Download Inhibition of Nitric Oxide Production Is Associated

From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Inhibition of Nitric Oxide Production Is Associated With Enhanced Weight
Loss, Decreased Survival, and Impaired Alloengraftment in Mice
Undergoing Graft-Versus-Host Disease After Bone Marrow Transplantation
By William R. Drobyski, Carolyn A. Keever, Gerald A. Hanson, Timothy McAuliffe, and Owen W. Griffith
The pathophysiologic role of nitric oxide (NO) in graft-versus-host disease (GVHD) was investigated in a murine bone
marrow (BM) transplantation modelwhere donor and recipient were H-2-matched but differed at multiple minor histocompatibility antigens. Host AKR/J (H-2K)mice received lethaltotalbodyirradiation
as pretransplant conditioning
followed by transplantationof donor B1O.BR (H-2K)BM cells
with or without spleen cells as a source of GVH-reactive T
cells. NO production, as assessed by serum nitrate and nitrite levels, was increased for up t o 3 weeks posttransplant
in animals undergoing both moderate and severe GVHD.
Administration of NO-methyl-L-arginine (L-NMA), an inhibitor of nitric oxide synthase, t o animals undergoing GVHD
resulted in effective suppression of NO production when
compared with saline-treated GVHD control animals. Suppression of NO production by L-NMA in GVHD animals was
associated with enhanced weight loss early posttransplant
and decreased overall survival. Histologic analysis of tissues
from L-NMA-treated and saline-treated GVHD animals
showed thatearly weight loss was notbecause of anexacer-
bation of GVHD, indicating that NO did not appear t o play
an immunosuppressive role in this experimental model. LNMA-treated animals with enhanced weight loss were observed t o have splenic atrophy, decreased extramedullary
hematopoiesis, and a reduction in BM cellularity when compared with GVHD control mice that were weight-matched
before transplant.Analysis of T-cell chimerism in the spleen
showed thatL-NMA treatment impaired donorT-cell repopdation. In vitro colony-forming unit (CFU) assays were performed t o further assess the role of NO on BM progenitor
cell growth. L-NMA added directly into culture had no
effect
on CFU-granulocyWmacrophage (CFU-GM) formation in
normal murineBM. In contrast, total CFU-GM from L-NMAtreated animals were significantly reduced when compared
with GVHD controls or BM control
animals who did notdevelop GVHD. Collectively, these data indicatethat inhibition
of NO impairs hematopoietic reconstitution and supportthe
premise that NO appears to play a novel role in thefacilitation of alloengraftment posttransplant.
0 1994 by The American Societyof Hematology.
RAFT-VERSUS-HOST disease (GVHD) is a major
cause of morbidity and mortality in patients undergoing allogeneic bone marrow transplantation (BMT). GVHD
is a complex pathophysiologic process that results from the
cooperative interaction of multiple effector cell populations
resident in the donor graft and persistent in the host after
conditioning.’ Although still incompletely understood, several fundamental tenets of GVH reactivity have been established using experimental animal models and clinical studies
in humans. A central requirement for donor-derived T cells
in the pathogenesis of GVHD has been substantiated by
studies in which removal of T cells from the donor BM
inoculum has resulted in the amelioration or complete prevention of the GVHD ~yndrome.’.~
Similarly, immunosuppressive agents designed to inhibit T-cell function and proliferation have been shown to mitigate GVHD.4 However,
while these studies indicate that T cells appear to be necessary for the induction of GVH reactivity, accumulating data
indicate that T cells only partially account for the pathophysiologic manifestations of GVHD.
Recent data support the premise that additional elements
of the immune network play an important role in GVHD.
Secondary immune effector cell populations that are recruited by T cells appear to serve as more proximate mediators of GVHD-induced tissue damage in certain experimental
settings. This hypothesis is supported by studies identifying
natural killer cells histopathologically in areas of tissue damage5 and correlating natural killer cell activity with GVHD
severity.6 Additionally, the pathologic discordance between
the degree of lymphocytic infiltration and observed tissue
damage has suggested that inflammatory mediators may also
play a role in GVHD. Emerging data indicate that dysregulation of cytokine production is critical in the pathogenesis of
GVHD andmay serve to amplify the immune response.
Specific cytokines such as interleukin-l (IL-l),7tumor necrosis factor (TNF),8 and y-interferon (Y-IFN)~
have all been
postulated to be mediators of the efferent phase of GVHD.
Each of these cytokines has a diversity of immunomodulatory functions, and the mechanism by which they contribute
to GVH reactivity has not been completely elucidated. Proposed common pathways by which cytokines might enhance
GVHD have included the upregulation of class I1 antigen
expression, the activation of additional effector cell populations, and the direct mediation of tissue damage. An alternative pathway by which cytokines might contribute to GVH
reactivity is via the induction of a common proximate mediator such as nitric oxide (NO).
NO is an unstable radical that has been shown to play an
important role in immune modulation,” neurotransmission,”
and regulation of vascular tone.” NO is generated from Larginine through an oxidation reaction that is catalyzed by
NO synthase (NOS). NO then decays in vivo into the stable
inorganic nitrogen oxides, nitrate and nitrite. NOS exists in
both a constitutive form present in vascular endothelial cells,
neurons, and platelet^'"'^ and in an inducible form that can
be present in macrophages, neutrophils, endothelial cells,
and hepatocyte^.'^"^ Expression of inducible NOS can be
G
Blood, Vol 84,
No 7
(October I), 1994:pp 2363-2373
From the Departments of Medicine, Pathology, Biostatistics and
Biochemistry, and the Bone Marrow Transplant Program, Medical
College of Wisconsin, Milwaukee, WI.
Submitted April 11, 1994; accepted June 9, 1994.
W.R.D. is supported by Grant No. CA01534 from the National
Cancer Institute.
Address reprint requests to William R. Drobyski,MD, Medical
College of Wisconsin, 8700 WWisconsin Ave, Milwaukee, W153226.
The publication costsof this article were defrayedin part by page
chargepayment. This article must therefore behereby marked
“advertisement” in accordance wirh 18 U.S.C. section 1734 solely to
indicate this fact.
0 1994 by The American Society of Hematology.
0006-4971/94/8407-0029$3.00/0
2363
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2364
DROBYSKI ET AL
Thy 1.2-/L3T4' and Thy
I .2 -/Lyt 2 '. Fluorescence activated cell
sorter (FACS) analysis were performed on spleen cell preparations
from chimeras at selected time points posttransplant. Erythrocytes
were lysed using hypotonic distilled water. Cells were stained with
monoclonalantibodies as recommended by themanufacturerfor
two-color analysis. Cells were analyzed on a FACS Analyzer (Bec30
ton Dickinson)equippedwithaFACSLitelaserandConsort
computer support.
Hisrologicsfudies. Tissues were obtained from control and experimentalanimals, fixed in 10% neutral bufferedformalin,and
processed into paraffin blocks. Four-micron-thick sections were prepared fromeachblockandwere
cut atthree levels to optimize
sampling.
For the evaluation of GVHD, tissue sections were screened with
theexaminer blinded to the treatment received by eachanimal.
