Hormones, Growth Factors, and Cell Signaling

Inducible Nitric Oxide Synthase Can Be Induced in the Absence of Active Nuclear Factor-κB in Rat Mesangial Cells

Involvement of the Janus Kinase 2 Signaling Pathway

OSAMU NAKASHIMA, YOSHIO TERADA, SEIJI INOSHITA, MICHIO KUWAHARA, SEI SASAKI and FUMIAKI MARUMO
JASN April 1999, 10 (4) 721-729;

This article requires a subscription to view the full text. If you have a subscription you may use the login form below to view the article. Access to this article can also be purchased.

Abstract

Abstract. The contribution of nuclear factor-κB (NF-κB) and interferon-γ (IFN-γ) signaling to nitric oxide generation is not completely understood. The effect of NF-κB release and its inhibition on nitrite production and the involvement of Janus kinase 2 (JAK2) in inducible nitric oxide synthase (iNOS) induction were investigated. The following assays were performed. (1) Nitrite produced by rat mesangial cells in primary culture was measured in incubations with tumor necrosis factor-α (TNF-α) or lipopolysaccharide (LPS), with or without IFN-γ. Cells were stimulated with TNF-α or LPS plus IFN-γ in the presence of NF-κB inhibitors, herbimycin A (HerA), or the more specific JAK2 inhibitor AG490. (2) Immunoblotting was performed against the p65 and p50 subunits of NF-κB and iNOS. (3) Electrophoretic mobility shift assays were performed against NF-κB in the presence of NF-κB inhibitors or AG490. (4) iNOS promoter activity was measured in the presence of AG490 or JAK2 antisense oligonucleotides. TNF-α or LPS alone did not induce nitrite production, but with IFN-γ these compounds did induce nitrite production. Pyrrolidine dithiocarbamate (PDTC), N-acetyl-L-cysteine, dexamethasone (Dex), HerA, and AG490 partially inhibited LPS/IFN-γ- or TNF-α/IFN-γ-induced nitrite production. p65 was inhibited by the three NF-κB inhibitors described above, whereas p50 was not. PDTC and Dex completely inhibited the p65/p50 heterodimer, but HerA and AG490 had little effect on p65/p50. AG490 and JAK2 antisense oligonucleotides suppressed iNOS promoter activity. It can be concluded that (1) iNOS can be induced without active NF-κB; (2) Dex, acetylsalicylic acid, and PDTC inhibit only p65; and (3) JAK2 is involved in iNOS induction, and the contribution of JAK2 to nitrite production is greater than that of NF-κB.

Lipopolysaccharide (LPS) and inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and interleukin-1β, stimulate nitric oxide (NO) production in a variety of cells (1,2,3,4,5). NO generation during inflammatory processes is associated not only with microbial and tumor cell killing (6,7), but also with destruction of normal tissue in rheumatoid arthritis (8) and refractory collapse of the vascular bed in septic shock (9). In chronic glomerulonephritis, mesangial cells produce relatively large amounts of NO, which may destroy renal tissue, worsen proteinuria, and produce deterioration of renal function. Although overproduction of NO is harmful to the body, complete inhibition of NO production is not beneficial because of the contribution of NO to homeostasis (10). NO has profound effects in the body, including vasorelaxation (11), inhibition of platelet aggregation (12), neurotransmission (12), and inhibition of sodium absorption at proximal tubules and cortical collecting ducts of the kidney (13,14). For example, administration of an NO synthase (NOS) inhibitor in rat chronic glomerulonephritis causes worsening of proteinuria (15), which may be attributable to nonselective inhibition of NOS. Modulation of NO generation in mesangial cells is of clinical significance. For the optimal regulation of NO production in mesangial cells, we must clarify which signaling cascade is most important for NO induction. The promoter region of the murine macrophage inducible NOS (iNOS) contains putative consensus sequences for TNF-α response element, nuclear factor-κB (NF-κB), γ-activated site, interferon-stimulated response element, activator protein-1, interferon-γ response element, nuclear factor for interleukin-6 expression, and octamer binding protein-1 (10). NF-κB, a transcriptional factor for inflammation-related proteins (16), has been shown to be the major transcriptional factor for iNOS induction in macrophages (9,17). Interferon-γ (IFN-γ), a cytokine that is secreted from activated T cells and macrophages, has been shown to enhance NO production (18). However, the role of NF-κB is not fully understood. NF-κB activated by LPS induces iNOS in mouse macrophages (18), whereas NF-κB generated by TNF-α fails to induce iNOS (19). LPS stimulates mouse macrophages to produce IFN-γ, which in turn may function in an autocrine manner to enhance NO production. We hypothesize that (1) NF-κB plays an important but not critical role and (2) the IFN-γ signaling cascade in concert with NF-κB is crucial for iNOS induction in rat mesangial cells (rMC).

We investigated the effects of NF-κB activation and inhibition on NO production and examined which IFN-γ signaling pathway is involved in iNOS induction. We demonstrated that (1) iNOS was induced even after total abolishment of active NF-κB, (2) inhibition of active NF-κB by dexamethasone (Dex), acetylsalicylic acid (ASA), or pyrrolidine dithiocarbamate (PDTC) was caused by suppression of only the p65 subunit of NF-κB, (3) the p50 subunit was constitutively expressed and was not inhibited by the aforementioned NF-κB inhibitors, (4) the effect of IFN-γ was not merely synergistic but was critical for iNOS induction, and (5) Janus kinase 2 (JAK2) played an essential role in IFN-γ signal transduction in rMC and the contribution of JAK2 to nitrite production was greater than that of NF-κB. These results suggest that a signaling pathway independent of NF-κB may exist for iNOS induction and that NF-κB activation is regulated by the p65 subunit.

