TGF-?1 induces the expression of type I collagen and SPARC, and enhances contraction of collagen gels, by fibroblasts from young and aged donors

JOURNAL OF CELLULAR PHYSIOLOGY 158:169-179 (1994)

TGF-Pl Induces the Expression of Type I Collagen and SPARC, and Enhances

Contraction of Collagen Gels, by Fibroblasts From Young and Aged Donors

M A Y J* REED, ROBERT B. VERNON, ITAMAR B. ABRASS, AND E. HELENE SAGE* Departments of Medicine (M.). R., 1. B.A.) and Biological Structure (R.B.V., E. H.S.),

University of Washington, Seattle, Washington 98 7 95

Fibroblasts have a major role in the synthesis and reorganization of extracellular matrix that occur during wound repair. An impaired biosynthetic or functional response of these cells to stimulation by growth factors might contribute to the delayed wound healing noted in aging. We, therefore, compared the responses of dermal fibroblasts from young and elderly individuals (26, 29, 65, 89, 90, and 92 years of age) to transforming growth factor-pl (TCF-P1) with respect to: ( 1 ) the synthesis of type I collagen and SPARC (two extracellular matrix proteins that are highly expressed by dermal fibroblasts during the remodeling phase of wound repair) and (2) the contraction of collagen gels, an in vitro assay of wound contraction. With the exception of one young donor, all cultures exposed for 44 hours to 10 ng/ml TCF-P1 exhibited a 1.6- to 5.5-fold increase in the levelsof secreted type I collagen and SPARC, relative to untreated cultures, and exhibited a 2.0- to 6.2-fold increase in the amounts of the corresponding mRNAs. Moreover, the dose-response to TGF-Pl (0.1-10 ng/ml), as determined by synthesis of type I collagen and SPARC mRNA, was as vigorous in cells from aged donors as in cells from a young donor. In assays of collagen gel contraction, fibroblasts from all donors were stimulated to a similar degree by 10 ng/ml TCF-P1. In conclusion, cells from both young and aged donors exhibited similar biosynthetic and contractile properties with exposure to TGF-P1. It therefore appears that the impaired wound healing noted in the aged does not result from a failure of their dermal fibroblasts to respond to this cyiokine.

It is generally accepted that aged cells retain their ability to respond to stimulation by growth factors but are less responsive than young cells. This belief is based primarily on experiments that measured proliferative responses of aged fibroblasts to a variety of cytokines that include insulin-like growth factors, epidermal growth factor, and platelet-derived growth factor (Plisko and Gilchrest, 1983; Phillips et al., 1984; Stanu- lis-Praeger et al., 1986). Diminished proliferative responses to these growth factors were exhibited by fibroblasts that were (1) aged in culture, (2) derived from aged normal donors, or (3) obtained from individuals with diseases of premature aging such as the progeroid syndromes (Harley et al., 1981; Tsuji et al., 1984). Moreover, recent work has demonstrated that aged smooth muscle cells are impaired in their antiproliferative response to the cytokine transforming growth factor-p (McCaffrey and Falcone, 1993). Little is known about how processes of aging might influence the effect of growth factors on other forms of cellular behavior, e.g., biosynthetic and contractile processes that are rel- evant to wound repair.

Transforming growth factor-pl (TGF-P1) has been studied extensively as a cytokine that regulates extracellular matrix production and cell-matrix interac- Q 1994 WILEY-LISS. INC

0 1994 Wilcy-Liss, Inc.

tions. In vitro, TGF-Bl appears to be multifunctional. For example, it inhibits the growth of several types of normal and malignant epithelial cells and it stimulates the proliferation of many mesenchymal cells (Moses et al., 1990; Sporn and Roberts, 1992). TGF-p1 stimulates fibroblasts to synthesize increased quantities of type I collagen and to contract collagen gels to a greater degree than untreated controls (Montesano and Orci, 1988; Penttinen et al., 1988). Synthesis of SPARC (secreted protein, acidic and rich in cysteine), which often coincides with increased production of type I collagen (Salonen et al., 1990; Vuorio et al., 1991; Reed et al., 1993a), was also increased by TGF-P1 in human gingi- val fibroblasts (Wrana et al., 1991). In vivo, TGF-pl is important in the regulation of connective tissue synthesis after wounding (Kane et al., 1991). Moreover, TGF-pl reverses the impairment in wound healing in aged rats (Reed et al., 1993b). Excessive levels of this cytokine might mediate scar formation and postradia- tion fibrosis (Canney and Dean, 1990; Border and Ruo-

Received June 7,1993; accepted August 26,1993. *To whom reprint requestslcorrespondence should be addressed.

170 REED ET AL.

slahti, 1992). Antibodies that block the activity of TGF-p1 have been shown to inhibit scarring in an animal model of wound repair (Shah et al., 1992).

Replacement of damaged extracellular matrix and wound contraction are critical elements of wound repair. Type I collagen accounts for >90% of the collagen found in skin and is requisite for maintenance of the structural integrity of the dermis. Studies in vivo and in vitro indicate that reduced levels of this protein might contribute to the atrophy of the skin and impaired healing that are associated with aging (Uitto et al., 1969; Shuster et al., 1975). Moreover, it has long been appreciated that wound contraction is delayed with advancing age (Orentreich and Selmanowitz, 1969). In support of this observation, fibroblasts that were aged in vitro or derived from aged donors were less able to contract collagen gels than early passage fibroblasts derived from very young donors (Kono et al., 1989).

SPARC is an extracellular glycoprotein that is secreted by numerous cell types that include platelets, endothelial cells, osteoblasts, and fibroblasts (Sage and Bornstein, 1991). Also known as osteonectin (Termine et al., 19Sl), it is the major noncollagenous protein in bone and exhibits affinities for Ca2+, platelet-derived growth factor, and components of the extracellular matrix that include type I collagen (Sage et al., 1989a; Raines et al., 1992). SPARC inhibits the spreading of cells in vitro and is important in processes that influence the proliferation, shape, and motility of cultured cells (Sage et al., 1989b; Funk and Sage, 1991,1993). In vivo, SPARC is expressed during normal development and at wound sites, where it might enhance the reorganization of connective tissue and stimulate angiogenesis (Sage et al., 1989~; Reed et al., 1993a; Sage, 1993). Moreover, in an animal model, the temporal and spatial expression of SPARC and type I collagen and their corresponding mRNA were concordant during all phases of dermal wound repair (Reed et al., 1993a).

