Cyclin-dependent kinase 4 (Cdk4) plays a central role in perinatal pancreatic β cell replication, thus becoming a potential target for therapeutics in autoimmune diabetes. Its hyperactive form, Cdk4R24C, causes β cell hyperplasia without promoting hypoglycemia in a nonautoimmune-prone mouse strain. In this study, we explore whether β cell hyperproliferation induced by the Cdk4R24C mutation balances the autoimmune attack against β cells inherent to the NOD genetic background. To this end, we backcrossed the Cdk4R24C knockin mice, which have the Cdk4 gene replaced by the Cdk4R24C mutated form, onto the NOD genetic background. In this study, we show that NOD/Cdk4R24C knockin mice exhibit exacerbated diabetes and insulitis, and that this exacerbated diabetic phenotype is solely due to the hyperactivity of the NOD/Cdk4R24C immune repertoire. Thus, NOD/Cdk4R24C splenocytes confer exacerbated diabetes when adoptively transferred into NOD/SCID recipients, compared with NOD/wild-type (WT) donor splenocytes. Accordingly, NOD/Cdk4R24C splenocytes show increased basal proliferation and higher activation markers expression compared with NOD/WT splenocytes. However, to eliminate the effect of the Cdk4R24C mutation specifically in the lymphocyte compartment, we introduced this mutation into NOD/SCID mice. NOD/SCID/Cdk4R24C knockin mice develop β cell hyperplasia spontaneously. Furthermore, NOD/SCID/Cdk4R24C knockin females that have been adoptively transferred with NOD/WT splenocytes are more resistant to autoimmunity than NOD/SCID WT female. Thus, the Cdk4R24C mutation opens two avenues in the NOD model: when expressed specifically in β cells, it provides a new potential strategy for β cell regeneration in autoimmune diabetes, but its expression in the immune repertoire exacerbates autoimmunity.

β cell cycle regulation is a key issue in diabetes research. β cell replication plays an essential role in maintaining β cell mass homeostasis in adults, and occurs actively during gestation and the perinatal period (1, 2), although in adult mammals, the rate of differentiated β cells undergoing replication is relatively low (3, 4, 5). Cyclin-dependent kinase 4 (Cdk4),3is involved in early G1 transition (6, 7) and has a central role in perinatal β cell replication (8, 9). Cdk4 binds to the D-type cyclins (D1, D2, and D3). The major function of the Cdk4/D cyclin holoenzyme is to phosphorylate the pocket protein pRb (6, 10) releasing E2F transcription factors responsible for cell cycle progression. R24C amino acid substitution in the Cdk4 kinase has been described in some familiar melanomas (11, 12). This mutation prevents the binding of the inhibitors of Cdk4 protein family to Cdk4 (7, 11, 12, 13), leading to Cdk4 hyperactivity. Knockin mice carrying the Cdk4R24C mutation develop β cell hyperplasia and pancreatic endocrine tumors in aged individuals from a nonautoimmune genetic background (9, 13). However, Cdk4R24C knockin mice do not exhibit hypoglycemia, and overall islet physiology is not altered (14, 15). These results suggest that the regulation of Cdk4 activity in β cells may become a potential target of therapeutic intervention in type 1 diabetes (T1D).

T1D is a chronic condition characterized by CD4 (16, 17) and CD8 (16, 18) T cell-dependent destruction of insulin-producing pancreatic β cells (19, 20). The NOD mouse is one of the best studied animal models of T1D (19, 20). NOD mice exhibit deficiencies in the induction of both central and peripheral tolerance (21, 22, 23, 24, 25, 26). Moreover, a wave of β cell apoptosis peaking around 2 wk of age occurs in healthy newborn mice (27) and though apoptosis is not normally associated with inflammation, the kinetics in the clearance of cell debris by NOD macrophages is slow compared with the nonautoimmune prone mouse strain BALB/c mice, thus providing an inflammatory environment triggering the autoimmune process in the NOD mouse model (27, 28, 29, 30).

Little is known about the immune reaction against β cell hyperplasia (31, 32), though it is believed that continuous exposure to certain Ags leads to tolerance by desensitization of Ag-specific T cells (33). Repetitive tolerogenic immunization induces anergy as a result of a diminished Th1 response in NOD mice (34, 35). However, it is not known whether continuous β cell replication due to Cdk4 hyperactivity (Cdk4/R24C) could lead to β cell survival to the autoimmune attack by tolerance induction to β cell Ags.

To explore whether the β cell mass hyperplasia induced by Cdk4 hyperactivity is resistant to the autoimmune attack in T1D and/or whether it allows a faster recovery of the islet β cell mass, we backcrossed the Cdk4R24C mice onto the NOD genetic background to study the development of both spontaneous and adoptively transferred autoimmune diabetes and to determine how both the immune system and the insulin-producing β cells are affected by hyperactivity of Cdk4.

Cdk4R24C knockin mice, originally in the mixed CD1/129Sv genetic background (9), were backcrossed onto the NOD background for 11 generations (N12). N12 NOD Cdk4R24C heterozygous mice, homozygous for all the NOD Idd alleles tested (17), were intercrossed to generate NOD mice either homozygous (R24C/HMZ) or heterozygous (R24C/HTZ) for the Cdk4R24C mutation, and NOD wild-type (WT) mice. The term NOD/Cdk4R24C knockin mice, unless specified in the text, comprehends both genotypes R24C/HMZ and R24C/HTZ mice. NOD/SCID mice were purchased from The Jackson Laboratory. All animal experimentation procedures performed were approved by the institutional animal ethics committee.

Mice were monitored weekly for the development of glycosuria with Medi-Test Glucose 3 (Macherey-Nagel). Diabetes was confirmed by measuring glycemia with the Accu-Check test strips (Accutrend; Roche Diagnostics); values over 200 mg/dl (>11.1 mM) were considered positive for diabetes.

Paraffin sections were counterstained with H&E (Sigma-Aldrich) for infiltration studies (Leica Qwin). To assess pancreatic infiltration, a minimum of six mice per experimental group and gender were used (≥180 islets counted/group). For infiltration score was: 0, no infiltration; 1, pancreatic infiltration or peri-insulitis; insulitis invading less; 2, or more; 3 than half of the islet area.

Paraffin sections were submitted to the ApoAlert DNA Fragmentation Assay kit (BD Clontech) and costained for insulin (DakoCytomation). A minimum of six females per experimental group was used (≥90 islets counted/group).

Pancreata paraffin sections were stained for insulin (ICN Pharmaceuticals) and detected by a peroxidated secondary Ab (Sigma-Aldrich Química). AEC chromogen was used as peroxidase substrate. Toluidine blue was used for islet counterstaining. The β cell area was quantified in a blinded fashion using optic microscopy and Leica IM 1000 QWin (Leica Microsystems). A minimum of six females per experimental group were used (≥180 islets counted/group).

NOD/SCID mice were transferred i.v. with 10 million spleen cells prepared as described (17) from prediabetic, 6- to7-wk-old gender-matched donors. Inhaled anesthetic was used (isoflurane; Abbot Laboratories).

