SEARCH

SEARCH BY CITATION

Keywords:

  • Alloantibodies;
  • clinical outcome;
  • donor-reactive DTH;
  • kidney;
  • simultaneous kidney-pancreas;
  • transplantation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Evidence of transplant recipient cellular sensitization towards donor antigens has rarely been directly measured. Rather, sensitization has been generally inferred by the presence of detectable allo-reactive or donor-reactive antibodies. In this study a newly developed delayed-type hypersensitivity assay was used to directly determine the incidence of post-transplant donor-reactive T-cell sensitization in a large cohort of kidney and simultaneous kidney-pancreas recipients. These results were compared with the presence of detectable circulating alloantibodies and with patient clinical outcome. We found an unexpectedly high incidence (52%) of donor-reactive delayed-type hypersensitivity reactivity in our study patients. Donor-reactive delayed-type hypersensitivity reactivity occurred at a much higher frequency than detectable alloantibodies (20%). Further, we found no correlation between the presence of alloantibodies and donor-reactive delayed-type hypersensitivity reactivity. We also found no correlation between the development of donor-reactive delayed-type hypersensitivity reactivity and the degree of donor and recipient HLA matching. Finally, the presence of detectable donor-reactive delayed-type hypersensitivity reactivity did not correlate with a worse clinical outcome at the time of these analyses. We conclude that in transplant recipients, the presence of circulating alloantibodies is a poor indicator of previous T-cell sensitization to donor antigens. We also conclude that our current immunosuppression strategies are relatively ineffective at blocking T-cell allosensitization, but are very effective at blocking the biological consequences of that allosensitization.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

In transplant recipients, graft alloantigens can stimulate naïve T cells to become activated, initiating a process that can result in acute rejection of the allograft. The progeny of these alloactivated T cells are distributed throughout the recipient immune system as ‘memory’ T cells, and the recipient becomes cellularly ‘sensitized’ to graft alloantigens. Clinical immunosuppression is designed to prevent the development of acute rejection by derailing this allosensitization process. When immunosuppression is suboptimal, allosensitization may occur and manifest itself as ‘subclinical’ acute rejection, leading to an accumulation of graft damage over a protracted period of time [reviewed by (1)]. Therefore, it would be of clinical interest to know whether or not a transplant recipient has become sensitized to their donor alloantigens. Such knowledge would help the physician understand the immune disposition of the recipient toward the allograft prior to the overt development of acute or chronic rejection.

In humans, the presence of allo-reactive T cells can be demonstrated with either mixed lymphocyte culture (MLC) or cytotoxic T-lymphocyte (CTL) analyses. These methods cannot discriminate between responses made by naïve vs. memory T cells. Thus they cannot discriminate between ‘alloreactivity’, the potential to make an immune response to alloantigens, and previous allosensitization, the result of prior functional contact with alloantigens. In fact, few studies have addressed the issue of how to detect allosensitized T cells within the pool of alloreactive T cells. Unfortunately, this is the key to understanding immune events that develop during the post-transplant period. One analytic method that may allow this is delayed-type hypersensitivity (DTH). DTH responses do not occur unless the host has been previously sensitized to the challenge antigens. Thus, DTH assays should provide a direct index of T-cell allosensitization to donor alloantigens.

The presence of donor-reactive alloantibodies in recipient serum has been widely accepted as an indirect index of T-cell sensitization. This is because IgG production by alloantigen-stimulated B cells requires input from alloantigen-reactive T cells [reviewed by (2)]. The presence of alloreactive IgG antibodies is easily detected in recipient serum. However, because there is no generally accepted method for direct determination of T-cell allosensitization, there are no previous studies to verify the relationship between alloreactive IgG production and T-cell allosensitization in transplant patients. Further, there are no previous studies that relate a direct index of T-cell allosensitization to the post-transplant clinical course of transplant patients.

In studies with murine cardiac allograft recipients, we observed that delayed type hypersensitivity responses (DTH) could be used to monitor T-cell allosensitization. Specifically, mice that rejected cardiac allografts displayed donor-reactive DTH responses, whereas naïve, nontransplanted mice, and immunosuppressed mice that accepted allografts failed to do so (3). In these mice, DTH reactivity did not always correlate with the production of donor-reactive IgG. While naïve, nontransplanted mice never displayed donor-reactive alloantibodies, cardiac allograft recipients display a wide range of donor-reactive IgG titers, ranging from very high to undetectable, whether or not the allograft is rejected (4). These studies suggested that alloantibody production might not be an accurate index of pro-inflammatory T-cell allosensitization in allograft recipients. They also confirmed that donor-reactive DTH reactivity might provide a reasonable index of pro-inflammatory donor-reactive T-cell allosensitization in transplant patients. However, the subcutaneous placement of donor alloantigens in allograft recipients to elicit a DTH response could result in unwanted allosensitization. To avoid this, we developed the trans-vivo DTH assay, in which human peripheral blood mononuclear cells (PBMC) are placed subcutaneously in the pinnae or footpads of mice, together with challenge antigens. If the PBMC donor was previously sensitized to the challenge antigens, the PBMC initiate a measurable DTH-like swelling response within 24 h. Human DTH reactivity to virtually any antigen can be tested in this manner, including alloantigens (4,5).

