Expression of the Farnesyltransferase β-Subunit Gene in Human Ovarian Carcinoma: Correlation to K-rasMutation

GYNECOLOGIC ONCOLOGY ARTICLE NO.

66, 308–312 (1997)

GO974750

Expression of the Farnesyltransferase b-Subunit Gene in Human Ovarian Carcinoma: Correlation to K-ras Mutation H. Tanimoto,* K. D. Mehta,† T. H. Parmley,* K. Shigemasa,‡ G. P. Parham,* J. Clarke,* and T. J. O’Brien*,† *Department of Obstetrics and Gynecology and †Department of Biochemistry and Molecular Biology, University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205-7199; and ‡Department of Obstetrics and Gynecology, Kure National Hospital, Kure, Japan Received November 11, 1996

The ras signaling protein requires a posttranslational modification to localize it to the inner surface of plasma membrane. In this state it can behave as a signal transduction mediator. Farnesyltransferase plays an important role in this posttranslational processing of ras by attaching a farnesyl group to the cysteine of the ras C-terminal tetrapeptide. In this study, we investigated the relationship of K-ras expression and mutation with farnesyltransferase b-subunit expression in 20 ovarian tumors (17 carcinomas and 3 low malignant potential tumors) and 4 normal ovaries. The expression level of mRNA was determined by using quantitative PCR and mutation analysis was performed by direct cDNA sequencing. K-ras mutations were found in 1 of 3 low malignant potential tumors and in 4 of 17 carcinoma cases. K-ras mRNA overexpression was found in 1 of 3 low malignant potential tumors (one with mutated ras) and in only 1 of 17 carcinoma cases. Farnesyltransferase b-subunit mRNA overexpression was found in 2 of 3 low malignant potential tumors and in 7 of 17 carcinoma cases. Interestingly, all K-ras mutation cases showed farnesyltransferase b-subunit overexpression. These findings suggest that there may be a direct relationship between K-ras (ras dysfunction) mutation and expression of farnesyltransferase b-subunit gene. q 1997 Academic Press

INTRODUCTION

The ras oncogenes encode monomeric GTPases that serve as relay signals from receptor tyrosine kinases to the nucleus, thereby stimulating cell proliferation [1]. Many studies have revealed that mutated ras genes are frequently identified in human cancers [2, 3]. Mutations in the ras peptide result in reduced GTPase activity, and consequently to proliferation of malignant cells [4]. Like the other monomeric GTPases, the ras protein requires a posttranslational modification to localize it to the inner surface of plasma membrane, where it functions as a signal transduction mediator. Farnesyltransferase plays an important role in this posttranslational processing of ras [4–6]. This processing includes farnesylation of the C-terminal cysteine residue, proteolytic removal of the three C-terminal amino acids, and carboxymethylation

Gyn 4750

/

6d1d$$$481

PATIENTS AND METHODS

Tissue Samples Fresh surgical specimens of ovarian tumor were obtained from 20 patients (3 cases of low malignant potential tumors and 17 cases of carcinomas). Clinical staging was determined according to the criteria of the International Federation of Gynecology and Obstetrics (FIGO). Normal ovaries were obtained from 4 patients who underwent surgery for benign gynecological disease. The materials were obtained immediately after the surgical procedure, frozen in liquid nitrogen, and stored at 0807C prior to mRNA isolation. mRNA Extraction and cDNA Synthesis Extraction of mRNA from the tissue specimen and cDNA synthesis were carried out according to the following methods as previously reported [8, 9]. mRNA was isolated by using a RiboSep mRNA isolation kit (Becton–Dickson Labware). In this procedure, poly A/ mRNA was isolated directly from the tissue lysate using the affinity chromatography media oligo(dT) cellulose. The amount of mRNA recovered was quantitated by UV spectrophotometry. cDNA was synthesized from 5.0 mg of mRNA by random hexamer priming using a 1st Strand cDNA synthesis kit (Clontech).

