- Email: [email protected]
NADPH oxidase but not myeloperoxidase protects lymphopenic mice from spontaneous infections
Biochemical and Biophysical Research Communications 355 (2007) 801–806 www.elsevier.com/locate/ybbrc
NADPH oxidase but not myeloperoxidase protects lymphopenic mice from spontaneous infections Dmitry V. Ostanin a, Shayne Barlow a, Deepti Shukla b, Matthew B. Grisham
a,*
a
b
Department of Molecular and Cellular Physiology, Louisiana State University Health Sciences Center, 1501 Kings Highway, P.O. Box 33932, Shreveport, LA 71130-3932, USA Department of Pathology, Louisiana State University Health Sciences Center, 1501 Kings Highway, P.O. Box 33932, Shreveport, LA 71130-3932, USA Received 5 February 2007 Available online 12 February 2007
Abstract The adaptive immune system plays an important role in host defense against invading micro-organisms. Yet, mice deficient in T- and B-cells are surprisingly healthy and develop few spontaneous infections when raised under specific pathogen-free conditions (SPF). The objective of this study was to ascertain what role phagocyte-associated NADPH oxidase or myeloperoxidase (MPO) plays in host defense in mice lacking both T- and B-cells. To do this, we generated lymphopenic mice deficient in either NADPH oxidase or MPO by crossing gp91 phox-deficient (gp91 ko) or MPO ko mice with mice deficient in recombinase activating gene-1 (RAG ko). We found that neither gp91 ko, MPO ko mice nor lymphocyte-deficient RAG ko mice developed spontaneous infections when raised under SPF conditions and all mice had life spans similar to wild-type (WT) animals. In contrast, gp91xRAG double-deficient (DKO) but not MPOxRAG DKO mice developed spontaneous multi-organ bacterial and fungal infections early in life and lived only a few months. Infections in the gp91xRAG DKO mice were characterized by granulomatous inflammation of the skin, liver, heart, brain, kidney, and lung. Addition of antibiotics to the drinking water attenuated the spontaneous infections and increased survival of the mice. Oyster glycogen-elicited polymorphonuclear neutrophils (PMNs) and macrophages obtained from gp91 ko and gp91xRAG DKO mice had no detectable NADPH oxidase activity whereas WT, RAG ko, and MPOxRAG DKO PMNs and macrophages produced large and similar amounts of superoxide in response to phorbol myristate acetate. The enhanced mortality of the gp91xRAG DKO mice was not due to defects in inflammatory cell recruitment or NO synthase activity (iNOS) as total numbers of elicited PMNs and macrophages as well as PMN- and macrophage-derived production of nitric oxide-derived metabolites in these mice were similar and not reduced when compared to that of WT mice. Taken together, our data suggest that that NADPH oxidase but not MPO (nor iNOS) is required for host defense in lymphopenic mice and that lymphocytes and NADPH oxidase may compensate for each other’s deficiency in providing resistance to spontaneous bacterial infections. 2007 Elsevier Inc. All rights reserved. Keywords: Chronic granulomatous disease; Lymphocytes; Neutrophils; Macrophages; Superoxide; Nitric oxide; Recombinase activating gene-1
The invasion of micro-organisms elicits both innate and adaptive immune responses by the host. These two arms of the body’s immune system are highly integrated with the innate immune system providing a rapid response to pathogens by producing inflammatory mediators and cytokines necessary to mount an adaptive/specific immune response to the microbes. In many cases, the innate immune system *
Corresponding author. Fax: +1 318 675 4156. E-mail address: [email protected] (M.B. Grisham).
0006-291X/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2007.02.029
can completely control an infection by mobilizing phagocytic leukocytes to the sites of infection where they ingest and destroy the infectious agent. Polymorphonuclear neutrophils (PMNs) and other phagocytic leukocytes such as monocytes and macrophages are considered to be the first line of leukocyte defense against pathogens as these cells phagocytose and kill invading bacteria via the elaboration of toxic reactive oxygen species (ROS) [1]. Indeed, a substantial portion of the anti-microbial activity of PMNs (and monocytes/macrophages) is mediated by the
802
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
multimeric flavoprotein complex, NADPH oxidase. When activated, this enzyme complex synthesizes large amounts of toxic ROS such as superoxide, hydrogen peroxide (H2O2) and, possibly, hydroxyl radical. In addition to these ROS, activated PMNs and monocytes secrete the green hemoprotein myeloperoxidase (MPO) into the phagocytic vacuole containing the ingested micro-organism where MPO catalyzes the H2O2-dependent oxidation of chloride to yield potent oxidizing and chlorinating agent, hypochlorous acid (HOCl) [1]. Together, these ROS are thought to kill the phagocytosed bacteria by oxidizing and chlorinating essential bacterial proteins involved in energy metabolism and cell division [9]. Indeed, chronic granulomatous disease (CGD) results from a defect in NADPH oxidase and thus patients with this disorder fail to produce superoxide, H2O2, HOCl, and N-chloramines [4,9]. This defect results in recurrent life-threatening bacterial and fungal infections of the lungs, skin, intestine, liver, and lymph nodes beginning early in infancy. CGD is caused by mutations in anyone of four genes encoding subunits of the nicotinamide dinucleotide phosphate (NADPH) oxidase complex including the gp91phox, p22phox, p47phox, and p67phox subunits [4]. Approximately 70% of all cases of CGD results from defects in the X-linked gene encoding the gp91phox subunit [7,12]. Over the past several years, gene targeted deletion has been used to generate different mouse models of CGD including the X-linked gp91phox-deficiency (gp91 ko) and the autosomal recessive p47phox-deficiency (p47 ko). Although earlier studies reported that both of these mutant mice possessed increased susceptibility to opportunistic pathogens, it is now clear that gp91 ko and p47 ko mice only rarely develop spontaneous infections when raised under specific pathogen-free conditions (SPF) [4,13]. Work by Shiloh et al. suggested that the inducible isoform of nitric oxide synthase (iNOS) may compensate for NADPH oxidase deficiency. Mice that are rendered deficient in both NADPH oxidase and iNOS become susceptible to spontaneous infections with commensal microorganisms, whereas, those deficient in only one of the two enzymes appeared reasonably free of infections [13]. Another protective pathway that may compensate for the lack of a functional NADPH oxidase in host defense is the adaptive arm of the immune system. Indeed, it is well-appreciated that both T- and B-cells interact with the innate immune system to provide protection against invading micro-organisms. Interestingly, lymphopenic mice such as severe combined immunodeficient (SCID) or recombinase activating gene-1-deficient (RAG ko) mice develop few spontaneous infections when raised under SPF conditions suggesting the presence of compensatory host defense mechanisms exist in these mice as well. The objective of the current study was to ascertain the role that phagocyte NADPH oxidase or MPO plays in host defense of mice that lack both T- and B-cells. To do this, we crossed gp91 ko or MPO ko mice with RAG ko animals to generate lymphopenic offspring deficient in either
