Overexpression of SPARC in stably transfected F9 cells mediates attachment and spreading in Ca(2+)-deficient medium.

Overexpression of SPARC in stably transfected F9 cells mediates attachment and spreading in ca2+-deficientmedium ELIZABETH A. EVERITT AND E. HELENE SAGE' Department of Biological Structure, School of Medicine, University of Washington, Seattle, WA 98195, U.S.A.

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/07/15 For personal use only.

Received July 10, 1992 EVERITT,E. A., and SAGE,E. H. 1992. Overexpression of SPARC in stably transfected F9 cells mediates attachment and spreading in c a 2 +-deficient medium. Biochem. Cell Biol. 70: 1368-1379. The ca2+-bindingprotein SPARC is one of a group of proteins that function in vitro to promote the rounding of cells. To assess whether the modulation of cell shape by SPARC is affected by extracellular c a 2 + , we used F9 cell lines that had been stably transfected with sense or antisense SPARC DNA. Sense-transfected (S) lines that overexpress SPARC are aggregated and rounded, whereas antisense (AS) lines that express low levels of the protein are flat and spread. We tested whether the cell lines would exhibit these altered morphologies in ca2+-deficient media. When cultured under these conditions, S lines attached and spread,whereas AS lines attached but remained round, with no subsequent spreading. Addition of CaCl, or purified SPARC to the cazc-deficient medium resulted in spreading of the AS and control lines and a reappearance of the altered morphologies. Expression of the ca2+-bindingcadherin uvomorulin by the cell lines correlated with neither their morphology nor their level of SPARC expression. We conclude that the altered phenotypes of the transected lines reflect, in part, the concentration of extracellular c a 2 + and that the spreading exhibited by the S lines under ca2+-deficientconditions is directly related to their enhanced expression of SPARC. SPARC might, therefore, mediate interactions between cells and matrix that are permissive for adhesion when levels of extracellular ca2+ are diminished. Key words: adhesion, cell shape, extracellular matrix, SPARC. EVERITT, E. A., et SAGE,E. H. 1992. Overexpression of SPARC in stably transfected F9 cells mediates attachment and spreading in ca2+-deficient medium. Biochem. Cell Biol. 70 : 1368-1379. La proteine SPARC liant le c a 2 + fait partie d'un groupe de proteines qui favorisent in vitro l'arrondissement des cellules. Pour Cvaluer si la modification de la forme des cellules par la SPARC est affectke par le c a 2 +extracellulaire, nous avons utilisC des lignees cellulaires F9 qui ont tte transfectees de f a ~ o nstable avec un DNA sens ou antisens de la SPARC. Les lignkes transfectees sens (S) qui surexpriment la SPARC sont agr6gCes et arrondies alors que les IignCes antisens (AS) qui expriment de faibles taux de la proteine sont aplaties et etalks. Nous avons vkrifie si les lignks cellulaires montrent ces m&mesmorphologies altCrCes dans les milieux deficients en c a 2 + .Lorsque cultivees dans ces conditions, les lignees S s'attachent et s'etalent alors que les lignkes AS s'attachent, mais demeurent arrondies sans ktalement subs& quent. L'addition de CaCl, ou de SPARC purifike au milieu deficient en c a 2 + provoque 17Ctalementdes ligntes AS et des lignees contr6les et une rtapparition des morphologies alttrkes. L'expression de la cadherine uvomoruline liant le c a 2 +par les lignees cellulaires ne correspond ni A leur morphologie ni a leur taux d'expression de la SPARC. Nous concluons que les phenotypes alterks des lignies transfecttes reflktent partiellement la concentration du c a 2 + extracellulaire et que l'ktalement exhibe par les lignees S dans des conditions de dkficience en ca2+ est directement relie a leur expression stimulee de la SPARC. La SPARC pourrait donc provoquer des interactions entre les cellules et la matrice qui permettraient l'adhbion quand les taux du c a Z +extracellulaire sont diminues. Mots cl&s : adhesion, forme cellulaire, matrice extracellulaire, SPARC. [Traduit par la redaction]

Introduction SPARC (secreted protein, acidic and rich in cysteine) is a secreted glycoprotein with two domains that have been shown to bind c a 2 +: the NH,-terminal. Glu-rich region and the COOH-terminal EF-hand (Engel et a/. 1987; Sage et a/. 1989a, 1989b; Vernon and Sage 1989). Fragments produced by trypsin cleavage, as well as regions represented by synthetic peptides of the NHz- and COOH-terminal domains of SPARC, have also been shown to bind ca2+ (Sage et al. 19896; Lane and Sage 1990). The C-terminal region contains an EF hand similar to those found in the ABBREVIATIONS: SPARC, secreted protein, acidic and rich in cysteine; S, sense; AS, antisense; DMEM, Dulbecco's modified Eagle's medium; FBS, fetal bovine serum; RA, retinoic acid; dicAMP, dibutyryl CAMP; PYS, parietal yolk sac endoderm cell line; MT, metallothionein; RIA, radioimmunoassay; PBS, phosphate-buffered saline; ELISA, enzyme-linked immunosorbent assay; SDS-PAGE, sodium dodecyl sulfate - polyacrylamide gel electrophoresis; IgG, immunoglobulin G; kDa, kilodalton(s); H/D, Ham's F12 - DMEM without CaCl, and MgCI,; BSA, bovine serum albumin. ' ~ u t h o rto whom all correspondence should be addressed. Prinred in Canada / Imprime au Canada

intracellular ca2+-binding proteins calmodulin, troponin C, and parvalbumin, and the extracellular protein fibrinogen (Kretsinger 1979; Dang et a/. 1985). The EF-hand in SPARC is a high-affinity site (Romberg et al. 1985) that binds one c a 2 + ,but there are also several low-affinity sites at the N-terminus that could theoretically bind an additional eight c a 2 + (Engel et al. 1987; Sage et al. 1989b). In the presence of c a 2 + , SPARC binds more strongly to extracellular matrix components, including types I, 111, and V collagen and thrombospondin (Sage et a/. 1989b). SPARC belongs to a group of proteins that exert antispreading effects on cells in culture (Sage and Bornstein 1991). Evidence for this function comes from the finding that SPARC and peptides from the ca2+-binding regions of SPARC induce cell rounding in spread monolayers of cultured endothelial cells and fibroblasts and prevent the spreading of newly plated cells. The antispreading effect is dose dependent and can be inhibited by antibodies specific for SPARC and the active peptides (Sage et al. 1989b; Lane and Sage 1990). Additional evidence for the antispreading activity of SPARC was recently provided by F9 teratocarcinoma cells that were stably transfected with S and AS

