le
UNIVERSITE PARIS VII
1986
THESE DE DOCTORAT D'ETAT
DE BIOLOGIE HUMAINE
Discipline
Physiologi
Présentée par Salimata Wade
Sujet :TRANSTHYRETINE (PREALBUMINE)
ET
THYMULINE DANS LA MALNUTRITION
PROTEINO-ENERGETIQUE
==---
CC>NSEll=AFRICAIN ET MALGACHE
POUR l'ENSEIGNEMENT SUPERIEUR
---
C. A. M. E. S. -
OUAGADOUGOU
00006
\\ Arrivée .:.1- 5· t1·A1,.19,~~ .(). ri ',SI
Soutenue le 16- l~-Enœ9Jg.e-e SO'J" n ·,1 0 U"
.
Devant le jury composé de :
NAVARRO Jean,
Président
LEMONNIER Daniel,
Directeur de la thèse
DARDENNE Mireille,
LESTRADET Henri,
MIGNON Michel,
.
WHITEHEAD Roger.
...

---

1
·1
Ce travail a été effectué à l'U1
INSERM,
Unité de
Recherches sur la Nutrition et
l'Alimentation que dirige le
1
Docteur Daniel Lemonnier.
C'est à lui que
je dois ma formation à
la recherche.
Après avoir dirigé mon DERBH,
il m'a de nouveau
1
accueillie dans son laboratoire.
Il
m'a fait bénéficier d'un
1
soutien matériel,
moral et financier sans lequel cette thèse
n'aurait pas eu lieu.
L'intérêt constant qu'il a manifesté dans
1
mes efforts a été pour moi
le plus grand des stimulants.
Qu'il
trouve ici l'expression de ma vive gratitude et de mon plus
1
profond attachement.
1
Je voudrais adresser mes respectueux remerciements à ceux
"1
qui ont bien voulu constituer le
jury de
cette thèse
:
Le Professeur Jean Navarro
(Service de Pédiatrie Générale
1
- Nutrition -
Gastro-entérologie,
Hôpital Bretonneau)
qui m'a fait
l'honneur de
juger ce travail et d'accepter la présidence du
jury,
1
1
Le Professeur Henri Lestradet
(Service de Pédiatrie,
Hôpital Hérold),
1
Le Professeur Michel Mignon
(Service de Gastro-
1
entérologie.
Hôpital Bichat).
,1
Le Docteur Roger Whitehead
(Dunn Nutrition Unit,
1
Cambridge)
que
je re~rcie aussi pour l'intérêt qu'il m'a toujours
manifesté.
J'espère que nos projets de collaborations futures
1
seront fructueux
et durables.
1
Je me sens particulièrement redevable au
Docteur Mireille
,j
1
Dardenne
(U25 INSERM,
Hôpital Necker),
d'avoir accepté de
r j
juger
bJ

---

f~1
!
1
travail, et d'avoir été pour moi une collaboratrice dynamique et
1
efficace. Je la remercie pour la sympathie qu'elle m'a toujours
témoignée, pour son aide et ~es conseils. Je lui renouvelle ici ma
1
sincère amitié.
1
1
1
1
1
1
1
1
1
1
1
1
l
1
1
,i
!;
1
j
;~
1
1
Il

---

1
1
Ce travail est le travail d'une équipe et il m'est très
1
agréable d'exprimer ma vive reconnaissance et mon amitié
indéfectible à Fanny Bleiberg-Daniel et à Béatrice Le Moullac, mes
1
compagnes de
tous
les
jours.
Fanny,
Béa et moi-même avons formé au
1
sein de
l'unité une équipe inséparable,
que,
j'espère, ~OOO km de
distance ne
sépareront pas.
Je
tiens à
leur dire à quel
point j'ai
1
apprécié leur amitié,
leur compétence et leur collaboration loyale
1
et efficace.
1
C'est aussi un travail de collaboration comme en
témoignent les signatures des articles.
La liste serait
longue si
1
je devais énumérer toutes
leSpersonnes ayant collaboré de près ou
de
loin à ce travail
; je les remercie tous ainsi que
les parents
1
des enfants qui ont participé à nos études,
les médecins et les
1
infirmiers
français
et
sénégalais qui
nous ont apporté
leur aide.
1
Mes remerciements s'adressent aussi à ceux qui au niveau
de
l'Unité 1 apportèrent leur concours
1
Merci Elvire Messialle pour ta gentillesse et ton
1
dynamisme
il a
fallu
toute
l'efficacité de
ton savoir-faire dans
la gestion financière et dans
les commandes pour surmonter les
1
difficultés qui auraient pu empêcher le déroulement normal de ce
1
travail.
1
Merci Jean-Pierre Suquet de m'avoir fait bénéficier de tes
1
conseils et de ta haute compétence dans
l'informatique et le
traitement statistique des données.
1

---

1
1
1
Merci Genevi~ve Mathon, notre "petite maman" pour ton
amitié et ton dynamisme
dans
la
fabrication
des
régimes.
1
1
Merci Abdourahmane N'Diaye pour l'entretien des animaux
qui est un travail
lourd et que tu
fais
avec dévouement.
1
Merci Jeannette Genevois,
Nadine Kaniewski,
Fatma Ardjoune
1
et Nadia SaIfi d'avoir bien voulu assurer
le
lourd travail
de
1
dactylographie.
1
Merci Claudie Flament pour avoir si gentiment accepté de
relire mon
texte
en français.
1
Je
remercie également tous mes collègues de
l'U1
pour leur
1
accueil et plus particulièrement ceux qui m'ont aidée par leurs
1
conseils,
leur amitié et leur soutien moral de
tous les jours.
Les
contacts que
j'ai eu au niveau du
laboratoire m'ont appris
1
beaucoup de choses sur la nature humaine et je l'espère, m'auront
enrichie autant que
les apports scientifiques.
1
1
1
1
1
1
1
1

---

1
1
1
1
Je
voudrais ici,
rendre hommage et remercier du fond
de
1
mon coeur une grande
darne Sénégalaise,
le Docteur Marie-Thérèse
Basse
(ancienne Directrice Générale de
l'Institut de Technologie
1
Alimentaire de Dakar
I.T.A.)
car,
sans
son aide et ses
1
encouragements, cette présente thèse n'aurait pas eu lieu.
Je
voudrais
lui
témoigner de nouveau ma profonde gratitude et mon
1
amitié.
Je
remercie également le Docteur Ousmane Kane qui lui
avait succédé à ce poste de
l'I.T.A. et qui m'avait toujours
1
encouragée et permis de poursuivre mes travaux.
1
C'est au Professeur Yves Ingenbleek que
je dois l'intérêt
1
de mes
recherches sur
la
préalbumine. Je
lui
renouvelle ma
profonde reconnaissance,
mon amitié et le remercie beaucoup de son
1
encouragement et de
son aide dans mes débuts
difficiles.
1
Je
voudrais dire merci
à tous mes amis et amies en
1
particulier au Professeur Thorkild Bog Hansen
(Protein Laboratory
Université de Copenhague)
"which I
thank very much
for his
1
excellent advices and
for
reviewing my english papers".
Merci
aussi à Pap,
à mes parents,
à mon frère Moustapha Wade.
1
1
1
Cette thèse est dédiée à tous
les étudiants en Nutrition
des pays en voie de développement pour les encourager,
quelles que
1
soient les difficultés
rencontrées,
à poursuivre leurs études et
1
travaux,
en vue d'une meilleure compréhension des problèmes
nutritionnels du Tiers-Monde.
1

---

1
-1-
1
1
SOMMAIRE
1
1
Pages
1
1
LISTE DES PUBLICATIONS CONSTITUANT LA THESE...............
2
1
INTRODUCTION. • • • • • • • • • . • • • • • • • • . • • • • • • • • • • • • • • • • • • • • • • • • • •
6
1
1
REFERENCES CITEES DANS L' INTRODUCTION. • • • • • • • • • • • • • • • • • • • •
24
1
CHAPITRE
l
:
TRANSTHYRETINE DANS LA MPE
ARTICLE 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
29
1
ARTICLE
2 ••••••••••••••••••••••••••••••••••••••.••
38
1
ARTICLE
3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
82
ARTICLE 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
116
1
ARTICLE 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
125
1
CHAPITRE
II
:
THYMULINE DANS
LA MPE
1
ARTICLE 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
129
ARTICLE
7 . . . . . • • . • . • . . . • . • • • • . . . • . . • . • • . • . • . . . • . . .
135
1
ARTICLE 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
141
ARTICLE 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
148
1
1
CONCLUSIONS GENERALES •••••••••••••••••••••••••••••••••••••
186
1
1

---

1
-2-
1
1
LISTE DES PUBLICATIONS CONSTITUANT LA THESE
1
Cette thèse qui comporte deux parties est basée sur des travaux
1
publiés ou articles soumis pour publication suivants:
1
1
CHAPITRE l
Transthyrétine
(préalbumine) dans la
Malnutrition Protéino-Energétique
(MPE).
1
1
A- Etudes sur l'animal
1
-
Article 1
1
F. BLEIBERG-DANIEL, R. VRANCKX,
S. WADE and E.A.NUNEZ (1985).
A
1
simplified method for the preparation of rat thyroxine-binding
prealbumin.
Factors influencing its circulating level.
Biochimica
1
et Biophysica Acta,
828,
270-277.
1
-
Article 2
1
WADE S., BLEIBERG-DANIEL F.,
LE MOULLAC B., BlOU D., GAUTHIER F.
and LEMONNIER D.
(1986). Value of serum transthyretin level in the
1
assessment of marginal
protein-energy malnutrition.
(soumis à
Clinical Chemistry).
1
1
-
Article 3
s. WADE, F. BLEIBERG-DANIEL, B.LE MOULLAC. (1986). Rat
1
transthyretin. The effects of acute short-term food deprivation on
the serum and cerebrospinal fluid concentration and on the hepatic
1
mRNA LEVEL.
(soumis à Journal of Nutrition).
1

---

1
1
-3-
1
B- Etudes sur l'homme
1
-
Article 4
1
F. BLEIBERG-DANIEL, S.WADE, C.
LABARRE, D. BALAGNY,
A.
FICHELLE,
1
J. BORY, J.M.
DESMONTS and D.
LEMONNIER (1985).
Variations in
plasma thyroxine-binding prealbumin
(TBPA)
in relation to other
1
circulating proteins in post-operative patients during rapid oral
refeeding.
Human Nutrition: Clinical Nutrition,
39C, 55-62.
1
-
Dans article 9 du chapitre II.
1
Transthyrétine chez des enfants Sénégalais cliniquement bien
1
portant et
vivant à Paris, et chez des enfants Sénégalais vivant
dans leur milieu et sélectionnés comme suit
: infectés et
1
souffrant de MPE sévère,
présentant des infections mineurs,
cliniquement bien portants.
1
1
-
Article 5
S. WADE, F. BLEIBERG-DANIEL,
and B.
LE MOULLAC
.
Transthyretin as
1
measured in capillary and venous plasma.
(à paraitre dans Clinical
Chemistry
, Novembre 1986).
1
1
CHAPITRE II
Thymuline dans la Malnutrition
1
Protéino-Energétique.
1
-
Article 6
1
B. MAIRE,
S. WADE, F.
BLEIBERG-DANIEL, M.
DARDENNE, G.
PARENT, P.
LE FRANCOIS and C. CARLES (1982).
Absence of variation in facteur
1
thymique sérique activity in moderately and severely malnourished
Senegalese children.
Am. J.
Clin.
Nutr., ~, 1129-1133.
1

---

1
1
-4-
1
1
-
Article 7
M.
DARDENNE, W.
SAVINO,
S. WADE, D.
KAISERLIAN, D.
LEMONNIER and
1
J.F. BACH (1984).
In vivo and in vitro studies of thymulin in
marginally zinc-deficient mice.
Eur. J.
Immunol., l!, 454-458.
1
1
-
Article 8
S. WADE, F.
BLEIBERG, A. MOSSE, J.
LUBETZKI,
H.
FLAVIGNY,
Ph.
1
CHAPUIS,
D. ROCHE,
D.
LEMONNIER and M.
DARDENNE (1985).
Thymulin
(Zn-facteur thymique serique)
activity in anorexia nervosa
1
patients.
Am. J.
Clin.
Nutr., ~, 275-280.
1
-
Article 9
1
S. WADE, G. PARENT, F.
BLEIBERG-DANIEL, B. MAIRE,
M.
FALL, D.
SCHNEIDER,
B.
LE MOULLAC and M.
DARDENNE (1986).
Thymulin
(Zn-FTS)
1
activity in protein-energy malnutrition : neW evidence for
the
1
interaction between malnutrition and infection on the Thymie
fonction.
(accepté dans Am. J.
Clin.
Nutr.).
1
1
1
1
1
1
1
1

---

1
1
-5-
1
1
1
1
1
1
1
1
INTRODUCTION
1
1
1
1
1
1
1
1
1
1
1

---

1
-6-
1
1
INTRODUCTION
1
La malnutrition a été définie comme un état pathologique
1
résultant de la carence ou de
l'excès,
relatif ou absolu,
d'un ou
de plusieurs nutriments essentiels, que cet état se manifeste
1
cliniquement ou ne soit décelable que par des analyses
1
biochimiques,
anthropométriques ou physiologiques (JELLIFFE,
1969).
1
LA MALNUTRITION PROTEINO-ENERGETIQUE (MPE)
1
1
L'existence de troubles nutritionnels liés à des carences
e
protéiques et/ou énergétiques est connue depuis le 16
siècle,
1
même si ce n'est qu'en 1959 que le terme MPE,
proposé par Jelliffe
fut unanimement adopté (JELLIFFE, 1959 ;
revue par Mc LAREN et al.
1
1982).
1
En 1930,
un Français nommé MARFAN (in Mc LAREN et al.
1982)
tenta de classer la MPE en désordres nutritionnels modérés
1
hypothrepsie
(état d'affaiblissement dû à la dénutrition),
et
sévères
: athrepsie
(état cachectique constituant
la phase ultime
1
de la dénutrition).
Cependant,
l'existence de formes modérées de
la MPE ne fut officiellement reconnue par la communauté
1
scientifique qu'en 1962 au cours du premier Symposium de la
1
Fondation suédoise de Nutrition,
qui avait alors pour thème "Mild-
to-moderate protein-caloric malnutrition"
(BLIX,
1963).
1
1
1
1

---

1
-7-
1
1
On peut assimiler la MPE modérée à un ou des états de
1
carences (ou de déficit)
protéino-énergétiques qui, contrairement
1
aux états de MPE sévères (marasme, kwashiorkor,
kwashiorkor-
marasmique),
n'entraînent pas ou très peu de modifications
1
cliniquement décelables.
1
CRITERES DE DEFINITION DE LA MPE
1
D'une manière générale,
les méthodes proposées pour la
1
détection de la MPE peuvent être classées en plusieurs catégories.
Leurs limites et leurs utilisations ont fait l'objet de plusieurs
1
revues générales (JELLIFFE, 1969
; SAUBERLICH et al. 1974 ;
1
ALLEYNE et al.
1979 ; GRANT et al.
1981 ; HAIDER
et al. 1984).
Brièvement, on peut citer:
1
1 -
les indices cliniques, détectables uniquement dans les
cas sévères de MPE.
1
2 -
les mesures anthropométriques
âge, poids,
taille, plis
1
cutanés, circonférence du bras.
3 -
les mesures biochimiques comprenant
:
1
. les mesures d'indices de composition corporelle
pour l'évaluation de la masse maigre ou des réserves protéiques,
1
excrétion urinaire
de créatinine sur des urines de 24 h ou 72 h,
1
excrétion urinaire de la 3-methylhistidine, de l'hydroxyproline ou
de l'hydroxylysine
1
.
la mesure des taux d'acides aminés circulants
1
1
1

---

r 1
1
-8-
1
1
· la mesure du taux circulant de certaines
protéines viscérales principalement sécrétées par le foie
1
albumine,
transferrine, thyroxine-binding prealbumine
(transthyrétine),
retinol-binding protein.
1
4 -
les mesures de certains paramètres de l'immunité
1
• taux d'immunoglobines circulants
• fractions du complément
1
· nombre et sous-populations de cellules T
circulantes
1
· tests d'hypersensibilité retardée
1
·
tests de transformation lymphoblastique.
5 -
les mesures de consommations individuelles
1
Notons cependant que chez l'homme,
au cours des enquêtes
de consommation, on a pris l'habitude de comparer les ingesta
1
mesurés aux apports ou allocations recommandés.
Le fait que ces
1
ingesta soient inférieurs aux allocations ne signifie pas
nécessairement une carence mais un risque de déficience, risque
1
qui croît d'autant plus que les valeurs mesurées sont en dessous
du niveau recommandé.
L'existence d'une déficience protéino-
1
énergétique peut seulement être mise en évidence par des examens
1
cliniques, anthropométriques et/ou biochimiques.
1
1
1
1
Il

---

1
-9-
1
1
On peut également signaler le développement récent de
1
nouvelles méthodes non invasives mais lourdes et complexes basées
1
surtout sur l'analyse de la composition corporelle (GARROW,
1982
BORKAN et al.
1982 ; MATTHEWS et al. 1983 ; BEDDOE et al. 1984 ;
1
LUKASKI et al.
1985)
1
-
Estimation de la masse maigre par :
1
• la mesure de la radioactivité naturelle du
potassium (40 K). Le potassium est une composante presque exclusive
1
du muscle et des viscères
· la mesure de la conductivité électrique du corps
1
ou impédance,
basée sur le fait que la masse maigre est plus
1
conductrice que la masse grasse
·
l'activation neutronique d'isotopes tel que
1
l'azote,
pour estimer l'azote corporel total.
-
Estimation de l'eau corporelle totale par:
1
· l'utilisation d'isotopes stables
(SH,18 0 ) selon
1
le principe de la dilution isotopique.
- Estimation de la masse grasse par
1
· la tomographie
·
l'utilisation de gaz rares tel que le Krypton,
1
plus soluble dans la graisse que dans
l'eau,
selon le principe de
1
la dilution isotopique.
La plupart de ces techniques sont en cours de validation ;
1
pour le moment,
elles ne sont pas applicables dans l'évaluation du
statut nutritionnel des enfants dans les pays en voie de
1
développement
(PVD).
1
1

---

1
1
-10-
1
1
De plus,
aucune des techniques citées, classiques ou
récentes,
ne peut à elle seule être considérée comme marqueur
1
unique de la MPE.
La combinaison de plusieurs d'entre elles serait
plus efficace pour déterminer l'état nutritionnel,
mais
jusqu'à
1
présent,
aucune combinaison appropriée n'a pu être proposée.
1
CRITERES DE DEFINITION DE LA MPE MARGINALE OU MODEREE
1
Du fait de la lourd
1
biochimiques
(excrétion uri
la
1
créatinine), de l'absence,
(index créatinine/taille),
(acides
1
aminés,
albumine) ou de spécificité (transferrin,
retinol binding
proteip) de certains marqueurs,
la MPE marginale ou modérée est
1
classiquement établie chez l'homme par référence à des mesures
1
anthropométriques
(GOMEZ, 1956
; JELLIFFE, 1966 ; WATERLOW, 1972,
1973 ; Mc LAREN et al.
1972).
1
Nous citerons ici deux exemples de classification de la
1
MPE marginale ou modérée couramment utilisée.
1
1. Classification selon GOMEZ (GOMEZ et al., 1956).
Le paramètre mesuré est le poids,
exprimé en pourcentage du poids
1
attendu par rapport à l'âge du sujet;
le critère optimal étant
ème
fourni par le 50
percentile des normes de HARVARD.
1
1
1
1

---

1
-11-
1
1
Classification
% Poids par rapport à l'âge
1
normal
> 90 %
1
er
MPE marginale
(1
degré)
89-75 %
ème
1
MPE modérée
(2
degré)
74-60 %
ème
MPE sévère
(3
degré)
< 60 %
1
2. Classification selon WATERLOW
(WATERLOW,
1972, 1973).
1
Les paramètres mesurés sont
le poids et
la taille exprimés
1
soit en pourcentage du poids attendu par rapport
à
la taille mesurée
(%poids/taille).
1
. soit en pourcentage de la taille attendue par
rapport à
l'âge du sujet
(%taille/âge).
1
eme
Les critères optimaux sont là aussi fournis par le 50
1
percentile des normes de HARVARD et
plus récemment des normes du
NCHS (National Center for Health Statistics). WATERLOW définit une
1
malnutrition
"récente"
(ou wasting)
en considérant
le paramètre
poids/taille, et une malnutrition "passée"
(ou stunting),
1
caractérisée par un retard de taille par rapport à l'âge.
Il considère dans chaque catégorie 4 degrés:
degré 0 = normal,
1
degré 1 = marginale, degré 2 = modérée, degré 3 = sévère. Ainsi,
1
pour le poids/taille: 0 = >90 %, 1 = 90-80 %, 2 = 80-70 %, 3 =
<70 % ; et pour la taille/âge
o = > 95 %, 1 = 95-90 %,
1
2 = 89-85 %, 3 = <85 %.
1
1
1
1

---

1
1
-12-
1
1
En se basant sur la classification proposé par GOMEZ,
BENGOA a estimé en 1974,
le nombre d'enfants de 0 à 4 ans
1
souffrant de MPE modérée ou sévère dans 3 régions du monde.
1
Amér ique Latine
MPE sévère
700.000
1
MPE modérée
9
.000.000
Afrique
MPE sévère
2.700.000
1
MPE modérée
16.000.000
Asie (sauf la
MPE sévère
6.000.000
1
Chine et le Japon)
MPE
modérée
64.000.000
1
Même si les subdivisions de la MPE en sévère, marginale ou
1
modérée,
proposées à partir des données de l'âge, du poids et de
la taille ont pu permettre de juger de l'ampleur de la MPE
1
modérée,
comme l'indique les données de BENGOA,
ces subdivisions
1
n'ont aucune base physiologique et sont très arbitraires surtout
en ce qui concerne la MPE marginale ou modérée
(SAUBERLICH, 1976
1
GOPALAN et al.
1984).
Ainsi apparaît-il le besoin de disposer d'indicateurs plus
1
sensibles que l'anthropométrie pour la détection des stades
1
précoces de la malnutrition. Ce besoin est double.
Il correspond à
des objectifs prioritaires de santé publique dans les pays en voie
1
de développement et à une demande en milieu hospitalier des pays
développés
(SIMOPOULUS, 1982). Ces indicateurs doivent être
1
simples à obtenir et à quantifier, et être utilisables sur le
terrain chez de
jeunes enfants qui représentent un groupe à haut
1
risque.
1
1

---

1
1
-13-
1
1
En 1972,
la transthyrétine a été proposée comme marqueur
sensible de la malnutrition dans
la mesure où sa concentration
1
plasmatique,
initialement basse chez des enfants sévèrement
1
dénutr~s, augmente plus rapidement que celle de l'albumine ou de la
transferrine au cours de la réalimentation.
1
1
BREF
RAPPEL SUR LA TRANSTHYRETINE
1
1
La transthyrétine,
anciennement dénommée préalbumine ou
thyroxine-binding préalbumine
(TBPA)
est une protéine vectrice des
1
hormones thyroïdiennes
(thyroxine et triiodothyronine)
et de la
vitamine A par l'intermédiaire de
la protéine liant
le
rétinol
1
(RBP).
Elle forme avec la RBP et le rétinol un complexe circulant
TBPA-RBP-rétinol dans un rapport molaire 1-1-1.
Sa liaison avec la
1
RBP est indépendante de celle des hormones thyroïdiennes,
d'où sa
1
nouvelle dénomination de "trans-thy-rétine".
Chez l'homme,
c'est
le deuxième transporteur des hormones thyroidien~es et
environ
1
40% des molécules de TBPA transportent le rétinol plasmatique.
1
1
lune revue bibliographique non exhaustive est
présentée après
l'introduction.
Nous avons voulu
éviter une répétition d'une
1
bibliographie déjà mentionnée dans les articles qui constituent
cette thèse.
Le lecteur, désirant des informations plus complètes
1
sur la transthyrétine et sur la thymuline pourra se rapporter
1
directement aux chapitres concernés.
1
1

---

1
-14-
1
1
La transthyrétine fut d'abord découverte dans le liquide
1
céphalo-rachidien (LCR)
, avant d'être reconnue présente dans le
1
sérum.
Les données actuelles montrent que la même protéine est à
la fois synthétisée par les plexus choroïdes
(TBPA du LCR),
et par
1
le foie
(TBPA du sérum).
Cependant son
(ou ses) site
(s) de
dégradation reste(nt)
inconnu(s).
1
C'est probablement la protéine la mieux étudiée du point
1
de vue structure. C'est un tétramère formé de quatre sous-unités
identiques de 127 acides aminés.
Elle a été cristallisée et sa
1
conformation cristallographique, de même que ses sites de liaisons
avec les hormones thyroïdiennes et avec la RBP,
sont parfaitement
1
connus.
C'est une protéine extrêmement stable, non glucosylée,
1
d'un poids moléculaire d'environ 55.000 daltons. Chez l'homme,
sa
demi-vie est de 2
jours et sa masse intravasculaire est
1
relativement
faible et représente environ la moitié de son taux
circulant. Son taux sérique, de 0.16 à 0.3 g/l dépend de l'âge et
1
du sexe. Jusqu'à présent, des taux sériques bas de transthyrétine
1
d'origine héréditaire n'ont pas été décrits.
Par contre la
concentration sérique de transthyrétine est abaissée par un état
1
inflammatoire et/ou infectieux,
ce qui lui a valu son nom de
protéine "négative" de la réaction inflammatoire
(llnegative acute-
1
phase reactant protein").
1
1
1
1
1

---

1
1
-15-
1
1
Récemment,
le gène codant pour la transthyrétine a été isolé
et un brin de DNA complémentaire de l'ARN messager spécifique de
1
la transthyrétine a pu être synthétisé.
La transthyrétine existe chez toutes les espèces et des
1
homologies de structure et de fonction ont été déterminées chez
1
certaines, dont
le rat et
l'homme.
En 1978, BURTON et al ont montré que la transthyrétine
1
avait une activité hormonale semblable à celle des hormones
thymiques dont la thymuline.
Par la suite, DARDENNE et al.
(1980)
1
ont suggéré qu'une partie de la thymuline circulante,
hormone
1
sécrétée par le thymus,
serait transportée par une protéine située
dans la zone de migration albumine-préalbumine.
On ne connait,
1
jusqu'à présent,
aucune autre étude ayant confirmé ou rejeté ces
hypothèses.
1
LE THYMUS DANS LA MPE
1
1
La malnutrition protéino-énergétique, qui est complexe car
souvent associée à des infections,
a des effets néfastes sur le
1
système immunitaire.
Dans les pays en voie de développement,
les
maladies infectieuses demeurent encore très fortes,
1
particulièrement chez les
jeunes enfants.
La mortalité et la
1
morbidité élevées qui en résultent sont supposées être dues,
en
grande partie,
par la moindre résistance de l'organisme causée par
1
la malnutrition.
Les études systématiques,
destinées à explorer en
1
1
1

---

1
-16-
1
1
détail l'immunocompétence dans la MPE,
les interactions nutrition-
1
infection-réponse immunitaire, ont débuté dans les années 50 et
1
depuis,
il existe un grand nombre de publications montrant les
effets de la MPE sur le système immunitaire.
Le bilan de ces
1
travaux a fait
l'objet de plusieurs revues générales citées dans
la chapitre II de ce travail.
1
Bien que les mécanismes impliqués dans l'altération de la
1
réponse immunitaire au cours de la MPE restent encore non
élucidés,
il ressort de ces travaux que le développement des
1
organes lymphoïdes et plus particulièrement du thymus est
fortement inhibé par la malnutrition. La MPE sévère et quelquefois
1
ses formes modérées affectent plus gravement
l'immunité à
1
médiation cellulaire que l'immunité humorale ou les défenses non
spécifiques.
Chez l'homme,
on observe particulièrement une
1
diminution ou une absence de réponse aux tests d'hypersensibilté
retardée,
une réduction in vitro de la prolifération
1
lymphoblastique des cellules T en présence de mitogènes, une
1
réduction in vitro du nombre de cellules T matures et une
perturbation dans la distribution des différentes sous-populations
1
de lymphocytes T (augmentation du nombre de cellules
ne formant
pas de rosettes E,
augmentation du nombre de cellules T
1
suppressives,
diminution du nombre de cellules T-auxilliaires).
1
1
1
1
1

---

1
1
-17-
1
1
Le thymus,
a été longtemps considéré comme le "baromètre"
de la malnutrition.
Chez
l'animal,
nous avons observé une atrophie
1
thymique quel que soit le type ou le degré de malnutrition
(WADE
et al . .1983).
L'influence de la MPE sur la taille et
la structure
1
du thymus est largement reconnue.
Les altérations histologiques,
1
décrites dans la littérature à partir d'études post-mortem chez
l'enfant dénutri,
consistent en une déplétion cellulaire partielle
1
ou totale des zones corticales et paracorticales du thymus,
une
réduction de l'épaisseur du cortex,
une augmentation de la masse
1
fibreuse interstitielle.
1
La composante épithéliale du thymus a été très peu
étudiée.
Cependant,
la trame du thymus aussi bien corticale que
1
médullaire est essentiellement constituée de cellules épithéliales
qui sont à
l'origine de la sécrétion d'hormones thymiques
1
nécessaires à la maturation et à la différenciation des
1
lymphocytes T (PAPIERNIK, 1986).
Plusieurs hormones thymiques ont été isolées et plus ou
1
moins partiellement caractérisées;
les plus connues sont la
thymosine fraction v,
la thymopoïétine et la thymuline
1
anciennement nommée Facteur Thymique Sérique
(FTS).
1
1
1
1
1
1

---

1
1
-18-
1
1
BREF RAPPEL SUR LA THYMULINE
1
La thymuline est un nonapeptide bien caractérisé, de poids
moléculaire environ 1000 daltons.
Elle a été découverte par
1
l'équipe de BACH à Necker, qui l'ont isolée à partir du sérum.
1
Elle a été purifiée sur sa capacité de faire apparaître l'antigène
Thy-l,
marqueur de maturation sur les cellules précurseurs de la
1
moelle osseuse.
Elle est exclusivement sécrétée par
l'épithélium
thymique, mais son activité biologique est détectable dans le
1
sérum de plusieurs espèces y compris de
l'homme.
Son niveau
1
sérique dépend de l'âge et de l'involution thymique.
Récemment, on
a montré que la molécule de thymuline contient du zinc et que ce
1
métal est indispensable à son activité biologique,
in vivo et in
vitro.
En fait,
la thymuline existe sous deux formes: une sans
1
zinc,
biologiquement inactive et une autre, complexée au zinc,
1
biologiquement active.
L'influence du zinc sur la conformation de
la thymuline a été récemment démontrée par résonance magnétique
1
nucléaire
(LAUSSAC et al., 1980).
La thymuline semble posséder plusieurs influences sur la
1
fonction des lymphocytes T,
la plus connue étant sa capacité
1
d'induire des marqueurs de différenciation sur les cellules,
capacité sur laquelle est basée son dosage biologique
(test des
1
rosettes).
Cependant,
l'existence d'un facteur appelé "facteur
1
1
1
1

---

1
-19-
1
1
allogénique"
(AF),
ayant la même activité que la thymuline sur le
1
test des rosettes,
a été démontré au cours de réactions
1
allogéniques. Ce facteur pourrait être produit par des cellules T
"activées".
Néanmoins, AF est discernable de la thymuline par sa
1
charge électrique. De plus,
il n'est pas sensible aux inhibiteurs
sériques de la thymuline
(molécules de haut poids moléculaire
1
d'environ 100.000 et 300.000 daltons).
Ce facteur n'est pas non
1
plus reconnu par les anticorps anti-thymuline.
1
BUTS ET OBJETS DU TRAVAIL
1
Notre travail comporte deux parties.
1
Dans la première partie intitulée transthyrétine dans la MPE
(chapitre 1),
nos objectifs ont été d'établir les effets
1
spécifiques de la malnutrition et de l'inflammation sur les taux
1
circulants et le métabolisme de la transthyrétine, de préciser son
rôle et son utilisation en tant que marqueur de la MPE marginale
1
ou modérée.
C'est un fait que de nombreux travaux ont été consacrés à
1
la transthyrétine en relation avec l'état nutritionnel. Ces
1
travaux émanent autant des pays en voie de développement (PVD) que
de milieux hospitaliers des pays développés.
Cependant,
les
1
carences nutritionnelles, et plus particulièrement celles des
enfants des PVD,
sont souvent associées à un état inflammatoire
1
1
1
1

---

1
-20-
1
1
et/ou infectieux.
On ne sait donc pas à quels facteurs doivent
1
être attribués des taux sériques bas de transthyrétine chez les
sujets dénutris.
1
Ces travaux cliniques nous ont fait
apparaître la
1
nécessité d'une expérimentation rigoureuse sur
l'animal qui
permette d'établir l'influence spécifique de la MPE sur la
1
transthyrétinémie.
Nous avons choisi le rat,
du fait de la
similitude presque parfaite de structure et de fonction entre la
1
transthyrétine de rat et celle de
l'homme.
Ces deux protéines sont
1
cependant antigéniquement différentes.
Afin de préciser les conditions d'utilisation de la
1
transthyrétine dans la malnutrition,
il nous a semblé également
important d'établir
entre les taux
1
à l'inflammation,
1
post-opératoire et
dont l'état nutritionnel peu
1
Dans la deuxième partie, thymuline dans la MPE (chapitre
1
II),
notre objectif a été d'étudier les répercussions de la MPE
1
sur la fonction thymique,
évaluées par la mesure de l'activité
biologique sérique de la thymuline.
Cette activité a été
1
déterminée dans différents états de MPE associés ou non à des
infections et dans la carence expérimentale en zinc.
1
1
1
1
1

---

1
1
-21-
1
1
En effet,
l'implication des facteurs thymiques, dans les
perturbations de la réponse immunitaire à médiation cellulaire
1
chez des enfants dénutris,
a été suggérée à partir de deux travaux
menés au Nigéria et au Bengladesh.
Ces études avaient montré que
1
des lymphocytes T circulants,
provenant d'enfants souffrant de MPE
1
sévère,
pouvaient acquérir des marqueurs de maturation,
après
incubation in vitro avec de la thymosine fraction V ou de la
1
thymopoiétine.
D'autres travaux ont montré par la suite que l'activité de
1
la thymuline était abaissée chez des enfants prématurés,
chez des
1
enfants souffrant de malnutrition sévère,
chez des animaux
énergétiquement restreints ou sévèrement carencés en zinc.
1
Cependant,
les désordres
nutritionnels,
en particulier
chez l'enfant,
peuvent avoir des causes multiples et
la plupart de
1
ces études n'ont tenu compte ni des divers degrés de MPE,
ni des
1
carences associées (carence en zinc en particulier).
Elle n'ont
pas non plus permis de définir l'influence spécifique de la MPE,
1
de la carence modérée en zinc,
ou des infections concomitantes sur
l'activité hormonale du thymus.
1
1
METHODOLOGIE
1
L'ensemble du travail a été mené sur l'animal
(principalemnt dans le chapitre 1),
et sur
l'homme.
Le matériel et
1
les méthodes utilisés sont décrits en détail dans les articles
constituant les deux chapitres.
1
1
,1

---

1
1
-22-
1
1
Dans le chapitre l
Les études sur l'animal portent sur
-
l'isolement et la purification de la
1
transthyrétine de rat et l'obtention d'un anticorps monospécifique
1
en vue de son dosage.
Ce préalable était nécessaire car cet
anticorps n'était pas disponible dans le commerce (article 1).
1
-
la définition du taux normal de
transthyrétinémie en fonction de l'âge, du sexe, du mode de
1
prélèvement sanguin (article 1).
1
-
la réponse de la transthyrétine à
différents degrés de malnutrition protéique et/ou énergétique
1
(article 2).
-
la réponse de la transthyrétine à des
1
périodes de
jeûne de un ou plusieurs
jours, et à une période de
jeûne de 24 h,
suivie d'une réalimentation
les conséquences
1
d'une baisse de la transthyrétinémie sur les hormones
1
thyroïdiennes et sur le taux hépatique de l'ARN messager
spécifique de la protéine (article 3).
1
Les études sur l'homme portent sur:
1
-
la relation entre les taux circulants
1
de transthyrétine et les taux circulants des protéines de la
réaction inflammatoire,
après un acte chirurgical mineur
1
(interventions orthopédiques>,
ne nécessitant pas de transfusion
sanguine, chez des patients bien nourris
(article 4).
1
1
1
1

---

1
1
-23-
1
1
-
l'application chez des enfants des
pays en voie de développement des résultats des expériences
1
précédentes (dans article 9 du chapitre II).
1
-
la comparaison entre les taux
plasmatiques de transthyrétine du sang veineux et capillaire
1
(artice 5).
1
Dans le chapitre II,
nous avons étudié:
1
-
d'abord
l'activité de la thymuline
chez des enfants infectés apparemment normaux,
modérément ou
1
sévèrement dénutris,
les différents degrés de MPE ayant été
définis à partir des mesures anthropométriques (article 6). Ce
1
travail
nous a conduit à considérer 3 facteurs dans
l'activité de
la thymul ine.
1
-
le rôle de la carence modérée en zinc
1
(article 7).
-
le rôle de la MPE non associée à des
1
infections (article 8).
-
le rôle de la MPE modérée et sévère
1
définies selon un ensemble de marqueurs nutritionnels
1
biochimiques,
le rôle des infections, et de la production de
"facteur allogénique" sur la fonction thymique (article 9).
1
1
1
1
1

---

1
1
-24-
1
1
REFERENCES
1
ALLEYNE,
G.A.O.,
HAY, R.W.,
PICOU,
D.I., STANFIELD, J.P.,
WHITEHEAD, R.G.
(1979) Protein-energy malnutrition,
2nd ed.,
1
Arnold,
London.
1
BEDDOE,
A.H.,
HILL,
G.L.
(1984) Clinical measurement of body
1
composition using in vivo neutron activation analysis.
J.P.E.N.
9 : 504-520.
1
1
BENGOA,
J.M.
(1974) The problem of malnutrition. WHO chronicle
28
: 3-7.
1
BLIX, J.
(1963) Symposia of the Swedish Nutrition Foundation 1.
1
(J. BLIX ed.), Almquist and Wiksells,
Uuppsala.
1
BORKAN,
G.A.,
GERZOF,
S.G., ROBBINS,
A.H.,
HULTS,
D.E., SILBERT,
1
C.K., SILBERT, J.E.
(1982) Assessment of abdominal
fat content
by computed tomography.
Am. J. Clin.
Nutr.
36 : 172-177.
1
1
BURTON, P.,
IDEN, S., MITCHELL,
K.,
WHITE,
A.
(1978) Thymie
hormone-like restoration by human prealbumin of azathioprine
1
sensitivity of spleen cells from thymectomized mice.
Proc.
Natl.
Acad.
Sei.
USA, 75 : 823-827.
1
1
DARDENNE, M.,
PLEAU, J.M., BACH, J.F.
(1980) Evidence of the
presence in normal serum of a carrier of the serum thymie
1
factor FTS.
Eur. J.
Immun., 10 :
83-86.
1

---

1
1
-25-
GARROW,
J.S.
(1982) New approaches to body composition.
Am. J.
1
Clin.
Nutr.
35
: 1152-1158.
1
GOMEZ,
F.,
RAMUS-GALVAN,
R., FRENK, S., CRAVIOTO,
J.M., CHAVEZ,
1
R., VASQUEZ, J.
(1956) Mortality in third degree malnutrition.
J. Trop.
Pediatr.
2 :
77.
1
1
GOPALAN, C., JAYA RAO,
K.S.
(1984) Classifications of
undernutrition -
their limitations and fallacies.
J. Trop.
1
Pediatr.
30 :
7-10.
1
GRANT, J.P., CUSTER,
P.B., THURLOW,
J.
(1981) Current techniques
1
of nutritional assessment.
Surgical Clinics of North America.
61 :
437-
463.
1
HAIDER,
M.,
HAIDER,
S.Q.
(1984) Assessment of protein-calorie
1
malnutrition.
Clin.
Chem.
30
: 1286-1299.
1
JELLIFFE,
D.B.
(1959) Protein-calorie malnutrition in tropical
1
preschool children
: a review of recent knowledge. J.
Pediat.,
54:
227.
1
1
JELLIFFE,
D.B.
(1966) The assessment of the nutritional status of
the community. WHO Monograph series nO 53, Geneva.
1
JELLIFFE,
D.B.
(1969) Appreciation de l'état nutritionnel des
1
populations.
Serie OMS,
Genève.
1
1
1

---

1
1
-26-
LAUSSAC, J.P., CUNG,
M.T., PASELOUP,
M., HARAN,
R.,
MARRAUD, M.,
1
LEFRANCIER,
P.,
DARDENNE,
M., BACH,
J.F.
(1986) NMR study of
1
thymulin,
a lymphocyte differentiating thymic nonapeptide. J.
Biol.
Chem., 261
:
7784-7790.
1
LUKASKI,
H.C., JOHNSON,. P.E., BOLONCHUK, W.W.,
LYKKEN, G.l.
(1985)
1
Assessment of fat-free mass using bioelectrical
impedance
1
measurements of the human body.
Am.
J.
Clin.
Nutr.
41 :
810-
817.
1
Mc LAREN,
D.S., READ, W.W.C.
(1972) Classification of nutritional
1
status in early childhood.
Lancet 2 : 146-148.
1
Mc LAREN,
D.S.,
BURMAN,
D.
(1982)
in Textbook of Paediatric
1
Nutrition: historical
introduction
(Mc LAREN D.S., BURMAN D.
eds),
pp 3-17, Churchill Livingstone, Edinburgh London
1
Melbourne and New York.
1
MATTHEWS,
D.E., BIER,
D.
(1983) Stable isotope methods for
1
nutritional
investigation.
Ann.
Rev.
Nutr.
3 : 309-339.
1
PAPIERNIK, M.
(1986)
Les organes lymphoïdes dans Immunologie par
1
Jean- François BACH,
pp 79-100,
Flammarion Médecine-Sciences.
1
SAUBERLICH,
H.E.,
SHALA, J.H.,
DOWDY,
R.P.
(1974)
Laboratory tests
for the assessment of nutritional status. CRe Press, Boca Raton,
1
FL.
1
•
SIMOPOULOS,
A.P.
(1982) Assessment of nutritional status.
Am. J.
1
Clin.
Nutr.
Suppl.
35 : 1089-1325.
1

