Characteristics of prostate-derived growth factors for cells of the osteoblast phenotype.

We examined the characteristics of mitogens extracted from human benign prostatic hyperplasia and prostatic adenocarcinoma tissue. Although mitogens for fetal rat skin fibroblasts as well as for rat calvarial osteoblasts and osterosarcoma cells were found, distinct entities that acted selectively in cells of the osteoblast phenotype could be obtained by sequential reverse-phase high performance liquid chromatography. Two peptides with apparent molecular weights of 10,000 and 13,000 D were derived from hyperplastic tissue, whereas a single moiety of 10,000 D was obtained from malignant tissue. These entities increased cell numbers and alkaline phosphatase activity in osteoblastlike cells consistent with effects on both growth and differentiation. Prostatic peptides did not stimulate adenylate cyclase in osteosarcoma cells. Mitogenic activity selective for osteoblastlike cells was identified in postpubertal but not prepubertal normal prostate. The results demonstrate the existence of osteoblastic growth factors in prostatic tissue whose presence may accompany postpubertal development.

Retention of chromosome 3 in extrapulmonary small cell cancer shown by molecular and cytogenetic studies.

In small cell lung carcinoma, one of the short arms of chromosome 3 is typically lost. To investigate chromosome 3 in extrapulmonary small cell carcinoma, we used DNA probes that detect restriction-fragment-length polymorphisms at loci on 3p. These probes were used to study DNA extracted from tumors and normal tissues and/or tumor cell lines from five patients with extrapulmonary small cell cancer. Tumor DNA from four of the five patients with extrapulmonary small cell cancer retained heterozygosity at loci on 3p. Cytogenetic studies of the tumor cell lines established from these four patients showed retention of both short arms of chromosome 3. We conclude that the loss of genetic material from 3p observed in small cell lung cancer is not typical in extrapulmonary small cell cancer.

Distinctive protein pattern in two-dimensional electrophoretograms of cancerous prostatic tissues.

In this report, we describe methods used to analyze the protein composition of sectioned frozen prostatic tissues by two-dimensional gel electrophoresis. Our results show a high degree of homology in two-dimensional electrophoretograms of proteins extracted from frozen sections of malignant prostate glands. Such homology was not apparent in protein patterns of benign hypertrophic prostatic tissue sections. Typically, 600 discrete proteins were resolved on two-dimensional electrophoretograms and 9 proteins were present in all patterns of prostate adenocarcinomatous tissues. These nine proteins were not observed in any of the protein electrophoretograms developed from nonmalignant prostate tissue. Three proteins were found common to nonmalignant prostate glands but were not present in prostatic adenocarcinoma.

Inability of cytoplasmic or nuclear androgen receptor content or distribution to distinguish benign from carcinomatous human prostate.

We used exchange saturation analysis at 15 degrees to quantitate total cytoplasmic and nuclear androgen receptor content of 70 patient specimens. Cytoplasmic androgen receptor contents (fmol/mg DNA) for eight specimens of clinically benign hyperplasia, 14 specimens of histologically hyperplastic prostate obtained at cystoprostatectomy, and carcinomatous and noncarcinomatous prostate obtained at radical prostatectomy for prostatic carcinoma, 48 specimens, respectively, were 830 +/- 165 (mean +/- S.E.), 890 +/- 445, 955 +/- 240, and 750 +/- 95. Nuclear androgen receptor contents of these same specimens, respectively, were 275 +/- 40, 235 +/- 30, 345 +/- 25, and 350 +/- 30; whereas, the values of the cytoplasmic/nuclear receptor content, respectively, were 3.25 +/- 0.55, 3.05 +/- 0.80, 2.50 +/- 0.50, and 2.80 +/- 0.40. Multiway analyses of variance of these cross-sectional data showed that there was no significant difference (p greater than 0.05) between group mean values. This result principally reflects the fact that the families of values for the four tissue groups were highly heterogenous with broad overlap. The results would not appear to be unduly influenced by carcinomatous epithelial cell content of the specimens, because cytoplasmic and nuclear androgen receptor content were not related to specimen carcinomatous epithelial cell content. Paired analyses of receptor content in carcinomatous and noncarcinomatous prostate specimens from the same prostate showed enhanced or unchanged receptor content in 58% (cytoplasmic) and 62% (nuclear) of specimens. Our studies show that cross-sectional analyses of androgen receptor content fail to distinguish carcinomatous prostate from noncarcinomatous prostate. However, paired analyses of these tissues from the same gland identify distinguishing differences. The clinical relevance of these observations remains to be examined.

