Human chorionic gonadotropin is an immune modulator and can prevent autoimmune diabetes in NOD mice
- First Online:
- Cite this article as:
- Khil, LY., Jun, HS., Kwon, H. et al. Diabetologia (2007) 50: 2147. doi:10.1007/s00125-007-0769-y
- 464 Downloads
Expression of T helper (Th)1 cytokine mRNA in pregnant women is known to be inversely correlated with serum human chorionic gonadotropin (hCG). Type 1 diabetes is a Th1-mediated autoimmune disease, in which intervention at an early stage of the autoimmune process can prevent disease progression. We hypothesised that immune modulation by treating young NOD mice with hCG may prevent diabetes.
Female NOD mice were treated with hCG or recombinant hCG from 3 to 15 weeks of age and the incidence of diabetes and development of insulitis was determined. CD4+ and CD8+ T cell populations, T cell proliferation, cytokine production and CD4+CD25+ regulatory T cells were examined and adoptive transfer experiments were performed.
Both purified and recombinant hCG prevented development of diabetes in NOD mice. hCG decreased the proportion and number of CD4+ and CD8+ T cells and inhibited T cell proliferative responses against beta cell antigens. hCG treatment suppressed IFN-γ production, but increased IL-10 and TGF-β production in splenocytes stimulated with anti-CD3 antibody. hCG treatment also suppressed TNF-α production in splenocytes stimulated with lipopolysaccharide. Furthermore, hCG treatment increased the CD4+CD25+/CD4+ T cell ratio in spleen and pancreatic lymph nodes. Depletion of CD4+CD25+ T cells from splenocytes of hCG-treated NOD mice abolished their preventive effect on diabetes transfer.
We conclude that hCG has an immunomodulatory effect by downregulating effector cells, including Th1 cells, CD8+ T cells and macrophages, and increasing the CD4+CD25+/CD4+ T cell ratio, thus preventing autoimmune diabetes in NOD mice.
KeywordsAutoimmune diseaseCytokinesHuman chorionic gonadotropinImmune regulationInsulitisNOD miceRegulatory T cellsT cellsTh1 immune responseType 1 diabetes
human chorionic gonadotropin
recombinant human chorionic gonadotropin
severe combined immunodeficient
Human chorionic gonadotropin (hCG) is a heterodimeric placental glycoprotein required to maintain pregnancy. The level of hCG increases during the first trimester of pregnancy and decreases to 10% of the peak value during the second and third trimesters . The symptoms of autoimmune diseases such as rheumatoid arthritis, Crohn’s disease and multiple sclerosis are attenuated during pregnancy [2–4]. In pregnant women, the expression of T helper (Th)1 cytokine mRNA such as IL2 and IFN-γ (also known as IFNG) is significantly decreased, while expression of IL18 mRNA, an inducer of IFN-γ in T lymphocytes and natural killer cells, is inversely correlated with serum levels of hCG . This suggests that hCG may have an immunoregulatory role in addition to its classical reproductive role in the maintenance of pregnancy. There is further evidence to support this view in animal models. Purified hCG and a 400–2,000 Da fraction of clinical grade hCG were shown to prevent diabetes in mice . Another study showed that treatment with hCG prevented the wasting and death of transgenic mice carrying human immunodeficiency virus genes . Recent studies have shown that this wasting syndrome is due in part to high levels of TNF-α, which can be suppressed by treatment with hCG . However, the mechanisms by which hCG prevents autoimmune disease are not clearly understood.
Type 1 diabetes results from the destruction of pancreatic beta cells by T cell-mediated autoimmune responses, which are thought to result from an autoimmune inflammatory process mediated by autoreactive Th1 cells and their secreted cytokines [9–11]. We hypothesised that treatment of prediabetic individuals with hCG might impede the autoimmune processes by modulating immune responses, resulting in the prevention of autoimmune type 1 diabetes. We tested this possibility in the NOD mouse, which is a widely used animal model for the study of human autoimmune type 1 diabetes. We report that hCG treatment downregulated Th1 cytokines and macrophage proinflammatory cytokines and increased Th2 cytokines and TGF-β. In addition, hCG decreased CD4+ and CD8+ T cells, but increased the ratio of CD4+CD25+ regulatory T cells/CD4+ T cells.
Female NOD and NOD/severe combined immunodeficient (NOD.SCID) mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). The animals were maintained under specific pathogen-free conditions and provided with free access to sterile food and water at the Animal Resources Centre, Faculty of Medicine, University of Calgary. The use and care of the animals were approved by the Animal Care Committee, Faculty of Medicine, University of Calgary.