As an approach to an understanding of the compromised wound healing noted in aged individuals, we asked whether the biosynthetic and contractile behav- iors of dermal fibroblasts obtained from aged donors were altered by exposure to TGF-p1 in vitro. We measured the ability of fibroblasts from aged donors to contract extracellular matrix and to synthesize type I collagen and SPARC in the presence of TGF-P1, and we found that the cells were not impaired in their response to this cytokine. This result indicates that the impaired healing of wounds in aging is not caused by refractori- ness of fibroblasts to TGF-P1.

MATERIALS AND METHODS Antibodies

Antibodies against type I collagen that was purified from lathyritic rat skin were produced in a guinea pig. Titer and specificity were determined by enzyme- linked immunoadsorbent assay (ELISA) and Western blot analyses of the purified antigen and of conditioned media from cultured human fibroblasts (L. Iruela- Arispe, M.J. Reed, and E.H. Sage, unpublished experiments). Antibodies against SPARC (osteonectin) that was purified from human platelets (Haematologic Technologies, Essex Junction, VT) were produced in a

guinea pig. Titer and specificity were determined by ELISA and Western blot analyses (T.F. Lane, M.J. Reed, and E.H. Sage, unpublished experiments).

Cell culture Cultures of human dermal fibroblasts isolated from

the mesial upper arms of male donors aged 26 (AG092661, 29 (AG04441A), 65 (AG040611, 89 (AG063081, 90 (AG06291A1, and 92 (AGO40641 years were obtained from the National Institute on Aging/ Aging Cell Repository of the Baltimore Longitudinal Study at the Coriell Institute for Medical Research. All experiments employed cells with passage numbers be- low T13. Fibroblasts were grown in a 1:l mixture of Dulbecco’s Modified Eagle’s Medium and Ham’s F12 medium (DMEM/F12) (Gibco, Grand Island, NY) with 7.5% heat-inactivated fetal calf serum (FCS) (HyClone, Logan, UT), 250 pg/ml amphotericin B (Sigma Chemi- cal Co., St. Louis, MO), 100 units/ml penicillin G, and 100 unitdm1 streptomycin sulfate. Cells were grown in low serum (DMEM/O.S% FCS) for 72 hours and were subsequently exposed to DMEM/5% FCS, with or without human recombinant TGF-Pl (Gibco), a t doses that ranged from 0.1-10 ngiml.

Analysis of DNA synthesis Fibroblasts from each donor were plated, in sextupli-

cate, into 24-well plates (Corning Glass Works, Corn- ing, NY), grown to confluence in DMEM/5% FCS, and placed in DMEM/0.5% FCS for 0,24,48,72, or 96 hours. At the end of each time period, cells were evaluated by light microscopy for evidence of detachment, and three wells were supplemented with 1 pCi/ml [meth~l-~HJ- thymidine (New England Nuclear, Boston, MA). After 4 hours, the labeled cells were washed twice with serum-free DMEM, fixed with ice-cold 10% trichloroacetic acid for 25 minutes, washed with cold ethanol, and allowed to air-dry. Material insoluble in trichloroacetic acid was hydrolyzed in 200 p1 of 0.4 M NaOH at 80°C for 20 minutes, neutralized with an equal volume of glacial acetic acid, combined with 3 ml of scintillation fluid, and analyzed for radioactivity in a scintillation counter. Cells in the three wells that were not supplemented with r3H1-thymidine were exposed to low serum in the same manner as cells in the radioactively- labeled wells. Cells were removed from the three unlabeled wells with trypsin and counted by means of a hemocytometer. Cell counts from unlabeled wells were used to normalize incorporation of r3H]-thymidine to cell number.

Determination of culture doubling time Fibroblasts were plated into 12-well plates (Corning)

at 2,500 cells/well and were grown in DMEM/5% FCS for 0,2,4,6,10,14,17, and 24 days. At the end of each time period, cells were removed from wells with trypsin and were counted by Coulter counter. Results were expressed as the average of triplicate wells. The average percentage of viable cells per triplicate was determined by a trypan blue exclusion assay. The time course of increase in viable cells was used to calculate the doubling time of the cultures.

TRANSFORMING GROWTH FACTOR-f3l AND AGED FIBROBLASTS 171

RNA extraction and Northern blot analysis Total RNA was isolated from fibroblasts by the

method of Chomczynski and Sacchi (19871, denatured in 50% formamide (vollvol), resolved by electrophoresis in a denaturing 1.2% agarose gel, and stained with 0.5 pglml ethidium bromide. Eight pg of total RNA was applied to each lane of the gel, RNA was transferred to Duralose-UV (Stratagene, La Jolla, CA) and was crosslinked by ultraviolet radiation.

Northern blots were prehybridized and hybridized as previously described (Iruela-Arispe et al., 1991a). Rela- tive levels of SPARC and type I collagen mRNA were measured by hybridization to the corresponding [32P]- cDNA probe. The SPARC probe was a 557 bp BamH1- EcoRl fragment from mouse cDNA (Mason et al., 1986; Vernon and Sage, 1989), and the crl(1) collagen probe was a 1.1 kb EcoR1-EcoR1 fragment from human cDNA (Chu et al., 1982). DNA inserts were isolated as previously described (Iruela-Arispe et al., 1991a). After hybridization, blots were washed at a final stringency of 0.1 x SSC (1 x SSC is 0.15 M NaC110.015 M sodium citrate, pH 6.8) containing 0.1% sodium dodecyl sulfate at 65°C. Radiolabeled bands were visualized by autoradiography. Intensity of the bands on the Duralose was determined by the PhosphorImager Facility of the Mar- key Molecular Medicine Center a t the University of Washington. Variations in the amount of total RNA applied to the gels were normalized by hybridization of the blots with a bovine cDNA probe for 28s rRNA (Iru- ela-Arispe et al., 1991a).

Metabolic labeling and Western blot analysis After 72 hours in low-serum conditions, all control

cells and cells treated with 10 ng/ml TGF-pl were incubated for 44 hours in DMEM/5% FCS with 50 pg/ml sodium ascorbate, 64 pg/ml p-aminopropionitrile fuma- rate, and 25 pCi/ml L-E2,3,4,5-3HJ proline (Amersham, Arlington Heights, IL). Culture media were collected, centrifuged to remove cells and debris, and supplemented with proteinase inhibitors (2 mM phenylmeth- ylsulfonyl fluoride, 10 mM N-ethylmaleimide, and 2.5 mM EDTA as final concentrations). Aliquots of media equivalent to lo5 cells were concentrated by rotary evaporation, solubilized in equal volumes of SDS- PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) sample buffer, reduced with 50 mM dithio- threitol, and resolved by SDS-PAGE on slab gels (stacking and resolving gels were 4% and 8% acryla- mide, respectively). One set of gels was dried and visualized by autoradiography. Radioactive bands on auto- radiograms were quantified by densitometric scanning within the respective linear range. Proteins from the second set of gels were transferred to nitrocellulose and were stained with amido black. Blots were blocked overnight in phosphate-buffered saline containing 1% nonfat dry milWO.OB% Tween-20, exposed to antibody against type I collagen, and incubated in 5 pCi of [12511- Protein A (New England Nuclear). Radioactive bands in blots were visualized by autoradiography and sub- jected to quantitative densitometry by the PhosphorIm- ager Facility.

