Influences of endogenous and exogenous TGF-/3 on elastin in rat lung fibroblasts and aortic smooth muscle cells STEPHEN E. MCGOWAN Department of Veterans Affairs Research Service and the University of Iowa College of Medicine, Iowa City, Iowa 52242 McGowan, Stephen E. Influences of endogenousand exogenousTGF-P on elastin in rat lung fibroblasts and aortic smooth musclecells.Am. J. Physiol. 263 (Lung Cell. Mol. Physiol. 7): L257-L263, 1992.- The factors that regulate elastin production during neonatal lung development have not been elucidated. Previous investigations suggestedthat transforming growth factor-p (TGF-/3) increaseselastin production by neonatal rat lung fibroblasts (LF). We examinedwhether this effect of TGF-P was unique to these cells or was evident in other neonatal cells, which constitutively produce elastin, or in cells from adults, whose constitutive elastin production is low. We have quantitated soluble elastin, elastin mRNA, and TGF-P production in primary cultures of smooth musclecells (SMC) and LF from neonatal and adult rats and have examined the alterations in solubleelastin and elastin mRNA that result from adding 100pM exogenousTGF-P, to these cultures. Unsupplemented cultures of LF and SMC obtained from neonatal rats exhibited higher steady-state levels of elastin mRNA and contained more soluble elastin in their culture medium than did cellsfrom adult animals.When neonatal LF were supplemented with 100pM TGF-P,, they showeda significant increasein the solubleelastin content of their culture mediumand their steadystate elastin mRNA. Neither LF obtained from adults nor SMC obtained from neonatal or adult rats significantly increased their soluble elastin or steady-state elastin mRNA after the addition of exogenousTGF-P. When neonatal LF were supplemented with an anti-TGF-P neutralizing antibody, the soluble elastin content of the culture medium decreasedsignificantly. Thesedata suggestthat the responsiveness of elastin expression to TGF-P is limited to neonatal LF and that endogenousTGF-P influences elastin production by neonatal LF. lung development; extracellular matrix PULMONARY ELASTIN synthesis in vivo is under strict developmental control and is most pronounced during the late fetal and early neonatal stages of mammalian development. Although elastin is abundantly produced in the aorta and lungs of neonates, in normal adults these organs contain very low steady-state levels of elastin mRNA. The chronological factors regulating elastin production and deposition are incompletely understood and may involve multiple steps, namely transcription, translation, export, and cross-linking. Previous investigations have shown that transcriptional regulation is an important control point for elastin production in the aorta and in the bovine nuchal ligament (8, 28). Although the developmental regulation of elastin synthesis has been less extensively investigated in the lung, elastin synthesis correlates with the level of elastin mRNA in this organ as well. Elastin deposition begins during the late glandular stage, intensifies during the canalicular and alveolar stages of development, and extends postnatally for - 14 days in rats and - 7 yr in humans (5). The newly formed elastin fibers assume a diffuse distribution during the late glandular stage but then localize into more discrete aggregates associated with collagen fibrils during the formation of alveoli. Recent studies by Heine
et al. (14) suggest that transforming growth factor (TGF)-P, colocalizes with collections of type III collagen in the clefts between branching ducts during the canalicular stage. They suggest that TGF-& may promote collagen formation in these clefts. Morphological studies by Fukuda et al. (13) indicate that elastin formation coincides with collagen formation in these cleft regions. However, there are currently no data available that directly relate TGF-P and elastin formation in the developing lung. Investigations by Liu and Davidson (21) and in our laboratory (22) have shown that supplementing cell cultures with TGF-& promotes elastin synthesis in neonatal porcine smooth muscle cells (SMC) and neonatal rat lung fibroblasts (LF) in vitro. These studies prompted three questions: 1) Is this TGF-P-mediated increase in elastin synthesis limited to a period of abundant constitutive elastin synthesis, as occurs in neonates, or can TGF-P also increase elastin synthesis in cells derived from adults when constitutive elastin synthesis is low? 2) Do primary cultures of LF and SMC obtained from neonates differ from those obtained from adults in their production of TGF-P? and 3) Does endogenous TGF-& influence elastin production by cultured neonatal cells? To address these questions, we have studied elastin and TGF-P in primary cultures of LF and SMC from neonatal and adult rats and have examined the alterations in soluble elastin and elastin mRNA that result from adding exogenous TGF-& to cultures of LF and SMC obtained from rats of these ages. METHODS CeZZ culture of rat LF and S2MC.After anesthetization with ketamine, the lungswere removed from Sprague-Dawleyrats at 5 days after birth or from 150-g adults, washedwith sterile Dulbecco’sphosphate-bufferedsaline (PBS), and then minced into 1-2 mm3 pieces. The tissue was incubated at 37°C with modified Hanks’ balanced salt solution (mHBSS) containing 0.5 mg of trypsin, 0.3 mg of collagenase(Sigma Chemical, type I), and 0.2 mg EDTA/ml. After 20 and 40 min, the mHBSS was removed, and the tissue was combined with Ham’s F-12 medium containing 10% fetal bovine serum (FBS), fresh mHBSS was added, and the incubation was continued for a total of 60 min. The mHBSS, containing the digestedtissue,was strained through a 150-pmfilter, and the cells that passedthrough the filter were seeededat a density of 1 x IO7 tells/75-cm2 tissue culture flask. After 1 h, the nonadherent cellswere removed,and the adherent cells were cultured in Ham’s F-12 medium containing 10% FBS, 100 U of penicillin, and 100 pg/ml of streptomycin. Removing nonadherent cells after 1 h provided a cell population that wasprimarily fibroblasts (6, 10). The cellswere maintained by these culture conditions for 8 days and then subcultured into 25-cm2flasks or 2.0-cm2wells at a density of 3 x lo* cells/cm2.The subculturedfibroblasts reachedconfluence in 3-4 days and were maintained in medium containing 10% FBS for a total of 8 days and then in medium containing 5% L257
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
L258
ELASTIN AND TGF-@ PRODUCTION
newborn calf serum and IO mM N-2-hydroxyethylpiperazineN’-2-ethanesulfonic acid (HEPES) for an additional 5-7 days. SMC were obtained from the aortas of 3-day-old rats after digestion with elastase and collagenase with a technique that has been described previously (23). Adult aortic SMC were cultured from explants that had been removed from 150-g rats. The rats were anesthetized with ketamine, and the thoracic and abdominal cavities were entered. After exsanguination, the aorta was removed in entirety, the fascia and associated tissues were removed, and the aorta was washed with PBS. A longitudinal incision was made to expose the intimal layer. The intimal and medial layers were separated from the adventia with a forceps and then minced. The l-mm3 pieces of the media were allowed to adhere to tissue culture dishes and then supplemented with Dulbecco’s minimum essential medium (MEM) supplemented with 20% FBS, 1 mM sodium pyruvate, MEM nonessential amino acids, 100 U/ml of penicillin, and 100 pg/ml of streptomycin. After lo-14 days the cells were subcultured as described previously (23). The production of elastin and TGF-P was analyzed using cells after one or two subcultivations. Quantitation of soluble elastin in the culture medium. Prior to quantitating elastin in triplicate wellsfor eachcondition, the LF were washedtwice with PBS and placed in medium containing 0.5% FBS and 25 pg ,&aminopropionitrile (BAPN)/ml for 24 h, and the SMC were placed in medium containing 5% newborn calf serum and BAPN. BAPN prevents cross-linking of tropoelastin (TE) and thereby facilitates the detection of soluble elastin in the culture medium. During the final 12 h of this period, some cultures remained unsupplementedwhile others were supplementedto a final concentration of 100 pM TGF-@ (TGF-P, from porcine platelets; R and D Systems,Minneapolis, MN). This concentration was previously shown to maximally increaseTE production by neonatal LF (22). The culture media were then removed, and fresh medium containing BAPN, with or without 100pM TGF-& was added,and the cells were incubated for an additional 24 h. The conditioned mediumthat was collectedafter this final 24 h period wasremoved,cooledto 4”C, centrifuged at 500 g for IO min, and the supernatants were stored at -70°C. The residualcell layers werewashedwith PBS and removed from the wells by scrapingand were sonicatedfor 45 s to achieve a homogeneousmixture. The sonicated cell layers were supplementedwith 0.2 M sodium acetate, pH 7.0, 1% aminonitrillo-sulfonic acid, and the protein wasremoved by extracting with phenol. Aliquots of the aqueousphase were treated with proteinaseE (Sigma,St. Louis, MO), and the DNA content was determined by quantitating the fluorescent intensity of the DNA-ethidium bromide complex, which was sensitive to deoxyribonycleaseI digestion (3). The solubleelastin in the conditioned mediawasquantitated usinga competitive enzyme-linked immunoassay(ELISA) similar to that describedby Davidson and Sephel (9). Immulon 2 microtiter plateswere coatedwith oxalic acid-solubilizedelastin (62.5 rig/well) that had been isolatedfrom adult rat aortas. The competing antigen was a mixture of TE moieties, which were isolatedfrom the conditioned mediumcollectedover 24 h, in the presenceof BAPN, from neonatal rat aortic SMC that had been subcultured 2 wk previously (12). The TE moieties coeluted from a Vydac C,, column. An aliquot was subjectedto sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) followed by either silver staining or immunoblotting, using goat anti-rat cw-elastinas the first antibody. The major immunoreactive and silver-stained bandsmigrated at 69 and 59 kDa, although other bands were evident. The protein concentration was determined using bicinchoninic acid (36). To generate a standard curve for the ELISA, serial dilutions of the competing antigen (0.8-12.2 ng) were mixed with goat anti-rat cu-elastin(Elastin Products, Pacific, MO) and incubated overnight at 4°C. The antigen-antibody mixture wasthen incubated with the a-elastin-coated plate under nonequilibrium condi-
BY RAT MESENCHYMAL
CELLS
tions. The free antibody that bound to the plate wasquantitated with peroxidase-labeledrabbit anti-goat immunoglobulin G UgG)
l
Because the ELISA detects other soluble elastin moieties besides75-kDa TE, cultures obtained from 5-day LF that had been exposedto 100 pM TGF-P for 24 h or had remained unexposedwere metabolically labeledwith [3H]proline. Twentyfour hours prior to labeling, the serum concentration of the culture medium was reducedto 0.5%, and BAPN was addedto achieve a final concentration of 25 pg/ml. After this 24 h preincubation, the culture medium was removed, the cells were washedthree times with PBS, and Dulbecco’sMEM (without serumor proline) was addedfor 30 min. This mediumwasthen changedto the samemediumexcept it now contained 30 &i/ml Z-[2,3,4,5,3H]proline, and the pulsewascontinued for 1 h. After removing the pulse medium, the cells were washed and then incubated in medium containing 1 mM proline for 1, 2, 4, or 6 h. The culture media and cell layers were collected and processedas describedpreviously (22). Newly labeledTE moieties in the culture media were immunoprecipitated with goat antirat cu-elastinfollowed by Staphlococcus aureus protein G bound to agarose(Gamma-bind G; Genex, Gaithersburg, MD). After washing,the TE moietieswereeluted from the agarosebeadsin a SDS-containing buffer and subjectedto SDS-PAGE (22). Autoradiograms were obtained as described and scannedwith a Shimadzu CS-9OOOUdensitometer. Analyzing elastin mRNA. The cell layers from four 25-cm2 flasksthat had beenexposedor remainedunexposedto 100pM TGF-P for 24 h were combined, and the RNA was isolated (1, 22). Northern analyseswere performed aspreviously described using 1.2% agarosegelscontaining 6% formaldehyde (22). After electrophoresis,the gels were stained with ethidium bromide, destained,and photographedto determinethe distancesthe 28s and 18s ribosomalRNA had migrated and to assurethat equal amountsof RNA had been loaded.The RNA wastransferred to a cationic nylon membrane (Nytran, Schleicher and Schuell, Keene, NH), which was prehybridized and then hybridized by incubating for 16 h at 42°C with 32P-labeledcHEL2 (32PcHEL2). The l.2-kb cDNA, cHEL2, was originally obtained from Dr. Joel Rosenbloom(University of Pennsylvania) and wasderived from the 5’ end of the human elastin gene(17). The cHEL2 was labeled with [a- 32P]dCTP (3,000 Ci/mmol Du Pont-New England Nuclear, Wilmington, DE) using the random primer technique (22). After hybridization and washing, the membranewas exposedto Kodak XOMAT-AR film using an intensifying screen.Exposure times were varied to achieve bandswith densitiesthat werewithin the detection limits of the film, and the resultant autoradiogramswere subjectedto densitometry. After removal of the 32P-cHEL2 and prehybridization, the membraneswere hybridized with 32P-labeledglyceraldehyde 3-phosphate dehydrogenase( [32P]cGAPDH) , a nearly full-length cDNA for rat glyceraldehyde3-phosphate dehydrogenase(11, 22). The plasmid containing this cDNA was a gift from Dr. Alan Goodridge (University of Iowa). The gene for GAPDH is not induced by TGF-P (38) and thus served as a control to normalize for differences in the quantities of RNA that were loadedin the various lanesof the gelsusedfor Northern analysis. Quantitating the biological activity of TGF-p in conditioned media. Neonatal and adult rat LF and SMC were cultured in 5% FBS in 9.3-cm2wells for 14 days after the first subcultivation.
