[ RESEARCH 52, 6877-6884, December 15, 1992] Regulation of Smooth Muscle a-Actin Promoter in ras-transformed Cells: Usefulness for Setting Up Reporter Gene-based Assay System for Drug Screening

C. Chandra Kumar, ~ Pierre Bushel, Sheela Mohan-Peterson, 2 and Fernando Ramirez Department of Tumor Biology, Schering-Plough Research Institute, Bloomfield, New Jersey 07003

ABSTRACT cancer have impaired GTPase activity, are insensitive to GAP, and remain bound to GTP leading to uncontrolled cell growth Oncogenic activation of ras results in changes in the transcription of (6). In addition, a number of posttranslational modifications of several genes leading to uncontrolled cell growth. In this paper, we ras such as palmitoylation, farnesylation, proteolytic cleavage, demonstrate that transformation of fibroblast cells by the ras oncogene leads to transcriptional repression of the smooth muscle a-actin pro- and carboxymethylation have been found to be essential for moter. Transient transfection analysis of plasmids containing the 5' biological activity of ras (9-11). upstream region of the human a-actin gene fused to human growth A number of approaches are being taken to find drugs that hormone or bacterial chloramphenicol acetyltransferase coding se- inhibit the accumulation of the GTP-bound form of the ras p21 quences into Rat-2 and ras-transformed Rat-2 (HO6) cells indicates that protein. These include assay systems designed to identify drugs a-actin promoter is repressed in ras-transformed cells. In addition, that inhibit the nucleotide exchange activity of ras, or stimulate stable rat fibroblast cell lines expressing human growth hormone or the GTPase activity of mutant ras, or inhibit ras:GAP interac- B-galactosidase under the control of ~,-actin promoter exhibit repressed tion, etc. (11). Recently, the involved in the modifica- reporter gene activity following transformation by the ras oncogene. tion of ras proteins have also been targeted for new drug dis- a-Actin promoter-driven B-galactosidase activity is derepressed in re- vertants of ras-transformed stable cell lines. This revertant cell line covery (11). Most of these assay systems are in vitro based, and expresses elevated levels of ras p21 protein and is resistant to retrans- the lead compounds in these assay systems have to be evaluated formation by Ki and Ha-ras oncogenes. The revertant may have either in secondary cell-based assay systems to determine their effi- a defective target protein whose activity is essential for the transforming cacy. In this paper, we describe a general cell-based assay sys- activity of ras or an activated tumor suppressor gene which can suppress tem capable of detecting compounds that inhibit the transform- the activity of ras. These results indicate that smooth muscle a-actin ing activity of ras, either directly or indirectly. promoter activity is a sensitive marker to follow phenotypic changes Previous studies have shown that human sm forms of myo- following transformation by ras and subsequent reversion. The advan- sin light chain 2 (MLC-2) and a-actin mRNA and protein tages of this a-actin promoter-reporter gene assay system to screen for levels are repressed in ras-transformed fibroblast cells (12, 13). drugs that inhibit the transforming activity of ras, either directly or These changes are part of the cytoskeletal changes that occur indirectly, are discussed. following neoplastic transformation of fibroblast cells (14). We have also shown that revertants of ras-transformed cells show INTRODUCTION normal levels of MLC-2 gene expression, suggesting that the expression of these cytoskeletal markers, such as sm MLC-2 ras proteins play a central role in neoplasia and growth con- and sm a-actin, is modulated by the ras-transformed pheno- trol. Activated ras oncogenes have been implicated in the onset type (12). In this paper, using plasmids containing 5' upstream of 20% of human and greater than 90% of pancreatic sequences of the human a-actin gene fused to different reporter and colon carcinomas (1-3). Hence, drugs that interfere with genes, we first demonstrate that a 906-base pair fragment is the function of ras, either directly or indirectly, will be useful in the treatment of a number of cancers. A great deal of effort has sufficient and necessary to confer responsiveness to the ras oncogene. Using stable rat fibroblast cell lines that express been focused in understanding the biochemical and structural characteristics of the ras p21 protein (4, 5). The mechanisms by LacZ reporter gene activity from the a-actin promoter, we which ras can induce transformation of cells have also been further show that revertants of ras-transformed cells exhibit studied in great detail (6). ras p21 protein exhibits GTPase derepression of promoter activity. The revertant cell line ex- activity and exists either in GTP-bound activated form or in presses elevated levels of the ras p21 protein and is resistant GDP-bound inactive form. Proteins that modulate the activity to retransformation by Ki and Ha-ras oncogenes. The useful- of ras consist of (a) a nucleotide exchange factor that facilitates ness of this promoter system to identify agents that can inhibit the exchange of GTP/GDP and thus promotes the activation of the transforming activity of ras, either directly or indirectly, is ras (7) and (b) a protein that acts as a GAP 3 and promotes discussed. inactivation of ras (8). Mutant forms of ras found in human MATERIALS AND METHODS

