| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
Programs in
* Immunology and Virology and
Molecular Medicine, and Departments of
Medicine and
Pathology, University of Massachusetts Medical School, Worcester, MA 01655
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The rejection of allogeneic bone marrow was
mediated by a radioresistant
CD8+TCR-+NK1.1- T cell
population. We conclude that a noncytopathic viral infection at the
time of transplantation can prevent engraftment of allogeneic bone
marrow and result in the death of sublethally irradiated mice treated
with costimulation blockade. Clinical application of stem cell
transplantation protocols based on costimulation blockade and tolerance
induction may require patient isolation to facilitate the procedure and
to protect recipients. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To overcome these potential toxicities, newer approaches to the generation of allogeneic hematopoietic chimerism have focused on inhibiting T cell costimulation (6, 7). Successful approaches in mice have included the following: sublethal irradiation combined with CTLA4-Ig and/or anti-CD154 mAb (6, 8), donor spleen cell transfusion plus anti-CD154 mAb (9), frequent injections of high doses of stem cells plus anti-CD154 mAb without myeloablative conditioning (10), and costimulation blockade combined with drug-induced myeloablation (11).
Allogeneic hematopoietic chimerism in each of these model systems has been documented to generate donor-specific transplantation tolerance in the absence of GVHD. In addition, our laboratory has extended the use of tolerance-based stem cell transplantation to the treatment of autoimmune diabetes; this usage requires not only the generation of allo-tolerance to islet grafts but also the prevention of recurrent islet-destructive autoimmunity (8). The potential clinical utility of allogeneic hematopoietic chimerism for allotransplantation (1), cancer immunotherapy (12), and the treatment of autoimmune disease (13) is clearly enormous.
Despite the promise inherent in stem cell transplantation protocols based on costimulation blockade, significant issues of safety and durability remain to be assessed. In particular, patients treated with partial myeloablation combined with costimulation blockade could be less resistant to viral infection and its associated pathophysiological effects. Many viral infections are known to impair hematopoiesis and to induce bone marrow failure in stem cell transplant recipients treated with conventional procedures; these include EBV (14), CMV (15, 16), human herpesvirus 6 and 7 (17, 18), and HIV (19, 20, 21), among others.
In mice, acute lymphocytic choriomeningitis virus (LCMV) infection has
been shown to induce transient anemia resulting from the production of
type 1 IFNs that suppress hematopoiesis (22). Persistent
infection with LCMV has also been shown to lead to aplastic anemia due
to chronic CD8+ T cell activation and induction
of IFN-
, TNF, and lymphotoxin- (23). Moreover, many
viral infections (24), including LCMV (25),
can enhance allograft rejection (26, 27). The effects of
viral infection on human stem cell graft recipients treated with
costimulatory blockade are unknown, but recent evidence suggests that
viral infection can prevent allogeneic bone marrow engraftment in mice
(28).
To analyze this problem, we used a mouse model system of allogeneic hematopoietic chimerism developed in our laboratory (8). This model combines sublethal irradiation and anti-CD154 mAb and generates durable allogeneic hematopoietic chimerism across an MHC mismatch at both major and minor histocompatibility loci. We now report that acute LCMV infection of allogeneic bone marrow recipients conditioned with sublethal irradiation, and anti-CD154 mAb prevents the establishment of allogeneic hematopoietic chimerism and is fatal. We further demonstrate that a fatal outcome appears to depend on type 1 IFNRs and that the failure to engraft allogeneic stem cells is dependent on a population of radioresistant alloreactive CD8+ T cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female C57BL/6 (H-2b, Ly5.2),
C57BL/6-Ly5.1, CBA/JCR (H-2k), and BALB/c
(H-2d) mice were obtained from the National
Cancer Institute (Frederick, MD). 129/Sv wild-type
(H-2b) and SV129 IFN-R knockout mice
(29) were obtained from a colony maintained by us. C57BL/6
mice in which the CD4, CD8, TCR-, or TCR-
lymphocyte
surface Ag gene was disrupted by homologous recombination were obtained
from The Jackson Laboratory (Bar Harbor, ME). All animals were
certified to be free of Sendai virus, pneumonia virus of mice, murine
hepatitis virus, minute virus of mice, ectromelia, lactate
dehydrogenase-elevating virus, mouse poliovirus, Reo-3 virus,
mouse adenovirus, LCMV, polyoma, Mycoplasma pulmonis, and
Encephalitozoon cuniculi. They were housed in a specific
pathogen-free facility in microisolator cages, given autoclaved food
and acidified water, and maintained in accordance with the guidelines
of the Institutional Animal Care and Use Committee of the University of
Massachusetts Medical School (Worcester, MA) and recommendations in the
Guide for the Care and Use of Laboratory Animals of the Institute of
Laboratory Animal Resources (National Research Council, National
Academy of Sciences).
