THE JOURNAL OF BIOLOGICAL CHEMISTRY
Minireview
Metabolic Programming: Causes
and Consequences*
Published, JBC Papers in Press, November 6, 2001,
DOI 10.1074/jbc.R100017200
Mulchand S. Patel‡ and Malathi Srinivasan
From the Department of Biochemistry, School of
Medicine and Biomedical Sciences, State University of
New York, Buffalo, New York 14214
The normal programmed development of a multicellular or-
ganism from the germ cell is a synchronized series of events
driven by genetic instructions acquired during conception. Dur-
ing the early critical periods in life the organism also has the
ability to respond to environmental situations that are alien to
normal development by adaptations at the cellular, molecular,
and biochemical levels. Such early adaptations to a nutritional
stress/stimulus permanently change the physiology and metab-
olism of the organism and continue to be expressed even in the
absence of the stimulus/stress that initiated them, a process
termed “metabolic programming” (1). A brief summary of the
findings from human epidemiological and animal studies is
presented below in support of the concept of metabolic pro-
gramming induced by nutritional experiences during critical
periods in development with consequences later in adulthood.
For detailed accounts the reader is referred to excellent reviews
on this subject (2–5). This minireview will focus on metabolic
programming with reference to a novel rat model developed in
our laboratory.
Evidence for Metabolic Programming from Human
Epidemiological Data
Extensive epidemiological findings indicate that metabolic
programming occurs in humans. Barker (6) was the first to
suggest from epidemiological studies that the disproportionate
size of the newborn resulting from maternal malnutrition cor-
related with an increased risk for adverse health outcomes
(type II diabetes, hypertension, and cardiovascular diseases)
later in adult life. These primary observations resulted in the
now widely recognized “fetal origins” hypothesis emphasizing
the importance of adequate maternal nutrition during preg-
nancy (4).
Evidence for Metabolic Programming in Animals
Nutritional programming has been demonstrated in animal
studies. In pioneering studies with rodents, McCance (7) dem-
onstrated by adjusting litter size that the quantity of food
consumed during early periods of postnatal life has long term
consequences on growth. The consequences of maternal mal-
nutrition induced by either a low protein diet or caloric restric-
tion during gestation and lactation cause major changes in the
structure and function of several organs in the offspring. Preg-
* This minireview will be reprinted in the 2002 Minireview Compen-
dium, which will be available in December, 2002. The work performed
in the authors’ laboratory and summarized in this article was supported
in part by NICHD Grant HD-11089 and NIDDK Grants DK-51601 and
DK-61518 from the National Institutes of Health.
‡ To whom correspondence should be addressed: Dept. of Biochemis-
try, 140 Farber Hall, SUNY, 3435 Main St., Buffalo, NY 14214. Tel.:
716-829-3074; Fax: 716-829-2725; E-mail: mspatel@buffalo.edu.
This paper is available on line at http://www.jbc.org
Vol. 277, No. 3, Issue of January 18, pp. 1629 –1632, 2002
© 2002 by The American Society for Biochemistry and Molecular Biology, Inc.
Printed in U.S.A.
nant rats fed a low protein diet produced pups with alterations
in pancreatic islets (8, 9). These include reduced islet vascular-
ization, ␤ cell proliferative capacity, and islet size with right-
ward shift (decreased sensitivity) in glucose-stimulated insulin
secretion and altered sensitivity to insulin in muscle (8, 9).
Furthermore, hypothalamic nuclei are malformed in these
weanling rats; this is accompanied by reduced vascularization
of the cerebral cortex in the progeny (10). Metabolic capacities
of the liver, muscle, and adipose tissue are compromised by
maternal protein restriction during gestation and lactation
with adverse adult onset outcomes (11). Elevated insulin con-
centrations during critical periods of development, as occurs
perinatally in the offspring of gestationally diabetic mothers,
lead to permanent malorganization of the ventromedial hypo-
thalamic nuclei followed by glucose intolerance in adult life
(12). Collectively, the results from animal models indicate that
even brief periods of dietary manipulation in early life have
implications for ill health outcomes, which become evident only
in adulthood.