308

0090-8258/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

of the cysteine [7]. The first farnesylation is catalyzed by farnesyltransferase and farnesyl diphosphate is utilized as a substrate. As a result of this series of reactions, the ras protein acquires a lipophilic domain at the C-terminus which mediates its binding to the inner surface of plasma membrane. In this study, we investigated the relationship of K-ras expression and mutation with farnesyltransferase b-subunit expression in ovarian tumors. The expression level of mRNA was determined by using quantitative PCR and mutation analysis was performed by direct cDNA sequencing.

07-14-97 22:44:16

goa

AP: GYN

FARNESYLTRANSFERASE GENE IN OVARIAN CARCINOMA

309

TABLE 1 Sequences of Amplification Primers Gene

Primer

K-ras FTase b-tubulin

5*-CTTGTGGTAGTTGGAGCT-3* 5*-AAGTCCTGAGCCTGTTTT-3* 5*-TCTTCAGTTCTTACAAGTTC-3* 5*-TCCTCGGTGCCAATGATGCA-3* 5*-TGCATTGACAACGAGGC-3* 5*-CTGTCTTGACATTGTTG-3*

Sense Antisense Sense Antisense Sense Antisense

Quantitative PCR mRNA expression of K-ras and farnesyltransferase bsubunit were determined using quantitative PCR. Quantitative PCR was performed according to the method of Noonan et al. [10] with some modification as previously reported [8, 9]. Primer sequences for amplification are listed in Table 1. The K-ras primer pair was designed to generate amplified fragments to include codons 12, 13, and 61. The farnesyltransferase primer pair was specific for its b-subunit sequence, because other prenylation enzymes share the same a-subunit. b-tubulin was utilized as an internal control. The PCR reaction mixture consists of cDNA derived from 50 ng of mRNA, 5 pmol of sense and antisense primers for both the target gene and the b-tubulin gene, 200 mmol of dNTPs, 5 mCi of a-32PdCTP, and 0.25 unit of Taq DNA polymerase with reaction buffer (Promega) in a final volume of 25 ml. The target sequences were amplified in parallel with the btubulin gene. Thirty cycles of PCR were carried out in a Thermal Cycler (Perkin–Elmer Cetus). Each cycle of PCR included 30 sec of denaturation at 957C, 30 sec of annealing at 607C, and 30 sec of extension at 727C. The PCR products were separated on 2% agarose gels and the radioactivity of each PCR product was determined using a phosphoimager (Molecular Dynamics). In the present study, we used the expression ratio (target gene/b-tubulin) measured by phosphoimager to determine gene expression and defined the value at mean / 2 SD as the cutoff value to determine overexpression. Overexpression is defined as an expression ratio above the mean for normal by 2 SD or more.

FIG. 1. cycles.

Linearity of the b-tubulin PCR product was constant over 30

monitored from 15 to 35 cycles over five cycle intervals and the abundance of amplification product was measured as the log 32P incorporation. The linearity of the PCR product was constant over 30 cycles (Fig. 1). Figure 2 shows an example of the phosphoimager autoradiograph of 32P-labeled PCR products coamplified using K-ras primers with internal control b-tubulin primers (all primer sequences are shown in Table 1). A comparison of normal, low malignant potential (LMP) tumor and carcinoma mRNA levels is shown in Fig. 2. The authors made every effort to optimize epithelial representation in normal tissue extractions by using wedges of ovarian tissue. Nevertheless, the authors appreciate the fact that some normal stromal tissue is included in the normal extracts. Analysis of the data quantified using the phosphoimager and compared as ratios of expression (K-ras/b-tubulin) indicates that, for these examples, in only one case is K-ras expression elevated over the cutoff value (mean / 2 standard deviations) compared to both control and other

Direct cDNA Sequencing Mutation analysis of K-ras was performed by direct sequencing of PCR amplified K-ras cDNA. Amplified cDNA samples were purified using a Wizard PCR Preps DNA purification system (Promega). The sequencing reaction was carried out using a PRISM Ready Reaction Dye Deoxy terminator cycle sequencing kit (Applied Biosystems). Applied Biosystems Model 373A DNA sequencing system was used for direct cDNA sequence determination. RESULTS

As a preface to this study, the authors evaluated the linearity of the PCR amplification. The PCR amplification was

AID

Gyn 4750

/

6d1d$$$482

07-14-97 22:44:16

FIG. 2. Quantitative PCR using b-tubulin and K-ras primers comparing normal and low malignant potential tumors and carcinomas. The expression ratios determined by phosphoimager were 0.012 (lane 1), 0.058 (lane 2), 0.020 (lane 3), 0.149 (lane 4), 0.601 (lane 5), 0.017 (lane 6), and 0.089 (lane 7).

goa

AP: GYN

310

TANIMOTO ET AL.