Received 6/19/92; accepted 9/30/92. Cell Culture and Transfections The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accord- Rat-2 fibroblast cells obtained from the American Type Culture Col- ance with 18 U.S.C. Section 1734 solely to indicate this fact. lection (CRL 1764) and stably transfected derivatives were grown in 1To whom requests for reprints should be addressed. 2 Present address: DNAX Research Institute, Palo Alto, CA 94304. DMEM supplemented with 2 mM glutamine and 10% (vol/vol) fetal 3 The abbreviations used are: GAP, GTPase-activating protein; sm, smooth bovine serum. Rat-6 fibroblasts, obtained from Dr. Bernard Weinstein muscle; DMEM, Dulbecco's modified Eagle's medium; TK, thymidine kinase; at Columbia University, were grown in DMEM supplemented with 2 PBS, phosphate-buffered saline; CAT, chloramphenicol acetyltransferase; hGH, human growth hormone; X-gal, 5-bromo-7-chloro-3-indolyl-~-i)-galactopyrano- mM glutamine and 10% bovine calf serum. To derive stable transfor- side; ONPG, O-nitrophenyl-/3-o-galactopyranoside; FITC, fluorescein isothiocyo- mants, cells were transfected with 10:1 mixtures of the reporter gene nate; IgG, immunoglobulin G; CHP, 4-cis-hydroxy-L-proline. plasmid (see below and Fig. 1) and plasmid pIBW DNA containing a 6877 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. REGULATION OF SMOOTH MUSCLE a-ACTIN PROMOTER IN ras-TRANSFORMED CELLS neomycin resistance marker fused to the herpes simplex virus TK pro- were washed extensively with PBS solution and photographed using a moter. Briefly, 5 x 105 cells, plated in 100-mm dishes, were incubated Zeiss IM-35 fluorescent microscope equipped with a 200-W mercury for 6 h with calcium phosphate DNA precipitates. Calcium phosphate light source. precipitates were prepared with 20 ~tg of plasmid DNA per 10-cm plate (15). The transfection was stopped by replacing the medium with the Isolation and Characterization of Revertant Cell Line standard culture medium after a brief rinse with PBS. Twenty-four h later, transfected cultures were split 1:10 and grown in the presence of To isolate the revertant cell line D-3, Rat-6 (2S-HO6) cells were the neomycin analogue G418 (Sigma) at 400 #g/ml. After 10 days, seeded at a density of 1 x 106 cells/100-mm dish and treated with 7 resistant clones were isolated using cloning cylinders. ~g/ml of 5-aza-cytidine for 24 h. After a recovery period of an addi- tional 48 h, cells were plated at a density of 1 x 105 cells for each 100-mm dish in the growth medium containing 200/~g/ml of CHP Construction of Plasmids (23, 24). The plates were fluid changed with DMEM containing CHP, The EcoRI-DraIII fragment consisting of-894 to + 12 base pairs of twice weekly for 3 to 4 wk, after which colonies made up of apparently the human a-actin gene was isolated from the paAS plasmid (16) and nontransformed cells were marked and isolated using cloning cylinders. ligated into the HindIII site of plasmids pCH 126 (17), pOGH (18), and Cells were further subcloned using the limiting dilution technique. pBasic (19) to derive paAP126, paAPGH, and paAPCAT plasmids, To determine the susceptibility of cells for transformation by retro- respectively. The structure of the plasmids was confirmed by DNA viruses containing Kirsten and Harvey ras oncogenes, Rat-6 (2S) and sequence analysis, pBasic and pOGH are promoterless plasmids con- D-3 revertant cells were plated at a density of 2 x 105 ceils in 100-mm taining CAT and hGH coding sequences, respectively (18, 19). Plasmid dishes, and virus stocks (a gift from Dr. Bassin, National Cancer Insti- pHO6T1 containing the T24 Ha-ras oncogene was a kind gift from Dr. tute) diluted with the growth medium were added to the cells in the Earl Ruley (Massachusetts Institute of Technology) (20). presence of 200 ug/ml of DEAE-dextran. Following infection for 2 h, the medium containing the virus was aspirated and replaced with fresh medium. The cells were fed with fresh medium at 3-day intervals, and Assays transformed foci were scored after 12 days by fixing the cells in meth- /~-Galactosidase Assay. To identify stable fibroblast cell lines ex- anol and staining with Giemsa reagent as described (24). pressing ~-galactosidase activity, cells were fixed for 30 min in 2% glutaraldehyde in 0.1 M piperazine-N-N'-bis[2-ethane-sulfonic acid] RESULTS buffer (pH 6.9) containing 1 mM MgCI2 and then incubated in a solu- tion containing 0.2% X-gal, 10 mM sodium phosphate buffer (pH 7.0), Transient Transfection Analysis of Plasmids Containing 150 mM sodium chloride, 1 mM MgCI2, and 3.3 mM K4Fe(CN)6. The a-Actin Promoter Fused to Reporter Genes in Rat-2 and ras- incubation was carried out at 370C for 12 h (21). transformed Rat-2 (HO6) Cells. The structures of the plasmids For the rapid and quantitative assessment of ~-galactosidase expres- used in the present study are shown in Fig. 1. A 906-base pair sion, the soluble substrate ONPG was used. Cells grown in 96-well microtiter plates were rinsed with PBS solution and lysed (50 gl/well) EcoRI-DralII fragment comprising the 5' upstream region of with 50 mM sodium phosphate buffer (pH 7.0) containing 5 mM B-mer- the human a-actin gene was fused to hGH, bacterial CAT, and captoethanol and 5% Nonidet P-40 for 15 min at room temperature. E. coli-derived/3-galactosidase (LacZ) coding sequences to de- The cell lysates were incubated with ONPG (1 mg/ml) in PBS solution rive paAPGH, paAPCAT, and paAP126 plasmids, respective- containing 1 mM MgCI2 (1 mg/ml). The reaction was terminated by ly. Plasmids pBasic and pOGH are promoterless vectors con- using 10 mM sodium bicarbonate solution (50 gl/well), and the absorp- taining CAT and hGH coding sequences, respectively. Plasmid tion was measured at 420 nm using a Titertek multiplate reader. pTKGH, containing TK promoter fused to hGH, and plasmid CAT Assay. The cell extracts were prepared 72 h after the transfec- pLW2 in which herpes immediate early promoter is fused to tion, and the CAT assays were performed as described (19) in a total CAT coding sequences were used as positive controls in the volume of 0.10 ml containing 250 mM Tris-HC! (pH 7.9), 4.4 mM acetyl transfection experiments. These plasmids were transfected into , 0.2 #Ci of [14C]chloramphenicol (40 to 60 Ci/mmol) (New the Rat-2 and Ha-ras-transformed derivative, Rat-2 (HO6) cell England Nuclear) (1 Ci -- 37 GBq), and cell extracts containing 50 #g of protein. The reactions were allowed to proceed for 30 min at 370C, lines. The results of transient transfection analysis using hGH and the analysis of the reaction was performed as described (19). The and CAT reporter gene plasmids, shown in Fig. 2, indicate that CAT activity was quantitated by counting regions of the chromatogram a-actin promoter-driven reporter gene activity is repressed in a liquid scintillation counter. about 9- and 30-fold, respectively, in ras-transformed Rat-2 Human Growth Hormone Assay. hGH secretion was measured in (HO6) cells compared to normal Rat-2 cells. The control plas- the supernatants of cell cultures using the Allegro HGH system from mid pTKGH exhibits 37% reduction in hGH secretion in ras- Nichols Institute (San Juan Capistrano, CA). Briefly, the procedure transformed cells compared to normal cells. The analysis of involved the use of two monoclonal antibodies against hGH to form a control plasmids pLW2 and pTKGH in these two cell lines soluble sandwich complex. One of the monoclonal antibodies is radio- indicates that the changes in reporter gene activities, observed labeled for detection, while the other monoclonal antibody is coupled to in ras-transformed cells, are specific to the a-actin promoter. . The addition to the reaction mixture of an avidin-coated plastic a-Actin Promoter-driven Reporter Gene Activity Is Re- bead allows for a specific and efficient binding of the sandwich complex pressed in Stable Rat Fibroblast Cells following Transforma- to a solid phase via the high-affinity interaction between biotin and avidin. These beads are then washed and counted in a gamma counter. tion by the ras Oncogene. We have isolated stable rat fibroblast cell lines that express hGH and/3-galactosidase activity from lmmunofluorescent Staining of ras p21 Protein the a-actin promoter. Subconfluent Rat-2 and Rat-6 fibroblast cells were cotransfected with either paAPGH or paAP 126 plas- Cells were grown in chamber slides overnight and fixed with acetone mid with plasmid plBW in which the aminoglycoside trans- for 10 min at -20~ Samples were incubated with purified rat mono- clonals against ras p21 protein (Y13-259; Oncogene Sciences, Min- ferase (neo § coding sequences are fused to TK promoter. The neola, NY) for 20 min (22). Bound antibody was detected by incubation Rat-2 or Rat-6 cells were selected for their ability to grow in the for 20 min with FITC-conjugated goat antiserum raised against rat IgG presence of the neomycin analogue G418. Individual clones (Cappel Laboratories, Durham, NC). As a control, cells were incubated were tested for the expression of hGH and ~-gal activity. In the with PBS solution containing IgG. After each incubation step, slides case of cells transfected with the LacZ reporter gene plasmid 6878 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. REGULATION OF SMOOTH MUSCLE a-ACTIN PROMOTER IN ras-TRANSFORMED CELLS

