Laszlo G. Boros

Mass isotopomer study of the nonoxidative pathways of the pentose cycle with [1,2-13C2]glucose

We present a single-tracer method for the study of the pentose phosphate pathway (PPP) using [1,2-13C2]glucose and mass isotopomer analysis. The metabolism of [1,2-13C2]glucose by the glucose-6-phosphate dehydrogenase, transketolase (TK), and transaldolase (TA) reactions results in unique pentose and lactate isotopomers with either one or two13C substitutions. The distribution of these isotopomers was used to estimate parameters of the PPP using the model of Katz and Rognstad (J. Katz and R. Rognstad. Biochemistry 6: 2227-2247, 1967). Mass and position isotopomers of ribose, and lactate and palmitate (products from triose phosphate) from human hepatoma cells (Hep G2) incubated with 30% enriched [1,2-13C2]glucose were determined using gas chromatography-mass spectrometry. After 24-72 h incubation, 1.9% of lactate molecules in the medium contained one 13C substitution ( m 1) and 10% contained two 13C substitutions ( m 2). A similar m 1-to- m 2ratio was found in palmitate as expected. Pentose cycle (PC) activity determined from incubation with [1,2-13C2]glucose was 5.73 ± 0.52% of the glucose flux, which was identical to the value of PC (5.55 ± 0.73%) determined by separate incubations with [1-13C] and [6-13C]glucose.13C was found to be distributed in four ribose isotopomers ([1-13C]-, [5-13C]-, [1,2-13C2]-, and [4,5-13C2]ribose). The observed ribose isotopomer distribution was best matched with that provided from simulation by substituting 0.032 for TK and 0.85 for TA activity relative to glucose uptake into the model of Katz and Rognstad. The use of [1,2-13C2]glucose not only permits the determination of PC but also allows estimation of relative rates through the TK and TA reactions.

Nonoxidative pentose phosphate pathways and their direct role in ribose synthesis in tumors: Is cancer a disease of cellular glucose metabolism?

Pentose phosphate pathways (PPP) are considered important in tumor proliferation processes because of their role in supplying tumor cells with reduced NADP and carbons for intracellular anabolic processes. Direct involvement of PPP in tumor DNA/RNA synthesis is not considered as significant as in lipid and protein syntheses. Currently, PPP activity in tumor cells is measured by lactate production, which shows a moderate activity: about 4% to 7% compared with glycolysis. Recent data generated in our laboratory indicate that PPP are directly involved in ribose synthesis in pancreatic adenocarcinoma cells, through oxidative steps (< 31%) and transketolase reactions (69%). These findings raise serious questions about the adequacy of lactate in measuring PPP activity in tumors. We hypothesize that ribose, not lactate, is the major product of PPP in tumor cells. Control of both oxidative and nonoxidative PPP may be critical in the treatment of cancer. PPP are substantially involved in the proliferation of human tumors, which raises the prospect of new treatment strategies targeting specific biochemical reactions of PPP by hormones related to glucose metabolism, controlling thiamine intake, the cofactor of the nonoxidative transketolase PPP reaction, or treating cancer patients with antithiamine analogues.

Inhibition of the oxidative and nonoxidative pentose phosphate pathways by somatostatin: a possible mechanism of antitumor action

Long-acting somatostatin analogs have recently become supplemental drugs in the treatment of neurofibroma because of their marked tumor growth inhibitory effect. Somatostatin is currently under extended evaluation in other cancers as a possible supplemental drug to the treatment protocols in use. The mode of action is not known. Somatostatin has been shown to cause glucose intolerance by inhibiting glucose-6-phosphate dehydrogenase (G6PD) in fish liver. Recent data generated in our laboratory indicate that it is this pathway and the transketolase reactions of the pentose cycle (PC) which are directly involved in the ribose synthesis process of pancreatic adenocarcinoma cells. In cell culture, somatostatin alone inhibited glucose carbon recycling through the PC by 5.7%, which was increased to 19.8% in combination with oxythiamine, a competitive inhibitor of transketolase. Oxythiamine produced strong apoptosis in in-vitro hosted tumor cells. We hypothesize that somatostatin- and oxythiamine-induced antiproliferative action is mediated by the inhibition of G6PD, transketolase, or both.

Dehydroepiandrosterone-sulfate inhibits pancreatic carcinoma cell proliferation in vitro and in vivo