Radioimmunoassay For assays of SPARC, media were prepared as de-

scribed for Western blots, with the deletion of EDTA. SPARC was measured by a single-antibody competition assay (Everitt and Sage, 1992). SPARC standards (0- 2,500 ng/ml) were grepared in DMEM/5% FCS. Twenty-five p1 of [12 Il-human platelet SPARC containing 20,000 cpm in 50 mM Tris/l50 mM NaClil mM CaC12/l% Tween-20/1% nonfat dry milk, pH 7.8 (buffer 1) was added to 150 p1 of experimental samples or to SPARC standards. The 1'251]-SPARC was iodinated by the chloramine-T method (Hunter and Greenwood, 1962). Guinea pig antibodies against human SPARC (25 p1 at 0.15 mg/ml) in buffer 1 were added to each tube and were incubated at 4°C overnight. After incubation, 200 p1 of 25% polyethylene glycol (molecular weight 8,000) in buffer 1 lacking milk (buffer 2) was added to each tube to enhance precipitation. The tubes were centrifuged for 2-3 minutes a t 10,000 rpm, the supernates were aspirated, and the pellets were washed with 400 pl of 12.5% polyethylene glycol in buffer 2. Radioactivity in the pellets was measured by a Beck- man 8000 gamma counter. Standards and samples were assayed in triplicate.

Collagen gel contraction assay Collagen gels were prepared by a combination of 13.3

ml of a solution of 3 mg/ml bovine type I collagen (Vit- rogenm, Celtrix Laboratories, Palo Alto, CAI, 18.7 ml DMEM, 2.2 ml sevenfold concentrated DMEM, and 1.8 ml FCS. The mixture was divided into six aliquots of 5.4 ml (one aliquot €or each cell type in the study). Each aliquot received 0.6 ml of fibroblasts suspended in DMEM/5% FCS. The aliquots were then divided into two equal volumes, one of which was supplemented with 10 ng/ml TGF-pl. Each volume was dispensed into three plastic tissue culture plates with wells of 35 mm diameter (Corning) that were precoated with a thin layer of agarose. Each well contained 1 ml of DMEM with 1 mg/ml type I collagen, 5% FCS, and 8 x lo4 fibroblasts. The solutions of collagen gelled after 30 minutes of incubation at 37°C in 5% C0,/95% air. Di- ameters of gels were recorded at intervals of 24 hours as the average of two measurements made at right angles. TGF-Pl was replaced at 48-hour intervals. After 7 days, gels were dissolved with collagenase and the cells were counted with a hemocytometer.

RESULTS Analyses of DNA synthesis

Studies of relationships between aging and cellular synthetic responses to TGF-P1 can be complicated by differences in basal rates of proliferation between young vs. aged cells. Therefore, we sought to establish conditions that would minimize age-related differences in proliferation prior to stimulation with TGF-P1. Con- trol cultures of fibroblasts grown in DMEM/S% FCS exhibited a decrease in DNA synthesis that was correlated with the increasing age of the donor (Fig. IA). Additional analysis of cell proliferation rates in two of the cultures showed that cells from the 89-year-old exhibited a 33% greater doubling time (4.8 days) than cells from the 26-year-old (3.6 days) (data not shown).

172 REED ET AL.

L .- 2 2000

A 5%FCS T T

T 000

0 26 29 65 89 90 92

B 0.5YoFCS

26 29 65 89 90 92

Fig. 1. Analyses of DNA synthesis by fibroblasts from young and aged donors, Fibroblasts cultured under standard growth conditions (DMEM/5%FCS) showed an age-related decrease in incorporation of [3H1-thymidine (A). Similar analysis of the cultures after 72 hours in a low concentration of serum (0.5%) showed that levels of DNA synthesis had decreased significantly and that differences among cultures were minimal (B).

After 72 hours in low concentrations of serum (0.5% FCS), levels of DNA synthesis in all cultures had decreased to 4 0 % of controls and exhibited minimal variation (Fig. 1B). Under these conditions, there was little evidence of cell death. After 96 hours in low serum, the fibroblasts displayed no further decrease in thymidine incorporation, but evidence of cell death (e.g., rounding and detachment) was apparent in cultures from the older donors. Therefore, we established 72 hours as the optimal period of exposure to low levels of serum, prior to stimulation with TGF-p1. In addition to a decreased proliferative capacity, cells from the aged donors exhibited other phenotypic characteristics of senescence, which included increased size and the presence of cytoplasmic inclusions (data not shown).

Analyses of type I collagen and SPARC mRNA We began our investigation of the age-related effects

of TGF-Pl on the production of type I collagen and SPARC by quantitation of mRNA from fibroblasts cultured in the presence or absence of this cytokine. Ini- tially, we compared the basal levels (i.e., levels in media containing 0.5% FCS and lacking TGF-P1) of type I collagen and SPARC mRNA among the different cell cultures by quantitative analysis of Northern blots. Fi- broblasts from young donors contained consistently high basal levels of mRNA for type I collagen and SPARC, whereas the older donors exhibited variability among the cultures (data not shown).