NADPH oxidase or MPO. We report in this communication that NADPH oxidase but not MPO (or iNOS) is required for host defense in lymphopenic mice and that lymphocytes and NADPH oxidase may compensate for each other’s deficiency in providing resistance to spontaneous bacterial infections. We propose that the gp91xRAG double-deficient mouse represents a novel model of chronic granulomatous disease that more closely mimics the human disease. Materials and methods Mice and genotyping protocols. Wild-type (WT; C57Bl/6), RAG-1deficient (RAG ko), and gp91phox-deficient (gp91 ko) mice were purchased from The Jackson Laboratories (Bar Harbor, Maine) whereas myeloperoxidase-deficient (MPO ko) mice were obtained from Dr. Stan Hazen (Cleveland Clinic) [2]. Male or female gp91 ko (or MPO ko) mice were bred with male or female RAG ko mice and the F1 heterozygous offsprings were then bred with each other. The F2 mice were interbred to produce heterozygote gp91+/XRAG/ or null homozygote gp91/XRAG/ double-deficient (gp91xRAG DKO) offspring. The same breeding scheme was used to generate MPOxRAG DKO. All mice were housed in the animal facility at the LSUHSC in micro-isolator cages under specific pathogen-free conditions. Animals received autoclaved bedding and food and individual pairs were kept in the LSUHSC breeding facility. All mice were monitored daily for signs of disease and tail snips were obtained from all offsprings for genotyping. Presence or absence of the gp91phox gene was determined using the genotyping protocol provided by The Jackson Laboratories (Bar Harbor, Maine) using the following primers: oIMR0517, 5 0 -AAg AgA AAC TCC TCT gCT gTg AA-3 0 ; oIMR0518, 5 0 -CgC ACT ggA ACC CCT gAg AAA gg-3 0 ; oIMR0519, 5 0 -gTT CTA ATT CCA TCA gAA gCT TAT Cg-3 0 . MPO genotyping was performed according to Brennan et al. [2] using the following primers: MPOF1, 5 0 -ACC TAG GCA TCC AAT GAC TC-3 0 ; MPOR1, 5 0 -CGC CAA TCA CTA ATGTAA ACC A-3 0 ; NEOF2, 5 0 -ATC GCC TTC TTG ACG AGT TC-3 0 . Cycling conditions were as follows: 95 C 45 s, 55 C 30 s, and 72 C 1 min plus 10 s auto-extension for 35 cycles. Tissue histopathology. Brain, lungs, heart, liver, kidney, intestine and spleen were removed and fixed overnight in 10% PBS-buffered formalin. Tissue was embedded in paraffin blocks, cut into 2-lm sections, stained with hematoxylin and eosin (H&E) and used for histopathological inspection. Leukocyte recruitment, superoxide production, and nitrite/nitrate production. Elicitation of peritoneal PMNs and macrophages was performed as previously described [3]. Briefly, mice were injected (i.p.) with 25 mg of oyster glycogen type II (Sigma, St. Louis, MO) in 0.5 ml of phosphatebuffered saline (PBS) and peritoneal exudates enriched with PMNs or macrophages were collected by peritoneal lavage with PBS containing 10 mM D-glucose at 16 or 72 h following injection, respectively. Cells were counted using a hemocytometer to determine the number of extravasated leukocytes at 16 and 72 h. Phorbol myristate acetate (PMA) stimulated superoxide production by PMNs or macrophages was quantified using the SOD-inhibitable reduction of cytochrome c as previously described [6]. Nitric oxide production by PMNs and macrophages was assessed by quantifying nitrate and nitrite production as previously described [5]. Briefly, 1 · 106 leukocytes in 0.5 ml of phenol-free Dulbecco’s modified Eagle’s medium (DMEM) supplemented with L-glutamine and 1 mM L-arginine were placed in wells from a 48-well flat-bottomed plate and stimulated with 1 lg/ml LPS (serotype O55:B5, Sigma, St. Louis, MO) and 1 U/ml IFN-c (R&D Systems Inc., Minneapolis, MN) for 24 h. The concentrations of nitrate and nitrite were quantified in cell-free supernatants using the Griess reagent as previously described [5]. Statistics. Data are presented as means ± SEM (standard error of mean). Statistical significance between experimental and control groups was evaluated using unpaired t test with Welch correction. Statistical significance between more than two groups was evaluated using a one-way
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
2.0
1.0
O
T
** gp K 91 O xR A G D M K PO O xR A G D K O
**
0.0 gp
91
R A G
K
W
B
Macrophages 1.2
**
**
1.0 0.8 0.6 0.4 0.2
O K
PO xR
A G
D
D A G M
gp 91 xR
R
A G
K
K
O
O
** K O
**
0.0 gp 91
Genotyping of all offsprings confirmed the presence or absence of the gp91phox, MPO, and/or RAG-1 genes. In order to confirm that mice deficient in both RAG-1 and gp91phox (or MPO) were in fact lymphopenic and lacked functional NADPH oxidase (or MPO) we first analyzed splenocytes from these mice for the presence of CD4, CD8, and B lymphocytes using flow cytometry. We found that gp91xRAG DKO or MPOxRAG DKO mice were devoid of CD3+CD4+ and CD3+CD8+ T-cells as well as B220+ B cells (data not shown). In addition, we found that PMA-stimulated gp91 ko and gp91xRAG DKO PMNs and macrophages did not produce detectable levels of superoxide whereas superoxide production by phagocytes obtained from RAG ko mice was similar or even higher than that produced by leukocytes obtained from WT mice (Fig. 1A and B). MPO was not detected in either the MPO ko or MPOxRAG DKO mice (data not shown). In our breeding colony over a period of several years of maintaining gp91 ko, MPO ko, and RAG ko mice under SPF conditions where mice were housed in micro-isolator cages with filtered air and given sterile food, water, and bedding, we have observed less than 5% mortality in these mice. However, when we crossed gp91 ko with RAG ko mice to generate the gp91xRAG DKO offspring we observed a dramatic increase in mortality with median survival for these DKO offspring of only 53 days (Fig. 2). Gross inspection of the moribund mice suggested that these animals were unusually susceptible to spontaneous bacterial (and fungal) infections even when maintained under SPF conditions. In many cases, we observed external abscesses on the paws, preputial gland, and face as well as severe ulcerative dermatitis in these DKO animals (Supplementary Fig. S1). In addition, we observed noticeable abscesses on the surface of the internal organs including lungs, liver, kidney, and spleen. Histopathological examination of the different tissues obtained from the gp91xRAG DKO mice revealed abscesses and inflammatory foci filled with mononuclear and polymorphonuclear (PMN) cells in virtually all tissue examined (Supplementary Fig. S2). In order to maintain and expand the breeding colony we placed all mice on triple antibiotic therapy (e.g. trimethoprim, tetracycline, and neomycin sulfate) in autoclaved drinking water. This prophylactic maneuver reduced the incidence of spontaneous infections and increased survival of the offsprings with their median survival time increased to 212 days (Fig. 2). Despite the improved health of the
**
3.0
W T
NADPH oxidase but not MPO protects lymphopenic mice from spontaneous multi-organ infections
Superoxide/min/106 cells (μM)
Results and discussion
PMNs
A
Superoxide/min/106 cells ( M) (μM)
analysis of variance (ANOVA). Statistical significance between selected groups was evaluated using Dunnett’s post hoc test. A probability (p value) of p < 0.05 was considered significant. All statistical analyses were done using GraphPad InStat (TreeStar Inc., ver. 3.06 for Windows) software.
