Inhaled isotretinoin (13-cis retinoic acid) is an effective lung cancer chemopreventive agent in A/J mice at low doses: a pilot study.

In previously treated head-and-neck cancer patients, p.o. administered isotretinoin (13-cis retinoic acid) reduced the occurrence of second aerodigestive tumors, including lung tumors, but side effects made chronic therapy problematic. We reasoned that inhaled isotretinoin might provide sufficient drug to the target cells for efficacy while avoiding systemic toxicity, and we proceeded with the pilot study reported here. Male A/J mice were given single i.p. doses of urethane, a common experimental lung carcinogen, or benzo[a]pyrene (BaP) or 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), putative major carcinogens in tobacco smoke. The following day, exposures to isotretinoin aerosols for 45 min daily at 1.3, 20.7, or 481 microg/l were initiated. After 2 weeks, the high dose caused severe toxicity on the snout skin, necessitating a reduction of dose frequency to twice a week. As a precaution, the mid dose was reduced to three exposures per week. The weekly total deposited doses after the dose frequency reductions were calculated to be 0.24, 1.6, and 24.9 mg/kg for the low, mid, and high doses, of which 16% was estimated to be deposited in the lungs. The weekly deposited pulmonary drug doses were calculated to be 0.01, 0.07, and 1.1% of a previously reported ineffective oral dose in urethane-treated A/J mice. After 10-16 weeks, mice were sacrificed to count areas of pulmonary hyperplasia and adenomas. For all carcinogens, the mice exposed to the high isotretinoin dose showed reductions of tumor multiplicity ranging from 56 to 80% (P < 0.005). The mid dose was associated with reductions of tumor multiplicity by 67 and 88% (P < 0.005) in BaP- and NNK-treated mice, respectively, and was tolerated until approximately 12 weeks, when both these and the high-dose mice began losing weight. The low-dose mice had nonsignificant reductions of 30% (P < 0.13) and 16% (P < 0.30) for BaP- and NNK-treated mice, respectively without any evidence of side effects. For BaP- and NNK-treated mice, numbers of hyperplastic areas directly correlated to dose level and inversely to tumor number, suggesting arrested progression. Inhaled mid-dose isotretinoin caused up-regulation of lung tissue nuclear retinoic acid receptors (RARs) relative to vehicle-exposed mice, RARalpha (3.9-fold vehicle), RARbeta (3.3-fold), and RARgamma (3.7-fold), suggesting that these receptors may be useful biomarkers of retinoid activity in this system. The encouraging results from this pilot study suggest that inhaled isotretinoin merits evaluation in people at high risk for lung cancer.

Inhibition of TPA and 12(S)-HETE-stimulated tumor cell adhesion by prostacyclin and its stable analogs: rationale for their antimetastatic effects.

We have investigated the regulatory role of PGI2 and its stable analogs, i.e., iloprost and cicaprost, on 12(S)-HETE- and TPA-enhanced tumor cell integrin expression and adhesion. Walker 256 carcinosarcoma cells express alpha IIb beta 3 integrin receptors, which mediate their adhesion to endothelium, subendothelial matrix and fibronectin. Adhesion is enhanced by treatment with exogenous 12(S)-HETE but not 12(R)-HETE or other lipoxygenase-derived hydroxy fatty acids, as well as by TPA. Both 12(S)-HETE and TPA enhanced alpha IIb beta 3 expression on W256 cells. PGI2 iloprost and cicaprost inhibited both 12(S)-HETE- and TPA-enhanced adhesion to endothelium and subendothelial matrix as well as alpha IIb beta 3 expression on W256 cells. The mechanism responsible for the effect of PGI2 was explored. Prostacyclin treatment of W256 cells resulted in an enhanced production of cAMP in a time- and dose-dependent manner. Pre-treatment of tumor cells with increasing concentrations of adenosine resulted in a dose-dependent decrease in the PGI2 effect on TPA or 12(S)-HETE-enhanced adhesion, suggesting that the PGI2 effect is mediated through PKA. Dibutyryl cAMP also blocked the 12(S)-HETE- or TPA-enhanced adhesion, and adenosine pre-treatment did not result in an inhibition of the dibutyryl cAMP effect. Collectively, our results suggest that the cyclooxygenase metabolite PGI2 can antagonize the lipoxygenase metabolite 12(S)-HETE- and TPA-enhanced alpha IIb beta 3 expression and tumor cell adhesion via activation of adenylate cyclase and elevation of intracellular levels of cAMP.