---

1
1
-27-
1
WADE, S., LEMONNIER, D., BLEIBERG, F., DELORME, J.
(1983) Early
nutritional experiments : effects on the humoral and cellular
1
immune responses in mice. J.
Nutr.
113 : 1131-1139.
1
WATERLOW, J.C.
(1972) Classification and definition of protein-
1
calorie malnutrition.
British Medical Journal,
3 : 566-569.
1
WATERLOW, J.C.
(1973) Note on the assessment and classification of
protein-energy malnutrition in children.
Lancet 2 : 87-89.
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
-28-
1
1
1
1
1
1
1
1
1
1
CHAPITRE l
TRANSTHYRETINE DANS LA MALNUTRITION PROTEINO-ENERGETIQUE
1
1
1
1
1
1
1
1
1
1

---

1
-29-
1
1
ART l C L E N° l
1
1
1
liA simplified method for the preparation of rat thyroxine-binding
prealbumin. Factors influencing its circulating level".
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-30-
1
1
1
270
BI/xirim,ca" Slapir.YI/ca Acta 828 (\\ 985\\ 270-2n
Ehev,er
1
BBA nus
1
A simplified method for the preparation of rat thyroxine-binding prealbumin.
Factors influencing its circulating level
1
Fanny BJeiberg u, Roger Vranckx h. Salimata Wadt" • and
Emmanuel A. N unez h
• (.'''11; d. R.cir.rrit'llllrla {""trll/a" .tl'Ailm.",alla". V.I INSERM. Hopl/ul Sicirut. 170 8d. .Vr,·. 7S0lR Prv'L <Utd
1
• Unl/. d. R.cit.rrn'I li. Siail,'!" d.ll"t.,aclla"I.\\o(ali-n.tlu'r.s ail CalUs dll O;~.lapp.m'''t. r.;.221INSER.\\o(. FU"lltt; d,
M;d.c,,,. Xa,·,.r·8icira,. 16 ru, H",rr·Huchur/i. nOl8 PUTlI' Frunc.,
(Rccelvrd Sclllemoer 191h. 1984\\
1RcvlSrd m~nu.cnl'l re<:etveQ J~nuarv 28rh. 1985\\
1
Kev .... ords: Thvro:une·blndin~ pre~lbumln: Pre~ihumln: Immunochcml~Ir:-: 1R.ll )cruml
1
Rat thyroxine-bindin~ prealbumin (TB PA) WOlS isolated in three simple steps by means of a serum
precipil:mon by a 5~ phenol solullon and rwo conseculive semi-preparall\\e pohac~'lamioe gel electro-
phoreses. The o\\'erail \\Idd ... as 15~ ;.lnd the TBPA preparation contained less than l~ impuriri~. ln ;.lddirion
1
a monospt'Ciiic lnllserum was r.lIsed in the rabbic. ln po" ac~'iamide gel. raI TBPA. ;lS ,,,th ilS hum an
counrerpart. migrated anodailv 10 Oilbumin ... hiie in a~arose ~el. its eiecrrophorelic mobili[\\' "'as ~imll:1r tO rhal
of aibumin. Serum TBPA me:lsured in aduit male Wisrar rats did nOI exhihil a circadi:ln rh\\lhm. Ho ..e'er.:1
signlficanl D% decrease "as obsen'ed berween 9 lnd 15 h. followed bv the reSlorallon of lhe iniliai ,aiue b\\
1
21 h. TBP.-\\ concenrrallon ..as me~ured in 1-. 15- and 13-Jav-<lid male ;lnd femaie pups as "el1 ;lS in .Jouit
rars. The le\\el of (his proreln increased from 1 lO 28 d:lVS of a~e and did nor discia\\' am se'l.uaJ dirrerenco!.
Ver. while TB PA concenlralions in aàuit maies \\Vere similar to Ihose recordeo in (he 13·<1a\\ ·i)id pups. for
aduit femaies. Ihey rerurned to the levels measured in Ihe l-JaY-Qld pups.
1
Inrroduetion
slmtiarltle:s. the rai see~s 10 be .1 sUll.Jbic: :nodei tO
1
slUùy the response oi TBPA lO speciic :lutntlOn.1i
Human .lS Il..dl .lS rat th\\'fo:\\IOe:,olOdlOi;; pre::l·
de:iiclencll:s. ln formui;llln~ our rese:n:;, aopro.Jch.
bumin (TBPA. commonl'\\.· n;lmc:ù pre;libumtnJ IS
lI.'e: first altempte:d tO Jevdop .1 slmaiiiieJ proce:·
of constù.:rable inlen:sl wllh resoet.:t [0 slructure
dure for prO\\'1din~ hi%rily purllÏed 1":11 TBPA. lS
1
and funcllons. as It lS .1 camer of bOlh vilamlO A
the few melhods of puniic::lllon re;:orted 10 the:
and Ihyro:\\lOe: (T.). ln man. the: prolelO IS llso
liter:lure ellher involve sever:l ste;:s or requlre :n
consldered as J ve~' senslllve marker of protelO-
affinllY medium not gener:lly J\\':uiable [3.4.71.
1
energy mainulmion (1.11. The molc:cubr struclure.
Other basiC dala are re;:orted as ,'..dl. including
Ihe amino acid compoSItion. Ihe: physlc:l prope:r-
slandard vaiuc:s for 1·. 15- and 13-J;lv old male:
tie:s and Ihe biologlc::i funcllons of human and rat
and femaie rats. aion~ wilh those: for aduit animais.
TBPA are very simlÏar [3.4\\. althou~h. in the rat.
As rat TBPA h:lli a ve~' short half·liie of 10 h (SI
1
TBPA is Ihe: major T, .:ame:r (5.61. ln VIe:W of Ihe:se:
or 30 h [71. circ:ldi;ln ,'ariallons were 10 be: e.'t.
pc:l.:led: consc:que:ndy. 'l''e alsll altempted 10 sec
• Tu whom ~u~Pùndcn~e ,h.... I<I he ~<I<1"",,..,.j.
whe:lher ;l 14 h rhythm ,n ,c:rum TBP.\\ .:unce:ntra-
Abh"",."",'n: TBP.... ,h,·n".nc.hln..hn~ pru.hunlln.
liuns mlght ..lI.:cur in ;luuit rats.
1
1
1

---

1
1
-31-
1
1
1
271
1
Malerials and Melhods
electrophoresis according to the method of Davis
[9] in a Bio-Rad model 155 gel electrophoresis ceil.
AflImals
using 7'l acrylamide in the running gel. The elec-
Adult male and female Wislar ratS nHa-Credo.
trophoresis was carried out at 4°C. The gels were
1
I"Arbresle. France) were used. They were fed ad
placed in a
2.10- 2
mg/ml solution of 1.8
libitum a commercial slock diet and had free access
anilinonaphtalene sulronic acid (ANS) in distiile:d
to waler. Three eltperimentS were conducted. Dif-
water. They were eltposed to ultraviolet light at
ferent melhods for coilecting blood from 7-week-
254 nm in order to eut up the TBPA-contalmng
1
old male raIS. divided in two groups of 8 animais
band \\3]. Ils prOlem content was elttracted at 4°C
each. were performed with and withoul diethyl
in an electrophoreuc concentralor (model 1750
ether aneslhesla in order to eltarmne the influence
electrophoretic sample concentrator. Isco. NE.
of the ongin of blood on TBPA level: the first
U.S.A.) using a 0.03 M Tris/0.06 M glycine buffer
1
group of raIs were decapllated. whereas the other
(pH 8.5). The protem e:ltlract was dialyzed and
group was aneslhetized with diethyl ether. In these
concentraled against 0.1 M NaO in a Mic:roprodi-
am mals. blood was coilecled successlvely from the
con c:oncentrator (Bioblock. Slrasbourg. France).
relro-orbilal plexus. the tail vem and the abdomi·
Second srmr·preparallve po/~'acn'Iamrdegelelec-
1
nal aOrla: then. they were decapitated. A group of
Irophoresrs. The TBPA solution th us obtained was
45 adult maie rats were divided in 9 subgroups of
subjected to a second round of semi-pre;Jarative
5 and decapllated at 3 h mtervals te see If circadian
electrophoresis. elttracted and coricentrated ac·
vanatlons ln serum TBPA rrught occur. Another
cording to the same procedures mentloned above.
1
group of ammais were mated. and' the pregnant
females as weil as thelr pups were fed a control
Qrht!r meQ!iuremenrs and procedures
diet pre!,ared ln the laboratorv. Pups were we:lned
TJt.~·ro.'Ctnl! bindtng capacUl·. An aliquOi of the
at 23 davs accordmg to current practlce m our
phenol supernatant and of the ..... hole ,er.lm wen:
1
laboratorv These rais were kliled at vanous ages
incubated overnll~ht at 4°C w\\th trace amoums oi
[0 deterrmne thelr serum TBP.-\\ <:oncentratlon. Al!
[1~IJlhyrolt1ne (s'Pee. JCt. ::00 ...Ci/!,g. Commis-
raIs were mamtamed al 2:: =lOC on a 12 h
sanat Energze Atomique: (CE".\\. Saclav (Francel.
light/ 1: h dark cycle i1ignts on 08:00 h and lights
Two concentrations of ['~~IIT•. 0.033 and 0..3.3
1
off :0:00 hl.
!'g/l00 ml phenol supernat~t or serum. wer~
used. The samples 1serum and ph.:nol suç:ernatanl
Puni/caHon procedures
labeled w\\th ('~IJT•. unlabeled ;e:rum and ;:hcnol
Phenul preClp,raHon of rhe serum. TBP.<-\\ was
supernatanll and a blank contammi ['~IJT. (:!'
1
isolated (rom 75 ml oi serum obtame:d from adult
10 - j !'g) but no prOtCIn were subJe:ct~d to analvlI-
maie raiS. Crystailine :"JaCl was adde:d .lt room
cal disc polvacryiJ.mldc gel dc:etrophoresl> .lccord-
temperature and dlssoived m the serum at :00 g; 1.
ing to DaVIS [91. The: gels were remùved and ,Ii<:e:d
An e:qu.l1 volume of S"ë (w/v) phenol in distlile:d
into 2·mm trans\\'erse sections. The: position of the:
1
wate:r (sdeeted bc:eause it opumail\\' increases the
albumm-.:ontaimng sli<:es. stamed wlth bromo-
intenSllv of the prole:m band anodal to albumlnl
phe:nol blue. was nOI.:d. E:u:h sli<.:e was ass.1ved for
was then slo",ly addc:d whlle: stimng.. The mlltture
r:zdioactlvity in a gamm:z counter. The uniabe:kd
was rapldlv centniuged:1I 17600 x .~ :lt 4°C for 30
samples were slained w\\th Coom~sle blue.
1
mm. The supe:rnatant was coilected and
th en
Preparatroft
of Q moftospl!crfic anlUeTUnI.
A
diaivzed elttenslvely for 48 h against runmng water
monospecific antiscrum to TBPA was pre!Jared in
and concentrated by lyophilisation.
two steps. 1 ml of a solution of punfied TBPA
Finr srmr·prepl.lram·e poll"ac,.,·ll.lmrde gel ele(uo-
(obtamed arter the fi~t semt-pre!Jarative poly·
1
phoreslS. The concentra le was dissolved in a 0.033
acrylamide gel electrophoresis) containing 140 !'g
M Tris/O.Ob M glycine buffer (pH 8.5) (10 ml
prOlem in 0.1 M NaCl was emuisified with an
buffe:r/75 ml raI serum,. ~d 20$ glycerol (v/v)
.:quai volume of complete Freund's :zdjuVOlnc. A
was the:n added to the solution. The mixture was
rab bic received le!J~ted sub~"Utaneous inJc:e:lIons
1
subjected to a $Cmi-pre!Jarative polyacrylamide gel
of lhe miltture. The rabbit was reimmuniud with
1
1
1

---

1
1
-32-
1
1
1
272
the same dose of antigen mixed with incomplete
1
extinction coefficient of the Lowry coler reaction
Freund's adjuvant 3 and 5 weeks after the primary
[14). Thus. in accordance with previous reports
injection. Blood was collected 1 week thereafter.
[4,1 41. the values obtained by the Lowry assay
The immunoglobulin fractions of the antiserum
were multiplied by 0.89. Serum TBPA level was
1
were obtained by (NH.>zSO. precipitation. This
quantified by electroimmunoassay [15] using the
antiserum was then used to immunize another
monospecific antiserum raised in our laboratory.
rabbit with precipitin lines according to the tech-
SrarlSlicai ana(~·sis. Ali results are expressed as
nique described by Kroll (10]; precipitin lines cor-
means ± S.E Differences in mean values between
1
responding to 100-150 /lg of purified TBPA were
groups were determined by analysis of variance
inoculated according to the method mentioned
and Student's r-test.
above.
Crossed immunoelecrrophoresrs. This technique
1
Results
was performed according to the method of Clarke
and Freeman [11]. Antibodies te who le rat serum
Isolarion and purificarion of rar TBP A
(Sebia. Issy les Moulineaux, France) or to rat
Purification according to the scheme outlined in
1
TBPA were used. The tank and agarose buffer
Table 1 has been repeated several times with simi-
contained Tris (365 mM). 5.5-diethylbarbituric acid
lar resu/ts. The sample of rat TBPA isolated by
(122 mM). calcium lactate (1 mM) and sodium
these procedures had been purified 170-fold. and
azide (10 mM). Prior to use. 1 part of the above
the overall yield was 15%. The first step. including
1
buffer was diluted wlth 4 parts of distilled water
the precipitation with 5% phenol. enriched the rat
(ionic strength. 0.02).
serum with a protein band anodal to albumin.
Auroradiography. Whole rat serum and purified
commonly called prealbumm 11-foid (Fig. lB).
1
TBPA were incubated 3 h at 4°C with trace
The identification of this band as TBPA was made
amounts of ['"'I]T. prior tO being subJected to
by its ability to bmd T•. Whatever the concentra-
crossed immunoelectrophoresis. At the end of the
tion of [Iè' I]T. used. the radioactl\\'tty was essen-
migration the plates were dried and exposed tO an
tially associated with the prealbulTUn band e~cept
1
X-ray
film
eH
Ultrofilm
LKB.
Stockholm.
for a certain amoum of free ['"'lIT. migrating with
Sweden) for 2-3 days according tO the technique
the tracking dye. The identification of this band as
previously used [121.
TBPA was later coniinned throu~h autoradio~ra­
Prorern derermInal/on. Total protein concentra-
phy of [Iè~ I]T.-labeJed whole ra~ serum run- in
1
tion was determmed by the method of Lowry et al.
crossed immunoelectrophoresis ag:unst the mono-
(13]. Th~ high content of tryptophan in TBPA as
sp~cific ami-TBPA serum obtained (Fig. 2A).
compared to bovine serum albumin increased the
Aft~r the first semi-preparatlve dectrophoreSIS
1
TABLE 1
PURIFICATION STEPS OF RAT THYROXl~E·BINDING PREALBUMI:-; (TBPAI
Mau:nalor
TOlal prolem
TBPA"
R~overed •
Yicld
1
procedure
(mgl
(mg,
TBPA
(~)
(%)
Whole serum (75 ml)
S377
31.8
\\00
100
S~ phenol precipilauon
346
21.5
70
70
1
finI seml-preparauve polvacrylamide
gd elec:trophoresis
10.8
11.6
S2
36
Second seml.preparative polyacrylamlde
sel el~lrophorests
4.5 '
4.9
42
15
1
• Values laler obllmed by el~troimmunoasuy.
~ Calcululed ror each Slep or the punfication.
1
, A con\\'ertins raclor wu ~ arler Lowry', us:ay (sec Malenal anu Melhodsl.
1
1