803
Fig. 1. Superoxide production by inflammatory PMNs and macrophages. (A) PMNs were elicited by injection (i.p.) of 25 mg of oyster glycogen in 0.5 ml of PBS. Exudates were collected by peritoneal lavage at 16 h following injection and PMA-stimulated superoxide production was quantified using the SOD-inhibitable reduction of cytochrome c as described in Materials and methods. (B) Macrophages were elicited by injection (i.p.) of 25 mg of oyster glycogen in 0.5 ml of PBS. Exudates were collected by peritoneal lavage at 72 h following injection and PMAstimulated production of superoxide was quantified as described above. All data are expressed as means ± SEM for triplicate samples from at least three different animals. **Indicates significant differences (p < 0.01) compared to those of WT as determined by ANOVA with Dunnett’s post-test.
antibiotic-treated gp91xRAG DKO mice, we continued to observe histopathological evidence of inflammation in the lungs of these mice (data not shown). The importance of possessing at least one copy of the gp91phox gene for resistance to infection in lymphopenic mice was illustrated by the fact that RAG ko mice heterozygous for the gp91 allele (gp91+/XRAG/) did not show increased susceptibility to spontaneous infections and essentially lived a normal life span (Fig. 2). In addition, we observed no histopathological evidence of inflammation in the lungs, liver, kidney, heart, and brain (Supplementary Fig. S3). These data clearly demonstrate that NADPH oxidase is required for host defense in lymphopenic mice. Further-
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
Leukocyte recruitment and nitric oxide production One possible mechanism that could account for the increased susceptibility to spontaneous infections in the
PMNs
A 14 12 10 8 6 4 2
91
xR
A G DK M PO O xR A G D K O
K O
0
Macrophages
B 18 15 12 9 6 3
M
D
K PO O xR A G D K O
0 G A 91 xR
more, data obtained in the current study suggest that lymphocytes and NADPH oxidase my compensate for each others’ deficiency as gp91 ko and RAG ko mice are healthy and are not susceptible to spontaneous infections when maintained under SPF condition. The role that T- and/or B-cells play in host defense is currently under investigation. Given the fact that gp91 ko mice do not develop spontaneous multi-organ infections, we propose that the gp91xRAG DKO mouse may represent a model of CGD that more closely mimics human disease. Unlike the gp91xRAG DKO mice, MPOxRAG DKO appeared healthy and did not develop life-threatening infections. Indeed, these mice lived a life span similar to WT, MPO ko and RAG ko mice suggesting that MPO (but not NADPH oxidase) is dispensable for host defense in lymphopenic mice (Fig. 2). At first glance, the lack of spontaneous infections in MPO ko or MPOxRAG DKO animals appear a bit surprising given the fact that the MPO–H2O2–Cl system in PMNs and monocytes has been shown to exert, such a potent anti-microbial system in vitro. However, it has been demonstrated that MPO deficiency may result in compensatory increases in ROS generation, degranulation, and phagocytosis thereby providing a ‘‘safety net’’ for killing of ingested microbes [9]. Indeed, our data demonstrate a significant increase in superoxide production by macrophages obtained from MPOxRAG DKO compared to that of WT mice (Fig. 1B). It should also be noted that although patients with hereditary MPO deficiency have few life-threatening infections their PMNs do have a demonstrable microbicidal defect in vitro and these individuals do appear to be more susceptible to certain fungal infections [9].
R A G
Fig. 2. Survival of gp91xRAG and MPOxRAG DKO mice. gp91xRAG DKO mice without antibiotic therapy (j, N = 7), gp91xRAG DKO with antibiotic therapy (h, N = 30), gp91+/ (het)XRAG/ without antibiotics (s, N = 13) and MPOxRAG DKO without antibiotics (, N = 6). Statistical differences between gp91xRAG DKO without antibiotic therapy and gp91xRAG DKO with antibiotic therapy was determined using the log-rank test.
gp
100 125 150 175 200
Days
KO
75
R A G
50
gp
25
O
0
K
gp91xRAG DKO - antibiotics
0
91
20
gp
p<0.0001
K O
40
gp 91
gp91xRAG DKO + antibiotics
60
W T
80
gp91xRAG DKO mice is that the phagocytic leukocytes in these animals are defective with respect to their ability to emigrate from the blood into the tissues in response to inflammatory stimuli. Therefore, we quantified recruitment of phagocytic cells in response to intraperitoneal injection of oyster glycogen in vivo in the WT and mutant mice. We found that the absolute numbers of elicited PMNs and macrophages among RAG ko, gp91xRAG DKO, and MPOxRAG DKO mice were all similar to those elicited in WT mice, suggesting defects in leukocyte extravasation could not account for the enhanced susceptibility of the gp91xRAG DKO mice to bacterial infection (Fig. 3A and B). Although not statistically significant, we observed a trend for increased cell numbers in gp91 ko mice compared to those of other groups of mice. These data are similar to those reported by Pollack et al. who demonstrated enhanced recruitment of PMNs in gp91 ko mice in response to intraperitoneal installation of thioglycollate [12].
W T
Percent survival
gp91 het xRAG ko - antibiotics
Exudate cell number/mouse (x106)
MPOxRAG DKO - antibiotics
100
Exudate cell number/mouse (x106)
804
Fig. 3. Leukocyte recruitment. (A) PMNs were elicited by injection (i.p.) of oyster glycogen and exudates were collected by peritoneal lavage at 16 h following injection. PMN viability was assessed using trypan blue exclusion and cell number was quantified using a hemocytometer. (B) Macrophages were elicited by injection (i.p.) of oyster glycogen and cells were collected at 72 h post-injection and quantified as described above. Data are expressed as means ± SEM for triplicate samples obtained from at least three different mice in each group.
