![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Department of Radiation Oncology, Duke University Medical Center, Durham, NC 27710 and * Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853
2To whom correspondence should be addressed. E-mail: waterland{at}radonc.duke.edu.
Human epidemiologic and experimental animal studies suggest strongly that prenatal and early postnatal nutrition influence adult susceptibility to diet-related chronic disease. To elucidate biologic mechanisms linking divergent early nutritional sufficiency to adult insulin axis function in an animal model of "metabolic imprinting," this research focused on the following two objectives: 1) identify a tissue responsible for effect persistence, and 2) identify genes showing sustained differential expression in that tissue. Newborn rats were assigned randomly to small (SL), control (C) or large litters (LL) until weaning. Glucose and insulin tolerance tests were conducted directly after weaning (age 26 d) and in adulthood (ages 110 and 255 d). Glucose-stimulated insulin secretion from isolated pancreatic islets was assessed at those ages. DNA microarrays were used to identify genes showing persistent between-group differential expression in isolated islets. Glucose and insulin tolerance tests suggested persistently reduced pancreatic glucose-responsiveness in SL and LL rats. Insulin tolerance tests showed no group differences in whole-body insulin-stimulated glucose uptake. These data support the hypothesis that the endocrine pancreas contributes to primary imprinting in this model. Persistent defects in glucose-stimulated insulin secretion from isolated islets also supported this hypothesis but only in SL rats. Of 13 named islet genes showing SL vs. C differential expression at age 26 d, 10 remained differentially expressed at age 110 d. These data indicate that the endocrine pancreas plays a primary role in the putative metabolic imprinting mechanism in SL rats.
KEY WORDS: • metabolic imprinting • programming • endocrine pancreas • postnatal nutrition • prenatal exposure delayed effect • rats
Human epidemiologic data from different populations (1
–3
) indicate that prenatal and early postnatal nutritional status may influence adult susceptibility to impaired glucose tolerance, cardiovascular disease and obesity. In many developing countries, changes in physical activity patterns and dietary overconsumption are supplanting infectious diseases as primary causes of mortality and morbidity. As generations exposed to suboptimal nutrition during early development increasingly experience dietary excess in adulthood, and overnutrition in early life becomes more common in all countries, understanding the biologic bases of persistent nutritionally induced susceptibility to diet-related chronic disease becomes more important.
Animal models also support the hypothesis that prenatal and early postnatal nutrition permanently affect metabolism (3
,4
). However, the specific mechanisms responsible for these phenomena remain unknown. Understanding the biologic mechanisms underlying such phenomena in animal models is a necessary step toward gauging their significance to human health and designing potential interventions to avoid or ameliorate their adverse consequences. Toward the goal of understanding these mechanisms, the term "metabolic imprinting" was proposed to focus attention on permanent responses to specific transient nutritional conditions early in life that are characterized by a limited period of susceptibility, a specific effect and a dose-response relation between exposure and outcome (3
).
Guided by this definition, the current study focused on the persistent effects of suckling-period litter size in rats. In this model, the offspring of several litters born on the same day are redistributed randomly to foster litters of divergent sizes. Pups assigned to small litters are overnourished during the suckling period, and those assigned to large litters are undernourished relative to control rats suckled in average-sized litters. Although all rats have free access to the same diet after weaning, suckling period litter size influences body composition (5
), cholesterol metabolism (6
) and insulin axis function (5
–8
) in adulthood.
Results of previous studies that focus on the effects of early nutrition on adult insulin axis function are consistent with a primary imprint in the endocrine pancreas and/or skeletal muscle. Primary imprint tissue may be defined as any tissue that accounts wholly or in part for the "metabolic memory" necessary for the effect persistence seen in models of metabolic imprinting. Due to the integrated nature of mammalian physiology, primary effects in one organ or tissue may lead to compensatory adaptations in other organs or tissues. Clearly, identifying the tissue or tissues housing a primary imprint is a logical first step in a study of the biological mechanisms operating in a model of metabolic imprinting.
