The value of glycated albumin for the prediction of graft outcome in the non-human primate porcine islet transplantation model.

BACKGROUND: The development of a precise and easy-to-use tool for monitoring islet graft function is important in clarifying the causes of graft loss, identifying appropriate therapy, and ensuring graft survival in the nonhuman primate (NHP) model of porcine islet transplantation (PITx). Glycated albumin (GA) is an indicator of intermediate-term changes in blood glucose control and is useful in clinical diabetes management. The validity of GA for monitoring graft function in NHP recipients of PITx was evaluated using a retrospective analysis of cohort samples. METHODS: Data from a total of 23 PITxs performed in 20 recipients (3 were retransplanted) were included in this study. Islet clusters purified from adult wild-type pigs were transplanted via the intraportal route into streptozotocin-induced diabetic rhesus monkeys with immune suppression. Blood samples were obtained once per week from the recipients until they lost insulin-independence. Blood samples were also obtained from 69 non-diabetic monkeys that served as a control group. The levels of GA and albumin in stored plasma aliquots were measured using each enzymatic method, and the GA result was expressed as the percentage of GA level to the total albumin level. RESULTS: The median level of GA in the recipients on the day of PITx (median 18.6%, 95% confidence interval [CI] 16.7%-20.4%) was significantly higher than that of healthy controls (median 9.14%, 95% CI 9.0%-9.3%, P < .0001). However, the level decreased after PITx and remained low or increased depending on the extent of residual graft function. The GA level at a nadir (median 11.6%, 95% CI 10.8%-13.0%) and the time to reach a nadir (median 43 days, 95% CI 21.7-69.3 days) both correlated with the duration of insulin-independence (rho [ρ] = -.605, P = .0028 and ρ = .662, P = .0008, respectively). The GA level strongly correlated with KG , the glucose disappearance rate during intravenous glucose tolerance testing (ρ = -.76, P < .0001). At post-transplant week (PTW) 3 and at PTW 4, the GA levels in recipients with long-term insulin-independence (>90 days) were significantly lower than those with short-term insulin-independence, which revealed the excellent performance for the prediction of long-term insulin-independence that is comparable to that of porcine C-peptide (historic data). CONCLUSIONS: As a surrogate indicator for graft function, serial measurement of GA may provide Supporting Information to that obtained from conventional monitoring techniques of graft function for assessing porcine islet grafts in NHP models.

D-dimer level, in association with humoral responses, negatively correlates with survival of porcine islet grafts in non-human primates with immunosuppression.

BACKGROUND: Several immunosuppression (IS) regimens achieve long-term graft survival in non-human primates (NHPs) after porcine islet transplantation (PITx), but their success rates vary. To understand the mechanism of graft loss, we investigated the relationships between graft survival and humoral or inflammatory responses for maintenance IS in NHPs after PITx. METHODS: Islets purified from adult wild-type pigs were intraportally transplanted into streptozotocin-induced diabetic rhesus monkeys. Three monkeys received an IS regimen without anti-CD154 monoclonal antibody (mAb, transplant [Tpl]-control) and 11 received IS with anti-CD154 mAb (Tpl-aCD154). Blood samples were obtained weekly from the recipients until graft function ceased and weekly from three healthy monkeys (non-Tpl-control) for 6 months. Levels of D-dimer, C-reactive protein (CRP), and anti-Galα1,3Gal (Gal) IgG, IgG1, IgG2, and IgM were measured. Liver biopsy sections were immunostained for fibrin, insulin, and human CD31. RESULTS: Tpl-control monkeys had higher time-weighted average levels (levelstwavrg ) of Δanti-Gal IgG (Δ, difference from level at day 0) and D-dimer than Tpl-aCD154 or non-Tpl-control. The levelstwavrg of Δanti-Gal IgG, IgG1, IgG2, and IgM did not differ between Tpl-aCD154 and non-Tpl-control. The levelstwavrg of D-dimer and Δanti-Gal IgG2 negatively correlated with graft survival. Liver biopsy sections revealed many spots of fibrin deposition inside islet grafts that were well vascularized by human CD31-positive cells. Level of D-dimer positively correlated with Δanti-Gal IgG1 in Tpl-control and with Δanti-Gal IgG2 in Tpl-aCD154. CONCLUSIONS: Intravascular coagulation, in association with immune responses against xenografts, may partly contribute to loss of islet grafts in NHPs after PITx.

