The Role of Human CD46 in Early Xenoislet Engraftment in a Dual Transplant Model

Abstract

Background

Membrane cofactor protein CD46 attenuates the complement cascade by facilitating cleavage of C3b and C4b. In solid organ xenotransplantation, organs expressing CD46 have been shown to resist hyperacute rejection. However, the incremental value of human CD46 expression for islet xenotransplantation remains poorly defined.

Methods

This study attempted to delineate the role of CD46 in early neonatal porcine islet engraftment by comparing Gal-knocked out (GKO) and hCD46-transgenic (GKO/CD46) islets in a dual transplant model. Seven rhesus macaques underwent dual transplant and were sacrificed at 1 hour (n=4) or 24 hours (n=3). Both hemilivers were recovered and fixed for immunohistochemistry (CD46, insulin, neutrophil elastase, platelet, IgM, IgG, C3d, C4d, CD68, Caspase 3). Quantitative immunohistochemical analysis was performed using the Aperio Imagescope.

Results

Within 1 hour of intraportal infusion of xenografts, no differences were observed between the two types of islets in terms of platelet, antibody or complement deposition. Cellular infiltration and islet apoptotic activity were also similar at 1 hour. At 24 hours, GKO/CD46 islets demonstrated significantly less platelet deposition (p=0.01) and neutrophil infiltration (p=0.01) compared to GKO islets. In contrast, C3d (p=0.38) and C4d (p=0.45) deposition was equal between the two genotypes.

Conclusions

Our findings suggest that expression of hCD46 on NPIs potentially provides a measurable incremental survival advantage in vivo by reducing early thrombo-inflammatory events associated with instant blood mediated inflammatory reaction (IBMIR) following intraportal islet infusion.

Introduction

Solid organ transplantation is the gold standard treatment for patients with end stage organ failure. The shortage of suitable donor organs, however, remains a significant obstacle in the field. Pig to human xenotransplantation offers the prospect of an unlimited supply of organs and therefore a potential solution to the donor shortage. Though historically viewed with skepticism given the myriad biologic incompatibilities between species,¹ xenotransplantation has garnered increasingly legitimate enthusiasm in the recent era, spurred in part by continuous improvements in genome editing technologies.² These technologies offer the promise of an ideal donor pig that is pathogen-free, physiologically compatible with humans, and relatively immunologically acceptable.^3-5 However, the assessment of specific gene modifications amongst numerous uncontrolled factors has been challenging, making rational selection of necessary donor modifications difficult.

Human membrane cofactor protein (hCD46) is among one of the first genes targeted for gene editing in the field of xenotransplantation, and is thus ripe for specific assessment. It belongs to a class of proteins called complement regulatory proteins (CRP) expressed on the surface of vascular endothelial cells. CD46 specifically attenuates the complement cascade by facilitating cleavage of C3b and C4b.⁶ However, CRPs exhibit homologous restriction,⁷ meaning that in the context of xenotransplantation, the recipient effector complement mechanisms cannot be efficiently countered by the donor CRPs. Xenografts invariably succumb to hyperacute rejection in this context. Hence, porcine organs that express human CD46 and thus can regulate human complement activation have a theoretical advantage over wild type organs. In solid organ xenotransplantation, organs expressing human CD46, when used in combination with multimodal immune suppression and other genetic modifications, have been shown to resist hyperacute rejection.^8,9 In fact, most promising pre-clinical studies on solid organ xenotransplantation (heart, lung, kidney and liver) have been performed using donors expressing human CRPs.^10-13