---

1
1
-33-
1
1
273
1
and extraction from the gel. the purity of TBPA
A
B
c
o
was assessed in analytical disc polyacrylamide gel
'.S-
electrophoresis (Fig. le). When a large amoum of
··ir ..:
.'z. i
protein (300 Ilg) was applied to the gel. three
1
<~,J,'
bands couId be detected: a major one (band a.
·r~
identified as the TBPA band) and two faim ones
~~~ ~
(bands band cl. When the phenol supematant was
not electrophoresed immediately but kept frozen
1
for several weeks. faint bands could be seen in the
f3 and y positions (Fig. le. bands dl. In addition.
the imensity of bands b and c was increased. Thus.
c-~~
b
. ~"
1
according to the procedure followed. the TBPA
band varied between 93 and 97~ of the total
proteins when assayed by densitometric scanning.
After the second serni-preparative polyacrylamide
1
gel electrophoresis. only the bands a and c were
jetectable (Fig. ID).
Fig. 1. Analvtlcal disc gel eleclrophoreSIS o( whole raI serum
ImmunizGlion and idenllficGlIon of band c afrer rhe
1
(A). phenol supemalanl (Bl and puniied TBPA aiter the (irsl
(Cl and the second (D) seml-prepar:ltIve polvacrylamlde gel
second semi-preparallVe po(vacrylamlde gel eleclro-
eleclrophoresls. The amounL o( protoln ( Jo' g) applied 10 ~ch get
phoresis
was: J60. 865. JOO and JOO ror the gels A. B. C and D.
The monosç cificity of the an ti-rat TBPA serum
respectlvely.
1
obtained was coniirrned in a crossed immunoe-
1
A
B
.•• 1"-
1
~-! l
- '·a 1
, .
t i
- t \\
. ;
1
1
1
1
....
.,.
1
Fig. l AUIQr:ldiugraphy o( [,:.slfT.·labeied raI ~erum and pun(jed TBPA (202 mg/mil a(ter crosscd immunocle..'lrophoTCSIS, ("111'1
u( rllt serum ( 1) e.,amme<J in 7'T. polyacrylamllJe gel in the fint ùimenslon against IO'ft (Y/v) munospc<:ific antl.TBPA serum in the
inlerme<Jiale gel and 10$ (v/vI antl·...hole raI serum in lhe upper gel. Arrow. TBPA pre..;pitate. (8)11'11)( rat serum 0) and 2 1'1 Il(
TBPA (21 run ln tanÙem in agarosc in lhe (inl dimenSion agamsl 10'" (y/v) anti-TBPA serum.
1
1
1

---

1
1
-34-
1
1
274
lectrophoresis system as shown in Fig. 3. Except
Eleetrophoretic mohi/ity of rat TBPA
for TBPA. the antiserum did neither recognize
In 7% polyacrylamide gel. rat TBPA migrated
1
proteins in whole serum nor in the purified pro-
anodally to albumin as observed in Fig. 2A. How-
tein. The rat serum and the purified protein dis-
ever. in agarose. its electrophoretic mobility was
played a continuous and complete cross-reactivity
similar to that of albumin: the TBPA peak was
1
(Fig. 3). Whole rat serum. eXamined al one in the
located under the albumin peak (Fig. 4). ln order
first dimension in polyacrylamide gel against anti-
to see whether the pattern of migration of rat
whole rat serum with the anti-TBPA serum as an
TBPA in agarose did not result from a calcium
intermediate gel. precipitated as a single peak in
dependent protein-protein interaction. this experi-
1
this latter gel (Fig. 2A). When the purified fraction
ment was carried out with the electrophoresis
and the whole rat serum were analyzed by tandem
buffer containing (Fig. 4). or deprived of. calcium
crossed immunoelectrophoresis followed by auto-
lactate (not shown). Whatever the buffer used.
1
radiography to detect their T binding properties.
similar patterns were observed.
4
TBPA from both fractions (purified preparation
and whole serum) were radiülabeled (Fig. 2B).
Influence of merhods for collecrrng blood on serum
1
This finding provides strong additional evidence
TBPA concentration
that the TBPA isolated by our procedure was
U nder diethyl ether anesthesia. the origin of
highly pure and still biologically active. When rat
blood did not significantly influence TBPA con-
TBPA obtained after the second semi-preparati"e
centration. When blüod was obtained by decapita-
1
eIectrophoresis
was
subjected
to
crossed
im-
tion. from the abdominal aona. from the rail vein
munoelectrophoresis against
anti-TBPA
serum.
using a pülyacrylamide gel in the first dimension. a
1
shoulder which corresponded to the band c was
observed. The shoulder and the peak correspond-
ing to the TBPA band (band a) displayed a com-
1
plete cross-reactivity (not shown). This shoulder
was not detectable wh en whole serum was run in
agarose gel which does not possess sieving proper-
1
ties.
1
t»
ln
1
o
-g0)g
1
1
o
--._. ---I-._--~~-
1
0 0
agarose
1
1 2
Fig. 4. Crossed immunodc:elrophoresis of 1 l'lof raI serum (1)
1
Fig. J. Crossed immunoc:lc:.:lrophorCSIS of 10 l'lof raI serum (1)
ellaminc:d in agarose gel againsl 10% (v/v) anll·TBP..\\ serum in
and 2 l'lof raI TBPA (2) run in tandem against 10$ (v/v)
the intermc:dlate gel (a) and 10't (v/v) ant ..... hole rat serum in
anll·TBPA serum.
the upper gel. Arrow. TBPA precipitale.
1
1

---

1
1
-35-
1
1
275
1
TABLE Il
but significantly, by 13% (P < 0.05). Subsequently.
SERUM TBPA CONCENTRATION
DURING
A 24 h
levels returned to their initial values by 21 :00 h
CYCLE
and remained fairly COnstant for the rest of the
1
45 5-month-old male rats with a mean welghl of 415 g were
24-h period (Table II).
divided in ni ne groups of 5. The rats were decapltated at 3 h
inlervals for a 24 h period beglnnlng al 09:00 h. The data are
Age- and sex-relared serum TBPA concenrrallon
presented as means ± S. E.
From 1 to 28 days of age. the TBPA concentra-
1
tion increased progressively and sigmficantly in
Hour
TBPA lmg;100 ml)
both sexes. For each age. the concentration was
09:00
40.2 ± 2.17
comparable between males and females. In male
12:00
36.7 ± 1.40
1
rats. the TBPA level recorded at 10 weeks of age
15:00
34.8 ± 1.87 ".b
18:00
35.7 ±0.79
was not significantly different from that obtained
21 :00
40.2:: 1.34
in
the 28-day-old pups.
In contrast. in adult
24:00
40.1 = 2.37
females. the TBPA concentration dropped signifi-
1
03:00
38.2 ± 1.83
cantly and reached values comparable to those
06:00
37.8;!: 1.20
originally found in the I-day-old pups (Table [Il).
09:00
38.0 = 2.70
When adult male rats were fed a commercial
1
• P < 0.05 vs. 09:00 (first group 1.
stock diet. the value determined for serum TBPA
b
P < 0.05 vs. 21 :00.
was: 39.2 ± 1.50 mg/lOO ml (II = 19). ThIS con·
centratIon was signlfil:antlv Iowa [han lha! meas·
or from the retro-orbital plexus. TBPA levels were
ured in adult male rats fed the diet prepared in the
1
39.9 ± 2.76. 40.4 ± 2.50. 39.3 ± 2.77 and 43.7 ±
laboratory.
2.53
mg/100
ml
respectIvely.
Diethyl
ether
anesthesia induced a slight decrease in TBP.-\\ levels
Discussion
1
without any significant difference occurring be-
tween the two groups of decapitated rats (39.9 ::
The method for purifil::llion ùf TBPA I\\hich
2.76 versus 45.2 ± 3.21 mg/100 ml).
has been developed presents the :.lCivantage of pro-
Iiding highly purified rat TBP.\\
1
ln
lhree sleps.
Saum TBPA concellCrarlOns durrng a :.+ il (Tele
usmg only lWO simple lechnlques. In aJdillùn. lhls
The overall variance between groups was not
procedure can easdy be reproduced in any labora-
statistically significant. However. From 09:00 h lO
tory \\\\.llhoUl sophislicaled equlpmenl. Differenliai
1
15:00 h. the TBPA concentration declined slightly.
precipilatIOn of serum IVllh phenol has been used
1
TABLE III
SERUM TB PA CONCENTRATION IN MALE AND FEMALE R.·\\TS AT VARIOUS AGES
1
The animais were bred ln our labor;J.torv. Pups "'ere "'.:aned when 28 davs old on the diet prevlouslv fed 10 lhe.r mOlhers. Blood was
.:olleclcJ either by decapltatlon (l·day·oIJ pupsl or from the relro-orbltal ple.,us. Means ln the same .;oiumn "'lIh dlfferent
super5<:nplS are siglllfi.:antly different (p < 0.01l. • Signlfi.:anllv hlgher ln males than in females (p < 0.01)
1
Age
Males
Females
n
body
TB PA
n
body
TBPA
1
we.ghl
(mg/lOO mil
weighl
(mylOO mil
(g)
(g)
1 day
18
6.3 ±0.13 •
26.3 ±0.9 "
III
5.9±0.13 •
:Id ±0.82·
15 days
12
241 =0.80 b
36.4 ± 1.11 h
\\0
23.9 ± 1.06 h
36.4 ±0.59 h
1
28 days
8
76.0 ± 2.40 •
55.9± 3.93'
12
75.0 ± 1.39 •
J'l.5 ± 1.94'
10 wecks
8
3585 ± 5.88 ... •
56.8 ± 0.45 •.•
12
212.3 ± 2.59 J
:9.9±1.31"
1
1

---

1
1
-36-
1
1
1
276
by some investigators to purify human TBPA
shows no TB PA labeling, suggesting that the puri-
[16-191. Our work indicates that this method can
fied TBPA preparation did not contain any detect-
1
also be applied to rat serum. Previous work has
able amount of retinol-binding protein. Thus. our
shown that T. is bound directly to TBPA [20J.
method for the isolation and purification of rat
Since the T. binding SIte on the TBPA molecule is
TBPA seems panicularly suitable 10 remove reti-
located in a cylindric channel that runs through
nol-binding protein. In Contrast. the radioacuve
1
the center of the protein molecule. we introduced
retinol bound to the whole serum. suggesting that
[I!~I]T. as a tracer in each step of lhe purification
dunng quantitative determination of lhe TBPA
to control the preservation of the native structure
contained in the serum. the anti·TBPA serum also
1
of the TB PA. The mobility of the protein and its
precipitates
the
TBPA-retinol-binding
protem
T.-binding capacity were used as criteria of identi·
complex.
fication of the punfied TBPA.
Rat TBPA isolated by the three simple steps
An analytical polyacrylamide gel electrophore·
descnbed in this paper was of high pumy and
1
sis was performed with an excess of protem to
appeared to be homogeneous according to electro-
provide maximal opponunlty for the detectlon of
phoresis in several systems. In addition. the T.-
impurities. In additIon to bands a (the TBPA
binding abiiity of the puniied prOlem strongly
1
band), the pattern obtamed after the final st,:p
suggests that its natIve structure and biological
showed a faint band (band cl. which bound T.
aCllvity were preserved.
and displaved a compi<:te cross-reactivity with the
Although rat TBPA shares many physlcal char·
TBPA band. This iinding \\S m accordance wllh the
aClenSllCS with its human counterpan [3A]. the
1
result of Branch et al. [21 J. L~cureuil et al. [191 and
study shows lhat it does nOl have lhe ~apld electro-
Socolow et al. [221 for human TBPA: they as-
phoretlc mobility of human TBPA especlJ.llv ln an
sumed that thls band c was a TBPA aggregate.
agarose medium: yet. Il mlgr:ltes disunctlv anod-
This hypothesls is aIse supponed by the f~ct -that
allv to alburrun in polyacrylamlde gel. These ~wo
1
band c increased ""hen [he procedure of puniica-
e1eclropnOretlc behavlOrs mlght be e.~plalned on
tlon included eplsodes of storage betwe~n any of
the basls oi differenuai e1ectnc charges oi bOlh
the different steps. Accordmgly. the percent age oi
proteln [3] .md/or on account oi vanations be-
1
puriiied TBPA (band a - band cl m dlfferent pre·
tween thelr ammo aCld seauence [4). The albumm-
parations amounted mmlmaily to 96";, ailer the
lik~ mobdit\\' of rat TBP-\\ ln agarose gel elec:ro-
first serru·preparatlve polyacrylamld~ gel declro-
phoresls must be tak~n lnto account ra aVOId anv
phoresls and ~xceed~d 99~ after the second ùne.
coniUSlOn as 10 the ldentltv of this prOleln [::6].
1
This result means that a satlsivmg degree of pUrifi-
Blood for rat TBPA me.::.surement mav be drawn
catIon can be obtamed ln only two steps. ..1,.ny
by various melhods wllhoUl any eife::: ùn TBP.-\\
attempl 10 idenufy bands band d \\\\ias unsuccess-
concentratlon. Like\\\\,se. diethyl el he, aneslhesia
fuI.
has no dfect on i ts lev~1.
1
In the plasma. part ùf the TBPA molo::cules
The slud\\' mdicates that TBPA is presenl ln the
circulate in
lhe
form
of
a
ternary
compk'l.
serum of 1-day-old rats regardl~ss ai sex. Dunng
(retinol-retmol-binding prolem- TBPA). Tlus com-
the suckling period. the graduai incre.::.se of serum
1
plex dissociates at low iomc strength [231 and
TBPA concentration is age- but not sex-depend·
during polyacrylamlde gel dectrophoresis [241. Rat
ent. at least until weantng (28 days). The slgntfi-
retinol-binding protein has an al mobility [25] and
cant fall in this protem. which occurs in female
no band migr:lting in lhis area was observed in
r:lts after we:1nÎng. might be due 10 hormonal
1
analytical disc elo::ctrophoresis aiter our punfica-
changes during this penod. Indeed. sorne aUlhors
tion procedure. In addition. a rocket immunoelec-
note that for humans. androgenic produCllon in
trophoresis of punfied TBPA as weil as whole
adolescent boys and male adults appears tO be
directly responsibJe for signific:mtly higher serum
1
serum incubated with [-' H]retinol (7200 cpm/l..d of
each sample) \\lias performed using our mono-
TBPA concentrations as compared to age-matched
specific anti-TBPA serum (10%. vIv). The result
girls and women [27]. However. the effo::ct of
obtained by autoradiography (unreported data)
estrogens on the TaPA level is more controverslal
[28.291·
1
1

---

1
1
-37-
1
1
1
277
Despite its short half-life. the circulating level
9 Davis. B.J. (1964) Ann. N.Y. Acad. Sei. 121. 404-427
1
of rat TBPA in adult animais was fairly constant
10 Kroll. J. (\\981) Methods Enzymol. B. 52-57
during' the 24 h cycle. Yct. a slight decrease that
Il Clarke. M.H.G. and Freeman. T. (1968) ain. Sei. 3S.
403-413
exceeded approx. 10% of the initial value was
12 EgloU. M.. Vranclut. R.. Tardivel·Lacombe. J. and Degrelle.
recordcd between 09:00 h and 21:00 h. This drop
1
H. (\\981) Steroids 37.455-462
must be taken into account in the timing of sam-
lJ Lowry. O.Hoo Rosebrough. N.Joo Fart'. A.L. and Randall.
pling. as small changes in the concentration of this
RJ. (1951) J. Biol. Chem. 193.265-275
protein arc to be expected.
14 Oppenheimer. J.Hoo Surks. M.I.. Smith. J.c. and SqueL R.
1
(\\965) J. Biol. Chem. 240.173-180
IS uurell. C.B. (\\9661 Anal. Biochem. 15. 21-36
Acknowledgement
16 Got. R. and Boumllon. R. (\\963) EJtpenentla 19. 411-49
17 De Nayer. P.• Van Den Schrieck. H.Goo Koch. Moo De
The study was supported by a gram from IN-
Visscher. M. (\\966) Biochim. Biophys. Acta 124. 411-4\\2
1
sERM. PRe No. 128020.
18 Tritsch. G.L. (\\ 972) J. Med. 3. 129-145
19 Lecureuil. Moo Leeureu". N. and CrouLlt·Reynes. G. CI 978)
Biochim. Biophys. Acta 532. 268-278
References
20 Blake. C.C.F. and Oatley. SJ. (\\977) Sature 268. 115-ln
1
21 Branch. W.T.. Robbins. 1.. Jr. and Edelhoch. H. 11971) J.
Ingenbleek. Y. (1982) in Marker PrOle,"s .n In/lammauon
Biol. Chem. 246. 6011-6018
(Allen. R.C.. Bienvenu. J.. Laurent. P. and Susk,nd. R.M ..
22 So<:olow. E.L.. Woeber. KJ.A .. Purci\\'. R.H .. Hollowav.
eds.l. pp. 405-414. Walter de GruYler & Co.. Berlin
MT and Ingbar. S.H. (1965) J. Clin. In'est. 44.1600-1609
1
2 Shettv. P.S .. WatraslewlCz. K.E. Jung. R.T.. James. W.P.T.
23 Peterson. P.A. (1971) J. Biol. Chem. :~. 34-43
(1979) Lancet Il. 230-232
24 Muto. Y. and Goodman. O.S. (197:) J. Biol. Chem. 247.
Pelerson.
P.......
Rask.
L.. 6stberg. L.. Andersson. L..
2533-2541
KOImwcndo. F. and Pertoft. H. (1973) J. B.ol. Chem. 248.
25 KOinal. M .. Raz. ..... and Goodman. O.S. (1968) J. Clin.
1
4009-~2::
1nvest. 47. 2025-:044
4 Navab. M.. Mailla. A.K .. Kand:l. Y. and Goodman. O.S.
26 Vranckx. R.. Ble,cerg. F.. Bena.ssavag. c.. Savu. L.. '.fart,".
(1917) J. Biol. Chem. 232. 5100-5106
M.E.. Zouaglu. H .. Wade. S. .1nd Sunez. E.A. (1983) Ab·
DaVIS. P.L Spauidlng. S. W. and Gr~erman. R.1. (\\ 9701
stra't PS·)·13. 4th International Cùngress of Electropnore·
1
Endocnnology 87. 978-986
SIS and Related T echnaC!ues. Tokvo. \\fJv 9- \\ 2. 198)
6 Sutherland. R.L and Brandon. M.R. 119761 Endocrlnology
27 lngenbleek. Y.. Van Den Hove. \\f.H. and Oerueile. '.t.
98. 91-98
(\\981) Clin. Ch,m..-\\cta 114. 219-::~
7 Oi'kson. P.\\V . Howlett. G.J. and Schrelcer. G. (1982) Eur.
28 Laurell. C.B.. Kullander. S. and Thore:!. J. (\\96;\\ Scand. J.
J. Biochem. 129. 239-29)
Clin. Llb. (nvest. 21. 337-)43
1
8 Peterson. P ...... :-liisson. S.F.. Ostberg. L.. R;uk. L. and
29 VahlqulSt. A.. Johnsson. A. and S~gre~. K.G. (!9791 ....m. J.
Vahlqulst. A. (197-1) Vitam. Horm. 32. 181-214
Clin. Nutr. 32. 14))-1438
1
1
1
1
[
1
1

---

•
-38-
1
1
1
1
1
ART 1 C L E N° 2
1
1
1
"Value of serum transthyretin level in the assessment of marginal
protein energy malnutrition. (soumis à Clinical Chemistry)."
1
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-39-
1
1
VALUE OF SERUM TRANSTHYRETIN LEVEL IN THE ASSESSMENT OF
1
MARGINAL PROTEIN-ENERGY MALNUTRITION.
1
Salimata WADE *, Fanny BLEIBERG-DANIEL, Béatrice LE MOULLAC,
1
2
Damien IYAKAREMYE, Daniel BIOUl, Francis GAUTHIER
and
1
Daniel LEMONNIER.
1
1
1
INSERM U1 Unité de Recherches sur la Nutrition et
1
l'Alimentation Hôpital Bichat,
1'70 bd Ney 75877 PARIS CEDEX
18,
1
1Laboratoire de Biochimie ERA CNRS 396,
1
92290 Châtenay-Malabry, and 2Laboratoire de Biochimie, Tours
(France).
1
1
1
1
*Correspondance should be addressed to Dr. WADE S. Unité de
1
Recherches sur la Nutrition et l'Alimentation, Hôpital
Bichat,
170 bd Ney 75877 PARIS CEDEX 18 (France).
1
1

---

1- . '!'
.,
'.
1
-40-
1
1
Abstract
1
The influence of a wide range of protein and/or energy
1
intakes on the serum level on rat transthyretin was studied.
1
Young and adult rats were fed ad libitum diets containing
18 %,
9 %, 6 %,
4 % and 0,5 % protein (w/w), or fed a
1
control diet in restricted amounts.
The transthyretin level
was reduced in young rats and was normal or slightly
1
enhanced in adult rats fed low prote in diets.
It was
1
progressively decreased in ail energy restricted rats.
Moderate energy restriction in rats fed equivalent amount of
1
protein also lowered its level. Rats with similar body
weight and similar protein intake showed marked differences
1
in serum transthyretin level depending on the amount of
energy consumed.
Serum transthyretin changes were discussed
1
in relation to the level of transthyretin in cerebrospinal
1
fluid and to the serum concentration of albumin, transferrin
and thyroid hormones.
The results show that serum
1
transthyretin is more closely related to the protein and
energy intakes than to the protein and energy content of the
1
diet. Our results suggest that serum transthyretin
1
measurement is a reliable Marker in the detection of
moderate and severe protein-energy restriction.
1
Additional Keyphrases : Rat Serum and cerebrospinal
1
fluid transthyretin. Albumin.
Transferrin. Total and free
1
thyroxine. Total and free triiodothyronine.
Protein and
energy restriction.
1

---

1
1
-41-
1
1
Introduction
1
Reliable methods for the assessment of marginal protein
1
energy-malnutrition (PEM) are needed but difficult to obtain
because of the imprecise boundaries between nonlal and
1
pathological states,
of the complex interactions among
1
dietary inadequacy, diseases and environment. A number of
biochemical parameters have been investigated in the hope
1
to show deterioration before any clinical sign of
malnutrition is detectable (1).
In man,
studies in
1
developing countries (2-9) as well as in developped
1
countries (10-14) have suggested the use of plasma
thyroxine-binding prealbumin (TBPA),
recently named
1
transthyretin (15),
to monitore mild to moderate PEM.
Human as weIl as rat TBPA is a protein with a dual
1
transporting function (16).
ft is a carrier protein for
1
thyroxine and an indirect vehicle for vitamin A by the
binding of retinol-binding protein (see ref.in 16). TBPA was
1
first observed in cerebrospinal fluid (CSF) by KABAT et al.
(17) and recent data indicate,
as earlier presumed, that
1
transthyretin mRNA was mainly found in the choroïd plexus
1
and in the liver (18-21).
1
1
1
1

---

1
l
1
-42-
1
1
Plasma (or serum) TBPA was found to decrease in human PEM
1
(2-9),
in obese restricted patients (10-12),
in inflammatory
1
reactions (22-26) and in liver diseases (27).
In
malnutrition, plasma TBPA is markedly reduced and changes
1
towards normal on repletion more rapidly than other visceral
proteins (3,4,9,10,14). But, obviously, when one measures
1
the plasma level in malnourished subjects often with
synergic infection or surgical complication,
i t is difficult
1
to determine which factor is most important in causing the
1
observed decrease.
Bence, the understanding of the
nutritional influences responsible for the lowered level of
1
plasma TBPA is important and needs to be fully defined. Such
studies can be more easily carried out in an animal model
1
than in man.
The rat was chosen in view of the similarities
1
in molecular structure,
amino acid composition,
physical
properties and biological functions between the rat and the
1
human TBPA (28,29). This high degree of homology was
recently confirmed by molecular cloning (30).
1
We undertook several studies to investigate the response of
1
serum TBPA to changes in various protein and/or energy
intakes.
In order to determine if nutritional factors also
1
regulate the TBPA concentration in CSF, we measured the
level of TBPA in CSF and compared i t to the level in serum
1
Abnormalities associated with decreased TBPA was also
1
assessed by the measurement of total and free fractions of
circulating thyroid hormones.
1
1·

---

1
1
-43-
1
1
In MOSt of our experiments, serum TBPA was determined
1
simultaneously with other nutritional markers (albumin,
transferrin) and with some infectious and/or inflammatory
1
markers (acute phase reactant proteins), namely orosomucoid
1
and a -macrog lobulin.
2
1
Matérials and Methods
1
1
AnimaIs and diets.
1
Male rats of the Wistar strain were used (lffa-Credo,
Arbresle, France). The animaIs were housed individually and
1
maintained at 22 ± 2 oc on a 12 h light/12 h dark cycle.
1
Water drinking was provided ad libitum. AlI rats were
previously fed a control diet for at least one week before
1
they were randomly assigned to experimental groups. The
study was carried out in several parts.
1
In ~xperiment 1, 4 week-old rats (mean weight about
1
80g) were divided into five groups of 6 and fed ad libitum
1
diets with varying protein/energy ratio
18 % protein
(Control group),
9 % protein (9 % group), 6 % prote in ( 6 %
1
group),
4 % protein (4 % group) and 0.5 % protein (0.5 % group). The
1
diets were isocaloric and the energy value of each was 3.1
1
1

---

1
1
-44-
1
1
kcal/g.
The composition is shown in Table 1.
1
Body weight and food consumption of the five groups were
recorded every 24 or 48 h.
Four groups of age-matched rats
1
were pair-fed to the 9 %,
6 %,
4 % and 0.5 % groups;
1
accordingly these groups were called 9 % PF,
6 % PF, 4 % PF
and 0.5 % PF.
The pair-fed rats received the control diet
1
(18 % protein) in amount equivalent to that the protein
deficient groups had consumed the preceding 24 or 48 h.
1
Pair-feeding was started when the food intake of the protein
1
deficient rats was decreased. Pair-fed animais were given
their food as a single meal between 1600 and 1700 hours to
1
avoid meal-feeding effects. The animais were fed for a
period of 6 weeks.
Blood was drawn from the retro-orbital
1
plexus by day 0, week 1, week 2 and week 4.
At the end of
the experiment, the rats were anesthetized with diethyl
1
ether and blood collected from the abdominal aorta.
Clear
1
cerebrospinal fluid (CSF),
free of blood contamination was
obtained by cisternal puncture (31). Serum and CSF were kept
1
at -
20 oC until analysed.
1
In ~x~~rim~nt Z, four groups of 8 adult rats (about
1
250g) were fed ad libitum the control diet (18 % protein),
and low protein diets containing 9 %, 6 % and 4 % protein
1
(Table 1). The rats were fed during 8 weeks.
Food
consumption was measured after 4 weeks of dietary treatment.
1
The rats were bled by day 0, week 2, week 4 and week 8.
In
1
1

---

1
1
-45-
1
1
an other protocol,
four groups of 8 adult rats (about 250g)
1
were formed.
The first group was fed ad libitum the control
diet.
The remaining groups received daily the control diet
1
in amount eaten by the adults rats fed 9 %,
6 % and 4 %
1
prote in diets
18g, 16g and 11g, respectively
these
animais were called "adult restricted rats".
Their food was
1
given as a single Meal between 1600 and 1700 hours.
They
were bled by day 0, week 2, week 4 and week 8.
1
1
ln ~~~~im~nt 3, the effect of meal-feeding, if any, on
the serum TBPA level of energy restricted rats was
1
investigated. We developped an automated feeder system
capable of delivering variable quantities of ground diet at
1
predetermined time intervals.
The system involves feeder
units which are easily placed on the coyer of an animal's
1
cage. Each feeder is connected to a control unit which
1
regulates the number of feeding installments. The control
unit is connected to a computer from which the number and
1
quantity of food delivered and the times of feeding can he
selected. A group of 8 adult rats was fed for 21 days
1
restricted amounts of the control diet (18 % protein,
1
Table 1) using this automated food dispenser.
The rats
received 80 % of the intake of age and weight-matched
1
control rats.
The feed deliveries were divided into one
feeding at 1600 hours and 11 hourly feeding during the night
1
(1900 to 0500 hours). A feeder unit connected to the control
1
1

---

,
1
1
-46-
1
1
unit and placed above an empty cage was used to control the
1
amount of food delivered per 24 h.
The rats were bled by day
0, day 8 and day 21.
1
1
In ~XPeriment 1, 6 week-old rats (about 180g) were
randomly divided into 4 groups.
The first group (9 rats) was
1
killed at this stage for base line values (control day 0).
The second group of 10 rats was fed ad libitum a diet
1
containing 9 % prote in (Table 1). This group was called
1
100 % energy fed (100 % EF). The third and fourth groups (12
animaIs per group) were energy but not prote in restricted
1
they received 90 % and 80 % of the intake of the 100 % EF
group,
respectively, but the same amount of protein.
This
1
was achieved by making isocaloric diets with higher
1
protein/energy ratio than the 9 % protein diet.
The 90 % and
80 % energy restricted groups (90 % EF and 80 % EF) were fed
1
diets containing 10 and 11.2 % protein,
respectively. After
11 days of dietary treatment, half of the animaIs in the
1
90 % EF and 80 % EF groups were killed.
The remaining rats
1
in both groups were refed ad libitum with the 9 % protein
diet.Twelve days after refeeding, aIl the animaIs (100 % EF,
1
90 % EF,
80 % EF groups) were killed under ether anesthesia
and blood drawn from the abdominal aorta.
1
[
In ~x~er1m~nt ~, adult rats (about 340g) were randomly
divided into 2 groups of 1 animaIs each. The first group
1
1

---

1
1
-47-
1
1
(control group) was fed a control diet containing 18 %
1
prote in and 3.7 kcal/g of diet (table 1). The second group
was fed a high calorie (HC) diet containing 18 % prote in and
1
4.76 kcal/g of diet. This diet was made by increasing the
1
corn-oil content of the control diet and decreasing the
amount of sucrose. Rats fed the HC diet were called HC
1
group. Rats were fed during a 21-day period. Body weights
and food intakes were recorded every 24 h or 48 h. The
1
animaIs were bled by day 0, day 8 and day 21.
1
Protein and thyroid hormone measurements.
1
Total thyroxine (total T4),
free thyroxine (free T4),
total
1
triiodothyronine (total T3) and free triiodothyronine (free
1
T3) were determined in serum by RIA (Kit Amersham,
France).
Purified rat transthyretin (TBPA) and a monospecific
1
antiserum against rat TBPA were made in our laboratory
according to previously published procedures (16). Serum and
1
cerebrospinal fluid (CSF) TBPA levels,
serum albumin and
1
a -macrog lobulin concentrations were measured by
2
electroimmunoassay (32) while transferrin and orosomucoid
1
concentrations were determined by radial immunodiffusion
(33). Rat transferrin and albumin as weIl as anti-rat
1
albumin and anti-rat transferrin were supplied by Cappel
Laboratories (West Chester, USA).
For the measurements
1
of orosomucoid and a -macrog lobulin,
serum obtained from
2
1
1

---

1
-48-
1
1
1
rats submitted to inflammation with turpentine oil was used
as "internaI standard".
1
Statistical analysis.
1
AlI results are expressed as means ± SEM. Differences in
1
mean values between groups were determined by analysis of
1
variance
(ANOVA)
and the paired Student·s t-test when only
two time points were compared.
1
Results.
1
1
The absence of an eventual
inflammatory reaction was
examined in rats fed low protein diets and in energy
1
restricted rats by the measurement of two acute-phase
reactant proteins.
Despite variations in the serum level of
1
orosomucoid and 1~2-macroglobulin between groups,
the values
1
obtained for both proteins were normal
and
markedly lower
than those expected in case of inflammation
(not shown).
1
Effects of feeding low protein diets ad libitum.
1
1
In the 4 week-old rats,
growth was significantly
reduced.
Table 2 shows that the lower the protein/energy
1
ratio of the diet,
the poorer the weight gain.
In the 0.5 'l.
group,
protein restriction was severe enough to stop growth
1
entirely.
1
1

---

1
1
-49-
1
1
Table 2 also shows the protein and energy intake per 100g
1
body weight of the rats during the 6 week-feeding period.
Compared to the control group, aIl the animaIs fed ad
1
libitum, except the 0.5 % group at week 1 and week 2,
1
consumed an excess of energy expressed per unit of body
weight.
1
The Mean ( ± SEM ) food intakes ( g/24h ) of the adult rats
fed ad libitum with the 18 %,
9 %,
6 % and 4 % protein diets
1
were 22.2 ± 0.85,
18.3 ± 0.97, 15.7 ± 0.84 and
1
11.3 ± 0.65, respectively. The body weight was also
significantly decreased in the adult rats fed low prote in
1
diets (not shown). Figure 1 illustrates the variations in
serum TBPA concentration of the young rats.
The serum TBPA
1
level of the 9 % group was comparable to that of the control
group throughout the study. The TBPA concentration decreased
1
significantly in the 6 % and 4 % groups by week 1 but
1
increased towards the control value from week 2 to week 6 ;
however the differences were still significant by week 2 and
1
4 but not by week 6.
In the 0.5 % group, the TBPA
concentration was markedly reduced from week 1 to week 6. By
1
contrast, the adult rats fed low protein diets exhibited a
1
normal or a slightly increased TBPA level thXoughout the
study (not shown).
1
Table 3 shows the effect of the 6 week-dietary treatment in
young rats on the concentration of TBPA in CSF and on the
1
serum albumin and transferrin level. Except in the Most
1
1

---

1
1
-50-
1
1
severely restricted animaIs (0.5 % group), the concentration
1
of TBPA in CSF was unchanged.
The albumin and transferrin
level was significantly decreased in the 4 % and 0.5 %
1
groups.
1
1
Effects of feeding a control diet in restricted amounts.
1
The body weight of the pair-fed rats was in between
1
that of the control rats and the corresponding animaIs fed
ad libitum low prote in diets. Their energy intake per 100 g
1
body weight was significantly lower than that of the
corresponding ad libitum fed animaIs (table 2). One week
1
after pair-feeding, the restricted rats exhibited
significantly decreased levels of TBPA (Fig.
1) which was
1
more pronounced in the 6 % PF,
4 % PF and 0.5 % PF groups
1
than in the 9 % PF group by week 4. After six weeks of
dietary treatment, the reduction was comparable in the 6 %
1
PF, 4 % PF and 9 % PF groups;
i t was more severe in the
0.5 % PF group (Fig.
1). Neither serum albumin level nor
1
TBPA in CSF was affected by the reduétion in energy
1
consumption.
Transferrin was low only in the 0.5 % PF group
(table 3).
1
The serum TBPA concentration in the adult rats fed for eight
weeks with about 80 %,
70 % and 50 % of the intake of the
1
control rats was also markedly decreased by week 2 (20 %,
1
1

---

1
1
-51-
1
1
24 'l. and 36 'l. drop,
respectively),
by week 4
(20 %,
20 % and
1
36 'l. drop,
respectively),
and by week 8
( 24 %,
35 % and
50 'l. drop,
respectively).
The more severe the restriction
1
was,
the lower the serum TBPA level.
An eventual effect of meal-feeding in energy restricted
1
rats was tested in a group of animaIs fed 80 'l. of the intake
1
of age and weight-matched control rats.
This was achieved by
using an automated feeder to mimic the usual meal pattern of
1
rats.
Delivering the food in several meals or as a single
meal did not prevent the decrease in the serum TBPA level,
1
previously observed in the 80 % energy restricted rats
1
(table 4).
The reduction was more severe after twenty one
(about 22 % drop)
than after eight days of restriction.
As
1
previously observed in restricted animaIs given their food
as a single meal,
the serum albumin and transferrin levels
1
were not modified at any time of the restriction
(table 4).
1
Effects of energy restriction and refeeding in rats fed
1
equivalent amount of protein.
1
Body weight was slightly but significantly reduced only
in the 80 % energy fed
(80 % EF)
group.
Furthermore,
the
1
body weight was regained after refeeding
(table 5).
The
1
100 'l. energy fed group
(100 % EF)
had received a 9 %
protein diet ad libitum.
As in experiment 1 and 2,
their
1
TBPA level was not affected.
Reducing the energy intake of
1
1

---

1
1
-52-
1
1
the animaIs,
whose diets provided equivalent amounts of
protein,
lowered the serum TBPA concentration :
after eleven
1
days of restriction the level
was significantly decreased in
1
the 90 % EF group as weIl
as in the 80 % EF group.
Albumin
and transferrin were not affected.
Twelve days of refeeding
1
increased the TBPA value towards the value observed in the
100 % EF group
(table 5).
1
In this experiment,
the influence of a decreased level
1
of TBPA on the thyroid hormones was tested.
The total and
free fraction of thyroxine
(T4)
and triiodothyronine
(T3)
1
were measured
(table 6).
Feeding a 9 % protein diet ad
libitum induced modifications in the amount of thyroid
1
hormones:
total
and free T4 were significantly lower in the
1
100 % EF group than in the control group.
In the 100 % EF group,
as weIl
as in the 90 % and 80 % EF
1
groups,
there was an increase in the free T3 level
while
the total
T3 amount was unchanged.
The 80 % EF and 90 % EF
1
groups exhibited values of free T3 significantly higher than
1
those of the 100 % EF group.
Total
T4 decreased markedly in
the energy restricted rats
(90 % EF and 80 % EF)
and
1
increased towards the value of the 100 % EF group after
refeeding.
The free part of this hormone was not influenced
1
by the energy restriction.
The observed decreases in total
1
T4 paralled those of TBPA.
1
1
1

---

1
1
-53-
1
1
1
Effect of feeding a high calorie diet.
1
The adult rats fed a diet with an energy content of
4.7 kcal/g decreased spontaneonly their daily food intake
1
(table 7)
but not their daily energy consumption.
Their body
1
weight as weIl as their serum TBPA level were comparable to
those of the control rats fed the control diet with less
1
calorie content.
1
Discussion
1
The major dietary factors
(excluding vitamins and
1
mineraIs)
that influence health and growth are the total
energy intake~ the concentration and quality of protein in
1
the diet,
and the efficiency with which theses factors are
utilized by the body.
Protein-energy malnutrition
(PEN)
is
1
the expression of deficiency in these major dietary factors.
1
In man,
no single method currently available adequately
assesses PEM.
Various tests are used including the
·1
measurement of certain visceral
proteins namely albumin,
transferrin,
transthyretin
(TBPA),
and retinol-binding
1
protein
(1).
Among the plasma proteins,
albumin is the most
1
extensively studied but i t does not respond to moderate and
short term changes in protein and energy intake.
TBPA has
1
been proposed as the most specifie and sensitive marker of
PEM
(2).
However studies in man give no indication of the
1
specifie influence of the major dietary factors on serum
1

---

1
1
-54-
1
1
TBPA and the response to various degrees of prote in and
1
energy restrictions has not been studied so far.
In Many developping contries,
in addition to the recognized
1
malnourished children with low TBPA level, there are large
numbers of children, who,
although apparently healthy, have
1
an abnormal level of serum TBPA (5-8),
significantly less
1
than the reported normal values for age (34,35).
Such
children are often used as control subjets in clinical
1
studies (6,7,9).
The decreased TBPA could be explained by
concomitant infections, but our own studies indicate that
1
most of these children were not infected (WADE et al.
1
unpublished data).In addition genetic variation has not been
reported for TBPA (3).
Such children might suffer from mild
1
to moderate malnutrition.
To gain some insight into the specifie influence of protein
1
and/or energy intakes on the concentration of TBPA, we have
1
turned to the animal model. Various degrees of protein
and/or energy restriction were induced in rats and
1
variations in the serum concentration of TBPA was recorded
at different times of the dietary treatments.
1
The effects of actual nutritional restriction from those of
1
disease or inflammatory processes were distinguished by the
measurement of orosomucoid and a -macrog lobulin, two main
2
1
acute-phase reactant proteins in rat (36,37). The data
showed that low TBPA levels reported in this study were only
1
related to the effect of nutritional deficiency per se and
1
not to an acute phase response.
1

---

1
1
-55-
1
1
It was apparent from the overall results that adaptative
1
changes in the serum TBPA level occured during protein
and/or energy restriction.
The concentration of serum TBPA
1
was greatly influenced by the diet, but undoubtly the data
need to be analysed in terms of protein and energy
1
requirements and utilization. The ability of the rat to
1
regulate the food intake in relation to its own energy
requirement has been weIl recognized.
In 1978, the National
1
Academy of Sciences reported the requirements of the
laboratory rat to achieve growth,
gestation and lactation,
1
to be 14 to 15 % dietary protein concentration (casein
1
supplemented with DL methionine) of a diet containing 5 %
fat,
little fibrous material and about 4 Kcal/g (38).
1
Protein requirements decline with age and the needs for
maintenance in the adult rat are much more less (38,39).
1
Lowering the recommanded dietary allowances does not
necessarily imply deficiency but increase the probability of
1
setting a state of deficiency.
For these reasons, the
1
protein content of our diets were decreased from 18 % to
9 %,
6 %,
4 % and 0.5 % (W/W).
1
Rats fed the nutritionally adequate diet (control diet,
18 %
protein) achieved an appropriate balance of protein and
1
energy intakes for maintenance and normal growth. Rats fed a
1
higher calorie diet containing an equivalent amount of
prote in , vitamins and mineraIs as the control diet,
1
decreased their food intake but consumed an equivalent
1
1

---

,
·1
1
-56-
1
1
amount of energy.
This agrees with earlier reports
1
indicating that the daily calorie intake remained constant
when the diet contained from 0 to 30 % fat (40). Rats fed
1
the high calorie diet also achieved an appropriate balance
of prote in and energy intakes for maintenance and normal
1
growth.
In both groups serum TBPA level was normal.
1
When low protein diets were fed ad libitum,
the rats
quickly increased their food intake. This happened during
1
the first two week exposure to the diets in all groups
except the 0.5 % group.
Nevertheless, growth was reduced but
1
not stopped except in the 0.5 % group.
In accordance with
1
previous findings (41,42), although an aversion to the diet
developped with prolonged exposure,
energy intake per g body
1
weight was still higher in the protein restricted than in
the control rats.
In their recent protein and energy balance
1
study,
Lunn and Austin (42) reported that in rats fed ad
1
libitum low protein diets,
dietary energy was present in
excess of the limited amount of growth allowed by the low
1
prote in intake. Our animaIs fed 9 % prote in diet ad libitum
have succeeded to ajust their protein needs by increasing
1
their daily food intake.
Their body weight was slightly Iess
1
than that of the control group but their weight gain during
the two first weeks of dietary treatment was comparable.
The
1
9 % group exhibited values of serum TBPA similar or slightly
higher than the control group.
1
1
1

---

1
1
-57-
1
1
Rats fed the 9 % protein diet ad libitum were compared
1
to rats fed an equivalent amount of protein, but energy
intake moderately lowered by restriction.
The serum TBPA
1
concentration was markedly reduced in the restricted rats.
The decreased TBPA level in the restricted rats was not an
1
effect of meal-feeding. The normal values were restored
1
after refeeding and the body weight was also improved.
Changes in the nutritional state of the animal was more
1
rapidly detected by the measurement of TBPA than by that of
body weight.
Neither albumin nor transferrin were affected.
1
The data indicate that when the protein intake is below the
1
requirements, an increase in the energy intake May improve
the nutritional state.
A decreasing energy intake leads to
1
the worsening of this state, primarily reflected in the
concentration of serum TBPA.
The data also suggest that
1
protein cannot be utilized efficiently if energy is lacking.
1
These data also allowed us to understand better the
results obtained in the first experiments. Serum TBPA was
1
decreased transiently in the 4 week-old rats fed ad libitum
diets containing 6 % and 4 % protein, then i t was increased
1
towards normal or over the normal value in adulthood.
It is
1
weIl recognized that in the rat, energy is used for growth
only after the requirement for maintenance has been met
1
(38).
In the young rats, there was an insufficient protein
consumption during the first four weeks of dietary
1
treatment ; but in contrast , the energy intake was in excess
1
1

---

1
1
-58-
1
1
of that required for maintenance and limited growth.
These
1
rats were unable to meet their protein needs and their serum
TBPA was decreased. As the animaIs became adult, protein
1
requirements declined.
For normal growth,
casein plus
DL-methionine levels can be decreased from 14 to 4 % of the
1
diet from 7 to 50 weeks (39),
suggesting that in the adult
1
