Pediatric Diabetes 2001: 2: 175177 Copyright C Munksgaard 2001

Printed in Denmark. All rights reserved

Pediatric Diabetes
ISSN 1399-543X

Original Article

Insulin in human milk and the prevention of

type 1 diabetes

Shehadeh N, Shamir R, Berant M, Etzioni A. Insulin in human milk and Naim Shehadeha, Raanan
the prevention of type 1 diabetes. Shamirb, Moshe Berantc and
Pediatric Diabetes 2001: 2: 175177. C Munksgaard, 2001 Amos Etzionic
a
Juvenile Diabetes Unit, bthe Unit of
Abstract: Although controversial, exclusive breast milk feeding was shown
Pediatric Gastroenterology and Nutrition,
to exert a protective effect in preventing type 1 diabetes. In contrast, an and the cDepartment of Pediatrics,
early introduction of cows milk-based formula in young infants may Rambam Medical Center and the B.
enhance the risk of disease, especially in genetically susceptible children, Rapapport Faculty of Medicine, Technion-
presumably by an increase of intestinal permeability to macromolecules Israel Institute of Technology, Haifa, Israel
such as bovine serum albumin and b-casein, which may arouse
autoimmunity. We have shown that human milk contains insulin in Key words: human milk insulin
substantial concentrations, while insulin is barely detectable (if at all) in intestinal maturation prevention type 1
infant formulas. Orally administered insulin was demonstrated to promote diabetes
gut maturation and to reduce intestinal permeability to macromolecules. Corresponding author: Naim Shehadeh
Furthermore, oral insulin may induce tolerance to insulin and protect MD, Pediatrics A, Rambam Medical
against the development of type 1 diabetes. We herewith raise a Center, Haifa 31096, Israel.
hypothesis that human milk is protective against the development of type Tel: 972-4-8542646;
1 diabetes by virtue of the effects of its substantial content of insulin. fax: 972-4-8542441; e-mail:
n_shehadeh/rambam.health.gov.il
Submitted 25 April 2001. Accepted for
publication 25 July 2001

