Inhibition of pleural mesothelial cell collagen synthesis by nitric oxide

Free Radical Biology & Medicine, Vol. 21, No. 5, pp. 601-607, 1996 Copyright © 1996 Elsevier ScienceInc. Printed in the USA. All rights reserved 0891-5849/96 $15.00 + .00 ELSEVIER

PII S0891-5849(96)00159-1

Original Contribution INHIBITION OF PLEURAL MESOTHELIAL CELL COLLAGEN SYNTHESIS BY NITRIC OXIDE

MICHAEL W . OWENS,* SHAWN A. MILLIGAN,* and MATTHEW B. GRISHAM* *Department of Medicine, Overton Brooks Veteran Affairs and Louisiana State University Medical Centers, Shreveport, Louisiana, USA; and *Department of Physiology and Biophysics, Overton Brooks Veteran Affairs and Louisiana State University Medical Centers, Shreveport, Louisiana, USA

(Received 28 August 1995; Revised 19 February 1996; Accepted 21 March 1996)

A b s t r a c t - - T h e pleural mesothelial cell has a critical role in repairing the mesothelium after injury via its ability to produce connective tissue macromolecules. We have recently shown that proinflammatory cytokines and lipopolysaccharide induce pleural mesothelial cells to produce nitric oxide. The present study examined the effect of nitric oxide on pleural mesothelial cell protein synthesis. Rat pleural mesothelial cells were exposed to various combinations of tumor necrosis factor, interleukin-1, interferon-y, and lipopolysaccharide or to the nitric oxide donors: 6morpholino-sydnonimine, S-nitroso-N-acetyl-o,L-penicillamine, sodium nitroprusside, and spermine-NO adduct for 2 4 - 4 8 h. Nitrate and nitrite (an index of nitric oxide production) and net collagen and noncollagen protein production (uptake of 3H-proline into collagenase-sensitive protein) were then determined. Net collagen production was significantly inhibited by the cytokine-lipopolysaccharide combinations tested. Collagen inhibition paralleled the time course of increased nitric oxide production. The inhibition of collagen production was also significantly reversed by the addition of N~-nitro-L-arginine methyl ester, and was reproduced by the addition of a 5:1 molar excess of Larginine to Nt-nitro-L-arginine methyl ester. Additionally, nitric oxide-generating compounds significantly inhibited collagen production in a dose-dependent manner compared to unexposed control cells. Net collagen production was inhibited to a greater degree than noncollagen protein synthesis. These results suggest that nitric oxide may be a significant mediator of PMC collagen production during conditions of significant pleural inflammation. K e y w o r d s - - N i t r i c oxide, Collagen, Mesothelial cell, Pleura, Inflammation, Free radical

repair process likely involve a number of the cytokines, growth factors, and bacterial products that may accumulate in a variety of combinations during these conditions. For instance, tumor necrosis factor (TNF) and interferon-y are produced and selectively concentrated at sites of active disease in patients with tuberculous pleuritis. 4 TNF has been shown to stimulate in vitro PMC collagen synthesis associated with increased steady-state levels of procollagen ~lI mRNAP Additionally, lipopolysaccharide (LPS) may also be present in the pleura and pleural space when bacterial pneumonia causes an influx of bacteria into the pleural space with the formation of an e m p y e m a or after the inhalation of asbestos particles contaminated with LPS. The effects of this complex inflammatory environment on the repair of the injured mesothelium by the PMC are not completely understood.

INTRODUCTION

The pleura consists of a single layer of mesothelial cells, beneath which is an extracellular matrix that is composed primarily of collagen and elastin. 1'2 A variety of systemic and pulmonary diseases, such as tuberculosis, bacterial pneumonia with empyema, rheumatoid arthritis, asbestosis, and malignancy, can affect the pleura resulting in significant inflammation and injury. The pleural mesothelial cell (PMC) has a critical role in repairing the pleura after injury via its ability to produce all of the connective tissue macromolecules found in the normal pleural extracellular matrix, particularly types I and III collagen) The factors that affect this Address correspondence to: Michael W. Owens, M.D., Pulmonary and Critical Medicine (111P), Overton Brooks VAMC, 510 E. Stoner Avenue, Shreveport, LA 71101-4295. 601

