| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Experimental Immunology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; and
Department of Immunology, Weizmann Institute of Science, Rehovot, Israel
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Earlier work characterizing the B6D2F1 model of GVHD demonstrated a significant parent anti-F1 CTL precursor frequency difference between B6 and DBA/2 anti-B6D2F1. Spleen cells from B6 mice exhibited CTL precursor frequency of 1/1450 against B6D2F1 cells and induced acute GVHD, whereas DBA/2 spleen cells had a CTL precursor frequency of 1/13,500 against B6D2F1 cells and induced chronic GVHD (2), indicating a role for CD8+ T cells in the development of acute GVHD. In further support of this idea, depletion of CD8+ T cells from the B6 inoculum ablated the acute form of GVHD and induced chronic GVHD (17, 18). Furthermore, repeated injection of DBA/2 cells into B6D2F1 mice resulted in a shift from chronic to acute disease (19). When considered together, these results were interpreted to mean that CTL precursor frequencies were responsible for determining the outcome for this model of GVHD.
To test the hypothesis that CD8+ T cell precursor frequencies determined GVHD outcome in the parent-into-immune-competent F1 model, we compared the disease in B6D2F1 mice injected with parental cells with the disease in (BALB/c x B6) F1 (CB6F1) mice injected with parental spleen cells. Both are H-2d/b F1 hybrids, involving MHC class I and II disparity, and should result in acute disease following injection of B6 cells if the hypothesis that CTL precursor frequency determines disease outcome is correct. Based on our definitions of acute vs chronic disease by parental T cell repopulation, CTL activity, and autoantibody production, we observed that B6-into-CB6F1 GVHD was reproducibly detected as acute disease (weeks 1–3), followed by a shift to chronic GVHD.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B6, DBA/2, B6D2F1, BALB/c, and CB6F1 female mice, 6–8 wk of age, were purchased from either Animal Production Area (Frederick, MD) or Charles River Breeding Laboratories (Wilmington, MA). Animals were cared for in accordance with the guidelines set up by the Institutional Animal Care and Use Committee. An animal protocol for studying GVHD was reviewed and approved by the National Cancer Institute Animal Care and Use Committee before starting experiments.
Preparation of murine cells
Single-cell suspensions were prepared from the spleens of mice in Dulbecco’s PBS (DPBS; Life Technologies, Grand Island, NY). Cell suspensions were filtered through a 70-µm sterile mesh screen (Becton Dickinson Labware, Franklin Lakes, NJ), and the cells were washed. For some applications, erythrocytes were lysed by using ACK Lysis Buffer (BioWhittaker, Walkersville, MD), or erythrocytes and dead cells were removed by using Lympholyte M (Accurate Chemical and Scientific, Westbury, NY). Cells were suspended in either DPBS or complete medium (CM) comprised of RPMI 1640 (Life Technologies) supplemented with 2 mM L-glutamine, penicillin, streptomycin, 5 x 10-5 M 2-ME (Life Technologies), 100 µM nonessential amino acids (Life Technologies), 5 µM HEPES (Life Technologies), and 10% heat-inactivated FBS (HyClone, Logan, UT) after the final wash, depending on use.
Induction of GVHD
Washed B6, DBA/2, BALB/c, B6D2F1, and CB6F1 splenocytes were suspended at 120 x 106 viable cells/ml in DPBS. GVHD was induced by the injection of 60 x 106 parental cells i.v. into CB6F1 or B6D2F1 mice. Control mice included uninjected age-matched mice and mice injected i.v. with 60 x 106 syngeneic F1 splenocytes. To maintain as much homogeneity between donor cell populations, both F1 combinations were injected on the same day using cells processed simultaneously under the same conditions. Both CB6F1 and B6D2F1 mice were injected from the same pool of B6 splenocytes.
