Vol. June

177,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

14, 1991

MOLECULAR

RESEARCH

COMMUNICATIONS Pages

CLONING

AND CHARACTERIZATION

867-873

OF 2B7, A RAT mRNA WHICH

DISTINGUISHES SMOOTH MUSCLE CELL PHENOTYPES IN VITRO AND IS IDENTICAL TO OSTEOPONTIN (SECRETED PHOSPHOPROTEIN I, 2aR)

Cecilia Giachelli,

Nancy Bae, Donna Lombardi,

Department

of Pathology SJ-60, University of Washington, Seattle, WA 98195

*Department Received

April

Mark Majesky*, and Stephen Schwartz

22,

of Pathology, Baylor College of Medicine, Houston, TX 77030 1991

We have identified a rat smooth muscle cell mRNA, 2B7, which distinguishes smooth muscle cell phenotypes in vitro. Sequence and tissue distribution data strongly suggest this mRNA to be identical to osteopontm (secreted phosphoprotein I, 2aR). In viva, 2B7 mRNA is expressed in normal rat aorta and carotid arteries at levels 50-60-fold greater than heart, and about 3-4 times the levels found in adult rat kidney on a per ~1 of total RNA basis. Of the other smooth muscle sources surveyed, significant levels o f 2B7 mRNA were detectable in total RNA prepared from rat uterus and stomach, but not intestine. 2B7 mRNA levels in both carotid and aortic arte increase with age, and are elevated approximate1 5-fold in the carotid artery 48 x after balloon angioplasty. The presence of osteopontin rsecreted phosphoprotein I, 2aR) in the normal artery wall and its increased expression after injury suggests a previously unappreciated role for this molecule in the a 1991Academic Press, Inc. vascular system.

We have been studying SMC phenotypic variation utilizing an in vitro rat aortic SMC model. Two distinct, yet stable, SMC phenotypes have been identified. In contrast to normal modulation changes seen when adult SMC de-differentiate in vitro, these differences are stable in culture over many generations and under different growth conditions. Cells isolated from rat pups (“P” cells) show a characteristic epithelioid morphology, secrete PDGF, and express high levels of cytochrome P450TA1, plateletderived growth factor B chain, and elastin (1,2,3,4, and unpublished observations). These features are shared by intimal SMC (“I” cells) isolated and cultured from the neointima formed in rat carotid arteries after balloon catheter injury (4,5, and unpublished data). In contrast to “P” or “I” SMC, SMC isolated and cultured from uninjured adult aortas or carotid arteries show a spindle-shaped morphology, low to undetectable PDGF secretion, and a greatly reduced cytochrome P450IAl and elastin mRNA expression (3,4,5, and unpublished data). This model system has been useful for assessing SMC properties which Abbreviations

SMC: Smooth muscle cells; SPPI: Secreted phosphoprotein I; SD: Sprague-Dawley; WKY: Wistar-Kyoto; SHR: Spontaneously Hypertensive Rats; SSC: Standard saline citrate. 0006-291X/91

867

$1.50

Copyright 0 1991 by Actrdetnic Press. Inc. All rights of reproduction ill an! form reserved.

Vol.

177,

No.

BIOCHEMICAL

2, 1991

AND

BIOPHYSICAL

appear to be relevant to the “P”-“I” phenotype (the rr phenotype) and immature SMC in general. In order to identify additional state-specific the 1~phenotype, we utilized a differential cDNAs

which were over-expressed

markers

Interestingly, following

COMMUNICATIONS

shared by intimal SMC

and/or regulatory

molecules for

screening technique and isolated a group of

in the pup compared to the adult SMC in vitro. One of

these clones is shown here to encode the rat osteopontin report that 2B7 is normally

RESEARCH

(SPPI, 2aR) mRNA.

We also

expressed at high levels in rat vascular smooth muscle in vivo.

the steady-state

2B7 mRNA

levels in the carotid artery increase rapidly

vascular injury.

