| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Medical Clinic III and Policlinic, Justus Liebig University, D-35385 Giessen, Germany
Address all correspondence and requests for reprints to: Thomas Linn, M.D., Medical Clinic III and Policlinic, Justus Liebig University, Rodthohl 6, D-35385 Giessen, Germany.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and to the
inhibition of T cells, which release interleukin-4 and -10 (4). In the
presence of insulitis, pancreatic interferon- messenger RNA (mRNA)
levels are correlated to ß-cell destruction (5). Recently, it has been demonstrated that ß-cell death by apoptosis preceded the lymphocytic infiltration of the islets (6). Thus, ß-cell susceptibility to apoptosis may be another major underlying variable influencing the occurrence of diabetes via immune or nonimmune mechanisms in the NOD mouse. Apoptosis is a highly regulated process of cell death, which must be initiated by a suitable stimulus and by modulation of endogenous signal transduction pathways, which need not necessarily involve an immune response.
Although many experimental conditions known to protect NOD mice from diabetes have been described in recent years, there has been little success in the identification of promoting factors for autoimmune diabetes. The frequency and severity of disease vary in the NOD colonies around the world; for example, in female NOD mice an incidence of 18–100% was found (7). Several studies suggest that multiple environmental factors, such as diet, modify the development of insulin-dependent diabetes (8, 9, 10, 11).
The protecting effect of dietary protein against diabetes, particularly for semisynthetic diets, has been established in the autoimmune NOD mouse model (8, 12, 13). We have recently reported that a high protein (HP) diet demands the insulin secretion capacity of ß-cells damaged by insulitis and accelerates the occurrence of overt diabetes in the NOD mouse (14). It is possible that diets rich in protein and/or fat result in reduced insulin action, which, in turn, challenges insulin secretion. In this study we used a high fat (HF) and a high fat-high protein (HFHP) diet to produce states of reduced insulin sensitivity. The use of these diets was based on the observation that insulin resistance resulted in two thirds of all mouse strains consuming more than 30% fat of the daily energy intake (15).
Based on our previous findings we wanted to address the following
questions. Firstly, could a HF or a HFHP diet modulate diabetes
incidence in this model of autoimmune diabetes mellitus? Secondly,
could these diets induce a state of impaired glucose utilization in the
NOD mouse? Thirdly, were the diets effective in significantly altering
apoptosis and/or interferon- expression in pancreatic islets?
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Diet
The experimental diets contained: casein, 200 g/kg;,
D,L-methionine, 3 g/kg; cellulose, 50 g/kg;
choline bitartrate, 2 g/kg; vitamin mixture, 10 g/kg; and mineral
mixture, 35 g/kg. The HF group’s feed contained 200 g/kg fat, and the
low fat (LF) group’s feed contained 50 g/kg. The fat was composed of
equal amounts of lard and soybean oil. The difference in energy content
was compensated for by carbohydrates. The HF diet contained 250 g
maize starch/kg and 250 g sucrose/kg (39% fat, 17% protein, 43%
carbohydrates, and 18,900 kJ/kg). The LF diet contained 325 g
maize starch/kg and 325 g sucrose/kg (12% fat, 21% protein, 68%
carbohydrates, and 16,170 kJ/kg).
A combined HFHP feed was composed to mimic the ordinary situation, where fat and protein are frequently consumed together. This feed consisted of: casein, 400 g/kg; fat, 200 g/kg; maize starch, 100 g/kg; sucrose, 100 g/kg; D,L-methionine, 3 g/kg; cellulose, 50 g/kg; choline bitartrate, 2 g/kg; vitamin mixture, 10 g/kg; and mineral mixture, 35 g/kg (43% fat, 38% protein, 19% carbohydrates, and 21,840 kJ/kg).
The components of the feed were purchased from Altromin Rodent Food (Lage, Germany). Intake was calculated by subtracting the weight of the uneaten feed from the preweighed amount added to the cage. The mice were given free access to feed and weighed twice weekly, and feed consumption was measured every 3–4 days. Because food intake and body weight had been found to be similar in the three groups previously, pair-feeding was not instituted.
