Constitutively active CCK2 receptor splice variant increases Src-dependent HIF-1 alpha expression and tumor growth.

Gastrointestinal (GI) cancers ectopically express multiple splice variants of the cholecystokinin-2 (CCK(2))/gastrin receptor; however, their relative contributions to the cancer phenotype are unknown. The aim of this study was to compare the effects of CCK(2) receptor (CCK(2)R) and CCK(2i4sv)R expression on cell growth both in vitro and in vivo using a human epithelial cell model, HEK239. In vitro, receptor variant expression did not affect cell proliferation either in the absence or presence of agonist. However, in vivo, the expression of CCK(2i4sv)R, but not CCK(2)R, increases HEK293 tumor growth in a constitutive, Src-dependent manner. Enhanced tumorigenicity of CCK(2i4sv)R is associated with an Src-dependent increase in the transcription factor, hypoxia-inducible factor-1alpha, its downstream target, vascular endothelial growth factor and tumor micro-vessel density, suggesting that CCK(2i4sv)R may contribute to the growth and spread of GI cancers through agonist-independent mechanisms that enhance tumor angiogenesis.

Protective action of luminal bile salts in necrotizing acute pancreatitis in mice.

Bile salts in the intestinal lumen act to inhibit the release of cholecystokinin (CCK). Recent studies have shown that CCK may play a permissive role in the development of acute pancreatitis. In this study, the amount of luminal bile salts in female Swiss Webster mice was either decreased by feeding 4% (wt/wt) cholestyramine or increased by feeding 0.5% sodium taurocholate for 1 wk. Plasma levels of CCK were stimulated by cholestyramine and inhibited by taurocholate. Then, acute pancreatitis was induced either by caerulein injections, or by feeding a choline-deficient, ethionine-supplemented (CDE) diet. Feeding of cholestyramine significantly decreased survival from 25% to 0% in the CDE pancreatitis, and increased the magnitude of elevation of serum amylase levels and the extent of pancreatic necrosis in both models of pancreatitis; CCK-receptor blockade with CR-1409 completely abolished the adverse effects of cholestyramine. In contrast, feeding of taurocholate significantly increased survival to 100% and decreased the elevation of serum amylase and pancreatic necrosis; CCK-8 antagonized these actions of taurocholate. Luminal bile salts appear to provide a physiologic protection against necrotizing pancreatitis, at least in part, both by inhibiting the release of CCK and by promoting resistance of the pancreas to CCK excessive stimulation in vivo.

Characterization of promoter elements required for cell-specific expression of the neurotensin/neuromedin N gene in a human endocrine cell line.

Expression of the gene encoding neurotensin/neuromedin N (NT/N) is mostly limited to the brain and specialized enteroendocrine cells (N cells) of the distal small intestine. We have analyzed the NT/N DNA sequences upstream of the RNA start site that direct cell-specific expression using a novel human endocrine cell line, BON, that resembles intestinal N cells in several important aspects, including NT/N precursor protein processing, ratios of different NT/N mRNA forms, and high levels of constitutive expression of the NT/N gene. Transient transfection assays with plasmids with progressive 5' deletions of the rat NT/N promoter identified the proximal 216 bp of 5' flanking sequences as essential for high-level constitutive NT/N expression in BON cells. In addition, a detailed mutational analysis defined multiple regions within the proximal 216 bp that contribute to cell-specific NT/N expression. These elements include a proximal cyclic AMP response element (CRE)/AP-1-like motif (TGACATCA) that binds c-Jun, JunD, CRE-binding (CREB), and ATF proteins, a near-consensus glucocorticoid response element, and a distal consensus AP-1 site that binds c-Fos, Fra-1, and JunD. In addition, elements contained within two 21-bp imperfect direct repeats play an important role in NT/N expression in BON cells and may bind novel factors that act as positive regulators of NT/N expression. DNase I footprinting and gel shift analyses demonstrate that the sites identified by mutational analysis, and at least one additional site, specifically bind BON cell nuclear proteins in vitro. We speculate that a complex pattern of regulation requiring interaction between a proximal CRE/AP-1-like motif and other upstream control elements play an important role in the high-level constitutive expression of NT/N in the human endocrine cell line BON. In addition, the BON cell line provides a unique model to further characterize the factors regulating cell-specific NT/N expression and to better understand the mechanisms responsible for the terminal differentiation of the N-cell lineage in the gut.