Dissociation between anti-porcine albumin and anti-Gal antibody responses in non-human primate recipients of intraportal porcine islet transplantation.

BACKGROUND: To understand humoral responses elicited after xenotransplantation, we compared the induction of anti-non-Gal antibodies vs. anti-Gal antibodies in non-human primates (NHPs) after intraportal porcine islet transplantation (PITX). METHODS: Anti-Gal and anti-non-Gal IgGs were analyzed in serial plasma samples of NHP recipients after PITX by enzyme-linked immunosorbent assay (ELISA) using synthetic Gal and by flow cytometry using α-1,3-galactosyltransferase gene knockout (GTKO) porcine endothelial cells, respectively. Anti-non-Gal IgG was detected in some recipients after PITX. The specificity of anti-non-Gal IgG was investigated by two-dimensional electrophoresis of the protein extract from GTKO porcine endothelial cells, Western blot analysis of recipient pre- and post-PITX plasma, and MALDI-TOF/TOF mass spectrometry, revealing albumin, a non-glycosylated protein in the serum supplement of the islets solution, as a putative antigen for anti-non-Gal IgG. The binding of IgG antibodies to human albumin (HA), bovine albumin (BA), porcine albumin (PA), and Gal was compared by ELISA in pre- and post-PITX plasma samples of 30 NHP recipients subjected to intraportal PITX, which were grouped according to the use of CD40-CD154 blockade and sirolimus. RESULTS: One of the immunoblot-matched spots was identified as BA by mass spectrometry. By ELISA, the plasma used in the immunoblot analysis revealed strong IgG binding to BA and PA, but not to HA. Anti-PA, anti-BA, and anti-Gal antibodies in NHP recipients 1 month after PITX were detected in 5 (100%), 3 (60%), and 5 (100%), respectively, of the 5 recipients receiving various immunosuppression (IS) without CD40-CD154 blockade (group I) and in 0 (0%), 0 (0%), and 4 (16%), respectively, of the 25 recipients receiving IS with CD40-CD154 blockade and sirolimus (group II). This finding revealed significant differences between the groups (P < 0.0001, P = 0.0011 and P = 0.0013, respectively). Interestingly, among 15 recipients achieving graft survival longer than 1 month in group II, anti-PA IgG was detected in only 1 recipient (6.7%) 180 days after PITX. However, an increase in anti-Gal IgG was detected in 7 recipients (46.7%) despite maintenance IS with anti-CD154 and sirolimus. This result indicates that anti-Gal IgG is more frequently induced than anti-PA IgG (P = 0.0352). Moreover, induction IS with anti-CD154 and sirolimus suppressed anti-Gal IgG, but not anti-PA and anti-BA IgG, responses in sensitized recipients given a repeat transplantation. CONCLUSIONS: In NHP recipients of PITX, anti-PA and anti-BA IgG antibodies are elicited by porcine serum included as a supplement in porcine islet preparation. IS including CD40-CD154 blockade and sirolimus suppresses these antibody responses in naïve recipients, but not in sensitized recipients. The elicitation of anti-xenogenic albumin antibodies, a humoral response to a model protein antigen, is distinct from that of anti-Gal antibodies, a response to carbohydrate antigen.

Investigation of the efficacy and safety of wild- type and triple-gene knockout pig RBC transfusions in nonhuman primates.