TABLE 2 Expression of K-ras and FTase Genes and Mutation of K-ras Gene in Ovarian Tumors mRNA expressionc Case No.

Histological typea

Stage/gradeb

K-ras

FTase

Mutationd K-ras

1–4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

n ovary s ad (LMP) m ad (LMP) m ad (LMP) s car s car s car s car s car s car s car s car s car s car s car s car s car s car s car m car m car

— 1/1 1/1 1/1 1/2 1/3 2/1 3/2 3/3 3/2 3/3 3/3 3/3 3/3 3/1 3/2 3/3 3/3 3/3 3/2 3/1

n n // n n n n n // n n n n n n n n n n n n

n // // n n n n n // // n n n n // // n // n // //

— w mut w w w w w w mut w w w w w mut w mut w mut w

a

n, normal; s ad, serous adenoma; s car, serous carcinoma; m car, mucinous carcinoma. b FIGO stage and histological grade. c n, normal range is equal to mean { 2 SD, //, positive is equal to mean /2 SD or greater. d w, wild type; mut, mutation.

carcinoma specimens (Fig. 2). Overall, K-ras expression was found to be elevated in two abnormal cases viz., case 6, a low malignant potential tumor, and in case 12, a serous carcinoma (Table 2). Direct sequencing of the PCR products derived from the complete tumor set indicates that mutations were present in 5 of the 20 tumor cases (Table 3). One low TABLE 3 K-ras Mutation in Ovarian Tumors Case No.

Histological typea

Stage

Codon

6

m adenoma (LMP)

I

12

a

13

s carcinoma

III

13

19

s carcinoma

III

13

21

s carcinoma

III

13

23

m carcinoma

III

13

Mutation GGT r GTT (Gly) (Val) GGC r GCC (Gly) (Ala) GGC r GCC (Gly) (Ala) GGC r GCC (Gly) (Ala) GGC r GCC (Gly) (Ala)

Gyn 4750

/

malignant potential mucinous tumor showed G-to-T transversion in codon 12. Three serous and one mucinous carcinoma showed G-to-C transversion in codon 13. These base changes alter the amino acid sequences such that codon 12 is mutated from glycine to valine and codon 13 from glycine to alanine (Table 3). Table 2 shows the summary of the histological type, FIGO stage, grade, expression of K-ras, and farnesyltransferase b-subunit genes and mutation of Kras gene in the cases studied. K-ras mutations were found in 1 of 3 low malignant potential tumors and 4 of 17 carcinoma cases. From a histological point of view, a majority of the tumors were serous cases. Three of these serous cases showed a K-ras mutation; of the 4 mucinous tumors studied, 2 cases showed a K-ras mutation. An example of the quantitative PCR using farnesyltransferase b-subunit primers with control b-tubulin primers is shown in Fig. 3. As in Fig. 2, we present a comparison of normal, LMP tumor and carcinoma mRNA levels to determine the relative expression of the farnesyltransferase bsubunit gene to b-tubulin. Analysis of the data quantified using the phosphoimager and compared as ratios of expression (farnesyltransferase/b-tubulin) indicates that in cases 5, 12, 19, and 13 farnesyltransferase b-subunit expression is elevated compared to both control and other carcinoma specimens. Overall, farnesyltransferase b-subunit mRNA overexpression was found in 2 of 3 low malignant potential tumors and 7 of 17 carcinoma cases (Table 4). All K-ras TABLE 4 Number of Cases with Overexpression and Mutation

Normal ovary LMP Carcinoma a

m, mucinous; s, serous.

AID

FIG. 3. Quantitative PCR of the b-subunit of farnesyltransferase compared to b-tubulin in normal and low malignant potential tumors and carcinomas. The expression ratios were 0.204 (lane 1), 0.156 (lane 2), 0.077 (lane 3), 2.127 (lane 4), 1.409 (lane 5), 1.597 (lane 6), and 3.326 (lane 7).