At that time, the medium was removed, the cells were washed with PBS, and medium without serum (HBlOl) was added. This serum-freemedium was conditioned by the cells for 48 h, chilled to 4”C, and centrifuged to remove nonadherent cells,and the supernatant was stored sterilely at 4°C. The TGF-P content of someof the culture media was analyzed without acidification. Transient (30 min) acidification to pH 3.0 dissociatesTGF-@ from its latency-associatedpeptide
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ELASTIN
AND
TGF-p
PRODUCTION
and assures that all of the TGF-P in the medium is biologically active at the time the assay is performed. However, it does not indicate what portion of the TGF-P was biologically active during the cell culture period. The TGF-@ activity detected in media that were not exposed to acid was at the lower limits of detection using a bioassay in which TGF-P inhibits mink lung cell growth (20). The unacidified samples contained only 7.5 t 3.6% (mean t SE, n = 11) of the TGF-P activity that wasfound in the respective acidified samplesfrom the growth-inhibition assay.Therefore, only the data obtained from acidified samples are presented. A receptor binding (radioreceptor) assaywas usedto quantitate TGF-/? (38). This assayrelies on the ability of proteins in the conditioned mediumto competewith TGF-P for binding to specific receptorsfor TGF-P. Therefore, it is more specific than growth inhibition or stimulation assaysas it dependson one biological event, namely receptor binding, rather than a cascade of events, including receptor binding, signal transduction, and ultimately DNA synthesis. Cells from a rat kidney fibroblast line (NRK-49F) were subcultured in a 24-well plate and maintained for 24 h. 1251-labeled TGF-P ( 1251-TGF-P)wasprepared using glucose oxidase and lactoperoxidase and carrier-free TGF-& from R & D Systems (24). TGF-@ (2 pg) was dissolved in 20 ~1of 4 mM HCl and supplementedwith 10 ~1 of 0.3 M sodiumphosphate,pH 7.2. Glucoseoxidaseand lactoperoxidase (50 ng, both) and 1 mCi of Na 1251 were added.The reaction was initiated by adding 100pg of @-D-&CO%? and terminated after 5 min by adding 10 ~1 of saturated tyrosine, 100 ~1 of 60 mM potassiumiodide, and 150 ~1of 8 M urea. The free Na1251was removed by elution from a PD-10 gel filtration column (Pharmacia, Piscataway, NJ) in 4 mM HCl, 0.075 M NaCl, 0.1% bovine serumalbumin. The 1251-TGF-@ had a specific radioactivity of -200 &i/pg and wasstored in aliquots at -20°C. The cell layers were incubated with various dilutions of conditioned mediumfor 2 h at 24°C washedand then incubated for 2 h with 1251-TGF-pThe cellsin somewellswereincubated with various concentrations (lo-200 pM) of nonradioactive TGF-P to compete with the 1251-TGF-pfor binding to receptors. After washing to remove unbound 1251-TGF-&, the bound 1251was solubilized and an aliquot was subjectedto gamma spectrometry. Nonspecific binding (1251-bindingin the presenceof a 300-fold excessof recombinant human TGF-& suppliedby Genentech) was subtracted from the total 1251bound for all wells and comprised 30-35% of the total 1251bound. Statistical analyses. The data are expressedas meansk SE. Comparisonsamong cultures that were exposed to different concentrations of TGF-P were madeby performing an analysis of variance. The multivariate generallinear hypothesis module of the SYSTAT (SYSTAT, Evanston IL) program was used. Where appropriate, Student’s t test for unpaired variables was used.Differences were consideredsignificant when P < 0.05. RESULTS
Effect of exogenous TGF-6 on soluble elastin production by LF and SMC. The first objective was to examine the
ability of exogenous TGF-@ to increase elastin production in cells obtained from rats at times when constitutive elastin synthesis was high (neonates) or low (adults). For each cohort of isolated cells, elastin production by cells exposed to 100 pM TGF-/3 was compared with control cells from the same primary isolation, which were not exposed to TGF-P. The quantity of soluble elastin in the media of control cells in the absence of exogenous TGF-p was 348.9 t 98.5 ng/pg of DNA (mean t SE, n = 3 experiments) during 24 h for LF obtained from 5-day-old rats and 55.9 t 8.2 ng/pg DNA (n = 3) for LF from adult rats (P < 0.01). The quantity of soluble elastin in the
BY RAT
MESENCHYMAL
L259
CELLS
media of control SMC was 3,721.l t 1,298.0 ng pg DNA-I. 24 h-l (n = 3) for SMC obtained at 3 days and 15.02 t 2.7 ngapg DNA-l.24 h-l (n = 3) for cells obtained from adult rats (P < 0.05). When cultures of LF from 5-day-old rats were made 100 pM in TGF-& a 2.31 t 0.31 fold increase (P c 0.01) in the soluble elastin content of the culture medium was observed compared with that in control cultures, which had been obtained from the same animals but had not received TGF-@ (Fig. 1). In contrast, TGF-@ did not significantly alter the soluble elastin content in the culture medium of LF, which were obtained from adult animals (P = 0.086). Adding TGF-P to cultures of SMC, which were obtained from either 3-day-old or adult rats, did not significantly alter the amount of soluble elastin in the culture medium (Fig. 2) *Because the antielastin antibody that was used in the ELISA assay detects soluble elastin moieties other than tropoelastin (TE), the ELISA does not specifically quantitate intact 75-kDa TE. Therefore, TE synthesis was also evaluated by pulse labeling LF from 5-day-old rats with [3H]proline. The results are shown in Fig. 3 and demonstrate that exposing cells to TGF-P is associated with a 2.0- and a U-fold increase in the quantity of 75-kDa TE present in the culture medium after chasing for 1 and 2 h, respectively. These data suggest that the TGF-P-mediated increase in soluble elastin detectable by the ELISA, at least partially, results from an increase in elastin synthesis. l
**
5 day
Age
q
protein
q
mRNA
Fig. I. Effects of exposure to transforming growth factor (TGF)-P on quantity of soluble elastin in medium (hatched bars) and elastin (stippled bars) mRNA in cultures of lung fibroblasts (LF) from &day-old and adult rats. Soluble elastin content of culture medium was quantitated by enzyme-linked immunoassay (ELISA, see text). Increase in soluble elastin and elastin mRNA in cultures of LF exposed to 100 pM TGF-P are expressed relative to those of LF obtained from same rats but that were not exposed to exogenous TGF-P (control, dashed line). Filters that were used for Northern analysis were first probed with a cDNA for elastin cHEL2 and then reprobed with a cDNA for glyceraldehyde 3-phosphate dehydrogenase. Autoradiograms were subjected to densitometry, and density of band representing elastin mRNA was normalized to density of corresponding band for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA for each sample. Ratios of density of elastin mRNA to GAPDH mRNA for samples from flasks that had been exposed to 100 pM TGF-P were expressed as fold increase from ratio for control samples that had not been exposed to TGF-0. Bars, mean (n = 3 sets of cultures from 3 separate animals for all groups, except n = 8 for mRNA from 5-day-LF); error bars, 1 SE. * P < 0.05, ** P < 0.01.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
L260
ELASTIN
AND TGF-P PRODUCTION
BY RAT MESENCHYMAL
ABCD
4
CELLS
EFGH
220
97
3 day
Adult Age m
protein
c]
mRNA
Fig. 2. Effects of exposure to TGF-P on quantity of soluble elastin in medium (hatched bars) and elastin mRNA (stippled bars) in cultures of aortic smooth muscle cells @MC) from 3-day-old and adult rats. Soluble elastin content of culture medium was quantitated by ELISA (see text). Increases in soluble elastin and elastin mRNA in cultures of SMC exposed to 100 pM TGF-fl are expressed relative to those by LF obtained from same rats but that were not exposed to exogenous TGF-0 (control, dashed line). Northern analyses were performed as described in Fig. 1 and in text. Bars, mean (n = 3 sets of cultures from 3 separate animals for all groups, except n = 4 for mRNA from adult SMC); error bars, 1 SE.