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
PMNs
**
90 80 70 60 50 40 30 20 10 0
**
Work reported in this study was supported by grants from the NIH (DK 64023) and the Yamanouchi USA Foundation.
A G D M K PO O xR A G D K O
gp
91
xR
R
A
gp 91
G
KO
K O
T
Appendix A. Supplementary data
B
Macrophages
175 150 125 100 75 50 25 D K O
KO
A G M PO xR
91 xR
AG
D
K O gp
R A G
K O
0 gp 91
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.bbrc.2007.02.029. References
**
200
W T
Nitrite/nitrate per hr/106 cells (μM)
PMNs obtained from lymphopenic ko animals. In addition to PMNs, we found that macrophages isolated from gp91xRAG DKO and MPOxRAG DKO animals produced amounts of nitrite and nitrate similar to those produced by WT and gp91 ko mice (Fig. 4B). However, we did observe significantly higher production of NO-derived nitrite and nitrate by macrophages obtained from RAG ko mice compared to that of all other groups (Fig. 4B). Taken together, our data clearly show that iNOS-derived NO, cannot protect NADPH oxidase-deficient mice from spontaneous infections in the absence of T- and/or B-cells. The reasons for this are not obvious, however, it may be observed that NO exerts a protective effect in NADPH oxidase-deficient mice by augmenting T- and B-cell effector functions as has been described by others [1,8,10,11]. Taken together, our data suggest that in the absence of T- and/or B-cells, NADPH oxidase but not MPO nor iNOS, is required for protecting mice from pathogenic infection. Acknowledgments
W
A
Nitrite/nitrate per hr/106 cells (μM)
Another possible mechanism which could account for the enhanced bacterial infections and increased mortality observed in the gp91xRAG DKO mice is a defect in the ability of phagocytic leukocytes to produce NO. In order to test this possibility, we quantified LPS/IFN-c-stimulated production of NO-derived nitrite and nitrate in vitro by elicited PMNs and macrophages obtained from WT and mutant mice. Interestingly, we found that extravasated PMNs isolated from gp91xRAG DKO and MPOxRAG DKO mice actually produced twice the nitrite and nitrate levels than did PMNs obtained from the other groups of mice (Fig. 4A). The reasons for this interesting observation are not clear, however, we speculate that the presence of NADPH oxidase-derived ROS or MPO may somehow down-regulate the production of iNOS-derived NO in
805
Fig. 4. Nitrite and nitrate production by inflammatory PMNs and macrophages. (A) Oyster glycogen-elicited PMNs (106 cells) were stimulated with 1 lg/ml LPS and 1 U/ml IFN-c for 24 h. The concentration of nitrite and nitrate was quantified in cell-free supernatants as described in the Materials and methods. (B) Oyster glycogen-elicited macrophages (106 cells) were stimulated with 1 lg/ml LPS and 1 U/ml IFN-c for 24 h and nitrite and nitrate were quantified as described in the Materials and methods. Data are expressed as means ± SEM for triplicate samples obtained from at least three animals per group. **Indicates significant differences (p < 0.01) compared to WT as determined by ANOVA with Dunnett’s post-test.
[1] A. Bang, E.R. Speck, V.S. Blanchette, J. Freedman, J.W. Semple, Recipient humoral immunity against leukoreduced allogenic platelets is suppressed by aminoguanidine, a selective inhibitor of inducible nitric oxide synthase, Blood 88 (1996) 2959–2966. [2] M.L. Brennan, M.M. Anderson, D.M. Shih, X.D. Qu, X. Wang, A.C. Mehta, L.L. Lim, W. Shi, S.L. Hazen, J.S. Jacob, J.R. Crowley, J.W. Heinecke, A.J. Lusis, Increased atherosclerosis in myeloperoxidasedeficient mice, J. Clin. Invest. 107 (2001) 419–430. [3] A. Cockrell, F.S. Laroux, D. Jourd’heuil, S. Kawachi, L. Gray, H.H. Van der heyde, M.B. Grisham, Role of inducible nitric oxide synthase in leukocyte extravasation in vivo, Biochem. Biophys. Res. Commun. 257 (1999) 684–686. [4] M.C. Dinauer, Chronic granulomatous disease and other disorders of phagocytic function, Hematol. Am. Soc. Hematol. Edu. Program (2005) 89–95. [5] M.B. Grisham, G.G. Johnson, J.R. Lancaster Jr., Quantitation of nitrate and nitrite in extracellular fluids, Methods Enzymol. 268 (1996) 237–246. [6] M.R. Hoffmeyer, S.P. Jones, C.R. Ross, B. Sharp, M.B. Grisham, F.S. Laroux, T.J. Stalker, R. Scalia, D.J. Lefer, Myocardial ischemia/ reperfusion injury in NADPH oxidase-deficient mice, Circ. Res. 87 (2000) 812–817. [7] S.H. Jackson, J.I. Gallin, S.M. Holland, The p47 phox mouse knockout model of chronic granulomatous disease, J. Exp. Med. 182 (1995) 751–758. [8] S.L. James, Role of nitric oxide in parasitic infections, Microbiol. Rev. 59 (1995) 533–547.
806
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
[9] S.J. Klebanoff, in: J.I. Gallin, R. Snyderman (Eds.), Inflammation Basic Principles and Clinical Correlates Oxygen metabolites from phagocytes, Lippincott Williams & Wilkins, Philadelphia, 1999, pp. 721–768. [10] H. Lindgren, L. Stenman, A. Tarnvik, A. Sjostedt, The contribution of reactive nitrogen and oxygen species to the killing of Francisella tularensis LVS by murine macrophages, Microbes Infect. 7 (2005) 467–475. [11] H. Ohmori, H. Egusa, N. Ueura, Y. Matsumoto, N. Kanayama, M. Hikida, Selective augmenting effects of nitric oxide on antigen-
specific IgE response in mice, Immunopharmacology 46 (2000) 55– 63. [12] J.D. Pollock, D.A. Williams, M.A. Gifford, L.L. Li, X. Du, J. Fisherman, S.H. Orkin, C.M. Doerschuk, M.C. Dinauer, Mouse model of X-linked chronic granulomatous disease, an inherited defect in phagocyte superoxide production, Nat. Genet. 9 (1995) 202–209. [13] M.U. Shiloh, J.D. MacMicking, S. Nicholson, J.E. Brause, S. Potter, M. Marino, F. Fang, M. Dinauer, C. Nathan, Phenotype of mice and macrophages deficient in both phagocyte oxidase and inducible nitric oxide synthase, Immunity 10 (1999) 29–38.
