| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Surgery and
Pathology, Washington University School of Medicine, St. Louis, MO 63110
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the present study, we characterize human CD4+ T cells generated against porcine islets both by in vitro stimulation with porcine islets as well as by the use of SCID mice reconstituted with human PBL (HuPBL-SCID)3 and subsequently transplanted with porcine islets.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CB/17 SCID mice were obtained from The Jackson Laboratory (Bar Harbor, ME) and bred at Washington University School of Medicine in a pathogen-free environment. All animals were 8–12 wk old and had total mouse Ig levels below 30 µg/ml, as measured by ELISA.
Human PBL reconstitution of SCID mice was accomplished as previously described (8). Briefly, human PBL from healthy donors were isolated by density gradient centrifugation on Ficoll-Paque (Pharmacia, Alameda, CA) from heparinized blood of healthy donors. Human PBL (30 x 106) were resuspended in a volume of 0.5 ml of PBS (Sigma, St. Louis, MO) and injected i.p. into SCID mice. An additional 15 x 106 PBL were cultured in the presence of 100 ng/ml OKT3 (Ortho, Raritan, NJ) in RPMI 1640 medium (Life Technologies, Grand Island, NY), supplemented with 25 mM HEPES buffer, 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, and 15% heat-inactivated FBS for 2 days at 37°C. OKT3-activated PBL were washed and injected 2 days after the initial injection. Serum was monitored weekly in these mice by ELISA for human IgG. Reconstitution was considered successful when human IgG levels reached >500 µg/ml within 3–5 wk.
Transplantation of porcine islets and isolation of graft-infiltrating lymphocytes (GIL)
Porcine islets were isolated from outbred Yorkshire sows by the Islet Isolation Core Facility of the Diabetes Research and Training Center of Washington University. Approximately 3000 islet equivalents were transplanted under the kidney capsule of the reconstituted mice as previously described (9). Two to three weeks after transplantation, the mice were sacrificed and the GIL were recovered by excising the site of the transplant from the kidney. The parenchyma was then minced into small pieces and incubated in media containing 20 U human rIL-2/ml and 5% T-Stim (Collaborative Biomedical Products, Bedford, MA). Cultures were maintained weekly by the addition of irradiated (3000 rads) porcine islets or splenocytes as an Ag source and autologous irradiated PBL as APC. After 6 wk of culture, the antiporcine T cell line was cloned by limiting dilution at 1 cell/well with the autologous, irradiated, EBV-transformed lymphoblastoid cell line (1 x 105) as feeder cells and relevant porcine splenocytes (1 x 104) as Ag in RPMI 1640 medium supplemented with 20 U/ml rIL-2.
In vitro generation of human T cells and clones against porcine islets
Four separate T cell lines were generated using 10 x 106 human PBL from healthy donors cultured with 100 islet-equivalent porcine islets in RPMI 1640 medium supplemented with 20 U/ml human rIL-2 and 5% T-Stim. Cultures were maintained by weekly addition of relevant irradiated porcine islets as Ag and autologous irradiated PBL as APC. After 3 wk of culture, the cell lines were cloned by limiting dilution at 1 cell/well with irradiated autologous PBL (1 x 104) as feeder cells (10) in the presence of 0.025% PHA (Difco Laboratories, Detroit, MI) in RPMI 1640 supplemented with 50 U/ml rIL-2.
Phenotype identification
The phenotype of cell populations in both the in vivo and in vitro studies was identified by flow cytometric analysis using anti-human CD3, CD4, and CD8 mAbs conjugated with peridinin chlorophyl protein, PE, and FITC, respectively (Becton Dickinson, San Jose, CA) (10). Briefly, 5 x 105 cells were incubated for 30 min at 4°C in a 1:50 dilution of Ab in PBS. The cells were then washed (three times) and analyzed using a FACscan with Lysis II software (Becton Dickinson).
