Differential Regulation of Collagenase and Stromelysin mRNA in Late Passage Cultures of Human Fibroblasts

Differential Regulation of Collagenase and Stromelysin mRNA in Late Passage Cultures of Human Fibroblasts

EXPERIMENTAL CELL RESEARCH 222, 150–156 (1996) Article No. 0019 Differential Regulation of Collagenase and Stromelysin mRNA in Late Passage Culture...

188KB Sizes 1 Downloads 40 Views

EXPERIMENTAL CELL RESEARCH

222, 150–156 (1996)

Article No. 0019

Differential Regulation of Collagenase and Stromelysin mRNA in Late Passage Cultures of Human Fibroblasts GUANGYUAN ZENG1

ALBERT J. T. MILLIS2

AND

Department of Biological Sciences, Center for Cellular Differentiation, University at Albany-State University of New York, 1400 Washington Avenue, Albany, New York 12222

Nontransformed human fibroblast cell cultures have been extensively studied as an in vitro model for cellular senescence. Recently there has been considerable interest in using the human fibroblast in the identification of genes relevant to the process of replicative senescence. We demonstrated that in comparison with early passage cultures the expression of collagenase and stromelysin mRNAs and proteins was increased ú81 in late passage cultures of human fibroblasts and, in addition, expression of Il-1a, a cytokine that regulates collagenase and stromelysin expression, was also significantly increased in late passage cell cultures. These findings suggested the hypothesis that constitutive Il-1a expression in late passage cells may coordinately regulate the age-associated increase in the expression of collagenase and stromelysin. To test this hypothesis we examined the effects of long-term Il-1a treatment, serum starvation, and cycloheximide inhibition on collagenase and stromelysin mRNA levels in early and late passage human fibroblast cell cultures. Here we report that in late passage cell cultures, collagenase and stromelysin mRNAs respond differentially to Il-1a, serum starvation, and cycloheximide addition. Continuous exposure to Il-1a reduced the half-life of stromelysin mRNA but had little effect on the half-life of collagenase mRNA. In contrast to stromelysin, the collagenase mRNA level is dependent on serum factors. Collagenase is induced during recovery from cycloheximide inhibition, but stromelysin expression is not affected. These results establish that collagenase and stromelysin mRNAs are differentially regulated in both early and late passage human fibroblasts and suggest that the mechanisms responsible for the age-associated increase in the two mRNAs are different. In addition, these studies support the conclusion that continuous long-term exposure to Il-1a, a condition that is char1 Present address: Hypertension and Endocrine Branch, NIHLB, Building 10, 8C-118, National Institutes of Health, Bethesda, MD 20892. 2 To whom correspondence and reprint requests should be addressed. Fax: (518) 442-4767. E-mail: [email protected].

12-12-95 23:39:40

q 1996 Academic Press, Inc.

INTRODUCTION

The limited proliferative potential and finite in vitro lifespan of normal human cells in culture have provided an important model for the study aging at the cellular level. The validity of the cell culture model is supported by data indicating that fibroblasts from individuals with premature aging syndromes and fibroblasts from aged donors have reduced division capacity [1–2]. Evidence that replicative senescence results from genetic rather than random events is supported by somatic cell hybridization studies which have shown that the senescent phenotype is dominant in hybrids formed from young and senescent cells [1, 3]. Replicative senescence is also correlated with changes in expression of cell cycle genes [4–8]. Cultured human skin fibroblasts have been used in the investigation of the molecular basis of cellular aging, in general, and cutaneous aging, in particular [3, 6, 9– 12]. Fibroblasts are the most numerous cell type in the dermis, have a well characterized role in connective tissue extracellular matrix metabolism, and have been widely studied as a model of in vitro cellular aging. In vivo, dermal fibroblasts are quiescent cells that retain the capacity to respond to external signals for replication, gene activation, and wound repair [13]. In vitro, cultured fibroblasts are quiescent under conditions of serum insufficiency or contact inhibition, but can be induced to proliferate in the presence of added factors, including cytokines and growth factors. The capacity to proliferate in response to added mitogenic factors is lost as a consequence of serial subcultivation and may be a factor critical to the phenomenon of replicative senescence. The absence of a mitogenic response to exogenous stimuli occurs even though senescent fibroblasts are able to bind otherwise mitogenic growth factors and ini-

