| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gene Knockout Donors1
Departments of
*
Pathology and
Immunology, University of Manitoba, Faculty of Medicine, Winnipeg, Manitoba, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
would influence the outcome of acute
graft-vs-host disease (GVHD). Graft-vs-host reactions were induced in
B6D2F1 hybrids using grafts from either IFN- gene
knockout (gko) or wild-type, C57BL/6J,
parental strain donors. GVHD was equally lethal in both groups, but
IFN- gko graft recipients developed a more protracted
form of the disease. These mice developed early wasting that persisted
until death. IFN- was present in spleen cell cultures from wild-type
graft recipients, but was absent in cultures from IFN-
gko graft recipients. Both recipient groups showed
macrophage priming for LPS-induced TNF-
release. Engraftment of
donor-derived CD4+ and CD8+ cells was greater
in IFN- gko graft recipients. Pathologic changes in
IFN- gko graft recipients were different from those
typically seen in acute GVHD. The syndrome developing in IFN-
gko recipients consisted of patchy alopecia, corneal
dryness and clouding, and lymphocytic infiltration of the liver,
pancreas, salivary gland, lung, and kidney. Lymphocytic infiltrates
were also present in the epidermis and the epithelium of both bile and
salivary gland ducts. Some of the lesions closely resembled those seen
in the "sicca"/Sjogren’s-like syndrome associated with chronic
GVHD; however, there was no evidence of immune complex deposition in
the kidney. These results indicate that GVHD in IFN- gko
graft recipients shares many features with acute GVHD, but both the
duration of the disease and its pathologic manifestations are
different. Our results suggest that IFN- plays a significant role in
the pathogenesis of acute GVHD by increasing the rate at which
mortality develops. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is thought to play a prominent role (10). Chronic GVHD has a
more indolent course, involves a wider range of organs, and has more
diverse pathologic manifestations. The clinical presentation can
resemble systemic lupus erythematosus, and scleroderma and is
characterized by the development of immune complex disease,
glomerulonephritis, and autoantibody formation. This form of GVHD may
in part be mediated by a Th2-type, humoral immune response
(11).
It has been suggested that the balance between Th1 and Th2 cytokines in
the initial stages of GVHD may be one factor determining whether the
disease follows an acute or a chronic course. Recent findings indicate
that all GVH reactions start out with the production of Th2 cytokines
and the activation of B cells. Early events that favor the development
of acute GVHD are engraftment of CD8+ cells and production
of IFN- by donor CD4+ cells. Otherwise, there is no
transition to acute GVHD, and the disease continues to evolve into the
chronic form (12). Some evidence suggests that donor-derived NK cells
may be instrumental in the development of acute GVHD, possibly by
producing IFN- early in the reaction, thereby promoting a Th1
response.4 Very recent data
indicate that early production of IL-12 may also be involved in this
process (13).
Exquisite sensitivity to endotoxin is a key feature of acute GVHD and
is central to understanding why it is almost always lethal. This
phenomenon was convincingly demonstrated in a study by Nestel et al.
(14), in which injections of endotoxin in doses insufficient to have a
discernible effect in normal mice were lethal in GVHD mice, killing
within 1 to 2 h. These injections caused very high levels of
TNF- to appear in the serum of GVHD mice. This effect was attributed
to priming of macrophages by Th1 cytokines, particularly IFN-. In
another study, Fowler et al. (15) showed that the sensitivity to
endotoxin in GVHD mice could be abrogated with the injection of
polarized Th2 cells, which resulted in inhibition of IFN-
production. While these investigations have indicated that IFN-
serves to promote acute GVHD, other studies have shown that exogenously
administered IFN- can mitigate some clinical manifestations and
reduce the mortality associated with the disease (16). Thus, the role
of IFN- in the pathogenesis of GVHD is still equivocal.
The purpose of our study was to explore the role of donor-derived
IFN- in GVHD by using IFN- gene knockout (IFN-
gko) donor mice in a parent
F1 hybrid
model and to determine how the use of these donors modified the course
and the outcome of the disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male and female C57BL/6J-Ifgtm1Ts
(IFN- gko) breeders were obtained from The Jackson
Laboratory (Bar Harbor, ME) and were used to found a breeding colony at
the University of Manitoba (Winnipeg, Canada). Offspring were used at
13 to 16 wk of age. These mice were housed in filter-topped, sterilized
cages and received sterilized food and water. Male and female C57BL/6J
(H-2b, hereafter referred to as wild-type) donors and
(C57BL/6J x DBA/2J)F1 hybrid recipients
(H-2b/d, abbreviated B6D2F1) were obtained
directly from The Jackson Laboratory.
Cell lines
The Moloney virus-induced lymphoma cell line YAC-1 (H-2a) was obtained from the American Type Culture Collection (Rockville, MD). The mouse T cell lymphoma BW1100 (H-2k, BW5147/M1100.129.237) was a gift from Dr. P. Marrack (Denver, CO). All cells were maintained in RPMI 1640 culture medium (Life Technologies, Grand Island, NY) containing glutamine (200 mM), sodium pyruvate (100 mM), and penicillin-streptomycin (5000 IU/ml to 5 mg/ml) and supplemented with 10% FCS.
Induction of GVHD
GVH reactions were induced in 13- to 16-wk-old
B6D2F1 hybrid recipients using either wild-type or
IFN- gko donors that were age and sex matched to
recipients. The method we used to induce GVHD has been described in
detail previously (17). Briefly, lymph nodes and spleens were harvested
from donors and then pooled and dissociated into a cell suspension by
pressing the organs through a stainless steel wire mesh. The cells were
washed, filtered through gauze, and adjusted to a final concentration
of 2 x 108 cells/ml of HBSS. Recipients were injected
via the tail vein with 60 x 106 cells suspended in
300 µl of HBSS.
