T cell activation is an energy-demanding process fueled by increased glucose consumption and accompanied by upregulation of the insulin receptor (INSR). In this article, we report that silencing the INSR in inducible knockdown rats impairs selective T cell functions but not thymocyte development. Glucose transport and glycolysis in activated CD4+ T cells were compromised in the absence of the INSR, which was associated with alterations in intracellular signaling pathways. The observed metabolic defects coincided with reduced cytokine production, proliferation, and migration, as well as increased apoptosis of CD4+ T cells. The cytotoxicity of CD8+ T cells in response to alloantigens was also diminished under these conditions, whereas the frequency and suppressive capacity of regulatory T cells were unaffected. The observed impairments proved to be decisive in vivo because silencing of the INSR attenuated clinical symptoms in animal models of acute graft-versus-host disease and multiple sclerosis. Taken together, our results suggest that upregulation of the INSR on T cells following activation is required for efficient adaptive immunity.

Resting T cells need little energy for normal function and survival. However, once they encounter their cognate Ag, they rapidly start to proliferate and differentiate into effector T cells. This is an energy-demanding process that involves migration to the site of inflammation and the production of cytokines and cytotoxic mediators. Hence, these changes are accompanied by metabolic reprogramming (13). Although resting T cells cover 96% of their energy demand by oxidative phosphorylation (OXPHOS), they switch to aerobic glycolysis upon activation (4). This phenomenon is known as the Warburg effect (5, 6). Consequently, glucose uptake has to be increased, which is presumably facilitated by the induction of the insulin receptor (INSR) (7). Hence it is conceivable that the INSR plays a critical role for T cell function and adaptive immunity (8, 9).

Activated T cells have to dispose of large amounts of ATP, which is accomplished by switching from OXPHOS to aerobic glycolysis, despite being a relatively inefficient metabolic pathway (10, 11). To meet the concomitant high glucose demand, the expression of transporter proteins and their surface localization are increased. Following T cell activation by APCs, CD28 signaling leads to the phosphorylation of Akt, which results in an upregulation of glucose transporter (Glut)1 and an increased activity of several glycolytic enzymes (12, 13). The importance of these changes was demonstrated by overexpressing Glut1 in mice, which led to increased proliferation and cytokine production by effector T cells (12, 14). Conversely, impaired glucose metabolism reduced T cell activation and function (11, 1416).

Glucose uptake is not equally important for all T cell subsets. Effector T cells, such as Th1 cells, Th17 cells, and CTLs, strongly increase glycolysis after activation (1719). In contrast, regulatory T cells (Tregs) are largely independent of glucose because they primarily use fatty acid oxidation for energy production (3). Consequently, Treg function was unaffected by Glut1 overexpression (14). Although it is well established that increased Glut1 surface levels after CD28 stimulation are important for the switch to aerobic glycolysis (8), a potential role for Glut3 and Glut4 in T cells and their regulation by the INSR are less clear. Some investigators reported Glut1 and Glut3, but not Glut4, expression in T cells, whereas others also found that Glut4 was expressed in activated T cells and even was augmented by insulin (20, 21). Notably, CD28 and the INSR are able to activate PI3K, which leads to the membrane recruitment of Akt and, eventually, stimulates glucose transport. Because both pathways use the same signaling molecules, it is conceivable that they act in a synergistic manner to ensure optimal glucose supply in T cells (8).

Most findings concerning the energy metabolism of lymphocytes were made in vitro, because Ag-specific T cells in inflammatory diseases are often rare and hard to separate from bystander cells. Notwithstanding these complications, the metabolic switch of T cells after activation could also be demonstrated in vivo in a mouse model of graft-versus-host disease (GvHD), which is a severe complication of hematopoietic stem cell transplantation (22). Allogeneic T cells become activated by MHC-mismatched host cells, leading to an inflammatory response that mainly damages the intestinal tract, liver, and skin. In the course of this process, effector T cells were found to switch from OXPHOS to aerobic glycolysis, which was accompanied by enhanced glucose uptake in the intestinal tract of mice and humans (23, 24). Insight into the metabolic requirements of T cells was also obtained in multiple sclerosis (MS), a neuroinflammatory disease leading to demyelination, axonal damage, and serious paralytic symptoms. Glucose availability in MS seems to matter even more than in GvHD, because the former is predominantly mediated by Th17 cells, which exclusively rely on glycolysis (3). Inhibition of this pathway in experimental autoimmune encephalomyelitis (EAE), a widely used animal model of MS, resulted in a reduced production of IL-17 and an amelioration of disease symptoms (25, 26). Taken together, elevated glucose consumption is a typical feature of inflammatory diseases and is required for full pathogenicity of activated T cells, highlighting a potential role for the INSR in adaptive immunity.

In this study, we took advantage of inducible INSR-knockdown rats, which express a specific small interfering RNA upon treatment with doxycycline (Dox), leading to a complete loss of INSR protein in most organs (27). As a consequence, glucose metabolism starts to derail, resulting in the development of type 2 diabetes. Although the rats eventually die from massive hyperglycemia, their T cells and hematopoietic stem cells can be studied in vitro or used for adoptive-transfer experiments in vivo. Taking this approach, we identified a crucial role for the INSR in T cell function and adaptive immunity.

Wild-type (wt) Lewis (LEW) and Dark Agouti (DA) rats were obtained from Harlan Laboratories (Rossdorf, Germany) or bred in our own animal facility. Inducible INSR-knockdown LEW rats (LEW.TG04(FH1t(INSR)UTG)/HRJB) and eGFP-transgenic LEW rats (LEW.TG3(FUGW)/HRJB), which served as controls, were reported previously (27, 28). Male and female rats were used at 8–14 wk of age; Dox was administered via food pellets containing 735 mg/kg doxycycline hyclate (ssniff Spezialdiaeten, Soest, Germany). Glucose levels were measured using an Ascensia Blood Glucose Meter CONTOUR. All experiments were conducted according to Lower Saxony state regulations and were approved by the responsible authorities (Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit).

Cell lysates were separated on 7.5% SDS-PAGE gels (20 μg protein per lane for tissue lysates; 2.5 × 105 cells per lane for cultured cells). After transfer to a nitrocellulose membrane, proteins were stained with primary Abs recognizing the INSR or ERK as a loading control (Santa Cruz Biotechnology, Heidelberg, Germany). Visualization was achieved by incubation with an HRP-conjugated secondary Ab in combination with ECL detection reagent (Roth, Karlsruhe, Germany) and a Chemilux Imager (Intas, Göttingen, Germany). Quantification was achieved using Gel-Pro Analyzer software (Media Cybernetics, Cambridge, U.K.).

Lymphocytes were stained with the following Abs (BD Biosciences, Heidelberg, Germany), as previously described (29): CD4 (W3-25), CD8α (Ox8), TCRβ (R73), CD25 (Ox39), and CD134 (Ox40). Nuclear staining of Foxp3 (eBioscience, Frankfurt, Germany) and intracellular staining of Ki-67 and phospho-Stat3 (pY705) (BD Biosciences) were performed following the manufacturers’ instructions. Samples were analyzed using a FACSCanto II (BD Biosciences) and FlowJo software (TreeStar, Ashland, OR).

Female INSR-knockdown rats were mated with male wt LEW rats to obtain fetuses of both genotypes. Pregnant rats were sacrificed at embryonic day 16.5, and the embryos were collected into PBS. Genotyping was performed by FACS analysis of eGFP in fetal livers, as described (30). Thymic lobes were dissected under a Leica M651 microscope and cleaned from nonthymic tissue debris using a Leica MS5 dissection stereomicroscope. The lobes were cultured on Nucleopore membranes floating on RPMI 1640 medium in 12-well flat-bottom plates, as described (30). Dox was administered at a concentration of 2 μg/ml, and the medium was changed every fourth day during the 12-d culture period. Several lobes from each genotype were combined, dispersed in FACS buffer with a syringe, centrifuged through a 40-μm nylon mesh, and analyzed by FACS.

CD4+ T cells were isolated from the lymph nodes by magnetic cell sorting with an autoMACS device (Miltenyi Biotec, Bergisch-Gladbach, Germany). In the case of costimulation, cell culture plates were coated with sheep anti-mouse–Ig (2.6 μg/ml) overnight at 4°C, washed with PBS, and incubated for 2 h with 4 μg/ml anti-TCR Ab R73 and 2 μg/ml anti-CD28 Ab JJ319 (BD Biosciences) in 500 μl of PBS/BSA at 37°C. T cells were added to RPMI 1640 medium with 10% FCS (1 × 106 cells/1 ml) and 2 μg/ml Dox and incubated for up to 3 d at 37°C and 5% CO2.

Glucose uptake was determined with the help of a fluorescently labeled deoxyglucose analog (2-NBDG) using a commercially available kit (Cayman Chemical, Ann Arbor, MI). Purified CD4+ T cells were stimulated with anti-TCR/CD28 Abs for 1 d, as described above, and washed twice with glucose-free medium. Cells were resuspended in 100 μl of glucose-free medium supplemented with 150 μg/ml 2-NBDG and incubated at 37°C for 5 h. The cells were washed twice with assay buffer, and glucose uptake was quantified by FACS analysis.

