The FASEB Journal • FJ Express Full-Length Article

Systemic NO production during (septic) shock depends on parenchymal and not on hematopoietic cells: in vivo iNOS expression pattern in (septic) shock Jennyfer Bultinck,* Patrick Sips,* Luc Vakaet,† Peter Brouckaert,* and Anje Cauwels*,1 *Department for Molecular Biomedical Research, Ghent University and VIB, Ghent, Belgium; and † Department of Radiotherapy, Ghent University Hospital, Ghent, Belgium Septic shock is the leading cause of death in noncoronary intensive care units and the 10th leading cause of death overall. Several lines of evidence support an important role for the vasodilator NO in hypotension, a hallmark of septic shock. However, NO may also positively or negatively regulate inflammation, apoptosis, and oxidative stress. These dual effects of NO may relate to its isoform specific production but also to differences in cellular and/or temporal expression. Via bone marrow transplantations, we examined the contribution of hematopoietic cells to the dramatically elevated NO levels seen in (septic) shock. Surprisingly, hematopoietic cells are not responsible at all for the production of circulating NO after systemic tumor necrosis factor or lipopolysaccharide challenge and contribute only marginally in a bacteremic (Salmonella) model of septic shock. Immunohistochemistry identified the nonhematopoietic sources of NO as hepatocytes, paneth cells, and intestinal and renal epithelial cells. In contrast, during granulomatous Bacillus Calmette-Gue´rin inflammation, the hematopoietic cell population represents the sole source of systemic NO. These mouse data demonstrate that, in contrast to the general conjecture, the dramatically elevated levels of NO during (septic) shock are not produced by hematopoietic cells such as monocytes/macrophages but rather by parenchymal cells in liver, kidney and gut.— Bultinck, J., Sips, P., Vakaet, L., Brouckaert, P., Cauwels, A. Systemic NO production during (septic) shock depends on parenchymal and not on hematopoietic cells: in vivo iNOS expression pattern in (septic) shock FASEB J. 20, E1619 –E1627 (2006) ABSTRACT

Key Words: inflammation 䡠 inducible NOS 䡠 bone marrow transplantation

NO is a free radical with pleiotropic functions in a wide variety of physiological systems. In the cardiovascular system, NO exerts vasodilatory, antiadhesive and antiproliferative effects (1). Three different isoforms of NO synthase (NOS) exist: endothelial (e), neuronal (n), and inducible (i) NOS. As opposed to the constitutive enzymes eNOS and nNOS, iNOS is synthesized de novo during inflammation, is Ca2⫹/calmodulin-inde0892-6638/06/0020-1619 © FASEB

pendent, and produces large amounts of NO over prolonged periods of time (2). NO derived from iNOS is essential for the survival of the host during infection, but the exact function(s) of induced NO in inflammatory settings like sepsis, systemic TNF treatment, inflammatory bowel disease, and rheumatoid arthritis remain unclear (3–5). TNF is a central regulator of immunity and inflammation. Systemic TNF treatment provokes a lethal shock syndrome, in which cardiovascular collapse is centrally orchestrated by NO (6). TNF interacts with TNF-R1 and TNF-R2, with most TNF activities being mediated by TNF-R1 (7, 8). We have previously reported that cardiovascular failure and mortality after TNF treatment can be prevented by inhibition of soluble guanylate cyclase activation, the prototype receptor for NO (9). However, iNOS-derived NO also exerts protective effects by tempering TNF-induced oxidative stress (10). Sepsis and septic shock are systemic inflammatory response syndromes resulting from and complicating an infection. They are the leading cause of death in noncoronary intensive care units and the incidence continues to increase (11). LPS, a constituent of the outer membrane of Gram-negative bacteria, mimics most of the septic effects and is widely used as an experimental model for septic shock (12). NO plays a controversial role in the pathophysiology of septic shock. Since it causes hemodynamic instabilities and tissue damage (13–16), NO may be considered detrimental to the host. However, NO may also confer protection to the host as an anti-inflammatory, antioxidant, and/or antiapoptotic agent (17). This ambiguity became remarkably clear when a clinical trial using NOS inhibitors in septic shock patients had to be prematurely terminated because of increased mortality associated with treatment, despite beneficial effects on systemic blood pressure (18, 19). Several studies also

1

Correspondence: Department for Molecular Biomedical Research, Ghent University/VIB, Technologiepark 927, Ghent 9052, Belgium. E-mail: [email protected]. doi: 10.1096/fj.06-5798fje E1619

suggest a variable role for iNOS depending on its cellular location and temporal expression (20 –24). Despite the fact that a multitude of cell types, both of hematopoietic and nonhematopoietic origin, have the ability to express iNOS in vitro as well as in vivo (25–30), it is still generally believed that macrophages, or immune cells in general, are the principle source of high systemic NO levels during septic shock. To determine the actual in vivo contribution of hematopoietic cells in systemic NO production during inflammatory or septic shock, we performed bone marrow transplantation (BMT) experiments between wild-type (WT) and iNOSor TNF-R1-deficient mice. Surprisingly, hematopoietic cells did not contribute to the production of systemic NO after TNF or LPS challenge and only minimally during Salmonella infection. In line with this, the major iNOS-expressing cells in these three shock models were identified as hepatocytes, paneth cells, and epithelial cells of the gastrointestinal tract and the kidney. In contrast, during BCG infection, a granulomatous inflammatory reaction model characterized by the formation of granulomas in reticulo-endothelial organs, hematopoietic cells were clearly the only source of systemic NO.

MATERIALS AND METHODS Laboratory Animals Female C57BL/6 mice were purchased from Janvier (France). iNOS-deficient (31; Jackson Laboratories) and TNF-R1-deficient mice (8) on a B6 background were bred as homozygotes in our facilities. All mice were housed in temperature-controlled, air-conditioned facilities with 14/10 h light/dark cycles, food and water ad libitum, and were used at 7–12 wk at the beginning of the experiment. All experiments were performed according to the guidelines of the animal ethics committee from the Ghent University, Belgium.

Blood Collection, IL-6, and NOxⴚ Measurement Blood was collected via cardiac puncture, allowed to clot at 37°C, and centrifuged to obtain serum. Interleukin (IL)-6 was measured using a 7TD1 bioassay (32). NO2⫺ ⫹ NO3⫺ (NOx⫺) concentration was determined using Griess reagent (33) as described before (9). BMT Recipient mice at 9 –12 wk of age were lethally irradiated (9 Gy for iNOS⫺/⫺ and 10 Gy for WT or TNF-R1⫺/⫺) and the next day injected intravenously with 107 BM cells isolated from femur and tibia of donor mice. The mice were left to recover for at least 8 wk. As a control for sufficient irradiation, one mouse was not reconstituted and died within 2 wk. Four days before irradiation, mice were housed in sterile cages. The drinking water was sterile, containing 0.2% neomycin sulfate. Control polymerase chain reaction for BMT Successful reconstitution of the hematopoietic lineage after irradiation and BMT was monitored by polymerase chain reaction (PCR). Mice were bled for DNA preparation from the WBC population, using the Puregene DNA Isolation Kit (Gentra Systems, Minneapolis, MN). The following primers were used: 5⬘-CCTCAGAGTCCTTCATGAAGCACAATGC-3⬘ and 5⬘-TGTTGAAGGCGTAGCTGAACAAGGTGG-3⬘ (wildtype iNOS); 5⬘-TAGGTCTCTAAGTCTGTGGACTTGGACC-3⬘ and 5⬘-ACTGCTCGACATTGGGTGGAAACATTCC-3⬘ (iNOS⫺/⫺, ref 31); 5⬘-TGCTGATGGGGATACATCCAT-3⬘ and 5⬘-CTTTGTGGCACTTGGTGCAGC-3⬘ (wild-type TNFR1); and 5⬘-CAACGCTATGTCCTGATAGCGGTCC-3⬘ and 5⬘CGTGTTCCGGCTGTCAGCGCAAGG-3⬘ (TNF-R1⫺/⫺). PCR amplification was conducted for 5 min at 95°C, followed by 39 cycles of 1 min at 95°C, 1 min at 69°C (wild-type iNOS) or at 60°C (i NOS⫺/⫺), 1.5 min at 72°C, and finally 10 min at 72°C; or 4 min at 94°C, followed by 34 cycles of 45 s at 94°C, 45 s at 57°C, 30 s at 72°C, and finally 7 min at 72°C (wild-type TNF-R1); or 4 min at 94°C, followed by 34 cycles of 1 min at 94°C, 1 min at 55°C, 3 min at 72°C, and finally 7 min at 72°C (TNF-R1⫺/⫺). Real time-Quantitative PCR