Carcinoembryonic antigen-like substance derived from human prostate.

Carcinoembryonic antigen-like substance, previously detected in large amounts in the medium from cultures of human prostatic epithelial cells, also is present in extracts of benign and malignant human prostate. By column chromatography, the prostate-derived carcinoembryonic antigen-like substance derived from cultured prostate is the same as that in tissue extracts and is distinctly different from colon-derived carcinoembryonic antigen. The molecular weight of prostate-derived carcinoembryonic antigen-like substance is estimated to be greater than 5 x 10(5). Prostate-derived carcinoembryonic antigen-like substance may be a prostate-specific substance.

Two-dimensional electrophoretic analysis of human prostatic fluid proteins.

The heterogeneous mixture of proteins present in expressed prostatic fluid from forty-nine men over 25 years old was analyzed on two-dimensional electrophoretic gels stained with silver. Samples were collected from men diagnosed with chronic prostatitis, urinary tract infection, benign prostatic hypertrophy (BPH), and prostatic carcinoma. Reference patterns were developed for all proteins observed in electrophoretograms. Over 1000 distinct proteins are contained in the comprehensive proteins pattern of prostatic fluid; however, only 18 proteins were common to all samples. No proteins unique to fluids from men with BPH or prostatic cancer were observed. Some proteins in the common proteins reference pattern of BPH patients were not seen in samples from men with prostatic cancer. Our preliminary results indicate that a series of proteins tentatively identified as variants of prostatic acid phosphatase appeared elevated in all BPH fluids but was absent or below limits of detection in samples from men with prostatic carcinoma. Our studies indicate that proteins secreted in prostatic fluid reflect physiological changes in the prostate, and certain changes may be useful indicators of malignant transformation.

Prognostic importance of prostate-specific antigen for monitoring patients with stages B2 to D1 prostate cancer.

To evaluate the prognostic value of prostate-specific antigen (PA) for detection of tumor growth after definitive therapy, 602 sera from 70 patients with stages B2 to D1 prostate cancer (26 of whom recurred) were analyzed in a blind study. Using Cox's proportional-hazards model, a highly significant association was found between serially measured PA and disease-free survival time (p = 0.0002). A positive predictive value of 100% was found for some markedly elevated PA levels and confirmed recurrence of disease. In fact, this study suggested that once a PA level of 88 ng/ml was reached, there was an average time of less than 2 months before a recurrence was clinically confirmed. Tumor growth in patients who recurred was indicated by a PA elevation before recurrence in 92% (24 of 26) as opposed to 20% (9 of 44) in disease-free patients. Additionally, in these 24 of 26 patients, levels of PA were elevated 12 months (mean lead time) before a confirmed disease recurrence. In patients who were still disease free, serial PA appeared to increase concurrently with putative tumor growth as shown by the initial surgical stage. Generally, the greater the PA level the more advanced was the stage of disease (B2 to D1). These data suggest that PA may be a useful adjuvant marker for monitoring tumor growth in patients with regionally confined prostate cancer.

Keratin immunoreactivity in the benign and neoplastic human prostate.

Keratin immunoreactivity in the benign and neoplastic human prostate was examined immunohistochemically using two monoclonal antibodies with differing specificities. One of these antibodies stained only the basal cells of the normal and hyperplastic prostatic epithelium, with no reactivity in tumor cells of prostatic adenocarcinoma. The other monoclonal antibody recognized a keratin protein present in all normal and hyperplastic columnar (secretory) epithelial cells, as well as in all cancer cells regardless of degree of tumor differentiation. In addition, the second antibody stained acinar and ductal epithelial cells exhibiting premalignant changes. Our findings indicate that keratin immunoreactivity differs among the epithelial cell populations of the human prostate, probably reflecting expression of different keratin proteins. The distinctive patterns of staining obtained with these two antibodies may assist in distinguishing hyperplastic from neoplastic prostatic epithelium, as well as in the recognition of basal cell hyperplasia, transitional cell metaplasia, and premalignant changes.

