Mutations of PTEN/MMAC1 in Primary Prostate Cancers from Chinese Patients¹

A candidate tumor suppressor gene designated PTEN, MMAC1, or TEP-1 (referred to as PTENhereafter) was identified (1, 2, 3) from the q23.3 region of chromosome 10, one of the most frequently deleted regions in prostate cancer (4). The PTEN gene has nine exons that encode a 403-amino acid protein of a dual-specific phosphatase with putative actin-binding and tyrosine phosphatase domains. Introduction of PTEN into cancer cells that lack PTEN function inhibits cell migration and induces cell cycle arrest and apoptosis via negative regulation of the phosphatidylinositol 3′-kinase/protein kinase B/Akt signaling pathway (5, 6, 7). Mutation and down-regulation of the PTEN gene have been detected in various human cancers including that of the prostate (8, 9, 10). In addition, germ-line mutations in PTEN are associated with Cowden disease (11),in which patients are at increased risk for certain cancers.

Thus far, PTEN appears to be the most frequently mutated gene in metastases of prostate cancer, occurring in at least 1 metastatic site in 12 of 19 (63%) patients who had multiple metastases (12) and in 9 of 15 (60%) cell lines and xenografts primarily derived from metastases of prostate cancer (13). These results indicate a role for PTEN in the progression of prostate cancer. Mutations of PTEN in localized prostate cancers have been found at lower frequencies including 1 of 28 (4%;Ref. 14), 1 of 25 (4%; Ref. 15), 1 of 40(2.5%; Ref. 16), 0 of 45 (17), and 1 of 22(5%; Ref. 18). Somewhat higher rates of mutations have been observed in other studies including 10 of 80 [12.5%; 10 of 23(43%) in cases with loss of heterozygosity at PTEN; Ref. 19], 5 of 37 (13.5%; Ref. 20), 8 of 60(13%; Ref. 21), and 1 of 10 (10%; Ref. 9). In hereditary prostate cancer, the role of PTEN has not been detected (22, 23).

The incidence of prostate cancer is lower in Asian men compared with Western men, but the specific genetic or environmental factors that are important are unknown (24, 25). Obviously, more cancers are detected in Western men because of screening with serum PSA³test. The frequency of PTEN mutations in prostate cancer from Asian men has been little studied. One study of 45 primary prostate cancers from Japanese patients did not detect any PTEN mutation (17). In this study, we analyzed primary prostate cancers from 32 Chinese patients, who were not diagnosed using the PSA test. Rather, they were diagnosed after showing clinical symptoms. We also analyzed six metastases from American patients who died of prostate cancer to document additional PTEN mutations in fatal prostatic disease.

Thirty-two formalin-fixed, paraffin-embedded prostate cancer specimens from radical prostatectomy from previously untreated Chinese patients were used in this study. These patients went to physicians after showing various symptoms of prostate cancer, e.g.,difficulty in voiding, urodynia, urgent and frequent urination, and hematuria. None of them were involved in PSA screening. Their prostates were examined by one or more of the following means: rectal ultrasound detection, digital rectal examination, computed tomography, and magnetic resonance imaging. Biopsy was performed for the patients who were suspected to have prostate cancer, and only those whose cancers were at stages B–C underwent radical prostatectomy. The prostatectomies were performed by four surgeons over a period of 5 years. All specimens were from archived paraffin blocks that had been used in routine diagnosis of cancer, and none of them were collected specifically for this study. In addition, DNA was available from six distant metastases from American patients who died of prostate cancer. The clinicopathological characteristics of the tumors are listed in Table 1. The exact tumor stage for the Chinese patients was not available. Tumor cells for DNA isolation were collected from 7 μm H&E-stained sections by microdissection using a protocol described previously (26), which typically ensured a minimum of 70% neoplastic cells for each sample. Nonneoplastic cells collected were present on the same slides as cancer cells and included stromal cells,lymphocytes, and urothelium; in most cases, they did not include nonneoplastic prostatic epithelium. For the cases of metastases,nonneoplastic cells were collected from lymph nodes or seminal vesicles. Use of the human specimens in this study was approved by the institutional human investigation committee.

