- Email: [email protected]
Krebs Cycle, Pentose Phosphate Pathway, and Glycolysis in the Uninvolved Gastric Mucosa of Peptic Ulcer and Gastric Cancer Patients
GASTROENTEROLOGY 73:1320-1325, 1977 Copyright © 1977 by the American Gastroenterological Association
Vol. 73, No. 6 Printed in U.S A .
KREBS CYCLE, PENTOSE PHOSPHATE PATHWAY, AND GLYCOLYSIS IN THE UNINVOLVED GASTRIC MUCOSA OF PEPTIC ULCER AND GASTRIC CANCER PATIENTS RALPH L. ORWELL, PH.D. , AND DOUGLAS w. PIPER, M.D., F.R.C.P. , F.R.A.C.P. The Department of Medicine, University of Sydney at the R oyal North Shore Hospital , Sydney, New South Wales, Australia
Uninvolved gastric mucosa from duodenal ulcer, gastric ulcer, and gastric cancer patients was incubated with [l-14C]glucose and [6- 14C]glucose in order to assess the relative contributions of the pentose phosphate pathway and Krebs cycle to glucose metabolism. [14C]Glucose counts retained by the tissue, glycolysis, and pyruvate formation were also measured. Tumor tissue from the cancer patients was included in the study. Less than 1.2% of the glucose entering the tissues was metabolized via the pentose phosphate pathway, suggesting that this pathway plays a minor role in energy production from glucose. The major determinant of energy production was the Krebs cycle. Its contribution to glucose rr. 3tabolism was greatest in the body mucosa of duodenal ulcer patients, less in the uninvolved body mucosa of gastric ulcer patients, and lower still in the corresponding body mucosa of gastric cancer patients. The low levels of Krebs cycle activity seen in the latter tissue resembled those of uninvolved antral mucosa. The smallest Krebs cycle contribution was seen in tumor tissue. [ 14C]Glucose counts retained by the tissue and glycolysis both tended to vary inversely with Krebs cycle activity among the tissues studied. Thus, both were small in the body mucosa of noncancer patients and somewhat larger in the body mucosa of cancer patients, in uninvolved antral mucosa and in tumor tissue. Disease states in gastric mucosa have been related to several criteria, for example, the extent to which body and antral mucosa are involved, 1 the number of parietal cells present, 2 and the extent of gastritis and intestinal metaplasia.3-5 More recently, evidence has been presented that the processes involved in glucose metabolism may be a relevant variable. 6 • 7 In the present study, the body and antral mucosa of chronic peptic ulcer patients, together with the body mucosa, antral mucosa, and tumor tissue of gastric cancer patients,
were compared in in vitro experiments as regards rates of glucose utilization, the relative contributions to glucose metabolism of the pentose phosphate pathway and Krebs cycle, and the rates of glycolysis within the tissues. Methods
Tissue procurement. Samples of gastric t issue were obtained from gastric ulcer and gastric cancer patients undergoing partial gastrectomy. These were sampled for uninvolved body and antral mucosa at least 3.0 em from any lesion. With Received April4, 1977. Accepted June 29, 1977. duodenal ulcer patients, the surgical procedure (vagotomy) limited sampling to uninvolved body mucosa; the latter was Address requests for reprints to: Professor D. W. Piper, Department of Medicine, Royal North Shore Hospital of Sydney, St. resected before the vagotomy step. Gastric cancer tissue was Leonards, New South Wales 2065, Australia. obtained from the outer sections of tumors, avoiding necrotic This study was supported by grants-in-aid from the National areas. Mucosa representative of every sample was preserved for Health and Medical Research Council of Australia, the New South Wales State Cancer Council, and the University of Sydney Cancer histological examination; the pathologist's findings were used to classify samples as body or antral mucosa, and to ensure Research Fund. Dr. Orwell is Research Officer, National Health and Medical that uninvolved mucosa from cancer patients was free from Research Council of Australia. His present address is: School of Life malignant cells. The cancer patients all possessed histologiSciences, The New