Effect of TGF-/I on elastin mRNA in cultures of LF and SMC. Northern analyses were performed to compare elastin expression in neonates and adults and to determine whether treatment with TGF-P increased elastin mRNA in LF and SMC. Figure 4 demonstrates that the steadystate levels of elastin mRNA in rat SMC (A) and LF (B) that were not exposed to exogenous TGF-P (lanes 1 and 4 in A; lanes 1 and 3 in B) are lower in adult animals than in neonates. The nature of the larger RNA (5.1 kb) has not been characterized, although it is larger than those attributable to alternative splicing (30). Multiple Northern analyses were performed using RNA from cells of the various ages to determine whether exposure to exogenous TGF-P increases steady-state elastin mRNA. Figure 4 also shows representative Northern analyses of RNA isolated from TGF-P-exposed SMC (lanes 2,3,5, and 6 in A) and LF (lanes 2 and 4 in B). Multiple analyses were performed to compare the steady-state elastin mRNA in LF that were exposed to 100 pM TGF-P with that of contol cells isolated from the same rats but that were not exposed to TGF-P. In Figs. 1 and 2, the ratios of the density of elastin mRNA to GAPDH mRNA for samples from flasks that had been exposed to TGF-P have been expressed as the fold increase from the ratio for control samples that had not been exposed to TGF-P. A mean, 1.6-fold (n = 8, P < 0.05) increase in elastin mRNA was observed in 5-day LF exposed to TGF-P (Fig. l), whereas TGF-P exposure did not significantly increase the steadystate level of elastin mRNA in cultures of lung fibroblasts from adult rats. Northern analyses of RNA that was isolated from SMC that had been exposed to TGF-@ or remained unexposed were compared in a similar fashion. A significant increase in elastin mRNA was not observed after TGF-P treatment of SMC obtained from rats at either age (Fig. 2).
66
43
Fig. 3. Effects of TGF-fi on incorporation of [“Hlproline into newly svnthesized tropoelastin (TE) in cultures of LF from 5dav-old rats after a l-h exposure to [3H]proline and then incubating with unlabeled proline. One group (lanes B, D, F, and ?Z) of flasks received TGF-@ during 24 h immediately preceeding pulse, and a second group (lanes A, C, E, and G) did not receive TGF-P. All flasks were incubated for 1 h with [3H]proline and were incubated for an additional period in medium not containing [3H]proline, which lasted for either 1 h (lanes A and B), 2 h (lanes C and D), 4 h (lanes E and F), or 6 h (lanes G and H). Media were immunoprecipitated with an anti-rat cu-elastin antibody prior to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), autoradiography, and densitometry.
TGF-p secretion by cultured rat LF and SMC. The biological activity of TGF-/3 was assayed in the acidtreated conditioned media from rat LF and SMC that were isolated from neonatal and adult animals. The radioreceptor assay demonstrated that the secretion of TGF-@ was 14.76 + 2.09 and 6.35 + 2.69 ng/pg DNA during 48 h for lung fibroblasts obtained from 5-day-old and adult rats, respectively (Fig. 5A). There was a trend toward higher secretion by neonatal cells (P = 0.06). SMC that were isolated from 3-day-old rats also trended toward greater secretion, 28.18 + 8.10 and 7.75 f 3.02 ng TGF-/3/pg DNA over 48 h compared with adult rats (P = 0.06; Fig. 5B). Effect of endogenous TGF-p on soluble elastin production by neonatal LF. These data demonstrate that LF and SMC obtained from neonatal rats secrete considerable quantities of TGF-fi. Neonatal lung fibroblasts endogenously secrete -15 ng of TGF-P/2-cm2 well (containing - 1 pg DNA and 1 ml medium), which is approximately sixfold more than the 2.5 ng exogenous TGF-@ (100 PM) that was shown to stimulate soluble elastin production. These findings suggest that TGF-fi, which was endogenously produced by LF, could potentially stimulate soluble elastin production by these cells. To test this
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ELASTIN
AND TGF-6’ PRODUCTION
BY RAT MESENCHYMAL
CELLS
./ .f(
1.4 kb
123
455
hypothesis, the effect of a polyclonal neutralizing antibody against TGIF-P on soluble elastin production by neonatal LF was examined. The culture media, collected over 24 h, from LF of 5-day-old rats contained 967.8 + 105.73 ng soluble elastin/pg DNA in the absence of rabbit antiTGF-@ (n = 4 experiments, each with 3 wells/group), whereas the media contained 682.6 + 107.2 ng in the presence of the antibody (see Fig. 6). This decrease was significant (P < 0.01) and suggests that endogenous TGF-fi influences the elaboration of soluble elastin by neonatal LF. Adding an irrelevant antibody, rabbit antimouse IgG, did not alter the soluble elastin content of the culture medium (data not shown). DISCUSSION
These studies demonstrate that, in cells cultured from rat lungs and aortas, elastin mRNA is higher in neonates than in adults. These findings are similar to those of other investigators (25). The ability of exogenous TGF-/3 to significantly increase elastin mRNA was limited to pulmonary fibroblasts during the neonatal period when the constitutive synthesis of elastin is high. The ability of exogenous TGF-P to increase soluble elastin in the conditioned medium was also limited to neonatal lung fibroblasts. Cultured cells obtained from neonates tended to secrete more TGF-P than did cells obtained from adults. Endogenous TGF-P appears to affect the quantity of soluble elastin in cultures of neonatal LF and this autocrine regulation could potentially influence the effects of exogenous TGF-/3. When cultures of LF from 5-day-old rats were exposed to 100 pM TGF-@, the quantity of soluble elastin in the culture medium increased by approximately twofold. Newly synthesized TE also increased by approximately twofold, suggesting that TGF-@ increases elastin production by these cells. Elastin mRNA underwent a smaller, 1.6-fold increase, which, although statistically significant,
1
3
4
L261
Fig. 4. A: Northern analysis of RNA isolated from SMC from 3-day-old (lanes l-3), and adult (lanes 4-6) rats. Control SMC (lanes 1 and 4) received no TGF-6, whereas 40 pM (lanes 2 and 5) or 100 pM TGF-R was added to SMC in lanes 3 and 6, respectively. Elastin mRNA was 3.5 kb and hybridized with 32P-labeled cHEL2 (32P-cHEL2). Autoradiogram shown in lanes l-3 was exposed for 9 h while autoradiogram shown in lanes 4-6 was exposed for 40 h. Same filter was reprobed with 32P-labeled cDNA (32P-cDNA) for glyceraldehyde 3-phosphate dehydrogenase (GAPDH), which hybridized with a 1.4-kb mRNA. B: Northern analysis of RNA isolated from lung fibroblasts (LF) from 5-day-old rats (lanes 1 and 2) and adult rats (lanes 3 and 4). RNA used in lanes 1 and 3 was isolated from LF that were not exposed to TGF-6, whereas lanes 2 and 4 contained RNA from LF that were exposed to 100 pM TGF-6. Analysis of elastin and GAPDH mRNA were performed as in A. Autoradiogram shown in lanes 1 and 2 was exposed for 30 h, whereas autoradiogram shown in lanes 3 and 4 was exposed for 120 h.
is near the limits of accurate detection by these methods and the biological significance of this finding is not established. It is possible that TGF-P may have translational or posttranslational effects on elastin production. In other cell culture systems, the ability of TGF-fl to decrease proteinase expression, while increasing antiproteinases, has been well documented (19, 27). TGF-@ appears to influence procollagen stability, as well as its synthesis and could also affect the stability of TE (16). However, the data in Fig. 3 suggest that the proportion of immunoreactive elastin that remains 75 kDa in size for 4 and 6 h after the pulse, and therefore has presumably not undergone proteolysis, is similar in TGF-P-exposed and unexposed cells. Kelley et al. (18) have examined the production of TGF-P by fetal and adult rat lung fibroblasts with an assay that quantitates TGF-P-mediated stimulation of anchorage-independent growth by NRK cells. They found that fibroblasts cultured from adult rats secreted 16.3 & 3.2 ng TGF-/3. lo6 cells-i*24 h-i. Using the radioreceptor assay, we have have detected 20.6 + 8.7 ng TGF-P/lo6 cells, based on the DNA content of rat cells determined by Puzos and Goodman (31) and interpolating from 48 to 24 h (31). Thus, like the data of Kelley and associates (38), our findings suggest that primary cultures of rat cells secrete much larger quantities than &ells obtained from humans or most continuous cell lines. This level of endogenous TGF-@ production (-400 PM) is higher than the concentration of exogenous TGF-& added to cultures of LF (100 PM). Although only 7.5% of the endogenous TGF-& was in the activated form,, it apparently can influence elastin production by LF from 5-day-old rats. The quantity of soluble elastin in the culture medium was significantly reduced by adding an antiTGF-P antibody to LF that had not been supplemented with exogenous TGF-@. Activated TGF-/3 from the endogenous pool may confound the effects of active TGF-@
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
L262
ELASTIN
AND TGF-P PRODUCTION
BY RAT MESENCHYMAL
CELLS
1,200
1,000
**
200 adult
5 day
0
Control
anti-TGF-beta
Fig. 6. Effect of anti-TGF-P neutralizing antibody on quantity of soluble elastin (determined by ELISA) in medium from cultures of &day LF not exposed to exogenous TGF-& Bars, means; error bars, 1 SE. ** P < 0.01.