Monitoring of mice with GVHD
We monitored GVHD in both experimental groups by daily
observation and periodic weighing. We also monitored splenomegaly in
mice sacrificed on several days postinduction. Spleen indexes (SI) were
calculated using the following formula:
 |
levels,
and splenic IFN- production. Mice from each group were also set
aside for monitoring mortality. Moribund animals in the agonal stages
of GVHD were sacrificed and autopsied. Tissue samples were collected
for histopathologic study by light and electron microscopy. Two groups
of control mice, one consisting of age- and sex-matched IFN-
gko donors and another of age- and sex-matched
B6D2F1 hybrid recipients were housed under the same
conditions as those mice undergoing GVH reactions. Detection of donor cells in recipient mice by flow cytometry
In a separate experiment we used flow cytometry to detect the
percentage of donor-derived cells in the spleens of recipient mice. We
also determined the proportion of these cells that expressed either CD4
or CD8. We used anti-H-2Dd to detect the MHC class
I of the opposite parent in the parentF1 hybrid strain
combination we used. Lymphocytes in the flow histograms that did not
express this marker were therefore deemed to be of donor origin. This
indirect method of detecting donor-derived, parental strain cells in
F1 hybrid hosts has been used by other investigators (12).
Recipients of grafts from both wild-type and IFN- gko
donors were assayed on days 4, 8, and 15. In the latter group, we also
performed an analysis on day 40. To demonstrate the specificity of
H-2Dd labeling, spleen cells from C57BL/6 and
B6D2F1 hybrid mice were analyzed.
Since we were interested primarily in the engraftment of T cells, spleen cell suspensions were passaged through nylon wool columns to remove adherent cells. Red cells were also removed by centrifugation on a Lympholyte-M gradient. Details of these methods were described previously (18). To perform flow cytometric analyses, cells were resuspended to a concentration of 5 x 106 cells/ml in PBS/1% BSA, and added to a V-bottom, 96-well, microtiter plate (Dynatech Laboratories, Chantilly, VA) at a volume of 100 µl/well. Plates were then centrifuged at 350 x g for 5 min at 4°C and resuspended in PBS/1%BSA containing 10 µg/ml of Ab. The following Abs were used: FITC-conjugated mouse anti-H-2Dd (34-5-8S, Cedarlane Laboratories, Hornby, Canada), PE-conjugated rat anti-CD4 (CT-CD4, Cedarlane), and PE-conjugated rat anti-CD8 (CT-CD8a, Cedarlane). FITC-conjugated mouse IgG2a (UPC-10, Caltag Laboratories, Burlingame, CA) and PE-conjugated rat IgG2a (LO-DNP-16, Caltag) were used as isotype controls. Coexpression of H-2Dd and CD4 or CD8 was determined by coincubating cells with 10 µg/ml of FITC-conjugated mouse anti-H-2Dd and 10 µg/ml of either PE-conjugated rat anti-CD4 or PE-conjugated rat anti-CD8. All incubations were performed for 30 min on ice, after which the cells were washed in PBS/1% BSA and resuspended in saline containing 2% paraformaldehyde. Two-color flow cytometric analyses were performed using an EPICS 753 fluorescence-activated cell sorter (Coulter, Hialeah, FL) with laser excitation set at 488 nm. Forward vs side light scatter histograms were used to define bitmap gates for single intact lymphocytes, with acquisition based on 6000 gated events. The FITC and PE fluorescence signals were split with a 550 dichroic filter and detected through 525- and 575-nm bandpass filters, respectively. Electronic compensation for spectral overlap was defined and verified with cell samples labeled separately with only FITC or PE. Data were collected in listmode format and analyzed using Coulter Elite software.
Determination of the percentage of CD4+ and
CD8+ T cells in grafts from wild-type and IFN-
gko grafts
Flow cytometry was used to verify that both grafts contained equal numbers of CD4+ and CD8+ cells. The protocols used to perform this analysis were identical with those described above, with the following exceptions: grafts consisted of pooled lymph node and spleen cells, and nylon wool purification was omitted.
Measurement of IFN- and IL-10 in spleen cell bulk cultures
Spleens were harvested asceptically in HBSS from recipients on
days 4, 8, and 10 postinduction. A sterile cell suspension was prepared
in 5% RPMI 1640 supplemented with HEPES (10 mM). The cell suspension
was adjusted to a concentration of 15 x 106
cells/ml and serially diluted to give final concentrations of 7.5,
3.75, and 1.875 x 106 cells/ml. Two milliliters of
suspension at each concentration was added to the wells of a 24-well
flat-bottom culture plate and incubated at 37°C in 5%
CO2. At 24, 48, and 72 h, 300 µl of supernatant was
removed from each well, frozen at -70°C, and later assayed for the
presence of IFN- by ELISA. The viability of the cultures over the
collection period was verified by daily inspection using phase contrast
microscopy. A sandwich ELISA using purified anti-IFN- mAb,
XMG1.2, and purified, biotinylated R4-6A2 (American Type Culture
Collection) in combination with streptavidin-alkaline phosphatase was
performed as described previously (19). Internal standards consisted of
IFN--containing Con A-stimulated mouse spleen cell supernatants
calibrated against World Health Organization-National Institute of
Allergy and Infectious Diseases international reference reagent
Gg02-901-533 (provided by Dr. C. Laughlin, National Institute of
Allergy and Infectious Diseases, National Institutes of Health).
Duplicate samples of supernatant were assayed using four twofold serial
dilutions. The lower limit of detection was 0.2 U/ml of IFN-, and
amounts were quantified at >0.5 U/ml in the linear portion of the
curve. The SE was <10%. Purified SXC1 and purified, biotinylated SXC2
were used in an ELISA to measure IL-10. Dr. T. Mosmann (University of
Alberta) initially provided the hybridomas. Each plate contained a
twofold serial dilution of standard rIL-10. The lower limit of
detection was 0.2 U/ml, and the amounts were quantified at >0.5 U/ml.