For the studies outlined in this report, we used the trans-vivo DTH assay in a large cohort of transplant recipients to: (i) determine the incidence of post-transplant donor-reactive T-cell reactivity; (ii) compare the expression of donor-reactive DTH responses with the presence of detectable alloantibodies; and (iii) evaluate the relationship between their post-transplant donor-reactive T-cell allosensitization status and the clinical outcome.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Patients

Included in this study were 180 primary recipients of a kidney or simultaneous kidney and pancreas organ(s) transplanted between 6/1983 and 6/2000. The demographics of this cohort of patients are shown in Table 1. One sample of PBMC was analyzed from each patient, obtained either during routine outpatient evaluation (n = 151) or during an admission to the hospital (n = 29). For the entire cohort of 180 patients, the PBMC samples were obtained a median of 2.5 years after transplantation (mean 3.9 years, range 5 months to 17 years). All PBMC donations were obtained following Institutional Review Board (IRB) approved informed consent.

Table 1. : Patient demographics
Age (years)a42.2 ± 12.8
  • a

    Mean ± standard deviation;

  • b

    AA – African-American;

  • c

    Kid-LRD – living donor kidney recipient;

  • d

    Kid-CAD – cadaveric donor kidney recipient;

  • e

    SPK – simultaneous kidney-pancreas recipient;

  • f

    HLA – human leukocyte antigen;

  • g

    IL2-R – interleukin 2 receptor, OKT3 – Muromonab-CD3, ATGAM – antithymocyte globulin

  • h

    CSA – cyclosporine, MMF – mycophenolate mofetil, AZA – azathioprine, pred – prednisone.

Gender (m/f)111/69
Race (AA/other)b 25/155
Type of donor organ(s)
 Kid-LRDc (n) 25
 Kid-CADd (n)110
 SPKe (n) 45
HLAf mismatcha4.4 ± 1.7
Inductiong
 IL2-R inhibitor 70
 OKT-3 48
 ATGAM 41
 none 21
Maintenanceh
 CSA/MMF/pred133
 CSA/AZA/pred 38
 CSA/pred  9
Kidney loss (n)  8
Pancreas loss (n)  5
Death (n)  2

Acute rejection

Acute rejection episodes were diagnosed clinically based on significant renal dysfunction as determined by a ≥ 25% rise in serum creatinine, and biopsy proven prior to treatment.

Chronic rejection

Chronic rejection was diagnosed by clinical criteria and confirmed by biopsy in 9 of 12 patients. All diagnosed patients experienced an otherwise unexplained rise in serum creatinine (mean of 4.7 ± 3.0, range 2.6–13). All patients were hypertensive at the time of diagnosis, requiring treatment with multiple antihypertensive medications (mean 2.9 ± 1.2, range 1–5). Significant proteinuria (> 500 mg/24 h) was present in 10/11 patients who had a 24-h urine available for analysis (mean 2.67 ± 3.25 g, range 0.41–12.2 g). Biopsy was available in 9 of 12 patients. Six biopsies revealed grade III, and 3 biopsies revealed grade II chronic/sclerosing allograft nephropathy using the Banff 97 working classification.

Peripheral blood mononuclear cell isolation

Blood was sterilely collected from transplant patients into citrate tubes. PBMC were isolated by Ficoll-Hypaque density gradient centrifugation. Cells were washed ×3 and resuspended in phosphate-buffered saline (PBS).

Subcellular donor antigen preparation

Antigen was prepared, as previously described (4), from donor cells isolated by Ficoll-Hypaque density gradient centrifugation of splenocyte or lymph node cell suspensions (cadaveric donors) or sterilely collected peripheral blood collected in citrate tubes (living donors). Briefly, the isolated cells were washed in PBS and adjusted to 120 × 106 cells/mL. The protease inhibitor phenyl methyl-sulfonyl fluoride (PMSF, Boehringer Manheim Corp., Indianapolis, USA) was added (33 mm/mL) just prior to sonication. The cells were disrupted using a VR5 Vibra-cell sonicator with a 2-mm microtip (Sonics & Materials, Inc., Danbury, CT, USA). The sonicate was centrifuged for 20 min at 14 000 g to remove debris. The protein content of the supernatant was determined using standard methods. Ten micrograms of protein was used for each injection.

Epstein-Barr virus (EBV) antigen lysate

Epstein-Barr virus antigen lysate was prepared from induced P3 H1 cells and was a kind gift from Bob Neagele (Viral Antigens, Inc., Memphis, TN, USA) as previously described. Ten micrograms of protein was used for each injection. Previous experiments have shown that EBV IgG seropositive individuals respond to the EBV lysate in the trans-vivo DTH assay (swelling response), whereas EBV IgG seronegative individuals do not.