Effects of cyclosporin A and alpha-difluoromethylornithine on the growth of hamster pancreatic cancer in vitro.

The growth and survival of hamster H2T pancreatic ductal adenocarcinoma in vitro are known to be significantly reduced by inhibitors of polyamine biosynthesis. alpha-Difluoromethylornithine (alpha-DFMO) is a specific and irreversible inhibitor of ornithine decarboxylase, the rate-limiting enzyme in polyamine biosynthesis. alpha-DFMO treatment inhibits the growth of H2T pancreatic cancer cells and decreases H2T cell survival in vitro and in vivo. In the present study, the effects of cyclosporin A (CsA) were examined on growth, survival, and polyamine levels in H2T pancreatic ductal adenocarcinoma in vitro. CsA had inhibitory effects on H2T pancreatic cancer growth similar to those of alpha-DFMO; these effects were blocked by the addition of the polyamine putrescine. Polyamine levels were found to be significantly altered in cells treated with CsA and/or alpha-DFMO. The combination of CsA (8.3 X 10(-4) mM) and alpha-DFMO (0.5 mM or 1.0 mM) inhibited H2T cell survival to a greater extent than either agent alone. These results suggest that CsA in combination with other agents that inhibit polyamine synthesis may be useful for the treatment of pancreatic cancer.

Treatment of hamster pancreatic cancer with alpha-difluoromethylornithine, an inhibitor of polyamine biosynthesis.

Polyamines are essential for cell division and growth. Inhibition of polyamine biosynthesis by alpha-difluoromethylornithine (DFMO) on the growth of hamster H2T pancreatic cancer was investigated both in vitro and in vivo. Cell-doubling time (TD) and survival fraction were determined after a single treatment with DFMO (5 mM). We examined the ability of putrescine to reverse the growth-inhibitory effect of DFMO. The TD for cells treated with DFMO in vitro was 49.6 +/- 5.7 versus 25.4 +/- 2.6 hours for control. The addition of putrescine to DFMO-treated H2T cells showed a reversal of the growth-inhibitory effect of DFMO. Cytotoxicity in vitro increased with prolonged treatment; the survival fraction after 24 hours of treatment was 32%; after 48 hours, 19%; after 72 hours, 13%; and after 92 hours, 8%. We performed two separate animal experiments. In experiment I, H2T cells were injected into the cheek pouch of male Syrian golden hamsters; controls did not receive DFMO. Continuous treatment with 3% DFMO in the drinking water was begun 7 days before, on the day of, or 7 days after tumor cell injection. In experiment II, 4 groups were treated identically to those in experiment I. An additional group of 10 hamsters received 3% DFMO and no tumor, and another additional group of 10 hamsters were housed individually with 3% DFMO begun 7 days after tumor cell injection. Tumor size, body weight, water, and food intake were measured. DFMO treatment in vivo significantly inhibited tumor size and inhibited growth of pancreatic cancer by as much as 50% of control. Our results demonstrate a significant antiproliferative effect of DFMO on the growth of pancreatic adenocarcinoma both in vitro and in vivo.

Loss in cell killing effectiveness of anticancer drugs in human gastric cancer clones due to recovery from potentially lethal damage in vitro.

The ability of human gastric cancer clones to recover from potentially lethal damage was studied. Recovery was greatest following treatments with bleomycin or Adriamycin; the recovery ratios (i.e., survival) increased almost 8-fold during a posttreatment incubation period. Recovery was also possible following treatments with actinomycin D, 1,2:5,6-dianhydrogalactitol, and diaziquone; however, the recovery ratios never increased above 2. No recovery was observed following treatment with 5-fluorouracil. Recovery from potentially lethal damage may be related to the heterogeneity in survival responses observed following treatment with some anticancer drugs. Bleomycin and Adriamycin treatments result in large heterogeneous survival fractions among these human stomach cancer clones, and the potentially lethal damage recovery ratios were larger (and variable). However, actinomycin D, diaziquone, and 1,2:5,6-dianhydrogalacticol produce very uniform killing effects in these cells and the recovery ratios are very much smaller and less variable. Finally the large amount of recovery observed after bleomycin or Adriamycin treatments resulted in the loss of cell killing effectiveness of the agents. Because the survival fractions increased during the recovery period, the net effect on cell killing was reduced to an amount normally obtained with doses that were up to six times smaller.