3 day
Age Fig. 5. Secretion of TGF-P by cultured lung fibroblasts (A) obtained from &day-old (n = 4) or adult (n = 3) rats and by cultured aortic SMC (B) obtained from 3-day-old (n = 3), and adult (n = 3) rats. Conditioned media were collected over 24 h and assayed by analyzing inhibition of 1251-TGF-p binding to NRK fibroblasts. Bars, means; error bars, 1 SE.
from an exogenous source. Differences in the level of endogenous production and activation of TGF-P by our neonatal LF cultures and those of others could explain why other investigators have not observed an increase in elastin m.RNA after supplemen ting cultures of neonatal LF with 40 pM TGF-P (2). Precedent exists for the regulation of other extracellula r matrix constituents in fibrotic diseases by endogenous TGF-P. In experimental models of pulmonary and hepatic fibrosis, increases in TGF-P mRNA and protein have been shown to precede increases in collagen (7, 16). Okuda et al. (26) have used antithymo #cyte serum- treated rats to demonstrate that endogenous TGF-0 modulates proteoglycan produc tion by isolated glomerul i. An antibody against TGF-P1 significantly reduced proteoglycan production by cultured glomeruli and mesangial cells. Autoregulation through TGF-P is not limited to pathological conditions and may be involved in physiological control
of endothelial cell growth and migration. Sato et al. (35) have shown that TGF-P-mediated increases in plasminogen activator inhibitor-l regulate the activation of latent TGF-@ in cocultures of endothelial cells and pericytes. These observations demonstrate the complexity of the effects of TGF-P on extracellular matrix production in vivo and that endogenous production of this cytokine may influence connective tissue protein synthesis at multiple levels (34). The influence of TGF-@ on the production of elastin in the lung remains to be defined. TGF-& is expressed in the lung during the canalicular stage when elastin production is beginning (13, 14). Expression of TGF-& and TGF-& has also been documented during lung development in mice (29). However these observations do not indicate that the TGF-@s are promoting elastin generation. Recent studies of excisional and incisional would healing suggest that exogenously supplied TGF-0, increases elastin gene expression in a subpopulation of fibroblasts in the superficial portion of the dermis (32,33). The exogenous TGF-& also increases TGF-& gene expression in the healing wound, and this endogenously produced TGF-P may perpetuate the formation of granulation tissue. It is possible that endogenous TGF-P may act similarly during fibrosis, remodeling, or repair in the adult lung (4). This work was supported by Grant ROl-HL-45135 from the National Heart, Lung, and Blood Institute and by the Department of Veterans Affairs Research Service. Address for reprint requests: S. E. McGowan, Pulmonary Div., Dept. of Internal Medicine, C33B GH, University of Iowa Hospitals and Clinics, Iowa City, IA 52242. Received 18 October 1991; accepted in final form 4 March 1992. REFERENCES 1. Auffray, C., and F. Rougeon. Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur. J. Biochem. 107: 303-314, 1980. 2. Berk, J. L., C. Franzblau, and R. H. Goldstein. Recombinant interleukin-lp inhibits elastin formation by a neonatal rat
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ELASTIN
AND TGF-fl PRODUCTION
lung fibroblast subtype. J. Biol. Chem. 266: 3192-3197, 1991. Blackburn, M. J., T. M. Andrews, and W. E. Watts. Measurement of DNA in isolated granulocytes by the ethidium technique. AnaL. Biochem. 51: l-10, 1973. 4. Broekelmann, T. J., A. H. Limper, T. V. Colby, and J. A. McDonald. Transforming growth factor-p1 is present at sites of extracellular gene expression in human pulmonary fibrosis. Proc.
BY RAT MESENCHYMAL 20.
3.
Natl.
Acad.
Burri,
Sci.
USA
88: 6642-6646,
1991.
and growth of the human lung. In: Handbook of Physiology. The Respiratory System. Bethesda, MD: Am. Physiol. Sot., 1985, sect. 3, vol. 1, chapt. 1, p. l-46. 6. Campagnone, R., J. Regan, C. B. Rich, M. Miller, D. R. Keene, L. Sakai, and J. A. Foster. Pulmonary fibroblasts: a model system for studying elastin synthesis. Lab. Invest. 56: 2245.
P. H. Development
230, 1987. 7.
8.
9.
10. 11.
Czaja, M. J., F. R. Weiner, K. C. Flanders, M. A. Giambrone, R. Wind, L. Biempica, and M. A. Zern. In vitro and in vivo association of transforming growth factor-p, with hepatic fibrosis. J. CeZZ Biol. 108: 2477-2482, 1989. Davidson, J. M., K. E. Hill, M. L. Mason, and M. G. Giro. Longitudinal gradients of collagen and elastin gene expression in the porcine aorta. J. Biol. Chem. 260: 1901-1908, 1985. Davidson, J. M., and G. C. Sephel. Regulation of elastin synthesis in organ and cell culture. In: Methods in EnzymoLogy, edited by S. P. Colowick and N. 0. Kaplan. San Diego, CA: Academic, 1987, vol. 144, p. 214-232. Floros, J., M. Post, and B. T. Smith. Glucocorticoids affect the synthesis of pulmonary fibroblast-pneumocyte factor at a pretranslational level. J. Biol. Chem. 260: 2265-2267, 1985. Fort, P., L. Marty, M. Piechaczyk, S. E. Sabrouty, C. Dani, P. Jeanteur, and J. M. Blanchard. Various rat adult tissues express only one major mRNA species from the glyceraldehyde-3-phosphate-dehydrogenase multigenic family. Nucleic Acids
Res. 13: 1431-1442,
1985.
1986.
Liu, J., and J. M. Davidson. The elastogenic effect of recombinant transforming growth factor-p on porcine aortic smooth muscle cells. Biochem. Biophys. Res. Commun. 154: 895-901, 1988. 22. McGowan, S. E., and R. McNamer. Transforming growth factor-p increases elastin production by neonatal rat lung fibroblasts. Am. J. Respir. Cell Mol. BioZ. 3: 369-376, 1990. 23. McGowan, S. E., and J. J. Murray. Direct effects of neutrophi1 oxidants on elatase-induced extracellular matrix proteolysis. 21.
Am.
17.
Ther.
246: 765-771,
1988.
Indik, Z., H. Yeh, N. Ornstein-Goldstein, P. Sheppard, N. Anderson, J. C. Rosenbloom, L. Peltonen, and J. Rosenbloom. Alternative splicing of human elastin mRNA indicated by sequence analysis of cloned genomic and complementary DNA. Proc.
Natl.
Acad.
Sci.
USA
84: 5680-5684,
1987.
Kelley, J., J. P. Fabisiak, K. Hawes, and M. Absher. Cytokine signaling in lung: transforming growth factor-p secretion by lung fibroblasts. Am. J. Physiol. 260 (Lung CeZZ. Mol. Physiol. 4): L123-L128, 1991. 19. Laiho, M., 0. Saksela, P. A. Andreasen, and J. Keski-Oja. Enhanced production and extracellular deposition of the endothelial-type plasminogen activator inhibitor in cultured human lung fibroblasts by transforming growth factor-p. J. CeZZ Biol. 103: 18.
2403-2410,
1986.
Rev. Respir.
Dis.
135: 1288-1293,
1987.
Murphy, M. J. 12”1-labelling of erythropoietin without loss of biological activity. Biochem. J. 159: 287-292, 1976. 25. Noguchi, A., R. Reddy, J. D. Kursar, W. C. Parks, and R. P. Mecham. Smooth muscle isoactin and elastin in fetal bovine lung. Exp. Lung Res. 15: 537-552, 1989. 26. Okuda, S., L. R. Languino, E. Rouslahti, and W. A. Border. Elevated expression of transforming growth factor-p and proteoglycan production in experimental glomerulonephritis. J. 24.
CZin. Invest.
86: 453-462,
1990.
Overall, C. M., J. Wrana, and J. Sodek. Independent regulation of collagenase, 72-kDa progelatinase, and metalloproteinase inhibitor expression in human fibroblasts by transforming growth factor-p. J. BioZ. Chem. 264: 1860-1869, 1989. 28. Parks, W. C., H. Secrist, L. C. Wu, and R. P. Mecham. Developmental regulation of tropoelastin isoforms. J. Biol. Chem. 27.
263: 4416-4423, 29.
1988.
Pelton, R. W., M. D. Johnson, E. A. Perkett, L. I. Gold, and H. L. Moses. Expression of transforming growth factor-&, -&, and -p3 mRNA and protein in the murine lung. Am. J. Respir. Cell Mol.
BioZ. 5: 522-530,
1991.
Pierce, R. A., S. B. Deak, C. A. Stolle, and C. D. Boyd. Heterogenity of rat tropoelastin mRNA revealed by cDNA cloning. Biochemistry 29: 9677-9683, 1990. 31. Puzas, E. J., and D. B. P. Goodman. A rapid assay for cellular deoxyribonucleic acid. Anal. Biochem. 86: 50-55, 1978. 32. Quaglino, D., L. B. Nanney, J. A. Ditesheim, and J. M. Davidson. Transforming growth factor-p stimulates wound healing and modulates extracellular matrix gene expression pig skin: incisional wound model. J. Invest. DermatoZ. 97: 34-42, 1991. 33. Quaglino, D., L. B. Nanney, R. Kennedy, and J. M. Davidson. Transforming growth factor-p stimulates wound healing and modulates extracellular matrix gene expression in pig skin. Lab. Invest. 34.
1991.
16. Hoyt, D. G., and J. S. Lazo. Alterations in pulmonary mRNA encoding procollagens, fibronectin, and transforming growth factor-p precede bleomycin-induced pulmonary fibrosis in mice. J Exp.
Like, B., and J. Massague. The antiproliferative effect of type p transforming growth factor occurs at a level distal from receptors for growth-activating factors. J. Biol. Chem. 261: 13426-13429,
30.
12. Franzblau, C., C. A. Pratt, B. Faris, N. M. Colannio, G. D. Offner, P. J. Mogayzel, Jr., and R. F. Troxler. Role of tropoelastin fragmentation in elastogenesis in rat smooth muscle cells. J. Biol. Chem. 264: 15115-15119, 1989. 13. Fukuda, Y., V. J. Ferrans, and R. G. Crystal. The development of alveolar septa in fetal sheep lung. An ultrastructural and immunohistochemical study. Am. J. Anat. 167: 405-439, 1983. 14. Heine, U. I., E. A. Munoz, K. C. Flanders, A. B. Roberts, and M. B. Sporn. Colocalization of TGF-P, and collagen I and III, fibronectin and glycosaminoglycans during lung branching morphogenesis. Development 109: 29-36, 1990. 15. Heino, J., and T. Heinonen. Interleukin-lp prevents the stimulatory effect of transforming growth factor-p on collagen gene expression in human skin fibroblasts. Biochem. J. 271: 827-830,
Pharmacol.
