DRI-C21041 and DRI-C21095 were synthesized either in-house or by WuXi AppTec (Kowloon, Hong Kong) as described before in detail (87, 88). Structures were confirmed by high-resolution mass spectrometry (HRMS) and NMR; purities were confirmed to be ≥95% by HPLC. The anti-CD40L mouse antibody (clone MR-1) was purchased from Bio X Cell (Lebanon, NH, USA).

Mice used for these studies were obtained from Jackson Laboratories (Bar Harbor, ME, USA). The following strains were used: NOD (NOD/ShiLtJ; strain # 001976), C57BL/6 (C57BL/6J; strain # 000664), and DBA/2 (DBA/2J; strain # 000671). All animal studies were reviewed and approved by the University of Miami Institutional Animal Care and Use Committee (IACUC). Procedures were conducted according to the guidelines of the Committee on Care and Use of Laboratory Animals, Institute of Laboratory Animal Resources (National Research Council, Washington, DC, USA). Animals were housed in micro-isolated cages with free access to autoclaved food and water in Virus Antibody Free (VAF) rooms managed by the Department of Veterinary Resources of the University of Miami.

For the diabetes prevention studies, young female NOD mice (n = 5/group) were subcutaneously (s.c.) treated twice daily (b.i.d.) between the ages of 5 and 18 weeks with DRI-C21095 and DRI-C21041 at 20 and 12.5 mg/kg, respectively, administered in 20% hydroxypropyl-β-cyclodextrin (HPβCD) solution as vehicle. A corresponding control group was treated with vehicle only. Mice were monitored for diabetes onset by glycosuria once a week starting from week 10 until 40 weeks of age. Once hyperglycosuria was detected, monitoring was switched to daily glycemia measurements using portable glucometers (Contour Next; Bayer, Mishawaka, USA). Animals with three consecutive readings of blood glucose (BG) >300 mg/dL were considered diabetic and humanely euthanized for tissue analysis. Animals remaining normoglycemic by week 40 of age were euthanized shortly thereafter and blood as well as pancreas, spleen, lymph nodes, and other organs were collected for various analyses.

Mouse splenocytes were collected by mashing spleens through a 70 μm cell strainer, subjected to ACK lysis to remove erythrocytes, and resuspended in Hank’s buffer. Cells were diluted in trypan blue and counted in a hemocytometer. One million viable cells were incubated with Live and Dead solution (Zombie UV, BioLegend, San Diego, CA, USA; cat. no. 423107) in PBS for 20 min at 4°C. Cells were washed and incubated with universal Fc blocker containing anti-mouse CD16/32 (clone 2.4G2) for 20 min at 4°C. Then, cells were incubated for 30 min at room temperature (RT) protected from light in a master mix for surface staining by the following fluorescence-labeled antibodies against NK-1.1 (BD Biosciences, Franklin Lakes, NJ, USA; cat. no. 741715); CD62L (BioLegend; cat. no. 104450); CD8a (BD Biosciences; cat. no. 553036); CD44 (BD Biosciences; cat. no. 562464); CD25 (BioLegend; cat. no. 102016); CD4 (BioLegend; cat. no. 100492); and CD3 (Invitrogen, Carlsbad, CA, USA; cat. no. 56-0032-82) at dilutions according to the recommendations of the suppliers. Cells were washed twice in PBS buffer supplemented with 1% BSA (VWR/Avantor, Radnor, PA, USA; cat. no. 0332), incubated with FIX/PERM (Thermo-Fisher Scientific, Waltham, MA, USA; kit cat. no. 00-5523-00) for 30 min at 4°C, and then washed twice with PERM buffer (Thermo-Fisher Scientific; kit cat. no. 00-5523-00);. Afterwards, cells were subjected to intracellular staining with anti-FoxP3 (Invitrogen; cat. on. 15-5773-82), in PERM buffer for 30 min at 4°C, then washed twice with wash buffer (1% BSA in PBS). After centrifugation, wash buffer was added to resuspend the cells before analysis. Single-color compensation samples for flow cytometry were prepared using the same antibodies with UltraComp eBeads (Thermo Fisher Scientific; cat. no. 01-3333-42). Data were acquired using a Cytek Aurora flow cytometer (Cytek, Fremont, CA, USA) and analyzed with Kaluza software (Beckman Coulter, Brea, CA, USA).