1369


EVERITT AND SAGE

SPARC cDNA. Cell lines that overexpressed SPARC were aggregated and rounded, whereas underexpressing lines were flat and spread; control F9 cells exhibited a morphology intermediate between those of S and AS cells (Everitt and Sage 1992). It therefore appears that SPARC acts extracellularly to modulate cell shape and cell-matrix interactions. It has recently been shown that SPARC does not round cells by abstraction of c a 2 + from proteins at the cell surface (Sage 1991), but the mechanism by which SPARC effects changes in shape is currently unknown. In this study, we used stably transfected F9 cell lines to ascertain whether the over- or under-expression of SPARC affected the adhesion, spreading, and (or) morphology of these cells when they were cultured in Ca2+-deficient media. Since it was formally possible that the altered expression of SPARC had influenced the production of another extracellular ca2+-bindingmolecule, we also asked whether the expression of uvomorulin (E-cadherin, cell CAM 120/80, L-CAM), a ca2+-dependent cadherin abundant on the surface of F9 cells (Yoshida and Takeichi 1982; Ozawa et al. 1990), was consistent with that of SPARC. We report that the altered morphologies normally exhibited by the stably transfected lines were not influenced by the levels of uvomorulin, but were instead a function of the concentration of c a 2 + in the media; the lines did not assume their typical morphologies in ca2+-deficient media. Moreover, the S lines exhibited spreading under ca2+-deficient conditions, a characteristic that appeared to be directly associated with the presence of SPARC in these cultures. Since the mitotic activity of many cells diminishes when extracellular ca2+ becomes low (Calabretta et al. 1986), and SPARC has been shown to induce a rounded cellular phenotype at physiological levels of ca2+ (Sage et al. 1989b), but a spread phenotype under ca2+-deficient conditions (reported here), cells in low ca2+ environments that are exiting the cell cycle might utilize SPARC to facilitate spreading.

Materials and methods Cell culture Murine F9 embryonal carcinoma stem cells were obtained from American Type Culture Collection. F9 stem cells and transfected clones were routinely cultured in DMEM (Sigma) containing high glucose and supplemented with 10% heat-inactivated FBS (Grover et al. 1983). F9 cells were grown on 100-mm tissue culture dishes (Corning) coated with 1% gelatin at 37°C in a humidified atmosphere containing 5% C02. For some experiments, cells were plated directly onto non-coated tissue culture plastic in Ham's F12, Ham's F12 - DMEM, or Ham's F12 - DMEM without CaCl, and MgC1, (Sigma). FBS was dialyzed against culture medium and used as described in the figure legends. Cells were passaged, induced to differentiate to parietal endoderm with RA and dicAMP, and stored as frozen stocks as previously described (Everitt and Sage 1992). PYS-2 cells were a gift from Dr. John Lehman (Albany Medical College, Albany, N.Y.) and were cultured as previously described (Sage et al. 1989b). Expression vectors, transfection, and characterization of cell lines Expression vectors designed to overexpress or inhibit the expression of SPARC were constructed and verified for orientation as previously described (Everitt and Sage 1992). Briefly, the entire cDNA insert for SPARC was ligated, in S or AS orientation, into two plasmid vectors, one containing the SV-40 promoter and polyadenylation signal sequences (pko-neo; Van Doren et al. 1984), and one which was inducible by ZnC1,. This cassette contained the mouse MT promoter and the human growth hormone polyadenylation site, and the resulting plasmid was termed pMT-SPARC. SV40

S-plasmids were termed pSV-SPARC and AS plasmids were termed pSV-CRAPS. Transfection of F9 cells was performed with DNA prepared as calcium phosphate precipitates (Graham and van der Eb 1973) and added to the cultures in the presence of ~ i ~ o f e c t (BRL). i n ~ ~ Cells were cotransfected with pko- or pMT-neo to allow for selection with the antibiotic G418 (Sigma). Stably transfected lines were routinely maintained in 100 pg G418/mL; no antibiotics were used on the F9 line. Stably transfected lines used in this study have been previously characterized by Southern and Northern analysis for SPARC DNA and poly(A) mRNA, respectively, and immunoblot analysis and RIA for SPARC protein (Everitt and Sage 1992). Nine cell lines were used in this study: (i) two lines served as controls, the parental F9 cells and Neo, the line isolated after transfection of pko-neo alone, (ii) four S lines, two from pSV-SPARC transfections (S-5 and S-8) and two from pMT-SPARC transfections (S-27, uninduced, and S-27 + ,induced in the presence of 40 pM ZnCl,), and (iii) three lines from pSV-CRAPS transfections (AS-1, -2, and -3). Selection of lines for the present study was based on the variation that they demonstrated in the amounts of integrated SPARC DNA and in the expression of SPARC mRNA and protein (Everitt and Sage 1992). +

Immunoblot analysis and RIA Immunoblot analysis of cell culture media was performed as previously described (Sage et al. 1989a, 1989b), with minor modifications. F9 and transfected cell lines were released with trypsin-EDTA in PBS, counted, and plated in equal numbers at high density under routine culture conditions. Cells were incubated for 24 h, media were collected and clarified, and a mixture of protease inhibitors was added. Antibodies were specific for SPARC (Sage et at. 1989a, 1989b) and have been previously characterized by ELISA, irnmunoblotting, and radioimmune precipitation (Sage et al. 1989a, 19896). SDS-PAGE was performed for immunoblot analysis and quantitation. RIA of culture medium was performed for each of the stable lines according to a protocol modified from Malaval et al. (1987). Briefly, equal numbers of cells were plated on uncoated cultureware under routine conditions for 6 h; medium was collected and clarified. To obtain a standard curve, SPARC protein was dissolved in DMEM containing 10% FBS or in a Trissaline buffer containing 2% nonfat dried milk at pH 7.9 (Tris buffer), and serial dilutions were made from 250 to 1 ng/mL. 12'1labelled SPARC in Tris buffer or DMEM containing 10% FBS was added to tubes containing sample or SPARC standard. To this mixture a polyclonal anti-SPARC antibody was added for an incubation overnight at 40°C. Immune complexes were precipitated in a 25% polyethylene glycol-8000 solution in Tris buffer without nonfat dried milk, and the radioactivity of the precipitates was determined by a Y-counter. Analyses of uvomorulin Immunoblot analysis of total uvomorulin was performed on cell layers of transfected and control cell lines. Cells were treated with trypsin-EDTA, counted in duplicate, and plated in equal numbers under routine culture conditions. After 6 h, the cells were solubilized in SDS-PAGE sample buffer (Laemmli 1970) and an aliquot was assayed for protein concentration with Coomassie protein assay reagent, according to the manufacturer's instructions (Pierce). Equal concentrations of protein from each cell line were subjected to SDS-PAGE, transferred to nitrocellulose, and incubated with a rat anti-mouse uvomorulin monoclonal antibody, DECMA-1 (Vestweber and Kemler 1985; Schuh et al. 1986; Sigma), at a dilution (by volume) of 1:1000. Blots were incubated with rabbit anti-rat IgG, followed by 1Z5~-labelled protein A Sepharose, both at dilutions of 1:1000. After several washes, blots were exposed to preflashed X-Omat AR film (Kodak) with intensifying screens at -70°C. For analysis of extracellular uvomorulin, cell lines were incubated with trypsin in the presence of c a 2 + ,a treatment which releases

