| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Center for Immunobiology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA 02115; and
Laboratoire d’Immunologie Experimentale (IMEX), Universite de Louvain, Brussels, Belgium
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B (7). Work from our laboratory has
demonstrated recently that the antiapoptotic genes A20,
bcl-2, bcl-xL, and A1 act as
inhibitors of NF-B, suppressing the expression of proinflammatory
genes in EC (8, 9). Given their dual role in protecting EC from
apoptosis and inhibiting EC activation, we refer to these genes as
protective genes (8). We have reasoned that under conditions that would
promote the expression of protective genes in EC, the proinflammatory
environment associated with xenograft rejection may be inhibited and
xenograft survival may be prolonged (8). We have demonstrated recently that long-term xenograft survival (>50 days) can be established in approximately 75% of hamster cardiac xenografts transplanted into rats treated with CyA and CVF (10), a modification of the protocol of Hasan and coworkers (11). Long-term xenograft survival is associated with: 1) expression in the xenograft endothelium of the protective genes heme oxygenase-1, A20, bcl-xL, and bcl-2, 2) an ongoing Th2 cytokine response, and 3) deposition of IgM EXA and activation of complement in the xenograft endothelium (10). We refer to the survival of these xenografts in the presence of anti-graft Abs and complement as accommodation (10, 12), a phenomenon first observed by Alexandre and colleagues in ABO blood group-incompatible allografts (13) and proposed by Bach and colleagues (12) to involve a physiologic modification of EC in their response to injury (8).
Hamster to rat xenografts undergoing rejection do not express
protective genes and have an ongoing Th1 cytokine response (10). In
addition to their role in activating EC (14), certain Th1 cytokines
such as IL-12 and IFN-
are involved in B cell isotype switch and
secretion of IgG2b (mouse IgG2a/IgG2b) and IgG2c (mouse IgG3) (15, 16, 17).
Rat IgG2b activates complement and mediates complement-dependent
cell-mediated cytotoxicity (CDC) as well as Ab-dependent cell-mediated
cytotoxicity, while IgG2c binds C1q, but does not mediate CDC or
Ab-dependent cell-mediated cytotoxicity (18). Most nucleated cells
actively counteract the cytotoxic effects of complement activation and
do not undergo necrosis (19). However, binding of C1q or generation of
sublytic levels of C5b-9 activates NF-B and up-regulates the
expression of proinflammatory genes in EC (20, 21).
The generation of EXA following transplantation has been suggested previously to contribute to xenograft rejection (22). However, a detailed analysis of the role of EXA in the pathogenesis of xenograft rejection has not been formally tested. In the present study, we have analyzed the role of EXA in the pathogenesis of xenograft rejection and investigated whether CyA and CVF administration suppress xenograft rejection by inhibiting the generation of EXA. We demonstrate that upon immunization with a hamster cardiac xenograft, untreated rats generate high levels of anti-hamster IgM and IgG EXA, which upon adoptive transfer into naive recipients induce xenograft rejection through complement activation. Administration of CyA, after transplantation, significantly decreased IgM and suppressed IgG EXA production to undetectable levels. Combination of CyA with CVF resulted in inhibition of complement activation during the transient rise of IgM EXA, occurring during 10 to 15 days after transplantation. Once complement returned to pretreatment levels, IgM XNA had dropped to basal levels, suggesting that xenograft EC may never be exposed to complement in the presence of high levels of IgM XNA, and therefore xenograft rejection does not occur. The expression of the protective genes heme oxygenase-1, A20, bcl-2, and bcl-xL in EC of the graft may then contribute to suppression of xenograft rejection by inhibiting EC activation upon complement activation by IgM XNA.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inbred male Lewis, ACI, and
rnu-/rnu- rats
(225–320 g) (Harlan Sprague-Dawley, Indianapolis, IN) were used as
xenograft recipients, and male Golden Syrian hamsters (70–120 g)
(Harlan Sprague-Dawley) were used as donors for cardiac xenografts.
Lewis rats express the Ig-1a light chain allotype, and ACI rats
express the Ig-1b light chain allotype (23). All animals were housed
in accordance with guidelines from American Association for Laboratory
Animal Care, and all research protocols were approved by Institutional
Animal Care and Use Committees of Beth Israel Deaconess Medical Center
(Boston, MA).
