PTP LAR expression compared to prognostic indices in metastatic and non-metastatic breast cancer.

Several prognostic indices in breast cancer, including c-erbB2, epithelial growth factor receptors (EGFR), estrogen and progesterone receptors are signal transduction molecules. Recently, expression of another signal transduction molecule, the protein tyrosine phosphatase LAR, has been suggested to be increased in breast cancer. The objective of the current investigation was to examine the relationship between LAR expression and prognostic parameters in breast cancer. LAR expression was associated with metastatic potential in the well-characterized 13762NF rat mammary adenocarcinoma clones. The metastatic MTLn3 and MTLn2 clones expressed sizable amounts of LAR. The essentially non-metastatic MTC clone had little LAR expression. C-erbB2 had highest expression in the highly metastatic MTLn3 clone, but c-erbB2 levels were sizeable in the weakly metastatic MTLn2 and non-metastatic MTC clone. EGFR expression had the strongest association with a clone's metastatic potential, being very high in MTLn3, weak in MTLn2, and undetectable in MTC. In human breast cancer specimens, LAR expression was strongly positive in 50% of metastatic cases but in only 21% of 'non-metastatic' cases. As with the 13762NF-derived clones, c-erbB2 expression was strongly positive independent of metastatic phenotype. However, 46% (6/13) of cases that were strongly positive for c-erbB2 were strongly positive for LAR. Only 17% (2/11) of negative or weakly c-erbB2 positive samples were strongly positive for LAR. All ER+ positive tumors (n = 15) were positive for LAR and 53% of these tumors were strongly positive for LAR. In ER negative cases, only 1 of 11 was strongly positive for LAR. While the current data indicate a strong association between ER and LAR expression in breast cancer tissue (p = 0.003), additional studies are warranted to further explore the relationship between LAR and prognostic indices of breast cancer progression.

Elevated expression of the neutrophil calcium-binding protein, MRP-14, in metastasis-enhancing neutrophils.

Tumor-elicited neutrophils (tcPMN) purified from 13762NF mammary adenocarcinoma tumor-bearing rats enhanced metastasis of syngeneic cells when co-injected intravenously; whereas, circulating (cPMN) and phorbol esteractivated (PMA-PMN) neutrophils did not [Welch et al. (1989) Proc. Natl. Acad. Sci. 86:5859-63]. We hypothesized that differential protein expression was responsible for functional differences between the neutrophil subtypes. Two-dimensional polyacrylamide gel electrophoresis was used to compare neutrophils (cPMN, PMA-PMN) purified from the peripheral blood of healthy, syngeneic nontumor-bearing rats, to tcPMN collected from rats with highly metastatic [clone MTLn3, subclone MTLn3(T44).5] or poorly metastatic [subclone MTLn3(T44).11] tumors growing in the mammary fat pads. Quantitative differences in polypeptide expression were observed between these functionally distinct PMN populations. Compared to cPMN, expression of a M(r) approximately 38.8 kDa (pl approximately 8) polypeptide was similar in tcPMN collected from poorly metastatic tumor-bearing rats, higher in PMA-PMN, and further increased in tcPMN from rats with highly metastatic tumors. Expression of two polypeptides, M(r) approximately 14.1 kDa (pl approximately 6) and M(r) approximately 43.3 kDa (pl approximately 5), was greater in tcPMN from rats with highly metastatic tumors compared to cPMN, PMA-PMN, or tcPMN from rats bearing poorly metastatic tumors. The latter two polypeptides thus appeared to be specifically increased in tcPMN from rats bearing highly metastatic tumors. Because it was most abundant and displayed the greatest differences between PMN subtypes, the M(r) approximately 14.1 kDa protein was further analyzed. Tryptic digests followed by internal sequence analyses of resulting peptide fragments revealed that the M(r) approximately 14.1 kDa contained amino acid sequences that were identical to those of MRP-14, a 14 kDa neutrophil calcium-binding protein belonging to the S-100 protein family of calcium-binding proteins. These results suggest a novel function for MRP-14 and suggest that MRP-14 may represent a marker for distinguishing phenotypically distinct subpopulations of neutrophils, particularly tcPMN with metastasis-enhancing abilities.

Metastasis suppressed, but tumorigenicity and local invasiveness unaffected, in the human melanoma cell line MelJuSo after introduction of human chromosomes 1 or 6.