Swine Leukocyte Ag (SLA) class I isolation and purification
MHC class I was isolated and purified following the procedure of Naziruddin et al. (11). Briefly, MHC class I molecules were affinity purified from detergent extracts of 10 g of spleen harvested from z/z haplotype of miniature swine, using two mAbs against porcine SLA class I framework determinants, 7.34.1 and 74.11.10 (American Type Culture Collection, Manassas, VA). Concentrations of 5 µg/ml and 10 µg/ml were used as an Ag source for proliferation assays.
Mixed lymphocyte islet reaction
Proliferation of T cells was done by mixed lymphocyte islet reaction using 4 x 104 responder cells (GILs or PBL) and stimulator cells (1–2 porcine islets/well or 1 x 104 porcine splenocytes) in presence of irradiated autologous APC (2 x 105 cells/well). Cells were incubated at 37°C for 4 days. During the last 18 h of culture, each well was pulsed with 1 µCi [3H]thymidine (1 µCi/well). Cells were then harvested onto glass fiber filters, and radioactivity was assessed by ß-scintillation counting. Results of quadruplicate cultures are reported as a stimulation index (SI), which was calculated using the following formula: SI = (mean cpm of experimental cultures)/(mean cpm of control cultures). Ab blocking experiments were done using W6/32 anti-HLA class I and KuIa2 anti-HLA class II mAbs at 5 µg/ml added at the initiation of culture.
Cell-mediated lympholysis (CML)
CML activity was measured using a standard 51Cr release assay with an E:T ratio of 25:1 for T cell bulk lines or 5:1 for T cell clones. Trypsin-dispersed relevant porcine islets and porcine splenocytes were incubated with 250 µCi of 51Cr for 120 and 60 min, respectively, at 37°C. T cells were incubated with labeled targets for 4 h in 96-well round-bottom plates. The supernatants were then harvested and assessed for 51Cr concentration. Results are expressed as a percent specific lysis, which was calculated using the following formula: % specific lysis = [(experimental lysis) - (spontaneous release)]/[(maximum lysis) - (spontaneous release)]. Spontaneous release was determined by incubation of target cells in medium alone, and the maximum lysis was determined by incubation of target cells in 1% Triton X-100.
Assay for cytokine-mediated cytotoxicity
Cytokine-mediated cytotoxicity was measured by culturing 2 x 105 T cell clones (effectors) plus 10 porcine islets (Ag) in the top section of a transwell culture system (Costar, Cambridge, MA). The bottom section contained a designated target cell: 15 porcine or human hand-picked islets, 1 x 105 lymphoblastoid cell line or kidney cell line, 1 x 105 PHA-blasted porcine lymphoid cells, 1 x 104 porcine endothelial cell line, or 1 x 104 human breast cancer cell line (HL-100) or lung cancer cell line (HTB-57). The cells were cultured at 37°C in RPMI 1640 medium supplemented with 20 U/ml human rIL-2. After 48 h, the top section of the transwell was removed, and supernatant was harvested. Targets were stained with fluorescein diacetate and propidium iodide, and percent viability was determined under a fluorescent microscope.
Cytokine measurement
Cytokine concentration in the supernatant was measured by ELISA.
TNF-
and IL-4 were determined using the Quantigen (PharMingen, San
Diego, CA) kit, and IFN-
was measured by a matched Ab pair
Sample-Pak (Endogen, Woburn, MA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Five GIL T cell lines were successfully developed from 10 mice that were reconstituted and transplanted. Haplotypes of human donors were as follows: (A30, 3 B60, 65 Cw3, 8 DR0102,0104), (A2, 29 B62,- Cw3,- DR0101,0404)x2, (A2,- B37, 44 Cw-, DR4, 10)x2. Phenotypic evaluation showed greater than 97% CD4+ T cells with less than 2% CD8+ T cells. CML assays performed on GIL T cell lines showed no specific lysis of relevant or third-party targets. Clones obtained from the T cell lines after limiting dilution were all CD4+.