MATERIALS

AND METHODS

Cell Isolation and Culture: Thoracic aortas from 12-15-day-old or 3-month-old male SD, WKY, or SHR rats were removed and stripped of endothelium and adventitia. SMC were obtained by collagenase and elastase digestion as previously described (4). Cells were rown under identical conditions in Waymouth’s medium supplemented with 10% adult %ovine serum (Hyclone), lOOpg/ml penicillin and 0.1 mg/ml streptomycin (Gibco). All RNA preparations were performed on cells at confluence as judged by light microscopy. SMC Librarv Prenaration. Screeninp and Clone Characterization: Total RNA was isolated from confluent cultures of SMC derived from lZday-old WKY pup rats by the method of Chomczynski and Sacchi (6). polyA+ RNA was prepared by ohgo-dT chromate raphy and converted to cDNA by oligo-dT priming by the method of Gubler and Hoffman a7) using a cDNA synthesis kit (Pharmacia). EcoRI adaptor arms were added and the cDNA was ligated into the lambda-zap II phage vector (Stratagene). The resulting cDNA library contained 300,000 recombinants. Subtracted probfipreparatton and differential screening were performed as follows. High specific activity P-cDNA was prepared from cultvd pup (WKY 12) or adult (WKY3M) SMC polyA+ RNA as previously described using PdCTP (3000 Ci/mmole; New England Nuclear) (8). Sequences common to each of the preparations were subtracted using an Invitrogen Subtraction Probe Preparation kit. Briefly, newborn or adult SMC polyA+ RNA was biotinylated using photobiotin ace.&te and irradiation. 10 pg biotinylated polyAf2RNA was hybridized with heterologous -PcDNA.for 2&h. Nonhybrtdized (unique) P-cDNA sequences were separated from hybridized P-cDNA as well as excess biotinylated RNA by streptavidin binding and phenol extraction. Between 5-20% of the unsub& acted input cpms remained using this subtraction scheme. The unique pup and adult P-cDNA subtracted probes thus generated were used to screen replicate filters of the pup SMC library at equivalent concentrations (ap rox. 20,000-50,000 cpm/ml). Filters were hybridized in Rapid Hybridization Buf Per (Amersham) at 65” C for 2-4 h, washed twice in 2XSSC/O.l% SDS for five min at 220 C, twice in 2XSSC/O.l% SDS for 20 min at 650 C, twice in 0.3XSSC/O.l% SDS for 20 min at 650 C, and autoradiography was performed for 16-36 h using Kodak XAR film and Lightning plus intensifying screens. Clones which gave at least lo-fold stronger hybridization signal with the pup subtracted probe compared with the adult subtracted probe were selected, rescreened, and plaque purified by standard techniques. Plasmid DNAs were rescued from the lambda zap II vector using R408 helper phage accordmg to the manufacturer’s directions. Sequence analysis was performed using a sequence kit (IBI). Seven percent acrylamide/6M urea gels were run on a Base Runner (IBI) sequencing apparatus. DNA sequence manipulations and data searches were performed using Intelligenetic Suite or PCGENE software (Intelligenetics). RNA Isolation. Northern Blotting and Hvbridization: 3M old SD rats (Simonson labs) weighmg a proximately 400 g were anesthetized with intramuscular injection of Innovarvet, 0.2 ml Pkg and intraperitoneal injection of pentobarbital, 5.4 mg/kg (Pitman-Moore). Endothelial denudation was performed using a balloon catheter as previously described (9). Total RNA was isolated from adult nonvascular rat tissues by the procedure of Chomczynski and Sacchi (6). Carotid and aortic RNAs were prepared by the modification of the above procedure as described by Majesky et al. (4). Northern analysis was 868

Vol.