Treatment with hCG and monitoring of diabetes onset
Recombinant hCG (rhCG) and hCG from the urine of pregnant women were obtained from Sigma (St Louis, MO, USA). Purity was determined by 12% SDS-PAGE followed by staining using a silver staining kit (Bio-Rad, Hercules, CA, USA). The pattern of protein bands in hCG was the same as in rhCG, indicating that the hCG did not contain other apparent protein contaminants compared with rhCG. NOD female mice (3 or 10 weeks of age) were injected i.p. with 50 IU hCG or rhCG (giving a blood concentration comparable with that in pregnant humans) in 200 μl PBS per mouse or with PBS alone, 5 times per week for 12 weeks. The development of diabetes was monitored by urine glucose measurements using Diastix (Bayer, West Haven, CT, USA) every week and positive glycosuria was confirmed by hyperglycaemia (blood glucose level >16.7 mmol/l). If mice became diabetic during the experiment, they were killed according to guidelines of the animal care protocol.
The pancreata were removed from hCG- or PBS-treated NOD mice at 15 or 30 weeks of age. The pancreata were fixed with 10% buffered formalin, embedded in paraffin and sectioned at 4.5 μm intervals. More than 500 serial sections were prepared from each pancreas and every 20th section was stained with haematoxylin and eosin . The degree of insulitis was scored as one of four categories: intact islet; <25% of the islet infiltrated; 25 to 50% of the islet infiltrated; 50 to 75% of the islet infiltrated; and >75% of the islet infiltrated. The degree of insulitis in 20 to 25 islets per mouse was independently evaluated by two investigators.
T cell proliferation
To examine the proliferation of T cells, splenocytes (1 × 105 cells) were cultured for 96 h in the presence or absence of anti-CD3 antibody (0.5 μg/ml) or mouse insulinoma (MIN)6N8a cell extract (25 μg/well) in 200 μl of complete RPMI medium containing 10% fetal bovine serum and antibiotics in a 96-well microplate. To prepare MIN6N8a cell extract, MIN6N8a cells were collected by trypsinisation, suspended in PBS and sonicated for 3 min. The cells were pulsed with [3H]thymidine (37 kBq/well) for 16 h of incubation and then harvested. The incorporation of [3H]thymidine was measured by liquid scintillation counting . A stimulation index was calculated by dividing the cpm from the stimulated group by the cpm from the unstimulated group.
Quantitative ELISA for cytokine production
Splenocytes (1 × 106 cells) from 10-week-old female NOD mice treated with PBS or hCG were stimulated with anti-CD3 antibody (0.5 μg/ml) for IFN-γ, IL-4, IL-10 and TGF-β or lipopolysaccharide (LPS; 10 ng/ml) for TNF-α for indicated times in 1 ml of complete RPMI medium. The supernatant fraction was collected and cytokine release was measured using a specific ELISA kit (Quantikine; R & D Systems, Minneapolis, MN, USA) according to the manufacturer’s protocol.
Splenocytes and pancreatic lymph node cells were isolated from hCG or PBS-treated female NOD mice at 10 weeks of age. To examine the CD4+ and CD8+ T cell populations, cells (1 × 106) were incubated with FITC-labelled anti-CD4 and phycoerythrin-labelled anti-CD8 antibodies (Pharmingen, Mississauga, ON, Canada) for 30 min at 4°C in staining buffer (1% fetal bovine serum and 0.1% sodium azide in PBS). To examine the macrophage population, cells were incubated with FITC-labelled anti-CD11b antibody (Pharmingen). To examine the CD4+CD25+ T cell population, cells were incubated with FITC-labelled anti-CD4 and phycoerythrin-labelled anti-CD25 antibodies. To examine the natural killer T and CD4+CD62L+ regulatory cell populations, splenocytes and lymph node cells were double-stained with FITC-labelled anti-CD3 and biotin-labelled DX5 antibodies or FITC-labelled anti-CD4 and biotin-labelled anti-CD62L antibodies, respectively, followed by phycoerythrin-labelled streptavidin. The cells were washed with staining buffer and analysed by FACS .