1370

BIOCHEM. CELL BIOL. VOL. 70, 1992

TABLE1 . S and AS cells express enhanced or diminished levels of SPARC


Cell line

FIG. 1. Stably transfected F9 cell lines exhibit enhanced or diminished levels of SPARC. Equal numbers of stably transfected cells were plated for 24 h in culture medium prior to resolution of culture medium proteins by SDS-PAGE, transfer to nitrocellulose, and incubation with polyclonal anti-SPARC IgG. Proteins were resolved on a 5% stacking - 10% separating gel in the presence of 50 mM dithiothreitol. C, control; AS, antisense transfectants; S, sense transfectants. Lane 1, SPARC from PYS cells; Iane 2, medium from F9 control line; Iane 3. Neo control lie; Iane 4, AS-3; lane 5, S-5; lane 6, S-27; lane 7, 5-27+ (sense line pMT-SPARC plus ZnCI,). The bands of apparent M, 43 000 (arrow) that appear in each lane correspond to control, reduced, or enhanced IeveIs of SPARC secreted from the transfected cell lines.

F9 Neo AS-1 AS-2 AS-3 S-5 S-8 S-27 S-27 +

SPARC (ng/mL)*

Yo of controlt

18 17 13 13 13 33 22 19

100+5 94+6 72+3 72+5 72 + 4 183+7 122+ 5 110+4 240

-z

*Determined by radioimmunoassay; n = 5 . ' ~ e a n? SEM. IRIA conducted on culture medium containing ZnCll was imprecise. Values for pMT-SPARC line S-27 + were derived from duplicate scans of autoradiograms representing an immune precipitation and an immunoblot, and were calculated relative to the levels of SPARC produced by untransfected F9 cells in those assays, i.e., 240% of control.

an 82- to 86-kDa fragment of uvomorulin from the F9 cell surface (Ringwald et al. 1987; Hyafil et al. 1981; Peyrieras et al. 1983). Briefly, equal cell numbers were plated as described above and allowed to attach, prior to incubation with 0.05% (w/v) trypsin (Worthington) in HEPES-buffered saline containing 2 mM CaC1, (trypsinates) for 1 h at 37°C (Hyafil et al. 1981; Peyrikras et al. 1983; Yoshida-Noro et al. 1984). Trypsinates were clarified, dialyzed against three changes of cold H,O containing 4 mM CaCI,, frozen, and lyophilized. Protein concentration of lyophilized material was determined as described above, and equal concentrations of protein were subjected to SDS-PAGE. Immunoblot analysis and autoradiography were performed as described above.

ment and spreading of parietal endoderm cells, control and transfected lines were incubated for 7 days with RA + dicAMP, treated with trypsin-EDTA, and subsequently plated as described above for the undifferentiated, transfected lines. SPARC, purified as previously described (Sage el al. 1989b), was added to culture wells at the various concentrations specified in the table and figure legends. Cell morphology was assessed by visual inspection with an inverted phase photomicroscope and the images were recorded on Ektachrome film (ET 135, Kodak). Multiple fields were photographed for each well whenever feasible. For quantitation of round and spread cells, cells that were overtly round or surrounded by a halo of refractility were classified as round, whereas cells that were spread and flattened, with minimal cellular refractility, were scored as spread. Cells were photographed as described above and a 7.5 x 12.5 cm black and white print was made from each Ektachrome slide. The total number of cells and the number of round and spread cells were subsequently counted in two different fields (100-200 cells each).

Experiments under ca2+-deficientconditions F9 cells and stably transfected lines were cultured in premixed DMEM, Ham's F12, Ham's F12 - DMEM, or H/D medium (Sigma); for some experiments this latter medium was supplemented with MgSO, to the level present in DMEM (0.8 mM). The H/D medium is also deficient in various amino acids, glucose, and NaHCO,; these compounds were supplemented to the levels present in the regular DMEM. FBS was dialyzed at 4OC for 24 h against two changes of H/D medium. Stock solutions of 1 M CaCl,, 1 M MgSO,, 0.1 M EDTA, 0.1 M EGTA, 50 mg BSA/mL (Sigma), and 1 mg/mL of the peptides GRGDSP and GRGESP (provided by Dr. Ron Heimark, ICOS, Bothell, Wash.) were prepared in distilled H,O or tissue culture-tested PBS (without ca2+ and M ~ , + Sigma), , the pH was adjusted to 7.4, and the solutions were sterile-filtered prior to their addition to the media. Addition of CaCI,, EDTA, EGTA, SPARC, BSA, GRGDSP, or GRGESP at the concentrations used in these experiments did not alter the pH of the culture media. Cells were treated with trypsin-EDTA, centrifuged, resuspended, counted, and plated onto non-coated plastic, 24- or 48-well culture clusters (Nunclon or Corning). Experiments were photographed at 3-4 h and (or) 20-24 h after plating or addition of the reagents. In some cases, the medium was replaced to eliminate cell debris. Cells incubated with chelators + c a 2 + were either plated in the presence of these agents or were allowed to attach for 24 h prior to addition of the chelating agents. S line 27 was not incubated with ZnC1, for these experiments, as it was determined to be deleterious to cell viability in H/D medium. For analysis of attach-