“Pup in a Cup” Rat Model
Most of the above cited animal studies deal with maternal
malnutrition during pregnancy leading to intrauterine growth
retardation and accompanying adaptations in the fetus for
survival under adverse conditions. Except for the effect of over-
nourishment caused by reducing litter size, the effect of an
altered nutrition during the suckling period has not been in-
vestigated extensively in the rat. This is because of experimen-
tal difficulties in successfully rearing rat pups on a modified
milk formula away from nursing mothers. The artificial rearing
technique as described by Hall (13) provides a means to study
the effects of altered nutrition during the suckling period in the
rat. We have adapted this technique to evaluate the conse-
quences of a switch in the “quality” of nutrition (from fat-rich
rat milk to high carbohydrate (HC)1 milk formula) during the
suckling period in the rat. The result is the “pup in a cup” rat
model (also referred to as the HC rat) where the newborn rats
are reared in Styrofoam cups floating in a temperature-con-
trolled water bath. Four-day-old rats are fed via intragastric
cannulas, introduced nonsurgically, raised on a HC milk for-
mula (56% of the calories derived from carbohydrate compared
with 8% in rat milk) up to day 24, and then weaned onto
laboratory chow. Four-day-old rats artificially reared on a high
fat milk formula (macronutrient composition similar to rat
milk to show that the artificial rearing technique per se has no
metabolic programming effects) and pups nursed by their own
mothers served as controls. The detailed protocol followed by
our laboratory for the experiments performed using the “pup in
a cup” rat model is described elsewhere (14, 15). Fig. 1A depicts
metabolic adaptations in the first generation HC rats.
Immediate Metabolic Adaptations in the
Suckling Period
An immediate outcome of the HC dietary intervention is the
onset of hyperinsulinemia within 24 h; this persists throughout
the suckling period and into adulthood even after the with-
1
The abbreviations used are: HC, high carbohydrate; HC rat, rat
raised on a high carbohydrate milk formula; PDX-1, pancreatic duode-
nal homeobox factor-1; SAPK-2, stress-activated protein kinase-2; PI3-
kinase, phosphatidylinositol 3-kinase; GLP-1, glucagon-like peptide-1.
1629
1630 Minireview: Metabolic Programming: Causes and Consequences

FIG. 1. Diet-induced metabolic programming in first and sec-

ond generation HC rats. Metabolic adaptations in first generation (A)
and second generation (B) HC rats in response to feeding a HC milk FIG. 2. Biochemical adaptations in pancreatic islets from neo-
formula to first generation neonatal rat pups are shown. GTT, glucose natal HC rats. A represents plasma insulin levels (pM) in neonatal HC
tolerance test. rats. B is the relative levels of preproinsulin and PDX-1 mRNAs in islet
from 12-day-old HC and mother-fed (MF) rats. C represents the insulin
secretory response to a glucose (Glu) stimulus by islets from 12-day-old
drawal of the HC milk formula at the time of weaning (15, 16) first generation HC rats at 60 min. Insulin secretion is expressed as
(Figs. 1A and 2A). During the suckling period there are no femtomoles of insulin/30 islets/60 min. D represents the activity of the
differences in body weights and in plasma glucose levels be- low Km hexokinase activity in supernatant (S) and pellet (P) fractions of
islet extracts of 12-day-old HC rats and mother-fed control rats. Activity
tween the HC and age-matched control groups (17). The over- is expressed as milliunits/mg of protein.
lap of the critical window for postnatal pancreatic development
with the high carbohydrate nutritional intervention in the HC
rat suggests that the endocrine pancreas is a target organ for
significant adaptations. We have observed significant alter-
ations at the cellular, molecular, and biochemical levels in
islets isolated from neonatal HC rats and have also observed
the programming of these adaptations into adulthood.
Cellular adaptations induced by dietary intervention during
the suckling period include an increased number of smaller
sized islets in the HC pancreas compared with controls (18).
Such HC islets have a larger immunopositive area for insulin
resulting in a net increase in the insulin-producing mass in HC
islets (18). Additionally, the rate of apoptosis and the expres-
sion of proliferating cell nuclear antigen are inversely altered
in islets and the ductal epithelium of the HC rat (18). One of the
factors contributing to the altered ontogeny has been attrib-
uted to the change in the expression of insulin-like growth
factor II (18).
