Posttransplant Cellular Immune Reactivity against Donor Antigen Correlates with Clinical Islet Transplantation Outcome: Towards a Better Posttransplant Monitoring

Abstract

The aim of the present study was to assess the efficiency of cell-based immune assays in the detection of alloreactivity after islet transplantation and to correlate these results with clinical outcome. Mixed lymphocyte cultures were performed with peripheral blood mononuclear cells from recipients (n = 14), donors, or third party. The immune reactivity was assessed by the release of IFN-γ (ELISpot), cell proliferation (FACS analysis for Ki67), and cytokine quantification (Bioplex). Islet function correlated with the number of IFN-γ-secreting cells following incubation with donor cells (p = 0.007, r = –0.50), but not with third party cells (p = 0.61). Similarly, a high number of donor-specific proliferating cells was associated with a low islet function (p = 0.006, r = −0.51). Proliferating cells were mainly CD3⁺CD4⁺ lymphocytes and CD3^-CD56⁺ natural killer cells (with low levels of CD3⁺CD8⁺ lymphocytes). Patients with low islet function had increased levels of CD4⁺Ki67⁺cells (p ≤ 0.0001), while no difference was observed in CD8⁺Ki67⁺ and CD56⁺Ki67⁺ cells. IFN-γ, IL-5, and IL-17 levels were increased in patients with low islet function, but IL-10 levels tended to be lower. IFN-γ-ELISpot, proliferation, and cytokines were similarly accurate in predicting clinical outcome (AUC = 0.77 ± 0.088, 0.85 ± 0.084, and 0.88 ± 0.074, respectively). Cellular immune reactivity against donor cells correlates with posttransplant islet function. The tested assays have the potential to be of substantial help in the management of islet graft recipients and deserve prospective validation.

Keywords

Monitoring Diabetes Islet Transplantation Cytokine ELISpot

Introduction

Over the recent years, noninvasive posttransplant monitoring has become one of the key challenges in the field of islet transplantation, as it could help in detecting and treating harmful events in a timely fashion.

Improving islet graft monitoring appears especially important in this area of transplantation, considering that islet biopsy cannot be used in routine clinical practice (islets can only be detected in about one third of needle liver biopsies) (30). In addition, the currently available monitoring techniques, based on positron emission tomography (PET), scintigraphy with glucagon-like peptide 1 (GLP1) analogues, and magnetic resonance imaging (MRI) are too invasive, time-consuming, and expensive for a close monitoring each time the patient is coming to clinic (4,23,33,34).

As a result, various investigators have tried to identify immune markers of the posttransplant islet graft fate (15). Such a monitoring is noninvasive and less expensive and has the potential to be used often and broadly both for detecting harmful events and tuning immunosuppression accordingly. Despite its potential, immune monitoring has been relatively slow to develop. The earliest investigations were based on the assessment of antibody. The level of auto-antibody [anti-glutamate decarboxylase (GAD), islet autoantigen (IA2), and/or insulin] has shown heterogeneous results when correlated to the posttransplant outcome (2,8,9,11,13,14,36). In addition, high pretransplant levels of anti-human leukocyte antigen (HLA) allo-antibody were associated to worse posttransplant outcomes, but peak posttransplant levels (assessed by flow beads or ELISA) only weakly correlated to outcome (3,6,29).

More recently, cell-based immune tests have been developed in the area of islet transplantation. They include the assessment of cell proliferation and cytokine release after stimulation with islet or donor antigens (15). These investigations have established that the level of pretransplant auto- and allo-immunity correlates accurately with the posttransplant outcome (9,11,19,24,25). On the other hand, the impact of a posttransplant immune reactivity, which is of potential use for the detection of harmful events, is less clear. The posttransplant levels of auto- (assessed by multimers or after stimulation with islet antigens) and of allo-reactive cells (determined as the number of cytotoxic T lymphocyte precursors) do correlate to a certain extend with outcome (11,12). However, these data have shown some heterogeneity regarding the value of posttransplant allo-immunity, and they deserve external validation. In addition, the enzyme-linked immunospot (ELISpot), easier and faster to perform, has not been assessed in the posttransplant setting, while we have previously shown that it can accurately predict the occurrence of rejection and recurrence of autoimmunity in a mouse models (32). Finally, the various potential tests should be compared and potentially refined with the assessment of lymphocyte subtypes.

The aim of the present study was to validate the assessment of posttransplant cell immune reactivity [proliferation, interferon-γ (IFN-γ)-ELISpot, and release of other cytokines] after stimulation with donor cells and correlated it with clinical outcome (assessed as the β-score) (26).

Patients and Methods

Study Design and Inclusion Criteria

Blood samples were prospectively collected between April and December 2010 from islet recipients at the University of Geneva, Switzerland. Islet recipients without type 1 diabetes (cystic fibrosis) were excluded. Healthy individuals were included as controls. The study has been reviewed and approved by the institutional research ethics board. Written consent was obtained from the included patients.

Islet Transplantation and Blood Sample Collection

All assessed patients were transplanted in Geneva University Hospital. Islets were isolated according to a modified Ricordi's technique and transplanted as previously described (28,31). All included patients had c-peptide-negative (<166 pmol/L) type 1 diabetes prior to transplantation. Transplantation outcome was assessed according to the β-score, as previously described in Ryan et al. (26).