150

0014-4827/96 $12.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4678f3022

acteristic of late passage cells, is not the factor responsible for the high levels of collagenase expression, but may be critical for stromelysin expression.

ecas

AP: Exp Cell

COLLAGENASE AND STROMELYSIN EXPRESSION

tiate changes in gene expression. Although senescent fibroblasts do not proliferate they are metabolically active. Comparison of early and late passage cultures of fibroblasts from newborn donors, aged via in vitro serial subcultivation, revealed changes in the expression of a number of genes, including ones involved in maintaining extracellular matrix integrity [2, 14–16]. For example, senescent fibroblasts overexpress several matrix metalloproteinases (MMP; collagenase (MMP-1) and stromelysin (MMP-3)), underexpress a MMP inhibitory protein (tissue inhibitor of metalloproteinases-1; TIMP-1), and underexpress procollagen mRNAs. Some of those changes have also been observed in fibroblasts cultured from aged donors and in strains of Werner syndrome cells [2, 14–15, 17–24]. In cultured cells collagenase synthesis and stromelysin synthesis are regulated by extracellular factors including Il-1 [20–21, 23, 25–28]. For example, the basal level of collagenase expression by early passage cultures of nontransformed human fibroblasts is characteristically low but can be increased in response to exogenous inflammatory and growth- promoting cytokines, phorbol esters, and agents that alter the cell surface or ligate specific integrin receptors on the cell surface [29]. Recently it was demonstrated that the effect of phorbol ester on collagenase expression was secondary to phorbol-mediated induction of Il-1a [30]. Recent reports of an age-associated increase in the expression of Il-1a and b mRNAs and protein of several human fibroblast and endothelial cell strains [24, 31– 33] and the likely correlation between Il-a expression and increased metalloproteinase expression suggested that the high level of expression observed in late passage cells may result from an autocrine effect of Il-1 on metalloproteinase expression [31] and that the late passage levels of collagenase and stromelysin mRNAs are coordinately regulated by Il-1. In this paper we examined the effects of several agents, including Il-1a, that regulate collagenase and stromelysin expression to determine if the two genes are coordinately regulated in late passage cell cultures. We report that added Il-1a increased the steady-state level of collagenase and stromelysin expression in both early and late passage cells, but its effect on collagenase and stromelysin expression is different. Serum contains various factors such as EGF3 [34] which are able to up-regulate collagenase and stromelysin expression [35]. Therefore, we examined the effect of serum removal on collagenase and stromelysin mRNAs in late 3

Abbreviations used: CL, collagenase; SL, stromelysin; Il-1a, interleukin-1a; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; Act D, actinomycin D; PDGF, platelet-derived growth factor; EGF, epidermal growth factor; CHX, cycloheximide; SDS, sodium dodecyl sulfate; PCR, polymerase chain reaction.