After exposure to 10 ngiml TGF-$1, both young and old fibroblasts, with the exception of O4441A (donor age 29 yrs), exhibited increases in the 4.8 kb and 5.8 kb

mRNAs for type I collagen and in the 2.2 kb mRNA for SPARC. A representative blot is shown in Figure 2A. Most of the cultures showed small increases in mRNAs for collagen and SPARC in the presence of DMEM/5% FCS lacking TGF-Pl. This effect was more pronounced in the younger cell lines. After 2 2 4 4 hours of exposure to TGF-P1, cultures of fibroblasts from 26-, 65-, 89-, 90-, and 92-year-old donors increased type I collagen mRNA from 2.3- to 5.1-fold and SPARC mRNA from 2.0- to 6.2-fold, in comparison to controls exposed to DMEM/5% FCS lacking TGF-$l (Fig. 2B). There was no correlation between the baseline levels of type I col-

Fig. 2. Levels of type I collagen and SPARC mRNA increase in fibroblasts exposed to TGF-61. In A, confluent cells from the 89-year- old donor were grown in 0.510 serum for 72 hours and were subsequently exposed to 10 ng/ml TGF-p1 for 0,1,22, or 44 hours. A North- ern blot of total RNA revealed 4.815.8 kb mRNA for the prooll(1) chain of type I procollagen and 2.2 kb mRNA for SPABC, after hybridization with the corresponding r3'P]-cDNA. B illustrates the magnitude of increase in levels of mRNA for type I procollagen and SPARC in cultures exposed to TGF-pl vs. unexposed controls, as derived from phosphorimage analyses of Northern blots. All data were normalized to the corresponding signal from 285 rRNA. The ordinates indicate the level of mRNA expression of cultures exposed to TGF-pl, divided by the mRNA expression of controls. The abscissae indicate the hours of exposure to TGF-p1. After 44 hours of exposure to TGF-pl, increases in levels of mRNAs for type I procollagen and SPARC ranged from 2.3- to 5.1-fold and 2.0- to 6.2-fold, respectively. Similar results were obtained from a separate series of experiments that was quantitated by densitometry. The two sets of data differed from each other by an average (for all six cultures) of no greater than 25%. Culture AG04441A did not change with respect to steady-state levels of proul(1) collagen and SPARC mRNA in the presence ofTGF-p1.

TRANSFORMING GROWTH FACTOR-B1 AND AGED FIBROBLASTS

4 -

3 -

2 -

1 -

173

.Bmm

6 -

5 -

4 .

4 .

3 .

2 - 3 - 2 -

1 -

0 - 0 2 2 2 4 4 0 2 2 2 4 p 22 43 0 2 22 44

1 .

n.

~

Type I Collagen SPARC

AGO9266 (26 yr)

3.

2 .

1 .

‘1

I

2 .

1 -

O

I n

” 0 2 2 2 4 4 0 ? 2 2 #


AGO4441 A (29 yr)

“1


AGO8308 (89 yr)


AGO6291 A (90 yr)

4 1 3 i


AGO4061 (65 yr) Type I CollaQen SPARC

AGO4064 (92 yr)

n

174 REED ET AL.

5, i

H young cells El aged cells

.r

0.1 1 .o 5.0 10.0 TGF-t31 (ng/ml)

Fig. 3. Cells cultured from aged and young donors express similar increases in the level of type I procollagen mRNA in response to a range of concentrations of TGF-PI. Cultures from the aged donors and the young donor were grown in 0.5% serum for 72 hours and were subsequently exposed to 0, 0.1, 1.0, 5.0, and 10 ng/ml TGF-pl. The magnitude of increase in levels of mRNA for type I collagen, as derived from phosphorimage analyses of Northern blots, is shown. All data were normalized to the corresponding signal from 28s rRNA. Similar data were obtained with respect to SPARC mRNA (not shown). For all cultures, levels of synthesis of type I procollagen mRNA increased with the concentration of TGF-PI to 5 ng/ml; there was no significant difference in response between 5 and 10 ng/ml TGF-p1.

lagen and SPARC mRNA and the degree of response to TGF-P1. After 44 hours of exposure to TGF-p1, levels of mRNA in the two young donors that responded were increased by an average factor of 3.7 2 0.6 (standard error of the mean) for type I collagen and 2.3 5 0.3 for SPARC, whereas among the three older donors, levels of mRNA were increased by an average factor of 3.7 f 0.8 for type I collagen and 4.6 ? 0.9 for SPARC. A separate series of experiments was quantitated by densitometry. The two sets of data exhibited identical trends and differed from each other (for all six cultures) by an average of no greater than 25%.

In a separate series of experiments, we compared the synthesis of type I collagen mRNA by fibroblasts from the three aged donors and the 26-year-old donor in response to different concentrations of TGF-pl (0.1-10 ng/ml). For all cultures, levels of synthesis of type I collagen increased up to 5 ng/ml TGF-P1; there was no significant difference in response between 5 and 10 ng/ml TGF-Pl. The profiles of collagen mRNA synthesis vs. dose of TGF-pl were similar between fibroblasts from the three aged donors and duplicate experiments from the young donor (Fig. 3). In these experiments, the synthesis of SPARC mRNA demonstrated a pattern similar to that of type I collagen (data not shown).

Analyses of secreted type I collagen and SPARC We next examined the influence of TGF-P1 on the

synthesis and secretion of type I collagen protein by

fibroblasts from the various donors. SDS-PAGE analyses were performed on media that contained proteins radiolabeled metabolically with [3Hl-proline. With the exception of AG04441A (donor age 29)) we observed that fibroblasts exposed to TGF-P1 for 44 hours increased the secretion of proteins corresponding to the p r o d and proo2 chains of type I procollagen from 2.0- to 7.5-fold and increased the secretion of fibronectin from 1.9- to 6.8-f0ld, when compared to unstimulated cultures (Fig. 4A,B). Quantitative analysis of Western blots of the processed chains of type I procollagen revealed that all cultures, with the exception of AG04441A, increased the secretion of type I collagen from 1.8 ? 0.03 (SEM) to 5.0 2 2.3-fold in response to TGF-P1 (Fig. 5A,B). There was no correlation between donor age and the ability of fibroblasts to augment synthesis of type I collagen in response to TGF-P1. We also measured the influence of TGF-P1 on the secretion of SPARC by radioimmunoassay. For all cell cultures, with the exception of AG04441A) exposure to TGF-pl for 44 hours resulted in a 1.6 2 0.08- to 5.5 * 0.8-fold increase in levels of SPARC protein, in comparison to unstimulated controls (Fig. 5B). As we found with respect to the secretion of type I collagen, there was no correlation between age of fibroblasts and secretion of SPARC, either by unstimulated cells or in cells exposed to TGF-f31. In the two young donors that responded, levels of type I collagen and SPARC protein were increased by average values of 3.5 * 1.6-fold and 3.9 * 1.6-fold, respectively, whereas among older donors, levels of type I collagen and SPARC protein were increased by average values of 2.9 ? 0.5-fold and 2.3 % 0.4-fold) respectively.