Introduction: Decreasing rates of blood donation and close margins between blood supply and demand pose challenges in healthcare. Genetically engineered pig red blood cells (pRBCs) have been explored as alternatives to human RBCs for transfusion, and triple-gene knockout (TKO) modification improves the compatibility of pRBCs with human blood in vitro. In this study, we assessed the efficacy and risks of transfusing wild-type (WT)- and TKO-pRBCs into nonhuman primates (NHPs). Methods: Blood from O-type WT and TKO pigs was processed to produce pRBCs for transfusion, which were transfused or not into NHPs (n=4 per group: WT, TKO, and control) after 25% total blood volume withdrawal: their biological responses were compared. Hematological, biochemical, and immunological parameters were measured before, immediately after, and at intervals following transfusion. Two months later, a second transfusion was performed in three NHPs of the transfusion group. Results: Transfusion of both WT- and TKO-pRBCs significantly improved RBC counts, hematocrit, and hemoglobin levels up to the first day post-transfusion, compared to the controls. The transfusion groups showed instant complement activation and rapid elicitation of anti-pig antibodies, as well as elevated liver enzyme and bilirubin levels post-transfusion. Despite the higher agglutination titers with WT-pRBCs in the pre-transfusion crossmatch, the differences between the WT and TKO groups were not remarkable except for less impairment of liver function in the TKO group. After the second transfusion, more pronounced adverse responses without any hematological gain were observed. Conclusions: WT- and TKO-pRBC transfusions effectively increased hematologic parameters on the first day, with rapid clearance from circulation thereafter. However, pRBC transfusion triggers strong antibody responses, limiting the benefits of the pRBC transfusion and increasing the risk of adverse reactions.

Initial investigation on the feasibility of porcine red blood cells from genetically modified pigs as an alternative to human red blood cells for transfusion.

The decline in blood donation rates and the ongoing shortage of blood products pose significant challenges to medical societies. One potential solution is to use porcine red blood cells (pRBCs) from genetically modified pigs as an alternative to human red blood cells (hRBCs). However, adverse immunological reactions remain a significant obstacle to their use. This study aimed to evaluate the compatibility of diverse genetically modified pRBCs with human serum. We acquired human complement-competent serum, complement 7 (C7)-deficient serum, and hRBCs from all ABO blood types. Additionally, we used leftover clinical samples from health checkups for further evaluation. pRBCs were collected from wild-type (WT) and genetically modified pigs: triple knockout (TKO), quadruple KO (QKO), and TKO/hCD55.hCD39 knockin (hCD55.hCD39KI). The extent of C3 deposition on RBCs was measured using flow cytometry after incubation in C7-deficient serum diluted in Ca++-enriched or Ca++-depleted and Mg++-enriched buffers. The binding of immunoglobulin (Ig) M/IgG antibody to RBCs after incubation in ABO-type human serum was evaluated using flow cytometry. Naïve human serum- or sensitized monkey serum-mediated hemolysis was also evaluated. Phagocytosis was assessed by incubating labeled RBCs with the human monocytic cell line THP-1 and measurement by flow cytometry. All three genetic modifications significantly improved the compatibility of pRBCs with human serum relative to that of WT pRBCs. The extent of IgM/IgG binding to genetically modified pRBCs was lower than that of WT pRBCs and similar to that of O-type hRBCs. Total and alternative pathway complement activation in all three genetically modified pRBCs was significantly weaker than that in WT pRBCs and did not differ from that in O-type hRBCs. The extent of serum-mediated hemolysis and phagocytosis of these genetically modified pRBCs was low and similar to that of O-type hRBCs. Sensitized monkey serum-mediated hemolysis in QKO and TKO/hCD55.hCD39KI pRBCs was higher than in O-type hRBCs but lower than in TKO pRBCs. The elimination of porcine carbohydrate antigens in genetically modified pigs significantly enhanced pRBC compatibility with naïve human sera, which was comparable to that of O-type hRBCs. These findings provide valuable insights into the development of pRBCs as potential alternatives to hRBCs.