Reagents Recombinant murine TNF was produced in E. coli and purified to homogeneity. Two different batches were used. For the kinetic and BMT experiments, specific activity was 1.6 ⫻ 108 IU/mg. In all other experiments, TNF had a specific activity of 1.2 ⫻ 108 IU/mg. The endotoxin content was ⬍0.02 ng/mg (Limulus amoebocyte lysate assay, NODIA BV, Belgium). E. coli (serotype 0111:B4) and S. abortus equi LPS were purchased from Sigma (St. Louis, MO, USA). Live BCG organisms were provided by the Institut Pasteur du Brabant (Brussels, Belgium) and the virulent S. enteritidis bacteria by the Ghent University Hospital (Ghent, Belgium). Injections and bacterial inoculations Injections of TNF, LPS, and BCG were done intravenously in 0.2 ml. Salmonella bacteria were grown at 37°C in LuriaBertani (Difco, Detroit, MI, USA) and inoculated intraperitoneally in 0.2 ml PBS. The number of viable bacteria was checked by dilution and pour plating. E1620

Vol. 20

November 2006

Total RNA was extracted from 30 mg tissue with the RNeasy Protect Mini Kit (Qiagen, Valencia, CA). During RNA purification, an on-column DNase digestion removed any residual genomic DNA. 1 ␮g of total RNA (measured spectrophotometrically) was reverse transcribed with random hexamers into cDNA using the Superscript First-Strand Synthesis System for reverse transcriptase (RT)-polymerase chain reaction (Invitrogen Life Technologies, Carlsbad, CA). The cDNA was diluted 40-fold [for iNOS, GAPDH, and hydroxymethylbilane synthase (HMBS) ]or 400-fold [for ribosomal protein l13a (Rpl13a)] before PCR amplifications, which were performed in a total volume of 25 ␮l, containing 10 ␮l cDNA sample, 10 ␮l qPCR Core Kit for SYBR Green I mix (Eurogentec, Seraing, Belgium) and 5 ␮l primer mix (for GAPDH 125 nM, for iNOS, HMBS, and Rpl13a 300 nM of each primer). A nontemplate control was always run and all PCR amplifications were performed in triplicate under the following conditions: 2 min at 50°C and 10 min at 95°C, followed by a total of 40 two-temperature cycles (15 s at 95°C and 1 min at 60°C). After amplification, a melting curve was acquired by heating the product to 95°C, cooling it to 60°C, and keeping it there for 15 s and then slowly heating it to 95°C. Fluorescence was

The FASEB Journal

BULTINCK ET AL.

measured through the slow heating phase. Melting curves were used to determine the specificity of PCR products. The following forward and reverse murine primers were used: 5⬘-CAGCTGGGCTGTACAAACCTT-3⬘ and 5⬘-CATTGGAAGTGAAGCGTTTCG-3⬘ (inducible NOS); 5⬘-GAAACTCTGCTTCGCTGCATT-3⬘ and 5⬘-TGCCCATCTTTCATCACTGTATG-3⬘ (HMBS); 5⬘-CCTGCTGCTCTCAAGGTTGTT-3⬘ and 5⬘-TGGTTGTCACTGCCTGGTACT T-3⬘ (Rpl13a); and 5⬘-ACCATCTTCCAGGAGCGAGAC-3⬘ and 5⬘-GCCTTCTCCATGGTGGTGAA-3⬘ (GAPDH). Relative RNA levels are presented, normalized against the mean of three reference housekeeping genes: HMBS, Rpl13a, and GAPDH. WB Analysis Frozen organs were homogenized and cleared by centrifugation; 25 ␮g of total protein was fractionated by SDS-PAGE, transferred to nitrocellulose, and incubated with a rabbit anti-inducible NOS antibody (Ab; Transduction Laboratories, Lexington, KY, 1/6000). Immune complexes were detected using a horseradish peroxidase-coupled anti-rabbit IgG Ab (Cell Signaling Technology, Beverly, MA) and the enhanced chemiluminescence (ECL) method (Western Lightning Chemiluminescence Reagent plus, PerkinElmer Life Science, Boston, MA). Tissue Embedding, Sectioning, and Staining Organs were fixed in 4% paraformaldehyde. After dehydration and paraffin embedding, 4 ␮m sections were prepared and incubated for 3 h at 37°C with rabbit anti-iNOS Ab (1/200, Chemicon, Temecula, CA; 1/500, Transduction Laboratories, Lexington, KY; 1/200, Santa Cruz Biotechnology, Santa Cruz, CA). Envision was used for 40 min at RT as secondary Ab, and visualization was done with an aminoethylcarbazole substrate (both from DAKO, Carpinteria, CA). Statistics Significance of IL-6 and NOx⫺ changes was examined using a one-way ANOVA Bonferroni test (with comparison of all pairs). NOx⫺ levels were compared with the corresponding treatment in the WT to WT group, except for the Salmonella infection experiment, where it is indicated separately.

RESULTS Kinetics of NOxⴚ production after different inflammatory stimuli To verify the optimal time point for measurement of systemic NOx⫺ (⫽NO2⫺⫹NO3⫺⫽biologically stable breakdown products of NO), kinetic studies were performed. As shown in Fig. 1A, NOx⫺ release in circulation was significantly detectable from 4 h after lethal TNF or LPS challenge on and further increased until 12 or 16 h, respectively. Three independent studies gave identical kinetic patterns. Hence, we decided to determine systemic NOx⫺ levels 8 h after TNF or LPS treatment in all future experiments, considering the absence of mortality and the significant amounts of NOx⫺ produced at that time point. For the bacteremic model of septic shock, mice were inoculated intraperitoneally with 800 cfu of S. enteritidis bacteria, causing SHOCK-ASSOCIATED iNOS AND SYSTEMIC NO IN MICE

Figure 1. A) TNF-, LPS-, or Salmonella-induced systemic NOx⫺ production. WT mice were injected iv with PBS, 15 ␮g TNF, or 300 ␮g S. abortus equi LPS, or ip with 800 cfu S. enteritidis. Numbers within figure are number of mice that survived/ total number of mice, at indicated time points. Only surviving mice were bled and used for NOx⫺ measurement. ***P ⬍ 0.001, **P ⬍ 0.01, *P ⬍ 0.05, as compared with PBS. B) NOx⫺ levels in serum of TNF- or LPS-treated WT, iNOS⫺/⫺ and BMT mice. Mice were injected iv with PBS, 12.5 ␮g TNF, 300 ␮g E. coli LPS or 150 ␮g S. abortus equi LPS. Serum NOx⫺ levels were determined 8 h later. ***P ⬍ 0.001, as compared with corresponding treatment in WT to WT group. Insert: PCR reactions, specific for WT or deficient iNOS allele, were performed on WBC DNA of BMT mice. I: WT to WT, II: iNOS⫺/⫺ to WT, III: WT to iNOS⫺/⫺, IV: iNOS⫺/⫺ to iNOS⫺/⫺, M: marker. C) NOx⫺ levels in serum of TNF-treated WT, TNF-R1⫺/⫺ and BMT mice. Mice were injected iv with PBS or 12.5 ␮g TNF. Systemic NOx⫺ was determined 8 h after treatment. Plotted are means ⫾ sd; numbers above bars are number of mice used. ***P ⬍ 0.001 as compared with corresponding treatment in WT to WT group.