602

M. W. OWENS et al. Table t. [3H]-Proline (% Control) 24 h IFN IFN 48 h IFN IFN

+ LPS + IL-I + LPS + TNF

95 _+ 14 114 + 17

+ LPS + IL-1 + LPS + TNF

60 _+ 11" 71 _+ l l *

CollageJ~Fotal Protein (% Control)

85 _+ 2 92 -- 2 54 + 4* 53 _+ 11"

[3H]-Leucine (% Control)

Nitrite (nmol per I(P Cells)

99 _+ 6 87 _+ 7

10 + 1 16 + 2

84 _+ 5 100 + 29

147 + 1" 142 _+ 1"

* p < .05, sample vs. control.

We have recently shown that specific combinations of proinflammatory cytokines and LPS can induce the PMC to produce large amounts of nitric oxide (NO). 6 NO is a free radical that is derived from the terminal guanidino nitrogen of L-arginine. 7 It has been shown to have a number of important intracellular and extracellular effects. 8-1~ NO is an activator of guanylate cyclase by binding with heme at its active site, with a resultant increase in cyclic-GMP production. ~2 NO also inhibits a number of other iron-containing enzymes such as nitrogenase, cyclooxygenase, and lipoxygenase; secondary to its ability to form complexes with iron] 3 NO has recently been shown to inhibit total protein synthesis in tumor cells, hepatocytes, and smooth muscle c e l l s . 14-16 In this article we demonstrate that NO, whether produced by the PMC or added exogenously, causes a selective inhibition of PMC collagen synthesis.

M A T E R I A L S AND METHODS

Reagents Human recombinant TNF (10 7 units/mg protein) was obtained from Amgen (Thousand Oaks, CA), murine recombinant interleukin- 1 (IL- 1) (3.5 × 10 5 units/ mg protein) from Genzyme (Cambridge, MA), and rat interferon-y (IFN-y) (4 × 10 6 units/rag protein) from Gibco (Gaithersburg, MD). S-nitroso-N-acetyl-D,Lpenicillamine (SNAP) was purchased from Affinity Bioreagents (Neshanic Station, N J). 6-Morpholinosydnonimine (SIN-I) was obtained from Cassela AG (Frankfurt, Germany). Spermine-NO adduct was purchased from Cayman Chemical Company (Ann Arbor, MI). [3H]-proline and [3H]-leucine were purchased from New England Nuclear (Boston, MA). H a m ' s F12 medium, phenol red-free Dulbecco' s modified Eagle medium (DMEM) and dialyzed fetal bovine serum

120

$ A

=m .~

10o

o1 c o o

80

-" 0

,,~

.,9~ C

~ C

60

2

40

~

20

o,. a..

.~

-t-

0

Control

Cytomix

Cytomix

Cytomix

+

+

L-NAME

L-NAME +

L-Arginine Fig. 1. Reversal of cytokine-induced inhibition of collagen production by L-NAME (1 mM). Cells were exposed to cytomix [TNF (50 ng/ml) + LPS (50/zg/ml) + IFN gamma (500 U/ml)] for 48 h. L-NAME (1 mM) was added to some wells to block NO production and determine if NO was causing inhibition of collagen production. An excess of L-arginine (5 raM) was added to competitively inhibit L-NAME and reproduce the effect. Results are expressed as mean + SEM (n = 3). Three separate experiments yielded similar results. *p < .05, samples vs. control.

Inhibition of pleural mesothelial cell collagen synthesis

603

120 A

i!o

lOO

=(.1 o

o'6

!!o !i

i

X 0

120

100 o c

G.