Results Cell lines and SPARC expression Seven cloned F9 cell lines stably transfected with SPARC or CRAPS cDNA were chosen for this study to represent a range of expression of SPARC mRNA and protein. Figure 1 is an immunoblot of SPARC secreted by several of the clones. For each line, equal numbers of cells were plated for 24 h in culture medium prior to SDS-PAGE. Relative to the control cell lines F9 and Neo (lanes 2 and 3), AS-3 (lane 4) showed diminished levels of SPARC, whereas lines S-5 (lane 5) and S-27+ (lane 7) exhibited increased amounts. Line S-27, which was not induced by ZnC12, showed a minimal increase in SPARC expression (lane 6), a result similar to that obtained by RIA (see also Table 1). The concentrations of SPARC in culture medium from cells of the various lines were also measured by RIA (Table 1). The S lines typically exhibited levels of SPARC that were 110-240% above those of control F9 and Neo lines (which were essentially equivalent), whereas SPARC production in all of the AS lines was 28% below the normally low levels exhibited by F9 cells. The S and AS lines consistently exhibited morphologies in vitro that were clearly altered from those of the control lines (Fig. 2A). S cells were rounded and aggregated (Fig. 2C), and AS cells were flat and spread (Fig. 2B; and

1371


EVERITT AND SAGE

FIG. 2. Undifferentiated S and AS lines display altered morphologies. Cells were treated with trypsin and EDTA, dispersed as single cells, counted, and replated at equal densities in DMEM - 10% FBS. (A) F9,untransfected; (B) antisense line AS-3; (C) sense line S-8. Magnification, 100 x .

1372


Everitt and Sage 1992). These altered morphologies were also apparent after the lines were induced to differentiate into parietal endoderm with RA dicAMP, but there were no changes in the attachment of the lines to fibronectin-coated substrates or in the levels of secreted metalloproteinases (Everitt and Sage 1992). Although we had proposed that SPARC was influencing cell morphology directly, it was possible that its effect was secondary to the perturbation of intercellular adhesion mechanisms. Thus, we investigated whether expression of the cell-surface cadherin of F9 cells, uvomorulin, had also been altered in the transfected F9 cell lines.


+

FIG. 3. Expression of uvomorulin by F9 and transfected cell lines does not correlate with levels of SPARC. (A) Immunoblot of total uvomorulin. Equal numbers of F9 stably transfected S and AS cells were incubated for 6 h in culture medium. Equal concentrations of protein extracted from the cells were resoIved by SDSPAGE,transferred to nitroceIlulose, and incubated sequentially with a monoclonat anti-mouse uvomorulin IgG and '"I-labelled protein A - Sepharose, S, sense lines; AS, antisense lina; C, control lines. Lane 1, S-27 lane 2, S-27; lane 3, AS-2; lane 4, AS-I; lane 5, S-8; lane 6, S-5: lane 7,AS-3; lanes 8 and 9, control lines F9 and Neo. The bands on the autoradiograph of apparent M, 120 000-123 000 that appear in each lane correspond to control, reduced, or enhanced levels of uvomorulin extracted from the transfected cell lines. (B) Immunoblot of cell surface uvomorulin. Equal numbers of F9 and stably transfected S and AS cells were incubated for 1 h in a weak solution of trypsin plus c a 2 + , and equal volumes of proteins extracted from the cell surfaces were resolved by SDS-PAGE and immunoblotted as described for A. Lane 1, S-27+; lane 2, AS-1; lane 3, S-5; lane 4, AS-3; lanes 5 and 6, control lines F9 and Neo. The bands on the autoradiograph of apparent M, 84 000-86 000 in each lane correspond to control, reduced, or enhanced levels of uvomorulin extracted from the surfaces of transfected cells. Positions of the molecular weight standards are indicated (M, x lop3).

+;

No correlation between levels of SPARC and uvomorulin F9 cells express high levels of uvomorulin, which mediates ca2+-dependent adhesion (Yoshida and Takeichi 1982; Vestweber and Kemler 1984; Ringwald et al. 1987; Ozawa et al. 1990). Changes in the distribution or availability of this molecule on S and AS cell surfaces could contribute to alterations in morphology: overexpressors of SPARC would be inclined to aggregate and underexpressors would remain apart. Immunoblots of total uvomorulin, analyzed as equal amounts of protein extracted from cells, are shown in Fig. 3A. Although there was variability among the cell lines in their production of uvomorulin, there was no apparent correlation with the expression of SPARC or the morphology of the lines. For example, AS-2 (lane 3) and AS-3 (lane 7) displayed extremely low levels of uvomorulin that were similar to those expressed by S-5 (lane 6), a transfected line that synthesizes high levels of SPARC. In contrast, line AS-1 (lane 4) produced levels of total uvomorulin that were similar to those of S-27 lines (lanes 1 and 2). To assess levels of uvomorulin present on cell surfaces, cells were incubated in a weak solution of trypsin in the presence of c a 2 + . The cell-surface proteins released by trypsin were separated by SDS-PAGE, transferred to nitrocellulose, and exposed to an anti-uvomorulin monoclonal antibody. As shown in Fig. 3B, there were variations in the amount of cell-surface uvomorulin that also did not correlate with the levels of SPARC expressed by the lines. The SPARC-transfected lines also exhibited a surface uvomorulin of greater apparent M, than the control lines F9 and Neo (cf. lanes 1-4 with lanes 5-6, Fig. 3B). The cause of this difference in molecular size is presently unknown. From these data it was apparent that the clonally derived cell lines produced variable amounts of uvomorulin that were not correlated with their expression of SPARC. Cell culture under ca2+-deficientconditions Since SPARC has been shown to bind c a 2 + as well as affect cell spreading in vitro, we asked whether the round and spread morphologies of lines secreting high and low levels of SPARC would be sustained in Ca2+-deficient media. Initially, control and transfected lines were plated in DMEM containing 2.5% FBS and 2 mM EGTA (Fig. 4). Under these conditions, cells of S lines attached and spread on uncoated tissue culture plastic (F) to a greater extent than AS (E) or control lines (D). Cells of control (D) and AS lines (E) were rounded and minimally attached in this medium after 24 h, a result obtained as early as 3 h after the cells were plated (Figs. 5A and 5C), or when either EGTA or EDTA was added to the cells 24 h after subculture (not shown). It should be noted that the morphologies displayed by the cells in DMEM containing a low concentration of

1373


EVERITT AND SAGE

FIG. 4. Spreading of control F9 and AS cells is abrogated in ca2+-deficientmedium. Equal numbers of F9, AS, and S cells were incubated in DMEM containing 2.5% FBS for 24 h (A-C; controls) or in the same medium containing 2 m M EGTA (D-F). AS, antisense lines; S, sense lines. A and D, F9; B and E, AS-1; C and F, S-27 + . Magnification, 1 0 0 ~ .


