Activation of protein kinase C stimulates collagenase production by cultured cells of the cervix of the pregnant . . gUInea pIg Mohammad R. Rajabi, MD, PhD," Samuel Solomon, PhD: and A. Robin Poole, PhDb C

Montreal, Quebec, Canada OBJECTIVE: Dilatation of the uterine cervix at parturition is achieved by an estrogen-induced, collagenase-mediated degradation of type I collagen in the cervix. The objective was to test the hypothesis that collagenase production in the cervix of pregnant guinea pig in culture is mediated by activation of protein kinase C. STUDY DESIGN: The effects of 17J3-estradiol, prostaglandin F2 or activation of protein kinase C by phorbol-12-myristate-13-acetate, 1-stearoyl-2-arachidonyl-sn-glycerol and phospholipase C on collagenase production was studied with primary monolayer cultures of cervical cells from Hartley guinea pigs at 50 days' gestation. Results are analyzed for statistical significance with analysis of variance. RESULTS: Collagenase production is increased twofold to threefold by 17J3-estradiol, prostaglandin F2., or activation of protein kinase C. The observed stimulation of collagenase production by 17J3-estradiol and prostaglandin F,. was blocked by the protein kinase C inhibitor 1-(5-isoquinoline sulfonyl) 2-methylpiperazine dihydrochloride. CONCLUSION: Collagenase production in cultured cervical cells of pregnant guinea pig is stimulated by activation of protein kinase C. (AM J OBSTET GVNECOL 1992;167:194-200.) 0>

Key words: Cervical dilatation, collagenase, estrogens, prostaglandins, protein kinase C Interstitial collagenase is a Ca 2 + and Zn 2 + -dependent llIetalloprotease, the action of which is an essential component of the overall metabolism of fibrillar collagen in the extracellular matrix. Collagenase cleaves collagen types I, II, and III, the predominant collagens of mammalian tissues. Its sole action is to cleave collagen by hydrolyzing the peptide bonds between Gly-Leu in the IX 2-chain and Gly-lle in the lXI-chain of type I collagen at a single locus in the native triple helical domain. I Once the collagen molecule is cleaved into two fragments (% length fragment TCA and 1/4 length fragment TC B ), the fragments denature and are further cleaved by gelatinase, PZ-peptidase, and dinitrophenyl-peptidase, followed by phagocytosis and further digestion , by lysosomal enzymes. I. 2 Collagen degradation is observed in biologic processes characterized by remodeling of connective tissue,

From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology," and the Departments of Surgery' and Biochemistry,' McGill University, and the joint Diseases Laboratory, Shriners Hospital for Crippled Children. Supported by the Medical Research Council of Canada, Shriners of North America, and Fonds de la Recherche en Sante du Quebec. Received for publication October 21, 1991; revised january 27, 1992; accepted February 13, 1992 Reprint requests: M. Rajabi, MD, PhD, Royal Vic/aria Hospital, Room F4.32, 687 Pine Ave. w., Montreal, Quebec, Canada H3A 1A1. 6/1137388

194

such as bone resorption, wound healing, embryogenesis, postpartum uterine involution, and ovulation and in diseases such as osteoarthritis, rheumatoid arthritis, corneal ulcers, epidermolysis bullosa, periodontal disease, and tumor invasion! Rapid collagen degradation is an underlying feature of cervical dilatation at parturition in both human and guinea pig."? We have been interested in the mechanism of cervical dilatation at parturition. We have previously shown that cervical dilatation is accompanied by a considerable increase in extractable collagenase in the uterine cervix in women and in the guinea pig at parturition. 3 ,? High levels of circulating collagenase are also present in women in premature labor. s Such studies have indicated that collagenase probably plays an important role in the rapid collagen degradation observed in normal parturition and in the pathophysiology of premature labor. We have chosen primary monolayer cultures of cells isolated from the cervices of pregnant guinea pigs at 50 days' gestation to study the mechanism of signal transduction of collagenase production for the following reasons: (I) there is an inadequate supply of cervical tissue from pregnant women, (2) cervical dilatation at parturition in both species is associated with a decrease in collagen (chemically and histologically) and a significant increase in extractable levels of collagenase in the cervix,H 7 and (3) the guinea pig's reproductive system resembles that of women in that it has a similar ovu-