Biological response of nonhuman primates to controlled levels of acute blood loss.

Introduction: The global shortage of human blood for medical use has prompted the development of alternative blood sources. Nonhuman primates (NHPs) are commonly used owing to their physiological similarities to humans. The objective of the current study was to establish a controlled-blood-loss model in NHPs to explore their clinical and biological responses. Methods: Blood was sequentially withdrawn from 10 cynomolgus monkeys (10, 14, 18, 22, and 25% of the total blood volume); their vital signs were monitored, and blood parameters were serially analyzed. Humoral mediators in the blood were measured using flow cytometry and enzyme-linked immunosorbent assays. Results: In NHPs subjects to 25% blood loss and presenting with related clinical symptoms, the systolic blood pressure ratio on day 0 after bleeding was significantly lower than that of the animals from the other groups (median: 0.65 vs. 0.88, P = 0.0444). Red blood cell counts from day 0-14 and hematocrit levels from day 0-7 were markedly decreased relative to the baseline (P < 0.01). These parameters showed a direct correlation with the extent of blood loss. The levels of creatine phosphokinase, aspartate aminotransferase, and alanine aminotransferase exhibited increases in response to blood loss and had a stronger correlation with the hemoglobin ratio than the volume of blood loss. The levels of C3a and C4a, as well as interleukin (IL)-1α and IL-15, displayed a strong correlation, with no apparent association with blood loss. Conclusion: The findings of the present study showed that only NHPs with 25% blood loss exhibited clinical decompensation and significant systolic blood pressure reduction without fatalities, suggesting that this level of blood loss is suitable for evaluating blood transfusion efficacy or other treatments in NHP models. In addition, the ratio of hemoglobin may serve as a more dependable marker for predicting clinical status than the actual volume of blood loss. Thus, our study could serve as a basis for future xenotransfusion research and to predict biological responses to massive blood loss in humans where controlled experiments cannot be ethically performed.

Increased human tumor necrosis factor-α levels induce procoagulant change in porcine endothelial cells in vitro.

BACKGROUND: Intravascular thrombosis and systemic coagulation abnormalities are major hurdles to successful xenotransplantation and are signs of acute humoral rejection. Increased expression of tissue factor (TF) is associated with the development of microvascular thrombosis in xenografts. To develop an effective strategy to prevent accelerated coagulation in xenografts, we investigated the mechanism by which porcine endothelial cells (PECs) become procoagulant after contact with human blood. METHODS: The changes in TF mRNA levels and activity in PECs after incubation with 20% human serum or human bioactive molecules, including C5a, tumor necrosis factor-α (TNFα) and interleukin (IL)-1α, were evaluated using real-time PCR and the factor Xa chromogenic assay, respectively. The procoagulant changes in PECs by these agonists were evaluated by measuring the coagulation time of human citrated plasma suspended with PECs pretreated with each agonist. TF expression and coagulation times were also assessed in PECs transfected with short interfering RNA (siRNA) designed to knock down porcine TF. We also examined the production of proinflammatory cytokines in human whole-blood or plasma after contact with PECs, which were screened using the cytometric bead array system. TNFα levels were measured using ELISA in whole-blood after contact with PECs, with or without the addition of xenoreactive antibodies or C1 esterase inhibitor. RESULTS: Porcine TF mRNA and activity in PECs were up-regulated in response to human TNFα and IL-1α but were not affected by C5a or 20% human serum. Up-regulation of TF expression by human TNFα or IL-1α shortened PEC-induced coagulation time, while siRNA-mediated knockdown of TF expression prolonged coagulation time. The incubation of PECs with human whole-blood led to a significant increase in human TNFα levels in the blood, which was promoted by the addition of xenoreactive antibodies and prevented by C1 esterase inhibitor. CONCLUSIONS: Human TNFα level increases in human blood after contact with PECs, which is attributed to xenoreactive antibody binding and subsequent complement activation. Human TNFα induces procoagulant changes in PECs with increased TF expression. This study suggests that human TNFα may be one of the mediators linking complement activation with procoagulant changes in the xenoendothelium.