Enhanced endothelial cell retraction mediated by 12(S)-HETE: a proposed mechanism for the role of platelets in tumor cell metastasis.

Platelets have been hypothesized to contribute to tumor cell metastasis, but the underlying mechanism(s) remain unknown. We demonstrate here that one mechanism whereby platelets may facilitate metastasis is by potentiating tumor cell-induced endothelial cell retraction, a prerequisite for the extravasation of most tumor cell types. The integrity of cultured microvascular endothelial cell (CD3 cells) monolayers was perturbed by 12[S]-hydroxyeicosatetraenoic acid (12(S)-HETE), a lipoxygenase metabolite of arachidonic acid, as well as by tumor cells (i.e., Lewis lung carcinoma cells or 3LL). 3LL cells induced a concentration- and time-dependent retraction of the CD3 monolayers, as assessed by quantitative binding assays as well as by phase-contrast microscopy. In contrast, normal murine fibroblasts (3T3) did not induce endothelial cell retraction. 3LL cell-induced endothelial cell retraction was potentiated, in a dose- and time-dependent manner, by homologous murine platelets while platelets alone did not induce endothelial cell retraction. Platelet-enhanced, tumor cell-induced endothelial cell retraction was inhibited by treating either tumor cells or platelets with the lipoxygenase inhibitors nordihydroguaiaretic acid or N-benzyl-N-hydroxy-5-phenylpentanamide (BHPP) as well as by PGI2 or its analogs iloprost and ZK96.480 (cicaprost), but not by the cyclooxygenase inhibitor aspirin (ASA). Tumor cells, upon adhesion to endothelium, initiated 12(S)-HETE biosynthesis, which was inhibited by pretreating tumor cells with BHPP but not with ASA. Additionally, 12(S)-HETE biosynthesis during tumor cell-endothelial cell adhesion was significantly enhanced by the addition of homologous platelets. Collectively, these results suggest that tumor cell-platelet-endothelial cell interactions lead to enhanced biosynthesis of 12(S)-HETE by tumor cells and/or platelets, which in turn induces endothelial cell retraction, thus facilitating tumor cell extravasation and metastasis.

The lipoxygenase metabolite, 12(S)-HETE, induces a protein kinase C-dependent cytoskeletal rearrangement and retraction of microvascular endothelial cells.

We previously reported that a lipoxygenase metabolite of arachidonic acid, 12(S)-hydroxyeicosatetraenoic acid [12(S)-HETE], induced large vessel endothelial cell (EC) retraction and increased tumor cell adhesion to exposed extracellular matrix (Honn et al., FASEB J. 3, 2285-2293, 1989). Here, we present evidence that 12(S)-HETE induces the retraction of microvascular ECs in a time- and concentration-dependent manner. The EC retraction was observed 15 min after 12(S)-HETE treatment and reached a peak level between 1 and 2 h. The monolayer reformed by 24 h. Silver staining and "gap-FRAP" experiments suggest that 12(S)-HETE altered the normally apposed cell junctions and impaired gap junction-mediated cell-cell communication. It appears that the 12(S)-HETE effect was mediated by cytoskeletal alteration. The first observed alteration in EC cytoskeleton following 12(S)-HETE stimulation is vimentin bundling, followed by the rearrangement and disruption of vinculin-containing adhesion plaques and/or simultaneous redistribution of alpha-actinin and disruption of spectrin. These changes are accompanied by progressive microfilament dissolution. During the same time interval, alpha-actinin is mobilized to the cell periphery at cell "ruffles." However, 12(S)-HETE showed little or no effects on actin-binding proteins filamin and tropomyosin or on microtubules. 12(S)-HETE effects on these cytoskeletal elements were fully reversible by 24 h and appeared to be mediated through enhancing protein phosphorylation. Following 12(S)-HETE (0.1 microM) treatment increased phosphorylation of proteins that comigrated with myosin light chain (20 kDa), actin (42 kDa), and vimentin (57 kDa) were observed. The enhanced phosphorylation of these cytoskeletal proteins was confirmed by 2D gel analysis. The phosphorylation-promoting effect of 12(S)-HETE on cytoskeletal proteins could be totally abolished by calphostin C, partially inhibited by staurosporine, but was not influenced by N-[2-(methylamine)ethyl]-5-isoquinolinesilfonamide dihydrochloride (HS), suggesting that the 12(S)-HETE effect was mediated via protein kinase C. This was further substantiated by quantitative experiments demonstrating that calphostin C, but not H8, inhibited 12(S)-HETE-induced EC retraction.

