Oncogenic ras results in increased cell kill due to defective thermoprotection in lung cancer cells.

BACKGROUND: The survival response of normal cells to heat stress is an upregulation of heat shock proteins and ras protein activation. We hypothesized that in lung cancer cells the presence of oncogenic ras interferes with thermoprotective mechanisms resulting in cell death. METHODS: An equal number of lung tissue culture cells (normal and cancerous) were subjected to either heat stress and then recovery (43 degrees C for 180 minutes, 37 degrees C for 180 minutes) or recovery alone (37 degrees C for 360 minutes). End points were surviving number of cells, cell-death time course, heat shock protein (HSP70, HSC70, HSP27) expression before and after heat stress, and time course for HSP70 expression during heat stress and recovery. Heated cells were compared with unheated control cells, then this difference was compared between cell types. RESULTS: Heat stress in normal cells caused an 8% decrease in cell number versus a 78% +/- 5% decrease in cancer cells (p < 0.05). In normal cells, heat stress caused a 4.4-fold increase in HSP70, no change in HSC70, and a 1.7-fold increase in HSP27. In contrast, cancer cells initially contained significantly less HSP70 (p < 0.05), and there was a 27-fold increase in HSP70 and a 2-fold increase in HSC70 with no HSP27 detected (comparison significant, p < 0.05). HSP70 time course in normal cells showed that HSP70 increased 100-fold, reaching a vertex at 2 hours and remaining elevated for 24 hours; in cancer cells, HSP70 maximum expression (100-fold) peaked at 5 hours,,then decreased to slightly elevated at 24 hours. CONCLUSIONS: Cancer cells with oncogenic ras have defective thermoprotective mechanism(s) causing increased in vitro cell death, which provides an opportunity for thermal treatment of lung cancer.

Effect of postmortem sampling technique on the clinical significance of autopsy blood cultures.

Our objective was to investigate the value of postmortem autopsy blood cultures performed with an iodine-subclavian technique relative to the classical method of atrial heat searing and antemortem blood cultures. The study consisted of a prospective autopsy series with each case serving as its own control relative to subsequent testing, and a retrospective survey of patients coming to autopsy who had both autopsy blood cultures and premortem blood cultures. A busy academic autopsy service (600 cases per year) at University of Texas Medical Branch Hospitals, Galveston, Texas, served as the setting for this work. The incidence of non-clinically relevant (false-positive) culture results were compared using different methods for collecting blood samples in a prospective series of 38 adult autopsy specimens. One hundred eleven adult autopsy specimens in which both postmortem and antemortem blood cultures were obtained were studied retrospectively. For both studies, positive culture results were scored as either clinically relevant or false positives based on analysis of the autopsy findings and the clinical summary. The rate of false-positive culture results obtained by an iodine-subclavian technique from blood drawn soon after death were statistically significantly lower (13%) than using the classical method of obtaining blood through the atrium after heat searing at the time of the autopsy (34%) in the same set of autopsy subjects. When autopsy results were compared with subjects' antemortem blood culture results, there was no significant difference in the rate of non-clinically relevant culture results in a paired retrospective series of antemortem blood cultures and postmortem blood cultures using the iodine-subclavian postmortem method (11.7% v 13.5%). The results indicate that autopsy blood cultures obtained using the iodine-subclavian technique have reliability equivalent to that of antemortem blood cultures.

Elevated levels of annexin I protein in vitro and in vivo in rat and human mammary adenocarcinoma.

