Nuclear matrix. Isolation and characterization of a framework structure from rat liver nuclei.

A nuclear framework structure termed the nuclear matrix has been isolated and characterized. This matrix forms the major residual structure of isolated nuclei and consists largely of protein with smaller amounts of RNA, DNA, carbohydrate, and phospholipid. The nuclear matrix can be further resolved by combined treatment with DNase and RNase. The remaining nuclear protein structure, after extraction of 90 percent of the nuclear protein, 99.9 percent of the DNA, and 98 percent of the RNA and phospholipid, is termed the nuclear protein matrix. Electron microscopy of this final nuclear protein matrix reveals an interior framework structure composed of residual nucleolar structures associated with a granular and fibrous internal matrix structure. The internal matrix framework is derived from the interchromatinic structures of the nucleus, and is connected to a surrounding residual nuclear envelope layer containing residual nuclear pore complex structures. Sodium dodecyl sulfate-acrylamide gel electrophoresis of the nuclear matrix proteins demonstrates three major polypeptide fractions, P-1, P-2, and P-3, with average molecular weights of approximately 69,000, 66,000 and 62,000, as well as several minor polypeptides which migrate at approximately 50,000 and at higher molecular weights (>100,000). Polypeptides with molecular weights identical to those of P-1, P-2 and P-3 are also components of isolated nuclear envelopes and nucleoli, whereas isolated chromatin contains no detectable matrix polypeptides. This suggests that the major matrix polypeptides are localized in specific structural regions of the nucleus, i.e., nuclear envelope, nucleoli, and interchromatinic structures. The presence of cytochrome oxidase activity in the isolated nuclear matrix indicates that at least some integral proteins of the nuclear membrane are associated with the matrix.

Cyclosporine A, an in vitro calmodulin antagonist, induces nuclear lobulations in human T cell lymphocytes and monocytes.

Cyclosporine A is a noncytotoxic, natural, 11 amino acid cyclic peptide used clinically as an immunosuppressant to prevent organ rejection after transplantation. Cyclosporine A is an in vitro calmodulin antagonist. At the low concentrations required to inhibit calmodulin-dependent phosphodiesterase in vitro, cyclosporine A causes a dramatic alteration in the nuclear morphology of 23% of human peripheral blood mononuclear leukocytes in vitro without loss of viability. The shape of the nucleus changes from ovoid to a distinctive, radially splayed lobulated structure. The changes occur in a dose-dependent manner in 60 min at 37 degrees C. Specific monoclonal antibodies to human leukocytes identify the cells susceptible to nuclear lobulation by cyclosporine A as OKT4 antigen-positive T cell lymphocytes and monocytes. The lobulated nuclei are 2N as determined by flow cytometric measurement of ethidium bromide fluorescence of DNA. The cyclosporine A-induced lobulation of T cell nuclei requires both physiologic temperature and metabolic energy. Although structurally different than cyclosporine A, the calmodulin antagonists R24571 and W-7 [N-(6-aminohexyl)-5-chloro-1-naphthalene-sulfonamide] also produce T cell nuclear lobulations that are indistinguishable from the nuclear lobulations caused by cyclosporine A. These data indicate that nonmitotic structural elements that govern normal nuclear morphology in a subset of mononuclear leukocytes appear to require a calmodulin-mediated process. Cyclosporine A may be a useful noncytotoxic inhibitor of calmodulin-dependent systems that influence nuclear structure and function.

Structural alteration in isolated rat liver nuclei after removal of template restriction by polyanions.

Specific polyanions release DNA template restrictions for DNA synthesis in isolated rat liver nuclei. The degree to which DNA synthesis is enhanced can be correlated with a spectrum of changes in nuclear structure Each polyanion which is effective in the release of template restriction produces a characteristic alteration in nuclear ultrastructure. Polyanions which have no effect on DNA synthesis do not appear to cause any change in nuclear organization or ultrastructure. Parallel measurements of nuclear DNA release and nuclear volume changes also indicate that template-activating polyanions cause remarkable changes in the structural organization of the treated nuclei. These results indicate that DNA template activation involves direct interactions between polyanions and nuclear constituents and suggest the possibility that naturally occurring polyanions might have a role in the control of gene activity

Nuclear protein matrix: association with newly synthesized DNA.