Phenotypic properties of cultured tumor cells: integrin alpha IIb beta 3 expression, tumor-cell-induced platelet aggregation, and tumor-cell adhesion to endothelium as important parameters of experimental metastasis.

The present study was undertaken to investigate the factors involved in determining the metastatic potential of cultured cells derived from solid tumors. We first investigated the effects of cell source and culture conditions on lung colony formation by i.v. injected B16a (B16 amelanotic melanoma) cells and inhibition of tumor colony formation by the thromboxane A2 synthase inhibitor, CGS14854. Prolonged culture resulted in a 10-fold decrease in the incidence of B16a lung colonies, whereas passage in vivo for 150 days did not affect lung colony formation by tumor cells isolated from enzymatic dispersates by centrifugal elutriation. Cultured B16a cells maintained at low density (LD) and harvested at low passage (LP) formed significantly more lung colonies than B16a cells harvested at high densities (HD) or high passage (HP). Over-confluent tumor cells produced even lower number of lung colonies. Lung colony formation by elutriated B16a cells (i.e., cells freshly isolated from tumor tissue) was consistently inhibited by CGS14854, whereas inhibition of lung colony formation by cultured B16a cells was dependent upon culture conditions. CGS14854 was ineffective or less effective against HD/HP B16a cells. The differences in lung colony formation between LD, HD and elutriated B16a cells were not due to differential cell-cycle distribution. Mechanistic studies indicated that LD/LP tumor cells induced aggregation of homologous platelets, whereas HD/HP B16a cells failed to induce significant platelet aggregation. Aggregation of homologous platelets correlated positively with lung-colonizing ability. Additionally, LD/LP cells demonstrated higher adhesion to endothelium than HD/HP B16a cells. Finally, LD/LP B16a cells expressed higher levels of alpha IIb beta 3 integrins than HD/HP tumor cells, as determined by flow cytometry and immunofluorescence.

12(S)-HETE promotes tumor-cell adhesion by increasing surface expression of alpha V beta 3 integrins on endothelial cells.

The present work was undertaken to investigate the regulatory role of 12(S)-HETE, a lipoxygenase metabolite of arachidonic acid, in the surface expression of alpha v beta 3 integrin receptors in endothelial cells (rat aortic endothelial cells, or RAEC). Several monoclonal and polyclonal antibodies localized alpha v beta 3 in focal adhesions in both subconfluent and post-confluent RAEC. RAEC alpha v beta 3 integrins were further characterized by immunoblotting and immunoprecipitation. 12(S)-HETE, but not 12(R)-HETE or other lipoxygenase-derived hydroxy fatty acids, induced a dose-dependent increase in alpha v beta 3 surface expression in RAEC, which was antagonized by prostacyclin or its analog iloprost as well as by 13-HODE, a 15-lipoxygenase product of linoleic acid. 12(S)-HETE promoted RAEC adhesion to vitronectin, an effect inhibited by antibodies against alpha v beta 3. 12(S)-HETE also promoted tumor-cell (W256 carcinosarcoma) adhesion to vitronectin, which was inhibited by various antibodies against alpha IIb beta 3 but not by an antibody against alpha v. W256 adhesion to 12(S)-HETE-treated RAEC demonstrated a significant increase, which was inhibited by anti-alpha v, -beta 3, or -alpha v beta 3 antibodies and by 13-HODE. Western blotting, immunoprecipitation and reverse transcription-polymerase chain reaction indicated that W256 carcinosarcoma cells expressed alpha IIb beta 3 integrins but not alpha v beta 3. The results suggest that the lipoxygenase metabolites [i.e., 12(S)-HETE and 13-HODE] play a significant role in modulating tumor-cell interactions with endothelium by enhancing endothelial cell integrin (e.g., alpha v beta 3) expression.