Analyses of collagen gel contraction We employed a collagen gel contraction assay (Kono

et al., 1989) to study the age-related effects of TGF-P1 on the ability of fibroblasts to contract connective tissue (Fig. 6). For all cell cultures, the rate of collagen gel contraction was accelerated in the presence of 10 nglml TGF-P1. Differences between unstimulated cultures and cultures stimulated by TGF-P1 were maximal after 2 days, a t which time stimulated fibroblasts from young and old donors had contracted gels to diameters that averaged 67 f 7.4% and 64 f 1.3%, respectively, of the diameters of gels that contained unstimulated fibroblasts. Cultures exposed to TGF-p1 between 2 and 4 days contracted the gels at a rate lower than that exhibited by control cultures. It appeared that the lower rate of gel contraction was caused by increased resistance of the compressed gels to further deforma- tion. After 4 days, control cultures and cultures exposed to TGF-P1 exhibited gel diameters that were similar. For all cultures, contraction of gels was modest between 4 and 7 days. After 7 days, fibroblasts within all cultures had contracted gels to a similar extent. Cell counts revealed that all collagen gels contained a similar number of cells after 7 days of culture.

DISCUSSION Fibroblasts have a major role in the synthesis and

reorganization of extracellular matrix that occur during wound repair (Pierce et al., 1991). Wound healing is slowed in aged individuals, especially those with

TRANSFORMING GROWTH FACTOR-pl AND AGED FIBROBLASTS 175

I

R . . fibronectin I 7

Age(;) 26 29 65 89 90 92

Fig. 4. TGF-p1 stimulates the secretion of fibronectin and type I procollagen by fibroblasts: analysis by SDS-PAGE. Cells were cultured 44 hours in media containing [3H]-proline, with (+) or without (-1 10 ng/ml TGF-P1. Proteins in media were resolved by SDS-PAGE and autoradiography (A). Volumes of culture supernate reflecting the secretory activity of 10' cells were applied to each lane. Radioactive bands corresponding to fibronectin (large arrows) and to the proal and proa.2 chains of type I collagen (small arrows and arrowheads, respectively) are indicated. Relative levels of fibronectin and collagen chains within the gel were measured by densitometry (B). With the exception of culture AG04441A, cultures exposed to TGF-P1 for 44 hours exhibited an increase from 2 . k to 7.5-fold in type I collagen, and from 1.9- to 6.Sfold in fibronectin.

chronic diseases or those >85 years of age. It is generally accepted that aged fibroblasts retain their ability to respond to stimulation by growth factors but are less responsive than young cells. A determination of the cytokine-responsive elements of cellular function that are compromised in aged fibroblasts, as well as those that are unaffected, is of benefit for the appropriate use

Age(yr) 26 29 65 89 90 92

Fig. 5. TGF-pl augments production of type I collagen and SPARC by fibroblasts: analysis by Western blot and radioimmunoassay. A A Western blot of supernates from fibroblasts cultured in the presence or absence of 10 ng/ml TGF-pl was incubated sequentially with antibody against type I collagen and 112511-Protein A. Proteins in equal volumes of culture supernate (reflecting the secretory activity of lo5 cells) were resolved electrophoretically. The antibody has bound to bands repre- senting the al(1) and a2fU chains of type I collagen (arrow and arrow- head, respectively). A graphical representation of the degree to which TGF-p1 stimulated the secretion of type I collagen (as determined by phosphorimaging of Western blots from two separate experiments) and the secretion of SPARC (as measured by 3 4 radioimmunoassays per culture) is shown in B. With the exception of culture AG04441A, cells exposed to TGF-pl for 44 hours exhibited a 1.8 k 0.03 @EM)- to 5.0 i- 2.3-fold increase in type I collagen, and a 1.6 ? 0.0% to 5.5 t O.8-fold increase in SPARC.

of growth and morphoregulatory factors in the treat- ment of age-related disorders.

Human cells that are aged in vitro by multiple pas- sages can express abnormal characteristics as a conse- quence of long-term exposure to conditions of cell culture. Problems of this nature can be minimized by the use of low-passage cell cultures derived from aged donors (i.e., cells aged in vivo) (Schneider et al., 1981). For our experiments, we obtained celIs from donors of widely varying ages. We studied fibroblasts from do-

176 REED ET AL.

AGO4061 (65 yr) AGO9266 (26 yr) AG04441A (29 yr)

15.

10. * 5. 5. -9

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7

AGOB291A (90 yr) AGO4064 (92 yr) AGO6308 (89 yr)

4 5. 5. 5 .

0 , , , , , , , , 0 4 , , , , , , 0, , , . , . . , 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7

Fig. 6. TGF-P1 accelerates the contraction of collagen gels by fibroblasts. In A, type I collagen gels containing AGO6308 fibroblasts from an 89-year-old donor were cultured for 2 days in the presence (wells 1-3 of top row) or absence (wells 4-6 of bottom row) of 10 ng/ml TGF-B1. All gels are contracting (two are indicated by arrows), but cells exposed to TGF-P1 are contracting gels to a greater degree than cells in wells lacking TGF-P1. In B, fibroblasts within well 3 of(A) are shown. Cells (arrows) are rounded and refractile. Bundles of collagen fibers, aligned by forces of cellular traction, connect adjacent cells (a

nors >85 years of age because an age-related impairment in the responses to growth factors should be exhibited most acutely in cells derived from individuals of advanced age. The in vivo aged fibroblasts used in this study exhibited the phenotypic characteristics and decreased proliferative capacity characteristic of fibroblasts aged in vitro.

In the present study, we have demonstrated that dermal fibroblasts, obtained from individuals represent- ing a wide range of ages, were able to increase the synthesis of type I collagen and SPARC protein, and their corresponding mRNA, in response to TGF-01. The

long, narrow bundle is indicated by arrowheads). In C, the effects of the presence (-), or absence of TGF-P1 on the contraction of collagen gels by fibroblasts from young and old donors are summa- rized graphically. The ordinates of the graphs indicate the diameters of the gels (in mm); the abscissae indicate the elapsed time of culture (in days). Each point represents the averaged diameters of three separate gels. TGF-P1 stimulates gel contraction by fibroblasts from all donors, an effect that is most apparent after 2 days in vitro. A, x 0.5; B, X 130.

level of stimulation was not decreased in cells from individuals of advanced age. It appeared that in fibroblasts from 89-, 90-, and 92-year-old donors, the temporal increase in mRNA synthesis in response to TGF-fJ1 lagged behind the response of younger donors. How- ever, after 44 hours of exposure to the growth factor, cells from all donors displayed similar levels of stimulation. It is noteworthy that fibroblasts from the 65-year- old donor responded to TGF-pl as rapidly as did fibroblasts from the 26-year-old donor. This result is not unexpected, since clinical studies indicate that both young individuals and individuals up t o the sixth de-

TRANSFORMING GROWTH FACTOR-P1 AND AGED FIBROBLASTS 177

cade of life exhibit similar capacities to respond to per- turbations such as wounding (US. Senate Special Com- mittee on Aging, 1987-88). In a separate series of experiments with a range of concentrations of TGF-P1, we found that cells from the aged donors exhibited a similar dose-response, in comparison to cells from the young donor, with respect to induction of type I collagen and SPARC mRNA.