To aid in distinguishing primary imprints from compensatory adaptations, two characteristics of primary imprint tissue are proposed herein. First, the primary imprint should show similar effects directly after the imprinting period and in adulthood. A compensatory adaptation, conversely, may arise after the imprinting period and may show variable effects as physiologic status changes over the life course. Second, effects demonstrated in vivo should be maintained when the primary imprint tissue is isolated and studied in vitro. Effects resulting from compensatory adaptation, on the other hand, may not be sustained once the tissue is removed from its organismic milieu.
The current study applied these two characteristics to identify a primary imprint tissue influencing adult insulin axis function in the rat postnatal litter size model. In vivo studies conducted directly after the imprinting period and in adulthood suggested that the endocrine pancreas (and not skeletal muscle) houses a primary imprint in this model. Functional and gene expression studies of isolated pancreatic islets further supported the hypothesis that the endocrine pancreas is a repository of "metabolic memory" in this model of metabolic imprinting.
Animals
All animal protocols were approved by the Cornell University Institutional Animal Care and Use Committee. Timed-pregnant Wistar rats (Charles River Laboratories, Wilmington, MA) were received on d 14 of gestation and were provided free access to stock diet3 (Prolab 1000, PMI Nutrition, Brentwood, MO) throughout gestation and lactation. At age 1 d, offspring of 9–14 litters born on the same day were redistributed randomly to small (SL,4 4 pups), control (C, 12 pups) or large litters (LL, 18 pups). Offspring were weaned at age 21 d, and had free access to stock diet and water thereafter. To eliminate potential sex-related outcome variance, only male offspring were studied.
Methods
Glucose and insulin tolerance tests.
Tests were conducted on rats aged 24–29 d (mean age 26 d), 102–118 d (mean age 110 d) and 252–259 d (mean age 255 d). Equal numbers of rats from each treatment group were tested on each study day, and within-day group ordering was rotated to balance possible time-of-day effects. Rats were not provided food overnight preceding the tests. All tests were conducted between 0900 and 1200 h. Each rat was anesthetized with sodium pentobarbitol (50 mg/kg, intraperitoneal). A baseline blood sample (
100 µL) was collected by amputation of the tail tip (age 26 d) or by tail vein venipuncture (age 110 and 255 d). Each rat then received an injection of glucose (0.5 g/kg at age 26 d, 1.0 g/kg at age 110 and 255 d, in 9 g/L saline) or bovine insulin (ICN Biomedicals, Costa Mesa, CA, 0.4 U/kg, in 9 g/L saline, 2.5 g/L bovine serum albumin) into a tail vein. Serial tail blood samples were collected after the injection. Blood samples were allowed to coagulate at room temperature for 45 min, then centrifuged (5000 x g, 10 min, 4°C). Sera were isolated and stored at -80°C until analysis.
Isolation of pancreatic islets.
Islets were isolated by collagenase digestion (9
). Rats were killed by CO2 asphyxiation in the morning in the fed state. Their pancreases were excised quickly, trimmed of adherent fat, minced in Krebs-Ringer bicarbonate buffered medium (KRB), pH 7.4 (10
), rinsed twice in KRB and digested at 37°C with hand shaking. The digest was rinsed 4 times in 40 mL of KRB. Islets of average size for each digest were then picked by micropipette under a dissecting microscope. Mean size of studied islets was estimated by measuring the diameter of a sample of 10 of the 100–150 islets selected per rat using a microscope eyepiece reticle. After isolation, islets were preincubated for 30 min at 37°C in KRB containing 3.0 mmol/L glucose before being studied.
Islet incubation. Incubation media consisted of KRB adjusted to 5.0, 10.0 and 15.0 mmol/L glucose. Incubation volume was 2.0 mL, and incubations were conducted in duplicate at each of the three glucose concentrations. The incubation media were equilibrated at 37°C, and incubations were initiated by transferring 10 islets to each tube. Incubations were conducted at 37°C, with gentle agitation. Samples of incubation media (100 µL) were collected at 10 and 60 min, frozen immediately and stored at -20°C until analysis. After the incubations, islet total insulin content was determined after HCl/ethanol extraction. Incubation medium containing the islets was adjusted to 77% (v/v) ethanol and 1% (v/v) HCl, vortexed, and incubated overnight at -20°C. The extracts were then diluted with KRB and stored at -20°C until analysis.