BACKGROUND Dehydroepiandrosterone-sulfate (DHEA-S) is a potent inhibitor of glucose-6 phosphate dehydrogenase, the rate limiting enzyme of the hexose monophosphate shunt, a biochemical pathway that provides substrate for DNA synthesis in neoplastic tissue. DHEA-S has been shown to inhibit the growth of neoplasms arriving from human skin, lung, colon, and mammary tissue. This study evaluates the effect of DHEA-S on human pancreatic cancer cell lines in vitro and in vivo. METHODS In vitro, the human pancreatic adenocarcinoma cell lines MiaPaCa-2, Capan-1, Capan-2, CAV and Panc-1 were treated with concentrations of 1.9 mumol/L to 1000 mumol/L DHEA-S in 1% dimethylsulfoxide (DMSO) for 5 consecutive days. Cell proliferation was determined by a nonradioactive cell proliferation assay and compared with DMSO treated controls. In vivo testing was performed by inoculating two cell lines, MiaPaCa-2 and Panc-I, into the flank of 40 male nude athymic mice in four study groups. After 1 week of growth, 667 mg/kg DHEA-S in 1% DMSO or 0.2 ml 1% DMSO alone in the control group was administered by daily intraperitoneal injection. Body weight and tumor size was recorded weekly, and tumor weight was measured after 3 weeks of treatment. RESULTS In vitro cell proliferation was decreased in the five cell lines by 36% to 62% of controls (p < 0.001) at 500 mumol/L DHEA-S. In vivo, after 2 weeks, tumor size was only 76% (p < 0.008) and 67% (p < 0.005) of the controls. After 3 weeks of treatment, tumor size was 73% (p < 0.001) and 53% (p < 0.001) of controls, and tumor weight was decreased by 73% in MiaPaCa-2 (p < 0.001) and 66% in Panc-1 (p < 0.001). Radioimmunoassay measurements of DHEA-S and testosterone from DHEA-S treated mouse plasma showed a significant increase in circulating levels of these hormones. CONCLUSIONS DHEA-S achieves high serum levels after intraperitoneal injection without elevation of serum testosterone levels and produces no significant toxicity. Treatment with DHEA-S results in a significant reduction of proliferation of human pancreatic cancer cells in culture and when grown as subcutaneous tumors in athymic nude mice.

Reversal of enhanced pancreatic cancer growth in diabetes by insulin.

BACKGROUND Epidemiologic and animal studies have linked pancreatic cancer growth with both diabetes and fat intake. This study examined the influence of insulin treatment on pancreatic cancer growth in diabetes. Diabetes-induced elevations in levels of glucose and free fatty acids were correlated with enhanced tumor growth both in vivo and in vitro. METHODS Hamsters were divided into three groups: control (n = 15), streptozocin-diabetic (n = 20), or insulin-treated diabetic (n = 20). Diabetes was induced with streptozocin and treated with a continuous subcutaneous infusion of insulin delivered via osmotic pumps. Five x 10(5) H2T hamster pancreatic cancer cells were implanted into the cheek pouch. Levels of plasma glucose and fatty acids were measured, and their effect on H2T cell division was assessed in vitro with a spectrophotometric cell proliferation assay. RESULTS Levels of plasma glucose and fatty acids were elevated in streptozocin-diabetic animals and normalized with insulin treatment. After 21 days of growth, tumor weight was 36 mg in the control group, 156 mg in the diabetic group (p < 0.01 versus other groups), and 33 mg in the insulin-treated diabetic group. In vitro dose-dependent promotion of cell growth was shown for glucose (250%), linoleic acid (287%), linolenic acid (169%), and oleic acid (98%). CONCLUSIONS Insulin ameliorated enhanced tumor growth in this model of diabetes. Glucose and free fatty acids mobilized during diabetes may serve as fuel for established pancreatic cancers.

Variable effect of streptozotocin-diabetes on the growth of hamster pancreatic cancer (H2T) in the Syrian hamster and nude mouse

BACKGROUND Streptozotocin-diabetes prevents induction of pancreatic tumors in several animal models and inhibits the growth of established human pancreatic cancer implants in nude mice. However, it also promotes growth of the hamster pancreatic cancer cell line, H2T, in the Syrian hamster. To test the hypothesis that these contradictory effects are due to tumor host differences, the growth of the H2T cell line was examined in the streptozotocin-diabetic nude mouse. METHODS H2T cells were implanted subcutaneously into streptozotocin-diabetic nude mice (n = 10) and untreated control mice (n = 10). After 21 days, tumors were excised and weighed. Plasma insulin and somatostatin were determined by radioimmunoassay. RESULTS After 3 weeks, tumors in the control group weighed 118 mg and tumors in the diabetic group weighed 28 mg (p < 0.001). Plasma insulin was significantly decreased in the streptozotocin-treated animals compared with control animals (insulin, 23 microU/ml vs 31 microU/ml; p < 0.001). In contrast, somatostatin was significantly elevated in the streptozotocin-diabetic group compared with the control group (somatostatin, 179 pg/ml versus 54 pg/ml, p < 0.001). Competitive binding studies revealed specific cell surface receptors for insulin (Kd, 15.5 nmol/L), and somatostatin (Kd, 2.5 nmol/L) on the H2T cells. In an in vitro cell proliferation assay, cell division was promoted by insulin (p < 0.01, maximum +11%) and inhibited by somatostatin (p < 0.01, maximum -18%). CONCLUSIONS The variable effect of streptozotocin-diabetes on pancreatic cancer growth is due to differences in the tumor host. The growth of pancreatic cancer, particularly in streptozotocin-diabetic nude mice, may be influenced by gut peptides in a receptor-dependent fashion.