A total of 200,000 responder cells were plated in 96-well plates. Ags (recombinant human insulin (Sigma-Aldrich); insulin B 12–25 peptide (VEALYLVCGERGFF) and GAD65p35 peptide (SRLSKVAPVIKARMMEYGTT; Sigma-Genosys) were added at 25 μg/ml. Agonistic anti-CD3 Ab (2C11 clone) was used at 1 μg/ml (BD Biosciences). Cells were cultured for 72 h, unless otherwise stated, in RPMI 1640 medium (The Broyle Ringmer; Biosera) with antibiotics, l-glutamine (2 mM), and 10% FCS. Tritiated methyl-thymidine was added at 1 μCi/well at day 3 (Cerdanyola; Amersham). Sixteen hours later, cells were harvested on filter mats (Cell Harvester; Wallac) and counted (beta plate; Wallac). Results are expressed as the percentage of increment in proliferation over basal values; this is calculated as the difference between total methyl-thymidine incorporation and basal incorporation divided by basal incorporation and multiplied by 100.

Cells were stained with fluorochrome-conjugated mAbs recognizing a series of surface markers (CD11b (myeloid lineage), CD11c (dendritic cells (DCs)), CD19 (B cells), CD25 (activation-regulation), CD3 (T cell), CD4 (Th), CD8 (T cytolytic), CD69 (activation), CD62L (naive), Fas (proapoptotic death receptor), IAd (MHC class II), Mac-3 (activated macrophages), F4/80 (macrophages)); Fc block (anti-CD16-CD32 24G2 clone) was performed before specific Ab staining. All Abs used were obtained from BD Pharmingen (BD Biosciences) with the exception of the F4/80 clone (Serotec). Acquisition was performed by flow cytometry (FACSCalibur; BD Biosciences). A total of 10,000 events were acquired for each sample.

Cells were stained for annexin V and 7-aminoactinomycin D (7AAD; Calbiochem) using the Annexin V Detection kit (BD Biosciences). Cells were acquired by flow cytometry.

A total of 2 × 106 total spleen cells were plated per well in 6-well plates and anti-CD3 Ab (2C11clone (1 μg/ml)) was added for either 0, 16, or 40 h. Cells were harvested and stained for both CD4 and CD8 lymphocytes before annexin V and 7AAD staining. Cells were acquired by flow cytometry.

Survival curves were analyzed using the log-rank test. To study the influences of genotype, stimulation, and time in culture on the variables analyzed, data were evaluated by ANOVA and analysis of covariance. When an effect or factor was statistically significant, the levels were pairwise compared with the multiple comparison method (36).

NOD/Cdk4R24C knockin male mice homozygous for the mutation (R24C/HMZ) exhibited a profound acceleration in the kinetics of the disease, starting at 10 wk of age compared with 15 wk in NOD WT littermates (Fig. 1,A). Furthermore, the cumulative incidence of the disease at 30 wk of age was notably higher in R24C/HMZ male (81%) than in WT littermates (24%) (Fig. 1,A). NOD/Cdk4R24C knockin male mice heterozygous for the mutation (R24C/HTZ) showed a tendency to have higher cumulative incidence of diabetes (38%) than WT males, this was lower than the R24C/HMZ male, suggesting a Cdk4R24C allele dosage effect (Fig. 1,A). The kinetics of diabetes onset in R24C/HTZ male mice was not accelerated (Fig. 1 A).

FIGURE 1.

Diabetes and islet infiltration is aggravated in NOD/Cdk4R24C mice. Cumulative incidence of diabetes (A and B) and infiltration scores (C and D) are shown. A, Male mice (WT: empty squares (n = 25); R24C HTZ: triangles (n = 26); and R24C HMZ: diamonds (n = 24)). B, Female mice (WT: empty squares (n = 25); R24C HTZ: triangles (n = 24); and R24C HMZ: diamonds (n = 22)). R24C HMZ and R24C HTZ were compared with the WT group. Pancreata from either male mice (C) or female (D) mice were scored for islet infiltration at different ages. The data shown are the result of three different section levels per mouse, and six mice per genotype and gender. ∗∗∗, p < 0.001, log-rank test.

FIGURE 1.

Diabetes and islet infiltration is aggravated in NOD/Cdk4R24C mice. Cumulative incidence of diabetes (A and B) and infiltration scores (C and D) are shown. A, Male mice (WT: empty squares (n = 25); R24C HTZ: triangles (n = 26); and R24C HMZ: diamonds (n = 24)). B, Female mice (WT: empty squares (n = 25); R24C HTZ: triangles (n = 24); and R24C HMZ: diamonds (n = 22)). R24C HMZ and R24C HTZ were compared with the WT group. Pancreata from either male mice (C) or female (D) mice were scored for islet infiltration at different ages. The data shown are the result of three different section levels per mouse, and six mice per genotype and gender. ∗∗∗, p < 0.001, log-rank test.

Close modal

WT, R24C/HTZ, and R24C/HMZ female mice showed similar cumulative incidence of diabetes (91, 91, and 81%, respectively) at 30 wk of age; however, the latency of the disease seems to be shortened in both R24C/HMZ and R24C/HTZ (9 and 10 wk of age, respectively) compared with WT (12 wk of age) in a Cdk4R24C allele dosage-dependent manner (Fig. 1 B).

To determine whether the diabetes exacerbation observed in R24C/HMZ mice was associated with an aggravation of islet infiltration, we examined histological sections of pancreata from either male or female NOD mice at different ages and the three possible genotypes (Fig. 1, C and D). The criteria used to select the ages for scoring the infiltration is based on the slower disease kinetics normally exhibited by NOD males compared with NOD females (20). Reflecting the disease exacerbation caused by the Cdk4R24C mutation in NOD males, infiltration scores were higher in mice carrying the Cdk4R24C mutation (38.86% of islets are heavily infiltrated in R24C/HMZ vs 16.69% in WT male mice) at 12 wk of age, when R24C/HMZ male mice start to develop diabetes (Fig. 1,C). At earlier ages (7–9 wk), the effect of the Cdk4R24C mutation was not as obvious (Fig. 1,C). Moreover, at 7 wk NOD/Cdk4R24C knockin females, both R24C/HMZ and R24C/HTZ, exhibited more profound islet injury than WT littermates (Fig. 1 D).

We proposed two possible, nonexclusive, hypotheses for the NOD/Cdk4R24C knockin phenotype: 1) in the NOD/Cdk4R24C knockin mice, the Cdk4R24C mutation is present in all cell types. Because macrophages and lymphocytes express Cdk4 (37, 38, 39, 40), and because this kinase is up-regulated upon TCR stimulation in T cells or BCR stimulation in B cells (37, 38, 39), the Cdk4R24C mutation would increase the basal and/or stimulated proliferative activity in the autoreactive repertoire. 2) Pancreatic β cells from NOD/Cdk4R24C knockin mice are more susceptible to autoimmune attack.