Preliminary studies had indicated that little histologic evidence o f
GVHD was observed in theskin or liverearlyposttransplant
in
animals receiving S X IOh spleen cells (GVHD control mice). Therefore, to facilitate a semiquantitative assessment of GVHD in control
MATERIALS ANDMETHODS
and experimental animals, gradingof GVHD was confined to analysis of the colon.
Mice. AKWJ(H-2L,MW,Thy
1.1+) andBlO.BR/SgSn (H-2',
Samples obtained from the colon of representative animals were
Mlsh, Thy 1.2+) mice were purchased from Jackson laboratories (Bar
evaluated in both axial and longitudinal sections. Glands were seHarbor. ME). All mice were housed in the American Association
lected that had the basal portion visibly resting in proximity to the
for Laboratory Animal Care-accredited Animal Resource Center of
muscularis mucosa. The integrity of the glandular epithelium and
the Medical College of Wisconsin (Milwaukee, WI). Mice received
the degree of cellular infiltration in the lamina propria were evaluregular mouse chow and acidified tap water ad libitum.
ated. The degree of changes in the epithelium was graded on a scale
Reagents. L-NMA
was
prepared
from
L-ornithine
(Sigma
of 1 to 4 with 1 being no abnormality, 2 being mild change consisting
N,S-dimethylthiopseuduronium
Chemical CO, StLouis,MO)and
ofoccasionalnecrobioticcellswithoutsignificantglandinjury,
3
iodide (Aldrich Chemical CO, Milwaukee, WI) by adaptation of the
beingmoderatechangewith
several glandshavingoneor
more
procedure of Corbin and Reporter.22 The
flavinate salt was convened
necrobiotic cells without gland injury, and 4 being severe change
to theacetatesalt by stirringwithDowex-l(hydroxide;BioRad,
with several glands having injury and/or destruction. Inflammatory
Hercules, CA) and by adjusting the resulting basic solution to pH 7
cell infiltrates in the lamina propria were graded on a scale of 0 to
with acetic acid. N"-methyl-D-arginine (D-NMA) was synthesized
4 reflecting no, mild, moderate,or marked infiltration. GVHD scores
in an identical fashion with the exception that D-ornithine (Sigma)
were normalized to that of BM controls to account for the effects
was used in place of L-ornithine. Both L-NMA and N"-methyl-Dof the conditioning regimen. The scores for epithelial integrity and
arginine (D-NMA) were dissolved in sterile distilled water andused
degree of infiltration in the lamina propria were added
to yield a
at a concentration of 35 mg/mL.
total GVHD score for individual mice.
Nitrate and nitrite levels were determined by the Griess reaction
Cohzy-jimtting unit (CFU)as.su.y.s. Femurs were harvested from
as described by Green et al.'? Nitrate was reduced to nitrite using a
mice,and BM was flushed fromthe BM cavity with Dulbecco's
copper-coated cadmium column. Quantitation of biologic samples
modified Eagle'smedium media. Cellswerecountedandresuswas by reference to a linear standard curve. Nitrate and nitrite stanpended at a concentration of 1.5 X 10' cells/mL in Iscove's modified
dards gave peaks of equal area showing
that nitrate was fully reDulbecco's medium. Viability always exceeded 9S% by trypan blue
duced.
dye exclusion. BM cells were plated in 35-mm tissue culture dishes
BMT. AKR recipient mice were treated with 900 cGy total body
containing 2.8% methylcellulose, (Fisher Scientific, Itasca, IL) 30%
irradiation(TBI)within4to8hoursbeforetransplantation.
TB1
fetal calf serum, erythropoietin ( I S p/mL; Toyobo, New York, NY).
was administered in a single dose using a Shepherd Mark I cesium
hemin (6.5 &mL: Sigma),mercaptoethanol (3.2 X 1 0 ~ 'molfl,),
irradiator(JLShepherdandAssociates,SanFernando,CA).The
and pokeweed mitogen-stimulated spleen cell conditioned medium.
dose rate was 83.3 cGy/min. BM was
flushed from donor femurs
with complete Dulbecco'smodified Eagle's medium (GIBCO, Grand Conditioned medium was obtained on the sixth day of culture and
used at a concentration of S%. This had been shown to be optimal
Island, NY) plus S% fetal bovine serum. The BM plugs were passed
for promoting progenitor cell growth in BM from normal, untreated
through a sterile mesh filter to obtain single cell suspensions. BM
AKR mice. Cultures were incubated at 37°C in S% CO2 for 14 days.
cells were then washed, resuspendedin fresh medium, and counted.
Colonies (greater than S O cells) were scored at day 7 and 14 under
Spleens were passaged through sterile mesh screens to obtain single
an inverted microscope. Data were expressed as CFU-granulocyte/
cell suspensions. Spleen cells were then treated with sterile distilled
macrophage (CFU-GM) per lo5 plated cells and total CFU-GM per
water to eliminate erythrocytes. BM and spleen cells were always
two femurs
greater than 90% viable by trypan bluedyeexclusion.Irradiated
Experimenfddesign.
AKR recipientswereadministered
900
recipient mice received a single intravenous injection containing 10
cGy TB1 as pretransplant conditioning. Based on the fact that IS of
X 10' BM cells with or without 5 to 20 X 10' spleen cells.
16 (94%) irradiated control mice (without BMT) died at a
median
Flowcytometricanalysis.
Two-colorimmunofluorescencewas
of I6 days after irradiation (range,I3 to 22 days), this was considered
used in some experiments to determine the extent
of donor T-cell
a lethal dose. Animals were transplanted with
BI0.BR BM (10 X
chimerism in the spleens of transplanted animals. Fluoroscein
isoIO" cells) with or without graded dosesof spleen cells( S to 20 X 10').
thiocyanate-conjugated anti-Thy 1.2 andphycoerythrin-conjugated
anti-L3T4 (CD4) and Lyt 2 (CD8) were obtained from Becton Dick- Irradiated micereceiving BM only wereconsideredexperimental
controls, because these animals do not develop GVHD because of
inson (Mountain View, CA). Donor-derived T cells were defined as
an insufficient number of T cells in the BM inoculum (typically 2%
Thy I .2+/L3T4' and Thy I .2'/Lyt 2'. Host T cells were defined as
stimulated by cytokines such as IL-I, TNF, 7-IFN, and IL2, resulting in the augmentation of NO production both in
vitro and in V ~ V O . " " ~ Enhanced NO productionhasbeen
undergoing a GVH reacrecently documented in animals
tion," but the pathophysiologic significance
of this molecule
is unknown. Because cytokine dysregulation is thought to
play an important role in GVHD, this raises the possibility
that NO might be a more proximate mediator of a cytokinefacilitated GVH response. Conversely,otherstudies
have
shown that NO has potent immunosuppressive properties in
vitro," suggesting that the molecule might play a role in the
downregulation of immune reactivity. To evaluate the role
of NO in GVHD, we used a specific NOS inhibitor (NGmethyl-L-arginine [L-NMA]) i n a murine model of allogeneic BMT where donor and recipient are H-2-matched but
differ at minor histocompatibility antigens.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
NO PRODUCTION
ALLOGENEIC
IN GVHD AFTER
2365
BMT
Table 1. Effect of Varvina lntensitv of GVH Reactivitv on NO Production
Nitrate/Nitrite Levels (prnollL)
Experiment no. 1
B M only ( N = 3)
BMS-20 (N
5)
Day 4
6
28.3 f 2.2
127.5 -C 18.2
30.6 t 4.7
84.168.3
f 29.5
13
8
11
18.2 f 3.1
98.6
2 14.0
15.6
19.6
f 3.1
f 59.9
rt 5.1
83.5 f 6.5
Nitrate/Nitrite Levels (pmol/L)
~~
Experiment no. 2
= 4)
B M only ( N 104
BMS-5 ( N
=
4)
Day 4
7
11
21
f 18.4
190.2 f62.6
24.5
42.3 f 3.0
t 7.6
39.6 t
30.0
4.8
71.2 f 8.9
14
18
t
49.4
9.7
81.7 t74.0
21.8
f
38.0
9.2
t69.8
38.4
f 8.9
t 38.2
Cohorts of mice were alternately bled at the defined intervals for determination of nitratehitrite levels. A total of 6 and 10 mice were in the
and BMS-20 groups, respectively (Experiment no. l ) , whereas 8 mice each were in the B M and BMS-5 groups (Experiment no. 2).