Progression of human melanoma toward increasing malignant behavior is associated with several nonrandom chromosomal aberrations, most commonly involving chromosomes 1, 6, 7, 9, and 10. We previously showed that introduction of human chromosome 6 into the highly metastatic human malignant melanoma cell line C8161 completely suppressed metastasis without altering tumorigenicity (Welch DR, Chen P, Miele ME, et al., Oncogene 9:255-262, 1994). Alterations of chromosome 1 are the most frequent chromosome abnormality observed in melanomas, and they frequently arise late in tumor progression. The purpose of the study presented here was to compare the effects of chromosomes 1 and 6 on malignant melanoma metastasis. By using microcell-mediated chromosome transfer, single copies of neo-tagged human chromosomes 1 or 6 were introduced into the human melanoma cell line MelJuSo. The presence of the added chromosome was verified by G banding of karyotypes, fluorescence in situ hybridization, and screening for polymorphic markers on each chromosome. The incidence and number of metastases per lung after intravenous or intradermal injection of parental MelJuSo cells was significantly (P<0.01) greater than those of hybrids containing either chromosome 1 or chromosome 6, although chromosome 1 was a less potent inhibitor of metastasis than chromosome 6. Cultures established from primary tumors and metastases remained neomycin resistant, suggesting that portions of the added chromosomes were retained. These results strengthen the evidence for the presence of a melanoma metastasis suppressor gene on chromosome 6. neo6/MelJuSo hybrids expressed 2.4- to 3.4-fold more of the melanoma differentiation-associated gene mda-6 (previously shown to be identical to WAF1/CIP1/Sdi1/CAP20) than parental metastatic cells. mda-6/WAF1 is among the candidate genes on chromosome 6. These results also demonstrate, for the first time, the existence of metastasis suppressor genes on human chromosome 1, although these genes appear to be less potent than the one encoded on chromosome 6.

Highly metastatic 13762NF rat mammary adenocarcinoma cell clones stimulate bone marrow by secretion of granulocyte-macrophage colony-stimulating factor/interleukin-3 activity.

Circulating neutrophil (polymorphonuclear leukocyte levels rise 50-fold in 13762NF tumor-bearing rats in proportion to the tumor's metastatic potential. Purified tumor-elicited neutrophils enhance metastasis of syngeneic tumor cells when co-injected intravenously; however, circulating and phorbol ester-activated polymorphonuclear neutrophils do not. The purpose of this study was to elucidate the source of tumor-elicited neutrophils in metastatic tumor-bearing rats. We examined the bone marrow in rats bearing tumors of poorly, moderately, and highly metastatic cell clones. Marrow from rats with highly metastatic tumors had increased cellularity (100%), myeloid to erythroid ratio (10:1), and megakaryocytes compared with control rats (cellularity, approximately 80%; myeloid to erythroid ratio, 5:1), with marrows from rats with moderately metastatic tumors having intermediate values. This suggested production of a colony-stimulating factor by the metastatic cells. To confirm this, bone marrow colony formation from control and tumor-bearing rats was compared. Colony number increased in proportion to the metastatic potential of the tumor. Conditioned medium from metastatic cells supported growth of the granulocyte-macrophage colony-stimulating factor/interleukin-3-dependent 32Dcl3 cell line, but media from nonmetastatic or moderately metastatic cells did not. Antibodies to murine granulocyte-macrophage colony-stimulating factor neutralized 32Dcl3 growth in tumor cell conditioned medium. These results suggest production of a granulocyte-macrophage colony-stimulating factor or interleukin-3-like activity by highly metastatic 13762NF clones and implicate a possible role for colony-stimulating factors in regulating the metastatic potential of mammary adenocarcinoma cell clones.

SOD2 (MnSOD) does not suppress tumorigenicity or metastasis of human melanoma C8161 cells.