death from 36 h on (not shown). Systemic NOx⫺ was detectable after 6 to 12 h postinfection, reaching a plateau 12–20 h after inoculation. However, in contrast to TNF or LPS, systemic NOx⫺ levels dropped again afterward, before the onset of lethality (Fig. 1A). Because the kinetics of NOx⫺ production varied considerably between different experiments, we followed systemic NOx⫺ levels at various time points in all E1621

subsequent Salmonella experiments. In BCG-infected mice, we determined NOx⫺ levels 2 h after challenge, since mortality occurred 2 to 3 h after TNF treatment due to hypersensitivity of the BCG-infected mice (34). Parenchymal cells are the only source of systemic NOxⴚ during TNF- or LPS-induced shock To investigate the involvement of the hematopoietic cell population in systemic NOx⫺ production after TNF or LPS, we performed BMT experiments. Irradiated iNOS⫺/⫺ mice were reconstituted with WT BM cells and vice versa. As a control, WT and iNOS⫺/⫺ mice were, after irradiation, transplanted with homologous BM cells. To check the BMT effectiveness, the genotype of circulating WBCs was determined. As shown in Fig. 1B (insert), virtually all the hematopoietic cells of the reconstituted mice were of the donor genotype. Three days later, the mice were injected intravenously with lethal doses of TNF, E. coli or S. abortus equi LPS. Systemic IL-6 levels 8 h after challenge were comparable, indicating an equivalent inflammatory reaction in all control and BM-chimeric mice (not shown). TNF or LPS did not induce any detectable NOx⫺ in circulation of iNOS⫺/⫺ mice with an iNOS⫹/⫹ hematopoietic cell population, while they clearly did in mice that express iNOS in every cell but the hematopoietic cells (Fig. 1B). Therefore, in TNF- or LPS-challenged mice, parenchymal cells are fully responsible for the systemic NOx⫺ production. For TNF, we confirmed this result with a BMT experiment between WT and TNF-R1⫺/⫺ mice (Fig. 1C). Also in a bacteremic model of septic shock, parenchymal cells are the primary source of systemic NOxⴚ The lack of involvement of hematopoietic cells in TNFor LPS-induced systemic NOx⫺ production was unexpected, considering the importance of these cells in immunity and inflammation and their in vitro ability to produce massive amounts of NOx⫺ in response to LPS (35–37). To substantiate the insignificance of immune cells in NOx⫺ production during shock, we set up an infectious Salmonella enteritidis model. As reported before (38, 39), iNOS⫺/⫺ mice were more susceptible to Salmonella infection (Fig. 2A). Since iNOS deficiency in either parenchymal or hematopoietic cells increased susceptibility, both cell types seem to be involved in NO-mediated host defense during salmonellosis (Fig. 2A). IL-6 and NOx⫺ levels were higher in all Salmonellainfected BMT animals (Fig. 2B, C), indicating increased inflammation due to the irradiation and transplantation procedure. Therefore, we compared systemic NOx⫺ levels among BMT mice only. Between 18 to 30 h after infection, significant amounts of circulating NOx⫺ were present in Salmonella-infected WT to WT mice (Fig. 2C), as well as in WT mice with iNOS⫺/⫺ BM. In contrast, only very low and insignificant levels of NOx⫺ were found in chimeric mice with iNOS-negative parenE1622

Vol. 20

November 2006

Figure 2. A) Lethality of Salmonella-infected BMT mice. WT, iNOS⫺/⫺, and BMT mice were injected ip with 400 cfu S. enteritidis. B and C) IL-6 and NOx⫺ levels in serum of Salmonella-infected BMT mice. WT, iNOS⫺/⫺, and BMT mice were injected ip with 400 cfu S. enteritidis. Plotted are means ⫾ sd; numbers above bars are number of mice used. ***P ⬍ 0.001, **P ⬍ 0.01, *P ⬍ 0.05, as compared with WT group at a single time point for IL-6 (B) or as indicated for NOx⫺ (C). † Indicates 100% lethality.

chymal cells (Fig. 2C). Hence, we conclude that in a bacteremic model of septic shock, there is undoubtedly a predominant role for parenchymal cells in systemic NOx⫺ production. Hematopoietic cells are responsible for systemic NOxⴚ production during granulomatous inflammation In an analogous BMT experiment between WT and iNOS⫺/⫺ mice, we found that hematopoietic cells are solely responsible for systemic NOx⫺ production during BCG infection (Fig. 3). All mice displayed a sub-

The FASEB Journal

BULTINCK ET AL.

Cellular expression pattern of iNOS via IHC

Figure 3. NOx⫺ levels in serum of BCG-infected BMT mice. BCG-infected WT, iNOS⫺/⫺ and BM-transplanted mice were injected iv with PBS or 5 ␮g TNF. Systemic NOx⫺ levels were determined 2 h later. Plotted are means ⫾ sd; numbers above bars are number of mice used. ***P ⬍ 0.001, **P ⬍ 0.01, as compared with corresponding treatment in WT to WT group.

stantial enlargement of spleen and liver, which provides evidence that the granulomatous inflammatory reaction is triggered by BCG. Only in mice with a WT parenchymal compartment, an extra TNF challenge caused an additional increase in NOx⫺ production (Fig. 3).

In control mice, none of the organs stained positive for iNOS (not shown). Identical iNOS expression patterns were observed 8 h after TNF or LPS (Fig. 5A–H), although less pronounced after LPS, correlating with the systemic NOx⫺ levels (Fig. 1). Positive iNOS staining was detected in hepatocytes, often specifically localized at the canalicular membrane (Fig. 5A, E). In the kidney, epithelial cells of pelvis (Fig. 5B, F) and tubuli (not shown) stained positive. Throughout the entire gastrointestinal tract, apical epithelial cells clearly expressed iNOS (Fig. 5C, G), as well as the paneth cells of the jejunum (Fig. 5D, H). In addition, after LPS we also observed some iNOS-positive Kupffer cells and occasional granulocytes and/or monocytes in liver, lung, spleen, kidney, and heart (not shown). Nevertheless, despite their ability to express iNOS protein in vivo, our BMT experiments indicate that these WBCs do not contribute significantly to the massive systemic NOx⫺

RNA and protein expression patterns of iNOS To elucidate the induction pattern of iNOS, we determined the iNOS mRNA levels in various organs via real time-quantitative PCR (Fig. 4A). It should be noted that the up-regulation of iNOS mRNA can be compared for the various treatments within a single organ only and not between different organs, since organ-specific fluctuations in the expression level of the reference genes influences the relative iNOS mRNA levels. In control mice, minimal traces of iNOS mRNA could be found in lung, spleen, kidney, intestinal tract, and heart. In liver, all three inflammatory stimuli were extremely potent inducers of iNOS expression. In the kidney, iNOS mRNA was found in an approximately equal amount after all three stimuli, and in the heart, a moderate up-regulation of iNOS was also observed. In lung and spleen, BCG was by far the best iNOS inducer, whereas in jejunum and colon, iNOS mRNA was predominantly found after TNF or LPS treatment, respectively. Because the presence of mRNA is no guarantee for subsequent translation into protein, we also performed WB analysis on organ homogenates. In contrast to mRNA, iNOS protein was never detected in control mice (Fig. 4B). After BCG infection, a very strong iNOS-specific band was observed in liver, lung, and spleen and a smaller one in kidney and heart (Fig. 4B), corresponding with the organs in which granulomas could be detected. After TNF or LPS, iNOS protein was always detectable in liver, kidney, jejunum, and colon. In the other organs, however, a very small band was detectable in 50% of the cases. WB on iNOS-deficient organs treated in parallel never showed the iNOSspecific band, as illustrated in Fig. 4B for LPS treatment. SHOCK-ASSOCIATED iNOS AND SYSTEMIC NO IN MICE