Annexin I is a phospholipid and actin binding protein which may play a role in signal transduction to the cytoskeleton. Previous work reported the differential expression of annexin I mRNA among rat adenocarcinoma cell lines of various metastatic potential (MTLn3, MTLn2, MTC.4: highest to lowest, respectively) (Pencil et al. 1993, Breast Cancer Res Treat, 25, 165-74). This relationship has been extended to the protein level in in vitro cultures using Western blotting and flow cytometry. Annexin I protein levels in MTLn3 cells are 3- to 5-fold higher than in MTC.4 cells. The weakly metastatic cell line MTLn2 shows levels 1.5- to 2.5-fold higher than MTC.4. In vivo tumors were produced by injecting 1 x 10(6) cells into mammary fat pads of syngeneic rats and necropsies were performed 40 days later. Semiquantitative immunohistochemical color image analysis was performed using a polyclonal rat annexin I specific antibody. Annexin I protein expression was highest in lung metastases of MTLn3, at 8-fold the levels observed in the MTC.4 primary tumors. MTLn3 cells in the primary tumor had an annexin I specific optical density 3-fold higher than that of cells in the MTC.4 primary tumor. MTLn2 primary tumors had an annexin I specific optical density 1.5-fold higher than MTC.4. A proportion of human mammary adenocarcinomas also have positive annexin I immunoreactivity, often with more uniform annexin I staining in the lymph node metastases. These results suggest that there may be survival advantages for nascent metastatic cells with high annexin I levels.

Induction of whole body hyperthemia with venovenous perfusion.

Whole body hyperthermia can be used for the treatment of metastatic cancer and human immunodeficiency virus infections. The therapeutic effects of hyperthermia are dependent upon the actual temperature of the target tissues. Therefore, homogeneous distribution of heat and precise control of temperature gradients is critical. To describe heat distribution during hyperthermia induced by venovenous perfusion, the authors used multiple channel temperature monitoring and a servo-regulated perfusion/heat exchange system. Young swine were randomly assigned to either a heated group (perfusion-induced hyperthermia, target core temperature at 43 degrees C, n = 6), or a control group (perfusion alone, target core temperature at 38 degrees C, n = 6). Blood was drained from the external jugular vein, heated with a computer assisted heat exchange system, and reinfused through the femoral vein at a flow of 10 ml/kg-1/min-1. Temperature probes in the esophagus, right and left tympanic canals, brain, pulmonary artery, arterial and venous blood, rectus spinae muscle, kidney, rectum, bone marrow, bladder, subcutaneous tissue, gluteus, and skin were simultaneously recorded. During the heat induction phase, the maximum water temperature was 54 degrees C, with a heating gradient of the blood (blood in-blood out) at 6 degrees C. The maximum temperature difference between tissues was 3.6 degrees C (kidney and esophagus) during heat induction, but decreased to 1.75 degrees C during maintenance. Bone marrow temperature was consistently 1-2 degrees C below the average core temperature of 43 degrees C throughout the experiment. The authors conclude that venovenous perfusion can predictably induce hyperthermia, but is associated with heterogenous temperature distribution among organs. Further studies are necessary to evaluate different perfusion and heating patterns to achieve homogenous hyperthermia.

Analysis of the complete sequence of the novel metastasis-associated candidate gene, mta1, differentially expressed in mammary adenocarcinoma and breast cancer cell lines.

To understand the genes and gene products involved in breast cancer invasion and metastasis, we previously isolated ten differentially expressed genes by differential cDNA library screening techniques, using the 13762NF rat mammary adenocarcinoma metastatic system. In this study, we further analysed a novel candidate metastasis-associated gene, mta1, previously designated clone 10.14. Northern blotting analyses showed that the steady-state mRNA expression level of mta1 was fourfold higher in a highly metastatic line (MTLn3) than in a nonmetastatic line (MTC.4). The mta1 gene was expressed at low levels in various normal rat organs, except testis, where it was expressed in high amounts. The mRNA expression levels of the human homologue of this gene were also examined in two human breast cancer metastatic systems; the ratios of mRNA were estimated to be MCF-7 (nonmetastatic):MCF7/LCC1 (invasive):MCF7/LCC2 (metastatic) = 1:2:4 and MDA-MB-468 (nonmetastatic):MDA231 (metastatic) = 1:4. Thus, the expression of this gene directly correlated with metastatic potential in two human systems, as well as in the rat metastatic system. Clone 10.14 was used to isolate a full-length cDNA clone for mta1, yielding the clone p10.14-C4.5, which was sequenced and analysed. Clone p10.14-C4.5 was 2756-bp long and contained a single open reading frame that could encode a protein of 703 amino acid (aa) residues. The aa sequence of mta1 was found to be novel by database homology search and contained possible phosphorylation sites for tyrosine kinase, protein kinase C and casein kinase II. A Pro-rich stretch was found at the C-terminal end that completely matched the consensus sequence for the SH3-binding motif.(ABSTRACT TRUNCATED AT 250 WORDS)

A novel candidate metastasis-associated gene, mta1, differentially expressed in highly metastatic mammary adenocarcinoma cell lines. cDNA cloning, expression, and protein analyses.

