Interleukin 4 Inhibition of Prostaglandin E2 Synthesis Blocks Interstitial ...

Vol. 267, No. 1, Issue of January 5,$: 515-519,1992 rrnted in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Interleukin 4 Inhibition of Prostaglandin E2 Synthesis Blocks Interstitial Collagenase and92-kDa Type IV Collagenase/Gelatinase Production byHuman Monocytes* (Received for publication, July 26,1991)

Marta L. CorcoranS, WilliamG. Stetler-Stevensons, Peter D. Browns, and Larry M. WahlSV From the $Cellular ImmunologySection, Laboratory of Immunology, National Institute of Dental Research and the §Tumor Invasion and Metastasis Section, Laboratory of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892

Activation of human monocytes results in the production of interstitial collagenase through a prostaglandin Ez(PGEz)-CAMP-dependent pathway. Inasbeen shown to inhibit much as interleukin 4 (IL-4) has PGEz synthesis by monocytes, we examined the effect of IL-4 on the production of human monocyte interstitial collagenase. Additionally, we also assessed the effect of IL-4 on the production of 92-kDa type IV collagenase/gelatinase and tissue inhibitor of metalloproteinase-1 (TIMP-1) bymonocytes. The inhibition of PGEz synthesis by IL-4 resulted in decreased interstitial collagenase protein and activity that could be restored by exogenous PGEz or dibutyryl cyclic AMP (BtzcAMP).IL-4 also suppressed ConA-stimulated 92kDa type IV collagenase/gelatinase protein and zymogram enzyme activity that could be reversed by exogenous PGEz or BtzcAMP. Moreover, indomethacin suppressed the ConA-induced production of 92-kDa type I V collagenase/gelatinase. These data demonstrate that, like monocyte interstitial collagenase, the ConAinducible monocyte 92-kDa type IV collagenase/gelatinase is regulated through a PGEz-mediated CAMPdependent pathway. In contrast to ConA stimulation, unstimulated monocytes released low levels of 92-kDa type IV collagenase/gelatinase that were not affected by IL-4, PGEz, or BtzcAMP, indicating that basal production of this enzyme is PGEz-CAMPindependent. IL4 inhibition of both collagenases was not a result of increased TIMP expression since Western analysis of 28.6-kDa TIMP-1 revealed that IL-4 did not alter the increased TIMP-1 protein in response to ConA. These data indicate that IL-4 may function in natural host regulation of connective tissue damage by monocytes.

Interleukin 4(IL-4),' originally described as a B cell growth factor (l), is a 20-kDa product of activated T cells that has been shown to modulate various functions of hematopoietic cells (2, 3). IL-4 has been noted to suppress many human monocyte functions, such as production of tumor necrosis factor a, IL-1 (4, 5), PGE2 (4), antileishmanial activity and hydrogen peroxide (6), superoxide (7), IL-6 (8), and IL-8 (9).

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ?l To whom correspondence should be addressed Bldg.30, Rm. 325, NIDR, NIH, Bethesda, MD 20892. Tel.: 301-496-9219. ' The abbreviations used are: IL, interleukin; TIMP, tissue inhibitor of metalloproteinase; PGE2, prostaglandin Ez; Bt2cAMP, dibutyryl cyclic AMP; DMEM, Dulbecco's modified Eagle's medium.

Inhibition of monocyte PGE2 synthesis by IL-4 suggested that this cytokine might also down-regulate interstitial collagenase production. Previous findings have demonstrated that suppression of PGE, synthesis of stimulated monocytes/ macrophages, by agents such as indomethacin (10, l l ) , dexamethasone (11, 12), or interferon-y (13), inhibits the activation sequence leading to the production of monocyte/macrophage interstitial collagenase. Moreover, the suppression of collagenase by these agents can be reversed with PGEz or Bt2cAMP, demonstrating a PGEz-mediated CAMP-dependent step (10-13) in the signal transduction pathway leading to monocyte/macrophage collagenase synthesis. In addition to producing interstitial collagenase (11, 1418), which degrades type I, 11, and I11 collagens (19), human peripheral blood monocytes have been shown to produce other enzymes and inhibitors that may have important roles in the degradation of connective tissue matrix. These include tissue inhibitors of metalloproteinases (TIMP) (18) and a 92-kDa type IV collagenase/gelatinase (20), which degrades basement membrane or type IV collagen, type V collagen, and gelatin (21). We therefore examined the effect of IL-4 on the production of interstitial collagenase, 92-kDa type IV collagenase/ gelatinase, and TIMP-1by human monocytes. MATERIALSANDMETHODS