Heparin binding affinity of rat prostatic growth factor in normal and cancerous prostates: partial purification and characterization of rat prostatic growth factor in the Dunning tumor.

The rat prostate contains two types of growth factors capable of stimulating DNA synthesis in BALB/3T3 cells. These rat prostatic growth factors (RPGF) were separable by a different affinity for heparin: low affinity type RPGF and high affinity (HiA) type RPGF. About 80% of the RPGF in the cytosol from normal prostates was low affinity type, whereas more than 80% in the cytosol from the Dunning tumors was HiA type. Elution profile of HiA-RPGF showed two peaks of activity eluted from the heparin-Sepharose column, one at 1.3-1.4 M NaCl (HiA1-RPGF) and the other at 1.6-1.7 M NaCl (HiA2-RPGF). HiA2-RPGF could be purified 1100-fold from the Dunning tumor (AT-3 subline) in about 20% recovery by heparin-Sepharose chromatography. The partially purified HiA2-RPGF in the Dunning tumor has a molecular weight of about 19,000 and isoelectric point of about 3.8, and stimulated DNA synthesis at a concentration of about 0.25 nM. The activity was lost by heat treatment at 70 degrees C for 5 min and by acid treatment, whereas it was stimulated by incubating with dithiothreitol. The HiA2-RPGF did not have transforming growth factor activity at a concentration of 250 ng/ml or lower in the presence of epidermal growth factor.

Cytokeratin characterization of human prostatic carcinoma and its derived cell lines.

Two murine monoclonal anti-cytokeratin antibodies with defined specificity were shown to distinguish between basal cells and luminal cells in human prostate tissue. Forty-one biopsies or transurethral resection specimens were characterized using these two antibodies. In cases of benign prostatic hyperplasia, focal loss of the basal cell layer was noted in areas of glandular proliferation. Ten cases of adenocarcinoma of the prostate, varying in Gleason's histological grade from 2 to 4, were also studied. In each case the carcinoma was shown to represent the luminal cell phenotype with no evidence of involvement of the basal cell phenotype. An analysis of three established metastatic prostatic carcinoma cell lines (DU-145, PC-3, and LNCaP) using two-dimensional electrophoresis showed that the cytokeratin complement of each cell line was slightly different but retained the phenotype of the luminal cell. It was concluded that during both hyperplasia and neoplastic transformation of the prostate, the luminal cell phenotype is primarily involved and that the basal cell phenotype does not appear to contribute to either intraluminal proliferation or invasive cell populations.

Subcellular distribution of androgen receptors in human normal, benign hyperplastic, and malignant prostatic tissues: characterization of nuclear salt-resistant receptors.

Two populations of nuclear androgen receptors have been characterized in human prostatic tissue, and the levels and proportions of each were found to differ in normal prostates, benign hyperplastic prostates (BPH), and malignant prostates. A significant percentage (35 to 50%) of total nuclear androgen receptors was associated with the salt-resistant nuclear matrix fraction. The remainder were easily extracted from nuclei by 0.6 M KCl. Optimal conditions for measuring receptors in both compartments involved the use of an inhibitor of proteolysis (phenylmethylsulfonyl fluoride) and the omission of dithiothreitol from buffers. In the presence of dithiothreitol, most of the nuclear salt-resistant receptors were rendered salt extractable. Cytosol androgen receptor levels were not significantly different in normal, BPH, or malignant prostatic tissues. In contrast, the levels and distribution of nuclear salt-extractable and salt-resistant androgen receptors exhibited characteristic patterns. Compared to normal prostatic tissue, nuclear salt-extractable receptors were significantly elevated in both BPH and cancer, whereas nuclear salt-resistant receptors were elevated in BPH but not in cancer. The ratio of salt-extractable to salt-resistant receptors was approximately 1:1 in both normal and BPH tissues and 2:1 in cancer. In addition, a microassay has been developed for the measurement of androgen receptors in the three subcellular compartments of needle biopsy specimens of prostatic cancer. Studies are in progress to determine whether the measurement of both nuclear salt-extractable and salt-resistant receptors may improve the usefulness of receptor levels to predict the hormonal responsiveness of prostatic cancer.