Materials and Methods

Reagents

Recombinant murine TNF-α and herbimycin A (HerA) were purchased from Life Technologies (Grand Island, NY). LPS from Escherichia coli 0111 B4, mouse recombinant IFN-γ, and ASA were from Sigma Chemical Co. (St. Louis, MO). AG490 was from Calbiochem (La Jolla, CA). Antisense phosphorothioate oligonucleotides against JAK2 and random control oligonucleotides were designed and synthesized by Biognostik (Goettingen, Germany). PDTC was from Wako Pure Chemicals (Osaka, Japan). N-Acetyl-L-cysteine (NAC) was from Katayama Chemicals (Tokyo, Japan). Rabbit polyclonal antibody against the p65 subunit of NF-κB, mouse monoclonal antibody against phosphotyrosine-containing proteins, rabbit polyclonal antibody against JAK2, and rabbit polyclonal antibody against iNOS were from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit polyclonal antibody against the p50 subunit of NF-κB was from Upstate Biotechnology (Lake Placid, NY). The ECL kit for Western blot detection was from Amersham (Buckinghamshire, England). The gel shift kit was from Stratagene (La Jolla, CA). The immunoprecipitation kit was from Boehringer Mannheim (Mannheim, Germany). The β-galactosidase enzyme assay system was from Promega (Madison, WI). LLC-PK1 cells were purchased from the American Type Culture Collection (Rockville, MD).

Cell Culture

rMC were cultured from isolated rat glomeruli as described (20). Briefly, glomeruli obtained from the renal cortices of Sprague Dawley rats by sequential sieving were resuspended in RPMI 1640 buffered with 20 mM Hepes at pH 7.4 and supplemented with 20% fetal calf serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. Cells were cultured at 37°C with 5% CO2. rMC used were in early passages.

Stimulation of rMC for Nitrite Production

Subconfluent rMC in 24-well plates were cultured for 24 h in RPMI 1640 with 5% fetal calf serum, followed by incubation for 24 h with TNF-α or LPS, with or without IFN-γ, NF-κB inhibitors, HerA, or AG490.

Nitrite Measurements

The assay was based on the Griess reaction (21). Briefly, culture supernatant (100 μl) was mixed with an equal volume of Griess reagent (0.1% N-[1-naphthyl]ethylenediamine in water/1% sulfanilamide in 5% phosphoric acid, 1:1) and the absorbance was read at 540 nm. Nitrite concentrations were determined using sodium nitrite as a standard.

Preparation of Nuclear Extracts

Nuclear extracts were prepared as described previously, with minor modifications (22). Briefly, 90% confluent rMC were incubated with each treatment cocktail for 3.5 h at 37°C, collected using trypsin, and centrifuged. Cell pellets were resuspended in 400 μl of ice-cold buffer A (10 mM Hepes, pH 7.9, 10 mM KCl, 1 mM dithiothreitol [DTT], 0.5 mM phenylmethylsulfonyl fluoride [PMSF]). Then 25 μl of a 10% Nonidet P-40 solution was added, and the tubes were vigorously shaken for 30 s on an orbital shaker and centrifuged for 30 s in a microfuge. Nuclear pellets were resuspended in 100 μl of ice-cold buffer C (20 mM Hepes, pH 7.9, 400 mM NaCl, 1 mM ethylenediaminetetraacetate, 1 mM ethylene glycol bis[β-aminoethyl ether]-N,N,N′,N′-tetraacetate, 1 mM DTT, 1 mM PMSF, 1 μg/ml aprotinin, 1 μg/ml pepstatin A). The tubes were vigorously rocked at 4°C for 15 min on a shaking platform. After centrifugation for 10 min at 4°C, supernatants were collected for use. Protein concentrations were determined using a colorimetric assay (23).

Preparation of Whole-Cell Lysates

rMC were harvested, pelleted by centrifugation, and resuspended in 100 μl of lysis buffer (20 mM Hepes, pH 7.5, 250 mM KCl, 0.1 mM ethylenediaminetetraacetate, 0.1 mM ethylene glycol bis[β-aminoethyl ether]-N,N,N′,N′-tetraacetate, 10% glycerol, 0.1% Nonidet P-40, 0.5 mM PMSF, 0.5 mM DTT, 1 μg/ml aprotinin, 1 μg/ml pepstatin A). The cell suspension was vigorously mixed on a platform with medium rotation for 30 min and was centrifuged at 15,000 rpm for 30 min. Protein concentrations were determined using a colorimetric assay (23).

Western Blotting

Nuclear extracts (10 μg of each) were analyzed on 10% polyacrylamide gels and whole-cell lysates (20 μg of each) on 10 to 20% gradient polyacrylamide gels. After transfer to polyvinylidene difluoride membranes, blocking procedures with 5% nonfat milk were performed for iNOS immunoblotting, followed by incubation with anti-rabbit polyclonal antibodies against the p65 or p50 NF-κB subunit or iNOS. After washing of the membranes, peroxidase-conjugated secondary antibody was applied and the protein of interest was identified by enhanced chemiluminescence (ECL kit), according to instructions provided by the manufacturer.

Electrophoretic Mobility Shift Assay

A gel shift kit was used, following the protocol described by the manufacturer. Briefly, nuclear extracts (10 μg of each) underwent reaction in premixed incubation buffer (included in the kit), with γ-32P-end-labeled oligonucleotides containing the NF-κB-binding consensus sequence found upstream of the β-Ig gene and in the enhancer region of HIV (16) (5′-GATCGAGGGGACTTTCCCTAGC-3′), for 30 min at room temperature. Excess unlabeled oligonucleotides (100-fold molar excess) were also added in one sample to confirm the binding specificity. For supershift analysis, nuclear extracts were incubated for 20 min with γ-32P-end-labeled oligonucleotides, followed by incubation for 10 min with specific polyclonal IgG antibody against the p65 or p50 subunit of NF-κB. After the reaction, the samples were analyzed on a 7% nondenaturing polyacrylamide gel. The gel was dried and exposed to x-ray film.

Transfection and Transient Expression of iNOS Promoter-Luciferase Reporter Gene Constructs

The plasmid containing cDNA for the iNOS promoter and the luciferase reporter gene (piNOS-luciferase) was a gift from Dr. Charles J. Lowenstein (Division of Cardiology, Johns Hopkins University) (17). piNOS-luciferase (20 μg) and the β-galactosidase expression plasmid (5 μg) were cotransfected into LLC-PK1 cells by electroporation. After 24 h, cells were stimulated for 16 h with TNF-α (100 ng/ml) plus IFN-γ (100 U/ml), in the presence or absence of JAK2 inhibitors (300 μM AG490, 2 μM JAK2 antisense oligonucleotides, or 2 μM random control oligonucleotides). The sequences used were 5′-GCTTGTGAGAAAGC-3′ for JAK2 antisense phosphorothioate oligonucleotides and 5′-GTCCCTATACGAAC-3′ for random control oligonucleotides. Cell lysates were prepared using reporter lysis buffer (Promega). The luciferase activity of 10 μl of cleared lysate was assayed using a Berthold LB 9501 luminometer, with injection of 50 μl of PicaGene™ (Toyo Inki, Tokyo, Japan); results were normalized for transfection efficiency, as determined by β-galactosidase activity.

