Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Nitric oxide (NO) is a gas formed by cleavage of the guanido group of arginine (1). This reaction is catalyzed by nitric oxide synthases (NOSs) and several cofactors, including the flavones, nicotinamide adenine dinucleotide phosphate (NADPH), and tetrahydrobiopterin. The inducible form of NOS (iNOS, type II NOS) is an enzyme not usually constitutively expressed. However, iNOS can be induced by inflammatory cytokines, such as a combination of tumor necrosis factor- (TNF-), interleukin-1 (IL-1), and interferon- (IFN-), generating large amounts of NO (2, 3). In asthma, increased levels of TNF-, IL-1, and IFN- have been detected in the respiratory tract or associated with cells within the respiratory tract (4), increased levels of NO have been detected in exhaled air (7, 8), and iNOS has been demonstrated in airway epithelial cells (9).

Asthma is a disorder associated with airway edema formation and bronchial inflammation (10). NO has been proposed to play a role in asthma by enhancing both airway edema formation and inflammation (13, 14). Antiinflammatory therapy for asthma includes corticosteroids, but also may include antineoplastic agents such as methotrexate (MTX) (10). Corticosteroids have been shown to reduce NO production by inhibiting iNOS transcription in lung epithelial cells (2, 3, 15). The effects of MTX on NO production are unclear. Studies have demonstrated a decrease in NO production by MTX as a result of its inhibition of synthesis of tetrahydrobiopterin (16), which is a required cofactor for NOS (17, 18). However, others have reported an increase in NO production after MTX (19). Furthermore, the effects of MTX alone or in combination with corticosteroids on NO production in airway epithelial cells are uncertain.

The purpose of this study was to investigate the effects of MTX on NO production by murine lung epithelial cells. The results show a modest decrease in NO production, which was reversed by tetrahydrobiopterin, but no effect on iNOS protein or mRNA. Furthermore, the decrease was additive with corticosteroids, which have previously been shown to decrease NO production by inhibiting iNOS mRNA transcription.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

LA-4 Cultures

The clonal murine lung epithelial-cell line LA-4 was purchased from the American Type Culture Collection (Rockville, MD) (20). LA-4 cells were grown on 35-mm, six-well culture plates in Ham's F-12 medium containing 10% fetal calf serum (FCS), L-glutamine (2 mM), and penicillin-streptomycin (100 U/ml-100 mg/ml). When confluent, the cells were washed twice and cultured in serum-free Ham's F-12 medium. The cells were stimulated to express iNOS by culturing with cytomix, a combination of recombinant human TNF-, human IL-1, and murine IFN- (Sigma Chemical Co., St. Louis, MO) (all at 10 ng/ml final concentration for most experiments) (2, 3). To some cultures, MTX (10³ to 10⁵ M), dexamethasone (10⁵ to 10⁹ M), cyclophosphamide (10³ to 10⁵ M), cytosine -D-arabinofuranoside (Ara-C, 10³ to 10⁵ M), and/or tetrahydrobiopterin (10⁴ to 10⁷ M) (all from Sigma) or bleomycin (Bleo, 10³ to 10⁵ M; Bristol-Myers, Evansville, IN) were added. Cells cultured in Ham's F-12 medium alone served as a negative control. The LA-4-culture supernatant fluids were collected and stored at 80°C for measurement of nitrite or nitrite + nitrate, and cellular protein or total cellular RNA was extracted from the adherent cells. In some cultures, cellular viability was assessed by measuring lactate dehydrogenase (LDH; Sigma) in the culture supernatant fluids.

Nitrite and Nitrite + Nitrate Measurements

Nitrite was converted to NO under acidic conditions, and the NO was measured with a sensitive NO chemiluminescence analyzer (Model 270B; Sievers, Boulder, CO) (2, 3, 21). The amount of nitrite was calculated by comparison reading from a standard curve of sodium nitrite, with the assay having a sensitivity of < 20 nM. Nitrite + nitrate was measured by reducing nitrate to nitrite, using Aspergillus nitrate reductase (final concentration: 0.1 U/ml; Boehringer Mannheim, Indianapolis, IN) in the presence of NADPH (final concentration: 0.086 mM) and flavin adenine dinucleotide (FAD) (final concentration: 0.011 mM) for 60 min at room temperature. The nitrate + nitrite was then measured as nitrite, using the foregoing techniques.