Despite the pre-clinical successes with CRP expressing organs, the incremental value of human CD46 expression has been challenging to assess, and it has not been directly examined in isolation in islet xenotransplantation. Unlike vascularized organs, islet xenografts face a different set of immunological barriers, one of which is related to the intraportal delivery of islet transplants and its invocation of an instant blood mediated inflammatory response (IBMIR). IBMIR was initially described in clinical islet transplantation, where rapid destruction of autologous or allogeneic islets was observed following exposure to human blood in vitro¹⁴ and in vivo,¹⁵ as well as in xenotransplantation.^16,17 A multifactorial process, IMBIR is generally considered to involve activation of platelets, as well as coagulation and complement cascades, which collectively lead to leukocyte infiltration. Many have postulated a role of CD46 (together with other CRPs) in limiting the extend of IBMIR.^18-20 A number of groups have even utilized porcine islets that express human CD46 or other CRPs such as CD55 and CD59, in non-human primate models with durable graft survival.^21-23 However, a head-to-head comparison of porcine islets lacking human CD46 and those expressing human CD46 has not been performed, and although the rationale for CD46 expression is sound, the redundancy of CRPs and the many other factors conspiring against a cellular xenograft make the question of the relative importance of a single gene modification reasonable.

Our group has previously established a dual islet transplant model,^24,25 in which we exploit the bifurcated portal venous system to compare the engraftment immunobiology of islets of two different genotypes within a single recipient. This model allows objective and rigorous measurements of the early immune responses immediately following portal delivery of islets, utilizing quantitative histological analysis, and controlling for the biology of the recipient, the quality of the islet prep and the exposure to immunosuppressive drugs for each paired experiment. In this current study, we attempted to explicitly delineate a role of human CD46 expression in early neonatal porcine islet engraftment by comparing Gal-knocked out (GKO) and human CD46-transgenic (GKO/CD46) islets in the dual transplant model.

Materials and Methods

Neonatal porcine islet procurement, isolation, and culture

Neonatal piglets of α1,3-galactosyl transferase knockout background, with (GKO/CD46) or without human CD46 transgene (GKO), were obtained from Revivicor Inc. (Blacksburg, VA, USA). Genotype of piglets was confirmed by both genetic sequencing and flow cytometry looking at human CD46 expression on porcine PBMCs (Figure 1D). Commercially available wildtype piglets were used as control for the weight of pancreas and NPI yield. Those wildtype piglets were acquired from Looper Farm (Granite Falls, NC, USA). Piglets between 1 to 7 days of age were transported to our institution and underwent terminal pancreatectomy. A previously described modified Korbutt technique²⁶ was used to isolate porcine islet cells from neonatal pancreatic tissue recovered. Neonatal porcine islets (NPIs) were maintained in culture using NPI culture media (Corning, Corning, NY, USA) supplemented with penicillin/streptomycin (Corning, Corning, NY, USA) for 6–8 days and were quantified in islet equivalents (IEQs) on day of transplantation using dithizone (Sigma-Aldrich, St. Louis, MO, USA). In addition, NPI viability was evaluated using fluorescein diacetate (Sigma-Aldrich, St. Louis, MO, USA) and propidium iodide (Sigma-Aldrich, St. Louis, MO, USA) live-dead counterstaining on the day of transplantation. All NPI preparations underwent endotoxin testing to rule out bacterial contamination prior to transplantation.²⁶ All procedures were performed in accordance with the “Guide for the Care and Use of Laboratory Animals” (Institute of Laboratory Animal Resources, National Research Council, DHHS), and approved by Duke Institutional Animal Care and Use Committee (IACUC).

Figure 1.

Weight of pancreas harvested (A) and yield of islet cells per gram of pancreas tissue (B) between the two genotypes and wild type piglets were not significantly different. (C) Technical success for islet segregation by contralateral hemiliver was confirmed by CD46 staining at both 1hr and 24hrs (p<0.01). (D) Genotype of NPIs was confirmed by sequencing (data not shown, by Revivicor Inc.) and flow cytometry using piglet PBMCs. Representative hCD46 IHC slides of the hemilivers where GKO islets were delivered (E) and where GKO/CD46 islets were delivered (F) at 1 hr.