Surgical procedures were conducted under anesthesia with pentobarbital (30–50 mg/kg i.p.) (Abbott, North Chicago, IL). Heterotopic hamster to rat cardiac transplantation was performed as described before (10).
Experimental protocols
Induction of accommodation was achieved by combination of CyA (Novartis Pharma, Basel, Switzerland) (15 mg/kg, i.m., daily from day 0 until the end of the experiments) and CVF administration (Quidel, San Diego, CA) (60 U/kg 1 day before xenotransplantation and 20 U/kg at the time of transplantation). To analyze the individual effect of CVF and CyA on the generation of anti-hamster EXA, Lewis and rnu-/rnu- rats were immunized with hamster cardiac xenografts and treated with CyA, CVF, or combination of CyA and CVF, as described above. Cardiac xenografts were removed at day 2 to avoid EXA immunoabsorption.
To analyze the role of rat anti-hamster EXA in xenograft rejection, ACI rats were immunized with a hamster cardiac xenograft and sacrificed 14 days after xenotransplantation for serum collection. IgM, IgG1, IgG2a, IgG2b, and IgG2c were purified using Sepharose-4b (Pharmacia, Upsala, Sweden)-coupled mouse anti-rat IgM (MARM-4), IgG1 (MARG1-2), IgG2a (MARG2a-1), IgG2b (MARG2b-8), and IgG2c (MARG2c-5) mAb (Experimental Immunology Unit, University of Louvain, Brussels, Belgium), as described before (24). Concentration of rat Ig was measured by ELISA, as described before (25). The relative amount of EXA in each purified fraction was titrated by a cellular ELISA described below and compared with the initial immunized serum, taking into account at least three serial dilutions in the linear range of the assay. Purified Ig were administered (i.v.) into naive recipients, 30 min after xenotransplantation, at a volume corresponding to the amount of EXA of the same isotype detected in 200 µl of the initial immunized serum, which caused HAR of hamster xenografts.
Pathology and immunohistology
Hamster cardiac xenografts were harvested 3 and 24 h after
Ig transfer, snap frozen, and stored at -70°C. Cryostat sections
were analyzed by immunoperoxidase using biotin-conjugated anti-rat
Ig-1b (LORK1b) mAb (Experimental Immunology Unit, UCL) to detect
transferred Ab bound to hamster xenografts, plus monospecific
polyclonal Abs to C3 and P-selectin, as described (26).
Endotoxin levels in purified Ig fractions were evaluated by a bioassay
based on the induction of E-selectin protein expression on the surface
of porcine EC, as measured by a cellular ELISA, as described before
(27). All samples revealed only minor endotoxin contamination (
5
ng/ml).
Complement hemolytic assay (CH50)
CH50 units were defined as the dilution of rat serum required to
produce 50% maximal lysis of Ab-sensitized sheep erythrocytes.
Briefly, Ab-sensitized sheep erythrocytes (1 x
108 cells/ml; Sigma Chemical, St. Louis, MO) were
incubated (30 min, 37°C) with rat serum in gelatin veronal buffer
(GVB2+) (Sigma Chemical). Cells were centrifuged and
hemoglobin release was measured (
= 550 nm). Background was measured
in the absence of sheep erythrocytes or in the absence of serum, and
subtracted from all samples.
Cellular ELISA
Serum levels of rat anti-hamster EXA were measured by
cellular based indirect enzyme-linked assay (ELISA). The Syrian hamster
kidney cell line HAK (ATCC, CCL-15; American Type Culture Collection,
Rockville, MD) was used as an antigenic target. Briefly, HAK cells were
cultured in DMEM (Life Technologies, Gaithersburg, MD), 10% FCS, 100
U/ml penicillin, 100 µg/ml streptomycin (Life Technologies).
Glutaraldehyde-fixed HAK cells were incubated (1 h, 37°C) in the
presence of rat serum serially diluted in 0.05% PBS/Tween-20 (Sigma
Chemical), and rat anti-hamster EXA were detected using mouse
anti-rat Ig light chain (MARK-1), Ig-1b light chain (RG11/15)
(PharMingen, San Diego, CA), IgM (MARM-4), IgG1 (MARG1-2), IgG2a
(MARG2a-1), IgG2b (MARG2b-8), or IgG2c (MARG2c-5) (Zymed, San
Francisco, CA) mAbs (1 µg/ml, 1 h, room temperature). Mouse
anti-rat mAbs were detected using horseradish peroxidase-labeled
goat anti-mouse F(ab')2 fragments depleted from
anti-rat Ig cross-reactivity (0.1 µg/ml, 1 h, room
temperature) (Pierce, Rockford, IL). Rat anti-hamster EXA of the
Ig-1b allotype were detected using biotin-labeled rat anti-rat
Ig-1b (LORK1b) mAb (Experimental Immunology Unit, UCL).