Flow cytometric analysis of parental cell engraftment
The spleens of GVHD mice were harvested at 1–4, 6, 8, and 12 wk
postinoculation. Single-cell suspensions of splenocytes were prepared,
and the number of cells per spleen was counted. Erythrocytes and dead
cells were removed using Lympholyte M. Splenocytes were stained with
FITC-conjugated anti-H-2d (clone
SF1-1.1), PE-conjugated
anti-H-2b (clone AF6-88.5) to distinguish
parental from F1 cells in the presence of the
FC
RIII clone 2.4G2 to block nonspecific staining. To further
identify the splenocyte populations anti-CD3 (clone 145-2C11),
anti-CD4 (clone RM4-5 or H129.19), anti-CD8
(clone 53-6.7),
or anti-CD19 (clone 1D3) conjugated to CyChrome or biotin plus
streptavidin-CyChrome were used. All mAbs were purchased from
PharMingen (San Diego, CA). Data were collected on a FACScan flow
cytometer and analyzed with CellQuest Software (Becton Dickinson, San
Jose, CA).
Autoantibody detection
ELISAs of serum Abs recognizing ssDNA (20) and
dsDNA (21, 22) were performed as described previously.
Briefly, for ssDNA assays, 96-well Maxisorp plates (Nalge Nunc
International, Roskilde, Denmark) were incubated with 10 µg/ml
methylated BSA (Sigma, St. Louis, MO) for 90 min at room temperature,
followed by incubation of the plates for 2 h with 10 µg/ml calf
thymus DNA (Sigma) that had been heated to >80°C for 15 min. Plates
were washed and blocked overnight with 5% FBS in DPBS. Mouse serum was
incubated on the plates for 2 h in 5-fold serial dilutions from
1:10 to 1:1250. Bound serum Abs were detected using goat anti-mouse
IgG conjugated to HRP (Cappel, Aurora, OH) and visualized using
2,2'-azinobis-(3-ethylbenzthiazoline-6-sulfonate) (Kirkegaard & Perry
Laboratories, Gaithersburg, MD) and measuring the OD at 405 nm. For
dsDNA assays, 96-well Maxisorp plates (Nalge Nunc International) were
incubated with 5 µg/ml poly-L-lysine for 2 h at room
temperature followed by incubation with 5 µg/ml 
phage DNA
(Boehringer Mannheim, Indianapolis, IN) for 2 h at room
temperature, then 4°C overnight. Other components of this assay were
performed as described above.
ELISAs to detect the presence of Ab to ENA were performed using plates from RELISA ENA single-well screening kits (Immuno Concepts, Sacramento, CA). Each microtiter well is coated with Sm, RNP, SS-A, SS-B, Scl-70, and Jo-1 autoantigens. Plates were blocked for 2 h at 37°C. Mouse serum was added to the plates in 5-fold serial dilutions from 1:10 to 1:1250. Bound serum Abs were detected with goat anti-mouse IgG-HRP (Cappel) and 3,3',5,5'-tetramethylbenzidine (Kirkegaard & Perry Laboratories) stopped with 2 N H2SO4. OD was measured at 450 nm.
Detection of Ig complexes in kidney sections
Detection of Ig complexes in kidney sections were performed as previously described (23). Briefly, kidneys were removed from mice following euthanasia and snap frozen in liquid nitrogen. Six-micrometer sections were cut, air dried, and fixed with acetone. Ig deposits were detected using FITC-labeled goat anti-mouse IgG (Sigma) incubated on slides for 30 min and extensively washed with PBS. A fluorescence microscope was used to visualize specific staining. Immune complexes were evaluated according to density and strength of staining with scores ranging from (-) to (+++). Scores were converted to a numerical value and a mean and SD for each group of mice was determined.
