Dysplasia of human prostate CD133(hi) sub-population in NOD-SCIDS is blocked by c-myc anti-sense.

BACKGROUND: The CD133(hi) sub-population of prostate epithelial cells has been demonstrated to possess tumor-initiating capacity consistent with that of the cancer stem cell theory. However, the involvement of oncogenes such as c-myc has not been fully elucidated in the CD133(hi) sub-population. METHODS: We have isolated primary prostate cell strains (IBC-10a) and immortalized them by transfection with hTERT. The in vitro and in vivo tumorigenic capacity of isolated CD133(hi) and CD133(lo) cells was evaluated with respect to c-myc expression using specific sense and anti-sense oligonucleotides. RESULTS: Freshly immortalized cells consisted of <3.3% CD133(hi)/CD24(hi) sub-population (SP). "Prostaspheres" generated from single CD133(hi) cells in the presence of EGF consisted of approximately 10% CD133(hi) SPs in 12-21 day cultures. A single Prostasphere generated from single CD133(hi) cells (6-10 cell stage at day 6 injected i.t.) produced dysplastic lesions in NOD-SCID mice (n = 4/5). Treatment of Prostaspheres from CD133(hi) SPs in vitro with c-myc or cyclin D1 anti-sense oligonucleotides totally blocked colony forming ability and growth. Furthermore, treatment of fully formed, 6-day Prostaspheres for 48 hr with c-myc anti-sense significantly reduced c-myc expression and their ability to generate lesions in NOD-SCIDs (n = 10 Prostaspheres injected i.t./mouse). CONCLUSIONS: These data demonstrate for the first time that a single CD133(hi) cell is competent to generate Prostaspheres in vitro and that CD133(hi) Prostaspheres require c-myc to grow and form dysplastic lesions in vivo.

Role of the VEGFR3/VEGFD receptor axis in TGFbeta1 activation of primary prostate cell lines.

BACKGROUND: Reports indicate that vascular endothelial growth factor receptor type 3 (VEGFR3) regulates cellular functions such as invasion, proliferation, and chemo-resistance. However, the exact function of the VEGFR3 signaling axis in prostate epithelial cells is poorly characterized. METHODS: The goal of this study was to evaluate whether TGFbeta1 in combination with VEGFD can promote pre-malignant invasive activities of intermediate basal cells (IBC-10a) isolated from human prostate cancer (Gleason score 6). RESULTS: hTERT immortalized IBC-10a cells normally grew as confluent "cobblestoned" monolayers, but treatment with TGFbeta1 (10 ng/ml for 2-6 hr) dissociated the cell-cell junctions and induced VEGFR3 translocation to the cell surface. This event was not inhibited by 10 microM cycloheximide or puromycin, indicating transcription and protein synthesis were not required. We further discovered that TGFbeta1 in combination with VEGFD induced a significant increase in the invasive activity of IBC-10a cells (>26% and 53% after 24 and 48 hr, respectively) in modified Boyden Chamber assays. TGFbetaRII receptor antibodies specifically blocked TGFbeta1 induction of VEGFR3 translocation to the cell surface and blocked VEGFD-induced invasion. Zymograms revealed that TGFbeta1 (and not VEGFR3) stimulated the secretion of MMP-2 and MMP-9, presumably to promote cell invasion. The cell invasion assays confirmed that antibodies specific for TGFbetaII receptor, MMP-2 and MMP-9 and VEGFR3, independently blocked TGFbeta1-induced invasion. CONCLUSIONS: For the first time, we have demonstrated the mechanism by which TGFbeta1 stimulates VEGFD/VEGFR3 receptor axis activation leading to increased cell migration and invasion by primary intermediate basal cell cultures.

High voltage electron microscopy studies of axoplasmic transport in neurons: a possible regulatory role for divalent cations.

