Differential outcomes in prediabetic vs. overtly diabetic NOD mice nonmyeloablatively conditioned with costimulatory blockade

Abstract

Objective: Autoimmune diabetes can be reversed with mixed chimerism. However, the myelotoxic agents currently required to establish chimerism have prevented the translation of this approach to the clinic. Here, we investigated whether multimodal costimulatory blockade would enhance chimerism and promote islet allograft tolerance in spontaneously diabetic nonobese diabetic (NOD) mice.

Materials and methods: Prediabetic and spontaneously diabetic NOD mice were preconditioned with anti-CD8 monoclonal antibody before conditioning with 500 cGy total body irradiation and transplantation with 30 × 10(6) B10.BR bone marrow cells. Overtly diabetic animals were conditioned similarly and transplanted with 300 to 400 B10.BR islets. After irradiation, both groups of recipients were treated with anti-CD154, anti-OX40L, and anti-inducible T-cell costimulatory monoclonal antibodies. Urine, blood glucose levels, and chimerism were monitored.

Results: Conditioning of NOD mice with costimulatory blockade significantly enhanced engraftment, with 61% of mice engrafting at 1 month. Eleven of 12 chimeric animals with engraftment at 1 month remained diabetes-free over a 12-month follow-up, whereas nonchimeric animals progressed to diabetes. In contrast, similar conditioning prolonged islet allograft survival in only 2 of 11 overtly diabetic NOD recipients. Chimerism levels in the 9 islet rejector animals were 0%.

Conclusions: Although nonmyeloablative conditioning reversed the autoimmune process in prediabetic NOD mice, the same regimen was significantly less effective in establishing chimerism and reversing autoimmune diabetes in spontaneously diabetic NOD mice.

Figure 1. Co-stimulatory blockade with anti-OX40L and anti-ICOS mAbs enhances B10.BR allogeneic engraftment in NOD mice

A: Mice were preconditioned with anti-CD8 mAb on day -3. On day 0, recipients were transplanted with 30 × 10⁶ B10.BR BMC 4 to 6 h after TBI as well as single daily dose of anti-CD154, anti-OX40L, and anti-ICOS mAbs on days 0, 1, 2, and 3. B: NOD mice received BMC with vehicle (n = 10), BMC with anti-CD8, anti-CD154, anti-OX40L (n = 13), or BMC with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS mAbs in addition to 500 cGy TBI (n = 19). Recipient mice were monitored for engraftment 30 days after BMC transplantation. C: NOD mice received BMC with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS in addition to a 500 cGy TBI dose (n = 12). Animals with transient donor chimerism were monitored monthly for 5 months (n = 6) (○). Animals with durable long-term chimerism (n = 6) (■) were monitored monthly for 10 months. Animals with no engraftment were monitored for 2 months (n=7) (▲). Data represents the means ± SE of combined data obtained from independent determinations. D–F: Multilineage typing of representative pre-diabetic chimeric NOD recipient treated with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS, donor BM and 500 cGy TBI and control B10.BR recipient in both bar graph (D) and dotplot format (E–F). Multilineage data were derived from PB 2 months after BM transplantation and analyzed using both lymphoid and myeloid gates.

Figure 1. Co-stimulatory blockade with anti-OX40L and anti-ICOS mAbs enhances B10.BR allogeneic engraftment in NOD mice