rats fed ad libitum the 4 % and 6 % prote in diets, the
prote in consumption was sufficient for maintenance and
1
limited body weight gain.
In these circumstances,
serum TBPA
was not affected.
In the growing animaIs fed the 0.5 %
1
protein diet,
both prote in and energy were limited. Growth
1
was stopped and the serum TBPA level was markedly decreased
even in adulthood.
To our knowledge, only another
1
experimental study has been published on the effect of
malnutrition on serum TBPA in rats fed a protein depleted
1
diet (43) and our findings agree weIl with this report.
1
However, the most notable result of this work was the
strong influence of energy restriction on the serum TBPA
1
level.
The TBPA concentration decreased in aIl pair-fed and
adult energy restricted rats. The pair-fed rats received the
1
same amount of food as that consumed by the protein
1
deficient animaIs but their body weight was in between that
of the corresponding protein deficient rats and the control
1
rats. At any time, their energy intake per g body weight was
less than that of the control rats and of the deficient
1
rats.
Energy rather than prote in appeared to be the factor
1
1

---

1
1
-59-
1
1
limiting growth in the pair-fed animais. Waterlow et al.
1
have reported that protein cannot be utilized if energy is
lacking and extra protein will be of no value (44).
This
1
could explain the high sensitivity of TBPA to energy
restriction. Furthermore,
rats with similar body weight and
1
similar protein intake (9 % and 0,5 % PF group) showed
1
marked differences in serum TBPA level depending on the
amount of energy consumed.
1
The concentration of TBPA in cerebrospinal fluid (CSF)
appeared to be less sensitive to dietary changes. Our
1
results which are in accordance with others (20) suggest
1
that the concentration of TBPA in CSF is regulated
independently from TBPA in serum. TBPA in CSF was not
1
affected by dietary intake,
unless by severe protein-energy
restriction.
1
The serum albumin concentration decreased with
decreasing dietary protein intake and was not influenced by
1
energy restriction,
not even by severe restriction.
These
1
results agree with those of Lunn et al.
(42) with only a
slight difference concerning the 6 % group. This difference
1
might be explained by variation in the duration of dietary
treatment : in the 6 % group,
serum albumin was
1
significantly low during the first two weeks of dietary
1
treatment (not shown) , then increased towards normal at the
end of the six week-period.
Lunn et al.
report that
1
hypoalbuminemia occurs with inadequate dietary prote in
1
1

---

1
1
-60-
1
1
intake and excess consumption of energy.
In aIl groups,
1
except the 0.5 % one, energy intake per unit body weight
decreased with prolonged exposure to the diets.
Thus i t is
1
possible that reduction in energy consumed in excess might
explain this increment.
1
Transferrin was less sensitive than TBPA to protein and
1
energy restriction but i t was lowered by severe prote in and
energy deficiency.
1
In this study, we also investigated the consequence of
decreased TBPA on the amount of circulating thyroid
1
hormones,
since TBPA is recognized to be the major T4
1
carrier in the rat (45,46).
In energy restricted rats,
total
T4 as weIl as TBPA levels were decreased and increased on
1
refeeding without change in the concentration of free T4,
suggesting that the observed decreased in TBPA was related
1
only to modification in the T4 bound fraction.
Independently
1
of changes in the concentration of the binding protein,
dietary prote in intakes per se had an effect on the
1
concentration of thyroid hormones. Both total T4,
free T4
and free T3 were influenced by the 9 % prote in diet.Serum
1
free T4 decreased markedly while serum free T3 was increased
1
and serum total T3 unchanged.
Previously, reduced free and
total T4 have been reported in rats fed low protein diets
1
(47,48) but results from the literature are conflicting.
Total T4 could either be normal or reduced (47-50).
It was
1
not our interest to look into the mechanism which controls
1
1

---

1
1
-61-
1
1
thyroid hormone secretion and distribution in protein-energy
1
restriction.
However,
i t is unlikely that the decreased
total T4 in the 9 % protein-fed rats could be the result of
1
a reduced thyroidal secretion or an age related-change in
the T4 concentration. Singh et al.
(51) have shown that
1
the protein content of a diet is not a limiting factor for
1
the rate of thyroxine secretion until i t is lower than 5 %
(w/w). Furthermore, T4 values recorded in the control rats
1
were within the normal range for age ; no age difference was
found in serum T4 in rats up to three months (52).
The
1
increase in serum free T3 was in keeping with the results of
1
Sawaya and Lunn (53) in rats fed low protein diets.
Elucidation of the physiological significance of the changes
1
of thyroid hormones as a result of feeding low prote in diets
or of decreased protein binding requires further studies.
1
Nevertheless, these results must be extrapolated with great
1
awareness of differences in the thyroxine-binding protein of
rat and man.
In man,
TBPA is not the major T4 binding
1
protein ; thyroxine binding globulin transports most of the
circulating T4 (54).
1
This study has clear1y demonstrated that serum TBPA is
1
a reliable marker in the detection of moderate and severe
protein-energy restriction. The simultaneous measurement of
1
major acute-phase reactant proteins allows a clearer
interpretation of serum TBPA values in nutritional
1
deficiency. The study also showed that TBPA is more
1
1

---

1
1
-62-
1
1
sensitive to energy restriction than albumin or transferrin.
1
The serum TBPA level seems to be more closely related to the
protein and energy intakes than the prote in or energy
1
content of the diet.
Although the precise mechanism by which
energy restriction lowered serum TBPA level requires
1
additionnaI works,
these experimental results May have
1
important implications for the assessment of marginal or
moderate PEH in human and particularly in children.
Dietary
1
surveys indicate that the vast majority of cases of
protein-energy deficiency in children are the result of
1
an inadequate intake of total energy (55). Moreover,
i t 1s
1
weIl recognized that anthropometric measurements alone as
proposed by the World Health Organizatlon underestimate
1
seriously the number of children in the developping
countries suffering from mild to moderate protein-energy
1
malnutrition.
1
1
1
1
1
1
1
1

---

1
1
-63-
1
1
References
1
1. Haider M,
Haider SQ. Assessment of prote in-calorie
1
malnutrition. Clin Chem 1984;30:1286-99.
1
2.
Ingenbleek Y,
De Visscher M,
De Nayer P. Measurement of
1
prealbumin as an index of prote in-calorie malnutrition.
Lancet 1912;ii:106-9.
1
3.
Ingenbleek Y,
Van Den 5chrieck HK, De Nayer P, De
1
Visscher M. Albumin, transferrin and the thyroxine-binding
1
prealbumin/retinol-binding protein (TBPA-RBP) complex in
assessment of malnutrition. Clin Chim Acta 1915;63:61-1.
1
4.
Large 5, Neal G, Glover J,
Thanangkul 0, OIson RE.
The
1
early changes of retinol-binding protein and prealbumin
1
concentrations in plasma of protein-energy malnourished
children after treatment with retinol and an improved diet.
1
Br J Nutr 1980;43:393-402.
1
5. Ogunshina SO,
Hussain MA. Plasma thyroxine-binding
prealbumin as an index of mild protein-energy malnutrition
1
in Nigerian children. Am J Clin Nutr 1980;33:194-800.
1
[
1
1

---

1
1!
1
-64-
1
1
6. Maire B, Wade S, Bleiberg F,
et al. Absence of variation
1
in facteur thymique serique activity in moderately and
severely malnourished Senegalese children. Am J Clin Nutr
1
1982;36:1129-33.
1
7. Delpeuch F, Cornu A,
Chevalier P. Detection of moderate
1
protein-energy malnutrition in pre-school children. Trans R
Soc Trop Med Hyg 1980;74:192-6.
1
8. Bleiberg F, Brun TA, Goihman S, Gouba E. Absence of
1
relationship between prealbumin concentration and weight for
1
height in children from Upper Volta. Nutr Rep Int
1980;21:123-6.
1
9. Smith FR, Suskind R,
Thanangkul 0,
Leitzmann C, Goodman
1
DS, OIson RE.
Plasma vitamin A,
retinol-binding protein and
1
prealbumin concentrations in protein-calorie malnutrition.
III-Response to varying dietary treatments. Am J Clin Nutr
1
1975;28:732-8.
1
10. Shetty PS, Watrasiewicz KE,
Jung RT,
James WPT. Rapid
turnover transport proteins : an index of subclinical
1
protein-energy malnutrition.
Lancet 1979;ii:230-2.
1
[
1
1

---

,
1
1
-65-
1
1
11. Kelleher PC,
Phinney sn, Sims EAH, Bogardus C, Horton
1
ES. Effects of carbohydrate-containing and
carbohydrate-restricted hypocaloric and eucaloric diets on
1
serum concentrations of retinol-binding protein,
thyroxine-binding prealbumin and transferrin. Hetabolism
1
1983;32:95-101.
1
12.
Hoffer LJ,
Bistrian BR, Young VR,
Blackburn GL,
1
Wannemacher HW. Hetabolic effects of carbohydrate in
low-calorie diets.
Metabolism 1984;33:820-5.
13. Young GA,
1
Chem C,
Collins JP,
Hill GL.
Plasma proteins in patients
1
receiving intravenous hyperalimentation after major surgery.
Am J Clin Nutr 1919;32:1192-9.
1
14. Bourry J, Hilano G,
Caldani C, Schneider H.
Assessment
1
of nutritional proteins during the parenteral nutrition of
1
cancer patients.
Ann Clin Lab Sci 1982;12:153-62.
1
15. Ne -
IUB and JCBN Newsletter. J
Biol Chem 1981;256:12-4.
1
16. Bleiberg F,
Vranckx R,
Wade S,
Nunez EA.
A simplified
1
meth6d for the preparation of rat thyroxine-binding
prealbumin.
Factors influencing its circulating level.
BBA
1
1985;828:210-1.
[
1
1

---

1
1
-66-
1
1
11. Kabat EA, Moore DH,
Landow H. An electrophoretic study
1
of the prote in components in cerebrospinal fluid and their
relationship to the serum proteins. J
Clin Invest
1
1942;21:511-1.
1
18. Soprano DR,
Herbert J, Soprano KJ,
Schon EA, Goodman DS.
1
Demonstration of transthyretin mRNA in the brain and other
extrahepatic tissues in the rat. J Biol Chem
1
1985;260:11193-8.
1
19. Dickson PW,
Howlett GJ,
Schreiber G. Rat transthyretin
1
(prealbumin) molecular cloning,
nucleotide sequence, and
gene expression in liver and brain. J Biol Chem
1
1984;260:8214-9.
1
20. Dickson PW,
Aldred AR, Marley PD,
Bannister D, Schreiber
1
G. Rat choroid plexus specializes in the synthesis and the
secretion of transthyretin (prealbumin).
Regulation of
1
transthyretin synthesis in choroid plexus is independent
from that in liver. J Biol Chem 1986;261:3415-8.
1
1
21.
Dickson PW, Aldred AR, Marley PD, Guo-Fen T,
Howlett GJ,
Schreiber G. High prealbumin and transferrin mRNA levels in
1
the choroid plexus of rat brain.
BBRC 1985;121:890-5.
1
1
1

---

1
1
-67-
1
1
22. Socolow EL,
Woeber KA, Purdy RH, Holloway MT,
Ingbar SH.
1
Preparation of 131I-Iabeled human serum prealbumin and its
metabolism in normal and sick patients. J Clin Invest
1
1965;44:1600-9.
1
23. Felding P, Fex G. Rates of synthesis of prealbumin and
1
retinol-binding protein during acute inflammation in the
rat. Acta Physiol Scand 1985;123:411-83.
1
24. Bleiberg-Daniel F,
Wade S,
Labarre C, et al.
Variations
1
in plasma thyroxine-binding prealbumin (TBPA) in relation to
1
other circulating proteins in post-operative patients during
rapid oral refeeding.
Hum Nutr : Clin Nutr 1985;39C:55-62.
1
25. Dickson PW,
Howlett GJ, Schreiber G.
Metabolism of
1
prealbumin in rats and change induced by acute inflammation.
1
Eur J Biochem 1982;129:289-93.
1
26. Schreiber G,
Howlett G, Nagashima M,
et al.
The acute
phase response of plasma prote in synthesis during
1
experimental inflammation. J Biol Chem 1982;251:10211-1.
1
21. Smith FR, Goodman DS. The effects of diseases of the
1
liver, thyroid,
and kidneys on the transport of vitamin A in
human plasma. J Biol Invest 1911;50:2426-36.
1
1
1

---

1
1
-68-
1
1
28. Peterson PA, Rask L, Ostberg L, Andersson L,
Kamwendo F,
1
Pertoft H. 5tudies on the transport and cellular
distribution of vitamin A in normal and vitamin A-deficient
1
rats with special reference to the vitam in A-binding plasma
protein. J Biol Chem 1973;248:4009-22.
1
1
29.
Navab M, Mallia AK,
Kanda Y, Goodman D5. Rat plasma
prealbumin.
Isolation and partial characterization. J Biol
1
Chem 1977;252:5100-6.
1
30. 5undelin J, Melhus H,
Das 5, et al.
The primary
1
structure of rabbit and rat prealbumin and a comparison with
tertiary structure of human prealbumin. J Biol Chem
1
1985;260:6481-7.
1
31. Jaffers WA, Griffith JQ.
The central nervous system.
1
In : Farris EJ, Griffith JQ,
eds.
The rat in laboratory
investigation,
2nd ed.
New York: Hafner Press,
1949;196-7.
1
32. Laurell CB. Electroimmunoassay. Scand J Clin Lab Invest
1
1972;29 Suppl 124:21-37.
1
33. Mancini G,
Carbonara AO, Heremans JF.
Immunochemical
1
quantitation of antigens by single radial immunodiffusion.
Int J
Immunochem 1965;2:235-59.
1
1
1

---

1
1
-69-
1
1
34. Vahlquist A,
Rask L,
Peterson PA, Berg T.
The
1
concentrations of retinol-binding protein, prealbumin, and
transferrin in the sera of newly delivered mothers and
1
children of various ages. Scand J Clin Lab Invest
1975;35:569-75.
1
1
35. Grant JP, Custer PB,
Thurlow J. Current techniques of
nutritional assessment. Surg Clin North Am 1981;61:437-63.
1
36. Eddington CL, Upchurch HF, Kampschmidt RF. Quantitation
1
of plasma a
AP globulin before and after stimulation with
2
1
leukocytic extracts.
Proc Soc Exp Biol Med 1972;139:565-9.
1
37. Shibata K, Okubo H,
Ishibashi H, Tsuda K. Rat al acid
glycoprotein.
Purification and immunological estimation of
1
its concentration.
BBA 1977;495:37-45.
1
38. National Research Council, Subcommittee on Laboratory
1
Animal Nutrition.
Nutrient Requirements of Laboratory
Animal. No.10, third revised ed. Washington:
National
1
Academy of Sciences,
1978.
1
39. Hartsook EW, Mitchell HH. The effect of age on the
1
protein and methionine requirement of the rat. J Nutr
1956;60:173-95.
1
1
1

---

1
1
-70-
1
1
40. Yoshida A,
Harper AE, Elvehjem CA. Effects of prote in
1
per calorie ratio and dietary level of fat on calorie and
protein utilization. J Nutr 1951;63:555-10.
1
41. Coward WA, Whitehead RC,
Lunn PG. Reasons why
1
hypoalbuminaemia May or May not appear in protein-energy
1
malnutrition.
Br J
Nutr 1911;38:115-26.
1
42.
Lunn PG,
Austin S. Dietary manipulation of plasma
albumin concentration. J
Nutr 1983;113:1191-802.
1
1
43. Peterson PA,
Nilsson SF, Ostberg L, Rask L,
Vahlquist A.
Aspects of the metabolism of retinol-binding protein and
1
retinol.
Vitam Horm 1914;32:181-214.
1
44.
Waterlow Je,
Payne PRo The protein gap.
Nature
1
1915;258:113-1.
1
45. Davis PJ,
Spaulding SW, Gregerman RI.
The three
thyroxine-binding proteins in rat serum : binding capacities
1
and effects of binding inhibitors. Endocrinology
1
1910;81:918-86.
1
46. Sutherland RL,
Brandon MR.
The thyroxine-binding
properties of rat and rabbit serum proteins. Endocrinology
1
1916;98:91-8.
1
1

---

1
1
-71-
1
1
47. Smallridge RC, Glass AR, Wartofsky L,
Latham KR,
Burman
1
KD.
Investigations into the etiology of elevated serum T3
levels in protein-malnourished rats. Metabolism,
1
1982;31:538-42.
1
48. Edozien JC,
Niehaus N, Mar MH, Makoui T,
Switzer BR.
1
Diet-hormone interrelationships in the rat. J
Nutr
1978;108:1767-76.
1
49. Cox MD,
Dalal SS, Heard CRC, Millward DJ.
Metabolic rate
1
and thyroid status in rats fed diets of different
1
protein-energy value:
the importance of free T . J Nutr
3
1984;114:1609-16.
1
50. Tulp üL,
Krupp PP,
Danforth E, Horton ES.
1
Characteristics of thyroid function in experimental protein
1
malnutrition. J Nutr 1979;109:1321-32.
1
51. ~ingh DV, Anderson RR, Turner CW. Effect of decreased
dietary protein on the rate of thyroid hormone secretion and
1
food consumption in rats.
J Endocr 1971;50:445-50.
1
52. Tang F. Effect of sex and age on serum aldosterone and
1
thyroid hormones in the laboratory rat.
Horm Metabol Res
1985;17:507-9.
1
1
1

---

1
1
-72-
1
1
53. Sawaya AL,
Lunn PG. Evidence suggesting that the
1
elevated plasma triiodothyronine concentration of rats fed
on protein deficient diets is physiologically active.
Br J
1
Nutr 1985;53:175-81.
1
54. Robbins J,
Cheng SY, Gershengorn MC, Glinoer D, Cahnmann
1
HJ, Edelnoch H. Thyroxine transport proteins of plasma.
Molecular properties and biosynthesis. Rec Prog Horm Res
1
1978;34:477-519.
1
1
55. Sukhatme PV.
Incidence of protein deficiency in relation
to different diets in India.
Br J
Nutr 1970;24:477-87.
1
1
1
1
1
1
1
1
1
1

---

-- ... _-----------------
Table 1. COMPOSITION OF DIETS (g/100g)
Component
Prote in content (w/w) of diet
18 %
9 %
6 %
4 %
0.5 %
Casein
19
9.5
6.5
4.1
1
-...j
DL-methionine
0.46
0.23
0.15
0.10
-
I..N
1
Yeast
4
2
1
1
1
Sucrose
62.64
74.3 rt
78.45
80.90
85.10
Corn oil
3
3
3
3
3
Cellulose
3.7
3.7
3.7
3.7
3.7
a
Vitamin mixture
2.2
2.2
2.2
2.2
2.2
b
Salt mixture
5
5
5
5
5
apreviously reported by Lemonnier D et al. Nutr Metab 1974;16:15-29
bReport of the AIN Ad Hoc Committee on Standards for Nutritional Studies"
J
Nutr 1977;107:1340-48.

---

- - - - - - - - - - - - - - - - - - - - -
Table 2. BODY WEIGHT AND FOOD CONSOMPTION OF RATS FED AD
LIBITUM THE CONTROL DIET (18 % PROTEIN).
LOW PROTEIN DIETS
(9 %, 6 %,
4 %, 0.5 %) OR THE CONTROL DIET
IN RE5TRICTED AHOUNT (PF GROUPS)
(Exp.
1 )
For weeks
0 - 1
1 - 2
2 - 4
4 - 6
a
b
Body weight
FI
PIc
Body weight
FI
PI
Body weight
FI
PI
Body weight
FI
PI
(g)
1
Groups
(g)
(g)
(g)
-.J
P1
C
131
11.8
2.06
171
10.8
1. 75
261
8.7
1. 38
320
6.8
1.16
9 %
126
13.7
1.19
164
11. 9
0.98
244
9.3
0.70
297
7.2
0.57
9 % PF
-
-
-
-
-
252
7.1
1. 35
302
6.1
1.13
6 %
113
14.3
0.79
134
12.8
0.67
170
10.5
0.59
218
8.4
0.50
6 % PF
-
-
-
-
-
231
6.9
1. 30
282
6.0
1.13
4 %
103
17.1
0.58
116
14.5
0.52
140
10.8
0.43
166
9.5
0.36
4 % PF
-
-
-
-
-
-
208
6.7
1. 20
253
5.7
1.03
0.5 %
74
10.1
0.05
68
10.8
0.06
66
9.8
0.06
60
8.8
0.05
0.5 % PF
108
7.0
1.29
121
6.2
1.16
121
5.8
0.99
107
5.6
0.93
a Mean body weight at the end of each Period. Food consumption was measured every 24 or 48 h.
b
c
FI = Mean Food intake (g) / 24 h / 100g body weight·
PI = Mean Prote in intake (g) / 24 h / 100g body weight·

---

---------------------
Table 3. EFFECTS OF THE 6-WEEK DIETARY TREATMENT
ON THE CONCENTRATION OF TBPA IN CEREBROSPINAL
FLUID (CSF) AND ON THE SERUM CONCENTRATION OF
ALBUMIN AND TRANSFERRIN (Exp.
1)
TBPA in CSF
Albumin
Transferrin
Groups
(mg/100mL)
(g/L)
(g/L)
---------------------------------------------------------------------------------------------------
C
1.26 !. 0.05
29.5 !. 0.5
4.2 !. 0.12
1
-..l
VI
9 %
1.19!.0.02
28.8 !. 0.8
4.3 + 0.10
1
9 % PF
1.39 !. 0.15
28.9 !. 1.1
4.2 !. 0.10
6 %
1.32 !. 0.06
29.5 ±. 0.7
4.0!.0.13
6 % PF
1. 29 !. 0.04
30.0 ±. 0.4
4.3 ±. 0.17
4 %
1.16.± 0.04
24.2 ± 0.9 **
3.9 + 0.13*
4 % PF
1.38 .± 0.05
28.7 ± 0.6
4.3 !. 0.19
0.5 %
1.05 !. 0.11 *
12.8 ±. 0.4 **
2.4 ±. 0.06 **
0.5 % PF
1.29.± 0.11
28. 9 .± 1. 8
3 . 5 !. 1. 20 **
Values are mean .± SEM n = 6 per group *p < 0.05, **p < 0.01 versus the control group (c)

---

---------------------
Table 4. EFFECTS OF ENERGY RESTRICTION ON BODY WEIGHT AND
SERUM PROTEIN LEVELS IN 8 RATS FED A CONTROL DIET WITH AN
AUTOMATED FEEDER
80 % energy restricted
day 0
day 8
day 21
animaIs
-------------------------------------------------------------------------------------
Body weight (g)
340 ± Il
341 ± 13
363 ± 12
1
.......
0'\\
1
TBPA (mg/100mL)
55.6 + 1.1
48.6 + 0.7*,a
**
43.3 + 1.7
Albumin (g/L)
32.6 + 0.3
33.0 + 0.8
32.9 + 0.5
Transferrin (g/L)
4.1 + 0.16
4.6 + 0.10
4.5 + 0.3
Values are mean ± SEM. the rats were fed 80 % of the amount of food consumed by age
and weight-matched control rats (for details see Materials and Methods). Paired
Student'st test was used for comparaisons. * p < 0.05, ** p < 0.01 Versus the mean by
day 0 ; aSignificant difference between day 8 and day 21 (p < 0.05).

---

- - - - - - - - - - - - - - - - - - - - -1
Table 5. EFFECTS OF ENERGY RESTRICTION AND REFEEDING ON BODY
WEIGHT AND SERUM PROTEIN LEVELS OF RATS FED EQUIVALENT AHOUNT OF PROTEIN.
Energy restriction
Ad libitum refeeding
a
groups
100 % EF
90 % EF
80 % EF
100 % EF
90 % EF
80 % EF
Parameters
Body weight (g)
230
217
210**
274
275
274
± 7
± 6
+ 3
± 7
± 6
± 7
1
-J
TBPA(mgj100mL)
53.6
41.3**
39.2**
-J
51.1
51.0
48.8
1
± 2.5
± 1. 8
+ 1. 13
± 0.4
± 2.0
± 2.0
Albumin (gjL)
30.7
31.1
31.4
31.3
31.1
30.3
± 0.5
±. 0.9
+ 1. 0
±. 0.7
± 0.2
±. 0.7
Transferrin (g/L»
4.5
4.6
4.5
4.4
4.5
4.6
± 0.18
± 0.04
+ 0.09
+ 0.12
± 0.11
± 0.12
a the 100 % EF (Energy Fed) group received ad libitum a diet containing 9 % protein (w/w).
The 90 % EF and 80 % EF groups were
fed the same amount of prote in as that consumed by the 100 % EF group with 90 and 80 % restrictions of energy intake. This was
achieved by making isocaloric diets containing 10 and Il.2 % proteins (w/w). After Il days, the energy restricted rats were
refed ad libitum with the 9 % protein diet for 12 days. Means ± SEM are show.
** p < 0.01 versus the 100 % EF group.

---

- - ... - - - - - - - - - - - - - - - - - -
Table 6. THYROID HORMONE CONCENTRATIONS IN THE
ENERGY-RESTRICTED RATS FED EQUIVALENT AMOUNT OF PROTEIN
----------------------------------------------------------------------------------------------------
Total T4
Free T4
Total T3
Free T3
Groups
(~g/mL)
(ng/100mL)
(ng/100mL)
(pg/mL)
----------------------------------------------------_.. _-----------------
Control day 0
5.7 :!. 0.3
2.26 :!. 0.10
50 !. 4
1. 9 !. 0.25
Energy-restricted (11 days)
a
90 % EF
3.6 ± 0.4 a ,b
1.42 ± 0.16
56 ± 3
2.7 ± 0.16a ,b
a
80 % EF
2.8 ±. 0.2 a ,b
1. 50 ± O. 10
54 ± 2
3.0 ±. 0.09 a ,b
1
Refeeding (12 days)
-...J
œ
4.9 ± 0.3a ,b
1.75 ± O.O~a,b
a
1
00 % EF
!>6 !
3
2.6 ±. 0.16
a
a
a
80 % EF
4.4 ±. 0.3
1.50 ±. 0.09
48 ± 3
2.4 ± O.11
Ad libitum fed (23 days)
a
a
a
100 % EF
4.2 ± 0.3
1. 46 .:t O. 06
48 ± 3
2.3 ± 0.13
Values are mean ± SEM. n = 6-10 rats per group. Statistically significant (a) versus the control group,
(b) versus the 100 % EF group,
p < 0.05 or less. 9 rats, previously fed a control diet were killed by
day 0 (control day 0). The 100 % EF group was fed ad libitum a 9 % prote in diet, the 90 % EF and 80 % EF
groups were energy restricted but received the same amont of protein as the 100 % EF group ; they were
refed ad libitum with the 9 % prote in diet.

---

- - ... - - - - - - - - - - - - - - - - - -
Table 7. BODY WEIGHT, FOOD INTAKE AND SERUM TBPA
LEVELS OF RATS FED A CONTROL DIET (18 % PROTE IN , 3.7
kcal/g
CONTROL GROUP) OR A HIGH CALORIC DIET
(18 % PROTEIN,
4.7 kcal/g : HC group)
day 0
day 8
day 21
Groups :
Control
HC
Control
HC
Control
HC
1
.......
-0
-------------------------------
1
Body weight
346 ± 6
340 ± 10
360 ± 7
369 ± 13
383 ± 7
386 ± 11
(g)
1
Food intake
-
19.7 ± 0.64
16.3** ± 0.64
20.2
± 0.47
15.4** ± 0.44
(g/24 h)
TaPA
54.5 ± 2.4
55.0 :!::. 1.6
49.8 ± 2.1
52.3 ± 2.9
47.7 ± 2.1
48.8 ± 2.3
(mg/100mL)
Value5 are mean :!::. SEM, n = 7 per group, ** p < 0.01 ver5US the corresponding control group. 1 average intake5 between 0 - 8
and 8 - 21 days.

---

1
1
-so-
I
1
%. variation
Co)
..
~
C1I
0)
en
co
1
(j
0
0
0
0
. 0
0
0
0
0
~
1
III
Q.
1
=
a-
-c
3
1
-•Q.
ca
1
~
0
C
''tI
....,
la
f-' .
..:l
1
C1I
~
C1I
,.....
C1I
~
1
0)
en
a-
...
~
~
~
~
1
~
~
~
a-"1
1
1
a-.
/i.
'tI
1
III
1
/
':
~
.
1 r:
1
1
/1
, :
-
•
1
/ /
}
Q.
,
1
/ /
ca
,
,
/ /
~
,
1
/ /
0
+
a-
q~a-
flll
c::
'tI
1
, \\
la
1
1
, \\
1
1
~
1
,\\
,\\
1
C1I
1
C1I
,\\ 1
1
:J:'
(1)
a-'
~~
1
0)
1
1
1
1

---

1
1
-81-
1
1
Figure Legends.
1
1
Fig.1
Variation in serum transthyretin levels of the 4
week-old rats fed ad libitum the control diet ( ... )
1
low prote in diets : 9 % (A
~ ). 6 % (0
0) •
1
4 % (0
0),
0 _5 % (e
e)
or the control diet in
restricted amount (Pair--fed groups)
: 9 % PF tr- - - ~ •
1
6 % PF 0- - - -0 ,
4 % PF 0- - - ~ JO. 5 % PF ... - - -e
The value 100 is assigned to the control group
1
a
b
p < 0.05,
p < 0.01 versus the control group_
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-82-
1
ART l C L E N° 3
1
1
"Rat transthyretin. The effects of acute short-term food, deprivation
1
on the serum and cerebrospinal fluid concentration and on the hepatic
mRNA LEVEL. (soumis à Journal of Nutrition).
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
-83-
1
1
1
RAT TRANSTHYRETIN. THE EFFECTS OF ACUTE SHORT-TERM
FOOD DEPRIVATION AND REFEEDING ON THE SERUM AND
1
CEREBROSPINAL FLUID CONCENTRATION AND ON THE
HEPATIC mRNA LEVEL
1
1
SALlMATA WADE, FANNY BLEIBERG-DANIEL, BEATRICE LE MOULLAC.
1
1
1
Unité de Recherches sur la Nutrition et l'Alimentation, U.l
INSERM, Hôpital Bichat, 110 Bd. Ney, 15018 Paris (France)
1
1
1
* To whom correspondence should he addressed.
1
1
1
1
Short running title
Rat Transthyretin during Fasting and
Refeeding
1
1
1
1

---

1
1
-84-
1
1
ABSTRACT
To gain some insight into the nutritional factors
1
which affect blood transthyretin level and its metabolism,
we have investigated the response of rat transthyretin to
1
one, two and three days of fasting and to 24 h of fasting
followed by refeeding.
The observed changes were compared to
1
the level of transthyretin in cerebrospinal fluid
(CSF) and
1
to the circulating amount of thyroid hormones.
Dot
hybridation of a hepatic mRNA-cDNA probe specifie for
1
transthyretin was used to measure the relative level of mRNA
transthyretin. Serum transthyretin decreased significantly
1
after fasting and the decrease was proportionnai to the
1
duration of the treatment.
When rats were fasted for 24 h,
then refed,
serum transthyretin levels remained low after
1
two days of refeeding.
The dot hybridation results suggested
that reduced liver synthesis was not the only mechanism
1
which could explain this long lasting effect of fasting.
1
Transthyretin level in CSF was not influenced by fasting.
In addition to the high sensitivity of serum transthyretin
1
to food deprivation,
the study also showed two distinct
influences of fasting on the thyroid hormones : a primary
1
effect which result in an inhibition of the conversion of T4
1
to T3 and a decreased T4 bound fraction as a result of
decrease serum level of transthyretin.
1
INDEXING KEY WORDS : Fasting. Rat Transthyretin.
1·
Thyroxine-binding prealbumin (TBPA).
Cerebrospinal fluid.
1
cDNA transthyretin.
Hepatic transthyretin mRNA. Thyroid
hormones.
1

---

1
1
-85-
1
1
Nutritional studies in man have suggested that plasma
1
thyroxine-binding prealbumin (TBPA),
recently named trans-
thyretin (1),
could be used as one of the most sensitive and
1
specifie nutritional markers (as reviewed in 2). The plasma
TBPA level decreases in protein-energy malnourished children
1
and in hospitalized adult patients (2).
However, Most of
1
these clinical studies have examined prolonged or
established nutritional depletion. Yet,
some hospitalized
1
patients,
in developped as weIl as in developing countries,
May suffer from acute short-term starvation. Moreover,
the
1
interpretation of the TBPA values in many clinical studies
1
is difficult because of the frequent presence of infection
or malignancy in the malnourished patients. During inflamma-
1
tory or infectious states, the liver switches its priority
for protein synthesis by diminishing its production of
1
albumin,
transferrin and particularly TBPA and accelerating
1
its production of various acute-phase reactant proteins
(3-5). We have demonstrated that this effect of the inflam-
1
matory response on plasma TBPA level is independent of
the dietary intake (6).
1
To get specifie information, we have studied the
1
response of rat TBPA to varying nutritional stress.
Previously, we have isolated and purified rat TBPA (7). Rat
1
as weIl as human TBPA is a carrier protein for thyroxine and
an indirect vehicle for vitamin A.
In view of the high
1
degree of structural and functional similarities between the
1
1

---

1
1
-86-
1
1
rat and the human TBPA (8-11),
the rat was chosen as an
1
experimental model.
In this work we have examined the
effects of various periods of food deprivation and of
1
acute short-term food starvation,
followed by refeeding.
The
inflammatory and infectious state of the animaIs,
if any,
1
was controlled by the measurement of two main acute-phase
1
reactant proteins in the rat : a
-macrog lobulin and a
-acid
2
l
glycoprotein (orosomucoid). We attempted to correlate the
1
serum TBPA level to the serum albumin and transferrin
concentrations,
to the level of TBPA in cerebrospinal fluid
1
(CSF) and to the hepatic level of mRNA TBPA.
The liver mRNA
1
level was measured using a cDNA probe specifie for TBPA
(11).
Also we have investigated whether changes in serum
1
TBPA resulted in a modification of the amount or
distribution of thyroid hormones (thyroxine and triiodo-
1
thyronine).
1
1
MATERIAL AND METHüDS
1
Animgls gnd di~t~
Three sets of experiments were conducted on adult male
1
Wistar rats (lffa-Credo,
l'Arbresle,
France). The rats were
1
housed individually and maintained at 22 + 2 oc on a 12 h
light /
12 h dark cycle.
AlI the animaIs had free access to
1
water.
1
1

---

1
1
-87-
1
1
In experiment l,
Three groups of 8 rats (mean weight
1
300 g) previously fed a commercial stock diet were formed
and fasted for one,
two and three days,
respectively. An
1
aliquot of blood was collected from the tail vein before
fasting (FO) and at the end of each fasting period (F1,
F2
1
and F3,
respectively).
1
In experiment 2,
a group of 8 rats was fasted for 24 h,
1
followed by six days of refeeding.
The animaIs received a
control diet containing 10% casein, 0.46% DL-methionine,
1
73.44% wheat flour,
4% yeast,
2.7% corn oil,
3.7% cellulose,
1
2.2% vitamin mixture (12) and 3.5% salt mixture (13).
Blood
was drawn from the tail vein before fasting (FO),
afterthe
1
fasting state (F1),
one,
two,
and six days after refeeding
(R1, R2 and R6,
respectively).
To differenciate the effects
1
of fasting from those of an eventual inflammatory state
1
induced by the repeated blood collection from the tail,
the
experiment included a control age-matched group.
This group
1
received the control diet and was bled in the same way as
the fasted rats.
The food intake and the body weight of the
1
animaIs were recorded daily.
1
Experiment 3 consisted of 30 rats divided into 5 groups
1
of 6 animaIs each. They were maintained on a control diet
until they reached about 200 g of body weight.
The composi-
1
tion of the diet was 19% casein, 0.46% DL-methionine, 62.64%
1
1

---

1
1
-88-
1
1
suc rose , 4% yeast,
3% corn oil,
3.1% cellulose,
2.2% vitamin
1
mixture (12) and 5% salt mixture (13).
The first group was
killed at this stage to provide base line values for serum
1
and CSF TBPA concentrations,
serum thyroid hormones and
liver mRNA TBPA levels (FO).
The remaining 4 groups were
1
fasted for one day. A group of 6 rats were killed after the
1
fasting state (F1).
The 3 others were further divided in
groups of 6 animals and were fed the control diet for one,
1
two and three days. Six rats were killed after each feeding
period (R1, R2 and R3,
respectively).
1
1
The animaIs were anesthetized with diethyl ether and
blood was collected from the abdominal aorta.
The liver was
1
removed, washed in sterile 0.9% NaCl and immediately frozen
in liquid nitrogen.
Clear cerebrospinal fluid (CSF),
free of
1
blood contamination was obtained by cisternal puncture (14).
1
Serum and CSF"were kept at -20 oC until analysed.
1
Er~in and t~rQid hQrmQn~ m~a~urem~nt~
Purified rat TBPA and a monospecific anti-rat TBPA
1
antiserum were made in our laboratory according to
1
previously published procedures (1). Serum and CSF TBPA,
albumin and 02-macroglobulin concentrations were measured by
1
electroimmunoassay (15) while transferrin and orosomucoid
concentrations were determined by radial immunodiffusion
1
(16). Rat transferrin and albumin as well as anti-rat
1
1

---

1
,
1
1
1
-89-
1
1
albumin and anti-rat transferrin were supplied by Cappel
1
Laboratories (West Chester, USA).
Total thyroxine (TT4), Free thyroxine (FT4) and total
1
triiodothyronine (TT3) were determined in plasma py RIA (Kit
1
Amersham, France). A free T4 index (FT4-I) was calculated
using the ratio TT4/TBPA.
1
Ex~raQ~lQn Qi ~Qkal 11yer RNA
AlI the experiments were performed under sterile conditions.
1
The solutions and the materials were sterilized by
1
autoclaving and gloves were worn at aIl stages.
Total RNA
extraction was carried out according to the method of Itoh
1
et al.
(17).
Briefly, 0.5 g aliquots of frozen liver were
homogenized at 4 oC in 5 ml Tris /
EDTA buffer
(TEB: O.lH
1
Tris-Cl,
pH 9.0,
10 mM EDTA) containing 2% SDS, O.lM
1
dithiotreitol and in 5 ml of a solution of
Phenol/chloroform/isoamyl alcohol (50:50:2) saturated with
1
TEB.
The homogenate was centrifuged at 4000 x g at 4 oC for
10 min.
The organic phase was extracted with 5 ml of TEB
1
containing 0.1 M dithiothreitol.
The combined aqueous phases
were reextracted three times with phenol/chloroform/isoamyl
1
alcohol.
The total nucleic acids were precipitated overnight
1
by addition of 0.1 vol.
cold 2 M sodium acetate, pH 5.7, and
2.5 vol.
cold ethanol,
centrifuged (12.000 x g,
4 oC,
20
1
min) and dissolved in distilled water (2 to 2.5 ml).
The RNA
content was precipitated for 2 h at 4 oC with 0.25 vol.
10 H
[
LiCl, collected by centrifugation (12.000 x g,
4 oC,
10 min)
1
1

---

1
1
-90-
1
1
washed with 70% cold ethanol then dissolved with distilled
1
water (0.5 ml).
0.5 to 1 ~g of total RNA of each sample was
analysed by electrophoresis in a 1% agarose gel.
1
1
I~QlatiQD~ ~urificatiQD and nick=translatiQn Qf th~ QDNA
TBfA ~rQbe
1
The cDNA clone for TBPA (PA-l,
509 bp long) was kindly
provided by Dr Pero A. Peterson, Department of Cell
1
Research,
The Wallenberg Laboratory,
University of Uppsala,
1
Sweden.
This clone was inserted into pUC9 at the site AWAI
and digested by the bacterial strain JM83 (11).
The plasmid
1
was isolated from the bacterial cell by alkaline extraction
(18).
After digestion by AWAI restriction enzyme, the cDNA
1
was electrophoresed in a 5% polyacrylamide gel,
isolated,
extracted at 4 oC in an electrophoretic concentrator (Iseo,
1
NE,
USA) and purified on a DEA-Sephacel Column (18).
The
1
509 bp rat TBPA cDNA was then nick-translated (19) in the
presence of [a _32 p ]
deoxynucleotides (specifie activity
1
3000 Ci/mmol,
Amersham,
France).
The nick-translated plasmid
was applied onto a Sephadex G-50 Column to separate the
1
unincorporated [a _32 p ] nucleotides.
The specifie activity
7
1
of the labeled fragments was about 5 X 10
cpm/~g.
1
RNA-cDHA dQt bYbridatiQn~
The method described by White and Brancroft (20) was
1
used.
50 ~g of total RNA in 62.5 ~l distilled water were
1
1

---

1
1
-91-
1
1
incubated with 25 ~l of 37% formaldehyde and 37.5 ~l of 20 x
1
SSC (3 M NaCl,
3 M Na+ citrate,
pH 7.0) at 60 oC for 15 min.
875 ~l of 20 x SSC were added. The samples were serially
1
diluted with 20 x SSC and 200 ~l of each were applied twice
into a nitrocellulose filter in a 96-well Biorad Apparatus
1
(BRL. Bethesda). Yeast tRNA and rat intestine total RNA were
1
used as negative controls. The filters were dried and baked
for 2 h at 80 oC. RNA bound to the filters were
1
prehybridized at 42 oC in 50% deionised formamide,
5 x SSC,
10 X Denhardt's,
20 mM NaH P0
PH 7.0,
100 ~g/ml sonicated
1
2
4
denatured salmon sperm DNA. Hybridation was carried out
6
1
overnight with addition of 2.5 x 10
cpm of 32P-labeled TBPA
cDNA per dot blot.
The filters were washed four to five
1
times at room tempe rature in 2 x SSC, 0.1% SDS for a total
of 40 to 50 min.
This was followed by 2 washes (for a total
1
of 2 x 60 min),
at 60 oC in 1 x SSC, 0.1% SDS. The filters
1
were dried and autoradiographed at -80 oC using Kodak
X-O-Mat film,
then analysed by spectrodensitometry.
1
Qt~~lcal anglY~i~
1
All results are expressed as means ± SEM. Differences
1
in mean values between groups were determined by analysis of
variance (ANOVA) and the paired Student's t-test.
1
1
1
1

---

1
1
-92-
1
1
RESULIS
1
EXEeriment 1. Starvation for l, 2 and 3 days.
1
At the onset of the experiment (FO) the 3 groups of
animals had comparable body weight (296 ± 3 g) which
1
declined by 7,11 and 13.5% following one,
two and three days
1
of starvation,
respectively. Serum TBPA concentration was
also identical in the groups at FO (41.3 ± 0.7 mg/100 ml)
1
and decreased significantly by 24 (F1),
40 (F2) and 48% (F3) at the end of each fasting period.
1
1
EXQeriment Z. Starvation and refeeding.
Initially, the Mean body weight between the control and
1
the experimental groups were not statistically different
(306 ± 1 versus 291 ± 4 g at FO respectively).
The body
1
weight declined in the fasted rats (273 ± 4 g at F1 but was
1
rapidly gained after two days of refeeding (298 ± 5 g at
R2).
During the refeeding period,
(fig.
1), the food intake
1
of the experimental rats,
expressed as g/24 h/100 g body
weight increased significantly after the first day of
1
refeeding (R1) but was restored to normal as early as the
1
second day of refeeding (R2).
As expected, the serum TBPA level was reduced after the
1
first day of fasting (fig.
1). However, despite the earlier
changes observed in body weight and food intake, the level
1
was still low after two days of refeeding (R2). The TBPA
1
1

---

1
1
-93-
1
1
concentration returned to prestarvation values (FO) after
1
six days of refeeding.
The paired Student's t
test indicated
that the TBPA level was also significantly decreased in the
1
control rats from day 0 to day three ; however, this
1
decrease was less pronounced in these rats compared to the
fasted rats.
The ANOVA test showed significantly higher
1
values in the control than in the experimental rats at Fl,
Rl and R2 (fig.
1). Although an elevation of the level of
1
a2-macroglobulin was neither observed in the control nor in
the fasted rats (values less than 0.02 g/l), the Mean
1
orosomucoid concentration raised significantly with the
1
repeated bleeding (298 ± 29 to 398 ± 31 mg/l00 ml, from day 0
to day six, p < 0.05, Paired-t test).
1
The TBPA variations were compared to those of albumin
and transferrin.
The serum albumin level was not affected by
1
the imposed acute short-term food deprivation : the mean
1
value observed at FO was 3.6 ± 0.12 g/100 ml and did not
change throughout the experiment.
The serum transferrin
1
concentration decreased significantly in the fasted rats
(4.4 ± 0.14 g/l at FO ; 3.6 ± 0.13 g/l at Fl, p < 0.01) and
1
increased towards base line values from Rl (4.1 + 0.16 g/l,
1
p < 0.05 versus FO) to R2 (4.3 ± 0.10 g/l).
1
E~riment 3. Fasting and refeeding with glucose.
In this experiment,
in addition to the fasted and water
[
drinking rats,
6 rats were fasted for one day but were given
1
1

---

1
1
-94-
1
1
free access to a solution of 15% glucose in place of the
1
drinking water.
The results observed in the experiment 2 were confirmed
1
in experiment 3. Serum TBPA was decreased in the one-day
fasted rats and remained low after two days of refeeding.
1
The prestarvation values were reached in the 3-day refed
1
rats (Table 1). The level of TBPA in CSF was unchanged
throughout the study (Table 1).
1
Figure 2 shows the linear relationship between the
blood level of TBPA and that of thyroxine.
The correlation
1
between both parameters was highly significant : r
= 0.64,
1
n = 30, p < 0.001. There was also a significant linear
correlation between the measured free T4 (FT4) and the
1
calculated free T4 index (not shown ; r
= 0.54 n = 30,
p < 0.01). Thyroid hormone concentrations determined after
1
each period are presented in Table 2. While the FT4-1 was
1
increased after the short period of food deprivation,
neither TT4 nor FT4 concentrations were significantly
1
different from the prestarvation values.
By contrast, the
level of TT3 was markedly decreased.
1
The serum TBPA concentration of the fasted and glucose
1
drinking rats was comparable to that of the fasted and water
drinking animaIs:
36.8 ± 26 and 38.8 ± 2.8 mg/100 ml.
1
respectively, but was significantly reduced in comparison to
the value obtained before fasting p < 0.01. The TT3 level of
1
the glucose drinking rats was significantly higher than that
1
1

---

1
1
-95-
1
1
of the water drinking animais : 50 ± 5 and 29 ± 3 mg/lOO ml,
1
respectively, p < 0.01. The TT3 value recorded in these
animais was not different from the value before fasting
1
(Table 2).
After the refeeding state RI and R2 (Table 2), TT4
1
levels were significantly lowered and the distribution of T4
1
between bound and free was also modified : At R1, the
estimated bound T4 fraction decreased and simultaneously the
1
TT3 increased.
At RI,
as observed in the experiment 2 the
food intake of the fasted animais was increased then
1
returned to normal at R2.
There was a reduction in ail the
1
fractions measured at R2. At these stages (RI and H2) the
value of FT4-1 was comparable to the prestarvation value.
1
Following three days of refeeding,
the plasma level of the
thyroid hormones,
T4 and T3,
and the calculated FT4-1 were
1
ail increased.
1
After isolation of the cDNA TBPA probe from the plasmid
pue 9 (Fig.
3), the specificity and purity was controlled.
1
Poly (A)-rich RNA was isolated from the liver of control
rats by oligo (dt)- cellulose affinity chromatography (2l)
1
and analysed by Northern blotting.
The Northern blots
1
revealed that the rat TBPA mRNA migrated as a single band
(not shown). Antoradiographic hybridation spots of the
1
hepatic mRNA TBPA are shown in Fig.
4. The intensity of
hybridation was linear from 1.25 ~g to 10 ~g of total RNA.
1
Hybridation signais were obtained neither with the sample of
1
1

---

1
1
-96-
1
1
RNA isolated from the small intestine nor with yeast tRNA.
1
The relative density of the hybridation signaIs was
calculated using a standard value refered to as 100%
this
1
standard was a sample of rat liver total RNA,
serially
diluted and applied twice in each dot.
Compared to the value
1
before fasting (FO) the relative percentage of the TBPA mRNA
1
was decreased by 24% after the fasting day (F1) and
by 8% and 10% after the refeeding days RI and R2,
1
respectively.
It was increased by 1% after three days of
refeeding (R3).
However,
significant differences (p < 0.05)
1