Volume Wi Number I

lation cycle, ovulates spontaneously, and has an actively secreting corpus luteum, and most importantly continuation of pregnancy is not dependent on ovarian hormones after the first half of pregnancy.9 These characteristics make the guinea pig an appropriate model for studies of human reproduction. 9 Collagenase is synthesized and rapidly secreted by fibroblasts and chondrocytes. I Synthesis and secretion are stimulated by platelet-derived growth factor,'O epidermal growth factor,'1 interleukin-I,'2 tumor necrosis factor, 13 and phorbol esters.14 We have previously shown that collagenase messenger ribonucleic acid and collagenase activity are stimulated by human recombinant interleukin-I~ and 17~-estradiol in monolayer cell cultures of cervixes of pregnant guinea pigs at 50 days gestation. 6 Furthermore, collagenase biosynthesis in culture is mediated by prostaglandins (PG) F2a and E2." The mechanism of signal transduction of collagenase production in the guinea pig cervix is not known. This study examines the role of protein kinase C on collagenase production in the cervix of the pregnant guinea pig. Hydrolysis of endogenous phosphatidylinositol bisphosphate by phospholipase C results in the release of inositol triphosphate and diacylglycerol. 15 The release of inositol triphosphate and diacylglycerol results in increasing intracellular calcium concentration and activation of protein kinase C, respectively.16 The present study extends our previous findings by providing evidence that activation of protein kinase C results in stimulation of collagenase production in cervical cell cultures of guinea pigs at 50 days' gestation. Material and methods

Material. The following reagents were obtained from the Sigma Chemical Company, St. Louis: tamoxifen, PGF"", 17~-estradiol, phorbol myristate acetate, I-Stearoyl-2-arachidonyl-sn-glycerol, phospholipase C, bacterial collagenase type AI, deoxyribonuclease type I, and bis-benzimide (Hoechst H33258). Trypsin 1: 250 was obtained from Difco Laboratories, Detroit. 1(5-Isoquinoline-sulfonyl)-2-methylpiperazine dihydrochloride (H-7) was obtained from Seikagaku America, St. Petersburg, Fla. Aquasol scintillation fluid and tritiated-methyl-thymidine were obtained from New England Nuclear, Dupont, Markham, Ontario. Animals. Adult Hartley guinea pigs (Charles River, Montreal) were housed in guinea pig cages (three to five per cage) with a 12-hours-light and 12-hours-dark cycle and fed guinea pig chow and water as desired. Timed gestation was induced by mating animals within 12 hours of delivery. Pregnant animals at 50 days of gestation (n = 10) were anesthetized with intramuscular ketamine hydrochloride (50 mg/kg) and xylazine (3 mg/kg). After approximately 15 minutes they were killed by intracardiac exsanguination. With aseptic