1374

BIOCHEM. CELL BIOL. VOL. 70. 1992

FIG. 5. Exogenous SPARC mediates spreading of control F9 and AS cells in ca2+-depleted culture medium. F9 and AS cells were incubated in DMEM containing 2.5% FBS and 2 mM EGTA for 3 h (A,C) or in the same medium containing 0.5 pM SPARC (B,D). A and B, F9; C and D, AS-3. Percentages of round versus spread cells were calculated from photomicrographs: A, 94% round, 6% spread; B, 40% round, 60% spread; C, 100% round; D, 34% round, 66% spread. Magnification, 1 0 0 ~ .

FBS (shown as controls in Figs. 4A-4C) were less distinct compared with those observed when S and AS lines were cultured in DMEM supplemented with 10-1 5% FBS (Fig. 2; and Everitt and Sage 1992). However, control and AS lines could be induced to spread in DMEM formulated with 2.5% FBS and 2 mM EGTA by the addition of 0.5 pM purified SPARC at the time of plating (Figs. 5B and 5D). S lines, which already overexpressed SPARC, exhibited little change in spreading or rounding after addition of SPARC to this medium (not shown). To eliminate chelating agents from the medium, subsequent experiments were performed in ca2+-free defined H/D medium containing 2.5% FBS. Under these conditions, S cells exhibited a greater degree of spreading compared with that of AS and control cells (data not shown). Cells appeared similar in H/D medium with either 2.5 or 10% FBS. Thus, the altered morphologies were most apparent in cell lines cultured in DMEM (Everitt and Sage 1992), Ham's F12, or

Ham's F12 - DMEM containing 10% FBS, but were minimized in ca2+-deficientH/D medium containing 10% FBS (data not shown). To ensure that dialysis of the FBS had not eliminated a factor requisite to the expression of the S or AS phenotype, we also performed experiments in DMEM containing 10% dialyzed FBS. Culture of the lines in this medium produced results similar to those obtained in media containing nondialyzed FBS (data not shown). It was, therefore, not the concentration of serum, but most likely the presence of ca2+ in nondialyzed FBS and regular DMEM, Ham's F12, or Ham's F12 - DMEM that accounted for the rounded or spread morphologies of the respective S and AS lines. When c a 2 + was added back to the H/D medium, there was a dose-dependent increase in the ability of the AS and control cell lines to spread on the culture substrate (data not shown). Moreover, in the presence of increasing concentrations of c a 2 + ,the lines achieved morphologies resembling

1375


AND SAGE

those found in DMEM containing FBS: AS lines appeared spread, whereas S lines were rounded and aggregated. These experiments, as well as the data shown in Fig. 5, indicated that SPARC and (or) c a 2 + could elicit spreading of the AS and control lines cultured in ca2+-deficient H/D medium. Since levels of cellular uvomorulin were correlated neither with the altered morphologies of the lines nor their levels of SPARC expression, we examined the role of SPARC itself in the process of cell spreading. Purified SPARC (final concentration, 1.0 pM), 30 pg/mL (0.5 mM) of BSA, or 100 pg/mL (17 pM) of GRGDSP or GRGESP was added to H/D medium containing either 5 or 10% FBS, at 0-3 h postsubculture. There was no difference in the appearance of cell lines incubated with BSA and control cell lines cultured in H/D medium containing 5% FBS (data not shown). Twenty-four hours after the addition of 1 pM SPARC, the number of spread cells in all the lines was greater than that of cells cultured in the absence of SPARC (Fig. 6 ) . This result was similar to that obtained with cells cultured in DMEM-EGTA and SPARC (Fig. 4). In contrast, the peptide GRGDSP prolonged the time required for attachment of all the cell lines in H/D medium containing 10% FBS. For as long as 8 h postsubculture, cells in the presence of GRGDSP remained unattached to the culture dish, whereas all lines were able to attach and S lines were able to spread within 3 h after the cells were plated in the same medium containing GRGESP (data not shown). Between 17 and 24 h later, all the cell lines were attached in the presence of GRGDSP and the S cells also exhibited spreading. Cells cultured with the control peptide GRGESP over the same interval of time appeared equivalent to control cells without peptide (data not shown). Thus, the peptide GRGDSP inhibited the spreading of the S lines in H/D medium only at early time points. We had previously shown that the morphologies of the S and AS lines persisted during differentiation to parietal endoderm (Everitt and Sage 1992). To determine whether S lines would maintain their ability to spread under ca2+-deficient conditions after differentiation to parietal endoderm, control and transfected cells were cultured for 7 days with RA + dicAMP in DMEM containing 10% FBS. After treatment with trypsin-EDTA and transfer into H/D medium containing 5% FBS, cells from all lines attached and exhibited partial spreading (Figs. 7A-7C). In the presence of 0.5 mM EGTA, however, control and AS cells detached and were aspirated when the medium was changed to optimize photographic conditions (Figs 7D and 7E). In contrast, S lines exhibited a greater degree of attachment and spreading, compared with control and AS lines, under conditions of c a 2 + depletion (Fig. 7F). The addition of 1 mM CaC12 to H/D medium containing 5% FBS and 0.5 mM EGTA induced attachment and spreading of all cell lines, and the morphologies typical of the differentiated lines in regular DMEM reappeared (Figs. 7G-71). Therefore, all the differentiated lines could attach and spread in ca2+-deficientmedium. In contrast to the S lines, however, AS and control lines were more easily detached in the presence of EGTA. Discussion In seven stably transfected F9 cell lines that overexpress or underexpress SPARC mRNA and protein, we found that (i) expression of uvomorulin was not correlated with either

FIG. 6. Transfected cell lines cultured in cazf -deficient H/D medium with exogenous SPARC exhibit a greater degree of spreading. Percent of round versus spread cells was calculated from photomicrographs of cells cultured 24 h with (+) or without ( - ) the addition of 1.0 pM SPARC to H/D medium with 2.5% FBS. Dark bars, percentage of round cells; striped bars, percentage of spread cells, SEM (n = 2). F9, control cell line; AS, antisense line; S, sense line.