Trans-vivo DTH analysis

To determine transplant recipient donor-reactive DTH responses, 6–9 × 106 recipient PBMC were injected, along with 10 µg of appropriate antigen, into the pinnae or footpads of naïve C57BL/6 mice as previously described (5). Briefly, 50 μL of solution containing recipient PBMC alone (negative control) or mixed with either EBV antigen (positive control) or donor antigen was injected using a 28-gauge needle. Care was taken to reduce the platelet contamination below 107 platelets/injection. Previous studies have shown that > 107 platelets/injection can result in nonspecific swelling (5). Swelling was measured after 24 h using a dial thickness gauge (Swiss Precision Instruments, Carlstadt, NJ, USA). Pre-injection measurement was subtracted from the 24- h measurement to calculate the change in thickness (reported in units of 10−4 inches). In all instances where naïve mice have been challenged with either tetanous toxoid or EBV antigen alone, no swelling response has been observed. For each patient, antigen-specific swelling responses were determined by subtracting the change in thickness from the PBMC alone injection (negative control) from the change in thickness obtained from PBMC plus antigen. Swelling responses ≥ 20 × 10−4 above negative control were considered positive. This value was chosen because it exceeds the mean plus 2 standard deviations of the swelling observed when saline alone or recipient PBMC alone are injected [reviewed in (5)].

Flow cytometric alloantibody detection

A commercially available pool of microparticle beads coated with various purified MHC antigens of known specificity, were used according to manufacturers instructions (FlowPRA, OneLambda, Canoga Park, CA, USA). Briefly, 20 μL of recipient sera was incubated with 5 μL of MHC class I plus 5 μL of MHC class II microparticle beads for 30 min at room temperature (RT). The beads were washed twice with buffer and centrifuged at 10 000 r.p.m. for 2 min. The beads were re-suspended in 100 μL of solution containing FITC-conjugated goat antihuman IgG and incubated for 30 min at RT. The wash step was repeated and the beads were re-suspended in 500 μL of wash buffer. Negative control serum using pooled sera from nontransfused males was similarly prepared. Samples were read with the aid of a Beckman Coulter XL2 flow cytometer. The fluorescence profile obtained with negative control sera was used as the baseline fluorescence. MHC class I and class II beads were readily distinguishable since they are fluorescent (excited at 488 nm and maximum emission at 580 nm) and have unique emission spectra. The positive/negative cutoff was empirically determined for each assay by setting a histogram region that excluded 98% of the peak obtained with the negative control serum. The median channel associated with this cut point was recorded for each assay. A test was deemed positive for alloantibody if there was noted a distinct peak or if there was a shift to the right in bead fluorescence of ≥ 6% to the right of the cutoff point.

Statistical analysis

Student's t-test and the Pearson chi-square test were used for statistical comparison of means (± SEM) and proportions between groups, respectively, where appropriate. Logistic regression analysis using a DTH positive result as the dependent variable was used where indicated. All statistical analyses performed using SPSS version 10.0.5 statistical software (Chicago, IL, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Representative results of the trans-vivo DTH assay

Representative results from the trans-vivo DTH assay, performed in triplicate for each patient, are shown in Figure 1. Patient no. 1 was chosen to exemplify the results obtained when DTH reactivity to donor antigens in this assay is lacking. Patient no. 2 was chosen to exemplify the results obtained when donor-reactive DTH reactivity is demonstrated. When PBMC derived from patient no. 1 were placed in the pinnae of a C57BL/6 mouse with no challenge antigen, only minimal background swelling occurred. This demonstrates that neither the human nor the murine cells respond detectably to each other in this assay. A minimal level of swelling also developed when the PBMC were challenged with donor alloantigens, indicating a lack of pro-inflammatory T-cell reactivity to donor antigens. This failure to mount a DTH response was not the result of immunosuppressive effects or an incompetent immune system, since the same PBMC challenged with EBV, a positive control antigen to which most individuals are sensitized, generated a strong swelling response. Because of this pattern of responses, this individual is considered to be DTH negative for donor alloantigens. Patient no. 2 demonstrates the alternative assay result. In the absence of challenge antigen, the PBMC from this patient again failed to induce a swelling response. However, the same PBMC promoted a strong swelling response when challenged with either EBV or donor alloantigens. This individual is considered to be DTH positive for donor alloantigens. Results of the trans-vivo DTH are easily reproducible, as is demonstrated with these two patients. In an additional six patients, repeat DTH assays were run less than 10 days apart. In 6/6 of these patients, the DTH results were the same for both tests (data not shown).

image

Figure 1. Post-transplant trans-vivo DTH assay results from two selected patients exemplifying the presence or absence of donor-reactive DTH reactivity. The results for each patient are expressed as a mean of triplicate testing. The change in ear swelling (in 10−4 inches, Y axis) for ears injected with recipient PMBC alone, recipient PBMC plus EBV antigen, and recipient PBMC plus sonicated donor antigens (see Methods). Patient no. 1 demonstrates EBV-reactive DTH reactivity but no donor-reactive DTH reactivity, whereas patient no. 2 demonstrates both EBV- and donor-reactive DTH reactivity. Patient no. 1 is designated as DTH negative to donor antigens and patient no. 2 is DTH positive to donor antigens.