Abbreviations: BMS-20. B M plus 20 X 106spleen cells; BMS-5. B M plus 5 X lo6 spleen cells.
BM
to 3%). Animals receiving BM plus spleen cells (BMS) received
treatment with either phosphate buffered saline (PBS) or L-NMA
(250 mgkg twice per day, intraperitoneally) to assess the effect of
L-NMA on GVHD and hematopoietic reconstitution. A dose of 250
mgkg was chosen based on a prior report that had shown that this
dose prevented histopathologic evidence of intestinal GVHD in a
L-NMA or saline was administered
parent + F1 murine
to animals for a total of 10 consecutive days, with the first dose
administered 4 to 6 hours before transplant. All dosing was based
on pretransplant weights of individual mice. BMS chimeras treated
with saline were designated as GVHD controls. BMS chimeras receiving either saline or L-NMA were stratified before transplant by
body weight so that the mean weight of each experimental group
was similar. Weights were obtained for every animal at least twice
weekly for 60 days. All mice were ear punched before transplant so
that individual animals could be serially assessed and to allow for
comparisons between L-NMA-treated and weight-matched GVHD
control animals. In some experiments, AKR hosts were preconditioned with 900 cGy TB1 and then transplanted with AKR BM (10
X lo6) with or without AKR spleen cells (20 X IO6) to assess the
effect of L-NMA in a syngeneic transplant model.
Sratisficat analysis. Data are reported as mean values -C 1 SD.
Descriptive statistics for weight profilesare reported as group means.
The group mean weights were based on decreasing numbers of
measurements over time because of expiring animals. Repeated measures analysis of variance was performed using individual animal
weight profiles. Analysis examined the questions of whether the
observed weight profiles showed a group difference, variation over
time, and a group-time interaction (ie, different time variation across
study groups). Analyses were performed at the two-sided 0.05 level
of significance. The Greenhouse-Geisser significance level adjustmentwasused for testing differences within animals over time.
Repeated measures analysis of variance was also performed to compare the effect of L-NMA on Cm-GM in normal murine BM and
28 1
l 90
80 -
o
i
-r? 70-
'5
5
m
3
60 50 -
Y
40
Y
30 -
3
18i
1
6
0
!
.
10
1
.
1
.
~
.
"
1
20
30
40
50
Days Post-Transplant
.
'
60
Fig 1. Serial weight curves showing lackof toxicity of L-NMA
administration in normal and transplanted AKR mice. Normal unirradiated AKR mice IN = 4; A) received a 10-day course of L-NMA (250
mglkg twice per day). Irradiated (900 cGy TBI) AKR animals were
transplanted with 10 x 10'B1O.BR B M and were then treated with
saline IN = 3; 0)
or L-NMA IN = 4; B) for 10 consecutive days.
-
::\
.
,
.
,
L
o
1
-%
tt.
.
,
.
,
.
I
0
0
10
20
30
40
50
;
60
Days Post-Transplant
Fig 2. Actuarial survival of BlO.BR/AKR chimeras. AKR mice were
conditionedwith 900 cGy T B 1 and were then administered either 10
x lo* B1O.BR BM cells alone IN = 10; 0)
or BM admixed with 5 =
10' B1O.BR spleen cells(BMS-5). BMS-5 chimeras receivedtreatment
with either saline (N = 28; 0 )or L-NMA (N= 35; 0)
for 10 consecutive
days. Data are derived from four experiments.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2366
DROBYSKI ET AL
mals ( P < ,001). This was observed as early as 4 days after
transplant and was well before any clinical signs of GVHD
wereevident.Elevated
nitrate and nitritelevelspersisted
24
26!i
throughout the period of testing in both experiments. In the
group receiving 20 X 10' spleencells, further testingwas
h
not possible because animalsbegan to dieof GVHD. Median
22
survival in this group was 18 days (n = IO), with no animals
surviving to day 60. The group receiving S X 10' spleen
S
cells had a lessintense GVH reaction as evidenced by a
.YJ
median survival of 28 days (n = 8), with 25% of animals
2o
surviving to day 60. In contrast, all animals transplanted
with BM only (n = 14) survived to day 60. NO production
18
was not significantly different between animals transplanted
with S X 10' versus 20 X IO6 spleencells.Unirradiated
normal AKR mice (n = 3) had serum nitratehitrite levels
of 15.7 ? 1.6 pmol/L.Thusnitratehitrite productionwas
increased in irradiated BM control animals when compared
with that for normal mice.
Days Post-Transplant
Administration of L-NMA inhibits NO production in recipients undergoing U GVH reaction. A study was performed
Fig 3. Serial weight curves for the first 60 days posttransplant.
to assess thetoxicity of L-NMA innormal unirradiated AKR
AKR mice were conditionedwith 900 cGy TB1 and transplanted with
B1O.BR BM IO) or B1O.BR BM plus 5 x 10' spleen cellsIBMS-5). BMSmice andinirradiated
AKR recipientsnot
developing
5 chimeras were treated with either PBS (e)or L-NMA (0)
for 10
GVHD. Irradiated AKR
hosts
were
transplanted
with
consecutive days. Values represent mean weights of surviving aniB1O.BR BM only and treated with either saline or L-NMA.
mals at each timepoint. Date are derived from the four experiments
No clinical signs of toxicitywere observed inany of the
shown in Fig 2.
three groups. All animalssurvived through day 70, at which
time they were killed.Asa
moreobjectiveparameter of
clinical well-being, serial weights were obtained in animals
to compare serial nitratehitrite levels in animals with and without
GVHD.
over the courseof this experiment (Fig I ) . Irradiated animals
Data representedby single measurements were analyzed by analy- in boththe L-NMA and saline groups showed
a characteristic
sis of variance methods in conjunction with Tukey's multiple comdecrease in weight, with nadirs occurring approximately S
parisonsmethod,atthe
95% confidencelevel, to identifygroups
to 6 days aftertransplant. There was nodifference in the
having significantly different population means. Actuarial survival
weight curvesbetween the two groups. Unirradiated animals
curves were compared using the log ranktest. Pair-wise comparison
treated with L-NMA slowly gained weight over the course
to identify groups having significantly different survival distributions
of the experiment.
were performed at the 0.025 level to preserve the overall .OS level
The effectiveness of this dose of L-NMA in inhibiting
of significance.Spleencellsizes,histologicGVHDscores,total
NO
production was assessed in a pilot study.Because a more
CFU-GM, and T-cell chimerism data were assessed using the paired
intense GVH reaction (20 X 10' spleen cells versus S X
Student'sf-test tocompareL-NMA-treatedanimalswith
salinetreated mice that had similar pretransplant weights.