6d1d$$$482

07-14-97 22:44:16

N

K-ras mutation

K-ras overexpression

FTase overexpression

4 3 17

— 1a (33%) 4a (24%)

0 (0%) 1 (33%) 1 (6%)

0 (0%) 2 (67%) 7 (41%)

All K-ras mutation cases showed FTase b-subunit overexpression.

goa

AP: GYN

FARNESYLTRANSFERASE GENE IN OVARIAN CARCINOMA

mutation cases showed farnesyltransferase b-subunit overexpression (Table 2). The authors have previously evaluated other internal controls including actin and G3PD and found b-tubulin to be the most consistently expressed in normal and ovarian tumors. mRNA expression levels of K-ras relative to b-tubulin in normal ovaries and ovarian tumors are shown in Fig. 4. In 2 of 20 tumor cases, the K-ras expression ratio was elevated above the cutoff value. Some cases were elevated above the mean for normal but these expression values were below the ú2 SD cutoff value. A comparison of normal versus tumor expression of the b-subunit of farnesyltransferase is shown in Fig. 5. Here 9 of 20 ovarian tumors expressed levels above the cutoff value (normal ovary mean / 2 SD), including 2 of 3 low malignant potential tumors. DISCUSSION

Activated ras oncogenes have been frequently identified in a variety of human carcinomas and it has been accepted that activation of ras through mutation is one of the common genetic alterations in carcinogenesis [1–3]. The ras gene encodes a monomeric GTPase that is located on the inner surface of the plasma membrane, which serves to relay signals from the receptor tyrosine kinases to the nucleus activating cell proliferation [1]. To become anchored to the inner surface of the plasma membrane, the ras protein requires a posttranslational modification in its C-terminus [4–6]. This C-terminal processing is obligatory for ras transforming activity. Unanchored ras, regardless of mutations, appears to have no role in signal transduction. In vitro examination of the transforming activity of ras using mutants with deletions within ras sequences showed that the mutants with a Cterminal deletion did not transform cells [11]. Recent studies have revealed that inhibition of farnesylation resulting in inactivation of ras can reduce tumor growth [12–14]. How-

FIG. 4. K-ras expression relative to b-tubulin in normal ovary and ovarian tumor.

AID

Gyn 4750

/

6d1d$$$482

07-14-97 22:44:16

311

FIG. 5. Farnesyltransferase b-subunit expression relative to b-tubulin in normal ovary and ovarian tumor.

ever, little is known about control of the expression of these prenylation enzymes in carcinoma cells and it is therefore pertinent to clarify the relationship of prenylation enzymes such as farnesyltransferase to ras expression and mutation in tumor cells. In this comparison study of ras integrity and function, we examined the expression of mRNA and mutation rate of Kras in ovarian tumor tissue compared to the expression of the farnesyltransferase b-subunit gene. K-ras mutations were found in 1 of 3 low malignant potential tumors and in 4 of 17 carcinoma cases, and interestingly, in all mutation cases overexpression of mRNA for the farnesyltransferase b-subunit gene was also noted (Table 2). These findings suggest a relationship between dysfunctional K-ras and the expression of farnesyltransferase b-subunit gene. Abrogation of the downstream processing of ras mediated signals could therefore reduce the feedback or down-regulation of farnesyl transferase. The net result would be overexpression of farnesyltransferase to enhance the ineffective ras signaling system. When we examined the overexpression of the K-ras gene, a lower incidence (2 of 20) of overexpression compared to mutation was observed. Some cases also showed farnesyltransferase b-subunit overexpression but not K-ras mutation (4 of 20). Such cases might be the result of other farnesylation events. It is known that a family of proteins such as nuclear lamins and transducin g-subunit are also farnesylated in order to achieve membrane localization [15, 16]. They also have the anchor consensus sequence at the C-terminus, CAAX (C, cysteine; A, an aliphatic amino acid; and X, any amino acid), and are dependent on farnesyltransferase for membrane anchoring. Although there are several proteins which are anchored by farnesyltransferase, the present study indicates a direct relationship between ras and farnesyltransferase overexpression. While dysfunctional ras signaling is likely not the primary insult in the malignant growth of ovarian tumors, it could

goa

AP: GYN

312

TANIMOTO ET AL.