Immunocytochemical assay for estrogen receptors applied to human prostatic tumors.

An immunocytochemical assay using a monoclonal antibody to estrogen receptor was applied to frozen sections of 43 prostatic tumors. In addition, in 16 tumors the results of the immunocytochemical assay were compared with those obtained using an enzyme immunoassay for estrogen receptor, a tritiated ligand binding assay, and a histochemical procedure using fluorescein-labeled estradiol conjugates. Specimens from 14 patients with benign prostatic hyperplasia and 29 patients with prostatic carcinoma were analyzed in assays in which human breast carcinoma specimens of high, medium, low, or negative estrogen receptor content acted as controls. The intensity of nuclear staining and the percentage of stained cells in the breast sample controls correlated well with estrogen receptor content as determined by both the enzyme immunoassay and the tritiated ligand binding assay. None of the fixed, frozen sections from the 43 prostatic tumors exhibited nuclear staining of either the benign or the malignant epithelial cells or of the stromal components. Negative results were also obtained with the tritiated ligand binding assay. The majority of prostatic samples were negative when using the enzyme immunoassay but four specimens had a mean value of 6.2 fmol/mg protein which is just above the sensitivity of the assay. Using fluorescein-labeled estradiol conjugates both cytoplasmic and nuclear binding was observed in the prostatic samples which did not correlate with the results obtained by the other three procedures.

Hormone dependency of a serially transplantable human prostatic cancer (HONDA) in nude mice.

Human prostatic cancer (HONDA) serially transplanted in nude mice grew well in male mice but not at all in untreated female mice or in castrated male mice. Progressive growth in female mice was obtained by i.m. administration of 1 mg of testosterone twice a week. Estradiol inhibited the growth of the tumor in male mice to some extent; however, some growth was observed. The tumor in untreated male mice retained the histological features of poorly differentiated adenocarcinoma. Tumors in castrated male mice showed reduction in size of tumor cell nests with relative overgrowth of stroma. The tumor in androgenized female mice consisted of columnar epithelial cells with large nuclei and more abundant cytoplasms and a large glandular lumen, showing histology of moderately differentiated adenocarcinoma. High levels of human prostatic acid phosphatase (PAP) were detected in sera from untreated male mice. Testosterone markedly increased the content of serum PAP of androgenized female mice. Estradiol reduced the levels of PAP in sera from untreated male mice regardless of the tumor weight. High-affinity androgen receptors were present in cytosol and in nuclear extract of the tumor in untreated male mice. No measurable amount of progesterone or estrogen receptors was present in cytosol from untreated male mice.

Heparin-binding growth factor gene expression and receptor characteristics in normal rat prostate and two transplantable rat prostate tumors.

Heparin-binding polypeptide growth factors (HBGF) are essential mitogens for isolated prostate cells. HBGF type one (HBGF-1) mRNA was expressed specifically in the epithelial cells of prostates from normal 6- to 8-week-old rats. Expression declined significantly at 14 weeks and was undetectable in 35-week-old animals. Slow-growing, androgen-responsive, nonmetastatic Dunning R3327PAP tumors, which are composed of a well-defined epithelium and stroma, expressed HBGF-1 mRNA constitutively in specifically the mesenchymal cells. A rapid-growing, androgen-independent, metastatic variant (Dunning R3327AT-3), which was composed of a single clonogenic cell type, expressed both HBGF-1 and HBGF type two (HBGF-2) mRNA. HBGF activity in the extracts of normal and tumor tissues correlated with mRNA levels. Epithelial cells from the R3327PAP tumor and the single cell type that composed the R3327AT-3 tumor exhibited alterations in HBGF receptor characteristics that correlated with increased sensitivity to mitogenic effects of HBGF. The results suggest that alterations in HBGF gene expression in both prostate epithelial and mesenchymal cells and in properties of the receptor in specifically epithelial cells may contribute to differential growth rates and malignancy of different prostatic tumors.