Download figure to PowerPoint

What is the incidence of donor-reactive DTH reactivity in transplant patients?

The trans-vivo DTH assay was used to test PBMC from 180 renal transplant patients for this study. As shown in Figure 2, we observed that 52% (93/180) of the patients were DTH positive for donor alloantigens, while 48% (87/180) were DTH negative. If these results are representative of all kidney and simultaneous kidney and pancreas recipients in our program, then about half of our patients develop T cells that are sensitized for DTH toward their donor antigens.

image

Figure 2. Relationship between the presence of donor-reactive DTH reactivity (DTH) and detectable circulating alloantibodies (Ab). Shown are the percentages of patients who are DTH–/Ab–, DTH–/Ab+, DTH+/Ab–, and DTH+/Ab+. The incidence of detectable alloantibodies is similar in patients who were DTH positive and DTH negative to donor antigens.

Download figure to PowerPoint

Does donor-reactive DTH reactivity correlate with the presence of detectable alloantibodies?

Currently, alloantibodies provide the only generally accepted index of donor-reactive allosensitization in transplant patients. To evaluate the relationship between detectable alloantibodies and donor-reactive DTH responses, sera from the 180 patients were tested for the presence of alloantibodies by the FlowBead PRA method. This assay measures the reactivity of serum IgG for microbeads that have been coated with various allelic forms of HLA class I or II molecules. As shown in Figure 2, we observed that of the 87 patients who were negative for donor-reactive DTH, 68 were also negative for alloantibodies, while 19 were positive for alloantibodies. Of the 93 who were positive for donor-reactive DTH, only 17 were positive for alloantibodies, while 76 were negative for alloantibodies. Thus donor-reactive DTH reactivity occurred much more frequently than did the development of detectable alloantibodies (52% vs. 20%). Therefore, a reliance on alloantibody detection as an index of allosensitization may seriously underestimate the incidence of allosensitization in transplant recipients. Further, the presence of alloantibodies did not correlate with donor-reactive DTH reactivity. Indeed, patients with detectable alloantibodies had roughly the same incidence of donor-reactive DTH reactivity as patients without detectable alloantibodies (47% and 53%, respectively). Apparently, at a single point in time circulating alloantibodies and donor-reactive DTH reactivity are unrelated indices of patient allosensitization.

Does donor-reactive DTH reactivity correlate with degree of HLA mismatch?

We reasoned that the likelihood of a transplant recipient developing donor-reactive DTH reactivity would probably be directly proportional to the degree of mismatch between donor and recipient HLA antigens. To test this hypothesis, the mean number of HLA mismatches between donor and recipient was determined for the DTH positive and negative recipients (Table 2). There was no statistical difference between the two groups of recipients. To analyze this further, patients were grouped according to the number of HLA mismatches to their donor organ(s). The incidence of a DTH positive result was compared between groups (0–6 antigen mismatched donor/recipient pairs) (Figure 3). In general, there was no relationship between the degree of HLA mismatch between donor and recipient and the percentage of patients who demonstrated donor-reactive DTH reactivity in the trans-vivo DTH assay.

Table 2. : DTH result vs. patient demographics, immunosuppressive management, and post-transplant timing of DTH testing
 DTH negative (n = 87)DTH positive (n = 93)UnivariMultivarl
  • a

    AA – African-American;

  • b

    PRA – panel reactive antibodies;

  • c

    LRD – living donor kidney recipient, CAD – cadaveric donor kidney recipient, SPK – simultaneous kidney-pancreas recipient;

  • d

    HLA – human leukocyte antigen;

  • e

    IL2-Ri – interleukin 2 receptor inhibitor, OKT3 – Muromonab-CD3, ATGAM – antithymocyte globulin;

  • f

    CMP – cyclosporine, mycophenolate mofetil, prednisone; CAP – cyclosporine, azathioprine, prednisone; CP – cyclosporine, prednisone

  • g

    CSA – cyclosporine;

  • h

    post-tx – post-transplantation;

  • i

    Signif – significance.

Age (years)41.6 ± 13.142.7 ± 12.6nsns
Gender (m/f)54/3357/36nsns
Race (AA)a (n)13% (11)15% (14)nsns
Donor organ(s) LRD/CAD/SPKc15/52/2010/58/25nsns
Last PRA > 10% (n/total tested)10% (8/81) 8% (7/91)nsns
High PRAb > 10% (n/total tested)24% (19/81)30% (27/91))nsns
HLAd mismatch (mean ± sd) 4.3 ± 1.9 4.5 ± 1.5nsns
Induction (IL2-Ri/OKT3/ATGAM/none)e32/21/24/1038/27/17/11nsns
Maintenance (CMP/CAP/CP)f56/25/677/13/30.020.02
Serum CSAg level (6 month) 232 ± 162 253 ± 120nsns
Serum CSA level (12 month) 199 ± 85 216 ± 69nsns
Time post-txh of DTH testing (years) 4.6 ± 4.3 3.4 ± 3.60.04ns
image

Figure 3. Comparison of the percentage of patients with detectable donor-reactive DTH reactivity (Y axis) vs. the number of HLA mismatches between recipient and donor (X axis). The number of patients within each category is included for each bar graph. Donor and recipient HLA A, B, and DR were available for analysis for 173 of the 180 donor/recipient pairs in this study.