enhance the malignant potential of a small proportion of these tumors (20%) and thereby offers an opportunity for future therapeutic intervention in those malignancies. ACKNOWLEDGMENTS We acknowledge the participation of the following divisions of the Cooperative Human Tissue Network (CHTN) in providing tumor tissues: Western Division, Case Western Reserve University, (Cleveland, OH); Midwestern Division, Ohio State University, (Columbus, OH); Eastern Division, NDRI, (Philadelphia, PA); Pediatric Division, Children’s Hospital, (Columbus, OH); and Southern Division, University of Alabama at Birmingham, (Birmingham, AL).

REFERENCES 1. Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson JD: Molecular Biology of the Cell. 3rd ed. New York: Garland, 1994, p. 763 2. Bos JL: ras oncogenes in human cancer. Cancer Res 49:4682–4689, 1989. 3. Bos JL, Fearon ER, Hamilton SR, Verlaan-de Vries M, van Boom JH, van der Eb AJ, Vogelstein B: Prevalence of ras gene mutations in human colorectal cancers. Nature 327:293–297, 1987 4. Newman CMH, Magee AI: Posttranslational processing of the ras superfamily of small GTP-binding proteins. Biochim Biophys Acta 1155:79–96, 1993 5. Casey PJ, Solski PA, Der CJ, Buss JE: p21 ras is modified by a farnesyl isoprenoid. Proc Natl Acad Sci USA 86:8323–8327, 1989 6. Hancock JF, Magee AI, Childs JE, Marshall CJ: All ras proteins are polyisoprenylated but only some are palmitoylated. Cell 57:1167– 1177, 1989

AID

Gyn 4750

/

6d1d$$$483

07-14-97 22:44:16

7. Hancock JF, Cadwallader K, Marshall CJ: Methylation and proteolysis are essential for efficient membrane binding of prenylated p21K-ras(B). EMBO J 10:641–646, 1991 8. Shigemasa K, Hu C, West CM, Moon SH, Parham GP, Parmley TH, Korourian S, Baker VV, O’Brien TJ: p21: A monitor of p53 dysfunction. Int J Gynecol Cancer, submitted for publication 9. Shigemasa K, Hu C, West CM, Clarke J, Parham GP, Parmley TH, Korourian S, Baker VV, O’Brien TJ: p16 overexpression: A potential early indicator of transformation in ovarian carcinoma. J Soc Gynecol Invest, in press, 1997 10. Noonan KE, Beck C, Holzmayer TA, Chin JE, Wunder JS, Andrulis IL, Gazdar AF, Willman CL, Griffith B, Von Hoff DD, and Roninson IB: Quantitative analysis of MDR1 (multidrug resistance) gene expression in human tumors by polymerase chain reaction. Proc Natl Acad Sci USA 87:7160–7164, 1990 11. Willumsen BM, Christensen A, Hubbert NL, Papageorge AG, Lowy DR: The p21 ras C-terminus is required for transformation and membrane association. Nature 310:583–586, 1984 12. Gibbs JB: ras C-terminal processing enzymes-new drug targets? Cell 65:1–4, 1991 13. Gibbs JB, Oliff A, Kohl NE: Farnesyltransferase inhibitors: ras research yields a potential cancer therapeutic. Cell 77:175–178, 1994 14. Sun J, Qian Y, Hamilton AD, Sebti SM: ras CAAX peptidomimetic FTI 276 selectively blocks tumor growth in nude mice of a human lung carcinoma with K-ras mutation and p53 deletion. Cancer Res 55:4243–4247, 1995 15. Fukuda Y, Takao T, Ohguro H, Yoshizawa T, Akino T, Shimonishi Y: Farnesylated g-subunit of photoreceptor G protein indispensable for GTP-binding. Nature 346:658–660, 1990 16. Glomset JA, Gelb MH, Farnsworth CC: Prenyl proteins in eukaryotic cells: A new type of membrane anchor. Trends Biochem Sci 15:139– 142, 1990

goa

AP: GYN