Total RNA was isolated using a Quick-RNA MiniPrep Kit (Zymo Research, Irvine, CA) or an RNeasy Mini Kit (QIAGEN, Hilden, Germany), and cDNA was prepared with an iScript Kit (Bio-Rad, Munich, Germany). Quantitative RT-PCR was performed on an ABI 7500 using SYBR MasterMix (both from Applied Biosystems, Darmstadt, Germany). Results were normalized to mRNA expression of β-actin and evaluated using the ΔΔ threshold cycle method. The following primer sequences were used: subunit A of lactate dehydrogenase (LDH), 5′-CGTGCACTAAGCGGTCCCA-3′ (forward) and 5′-GTTCTGGGGGACCTGTTCTTC-3′ (reverse); β-actin, 5′-AGCTCCTCCGTCGCCGGTC-3′ (forward) and 5′-CCACCATCACACCCTGGTGCCT-3′ (reverse); Glut1 (slc2a1), 5′-GAGTGTACTGTGGCCTGA-3′ (forward) and 5′-GTCTAAGCCGAACACCTG-3′ (reverse); Glut3 (slc2a3), 5′-TTTGGCAGACGCAACTCCAT-3′ (forward) and 5′-GATGCCAATAATCAGGCGGC-3′ (reverse); Glut4 (slc2a4), 5′-GGTTGGGAAGGAAAAGGG-3′ (forward) and 5′-AGTAGGCGCCAATGAGGA-3′ (reverse); leptin receptor (LEPR), 5′-CCAGTACCCAGAGCCAAAGT-3′ (forward) and 5′-GGGCTTCACAACAAGCATGG-3′ (reverse); and IFN-γ, 5′-ACGCCGCGTCTTGGTTTTGC-3′ (forward) and 5′-TACCGTCCTTTTGCCAGTTCCTCCA-3′ (reverse).

The rate of glycolysis was assessed based on the release of l-lactate into the medium. CD4+ T cells were stimulated with anti-TCR/CD28 Abs for 1 d, as described above, the supernatant was collected, and the amount of lactate was measured by a fluorescence-based method using a commercially available kit (Cayman Chemical). In brief, the cell culture supernatant was deproteinated with metaphosphoric acid and neutralized, and the salts were precipitated. The samples were diluted and placed in black 96-well microtiter plates. Lactate was subsequently converted to pyruvate in an enzymatic reaction, concomitantly yielding a highly fluorescent product detected with an excitation wavelength of 530–540 nm and an emission wavelength of 585–595 nm using an Infinite 200 Pro reader (Tecan, Männedorf, Switzerland).

CD4+ T cells from rats of both genotypes (n = 4) were stimulated in triplicates in 96-well plates with anti-TCR/CD28 Abs for 1 d and then lysed using Protein Extraction Buffer (Full Moon BioSystems, Sunnyvale, CA). Differential activation of INSR-deficient T cells was confirmed by FACS analysis of CD25 and CD134. A total of 100 μg of total protein per sample was used to hybridize the PIG219 Ab Array (Full Moon Biosystems) covering major signaling molecules of the INSR and mammalian target of rapamycin (mTOR) pathways. Labeling, detection, and array scanning were performed by BioCat (Heidelberg, Germany), according to standard protocols. Data were normalized to the average signal intensity on each array and are depicted as fold change between samples of both genotypes.

IFN-γ, IL-4, and IL-10 levels in cell culture supernatants were measured by cytokine bead array, according to the manufacturer’s instructions (BD Biosciences). In brief, 20 μl of cell culture supernatant was incubated with the cytokine-specific beads for 1 h, followed by the addition of the detection Abs and incubation for another hour. Analysis was performed using a FACSCanto II and FCAP Array v3.0 software (both from BD Biosciences).

A Boyden chamber assay was performed as described previously (31). CD4+ T cells were stimulated with anti-TCR/CD28 Abs for 1 d, as outlined above, seeded in the upper part of a 5-μm Transwell insert (1 × 106 cells), and allowed to migrate spontaneously into the lower chamber for 3 h. The cells in the lower part were counted, and the migration relative to the input was calculated, as previously described (31). The relative migration rate of control T cells was set as 1.

CD4+ T cells were stimulated with anti-TCR/CD28 Abs for 1 d, as described. To study apoptosis induced by serum deprivation, T cells were cultured for another day in RPMI 1640 medium with only 1% FCS. To analyze activation-induced cell death (AICD), T cells were washed and incubated for another day with 4 μg/ml of the anti-TCR Ab R73 but without adding anti-CD28 Ab JJ319. In both assays, Dox was present during the entire culture period. Apoptotic cells were detected by staining with annexin V (AxV)–Cy5 in combination with 7-aminoactinomycin D (7-AAD; both from BD Biosciences), followed by FACS analysis.

CD4+ or CD8+ T cells were isolated from the lymph nodes of rats on a LEW background by magnetic cell sorting (Miltenyi Biotec), and the concentration was adjusted to 2.5 × 106 cells per milliliter. These cells were mixed in a 4:1 ratio with irradiated allogeneic splenocytes isolated from DA rats (30 Gy) and cocultured for up to 3 d. Proliferation was determined by [3H] thymidine incorporation assay (32), and gene expression was analyzed by quantitative RT-PCR.

Rats on a LEW background were immunized with allogeneic splenocytes isolated from DA rats by injecting 1.5 × 107 cells in a volume of 50 μl slightly above the footpad of the hind limbs. Five days later, Dox treatment was initiated. After 10 d, the draining lymph nodes were isolated, and the lymphocytes were restimulated at a 1:50 ratio with irradiated allogeneic splenocytes (30 Gy) from DA rats in DMEM medium containing 10% FCS, 1% Penicillin/Streptomycin, 20% ConA supernatant, and 10 ng/ml mouse IL-2 (PeproTech, Hamburg, Germany). In parallel, lymphocyte blasts serving as target cells were generated by culturing splenocytes from DA rats in the presence of 2.5 μg/ml ConA and 2 μg/ml mouse IL-2. After 5 d, the cells were harvested and mixed with the restimulated lymphocytes. The specific lysis by CTLs was determined using a [51Cr]-release assay, as described (33).

CD4+ CD25 indicator T cells were isolated by magnetic cell sorting and cultured with irradiated splenocytes (30 Gy) in the presence of 2 μg/ml ConA (Sigma, Taufkirchen, Germany). CD4+CD25+ Tregs were isolated from the lymph nodes of rats that were pretreated with 1 mg of the superagonistic anti-CD28 Ab JJ316 (3 d) and Dox (5 d) and added to the indicator cells at different ratios, as described previously (34). Indicator cells that were cultured individually served as a control. Proliferation was assessed by [3H]-thymidine incorporation assay (32).

The procedure used for disease induction was adapted from previously published protocols (35, 36). wt DA rats were sublethally irradiated (12.5 Gy); 1 d later, they were injected i.v. with 3 × 107 bone marrow cells and 4.5 × 106 total lymph node cells (corresponding to ∼3 × 106 T cells) isolated from Dox-treated transgenic or control rats on a LEW background. Clinical symptoms were scored on a daily basis by an investigator who was blinded to the genotype of the rats. This was accomplished using Cooke’s score (37), which is based on five parameters (body weight, activity, fur, skin, posture), each of which was assigned a value between 0 and 2 and subsequently added up to a maximal score of 10. Animals were treated with pain-reducing medication and euthanized for ethical reasons at a score of 7 or a weight loss > 20%. Deceased animals were assigned a score of 10. All rats received Dox-containing food during the entire experiment. In one experiment, rats were sacrificed at day 10 after GvHD induction, and the splenocytes were analyzed by FACS and quantitative RT-PCR.

T cells were depleted in wt LEW rats by i.v. injection of depleting Abs specific for CD8 (Ox8, 1 mg/kg) and CD4 (Ox38, 10 mg/kg). Two days later, the recipient rats were sublethally irradiated (12.5 Gy) and received 1 × 107 bone marrow cells isolated from transgenic or control rats the following day. Reconstitution of the immune system was tested by FACS analysis and found to be >90% in all cases.

Transgenic and control rats were immunized with guinea pig myelin oligodendrocyte glycoprotein (gpMBP) by s.c. injection of a paste composed of 2 mg/ml gpMBP emulsified in 2 mg/ml CFA (1:1 ratio) slightly above the footpad of either hind limb (50 μl each). In the case of active EAE, all rats received Dox-containing food starting 5 d prior to disease induction and were weighed and scored daily according to a 10-grade scale to assess the degree of paralytic symptoms for 30 d, as previously described (38). To generate encephalitogenic T cells, the draining lymph nodes of transgenic rats were isolated at day 10, followed by repeated restimulation and expansion in vitro (38). Adoptive-transfer EAE (AT-EAE) was induced by i.v. injection of 4 × 106 encephalitogenic T cells, which were cultured in the presence or absence of Dox, for 6 d into 10–14-wk-old wt LEW rats, followed by daily weighing and scoring for 15 d. During AT-EAE, the rats received Dox-containing food or a standard diet.

Data were analyzed using Prism software (GraphPad, San Diego, CA). The unpaired t test was used in most cases, with the exception of the cytotoxicity assay, for which two-way ANOVA was used; the Kaplan–Meier survival curve, for which the log-rank Mantel–Cox test was used; and GvHD and EAE disease courses, for which the Kruskal–Wallis test was used. All data are shown as mean ± SEM.