To test the first hypothesis and evaluate the diabetogenic capacity of NOD/Cdk4R24C lymphocytes, we performed adoptive transfer experiments of total splenocytes into NOD/SCID recipients (Fig. 2). NOD/SCID male adoptively transferred with R24C/HMZ male splenocytes exhibited the tendency to develop accelerated diabetes (onset at 5 wk posttransfer) compared with those transferred with WT splenocytes (onset at 8 wk posttransfer) (Fig. 2,A); however, the risk of becoming diabetic is not significantly different between both experimental groups. The cumulative incidence of diabetes was higher in the NOD/SCID male adoptively transferred with R24C/HMZ spleen cells (87%) than in the NOD/SCID male transferred with WT splenocytes (63%) (Fig. 2 A).

FIGURE 2.

NOD/Cdk4R24C spleen cells are more diabetogenic than their WT partners. NOD/SCID mice were adoptively transferred with spleen cells from (more than or equal to three donors per group). A, WT: (empty squares) (n = 11); R24C HTZ (triangles) (n = 10); and R24C HMZ (diamonds) (n = 8) male donors. B, WT: (empty squares) (n = 9); R24C HTZ (triangles) (n = 14); and R24C HMZ (diamonds) (n = 8) female donors. R24C HMZ and R24C HTZ groups were compared with the WT group. ∗, p < 0.05; ∗∗, p < 0.01, log-rank test.

FIGURE 2.

NOD/Cdk4R24C spleen cells are more diabetogenic than their WT partners. NOD/SCID mice were adoptively transferred with spleen cells from (more than or equal to three donors per group). A, WT: (empty squares) (n = 11); R24C HTZ (triangles) (n = 10); and R24C HMZ (diamonds) (n = 8) male donors. B, WT: (empty squares) (n = 9); R24C HTZ (triangles) (n = 14); and R24C HMZ (diamonds) (n = 8) female donors. R24C HMZ and R24C HTZ groups were compared with the WT group. ∗, p < 0.05; ∗∗, p < 0.01, log-rank test.

Close modal

The R24C/HMZ female splenocytes have a stronger diabetogenic action when injected into NOD/SCID females: the cumulative incidence of the disease rose dramatically (92%) and its onset was earlier (4 wk posttransfer) than in WT donors (44% and 8 wk posttransfer) (see Fig. 2 B). These observations confirm the first hypothesis because R24C/HMZ splenocytes, especially females, had higher diabetogenic capacity. To identify the origin of the enhanced autoreactivity in NOD/Cdk4R24C knockin mice, we further explored the behavior of NOD/Cdk4R24C splenocytes in the presence of different stimuli in vitro.

Spleen cells from 6- to 7-wk-old donors were cultured in the presence or absence of a series of T1D autoantigenic stimuli (recombinant human insulin, insulin B chain p12–25 peptide and GAD65 p35 (41, 42), and the anti-CD3 agonistic polyclonal stimulus as positive control of stimulation. Neither male nor female NOD/Cdk4R24C splenocytes (HMZ and HTZ) showed increased response toward any of the autoantigens tested (Fig. 3, A and C, respectively). However, both male and female R24C/HMZ splenocytes had significantly enhanced basal proliferation compared with WT control spleen cells (Fig. 3, B and D, respectively). Last but not least, R24C/HMZ female spleen cells had a much weaker response to anti-CD3 stimulation than WT and R24C/HTZ female splenocytes, suggesting that R24C/HMZ female lymphocytes are more sensitive to AICD.

FIGURE 3.

Spleen cell response after in vitro stimulation. Spleen cells from prediabetic, 6- to 7-wk-old male (A and B) or female mice (C and D) were cultured in the presence of different stimuli. The different genotypes are plotted as follows: WT (□), R24C HTZ (▦), and R24C HMZ (▪). Results are expressed as percentage of increment in proliferation over basal values (A and C), or basal proliferation (cpm) (B and D); n ≥ 6 independent experiments. ∗∗, p < 0.01; ∗∗∗, p < 0.005.

FIGURE 3.

Spleen cell response after in vitro stimulation. Spleen cells from prediabetic, 6- to 7-wk-old male (A and B) or female mice (C and D) were cultured in the presence of different stimuli. The different genotypes are plotted as follows: WT (□), R24C HTZ (▦), and R24C HMZ (▪). Results are expressed as percentage of increment in proliferation over basal values (A and C), or basal proliferation (cpm) (B and D); n ≥ 6 independent experiments. ∗∗, p < 0.01; ∗∗∗, p < 0.005.

Close modal

As shown above, female R24C/HMZ spleen cells had a superior diabetogenic activity than female WT spleen cells, which correlates with their enhanced activation status (Figs. 2,B and 3,D). However, female R24C/HMZ spleen cells proliferate less in response to anti-CD3 treatment in vitro (Fig. 3,C). We hypothesized that female R24C/HMZ T lymphocytes express more CD3 coreceptor molecules on the plasma membrane, and this would lead to both a dramatic activation and a drastic AICD upon CD3 engagement. Confirming our hypothesis, R24C/HMZ T lymphocytes had a significantly higher mean intensity of fluorescence due to CD3 staining than WT ones (93.7 ± 2.6 (n = 3) vs 76.3 ± 3.2 (n = 4), respectively). We further assayed the susceptibility of cultured T lymphocytes to AICD by annexin V staining at different time points following anti-CD3 stimulation (Fig. 4). Total spleen cells from female R24C/HMZ, R24C/HTZ, or WT individuals were cultured in the presence or absence of anti-CD3 stimulation. Samples were withdrawn at 0, 16, and 40 h poststimulation and stained for CD4 or CD8, annexin V (early apoptosis) and 7AAD (dead cells). Only the percentages of live apoptotic T lymphocytes are shown (Fig. 4). Spleen CD8 T cells from both R24C/HMZ and R24C/HTZ knockin females were more likely than WT CD8 T cells to undergo apoptosis shortly after stimulation (16 h) (Fig. 4,A, middle panel); at longer incubation periods (40 h), only spleen CD8 T cells from R24C/HMZ mice displayed higher levels of AICD than WT CD8 T cells (Fig. 4,A, lower panel). Regarding CD4 T cells, only at the longest incubation time tested (40 h) was the degree of AICD in CD4 T cells from both R24C/HMZ and R24C/HTZ knockin females significantly different from that of the WT group (Fig. 4 B, lower panel).

FIGURE 4.

Female NOD/Cdk4R24C T cells are more susceptible to AICD. The percentage of apoptotic either CD8 (left) or CD4 (right) T cells at 0, 16, and 40 h of in vitro anti-CD3 stimulation is represented. Mice were prediabetic, 6- to 7-wk-old females. #, Compares anti-CD3 (A-CD3) treated cells to their nonstimulated (NS) partners of the same genotype and time condition. ∗, Compares either the anti-CD3-treated (A-CD3) R24C (HTZ or HMZ) cells to the (A-CD3) WT cells at the same time point; or the nonstimulated (NS) R24C (HTZ or HMZ) cells to the (NS) WT cells at the same time point; n = 3. ∗, #, p < 0.05; ∗∗, ##, p < 0.01; ∗∗∗, ###, p < 0.005.