/ m4678f3022

12-12-95 23:39:40

ecas

151

passage cells and report that collagenase and stromelysin mRNA levels are differentially affected by removal of serum. The high level of collagenase mRNA characteristic of late passage cells decreased rapidly, but the high level of stromelysin mRNA was unchanged. Examination of the effects of cycloheximide, an inhibitor of protein synthesis that affects Il-1a expression [31, 33], also supports the conclusion that the two metalloproteinases are differentially regulated. Considered collectively, these results establish for the first time that the collagenase and stromelysin genes are regulated differently and independently in late passage cell cultures. MATERIALS AND METHODS Materials. Medium 199 was obtained from Life Technologies, Inc. (Grand Island, NY); fetal bovine serum was from HyClone Laboratories, Inc. (Logan, UT); [32P]dCTP (3000 Ci/mmol) was from Amersham, Inc. (Arlington Heights, IL); the Klenow fragment of DNA polymerase was from U.S. Biochemicals (Cleveland, OH); Il-1I was from Genzyme (Cambridge, MA); actinomycin D was from Sigma Chemical Co. (St. Louis, MO); GeneSceen Plus was from Biotech System (Boston, MA); electrophoresis reagents were from Bio-Rad (Richmond, CA); and TRI reagent was from Molecular Research Center, Inc. (Cincinnati, OH). Cells and cell culture. Human fibroblast strain RIG was derived from newborn foreskin and cultured in medium 199 supplemented with 10% fetal bovine serum as previously described [21]. Cultures were routinely seeded at 6.67 1 103 cells/cm2 on tissue culture plastic. The number of cells recovered at each passage was used to calculate the population doubling level [23, 36]. Early passage cultures are defined as cultures that have completed less than 40% of their in vitro lifespan, and late passage cultures are those that have completed more than 80% of their in vitro lifespan [21, 23, 36]. To generate cultures representing different points in the in vitro lifespan fibroblast cultures were serially subcultivated at 7-day intervals and the total number of cells was computed at each subcultivation. The sum of all the cells obtained at all subcultivation points was used to calculate the percentage of lifespan completed for each experiment. Typically, fibroblasts were cultured for 4 days in medium 199 plus 10% fetal bovine serum before transfer to serum-free medium and addition of Il-1a (100 pg/ml), actinomycin D (5 Tg/ml), or cycloheximide (10 mg/ml). RNA extraction. TRI reagent was used to extract total RNA based on the protocol recommended by the manufacturer. Briefly, the cells in each 75-cm2 culture flask were homogenized in 2 ml of TRI reagent. After 5 min at room temperature, 0.4 ml of chloroform was added, mixed, and stored for 2–15 min at room temperature. After centrifugation, the aqueous RNA-containing phase was precipitated with isopropanol, washed with 75% ethanol, and dissolved in distilled water. Northern blot analysis. Total RNA was quantitated spectrophotometrically before being denatured in 10% glyoxal and 50% formamide by heating to 657C for 5 min. Typically, 8 mg total RNA from late passage fibroblasts and in some experiments 8–16 mg total RNA from early passage cells were loaded in each lane of a 1% agarose slab gel and separated by electrophoresis. After electrophoresis each gel was treated with 200 mM NaOH, vacuum-transferred to GeneSceen Plus membrane, and neutralized with 0.5 M Tris–HCl at pH 7.4. After ultraviolet cross-linking (Stratelinker, Stratagene, Inc., La Jolla, CA) the membranes were prehybridized in 8% Denhardt’s

AP: Exp Cell

152

ZENG AND MILLIS

solution (1% Ficoll and 1% polyvinypyrrolidone), 43% formamide, 0.69 M NaCl, 0.09% Na pyrophosphate, 0.9% SDS, 0.7% dextran sulfate, 1.74 mg/ml heparin, and 86 mg/ml denatured salmon sperm DNA for at least 3 hr at 427C. Each membrane was incubated in hybridization buffer overnight using 1 1 106 cpm/ml of radiolabeled probe before being washed in 11 SSC containing 0.2% SDS at ambient temperature for 30 min. Subsequently each was washed with 0.11 SSC containing 0.2% SDS for an additional 30 min at 507C. Autoradiography was performed at -707C. Quantitation of each radioactive signal was performed using a Betascope 603 (Betagen). Collagenase and stromelysin mRNAs were detected using a 1.7kb collagenase cDNA and a 1.7-kb stromelysin cDNA provided by Dr. Constance Brinckerhoff (Dartmouth Medical College, Hanover, NH) and subsequently cloned into the EcoRI site of pUC19. Plasmids were isolated by the alkaline lysis method [37]. The restriction fragments were recovered from the plasmids and random primer labeled with [32P]dCTP using the Klenow fragment of DNA polymerase. JActin was probed with a PCR product generated from human foreskin fibroblast RNA using primers purchased from Clontech, (Palo Alto, CA). GAPDH cDNA was generously provided by Jim Vinci (University at Albany, Albany, NY).