Western Blotting

Cells were evaluated for iNOS protein expression by Western blotting, as previously described (22). Briefly, the cells were lysed in 10 mM Tris-HCl, pH 7.4, and 1% sodium dodecyl sulfate (SDS) buffer containing protease inhibitors. Samples were electrophoresed on 7.5% SDS-polyacrylamide gels and were then blotted to polyvinylidene difluoride membranes. The membranes were probed with a monoclonal antimacrophage iNOS antibody that does not recognize constitutive NOSs (Transduction Laboratories, Lexington, KY). The blots were then washed and incubated with an antimouse alkaline phosphatase-conjugated antibody. After washing, the blots were developed with 5-bromo-4-chloro-3-indolyl phosphate blue tetrazolium. After drying, the blots were photographed and the intensity of the bands was quantified through densitometry.

Northern Blot Analysis

Total cellular RNA was extracted from adherent cells, using a modification of the methods of Chomczynski and Sacchi (23). RNA samples were applied to 1% denaturing agarose gels, electrophoresed, and blotted onto Hybond-N filters (Amersham, Arlington Heights, IL). The blotted filters were hybridized with a ³²P-labeled murine iNOS complementary DNA (cDNA) probe (1 × 10⁶ cpm/ml) generated by random priming, using a multiprime DNA labeling system (Amersham). The murine iNOS cDNA probe was the 1.8-kb cDNA fragment spanning the presumed coding region of the iNOS cDNA clone (EcoRI fragment) (2, 24). The filters were washed at a final stringency of 0.1× standard saline citrate (SSC)/0.1% SDS at 55°C, and were exposed to Kodak X-OMAT S film (Kodak, Rochester, NY) at 70°C for 1 to 3 d. After autoradiography, the filters were stripped and hybridized with a ³²P-labeled glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The GAPDH cDNA probe was the 1,272-bp Pst I segment from rat GAPDH cDNA (2). Autoradiographs were assessed with scanning laser densitometry. Specific iNOS mRNA levels were calculated as the ratio of iNOS to GAPDH mRNA expression.

Statistical Analysis

Data are expressed as means ± SEM. Comparisons were done with Scheffe's analysis of variance (ANOVA). A value of P < 0.05 was considered significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Methotrexate Inhibition of Nitrite and Nitrite + Nitrate Accumulation

MTX caused a dose-dependent inhibition of nitrite and nitrite + nitrate in the culture supernatant fluids (Figure 1A. Nitrite: cytomix alone = 36.4 ± 0.6 µM; cytomix + MTX at 10³ M = 29.6 ± 1.1 µM; cytomix + MTX at 10⁴ M = 31.3 ± 0.8 µM; cytomix + MTX at 10⁵ M = 32.7 ± 0.7 µM. Figure 1B. Nitrite + nitrate: cytomix alone = 53.5 ± 5.8 µM; cytomix + MTX at 10³ M = 36.8 ± 0.8 µM; cytomix + MTX at 10⁴ M = 39.9 ± 1.0 µM; cytomix + MTX at 10⁵ M = 44.1 ± 2.7 µM). Similarly, MTX caused a time-dependent inhibition of nitrite and nitrite + nitrate (Figure 2A. Nitrite: cytomix alone at 8 h = 23.6 ± 1.6 µM versus cytomix + MTX at 10³ M at 8 h = 19.6 ± 0.3 µM; cytomix alone at 24 h = 58.9 ± 0.6 µM versus cytomix + MTX at 10³ M at 24 h = 42.7 ± 4.9 µM. Figure 2B. Nitrite + nitrate: cytomix at 8 h = 53.4 ± 3.2 µM versus cytomix + MTX at 10³ M at 8 h = 39.2 ± 2.0 µM; cytomix alone at 24 h = 94.7 ± 1.3 µM versus cytomix + MTX at 10³ M at 24 h = 64.0 ± 2.9 µM). The results were not explained by a decrease in cell viability because LDH was not increased in the culture supernatant fluids by MTX at 10³ M (medium: 39.4 ± 7.9 U/liter versus MTX: 59.6 ± 19.7 U/liter; n = 3 each condition; P > 0.05), and the effects were reversed by tetrahydrobiopterin (see the subsequent discussion).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Dose-responsive inhibition of nitrite (A) and nitrite + nitrate (B) by methotrexate (MTX: 103 to 105 M). LA-4 cells were cultured for 24 h in the presence of medium alone; a combination of human TNF-, human IL-1 , and IFN- (cytomix, all at 10 ng/ml); or cytomix with MTX at the indicated dose; n = 3 each condition; *P < 0.05 compared with cytomix alone.