Protein kinase C stimulates collagenase

195

technique a midline laparotomy and a total abdominal hysterectomy were performed according to a protocol approved by the Royal Victoria Hospital Research Animal Care Sub-Committee. Cervical cell isolation and culture. Cervices obtained from 10 pregnant guinea pigs at 50 days' gestation were cut into 2 to 4 mm pieces and digested with 0.05% trypsin I: 250/0.02% ethylenediaminetetraacetic acid (EDTA) in calcium and magnesium-free phosphate-buffered saline solution (for 15 minutes on ice, followed by digestion with bacterial collagenase type AI, 0.5 mg/ml) and deoxyribonucleas type I (0.05 mg/ml) in Dulbecco's modified Eagle medium for 4 to 6 hours at 37° C with gentle agitation. The digested mixed cervical cells were recovered by centrifugation at 365g for 10 minutes at room temperature every I to 2 hours followed by washing once with Hanks' balanced salt solution. The process was repeated until the tissue was completely digested (usually within 6 hours). The released cells were filtered twice through nylon monofilament with a pore size of 400 ILmollL (Swiss Nitex, Thompson, Montreal) and recovered by centrifugation at 365g for 10 minutes at room temperature. The cells were washed three times with Hanks' balanced salt solution followed by suspension in Dulbecco's modified Eagle medium containing heat-inactivated (60° C for 30 minutes) 10% fetal calf serum, penicillin G sodium (100 U/ml), streptomycin sulfate (100 ILg/ml), and amphotericin B (1.25 ILg/ml). Cell viability exceeded 90% as determined by the trypan blue exclusion test (approximately 3 X 106 cells per cervix). For each experiment three to four pregnant guinea pigs at 50 days of gestation were used. Cells were counted with a hemocytometer slide. Cells were plated in Limbro plates containing 96 flat-bottom wells, (Flow Laboratories, McLean, Va.) at I x 105 cells per well in 300 ILl medium and incubated in a humidified atmosphere of 5% carbon dioxide in air at 37° C. After 3 days almost all cells had adhered. The medium and unattached cells were removed and replaced with fresh medium. Cells were then grown to confluency (usually within 3 days). Immediately before analysis of collagenase production, the cells were washed once in Hanks' balanced salt solution, followed by the addition of different concentrations of the chemicals and hormones to be tested in serum-free Dulbecco's modified Eagle medium containing penicillin, streptomycin, and amphotericin B at the concentrations described above with appropriate controls containing ethanol, which was used as the solvent at 0.1 % to 0.3% final concentration of ethanol in Dulbecco's modified Eagle medium. After a further 72 hours culture media were collected and frozen at - 20° C until assayed for collagenase. The cells were collected after a brief exposure to 100 ILl per well of 0.05% trypsin and 0.02% EDTA in calcium and mag-

196

Rajabi. Solomon. and Poole

July 1992

,\ III

J Obstet Gynecol

Table I. Tritiated methyl thymidine incorporation Incorporation Counts/min (mean ± SEM)

Control Dulbecco's minimum essential medium Phorbol-12-myristate-13-acetate (10- 8 mollL) I-Stearoyl-2-arachidonyl-sn-glycerol (10- 7 moll L) Phospholipase C (0, I U / ml) 17~-Estradiol (10- 7 mollL) PGF'a (10- 6 mollL)

62,753 57,025 55,025 43.849 66,683 51,080

± 3,393

± ± ± ± ±

1,204 7,552 8,425 5,859 8,862

% 5.76 5.24 5,11 4.03 6.12 4.69

Tritiated methyl thymidine added to each well contained 1.1 x 10" counts/min. Each value gives the mean value ± SEM for quadruplicate wells. There were no significant differences.

nesium-free phosphate buffered saline solution (5 to 10 minutes, 37° C). Representative wells were counted with a hemocytometer slide. Cells from the remaining wells were frozen at - 20° C for deoxyribonucleic acid (DNA) determinations. Collagenase assay. Acid-soluble collagen was purified from the skins of adult female Sprague-Dawley rats, and the collagenase substrate tritiated telopeptidefree collagen type I was prepared as previously described. 6 Collagenase was assayed with the specific collagenase method described by Dean and Woessner!7 with some modifications. Briefly, enzyme samples (up to 100 /LI in assay buffer) were incubated with 10 /Llof tritiated telopeptide-free collagen substrate (specific activity 3.1 x 106 counts/min/mg collagen at 2.24 mg/ml, in 50 mmollL Tris hydrochloride, pH 7.6, 0.3 moUL sodium chloride) in 1.5 ml microfuge tubes (39 x 10 mm, Sarstedt, Germany). Aminophenylmercuric acetate (0.5 mmollL) was added to activate procollagenase. 1,1O-Phenanthroline (l mmollL) was added to inhibit collagenase and served as a blank control. After 18 to 48 hours of incubation at 30° C, the reaction was terminated by the addition of EDTA, to a final concentration of 40 mmollL. Collagenase cleavage products (TC A , TC B ) were further digested by trypsin and chymotrypsin in the presence of type I collagen as a carrier protein for 2 hours at 30° C. The undigested tritiated telopeptide-free substrate was precipitated in 10% trichloroacetic acid at 0° C for 30 minutes in the presence of 250 /Lg bovine serum albumin (10 mg/ml assay buffer) followed by centrifugation in a fixed speed Eppendorf centrifuge 5413 (Brinkman, Westbury, N.Y.) for 15 minutes. A 100 /LI aliquot of the supernatant was mixed with 5 ml Aquasol scintillation fluid and counted in a Beckman scintillation spectrometer (Beckman, Palo Alto, Calif.). Mixing of these volumes resulted in minimal quenching of radioactivity (data not shown). The percent digestion was calculated as the total counts per minute in the supernatant minus counts per minute in the 1,10-phenanthroline blanks divided by original counts in the tritiated telopeptidefree substrate (in 100 /Ll of 10% trichloroacetic acid in