Test Plasmids Control Plasmids

A p li' H Sa Xb po.APCAT -~ -~-aeC~ [~ CAT p pBasic [ [ [ I CAT ~- -894 +12 A p Pp Av li' I,_ poAPGH -f o..o h" p PLW2 --~ IE ~'~ CAT '~-- -894 +12 -91 +119 A p I P" I.. H Sa Xb n poAP126 P p~GH l t ' I I hGH p -894 +12

E M

-f TK hGH -149 +51 Fig. 1. Various reporter gene plasmids used in the present study. A 906-base pair EcoRI-DraIII fragment, containing 5' upstream sequences of the human a-actin gene, was blunt ended and subcloned into the HindIII site of pCHI26 (LacZ), pBasic, and pOGH plasmids to derive paAP126, paAPCAT, and paAPGH plasmids, respectively. Plasmid pLW2 containing herpes immediate early promoter fused to CAT coding sequences and pTK GH containing thymidine kinase promoter fused to hGH coding sequences were used as positivecontrols in transfectionexperiments. Plasmids pBasic and pC)GH are promoterless plasmids containingCAT and hGH coding sequences, respectively,and were used as negativecontrols. Restriction sites are E, EcoRI; A, Accl; P, PstI; B, BamMI; M, Mlul, H, HindIII; Sa, Sall; Xb, XbaI; ,4v, Aval. The numbers correspond to the base pairs relative to the transcription start site.