Dual islet transplantation

Ten rhesus macaques (Macaca mulatta) weighing 3–10 kg were selected as NPI xenograft recipients. No attempt was made to select animals based on xeno-specific antibody levels. All procedures and care of animals was performed in accordance with the Guide for the Care and Use of Laboratory Animals and approved by the Emory University and Duke University Institutional Animal Care and Use Committees. Individual islet infusions were prepared using 20mL CMRL 1066 without phenol red (Corning, Corning, NY, USA), 100 units of heparin sodium, and 1.5 mg/kg of etanercept (Enbrel Immunex Corp, Thousand Oaks, CA). Islet transplantation was performed as previously described.^24,25 In brief, two islet preparations were made for each transplant recipient, one for each islet phenotype, and delivered to hemi-livers via left or right portal vein. The side of the infusion was randomized to control for any potential differences segregating with lateralization. Ten animals underwent dual transplant of GKO and GKO/CD46 islets without immunosuppression with endpoints of 1 hour (n=5) and 24 hours (n=5) (Table 2).

Table 2.

A table of experiments performed in the present study. Five animals each in 1-hour, 24-hour and 7-day groups. Technical success was defined as successfully delivering GKO or GKO/CD46 islet preparation to its corresponding hemiliver without crossing over. We were able to achieve 4 and 3 successes in 1-hour and 24-hour groups respectively.

Timepoints	Immunosuppression	# of animals	# of technical successes
1 hour	N	5	41
24 hours	N	5	32
7 days	Y	5	23

^1:RNH5 no islets on L side

^2:FA10 no islets on L side, RUS8 few islets on L side, mixed genotypes on R side

^3:G25G no islets on L side, H15X no islets on R side, G20O mixed genotypes on L side

Extended dual islet transplantation

Rhesus macaques were rendered diabetic through streptozocin induction (STZ, 1250 mg/m² IV; Sigma, St Louis, MO, USA) and diabetes was confirmed by significant reduction and poor responsiveness of rhesus C-peptide levels (C-peptide ELISA assay; Abcam, Cambridge, MA, USA) to dextrose bolus. As NPIs generally take weeks to months to reach maturity, our 7-day extended dual islet transplantation was not designed to achieve normoglycemia in NHP recipients. The use of diabetic recipients was not essential for our study but selected to better simulate the clinical scenario. Additionally, the host immune responses to xenograft in a diabetic animal and a normoglycemic animal may differ.

Five diabetic recipients underwent GKO and GKO/CD46 xenoislet dual transplantation (Table 2) and given an immunosuppressive regimen consisting of Basiliximab at 0.3mg/kg on day of transplant and Post-operative day (POD) 2, Belatacept at 20mg/kg on POD 0 and 4, and Tacrolimus at 0.05mg/kg twice a day starting at POD 1 with a target trough level of 8–12ng/ml. As neonatal porcine islet requires at least 3 weeks to mature in vivo and recipients following NPI infusion do not achieve insulin independence until weeks to months posttransplantation, recipient blood glucose measurements were not indicative of graft function and were taken only to monitor for clinical consequences of hypo- or hyperglycemia.

Tissue collection and immunohistochemistry

Islet transplant recipients were euthanized at 1-hour, 24-hour or 7-day following islet cell infusion and livers were recovered for detailed immunohistochemical (IHC) analysis. The explanted liver was divided into right and left lobes. Approximately 1.0cm of central liver tissue was excluded from analysis to minimize crossover contamination between islet phenotypes. Each hemiliver was divided into 5 sections and preserved in 10% neutral buffered formalin for histological evaluation, as described ²⁵. Tissue samples were embedded in paraffin and underwent subsequent sectioning and hematoxylin and eosin (H&E) staining for morphologic observation. If adequate numbers of islet cell clusters (≥5) were seen, these tissue slides subsequently underwent IHC staining for human CD46 to ensure complete segregation of genotypes before additional immunohistochemical analyses were performed. Between the 1-hour and 24-hour groups, the number of islets analyzed on each histology slides ranged from 5 to 102, with a mean of 23.1, a median of 22.5 and a mode of 23 (Figure S1). A list of antibodies used in this study are shown here (Table 1). Whole slide digital images were captured using the Aperio ScanScope XT (Leica Biosystems, Vista, CA, USA) slide scanner system with 20x objective magnification. Quantitative IHC analysis was performed using Aperio Imagescope (Leica Biosystems, Vista, CA, USA) digital pathology software.²⁵ Individual slides were scanned and areas of the liver parenchyma where islet clusters were deposited were manually selected and analyzed using an optimized positive pixel algorithm to obtain a percent pixel positivity of the individual stain per islet. The sum of individual islets within a hemiliver was compiled to compute a percent positive staining representative of each hemiliver, termed the pixel positivity index. Data for a specific stain were excluded from an animal if the total number of islets found in a hemiliver was less than five.