Studies on the role of platelet eicosanoid metabolism and integrin alpha IIb beta 3 in tumor-cell-induced platelet aggregation.

Platelet eicosanoid metabolism resulting from tumor-cell-induced platelet aggregation (TCIPA) was examined in a homologous in vitro system. Rat Walker 256 carcinosarcoma cells induced the aggregation of rat platelets via a thrombin-dependent mechanism with concomitant production of eicosanoid metabolites (e.g., 12-HETE, TXA2). TCIPA was dependent on the concentration of tumor cells inducing aggregation, as well as cyclooxygenase and lipoxygenase products. Cyclooxygenase inhibitors, but not lipoxygenase inhibitors, blocked platelet aggregation induced in vitro by a low concentration of agonist. At a high agonist concentration, neither cyclooxygenase nor lipoxygenase inhibitors alone affected platelet aggregation; however, the combined inhibition of both the cyclooxygenase and lipoxygenase pathways resulted in subsequent inhibition of platelet aggregation regardless of agonist concentration. The extent of platelet TXA2 and 12-HETE biosynthesis was likewise dependent on and correlated with agonist concentration. The inhibitors used in this study did not significantly inhibit protein kinase C activity at the doses tested. Platelet surface glycoprotein alpha IIb beta 3 play an important role in platelet aggregation. The effect of platelet cyclooxygenase and lipoxygenase inhibition in regulating alpha IIb beta 3 surface expression was examined by flow cytometric analysis. Thrombin stimulation of washed rat platelets resulted in significantly increased surface expression of platelet alpha IIb beta 3 integrin complex. The enhanced surface expression was not inhibited by a cyclooxygenase inhibitor (aspirin), a thromboxane synthase inhibitor (CGS-14854) or a thromboxane receptor antagonist (SQ 29,548), nor was it stimulated by a thromboxane A2 mimic (pinane-thromboxane A2). However, alpha IIb beta 3 expression was blocked by lipoxygenase inhibition and stereospecifically increased by the platelet lipoxygenase metabolite 12(S)-HETE. These results suggest that both the platelet lipoxygenase and cyclooxygenase pathways are important for TCIPA but that different mechanisms of action are involved.

Identification of the alpha IIb beta 3 integrin in murine tumor cells.

Previously, we have identified an alpha IIb beta 3-like integrin in tumor cells by using antibodies against platelet alpha IIb beta 3. However, alpha IIb beta 3 was considered to be expressed strictly in megakaryocyte lineage cells. In order to resolve this controversy, the alpha IIb beta 3-like integrin in murine B16 amelanotic melanoma (B16a) cells was characterized at DNA, RNA, and protein levels. The presence of alpha 5, alpha v, alpha IIb, beta 1, and beta 3 genes in B16a cells was confirmed by Southern analysis. mRNAs for all these integrins except alpha v were detectable by Northern blotting. The alpha IIb beta 3 protein was identified by Western blotting using subunit-specific antibodies and by immunoprecipitation using complex-specific antibody. The alpha IIb beta 3 integrin was localized intracellularly by immunocytochemistry. Finally, alpha IIb and beta 3 mRNAs were amplified by reverse transcription-polymerase chain reaction and the identity of alpha IIb was verified by sequencing. Partial DNA and deduced amino acid sequence analysis showed that B16a alpha IIb shares approximately 80% homology with the human alpha IIb and approximately 90% homology with the rat alpha IIb, whereas B16a alpha IIb shares only approximately 26% homology with the human alpha v. These experiments indicate that the alpha IIb beta 3-like protein in B16a cells is the authentic alpha IIb beta 3 and demonstrate, for the first time, that integrin alpha IIb beta 3 is not confined to megakaryocyte lineage cells.

Alpha IIb beta 3 integrin expression and function in subpopulations of murine tumors.