Biotin-labeled rat mAb was detected using horseradish
peroxidase-labeled streptavidin (0.2 µg/ml) (Sigma Chemical).
Horseradish peroxidase was revealed using orthophenyldiamine (OPD;
Sigma Chemical) and H2O2 (0.03%) in citrate
buffer (pH 4.9). Absorbance was measured at = 490 nm. The relative
amount of circulating EXA in the serum was expressed as OD ( = 490)
taken from one serial dilution in the linear range of the assay
(1/32–1/1024).
Binding of rat C3 to hamster cells was measured by a modified cellular ELISA using HAK cells as antigenic targets. Briefly, nonfixed HAK cells were incubated in presence of rat serum serially diluted in GVB2+ buffer (1 h, 37°C). Cells were fixed in PBS and 0.05% glutaraldehyde, and rat C3 deposition was detected using a mouse anti-rat C3 mAb (Serotec, Oxford, U.K.).
Complement-mediated cytotoxicity
Preconfluent HAK cells were incubated with purified rat
anti-hamster EXA (1–2 h, 37°C) in DMEM, 10% FCS, 100 U/ml
penicillin, and 100 µg/ml streptomycin. Cells were washed in PBS,
incubated with 10% baby rabbit serum (Accurate Chemicals, Westbury,
NY) in GVB2+ buffer (1 h, 37°C), collected by
trypsin digestion (Life Technologies), and resuspended in PBS, 3% FCS,
and 10 µg/ml propidium iodide (15 min, room temperature) (Sigma
Chemical). Cells were analyzed for viability by flow cytometry (1
x 104 cells/sample) using a FACScan cytometer (Becton
Dickinson, Mountain View, CA) equipped with Cell Quest software (Becton
Dickinson). Alternatively, CDC was measured by a cytotoxicity assay
based on cleavage of the tetrazolium salt MTT (Sigma Chemical).
Briefly, preconfluent HAK cells were incubated in the presence of rat
anti-hamster purified EXA (1–2 h, 37°C), washed in PBS,
incubated with 10% baby rabbit serum in GVB2+ (1 h,
37°C), washed in PBS, and incubated (2 h, 37°C) with MTT (300
µg/ml). Cells were washed in PBS and lysed in 100% ethanol to
dissolve MTT crystals. Absorbance was measured at = 570 nm and the
percentage of cytotoxicity was calculated as follows: 100 - (100
x [ODsample - OD100% lysis]/[OD0%
lysis - OD100% lysis]). The spontaneous
cytotoxicity by baby rabbit serum in the absence of rat EXA was
subtracted from all experimental samples. One hundred percent cell
lysis was calculated after lysis of HAK cells in lysis solution
(Promega, Madison, WI) (10 µl, 1 h). Zero percent cell lysis was
calculated from HAK cells in culture medium.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rat IgM was immunopurified from the serum of naive or immunized
rats, as described in Materials and Methods. Myeloma IgM was
used as a negative control. IgM concentrations were measured by ELISA
(data not shown). IgM purified from control or immunized rat sera
recognized the hamster epithelial cell line HAK (Fig. 1
A). Recognition of
hamster cells was specific since myeloma IgM (IR202) recognize hamster
HAK cells only weakly, due to nonspecific binding of rat IgM in this
assay (Fig. 1A). In contrast to IgM purified from
immunized serum, IgM purified from control serum or from myeloma
ascites did not activate complement significantly and did not mediate
CDC of hamster cells (Fig. 1B).
|
Unlike IgM, circulating XNA of the IgG isotype remained at
background levels before xenotransplantation (Fig. 2). Following xenotransplantation, rats
generated high levels of anti-hamster IgM, IgG1, IgG2a, IgG2b, as
well as IgG2c EXA that were directed against hamster endothelium and
recognized HAK cells (Fig. 2). Anti-hamster IgM EXA reached a maximal
serum level at day 6 after transplantation and decreased thereafter.