Abs and flow cytometry
The MR1 hamster anti-mouse CD154 mAb was purified from ascites, diluted in PBS to a final concentration of 1 mg/ml (30), and injected at a dose of 0.5-mg i.p. on days 0 and +3 relative to bone marrow transplantation. Depleting Abs were used as follows: 1 mg anti-NK1.1 (clone PK-136; Ref. 31) i.p. on days -8, -1, and +6; 0.5 mg anti-CD4 (GK1.5) and anti-CD8 (2.43) i.p. on days -3, -2, and -1. Each mAb was documented by flow microfluorometry to deplete >97% of its target cell population.
FITC-conjugated anti-H-2Kb (AF6-88.5) and
anti-Ly5.2 (104); PE-conjugated
anti-H-2Kd (SF1-1.1), anti-Ly5.1 (A20),
anti-CD4 (L3T4), anti-CD8 -chain (53-6.7), anti-TCR
-chain (H7-597), anti-CD45R/B220 (RA3-6B2), anti-CD11b/Mac1
(M1/70), anti-NK1.1 (PK136), and anti-GR1 (RB6-8C5); and
biotinylated anti-H-2Kd (SF1-1.1) and
anti-Ly5.1 (A20) mAbs were all obtained from BD PharMingen (San
Diego, CA). Flow microfluorometry was performed as described
(8). Briefly, single cell suspensions were labeled with
Ab, rinsed, washed, fixed in 1% paraformaldehyde, and analyzed on a
FACScan (BD Biosciences, Sunnyvale, CA). Forward angle and side scatter
were used to distinguish lymphocytes, monocytes, and granulocytes. Dead
cells and erythrocytes were excluded by electronic gating. At least
104 events were analyzed for each sample. The
relative percentages of host- and donor-origin cells in the C57BL/6
(H-2Kb, Ly5.2) recipients of BALB/c
(H-2Kd) or C57BL/6-Ly5.1 (Ly5.1) bone marrow were
determined by flow microfluorometry. In preliminary experiments, known
mixtures of donor and host PBMC were analyzed, and it was determined
that the lower limit of sensitivity of the assay for detecting either
donor (H-2Kd or Ly5.1) or host
(H-2Kb or Ly5.2) cells was 0.5%. Because not all
hematopoietic cells express MHC class I Ag, the relative percentage of
donor-origin cells in chimeric mice recipients was calculated as
follows: [percentage of donor cells/(percentage of donor cells +
percentage of host cells)] x 100% (8).
Cell preparation and bone marrow transplantation
Recipient mice were treated with 6 Gy whole body irradiation
using a 137Cs source (Gammacell 40; Atomic Energy
of Canada, Ottawa, Ontario, Canada). This dose was documented in
preliminary experiments to be nonlethal for C57BL/6 mice. Within 1–3 h
of irradiation all recipients received a single i.v. injection of
18–25 x 106 donor bone marrow cells in a
volume of 0.5 ml via the lateral tail vein. Donor femurs and tibias
(female BALB/c or C57BL/6-Ly5.1 mice 
6 wk of age) were flushed with
RPMI medium using a syringe and 24-gauge needle. Cells were filtered
through sterile nylon mesh (70 µm; BD Biosciences, Franklin Lakes,
NJ), counted, and resuspended in RPMI. Recipient mice were females 6
wk of age. To assess chimerism, blood samples were obtained from mice
given donor bone marrow 2–4 wk earlier. Additional samples were
obtained periodically as described in Results. Hematopoietic
chimerism was defined as the presence of >2% MHC class
1+ donor-origin cells in peripheral blood.
Suspensions of spleen cells were filtered through sterile nylon mesh (70 µm) and centrifuged. Erythrocytes were lysed with hypotonic NH4Cl, and the spleen cells were resuspended in RPMI and counted using a hemocytometer.
LCMV infection and assay for infectious units of LCMV
Mice were inoculated i.p. with 5 x 104 PFU of LCMV, strain Armstrong, propagated in baby hamster kidney cells (25). Mice were inoculated with LCMV immediately after bone marrow injection, or 2 or 7 wk posttransplantation as described. LCMV viral titers were measured by LCMV viral plaque assay as described (32). Results are expressed as geometric mean titers, i.e., the arithmetic mean of the log10 values.