were found only in the fasted rats compared to refed rats at
1
R3.
1
DISCUSSION
1
Transthyretin (TBPA) has been proposed as one of the
1
Most sensitive nutritional markers (2),
therefore i t is of
interest to define fully the nutritional
factors which may
1
affect its blood level.
In man,
various influences lower the
serum level of TBPA,
including inflammation and fasting
1
(6,22). There are only few works on the specifie effect of
1
acute food deprivation on serum TBPA in rat,
as weIl as in
man,
and these reports involve at least 3 to 5 days of
1
fasting (5,22).
To our knowledge,
no information is
available concerning how acute short-term starvation and
1
refeeding affects the concentration and the metabolism of
1
1

---

1
1
-97-
1
1
this protein.
This study was undertaken to investigate the
1
specifie response of TBPA to one,
two and three days of
fasting and the response to one day of fasting,
followed by
1
refeeding.
Different times of fasting allowed a clear definition
1
of the response of TBPA to acute food deprivation.
The serum
1
level of the protein progressively decreased with the
severity of starvation.
In the three-day fasted rats,
TBPA
1
values observed were in agreement with data obtained in
12-75 h fasted rats (5).
In some experiments.
rats were
1
fasted for 9 or 12 h
(unreported results).
In these rats.
a
1
slight but significant decrease in serum TBPA could be
observed after 12 h but not after 9 h food deprivation.
The
1
results. taken together indicate a high sensitivity of rat
TBPA to acute short-term food deprivation.
In the one-day
1
fasted and refed rats,
serially bled from the tail vein.
1
TBPA levels were low even after two days of refeeding.
Part
of this reduction was the reflection of an acute-phase
1
response.
The repeated bleeding does induce a slight
inflammatory reaction and this finding was supported by
1
the simultaneous measurement of serum TBPA and orosomucoid
1
in the control and experimental rats.
Nevertheless,
the only
effect of this inflammatory reaction was unable to explain
1
the over aIl reduction in TBPA levels observed in the fasted
and refed animaIs.
The reason for the long lasting effect of
1
starvation on the serum concentration of TBPA was not
1
1

---

1
1
-98-
1
1
apparent.
It could be the result of either decreased
1
synthesis,
altered distribution or increased catabolism.
Comparable results have been reported during the refeeding
1
of five-day starved subjects (22). Another work (5)
suggested that only decreased synthesis is responsible for
1
the decreased TBPA concentration during fasting.
However,
1
our results could not be explained on this basis.
In view of
the very short half life of rat TBPA.
la to 30 h (23,24). it
1
might be expected that synthesis and distribution and/or
catabolism were altered. To further investigate if a reduced
1
rate of synthesis of TBPA in the liver was the underlying
1
mechanism which controls the blood level during fasting and
refeeding,
the level of liver mRNA specifie for TBPA was
1
measured. This level was comparable in aIl the refed rats at
any time of the refeeding period (RI, R2 and R3)
; It was
1
decreased only in the fasted rats.
These results,
suggest
1
that a decreased level of hepatic TBPA mRNA and hence a
decreased synthesis most probably participate in the
1
decreased serum level after fasting.
but was not the only
regulatory mechanism of serum TBPA level during acute
1
short-term food deprivation and refeeding.
1
The observed changes in serum TBPA were correlated to
changes in albumin and transferrin. two plasma proteins also
1
used as nutritional markers.
A significant decrease of the
serum transferrin concentration was observed after the
1
fasting state whereas albumin was not influenced.
In view of
1
1

---

1
1
-99-
1
1
the large albumin pool,
the lack of variation in the
1
concentration of this protein might he ascribed to a shi ft
from the extra to the intravascular compartment. However,
1
while TBPA was still low after two days of refeeding,
transferrin levels increased towards normal after the first
1
day of refeeding,
suggesting that a decreased rate of
1
transferrin synthesis was probably responsible for the
decreased plasma level after fasting.
1
Recently,
i t has been demonstrated clearly that the two
main sites of TBPA synthesis are the liver and the choroid
1
plexus,
and growing evidence suggests that the choroid
1
plexus is the main source of TBPA in cerebrospinal fluid
(CSF)
(23,25).
Therefore,
the response of TBPA was
1
simultaneously measured in CSF and in serum.
The protein
level in CSF was not influenced by fasting.
Similar results
1
were observed in protein-energy restricted rats (Wade et
1
al.,
unpublished data).
Our data indicate that TBPA in CSF
is regulated independently of TBPA in serum.
Parallel
1
findings were reported in rats submitted to acute
inflammation : in these rats,
the liver TBPA mRNA level
1
decreased significantly while that of the choroid plexus was
1
not modified (25).
TBPA has been recognized as the major plasma carrier of
1
thyroxine (T4) in the rat (26,21).
In view of this
predominant role of TBPA in the binding of thyroid hormones,
1
the relationship between serum TBPA and thyroxine (T4)
1
1

---

1
1
-100-
1
1
levels and the consequence of decreased TBPA concentration
1
on total thyroxine,
free thyroxine and total
triiodothyronine (T3) were examined. Serum TBPA and total
1
Thyroxine were found to correlate markedly over a wide range
of concentrations.
The free part of the hormone (FT4) was
1
linearly related to the calculated free T4 index as weIl.
1
After one day of fasting,
the amount of total thyroxine in
the serum did not change while that of total
1
triiodothyronine was significantly decreased.
Yet, the
distribution of thyroxine between the bound and free form
1
was modified presumably as a result of the fall in serum
1
TBPA after fasting.
The reduction of the total
triiodothyronine level after the fasting state could reflect
1
either decreased thyroidal secretion,
decreased TBPA binding
or decreased peripheral conversion of T4 to T3 (28-30).
1
BALSAM and INGBAR (28) have reported that administration of
1
glucose as the sole nutritional source prevents the impaired
generation of T4 to T3 in 48 h fasted rats.
In accordance
1
with these reports,
our data on water and glucose drinking
rats also suggested that the conversion,
rather than the
1
alteration in TBPA binding to T3,
or the secretion of T4,
1
was responsible for the reduced T3 level observed in the
one-day fasted rats.
1
It is difficult to conceive that the refed rats (at RI,
R2 and R3) could have an abnormal flux of T4 into the plasma
1
(31). Hence,
the reduced TT4 levels at these stages could
1
1

---

1
1
-101-
1
1
on1y be exp1ained by disorders associated with abnormatities
1
in the concentration of TBPA.
In man,
the disorders in the
distribution of thyroid hormones associated with a decrease
1
in thyroxine-binding globulin are comparable to those
observed in this study : at Rl,
as serum TBPA was still
1
decreased, the number of unoccupied binding sites was
1
decreased,
resulting in a shift of T4 from the bound to the
free state and on an increased conversion of T4 to T3.
The
1
differences between results obtained at RI and R2 for FT4
and TT3 were Most probably explained by the higher level of
1
the food intake of the rats at RI th an at R2,
and the
1
composition of the diet given (high sucrose content)
(30).
At R3,
TT4 increased as a result of both a normal flux of T4
1
into the plasma and an increase in the number of unoccupied
binding sites (raise in TBPA level).
1
As in man,
starvation in the rat induces a modification
1
of the distribution of thyroid hormones (31).
However,
in
Many reports,
studies were performed with rat fasted during
1
two days or more.
The effect of acute short-term food
deprivation (one day of fasting) and refeeding on thyroid
1
hormones was less investigated ; moreover,
TBPA was not
1
measured in MOSt of these experiments. Our study clearly
shows that serum TBPA is highly sensitive to food
1
deprivation.
It also shows two distinct influences of
starvation on the thyroid hormone.
A primary effect of
1
fasting which greatly affects the amount of circulating T3
1
1

---

1
1
-102-
1
1
by inhibiting the conversion of T4 to T3,
and abnormalities
1
in the binding and distribution of T4 directly related to
the circulating amount of TBPA.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-103-
1
1
1
1
ACKNOWLEDGMENTS
1
1
The authors are grateful to Dr. Per A.
Peterson who
kindly provided the cDNA transthyretin insert. They also
1
thank Dr. Monique Thomasset and Nelly Gouhier (INSERM U.120)
for their good advice and expert technical assistance with
1
respect to molecular biology experiment.
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-104-
1
1
REFERENCES
1
1
1. NC-IUB and JCBN Newsletter.
(1981) J.
Biol. Chem.
256,
12-14.
1
1
2.
Ingenbleek, Y.
(1985)
In : Pediatric Adolescent
Endocrinologie (Zvi Laron,
Petah Tikva eds.), vol.
14, pp
1
345-368, Karger,
Basel.
1
3. Young,
G.A.,
Chem, C.,
Collins, J.P.
and Hill, G.L.
1
(1979) Am. J. Clin. Nutr. 32,
1192-1199.
1
4. Carpentier, Y.A.,
Barthel, J.
and Bruyns, J.
(1982) Proc.
Nutr.
Soc. 41,
405-417.
1
1
5. Felding, P.
and Fex, G.
(1985) Acta Physiol.
Scand.
123,
477-483.
1
6. Bleiberg-Daniel, F., Wade, S.,
Labarre, C., Balagny, D.,
1
Fichelle, A.,
Bory, J.,
Desmonts, J.M.
and Lemonnier,
D.
1
(1985) Ruman Nutrition: Clin. Nutr.
39C,
55-62.
1
7. Bleiberg, F.,
Vranckx, R., Wade, S. and Nunez, E.A.
(1985) BBA 828,
270-277.
1
1
1

---

1
\\
\\
1
-105-
1
1
8. Peterson, P.A.,
Rask,
L., Ostberg,
L.,
Anderson,
L.,
1
Kamwendo,
F. and Pertoft,
H.
(1973) J.
Biol. Chem.
248,
4009-4022.
1
9. Kanda, Y., Goodman,
D.S., Canfield, R.E.
and Morgan,
F.J.
1
(1974) J.
Biol. Chem.
249,
6796-6805.
1
10. Navab, M.,
Mallia, A.K.,
Kanda,
Y.
and Goodman,
D.S.
1
(1977) J. Biol. Chem.
252,
5100-5106.
1
11. Sundelin, J.,
Melhus,
H.,
Das,
5., Eriksson, il., Lind,
1
P., Tragardh,
L.,
Peterson,
P.A.
and Rask,
L.
(1985) J.
Biol. Chem.
260,
6481-6487.
1
12. Lemonnier,
D.,
De Gasquet, P., Griglio, S.,
Naon,
R.,
1
Reynouard,
F.
and Tremolières, J.
(1974) Nutr. Metab.
1
16,
15-29.
1
13. American Institute of Nutrition. Report of the AIN Ad
Hoc Committee on Standards for Nutritional Studies
1
(1977). J.
Nutr.
107, 1340-1348.
1
14. Farris, E.J. and Griffith, J.Q.
(1949) In : The Rat in
1
Laboratory Investigation (Farris, E.J. and Griffith,
J.Q. eds.),
pp 196-197, Hafner Press New York N.Y.
1
1
1

---

1
1
-106-
1
1
15.
Laurell,
C.B.
(1966) Anal.
Biochem.
15,
21-36.
1
16. Mancini, G.,
Carbonara,
A.O.
and Heremans,
J.F.
(1965)
1
Int.
J.
Immunochem.
2,
235-259.
1
17.
Itoh,
N.,
Nose,
K.
and Okamoto,
H.
(1979) Eur.
J.
1
Biochem.
97,
1-9.
1
18. Maniatis,
T.,
Fritsch, E.F.
and Sambrook,
J.
(1982) In
1
Molecular Cloning,
A Laboratory Manual,
Cold Spring
Harbor,
N.Y.
1
19. Rigby,
P.M.J.,
Diekmann,
H.,
Rhodes,
C.
and Berg,
P.
1
(1977) J.
Mol.
Biol.
113,
237-251.
1
20.
White,
B.A.,
Bancroft, F.C.
(1~)82) J.
Biol.
Chem.
257,
1
8569-8572.
1
21. Aviv,
H.
and Leder,
P.
(1972) Proc.
Natl.
Acad. Sei.
69,
1408-1412.
1
1
22. Gofferje,
H.
and Kozlik,
V.
(1977)
Infusionstherapie 4,
320-324.
1
23. Dickson,
P.W.,
Howlett, G.J.
and Schreiber, G.
(1982)
1
Eur. J.
Biochem.
129.
289-293.
1
1

---

1
1
-107-
1
1
24. Peterson, P.A., Nilsson,
S.F., Ostberg,
L., Rask,
L.
and
1
Vahlquist A.
(1974) Vitam.
Horm.
32,
181-214.
1
25. Dickson, P.W., Alfred, A.R.,
Marley, P.D.,
Bannister,
1
D., Schreiber, G.
(1986) J.
Biol. Chem.
261,
3475-3478.
1
26. Davis, P.J., Spaulding, S.W., Gregerman, R.I.
(1970)
Endocrinology.
87,
978-986.
1
1
27. Sutherland, R.L.,
Brandon, M.R.
(1976) Endocrinoly.
98,
91-98.
1
28. Balsam, A.,
Ingbar,
S.H.
(1979) J.
Clin.
Invest.
63,
1
1145-1156.
1
29. Kinlaw, W.B., Schwartz, H.L., Oppenheimer,
J.H.
(1985)
1
J. Clin.
Invest.
75,
1238-1241.
1
30. Burger, A.G.,
Berger, M., Wimpfheimer, K.,
Danforth E.
(1980) Acta Endocrinologica 93,
322-331.
1
1
31.
Ingbar, S.H., Woeber,
K.A.
(1974) In : Textbook of
endocrinology (Willam, R.H.,
ed.) pp 95-232, WB
1
Saunders, Philadelphia-London-Toronto.
1
1
1

---

- - -- - - -- - - - - - - - - - - - - -
Table 1
Effect of fasting and refeeding on body weight,
serum and
cerebrospinal fluid (CSF) TBPA concentrations.
Periods
Body weight
Serum TBPA
TBPA in CSF
(g)
(mg/100ml)
(mg/100ml)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
1
FO
209 ± 4.2 .
48.3 :t 0.9
1.16 :t 0.03
i-'
a
CD
b
b
1
F1
191 ± 2.9
38.8 :t 2.8
1.09 :t 0.05
b
R1
212 ± 3.4
35.1 :t 1.4
1.18 :t 0.05
a
b
R2
219 ± 3.4
38.7 :t 1.1
1. 20 !
0.03
b
R3
226 ± 2.7
46.6 ± 2.0
1.11 :t 0.06
Measurements were made during the following periods : before fasting (FO),
after
the fasting day (F1),
l,
2 and 3 days after refeeding (R1, R2, R3,
respectively).
a
b
Results are expressed as means ± SEM.
p < 0.05,
p < 0.01 versus the control group
(at FO). n = 6 animais for each period.

---

- - - - - - - - - - - - - - - - - - - - -
Table 2
Serum thyroid hormone concentrations in rats fasted for 24 h
and fed ad libitum for 3 consecutive days.
Periods
Total thyroxine (TT4)
Free thyroxine (FT4)
Total triodothyronine (TT3)
FT41
(lJ.g/I00ml)
(ng/l00ml)
(ng/l00ml)
FO
5.46
2.26
45
1. 12
± 0.24
± 0.14
± 4
± 0.04
b
b
FI
5.51
2.59
29
1. 45
1
± 0.16
+ 0.15
+ 3
± 0.08
f--'
-
-
0
'Ü
b
a
1
RI
4.08
2.16
54
1. 16
± 0.10
± 0.13
± 5
± 0.05
b
b
a
R2
4.08
1.63
37
1. 10
± 0.20
± 0.11
± 2
± 0.05
b
b
a
R3
6.23
2.60
63
1.35
± 0.12
± 0.11
± 1
± 0.04
Measurements were made before fasting (FO), after the fasting state (Fl),
l,
2 and 3 days after
refeeding (RI, R2, R3,
respectively).
n = 6 animaIs for each period.
Results are expressed as
a
b
means ± SEM.
p < 0.05,
p < 0.01 versus the control group (at FO). Free T4 index (FT4I) 1s the
ratio total thyroxine (lJ.g) x 100
TBPA (lJ.g/ml)

---

1
1
-110-
1
1
1
food intake
TSPA
. '(9/24h/100 9
body_weight)
1
(% va ria t i on)
1
1
1
1
o
o
(JI
~
w
t.)
~
o
o
0
0
o
o-r1
o
1
,
,
,
,
,
1
,1
1
"T'I
.....
1
1.0
1
1
r:r
1
:c
r:r
al
1:-
....
n
J
1
1
1
1
('al
1
,
1
1
,
1
1
l
' . '
:c
,
l,',' l,',
Il.
1
l
",
w
\\
1
\\
..
\\
\\
(JI
\\
\\
1
1
:c
1
0)
"
1
"
Q.
m
~
...
'<
0
Q.
'"
1
'"
1
1
1
1

---

1
1
. -lll-
1
1
1
Total
thyroxine TT4 (J.)9/1ooml)
(J'l
1
"-J
(J'l
o
.."
~.
1
1.0
N
l.rJ
0
1
0
"'""f
1
w
te
U1
"'0
0
>
1
't:
te
....
3
1
--
1
1
1
01
(J'l
o
1
1
1
1
1
1
Il

---

-112-
Fiq. 3
a b c
d
e
f

---

-113-
Fig. 4
"
l
2
3
4
'':1
e a
b
,
.
1
.0;;
'. '. ,. âl. 1 C
1
~
1
.
1
;
.....
'• • . 4t
"~
d
• •
~,
,
e
••• . '~.:': f
.' •. 1!1.''-.1 9

---

1
1
-114-
1
1
1
LEGEND Ta FIGURES
1
1
Fig 1.
Food intake and serum TBPA variations after the
short-term acute food deprivation and refeeding
1
Experimental group
Control group
1
a p < 0.05, b p <0.01 versus the prestarvation
c
1
values (at Fa);
p < 0.01 versus the control
group.
1
Fig 2.
Correlation between the serum concentration of
1
TBPA and total thyroxine (TT4). Correlation
,1
coeficient : r
= 0.6420~ n = 30, p < 0.001,
y = 1.23 ~ 0.0092 x
1
Fig 3.
Analytical agarose gel electrophoresis of the
1
509 bp PA-l (cDNA TBPA clone) inserted into the
plasmic pUC9 at the site AWAI. the plasmid was
1
isolated from the bacterial strain JM83 by
1
alkaline extraction and cesium chloride gradient
density centrifugation.
Lane a to d : migration
1
forms of the plasmid digested by AWAI restriction
enzyme
lane e
: migration forms of the plasmid
1
before digestion and lane f
: DNA molecular
1
weight
1

---

1
1
-ll~-
1
1
Fig 4.
Autoradiographs of the mRNA-cDNA TBPA hybridation
1
spots from rat liver before fasting (a,c), after
one day of fasting (d) and after one (e), two (f)
1
and three days (g) of refeeding.
From lane 1 to 4
1
10 ~g, 5 ~g,
2.5 ~g and 1.25 ~g of total RNA were
applied. mRNA TBPA was not found in the small
1
intestine (b).
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
-116-
1
1
ART l C L E N° 4
1
1
1
"Variations in plasma thyroxine-binding prealbumin (TBPA) in
relation to other circulating proteins in post-operative patients
during rapid oral refeeding".
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-117-
1
..
1
Human Nutrition: Clinical Nutrition (1985) 39C, 55-62.
Received 31 Ju1y, 1984; accepted 80ctober, 1984.
1
VARI ATIONS IN PLASMA THYROXINE-BINDING PREALBUMIN
(TBPA) IN RELATION TO OTHER CIRCULATING PROTEINS IN
POST-OPERATIVE PATIENTS DURING RAPID ORAL REFEEDING
1
F. BLEIBERG-DANIEL*, S. WADE*, C. LABARRE**, D. BALAGr\\Y***, A. FICHELLE***.
J. BORY*' J.M. DESMOr\\TS*** and D. LEMOr\\r\\IER*
1
*Vnité de Recherches sur la Nutritiun et l'Alimentatiun, V.1 INSER,\\J, **Laburatuire d'lmmllllUchimie
and ***Dépanement d'Anesthésiolugie, Hôpital Bichat, liO Bd Ney, 75018 Paris, Frallce.
1
Variations in plasma thyroxine-binding prealbumin (TBPA) were imestigated in 15 well-
nourished patients who underwent minor orthopaedic surgery and resumed normal oral
1
feeding on the first post-operative day. TBPA fluctuations were analysed together with those
of other nutritional and inflammatory markers including album in (ALB), sorne acute-phase
reactant proteins. C-reacti\\'e protein (CRP). orosomucoid also named al-acid-glycoprotein
1
(aIGP), al-antitrypsin (aIAT) as weil as cortisol and haematocrit. :\\leasurements were
conducted the day before operation. after the aclministration of anaesthesia. 2 h after the
patient regained consciousness and then dailv for a period of one week (days 1 to i).
Assays showed that TBPA and ALB levels began to decline by day 1 reaching minimum
1
values by day 3 (with 35 and 15 pel' cent total decreases respecti\\'elv compared to the initial
levels). CRP levels began to rise on day 1 reaching maximum le\\'els by clay 2. alGP and alAT
started to increase on day 1 and 2 respectively ancl displa\\'ed maximum concentrations b\\, da\\'
1
3. Cortisol. on the other hand. showed a rapid. \\'et short-li\\'ed increase after the patient
regained consciousness on the da\\' of operation. Haematocrit levels decreased from da~' 1 to
day 3 and these low values were maintained until the end of the study. Aside l'rom cortisol ancl
CRP. the initial levels of the other parameters were not \\et restored b\\, da\\' i.
1
These results show that clespite an earlv retUrI1 to normal oral feedin~ in post-operati\\'e
patients. the commonl\\' obsen'ed pattern of \\'ariation in TB P.-\\. le\\'els persisted. indicatln~ that
such a pattern seems to be mainl~' inf1uenced bv stress-incluced post-operati\\'e response'i
1
rather than bv nutritional suppl\\'.
Introductioll
1
Several clinical studies have Ïrwestigated the effect of surgery on the plasma
concentration of various nutritional and inflammatory markers (\\-Verner &
Odenthal, 196ï; Werner & Cohnen, 1969; Clarke, Freeman & Pryse-Phillips.
1
19ï 1; Aronsen et al., 1972; Ramsden et al., 19ï8; Young et al., F9ï9; Herold,
et al., 1979; Carpentier. et al. 1982; Colley et al., 1983)..Results show
that under whatever type of operation or different post-operative nutritional
1
conditions, similar patterns of \\'ariation seem to occur. There are transient
decreases in the concentrations of albumin (ALB). transferrin (Tf) and
prealbumin (TBPA) with minimal levels usually being reached by the third
1
or fourth day after the operation. However, by contrast, there are marked
rises in the serum levels of various acute-phase reactant proteins (APRP) \\\\'ith
each one following its o\\\\'n specifie time sequence. Apan from the influence
1
of several non-specifie factors studied for ALB in particular, investigators
contend that the decreases in the nutritionally related .parameters in the
patients are mainly a side-effect of the inflammatory response which induces
1
55
1

---

1
-118-
1
in the liver a preferential synthesis of the APRP (inflammatory markers) at
1
the expense of the nutritional markers namely ALB, Tf and TBPA (Young et
al., 1979; Carpentier et al., 1982). Compared to ALB and Tf, TBPA has a
very short half-life and is known to react rapidly to fasting, defi.ciencies of
1
protein and/or energy and refeeding (Gofferje & Kozlik, 1977; Shetty et al.,
1979; Kelleher et al., 1983), Hence, our objective was to see if food supply to
the patient during the immediate post-operative period wou Id not signifi-
1
cantly influence the plasma concentration of this protein.
Until now, few investigators have studied homogeneous groups of surgical
patients (with regard to the type and severity of injury and the preoperative
1
nutritional status of the patients), the consequent side-effects of the surgery
itself which can modify the plasma TBPA levels during the first post-
operative days (haemorrhage and whole blood transfusion) and the effects of
1
the operation on TBPA concentrations independently from the influence of
partial starvation. The surgical patients previously studied often received
electrolytes during the first few days post-operation before starting total
1
parenteral or oral alimentation, or were post-operatively given parenteral
feeding which supplied less energy than would a normal diet. Thus, we
seleeted a group of well-nourished patients who were to undergo minor
1
orthopaedic surgery with hopefully minimal and controlled blood losses and
who \\Vere to resume normal oral feeding the day following the intervention.
Plasma TBPA was measured simultaneously with the circulating levels of
1
ALB and APRP. Furthermore, because TBPA synthesis is increased by
glucocorticoid treatment (Oppenheimer & \\Verner, 1966; Burr et al., 1976)
and also most probably by enhanced adreno-cortical activity (Jacobsen,
1
Peitersen & Hummer, 19ï9). cortisol concentrations \\Vere also assaved.
5 uhjects and methods
1
The study was conducted in 13 men and 2 women hospitalized in an
orthopaedic department. The average age \\Vas 31 years old \\\\'ith a range
1
between 18 and 52. Selection was based on the following criteria: the patient
hacl to he in good nutritional status and not be affeeted by any pathology or
1
treatment that might influence the circulating TBPA le\\'els. These might
include such problems as thyroid dysfunction, liver disease, nephropathy,
1
carcinoma, infeetious diseases or steroid treatment. Selection was carried out
through a questionnaire, clinical examinations and biological dererminations
1
(erythrocyte sedimentation time, serum ion concentration, transansaminase
activity, plasma creatinine and urea nitrogen concentrations). The last
criterion was that the patient underwent minor surgery that wouId not
1
require blood transfusions and that he or she could rapidly resume normal
oral feeding. Out of the chosen patients, 5 underwent reparative surgery on
the brachial plexus while another 8 had peripheral nervous grafts. The last 2
1
received treatments for a recurring luxation of the shoulder and a discal
hernia respectively. In aH cases the patient's consent was obtained after
explaining the aims of the study.
1
Ali patients were premedicated. Nine received 10 mg of diazepam orally
while 6 received 120 mg of pentobarbital rectally. Ali 15 then had
1
56
1
1

---

1
1
-119-
intramuscular injections of atropin (0.5 mg). Al'ter a 12-h fast, anaesthesia
1
was induced using thiopental (7 mg/kg), succinyldicholine (1 mg/kg) and
through intubation with halothane (0.05 per cent) and nitrous oxide (70 per
cent). Phenoperidine was used as a pain-killer. In aIl cases surgery was
1
performed between 8 and 10 a.m. The most significant blood losses were
recorded l'rom the patients sllffering From the discal hernia (400 ml) and the
recurrent luxation of the shoulder (300 ml). The range for the other patients
1
was between 150 and 200 ml. AlI patients received a 5 per cent glucose
solution intravenollsly at 300 ml/ho
After the operation, patients received a standard diet of 75 g protein, 75 g
1
lipids and 225 g carbohydrates daily. On the average this supplied 478 kJ
(2000 kcal)/d. Food consumption was recorded daily. The patients started to
eat on the day of the operation, but the food supplied ta them \\Vas not eaten
1
completely. However, al'ter the first post-operative day and for the duration
of the study, aIl the patients consumed their meals completely.
Fasting blood \\Vas collected before surgery (day - 1) and then again every
1
day at 8 a.m. until at least the 4th day although preferably through to the 7th
day. In addition, blood samples \\Vere taken on the day of operation
immediately al'ter induction of anaesthesia (day 01) and then 2 h al'ter the
1
patients regained consciousness (day 02). Plasma was stored at -20 oC until
processed.
ALB, TBPA, aIGP, a,AT and CRP concentrations \\Vere assaved bv laser
1
,
,
nephelometry using a Behring nephelometer and monospecific antisera and
control sera (Behring-Werke, ~larburg/Lahn,\\\\'est German~) Cortisolle\\'els
were determined
by radioimmunoassay (Cortctk-125, Commissariat à
1
l'Energie Atomique, GyflYvette, France).
Data are expressed as means ± s.e.m. The paired Student's t-test was llsed
1
for the evaluation of the results.
Resu!ts
1
The parameters measured and their \\"ariations expressed as percentage of
the initial le\\'els are shO\\\\'T1 in the Table and in the Figure respecti\\"ely.
1
The mean haematocrit \\'alue decreased significantly l'rom day 1 onwards
reaching a fairl)' constant le\\'el bet\\Veen days 3 and ï. ~Iean preoperati\\'e
levels of ALB and TBP.-\\ were within normal limits. The concentration of
1
both proteins declined uniformly and significantly l'rom day 1. TBP.-\\ and
ALB levels decreased by 35 and 15 per cent respecti\\'el)' by day 3 compared
ta day -1. Despite marked upward trends. at least for TBP.-\\. initial
1
concentrations had not been restored by day ï.
CRP reacted more strongly than the oiher 'APRP; its response \\Vas initiated
by day .l, peaking at a 35-fold increase over initiallevels on day 2. The initial
1
concentration \\Vas restorecl by day 5. e\\'en though CRP concentrations \\"aried
widely bet\\Veen indi\\"idual"s. a IGP and a 1AT displayed clear-cut increases
within a period of 24 and 48 h respectively. l\\laximum concentrations \\Vere
1
reached on da)' 3 with 1.6- and lA-fold increases over day -1 respectively.
From day 3 to ï levels of neither protein changed significantly.
1
57
1
1

---

- - - - - - - - - - - - - - - - - - - - -
(Jo
00
Tahle. SeriaI c!UlU{.;l'S in IUlnl/a/ucri/ (lit), (/l/llilI/il/ (AUI). thWllxinl'-bill/ling prml/lUlIlill (TBPA), orusumucuù[ (a/CP), ara1ltitrypsin (a/AT),
c- l'mc/il'{' l}fo/I'il/ (CRI') (/I/d cor/i.1II1 "'l'l'II I/II//Ill'il/g or/hollll/'die .\\11Iger)' (II/mm ± s.l'.IIl.)
~!ariaMI's II//,//\\I/Iï,d
/)U)' -1
1hn/} 1
/)ar/}2
1hl)' 1
Day 2
Day 3
Day -1
Day 5
Day 6
Day 7
III ('if)
'I-I.:~
-
-
-II.H
41.0
40.0
:19.9
3~U
39.8
39.7
± O.:>
± Il.:)
± (1.()
± (1.()
± 0.7
± 0.7
± 0.8
± 0.6
1'"
**
**
**
**
**
**
**
;\\ I.B
:LHI
:~.H:\\
:U;H
:\\.:>:>
:HO
:\\.~7
:\\.:\\~
:L11
:L15
3.18
(g/IOO ml)
± 0.11
± 0.11
± 0.17
± O. 10
± 0.11
± (I.O~)
± 0.12
± 0.12
± 0.15
± 0.10
l'
II.S.
Il.S.
**
**
**
**
**
**
**
TBPA
:n.~)
:n.(;
:\\·I.H
~~).·l
25.~
n.o
~4 .()
25.0
27.1
27.3
1
(lIlg1l00 1111)
± 1.7
± 1.:>
± 1.:>
± I.~
± I.:~
± 1.1
± 1.:\\
± I.~
± O.G
± 1.0
1--"
N
l'
II.S.
II.S.
**
**
**
**
**
*
*
0
1
alCP
H:>
HI
71l
~)~ )
120
121
118
114
122
124
(III gll 00 1111)
± H
+ -
-
1
± H
± Il
± Il
± 10
± Il
± 9
± Il
± 14
l'
II. ...
II. S.
**
**
**
**
**
*
*
alXr
~~I
~ 1:\\
~II
~:>.~
:1:17
:~48
:119
313
339
317
(lIIg/100 ml)
± 1:\\
± 1()
± ~I
± III
± 17
± IH
± 14
± 17
± 18
± 17
1)
II.S.
II. S.
II.S.
**
**
**
**
**
**
CRplt
0.·1 Il
Il.Y>
(UI
4.8:>
IOAO
H.:W
4. 10
2.20
1.65
1.40
(III gll 00 111\\)
± O.llq
± Il.0:>
± O.1l 1
± I.~·I
± ~.II
± I.~)()
± 1.31
± I.Ot)
± 0.88
± 0.76
1)
II.S.
II. S.
**
**
**
*
n.s.
n.s.
n.s.
Cori iS/l1
1~O
118
~:-)~)
IHH
157
153
158
165
145
137
(fJ. gll 00 III 1)
± 1:>
± IIi
± :t\\
± Iq
± Il;
± 1:)
± 12
± IH
± 15
± 19
l'
II. ...
**
**
II.S.
*
*
**
n.s.
n.s.
Il
H
1:->
7
1:)
12
15
12
9
8
8
,1
Signilicallce or dillen'IIl'(' bl'l'Vl'l'lI pre- ;11\\(1 pOSI-opeLlIivl' lIlt'all v;t1l1es: *1' < 0.05; **1' < 0.01; n.s.: nol signilicam
It Each lillle 1Ill" CRP COI\\( elllr;l.ioli \\\\'; ... < 0.1 IlIg/IOO 1111, the vaille 0.:1 ",as IIsed l'or Ihe calcu(;lIiolls.

---

1
-121-
1
a
1
2
·100
•
1
.'.
:
...
"
·
=
.
....
·
<
"
'...,
.
.
=-
1
~ .50
1.~~
.......... "O"....,,;j 8
....
z
._~~-
_../~t Q.-lnt,trypSIA
....
1
<::>
o~~\\,
....
.
z
.....
~
\\..
~ CortIsol
<::>
<::>
.
1
!..
•......
o
~
0---
~~IAlb.lli.
1
0 - - -
' j• ••••...:'
TSPA
--::r--~~
-.-. __ •• - •• CR P
1
• 50
OP
_ _""""_ _
_ _""""_ _'--_....J.
•
~~c!
~ _ . . . . . . . L .
0,
0, 1
2
J
4
5
6
1
OAYS
AFTER OPERATION
Figure. Clu/lIgl's ill Ji III 111111 /(JII(I'/IlmlllJ/I ol ri/li//il/II/. !i/Ni/lnllllill. um,\\olfll/(uirl. Ci.rfl/llllrY!i,lIfl,
C- IN/rll,'1' !Jllill'III (CR?) rilui (orlilul rtl/rl/lg Iftl' il'l'l'k !ol/u'l'Ing llill/or ulll/()!J{/l'd,r SI/rgn)'
1
:\\mong the \\'arious proteins measured. none of them \\\\'as signifîcanth'
1
affeeted by the 12-h fast (day () 1) or by the lïrst hours of post-surgical stress
(da\\' (2). Cortisol concentration. on the other hand. reached its maximal
\\'al~le on clay 02 sho",ing a 2. 1j·folcl increase o\\'er cla\\' - 1: its le\\'els remained
1
significantly higher b\\' day 1 than by d~l\\ - 1. The preoperati\\'e \\'alue "'as
reestablished b\\' da\\' G.
1
Distllssi(J/i
In this study. the post-operati\\'e pattern of \\'ariation for nutritional and
1
inflammator\\' markers \\\\'as simi1ar to those commonl\\' obsen'ed (\\\\'erner &:
Odenthal. 196ï: Werner &: Colmen. 196~): Clarke et ai., 19ï 1; Aronsen el a/..
19ï2: Young fi a/.. 19ï9; Carpentier el al., 1982); decreases in ALB and
1
TBP.-\\. le\\'els with minimal \\'alues being reached by the 3rd day after the
operation occur concomitantly ",ith increases in APRP concentrations, the
serum CR? le\\'e! showing the most marked increase. Howe\\'er, part of the
1
decrease in the ALB and TBPA le\\'els could ha\\'e been due to haemodilution
as a result of the intra\\'enous injection of.5 per cent glucose solution as weil as
blood loss. Yet, haemodilution is not the only factor responsible for the
1
important clecrease in TBPA concentration. The TBPA concentration was
59
1
1
1

---

1
-122-
1
not influenced by the short preoperative perioc\\ of fasting. Gofferje & Kozlik
1
(1977) have observed th.ù clecrease in plasma TBPA occurs only 48 h after
fasting or severe protein-energy deficiency (15 g of essential aminoacids ancl
80 g of carbohydrates (93 kJ/d. ~~90 kcalld). Thus. it is unlikely that the
1
periods of fasting and partial protein-energy restriction immediately after
the operation could have had sOl11e effeet on TB PA le\\'el. The fact that
normal oral nutrition \\\\'as resumed on the first post-operati\\'e day \\\\'hile the
1
circulating levels of TBPA continued ta decline strongly suggests that the
nutritional conditions \\\\'ere not a cIetermining factor that coulcI explain
either the systematic decrease of this protein or its pattern of variation. ln
1
our stucIy, the cIecrease in TBPA level amountecl to 35 per cent of the initial
concentration by day 3. Despite the large cIifferences reportecI in the
literature on the percentage cIecrease in TBPA From initial le\\'els for
1
cIifferent or e\\'en sil11ilar types of operations, the values usually range from
40 to 55 per cent bel 0\\\\' preoperati\\'e levels ancl, therefore, are greater than
those observecI in the orthopaeclic patients. Consequently, this relati\\'e
1
recIuetion in the rate of cIecrease in TBPA in our stucIy may result either
from the limitecI tissue injury (minor surgery) ancI bloocI losses or by some
protein-sparing effeet introcIucecI by early refeecIing.
1
The fincIing that aclequate lllltritional suppl Y failecI to pre\\'ent the TBPA
cIrop after sllrgen' is also supported by stuclies performed in meclical
patients suflering from multiple traumas \\\\'ho benefitecl From intravenous
1
hyperalil11entation (Shenkin fi a/., 19HO). ln these situations. the pattern of
variation in protein as re\\'ealecl by nutritional and inflammatory markers
1
closely
resembJes
those obsen'ecI aher surgical stress. These authors
cIel110nstratecI that 11\\peralil11entation initiatecI the morning after operation
impro\\'ecI the nitrogen balance better than an intra\\'enous perfusion of
1
electrolytes. but that it significantly alterecI neither the percentage cIecrease
in ALB. TBPA and Tf nor their pattern of \\'ariation o\\'er time.
l'rinary losses of TBPA cannot explain the plasma fall in this protein in
1
our patients since Ramsden et a/. (197H) ha\\'e ShO\\\\'l1 that its urinary excretion
\\\\'as not enhanced aher surgery. The post-operati\\'e decrease in ALB le\\'els
has been partially ascribecl to non-specifie factors such as a shift to\\\\'ards the
1
extra\\'ascular space. especialh' that foulld at the \\\\'ound site C\\louridsen.
19(7). Such a l11ech~1l1ism has Ilot been il1\\'estigatecI for TB P.-\\ but Gll1llOt be
excIucIecI as its molecular \\\\'eighr is lo\\\\'er than that of ALB.
1
Oppenheimer et al.
(1965a)
reportecI
that after electi\\'e surgery a
diminishecl synthesis of TBPA coupled \\\\'ith its short half-life "'as responsible
for its rapid decrease, ln ail the surgieal studies reponecl in the literature
1
incIucIing this one. the clecrease in TBPA was initiatecl on the first post-
operati\\'e day whereas cluring fasting and acute protein-energy clepri\\'ation
the fall occurred onl\\' on the second day of the dietary experiment (Gofferje
1
& Kozlik. 1977). Thus. the reaction of TBPA to a surgical stress, e\\'en \\Vith a
rapid return tu normal oral feeding, was t\\Vice as rapid as that recordecl after
se\\'ere food restriction. This finding suggests that the mechanisms responsi-
1
ble for the drop in the plasma le\\'el of TBPA are different in both cases. ln
surgical trauma. at least until the 3rd post-operati\\'e clay, such a mechanism
l,
1
60
1
1
1
1
t

---

1
- -123-
1
might ue related to the inflammatory response, as the post-operative
1
elevation of sorne acute-phase proteins within 24 to 48 h coincides with the
decrease in nutritional markers, especially with that of TBPA; this assump-
tion is supported by our data as weIl as those of other investigators (Clarke et
1
al., 1971; Aronsen et al., 1972; Carpentier et aL, 1982).
The hypothesis in which the inflammatory response wou Id indirectly
produce a fall in the nutritional markers, namely ALB and TBPA (Young et
1
al., 1979; Carpentier et al., 1982) is supported by experimental studies: in
rats injected \\Vith turpentine, Northemann et al. (1983) measured the
translatable mRNAs for orosomucoid, <X2M, Tf, <XI-proteinase inhibitor
1
and ALB. For ALB, translatable mRNA dropped to about 30 per cent of
control values whereas translatable mRNA increased for the other proteins.
This result suggests that a marked and rapid enhancement of APRP
1
synthesis is compensated for by a reduction in the synthesis of visceral
proteins such as ALB. To our knowledge, similar studies have not been
performed for TBPA; however, this regulatory mechanism seems to hold
1
true for TBPA since its rate of synthesis is severely depressed during acute
experimental inflammation (Dickson, Howlett & Schreiber, 1982; Felding &
Fex, 1983).
1
In our study, no clear relationship was found between cortisol and TBPA
levels over time. Cortisol concentration exhibited a transie nt and rapid
increase which did not seem to interfere \\Vith the TBPA variations.
1
ArJ{//(l1l'll'Ilgl'Illl'l/l.I-The stuch' \\\\as supponed Il\\' a grant from the Conseil Scientifique de la
Faculte Xa\\'ier-Richat and frolll the P.R.C. f:\\SER\\[ :\\0. 128020. The authors wish to thank
1
\\[s F. Lionnet and :\\. LefeHe. and \\Ir \\1. »opon liJr their technical assistance.
RI'/I'n'lla.1
1
ArollSen. K. F.. Ekelund. (; .. Kindmark. c.o. &: Laurel!. C.B. (197~): Seqllelltial changes of
plasma proteins alter surgical trauma. SII/Ilt!. .J. Clill. Lllb. 1111'1'.11. 29, suppl. 12-t. 127-136.
Bmr. \\\\'.:\\ .. Ramsden. D.B .. CrilTiths. R.S .. Rial k. LC .. Hoilenherg. R.. \\leinhold. H. &
\\\\·ent.el. K.\\\\·. (1 ~)71;): [lIeu 01;\\ single dose ofdexalllethasone on serum concentrations of
1
tll\\Toid hormones. I.IIIIII'! 2, jH·() 1.
Carpentier. Y..-\\ .. Barthel. J. &: Brll\\ns . .J, (19H2): Plasma protein concelltration in nutritional
assesslllent. PmI'. .\\'1111', Sul'. 41, -to:)·-t 17.
1
Clarke. H.G.\\1.. freeman. T. & Pryse-Phillips. \\\\'. (1971): Serum protein changes al'ter injlll'\\'.
Clill. Sei. 40, :3:\\7-3-t-t.
Colin. L\\l.. Fleck ..-\\ .. Goode. A.\\\\'.. \\luller. B.R. & \\hers. \\1..-\\. (1983): Earh' time course of
1
th'e anlle phase protein response in man.}. CI/II. PI/ih. 36. 2ü3-207.
Dickson. P. W., Howlett. G.J. & Schreiber. G. (1982): \\Ietabolism of prealbumin in rats and
changes indllced :',' dcute inl1amll1ation. Eur.). B/uchem. 129, 289-293.
Felding. P. & Fex. G. (198:~): Factors responsible for the c1ecreased plasma concentration of
1
prealbumin c1uril.~ ;;lùte inl~ammationand fasting in the rat. Aclll Phys/u/. S(((fltl. 117, 3ï7-
~)H:~.
GofTelje. H. & K011ik, \\'. (19ïï): Proteinstatlls bei kurzfristigem Fasten und bei zufuhr
1
essentielle .. Aminosauren. IIIJilS/ulI.\\lhemp/e 4, 320·32o.l.
Herold. G .. Stephan. B. & \\lenzeL T. (19ï9): Die Spiegel der Plasma proteine TransferrÎn.
retillol-bindendes Protein 'Incl Pr~ialbumin in der postoperati\\'en parenteralell Enührung
bei ullterschiedlich dosiener Aminos~iurenzufuhr.Illfllsionsthempie 6, 12-16.
1
Jacobsen. B.B .. Peitersen. B. & Hummel'. L. (19ï9): Serum concentrations of thyrotropin.
th ~Toid hormones and t hHoicl hormone-binding proteins durillg acute ancl reco"en' stages
of idiopathie respiraton distress s\\'ndrome. Acta Pal'l/iatr. Scand. 68, 257-26o.l.
1
61
1
1

---

1
-124-
Kelleher, P.C., Phinney, S.D., Sims, E.A.H., Bogardus, C., Horton, E.,S., Bistrian, B.R.,
1
Amatruda, J.M. & Lockwood, D.H. (1983): Effects of carbohydrate-containing and
carbohydrate-restricted hypocaloric and eucaloric diets on serum concentrations of retinol-
binding protein, thyroxine-binding prealbumin and transferrin. Metabolisln 32, 95-101.
1
Mouridsen, H.T. (1967): Turnover of human serum albumin before and al'ter operations.
Clin. Sei. 33, 345-354.
Northemann, W., Andus, T., Gross. V.. fl.:agashima. M., Schreiber, G. & Heinrich, P.e.
(1983): Messenger R:"A activities of four acute phase proteins during inflammation. Febs
1
letters 161, 319-322.
Oppenheimer, J.H.,
Bernstein,
G.,
Smith, j.e.
& Surks,
~I.I. (1965a): Effect of
nonthyroidal disease and surgica1 trauma on the turnover of 11:11 labeled thyroxine-
1
binding prea1bumin (TB PA) (abstract) . .J. Clin. [Ilves/. 44, 1082-1083.
Oppenheimer, J.H .. Surks, ~r.i., Bernstein. G. & Smith. J.e. (1965b): ~letabolism of iodine-
131-1abeled thyroxine-binding prealbumin in man. ScÎl'IICI' 149, 748-751.
Oppenheimer, J.H. & \\Verner, S.c. (1966): Effect of prednisone on thyroxine-binding
1
proteins. j. Clin. E I/{locr. 26, 715-721.
Ramsden. D.B., Prince. H.P .. BUlT. \\V.A .. Bradwell. A.R., Black, E.G .. hans. A.E. &
Hoffenberg. R. (1978): The inter-relationship of thyroid hormones. \\'itamin A and their
'1
binding proteins follo\\\\"ing acute stress. Clill. Elldocr. 8, 109-122.
Shenkin. A., Keuhauser, ~r., Bergstrom. j., Chao. L., \\ïnnars. E.. Larsson. j.. Liljedahl. 5.0 ..
Schilde B. & Fürst, P. (1980): Biochemical changes associated wih se\\'ere trauma. A1/1. J.
Clin. XII/l'. 33,2119-2127.
1
Sheny, P.S., Watrasie\\\\"icz. K.E .. Jung. R.T. &James. \\V.P.T. (1979): Rapid turnO\\'er transport
proteins: an index of subclinical protein-energy malnutrition. LallCft 2, 230-232.
Werner, ~1. & Col1l1en. (;. (1969): Changes in serum proteins in the immediate postoperati\\'e
1
period. Clill. Sri. 36, 173-184.
Werner. ~1. & Odenthal. D. (1967): Serum protein changes aher gastrectomy as a model of
acute phase reaetion. J. Lab. CI/II ..\\let/. 70, 302-3\\ O.
Young, G.A .. Chem. C.. CollillS. J .P, &: Hill. G.L. (1979): Plasma proteills in patients recei\\ing
intra\\'enous hyperalimentation artel' major surgen'. A1/1. J. Clill . .\\'/ltr, 32, II ~)2-1199.
1
1
1
1
1
1
1
1
62
1
1

---

\\1.
1
il
-125-
1
ART l C L E N° 5
1
1
1
"Transthyretin as Measured in Capillary and Venous Plasma".
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-126-
038A8135
1
CLIN·MED
CW8972;28/08,10: 19
1
3/mms. 1·2
002809
1
002~ 00:16.6
2,8·AUG·86
10:30:40
1
1
(
1
Transthyretfn as Measured ln Capillary and Venous
Plasma, S. Wade, F. Bleiberg-DanÜ!l, and B. Le MouLZac
1
1
<Unité de Recherches sur la Nutrition et l'Alimentation,
U.l IN8ER..\\1, hôpital Bichat, 170 Bd. Ney, 75877 Paris
Cedex 18, France)
1
Thyt'oxio-binding prealbumin (TBPA), aIso called
transthyt'enn, has been proposed as one of the most sensi-
tive nutritional markers (Clin Chim Acta 1975;63:61-7;
1
Lancet 1979;ii:230-2) and is \\V'ÏdeIy used ta monitor the
efficiency of nutritional rehabilitation in m.alnourished pa-
tientS (Clin Chîm Acta 1975;65:61-7, Am J Clin Nuer
1
1975:28:732-9, Ann Clin Lab Sei 1982;12(3):153-621. Week-
ly sampling is usuaily required far this purpose. TBPA
concentration can be measureà in 5 JLL of serum. It is thus
1
pre!ènble ta avoià venipunC:::-JIe, especially in malnour-
ish.eci. children. Here we comcare values ofTBPA in samcles
ootalneà by venipunt::"JIe ~à by skin-punc:r:ure in yo~g
1
ci-.ilciren.
Our population sampie consisψ of 45 Senegalese ci-il-
dren \\24 boys. 21 gir~s) living in a viilage locaœd about 100
1
kI= zorn Da.i.;:ar (Senegal, West Amcal. Their ages rangeci
frem si."<: ta 43 rconth..s (mean 24.6 monr..;".5l. In the morning,
biooci was coil~..ed by venipunc:ure inm 3-mL Vacutainer
1
Tu:es cont.llning EDT_;'. then by 1Înger prick accorriing ta
the :ecoIr~enciarionsof the Commirœe on Pediatrie Clini-
ca.i Chemisr:-:.- of the Americ:m Association for Clinical
C~e~tI""; \\C:in Chem 1979:25:183-9), sœriie Manolet Lan·
1
ce!:s being useci. Blaod was collecœd inm SOO-1J.1. Beaon-
Dici:d..'"lson MiC'Otainer Tubes containing EDTA. Between