Effect of Factor H on Complement Alternative Pathway Activation in Human Serum Remains on Porcine Cells Lacking N-Glycolylneuraminic Acid.

Background: Triple knockout (TKO) donor pigs lacking alpha-1,3-galactose (Gal), N-glycolylneuraminic acid (Neu5Gc), and Sd(a) expressions were developed to improve the clinical success of xenotransplantation. Neu5Gc, a sialic acid expressed on cell surfaces, recruits factor H to protect cells from attack by the complement system. Lack of Neu5Gc expression may cause unwanted complement activation, abrogating the potential benefit of gene-modified donor pigs. To investigate whether TKO porcine cells display increased susceptibility to complement activation in human serum, pathway-specific complement activation, apoptosis, and human platelet aggregation by porcine cells were compared between alpha-1,3-galactosyltransferase gene-knockout (GTKO) and TKO porcine cells. Methods: Primary porcine peripheral blood mononuclear cells (pPBMCs) and endothelial cells (pECs) from GTKO and TKO pigs were used. Cells were incubated in human serum diluted in gelatin veronal buffer (GVB++) or Mg++-EGTA GVB, and C3 deposition and apoptotic changes in these cells were measured by flow cytometry. C3 deposition levels were also measured after incubating these cells in 10% human serum supplemented with human factor H. Platelet aggregation in human platelet-rich plasma containing GTKO or TKO pECs was analyzed. Results: The C3 deposition level in GTKO pPBMCs or pECs in GVB++ was significantly higher than that of TKO pPBMCs or pECs, respectively, but C3 deposition levels in Mg++-EGTA-GVB were comparable between them. The addition of factor H into the porcine cell suspension in 10% serum in Mg++ -EGTA-GVB inhibited C3 deposition in a dose-dependent manner, and the extent of inhibition by factor H was similar between GTKO and TKO porcine cells. The percentage of late apoptotic cells in porcine cell suspension in GVB++ increased with the addition of human serum, of which the net increase was significantly less in TKO pPBMCs than in GTKO pPBMCs. Finally, the lag time of platelet aggregation in recalcified human plasma was significantly prolonged in the presence of TKO pECs compared to that in the presence of GTKO pECs. Conclusion: TKO genetic modification protects porcine cells from serum-induced complement activation and apoptotic changes, and delays recalcification-induced human platelet aggregation. It does not hamper factor H recruitment on cell surfaces, allowing the suppression of alternative complement pathway activation.

The role of phagocytosis in IL-8 production by human monocytes in response to lipoproteins on Staphylococcus aureus.

Cytokine responses to microbes are triggered by pattern recognition receptors, such as Toll-like receptors (TLRs), which sense pathogen-associated molecular patterns. Cell wall-associated triacylated lipoproteins in Staphylococcus aureus are known to be native TLR2 ligands that mediate host inflammatory responses against S. aureus. However, the mechanism by which these lipidated lipoproteins, which are buried under the thick S. aureus cell wall, work to stimulate TLR2 remains unclear. Heat-killed wild type S. aureus cells activated human monocytic THP-1 cells to produce proinflammatory cytokines, including interleukin (IL)-8, whereas the lipoprotein lipidation-deficient lgt mutant induced less than an eighth of the amount of IL-8 induced by the wild type. IL-8 induction in response to heat-killed S. aureus cells in THP-1 cells was not inhibited by a blocking antibody against cell surface TLR2, suggesting that intracellular TLR2 might be involved in the induction of IL-8 by S. aureus lipoprotein. The relationship between phagocytosis and IL-8 production in THP-1 cells was analyzed on a single-cell level by flow cytometry using fluorescein-labeled S. aureus cells and phycoerythrin-labeled anti-IL-8 antibody. Production of intracellular IL-8 was correlated with phagocytosis of S. aureus cells in THP-1 cells and in human peripheral blood mononuclear cells. Opsonization of S. aureus cells enhanced both the phagocytosis of S. aureus cells and the production of intracellular IL-8 in THP-1 cells. These results suggest that lipidated lipoproteins on S. aureus cells stimulate human monocytes after phagocytosis.