10' spleen cells) was associated with higher mortality
(see
above), the efficacy ofL-NMA in animals receiving 20 X IO6
RESULTS
spleen cells was initiallytested.Nitratelnitritelevelswere
measured on day 9, near the end of L-NMA administration,
NO production is increased in GVHD induced across miat a time when levels were expected to be elevated (Table 1).
nor histocompatibility antigens. Initial studies wereperformed to determine the effectof GVHD on NO production Blood samples wereobtained by retroorbitalvenipuncture
before the morning L-NMA dose so that the nitratehitrite
in this murine BMT model. The intensity of the GVH reaclevelswould be reflective of atrough L-NMA level. The
tion was varied to assess how this affected NO production.
control group consisted of animals transplanted with BM
AKR recipients were administered 900 cGy TB1 and then
cells only. Nitratehitritelevels were higher in saline-treated
transplanted with 10 X lo6 BM cells admixed with S X lo6
mice (mean, 52.9 t 24.5 pmol/L; n = 4) than in L-NMAor 20 X lo6spleencellsfrom
B1O.BR donorstoinduce
treated animals (mean, 22.8 t 15.7 pmol/L; n = S) or animoderate or severeGVHD, respectively.Control animals
mals receivingBMonly (mean, 33.0 ? 19.4 pmol/L; n =
received TB1 and BM cells but no spleen cells so that the
impact of the conditioning regimen alone on NO production 3), indicating that L-NMA was able to suppress NO production in animals undergoing an intense GVH reaction to the
could be determined. Peripheral
blood was obtainedby retrolevel of BM controls.
orbital venipunctureat defined intervals posttransplant to
Administration of L-NMA exacerbates weight loss and
determine the time course of NO production during a GVH
adversely affects survival in mice with GVHD. The effect
reaction.
of L-NMA administration on survival of mice undergoing a
The results of these experiments are detailed in Table 1.
GVH reaction was investigated. AKR recipients were condiNitrateand nitritelevels,reflecting
NO production, were
tionedwith 900 cGy TB1 and administered 10 X 10' BM
significantly elevated compared with those of control ani-
8
S
i46
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
NO PRODUCTION IN GVHD AFTER ALLOGENEIC BMT
2367
Table 2. Effect of L-NMA on NO Production and Survival After Syngeneic
(AKR -t AKRI BMT
~~
~~
~~~________
Group No.
No.
Inoculum
of Mice
1
2
3
4
5
5
4
3
Spleen
Survival
Cell Bo-Day
Nitrate/Nitrite
Adminisiration
L-NMA
day) per(250
twice
mglkg
(% Surviving
Animals1
bmolk)
+
20 x 106
20 x 106
-
-
-
100
+100
12.3
26.4
11.8
12.8
2 1.8
f 12.2
2 7.1
2 0.7
100
l00
AKR mice were conditioned with 900 cGy TB1 and transplanted with AKR BM (10 x 10' cells) with o r without 20
Nitratehitrite levels were determined on day 4 posttransplant. All animals were then followed up for survival.
cells plus 5 X lo6 spleen cells from B1O.BR donors. Mice
received treatment for 10 days with either PBS or L-NMA
(250 mgkg twice per day, intraperitoneally). A control group
received 900 cGy T B 1 plus lo7BM cells. Nitratehitrite values on the ninth day posttransplant were determined in all
experimental groups. Saline-treated animals (n = 28)had
significantly higher nitratehitrite levels (mean, 55.6 2 12.6
PmoVL) compared with those of L-NMA-treated mice (n
= 33; mean, 42.2 f 12.4 pmoyL; (P = .OOl). Nitratehitrite
levels in control animals (n = 10) averaged 43.6 f 7.5
pmoliL, which did not differ from L-NMA-treated animals.
Survival of these animals is shown in Fig 2. Sixty-day surviva1 for saline-treated animals was significantly better than
for L-NMA-treated animals (50% v 17%;P < .001). Poorer
survival in L-NMA-treated animals was primarily because
of substantial early mortality (median survival, 12 days in
L-NMA-treated animals versus 58 days in saline-treated
animals). L-NMA-treated animals were observed to have
significantly greater weight loss (Fig 3) within the first 10
days posttransplant when compared with saline-treated mice
( P < .01). After the initial 2 weeks, saline-treated animals
281
7
26
h
E
24
22
o
IO
3020
40
50
Days Post-Transplant
60
Fig 4. Serial weight curves for the first 60 days posttransplant.
AKR mice were conditionedwith 900 cGy TB1 and transplantedwith
AKR B M alone orAKR B M plus 20 = l@
spleen cells (BMS-20). Mice
transplanted with BM only were treatedwith either saline (U) or LNMA (U1 for 10 consecutive days. BMS-20 chimeras also received
treatment with either saline(0)or L-NMA (01over the same interval.
Values represent mean weights of surviving animals at each timepoint.
x
10' AKR spleen cells.
and surviving L-NMA-treated animals were both able to
regain weight and did not differ significantly from each other
thereafter. In contrast, L-NMA-treated animals who received BM only (without spleen cells) showed no disproportionate loss in weight when compared with saline-treated
BM controls (data not shown).
To confirm that the effect of L-NMA administration was
enantiomer-specific, identically transplanted mice (n = 10)
received a 10-day course of D-NMA (250 mgkg twice per
day, intraperitoneally). Nitratehitrite levels in this group averaged 54.1 2 10.6 pmol/L, indicating that D-NMA was
unable to inhibit NO production. Weight curves were similar
to those of saline-treated animals (data not shown). Sixtyday survival was 80% with no early mortality, showing that
the effect observed with L-NMA was enantiomer-specific.
Inhibition of NO production is associated with weight loss
but not with exacerbation of GVHD early in the posttransplant period. Weight loss is known to be a sensitive indicator of GVHD in mice. Therefore, it was possible that the
observed early weight loss in L-NMA-treated animals was
because ofan exacerbation of GVHD resulting from the
suppression of NO production. To evaluate this possibility,
tissues from saline- and L-NMA-treated mice that had received BM plus 5 x lo6 spleen cells were analyzed at 9 to
10 days posttransplant for histologic evidence of GVHD. LNMA-treated animals showing the most
profound weight
loss were selected along with a group of saline-treated animals (GVHD controls) that
had
similar pretransplant
weights.