CTL assays
Responding T cells from control and GVHD mice were prepared from pooled spleens, filtered, and washed with DPBS. Stimulator cells were splenocytes from normal F1 mice treated with ACK lysing buffer and irradiated with 2000 cGy. Final cell concentrations were 2.5 x 106 responding cells/ml and 1.25 x 106 stimulators/ml in CM. Cultures were incubated at 37°C with 5% CO2 for 5 days. Pooled effector cells were harvested and counted on day 5 for use in the CTL assays. Target cells were P815 (H-2d) and EL-4 (H-2b) cell lines labeled with 300 µCi 51Cr for 90 min at 37°C and washed extensively. Target and effector cells were incubated together at 37°C for 4 h at various E:T ratios and compared with target cells and CM for spontaneous release and target cells plus 3% Triton X-100 for maximum release. Supernatants were harvested using Skatron (Sterling, VA) harvesting filters and frames. The percent specific lysis equals [(experimental - spontaneous)/(maximum - spontaneous)] x 100.
Limiting dilution CTL assays
Single-cell suspensions were prepared from the spleens of B6, DBA/2, BALB/c, B6D2F1, and CB6F1 mice. Stimulator cells were CB6F1 or B6D2F1 splenocytes irradiated with 2000 cGy and used at 1 x 106 cells/well in 96-well round-bottom plates (Costar, Corning, NY). CM was supplemented with recombinant murine IL-2 (BioSource International, Camarillo, CA) at a final concentration of 20 U/ml for these assays. Cells were cultured for 8 days at 37°C with 5% CO2. Target cells were P815 (H-2d) and EL-4 (H-2b) cell lines labeled as above. Target cells were added to the limiting dilution wells at 3 x 103 cells/well. After the addition of target cells, plates were briefly spun to collect cells at the bottom of wells. Plates were incubated for 4 h at 37°C.
Limiting dilution assays using B6 anti-CB6F1 and B6 anti-B6D2F1 were performed using 2-fold serial dilutions of responding cells in the range of 1.6 x 104 to 125 cells/well. Assays using DBA/2 anti-B6D2F1 and BALB/c anti-CB6F1 were set up using 2-fold serial dilutions of responding cells in the range of 5 x 104 to 781.25 cells/well. All cell concentrations were tested using 24 replicates. Spontaneous release of 51Cr was determined by incubating target cells for 4 h in wells that contained stimulator cells stimulated with CM and IL-2 for 8 days. All of the supernatant was harvested as above. Wells were scored positive for CTL activity if the cpm for a well was greater than the mean spontaneous release plus 3 SDs.
Both CTL precursor frequencies and Th precursor frequencies described below were determined by the least squares method using a computer program supplied by Dr. Charles Orostz (Ohio State University, Columbus, OH).
Limiting dilution proliferation assays
Single-cell suspensions of splenocytes from B6,
B6D2F1, and CB6F1 mice were
prepared as described above and used as responder cells. Stimulator
cells were B6, B6D2F1, or
CB6F1 splenocytes treated with 50 µg/ml
mitomycin C (Sigma) per 50 x 106 cells for
30 min at 37°C. Assays were conducted in flat-bottom 96-well plates,
using 2.5 x 105 stimulator cells/well and
variable numbers of responder cells in CM. Responder cells were used in
0.75-fold serial dilutions from 1 x 105 to
7508.5 cells/well in 48-well replicates. Four milliliters of responder
cell suspension at 1 x 106 cells/ml was
added to 2 ml of medium in a fresh tube. The new cell suspension was
mixed thoroughly, and 4 ml was removed to a fresh tube containing 2 ml
of media. This process was continued until eight cell dilutions were
ready for use in the assay. One hundred microliters of responder cells
and stimulator cells were used per well. Plates were incubated for 4
days at 37°C, pulsed with 1 µCi/well
[3H]thymidine, and incubated overnight. Plates
were harvested using a Tomtec plate washer (Wallac, Gaithersburg, MD)
and counted with a 
plate reader (Wallac).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It was previously demonstrated that injection of B6 splenocytes
into immunocompetent B6D2F1 mice resulted in high
levels of parental cell engraftment and a short lethal disease course
(2). Injection of DBA/2, the other parent, resulted in a
chronic progressive disease with little parental cell engraftment
(2). Typical engraftment profiles for acute and chronic
GVHD obtained in this study are shown for this strain combination in
Fig. 1
A. Injection of BALB/c
splenocytes into CB6F1 resulted in the
engraftment pattern expected for chronic GVHD, with little engraftment
of parental lymphocytes (Fig. 1B). Injection of B6
splenocytes into CB6F1 resulted in an initial
burst of parental cell engraftment similar to that of
B6-into-B6D2F1. However, between the third and
fourth week of GVHD, the percentage of parental cells unexpectedly
began to decrease. This decline in parental B6 cells continued until
week 12, at which time the disease mirrored chronic GVHD (Fig. 1B). In contrast to the acute GVHD of
B6-into-B6D2F1 mice that die within 4–8 wk (Ref.