Blood samples were collected in three 8.5-ml acid citrate dextrose tubes (BD Vacutainer, Franklin Lakes, NJ) during routine outpatient clinics. Samples were collected between 1 and 9 years after transplantation at the closest outpatient clinic date from the start of the study. The HLA A, B, DR typing of both recipients and donors was performed by molecular biology technology (luminex SSO or SSP).

Anti-HLA Antibody Detection Assays

Sera were analyzed for the presence of anti-HLA class I and class II IgG antibodies by solid phase assays ELISA for pretransplantation screening (LAT^™-M, OneLambda, Canoga Park, CA) and by LABScreen® Mixed assay for posttransplantation data (LSM12®, OneLambda). Test interpretation was performed using HLA Visual® software (OneLambda) on the LABScan100^™ flow cytometer (Luminex, Inc., Austin, TX) with a positive cut-off of 3.0. All positive results were tested to identify the specificity of anti-HLA antibodies with a high-definition LABScreen® Single Antigen class I and class II assay (OneLambda). A mean fluorescence intensity (MFI) of 500 is used as threshold.

Mixed Lymphocyte Cultures (MLCs)

Fresh peripheral blood mononuclear cells (PBMCs) isolated using Ficoll Paque (GE Healthcare, Uppsala, Sweden) were counted and used as responder cells. Donor and third party spleen cells used as stimulator cells were thawed, washed, and counted. Third party cells were selected as having no common A, B, and DR human leucocyte antigens with donor and recipient cells. One MLC was performed for each donor, leading to the number of 28 assays (n = 28) for the whole group (n = 14).

MLCs were performed in 96-well plates in 200 μl RPMI 1640 (Gibco Invitrogen, Carlsbad, CA), 10% fetal calf serum, HEPES (1 M, Gibco Invitrogen), and penicillin–streptomycin with l-glutamine (1 U/ml, 1 μg/ml, and 0.29 mg/ml, respectively, Gibco). Responder cells (5 × 10⁵) were incubated with 5 × 10⁵ irradiated stimulator cells (5,000 rad) at 37°C/5% CO₂. Cells were transferred to ELISpot plates in duplicate for 24-h incubation or into 96-well flat plates for 72-h incubation in view of the proliferation assessment. Anti-CD3/CD28 beads (1 μl/well, Dynabeads, Invitrogen) were used as positive control.

Enzyme-Linked Immunosorbent Spot

The release of IFN-γ was assessed by ELISpot (eBioscience, San Diego, CA) according to the manufacturer's instructions. Briefly, wells were coated overnight with capture anti-IFN-γ antibody prior to culture. After 24 h of culture, wells were washed and biotinylated detection antibody was added. Following incubation with avidin-horseradish peroxidase, the substrate solution (3-amino-9-ethyl carbazole, AEC) was added and spots were read using an automated Immunospot analyzer (Cellular Technology Ltd., Bonn, Germany). Results were expressed as the number of IFN-γ-secreting cells/5 × 10⁵ total cells using the following formula: (number of spots/well – background measured in the negative control wells).

Intracellular IFN-γ and Proliferation Flow Cytometry Assessment

Intracellular IFN-γ content and cell proliferation were assessed by flow cytometry. Cells were centrifuged, and the supernatant was removed and frozen. The intracellular IFN-γ content was assessed after 20 h of culture. Cells were incubated for 3 h with a protein transport inhibitor (GolgiPlug, 0.25 μl/well, BD Biosciences, San Diego, CA). Staining for intracellular cytokine [allophycocyanin (APC) anti-IFN-γ (B27)] was performed using a fixation/permeabilization solution kit (BD Biosciences).

The level of proliferation was assessed after 72 h of culture. Fixation and permeabilization were performed using forkhead box P3 (Foxp3) staining buffer set according to the manufacturer's protocol (eBiosciences). For intranuclear labeling of proliferation, cells were incubated with fluorescein isothiocyanate (FITC) anti-Ki67 antibody (BD Biosciences). Proliferation results were obtained by dividing each specific data by the level of proliferation in the negative control.

Cell surface antigens were labeled with the following antibodies: phycoerythrin (PE) anti-CD8 (RPA-T8), peridinin chlorophyll protein complex (PerCP) anti-CD3 (UCHT1), PE/cyanine 7 (Cy7), anti-CD56 (HCD56), APC/Cy7 anti-CD4 (RPA-T4) (BioLegend, San Diego, CA). Cell labeling was acquired and analyzed with a FACSCanto flow cytometer (BD Biosciences).

Quantification of Cytokine Production in MLC

The production of various cytokines was assessed by Bioplex technology using a human Th1/Th2 Multiplex cytokine kit including interleukin (IL)-5, IL-10, and IFN-γ and a human IL-17 Singleplex cytokine kit, according to the manufacturer's protocol (Invitrogen). Briefly, antibody-coated cytokine-specific beads were prepared in 96-well plates. Supernatants (retrieved after a 3-day MLC) were added to the wells for a 45-min incubation at room temperature in the dark with a 500-Hz shaking, allowing cytokines to bind to the cytokine-specific beads. Plates were subsequently washed and incubated with biotinylated anti-cytokine detection antibody for 60 min. After washing, streptavidin-PE was added for 30 min to bind to the detection antibody. The level of fluorescence of the cytokine-specific beads was analyzed by a double-laser Bio-plex reader. Cytokine concentration was determined based on a standard curve included in each plate, using cytokine standards provided by the manufacturer.