To understand the genes involved in breast cancer invasion and metastasis, we analyzed a novel candidate metastasis-associated gene, mta1, which was isolated by differential cDNA library screening using the 13762NF rat mammary adenocarcinoma metastatic system. Northern blot analyses showed that the mRNA expression level of the mta1 gene was 4-fold higher in the highly metastatic cell line MTLn3 than in the nonmetastatic cell line MTC.4. The mta1 gene was expressed in various normal rat organs, especially in the testis, suggesting its essential normal function. The mRNA expression levels of the human homologue of this gene also correlated with the metastatic potential in two human breast cancer metastatic systems. The full-length mta1 cDNA sequence contained an open reading frame encoding a protein of 703 amino acid residues, and sequence analysis by data base homology search indicated that mta1 is a novel gene. The Mta1 protein contained several possible phosphorylation sites, and a proline-rich amino acid stretch at the carboxyl-terminal end completely matched the consensus sequence for the src homology 3 domain-binding motif. Using antibodies raised against glutathione S-transferase-Mta1 fusion protein or a synthetic oligopeptide, Western blots showed that the molecular mass of the Mta1 protein was approximately 80 kDa, and the levels of the Mta1 protein also correlated with the metastatic potential, results similar to those obtained from the Northern analyses. Thus, the novel gene mta1 may encode a molecule that is functional in normal cells as well as in breast cancer invasion and metastasis.

Candidate metastasis-associated genes of the rat 13762NF mammary adenocarcinoma.

Differential hybridization was used to isolate genes potentially involved in the process of metastasis. Ten complementary DNAs (cDNAs) that were differentially expressed between a highly metastatic (MTLn3) and a nonmetastatic (MTC.4) line of the rat 13762NF mammary adenocarcinoma were isolated and sequenced. Examination of the EMBL/GenBank database revealed that one of the genes had a high degree of homology (98.8%) to annexin I (also known as calpactin II). Quantitative analysis of Northern blot hybridizations showed that the annexin I-like sequence was expressed 4- to 7-fold higher in MTLn3 than in MTC.4 cells. Steady state mRNA levels were also low in MTLn2, a cell line of low metastatic potential closely related to MTLn3, but were not related to metastatic potential in colon adenocarcinoma or melanoma cells. Two of the cDNAs (designated 8.11 and 10.14) were found to be novel. The expression of 10.14 mRNA (3.2 kb) was 4-fold higher in MTLn3 than in MTC.4 cells. Sequencing of the 10.14 cDNA (2.2 kb) revealed a putative open reading frame of 583 amino acids that was also novel. Expression of 8.11 mRNA (> 7 kb) inversely correlated with metastatic potential. Another differentially transcribed gene was highly homologous to ERK2 (extracellular signal related kinase 2), a mitogen-activated protein kinase (MAPK). Northern analysis of ERK2 expression revealed 3-fold higher amounts of a 1.3 kb mRNA in MTLn3 than in MTC.4 cells. Higher levels of ERK2 mRNA were generally seen in the more metastatic human colon but not in melanoma cell lines. We also corroborated the work of Taniguchi (Nucl Acids Res 19:6949, 1991) by independently identifying EF-1 alpha as a putative metastasis-associated gene.

Evidence against involvement of calcium in carbon tetrachloride-dependent inhibition of lipid secretion by isolated hepatocytes.

Carbon tetrachloride (CCl4)-induced inhibition of very low density lipoprotein (VLDL) secretion was studied in isolated hepatocytes. The hypothesis that inhibition of secretion is due to altered calcium homeostasis following CCl4-dependent inhibition of endoplasmic reticulum calcium sequestration was investigated. Inhibition of VLDL secretion by CCl4 was not dependent on extracellular calcium, since inhibition occurred when extracellular calcium was reduced to 0.1 microM. CCl4 inhibited hepatocyte VLDL secretion more rapidly than it inhibited microsomal calcium sequestration. Further, the concentration of CCl4 that produced half-maximal inhibition of VLDL secretion was about one-half the concentration required to produce half-maximal inhibition of microsomal calcium sequestration. The calcium ionophore A23187 did not mimic the action of CCl4 in inhibiting VLDL secretion under conditions in which A23187 altered cellular calcium homeostasis. The results that an alteration of calcium homeostasis is not involved in inhibition of VLDL secretion by carbon tetrachloride.