Manganese superoxide dismutase (MnSOD, encoded by the SOD2 gene mapping to chromosome 6q25) has been implicated as a tumor suppressor and as a metastasis suppressor in some tumor cell lines. We showed that introduction of an intact chromosome 6 into the metastatic melanoma cell line C8161 completely suppressed metastasis but did not affect tumorigenicity (Welch et al., (1994) Oncogene 9:255). The purpose of this study was to test whether SOD2 is the gene responsible for metastasis suppression. MnSOD protein levels of C8161 (measured by Western blot), before and after transfer of chromosome 6, showed no correlation with metastatic potential. To determine whether the lack of correlation was due to mutant, nonfunctional SOD2, a highly metastatic subclone of C8161 (C8161c1.9) was transfected with functional SOD2 or vector control (pSFFV). Metastatic potential and tumorigenicity were unchanged. Southern and Northern blots confirmed the presence of the transfected SOD2; however, total MnSOD protein and antioxidant activity were not significantly altered. These results suggest that levels of MnSOD are highly regulated within C8161 melanoma cells and that SOD2 does not suppress tumor formation nor metastatic potential in all human melanomas.

Microcell-mediated transfer of chromosome 6 into metastatic human C8161 melanoma cells suppresses metastasis but does not inhibit tumorigenicity.

Structural alterations of chromosome 6, including del(6q), are often associated with metastatic melanoma; therefore, we hypothesized that a metastasis-suppressor gene could be coded on human chromosome 6. Highly metastatic C8161 human malignant melanoma cells exhibit chromosomal changes typical of late-stage melanomas. Using microcell-mediated chromosome transfer, a copy of a normal human chromosome 6 was introduced into C8161. Three randomly selected hybrid clones (neo6/C8161.1, neo6/C8161.2 and neo6/C8161.3) were assayed for metastasis in athymic nude mice. All controls - parental C8161 cells, randomly-selected single cell clones, neo-transfected cell clones, neo11/C8161.2 and neo11/C8161.3 - were tumorigenic (270/272 mice) and metastatic (208/272 mice). neo6/C8161 hybrid cells were still tumorigenic (91/93 mice) but were not metastatic (0/195 mice). The presence of the added chromosomes was verified in cultured and tumor cells by amplification of polymorphic (CA)n markers using PCR-RFLP. The neo6/C8161 hybrids display growth and morphological patterns of more differentiated cells than C8161. In Northern blot analysis an inverse relationship between metastatic ability and metastasis-suppressor gene, nm23-H1, expression is observed - with clone neo6/C8161.1 expressing the highest level of nm23 transcripts, neo6/C8161.2 and neo6/C8161.3 expressing intermediate levels, and barely detectable levels are seen in C8161. Collectively, these results suggest that a malignant melanoma metastasis-regulatory gene may be located on human chromosome 6. These results further demonstrate that tumorigenicity and metastatic ability are distinct phenotypes.

Degradation and intracellular accumulation of a residualizing hyaluronan derivative by liver endothelial cells.

The release and intracellular accumulation of 125I-hyaluronan degradation products was studied in cultured liver endothelial cells with hyaluronan oligosaccharides (relative molecular mass = approximately 44,000) uniquely modified and radiolabeled at the terminal reducing sugar. Two methods were combined to measure 125I-hyaluronan degradation by liver endothelial cells. (a) Cetylpyridinium chloride precipitation of hyaluronan oligosaccharides was used as a rapid, convenient assay to monitor the appearance of hyaluronan degradation products. Hyaluronan oligosaccharides less than 54 to 60 monosaccharides in length were not precipitated with cetylpyridinium chloride and thus were assessed as degraded. (b) Gel filtration chromatography was used to estimate the size range of oligosaccharides produced by liver endothelial cells. After internalization of 125I-hyaluronan, liver endothelial cells released radioactive degradation products into the culture media after a lag period of 2.5 to 3.0 hr. The intracellular accumulation of degraded 125I-hyaluronan was linear for at least 2 hr even though no degradation products were released. The long lag before release of degraded 125I-hyaluronan is likely caused by the modified chemical structure at the reducing end of the hyaluronan derivative; the derivative acts like a residualizing label. After this lag the release of degraded 125I-hyaluronan proceeded linearly for up to 12 hr. The extracellular 125I-hyaluronan degradation products eluted with a distribution coefficient of 1.3 on a gel filtration column. The major intracellular 125I-labeled degradation product showed the same retardation (distribution coefficient = 1.3). This retention may be caused by the hydrophobic aromatic and alkyl modifications to the former reducing sugar, also characteristics of a residualizing label. In addition, at least two larger minor intermediates were observed intracellularly. The rate of intracellular 125I-hyaluronan degradation was dependent on hyaluronan concentration and reached a maximal rate (159 molecules/cell/sec) at 2 x 10(-7) mol/L. This was about half the maximal rate of endocytosis (285 molecules/cell/sec) at a hyaluronan concentration of 1.3 x 10(-7) mol/L. The apparent ligand concentration that gives half-maximal responses for endocytosis and intracellular degradation was 0.6 x 10(-7) and 1.0 x 10(-7) mol/L, respectively.