Immunoprecipitation

rMC were incubated with LPS (100 μg/ml) plus IFN-γ, in the presence or absence of AG490 (100 μM), for 15 min. Using an immunoprecipitation kit and following the instructions provided by the manufacturer, whole-cell lysates were prepared and immunoprecipitation was performed with a monoclonal antibody against phosphotyrosine-containing proteins, followed by Western blotting with a polyclonal anti-JAK2 antibody.

Statistical Analyses

Experimental conditions were assayed in triplicate within each experiment (in Figures 1, 2, 3, and 7A) or in duplicate within each experiment (in Figure 8). Comparisons were made using one-way ANOVA (in Figures 1 and 2), the Scheffe F test (in Figure 8), and the Fisher protected least-significant difference test (in Figures 3 and 7A). The level for statistically significant differences was defined as P < 0.05.

Figure 1.
Figure 1.

Failure of rat mesangial cells (rMC) to produce nitrite under the activation of nuclear factor-κB (NF-κB). rMC were incubated for 24 h with 5% fetal calf serum-containing medium. Control (cont), 1.2 ± 0.61 μM (n = 5); lipopolysaccharide (LPS) at 1 μg/ml, 1.2 ± 1.2 μM (n = 3); at 10 μg/ml, 1.36 ± 1.3 μM (n = 3); at 100 μg/ml, 2.37 ± 2.73 μM (n = 4); tumor necrosis factor-α (TNF-α) at 100 ng/ml, 1.09 ± 1.1 μM (n = 3). The data are mean ± SEM. There were no significant differences between control and LPS or TNF-α treatment groups (P < 0.05).

Figure 2.
Figure 2.

Nitrite production by rMC stimulated with the combination of interferon-γ (IFN-γ) and LPS. rMC were stimulated for 24 h with 5% fetal calf serum-containing medium. Control, 2.28 ± 0.5 μM (n = 5); with LPS at 1 μg/ml, 12.2 ± 5.9 μM (n = 3); at 100 μg/ml, 20.7 ± 2.4 μM (n = 4); TNF-α at 1 ng/ml, 5.2 ± 0.3 μM (n = 3); at 10 ng/ml, 31.9 ± 3.6 μM (n = 3); at 100 ng/ml, 36.8 ± 1.5 μM (n = 5). All treatment cocktails contained IFN-γ (100 U/ml). The data shown are mean ± SEM. There were significant differences in nitrite accumulation between control and treatment groups (P < 0.05).

Figure 3.
Figure 3.

Effects of NF-κB inhibitors on nitrite production by rMC stimulated with TNF-α + IFN-γ (A) or LPS + IFN-γ (B). (A) rMC were incubated for 24 h with TNF-α (100 ng/ml) + IFN-γ (100 U/ml) in combination with various concentrations of NF-κB inhibitors. The nitrite accumulation shown is a percentage of TNF-α + IFN-γ treatment values. Control, 13.2 ± 1.8%; TNF-α + IFN-γ + pyrrolidine dithiocarbamate (PDTC) at 0.01 μM, 82.8 ± 4.1% (n = 4); at 0.1 μM, 81.1 ± 3.0% (n = 4); at 0.5 μM, 78.0 ± 3.2% (n = 4); TNF-α + IFN-γ + N-acetyl-L-cysteine (NAC) at 0.1 mM, 79.8 ± 3.2% (n = 4); at 0.3 mM, 77.4 ± 3.7% (n = 4); at 3 mM, 72.2 ± 3.1% (n = 4); TNF-α + IFN-γ + dexamethasone (Dex) at 0.01 μM, 85.8 ± 3.9% (n = 4); at 0.1 μM, 77.0 ± 2.4% (n = 4); at 1 μM, 75.3 ± 0.8% (n = 4). The data are mean ± SEM. There were significant differences between control or TNF-α + IFN-γ and each inhibitor group, between PDTC at 0.01 μM and PDTC at 0.5 μM (*), between NAC at 0.1 mM and NAC at 3 mM (**), and between Dex at 0.01 μM and Dex at 0.1 μM (***) (P < 0.05). (B) rMC were incubated for 24 h with LPS (100 μg/ml) + IFN-γ (100 U/ml) in combination with the NF-κB inhibitors shown in A. The nitrite accumulation shown is a percentage of LPS + IFN-γ treatment values. Control, 8.4 ± 0.98% (n = 4); LPS + IFN-γ + PDTC at 0.01 μM, 85.2 ± 4.8% (n = 4); at 0.1 μM, 77.2 ± 6.1% (n = 4); at 0.5 μM, 66.0 ± 1.4 (n = 4); LPS + IFN-γ + NAC at 0.1 mM, 86.8 ± 2.4% (n = 4); at 0.3 mM, 81.5 ± 1.9% (n = 4); at 3 mM, 73.7 ± 3.2% (n = 4); LPS + IFN-γ + Dex at 0.01 μM, 60.9 ± 0.8% (n = 4); at 0.1 μM, 46.2 ± 1.1% (n = 4); at 1 μM, 45.3 ± 1.1% (n = 4). The data are mean ± SEM. There were significant differences between control or LPS + IFN-γ and each inhibitor group, between PDTC at 0.1 μM and PDTC at 0.5 μM (*), between NAC at 0.1 mM and NAC at 3 mM (**), and between Dex at 0.01 μM and Dex at 0.1 μM (***) (P < 0.05 for all comparisons).

Figure 7.
Figure 7.