assay buffer) x 100. One unit of collagenase is defined as the amount of collagenase that digests 1 /Lg of tritiated telopeptide-free collagen type I in 1 minute at 30° C. Inactivation of tissue inhibitor of metalloproteinase. To prevent the inactivation of activated collagenase by tissue inhibitor of metalloproteinase, the latter was inactivated by reduction of its six disulfide bridges in 2 mmol/L dithiothrietol at 37° C for 30 minutes. To prevent spontaneous reactivation of tissue inhibitor of metalloproteinase, the reduced form was alkylated with 5 mmoUL iodoacetamide at 37° C for 30 minutes. The samples were dialyzed against assay buffer for 18 hours at 4° C, then assayed for collagenase. This treatment increases the detection of collagenase and does not interfere with collagenase activity.6.7 DNA determination. DNA was determined as previously described. 6 Briefly, cervical tissues were digested with proteinase K (0.5 mg/ml) for 24 hours at 56° C in 0.1 moll L sodium phosphate buffer, pH 6.5, containing 0.01 % EDTA. DNA standard curves were prepared from calf thymus DNA (0.1 to 1.0 /Lg) in 0.1 moUL phosphate and EDTA buffer. To 100 /Ll of tissue digest was added 2 ml of a solution containing 1.0 /Lg/ml hisbenzimide in 0.05 mollL sodium phosphate, pH 7.4, containing 2 moUL sodium chloride. Fluorescence was recorded with an excitation at 356 nm and emission at 458 nm with a fluorescence spectrophotometer (model 650-IOS, Perkin-Elmer Corporation, Norwalk, Conn.). Tritiated thymidine incorporation. Primary monolayer cell cultures of cervices from guinea pigs at 50 day's gestation were maintained in 96-well plates in Dulbecco's modified Eagle medium with 10% fetal calf serum until confluency, as described above. The cells were washed twice in Hanks' balanced salt solution followed by the addition of 75 /LI of serum-free Dulbecco's modified Eagle medium (in the presence or absence of hormones and chemicals to be tested). Twenty-five microliters of serum-free Dulbecco's modified Eagle medium containing 1 /LCi of tritiated methyl thymidine (6.7 Ci/mmol) were added during the last 18 hours in

Protein kinase C stimulates collagenase

Volume 167 Number I

culture. Cells were incubated in an atmosphere of 5% carbon dioxide in air at 37° C. Culture media were removed, and cells were suspended by brief exposure to 100,...1 per well of 0.05% trypsin and 0.02% EDTA in calcium-and magnesium-free phosphate-buffered salt solution for 5 minutes at 37° C. Cells were harvested on filter mats with a cell harvester (Titertek Cell Harvester, Skraton Inc., Sterling, Va.). Filters were added to 10 ml Aquasol scintillation fluid and counted in a Beckman scintillation spectrometer. Statistical analyses. Results are expressed as the mean ± I SEM unless otherwise indicated. Statistical analyses were performed by analysis of variance for multiple observations with RSI 1 Statistical Tools (BBN Software Products Corporation, Cambridge, Mass.). Statistical differences were determined with Bonferroni simultaneous confidence intervals for all comparisons, when applicable. Differences between groups were considered significant when p:S 0.05. Dose-response curve fitting was obtained with DeltaGraph version 1.5 with polynomial of degree setting at 3 (DeltaPoint, Inc., Monterey, Calif.).