Thymocytes constitutively express the INSR and respond to insulin exposure with increased nutrient uptake (39). Because thymocytes are highly proliferative cells, we wondered whether absence of the INSR impaired T cell development. To address this question, immunologically naive inducible INSR-knockdown (hereafter referred to as transgenic) and control rats were treated with Dox for 5 d. This led to an increased blood glucose level in transgenic rats because of the insulin insensitivity of metabolic organs (27), as well as an almost complete ablation of the INSR in the liver, collectively indicating that the knockdown was successful (Fig. 1A, 1B). However, thymocyte composition in transgenic rats was unaltered compared with control rats, although the INSR was highly expressed in the wt thymus and decreased to almost undetectable levels in transgenic rats following Dox treatment (Fig. 1B, 1C). To study T cell development in a more dynamic manner, we performed fetal thymus organ culture (FTOC). Thymi composed exclusively of double-negative (DN) cells were dissected from embryonic day 16.5 embryos of both genotypes and cultured in the presence of Dox for 12 d. This procedure resulted in an almost normal thymocyte repertoire composed primarily of double-positive (DP) and single-positive (SP) cells, regardless of the presence of the INSR (Fig. 1D). We conclude that INSR signaling is dispensable for thymic T cell development.

FIGURE 1.

Silencing of the INSR causes severe hyperglycemia but has no impact on thymocyte development. (A) Transgenic (tg) and control (wt) rats received Dox-containing food pellets for 5 d. Blood glucose levels were determined daily by tail tip puncture and are depicted as the mean ± SEM (n = 5). (B) Transgenic and control rats were fed Dox-containing food for 5 d, followed by dissection of liver and thymus. INSR protein levels were analyzed by Western blot; ERK expression served as a loading control. The result of an representative experiment (of three) is shown for lysates from each organ of two rats per genotype. (C) FACS analysis of thymocytes from 8-wk-old transgenic and control rats that received a Dox-containing diet for 5 d. Mean (± SEM) percentages of CD4CD8 (DN), CD4+CD8+ (DP), CD4+CD8 (CD4 SP), and CD4CD8+ (CD8 SP) thymocytes are depicted for each genotype (n = 4). (D) FACS analysis of FTOCs from transgenic and control embryos after 12 d of culture in the presence of Dox. Depicted are the mean percentages ± SEM of DN, DP, CD4 SP, and CD8 SP thymocytes (n = 3). Statistical analysis was performed with an unpaired t test and refers to the comparison between genotypes. n.s., not significant.

FIGURE 1.

Silencing of the INSR causes severe hyperglycemia but has no impact on thymocyte development. (A) Transgenic (tg) and control (wt) rats received Dox-containing food pellets for 5 d. Blood glucose levels were determined daily by tail tip puncture and are depicted as the mean ± SEM (n = 5). (B) Transgenic and control rats were fed Dox-containing food for 5 d, followed by dissection of liver and thymus. INSR protein levels were analyzed by Western blot; ERK expression served as a loading control. The result of an representative experiment (of three) is shown for lysates from each organ of two rats per genotype. (C) FACS analysis of thymocytes from 8-wk-old transgenic and control rats that received a Dox-containing diet for 5 d. Mean (± SEM) percentages of CD4CD8 (DN), CD4+CD8+ (DP), CD4+CD8 (CD4 SP), and CD4CD8+ (CD8 SP) thymocytes are depicted for each genotype (n = 4). (D) FACS analysis of FTOCs from transgenic and control embryos after 12 d of culture in the presence of Dox. Depicted are the mean percentages ± SEM of DN, DP, CD4 SP, and CD8 SP thymocytes (n = 3). Statistical analysis was performed with an unpaired t test and refers to the comparison between genotypes. n.s., not significant.

Close modal

Quiescent T cells do not express the INSR but start to upregulate it briefly after stimulation (40). However, its function in T cell activation was unknown. To address this issue, we isolated CD4+ T cells from rats that had received Dox for 5 d and stimulated them with anti-TCR/CD28 Abs in the continuous presence of Dox. Notably, the hyperglycemia that develops in transgenic rats after Dox treatment in vivo (Fig. 1A) had no impact on the numbers or composition of circulating T cells (data not shown), because they are mostly naive and do not express the INSR or high levels of Gluts. However, when activated in vitro, protein levels of the INSR were strongly increased in CD4+ T cells from control rats within 3 d, whereas the INSR in cells from transgenic rats remained barely detectable (Fig. 2A, 2B). To determine whether absence of the INSR led to an impaired energy metabolism, CD4+ T cells were isolated from Dox-treated rats, stimulated with anti-TCR/CD28 Abs in vitro for 1 d, and analyzed for the expression of Gluts. Glut1, Glut3, and Glut4 mRNA levels were strongly increased after T cell activation (Fig. 2C). However, upregulation of Glut3 and Glut4 was less pronounced when the INSR was silenced, which was not the case for Glut1 (Fig. 2C). To explore functional consequences of the altered gene expression, activated CD4+ T cells were incubated with the fluorescent nondegradable glucose-analog 2-NBDG and analyzed by FACS. In accordance with the impaired upregulation of Glut3 and Glut4, glucose uptake was significantly lower in the absence of the INSR (Fig. 2D). In addition, the synthesis of lactate as a measure of glycolysis was diminished after silencing the INSR, and the mRNA levels of LDH were also reduced (Fig. 2E, 2F). These observations indicate that INSR deficiency leads to an inhibition of aerobic glycolysis caused by a shortage of glucose and a direct impact on the glycolytic pathway. Therefore, CD4+ T cells appear to depend on the INSR to tap the full potential of their glycolytic capacity.

FIGURE 2.

Glucose uptake and glycolysis during T cell activation are impaired in the absence of the INSR. CD4+ T cells were purified from the lymph nodes of transgenic (tg) and control (wt) rats that received Dox-containing food pellets for 5 d and stimulated with anti-TCR/CD28 Abs in the continuous presence of Dox. (A) Protein lysates were prepared before or after costimulation for up to 3 d. INSR protein levels were determined by Western blot analysis; ERK expression served as loading control (one representative analysis of four is depicted). (B) The quantification of INSR protein levels depicted as mean ± SEM refers to experiments as in (A). Values at day 0 were arbitrarily set to 1 (n = 4). (C) Quantitative RT-PCR was used to determine mRNA levels of Glut1, Glut3, and Glut4 before and 1 d after costimulation. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples obtained on day 0 were arbitrarily set to 1 (n = 4). (D) CD4+ T cells were costimulated for 1 d, followed by incubation with 2-NBDG for 5 h. Glucose uptake was quantified by FACS analysis and is depicted as the mean fluorescence intensity (MFI) ± SEM of the cells (n = 12/9 for wt/tg). (E) CD4+ T cells were costimulated for 1 d, and the mean concentration ± SEM of lactate was measured in the cell culture supernatant (n = 6). (F) CD4+ T cells were harvested after 1 d of costimulation, and mRNA levels of LDH were analyzed by quantitative RT-PCR. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples were arbitrarily set to 1 (n = 5/6 for wt/tg). Statistical analysis in all panels was performed by the unpaired t test and refers to the comparison between genotypes. *p < 0.05, **p < 0.01, ***p < 0.001. n.s., not significant.

FIGURE 2.

Glucose uptake and glycolysis during T cell activation are impaired in the absence of the INSR. CD4+ T cells were purified from the lymph nodes of transgenic (tg) and control (wt) rats that received Dox-containing food pellets for 5 d and stimulated with anti-TCR/CD28 Abs in the continuous presence of Dox. (A) Protein lysates were prepared before or after costimulation for up to 3 d. INSR protein levels were determined by Western blot analysis; ERK expression served as loading control (one representative analysis of four is depicted). (B) The quantification of INSR protein levels depicted as mean ± SEM refers to experiments as in (A). Values at day 0 were arbitrarily set to 1 (n = 4). (C) Quantitative RT-PCR was used to determine mRNA levels of Glut1, Glut3, and Glut4 before and 1 d after costimulation. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples obtained on day 0 were arbitrarily set to 1 (n = 4). (D) CD4+ T cells were costimulated for 1 d, followed by incubation with 2-NBDG for 5 h. Glucose uptake was quantified by FACS analysis and is depicted as the mean fluorescence intensity (MFI) ± SEM of the cells (n = 12/9 for wt/tg). (E) CD4+ T cells were costimulated for 1 d, and the mean concentration ± SEM of lactate was measured in the cell culture supernatant (n = 6). (F) CD4+ T cells were harvested after 1 d of costimulation, and mRNA levels of LDH were analyzed by quantitative RT-PCR. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples were arbitrarily set to 1 (n = 5/6 for wt/tg). Statistical analysis in all panels was performed by the unpaired t test and refers to the comparison between genotypes. *p < 0.05, **p < 0.01, ***p < 0.001. n.s., not significant.

Close modal

The INSR is part of a complex signaling network that includes LEPR and mTOR. Therefore, we investigated whether silencing of the INSR had an impact on signaling pathways controlling metabolic activity and immunological functions in CD4+ T cells stimulated for 1 d with anti-TCR/CD28 Abs. Although insulin and leptin signaling closely cooperate in the regulation of energy metabolism, mRNA levels of LEPR and Stat3 phosphorylation were not altered in the absence of the INSR (Fig. 3). In contrast, several signaling proteins of pathways downstream of TCR/CD28 or INSR/LEPR, including the RAS/ERK, PI3K/Akt, and mTOR pathways, were downregulated when the INSR was silenced (Table I). Moreover, the abundance of the proapoptotic protein BAD, which is a target of PI3K/Akt signaling, was increased in INSR-deficient CD4+ T cells (Table I). The overall changes in protein levels were only 10–20%; however, this is in the same range as many other metabolic and immunological differences observed between T cells of both genotypes. Taken together, silencing of the INSR resulted in a moderate blunting of downstream, but not parallel, signaling pathways.

FIGURE 3.