Surgical procedure
After a 12-h fast, mice were anesthetized with sodium
pentobarbital (50 mg/kg). Anesthesia during the clamp was maintained by
a low dose of isoflurane (1–2% volume). Cannulas were surgically
implanted into the jugular vein and carotid artery with the assistance
of a Leica Corp. StereoZoom dissecting microscope
(Leica Corp., Wetzlar, Germany). Because of the small
diameter of the artery, special tubes with an external diameter of 0.5
mm (Kalensee, Giessen, Germany) were used. This was the optimal size
for unrestricted blood flow through the cannula. Tubes with 0.58 mm
external diameter for connecting cannulas to infusion pumps were
inserted around the 0.5-mm cannulas and secured with suture and
adhesive (Histoacryl-Braun, Melsungen, Germany). The arterial cannula
was attached to a 23-gauge adaptor of the sample port and the jugular
cannula to the infusion line. For the arterial cannula, the distance
from the sampling port to the site of entry into the animal was
approximately 3 cm. Infusions of insulin and glucose were made into the
jugular vein; the carotid arterial cannula was used for blood sampling.
A sample port was adapted to minimize blood loss while allowing for an
accurate sample (
20 µl) to be taken.
In vivo tracer kinetics
[6,62H]Glucose (Promochem, Wesel, Germany) was
continuously infused (0.12 mg/kg·min) via the jugular venous catheter
(steady state after 20 min). Blood samples (20 µl) were obtained at
20, 25, and 30 min for determination of the relative content of
[6,6-2H]glucose. The plasma glucose concentration was
determined by a Beckman Coulter, Inc. glucose analyzer.
For determination of [6,6-2H]glucose, plasma samples were
deproteinized by mixing with 0.05 ml ethanol. After centrifugation, the
supernatant was evaporated to dryness in a stream of helium, and an
aldonitrile pentaacetate derivative of glucose was prepared. An aliquot
of the solution (1 µl) was injected into a gas chromatograph-mass
spectrometer (Unica 304, Philipps, Rahway, NJ) system equipped with a
60-m fused silica capillary column (OV-1). The peak abundance of
m/z 454 for [6,6-2H]glucose was monitored to calculate
the plasma enrichment of the labeled glucose. The precision of glucose
determination in plasma by isotope dilution showed a variation of
0.86–0.99%. Glucose appearance and disposal were calculated by the
nonsteady state equations of Steele (16).
Euglycemic clamp
After equilibration of the system (15 min), insulin (25 min, 1
mU/kg·min and 25 min, 10 mU/kg·min) was infused at a rate of 10
µl/min after a priming dose of 20 µl/min for the first 10 min.
Blood glucose levels were determined every 5 min by a Beckman Coulter, Inc. glucose analyzer (Palo Alto, CA). A variable rate
of glucose (2.5% dextrose) was infused to maintain the blood glucose
level at 5 mmol/liter. At the conclusion of the experiment, animals
were transferred for perfusion of the isolated pancreas.
In vitro pancreas perfusion
Details of this procedure were described previously (14). The
pancreas remained in situ during the experiment, and
catheters were fixed by Histoacryl (Braun, Melsungen, Germany). Basal
medium for perfusion (pH 7.4) consisted of 2.53 mmol/liter
CaCl2, 4.74 mmol/liter KCl, 1.19 mmol/liter
KH2PO4, 1.19 mmol/liter MgSO4,
118.5 mmol/liter NaCl, 25 mmol/liter NaHCO3, and 0.1% BSA
(fraction V, Sigma Chemical Co., Deisenhofen, Germany).
Perfusion media were warmed to 37 C, and glucose and arginine were
added, respectively (first and third phases, 5 mmol/liter glucose;
second phase, 30 mmol/liter glucose; fourth phase, 5 mmol/liter glucose
plus 19 mmol/liter arginine). After a 20-min equilibration at 5
mmol/liter glucose, the perfusion effluent was collected via a catheter
placed in the vena portae. Samples were immediately frozen (-20 C).
Insulin and glucagon were determined by RIA (RIA-gnost Insulin,
Behring, Marburg, Germany), with rat insulin and glucagon as standards
(Novo, Mainz, Germany). Insulin secretion was calculated by subtracting
mean insulin concentrations at 5 mmol/liter glucose (15–20 min; Fig. 2
) from mean insulin concentrations at 30 mmol/liter (21–35 min).
|
In situ detection of apoptotic and proliferative islet cells
Apoptosis was searched for in islets of NOD mice by in
situ labeling the 3'-hydroxyl termini of DNA breaks with modified
deoxy-UTP. This method was used on four mice of each diet group at 10,
20, and 30 weeks of age. DNA breaks in ß-cells were identified by
incubating Paraplast-embedded pancreatic sections with a mixture
containing the enzyme terminal deoxynucleotidyl transferase and
fluorescein-deoxy-UTP (catalogue no. 1767291 and 1767305,
Boehringer Mannheim, Mannheim, Germany) for 1 h at
room temperature. After washing with Tris-HCl (pH 7.6), sections were
incubated with peroxidase antifluorescein antibody for 30 min and
stained with 3,3'-diaminobenzidine tetrahydrochloride using a
peroxidase substrate kit (Boehringer Mannheim).