Again, the SE was <10%.
Measurement of TNF- in serum following injection of LPS
LPS-induced TNF- release was assayed in recipients of grafts
from either wild-type or IFN- gko donors on days 8 and 10
postinduction. Additional measurements were performed in IFN-
gko graft recipients on days 40 and 70, a time in the
reaction when all recipients of wild-type grafts had already succumbed
to GVHD. Controls consisted of untreated B6D2F1 hybrids.
LPS was prepared from a stock solution at a concentration of 1 mg/ml in
PBS. The stock was stored at -70°C and diluted 1/20 before
injection. Three recipients from each GVHD group and three normal
control B6D2F1 hybrids were injected i.v. with 10 µg of
LPS (Sigma, St. Louis, MO). Ninety minutes after injection the mice
were bled from the tail vein. Groups of three uninjected recipients
from each group were also bled at each time point to determine the
level of TNF- in the serum without LPS. The 90-min interval between
injection of LPS and collection of blood had been determined to be
optimal in pilot experiments in which normal mice were injected with
0.1 mg of polyinosinic:polycytidylic acid (poly I:C; ICN, Costa Mesa,
CA) i.p. to prime macrophages and then assayed for LPS-induced TNF-
release.
Blood samples were allowed to clot overnight at 4°C. The serum was
then collected and stored at -70°C. The ELISA used to measure
TNF- in the samples was performed using murine anti-TNF- mAb
(clone MP6-XT22; PharMingen, San Diego, CA) diluted to 4 µg/ml in 0.1
M NaC03, pH 8.2. After an overnight incubation with Ab, the
wells were blocked for 2 h using PBS/3% BSA. The assay was
standardized with recombinant murine TNF- (10 µg/ml; R&D Systems,
Minneapolis, MN) diluted to 4 ng/ml in PBS/3% BSA. Doubling dilutions
of serum samples and standard were performed in PBS/3% BSA starting at
1/1. Each well received 100 µl of standard, serum sample, or dilution
buffer. After an overnight incubation, the wells were washed and
incubated with 100 µl of biotinylated rabbit anti-mouse TNF-
polyclonal Ab (0.5 mg/ml; PharMingen) diluted to 4 µg/ml in PBS/3%
BSA. This was followed by a further incubation with avidin-peroxidase
(100 µl at 2 mg/ml; Sigma) diluted 1/2000 in PBS/3% BSA. Substrate
was prepared by dissolving ABTS (Sigma; 300 µg/ml) in 0.1 M citric
acid, pH 4.5, and adding 30% H2O2 at a
concentration of 0.9 µl/ml immediately before use. One hundred
microliters was then added to each well. After 30 min, ODs were read at
405 nm. The lower limit of detection was 60 pg/ml; measurements were
taken only from the linear portion of the curve. The SE was <10%.
NK cell assays
Splenic NK activity was measured in recipients of grafts from
either wild-type or IFN- gko donors on days 5, 8, and 10
postinduction using a 4-h 51Cr release assay as described
in detail previously (18). Suspensions of splenic effector cells were
incubated on nylon wool. Nonadherent cells were further purified by
density gradient centrifugation on Lympholyte-M (Cedarlane), washed,
adjusted to a concentration of 107 cells/ml, and serially
diluted 4 times to provide E:T cell ratios ranging from 100:1 to
12.5:1. YAC-1 and BW1100 target cells were labeled with
Na251CrO4 (Amersham, Oakville,
Canada) at a dose of 50 µCi/1 x 106 cells for 60
min, washed three times in 5% RPMI 1640, and resuspended to a final
concentration of 105 cells/ml. Cytotoxicity at each E:T
cell ratio was measured in triplicate cultures consisting of 100 µl
of effector cell suspension and 100 µl of target cell suspension
combined in wells of a plastic 96-well V-bottom microtiter plate.
Spontaneous release of 51Cr was measured in supernatants
from cultures containing of 100 µl of target cells and 100 µl of
medium without effector cells. Maximum 51Cr release was
measured in cultures containing 100 µl of target cells and 100 µl
of medium and no effector cells, and was collected by thoroughly mixing
the culture and removing 100 µl of the suspension. The plates were
incubated at 37°C for 4 h in 5% CO2. After
centrifugation, 100 µl of supernatant was harvested from each well
and counted for 2 min in an LKB gamma counter (LKB, Rockville, MD). The
percent lysis for each sample was calculated as follows:
 |
Donor/host origin of NK and NK-like cytotoxic activity
The rationale for the method used to detect the relative
contribution of donor and host cells to NK and NK-like cytotoxicity
observed in recipient mice is identical with that used in the
engraftment experiment described earlier. We used
anti-H-2Dd and complement to deplete host-derived
cytotoxic activity directed against YAC-1 and BW1100 target cells. The
activity remaining after depletion was deemed to be of donor origin.
Details of this method were provided previously (21). Because NK and
NK-like activities were maximal on different days in the two recipient
groups, we only assayed for the donor/host origin of this activity on
the days on which it was greatest. NK activity (YAC-1-directed lysis)
was maximal on day 4 in both recipient groups, whereas NK-like activity
(BW1100-directed lysis) was highest on day 8 in recipients of wild-type
grafts and on day 4 in recipients of IFN- gko grafts.
Purified effector cells (prepared as described in the previous section)
were incubated at a concentration of 2 x 107 cells/ml
for 1 h on ice with anti-H-2Dd (34-5-8S;
Cedarlane) diluted 1/40. This was followed by an incubation at a
concentration of 2 x 107 cells/ml for 1 h at
37°C with lyophilized Low-Tox-M Rabbit Complement (Cedarlane),
reconstituted in 1.0 ml of ddH2O, and diluted 1/9. Negative
and positive controls consisted of effector cells incubated with either
complement only or anti-ASGM1 rabbit antiserum (Wako
Chemicals, Dallas, TX) reconstituted in 1.0 ml of ddH2O and
diluted 1/100, and complement.