The LEPR/Stat3 signaling pathway is unaffected by INSR silencing. CD4+ T cells were purified from the lymph nodes of transgenic (tg) and control (wt) rats that received Dox-containing food pellets for 5 d and were subsequently stimulated for 1 d with anti-TCR/CD28 Abs in the continuous presence of Dox. (A) Quantitative RT-PCR was used to determine mRNA levels of LEPR. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples were arbitrarily set to 1 (n = 4). (B) The mean fluorescence intensity (MFI) ± SEM of intracellular phospho-Stat3 (pY705) in CD4+ T cells was determined by FACS analysis. n.s., not significant.

FIGURE 3.

The LEPR/Stat3 signaling pathway is unaffected by INSR silencing. CD4+ T cells were purified from the lymph nodes of transgenic (tg) and control (wt) rats that received Dox-containing food pellets for 5 d and were subsequently stimulated for 1 d with anti-TCR/CD28 Abs in the continuous presence of Dox. (A) Quantitative RT-PCR was used to determine mRNA levels of LEPR. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples were arbitrarily set to 1 (n = 4). (B) The mean fluorescence intensity (MFI) ± SEM of intracellular phospho-Stat3 (pY705) in CD4+ T cells was determined by FACS analysis. n.s., not significant.

Close modal
Table I.
Impact of INSR ablation on proteins of relevant signaling pathways
PathwayProteinDetection AbProtein Level in tg/Dox Versus wt/DoxFold Change
INSR, TCR RAS p21 H and K Down 0.90 
 MEK1 Ab-286 Down 0.81 
 ERK1 (p44/42) Ab-202 Down 0.86 
INSR, LEPR, IRS-1 Ab-307 Down 0.88 
CD28 PI3K (p85α/γ) Ab-467/199 Unaltered 1.05 
 PTEN Ab-370 Down 0.85 
 PDK1 Ab-241 Down 0.82 
 Akt Ab-473 Down 0.82 
 BAD Ab-136 Up 1.19 
 PKCζ Ab-410 Down 0.80 
mTOR mTOR Ab-2481 Unaltered 0.96 
 4E-BP1 Ab-65 Up 1.20 
 eIF4E Ab-209 Down 0.88 
 p70S6K Ab-411 Down 0.81 
PathwayProteinDetection AbProtein Level in tg/Dox Versus wt/DoxFold Change
INSR, TCR RAS p21 H and K Down 0.90 
 MEK1 Ab-286 Down 0.81 
 ERK1 (p44/42) Ab-202 Down 0.86 
INSR, LEPR, IRS-1 Ab-307 Down 0.88 
CD28 PI3K (p85α/γ) Ab-467/199 Unaltered 1.05 
 PTEN Ab-370 Down 0.85 
 PDK1 Ab-241 Down 0.82 
 Akt Ab-473 Down 0.82 
 BAD Ab-136 Up 1.19 
 PKCζ Ab-410 Down 0.80 
mTOR mTOR Ab-2481 Unaltered 0.96 
 4E-BP1 Ab-65 Up 1.20 
 eIF4E Ab-209 Down 0.88 
 p70S6K Ab-411 Down 0.81 

CD4+ T cells from transgenic rats treated with Dox (tg/Dox) were compared with CD4+ T cells from wt rats treated with Dox (wt/Dox). The analysis was performed on day 1 after T cell stimulation with anti-TCR/CD28 Abs in the continuous presence of Dox using a commercial Ab array.

To uncover whether the impaired energy metabolism of INSR-deficient T cells had an impact on their activation, we isolated CD4+ T cells from the lymph nodes of Dox-treated transgenic and control rats and stimulated them in vitro with anti-TCR/CD28 Abs in the continuous presence of Dox. Abundance of the CD25+ CD134 subpopulation, which represents an early stage of T cell activation in rats, was reduced in cultures of transgenic CD4+ T cells compared with controls (Fig. 4A). In addition, IFN-γ, IL-4, and IL-10 levels in the supernatant were lower at all time points (Fig. 4B–D). Interestingly, the INSR does not seem to impact T cell polarization, because Th1 and Th2 cytokines were affected in a similar manner. When we measured proliferation and migration of activated CD4+ T cells 1 d after stimulation, we found that transgenic cells were significantly less active (Fig. 4E). In parallel, the sensitivity to apoptosis induction by serum deprivation (1% FCS) or AICD was increased (Fig. 4F). We conclude that silencing of the INSR affects T cell activation, as well as T cell function and survival.

FIGURE 4.

INSR expression in CD4+ T cells is required for full activity. CD4+ T cells were purified from the lymph nodes of transgenic (tg) and control (wt) rats that received Dox-containing food pellets for 5 d and were stimulated with anti-TCR/CD28 Abs in the continuous presence of Dox. (A) CD4+ T cells were analyzed before or after costimulation for up to 3 d, and the mean percentage ± SEM of CD25+CD134 cells was determined by FACS (n = 8). CD4+ T cells were costimulated for up to 3 d, followed by analysis of IFN-γ (B), IL-4 (C), and IL-10 (D) levels in the cell culture supernatant by cytokine bead array. Cytokine concentrations are depicted as the mean ± SEM (n = 4–8). (E) Analysis of proliferation and migration of CD4+ T cells costimulated for 1 d. Proliferation was determined by [3H]thymidine-incorporation assay and is depicted as the fold change of cpm ± SEM normalized to wt samples (left, n = 12). Migration was assessed using a Boyden chamber assay. For this, the cells were allowed to migrate spontaneously for 3 h, and the amount of cells in the lower chamber was quantified by FACS. The relative migration ± SEM is depicted; wt samples were arbitrarily set to 1 (right, n = 4–6). (F) CD4+ T cells were costimulated for 1 d. Subsequently, they were subjected to serum deprivation with 1% FCS (left, n = 4) or AICD (right, n = 8) for another day. The percentage of apoptotic cells, defined as AxV+7-AAD+, was determined by FACS analysis. *p < 0.05, **p < 0.01, ***p < 0.001, unpaired t test, refers to the comparison between genotypes. n.s., not significant.

FIGURE 4.

INSR expression in CD4+ T cells is required for full activity. CD4+ T cells were purified from the lymph nodes of transgenic (tg) and control (wt) rats that received Dox-containing food pellets for 5 d and were stimulated with anti-TCR/CD28 Abs in the continuous presence of Dox. (A) CD4+ T cells were analyzed before or after costimulation for up to 3 d, and the mean percentage ± SEM of CD25+CD134 cells was determined by FACS (n = 8). CD4+ T cells were costimulated for up to 3 d, followed by analysis of IFN-γ (B), IL-4 (C), and IL-10 (D) levels in the cell culture supernatant by cytokine bead array. Cytokine concentrations are depicted as the mean ± SEM (n = 4–8). (E) Analysis of proliferation and migration of CD4+ T cells costimulated for 1 d. Proliferation was determined by [3H]thymidine-incorporation assay and is depicted as the fold change of cpm ± SEM normalized to wt samples (left, n = 12). Migration was assessed using a Boyden chamber assay. For this, the cells were allowed to migrate spontaneously for 3 h, and the amount of cells in the lower chamber was quantified by FACS. The relative migration ± SEM is depicted; wt samples were arbitrarily set to 1 (right, n = 4–6). (F) CD4+ T cells were costimulated for 1 d. Subsequently, they were subjected to serum deprivation with 1% FCS (left, n = 4) or AICD (right, n = 8) for another day. The percentage of apoptotic cells, defined as AxV+7-AAD+, was determined by FACS analysis. *p < 0.05, **p < 0.01, ***p < 0.001, unpaired t test, refers to the comparison between genotypes. n.s., not significant.

Close modal

Individual T cell subpopulations differ in their metabolic features. Hence, it was conceivable that they also have varying requirements for the INSR. To address this issue, we initially performed an MLR, which represents a more physiological situation compared with stimulation with anti-TCR/CD28 Abs. CD4+ T cells were isolated from transgenic and control LEW rats after treatment with Dox in vivo, mixed with allogeneic splenocytes from DA rats in vitro, and cultured for up to 3 d in the continuous presence of Dox. INSR-deficient cells proliferated significantly less than did control cells at all time points, which was accompanied by a reduced expression of IFN-γ (Fig. 5A, 5B). CD8+ T cells were also analyzed in an MLR and showed a similarly reduced proliferation rate when the INSR was silenced (Fig. 5C). In addition, the cytotoxicity of CD8+ T cells was tested by immunization of control and transgenic LEW rats with splenocytes from wt DA rats, followed by restimulation of lymphocytes from regional lymph nodes with allogeneic splenocytes in vitro. Importantly, transgenic CTLs were less efficient at lysing allogeneic target cells than were control CTLs (Fig. 5D). Collectively, these experiments confirm that the INSR was necessary for full T cell activity in a physiological situation.

FIGURE 5.