Assessing islet global gene expression
Reproducibility studies.
In each of two independent experiments, islets were isolated from two age-matched groups of 3–6 Wistar rats in the morning after overnight food deprivation. Islets (75/rat) were homogenized immediately in denaturing reagent (RNAzol B, Tel-Test, Friendswood, TX), snap-frozen in liquid nitrogen and pooled within each group. RNA was isolated by guanidium-thiocyanate-phenol-chloroform extraction (11
) followed by digestion of genomic DNA by RNase-free DNase (Genhunter, Nashville, TN) and quantified spectrophotometrically.
Radiolabeled cDNA target generation and hybridization were conducted according to instructions of the DNA microarray manufacturer (Research Genetics, Huntsville, AL). Radioactive target was generated by reverse-transcribing 1 µg of total RNA (Superscript II, Invitrogen, Carlsbad, CA) in the presence of 33P-dCTP (Perkin Elmer Life Sciences, Boston, MA). Target cDNA was purified by column chromatography (BioSpin 6, Bio-Rad Laboratories, Hercules, CA) and hybridized to a Research Genetics GF300 rat DNA microarray (12
). Of the > 5000 genes on the GF300 (the "probes"), 1500 are named genes, and the remainder are expressed sequence tags (EST). A complete listing of genes on the GF300 microarray is available on the Research Genetics website at ftp://ftp.resgen.com/pub/genefilters. The membrane was washed and imaged on a phosphor imaging system (Molecular Dynamics, Sunnyvale, CA or Packard Bioscience, Meriden, CT). To compare hybridization intensities between two groups of rats, after the first hybridization was imaged, the membrane was stripped and rehybridized with target from the second group. Images were analyzed by Pathways software (Research Genetics). The software normalized the overall hybridization intensity of every spot on the membrane so that the expression intensity of each gene could be compared quantitatively between the two groups. A different microarray was used for each of the two reproducibility studies.
Glucose induction experiment.
Age-matched adult Wistar rats were denied access to food for 48 h before receiving intraperitoneal injections of glucose (2 g/kg, in 9 g/L saline, n = 3 rats) or saline only (n = 3 rats) at time 0 and 3 h. At time 6 h, islets (150/rat) were isolated as described above and pooled within each group. RNA isolation, target generation, microarray hybridization and analysis were conducted as described above.
Comparing islet gene expression among litter size groups. Islet gene expression comparisons among the three litter-size groups were conducted at mean ages of 26 and 110 d. Islets for RNA isolation (50/rat) were collected from each rat included in the d 26 and 110 studies of islet function (n = 6–8 rats/group). The islets were homogenized immediately in denaturing reagent, snap-frozen in liquid nitrogen and pooled within litter-size groups. RNA isolation, target generation and microarray hybridization were conducted as described above.
Biochemical analyses
Serum glucose concentrations were measured by spectrophotometric assay (Trinder method, Sigma Diagnostics, St. Louis, MO). The within- and between-assay CV for the glucose assay were 3 and 5%, respectively. Serum and islet incubation medium insulin concentrations were measured by RIA (Linco Research, St. Louis, MO). The regular rat insulin kit (range 30–1500 pmol/L) and the sensitive rat insulin kit (range 3–150 pmol/L) were used as appropriate. The within- and between-assay CV for the regular rat insulin kit were 4 and 15%, and for the sensitive rat insulin kit, 3 and 13%, respectively.