7 ; current study), we have observed
B6-into-CB6F1 GVHD mice surviving for >30
wk.
|
88% (±1.9%) donor lymphocytes
and 11.2% (±5.6%) host lymphocytes. The majority of donor cells in
the affected spleens are CD4+, making up 37% of
the total cell number (Fig. 2A).
B6-into-CB6F1 mice exhibited a similar pattern of
donor/host cell changes in the spleens of affected mice. Donor
lymphocytes expanded over the first 3 wk of disease, with donor cells
peaking at 62.2% (±11.7%). Donor CD4+ T cell
numbers increased to 16.5% of the spleen and donor
CD8+ T cell numbers increased to 10.6% of the
spleen at week 3 (Fig. 2B).
|
Cellular changes in GVHD mice
Although the percent changes in cells give an overview of the cell
populations over time, they do not take into account the dynamics of
changes in spleen size and cellularity during the course of disease. To
obtain a more complete understanding of the changes within the spleens
of the GVHD mice, the number of parental and F1
splenocytes was determined by multiplying the percentage of each cell
type in a given spleen by the total number of cells recovered from that
spleen. In acute GVHD there was a drastic decrease in
B6D2F1 cell number by week 2 (Fig. 3A). This dramatic decrease
seen in host spleen cell numbers was due to the elimination of
B6D2F1 CD4+,
CD8+, and CD19+ cells (Fig. 4A). Concurrently, the B6
cells that were injected into the F1 hosts
expanded and gradually became the major cell population in the
F1 spleens. At 1 wk post-GVHD, there were
1.56 x 107 B6 CD4+ T
cells in spleens of B6-into-B6D2F1 mice. This
number decreased to 3.34 x 106 cells at 4
wk and rebounded to 3.82 x 107 cells at
week 8 (Fig. 5A).
CD8+ T cells of B6 origin followed a similar
pattern of expansion, regression, and expansion during acute GVHD (Fig. 5A) with cell numbers ranging from 3.96 x
106 to 1.21 x
107.
|
|
|
, A and
B). Converting this percentage to cell numbers resulted in a
similar picture to that seen with percentages. Both DBA/2 and BALB/c
parental cells were detected at low levels in their respective
F1 hosts throughout all the time points tested
(Fig. 3, B and D). The parental cells that were
detected in the spleens of affected mice were predominantly
CD4+ T cells (data not shown) as previously
reported (1, 16). These CD4+ T cells
have been suggested to provide help for driving host B cells and the
autoimmune symptoms that are manifested later in the disease
(16). F1 CD4+
and CD8+ T cell levels remained relatively
constant over the course of GVHD, whereas CD19+ B
cells levels transiently increased during the initial phase of the
disease, returned to normal levels, and then began to increase again at
the later time points (Fig. 4, B and D).