200 and 500 .ur. oÏ biooci was obtained from a singie finger.
1
Blooà scecicer.5 were keot at 4, oC for a few hours, then
cen=_'Ûgeci at 10 000 x i for 2 min in a cool, hign.,stJeeà
cen~.fuge.
1
1
1
1

---

Il
1
-127-
1
The TEPA in plasma was quantified by the radial
immunodiifusion technique of Mancini, by using an antise-
I"'UI:l against human TEPA (Behringwerke AG, Marburg,
F.R.G'). Mean plasma concentrations
1
of TBPA (mgIL) in
sam~les obtained by venipuneture and skin·punc:ture were
141 (SEM) 4.9 (range
1
70-224) and 142 (SEM) 4.7 (range 64-
224), respectively. No signiiicant clliference was
1
observed in
the paired samples obtained by bath techniques.
Thus, one aï the most sensitive markers of nutriticnal
stat".:s can he determined accurately from a -finger skin·
1
punc::'.JIe. 'This teclmique, well accepted by the children and
their mcthers, is particularly suitable in young children,
especi:illy when frequent sampling is required. Skm-punc.
1
t'Ure is recommended in developing countries because it does
net aggravaœ anemia, which is comman, even in apparently
heslthy children.
1
\\
1
1
1
1
1
1
1
1
t
1
1
1
1
1
1
1
1
\\

---

'1
1
-128-
.
1
1
1
1
i
1
1
1
1
1
1
1
CHAPITRE II
1
THYMULINE DANS LA MALNUTRITION PROTEINO-ENERGETIQUE
1
1
1
1
1
1
t
1
f1
'1
1
1
1
1
1

---

1
-129-
1
1
ART l C L E N° 6
1
1
1
"Absence of variation in facteur thymique sérique activity
in moderately and severely malnourished Senegalese children".
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-130-
1
Absence of variation in facteur thymique sérique
1
activity in nloderately and severely malnourished
Senegalese children 1.2
1
Bernard Maire, PhD, Salimata Wade, PhD. Fanny Bleiberg. PhD, Mireille Dardenne, MD,
Gérard Parent, MD, Patrice Le François, PhD, and Christian Caries
1
ABSTRACT
Facteur thymique sérique activity was evaluated in relation to differenttypes of
malnutrition in Senegalese chiJdren aged 5 to 42 months. They were c1assified in four groups:
controls. moderate malnulrition. marasmus. and kwashiorkor. according to anlhropometric meas-
1
uremenls and c1inical examination. The two latter groups were characterized by very depressed
levels of total protein. albumin. lransferrin and prealbumin. and by high cortisol concentrations.
Zinc stalus was marginal in ail cllildren. Facteur thymique sérique activity. determined by the·
rosette assay. was normal in the malnourished patients suggesting that moderate as weil as severe
1
malnutrition is not necessari1y associated with depressed levels of circulating thymic hormone.
These results are discussed in relation to zinc stalus and infections.
Am J C/in Nutr
1982:36: 1129-1133.
1
KEY WORDS
PrOlein-energy malnutrition. facteur thymique sérique activity. zinc. cortisol
1
Introduction
decrease in FTS activity in energy-restricted
rats (6), in malnourished children (7), and in
1
The immune responses of children suffer-
small for gestational age infants (8). Ho\\vever.
ing from protein-energy malnutrition have
the effect of malnutrition on the immune
been extensivelv investigated during the past
response depends on the severity, duration.
few years. Protéin-energy malnutrition often
1
and type of malnutrition. and on the specific
results in the impairment of functions related
nutritional deficiencies involved. Thus. a de-
to cell-mediated immunitv (1) but the mech-
crease in FTS aetivity has been reported in
anisms involved are unknown. Previous stud-
rats and mice fed a zine-deficient diet Jnd in
1
ies have indicated that thvmic hormones
pyridoxine-restricted rats (9. 10). Studies car-
probably rlay a role in the d'ifferenciation of
ried out in malnourished human subjects
T cell subpopulations. ln Nigeria. Olusi et al
have not taken into account the influence of
(2) have observed that thymosin fraction 5
1
the severity and type of malnutrition on FTS
increases the low percentage of E rosettes in
activity.
children suffering from severe kwashiorkor.
The present work was undertaken to deter-
Similarly. Jackso';, and Zaman (3). in Bang-
mine the FTS activity in children suffc:ring
1
ladesh reported an inerease in T cell rosettes
from marginal to severe malnutrition in re-
\\Vith lhymopoietin in 10 of 16 marasmie or
lation to the plasma protein pattern. the zinc
marasmic-kwashiorkor childrcn. These au-
level. and the concentration of cortisol which
thors have raised the possibility to use thymie
is known to be Iympholytie (II).
1
hormones in the treatment of malnourished
children to restore the impaired immune re-
1 From the UI INSERM. Hôpital BICHAT. 170 Bd
sponses. A eireulating thymie hormone. fac-
Ney. 75018 Paris. France: ORSTOM and ORANA. BP
1
teur thymique sérique (FTS) has been iso-
2089.
Dabr.· Sénégal:
U25
INSERM.
Hopital
bted by Bach et al (4). FTS aetivity has been
NECKER. 161 rue de Sèvre~. 75015 Pan~. France.
~ Address reprint requesls
found in the serum of various spccies includ-
10: Bernard Maire. OR-
STOM and ORANA. BP 2089. Dakar. Sénégal.
ing man. and its presence is c1early thymus
Received January 21, 1982.
1
dependent (5). Recent studies have shown a
Acceplcd fM publication June 1. 1982.
The Americaff JOl/m1l1 nf Clifficnl NllIrilioff 36: DECEMDER 19112. pp 1129-1133. Printed in USA
1L!9
1
© IlJR2 American SClciely for Clinical NlIlriliCln
1
1

---

1
1
-131-
1130.
MAIRE ET EL
1
Patients and methOlls
FTS inhibilor). The ullrafihrales were incubaled for 90
min al 37°C wilh spleen ceHs from mice IhymeclOmized
1
Patients
10 to 15 days before Ihe lesl and Az (\\0 !.d/ml) was
added simullaneously. Al the: end oflhe incubation lime
Fony-seven Senegalese children whose ages ranged
12 X 10" sheep red blood cells were added 10 Ihe cells.
from 5 1042 monlhs, were invesligaled. 33 oflhe children
This cellular preparalion
were hospilalized for various diseases including malnu-
WolS centrifugcd (al 4oC 150 X
1
g, for 5 min) and resuspended by low spe:e:d rOlalion (10
Irition. They were classified as moderalely or severely
rpm) on a roller. Roselle-forming cells were counled in
malnourished using Ihe Walerlow's c1assificalion (12).
a hemocYlOmeler. ln Ihe presence of FTS. Az inhlbiled
The severe cases were discriminaled in k.washiork.or or
rose Ile formation al a conccOlralion of 10 I-lg/m1. The
marasmus according 10 a clinicaJ examination. The con·
highesl dilulion of Ihe plasma sam pIe which induced the
1
Irols were selected among children who belonged to the
same families as the patienls.
roselle inhibition by Az WolS considered as Ihe active
dilution. Ail Ihe delerminalions were carried out in
Group l-contro/s. Their weighl for heighl WOlS more
duplicale. and the results were expressed as log.2 recip-
Ihan 80% and height for age over 90% of the NCHS
rocal titer.
1
slandard (13).
Group l-moderale/y ma/nourished chi/dren. Their
SlaliSlica/ ana/ysis of daca
weighl for heighl ranged from 70 tO llO% of Ihe same
slandard with a height for age below or more than 90%
Differences in mean values between the groups were
of Ihe reference value.
delermined by analysis of variance.
1
Group 3-marasmic chi/dren. They showed a very
reduced subcutaneous fal tissue and muscular alrophy.
Results
Their weight for helght was under 70% of the NCHS
slandard and their helghl for age ranged from ilS tO 95%
The average age was similar in the three
of the reference value. indicating an acule or chronic
groups of malnourished chiluren: 17.5 ± 2.12,
evolution.
Group 4-kwashiorkor chi/dren .....ilh or
19.1 ± 2.47. 20.1 ± 3.78 months in groups 2,
WÙ}WUI skin
alld hair changes. They were ail edemic. Their weighl ror
3. and 4, respectively. Compared lO the con-
height ranged between 63 and 80% of the NCHS stan-
trol group (9.6 ± 1.09) they were significantly
dard. Only one patient WOlS stunted (height for age less
aider (p < 0.05).
than 90% of the slandard).
The varia us degrees of malnutrilion in our
No clinical signs of vltamlO deliciency were noliced
except a gingivitis in one child wllh kwashiorkor. Anemia
subjects are ref1ected in lhe values oblained
assoclated wilh hypos,deremia WolS present in 40'è of the
for plasma proleins (Table 1).
children. in controls as weil as in malnoufl,hed groups
ln moderalely malnourished chilJrcn. only
without any difference 10 prevalence.
lhe level of prealbumin was reduced. whc:reas
Most of the subjecis. Including the controls. were
1
suffering from various Înfecllons. mainly re,piralllry and
the marasmic and kwashiorkor groups ex-
gaslrointesllnal infections. Four marasmlc clllidren dled
hibiled very depressed values fûr .dl the pa-
durill!l hospllallzallon from grave. Jcute. supeflmpo,ed
rameters lested. compan~d la the control
infections (myocarditis. mea,les. and broncho pneumo·
group. The level of p\\;lsma prolein~ ([oul
1
patllies).
prOlein,
albunun,
anu
translàrin)
w:.l~
slighlly highc:r lhan expecteu in lhe marasmlC
MecllUds'
children. Thùe was no difference in lhe Illean
1
Blood was wllhdrawn by vempuncture bdore any
plasma prolein values belween lhis group ,lnd
nutritional and antl-inli:cllous treatmem was >Ianed.
the kwashiorkor group.
Blood was collccted 10 cooled (4°C) hepanmled lubes
between ~ and 10 A~t to take 1010 account Ihe (IrcaJian
The plasma zinc concentra lions ùf groups
rhythm of COrll,ol. Pla,ma was Immediately .eparated
2 (629 ± 52.1 /l/l) and J (700 :t 66.5) were
1
and kepl frozen at -20°C umil analyzed.
comparable [0 thosc of lhe conlrals (055 :t
The total protclO concentrallon ..... as delermineJ U,IO!l
90.7). Thc decrcased zinc Ieve! observed in
Ihe biurel melhod and plasma albunlln. tran,fcrfllI. and
group 4 (400 ± 116.3) was not stalislically
prealbumin levds according III fhe radiat Îmmun,'diffu·
1
sion method of Mancini et al (14).
signiticanl.
Plasma ZIOC was assayed by atùmic absorption spec-
The cortisol concentralion was higha 1I1
tropholometry.
the severel)' malnourished children (groups J
Cùnisol WOlS measured by radioinullunoassa y (K il
and 4) lhan in lhe other groups. 1I0WCYCf. Ihc
1
Conctk 125-0ris. CL",-. Gif·sur-Yvelle. Francel.
FTS activity was ,!uanlilied in plasma accordlOg
only significant dirfcrcnce was obscrved tll
III
Ihe method prevlOusly Jescnbcd by Dardcnne and Uach
group 4 compared lO group 1 {Fi~ L).
(15): Ihis assay analyzes Ihe conversion of relatlvely
Figure 2 shows the plasma Ievels of fTS
1
azathioprine (Al) resistant spleen cells of adult thymec-
aClÎvity. When expresseu as log-2 rccipmcal
tllllliled nuce tù e pùsillve ro,elle-forming ceHs Ihat are
tÎlcrs lhe values recorued for normal subJccts
more sensitive III Al. llnel1y. plasma samplcs \\\\ere Iii-
lered by centnfugatiùn al 4°C
ral1ged belween 4.5 and 7
,lIl Cf 50 AmÎ.:l\\n mem-
(IllC:11l 5.7 :t 0.2U).
branes (nlOlccular weighl cul off al 50,000 III remù\\e an
ln moue ra le anu severe malnUlriliùn {maras-
1
1
1

---

1
1
-132-
FTS ACTIVITY IN MALNUTRITION
IIJI
1
TABLE 1
Plasma protein levels· of Senegale~e children grouped according
tO differenttypes of malnutrition
Conlrol. (n - 14)
Moderale malnulfilion (n - 14)
Marasmus (n - Il)
KwashIorkor ln - 8)
1
4
Total prolein
67.4 ± 2.15
62.9 ± 2.32
56.8 ± 2.55t. t
51.8 ± 4J5t. t
Albumin
33.3 ± 1.58
JI.5 ± 1.58
22.2 ± l.13t. t
19.9 ± 1.38t. t
Prealbumin
0.14 ± 0.015
0.11 ± O.OOlt
0.08 ± O.004t
0.09 ± 0.oo8t
1
Transferrin
J.2 ± 0.25
2.9 ± 0.23
1.5 ± 0.27t. t
1.2 ± 0.33t. t
• Values e~rressed per g/I (mean ± SEM).
1
t p < 0.05. versus the group 1.
t p < 0.05. versus the group 2.
30n
Discussion
1
For practical and ethical reasons. well-
r
nourished patients were selected among sib-
1
lings. Reference values for sorne plasma pro-
1
teins given by the manufacturer (Behring-
(O[
werke
AG
Marburg.
Germany) indicate
slightly lower values below 1 yr of age. Our
~
1
controls were younger than the malnourished
1
u
100~
groups so that the differences observed were
1
not age dependent.
Prealbumin was the only protein that al-
1
lowed a distinction between normal and mod-
aL
erately malnourished children. This finding
2
3
4
is in keeping with the results of Ogunshina et
Groups 01 chlldren
al (16). However. its level was not further
1
FIG. 1. Plasma cortisol levels of chilJren with vary-
decreased in groups 3 and 4 in contrast wilh
in!; Jegrees l,f malnutritiun. Groups of children: 1. con-
total protein. albumin. and transferrin values.
tmls: 2. l11l'uerate malnutrition: 3. marasmus: 4. kwashi·
orkor. /l.leans + SEM are shown .• p <
The kwashiorkor cases were similar to those
0.01 versus the
1
control grour: t p < 0.05 versus the moderately mal-
described by Lunn et al (17) in the Gambia
nou risheu grou p
where infections are precipitating factors for
the appearance of edema in chtldren who
Br
1
would otherwise be regarded as cases of ma-
1
7 ~
rasmus. This explains the rather high levels
1
i
of cortisol observed in these children. The
r:;-
6f-
...
.",
- ...-
fact that the levels of the various plasma
:§.
5~
proteins were lower than expected in the
1
i!:
1
marasmic children (group 3) suggested a pos-
:2
4~
ü
sible evolution to a marasmic-kwashiorkor
'"
(,f)
state. As previously reported for normal hu-
1-
1
....
mans. fTS found in the sera of children less
:~
than 20 yr of age was active between the
1L
concentrations of 1/16 and 1/28 (18. 19). Our
1
ol
control values for FTS activity paralleled this
COO1'OIS
Moderale
Marasmus
KwashIorkor
finding. In our patients, moderate and severe
ma'nulnlron
malnutrition were not associated with a de-
FIG. 2. Plasma FTS activit)' in normal and mal-
nourished children.
crease in FTS activity. In contrast. Chandra
1
(7) reported elevated FTS titers in his controls
mus and Kwashiorkor) FTS activity was sim-
and reduced levels of FTS activity in children
ilar to thal of the control group (5.6 ± 0.22.
suffering from protein-energy malnutrition.
1
6.0 ± 0.17. 5.1 ± 0.44, in groups 2.3. and 4,
Il is known that protein-energy malnutri-
respectively.
tion and high cortisol (or corticosterone)
1
1
1

---

1
1
-133-
1
1132
MAIRE ET EL
1
levels may induce thymic atrophy. The strict
our patients. The mean zinc concentrations
thymic dependency of FTS, already shown
reached borderline values in groups 1,2, and
1
by its disappearance aCter thymectomy and
3, and feU below normal levels in group 4
its presence in thymic extracts (20) was re-
(29). Yet, the variations between the individ-
cently confirmed using indirect immunofluo-
ual kwashiorkor children did not result in a
rescence or immunoperoxidase binding to re-
statisticaUy significant difference between the
1
ticuloepithelial cells of an antibody produced
four groups. When we looked at FTS levels
against synthetic FTS (21). Similar findings
in individual patients in relation to zinc con-
have been reported for human (22) and mu-
centrations, we did not observe any correla-
1
rine thymuses, using anti-FTS monoclonal
tion of (r = 0.12). However, Iwata et al (9)
antibodies (23). The localization of FTS in
reported that a state of marginal deficiency
thymic epithelial cells was further confirmed
in mice did not lower their FTS activity while
by immunoelectron microscopic studies (24).
a 0% zinc deficient diet markedly reduced the
1
In malnutrition, various reports have indi·
FTS level. The rosette assay used to evaluate
cated that the thymic involution involves
the FTS levels is based on the biological
mainly the lymphoid part; in a postmortem
activity of this hormone. Natural as well as
1
study of kwashiorkor, Mugerwa (25) ob-
synthetic FTS are active only when sorne
served weU-preserved thymic epithelial ceUs.
minerais, particularly zinc, are bound to the
Therefore, a normal FTS activity would not
peptide: FTS loses its biological activity after
be inconsistent with protein-energy malnutri-
passage on a chelating agent and recovers it
1
tion.
aCter addition of zinc (30). So, the possible
In addition to their lympholytic effect, glu-
influence of the zinc status in the mainte-
cocorticoid hormones might also inl1uence
nance of normal circulating thymic hormone
1
the circulating thymic hormone. Steriod
must be considered although the minimum
treatment of normal mice results in a drop of
zinc level required is not yet established.
FTS activity; however, this effect is induced
The effects of protein-energy malnutrition
by high doses and is transient (26). Under
on immunological processes are often con-
1
physiological conditions, adrenal corticoid
flicling. According to the nutritional, sani-
hormones do not seem to interfere with FTS
tary, and cultural environment, numerous
levels since the high cortisol concentrations
factors may produce heterogenous results.
Our study demonstrates that moderate as
1
recorded in the marasmic and kwashiorkor
children were not negatively correlated with
weU as severe malnutrition induding maras-
the levels of circulating thymic hormone.
mus and kwashiorkor were not a;sociated
The normal FTS activity in moderately
with a decrease in FTS activity. One nuy
1
and severely malnourished children might
hope that further experimental studies WOll!l!
possibly be explained by the status of their
be useful to clarify this point and we sugge~t
concurrent infections. Studies have shown
that the unaffected level of circulatine. hor-
.1
that activated T cells secrete an a//ogenic
mone might be explained eith~r by th~ zinc
factor which is active in the rosette assay (27,
status of the children or by the stimulalOry
28). Our subjects sustained many infectious
effect of their concurrent infections.
~
diseases, therefore, we assumed that the pro-
1
duction of this factor in these children might
References
artificiaUy enhance the FTS activity. This
1. B~ll RG. Undernutrilion. inf~clioll anJ immunilV:
could probably explain the discrepancy be-
th~ rol~ of parasites. Papua N~w Guinea McJ'J
1
tween our results and those of Chandra (7)
1978;21:43-5.
who did not mention any apparent acute
2. Olusi SO, Thurman G B. Goldslein AL. Efr~cl of
lhymosin on T lymphocyte rosene lormalion in chil·
infections in his patients.
drc:n wilh kwashiorkor. Clin lmmunol Immunù-
However, from our point of view, the most
pathol 1980; 15:687-91.
critical factor to be considered is the possible
3. Ja.:kson TM, Zaman SN. The in vitro effeci ùf Ih~
role of a zinc deficiency in malnourished
thymi.: factor thymopoc:ilin on .1 subpopulatiùn of
children. Rats and mice fed a zinc-deficient
lymphocytes from severely malnùurishc:d duldren.
Clin Exp Immunol 1980;39:717-21.
diet show depressed FTS activity (9, 10). ln
·1
4. Bach JF. Dardc:nne M, Plc:au H.\\. Biochemi.:al char-
contrast to Chandra's report (7), we have
acterization of a serum thymi.: factor.
Nalure
determined the plasma zinc con~entration in
1977;266:55-6.
1
1

---

1
1
-134-
1
FTS ACTIVITY IN MALNUTRITION
1
1133
5. Bach J F. Dardenne M. Pleau J M, et al. Isolation.
18. Bach JF. DanIenne M. Papiernik M, et al. Evidence
biochemical characteristics. and biological activity of
for a serum factor secreted by the human thymus.
1
a eirculating thymic hormone in the mouse and in
Lancet 1972;2: 1056-8.
the hum an. Ann NY Acad Sei 1975;249: 186-210.
19. Iwata T, Incefy GS, Cunningham-Rundles S, et al.
6. Heresi G. Chandra RK. Effects of severe calorie
Circulating thymic hormone activity in patients with
restriction on thymic factor activity and lymphocyte
primary and secondary immunodeficiency diseases.
1
stimulation response in rats. J Nutr 1980;110: 1888-
Am J Med 1981;71:385-94.
93.
20. Dardenne M. Pleau JM, Blouquit JY and Bach JF.
7. Chandra RK. Serum thymic hormone activity in
Characterization of facteur thymique sérique (FTS)
protein-energy malnutrition. Clin Exp Immunol
in the thymus. Il. Direct demonstration of the pres-
1
1979;31\\:228-30.
ence of FTS in thymosin fraction V. Clin Exp Im-
8. Chandra RK. Serum thymic hormone activity and
munol 1980;42:477-82.
cel1 mediated immllnity in healthy neonates. preterm
21. Monier JC, Dardenne M. Pleau JM. et al. Charac-
infants. and smal1-for-gestional age infants. Pediat-
terization of facteur thymique sérique (ITS) in the
ries 1981;67:407 -Il.
thymus. 1. Fi:t:ation of anti-FTS antibodies on thymic
1
9. (wata T. lncefy GS. Tanaka T, et al. Circulating
reticulo-epithelial ceUs. Clin Exp Immunol 1980;
thymic hormone levels in zinc deficiency. Cell Im-
42:470-6.
munol 1979:47: 100-5.
22. Jambon B. Montagne B, Bene MC, et al. Immuno-
10. Chandra RK. Heresi G, Bing AU. Serum thymic
histological localization of "facteur thymique sé-
1
factor activity in deficiencies of calories. zinc. vi-
rique" (FTS) in human thymic epithelium. J Im-
tamin
A and
pyrido:t:ine.
Clin
Exp
Immunol
munol 1981;127:2055-9.
19R0:42:332-5.
23. Dardenne M. Pleau JM. Savino W, et al. Monoclonal
Il. Harris AW. Ba:t:ter JD. Variations in cellular sensi-
antibody against the serum thymic factor (FTS).
1
tivity to glucocorticoids: observations and mecha-
lmmunol Lett (in press).
nisms. In: Glucocorticoid hormone action. Berlin:
24. Schmitt D. Monier JC, Dardenne M. el al. Cylo-
Springer- Verlag.. BHler and Rousseau. 1979:423-48.
plasmic locaiization of FTS (facteur thymique sé-
12. \\Vaterlow Je. Classilication and definition of pro-
rique) in thymic epithelial cells. An immunoelectron-
microscopical study. Thymus 1980;2:177-86.
1
tein-calorie malnutrition. Br Med J 1972;2:566-9.
13. ~ational
Center For Health Statistics: NCHS.
25. Mugerwa JW. The Iymphoreticular system in kwash-
Growth charts monthly vilal statistics report. Vol 25.
iorkor. J Pathol 1971;105:105-9.
no J. <lJppl (liRAI 76-1120. Rockville. MD: Health
26. Bach JI", Duval D. Dardenne M. et al. The effects of
Resources Auministration. June 1976.
steriods on T cells. Transplant Proc \\975;7:25-30.
1
14. Mancini G. Carhonara AO. Heremans JF. Immu-
27. Dardenne ~. rlach JF. Demonstration and charac-
nochemlcal Guanlitatiun of antigens by single radial
terilation of a serum factor produced by actl\\'aled T
immunodlffuslon. Int J lmmunochem 19f15;2:235-
cells. Immunology 1977:33:643-51.
59.
28. Safai R. DanJenne M. Incefy GS. et al. Circulating
1
15. DJrdenne M. rlach .IF The sheep cell roselle assay
Ihymic factor. facteur lhymiGue sérique (FTS), ln
for the evalualiL'n of thvmic hormones. Koovker.
mvcosis fungolùes and Sezarv syndrome. Clin Im-
R(ltterdam~ Van Ilekkum: 1975;235-43.
.
m~nol Imm~nopathol 1979;13:402-6.
16. O~unsillna sa. Hussain MA. Plasma thyroxine
29. Unùerwood EJ. Zinc ln: Trace elements in human
1
blmJlng prealhllmln as an index of mild proteln-
and animai nutrillon. New York: NY: Academic
energy malnulntll'n ln Nigerian chituren. Am J Clin
Press. 1977: 196-242.
l';lIlr 1980:}J:7'l4-~n().
30. Dardenne M. Pleau JM. Lefrancier P. et al. RC,le du
17. Lunn PG. Wilileheaù RG. Coward WA. Two path-
llnC et d'aulres métau.\\ dans l'actiVité biologique du
ways to kwasI1ll1rkC'r? Trans Roy S0C Trop Med Hyg
FTS (thymuline). RC Acad Sci Paris 1981;292:
1
1979;73:438-44
793-6.
1
1
1
t
1
1
f
1
1

---

1
1
-135-
1
ART l C L E N° 7
1
1
1
"In vivo and in vitro studies of thymulin in marginally
zinc-deficient mice".
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
,
1
-136-
:1
1
Eur. J. Immuno1. 1984. 14 : 454-458
1
te
1
Mireille Dardenne.o., Wilson Savino.o.,
In vivo and in vitro studies of thymulin ia marginally
Sali \\\\ ade+, Dominique Kaiserlian.o.,
Daniel Lemonnier+ and Jean-François
zinc-deficient mice*
Bach.o.
1
INSERM U 25, Hôpital Necke~ and
Thymulin (or serum thymic factor, FTS-Zn), a well-defined thymic hormone previ-
Hôpital Bi~hat+, Paris
ously shown ta be a nonapeptide binding the metal zinc, was studied in mice subjected
to a long-term marginally Zn-deficient diet. In spite of the absence of thymic atrophy.
1
we observed a significant decrease in the serum Ievels of thymulin as early as two
months after the onset of treatment. However, these levels could be consistently
restored after in vitro addition of ZnCI!. The analysis of thymuses from Zn-deficient
mice showed that, despite the apparently normal network of epithelial cells. there was
1
a progressive increase in the number of thymulin-containing cells (assessed by
immunofluorescence with anti-thymulin mono:!on:-.i antibodies) that was already sig-
nificant after two mon~hs of treatment. Thes~ results are in keeping with those of
previous investigators, showing a specific. altered, thymic endocrine function follow-
ing Zn de privation. Nonetheless. our results strongly suggest that the nonaetive Zn-
1
deprived peptide is secreted under these experimental conditions. Furthermore, the
fact that the augmented numbers of thymulin-containing cells were observed in the
thymuses following a decrease in the peripheral thymulin (biologically active) brings
further evidence for the existence of a feedback mechanism for the secretion of tbis
1
hormone.
1 Introduction
binding to the molecule [9]. The binding affinity was calcu-
1
lated ta be around 1~ M in equilibrium chromatography [10].
There is a multiplicity of information indicating that Zn!+
The interaction between Zn:+ and the peptide was further
exerts a powerful and apparently specific influence on the thy-
suggested by electron probe microanalysis demonstrating the
mus, on T lymphoC}1es and on cellular immunity, in both
presence of the metal ion in thymic reticulo-epithelial ceUs [91
1
human and animal systems (recently reviewed by Schloen et
which are known to contain the hormone [Il].
al. [1]. Good [2] and Bach [3]).
These results indicated the existence of two forms of the hor-
Il is apparent that Zn!+ deprivation induces a marked thymic
mone: the first one deprived of Zn> and biologically inactive.
1
atrophy, whose effects closely resemble those of thymectomy:
the second one containing Zn~+ and biologically active. for
either neonatal thymectomy. or young adult thymectomy
which we proposed the name of thymulin (or FTS-Zn).
when it is imposed in adulthood [~J. More recently. several
investigators evoked the possibility that the epithelial function
To further investigate the relationship between thymic hor-
1
of the thymus may be specifically altered by this nutritional
mone a?d Zn>, we examined the consequences of a margin-
deficiency. They reported that a Zn~+ deficit induced a signifi-
ally Zn-- -deprived diet on the serum level and thymic eplthe-
cant lowerin!! of the serum level of thvmic hormone. thvmulin
liai content of the hormone. in the absence of sigruficant thv-
(previously ;alled serum thymic factôr. FTS), both in ~nimal
mic atrophy.
- .
1
models such as mice and rats [5. 61 and in patients with com-
mon variable immunodeficiency [7] or nephrotic syndrome
In this study. we report our observations concerning the alter·
with zinc deficiency [8]. ~10reover. Zn administration restored
ations of thvmic hormone secretion and T cell dvsfuncnon in
normal hormone levels. However. these results remained
mice maintiined on such a margmally Zn:- -defici~ntdiet for a
1
unexplained until recent data from our laboratory prov~ded a
period of 6 months.
molecular basis to the putative relationship between Zn-- and'
thymulin. We showed that thymulin. utilized in its synthetic or
natural form loses its biololtical' actiVÎtv in a rosette assav after
treatment with a cheJating-agent. Ch~lex 100. This activity is
2 Materials and methods
1
restored by the addition of Zn~· salts and. ta a certain extent.
other metals. Experiments using 'radiolabeled Zn~- and thr'
2.1 Chemicals
mulin revealed that peptide activation is secondary to Zn'-
Hanks' medium and sheep red blood cells (SRBC) were
1
[1 ~3761
obtained from Pasteur Institute. Azathioprine (Azl. used in ItS
• This work was supported by INSER:>.f and Fondauon pour la
sodium salt form. was obtained from Burroughs Wellcome
Recherche Médicale.
(Research Triangle Park. NC). Zinc chloride (Zna~) of the
highest reagent grade was purchased from :Vlerck. Darmstadt.
1
Correspondence: Mireille Dardenne. I:"SER:>'f U 15. H0pital Necker.
FRG. Metal ion-chelating agent. Chelex 100. was produced by
161. rue de Sèvres. F-7S730 Paris Cede~ 15. France
Bio-Rad Laboratories. Richmond. CA. Anti-thymulin mono-
Abbreviations:
FTS: Serum thymlc [actor'
clonal antibodies were obtamed bv cell fusion accordinl! to a
Az: Azathiopnnc
T\\:
Thymeclomiled PUS: Phosphalc-buikrcd saline Ga)l: Goal anli·
method previously reported [121.' Goat ann-rabbit immuno-
1
mouse GaR: Goat anti-rabbil FITC: F1uorescein iSOlhiocyanah:
globulins bound to l1uorescein isothiocyanate (FITC-GaR)
SRFC: Spontaneous rosette-iormmg cells SRUC: Shcep rcd blood
and purified goat anti-mouse IgG~, serum bound to FITC
ceIls
(FITC-GaM [[gG~,]) were provided by Nordic Laboratories.
1
1
OO14-29110/R·W505-~S4S():.50 0
~ "criaI! Chemie GmhH. D-(,Q40 Wcinhcim. I<JR4
\\
1
1

---

1
1
-137-
1
Eur. J. Immuno!. 1984.14:454-458
Thymulin in zinc·deficient mice
455
Tilburg, The Netherlands, and absorbed with rat tissue pow-
functional effeets of Zn2+ deprivation on the T lymphocyte
der prior to use. Anti-keratin antiserum was obtained from
1
spleen population [16].
Oinisciences, Sées, France.
2.5 Chelation and metal treatment
1
2.2 Animals and diets
Sera (100 J.Ù) from Zn2+-deprived and normal or Tx mice. were
Three-week-old C57BU6 femaJe mice were obtained from a
incubated for 30 min at room temperature with an equal vol-
commercial source (IFFA-CREDO, L'Arbresle, France).
ume of Chelex 100 at 50 mg/ml in distilled water. At the end of
1
They were divided into two groups and rested for 1 'week
the incubation, the mixture was centrifuged at 12000 x g for
before being fed ad libitum a basal diet with adequate (52
2 min to eliminate the chelating resin and the biological activ-
ppm) or
ity in tbe supematant of Chelex 100-treated serum was mea-
mar~nally deficient (7.7 ppm) amounts of dietary
Zn2+. The Zn + content ofthe diet was determined by' atomic
sured. Ten I4g of ZnCI2 were tben added to 100 J.Ù of Chelex-
absorption spectrophotometry. All mice were housed in plas-
tfl.::ated serum. The mixture was then incubated for 15 min at
1
tic cages, 4 per cage at an ambient temperature of 22
room temperature and its biological activity was measured in
± 2oc.
the rosette assay.
Precautions were taken conceming the Zn2+-deficient group
to avoid contaminations with Zn2+. This group was housed in
polycarbonate cages with stainless steel bars and covers.
1
2.6 ImmunoOuorescence studies
Pathogen-free conditions were maintained and deionized dis-
tilled water was available ad libitum. The cages, feeding jars
Immediately after killing of each animal, the thymus was fro-
and water bottles were washed with deionized water to remove
zen and stored in liquid nitrogen. Unfixed cryostatic sections
all residual Zn2+. Dietary intake by boch groups of animals
1
(2 l4II1 thick) were incubated with an anti-thymulin monoclonal
was monitored after 6 weeks of dietary treatrnent.
antibody of the IgG2.1 subclass. Previous studies have demon-
strated the strict specificity of tbis antibody for a subpopula-
tion of epithelial cells in mouse thymus [11]. After being incu-
2.3 Serum quantifications
1
bated for 30 min at room temperature with the anti-thymulin
monoclonal (used at the concentration of 0.5 mg/ml) sections
Mice were bled from the orbital sinus and killed by cervical
were washed in phosphate-buffered saline (PBS) for 15 min
dislocation between 9:30 and 10:20 a.m. at different periods
and exposed to mC-GaM (lgG2.1) used at a final dilution of
during dietary Zn2+ treatrnent: 2. 9, 18 and 30 weeks. Blood
1
1120. Following a second washing in PBS, the slides were
was collected in cooled tubes and sera were kept frozen until
mounted in a g1ycerollPBS solution. The number of thymulin-
analvzed. Zn2+ serum levels were assayed by atomic absorp-
containing cells in each thymus was assessed by counting 300
tion .spectrophotometry on O.l-ml aliquots of serum after 9
microscopic fields of 29400 14M2 in 3-4 sections obtained at
weeks of dietary control. Total serum proteins were measured
different levels of the organ [11]. To verity the general aspect
1
before the last experiment by the Lowry method [131 and the
of the thymic epithelial cell network. some sections of each
albumin fraction was determined by electroimmunoassay
thymus were incubated with a xenogenic anti-keratin anti-
according to Laurell [14]. Such protein measurements were
serum diluted 1160. The keratin specificity of this antiserum
performed to confirm the absence of prote in caloric denutri-
has been previously demonstrated [17].
1
tion.
3 Results
2.4 Determination oC thymulin serum levels
1
3.1 Body and thymus weights
The serum level of biologically acti\\'e thymulin was evaluated
by a rosette assay described in detail elsewhere [15]. Briefly,
After 2 weeks of dietary regimen, the Zn2+ -deficient mice
sera under study were filtered through Amicon (Lexington. • weighed less than the control ones. It was verified that daily
1
MA)
membranes
to
eliminate
high
molecular
weight
molecules, previously shown to behave as thymulin inhibitors.
The ultrafiltrates were incubated for 90 min at 37°C with
_
5 ~
~
! '-"
spleen cells from C57BL'6 mice thymectomized (Tx) 10 to 15
.§: 4
1
davs before the test and azathioprine (Al) at the concentration
~
of iD l4g/ml. a concentration which inhibits rosette formation in
1 3
normal mice but not in Tx mice (16]. Rosettes were then
formed by centrifugation with SRBC and enumerated in a
j
~...._._-,-tlrt"'1d
hematocytometer after gentle resuspension. In the presence of
~
1
thymulin-containing sera. rosette formation was inhibited by
~
oc
Al. The results were expressed as the log2 of reciprocal highest
05
2
4
6
i
serum dilution conferrim! sensitivitv to Al inhibition upon
Ou'allon 01 Zn deplellon Imonthsi
spleen cells from adult Ti (ATx) mite.
1
Figure 1. Relative Ihymie weighls of marginally Zn!> -deficienl mlce
Using the same procedure. we studied the sensitivity to Al
(e--e) compared 10 connol animais (0---0) during 6 monlhs of
1
demonstrated by spontaneous rosette forming cells (sRFC) in
treatment. Each point represents the average of the results found in 6
!
the spleen of Zn2+-deficient and control mice after 2, 4 and 6
micc. No significant differenccs were observed bc:tween the
1
IWO
f
months of diet. This study was performed to e\\'aluate sorne
groups Ihroughout the expcrimental period.
~f
1
1
1

---

1
1
-138-
1
456
M. Dardenne, W. Savino. S. Wade el al.
Eur. J.lmmunol. 191'4.].1: 454-458
food consumption by the two groups did not differ after a 6-
6
1
week feeding period: 3.9 ± 0.16 vs. 3.9 ± 0.21 for the Zn2".
1
deficient and control animais. respeclÎ\\'ely. Therefore. ad
N
co
libitllm feeding was maintained throughout the study. Mice
oS!1
marginally deprived of Zn~" became partially hairless after 15
..
weeks of dietary treatment. Dietary Zn~" depletion did not . j 4
1
cause any significant alteration of thymie weight. Both control
5
:;
and experimental animais showed nonnal age-dependent thy·
mic involution (Fig. 1). After 4 and 6 months of treatment, we
j 3
~
observed a slight difference in relative thymie weight between
1
control and Zn2+-depleted mice, the latter presenting inferior
values. This difference. however, was not statistically signifi-
,
,
canto
Sera
Seri
Sm
_Z.CI>
.Z.CI>
..",.,," ZnCb
• ZnCb
4 months
6 monlhs
1
3.2 Serum lnH leveb
Figurt 3. Resloralion of normal Ihymulin activily in sera of margine
Serum Zn2+ values showed a moderate but nonsignificant t alIy Znz•·deficienl mice after in vitrO addition of ZnClz. One hundred
(p ~ 0.05) decline after 2 months of marginal Zn!" depriva-
1
III of mouse serum were incubated 1 h al 37·C wilh 10 Dg of ZnCl:.
tion: 89 ± 6.0 !1g1100 ml in controls, 74 ± 6.2 !1g1100 ml in test
The thymulin determinatioD was done before and afler in vitro Zn2+
mice. The levels of total serum proteins and albumin measured
addition on individual samples.
two weeks before the last killing were not different from one
group to another: 7.9 ± 0.26 gl100 ml. 8.0 ± 0.30 gl100 ml for
1
the total serum proteins; 4.1:t 0.13 gl100 ml. 4.1 ± 0.05 gI
20r
100 ml for the serum albumin of the control and Zn2+-defi·
cient groups, respectively.
~ 151
1
~ lO~
-
!
3.3 Thymulin levels
.;z
3.3.1 ln vivo determinatioDS
1:<>"'",,,
1
1
- - . _ - - . . . . ..o__ ......._.~__ ..
'O
o
2
4
6
At the beginning of the experiment. as is usually observed in
Months
young animaIs. thymulin levels were high in both control-fed
and Zn2+-deprived mice. These high levels were maintained in
1
control animaIs throughout the entire experimental period. In
Fîgurt 4. Sensilivity 10 Az expressed as Ihe minimal inhibitiOD con·
Zn2" -deprived animals. thymulin levels began to fail signifie
centration (MIC) in ~glnù of Ihe sRFC from Zn~+ -deficienl (~.)
and control mice (0---0). Spleen ceUs were incubaled Voith serial
cantly two months after the beginning of the restricted diet and
dilutions of Az from 0.5 10 100 ~glml. The number of sRFC was
reached still lower values after 4 months of Zn2" deprivation
delennined al each concenlralion of Az. Each poinl represenl5 a
1
(Fig. 2).
mean ± SE (mean of 6 mice).
3.3.2 ln vitro studies
containing ZnH • anè! the other inactive and deprived of the
1
Metal ion [9]. To confinn the possibility that the peptide could
As previously mentioned, recent data suggested that natural
be present in the serum from Zn2+-deprived mice in an inac-
molecule could exist in two fonns. one biologically active,
tive form, we perfonned supplementary experiments adding
ZnC12 to the mouse sera under study. This addition was done
ï ...,.....
1
. . . . . .I ..
after serum chelation to eliminate other metals present in the
6
..
"
sera, which could compete with Zn2+ for the binding site on
" "1-------,,? c.n-
the peptide [9].
As observed in Fig. 3, after chelation and incubation with
1
ZnC12, the thymulin serum levels found in Zn2+-deprived mice
reached nonnal values Iike those found in control animais.
The same results were observed after 2, 4 and 6 months of
Zn2" -deprived diet. The same experiments performed with Tx
1
mouse sera never induced any increase in thymulin levels. No
D5
2
4
6 Months
change in the level of the hormone was observed after incuba-
Iill!ül
tion of nonnal mouse serum with ZnC12•
Figurt
1
2. Thymulin levels in marginally Znz• -delicient (~) and
control mice (0---0). Thymulin serum levels were evaluated by an
inhibition rosette assay and expressed as the iog
3.4 Sensitivity to Az of spleen sRfC
z reciprocaltiter of the
highest active dilution. A significant decrease of thymulin leveIs was
observed in Znz• -deficient mice aCter 2 months of diet. Each dot
The number of sRFC in the spleen of Zn2+·deficient mice was
1
represents a mean ± SE (6 inice tested).
not modified when compared to control mice. But, as previ-
1
1

---

1
1
-139-
1
Eur. J. Immuno!. 1984.14: 454-458
Thymulin in zinc-deficient mice
457
1
l
4
30n
Discussion
1
In tbis study, we report the consequences of a marginal Zn!"
deprivation on mouse thymie function as assessed by the blood
level and thymie content of the thymie hormone, thymulin
(FIS-Zn) which has been shown to be produced by thymie
1
epithelial cells [Il].
As previously reported by Iwata et al. [5] in mice with moder-
ate and severe Zn:" deprivations, the blood levels of thymulin
o
1
4
6 Months
are significantly decreased as early as 2 months after the onset
of the Zn!+ -deficient diet. This effect is still more pronounced
after 4 and 6 months of the restricted diet, since at that time
Fig~re 5. Number of thymulin<Q:Haining cellslloo microscopic fields
the peripheral level of the hormone is still detectable but 8
of 29400 IUDl in marginally Zn'- -de ficie nt (.) and control mouse
1
limes lower, than in control animais.
thymuses (0).
However, at variance with previous observations, tbis rapid
drop in the thymie hormone level occurs in the absence of
1
significant thymie atrophy, a parameter wbich was constant in
ail previous reports. Low thymie hormone levels had been
interpreted as a consequence of thymic histological involution
[201. In our experiments, however, this hypothesis is unlikely,
1
since the thymus is not histologically modified and the epithe-
'.
liai ne!Work retains a normal aspect. If slight modifications of
'.....'""&...
sorne epithelial cells have been observed in ultrastrucrural
......................
studies (submitted to publication), they are not sufficient to
...._-- ............-.....
1
explain the major decline of the hormone measured in
peripheral blood.
Another Înterpretation of this hormonal decay is raised in our
recent biochemical studies showing that thymulin is a metal-
1
lopeptide containing Zn:+, which is necessary for its biological
Figure 6. Thymulin serum levels and lhymulin-containing cells. in
marginally Zn'--deficient mice. ~umber of thymulin-containing ceilS!
activitv. The hormone has been shown to exist in !wo forms;
100 microscopic fields (~l. Thymulin serum levels ("'---l as
one bi'ologically inactive. deprived of the metal ion, and one
assessed by inhibition of rosette formation are expressed as log, recip-
biologically active. containing Zn:· which binds the peptide
1
rocal titer of the highest active dilution. Each dot represents a Mean of
with an affinity caJculated in the order of 10-<' M [101.
6 mice.
We hypothesized that in Zn:· -deprived mice the hormone
cou Id still be produced by the thymie epithelium. but in its
1
inactive form. deprived of Zn:·. To assess this hypothesis. we
performed in vitro activation experiments using ZnCI:. We
ously obser\\"ed by Iwata et al. [5], their sensitivity to Az was
showed that the in vitro chelation. followed by incubation of
drasticallv decreased :!. 4. and 6 months after the onset of the
Zn:+ .deprived mice sera with ZnCI: induced a full recovery of
restricted diet. Results are shown in Fig. 4. The values
1
thymulin biological activity. Similar observations were made
observed are very simtlar to those shown after ATx [16].
with sera from children presenting nephrotic syndrome with
Zn:" deficiency [19]. a disease in which we observed a low
level of thvmulin acli\\itv that could be restored to normal after