Immunohistochemical localization of insulin and DNA strand breaks
confirmed the ß-cell origin of apoptosis. Variation of apoptotic cell
count from one pancreas to the other was 6–15% (n = 12; median,
9%).
ß-Cell proliferation was studied after in vivo labeling of four mice in each group with BrdU from Sigma Chemical Co. (50 mg/kg, ip, 1 h before pancreas dissection). Paraffin-embedded pancreatic sections were double stained for BrdU and insulin. Immunostaining for BrdU was performed with a cell proliferation kit (catalogue no. 1758756, Boehringer Mannheim). Briefly, sections were incubated with a mixture of nuclease and a mouse alkaline phosphatase-conjugated anti-BrdU monoclonal antibody for 1 h at room temperature, washed for 15 min with Tris-HCl, and incubated with a peroxidase antimouse IgG (IgG2) for 30 min and stained with 3,3'-diaminobenzidine tetrahydrochloride using peroxidase substrate kit. After BrdU labeling, the same sections were washed and then stained for insulin. In these double stained sections, ß-cells exhibited brown cytosol, and BrdU-positive cells exhibited brown nuclei. To estimate the proliferation rate, ß-cells and BrdU-positive ß-cells were counted using a standard light microscope (Carl Zeiss, Wetzlar, Germany). ß-Cells exhibited brown cytosol, and BrdU-positive cells also exhibited brown nuclei. BrdU-negative cells had blue nuclei due to hematoxylin stain.
The mean number of ß-cells inspected for apoptosis or proliferation in a single pancreas was 9571 ± 2077. Results were expressed as the mean number of these cells per 1000 ß-cells.
Analysis of interferon- expression within the endocrine
pancreas
Total RNA was isolated from fresh pancreatic tissue by acid
guanidinium thiocyanate-phenol-chloroform extraction. Detection of mRNA
was performed by RT-PCR (18). For this purpose specific primers were
used for interferon- and cyclophilin (CLONTECH Laboratories, Inc., Palo Alto, CA). After 30 cycles at 94 C for 20 s and
60 C for 20 s in a GeneAmp PCR System 9600 (Perkin Elmer/Cetus, Norwalk, CT), the products were subjected to
electrophoresis followed by hybridization with specific
32P-labeled probes binding at sites between the primer
sequences. Quantification of the signals was performed by measuring the
32P-stimulated luminescence with a phosphorimager (Fujix
BAF 1000, Raytest, Staubenhardt, Germany). The values obtained for the
interferon- PCR product were normalized as the percentage of
32P incorporated into the cyclophilin PCR product.
Statistics
Experimental differences in diabetes incidence were assessed by
Kaplan-Meier life table analysis, using a log-rank test. ANOVA was used
for all comparisons among the three dietary groups followed by the
Newman-Keuls post-hoc test when appropriate. Statistical
tests were performed with the Statistical Package of the Social
Sciences (version 6.1.2, Munich, Germany).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
demonstrates the acceleration of diabetes incidence over the course of
time.
|
- and ß-cells, the
pancreas of NOD mice was perfused with glucose and arginine as
secretagogues. At 10 and 20 weeks, insulin and glucagon secretory
responses were increased in the HFHP group compared with those in the
HF- and LF-fed mice (20 weeks; Fig. 2
, a and b). At 20 weeks mean insulin secretion in the perfusate after
stimulation with 30 mmol/liter glucose was 189 ± 109 pmol/liter
for the LF mice, 368 ± 198 pmol/liter for the HF group
(P < 0.05), and 834 ± 243 pmol/liter for the
HFHP mice (P < 0.02 vs. LF). The
differences among the groups were even more obvious after stimulation
with arginine and glucose together.
At 30 weeks of age, mean insulin secretion (30 mM glucose)
had declined significantly in the HFHP group, indicating functional
ß-cell loss subsequent to the relative hypersecretion of insulin
shown in Fig. 2a (30 weeks; LF, 167 ± 89; HF, 275 ± 113;
HFHP, 156 ± 102 pmol/liter; P < 0.01 for HFHP at
20 weeks vs. 30 weeks).