FIGURE 4.

Female NOD/Cdk4R24C T cells are more susceptible to AICD. The percentage of apoptotic either CD8 (left) or CD4 (right) T cells at 0, 16, and 40 h of in vitro anti-CD3 stimulation is represented. Mice were prediabetic, 6- to 7-wk-old females. #, Compares anti-CD3 (A-CD3) treated cells to their nonstimulated (NS) partners of the same genotype and time condition. ∗, Compares either the anti-CD3-treated (A-CD3) R24C (HTZ or HMZ) cells to the (A-CD3) WT cells at the same time point; or the nonstimulated (NS) R24C (HTZ or HMZ) cells to the (NS) WT cells at the same time point; n = 3. ∗, #, p < 0.05; ∗∗, ##, p < 0.01; ∗∗∗, ###, p < 0.005.

Close modal

Therefore, female R24C/HMZ lymphocytes are more susceptible to AICD upon engagement of the CD3 coreceptor, whose expression is, in turn, up-regulated in the knockin female mice.

In parallel, we performed proliferation assays in which tritiated methyl-thymidine was added at different time points (0, 16, 40, and 72 h) after the stimulation with anti-CD3 agonistic Ab (Fig. 5). Female R24C/HMZ spleen cells exhibit higher basal proliferation than NOD/WT spleen cells only after 72 h of stimulation (4581 ± 558 vs 2594 ± 414, in R24C/HMZ and WT, respectively) (Fig. 5,A). This enhanced basal proliferation in R24C/HMZ spleen cells is also translated into a significantly lower percentage of increment in proliferation over basal values upon polyclonal stimulation at 72 h (201.98 ± 28.68 vs 561.63 ± 79.96, in R24C/HMZ and WT, respectively) (Fig. 5 B). No differences between both genotypes were found at the other three time points tested.

FIGURE 5.

Spleen cell response after in vitro polyclonal stimulation is altered by the Cdk4R24C mutation. Spleen cells from prediabetic, 5- to 6-wk-old female were cultured in the presence of anti-CD3 agonistic stimulation (1 μg/ml) for different incubation periods (0, 16, 40, and 72 h) before the addition of tritiated methyl-thymidine. The different genotypes are plotted as follows: WT: (empty squares) and R24C HMZ (diamonds). Results are expressed as percentage of increment in proliferation over basal values (A) or basal proliferation (cpm) (B); n ≥ 3 independent experiments; ∗, p < 0.05.

FIGURE 5.

Spleen cell response after in vitro polyclonal stimulation is altered by the Cdk4R24C mutation. Spleen cells from prediabetic, 5- to 6-wk-old female were cultured in the presence of anti-CD3 agonistic stimulation (1 μg/ml) for different incubation periods (0, 16, 40, and 72 h) before the addition of tritiated methyl-thymidine. The different genotypes are plotted as follows: WT: (empty squares) and R24C HMZ (diamonds). Results are expressed as percentage of increment in proliferation over basal values (A) or basal proliferation (cpm) (B); n ≥ 3 independent experiments; ∗, p < 0.05.

Close modal

In conclusion, the data from annexin V staining (Fig. 4), as well as the time course of methyl-thymidine incorporation (Fig. 5), suggest that the lower proliferative response in front of anti-CD3 stimulation detected in female R24C/HMZ spleen cells at 72 h compared with WT cells is due to higher earlier AICD accompanied by increased basal proliferation which contributes to significantly diminish the stimulation calculated as a percentage of increment over basal values.

We next characterized R24C/HMZ spleen cells by staining them for surface markers to explore their activation state and/or susceptibility to apoptosis. Only the markers that presented significant differences between the HMZ and WT genotypes either in male and/or female cells are shown (Tables I and II).

Table I.

Surface marker profiling from male NOD/Cdk4R24C splenocytesa

Spleen LymphocytesSpleen Macrophages
CD19+CD25+CD4+CD25+dCD4+CD62L+CD8+CD62L+CD4+CD69+bCD3+CD95+dCD4+CD95+bCD8+CD95+dCD11b+F4/80+dCD11b+Mac-3+b
WT 1.04 ± 0.17 2.61 ± 0.29 22.28 ± 5.42 1.6 ± 0.73 0.38 ± 0.09 35.69 ± 4.75 20.68 ± 2.54 10.65 ± 1.00 8.34 ± 3.86 21.89 ± 11.47 
HMZ 0.72 ± 0.37 6.58 ± 0.64 15.5 ± 6.51 4.09 ± 1.85 2.71 ± 0.77 15.16 ± 1.74 10.41 ± 4.73 3.92 ± 1.12 26.50 ± 2.99 46.05 ± 4.75 
Spleen LymphocytesSpleen Macrophages
CD19+CD25+CD4+CD25+dCD4+CD62L+CD8+CD62L+CD4+CD69+bCD3+CD95+dCD4+CD95+bCD8+CD95+dCD11b+F4/80+dCD11b+Mac-3+b
WT 1.04 ± 0.17 2.61 ± 0.29 22.28 ± 5.42 1.6 ± 0.73 0.38 ± 0.09 35.69 ± 4.75 20.68 ± 2.54 10.65 ± 1.00 8.34 ± 3.86 21.89 ± 11.47 
HMZ 0.72 ± 0.37 6.58 ± 0.64 15.5 ± 6.51 4.09 ± 1.85 2.71 ± 0.77 15.16 ± 1.74 10.41 ± 4.73 3.92 ± 1.12 26.50 ± 2.99 46.05 ± 4.75 
a

Percentage of 6- to 7-wk-old spleen cells exhibiting the marker combination in the corresponding column; more than or equal to three experiments per experimental group. Those cell subpopulations statistically different between both genotypes are shown in bold.

b

, p < 0.05;

c

, p < 0.01;

d

, p < 0.005.

Table II.