The anti-hamster IgG EXA response was slightly delayed as compared
with the IgM response, and once it reached a maximal level
(approximately day 7–14) remained stable (Fig. 2). Increased serum
levels of IgM and IgG EXA correlated with increased C3 deposition on
the xenograft vascular endothelium (data not shown) and increased
complement activation in vitro, as revealed by the detection of C3
deposition on the surface of HAK cells (data not shown), as well as by
CDC of HAK cells (Fig. 2).
|
To evaluate the role of EXA in the pathogenesis of xenograft
rejection, IgM, IgG1, IgG2a, IgG2b, and IgG2c were purified from the
serum of untreated and immunized ACI rats (expressing the -1b Ig
light chain allotype) and transferred into naive Lewis rats (expressing
the -1a Ig light chain allotype), 30 min after xenotransplantation.
Immunization of ACI rats resulted in production of high levels of
anti-hamster EXA, as detected by mAbs specific for the rat Ig-1b
light chain allotype (LORK-1b) or the rat Ig-1a and -1b light chain
allotypes (MARK-1) (Fig. 3A).
|
B). The relative amount of
EXA in each isotype fraction was evaluated (Fig. 3B),
and the volume of purified Ig transferred was adjusted for each isotype
to correspond to the relative amount of EXA of the same isotype present
in 200 µl of the initial immunized serum (Fig. 3B),
which reproducibly induced HAR of hamster cardiac xenografts (Table I). Purified Ig fractions revealed only
minor cross-contamination with other Ig isotypes, as tested by cellular
ELISA (Fig. 3C).
|
).
The ability of IgM to activate complement was twofold higher than that
of IgG2b. Low but detectable CDC was also observed upon binding of
IgG1, and to a lesser extent IgG2c to HAK cells (Fig. 4). However, IgG1
and IgG2c mediated CDC 10- to 20-fold less efficiently than IgG2b or
IgM, respectively (Fig. 4).
|
are the results of administering unfractionated serum
or various purified Ig isotype fractions to rats having received a
xenograft 30 min earlier. Immunized serum induced HAR of hamster
xenografts in 7 to 15 min, while rats receiving control serum rejected
xenografts in 72 h (Table I). Transfer of IgM purified from
immunized serum induced xenograft rejection at 48 h, while
rats receiving an equivalent amount of IgM purified from nonimmunized
serum rejected xenografts in 72 h. Transfer of IgG1 or IgG2a
purified from immunized serum had no significant effect in terms of
xenograft survival. Transfer of IgG2b purified from immunized serum
induced xenograft rejection at 26 to 62 min, while transfer of IgG2b
purified from nonimmunized serum had no significant effect in terms of
xenograft survival. Transfer of IgG2c purified from immunized serum
induced xenograft rejection in 3 to 24 h. This effect was,
however, not observed in 2 of 10 animals analyzed. Transfer of
equivalent amounts of control IgG2c had no significant effect in
inducing xenograft rejection.
To evaluate the role of complement activation in the induction of
xenograft rejection, IgM, IgG2b, or IgG2c EXA were transferred into
CVF-treated rats, 30 min after xenotransplantation. Transfer of EXA had
no significant effect in inducing xenograft rejection, as compared with
recipients treated with CVF alone (Table I). These data demonstrate
that induction of xenograft rejection by IgM, IgG2a, or IgG2c EXA are
dependent on complement activation (Table I).
Detection of Ig-1b EXA after transfer of purified IgM,
IgG1, IgG2a, IgG2b, and IgG2c
Serum levels of Ig of the -1b allotype were evaluated 15 min
after transfer of Ig of the -1b allotype, purified from immunized
ACI rats, into Lewis rats expressing Ig of the -1a allotype. Ig of
the -1b light chain allotype remained at undetectable levels before
EXA transfer (Fig. 5, A and
B). The levels of circulating EXA of the -1b
allotype were similar in recipients receiving IgM, IgG1, IgG2a, IgG2b,
or IgG2c EXA, indicating that distinct ability of these EXA isotypes to
cause xenograft rejection was not related to variations in their serum
concentration.