¢

& t o U

2o o 0

2,50

500

1000

S N A P (,.M)

Fig. 2. Effect of SNAP on PMC collagen production. Cells were exposed to SNAP for 48 h. Results are expressed as the mean + SEM (n = 3). Three separate experiments yielded similar results. *p < .05, samples vs. control. were purchased from Gibco (Grand Island, NY). LPS was obtained from Difco Laboratories (Detroit, MI). Superoxide dismutase (SOD) was obtained from Pharmacia (Uppsala, Sweden). Sodium nitroprusside (SNP), hemoglobin, sulfanilamide, naphthalene diamine dihydrochloride, and NG-nitro-L-arginine methyl ester (L-NAME) were obtained from Sigma (St. Louis, MO), as were all other chemical reagents. Oxyhemoglobin was prepared as previously described. 17

Nitrite and nitrate production

Cell culture and treatment

The measurement of nitrite and nitrate using the Greiss reagent was used as an index of NO production as previously described. 6

Rat PMC were routinely grown in H a m ' s F-12 medium supplemented with 10% heat-inactivated fetal bovine serum, glutamine (2 mM) and gentamicin (50 #g/ ml) in a 95% air 5% CO2, humidified atmosphere as

previously described. 6 Confluent monolayers of PMC were exposed to IFN-T + LPS + TNF, IFN-T+ LPS + IL-1, SNAP, SIN-l, SNP, spermine-NO, or control solutions (medium only with 1% fetal bovine serum) for 2 4 - 4 8 h. After incubation samples were harvested for collagen, total protein, and nitrite/nitrate measurement as described below.

Collagen and total protein production Collagen production by confluent monolayers of PMC was determined by measuring the incorporation

604

M . W . OWENS et al. 120 -

$ ¢1 ~

"6 ~ O

lOO-

80-

o

9C ~~ m

~_

60-

40-

"6'~ L. r,

20,

¢~

e~

Control

SNAP

SNAP +

HEMOGLOBIN

Fig. 3. Effect of oxyhemoglobin on SNAP-induced PMC collagen production inhibition. Cells were exposed to SNAP _+ oxyhemoglobin (25/zM) for 48 h. Results are expressed as the mean _+ SEM (n = 3). Three separate experiments yielded similar results. *p < .05, samples vs. control.

of [3H]-proline (5 #Ci/ml for the final 12 h of treatment) into bacterial collagenase-sensitive protein as previously described. 5 Collagen production is expressed as degradations per minute per 106 cells (as determined from a parallel plate), and then expressed as a percent of control and as a percent of total protein synthesis. The incorporation of [3H]-leucine (5 #Ci/ ml for the final 12 h of treatment) was used as an additional method to assess total protein synthesis in some experiments)

Statistical analysis Data is expressed as mean ± standard error of the mean. The significance of differences between experimental groups was determined by Student's t-test. Statistical significance was defined as a p <- .05. RESULTS

Exposure of PMC to cytokine-LPS combinations for 24 h did not significantly affect the uptake of [3H]proline into collagenase-sensitive protein and was associated with minimal levels of NO production (Table 1). However, net collagen production was significantly inhibited by exposure to cytokine-LPS combinations for 48 h as demonstrated by both the uptake of [3H]proline into collagenase-sensitive protein and the collagen/total protein ratio (Table 1). This inhibition of collagen production was temporally associated with higher levels of NO production ( > 140 nmol 106 cells)

(Table 1). Additionally, the inhibition of collagen production was significantly reversed by L-NAME, a competitive inhibitor of nitric oxide synthase, and was reproduced by the addition of a 5:1 molar excess of L-arginine to L-NAME, suggesting that the observed inhibition of collagen production was due to PMC NO production (Fig. 1). Total protein synthesis measured by the uptake of [3H]-leucine was not significantly affected, establishing in an independent manner that the observed effects are specific for collagen (Table 1). The decreases in collagen were not secondary to cell death because these same combinations did not decrease viable cell counts using the trypan blue exclusion method. 6 In addition to endogenous NO, a variety of exogenous NO donors were also tested for their ability to affect collagen synthesis. SNAP caused a dose-dependent inhibition of both the uptake of [3H]-proline into collagenase-sensitive protein and the collagen/total protein ratio, resulting in a maximal inhibition of 58 and 50% relative to control, respectively (Fig. 2). The decrease in collagen production by SNAP was inhibited by oxyhemoglobin, which binds and inactivates NO (Fig. 3). To further establish that this effect was due to the release of NO, SIN-1 was also studied. SIN-1 significantly decreased collagen production (Fig. 4). SOD further augmented the inhibitory effect of SIN- 1 on collagen production Fig. 5). It is known that SIN-1 releases both superoxide radicals and NO, which may react to yield peroxynitrite.18 The addition of SOD may have decreased collagen production by scavenging the

Inhibition of pleural mesothelial cell collagen synthesis

605

120

==o

100

m 0 0

o"6

80

v

60 Q,G. ~) 4) • >

40

"e

20

~

,I-

0 120,

~0 u

n

IO0

8o c

G.