(A) The protein tyrosine kinase inhibitor herbimycin A (HerA) and the relatively specific Janus kinase 2 (JAK2) inhibitor AG490 markedly reduced nitrite production by rMC stimulated with LPS + IFN-γ. (B) AG490 had little effect on NF-κB and its consensus sequence oligonucleotide complex. (C) AG490 suppressed phos-phorylation of JAK2. (A) HerA (at 10, 100, or 1000 ng/ml) or AG490 (at 0.3, 3, 30, or 300 μM) was added to LPS (100 μg/ml) + IFN-γ (100 U/ml), and cells were incubated for 24 h. Nitrite accumulation is shown as a percentage of LPS/IFN-γ-stimulated nitrite production. LPS + IFN-γ, 100% (n = 4); LPS + IFN-γ + HerA at 10 ng/ml, 83.3 ± 6.6% (n = 4); at 100 ng/ml, 35.1 ± 4.2% (n = 4); at 1000 ng/ml, 21.8 ± 3.7% (n = 4); LPS + IFN-γ + AG490 at 0.3 μM, 87.8 ± 4.1% (n = 4); at 3 μM, 67.9 ± 3.1% (n = 4); at 30 μM, 48.8 ± 1.4% (n = 4); at 300 μM, 24.8 ± 3.3% (n = 4); control, 15.3 ± 5.2% (n = 4). The data are mean ± SEM. There were significant differences between HerA groups, between AG490 groups, and between the LPS + IFN-γ group and each inhibitor group (P < 0.05), whereas there were no significant differences between control and LPS + IFN-γ + HerA (1000 ng/ml) or LPS + IFN-γ + AG490 (300 μM) (P < 0.05). (B) AG490 (30 and 3 μM) was incubated with LPS (100 μg/ml) + IFN-γ (100 U/ml) for 3.5 h, and then nuclear extracts were obtained. This is representative of three independent experiments. See Figure 5 for details on electrophoretic mobility shift assay. The radioactivity to the left of the band in the control lane is an artifact. These results showed that the JAK2 pathway played an important role in the induction of iNOS in the IFN-γ signaling cascade, without affecting NF-κB. (C) rMC were stimulated with LPS (100 mg/ml) + IFN-γ (100 U/ml), in the presence or absence of AG490 (100 μM), for 15 min. Using an immunoprecipitation kit, according to the instructions provided by the manufacturer, whole-cell lysates were prepared and immunoprecipitation was performed with 1 μg of monoclonal antibody against phosphotyrosine-containing proteins, followed by Western blotting with polyclonal anti-JAK2 antibody. The results showed that JAK2 was phosphorylated by LPS + IFN-γ, and the phosphorylation was suppressed by AG490. This is representative of three independent experiments.

Figure 8.
Figure 8.

Suppression of TNF-α/IFN-γ-stimulated iNOS promoter activity by AG490 and JAK2 antisense oligonucleotides in LLC-PK1 cells. LLC-PK1 cells were contransfected with piNOS-luciferase (20 μg) and β-galactosidase expression vector (5 μg) and were stimulated with TNF-α (100 ng/ml) and IFN-γ (100 U/ml) for 16 h, with or without JAK2 inhibitors. The cells were then lysed and assayed for luciferase activity. TNF-α + IFN-γ, 2.12 ± 0.047 (n = 4); TNF-α + IFN-γ + AG490 (300 μM), 1.39 ± 0.07 (n = 4); TNF-α + IFN-γ + JAK2 antisense oligonucleotides (2 μM), 1.19 ± 0.14 (n = 4); TNF-α + IFN-γ + random control oligonucleotides (2 μM) (cont. oligo), 1.73 ± 0.10 (n = 4). The data shown are mean ± SEM. The values were normalized to β-galactosidase activity and are shown as fold increases, compared with control luciferase activity. There were significant differences between TNF-α + IFN-γ and TNF-α + IFN-γ + AG490 (*) or TNF-α + IFN-γ + JAK2 antisense oligonucleotides (**) and also between TNF-α + IFN-γ + JAK2 antisense oligonu-cleotides and TNF-α + IFN-γ + control oligonucleotides (***) but not between TNF-α + IFN-γ + AG490 and TNF-α + IFN-γ + JAK2 antisense oligonucleotides (P < 0.05 for all comparisons). There was no significant difference between TNF-α + IFN-γ and TNF-α + IFN-γ + control oligonucleotides and no significant difference between control and TNF-α + IFN-γ + AG490 or JAK2 antisense oligonucleotides. A decrease in iNOS promoter activity is consistent with the results in Figure 7A. For details, see Materials and Methods.

Results

Insufficiency of NF-κB Activation for the Production of NO

We hypothesized that even full induction of NF-κB would not generate nitrite in rMC by itself. rMC were incubated with LPS (100 μg/ml) or TNF-α (100 ng/ml) to release NF-κB from its inactive form (Figure 1). There were no significant differences between control and LPS or TNF-α treatment groups. rMC were also incubated with IFN-γ (100 U/ml) as a positive control, to exclude the possibility that LPS, TNF-α, or rMC did not function properly (Figure 2). Cells exhibited nitrite accumulation in a dose-dependent manner, with peak levels of 20.7 ± 2.4 μM (100 μg/ml LPS + 100 U/ml IFN-γ, n = 4) and 36.8 ± 1.5 μM (100 ng/ml TNF-α + 100 U/ml IFN-γ, n = 5). We confirmed that TNF-α, LPS, IFN-γ, and NF-κB inhibitors were not toxic to rMC at their working concentrations, using the dimethylthiazolyldiphenyltetrazolium bromide assay (data not shown). These data suggested that activation of NF-κB did not stimulate nitrite production by itself and, unlike in other cells, IFN-γ signaling was not synergistic but was essential in rMC.