Light and high voltage electron microscopy (HVEM) procedures have been employed to examine the processes regulating saltatory motion in neurons. Light microscope studies demonstrate that organelle transport occurs by rapid bidirectional saltations along linear pathways in cultured neuroblastoma cells. HVEM stereo images of axons reveal that microtubules (Mts) and organelles are suspended in a continuous latticework of fine microtrabecular filaments and that the Mts and lattice constitute a basic cytoskeletal structure mediating the motion of particles along axons. We propose that particle transport depends on dynamic properties of nonstatic microtrabecular lattice components. EXperiments were initiated to determine the effects of changes in divalent cation concentrations (Ca2+ and Mg2+) on: (a)the continuation of transport and (b) the corresponding structural properties of the microtrabecular lattice. We discovered that transport continues or is stimulated to a limited extent in cells exposed to small amounts of exogenously supplied Ca2+ and Mg2+ ions (less than 0.1 mM). Exposure of neurons to increased dosages of Ca2+ and Mg2+ (0.2-1.0 mM) stimulates transport for 2-4 min at 37 degrees C, but after a 5- to 20-min exposure the saltatory movements of organelles are observed gradually to become shorter in duration and rate particle motion ceases to occur. HVEM observations demonstrated that Ca2+ - and with the cessation of motion. Ca2+-containing solutions produced contractions of the microtrabecular filaments, whereas Mg2+-containing solutions had the opposing effect of stimulating an elongation and assembly (expansion) of microtrabeculae. On the basis of these observations we hypothesize that cycles of Ca2+/Mg2+-coupled contractions and expansions of the microtrabecular lattice probably regulate organelle motion in nerve cells.

A functional in vitro model for studies of intracellular motility in digitonin-permeabilized erythrophores.

Phase contrast cine results demonstrate that erythrophores maintain saltatory particle motion for hours after permeabilization with 0.001% digitonin in a cytoskeletal stabilizing solution at 23 degrees C. High voltage electron microscopy (HVEM) studies reveal that cytoskeletal elements are retained intact, except in immediate subplasmalemmal regions where the plasma membrane is punctured by digitonin. During digitonin treatments, cells are permeable to ions, small molecules, and antibodies. We find that motion is Ca2+ and ATP-sensitive, and optimal in PIPES buffer (pH 7.2 containing 1 mM Mg2+/ATP and EGTA-CA2+ (10(-7) M Ca2+) at 37 degrees C. Experiments testing the inhibitory effects of vanadate (0.4-10 microM), ouabain (100-600 microM), N-ethyl maleimide, and the cytochalasins B and D indicate that a dyneinlike ATPase may provide the motive force for driving saltatory pigment motion in erythropores.

Estramustine binds a MAP-1-like protein to inhibit microtubule assembly in vitro and disrupt microtubule organization in DU 145 cells.

The twofold purpose of the study was (a) to determine if a MAP-1-like protein was expressed in human prostatic DU 145 cells and (b) to demonstrate whether a novel antimicrotubule drug, estramustine, binds the MAP-1-like protein to disrupt microtubules. SDS-PAGE and Western blots showed that a 330-kD protein was associated with microtubules isolated in an assembly buffer containing 10 microM taxol and 10 mM adenylylimidodiphosphate. After purification to homogeneity on an A5m agarose column, the 330-kD protein was found to promote 6 S tubulin assembly. Turbidimetric (A350), SDS-PAGE, and electron microscopic studies revealed that micromolar estramustine inhibited assembly promoted by the 330-kD protein. Similarly, estramustine inhibited binding of the 330-kD protein to 6-S microtubules independently stimulated to assemble with taxol. Immunofluorescent studies with beta-tubulin antibody (27B) and MAP-1 antibody (MI-AI) revealed that 60 microM estramustine (a) caused disassembly of MAP-1 microtubules in DU 145 cells and (b) removed MAP-1 from the surfaces of microtubules stabilized with 0.1 microM taxol. Taken together the data suggested that estramustine binds to a 330-kD MAP-1-like protein to disrupt microtubules in tumor cells.

Cytoplasmic microtubule organizing centers isolated from Polytomella agilis.

Basal body rootlets in Polytomella function as organizing centers for cytoplasmic microtbules in vivo. A method is described to isolate intact basal body-rootlet complexes. The integrity of the isolated complexes is confirmed by electron microscopy, and the rootlets are shown to be competent as initiation sites for the in vitro polymerization of brain microtubule protein.