Table 1.

A list of antibodies used in the present study.

Target of antibody	Cross-reactivity reported by manufacturer	Manufacturer	Clone
CD3	Human	Dako, Carpinteria, CA, USA	A0452
CD46	Human, Rabbit	Abcam, Cambridge, MA, USA	Ab108307
Insulin	Rat, Horse, Canine, Rabbit, Guinea pig, Monkey, Human, Bovine, Feline, Sheep, Pig	Sigma-Aldrich, St. Louis, MO, USA	I2018
C3d	Human	Abcam, Cambridge, MA, USA	Ab136916
C4d	Human	American Research Products, Waltham, MA, USA	12-5000
IgG	Monkey	Sigma-Aldrich, St. Louis, MO, USA	A2054
IgM	Monkey	KPL, Gaithersburg, MD, USA	071-11-031
Neutrophil elastase	Human	Dako, Carpinteria, CA, USA	M0752
CD68	Human	Dako, Carpinteria, CA, USA	M0814
CD61	Human	Dako, Carpinteria, CA, USA	M0753
Cleaved caspase 3	Human, Mouse, Rat, Monkey	Cell Signaling Technology, Danvers, MA, USA	9664s

Statistical analysis

Pixel positivity index was calculated within each hemiliver and paired against the contralateral hemiliver positivity for each animal. As positivity was measured in a percentage format, a paired t-test with logarithmic transformation was performed to analyze stain differences between islet preparations at the study time points of 1 and 24 hours. An unpaired t-test was performed to measure differences in positivity from 1 to 24 hours for GKO and GKO/CD46. Statistical analysis was performed using GraphPad Prism (GraphPad Software Inc., La Jolla, CA, USA) version 7 statistical software. A p ≤ 0.05 was determined to be statistically significant.

Results

Human CD 46 expression on porcine cells does not change the weight or cellular composition of the pancreas

To determine whether that the lack of Gal and/or the presence of hCD46 significantly influenced pancreatic development, we compared the weights of individual pancreata and the yield of islets per gram of pancreatic tissue among wild type, GKO and GKO/CD46 piglets (Figure 1A-B). The weight of pancreas at recovery (p=0.84) and NPI yield (p=0.81) were comparable (GKO: 3.02±0.23 g/piglet and 11,829±2,164 IEQs/g; GKO/CD46: 3.23±0.09 g/piglet and 12,976±1,866 IEQs/g; Wild type: 3.07±0.38 g, 11,226±1,927 IEQs/g).

Islet segregation was achieved by the dual islet transplant model

We performed a total of fifteen dual islet transplantations and collected tissue for histological analysis at 1-hour (n=5), 24-hour (n=5) and 7-day (n=5). Technical success was defined as successfully delivering GKO or GKO/CD46 NPIs to its corresponding hemiliver without islet crossover. We were able to achieve 4 and 3 successes in 1-hour and 24-hour groups respectively (Table 2) and only technical successes were included in formal statistical analysis.

Among the five animals that received extended dual islet transplant, we were only able to complete two technical successes. Two animals did not have NPIs seen on histology in either liver lobe and one had significant genotype crossover in the left lobe. Given that we were able to achieve only two technical successes in the 7-day group (Table 2), this group was subsequently removed for further statistical analysis due to small sample size. The results of the long-term animals are included descriptively.

IHC staining of human CD46 was analyzed to ensure adequate segregation of GKO and GKO/CD46 NPIs in the dual transplant model (Figure 1C-F). At 1-hour (p<0.01, n=4) and 24-hour (p<0.01, n=3), human CD46 staining was significantly more positive on the side where GKO/CD46 NPIs were delivered. As mentioned earlier, only two technical successes were achieved in the 7-day group and histology confirmed complete segregation of GKO and GKO/CD46 NPIs to opposite lobes of the liver in those two animals.