Correlation between clinical response to hormone therapy and steroid receptor content in prostatic cancer.

Hormonal therapy is the dominating form of treatment for prostatic carcinoma. The majority of cases (80%) are well controlled for varying times with this regimen. However, thus far there have been no adequate methods to predict in which cases hormonal therapy is of less benefit. Measurement of cancer tissue content of intracellular hormone receptors constitutes progress toward a more individualized therapy in prostatic carcinoma. In this study biopsies from 16 cancer patients were taken before therapy was given, and the specimens were analyzed with regard to content of specific methyltrienolone-binding sites. A correlation has been made between receptor content and clinical response to hormonal therapy in each case. Twelve specimens contained measurable amounts of steroid receptors. Of these, one patient died during irradiation therapy before onset of hormonal treatment. However, of the remaining 11 patients, 9 responded well to hormones (9/11 approximately 82%). The two receptor-positive nonresponders had the lowest measurable receptor levels in the series. Four specimens contained no detectable amounts of receptors. Three of these patients showed no response to therapy (3/4 = 75%) but one was "false negative." Our data indicate that steroid receptor analysis may become a valuable diagnostic tool in individualizing the therapy for prostatic cancer.

Immunocytochemical evaluation of human prostatic carcinomas for carcinoembryonic antigen, nonspecific cross-reacting antigen, beta-chorionic gonadotrophin, and prostate-specific antigen.

The unlabeled antibody peroxidase-antiperoxidase technique was used to examine human malignant prostatic tissue (primary tumors) for the presence of prostate-specific antigen (PSA), carcinoembryonic antigen (CEA), nonspecific cross-reacting antigen (NCA), and beta-chorionic gonadotrophin (HCG). The results were compared to those obtained with normal and hyperplastic prostate tissue (BPH). All specimens of neoplastic, hyperplastic, and normal prostate tissue showed immunostaining reactions for PSA. Immunostaining for PSA was relatively uniform among samples of normal and BPH tissue, but variations with respect to intensity of PSA immunostaining were noted among prostate tumors as well as between the neoplastic cells of individual tumors. Some areas of normal or hyperplastic prostatic epithelium within tumors showed stronger staining reactions for PSA than the tumor cells themselves. Using an antiserum which was able to detect both NCA and CEA, it was found that 16 of 38 tumors (42%) had positive immunostaining reactions. Of these, 15 were subsequently shown to contain only NCA immunoreactivity, and 1 tumor had both NCA and CEA immunoreactivity. NCA, but not CEA, immunoreactivity was identified in hyperplastic prostate tissue within tumor specimens and in BPH specimens. Neither antigen was detected in normal prostatic epithelium. Three of 38 tumors (8%) were found to contain neoplastic cells with HCG immunoreactivity. HCG immunoreactivity was not identified in BPH or normal prostatic tissue. Therefore, HCG and CEA immunoreactivity appear to be tumor-associated antigens in prostate cancer which are expressed with a low incidence. The results of the study identified prostate tumors with different patterns of immunocytochemical markers: 22 of 38 tumors (58%) contained only PSA immunoreactivity; 13 of 38 tumors (34%) contained PSA and NCA immunoreactivity; 2 of 38 tumors (5%) were positive for PSA, NCA, and HCG immunoreactivity; and 1 of 38 tumors (3%) contained PSA, NCA, HCG, and CEA immunoreactivity. Apart from PSA, which was present in all tumors, the markers studied here appeared to be more frequently expressed in well-differentiated tumors than in less-differentiated tumors. Our results suggest the possibility of subclassifying prostate tumors by means of immunocytochemistry.

The modified nucleotide constituents of human prostatic cancer cell (MA-160) poly(A)-containing RNA.