177,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

performed with ZetaProbe (New England Nuclear) membranes and using blotting, hybridization, and washing conditions which have previously been described (3). Densitometry was performed using a Beckman Model R-112 densitometer. DNA Probes and Radiolabeling: A plasmid containing a 1.3 kilobases fragment of the lyceraldehyde-3-phosphate dehydrogenase (pHc GAP) has previously been described ‘i 10). For Northern analyses, whole plasmi&, or cDNA inserts, were radiolabeled using a Schleicher and Schuell multtprime kit and -PdCT’P (800 Ci/mmole, New England Nuclear). RESULTS Molecular distinguished

Clonine of 2B7: We wished to isolate genes whose expression pup from adult vascular SMC in vifro. A cDNA

levels

library was prepared from

SMC cultured from pup rats and 500,000 plaques were differentially screened with either pup- or adult-subtracted probes as described in “Materials and Methods”. Plaques which hybridized more strongly with pup-subtracted probe compared with adult-subtracted probe were isolated and put through two additional rounds of differential screening as above. 25 cDNAs were isolated in this manner and chosen for further analysis. Four of these clones appeared to code for the same or highly related mRNA based on hybridizing mRNA size, sequence, and tissue distribution (not shown). We have named the mRNA encoded by these clones 2B7 and describe its properties

in this report.

2B7 mRNA Levels Distimmish SMC Phenotvnes In T/itro: Northern blots were prepared from RNAs isolated from cultured pup and adult SMC from various rat strains and probed with the 2B7 cDNA clone. The resulting autoradiogram is shown in Figure 1. In all samples, 2B7 interacted with a prominent 1.6 kb mRNA. As expected based on the cloning criteria, 2B7 mRNA levels were higher in cultured SMC derived from pup WKY rat aortas compared with cells derived from adult rat aortas. Cultured SMC derived from both SD and SHR rat strains showed the same relative difference dependent on the age of the donor. To further characterize and injury-specific mRNA expression in vivo was explored.

in 2B7 mRNA

levels

2B7, tissue-, developmental-,

Distribution of 2B7 in Various Tissues Of Rat In l&o: To determine the tissue distribution of 2B7, total RNA was isolated from predominantly muscular and nonmuscular

Firmre 1. Steady-state 2B7 mRNA levels distinguish SMC phenotypes in vitro. Total RNA was i lated from cultured SMC and 12.5 pg were Northern blotted. Filters were probed with s’ P-2B7, washed and autoradio raphy was performed for 16 h. SD3MTS and SD12T15, SD 3-month-old, passage 4 and 12 --day old, passage 1.5,respectively; SHR3MTS and SHRlST6, SHR 3-month-old, passage 5 and E-day-old, passage 6, respectively, and WKY3MT16 and WKY 12T16, WKY 3-month-old, passage 16 and 12-day-old, passage 16, respectively.

869

Vol.

177,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Figure 2. A) 2B7 mRNA expression in various muscular tissues of the rat. Total RNA was isolated f om the indicated adult rat tissues and 12.5 pg were Northern blotted and probed with the $2 -P-2B7 probe. E osure time was 16 h. Prolonged exposure (four days) allowed detection of a faint 2B7 ban 7 m uterus and heart, but not in skeletal muscle or intestine. B) Relative levels of 2B7 mRNA in rat tissues. Total RNA was isolated from the indicated tissue sources, of uninjured 39 ult rats and S-15 pccgwere blotted to a nylon membrane. The blot was hybndtzed first to P-2B7, then dehybridized and probed with a cDNA probe for glyceraldehyde-3-phosphate dehydrogenase (GAPD). Densttometry was performed on several different e osures of the films to assess the relative levels of 2B7 mRNA in each of the samples. Va“p ues are given as the 2B7/GAPD ratio in arbitrary densitometric units (ADU).

tissues of the adult rat, Northern blotted, and probed with 2B7. As shown in Figure 2A, the 2B7 probe hybridized strongly with a 1.6kb mRNA species in aorta and carotid RNAs, and more weakly to the same size species in stomach and uterus. Northern analysis of total RNAs isolated from the medial, adventitial and periadventitial layers of adult rat aorta indicated that 2B7 mRNA was predominantly expressed by medial SMC, with little to no mRNA

detected

in the adventitial

or periadventitial

layers of this artery

(data not shown).