Figure 4. iNOS mRNA and protein expression pattern. A) Real time-quantitative PCR was performed on RNA extracted 8 (TNF, black bars or LPS, hatched bars) or 2 (BCG, cross-hatched bars) h after challenge. Relative mRNA levels, normalized against mean of 3 reference genes, are shown. Up-regulation of iNOS mRNA is presented as a percentage of the response in PBS (open bars) mice. B) WB analysis. Data shown are from 4 individual experiments. Li: liver; Lu: lung; S: spleen; K: kidney; J: jejunum; C: colon; H: heart. E1623

Figure 5. iNOS-specific IHC. Organs were collected 8 h after 5 ␮g TNF or 150 ␮g S. abortus equi LPS, 24 h after 800 cfu of S. enteritidis, or 2 h after PBS in BCG-infected mice. Arrows in A indicate canalicular membranes of hepatocytes. Magnification ⫽ ⫻ 40 (J, N), ⫻ 60 (A, C, E, G), or ⫻ 100 (B, D, F, H, I, K, L, M, O, P).

production. Immunohistochemical analysis in LPStreated iNOS⫺/⫺ mice with WT BM confirmed this: an occasional granulocyte and/or monocyte stained iNOSpositive (not shown), although no systemic NOx⫺ could be detected (Fig. 1B). Identical results were obtained 16 h after TNF or LPS (not shown). One day after the inoculation of 800 cfu of Salmonella enteritidis, similar iNOS expression patterns were detected as for LPS (Fig. 5I–L). In addition, some cells of the renal papilla and white pulpa of the spleen also expressed iNOS (not shown). Salmonella infection also occasionally induced some small iNOS-positive granulomas (not shown). Other iNOS-expressing hematopoietic cells were Kupffer cells (Fig. 5I), cells in the red pulpa of the spleen and granulocytes or monocytes in liver, lung, kidney, and spleen (not shown). BCG infection induces the formation of larger granulomas, particularly in liver, lung, and spleen, and occasionally in kidney and heart. IHC revealed iNOS expression exclusively in the epithelioid macrophages within those granulomas (Fig. 5M–P). Immunostaining of sections derived from iNOS-deficient TNF-, LPS-, Salmonella-, or BCG-treated mice was always negative (not shown). Identical results were obtained with anti-iNOS antibodies from Chemicon and Transduction Laboratories. The use of the antiiNOS Ab from Santa Cruz Biotechnology was abroE1624

Vol. 20

November 2006

gated, because of substantial aspecific binding on sections from iNOS⫺/⫺ mice (not shown).

DISCUSSION Increased production of NO has been demonstrated in both experimental and clinical infection, inflammation, and sepsis (9, 12, 40 – 42). Although NO is clearly involved in the development of hypotension during septic shock, many other harmful as well as beneficial effects of NO may play an important role. Some of these dual effects of NO may simply relate to its production by different NOS isoforms. However, also differences in cellular and/or temporal expression patterns might be critical. Systemic NOx⫺ production in LPS-treated mice was originally attributed to peritoneal macrophages (35). Later on it became clear that LPS could also induce (iNOS-derived) NO production in other hematopoietic cells like neutrophils (36), eosinophils (37), and lymphocytes (43). Since both WBCs and NO have important roles in immunity and inflammation (3, 44), it was accepted for a long time that WBCs are the main source of systemic NO in infectious and inflammatory diseases, including septic shock. This belief still generally prevails, despite the fact that also a wide range of parenchymal cells, such as hepatocytes, epithelial, vascular smooth muscle, and

The FASEB Journal

BULTINCK ET AL.

endothelial cells have the ability to express iNOS, both in vitro after LPS and/or cytokine stimulation (25, 26, 45– 47), as well as in vivo during endotoxic and/or septic shock (26 –30, 48). Moreover, despite the fact that LPS-treated Tyk2⫺/⫺ macrophages are unable to produce NO, identical NO levels can be found in circulation of endotoxic Tyk2⫺/⫺ and WT mice, already suggesting that systemic LPS-induced NO does not originate from macrophages (49). More recently, LPS treatment of iNOS-chimeric mice indicated that indeed parenchymal cells are the most important source of iNOS during endotoxemia (22). However, these results were not confirmed in an infectious shock model, which is not only clinically more relevant, but may also affect circulating hematopoietic cells to a much greater extent than LPS. In addition, the cellular source of parenchymal iNOS was not determined (ND). In our study, we found that even in an infectious and lethal septic shock model, the hematopoietic cell population does not contribute significantly to systemic NO production. We also identified iNOS-expressing hepatocytes, paneth cells, and epithelial cells of the gut and kidney as the parenchymal source of systemic NO after TNF, LPS or S. enteritidis challenge. NO produced by hepatocytes and paneth cells can easily reach the circulation via sinusoids and intestinal venules, respectively. Apical intestinal epithelium produces NO that is likely to diffuse into the intestinal lumen. Still, intestinal villi contain a complex network of blood vessels designed to absorb nutrients from the lumen, making it possible for NO to reach the blood via this way. Since liver and intestinal tract represent two of the largest organs within the mammalian body, they correspond to a vast amount of iNOS-expressing cells. iNOS-expressing hepatocytes, Kupffer cells, and gastro-intestinal epithelial cells have been described in endotoxic (26, 28 –30, 50) and bacteremic (51) rodents. Also the specific canalicular staining pattern for iNOS within hepatocytes has been observed before (50). Interestingly, we also found distinctive iNOS expression in paneth cells during TNF, LPS, or bacterial inflammation, which has never been reported before. It is not until recently that great interest arose for these important cells of innate mucosal immunity. Several antimicrobial peptides and proteins are released by paneth cells on LPS challenge (52, 53). The well-defined function of iNOS-derived NO in innate immunity (3) makes iNOS induction in paneth cells very plausible. As evidenced by our immunohistochemical study and that of others (28 –30, 54), LPS treatment of rodents induces the expression of iNOS in monocytes, in infiltrated and resident macrophages, and in granulocytes. However, and surprisingly, those cells do not contribute significantly to systemic NO production, as demonstrated by our BMT experiments. Several possible explanations may be envisioned. It could be that those few cells are simply insufficient to contribute significantly. Other explanations could be that the NO is locally captured in a rapid reaction like that with other radicals SHOCK-ASSOCIATED iNOS AND SYSTEMIC NO IN MICE

or that iNOS expression in these WBCs does not correlate with NO production. Indeed, when L-arg is limited, for example, iNOS produces superoxide rather than NO in a so-called “uncoupled ” reaction (55). It has also been reported that in an inflammatory setting, WBCs produce NO only after engagement with integrins and therefore need to infiltrate into tissue first (56, 57). In contrast to rodents, human macrophage cell lines are incapable of inducing iNOS activity in response to LPS, not even in combination with cytokines (57–59). Moreover, iNOS activity can hardly be detected in macrophages isolated from septic patients (60 – 62). Nevertheless, in both species, systemic NO levels are elevated during infection, inflammation, and septic shock. It has always been believed that this discrepancy could be attributed to an inherent difference between mice and men. However, our data indicate that the role of monocytes/macrophages as the source of shockinducing NO is much less prominent than generally assumed. This is further supported by a study of Clark and colleagues (63), which clearly shows parenchymal iNOS expression in human malaria and sepsis. Whereas parenchymal cells are the major source of systemic NOx⫺ during TNF-, LPS- or Salmonella-induced shock, hematopoietic cells represent the main source of systemic NOx⫺ during BCG infection. IHC revealed the epithelioid macrophages of the BCG-induced granulomas as the only iNOS source, which is in correspondence with previous data obtained from Mycobacteriuminfected rodents and men (64 – 66). In addition, during other intracellular infectious diseases such as Leishmania, Listeria, or Toxoplasma infection, the expression of iNOS is restricted to the hematopoietic lineage (67– 69). In conclusion, our in vivo study revealed a differential role for hematopoietic cells in systemic NO production. Although hematopoietic cells are crucial for the production of systemic NO during granulomatous inflammation, they do not contribute at all to the production of circulating NO after systemic TNF or LPS challenge. In addition, in a bacteremic model of septic shock, cells of hematopoietic origin did not significantly add to the enormous systemic NO production. Instead, hepatocytes, paneth cells, and epithelial cells of gut and kidney represent the parenchymal source of systemic NO. Interestingly, our data also suggest that differences between in vitro and in vivo observations, rather than species diversities, might be responsible for the long-standing contradiction between the significant amounts of systemic NO detectable in a wide range of human inflammatory diseases vs. the inability of human monocytes/macrophages to produce NO in vitro (70, 71). The authors thank Linda Van Geert, Joris De Backer, and Geert Versporten for animal care; Dr. H. Bluethmann (Roche, Basel) for the generous gift of TNF-R1⫺/⫺ mice; and Dr. C. Cuvelier for discussions about histology. This research was supported by grants from the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen, the Interuniversity Attraction E1625

Pole program and Universiteit Gent-Geconcerteerde Onderzoeks Acties.