Statistical methods
For the insulin and glucose tolerance tests, unpaired t tests were used to detect significant group differences in baseline serum glucose and insulin concentrations. To allow lower-order polynomial models to fit the data on serum glucose and insulin concentrations vs. time, baseline values were omitted from the data before conducting temporal ANOVA (except for analysis of serum glucose concentrations during insulin tolerance tests). Repeated-measures ANOVA (PROC mixed, SAS system for Windows version 6.12, SAS Institute, Cary, NC) was used to identify significant group differences in overall responses of serum glucose and insulin vs. time (in vivo studies) or in overall insulin accumulation vs. glucose concentration (islet studies). The models assumed an autoregressive covariance structure among repeated measures within each rat. The analyses were conducted in a "step-down" fashion, sequentially omitting nonsignificant highest-order terms (P > 0.10) until the simplest model that adequately explained the variance in the data was identified. The direct relation between serum glucose and insulin concentrations during the glucose tolerance tests was evaluated by repeated-measures ANOVA using the model:
 |
 |
An 
level of 0.05 was used to identify significant effects.
Body weight.
At age 21 d, SL rats were 9% heavier and LL rats 34% lighter than C rats (Fig. 1
). The weight difference between LL and C, but not SL and C, persisted into adulthood.
|
Insulin and glucose tolerance tests were conducted directly after weaning and in adulthood to discriminate between potential primary imprints in the endocrine pancreas and skeletal muscle. The insulin tolerance tests showed no significant group differences in whole-body insulin-responsive glucose uptake at either age 26 or 110 d (data not shown). Hence, these findings did not support primary imprinting of skeletal muscle insulin sensitivity in this model.
The glucose tolerance tests showed no significant group differences in serum glucose dynamics at age 26, 110 or 255 d (data not shown). However, at age 26 d, LL rats had significantly lower serum insulin concentrations in the glucose tolerance tests relative to C rats (Fig. 2A
). The direct dependence of serum insulin on serum glucose concentrations during the d 26 glucose tolerance tests was evaluated by repeated-measures ANOVA. In 26-d-old C rats, serum insulin increased by 19.2 pmol/L for every 1 mmol/L increase in serum glucose (Table 1
). The significant glucose · group terms indicate that in LL and SL rats, the slope of this relation was reduced, by 9.8 and 6.8 pmol insulin per mmol glucose, respectively.
|
|
B). Furthermore, the slope of the relation between serum glucose and insulin concentrations during the d 110 glucose tolerance tests (Table 1)
was reduced by about half in LL and SL rats, relative to that in C rats. This result is quantitatively similar to that found at age 26 d. A significant reduction in the slope of this relation was also observed in the results of the glucose tolerance tests at age 255 d, but only in SL rats (Table 1)
. In summary, the results of the insulin tolerance tests did not support primary imprinting of skeletal muscle insulin sensitivity. However, the glucose tolerance tests indicated an early and sustained defect in glucose-responsiveness of pancreatic insulin secretion in both the LL and SL groups. These findings suggested primary imprinting in the endocrine pancreas.
Studies of islet insulin secretion.
If the endocrine pancreas maintains a primary imprint in this model, the defect in glucose-stimulated insulin secretion should be maintained when pancreatic islets are studied in isolation. To test this hypothesis, glucose-stimulated insulin secretion of isolated islets was studied. Neither the mean diameter nor the total insulin content of isolated islets differed significantly among the three groups at any age (data not shown). Glucose-stimulated insulin secretion of LL islets was not significantly different from that of C islets at age 26 or 110 d (Fig. 2
C and D). At age 280 d, LL islets showed a significant decrease in glucose-stimulated insulin secretion relative to C islets (Fig. 3
).
|
C). At age 110 d, SL islets exhibited a left-shift of the glucose response curve (Fig. 2
D), in that maximal 2nd-phase insulin secretion was reached at 10 mmol/L glucose (glucose responsiveness of isolated rat islets normally reaches a plateau at 15–20 mmol/L glucose). The defect in SL islets persisted into late adulthood. At age 280 d (Fig. 3)
the SL group had a diminished ability to increase 1st-phase insulin secretion in response to elevated glucose concentrations. Islet gene expression studies.