Download figure to PowerPoint

Does donor-reactive DTH reactivity correlate with other clinical variables?

We next compared the demographics between graft recipients who were DTH positive and DTH negative and failed to find any significant differences in any of the parameters evaluated (Table 2). These parameters included recipient demographics, the type of organ(s) received and humoral evidence of pretransplant allosensitization (pretransplant PRA). To evaluate the possibility that post-transplant immunosuppressive management influenced recipient donor-reactive DTH reactivity, the agents used for induction treatment and maintenance immunosuppression were determined for each patient. Additionally, cyclosporine exposure was also estimated, based on trough whole-blood cyclosporine levels obtained at 6 months and 1 year post-transplant. These data, along with the time of DTH testing post-transplant, were compared between patients who were DTH positive and those who were DTH negative. The agent used for induction therapy at the time of engraftment and the mean serum trough levels at 6 and 12 months after transplantation did not significantly differ between the two groups (Table 2). However, the DTH positive patients were more recently transplanted than their DTH negative counterparts. Thus, their time from transplantation to DTH testing was significantly shorter (p = 0.04), and significantly more of these patients received mycophenolate mofetil rather than azathioprine as part of their maintenance immunosuppressive regimen (p = 0.02). In a multivariate model that included mycophenolate mofetil use and the time interval between transplantation and DTH testing, only mycophenolate use was a significant risk factor for a positive DTH result. This suggests that mycophenolate administration, rather than the time of post-transplant DTH testing, is the variable which correlates with a positive DTH result. Serial post-transplant analysis of patients over time is under way to more accurately determine the influence of time post-transplant on the results of donor-reactive DTH analysis. Results of these studies will form the basis of a future manuscript.

Does donor-reactive DTH reactivity correlate with a worse clinical outcome?

We compared the post-transplant clinical outcome of the DTH positive patients to that of their DTH negative counterparts. The 180 patients were grouped by DTH result and various clinical parameters were then compared (Table 3). We observed that both groups had a similar incidence of previous acute rejection, a similar incidence of chronic rejection, a similar period of time to their first acute rejection episode and similar mean serum creatinine level at 6 months, 1 year, 2 years and 3 years post transplant. Also, the incidence of graft loss was similar in both groups. Thus, the expression of donor-reactive DTH did not appear to correlate with any of the measured clinical outcome parameters.

Table 3. : DTH result vs. clinical outcome for 180 kidney (K) and simultaneous kidney-pancreas (SPK) recipients
 Neg DTH (n = 87)Pos DTH (n = 93)Significance
  • a

    AR = acute rejection.

  • b

    CR = chronic rejection.

  • c

    yrs = years.

Incidence of previous ARa
 0 68% (59/87) 77% (72/93) 
 1 25% (22/87) 14% (13/93)p = 0.16
 > 1  7% (6/87)  9% (8/93) 
Incidence of CRb  6% (5)7.5% (7)ns
Incidence of biopsy-proven CR4.6% (4)5.4% (5)ns
Time to 1st AR episode (days)  97 ± 185 140 ± 226ns
Time from last AR episode (years)c4.41 ± 4.393.10 ± 3.58p = 0.03
Serum creatinine
 6 months (n = 175) 1.7 ± 0.4 1.6 ± 0.5ns
 1 years (n = 160) 1.7 ± 0.5 1.7 ± 0.8ns
 2 years (n = 113) 1.9 ± 1.1 1.6 ± 0.6ns
 3 years (n = 87) 1.8 ± 0.7 1.6 ± 0.5ns
Kidney loss (n)  4% (4) 6% (5)ns
Pancreas loss (n)  2% (2) 4% (3)ns
Death (n)  1% (1) 1% (1)ns

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

In these studies, we used the trans-vivo DTH assay to provide a direct index of T-cell sensitization to donor alloantigens in transplant patients. A critical feature of the DTH assay is that it can selectively measure recall responses, i.e. T-cell responses to antigens previously encountered by the immune system. Most assays of T-cell alloreactivity, including the mixed lymphocyte response (MLR) cannot discriminate between alloreactivity mediated by naïve T cells and alloreactivity mediated by memory T cells. Thus, they cannot separate allosensitization, the previous response to alloantigens, from alloreactivity, the potential for an alloimmune response. To our knowledge, the only other analytic system previously used clinically to discriminate between T-cell alloreactivity and T-cell allosensitization is the 24-h cytokine ELISPOT assay (6). As far as we are aware, that assay has not been used to evaluate T-cell allosensitization in a large population of transplant recipients. This DTH-based analysis of T-cell allosensitization is well suited for studies designed to monitor the changing disposition of allograft recipients' T cells toward graft alloantigens during the post-transplant period.