|
Analysis of hamster cardiac xenografts 3 h after
transfer of purified Ig showed in each case deposition of EXA of the Ig
-1b allotype on the xenograft endothelium (Fig. 6, a, d,
g, j, and m) whereas the extent
of concomitant deposition of C3 (Fig. 6, b, e,
h, k, and n) and expression of
P-selectin (Fig. 6, c, f, i,
l and o) lm l lp o varied considerably
between the isotypes of Ig transferred. In recipients receiving
purified IgM, IgG2b, or IgG2c EXA, xenografts showed endothelial
deposition of C3, indicating that IgM, IgG2b, and IgG2c EXA activate
complement upon binding to the xenograft endothelium (Fig. 6, b, k, and n). Moderate to
marked microvascular expression of P-selectin was associated with
platelet aggregation and plugging of the microvessels (Fig. 6, c, l, and o). By contrast,
transfer of purified IgG1 or IgG2a EXA was associated with only minimal
or no C3 deposition in the xenograft endothelium, suggesting that upon
binding to the xenograft endothelium, IgG1 and IgG2a EXA do not
activate complement efficiently (Fig. 6, e and
h). P-selectin labeling was also absent upon transfer
of EXA of the IgG1 or IgG2a isotypes (Fig. 6, f and
I). Analyses of hamster xenografts 24 h after Ig
transfer no longer showed deposition of EXA of the -1b allotype on
cardiac endothelium, whereas endogenous EXA of the Ig-1a allotype
were detected (data not shown). This data indicate that upon binding to
the xenograft endothelium, the t1/2 of EXA is
shorter than 12 h.
|
We have demonstrated previously that rejection of hamster to rat
cardiac xenografts is suppressed under the administration of the T cell
immunosuppressant CyA and the complement inhibitor CVF (10). In our
previous study, CVF was administered during 11 days, i.e., 1 day before
xenotransplantation and during 10 days thereafter (10). To decrease the
toxicity and immunogenicity associated with the administration of CVF,
we have reduced the duration of the CVF treatment to 2 days, i.e., 1
day before xenotransplantation and at the time of xenotransplantation.
Combination of CyA and CVF resulted in long-term survival of hamster to
rat cardiac xenografts (>50 days), while xenograft survival was not
prolonged significantly in recipients treated with CyA alone or CVF
alone, as compared with untreated recipients (Table II). Xenograft survival was associated
with the expression of the protective genes A20, bcl-2,
bcl-xL, and heme oxygenase-1 in the xenograft
endothelium, as well as with an ongoing Th2 cytokine response (data not
shown). IgM and C3 were detected in the endothelium of xenografts
undergoing long-term survival (data not shown).
|
Administration of CyA alone or in combination with CVF
significantly decreased the generation of IgM EXA as compared with
untreated or CVF-treated rats (Fig. 7).
Administration of CVF alone or in combination with CyA had no
significant inhibitory effect on the generation of IgM EXA as compared
with controls or CyA-treated rats (Fig. 7). Administration of CVF led
to depletion of complement hemolytic activity (CH50) to undetectable
levels by the time of xenotransplantation and during 3 days thereafter,
regaining pretreatment levels 14 to 17 days after transplantation (Fig. 7). In rats treated with the combination of CyA and CVF, IgM XNA
decreased to preimmunization serum levels between days 3 and 7, at
which time the CH50 was still suppressed by 40 to 60% of pretreatment
levels (Fig. 7).
|
To measure rat complement activation by hamster cells, we
developed a semiquantitative cellular ELISA for the detection of rat C3
deposition on the surface of hamster HAK cells. Consistent with the
observation that serum from naive rats has preformed anti-hamster
IgM XNA (Fig. 1), naive rat serum exhibits low but detectable levels of
complement activation in the presence of hamster cells (Fig. 8). In untreated rats, a significant
increase in complement activation was detected 3 to 7 days after
transplantation (Fig. 8), which was correlated with detection of C3
deposition on the xenograft vascular endothelium (data not shown) and
xenograft rejection (Table II). In CyA-treated rats, a modest but
detectable increase in complement activation was observed 3 to 7 days
after transplantation (Fig. 8), which correlated with C3 deposition in
the xenograft endothelium (data not shown) as well as with xenograft
rejection (Table II). In CVF-treated rats, complement activation
remained undetectable until day 3 (Fig. 8) and increased rapidly
thereafter (Fig. 8), which was consistent with xenograft rejection
(Table II). In recipients treated with CyA and CVF, complement
activation remained at undetectable levels after transplantation (Fig. 8), which was consistent with the long-term survival (>50 days) of
hamster xenografts in CyA/CVF-treated rats (Table II).