Role of gastrin and gastrin receptors on the growth of a transplantable mouse colon carcinoma (MC-26) in BALB/c mice.

We recently reported trophic response of transplantable mouse colon cancer cells (MC-26) to pentagastrin, in vivo, and demonstrated gastrin receptors on MC-26 cells, in vitro. In the present study, growth of MC-26 cells in mice, in response to pentagastrin, was studied in relation to binding kinetics and capacity of gastrin receptor. Gastrin receptor levels on mouse fundic and colonic membranes and on MC-26 cellular membranes were determined before MC-26 cell inoculation and designated as Day 0 levels. Four groups of mice were next inoculated with MC-26 cells and given injections of either pentagastrin (treated) or normal saline (control) for 10 or 15 days. At the end of the treatment periods, body, tumor, fundic, and colon weights were noted and gastrin receptor measured. tumor and fundic weights increased significantly within 15 days of pentagastrin treatment, compared to control values. In control (non-pentagastrin treated) mice, the binding affinity of gastrin receptor on tumor membranes was significantly decreased and associated with the complete loss of high-affinity gastrin receptor (Kd = less than 0.5 nM) by Day 15 of tumor growth. On the other hand, both the binding affinity and gastrin receptor levels of tumor membranes were maintained at Day 0 values by pentagastrin treatment. Endogenous gastrin was therefore ineffective in maintaining high-affinity gastrin receptor on control tumors. A significant number of low-affinity gastrin-binding sites (Kd = less than 2 nM) appeared in control tumors by Day 15, which could reflect rapid dedifferentiation or conformational changes of gastrin receptor in the absence of high levels of normal regulatory hormones. These studies demonstrate that the trophic effects of gastrin on MC-26 cells are probably mediated by its regulation and maintenance of the binding affinity and capacity of gastrin receptor on the cancer cells, in vivo.

Specific binding and growth effects of bombesin-related peptides on mouse colon cancer cells in vitro.

In the present study, we characterized specific binding of bombesin (BBS)/gastrin-releasing peptide (GRP) to mouse colon cancer (MC-26) cells. MC-26 cells were inoculated into male BALB/c mice subdermally, and tumors were harvested from mice 21-28 days postinoculation. Tumor membranes were analyzed for binding to GRP-related peptides, using either 125I-GRP or 125I-tyrosine4-BBS. Under optimal binding assay conditions, BBS displaced specific binding of both 125I-GRP and 125I-tyrosine4-BBS in a dose-dependent manner, and a curvilinear displacement resulted. Specific binding data, analyzed by either a Scatchard or a Lineweaver-Burk plot, demonstrated presence of 2 classes of specific binding sites, arbitrarily named type I and type II sites. Type I sites had a high binding affinity [Kd 0.45 +/- 0.05 nM (SE)] and a relatively low capacity (226 +/- 27 fmol/mg membrane protein), whereas type II sites had a 10-20-fold lower binding affinity and approximately 6-7-fold higher capacity. BBS/GRP binding sites were specific for GRP-related peptides and demonstrated no significant binding affinity for all other unrelated peptides tested. Relative binding affinity of GRP analogues was in the order of GRP (14-27) greater than neuromedin C greater than or equal to BBS greater than or equal to GRP (1-27) greater than neuromedin B (for the later, P greater than 0.05 versus other peptides). Two BBS receptor antagonists, [D-Arg1,D-trp7,9,Leu11]-substance P (spantide) and [Leu13-psi-(CH2NH)Leu14]BBS also inhibited specific binding of 125I-GRP in a dose-dependent manner. Molecular weight of GRP/BBS binding proteins on tumor membranes was determined by cross-linking methods. A major molecular form (greater than 80-90%) (Mr approximately 75,000) and a minor Mr approximately 180,000 band were evident, both under reducing and nonreducing conditions. BBS (0.5-50 nM) demonstrated a significant dose-dependent growth effect on MC-26 cells in vitro, in terms of [3H]thymidine and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide uptake; these studies indicate that the BBS/GRP binding sites on MC-26 cells may serve as functional receptors and mediate the growth effects of BBS on MC-26 cells.