21. 22.

REFERENCES 1. 2. 3. 4. 5. 6.

7.

8.

9.

10.

11.

12. 13. 14. 15. 16. 17. 18.

19.

20.

23.

Bateman, R. M., Sharpe, M. D., and Ellis, C. G. (2003) Benchto-bedside review: microvascular dysfunction in sepsis– hemodynamics, oxygen transport, and nitric oxide. Crit. Care 7, 359 –373 Stuehr, D. J. (1999) Mammalian nitric oxide synthases. Biochim. Biophys. Acta 1411, 217–230 Bogdan, C. (2001) Nitric oxide and the immune response. Nat. Immunol. 2, 907–916 Coleman, J. W. (2001) Nitric oxide in immunity and inflammation. Int. Immunopharmacol. 1, 1397–1406 Colasanti, M., and Suzuki, H. (2000) The dual personality of NO. Trends Pharmacol. Sci. 21, 249 –252 Kilbourn, R. G., Gross, S. S., Jubran, A., Adams, J., Griffith, O. W., Levi, R., and Lodato, R. F. (1990) NG-methyl-L-arginine inhibits tumor necrosis factor-induced hypotension: implications for the involvement of nitric oxide. Proc. Natl. Acad. Sci. U. S. A. 87, 3629 –3632 Peschon, J. J., Torrance, D. S., Stocking, K. L., Glaccum, M. B., Otten, C., Willis, C. R., Charrier, K., Morrissey, P. J., Ware, C. B., and Mohler, K. M. (1998) TNF receptor-deficient mice reveal divergent roles for p55 and p75 in several models of inflammation. J. Immunol. 160, 943–952 Rothe, J., Lesslauer, W., Lotscher, H., Lang, Y., Koebel, P., Kontgen, F., Althage, A., Zinkernagel, R., Steinmetz, M., and Bluethmann, H. (1993) Mice lacking the tumour necrosis factor receptor 1 are resistant to TNF-mediated toxicity but highly susceptible to infection by Listeria monocytogenes. Nature 364, 798 – 802 Cauwels, A., Van Molle, W., Janssen, B., Everaerdt, B., Huang, P., Fiers, W., and Brouckaert, P. (2000) Protection against TNFinduced lethal shock by soluble guanylate cyclase inhibition requires functional inducible nitric oxide synthase. Immunity. 13, 223–231 Cauwels, A., Bultinck, J., and Brouckaert, P. (2005) Dual role of endogenous nitric oxide in tumor necrosis factor shock: induced NO tempers oxidative stress. Cell Mol. Life Sci. 62, 1632–1640 Sands, K. E., Bates, D. W., Lanken, P. N., Graman, P. S., Hibberd, P. L., Kahn, K. L., Parsonnet, J., Panzer, R., Orav, E. J., Snydman, D. R., et al. (1997) Epidemiology of sepsis syndrome in 8 academic medical centers. JAMA 278, 234 –240 Feihl, F., Waeber, B., and Liaudet, L. (2001) Is nitric oxide overproduction the target of choice for the management of septic shock? Pharmacol. Ther. 91, 179 –213 Thiemermann, C. (1997) Nitric oxide and septic shock. Gen. Pharmacol. 29, 159 –166 Ceppi, E. D., Smith, F. S., and Titheradge, M. A. (1997) Nitric oxide, sepsis and liver metabolism. Biochem. Soc. Trans. 25, 929 –934 Knowles, R. G., and Moncada, S. (1992) Nitric oxide as a signal in blood vessels. Trends Biochem. Sci. 17, 399 – 402 Kilbourn, R. G., Traber, D. L., and Szabo, C. (1997) Nitric oxide and shock. Dis. Mon. 43, 277–348 Johnson, M. L., and Billiar, T. R. (1998) Roles of nitric oxide in surgical infection and sepsis. World J. Surg. 22, 187–196 Avontuur, J. A., Tutein Nolthenius, R. P., van Bodegom, J. W., and Bruining, H. A. (1998) Prolonged inhibition of nitric oxide synthesis in severe septic shock: a clinical study. Crit. Care Med. 26, 660 – 667 Lopez, A., Lorente, J. A., Steingrub, J., Bakker, J., McLuckie, A., Willatts, S., Brockway, M., Anzueto, A., Holzapfel, L., Breen, D., et al. (2004) Multiple-center, randomized, placebo-controlled, double-blind study of the nitric oxide synthase inhibitor 546C88: effect on survival in patients with septic shock. Crit. Care Med. 32, 21–30 Laubach, V. E., Foley, P. L., Shockey, K. S., Tribble, C. G., and Kron, I. L. (1998) Protective roles of nitric oxide and testosterone in endotoxemia: evidence from NOS-2-deficient mice. Am. J. Physiol. 275, 2211–2218

E1626

Vol. 20

November 2006

24.

25.

26.