IL-4"Human recombinant IL-4 was obtained from Sterling Pharmaceutical (Malvern, PA). The specific activity of this preparation was 1.5 X lo7 units/mg. Human Monocytes-Human peripheral blood mononuclear cells were obtained by leukapheresis of normal volunteers at the blood bank of the National Institutesof Health. After density sedimentation of the mononuclear cells with lymphocyte separation medium (Organon Teknika, Durham, NC),the monocytes were purified by counterflow centrifugal elutriation, as previously described (22), except that pyrogen-free phosphate-buffered saline was used in the elutriation procedure. Monocytes were enriched to >go%, as determined by morphology, nonspecific esterase staining, and flow cytometry (22). Moreover, the purification procedure did not activate the monocytes, as shown by the fact that following overnight incubation at 37 "C in suspension, less than 4% of these cells were IL-2 receptor positive, a sensitive marker of monocyte activation (23). Culture Conditions-Purified monocytes were cultured either adherent in 75-cm2 tissue culture flasks at a concentration of 4 X lo6 cells/ml of serum-free Dulbecco's modifiedEagle's medium (DMEM; Mediatech Inc., Herndon, VA) or in 35-mm Petri dishes at 4 X lo6 cells/ml. Control or concanavalin A (Conk Ca1biochem)-stimulated monocytes were cultured in the presence or absence of IL-4, which was added a t the time of plating and 1 h prior to the addition of ConA. Bt2cAMP, PGE2, and indomethacin (Sigma) were also added to some of the cultures. The culture media were harvested at 48 h and stored at -20 'C. PGEz Assay-PGE2 levels in the media supernatants from monocyte cultures were determined by radioimmunoassay, as described