Surface marker profiling from female NOD/Cdk4R24C splenocytesa

Spleen LymphocytesSpleen Macrophages
CD19+CD25+bCD4+CD25+CD4+CD62L+cCD8+CD62L+bCD4+CD69+CD3+CD95+CD4+CD95+CD8+CD95+cCD11b+F4/80+CD11b+Mac-3+
WT 0.09 ± 0.13 4.37 ± 0.53 27.21 ± 0.90 7.67 ± 0.85 1.68 ± 0.46 15.38 ± 3.72 9.63 ± 2.46 3.24 ± 1.03 30.30 ± 3.38 51.56 ± 6.63 
HMZ 0.41 ± 0.06 4.12 ± 0.28 21.03 ± 2.13 5.96 ± 1.08 2.41 ± 1.20 19.75 ± 4.33 13.03 ± 2.68 6.11 ± 1.38 12.54 ± 3.99 39.24 ± 14.32 
Spleen LymphocytesSpleen Macrophages
CD19+CD25+bCD4+CD25+CD4+CD62L+cCD8+CD62L+bCD4+CD69+CD3+CD95+CD4+CD95+CD8+CD95+cCD11b+F4/80+CD11b+Mac-3+
WT 0.09 ± 0.13 4.37 ± 0.53 27.21 ± 0.90 7.67 ± 0.85 1.68 ± 0.46 15.38 ± 3.72 9.63 ± 2.46 3.24 ± 1.03 30.30 ± 3.38 51.56 ± 6.63 
HMZ 0.41 ± 0.06 4.12 ± 0.28 21.03 ± 2.13 5.96 ± 1.08 2.41 ± 1.20 19.75 ± 4.33 13.03 ± 2.68 6.11 ± 1.38 12.54 ± 3.99 39.24 ± 14.32 
a

Percentage of 6- to 7-wk-old spleen cells exhibiting the marker combination; more than or equal to three experiments per experimental group. Those cell subpopulations statistically different between both genotypes are shown in bold.

b

, p < 0.05;

c

, p < 0.01;

d

, p < 0.005.

Both CD95+CD4+ and CD95+CD8+ T cells were less frequent in R24C/HMZ male mice compared with WT male mice (Table I). Accordingly, total cell numbers in HTZ and HMZ male spleens were significantly lower than in WT male mice (data not shown). CD25+CD4+ T cells and CD69+CD4+ T cells were more frequent in male R24C/HMZ spleens than in male WT spleens, indicating that R24C/HMZ spleen cells are more activated compared with those from WT male mice. In addition, macrophage (CD11b+F4/80+ cells) and activated macrophages (CD11b+Mac-3+ cells) were more represented in R24C/HMZ spleen cells than in WT ones (Table I). Summarizing, the R24C/HMZ male splenocytes had a more activated phenotype than WT splenocytes and T cells seem to be less susceptible to Fas (CD95)-induced apoptosis.

We did not find any difference in female total spleen cell count between the three genotypes (data not shown). In agreement with our results of the proliferation studies, female R24C/HMZ spleens had a higher ratio of CD8+ T cells expressing Fas on their surface compared with WT CD8+ T cells. Moreover, the ratio of R24C/HMZ CD4+ T cells expressing CD62L on their membrane was decreased compared with WT CD4+ T cells, suggesting enhanced activation state in female R24C/HMZ splenocytes (Table II). In contrast, the frequency of CD25+, activated, B lymphocytes was decreased in female R24C/HMZ mice (Table II). Thus, female NOD/Cdk4R24C spleen CD8 T cells appeared more susceptible to apoptosis and T cells had a more activated phenotype compared with gender-matched WT splenocytes.

Because draining PLN host the effective mounting of the immune response against islet Ags, we determined the proliferation of total cells from pancreatic lymph nodes from prediabetic mice (6–7 wk of age) in response to various stimuli (Fig. 6).

FIGURE 6.

Response of PLN cells after in vitro stimulation. PLN cells from prediabetic, 6- to 7-wk-old males (A and B) or females (C and D) were cultured in the presence of different stimuli. The different genotypes are plotted as follows: WT (□), R24C HTZ (▦), and R24C HMZ (▪); n ≥ 6 independent experiments for each point. ∗, p < 0.05; ∗∗∗, p < 0.005.

FIGURE 6.

Response of PLN cells after in vitro stimulation. PLN cells from prediabetic, 6- to 7-wk-old males (A and B) or females (C and D) were cultured in the presence of different stimuli. The different genotypes are plotted as follows: WT (□), R24C HTZ (▦), and R24C HMZ (▪); n ≥ 6 independent experiments for each point. ∗, p < 0.05; ∗∗∗, p < 0.005.

Close modal

Neither male nor female NOD/Cdk4R24C splenocytes (HMZ and HTZ) showed increased response toward any of the autoantigens tested compared with WT cells (Fig. 6, A and C, respectively). Reflecting the phenotype shown by their homologue spleen cells, proliferation of female NOD/Cdk4R24C PLN cells (both R24C/HMZ and R24C/HTZ) is impaired upon CD3 engagement in comparison to WT PLN cells (Fig. 6,C). We only found changes in basal proliferation in female R24C/HMZ PLN cells (Fig. 6, B and C).

Our second hypothesis was that an increased apoptosis undergone by NOD/Cdk4R24C β cells accounts, at least in part, for the exacerbated diabetes shown by NOD/Cdk4R24C mice. To test this hypothesis, we examined the levels of cell apoptosis, quantified by the TUNEL technique on histological sections of pancreata from male (7, 9 and 12 wk of age) and female mice (3, 5 and 7 wk of age) of the three possible genotypes. No significant differences were found in any of the groups (data not shown).

We conclude that the exacerbated diabetic phenotype exhibited by NOD/Cdk4R24C knockin mice is caused by an abnormally autoreactive lymphocytic compartment and not to an increased susceptibility to apoptosis by NOD/Cdk4R24C pancreatic β cells.

We therefore tested whether the hyperreplicative activity of NOD/Cdk4R24C β cells, in the absence of the enhanced autoreactivity of the NOD/Cdk4R24C immune repertoire, would render CDK4R24C mice resistant to the autoimmune attack inflicted by the “normal” NOD immune repertoire. To this end, we introduced the R24C mutation in Cdk4 in the NOD/SCID genetic background, which is deficient in T and B lymphocytes, and studied both spontaneous and adoptively transferred diabetes onset in female individuals (Figs. 7 and 8). Neither NOD/SCID R24C/HMZ nor NOD/SCID WT female mice developed spontaneous diabetes (data not shown) indicating that the immune system remaining in the NOD/SCID R24C/HMZ female (macrophages, DCs, NK cells, etc.) is unable to cause the disease in the absence of lymphocytes. As expected, 6-wk-old NOD/SCID R24C/HMZ female mice exhibit spontaneous β cell mass hyperplasia compared with NOD/SCID WT female mouse littermates (see Fig. 7).

FIGURE 7.

NOD/SCID R24C HMZ mice develop β cell hyperplasia. Pancreata from either 6- to 7-wk-old NOD/SCID WT or NOD/SCID R24C HMZ female mice were submitted to morphometric analysis. The average percentage of β cell area in front of the whole islet area is represented. The different genotypes are plotted as follows: WT (□) and R24C HMZ (▪). Six mice have been analyzed for each experimental group (a minimum of 30 islets/mouse have been counted). ∗∗∗, p < 0.005 two-tailed t test.

FIGURE 7.

NOD/SCID R24C HMZ mice develop β cell hyperplasia. Pancreata from either 6- to 7-wk-old NOD/SCID WT or NOD/SCID R24C HMZ female mice were submitted to morphometric analysis. The average percentage of β cell area in front of the whole islet area is represented. The different genotypes are plotted as follows: WT (□) and R24C HMZ (▪). Six mice have been analyzed for each experimental group (a minimum of 30 islets/mouse have been counted). ∗∗∗, p < 0.005 two-tailed t test.