a) 80-, b) Fig. 2. Transient transfection analysis of hGH and CAT reporter-gene plasmids in ~4 pLW2 pBasic po~ACAT Rat-2 and ras-transformed Rat-2 (HO6) cells. Approximately 5 • 105 cells were transfected Cell line with various plasmid DNAs indicated in the 60 [] Rat-2cells figure. In a, 20 ,g of paAPGH and pTKGH _0' [] Rat-2HO6 cells were used for transfecting Rat-2 and Rat-2 o| (HO6) cells. Cells were exposed to DNA/ calcium phosphate precipitates for 6 h before = rinsing them with PBS twice and adding fresh ~ 40 medium. GH levels in the supernatants were -7- determined 72 h later as described in "Mate- o rials and Methods" and corrected for the via- ble cells at the time of the assay. The number 20 of viable cells was determined by using trypan tt tt blue dye. In b, 20 .g of pBasic and paACAT and 2 .g ofpLW2 plasmids were used for each o ~ o o transfection experiment. CAT activities were determined in cell extracts with equal amounts 0 pTI

(paAP126), a duplicate set of G418-resistant colonies were Fig. 3, a and b, respectively. The three stable cell lines stained with X-gal substrate to identify the ones expressing (APGHB2, Y2, and 2S) were transformed with the Ha-ras on- ~-gal activity. In an independent experiment, cotransfection of cogene by transfecting them with the plasmid pHO6T1 that promoterless vector plasmid pCH126 with plasmid pIBW gave contains human bladder carcinoma-derived Ha-ras oncogene as rise to G418-resistant colonies that remained colorless when well as cis-acting enhancer sequences from SV40 and Moloney stained with X-gal substrate, indicating that the expression of murine sarcoma virus (20). Transformed foci were isolated and ~-gal activity is dependent on the promoter sequences (data not expanded. Analysis of hGH secretion in the supernatants of shown). The level of hGH secreted or ~-gal activity expressed by four independent ras transformants of the APGH B2 cell line different clones differed somewhat among several cell lines. (APGH H2, M2, N4, and F2) (Fig. 3a) indicates that it is One clone, APGHB2, that was characterized by a higher level repressed in all the ras transformants. Similarly, analysis of ten of hGH secretion was selected for further experiments de- different transformed foci from each stable Rat-2 (Y2) and scribed below. Similarly, clones Y2 and 2S, expressing B-gal Rat-6 (2S) clone revealed that the B-gal activity is repressed in activity under the control of a-actin promoter, were selected all the ras transformants. Typical results of two ras transfor- among the stable clones of Rat-2 and Rat-6 cells, respectively. mants of Y2 and 2S clones Rat-2 (Y2-HO6) and Rat-6 (2S- The level of hGH secreted by the APGHB2 clone and the B-gal HO6) are shown in Fig. 3b. These results clearly demonstrate activities expressed by Y2 and 2S clonal cell lines are shown in that sm ot-actin promoter is repressed following transformation 6879 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. REGULATION OF SMOOTH MUSCLE a-ACTIN PROMOTER IN ras-TRANSFORMED CELLS a) b)

20O 0.3-

,= I I I

m

Cell line E 0.2, e.

e~ O o N =-m ~- Rat-2 Y2 [] APGHH2 E ~- lOO a //// Rat-2(Y2-Ho6) O [] APGHM2 .c d - - "0- - Rat-6 (2S) [] APGHN4 "--'O"- Rat-6(2S-Ho6) [] ,V'GHF2 i I 0.1

9 '<1" l~ .=_.