Subpopulations of B16 amelanotic melanoma (B16a) cells, isolated by centrifugal elutriation from enzymatically dispersed solid tumors, demonstrated different abilities to form lung colonies when injected intravenously. In contrast, no differences in experimental metastasis were observed among subpopulations obtained from Lewis lung (3LL) tumors. Lung colonization by B16a and 3LL subpopulations correlated positively with observed differences (B16a) or lack of differences (3LL) in tumor cell ability to induce aggregation of homologous platelets, to adhere to subendothelial matrix or fibronectin, and with the percentage of cells in the G2/M phase of the cell cycle. Both B16a and 3LL cells express alpha IIb beta 3 integrin receptors; however, differences in the receptor expression level were found only among B16a subpopulations. Comparison of the amount of alpha IIb beta 3 receptor expressed on cell surface with tumor cell ability to induce platelet aggregation (TCIPA) and to adhere to fibronectin or subendothelial matrix revealed a positive correlation. Pretreatment of tumor cells with alpha IIb beta 3-specific antibodies inhibited tumor cell matrix adhesion, TCIPA, and lung colony formation. We propose that alpha IIb beta 3 integrin receptor expression, tumor cell matrix adhesion, and tumor cell-induced platelet aggregation can be important parameters to indicate the metastatic potential of some tumor cells and that the alpha IIb beta 3 is a multifunctional receptor involved in both tumor cell-matrix and tumor cell-platelet interactions. Further, the correlation among cell cycle phase, metastatic ability, and receptor expression suggests that metastatic propensity may be transiently expressed and/or increased in some tumor cell subpopulations.

Increased expression of alpha IIb beta 3 integrin in subpopulations of murine melanoma cells with high lung-colonizing ability.

Four subpopulations of B16 amelanotic melanoma cells, possessing different abilities to induce platelet aggregation (TCIPA) and to form lung colonies, were isolated by centrifugal elutriation. The expression of alpha IIb beta 3, alpha v beta 3 and alpha 5 beta 1 integrins was examined in the 4 subpopulations in order to determine the relationship between integrin receptor expression and tumor-cell metastatic potential. The mRNA of alpha IIb, alpha 5, beta 1 and beta 3 was detectable in the 4 subpopulations by Northern blotting. A gradual increase in mRNAs and cell-surface immunoreactivity of the alpha IIb beta 3 receptor, but not in their gene copies, was observed from the low to the high metastatic subpopulations. The ability of tumor cells to adhere to fibronectin and subendothelial matrix (SEM) increased in parallel. In the high metastatic cells, the alpha IIb beta 3 receptors, but not the alpha 5 beta 1 receptors, were localized to focal adhesion plaques. Incubation of the high metastatic cells with alpha IIb beta 3-specific antibodies reduced their matrix adhesion, TCIPA and lung-colonizing abilities. In contrast, in the low met- astatic cells, SEM adhesion and lung-colony formation were not affected by anti-alpha IIb beta 3 antibody treatment. Incubation of either the low or the high metastatic subpopulation with an alpha 5 beta 1-specific antibody had no effect in vitro and showed a slight inhibition of lung colonization in vivo. Our results suggest that several phenotypic characteristics of the enhanced metastatic potential of B16a subpopulations may be mediated by increased expression of alpha IIb beta 3 receptors and that expression of these receptors may be regulated at the transcriptional level.

Is there a role for the tumor cell integrin alpha IIb beta 3 and cytoskeleton in tumor cell-platelet interaction?

In vitro tumor cell-platelet interaction was examined using B16 amelanotic (B16a) melanoma cells. These tumor cells express the alpha IIb beta 3-type cytoadhesin. Aggregation studies demonstrated that tumor cell surface alpha IIb beta 3 mediates the recognition of platelets since pretreatment of tumor cells with antibody against alpha IIb beta 3 prevents platelet-tumor cell interaction as well as platelet activation measured by aggregometry, platelet eicosanoid metabolism and ultrastructural analysis. In B16a cells, disruption of the microfilaments and intermediate filaments inhibits mobility of alpha IIb beta 3 on the cell surface. Microtubules do not play a role in receptor mobility, because B16a cells do not possess well-defined microtubules in interphase and colchicine does not affect receptor mobility. Disruption of microfilaments or intermediate filaments results in an inhibition of tumor cell-platelet interaction as evidenced by aggregometry studies and ultrastructural analysis. We suggest that platelet interaction with tumor cells begins with alpha IIb beta 3-mediated receptor recognition followed by not only platelet activation but also microfilament- and vimentin intermediate filament-dependent tumor cell activation.