3.5 Thymulin-eontaining ceIls and the epithelial network
1
in vitro ch'elation and iricubalion with ZnCl2 [8]. These results
confirm the presence of the inactive hormone in the serum of
Long-term Zn:" depletion was shown to produce an increase
Zn:" -deficient mice or individuals and its potential activation
in the number of thymulin-eontaining cells. Two weeks of
following Zn:+ addition. The specificity of these results was
treatment did not cause an\\" difference in the number of these
confirmed by the negativity of the activation experiments per-
1
cells. However, after 4 months of a Zn:· ·depleted diet. the
formed with sera from Tx mice or Di George's syndrome
absolute number of cells containing thymulin was significantly
patients in which the hormone is lac king.
bigher (277 ± 42) than the average value found in control mice
(148 ± 12). Similar differences were still obser\\"ed after 6
It can be argued. however, that in marginally Zn:+ -deficient
1
months of Zn:" depletion. These results are summarized in
mice. plasma Zn!+ levels are only slightly, but not significantly
Fig. 5. When comparing the number of thymulin-containing
lowered when compared to control animais. suggesting that
celis and thvmulin serum levels in Zn:" ·deficient mice. it
• the circulating hormone should be capable of binding the Zn:~
appears that 'the decrease in thymulin levels precedes increase
present in the serum. this binding inducing the reactivation of
1
in the number of hormone·containin!! cells (Fil!. 6). Interest-
the molecule. However. as mentioned above, the affini!" of
ingly, the thymic epithelial network -;s revealc-d by the anti-
Zn:+ for the peptide is low (10-<> Ml: this finding could explain
keratin antiserum, showed a normal aspect (with easily distin·
thymulin's high sensitivity to the Zn:+ concentration due to
guishable medullary and cortical areasl, during the entire
the competition with other proteins whieh. in the plasma bind
1
experimental period.
this metal y,ith higher affinity. Furthermore, as previously dis·
1
1

---

!I
1
1
-140-
1
458
M. Dardenne, W. Savino. S. Wade et al.
Eur. J. Immunol. 1984. 14: 454-458
cussei by others [20], a single estimate of total plasma Zn2+
5 References
1
cannot be used as a sensitive parameter to predict Zn2+ status.
since the metal ion is mainly bound to various plasma proteins
1 SchJoehn, L. H.. Fernandez. G., Garofalo, J. A. and Good. R.
and, depending upon its affinity for its carriers, minor changes
A., Clin. Bull. 1979. 9: 63.
in plasma levels can induce important variations in Zn 2+ to
2 Good, R. A.. J. Clin. Immunol. 1981. 1: 3.
1
3
protein binding. In parallel with thymic endocrine function,
Bach. J.-F., Immunol. Today 1981. 2: 225.
4 Nash. L., Iwata. T.. Fernandes. G.. Good, R. A. and Incefy. G.•
we compared the sensitivity of sRfC to Az in spleens from
Cel/. Immunol. 1979. 48: 238.
normal and Zn2+-deprived mice. As shown by previous inves-
5 Iwata, T., Incefy, G. S., Fernandez. T. G., Mendendez-Botet, C.
tigators [5] a significant decrease in the sensitivity to Az was
J., Pih, K. and Good. R. A., Cel/. Immunol. 1979. 47: 100.
1
induced by Zn2+ deprivation. This result must be considered in
6 Chandra. R. K., Heresi. G. and Au, B.. Clin. EJ:p. Immunol.
light of the bigh dependency of tbis spleen RfC population on
1980. 42: 332.
thymic hormone and is comparable to the effects of adult
7 Cunningham·Rundles, c., Cunningham-Rundles, C., Iwata. T.,
th)m~etomy (16]. It can be interpreted as a consequence of
Incefy, G., Garofalo. J. A.. Menendez-Botet. C., Lewis. V.,
thymulin decay and not a direct effect of Zn2
1
+ -deficiency on
Twomey, J. J. and Good, R. A., Clin. Immunol. Immunopathol.
1981. 21: 387.
the lymphocyte.
8 Bensman. A., Dardenne. M., Morgant. G., Vasmant. D. and
Fmally, analysis of thymuses from Zn2
Bach. J.-F., Eur. J. PediatT. 1983.140: 218.
+-deprived mice showed
9 Dardenne. M., Pléau, J. M., Nabarra. B.. Lefrancier. P., Derrien.
that, in spite of the absence of major alterations in thymic
1
M.. Choay, J. and Bach, J.-F., Proc. Natl. Acad. Sei. USA 1982.
epithelial cells (as assessed by the normal pattern observed
79: 5370.
after anti-keratin staining), there was a progressive increase in
10 Gastinel. L. N., Pléau. J. M., Dardenne, M. and Bach. J.-F.,
the number of thymulin-eontaining cells, thus suggesting an
Biochem. BiophYJ. Acta 1984. 797: 147.
increase in the production of the hormonal peptide.
Il Savino. W., Dardenne. M., Papiernik. M. and Bach. J.-F.,!. EJ:p.
1
Med. 1982. 156: 628.
Our observations indicate a compensatory phenomenon. simi-
12 Dardenne. M., Pléau, J. M., Savino. W. and Bach. J.-F..
lar to a classical feedback response. We have recently shown in
Immunol. Lett. 1982. 4: 79.
normal mice that an e:qJerimentai decrease of circulating thy-
13 Lowry, O. H.. Rosebrough. N. J., Fare, A. L. and Randall. R. J..
1
J. Biol. Chem. 1951. 193: 265.
mutin (either by injections of anti-thymulin monoclonal anti-
14 LaureU. C. B.. Anal. Biochem. 1965. 10: 358.
bodies or by promoting endogenous antibody production after
15 Dardenne. M. and Bach. J.-F.. BiologicaJ Acrivi~' of Th.vmic Hor-
immunization with thymulin coupled to bovine serum albumin
mones, Kooyker Scientific Publications. Rotterdam 1975. p. 235.
induced an increase in the number of thymulin-contaimng cells
16 Bach. J.-F. and Dardenne. M.. Immunology 1973.25. 353.
1
[21]. We think that in the case of a marginal Zn2- -deficiency as
i7 Viac.1.. Staquet. M. J., Thivolet. J. and GouJon. G.. Arch. Der-
studied here. tbis feedback effect could be stimulated by the
1I'Ultoi. Res. 1980.267: 179.
decline of the hormone in its active form, suggesting that the
18 Tanaka. L.. Fernandez, G., Tsao. C.. Pih. K. and Good. R. A..
biological regulatory system responsible. strictly recognizes
Fed. Proc. 1978. 37: 931.
1
the peptide containing Zn2
19 Bensman. A.. Morgant. G.. Hasaert, D. and La.sfargues. G..
". This hypothesis is in agreement
Nouv. PreJJe ,\\{ed. 1981. /0: -16.
with the fact that. so far. ail previously investigated functions
20 Meadow. S.. Smith. P. W. N.. Keeling, N. W., Ruse. 1.. Day. J..
of thymulin require the presence of Zn2+ in the molecule of
Scopes. J. W.. Thompson. R. P. H. and Blodcam. B. L.. Luncet
the hormone.
1981. ii: 1135.
1
21 Sa~ino. W., Dardenne. M. and BJch. J.-F., Clin. Exp. lmmunol.
We thank M. C. Gagnerault and C. Rieumailhol for rheir Jkillful rechm-
1983.52: 7.
.
cal asJisrance and J. JacobJon for reviewmg rhe manusCTipt.
Received December 14. 1983.
1
1
1
1
1
1
1
r
1
1
1
l

---

1
-141-
1
1
ART 1 C L E N° 8
1
1
1
"Thymulin (Zn-Facteur Thymique Serique) activity in
anorexia nervosa patients".
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

1
1
-142-
1
Thymulin (Zn-Facteur Thymique Serique) activity
1
in anorexia nervosa patients1-3
1
Salimata Wade, Fanny Bleiberg, Arlette Massé, Jean Lubetzki. Hubert Flavigny,
Philippe Chapuis. Daniel Roche. Daniel Lemonnier and Mireille Dardenne
1
ABSTRACf
Thymulin (or FTS-Zn) a well-defined thymic honnone was studied in fifteen
female patients hospitalized for imorexia nervosa. The circulating honnone was measured together
1~.
with the plasma levels of thyroid honnones, cortisol and zinc. Thymulin activity detennined by
1
the rosette assay was significantly reduced in the anorexia nervosa patients compared to sex- and
age-matched healthy control subjects. The patients were characterized by very depressed plasma
levels of triiodothyronine (TJ) but exhibited nonnal concentrations of thyroxine (T.), thyroxine-
binding g10bulin (TBG), cortisol and zinc. The distribution of their peripheral lymphocyte cells
1
into several subsets was not alfected. The observed decrease of thymulin activity in this illness
might he the consequence of thymic atrophy secondary to malnutrition and/or honnonal
disturbances. Our results suggested that the fall in thymulin level might explain the variability of
cellular immune responses in anorexia nervosa patients and occurrence of anergy when their
1
weight loss is far advanced.
Am J Clin NUlr 1985;42:275-280.
KEY WORDS
Anorexia nervosa. thymulin (FTS-Zn) activity, zinc, cortisol. thyroid honnones.
1
T-lymphocyte subpopulations
1
Introduction
delayed hypersensitivity skin responses (S,
Il), but when their weight loss is far ad·
Protein-energy malnutrition, at least in
vanced, a significant deerease in cell-mediated
1
children, is known to be associated with
immunity can occur (S).
thymie atrophy, depression in T-cell numbers
Recently, several investigators have indi-
and functions. This apparently powerful and
cated the possibility that the epithelial func-
1
specifie effect of malnutrition on the thymus
tion of the thymus may be altered by nutri-
has been extensively studied in malnourished
tional deficiencies ( 12-15). In order to further
individuals from the developing countries.
investigate the effect of anorexia nervosa on
However. in these populations, infections
thymie function, a circulating thymie hor-
1
are commonly associated with malnutri-
tion and can interfere with the immune
responses (1-4).
1 From
the U 1 INSERM, Hôpital Bichat, 170 Bd
1
Anorexia nervosa is a syndrome of pro-
Ney, 75018 Paris: U25 INSERM, Hôpital Necker. 161
rue de Sèvres. 75015 Paris: Service de Psychiatrie, Hôpital
found weight loss in young persons in which
International de l'Université de Paris, 42 Bd Jourdan.
the refusai to eat is directly related to psychic
75014 Paris: Service de Médecine Interne et d'Endocri-
1
disturbances. Despite their severe body-wast-
nologie-Nutrition, Laboratoire de Biochimie. Laboratoire
ing, anorexia nervosa patients are described
de Radio-Immunologie, Hôpital Lariboisière. 2 rue Am-
as exceptionally free from infections (5). Ex-
broise-Paré, 75010 Paris.
2 Supported by grants from the Conseil Scientifique
tensive literature involving the clinical and
de l'UER Xavier-Bichat and from the PRC INSERM
1
sorne biochemical changes in this illness has
W 128020.
been widely reported in the past (6, 7), but
J Address
reprint request to: Salimata Wade, U 1 .
the effect of anorexia nervosa on the immune
INSERM. Unité de Recherches sur la Nutrition et
l'Alimentation, Hôpital Bichat. 170 Bd Ney, 75018
1
function remains conflicting and unclear (S-
Paris-France.
Il): thus, patients with anorexia nervosa
Received May 10, 1984.
display a relatively modest impainnent of
Accepted for publication January 29, 1985.
t
The American JOl/rnal of Clinical Nwrilion 42: AUGUST 1985, pp 275-280. Printed in USA
275
f
©
1
1985 American Society for Oinical Nutrition
1
1
1
1

---

1
1
-143-
1
276
WADE ET AL
1
mone, thymulin (a zinc-facteur thymique
weight, a persistance in an idealized slim self·image,
serique complex previously called FfS) was
amenorrhea and an absence of other medical and psy-
chiatric ilIness. Their clinical features are reported in
quantified in anorexia nervosa patients and
Table 1. Calculations for ideal body weight (IBW) were
in healthy control subjects. Thymulin is a
1
based on Lorentz' formula (21): height - 100 - [(height
nonapeptide exdusively produced by the
- 150)/2.5]. The control group (thirteen females) corn·
thymie epithelial cells but its activity is de-
prised students and nurses and were age·matched. None
tectable in peripheral blood. Thymulin level
of the subjects, including the controls had clinical evidence
of infection.
1
is thus a good marker of thymie epithelial
Consent for venipuncture was obtained from the
funetion.
patients and the medical staff after the purpose and
Thymulin activity has been found in the
nature of the study had been explained. Our anorexia
serum of various species induding man. Its
nervosa patients were tested within 1 to 5 days of
1
admission. After overnight fasting, blood was drawn and
level is relatively constant in humans during
collected in cooled (4°C) heparinized tubes. Because of
ontogeny but declines with advancing age
the cortisol assay, samples were obtained between 8 and
(16-18). Thymulin possesses several biologicaI
8:30 AM. Plasma was immediately separated and kept
frozen at -20°C until analyzed.
1
activities (16, 18). It has aIso been shown
that it acts on the T-cells involved in delayed
Plasma zinc was assayed by atomic absorption spec-
trophotometry.
type hypersensitivity (18, 19).
Cortisol was quantilied by radioimmunoassay (RIA)
(Kit Cortctk 125-0ri5, CEA, Gif·sur-Yvette, France).
1
Patients and methods
Thyroxine (T.), triiodothyronine (Tj ) and thyroxine-
binding globulin (TBG) were determined in plasma
Patients
samples by RIA (Kit CEA, Gif·sur-Yvette, France).
The study, conducted in accordance with the INSERM
Immune studies
1
ethical committee for human experimentation, involved
lifteen female patients (14 ta 35 years of age) suffering
Lymphocytes were isolated from peripheral blood by
from anorexia nervosa. They were admitted for treatment
Ficoll·Hypaque gradient density centrifugation. Mono-
either at the psychiatric unit of the Hôpital International
clonal anti-T cell antibodies were used to evaluate the
1
de l'Université de Paris or the Medicine and Endocrine
proportion of various T-Iymphocyte cell subsets by in-
Unit of the Hôpital Lariboisière. Paris. The patients
direct immunonuorescence: the percentage of mature T-
satislied the diagnostic criteria of Feighner et al (20);
ceIls, helper and cytotoxic·suppressor cells were deter-
they were characterized by a distorted attitude towards
mined using OKTl , OKT. and OKTs respectively (22).
food, eating and weight, a greater than 30% loss in body
slg-bearing lymphocytes (B-Iymphocytes) were quan-
1
TABLE 1
Clinical features of patients with anorexia nervosa
1
Durallon of
Body "Clght al
Ideal body· WClghl
" "'e.ghl
Pallents
Age
anorUla ncrvosa
hospÎtalllatlon
(IBW)
'050
(.1'''1
(yr.')
(klii
Iklii
RB
F
15
1
15 1
34
50.6
33
1
lePA
F
20
5
151
28
50.6
45
SC
F
18
4
149
31.5
49.4
37
NF
F
15
0.5
155
38
53
29
PMC
F
18
0.3
162
34
57.2
46
1
GN
F
16
1
163
34.8
57.8
40
KJ
F
35
23
160
38.6
56
31
BJ
F
29
12
161
31.3
56.6
45
VBt
F
18
4
147
27
48.2
44
BF
F
20
0.7
167
45
60.2
25
1
MM
F
26
9
161
38
56.6
33
Ait
F
30
2
175
45
65
31
CL
F
19
0.5
167
39.5
60.2
35
Ge
F
14
1.5
138
20.6
42.8
52
1
AV
F
16.5
0.75
158
40
54.8
27
Mean ± SEM
20.6 ± 1.64
4.4 ± 1.6
158 ± 2.41
35.0 ± 1.72
54.6 ± 1.45
36.9
• Ideal body weight (IBW) was calculated acccording to Lorentz' formula: JBW = height _ 100 _ height ; 150.
1
t Patients edemic on admission.
2.
1
1
1

---

1
1
-144-
1
THYMIC FUNCTION IN ANOREXIA NER VOSA
277
25
Results
At the time of the study the mean weight
1
E 20
and height of the patients were 35.0 ± 1.7
0
E
0
0
kg and 158 ± 2.4 cm, respectively. The du-
~
0
ration of the syndrome varied between 0.3
1
- 15
~
l:II
~
-CJl
to 23 yr suggesting that recurrent episodes of
~
the illness had occurred in sorne patients.
.....
10
40
However only two of them were edemic
0
U
!Il
Z
(c1inical observation). Body weight loss was
1
l-
N
greater than 35% (Table 1).
a:
5
20
0
The healthy control group was 21.2 ± 1.6
U
(range 15 to 35) yr old: the mean weight and
0
0
height were 54 ± 1.6 kg (range 46 to 60) and
1
FIG 1. Plasma cortisol and zinc level in anorexia
163 ± 1.6 cm (range 152 to 173), respectively.
nervosa patients 0 and matched healthy controls m
Their weight/height index was more than
Means + SEM are shown.
90% of the standard (21, 24).
1
The plasma zinc concentration (Fig 1) was
unchanged in comparison with healthy sub-
tified by direct immunotluorescence with a t1uorescein
jects. The mean plasma cortisol was slightly
conjugated Ig-antisera.
and not significantly higher in the patients
Thymulin activity was measured by the method de-
1
scribed by Dardenne and Bach (23); the assay analyzes
than in the control group (Fig 1). T4 concen-
the conversion of relatively azathioprine- (Az) resistant
tration and free T4 index were comparable
spleen cells of adult thymectomized mice to li-positive
in the anorexia nervosa patients and in the
rosette-forming cells that are more sensitive to Az.
controls, while T
1
3 was significantly reduced.
Brietly, plasma samples were filtered by centrifugation
at 4°C on CF~o Amicon membranes (molecular weight
No statistical difference was observed in
eut off at 50.000 to remove an FTS inhibitor). The
TBG, although mean TBG levels were slightly
ultrafiltrates were incubated for 90 min at 37°C with
higher in the control than in the patient
1
spleen cells from mice thymectomized 10 to 15 days
groups (Table 2). This slight elevation was
before the test. and Az (10 Ill/ml) was added simulta·
probably due to oral contraception.
neously. At the end of the incubation lIme 12 X 106
sheep red blood cells were added to the cells. This cellular
Thymulin activity, expressed as log-2 re-
preparation was centrifuged (at 4oC 150 x g. 5 min) and
ciprocal titers is shown in Figure 2. Because
1
resuspended by low-speed rotation ( 10 rpm) on a roller.
of the age-related decline of thymulin le\\'el.
Rosette-forming cells were counted in a hemocytometer.
results are shown according to age. Plasma
ln the presence of thymulin. Az inhibited rosette for-
mation at a concentralIon of 10 Ilg/ml. The highest
thymulin level was significantly (p < 0.01)
1
dilution of the plasma sample which induced the rosette
reduced whatever the duration of anorexia
inhibition by Az was considered as the active dilution.
nervosa. Yet, the distribution of T-Iympho-
Ali the determinations were canied out in duplicate and
cytes in the peripheral blood into mature.
the results were expressed as log-2 reciprocal titer.
helper. cytotoxic-suppressor cells remained
1
Statistical analysis of data
unchanged as weil as the ratio helper/C)10-
Differences in mean values between the groups were
toxic-suppressor cells. The total B-lymphoC)1e
determined using the Student's t test.
count was also normal (Table 3).
1
TABLE 2
Concentration of thyroxine (T.). triiodothyronine (Tl) and thyroxine binding globulin (TBG)
1
TBG
II'X//OO mil
1
Anorexia nervosa patients
86.5 ± 6At
9.5 ± 0.8
23.7 ± lA
4.1 ± 0.3
Controls
129.8 ± 7.5
9.2 ± 0.4
25.8 ± 1.6
3.7 ± 0.2
.
T.
• Ratio TBG x 10.
1
Means ± SEM; t P < 0.01 versus the control group.
1
1
1
1
1
1

---

1
1
-145-
1
278
WADE ET AL
N
8
suggests that the decreased thymulin activity
1
observed in anorexia nervosa did not result
1 7
from zinc deficiency. The reduction of thy-
1
> 6
mulin activity in this syndrome might be the
~
i
:> 5
••••
••
consequence of protein-energy malnutrition
1
...........
without superimposed infections via thymus
§4
•••••
1
atrophy and/or other honnonal disturbances.
Z
3
•••
•
The immune system, although regulated to
::i 2
;:)
-......-
_...-
a large extent by intrinsic cellular and hu-
::1
•
>
••
moral
events,
is
sensitive to
honnonal
1
~ 0 ........~~......---'
changes. In recent years, several investigators
15 • 21' years
25· 35 years
have demonstrated that honnones such as
-~-
Patients Controls
Patients Control.
glucocorticoids, growth honnone. thyroxine
and sexual honnones may exert their action
1
FIG 2. Plasma thymulin activity in anorexia nervosa
on the Iymphoid system (28). It has been
patients and healthy controls grouped according to age.
Individual values were reported. p < 0.01. patients versus
shown that high glucocorticoid levels may
the corresponding control group.
decrease thymulin concentration (29). Our
1
patients exhibited a nonnal cortisol concen-
tration suggesting that this factor was not
Discussion
involved in the depression of their circulating
thymic honnone.
1
Plasma thymulin activity was significantly
In humans, alterations of thyroid turnover
1
reduced in anorexia nervosa. We have pre-
(low T3 syndrome} cause reduction of thy-
viously reported that thymulin level was nor-
mulin activity (28). Experimental thyroxine-
mal in moderately as well as in severely
deprived animais show a significant decrease
1
malnourished children with acute infections
in thymulin activity. which is restored by
(25).
However nutritional deficiency
has
daily injection with exogenous triiodothyro-
complex effects on the immune system de-
nine (30). Low serum T3 and increased serum
1
pending upon the age at which the nutritionai
reverse T3 levels have been previously re-
insult occurs, variations in nutritional re-
ported by others (31-33) in anorexia nervosa.
quirements and the presence or absence of
The patients we studied exhibited nonnal T4
infections. Decreased thymulin activity has
and TBG levels but a significant drop in
1
been shown in mainourished children without
plasma T3 concentrations compared to the
infections (12) and in experimental animais
healthy controls. The value recorded for T3
fed nutrient-deficient (zinc. pyridoxine. vita-
in the presence of unchanged T4 and TBG
mins) diets (15, 26, 27).
concentrations suggests a decrease in frac-
1
The presence of zinc is essential for the
tional conversion of T4 to T3 . Virtually ail
biological activity of the circulating thymie
the metabolic action of T 3 is derived from
honnone (\\ 7). In our study, the plasma zinc
T4 generated from it. In nonnal subjects a
1
concentrations of the patients were nonnal
minimum of about 40% of T4 is metabolized
and there was no correlation between its level
via monodeiodination (34). These data con-
and that of thymulin whatever the degree
finn previous findings (33) indicating that
and duration of patient's malnutrition. This
patients with anorexia nervosa were euthy-
1
1
TABLE 3
Distribution of peripheral lymphocyte cells into several subsets*
CytotO..1C.
.I,-bun",
~bture
Helper
suppressor
(.,
•
Rauo: -
1
B-cells
T<ells
lb)
b
Anorexia nervosa patients
13 ± 1.2
64 ± 2.9
39 ± 2.3
23 ± 1.8
1
1.8 ± 0.20
Controls
II ± 1.4
S9 ± 4.S
40 ± 2.7
22 ± 1.6
1.9 ± 0.27
1
t
• Results expressed as %.
1
Means ± SEM.
1
t"r
1
1
1
i
1
1
1

---

1
1
-146-
THYMIC RlNCTlON IN ANOREXIA NERVOSA
1
279
roid. However the decline in plasma Tl levels
anorexia nervosa may have a much greater
observed in the present cases might suggest
impact in functional than in phenotypic terms
1
that Tl can promote important changes in
(surface antigens). Additional studies ofhelper
plasma thymulin activity in anorexia nervosa,
T-ceU functions in this syndrome appear
although we did not find any linear correla-
warranted to further define the functional
tion between the two factors.
immunological status of anorexia nervosa
1
The sexual honnonal status of the patients
patients. Most of the immune studies per-
must also be considered, since there is a
fonned in such patients involve delayed hy-
relationship between gonadotrophic and go-
persensitivity skin tests. Our data demon-
1
nadal steroid honnones and the thymus
strated that short-tenn as weU as long-tenn
gland. Abnonnalities in gonadotrophic secre-
anorexia nervosa induces a thymie dysfunc-
tion have been observed in anorexia nervosa
tion which is reflected by a decreased thy-
female patients due to the prominence of
mulin production when body wasting is
1
menstrual dysfunction (7). Dardenne et al
advanced.
n
(unpublished results) have demonstrated that
castrated female or male mice show a strong
The author.; acknowledge the anorexia nervosa patientS
reduction ofthymulin activity. However, fur-
1
and the control subjects for their cooperation. They
ther studies are needed to define which of
thank the Medical Staff of the Psychiatric U nit (Hôpital
the sexual honnones is directly involved in
International de I·Univer.;ité de Paris) especially ~r.; N
such a mechanism.
Bonnet and the Medical Staff of the Medicine and
Endocrine Unit (Hôpital Lariboisière).
1
Mature T-œil numbers and subpopulations
were normal in our patients. Yet, their thy-
mulin activity did not correlate with the
References
number and percentage of T-œU subsets.
1
1. Bell RG. Undernutrition. infection and immunity:
Pertschuk et al (8) reported that in anorexia
the role of parasites. Papua New Guinea ~ed J
nervosa. anergy becomes a rule when weight
1978:21 :43-55.
loss is advanced (weight less than 60% IBW).
2. Suskind RM. Malnutrition and the immune mponse.
1
This was recently confinned by the study of
New York. NY: Raven Press. 1977.
3. Chandra RK. :-.rewberne PM. :-.rutrition. immunitv
Russell et al (35). Anergy probably occurs in
and infection. :-Iew York. NY: Plenum. 1977.
.
our patients since most of them were severely
4. Gross RL. Newberne PM. Role of nutntion in
depleted. Delayed-type hypersensitivity is a
immunologic functions. Physiol Rev 1980:60: 188-
1
T-cell-dependent phènomenon which mainly
302.
involves helper T-cell function. Thus anergy
5. Bowers TK. Eckert E. Leucopenia in anorexla ner·
vosa: lack of incre:lSCd risk of Întecllon. Arch Intern
may result from abnonnalities in function
Med 1978:138:15::0-3.
1
and/or in number of helper T-cell subset.
6. Warren MP. Van de Wiele RL. Clinical and :neta-
Accordingly a possible abnonnal response
bohc features of anorexia nervosa. Am J Obstet
can coexist with unchanged T-cell number
Gynecol 1973:117:435-49.
7. Lupton M. Simon L Barrv V. et al. BlOloilCai
(or phenotype as defined by monoclonal an-
1
aspects of anore:l1a nervo~. Life Sci 1976:18:
tibodies). The lack of correlation between T-
1341-8.
cell functions and phenotypes has been re·
8. Penschuk Ml. Crosby LO. Barot L et al. Immune-
ported in other diseases (36). The coexistence
competency in anorexia nervosa. Am 1 Clin :-.rutr
198:::35:968-7::.
1
of nonnal T-cell number and reduced cellular
9. Wyatt RJ. Farrell M. Berry PL et al. Reduced
immune capacity in anorexia nervosa has
alternative compicment pathway control protem lev.
also been found by previous investigators (8.
els in anorexia nervosa: response to parenteral ali-
10, 11, 35). Such discrepancy may indicate
mentation. Am 1 Gin Nutr 1982:35:973-80.
1
that the explanation for the impaired cellular
\\0. Armstrong-Esther CA. Lacey JH. Crisp AH et al.
An investigation of the immune response of patients
immunity in this illness is not the reduction
suffering from anorexia nervosa. Postgraduate ~ed
in the proportion and number of T-Iympho-
J 1978:54:395-9.
1
cyte subjects in the peripheral blood. Thy-
II. Golla JA, L1non LA. Anderson CF et al. An
mulin is known to promote most T-<:ell
immunological asscssment of patients with anorexia
functions and helper T-<:eU stimulation by
ner...osa. Am J Gin Nutr 1981 :34:2756-62.
12. Chandra RK. Serum thymic hormone acti\\ity in
thymulin has been demonstrated (18, 19);
protein~nergy
1
malnutrition. Oin Exp Immunol
hence the decrease in thymulin activity in
1979:38:228-30.
1
1
1

---

1
1
-147-
1
1
280
WADE ET AL
13. Olusi SO, Thunnan GB, Goldstein AL. Effects of
moderately and severely malnourished Senegalese
1
thymosin on T·lymphocyte rosette fonnation in
children. Am J Clin Nutr 1982;36: 1129-33.
children with kwashiorkor. Clin Exp Immunopathol
26. Iwata T, Incefy GS. Tanaka T et al. Circulating
1980:15:687-91.
thymic honnone levels in zinc deficiency. Cell Im-
14. Jackson TM, Zaman SN. The in vitro effect of the
munol 1979;47: 100-5.
1
thymic factor thymopoeitin on a subpopulation of
27. Chandra RK, Heresi G, Bing AU. Serum thymic
lymphocytes from severely malnourished children.
factor activity in defi.ciencies of calories, zinc. vila-
Qin Exp Immunol 1980;39:717-21.
min A and pyridoxine. Clin Exp Immunol 1980;42:
15. Dardenne M, Savino W, Wade S et al. In vivo and
332-5.
1
in vitro studies of thymulin in marginally zinc·
28. Fabris N, Garaci E, Hadden J et al. Immunoregu-
deficient mice. Eur J Immunol 1984; 14:454-8.
lation. London: Plenum Press. 1982.
16. Bach JF. Dardenne M, Pleau JM et al. Isolation.
29. Bach JF, Duval D, Dardenne M et al. The elfects of
biochemical characteristics and biological activity of
steroids on T cells. Transplant Pree 1975;7:25-30.
a circulating thymic honnone in the mouse and in
1
30. Savino W, Wolf B. Araun-Spire S. Thymic honnone
the human. Ann NY Acad Sci 1975;249:186-210.
containing cells. IV. Auctuations in the thyroid
17. Dardenne M, Pleau JM. Nabarra B et al. Contribution
honnone levels in vivo can modulate the secretion
of zinc and other metals to the biological activity of
of thymulin by the epithelial cells of young mouse
the serum thymic factor. Proc Nat! Acad Sci USA
thymus. Clin Exp Immunol 1984;55:629-35.
1
1982:79:5370-3.
31. Croxson MS, Ibbertson HK. Low serum triiodothy-
18. Bach JF. Thymulin (FIS-Zn). Gin Immunol Allergy
ronine (Tl) and hypothyroidism in anorexia nervosa.
1983;3: 133-56.
.
J Clin Endocrinol Metab 1977;44:167-74.
19. Erard D. Charreire J, Auffredou MT et al. Regulation
32. Richard JL. Bringer 1. Rigaud P et al. Modifications
1
of contact sensillvity to DNFB in the mouse: eifects
du profil honnonal de ('anorexie mentale au cours
of adult thymectomy and thymic factor. J Immunol
de la renutntion. Corrélations avec les paramètres
1979: 113: 1573-6.
nutritionnels. Ann EndocnnoJ 1982:43: 164-5.
20. Feighner JP, Robins E. Guze SB et al. Diagnostic
critena for use in psychiatric research. Arch Gen
33. K\\'etny J. Thyroxine binding and cellular metJbolism
1
Psvchlatrv 1972:26:57-63.
of thyroxine in mononuclear blood cells from patients
21. Lorenu ru. Der Konstitutionsindex der Frau. Klin
with anorexla nervosa. J Endocr 1983:98:Jn-50.
Wochenshr 1929:16:734-6.
34. lngbar SH. Woeber KA. The thyroid gland. In:
22. Bach ~A. Bach JF. The use of monoclonal antl-T
William RH. cd. Textbook of endocrinology, Phil·
1
cell anubodies to studv T cell imbaiances in human
adelphia-London-Toronto: WB Saunders. 1974:95-
diseases. Clin Exp Im.:ounol 1981 ;45:449-56.
232.
23. Dardenne M. Bach JF. The sheep cell rosette assay
35. Russell D McR. Prendergast Pl. Darby PL et al. A
for the evaluation of thymlc honnones. Kooyker
comparison between muscie function and body
1
Rotterdam: Van BeHum 1975:235-43.
composition in anorexia ner\\'osa: the eifec~ of re-
24. Blackburn GL. Bistnan BR. Baltej SM et al. :-Jutri-
feeding. Am J Clin Nutr 1983:38:129-37.
tional and metabolic assessment of the hospitalized
36. Tjernlund U. Cesare P. Tournier E et al. T-cell
patient. JPEN 1977:1:11-22.
subsets in multiple scierosis: a comparative study
25. Maire B. Wade S. Bleiberg F et al. Absence of
between cell surface antigens and functlon. Clin
1
vanauon 10 Facteur Thymique Serique actlvity in
Immunol Immunopath 198·U2: 185-97.
1
1
1
1
1
1
1

---

1
1
-148-
1
1
;1
ART l C L E N° 9
1
1
i
1
1
1
"Thymulin (Zn-FTS) aetivity in protein-energy malnutrition
new evidence for the interaction between malnutrition and
1
infection on the thymie function".
1
1
1
1
1
1
1
1
1
1f
1
1
,1!
f
1
1
1
1

---

1
1
-149-
1
1
THYMULIN
<Zn-FTS) ACTIVITY IN PROTEIN-ENERGY MALNUTRITION :
1
NEW EVIDENCE FOR THE INTERACTION BETWEEN MALNUTRITION AND
INFECTION ON THE THYMIC FUNCTION
1
1
Foot-line :
Thymic function in malnourished and infected
1
children
1
Salimata WADE, Gérard PARENT, Fanny BLEIBERG-DANIEL, Bernard
1
MAIRE,
Mouhamadou FALL, Dominique SCHNEIDER, Béatrice LE
1
MOULLAC and Mireille DARDENNE
1
1
1 From the U.1 INSERM, Hôpital Bichat, 170 bd. Ney, 75018
1
PARIS; ORSTOM, ORANA, Dakar - Sénégal
; Centre Hospitalo-
1
Universitaire, Dakar; U.25 INSERM,
Hôpital Necker,
161 rue
de Sèvres, 75015 PARIS.
1
2 Supported in Part by grants from the Nestlé Nutrition
1
Research Grant Programme N° 84/59.
1
3 Address reprint request to : Salimata Wade, U.1 INSERM,
1
Unité de Recherches sur la Nutrition et I~Alimentation,
Hôpital Bichat,
170 bd.
Ney, 75018 Paris -
France.
1
1

---

1
1
-150-
1
1
ABSTRACT
The combined effects of protein-energy
malnutrition (PEM) and infection on the thymie function,
1
evaluated by the specifie plasma thymulin activity were
studied in Senegalese children:
29 hospitalized in Dakar for
1
malnutrition and various diseases (these children were
1
severely malnourished and infected); 9 infected without sign
of severe PEM,living in Dakar;
13 apparently healthy,
1
uninfected living in Dakar;
7 apparently healthy, uninfected
living in Paris. Most of the free-living children in Dakar
1
suffered from mild to moderate protein-energy malnutrition.
1
The results showed an absence of variation in the total
plasma thymulin activity in the moderately as weIl as in the
1
severely malnourished children. Yet, the specifie thymu1in
activity (total plasma activity minus the activity recorded
1
after adsorption of the plasma with a monoclonal
1
anti-thymulin antibody ) was almost undetectable in the
infected children and was normal only in
the children
1
living in Paris. The study demonstrated that the specifie
thymulin activ±ty might be decreased by moderate and severe
1
PEH, but episodes of infection induce the production of
1
factor(s) which enhance artificially the plasma thymulin
level. The results also suggested that severe malnutrition
1
is not the only underlying cause of depressed level of
thymulin in malnourished children from the Third World.
1
Concurrent infections are important factors to be taken into
1
account.
1
KEY WORDS
Protein-energy malnutrition; thymulin (Zn-FTS)
activity;
lymphocyte subpopulations;
infections.
1

---

1
1
-151-
1
1
INTRODUCTION
1
Malnutrition and infectious diseases are the two major
1
health problems of developing countries. Since the report of
SCRIMSHAWet al in 1968 (1), malnutrition is now widely
1
recognized to synergize with infection to create a vicious
cycle in children. The high frequency of infections in
1
protein-energy malnourished children has long suggested an
1
impairement of their cell-mediated immune responses (2 - 6).
Severe thymie atrophy is a striking and consistant finding
1
in undernourished children : involution of the thymus gland
in such children has led to the term "nutritional
1
thymectomy". However much remains to be learnt about the
1
underlying mechanisms. Many studies indicated that T-
lymphocyte division and proliferation are severely
1
restricted (2 - 6), but very little information is available
regarding the epithelial structure of the thymus. MUGERWA
1
(7) observed well-preserved thymie epithelial cells in
children with kwashiorkor.
In contrast, JAMBON et al (8) in
1
a postmortem study, examined the thymus of children with
1
various degrees of malnutrition and reported marked
modification in the severely malnourished children, who were
1
nearly completely depleted of thymie reticulo-epithelial
cells. Evidence of thymie epithelium abnormalities in
1
!
malnourished mice has also been reported by MITTAL et al
1
1
1
(9,10).
f
In vitro studies with peripheral T-Iymphocytes from
1
1
1
l

---

1
1
-152-
1
1
1
children with severe malnutrition indicated that some thymie
1
1
hormones such as thymopoïetin as well as thymosin fraction V
1
increased the percent age of T-cell rosettes
(11,
12).
•
1
Concentration of thymie hormones might be altered but
the few reports in the litterature were conflicting.Thymulin
1
levels were also studied in malnutrition.
Thymulin
(or
1
Zn-FTS)
is a basic nonapeptide with a molecular weight of
about 1,000,
10 times lower than that of thymosin but 10 to
1
1,000 times more active.lt is exclusively secreted by the
thymie epithelium;
its activity which is age dependant, has
1
been found in the serum of various species (13).
The strict
1
thymie dependency of thymulin has been confirmed by many
studies (14 -
17). Thymulin is known to promote most T-cell
1
functions
(18).
In the abscence of an available radioimmuno-assay,
the
1
1
determination of blood thymulin activity is achieved by a
1
1
1
bioassay (rosette assay)
developped by DARDENNE et al
(19).
Using this rosette inhibition assay, CHAN DRA reported
1
1
reduced levels of thymulin in children with severe
malnutrition (20)
and in small for gestational-age infants
1
(21). The decreased activity was restored to normal after
6-8 weeks of nutritional therapy
(22).
In contrast, we found
1
no significant ~ffect in moderately or severely malnourished
1
and infected children
(23). As i t is
well established that
1
1
1
i
[
1
l

---

1
1
-153-
1
1
during antigen-induced T-cell activation,
an l'allogene~c"
1
factor
(AF)
is produced with a thymulin-like activity on the
rosette assay
(24,
25),
we proposed differences in the rate
1
of concomitant infections to explain the discrepancy between
our results and those of CHANDRA.
1
The present study was designed to further investigate
1
the effect of malnutrition/infection interaction on thymulin
activity and to examine the production of AF in groups of
1
Senegalese children with various degrees of malnutrition and
infection. AF production was measured after specifie
1
immunoadsorption of the plasma samples with monoclonal
1
anti-thymulin antibody.
1
SUBJECTS AND METHODS
1
Subjects
1
A total of fifty-eight Senegalese children aged between
4 and 30 months
(mean age 18 months)
were investigated.
The
1
children were selected according to a carefull clinical
examination and were grouped as follows :
1
- 29 children hospitalized in Dakar (Senegal,
West
1
Africa)
for various diseases including malnutrition ; most
of these subjects were mainly suffering from respiratory
1
diseases, gastrointestinal infections and measles (group M)
1
1
1

---

1
1
-154-
1
1
-
22 age-matched children coming for vaccination
1
in a MaternaI Child Care Center of Dakar were divided into
two groups:
1
the first one
(group I)
comprised 9 children without
clinical signs of either kwashiorkor or marasmus but were
1
diagnosed as suffering from various common infectious
1
diseases
(diseases of the lungs, measles, diarrhea, otitis,
cold).
1
the second one (group CD)
involved 13 apparently
healthy children free from any evidence of recognizable
1
clinical diseases.
1
7 healthy aged-matched Senegalese children were
1
screened in Paris (France)
and served as control subjects
(group CP>.
1
The mothers and the medical staff were informed about the
1
object of the study and their full
consent was obtained. The
study was conducted in accordance with the Senegalese
1
ethical committee for human experimentation.
Ages, weights were recorded and compared to the expected
1
weight for age, weight for height and height for age using
1
the NCH8 standards (26).
1
Methods
General measurements
1
Blood was withdrawn by venipuncture and collected in
1
cooled
(4°C)
heparinized tubes before any nutritional and
1

---

1
1
1
-155-
1
1
1
1
anti-infectious treatment was started. Plasma was
1
immediately separated and kept frozen at -70°C until
analyzed. An aliquot of blood was used to determine the
1
hemoglobin levels by the cyanmet-hemoglobin method,
1
hematocrit, red blood cell and white blood cell counts by
standard techniques.
1
To standardize aIl the measurements and to take into
account unknown environmental factors,
blood was also drawn
1
from a healthy white adult subject living in Dakar more than
1
1 year.
This subject was used as "internaI reference."Plasma
albumin,
prealbumin, transferrin, orosomucoid and c-reactive
1
protein levels were measured according to the radial
immunodiffusion method of MANCINI et al
(27).
1
Immune studies
1
Lymphocytes were isolated from peripheral blood by
1
Ficoll-Hypaque gradient density centrifugation. Monoclonal
anti-T cell antibodies (Ortho Diagnostics Systems Inc.,
1
N.J.) were used to evaluate the proportion of various T-
lymphocyte subsets by indirect immunofluorescence : the
1
percentage of mature T-cells, helper/inducer and
1
cytotoxic/suppressor cells were determined using OKT3, OKT4
and OKT8 respectively (28).
sIg-bearing lymphocytes (B-
1
lymphocytes) were quantified by direct immunofluorescence
with a fluorescein conjugated Ig-antiserum.
1
1
1

---

1
l ,
1
-156-
1
f
1
1
1
l
1
1
Thymulin activity of the plasma samples was measured
1
using the rosette assay described by DARDENNE et al
(19) and
1
reported in detail elsewhere (23). This test analyses the
i
1
1
conversion of relatively azathioprine (Al)
and anti-thy1
1
l
f
1
serum resistant rosette forming cells (RFC)
from adult
!
l
thymectomized
(ATX)
mice into thy1 positive RFC which are
1
more sensitive to inhibition by Al.
Briefly spleen
rosette-forming cells from ATX mice have a low sensitivity
1
to inhibition by Al in comparison with normal spleen RCF.
1
Thymie peptides or thymulin content of serum correct this
abnormality after 60 min in vitro incubation with spleen
1
cells. Normally several dilutions of samples are used.
The
highest dilution of a serum sample which induces
the
1
rosette inhibition by Al is considered as the active
dilution.
1
To differenciate thymulin activity from that of AF,
1
1
each plasma sample has been analyzed directly and after
specifie immunoadsorption with a monoclonal anti-thymulin
1
1
t
antibody obtained in mice (29). Briefly sera under study
Î
were incubated during 30 min at 37°C with the monoclonal
1
antibody.
The residual activity was evaluated in the rosette
1
1
assay after filtration of the mixture through Amicon
1
~
Membrane. AlI the determinations wers carried out in
!i
1
1
duplicate and the results were expressed as 10g-2 reciprocal
!t
titer of the highest active dilution.
i
1
!i1
f
1
l
1
1

---

1
l
1
-157-
1
1
f1
1
1
Statistical analysis of data:
Differences in mean values
1
1
between the groups were determined by analysis of variance.
1
Specifie tests such as the Duncan and the Tukey tests were
1
1
also used
(30 ).
t
!f
1
!
RESULTS
~
t
1
Anthropometrie and haematologic measurements of the
1
1
children are shown in table 1. The average age was similar
t
1
in the four groups.
According to the classification proposed
f
by Waterlow (31), nearly aIl the children in the group M
1
1
were wasted : most of them exhibited weight for height
r
1
1
values less than 80 'l. of the NCHS standards, except 3 which
1
were edemic and 1 who suffered from measles. Eighteen
1
children from this group were wasted and stunted
(height for
age less than 95 'l. of the reference). For the parameters
1
1
weight/age and weight/height, no significant differences
1
1
were found between the
group 1 and CD ; for these
measurements, the Tukey's test separated the four groups
1
into 3