Skin transplantation
Full thickness skin grafts 
1 cm in diameter were
procured and transplanted as described (30). Grafts were
examined three times weekly, and rejection was defined as the first day
on which the entire graft surface appeared necrotic (30).
Grafts adherent to the bandage or fully necrotic on day 7 were deemed
technical failures and were excluded from analysis
(33).
Histology
Samples of transplanted skin, host skin, small intestine, large intestine, femur, spleen, and liver were recovered from selected experimental mice, fixed and stored in 10% buffered formalin, embedded in paraffin, processed for light microscopy, and stained with H&E. A qualified pathologist (B. A. Woda), who was unaware of the treatment status of specimen donors, performed histological analyses.
Statistical analysis
Parametric data are given as the arithmetic mean ± 1 SD. Duration of graft survival is given as the median survival time (MST). Graft survival among groups was compared using the method of Kaplan and Meier (34) and the log rank statistic (35). Values of p <0.05 were considered significant. Analysis of 2 x 2 tables used the Fisher exact statistic (36).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We first identified doses of sublethal irradiation and
anti-CD154 mAb that permitted the generation of mixed hematopoietic
chimerism in C57BL/6 (H-2b) mice transplanted
with either syngeneic (C57BL/6-Ly5.1) or fully allogeneic BALB/c
(H-2d) bone marrow. As shown in Table I
, mice that received 18–25 x
106 syngeneic bone marrow cells plus 6 Gy of
radiation uniformly became chimeric. Addition of anti-CD154 mAb
treatment was not required for the generation of syngeneic
hematopoietic chimerism and had no effect on the percentage of
syngeneic donor-origin cells present.
|
,
were reanalyzed at intervals up to 27 wk after transplantation. The
percentage of donor-origin cells remained stable and high for the
entire period of observation, >80% in 12 of 13 mice analyzed at the
end of the experiment.
A subset of the stably allochimeric mice shown in Table I was also
analyzed by flow microfluorometry to determine the percentage of
donor-origin T cells, B cells, macrophages, and granulocytes present in
peripheral blood. As shown in Fig. 1, donor-origin cells representing each of these four lineages were
present in allochimeric C57BL/6 recipients throughout the 9-wk period
of observation. The percentage of donor-origin T and B cells rose
monotonically over time, whereas the percentage of donor-origin
granulocytes was maximal 2 wk after transplantation and declined
thereafter. As expected, donor-origin cells representing all four
lineages were also present in the syngeneic chimeric recipients
generated using Ly5 congenic marrow; these lineages exhibited the same
temporal changes in percentage (data not shown).
|
Donor-specific skin allografts uniformly survive on mixed chimeric mice that receive anti-CD154 Ab
Having generated mixed hematopoietic chimerism in the absence of GVHD, we next documented the presence of donor-specific transplantation tolerance. To do so, a total of 20 C57BL/6 mice received BALB/c skin grafts 8–17 wk after transplantation of BALB/c bone marrow. Among these bone marrow recipients, nine had been conditioned with sublethal irradiation alone and were nonchimeric. The other 11 had been conditioned with sublethal irradiation and anti-CD154 mAb as described above, and at the time of skin grafting all were chimeric. The percentage of donor-origin PBMC in these recipients was 58–96%. All 11 skin grafts were still intact at the time the allochimeric animals were electively killed, 72–251 days after transplantation. In contrast, the MST of skin grafts in the nonchimeric mice was significantly shorter (n = 9; MST, 12 days; range, 10–12 days; p < 0.001 vs allochimeric recipients).
Three of the allochimeric mice with donor-specific (H-2d) skin grafts that had been in place for 30 days were selected at random and given a third party CBA/JCR (H-2k) skin graft on the contralateral flank. Survival of the CBA/JCR skin allografts was brief (MST, 11 days; range, 10–11 days). The BALB/c skin grafts on these mice were still intact at the conclusion of the experiment 127 days after the CBA/JCR skin allografts had been rejected. One additional allochimeric mouse received only a CBA/JCR skin allograft, and survival of that graft was brief (11 days), demonstrating that T cell function was present and that the mice were specifically tolerant to H-2d-expressing cells.
Histological analysis of transplanted skin was performed on a subset of two chimeric mice with healed-in BALB/c skin grafts that had survived intact for 205 and 212 days. In neither instance was there evidence of inflammation suggestive of graft rejection.