RESULTS

Collagenase and Stromelysin mRNAs Are Independently Regulated by Il-1a Previously we established that Il-1a, collagenase, and stromelysin protein levels were each increased in late passage cell cultures, suggesting that the increase in collagenase and stromelysin resulted from continuous exposure to Il-1a [24, 31]. To test this possibility late passage cultures were exposed to 100 pg/ml Il-1a for 57 hr. Figure 1A shows that in response to added Il-1a, both metalloproteinase mRNAs were increased to maximum levels after 23 hr of Il-1a treatment. At longer times of exposure, however, the steady-state level of collagenase mRNA decreased, and by 35–48 hr of exposure it was not detected. In contrast, stromelysin mRNA showed only a slight decrease within the 57hr time period studied. This result establishes that the two metalloproteinase genes are not coordinately regulated in late passage cells and that continuous exposure to added Il-1a has different consequences for each gene. To examine the possibility that the absence of serum contributed to the Il-1a-mediated decrease in collagenase mRNA, we examined the effects of Il-1a added to cultures in medium supplemented with 10% serum. In the presence of serum, the decrease in collagenase mRNA still occurred, but at a slower rate (Figs. 1B and 1C), indicating that serum factors influence the Il-1a effect on the steady-state level of collagenase mRNA. In contrast, the effect of Il-1a on stromelysin is independent of serum factors. To determine if the Il-1a-mediated decrease in the collagenase mRNA level resulted from a change in the mRNA half-life, the half-lives of collagenase and stromelysin mRNAs in late passage cells were calculated after actinomycin D treatment. In cultures not exposed

/ m4678f3022

12-12-95 23:39:40

ecas

FIG. 1. Effects of Il-1I on collagenase and stromelysin mRNA expression in late passage cell cultures. Cultures were placed in serum-free medium (A) or in medium containing 10% FBS (B) at 0 hr and simultaneously treated with 100 pg/ml of Il-1a. Total RNA was extracted at the times indicated and 8 Tg of total RNA was loaded in each lane. After transfer each membrane was probed sequentially with collagenase (CL), stromelysin (SL), or GAPDH cDNA. (C) The decay in collagenase and stromelysin mRNAs after 23 hr of exposure to Il-1a. The data are from A and B, normalized to the GAPDH signal. The 23-hr values are represented as 1.0. Symbols used: serum free medium (h, j); serum containing medium (s, l); collagenase (open symbols: h, s); and stromelysin (filled symbols: j, l).

to added Il-1a the mRNA half-lives were ú34 hr for each metalloproteinase (Figs. 2Aand2C). However, as shown in Figs. 2B and 2C Il-1a treatment had no effect on the half-life of collagenase mRNA, but the half-life of stromelysin mRNA was reduced to 24 hr. In the presence of actinomycin D the decrease in collagenase mRNA after Il-1a treatment (Fig. 2A) was not observed (Fig. 2B), indicating that mRNA synthesis is required for the decrease in collagenase mRNA but not for the decrease in stromelysin mRNA. The effects of Il-1a on collagenase and stromelysin mRNAs in early passage cells were similar to the effects on late passage cells ( Fig. 3A, data for stromelysin not shown). Further, as shown in Fig. 2 for late passage

AP: Exp Cell

COLLAGENASE AND STROMELYSIN EXPRESSION

153

and stromelysin are differentially regulated in late passage fibroblasts. Effects of Cycloheximide on Collagenase and Stromelysin mRNA