None of the animals in either group were found to have
significant pathologic changes of GVHD in the skin or liver
on the ninth day posttransplant (data not shown). Therefore,
histologic analysis was confined to the colon. Saline-treated
animals were observed to have pathologic changes in the
colon (eg, crypt cell necrosis and lymphocytic infiltration in
Table 3. Effect of L-NMA on Spleen Size in BMS Chimeras
Underaoina GVHD
Weight (g)
Pretransplant
Day
Spleen
Total
9 to 10
Size
(no. of cells x
Saline-treated
( n = 9)
L-NMA-treated
( n = 9)
24.8 f 0.8
21.8 f 1.3
24.2
16.3 t 1.4
?
1.8
23.2 f 8.6
6.98.1
Weights and spleen sizes reflect the mean values for each group.
Results are derived from three separate experiments.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2368
DROBYSKI ET AL
Fig 5. Photomicrographs of spleen sectionsfrom
AKR mice preconditionedwith 900 cGy TB1 and then
transplanted with 10 x lo6 BM plus 5 x lo6 spleen
cells from B1O.BR donors. Spleens were harvested
from animals9 to 10 days after transplant. (A) Spleen
section from a saline-treated animal showing areas
of white pulp (W) and red pulp (R). Prominent extramedullary hematopoiesis is shown in the red pulp;
note dark staining cells indicative of
erythroid hyperplasia (hematoxylin eosin stain). (B) Spleen from LNMA-treated mouse showing marked reduction in
extramedullaryhematopoiesis in the red pulp (hematoxylin eosin stain).
the lamina propria) consistent with GVHD. Although LNMA-treated animals overall had less evidence of GVHD,
differences in the histologic GVHD scores for saline (mean,
2.7 2 1.9) and L-NMA mice (mean, I .2 0.9; N = 8 mice
per group) were statistically nonsignificant (P = .IO). These
findings suggested that administration ofL-NMA did not
exacerbate GVHD and that the enhanced weight loss in LNMA-treated animals could not be ascribed to GVHD. No
other histologic abnormalities were observed in the colon,
skin, or liver of L-NMA-treated animals to account for
weight loss. The majority of surviving saline-, L-NMA-,
andD-NMA-treated animals transplanted withBMShad
clinical evidence of GVHD byday 60, indicating that LNMA,when administered over a IO-day period, wasnot
able to abrogate the eventual development of GVHD.
Weight loss was observed only in irradiated AKR recipients that had received BMS from histoincompatible B I O.BR
donors. To determine whether spleen cells alone, in the absence of histocompatibility differences between donor and
recipient, could effect weight loss, experiments were performed using a syngeneic model. AKR recipients were irra-
diated with 900 cGy and transplanted with AKR BM with
or without spleen cells (20 X 10'). Control animals received
BM only and were treated with either saline or L-NMA. The
results of this experiment are shown in Table 2. L-NMAtreated recipients of AKR BMS had identical survival compared with that of saline-treated animals. Moreover, there
was no difference in weight curves posttransplant between
any of the groups (P = .90; see Fig 4). NO production at 9
days posttransplant was not increased above that of unirradiated AKR controls. These data indicated that a histocompatibility difference between donor and recipient plus the presence of spleen cells in the donor inoculumwereboth
necessary to augment NO production andto effect enhanced
weight loss in the posttransplant period.
L-NMA treatment impairs hematopoietic reconstitution in
animals undergoing GVHD. During histologic studies for
analysis of GVHD, it was observed that the spleens of LNMA-treated chimeras showing marked weight loss were
noticeably smaller than those of saline-treated animals, suggesting that L-NMA might interfere with posttransplant hematopoietic reconstitution. To quantify this observation,
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2369
NOPRODUCTION IN GVHD AFTER ALLOGENEIC BMT
c 1.0
0
Ql
a,
.-C
C
0.9
0.8
..................................................................
................................................
I
8
m
m
mm
l
m
m
m
l
mm
U
S
-E 0.34
S
I
1
I
BM
BMS-5
Saline
BMS-5
L-NMA
Fig 6. Effectr of L-NMA administrationon donor T-cell chimerism.
AKR mice were conditionedwith 900 cGy TB1 and transplanted with
10 x lo* B1O.BR BM cells or B1O.BR BM plus 5 x lo6 spleen cells
(BMS-5). BMS-5 chimeras received al 0 day courseof treatment with
either saline or L-NMA. Animals were killed at day 9 to l 0 to assess
the proportion of donor T cells in the spleen. The horizontal line
represents the lower threshold for complete donorT-cellchimerism.'O In one experiment, animals were precondkioned with 1,100
cGy TBI. Data are derived from three experiments.
spleen sizes of both groups were assessed (Table 3). LNMA-treated mice with accelerated weight loss were killed
onday 9 posttransplant and compared with pretransplant
weight-matched saline-treated animals. The cellular content
of spleens from L-NMA-treated animals was significantly
less than that of saline-treated mice (mean, 8.1 X lo6 cells
v 23.2 X lo6 cells; P < .003). Histologic analysis of spleens
from saline-treated and L-NMA-treated animals showed
that L-NMA-treated mice had a substantial reduction in
extramedullary hematopoiesis which was most noticeable in
the red pulp (Fig 5). To evaluate donor cell repopulation in
these chimeras, flow cytometric analysis was performed to
determine the extent of donor T-cell chimerism in the spleens
of L-NMA- and saline-treated mice (Fig 6). The proportion
of donor T cells was determined by the number of CD4'
and CD8+ T cells coexpressing Thy-1.2. There was a significantly greater percentage of donor T cells in spleens of
saline-treated animals (0.63 ? 0.13) as compared with LNMA-treated mice (0.47 ? 0.13; P = .02) when evaluated
9 to 10 days after transplant.
Histologic studies of BM were then performed to determine the effect of L-NMA on BM reconstitution. Representative BM from saline- and L-NMA-treated mice are shown
in Fig 7 and show a marked reduction in cellularity in LNMA-treated animals. To further evaluate the role of NO
in hematopoietic reconstitution posttransplant, in vitro assays
were performed to assess the effect of L-NMA on progenitor
cell growth. Addition of L-NMA to C N assays had no
affect on CFU-GM ( P = .76), indicating that L-NMA was
not directly toxic or suppressive to progenitor cells (Table
4). Because it was possible that L-NMA might indirectly
affect hematopoieic reconstitution, progenitor cell growth
was investigated using BM from L-NMA- and salinetreated GVHD animals along with BM from control mice
without GVHD. AKR animals were conditioned with 900
cGy TB1 and transplanted with BM plus either 5 X lo6 or
10 X lo6 B1O.BR spleen cells to induce moderate GVHD.
GVHD animals received 10 days of treatment with either LNMA or saline. Irradiated control animals received BM only
and no other treatment. The results of these studies are shown
in Table 5. Total CFU-GM from L-NMA-treated animals
were noted to be significantly reduced when compared with
CFU-GM from saline-treated animals or from animals transplanted withBM only. The differences in total C N - G M
between saline- and L-NMA-treated animals were significantly different at day 7 (P = .OS)and at day 14 ( P < .02)
of the assay.