NADPH oxidase but not myeloperoxidase protects lymphopenic mice from spontaneous infections Dmitry V. Ostanin a, Shayne Barlow a, Deepti Shukla b, Matthew B. Grisham
a,*
a
b
Department of Molecular and Cellular Physiology, Louisiana State University Health Sciences Center, 1501 Kings Highway, P.O. Box 33932, Shreveport, LA 71130-3932, USA Department of Pathology, Louisiana State University Health Sciences Center, 1501 Kings Highway, P.O. Box 33932, Shreveport, LA 71130-3932, USA Received 5 February 2007 Available online 12 February 2007
Abstract The adaptive immune system plays an important role in host defense against invading micro-organisms. Yet, mice deficient in T- and B-cells are surprisingly healthy and develop few spontaneous infections when raised under specific pathogen-free conditions (SPF). The objective of this study was to ascertain what role phagocyte-associated NADPH oxidase or myeloperoxidase (MPO) plays in host defense in mice lacking both T- and B-cells. To do this, we generated lymphopenic mice deficient in either NADPH oxidase or MPO by crossing gp91 phox-deficient (gp91 ko) or MPO ko mice with mice deficient in recombinase activating gene-1 (RAG ko). We found that neither gp91 ko, MPO ko mice nor lymphocyte-deficient RAG ko mice developed spontaneous infections when raised under SPF conditions and all mice had life spans similar to wild-type (WT) animals. In contrast, gp91xRAG double-deficient (DKO) but not MPOxRAG DKO mice developed spontaneous multi-organ bacterial and fungal infections early in life and lived only a few months. Infections in the gp91xRAG DKO mice were characterized by granulomatous inflammation of the skin, liver, heart, brain, kidney, and lung. Addition of antibiotics to the drinking water attenuated the spontaneous infections and increased survival of the mice. Oyster glycogen-elicited polymorphonuclear neutrophils (PMNs) and macrophages obtained from gp91 ko and gp91xRAG DKO mice had no detectable NADPH oxidase activity whereas WT, RAG ko, and MPOxRAG DKO PMNs and macrophages produced large and similar amounts of superoxide in response to phorbol myristate acetate. The enhanced mortality of the gp91xRAG DKO mice was not due to defects in inflammatory cell recruitment or NO synthase activity (iNOS) as total numbers of elicited PMNs and macrophages as well as PMN- and macrophage-derived production of nitric oxide-derived metabolites in these mice were similar and not reduced when compared to that of WT mice. Taken together, our data suggest that that NADPH oxidase but not MPO (nor iNOS) is required for host defense in lymphopenic mice and that lymphocytes and NADPH oxidase may compensate for each other’s deficiency in providing resistance to spontaneous bacterial infections. 2007 Elsevier Inc. All rights reserved. Keywords: Chronic granulomatous disease; Lymphocytes; Neutrophils; Macrophages; Superoxide; Nitric oxide; Recombinase activating gene-1
The invasion of micro-organisms elicits both innate and adaptive immune responses by the host. These two arms of the body’s immune system are highly integrated with the innate immune system providing a rapid response to pathogens by producing inflammatory mediators and cytokines necessary to mount an adaptive/specific immune response to the microbes. In many cases, the innate immune system *
Corresponding author. Fax: +1 318 675 4156. E-mail address: [email protected] (M.B. Grisham).
0006-291X/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2007.02.029
can completely control an infection by mobilizing phagocytic leukocytes to the sites of infection where they ingest and destroy the infectious agent. Polymorphonuclear neutrophils (PMNs) and other phagocytic leukocytes such as monocytes and macrophages are considered to be the first line of leukocyte defense against pathogens as these cells phagocytose and kill invading bacteria via the elaboration of toxic reactive oxygen species (ROS) [1]. Indeed, a substantial portion of the anti-microbial activity of PMNs (and monocytes/macrophages) is mediated by the
802
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
multimeric flavoprotein complex, NADPH oxidase. When activated, this enzyme complex synthesizes large amounts of toxic ROS such as superoxide, hydrogen peroxide (H2O2) and, possibly, hydroxyl radical. In addition to these ROS, activated PMNs and monocytes secrete the green hemoprotein myeloperoxidase (MPO) into the phagocytic vacuole containing the ingested micro-organism where MPO catalyzes the H2O2-dependent oxidation of chloride to yield potent oxidizing and chlorinating agent, hypochlorous acid (HOCl) [1]. Together, these ROS are thought to kill the phagocytosed bacteria by oxidizing and chlorinating essential bacterial proteins involved in energy metabolism and cell division [9]. Indeed, chronic granulomatous disease (CGD) results from a defect in NADPH oxidase and thus patients with this disorder fail to produce superoxide, H2O2, HOCl, and N-chloramines [4,9]. This defect results in recurrent life-threatening bacterial and fungal infections of the lungs, skin, intestine, liver, and lymph nodes beginning early in infancy. CGD is caused by mutations in anyone of four genes encoding subunits of the nicotinamide dinucleotide phosphate (NADPH) oxidase complex including the gp91phox, p22phox, p47phox, and p67phox subunits [4]. Approximately 70% of all cases of CGD results from defects in the X-linked gene encoding the gp91phox subunit [7,12]. Over the past several years, gene targeted deletion has been used to generate different mouse models of CGD including the X-linked gp91phox-deficiency (gp91 ko) and the autosomal recessive p47phox-deficiency (p47 ko). Although earlier studies reported that both of these mutant mice possessed increased susceptibility to opportunistic pathogens, it is now clear that gp91 ko and p47 ko mice only rarely develop spontaneous infections when raised under specific pathogen-free conditions (SPF) [4,13]. Work by Shiloh et al. suggested that the inducible isoform of nitric oxide synthase (iNOS) may compensate for NADPH oxidase deficiency. Mice that are rendered deficient in both NADPH oxidase and iNOS become susceptible to spontaneous infections with commensal microorganisms, whereas, those deficient in only one of the two enzymes appeared reasonably free of infections [13]. Another protective pathway that may compensate for the lack of a functional NADPH oxidase in host defense is the adaptive arm of the immune system. Indeed, it is well-appreciated that both T- and B-cells interact with the innate immune system to provide protection against invading micro-organisms. Interestingly, lymphopenic mice such as severe combined immunodeficient (SCID) or recombinase activating gene-1-deficient (RAG ko) mice develop few spontaneous infections when raised under SPF conditions suggesting the presence of compensatory host defense mechanisms exist in these mice as well. The objective of the current study was to ascertain the role that phagocyte NADPH oxidase or MPO plays in host defense of mice that lack both T- and B-cells. To do this, we crossed gp91 ko or MPO ko mice with RAG ko animals to generate lymphopenic offspring deficient in either