South Wales Institute of Technology, West- cally confirmed adenocarcinomas. Biochemical methods . Gastric mucosa was placed, within bourne Street, Gore Hill, New South Wales 2065, Australia. The authors acknowledge with thanks the expert assistance of 20 min of resection, in ice-cold Krebs-Ringer phosphate methe staff of the Department of Pathology, Royal North Shore Hospi- dium, pH 7.4,8 and dissected into intact mucosal segments of tal, Sydney, who performed the histological examinations, and the area approximately 4 mm. 2 Tumor tissue was sliced into cooperation of our surgical colleagues, Dr. V. H. Cumberland, Dr. segments of similar dimensions. Portions (0.1 g) of prepared tissue were preincubated 15 min in Warburg flasks without R. M. Hollings, and Associate Professor G. A. E. Coupland in labeled substrate and then incubated for 3 hr at 37°C with providing gastric tissue. 1320
December 1977
1321
GLUCOSE OXIDATION IN STOMACH MUCOSA
agitation in 2.0 ml of the same medium containing 5.0 mM unlabeled glucose and 0.20 1-LCi of 14C-labeled glucose per flask. All flasks were gassed with 100% 0 2 ; the center wells contained filter paper wicks impregnated with 0.1 ml of 10% KOH (w/v). Incubation was stopped by chilling flasks to 4°C and adding 0.50 ml of 1 M perchloric acid via the sidearm. After 1.5 hr with occasional agitation to collect 14 C0 2 , the suspension was centrifuged for 10 min at 600 x g and the pellet was washed by resuspending in 1.0 ml of water and recentrifuging. The pellet was then dried over P 2 0 5 , weighed, rehydrated with 0.3 ml of water, and finally digested in 2.0 ml of 1 M hyamine hydroxide in methanol (50°C, 1 hr). 9 A 0.40-ml aliquot was analyzed for radioactivity. The supernatant and washing were combined, neutralized with 0.60 ml of 1.0 M potassium bicarbonate and, after removal of potassium perchlorate, stored at -30°C. Glucose and pyruvate were assayed in extracts from duplicate flasks by the methods of Lowry et al. 10 Lactate was similarly assayed using Hohorst's method. 11 The paper wicks, together with 0.10 ml of water used to rinse each center well, were transferred to counting vials containing 0.40 g of Cab-O-SiP 2 and 10 ml of scintillation fluid (naphthalene 120 g; 2,5-diphenyloxazole (PPO) 4.0 g; 1,4-bis[2-(5-phenyloxazolyl)]benzene (POPOP) 0.05 g; toluene 400 ml; 1,4-dioxan 300 ml; 2-methoxyethanol300 ml). Counting was performed in a Packard Tri-Carb model 3375 counter with 45 to 50% efficiency. Hyamine hydroxide digests were counted in the same system without Cab-0-Sil at 50 to 55% efficiency. In all cases quenching was corrected for by means of external standards. Net changes in glucose, lactate, and pyruvate were obtained by subtracting the values given by duplicate tissue samples which received only the preincubation step. Yields of 14 C0 2 from [1- 14 C]glucose and [6- 14 C]glucose were used firstly to calculate the pentose phosphate pathway contributions to glucose metabolism, using specific 14 C0 2 yields as defined by Katz and Wood. 13 Secondly, the 14 C0 2 yield from [6- 14 C]glucose was used as a direct measure of over-all Krebs cycle activity, although it is more accurately an estimate of glucose carbon entering the cycle. Figures given for 14 C label retained by the tissue refer to experiments in which [6- 14 C]glucose was used. All results were calculated on the basis of the dry weight of tissue recovered after the experiment and expressed as micromoles per gram of dry tissue per hour. These data were further analyzed by expressing the glucose equivalent of each metabolite as a percentage of the glucose utilized.
Reagents and chemicals . [V 4 C]Glucose and [6- 14 C]glucose were obtained from The Radiochemical Centre, Amersham, United Kingdom. PPO, POPOP, and Cab-0-Sil were purchased from Consolidated Nucleonics, Sydney, Australia, and hyamine hydroxide was from Calbiochem (Aust.) Ltd., Sydney, Australia. Enzymes, cofactors, and substrates were from Boehringer Mannheim (Aust. ) Ltd., Melbourne, Australia. All other reagents were analytical reagent grade. Statistical analysis. Student's t-test was used for all comparisons.