IBMIR response was not abolished by human CD46 expression on NPI alone

Despite expression of human CD46 on NPIs, IBMIR was evident in vivo following portal infusion of NPIs. Heavy deposition of platelets, IgM, IgG, C3d and C4d were observed at 1-hour post-transplant, coinciding with abundant graft-infiltrating neutrophils (Figure 2A, D, G). In contrast, only scant infiltration of macrophages (CD68 stain) was seen at 1-hour within the xenografts. By 24-hour post islet infusion, IgM (p=0.02) and IgG (p=0.04) staining were significantly reduced, while the presence of graft-infiltrating macrophages increased (p<0.01).

Figure 2.

Development of instant blood mediated inflammatory response (IBMIR) observed in vivo over time in GKO/CD46 NPIs. Representative neutrophil IHC slides of hemilivers (GKO/CD46 side) procured at 1-hour (A), 24-hour (B) and 7-day (C). Representative platelet immunohistochemistry slides at various time points shown in (D-F). Neutrophil infiltration (G) and platelets deposition (H) were heavily featured early in IBMIR but their involvement diminished over time. (I) Other components of IBMIR observed at various early time points following portal infusion of islets.

Platelet deposition was minimal on histology by 7 days (Figure 2G), while macrophage and T cells appeared more developed in keeping with the burgeoning adaptive phase of the host immune response.

Human CD46 expression reduced xenograft platelet deposition and neutrophil infiltration at 24 hours

Detailed IHC analysis was undertaken to investigate differences in host immune responses to GKO and GKO/CD46 NPIs. At 1-hour post-transplant, we observed comparable platelet deposition (p=0.85) and neutrophil infiltration (p=0.81) between GKO and GKO/CD46 NPIs. However, GKO/CD46 NPIs suffered significantly less platelet deposition (p=0.01) and neutrophil infiltration (p=0.01) compared to GKO islets at 24-hour posttransplant (Figure 3).

Figure 3.

The role of hCD46 in islet xenotransplantation. Significantly less neutrophil infiltration (p=0.01) and platelet deposition (p=0.01) were observed with GKO/CD46 islets compared to GKO islets at 24 hours following intraportal infusion, even though no differences was seen at 1 hour (A, D). Representative immunohistochemistry images on both L and R hemilivers precured from the same animal at 24 hours stained for neutrophil elastase (B-C) and platelet (E-F).

As CD46 downregulates the complement cascade, we hypothesized that the aforementioned differences in platelet deposition and neutrophil infiltration were due to CD46’s regulatory effect on the host complement system and its associated chemotaxtic and opsonizing effects. C4d was applied as a measure for complement activation through the classical pathway and C3d, alternative pathway respectively. However, no difference in C4d or C3d were observed between two types of NPIs at 1-hour (C4d: p=0.21; C3d: p=0.30) and 24-hour (C4d: p=0.45; C3d: p=0.38). Similarly, analysis of the two types of NPI in antibody deposition, T cell and macrophage infiltration and NPI apoptosis at both 1-hour and 24-hour did not reveal any statistically significant differences (Figure 4).

Figure 4.

Other measured variables, including T cell and macrophage infiltration, C3d, C4d, IgM and IgG deposition were comparable between GKO and GKO/CD46 islets at both 1hour (A) and 24 hours (B).

Discussion

One of the major hurdles to islet transplantation is the early loss of islets immediately following infusion of cells into the portal venous system. One study estimated that at least 25% of a transplanted islet mass is lost within minutes.²⁷ This loss is associated with IBMIR, which is believed to be largely triggered by unopposed activation of the complement and coagulation cascades. Here we describe a series of experiments designed to better understand the role of human CD46 in islet xenotransplantation. In order to isolate the effect of human CD46 expression, we applied the dual transplant model and performed a head-to-head comparison between GKO NPIs deficient in human CD46 and those expressing human CD46. We demonstrate that expression of the human complement regulatory protein CD46 is associated with a significantly reduction in platelet deposition and neutrophil infiltration during early engraftment. Although IBMIR remains evident, these data suggest that the effects of CD46 incompatibility are contributory to the barriers aligned against xenoislet engraftment, and that this modification is a rational one in the development of a uniform xenoislet donor.