|
Administration of CyA resulted in total suppression of
anti-hamster IgG EXA of all subclasses including IgG2c, which was
previously described to be mainly T cell independent (28) (Fig. 9). Administration of CVF significantly
decreased the serum levels of IgG1 and IgG2c EXA as compared with
untreated rats (Fig. 9).
|
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our results show that rats have low but detectable levels of
circulating preformed anti-hamster XNA of the IgM isotype, which
activate complement weakly (Fig. 1) and presumably do not initiate HAR
of hamster cardiac xenografts because of the weakness of complement
activation (Fig. 2). Following xenotransplantation, rats produce high
levels of circulating EXA of the IgM, IgG1, IgG2a, IgG2b, and IgG2c
isotypes, which activate complement efficiently and thus are likely to
be involved in xenograft rejection (Fig. 2). To analyze the ability of
different EXA isotypes to initiate xenograft rejection, we adoptively
transferred allotype-specific purified EXA into naive recipients that
received a cardiac xenograft 30 min earlier. Transfer of EXA of the
IgM, IgG2b, and IgG2c isotypes, but not of the IgG1 and IgG2a isotypes,
induced xenograft rejection in a complement-dependent manner (Table I).
These data are consistent with detection of C3 deposition in the
xenograft endothelium in recipients receiving IgM, IgG2b, or IgG2c EXA,
but not IgG1 or IgG2a EXA (Fig. 6), and with the observation that IgM,
IgG2b, and IgG2c EXA did not cause xenograft rejection when transferred
into complement-depleted rats (Table I).
It should be noticed that IgG2c EXA mediated xenograft rejection
through complement activation, while this IgG subclass did not mediate
CDC of hamster cells in vitro (Figs. 4 and 6) (18). One possible
explanation as to why this occurs may result from the fact that IgG2c
binds C1q as efficiently as other IgG subclasses that mediate CDC,
i.e., IgG2b, without activating the entire complement cascade (29).
Therefore, IgG2c may initiate xenograft rejection by promoting
endothelial cell activation through C1q, which has been shown to occur
in vitro (21).
An alternative explanation may be that IgG2c does not mediate CDC in vitro because of a species incompatibility between rat IgG2c and rabbit complement, used as a source of complement in the CDC assay. Therefore, it is possible that IgG2c may activate rat complement in vivo, and thus could initiate xenograft rejection through complement activation.
Taken together, our present data support the following model for the
rejection of hamster to rat cardiac xenografts. Following
xenotransplantation, rats generate high levels of EXA, among which IgM,
IgG2b, and IgG2c activate complement (Fig. 4) to mediate EC activation,
platelet aggregation, P-selectin expression, and microvasculature
congestion (Fig. 6). These events result in rapid generation of
microvascular hypoxia and provide a nidus for recruitment and
activation of host leukocytes, which will presumably contribute to
xenograft rejection (30, 31).
Our present data may also be relevant for the mechanism of pathogenesis of acute vascular rejection of allografts. Generation of T cell-dependent complement-fixing alloantibodies, mainly directed against MHC, may also contribute to the pathogenesis of allograft acute vascular rejection. However, in contrast to xenografts, complement inhibitor proteins expressed in allograft endothelium (CD46, CD55, and CD59) may, at least to some extent, inhibit complement activation by cytotoxic alloantibodies. This may not occur as efficiently in xenografts because the interaction of complement inhibitors with components of the complement cascade is species specific. Therefore, the threshold level of complement activation needed to initiate acute vascular rejection of allografts may be higher than that necessary to induce xenograft rejection.
Having established that EXA play a central role in the pathogenesis of
xenograft rejection, we asked whether the mechanism by which
administration of CVF and CyA led to long-term xenograft survival and
accommodation involved the inhibition of EXA (Table II). Our findings
show that in rats treated with CVF for 2 days, complement is suppressed
to undetectable levels during 3 days after transplantation and does not
return to normal levels until 14 days after transplantation (Fig. 7).