In the trans-vivo DTH assay, detection of allosensitized patients is relatively straightforward, while detection of nonsensitized patients is not. Negative DTH responses may be due to lack of allosensitization, but could also be due to the lack of functioning T cells, or residual effects of in vivo administered immunosuppression. To address this, the patients were concurrently tested for DTH reactivity to a positive control antigen to which most humans in the region are sensitized, such as EBV, CMV or tetanus toxoid. Most of the patients in this study were tested for reactivity to EBV. If patients' PBMC respond to the control antigen, but not to donor alloantigens, they are considered immunologically responsive, but nonsensitized to donor alloantigens. If the patients fail to respond to either control antigen, the test is considered to be invalid. This is uncommon (0.4% of all tests performed to date, data not shown), suggesting that pharmacological immunosuppression and/or limited immune capacity rarely cause invalid results in this analytic system.

When we tested 180 transplant patients using the trans-vivo analysis for donor-reactive DTH reactivity, we were surprised to find a relatively high incidence of donor reactivity (Figure 1). We had expected a somewhat lower proportion, perhaps similar to the proportion of patients with a history of acute rejection, which was 27% for this patient population. Apparently, the adoption of a pro-inflammatory T-cell disposition toward donor alloantigens is relatively common among our transplant recipients. In the DTH positive patients, it would appear that pharmacological immunosuppression is effective at blocking T-cell dependent acute allograft rejection, but not the development of pro-inflammatory T-cell allosensitization. In turn, this would suggest that immunosuppressive agents thought to block acute rejection by blocking T-cell activation may actually block the consequences of T-cell sensitization, rather than the sensitization event itself. If true, a reduction or complete withdrawal of immunosuppression in DTH positive patients may carry a higher risk of a subsequent acute rejection episode compared to DTH negative patients.

We note that despite this high incidence of DTH reactivity, we may still be underestimating the true incidence of T-cell allosensitization in these patients. By design, patient PBMC were challenged for alloreactive DTH reactivity with a subcellular sonicate of donor leukocytes. Presumably, this assay stimulates T-cell responders primarily via the indirect pathway of allorecognition. However, it remains possible that the cell-free membrane fragments in the sonicated alloantigen preparation donate intact MHC molecules for display by responder antigen presenting cells (APCs), and that a small amount of direct pathway T-cell stimulation may occur in this assay. We cannot test for this possibility because the subcutaneous injection of mice with mixed allogeneic leukocytes always results in a strong DTH-like swelling response (unpublished observation). Further, we presume that both allogeneic MHC class I and class II molecules can prime human T cells for DTH responses, but we do not know how well MHC class I or II molecules are represented in the donor leukocyte sonicate. Relatively poor representation of MHC class I or II molecules during DTH challenge may again lead to an underestimate of T-cell allosensitization in these transplant patients. Studies are currently underway to address this issue.

It is interesting to note that the incidence of a DTH positive result did not increase with an increasing degree of HLA mismatch in these study patients. In fact, the incidence for patients with no mismatches to their donor for HLA class I, HLA class II, or both class I and class II antigens was not statistically significantly different than for their mismatched counterparts (Figure 3). There are several possible explanations for this finding. It may be that donor-reactive DTH reactivity is elicited in our non-mismatched recipients by subtle allelic mismatches at the nucleotide level that were undetected using the traditional tissue typing techniques. Strong mixed lymphocyte responses are known to routinely occur under such circumstances (7). Our recent adoption of high-resolution DNA-based HLA identification will address this issue. Alternatively, minor histocompatibility antigens could drive the T-cell responses in the trans-vivo DTH assay. This has been previously described in a murine corneal allograft model (8), but to our knowledge has not been reported in human transplant recipients. Finally, a positive DTH result could be due to allostimulation of T cells previously primed by environmental antigen(s) that are cross-reactive to donor alloantigens. T cells primed by exposure to environmental antigens (such as viral infections) have been shown to cross-react with alloantigens (9) as well as autoantigens [reviewed in (10)]. In this regard, it becomes important to determine if the donor-reactive DTH responses were acquired post-transplant.

Since donor-reactive T cells are required to generate donor-reactive IgG, the incidence of detectable donor-reactive IgG is generally considered to reflect the incidence of T-cell allosensitization. We tested this relationship in this study (Figure 2). It must be noted that for this study alloreactive IgG was detected by FlowBead PRA analysis. This approach has two potential problems. It may detect alloantibodies that are not donor-reactive, and it may miss donor-reactive alloantibodies. However, it was necessary to use the FlowBead PRA method due to limited donor alloantigen availability. We are developing a donor-specific ELISA method of antibody detection and have compared the ELISA results with the FlowBead PRA results. In general, we found an 85% correlation; that is, most bead-reactive antibody is donor-reactive (manuscript in preparation). Until donor antigens are more generally available, the FlowBead PRA method of alloantibody detection will have to suffice. It is interesting to note that some patients had detectable circulating IgG alloantibodies but lacked evidence of donor-reactive DTH reactivity. Presumably B-cell antidonor IgG production requires interaction with donor-sensitized T cells. However, antibody production in the absence of peripheral DTH antigen reactivity has been observed under certain conditions following an intravenous antigen challenge (11,12). Perhaps similar mechanisms are operative in our patient population.