Clinical significance of gastrin receptors in human colon cancers.

We have measured gastrin receptors (GR) in surgical specimens from 67 patients with primary colon cancers in order to determine the clinical significance of GR in colon cancer. GR analysis was performed on these specimens, and 22 cancers (32.8%) had no detectable GR. Thirty-eight cancers (56.7%) had high-affinity (Kd less than 1.0 nM) levels of GR. Seven cancers (10.4%) had only low-affinity GR (Kd greater than 1.0 nM). Twenty patients (29.9%) had cancers with GR greater than 10 fmol/mg protein. Mean GR content was significantly greater (11.8 +/- 2.9 fmol/mg protein) in Dukes' Stage A and B cancers when compared to Stage C and D cancers (6.2 +/- 1.6 fmol/mg protein). A significantly greater percentage (52.4%) of patients in the early stages (A and B) had tumors with greater than 10 fmol/mg protein compared to patients with more advanced (C and D) cancers (19.6%). GR content did not correlate with histological differentiation, patient age, or preoperative carcinoembryonic antigen levels. No difference in the GR content was noted between left and right colon cancers or in patients of different sex or race. GR content of normal colon mucosa correlated with the GR content of colon cancers from the same surgical specimen, suggesting that these tumors maintain their normal complement of GR. In the early period of follow-up, 12 of 43 (28%) Stage C and D patients with GR less than 10 fmol/mg protein have died, whereas all 8 Stage C and D patients with GR greater than 10 fmol/mg protein are alive. GR content of colon cancers may have prognostic significance and may identify a group of patients with colon cancer that may benefit from hormonal therapy with antigastrin drugs.

Effects of gastrin on 3',5'-cyclic adenosine monophosphate, intracellular calcium, and phosphatidylinositol hydrolysis in human colon cancer cells.

Gastrin is a trophic factor for some human colon cancer cells. However, the signal-transduction pathways by which gastrin regulates growth are still unknown. We examined the effect of synthetic human gastrin-17 (G-17) on signal-transduction pathways and cell growth using 4 different human colon cancer cell lines (LoVo, COLO 320, HT-29, and HCT116). G-17 stimulated the production of cyclic AMP in LoVo, COLO 320, and HCT116 cells, while G-17 stimulated phosphatidylinositol hydrolysis and mobilization of intracellular calcium in HT-29 cells. The growth-regulatory effect of G-17 on these colon cancer cells (stimulatory on LoVo, COLO 320, and HT-29 cells; inhibitory on HCT116 cells) was well correlated with the effect of G-17 on the signal-transduction pathway in each cell line. We further examined the effect of a selective cholecystokinin-B type receptor antagonist, JMV 320, on G-17-induced signal-transduction pathways and G-17-regulated growth. In each cell line, the effect of JMV 320 on G-17-induced signal-transduction pathways was well correlated with that on G-17-regulated growth. G-17 appears to regulate, at least to some extent, growth of human colon cancer cells through gastrin receptor-linked signal-transduction pathways that are cell-specific.

Selective reduction of human tumor cell populations by human granulocytes in vitro.

Human peripheral blood granulocytes at a 10:1 effector cell:target cell ratio were shown by an in vitro semiquantitative staining procedure to reduce the number of human tumor cells, but not that of normal cells. Microscopic observations revealed that this selective reduction of tumor cells by granulocytes was a function of both detachment and cytolysis. The cytotoxic effect of granulocytes on the tumor cells was confirmed by a quantitative 5-[125I]iodo-2-deoxyuridine release assay. The data indicate that human granulocytes at a relatively low effector:target cell ratio (10:1) have the capacity to recognize and destroy human tumor cells in vitro.

Effects of bombesin on growth of human small cell lung carcinoma in vivo.