*

SPARC expression or the morphology of the cell line, (ii) the altered morphologies of the AS and S lines did not appear in ca2+-deficient medium, and (iii) SPARC and (or) c a 2 + facilitated cell spreading in ca2+-deficient medium. If surface expression of uvomorulin was diminished in AS cells, the associated reduction in intercellular adhesion could contribute to cell separation and spreading on the substrate. In contrast, higher levels of uvomorulin might cause the cells to adhere more to each other than to the substrate. Although the levels of uvomorulin were markedly different among the lines, the variation did not correlate with the respective levels of SPARC. There appeared to be significant intercellular variation in the amount of uvomorulin present in cultures of F9 cells, as we found that SPARC transfectants, each cloned from single cells, exhibited variable levels of this protein. It, therefore, appeared that the F9 and transfected cell lines expressed at least two different secreted ca2+-binding proteins that influenced cell shape and (or) intercellular adhesion. We suggest that these proteins have mutually exclusive functions for these cells and are regulated independently. The distinctive morphologies of the S and AS cells were not apparent under ca2+-deficient conditions, although regular DMEM, Ham's F12, or Ham's F12 - DMEM readily promoted spreading (AS cells) or rounding and aggregation (S lines). FBS modulated the morphology of all the cell lines (including untransfected F9 cells) in ca2+-containing media. However, ca2+-deficientmedia and FBS failed to support the distinct morphologies described for S and AS cells. Although components of FBS other than c a 2 + could have influenced the shape and adhesion of the lines, our results indicated that the levels of c a 2 + in the medium significantly affected the phenotype of SPARC-transfected cells. The addition of CaC12 to culture medium containing EGTA probably also improved cell viability, since a deficiency of c a 2 + compromises many cellular metabolic

1376


EGTA+CaClq


EGTA

FIG. 7. Morphology of differentiated F9 control and transfected cells in ca2+-deficientmedium. Cells were treated with RA/dicAMP for 7 days and replated into H/D medium containing 5% FBS. (A-C) Controls. Differentiated cells were plated in H/D medium containing 5% FBS. (D-F) Cells were plated as in A, except that 0.5 mM EGTA was added to cultures. (G-I) Cells were plated as in D-F with the addition of 1 mM CaCl,. AS, antisense lines; S, sense lines. A, D, and G, control (neo-transfected); B, E, and H, AS-1; C, F, and I, S-5.


EVERITT AND SAGE

pathways. Therefore, the increase in cell spreading that we observed in the presence of CaC12 could have been due largely to the restoration of more optimal culture conditions. Ca2+ might in fact be a necessary cation for the renewal of cell surface proteins that were removed by trypsin during subculture. Increased spreading of all lines was also observed with the addition of SPARC to H/D medium or to medium containing EGTA or EDTA. The ca2+ bound to the EF hand in SPARC would presumably not have been available to facilitate cell spreading, but it is possible that several ca2+ associated with the N-terminus would dissociate more readily from SPARC under conditions of diminished ionic strength (Engel et al. 1987). Our results are consistent with the hypothesis that SPARC is an antispreading protein. Circular dichroism studies have shown that SPARC undergoes reversible conformational changes coincident with the binding of c a 2 + . The ca2+-induced transition toward an increase in a-helicity indicated that > 1 c a 2 + was bound to SPARC and that ca2+-binding was cooperative (Engel et al. 1987). If some of the N-terminal, low-affinity ca2+-binding sites were unoccupied in ca2+-deficientmedium, the SPARC used in some of the experiments and secreted by the cells might have been denatured and thus unable to exert an antispreading effect (Sage et al. 1989b). In ca2+-deficient medium, S lines, which overexpress SPARC, should then exhibit more spreading compared with normal culture conditions, whereas control and AS lines should exhibit morphologies similar to those seen when chelating agents are added to DMEM. These predicted results were in fact obtained: (i) cells appeared similar in both ca2+-deficientH/D medium and in DMEM containing EGTA or EDTA, and (ii) the addition of exogenous SPARC to ca2+-deficient medium increased the number of spread cells in all the lines. These experiments support the notion that cells can spread in vitro in the presence of partially denatured (inactive) SPARC. This prediction, however, is based on the premise that cells normally secreting SPARC in vitro maintain mechanisms and (or) produce molecules that override or interact with SPARC to eliminate its antispreading effect. The net result would be culture conditions permissive for spreading. Although there is currently no evidence for such a mechanism, cells that normally secrete SPARC in vitro are able to attach and spread. One of several adhesive extracellular matrix proteins (fibronectin, vitronectin) or their receptors (integrins) might be involved in interactions which could abrogate the effect of SPARC. Although it is assumed that SPARC functions as an extracellular ca2+-binding protein, there is presently no evidence showing that SPARC changes cell shape by modulating c a 2 + at the cell surface (Sage 1991). We have shown that native, a-helical SPARC and synthetic peptides representing its two ca2+-binding domains caused cell rounding in regular media (Sage et al. 1989b; Lane and Sage 1990). However, it is also possible that c a 2 + merely stabilizes the native conformation of SPARC, as suggested by Engel et al. (1987), and that SPARC might subserve an additional function in the absence of c a 2 + . There is precedence for Ca2+-bindingproteins to perform cellular functions other than those executed in the presence of bound ligand. For example, calmodulin has been shown to stimulate extracellular adenylate cyclases in a ca2+-independent manner (Greenlee et al. 1982), and mutant calmodulins that

1377

do not bind c a 2 + functioned as wild-type calmodulin in yeast to support growth, growth arrest during N2 starvation, and recovery from heat shock (Geiser et al. 1991). SPARC and other secreted ca2+-bindingproteins, such as fibrinogen (Dang et al. 1985) and uvomorulin (Ringwald et al. 1987), might perform extracellular functions that are not contingent upon the presence of bound c a 2 + . Alternatively, c a 2 + could stabilize these molecules against protease digestion or heat denaturation (Hyafil et al. 1981; Marguerie et al. 1977). Finally, the spreading exhibited by S cell lines in ca2+-deficient media might have been mediated by another protein that was altered secondarily by the enhanced expression of SPARC. SPARC is believed to affect cell spreading through the disruption of interactions between cells and their extracellular matrix (Sage and Bornstein 1991). In our experiments, substrate-bound vitronectin and fibronectin, present in FBS, presumably mediated cellular attachment in conjunction with heparan sulfate proteoglycans and (or) integrins (Norris et al. 1990). However, the cell lines might have utilized these proteins differentially during attachment or might not have expressed appropriate receptors for either ligand. The divalent cation requirement of the integrins for ligand binding should have been satisfied by the presence of ~ g ' ' (Smith and Cheresh 1988). We were able to show that the cells remained rounded in the presence of GRGDSP for a maximum of 8 h; therefore, the rounding effect of the RGD sequence overcame the spreading effect of SPARC on the S lines. At this point, however, there is no direct evidence linking the apparent mechanisms by which SPARC and RGD-containing peptides perturb cell-matrix interactions. Reversion of the lines to a control morphology after 8 h indicated that the RGD peptide had been degraded or that the cells had synthesized additional integrins. All of the cell lines were able to spread in c a 2 +-deficient medium after differentiation to parietal endoderm. The process of F9 cell differentiation in vitro, however, proceeds with the expression of many secreted proteins, such as laminin, type IV collagen, and SPARC (Strickland et al. 1980; Mason et al. 1986). In contrast, expression of some cell surface proteins such as uvomorulin ceases upon differentiation (Vestweber and Kemler 1984). Either the deposition of laminin and type IV collagen or the absence of uvomorulin could account for the enhanced attachment and spreading of the differentiated lines. Conversely, the presence of denatured SPARC in the ca2+-deficient medium could have facilitated cell spreading, an explanation already provided for the results obtained with the undifferentiated F9 cell lines. Since only S cells maintained contact with the substrate in the presence of EGTA, overexpression of SPARC might continue to play a role in mediating the enhanced attachment and spreading of parietal endoderm cells under ca2+-deficient conditions. The tissue distribution of SPARC mRNA and protein (Sage et al. 1989a; Holland et al. 1987; Wewer et al. 1988), as well as the identification of antispreading activity (Sage et al. 1989b), indicates a potential involvement of SPARC with remodelling and (or) morphogenesis. Cellular events associated with shape change vary markedly in their c a 2 + requirements. For example, mitosis requires high levels of extracellular c a 2 + , whereas cellular transformation and metastasis are associated with decreased requirements for c a 2 + (Means and Rasmussen 1988). Since increased