Regulation of kinesin expression and type IV collagenase secretion in invasive human prostate PC-3 tumor sublines.

PC-3 human prostatic tumor sublines have been previously isolated which exhibit striking differences in their invasive and metastatic phenotypes. This work has been extended here to measure and compare the levels of kinesin, a microtubule-dependent translocator molecule, in the PC-3 sublines. Western blots, slot blots, radiolabeling, and immunoprecipitation analysis showed that kinesin was expressed in the highly invasive and metastatic sublines at levels which were elevated above the base-line levels observed in the parent PC-3 cells. In comparison, kinesin was not expressed in detectable amounts in the noninvasive cell lines. The conditioned medium of the metastatic PC-3 sublines contained a heat- and trypsin-sensitive protein which exhibited a dosage-dependent capacity to stimulate increased kinesin expression, type IV collagenase secretion, and invasion of Matrigel by the metastatic sublines. The noninvasive sublines failed to secrete a similar stimulatory factor(s) or respond to the conditioned medium of metastatic sublines. Various growth factors and cytokines tested (platelet-derived growth factor, epidermal growth factor, insulin-like growth factor, formylmethionineleucinephenylalanine) had no significant effect on either kinesin expression or protease secretion and invasion. Pertussis toxin blocked the stimulatory effects of the conditioned medium, but other agents known to interfere with adenylate cyclase pathways (i.e., cholera toxin, forskolin, 8-bromoadenosine) failed to block stimulation. The data show for the first time that kinesin, protease secretion, and the resulting invasion process may be regulated in a coordinated manner by an autocrine factor(s) which activates G-protein-dependent processes.

Blocking of collagenase secretion by estramustine during in vitro tumor cell invasion.

The effects of the antitumorigenic drug estramustine on tumor cell membrane penetration (invasion) were investigated in vitro by utilizing a synthetic basement membrane system (a modified Boyden chamber). Tumor cells were plated on a "partition barrier," consisting of a porous filter (8-micron pores) which was coated with a reconstituted basement membrane matrix (Matrigel), and induced to migrate across the barrier with conditioned medium obtained from 9DU 145 human prostatic tumor cells (passage 9). Quantitative radiolabeling studies demonstrated that specially isolated lines (isolated by several passages through the Matrigel) of DU 145 cells, A2058 melanoma, and B16-F10 melanoma cells were highly invasive such that 15 to 20% migrated across a 1-mm-thick Matrigel layer within 5 h at 37 degrees C. NIH-3T3 cells, mouse fibroblasts, and 20DU 145 cells (passage 20) exhibited little or no membrane invasive behavior. Micromolar concentrations of estramustine (30 to 120 microM) inhibited invasion by the invasive cell lines in a dosage-dependent fashion. Quantitative enzymatic assays and radioimmune assays demonstrated that estramustine inhibited membrane invasion by blocking type IV collagenase secretion. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blots confirmed that 30 to 60 microM estramustine blocked secretion of a Mr 105,000 collagenase protein. Indirect studies showed that a collagenase antibody raised against the Mr 105,000 protein and inhibitors of proteinase activity, including a metalloproteinase inhibitor, and 1,10-phenanthroline, blocked invasion. Because the antibodies inhibited type IV collagenase digestion of 3H-mouse type IV collagen, and invasion simultaneously, it is proposed that collagenolytic activity is involved in invasion. These data demonstrate that estramustine blocks proteinase secretion, and suggest that estramustine may be a useful therapeutic drug for the prevention of metastasis.

Antimicrotubule effects of estramustine, an antiprostatic tumor drug.