No signal was detected in intestinal or skeletal muscle even after prolonged autoradiography. Total RNA samples isolated from whole adult rat kidney, lung, and liver were also analyzed by Northern analysis for 2B7 gene expression (not shown). Steady-state 2B7 levels normalized to glyceraldehyde-3-phosphate dehydrogenase for these tissues as well as those indicated in Figure 2A are plotted in Figure 2B. Of the nonmuscle tissues surveyed, the kidney showed the highest level of 2B7 mRNA. However, the carotid and aorta both contained approximately 4-fold greater 2B7 mRNA levels than the kidney. Developmental Expression of 2B7 in Rat Arteries: Since 2B7 was cloned as a gene whose expression in SMC in vitro depends on the age of the donor, it was of interest to 870

Vol.

177,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

dineafter Ww 2B7

03

GAPD

6h

8h

24h

COMMUNICATIONS

48h

7d

2B7

0 4

Figure 3. Developmental expression of 2B7 analysis was performed on 1Opg total RNA or adult (3-month-old) carotids O.~$y-W P-2B7 (upper panel) or GAPD (lower and 16 h for GAPD. Figure carotid to the (upper shown

RESEARCH

GAPD

in rat carotid and aorta. Northern blot isolated from fetal (gestational day 19), pup or aortas. After transfer, blots were probed panel). Exposure times were 16 h for 2B7

4. Increased expression of 2B7 mRNA following balloon catheter injury to.rat. . artery. A) Total RNA was isolated at the indicated times following denu adult rat carotid. 15 pg total RNA was Njsthern blotted and probed with 9;:;&-y panel), dehybridized and reblotted with -P-GAPD (lower panel). Exposure times were 5 h for 2B7 and 48 h for GAPD.

determine whether this developmental difference was also seen in arterial SMC in viva. Total RNA was isolated from rat carotids and aortas at various ages and probed with 2B7. The resulting Northern blot is shown in Figure 3. Surprisingly, 2B7 mRNA expression was lower in fetal aorta, as well as in pup aorta and carotid arteries, compared to the same tissues isolated from 3-month-old rats (upper panel). The same filter was dehybridized and probed for glyceraldehyde-3-phosphate dehydrogenase and the resulting autoradiogram is shown in the lower panel of Figure 3. Increased Expression of 2B7 mRNA After Carotid Iniury: Work done by others (11,12) as well as our own lab (13,14) has established that an endothelial denuding injury to the carotid artery initiates a series of events which culminates in the formation of an intimal layer. This layer consists of medial SMC which have migrated into the intimal region, many of which go on to proliferate there. The intimal SMC in vivo resemble rr SMC in several important ways, including loss of contractile apparatus, increased synthesis of extracellular matrix molecules and increased proliferative activity (9,15,16). Since 2B7 was isolated from cells displaying the rr SMC phenotype similar to the intimal cell, it was of interest to determine whether its expression was modulated by the injury process. As shown in Figure 4 (upper panel), steady-state 2B7 mRNA levels are increased after carotid injury. In contrast, glyceraldehyde-3-phosphate dehydrogenase mRNA levels did not vary after injury (lower panel). Densitometric scanning of the autoradiograms in Figure 4 and normalizing to GAPD mRNA levels suggests a 5fold increase in steady-state 2B7 mRNA levels 48 h after injury. 2B7 mRNA levels remained elevated seven days after injury. 2B7 mRNA levels were essentially unchanged between 0.5 h and 6 h post-injury at levels equivalent to those in uninjured carotid artery (not shown). Homologv Between 2B7 and Gsteopontin: To determine whether 2B7 showed homology to any previously characterized molecules, we performed partial sequence 871

Vol.

177,

No.