The residual structural framework of the cell nucleus termed the nuclear protein matrix, is associated with newly synthesized DNA in regenerating rat liver. One minute after rats are injected with [3-H] thymidine, more than 90 percent of the total tritium in nuclear DNA is associated with the matrix DNA although this DNA comprises only 25 percent of the total nuclear DNA. In contrast, the bulk DNA, 75 percent of total nuclear DNA, contains less than 8 percent of the total labeled DNA. The percentage of total labeled DNA associated with the bulk DNA increases for 30 minutes after injection and decreases correspondingly in the matrix DNA.

Comparison of spontaneous and experimentally induced canine prostatic hyperplasia.

Spontaneous prostatic hyperplasia in the beagle appears to progress with age from a glandular to a cystic histological appearance. Prostatic hyperplasia can be induced in young beagles with intact testes by treatment for 4 mo with either dihydrotestosterone or 5 alpha-androstane-3 alpha, 17 beta-diol, alone, or with either of these steroids in combination with 17 beta-estradiol. In contrast, the induction of prostatic hyperplasia in young castrated beagles, in which the gland had been allowed to involute for 1 mo, requires the administration of both 17 beta-estradiol and either 5 alpha-androstane-3 alpha, 17 beta-diol or dihydrotestosterone. Testosterone and 17 beta-estradiol, either singly or in combination, did not produce the hyperplastic condition in intact or castrated beagles. The experimentally induced prostatic hyperplasia is identical in pathology to the glandular hyperplasia that occurs naturally in the aging dog with intact testes. However, cystic hyperplasia was not produced by any of the treatments tested in young animals.

Spontaneous benign prostatic hyperplasia in the beagle. Age-associated changes in serum hormone levels, and the morphology and secretory function of the canine prostate.

This paper is a cross-sectional study of spontaneous benign prostatic hyperplasia (BPH) in a single canine species. The effects of aging and hormonal changes on the growth, histology, and glandular secretory function of the canine prostate were studied in 42 male beagles ranging in age from 8 mo to 9 yr. The beagle prostate enlarges for at least 6 yr, whether normal or hyperplastic. In contrast, prostatic secretory function, determined by ejaculate volume and total ejaculate protein, declines markedly after 4 yr of age. These reciprocal growth and functional changes in the prostate are closely associated with a progressive increase in the incidence of BPH, which is already apparent in some dogs by age two. With age there is a modest decrease in serum androgen levels with no apparent change in serum 17 beta-estradiol levels. This suggests that the growth and functional changes that are associated with the development of BPH and are initiated very early in life reflect an altered sensitivity of the prostate to serum androgens or a response to the relative decrease in the serum androgen to estrogen ratio.

Preneoplastic alterations in nuclear morphology that accompany loss of tumor suppressor phenotype.

Alterations of nuclear shape are frequently observed in tumor cells, but the genes controlling these changes and the stage in the neoplastic process at which they occur are unknown. We have studied nuclear shape changes in chemically immortalized, nontumorigenic Syrian hamster embryo cell clones that had either retained (supB+) or lost (supB-) the ability to suppress the tumorigenic phenotype when they were hybridized with a tumor cell line (BP6T). Quantitative morphometric analysis of the nuclei of cells from each of two pairs of supB+/supB- variants indicated that the nuclei of supB- cells were significantly more out of round than those of their corresponding supB+ clones. These data indicate that modification of nuclear structure may represent an early, preneoplastic event in multistep chemical carcinogenesis and that loss of a tumor suppressor gene function may regulate alterations in nuclear morphology.

Organ site specificity for cancer in chromosomal instability disorders.

DNA rearrangement rather than point mutation is an emerging hypothesis for human carcinogenesis. Although there is no direct evidence for this hypothesis, indirect evidence is provided by cancer cytogenetics and genetics. It has been suggested that patients with Bloom's syndrome, a disorder of spontaneous chromosomal rearrangement, develop the common fatal internal cancers and thus that genetic rearrangements, rather than chemical mutagens, cause most internal human cancers. To test this observation, we have derived age- and sex-adjusted cancer incidence rate ratios for specific organ sites in patients with three chromosomal instability disorders (Bloom's syndrome, xeroderma pigmentosum, and dyskeratosis congenita) and have found that the increased incidence of cancer in all three disorders is limited to specific and often uncommon organ sites. We conclude that chromosomal instability disorders do not predispose patients to the common fatal internal cancers. Although DNA rearrangement remains a promising concept in human carcinogenesis, the organ site specificity of cancers associated with Bloom's syndrome, xeroderma pigmentosum, and dyskeratosis congenita cannot be used as evidence to implicate this mechanism.

Adaptation versus selection as the mechanism responsible for the relapse of prostatic cancer to androgen ablation therapy as studied in the Dunning R-3327-H adenocarcinoma.