Bombesin-like peptides are found in many different human tumors and are thought to function as an autocrine growth factor for small cell lung cancer in humans. In this study, a human small cell lung carcinoma (NCI-H69) was s.c. implanted bilaterally into the flanks of 12 nude mice. The mice were randomized and divided into two groups and given either bombesin (20 micrograms/kg) or saline i.p. 3 times a day. Tumor areas were measured twice weekly for 6 wk. At sacrifice, the tumors and normal pancreas were excised, weighed, and assayed for DNA, RNA, and protein content. Significant stimulation of tumor growth was observed at weeks 4, 5, and 6. Tumor weight at sacrifice was significantly elevated (77%) above the control, as was DNA content (78%). Bombesin significantly increased the weight (42%), DNA (48%), and protein (61%) contents of the normal mouse pancreas. We conclude that bombesin may act as an autocrine growth factor, or indirectly through the release of other growth factors, on human small cell lung carcinoma.

Inhibition of pentagastrin-stimulated up-regulation of gastrin receptors and growth of mouse colon tumor in vivo by proglumide, a gastrin receptor antagonist.

We have recently demonstrated that gastrin stimulates growth of mouse colon cancer (MC-26) in vivo by regulation of gastrin receptors (GR). In the present study, we have tested the effect of proglumide (PGL), a GR antagonist, on the trophic and GR-regulatory effects of gastrin on MC-26 tumors. Four groups of 12 mice each were inoculated with 5 X 10(4) MC-26 cells and given injections of either normal saline (control), pentagastrin (PG), PGL, or both PG + PGL for 21 days. At the end of the treatment period, body, tumor, fundic, and colon weights were noted and GR measured. Two types of specific gastrin-binding sites were found on tumor cell membranes of control mice, one with high binding affinity (Kd = less than 1.0 nM) and low capacity (GR), and the other with a very high capacity and a low affinity (Kd = greater than 0.1 microM) (type 2 gastrin-binding sites). Only the type 1 GR were observed on the fundic mucosal and colon membranes. PG treatment resulted in a significant weight increase of the tumors with an up-regulation of only type 1 GR. On the other hand, PG had no significant effect on fundic mucosal and colonic GR levels, but caused a significant increase in fundic mucosal weights. PGL completely inhibited both the trophic and GR up-regulatory effects of PG on tumors, but incompletely reduced the PG-stimulated fundic mucosal weight gain, indicating differential sensitivity of tumor and normal tissues to PGL. PGL, in the absence of PG, was slightly trophic for normal fundic mucosa, but had no effect on MC-26 tumors and normal colon. The one striking effect of PGL, in the presence of PG, was the significant lowering of the binding affinity of type 1 GR for gastrin on both the tumor and normal gastrointestinal tissues. This effect may be another mechanism by which PGL interferes with the actions of PG on MC-26 tumors and fundic mucosa of mice.

Human carcinoid cell production of paracrine growth factors that can stimulate fibroblast and endothelial cell growth.

A serotonin-secreting human pancreatic carcinoid cell line (BON) is demonstrated to express transcripts for all three mammalian types of transforming growth factor beta (TGF beta 1, 2, and 3). Similarly, freshly excised carcinoid tumors from six patients were also found to express mRNA for all three of the type-beta TGFs. Medium conditioned by BON cells had detectable TGF beta activity, although most of the activity was latent as determined by radioreceptor assay with and without prior acid treatment. However, nonactivated BON-conditioned medium stimulated DNA synthesis, soft agar growth, and an increase in TGF beta 1 and fibronectin mRNA expression in AKR-2B fibroblasts. In addition, BON-conditioned medium had a potent endothelial cell growth-stimulatory activity. Since the TGF beta s inhibit growth of endothelial cells, the presence of other growth factors was suspected. TGF alpha, c-sis, and basic fibroblast growth factor transcripts were also found to be expressed by the BON carcinoid cells. These data indicate that multiple peptide growth factors may have a paracrine role in the desmoplastic reaction accompanying carcinoid tumors.

Intratumor variability in prognostic indicators may be the cause of conflicting estimates of patient survival and response to therapy.