One of the major effects of breast-feeding is to sup-
port gut maturation through the provision of growth
hormones, such as insulin. Insulin can interact with
intestinal mucosa when given both by the systemic
and by the oral route, since insulin receptors are
found on enterocyte apical and basolateral mem-
branes (1). The positive effects of orally administered
insulin on gut maturation and mucosal enzyme ex-
pression have been demonstrated in suckling mice (2),
suckling rats (3), newborn piglets (4), newborn pigs
(5) and newborn calves (6).
We have recently demonstrated that a significantly
higher insulin concentration is present in human milk
(60.2 41.1 mU/mL, mean SD, n 42) than in fresh
pooled cows milk (16.3 6.0 mU/mL, n 4), and
found that insulin is barely detectable in infant for-
mulas. Insulin concentrations were determined by ra-
dioimmunoassay with a commercial kit (Biomedica,
Sorin) using human insulin as standard, with a speci-
ficity of 100% for human and bovine insulin (7).
Observations on lactating dams and suckling rats
have shown that milk-borne insulin is biologically ac-
tive, and that immature enterocytes have an increased
responsiveness to insulin (8). It is therefore reason-
able to propose that breast milk insulin might well
have a beneficial influence on gut maturation. Al-
though insulin is a macromolecule that is usually not
absorbed in the gut (9), and its effect may be local
and limited to the suckling period (1), we have re-
cently demonstrated that oral insulin supplementa-
tion in non-suckling mice also has an effect on serum
lipid levels (10), which indicates that absorption has
taken place. This is in agreement with observations
by immunocytochemical methods in adult rats, on a
transcellular (but not paracellular) intestinal trans-
port of insulin (11).
Several investigators have suggested that there is
an association between type 1 diabetes and neonatal
feeding practices namely, exclusive breast-feeding as
being associated with a protective effect, and an early
introduction of cows milk-based formula into the
diet as being associated with an increased risk (12
15), although a short period of exclusive breast-feed-
ing (less than 2 months) appears to predispose genet-
175
Shehadeh et al.
ically susceptible young children to progressive signs
of beta-cell autoimmunity (16). Virtanen et al. (17)
have shown that early introduction of dairy products
is independently associated with an increased risk of
type 1 diabetes, although contradictory findings have
also been published (18). Most publications, however,
emphasize the role of cows milk components, such
as bovine serum albumin and b-casein, in the devel-
opment of type 1 diabetes (19, 20). Consequently, a
casein hydrolysate infant formula (Nutramigen),
which was found to protect diabetes-prone non-obese
diabetic (NOD) mice from developing the disease, is
already under study in a primary type 1 diabetes pre-
vention trial to test whether the avoidance of intact
bovine protein during the first 6 months of life re-
duces the development of type 1 diabetes in high-risk
groups (19).
The fact that human milk is rich in insulin, while
infant formulas lack this important hormone, has not
been hitherto considered in discussions on the patho-
genesis and prevention of type 1 diabetes. We here-
with raise a hypothesis that human milk is protective
against the development of type 1 diabetes by virtue
of the effects of its substantial content of insulin.
Oral insulin and intestinal permeability
Intestinal permeability and intestinal epithelial pas-
sage of antigenic macromolecules may play an im-
portant role in the development of autoimmune dia-
betes (21). In the biobreeding (BB) rat, autoimmune
diabetes develops spontaneously and is accompanied
by evidence of insulitis and other autoimmune
phenomena. Data from this model indicate that die-
tary antigens may play a role in disease initiation,
since feeding these animals a hydrolyzed casein diet
reduces the incidence and delays the onset of the dis-
ease (22), which could be attributed to the removal of
a disease-promoting protein from the diet. In effect,
increased gastric and intestinal permeability has been
demonstrated before the development of insulitis and
clinical diabetes in the BB rat (21); this increased per-
meability may allow the increased delivery of a lu-
minal compound or compounds to the mucosal im-
mune system, which appears to be an important fac-
tor in the genesis of the disease.
Orally administered insulin plays an important role
in maturation of the regulatory mechanisms gov-
erning intestinal absorption of macromolecules as
demonstrated, for example, by a suppression of IgG
absorption in adrenalectomized suckling rats re-
ceiving oral insulin supplementation (23). This effect
of oral insulin can reduce the exposure of the immune
system to triggering proteins, such as bovine albu-
min and casein that are present in infant formulas,
while the absence of insulin in these formulas may
176
thus favor the exposure of the intestinal mucosal im-
mune system to foreign dietary proteins.
Insulin as a primary antigen in the process of
autoimmune diabetes
Although the critical events that trigger the autoreac-
tive process in type 1 diabetes are not clear, destruction
of insulin-producing b-cells appears to be mediated by
the activation of autoreactive T cells that recognize sev-
eral islet b-cell antigens, including insulin, glutamic
acid decarboxylase (GAD) 65 and 67 isotypes, heat
shock protein 60, and some uncharacterized b-cell
antigens (24). These antigen specificities have been de-
fined in primary T-cell assays and by the generation of
T-cell lines and clones from type 1 diabetic patients and
high-risk first-degree relatives, and from lymph nodes,
spleens, and pancreata of NOD mice. The majority of
pathogenic CD4 T-cell clones derived from pancreata
of NOD mice with insulitis or frank diabetes not only
recognize insulin, but react specifically with the 9-23
peptide region of the B chain. In a very recent publi-
cation, Alleva et al. (25) have demonstrated for the first
time a cellular response to the B 9-23 insulin peptide in
human type 1 diabetes. These data suggest that insulin
may be the primary initial target of the immune system
in the autoimmune process that leads to type 1 diabetes
(26), and that therapies directed at this autoantigenic
response might be of benefit in preventing type 1 dia-
betes. In animal models, oral administration of insulin
can stimulate the intestinal immune system and arouse
active cellular mechanisms that suppress the develop-
ment of autoimmune diabetes (27). On the other hand,
bovine insulin, which is present in cows milk and dif-
fers from human insulin by only three amino acids,
may disrupt the induction of tolerance to human insu-
lin and lead to the development of type 1 diabetes (28).
Fig. 1. We hypothesize that insulin found in human milk plays a
major role in the primary prevention of type 1 diabetes by reducing
intestinal penetration of triggering dietary immunopathogenic
peptides, and by inducing active cellular mechanisms that suppress
the development of autoimmune diabetes.
Pediatric Diabetes 2001: 2: 175177
 Insulin in human milk and type 1 diabetes