Close modal
FIGURE 8.

The NOD/Cdk4R24C mutation confers resistance to autoimmunity. Total spleen cells from WT female donors (more than or equal to three donor mice per experimental group) were adoptively transferred into either NOD/SCID R24C HMZ (diamonds) (n = 11); or NOD/SCID WT (empty squares) (n = 8) recipients. ∗, p < 0.05, log-rank test.

FIGURE 8.

The NOD/Cdk4R24C mutation confers resistance to autoimmunity. Total spleen cells from WT female donors (more than or equal to three donor mice per experimental group) were adoptively transferred into either NOD/SCID R24C HMZ (diamonds) (n = 11); or NOD/SCID WT (empty squares) (n = 8) recipients. ∗, p < 0.05, log-rank test.

Close modal

When prediabetic total spleen cells from WT female donors were adoptively transferred into NOD/SCID R24C/HMZ, the cumulative incidence of diabetes was significantly lower than in NOD/SCID WT recipients (81 vs 100%, respectively) (Fig. 8). Moreover, the most pronounced difference was observed between 13 and 14 wk posttransfer, by this point, WT recipients already showed a cumulative incidence of 100% whereas only 54% of R24C/HMZ had developed the disease. These data clearly demonstrate that R24C/HMZ islets are more resistant to the autoimmune attack than WT islets.

One of the main aims in T1D research is to engineer an unlimited, easily producible source of β cells. Because T1D therapy currently rests on allogenic cadaveric islet transplantation (43), the limited existing resources make this strategy unavailable to all T1D patients. Moreover, the prescriptive use of generalized immunosuppression following transplantation throughout a lifespan, with all its associated deleterious effects, makes solitary islet transplantation advisable only for those T1D patients with labile glycemic control (43).

An alternative to explore is the enhancement of autologous β cell replication. Because Cdk4 is an essential enzyme in perinatal β cell replication (13), in this study, we explored whether autoimmunity would be balanced by β cell mass hyperplasia in NOD mice. We found that NOD/Cdk4R24C knockin mice exhibited exacerbated disease. This enhancement of autoimmunity in NOD/Cdk4R24C knockin mice was due to the hyperactivity of the NOD immune repertoire, and not to β cell hyperplasia caused by the R24C mutation, because in NOD/Cdk4R24C individuals the proliferative immune responses toward typical T1D Ags were not increased, and basal proliferation values (non-Ag driven) were already higher compared with WT mice. Moreover, the adoptive transfer experiments of either NOD/Cdk4R24C or NOD/WT donors into NOD/SCID recipients corroborate that the NOD/Cdk4R24C immune system is hyperactive.

The aggravation of the diabetic phenotype is more obvious in NOD male mice than in female littermates and it seems to be R24C-dosage dependent. This could be explained by the fact that NOD (WT) male mice, per se, have a reduced and delayed incidence of diabetes compared with NOD (WT) female mice. NOD male thymocytes are more susceptible to dexamethasone-induced apoptosis; hence, autoreactive thymocytes may be more easily removed in male NOD thymuses by negative selection (44, 45) than those from NOD females. Therefore, our results suggest that the Cdk4R24C mutation demolishes the hormonal barrier protecting NOD WT males from autoimmunity. It has been reported that dividing T cells are more susceptible to AICD than activated, nondividing T cells (46).

We propose an explanation for the apparent contradiction between the lower proliferative response to anti-CD3 stimulation, and the enhanced diabetogenicity of female R24C/HMZ splenocytes: female splenocytes are more susceptible to activation by agonistic anti-CD3 treatment, due to the higher CD3 expression on their surface, and hence, more sensitive to AICD. Moreover, basal proliferation is enhanced in R24C/HMZ compared with WT cells, leading to lower values for the percentage of increment in proliferation over basal values.

Both, male and female R24C/HMZ splenocytes exhibit higher basal proliferation compared with WT cells, which indicates a state of enhanced basal activation due to the R24C mutation, and explains the exacerbated autoimmune phenotype in the R24C/HMZ mice.

Moreover, Cdk4 is clearly required for cytokine responsiveness in T cells (47). Surface markers on the lymphocyte and myeloid compartments in the prediabetic period reveal that both male and female NOD/Cdk4R24C cells present a more activated phenotype than WT gender-matched cells. The lower spleen cell count in both R24C/HTZ and R24C/HMZ compared with WT male individuals (data not shown) may reflect an enhanced early AICD, because there is a lower ration of CD95-expressing T lymphocytes in R24C/HMZ male at the age tested. However, while the total cell count for Mac-3+ cells is significantly decreased in R24C/HMZ male (data not shown), the ratio of activated macrophages is increased, resulting in a positive effect of the R24C mutation in the activation of spleen macrophages.

On the contrary, we have seen that pancreatic β cells are positively influenced by the Cdk4R24C mutation in NOD females. Enhanced β cell replication is not associated with an increment in β cell death, but with a protection against cytokine-induced β cell apoptosis in adoptive transfer experiments in which WT NOD spleen cells are transferred into NOD/SCID recipients carrying the R24C (CDK4R24C NOD/SCID) mutation or into WT NOD/SCID recipients as a control group. This may be due either to a higher number of β cell targets due to β cell hyperplasia in R24C mice and/or enhanced intrinsic resistance of these β cell targets to the autoimmune attack. In contrast, cell cycle activation is normally related to an increased susceptibility either to apoptosis in lymphocytes (46, 48) or to cytotoxic stimuli in β cells (49). Further studies need to be performed to determine whether Cdk4 hyperactivity induces resistance to apoptosis in β cells and whether tolerance to β cell Ags is promoted by the Cdk4 hyperactivity in NOD/SCID recipients.

In summary, here we demonstrate that Cdk4 hyperactivation affects the autoimmune repertoire, exacerbating its self-reactivity and pathogenicity, while the Cdk4 hyperactivation in the β cell mass causes hyperplasia and seems to counteract autoimmunity in a WT NOD immune repertoire. This is the first time in which the role of Cdk4 in autoimmunity has been analyzed. In addition, we have found expression of Cdk4 in human pancreatic islets (data not shown), which points to Cdk4 as a potential target in human T1D therapy.

In conclusion, research on Cdk4 activity, its regulation, and regulation of its expression is opening up a promising field of study in the development of new therapeutic strategies for T1D therapeutics by designing approaches to induce its activation in a controlled, cell type-specific fashion.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by an Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS) predoctoral fellowship (to N.M.); the European Foundation for the Study of Diabetes (EFSD)/Paul Langerhans/Amylin Pharmaceuticals Award 2004 (to C.M.); an EFSD/Lilly Research Fund grant (to C.M.); the Ministerio de Ciencia y Tecnología (Ramon y Cajal Program (to C.M.) and SAF 2003-06139 to A.G.)); Ministerio de Educacion y Ciencia Grant BMC2000-0130 (to S.O.); Ministerio de Educacion y Ciencia Grant SAF2004-02666 (to T.S.); Ministerio de Sanidad y Consumo PI040587 (to T.S.); the European Commission (MIRG-CT-2004-012692) (to T.S.); and RCMN 03/08 and RGDM 03/212, FISPI 051264, FISPI 051241 by the Instituto de Salud Carlos III, Madrid (to R.G.). C.M. is an investigator at the IDIBAPS.