Pancreatic islets used for transplantation were obtained by enzymatic digestion of DBA/2 donor mouse pancreata, followed by purification on density gradients using protocols standardized at the Preclinical Cell Processing and Translational Models Core of our institute (Diabetes Research Institute, DRI) (37, 40). After overnight culture, isolated islets were implanted in fully anesthetized male and female recipient C57BL/6 mice under the kidney subcapsular space, as previously described (37, 40). Prior to transplantation, diabetes was induced with a single dose of streptozotocin (STZ) treatment (200 mg/kg, i.v.) and mice confirmed to be hyperglycemic (i.e., three consecutive non-fasting BG readings > 300 mg/dL) were transplanted with 500 islet equivalents (IEQ) under the kidney capsule. DRI-C21095 and DRI-C21041 treatments (10, 12.5, 15, and 20 mg/kg) were administered s.c. in 20% HPβCD vehicle twice daily (b.i.d.) starting the day before transplant (day -1) until day 30 post-transplant. Mice treated with anti-CD40L antibody (MR-1) received 250 μg administered once a day (s.c.) on days -1, 0, 3, and 7. Transplanted mice remaining normoglycemic at 60 days post-transplant underwent unilateral nephrectomy of the graft-bearing kidney to confirm function of the islet graft and were euthanized shortly thereafter.

In these studies, normoglycemic C57BL/6 mice were transplanted with allogeneic DBA/2 donor islets in the anterior chamber of the eye (ACE). Recipient mice were not pretreated with STZ since the ACE-islet grafts are monitored directly through the transparent cornea. The recipients also expressed green fluorescent protein (GFP) in either macrophages or CD8 T cells; these mice were generated by crossing FVB-Tg(Csf1r-icre)1Jwp/J (strain # 021024; Jackson Laboratories) or C57BL/6-Tg(Cd8a-cre)1Itan/J (strain # 008766; Jackson Laboratories), respectively, with B6.Ai38(RCL-GCaMP3)^lox/lox (strain # 014538; Jackson Laboratories) and breeding for >10 generations to reach homozygosity. Homozygous mice received 50 IEQs in one eye (DBA/2 islets), as previously described in detail (89). Five to seven mice were used per treatment group. MR-1 treatment was as described above. Mice treated with DRI-C21095 (20 mg/kg) or DRI-C21041 (12.5 mg/kg) received s.c. injections twice daily (b.i.d.) in 20% HPβCD vehicle up to day 60 post-transplantation. The structural integrity of ACE-transplanted islets was longitudinally monitored by direct visualization and their survival was assessed by quantitative analysis of individual islets volume, as previously described in detail (52, 89–92). In brief, islets engrafted on top of the iris were mapped in digital images of the eye acquired during the first week after transplantation and revisited during the repeated imaging sessions in these longitudinal studies. The islets were imaged noninvasively through the cornea by confocal microscopy using the 633 nm laser backscatter (reflection) (90), and the islet-infiltrating immune cells (macrophages and T cells) were visualized based on the GFP expression. Three-dimensional (3D) confocal micrographs were acquired using a 20× water immersion objective in z-stacks spanning the full height of the ACE-transplanted islets, and analysis of the individual islet volume was performed in the 3D images. Survival of individual islets was defined when the individual islet maintained its volume above 70% relative to its corresponding baseline (acquired during the first week after transplantation). Immune cell infiltration into the islets was quantified based on the GFP volume inside individual islets and expressed as the percentage of the corresponding islet volume. Volumetric analysis in the images was performed using Volocity software version 6.3.1 (Quorum Technologies Inc.; Puslinch, ON, Canada).