View larger version (14K): [in this window] [in a new window]   Figure 2.   Time course of nitrite (A) and nitrite + nitrate (B) accumulation in the presence of MTX. Cells were cultured in the presence of medium alone, cytomix, or cytomix + MTX (103 M); n = 3 each condition; *P < 0.05 compared with cytomix alone.

Inhibition of nitrite and nitrate accumulation by MTX was statistically significant, though modest. However, cytomix with each of the cytokines at 10 ng/ml represented a near-maximal stimulus for the expression of iNOS in these murine lung epithelial cells (2). To determine whether a greater effect was observed at submaximal levels of stimulation, separate experiments were done with murine cells cultured with decreasing doses of cytomix (i.e., TNF-, IL-1, and IFN-) at doses of 1 ng/ml and 0.1 ng/ml final concentration for each cytokine, and the effect of MTX 10³ M was assessed. When expressed as a percent inhibition, the inhibition of nitrite accumulation was less pronounced with cytokine stimulation at 10 ng/ml (Figure 3A; 7% inhibition, 39.2 ± 2.8 versus 36.4 ± 1.9 µM, n = 6) than at 1 ng/ ml (24% inhibition, 36.7 ± 1.8 versus 27.9 ± 1.8 µM, n = 6) or at 0.1 ng/ml (39% inhibition, 14.5 ± 1.9 versus 8.9 ± 0.9 µM, n = 9). Similar results were obtained with nitrite + nitrate with cytokine stimulation at 10 ng/ml (Figure 3B; 13% inhibition, 92.8 ± 0.5 versus 80.8 ± 1.3 µM, n = 6) as compared with 1 ng/ml (17% inhibition, 79.9 ± 4.2 versus 66.5 ± 3.2 µM, n = 6) and 0.1 ng/ml (61% inhibition, 39.2 ± 0.9 versus 15.4 ± 0.6 µM, n = 9).

View larger version (13K): [in this window] [in a new window]   Figure 3.   Inhibition of nitrite (A) and nitrite + nitrate (B) by MTX 103 M (open circles, dotted line) compared with differing doses of cytomix (solid circles, solid line). LA-4 cells were cultured for 24 h with human TNF-, human IL-1, and IFN- all at 10 ng/ml (cytomix 10), 1 ng/ml (cytomix 1), or 0.1 ng/ml (cytomix 0.1), with or without MTX; n = 5 to 9, each condition; *P < 0.05 compared with cytomix at same dose alone.

In contrast to MTX, cyclophosphamide, bleomycin, and Ara-C, other antineoplastic agents, caused no reduction in nitrite or nitrite + nitrate accumulation in the culture supernatant fluids after 24 h (Figure 4; P > 0.05 all comparisons).

View larger version (43K): [in this window] [in a new window]   Figure 4.   Lack of effect of cyclophosphamide (Cyclo), bleomycin (Bleo), or Ara-C on nitrite (A) and nitrite + nitrate (B) accumulation. Cells were cultured for 24 h in the presence of medium alone, cytomix, or cytomix + Cyclo, Bleo, or Ara-C at the concentration indicated; n = 3 each condition. None of the supernatant fluids from cultures containing cytomix and antimetabolites differed compared with cytomix alone (P > 0.05, all comparisons).

MTX and Corticosteroid Inhibition of Nitrite and Nitrite + Nitrate Accumulation

In accord with previous reports, dexamethasone (10⁵ to 10⁹ M) caused a dose-dependent decrease in nitrite and nitrite + nitrate in the culture supernatant fluids after 24 h (Figure 5). Addition of MTX 10³ M caused a further reduction in nitrite and nitrite + nitrate at each concentration (P < 0.05, all comparisons except in nitrite at 10⁹ M and nitrite + nitrate at 10⁵ M).