A: Mice were preconditioned with anti-CD8 mAb on day -3. On day 0, recipients were transplanted with 30 × 10⁶ B10.BR BMC 4 to 6 h after TBI as well as single daily dose of anti-CD154, anti-OX40L, and anti-ICOS mAbs on days 0, 1, 2, and 3. B: NOD mice received BMC with vehicle (n = 10), BMC with anti-CD8, anti-CD154, anti-OX40L (n = 13), or BMC with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS mAbs in addition to 500 cGy TBI (n = 19). Recipient mice were monitored for engraftment 30 days after BMC transplantation. C: NOD mice received BMC with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS in addition to a 500 cGy TBI dose (n = 12). Animals with transient donor chimerism were monitored monthly for 5 months (n = 6) (○). Animals with durable long-term chimerism (n = 6) (■) were monitored monthly for 10 months. Animals with no engraftment were monitored for 2 months (n=7) (▲). Data represents the means ± SE of combined data obtained from independent determinations. D–F: Multilineage typing of representative pre-diabetic chimeric NOD recipient treated with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS, donor BM and 500 cGy TBI and control B10.BR recipient in both bar graph (D) and dotplot format (E–F). Multilineage data were derived from PB 2 months after BM transplantation and analyzed using both lymphoid and myeloid gates.

Figure 1. Co-stimulatory blockade with anti-OX40L and anti-ICOS mAbs enhances B10.BR allogeneic engraftment in NOD mice

A: Mice were preconditioned with anti-CD8 mAb on day -3. On day 0, recipients were transplanted with 30 × 10⁶ B10.BR BMC 4 to 6 h after TBI as well as single daily dose of anti-CD154, anti-OX40L, and anti-ICOS mAbs on days 0, 1, 2, and 3. B: NOD mice received BMC with vehicle (n = 10), BMC with anti-CD8, anti-CD154, anti-OX40L (n = 13), or BMC with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS mAbs in addition to 500 cGy TBI (n = 19). Recipient mice were monitored for engraftment 30 days after BMC transplantation. C: NOD mice received BMC with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS in addition to a 500 cGy TBI dose (n = 12). Animals with transient donor chimerism were monitored monthly for 5 months (n = 6) (○). Animals with durable long-term chimerism (n = 6) (■) were monitored monthly for 10 months. Animals with no engraftment were monitored for 2 months (n=7) (▲). Data represents the means ± SE of combined data obtained from independent determinations. D–F: Multilineage typing of representative pre-diabetic chimeric NOD recipient treated with anti-CD8, anti-CD154, anti-OX40L, and anti-ICOS, donor BM and 500 cGy TBI and control B10.BR recipient in both bar graph (D) and dotplot format (E–F). Multilineage data were derived from PB 2 months after BM transplantation and analyzed using both lymphoid and myeloid gates.

Figure 2. Incidence of Diabetes in age-matched naïve and conditioned NOD mice

The incidence of diabetes was compared between the various treatment groups listed in the figure legend. The most robust diabetes-prevention was in the group that was durably chimeric. The progression of diabetes was monitored weekly in control mice and all treatment groups. Mice with blood glucose levels greater than 250 mg/dL after 3 consecutive measurements were considered diabetic.

Figure 3. Diabetes is reversed in some conditioned diabetic mice transplanted with islet allografts

A: Diabetic NOD recipients were preconditioned with anti-CD8 mAb on day -3, transplanted with 30×106 B10.BR donor BMC on day 0 and received a 500 cGy TBI dose 6 h later. Twenty-four hours after BM transplant, NOD mice were transplanted with 300–400 B10.BR donor islet allografts, and on days 0–3 received anti-CD154, anti-OX40L and anti-ICOS. B: Blood glucose measurements were carried out in islet allograft acceptor mice (○) (n = 2), islet allograft rejector mice (●) (n = 9), and control mice not transplanted with donor islets (△) (n = 2) to assess islet allograft survival in transplanted recipients. Rejection was confirmed with 3 consecutive non-fasting blood glucose readings over 250 mg/dL. Values are expressed as mean ± SE.