Estramustine [17 beta-estradiol 3 N bis(2-chloroethyl)carbamate; EM] is a stable conjugate of estradiol and nor-nitrogen mustard that is used for the treatment of human prostatic carcinoma. We have studied the cytotoxic effects of EM on the cytoskeletal organization of squirrelfish pigment cells (erythrophores) and human prostatic tumor cells (DU 145) in culture. Light and whole-mount electron microscopy studies reveal that, at microM levels (60 to 120 microM), EM has a dose-dependent disruptive effect on cell shape, cytoskeletal organization, and intracellular transport. Upon removal of the drug, the cytological effects of EM are rapidly reversible in fish cells but not DU 145s. Immunofluorescent studies reveal that EM produces microtubule disassembly in fish erythrophores and DU 145 cells. A concomitant disruption of actin-microfilament arrays also occurs in DU 145 cells. These morphological data suggest that EM, in contradistinction to its constituent estradiol: nitrogen mustard species, induces cytotoxicity as an antimicrotubule drug. The observed disruption of the microtubules and cytomatrix of interphase cells is not reversible in the prostatic carcinoma cells. The disruptive action of EM on the cytoskeleton could ultimately produce a cytotoxic antimitotic effect in dividing cells.

Type IV collagenase (M(r) 72,000) expression in human prostate: benign and malignant tissue.

The expression of type IV collagenase (M(r) 72,000) has been examined in tissues from patients with benign prostatic hyperplasia (6 patients) and varying Gleason grades of malignant prostate cancer (18 patients). Immunoperoxidase labeling indicated that expression of the type IV collagenase was weak or nonexistent in benign tissue but consistently strong in the glandular and ductal epithelial cells of prostate tumors diagnosed at Gleason grades 1-8. In moderate to advanced cancer (i.e., Gleason grades 2 to 8), invasive tumor foci in the stromal tissue produced relatively modest amounts of type IV collagenase. The normal stromal tissue (i.e., fibroblasts) uniformly failed to produce detectable levels of type IV collagenase in the 24 patients examined. Northern and quantitative slot blot hybridization assays demonstrated that collagenase type IV mRNA levels were low in benign tissue and high in malignant tumors. In contrast, the stromal cells did not express significant amounts of type IV collagenase mRNA. Enzyme-linked immunosorbent assays demonstrated that the amounts of type IV collagenase protein correlated directly with the mRNA levels in the tumor tissue. The studies suggest that type IV collagenase may be selectively overexpressed by malignant, preinvasive prostatic epithelial cells.

Taxol blocks processes essential for prostate tumor cell (PC-3 ML) invasion and metastases.

We have examined the antimetastatic effects of taxol on a PC-3 human prostatic tumor variant (PC-3 ML) which metastasizes to the lumbar vertebrae in severe combined immunodeficiency-carrying (SCID) mice. Immunofluorescence labeling indicated that taxol (0.5 to 1.0 microM for 6 h) produced an abnormal bundling of microtubules in a dosage-dependent manner. Slot blotting and gelatinase assays revealed that taxol inhibited secretion of the M(r) 72,000 and M(r) 92,000 type IV collagenases plus a M(r) 57,000 gelatinase. Radioimmunoprecipitation measurements confirmed that the drug inhibited both the secretion and the synthesis of the M(r) 72,000 collagenase. Taxol also blocked total protein secretion but did not influence total protein synthesis or turnover. Boyden chamber chemotactic studies further showed that taxol (0.5 to 1.0 microM) inhibited invasion of Matrigel. More importantly, studies in SCID mice demonstrated that taxol (50 to 250 mg/m2/day) blocked the establishment, growth, and long-term survival of PC-3 ML cells.

Liarozole and 13-cis-retinoic acid anti-prostatic tumor activity.

Liarozole fumarate (R85,246), a novel benzimidazole derivative, reduced s.c. and bone metastasis tumor growth by the androgen-independent PC-3ML-B2 human prostatic carcinoma clone in SCID mice. The drug inhibited cell invasion of Matrigel in Boyden chamber chemotactic assays and the secretion of type IV collagenase. In vitro, liarozole failed to inhibit cell proliferation and cell attachment to various substrates (Matrigel, laminin, type IV collagen, and fibronectin). In vivo, the drug also blocked type IV collagenase production in established s.c. tumors. Liarozole has been postulated by others (R. De Coster, W. Wouters, R. Van Ginckel, D. End, et al. J. Steroid Biochem. Mol. Biol., 43: 197-201, 1992) to inhibit retinoic acid catabolism. Our data indicate that liarozole treatment can increase the tumor retinoic acid levels in vivo. Studies of retinoic acid revealed that the drug independently reduced tumor growth in vivo and inhibited cell invasion of Matrigel and the secretion of collagenase IV. Surprisingly, liarozole and retinoic acid failed to exhibit measurable synergistic activity both in vitro and in vivo. Taken together these data suggest that liarozole might inhibit retinoic acid catabolism in vivo and consequently have significant therapeutic value as an anti-prostatic tumor agent.