515

516

Inhibition of Human Monocyte

(24), using rabbit anti-PGEZ antiserum obtained from Dr. L. Levine (Brandeis University, Waltham, MA). Interstitial Collagenase Activity Assay-Supernatants from monocyte cultures were dialyzedfor 3 days against daily changes of distilled H,O buffered to pH 7.5 with Tris and lyophilized. Following resuspension of the concentrated media products in 300 p1 of 0.05 M TrisHCl, pH 7.5, containing 0.2 M NaCl and 5 mM CaCl,, the samples were assayed for collagenase activity on acid-extracted [14C]glycinelabeled collagen fibrils, as previously described (25). Collagenase activity was determined by the amount of ["Clglycine solubilized minus the [14C]glycineliberated by 0.01% trypsin. One unit of collagenase activity represents the hydrolysis of 1 pg of collagen/min at 35 "C. WesternAnalysis of Metalloproteinases and TZMP-I-Supernatants from monocyte cultures that had been dialyzed and lyophilized, as described above, were resuspended in loading dye (0.5 M Tris-HC1 (pH 6.8), 10% (w/v) sodium dodecyl sulfate, 0.1% bromphenol blue, 20% glycerol), reduced (1% p-mercaptoethanol), heated for 2 min at 95 "C, loaded, and electrophoresed on adiscontinuous 10%polyacrylamide gel. After electrophoresis, the proteins were transferred in electroblotting buffer (20 mM Tris (pH 8.0), 150 mM glycine, 20% MeOH) overnight a t 25-30 V. Western blots were blocked in agelatin, bovine serum albumin, milk solution (3% gelatin, 3% bovine serum albumin, 5% nonfat dry milk, 150 mM NaCl, 50 mM Tris, 0.05% Tween) a t room temperature or 4 "C until solidified. The blots were exposed to primary antibodies (interstitial collagenase, 92-kDa type IV collagenase/gelatinase and TIMP-1) andsubsequently developed by using goat anti-rabbit IgG horseradish peroxidase in conjunction with an enhanced chemiluminescence Western blotting detection system (Amersham Corp.) Preparation of Antibodies-Synthetic peptides corresponding to specific amino acid sequences found in interstitial collagenase and 92-kDa type IV collagenase/gelatinase were synthesized on a Biosearch 9600 peptide synthesizer using t-butoxycarbonyl amino acid methodology. Peptides were coupled to keyhole limpet hemocyanin using 0.14% gluteraldehyde, and antibodies to these peptides were generated by injecting them into New Zealand Whiterabbitsin Freund's adjuvant, as previously described (26). Antibody 125, which recognizes the latent and active forms of interstitial collagenase, was made against peptide 125 with the amino acid sequence VLTEGNPRWEQTHLRYRI, which corresponds to amino acids 102-119of the interstitial procollagenase sequence (19). Antibody 110 recognizes 92-kDa type IV collagenase/gelatinase and was made against peptide 110 with the amino acid sequence EALMYPMYRFTEGPPLHK, which corresponds to amino acids 397-414of this enzyme. The antibodies were affinity-purified by applying the 50% (NH4),S04 precipitate from heat-inactivated antiserato peptide affinity columns in which the antigen peptide was coupled to Affi-Gel 10 (Bio-Rad), according to the manufacturer's directions. Antibodies were eluted using 1 M acetic acid, neutralized with 2 M Tris, dialyzed against phosphate-buffered saline, and stored at -80 "C. Zymogram Analysis of 92-kDa Type ZV CollagenaselGelatineConcentrated monocyte supernatants were prepared in loading dye as described above for Western analysis, with the exception that the samples were not reduced or heated, and loaded on a discontinuous 10% polyacrylamide gel containing 0.1% gelatin. Conditioned media from phorbol ester-stimulated HT1080 fibrosarcoma cells that overproduce the 92-kDa type IV collagenase/gelatinase was used as a standard. Following electrophoresis, the gel was incubated in collagenase buffer (described above under interstitial collagenase assay) containing 2.5% Triton X-100 for 30-60 min at room temperature and subsequently incubated in collagenase buffer for 30-60 min at 37 "C. The gels were stained overnight in Coomassie Blue (0.25% Coomassie Blue, 45.4% methanol, and 9.2% glacial acetic acid), destained (75% ethanol and 25% glacial acetic acid), and dried. The relative degree of digestion on the zymograms was analyzed using an Ultroscan XL laser densitometer (Pharmacia LKB Biotechnology Inc.). The relative densitometric areas were calculated by integrating the area under each peak and theareas expressed as arbitrary units. RESULTS

Effect of IL-4 on Monocyte PGE, and InterstitialCollagenase-The data presented in Fig. 1 confirm previous findings (4) that IL-4 is a potent inhibitor of PGEz production, as shown by the significant inhibition of ConA-induced PGE, production by 15 units/ml (1 ng/ml) of IL-4. Depending on

Collagenase Production by

Control

IL-4

Con

A

0.15 1.5

15

Con AtIL-4 lunits/nl)

FIG. 1. Effect of IL-4 on monocyte PGEa production. Purified monocytes were cultured in DMEM at 4 X 106/2m1/35-mm Petri dish in the presence or absence of IL-4 (at the indicated concentrations) for 1 h, and then ConA (20 pg/ml) was added to some of the cultures. The media supernatants were harvested 24 hafter the addition of ConA and assayed for PGE, by radioimmunoassay. The data represent the mean (eS.D.) of duplicate samples. Suppression of PGE, production by IL-4 confirms previous findings (4).

* -

:.

.n

2 w

700 600

5

500

: 6 m~

400

u

0

5 c 300 ola m -

a .