The complexity of pancreatic islet glucose-stimulated insulin secretion (13
,14
) suggested many potential regulatory targets that might underlie the sustained functional defects identified in SL islets. DNA microarrays, which enable simultaneous expression monitoring of thousands of genes, were chosen as an effective way to identify potential regulatory targets and also focus future studies.
Before using the microarrays to compare islet gene expression among groups, reproducibility studies were conducted to establish criteria for identifying measurable expression changes. In two independent experiments, each comparing islet gene expression between two age-matched groups of untreated rats, many genes expressed at low levels showed artifactually high between-group expression ratios (data not shown). To work around this "signal-to-noise" problem, group comparisons of gene expression were limited to the 2000 genes with the highest mean expression levels in a given comparison. When genes with very low expression levels were thereby excluded from the results of the two reproducibility experiments, not a single gene showed an absolute differential expression ratio > 1.5 in both comparisons. Therefore, the following criteria were used to identify genes showing between-group differential expression: 1) mean expression intensity ranking in the top 2000 and 2) between-group expression ratio > 1.60.
An in vivo glucose induction experiment was conducted to validate the use of the microarrays to study islet gene expression changes. Compared with islets from vehicle-injected rats, only three named genes and five EST were expressed differentially in islets from glucose-injected rats; all were up-regulated in glucose-treated rats (Table 2
). The increase in insulin 2 mRNA levels was, of course, anticipated (15
). The conservative nature of the criteria used to determine differential expression is reflected by the relatively small number of genes identified as responding to an acute glucose challenge.
|
). Only one named gene, carbonyl reductase, showed LL vs. C differential expression at both 26 and 110 d. Notably, the expression of 5 EST was up-regulated in LL islets at both 26 and 110 d (Table 3)
.
|
). Of the named genes, insulin 2 showed the highest C:SL expression ratio of 2.90, indicating a 65% reduction in SL relative to C islets. SL vs. C islet gene expression differences of named genes at age 110 d were strikingly similar to those found in the postweaning studies. A total of 10 named genes and 48 EST showed differential expression at both ages (Table 4)
. Notably, of the 13 named genes that were expressed differentially between C and SL islets at age 26 d, the 10 with the highest d 26 expression ratios continued to show differential expression at age 110 d. The relative ordering by between-group expression ratios was very similar at both ages. The striking quantitative stability over time of the islet expression differences between C and SL islets provided further evidence in support of primary imprinting within the endocrine pancreas in this model.
|
The studies of isolated islets were crucial to testing this possibility because of the requirement that in vivo effects be maintained in vitro by primary imprint tissues. Reduced glucose-stimulated insulin secretion was found only in isolated SL islets directly after the imprinting period and throughout the period of adulthood that was studied. Thus, the studies of isolated islets provided support for primary imprinting in the endocrine pancreas in SL, but not LL rats.
The insulin tolerance tests found no group differences in whole-body insulin-responsive glucose uptake. However, during the glucose tolerance tests, the SL and LL rats exhibited normal glucose tolerance despite blunted insulin secretion, suggesting increased insulin sensitivity in those groups. These apparently contradictory results merit comment. One possible explanation is that even the blunted insulin response of the SL and LL groups would normally be sufficient to stimulate maximal glucose uptake by peripheral tissues. In this case, although C rats secreted more insulin in response to the glucose load, their rate of glucose uptake would be similar to that of SL and LL rats.
The second objective of the current research was to gain insight into the molecular mechanisms that potentially mediate persistent differences in primary imprint tissue. The complexity of pancreatic islet glucose-stimulated insulin secretion (13
,14
) and the limited quantity of RNA obtainable from isolated islets made DNA microarrays a very attractive approach to meet this objective. Comparing C and SL islets, 10 named genes were expressed differentially in the same direction at both ages (Table 4)
. Despite the many developmental changes between weaning and adulthood likely to affect islet gene expression, differential expression persisted through d 110 in >75% of the named genes differentially expressed directly after the imprinting period, i.e., d 26. The gene expression data thus provide highly supportive evidence that primary imprinting occurs in the endocrine pancreas of SL rats.