FIG. 2. Effect of Il-1a on collagenase and stromelysin mRNA half-lives in late passage fibroblasts. Fibroblast cultures in serumfree culture medium were exposed to 5 mg/ml actinomycin D in the presence or absence of Il-1a. (A) Equal amounts of total RNA extracted from cultures treated with actinomycin D for the indicated times and probed as indicated. (B) Equal amounts of total RNA extracted from cultures treated with Il-1a for 23 hr and subsequently with actinomycin D for the times indicated. (C) Comparison of collagenase and stromelysin mRNA half-lives, from the data in A and B, and normalized to the GAPDH or b-actin signal. The control level is set at 1.0 for each curve. Symbols used: collagenase (h) and stromelysin (j) from cultures not exposed to Il-1a; collagenase (s) and stromelysin (l) from cultures exposed to Il-1a for 23 hr before the addition of actinomycin D. After transfer each membrane was probed with collagenase (CL), stromelysin (SL), b-actin (ACTIN), or GAPDH cDNA.

We previously reported that cycloheximide stabilized Il-1 mRNAs [31]. To determine if the characteristically low steady-state mRNA levels of collagenase and stromelysin in early passage human fibroblasts were each affected by cycloheximide, fibroblast cultures were treated with 10 mg/ml cycloheximide for 4 hr and then incubated in medium without cycloheximide for 0–22 hr. As shown in Fig. 5, the steady-state level of collagenase mRNA increased within 3 hr of cycloheximide removal and remained at a level 81 the untreated control for at least 22 hr. In contrast, the steady-state level of stromelysin mRNA was not affected by the cycloheximide treatment, further indicating that the two metalloproteinases are differentially regulated in early passage cell cultures (Figs. 5A and 5C). This supports the finding of independent regulation of collagenase and stromelysin reported for early passage rheumatoid synovial cell cultures [25]. To determine if cycloheximide treatment also affects collagenase and stromelysin mRNA levels in late passage cells, those cultures were also exposed to cycloheximide following the same protocol. Figures 5B and 5C shows that the response of late passage cell cultures to

cells, Il-1a treatment resulted in a reduction in the stromelysin mRNA half-life to 11 hr without modification of collagenase mRNA half-life (Figs. 3B and 3C). These data further confirm that collagenase and stromelysin mRNAs are differentially regulated by Il-1a and that each mRNA is similarly regulated by Il-1a in early and late passage cells. Effects of Serum Removal on Collagenase and Stromelysin mRNA Level in Late Passage Cultures Fetal bovine serum contains factors that affect collagenase expression [30]. To determine if collagenase expression and stromelysin expression in late passage cells cultures were similarly dependent on serum we examined the effects of serum withdrawal on mRNA expression. Figure 4 shows that after 23 hr without serum the characteristically high steady-state level of collagenase mRNA decreased and was reduced to 50% of the control value within 40 hr of serum removal. In contrast, the stromelysin mRNA level remained high over the entire 57-hr serum-free period. It appears that stromelysin mRNA is not dependent on available serum factors and this provides evidence that collagenase

/ m4678f3022

12-12-95 23:39:40

ecas

FIG. 3. Effect of Il-1a on collagenase and stromelysin mRNAs and on collagenase and stromelysin mRNA half-lives in early passage fibroblasts. (A) Fibroblast cultures in serum-free culture medium were treated with 100 pg/ml Il-1a for the indicated times. (B) Equal amounts of total RNA extracted from cultures treated with Il-1a and actinomycin D for the indicated times. (C) Comparison of collagenase and stromelysin mRNA half-lives, from the data in the presence or absence of Il-1a and normalized to the b-actin signal. The control level is set at 1.0 for each curve. Symbols used: collagenase (h) and stromelysin (j) from cultures not exposed to Il-1a; collagenase (s) and stromelysin (l) from cultures exposed to Il-1a and actinomycin D. After transfer each membrane was probed with collagenase (CL), stromelysin (SL), b-actin (actin,) or GAPDH cDNA.