As expected, the donor-specific (C57BL/6-Ly5.1) skin grafts on chimeric C57BL/6 recipients of syngeneic C57BL/6-Ly5.1 bone marrow survived indefinitely. This was true both for recipients conditioned with radiation alone (MST, 138; range, 32–164 days; n = 5) and for recipients conditioned with both radiation and anti-CD154 mAb (MST, 156; range, 60–297 days; n = 23).
These data document a model system characterized by mixed hematopoietic chimerism and donor-specific transplantation tolerance in the absence of GVHD and minimal preparative risk to the recipient. Because the system accurately models an approach that could well be put into clinical practice, it was deemed appropriate for use in analyses of safety and durability in the presence of viral infection, which is a common complication of clinical bone marrow transplantation.
Early but not late LCMV infection abrogates allogeneic hematopoietic chimerism and is fatal
We addressed the issue of safety by studying the effects of viral infection on mice undergoing treatment to induce hematopoietic chimerism. We also addressed issues of both safety and durability by examining the effect of delayed exposure to virus on mice in which mixed chimerism had been successfully established 15 or 50 days earlier. We first studied the effect of LCMV infection at the time of tolerization and bone marrow transplantation. In two separate experiments, C57BL/6 mice were randomized into two groups and treated with radiation, anti-CD154 mAb, and either syngeneic C57BL/6-Ly5.1 or allogeneic BALB/c bone marrow as described above. The transplanted mice in both groups were then randomly assigned to one of four subgroups. The first subgroup received no further treatment. Mice in the remaining three subgroups were given an i.p. injection of LCMV, strain Armstrong, on the same day as transplantation, or on days 15 or 50 after transplantation.
As shown in Table II, there was no effect
of LCMV infection at any time point on the recipients of syngeneic
C57BL/6-Ly5.1 bone marrow with respect to the number of mice becoming
chimeric or the percentage of donor-origin cells present 2–9 wk after
transplantation. The percentage of donor-origin cells at each time
point was comparable to that observed in the uninfected control mice
(Table I, line 2). None of the mice in any group appeared sick or died
during the period of observation.
|
), indicating that the viral infection had blocked the development of
allogeneic hematopoietic chimerism. To determine to what extent these strikingly different outcomes were due to deleterious effects of allogeneic bone marrow vs rescue by syngeneic marrow, an additional cohort of control C57BL/6 mice was treated with radiation, anti-CD154 mAb, and LCMV but no bone marrow; 13 of 19 mice (68%) survived for >4 wk. This rate of survival was significantly higher than that of the LCMV-infected allogeneic bone marrow recipients described above (0%; n = 15; p < 0.0001) and statistically similar to the survival rate in the LCMV-infected recipients of syngeneic bone marrow (100%; n = 10; p = 0.07, Fisher exact statistic). In addition, nearly all control mice survived >4 wk (eight of nine treated with radiation and LCMV, five of five treated with radiation alone, five of five treated with radiation plus anti-CD154 mAb, and five of five treated with anti-CD154 mAb plus LCMV). These data indicate that the allogeneic bone marrow cells have a role in the LCMV-induced death of sublethally irradiated hosts treated with anti-CD154 mAb.
Durability of the chimeric state following delayed infection with LCMV
We next studied the durability of our chimeric state by delaying
LCMV infection until 2 or 7 wk after the establishment of mixed
allogeneic hematopoietic chimerism. Among the mice randomized to
receive LCMV 15 days after transplantation, one died before infection.
Among the remaining nine mice, all were chimeric on the day before
infection. Subsequent to infection, the percentage of donor-origin PBMC
declined by 17% on day 28 and remained at approximately this same
level on days 49 and 63 (Table II). None of these mice appeared ill and
none died. At each time point after infection, the percentage of
donor-origin PBMC in the LCMV-infected mice (Table II) was somewhat
lower than in the uninfected controls (Table I, line 4). This
experiment indicates that the deleterious effects of LCMV infection on
host and graft survival are confined to a narrow window of time during
the tolerization and transplantation process.
Among the mice randomized to receive LCMV 50 days after
transplantation, 9 of 10 were chimeric on the day before infection.
After infection, the percentage of donor-origin PBMC in the chimeric
mice declined by 11% on day 63 (Table II). None of the 10 mice
appeared ill and none died. At corresponding time points after
infection, the percentage of donor-origin PBMC in the LCMV-infected
mice (Table II) was again lower than in the uninfected controls (Table I, line 4).