:: 200

- 3 E

I

100

0

Control Con A

0.15 1.5 150 15 t IL-4 lunits/m11

Con A

FIG. 2. Effect of IL-4 on monocyte interstitial collagenase activity. Purified monocytes were cultured in 75-cm2culture flasks a t 4 X 107/10 ml of DMEM. 1L-4 was added to some of the cultures just prior to the addition of ConA (20 pg/ml). Collagenase activity, expressed in milliunits ( m u ) ,was determined in 48 h, as described under "Materials and Methods," with 1 unit of activity representing the digestion of 1pg of collagen/min at 35 "C.The datarepresent the mean (k S.D.) of duplicate cultures.

the donors, IL-4 inhibited 70-95% of the ConA-induced PGE, production. The ability of IL-4 to block PGEz production suggested that PGE,-cAMP-dependent monocyte interstitial collagenase production would also be inhibited. As shown in Fig. 2, IL-4 caused adose-dependentinhibition of ConAinduced monocyte interstitial collagenase bioactivity. In this experiment, IL-4 (15 units/ml) inhibited approximately 50% of the collagenase activity induced by ConA. Addition of IL4 1h before ConA resulted in a greater suppression of interstitial collagenase (as shown in Fig. 3). Restoration of Interstitial Collagenase Productionin IL-4-treated Monocytes by PGE, and Bt2cAMP-PGE2 or BtzcAMP were added to IL-4-treated monocytes to determine if the mechanism by which IL-4 inhibited collagenase synthesis was related to the suppression of the PGE2-CAMPpathway. Incubation of monocytes with IL-4 for 1 hprior to stimulation with ConA resulted in a significant inhibition of interstitial collagenase activity that could be restored with the addition of PGEz (Fig. 3). In addition to examining the effect of IL-4 on interstitial collagenase enzymatic activity on radioactive collagen gels, interstitial collagenase protein production was also evaluated by Western analysis. As shown in Fig. 4, ConA induced substantial amountsof interstitial collagenase protein at molecular weights of approximately 47,000 and 42,000, corresponding to the partially active and active forms of the fibroblast enzyme, respectively (27). In contrast, supernatants from control cells contained only small amounts of prointerstitial collagenase (57 and 52 kDa) with no detect-

Inhibition of Human Monocyte Collagenase Production

by IL-4

-97

t

IL-4

517

t

PGEZ

-66

(M)

FIG.3. Restoration of interstitial collagenase activity in IL4-treated monocytes by PGEZ. 4 X 10' monocytes were cultured in 10 ml of DMEM in 75-cm' culture flasks. Some of the cultures received IL-4 (15 units/ml)at the time of plating, and 1h later, ConA (20 pg/ml) was added in the presence or absence of the indicated concentrations of PGE2. The culture media supernatants were harvested a t 48h, dialyzed, lyophilized, and assayed for collagenase activity. The data represent the mean(k S.D.) of duplicate cultures. M , molar; mu,milliunits.

1 2 3 4 5

4' -97 I46

45 31

21 FIG.4. Restoration of interstitial collagenase protein in IL4-treated monocytes by PGEZ and BtzcAMP. Media from 4 X

B

-45

-31

FIG.5. Effect of IL-4 on 92-kDa type IV collagenase/gelatinase. Media from 4 x 10' monocytes cultured for 48 h in 10 ml of DMEM in 75-cm2 culture flasks were dialyzed and lyophilized, and the resuspended supernatants, equivalent to 2 x IO', were analyzed for type IV collagenase by zymogram ( A )and supernatantsequivalent to 3 X lo7 were analyzed by Western blot ( B ) . The experimental conditions examined in each lane are control(lane I), control + 0.15 unit/ml of IL-4 (lane 2), control + 1.5 unit/ml of IL-4 (lane 3 ) , control 15 units/ml of IL-4 (lane 4 ) , control 150 units/ml of IL4 (lane 5 ) , ConA (20 pg/ml) (lane 6 ) , ConA 0.15 unit/ml of IL-4 (lane 7), ConA 1.5 units/ml of IL-4 (lane 8 ) , ConA + 15 units/ml of IL-4 (lane 9),and ConA 150 units/ml of IL-4 (lane 10). IL-4 was added 1 h prior to the addition of ConA. The relative enzyme activity detected on the zymogram was measured by densitometry, and the arbitrary unitsfor each lane are4.3 (lane Z), 3.8 (lane 2), 3.8 (lane 3 ) . 3.7 (lane 4 ) , 3.2 (lane 5 ) , 16.7 (lane 6 ) , 15.7 (lane 7), 14.1 (lane 8 ) , 8.3 (lane 9),and 4.9 (lane 10). A standard ( S ) for 92-kDa type IV collagenase/gelatinasefrom HT1080 fibrosarcomacells is shown on the zymogram, and the location of the molecular weight standards are shown on the Western.