The endocytic hyaluronan receptor in rat liver sinusoidal endothelial cells is Ca(+2)-independent and distinct from a Ca(+2)-dependent hyaluronan binding activity.

Isolated and cultured rat liver sinusoidal endothelial cells (LECs) retain the ability to specifically bind 125I-hyaluronan (HA) and internalize it using a coated pit pathway [Biochem J, 257:875-884, 1989]. Here we have determined the effect of Ca+2 on the binding and endocytosis of HA by LECs. 125I-HA binding to intact LECs at 4 degrees C occurred both in the absence (10 mM EGTA) or the presence of physiologic concentrations of Ca+2 (1.8 mM). However, the specific binding of 125I-HA to LECs increased linearly with increasing Ca+2 concentrations. After permeabilization with the nonionic detergent digitonin, the Ca(+2)-independent HA binding activity increased approximately 743%, while the Ca(+2)-dependent binding activity was enhanced only approximately 46%. Therefore, the Ca(+2)-dependent HA binding activity appears not to be intracellular, whereas the Ca(+2)-independent HA receptor is found both inside LECs and on the cell surface. When LECs were allowed to endocytose 125I-HA at 37 degrees C in 10 mM EGTA or in 1.8 mM Ca+2, no differences were seen in the extent or rate of endocytosis. When LECs were allowed to endocytose 125I-HA in the presence of 10 mM Ca+2, the amount of cell-associated radioactivity increased approximately 20-50-fold. However, this additional cell-associated 125I-HA was not sensitive to hyperosmolarity and was removed by washing the cells in 10 mM EGTA at 4 degrees C. Therefore, the Ca(+2)-dependent cell-associated 125I-HA had accumulated on the cell surface and had not been internalized. From these studies we conclude that LECs have at least two types of specific HA binding sites.(ABSTRACT TRUNCATED AT 250 WORDS)

Endocytosis of hyaluronic acid by rat liver endothelial cells. Evidence for receptor recycling.

Hyaluronic acid (HA) is cleared from the blood by liver endothelial cells through receptor-mediated endocytosis [Eriksson, Fraser, Laurent, Pertoft & Smedsrod (1983) Exp. Cell Res. 144, 223-238]. We have measured the capacity of cultured rat liver endothelial cells to endocytose and degrade 125I-HA (Mr approximately 44,000) at 37 degrees C. Endocytosis was linear for 3 h and then reached a plateau. The rate of endocytosis was concentration-dependent and reached a maximum of 250 molecules/s per cell. Endocytosis of 125I-HA was inhibited more than 92% by a 150-fold excess of non-radiolabelled HA. HA, chondroitin sulphate and heparin effectively competed for endocytosis of 125I-HA, whereas glucuronic acid, N-acetylglucosamine, DNA, RNA, polygalacturonic acid and dextran did not compete. In the absence of cycloheximide, endothelial cells processed 13 times more 125I-HA in 6 h than their total (cell-surface and intracellular) specific HA-binding capacity. This result was not due to degradation and rapid replacement of receptors, because, even in the presence of cycloheximide, these cells processed 6 times more HA than their total receptor content in 6 h. Also, in the presence of cycloheximide, no decrease in 125I-HA-binding capacity was seen in cells processing or not processing HA for 6 h, indicating that receptors are not degraded after the endocytosis of HA. During endocytosis of HA at 37 degrees C, at least 65% of the intracellular HA receptors became occupied with HA within 30 min. This indicates that the intracellular HA receptors (75% of the total) function during continuous endocytosis. Hyperosmolarity inhibits endocytosis and receptor recycling in the asialoglycoprotein and low-density-lipoprotein receptor systems by disrupting the coated-pit pathway [Heuser & Anderson (1987) J. Cell Biol. 105, 230a; Oka & Weigel (1988) J. Cell. Biochem. 36, 169-183]. Hyperosmolarity inhibited 125I-HA endocytosis in liver endothelial cells by more than 90%, suggesting use of a coated-pit pathway by this HA receptor. We conclude that liver endothelial cell HA receptors are recycled during the continuous endocytosis and processing of HA.