Like the two early studies in this dual transplant series, we observed a nonspecific milieu of antibody, complement and coagulation immediately following intraportal infusion of NPIs.^24,25 At 1-hour, human CD46 expression on NPIs did not provide any perceived advantage at this stage. We postulate that this initial wave of damage may be mediated by a nonspecific innate immune response that would likely require pharmacological intervention or additional donor genetic engineering.

CD46 attenuates the complement cascade by acting as a cofactor for serum factor I, a serine protease that by cleaving C3b and C4b. Thus, NPIs expressing host-compatible CD46 could be expected to experience less graft damage related to complement activation. However, there is immense redundancy in the factors leading to complement activation, and thus it is not surprising that we did not observe a measurable difference in complement activation at both 1-hour and 24-hour posttransplant between GKO and GKO/CD46 NPIs. Furthermore, we acknowledge that given the numerous biological perturbations referable to islet infusion into the portal system, and the modest quantitative capacity of IHC, this may be attributed to a saturation effect. The complement activation could be simply too overwhelming that the local expression of human CD46 on NPIs is not sufficient in reversing the local deposition. Additionally, C3d and C4d staining during early engraftment may not accurately reflect the magnitude of complement activation, especially in the setting of IBMIR. Because NPIs expressing human CD46 can break down C4b into membrane-bound C4d more efficiently, conceivably, C4d may become paradoxically more prominent in xenograft where local complement cascade is more controlled. That the effects seen in this series of experiments relate more to cellular accumulation, it suggests that a reduction in the chemotactic effects of non-bound complement breakdown products may be the most relevant in this setting. Interestingly, we have previously reported on increased staining of C4d in allogeneic islets compared to NPIs 7 days post-infusion.²⁴ Although C3d and C4d are widely used as marker for complement activation, they have not been verified in the setting of IBMIR.^21,23 C3a and C3c have been used as a measure of complement activation in vitro in studying IBMIR;^18,28,29 however, both are soluble molecules and thus not suitable for quantitative IHC analysis. Finally, we did not select for recipients based on IgM levels or the amount of xeno-specific antibody. There is substantial variability in primates that has been exploited to more consistently reduce the humoral barriers to xenotransplantation,³⁰ and this approach could perhaps reveal a more impactful effect of CRP modification.

Although no difference was observed in complement activation between GKO and GKO/CD46 NPIs, we found significant less platelet deposition and neutrophil infiltration in GKO/CD46 NPIs at 24-hour, a difference that was not seen at 1-hour post infusion. This implies NPIs expressing human CD46 are better at controlling the early thrombo-inflammatory responses associated with IBMIR. This may be achieved either via its known mechanism at promoting cleavage of C3b and C4b, or via other unknown mechanisms related to platelet activation and leukocyte adhesion.

We had difficulty in achieving sufficient technical successes in the extended dual islet transplant group. Unlike our last report in this series, where we used an experimental drug regimen including anti-CD154 and anti-LFA1³¹ in the extended dual transplant arm,²⁴ only clinically available immunosuppressive drugs were used in these studies. To date there is no regimen that is limited to clinically available agents that consistently achieves xenoislet engraftment. At 7 days, significant antibody deposition and dense T cell and macrophage infiltrates were observed on histology in most recipients, consistent with acute rejections. This rapid rejection may have also contributed to the inadequate islet sampling and histology from this series.

Although many groups have experimented with xenotransplantation using multi-transgenic donor organs, most of the pre-clinical studies lack a means of parsing-out the relative contributions of each element used to promote engraftment. This is referable both to the limited availability of transgenic donors and the cost inherent in large animal studies. The dual transplant model provides a unique platform where NPIs with various genetic modifications can be compared within a single recipient. In this study our findings illustrate the benefit of hCD46 in reducing early thrombo-inflammatory events associated with IBMIR. However, it made clear that, anticipatably, local hCD46 expression alone does not abolish IBMIR. Along with CD46, NPIs expressing the two other human CRPs—CD55 and CD59—have also been shown to partially modify IBMIR in vivo.¹⁸ Conceivably the expression of all three CRPs targeting at different levels of the complement cascade could promote a complete effect. Interestingly, human islets express CD59 mainly, CD46 partially and do not express CD55 at all²⁹. Given that allogeneic islets are vulnerable to IBMIR, NPIs expressing all three CRPs could be envisioned to be even superior to allogeneic islets concerning local control of complement activation.