BIOCHEMICAL

2, 1991

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

analysis of several 2B7 clones. 134 bp of the 3’ end and 119 bp of the 5’ end of the longest 2B7 clone was obtained and used to search GenBank version 66 and EMBL version 25 for possible homologous sequences. A 99% match was found with the sequence reported for rat osteopontin

mRNA

(17). The 1% mismatch is most likely due to sequencing error.

This 2B7 clone represents a nearly full-length cDNA which starts 55 bp upstream of the start methionine and ends 62 bp after the third polyadenylation signal reported by Oldberg et al. (17) for rat osteopontin.

Three smaller 2B7 cDNAs

were also isolated, and partial

sequence analysis indicated that they also encode osteopontin

mRNA

(data not shown).

DISCUSSION We utilized a subtraction

cloning and differential

screening strategy in order to identify

genes that mark or are involved in maintenance of the common phenotype of cells cultured from immature vessels or from the neointima found after injury.

Using this strategy we

have identified a set of clones which appear to code for the same mRNA we have termed 2B7. Based on partial sequence analysis, mRNA size and tissue distribution data, we conclude that 2B7 is probably identical to osteopontin (SPPI, 2aR). Osteopontin is a highly acidic, phosphorylated sialoprotein originally isolated as a matrix molecule in bone (l&19). Osteopontin has also been called SPPI (20), 2aR (21) BSP-I (l&19), and 44 kDa1 bone phosphoprotein (22), depending on the group and system being studied. In addition to calcified tissues, sources for the protein include normal cells in the kidney, inner ear, placenta, decidua, brain and bone marrow (23). Intriguingly, all transformed cells thus far examined, regardless of origin, synthesize significantly higher osteopontin levels than their untransformed counterparts (19,24). The presence of osteopontin in vascular SMC has not previously been reported. show here that 2B7 mRNA

We is normally expressed at significant levels in adult rat aorta and

carotid arteries. Only trace or undetectable levels were evident in other muscular tissues of the rat, i.e., stomach, heart, uterus, intestine, and skeletal muscle. In agreement with Yoon et al. (25), we found high levels of osteopontin in rat kidney, but little expression in rat lung or liver. We have also demonstrated that steady-state 2B7 mRNA levels are regulated during vascular development and increase in response to balloon catheter injury. These studies are the first to demonstrate osteopontin expression in vascular SMC, and suggest a role for this adhesive protein in the normal development of large arteries as well as injury related processes. The unclear. A role in cell adhesion include an Arg-Gly-Asp-Ser cell vitronectin, and thrombospondin

function of osteopontin, however, in any tissue is as yet is suggested by structural features of the protein which recognition sequence similar to those found in fibronectin, (26). In support of this hypothesis, osteopontin promotes

cell attachment and spreading in bone and ligament cells in an RDG-dependent fashion (17,27,28,29). A role in Ca+ + metabolism is suggested not only by tissue distribution, but the presence of a putative Ca+ + binding site in the osteopontin protein sequence (30) and its high affinity for hydroxyapatite (17). It is intriguing to consider the possibility that osteopontin plays a role in determining the distinctive morphology ofn cells in culture and, 872

Vol.

177,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

possibly, the organization of the SMC in viva as they migrate across the internal elastica after injury and organize to form the distinct morphologic structures of the neointima. Acknowledements We thank Hiro Okazaki and Chris Peabody for help with the animal studies, and Holly Kabinoff for manuscript preparation. This work was supported by NIH Grants HL03174, HL26405, and HL073 12. REFERENCES ;*

3: 4. 5. !: ;: 10. 11. 12. 13. 14. 15. 16. 17. 18. :;: Z: i-i: 25. 26. 27. 28. 29. 30.