Clearance kinetics of LCMV in infected hematopoietic chimeras
We next hypothesized that the differential survival of syngeneic
vs allogeneic bone marrow recipients was due to differential ability to
clear LCMV. To test this hypothesis we measured LCMV titers in chimeric
mice infected with LCMV at different times after transplantation. As
shown in Table III, all recipients of
allogeneic bone marrow infected on the day of transplantation had
failed to clear virus during the first 2 wk and died soon after. In
contrast, syngeneic chimeras infected on the day of transplantation
survived, but thereafter they were persistent carriers of virus. Viral
titers 2 wk after transplantation and infection were similar in the
allogeneic bone marrow recipients that died and in the syngeneic bone
marrow recipients that survived. The data suggest that neither viral
load per se nor ability to completely clear virus was the determinant
of differential survival.
|
Immunocompromised mice are known to become persistent virus carriers due to clonal exhaustion of LCMV-specific T cells (37). The persistent carrier state in allograft recipients infected on day 15 suggests that they may be immunocompromised at that time point.
The bone marrow and lymphoid compartments of allogeneic bone marrow recipients given anti-CD154 mAb and LCMV infection are markedly hypoplastic
To determine the cause of death in allogeneic bone marrow
recipients infected with LCMV on the day of bone marrow
transplantation, cohorts of control and infected bone marrow recipients
were killed 7 or 14 days after transplantation. Light microscopic
analysis of sections of spleen and femurs of infected mice
(n = 8) revealed severe reductions in the number of all
hematopoietic populations (Fig. 2).
Reductions in marrow cellularity averaged 86 ± 12% (range,
60–97%). Histologic examination of the spleens revealed lymphoid
depletion in all cases, and, with the exception of a single splenic
nodule in one mouse that showed regenerative activity, there was no
evidence of extramedullary hematopoiesis.
|
, cell counts in both
tissues were depressed immediately after irradiation and
transplantation. In synchimeric bone marrow recipients, cell counts
thereafter recovered, irrespective of LCMV infection at the time of
transplantation. Recovery in the spleen cell compartment of the
LCMV-infected mice was somewhat slower than in the bone marrow
compartment. Control uninfected allogeneic bone marrow recipients also
showed cellular recovery in both spleen and bone marrow. However, in
contrast to all other groups, cell counts in recipients of allogeneic
bone marrow that were infected with LCMV on the day of transplantation
never recovered, and all mice died before day 21.
|
R expression is required for LCMV-induced hypoplasia
and death in allogeneic bone marrow recipients given anti-CD154 mAb
Having discovered severe hypoplasia restricted to recipients of
allogeneic bone marrow and LCMV infection, we next sought to determine
its cause. We first hypothesized that the cause was related to cytokine
release. Reversible depression of hematopoiesis is known to occur early
in the course of LCMV infection and has been reported to be a direct
effect of IFN- (22). To determine the role of
IFN- in our model system, we repeated our experiments using
129/Sv IFN-R knockout mice (29).
129/Sv+/+ and 129/Sv IFN-R knockout mice
were irradiated (600 rad) and given 25 million BALB/c bone marrow cells
and two injections of anti-CD154 mAb on days 0 and +3. Half were
then infected with LCMV at the time of transplantation. As shown in
Table IV, control
129/Sv+/+ and 129/Sv IFN-R knockout
recipients readily accepted BALB/c bone marrow and were chimeric. As
expected, LCMV infection of control 129/Sv+/+
recipients led to failure of engraftment and death. In contrast, LCMV
infection of 129/Sv IFN-R knockout recipients also led to failure
of the bone marrow allograft, but all of the mice survived. As was the
case for LCMV-infected C57BL/6 recipient mice (Fig. 2), histologic
study of LCMV-infected SV129+/+ recipients
revealed bone marrow hypoplasia and splenic lymphopenia. In contrast,
the spleen and bone marrow of SV129 IFN-R knockout mice treated
in a similar way showed normal cellularity. This experiment suggests
that death of the host was the consequence of a type 1 IFN-mediated
process, but rejection of the allogeneic marrow graft was due to a
different mechanism.
|
+NK1.1- cells
prevent allogeneic bone marrow engraftment in recipients treated with
LCMV infection and anti-CD154 mAb
Although interference with the function of IFN-Rs prevented
death in LCMV-infected allogeneic bone marrow recipients treated with
anti-CD154 mAb, the allogeneic marrow still did not engraft and
survival depended on the recovery of the host marrow. Therefore, we
questioned whether failure of allogeneic bone marrow engraftment in the
presence of LCMV infection was the result of cell-mediated rejection.