Histopathology
Mice were euthanized by CO2 asphyxiation. Samples of skin, lung, liver, spleen, lymph node, salivary gland, pancreas, and kidney were collected, fixed in 10% neutral buffered formalin for 24 h, machine processed through graded alcohol, and embedded in paraffin. Four-micron sections were cut and stained with hematoxylin and eosin. Samples of kidney were also taken for electron microscopy. This tissue was fixed in 2% buffered glutaraldehyde for 2 h, rinsed in phosphate buffer, postfixed in buffered osmic acid for 2 h and stained for 20 min in 2% aqueous uranyl acetate. After dehydration in graded ethanol the tissue was embedded in Spurr (J.B.EM Services, Dorval, Canada). Ultrathin sections were cut, stained with lead citrate for 5 min, and examined in a Philips EM 201 (Philips, Mahway, NJ).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Data comparing these indexes of GVHD are shown in Figures 1
and 2. Mortality was 100% in both
groups. As illustrated in Fig. 1A, most recipients of grafts
from wild-type donors died 15 to 20 days postinduction, and all had
succumbed by day 40. In contrast, the first recipient in the group
receiving grafts from IFN- gko donors died on day 4
postinduction. The remaining mice died between days 50 and 90. There
were no survivors beyond day 90. Weight loss data are shown in Figure 1B. Recipients of grafts from wild-type donors started to
lose weight on day 15. The most rapid reduction occurred between days
15 and 20, corresponding to the period of greatest mortality. IFN-
gko graft recipients experienced a transient episode of
rapid and severe weight loss early in the course of the disease,
between days 2 and 15. They then recovered, with the group mean
returning to preinduction levels by day 25. This was followed by a
second period of wasting that was slower and sustained over the
remaining course of the disease. Figure 2
shows that splenomegaly developed in both groups, but occurred earlier
in IFN- gko graft recipients.
|
|
We used flow cytometry to determine the percentage of cells in the
spleen that expressed the H-2Dd haplotype and were
therefore of host origin. Cells in the recipient that did not express
H-2Dd were considered to have come from the donor. Control
experiments showed that 100% of spleen cells from B6D2F1
hybrid mice expressed H-2Dd, whereas no
H-2Dd-positive cells were detected in either wild-type or
IFN- gko donors (data not shown). Figure 3 shows representative flow histograms
from one individual from each of the two groups of recipients on day 4
postinduction. Data comparing the rate at which donor-derived cells
from either wild-type or IFN- gko donors populated the
spleens of recipient mice are shown in Table I. Recipients of grafts from IFN-
gko donors showed a greater percentage of donor-derived
cells as early as day 4 postinduction. With the exception of day 4, the
number of donor-derived cells in the spleens of IFN- gko
graft recipients was more than twice that seen in recipients of
wild-type grafts. The percentage of donor-derived CD4+ and
CD8+ cells was also greater in recipients of grafts from
IFN- gko donors. In recipients of wild-type grafts, the
percentage of donor CD4 cells declined steadily from day 4 to day 15. A
similar decrease was also seen in IFN- gko graft
recipients from days 4 to 8. By day 15, however, the percentage of
CD4+ cells in this group began to increase. Both groups
showed a steady increase in the number of donor CD8+ cells,
but the rate at which the number of these cells increased was greater
in IFN- gko graft recipients. By day 15, the percentage
of donor CD8+ cells in IFN- gko graft
recipients was 5 times that seen in recipients of wild-type grafts.
|
|
gko grafts, we determined the percentages of these cells in
each of the grafts. Data shown in Table II indicate that the numbers of these
cells were similar in both groups, an observation consistent with that
reported previously by Dalton et al. (22).
|
and IL-10 production by spleen cell cultures
Table III shows the amount of
IFN- and IL-10 produced in spleen cell culture supernatants from GVH
mice that received grafts from either wild-type or IFN-
gko donors. The measurements were made on days 4, 8, and 15
postinduction. Culture supernatants derived from recipients of
wild-type grafts contained 21.6 U/ml of IFN- on day 8 postinduction.
No IFN- production was detected in cultures from IFN-
gko graft recipients on any of the days assayed, nor was any
seen in cultures from normal B6D2F1 hybrid controls. IL-10
production was observed in both groups of recipients on day 4 only. The
level seen in recipients of wild-type grafts was somewhat higher than
that in recipients of IFN- gko grafts. No IL-10
production was observed in controls.
|
following injection of endotoxin
Results are shown in Figure 4. Sera
from normal control mice did not contain any detectable TNF-, nor
did injection of these animals with 10 µg of endotoxin produce any
increase in serum levels. Sera from recipients of grafts from wild-type
donors showed no detectable TNF- on day 8 and slightly elevated
levels on day 10. After injection of LPS, there was a dramatic increase
in the amount of TNF- detected in the serum. Sera from uninjected
IFN- gko graft recipients contained no detectable amounts
of TNF- on either day 8 or day 10. The effect of injecting 10 µg
of LPS was similar to that seen in recipients of wild-type grafts, with
a very marked increase in the serum levels of TNF- on both days. The
levels observed in IFN- gko graft recipients were,
however, smaller than those in recipients of grafts from wild-type
donors. Recipients of grafts from IFN- gko donors showed
augmented LPS-induced TNF- release on day 40, but not on day 70,
postinduction.
|
These results are shown in Figure 5.