We observed that the incidence of donor-reactive DTH reactivity (52%) was much higher than the incidence of alloantibody production (20%). This suggests that DTH reactivity and IgG production do not correlate well in this patient population. Indeed, alloreactive IgG production occurred with similar frequency in DTH positive and DTH negative patients (Figure 2). Reciprocally, DTH reactivity occurred as frequently in patients who made alloantibodies as in those who did not. By these criteria, it appears that detectable DTH reactivity and circulating alloantibodies behave as independent allosensitization events, and therefore alloantibody production is a poor index of the true incidence of patient allosensitization. This observation is at odds with the relationship that has been observed in animal models, where antidonor alloantibodies are usually detectable when solid-organ transplant recipients demonstrate donor-reactive DTH reactivity (3,13). If production of alloreactive IgG and the expression of donor-reactive DTH reactivity are both evidence of previous donor sensitization, then their additive incidence of 62% represents the true incidence of allosensitization (either humoral or cellular) in our transplant patients. Thus, the incidence of allosensitization is quite high, indicating that the majority of transplant recipients develop allosensitization despite clinically efficacious immunosuppression.

This leads to questions regarding the clinical importance of T-cell allosensitization. Recent studies have clearly shown an adverse effect of post-transplant humoral allosensitization on allograft survival (14–27). We presumed that T-cell allosensitization would similarly correlate with an adverse clinical outcome. Thus, we were surprised to find that donor-reactive DTH reactivity did not correlate with any of the clinical outcomes examined (Table 3). The lack of correlation between T-cell donor sensitization and clinical outcome has led us to hypothesize that ongoing immunosuppression is relatively ineffective at blocking T-cell allosensitization, but quite effective at blocking the biologic result of that allosensitization. Since complete withdrawal of immunosuppression is not clinically practical at this time, we cannot directly test this hypothesis. Perhaps this method for testing recipient T-cell donor sensitization post transplantation will be more useful and informative in the future for designing tolerogenic immunosuppressive strategies or strategies designed to avoid over-immunosuppression.