To identify the cell type responsible for graft failure we conducted a
series of cell deletion studies focused on T cell subsets and NK cells.
NK cells were of particular interest because they reportedly play a
pivotal role in rejection of murine allogeneic bone marrow transplants
in irradiated hosts (38, 39, 40).
Studies in mice treated with cell-depleting reagents
We first studied cell-depleting reagents in uninfected mice. As
shown in the upper half of Table V,
pretransplantation administration of anti-NK1.1 mAb, anti-CD4
mAb, or anti-CD8 mAb had little or no effect on subsequent
hematopoietic chimerism or survival in C57BL/6 recipients of BALB/c
bone marrow and anti-CD154 mAb in the absence of LCMV.
|
. The principal finding was that treatment
with anti-CD8 mAb completely prevented the deleterious effect of
LCMV infection, as the great majority of treated mice both survived and
exhibited allogeneic hematopoietic chimerism. In contrast, treatment
with anti-NK1.1 mAb had no effect on LCMV-infected mice, as all
tested mice failed to engraft (n = 4) and most (eight
of nine) died. LCMV-infected mice treated with anti-CD4 mAb
displayed improved survival vs untreated LCMV-infected controls, but
(with the exception of a single mouse) allogeneic bone marrow still did
not engraft. As expected, and consistent with the results in Table I,
control recipients treated only with sublethal irradiation,
anti-CD154 mAb, and LCMV showed no evidence of allogeneic bone
marrow engraftment (n = 13), and the great majority (16
of 17) died. Studies in knockout mice
Because our data suggested that a CD8+
non-NK cell was responsible for the failure of bone marrow engraftment
in LCMV-infected mice, we next retested this hypothesis using
appropriate knockout mice as graft recipients. As shown in the upper
half of Table VI, the absence of cell
surface expression of CD4, CD8, TCR-, or TCR- had little or
no effect on hematopoietic chimerism or survival in C57BL/6 knockout
recipients of BALB/c bone marrow and anti-CD154 mAb in the absence
of LCMV infection.
|
knockout
mice resulted in greatly enhanced survival and robust hematopoietic
chimerism. Transplantation into LCMV-infected CD4 and TCR-
knockout mice resulted in slightly but not statistically significantly
improved survival, but there was no engraftment of donor bone marrow
(Table VI). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). These studies clearly document the risk posed to stem cell graft recipients by viral infection when it occurs at the time of the procedure. Although there have to date been no reports of adverse virus-associated events in clinical trials of costimulation blockade therapy of autoimmune diseases, adaptation of costimulation blockade to clinical stem cell transplantation will require careful attention to this risk and documentation of safety (41). Viral infection is one of the most important risks faced by allogeneic stem cell recipients. It may arise from infected transplanted tissue, from reactivation of latent host viruses as a consequence of an allogeneic stimulus and immunosuppressive treatment, or from exposure of an immunosuppressed host to exogenous environmental pathogens (42, 43, 44, 45, 46).
With respect specifically to hematopoietic stem cell transplantation based on costimulation blockade, viral infection can have at least two potentially serious consequences. The first is overwhelming infection leading to death in the context of an immune system that cannot generate a robust immune response. The present data clearly show that viral infection at the time of allogeneic stem cell transplantation and costimulatory blockade can impair virus clearance and be fatal even in the absence of lethal conditioning. Of particular note in this study is the LCMV strain we used. LCMV-Armstrong strain is relatively noncytopathic and can maintain persistent infection in C57BL/6 mice that remain otherwise asymptomatic (37). However, in our experiments, infection with LCMV-Armstrong strain was lethal, but only if hosts receive allogeneic (not syngeneic) bone marrow in addition to sublethal irradiation and anti-CD154 mAb. Fortunately, we observed that vulnerability to viral infection was confined to a narrow window of time after tolerization and bone marrow cell engraftment.