Cytotoxic activity directed at YAC-1 target cells was greater in
IFN- gko graft recipients. The differences were greatest
on days 4 and 8. The very slight difference seen on day 10 was not
considered significant. Lysis of BW1100 target cells, used as
definitive targets for NK-like activity, was present in both groups of
recipients. It was seen on day 4 in IFN- gko graft
recipients and on day 8 in wild-type C57BL/6J graft recipients. Results
from experiments to determine the relative contributions of host and
donor cells to the NK and NK-like activities are shown in Table IV. Approximately half of the day 4
YAC-1-directed lysis in wild-type graft recipients was donor in origin,
whereas almost 70% of the NK-like activity was donor derived. In
IFN- gko graft recipients, one-third of the day 4
YAC-directed lysis and approximately 25% of the BW1100-directed
cytotoxicity could be attributed to donor cells. The data indicate a
comparatively smaller contribution to both NK and NK-like activity by
donor-derived cells in these recipients.
|
|
gko donors
Beginning about day 50 postinduction, IFN- gko graft
recipients developed a syndrome characterized by patchy allopecia on
the head and neck, with focal excoriation and ulceration of the skin
around the snout, ears, and back of the neck. Several of the animals
developed eye lesions consisting of retraction of the eyelids,
protrusion of the eyeball from the orbit (exopthalmos), and clouding
and desiccation of the cornea and conjunctiva. Some of these features
are shown in Figure 6. When these animals
were autopsied, gross examination of their internal organs demonstrated
marked splenomegaly (SI of 3.5–4.5). Two of the animals autopsied
showed unilateral hydronephrosis with obstruction in the lower
one-third of the ureter. There was enlargement and pallor of the liver.
Microscopic examination of the spleen revealed marked lymphoid
hyperplasia. The liver showed expansion of the portal tracts by a
cellular infiltrate consisting mostly of lymphocytes (Fig. 7A). In two animals
neutrophils and eosinophils could be identified in the infiltrates.
These lesions were often very large but were confined mainly to the
portal areas. Occasionally they extended through the limiting plate
into the lobules. No hepatocellular necrosis was observed.
Intracanalicular bile stasis was present. In many portal areas bile
ducts could not be identified. In many intact bile ducts we observed
lymphocytic infiltration and disruption of the epithelium (Fig. 7, B and C).
|
|
D) and consisted mostly of lymphocytes. In some
animals, neutrophils could also be identified in the infiltrates. In
some salivary glands, invasion of the duct epithelium by lymphocytes
was conspicuous. Infiltrates were also present in the lung, where they
were found primarily around bronchi and pulmonary blood vessels. In sections of skin we observed ulceration associated with chronic inflammation and a granulation tissue reaction in the ulcer beds. Sampling away from the ulcerated areas revealed lymphocytic infiltration of the dermis. The epidermis showed edema and mononuclear cell infiltration. Dyskeratotic epidermal cells were occasionally observed. Lymphocytic infiltration was also present in the epithelium surrounding the hair follicles.
Sections of kidney demonstrated focal lymphocytic infiltrates in the
interstitium of the cortex and medulla. The glomeruli appeared normal
by light microscopy. Electron microscopic examination of the glomeruli
showed no evidence of immune complex deposition in the glomerular
basement membrane (Fig. 8).
|
gko donor mice housed under the same conditions as
IFN- gko graft recipients did not develop any overt signs
of disease over the course of the experiment. Autopsies of these
animals revealed no histopathologic abnormalities. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gko donors. The C57BL/6J-Ifgtm1Ts
mutant used in these experiments was developed by Dalton et al. (22).
These mice thrive if housed in a clean environment, but show a variety
of immune defects, including impaired production of antimicrobicidal
products, decreased expression of MHC class II Ags by macrophages, and
an increased susceptibility to intracellular pathogens such as
Mycobacterium bovis. They also demonstrate uncontrolled
proliferation of splenocytes in response to both mitogen and
alloantigen as well as increased T cell cytolytic activity against
allogeneic target cells in mixed lymphocyte reactions (22).
In our experiments IFN- gko graft recipients showed a
pattern of mortality and weight loss different from that seen in
recipients of grafts from wild-type donors. Although GVHD in both
groups was equally lethal, the course was more prolonged in IFN-
gko graft recipients, with the time to 100% mortality being
more than twice that seen in recipients of wild-type grafts. Both
groups developed weight loss early in the disease, but this began
earlier in IFN- gko recipients. Whereas cachexia in
wild-type graft recipients was unremitting, IFN- gko
graft recipients recovered from their initial episode of cachexia and
then experienced a second round of wasting later in the disease.
Because TNF- is allegedly involved in the GVHD-associated cachexia,
we measured serum TNF- levels. A small amount of TNF- was
observed in the serum of wild-type graft recipients on day 10
postinduction in the absence of LPS injection, whereas none was seen in
recipients of grafts from IFN- gko donors on either day 8
or 10 postinduction. This might suggest that factors other than TNF-
are involved in producing weight loss in GVH mice. It would not be
surprising if anorexia in these severely ill mice were a major
factor.
Our flow cytometric data demonstrated some interesting differences in
the pattern of T cell engraftment of the two recipient groups. The
number of donor-derived, nonadherent, spleen cells in recipients of
wild-type grafts never exceeded 20%. The percentage of
CD4+ cells was 13% on day 4, but gradually declined to 3%
on day 15. CD8+ cells showed a modest increase from 2 to
6% over the same interval. The findings were very different in
recipients of IFN- gko grafts. Even in the very early
stages of the disease, the level of engraftment was much greater
(33% on day 4) and increased steadily to 77% by day 40. The
percentage of CD4+ cells on day 4 was also considerably
greater (23%). This value declined slightly, but by day 40 had
returned to 20%. The number of CD8+ cells increased
steadily to reach 31% by day 40. These engraftment data are to some
extent mirrored by the changes in the size of the spleen. Whereas
GVH-induced splenomegaly in wild-type graft recipients was a transient
phenomenon, starting early in the course of GVHD and then subsiding,
IFN- gko graft recipients developed splenomegaly that
persisted to the end of the disease. We know from the work of Dalton et
al. (22) that splenocytes from IFN- gko mice develop an
exaggerated proliferative response to allogeneic stimulator cells in
MLR. This may explain why we observed both the greater accumulation of
donor-derived T cells and the persistent splenomegaly in IFN-
gko graft recipients. It also provides a cogent explanation
for the very pronounced lymphocytic infiltrates in target organs of
recipients of IFN- gko grafts.