27. 28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Andrew, P. J., and Mayer, B. (1999) Enzymatic function of nitric oxide synthases. Cardiovasc. Res. 43, 521–531 Hickey, M. J., Sihota, E., Amrani, A., Santamaria, P., Zbytnuik, L. D., Ng, E. S., Ho, W., Sharkey, K. A., and Kubes, P. (2002) Inducible nitric oxide synthase (iNOS) in endotoxemia: chimeric mice reveal different cellular sources in various tissues. FASEB J. 16, 1141–1143 Poon, B. Y., Raharjo, E., Patel, K. D., Tavener, S., and Kubes, P. (2003) Complexity of inducible nitric oxide synthase: cellular source determines benefit versus toxicity. Circulation 108, 1107– 1112 Veihelmann, A., Hofbauer, A., Krombach, F., Dorger, M., Maier, M., Refior, H. J., and Messmer, K. (2002) Differential function of nitric oxide in murine antigen-induced arthritis. Rheumatology 41, 509 –517 Nussler, A. K., Geller, D. A., Sweetland, M. A., Di Silvio, M., Billiar, T. R., Madariaga, J. B., Simmons, R. L., and Lancaster, J. R., Jr. (1993) Induction of nitric oxide synthesis and its reactions in cultured human and rat hepatocytes stimulated with cytokines plus LPS. Biochem. Biophys. Res. Commun. 194, 826 – 835 Geller, D. A., Di Silvio, M., Nussler, A. K., Wang, S. C., Shapiro, R. A., Simmons, R. L., and Billiar, T. R. (1993) Nitric oxide synthase expression is induced in hepatocytes in vivo during hepatic inflammation. J. Surg. Res. 55, 427– 432 Knowles, R. G., Merrett, M., Salter, M., and Moncada, S. (1990) Differential induction of brain, lung and liver nitric oxide synthase by endotoxin in the rat. Biochem. J. 270, 833– 836 Sato, K., Miyakawa, K., Takeya, M., Hattori, R., Yui, Y., Sunamoto, M., Ichimori, Y., Ushio, Y., and Takahashi, K. (1995) Immunohistochemical expression of inducible nitric oxide synthase (iNOS) in reversible endotoxic shock studied by a novel monoclonal antibody against rat iNOS. J. Leukoc. Biol. 57, 36 – 44 Buttery, L. D., Evans, T. J., Springall, D. R., Carpenter, A., Cohen, J., and Polak, J. M. (1994) Immunochemical localization of inducible nitric oxide synthase in endotoxin-treated rats. Lab. Invest. 71, 755–764 Cook, H. T., Bune, A. J., Jansen, A. S., Taylor, G. M., Loi, R. K., and Cattell, V. (1994) Cellular localization of inducible nitric oxide synthase in experimental endotoxic shock in the rat. Clin. Sci. 87, 179 –186 Laubach, V. E., Shesely, E. G., Smithies, O., and Sherman, P. A. (1995) Mice lacking inducible nitric oxide synthase are not resistant to lipopolysaccharide-induced death. Proc. Natl. Acad. Sci. U. S. A. 92, 10688 –10692 Van Snick, J., Cayphas, S., Vink, A., Uyttenhove, C., Coulie, P. G., Rubira, M. R., and Simpson, R. J. (1986) Purification and NH2-terminal amino acid sequence of a T-cell-derived lymphokine with growth factor activity for B-cell hybridomas. Proc. Natl. Acad. Sci. U. S. A. 83, 9679 –9683 Green, L. C., Wagner, D. A., Glogowski, J., Skipper, P. L., Wishnok, J. S., and Tannenbaum, S. R. (1982) Analysis of nitrate, nitrite, and (15N)nitrate in biological fluids. Anal. Biochem. 126, 131–138 Cauwels, A., Brouckaert, P., Grooten, J., Huang, S., Aguet, M., and Fiers, W. (1995) Involvement of IFN-gamma in Bacillus Calmette-Guerin-induced but not in tumor-induced sensitization to TNF-induced lethality. J. Immunol. 154, 2753–2763 Stuehr, D. J., and Marletta, M. A. (1985) Mammalian nitrate biosynthesis: mouse macrophages produce nitrite and nitrate in response to Escherichia coli lipopolysaccharide. Proc. Natl. Acad. Sci. U. S. A. 82, 7738 –7742 Kolls, J., Xie, J., LeBlanc, R., Malinski, T., Nelson, S., Summer, W., and Greenberg, S. S. (1994) Rapid induction of messenger RNA for nitric oxide synthase II in rat neutrophils in vivo by endotoxin and its suppression by prednisolone. Proc. Soc. Exp. Biol. Med. 205, 220 –229 Oliveira, S. H., Fonseca, S. G., Romao, P. R., Figueiredo, F., Ferreira, S. H., and Cunha, F. Q. (1998) Microbicidal activity of eosinophils is associated with activation of the arginine-NO pathway. Parasite Immunol. 20, 405– 412 Alam, M. S., Akaike, T., Okamoto, S., Kubota, T., Yoshitake, J., Sawa, T., Miyamoto, Y., Tamura, F., and Maeda, H. (2002) Role of nitric oxide in host defense in murine salmonellosis as a function of its antibacterial and antiapoptotic activities. Infect. Immun. 70, 3130 –3142

The FASEB Journal

BULTINCK ET AL.

39.

40. 41. 42.

43. 44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55. 56.

Mastroeni, P., Vazquez-Torres, A., Fang, F. C., Xu, Y., Khan, S., Hormaeche, C. E., and Dougan, G. (2000) Antimicrobial actions of the NADPH phagocyte oxidase and inducible nitric oxide synthase in experimental salmonellosis. II. Effects on microbial proliferation and host survival in vivo. J. Exp. Med. 192, 237–248 Lai, C. S., and Komarov, A. M. (1994) Spin trapping of nitric oxide produced in vivo in septic-shock mice. FEBS Lett. 345, 120 –124 Evans, T., Carpenter, A., Kinderman, H., and Cohen, J. (1993) Evidence of increased nitric oxide production in patients with the sepsis syndrome. Circ. Shock 41, 77– 81 Tracey, W. R., Tse, J., and Carter, G. (1995) Lipopolysaccharideinduced changes in plasma nitrite and nitrate concentrations in rats and mice: pharmacological evaluation of nitric oxide synthase inhibitors. J. Pharmacol. Exp. Ther. 272, 1011–1015 Victor, V. M., Rocha, M., and De la Fuente, M. (2003) N-acetylcysteine protects mice from lethal endotoxemia by regulating the redox state of immune cells. Free Radic. Res. 37, 919 –929 Arkovitz, M. S., Wispe, J. R., Garcia, V. F., and Szabo, C. (1996) Selective inhibition of the inducible isoform of nitric oxide synthase prevents pulmonary transvascular flux during acute endotoxemia. J. Pediatr. Surg. 31, 1009 –1015 Asano, K., Chee, C. B., Gaston, B., Lilly, C. M., Gerard, C., Drazen, J. M., and Stamler, J. S. (1994) Constitutive and inducible nitric oxide synthase gene expression, regulation, and activity in human lung epithelial cells. Proc. Natl. Acad. Sci. U. S. A. 91, 10089 –10093 Robertson, D. A., Hughes, G. A., and Lyles, G. A. (2004) Expression of inducible nitric oxide synthase in cultured smooth muscle cells from rat mesenteric lymphatic vessels. Microcirculation 11, 503–515 Wu, F., Cepinskas, G., Wilson, J. X., and Tyml, K. (2001) Nitric oxide attenuates but superoxide enhances iNOS expression in endotoxin- and IFNgamma-stimulated skeletal muscle endothelial cells. Microcirculation 8, 415– 425 Hoffman, R. A., Zhang, G., Nussler, N. C., Gleixner, S. L., Ford, H. R., Simmons, R. L., and Watkins, S. C. (1997) Constitutive expression of inducible nitric oxide synthase in the mouse ileal mucosa. Am. J. Physiol. 272, G383–392 Karaghiosoff, M., Neubauer, H., Lassnig, C., Kovarik, P., Schindler, H., Pircher, H., McCoy, B., Bogdan, C., Decker, T., and Brem, G., et al. (2000) Partial impairment of cytokine responses in Tyk2deficient mice. Immunity 13, 549 –560 Vos, T. A., Gouw, A. S., Klok, P. A., Havinga, R., van Goor, H., Huitema, S., Roelofsen, H., Kuipers, F., Jansen, P. L., and Moshage, H. (1997) Differential effects of nitric oxide synthase inhibitors on endotoxin-induced liver damage in rats. Gastroenterology 113, 1323–1333 Parmely, M. J., Wang, F., and Wright, D. (2001) Gamma interferon prevents the inhibitory effects of oxidative stress on host responses to Escherichia coli infection. Infect. Immun. 69, 2621–2629 Ayabe, T., Satchell, D. P., Wilson, C. L., Parks, W. C., Selsted, M. E., and Ouellette, A. J. (2000) Secretion of microbicidal alpha-defensins by intestinal Paneth cells in response to bacteria. Nat. Immunol. 1, 113–118 Ayabe, T., Satchell, D. P., Pesendorfer, P., Tanabe, H., Wilson, C. L., Hagen, S. J., and Ouellette, A. J. (2002) Activation of Paneth cell alpha-defensins in mouse small intestine. J. Biol. Chem. 277, 5219 –5228 Stumm, M. M., D’Orazio, D., Sumanovski, L. T., Martin, P. Y., Reichen, J., and Sieber, C. C. (2002) Endothelial, but not the inducible, nitric oxide synthase is detectable in normal and portal hypertensive rats. Liver 22, 441– 450 Xia, Y., Roman, L. J., Masters, B. S., and Zweier, J. L. (1998) Inducible nitric-oxide synthase generates superoxide from the reductase domain. J. Biol. Chem. 273, 22635–22639 Miles, A. M., Owens, M. W., Milligan, S., Johnson, G. G., Fields, J. Z., Ing, T. S., Kottapalli, V., Keshavarzian, A., and Grisham,

SHOCK-ASSOCIATED iNOS AND SYSTEMIC NO IN MICE

57.