In all likelihood, other biologically important between-group differences in islet gene expression went undetected in these studies. Of the >5000 genes on the array, the analyses were limited to only the 2000 showing the highest expression levels in islet tissue, and genes showing (potentially significant) expression differentials < 1.60 were excluded from consideration. Furthermore, gene expression was assessed only in islets from fed rats in the morning. It is possible that different results may have been obtained under different temporal or dietary conditions, for example, after a dietary challenge.
On the other hand, limitations of the microarray data must be acknowledged. Ideally, differential expression of genes identified in microarray experiments would be verified by more traditional methods such as Northern hybridization. However, in the current study, the quantity of islet RNA isolated was insufficient to allow verification by Northern blots. To minimize concerns raised by this problem, reproducibility experiments were conducted at the outset to define realistic criteria for identifying differential expression in rat pancreatic islets. Further, the analysis focused on genes expressed differentially at both age 26 and 110 d. Hence, the list of genes showing persistent differential expression is the result of replicate microarray experiments conducted independently at the two ages.
The most salient finding of the microarray experiments was the persistent down-regulation of insulin 2 mRNA concentrations in SL islets. Not only is insulin the definitive and unique product of adult ß cells, but the persistently reduced levels of the transcript in SL rats is generally consistent with the defect in glucose-stimulated insulin secretion found in that group. Of all the genes showing persistent differential expression, insulin 2 showed the highest SL vs. C expression ratio directly after the imprinting period and in adulthood. Several of the other named genes showing SL vs. C persistent differential expression are involved in the regulation of insulin synthesis and secretion; thus their coordinated regulation with insulin 2 is plausible. These include mitochondrial ATP synthase, calcium ATPase, cholecystokinin (CCK) and neuronatin.
The increase in ATP:ADP ratio generated by mitochondrial oxidative phosphorylation links ß-cell glucose metabolism to insulin exocytosis. Hence, the persistent down-regulation of a mitochondrial ATP synthase subunit is consistent with the decreased glucose-sensitive insulin secretion found in the SL group. The potential role of this synthase is illustrated by the diabetic BHEcdb rat. In that model, the primary molecular defect causing insulin hyposecretion is a point mutation in one subunit of mitochondrial ATP synthase (16
). Calcium signaling also plays a central role in regulating insulin exocytosis. Calcium ATPase in the sarcoplasmic reticulum is responsible for pumping the divalent cation into that organelle. Here too, the down-regulation of the calcium ATPase in the SL group is consistent with their decreased glucose-responsive insulin secretion. For example, decreased activity and expression of the transporter have been found in rat models of type II diabetes (17
,18
). CCK is a peptide hormone secreted by the intestinal mucosa and neurons. CCK receptors localize to the ß cells in immunocytochemical studies of islet tissue, and the hormone is a known insulin secretagogue. Furthermore, CCK-expressing neurons innervate pancreatic islets (19
) and likely play a role in paracrine stimulation of islet hormone secretion. Accordingly, decreased islet CCK expression in SL islets is consistent with the persistent reduction in glucose-stimulated insulin secretion found in that group. Neuronatin, implicated in neurological development, is one of only 10 genes that, together with insulin, were found to be expressed at much higher levels in pancreatic ß cells than in cells (20
). Hence, this protein probably either contributes to the ß-cell specificity of insulin gene expression or is regulated by some of the same factors that regulate insulin gene transcription. In either case, its down-regulation in SL islets is consistent with the decreased levels of insulin 2 mRNA found in those rats.
The impressive quantitative stability of the gene expression differences between C and SL islets suggests that early postnatal nutrition may affect the establishment or maintenance of epigenetic mechanisms that perpetuate cell-specific patterns of gene activity throughout the life course. Moreover, these data suggest that genomically imprinted genes have an elevated susceptibility to environmental perturbation that increases the likelihood of their involvement in metabolic imprinting. Genomic imprinting is an epigenetic phenomenon in which the two alleles of a gene are expressed differentially depending on their parental origin (21
). Because genomically imprinted genes are effectively regulated by their previous environment, it has been proposed that they may represent susceptibility loci for epigenetic dysregulation (22
). Thus, it is compelling that of the 10 named genes showing persistent differential expression between the SL and C groups, insulin 2 and neuronatin are 2 of only 50 genes known to be genomically imprinted in the mouse (23
).