a W

Cf)

et Z Wet

Effects of phorbol myristate acetate, diacylglycerol, phospholipase C, 1713-estradiol, and PGF.« on DNA replications. Confluent cells were exposed to these chemicals and hormones in serum-free Dulbecco's modified Eagle medium in the presence of tritiated methyl thymidine. There was no significant effect on tritiated thymidine incorporation into DNA when compared with control under these conditions (Table I). Stimulation of procollagenase by known activators of protein kinase C. The addition of the tumor promoter phorbol-12-myristate-13-acetate (Fig. I, A) or the diacylglycerol I-stearoyl-2-arachidonyl-sn-glycerol (Fig. 1, B) resulted in a twofold stimulation of procollagenase production. Each of these activators, at all concentrations tested, resulted in an increase in pro collagenase production. Dose-response effects were observed with a maximum response at 10- 8 mollL for phorbol-12-myristate-13-acetate and 10 -7 moll L for 1stearoyl-2-arachidonyl-sn-glycerol. Similarly, the addition of bacterial phospholipase C (0.01 to 10 U Iml Dulbecco's modified Eagle medium) also resulted in a threefold stimulation of procollagenase production with a maximum response at 0.1 U/ml (Fig. 2). Evidence for role of protein kinase C in mediating procollagenase stimulation of prostaglandins and estrogen. PGF 2• (10- 6 mollL) (Fig. 3, A) and 17[3-estradiol (10- 7 mollL) (Fig. 3, B) stimulate the production of procollagenase in culture. The observed stimulation by both hormones was completely blocked by 1-(5-isoquinoline sulfonyl)-2-methylpiperazine dihydrochloride, H-7, an inhibitor of protein kinase C.'s The stimulatory

Phorbol-12-Myristate-13-Acetate

*

f

60

(!)z

eto 40

..J..J::J

OE

0

0

a: Q.

b W

Cf)

20

0 80

I

i

1-Stearoyl-2-Arachidonyl-sn-Glycerol

*

*

987

6

60

et Z Wet

(!)z

eto

..J..J::J

40

OE 0

0

Results

80

197

a: Q.

20

0+--"'0-11

~O

i

i

i

5

i

4

-log [MJ Fig. 1. Effects of activation of protein kinase C by phorbol12-myristate-13-acetate (a) and by diacylglycerol l-stearoyl-2arachidonyl-sn-glycerol (b) on procollagenase production from primary monolayer cervical cell cultures of four pregnant guinea pigs at 50 days' gestation. Culture medium from each well was reduced with dithiothreitol, alkylated with iodoacetamide, dialyzed, and assayed for collagenase in duplicate in the presence of aminophenylmercuric acetate (0.5 mmol/L, added to activate procollagenase). Samples assayed for collagenase in presence of 1,10-phenanthroline (l mmol/L, added to inhibit collagenase) served as blank controls. Mean ± SEM. Each point determination represents mean of quadruplicate wells assayed for collagenase. - Log[Mj, Negative logarithm of molar concentration; asterisk, p < 0.05 versus control (analysis of variance).

effect of 17[3-estradiol but not that of PGF,. was completely abolished by the estrogen antagonist tamoxifen. Comment Interstitial collagenase is believed to play an important role in the cleavage of collagen during cervical dilatation at parturition in the human. 3 We have shown that the pregnant guinea pig can be used to study the regulation of this process!' In this study we have extended our previous findings to examine the mechanism of signal transduction of collagenase production in cultured cervical cells of the guinea pig at 50 days'

198

Rajabi, Solomon, and Poole

July 1992 Am J Obstet Gynecol

100

w

en

et Zet

*

80

7

*

w

c:(M ZO

(J)_

wz 60

w"-

CJo

eten

5:;

..1:::1.

uE

40

0

a:

0.

20

~

0 Control

(!)~ «-1Q)

6

5 4

-1~ 3

i i i

0.01

0.1

1

i

10

OE Ofr 0 _ 2 a:: 1 a..