In rats treated with CyA and CVF, it is during this period of
complement depletion that there is an increase in the serum levels of
IgM XNA (Fig. 7). Once IgM XNA returned to preimmunization levels, by
day 14 after transplantation, complement regained pretreatment levels
(Fig. 7), suggesting that there is no time at which high levels of IgM
XNA and normal levels of complement are present simultaneously. These
events result in the absence of complement activation on the surface of
hamster cells (Fig. 8) and, presumably for this reason, in the absence
of delayed xenograft rejection (Table II).
It has been suggested recently by Lin et al. (32) that xenograft rejection can occur once the production of IgM EXA is suppressed by Leflunomide administration after transplantation. Our own findings using an anti-µ Ab, which we have previously shown suppresses circulating IgM to undetectable levels in adult rats (33, 34), lead to a different interpretation. We find that in the absence of circulating IgM, rats receiving only CyA do not reject hamster cardiac xenografts that undergo long-term survival (>100 days) (Sato et al., manuscript in preparation). Similar results were obtained in rnu-/rnu- rats treated with anti-µ mAb, indicating that IgM EXA play an essential role in the pathogenesis of xenograft rejection (Takigami et al., article in preparation). Furthermore, these data also indicate that under T cell immunosuppression, xenograft rejection does not occur in absence of IgM XNA. The difference between our results and those reported by Lin et al. (32) may result from the fact that in the study of Lin et al., preexisting IgM XNA were not depleted. These Abs may have a cumulative effect with time, which we suggest would lead to low levels of complement activation responsible for EC activation and xenograft infiltration by host macrophages and NK cells, leading to rejection.
Our data also indicate that a significant portion of the
anti-hamster IgM response is T cell dependent, which contrasts with
previous reports suggesting that anti-hamster IgM EXA are generated
exclusively in a T cell-independent manner (35, 36). This conclusion is
based on two observations: 1) In euthymic rats, but not in
rnu-/rnu- rats, the T
cell-immunosuppressant CyA significantly decreased the production of
IgM EXA after transplantation (Fig. 7), and 2) the levels of IgM EXA
produced after transplantation of a hamster heart in
rnu-/rnu- rats were
significantly lower than those detected in euthymic rats (Figs. 7 and 10).
The effects of CyA and CVF in inhibiting IgM EXA and complement may
contribute to avoid rejection in the first 3 to 6 days after
transplantation, but cannot account for long-term xenograft survival
(>50 days). However, CyA has additional effects that explain long-term
xenograft survival. Administration of CyA suppressed the generation of
all IgG EXA subclasses to undetectable levels, including those we have
shown to be involved in xenograft rejection, i.e., IgG2b and IgG2c
(Table I; Figs. 6 and 9). An additional effect of CyA that potentially
contributes to suppress xenograft rejection is to deviate T cell
activation toward a Th2 cytokine response (data not shown) (10).
Th2-driven cytokines inhibit the production of proinflammatory
Th1-driven cytokines such as IL-12 and IFN- (37), and may therefore
account for suppression of IgG2b and IgG2c EXA, since production of
these Ig isotypes is dependent on Th1 cytokines such as IL-12 and
IFN- (15, 16, 17).
In the present study, we were interested in investigating which classes and subclasses of EXA were responsible for rejection, and whether the production of the implicated Ig isotypes could be related to the accompanying T cell cytokine response observed during xenograft rejection. This question was founded in previous data showing that certain IgG subclasses, i.e., IgG2b, are associated with Th1 cytokines, whereas others are dependent on Th2-type cytokines, i.e., IgE and IgG1 (37, 38). We found, by adoptively transferring specific Ig subtypes, that IgM, IgG2b, and IgG2c were capable of inciting rejection in the presence of complement, while IgG1 and IgG2a did not lead to rejection. These responses correlate precisely with the pattern of Th1 vs Th2 cytokines present, i.e., long-term xenograft survival is associated with an ongoing Th2 cytokine response, while xenograft rejection is associated with an ongoing Th1 cytokine response (10). These data indicate that Th1-driven cytokines may be implicated indirectly in xenograft rejection by promoting the production of T cell-dependent complement-fixing EXA that will cause rejection upon binding to the xenograft endothelium. Given these findings, we tested how CyA and CVF were implicated in leading to long-term survival (>50 days). To allow xenograft survival for the first week, CVF had to be given only twice, on day -1 and on the day of transplantation. This treatment suppressed complement for a sufficiently long time that the increased levels of IgM EXA were not accompanied by a competent complement system capable of initiating rejection. By the time complement recovered to pretreatment levels, the IgM EXA titers had decreased to background, and thus complement activation on the surface of hamster cells did not occur. With regard to the long-term survival (>50 days) of the xenograft in the presence of normal IgM XNA levels and normal levels of complement, we found that CyA blocked all EXA IgG to undetectable levels, including those subclasses involved in xenograft rejection. Finally, under CyA administration, EC in the xenograft expressed a series of protective genes, such as heme oxygenase-1, A20, bcl-2, and bcl-xL, which might explain the lack of EC response to the sublytic effects of IgM and complement.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Miguel P. Soares, Center for Immunobiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 99 Brookline Avenue, Boston, MA 02115. E-mail address:
3 Abbreviations used in this paper: XNA, preformed xenoreactive natural Abs; CDC, complement-dependent cytotoxicity; CVF, cobra venom factor; CyA, cyclosporin A; EC, endothelial cell; EXA, elicited xenoreactive Ab; GVB, gelatin veronal buffer; HAR, hyperacute rejection; C18H16N5SBR; NF-B, nuclear factor-B.
