The Prostate 22:23-30 (1993)

Mutant p53 Expression in Prostate Carcinoma Peter J. Van Veldhuizen, Raj Sadasivan, Fernando Garcia, Mark S. Austenfeld, and Ronald L. Stephens

Division of Oncology (P.J. V.V., R.S., R.L.S.), Department of Pathology (F.G.), and Section of Urology (M.S.A.), University of Kansas Medical Center, Kansas City

The expression of the mutant p53 tumor suppressor gene was evaluated in 33 human prostate carcinomas. Using an immunohistochemical method with monoclonal antibodies PAb 1801 and PAb 240, 26 (79%) tumors demonstrated positive immunostaining for mutant p53. Only areas of glandular tumor were positive, with adjacent stromal elements and areas of glandular hyperplasia being negative. The predominant staining pattern was cytoplasmic. This pattern may be related to p53 binding to certain heat shock proteins (HSP 72/73), as a monoclonal antibody to these proteins demonstrated a cytoplasmic location as well. These results demonstrate that abnormal p53 expression is a frequent event in prostate cancer. 0 1993 Wiley-Liss, Inc.

Key words: PAb 1801, PAb 240, cytoplasmic, heat shock

INTRODUCTION

Prostate cancer is the most common malignancy among men in the United States, and is the second leading cause of cancer mortality. The genetic targets leading to transformation from normal prostate to malignancy have not been well characterized. Mutations of the p53 gene have rapidly become one of the most commonly identified genetic changes in human malignancies [ l ] . The p53 gene encodes a phosphoprotein that is important in the regulation of cell growth. A mu- tation of this gene can result in either the loss of its normal function (i.e., the loss of the ability to block cell division) or in the gain of an abnormal function (e.g., the ability to transform cells directly) [2-41.

Somatic mutations of p53 have been implicated as causal events in a number of common tumors, including breast [5-71, lung [8,9], esophageal [lo], bladder [ 1 I], and colon carcinomas [ 12,131. Mutations have also been identified in certain prostate cancer cell lines [14,15]. Germ line mutations have been identified in certain cancer- prone families, including those with the Li-Fraumeni syndrome [ 16,171.

The p53 gene is located on the short arm of chromosome 17, band p13. A single point mutation can create a carcinogenic p53 and mutations are usually missense resulting in the production of an abnormal protein product. Point mutations have been the most frequently identified change, although allelic loss, rearrangements, and

Received for publication May 12, 1992; accepted August 17, 1992.

Address reprint requests to Dr. Raj Sadasivan, University of Kansas Medical Center, Division of On- cology, 3901 Rainbow Blvd., Kansas City, KS 66160-7353.

0 1993 Wiley-Liss, Inc.

24 Van Veldhuizen et al.

deletions also occur [ 1,181. Mutations have been identified at several different re- gions within the gene, but the majority have been confined to exons 5-8 [l]. To a degree, the specific point of mutation varies between the different tumor types, and the different mutants appear to have distinct biological activities [8, 19,201.

The wild-type p53 has a short intracellular half-life of only 15 minutes, whereas mutant p53 proteins have a half-life of several hours. This extension of the half-life results in a sufficient increase on the amount of intracellular protein for it to become immunohistochemically detectable [2 11. Immunostaining with specific monoclonal antibodies to the mutant p53 protein correlates well with mutations found at the DNA level [5,9,22]. The wild type is localized predominantly to the cell nucleus; however, certain mutants have been shown to complex with the 70-KD family of heat shock proteins, resulting in a change of the cellular distribution of that mutant [23,24].

Using an immunohistochemical method and specific monoclonal antibodies, we report the frequency of abnormal p53 expression in human prostate carcinomas, and correlate this expression with tumor stage and Gleason’s score.

MATERIALS AND METHODS

Prostate carcinoma specimens were obtained from patients undergoing radical prostatectomy for a known carcinoma. Specimens were obtained at histopathologic examination of the gross tissue received from the operating room and were quickly frozen at - 70°C. These samples were saved for batch processing. Tissue samples were stained with hematoxylin and eosin (H&E), and there was an average of one section per patient. These slides were used to facilitate histopathologic correlation. Immunohistochemical staining for mutant p53 expression was accomplished using mouse monoclonal antibodies PAb 1801 and PAb 240 (Oncogene Science, Manhas- set, NY) and using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) [25]. PAb 1801 recognizes both the wild-type and mutant p53 protein products and reacts with an epitope located near the amino end of all known forms of p53. PAb 240 detects only mutant forms and recognizes an evolutionarily conserved epitope lying between amino acids 156-214 on murine p53.