I-e,

4) OD _m 0

o

20 0

0

100

500

1000

SIN-1 (pM) Fig. 4. Effect of SIN-1 on PMC collagen production. Cells were exposed to S1N-1 for 48 h. Results are expressed as the mean ± SEM (n = 3). Three separate experiments yielded similar results. *p < .05, samples vs. control.

superoxide radicals, thereby prolonging the activity of NO. Conversely, the addition of oxyhemoglobin significantly inhibited the decrease in collagen synthesis by SIN-1 (Fig. 5). SNP (500 #M) and spermine-NO (1000 #M) also significantly inhibited collagen production, resulting in a maximal inhibition of 50 _+ 2 and 46 _ 3% relative to control, respectively (p < .05). DISCUSSION

In the present study two different cytokine-LPS combinations caused a selective inhibition of collagen production that was dependent on the formation of NO. Additionally, a similar degree of inhibition of collagen production was seen when NO was added in the form of four structurally different NO-generating compounds. These results demonstrate that NO, whether

produced endogenously by the cell or added exogenously to the cell, can cause a significant and selective inhibition of collagen synthesis in PMC. These results suggest that NO produced by the PMC itself or by another cell in the local environment, such as a macrophage, could significantly affect the biosynthesis of matrix macromolecules by the PMC. NO has been shown to inhibit total protein synthesis in tumor cells, smooth muscle cells and hepatocytes in vitro, requiring relatively high concentrations ( > 100 #M/24 h for hepatocytes). 14-16Recent studies have suggested that NO may also be capable of affecting the synthesis of specific matrix proteins. NO has been shown to mediate the suppression of cartilage proteoglycan synthesis by interleukin-1.19 Rat mesangial cells exposed to IFN-7 and LPS had an inhibition of collagen and fibronectin synthesis and a stimulation of lam-

606

M . W . OWENS et al.

=~

4

o

£.E a.~ ~,£

2

v

CONTROL

!i SIN-1

BB SIN-1

SIN-1

+

+

SOD

SOD +

HEMOGLOBIN

Fig. 5. Effect of SIN-I on PMC collagen production. Cells were exposed to SIN-1 for 48 h. Some wells with SIN-I also received superoxide dismutase (100 #g/ml) in order to scavenge superoxide radicals and prevent their interaction with NO (prolonging the activity of NO). Some wells received oxyhemoglobin (25 #M). Results are expressed as the mean _+ SEM (n - 3). Three separate experiments yielded similar results. *p < .05, samples vs. control; *p < .05, SIN-1 + SOD + oxyhemoglobin vs. SIN1 + SOD.

inin synthesis, all of which was at least partially dependent on the formation of N O . 2° However, another study in smooth muscle cells exposed to NO donors did not find a relative decrease in collagen synthesis compared to total protein synthesis (collagen synthesis was decreased, but total protein synthesis was decreased proportionately). While the reasons for these differences are unclear, it is known that IFN-y can inhibit collagen synthesis in ways not dependent on the production of NO. 2~ However, in the present study PMC were exposed to NO donors, as well as cytokineinduced NO, suggesting that NO was the mediator of the observed effects. The mechanisms of the inhibition of protein synthesis by NO are only speculative at this point. In hepatocytes, the inhibition of general protein synthesis does not appear to involve NO-mediated decreases in guanylate cyclase activity or alterations of energy utilization by mitochondria, as do the vasodilatory and antiproliferative properties of N O . j5'22 Interestingly, the inhibition of protein synthesis in both tumor cells and hepatocytes is associated with a loss of intracellular iron, the significance of which is unclearY Inhibition of collagen synthesis could occur at various levels, including a decrease in collagen gene transcription, a decrease in the translation of procollagen mRNA, or an increase in the degradation of collagen. NO-mediated decreases in hepatocyte albumin synthesis (representative of total protein synthesis inhibition) are not associated with changes in steady-state levels of albumin