1378

BIOCHEM. CELL BIOL. VOL. 70.

amounts of S P A R C have been identified with certain invasive tumors in vivo (Mann et al. 1987; Columbo et al. 1989), a ca2+-deficient extracellular milieu might create a n environment in which S P A R C could promote transient cell adhesion.


Acknowledgments Thanks are due P. Hasselaar a n d J. C. Yost for purified SPARC, W. J. Nelson for suggestions regarding the uvomorulin experiments, a n d M. Nameroff for comments a n d advice throughout these experiments. This work was supported by a Public Health Service National Research Service Award 5T32-GM07270 grant from National Institutes of Health t o E.A.E. a n d a National Institutes of Health grant GM-40711 t o E.H.S.

Calabretta, B., Battini, R., Kaczmarek, L., de Riel, J.K., and Baserga, R. 1986. Molecular cloning of the cDNA for a growth factor-inducible gene with strong homology to S-100, a calciumbinding protein. J. Biol. Chem. 261: 12 628 - 12 632. Columbo, M.P., Biondi, G., Galasso, D., Baracetti, P., Howe, C.C., and Parmiani, G. 1989. Osteonectin transcript and metastatic behavior in V-Ki-ras transformed fibroblasts. Int. J. Cancer (Suppl.), 4: 76-77. Dang, C.V., Ebert, R.F., and Bell, W.R. 1985. Localization of a fibrinogen calcium binding site between g-subunit positions 31 1 and 336 by terbium fluorescence. J. Biol. Chem. 260: 9713-9719. Engel, J., Taylor, W., Paulsson, M., Sage, H., and Hogan, B. 1987. Calcium binding domains and calcium-induced conformational transition of SPARC/BM-40/osteonectin, an extracellular glycoprotein expressed in mineralized and nonrnineralized tissues. Biochemistry, 26: 6958-6965. Everitt, E.A., and Sage, E.H. 1992. Expression of SPARC is correlated with altered morphologies in transfected F9 embryonal carcinoma cells. Exp. Cell Res. 199: 134-146. Geiser, J.R., van Tuinen, D., Brockerhoff, S.E., Neff, M.M., and Davis, T.N. 1991. Can calmodulin function without binding calcium? Cell, 65: 949-959. Graham, F.L., and van der Eb, A. 1973. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology, 52: 456-467. Greenlee, D.V., Andreasen, T.J., and Storm, D.R. 1982. Calciumindependent stimulation of Bordatellapertussis adenylate cyclase by calmodulin. Biochemistry, 21: 2759-2764. Grover, A., Oshima, R.G., and Adamson, E.D. 1983. Epithelial layer formation in differentiating aggregates of F9 embryonal carcinoma cells. J. Cell Biol. 96: 1690-1696. Holland, P. W.H., Harper, S.J., McVey, J.H., and Hogan, B.L.M. 1987. In vivo expression of mRNA for the C a t +-binding protein SPARC (osteonectin) revealed by in situ hybridization. J. Cell Biol. 105: 473-482. Hyafil, F., Babinet, C., and Jacob, F. 1981. Cell-cell interactions in early embryogenesis: a molecular approach to the role of calcium. Cell, 26: 447-454. Kretsinger, R.H. 1979. The informational role of calcium in the cytosol. Adv. Cyclic Nucleotide Res. 11: 1-26. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London), 277: 680-685. Lane, T.F., and Sage, E.H. 1990. Functional mapping of SPARC: peptides from two distinct Ca++-binding sites modulate cell shape. J. Cell Biol. 111: 3065-3076. Malaval, L., Fournier, B., and Delmas, P.D. 1987. Radioimmunoassay for osteonectin. Concentrations in bone, nonmineralized tissues and blood. J. Bone Miner. Res. 2: 457-465. Mann, K., Deutzmann, R., Paulsson, M., and Timpl, R. 1987. Solubilization of protein BM-40 from a basement membrane with