In brief, thin sections of 8 pm were made from the tissue samples and mounted on poly L-lysine-coated glass slides. The slides were air dried and fixed in acetone for 10 min. They were washed in PBS with Tween 0.02% for 10 min. Sections were incubated in 0.1 % hydrogen peroxide to quench endogenous peroxidase activity. The slides were then treated with a 1:200 dilution of normal horse serum for 20 min. Slides were then washed in PBS with Tween, and the different monoclonal antibodies were applied to separate slides at a concentration of 10 pg/ml. Omission of the primary antibody was used as a negative control. All slides were incubated overnight at 4°C in a humidified chamber. With continued washing with PBS between each step, the slides were then treated with a biotinylated goat antimouse IgG for 30 min. This was followed by a 60-min incubation with an avidin-horseradish peroxidase (HRP) complex. The color reaction was completed after incubation with DAB (di- aminobenzidine tetrachloride) for 3 min. The sections were counterstained with he- matoxylin and mounted.

Using the same immunohistochemical technique but with monoclonal antibody HSP 72/73 (Oncogene Science), prostate specimens were evaluated for the presence

p53 Expression in Prostate Carcinoma 25

and cellular location of these heat shock proteins in conjunction with mutant p53 proteins.

Histologic grading of H&E slides was done according to the Gleason’s system. In cases in which there was only one histologic pattern that was discernible, the Gleason’s score was computed by doubling the Gleason’s grade.

RESULTS

Thirty-three fresh specimens were evaluated for the expression of a mutant p53 product. The mean age of the patients was 68.7 years within a range of 5 1-81. There were 31 white and 2 black patients. Of 33 tumors evaluated, 26 (79%) of the tumors evaluated expressed detectable amounts of p53, as evidenced by a positive immuno- histochemical stain.

In all cases, the predominant staining pattern was cytoplasmic, although in some cases varying degrees of nuclear staining in addition to the cytoplasmic pattern were seen as well (Fig. 1). The intensity of the cytoplasmic staining made assessment of the degree of nuclear staining difficult in these cases. In all positive specimens, greater than 50% of tumor cells stained with a moderate to intense staining pattern. All negative specimens either had minimal or no detectable staining.

The staining pattern was homogeneous with the majority of tumor cells staining positively. Specimens that contained normal areas of prostate or areas of glandular hyperplasia stained positively only in the areas in which tumor was identified on the correlative H&E section. The stromal elements and the histologically normal-ap- pearing glands either had very minimal or no detectable staining.

The results from using the two different anti-p53 monoclonal antibodies, PAb 240 and PAb 1801, were equivalent, but with two exceptions. First, PAb 240 in general stained the tumor cells more intensely with less background staining. Second, one patient with a stage Dl tumor stained positively with PAb240, but had minimal cytoplasmic staining with PAB 180 1.

Although the numbers are small, higher-stage tumors were less likely to show a positive staining pattern (Table I). Table I1 shows the expression of mutant p53 proteins as a function of the histologic grade (Gleason’s score). The histologic grade appears not to be correlated with mutant p53 immunostaining, although there is a trend of decreased staining with high-grade (Gleason’s score 8 -10) tumors.

Heat shock proteins were detectable in all twelve samples evaluated, and in the areas with tumor involvement, the staining pattern was similar to that seen with the p53 monoclonal antibodies. In these sections, stromal areas as well as tumor areas stained positively.

DISCUSSION

This study demonstrates that abnormal expression of the p53 gene is a common occurrence in human prostate cancer. This occurrence may be an important step in the series of genetic events that ultimately lead to the development of a prostate carci- noma. The 79% frequency found in this study is higher than that seen in some malignancies [5,6,10,11], but correlates with the frequency identified in others, including malignant melanoma [26] and small cell carcinoma of the lung [S].