0

PHOSPHOLIPASE C Ulml

Fig. 2. Effects of hydrolysis of endogenous phosphatidylinositol bisphosphate by bacterial phospholipase C on procollagenase production from primary monolayer cervical cell cultures of three pregnant guinea pigs at 50 days' gestation. Culture medium from each well was reduced with dithiothreitol, alkylated with iodoacetamide, dialyzed, and assayed for collagenase in duplicate in the presence of amino phenylmercuric acetate (0.5 mmollL, added to activate procollagenase). Samples assayed for collagenase in presence of I,IO-phenanthroline (l mmollL, added to inhbit collagenase) served as blank controls. Mean ± SEM. Each point determination represents mean of quadruplicate wells. Asterisk, p < 0.05 versus control (analysis of variance). gestation. Collagenase is not stored in fibroblasts, and biosynthesis is accompanied by prompt secretion of the enzyme into the extracellular matrix.' Therefore secretion of the enzyme is a measure of collagenase biosynthesis. 6 Ordinarily, activated procollagenase is inactivated by available tissue inhibitor of metalloproteinase as soon as the enzyme is activated. With a recently developed sensitive and specific assay to detect latent procollagenase and a method to inactivate tissue inhibitor of metalloproteinase, it was possible to measure the total latent enzyme produced by these cells. It was also important to determine the mitogenic effect of the chemicals and hormones used in this study on cervical cells. As shown in Table I, there was no significant effect on cell division as determined by tritiated thymidine incorporation of any of the reagents tested compared with the controls when using confluent cultures and serum-free medium. The dissociation between mitogenic activity and protein synthesis under these conditions allowed us to investigate the effect of these reagents on procollagenase biosynthesis in culture. The role of protein kinase C activation in procollagenase biosynthesis was studied. Hydrolysis of endogenous phosphatidylinositol bisphosphate by bacterial phospholipase C 15 and activation of protein kinase C by phorbol-12-myristate-13-acetate and diacylglycerol resulted in twofold to threefold stimulation of procollagenase production. These results indicate that production of procollagenase may be mediated by activation of endogenous phospholipase C resulting in

~ 0



7

w

«M ZO

(J)_

w"-

(!)~ «-1Q)

6

b

tI-

~~

~~

0..-

~~

(!)~

*

~O

tI-

(\IC

u.~

(!)

'x

a..~

--

~~ ~O

tI-

~'i-

~~

S +

5 4

-1~ 3 OE

00. 2

O~

a::

a..

1

0

g

8

~

e. ~

ul"

~~

e. c

~O

ul"i

3+

~~ ~

0

e.~

ul"±+

Fig. 3. Effects of PGF20 (a) and 17!3-estradiol (E2 ) (b) as modified by estrogen antagonist tamoxifen and protein kinase C inhibitor 1-(5-isoquinoline sulfonyl)-2-methylpiperazine dihydrochloride (H-7) on procollagenase production in primary monolayer cervical cell cultures of three pregnant guinea pigs at 50 days' gestation. Each point determination represents mean of quadruplicate wells assayed for collagenase in duplicate after reduction with dithiothreitol, alkylation with iodoacetamide, and dialysis in presence of 0.5 mmoll L aminophenylmercuric acetate. Asterisk, p < 0.05 versus control; asterisks, p < 0.05 versus PGF2a (a) and 17f3-estradiol (b) (analysis of variance).

hydrolysis of phosphatidylinositol biphosphate, production of diacylglycerol, and activation of protein kinase C. We have previously shown that 1713-estradiol stimulates procollagenase production in cultured cervical cells of guinea pigs at 50 days' gestation through a prostaglandin intermediate with an optimum response at 10- 7 mollL. 6 PGF 2• also stimulates procoIlagenase production with an optimum response at 10- 6 moJlL.6 In this study the stimulatory effect of 1713estradiol and not that of PGF2• was completely blocked by the estrogen antagonist tamoxifen, indicating a specific effect of estrogen. The observed stimulatory ef-

Protein kinase C stimulates collagenase

Volume 16 7 Number I

199

Extracellular Cell membrane

~

~

~.

C

B

~ca~I~)

>

Activation of protein kinase C stimulates collagenase production by cultured cells of the cervix of the pregnant guinea pig.

Dilatation of the uterine cervix at parturition is achieved by an estrogen-induced, collagenase-mediated degradation of type I collagen in the cervix...
2MB Sizes 0 Downloads 0 Views