3

Abbreviations used in this paper: Cdk4, cyclin-dependent kinase 4; T1D, type 1 diabetes; WT, wild type; DC, dendritic cell; AICD, apoptosis-induced cell death; PLN, pancreatic lymph node; HMZ, homozygous; HTZ, heterozygous.

1
Dor, Y., J. Brown, O. I. Martinez, D. A. Melton.
2004
. Adult pancreatic β-cells are formed by self-duplication rather than stem-cell differentiation.
Nature
429
:
41
-46.
2
Kassem, S. A., I. Ariel, P. S. Thornton, I. Scheimberg, B. Glaser.
2000
. β-cell proliferation and apoptosis in the developing normal human pancreas and hyperinsulinism of infancy.
Diabetes
49
:
1325
-1333.
3
Finegood, D. T., L. Scaglia, S. Bonner-Weir.
1995
. Dynamics of β-cell mass in the growing rat pancreas: estimation with a simple mathematical model.
Diabetes
44
:
249
-256.
4
Teta, M. S., S. Y. Long, L. M. Wartschow, M. M. Rankin, J. A. Kushner.
2005
. Very slow turnover of β-cells in aged adult mice.
Diabetes
54
:
2557
-2567.
5
Butler, A. E., J. Janson, S. Bonner-Weir, R. Ritzel, R. A. Rizza, P. C. Butler.
2003
. β-cell deficit and increased β-cell apoptosis in humans with type 2 diabetes.
Diabetes
52
:
102
-110.
6
Cozar-Castellano, I., M. Weinstock, M. Haught, S. Velázquez-García, D. Sipula, A. F. Stewart.
2006
. Evaluation of β-cell replication in mice transgenic for hepatocyte growth factor and placental lactogen: comprehensive characterization of the G1/S regulatory proteins reveals unique involvement of p21cip.
Diabetes
55
:
70
-77.
7
Mettus, R. V., S. G. Rane.
2003
. Characterization of the abnormal pancreatic development, reduced growth and infertility in Cdk4 mutant mice.
Oncogene
22
:
8413
-8421.9.
8
Martin, J., S. L. Hunt, P. Dubus, R. Sotillo, F. Nehme-Pelluard, M. A. Magnuson, A. F. Parlow, M. Malumbres, S. Ortega, M. Barbacid.
2003
. Genetic rescue of Cdk4 null mice restores pancreatic β-cell proliferation but not homeostatic cell number.
Oncogene
22
:
5261
-5269.
9
Rane, S. G., P. Dubus, R. V. Mettus, E. J. Galbreath, G. Boden, E. P. Reddy, M. Barbacid.
1999
. Loss of Cdk4 expression causes insulin-deficient diabetes and Cdk4 activation results β-islet cell hyperplasia.
Nat. Genet.
22
:
44
-52.
10
Nevins, J. R..
2001
. The Rb/E2F pathway and cancer.
Hum. Mol. Genet.
10
:
699
-703.
11
Zuo, L., J. Weger, Q. Yang, A. M. Goldstein, M. A. Tucker, G. J. Walker, N. Hayward, N. C. Dracopoli.
1996
. Germline mutations in the p16INKa binding domain of CDK4 in familial melanoma.
Nat. Genet.
12
:
97
-99.
12
Wolfel, T., M. Hauer, J. Scheneider, M. Serrano, C. Wolfel, E. Klehmann-Hieb, E. De Plaen, T. Anclen, K. H. Meyer zum Buschenfelde, D. Beach.
1995
. A p16INK4a-insensitive Cdk4 mutant targeted by cytolytic T lymphocytes in human melanoma.
Science
269
:
1281
-1284.
13
Sotillo, R., P. Dubus, J. Martín, E. De la Cueva, S. Ortega, M. Malumbres, M. Barbacid.
2001
. Wide spectrum of tumors in knock-in mice carrying a Cdk4 protein insensitive to INK4 inhibitors.
EMBO J.
20
:
6637
-6647.
14
Marzo, N., C. Mora, M. E. Fabregat, J. Martín, E. F. Usac, C. Franco, M. Barbacid, R. Gomis.
2004
. Pancreatic islets from cyclin-dependent kinase 4/R24C (Cdk4) knockin mice have significantly increased β cell mass and are physiologically functional, indicating that Cdk4 is a potential target for pancreatic β cell mass regeneration in Type 1 diabetes.
Diabetologia
47
:
686
-694.
15
Hino, S., T. Yamaoka, Y. Yamashita, T. Yamada, J. Hata, M. Itakura.
2004
. In vivo proliferation of differentiated pancreatic islet β cells in transgenic mice expressing mutated cyclin-dependent kinase 4.
Diabetologia
47
:
1819
-1830.
16
Katz, J., C. Benoist, D. Mathis.
1993
. Major histocompatibility complex class I molecules are required for the development of insulitis in non-obese diabetic mice.
Eur. J. Immunol.
23
:
3358
-3360.
17
Mora, C., F. S. Wong, C. H. Chang, R. A. Flavell.
1999
. Pancreatic infiltration but not diabetes occurs in the relative absence of MHC class II-restricted CD4 T cells: studies using the NOD/CIITA-deficient mice.
J. Immunol.
162
:
4576
-4588.
18
Serreze, D. V., E. H. Leiter, G. J. Christianson, D. Greiner, D. C. Roopenian.
1994
. Major histocompatibility complex class I-deficient NOD-B2null mice are diabetes and insulitis resistant.
Diabetes
43
:
505
-509.
19
Bergholdt, R., P. Heding, K. Nielsen, R. Nolsoe, T. Sparre, J. Storling, J. Nerup, F. Pociot, T. Mandrup-Poulsen.
2004
. Type 1 diabetes mellitus: an inflammatory disease of the islet.
Adv. Exp. Med. Biol.
552
:
129
-153.
20
Delovitch, T. L., B. Singh.
1997
. The nonobese diabetic mouse as a model of autoimmune diabetes: immune dysregulation gets the NOD.
Immunity
7
:
727
-738.
21
Kishimoto, H., J. Sprent.
2001
. A defect in central tolerance in NOD mice.
Nat. Immunol.
2
:
1025
-1031.
22
Zucchelli, S., P. Holler, T. Yamagata, M. Roy, C. Benoist, D. Mathis.
2005
. Defective central tolerance induction in NOD mice: genomics and genetics.
Immunity
22
:
385
-396.
23
Salomon, B., D. J. Lenschow, L. Rhee, N. Ashourian, B. Singh, A. Sharpe, J. A. Bluestone.
2000
. B7/CD28 costimulation is essential for the homeostasis of the CD4+CD25+ immunoregulatory T cells that control autoimmune diabetes.
Immunity
12
:
431
-440.
24
Gregori, S., N. Giarratana, S. Smiroldo, L. Adorini.
2003