Several molecular adaptations are observed in neonatal is-
lets in response to the HC dietary intervention. Increases in
insulin biosynthesis and in gene expression of preproinsulin
are observed in islets of neonatal HC rats (19) (Fig. 2B). Addi-
tionally, mRNA levels of transcription factors such as pancre- FIG. 3. Putative pathway for the regulation of preproinsulin
atic duodenal homeobox factor-1 (PDX-1, also known as soma- gene transcription in neonatal HC islets. This scheme indicates the
tostatin transcription factor-1) (Fig. 2B), islet factor-1 (Isl-1), various responses induced in the HC rat to the high carbohydrate
dietary intervention, the possible interactions among them, and the
upstream stimulatory factor-1, regenerating factor-3 (reg-3) consequent regulation of preproinsulin gene transcription. USF-1, up-
(19, 20), and Beta2/NeuroD and hepatocyte nuclear factor ␤-32 stream stimulatory factor-1.
are significantly increased in neonatal HC islets. mRNA levels
of stress-activated protein kinase-2 (SAPK-2), phosphatidyl- sulin promoter activity (21). DNA binding activity of PDX-1 has
inositol 3-kinase (PI3-kinase), acetyl-CoA carboxylase, glucose been proposed to be modulated by glucose via a phosphoryla-
transporter 2, and insulin receptor substrate-1 and -2 are also tion cascade(s) involving SAPK-2 and PI3-kinase, which facil-
significantly higher in these HC islets (19, 20). PDX-1 is an itates the translocation of the 46-kDa, phosphorylated active
important transactivator of the insulin gene and is an essential form to the nucleus resulting in increased insulin gene tran-
component of the mechanisms whereby glucose modulates in- scription (21) (Fig. 3). The possible correlation between the
observed effects of the HC dietary intervention in neonatal HC
2
M. S. Patel, M. Srinivasan, F. Song, and R. Aalinkeel, unpublished rats and preproinsulin gene transcription is shown in Fig. 3. In
observations. addition PDX-1 is an important transcription factor that com-
 Minireview: Metabolic Programming: Causes and Consequences
FIG. 4. The three phases of metabolic programming in the HC rat. Possible mechanisms for the adaptations to the HC dietary intervention
in the suckling period (Adaptive Phase), the persistence into adulthood of these adaptations (Persistence Phase), and for their transmission to the
progeny (Transmission Phase) are shown.
mits the progression of the pluripotent ductal cell into an
endocrine cell (22). The significant increase in the gene expres-
sion, DNA binding activity, and protein content of PDX-1 in
neonatal HC islets suggests that it plays a pivotal role in the
cellular adaptations as well as in the onset and maintenance of
hyperinsulinemia in this rat model (19). Transcription factors
like Isl-1, Beta2/NeuroD, reg-3, and hepatocyte nuclear factor
␤-3 also contribute to pancreatic organogenesis (22–24). The
increases in the gene expression of these factors suggest that
they too contribute to the modification of the islet architecture
in the neonatal HC rats. cDNA array analysis has indicated
significant changes in global gene expression patterns in neo-
natal HC islets as well as in adult HC islets suggesting that a
wide range of molecular alterations is essential for the onset
and persistence of the HC phenotype in HC rats (20).
Significant biochemical adaptations summarized below ini-
tiate and sustain hyperinsulinemia in neonatal HC rats. A
characteristic feature of neonatal HC rats is about a 6-fold
increase in the concentration of circulating insulin (16, 25) (Fig.
2A). There is a distinct leftward shift (increased sensitivity) in
the glucose-stimulated insulin secretory response in HC islets
(Fig. 2C), and this is associated with increases in the low Km
hexokinase activity (Fig. 2D) and an increase in glucose trans-
porter 2 protein content (25). Insulin secretion by islets is
regulated by three pathways (the KATP channel-dependent
pathway, the KATP channel-independent augmentation path-
way, and the Ca2⫹ channel-independent augmentation path-
way (26)), and these pathways are up-regulated in HC islets
(27). Circulating levels of glucagon-like peptide-1 (GLP-1) are
significantly higher in HC rats suggesting that GLP-1-medi-
ated events contribute significantly to the hyperinsulinemia of
HC rats (27). In addition to its insulinotropic effects, GLP-1
stimulates transcription of the preproinsulin gene and PDX-1
gene and also the proliferation and neogenesis of ␤ cells from
ductal epithelium in rodents (28 –30). In light of the above, the
significant increase in circulating levels of GLP-1 in 12-day-old
HC rats has important implications for the HC phenotype in
suckling HC rats.