NO Generation in the Absence of Active NF-κB

If NF-κB plays an important but not critical role for iNOS induction, nitrite could be produced without active NF-κB. rMC were incubated with cytokines in the presence of NF-κB inhibitors (Figure 3). PDTC and NAC act as antioxidants and interfere with I-κB degradation, and Dex functions through the induction of I-κB synthesis. Nitrite accumulation stimulated by TNF-α + IFN-γ was inhibited in a dose-dependent manner, by 70 to 80%, with addition of nontoxic concentrations of PDTC, NAC, or Dex (Figure 3A). Nitrite production stimulated by LPS + IFN-γ was also not completely inhibited by NF-κB inhibitors (Figure 3B). To exclude the possibility that the incomplete inhibition was the result of short inhibitor half-lives, we incubated rMC with LPS + IFN-γ in the presence of PDTC, NAC, or Dex for 8 and 17 h and observed almost the same inhibitory effects (data not shown). Also, Western blotting was performed to determine whether iNOS was induced without NF-κB activation (Figure 4). rMC were incubated with TNF-α + IFN-γ + Dex or ASA. The result showed that TNF-α alone did not induce iNOS, and Dex failed to make complete the inhibition of iNOS induced by TNF-α + IFN-γ. ASA did not inhibit iNOS induction. Similar results were obtained from incubations with LPS + IFN-γ (Figure 4B).

Figure 4.
Figure 4.

Failure of Dex to produce complete inhibition of inducible nitric oxide synthase (iNOS) induced by TNF-α + IFN-γ (A) or LPS + IFN-γ (B). Ninety percent confluent rMC were incubated with TNF-α (100 ng/ml), TNF-α (100 ng/ml) + IFN-γ (100 U/ml), TNF-α + IFN-γ + Dex (1 μM), or TNF-α + IFN-γ + acetylsalicylic acid (ASA) (1 mM) (A) or with LPS (100 μg/ml) instead of TNF-α (B) for 20 h at 37°C and were collected to yield whole-cell lysates. Each lysate (20 μg) was loaded on a 10% polyacrylamide gel. After proteins were transferred to polyvinylidene difluoride membranes, the membranes were incubated with anti-rabbit polyclonal iNOS antibody and then with secondary antibody. Neither TNF-α nor LPS induced iNOS by itself under the activation of NF-κB, and Dex and ASA failed to abolish iNOS completely. Each result is representative of three independent experiments.

Demonstration of Complete Inhibition of Active NF-κB

The question of whether NF-κB is completely blocked by NF-κB inhibitors arises. To address this issue, we performed electrophoretic mobility shift assay (EMSA) (Figure 5). There were two weak bands in control samples. After stimulation with LPS or LPS + IFN-γ, the two bands were increased. Supershift analysis revealed that the upper band was a p65/p50 heterodimer and the lower band was a p50/p50 homodimer. The upper p65/p50 heterodimer was completely inhibited, with the signal below control levels, whereas the level of the lower p50/p50 homodimer was not changed. The two bands disappeared with the addition of a 100-fold greater concentration of nonradiolabeled oligonucleotides containing the NF-κB consensus sequence, which confirmed that these bands were specific. In combination with Figure 4, these results showed that iNOS was induced despite complete abolishment of NF-κB/DNA interactions and that LPS alone did not induce iNOS despite NF-κB activation.

Figure 5.
Figure 5.

Complete inhibition by PDTC and Dex of NF-κB-oligonucleotide complexes produced by LPS + IFN-γ. Ninety percent confluent rMC were incubated with LPS (100 μg/ml), LPS (100 μg/ml) + IFN-γ (100 U/ml), LPS + IFN-γ + PDTC (10 μM), or LPS + IFN-γ + Dex (1 μM) for 3.5 h at 37°C and were collected to yield nuclear extracts. Nuclear extracts (10 μg) were reacted for 20 min at room temperature with γ-32P-end-labeled oligonucleotides containing the NF-κB-binding consensus sequence. A 100-fold molar excess of unlabeled oligonucleotides containing the NF-κB consensus sequence was also added to the LPS + IFN-γ samples. For supershift analysis, nuclear extracts were preincubated with radiolabeled probe for 20 min, followed by reaction with specific antibody for an additional 10 min. The radioactivity to the left of the band in the control lane is an artifact. PDTC and Dex completely abolished the p65/p50 heterodimer of NF-κB, to values below the control levels. This is representative of three independent experiments. *100 cold oligo, 100-fold molar excess of unlabeled oligonucleotides containing the NF-κB consensus sequence; p65 Ab, LPS + IFN-γ + radiolabeled NF-κB consensus sequence oligonucleotides incubated with antibody against the p65 subunit of NF-κB; p50 Ab, LPS + IFN-γ + radiolabeled NF-κB consensus sequence oligonucleotides incubated with antibody against the p50 subunit of NF-κB.

Active NF-κB Formation in an I-κB-Independent Manner

NF-κB is activated by degradation of I-κB and is translocated to the nucleus, where it binds to the consensus sites on DNA (24,25). We used Western blotting against the p65 and p50 subunits in nuclear extracts to confirm that this mechanism also applied in rMC (Figure 6). If the mechanism applied, then both p65 and p50 subunits would disappear after incubation with NF-κB inhibitors. p50 was constitutively expressed in the control group. p65 was completely abolished after incubations with PDTC, Dex, or ASA. However, surprisingly, p50 was not affected by NF-κB inhibitors. EMSA in combination with supershift assays revealed that the p65/p50 heterodimer was suppressed to control levels, whereas the p50/p50 homodimer was not inhibited, which supported the Western blotting results described above.

Figure 6.
Figure 6.

Inhibition of only the p65 subunit of NF-κB by PDTC, Dex, and ASA. We performed Western blotting against the p65 (A) and p50 (B) subunits of NF-κB. Briefly, 90% confluent rMC were incubated with LPS (100 μg/ml), LPS (100 μg/ml) + IFN-γ (100 U/ml), LPS + IFN-γ + PDTC (10 μM), LPS + IFN-γ + Dex (1 μM), or LPS + IFN-γ + ASA (1 mM) for 3.5 h at 37°C and were collected to yield nuclear extracts. Each nuclear extract (10 μg) was analyzed on a polyacrylamide gel. After blocking with 5% nonfat milk, the membranes were incubated with anti-rabbit polyclonal p65 (A) or p50 (B) antibody and then with secondary antibody. Each result is representative of three independent experiments. For details, see Materials and Methods. These results showed that the aforementioned NF-κB inhibitors inhibited only the p65 subunit.