Statistical Analysis

Groups were compared using the Mann–Whitney, Spearman, and Kruskal–Wallis tests. Tables reported medians and quartiles. In order to report on a view of central tendency, figures displayed means and SEMs. The accuracy of the immune monitoring to predict clinical outcome was assessed utilizing a receiver operating characteristic (ROC) curve and determining its area under the curve (AUC). A standard a level of 0.05 was selected. Analyses were performed with the SPSS software package (SPSS 15.0, SPSS, Chicago, IL).

Results

Patient Characteristics

Blood samples from 14 patients were available for the study (Table 1). A median of 13.8 × 10³ IEQ/kg (islet equivalent) was transplanted per patient in two to six infusions. Of note, one patient has lost a first set of transplant and subsequently received a new transplantation with a total of six infusions (the inclusion or the exclusion of this patient did not change results, and he was kept in the whole analysis). The transplantations were performed as islet alone (n = 5), islet after kidney (n = 6), or simultaneous islet/kidney (n = 3) transplantations. Nine patients were on a tacrolimus/mycophenolate mofetil (MMF) maintenance protocol at the time of assessment. According to the β-score, half of the patients had a good islet function (β-score > 4) at the time of the immune assessment. Figure 1 provides the rates of insulin independence and c-peptide positivity in both groups. Only two patients were c-peptide negative and were assessed 72 and 78 months after islet loss. None of the 14 patients demonstrated anti-HLA class I or class II antibodies prior to transplantation (Table 2). At the time of cellular reactivity assessment, anti-HLA class I or class II antibodies were detected in some recipients, but no trend was observed between patients with good or bad islet function and only one patient demonstrated high level of anti-donor antibodies. Similar observations were made with a longitudinal assessment including peak levels.

Figure 1.

C-peptide positivity (A) and insulin independence (B) in the studied patient groups.

Table 1.

Patient Characteristics at the Time of the Assay

	β-Score
	Score ≤ 4	Score > 4	Total
Patients (n)	7	7	14
Gender (male/female)	4/3	2/5	6/8
Body weight (kg)	61.7 (56.25–82.95)	58.2 (57.4–67.7)	60.8 (57.2–78.2)
Number of infusions^a	2 (2–3)	3 (2.5–3)	3 (2–3)
Total transplanted islets (IEQ/kg)^a	12,813 (9,862–17,370)	15,094 (11,076–17,071)	13,886 (11,046–17,349)
Years post last transplant	7 (5.5–8)	4 (1.5–4.5)	5 (4–6.75)
Immunosuppressive treatment:
Induction treatment (thymoglobuline/anti-CD25)	0/7	4/4	4/11
Tacrolimus - MMF	4	5	9
Tacrolimus - Sirolimus	2	1	3
MMF - Sirolimus	1	–	1
MMF	–	1	1
Status:
Islet transplant alone	1	4	5
Islet transplant after kidney transplant	5	1	6
Simultaneous islet and kidney transplant	1	2	3
β-score:
Fasting plasma glucose (mmol/L)	7.6 (6.3–10.25)	6.2 (5.75–6.35)	6.3 (5.9–7.55)
HbA1c (%)	6.6 (6.5–7.9)	5.6 (6.5–6.0)	6.3 (5.7–6.6)
Daily insulin requirement (units/kg)	0.38 (0.33–0.44)	0	0.12 (0–0.37)
Basal C-peptide (nmol/L)	0.22 (0.14–0.27)	0.43 (0.34–0.55)	0.31 (0.24–0.39)
Stimulated C-peptide (nmol/L)	0.19 (0.09–0.3)	–	–
OHA use	–	–	–

Two patients have lost a first set of transplant and subsequently received a new transplant (explaining the high number of infusions and the high number of transplanted islets in these two individuals). MMF, mycophenolate mofetil; HbAlc, glycated hemoglobin; OHA, oral hypoglycemic agents.

Table 2.

Anti-HLA Antibodies Before and After Islet Transplantation

		Pretransplant	Closed to Cellular Assessment
	Patient	Anti-Class I/II IgG	Anti-Class I IgG	Anti-Class II IgG	DSA (MFI)
β-score>4	1	No/No	No	Yes	No A-B-DR
	2	No/No	No	No
	3	No/No	No	No
	4	No/No	No	No
	5	No/No	No	No
	6	No/No	No	Yes	No A-B-DR
	7	No/No	No	No
β-Score<4	8	No/No	Yes	Yes	A26 (866)
	9	ND	No	No
	10	No/No	Yes	No	No
	11	No/No	Yes	Yes	A24 (4539) B35(13439)/B38(7971)/B41(4333)
	12	No/No	No	No
	13	ND	No	No
	14	No/No	No	No

ND, no data; HLA, human leukocyte antigen; DSA (MFI), donor specific antibody (mean fluorescence intensity).