DISCUSSION
In this study, we have examined the pathophysiologic significance of NO in GVHD using a murine BMT model where
donor and recipient differ for minor histocompatibility antigens. This model is analogous to the majority of transplants
in humans where donor and recipient are HLA-identical but
differ at multiple minor histocompatibility loci. NO production was increased under experimental conditions, which resulted in both moderate and severe GVHD. Nitratehitrite
levels remained elevated for several weeks, suggesting that
an ongoing GVH reaction was responsible for enhanced NO
production. Nitratehitrite measurements in irradiated control
animals were observed, in some cases (Table l), to be greater
than those in normal untreated AKR mice, suggesting that
the irradiation regimen may have played a contributory role
in addition to the alloimmune effect. This is consistent with
previously published dataz5and is further supported by the
observation that nitrateinitrite levels were highest on day 4,
during which time conditioning regimen toxicity was maximal.
This study showed that L-NMA, a potent inhibitor of nitric
oxide synthase, could be administered immediately posttransplant without obvious toxicity to irradiated animals not
undergoing GVHD. In animals with GVHD, administration
of L-NMA resulted in the reduction of nitratehitrite levels
to that of BM controls as assessed by measurements taken
9 days posttransplant. Suppression was shown to be enantiomer-specific, because treatment with D-NMA didnot inhibit
NO production. Nitratehitrite measurements were deliberately taken immediately beforeL-NMA dosing. This was
done so that measurements would be performed when LNMA levels in individual mice would presumably be lowest
and would, therefore, correspond to a period of least NOS
suppression. Thus, the fact that NO production was suppressed at this point makes it likely that suppression was
continuous during the period of dosing.
Prior studies in rodents have shown that NO is able to
suppress immune reactivity in vitro. This conclusion is based
on observations that inhibition of NO production with specific NOS inhibitors enhanced mitogen- and alloantigen-induced lymphocyte proliferation." Suppression of proliferation by NO appeared to be mediated in part by macrophages,
because macrophage depletion before culture was able to
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
DROBYSKI ET AL
2370
‘I
,
”
augment a proliferative response.’6 NO has also been implicated as having a role in mediating some of the immune
suppression observed in in vitro studies performed in animals
undergoing GVHD.” Although these data support an immunosuppressive role for NO in vitro, other studies have implied that NO may be able to potentiate immune reactivity
in vivo. Lander and coworkers” showed that NO donor compounds were able to augment selected activation parameters
in peripheral blood mononuclear cells, suggesting that NO
might facilitate recruitment of other effector populations
thereby enhancing immune reactivity. Similarly, NO has
been shown to potentiate islet cell destruction caused by IL1 administration in an autoimmune model of diabetes.” More
recently, McCartney-Francis et aI3” reported that inhibition
of NOameliorated the manifestations of inflammatory arthritis. These latter studies suggest that NO plays a proinflammatory role and serves as a more proximate mediator of
tissue damage in vivo.
Although NO production has been documented to be increased in GVHD, the pathophysiologic role of NO in
GVHD has not been extensively studied. Garside and co-
Fig 7. Photomicrographsof BM obtained from
AKR mice preconditionedwith 900 cGy TB1 and then
transplanted with B1O.BR BM (10 x lo6) plus 5 x
lo6 spleen cells. BM samples wera harvested from
femurs of mice 9 to 10 days after transplant. (A) BM
from saline-treated animal showing early hematopoietic reconstitution with three cell line engraftment (hematoxylin eosin stain). (B) BM from LNMA-treated mouse showing marked reduction in
cellularity with prominent central sinusoidal dilatation reflecting an overall reduction in hematopoietic
elements (periodic
acid
Schiff
stain).
workersz4showed that treatment with L-NMA ameliorated
the pathologic manifestations of intestinal GVHD after
haploidentical BMT inmice. However, this studydidnot
verify that L-NMA effectively blocked NO production, nor
was the effect of L-NMA on survival of animals undergoing
GVHD assessed. In the present study, L-NMA administration was associated with enhanced weight loss, initially suggesting that inhibition of NO production might exacerbate
GVHD. However, histologic analyses of colonic tissue samples fromL-NMA-and
saline-treated GVHD controls
clearly indicated thatL-NMAdidnot
exacerbate GVHD
in this experimental model. This finding is consistent with
previous in vivo observations in other disease ~ t a t e s ’ ~and
.~
supports the premise thatNO does not appear toplayan
immunosuppressive role in GVHD. Amelioration of GVHD
by inhibition of NOS early after BMT wassuggested by
our data, but the study was not conclusive on thispoint.
Assessment of a potential saluatory effect ofL-NMAon
GVHD was confounded by the fact that L-NMA adversely
affected alloengraftment which is a prerequisite for the induction of GVHD. The observation that the majority of sur-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
NOPRODUCTION IN GVHDAFTERALLOGENEIC
2371
BMT
Table 4. Effect of In Vitro L-NMA Administration on CFU-GM
CFU-GM
Experiment no. 1
NO L-NMA
1 mmol/L L-NMA
Experiment no. 2
NO L-NMA
1 mmol/L L-NMA
Experiment no. 3
No L-NMA
1 mmol/L L-NMA
Day 7
Day 14
151 -C322
16.5
134 2 18.6
300
2
34.8
t 18.2
198 191
t 12.4
122 t 25.4
? 21.3
199 2 34.5
139 2 15.8
158 2 20.0
161 ? 21.8
176 t 22.4
BM was obtained from normal AKR mice and assayed in vitro for
the formation of CFU-GM in the presence or absence of 1 mmol/L LNMA. Assays were scored on days 7 and 14. Data are presented as
CFU-GM per 1 x lo5 plated cells 2 1 SD. Each experiment represents
a replicate assay.
viving L-NMA-treated animals went on to develop clinical
signs of GVHD indicated that L-NMA, as administered in
this study, was not able to abrogate the eventual development
of GVHD. Whether alternative routes of administration or
more prolonged therapy might mitigate GVHD without resulting in early mortality requires further study.
The observation of enhanced weight loss in L-NMAtreated animals was unexpected and appeared to be a major
contributing factor to early mortality. The finding of weight
loss was consistent with prior studies by Evans and colleagues3' who reported a wasting syndrome in Leishmaniasis-infected animals treated with L-NMA. These investigators attributed this finding to a reduction in food intake and
speculated that elevated levels of cytokines such as TNF
may have played a role in the development of cachexia. In
this study, weight loss was not observed in syngeneic recipients or allogeneic recipients of BM. Weight loss was observed only in irradiated animals transplanted with histoincompatible BMS, indicating that an allogeneic immune
response was a prerequisite for the induction of enhanced
weight loss. Although GVHD wasnot the cause of pronounced weight loss, the intensity of the GVHD reaction
appeared to influence the degree of weight loss and early
mortality in L-NMA-treated animals. Specifically, we have
observed that, in very mild GVH reactions, weight loss and
early mortality are mitigated (unpublished observations).