Identification of the receptor for a novel M(r) 78,000 "invasion stimulating factor" from metastatic human prostatic PC-3 ML clones.

A M(r) 78,000 protein (reduced), termed invasion stimulating factor (ISF), was purified from the conditioned medium of a bone metastasizing human prostatic PC-3 ML clone (M. E. Stearns and M. Stearns. Cancer Metastasis Rev., 12: 39-52, 1993). Scatchard analysis and affinity cross-linking studies revealed that the producer PC-3 ML cells expressed a receptor binding site (M(r) approximately 115,000). We found a Kd approximately 425 pM and about 22,000 sites/cell. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis autoradiograms confirmed that the PC-3 ML cells expressed the receptor, whereas the ISF non-producing, noninvasive PC-3 clones (i.e., 3-4 x N.I. PC-3 cells) failed to express the ISF receptor. We conclude that a unique ISF autocrine loop characterizes the bone metastatic PC-3 ML cells.

Antimestatic and antitumor activities of interleukin 10 in transfected human prostate PC-3 ML clones: Orthotopic growth in severe combined immunodeficient mice.

We have permanently transfected human prostate PC-3 ML tumor cells and examined the influence of interleukin 10 (IL-10) production on tumor growth and metastasis following orthotopic implantation in the prostate gland of severe combined immunodeficient mice. Measurements of tumor volume after 5, 8, and 12 weeks indicated that tumor volume was negatively correlated with the amount of IL-10 production. Likewise, the extent of metastasis was inversely related to the amount of IL-10 produced. Following i.v. injection, the IL-10-expressing clones also failed to metastasize to the bone marrow. Controls showed that PC-3 ML and PC-3 ML mock clones grew rapidly and metastasized when implanted orthotopically or injected i.v. via the tail vein. Mouse survival curves showed that all of the mice injected orthotopically with the PC-3 ML clones died by about 14-16 weeks, whereas the PC-3 ML-IL10a or PC-3 ML-ILl0b clones induced only 10-20% death after 23-24 weeks. Likewise, survival studies showed a high death rate by approximately 30 days with PC-3 ML mock cells but <10% death by 12 weeks with the IL-10-transfected clones injected i.v. via the tail vein. The data strongly suggest that IL-10 production blocks tumor growth and metastasis in severe combined immunodeficient mice.

Interleukin 10 (IL-10) inhibition of primary human prostate cell-induced angiogenesis: IL-10 stimulation of tissue inhibitor of metalloproteinase-1 and inhibition of matrix metalloproteinase (MMP)-2/MMP-9 secretion.

In in vitro angiogenesis assays, aggregates of human papilloma virus (HPV)-18-immortalized primary human prostate cancer cells (HPCA-5aHPV-18 or HPCA-10aHPV-18 cells) induced human bone marrow endothelial cells (HBMCE-1 cells) to form microvessels in three-dimensional collagen I gels after 1-2 days incubation at 37 degrees C. The microvessels aligned perpendicular to the tumor aggregates and abutted on the edges of the aggregates. The number and length of the microvessels increased significantly from day 1 to 2 (i.e., by approximately 30%). ELISAs showed that the HPCA-5aHPV-18 cells normally secreted low levels of tissue inhibitor of metalloproteinase (TIMP)-2, matrix metalloproteinase (MMP)-2, and MMP-9 but relatively high levels of TIMP-1. In contrast, HPCA-10aHPV-18 cells secreted high levels of MMP-2 and MMP-9 (>40 pg/microg protein) but low levels of TIMP-1 and TIMP-2 (<5 pg/microg protein). Interleukin 10 (IL-10) (15 ng/ml) induced TIMP-1 production (>15 pg/microg protein) but reduced MMP-2 and MMP-9 secretion (<5 pg/microg protein) by the HPCA-5aHPV-18 and HPCA-10aHPV-18 cells. IL-10 (15 ng/ml) and MMP-9/MMP-2 antibodies all blocked induction of microvessel formation in the coculture experiments. In contrast, IL-10 receptor antibodies and TIMP-1 antibodies countered IL-10's effects and promoted angiogenesis. The data demonstrated that IL-10 stimulation of TIMP-1 and inhibition of MMP-2 and MMP-9 secretion by prostate tumor cells can control induction of angiogenesis in vitro.