The suppression of -cells of the endocrine pancreas by 30 mmol/liter
glucose was incomplete for the HFHP group, but not for the diet with HF
alone [mean glucagon release of the HFHP group, 78 ± 29
pmol/liter (P < 0.01 vs. LF group); HF,
38 ± 15 pmol/liter (P = NS); LF, 31 ± 17
pmol/liter]. Figure 2b also demonstrates that glucagon stimulation by
arginine was increased in the HFHP group compared with that in the two
other cohorts.
Endogenous glucose production and glucose disposal
The rate of glucose appearance in the basal state was
significantly greater in the HFHP group (3.2 ± 0.3 mg/kg·min in
the LF group; 3.5 ± 0.4 in the HF group; and 4.9 ± 0.4 in
the HFHP group; P < 0.05; Fig. 3). The effect of two concentrations of
insulin on glucose utilization was evaluated in 12-h fasted mice. The
glucose level was clamped at 5.1 ± 0.2 mmol/liter. The glucose
infusion rate increased 5 min after the beginning of the insulin
infusion relative to the concentration of infused insulin. The glucose
infusion rate was reduced similarly (34% with 1 mU/kg·min insulin
infusion rate and 25% with 10 mU/kg·min insulin infusion rate) in
the HF and HFHP groups compared with that in the LF group.
|
; P < 0.02). However,
an overall trend for reduction of ß-cell mass was found
morphometrically in all dietary groups with increasing age. LF-fed NOD
mice lost 5%, the HF group lost 22%, and HFHP-fed mice lost 68% of
their ß-cells between the 10th and 30th weeks. BrdU labeling of
ß-cells varied between 3.1/1000 cells at 10 weeks and 1.2/1000
ß-cells at 30 weeks, and there was no significant difference among
the three dietary groups. Figure 5 shows
a single islet with insulitis, and with an apoptotic ß-cell. The
number of apoptotic cells exceeded the number of proliferating
ß-cells in the HFHP group at 20 and 30 weeks of age [5.1 ± 0.6
apoptotic cells/1000 ß-cells in HFHP mice; 3.2 ± 0.4 in HF mice
(P < 0.05 vs. LF), and 2.8 ± 0.4 in
LF mice at 30 weeks (P < 0.01 vs. LF); Fig. 6]. Pancreatic interferon- mRNA as a
percentage of the cyclophilin mRNA was 2.1 ± 0.9% for the LF,
2.8 ± 1.1% (P = NS vs. LF) for the
HF, and 7.5 ± 1.7% for the HFHP diet (P < 0.05
vs. LF) at 20 weeks of age. At 30 weeks interferon- mRNA
was 5.9 ± 1.9% in the HFHP mice, not significantly different
from values in the LF and HF mice.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Role of impairment of insulin-stimulated glucose disposal
Progression to diabetes in HF and HFHP NOD mice was associated
with impairment of insulin-stimulated whole body glucose disposal. It
is plausible to assume that apart from hyperglycemia there is no
specific mechanism for reduced insulin action in type 1 diabetes. The
following findings support the possibility that insulin action may also
be relevant to diabetes progression even in the presence of
normoglycemia: 1) insulin-stimulated glucose disposal of 20-week-old
prediabetic NOD mice was reduced in the euglycemic clamp study; 2)
substrate-stimulated insulin release from the in vitro
perfused pancreas was enhanced in proportion to the suppression of
overall insulin sensitivity; 3) the number of apoptotic ß-cells rose
3-fold in the insulin-resistant HFHP group; and 4) at the age of 30
weeks mice with the most marked inhibition of glucose disposal had lost
two thirds of their initial ß-cell mass, whereas mice with normal
insulin sensitivity had preserved 90% of their ß-cell mass. Insulin
resistance, although not eliciting a stimulation of
interferon--producing T lymphocytes, may contribute to the
progression of diabetes by imposing a higher insulin demand on the
ß-cell. Conversely, studies have demonstrated that daily injections
of exogenous insulin prevented diabetes by inducing antigenic tolerance
and/or by relieving ß-cell insulin challenge (24, 25).
Although reduction of insulin-stimulated glucose disposal in mice fed a HF diet and in mice fed a HFHP diet was not different, and HF diet alone showed a trend toward elevated diabetes incidence, the latter effect became statistically significant only in the HFHP group. Therefore, suppression of glucose disposal was not sufficient to explain the diabetogenicity especially of the HP diet.