The DNA index, percentage of S-phase cells, proliferation fraction, and glutathione (GSH) content were determined at more than 1100 separate sites in 140 human tumors and 140 normal tissues. The study showed that the variability was so great from site to site within a tumor that there was only a 61% chance of identifying an aneuploid tumor clone (when present) if only a single site sample was analyzed for DNA content. Similar broad variability was observed in the percentage of S-phase cells, proliferation fraction, and glutathione content. Since these tumor characteristics are often used to predict the outcome of therapy and patient survival, the inaccuracy and underestimation of the test results may cause conflicting or erroneous predictions. The probability of finding an aneuploid clone or elevated percentage of S-phase cells proliferation fraction and GSH content increased dramatically as the number of sample sites studied per tumor was increased. Statistical analyses indicated that in order to achieve a 90% probability that the test results for these parameters were representative of the whole tumor: (a) all single site testing should be abandoned; (b) assays should be performed on samples taken from 3-7 different sites within each tumor; or (c) samples from each tumor should be pooled and the analyses run on a thoroughly mixed or homogenized aliquot of the multisite sample.

Establishment and characterization of an in vitro model system for human adenocarcinoma of the stomach.

Ten permanent clones derived from a single biopsy specimen of an untreated human adenocarcinoma of the stomach were established and characterized in vitro. Tissue culture growth properties, doubling times, plating efficiencies, growth fractions, cell cycle phase distributions, DNA indices, modal chromosome numbers, and ploidies were determined. Growth fractions were nearly 100%, and doubling times ranged from 23 to 37 hr. The plating efficiencies were generally high for tumor cells in culture, ranging up to 70%. Modal chromosome numbers varied from 45 to 48, with a wider range of variability in about 25% of the cells studied in each clone. In addition, the parent cell line (from which the clones were isolated) was shown to grow in athymic mice and to have the same histochemical and cytological characteristics as the specimen taken from the patient. It is important to characterize human tumor cells in vitro in this detailed manner, since they serve as excellent model systems for other studies involving the heterogeneous responses to drugs and radiation. The identification of mechanisms of drug sensitivity and resistance and the testing of drug and radiation combination treatment schedules in such in vitro systems can provide valuable insight into the design of clinical protocols for treatment of stomach cancer in humans.

Inhibition of human gastric adenocarcinoma xenograft growth in nude mice by alpha-difluoromethylornithine.

We studied the effect of inhibition of polyamine biosynthesis by alpha-difluoromethylornithine on the growth of a human gastric adenocarcinoma (CLEES) xenotransplanted in nude mice. CLEES is a well-differentiated gastric adenocarcinoma of the intestinal type. The doubling time has ranged from 7 to 10 days through 11 passages. Electron microscopic and immunohistochemical studies comparing the original tumor and xenotransplants showed similar structure and similar amounts of carcinoembryonic antigen. Polyamine biosynthesis is required for cell division. alpha-Difluoromethylornithine inhibits ornithine decarboxylase, the rate-limiting enzyme in polyamine biosynthesis. In this study, 48 athymic mice were used in two experiments. In the first experiment, two groups of 12 mice each were inoculated with CLEES tumor cells and received either tap water or a 3% alpha-difluoromethylornithine solution as drinking water. Tumor size was measured twice weekly. Tumor size was significantly decreased from controls by the fourth week of treatment and at all points of analysis thereafter for 7 wk. In the second experiment, alpha-difluoromethylornithine significantly reduced tumor concentrations of the polyamines putrescine and spermidine. In addition, the tumor content of DNA was significantly reduced in treated mice (0.64 +/- 0.16 mg) compared to controls (4.76 +/- 0.92 mg). Our data suggest that inhibition of polyamine biosynthesis may be a useful component of multidrug chemotherapy for human gastric adenocarcinoma. Establishment of tumor lines such as this gastric adenocarcinoma will facilitate further studies on the biological behavior of human gastric cancer and its response to chemotherapeutic manipulation in vivo.

Changes in glutathione content and resistance to anticancer agents in human stomach cancer cells induced by treatments with melphalan in vitro.