Harvested tissues were immediately placed in a 15 mL Falcon tube containing 4% paraformaldehyde (PFA), incubated at room temperature (RT) for 4 h, and washed three times with PBS for a total of 15 min. They were transferred into a tube containing 30% (w/v) sucrose solution and incubated at 4°C overnight. Eyes were then mounted in Tissue-Tek O.C.T. (Sakura Finetek, Torrance, CA, USA) and snap frozen by placing the tissue mold on dry ice. Tissue sections (14 μm thick) were collected on glass slides and kept at –80°C until analysis. For immunofluorescence staining, samples were subjected to antigen retrieval by standard protocol, washed three times with PBS, and permeabilized for 1 h at RT with 0.3% Triton X-100 in 10% Fc Block (SuperBlock Blocking Buffer, Pierce, Rockford, IL, USA) or normal goat serum. After blocking for at least 1 h, samples were incubated overnight at 4°C with anti-insulin (guinea pig, 1:500; Dako, Carpinteria, CA, USA) and anti-glucagon (rabbit, 1:500; Abcam, Cambridge, MA, USA) primary antibodies. Sections were washed 5 times with PBS plus 0.05% Tween 20 and then incubated with the secondary antibodies: goat anti-rabbit AlexaFluor 488 and goat anti-guinea pig AlexaFluor 647 (both at 1:200; Life Technologies, Carlsbad, CA, USA). Nuclear counterstain was performed by DAPI (1:2000; Life Technologies). Slides were washed another 5 times with PBS plus 0.05% Tween 20 and 2 times with PBS alone before applying mounting media (Clear-MOUNT with TRIS, Electron Microscopy Science, Hatfield, PA, USA) and cover slips. Immunofluorescence imaging was performed using a Leica Stellaris 5 confocal microscope (Leica, Deerfield, IL, USA).

Pharmacokinetics (PK) in mice was evaluated by quantifying plasma concentration in blood samples (50 μL) drawn from the orbital sinus at predefined time points (0, 0.5, 1, 2, 4, 6, and 24 h) following administration of a single dose (30 mg/kg, s.c.) to C57BL/6 mice (n = 4; 10-week-old males and females). Plasma concentrations were obtained using LC/MS performed at the Mass Spectrometry Laboratory, Department of Chemistry, University of Florida (Gainesville, FL, USA) on a ThermoFinnigan, LTQ X mass spectrometer with a ThermoScientific Hypersil GOLD aQ (150 x 2.1 mm, 3 μm) column (Thermo Fisher Scientific, Waltham, MA, USA) using mobile phases containing 0.1% formic acid in water and 0.1% formic acid in acetonitrile. Concentrations were calculated from the corresponding peak areas following a calibration curve in the 0.8 to 100 μg/mL range. Experimental concentration data obtained this way were fitted with a standard one-compartment, first-order absorption, first-order elimination PK model (87, 93) implemented as a custom equation in GraphPad Prism (GraphPad, La Jolla, CA, USA; version 10.1.2) to estimate the elimination rate constant k ₁₀ and the corresponding elimination half-life, t _1/2 = ln(2)/k ₁₀.

The percent of functioning islet allografts in the transplantation models and the diabetes-free survival of NOD mice in the T1D prevention model were shown as Kaplan-Meier survival plots with asterisks denoting statistically significant differences versus the vehicle-treated group as determined by the log rank (Mantel-Cox) test using GraphPad Prism (GraphPad; version 10.1.2) (^* p < 0.05, ^** p < 0.01).