Each of the primary prostate cancers was first screened for mutation by using the PCR-SSCP approach. Primers used for each PTEN exon were the same as described previously (16). PCRs for the SSCP contained 5–10 ng of genomic DNA, 1× PCR buffer [20 mm Tris-HCl (pH 8.8), 10 mmKCl, 10 mm(NH₄)₂SO₄,2 mm MgSO₄, 0.1% Triton X-100, and 0.1 mg/ml BSA], 1 μm of each primer, 3 μm of each deoxynucleotide triphosphate, 1 μCi of [α-³²P]dCTP (3000 Ci/mmol), 0.6 unit of Taq DNA polymerase, and 0.1 unit of Pfu DNA polymerase and was incubated at 95°C for 5 min followed by 30 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 1 min. The ³²P-labeled PCR products were electrophoresed at 5 W overnight at room temperature in a 6% nondenaturing polyacrylamide gel in 1× TPE buffer [30 mm Tris,20 mm PIPES (1,4-piperazinediethanesulfonic acid), and 1 mm Na₂EDTA (pH 6.8)] as described previously (27). PCR products were also analyzed in a 0.25× MDE gel(FMC BioProducts, Rockland, ME) containing 10% glycerol, which was also run at 5 W overnight at room temperature. After electrophoresis,the gels were dried and exposed to X-ray film for 1–2 days. Samples showing a bandshift for a specific exon were reamplified for both tumor DNA and matched nonneoplastic cells to confirm the shift, using the same conditions.

For the samples which repeatedly showed a bandshift in the SSCP analysis, shifted bands were cut and immersed in 20 μl of H₂O, following the protocol described by Kukita et al. (27). Two μl of the released DNA were amplified by PCR using the same primers, as in SSCP analysis, in a 50-μl of reaction. The PCR conditions were the same except that 200μ m of each deoxynucleotide triphosphate and no[³²P]dCTP were used. These PCR products were purified by using the QIAquick PCR purification kit (Qiagen, Valencia,CA), and were sequenced by using the ThermoSequenase Cycle Sequencing kit (USB) following the manufacturer’s instructions. Sequencing data were collected and analyzed by using the ScanDNASIS and MacDNASIS software (Hitachi Software, San Bruno, CA).

For the six metastases of prostate cancer, which tended to be more homogeneous in neoplastic cells, their DNAs were amplified by PCR for each of the PTEN exons, and the resultant PCR products were purified and directly sequenced by the same procedure as described above. For an exon showing a mutation, the PCR sequencing procedure was repeated to confirm the mutations. Once confirmed, matched normal DNA for a specific exon was also amplified by PCR and sequenced to determine whether a mutation was somatic or germ line.

The difference in the frequency of PTEN mutations between primary tumors in the current study and that of our previous study (16) was analyzed statistically by the use of Fisher’s exact test (two-tailed; Ref. 28).

Seventy % of the 32 primary prostate cancers from Chinese men were Gleason scores 8–10, whereas 30% were Gleason score 7. PCR-SSCP and direct DNA sequencing analyses of these samples revealed PTEN sequence alterations in 7 cases. Examples of bandshifts for tumors in SSCP assay, which indicated the existence of sequence alterations in the PTEN gene, are shown in Fig. 1, and examples of DNA sequencing ladders that identify PTENmutations are shown in Fig. 2. Tumor cases and their PTEN mutation status are listed in Table 1. Although 2 of the 7 cases had alterations that did not change the PTEN polypeptide, five cases (16%) had mutations that could potentially change PTEN function (Table 1). Case 113 had a nonsense mutation at codon 20 that would truncate the majority of the PTEN protein. Case 92 had two missense mutations in its exon 5, which changed codons 101 and 135 from isoleucine to alanine and valine,respectively. Cases 91, 109, and 114 showed missense mutations that changed codons 55, 150, and 272 from arginine, glutamine, and histidine to glycine, glycine, and tyrosine, respectively.

We also analyzed six metastases of prostate cancer from American men,using the methods of PCR amplification and direct DNA sequencing. Two cases showed PTEN mutations. Case 49 had a nonsense mutation at codon 110 in exon 5 that would truncate the PTEN protein, and case 47 had three missense mutations in exon 9 of PTEN, changing codon 344 from lysine to arginine, codon 348 from threonine to isoleucine, and codon 382 from threonine to serine.