The augmentation of LPS-induced TNF- release in IFN-
gko graft recipients was unexpected. This experiment was
modeled after that of Nestel et al. (14), who showed that mice with
acute GVHD secrete very large amounts of TNF- into the serum in
response to injections of endotoxin in doses insufficient to cause any
increase in serum levels in normal control animals. They attributed
this to macrophage priming by IFN- released from donor-derived T and
NK cells in response to alloantigen. We predicted that if donor-derived
IFN- was responsible for this effect, augmented LPS-induced TNF-
release should not be observed in IFN- gko graft
recipients. Our results, however, showed the opposite, challenging the
hypothesis that IFN- is solely responsible for macrophage priming in
acute GVHD. We also found that despite the absence of IFN-,
augmented LPS-induced TNF- release was still present in IFN-
gko graft recipients on day 40 of the reaction. To make
certain that no IFN- was produced in IFN- gko graft
recipients, we compared the level of this cytokine in spleen cell
cultures from both experimental groups on several days postinduction.
As expected, cultures from IFN- gko graft recipients
contained no measurable IFN-, even in the very early stages of GVHD
(day 4). IFN- production was observed on day 8 in recipients of
wild-type grafts. The absence of IFN- production in recipients of
IFN- gko grafts demonstrates that the IFN- produced in
recipients of wild-type grafts is derived entirely from the graft and
not the host. These findings suggest that a cytokine other than IFN-
is involved in the macrophage priming effect in GVHD. Whether this
cytokine supplants IFN- in IFN- gko mice or whether it
is indeed the real priming factor remains unknown. Experiments to
address this matter are in progress.
Conventional NK cell activity, as measured by YAC-1-directed lysis,
appeared in both recipient groups, but was higher in IFN-
gko graft recipients. We found that there was a
comparatively smaller contribution to conventional NK activity by
donor-derived cells in recipients of grafts from IFN- gko
donors. The reason for this disparity is not known. Dalton et al. (22)
found that resting NK activity (YAC-1-directed lysis) is significantly
lower in IFN- gko mice than it is in wild-type mice. This
finding may in part explain the considerably smaller contribution of
donor-derived cells to the YAC-1-directed cytotoxic activity in
recipients of IFN- gko grafts. It does not, however,
explain why the overall level of conventional NK activity was higher in
these recipients. It is possible that other NK-activating cytokines may
be secreted in greater quantity during GVH reactions in which IFN-
is absent, thereby increasing the overall level of NK activity.
Cytotoxicity directed at BW1100 targets was also present in both groups
of recipients. It was slightly higher in those that received wild-type
grafts, but it appeared earlier in IFN- gko graft
recipients. We know from previous studies that lysis of BW1100 target
cells is mediated by a population of cells functionally and
phenotypically distinct from conventional NK cells (18). These cells,
referred to as NK-like, are
CD3+/CD4-/CD8-/NK1.1+/ASGM1+
and have the ability to kill a range of tumor target cells (e.g.,
BW1100 and P815) that are normally insensitive to NK lysis. Spontaneous
NK-like activity is not detectable in the spleen and lymph nodes, but
appears after stimulation with the IFN inducers such as poly I:C (23).
GVH-induced NK-like activity is mediated in part by
CD3+/CD4-/CD8-/NK1.1+/ASGM1+
TCR+ cells (24). NK-like activity appears as a
transient response early in the course of acute GVHD in mice, but does
not occur in mice with chronic GVHD (22). Our present data reconfirm
our previously published observation that most NK-like activity in
recipients of wild-type grafts is donor derived (21). The role of
NK-like cells in the pathogenesis of GVHD is unclear, but it has been
suggested they may be involved in mediating tissue injury. Although the
level of NK-like activity was somewhat less in recipients of grafts
from IFN- gko donors, the fact that NK-like activity was
present in these recipients suggests that the activation of the cells
does not necessarily depend on IFN-. This conclusion is supported by
findings in another study showing that activation of NK-like cells in
mice with GVHD is coextensive with the production of IFN-ß, rather
than IFN- (18). It is interesting that donor-derived cells
contributed approximately 60% less to this activity than they did in
recipients of wild-type grafts. We postulate that, as with conventional
NK activity, this may also be due to lower levels of NK-like cytotoxic
activity in IFN- gko donor mice. Our own experiments
using poly I:C to induce NK-like activity in IFN- gko
donors indicate that this activity is approximately one-third less than
that seen in wild-type mice (C. A. Ellison and J. G. Gartner,
unpublished observation).
The pathology of GVHD in IFN--gko graft recipients was
different in both the range of organ involvement and the size of the
lesions from that in wild-type graft recipients. The most striking
findings were the involvement of skin and eye in IFN- gko
graft recipients as well as the prominent cellular infiltrates in lung,
salivary gland, and pancreas. Another notable feature was the presence
of neutrophils and eosinophils in the cellular infiltrates in liver and
salivary gland in some IFN- gko recipients. These cells
are not usually seen in GVH-induced cellular infiltrates. Whether the
presence of neutrophils reflects a reaction to the tissue damage, a
superimposed bacterial infection, or an idiosyncrasy of GVH lesions in
the absence of IFN- is not known.