We evaluated the immunosuppressive efficacy of our SMIs targeting the CD40–CD40L PPI (DRI-C21041 and DRI-C21095; Figure 1 ) in an allogeneic islet transplantation model with full MHC-mismatch between donors and recipients. DBA/2 donor islets (MHC Class-I H-2K^d) were transplanted under the kidney capsule of STZ-induced diabetic C57BL/6 (MHC Class-I H-2K^b) mouse recipients that were then treated with various drug regimens. The ability of these recipients to maintain normoglycemia following the transplant was assessed via monitoring of blood glucose levels. DRI-C21095 and DRI-C21041 were administered at doses of 10, 12.5, 15, and 20 mg/kg (b.i.d., s.c.) from days -1 to 30 after transplant. Matched control mice treated with vehicle only (20% HPβCD) or the mouse anti-CD40L monoclonal antibody (MR-1; on days -1, 0, 3, 7 after transplant) were included as negative and positive controls, respectively ( Figure 2 ). These studies revealed a noticeable dose-dependency: the two lowest doses of DRI-C21095 produced no effect, whereas the highest doses of both compounds produced significant prolongation of graft survival and function in 40%–50% of recipients even well after termination of treatment in a manner similar to the positive control MR-1 antibody. Unilateral nephrectomy of the graft-bearing kidney in DRI-C treated mice that remained non-diabetic (normoglycemic) as of day 60 post-transplant resulted in these mice becoming hyperglycemic (BG > 300 mg/dL), thereby confirming the sustained survival and function of the islet allografts in association with the DRI-C treatments.

We also evaluated the immunosuppressive efficacy of our SMIPPIs in a different model of islet transplantation using the anterior chamber of the eye (ACE) as transplant site (52, 89–92). DRI-C21095 (20 mg/kg) and DRI-C21041 (12.5 mg/kg) were administered (b.i.d., s.c.) starting on the day of transplant (day 0) and maintained until day 60 post-transplant ( Figure 3 ). As before, DBA/2 mice were used as islet donors and C57BL/6 mice as recipients to maintain full MHC-mismatch. In these studies, the recipients were normoglycemic and expressed GFP in either macrophages or T cells – this allowed direct in vivo visualization of these cells and the tracking of their infiltration into the ACE-transplanted islets. Survival of the islet allografts was monitored via longitudinal noninvasive imaging and quantitative volume analysis, taking advantage of the transparency of the transplant site ( Figures 4A–C ). In this model, DRI-C21041 was particularly effective: 80% of recipients maintained the structural integrity and function of their islet allografts up to the end of follow-up (115 days post-transplant), significantly longer than did the vehicle-treated and untreated control mice that rejected their grafts with a median of 28 days ( Figure 3 ). Rejection in the controls was similar to that in other controls in our previous ACE studies (median survival of 21 days) (52). While the overall graft survival in the MR-1 treated mice in this study was for some reason less than what we previously observed in a similar model (52), MR-1 treatment still resulted in significant prolongation in the survival of the ACE-transplanted islet allografts compared to the control groups ( Figure 3 ).

A distinctive advantage of the ACE-transplantation is that graft integrity can be monitored by direct noninvasive visualization, and longitudinal images of the same individual islets can be used to document the survival of the graft. This is illustrated in Figures 4A, B with two representative sets of longitudinal images showing the whole eye with pancreatic islets engrafted on top of the iris and the corresponding confocal micrographs of representative islets from an MR-1 treated mouse, which rejected its islets by day 56, and a DRI-C21041 treated mouse that did not reject its islet grafts up to the end of the follow-up on day 115 post-transplant. For the rejected islet shown, a reduction of its volume was already evident by day 35 (POD35), and the islet was no longer visible by day 56 post-transplant due to its complete immune destruction ( Figure 4A ). In contrast, the structural integrity of the islet in the tolerant DRI-C21041 treated mouse was maintained, and its volume remained unchanged until the end of the experiment ( Figure 4B ). In vivo volumetric analysis in confocal z-stacks (i.e., in 3D) of the individual islets in the DRI-C21041 and MR-1 treated mice highlighted the difference in the average change in islet volume between the two groups over time ( Figure 4C ). Further, ex vivo analysis in such islets (after necropsy) by immunostaining for glucagon and insulin also confirmed the sustained graft function in the DRI-C21041 treated recipients until the end of the studies on day 115 post-transplant ( Figure 4D ).