View larger version (24K): [in this window] [in a new window]   Figure 5.   Additive inhibition by MTX (103 M) with corticosteroids of nitrite (A) and nitrite + nitrate synthesis (B). LA-4 cells were cultured for 24 h in the presence of dexamethasone at the concentrations indicated, either alone (shaded bars) or with MTX (open bars); n = 3 each condition; *P < 0.05 compared with cytomix alone.

Mechanism of Action of MTX Inhibition

Tetrahydrobiopterin (10⁴ to 10⁷ M) reversed the MTX inhibition of nitrite and nitrite + nitrate accumulation in the culture supernatant fluids (Figure 6). Higher doses of tetrahydrobiopterin ( 10³ M) appeared to be toxic to the cells, resulting in cell detachment and a reduction in nitrite and nitrite + nitrate (data not shown). Tetrahydrobiopterin at 10⁴ M caused no reversal in dexamethasone (10⁵ M) inhibition (P > 0.05), which was significantly decreased as compared with exposure to cytomix alone (22.8 ± 0.5 µM versus 34.5 ± 1.6 µM, n = 3 each condition; P < 0.05).

View larger version (17K): [in this window] [in a new window]   Figure 6.   Attenuation of MTX inhibition of nitrite (A) and nitrite + nitrate (B). LA-4 cells were cultured for 24 h in the presence of medium alone, cytomix alone, cytomix + MTX 103 M, or cytomix + MTX + tetrahydrobiopterin at the concentrations indicated; n = 3 each condition; *P < 0.05 compared with cytomix + MTX.

MTX did not alter protein expression as shown by Western blot analysis when this expression was analyzed with scanning densitometry (cells cultured in medium: protein undetectable; cells cultured with cytomix for 24 h: protein 10,832 ± 812 mg/ml; cells cultured with cytomix and MTX 10³ M: protein 10,919 ± 703 mg/ml; n = 3 each condition; P > 0.05 cytomix versus cytomix + MTX). Similarly, MTX did not alter iNOS mRNA levels expressed as a ratio to GAPDH mRNA (cells cultured in medium: iNOS mRNA undetectable; cells cultured with cytomix for 24 h: iNOS mRNA 0.16 ± 0.02; cells cultured with cytomix and MTX at 10³ M: iNOS mRNA 0.17 ± 0.03; n = 5 each condition; P > 0.05 cytomix versus cytomix + MTX).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

These studies demonstrate the reduction of NO production by MTX, and an additive effect of MTX with corticosteroids. MTX caused a dose- and time-dependent inhibition of nitrite and nitrate, two stable end products of NO production, in the culture supernatant fluids of murine lung epithelial cells. This effect did not appear to be a nonspecific effect, because cyclophosphamide, another antineoplastic agent used as an antiinflammatory agent in some lung disorders, did not inhibit nitrite and nitrite + nitrate accumulation. Furthermore, the mechanism of the effect is consistent with MTX inhibition of tetrahydrobiopterin synthesis by the dihydrofolate reductase-dependent pathway, because tetrahydrobiopterin reversed the MTX inhibition of nitrite and nitrite + nitrate, and there was no detectable effect on iNOS protein or mRNA expression.

Studies have found differing effects of MTX on NO production. Gao and colleagues (19) have reported an increase in bradykinin-induced plasma exudation with MTX in the hamster cheek-pouch model. The increase observed with MTX was attenuated with N^G-L-arginine methyl ester (L-NMMA), an NOS inhibitor, suggesting that NO was increased by MTX. Therefore, these experiments reflect alteration of a constitutive NOS. Other studies, with lipopolysaccharide (LPS)-stimulated rat aortic smooth-muscle cells, have shown a 10 to 40% reduction in nitrite accumulation with MTX (16). These studies would be most compatible with alteration of NO production by the inducible rather than by a constitutive NOS, and are consistent with the findings in our study.