Silencing of the INSR differentially impacts functional features of T cell subsets. (A) CD4+ T cells were isolated from transgenic (tg) and control (wt) LEW rats that received Dox-containing food pellets for 5 d and were subjected to an MLR in the continuous presence of Dox for up to 3 d using irradiated splenocytes from DA rats. Proliferation was measured by [3H]thymidine-incorporation assay and is depicted as the mean cpm ± SEM (n = 4–9). (B) IFN-γ mRNA levels were determined in MLRs after 3 d of culture by quantitative RT-PCR. Values were normalized to β-actin and are depicted as the mean ± SEM (n = 3/4 for wt/tg). (C) CD8+ T cells were isolated from transgenic (tg) and control (wt) LEW rats and subjected to an MLR, as described above. Proliferation was determined by [3H]thymidine-incorporation assay and is depicted as the mean cpm ± SEM (n = 4–9). (D) Alloreactive CTLs were generated from transgenic (tg) and control (wt) LEW rats, fed Dox-containing pellets, by immunization with splenocytes from DA rats. After 10 d, lymphocytes were isolated from regional lymph nodes and restimulated in vitro. The mean specific lysis ± SEM of DA blasts serving as target cells of the CTLs was determined by [51Cr]-release assay at different E:T ratios (n = 11–14). (E) Transgenic (tg) and control (wt) rats were treated with Dox for 5 d in vivo, and lymph node cells were analyzed by FACS. The percentage of CD4+CD25+Foxp3+ Tregs among all CD4+ T cells is depicted (left, n = 4). Proliferating Tregs were identified by intracellular staining with Ki-67. The frequency of Ki-67+ cells among all Tregs is depicted (right, n = 4). (F) Treg function was investigated using an in vitro suppression assay. CD4+CD25+ Tregs were isolated from transgenic (tg) and control (wt) rats that received Dox-containing food for 5 d and were added to stimulated CD4+CD25 indicator T cells at different ratios. The cultures were pulsed 2 d later, and cells were harvested and analyzed by [3H]thymidine-incorporation assay another 16 h later. Proliferation of indicator cells cultured without the addition of Tregs served as a control. Proliferation is depicted as the mean cpm ± SEM (n = 6–9). *p < 0.05, **p < 0.01, unpaired t test, refers to the comparison between genotypes (A–C, E, and F); two-way ANOVA (D). n.s., not significant.

FIGURE 5.

Silencing of the INSR differentially impacts functional features of T cell subsets. (A) CD4+ T cells were isolated from transgenic (tg) and control (wt) LEW rats that received Dox-containing food pellets for 5 d and were subjected to an MLR in the continuous presence of Dox for up to 3 d using irradiated splenocytes from DA rats. Proliferation was measured by [3H]thymidine-incorporation assay and is depicted as the mean cpm ± SEM (n = 4–9). (B) IFN-γ mRNA levels were determined in MLRs after 3 d of culture by quantitative RT-PCR. Values were normalized to β-actin and are depicted as the mean ± SEM (n = 3/4 for wt/tg). (C) CD8+ T cells were isolated from transgenic (tg) and control (wt) LEW rats and subjected to an MLR, as described above. Proliferation was determined by [3H]thymidine-incorporation assay and is depicted as the mean cpm ± SEM (n = 4–9). (D) Alloreactive CTLs were generated from transgenic (tg) and control (wt) LEW rats, fed Dox-containing pellets, by immunization with splenocytes from DA rats. After 10 d, lymphocytes were isolated from regional lymph nodes and restimulated in vitro. The mean specific lysis ± SEM of DA blasts serving as target cells of the CTLs was determined by [51Cr]-release assay at different E:T ratios (n = 11–14). (E) Transgenic (tg) and control (wt) rats were treated with Dox for 5 d in vivo, and lymph node cells were analyzed by FACS. The percentage of CD4+CD25+Foxp3+ Tregs among all CD4+ T cells is depicted (left, n = 4). Proliferating Tregs were identified by intracellular staining with Ki-67. The frequency of Ki-67+ cells among all Tregs is depicted (right, n = 4). (F) Treg function was investigated using an in vitro suppression assay. CD4+CD25+ Tregs were isolated from transgenic (tg) and control (wt) rats that received Dox-containing food for 5 d and were added to stimulated CD4+CD25 indicator T cells at different ratios. The cultures were pulsed 2 d later, and cells were harvested and analyzed by [3H]thymidine-incorporation assay another 16 h later. Proliferation of indicator cells cultured without the addition of Tregs served as a control. Proliferation is depicted as the mean cpm ± SEM (n = 6–9). *p < 0.05, **p < 0.01, unpaired t test, refers to the comparison between genotypes (A–C, E, and F); two-way ANOVA (D). n.s., not significant.

Close modal

Tregs have a different metabolic profile compared with other T cell subsets and show a particularly high proliferation rate in vivo (41, 42). To investigate whether Tregs were affected by the absence of the INSR, transgenic and control rats were treated with Dox in vivo. The frequency of Foxp3+CD25+CD4+ cells and the percentage of proliferating Ki-67+ cells among them were the same, regardless of the genotype (Fig. 5E). The function of Tregs was tested using an in vitro suppression assay. It turned out that Tregs from transgenic and control rats were equally efficient in inhibiting the proliferation of conventional T cells, suggesting that the INSR was dispensable for Treg activity (Fig. 5F). Thus, our results indicate that the differential INSR dependency of T cell subsets is in line with their proposed metabolic requirements.

To test whether the INSR in CD4+ and CD8+ T cells is relevant in vivo, we used a rat model of GvHD (35, 36). Bone marrow and lymph node cells containing mature T lymphocytes were isolated from transgenic and control rats on the LEW background and transferred into sublethally irradiated DA rats. When GvHD was induced by control cells, clinical symptoms were initially observed around day 6 and worsened rapidly thereafter (Fig. 6A). Consequently, the majority of the rats had died by the end of the third week or had to be sacrificed for ethical reasons (Fig. 6B). Conversely, transfer of INSR-deficient bone marrow and lymph node cells resulted in an attenuated disease course; more than half of the rats were still alive after 3 wk (Fig. 6A, 6B).

FIGURE 6.

The INSR is required for full pathogenicity of T cells in a rat model of GvHD. (A and B) GvHD was induced in sublethally irradiated wt DA rats by transferring bone marrow and lymph node cells from transgenic (tg) and control (wt) LEW rats that received Dox-containing food pellets for 5 d. Rats injected only with bone marrow (BMonly) served as controls. All rats received a Dox-containing diet during the entire experiment. (A) The clinical score was determined daily in a blinded fashion based on five parameters (body weight, activity, fur, skin, and posture) and is depicted as the mean ± SEM. **p < 0.01, Kruskal–Wallis test. (B) Kaplan–Meier survival curves are depicted that correspond to the same rats shown in (A). n = 14 or 15 (GvHD), n = 6 (BMonly), pool of three independent experiments. **p < 0.01, log-rank Mantel–Cox test. (C and D) GvHD was induced as described above. At day 10, the spleens of the recipient rats were removed and used for analysis. (C) T cell proliferation was measured by intracellular Ki-67 staining and is depicted as the percentage of Ki-67+ cells among TCRβ+ cells (left, n = 4). T cell apoptosis is depicted as the percentage of AxV+7-AAD+ cells among TCRβ+ cells (right, n = 4). (D) Splenocytes were analyzed for mRNA levels of Glut1, Glut3, and Glut4 by quantitative RT-PCR. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples were arbitrarily set to 1 for each gene (n = 4/5 for wt/tg). Statistical analyses were performed using an unpaired t test and refer to the comparison between genotypes. n.s., not significant.

FIGURE 6.

The INSR is required for full pathogenicity of T cells in a rat model of GvHD. (A and B) GvHD was induced in sublethally irradiated wt DA rats by transferring bone marrow and lymph node cells from transgenic (tg) and control (wt) LEW rats that received Dox-containing food pellets for 5 d. Rats injected only with bone marrow (BMonly) served as controls. All rats received a Dox-containing diet during the entire experiment. (A) The clinical score was determined daily in a blinded fashion based on five parameters (body weight, activity, fur, skin, and posture) and is depicted as the mean ± SEM. **p < 0.01, Kruskal–Wallis test. (B) Kaplan–Meier survival curves are depicted that correspond to the same rats shown in (A). n = 14 or 15 (GvHD), n = 6 (BMonly), pool of three independent experiments. **p < 0.01, log-rank Mantel–Cox test. (C and D) GvHD was induced as described above. At day 10, the spleens of the recipient rats were removed and used for analysis. (C) T cell proliferation was measured by intracellular Ki-67 staining and is depicted as the percentage of Ki-67+ cells among TCRβ+ cells (left, n = 4). T cell apoptosis is depicted as the percentage of AxV+7-AAD+ cells among TCRβ+ cells (right, n = 4). (D) Splenocytes were analyzed for mRNA levels of Glut1, Glut3, and Glut4 by quantitative RT-PCR. Values were normalized to β-actin and are depicted as the mean ± SEM; wt samples were arbitrarily set to 1 for each gene (n = 4/5 for wt/tg). Statistical analyses were performed using an unpaired t test and refer to the comparison between genotypes. n.s., not significant.

Close modal

To obtain insights into the mechanisms that may explain the reduced pathogenicity of INSR-deficient T cells, we analyzed rats of both genotypes at day 10 after GvHD induction. Ex vivo analysis of splenocytes revealed a reduction in T cell proliferation but an unaltered apoptosis rate in rats that had been transplanted with INSR-deficient lymph node cells (Fig. 6C). Concomitantly, Glut3 and Glut4 mRNA levels were reduced (Fig. 6D). The sensitivity of these assays was limited by the fact that donor T cells and residual splenocytes of the recipients could not be distinguished, offering an explanation for why statistical significance was not reached, despite obvious trends. Taken together, our findings suggest that cell-intrinsic INSR expression boosts T cell activity and, thereby, impacts the strength of adaptive immune responses in vivo.

We used AT-EAE to corroborate our findings made in GvHD. Ag-specific CD4+ effector T cells used for disease induction were generated by immunization of transgenic rats with gpMBP, followed by restimulation in vitro in the presence of Dox to silence the INSR or in the absence of Dox. Adoptive transfer of both types of encephalitogenic T cells in the continuous presence of Dox resulted in a monophasic disease course with a peak on day 5, as expected (Fig. 7A). However, there was no difference in clinical symptoms, regardless of whether the cells and rats had been treated with Dox (Fig. 7A). Surprisingly, Western blot analysis revealed that encephalitogenic T cells did not express the INSR at all, even before treatment, which explains the absence of any observable differences between both experimental groups in this model (Fig. 7B).