The Posttransplant Donor-Specific Cellular Immunity Correlated with the β-Score

We first assessed the level of the immune reactivity of islet-transplanted patients against irradiated donors and third party cells. IFN-γ-ELISpot and proliferation assays were used to determine this level. These data were correlated to the clinical outcome assessed by the β-score (Fig. 2). The presence of a donor-specific cellular immunity detected was associated with a low β-score (IFN-γ-ELISpot; p = 0.0073, r = −0.495 and proliferation; p = 0.0061 r = −0.505). In order to demonstrate the specificity of the detected immune reactivity, the same analysis was performed after third party stimulation and no correlation could be found (ELISpot, p = 0.61, r = 0.146; proliferation, p = 0.83, r = −0.0604). In three tested islet after kidney (IAK) patients, no immune reactivity was detected against kidney donors.

Figure 2.

Correlation between posttransplant donor-specific cellular immunity and β-score level. The assessment of cell activity was performed by interferon-γ (IFN-γ)-ELISpot (A) and proliferation (B). Correlation was assessed with the Spearman test (n = 28 with several recipients tested against multiple donors).

Phenotype of Donor-Specific Reactive Cells

Despite the use of immunosuppression, patients demonstrated cell ratios corresponding to those of control subjects, with a small decrease in the CD4-expressing subset (CD3⁺CD4⁺) (Fig. 3A). This observation is likely related to the long time between transplantation and assessment.

Figure 3.

Phenotype of donor-specific reactive cells. (A) Proportion of CD3⁺CD4⁺, CD3⁺CD8⁺, and CD3^–CD56⁺ cells in peripheral blood mononuclear cells (PBMCs) of islet transplant patients (n = 14, white bars) compared to healthy individuals (n = 9, gray bars, ∗p = 0.02). (B) Proportion of each cell subset in the population of IFN-γ-producing cells (left) and proportion of IFN-γ-producing cells in each cell subset (right), n = 5. (C) Proportion of each cell subset in the population of proliferating cells (left) and proportion of proliferating cells in each cell subset (right), n = 28. (D) Proportion of each cell subset in the population of proliferating cells compared between patients with low (<4) and high (>4) β-scores (left, ∗∗∗p = 0.0001). Proportion of proliferating cells in each cell subset compared between patients with low (<4) and high (>4) β-scores (right, ∗∗∗p = 0.009). Groups were compared using the Mann–Whitney test.

In order to assess which cell populations were activated in presence of donor antigens, flow cytometry studies were performed. IFN-γ-producing cells were not easily detected by flow cytometry, and only few patients with a very high cytokine secretion level could be analyzed (five patients). In these patients, IFN-γ production was mainly driven by CD4⁺ cells (35.9 ± 3.4%) (Fig. 3B, left). The proportion of IFN-γ-producing cells was enhanced in the natural killer (NK) cells subset (CD3^-CD56⁺) and only represented a small part of the CD4⁺ and CD8⁺ subsets (0.23%, 0.079%, 0.069%) (Fig. 3B, right).

Regarding the level of proliferation, Ki67⁺ cells were mainly CD4⁺ lymphocyte and NK cells (24.6 ± 2.7% and 17.3 ± 2.1%, respectively) (Fig. 3C, left). Only 6.8 ± 1.2% of proliferating cells were CD3⁺CD8⁺. Similar to the IFN-γ-producing cells, proliferating cells were more frequent in the NK cell subset, compared to the CD4⁺ and CD8⁺ cell subsets (6.19%, 0.80%, and 0.48%, respectively) (Fig. 3C, right).

In an effort to understand whether an alteration of these populations would correlate to the clinical outcome, low (≤4) and high (>4) β-score patients were further compared (Fig. 3D, left). Patients with the worse islet function showed increased levels of CD4⁺Ki67⁺ cells (42.8% vs. 15.7%, p = 0.0001). No significant differences were observed among CD8⁺Ki67⁺ and CD56⁺Ki67⁺ cells (6.7% vs. 4.2%, p = 0.91 and 16.3% vs. 15.8%, p = 0.32).

Finally, to ensure that this increase was not related to a disruption of the homeostasis in the total lymphoid subset, the proportion of Ki67-expressing cells in each cell subset was assessed (Fig. 3D, right). The proportion of Ki67⁺ cells observed in the CD4⁺ population after donor-specific stimulation was increased in patients with a low β-score (0.668% vs. 0.218%, p = 0.009). No significant differences were observed in the CD8⁺ and CD56⁺ cell populations (p = 0.78 and p = 0.64, respectively). No difference was found after stimulation with third party in all cell subsets.

Cytokine Profile

In order to better understand the alloimmune profile of the patients, we analyzed a panel of cytokines in the supernatants following a 72-h incubation with donor cells. IL-5, IL-17, IFN-γ, and IL-10 were analyzed (Fig. 4). Similar to the ELISpot data, significantly higher IFN-γ levels were detected in patients with low islet function (260 pg/ml vs. 16.6 pg/ml, p = 0.029). The secretion of IL-5, a Th2-type cytokine, was also increased in patients with poor function (48.2 pg/ml vs. 1.77 pg/ml, p = 0.026).

Figure 4.