The Dunning R-3327-H rat prostatic adenocarcinoma is a well-differentiated, slow-growing, serially transplantable tumor of spontaneous origin. When intact male rats bearing such an exponentially growing H-tumor s.c. are castrated, tumor growth abruptly stops, demonstrating the initial androgen sensitivity of this tumor. Eventually, however, after an extended period, the tumor invariably relapses and once again appears to grow exponentially. At the time of relapse, the tumor is no longer androgen sensitive but has irreversibly progressed to a completely insensitive state. The mechanism responsible for this irreversible progression has been demonstrated by fluctuation analysis not to be due to environmentally induced adaptation of initially androgen-dependent H-tumor cells to a new androgen-independent state. Instead, the progression is due to the basic heterogeneity of the original H-tumor (i.e., it is composed of a mixture of preexisting clones of both androgen-dependent and androgen-independent tumor cells). Following castration, only the preexisting clones of androgen-independent tumor cells are able to continue exponential growth; the androgen-dependent tumor cells stop proliferating and die. Thus, androgen ablation creates a host environment in which the androgen-independent tumor cells have a highly selective growth advantage over the androgen-dependent cells. Eventually, with time, this selective growth advantage results in a tumor which is completely composed of androgen-independent cells. It is the continuous proliferative growth of these androgen-independent tumor cells which leads to the relapse phenomenon.

Synergistic enhancement by tumor necrosis factor of in vitro cytotoxicity from chemotherapeutic drugs targeted at DNA topoisomerase II.

Recombinant human tumor necrosis factor (rHTNF) alone had no effect on L929 tumor cells at 100 units/ml for 20 h of continuous exposure. However, under the same conditions, rHTNF markedly enhanced the cytotoxicity of Adriamycin, actinomycin D, 4'-(9-acridinylamino)-methanesulfon-m-anisidide, teniposide (VM 26), and etoposide (VP 16), all targeted at DNA topoisomerase II. The rHTNF had a minimally enhancing effect on the cytotoxicity of bleomycin, hydroxyurea, and 1-beta-D-arabinofuranosylcytosine and no effect on the cytotoxicity of cis-platinum, mitomycin C, vincristine, and vinblastine, all chemotherapeutic drugs with dose-related cytotoxic effects on L929 cells but mechanisms of action which do not appear to involve topoisomerase II. Treatment with rHTNF first and then topoisomerase-targeted drugs yielded no enhanced cytotoxicity, whereas pretreatment with drug followed by rHTNF yielded marked enhancement of cytotoxicity. Topoisomerases have previously been implicated in cell kill phenomena following treatment with certain chemotherapeutic agents [K.M. Tewey, et al., Science (Wash. DC), 226:466-468, 1984]. The data suggest that the lethality to the cell from topoisomerase-targeted drug treatment is increased by rHTNF in vitro. We suggest that rHTNF may be a useful adjuvant to this class of drugs which has well-known antitumor activity.

Models for development of nonreceptor methods for distinguishing androgen-sensitive and -insensitive prostatic tumors.

From the original Dunning R-3327 rat prostatic adenocarcinoma, several distinct sublines have been obtained. These sublines include a well-differentiated, slow-growing, androgen-sensitive tumor (R-3327-H); a well-differentiated, slow-growing, androgen-insensitive tumor (R-3327-HI); and a fast-growing, androgen-insensitive, anaplastic tumor (R-3327-AT). These three sublines were compared in order to develop new model methods for the prediction of the androgen sensitivity and the degree of differentiation of prostatic adenocarcinomas. The R-3327-AT was very distinct in all parameters examined except the tissue protein electrophoretic patterns which contained a uniform pattern in all tumors. The significant differences between R-3327-H and -HI sublines were (a) the inability of testosterone to stimulate DNA synthesis in the R-3327-HI tumor and (b) the difference in the enzymatic profiles of these sublines. The specific activity of three enzymes (3 alpha-hydroxysteroid dehydrogenase, leucine aminopeptidase, lactic dehydrogenase) increased while the activity of another three enzymes (6 alpha,7 alpha-hydroxylase, 5 alpha-reductase, alkaline phosphatase) decreased in the sublines which are androgen insensitive and less differentiated. An arbitrary index was constructed, based upon these enzyme differences, which clearly discriminates the degree of androgen sensitivity and differentiation of these R-3327 rat prostatic adenocarcinomas.

Chaotic oscillations in cultured cells: rat prostate cancer.