Therefore, a threshold level of GVHD appears to be necessary for this outcome. It appears that TB1 is also a critical
factor, because weight loss occurred early posttransplant,
after which time, some L-NMA-treated animals were able
to regain weight as they recovered from the conditioning
regimen. The fact that this occurred despite an ongoing GVH
reaction suggests that GVHD alone was insufficient to effect
continued weight loss. Because both TB13*and GVHD'.' are
able to cause enhanced cytokine production, this would be
consistent with the hypothesis that weight loss was cytokinemediated. Studies are ongoing to define this mechanism that
may represent a novel approach for the study of the phenomenon of cachexia in inflammatory conditions.
We cannot exclude the possibility that the microbiologic
flora of animals was in part responsible for weight loss and
early mortality in L-NMA-treated mice. Infection is a major
cause of death in mice undergoing GVHD, and the microbiologic status of animals can modulate the intensity of GVH
reactivity. This is born out by studies that have shown that
GVHD severity can be ameliorated or completely prevented
by transplanting animals in germ-free environment^.^^ NO
is known to be important in the host defense against selected
parasites and intracellular
although its full
role in the spectrum of host immune responses has not been
completely defined. It is possible that early mortality in LNMA-treated animals may have been because of a reduced
ability to eradicate infection or that the process predisposing
animals to weight loss may have enhanced host susceptibility
to infectious complications. It is unlikely that deaths were
exclusively caused by impaired hematopoietic reconstitution,
because irradiated control animals survived longer than LNMA-treated mice.
NO has been postulated to play a role in vitro in the
growth and differentiation of leukemia
suggesting
that NO may be important in hematopoiesis. L-NMAtreated mice with early weight loss were noted to have
marked reduction in BM cellularity and extramedullary hematopoiesis when compared with weight-matched salinetreated GVHD controls. Donor T-cell repopulation in the
spleens of these animals was also reduced (Fig 6 and Table
3). Collectively, these data suggest that inhibition of NOS
was able to compromise hematopoietic reconstitution posttransplant. To further evaluate this observation, we performed in vitro CFU assays to assess the effect of L-NMA
on normalmurine BM. Addition of L-NMA to colony assays
had no effect on CFU-GM, indicating that L-NMA did not
directly inhibit progenitor cell growth. In contrast, there was
a significant reduction in CFU-GM inBM obtained from
cachectic L-NMA-treated animals when compared with either GVHD or BM control animals, indicating that inhibition
of NO production indirectly effected BM suppression. The
reason inhibition of NO production was able to impair hematopoiesis is not clear. Graft rejection has not been observed
in B 10.BWAKR donorhost combination^,^^ because animals
transplanted with BM depleted of mature T cells become
mixed chimeras. This is presumably because of the lack of
mixed lymphocyte culture reactivity in the host-versus-graft
Table 5. Effect of In Vivo L-NMA Administration on CFU-GM
Total CFU-GM'
EMS L-NMA
EMS saline
BM onlv
No. of Mice
Day 7
Day 14
4
4
3
6,319t 1,877
13,297 ? 3,692
18.011 ? 1.270
4,347 t 1,955
11,860? 3,694
11.371 2 1.439
BM was obtained from AKR hosts that had been preconditioned
with 900 cGy TB1 and transplanted with B1O.BR BM with or without
5 x IO6 spleen cells. EMS chimeras received treatment with saline or
L-NMA for 10 consecutive days. BM cells were harvested 8 to 9 days
posttransplant and assayed in vitro for formation ofCFU-GM. Assays
were scored on days 7 and 14. Data are derived from two separate
experiments.
* Data expressed per two femurs 2 1 SD.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
DROBYSKI
2372
d i r e ~ t i o nTherefore,
.~~
impaired hematopoietic reconstitution
was not likely to be due to augmentation of a host-antidonor
immune response. It is tempting to speculate that L-NMA
may have augmented release of cytokines such as TNF and
y-IFN that are known to have BM suppressive effects in
vivo. However, there ispresentlyno
data to support the
premise that inhibition of NO production is able to enhance
cytokine release. It is also possible that L-NMA administration may have facilitated the induction of other soluble factors or effector cell populations which played more proximate BM suppressive roles. It should also be noted that these
observations may be due in part to the strain combinations
used in this study and that other strains will needto be
evaluated to confirm the generalizability of these results.
In conclusion, this study showed that NO production is
increased in GVHD induced across minor histocompatibility
antigens. Administration of L-NMA was able to suppress
NO production in animals undergoing GVHD but was associated with pronounced early weight loss and decreased survival. L-NMA treatment did not exacerbate GVHD early in
the posttransplant period, indicating that NO does not appear
to play a critical immunosuppressive role in this experimental model. Inhibition of NO production was associated with
impaired hematopoietic reconstitution, suggesting that NO
may play a role in facilitating engraftment in allogeneic
BMT recipients. Efforts to understand the mechanisms by
which inhibition of NO production impairs hematopoietic
reconstitution may provide new insights into understanding
the process of alloengraftment and warrants further study.
ACKNOWLEDGMENT
We thank David Majewski, Michael Chaltry, and Michael Hayward for technical assistance and Jennifer Olson for preparation of
the manuscript.
REFERENCES
1. Burakoff SJ, Deeg HJ, Ferrara J, Atkinson K: Graft-VersusHost Disease. New York, NY, Marcel Dekker, Inc, 1990
2. Vallera DA, Youle RJ, Neville DM, Kersey JH: Bone marrow
transplantation across major histocompatibility barriers
V. Protection
ofmicefromlethalgraft-versus-hostdiseasebypretreatment
of
donor cells with monoclonal anti-Thy1.2 coupled to the toxin ricin.
J Exp Med 155:949, 1982
3. Waldmann H, Hale G, Cividalli
G, Weshler Z, Manor D, Rachmilewitz EA, PolliakA, Or R, WeissL, SamuelS, Brantbar C, Slavin
S: Elimination of graft-versus-host disease by in vitro depletion of
alloreactive lymphocytes with a monoclonalrat anti-human lymphocyte antibody (Campath-l). Lancet 1:483, 1984
4. Gluckman E: Immunosuppressive drugs for the prevention of
acute graft-versus-host disease, in Burakoff SJ, Deeg HJ, Ferrara J,
Atkinson SK (eds):Graft-Versus-HostDisease,NewYork,NY.