Each of the above mutations was somatic, as the matched nonneoplastic cells showed no mutations. The difference in the frequency of 16% for PTEN mutation in the cancers from Chinese patients compared with the frequency of 2.5% in our prior analysis of 40 resected primary tumors detected in American men after PSA test and biopsy (16) showed a trend in significance (P =0.08).

The PTEN gene was isolated from the q23 region of chromosome 10, one of the most frequently deleted regions in prostate cancer (4, 29, 30). Mutations of the gene have been detected in various human cancers including that of the prostate (9, 12, 13, 19, 20, 21), implicating PTEN in the development and/or progression of prostate cancer. It is thus far the most frequently mutated gene in prostate cancer. Our finding of PTEN mutations in 5 of 32 primary, high-grade prostate cancer specimens confirms that PTEN is a major gene, if not the target gene, for the 10q23 region of deletion in a subset of prostate cancers.

Mutation frequencies of PTEN in prostate cancer differ among studies, largely because of differences in tumor grade and stage in the study populations. Mutations up to 60% have been detected in studies of prostate cancer cell lines and xenografts from metastases (13), whereas in some studies of localized disease, few or no mutations have been detected (16, 17). In this study,we detected PTEN mutations in 5 of 32 (16%) primary prostate cancers from Chinese patients who were diagnosed with clinical symptoms but without the aid of the serum PSA screening test. This frequency was higher than that (1 of 40 or 2.5%) detected in primary prostate cancers from American patients who were diagnosed by PSA test in our previous study (16). The majority of tumors from the Chinese patients were high grade (Gleason scores 8–10), whereas the majority of tumors in the American patients were lower grade(Gleason scores 5–7), indicating that PTEN mutations occur more often in tumors with high Gleason scores, even in those that are primary lesions. This conclusion is consistent with published studies of primary prostate cancers (15, 17, 20). In one study of 37 primary tumors with 20 (54%) high-grade and 17 (46%) lower grade lesions, five cases, four of which were high-grade tumors, had PTEN mutations (20). In another study of 45 primary tumors that were mainly low-grade cancers [30 (67%) lower grade cases and 15(33%) high-grade cases], no PTEN mutations were found (17). Summarizing five studies in which both tumor grade and PTEN mutations were available (15, 16, 17, 20), we found that 9 of 67 (13.4%) high-grade tumors showed PTEN mutations, whereas only 3 of 117 (2.6%) lower grade cases showed mutations. The former rate is significantly higher than the latter (P = 0.01)using the χ² analysis-of-contingency table (28). Consistent with mutation studies, loss of PTEN expression has also been shown to correlate with high grade of primary prostate cancer (9, 10).

It has been reported that prostate cancer incidence is lower in Asian men than in Western men (24, 25). Although one study of Japanese patients did not detect any PTEN mutations in 45 primary tumors that were mainly low-grade cancers (17), we found more frequent PTEN mutations in a group of Chinese patients that had mainly high-grade tumors in this study; the latter is consistent with studies in Western men (20). These results suggest that PTEN is likely not a genetic factor contributing to the racial difference in prostate cancer incidence. This conclusion is further supported by the fact that all of the PTEN mutations were detected in prostate cancer cells only and not in their matched nonneoplastic cells. Also, no PTENmutation has been detected in familial prostate cancers (22, 23). The differences in PTEN mutation rates in our study compared with that of Orikasa et al. (17) may be attributable to differences in the distribution of tumor grades between the study samples.

We detected multiple mutations for PTEN in two tumors, i.e., case 92 had two missense mutations in exon 5 and case 47 had three missense mutations in exon 9 (Table 1). The heterogeneous nature of prostate cancer is well known (31); therefore,it is likely that multiple mutations of PTEN in one tumor may come from different subclones of tumor cells. In an analysis of metastases involving multiple organ sites in patients who died of prostate cancer, Suzuki et al. (12) found that different metastases within the same patient had different PTEN mutation status, indicating a complex genetic relationship between various subclonal lineages of prostate cancer cells. Mutation of exon 5 appears to be more frequent than that of other exons in both Cowden disease and various somatic cancers (8).

In summary, PTEN mutations were seen more often in primary prostate cancers from Chinese men compared with localized tumors from American patients. This difference is likely attributable to the presence of an excess of high-grade cancers in the Chinese patients. Whether primary prostate tumors with PTEN mutations have a greater proclivity to metastasize than those of similar grade and stage without mutations remains to be determined.