Cytokine profile following 3-day mixed lymphocyte reaction with donor cells. (A) Concentration of cytokines [IFN-γ, interleukin 5 (IL-5), IL-17, and IL-10] in supernatants were compared between patients with low (<4, white bars) and high (>4, black bars) β-scores [n = 28, ∗p = 0.02 (IFN-γ), p = 0.02 (IL-5)]. (B) Cytokine ratios (IFN-γ/IL-10, IL-5/IL-10, and IL-17/IL-10) were compared between patients with low (<4) and high (>4) β-scores [∗∗∗p = 0.0006 (IFN-γ/IL-10), ∗p = 0.014 (IL-5/IL-10)]. Groups were compared using Mann–Whitney test.

Interestingly, IL-17 levels tended to be increased in patients with low function, but this pattern could only be assessed in a limited number of patients leading to the absence of statistically significant difference (3.74 pg/ml vs. 1.77 pg/ml, p = 0.4). Conversely, a trend towards higher IL-10 levels was detected in patients with good islet function (1.79 pg/ml vs. 6.14 pg/ml, p = 0.7).

Overall, the ratio between activating cytokines and IL-10, which is considered as an immunoregulatory cytokine, could predict the immune status of the patients. IFN-γ/IL-10 ratio was higher in patients with low β-score (144.9 vs. 2.7, p = 0.0006). Similarly, the IL-5/IL-10 ratio was significantly increased in these patients (26.8 vs. 0.28, p = 0.014), but no significant differences were observed in IL-17/IL-10 ratio (2.08 vs. 0.29, p = 0.14).

Absence of Significant Correlation Between the Level of Cell Immune Reactivity and the Number of Islet Donors

The impact of repeated islet infusions was evaluated by the donor-specific reactivity associated to the first (Tx1), second (Tx2), or third (Tx3) transplantation. The degree of cellular reactivity tended to increase with the number of islet infusions as tested by IFN-γ-ELISpot (Tx1: 35.7 ± 23.3 to Tx3: 72.08 ± 50.7) (Fig. 5A) and proliferation (Tx1: 5.59 ± 2.48 to Tx3: 12.8 ± 4.46) (Fig. 5B), but no significant difference could be observed (Kruskal–Wallis test, p = 0.895 and 0.524, respectively). This lack of correlation may in part be related to the small sample size.

Figure 5.

IFN-γ-secreting cells (A) and proliferation index (B) were assessed according to the number of transplantation (Tx1, Tx2, Tx3)(n = 28). Reactivity against third party was used as control (n = 14). Groups were compared using the Kruskal–Wallis test.

Accuracy of the IFN-γ-ELISpot and Proliferation Assays in Predicting Clinical Outcome

The accuracies of the IFN-γ-ELISpot, all immune cell proliferation, and cytokine ratio in predicting clinical outcome were assessed using an ROC curve (β-scores of <4 and >4 were entered in the model, n = 28) (Fig. 6). They demonstrated similar accuracies in predicting clinical outcome (receiver operative characteristics curve AUC = 0.77 ± 0.088, 0.85 ± 0.084, and 0.88 ± 0.074, respectively).

Figure 6.

Receiver operating characteristic (ROC) curves assessing the accuracy of IFN-γ-ELISpot [area under curve (AUC) = 0.77 ± 0.088], immune cell proliferation (0.85 ± 0.084), and IFN-γ/IL-10 (0.88 ± 0.074) ratio in predicting clinical outcome.

With a cut-off at 9 spots per 5.10⁵ cells, the IFN-γ-ELISpot assay had a sensitivity of 72% and a specificity of 70%. With a cut-off at 8 (stimulation index), the proliferation assay had a sensitivity of 72% and a specificity of 88%. With a cut-off at 15, the IFN-γ/IL-10 ratio assay had a sensitivity of 84% and a specificity of 80%.

Discussion

This study demonstrates that the level of cellular immune reactivity assessed against donor antigens by ELISpot, cell proliferation, and cytokine profile correlates with the clinical outcome after allogeneic islet transplantation.

The present study was based on previously published data suggesting that donor antigen-stimulated IFN-γ-ELISpot can allow for an accurate immune monitoring after mouse islet transplantation (32). Until now, the clinical use of the ELISpot technology in the field of islet transplantation (with the detection of IFN-γ and IL-2-producing cells) has almost exclusively been reported for a pretransplant assessment, and the level of pretransplant cellular alloreactivity has been associated to worse posttransplant outcomes (19). The present study assessed the use of a posttransplant immune monitoring as a potential marker of islet loss. ELISpot, cell proliferation, and cytokine ratio demonstrated similar accuracies in predicting islet function. However, while the ELISpot technology is less time-consuming (one day incubation against three for the other two) and easier to perform (no need of flow cytometry, radioactivity, or bioplex reader), we would favor this assay for a broader validation and use. Of note, these islet transplant observations are in keeping with data suggesting a good correlation between donor-ELISpot reactivity and creatinine level after kidney transplantation (10,22).

IFN-γ and IFN-γ-producing cells are known as the main promoters of rejection, and we favor the detection of this cytokine for an accurate posttransplant immune monitoring (7). However, inflammation or autoimmune recurrence can lead to the secretion of other types of cytokines, which contribute to the loss of islet. In this regard, IL-5 levels were increased in patients with low islet function. The presence of this Th2-associated cytokine may be related to the observation that islet allograft rejection can be mediated by CD4^- cells with both Th1 and Th2 cytokine phenotypes (1). In addition, a Th2 response appears predominant in a chronic allograft rejection, which may have been present in some studied patients (18,21).