GIL T cells showed significant (SI > 80) proliferation to both
relevant (autologous to transplanted islets) and third-party (other
outbred) porcine Ag only in the presence of autologous APC (Fig. 1
). This proliferation was blocked by the
addition of the KuIa2 mAb, indicating Ags expressed on
porcine-stimulating cells were processed and presented by self-HLA
class II molecules. Addition of control mouse IgG showed no blocking of
the proliferation.
|
shows that the GIL responded to both
porcine islets as well as purified SLAz/z class I
molecules isolated from porcine splenocytes only in the presence of
self-APC, and this proliferation was significantly blocked by the
addition of the KuIa2 mAb.
|
).
|
Four different human antiporcine islet T cell lines were generated
against different outbred pig islets. Haplotypes of donors were as
follows: (A2,- B37, 44 Cw-, DR7, 10)x2, (A2, 3 B35, 44 Cw4, 5 DR4,
9)x2. Two cultures contained only CD4+ T cells,
whereas the remaining two showed 12% and 15%
CD8+ T cells. Fig. 4 shows the proliferative response of a
representative T cell line in response to porcine Ag. This response was
also dependent on the presence of autologous APC and was blocked by the
KuIa2, the anti-class II mAb. This result, along with the data
obtained from GILs generated from HuPBL-SCID mice transplanted with
porcine islets, indicates a predominant role for indirect Ag
presentation of xenoantigens for the generation of human antiporcine
immune response.
|
). Clone no. 33 substantially
proliferated to relevant porcine islets with a SI of 40 times that of
the other clones. These results again indicate that porcine Ags are
presented by human APC to the human T cells. In general, human islets
failed to stimulate three of the four clones. Proliferation was seen in
clone no. 18 with and without APC when human islets were used as Ag
(Fig. 5). It is likely that some of the human and porcine Ags may
cross-react because clone no. 18 demonstrated some proliferation with
human islets as the Ag source.
|
Human T cell clones were tested for their ability to lyse porcine
islets by standard CML assay. All of the clones were
CD4+ and were tested at an E:T ratio of 5:1.
Seventeen of the 24 clones demonstrated >8% lysis of porcine islet
target with six of the clones showing lysis of porcine islet targets
>10% (Fig. 6). Initial testing of other
porcine and human cell types for lysis by the same clones demonstrated
no lysis (data not presented). This suggests that cytotoxicity of the
islets may be mediated by cytokine-induced lysis, considering the high
susceptibility of pancreatic islets to cytokine-mediated damage
(12, 13).
|
To understand the mechanism of killing of porcine islets by
CD4+ T cells clones, experiments were designed
utilizing a transwell culture system in which effectors and targets are
separated by a filter barrier and therefore are not in direct contact.
The T cell clones were cultured with porcine islets in the top section
of the transwell culture system with the bottom section containing the
different targets. After 48 h, the target cells were analyzed for
viability. Fig. 7A shows that
clones #20 and #18 significantly lysed the porcine as well as human
islets. In addition, there was no cytotoxicity for the other human or
porcine targets previously described by any of the three clones (no
lysis was observed by clone #33). It is of interest that in some cases,
even if APCs were not added to the top of the transwell at the time of
performing the assay, these clones produced cytokines that lysed the
islets on the bottom portion of the transwell. We feel that these
clones, which were stimulated with porcine islets in the presence of
APC on a weekly basis, remain in an activated state, resulting in
continued production of cytokines. Analysis of the culture media during
the cytotoxicity studies revealed the cytokine profile of each of the
three clones investigated. Fig. 7B shows that clone #18
produced high concentrations of NF- (150 pg/ml) and IFN- (110.9
pg/ml) and negligible IL-4 (6.6 pg/ml). Clone #20 produced moderate
amounts of IL-4 (85.5 pg/ml) and IFN- (53 pg/ml) and low levels of
TNF- (14.2 pg/ml), and clone #33 produced extremely high amounts of
IL-4 (532 pg/ml), moderate levels of TNF- (79.4 pg/ml), and very low
quantities of IFN- (12.4 pg/ml). Preliminary results of TNF-
blocking show no killing of islets in the presence of anti-TNF-
(data not shown). These data are consistent with clone #18 being of Th1
phenotype, clone #20 being of Th0 phenotype, and clone #33 being of Th2
phenotype. These results also indicate the high susceptibility of both
human and porcine islets for cytokine-mediated damage by IFN-.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Knowledge of porcine islet rejection mechanisms mediated by human T
lymphocytes is necessary for successful long-term xenograft survival.