Type I collagen is the predominant collagen found in human skin and is particularly important in the archi- tectural support of the reticular portion of the dermis. Levels of dermal type I collagen decrease with aging in vivo and among fibroblasts aged in vitro. Reduction of type I collagen in the dermis is thought to contribute to both age-related atrophy of skin and its decreased ability to respond to pathologic processes that include wounding (Johnson et al., 1986; Love11 et al., 1987; Martin et al., 1990; Takeda et al., 1992). Decreases in the level of type I collagen in the skin with aging are a function of increased degradation as well as of decreased synthesis of this protein (Mays et al., 1991). In this regard, it is noteworthy that under similar culture conditions, dermal fibroblasts aged in vitro synthesize higher levels of collagenase than early-passage fibroblasts (West et al., 1989). During wound repair, TGF-P1 stimulates fibroblasts to proliferate and to synthesize increased amounts of extracellular matrix proteins that include specific collagens, glycoproteins, and pro- teoglycans (Ignotz and Massague, 1986; Roberts et al., 1986; Raghow et al., 1987; Pierce et al., 1989, 1992; Quaglino et al., 1991). Moreover, TGF-P1 inhibits collagenase production by fibroblasts (Edwards et al., 1987). Although modulation of many of these proteins might be beneficial to the healing process, the therapeutic utility of TGF-P1 has not been determined. The demonstration in the present study of the ability of TGF-P1 to stimulate the synthesis of type I collagen in cells from aged donors in vitro correlates with data in vivo, in which the topical application of TGF-P1 reversed the impairment in wound repair in aged rats (Reed et al., 1993b).

Although others have shown that TGF-P1 increases the synthesis of SPARC by fibroblasts (Wrana et al., 19911, the influence of aging on the expression of SPARC by cells stimulated with growth factors has not been studied systematically. Recent studies have shown that SPARC was present a t high levels relative to other cDNAs in a library obtained from cells derived from a patient with Werner’s syndrome, a disease of premature aging (Murano et al., 1991). Since SPARC is a major constituent of bone and binds CaZt, hydroxyap- atite, and type I collagen, it has been suggested that changes in the level of this protein in bone matrix might be of significance in diseases of aging bone, such as osteoporosis (Termine, 1990). A recent study compared levels of SPARC synthesis in bone cells derived from human sources ranging from fetuses to adults 60 years of age. Highest expression of SPARC was seen in osteoblasts from fetal and prepubescent (10-15-year- old) donors, followed by a decrease to 25% of maximal levels at age 30, with no further decreases between ages 30 and 60 (Fedarko et al., 1992). SPARC is associated with morphogenetic processes that involve changes in the shape, attachment, and motility of cells. Its expression is increased during angiogenesis in vitro and in

blood vessels of the developing mouse embryo in vivo (Iruela-Arispe et al., 1991a,b; Sage, 1993). Moreover, SPARC and/or fragments of SPARC are angiogenic in vitro and in vivo. We have observed that SPARC mRNA and protein are increased during the formation of granulation tissue in wound repair, results that are consistent with a role for SPARC in the remodeling of tissue and induction of angiogenesis (Reed et al., 1993a). Since SPARC is bound by type I collagen, the induction of both type I collagen and SPARC by TGF-P1 that we observed in vitro could reflect the physical as- sociation of the two proteins that occurs in wounded tissue (Reed et al., 1993a). Type I collagen might pro- vide a substrate upon which SPARC can attach and thereby be retained within sites of injury.

We also observed that dermal fibroblasts from young and aged donors exhibited a similar ability to contract collagen gels in the absence of TGF-P1 and to respond with enhanced contraction in the presence of this cytokine. The results of this model of wound contraction in vitro demonstrate that impaired healing of wounds in older individuals is not due to a decreased capacity of fibroblasts to apply traction forces to extracellular matrix.

In conclusion, we find that fibroblasts from aged donors, which exhibit the phenotypic characteristics of senescent cells, retain a significant capacity to respond to TGF-P1 with respect to biosynthetic and contractile functions that are important in wound repair. Since it appears that impaired wound healing in aged individuals does not result from a generalized inability of their cells to respond to TGF-P1, this cytokine may be of therapeutic benefit in this segment of the population. Other factors that might contribute to an age-related decline in wound repair include inhibition of cellular proliferation, inadequate production of cytokines and/or their receptors, or insufficient circulation for the delivery of nutrients and biologically active molecules.

ACKNOWLEDGMENTS The authors express gratitude to Dr. Luisa Iruela-

Arispe for the antibody against type I collagen and her assistance throughout the project, and to Dr. Timothy F. Lane for the antibody against SPARC and iodination of SPARC. We also thank Dr. Tom Norwood, Dr. Dan Twardzik, Cherry Tamblyn, Jane Ranchalis, and Tony Saulewicz for their assistance in cell culture. This work was supported by National Institutes of Health grant GM-40711. M.J.R. is supported by the PfizedAmerican Geriatrics Society Postdoctoral Fellowship Program.

LITERATURE CITED Border, W.A., and Ruoslahti, E. (1992) Transforming growth factor+

in disease: The dark side of tissue repair. J. Clin. Invest., 9O:l-7. Canney, P.A., and Dean, S. (1990) Transforming growth factor beta: a

promoter of late connective tissue injury following radiotherapy? Br. J. Radiol., 63:620-623

Chomczynski, P., and Sacchi, N. (1987) Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162;156159.

Chu, M.L., Myers, J.C., Bernard, M.P., Ding, J.F., and Ramirez, F. (1982) Cloning and characterization of five overlapping cDNAs specific for the human pro 01 (11 collagen chain. Nucleic Acids Res , 10:5925-5934.

Edwards, D.R., Murphy, G., Reynolds, J.J., Whitham, S.E., Docherty, A.J.P., Angel, P., and Heath, J.K. (1987) Transforming growth fac-

178 REED ET AL.

tor beta modulates the expression of collagenase and metalloprotei- nase inhibitor. EMBO J. , 6:189!&1904.