IL-17 demonstrated a trend towards higher levels in patients with low islet function (and a statistical significance may have been reached with a larger sample). This observation may reflect some degree of autoimmune recurrence in the studied patients, as higher levels of IL-17 have been detected in various clinical autoimmune diseases and at the time of autoimmune recurrence after islet transplantation in nonobese diabetic (NOD) mice (5,32,38). Of further note, IL-17 has also been detected in case of pure allogeneic rejection, which could also explain the presence of this cytokine (17,38). Overall, a combined assessment of both allo- and autoimmunity should be further explored, which may also be performed with the use of multimer assays (37).

IL-10 is commonly considered as one of the major cytokines associated with the presence of regulatory cells and the active suppression of the immune response (27). The IFN-γ/IL-10 ratio has been reported as an important measure of the immune reactivity in type 1 diabetes and islet transplantation (16,35). Huurman et al. have demonstrated lower posttransplant IFN-γ/IL-10 ratios in patients reaching insulin independence (12). This observation is confirmed by the present data, and the assessment of this ratio could potentially improve the ELISpot accuracy.

While immunosuppression could impact the outcome of ex vivo immune tests, the present data and results from others suggest that accurate results can be achieved despite the presence of immunosuppressive drugs (20). In addition, an accurate assessment could be achieved with the study of the relative proportion of proliferating (Ki67-expressing) cells among the CD3⁺CD4⁺ cell subtype. Such an assessment has the potential to be performed even after the use of depleting agents and deserves further validation.

The main limitation of the study is its transversal assessment of the patients, with some of them being several years after the islet loss. Along the same line, the duration of follow-up was different between groups, potentially leading to differing levels of immunosuppression. Overall, it is impossible to tell whether the detected alloreactivity was a cause or a consequence of the islet loss, and a further prospective longitudinal validation is now required. However, it is of great interest to see that the fingerprint of alloreactivity can persist even years after the loss of islet graft function. This suggests that our in vitro assays can be used at various time points after transplantation to monitor the level of immunity and better tune immunosuppression.

Overall, the level of cellular immune reactivity assessed against donor antigens by ELISpot, proliferation, and cytokine profile correlates with the clinical outcome after allogeneic islet transplantation. These strategies (and especially the easier and less time-consuming ELISpot) have the potential to improve the immune monitoring after islet transplantation and deserve a prospective longitudinal validation.

Footnotes

Acknowledgments

C.T. was supported by the Swiss National Science Foundation (SCORE grant 3232230-126233). The study was funded by a grant “Projet Recherche et Développement” from the Geneva University Hospitals. The authors declare no conflicts of interest.

References

Barbara

J. A.

Turvey

S. E.

Kingsley

C. I.

Spriewald

B. M.

Hara

Witzke

Morris

P. J.

Wood

K. J.

Islet allograft rejection can be mediated by CD4⁺, alloantigen experienced, direct pathway T cells of TH1 and TH2 cytokine phenotype. Transplantation 70: 1641–1649; 2000.

Bosi

Braghi

Maffi

Scirpoli

Bertuzzi

Pozza

Secchi

Bonifacio

Autoantibody response to islet transplantation in type 1 diabetes. Diabetes 50: 2464–2471; 2001.

Campbell

P. M.

Salam

Ryan

E. A.

Senior

Paty

B. W.

Bigam

McCready

Halpin

Imes

Al Saif

Lakey

J. R. T.

Shapiro

A. M. J.

Pretransplant HLA antibodies are associated with reduced graft survival after clinical islet transplantation. Am. J. Transplant. 7: 1242–1248; 2007.

Eich

Eriksson

Lundgren

Visualization of early engraftment in clinical islet transplantation by positron-emission tomography. N. Engl. J. Med. 356: 2754–2755; 2007.

Emamaullee

J. A.

Davis

Merani

Toso

Elliott

J. F.

Thiesen

Shapiro

A. M. J.

Inhibition of Th17 cells regulates autoimmune diabetes in NOD mice. Diabetes 58: 1302–1311; 2009.

Ferrari-Lacraz

Berney

Morel

Marangon

Hadaya

Demuylder-Mischler

Pongratz

Pernin

Villard

Low risk of anti-human leukocyte antigen antibody sensitization after combined kidney and islet transplantation. Transplantation 86: 357–359; 2008.

Halloran

P. F.

de Freitas

D. G.

Einecke

Famulski

K. S.

Hidalgo

L. G.

Mengel

Reeve

Sellares

Sis

The molecular phenotype of kidney transplants. Am. J. Transplant. 10: 2215–2222; 2010.

Hering

B. J.

Kandaswamy

Ansite

J. D.

Eckman

P. M.

Nakano

Sawada

Matsumoto

Ihm

S.-H.

Zhang

H.-J.

Parkey

Hunter

D. W.

Sutherland

D. E. R.

Single-donor, marginal-dose islet transplantation in patients with type 1 diabetes. JAMA 293: 830–835; 2005.