A clone of a human gastric carcinoma cell line was used to determine whether cells which had survived a treatment with Melphalan would express altered survival responses when treated again with this agent 1 week or more later. Cells were treated for 1 h each week with 2 micrograms/ml (99% lethal dose). After the first Melphalan treatment, the cells exhibited a 10-fold reduction in sensitivity to Melphalan. This was preceded by a 2-fold increase in intracellular glutathione content. By the end of 10 weekly treatments, the cells were 50 times more resistant than controls (based on changes in survival fractions). They also demonstrated collateral resistance to Actinomycin D, 1-(2-chloroethyl)-3-(4-methylcyclohexyl)-1-nitrosourea, galactitol, and X-rays, but showed no change in sensitivity to 5-fluorouracil, bleomycin, and Adriamycin. The resistance to Melphalan was not reversible when treatment was withheld for 4 weeks on two different occasions. The results suggest that treatment with a high dose of Melphalan either selects an existing population of cells with a high GSH content or induces mutations leading to increased GSH content or both, and this results in the expression of greater Melphalan resistance at the time of other treatments. Furthermore, Melphalan treatment stimulates a 50% increase in GSH content in resistant cells in just 6 h, an 85% increase in 36 h, and a 150% increase in 72 h. L-Buthionine sulfoximine partially reversed the expression of resistance to Melphalan by inducing a 60% reduction in intracellular glutathione content.

Treatment-induced changes in sensitivity in a multiclonal human tumor mixture model in vitro.

An in vitro model has been devised so that mixtures of human tumor cells can be grown together for studies related to drug-induced or -selected changes in sensitivity. In the studies reported here, two human astrocytoma clones, one sensitive and one resistant to 1-(2-chloroethyl)-3-(4-methylcyclohexyl)-1-nitrosourea (MeCCNU), were carefully matched for doubling times, cell cycle phase distributions, and colony-forming efficiencies. The clones were mixed and grown together, and after only three weekly treatments with MeCCNU (10 micrograms/ml for 1 h each week) the sensitive cells in the mixture were killed, leaving behind a population that was almost 100% resistant to further exposures to MeCCNU. The loss of the sensitive cells from the mixture each week was easily detected by visual observation of flow microfluorometry histograms since the clones had different DNA indices. Repeated weekly exposures of the unmixed resistant clone (AST 1-1) to MeCCNU caused very little accumulated cell kill. Similar exposures of the unmixed sensitive clone (AST 3-4) produced a linear decrease in survival over the first three weekly treatments with 10 micrograms MeCCNU/ml, but after that time these cells became progressively more resistant to MeCCNU. It is unlikely that the change to resistance in the AST 3-4 clone occurred because of contamination with the resistant AST 1-1 cells, because their DNA index remained stable. These data show that repeated treatments with a single agent can cause a tumor cell population to become more resistant. It remains to be tested whether this resistance was the result of cellular interactions, drug-induced changes in sensitivity, or selection for resistant cells already present in the populations. This mixture model may be useful in studies on how cellular interactions influence growth and drug sensitivity in tumor and normal cell populations.

Use of 1,2:5,6-dianhydrogalactitol in studies on cell kinetics-directed chemotherapy schedules in human tumors in vivo.

Recently, it has been shown that 1,2:5,6-dianhydrogalactitol (DAG) can cause reversible alterations in cell cycle kinetics. Following treatment of CHO cells in vitro and Ehrlich ascites tumor cells in vivo, significant increases in the fraction of cells in S phase were observed to occur, and this was followed by an increase in the fractions of cells in G2 and mitosis. Treatments with S or G2-M phase-specific drugs at the peak enrichment times after DAG was given resulted in greater cell kills than when given by any other schedule. We have extended these kinetics-directed drug schedule studies to human tumors in vivo. The first phase was to determine whether DAG could be used to perturb cell kinetics in vivo as effectively in patients as it was in vitro. In 14 of 17 tumors studied, increases in the S-phase fractions were observed (ranging from 30 to 240% increases). The hr at which the S-phase peaks were observed (post-DAG treatment) was variable among the patients and among the tumors studied. However, this points out the value of obtaining actual cell kinetics data from serially biopsied tumors growing on the body surface and illustrates the importance that these data may have in helping to select an optimal time at which to give an S phase-specific drug. If such tumor cell kinetics-directed scheduling is ultimately shown to be effective, it will represent a means of individualizing therapy for a large fraction of tumor patients whose tumors are growing on or near the surface of the body. The tumors utilized in these studies were squamous carcinomas of the head and neck, skin, anus, and cervix; adenocarcinomas of the breast and rectum; and malignant melanoma. The second phase of this study will be to determine the tumor responses in patients treated with such kinetics-directed schedules.