Figure 4. Histological analysis, donor chimerism, and glucose challenge in islet acceptor NOD mice

Hematoxylin and eosin staining was carried out on pancreatic tissue obtained either 6 months after islet transplantation in acceptor animals (A–B) or 2–3 months in rejector animals (C–D). H&E staining revealed mononuclear infiltration with preservation of islet architecture (200×) in pancreatic tissue from acceptor animals, while the majority of beta cells from rejected pancreatic tissue showed strong mononuclear infiltration (C) and large to small vacuoles indicating destruction of islet architecture (D) (200×). E: Islet allograft transplanted animals were monitored for donor chimerism 1 to 6 months post-BMT in peripheral blood of islet acceptor animals (○) (n = 2) and islet rejector animals (●) (n = 9). F: Glucose tolerance tests were carried out in islet allograft acceptor mice (○) (n = 2) 16 weeks after B10.BR BM and islet allograft transplantation. Normal naïve C57BL/6 mice (△) (n = 5), pre-diabetic naïve NOD mice (■) (n = 9), and diabetic NOD mice (●) (n = 4) were evaluated as controls for comparison. Values are expressed as mean ± SE (blood glucose (mg/dL).

Figure 4. Histological analysis, donor chimerism, and glucose challenge in islet acceptor NOD mice

Hematoxylin and eosin staining was carried out on pancreatic tissue obtained either 6 months after islet transplantation in acceptor animals (A–B) or 2–3 months in rejector animals (C–D). H&E staining revealed mononuclear infiltration with preservation of islet architecture (200×) in pancreatic tissue from acceptor animals, while the majority of beta cells from rejected pancreatic tissue showed strong mononuclear infiltration (C) and large to small vacuoles indicating destruction of islet architecture (D) (200×). E: Islet allograft transplanted animals were monitored for donor chimerism 1 to 6 months post-BMT in peripheral blood of islet acceptor animals (○) (n = 2) and islet rejector animals (●) (n = 9). F: Glucose tolerance tests were carried out in islet allograft acceptor mice (○) (n = 2) 16 weeks after B10.BR BM and islet allograft transplantation. Normal naïve C57BL/6 mice (△) (n = 5), pre-diabetic naïve NOD mice (■) (n = 9), and diabetic NOD mice (●) (n = 4) were evaluated as controls for comparison. Values are expressed as mean ± SE (blood glucose (mg/dL).

Figure 5. Expression of CD8⁻/CD4⁺/CD25⁺/FoxP3⁺ in PB, BM and spleens of islet allograft rejector animals and recipients carrying long-term surviving grafts

Peripheral blood, BM, and splenocytes were isolated from naïve NOD, diabetic NOD, NOD recipients with long-term surviving islet allografts, or from NOD recipients that rejected their islet allografts and stained for surface markers: CD8, CD4, and CD25. Cells were fixed and permeabilized overnight and stained for FoxP3 the following day. Stained cells were then analyzed by flow cytometry. Percentages (A) and absolute number (B) of CD8⁻/CD4⁺/CD25⁺/FoxP3⁺ T_reg and newly activated CD8⁻/CD4⁺/CD25⁺/FoxP3⁺ T_eff. The values are expressed as mean ± SE of independent experiments from 3 naïve animals, 3 diabetic animals, 2 islet acceptor animals, and 4 islet rejector animals.

Figure 5. Expression of CD8⁻/CD4⁺/CD25⁺/FoxP3⁺ in PB, BM and spleens of islet allograft rejector animals and recipients carrying long-term surviving grafts

Peripheral blood, BM, and splenocytes were isolated from naïve NOD, diabetic NOD, NOD recipients with long-term surviving islet allografts, or from NOD recipients that rejected their islet allografts and stained for surface markers: CD8, CD4, and CD25. Cells were fixed and permeabilized overnight and stained for FoxP3 the following day. Stained cells were then analyzed by flow cytometry. Percentages (A) and absolute number (B) of CD8⁻/CD4⁺/CD25⁺/FoxP3⁺ T_reg and newly activated CD8⁻/CD4⁺/CD25⁺/FoxP3⁺ T_eff. The values are expressed as mean ± SE of independent experiments from 3 naïve animals, 3 diabetic animals, 2 islet acceptor animals, and 4 islet rejector animals.