B6-into-CB6F1 GVHD presents complex cellular
interaction patterns not typical of either of the other two forms of
GVHD. The number of CB6F1 cells increased during
the first 2 wk. A dramatic decrease in the number of
CB6F1 cells was observed between weeks 2 and 3,
simultaneous with an increase in the number of parental B6 cells (Fig. 3C). The loss of CB6F1 cells included
CD4+, CD8+, and
CD19+ cells (Fig. 4C), whereas the
increase in B6 cells involved CD4+ and
CD8+ T cells (Fig. 5B). By week 4,
these spleens had decreased in size from a peak at week 2 of 200
x 106 to 60 x 106
(Fig. 3C). The continued decrease in spleen cell number
between weeks 3 and 4 was due to a decrease in B6 cells (Fig. 5B). B6-into-CB6F1 GVHD continued to
diverge from acute GVHD as disease progressed. The spleens reached a
maximum size of 250 x 106 cells by 6 wk
postdisease induction. During this 2-wk period, massive expansion of
CB6F1 cells occurred. Both
F1 CD4+ and
CD19+ cells expanded to greater than normal
levels, whereas the F1 CD8+
T cells returned to normal levels (Fig. 4C). Concurrently,
parental B6 cells underwent a small transitory increase in number (Fig. 5B). By 12 wk postdisease induction, the spleens in
B6-into-CB6F1 GVHD mice exhibited a cellular
consistency that was similar to that of chronic GVHD mice.
Limiting dilution CTL assays
It was important to compare the CTL precursor frequencies of B6
anti-B6D2F1 and B6
anti-CB6F1 because the
B6-into-CB6F1 GVHD resulted in the unexpected
sequential acute to chronic pattern of GVHD (on the basis of host
spleen repopulation and survival patterns), and acute vs chronic GVHD
models were associated with differences in parent
anti-F1 CTL precursor frequencies
(2). Therefore, we repeated and verified the earlier
limiting dilution experiments indicating that the acute
B6-into-B6D2F1 GVHD vs the chronic
DBA/2-into-B6D2F1 GVHD could be accounted for by
differences in parent anti-F1 CTL precursor
frequencies. Our frequencies were 1/2,295 for B6
anti-B6D2F1 (compared with the earlier
1/1,450), and 1/19,510 for DBA/2 anti-B6D2F1
(compared with 1/13,500) (2) (Fig. 6). We also simultaneously compared the
above parental anti-F1 CTL precursor
frequencies in the B6 anti-CB6F1 (1/2,370)
and BALB/c anti-CB6F1 (1/19,230) CTL
precursor frequencies (Fig. 6). Despite the differences noted above in
the GVHD profiles of B6-into-B6D2F1 and
B6-into-CB6F1 (
Figs. 1–5), the B6
anti-F1 CTL precursor frequencies were
indistinguishable. Furthermore, the CTL precursor frequencies for DBA/2
anti-B6D2F1 and BALB/c
anti-CB6F1 were indistinguishable from each
other, although they were 9-fold lower than the B6
anti-F1 frequencies.
|
Acute GVHD is characterized by the expansion of parental
CD8+ T cells and the ability to isolate
anti-F1 and
anti-H-2d cytotoxicity from the spleens of
affected mice (2, 3, 5, 23). To further characterize the
B6-into-CB6F1 GVHD, spleens were removed at 2-wk
intervals and tested for anti-H-2d CTL
activity. At 2 and 4 wk post-GVHD induction, similar
anti-H-2d CTL activity was detected in
spleens from both acute and B6-into-CB6F1 GVHD
mice (Fig. 7, A and
B). This CTL activity could be detected at all time points
tested in mice with acute GVHD (Fig. 7, C and D).
Anti-H-2d CTL activity could be detected in
B6-into-CB6F1 GVHD during the first two, but not
at the later two, time points (Fig. 7, C and D).
B6D2F1 and CB6F1 mice with
chronic GVHD were also tested at all four time points for the presence
of anti-H-2b CTL activity. At no time during
the 8 wk was anti-H-2b activity detected
(data not shown).
|
Chronic GVHD is associated with the production of Abs recognizing
ssDNA (11, 12), dsDNA (11, 12, 13, 14, 15), and ENA
(11, 15, 16). GVHD mice were bled at 16 wk post-GVHD
induction, and the serum was tested for autoantibody production.