Gordon, D., Mohai, L.G., and Schwartz, S.M. (1986) Circ. Res. 59,633-644. Seifert, R.A., Schwartz, S.M., and Bowen-Pope, D.F. (1984) Nature 311,669-671. Giachelli, CM., Maiesky, M.W., and Schwartz, S.M. (1991) J. Biol. Chem. 266,39813986. - Majesky, M.W., Benditt, E.P., and Schwartz, SM. (1988) Proc. Natl. Acad. Sci. USA 85.1524-1528. Wglker, L.N., Bowen-Pope, D.F., Ross, R., and Reidy, M.A. (1986) Proc. Natl. Acad. Sci. USA 83.7311-7315. ChomczynsGi, P., and Sacchi, N. (1987) Anal. Biochem. 162,156-159. Gubler, U., and Hoffman, B. (1983) Gene (Amst.) 25,263-269. Sargent, T.D. (1987) Methods in Enzymology 152,423-432. Daemen, M.J.A.P., Lombardi, D.M., Bosman, F.T., and Schwartz, S.M. (1991) Circ. Res. 68,450-456. Tso, J.Y., Sun, X-H., Kao, T.H., Reece, KS., and Wu, R. (1985) Nucleic Acids Res. 13,2485-2502. Clowes, A.W., Reidy, M.A., and Clowes, M.M. (1983) Lab. Invest. 49,327-333. Reidy, M.A., Fingerle, J., and Majesky, M.W. (1988) In Hyperlipidaemia and Atherosclerosis (K.E. Suckling and P.H.E. Groot, Eds.), pp. 149-164. Academic Press, Ltd., London. Clowes, A.W., and Schwartz, S.M. (1985) Circ. Res. 56,139-145. Schwartz! S.M., Stemerman, M.B., and Benditt, E.P. (1975) Am. J. Pathol. 81,15-42. Gabbiam, G., Kocher, O., Bloom, W.S., Vandekerckhove, J., and Weber, K. (1984) J. Clin. Invest. 73,148-152. Boyd, C.D., Kniep, A.C., Pierce, R.A., Deak, S.B., Karboski, C., Miller, D.C., Parker, M.I., MacKenzie, J.W., Rosenbloom, J., and Scott, G.E. (1988) Connect. Tissue Res. 18,65-78. Oldberg, A., Franzen, A., and Heinegard, D. (1986) Proc. Natl. Acad. Sci. USA 83,8819-8823. Franzen, A., and Heinegard, D. (1985) In The Chemistry and Biology of Mineralized Tissues (W.T. Butler, Ed.), pp. 132-141. EBSCO Media, Birmingham. Franzen, A., and Hemegard, D. (1985) Biochem. J. 232,715-724. Senger, D.R., and Perruzzi, C.A. (1985) Cancer Res. 45,5818-5823. Smith, J.H., and Denhardt, D.T. (1987) J. Cell Biochem. 34,13-22. Prince, C.W., Oosawa, T., Butler, W.T., Tomana, M., Bhown, A.S., Bhown, M., and Schrohenloher, R.E. (1987) J. B~ol. Chem. 262,2900-2907. Butler, W.T. (1989) Connect. Tissue Res. 23,123-136. Craig, A.M., Nemir, M., Mukherjee, B.B., Chambers, A.F., and Denhardt, D.T. (1988) Biochem. Biophys. Res. Commun. 157,166-173. Yoon, K., Buenaga, R., and Rodan, G.A. (1987) Biochem. Biophys. Res. Commun. 148,1129-1136. Ruoslahti, E., and Pierschbacher, M.D. (1987) Science 238,491-497. Sauk, J.J., Van Kampen, C.L., Norris, K., Moehring, J., Foster, R.A., and Somerman, M.J. (1990) Exp. Cell Res. 188,105-110. Butler, W.T., Prince, C.W., Oosawa, T., Somerman, M.J., Foster, R.A., and Sauk, J.J. (1987) In Calcium Regulation and Bone Metabolism (D.V. Cohn, T.J. Martin, and P.J. Meunier, Eds.), Excerpta Medical International Congress Series, Amsterdam. Somerman, M.J., Prince, C.W., Sauk, J.J., Foster, R.A., and Butler, W.T. (1987) J. Bone Miner. Res. 2,259-265. Prince, C.W. (1989) Connec. Tissue Res. 21,15-20. 873