FIGURE 7.

The INSR is only expressed on newly activated T cells and is required for full-blown EAE after active immunization. (A) AT-EAE was induced in wt LEW rats using transgenic (tg) encephalitogenic T cells restimulated and expanded in vitro in the absence (con) or presence of Dox. Similarly, recipient rats were fed a Dox-containing diet or standard food pellets (con). Mean clinical scores ± SEM were determined daily in a blinded manner (n = 4). (B) wt CD4+ T cells were analyzed directly after isolation from the lymph nodes (con) or after their costimulation with anti-TCR/CD28 Abs for 2 d (stim). In addition, two lines of encephalitogenic T cells repeatedly restimulated and expanded with gpMBP in vitro were tested. INSR expression was determined by Western blot analysis; ERK served as a loading control. One representative blot of four is depicted. (C) Bone marrow chimeric rats were generated by injection of wt LEW rats with bone marrow from transgenic (tg) or control (wt) LEW rats after irradiation and T cell depletion with monoclonal anti-CD4/8 Abs. Following completion of immune reconstitution, EAE was actively induced by immunization with gpMBP in rats harboring a transgenic or control immune system and were fed a Dox-containing diet for the entire experiment. Mean clinical scores ± SEM were determined in a blinded manner daily (n = 7 or 8). Statistical analyses of the disease course (A and C) were performed using the Kruskal–Wallis test and refer to the comparison between genotypes. *p < 0.05. n.s., not significant.

FIGURE 7.

The INSR is only expressed on newly activated T cells and is required for full-blown EAE after active immunization. (A) AT-EAE was induced in wt LEW rats using transgenic (tg) encephalitogenic T cells restimulated and expanded in vitro in the absence (con) or presence of Dox. Similarly, recipient rats were fed a Dox-containing diet or standard food pellets (con). Mean clinical scores ± SEM were determined daily in a blinded manner (n = 4). (B) wt CD4+ T cells were analyzed directly after isolation from the lymph nodes (con) or after their costimulation with anti-TCR/CD28 Abs for 2 d (stim). In addition, two lines of encephalitogenic T cells repeatedly restimulated and expanded with gpMBP in vitro were tested. INSR expression was determined by Western blot analysis; ERK served as a loading control. One representative blot of four is depicted. (C) Bone marrow chimeric rats were generated by injection of wt LEW rats with bone marrow from transgenic (tg) or control (wt) LEW rats after irradiation and T cell depletion with monoclonal anti-CD4/8 Abs. Following completion of immune reconstitution, EAE was actively induced by immunization with gpMBP in rats harboring a transgenic or control immune system and were fed a Dox-containing diet for the entire experiment. Mean clinical scores ± SEM were determined in a blinded manner daily (n = 7 or 8). Statistical analyses of the disease course (A and C) were performed using the Kruskal–Wallis test and refer to the comparison between genotypes. *p < 0.05. n.s., not significant.

Close modal

As an alternative approach to tackle INSR function in EAE, we induced the disease by active immunization with gpMBP. Because a complete knockdown of the INSR in all organs causes lethal diabetes within a few days (27), we generated chimeric rats using bone marrow from transgenic or control rats. To ensure complete eradication of mature T cells in the recipient rats, they were treated with depleting anti-CD4/CD8 Abs, followed by sublethal irradiation and bone marrow transfer. This procedure resulted in the successful reconstitution of the recipient’s immune system after 12 wk, with almost all T cells being derived from the graft. Finally, 7 d of Dox treatment led to ablation of the INSR in hematopoietic cells of rats that received transgenic bone marrow, whereas it was still present in the immune system of rats reconstituted with control bone marrow. Induction of EAE caused the expected disease course, with first clinical symptoms around day 10 and the peak of the disease around day 15. INSR deficiency in the hematopoietic system attenuated EAE; the onset of the disease was delayed and the overall severity was reduced in rats harboring a transgenic immune system compared with control rats (Fig. 7C). This finding indicates that the INSR plays an important role during T cell priming, as well as the subsequent differentiation to effector T cells, and corroborates the notion that it is required for full activity of T cells in adaptive-immune responses in vivo.

The immune system has the ability to respond rapidly to the presence of pathogens by exponentially increasing the number of Ag-specific lymphocytes, followed by their differentiation into effector cells. The enormous energy demand caused by such a response cannot be met by OXPHOS, which is a relatively slow process. Therefore, most lymphocytes switch to aerobic glycolysis, a process initially described for tumor cells by Warburg et al. (43, 44). The altered metabolism, for instance in T cells, necessitates a huge increase in glucose uptake, which is promoted by Glut1 (8, 13). In addition, activated T cells start to express the INSR, which fosters an increase in Glut3 and Glut4, as well as an upregulation of glycolytic enzymes (8, 9, 21). Because INSR induction coincides with the metabolic switch, it was reasonable to speculate that this mechanism plays a critical role in T cell function.

Initially, we sought to analyze thymic T cell development, which involves massive proliferation and, thus, is an energy-demanding process. Surprisingly, there was no difference in thymocyte composition after silencing of the INSR in vivo, despite the fact that it is constitutively expressed in the thymus. Therefore, it appears that the absence of the INSR is compensated by other mechanisms. Of note, rat thymocytes stimulated with mitogens in vitro were shown to switch from OXPHOS to anaerobic glycolysis (15), but it is unknown whether such a switch also occurs during thymocyte development.

The role of the INSR in the immune system is ill-defined. Our data now indicate that activated T cells need the INSR to cover their large glucose demand, which then allows them to acquire full effector functions. In essence, polyclonal activation of CD4+ T cells lacking the INSR was delayed, cytokine production was diminished, and migration and proliferation were reduced. In addition, apoptosis induction by serum deprivation and AICD were enhanced, which is in line with the increased expression of the proapoptotic protein BAD, a downstream target of the INSR signaling pathway. The observed deficits in T cell function coincided with a reduction in Glut3, Glut4, and LDH mRNA levels, glucose uptake, and lactate production. Notably, glucose uptake by activated CD4+ T cells was only moderately diminished in the absence of the INSR, which fits well with our observation that Glut1 levels were unaltered. However, the fact that Glut3 and Glut4 expression was increased in stimulated rat T cells is important for the translational relevance of our data. Although mouse T cells were reported not to rely on Glut4 at all (20), human T cells were shown to express Glut3 and Glut4 in an insulin-dependent manner upon activation (21). Therefore, rats might reflect certain aspects of human physiology more faithfully than mice. Based on previous data and our data, we are inclined to believe that the roles of different Gluts in T cells have been underestimated. In our rat model, Glut1 expression appears to be exclusively regulated by TCR/CD28 signaling, whereas induction of Glut3 and Glut4 also requires the INSR. This notion would be in line with our analysis of the signaling network controlling T cell metabolism and function. In transgenic CD4+ T cells, we found a broad, albeit moderate, reduction in many proteins downstream of the INSR and the TCR/CD28 complex (e.g., RAS/ERK, PI3K/Akt, and mTOR pathways). In contrast, the LEPR/Stat3 pathway was unaffected by INSR silencing. We propose that insulin and costimulatory signals share similar intracellular signaling molecules through which they cooperate in regulating glucose uptake and immunological functions. Thus, our results support a critical role for the INSR in securing sufficient energy supply to effector T cells.

Previous work revealed that CD4+ and CD8+ effector T cells predominantly meet their energy demand by aerobic glycolysis (1719), whereas Tregs are largely independent of glucose because they are fueled by fatty acid oxidation (14, 45). Experiments described in this article show that in vitro alloreactive INSR-deficient effector T cells proliferated at a lower rate, produced less proinflammatory cytokines, and were less cytotoxic. However, the frequency, proliferation, and repressive activity of INSR-deficient Tregs were unaltered. Thus, our data identified functional deficits in selected T cell subsets that match the metabolic processes used to meet their energy demand. Although earlier studies indicated that metabolic pathways (1), including insulin signaling (46), influence the Th1/Th2 polarization of T cells, our findings did not reconfirm this notion.

Central to our work is the finding that the absence of the INSR in T cells impacts adaptive immunity in vivo, despite having no influence on the homeostasis of circulating naive T cells. The role of the INSR in adaptive-immune responses was investigated in two models of inflammatory diseases that involve CD4+ and CD8+ T cells (36, 47). GvHD is caused by allogeneic T cells that are activated after transfer into an MHC-mismatched host. As expected from our in vitro results, clinical symptoms of GvHD were milder and survival was significantly cantly better when we transferred transgenic lymphocytes instead of control ones. This finding indicates that full-blown GvHD is not possible without the INSR, which is in line with the observation that disease activity in mice and men is linked to enhanced glucose uptake (23, 24) and that glycolysis becomes the major energy source of pathogenic T cells in GvHD after metabolic reprogramming (22). Our analyses of splenocytes isolated from rats with overt disease symptoms further suggested that diminished glucose uptake and reduced proliferation might account for the lower disease activity of INSR-deficient T cells. However, an increased apoptosis rate of INSR-deficient T cells might have been masked by the rapid clearance of apoptotic cells in vivo. Whether altered cytokine production or a reduced cytotoxicity of alloreactive CTLs also contributes to this effect remains to be determined.