Splenocytes were isolated from hCG-, rhCG- or PBS-treated mice at 15 weeks of age, checked for viability under the microscope and transferred (i.v.) into 6-week-old NOD.SCID mice as described elsewhere . To examine the preventive effect, splenocytes from hCG- or PBS-treated mice were transferred along with same number of diabetic splenocytes. To determine whether CD4+ T cell population contained regulatory T cells, splenocytes were isolated from hCG- or PBS-treated NOD mice and CD8+ T cells were depleted with anti-CD8 antibody-linked micromagnetic beads (Miltenyi Biotec, Auburn, CA, USA). The CD8+ T cell-depleted splenocytes (1 × 107 cells) were co-transferred with splenocytes from diabetic NOD mice (0.5 × 107 cells, 2:1 ratio). To examine the role of CD4+CD25+ T cells, splenocytes were isolated from hCG-treated NOD mice and CD25+ T cells were depleted using anti-CD25 antibody-linked micromagnetic beads (Miltenyi Biotec). The CD25+ T cell-depleted splenocytes or total splenocytes (1 × 107 cells per recipient) were transferred. In addition, CD8+- and CD25+-depleted splenocytes were co-transferred with splenocytes from diabetic NOD mice. As a positive control, splenocytes from acutely diabetic mice were injected into NOD.SCID mice. The development of diabetes in NOD.SCID recipients was monitored by measuring urine glucose twice weekly and confirmed by measurement of blood glucose levels.
The statistical significance of the differences between groups was analysed by two-tailed Student’s t test. Normality of data was verified by the Kolmogorov–Smirnov test. A level of p < 0.05 was accepted as significant. For comparison of incidence of diabetes, the two-tailed log–rank test was used.
Purified hCG prevents autoimmune diabetes in NOD mice
hCG treatment decreases the production of IFN-γ and TNF-α and increases the production of IL-10 and TGF-β
hCG treatment decreases the proportion and number of CD4+ and CD8+ T cells, but increases the ratio of CD4+CD25+/CD4+ T cells
CD4+ and CD4+CD25+ T cells from hCG-treated mice protect against the development of diabetes
To determine whether CD4+CD25+ T cells are responsible for the prevention of diabetes, we depleted the CD8+ and CD25+ cell populations from splenocytes of hCG-treated NOD mice and co-transferred the remaining cells along with splenocytes from diabetic NOD mice into NOD.SCID recipients. All mice (6 of 6) that received CD8+ cell- and CD25+ cell-depleted splenocytes from hCG-treated NOD mice along with diabetogenic splenocytes developed diabetes (Fig. 6b), indicating that CD4+CD25+ cells play an important role in the prevention of diabetes. We also depleted the CD25+ T cell population from splenocytes of hCG-treated NOD mice and transferred the remaining cells into NOD.SCID mice. All the recipients of CD25+ T cell-depleted splenocytes developed diabetes, whereas none of the recipients of splenocytes that were not depleted of CD25+ T cells developed diabetes (Fig. 6c).
Recombinant hCG prevents autoimmune diabetes in NOD mice
Comparison of recombinant hCG with purified hCG with regard to control of autoimmune diabetes in NOD mice
Diabetes incidence by adoptive transferb
Production of cytokines (ng/ml)c
IFN-γ (n = 9)
1.400 ± 0.115
0.764 ± 0.077***
0.790 ± 0.0921**
IL-4 (n = 7)
0.129 ± 0.011
0.131 ± 0.012
0.133 ± 0.013
TNF-α (n = 5)
0.479 ± 0.076
0.236 ± 0.033*
0.287 ± 0.030*
hCG is required for the maintenance of pregnancy [1, 15]. The symptoms of autoimmune disease can be attenuated in pregnant women [2–4, 16, 17]. It was previously found that purified hCG or a fraction of clinical grade hCG can prevent autoimmune diabetes in animal models . However, the mechanisms by which hCG prevents autoimmune disease are not clearly known.
Here, we first examined the effect of hCG on the prevention of diabetes. Treatment of NOD mice with hCG before the development of insulitis resulted in complete prevention of diabetes, while treatment after the development of insulitis was partially effective, suggesting that hCG has a preventive effect before and after the onset of insulitis. In contrast to earlier reports , we also found that rhCG could prevent the development of diabetes.