The specific interaction between fibrin(ogen) and hyaluronan: possible consequences in haemostasis, inflammation and wound healing.

We have proposed that fibrin and hyaluronan (HA) are macromolecular regulators during inflammation and wound healing. Here we extend our studies to characterize the specific interaction between fibrin(ogen) and HA. 125I-labelled HA (Mr approximately 32,000) was bound by plastic surfaces coated with human fibrinogen but not bovine serum albumin, ovalbumin, beta-lactoglobin or rabbit immunoglobulin G. 125I-labelled fibrinogen bound to a unique hexylamine derivative of HA coupled to Sepharose and was eluted specifically by HA oligosaccharides in a size-dependent manner. A dot blot assay, in which proteins are adsorbed to nitrocellulose and probed with 125I-HA, also showed specific binding to human fibrinogen. This assay was used to examine fibrinogens from other mammalian species. No specific 125I-HA binding was observed with the protein from horse, rat or cow. Significant binding was detected with human, sheep, rabbit, dog, baboon, goat and pig fibrinogens. Thrombin-induced formation of fibrin clots is also affected by HA, which decreases the lag time before clotting and increases the rate of clot formation. The rate of fibrin polymerization increased over 500% in the presence of 60 microM HA. Furthermore, the structure of the fibrin gel, as assessed by light scattering, was altered by HA or chondroitin sulphate in a concentration-dependent manner. The results support the proposed wound-healing model and indicate that an increase in circulating HA levels could adversely affect haemostasis and increase the risk of thrombosis or bleeding. The interaction between HA and fibrinogen emphasizes the importance of the liver endothelial cell HA receptor in the removal of glycosaminoglycans from the blood. Cultured cells continuously endocytosing 125I-HA for 4 h reutilized their total cellular HA receptors at least once every 50 min even in the presence of cycloheximide. This endocytotic receptor was therefore shown to be part of a recycling system.

Specific intracellular hyaluronic acid binding to isolated rat hepatocytes is membrane-associated.

Intact isolated rat hepatocytes show a small amount of specific 125I-labeled hyaluronic acid (HA) binding. However, in the presence of digitonin, a very large increase in the specific binding of 125I-HA is observed. Chondroitin sulfate, heparin and dextran sulfate were as effective as unlabeled HA in competing for 125I-HA binding to permeabilized hepatocytes, indicating that the binding sites may have a general specificity for glycosaminoglycans. After rat hepatocytes had been homogenized in a hypotonic buffer, more than 98% of the 125I-HA binding activity could be pelleted by centrifugation at 100,000 x g for 1 h. Mild alkaline treatment of hepatocyte membranes did not release 125I-HA binding activity, suggesting that the HA binding site is an integral membrane molecule. Furthermore, trypsin treatment of deoxycholate-extracted membranes destroyed the binding activity, as assessed by a dot-blot assay. This suggests that a protein component in the membrane is necessary for 125I-HA binding activity. Rat fibrinogen could be a possible candidate for the HA binding activity because HA binds specifically to human fibrinogen (LeBoeuf et al. (1986) J. Biol. Chem. 261, 12 586). Also, fibrinogen can be found in a quasi-crystalline form in rat hepatocytes and could be pelleted with the membranes. Rat fibrinogen was not responsible for the 125I-HA binding activity, since (1) purified rat fibrinogen did not bind to 125I-HA, and (2) immunoprecipitation of rat fibrinogen from hepatocyte extracts did not decrease the 125I-HA binding of these extracts. We conclude that the internal HA binding sites are membrane- or cytoskeleton-associated proteins and are neither cytosolic proteins nor fibrinogen.

Affinity and distribution of surface and intracellular hyaluronic acid receptors in isolated rat liver endothelial cells.