Diet influences the balance of apoptosis and proliferation of
ß-cells
Both apoptosis and proliferation were rare events in islets of
postweaning NOD mice (<0.1% of the total ß-cells). It has been
reported in the NOD mouse that proliferation of pancreatic duct cells
was more frequent than proliferation of islet cells; however, overall
cell proliferation was not stimulated by insulitis (26). The number of
apoptotic cells in pancreatic sections was comparable with the results
of cell culture experiments. An apoptotic rate of only 3% caused a
cell population to regress at 25%/day, and cell deletion by apoptosis
was completed rapidly, with apoptotic cells remaining visible for only
a few hours (27).
It was striking that after the first 6 weeks of feeding the HFHP diet, ß-cell mass was significantly higher compared with that in the other experimental groups. This could mean that neogenesis prevailed over death of ß-cells in the HP group only a few weeks after weaning and was not detectable at 10 weeks of age. We and others have reported previously that diets rich in protein and energy promoted ß-cell expansion and insulin biosynthesis, whereas lack of nutrients at an early age prevented islets from sufficient growth (14, 28). Islet cell mass seemed to be directly correlated with growth velocity at a young age and in fetal or prepubertal conditions. On the other hand, repair mechanisms failed to compensate for ß-cell defects triggered by autoimmune insulitis in most experimental settings (29).
Several prior investigations concluded that circulating metabolites tip the kinetic balance in favor of ß-cell death (30, 31, 32, 33). Other mechanisms, for instance putative intraislet growth factors, may be effective; however, they cannot be proved by our investigation. Several studies have suggested a link between dietary protein availability and the capacity of ß-cells to grow and proliferate. Although the HFHP diet was found to initially increase ß-cell mass, NOD mice on this diet had lost most of their ß-cells and insulin secretory capacity, as determined by the isolated perfused pancreas at the end of the observation period. Insulin release was also reduced after 30 weeks of feeding the HF diet compared with that in the LF group. Age-dependent reduction of islet mass superimposed the loss of ß-cells induced by diet in all three groups.
-Cells and glucagon secretion
An association between glucose-induced insulin release and islet
cell mass was previously established in the NOD mouse (34). This
correlation was also true in our study, as increased ß-cell mass in
the HFHP group was accompanied by greater insulin and glucagon release
at 20 weeks of age compared with that in the other dietary groups.
Glucagon secretion stimulated by arginine was significantly elevated in
the HFHP group, but not in the HF mice. Only the proliferation of
ß-cells was measured by us in the endocrine pancreas, but -cell
growth could also be stimulated by protein feeding. NOD ß-cells were
exposed to a destructive process from the 10th week on. This process
was promoted by the protein and the fat components of the experimental
diets. It is established in all animal models of type 1 diabetes that
destruction is selective for ß-cells, with -cells being spared
from death. Consequently, a relative overproportion of -cells must
be assumed in the islets of NOD mice fed the HFHP diet. Glucagon
per se may even represent a diabetogenic factor in the NOD
mouse, as increased glucagon plasma levels are associated with
hyperglycemia.
Nonimmune mechanism in the reduction of ß-cell mass
Another line of evidence demonstrates that there may be more
nonimmune factors involved in the destruction of ß-cells. Studies in
partially pancreatectomized rats identified functional changes in the
remaining ß-cells (35). First, glucose sensitivity increased,
followed a couple of weeks later by a fall in glucose-induced insulin
secretion and glucose potentiation, with the postulated mechanism being
reduced ß-cell stores of insulin. The working model is that ß-cell
hypersecretion caused by raised glucose sensitivity is a key pathogenic
factor, being responsible for a fall in insulin stores. It was also
reported for the NOD mouse that glucose sensitivity of ß-cells
normalized when hyperglycemica was corrected (36). Studies in the BB
rat confirm that postprandial hyperglycemia alone produces a state of
ß-cell hypersensitivity to glucose predating impaired glucose
responsiveness of the ß-cell (37). The resulting insulin
hypersecretion could be an early, critical step in the development of
the impaired glucose-potentiated insulin responses. In the
pancreatectomy model, increased hexokinase activity caused ß-cell
hypersensitivity to glucose, resulting in hyperproinsulinemia (38).
ß-Cell hypersecretion in a state of latent diabetes is a pivotal step
in ß-cell dysfunction, and this mechanism may also be effective in
autoimmune models of diabetes. The suggestion is that diets causing
glucose intolerance would promote ß-cell dysfunction, whereas diets
or agents lowering insulin secretion would protect against impaired
ß-cell function.