B6D2F1 mice injected with DBA/2 cells elicited
Abs to all Ags tested (Fig. 8, A–C). The
BALB/c-into-CB6F1 chronic GVHD mice and the
B6-into-CB6F1 GVHD mice resulted in Abs being
produced to both ssDNA (Fig. 8A) and dsDNA (Fig. 8B) at higher titers than seen with
DBA/2-into-B6D2F1. At 16 wk post-GVHD induction,
B6-into-CB6F1 and both forms of chronic GVHD
produced Abs recognizing ENA (Fig. 8C). These Abs were not
detected in syngeneic injected F1 controls.
|
A;
CB6F1, Fig. 9C). Both models of
chronic GVHD, DBA/2-into-B6D2F1 (Fig. 9B) and BALB/c-into-CB6F1 (Fig. 9D), had significant Ig deposits. The
B6-into-CB6F1 GVHD also had significant numbers
of Ig deposits within the kidneys (Fig. 9E). Eight mice in
each group were analyzed for Ig deposits, and numerical values were
assigned to each score to compare staining between groups (Table I). These results indicated that both
models of chronic GVHD, DBA/2-into-B6D2F1 and
BALB/c-into-CB6F1, resulted in very similar
kidney-staining patterns and intensity.
B6-into-CB6F1 mice also had Ig deposits in the
kidney, but the mean score of the group was lower than that seen in
either of the chronic models of GVHD. Because the
B6-into-CB6F1 mice undergo a period of acute
disease before progressing on to chronic disease, differences seen in
kidney-staining patterns might be due to a delay in this group in
developing renal pathology. Although the initiating events of chronic
and B6-into-CB6F1 GVHD are very different, the
end result of these two pathways is an autoimmune disease similar to
lupus.
|
|
Because CTL precursor frequencies alone did not account for the
differences between acute and B6-into-CB6F1 GVHD,
another mechanism must exist. Parental Th cells play a role in driving
the F1 B cell to hyperactivity in chronic GVHD
and are presumed to play a supportive role for the CTL in acute GVHD
(24). Therefore, it is also possible that differences in
the parent anti-F1 Th cell frequency
contribute to determining the final disease outcome. Limiting dilution
proliferation assays were performed to test for differences in the B6
response against mitomycin C-treated B6D2F1 and
CB6F1. The Th cell limiting dilution curves for
B6 anti-CB6F1 and B6
anti-B6D2F1 were identical and had a
frequency of 1/79,120 (data not shown). It is possible that these
proliferative Th cell frequencies are not different or that the
stimulating alloantigens were limiting on F1
stimulator cells, resulting in the Th frequencies appearing to be
indistinguishable. Therefore, these limiting dilution proliferation
assays were repeated using mitomycin C-treated homozygous BALB/c and
DBA/2 stimulator cells. We observed a >3-fold difference in Th
precursor frequency using allogeneic homozygous
H-2d stimulator cells (Fig. 10). B6 anti-DBA/2 exhibited a Th
precursor frequency of 1/19,420 spleen cells, whereas B6
anti-BALB/c had a Th precursor frequency of 1/65,900.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, 5), loss of
F1 lymphocytes (
Figs. 2–4), and detectable
parent anti-F1 CTL activity in the spleens of
affected mice (Fig. 7). As the disease continues, distinct differences
become apparent between B6-into-CB6F1 and
B6-into-B6D2F1 mice. In
B6-into-B6D2F1 mice, the CTL activity continues
to be detectable through week 8 (Fig. 6), and the level of donor cell
engraftment remains high (
Figs. 1–3, 5), correlating with data
published using other acute parent-into-F1
combinations (1). In B6-into-CB6F1
mice, the F1 cells begin to proliferate between
weeks 4 and 6 (Figs. 2 and 3), resulting in a drastic decrease in the
percentage of parental cells in the spleen (Fig. 1), although the
absolute number of B6 cells increases during this same time frame. The
number of CB6F1 cells increases from 2.9 x
107 to 1.9 x 108
cells/spleen between weeks 4 and 6, whereas the number of parental
cells increases from 1.1 x 107 to 5.7
x 107 cells/spleen. This results in the
percentage of parental cells dropping from 44.8 to 22.4%. The
percentage of parental cells continues to decline until week 12, when
12% of the cells are of parental origin.