In conclusion, this study provides support for the hypothesis that nutritional stimuli during critical periods of development can modify adult chronic disease susceptibility. Specifically, the data indicate that even moderate overnutrition in the postnatal period may permanently alter endocrine pancreas gene expression. By evaluating logical characteristics of primary imprint tissue, evidence was obtained supporting a role for the endocrine pancreas in maintaining effect persistence in the suckling-period litter size model. Microarray results demonstrated quantitatively persistent differential gene expression and enabled the rapid identification of a small subset of genes that may participate in the mechanism(s) of metabolic imprinting in this model. Furthermore, this study provides evidence directly suggestive of an important role for early nutrition in influencing the epigenetic regulation of genomically imprinted genes.
1 Supported by National Institutes of Health Training Grant 5 T32 DK07158–23 to R.A.W.
3 Proximate composition: protein, 164 g/kg; metabolizable energy, 14.1 MJ/kg; fat, 71 g/kg; fiber, 146 g/kg.
4 Abbreviations used: C, control litter size group; CCK, cholecystokinin; EST, expressed sequence tag; LL, large litter size group; KRB, Krebs-Ringer bicarbonate buffered medium; SL, small litter size group.
Manuscript received 25 October 2001. Initial review completed 16 November 2001. Revision accepted 5 December 2001.
1. Leon, D. A. (1998) Fetal growth and adult disease. Eur. J. Clin. Nutr. 52:S72-S82.
2. Lucas, A. (1991) Programming by early nutrition in man. Ciba Found. Symp. 156:38-50.[Medline]
3.
Waterland, R. A. & Garza, C. (1999) Potential mechanisms of metabolic imprinting that lead to chronic disease. Am. J. Clin. Nutr. 69:179-197.
4. Barker, D. J. (1994) Programming the baby. Barker, D. J. eds. Mothers, Babies, and Disease in Later Life 1994:14-36 BMJ Publishing Group London. .
5. Cryer, A. & Jones, H. M. (1980) The development of white adipose tissue. Effect of litter size on the lipoprotein lipase activity of four adipose-tissue depots, serum immunoreactive insulin and tissue cellularity during the first year of life in male and female rats. Biochem. J. 186:805-815.[Medline]
6. Hahn, P. (1984) Effect of litter size on plasma cholesterol and insulin and some liver and adipose tissue enzymes in adult rodents. J. Nutr. 114:1231-1234.
7. Aubert, R., Suquet, J. P. & Lemonnier, D. (1980) Long-term morphological and metabolic effects of early under- and over-nutrition in mice. J. Nutr. 110:649-661.
8. Plagemann, A., Heidrich, I., Gotz, F., Rohde, W. & Dorner, G. (1992) Obesity and enhanced diabetes and cardiovascular risk in adult rats due to early postnatal overfeeding. Exp. Clin. Endocrinol. 99:154-158.[Medline]
9. Lacy, P. E. & Kostianovsky, M. (1967) Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes 16:35-39.[Medline]
10. Noda, M., Komatsu, M. & Sharp, G.W.G. (1996) The BHC-9 pancreatic ß-cell line preserves the characteristics of progenitor mouse islets. Diabetes 45:1766-1773.[Abstract]
11. Adamovicz, J. & Gause, W. C. (1995) RNA purification. Dieffenbach, C. W. Dveksler, G. S. eds. PCR Primer 1995:117-119 Cold Spring Harbor Laboratory Press Cold Spring Harbor, NY. .