The lipoxygenase metabolite 12(S)-HETE induces a cytoskeleton-dependent increase in surface expression of integrin alpha IIb beta 3 on melanoma cells.

Integrin receptors are mediators of cell-extracellular matrix and cell-cell interactions. Biochemical and immunocytochemical evidence shows that the platelet integrin receptor alpha IIb beta 3 is present on the cell surface, at focal adhesion plaques and in the perinuclear region of metastatic B16a murine melanoma cells. Antibody to the fibronectin receptor alpha 5 beta i, inhibits basal adhesion by approx. 30%, whereas antibodies to alpha IIb beta 3 are ineffective. The surface immunoreactivity of tumor cells for alpha IIb beta 3 can be enhanced by pre-treatment (5 min) with a lipoxygenase metabolite of arachidonic acid [i.e. 12-(S)-HETE] in a dose-dependent manner (max. effect approx. 0.1 microM). Other lipoxygenase metabolites are ineffective. B16a cells possess a large intracellular pool of alpha IIb beta 3, from which the receptor complex translocates to the cell surface following 12-(S)-HETE pretreatment. This pre-treatment of tumor cells enhances their adhesion to fibronectin, which is mediated exclusively by alpha IIb beta 3 receptors. 12-(S)-HETE also facilitates the redistribution of alpha IIb beta 3 in the plasma membrane with localization at the focal adhesion plaques. The cytoskeleton of the B16a cell is characterized by an absence of distinct microtubules in interphase cells and the presence of prominent microfilaments and vimentin intermediate filaments. In B16a cells, the disruption of intermediate filaments and/or microfilaments prevents the 12-(S)-HETE-induced increase in plasma membrane alpha IIb beta 3 and enhanced tumor-cell adhesion to fibronectin. The microtubule-disrupting agent, colchicine, is ineffective in both respects. We conclude that the lipoxygenase metabolite of arachidonic acid, 12-(S)-HETE, regulates the surface expression and function of the alpha IIb beta 3 integrin in B16a cells. Further, these data support the hypothesis that microfilaments and intermediate filaments have a profound role in regulating the expression of a multifunctional integrin in B16a tumor cells.

Enhanced tumor cell adhesion to the subendothelial matrix resulting from 12(S)-HETE-induced endothelial cell retraction.

A 12-lipoxygenase metabolite of arachidonic acid, 12(S)-hydroxyeicosatetraenoic acid (12[S]-HETE), which is produced by platelets and tumor cells, was tested for its ability to induce retraction of endothelial cell monolayers. The induction of endothelial cell retraction is a critical step in tumor cell metastasis. Endothelial cells demonstrated reversible retraction in response to 12(S)-HETE, but did not respond to the stereoisomer 12(R)-HETE or to unrelated 5-lipoxygenase (i.e., 5[S]-HETE) or 15-lipoxygenase (i.e., 15[S]-HETE) metabolites. Endothelial cells did not demonstrate loss of viability in response to 12(S)-HETE. The induction of retraction was both dose and time dependent. Scanning electron microscopy confirmed that 12(S)-HETE induced endothelial cell retraction and revealed collapsed filopodia on their surface, the appearance of spaces between endothelial cells and the underlying subendothelial matrix, in addition to large gaps between adjacent endothelial cells. Tumor cell adhesion to endothelial cell monolayers was enhanced 1 h after pretreatment of monolayers with 12(S)-HETE but not after pretreatment with other lipoxygenase metabolites. Tumor cell adhesion to endothelial cell monolayers 36 h after pretreatment with 12(S)-HETE was not different from adhesion to untreated monolayers. Therefore we suggest that 12(S)-HETE generated during tumor cell-platelet-endothelial cell interactions may induce reversible endothelial cell retraction, allowing tumor cell access to the subendothelial matrix, which is a critical step in their eventual extravasation from the microvasculature during hematogenous metastasis.