Hilbrands

Huurman

V. A. L.

Gillard

Velthuis

J. H. L.

De Waele

Mathieu

Kaufman

Pipeleers-Marichal

Ling

Movahedi

Jacobs-Tulleneers-Thevissen

Monbaliu

Ysebaert

Gorus

F. K.

Roep

B. O.

Pipeleers

D. G.

Keymeulen

Differences in baseline lymphocyte counts and autoreactivity are associated with differences in outcome of islet cell transplantation in type 1 diabetic patients. Diabetes 58: 2267–2276; 2009.

10.

Hricik

D. E.

Rodriguez

Riley

Bryan

Tary-Lehmann

Greenspan

Dejelo

Schulak

J. A.

Heeger

P. S.

Enzyme linked immunosorbent spot (ELISpot) assay for interferon-gamma independently predicts renal function in kidney transplant recipients. Am. J. Transplant. 3: 878–884; 2003.

11.

Huurman

V. A. L.

Hilbrands

Pinkse

G. G. M.

Gillard

Duinkerken

van de Linde

van der Meer-Prins

P. M. W.

Versteeg-van der Voort Maarschalk

M. F. J.

Verbeeck

Alizadeh

B. Z.

Mathieu

Gorus

F. K.

Roelen

D. L.

Claas

F. H. J.

Keymeulen

Pipeleers

D. G.

Roep

B. O.

Cellular islet autoimmunity associates with clinical outcome of islet cell transplantation. PLoS ONE 3: e2435; 2008.

12.

Huurman

V. A. L.

Velthuis

J. H. L.

Hilbrands

Tree

T. I. M.

Gillard

van der Meer-Prins

P. M. W.

Duinkerken

Pinkse

G. G. M.

Keymeulen

Roelen

D. L.

Claas

F. H. J.

Pipeleers

D. G.

Roep

B. O.

Allograft-specific cytokine profiles associate with clinical outcome after islet cell transplantation. Am. J. Transplant. 9: 382–388; 2009.

13.

Jaeger

Brendel

M. D.

Hering

B. J.

Eckhard

Federlin

Bretzel

R. G.

IA2 antibodies are only positive in association with GAD 65 and islet cell antibodies in islet transplanted insulin-dependent diabetes mellitus patients. Transplant. Proc. 30: 659–660; 1998.

14.

Keymeulen

Ling

Gorus

F. K.

Delvaux

Bouwens

Grupping

Hendrieckx

Pipeleers-Marichal

Van Schravendijk

Salmela

Pipeleers

D. G.

Implantation of standardized beta-cell grafts in a liver segment of IDDM patients: Graft and recipients characteristics in two cases of insulin-independence under maintenance immunosuppression for prior kidney graft. Diabetologia 41: 452–459; 1998.

15.

Lacotte

Berney

Shapiro

A. J.

Toso

Immune monitoring of pancreatic islet graft: Towards a better understanding, detection and treatment of harmful events. Expert Opin. Biol. Ther. 11: 55–66; 2011.

16.

Lindley

Dayan

C. M.

Bishop

Roep

B. O.

Peakman

Tree

T. I. M.

Defective suppressor function in CD4(+)CD25(+) T-cells from patients with type 1 diabetes. Diabetes 54: 92–99; 2005.

17.

Loverre

Tataranni

Castellano

Divella

Battaglia

Ditonno

Corcelli

Mangino

Gesualdo

Schena

F. P.

Grandaliano

IL-17 expression by tubular epithelial cells in renal transplant recipients with acute antibody-mediated rejection. Am. J. Transplant. 11: 1248–1259; 2011.

18.

Mhoyan

G. D.

Kakoulidis

T. P.

Que

Yolcu

E. S.

Cramer

D. V.

Shirwan

Predominant expression of the Th2 response in chronic cardiac allograft rejection. Transpl. Int. 16: 464–473; 2003.

19.

Mohanakumar

Narayanan

Desai

Ramachandran

Shenoy

Jendrisak

Susskind

B. M.

Olack

Benshoff

Phelan

D. L.

Brennan

D. C.

Fernandez

L. A.

Odorico

J. S.

Polonsky

K. S.

A significant role for histocompatibility in human islet transplantation. Transplantation. 82: 180–187; 2006.

20.

Monti

Scirpoli

Maffi

Ghidoli

De Taddeo

Bertuzzi

Piemonti

Falcone

Secchi

Bonifacio

Islet transplantation in patients with autoimmune diabetes induces homeostatic cytokines that expand autoreactive memory T cells. J. Clin. Invest. 118: 1806–1814; 2008.

21.

Nath

D. S.

Tiriveedhi

Basha

H. I.

Phelan

Moazami

Ewald

G. A.

Mohanakumar

A role for antibodies to human leukocyte antigens, collagen-v, and k-α1-tubulin in antibody-mediated rejection and cardiac allograft vasculopathy. Transplantation 91: 1036–1043; 2011.

22.

Nickel

Presber

Bold

Biti

Schönemann

Tullius

S. G.

Volk

H.-D.

Reinke

Enzyme-linked immunosorbent spot assay for donor-reactive interferon-gamma-producing cells identifies T-cell presensitization and correlates with graft function at 6 and 12 months in renal-transplant recipients. Transplantation 78: 1640–1646; 2004.