At later time points, B6-into-CB6F1 GVHD exhibits
characteristics of chronic GVHD, a lupus-like illness. This condition
manifests itself in the production of autoantibodies recognizing ssDNA,
dsDNA, and ENA. But more importantly, the disease continues on to a
glomerulonephritis with Ig deposits being readily detectable in kidneys
of both models of chronic GVHD and the sequential acute to chronic
GVHD, B6-into-CB6F1 (Fig. 9).
Previous reports have involved models of GVHD that can be manipulated
to manifest either signs of acute or chronic GVHD. In a bone marrow
transplant model of GVHD, lethally irradiated B6 recipients can develop
symptoms of acute or chronic GVHD depending on the number of LP/J cells
used to reconstitute the host B6 mouse (25). A second
model is the injection of C3H splenocytes into (B6 x
C3H)F1 mice. Cell doses of >2 x
107 but <1 x 108
cells result in chronic GVHD and the associated production of
autoantibodies, whereas injection of 
1 x
108 cells results in acute GVHD
(26). The B6-into-CB6F1 GVHD differs
from these models of GVHD because with the same cell dose
B6-into-CB6F1 mice exhibit acute and chronic GVHD
sequentially. In further support of a model of GVHD that induces
symptoms of both acute and chronic GVHD, a mixture of pathogenic and
nonpathogenic T cell clones isolated from cyclosporine-induced GVHD in
rats when injected into the footpad of naive rats can cause a localized
reaction that exhibits chronological pathological changes of acute and
chronic GVHD (27).
It should be noted that BALB/c-into-CB6F1 induced the expected chronic GVHD pattern that was indistinguishable from DBA/2-into-B6D2F1. Both of these chronic GVHD models, as well as B6-into-CB6F1 mice, resulted in production of serum Abs to ssDNA, dsDNA, and to ENA. Furthermore, renal Ig deposits were detected in all three of these GVHD combinations. Thus, a lupus-like autoimmune condition was observed in B6-into-CB6F1 GVHD, similar to the two H-2d-into-H2d/b F1 models. We cannot exclude the possibility that the acute B6-into-B6D2F1 GVHD would also have resulted in a similar long-term autoimmune disease pattern had they survived their acute disease.
This study demonstrates that similarities in H-2 and in donor
anti-host CTL precursor frequencies do not necessarily predict
whether parent-into-F1 GVHD will result in
chronic or acute disease. B6-into-B6D2F1 and
DBA/2-into-B6D2F1 models that elicit acute
(1, 2, 3, 4, 5, 6, 7, 8) and chronic GVHD (1, 4, 9, 10, 11, 12, 13, 14, 15, 16),
respectively, exhibit a 9-fold difference in donor anti-host CTL
precursor frequencies. The CTL comparisons also included DBA/2 and
BALB/c anti-H-2b responses on the B6
background. We observed a 9-fold higher frequency in the B6
anti-H-2d/b CTL response, irrespective of
whether the H-2d allele was provided by the DBA/2
(1/2,295) or BALB/c (1/2,370) than in the DBA/2 and BALB/c
anti-H-2d/b frequency (1/19,510 and 1/19,230,
respectively) (see Fig. 6). This 9-fold difference in CTL precursor
frequency was similar to the 9-fold difference reported earlier for
B6-anti-B6D2F1 vs DBA/2
anti-B6D2F1 (2). These results
indicate that the difference in B6-into-B6D2F1
acute GVHD and B6-into-CB6F1 GVHD cannot be
accounted for by parent anti-F1, H-2
allogeneic CTL precursor frequencies. Nevertheless, the differences
between acute GVHD and the two examples of
parent-into-F1 chronic GVHD
(anti-H-2b) are consistent with the 9-fold
difference in parent anti-F1 CTL precursor
frequencies.