+

+

+ +

+

lo7 monocytes cultured for 48 h in 10ml of DMEM in 75-cm2culture the legend for Fig. 5 ) , the Con A-induced 4-fold increase in flasks were dialyzed and lyophilized, and the resuspended superna- 92-kDa type IV collagenase/gelatinase was reduced in a dosetants, equivalent to 3 X IO' monocytes, were analyzed for interstitial dependent manner by IL-4 until it reached control levels at collagenase protein by Western analysis. The experimental conditions 150 units/ml of IL-4. examined in each lane are control (lane l ) , ConA (20 pg/ml) (lane Although the monocyte 92-kDa type IV collagenase/gela2),ConA + IL-4 (15 units/ml) (lane 3 ) , ConA + IL-4 + PGEZ tinase was predominantly expressed in the latent form (92M ) (lane 4 ) , and ConA IL-4 Bt2cAMP M) ( l a n e 5 ) . IL-4 kDa), asseen in thezymogram, media from ConA-stimulated was added 1 h prior to the addition of the other reagents.

+

+

monocytes also contained activity a t approximately 84 kDa (band below the 92 kDa band in lanes 6-a), the molecular able active enzyme. IL-4 caused a significant decrease of mass of the active form of the enzyme (21). Supernatants ConA-induced monocyte interstitial collagenase protein that from control monocytes contained low levels of 92-kDa type could be reversed by PGEz (1O"j M ) (lane 4 ) or Bt2cAMP IV collagenase/gelatinase that were unaffected by IL-4 treatM ) ( l a n e 5 ) . Addition of PGE, or BtzcAMP to unstimment. ulated monocytes did not induce pro- or active interstitial Restoration of Monocyte 92-kDa Type IV Collagenase/Gelcollagenase (datanot shown). Theability of PGE,or atinase Production in IL-4-treated Monocytes by PGEz and Bt2cAMP to restore interstitial collagenase production in ILBt2cAMP-PGE2 and Bt2cAMP were added to IL-4-treated 4-treated monocytes demonstratesthatIL-4inhibition of monocytes to determine if the inhibitionof monocyte 92-kDa PGEz interferes with the requiredPGE,-CAMP-dependent type IV collagenase/gelatinase by IL-4was, like that of interstep for interstitial collagenase production. stitial collagenase, also related to interruption of a PGEzEffect of IL-4 onthe Production of 92-kDa Type IV Collagenase/Gelatinaseby Human Monocytes-Monocytes have CAMP-dependent step. As shown in Fig. 6, A and B, PGEZor previously been shown to produce a 92-kDa type IV collagen- BhcAMP reversed the IL-4-mediated inhibition (lune 9 ) of ase (X),and the production of this metalloproteinase was ConA-induced92-type IV collagenase/gelatinase (lane 8). examined to determine if IL-4 regulated it ina fashion similar Restoration of both 92-kDa type IV collagenase, as deterto thatof interstitial collagenase. IL-4 caused a dose-depend- mined by zymogram (Fig. 6A) and protein(Fig. 6 B ) ,by PGEZ M (lane IO) withmaximal restoration at ent inhibition of the ConA-induced monocyte 92-kDatype IV was detected a t (lune 11) to M (lune 12). Additionally, 92-kDatype collagenase/gelatinase, as determinedby zymogram (Fig. 5A) and protein analysis (Fig. 5B). Some inhibition of ConA- IV collagenase/gelatinase was also restored in IL-4-treated induced type IV collagenase was detected in both assays a t monocytes by Bt2cAMP, asshown by zymogram and Western 1.5 units/ml of IL-4 ( l a n e a), whereas there was a significant analysis (lune 13).The relative enzyme activity, as determined decrease in activityas detected by zymogram and protein with by densitometry, for these conditions is given in the legend 15 (lane 9)and 150 (lane 10) units/ml of IL-4. As determined for Fig. 6. The low levels of 92-kDa type IV collagenase/ by densitometric analysisof the zymogram (see the values in gelatinase in unstimulated monocyte cultures were only de-