Because of chronic hyperinsulinemia the capacity for hepatic
lipogenesis is enhanced in neonatal HC rats (16). Enzymes
such as glucokinase and malic enzyme that normally appear in
the liver at weaning are precociously induced by this dietary
treatment suggesting that their appearance is not development-
dependent but is controlled by the presence of the stimulus
(diet-induced hyperinsulinemia) (16). It is evident from the
above that feeding rats an HC milk formula, instead of moth-
er’s milk, during the suckling period elicits significant alter-
ations in islet functions as well as in metabolic responses of
1631
peripheral tissues, which are important for the development of
adult-onset obesity (Fig. 4, Adaptive Phase).
Programming of Early Onset Adaptations into
Adulthood
Hyperinsulinemia persists into adulthood of HC rats despite
withdrawal of the HC nutritional intervention on day 24 (Fig.
1A). Alterations in the insulin secretory pathway observed in
the neonatal HC islets are programmed into adult islets (31,
32). The molecular changes (increase in the preproinsulin gene
transcription) observed in islets from neonatal HC rats are also
observed in adult HC islets (32). There is an increase in the
insulin-producing mass of the adult HC pancreas (17). Chronic
hyperinsulinemia is accompanied by an increase in the body
weight of the animals from day 55 onward and full blown
obesity by day 100 (15) (Fig. 1A). Although the adult HC rats
maintain normoglycemia, an abnormal response to an oral
glucose tolerance test is observed in HC animals on approxi-
mately day 75 (17) (Fig. 1A). In addition, liver and adipose
tissue show increased lipogenic capacities in adulthood (15). An
increase in the cell size in epididymal adipose tissue reflects
the adiposity observed in adulthood (15). Glycogen content and
the glycogen synthase activation cascade are down-regulated in
liver and skeletal muscle (33) and up-regulated in epididymal
adipose tissue of adult HC rats (34). Taken together these
findings indicate that early adaptations are programmed and
are accompanied by additional changes probably triggered by
adult-onset factors (Fig. 4, Persistence Phase).
Nutritionally Induced Generational Effect on
Metabolic Programming
An important and unexpected consequence of the nutrition-
ally induced metabolic programming noted above is the trans-
mission of the phenotype from the HC mother to the progeny
(see Fig. 1B). Female rats fed an HC milk formula during their
suckling period spontaneously transmitted their metabolic
characteristics to their progeny without the pups themselves
having to undergo any nutritional treatment (35). The second
generation HC rats display hyperinsulinemia and an altered
insulin secretory pattern within 48 h after weaning them onto
laboratory chow on day 24.2 Molecular adaptations (increases
in mRNA levels of preproinsulin, PDX-1, upstream stimulatory
factor-1, SAPK-2, and PI3-kinase) are also programmed in
second generation HC rats.2 The growth pattern of HC rats in
the second generation parallels that of first generation HC rats
(35) (Fig. 1B). Cross-breeding experiments have demonstrated
that only HC females transmit these traits to the progeny,
suggesting that the intrauterine experience may be essential
for the transmission (Fig. 4, Transmission Phase).
1632 Minireview: Metabolic Programming: Causes and Consequences
Potential Mechanisms of Metabolic Programming
Metabolic programming is an “adaptive process” that occurs
in response to a nutritional stimulus/insult during a vulnerable
period of susceptibility early in life. Direct evidence for mech-
anisms responsible for metabolic programming is not yet avail-
able. Potential mechanisms have been discussed by Lucas (1)
and Waterland and Garza (5). The immediate effects of a nu-
tritional stress on structural development with long term con-
sequences have been demonstrated in rats (8, 10). Experimen-
tally induced hyperinsulinemia in neonatal rats during the
critical period of brain development results in alterations in
body weight, blood pressure, and glucose metabolism in adult
life because of disorganization of the ventromedial hypotha-
lamic nuclei (12). The HC milk formula induces responses in
pancreatic islets and the small intestine as evidenced by in-
creases in the circulating plasma levels of insulin and GLP-1 in
neonatal HC rats. In the context of reports on the effects of a
nutritional modification on the brain and the effects of the HC
dietary intervention on the islets and gut in neonatal HC rats,
it is tempting to suggest that cross-talk may occur among the
pancreas, small intestine, and brain resulting in the onset and
persistence of hyperinsulinemia in HC neonates.