12. V SM, R L, A A et al. Infant feeding in

Hypothesis Finnish children 7 yr of age with newly diagnosed
IDDM. Diabetes Care 1991: 14: 415417.
Based on the data presented above, we hypothesize 13. N JM, S FW. A meta-analysis of infant diet and
that the insulin content of human milk has an im- insulin dependent diabetes mellitus: do biases play a role?
portant role in the primary prevention of type 1 dia- Epidemiology 1996: 7: 8792.
betes, by inducing active cellular mechanisms that 14. K J. What can epidemiology tell us about the role
suppress the development of autoimmune diabetes, of infant diet in the etiology of IDDM? Diabetes Care
1995: 17: 8791.
and by reducing intestinal transmission of triggering 15. V C, H N, I L et al. Environmental fac-
dietary peptides and thus minimizing the intestinal tors in childhood IDDM. Diabetes Care 1994: 17: 1381
mucosal immune response against such immunopa- 1388.
thogenic peptides (Fig. 1). 16. K T, E M, K A et al. Short-term
exclusive breastfeeding predisposes young children with
increased genetic risk of type 1 diabetes to progressive
beta-cell autoimmunity. Diabetologia 2001: 44: 6369.
References 17. V SM, R L, Y K et al. Early intro-
duction of dairy products associated with increased risk
1. B JP, D K N, M S et al. Expression of
of IDDM in Finnish children. Diabetes 1993: 42: 1786
insulin receptors and of 60-kDa receptor substrate in rat
1790.
mature and immature enterocytes. Am J Physiol Gastro-
intest Liver Physiol 1997: 273: G217226. 18. A HK, K M. Putative environmental factors
in type 1 diabetes. Diabetes Metab Rev 1998: 14: 3167.
2. A P, M D. Insulin influences the matu-
19. K W, H-MK D, C RK et al.
ration and proliferation of suckling mouse intestinal mu-
Immunological aspects of nutritional diabetes prevention
cosa in serum free organ culture. Biol Neonate 1984: 46:
in NOD mice. Diabetes 1997: 46: 557564.
229236.
20. K J, M JM, K M et al. A bovine
3. M S, D K N, S A et al. Insulin signal
albumin peptide as a possible trigger of insulin-dependent
transduction in rat small intestine: role of MAP kinases
diabetes mellitus. N Engl J Med 1992: 327: 302307.
in expression of mucosal hydrolases. Am J Physiol Gastro-
intest Liver Physiol 2001: 280: G229240. 21. M JB, J J, U SJ, H J, G
DG. Increased gastrointestinal permeability is an early
4. S RJ. Oral insulin increases small intestinal mass lesion in the spontaneously diabetic BB rat. Am J Physiol
and disaccharidase activity in the newborn miniature pig. Gastrointest Liver Physiol 1999: 39: G951957.
Pediatr Res 1990: 28: 171175.
22. S FW, E RB, K H. Diet and autoimmunity:
5. A JM, G RH, P PH. Absorption of prospects of prevention of type 1 diabetes. Diabetes Nutr
colostral gamma-globulins and insulin by the newborn Metab 1989: 2: 6173.
pig. J Anim Sci 1962: 21: 412413.
23. H E, S B. Precocious cessation of intestinal
6. P AE, R PC, S B. Absorption of orally macromolecular transmission and sucrase development in-
administered insulin by the newly born calf. J Physiol duced by insulin in adrenalectomized suckling rat. Comp
1964: 171: 203215. Biochem Physiol A 1991: 99: 327331.
7. S N, G L, B S, P R, E- 24. A MA, M NK. The pathogenesis of insu-
A. The importance of insulin content in infant diet: lin-dependent diabetes mellitus. N Engl J Med 1994: 331:
suggestion for a new infant formula period. Acta Paediatr 14281436.
2001: 90: 9395. 25. A DG, C PD, J L, K WW et al. A dis-
8. B JP, D K N, S EM, M S. Oral insu- ease-associated cellular immune response in type 1 dia-
lin is biologically active on rat immature enterocytes. J betes to an immunodominant epitope of insulin. J Clin
Pediatr Gastroenterol Nutr 1997: 25: 230232. Invest 2001: 107: 173180.
9. L M. Out with the Jab, in with painless pills. Lancet 26. W DR, E GS. Its insulin. J Autoimmun
1997: 349: 1676. 2000: 15: 286291.
10. S T, H R, C M et al. Atherosclerosis in 27. Z ZJ, D L, E G, W HL. Sup-
apolipoprotein E-deficient mice is attenuated by enterally pression of diabetes in nonobese diabetic mice by oral ad-
given insulin. Gastroenterology 2001: 120: A677. ministration of porcine insulin. Proc Natl Acad Sci USA
11. B M, Z E, G D et al. Biochemical and 1991: 88: 1025210256.
morpho-cytochemical evidence for the intestinal absorp- 28. V O, P J, O T, A HK.
tion of insulin in control and diabetic rats. Comparison Cow milk feeding induces antibodies to insulin in
between the effectiveness of duodenal and colon mucosa. children a link between cow milk and insulin dependent
Diabetologia 1994: 37: 119126. diabetes mellitus? Scand J Immunol 1998: 47: 131135.

Pediatric Diabetes 2001: 2: 175177 177