It is notable that the cellular location of the p53 protein identified in these

26 Van Veldhuizen et al.

Fig. 1. A: Hematoxylin and eosin-stained moderately differentiated adenocarcinoma of the prostate. x 250. B: Immunostain of this tumor using a monoclonal antibody for p53. There is intense staining of the glandular areas of tumor, with the surrounding stroma being negative. X400.

p53 Expression in Prostate Carcinoma 27

TABLE I. Expression of Mutant p53 According to Tumor Stage

Stagea samples positiveb

A2 1 0 (0%) B1 11 11 (100%) B2 16 12 (75%) C 3 2 (66%) D1 2 1(50%)

No. of No.

"Pathologic stage. bAll positive samples had staining of >50% of tumor cells, with a moderate to intense staining pattern.

TABLE 11. Expression of Mutant p53 According to Gleason's Score

Gleason's No. of No. score samules Dositive"

0-4 6 5 (83%) 5 -7 21 17 (81%) 8-10 6 4 (66%)

aResults of immunostaining were similar for PAb 180 1 and PAb 240.

samples is predominantly cytoplasmic. Cytoplasmic staining has been reported in colorectal cell lines [13], as well as in certain breast [7] and lung [9] carcinomas, although the predominant staining pattern in these series was nuclear. In a series of melanoma patients 1261, 29 of 45 melanomas demonstrated a combined nuclear and cytoplasmic staining pattern. In addition, 8 of the 45 tumors demonstrated only cytoplasmic staining. A more recent report examined p53 expression in a series of 11 1 primary breast tumors, and 59 (53%) were positive using monoclonal antibody PAb 240 [27]. In this study, the predominant staining pattern was also cytoplasmic, in contrast to several series of breast carcinomas in which nuclear staining was the predominant pattern. Nuclear disruption was not the cause of the cytoplasmic location of p53 in this series, and the precise reason for the different staining patterns remains unclear.

The variation of the cellular location may depend on the characteristic of a particular mutant. The p53 phosphoprotein is synthesized in the cytoplasm and then transported to the nucleus. Nuclear localization is thought to be needed in order for p53 activity. Using rat embryo fibroblasts, Shaulsky et al. [28] demonstrated that the cellular location of the mutant p53 depended on the part of the mutant protein affected. Mutations that altered the carboxy terminus of the p53 protein resulted in a cytoplasmic location of the mutant, whereas alterations of the central or amino ter- minus resulted in a predominantly nuclear location.

Another possible explanation for cytoplasmic staining is that certain mutant p53 proteins form stable complexes with cellular heat shock proteins [23,24]. Little is known about the precise function 1291 of these proteins, but they are known to concentrate in the nucleus during periods of heat shock or stress to the cell, but are

28 Van Veldhuizen et al.

otherwise cytoplasmically located [30]. Further investigation is needed to determine whether a complex is formed with mutant p53 in the case of prostate carcinomas.

Gusterson et al. [31] demonstrated that the method of tissue processing may affect the cellular location. Using squamous carcinoma cell lines, these investigators found predominantly nuclear localization when nonaqueous fixatives (methanol, ac- etone, and methacarn) were used, as well as a shift toward a predominant cytoplasmic localization when either a 70% aqueous methanol or formol saline fixative was used.

We used monoclonal antibodies PAb 240 and PAb 1801. PAb 1801 detects both the wild-type as well as mutant p53 protein, but PAb 240 is believed to be mutant specific [32]. However, denaturation of the wild type protein could expose the epitope that this antibody detects. If p53 is expressed normally, it should still be immunohistochemically undetectable. Overexpression of the wild type could con- ceivably raise the cellular concentration enough to allow detection immunohis- tochemically . Denaturation in this setting, whether it occurs naturally or in tissue processing, could possibly result in a positive immunostain with PAB 240 without an underlying mutation.

One published series [33] has found p53 positivity in 5/24 (17%) of prostate carcinomas, as evidenced by positive staining with PAb 240. They also evaluated 34 specimens of benign prostatic hypertrophy which were all negative. This series con- tained predominantly high-stage and high-grade tumors with all, but three of the carcinomas having a Gleason’s score of 8 or greater compared with our series, which contains predominantly patients with stage B and Gleason’s score 5-7. This in part may explain the different positivity rate between these two studies. One recently published abstract (341 correlates with our findings, demonstrating p53 positivity using PAb 240 in 77% of prostatic carcinomas. Abnormal p53 mRNA production has also been demonstrated in 75% of prostate tumors [15].