The MTX inhibition of NO production by iNOS was reversed by tetrahydrobiopterin. However, Gross and Levi (16) have reported that MTX inhibition was not reversed with tetrahydrobiopterin in doses up to 300 µM, leading to the hypothesis that tetrahydrobiopterin may enter the cell as dihydrobiopterin requiring reduction back to tetrahydrobiopterin for utilization by NOS. The differences between the results of the present study and the previous study are not clear, but could be potentially explained by differences in the cell types or the experimental conditions.

The mechanism of corticosteroid inhibition of iNOS may be multifactorial. We have previously reported that dexamethasone inhibits iNOS mRNA transcription (2, 3). Simmons and coworkers (25), using cardiac microvascular endothelial cells, also reported a decrease in iNOS with dexamethasone in serum-starved cells. However, they also reported a decrease in intracellular tetrahydrobiopterin levels with dexamethasone. However, in their experiments, addition of tetrahydrobiopterin at 10⁴ M did not significantly increase interferon-- and IL-1-stimulated nitrite accumulation in the culture supernatant fluids over that in serum-starved controls treated with these cytokines. Furthermore, nitrite accumulation in the culture supernatant fluids was lower with dexamethasone treatment regardless of whether the cells were serum-starved or supplemented with tetrahydrobiopterin. These results are consistent with the results in this study. An explanation may come from Rosenkranz-Weiss and colleagues (26), who reported that inflammatory cytokines increase intracellular tetrahydrobiopterin levels in serum-cultured cells. Dexamethasone may limit tetrahydrobiopterin availability in tetrahydrobiopterin-deficient cells, but any effect may be abrogated by increased tetrahydrobiopterin synthesis with cytokine stimulation in the presence of sufficient substrate.

Asthma is a disorder associated with airway edema formation (11, 12). The inducible form of NOS is expressed by the bronchial epithelium, and exhaled NO is increased in asthmatic individuals (7). The increased NO has been hypothesized to play a role in the airway edema formation of asthma. If increased levels of NO play a role in asthma, then strategies to decrease NO production are important in therapy of this disease. The study described here suggests that NO from iNOS can be inhibited by corticosteroids, which are known to inhibit NO production by reducing transcription of iNOS mRNA (2, 3, 15). In addition, MTX has an additive effect and can further decrease NO production. This effect was not secondary to a reduction in transcription, and did not appear to be secondary to a nonspecific antimetabolic effect, because cyclophosphamide, another antimetabolite, did not have a similar effect. Rather, the reduction in NO production appeared to be secondary to reduced enzyme activity, because the effect was attenuated with tetrahydrobiopterin. Therefore, the effect of MTX on NO production occurs at a different step than the effect of corticosteroids, and the two effects would be expected to be additive. These observations suggest that strategies combining corticosteroids with MTX in vivo may produce a reduction similar to the reduction observed in vitro.

Cyclophosphamide must first be hydroxalated to be active (27). It is uncertain whether the lung epithelial cells used in this study activate cyclophosphamide. Therefore, the lack of effect of cyclophosphamide may have been due to a lack of its activation or lack of an effect on iNOS activity. However, bleomycin and Ara-C, which do not require activation (27), had no effect on nitrite accumulation, further suggesting that the effect of MTX on NO production is not due to a nonspecific antimetabolic effect.

Another potential limitation of our study was the use of the murine lung epithelial-cell line LA-4. These cells were derived from an alveolar epithelial cell (20). However, the cells were chosen because of the ease of induction of iNOS and because their response is qualitatively similar to that of human bronchial epithelial cells (2, 3).

A number of antiinflammatory therapies have been proposed for asthma when corticosteroids produce an inadequate clinical response (28). Interestingly, MTX has been the antimetabolite most commonly used in asthma, despite the use of other antimetabolites for corticosteroid-refractory pulmonary disorders such as idiopathic pulmonary fibrosis (29, 30). Whether the apparent preference for MTX in asthma is due to an effect on iNOS is unknown.

The data presented here show that MTX inhibits NO production by murine lung epithelial cells in vitro, and produces added inhibition when combined with corticosteroids. Furthermore, the results of our study imply a potential strategy for reduction of NO production in vivo by combining therapies that inhibit iNOS at different points along the pathway of NO synthesis.