Everitt, E., and Sage, E.H. (1992) Expression of SPARC is correlated with altered morphologies in transfected F9 embryonal carcinoma cells. Exp. Cell Res., 199:13&146.

Fedarko, N., Vetter, U., Weinstein, S., and Robey, P. (1992) Age- related changes in hyaluronan, proteoglycan, collagen, and osteonectin synthesis by human bone cells. J. Cell. Physiol., 151r215- 227.

Funk, S.E., and Sage, E.H. (1991) The Ca2+-binding glycoprotein SPARC modulates cell cycle progression in bovine aortic endothelial cells. Proc. Natl. Acad. Sci. USA, 88t264S2652.

Funk, S.E., and Sage, E.H. (1993) Differential effects of SPARC and cationic SPARC peptides on DNA synthesis by endothelial cells and fibroblasts. J . Cell. Physiol., 154.53-63.

Harley, C.B., Goldstein, S., Posner, B.I., and Guyda, H. (1981) De- creased sensitivity of old and progeric human fibroblasts to a preparation of factors with insulinlike activity. J. Clin. Invest., 68:98€L 994.

Hunter, W.M., and Greenwood, F.C. (1962) Preparation of Iodine-131 labelled human growth hormone of high specific activity. Nature, 194:495-496.

Ignotz, R.A., and Massague, J . (1986) Transforming growth factor-p stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J . Biol. Chem., 261 t4337- 4345.

Iruela-Arispe, M.L., Hasselaar, P., and Sage, E.H. (1991a) Differential expression of extracellular proteins is correlated with angiogenesis in vitro. Lab. Invest., 64:174-186.

Iruela-Arispe, M.L., Diglio, C.A., and Sage, E.H. (1991b) Modulation of extracellular matrix proteins by endothelial cells undergoing angiogenesis in vitro. Arterioscler. Thromb., I1 :805-815.

Johnson, B.D., Page, R.C., Narayanan, AS., and Pieters, H.P. (1986) Effects of donor age on protein and collagen synthesis in vitro by human diploid fibroblasts. Lab. Invest., 5t490-496.

Kane, C., Hebda, P., Mansbridge, J., and Hanawalt, P. (1991) Direct evidence for spatial and temporal regulation of transforming growth factor p l expression during cutaneous wound healing. J. Cell. Phys- iol., 148:157-173.

Kono, T., Tanii, T., Furukawa, M., Mizuno, N., Kitajima, J., Ishii, M., and Hamada, T. (1989) Correlation between ageing and collagen gel contractility of human fibroblasts. Acta. Derm. Venereal. (Stockh.), 70r241-244.

Lovell, C.R., Smolenski, K.A., Duance, V.C., Light, N.D., Young, S., and Dyson, M. (1987) Type I and I11 collagen content and fibre distribution in normal human skin during ageing. Br. J. Dermatol., 11 7:419428.

Martin, M., El Nabout, R., Lafuma, C., Crechet, F., and Remy, J. (1990) Fibronectin and collagen gene expression during in vitro ageing of pig skin fibroblasts. Exp. Cell Res., 191 :8-13.

Mason, I.J., Taylor, A,, Williams, J.G., Sage, E.H., and Hogan, B.L.M. (1986) Evidence from molecular cloning that SPARC, a major prod- uct of mouse embryo parietal endoderm, is related to an endothelial cell “culture shock” glycoprotein. EMBO J., 5t1465-1472.

Mays, P.K., McAnulty, R.J., Campa, J.S., and Laurent, G.J. (1991) Age-related changes in collagen synthesis and degradation in rat tissues. Biochem. J., 276r307-313.

McCaffrey, T.A., and Falcone, D.J. (1993) Evidence for a n age-related dysfunction in the antiproliferative response to transforming growth factor-p in vascular smooth muscle cells. Mol. Biol. Cell., 4~315-322.

Montesano, R., and Orci, L. (1988) Transforming growth factor p stimulates collagen-matrix contraction by fibroblasts: Implication for wound healing. Proc. Natl. Acad. Sci. USA, 85:4894-4897.

Moses, J., Yang, E., and Pietenpol, J. (1990) TGF-p stimulation and inhibition of cell proliferation: new mechanistic insights. Cell, 63:245-247.

Murano, S., Thweatt, R., Schmookler-Reis, R.J., Jones, R.A., Moer- man, E.J., and Goldstein, S. (1991) Diverse gene sequences are over- expressed in Werner syndrome fibroblasts undergoing premature replicative senescence. Mol. Cell. Biol., 11 :3905-3914.

Orentreich, N., and Selmanowitz, V. (1969) Levels of biological functions with ageing. Trans. N. Y. Acad. Sci., 2.992-1012.

Penttinen, R.P., Kobayashi, S., and Bornstein, P. (1988) Transforming growth factor p increases mRNA for matrix proteins both in the presence and in the absence of changes in mRNA stability. Proc. Natl. Acad. Sci. USA, 85;1105-1108.

Phillips, P.D., Kaji, K., and Cristofalo, V.J. (1984) Progressive loss of the proliferative response of senescing WI-38 cells to platelet-derived growth factor, epidermal growth factor, insulin, transferrin,

Pierce, G., Brown, D., and Mustoe, T.A. (1991) Quantitative analysis of inflammatory cell influx, procollagen type I synthesis, and collagen cross-linking in incisional wounds: Influence of PDGF-BB and TGF-p1 therapy. J. Lab. Clin. Med., 11 7r373-382.

Pierce, G., Tarpley, J., Yanagihara, D., Mustoe, T.A., Fox, G., and Thomason, A. (1992) Platelet-derived growth factor (BB ho- modimer), transforming growth factor-pl, and basic fibroblast growth factor in dermal wound healing. Am. J . Pathol., 140:1375- 1388.

Pierce, G.F., Mustoe, T.A., Lingelbach, J., Masakowski, V.R., Gra- mates, P., and Deuel, T.F. (1989) Transforming growth factor-p reverses the glucocorticoid-induced wound-healing deficit in rats: Possible regulation in macrophages by platelet-derived growth factor. Proc. Natl. Acad. Sci. USA, 86:2229-2233.

Plisko, A., and Gilchrest, B.A. (1983) Growth factor responsiveness of cultured human fibroblasts declines with age. J. Gerontol., 38.513- 518.

Quaglino, D., Nanney, L., Ditesheim, J., and Davidson, J. (1991) Transforming growth factor-beta stimulates wound healing and modulates extracellular matrix gene expression in pig skin: Inci- sional wound model. J. Invest. Dermatol., 97:3p42.