Some of the histopathologic changes we observed in IFN-
gko graft recipients are similar to those seen in BMT
recipients with chronic GVHD. This is best exemplified by the
lymphocytic infiltrates we observed in the salivary gland ducts. In
chronic GVHD, destruction of the excretory ducts in the salivary and
lacrimal glands by infiltrating lymphocytes causes a sicca syndrome,
resembling that seen in Sjogren’s disease, to develop. Although we did
not examine the lacrimal glands, it is quite possible that involvement
of these organs may have been instrumental in causing the eye lesions
we observed in IFN- gko graft recipients.
The development of an autoimmune/systemic lupus erythematosus-like
syndrome, characterized by autoantibody formation and immune complex
disease, is also a feature of chronic GVHD (25). However, as our
electron microscopic findings in the kidney demonstrated, IFN-
gko graft recipients did not develop immune complex disease
and glomerulonephritis as part of their syndrome. Rus et al. (12) have
attributed the development of immune complex disease in chronic GVHD to
B cell hyperplasia induced by Th2 cytokines. They have suggested that
in GVH reactions destined to produce acute GVHD, donor-derived CD8
cells stem the development of the autoimmune syndrome by eliminating
activated B cells. Dalton et al. (22) found that splenocytes from
IFN- gko mice develop very high levels of CTL activity
against allogeneic target cells in MLR. In our flow cytometry
experiments, we also observed a very large percentage of donor-derived
CD8 cells in recipients of grafts from IFN- gko donors.
Considering Dalton’s MLR findings, this is not surprising. Since Rus
et al. suggest that CD8 cells prevent the development of Th2-mediated
chronic GVHD, the increase in CD8 cells we observed in IFN-
gko graft recipients might explain why these animals did not
develop immune complex disease as part of their syndrome. The fact that
IL-10 levels were much lower than those in wild-type graft recipients
provides further evidence that these mice did not develop a Th2
response in the absence of IFN-.
In summary, the lethal GVHD developing in recipients of grafts from
IFN- gko donors has many features in common with GVHD
seen in recipients of wild-type grafts, but there are also several
significant differences. The course of GVHD in IFN- gko
graft recipients is more prolonged, and there is a greater range of
organ involvement. The lesions that develop are more extensive and are
associated with lymphocytic infiltration of epithelium. Many of the
observed changes are similar to those seen in patients with chronic
GVHD. However, the absence of immune complex disease in the kidney
suggests that a Th2-mediated lupus-like syndrome does not develop.
Paradoxically, the use of IFN- gko donors does not
prevent macrophage priming for LPS-induced TNF- release, a
phenomenon thought to be involved in the pathogenesis of the rapidly
progressive form of septic shock that often complicates acute GVHD.
Overall, our results support the idea that donor-derived IFN- does
play an important role in the pathogenesis of acute GVHD. The
enthusiasm with which we draw this conclusion is to some degree
tempered by the caution that should be observed when using gene
knockout models. The absence of the gene in question during ontogeny
can affect the development or function of other components of the
immune system. Thus, an effect observed when a gene knockout is used
may not necessarily be due to the absence of the gene or gene product
per se. This said, our findings with IFN- gko donors
suggest that one of the major effects of IFN- is to cause an
acceleration of GVH-induced mortality. Our findings unfortunately do
not disclose exactly where this cytokine acts in the mechanism of the
disease.
|
Acknowledgments |
|---|
ELISA assays, and Charmaine Hedgecock for
preparation of specimens for electon microscopy. We are also grateful
for the assistance provided by Dr. Edward Rector. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. John G. Gartner, Department of Pathology, University of Manitoba, 770 Bannatyne Ave., Winnipeg, Manitoba, Canada R3E 0W3.
3 Abbreviations used in this paper: GVHD, graft-vs-host disease; BMT, bone marrow transplantation; gko, gene knockout; IFN- gko graft recipients, recipients of grafts from C57BL/6J-Ifgtm1Ts donors; C57BL/6J-Ifgtm1Ts, IFN- gene knockout donors; BW1100, BW5147/M1100.129.237; SI, spleen index; PE, phycoerythrin; poly I:C, polyinosinic:polycytidylic acid; LU10/107, lytic units per 107 effector cells.
4 C. A. Ellison, K. T. HayGlass, J. M. M. Fischer, G. C. MacDonald, and J. G. Gartner. 1998. Depletion of NK1.1+ cells from the graft reduces IFN- levels and LPS-induced TNF- release in F1-hybrid mice with acute graft-vs.-host disease. Transplantation. In press.