Specific transcription factors prognostic for prostate cancer progression.

We have previously identified (M. Wang et al., Oncol. Res., in press, 1998) an enhancer element [human tissue inhibitor of metalloproteinase-1 enhancer-1 (HTE)] for the human tissue inhibitor of metalloproteinase-1 promoter that binds a novel zinc finger, cysteine-rich transcription factor (CRTF). In this study, we have used electrophoretic mobility shift assays to examine the relative level of expression of CRTF, jun/fos, and IFN-gamma responsive signal transducer activators of transcription (STATs) that bind specific HTE, activator protein, and IFN-gamma (Fcy and interferon regulatory factor) response motifs in tumor lines and human prostate tissue [i.e., normal (n = 3); benign prostatic hyperplasia (BPH; n = 12); high grade prostate intraepithelial neoplasia (PIN; n = 10); and prostate cancer adenocarcinoma (PCA; n = 61) plus seminal vesicle (n = 10) tissues]. The data showed that CRTF was overexpressed in PCA (Gleason's score, 10>8>6>5>4) compared with BPH, PIN, seminal vesicle, and normal tissues. To a much lesser degree, jun/fos and STAT 1 were also elevated in PCA compared to BPH, PIN, and normal tissues. In addition, blinded studies showed that CRTF and jun/fos were present at low levels in organ-confined specimens but at significantly elevated levels (P < 0.001) in samples exhibiting capsular penetration and localized spread, which indicated that CRTF and perhaps jun/fos were markers for cancer progression.

Role of interleukin 10 and transforming growth factor beta1 in the angiogenesis and metastasis of human prostate primary tumor lines from orthotopic implants in severe combined immunodeficiency mice.

Transfection of primary human prostate tumor cells (i.e., HPCA-10a, 10b, 10c, and 10d lines) with the transforming growth factor (TGF)-beta1 gene stimulated anchorage-independent growth and promoted tumor growth, angiogenesis, and metastasis after orthotopic implantation in severe combined immunodeficiency mice. In contrast, interleukin (IL)-10 transfected cells or cells cotransfected with these two genes exhibited reduced growth rates and significantly reduced angiogenesis and metastasis after 8, 12, and 16 weeks. Enzyme-linked immunosandwich assays confirmed that the respective tumors expressed elevated levels of TGF-beta1 and IL-10 in vivo. ELISAs further showed that TGF-beta1 expression induced matrix metalloproteinases-2 (MMP-2) expression, whereas IL-10 down-regulated MMP-2 expression while up regulating TIMP-1 in the transfected cells. Also, tumor factor VIII levels correlated with TGF-beta1 and MMP-2 expression and inversely with IL-10 and TIMP-1 levels. More importantly, mouse survival was zero after 4-6 months in mice bearing TGF-beta1- and MMP-2-expressing tumors and increased significantly in mice implanted with IL-10- and TIMP-1-expressing tumors (i.e., to >80% survival). Analysis of the metastatic lesions showed that they expressed TGF-beta1 and MMP-2 but barely detectable levels of IL-10 or TIMP-1, suggesting that IL-10 and TIMP-1 might normally block tumor growth, angiogenesis, and metastasis.

Alendronate blocks TGF-beta1 stimulated collagen 1 degradation by human prostate PC-3 ML cells.