Inhibition of Human Monocyte

518

-

""

Collagenase Production by

IL-4

II

-97

-66

6

-45

-31

I

ql

FIG.6. Restoration of 92-kDa type IV collagenase in IL-4treated monocyte cultures by PGE2 andBt2cAMP. Media from monocyte cultures, as described in the legend to Fig. 5, were analyzed for type IV collagenase protein by zymogram ( A ) and Western analysis ( R ) . The experimental conditions examined in each lane are control (lane I ) , control + IL-4 (15 units/ml) (lane 2), control IL4 lo-* M PGE, (lane 3 ) , control IL-4 10" M PGE, (lane 4), M control IL-4 lo-" M PGE2 (lane 5),control IL-4 5 X Bt2cAMP (lane 6 ) ,control + IL-4 + 10" M Bt2cAMP (lane 7),ConA (20 pglml) (lane 8 ) , ConA IL-4 (15 units/ml) (lane 9), ConA IL4 + lo-' M PGE2 (lane I O ) , ConA + IL-4 10" M PGE2 (lane II), ConA IL-4 M PGE, (lane 12), ConA IL-4 5 X M Bt2cAMP (lane 13), and ConA + IL-4 10" M Bt2cAMP (lane 14). IL-4 was added 1 h prior to ConA, PGE2, and Bt2cAMP. The relative enzyme activity detected on the zymogram was measured by densitometry, and the arbitrary units for each lane are 1.1 (lane I ) , 1.6 (lane 2), 2.3 (lane 3 ) , 1.0 (lane 4), 1.2 (lane 5), 1.2 (lane 6 ) ,0.6 (lane 7),9.5 (lane 8 ) , 3.7 (lane 9), 5.0 (lane I O ) , 7.1 (lane II), 6.6 (lane 12), 4.8 (lane 13), and 3.6 (lane 14). Relative enzyme densitometric units for lanes 9-14 of the Western are 16.1 (lane 9), 0.9 (lane I O ) , 6.4 (lane 11). 13.0 (lane I2),5.2 (lane I3), and 7.3 (lane 14). A standard ( S ) for 92-kDa type IV collagenase/gelatinase from HT1080 fibrosarcoma cells is shown on the zymogram, and the location of the molecular weight standards are shown on the Western.

+

+

-45

+

+

+

+

+

+

+

+

,, ,

FIG. 7. Effect of IL-4 on tissue inhibitor of TIMP-1. Dialyzed and lyophilized media supernatants from 48-h monocyte cultures (equivalent to 3 X 10' monocytes) were analyzed for TIMP-1 by Western analysis. IL-4 (15 units/ml) was added 1 h prior to the addition of Con A (20 pg/ml).The experimental conditions examined in each lane are control (lane A ) , control IL-4 (lane R ) , control + IL-4 10" M PGE, (lane C), ConA (lane D),ConA + IL-4 (lane E ) , ConA + IL-4 + M PGEz (lane F),and ConA IL-4 + 5 X M BbcAMP (lane G ) .