NADPH oxidase or MPO. We report in this communication that NADPH oxidase but not MPO (or iNOS) is required for host defense in lymphopenic mice and that lymphocytes and NADPH oxidase may compensate for each other’s deficiency in providing resistance to spontaneous bacterial infections. We propose that the gp91xRAG double-deficient mouse represents a novel model of chronic granulomatous disease that more closely mimics the human disease. Materials and methods Mice and genotyping protocols. Wild-type (WT; C57Bl/6), RAG-1deficient (RAG ko), and gp91phox-deficient (gp91 ko) mice were purchased from The Jackson Laboratories (Bar Harbor, Maine) whereas myeloperoxidase-deficient (MPO ko) mice were obtained from Dr. Stan Hazen (Cleveland Clinic) [2]. Male or female gp91 ko (or MPO ko) mice were bred with male or female RAG ko mice and the F1 heterozygous offsprings were then bred with each other. The F2 mice were interbred to produce heterozygote gp91+/XRAG/ or null homozygote gp91/XRAG/ double-deficient (gp91xRAG DKO) offspring. The same breeding scheme was used to generate MPOxRAG DKO. All mice were housed in the animal facility at the LSUHSC in micro-isolator cages under specific pathogen-free conditions. Animals received autoclaved bedding and food and individual pairs were kept in the LSUHSC breeding facility. All mice were monitored daily for signs of disease and tail snips were obtained from all offsprings for genotyping. Presence or absence of the gp91phox gene was determined using the genotyping protocol provided by The Jackson Laboratories (Bar Harbor, Maine) using the following primers: oIMR0517, 5 0 -AAg AgA AAC TCC TCT gCT gTg AA-3 0 ; oIMR0518, 5 0 -CgC ACT ggA ACC CCT gAg AAA gg-3 0 ; oIMR0519, 5 0 -gTT CTA ATT CCA TCA gAA gCT TAT Cg-3 0 . MPO genotyping was performed according to Brennan et al. [2] using the following primers: MPOF1, 5 0 -ACC TAG GCA TCC AAT GAC TC-3 0 ; MPOR1, 5 0 -CGC CAA TCA CTA ATGTAA ACC A-3 0 ; NEOF2, 5 0 -ATC GCC TTC TTG ACG AGT TC-3 0 . Cycling conditions were as follows: 95 C 45 s, 55 C 30 s, and 72 C 1 min plus 10 s auto-extension for 35 cycles. Tissue histopathology. Brain, lungs, heart, liver, kidney, intestine and spleen were removed and fixed overnight in 10% PBS-buffered formalin. Tissue was embedded in paraffin blocks, cut into 2-lm sections, stained with hematoxylin and eosin (H&E) and used for histopathological inspection. Leukocyte recruitment, superoxide production, and nitrite/nitrate production. Elicitation of peritoneal PMNs and macrophages was performed as previously described [3]. Briefly, mice were injected (i.p.) with 25 mg of oyster glycogen type II (Sigma, St. Louis, MO) in 0.5 ml of phosphatebuffered saline (PBS) and peritoneal exudates enriched with PMNs or macrophages were collected by peritoneal lavage with PBS containing 10 mM D-glucose at 16 or 72 h following injection, respectively. Cells were counted using a hemocytometer to determine the number of extravasated leukocytes at 16 and 72 h. Phorbol myristate acetate (PMA) stimulated superoxide production by PMNs or macrophages was quantified using the SOD-inhibitable reduction of cytochrome c as previously described [6]. Nitric oxide production by PMNs and macrophages was assessed by quantifying nitrate and nitrite production as previously described [5]. Briefly, 1 · 106 leukocytes in 0.5 ml of phenol-free Dulbecco’s modified Eagle’s medium (DMEM) supplemented with L-glutamine and 1 mM L-arginine were placed in wells from a 48-well flat-bottomed plate and stimulated with 1 lg/ml LPS (serotype O55:B5, Sigma, St. Louis, MO) and 1 U/ml IFN-c (R&D Systems Inc., Minneapolis, MN) for 24 h. The concentrations of nitrate and nitrite were quantified in cell-free supernatants using the Griess reagent as previously described [5]. Statistics. Data are presented as means ± SEM (standard error of mean). Statistical significance between experimental and control groups was evaluated using unpaired t test with Welch correction. Statistical significance between more than two groups was evaluated using a one-way
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
2.0
1.0
O
T
** gp K 91 O xR A G D M K PO O xR A G D K O
**
0.0 gp
91
R A G
K
W
B
Macrophages 1.2
**
**
1.0 0.8 0.6 0.4 0.2
O K
PO xR
A G
D
D A G M
gp 91 xR
R
A G
K
K
O
O
** K O
**
0.0 gp 91
Genotyping of all offsprings confirmed the presence or absence of the gp91phox, MPO, and/or RAG-1 genes. In order to confirm that mice deficient in both RAG-1 and gp91phox (or MPO) were in fact lymphopenic and lacked functional NADPH oxidase (or MPO) we first analyzed splenocytes from these mice for the presence of CD4, CD8, and B lymphocytes using flow cytometry. We found that gp91xRAG DKO or MPOxRAG DKO mice were devoid of CD3+CD4+ and CD3+CD8+ T-cells as well as B220+ B cells (data not shown). In addition, we found that PMA-stimulated gp91 ko and gp91xRAG DKO PMNs and macrophages did not produce detectable levels of superoxide whereas superoxide production by phagocytes obtained from RAG ko mice was similar or even higher than that produced by leukocytes obtained from WT mice (Fig. 1A and B). MPO was not detected in either the MPO ko or MPOxRAG DKO mice (data not shown). In our breeding colony over a period of several years of maintaining gp91 ko, MPO ko, and RAG ko mice under SPF conditions where mice were housed in micro-isolator cages with filtered air and given sterile food, water, and bedding, we have observed less than 5% mortality in these mice. However, when we crossed gp91 ko with RAG ko mice to generate the gp91xRAG DKO offspring we observed a dramatic increase in mortality with median survival for these DKO offspring of only 53 days (Fig. 2). Gross inspection of the moribund mice suggested that these animals were unusually susceptible to spontaneous bacterial (and fungal) infections even when maintained under SPF conditions. In many cases, we observed external abscesses on the paws, preputial gland, and face as well as severe ulcerative dermatitis in these DKO animals (Supplementary Fig. S1). In addition, we observed noticeable abscesses on the surface of the internal organs including lungs, liver, kidney, and spleen. Histopathological examination of the different tissues obtained from the gp91xRAG DKO mice revealed abscesses and inflammatory foci filled with mononuclear and polymorphonuclear (PMN) cells in virtually all tissue examined (Supplementary Fig. S2). In order to maintain and expand the breeding colony we placed all mice on triple antibiotic therapy (e.g. trimethoprim, tetracycline, and neomycin sulfate) in autoclaved drinking water. This prophylactic maneuver reduced the incidence of spontaneous infections and increased survival of the offsprings with their median survival time increased to 212 days (Fig. 2). Despite the improved health of the
**
3.0
W T
NADPH oxidase but not MPO protects lymphopenic mice from spontaneous multi-organ infections
Superoxide/min/106 cells (μM)
Results and discussion
PMNs
A
Superoxide/min/106 cells ( M) (μM)
analysis of variance (ANOVA). Statistical significance between selected groups was evaluated using Dunnett’s post hoc test. A probability (p value) of p < 0.05 was considered significant. All statistical analyses were done using GraphPad InStat (TreeStar Inc., ver. 3.06 for Windows) software.