Received for publication September 30, 1997. Accepted for publication December 10, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Wang, J. Arp, W. Liu, S. J. Faas, J. Jiang, D. R. Gies, S. Ramcharran, B. Garcia, R. Zhong, and R. P. Rother Inhibition of Terminal Complement Components in Presensitized Transplant Recipients Prevents Antibody-Mediated Rejection Leading to Long-Term Graft Survival and Accommodation J. Immunol., October 1, 2007; 179(7): 4451 - 4463. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hasegawa, K. Okada, Y. Takano, Y. Hiraishi, and Y. Okita Autologous fibrin-coated small-caliber vascular prostheses improve antithrombogenicity by reducing immunologic response J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1268 - 1276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Y. Ma, N. Nakamura, V. Dlabac, H. Naito, S. Yamaguchi, M. Ishikawa, M. Nonaka, M. Ishiguro, N. Kawasaki, S. Oka, et al. Isolation, Cloning, and Characterization of a Novel Phosphomannan-binding Lectin from Porcine Serum J. Biol. Chem., April 27, 2007; 282(17): 12963 - 12975. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Saadi, T. Takahashi, R. A. Holzknecht, and J. L. Platt Pathways to Acute Humoral Rejection Am. J. Pathol., March 1, 2004; 164(3): 1073 - 1080. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Sato, J. Balla, L. Otterbein, R. N. Smith, S. Brouard, Y. Lin, E. Csizmadia, J. Sevigny, S. C. Robson, G. Vercellotti, et al. Carbon Monoxide Generated by Heme Oxygenase-1 Suppresses the Rejection of Mouse-to-Rat Cardiac Transplants J. Immunol., March 15, 2001; 166(6): 4185 - 4194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Lin, M. P. Soares, K. Sato, E. Csizmadia, S. C. Robson, N. Smith, and F. H. Bach Long-Term Survival of Hamster Hearts in Presensitized Rats J. Immunol., May 1, 2000; 164(9): 4883 - 4892. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mahboubi, B. C. Biedermann, J. M. Carroll, and J. S. Pober IL-11 Activates Human Endothelial Cells to Resist Immune-Mediated Injury J. Immunol., April 1, 2000; 164(7): 3837 - 3846. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Lin, M. P. Soares, K. Sato, K. Takigami, E. Csizmadia, N. Smith, and F. H. Bach Accommodated Xenografts Survive in the Presence of Anti-Donor Antibodies and Complement That Precipitate Rejection of Naive Xenografts J. Immunol., September 1, 1999; 163(5): 2850 - 2857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Brouard, B. Vanhove, K. Gagne, A. Neumann, P. Douillard, A. Moreau, C. Cuturi, and J. P. Soulillou T Cell Repertoire Alterations of Vascularized Xenografts J. Immunol., March 15, 1999; 162(6): 3367 - 3377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Lin, M. P. Soares, K. Sato, K. Takigami, E. Csizmadia, J. Anrather, and F. H. Bach Rejection of Cardiac Xenografts by CD4+ or CD8+ T Cells J. Immunol., January 15, 1999; 162(2): 1206 - 1214. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