Marcel Dekker, p 389
5. Guillen FJ, Ferrara J, Hancock W, Messadi D, Fonferko E,
Burakoff SJ, Murphy GI: Acute cutaneous graft-versus-host disease
to minor histocompatibility antigens in a murine model. Evidence
that largegranularlymphocytesareeffectorcells
in the immune
response. Lab Invest 55:35, 1986
6. Ghayur T, Seemayer TA, Lapp WS: Kinetics of natural killer
cell cytotoxicity during the graft-versus-host reaction. Transplantation 44254. 1987
7. McCarthy PL, Abhyankar S, Neben S, Newman G, Sieff C,
ET AL
Thompson RC, Burakoff SJ, FerraraJLM: Inhibition of interleukin1 by an interleukin-1 receptor antagonist prevents graft-versur-hort
disease. Blood 78:1915, 1991
8. PiguetPF,Grau
GE, AlletB,Vassalli
P: Tumournecrosis
factodcachetin is an effector of skin and gut lesions
of the acute
phase of graft-versus-host disease. J Exp Med 166:1280. 1987
9.MowatAM:Antibodies
to TFN-y preventimmunologically
mediated intestinal damage in murinegrdft-VerSUS-hoSt reaction. Immunology 68: 18, 1989
10. Nathan CF, Hibbs JB: Roleof nitric oxide synthesisin macrophage antimicrobial activity. Cum Opin Immunol 3:65, I991
1 I . Garthwaite J, Charles SL, Chess-Williams R: Endotheliumderivedrelaxingfactorreleaseonactivation
of NMDAreceptors
336385,
suggests role as intercellular messenger in the brain. Nature
1988
12. Aisaka K, Gross SS, Griffith OW, Levi R: N"-methylarginine,
an Inhibitor of endothelium-derived nitric oxide synthesis,
is a potent
pressor agentintheguinea
pig: Doesnitricoxideregulate
blood
pressure invivo'! Biochem Biophys Res Commun 160:881. 1989
13. Nathan C: Nitric oxide as a secretory product of mammals
and cells. FASEB J 6:3051, 1992
14. Moncada S, Palmer RMJ, Higgs EA: Nitric oxide: Physiology,pathophysiologyandpharmacology.Pharmacol
Rev 43: 109,
1991
1 S. Ding A, Nathan CF, Stuehr DJ: Release of reactive nitrogen
intermediates and reactive oxygen intermediates from mouseperitoneal macrophages: Comparison of activating cytokines and evidence
for independent production. J Immunol 141:2407, 1988
16. NusslerAK,DiSilvio M, BilliarTR,HoffmanRA,Geller
DA, Selby R, Madariaga J, Simmons RL: Stimulation of the nitric
oxidesynthasepathway
in humanhepatocytes by cytokinesand
endotoxin. J Exp Med 176:261, 1992
17. Kilbourn RG, Belloni P: Endothelial cell production of nitrogen oxides in response to interferon-y in combination
with tumor
necrosis factor, interleukin-l or endotoxin. J Natl Can Inst 82:772.
I990
18. Drapier JC, Weitzerbin J, Hibbs JB: Interferon-y and tumor
necrosis factor induce the L-arginine dependent cytotoxlc effector
mechanism in murine macrophages. Eur J Immunol 18:1587, 1988
19. Hibbs JB, Westenfelder C, Taintor R, Vavrin 2, Kablitz C,
BaranowskiRL,Ward JH, MenloveRL, McMurry MP,Kushner
JP,SamlowskiWE:Evidenceforcytokine-inducible
nltric oxide
synthesis from L-argininein patients receiving interleukin-2 therapy.
J Clin Invest 89:867, 1992
20. LangrehrJM,MuraseN,MarkusPM,Cai
X, Neuhaus P.
Schraut W, Simmons RL, Hoffman RA: Nitric oxide production in
host-versus-graft and graft-versus-host reactions in the
rat. J Clin
Invest 90:679, 1992
2 1. Hoffman RA, Langrehr JM, Billiar TR, CurranRD, Simmons
RL: Alloantigen-induced activation of rat splenocytes is regulated
by theoxidativemetabolism of L-arginine.JImmunol145:2220,
l990
22. Corbin JL, Reporter M: N"-Methylated arginines; A convenient preparationof N"-Methylarginine. Anal Biochem57.3 IO, 1974
23. Green LC, Wagner DA, Glogowski J , Skipper PL, Wishnok
IS, Tannenbaum SR: Analysis of nitrate, nitrite, and ["NI nitrate in
biological fluids. Anal Biochem 126:131, 1982
24. Garside P, Hutton AK,Severn A, Liew FY, Mowat AM:
Nitric oxide mediates intestinal pathology in graft-versus-host disease. Eur J Immunol 22:2141, 1992
25. Voevodskaya NV, Vanin AF: Gamma irradiation potentiates
L-arginine dependent nitric oxide formation in mice. Biochem Biophys Res Comm 186:1423, 1992
of "suppressor"macrophages.
26. Mills CD:Molecularbasis
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
NO PRODUCTION IN GVHD AFTER ALLOGENEIC BMT
Arginine metabolism via the nitric oxide synthetase pathway. J Immunol 146:2719, 1991
27. Hoffman RA, Langrehr JM, Wren SM, Dull ICE, Ildstad ST,
McCarthy SA, Simmons RL: Characterization of the immunosuppressive effects of nitric oxide in graft-versus-host disease. J Immuno1 151:1508, 1993
28. Lander HM, Sehajpal P, Levine DM, Novogrodsky A: Activation of human peripheral blood mononuclear cells by nitric oxide
generating compounds. J Immunol 150:1509, 1993
29. Bergmann L, Kroncke KD, Suscheck C, Kolb H, Kolb-Bachofern V: Cytotoxic action of IL-1B against pancreatic islets is mediated via nitric oxide formation and is inhibited by NG monomethyl
L-arginine. FEBS Lett 299: 103, 1992
30. McCartney-Francis N, Allen JB, Mizel DE, Albina JE, Xie
Q, Nathan CF, Wahl SM: Suppression of arthritis by an inhibitor of
nitric oxide synthase. J Exp Med 178:749, 1993
31. Evans TG, Thai L, Granger DL, Hibbs JL: Effect of in vivo
inhibition of nitric oxide production in Leishmaniasis. J Immunol
151:907, 1993
32. Xun CQ. Thompson JS, Jennings CD, Brown SA, Widmer
2373
MB: Effect of TBI, BuCy2 or Cy5 conditioning on inflammatory
cytokine release and development of acute and chronic graft-versushost disease in H-2 incompatible transplanted SCID mice. J Cell
Biochem 18B:78, 1994
33. Jones JM, Wilson R, Bealmear PM: Mortality and gross pathology of secondary disease in gem-free mouse radiation chimeras.
Radiat Res 45:577, 1971
34. James SL, Glaven J: Macrophage cytotoxicity against schistosomula of schistosoma mansoni involves arginine-dependent production of reactive nitrogen intermediates. J Immunol 143:4208, 1989
35. Magrinat G , Mason SN, Shami PJ, Weinberg JB: Nitric oxide
modulation of human leukemia cell differentiation and gene expression. Blood 80:1880, 1992
36. Truitt RL, Atasoylu AA: Impact of pretransplant conditioning
and donor T cells on chimerism, graft-versus-host disease, graftversus-leukemia reactivity, and tolerance after bone marrow transplantation. Blood 11:2515, 1991
37. Janeway CA, Yagi J, Conrad PJ, Katz ME, Jones B, Vroegop
S, Busser S: T-cell responses to MLs and to bacterial proteins that
mimic its behavior. Immunol Rev 107:61, 1989
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1994 84: 2363-2373
Inhibition of nitric oxide production is associated with enhanced
weight loss, decreased survival, and impaired alloengraftment in
mice undergoing graft-versus-host disease after bone marrow
transplantation
WR Drobyski, CA Keever, GA Hanson, T McAuliffe and OW Griffith
Updated information and services can be found at:
http://www.bloodjournal.org/content/84/7/2363.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.