A second serious consequence of viral infection in allogeneic stem cell graft recipients is alteration of the host cellular milieu leading to host-vs-graft reaction and graft failure. Tolerance induced by a donor-specific transfusion and a short course of anti-CD154 mAb (47), like other costimulation blockade protocols that induce tolerance (48), is dependent on the deletion of host alloreactive CD8+ T cells. Consistent with concerns that LCMV infection could activate the host immune system and abrogate tolerance, we previously documented that infection with LCMV at the time of transplantation induced skin allograft rejection in mice treated with donor-specific transfusion and anti-CD154 mAb (25). We subsequently demonstrated that skin graft rejection in those LCMV-infected mice was due to the ability of LCMV infection to abrogate host alloreactive CD8+ T cell deletion (49). Others have shown that mice conditioned with busulfan and treated with anti-CD154 mAb, CTLA-4-Ig, and a bone marrow allograft lose that graft (but do not die) if infected with LCMV (28). In that report, the mechanism of allogeneic stem cell graft failure was not identified, but evidence was presented to suggest that it was dependent on dendritic cell activation. In this work we report that allograft failure is due to a radioresistant alloreactive CD8+ T cell that appears to be activated by the viral infection and that is not deleted by costimulation blockade in LCMV-infected hosts.
With respect to underlying mechanisms, our data set provides insight
into the distinct and independent factors that determine whether or not
the allogeneic stem cell graft will survive and whether or not the
recipient will survive irrespective of the success or failure of the
stem cell allograft. Our data show that allogeneic bone marrow is
unable to engraft in recipient mice treated with sublethal irradiation,
anti-CD154 mAb treatment, and an LCMV infection at the time of
transplantation. The mice appear to die as the result of bone marrow
hypoplasia 2–3 wk after transplantation. In contrast, syngeneic bone
marrow readily engrafts in identically treated LCMV-infected mice. The
simplest hypothesis to explain these findings was that LCMV-infected
mice receiving allogeneic bone marrow died due to the complete lack of
donor cell engraftment. We believe this hypothesis is not correct for
two reasons. First, we showed that 89% of LCMV-infected mice that
received irradiation but no bone marrow transplant survived. Second, we
showed that BALB/c stem cells were unable to engraft in IFN-R
knockout 129/Sv mice, yet these knockout mice survived for the duration
of the experiment (up to 7 wk). These data clearly indicate that simple
failure of allogeneic stem cells to engraft is, in and of itself,
insufficient to kill sublethally irradiated, LCMV-infected mice. The
data further document that the allogeneic bone marrow graft itself was
important in the death of the host.
We hypothesize that the donor allogeneic bone marrow contributed to a
fatal outcome in LCMV-infected recipients by initiating a graft-vs-host
reaction (GVHD) mediated by alloreactive T cells in the donor bone
marrow inoculum, thereby activating the host immune system
(50, 51, 52). The allogeneic marrow may also have induced a
host response against donor alloantigens. Host immune systems activated
by GVHD or allogeneic cells plus LCMV infection would be expected to
induce high levels of inflammatory cytokines in the host. Our data show
that IFN-R knockout mice that were given allogeneic marrow,
costimulation blockade, and LCMV infection did not die. It is known
that LCMV-infected mice produce high levels of IFN-
(53, 54, 55, 56), and in a preliminary study we confirmed
that IFN- can readily be detected in the serum of both wild-type
and IFN-R knockout C57BL/6 mice 2 and 4 days after LCMV infection
of recipients transplanted with either syngeneic or allogeneic bone
marrow (D. Forman, unpublished observations). IFN- is produced by
several cell types including dendritic-like cells (57, 58)
and monocytes/macrophages (59). It inhibits the
generation of both CFU and burst-forming units in long-term human bone
marrow cultures (60) and leads to bone marrow dysfunction
in vivo in both mice and humans (22, 61). In addition, the
transient pancytopenia that follows LCMV infection in normal mice
(22) does not occur in IFN-R knockout mice
(22).
If IFN- is the agent of lethality but not allograft rejection in
LCMV-infected mice, another mechanism must prevent allogeneic stem
cells from engrafting in mice that receive irradiation, anti-CD154
mAb, and LCMV infection. Because NK cells are activated by LCMV
infection (62) and play a key role in allogeneic bone
marrow graft rejection (63), we examined their role in our
system. We observed that depletion of NK1.1+
cells had no effect on allogeneic bone marrow engraftment or on
survival of LCMV-infected recipients of allogeneic bone marrow. These
data suggest that the cell type responsible for preventing allogeneic
bone marrow engraftment is a NK1.1-
cell.