Cytoplasmic poly A(+) RNA from human prostatic cancer cells grown in the presence of 32P was isolated by affinity chromatography on columns of oligo(dT)-cellulose. The RNA was digested with RNAase T2 and the products of digestion were fractionated by two-dimensional electrophoresis. The resulting autoradiograms revealed the presence of many different cap groups as well as two internal modified nucleotide components. 19 different type 1 and type 2 'cap' groups were identified. The internal modified nucleotides were N6-methyl adenosine and a 2'-O-methyl nucleotide possessing an unusual modified base.

Molybdate and the measurements of androgen receptors in prostate cancer.

Using the sucrose density gradient technique, we have compared androgen receptor values obtained in the presence and absence of sodium molybdate from cytosols of surgically obtained human prostate cancer tissue. In all nine prostates examined, inclusion of molybdate in the buffers resulted in an increase in the amount of androgen receptor detected, in some cases dramatically. Since accurate androgen receptor assays could be valuable for therapeutic decision making in prostate cancer, we advocate the inclusion of sodium molybdate in assay buffers to guard against the possibility of false negative or artificially low values.

Evaluation of receptors for somatostatin in various tumors using different analogs.

The binding characteristics of several somatostatin (SS-14) analogs developed in our laboratory were examined in various human and animal tumors and normal tissues. In rat cerebral cortex and human breast cancer membranes the interaction of SS-14 with its binding sites was rapid, specific, saturable, linear with protein concentrations, and dependent on time and temperature. Analysis of kinetic and equilibrium experimental data showed that the interaction of [125I-Tyr11]SS-14 with the binding sites in all normal and tumoral tissue specimens was consistent with the presence of a single class of noncooperative binding sites. Superactive octapeptide analogs of somatostatin-containing hexapeptide sequences Cys-Phe-D-Trp-Lys-Thr-Cys or Cys-Tyr-D-Trp-Lys-Val-Cys showed significant binding affinities to SS-14 receptors. Among these analogs, D-Trp-Cys-Phe-D-Trp-Lys-Thr-Cys-Thr-NH2 (RC-98-I) showed the highest binding affinity to normal human pancreatic tissue and human pancreatic adenocarcinoma. In contrast, Sandostatin (SMS 201-995) bound only to normal pancreas, not to human pancreatic cancers. Analog RC-98-I also showed a high binding to human and rat prostate cancers. In human epithelial ovarian cancers and an arrhenoblastoma, analogs D-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-Trp-NH2 (RC-95-I), D-Phe-Cys-Tyr-D-Trp-Lys-Val-Cys-Thr-NH2 (RC-121) and D-Phe-Cys-Tyr-D-Trp-Lys-Val-Cys-Trp-NH2 (RC-160) appeared to be the most potent in displacing labeled SS-14. Analogs Ac-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-Thr-NH2 (RC-101-I) as well as RC-121, RC-160, and RC-95-I, but not SMS-201-995, showed high binding affinity in human breast cancers. In specimens of human meningioma the highest binding was found with analogs RC-121, RC-95-I, and RC-101-I. Since marked variations in binding affinities were noted for several analogs in the tissues of origin and the tumors, this suggest that differences may exist between somatostatin receptors not only in normal vs. cancerous tissues, but also among various tumors. Our findings also imply that some analogs could be therapeutically superior to others in the treatment of certain tumors.

High level of expression in the prostate of a human glandular kallikrein mRNA related to prostate-specific antigen.

Using a synthetic oligonucleotide primer complementary to human prostate-specific antigen mRNA, we found that an additional sequence possibly similar to human glandular kallikrein-1 could be read by a primer-extension sequencing technique. We were able to confirm the identity of that additional sequence with another oligonucleotide primer complementary to a specific region of the human glandular kallikrein-1 mRNA sequence. Northern blot analysis with 2 oligonucleotide probes respectively specific for prostate-specific antigen and human glandular kallikrein-1 mRNAs showed that the length of both mRNAs was similar at 1.5 kb. The level of human glandular kallikrein-1 mRNA relative to that of prostate-specific antigen could be estimated as approx. 10-20%. This study constitutes the first evidence that the human glandular kallikrein-1 gene is expressed at a high level in a human tissue.