803
Fig. 1. Superoxide production by inflammatory PMNs and macrophages. (A) PMNs were elicited by injection (i.p.) of 25 mg of oyster glycogen in 0.5 ml of PBS. Exudates were collected by peritoneal lavage at 16 h following injection and PMA-stimulated superoxide production was quantified using the SOD-inhibitable reduction of cytochrome c as described in Materials and methods. (B) Macrophages were elicited by injection (i.p.) of 25 mg of oyster glycogen in 0.5 ml of PBS. Exudates were collected by peritoneal lavage at 72 h following injection and PMAstimulated production of superoxide was quantified as described above. All data are expressed as means ± SEM for triplicate samples from at least three different animals. **Indicates significant differences (p < 0.01) compared to those of WT as determined by ANOVA with Dunnett’s post-test.
antibiotic-treated gp91xRAG DKO mice, we continued to observe histopathological evidence of inflammation in the lungs of these mice (data not shown). The importance of possessing at least one copy of the gp91phox gene for resistance to infection in lymphopenic mice was illustrated by the fact that RAG ko mice heterozygous for the gp91 allele (gp91+/XRAG/) did not show increased susceptibility to spontaneous infections and essentially lived a normal life span (Fig. 2). In addition, we observed no histopathological evidence of inflammation in the lungs, liver, kidney, heart, and brain (Supplementary Fig. S3). These data clearly demonstrate that NADPH oxidase is required for host defense in lymphopenic mice. Further-
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
Leukocyte recruitment and nitric oxide production One possible mechanism that could account for the increased susceptibility to spontaneous infections in the
PMNs
A 14 12 10 8 6 4 2
91
xR
A G DK M PO O xR A G D K O
K O
0
Macrophages
B 18 15 12 9 6 3
M
D
K PO O xR A G D K O
0 G A 91 xR
more, data obtained in the current study suggest that lymphocytes and NADPH oxidase my compensate for each others’ deficiency as gp91 ko and RAG ko mice are healthy and are not susceptible to spontaneous infections when maintained under SPF condition. The role that T- and/or B-cells play in host defense is currently under investigation. Given the fact that gp91 ko mice do not develop spontaneous multi-organ infections, we propose that the gp91xRAG DKO mouse may represent a model of CGD that more closely mimics human disease. Unlike the gp91xRAG DKO mice, MPOxRAG DKO appeared healthy and did not develop life-threatening infections. Indeed, these mice lived a life span similar to WT, MPO ko and RAG ko mice suggesting that MPO (but not NADPH oxidase) is dispensable for host defense in lymphopenic mice (Fig. 2). At first glance, the lack of spontaneous infections in MPO ko or MPOxRAG DKO animals appear a bit surprising given the fact that the MPO–H2O2–Cl system in PMNs and monocytes has been shown to exert, such a potent anti-microbial system in vitro. However, it has been demonstrated that MPO deficiency may result in compensatory increases in ROS generation, degranulation, and phagocytosis thereby providing a ‘‘safety net’’ for killing of ingested microbes [9]. Indeed, our data demonstrate a significant increase in superoxide production by macrophages obtained from MPOxRAG DKO compared to that of WT mice (Fig. 1B). It should also be noted that although patients with hereditary MPO deficiency have few life-threatening infections their PMNs do have a demonstrable microbicidal defect in vitro and these individuals do appear to be more susceptible to certain fungal infections [9].
R A G
Fig. 2. Survival of gp91xRAG and MPOxRAG DKO mice. gp91xRAG DKO mice without antibiotic therapy (j, N = 7), gp91xRAG DKO with antibiotic therapy (h, N = 30), gp91+/ (het)XRAG/ without antibiotics (s, N = 13) and MPOxRAG DKO without antibiotics (, N = 6). Statistical differences between gp91xRAG DKO without antibiotic therapy and gp91xRAG DKO with antibiotic therapy was determined using the log-rank test.
gp
100 125 150 175 200
Days
KO
75
R A G
50
gp
25
O
0
K
gp91xRAG DKO - antibiotics
0
91
20
gp
p<0.0001
K O
40
gp 91
gp91xRAG DKO + antibiotics
60
W T
80
gp91xRAG DKO mice is that the phagocytic leukocytes in these animals are defective with respect to their ability to emigrate from the blood into the tissues in response to inflammatory stimuli. Therefore, we quantified recruitment of phagocytic cells in response to intraperitoneal injection of oyster glycogen in vivo in the WT and mutant mice. We found that the absolute numbers of elicited PMNs and macrophages among RAG ko, gp91xRAG DKO, and MPOxRAG DKO mice were all similar to those elicited in WT mice, suggesting defects in leukocyte extravasation could not account for the enhanced susceptibility of the gp91xRAG DKO mice to bacterial infection (Fig. 3A and B). Although not statistically significant, we observed a trend for increased cell numbers in gp91 ko mice compared to those of other groups of mice. These data are similar to those reported by Pollack et al. who demonstrated enhanced recruitment of PMNs in gp91 ko mice in response to intraperitoneal installation of thioglycollate [12].
W T
Percent survival
gp91 het xRAG ko - antibiotics
Exudate cell number/mouse (x106)
MPOxRAG DKO - antibiotics
100
Exudate cell number/mouse (x106)
804
Fig. 3. Leukocyte recruitment. (A) PMNs were elicited by injection (i.p.) of oyster glycogen and exudates were collected by peritoneal lavage at 16 h following injection. PMN viability was assessed using trypan blue exclusion and cell number was quantified using a hemocytometer. (B) Macrophages were elicited by injection (i.p.) of oyster glycogen and cells were collected at 72 h post-injection and quantified as described above. Data are expressed as means ± SEM for triplicate samples obtained from at least three different mice in each group.