AP: Exp Cell

154

ZENG AND MILLIS

collagenase was higher in cells from old donors than in those from young donors [39]. Because Il-1a regulates collagenase and stromelysin synthesis [38] and late passage fibroblasts synthesize and secrete Il-1a, we postulated that continuous exposure to endogenous Il1a was critical for the increased expression of both collagenase and stromelysin in those cells [31]. Here, to evaluate the feasibility of that model, we exposed early and late passage fibroblast cultures to 100 pg/ml Il-1a continuously for 0–57 hr. We report that the steadystate levels of collagenase and stromelysin mRNA are differentially affected by long-term exposure to added Il-1a and, in contrast to studies of the collagenase response to short-term exposure to Il-1a, long-term (ú23 hr) exposure actually resulted in decreased collagenase mRNA expression. In contrast, Il-1a-mediated en-

FIG. 4. Effects of serum starvation on collagenase and stromelysin mRNA levels in late passage cell culture. Late passage cell cultures were exposed to serum-free culture medium at 0 hr and total RNA was isolated at the indicated times. Each lane contained 8 mg of total RNA. (B) A quantitation of the Northern assays shown in A. The level at 23 hr is set at 1.0 for each curve. The membrane was probed sequentially with collagenase cDNA (CL), stromelysin cDNA (SL), and (GAPDH). The open boxes indicate collagenase (h) and the filled boxes indicate stromelysin (j).

cycloheximide treatment was similar to that of the early passage cell cultures. Within 25 hr after removal of cycloheximide there was a 21 increase in the steadystate collagenase mRNA level and no apparent increase in the stromelysin mRNA level. The effect of cycloheximide on collagenase expression requires RNA synthesis. Figure 5D shows comparisons of collagenase mRNA levels in cycloheximide-treated cultures exposed to the RNA inhibitor actinomycin D. The increase in the collagenase mRNA level was inhibited by actinomycin D in both the 3- and 22-hr cultures. These results further indicated that collagenase and stromelysin mRNAs are independently regulated in late passage cells. DISCUSSION

Previous studies demonstrated that the basal levels of expression of Il-1a, collagenase, and stromelysin were each increased as a consequence of in vitro cellular aging [24, 31] and that the promotor activity of

/ m4678f3022

12-12-95 23:39:40

ecas

FIG. 5. Effects of cycloheximide withdrawal on collagenase and stromelysin mRNAs. Early passage cell cultures (A) and late passage cell cultures (B) were either not treated (-CHX) or treated with 10 mg/ml cycloheximide (CHX) for 4 hr and then incubated in fresh medium without cycloheximide for the additional times indicated. Each lane contains 8 mg of total RNA and the Northern assays were probed sequentially with collagenase (CL), stromelysin (SL), or GAPDH. (C) Quantitation of the radioactive signals for collagenase and stromelysin, normalized to the GAPDH signal, as shown in A and B. The control level (not exposed to cycloheximide) is set at 1.0 for each curve. Symbols used: early passage collagenase (h), stromelysin (j); late passage collagenase (s), stromelysin (l). (D) The effect of 5 mg/ml actinomycin D on steady-state mRNA levels in the cycloheximide-treated cultures at 3 and 22 hr after withdrawal of cycloheximide.