Raghow, R., Postlethwaite, A.E., Keski-Oja, J., Moses, H.L., and Kang, A.H. (1987) Transforming growth factor-p increases steady state levels of type I procollagen and fibronectin messenger RNAs posttranscriptionally in cultured human dermal fibroblasts. J. Clin. Invest., 79r1285-1288.

Raines, E., Lane, T., Iruela-Arispe, M.L., Ross, R., and Sage, E.H. (1992) The extracellular glycoprotein SPARC interacts with platelet-derived growth factor (PDGF)-AB and -BB and inhibits the binding of PDGF to its receptors. Proc. Natl. Acad. Sci. USA, 89t1281- 1285.

Reed, M.J., Puolakkainen, P., Lane, T.F., Dickerson, D., Bornstein, P., and Sage, E.H. (1993a) Differential expression of SPARC and thrombospondin 1 in wound repair: Immunolocalization and in situ hybridization. J. Histochem. Cytochem. 41 :1467-1477.

Reed, M.J., Puolakkainen, P., Abrass, I.B., and Sage, E.H. (1993b) Transforming growth factor beta 1 reverses the impaired wound healing in aged rats. J. Amer. Ger. SOC., 41s: SA1.

Roberts, A.B., Sporn, M.B., Assoian, R.K., Smith, J.M., Roche, N.S., Wakefield, L.M., Heine, U.I., Liotta, L.A., Falanga, V., Kehrl, J.H., and Fauci, A.S. (1986) Transforming growth factor type p: Rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc. Natl. Acad. Sci. USA, 83t4167- 4171.

Sage, E.H. (1993) SPARC and type I collagen are functional proteins in angiogenesis. J . Cell Biochem., 17E:144.

Sage, E.H., and Bornstein, P. (1991) Extracellular proteins that mod- ulate cell-matrix interactions: SPARC, tenascin, and thrombospondin. J . Biol. Chem., 266:14831-14834.

Sage, H., Vernon, R., Funk, S., Everitt, E., and Angello, J. (1989a) SPARC, a secreted protein associated with cellular proliferation, inhibits cell spreading in vitro and exhibits Ca2’-dependent binding to the extracellular matrix. J. Cell. Biol., 109:341356.

Sage, E.H., Decker, J., Funk, S.E., and Chow, M. (198913) SPARC: a Ca2+-binding extracellular protein associated with endothelial cell injury and proliferation. J. Mol. Cell. Cardiol., 21:13-22.

Sage, E.H., Vernon, R.B., Decker, J., Funk, S.E., and Iruela-Arispe, M.L. (1989~) Distribution of the calcium-binding protein SPARC in tissues of embryonic and adult mice. J. Histochem. Cytochem., 37t819-829.

Salonen, J., Domenicucci, C., Goldberg, H.A., and Sodek, J. (1990) Immunohistochemical localization of SPARC (osteonectin) and denatured collagen and their relationship to remodelling in rat dental tissues. Arch. Oral Biol., 35t337-346.

Schneider, E.L., Monticone, R., Smith, J., Braunschweiger, K., and Roberts, T. (1981) Skin fibroblast cultures derived from members of the Baltimore Longitudinal Study: A new resource for studies o f cellular aging. Cytogenet. Cell Genet., 31 :4046.

Shah, M., Foreman, D., and Ferguson, M. (1992) Control of scarring in adult wounds by neutralising antibody to transforming growth factor p. Lancet, 339r213-214.

Shuster, S., Black, M.M., and McVitie, E. (1975) The influence of age and sex on skin thickness, skin collagen and density. Br. J. Derma- tol., 93539-643.

Sporn, M.B., and Roberts, A.B. (1992) Transforming growth factor-p: Recent progress and new challenges. J . Cell Biol., 119t1017-1021.

Stanulis-Praeger, B.M., and Gilchrest, B.A. (1986) Growth factor responsiveness declines during adulthood for human skin-derived cells. Mech. Ageing Dev., 35t185-198.

Takeda, K., Gosiewska, A., and Peterkofsky, B. (1992) Similar, but not and dexamethasone. J; Gerontoc, 39;ll-17. identical, modulation of expression of extracellular matrix compo-

TRANSFORMING GROWTH FACTOR-61 AND AGED FIBROBLASTS 179

nents during in vitro and in vivo aging of human skin fibroblasts. J. Cell Physiol., 153r450-459.

Termine, J.D. (1990) Cellular activity, matrix proteins, and aging bone. Exp. Gerontol., 25.917-221.

Termine, J.D., Kleinman, H.K., Whitson, S.W., Conn, K.M., Mc- Gamey, M.L., and Martin, G.R. (1981) Osteonectin, a bone-specific protein linking mineral to collagen. Cell, 26r99-105.

Tsuji, Y., Ide, T., Ishibashi, S., and Nishikawa, K. (1984) Loss of responsiveness in senescent human TIG-1 cells to the DNA synthesis-induc- ing effect of various growth factors. Mech. Ageing Dev., 27:219-232.

Uitto, J., Halme, J., Hannuksela, M., Peltokallio, P., and Kivirikko, K.I. (1969) Protocollagen proline hydroxylase activity in the skin of normal human subjects and of patients with scleroderma. Scand. J. Clin. Lab. Invest., 23241-247.

US. Senate Special Committee on Aging. (1987-1988) Aging Amer- ica: Trends and Projections. Dept. Health & Human Services, Wash- ington, DC.

Vernon, R.B., and Sage, E.H. (1989) The calcium-binding protein SPARC is secreted by Leydig and Sertoli cells of the adult mouse testis. Biol. Reprod., 40r1329-1340.

Vuorio, T., Kahari, V.-M., Black, C., and Vuorio, E. (1991) Expression of osteonectin, decorin, and transforming growth factor-81 genes in fibroblasts cultured from patients with systemic sclerosis and mor- phea. J. Rheumatol., 18.247-251.

West, M.D., Pereira-Smith, O.M., and Smith, J.R. (1989) Replicative senescence of human skin fibroblasts correlates with a loss of regulation and overexpression o f collagenase activity. Exp. Cell Res.,

Wrana, J.L., Overall, C.M., and Sodek, J. (1991) Regulation of the expression of a secreted acidic protein rich in cysteine (SPARC) in human fibroblasts by transforming growth factor p. Eur. J. Bio- chem., 197.319428.

~ : i 3 a i 4 7 .

Documents

TGF-?1 induces the expression of type I collagen and SPARC, and enhances contraction of collagen gels, by fibroblasts from young and aged donors