During graft rejection, mononuclear cells infiltrate the graft, and an
array of cytokines is produced that is known to have deleterious
effects on the transplanted organ. This process is likely to be more
significant in islet allograft rejection because it is known that
islets are highly susceptible to cytokine-mediated damage
(13). Cytokines such as TNF-, IFN-, and IL-4 have
been shown to regulate the expression of various cell-surface molecules
involved in immune responses, such as adhesion molecules and MHC class
I and II Ags of the transplanted organs (15).
Studies have shown that human T cell response against porcine Ags is at least as strong as the alloresponse and that both CD4+ and CD8+ T cells play a role in this process (16, 17). It has also been proposed that the xenoresponse is primarily a CD4+ T cell-dependent process and that indirect Ag presentation of xenoantigens plays an important part in this response (18). However, the mechanism by which CD4+ T cells destroys the xenograft is poorly understood (19). Although it is generally assumed that cytokines play a major role in this process, the results reported here indicate that cell-mediated cytotoxicity of porcine islets by CD4+ T cells is primarily due to Th1-type cytokines.
Graft-infiltrating CD4+ T cells derived from the
porcine islet-transplanted HuPBL-SCID mice were primarily
CD4+ within 3 wk of culture. Culture conditions
were general and did not promote one phenotype over another. GIL
CD4+ T cells proliferated when porcine Ag was
provided as a stimulus in conjunction with autologous APC (Fig. 1
). It
is not surprising that all of the CD4 T cell clones tested needed
self-APC because islets normally do not express MHC class II Ags. Work
by Bretzel et al. (20) states that MHC class II is
expressed only on vascular endothelium, macrophages, and monocytes and
not on porcine islet cells. In addition, Pavlovic et al.
(21) and Swift et al. (22) indicate the
absence or lack of impact of MHC class II expression on islets with
regard to T cell stimulation, even with the addition of high-dose
IFN- and/or TNF-. However, it was surprising that the major
porcine Ag recognized by these GILs is MHC molecule, because the human
CD4+ T cells not only proliferated when porcine
islets were given as Ag but also when purified SLA class I molecules
were given as Ag (Fig. 2). This strongly indicates that SLA molecules
present on porcine xenografts are the major Ags that are recognized by
human T cells and that indirect presentation of these molecules
constitutes a major pathway of xenorecognition. Recent studies both in
animal models and in human solid organ transplantation have shown that
indirect Ag presentation of peptides derived from mismatched MHC class
I and II molecules plays an important role in chronic allograft
rejection (23, 24, 25). Results presented in this study
indicate that a similar mechanism may also play a role in xenograft
rejection. Human CD4+ T cell clones generated in
vitro against porcine islets demonstrated specificity identical with
those obtained by GILs. APC were needed for the proliferation, and the
MHC class II restriction of these T cell clones were further documented
in this study by the inhibition of proliferation by the KuIa2
mAb.
Based on their cytokine production, CD4+ T cells
have been grouped into three kinds. Th1 cells primarily produce IL-2,
IFN-, and TNF- and regulate cellular immune response to viral and
intracellular pathogen infections. Th2 cells primarily secrete IL-4,
IL-5, IL-10, and IL-13 and regulate humoral immune responses. Th0
clones are defined as a subset that produce both Th1 and Th2 cytokines
and that are often considered as a precursor cell. Immune responses to
antigenic stimuli are polarized with respect to the types of T cells
that mediate these responses. For example, type 1 diabetes mellitus is
thought to be mediated by the Th1 subset, whereas the Th2 cells may
play a protective role (26). Allograft rejection is often
associated with the Th1 response, whereas the long-term graft survival
has been shown to be associated with a Th2 response (27, 28). In the present study, CD4+ T cell
clones derived after coculturing normal human PBL with porcine islets
exhibit three cytokine profiles consistent with Th1, Th0, and Th2
phenotypes. Clone #18 was strongly cytotoxic to the pig as well as to
the human islets and exhibited a characteristic Th1 cytokine profile
producing large amounts of TNF- and IFN- with no IL-4 secretion.