12. Bowtell, D. D. (1999) Options available—from start to finish—for obtaining expression data by microarray [published erratum appears in Nat. Genet. 21: 241]. Nat. Genet. 21:25-32.[Medline]
13. Schuit, F. C. (1996) Factors determining the glucose sensitivity and glucose responsiveness of pancreatic beta cells. Horm. Res. 46:99-106.[Medline]
14. Ashcroft, F. M., Proks, P., Smith, P. A., Ammala, C., Bokvist, K. & Rorsman, P. (1994) Stimulus-secretion coupling in pancreatic beta cells. J. Cell Biochem. (suppl.) 55:54-65.
15. Giddings, S. J., Chirgwin, J. & Permutt, M. A. (1982) Effects of glucose on proinsulin messenger RNA in rats in vivo. Diabetes 31:624-629.[Abstract]
16. Mathews, C. E., McGraw, R. A., Dean, R. & Berdanier, C. D. (1999) Inheritance of a mitochondrial DNA defect and impaired glucose tolerance in BHE/Cdb rats. Diabetologia 42:35-40.[Medline]
17. Levy, J., Zhu, Z. & Dunbar, J. C. (1998) The effect of glucose and calcium on Ca2+-adenosine triphosphatase in pancreatic islets isolated from a normal and a non-insulin-dependent diabetes mellitus rat model. Metabolism 47:185-189.[Medline]
18. Varadi, A., Molnar, E., Ostenson, C. G. & Ashcroft, S. J. (1996) Isoforms of endoplasmic reticulum Ca(2+)-ATPase are differentially expressed in normal and diabetic islets of Langerhans. Biochem. J. 319:521-527.
19. Larsson, L. I. (1979) Innervation of the pancreas by substance P, enkephalin, vasoactive intestinal polypeptide and gastrin/CCK immunoractive nerves. J. Histochem. Cytochem. 27:1283-1284.[Abstract]
20. Arava, Y., Adamsky, K., Ezerzer, C., Ablamunits, V. & Walker, M. D. (1999) Specific gene expression in pancreatic beta cells: cloning and characterization of differentially expressed genes. Diabetes 48:552-556.[Abstract]
21.
Barlow, D. P. (1995) Gametic imprinting in mammals. Science (Washington DC) 270:1610-1613.
22. Pembrey, M. (1996) Imprinting and transgenerational modulation of gene expression; human growth as a model. Acta Genet. Med. Gemellol. (Roma) 45:111-125.[Medline]
23. Tilghman, S. M. (1999) The sins of the fathers and mothers: genomic imprinting in mammalian development. Cell 96:185-193.[Medline]
This article has been cited by other articles:
|
|
 |
|
  P. J. Martens and L. Romphf Factors Associated With Newborn In-Hospital Weight Loss: Comparisons by Feeding Method, Demographics, and Birthing Procedures J Hum Lact, August 1, 2007; 23(3): 233 - 241. [Abstract] [PDF]
|
|
|
|
 |
|
 N. Stettler, V. A. Stallings, A. B. Troxel, J. Zhao, R. Schinnar, S. E. Nelson, E. E. Ziegler, and B. L. Strom Weight Gain in the First Week of Life and Overweight in Adulthood: A Cohort Study of European American Subjects Fed Infant Formula Circulation, April 19, 2005; 111(15): 1897 - 1903. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. C. Mcmillen and J. S. Robinson Developmental Origins of the Metabolic Syndrome: Prediction, Plasticity, and Programming Physiol Rev, April 1, 2005; 85(2): 571 - 633. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Chu and M.-J. Tsai Neuronatin, a Downstream Target of BETA2/NeuroD1 in the Pancreas, Is Involved in Glucose-Mediated Insulin Secretion Diabetes, April 1, 2005; 54(4): 1064 - 1073. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A Waterland Commentary: The global relevance of 'biological Freudianism' Int. J. Epidemiol., February 1, 2005; 34(1): 15 - 17. [Full Text] [PDF]
|
|
|
|
 |
|
 N. Stettler, S. K Kumanyika, S. H Katz, B. S Zemel, and V. A Stallings Rapid weight gain during infancy and obesity in young adulthood in a cohort of African Americans Am. J. Clinical Nutrition, June 1, 2003; 77(6): 1374 - 1378. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