EAE mimics pathophysiological features of MS and can be induced by transfer of Ag-specific effector T cells. Unfortunately, their repeated restimulation during in vitro culture resulted in a complete loss of INSR protein independently of Dox treatment. To circumvent artifacts introduced by maintaining encephalitogenic T cells ex vivo, we induced EAE by active immunization of bone marrow chimeric rats in which cells of the immune system expressed the INSR or not. EAE was attenuated in rats with a hematopoietic deficiency of the INSR, confirming its importance in vivo. Although we cannot exclude a contribution by leukocyte subsets other than T cells, it is likely that the difference in disease severity is predominantly caused by an impaired function of CD4+ and CD8+ effector T cells. Because Ag-dependent T cells need to proliferate and differentiate after immunization, we propose that efficient glycolysis is necessary for full-blown EAE. This notion is supported by a report showing that inhibition of glucose metabolism with 2-deoxyglucose suppressed EAE by interfering with the generation of Th17 cells (25).

Currently, ∼350 million people worldwide suffer from type 2 diabetes, which is characterized by insulin insensitivity (48) and may lead to kidney failure, heart disease, and neuropathies. Type 2 diabetes patients also have impaired clearance of infections, but the underlying mechanisms have not been fully explored (49). Although many factors, such as disturbed blood flow or abnormally high glucose levels, may contribute to this phenotype, our findings now reinforce the notion that an impaired T cell function may account, at least in part, for the increased susceptibility of these patients to infectious diseases (9). Therefore, it is likely that the impact of deregulated insulin signaling on immunity is being underestimated. Our finding that the absence of insulin responsiveness impairs adaptive immunity has significant socioeconomic consequences: the immune system of patients suffering from type 2 diabetes should be carefully monitored, and therapeutic interventions that target the energy metabolism of T cells should be considered with the potential to decrease morbidity and mortality in the future.

We thank Amina Bassibas, Julian Koch, and Leslie Elsner for expert technical assistance and Dr. Anja Siepert for providing monoclonal anti-CD4/CD8 Abs.

This work was supported by grants from the Deutsche Forschungsgemeinschaft (RE1631/10-1, RE1631/15-1, and SFB/TRR43-B13 to H.M.R.) and the European Union (FP7-PEOPLE-2012-ITN-315963 CELLEUROPE to R.D.).

Abbreviations used in this article:

7-AAD

7-aminoactinomycin D

AICD

activation-induced cell death

AT-EAE

adoptive-transfer EAE

AxV

annexin V

DA

Dark Agouti

DN

double-negative

Dox

doxycycline

DP

double-positive

EAE

experimental autoimmune encephalomyelitis

FTOC

fetal thymus organ culture

Glut

glucose transporter

gpMBP

guinea pig myelin oligodendrocyte glycoprotein

GvHD

graft-versus-host disease

INSR

insulin receptor

LDH

lactate dehydrogenase

LEPR

leptin receptor

LEW

Lewis

MS

multiple sclerosis

mTOR

mammalian target of rapamycin

OXPHOS

oxidative phosphorylation

SP

single-positive

Treg

regulatory T cell

wt

wild-type.