125I-Hyaluronic acid (HA) uniquely modified only at the reducing end (Raja, R.H., LeBoeuf, R. D., Stone, G.W., and Weigel, P.H. (1984) Anal. Biochem. 139, 168-177) binds specifically to rat liver endothelial cells in suspension or in culture. About 67-85% of the HA binding sites in isolated cells in suspension and 50% in cultured cells were intracellular, since they were exposed after permeabilizing cells with digitonin. Specific 125I-HA binding at 4 degrees C varied from 60 to 80% for intact cells and from 70 to 90% for permeabilized cells. Freshly isolated permeabilized cells bound about 500,000 HA molecules/cell at saturation. Within 5 h of culture, however, total HA binding decreased to 250,000 molecules/cells and then remained constant for at least 36 h. Surface HA receptor activity was essentially the same on cultured cells or cells in suspension (approximately 10(5)/cell). Cultured cells had 1.8 x 10(5) fewer intracellular receptors/cell. The affinities of surface and intracellular receptors of cells in culture and in suspension were essentially the same. The average Kd, determined by equilibrium binding studies, was 5.8 +/- 2.8 x 10(-8) M (n = 12). Dissociation of bound 125I-HA from permeable cultured cells was rapid (t1/2 = 30.9 min;kappa off = 3.7 x 10(-4) s-1). A variety of carbohydrates had essentially identical effects on 125I-HA binding to surface or total cellular receptors in cells in culture or in suspension. Chondroitin sulfate and heparin competed almost as effectively as unlabeled HA for 125I-HA binding at 4 degrees C. Other saccharides including polygalacturonic acid, dextran, glucuronic acid, and N-acetylglucosamine competed poorly or not at all. We conclude that (i) the 125I-HA binding sites within liver endothelial cells are HA receptors, identical in affinity and specificity to those on the cell surface; (ii) the distribution of cellular HA receptors is similar to other receptor systems with about 50-80% being intracellular; (iii) the liver endothelial cell HA receptor recognizes several glycosaminoglycans; and (iv) the liver endothelial receptor is different in function and characteristics than the fibroblast HA receptor.

The role of hyaluronic acid in inflammation and wound healing.

We have previously hypothesized that hyaluronic acid (HA) and fibrin specifically interact and help orchestrate the early stages of the inflammatory response and wound healing (J. Theoretic. Biol. 119, 219, 1986). In the present work we have extended studies to confirm this prediction. Several approaches were used to show that human fibrinogen specifically binds to hyaluronic acid. In addition, this latter glycosaminoglycan greatly stimulated the in vitro formation of fibrin clots induced by thrombin. The presence of hyaluronic acid also altered the structure of the final fibrin gel. Removal of circulating hyaluronic acid in blood is therefore probably vital in order to maintain normal haemostasis. Evidence is presented to suggest that an endocytic receptor recycling process is responsible for the ability of liver endothelial cells to perform this function and remove HA from the blood. Although hyaluronic acid levels are initially low in a blood clot, the proposed wound-healing model explains why and how HA levels increase and the functional and structural significance of this polysaccharide during wound healing.

The effects of hepatocyte stimulating factor on fibrinogen biosynthesis in hepatocyte monolayers.

The biosynthesis of fibrinogen increased at least eightfold in primary hepatocytes when incubated in the presence of monocyte/macrophage-derived hepatocyte stimulating factor (HSF). The large increase in fibrinogen production is due to increased availability of the mRNAs for the protein since cytodot analysis of cellular RNA showed a 10-12-fold increase in each of the fibrinogen mRNAs. Pulse-chase experiments showed that the time for fibrinogen synthesis, assembly, and secretion was 40-50 min for both control and stimulating conditions. This indicates that the increased production was due principally to the presence of greater amounts of fibrinogen mRNA rather than translation or secretion-specific events. Three lines of evidence indicate that the increase in fibrinogen production was due to HSF effects on transcription: (a) analysis of cytoplasmic levels of each of the fibrinogen mRNAs showed that all three increased at the same rate and to the same extent, demonstrating that HSF affects the three gene products coordinately; (b) Northern gel analysis of cytoplasmic RNA isolated after very brief exposures to HSF showed increases in a large molecular weight fibrinogen RNA precursor; and (c) actinomycin D blocked the HSF-stimulated increase in fibrinogen mRNA species. Furthermore, experiments in which protein synthesis was inhibited by cycloheximide failed to inhibit the increase in fibrinogen mRNAs, indicating new protein synthesis is not required for the HSF stimulation of fibrinogen mRNA. These results are consistent with our hypothesis that HSF is exerting its control of fibrinogen at the level of gene transcription.