0 ~ 0.0 I I I 2 5 2 4 6

Days after plating Days after plating Fig. 3. Analysis of a-actin promoter-driven hGH and B-gai activities in stable rat fibroblast cell lines before and after transformation by the ras oncogene. Stable rat fibroblast cell lines expressing hGH and ~-galactosidase activity were identified using the hGH radioimmunoassay and the X-gal staining procedure, respectively. Stable Rat-2 cells expressing hGH (APGHB2) (a) or ~-gal activity [Rat-2 (}'2)] and Rat-6 cells expressing/3-gal activity [Rat-6 (2S)] (b) were transformed with the Ha-ras oncogene, and transformed foci were isolated and expanded for further analysis of hGH and ~-gal activities. In a, APGH H2, APGH 3112, APGH N4, and APGH 1:2 are four independent ras transformants of the APGH B2 clone. In b, Rat-2 (Y2-H06) and Rat-6 (2S-H06) are ras transformants of Rat-2 (Y2) and Rat-6 (2S) clones, respectively. GH levels were measured as described in the legend to Fig. 2. ~-Galactosidase activity was measured by plating cells at a density 1 • 103 cells/well in a 96-well microtiter plate and monitoring the conversion of ONPG at 420 nm on various days. The typical results of several independent experiments are shown in the figure. O.D. absorbance. by the ras oncogene and that the 906-base pair region upstream subcloning of the D-3 cell line. The phase-contrast micrographs of the transcription start site contains all the information nec- of the revertant D-3 cell line together with the parental Rat-6 essary to confer ras responsiveness. (2S) and (2S-HO6) cell lines shown in Fig. 4a indicate that the Derepression of a-Actin Promoter-driven Reporter Gene revertant is morphologically altered and exhibits a contact- Activity in a Revertant of the ras-transformed Cell Line. To inhibited growth pattern. The B-gal activity measurements of determine whether the repression of a-actin promoter-driven the different cell lines shown in Fig. 4b indicate that the a-actin reporter gene activity is reversible, a revertant of the ras-trans- promoter activity is derepressed in the revertants, suggesting formed stable Rat-6 (2S-HO6) cell line, expressing ~-gal activ- that its activity is modulated by phenotypic changes. ity under the influence of a-actin promoter, was isolated. The Characterization of the Revertant Cell Line. Next, we isolation of the revertant of ras-transformed Rat-6 (2S-HO6) wanted to examine the levels of ras p21 protein in the revertant cells involved treating the cells with 5-aza-cytidine as a mutagen D-3 cell line and compare them with those of the parental cell and selecting the flat revertants after selection with CHP, as lines Rat-6 (2S) and (2S-HO6). The level of ras p21 protein was described by Bassin and Benade (23). Approximately 1 --~ 106 assessed by indirect immunofluorescence using monoclonal an- Rat-6 (2S-HO6) cells were treated with 5-aza-cytidine for 24 h tibody Y13-259 (22). As shown in Fig. 5, the revertant clone and allowed to recover for 2 days before fluid changing into a D-3 contains an elevated level of ras p21 protein similar to the medium containing 200 ~tg/ml of CHP. After 4 wk of selection ras-transformed Rat-6 (2S-HO6) cell line. These results suggest with CHP, flat colonies consisting of apparently nontrans- that the revertant may have one or more defective target pro- formed cells were identified microscopically. These flat colonies teins that are required for transformation by ras or they may were isolated using cloning cylinders and further subcloned have acquired an activated tumor suppressor gene that is capa- using the limiting dilution technique. One revertant line D-3, ble of suppressing the transforming activity of ras. Alterna- which showed stable flat morphology over several passages, was tively, the overexpressed ras p21 protein may be biologically expanded, and another subclone D-3A was obtained by further inactive. To analyze these possibilities, we wanted to determine 6880 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. REGULATION OF SMOOTH MUSCLE a-ACTIN PROMOTER IN ras-TRANSFORMED CELLS

Rat-6(2S) Untransformed (Parental) (2S-806) Transformed (D-a) Revertant

0.20 -

D-3 A Subclone (Revertant) D-3 (Revertant) Rat-6 (2S) (Untransformed) ~" 0.14 / C ,/ 0 04 i/ A a o / / "-">, 0.10

< / ,r

/ (2S - H06) (Transformed) 0.04

I i I I I I I I I 2 4 6 9 Days After Plating Fig. 4. Derepression of a-actin promoter-drivenB-gal activity in a revertant of ras-transformed Rat-6 (2S-HO6) cells, ras-transformed Rat-6 (2S-HO6) cells were mutagenized with 5-aza-cytidine and grown in medium containing 200 t~g/ml of CHP. Revertant clones, surviving CHP selection and exhibiting consistent flat morphology,were isolated, expanded, and further subclonedby the limiting dilution technique. D-3 is a revertant clone isolated from ras-transformedRat-6 (2S-HO6) cells, and D-3 .4 is a subclone of the revertant isolated by the limiting dilution technique. The phase-contrast micrographs of the different cells are shown at a magnification of 1 x 100. B-Galactosidase activity was measured as described in "Materials and Methods." After 6 days of incubation, the ras-transformed 2S-HO6 cells overgrow the well and peel off from the plate.

if the revertant cell line is susceptible to retransformation by tiffed by their ability to derepress the a-actin promoter-driven ms. If the revertant displays a defect in transmitting the signal reporter gene activity in these ras-transformed cells. derived from the oncogenic ras protein, it should be capable of resisting retransformation by the ras oncogene. To examine DISCUSSION this, Rat-6 (2S) and D-3 cells were plated in 100-mm dishes and infected with different dilutions of retrovirus stocks containing In this paper, we show that the 906-base pair fragment posi- Ki-ras and Ha-ras oncogene. Foci were scored after 12 days. tioned 5' to the transcription start site of the human sm a-actin The results shown in Table 1 demonstrate that the revertant cell gene has all the information necessary for mediating its tran- line D-3 is able to resist retransformation by ras oncogenes. scriptional repression in ras-transformed cells. Revertants of Hence, the revertant cell line may have a defective target pro- transformed cells show derepression of the a-actin promoter tein whose activity is essential for the transforming activity of activity indicating that the pr0m0ter activity can be modulated ras or has an activated suppressor gene that can negate the concomitantly by phenotypic changes in fibroblast cells. The activity of ras. In either case, it is clear that a-actin promoter- revertant cell line that we have isolated expresses elevated levels driven B-gal activity is derepressed in revertants of ras-trans- of presumably biologically active ras p21 protein and resists formed cells irrespective of the mechanisms by which revertants retransformation by the ras oncogene. The revertants could arise. These results suggest that agents that inhibit the trans- arise by alterations in protein molecules whose activity is an forming activity of ras, either directly or indirectly, can be iden- indispensable requirement for transformation by ras and other 6881 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. REGULATION OF SMOOTH MUSCLE a-ACTIN PROMOTER IN ras-TRANSFORMED CELLS Rat-6(2S) (2S-HO6) Transformed