AP: Exp Cell

COLLAGENASE AND STROMELYSIN EXPRESSION

hancement of stromelysin mRNA expression is maintained for as long as 72 hr. Further, the steady-state levels of collagenase and stromelysin mRNAs are also differentially affected by serum withdrawal and by cycloheximide inhibition and recovery; agents known to regulate metalloproteinase expression [35, 38, 40– 42]. These results suggest that long term exposure to Il-1 contributes to increased stromelysin expression, but that additional regulatory factors must also contribute to the high basal levels of collagenase mRNA that are characteristic of the late passage cell culture phenotype. Experiments to identify those factors are currently in progress. The effects of Il-1a on collagenase and stromelysin mRNA levels during the process of cellular aging appear to be complex. The results presented here demonstrate that after exposure of less than 23 hr both mRNA levels increased in response to Il-1a. However, after longer exposure (34 hr) the stromelysin mRNA level diminished by 75% and the collagenase mRNA level diminished by 20% in early passage cultures. Computation of T1/2 indicates that Il-1a treatment resulted in reduction of the stromelysin half-life from ú34 hr to 11 hr, while the half-life of collagenase mRNA remained at ú34 hr. Similarly, in late passage cultures long-term exposure to Il-1a resulted in a reduction in the stromelysin mRNA half-life to 24 hr, while the collagenase mRNA half-life was virtually unchanged. The steady state level of stromelysin mRNA remains high even though the mRNA half-life decreased, suggesting that Il-1a must simultaneously induce stromelysin mRNA transcription while reducing its stability. Similar effects of Il-1a were previously reported for mRNAs encoding the Il-1 type I receptor and osteocalcin [43, 44]. The biological role of increased Il-1a production and subsequent metalloproteinase expression in late passage fibroblasts is not completely understood. However, the presence of this cytokine and its effect on stromelysin production suggest that it may have a critical role in regulating the degradation and processing of extracellular matrix components that may account for ageassociated changes in cell growth and adhesion [16, 45]. Stromelysin is likely to be a potent force in matrix degradation during cellular aging since its reported activities include activation of collagenase and degradation of matrix glycoproteins and proteoglycans [46, 47]. This work was supported by NIH Grant AG009279. We thank Heather McCue for her assistance with many of these experiments; and Lisa Shackelton and Casey Moulson for critically reading the manuscript.

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

30. 31. 32. 33.

REFERENCES 34. 1. Martin, G. M., Sprague, C. A., and Epstein, C. J. (1971) Lab. Invest. 23, 86–92.

/ m4678f3022

12-12-95 23:39:40

ecas

155

Goldstein, S. (1990) Science 249, 1129–1133. Hayflick, L. (1965) Exp. Cell Res. 37, 614–636. Sager, R. (1989) Science 246, 1402–1412. Stein, G. H., Beeson, M., and Gordon, L. (1990) Science 249, 666–669. Seshadri, T., and Campisi, J. (1990) Science 247, 205–209. Dimer, G., and Campisi, J. 1994) Exp. Cell Res. 212, 132–140. Ritting, S. R., Brooks, K. M., Cristofalo, V. J., and Baserga, R. (1986) Proc. Natl. Acad. Sci. USA 83, 3316–3320. West, M. D. (1994) Arch.Dermatol. 130, 87–95. Cristofalo, V. J., Gerhard, G. S., and Pignolo, R. J. (1994) Surg. Clin. North Am. 74, 1–21. Riabowol, K., Schiff, J., and Gilman, M. Z. (1992) Proc. Natl. Acad. Sci. USA 89, 157–161. Riabowol, K. T. (1992) Biochem. Cell Biol. 70, 1064–1072. Agren, M. S., Taplin, C. J., Woessner, J. F., Eaglstein, W. H., and Mertz, P. M. (1992) J. Invest. Dermatol. 99, 709–714. Sottile, J., Hoyle, M., and Millis, A. J. T. (1987) J. Cell. Physiol. 131, 210. West, M. D., Pereira-Smith, O. M., and Smith, J. R. (1989) Exp. Cell Res. 184, 138–147. Millis, A. J. T., and Mann, D. (1989) Fibronectin and Aging. In Fibronectin in Disease (Carsons, S., Ed.), pp. 37–48. CRC Press, Boca Raton, FL. Brown, W. T. (1990) Annu. Rev.Gerontol. Geriatrics 10, 23–42. Bauer, E. A., Silverman, N., Busiek, D. F., Kronberger, A., and Deuel, T. F. (1986) Science 234, 1240–1243. Furth, J. J. (1981) J. Gerontol. 40, 8122–8124. Johnson-Wint, B., and Gross, J. J. (1984) Cell Biol. 98, 90–96. Sottile, J., Hoyle, M., and Millis, A. J. T. Coll. Rel. Res. 8, 361– 374. Sottile, J., Mann, D. M., Diemer, V., and Millis, A. J. T. (1989) J. Cell. Physiol. 138, 281–290. Millis, A. J. T., Hoyle, M., McCue, H. M., and Martini, H. (1992) Exp. Cell Res. 201, 373–379. Kumar, S., Vinci, J. M., Millis, A. J. T., and Baglioni, C. (1990) Exp. Gerontol. 28, 505–513. MacNaul, K. L., Chartrain, N., Lark, M., and Tocci, M. J. 1990) J. Biol. Chem. 265, 17238–17245. Vater, C. A., Nagase, H., and Harris, E. D., Jr. (1983) J. Biol. Chem. 258, 9374–9382. Postlethwaite, A. E., Lachman, L. B., Mainardi, C. L., and Kang, A. H. (1983) J. Exp. Med. 157, 801–806. Brinckerholff, C. E., Benoit, M. C., and Culp, W. J. Proc. Nat. Acad. Sci. USA 82, 1916–1920, 1995. Alexander, C. M., and Werb, Z. (1991) Extracellular Matrix Degradation in Cell Biology of Extracellular Matrix (Hay, E., Ed.), 2nd ed., pp. 225-304. Plenum, New York. Fini, M. E., Strissel, K. J., Girard, M. T., Mays, J. W., and Rinehart, W. B. (1994) J. Biol. Chem. 269, 11291–11298. Kumar, S., Millis, A. J. T., and Baglioni, C. (1992) Proc. Natl. Acad. Sci. USA 89, 4683–4687. Maier, J., Voulalas, P., Roeder, D., and Maciag, T. (1990) Science 249, 1570–1574. Garfinkel, S., Brown, S., Wessendolf, J. H. M., and Maciag, T. (1994) Proc. Natl. Acad. Sci. USA 91, 1559–1563. Perheentupa, J., Lakshmanan, J., Hoath, S. B., Beri, U., Kim, H., Mascaso, T., and Fisher, D. A. (1985) Am. J. Physiol. 248, E391-E396.