In our samples, p53 was expressed less frequently in higher stage and grade of tumor. It is hypothesized that these higher grade tumors may have undergone an additional mutation or deletion in p53 and subsequently no longer express the gene product. These mutations or deletions would then not be detected by immunostaining

In colon epithelia, mutation of the p53 gene appears to be a late event. It appears to be the final genetic insult, as an adenoma is converted to an overt malignancy [36]. In prostate epithelia, it is unclear as to whether p53 mutations are early or late events in the development of neoplasia from premalignant conditions. Further investigation of this question may be helpful in understanding the development of prostate cancer. In addition, further quantitation of the level of p53 expression and DNA sequencing is needed to further characterize p53 expression in this tumor.

[361.

REFERENCES

1. Hollstein M, Sidransky D, Vogelstein B, Harris CC: p53 mutations in human cancers. Science

2. Lewis R: The many faces of p53. J NIH Res. 3:56-59, 1991. 3. Sager R: Tumor suppressor genes: The puzzle and the promise. Science 246:1406-1412, 1989. 4. Hollingsworth RE, Lee W: Tumor suppressor genes: New prospects for cancer research. JNCI

5 . Cattoretti G , Rilke F, Andreola S, D’Amato L, Delia D: p53 expression in breast cancer. Int J Cancer

253:49-53, 1991.

83:91-96, 1991.

41: 178-1 83, 1988.

p53 Expression in Prostate Carcinoma 29

6. Varley JM, Brammer WJ, Lane DP, Swallow CD, Dolan C, Walker RA: Loss of chromosome 1 7 ~ 1 3 sequences and mutation of p53 in human breast carcinomas. Oncogene 6:413-421, 1991.

7. Bartek J, Bartkova J, Vojtesek B, Staskova Z, Rejthar A, Kovarik J, Lane DP: Patterns of expression of the p53 tumour suppressor in human breast tissues and tumours in situ and in vitro. Int J Cancer 46:839-844, 1990.

8. Takashi T, Suzuki H, Hida T, Sekido Y, Ariyoshi Y, Ueda R: The p53 gene is very frequently mutated in small cell lung cancer with a distinct nucleotide pattern. Oncogene 6:1775-1778, 1991.

9. Iggo R, Gatter K, Bartek J, Lane D, Harris AL: Increased expression of mutant forms of p53 oncogene in primary lung cancer. Lancet 2:675-678, 1990.

10. Hollstein MC, Metcalf RA, Welsh JA, Montesano R, Hams CC: Frequent mutation of the p53 gene in human esophageal cancer. Proc Natl Acad Sci USA 87:9958-9961, 1990.

11. Sidransky D, VonEschenbach A, Tsai Y, Jones P, Summerhayes I, Marshall F, Paul M, Green P, Hamilton SR, Frost P, Vogelstein B: Identification of p53 gene mutations in bladder cancers and urine samples. Science 252:706-709, 1991.

12. Rodrigues NR, Rowan A, Smith ME, Kerr IB, Bodmer WF, Gannon JV, Lane D: p53 mutations in colorectal cancer. Proc Natl Acad Sci USA 87:7555-7559, 1990.

13. Purdie CA, O’Grady J, Pins J, Wyllie AH, Bird CC: p53 expression in colorectal tumors. Am J Pathol 138:807-812, 1991.

14. Rubin SJ, Hallahan D, Ashman C, Brachman D, Beckett M, Virudachalam S, Weichselbaum R: Two prostate carcinoma cell lines demonstrate abnormalities in tumor suppressor genes. J Surg Oncol 46:31-36, 1991.

15. Cooke DB, Quarmby VE, Mickey D, Isaacs JT, French FS: Oncogene expression in prostate cancer: Dunning R3327 rat dorsal prostatic adenocarcinoma system. Prostate 13:263-272, 1988.

16. Malkin D, Li FP, Strong LC, Fraumeni JF, Nelson CE, Kim D, Kassel J, Gryka M, Bischoff FZ, Tainsky M, Friend S: Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 250:1233-1238, 1990.