+

+

+

genase/gelatinase in IL-4-treated monocytes. Moreover, inhibition of PGE, synthesis by indomethacin also blocked the production of 92-kDa type IV collagenase/gelatinase (data + not shown) as haspreviously been shown for monocyte interstitial collagenase (11).These results are the first demonstration that the signal transduction pathway for the monocyte 92-kDa type IV collagenase involves a PGE,-CAMP-dependent step similar to that for the monocyte/macrophage interstitial collagenase (10, 11). IL-4 was more effective in the suppression of PGE,-mediated metalloproteinase production if it was added 1 h prior to stimulation (Fig. 3), as compared to the simultaneous addition of ConA and IL-4 (Fig. 2). This suggested preincubation was required for IL-4 to exert its major effect on the tected by zymogram analysis in this experiment and were not suppression of prostaglandin synthesis. This conclusion is significantly altered by PGE,orBt2cAMP.Theseresults supported by our recent data that preincubation with IL-4 demonstrate that ConA-induced 92-kDa type IV monocyte caused a significant decrease in ConA-induced membranecollagenase is, like monocyte interstitial collagenase, also associated prostaglandin H synthetase.' regulated through a PGE,-CAMP step. This was further subAlthough PGE, or BbcAMP reversed the IL-4 inhibition stantiated by the ability of indomethacin, a cyclooxygenase of collagenase production in ConA-stimulated monocyte culinhibitor, to suppress 92-kDa typeIV collagenase/gelatinase tures, in the absence of ConA, these compounds failed to (data not shown). increase thebasal levels of 92-kDa type IV collagenase/ Effectof IL-4on the Production of TIMP-1 by Monocytesgelatinase either in the presence (Fig. 6) or absence of IL-4 In addition to the ability of IL-4 to directly suppress the (data not shown). This is in agreement with previous studies protein levels of interstitial and92-kDa type IV collagenases, on monocyte/macrophage interstitial collagenase that demIL-4 may also inhibit the activity of these enzymes by the onstrated the requirement of a primary stimulus,such as induction of TIMP. However, althoughstimulationwith ConA, in order for PGE'-mediated increases in intracellular ConA did increase the 28.5-kDa TIMP-1 protein(Fig. 7, lane levels of cAMP to be effective in initiating collagenase proD )as compared to control monocytes (lane A ) , IL-4 had no duction (28). The primary stimulusmay be required to induce effect on the expression of ConA-induced TIMP-1 (lane E ) changes in intracellular proteins, such as the cytoskeleton, nor did PGE, (lane F ) or Bt2cAMP (lane G). Similarly, the that arenecessary forthe transductionof the CAMP-mediated expression of TIMP-1 by unstimulated monocytes was unaf- signal (12). fected by IL-4 (lune B ) or PGE, (lune C). The low levels of latent 92-kDa type IV collagenase/gelatinase as determined by zymogram and Western analysis in DISCUSSION control monocytes were not affected by the addition of PGE, Our results demonstrate that IL-4 isa potent inhibitor of or Bt'cAMP. This suggests that the constitutive release of ConA-induced production of monocyte interstitial collagen- this enzyme in unstimulated monocytes is independent of ase, as well as the monocyte 92-kDa type IV collagenase/ PGE, regulation, whereas the inducible 92-kDa type IV colgelatinase. The mechanism of inhibition is related to the lagenase/gelatinase is PGE,-cAMP-dependent. The low conability of IL-4 to inhibit PGE, synthesis, thereby blocking the stitutive release of latent 92-kDa type IV collagenase/gelatinPGE,-mediated increase in cAMP required for production of ase may facilitate passage of unactivated monocytes through interstitial collagenase and 92-kDa type IV collagenase/gela- the basement membrane of blood vessels into tissues or intinase by activated monocytes. This was demonstrated by the flammatory sites, where they may subsequently be activated ability of PGE, and Bt,cAMP to restore the proteinlevels of to produce increased levels of latent andactive enzyme. both enzymes, as well as the enzymatic activity of interstitial collagenase and zymogram activity of 92-kDa type IV collaM. L. Corcoran and L.M. Wabl, manuscript in preparation.

+

+

+

+

+

Inhibition of Human MonocyteCollagenase Production by IL-4 The IL-4-mediated decrease in the expression of the ConAinduced interstitial collagenase and 92-kDa type IV collagenase/gelatinaseprotein was responsible for the decrease in interstitial enzymatic activity and in the latent and active 92kDa type IV collagenase/gelatinase, as shown by zymogram, rather than an enhancement of TIMP. This was shown by Western analysis of TIMP-1, which demonstrated that the ConA-induced increase in TIMP-1 was unaffected by IL-4. Thus, the inhibitory effect of IL-4 is primarily due to the suppression of PGE2, which interferes with the requisite PGE2-CAMP step inducing in metalloproteinase secretion and activation. This is further demonstrated by the ability of PGE2 or Bt2cAMP to restore the protein and activity levels for these collagenases. Western analysis showed that unstimulated monocytes produce only small amounts of interstitial procollagenase. The inability to detect significant amounts of interstitial procollagenase protein demonstrates that these cells were not activated by the isolation procedure. This is in agreement with the assessment of cell activation by monitoring IL-2receptor expression (23). Following overnight incubation at 37 "C in suspension,