(MIl, CD and CP ) and 2
(M,
and IICDICP ) distinct
categories, respectively.
Three children in the group land
1
2 in the group CD were mildly wasted
<weight for height
1
ranging between 80 - 90 'l. of the reference).
None of the
children in both groups was stunted.
1
Anthropometrie measurements of the group CP were within
the normal range.
Physical development was better achieved
1
in this group compared to the group CD.
1
1
1

---

1
l
1
-158-
1
1
fi
1
1
Most of the children, excluding the group CP were
•
!
1
anemic ; however,
anaemia was more pronounced in children
from the group M and CO than in those from the group 1. Mean
1
values for red blood cell counts were significantly less in
the group M than in the group l, CO or CP, while the white
1
blood cell counts were significantly higher only in the
1
groups M and 1.
The various degrees of malnutrition and infection are
1
reflected in the values obtained for plasma proteins. Table
2 summarizes the measures for the four groups.
Oepressed
1
levels of albumin,
prealbumin and transferrin and increased
1
levels of orosomucoid and C-reactive protein
(CRP)
were
f
found in the group M.
There was no difference in the mean
1
1
plasma albumin level between the groups CC and CP. However,
the prealbumin level was significantly lower in the group CC
1
than in the group CP.
Albumin and prealbumin concentrations
were significantly less in the group 1 compared to the group
1
CD or CP.
The plasma transferrin level was decreased only in
1
the group M ;
i t was increased in the group CD.
As observed in the group M,
increased values of
1
orosomucoid and CRP were also apparent in the group 1.
In
the clinically uninfected children
(CD and CP ) CRP was
1
undetectable.
The mean orosomucoid concentration was
1
significantly lower in the group CP than in the group CD
(P<O.OS).
1
1
1

---

1
1
-159-
1
1
Plasma protein concentrations of the subject used as
1
"internaI reference" are reported in table 3. AlI the
measurements were within the normal range for age.
1
Tables 4-1 and 4-2 show the distribution of lymphocytes
from the peripheral blood into B-cells, T-mature,
1
helper-inducer, cytotoxic-suppressor cells and the ratio
1
helper-inducer/cytotoxic-suppressor. As i t was necessary
that the immunofluorescence subpopulation counts be done by
1
one and the same technician, these counts were carried out
only in children living in Dakar
< groups M,land CD ).
1
Wide variations were observed within the groups, so that
1
individual and mean values were shown. There were no
significant differences in the mean values between the 3
1
groups for the percentage of B,
T lymphocytes or their
subsets, as weIl as for the ratio helper/cytotoxic. The
1
Tukey's test was unable to separate the groups into several
1
categories. However,
within each group, anomalies of
helper/cytotoxic ratio were observed :
this ratio was either
1
markedly depressed
<range 0.1 to 0.8) or increased
<)2). The
Band T-cell counts of the subject used as "internaI
1
reference" were within the normal range <table 3).
1
Results of the rosette assay are reported in fig.l and
2 and are expressed as 10g-2 reciprocal titers.
No
1
significant differences were observed in the total plasma
thymulin-like activity between the groups M,land CD and
1
1
1

---

1
1
-160-
1
1
between the groups l, CD and CP ; this activity is
1
significantly decreased in the group M only wh en compared to
the group CP (P<O.03).
A significant residual thymulin-like
1
activity ("allogeneic"
factor),
measured after adsorption of
the plasma samples with monoclonal anti-thymulin antibodies,
1
was noticed in the groups M,land CD compared to the group
1
CP ; i t was also observed in the subject used as "internaI
reference"
(table 3).
In contrast, the specifie thymulin
1
activity (fig.2)
was markedly reduced in the groups M and 1
/
and most of these children exhibited undetectable
1
activities. Thymulin was present in the plasma of children
1
from the group CD but the level was low compared to that of
children from the group CP.
1
DISCUSSION
1
We have previously published an absence of variation in
1
thymulin activity
(Zn-FTS)
in moderately and severely
malnourished and infected Senegalese children
(23).
In this
1
previous study AF was not measured. The present
investigation is thus an attempt to further understand the
1
effects of malnutrition/infection interaction on specifie
1
plasma thymulin activity and AF production.
One of the intrinsic difficulties in evaluating the
1
immune response of undernourished children from developing
countries is to separate the effects of malnutrition from
1
that of synergie infections. The complexity of the
1
1
J

---

1
1
-161-
1
1
interaction between dietary inadequacy, diseases and
1
environmental context makes i t difficult to establish
whether the impairment observed is a specifie result of a
1
nutritional problem or whether i t is secondary to the
presence of infection and/or environmental stimuli. The
1
present investigation comprises Senegalese children living
1
in their local environment in Dakar and age-matched
Senegalese children living in Paris.
Four groups of children
1
were defined,
first by clinical examination and further by
anthropometric and biochemical measurements. The degree of
1
weight for age loss, height and weight for height
1
retardations plus the determination of plasma transferrin,
albumin and prealbumin levels were used to diagnose
1
1
protein-energy malnutrition.
Presence or absence of
è
t
1
1
infection and/or inflammatory states was confirmed by
1
informations from the white blood cell counts and the
11
1
measurement of two acute phase reactant proteins
1
(
(orosomucoid and CRP).
The acute phase proteins are plasma
1
proteins submitted to a characteristic pattern of variation
1
which occurs in response to different forms of infection,
!
1
1
inflammation or tissue damage (32). They are consistently
1
[
1
elevated during these states. A recent study has shown that
increased levels of orosomucoid and CRP correlates with the
t
1
1
presence of infection or inflammatory reaction with a 99 1-
1
certainty
(33).
The uninfected children
(CD, and CP groups)
l
1
f
!!
1
t
1
1

---

1
l
,1
-162-
1
1
1
J
,ff
1
t
of our study did not show elevation in either CRP or
1
orosomucoid. Although the mean orosomucoid concentration was
1
significantly higher in the apparently healthy chidren
1
living in Dakar than in those living in Paris,the values
exhibited were within the normal range
(39,
40).
1
Anaemia was common in aIl children,
except in the
1
children living in Paris
(group CP).
In accordance with
f
previous reports
(34,
35), the plasma transferrin level was
!
1
markedly reduced in the group M (severely malnourished)
and
~
was high only in the most anemic,
apparently healthy
1
children
(group CD).
[
1
Among the visceral proteins used as nutritional
markers,
prealbumin has been proposed as the most sensitive
1
for early detection of malnutrition
(35 -
37).
Yet,
in the
presence of superimposed inflammatory (38)
or infectiou$
1
(33)
states,
lowered plasma prealbumin corresponded to
1
elevated levels of acute phase reactant proteins.
This often
makes difficult the evaluation of the specifie effect of
1
nutritional deficiency.
However,
in this study the group CD
(apparently healthy children living in Dakar), who could
1
otherwise be regarded as a control group,
showed reduced
1
levei of prealbumin without elevation in either CRP or
orosomucoid.
Thus,
the decreased plasma prealbumin in the
1
group CD was not a reflection of an acute phase response.
Information about the individual values of prealbumin,
1
1
1

---

1
1
-163-
!
1
1
f
1
orosomucoid and CRP suggested that most but not aIl children
1
in the group CD suffered from mild to moderate
protein-energy malnutrition. Support for this view comes
1
1
from our recent experimental studies :
in animal model of
f[
pure moderate protein-energy malnutrition,
the plasma
l
1
t
concentration of prealbumin decreases but the acute phase
1
1
proteins are not elevated unless the animaIs are also
infected
(Wade et al unpublished data).
Thus, we suggest,in
1
agreement with previous reports
(33), that new information
can be gained by simultaneous measurements of prealbumin,
1
r
CRP and orosomucoid in the evaluation of nutritional status.
1
1
Children in the group l were marginally
malnourished and
also infected. Their concurrent infections explained the
1
rather low level of prealbumin observed.
Anthropometrie and biochemieal data showed that children in
1
the group M were severely malnourished and chronically
infected. The children in the group CP were healthy,
well-
1
nourished and uninfected.
1
In accordance with our previous study
(23), moderate
1
or severe malnutrition was not associated with changes in
1
1
total plasma thymulin-like activity.
Decreased thymulin
level has been reported in
malnourished children
(20,
22)
1
1
[
and in anorexia nervosa patients without superimposed
1
infections (41).
In both studies,
thymulin activity was
determined by measuring the total activity expressed in the
1
t
1
1
1

---

1
1
-164-
1
l
1
1
rosette assay.
However,
activated T-cells may secreted an
1
1
"allogeneic" factor
(AF)
which is a peptide of low molocular
weight.
AF is distinct from thymulin antigenically and by
1
electric charge.
AIso,
i t is not inhibited by the specifie
serum thymulin inhibitions (24).
However,
in the rosette
1
assay,
AF and thymulin are indistinguishable. Therefore,
1
presence of AF in the serum of children may be interpreted
as an increase of thymulin.
We overcame this error by
1
introducing adsorption with monoclonal anti-thymulin
antibody.
Hence we could measure specifically the AF level.
1
This study demonstrated the significant contribution
1
of AF on the total activity measured : Specifie thymulin
activity
(total activity minus AF activity) was nearly
1
undetectable in the severely malnourished and infected
children as weIl as in the moderately malnourished and
1
infected children,
while their total
activity was similar to
1
those of the control children
(group CP).
However,
other
environmental stimuli may also induce AF production since AF
1
was detectable in few children from the healthy group
(CP)
and in aIl children from the group CD;
these children,
1
although uninfected produced a significant amount of AF,
1
quite similar to that recorded in the subject used as
"internaI reference".
This subject was a healthy adult man
1
from a high socio-economic class living in Dakar.
1
1
1

---

1
-165-
1
f
1
1
1
Nevertheless, specifie thymulin activity was lower in the
1
apparently healthy children living in Dakar than in the
healthy children living in Paris. This might be the result
1
of the effect of moderate malnutrition, but such assumption
needs to be confirmed by screening well-nourished,
1
uninfected children living in the same environment as
1
age-mached uninfected, moderately malnourished children.
In
addition, although the children in the group 1 were markedly
1
less undernourished than those in the group M,
their
thymulin activity was similarly depressed.
1
The percentage of B-Iymphocytes, T-Iymphocytes and the
1
subsets,
identified by surface markers bore little relation
to the plasma specifie thymulin activity.
It seems likely
1
that many variables may influence the number of T-cells and
their distribution. We have reported such discrepancy in
1
anorexia nervosa patients (41)
and lack of correlation
1
between T-cell function and phenotype has been shown by
others <42,
43). These results suggested that,
although
1
thymie hormones may promote most T-cell functions in
malnourished subjects <11,
12) or animaIs
<44,
45),
1
decreased plasma activity is not necessarily associated with
1
phenotypic modifications.
Our study demonstrated that in protein energy-
t
f
1
malnutrition, the level of thymulin activity is highly
t
influenced by concurrent infections and/or stimulation by
1
1
1

---

r}
1
1
-166-
1
1
t1
1
local environmental factors.
It also suggested that severe
1
malnutrition is not the only underlying mechanism of
depressed level of thymulin in malnourished and infected
1
1
children.
1
1
!
l
1
1
1
1
1
~i
1
1
1
f[
1
1
1
1
1
1
1
1

---

1
l
'1
-167-
f
!
1
r
1
1
'1
1!
1
1
t
ACKNOWLEDGMENTS
1
1
t!\\
1
f
The authors acknowledge the parent·s children for their
~!
1
cooperation.
They thank the Medical Staff of the PMI
(Protection
Maternelle et Infantile) Medina in Dakar especially Dr Nakoulima,
Dr Mbow and Dr Santos. They are also grateful to Dr Desjeux JF
1
and Mrs Maud G for their medical assistance in Paris.
1
1
1
1
1
1
1·
1
1
1
1
1
f
1
1
1

---

1
1
-168-
1
1
1
References
1
1
1
Scrimshaw NS,
Taylor CE, Gordon JE.
Interactions
1
of Nutrition and Infection.
Geneva ~
World Health
Organization.
Monogr Ser No 57,1968.
1
2
Suskind RM.
Malnutrition and the immune
1
response.New-York,
NY:Raven Press~ 1977.
1
3
Chandra RK,
Newberne PM.
Nutrition,
immunity and
1
infection.
New-York,NY:Plenum,
1977.
1
4
Sell RG.
Undernutrition,
infection and immunity :
the role of parasites.
Papua New Guinea Med J
1978
1
;21:43-55.
1
5
Gross RL Newberne PM.
Role of nutrition in
1
immunologie functions.
Physiol Rev
1980;60:188-302.
1
1
6
MC Murray DN.
Cell-mediated immunity in
nutritional deficiency.
Progress in Food and Nutr
1
Sc 1984;8:193-228.
1
1

---

1
1
-169-
1
1
1
7
Mugerwa JW.
The lymphoreticular system in kwash-
1
iorkor.
J Pathol
1971;105-9.
1
8
Jambon B,
Zeigler 0,
Duheille J et al.
Fonction
1
hormonale lymphodifferenciatrice du thymus et
malnutrition protéino-énergétique chez l'enfant.
1
Serie colloques INSERM,
1986.
(In Press).
1
9
Mittal A,
Woodward B.
Thymie epithelial cells of
1
severely undernourished mice
Accumulation of
cholesteryl esters and absence of cytoplasmic
1
vacuoles.
Proc Soc Exp Biol Med 1985;178:385-91.
1
10
Mittal A,
Woodward B.
Ultrastructural and
morphometric analysis of thymie epithelial
1
secretory vacuoles in severely protein-energy
1
malnourished weanling mice.
Nutr Res
1986;6:663-71.
1
1
11
Jackson TM,
Zaman SN.
The in vitro effect of the
1
f
thymie factor Thymopoietin on a subpopulation of
1
lymphocytes from severely malnourished children.
!
Clin Exp Immunol
1980;39:717-21.
1
1
,
1
1
1

---

1
1
-170-
1
1
1
12
Olusi SO,
Thurman GB, Goldstein AL.
Effect of
1
thymosin on T-Iymphocyte rosette formation in
children with kwashiorkor. Clin Exp Immunopathol
1
r
1980;15:687-91.
1
1
13
Bach JF,
Dardenne M,
Pleau JM et al.
Isolation,
1
l
biochemical characteristics, and biological
activity of a circulating thymic hormone in the
1
mouse and in the human. Ann NV Acad Sci
1
1975;249:186-210.
1
f
1
14
Dardenne M,
Pleau JM,
Blouquit JV and Bach JF.
f
Characterization of facteur thymique sérique (FTS)
1
in the thymus.
II. Direct demonstration of the
1
presence of FTS in thymosin fraction V. Clin Exp
Immunol
1980;42:477-82.
1
15
Monier JC,
Dardenne M,
Pleau JM et al.
1
Characterization of facteur thymique sérique (FTS)
1
in the thymus.
1. Fixation of anti-FTS antibodies
on thymic reticulo-epithelial cells. Clin Exp
1
Immunol
1980;42:470-6.
1
1
1
!
1
1
i
1
1

---

1
l
1
-171-
1
1
!
1
1
16
Jambon M, Montagne B,
Bene MC et al.
1
Immuno-histological
localization of "facteur
thymique sérique"
(FTS)
in human thymie
1
epithelium. J
Immunol
1981;127:2055-9.
1
17
Schmitt D,
Monier JC,
Dardenne M et al.
1
Cytoplasmic localization of FTS (facteur thymique
sérique)
in thymie epithelial cells. An
1
immunoelectron-microscopical study.
Thymus
1
1980;2:177-86.
1
18
Bach JF.
Thymulin
(FTS-Zn). Clin Immunol Allergy
1983;3:133-56.
1
19
Dardenne M,
BACH JF.
The sheep cell rosette assay
1
for the evaluation of thymie hormones. Kooyker,
1
Rotterdam:Van Bekkum,
1975;235-43.
1
1
20
Chandra RK. Serum thymie hormone activity in
1
!
protein-energy malnutrition. Clin Exp Immunol
1
1979;38:228-30.
1
1
1
1
1
/[
1
1

---

1
1
-172-
1
t
1
1
f
1
~
1
21
Chandra RK. Serum thymie hormone activity and cell
1
mediated immunity in healthy neonates,
preterm
infants,
and small-for-gestional age infants.
1
Pediatries 1981;67:407-11.
1
22
Chandra RK.
Numeral and functional deficiency in T
1
helper cells in protein energy malnutrition.
Clin
Exp Immunol
1983;51:126-32.
1
1
23
Maire B, Wade S, Bleiberg F et al.
Absence of
variation in Facteur Thymique Serique activity in
1
moderately and severely malnourished Senegalese
children.
Am J Clin Nutr 1982;36:1129-33.
1
1
24
Dardenne M,
Bach JF.
Demonstration and
characterization of a serum factor produced by
1
activated T cells.
Immunology 1977;33:643-51.
1
25
Safai B, Dardenne. M.
Incefy GS et al.Circulating
1
thymie factor,
facteur thymique sérique (FTS) ,
in
mycosis fungoides and Sezary syndrome.
Clin
1
Immunol
Immunopathol 1979;13:402-6.
1
1
1

---

1
1
-173-
ti
1
f
1
1
1
1
f
26
National Center For Health Statistics : NCHS.
1
Growth charts monthly vital statistics report.
Vol
1
f
25,
nO 3, suppl
(HRA)
76-1120.
Rockville,
MD
f
1
f
Health Resources Administration, June 1976.
t
f
1
27
Mancini G, Carbonara AD,
Heremans JF.
1
Immunochemical quantitation of antigens by single
l
radial
immunodiffusion.
Int J
Immunochem
1
1
1
1965;2:235-59.
1
1
t
28
Bach MA,
Bach JF.
The use of monoclonal anti-T
r
t
1
cell antibodies to study T cell imbalances in
human diseases. Clin Exp Immunol
1981;45:449-56.
1
1
1
29
Dardenne M,
Pleau JM, Savino W et al.
Monoclonal antibody
1
t
against the serum thymie factor
(FTS).
Immunol letters
f
1
!
1982;48:61-4.
[
f
1
30
Tukey JW.
In Bennett CA, Franklin NL.
Statistical
1
Analysis in Chemistry and the chemical
Industry.
1
1
!
New York:
John Wiley,
1954;339-44.
!
1
1
1
1
1
1

---

1
1
-174-
1
1
1
31
Waterlow JC.
Classification and definition of
1
protein-calorie malnutrition.
Br Med J
1972;3:566-9.
1
1
32
Pepys MB,
Baltz ML.
Acute phase proteins with
Special reference to c-reactive protein and
1
related proteins
(Pentaxims)
and serum amyloid A
protein.
Advances in Immunol
1983;34:141-212.
1
1
33
Ingenbleek Y,
Carpentier VA.
A pronostic
inflammatory and nutritional
index Scoring
1
critically ill patients.
Internat J Vit Nutr Res
1985;55:91-101.
1
1
34
Reeds PJ, Laditan AAO.
Serum albumin and
transferrin in protein-energy malnutrition Br J
1
Nutr 1976;36:255-63.
1
35
Ingenbleek Y,
Van Den Schrieck HG,
De Nayer PH et
1
al.
Albumin,
transferrin and the thyroxine-binding
prealbumin/retinol-binding protein
(TBPA-RBP)
1
complex in assessment of malnutrition.
Clin chim
Acta 1975;63:61-7.
1
1
1

---

1
1
-175-
1
1
1f
1
1
36
Smith FR, Suskind R,
Thanangkul 0 et al. Plasma
1
vitamin A, retinol-binding protein and prealbumin
1
concentrations in protein-calorie malnutrition.
1
III-Response to varying dietary treatments.
Am J
1
Clin Nutr 1975;28:732-8.
1
37
Shetty PS, Watrasiewicz KE,
Yung RT et al. Rapid
turnover transport proteins : an index of
1
subclinical protein-energy malnutrition.
The
1
Lancet 1979;ii:230-2.
1
38
Bleiberg-Daniel F, Wade S, Labarre C et al.
Variations in plasma thyroxine-binding prealbumin
1
(TBPA)
in relation to other circulating proteins
1
in post-operative patients during rapid oral
refeeding.
Human Nutrition: Clin Nutr 1985;39C:55-62.
1
39
Liappis N, Robmoller M, Dorner K et al.
1
Referenzwerte der IgA,
Ig6,
IgM, a -Makroglobulin
2
1
a -antitrypsin, C3-Komplement, C4-Kompl ement ,
1
Coeruloplasmin, Haptoglobin,
properdin faktor B,
1
Saur en a -glykoprotein and Transferrin
1
Konzentration im Serum von gesunden Kindern.
Klin
1
Padiat 1983;195:107-16.
1
1

---

'1
-176-
1 Il
il
1
II
f
40
Herbeth B,
Henny J, Siest G.
Variations
1
biologiques et valeurs de reference de la
1
transferrine,
des immunoglobulines A et de
l'orosomucoïde.
Ann Biol Clin 1983;41:23-32.
1
41
Wade S,
Bleiberg F,
Mosse A et al.
Thymulin
1
<Zn-Facteur Thymique Serique)
activity in anorexia
nervosa patients. Am J Clin Nutr 1985;42:275-80.
1
42
Bach JF,
Chatenoud L.
The significance of T-cell
subsets defined by monoclonal antibodies in human
1
diseases. Ann Immunol
<Inst.
Pasteur)
1982;1330:131-6.
1
1
43
Tjernlund U,
Cesaro P,
Tournier E et al.
T-cell
subsets in multiple sclerosis :
a comparative
1
study between cell surface antigens and function.
1
Clin Immunol
Immunopath 1984;32:185-97.
1
44
Petro TM, Chien G, Watson RR.
Alteration of
cell-mediated immunity to Listeria monocytogenes
1
in protein-malnourished mice treated with thymosin
fraction V.
Infect Immun 1982;37:601-8.
1
1
45
Watson RR, Chien G, Chung C.
Thymosin treatment
serum corticosterone and lymphocyte mitogenesis in
1
moderately and severely protèin-malnourished mice
J Nutr 1983;113:483-93.
1

---

~"F'."
' " .
~ ;",",'
',-
~':r--"'"
..~'
' ..,' -.-
..
..,..,
-_ •• ~ •• --------------
TABLE 1
Description of the subjets grouped according to their clinical exalination (Ail subjects are Senegalese children)
================================================================================================================================
U
U
Groups •
Age
Sex ratio
lleight
lleight
Height U
Red blood
Helatocrit Heloglobin
Leucocyte
for age
for height
for age
cells
counts
3
(Ionths)
IOb
3
"If
x
x
x
/l13
1
g/dl
10 /11
t
t
t
t
t
t
t
"
17.3
17112
SB
b7
92
3.73
2B
B.5
14.7
In=29)
~ 1.3
~ 2.2
! 1.9
! 1.1
~ O.IB
! 1.2
! 0.3B
! 1. 7
--------------------------------------------------------------------------------------------------------------------------------
1
t
t
t
......
-..-J
1
lb.5
b/3
90
93
97
3.9b
32
9.9
14.1
-..-J
1
(n=9)
+ 1.6
! 3.2
! 2.2
~ 0.9
! O.IB
! 1.1
! 0.5b
+ 1.7
t
t
CD
15.3
bl7
9b
99
99
4.23
31
9.0
B.9
(n=13)
! 1.5
! 3.2
+ 2.5
! 1.0
! 0.16
! 1.4
! 0.51
! 0.6
CP
IB.I
5/2
110
105
104
4.2B
33
11.3
9.9
In=7)
! 2.7
+ 1.9
+ l.b
! 0.9
! 0.11
! 1.0
! 0.34
! 1.7
"ean ! SE" t P < 0.05
t P < 0.01
versus the group CP
'Group " =children hospitalized for severe lalnutrition i group 1 =children diagnosed as suffering frol cOllon infectious di-
-seases without signs of eitheir khashiokor or "araslus i group CD = Apparently healthy children living in Dakar i group
CP =Apparently healthy children living in Paris.
"1 of the soth percentile of the NCHS values.
_--~_"""_.,_.' __'~~'~_' ., ",'~~"_"""""""'"",,,""~
" ,
....• _.
• "..._~..,. .,'
-,,,,,,
=,,,,,_,,~,,,•.""~'''''''''''''''''''' __-~''''
-''''''___'
....,J

---

- - - - - - - - - - - - - - - - - - - - -
TABLE 2
Plasma protein levels
=====================================================~
==================================================================
Groups
Albumin
Prealbumin
Orosomucoid
C-reactive protein*
Transferrin
(g/l)
(mg/l)
(g/l)
(mg/l)
(g/l)
M
19.2-f.
63-f.
2.9l-f.
6 -- 155
1.1-f.
(n = 29)
.± 1. 3
± 5.6
± 0.13
Undetectable
± 0.11
1
f-'
in 4 subjects
-...J
CD
1
-f.
-f.
l
106-f.
31.6
1. 45
13 -- 139
3.5
(n = 9)
.± 2.0
± 9.1
± 0.18
undetectable
± 0.23
in 4 subjets
CD
35.0
149J:.
0.87r-
undetectable
4.5-f.
(n = 13)
.± 1.1
± 7.1
± 0.06
± 0.11
CP
31.9
194
0.57
3.2
(n = 7)
.± 1. 1
± 9.9
± 0.04
undetectable
± 0.21
Means ± SEM
t p < 0.5
-f. p < 0.01
Versus the group CP
*Minimum and maximum values are reported .
~"'"
~
_"_"_~._. .~",,,""_'
,~_~~.~~ _.~
.'_~ n ~
...
".
' P , . " ,
•
_
,
_
,••_ . _ "
..•..•
M
• •
_ _
. . .
. .

---

---------------------
TABLE 3
Biochelical and illunological leasurelents of the subject used as -internai reference-.
=====================================================:=====:===================================:==========================================
Red blood
Heutocri te
Heloglobin
Leucocyte
Albulin
Prealbulin
Transferrin
Orosolucoid
cells
counts
6
3
3
3
10 /11
1
g/dl
10 /11
gll
Ig/l
g/l
g/l
1
4.2
43
15
4.3
46.3
324
2.4
0.51
1-'
-.J
\\D
1
CRP
Total rosette
AF activity
51g- Bearing
nature
Helper
cytotoxic
Ratio
acti vit y
B-cells
T-cells
(a)
5uppressor (b)
~
b
Igl1
Log-2
Log-2
1
1
1
1
o
5
3
8
88
42
26
1.6
.......
~.""""i"''''''_'''"''''''''''"''"''f.'''''''~_''",
,,~
~~_,
""'" "
1"". " • . ''''!'. • . K~Jl'!"";O
_, M W , \\ < @ . ' P A " - - · - · - - " _ " ! r , , , ! $ , * , , , · · m ,
MX
.~.o"'
p
~,~.",,,,,
LI~"'7lII!M".",,,,."'r or llF,~ ,,-,,)1... li! i..
))%1

---

---------------------
TABLE 4.1
Dislribulion of peripheral ly.phocyles inlo several subsets
=====================::=========:==============================================================
Groups
5lg-Bearing
!tature
Helper
Cytotoxic
Ratio!
8-cell s
T-cells
la)
suppressor lb)
b
"
1
9
45
38
17
2.2
2
24
52
49
30
1. b
3
7
79
50
3b
1.3
4
41
70
18
33
0.5
1
5
2b
58
55
lb
3.3
i-'
co
b
12
28
5
19
0.2
0
1
7
Il
44
30
19
1.5
8
10
b8
5b
Il
5.3
9
19
81
3
30
0.1
10
9
77'
3
29
0.1
Il
lb
42
7
28
0.2
12
15
58
53
24
2.2
IJ
37
b3
58
14
4.2
If
9
33
25
28
0.8
15
19
59
13
4b
0.3
lb
10
37
3
Il
0.2
17
7
If
Il
12
0.8
18
lb
22
10
17
0.5
19
8
42
b
5
1.0
20
3
b8
12
29
0.4
21
19
58
14
34
0.4
22
lb
32
18
2b
O.b
23
12
55
9
25
0.3
-------------------------------------------------------.---------------------------------------
!teans
15
52
24
23
1.2
! SE"
! 1.9
! 3.9
! 4.2
! 2.0
! 0.29
~ M~,~~,
----::-t-~---:~:~,---------:------------p------~:.-::--_:---7--7:--~---~~:~~;---------;-:~~:,~,:-----::.~-~
_ _ _ _ _ _,
,.
ft
.9##
Qi,
:
;
*
~,&~~

---

---------------------
TABLE 4.2
Distribution of peripher.l ly.phocytes into several subsets
===========================================================================================
Groups
Slg-Bearing
"ature
Helper
Cytotoxic
btio!
B-cells
T-cel1s
(a)
suppressor(bl
b
-------------------------------------------------------------------------------------------
1
7
85
66
13
4.9
2
9
68
60
20
3.0
3
6
40
4
34
0.1
4
21
30
9
19
0.4
5
17
72
41
28
1.4
6
17
64
45
35
1.3
1
......
7
11
87
42
20
2.0
co
......
8
31
41
33
11
2.9
1
-------------------------------------------------------------------------------------------
.
"eans
15
61
38
23
2.0
~ SE"
~ 2.9
~ 7.6
~ 7.8
~ 3.1
~ 0.55
--------------------------------------------------------------------------------------.----
CO
1
18
69
42
36
1.1
2
36
55
27
25
1.0
3
13
80
42
19
2.1
4
27
50
45
19
2.3
5
23
62
14
58
0.2
6
44
71
47
27
1.7
7
28
63
49
24
2.1
8
20
51
39
15
2.6
9
4
83
44
39
1.1
10
4
50
30
20
1.5
"uns
21
63
38
28
1.6
! SE"
+ 4.2
~ 3.9
~ 3.5
~ 4.1
~ 0.23

---

---

1
-183-
1
Fig. 2
1
Thymulin activity (log-2)
1
o
~
~
w
~
~
-~-~i----,..--~i-....."
r - i
~I
1
•
1
~ 1
1
1
1
•
•
1
1
1
•
1
1
1
1
1
•
~ 1
1
1
1
1
1
1
1
1

---

1
1
-184-
1
1
1
1
LEGEND FIGURES
,1
Fig.
1
Total plasma activity on the rosette a5say (AF and
1
thymulin activities).
Individual values were
reported.
1
Fig.
2
Specifie plasma thymulin activity (total activity
minus AF activity).
Individual values were
reported.
1
1
1
1
1
1
1
1
1
1 •
1
1

---

•
-185-
1
1
1
1
1
1
1
1
1
1
CONCLUSIONS GENERALES
1
1
1
1
1
1
1
1
1
1

---

1
,
-186-
1
1
1
1
r
1
CONCLUSIONS GENERALES
1
1
INTERET DU DOSAGE SERIQUE DE LA TRANSTHYRETINE DANS LA MPE
1
MARGINALE OU MODEREE
1
L'isolement et la purification de la transthyrétine de rat
ont permis la fabrication d'un anticorps mono-spécifique anti-
1
transthyrétine de rat et la mise au point de son dosage dans le
1
sérum et dans le liquide céphalorachidien.
La transthyrétine de rat isolée et purifiée se présente
1
comme une espèce moléculaire unique, bien définie et très
similaire de la transthyrétine humaine.
La protéine isolée
1
conserve sa capacité de liaison avec la thyroxine et donc sa
1
fonction de transporteur.
Comme chez
l'homme,
les taux sériques de transthyrétine
1
sont plus faibles à la naissance qu'à l'âge adulte.
De l
à 28
jours d'âge,
les rats mâles et femelles ont des taux comparables
1
qui,
par la suite,
chutent uniquement chez les femelles.
1
Au cours de la restriction alimentaire protéique ou
1
énergétique,
la concentration sérique de la transthyrétine baisse
t
1
en dehors de tout état inflammatoire ou infectieux,
apprécié par
1
le dosage de l'orosomucoïde et de l'a
-macroglobuline, protéines
1
2
majeures de la réaction inflammatoire chez le rat.
Cette baisse,
1
1
1
contrairement à d'autres marqueurs nutritionnels
(albumine,
transferrine),
est directement reliée aux ingesta protéiques et
1
énergétiques,
et non pas à la composition du régime.
1

---

1
-187-
La concentration sérique de la transthyrétine est très
1
influencée par une insuffisance d'apport énergétique et ceci, même
1
si l'apport protéique est adéquat. Ce fait est d'autant plus
important que les enquêtes nutritionnelles dans les pays en voie
1
de développement
(PVD)
montrent que la grande majorité des enfants
souffrent d'une insuffisance énergétique globale.
1
Contrairement au taux sérique,
le taux de transthyrétine
1
dans le liquide céphalorachidien est peu sensible aux apports
nutritionnels.
1
La transthyrétine du rat est très influencée par la
1
restriction alimentaire totale.
Elle est abaissée après 24 h de
jeûne et reste basse après deux jours de réalimentation.
Le taux
1
hépatique d'ARN messager, mesuré à l'aide d'une sonde d'ADN
complémentaire spécifique de la transthyrétine,
augmente dès le
1
premier jour de réalimentation,
suggérant que la restriction
1
alimentaire totale ne modifie pas seulement la synthèse hépatique
de la transthyrétine,
mais probablement aussi sa distribution
1
et/ou son catabolisme.
La transthyrétine est le principal transporteur des
1
hormones thyroïdiennes chez le rat. Sa chute,
lors de la carence
1
énergétique ou de la restriction alimentaire totale,
s'accompagne
d'une modification de la distribution des hormones thyroïdiennes
1
et particulièrement de la fraction liée de la thyroxine.
1
Chez l'homme,
au cours d'une inflammation sans
1
dénutrition,
la concentration plasmatique de la transthyrétine
1
baisse en même temps qu'augmente
(en image miroir)
celle de
l 'orosomucoïde, de la protéine C-réactive, de l'a l-anti-trypsine.
t
1
La mesure de ces protéines est un moyen d'apprécier l'influence
l!
1
concomitante de l'inflammation et/ou de l'infection sur l'état
nutritionnel.
t
1
L'ensemble de ces résultats suggère que le dosage de la
1
1
l
1
i

---

1
-188-
1
transthyrétine sérique est un indicateur sensible et spécifique de
-4
la MPE marginale ou modérée.
1
Comme la plupart des protéines plasmatiques,
le dosage de
1
la transthyrétine est simple, et peut se réaliser par une micro-
méthode à partir d'un prélèvement de sang capillaire.
1
La détermination de l'état nutritionnel des enfants dans
les PVD devrait comporter,
en plus de l'anthropométrie (âge,
1
poids, taille) et du dosage de l'albumine sérique, la
1
détermination simultanée des taux sériques de la transthyrétine,
de l'orosomucoïde et de la C-réactive protéine. Ces mesures
1
pourraient permettre de mettre en évidence un état de MPE précoce
chez des enfants apparemment bien portants.
1
1
FONCTION THYMIQUE DANS LA MPE
1
Au cours de la carence marginale en zinc, les cellules
épithéliales du thymus augmententleur capacité à produire de la
1
thymuline qui reste cependant biologiquement inactive.
L'addition
1
de zinc in vitro dans le sérum des animaux carencés restaure
l'activité biologique de la thymuline,
suggérant que cette
1
activité est fortement dépendante en zinc.
Chez des enfants dénutris et infectés, on n'observe aucune
1
variation de l'activité de la thymuline,
lorsque le dosage
1
biologique est effectué directement sur le sérum total, et ceci,
quel que soit le degré de malnutrition.
Cependant,
la thymuline
1
est abaissée dans l'anorexie mentale, MPE non associée à des
1
f
infections; cette baisse n'est pas due à une carence en zinc.
f
1
1
t
L'absence de variation de l'activité de l'hormone
1
i
1
thymique, observée chez les enfants dénutris et infectés,
résulte
de la présence des infections concomitantes. En effet
leurs
,
1
cellules T "activées" sécrètent un facteur appelé "facteur
allogénique" ayant la même activité biologique que la thymuline
1
1

---

1
-189-
[
1
dans le test des rosettes.
f(
En l'absence d'une méthode de détermination non biologique
1
directe de la thymuline,
l'activité spécifique de cette hormone
1t
1
dans le sérum est évaluée indirectement après adsorption du sérum
t
i
par un anticorps anti-thymuline. Chez des enfants normaux ou
1
~
présentant une MPE sévère ou modérée, associée ou non à des
infections,
l'adsorption du sérum par un anticorps monoclonal
1
1
anti-thymuline permet de mettre en évidence:
1
1- une chute de l'activité spécifique de la thymuline dans
la MPE sévère et modérée.
1
2- la présence d'une quantité importante de "facteur
allogénique" dans le sérum des enfants infectés.
i
1
n'est pas le
1
seul
1
trouve aucune
1
1
différence de l'activité
thymique chez des
.=.......;;;;.,~~....:::;;.,.~-..;;;;...IO:....:~~~~
1
enfants sévèrement dénutris et
à des enfants
1
1
dénutris,
cette activité spécifique est plus importante chez les
non-infectés que chez les infectés, suggérant que la présence
1
d'infections est un élément important qui influe fortement sur la
fonction épithéliale du thymus.
1
L'évaluation directe de l'activité biologique de la
1
thymuline dans le sérum peut donc être faussée par la présence
d'infections concomitantes.
1
La MPE, de même que les infections,
affectent la fonction
hormonale du thymus.
La diminution de l'activité de la thymuline
1
pourrait expliquer l'atteinte de certaines fonctions de l'immunité
1
à médiation cellulaire dans la MPE.
1
1

---

~I
'1
l'
1
!I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

---

--
TRANSTHYRETINE ( PREALBUMINE ) ET THYMULINE
DANS LA MALNUTRITION PROTEINO-ENERGETIQUE
RESUME DES TRAVAUX
'.

---

~
1
Chapitre l
Transthvrétine dans la malnutrition protéino-
1
[
énergétiaue
(MPE)
lff
t
Depui~ de nombreuses alnndées, des scien~ifiqUe~. travaillant 1:.•
sur les problemes nutritionne s
es Pays en V01e de Developpement
(PVD),
recherchent des indicateurs suffisamment sensibles pour
1
àétecter l'état de malnutrition protéino-énergétique
(MPE) à un
t
stade précoce avant l'apparition de signes cliniques.
i
La thyroxine-binding prealbumin
(TBPA)
ou préalbumine,
~l
récemment nommée transthyrétine a été proposée comme indicateur
~î~
sensible de la MPE,
dans
la mesure o~ sa concentration
plasmatique,
initialement basse chez des enfants dénutris,
augmente plus rapidement que celle de
l'albumine ou de la
transferrine au cours de la réalimentation.
Cependant,
les
carences nutritionnelles des enfants dans les PVD sont souvent
associées à un état inflammatoire ou infectieux. On ne sait donc
pas à quels facteurs doivent être attribués ces taux bas de
transthyrétine.
Nos objectifs sont d'établir les effets spécifiques de la
malnutrition et de
l'inflammation sur les taux circulants et le
métabolisme de la transthyrétine, et de préciser son rôle et son
utilisation en tant que marqueur de la MPE marginale ou modérée.

---

2.
Ces études ont été menées parallèlement chez l'animal et chez
l'homme.
L'influence spécifique de divers niveaux d'ingesta
protéiques et/ou énergétiques sur la transthyrétine a été testée
chez le rat.
La transthyrétine du rat offre l'avantage remarquable
de présenter,
comme chez l'homme, une espèce moléculaire unique
bien définie et très similaire,
tant par sa structure que par sa
fonction.
C'est une protéine que lIon retrouve dans le sérum et
dans le liquide céphalo-rachidien,
vectrice des hormones
thyroïdiennes
(T3, T4)
et de
la vitamine A par l'intermédiaire de
la protéine liant le rétinol. Toutefois,
la transthyrétine du rat
et celle de
l'homme sont antigéniquement différentes.
Nous avons
isolé et purifié la transthyrétine de rat et préparé chez le lapin
un anticorps monospécifique
(article 1).
La protéine isolée
conserve sa capacité de liaison avec la thyroxine
(T4) et donc sa
fonction de transporteur
(article 1). Comme chez
l'homme,
les taux
sériques de transthyrétine sont plus faibles
à la naissance qu'à
l'âge adulte.
De 1 à 28
jours d'âge,
les rats mâles et femelles
ont des taux comparables qui,
par la suite, chutent uniquement
chez les femelles
(article 1).
Au cours de la restriction alimentaire protéique ou
énergétique,
la concentration sérique de la transthyrétine baisse
en dehors de tout état inflammatoire ou infectieux,
apprécié par
le dosage de l'orosomucoïde et de 11 az-macroglobuline,
protéines
majeures de la réaction inflammatoire chez le rat
(article 2).
Cette baisse,
contrairement à d'autres marqueurs nutritionnels
(albumine,
transferrine)
est directement
reliée aux ingesta

---

3 •
protéique et énergétique des régimes
(article 2).
La concentration
sérique de la transthyrétine est très influencée par une
insuffisance d'apport énergétique et ceci,
même si l'apport
protéique est adéquat
(article·2).
Ce fait
est d'autant plus
important que les enquêtes nutritionnelles dans les PVD montrent
que, la grande majorité des enfants souffrent d'une insuffisance
énergétique globale.
Contrairement au taux sérique,
le taux de transthyrétine
dans le liquide céphalo-rachidien est peu sensible aux apports
nutritionnels
(articles 2 et 3).
La transthynétine du rat est très
influencée par la restriction alimentaire totale.
Elle est
abaissée après 2 h de
jeûne et reste basse après deux
jours de
réalimentation
(article 3)
;
le taux hépatique d'ARN messager,
mesuré à l'aide d'une sonde d'ADN complémentaire spécifique de la
t:ansthyrétine,
augmente dès
le premier
jour de réalimentation,
suggérant que la restriction alimentaire totale ne modifie pas
seulement
la synthèse hépatique de la transthyrétine,
mais
probablement aussi sa distribution et/ou son catabolisme (article
3 ) .
La transthyrétine est le principal
transporteur des
hormones thyroïdiennes chez
le rat.
Sa chute,
lors de la carence
énergétique ou de la restriction alimentaire totale,
s'accompagne
d'une modification de la distribution des hormones thyroïdiennes
et particulièrement de la fraction liée r de la thyroxine (articles
2 et 3).
L'ensemble de ces résultats suggère que le dosage de la
transthyrétine sérique est un indicateur sensible et spécifique de

---

4.
la MPE marginale ou modérée.
Au cours d'une inflammation sans dénutrition,
la
concentration plasmatique de la transthyrétine baisse en même
temps qu'augmente
(en image miroir) celle de 1'orosomucoïde, de la
protéine C-réactive, de l' ~-antitrypsine (article 4). La mesure
de ces protéines est un moyen d'apprécier
l'influence concomitante
de l'inflammation et/ou de l'infection sur l'état nutritionnel
(article 9).
Comme la plupart des protéines plasmatiques, le dosage de
la transthyrétine est simple,
et peut se réaliser par une micro-
méthode ~ partir d'un prél~vement de sang capillaire (article 5).
En conclusion,
la détermination de
l'état nutritionnel des
enfants dans les PVD devrait comporter, en plus de l'anthropo-
métrie (âge,
poids,
taille)
et du dosage de
l'albumine sérique,
la
àétermination simultanée des taux sériques de la transthyrétine,
de
l'orosomucoïde et de la C-réactive protéine.
Chapitre II
Thvmuline dans la MPE
La thymuline ou complexe zinc-Facteur Thymique Sérique
(Zn-FTS)
est une hormone sécrétée par les cellules épithéliales du
thymus, capable d'induire des marqueurs de maturation sur les
lymphocytes T.
Elle poss~de une activité sérique que l'ont peut
mesurer par un test biologique. Cette activité est zinc dépendante
: des animaux carencés en zinc présentent une activité sérique
abaissée, qui est restaurée apr~s addition in vitro de chlorure de
zinc
(article 7).

---

•
•
5 •
•
Les hormones thymiques ont été impliquées dans les
mécanismes de dépression de certaines fonctions de
l'immunité à
~
médiation cellulaire observées dans la MPE,
mais la plupart de ces
f
travaux sont basés sur des Tests in vitro.
Dans ce travail,
les
\\
répercussions de la MPE sur la fonction thymique, évaluée par la
1
mesure de
l'activité biologique sérique de la thymuline, sont
étudiées.
Sachant qu'il existe des interactions MPE,
infections et
immunocompétence, cette activité a été déterminée dans différents
états de MPE associés ou non à des infections.
Chez des enfants dénutris et infectés,
on n'observe aucune
variation de l'activité de la thymuline,
lorsque le dosage
biologique est effectué directement sur'le sérum total, et ceci,
quel que soit le degré de malnutrition (articles 6 et 9).
Cependant,
la thymuline est abaissée dans
l'anorexie mentale, MPE
non associée à des infections; cette baisse n'est pas due à une
carence en zinc
(article 8).
L'absence de variation de l'activité de l'hormone thymique
observée chez les enfants dénutris et infectés pouvait donc
te
résulter de la présence des infections concomitantes.
En effet,
f
les cellules T "activées" sécrètent un facteur appelé" facteur
\\
allogènique" ayant la même activité biologique que la thymuline
1
dans le test biologique.
En l'absence d'une méthode de détermination non biologique
directe de la thymuline,
l'activité spécifique de cette hormone
dans le sérum peut être évaluée indirectement après absorption du
sérum par un anticorps anti-thymuline.
Chez des enfants normaux ou
présentant une MPE sévère ou modérée,
associée ou non à des

---

•
•
6.
infections
(article 9),
l'absorption du sérum par un anticorps
monoclonal anti-thymuline permet de mettre en évidence:
1- Une chute de l'activité spécifique de la thymuline dans
la MPE sévère et modérée.
2- La présence d'une quantité importante de "facteur
allogénique" dans le sérum des enfants infectés.
Cependant, la carence protéino-énergétique n'est pas le
seul mécanisme pouvant expliquer l'atteinte de la fonction
hormonale du thymus dans la MPE. En effet, on ne trouve aucune
différence de l'activité soécifiaue de l'hormone thymique chez des
.enfants sévèrement dénutris et infectés comparés à des enfants
modérément dénutris et infectés.
Chez des enfants modérément
dénutris, cette activité spécifique est plus importante chez les
non-infectés que chez les infectés
(article 9),
suggérant que la
présence d'infections est un élément important qui influe
fortement sur la fonction épithéliale du thymus.
En conclusion,
l'évaluation directe de l'activité
biologique de la thymuline dans le sérum peut être faussée par.la
présence d'infections concomitantes.
La MPE,
de même que les infections,
affectent la fonction
hormonale du thymus.
La diminution de l'activité de la thymuline
pourrait expliquer l'atteinte de certaines fonctions de l'immunité
à médiation cellulaire dans la MPE.

---