1992

chelating agents and evidence for its identity with osteonectin and SPARC. FEBS Lett. 218: 167-172. Marguerie, G., Chagniel, G., and Suscillon, M. 1977. The binding of calcium to bovine fibrinogen. Biochim. Biophys. Acta, 490: 94-103. Mason, I.J., Taylor, A., Williams, J.G., Sage, H., and Hogan, B.L.M. 1986. Evidence from molecular cloning that SPARC, a major product of mouse parietal endoderm, is related to an endothelial cell "culture shock" glycoprotein of M, 43 000. EMBO J. 5: 1465-1472. Means, A.R., and Rasmussen, C.D. 1988. Calcium, calmodulin and cell proliferation. Cell Calcium, 9: 3 13-3 19. Norris, W.D., Steele, J.G., Johnson, G., and Underwood, P.A. 1990. Serum enhancement of human endothelial cell attachment to and spreading on collagens I and IV does not require serum fibronectin or vitronectin. J. Cell Sci. 95: 255-262. Ozawa, M., Engel, J., and Kemler, R. 1990. Single amino acid substitutions in one ca2+binding site of uvomorulin abolish the adhesive function. Cell, 63: 1033-1038. Peyrieras, N., Hyafil, F., Louvard, D., Ploegh, H.L., and Jacob, F. 1983. Uvomorulin: a nonintegral membrane protein of early mouse embryo. Proc. Natl. Acad. Sci. U.S.A. 80: 6274-6277. Ringwald, M., Schuh, R., Vestweber, D., Eistetter, H., Lottspeich, F., Engel, J., Dolz, R., Jiihnig, F., Epplen, J., Mayer, S., Miiller, C., and Kemler, R. 1987. The structure of cell adhesion molecule uvomorulin. Insights into the molecular mechanism of ca2+-dependentcell adhesion. EMBO J. 6: 3647-3653. Romberg, R.W., Werness, P.G., Lollar, P., Lawrence-Riggs, B., and Mann, K.G. 1985. Isolation and characterization of native adult osteonectin. J. Biol. Chem. 260: 2728-2736. Sage, E.H. 1991. Modulation of endothelial cell shape by SPARC does not involve chelation of extracellular c a 2 + and ~ g ' + . Biochem. Cell Biol. 70: 56-62. Sage, E.H., and Bornstein, P. 1991. Extracellular proteins that modulate cell-matrix interactions: SPARC, tenascin and thrombospondin. J. Biol. Chem. 266: 14 831 - 14 834. Sage, E.H., Vernon, R.B., Decker, J., Funk, S., and Iruela-Arispe, M.L. 1989a. Distribution of the calcium-bindingprotein SPARC in tissues of embryonic and adult mice. J. Histochem. Cytochem. 37: 819-829. Sage, E.H., Vernon, R.B., Decker, J., Funk, S., Everitt, E.A., and Angello, J. 1989b. SPARC, a secreted protein associated with cellular proliferation, inhibits cell spreading in vitro and exhibits ca2+-dependent binding to the extracellular matrix. J. Cell Biol. 109: 341-356. Schuh, R., Vestweber, D., Riede, I., Ringwald, M., Rosenberg, U.B., Jackle, H., and Kemler, R. 1986. Molecular cloning of the mouse cell adhesion molecule uvomorulin: cDNA contains a B1-related sequence. Proc. Natl. Acad. Sci. U.S.A. 83: 1364-1368. Smith, J.W., and Cheresh, D.A. 1988. The Arg-Gly-Asp binding domain of the vitronectin receptor. J. Biol. Chem. 263: 18 726 - 18 731. Strickland, S., Smith, K.K., and Marotti, K.R. 1980. The hormonal induction of differentiation in teratocarcinoma stem cells: generation of parietal endoderm by retinoic acid and dibutyryl CAMP. Cell, 21: 347-355. Van Doren, K., Hanahan, D., and Gluzman, Y. 1984. Infection of eukaryotic cells by helper-independent recombinant adenoviruses: early region 1 is not obligatory for integration of viral DNA. J. Virol. 50: 606-614. Vernon, R.B., and Sage, H. 1989. The calcium-binding protein SPARC is secreted by Leydig and Sertoli cells of the adult mouse testis. Biol. Reprod. 40: 1329-1340. Vestweber, D., and Kemler, R. 1984. Rabbit antiserum against a purified surface glycoprotein decompacts mouse preimplantation embryos and reacts with specific adult tissues. Exp. Cell Res. 152: 169-178. Vestweber, D., and Kemler, R. 1985. Identification of a putative cell adhesion domain of uvomorulin. EMBO J. 4: 3393-3398.

EVERI'lT AND SAGE


Wewer, U.M., AIbrechtsen, R., Fisher, L.W., Young, M.F., and Terrnine, J.D. 1988. Osteonectin/SPARC/BM-40 in human decidua and carcinoma, tissues characterized by de novo formation of basement membrane. Am. J. Pathol. 132: 147-157. Yoshida. C., and Takeichi, M. 1982. Teratocarcinoma cell adhesion: identification of a cell-surface protein involved in calciumdependent cell aggregation. Cell, 28: 217-224.

1379

Yoshida-Noro, C., Suzuki, N., and Takeichi, M. 1984, Molecular nature of the calcium-dependent d-cell adhesion system. Mouse teratocarcinoma and embryonic cells studied with a monoclonal antibody. Dev. Biol. 101: 19-27.

Expression of SPARC is correlated with altered morphologies in transfected F9 embryonal carcinoma cells.

Distance effect of matrix attachment regions on transgene expression in stably transfected Chinese hamster ovary cells.

SPARC Expression Is Selectively Suppressed in Tumor Initiating Urospheres Isolated from As+3- and Cd+2-Transformed Human Urothelial Cells (UROtsa) Stably Transfected with SPARC.

Connexin32 gap junction channels in stably transfected cells. Equilibrium and kinetic properties.

Determination of transgene copy number in stably transfected mammalian cells by PCR-capillary electrophoresis assay.

Transepithelial resistance in human bestrophin-1 stably transfected Madin-Darby canine kidney cells.

Signal transduction by the platelet integrin alpha IIb beta 3: induction of calcium oscillations required for protein-tyrosine phosphorylation and ligand-induced spreading of stably transfected cells.

Fibroblast-conditioned medium contains cell surface proteins required for cell attachment and spreading.

Extracellular matrix and hormones transcriptionally regulate bovine beta-casein 5' sequences in stably transfected mouse mammary cells.

Gating of gap junction channels as revealed in cells stably transfected with wild type and mutant connexin cDNAs.

SPARC overexpression alters microRNA expression profiles involved in tumor progression.

Mechanism of regulation of thyrotropin-releasing hormone receptor messenger ribonucleic acid in stably transfected rat pituitary cells.

Establishment of stably transfected cells constitutively expressing the full-length and truncated antigenic proteins of two genetically distinct mink astroviruses.

Myostatin knockdown and its effect on myogenic gene expression program in stably transfected goat myoblasts.

Inhibition of cAMP accumulation by intracellular calcium mobilization in C6-2B cells stably transfected with substance K receptor cDNA.

Matrix attachment region combinations increase transgene expression in transfected Chinese hamster ovary cells.

Assembly intermediates of the mouse muscle nicotinic acetylcholine receptor in stably transfected fibroblasts.

Targeted surface expression of an exogenous antigen in stably transfected Babesia bovis.

Inositol trisphosphate mediates cloned muscarinic receptor-activated conductances in transfected mouse fibroblast A9 L cells.

Transmembrane signalling by the N-formyl peptide receptor in stably transfected fibroblasts.

Attachment and spreading of baby hamster kidney cells to collagen substrata: effects of cold-insoluble globulin.

Radiolocalisation and imaging of stably HPLAP-transfected MO4 tumours with monoclonal antibodies and fragments.

Secreted Protein Acidic and Rich in Cysteine (SPARC) Mediates Metastatic Dormancy of Prostate Cancer in Bone.

Maintenance and use of F9 teratocarcinoma cells.