Bidirectional control of membrane expression and/or activation of the tumor cell IRGpIIb/IIIa receptor and tumor cell adhesion by lipoxygenase products of arachidonic acid and linoleic acid.

Lewis lung carcinoma cells express a plasma membrane receptor (i.e., IRGpIIb/IIIa) which is immunologically and functionally related to the platelet aggregation receptor complex (i.e., GpIIb/IIIa). Both fluorescence microscopy and flow cytometric analysis reveal that surface expression and/or activation of this tumor cell receptor is enhanced by a phorbol ester [i.e., 12-O-tetradecanoylphorbol-13-acetate (TPA)] and a lipoxygenase metabolite of arachidonic acid; 12-hydroxyeicosatetraenoic acid (i.e., 12-HETE). TPA-enhanced expression appears to be mediated by a lipoxygenase metabolite, as this effect can be reversed by lipoxygenase inhibitors but not by cyclooxygenase inhibitors. In parallel with these results both TPA and 12(S)-HETE [but not 12(R)-HETE] enhance tumor cell adhesion to endothelial cells, subendothelial matrix and fibronectin, but not to type IV collagen. TPA-enhanced adhesion can be reduced by lipoxygenase inhibitors but not by cyclooxygenase inhibitors and in addition, stimulated adhesion can be blocked by pretreatment of tumor cells with specific polyclonal or monoclonal antibodies which react against IRGpIIb/IIIa. 12(S)-HETE-enhanced adhesion can also be inhibited by these same antibodies. In contrast, a lipoxygenase product of linoleic acid, 13(S)-hydroxyoctadecadienoic acid, inhibited TPA and 12(S)-HETE-enhanced tumor cell adhesion to endothelial cells, subendothelial matrix, and fibronectin. These results suggest that (a) IRGpIIb/IIIa is a multifunctional receptor which mediates tumor cell adhesion to a variety of biological substrata, (b) TPA enhances surface expression and/or activation of this receptor possibly via a lipoxygenase metabolite of arachidonic acid, and (c) these effects are opposed by a lipoxygenase metabolite of linoleic acid.

Lipoxygenase products regulate IRGpIIb/IIIa receptor mediated adhesion of tumor cells to endothelial cells, subendothelial matrix and fibronectin.

Tumor cell adhesion to endothelial cells, subendothelial matrix, and fibronectin is stimulated by the lipoxygenase metabolite of arachidonic acid, 12(S)-HETE, but not by 12(R)-HETE, 5-HETE or 15-HETE. Adhesion is also stimulated by the phorbol ester TPA, an effect inhibited by lipoxygenase but not cyclooxygenase inhibitors. TPA and 12(S)-HETE mediated adhesion is due, in part, to an integrin receptor (i.e., IRGpIIb/IIIa) related to the platelet glycoprotein IIb/IIIa complex and is inhibited by specific monoclonal and polyclonal antibodies against platelet IIb/IIIa. TPA and 12(S)-HETE stimulated adhesion is also inhibited by a lipoxygenase product of linoleic acid; i.e., 13-HODE. These results suggest bidirectional control of tumor cell adhesion by lipoxygenase products of arachidonic acid (increase) and linoleic acid (decrease).

Role of tumor cytoskeleton and membrane glycoprotein IRGpIIb/IIIa in platelet adhesion to tumor cell membrane and tumor cell-induced platelet aggregation.

Components of the tumor cell cytoskeleton (i.e., microtubules, microfilaments, and intermediate filaments) have been reported to affect metastatic ability, since disruption of these components leads to a decrease in metastasis. One mechanism of metastasis which has not been previously considered is the decreased interaction of tumor cells with platelets. We present evidence that disruption of the tumor cell cytoskeleton decreases the ability of tumor cells to aggregate homologous platelets. This effect is dependent upon the disruption of microfilaments/intermediate filaments but not disruption of microtubules. In addition, tumor cell platelet interactions require the lateral mobility of specific receptors (i.e., clustering) on the tumor cell plasma membrane. A membrane glycoprotein immunologically related to the platelet glycoprotein IIb/IIIa complex was identified on Walker 256 carcinosarcoma cells using specific polyclonal and monoclonal antibodies and Northern blot analysis using complementary DNA probes for IIb and IIIa. Mobility of this receptor is dependent upon tumor cell microfilaments/intermediate filaments, but not microtubules. Furthermore, treatment of tumor cells with specific antibodies to the platelet glycoprotein IIb/IIIa complex inhibits tumor cell-platelet interaction at the macroscopic level (i.e., aggregation) and at the ultrastructural level (i.e., platelet adhesion to the tumor cell surface). These results suggest that this immunologically related glycoprotein IIb/IIIa is a receptor for platelet binding to the tumor cell surface, an event which precedes overt platelet aggregation and is dependent upon an intact tumor cell microfilament and intermediate filament network. Therefore, the decreased metastasis observed by others following disruption of the tumor cell cytoskeleton may be due, in part, to a decreased tumor cell-platelet interaction.