Assessment of the role of calcium ion in halocarbon hepatotoxicity.

Halogenated hydrocarbons (CCl4, BrCCl3, 1,1-dichloroethylene, bromobenzene) cause a wide spectrum of dysfunction and injury in liver cells. An early effect of CCl4, BrCCl3, and 1,1-dichloroethylene is destruction of the Ca2+-sequestering ability of the endoplasmic reticulum, and it has been suggested that this lesion leads to subsequent disruption of other cell functions. Work to test this hypothesis has begun in this and other laboratories. While it appears that redistribution of intracellular Ca2+ does occur following these agents, the importance of this in cell injury is not fully resolved. Current results suggest Ca2+ redistribution may be involved in some cases (e.g., surface blebbing caused by bromobenzene), but not in others (e.g., inhibition of lipid secretion by CCl4).

Carbon tetrachloride-dependent inhibition of lipid secretion by isolated hepatocytes. Characterization and requirement for bioactivation.

Secretion of lipid as very low density lipoprotein (VLDL) by isolated hepatocytes was studied in a system in which the cells were exposed to a constant concentration of carbon tetrachloride (CCl4) throughout the duration of the incubation. Inhibition of secretion was characterized in terms of CCl4 concentration and duration of incubation. Half-maximum inhibition of VLDL secretion occurred at 80 microM CCl4. At 390 microM CCl4, VLDL secretion was inhibited 40% in 2 min and was suppressed completely in 5 min. No CCl4-dependent release of cellular glutamic oxaloacetic transaminase occurred under the conditions studied. Evidence is presented indicating that the metabolism of CCl4 is required for the expression of CCl4-dependent inhibition of lipid secretion. With respect to the parameters studied, the isolated hepatocyte system closely mimics the hepatic response of the intact animal to CCl4.

Evidence against a role for disturbed hepatocellular calcium homeostasis in the fatty liver of carbon tetrachloride hepatotoxicity.

A consensus does not exist regarding the nature of mechanisms linking the initial events of CCl4 metabolism to emergence of the classical indices of CCl4 liver cell injury. The possibility that a CCl4-dependent disturbance of intrahepatocellular calcium homeostasis might be a linking mechanism was investigated with isolated hepatocytes in suspension. CCl4-dependent inhibition of very low density lipoprotein secretion was studied. On the basis of kinetic data, dose-response data, and failure of elevated cytosolic calcium levels to inhibit lipid secretion, it was concluded that disturbed intracellular calcium homeostasis probably is not important in CCl4-dependent inhibition of secretion of very low density lipoproteins.

Loss of calcium sequestration capacity in endoplasmic reticulum of isolated hepatocytes treated with carbon tetrachloride.

A method for treating isolated hepatocytes with a known and toxicologically meaningful concentration of CCl4 was used to study the effect of the haloalkane on the calcium sequestration capacity of microsomal vesicles derived from the hepatocytes. The essence of the method is to allow a very small volume of CCl4 to come to diffusion equilibrium in a closed system characterized by a gas phase that is large relative to the aqueous phase in which hepatocytes are to be suspended. The final equilibrium concentration of CCl4 in the aqueous phase is 174 microM. When hepatocytes are added to the aqueous phase, after at least 1 hr of incubation of the resealed system at 37 degrees C, the cells do not liberate GOT to the medium above control values, and their capacity to exclude trypan blue is unimpaired. However, covalent binding of the 14C from 14CCl4 is maximal within the first 10 min of incubation, and the calcium sequestering capacity of microsomal vesicles derived from hepatocytes incubated in the presence of CCl4 is reduced to 30% of control levels after 20 min. This system should prove useful in study of early effects of CCl4-dependent derangement of hepatocellular calcium homeostasis.