Normal prostate epithelial cells exhibit uniformity of structure, function, and DNA content. This uniformity is dramatically perturbed in cancer with the development of variance associated with tumor cell heterogeneity. The development of this kind of diversity is paralleled in models of chaotic oscillators that produce multiple pseudosteady states. We have tested prostatic cancer cells in culture for the presence of chaos by comparing the micromotion of two related rat prostate cancer cell lines that exhibited large differences in motility and metastatic potential. In these extremes of cancer cell types, our data suggest that the three criteria which characterize a chaotic oscillation are fulfilled by their cellular micromotions: (a) absence of defined regularity in the time series as evidenced using Fourier analysis and visual inspection; (b) determinism as evidenced by attractor reconstruction; and (c) sensitive dependence on initial conditions as evidenced by a positive Lyapunov exponent. Cellular motion was studied by using an electronic cell impedance sensor which records, in real time, the fluctuations in the resistive and capacitive properties of cells cultured on recording electrodes. Our data and a preliminary screen of other cell types support a model of established cell lines in culture as chaotic oscillators.

Metastatic potential prediction by a visual grading system of cell motility: prospective validation in the Dunning R-3327 prostatic adenocarcinoma model.

A method for accurate prediction of prognosis in individual patients with prostatic carcinoma does not exist. The limitations of pathological grading systems may result from the failure of standard pathological examination of fixed dead tissue to accurately assess the biological and metastatic behavior of live tumor cells. Many of the sublines of the Dunning R-3327 rat prostatic adenocarcinoma are histologically similar yet differ in metastatic potential. Cells from the Dunning model were grown in culture and filmed by time-lapse videomicroscopy. These cells exhibited characteristic membrane ruffling, pseudopodal extension, and cellular translation that could be graded with 80% reproducibility. Individual cells from sublines with high metastatic potential were separated from cells from sublines of low metastatic potential in 96% of cases. We have applied our cell motility grading system to prospectively classify the metastatic potential of neoplastic cells. The mean motility grades of sublines of high and low metastatic potential differed significantly (Mann-Whitney-Wilcoxon, P less than 0.0005). Among seven sublines in which the grading system was developed, individual cells were correctly classified as high or low metastatic in 71% of cases by ruffling or pseudopodal extension, 73% of cases by translation, and 75% of cases by motility index, an average of the three parameters of motility. Among four newly tested sublines, cells from a low metastatic and high metastatic sublines were perfectly classified. Cells from two other low metastatic sublines were misclassified. When all 88 cells from the 11 sublines were classified, high metastatic cells were detected with 94% sensitivity and 50% specificity. The predictive value of a determination of low metastatic was 93%, whereas the predictive value of an assignment of high metastatic was 52%. The ability to detect and accurately classify most highly metastatic cells while rarely erring in a classification of low metastatic potential suggests that a grading system of cancer cell motility should be evaluated in human prostatic carcinoma. The motility of live prostatic carcinoma cells may predict patient prognosis better than standard pathological grading systems.

Early cell motility changes associated with an increase in metastatic ability in rat prostatic cancer cells transfected with the v-Harvey-ras oncogene.

The development of metastatic ability by cancer cells is a multifactorial process whose temporal events are complex and poorly understood. One step in the metastatic process may involve cell motility. Previous studies reported correlations between motility and metastatic ability. Whether this correlation, seen in cancer cells maintained for long periods of time, is an epiphenomenon developing late in the growth of the cancer as a selection artifact of continuous passage, or is critically required for the acquisition of metastatic ability is unknown. To investigate the relationship between cell motility and the acquisition of metastatic ability, advantage was taken of recently developed DNA transfection methods for inducing high metastatic ability in initially low metastatic cancer cells. The Dunning AT2.1 cell line, a clonal rat prostatic cancer cell line with low metastatic ability, was transfected with a plasmid containing the neomycin resistance gene alone or in combination with the v-Harvey-ras oncogene. A series of the transfected cells was isolated by limiting dilution. After the first in vitro passage following transfection, cells were inoculated into rats to characterize their metastatic ability. The same transfectants were simultaneously studied using our visual grading system of cell motility to study the early motility changes associated with newly acquired metastatic ability. The data demonstrate increased membrane ruffling, pseudopodal extension, and cell translation (translocation) in the v-H-ras-transfected cell lines with high metastatic potential.

Animal models of the hormone-sensitive and -insensitive prostatic adenocarcinomas, Dunning R-3327-H, R-3327-HI, and R-3327-AT.