Because islets transplanted in the ACE can be followed via direct noninvasive imaging, we monitored in situ the graft infiltration by GFP-expressing macrophages and T cells (see Methods for details) and quantified the in vivo infiltration kinetics in mice that either rejected or tolerated their islet allografts. Regardless of treatment, macrophages initially infiltrated islets to similar degrees in recipients that ended up either rejecting or tolerating their allografts, but their relative abundance noticeably increased in rejected islets whose volume was progressively reduced ( Figure 5 ). Additional studies in mice with GFP-expressing T cells further showed that blocking the CD40–CD40L interaction with either our SMIs or MR-1 decreased the T cell infiltration into the islet allografts. Interestingly, while the abundance of macrophages within tolerated islets remained relatively unchanged in the DRI-C21041 and MR-1 treated recipients until the end of follow-up on post-transplant day 112 ( Figure 5 ), tolerant mice treated with our DRI-C compounds had markedly reduced initial T cell infiltration in the first 21 days after the transplant compared to the MR-1 treated counterparts, and this persisted until the end of the follow-up on day 112 post-transplant in DRI-C21041 treated mice ( Figure 6 ).

In addition to evaluating the immune suppressive effects of our SMIPPIs in allogeneic islet transplant models, we also assessed their ability to prevent the onset of autoimmune T1D in NOD mice in an exploratory study using female animals as customary due to their higher incidence rate of spontaneous diabetes. In light of previous publications studying CD40–CD40L inhibition in the same model (54, 60, 94), we administered treatments here from 5 to 18 weeks of age and monitored the diabetes incidence by regular glycemia measurements during an extended follow-up period until the mice reached 40 weeks of age. Results indicated that while hyperglycemia (i.e., diabetes onset) occurred in the typical 80% range in the control-treated animals, the incidence was reduced in the groups treated with our SMIs: it was reduced to 60% and 25% in the DRI-C21041- and DRI-C21095-treated mice, respectively ( Figure 7 ). The median age of diabetes-free survival was 26 weeks in the vehicle-treated controls versus 31 weeks in those treated with DRI-C21041 and >40 weeks in the DRI-C21095 treated mice since they did not cross the 50% threshold (“undetermined”). We also performed exploratory immunophenotyping to identify indications of possible lasting treatment effects on immune cell subpopulations in the mice that did not become diabetic. Results can be considered as indicative only since they are from a small number of mice (one or two per group) and well after the termination of treatment (at week 40 vs 18 of age), nevertheless, they suggested possible persistent reduction in natural killer cells (NK; NK1.1⁺CD3^–) and central memory T cells (CD4 and CD8 Tcm; CD44⁺CD62L⁺) and a concomitant increase in regulatory T cells (Treg; CD25⁺FoxP3⁺) ( Supplementary Figure 1 ).

Because DRI-C21041 performed unexpectedly well in the ACE allo-transplant model compared to the ester-containing DRI-C21095 ( Figure 1 ), which has shown higher potency in vitro (87, 88) and performed well in the kidney allo-transplant and diabetes prevention models, we carried out a brief pharmacokinetic (PK) evaluation in C57BL/6 mice. A single dose (30 mg/kg) was administered s.c. (same route as in the studies used for efficacy assessments), and plasma concentrations were quantified in samples collected at predefined time-points after the injection (0, 0.5, 1, 2, 4, 6, and 24 h). Results indicated that DRI-C21041 has an acceptably long half-life (t _1/2 = 10.8 h) ( Figure 8 ), considerably better than the ester-containing DRI-C21045 that we evaluated in an earlier study (t _1/2 ≈ 2 h) (87). This is not surprising as ester-containing compounds are susceptible to quick metabolic degradation by abundantly present esterases, often resulting in shorter half-lives – an effect further exacerbated in rodents as they tend to metabolize ester-containing drugs (especially aliphatic esters) much faster than humans due to their known higher esterase activity (95).