L263
CELLS
63: 307-319,
1990.
Roberts, A. B., K. C. Flanders, U. I. P. Kondiah, S. J. Kim, and M. B. growth factor-p: multifunctional regulator development. Philos. Trans. R. Sot. Lond.
Heine, S. Jakowlew, Sporn. Transforming of differentiation and B Biol.
Sci. 327: 145-
154, 1990.
Sato, Y., R. Tsuboi, R. Lyons, H. Moses, and D. B. Rifkin. Characterization of the activation of latent TGF-P by cocultures of endothelial cells and pericytes or smooth muscle cells: a selfregulating system. J. CeZl Biol. 111: 757-763, 1990. G. T. Hermanson, A. K. Mallia, 36. Smith, P. K., R. I. Krohn, F. H. Gartner, M. D. Provenzano, E. K. Fujimoto, N. M. Goeke, B. J. Olson, and D. C. Klenk. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150: 76-85, 1985. 37. Van-Obberghen Schilling, E., N. S. Roche, K. C. Flanders, M. B. Sporn, and A. B. Roberts. Transforming growth factor-& positively regulates its own expression in normal and transformed cells. J. Biol. Chem. 263: 7741-7746, 1988. 38. Wakefield, L. M., D. M. Smith, T. Masui, C. C. Harris, and M. B. Sporn. Distribution and modulation of the cellular receptor for transforming growth factor-p. J. Cell BioZ. 105: 965-975, 1987. 35.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
et al. (14) suggest that transforming growth factor (TGF)-P, colocalizes with collections of type III collagen in the clefts between branching ducts during the canalicular stage. They suggest that TGF-& may promote collagen formation in these clefts. Morphological studies by Fukuda et al. (13) indicate that elastin formation coincides with collagen formation in these cleft regions. However, there are currently no data available that directly relate TGF-P and elastin formation in the developing lung. Investigations by Liu and Davidson (21) and in our laboratory (22) have shown that supplementing cell cultures with TGF-& promotes elastin synthesis in neonatal porcine smooth muscle cells (SMC) and neonatal rat lung fibroblasts (LF) in vitro. These studies prompted three questions: 1) Is this TGF-P-mediated increase in elastin synthesis limited to a period of abundant constitutive elastin synthesis, as occurs in neonates, or can TGF-P also increase elastin synthesis in cells derived from adults when constitutive elastin synthesis is low? 2) Do primary cultures of LF and SMC obtained from neonates differ from those obtained from adults in their production of TGF-P? and 3) Does endogenous TGF-& influence elastin production by cultured neonatal cells? To address these questions, we have studied elastin and TGF-P in primary cultures of LF and SMC from neonatal and adult rats and have examined the alterations in soluble elastin and elastin mRNA that result from adding exogenous TGF-& to cultures of LF and SMC obtained from rats of these ages. METHODS CeZZ culture of rat LF and S2MC.After anesthetization with ketamine, the lungswere removed from Sprague-Dawleyrats at 5 days after birth or from 150-g adults, washedwith sterile Dulbecco’sphosphate-bufferedsaline (PBS), and then minced into 1-2 mm3 pieces. The tissue was incubated at 37°C with modified Hanks’ balanced salt solution (mHBSS) containing 0.5 mg of trypsin, 0.3 mg of collagenase(Sigma Chemical, type I), and 0.2 mg EDTA/ml. After 20 and 40 min, the mHBSS was removed, and the tissue was combined with Ham’s F-12 medium containing 10% fetal bovine serum (FBS), fresh mHBSS was added, and the incubation was continued for a total of 60 min. The mHBSS, containing the digestedtissue,was strained through a 150-pmfilter, and the cells that passedthrough the filter were seeededat a density of 1 x IO7 tells/75-cm2 tissue culture flask. After 1 h, the nonadherent cellswere removed,and the adherent cells were cultured in Ham’s F-12 medium containing 10% FBS, 100 U of penicillin, and 100 pg/ml of streptomycin. Removing nonadherent cells after 1 h provided a cell population that wasprimarily fibroblasts (6, 10). The cellswere maintained by these culture conditions for 8 days and then subcultured into 25-cm2flasks or 2.0-cm2wells at a density of 3 x lo* cells/cm2.The subculturedfibroblasts reachedconfluence in 3-4 days and were maintained in medium containing 10% FBS for a total of 8 days and then in medium containing 5% L257
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
L258
ELASTIN AND TGF-@ PRODUCTION
newborn calf serum and IO mM N-2-hydroxyethylpiperazineN’-2-ethanesulfonic acid (HEPES) for an additional 5-7 days. SMC were obtained from the aortas of 3-day-old rats after digestion with elastase and collagenase with a technique that has been described previously (23). Adult aortic SMC were cultured from explants that had been removed from 150-g rats. The rats were anesthetized with ketamine, and the thoracic and abdominal cavities were entered. After exsanguination, the aorta was removed in entirety, the fascia and associated tissues were removed, and the aorta was washed with PBS. A longitudinal incision was made to expose the intimal layer. The intimal and medial layers were separated from the adventia with a forceps and then minced. The l-mm3 pieces of the media were allowed to adhere to tissue culture dishes and then supplemented with Dulbecco’s minimum essential medium (MEM) supplemented with 20% FBS, 1 mM sodium pyruvate, MEM nonessential amino acids, 100 U/ml of penicillin, and 100 pg/ml of streptomycin. After lo-14 days the cells were subcultured as described previously (23). The production of elastin and TGF-P was analyzed using cells after one or two subcultivations. Quantitation of soluble elastin in the culture medium. Prior to quantitating elastin in triplicate wellsfor eachcondition, the LF were washedtwice with PBS and placed in medium containing 0.5% FBS and 25 pg ,&aminopropionitrile (BAPN)/ml for 24 h, and the SMC were placed in medium containing 5% newborn calf serum and BAPN. BAPN prevents cross-linking of tropoelastin (TE) and thereby facilitates the detection of soluble elastin in the culture medium. During the final 12 h of this period, some cultures remained unsupplementedwhile others were supplementedto a final concentration of 100 pM TGF-@ (TGF-P, from porcine platelets; R and D Systems,Minneapolis, MN). This concentration was previously shown to maximally increaseTE production by neonatal LF (22). The culture media were then removed, and fresh medium containing BAPN, with or without 100pM TGF-& was added,and the cells were incubated for an additional 24 h. The conditioned mediumthat was collectedafter this final 24 h period wasremoved,cooledto 4”C, centrifuged at 500 g for IO min, and the supernatants were stored at -70°C. The residualcell layers werewashedwith PBS and removed from the wells by scrapingand were sonicatedfor 45 s to achieve a homogeneousmixture. The sonicated cell layers were supplementedwith 0.2 M sodium acetate, pH 7.0, 1% aminonitrillo-sulfonic acid, and the protein wasremoved by extracting with phenol. Aliquots of the aqueousphase were treated with proteinaseE (Sigma,St. Louis, MO), and the DNA content was determined by quantitating the fluorescent intensity of the DNA-ethidium bromide complex, which was sensitive to deoxyribonycleaseI digestion (3). The solubleelastin in the conditioned mediawasquantitated usinga competitive enzyme-linked immunoassay(ELISA) similar to that describedby Davidson and Sephel (9). Immulon 2 microtiter plateswere coatedwith oxalic acid-solubilizedelastin (62.5 rig/well) that had been isolatedfrom adult rat aortas. The competing antigen was a mixture of TE moieties, which were isolatedfrom the conditioned mediumcollectedover 24 h, in the presenceof BAPN, from neonatal rat aortic SMC that had been subcultured 2 wk previously (12). The TE moieties coeluted from a Vydac C,, column. An aliquot was subjectedto sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) followed by either silver staining or immunoblotting, using goat anti-rat cw-elastinas the first antibody. The major immunoreactive and silver-stained bandsmigrated at 69 and 59 kDa, although other bands were evident. The protein concentration was determined using bicinchoninic acid (36). To generate a standard curve for the ELISA, serial dilutions of the competing antigen (0.8-12.2 ng) were mixed with goat anti-rat cu-elastin(Elastin Products, Pacific, MO) and incubated overnight at 4°C. The antigen-antibody mixture wasthen incubated with the a-elastin-coated plate under nonequilibrium condi-
BY RAT MESENCHYMAL
CELLS
tions. The free antibody that bound to the plate wasquantitated with peroxidase-labeledrabbit anti-goat immunoglobulin G UgG)
l
Because the ELISA detects other soluble elastin moieties besides75-kDa TE, cultures obtained from 5-day LF that had been exposedto 100 pM TGF-P for 24 h or had remained unexposedwere metabolically labeledwith [3H]proline. Twentyfour hours prior to labeling, the serum concentration of the culture medium was reducedto 0.5%, and BAPN was addedto achieve a final concentration of 25 pg/ml. After this 24 h preincubation, the culture medium was removed, the cells were washedthree times with PBS, and Dulbecco’sMEM (without serumor proline) was addedfor 30 min. This mediumwasthen changedto the samemediumexcept it now contained 30 &i/ml Z-[2,3,4,5,3H]proline, and the pulsewascontinued for 1 h. After removing the pulse medium, the cells were washed and then incubated in medium containing 1 mM proline for 1, 2, 4, or 6 h. The culture media and cell layers were collected and processedas describedpreviously (22). Newly labeledTE moieties in the culture media were immunoprecipitated with goat antirat cu-elastinfollowed by Staphlococcus aureus protein G bound to agarose(Gamma-bind G; Genex, Gaithersburg, MD). After washing,the TE moietieswereeluted from the agarosebeadsin a SDS-containing buffer and subjectedto SDS-PAGE (22). Autoradiograms were obtained as described and scannedwith a Shimadzu CS-9OOOUdensitometer. Analyzing elastin mRNA. The cell layers from four 25-cm2 flasksthat had beenexposedor remainedunexposedto 100pM TGF-P for 24 h were combined, and the RNA was isolated (1, 22). Northern analyseswere performed aspreviously described using 1.2% agarosegelscontaining 6% formaldehyde (22). After electrophoresis,the gels were stained with ethidium bromide, destained,and photographedto determinethe distancesthe 28s and 18s ribosomalRNA had migrated and to assurethat equal amountsof RNA had been loaded.The RNA wastransferred to a cationic nylon membrane (Nytran, Schleicher and Schuell, Keene, NH), which was prehybridized and then hybridized by incubating for 16 h at 42°C with 32P-labeledcHEL2 (32PcHEL2). The l.2-kb cDNA, cHEL2, was originally obtained from Dr. Joel Rosenbloom(University of Pennsylvania) and wasderived from the 5’ end of the human elastin gene(17). The cHEL2 was labeled with [a- 32P]dCTP (3,000 Ci/mmol Du Pont-New England Nuclear, Wilmington, DE) using the random primer technique (22). After hybridization and washing, the membranewas exposedto Kodak XOMAT-AR film using an intensifying screen.Exposure times were varied to achieve bandswith densitiesthat werewithin the detection limits of the film, and the resultant autoradiogramswere subjectedto densitometry. After removal of the 32P-cHEL2 and prehybridization, the membraneswere hybridized with 32P-labeledglyceraldehyde 3-phosphate dehydrogenase( [32P]cGAPDH) , a nearly full-length cDNA for rat glyceraldehyde3-phosphate dehydrogenase(11, 22). The plasmid containing this cDNA was a gift from Dr. Alan Goodridge (University of Iowa). The gene for GAPDH is not induced by TGF-P (38) and thus served as a control to normalize for differences in the quantities of RNA that were loadedin the various lanesof the gelsusedfor Northern analysis. Quantitating the biological activity of TGF-p in conditioned media. Neonatal and adult rat LF and SMC were cultured in 5% FBS in 9.3-cm2wells for 14 days after the first subcultivation.