Essential Role of JAK2 for the Induction of iNOS

Next we sought to determine how IFN-γ affected nitrite production (Figure 7A). To examine whether protein tyrosine kinase was involved in IFN-γ-induced nitrite production, rMC were costimulated with LPS (100 μg/ml) + IFN-γ (100 U/ml) in the presence of HerA and a more specific JAK2 inhibitor (AG490) for 24 h (Figure 7A). HerA and AG490 reduced nitrite accumulation in a dose-dependent manner. HerA (1000 ng/ml) and AG490 (300 μM) completely inhibited nitrite production (Figure 7A). To confirm that AG490 functioned at least in part through inhibition of JAK2, we performed immunoprecipitation with an antibody against phosphotyrosine-containing proteins, followed by Western blotting against JAK2, which revealed that AG490 actually suppressed JAK2 phosphorylation (Figure 7C). These results were supported by iNOS promoter activities measured using the luciferase assay (Figure 8). There were significant differences between TNF-α (100 ng/ml) + IFN-γ (100 U/ml) and TNF-α + IFN-γ + AG490 (300 μM) or TNF-α + IFN-γ + JAK2 antisense oligonucleotides (2 μM), but not between TNF-α + IFN-γ + AG490 (300 μM) and TNF-α + IFN-γ + JAK2 antisense oligonucleotides (2 μM). There was no significant difference between TNF-α + IFN-γ and TNF-α + IFN-γ + random control oligonucleotides, which demonstrated that the effect of JAK2 antisense oligonucleotides was neither nonspecific nor toxic. The data in Figures 7A and 8 indicate that inhibition of JAK2 causes complete abrogation of nitrite, which suggests that the contribution of JAK2 to nitrite production is greater than that of NF-κB. EMSA showed that AG490 had almost no effect on the NF-κB-consensus sequence oligonucleotide complex (Figure 7B). These data suggested that JAK2 plays an essential role in the induction of iNOS in the IFN-γ signaling cascade, without affecting NF-κB.

Discussion

The mechanism by which inflammatory cytokines induce iNOS has been thought to involve activation of NF-κB (17). In RAW 264.7 cells, LPS increased iNOS mRNA and promoter activity (9,18). In contrast, LPS failed to induce iNOS in murine proximal convoluted tubule cells (19). In our experiments, even high concentrations of LPS (100 μg/ml) and TNF-α (100 ng/ml) failed to generate nitrite in rMC, whereas nitrite was produced with the addition of IFN-γ to LPS. Taken together, these results suggest that NF-κB activation is not sufficient for nitrite accumulation in rMC.

In this study, we established that in rMC NF-κB induced by TNF-α or LPS did not promote nitrite production and that different kinds of NF-κB inhibitors, i.e., PDTC (which is a relatively specific inhibitor and does not affect other transcription factors, such as specificity protein-1, activator protein-1, or cAMP response element binding protein) (26), NAC, and Dex (27,28,29), produced only partial inhibition of nitrite accumulation at maximal nontoxic concentrations. On the basis of these data, we proposed that NF-κB played an important but not critical role in rMC, because iNOS was induced even in the absence of active NF-κB in rMC. Endothelin-1 blocked cyto-kine-induced iNOS expression under the activation of NF-κB (30), which supported our results. This study does not conflict with the report by Saura et al. (31), which demonstrated the importance of NF-κB in the production of NO, because NF-κB is important in terms of its contribution to nitrite production even if iNOS can be induced in the absence of active NF-κB. We also discovered that in rMC the p65 subunit of NF-κB was not detectable in nuclear extracts, whereas p50 was still observed in the presence of NF-κB inhibitors. Ziegler-Heitbrock et al. (26) suggested but did not demonstrate this phenomenon in mono Mac 6 cells. These data suggest that p50 is regulated in ways other than via the I-κB degradation pathway previously described (24,25).

Distinct NF-κB heterodimers have been shown to regulate iNOS induction. The p65/p50 heterodimer confers LPS inducibility of the iNOS in murine medullary thick ascending limb cells and murine proximal convoluted tubule cells (19,32), whereas the c-Rel/p50 heterodimer mediates this response in mouse macrophages (9). In our studies, we examined only the p65 and p50 subunits of NF-κB when rMC were activated by LPS or TNF-α, because Saura et al. (31) reported that p65 and p50 played dominant roles in the production of nitrite in rMC, whereas c-Rel contributed little, if at all. It is interesting that different cells adopt different mechanisms to regulate iNOS. This is a ubiquitous enzyme that is expressed by a number of mammalian cells (6), but the degree of nitrite production varies, even in response to the same stimuli (33). This regulatory complexity presumably reflects the inducibility of the iNOS gene, with one NF-κB heterodimer being a much stronger driving force for iNOS induction than the other. This hypothesis may explain why macrophages produce larger amounts of nitrite than do mesangial cells in response to the same kinds and same concentrations of inflammatory cytokines. The cell type-specific control of NO production is closely associated with the cell type-specific function, i.e., macrophages for bactericidal purposes and cortical collecting duct cells for sodium diuresis (13).

To exclude the possibility that traces of the p65/p50 heterodimer existed in our experiment, we performed EMSA. We found that Dex totally abolished the probe-p65/p50 NF-κB complex, to less than control levels, confirming that the NF-κB inhibitor functioned fully and properly. In contrast to the report of Amoah-Apraku et al. (19), nitrite was still produced after the total abolishment of active NF-κB, suggesting that a signaling pathway independent of NF-κB exists for the induction of nitrite. Addition of ASA to LPS + IFN-γ inhibited p65 (34) but, surprisingly, induction of iNOS was more enhanced than without ASA. This evidence suggests that prostaglandins may have negative effects on iNOS induction, at least in rMC.