When the precursor frequency experiments were extended to include
parent anti-F1 Th cell analysis, a >3-fold
difference was observed between B6 anti-DBA/2 and B6
anti-BALB/c (see Fig. 10). It is possible that the 3-fold higher Th
precursor frequency of B6 anti-DBA/2 provided an initial advantage
that permitted the acute disease to develop and be maintained long
enough to result in morbidity by 30 days. Multiple minor
histocompatibility differences exist between DBA/2 and BALB/c,
including differences in the expression of the Mls Ag.
Mls Ag are superantigens encoded by endogenous murine retroviruses that
stimulate a high proportion of T cells bearing a specific TCR V
family. DBA/2 cells (and F1s on this background)
bear Mls 1a, a strong Mls Ag that stimulates the
TCR V 6 and 8.1 and is not present on BALB/c cells (and
F1s on this background). The use of Mls
1a-bearing cells as stimulator cells in vitro
results in the production of a V 6+,
CD4+, Th1 cell responder population
(28). This was confirmed in vivo; furthermore, the
development of a maximal Th1 response was dependent on the presence of
endogenously produced IFN- (28). Also,
CD8+ T cells bearing these same TCR V families
can respond to Mls 1a by producing IFN-
(29). It is possible that the strong response of B6
anti-Mls 1a of the DBA/2 is sufficient to
initiate and sustain the strong cytokine response necessary for driving
CTL, whereas the weaker Mls Ag expressed by the BALB/c (Mls
1b) is adequate to initiate but not sustain the
CTL-effected acute disease. In further support of the possible role of
Mls Ag during acute GVHD, studies have shown that Mls
1a-reactive donor V 6 and 8.1 are expanded
during acute GVHD (30). Studies are in progress to
determine the possible role of Mls for the acute vs sequential
acute-to-chronic in vivo models of GVHD.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: GVHD, graft-vs-host disease; B6, C57BL/6; B6D2F1, (B6 x DBA/2)F1; CB6F1, (BALB/c x B6) F1; CM, complete medium; ENA, extractable nuclear Ag; DPBS, Dulbecco’s PBS.
Received for publication June 21, 2000. Accepted for publication August 24, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
during graft-versus-host disease. J. Exp. Med. 175:405. production. J. Immunol. 161:2848. secretion by specifically primed CD8+ cells but fails to trigger cytotoxicity. Eur. J. Immunol. 22:2789.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. Puliaev, I. Puliaeva, L. A. Welniak, A. E. Ryan, M. Haas, W. J. Murphy, and C. S. Via CTL-Promoting Effects of CD40 Stimulation Outweigh B Cell-Stimulatory Effects Resulting in B Cell Elimination and Disease Improvement in a Murine Model of Lupus J. Immunol., July 1, 2008; 181(1): 47 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Giorgini and A. Noble Blockade of chronic graft-versus-host disease by alloantigen-induced CD4+CD25+Foxp3+ regulatory T cells in nonlymphopenic hosts J. Leukoc. Biol., November 1, 2007; 82(5): 1053 - 1061. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Hurchla, J. R. Sedy, and K. M. Murphy Unexpected Role of B and T Lymphocyte Attenuator in Sustaining Cell Survival during Chronic Allostimulation J. Immunol., May 15, 2007; 178(10): 6073 - 6082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Zheng, J. H. Wang, M. N. Koss, F. Quismorio Jr., J. D. Gray, and D. A. Horwitz CD4+ and CD8+ Regulatory T Cells Generated Ex Vivo with IL-2 and TGF-{beta} Suppress a Stimulatory Graft-versus-Host Disease with a Lupus-Like Syndrome J. Immunol., February 1, 2004; 172(3): 1531 - 1539. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