At that time, the medium was removed, the cells were washed with PBS, and medium without serum (HBlOl) was added. This serum-freemedium was conditioned by the cells for 48 h, chilled to 4”C, and centrifuged to remove nonadherent cells,and the supernatant was stored sterilely at 4°C. The TGF-P content of someof the culture media was analyzed without acidification. Transient (30 min) acidification to pH 3.0 dissociatesTGF-@ from its latency-associatedpeptide
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ELASTIN
AND
TGF-p
PRODUCTION
and assures that all of the TGF-P in the medium is biologically active at the time the assay is performed. However, it does not indicate what portion of the TGF-P was biologically active during the cell culture period. The TGF-@ activity detected in media that were not exposed to acid was at the lower limits of detection using a bioassay in which TGF-P inhibits mink lung cell growth (20). The unacidified samples contained only 7.5 t 3.6% (mean t SE, n = 11) of the TGF-P activity that wasfound in the respective acidified samplesfrom the growth-inhibition assay.Therefore, only the data obtained from acidified samples are presented. A receptor binding (radioreceptor) assaywas usedto quantitate TGF-/? (38). This assayrelies on the ability of proteins in the conditioned mediumto competewith TGF-P for binding to specific receptorsfor TGF-P. Therefore, it is more specific than growth inhibition or stimulation assaysas it dependson one biological event, namely receptor binding, rather than a cascade of events, including receptor binding, signal transduction, and ultimately DNA synthesis. Cells from a rat kidney fibroblast line (NRK-49F) were subcultured in a 24-well plate and maintained for 24 h. 1251-labeled TGF-P ( 1251-TGF-P)wasprepared using glucose oxidase and lactoperoxidase and carrier-free TGF-& from R & D Systems (24). TGF-@ (2 pg) was dissolved in 20 ~1of 4 mM HCl and supplementedwith 10 ~1 of 0.3 M sodiumphosphate,pH 7.2. Glucoseoxidaseand lactoperoxidase (50 ng, both) and 1 mCi of Na 1251 were added.The reaction was initiated by adding 100pg of @-D-&CO%? and terminated after 5 min by adding 10 ~1 of saturated tyrosine, 100 ~1 of 60 mM potassiumiodide, and 150 ~1of 8 M urea. The free Na1251was removed by elution from a PD-10 gel filtration column (Pharmacia, Piscataway, NJ) in 4 mM HCl, 0.075 M NaCl, 0.1% bovine serumalbumin. The 1251-TGF-@ had a specific radioactivity of -200 &i/pg and wasstored in aliquots at -20°C. The cell layers were incubated with various dilutions of conditioned mediumfor 2 h at 24°C washedand then incubated for 2 h with 1251-TGF-pThe cellsin somewellswereincubated with various concentrations (lo-200 pM) of nonradioactive TGF-P to compete with the 1251-TGF-pfor binding to receptors. After washing to remove unbound 1251-TGF-&, the bound 1251was solubilized and an aliquot was subjectedto gamma spectrometry. Nonspecific binding (1251-bindingin the presenceof a 300-fold excessof recombinant human TGF-& suppliedby Genentech) was subtracted from the total 1251bound for all wells and comprised 30-35% of the total 1251bound. Statistical analyses. The data are expressedas meansk SE. Comparisonsamong cultures that were exposed to different concentrations of TGF-P were madeby performing an analysis of variance. The multivariate generallinear hypothesis module of the SYSTAT (SYSTAT, Evanston IL) program was used. Where appropriate, Student’s t test for unpaired variables was used.Differences were consideredsignificant when P < 0.05. RESULTS
Effect of exogenous TGF-6 on soluble elastin production by LF and SMC. The first objective was to examine the
ability of exogenous TGF-@ to increase elastin production in cells obtained from rats at times when constitutive elastin synthesis was high (neonates) or low (adults). For each cohort of isolated cells, elastin production by cells exposed to 100 pM TGF-/3 was compared with control cells from the same primary isolation, which were not exposed to TGF-P. The quantity of soluble elastin in the media of control cells in the absence of exogenous TGF-p was 348.9 t 98.5 ng/pg of DNA (mean t SE, n = 3 experiments) during 24 h for LF obtained from 5-day-old rats and 55.9 t 8.2 ng/pg DNA (n = 3) for LF from adult rats (P < 0.01). The quantity of soluble elastin in the
BY RAT
MESENCHYMAL
L259
CELLS
media of control SMC was 3,721.l t 1,298.0 ng pg DNA-I. 24 h-l (n = 3) for SMC obtained at 3 days and 15.02 t 2.7 ngapg DNA-l.24 h-l (n = 3) for cells obtained from adult rats (P < 0.05). When cultures of LF from 5-day-old rats were made 100 pM in TGF-& a 2.31 t 0.31 fold increase (P c 0.01) in the soluble elastin content of the culture medium was observed compared with that in control cultures, which had been obtained from the same animals but had not received TGF-@ (Fig. 1). In contrast, TGF-@ did not significantly alter the soluble elastin content in the culture medium of LF, which were obtained from adult animals (P = 0.086). Adding TGF-P to cultures of SMC, which were obtained from either 3-day-old or adult rats, did not significantly alter the amount of soluble elastin in the culture medium (Fig. 2) *Because the antielastin antibody that was used in the ELISA assay detects soluble elastin moieties other than tropoelastin (TE), the ELISA does not specifically quantitate intact 75-kDa TE. Therefore, TE synthesis was also evaluated by pulse labeling LF from 5-day-old rats with [3H]proline. The results are shown in Fig. 3 and demonstrate that exposing cells to TGF-P is associated with a 2.0- and a U-fold increase in the quantity of 75-kDa TE present in the culture medium after chasing for 1 and 2 h, respectively. These data suggest that the TGF-P-mediated increase in soluble elastin detectable by the ELISA, at least partially, results from an increase in elastin synthesis. l
**
5 day
Age
q
protein
q
mRNA
Fig. I. Effects of exposure to transforming growth factor (TGF)-P on quantity of soluble elastin in medium (hatched bars) and elastin (stippled bars) mRNA in cultures of lung fibroblasts (LF) from &day-old and adult rats. Soluble elastin content of culture medium was quantitated by enzyme-linked immunoassay (ELISA, see text). Increase in soluble elastin and elastin mRNA in cultures of LF exposed to 100 pM TGF-P are expressed relative to those of LF obtained from same rats but that were not exposed to exogenous TGF-P (control, dashed line). Filters that were used for Northern analysis were first probed with a cDNA for elastin cHEL2 and then reprobed with a cDNA for glyceraldehyde 3-phosphate dehydrogenase. Autoradiograms were subjected to densitometry, and density of band representing elastin mRNA was normalized to density of corresponding band for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA for each sample. Ratios of density of elastin mRNA to GAPDH mRNA for samples from flasks that had been exposed to 100 pM TGF-P were expressed as fold increase from ratio for control samples that had not been exposed to TGF-0. Bars, mean (n = 3 sets of cultures from 3 separate animals for all groups, except n = 8 for mRNA from 5-day-LF); error bars, 1 SE. * P < 0.05, ** P < 0.01.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
L260
ELASTIN
AND TGF-P PRODUCTION
BY RAT MESENCHYMAL
ABCD
4
CELLS
EFGH
220
97
3 day
Adult Age m
protein
c]
mRNA
Fig. 2. Effects of exposure to TGF-P on quantity of soluble elastin in medium (hatched bars) and elastin mRNA (stippled bars) in cultures of aortic smooth muscle cells @MC) from 3-day-old and adult rats. Soluble elastin content of culture medium was quantitated by ELISA (see text). Increases in soluble elastin and elastin mRNA in cultures of SMC exposed to 100 pM TGF-fl are expressed relative to those by LF obtained from same rats but that were not exposed to exogenous TGF-0 (control, dashed line). Northern analyses were performed as described in Fig. 1 and in text. Bars, mean (n = 3 sets of cultures from 3 separate animals for all groups, except n = 4 for mRNA from adult SMC); error bars, 1 SE.
Effect of TGF-/I on elastin mRNA in cultures of LF and SMC. Northern analyses were performed to compare elastin expression in neonates and adults and to determine whether treatment with TGF-P increased elastin mRNA in LF and SMC. Figure 4 demonstrates that the steadystate levels of elastin mRNA in rat SMC (A) and LF (B) that were not exposed to exogenous TGF-P (lanes 1 and 4 in A; lanes 1 and 3 in B) are lower in adult animals than in neonates. The nature of the larger RNA (5.1 kb) has not been characterized, although it is larger than those attributable to alternative splicing (30). Multiple Northern analyses were performed using RNA from cells of the various ages to determine whether exposure to exogenous TGF-P increases steady-state elastin mRNA. Figure 4 also shows representative Northern analyses of RNA isolated from TGF-P-exposed SMC (lanes 2,3,5, and 6 in A) and LF (lanes 2 and 4 in B). Multiple analyses were performed to compare the steady-state elastin mRNA in LF that were exposed to 100 pM TGF-P with that of contol cells isolated from the same rats but that were not exposed to TGF-P. In Figs. 1 and 2, the ratios of the density of elastin mRNA to GAPDH mRNA for samples from flasks that had been exposed to TGF-P have been expressed as the fold increase from the ratio for control samples that had not been exposed to TGF-P. A mean, 1.6-fold (n = 8, P < 0.05) increase in elastin mRNA was observed in 5-day LF exposed to TGF-P (Fig. l), whereas TGF-P exposure did not significantly increase the steadystate level of elastin mRNA in cultures of lung fibroblasts from adult rats. Northern analyses of RNA that was isolated from SMC that had been exposed to TGF-@ or remained unexposed were compared in a similar fashion. A significant increase in elastin mRNA was not observed after TGF-P treatment of SMC obtained from rats at either age (Fig. 2).