Immunoglobulin M and Immunoglobulin G Subclass Distribution of Anti-galactose-Alpha-1,3-Galactose and Anti-N-Glycolylneuraminic Acid Antibodies in Healthy Korean Adults.

BACKGROUND: Human preformed antibodies (Abs), anti-galactose-alpha-1,3-galactose (Gal) and anti-N-glycolylneuraminic acid (Neu5Gc), can react with porcine antigens of wild-type pigs. To provide basic population data of the Abs for potential application in clinical xenotransplantation, we developed enzyme-linked immunosorbent assay methods and investigated the serum titers of anti-Gal and anti-Neu5Gc Abs, including immunoglobulin (Ig) M and IgG along with its subclasses, in humans. METHODS: Anti-Gal and anti-Neu5Gc Abs serum titers were measured in 380 healthy Korean adults using the in-house enzyme-linked immunosorbent assays. The frequency and median values of anti-Gal and anti-Neu5Gc were measured, and their class and subclass distribution were evaluated. RESULTS: The detection frequencies of anti-Gal were 99.2%, 95.0%, 23.2%, 94.5%, 12.4%, and 3.4% for IgM, IgG, IgG1, IgG2, IgG3, and IgG4, respectively. The detection frequencies of anti-Neu5Gc Abs were 87.4%, 96.6%, 1.6%, 46.3%, 0.0%, and 0.0% for IgM, IgG, IgG1, IgG2, IgG3, and IgG4, respectively. The median values of anti-Gal IgM (1001.6 ng/mL) and IgG (1198.3 ng/mL) were significantly higher than those of anti-Neu5Gc Abs (IgM, 328.4 ng/mL; IgG, 194.7 ng/mL; P < .001). IgG2 titers of both anti-Gal and anti-Neu5Gc Abs correlated better with the IgG class than the titers of other IgG subclasses. CONCLUSIONS: The titers of anti-Gal Abs were higher than those of anti-Neu5Gc Abs. IgG2 was the main IgG subclass in both anti-Gal and anti-Neu5Gc Abs. Variation in the titers of anti-Gal or anti-Neu5Gc Abs may partly explain the biological and immunologic changes that occur in recipients of xenotransplants.

Early Interferon-Gamma Response in Nonhuman Primate Recipients of Solid-Organ Xenotransplantation.

BACKGROUND: To understand changes in biological responses in nonhuman primate (NHP) recipients of xenotransplantation (XTP), we retrospectively investigated chronological changes in cytokine profiles of NHP recipients after solid-organ XTP. METHODS: Plasma samples were collected from 7 NHP recipients of pig heart or kidney XTP with α-1,3-galactosyltransferase gene knockout (GTKO) under anti-CD154-based immune suppression at the following time points: immediately before; 2 hours, 3 days, and 7 days after XTP; and weekly thereafter until the graft failed. The plasma levels of the following cytokines were measured: interleukin (IL)-1α, IL-1ß, IL-6, IL-12p70, IL-8, IL-10, IL-15, tumor necrosis factor, interferon gamma (IFN-γ), D-dimer, C3a, and histone-complexed DNA fragments. For in vitro experiments, human natural killer (NK) cells were cocultured with wild-type porcine endothelial cells (PECs), GTKO-PECs, and human umbilical vein endothelial cells, with or without anti-CD154 antibody. IFN-γ levels in the culture supernatants were compared. RESULTS: IFN-γ levels peaked on day 7 or 10 of XTP and then decreased to basal levels, whereas proinflammatory cytokine levels increased along with the elevation of histone-complexed DNA fragments and were sustained until xenograft failure. In vitro, human NK cells produced more IFN-γ when in contact with wild-type PECs than with human umbilical vein endothelial cells, which was not reduced by the use of GTKO-PECs or addition of anti-CD154 antibody to the mixture. CONCLUSIONS: In NHP recipients of XTP, the early peak of IFN-γ priming subsequent inflammatory responses may be attributed to NK cell activation in response to xenografts.