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
PMNs
**
90 80 70 60 50 40 30 20 10 0
**
Work reported in this study was supported by grants from the NIH (DK 64023) and the Yamanouchi USA Foundation.
A G D M K PO O xR A G D K O
gp
91
xR
R
A
gp 91
G
KO
K O
T
Appendix A. Supplementary data
B
Macrophages
175 150 125 100 75 50 25 D K O
KO
A G M PO xR
91 xR
AG
D
K O gp
R A G
K O
0 gp 91
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.bbrc.2007.02.029. References
**
200
W T
Nitrite/nitrate per hr/106 cells (μM)
PMNs obtained from lymphopenic ko animals. In addition to PMNs, we found that macrophages isolated from gp91xRAG DKO and MPOxRAG DKO animals produced amounts of nitrite and nitrate similar to those produced by WT and gp91 ko mice (Fig. 4B). However, we did observe significantly higher production of NO-derived nitrite and nitrate by macrophages obtained from RAG ko mice compared to that of all other groups (Fig. 4B). Taken together, our data clearly show that iNOS-derived NO, cannot protect NADPH oxidase-deficient mice from spontaneous infections in the absence of T- and/or B-cells. The reasons for this are not obvious, however, it may be observed that NO exerts a protective effect in NADPH oxidase-deficient mice by augmenting T- and B-cell effector functions as has been described by others [1,8,10,11]. Taken together, our data suggest that in the absence of T- and/or B-cells, NADPH oxidase but not MPO nor iNOS, is required for protecting mice from pathogenic infection. Acknowledgments
W
A
Nitrite/nitrate per hr/106 cells (μM)
Another possible mechanism which could account for the enhanced bacterial infections and increased mortality observed in the gp91xRAG DKO mice is a defect in the ability of phagocytic leukocytes to produce NO. In order to test this possibility, we quantified LPS/IFN-c-stimulated production of NO-derived nitrite and nitrate in vitro by elicited PMNs and macrophages obtained from WT and mutant mice. Interestingly, we found that extravasated PMNs isolated from gp91xRAG DKO and MPOxRAG DKO mice actually produced twice the nitrite and nitrate levels than did PMNs obtained from the other groups of mice (Fig. 4A). The reasons for this interesting observation are not clear, however, we speculate that the presence of NADPH oxidase-derived ROS or MPO may somehow down-regulate the production of iNOS-derived NO in
805
Fig. 4. Nitrite and nitrate production by inflammatory PMNs and macrophages. (A) Oyster glycogen-elicited PMNs (106 cells) were stimulated with 1 lg/ml LPS and 1 U/ml IFN-c for 24 h. The concentration of nitrite and nitrate was quantified in cell-free supernatants as described in the Materials and methods. (B) Oyster glycogen-elicited macrophages (106 cells) were stimulated with 1 lg/ml LPS and 1 U/ml IFN-c for 24 h and nitrite and nitrate were quantified as described in the Materials and methods. Data are expressed as means ± SEM for triplicate samples obtained from at least three animals per group. **Indicates significant differences (p < 0.01) compared to WT as determined by ANOVA with Dunnett’s post-test.
[1] A. Bang, E.R. Speck, V.S. Blanchette, J. Freedman, J.W. Semple, Recipient humoral immunity against leukoreduced allogenic platelets is suppressed by aminoguanidine, a selective inhibitor of inducible nitric oxide synthase, Blood 88 (1996) 2959–2966. [2] M.L. Brennan, M.M. Anderson, D.M. Shih, X.D. Qu, X. Wang, A.C. Mehta, L.L. Lim, W. Shi, S.L. Hazen, J.S. Jacob, J.R. Crowley, J.W. Heinecke, A.J. Lusis, Increased atherosclerosis in myeloperoxidasedeficient mice, J. Clin. Invest. 107 (2001) 419–430. [3] A. Cockrell, F.S. Laroux, D. Jourd’heuil, S. Kawachi, L. Gray, H.H. Van der heyde, M.B. Grisham, Role of inducible nitric oxide synthase in leukocyte extravasation in vivo, Biochem. Biophys. Res. Commun. 257 (1999) 684–686. [4] M.C. Dinauer, Chronic granulomatous disease and other disorders of phagocytic function, Hematol. Am. Soc. Hematol. Edu. Program (2005) 89–95. [5] M.B. Grisham, G.G. Johnson, J.R. Lancaster Jr., Quantitation of nitrate and nitrite in extracellular fluids, Methods Enzymol. 268 (1996) 237–246. [6] M.R. Hoffmeyer, S.P. Jones, C.R. Ross, B. Sharp, M.B. Grisham, F.S. Laroux, T.J. Stalker, R. Scalia, D.J. Lefer, Myocardial ischemia/ reperfusion injury in NADPH oxidase-deficient mice, Circ. Res. 87 (2000) 812–817. [7] S.H. Jackson, J.I. Gallin, S.M. Holland, The p47 phox mouse knockout model of chronic granulomatous disease, J. Exp. Med. 182 (1995) 751–758. [8] S.L. James, Role of nitric oxide in parasitic infections, Microbiol. Rev. 59 (1995) 533–547.
806
D.V. Ostanin et al. / Biochemical and Biophysical Research Communications 355 (2007) 801–806
[9] S.J. Klebanoff, in: J.I. Gallin, R. Snyderman (Eds.), Inflammation Basic Principles and Clinical Correlates Oxygen metabolites from phagocytes, Lippincott Williams & Wilkins, Philadelphia, 1999, pp. 721–768. [10] H. Lindgren, L. Stenman, A. Tarnvik, A. Sjostedt, The contribution of reactive nitrogen and oxygen species to the killing of Francisella tularensis LVS by murine macrophages, Microbes Infect. 7 (2005) 467–475. [11] H. Ohmori, H. Egusa, N. Ueura, Y. Matsumoto, N. Kanayama, M. Hikida, Selective augmenting effects of nitric oxide on antigen-
specific IgE response in mice, Immunopharmacology 46 (2000) 55– 63. [12] J.D. Pollock, D.A. Williams, M.A. Gifford, L.L. Li, X. Du, J. Fisherman, S.H. Orkin, C.M. Doerschuk, M.C. Dinauer, Mouse model of X-linked chronic granulomatous disease, an inherited defect in phagocyte superoxide production, Nat. Genet. 9 (1995) 202–209. [13] M.U. Shiloh, J.D. MacMicking, S. Nicholson, J.E. Brause, S. Potter, M. Marino, F. Fang, M. Dinauer, C. Nathan, Phenotype of mice and macrophages deficient in both phagocyte oxidase and inducible nitric oxide synthase, Immunity 10 (1999) 29–38.