66
43
Fig. 3. Effects of TGF-fi on incorporation of [“Hlproline into newly svnthesized tropoelastin (TE) in cultures of LF from 5dav-old rats after a l-h exposure to [3H]proline and then incubating with unlabeled proline. One group (lanes B, D, F, and ?Z) of flasks received TGF-@ during 24 h immediately preceeding pulse, and a second group (lanes A, C, E, and G) did not receive TGF-P. All flasks were incubated for 1 h with [3H]proline and were incubated for an additional period in medium not containing [3H]proline, which lasted for either 1 h (lanes A and B), 2 h (lanes C and D), 4 h (lanes E and F), or 6 h (lanes G and H). Media were immunoprecipitated with an anti-rat cu-elastin antibody prior to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), autoradiography, and densitometry.
TGF-p secretion by cultured rat LF and SMC. The biological activity of TGF-/3 was assayed in the acidtreated conditioned media from rat LF and SMC that were isolated from neonatal and adult animals. The radioreceptor assay demonstrated that the secretion of TGF-@ was 14.76 + 2.09 and 6.35 + 2.69 ng/pg DNA during 48 h for lung fibroblasts obtained from 5-day-old and adult rats, respectively (Fig. 5A). There was a trend toward higher secretion by neonatal cells (P = 0.06). SMC that were isolated from 3-day-old rats also trended toward greater secretion, 28.18 + 8.10 and 7.75 f 3.02 ng TGF-/3/pg DNA over 48 h compared with adult rats (P = 0.06; Fig. 5B). Effect of endogenous TGF-p on soluble elastin production by neonatal LF. These data demonstrate that LF and SMC obtained from neonatal rats secrete considerable quantities of TGF-fi. Neonatal lung fibroblasts endogenously secrete -15 ng of TGF-P/2-cm2 well (containing - 1 pg DNA and 1 ml medium), which is approximately sixfold more than the 2.5 ng exogenous TGF-@ (100 PM) that was shown to stimulate soluble elastin production. These findings suggest that TGF-fi, which was endogenously produced by LF, could potentially stimulate soluble elastin production by these cells. To test this
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ELASTIN
AND TGF-6’ PRODUCTION
BY RAT MESENCHYMAL
CELLS
./ .f(
1.4 kb
123
455
hypothesis, the effect of a polyclonal neutralizing antibody against TGIF-P on soluble elastin production by neonatal LF was examined. The culture media, collected over 24 h, from LF of 5-day-old rats contained 967.8 + 105.73 ng soluble elastin/pg DNA in the absence of rabbit antiTGF-@ (n = 4 experiments, each with 3 wells/group), whereas the media contained 682.6 + 107.2 ng in the presence of the antibody (see Fig. 6). This decrease was significant (P < 0.01) and suggests that endogenous TGF-fi influences the elaboration of soluble elastin by neonatal LF. Adding an irrelevant antibody, rabbit antimouse IgG, did not alter the soluble elastin content of the culture medium (data not shown). DISCUSSION
These studies demonstrate that, in cells cultured from rat lungs and aortas, elastin mRNA is higher in neonates than in adults. These findings are similar to those of other investigators (25). The ability of exogenous TGF-/3 to significantly increase elastin mRNA was limited to pulmonary fibroblasts during the neonatal period when the constitutive synthesis of elastin is high. The ability of exogenous TGF-P to increase soluble elastin in the conditioned medium was also limited to neonatal lung fibroblasts. Cultured cells obtained from neonates tended to secrete more TGF-P than did cells obtained from adults. Endogenous TGF-P appears to affect the quantity of soluble elastin in cultures of neonatal LF and this autocrine regulation could potentially influence the effects of exogenous TGF-/3. When cultures of LF from 5-day-old rats were exposed to 100 pM TGF-@, the quantity of soluble elastin in the culture medium increased by approximately twofold. Newly synthesized TE also increased by approximately twofold, suggesting that TGF-@ increases elastin production by these cells. Elastin mRNA underwent a smaller, 1.6-fold increase, which, although statistically significant,
1
3
4
L261
Fig. 4. A: Northern analysis of RNA isolated from SMC from 3-day-old (lanes l-3), and adult (lanes 4-6) rats. Control SMC (lanes 1 and 4) received no TGF-6, whereas 40 pM (lanes 2 and 5) or 100 pM TGF-R was added to SMC in lanes 3 and 6, respectively. Elastin mRNA was 3.5 kb and hybridized with 32P-labeled cHEL2 (32P-cHEL2). Autoradiogram shown in lanes l-3 was exposed for 9 h while autoradiogram shown in lanes 4-6 was exposed for 40 h. Same filter was reprobed with 32P-labeled cDNA (32P-cDNA) for glyceraldehyde 3-phosphate dehydrogenase (GAPDH), which hybridized with a 1.4-kb mRNA. B: Northern analysis of RNA isolated from lung fibroblasts (LF) from 5-day-old rats (lanes 1 and 2) and adult rats (lanes 3 and 4). RNA used in lanes 1 and 3 was isolated from LF that were not exposed to TGF-6, whereas lanes 2 and 4 contained RNA from LF that were exposed to 100 pM TGF-6. Analysis of elastin and GAPDH mRNA were performed as in A. Autoradiogram shown in lanes 1 and 2 was exposed for 30 h, whereas autoradiogram shown in lanes 3 and 4 was exposed for 120 h.
is near the limits of accurate detection by these methods and the biological significance of this finding is not established. It is possible that TGF-P may have translational or posttranslational effects on elastin production. In other cell culture systems, the ability of TGF-fl to decrease proteinase expression, while increasing antiproteinases, has been well documented (19, 27). TGF-@ appears to influence procollagen stability, as well as its synthesis and could also affect the stability of TE (16). However, the data in Fig. 3 suggest that the proportion of immunoreactive elastin that remains 75 kDa in size for 4 and 6 h after the pulse, and therefore has presumably not undergone proteolysis, is similar in TGF-P-exposed and unexposed cells. Kelley et al. (18) have examined the production of TGF-P by fetal and adult rat lung fibroblasts with an assay that quantitates TGF-P-mediated stimulation of anchorage-independent growth by NRK cells. They found that fibroblasts cultured from adult rats secreted 16.3 & 3.2 ng TGF-/3. lo6 cells-i*24 h-i. Using the radioreceptor assay, we have have detected 20.6 + 8.7 ng TGF-P/lo6 cells, based on the DNA content of rat cells determined by Puzos and Goodman (31) and interpolating from 48 to 24 h (31). Thus, like the data of Kelley and associates (38), our findings suggest that primary cultures of rat cells secrete much larger quantities than &ells obtained from humans or most continuous cell lines. This level of endogenous TGF-@ production (-400 PM) is higher than the concentration of exogenous TGF-& added to cultures of LF (100 PM). Although only 7.5% of the endogenous TGF-& was in the activated form,, it apparently can influence elastin production by LF from 5-day-old rats. The quantity of soluble elastin in the culture medium was significantly reduced by adding an antiTGF-P antibody to LF that had not been supplemented with exogenous TGF-@. Activated TGF-/3 from the endogenous pool may confound the effects of active TGF-@
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
L262
ELASTIN
AND TGF-P PRODUCTION
BY RAT MESENCHYMAL
CELLS
1,200
1,000
**
200 adult
5 day
0
Control
anti-TGF-beta
Fig. 6. Effect of anti-TGF-P neutralizing antibody on quantity of soluble elastin (determined by ELISA) in medium from cultures of &day LF not exposed to exogenous TGF-& Bars, means; error bars, 1 SE. ** P < 0.01.
3 day
Age Fig. 5. Secretion of TGF-P by cultured lung fibroblasts (A) obtained from &day-old (n = 4) or adult (n = 3) rats and by cultured aortic SMC (B) obtained from 3-day-old (n = 3), and adult (n = 3) rats. Conditioned media were collected over 24 h and assayed by analyzing inhibition of 1251-TGF-p binding to NRK fibroblasts. Bars, means; error bars, 1 SE.