We have previously shown that alendronate can prevent human PC-3 ML tumor cell metastasis to the bone (Wang and Stearns, 1991, Differentiation, 48, 115-25). In this paper, ELISAs and Western blots showed that TGF-beta1 stimulated the secretion of a 72 kDa matrix metalloproteinase 2 (MMP-2) to enhance the solubilization of radiolabeled collagen 1 by metastatic human prostate PC-3 ML cells. A potent bisphosphonate compound, alendronate, inhibited MMP-2 secretion to block solubilization of collagen 1. Alendronate failed to inhibit MMP-2 activity directly, but instead appeared to block cellular secretion of MMP-2. Alendronate failed to inhibit secretion of tissue inhibitor of metalloproteinase-2 (TIMP-2; the inhibitor of MMP-2) and the decrease in collagen 1 solubilization observed may occur, in part, from changes in the molar stoichiometry of TIMP-2 to MMP-2. We conclude that alendronate may be a potent inhibitor of bone resorption based on its ability to block MMP-2 secretion by tumor cells.

Alendronate blocks metalloproteinase secretion and bone collagen I release by PC-3 ML cells in SCID mice.

We have previously shown that alendronate, a potent bisphosphonate compound, can prevent human PC-3 ML tumor cell metastasis to the bone (Stearns and Stearns, 1996, Oncol Res, 8, 69-75). In this paper, tumor cells were injected into the bone medullary cavity of SCID mice femurs both in vivo and following isolation in vitro. ELISAs showed that the amount of collagen I released in the bone marrow (i.e. in in vitro experiments) and the blood plasma (i.e. in in vivo experiments) was a function of the time of incubation or the number of cells injected in the femurs. ELISAs also showed that the levels of matrix metalloproteinase (MMP-2 and MMP-9) secreted in the bone medullary cavity of the femurs directly correlated with the extent of collagen 1 release. In vitro experiments carried out with 'live' and 'devitalized bone' yielded similar results suggesting that the tumor cells (not the osteoclasts) were primarily responsible for the bone solubilization observed. Alendronate pretreatment of the SCID mice (0.1 mg/kg biweekly for 3 weeks) (or the tumor cells) blocked both MMP production by the tumor cells (and the osteoclasts) and collagen I release, providing direct evidence that alendronate might be utilized to prevent bone destruction by metastatic tumor cells. Zymography indicated that MMP-2 activation might be responsible for bone solubilization. In addition, the data suggest that the plasma levels of collagen I might be a marker of bone metastasis and osteolysis.

In situ hybridization studies of metalloproteinases 2 and 9 and TIMP-1 and TIMP-2 expression in human prostate cancer.

The expression of MMP-2, MMP-9, TIMP-1, TIMP-2, and the urokinase receptor were examined in fetal and normal prostate tissues, benign prostatic hyperplasia and prostate cancer (n = 117). In situ hybridization with digoxigenin-labeled oligonucleotide probes demonstrated that TIMP-1 and TIMP-2 were expressed at elevated levels in the stroma of Gleason sum 5 tissues, whereas MMP-2 and MMP-9 were expressed at relatively low levels. In higher Gleason sum tissues (GS 8-10), TIMP-1 and TIMP-2 were not expressed, whereas MMP-2 and MMP-9 were intensely expressed. Furthermore, TIMP-1 and TIMP-2 expression was high in organ-confined specimens (OC, n = 43), somewhat lower in specimens with capsular penetration (CP, n = 29), and low or negative in samples with surgical margin/seminal vesicle (M/SV, n = 17) and lymph node (LN, n = 13) involvement. In contrast, MMP-2 and MMP-9 expression was low in the OC tissues; and noticeably higher in CP, M/SV, and LN specimens. Finally, correlation of TIMP and MMP expression with GS and pathological stage versus cure rate further revealed that a high percentage of organ-confined, GS 5 specimens expressing TIMP and little MMP were cured. In comparison, few of the GS 7-10 patients with capsular penetration and expressing MMP and little TIMP were cured. The data suggest that TIMP-1 (and TIMP-2) and MMP-2 (and MMP-9) are independent predictors of outcome.