Role of tumor cell glycoproteins immunologically related to glycoproteins Ib and IIb/IIIa in tumor cell-platelet and tumor cell-matrix interactions.

A panel of monoclonal and polyclonal antibodies raised against human platelet GpIb or the GpIIb/IIIa complex were used to detect immunologically related molecules on two cell lines derived from human solid tumors. Human cervical carcinoma (MS751) and human colon carcinoma (clone A) expressed molecules immunologically related to platelet GpIb and GpIIb/IIIa complex. These molecules were localized to their plasma membranes by immunofluorescence and immunocytochemistry. The immunologically related GpIb was evenly distributed on the tumor cell membrane with occasional areas of aggregates, whereas the immunologically related GpIIb/IIIa had a pronounced punctate distribution of aggregates in prefixed cells. When MS751 or clone A cells were pretreated with antibodies against platelet GpIb and/or the GpIIb/IIIa complex, their ability to induce platelet aggregation was significantly inhibited. In addition, when tumor cells were pretreated with antibodies against the platelet IIb/IIIa complex, adherence to fibronectin-coated plates was also significantly inhibited. These results suggest a role for these immunologically related tumor cell glycoproteins in tumor cell-host cell (i.e., platelet, endothelial cells) interactions, tumor cell interactions with components of the subendothelial matrix, and subsequent tumor metastasis.

Separation of high and low metastatic subpopulations from solid tumors by centrifugal elutriation.

We have isolated from murine solid tumors (B16a) subpopulations of cells possessing high and low metastatic potential. Tumors were dispersed by collagenase treatment. The resulting heterogeneous population of cells (i.e., viable and non-viable tumor cells and host cells) were separated by centrifugal elutriation. Four of the fractions (100, 180, 260, 340) contained tumor cells of high viability (greater than 95%) and high purity (less than 1% host cell contamination). The four fractions were characterized by flow cytometry and found to differ in distribution of cells in G1, S and G2. The cell populations were also found to differ in metastatic potential as determined by their ability to form lung colonies following intravenous injection. The 340 fraction was approximately 5-fold more metastatic than the 100 fraction. We also observed that cells from the 100 fraction failed to induce platelet aggregation whereas cells from the 340 fraction induced significant platelet aggregation. These observations demonstrate that cells of B16a tumors are heterogeneous for phenotypic characteristics (i.e., metastatic potential; platelet aggregation, etc.) and that their ability to induce platelet aggregation is positively correlated with metastatic potential.

Inhibition of human tumor cell induced platelet aggregation by antibodies to platelet glycoproteins Ib and IIb/IIIa.

Tumor cell induced platelet aggregation was shown to be inhibited in a dose dependent manner by preincubation of human platelets with antibodies to platelet glycoprotein Ib and the IIb/IIIa complex. Combination of antibody to Ib and antibody to the IIb/IIIa complex at concentrations which produced half maximal inhibition of platelet aggregation alone caused complete inhibition of tumor cell induced platelet aggregation. Antibodies to platelet glycoproteins Ib and the IIb/IIIa complex also inhibited platelet synthesis of thromboxane A2, but not synthesis of 12-hydroxyeicosatrienoic acid. Inhibition of tumor cell induced platelet aggregation with antibodies against platelet glycoproteins suggests a role for these glycoproteins in tumor cell-platelet interactions and possibly platelet facilitated tumor cell metastasis.