The Dunning R-3327-H is a well-differentiated transplantable rat prostatic adenocarcinoma that contains both hormone-sensitive and -insensitive cells. The component composed of hormone-insensitive cells has been permitted to grow in a castrated male, and a new slow-growing, well-differentiated hormone-insensitive subline of the tumor has been established and designated R-3327-HI. In addition, a rapidly growing hormone-insensitive anaplastic tumor has been developed, R-3327-AT. These three tumor lines have been characterized, and their histological and biochemical profiles are compared.

Genetic instability coupled to clonal selection as a mechanism for tumor progression in the Dunning R-3327 rat prostatic adenocarcinoma system.

The androgen-sensitive Dunning R-3327-H prostatic adenocarcinoma has been maintained by continuous serial passage in intact male rats for many years. While it has been possible to maintain the original characteristics of the well-differentiated H tumor over 16 years, there have evolved spontaneously, however, more aberrant sublines from this tumor at several subpassages in intact male rats. Serial passage of these individual sublines has established five additional R-3327 tumors each with distinct phenotypes and each more aberrant than the parent H tumor. In addition, it has been possible by passage of the H tumor in castrated male rats to obtain a well-differentiated slow-growing adrogen-insensitive tumor termed the Hl-S tumor. The continuous serial passage of this Hl-S tumor has likewise resulted in the emergence of three new types of Dunning tumors. The results from the biochemical and chromosomal studies presented demonstrate that there is a consistent association in each of these tumor progressions between the expression of genetic instability, which results in the addition of phenotypically new clones of cells to the tumor population, and the subsequent selection of these newly developed clone. These results suggest that the process of genetic instability coupled to clonal selection is one mechanism for the change in tumor phenotype characteristically associated with tumor progression within this system of prostatic tumors.

Identification of nuclear matrix proteins in the cancer and normal rat prostate.

The nuclear matrix is the structural component of the nucleus that determines nuclear morphology and organizes the DNA in a three-dimensional fashion that is tissue specific. Previously, some of the nuclear matrix proteins have been reported to be both tissue and cell type specific and are altered with the state of differentiation and transformation. This study demonstrates that the nuclear matrix is specific for the individual lobes of the normal rat prostate and that the nuclear matrix undergoes changes in protein composition in the Dunning prostate cancer tissue. Additionally, in the Dunning rat prostate adenocarcinoma cell lines, there is a range of tumor phenotypes and the nuclear matrix varies in composition in each tumor cell type. These differences in the nuclear matrix proteins are associated with quantitative changes in nuclear morphology that form the pleiomorphic state of the cancer nucleus.

A novel M(r) 32,000 nuclear phosphoprotein is selectively expressed in cells competent for self-renewal.

We investigated the association between expression of a novel M(r) 32,000 nuclear phosphoprotein (pp32) and cell proliferation in vivo using the well-characterized physiological model of androgen-dependent regeneration of prostate in orchiectomized rats, pp32 is expressed at high levels in neoplastic cell lines and in certain anatomically defined stem cell compartments of normal human tissues such as intestinal crypt epithelial cells. Immunohistochemistry and in situ hybridization were used to monitor pp32 expression in rat ventral prostatic epithelium following castration and androgen restoration. Castrated rats retained only 6% of prostate wet weight compared to intact controls but were capable of complete gland restoration upon androgen replacement. In intact controls, pp32 expression localized to small acini at the periphery of the gland and to rare basal cells in the central regions. Ten days following castration, there was a 3,5-fold enrichment in the frequency of pp32-positive cells with greater than 56% of remaining epithelial cells expressing pp32 protein. In situ hybridization showed that all remaining epithelial cells contained pp32 mRNA. Upon testosterone replacement, pp32 expression and localization returned to that of intact controls. In order to determine the association between pp32 expression and cell division, DNA synthesis was monitored by bromodeoxyuridine incorporation during prostate involution and regeneration. Bromodeoxyuridine incorporation peaked 3 days after androgen replacement and occurred diffusely throughout the gland. Thus, pp32-positive cells are anatomically distinguishable from the population of terminally differentiating cells undergoing rapid expansion. Preliminary immunohistochemical studies of human prostatic neoplasia demonstrated increased expression of pp32 in human prostatic adenocarcinoma and prostatic intraepithelial neoplasia compared to benign prostatic hypertrophy and normal human prostate. The highest degree of expression occurred in the higher Gleason grades and prostatic intraepithelial neoplasia. This work suggests that pp32 is a nuclear protein which has a selective but presently undefined role in cells competent for self-renewal.