Epigenetic mechanisms that mediate phenomena such as
genomic imprinting may also contribute to programming (36).
Epigenetic modification is triggered by changes in environment
and can occur in both somatic and germ cell lineage during
development (37). Altered DNA methylation patterns have
been shown to be caused by protein deficiency and folate de-
pletion (38). Nutritional alterations early in life may modify
cell-specific DNA methylation patterns leading to altered levels
of gene expression in specific tissues. In addition, because the
altered DNA methylation patterns in specific cells are trans-
mitted to the daughter cells by replication, the initial modifi-
cations are immortalized (Fig. 4).
Concluding Remarks
The “pup in a cup” rat model is unique in that it permanently
programs the metabolism of an adult rat by merely modifying
the composition of the milk fed to the animal during its suck-
ling period. In addition, this type of metabolic programming is
transmitted to the next generation by the mother. It is evident
from our studies with rats that the nature and the timing of the
dietary treatment programs the onset of pathological condi-
tions in the adult that mimic major metabolic diseases noted in
humans, such as obesity and type II diabetes. Evidence points
to diet-induced hyperinsulinemia as the primary event in met-
abolic programming in the HC rat.
In a recent report Mokdad et al. (39) have pointed out that
the epidemic of obesity is a critical public health threat in the
United States (nearly one of five Americans is obese). Because
there has not been a significant change in the gene pool in the
United States in recent decades (39), it is unlikely that genes
related to obesity are involved in the increased incidence in this
disease. There has also been a 33% increase in the incidence of
diagnosed diabetes in the past decade, and this has been highly
correlated with the increase in the incidence of obesity, sug-
gesting that obesity is a major risk factor for chronic diseases
(39). It is becoming clear that the origins of chronic diseases are
not limited to only inherited genes and/or sedentary life styles.
The results from the HC rat model suggest that nutritional
experiences of infants during the immediate postnatal life such
as overfeeding of formula and early introduction of supplemen-
tal weaning foods high in carbohydrates (e.g. cereals, fruits,
juices, etc.) may contribute to metabolic programming leading
to adult-onset diseases like obesity and diabetes. The newly
emerging field of metabolic programming therefore offers an
additional route to examine the etiology of adult-onset chronic
diseases.
Acknowledgments—We sincerely thank Dr. Richard W. Hanson of
Case Western Reserve University for fruitful discussions and for critical
reading of the manuscript and Dr. Gail Willsky of this department for
comments on the manuscript.