mRNA, suggesting that decreases in total protein synthesis may be mediated at the level of translation or posttranslation] 5 The effect of NO on PMC collagen gene transcription may be different, because collagen is selectively inhibited. NO has been shown to increase m R N A levels for TNF-c~ and IL-1/3 and decrease mRNA for proto-oncogenes c-myc and c-myb in human leukemia cells, demonstrating that NO can have selective effects on gene transcription. 24 Additionally, recent studies have demonstrated that chondrocytes stimulated with IL-1 have an increase in gelatinase synthesis that is dependent on the generation of " l o w " levels of NO ( < 125 nanomoles/106 cells), suggesting that NO could contribute to a net decrease in collagen production by stimulating metalloproteinase activity. 25"26 Cytokines released within the lung interstitium, pleura, and pleural space are known to initiate further manifestations of pleural inflammation through their direct effect on the PMC. 5-6'27We have previously shown that TNF, epidermal growth factor, and platelet-derived growth factor increase collagen synthesis in PMC (none of these factors stimulate PMC NO production when added alone), s'27 While these individual factors increase collagen synthesis, it is more likely in vivo that the PMC will be exposed to multiple factors in a variety of combinations, depending on the relative tempo with which an inflammatory response develops. The PMC may respond to this complex inflammatory environment in a manner that is entirely different than would be predicted from studies of individual factors,

Inhibition of pleural mesothelial cell collagen synthesis

for example, NO production with subsequent collagen inhibition as opposed to collagen stimulation. These results suggests that NO may compromise the ability of the PMC to repair the pleura during conditions associated with significant pleural inflammation. Alternatively, it could be postulated that the formation of NO might have a role in preventing the excessive formation of fibrotic tissue. Acknowledgements - - This work was supported by a Merit Review Grant from the Veterans Affairs Medical Research Service (M.W.O.) and National Institute of Diabetes and Digestive and Kidney Disorders Grant DK-43785 (M.B.G.).

14. 15.

16.

17.

18. REFERENCES

1. Agostoni, E. Mechanics of the pleural space. Physiol. Rev. 52:57-128; 1972. 2. Miller, W. S..The lung. Springfield, IL: Thomas; 1947. 3. Rennard, S. I.; Jaurand, M.; Bignon, J.; Kawanami, O.; Ferrans, V. J.; Davidson, J.; Crystal, R. G. Role of mesothelial and submesothelial stromal cells in matrix remodeling following pleural injury. Am. Rev. Respir. Dis. 130:267-274; 1984. 4. Barnes, P. F.; Fong, S.; Brennan, P. J.; Twomey, P. E.; Mazumder, A.; Modlin, R. L. Local production of tumor necrosis factor and interferon gamma in tuberculous pleuritis. J. lmmunol. 145:149-154; 1990. 5. Owens, M. W.; Grimes, S. R. Pleural mesothelial cell response to inflammation: Tumor necrosis factor-induced mitogenesis and collagen synthesis. Am. J. Physiol 265:L382-L388; 1993. 6. Owens, M. W.; Grisham, M. B. Nitric oxide synthesis by rat pleural mesothelial cells: Induction by cytokines and lipopolysaccharide. Am. J. Physiol. 265:L110-L116; 1993. 7. Palmer, R. M. J.; Ashton, D. S.; Moncada, S. Vascular endothelial cells synthesize nitric oxide from L-arginine. Nature 333:664-666; 1988. 8. Ignarro, L. J. Biosynthesis and metabolism of endothelium-derived nitric oxide. Annu. Rev. Pharmacol. ToxicoL 30:535-560; 1990. 9. Marietta, M. A.; Tayeh, M. A.; Havel, J. M. Unraveling the biological significance of nitric oxide. BioFactors 2:219-225; 1990. 10. Moncada, S.; Palmer, R. M. J.; Higgs, E. A. Nitric oxide: Physiology, pathophysiology, and pharmacology. PharmacoL Rev. 43:109-142; 1991. 11. Nathan, C. F.; Hibbs, J. B., Jr. Role of nitric oxide synthesis in macrophage antimicrobial activity. Curr. Opin. Immunol. 3:6570; 1991. 12. Palmer, R. M. J.; Ferrige, A. G.; Moncada, S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature 327:524-526; 1987. 13. Kanner, J.; Harel, S.; Granit, R. Nitric oxide, an inhibitor of lipid

19.