Increase in anti-Gal IgM level is associated with early graft failure in intraportal porcine islet xenotransplantation.

BACKGROUND: Anti-Gal is a major antibody induced in non-human primates (NHPs) after xenotransplantation. To understand the mechanism of graft rejection, we investigated the association between anti-Gal responses and graft failure in NHP recipients of porcine islet transplantation (PITx). METHODS: Intraportal PITx was performed in 35 diabetic NHPs, and graft function was monitored. Early graft failure (EGF) was defined as loss of graft function within a month after PITx. Seven, 19, nine NHPs received immunosuppression (IS) without CD40 pathway blockade (Group I), with anti-CD154 (Group II), and with anti-CD40 (Group III), respectively. The anti-Gal levels on day 0 and day 7 of PITx were measured by ELISA. RESULTS: The frequency of EGF was significantly lower in Group II (26.3%) than in Group I (100%, P=0.0012) and Group III (77.8%, P=0.0166). While levels of anti-Gal IgG in Group I and anti-Gal IgM in Group III increased on day 7 compared with day 0 (P=0.0156 and 0.0273), there was no increase in either on day 7 in Group II. The ratio of anti-Gal IgM or IgG level on day 7 to that on day 0 (Ratio7/0) was significantly higher in recipients with EGF than without EGF (P=0.0009 and 0.0027). ROC curve analysis of anti-Gal IgM Ratio7/0 revealed an area under the curve of 0.789 (P=0.0003). CONCLUSIONS: IS with anti-CD154 suppressed anti-Gal responses and prevented EGF in PITx. Anti-Gal IgM Ratio7/0, being associated with EGF, is a predictive marker for EGF.

The role of the alternative complement pathway in early graft loss after intraportal porcine islet xenotransplantation.

BACKGROUND: Intraportal islet transplantation (ITx) causes instant blood-mediated inflammatory reaction (IBMIR), resulting in an early loss of transplanted islets. Porcine islets, transplanted intraportally into nonhuman primates (NHPs), induce complement activation, contributing to the development of IBMIR; however, the exact mechanism is not clear. METHODS: Complement activation were compared after incubation of purified adult porcine islets in 20% human serum with or without complement inhibitors, C1 esterase inhibitor (C1E-inh), anti-factor B, and purified human factor H. Intraportal porcine ITx was performed in diabetic NHPs to which cobra venom factor (CVF), factor H, or none of complement inhibitor was administered during the peritransplant period. The extent of complement activation and function of islet grafts were monitored after ITx. RESULTS: The incubation of porcine islets with human serum resulted in generation of C3a, C4d, and factor Bb in the fluid phase. However, the generation of C3a after incubation was suppressed by anti-factor B or factor H, but not by C1E-inh. Moreover, in NHPs with porcine ITx, the administration of CVF or factor H ameliorated the increase in plasma C3a and factor Bb levels, as well as early release of porcine C-peptide after ITx. Furthermore, the functional survival of islet grafts was prolonged in the recipients of the CVF group compared to those in the control group. CONCLUSIONS: The alternative complement pathway contributes to the development of IBMIR and the early loss of grafts in NHPs with porcine ITx. Complement inhibition during the peritransplant period may be beneficial for the survival of islet grafts.