23.

Pattou

Kerr-Conte

Wild

GLP-1-receptor scanning for imaging of human beta cells transplanted in muscle. N. Engl. J. Med. 363: 1289–1290; 2010.

24.

Roelen

D. L.

Huurman

V. A. L.

Hilbrands

Gillard

Duinkerken

van der Meer-Prins

P. W. M.

Versteeg-van der Voort Maarschalk

M. F. J.

Mathieu

Keymeulen

Pipeleers

D. G.

Roep

B. O.

Claas

F. H. J.

Relevance of cytotoxic alloreactivity under different immunosuppressive regimens in clinical islet cell transplantation. Clin. Exp. Immunol. 156: 141–148; 2009.

25.

Roep

B. O.

Stobbe

Duinkerken

van Rood

J. J.

Lernmark

Keymeulen

Pipeleers

Claas

F. H.

de Vries

R. R.

Auto- and alloimmune reactivity to human islet allografts transplanted into type 1 diabetic patients. Diabetes 48: 484–490; 1999.

26.

Ryan

E. A.

Paty

B. W.

Senior

P. A.

Lakey

J. R. T.

Bigam

Shapiro

A. M. J.

Beta-score: An assessment of beta-cell function after islet transplantation. Diabetes Care 28: 343–347; 2005.

27.

Sabat

Grütz

Warszawska

Kirsch

Witte

Wolk

Geginat

Biology of interleukin-10. Cytokine Growth Factor Rev. 21: 331–344; 2010.

28.

Toso

Baertschiger

Morel

Bosco

Armanet

Wojtusciszyn

Badet

Philippe

Becker

C. D.

Hadaya

Majno

Bühler

Berney

Sequential kidney/islet transplantation: Efficacy and safety assessment of a steroid-free immunosuppression protocol. Am. J. Transplant. 6: 1049–1058; 2006.

29.

Toso

Edgar

Pawlick

Emamaullee

Merani

Dinyari

Mueller

T. F.

Shapiro

A. M. J.

Anderson

C. C.

Effect of different induction strategies on effector, regulatory and memory lymphocyte subpopulations in clinical islet transplantation. Transpl. Int. 22: 182–191; 2009.

30.

Toso

Isse

Demetris

A. J.

Dinyari

Koh

Imes

Kin

Emamaullee

Senior

Shapiro

A. M. J.

Histologic graft assessment after clinical islet transplantation. Transplantation 88: 1286–1293; 2009.

31.

Toso

Oberholzer

Ris

Triponez

Bucher

Demirag

Andereggen

Buehler

Cretin

Fournier

Majno

Hong

Lou

Morel

Factors affecting human islet of Langerhans isolation yields. Transplant. Proc. 34: 826–827; 2002.

32.

Toso

Pawlick

Lacotte

Edgar

Davis

McCall

Morel

Mentha

Berney

Shapiro

A. M. J.

Detecting rejection after mouse Islet transplantation utilizing islet protein-t ELISpot. Cell Transplant. 20: 955–962; 2011.

33.

Toso

Vallee

J.-P.

Morel

Ris

Demuylder-Mischler

Lepetit-Coiffe

Marangon

Saudek

James Shapiro

A. M.

Bosco

Berney

Clinical magnetic resonance imaging of pancreatic islet grafts after iron nanoparticle labeling. Am. J. Transplant. 8: 701–706; 2008.

34.

Toso

Zaidi

Morel

Armanet

Andres

Pernin

Baertschiger

Slosman

Bühler

L. H.

Bosco

Berney

Positron-emission tomography imaging of early events after transplantation of islets of Langerhans. Transplantation 79: 353–355; 2005.

35.

van den Boogaardt

D. E. M.

van Miert

P. P. M. C.

de Vaal

Y. J. H.

de Fijter

J. W.

Claas

F. H. J.

Roelen

D. L.

The ratio of interferon-gamma and interleukin-10 producing donor-specific cells as an in vitro monitoring tool for renal transplant patients. Transplantation 82: 844–848; 2006.

36.

van Kampen

C. A.

van de Linde

Duinkerken

van Schip

J. J.

Roelen

D. L.

Keymeulen

Pipeleers

D. G.

Claas

F. H. J.

Roep

B. O.

Alloreactivity against repeated HLA mismatches of sequential islet grafts transplanted in non-uremic type 1 diabetes patients. Transplantation 80: 118–126; 2005.

37.

Velthuis

J. H.

Unger

W. W.

Abreu

J. R. F.

Duinkerken

Franken

Peakman

Bakker

A. H.

Reker-Hadrup

Keymeulen

Drijfhout

J. W.

Schumacher

T. N.

Roep

B. O.

Simultaneous detection of circulating autoreactive CD8⁺ T-cells specific for different islet cell-associated epitopes using combinatorial MHC multimers. Diabetes 59: 1721–1730; 2010.

38.

Wang

Yang

Sheng

Chen

Tang

Sheng

Qin Zang

T-cell vaccination leads to suppression of intrapancreatic Th17 cells through Stat3-mediated RORγt inhibition in autoimmune diabetes. Cell Res. 21: 1358–1369; 2011.