58.

59.

60.

61.

62.

63.

64.

65.

66. 67.

68.

69.

70. 71.

M. B. (1995) Nitric oxide synthase in circulating vs. extravasated polymorphonuclear leukocytes. J. Leukoc. Biol. 58, 616 – 622 Weinberg, J. B., Misukonis, M. A., Shami, P. J., Mason, S. N., Sauls, D. L., Dittman, W. A., Wood, E. R., Smith, G. K., McDonald, B., Bachus, K. E., et al. (1995) Human mononuclear phagocyte inducible nitric oxide synthase (iNOS): analysis of iNOS mRNA, iNOS protein, biopterin, and nitric oxide production by blood monocytes and peritoneal macrophages. Blood 86, 1184 –1195 Jungi, T. W., Adler, H., Adler, B., Thony, M., Krampe, M., and Peterhans, E. (1996) Inducible nitric oxide synthase of macrophages. Present knowledge and evidence for species-specific regulation. Vet. Immunol. Immunopathol. 54, 323–330 Arias, M., Zabaleta, J., Rodriguez, J. I., Rojas, M., Paris, S. C., and Garcia, L. F. (1997) Failure to induce nitric oxide production by human monocyte-derived macrophages. Manipulation of biochemical pathways. Allergol. Immunopathol. (Madr). 25, 280 –288 Hersch, M., Scott, J. A., Izbicki, G., McCormack, D., Cepinkas, G., Ostermann, M., and Sibbald, W. J. (2005) Differential inducible nitric oxide synthase activity in circulating neutrophils vs. mononuclears of septic shock patients. Intensive Care Med. 31, 1132–1135 Kobayashi, A., Hashimoto, S., Kooguchi, K., Kitamura, Y., Onodera, H., Urata, Y., and Ashihara, T. (1998) Expression of inducible nitric oxide synthase and inflammatory cytokines in alveolar macrophages of ARDS following sepsis. Chest 113, 1632–1639 Annane, D., Sanquer, S., Sebille, V., Faye, A., Djuranovic, D., Raphael, J. C., Gajdos, P., and Bellissant, E. (2000) Compartmentalised inducible nitric-oxide synthase activity in septic shock. Lancet 355, 1143–1148 Clark, I. A., Awburn, M. M., Whitten, R. O., Harper, C. G., Liomba, N. G., Molyneux, M. E., and Taylor, T. E. (2003) Tissue distribution of migration inhibitory factor and inducible nitric oxide synthase in falciparum malaria and sepsis in African children. Malar. J. 2, 6 Ehlers, S., Kutsch, S., Benini, J., Cooper, A., Hahn, C., Gerdes, J., Orme, I., Martin, C., and Rietschel, E. T. (1999) NOS2derived nitric oxide regulates the size, quantity and quality of granuloma formation in Mycobacterium avium-infected mice without affecting bacterial loads. Immunology 98, 313–323 MacMicking, J. D., North, R. J., LaCourse, R., Mudgett, J. S., Shah, S. K., and Nathan, C. F. (1997) Identification of nitric oxide synthase as a protective locus against tuberculosis. Proc. Natl. Acad. Sci. U. S. A. 94, 5243–5248 Schon, T., Elmberger, G., Negesse, Y., Pando, R. H., Sundqvist, T., and Britton, S. (2004) Local production of nitric oxide in patients with tuberculosis. Int. J. Tuberc. Lung Dis. 8, 1134 –1137 Schluter, D., Kaefer, N., Hof, H., Wiestler, O. D., and DeckertSchluter, M. (1997) Expression pattern and cellular origin of cytokines in the normal and Toxoplasma gondii-infected murine brain. Am. J. Pathol. 150, 1021–1035 Stenger, S., Thuring, H., Rollinghoff, M., and Bogdan, C. (1994) Tissue expression of inducible nitric oxide synthase is closely associated with resistance to Leishmania major. J. Exp. Med. 180, 783–793 Jungi, T. W., Pfister, H., Sager, H., Fatzer, R., Vandevelde, M., and Zurbriggen, A. (1997) Comparison of inducible nitric oxide synthase expression in the brains of Listeria monocytogenesinfected cattle, sheep, and goats and in macrophages stimulated in vitro. Infect. Immun. 65, 5279 –5288 Denis, M. (1994) Human monocytes/macrophages: NO or no NO? J. Leukoc. Biol. 55, 682– 684 Albina, J. E. (1995) On the expression of nitric oxide synthase by human macrophages. Why no NO? J. Leukoc. Biol. 58, 643– 649 Received for publication February 3, 2006. Accepted for publication June 23, 2006.

E1627

The FASEB Journal • FJ Express Summary

Systemic NO production during (septic) shock depends on parenchymal and not on hematopoietic cells: in vivo iNOS expression pattern in (septic) shock Jennyfer Bultinck,* Patrick Sips,* Luc Vakaet,† Peter Brouckaert,* and Anje Cauwels*,1 *Department for Molecular Biomedical Research, Ghent University and VIB, Ghent, Belgium; and † Department of Radiotherapy, Ghent University Hospital, Ghent, Belgium To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.06-5798fje

SPECIFIC AIMS Three different isoforms of NO synthase (NOS) exist: endothelial (e), neuronal (n), and inducible (i) NOS. As opposed to the constitutive enzymes eNOS and nNOS, iNOS is synthesized de novo during inflammation and produces large amounts of NO over prolonged periods of time. Septic shock is the leading cause of death in noncoronary intensive care units. Lipopolysaccharide (LPS), a constituent of the outer membrane of Gram-negative bacteria, mimics most of the septic effects and is widely used as an experimental model for septic shock. NO plays a controversial role in the pathophysiology of (septic) shock. Since it causes hypotension and tissue damage, NO may be considered detrimental to the host. However, NO may also confer protection to the host as an anti-inflammatory, antioxidant, and/or anti-apoptotic agent. This ambiguity may relate to its isoform-specific production but also to differences in cellular and/or temporal iNOS expression. The iNOS enzyme was originally identified in macrophages. Inflammatory stimuli may induce the production of (iNOS-derived) NO in several hematopoietic cell lines, such as macrophages, neutrophils, eosinophils and lymphocytes. Hence, and because of their well-known role in inflammatory diseases such as septic shock, immune cells have generally been considered the principal in vivo source of systemic high-output NO. To verify this, we performed bone marrow transplantation (BMT) experiments between wild-type (WT) and iNOS- or TNF-R1-deficient mice and challenged these chimeric mice with shock-inducing TNF, LPS, or live Salmonella. In comparison, we also infected mice with BCG, which causes a granulomatous inflammatory reaction. In addition, the in vivo expression pattern of iNOS mRNA and protein was investigated. PRINCIPAL FINDINGS 1. Parenchymal cells are the only source of systemic NOxⴚ in TNF- or LPS-induced shock and in a bacteremic model of septic shock To investigate the involvement of hematopoietic cells in systemic NOx⫺ production after TNF or LPS, 0892-6638/06/0020-2363 © FASEB

irradiated iNOS⫺/⫺ mice were reconstituted with WT BM cells and vice versa. As a control, WT and iNOS⫺/⫺ mice were, after irradiation, transplanted with homologous BM cells. To check the BMT effectiveness, the genotype of circulating WBCs was determined. As shown in Fig. 1 (insert), virtually all the hematopoietic cells of the reconstituted mice were of the donor genotype. Three days later, the mice were intravenously injected with lethal doses of TNF, E. coli, or S. abortus equi LPS. Systemic interleukin (IL)-6 levels 8 h after challenge were comparable (not shown). However, TNF or LPS did not induce any detectable NOx⫺ in circulation of iNOS⫺/⫺ mice with an iNOS⫹/⫹ hematopoietic cell population, while they clearly did in mice that express iNOS in every cell but the hematopoietic cells (Fig. 1A). Therefore, in TNF- or LPS-challenged mice parenchymal cells are fully responsible for the systemic NOx⫺ production. For TNF, we confirmed this result with a BMT between WT and TNF-R1⫺/⫺ mice (not shown). To substantiate the insignificance of immune cells in NOx⫺ production during shock, we set up an infectious Salmonella enteritidis model. As reported before, iNOS⫺/⫺ mice were more susceptible to Salmonella infection (not shown). Between 18 to 30 h after infection, significant amounts of circulating NOx⫺ were present in Salmonella-infected WT to WT mice, as well as in WT mice with iNOS⫺/⫺ BM. In contrast, however, only very low and insignificant levels of NOx⫺ were found in chimeric mice with iNOS-negative parenchymal cells (not shown), indicating that in a bacteremic model of septic shock there is also a predominant role for parenchymal cells in systemic NOx⫺ production.