. Dynamics of pathogenic and suppressor T cells in autoimmune diabetes development.
J. Immunol.
171
:
4040
-4047.
25
Ridgway, W. M., M. Fasso, C. G. Fathman.
1999
. A new look at MHC and autoimmune disease.
Science
284
:
749
-751.
26
Kanagawa, O., B. A. Vaupel, G. Xu, E. R. Unanue, J. D. Katz.
1998
. Thymic positive selection and peripheral activation of islet antigen-specific T cells: separation of two diabetogenic steps by and I-A(g7) class II MHC β-chain mutant.
J. Immunol.
161
:
4489
-4492.
27
Trudeau, J. D., J. P. Dutz, E. Arany, D. J. Hill, W. E. Fieldus, D. T. Finegood.
2000
. Neonatal β-cell apoptosis: a trigger for autoimmune diabetes?.
Diabetes
49
:
1
-7.
28
O′Brien, B. A., Y. Huang, X. Geng, J. P. Dutz, D. T. Finegood.
2002
. Phagocytosis of apoptotic cells by macrophages from NOD mice is reduced.
Diabetes
51
:
2481
-2488.
29
Eizirik, D. L., T. Mandrup-Poulsen.
2001
. A choice of death-the signal- transduction of immune-mediated β-cell apoptosis.
Diabetologia
44
:
2115
-2133.
30
Thomas, H. E., T. W. Kay.
2000
. β cell destruction in the development of autoimmune diabetes in the non-obese diabetic (NOD) mouse.
Diabetes Metab. Res. Rev.
16
:
251
-261.
31
Speiser, D. E., R. Miranda, A. Zakarian, M. F. Bachmann, B. McKall-Faienza, K. Odermatt, D. Hanahan, R. M. Zinkernagel, P. S. Ohashi.
1997
. Self antigens expressed by solid tumors do not efficiently stimulate naïve or activated T cells: implications for immunotherapy.
J. Exp. Med.
186
:
645
-653.
32
Schulz, B., B. Hehmke, E. Zander, B. Ziegler.
1990
. Autoimmune reactions in a patient with malignant insulinoma treated by multiple low dose streptozotocin.
Exp. Clin. Endocrinol.
95
:
77
-82.
33
Anderson, P. O., B. A. Manzo, A. Sundstedt, S. Minaee, A. Symonds, S. Khalid, M. E. Rodriguez-Cabezas, K. Nicolson, S. Li, D. C. Wraith, P. Wang.
2006
. Persistent antigenic stimulation alters the transcription program in T cells, resulting in antigen-specific tolerance.
Eur. J. Immunol.
36
:
1374
-1385.
34
Polanski, M., N. S. Melican, J. Zhang, H. L. Weiner.
1997
. Oral administration of the immunodominant B-chain of insulin reduces diabetes in a co-transfer model of diabetes in the NOD mouse and is associated with a switch from Th1 to Th2 cytokines.
J. Autoimmun.
10
:
339
-346.
35
Hartmann, B., K. Bellmann, I. Ghiea, R. Kleemann, H. Kolb.
1997
. Oral insulin for diabetes prevention in the NOD mice: potentiation by enhancing Th2 cytokine expression in the gut through bacterial adjuvant.
Diabetologia
40
:
902
-909.
36
Miller, R G., Jr.
1966
.
Simultaneous Statistical Inference
McGraw-Hill, New York.
37
Baksh, S., H. R. Widlund, A. A. Frazer-Abel, J. D. Du, S. Fosmire, D. E. Fisher, J. A. DeCaprio, J. F. Modiano, S. J. Burakoff.
2002
. NFATc2-mediated repression of cyclin-dependent kinase 4 expression.
Mol. Cell.
10
:
1071
-1081.
38
Tanguay, D. A., T. C. Cjiles.
1996
. Regulation of the catalytic subunit (p34PSK-J3/Cdk4) for the major D-type cyclin in mature B lymphocytes.
J. Immunol.
156
:
539
-548.
39
Kovalev, G. I., D. S. Franklin, V. M. Coffield, Y. Xiong, L. Su.
2001
. An important role of CDK inhibitor p18(INK4c) in modulating antigen receptor-mediated T cell proliferation.
J. Immunol.
167
:
3285
-3292.
40
Shin, H. H., S. J. Kim, H. S. Lee, H. S. Choi.
2004
. The soluble glucocorticoid-induced tumor necrosis factor receptor causes cell cycle arrest and apoptosis in murine macrophages.
Biochem. Biophys. Res. Commun.
316
:
24
-32.
41
Daniel, D., D. R. Wegmann.
1996
. Protection of nonobese diabetic mice from diabetes by intranasal or subcutaneous administration of insulin peptide B-(9–23).
Proc. Natl. Acad. Sci. USA
93
:
956
-960.
42
Zekzer, D., F. S. Wong, O. Ayalon, I. Millet, M. Altieri, S. Shitani, M. Solimena, R. S. Sherwin.
1998
. GAD-reactive CD4+ Th1 cells induce diabetes in NOD/SCID mice.
J. Clin. Invest.
101
:
68
-73.
43
Ryan, E. A., B. W. Paty, P. A. Senior, D. Bigam, E. Alfadhli, N. M. Kneteman, J. R. Lakey, A. M. Shapiro.
2005
. Five-year-follow-up after clinical islet transplantation.
Diabetes
54
:
2060
-2069.
44
Fitzpatrick, F., F. Lepault, F. Homo-Delarche, J. F. Bach, M. Dardenne.
1991
. Influence of castration, alone or combined with thymectomy, on the development of diabetes in the nonobese diabetic mouse.
Endocrinology
129
:
1382
-1390.
45
Casteels, K. M., C. A. Gysemans, M. Waer, R. Bouillon, J. M. Laureys, J. Depovere, C. Mathieu.
1998
. Sex difference in resistance to dexamethasone-induced apoptosis in NOD mice: treatment with 1,25(OH)2D3 restores defect.
Diabetes
47
:
1033
-1037.
46
Boissonnas, A., B. Combadiere.
2004
. Interplay between cell division and cell death during TCR triggering.
Eur. J. Immunol.
34
:
2430
-2438.
47
Modiano, J. F., J. Mayor, C. Ball, M. K. Fuentes, D. S. Linthicum.
2000
. CDK4 expression and activity are required for cytokine responsiveness in T cells.
J. Immunol.
165
:
6693
-6702.
48
Karas, M., T. Z. Zaks, J. L. Liu, D. LeRoith.
1999
. T cell receptor-induced activation and apoptosis in cycling human T cells occur throughout the cell cycle.
Mol. Biol. Cell
10
:
4441
-4450.
49
Ritzel, R. A., P. C. Butler.
2003
. Replication increases β-cell vulnerability to human islet amyloid polypeptide-induced apoptosis.
Diabetes
52
:
1701
-1708.