D-3 Revertant Conjugate (Control)

Fig. 5. l nmmnol'luoresccncv,st~tining of ras p21 protein in Rat-e, (25). ra~-trallslormed 12S-tt()~). and rc~crhml I) ~ ,eli lines. (ells. gro~vn in chamber slides, ~ere fi\ed ,~ith acetone and incubated ~ith rat monoclonal antibodie, ih 13-259)aK4inst ras p21 protein (()ncogcnc Science',. \l~nne~)la, "x.~ }. Bound antibody ~as detected b\ incubation ~ith FIT( conlt, e,aled goat ;mti-lg(; rai,,cd against rat Ia(i. (ell, x~cre ph~)t(~raphcd usin~ 'a Zeis~ Im ~, lluorcscenI microscope..-Xs a control. Rat-6 (2N) ccll~ were incut~'atcd ~ ith PBN soluli(m containina nil let; ahmc t,cl~*rc addin~ I-lI ( conju.aah'd goat anti ral 1~2{, :~ntibodies. oncogenes..\lternativelx, the revertants could also arise b; the the signal transduclion path~ a.~ initiated by activated oncogene action of dominant tumor suppressor genes ~vhosc products products. \\e ha~e rcccnllx shov, n that revertants of ras-trans- may suppress the synthesis or function of oncogencs or other formed K-t lOS cells, which lack the viral ras sequences, exhibit target genes (23). In almost every stud\ in ~hich resistance of derepression of sm XlI,C-2 mRNA levels (12). Similarly, a'-ac- revcrtants to retransformation has been measured, lhe rcxertant tin mRNA expression ~as found io be restored in a flat rever- cells have become resistant not only to the parental transform- rant of Ha-ras oncogene-transforined NIH/3T3 cells, which is ing gene and other structurally related oncogenes, but also to resistant to transformation by various oncogenes (26). Even others that share neither structural homolog.~ nor apparent though we have not analyzed a-actin promoter activity in a functional homologv. This is usually interpreted to mean lhat series of rcvertants, it seen> reasonable to conclude that sm the resisted oncogenes function by v,a~ of convergent pathways. ~,-actin promoter activity is derepressed in revertant cells irre- These results also indicate that altering or inhibiting one or, at spective of the mechanisms by which they arise. most, a few critical functions may result in resistance to many The major application of this system, we believe, would be oncogenes. Recently, a/tans-dominant deletion mutailt of x-jun the identification of agents that can suppress the ras-trans- lacking its activation domain has been shown to haxe a tumor formed phenot.vpe either bx acting directly on the ras p21 pro- suppressor activity against Ira-transformed cells (25). It ~as tcin c~r indircctlx on other molecules in the cell which mediate hypothesized that this deletion mutant ot'/~,z could block trans- the transforming activity of ra.s..-ks discussed above, since many formation by competing with c-jun to bind to c-fos. These re- oncogenes stein to function by way of convergent pathways, suits indicate that revertants could also arise by inhibition of the this assay system would be capable of identifying a broad range activities of transcription factors that mediate or participate in of compounds which can inhibit the transforming activity of not 6882 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. REGULATION OF SMOOTH MUSCLE ~-ACTIN PROMOTER IN ras-TRANSFORMED CELLS