1

Correspondence: Department for Molecular Biomedical Research, Ghent University/VIB, Technologiepark 927, Ghent 9052, Belgium. E-mail: [email protected] doi: 10.1096/fj.06-5798fje 2363

ing BCG infection. Only in mice with a WT parenchymal compartment, an extra TNF challenge caused an additional increase in NOx⫺ production (not shown). 3. Expression patterns of iNOS

Figure 1. Serum NOx⫺ levels in TNF- or LPS-treated WT, iNOS⫺/⫺, and BMT mice. Mice were injected iv with PBS, 12.5 ␮g TNF, 300 ␮g E. coli LPS, or 150 ␮g S. abortus equi LPS. Serum NOx⫺ levels were determined 8 h later. ***P ⬍ 0.001, as compared with corresponding treatment in the WT to WT group. Plotted are means ⫾ sd; numbers above bars are number of mice. Insert: PCR reactions, specific for the WT or deficient iNOS allele, were performed on WBC DNA of BMT mice. I: WT to WT, II: iNOS⫺/⫺ to WT, III: WT to iNOS⫺/⫺, IV: iNOS⫺/⫺ to iNOS⫺/⫺, M: marker.

2. Hematopoietic cells are responsible for systemic NOxⴚ production during granulomatous inflammation In an analogous BMT experiment between WT and iNOS⫺/⫺ mice, we found that hematopoietic cells are solely responsible for systemic NOx⫺ production dur-

To elucidate the induction pattern of iNOS, we first determined the iNOS mRNA and protein levels in various organs via real time-quantitative polymerase chain reaction (PCR) and WB analysis, respectively. After BCG infection, iNOS was strongly detected in liver, lung, and spleen and weakly in kidney and heart, corresponding with the organs in which granulomas could be detected (not shown). After TNF or LPS, iNOS was strongly detected in liver, kidney, jejunum, and colon. To examine the cellular expression pattern of iNOS, we performed immunohistochemistry (IHC). Identical iNOS expression patterns were observed after TNF or LPS (Fig. 2A–H), although less pronounced after LPS, correlating with the systemic NOx⫺ levels. Positive iNOS staining was detected in hepatocytes, often specifically at the canalicular membrane (Fig. 2A, E). In the kidney, epithelial cells of pelvis (Fig. 2B, F) and tubuli (not shown) stained positive. Throughout the gastrointestinal tract, apical epithelial cells clearly expressed iNOS (Fig. 2C, G), as well as the paneth cells of the jejunum (Fig. 2D, H). In addition, after LPS we also observed some iNOS-positive Kupffer cells and

Figure 2. iNOS-specific IHC. Organs were collected 8 h after 5 ␮g TNF or 150 ␮g S. abortus equi LPS, 24 h after 800 cfu of S. enteritidis, or 2 h after PBS in BCG-infected mice. Arrows in A indicate canalicular membranes of hepatocytes. Magnification ⫽ ⫻40 (J, N), ⫻60 (A, C, E, G), or ⫻100 (B, D, F, H, I, K, L, M, O, P). 2364

Vol. 20

November 2006

The FASEB Journal

BULTINCK ET AL.

Figure 3. Schematic presentation of in vivo cellular source of iNOS-derived NO in various inflammatory settings. BM transplantations between WT and iNOS⫺/⫺ mice revealed that hematopoietic cells are fully responsible for systemic NO production during BCG infection. However, after TNF, LPS or Salmonella treatment, most, if not all, systemic NO originates from non-hematopoietic cells, which is in contrast with the general assumption, based on in vitro studies, that macrophages are the principle source of iNOS-derived NO during inflammatory shock.

occasional granulocytes and/or monocytes in liver, lung, spleen, kidney and heart (not shown). Nevertheless, despite their ability to express iNOS protein in vivo, our BMT experiments indicate that these WBCs do not contribute significantly to the massive systemic NOx⫺ production. One day after the inoculation of Salmonella enteritidis, similar iNOS expression patterns were detected as for LPS (Fig. 2I–L). Salmonella infection also occasionally induced some small iNOS-positive granulomas (not shown). BCG infection induces the formation of larger granulomas, particularly in liver, lung, and spleen, and occasionally in kidney and heart. IHC revealed iNOS expression exclusively in the epithelioid macrophages within those granulomas (Fig. 2M–P).

CONCLUSIONS AND SIGNIFICANCE For a long time, it has been accepted that leukocytes are the main source of systemic NO in infectious and inflammatory diseases, including septic shock. However, we here provide conclusive in vivo evidence to

SHOCK-ASSOCIATED INOS AND SYSTEMIC NO IN MICE

refute this general assumption. Our study clearly establishes a predominant role for parenchymal cells in systemic NOx⫺ production during endotoxic and bacteremic (septic) shock. We also identified iNOS-expressing hepatocytes, paneth cells, and epithelial cells of the gut and kidney as the parenchymal source of systemic NO after TNF, LPS, or Salmonella infection. Since liver and intestinal tract represent two of the largest organs within the mammalian body, they correspond with a vast amount of iNOS-expressing cells. iNOS-expressing hepatocytes (including the specific canalicular staining), Kupffer cells, and gastrointestinal epithelial cells have been described before. Interestingly, we also found distinctive iNOS expression in paneth cells, which has never been reported before. Paneth cells release several antimicrobial peptides and proteins and thus have an important role in innate mucosal immunity. The well-defined function of iNOS-derived NO in innate immunity makes iNOS induction in paneth cells very plausible. As evidenced by our immunohistochemical study and that of others, LPS treatment of rodents induces the expression of iNOS in monocytes, macrophages, and granulocytes. However, and surprisingly, those cells do not contribute significantly to systemic NO production, as taught by our BMT experiments. Several possible explanations may be envisioned. It could be that those few cells are simply insufficient to contribute significantly. Other explanations could be that the NO is locally captured in a rapid reaction like that with other radicals or that iNOS expression in these WBCs does not correlate with NO production. Indeed, when l-arg is limited, for example, iNOS produces superoxide rather than NO in a so-called “uncoupled” reaction. Whereas parenchymal cells are the major source of systemic NO during TNF-, LPS-, or Salmonella-induced shock, hematopoietic cells (more specifically the granulomar epithelioid macrophages) represent the main source of systemic NO during BCG infection. In conclusion, our in vivo study reveals a differential role for hematopoietic cells in systemic NO production. It may be interesting to add that the generation of induced NO by human monocytes/macrophages has long been a subject of great controversy. Although humans produce significant quantities of NO during infection or inflammation, in vitro treatment of human monocytes/macrophages failed to elicit NO release, in contrast to murine macrophages. It has always been believed that this discrepancy could be attributed to an inherent difference between mice and men. Our data indicate that differences between in vitro and in vivo observations, rather than species diversities, are responsible for this discrepancy and that in both species the role of monocytes/macrophages as the source of shockinducing NO is much less prominent than generally assumed.

2365