IFN-γ has been shown to have a synergistic action when used with TNF-α or LPS in other cells (18,33); however, judging from our data, the effect is not just synergistic, because neither LPS nor IFN-γ yielded nitrite if used alone. Macrophages derived from interferon regulatory factor-1-deficient mice also do not generate nitrite (35). This evidence suggests that IFN-γ has a crucial role in iNOS induction, in concert with NF-κB. We measured IFN-γ-stimulated nitrite production and found that IFN-γ alone did not yield nitrite (n = 5). The explanation for this result is as follows. Even if NF-κB is completely inhibited, an LPS- or TNF-α-activated signaling pathway other than that involving NF-κB may coordinate with IFN-γ to induce iNOS, because LPS or TNF-α signaling is not confined only to NF-κB. Therefore, it is probable that detectable levels of nitrite would not be produced by IFN-γ alone. We found that both HerA (1000 ng/ml) and AG490 (300 μM) completely inhibited nitrite production. These data suggest that JAK2 activation is critical but not sufficient for nitrite production. Of all the JAK isoforms, JAK2 is located downstream of the IFN-γ receptor. After receptor stimulation, IFN-γ activates JAK1 and JAK2, which in turn phosphorylate signal transducers and activators of transcription 1α (STAT1α), leading to dimerization of STAT1α and translocation to the nucleus (36,37). The phosphorylated STAT1α dimer then binds to the consensus motif of the interferon regulatory factor promoter and initiates transcription (18,38). HerA exhibits broad specificity (39,40), whereas AG490 is more specific for JAK2 inhibition (41). In hepatocytes, which also contain iNOS, AG490 inhibits promoter activation of the iNOS gene induced by IFN-γ + LPS via suppression of NF-κB binding activity (42). In our results, the accumulation of nitrite was totally inhibited, without interference with NF-κB activation, and iNOS promoter activity was completely inhibited by AG490 and JAK2 antisense oligonucleotides, suggesting that in rMC AG490 functioned not via NF-κB inhibition but via JAK2 inhibition.

In conclusion, NF-κB has an important but not critical role in iNOS induction in rMC. NF-κB inhibitors function through suppression of the p65 subunit, rather than p50. IFN-γ functions at least in part through JAK2, which plays a critical role in nitrite production. The contribution of JAK2 to nitrite production is greater than that of NF-κB.