17. Srivastava S, Zou Z, Pirollo K, Blattner W, Chang EH: Germ-line transmission of a mutated p53 gene in a cancer prone family with the Li-Fraumeni syndrome. Nature 348:747-749, 1990.

18. Levine AJ, Momand J, Finlay CA: The p53 tumour suppressor gene. Nature 351:453-455, 1991. 19. Bennett WP, Hollstein MC, He A, Zhu SM, Resau JH, Trump BF, Welsh JA, Midgley C, Lane DP,

Harris CC: Archival analysis of p53 genetic alterations in Chinese esophageal carcinoma. Oncogene 6:1779-1784, 1991.

20. Halevy 0, Michalovitz D, Oren M: Different tumor derived p53 mutants exhibit distinct biological activities. Science 250:113-116, 1990.

21. Lane DP, Benchimol S: p53: Oncogene or antioncogene? Genes Devel 4:l-8, 1990. 22. Barick J, Iggo R, Gannon J, Lane DP: Genetic and immunochemical analysis of mutant p53 in human

breast cancer cell lines. Oncogene 5:893-899, 1990. 23. Pinhasi 0, Michalovitz D, Ben-Zeev A, Oren M: Specific interaction between the p53 cellular

tumour antigen and major heat shock proteins. Nature 320:182-184, 1986. 24. Sturzbecher H, Chumakov P, Welch WJ, Jenkins J: Mutant p53 proteins bind hsp 72/73 cellular heat

shock protein in SV 40-transformed monkey cells. Oncogene 1:201-211, 1987. 25. Hsu SM, Raine L, Fanger H: A comparative study of the peroxidase-antiperoxidase method and an

avidin-biotin complex method for studying polypeptide hormones with radioimmunoassay antibod- ies. Am J Clin Pathol 75:734-738, 1981.

26. Stretch JR, Gatter KC, Ralkiaer E, Lane DP, Hams AL: Expression of mutant p53 in melanoma. Cancer Res 515976-5979, 1991.

27. Horak E, Smith K, Bromley L, LeJeune S, Greenall M, Lane D, Harris AL: Mutant p53, EGF receptor and c-erbB-2 expression in human breast cancer. Oncogene 6:2277-2284, 1991.

28. Shaulsky G, Goldfinger N, Ben-Zeev A, Rotter V: Nuclear accumulation of p53 protein is mediated by several nuclear localization signals and plays a role in tumorigenesis. Mol Cell Biol 10:6565- 6577, 1990.

29. Schlesinger MJ: Heat shock proteins: The search for functions. J Cell Biol 103:321-325, 1986. 30. Velazquez JM, Lindquist S: hsp70: nuclear concentration during environmental stress and cytoplas-

mic storage during recovery. Cell 36:655-662, 1984. 31, Gusterson BA, Anbazhagan R, Warren W, Midgley C, Lane DP, O’Hare M, Stamps A, Carter R,

Jay yatilake H: Expression of p53 in premalignant and malignant squamous epithelium. Oncogene 6:1785-1789, 1991.

30 Van Veldhuizen et al.

32. Cannon JV, Greaves R, Iggo R, Lane DP: Activating mutations in p53 produce a common confor- mational effect; a monoclonal antibody specific for the mutant form. EMBO J 9:1595-1602, 1990.

33. Narayan KS, Meyer PR, Jones LW: Archival studies of p53 mutant gene expression and prostate specific antigen (PSA) levels in prostate cancer. J Urol 147:214A, 1992.

34. Meyers FJ, Chi SC, Chamberlain S. deVere White RW, Gumerlock PH: Intra-exonic deletion of p53 in human benign prostatic hyperplasia (BPH). Proc AACR 33:380, 1992.

35. Mellon K, Thompson S, Charlton RG, Marsh C, Robinson M, Lane DP, Hams AL, Wilson Horne CH, Neal DE: p53, c-erbB-2 and the epidermal growth factor in the benign and malignant prostate. J Urol 147:496-499, 1992.

36. Fearon ER, Vogelstein B: A genetic model for colorectal tumorigenesis. Cell 61:759-767, 1990.