Received for publication September 9, 1997. Accepted for publication March 16, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
during graft-versus-host disease. J. Exp. Med. 175:405. T cells in the pathobiology of murine acute graft- versus-host disease: evidence that T cells mediate natural killer-like cytotoxicity in the host and that elimination of these cells from donors significantly reduces mortality. J. Immunol. 155:4189.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  G. Cui, X. Qin, Y. Zhang, Z. Gong, B. Ge, and Y. Q. Zang Berberine Differentially Modulates the Activities of ERK, p38 MAPK, and JNK to Suppress Th17 and Th1 T Cell Differentiation in Type 1 Diabetic Mice J. Biol. Chem., October 9, 2009; 284(41): 28420 - 28429. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Wang, W. Asavaroengchai, B. Yong Yeap, M.-G. Wang, S. Wang, M. Sykes, and Y.-G. Yang Paradoxical effects of IFN-{gamma} in graft-versus-host disease reflect promotion of lymphohematopoietic graft-versus-host reactions and inhibition of epithelial tissue injury Blood, April 9, 2009; 113(15): 3612 - 3619. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Burman, T. Banovic, R. D. Kuns, A. D. Clouston, A. C. Stanley, E. S. Morris, V. Rowe, H. Bofinger, R. Skoczylas, N. Raffelt, et al. IFN{gamma} differentially controls the development of idiopathic pneumonia syndrome and GVHD of the gastrointestinal tract Blood, August 1, 2007; 110(3): 1064 - 1072. [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]
|
|
|
|
|
|
M. M. Hauri-Hohl, M. P. Keller, J. Gill, K. Hafen, E. Pachlatko, T. Boulay, A. Peter, G. A. Hollander, and W. Krenger Donor T-cell alloreactivity against host thymic epithelium limits T-cell development after bone marrow transplantation Blood, May 1, 2007; 109(9): 4080 - 4088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. W. Blaser, N. R. Schwind, S. Karol, D. Chang, S. Shin, S. Roychowdhury, B. Becknell, A. K. Ferketich, D. F. Kusewitt, B. R. Blazar, et al. Trans-presentation of donor-derived interleukin 15 is necessary for the rapid onset of acute graft-versus-host disease but not for graft-versus-tumor activity Blood, October 1, 2006; 108(7): 2463 - 2469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Komiyama, S. Nakae, T. Matsuki, A. Nambu, H. Ishigame, S. Kakuta, K. Sudo, and Y. Iwakura IL-17 Plays an Important Role in the Development of Experimental Autoimmune Encephalomyelitis J. Immunol., July 1, 2006; 177(1): 566 - 573. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Porcellini, E. Traggiai, U. Schenk, D. Ferrera, M. Matteoli, A. Lanzavecchia, M. Michalak, and F. Grassi Regulation of peripheral T cell activation by calreticulin. J. Exp. Med., February 20, 2006; 203(2): 461 - 471. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Roychowdhury, B. W. Blaser, A. G. Freud, K. Katz, D. Bhatt, A. K. Ferketich, V. Bergdall, D. Kusewitt, R. A. Baiocchi, and M. A. Caligiuri IL-15 but not IL-2 rapidly induces lethal xenogeneic graft-versus-host disease Blood, October 1, 2005; 106(7): 2433 - 2435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Varona, V. Cadenas, L. Gomez, C. Martinez-A, and G. Marquez CCR6 regulates CD4+ T-cell-mediated acute graft-versus-host disease responses Blood, July 1, 2005; 106(1): 18 - 26. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Puliaev, P. Nguyen, F. D. Finkelman, and C. S. Via Differential Requirement for IFN-{gamma} in CTL Maturation in Acute Murine Graft-versus-Host Disease J. Immunol., July 15, 2004; 173(2): 910 - 919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ichiba, T. Teshima, R. Kuick, D. E. Misek, C. Liu, Y. Takada, Y. Maeda, P. Reddy, D. L. Williams, S. M. Hanash, et al. Early changes in gene expression profiles of hepatic GVHD uncovered by oligonucleotide microarrays Blood, July 15, 2003; 102(2): 763 - 771. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Reddy, T. Teshima, G. Hildebrandt, D. L. Williams, C. Liu, K. R. Cooke, and J. L.M. Ferrara Pretreatment of donors with interleukin-18 attenuates acute graft-versus-host disease via STAT6 and preserves graft-versus-leukemia effects Blood, April 1, 2003; 101(7): 2877 - 2885. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-G. Yang, J. Qi, M.-G. Wang, and M. Sykes Donor-derived interferon gamma separates graft-versus-leukemia effects and graft-versus-host disease induced by donor CD8 T cells Blood, May 13, 2002; 99(11): 4207 - 4215. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Via, A. Shustov, V. Rus, T. Lang, P. Nguyen, and F. D. Finkelman In Vivo Neutralization of TNF-{alpha} Promotes Humoral Autoimmunity by Preventing the Induction of CTL J. Immunol., December 15, 2001; 167(12): 6821 - 6826. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Liu, B. E. Anderson, M. E. Robert, J. M. McNiff, S. G. Emerson, W. D. Shlomchik, and M. J. Shlomchik Selective T-cell subset ablation demonstrates a role for T1 and T2 cells in ongoing acute graft-versus-host disease: a model system for the reversal of disease Blood, December 1, 2001; 98(12): 3367 - 3375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Snider and H. Liang Early Intestinal Th1 Inflammation and Mucosal T Cell Recruitment During Acute Graft-Versus-Host Reaction J. Immunol., May 15, 2001; 166(10): 5991 - 5999. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. P. Nestel, R. N. Greene, K. Kichian, P. Ponka, and W. S. Lapp Activation of macrophage cytostatic effector mechanisms during acute graft-versus-host disease: release of intracellular iron and nitric oxide-mediated cytostasis Blood, September 1, 2000; 96(5): 1836 - 1843. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Miura, D. Mizuki, S. Sasaki, S. Hasegawa, H. Sashinami, and A. Nakane Host Resistance to Listeria monocytogenes Infection Is Enhanced but Resistance to Staphylococcus aureus Infection Is Reduced in Acute Graft-versus-Host Disease in Mice Infect. Immun., July 1, 2000; 68(7): 4340 - 4343. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Krenger, S. Rossi, L. Piali, and G. A. Hollander Thymic atrophy in murine acute graft-versus-host disease is effected by impaired cell cycle progression of host pro-T and pre-T cells Blood, July 1, 2000; 96(1): 347 - 354. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Sloand, S. Kim, J. P. Maciejewski, F. Van Rhee, A. Chaudhuri, J. Barrett, and N. S. Young Pharmacologic doses of granulocyte colony-stimulating factor affect cytokine production by lymphocytes in vitro and in vivo Blood, April 1, 2000; 95(7): 2269 - 2274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Zeng, D. Lewis, S. Dejbakhsh-Jones, F. Lan, M. Garcia-Ojeda, R. Sibley, and S. Strober Bone Marrow NK1.1- and NK1.1+ T Cells Reciprocally Regulate Acute Graft versus Host Disease J. Exp. Med., April 5, 1999; 189(7): 1073 - 1081. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or IFN-
) donors into B6D2F1 hybrid recipients.
Error bars indicate the SE of the mean body weight in the two treatment
groups on each day. We included the last known weight of any mice dying
from the effects of GVHD (phantom weights) along with the weights of
survivors in our calculation of the mean for each day.