Type 1 diabetes is a T cell-mediated autoimmune disease in both humans and NOD mice, in which activated Th1-type CD4+ and CD8+ T cells are believed to kill pancreatic beta cells by apoptosis [9, 10, 18]. Therefore, we examined the T cell population in hCG-treated NOD mice. Treatment of NOD mice with hCG decreased the number of CD4+ and CD8+ T cells. Consistent with this result, antigen-specific and non-specific T cell proliferation was inhibited by hCG. In addition, we found that hCG treatment decreased IFN-γ production and increased the production of IL-10 and TGF-β, suggesting that hCG may decrease the Th1 immune response. Taken together, these results indicate that treatment of NOD mice with hCG downregulates Th1-type CD4+ and CD8+ T cells. In addition to T cells, macrophages are important contributors to the destruction of beta cells in NOD mice. They achieve this by creating an immune environment that potentiates the Th1 response and by damaging beta cells with macrophage-derived soluble mediators such as IL-1β, TNF-α and IFN-γ . We found that hCG treatment decreased LPS-stimulated TNF-α production, indicating that macrophage function is also downregulated, which may then decrease the Th1 immune response. In addition, the reduction of beta cell-toxic substances from macrophages may synergistically act to protect beta cells from destruction. It is known that TNF-α upregulates the transcription factors nuclear factor-κB and activator protein-1, resulting in the expression of a number of inflammatory response genes downstream of TNF-α, and that in vivo treatment of cells with hCG downregulates TNF-α-induced nuclear factor-κB and activator protein-1 . This could explain, in part, the preventive effects of hCG on beta cell destruction.
CD8+ T cell-depleted splenocytes from hCG-treated mice were able to substantially prevent the transfer of diabetes by splenocytes from acutely diabetic NOD mice (Fig. 6a), achieving a similar degree of prevention to that seen in recipients of total splenocytes from hCG-treated NOD mice (Fig. 2b). This suggests that hCG treatment may induce a regulatory T cell population that can prevent diabetes transfer and that the CD4+ T cell population may contain these regulatory T cells. Regulatory T cell numbers and function have been shown to be impaired in NOD mice [20, 21] and in human patients with type 1 diabetes or other autoimmune diseases . Thus, we examined whether there were any changes in regulatory T cell populations, such as natural killer T cells, CD4+CD62L+ T cells and CD4+CD25+ T cells. We found that only the CD4+CD25+ T cell population was significantly increased in hCG-treated NOD mice. An increase in the CD4+CD25+ T cell population was also found in male NOD mice treated with hCG (data not shown). Although hCG treatment did not increase the absolute number of CD4+CD25+ regulatory T cells, the increase in the ratio of CD4+CD25+ T cells relative to CD4+ T cells was able to inhibit the proliferation of effector T cells, as CD4+CD25+ regulatory T cells are potent suppressors of activation of both CD4+ and CD8+ T cells [22, 23]. Adoptive transfer of CD4+CD25+ T cells can prevent the development of autoimmune thyroiditis, gastritis, inflammatory bowel disease and diabetes [24–26]. Moreover, CD4+CD25+ T cells have been shown to play an important role in controlling the progression of type 1 diabetes in NOD mice [27–29]. Therefore, hCG may increase the ratio of CD4+CD25+/CD4+ T cells, which could restore the immune balance between regulatory and effector T cells, resulting in the prevention of autoimmune disease. Depletion of CD4+CD25+ T cells resulted in complete loss of the preventive effect of splenocytes from hCG-treated NOD mice, indicating that CD4+CD25+ T cells play a role in the prevention of autoimmune diabetes by hCG treatment. It is possible that the increase in IL-10 and TGF-β in hCG-treated mice may contribute to the increased ratio of CD4+CD25+/CD4+ T cells, as it is known that these cytokines play an important role in the generation and function of CD4+CD25+ regulatory T cells [30, 31]. There is growing evidence that hCG can have a direct immunomodulatory effect on immune cells [6, 32]. In this regard, we found from in vitro experiments that treatment of immune cells with hCG inhibited IFN-γ and TNF-α production (L.-Y. Khil, H.-S. Jun and H. Kwon, unpublished results). However, we cannot exclude the possibility that hCG, which is a pleiotropic hormone, might also exert its immunosuppressive effects by modulating the level of other reproductive hormones such as progesterone and testosterone [1, 33, 34] and thereby prevent the development of type 1 diabetes in NOD mice.
In this report, we have shown that hCG treatment: (1) downregulates Th1 cells, CD8+ T cells and macrophages; (2) upregulates Th2 cells; and (3) increases the ratio of CD4+CD25+/CD4+ T cells in the spleen and pancreatic lymph nodes, contributing to the prevention of autoimmune diabetes in NOD mice. These results raise the possibility that hCG, a subunit of hCG (e.g. beta hCG) or a molecule associated with hCG [6, 35–39] might be useful in the treatment of diseases characterised by immune deviation or abnormal cytokine release. Although NOD mice are widely used as an experimental model of human type 1 diabetes, caution must be exercised when extrapolating results from mice to humans.
The authors thank B. Sun for islet histology and A. Kyle for editorial assistance. This work was supported by grants from the Canadian Institutes of Health Research (MA9584) and American Diabetes Association (1-04-ISLET-31) to J.-W. Yoon and H.-S. Jun.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.