Table 1 Resistance of revertant (D-3) cells to retransformation by the tion but are specific to the oncogenes (12). Thus, it is reasonable Ki- and Ha-ras oncogenes to propose that transformation by specific oncogenes initiates a Rat-6 (2S) and D-3 cells were infected with different dilutions of Harvey (Ha-MuSV) and Kirsten murine sarcoma virus (Ki-MuSV) stocks in the presence cascade of events that ultimately converge, resulting in changes of 200 #g/ml of DEAE-dextran. Following infection for 2 h, the medium was in the activities of specific transcription factors required for the aspirated and replaced with fresh medium. The cells were fed with fresh medium expression of specific genes. Identification of cis- and trans- at 3-day intervals, and transformed foci were scored after 12 days by staining with Giemsa reagent. acting factors that mediate the down regulation of a-actin gene No. of transformed foci expression may lead to important insights into the molecular mechanisms involved in transformation by ras. Virus dilutions Control Virus stock Cell line (uninfected) 10-1 10 -2 10 -3 ACKNOWLEDGMENTS Ki-MuSV Rat-6 (2S) 0 <200 <200 80 D-3 5 8 6 6 We are grateful to Dr. Earl Ruley for providing plasmids paAS, Ha-MuSV Rat-6 (2S) 0 <200 80 12 paAPCAT, and pHO6TI and also for many valuable suggestions dur- D-3 6 6 7 5 ing the course of this study. We wish to thank Dr. Yoshito Kaziro, Dr. Louis Perkins, and Dr. Angela Yang-Yen for their comments on the manuscript; Dr. Claude Nash and Joseph Catino for support and en- just ras alone, but a number of other oncogenes as well. These couragement; and Rita Cunniff for typing the manuscript. agents, which could be small molecules or genetic sequences (either complementary DNA or genomic sequences), can be REFERENCES identified by their ability to derepress the a-actin promoter- 1. Barbacid, M. ras oncogenes: their role in neoplasia. Eur. J. Clin. Invest., 20: driven reporter gene activity in ras-transformed cells. Alterna- 225-235, 1990. tively, selection systems such as the hypoxanthine:amino- 2. Bos, J. L., Fearson, E. R., Hamilton, S. R., Verlaan-deVires, M., Van Boom, J. H., Vander Eb, A. J., and Vogelstein, B. Prevalance of ras gene mutations pterine:thymidine selection medium, which is dependent on TK in colorectal cancers. Nature (Lond.), 327: 293-297, 1987. expression in TK- cells, could also be used to select tumor 3. Bos, J. L. ras oncogenes in human cancer: a review. Cancer Res., 49: 4682- 4689, 1989. suppressor genes. Hence, a-actin promoter fused to reporter 4. Milburn, M. V., Tong, L., de Vos, A. M., Briinger, A., and Yamaizumi, Z. genes would provide an activity marker for the identification Molecular switch for : structural differences between ac- and isolation of drugs that can suppress the ras-transformed tive and inactive forms of protooncogenic ras proteins. Science (Washington DC), 247: 939-945, 1990. phenotype. Indeed, the colorimetric assay for expression of 5. Pal, E. F., Krengel, V., Petsko, G. A., Goody, R. S., Kabsch, W., and Wit- /~-gal activity or immunological methods for estimating hGH tinghofer, A. Refined crystal structure of the triphosphate conformation of levels in the supernatants of the cells, performed in 96-well H-ras p21 at 1.35 A resolution: implications for the mechanism of GTP hydrolysis. EMBO J., 9: 2351-2359, 1990. dishes, provides a pharmacological screening assay which 6. Ballag, G., and McCormick, F. Regulators and effectors of ras proteins. would be quite general. Another major advantage of this assay Annu. Rev. Cell Biol., 7: 601-632, 1991. 7. West, M., Kung, H. F., and Kamata, T. A novel membrane factor stimulates system is that cytotoxic agents such as nonspecific inhibitors of guanine nucleotide exchange reaction of ras proteins. FEBS Lett., 259: 245- DNA, RNA, and protein synthesis will not be scored as false 248, 1990. positives, since the derepression of the reporter-gene activ- 8. Trahey, M., and McCormick, F. A cytoplasmic protein stimulates normal N-ras p21GTPase but does not affect oncogenic mutants. Science (Washing- ity would require the cells to actively synthesize mRNA and ton DC), 238: 542-545, 1987. proteins. 9. Clarke, S., Vogel, J. P., Deschenes, R. J., and Stock, J. Posttranslational The other usefulness of the a-actin promoter system would be modifications of the Ha-ras oncogene protein: evidence for a third class of protein carboxyl methyltransferases. Proc. Natl. Acad. Sci. USA, 85: 4643- to understand the molecular mechanisms involved in cytoskel- 4647, 1988. etal changes that occur in transformed cells. Specifically, the 10. Reiss, Y., Goldstein, J. L., Seabra, M. C., Casey, P. J., and Brown, M. S. Inhibition of purified p21 ras Farnesyl:protein transferase by cys-AAX tet- mechanisms by which activated ras oncogene expression leads rapeptide. Cell, 62: 81-88, 1990. to down regulation of specific cellular genes have not been 11. Gibbs, J. B. ras C-terminal processing enzymes--new drug targets? Cell, 65: understood. Studies on thyroglobulin gene expression in thy- 1-4, 1991. 12. Chandra Kumar, C., and Chang C. Human smooth muscle myosin light roid cells indicate that transformation by ras inactivates specific chain-2 gene expression is repressed in ras transformed fibroblast cells. Cell trans-acting factor(s). The repression in the transformed cells of Growth & Differ., 3: 1-10, 1992. the reporter gene joined to the thyroglobulin promoter can be 13. Leavitt, J., Gunning, P., Kedes, L., and Jariwala, R. Smooth muscle a-actin is a sensitive marker for the transformed phenotype. Nature (Lond.), 316: reversed by fusion with normal differentiated thyroid cells 840-842, 1985. (27). A number of promoters such as that of c-fos, collagenase, 14. Pollack, R., Chen, S., Powers, S., and Verderame, M. Transformation mech- anisms at the cellular level. Adv. Viral Oncol., 4: 3-27, 1984. and polyoma enhancer have been shown to be stimulated by the 15. Wigler, M., Sweet, R., Sim, G. K., Wold, B., Peilicer, A., Lacey, E., Maniatis, ras oncogene (28-31). In the case of polyoma enhancer, it has T., Silverstein, S., and Axel, R. Transformation of mammalian cells with been shown that the three nuclear oncogene-encoded proteins genes from procaryotes and eucaryotes. Cell, 16: 777-785, 1979. 16. Reddy, S., Ozgur, K., Lu, M., Chang, W., Mohan, S. R., Chandra Kumar, C., c-fos, c-jun, and c-ets-1 cooperate for activation by the ras on- and Ruley, H. E. Structure of the human smooth muscle a-actin gene. J. Biol. cogene. The induction of c-fos by the ras oncogene has been Chem., 265: 1683-1687, 1990. shown to be mediated by the serum response factor and CArG 17. Hall, C. V., Jacob, E., Ringold, G. M., and Lee, F. Expression and regulation of Escherichia coil Lac Z gene fusions in mammalian cells. J. Mol. Appl. (CC[A/T]6GG) box interaction (30). The los gene is also nega- Genet., 2: 101-109, 1983. tively regulated by its own product through the CArG box (30). 18. Selden, R. F., Burke Howie, K., Rowe, M. E., Goodman, H. M., and Moore, D. D. Human growth hormone as a reporter gene in regulation studies The CArG box sequences are present in the promoter regions of employing transient gene expression. Mol. Cell. Biol., 6: 3173-3179, 1986. a number of muscle-specific genes, including human a-actin 19. Gorman, C. M., Moffatt, L. F., and Howard, B. H. Recombinant genomes (12). It would be interesting to analyze the role of CArG box which express chioramphenicol acetyl transferase in mammalian cells. Mol. Cell. Biol., 2: 1044-1050, 1983. sequences in regulating o~-actin promoter activity in ras-trans- 20. Spandidos, D. A., and Wilkie, N. M. Malignant transformation of early formed cells. passage rodent cells by a single mutated oncogene. Nature (Lond.), 310: Our previous studies, together with those of others, indicate 469-475, 1984. 21. Craven, G. R., Steers, E., Jr., and Anfinsen, C. B. Purification, composition, that the changes in cytoskeletal components, observed in trans- and molecular weight of the/3-galactosidase of Escherichia coil K12. J. Biol. formed cells, are not an inevitable consequence of transforma- Chem., 240: 2468-2477, 1965. 6883 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. REGULATION OF SMOOTH MUSCLE ,-ACTIN PROMOTER IN ras-TRANSFORMED CELLS