The Th0 type clone, clone #20, produced intermediate levels of TNF-,
IFN-, and IL-4. This clone also lysed both human and porcine islets.
However, the Th2 type clone (clone #33) did not show any cytotoxicity
against the target, and it secreted large quantities of IL-4, an
intermediate amount of TNF-, and no IFN-. Many studies have
identified that graft acceptance is associated with preferential
inhibition of Th1 cells and activation of Th2 cells (29, 30). The potential central role of IL-4 in tolerance induction
and maintenance has been proposed to be related to its ability to
regulate the commitment of Th0 cells to the Th2 phenotype and to
inhibit the development and function of Th1 cells. Note that clone #33
showed a 40-fold SI compared with that of the other clones in the
proliferation assay, supporting the fact that IL-4 increases T cell
differentiation. Our data, based on a limited number of clones
analyzed, supports such mechanisms in porcine islet rejection. These
results indicate that cytokines secreted by Th1 clones may cause damage
to porcine islets similar to what has been reported for islet allograft
rejection (12). However, Th2 clones producing IL-4 are
protective to the xeno islets, and results from clone #33 demonstrate
that it can neutralize the effect of TNF- because no islet lysis was
detected. Therefore, a balance of various cytokines in situ is
significant in determining islet survival.
Several studies have shown that indirect presentation of antigenic peptide by host APC constitute a pathway leading to the xenograft rejection (19). Our laboratory has demonstrated that human CD8+ T cells stimulated by porcine endothelial cells are capable of recognizing porcine Ag presented by porcine MHC class I molecules in a manner similar to an allogeneic response (16). This may be due to a close homology between porcine and human class I Ag at the molecular level. Further work from our laboratory has also shown that human CD4+ T cells recognize xenoantigens of porcine endothelial cells mostly by the indirect pathway as in the context of self-APC involvement (17). Results presented in this report using porcine islets agree with our earlier findings and clearly point out the important role of CD4+ T cells in the rejection of islet xenografts. Our results have also shown that the primary xenoantigen recognized is indeed the SLA class I molecule. This is most likely because of the fact that the porcine islets used in both our in vivo and our in vitro studies were cryopreserved and/or cultured for a period of time, and therefore MHC class II-expressing cells may have been depleted. However, if one performs transplantations of fresh islets, it is likely that peptides derived from SLA class II molecules may also be taken up and presented by self-APC. We are currently investigating this possibility. Also, we are currently identifying the epitopes of the SLA class I molecule recognized by the human T cells and by the restricting elements of the human APCs.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Thalachallour Mohanakumar, Washington University School of Medicine, Department of Surgery, Box 8109-3328 CSRB, 660 South Euclid Avenue, St. Louis, MO 63110.
3 Abbreviations used in this paper: HuPBL-SCID, SCID mice reconstituted with human PBLs; GIL, graft-infiltrating lymphocyte; SI, stimulation index; CML, cell-mediated lympholysis; SLA, swine leukocyte Ag.
Received for publication November 23, 1999. Accepted for publication May 16, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and glucose on major histocompatibility complex class I and II expression by pancreatic ß- and non-ß-cells. J. Clin. Endocrinol. Metab. 82:2329.This article has been cited by other articles:
|
|
 |
|
  G. R. Rayat, Z. A. Johnson, J. N. Beilke, G. S. Korbutt, R. V. Rajotte, and R. G. Gill The Degree of Phylogenetic Disparity of Islet Grafts Dictates the Reliance on Indirect CD4 T-Cell Antigen Recognition for Rejection Diabetes, June 1, 2003; 52(6): 1433 - 1440. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sharland, A. Patel, J. H. Lee, A. E. Cestra, S. Saidman, and G. L. Waneck Genetically Modified HLA Class I Molecules Able to Inhibit Human NK Cells Without Provoking Alloreactive CD8+ CTLs J. Immunol., April 1, 2002; 168(7): 3266 - 3274. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