from an exogenous source. Differences in the level of endogenous production and activation of TGF-P by our neonatal LF cultures and those of others could explain why other investigators have not observed an increase in elastin m.RNA after supplemen ting cultures of neonatal LF with 40 pM TGF-P (2). Precedent exists for the regulation of other extracellula r matrix constituents in fibrotic diseases by endogenous TGF-P. In experimental models of pulmonary and hepatic fibrosis, increases in TGF-P mRNA and protein have been shown to precede increases in collagen (7, 16). Okuda et al. (26) have used antithymo #cyte serum- treated rats to demonstrate that endogenous TGF-0 modulates proteoglycan produc tion by isolated glomerul i. An antibody against TGF-P1 significantly reduced proteoglycan production by cultured glomeruli and mesangial cells. Autoregulation through TGF-P is not limited to pathological conditions and may be involved in physiological control
of endothelial cell growth and migration. Sato et al. (35) have shown that TGF-P-mediated increases in plasminogen activator inhibitor-l regulate the activation of latent TGF-@ in cocultures of endothelial cells and pericytes. These observations demonstrate the complexity of the effects of TGF-P on extracellular matrix production in vivo and that endogenous production of this cytokine may influence connective tissue protein synthesis at multiple levels (34). The influence of TGF-@ on the production of elastin in the lung remains to be defined. TGF-& is expressed in the lung during the canalicular stage when elastin production is beginning (13, 14). Expression of TGF-& and TGF-& has also been documented during lung development in mice (29). However these observations do not indicate that the TGF-@s are promoting elastin generation. Recent studies of excisional and incisional would healing suggest that exogenously supplied TGF-0, increases elastin gene expression in a subpopulation of fibroblasts in the superficial portion of the dermis (32,33). The exogenous TGF-& also increases TGF-& gene expression in the healing wound, and this endogenously produced TGF-P may perpetuate the formation of granulation tissue. It is possible that endogenous TGF-P may act similarly during fibrosis, remodeling, or repair in the adult lung (4). This work was supported by Grant ROl-HL-45135 from the National Heart, Lung, and Blood Institute and by the Department of Veterans Affairs Research Service. Address for reprint requests: S. E. McGowan, Pulmonary Div., Dept. of Internal Medicine, C33B GH, University of Iowa Hospitals and Clinics, Iowa City, IA 52242. Received 18 October 1991; accepted in final form 4 March 1992. REFERENCES 1. Auffray, C., and F. Rougeon. Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur. J. Biochem. 107: 303-314, 1980. 2. Berk, J. L., C. Franzblau, and R. H. Goldstein. Recombinant interleukin-lp inhibits elastin formation by a neonatal rat
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ELASTIN
AND TGF-fl PRODUCTION
lung fibroblast subtype. J. Biol. Chem. 266: 3192-3197, 1991. Blackburn, M. J., T. M. Andrews, and W. E. Watts. Measurement of DNA in isolated granulocytes by the ethidium technique. AnaL. Biochem. 51: l-10, 1973. 4. Broekelmann, T. J., A. H. Limper, T. V. Colby, and J. A. McDonald. Transforming growth factor-p1 is present at sites of extracellular gene expression in human pulmonary fibrosis. Proc.
BY RAT MESENCHYMAL 20.
3.
Natl.
Acad.
Burri,
Sci.
USA
88: 6642-6646,
1991.
and growth of the human lung. In: Handbook of Physiology. The Respiratory System. Bethesda, MD: Am. Physiol. Sot., 1985, sect. 3, vol. 1, chapt. 1, p. l-46. 6. Campagnone, R., J. Regan, C. B. Rich, M. Miller, D. R. Keene, L. Sakai, and J. A. Foster. Pulmonary fibroblasts: a model system for studying elastin synthesis. Lab. Invest. 56: 2245.
P. H. Development
230, 1987. 7.
8.
9.
10. 11.
Czaja, M. J., F. R. Weiner, K. C. Flanders, M. A. Giambrone, R. Wind, L. Biempica, and M. A. Zern. In vitro and in vivo association of transforming growth factor-p, with hepatic fibrosis. J. CeZZ Biol. 108: 2477-2482, 1989. Davidson, J. M., K. E. Hill, M. L. Mason, and M. G. Giro. Longitudinal gradients of collagen and elastin gene expression in the porcine aorta. J. Biol. Chem. 260: 1901-1908, 1985. Davidson, J. M., and G. C. Sephel. Regulation of elastin synthesis in organ and cell culture. In: Methods in EnzymoLogy, edited by S. P. Colowick and N. 0. Kaplan. San Diego, CA: Academic, 1987, vol. 144, p. 214-232. Floros, J., M. Post, and B. T. Smith. Glucocorticoids affect the synthesis of pulmonary fibroblast-pneumocyte factor at a pretranslational level. J. Biol. Chem. 260: 2265-2267, 1985. Fort, P., L. Marty, M. Piechaczyk, S. E. Sabrouty, C. Dani, P. Jeanteur, and J. M. Blanchard. Various rat adult tissues express only one major mRNA species from the glyceraldehyde-3-phosphate-dehydrogenase multigenic family. Nucleic Acids
Res. 13: 1431-1442,
1985.
1986.
Liu, J., and J. M. Davidson. The elastogenic effect of recombinant transforming growth factor-p on porcine aortic smooth muscle cells. Biochem. Biophys. Res. Commun. 154: 895-901, 1988. 22. McGowan, S. E., and R. McNamer. Transforming growth factor-p increases elastin production by neonatal rat lung fibroblasts. Am. J. Respir. Cell Mol. BioZ. 3: 369-376, 1990. 23. McGowan, S. E., and J. J. Murray. Direct effects of neutrophi1 oxidants on elatase-induced extracellular matrix proteolysis. 21.
Am.
17.
Ther.
246: 765-771,
1988.
Indik, Z., H. Yeh, N. Ornstein-Goldstein, P. Sheppard, N. Anderson, J. C. Rosenbloom, L. Peltonen, and J. Rosenbloom. Alternative splicing of human elastin mRNA indicated by sequence analysis of cloned genomic and complementary DNA. Proc.
Natl.
Acad.
Sci.
USA
84: 5680-5684,
1987.
Kelley, J., J. P. Fabisiak, K. Hawes, and M. Absher. Cytokine signaling in lung: transforming growth factor-p secretion by lung fibroblasts. Am. J. Physiol. 260 (Lung CeZZ. Mol. Physiol. 4): L123-L128, 1991. 19. Laiho, M., 0. Saksela, P. A. Andreasen, and J. Keski-Oja. Enhanced production and extracellular deposition of the endothelial-type plasminogen activator inhibitor in cultured human lung fibroblasts by transforming growth factor-p. J. CeZZ Biol. 103: 18.
2403-2410,
1986.
Rev. Respir.
Dis.
135: 1288-1293,
1987.
Murphy, M. J. 12”1-labelling of erythropoietin without loss of biological activity. Biochem. J. 159: 287-292, 1976. 25. Noguchi, A., R. Reddy, J. D. Kursar, W. C. Parks, and R. P. Mecham. Smooth muscle isoactin and elastin in fetal bovine lung. Exp. Lung Res. 15: 537-552, 1989. 26. Okuda, S., L. R. Languino, E. Rouslahti, and W. A. Border. Elevated expression of transforming growth factor-p and proteoglycan production in experimental glomerulonephritis. J. 24.
CZin. Invest.
86: 453-462,
1990.
Overall, C. M., J. Wrana, and J. Sodek. Independent regulation of collagenase, 72-kDa progelatinase, and metalloproteinase inhibitor expression in human fibroblasts by transforming growth factor-p. J. BioZ. Chem. 264: 1860-1869, 1989. 28. Parks, W. C., H. Secrist, L. C. Wu, and R. P. Mecham. Developmental regulation of tropoelastin isoforms. J. Biol. Chem. 27.
263: 4416-4423, 29.
1988.
Pelton, R. W., M. D. Johnson, E. A. Perkett, L. I. Gold, and H. L. Moses. Expression of transforming growth factor-&, -&, and -p3 mRNA and protein in the murine lung. Am. J. Respir. Cell Mol.
BioZ. 5: 522-530,
1991.
Pierce, R. A., S. B. Deak, C. A. Stolle, and C. D. Boyd. Heterogenity of rat tropoelastin mRNA revealed by cDNA cloning. Biochemistry 29: 9677-9683, 1990. 31. Puzas, E. J., and D. B. P. Goodman. A rapid assay for cellular deoxyribonucleic acid. Anal. Biochem. 86: 50-55, 1978. 32. Quaglino, D., L. B. Nanney, J. A. Ditesheim, and J. M. Davidson. Transforming growth factor-p stimulates wound healing and modulates extracellular matrix gene expression pig skin: incisional wound model. J. Invest. DermatoZ. 97: 34-42, 1991. 33. Quaglino, D., L. B. Nanney, R. Kennedy, and J. M. Davidson. Transforming growth factor-p stimulates wound healing and modulates extracellular matrix gene expression in pig skin. Lab. Invest. 34.
1991.
16. Hoyt, D. G., and J. S. Lazo. Alterations in pulmonary mRNA encoding procollagens, fibronectin, and transforming growth factor-p precede bleomycin-induced pulmonary fibrosis in mice. J Exp.
Like, B., and J. Massague. The antiproliferative effect of type p transforming growth factor occurs at a level distal from receptors for growth-activating factors. J. Biol. Chem. 261: 13426-13429,
30.
12. Franzblau, C., C. A. Pratt, B. Faris, N. M. Colannio, G. D. Offner, P. J. Mogayzel, Jr., and R. F. Troxler. Role of tropoelastin fragmentation in elastogenesis in rat smooth muscle cells. J. Biol. Chem. 264: 15115-15119, 1989. 13. Fukuda, Y., V. J. Ferrans, and R. G. Crystal. The development of alveolar septa in fetal sheep lung. An ultrastructural and immunohistochemical study. Am. J. Anat. 167: 405-439, 1983. 14. Heine, U. I., E. A. Munoz, K. C. Flanders, A. B. Roberts, and M. B. Sporn. Colocalization of TGF-P, and collagen I and III, fibronectin and glycosaminoglycans during lung branching morphogenesis. Development 109: 29-36, 1990. 15. Heino, J., and T. Heinonen. Interleukin-lp prevents the stimulatory effect of transforming growth factor-p on collagen gene expression in human skin fibroblasts. Biochem. J. 271: 827-830,
Pharmacol.
L263
CELLS
63: 307-319,
1990.
Roberts, A. B., K. C. Flanders, U. I. P. Kondiah, S. J. Kim, and M. B. growth factor-p: multifunctional regulator development. Philos. Trans. R. Sot. Lond.
Heine, S. Jakowlew, Sporn. Transforming of differentiation and B Biol.
Sci. 327: 145-
154, 1990.
Sato, Y., R. Tsuboi, R. Lyons, H. Moses, and D. B. Rifkin. Characterization of the activation of latent TGF-P by cocultures of endothelial cells and pericytes or smooth muscle cells: a selfregulating system. J. CeZl Biol. 111: 757-763, 1990. G. T. Hermanson, A. K. Mallia, 36. Smith, P. K., R. I. Krohn, F. H. Gartner, M. D. Provenzano, E. K. Fujimoto, N. M. Goeke, B. J. Olson, and D. C. Klenk. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150: 76-85, 1985. 37. Van-Obberghen Schilling, E., N. S. Roche, K. C. Flanders, M. B. Sporn, and A. B. Roberts. Transforming growth factor-& positively regulates its own expression in normal and transformed cells. J. Biol. Chem. 263: 7741-7746, 1988. 38. Wakefield, L. M., D. M. Smith, T. Masui, C. C. Harris, and M. B. Sporn. Distribution and modulation of the cellular receptor for transforming growth factor-p. J. Cell BioZ. 105: 965-975, 1987. 35.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (128.205.114.091) on December 1, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