22. Furth, M. E., Davis, L. J., Fluerdelys, B. S., and Scolnick, E. M. Monoclonal 27. Avvedimento, V. E., Musti, A., Fusco, A., Bonapace, M. J., and Di Lauro, R. antibodies to the p21 products of the transforming gene of Harvey murine Neoplastic transformation inactivates specific-transacting factor(s) required sarcoma virus and of the cellular ras gene family. J. Virol., 43: 294-304, for the expression of the thyroglobulin gene. Proc. Natl. Acad. Sci. USA, 85: 1982. 1744-1748, 1988. 23. Bassin, R. H., and Benade, L. E. Defining the mechanisms of transformation 28. Wasylyk, C., Imler, J. L., Perez-Mutul, J., and Wasylyk, B. The c-Ha-ras through analysis of revertant cells. In: G. Klein (ed.), Tumor Suppressor oncogene and a tumor promoter activate the polyoma virus enhancer. Cell, Genes, pp. 15-47. New York: Marcell-Dekker Publishing Co., 1990. 48: 525-534, 1987. 24. Bassin, R. H., and Noda, M. Oncogene inhibition by cellular genes. Adv. 29. lmler, J. L., Schatz, C., Wasylyk, C., Chatton, B., and Wasylyk, B. A Harvey- Viral Oncol., 6: 103-127, 1987. ras responsive transcription element is also responsive to a tumor-promoter 25. Lloyd, A., Yancheva, N., and Wasylyk, B. Transformation suppressor activity and serum. Nature (Lond.), 332: 275-278, 1988 of a jun transcription factor lacking its activation domain. Nature (Lond.), 30. Gutman, A., Wasylyk, C., and Wasylyk, B. Cell-specific regulation of onco- 352: 635-638, 1991. gene-responsive sequences of the c-fos promoter. Mol. Cell. Biol., I1: 5381- 26. Miillauer, L., Fujita, H., Suzuki, H., Katabami, M., Hitomi, Y., Ogiso, Y., 5387, 1991. and Kuzumaki, N. Elevated levels of gelsolin and a-actin expression in a flat 31. Schonthal, A., Herrlich, M. J., Rahmsdorf, P., and Ponta, H. Requirements revertant R 1 of Ha-ras oncogene-transformed NIH/3T3 cells. Biochem. Bio- for los gene expression in the transcriptional activation of collagenase by phys. Res. Commun., 171: 852-859, 1990. other oncogenes and phorbol esters. Cell, 54: 325-334, 1988.

6884 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research. Regulation of Smooth Muscle α-Actin Promoter in ras -transformed Cells: Usefulness for Setting Up Reporter Gene-based Assay System for Drug Screening

C. Chandra Kumar, Pierre Bushel, Sheela Mohan-Peterson, et al.

Cancer Res 1992;52:6877-6884.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/52/24/6877

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/52/24/6877. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1992 American Association for Cancer Research.