AP: Exp Cell

156

ZENG AND MILLIS

35. Kerr, L. D., Miller, D. B. and Matrisian, L. M. (1990) Cell 61, 267–278. 36. Zeng, G., and Millis, A. J. T. (1994) Exp. Cell Res. 213, 148– 155. 37. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 38. Frish, S. M., and Ruly, H. E. (1987) J. Biol. Chem. 262, 16300– 16304. 39. Burke, E. M., Horton, W. E., Pearson, J. D., Crow, M. T. and Martin, G. R. (1994) Exp. Gerontol. 29, 37–53. 40. Schonthal, A., Herrlich, P., Rahmsdorf, H. J., and Ponta, H.. (1988) Cell 54, 325–334.

41. Gack, S., Vallon, R., Schaper, J., Ruther, U., and Angel, P. (1994) J. Biol. Chem. 269, 10363–10369. 42. Chua, C. C., Geiman, D. E., Keller, G. H., and Ladda R. L. (1985) J. Biol. Chem. 260, 5213–5216. 43. Li, Y-P., and Stashenko, P. (1992) J. Immunol. 148, 788–794. 44. Ye, K., Koch, K. C., Clark, B. D. and Dinarello, C. A. (1992) Immunology 75, 427–434. 45. Girard, M. T., Matsubara, M., Kublin, C., Tessier, M. J., Cintron, C., and Fini, M. E. (1993) J. Cell Sci. 104, 1001–1011. 46. Wilhelm, S. M., Collier, I. E., Kronberger, A., Eisen, A. Z., Marmer, B. L., Grant, G. A., Bauer, E. A., and Goldberg, G. I. (1987) Proc. Natl. Acad. Sci. USA 84, 6725–6729. 47. Quintavalla, J. C., Berg, R. A., Beavis, A. J., Piccoli, S. P., Rediske, J. J., Kurkinen, M., Patrick, R. A., and Robertson, F. M. (1993) J. Cell. Physiol. 154, 113–121.

Recieved March 21, 1995 Revised version received August 28, 1995

/ m4678f3022

12-12-95 23:39:40

ecas

AP: Exp Cell