Interleukin‐1α gene expression during wound healing

Interleukin‐1α is known to be constitutively produced by epidermal keratinocytes under normal conditions, and injection of this cytokine enhances wound reepithelialization. However, no studies have characterized the temporal sequence of interleukin‐1α gene expression over the time course of wound healing, and the cellular sources of this cytokine have not been identified. In the present studies, levels of interleukin‐1α messenger RNA in wound tissue isolated from SKH‐1 hairless mice were characterized and the cells that produced interleukin‐1α immunoreactive protein over a 10‐day time course of wound healing were defined. A time‐dependent upregulation in interleukin‐1α gene expression occurred immediately (4 hours) after a full‐thickness wound was made, which represented a four‐fold increase over levels of cytokine gene expression detected in nonwounded skin. Upregulation of cytokine gene expression correlated with an immediate increase in plasma interleukin‐1α levels and was followed by an increase in interleukin‐1α immunoreactive protein localized to keratinocytes within the leading edge of the wound and epidermis, as well as to neutrophils within the dermis. The rapid increase in local and systemic interleukin‐1α levels correlated with the infiltration of a significant number of neutrophils into the wound site and with the proliferation of both basal keratinocytes and dermal fibroblasts. Given the known ability of interleukin‐1α to regulate proliferation and migration of epidermal keratinocytes and to indirectly induce leukocyte chemotaxis, the results of the present studies suggest that interleukin‐1α may be an important cytokine with both local and systemic actions that are linked to the initiation of critical cellular events early in wound healing.

CD-ROM source data uploaded to the operating and storage devices of an IBM 3090 mainframe through a PC terminal

A powerful method of processing MEDLINE and CINAHL source data uploaded to the IBM 3090 mainframe computer through an IBM/PC is described. Data are first downloaded from the CD-ROMs PC devices to floppy disks. These disks then are uploaded to the mainframe computer through an IBM/PC equipped with WordPerfect text editor and computer network connection (SONNGATE). Before downloading, keywords specifying the information to be accessed are typed at the FIND prompt of the CD-ROM station. The resulting abstracts are downloaded into a file called DOWNLOAD.DOC. The floppy disks containing the information are simply carried to an IBM/PC which has a terminal emulation (TELNET) connection to the university-wide computer network (SONNET) at the Ohio State University Academic Computing Services (OSU ACS). The WordPerfect (5.1) processes and saves the text into DOS format. Using the File Transfer Protocol (FTP, 130,000 bytes/s) of SONNET, the entire text containing the information obtained through the MEDLINE and CINAHL search is transferred to the remote mainframe computer for further processing. At this point, abstracts in the specified area are ready for immediate access and multiple retrieval by any PC having network switch or dial-in connection after the USER ID, PASSWORD and ACCOUNT NUMBER are specified by the user. The system provides the user an on-line, very powerful and quick method of searching for words specifying: diseases, agents, experimental methods, animals, authors, and journals in the research area downloaded. The user can also copy the TItles, AUthors and SOurce with optional parts of abstracts into papers under edition. This arrangement serves the special demands of a research laboratory by handling MEDLINE and CINAHL source data resulting after a search is performed with keywords specified for ongoing projects. Since the Ohio State University has a centrally founded mainframe system, the data upload, storage and mainframe operations are free.

An automatized computer-method utilizing Procomm Plus and DataEase (4.2) PC and SAS (6.06) mainframe software for isolated, perfused guinea-pig heart studies

A powerful, time sharing and automatized method of a comprehensive data analysis for isolated, perfused guinea-pig heart studies is described. Data are collected using DataEase PC software (version 4.2) into forms with data fields specified for vital parameters consistently recorded in isolated, perfused heart studies (HR, CBF, PEAKPRESSURE, DPDT, MVO2). After running, DataEase reports the data and information is uploaded to an IBM 3081D mainframe computer on each day of heart experiment and data collection. The uploading process, the data archival and the statistical analyses are automatized by Procomm Plus commands written in Aspect Source Program (.ASP) Files for logging, data transforming and file management procedures. The ASPCOMP.EXE compiler compiles these .ASP files into Aspect Script eXecutable (.ASX) programs, which run on the PC in our laboratory and activate WYLBUR (IBM 3081D Batch-job service and Command file processor) edited files in the mainframes electronic devices then upload, backup and save data into these files. SAS EXE files containing program instructions for the data analyzing system are then forced by Procomm Plus to operate over the data just uploaded. SAS reads the DATA files by its INFILE facility and performs comprehensive statistical analyses and produces hard output including graphics and JOB reports of dose-response- and logaritmic scale curves for delivery to team members. This computerized and automatized method developed for isolated, perfused guinea-pig heart studies is capable of performing multiple file transfer, sophisticated statistical analyses and graphic procedures after one keystroke on the PC (Alt-F5 in Procomm Plus section) and also facilitates a consistent and convenient method for planning, controlling and standardizing experiments. The method is based on an interactive computer conversation between the PC in the laboratory and the remotes WYLBUR editor. No human presence is needed; however, in case of failure, Procomm Plus gives one of the team members supervising the system a phone call in order to get human help.