REFERENCES
1. Lucas, A. (1991) CIBA Found. Symp. 156, 38 –50
2. Whitaker, R. C., and Dietz, W. H. (1998) J. Pediatr. 132, 768 –776
3. Hoet, J. J., and Hanson, M. A. (1999) J. Physiol. (Lond.) 514, 617– 627
4. Godfrey, K. M., and Barker, D. J. P. (2000) Am. J. Clin. Nutr. 71, (suppl.)
1344S–1352S
5. Waterland, R. A., and Garza, C. (1999) Am. J. Clin. Nutr. 69, 179 –197
6. Barker, D. J. P. (1995) Br. Med. J. 311, 171–174
7. McCance, R. A. (1962) Lancet ii, 271–272
8. Berney, D. M., Desai, M., Palmer, D. J., Greenwald, S., Brown, A., Hales, C. N.,
and Berry, C. L. (1997) J. Pathol. 183, 109 –115
9. Latorraca, M. Q., Reis, M. A. B., Carneiro, E. M., Mello, M. A. R., Velloso, L. A.,
Saad, M. J. A., and Boschero, A. C. (1998) J. Nutr. 128, 1643–1649
10. Bennis-Taleb, N., Remacle, C., Hoet, J. J., and Reusens, B. (1999) J. Nutr. 129,
1613–1619
11. Ozanne, S. E., and Hales, C. N. (1999) Proc. Nutr. Soc. 58, 615– 619
12. Harder, T., Plagemann, A., Rohde, W., and Dorner, G. (1998) Metabolism 47,
855– 862
13. Hall, W. G. (1975) Science 190, 1313–1315
14. Hiremagalur, B. K., Johanning, G. L., Kalhan, S. C., and Patel, M. S. (1992) J.
Nutr. Biochem. 3, 474 – 480
15. Hiremagalur, B. K., Vadlamudi, S., Johanning, G. L., and Patel, M. S. (1993)
Int. J. Obes. 17, 495–502
16. Haney, P. M., Raefsky-Estrin, C., Caliendo, A., and Patel, M. S. (1986) Arch.
Biochem. Biophys. 244, 787–794
17. Vadlamudi, S., Hiremagalur, B. K., Tao, S., Kalhan, S. C., Kalaria, R. N.,
Kaung, H. C., and Patel, M. S. (1993) Am. J. Physiol. 265, E565–E571
18. Petrik, J., Srinivasan, M., Aalinkeel. A., Coukell, S., Arany, E., Patel, M. S.,
and Hill, D. J. (2001) Pediatr. Res. 49, 84 –92
19. Srinivasan, M., Song, F., Aalinkeel, R., and Patel, M. S. (2001) J. Nutr.
Biochem. 12, 575–584
20. Song, F., Srinivasan, M., Aalinkeel, R., and Patel, M. S. (2001) Diabetes 50,
2053–2060
21. Macfarlane, W. M., Smith, S. B., James, R. F. L., Clifton, A. D., Doza, Y. N.,
Cohen, P., and Docherty, K. (1997) J. Biol. Chem. 272, 20936 –20944
22. Huang, H-P., and Tsai, M-L. (2000) J. Biomed. Sci. 7, 27–34
23. Okamoto, H. (1999) J. Hepato-Biliary-Pancreatic Surgery 6, 254 –262
24. Wu, K-L., Gannon, M., Peshavaria, M., Offield, M. F., Henderson, E., Ray, M.,
Marks, A., Gamer, L. W., Wright, C. V. E., and Stein, R. (1997) Mol. Cell.
Biol. 17, 6002– 6013
25. Aalinkeel, R., Srinivasan, M., Kalhan, S. C., Laychock, S. G., and Patel, M. S.
(1999) Am. J. Physiol. 277, E1061–E1069
26. Komatsu, M., Schermerhorn, T., Noda, M., Straub, S. G., Aizawa, T., and
Sharp, G. W. G. (1997) Diabetes 46, 1928 –1938
27. Srinivasan, M., Aalinkeel, R., Song, F., Lee, B., Laychock, S. G., and Patel,
M. S. (2000) Am. J. Physiol. 279, E1347–E1357
28. Fehmann, H. C., and Habener, J. F. (1992) Endocrinology 130, 159 –166
29. Stoffers, D. A., Kieffer, T. J., Hussain, M. A., Drucker, D. J., Bonner-Weir, S.,
Habener, J. F., and Egan, J. M. (2000) Diabetes 49, 741–748
30. Xu, G., Stoffers, D. A., Habener, J. F., and Bonner-Weir, S. (1999) Diabetes 48,
2270 –2276
31. Laychock, S. G., Vadlamudi, S., and Patel, M. S. (1995) Am. J. Physiol. 269,
E739 –E744
32. Aalinkeel, R., Srinivasan, M., Song, F., and Patel, M. S. (2001) Am. J. Physiol.
281, E640 –E648
33. Srinivasan, M., Vadlamudi, S., and Patel, M. S. (1996) Int. J. Obes. 20,
981–989
34. Srinivasan, M., and Patel, M. S. (1998) J. Nutr. Biochem. 9, 81– 87
35. Vadlamudi, S., Kalhan, S. C., and Patel, M. S. (1995) Am. J. Physiol. 269,
E731–E738
36. Thorvaldsen, J. L., and Bartolomei, M. S. (2000) Science 288, 2145–2146
37. Monk, M. (1995) Dev. Genet. 17, 188 –197
38. Jacob, R. A. (2000) Am. J. Clin. Nutr. 72, 903–904
39. Mokdad, A. H., Serdula, M. K., Dietz, W. H., Bowman, B. A., Marks, J. S., and
Koplan, J. P. (2000) J. Am. Med. Assoc. 284, 1650 –1651