1
Wang
R.
,
Green
D. R.
.
2012
.
Metabolic checkpoints in activated T cells.
Nat. Immunol.
13
:
907
915
.
2
Pearce
E. L.
,
Pearce
E. J.
.
2013
.
Metabolic pathways in immune cell activation and quiescence.
Immunity
38
:
633
643
.
3
Wahl
D. R.
,
Byersdorfer
C. A.
,
Ferrara
J. L.
,
Opipari
A. W.
 Jr.
,
Glick
G. D.
.
2012
.
Distinct metabolic programs in activated T cells: opportunities for selective immunomodulation.
Immunol. Rev.
249
:
104
115
.
4
Guppy
M.
,
Greiner
E.
,
Brand
K.
.
1993
.
The role of the Crabtree effect and an endogenous fuel in the energy metabolism of resting and proliferating thymocytes.
Eur. J. Biochem.
212
:
95
99
.
5
Chang
C. H.
,
Curtis
J. D.
,
Maggi
L. B.
 Jr.
,
Faubert
B.
,
Villarino
A. V.
,
O’Sullivan
D.
,
Huang
S. C.
,
van der Windt
G. J.
,
Blagih
J.
,
Qiu
J.
, et al
.
2013
.
Posttranscriptional control of T cell effector function by aerobic glycolysis.
Cell
153
:
1239
1251
.
6
Pearce
E. L.
2010
.
Metabolism in T cell activation and differentiation.
Curr. Opin. Immunol.
22
:
314
320
.
7
Helderman
J. H.
,
Strom
T. B.
.
1978
.
Specific insulin binding site on T and B lymphocytes as a marker of cell activation.
Nature
274
:
62
63
.
8
Frauwirth
K. A.
,
Thompson
C. B.
.
2004
.
Regulation of T lymphocyte metabolism.
J. Immunol.
172
:
4661
4665
.
9
Stentz
F. B.
,
Kitabchi
A. E.
.
2003
.
Activated T lymphocytes in Type 2 diabetes: implications from in vitro studies.
Curr. Drug Targets
4
:
493
503
.
10
Fox
C. J.
,
Hammerman
P. S.
,
Thompson
C. B.
.
2005
.
Fuel feeds function: energy metabolism and the T-cell response.
Nat. Rev. Immunol.
5
:
844
852
.
11
Maciver
N. J.
,
Jacobs
S. R.
,
Wieman
H. L.
,
Wofford
J. A.
,
Coloff
J. L.
,
Rathmell
J. C.
.
2008
.
Glucose metabolism in lymphocytes is a regulated process with significant effects on immune cell function and survival.
J. Leukoc. Biol.
84
:
949
957
.
12
Jacobs
S. R.
,
Herman
C. E.
,
Maciver
N. J.
,
Wofford
J. A.
,
Wieman
H. L.
,
Hammen
J. J.
,
Rathmell
J. C.
.
2008
.
Glucose uptake is limiting in T cell activation and requires CD28-mediated Akt-dependent and independent pathways.
J. Immunol.
180
:
4476
4486
.
13
Frauwirth
K. A.
,
Riley
J. L.
,
Harris
M. H.
,
Parry
R. V.
,
Rathmell
J. C.
,
Plas
D. R.
,
Elstrom
R. L.
,
June
C. H.
,
Thompson
C. B.
.
2002
.
The CD28 signaling pathway regulates glucose metabolism.
Immunity
16
:
769
777
.
14
Michalek
R. D.
,
Gerriets
V. A.
,
Jacobs
S. R.
,
Macintyre
A. N.
,
MacIver
N. J.
,
Mason
E. F.
,
Sullivan
S. A.
,
Nichols
A. G.
,
Rathmell
J. C.
.
2011
.
Cutting edge: distinct glycolytic and lipid oxidative metabolic programs are essential for effector and regulatory CD4+ T cell subsets.
J. Immunol.
186
:
3299
3303
.
15
Greiner
E. F.
,
Guppy
M.
,
Brand
K.
.
1994
.
Glucose is essential for proliferation and the glycolytic enzyme induction that provokes a transition to glycolytic energy production.
J. Biol. Chem.
269
:
31484
31490
.
16
Vander Heiden
M. G.
,
Plas
D. R.
,
Rathmell
J. C.
,
Fox
C. J.
,
Harris
M. H.
,
Thompson
C. B.
.
2001
.
Growth factors can influence cell growth and survival through effects on glucose metabolism.
Mol. Cell. Biol.
21
:
5899
5912
.
17
Dang
E. V.
,
Barbi
J.
,
Yang
H. Y.
,
Jinasena
D.
,
Yu
H.
,
Zheng
Y.
,
Bordman
Z.
,
Fu
J.
,
Kim
Y.
,
Yen
H. R.
, et al
.
2011
.
Control of T(H)17/T(reg) balance by hypoxia-inducible factor 1.
Cell
146
:
772
784
.
18
He
S.
,
Kato
K.
,
Jiang
J.
,
Wahl
D. R.
,
Mineishi
S.
,
Fisher
E. M.
,
Murasko
D. M.
,
Glick
G. D.
,
Zhang
Y.
.
2011
.
Characterization of the metabolic phenotype of rapamycin-treated CD8+ T cells with augmented ability to generate long-lasting memory cells.
PLoS One
6
:
e20107
.
19
Michalek
R. D.
,
Gerriets
V. A.
,
Nichols
A. G.
,
Inoue
M.
,
Kazmin
D.
,
Chang
C. Y.
,
Dwyer
M. A.
,
Nelson
E. R.
,
Pollizzi
K. N.
,
Ilkayeva
O.
, et al
.
2011
.
Estrogen-related receptor-α is a metabolic regulator of effector T-cell activation and differentiation.
Proc. Natl. Acad. Sci. USA
108
:
18348
18353
.
20
Macintyre
A. N.
,
Gerriets
V. A.
,
Nichols
A. G.
,
Michalek
R. D.
,
Rudolph
M. C.
,
Deoliveira
D.
,
Anderson
S. M.
,
Abel
E. D.
,
Chen
B. J.
,
Hale
L. P.
,
Rathmell
J. C.
.
2014
.
The glucose transporter Glut1 is selectively essential for CD4 T cell activation and effector function.
Cell Metab.
20
:
61
72
.
21
Maratou
E.
,
Dimitriadis
G.
,
Kollias
A.
,
Boutati
E.
,
Lambadiari
V.
,
Mitrou
P.
,
Raptis
S. A.
.
2007
.
Glucose transporter expression on the plasma membrane of resting and activated white blood cells.
Eur. J. Clin. Invest.
37
:
282
290
.
22
Nguyen
H. D.
,
Chatterjee
S.
,
Haarberg
K. M.
,
Wu
Y.
,
Bastian
D.
,
Heinrichs
J.
,
Fu
J.
,
Daenthanasanmak
A.
,
Schutt
S.
,
Shrestha
S.
, et al
.
2016
.
Metabolic reprogramming of alloantigen-activated T cells after hematopoietic cell transplantation.
J. Clin. Invest.
126
:
1337
1352
.
23
Gatza
E.
,
Wahl
D. R.
,
Opipari
A. W.
,
Sundberg
T. B.
,
Reddy
P.
,
Liu
C.
,
Glick
G. D.
,
Ferrara
J. L.
.
2011
.
Manipulating the bioenergetics of alloreactive T cells causes their selective apoptosis and arrests graft-versus-host disease.
Sci. Transl. Med.
3
:
67ra8
.
24
Stelljes
M.
,
Hermann
S.
,
Albring
J.
,
Köhler
G.
,
Löffler
M.
,
Franzius
C.
,
Poremba
C.
,
Schlösser
V.
,
Volkmann
S.
,
Opitz
C.
, et al
.
2008
.
Clinical molecular imaging in intestinal graft-versus-host disease: mapping of disease activity, prediction, and monitoring of treatment efficiency by positron emission tomography.
Blood
111
:
2909
2918
.
25
Shi
L. Z.
,
Wang
R.
,
Huang
G.
,
Vogel
P.
,
Neale
G.
,
Green
D. R.
,
Chi
H.
.
2011
.
HIF1alpha-dependent glycolytic pathway orchestrates a metabolic checkpoint for the differentiation of TH17 and Treg cells.
J. Exp. Med.
208
:
1367
1376
.
26
Yang
K.
,
Chi
H.
.
2014
.
Metabolic control of Th17 cell generation and CNS inflammation.
J. Neurol. Neurophysiol.
Suppl. 12
:
S12
-
004
.
27
Herold
M. J.
,
van den Brandt
J.
,
Seibler
J.
,
Reichardt
H. M.
.
2008
.
Inducible and reversible gene silencing by stable integration of an shRNA-encoding lentivirus in transgenic rats.
Proc. Natl. Acad. Sci. USA
105
:
18507
18512
.
28
van den Brandt
J.
,
Wang
D.
,
Kwon
S. H.
,
Heinkelein
M.
,
Reichardt
H. M.
.
2004
.
Lentivirally generated eGFP-transgenic rats allow efficient cell tracking in vivo.
Genesis
39
:
94
99
.
29
van den Brandt
J.
,
Kwon
S. H.
,
Hünig
T.
,
McPherson
K. G.
,
Reichardt
H. M.
.
2005
.
Sustained pre-TCR expression in Notch1IC-transgenic rats impairs T cell maturation and selection.
J. Immunol.
174
:
7845
7852
.
30
van den Brandt
J.
,
Voss
K.
,
Schott
M.
,
Hünig
T.
,
Wolfe
M. S.
,
Reichardt
H. M.
.
2004
.
Inhibition of Notch signaling biases rat thymocyte development towards the NK cell lineage.
Eur. J. Immunol.
34
:
1405
1413
.
31
Schweingruber
N.
,
Fischer
H. J.
,
Fischer
L.
,
van den Brandt
J.
,
Karabinskaya
A.
,
Labi
V.
,
Villunger
A.
,
Kretzschmar
B.
,
Huppke
P.
,
Simons
M.
, et al
.
2014
.
Chemokine-mediated redirection of T cells constitutes a critical mechanism of glucocorticoid therapy in autoimmune CNS responses.
Acta Neuropathol.
127
:
713
729
.
32
Wüst
S.
,
van den Brandt
J.
,
Tischner
D.
,
Kleiman
A.
,
Tuckermann
J. P.
,
Gold
R.
,
Lühder
F.
,
Reichardt
H. M.
.
2008
.
Peripheral T cells are the therapeutic targets of glucocorticoids in experimental autoimmune encephalomyelitis.
J. Immunol.
180
:
8434
8443
.
33
Taurog
J. D.
,
Dorris
M. L.
,
Satumtira
N.
,
Tran
T. M.
,
Sharma
R.
,
Dressel
R.
,
van den Brandt
J.
,
Reichardt
H. M.
.
2009
.
Spondylarthritis in HLA-B27/human beta2-microglobulin-transgenic rats is not prevented by lack of CD8.
Arthritis Rheum.
60
:
1977
1984
.
34
van den Brandt
J.
,
Fischer
H. J.
,
Walter
L.
,
Hünig
T.
,
Klöting
I.
,
Reichardt
H. M.
.
2010
.
Type 1 diabetes in BioBreeding rats is critically linked to an imbalance between Th17 and regulatory T cells and an altered TCR repertoire.
J. Immunol.
185
:
2285
2294
.
35
Zinöcker
S.
,
Sviland
L.
,
Dressel
R.
,
Rolstad
B.
.
2011
.
Kinetics of lymphocyte reconstitution after allogeneic bone marrow transplantation: markers of graft-versus-host disease.
J. Leukoc. Biol.
90
:
177
187
.
36
Theiss-Suennemann
J.
,
Jörß
K.
,
Messmann
J. J.
,
Reichardt
S. D.
,
Montes-Cobos
E.
,
Lühder
F.
,
Tuckermann
J. P.
,
Wolff
H. A.
,
Dressel
R.
,
Gröne
H. J.
, et al
.
2015
.
Glucocorticoids attenuate acute graft-versus-host disease by suppressing the cytotoxic capacity of CD8(+) T cells.
J. Pathol.
235
:
646
655
.
37
Cooke
K. R.
,
Kobzik
L.
,
Martin
T. R.
,
Brewer
J.
,
Delmonte
J.
 Jr.
,
Crawford
J. M.
,
Ferrara
J. L.
.
1996
.
An experimental model of idiopathic pneumonia syndrome after bone marrow transplantation: I. The roles of minor H antigens and endotoxin.
Blood
88
:
3230
3239
.
38
Tischner
D.
,
Weishaupt
A.
,
van den Brandt
J.
,
Müller
N.
,
Beyersdorf
N.
,
Ip
C. W.
,
Toyka
K. V.
,
Hünig
T.
,
Gold
R.
,
Kerkau
T.
,
Reichardt
H. M.
.
2006
.
Polyclonal expansion of regulatory T cells interferes with effector cell migration in a model of multiple sclerosis.
Brain
129
:
2635
2647
.
39
Goldfine
I. D.
,
Gardner
J. D.
,
Neville
D. M.
 Jr.
1972
.
Insulin action in isolated rat thymocytes. I. Binding of 125 I-insulin and stimulation of -aminoisobutyric acid transport.
J. Biol. Chem.
247
:
6919
6926
.
40
Helderman
J. H.
,
Strom
T. B.
.
1979
.
Role of protein and RNA synthesis in the development of insulin binding sites on activated thymus-derived lymphocytes.
J. Biol. Chem.
254
:
7203
7207
.
41
Vukmanovic-Stejic
M.
,
Zhang
Y.
,
Cook
J. E.
,
Fletcher
J. M.
,
McQuaid
A.
,
Masters
J. E.
,
Rustin
M. H.
,
Taams
L. S.
,
Beverley
P. C.
,
Macallan
D. C.
,
Akbar
A. N.
.
2006
.
Human CD4+ CD25hi Foxp3+ regulatory T cells are derived by rapid turnover of memory populations in vivo.
J. Clin. Invest.
116
:
2423
2433
.
42
Booth
N. J.
,
McQuaid
A. J.
,
Sobande
T.
,
Kissane
S.
,
Agius
E.
,
Jackson
S. E.
,
Salmon
M.
,
Falciani
F.
,
Yong
K.
,
Rustin
M. H.
, et al
.
2010
.
Different proliferative potential and migratory characteristics of human CD4+ regulatory T cells that express either CD45RA or CD45RO.
J. Immunol.
184
:
4317
4326
.
43
Warburg
O.
,
Gawehn
K.
,
Geissler
A. W.
.
1958
.
Metabolism of leukocytes
.
Z. Naturforsch. B
13B
:
515
516
.
44
Warburg
O.
1956
.
On the origin of cancer cells.
Science
123
:
309
314
.
45
Barbi
J.
,
Pardoll
D.
,
Pan
F.
.
2013
.
Metabolic control of the Treg/Th17 axis.
Immunol. Rev.
252
:
52
77
.
46
Viardot
A.
,
Grey
S. T.
,
Mackay
F.
,
Chisholm
D.
.
2007
.
Potential antiinflammatory role of insulin via the preferential polarization of effector T cells toward a T helper 2 phenotype.
Endocrinology
148
:
346
353
.
47
Camara
M.
,
Beyersdorf
N.
,
Fischer
H. J.
,
Herold
M. J.
,
Ip
C. W.
,
van den Brandt
J.
,
Toyka
K. V.
,
Taurog
J. D.
,
Hünig
T.
,
Herrmann
T.
, et al
.
2013
.
CD8+ T cell help is required for efficient induction of EAE in Lewis rats.
J. Neuroimmunol.
260
:
17
27
.
48
Nathan
D. M.
2015
.
Diabetes: advances in diagnosis and treatment.
JAMA
314
:
1052
1062
.
49
Müller
L. M.
,
Gorter
K. J.
,
Hak
E.
,
Goudzwaard
W. L.
,
Schellevis
F. G.
,
Hoepelman
A. I.
,
Rutten
G. E.
.
2005
.
Increased risk of common infections in patients with type 1 and type 2 diabetes mellitus.
Clin. Infect. Dis.
41
:
281
288
.

The authors have no financial conflicts of interest.