Finally, it will be interesting to observe in our program whether the incidence of pro-inflammatory T-cell allosensitization changes over the next few years, due to the relatively recent clinical use of newer, more effective immunosuppressive drugs that have significantly lowered the incidence of acute rejection. Also, due to wide variations in clinical practices among the different transplant centers, we would predict that varying patterns of donor sensitization would be observed by different centers. The trans-vivo DTH assay should be useful in expanding our appreciation of post-transplant allosensitization in transplant recipients.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This study was supported by National Institutes of Health grants PO1-AI/HL40150, RO1-HL61966, PO1-HL70294, a Medical Research Development Fund (MRDF) grant, and in part by grant P30-CA16058, National Cancer Institute, Bethesda MD (cgo). We are grateful to Marsha Stalker for preparation of the manuscript and to Irene DeAndero RN, BSN, CCTC, Melissa Knox LPN, CCTC, Becky Miller LPN, CCTC, Mary Ann Pettit RN, BSN, CCTC, M.J. Sprague RNC, CCTC, and Carol Wheeler RN, BSN for their help in sample acquisition. We would also like to thank Maria Belizzi, Alice Bickerstaff MS, and Jake Jansen BS for their technical assistance.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Rush DN, Karpinski ME, Nickerson P, Dancea S, Birk P, Jeffery JR. Does subclinical rejection contribute to chronic rejection in renal transplant patients? Clin Transplant 1999; 13: 441446.
  • 2
    Purkerson J, Isakson P. A two-signal model for regulation of immunoglobulin isotype switching. FASEB J 1992; 6: 32453252.
  • 3
    Sirak J, Orosz C, Wakely E, VanBuskirk A. Alloreactive delayed-type hypersensitivity in graft recipients. Transplantation 1997; 63: 13001307.
  • 4
    VanBuskirk A, Wakely E, Sirak J, Orosz C. Patterns of allosensitization in allograft recipients: long term cardiac allograft acceptance is associated with active alloantibody production in conjunction with active inhibition of alloreactive DTH. Transplantation 1998; 65: 11151123.
  • 5
    Carrodeguas L, Orosz CG, Waldman WJ, Sedmak DD, Adams PW, VanBuskirk AM. Transvivo analysis of human-delayed type hypersensitivity reactivity. Hum Immunol 1999; 60: 640651.
  • 6
    Heeger PS, Greenspan NS, Kuhlenschmidt S et al. Pretransplant frequency of donor-specific, IFN-g-producing lymphocytes is a manifestation of immunologic memory and correlates with the risk of posttransplant rejection episodes. J Immunol 1999; 163: 22672275.
  • 7
    Baxter-Lowe LA, Eckels DD, Ash R, Casper J, Hunter JB, Gorski J. The predictive value of HLA-DR oligotyping for MLC responses. Transplantation 1992; 53: 13521357.
  • 8
    Sonoda Y, Sano Y, Ksander B, Streilein J. Characterization of cell-mediated immune responses elicited by orthotopic corneal allografts in mice. Invest Ophthalmol Visual Sci 1995; 36: 427434.
  • 9
    Lechler R, Heaton T, Barber L, Bal V, Batchelor J, Lombardi G. Molecular mimicry by major histocompatibility complex molecules and peptides accounts for some alloresponses. Immunol Lett 1992; 34: 6369.
  • 10
    Benoist C, Mathis D. Autoimmunity provoked by infection: how good is the case for T cell epitope mimicry? Nature Immunol 2001; 2: 797801.
  • 11
    Hahn H, Kaufmann SH, Falkenberg F, Chahinin M, Horn W. Peritoneal exudate T lymphocytes with specificity to shepp red blood cells. II. Inflammatory helper T cells and effector T cells in mice with delayed-type hypersensitivity and in suppressed mice. Immunology 1979; 38: 5155.
  • 12
    Milon G, Marchal G, Seman M, Truffa-Bachi P, Zilberfarb V. Is the delayed-type hypersensitivity observed after a low dose of antigen mediated by helper T cells? J Immunol 1983; 130: 11031107.
  • 13
    Vella JP, Vos L, Carpenter CB, Sayegh MH. Role of indirect allorecognition in experimental late acute rejection. Transplantation 1997; 64: 18231828.
  • 14
    Monteiro F, Buelow R, Mineiro C, Rodrigues H, Kalil J. Identification of patients at high risk of graft loss by pre-and posttransplant monitoring of anti-HLA class I IgG antibodies by enzyme-linked immunosorbent assay. Transplantation 1997; 63: 542546.
  • 15
    Christiaans MH, Nieman F, Van Hooff JP, Van Den Berg-Loonen EM. Detection of HLA class I and II antibodies by ELISA and complement-dependent cytotoxicity before and after transplantation. Transplantation 2000; 69: 917927.
  • 16
    McKenna RM, Takemoto SK, Terasak PI. Anti HLA antibodies after solid organ transplantation. Transplantation 2000; 69: 319326.
  • 17
    Martin S, Mallick N, Gokal R, Harris R, Johnson R. Posttransplant antidonor lymphocytotoxic antibody production in relation to graft outcome. Transplantation 1987; 44: 5053.
  • 18
    Scornik JC, Lim P, Howard R, Pfaff W. Posttransplant antidonor antibodies and graft rejection. Transplantation 1989; 47: 287290.
  • 19
    Suciu-Foca N, Reed E, D'Agati VD et al. Soluble HLA antigens, anti-HLA antibodies, and antiidiotypic antibodies in the circulation of renal transplant recipients. Transplantation 1991; 51: 593601.
  • 20
    Halloran PF, Schlaut J, Solez K, Srinivasa NS. The significance of the anti-class I response. II. Clinical and pathologic features of renal transplants with anti-class I-like antibody. Transplantation 1992; 53: 550555.
  • 21
    Al-Hussein KA, Shenton BK, Bell A et al. Characterization of donor-directed antibody class in the post-transplant period using flow cytometry in renal transplantation. Transplant Int 1994; 7: 182189.
  • 22
    Lobo PI, Spencer CE, Stevenson WC, Pruett TL. Evidence demonstrating poor kidney graft survival when acute rejections are associated with IgG donor-specific lymphocytotoxin. Transplantation 1995; 59: 357360.
  • 23
    Trpkov K, Campbell P, Pazderka F, Cockfield S, Solez K, Halloran P. Pathologic features of acute renal allograft rejection associated with donor-specific antibody: analysis using the Banff grading schema. Transplantation 1996; 61: 15861592.
  • 24
    Abe M, Kawai T, Futatsuyama K et al. Postoperative production of anti-donor antibody and chronic rejection in renal transplantation. Transplantation 1997; 63: 16161619.
  • 25
    Schonemann CGJ, Leverenz S, May G. HLA class I and class II antibodies. Transplantation 1998; 65: 15191523.
  • 26
    Piazza A, Adorno D, Poggi E et al. Flow cytometry crossmatch. A sensitive technique for assessment of acute rejection in renal transplantation. Transplant Proc 1998; 30: 17691771.
  • 27
    Christiaans MHL, Overhof-De Roos R, Nieman F, Van Hooff JP, Van Den Berg-Loonen E. Donor-specific antibodies after transplantation by flow cytometry: relative change in fluorescence ratio most sensitive risk factor for graft survival. Transplantation 1998; 65: 427433.