20.

21.

22.

23.

24. 25.

26.

27.

607

oxidation by lipoxygenase, cyclooxygenase and hemoglobin. Lipids 27:46-49; 1992. Stuehr, D. J.; Nathan, C. F. Nitric oxide. A macrophage product responsible for cytostasis and respiratory inhibition in tumor target cells. J. Exp. Med. 169:1543-1555; 1987. Curran, R. D.; Ferrari, F. K.; Kispert, P. H.; Stadler, J.; Stuehr, D. J.; Simmons, R. L.; Billiar, T. R. Nitric oxide and nitric oxidegenerating compounds inhibit hepatocyte protein synthesis. FASEB J. 5:2085-2092; 1991. Kolpakov, V.; Gordon, D.; Kulik, T. J. Nitric oxide-generating compounds inhibit total protein and collagen synthesis in cultured vascular smooth muscle cells. Circ. Res. 76:305-309; 1995. Martin, W.; Villani, G. M.; Jothianandan, D.; Furchgott, R. F. Selective blockade of endothelium-dependent and glyceryl trinitrate-induced relaxation by hemoglobin and by methylene blue in the rabbit aorta. J. PharmacoL Exp. Ther. 232:708-716; 1985. Feelisch, M.; Ostrowsli, J.; Noack, E. On the mechanism of NO release from sydnonimines. J. Cardiovasc. Pharm. 14(Suppl. 11):S13-$22; 1989. Taskiran, D.; Stefanovic-Racic, M.; Georgescu, H.; Evans, C. Nitric oxide mediates suppression of cartilage proteoglycan synthesis by interleukin-1. Biochem. Biophys. Res. Commun. 22:142-148; 1994. Trachtman, H.; Futterweit, S.; Singhal, P. Nitric oxide modulaltes the synthesis of extracellular matrix proteins in cultured rat mesangial cells. Biochem. Biophys. Res. Commun. 207:120-125; 1995. Rosenbloom, J.; Feldman, N. G.; Freundlich, B.; Jimenez, S. A. Transcriptional control of human diploid fibroblast collagen synthesis by "y-interferon. Biochem. Biophys. Res. Commun. 123:365-372; 1984. Stadler, J.; Curran, R. D.; Ochoa, J. B.; Harbrecht, B. G.; Hoffman, R. A.; Simmons, R. L.; Billiar, T. R. Effect of endogenous nitric oxide on mitochondrial respiration of rat hepatocytes in vitro and in vivo. Arch. Surg. 126:186-191; 1991. Billiar, T. R.; Curran, R. D.; Stuehr, D. J.; Ferrari, F. K.; Simmons, R. L. Evidence that activation of Kupffer cells results in production of L-arginine metabolites that release cell-associated iron and inhibit hepatocyte protein synthesis. Surgery 106:364372; 1989. Magrinat, G.; Mason, S. N.; Shami, P. J.; Weinberg, J. B. Nitric oxide modulation of human leukemia cell differentiation and gene expression. Blood 80:1880-1884; 1992. Stadler, J.; Stefanovic-Racic, M.; Billiar, T. R.; Curran, R. D.; Mclntyre, L. A.; Georgescu, H. I.; Simmons, R. L.; Evans, C. H. Articular chondrocytes synthesize nitric oxide in response to cytokines and lipopolysaccharide. J. lmmunol. 147:3915-3920; 1991. Murrell, G. A. C.; Jang, D.; Williams, R. J. Articular chondrocytes synthesize nitric oxide in response to cytokines and lipopolysaccharide. Biochem. Biophys. Res. Commun. 206:15-21; 1995. Owens, M. W.; Milligan, S. A. Growth factor modulation of rat pleural mesothelial cell mitogenesis and collagen synthesis. Effects of epidermal growth factor and platelet-derived growth factor. Inflammation 18:77-89; 1994.