Footnotes

  • American Society of Nephrology

References

  1. ↵
    Bredt DS, Snyder SH: Nitric oxide: A physiologic messenger molecule. Annu Rev Biochem 63:175 -195, 1994
    OpenUrlCrossRefPubMed
  2. ↵
    Geller DA, Nussler AK, Di Silvio M, Lowenstein CJ, Shapiro RA, Wang SC, Simmons RL, Billiar TR: Cytokines, endotoxin, and glucocorticoids regulate the expression of inducible nitric oxide synthase in hepatocytes. Proc Natl Acad Sci USA 90:522 -526, 1993
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Imai T, Hirata Y, Kanno K, Marumo F: Induction of nitric oxide synthase by cyclic AMP in rat vascular smooth muscle cells. J Clin Invest 93: 543-549,1994
  4. ↵
    Kunz D, Muhl H, Walker G, Pfeilschifter J: Two distinct signaling pathways trigger the expression of inducible nitric oxide synthase in rat renal mesangial cells. Proc Natl Acad Sci USA91 : 5387-5391,1994
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Lyons CR, Orloff GJ, Cunningham JM: Molecular cloning and functional expression of an inducible nitric oxide synthase from a murine macrophage cell line. J Biol Chem 267:6370 -6374, 1992
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Nathan C: Nitric oxide as a secretory product of mammalian cells.FASEB J 6:3051 -3064, 1992
    OpenUrlAbstract
  7. ↵
    Velazquez L, Fellous M, Stark GR, Pellegrini S: A protein tyrosine kinase in the interferon α/β signaling pathway. Cell70 : 313-322,1992
    OpenUrlCrossRefPubMed
  8. ↵
    Evans CH, Stefanovic-Racic M, Lancaster J: Nitric oxide and its role in orthopaedic disease. Clin Orthop312 : 275-294,1995
  9. ↵
    Xie QW, Kashiwabara Y, Nathan C: Role of transcription factor NF-κB/Rel in induction of nitric oxide synthase. J Biol Chem 269: 4705-4708,1994
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Kone BC, Baylis C: Biosynthesis and homeostatic roles of nitric oxide in the normal kidney. Am J Physiol272 : F561-F578,1997
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Palmer RM, Ferrige AG, Moncada S: Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature (Lond) 327: 524-526,1987
    OpenUrlCrossRefPubMed
  12. ↵
    Radomski MW, Palmer RM, Moncada S: The anti-aggregating properties of vascular endothelium: Interactions between prostacyclin and nitric oxide.Br J Pharmacol 92:639 -646, 1987
    OpenUrlCrossRefPubMed
  13. ↵
    McKee M, Scavone C, Nathanson JA: Nitric oxide, cGMP, and hormone regulation of active sodium transport. Proc Natl Acad Sci USA91 : 12056-12060,1994
    OpenUrlAbstract/FREE Full Text
  14. ↵
    Stoos BA, Garcia NH, Garvin JL: Nitric oxide inhibits sodium reabsorption in the isolated perfused cortical collecting duct. J Am Soc Nephrol 6: 89-94,1995
    OpenUrlAbstract
  15. ↵
    Uhlenius N, Tikkanen I, Tikkanen T, Miettinen A, Tornroth T, Fyhrquist F: Chronic inhibition of nitric oxide synthase in Heymann nephritis.Nephron 74:144 -149, 1996
    OpenUrlPubMed
  16. ↵
    Sen R, Baltimore D: Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Cell46 : 705-716,1986
    OpenUrlCrossRefPubMed
  17. ↵
    Lowenstein CJ, Alley EW, Raval P, Snowman AM, Snyder SH, Russell SW, Murphy WJ: Macrophage nitric oxide synthase gene: Two upstream regions mediate induction by interferon gamma and lipopolysaccharide. Proc Natl Acad Sci USA 90:9730 -9734, 1993
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Lorsbach RB, Murphy WJ, Lowenstein CJ, Snyder SH, Russell SW: Expression of the nitric oxide synthase gene in mouse macrophages activated for tumor cell killing: Molecular basis for the synergy between interferon-gamma and lipopolysaccharide. J Biol Chem268 : 1908-1913,1993
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Amoah-Apraku B, Chandler LJ, Harrison JK, Tang SS, Ingelfinger JR, Guzman NJ: NF-κB and transcriptional control of renal epithelial-inducible nitric oxide synthase. Kidney Int48 : 674-682,1995
    OpenUrlCrossRefPubMed
  20. ↵
    Terada Y, Yamada T, Nakashima O, Tamamori M, Ito H, Sasaki S, Marumo F: Overexpression of cell cycle inhibitors (p16INK4 and p21Cip1) and cyclin D1 using adenovirus vectors regulates proliferation of rat mesangial cells. J Am Soc Nephrol 8:51 -60, 1997
    OpenUrlAbstract
  21. ↵
    Green LC, Wagner DA, Glogowski J, Skipper PL, Wishnok JS, Tannenbaum SR: Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids. Anal Biochem 126:131 -138, 1982
    OpenUrlCrossRefPubMed
  22. ↵
    Schreiber E, Matthias P, Muller MM, Schaffner W: Rapid detection of octamer binding proteins with “mini-extracts,” prepared from a small number of cells [Abstract]. Nucleic Acids Res17 : 6419,1989
    OpenUrlFREE Full Text
  23. ↵
    Bradford MM: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248 -254, 1976
    OpenUrlCrossRefPubMed
  24. ↵
    Chen ZJ, Parent L, Maniatis T: Site-specific phosphorylation of IκBα by a novel ubiquitination-dependent protein kinase activity.Cell 84:853 -862, 1996
    OpenUrlCrossRefPubMed
  25. ↵
    Janosch P, Schellerer M, Seitz T, Reim P, Eulitz M, Brielmeier M, Kolch W, Sedivy JM, Mischak H: Characterization of IκB kinases: IκBα is not phosphorylated by Raf-1 or protein kinase C isozymes, but is a casein kinase II substrate. J Biol Chem271 : 13868-13874,1996
    OpenUrlCrossRefPubMed
  26. ↵
    Ziegler-Heitbrock HW, Sternsdorf T, Liese J, Belohradsky B, Weber C, Wedel A, Schreck R, Bauerle P, Strobel M: Pyrrolidine dithiocarbamate inhibits NF-κB mobilization and TNF production in human monocytes.J Immunol 151:6986 -6993, 1993
    OpenUrlAbstract
  27. ↵
    Auphan N, DiDonato JA, Rosette C, Helmberg A, Karin M: Immunosuppression by glucocorticoids: Inhibition of NFκB activity through induction of IκB synthesis. Science270 : 286-290,1995
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Scheinman RI, Cogswell PC, Lofquist AK, Baldwin AS Jr: Role of transcriptional activation of IκBα in mediation of immunosuppression by glucocorticoids. Science270 : 283-286,1995
    OpenUrlAbstract/FREE Full Text
  29. ↵
    Scheinman RI, Gualberto A, Jewell CM, Cidlowski JA, Baldwin AS Jr: Characterization of mechanisms involved in transrepression of NF-κB by activated glucocorticoid receptors. Mol Cell Biol15 : 943-953,1995
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Beck KF, Sterzel RB: Cloning and sequencing of the proximal promoter of the rat iNOS gene: Activation of NFκB is not sufficient for transcription of the iNOS gene in rat mesangial cells. FEBS Lett 394: 263-267,1996
    OpenUrlCrossRefPubMed
  31. ↵
    Saura M, Zaragoza C, Diaz-Cazorla M, Hernandez-Perera O, Eng E, Lowenstein CJ, Perez-Sala D, Lamas S: Involvement of transcriptional mechanisms in the inhibition of NOS2 expression by dexamethasone in rat mesangial cells. Kidney Int 53:38 -49, 1998
    OpenUrlCrossRefPubMed
  32. ↵
    Kone BC, Schwobel J, Turner P, Mohaupt MG, Cangro CB: Role of NF-κB in the regulation of inducible nitric oxide synthase in an MTAL cell line. Am J Physiol 269:F718 -F729, 1995
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Lau KS, Nakashima O, Aalund GR, Hogarth L, Ujiie K, Yuen J, Star RA: TNF-α and IFN-γ induce expression of nitric oxide synthase in cultured rat medullary interstitial cells. Am J Physiol269 : F212-F217,1995
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Kopp E, Ghosh S: Inhibition of NF-κB by sodium salicylate and aspirin. Science 265:956 -959, 1994
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Kamijo R, Harada H, Matsuyama T, Bosland M, Gerecitano J, Shapiro D, Le J, Koh SI, Kimura T, Green SJ: Requirement for transcription factor IRF-1 in NO synthase induction in macrophages. Science263 : 1612-1615,1994
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Darnell JE Jr, Kerr IM, Stark GR: Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264:1415 -1421, 1994
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Ihle JN: STATs: Signal transducers and activators of transcription.Cell 84:331 -334, 1996
    OpenUrlCrossRefPubMed
  38. ↵
    Wang Y, O'Neal KD, Yu-Lee L: Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Mol Endocrinol11 : 1353-1364,1997
    OpenUrlCrossRefPubMed
  39. ↵
    Akiyama T, Ishida J, Nakagawa S, Ogawara H, Watanabe S, Itoh N, Shibuya M, Fukami Y: Genistein, a specific inhibitor of tyrosine-specific protein kinases. J Biol Chem 262:5592 -5595, 1987
    OpenUrlAbstract/FREE Full Text
  40. ↵
    Uehara Y, Murakami Y, Suzukake-Tsuchiya K, Moriya Y, Sano H, Shibata K, Omura S: Effects of herbimycin derivatives on src oncogene function in relation to antitumor activity. J Antibiot (Tokyo)41 : 831-834,1988
    OpenUrlPubMed
  41. ↵
    Levitzki A, Gazit A: Tyrosine kinase inhibition: An approach to drug development. Science 267:1782 -1788, 1995
    OpenUrlAbstract/FREE Full Text
  42. ↵
    Lee BS, Kang HS, Pyun KH, Choi I: Roles of tyrosine kinases in the regulation of nitric oxide synthesis in murine liver cells: Modulation of NF-κB activity by tyrosine kinases. Hepatology25 : 913-919,1997
    OpenUrlCrossRefPubMed
View Abstract
Back to top

In this issue

View Selected Citations (0)
Print
Download PDF
Article Alerts
Email Article
Citation Tools
Request Permissions
Inducible Nitric Oxide Synthase Can Be Induced in the Absence of Active Nuclear Factor-κB in Rat Mesangial Cells
OSAMU NAKASHIMA, YOSHIO TERADA, SEIJI INOSHITA, MICHIO KUWAHARA, SEI SASAKI, FUMIAKI MARUMO
JASN Apr 1999, 10 (4) 721-729;
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Recent Tweets about the Article