To identify the cell that prevents allogeneic bone marrow engraftment
in LCMV-infected mice, we used C57BL/6 knockout mice and C57BL/6 mice
that had been depleted of various cell types. The critical observation
was that, after infection with LCMV, only three types of recipients
survived and became chimeric: mice depleted of
CD8+ T cells, CD8 knockout mice, and TCR-
knockout mice. These data indicate that the mediator of bone marrow
allograft destruction in LCMV-infected mice treated with costimulatory
blockade is a radioresistant
CD8+NK1.1-TCR+
T cell. This conclusion is supported by the observation that
anti-CD8 mAb treatment facilitates the induction of mixed
hematopoietic chimerism in sublethally irradiated mice given
anti-CD154 mAb plus allogeneic bone marrow grafts
(64), indicating an important role for alloreactive
CD8+ T cells in allogeneic bone marrow rejection.
In this work we show that this host alloreactive activity can be
greatly amplified by virus infection to overcome costimulation blockade
and prevent allogeneic marrow engraftment. The infection of donor bone
marrow cells by LCMV may also have contributed to this effect by
rendering them highly susceptible to a host CD8-mediated antiviral
immune response. Interestingly, the data suggest that a bone marrow
CD8+ facilitator cell (65, 66) may
not be required for allogeneic bone marrow engraftment in this
model.
Finally, in a previous report using nonobese diabetic and BALB/c mice as recipients (8), we showed that sublethal irradiation and costimulation blockade led to the generation of stable but complete allogeneic hematopoietic chimerism. The present study demonstrates that the same methods can be used to generate mixed hematopoietic chimerism. A state of complete chimerism has several theoretical disadvantages in clinical application. First, it retards and may compromise the immunocompetence of the recipient (67, 68). Second, it could increase the likelihood of GVHD (69, 70). However, even in recipients of "mini-transplants" who have become mixed chimeras following treatment with immunosuppression, sublethal myeloablation, and allogeneic stem cell transplantation, GVHD remains a major complicating factor (2, 3). Our data suggest that costimulation blockade can prevent GVHD but, in the presence of viral infection, can lead to different and potentially fatal complications.
We conclude that viral infection at the time of transplantation in mice
treated with anti-CD154 Ab can prevent the engraftment of
allogeneic bone marrow and lead to the death of recipients. The
mechanism of graft destruction appears to be mediated by a
radioresistant alloreactive CD8+ T cell. Graft
loss is associated with a fatal outcome despite the fact that
recipients received sublethal conditioning. The mechanism of death
appeared to depend on IFN-R expression on host cells. Finally, it
is important to note that both graft loss and a fatal outcome occurred
after challenge with a virus that is noncytopathic. It is of concern
that more virulent agents might have similar adverse consequences in
the context of less stringent conditioning, or at later time points
after bone marrow transplantation. Clinical application of stem cell
transplantation protocols based on costimulation blockade and tolerance
induction may require patient isolation to facilitate the procedure and
to protect recipients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Dale L. Greiner, Division of Diabetes, University of Massachusetts Medical School, 373 Plantation Street, Biotech 2, Suite 218, Worcester, MA 01605. E-mail address: dale.greiner{at}umassmed.edu
3 Abbreviations used in this paper: GVHD, graft-vs-host disease; LCMV, lymphocytic choriomeningitis virus; MST, median survival time.
Received for publication March 1, 2002. Accepted for publication April 24, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and TGF-1. Bone Marrow Transplant. 19:471.[Medline]
/ during acute infection with the noncytopathic lymphocytic choriomeningitis virus. J. Exp. Med. 185:517., and CTLA4. J. Clin. Invest. 101:2446.[Medline]
and IL-4 in acute graft-versus-host disease after allogeneic bone marrow transplantation in mice. J. Clin. Invest. 102:1742.[Medline]
is required for inhibition of acute graft-versus-host disease by interleukin 12. J. Clin. Invest. 102:2126.[Medline]
/ and interleukin 12-mediated pathways in promoting T cell interferon responses during viral infection. J. Exp. Med. 189:1315. interferon of macrophages infected with lymphocytic choriomeningitis virus clone 13: susceptibility to histoplasmosis. Infect. Immun. 63:1468.[Abstract]
-producing CD11c-, myeloid CD11c+, and mature interdigitating dendritic cells. J. Clin. Invest. 107:835.[Medline]
and - inhibit the in vitro differentiation of immunocompetent human dendritic cells from CD14+ precursors. Blood 96:210. interferon on human long-term bone marrow culture. Leuk. Res. 14:525.[Medline]
on normal human hematopoiesis: an immunohistochemical and morphometric study on trephine biopsy specimens. J. Interferon Cytokine Res. 18:247.[Medline]
-chain heterodimer expressed on a CD8+ bone marrow subpopulation that promotes allogeneic stem cell engraftment. Nat. Med. 6:904.[Medline]