In addition, several other genetic modifications, developed by needs in solid organ xenotransplantation, can target different components of the IBMIR and may further reduce the early loss of islet mass following transplantation. For example, donor animals deficient in non-gal xenoantigens will minimize epitopes for immunoglobulin binding.⁴ Expression of human CD39 and CD73 and deletion of porcine von Willebrand factor (vWF) may further reduce platelet aggregation and deposition on NPIs, while various human coagulation regulatory proteins, including EPCR, TM and TFPI, may inhibit downstream clotting.^32-36 Lastly, expression of human HLA-E and CD47 on NPIs reduces NK and macrophage infiltration.^37,38 Although germline genetic modifications are generally required in solid organ xenotransplantation, genetic modifications may also be achieved temporarily in NPIs via viral vectors³⁹. For instance, overexpression of anti-apoptotic protein XIAP in islets has been shown to avoid early islets loss.^40-42 This strategy is especially relevant given the innate and short-lived nature of IBMIR. Alternatively, IBMIR may be controlled through systemic administration of clinically available complement inhibiting therapeutics. Cobra venom factor (CVF) has been historically used in pre-clinical models to deplete complement factors with good efficacy,⁴³ however, there are two FDA-approved complement inhibitors available (C1 esterase inhibitor and C5a inhibitor eculizumab) and around two dozen more in the pipeline.⁴⁴ In fact, eculizumab is currently being tested in a clinical trial to evaluate its role in preventing early destruction of islets after portal infusion ().

This study marks the third in our dual transplant model series. So far we have compared allogeneic islets to wild type NPIs, wild type NPIs to GKO NPIs, and now GKO NPIs to GKO/CD46 NPIs. While the dual transplant model is the only in vivo models that allows a head to head comparison between NPIs of two different genotypes, it is not without limitations. Firstly, the dual transplant model is complex and technically demanding. Secondly, as mentioned early, soluble factors, such as complement activation byproduct and cytokines, are likely involved in the early inflammatory process. These soluble factors, once released, circulate systemically and therefore not amenable to measurement in our current model. Thirdly, the dual transplant model allows comparison of two genotypes at a time while there are at least 10–15 different genetic modifications available to be tested. A high throughput in vitro screening method such as the surface coated loop device used in Bennet’s early studies,¹⁴ may be utilized before rigorous in vivo testing is conducted.

In conclusion, the expression of human CD46 partially protects xenoislets from early platelet deposition and neutrophil infiltration following intraportal infusion of NPIs. Our findings underscore the importance of CRPs in IBMIR and endorse the inclusion of hCD46 in future preclinical and clinical islet xenotransplantation studies.

Supplementary Material

Supp FigS1

Figure S1. The number of islets analyzed on human CD46 IHC slides across different experimental groups is shown (on one animal, the number of islets on the insulin IHC slides were used as the determining stain). As slides used for additional IHC staining was performed on serial slide sections adjacent to the human CD46 staining slides, the number of islets analyzed on the human CD46 stained slides reflects those on other IHC slides generated from the same section of the liver specimen. The number of islets analyzed on each slide did not differ significantly between the GKO and GKO/CD46 sides in either the 1hr or 24hrs groups. Overall, the number of islets analyzed ranged from 5 to 102 (mean=23.1, median=22.5, mode=23).

Abbreviations:

NHPs

Non-human primates

IBMIR

Instant blood mediated inflammatory reaction

GKO

α1,3-galactosyltransferase knockout

hCD46

human CD46

WT

Wild-type

NPIs

Neonatal porcine islets

IEQs

Islet equivalents

IVGTT

IV glucose tolerance test

Gal

galactose-α1,3-galactose

IHC

immunohistochemistry