Diabetogenesis caused by the HFHP diet is not mediated by
interferon-
A higher cytokine ratio of interferon- and interleukin-4 does
not occur until the onset of diabetes in NOD mice. Therefore, it was
concluded that overt diabetes in the NOD mouse progresses as a
predominant inflammatory ß-cell dysfunction without actual ß-cell
destruction until late in the disease process (39). However, this
concept was recently challenged by the finding that spontaneous
ß-cell death was present in NOD mice at 3 weeks of age,
i.e. when insulitis had not yet appeared, and represented
the earliest and sole mode of cell death (6). Reduction of islet area
by diet before the onset of insulitis was also reported for the BB rat
(40). We and others have shown that diet-induced changes in target cell
expression of major histocompatibility complex class I antigens occur
at a very early stage before classic insulitis (17, 41) and that
macrophages preferentially attack hyperfunctional ß-cells at this
stage (42). These lines of evidence indicate that diets act through
early effects on the target cells and antigen-presenting cells. This
early interaction may not be sufficient to result in overt diabetes, as
this outcome would require long term exposure to putative food
diabetogens to cause apoptosis of the ß-cells, which due to flow of
ß-cell autoantigens into the circulation attracts infiltrating
leukocytes that damage more ß-cells. Decreasing insulin action gives
an additional insulin secretion stimulus that might well be a potent
trigger for the destruction of ß-cells. Recently, it was reported
that late stage, age-specific immunotherapy was successful in reducing
diabetes incidence in the NOD mouse (43). This could mean that diets
reducing insulin demand on the ß-cell are capable of preventing
manifest diabetes in late stage insulitis. Interferon- expression at
30 weeks was not different in the dietary groups, nor was the area
analysis of the islet infiltrate by imaging (data not shown),
indicating that with the present experimental approach there was no
difference in the recruiting of immune cells. These results underline
the importance of nonimmune mechanisms in feeding HP and HF diets.
There are thus several conclusions. The combination of high protein and
high fat in the feed was diabetogenic for young NOD mice.
Interferon- expression in islets was different among the dietary
groups only at 20 weeks, not at 30 weeks. Apoptosis of ß-cells was
significantly increased at 20 and 30 weeks of age in the HF and HPHF
groups. Potential nonimmune mechanisms resulting in ß-cell apoptosis
were 1) relative overproportion of glucagon release due to
morphological and/or functional stimulation of -cells by the protein
component, 2) intermittent hyperglycemia because of reduced peripheral
insulin sensitivity and augmented hepatic glucose production by fat and
protein, and 3) putative ß-cell stress factors produced by the
enteral digestion of protein.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received October 20, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene expression in pancreatic
islet-infiltrating mononuclear cells correlates with autoimmune
diabetes in nonobese diabetic mice. J Immunol 154:4874–4882[Abstract]
transcription. J Clin Invest 95:628–634
-Cell neogenesis in an animal model of
IDDM. Diabetes 46:599–606[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  A. Chamson-Reig, E. J Arany, K. Summers, and D. J Hill A low protein diet in early life delays the onset of diabetes in the non-obese diabetic mouse J. Endocrinol., May 1, 2009; 201(2): 231 - 239. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Cerf, K. Williams, X. I. Nkomo, C. J. Muller, D. F. Du Toit, J. Louw, and S. A. Wolfe-Coote Islet cell response in the neonatal rat after exposure to a high-fat diet during pregnancy Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2005; 288(5): R1122 - R1128. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Pighin, L. Karabatas, C. Pastorale, E. Dascal, C. Carbone, A. Chicco, Y. B. Lombardo, and J. C. Basabe Role of lipids in the early developmental stages of experimental immune diabetes induced by multiple low-dose streptozotocin J Appl Physiol, March 1, 2005; 98(3): 1064 - 1069. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Peppa, C. He, M. Hattori, R. McEvoy, F. Zheng, and H. Vlassara Fetal or Neonatal Low-Glycotoxin Environment Prevents Autoimmune Diabetes in NOD Mice Diabetes, June 1, 2003; 52(6): 1441 - 1448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Prentki, E. Joly, W. El-Assaad, and R. Roduit Malonyl-CoA Signaling, Lipid Partitioning, and Glucolipotoxicity: Role in {beta}-Cell Adaptation and Failure in the Etiology of Diabetes Diabetes, December 1, 2002; 51(90003): S405 - 413. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