Results The results for the six types of tissue studied will first be expressed in terms of the molar quantities of glucose utilized and metabolites formed, and secondly, in terms of the proportions of the glucose metabolized by various pathways. Finally, data on the relative importance of aerobic and anaerobic pathways for energy production in the mucosa is presented.
Molar Quantities of Glucose Utilized and Metabolites Formed in Gastric Mucosa and Tumor Tissue These data are given in table 1, expressed as micromoles per gram of dry tissue per hour. Noteworthy features are: Glucose utilization. The body mucosa of duodenal ulcer patients utilized glucose more rapidly than any of . the other tissues studied (table 1; P < 0.05). Among the latter, similar rates of glucose utilization were observed (P = 0.15 to 0.89). Lactate and pyruvate production. Lactate production was higher in duodenal ulcer patients' body mucosa than in gastric ulcer patients' body (P < 0.005) and antral (P < 0.02) mucosa, and in gastric cancer patients' body mucosa (P < 0.02) (table 1). It was also higher in · tumor tissue than in gastric ulcer group body mucosa (P < 0.01), gastric ulcer antral mucosa (P < 0.05), and in cancer group body mucosa (P < 0.03). The cancer patients' antral mucosa gave a relatively high level of lactate formation, which exceeded that found in gastric ulcer patients' body mucosa (P < 0.05). Apart from these differences, similar amounts of lactate were formed by the tissues studied (P = 0.12 to 0.80).
TABLE 1. Utilization of 14C-labeled glucose by segments of gastric mucosa and slices of gastric cancer tissue from duodenal ulcer (DUJ, gastric ulcer (GUJ, and gastric cancer (GCaJ patients; means ± BE. Segments or slices (1 00 mg) were incubated for 3 hr at 3 7oC in 2 .0 ml of Krebs-Ringer phosphate medium containing 5 .0 mM glucose and 0 .20 JLCi of {1-'
Uninvolved antral mucosa
Uninvolved body mucosa
No. ofpatients Glucose utilized C0 2 from C-1 [1- 14 C]glucose C0 2 from C-6 [6-' 4 C]glucose Lactate" Pyruvate" C-6 retained by tissue [6-' 4 C]glucose) a
DU
GU
GCa
GU
GCa
Gastric cancer tissue
Tissue 1
Tissue 2
Tissue 3
Tissue 4
Tissue 5
Tissue 6
12 ± ± ± ± ± ±
15 46.0 ± 3.6 6.29 ± 0.81 5.91 ± 0.88 38.6 ± 7.1 0.11 ± 0.05 7.94 ± 0.87
13 43.1 ± 4.2 4.48 ± 0.90 3.75 ± 0.77 44.6 ± 7.6 O.o7 ± 0.03 7.78 ± 0.92
72.9 13.63 13.07 81.3 0.22 6.48
7.4 1.56 1.60 11.6 O.o7 0.74
Includes contribution from endogenous substrates. _
14 39.0 ± 2.41 ± 1.92 ± 47.9 ± 0.30 ± 10.21 ±
3.7 0.47 0.37 6.9 0.18 1.27
10 45.2 ± 4.1 3.58 ± 0.62 2.73± 0.54 67.8 ± 13.1 0.09 ± 0.04 10.38 ± 1.15
47.1 3.06 1.78 77.0 0.29 7.56
8 ± 8.0 ± 0.69 ± 0.56 ± 11.3 ± 0.12 ± 1.43
1322
Vol . 73, No.6
ORWELL AND PIPER
Duodenal ulcer patients' body mucosa and tumor tissue both yielded significantly more pyruvate than cancer patients' body mucosa (P = 0.05). The high pyruvate yields in the first-mentioned tissues parallel the lactate results and point to the coupling of lactate and pyruvate levels in the tissue via lactate dehydrogenase. A high level of pyruvate production was also seen in gastric ulcer group antral mucosa, although this result was not significantly different from values obtained for the other tissues (P = 0.22 to 0.94). Elsewhere, no differences in pyruvate formation were seen (P = 0.12 to 0.94).
C0 2 production from C-1 and C-6 of glucose. C0 2 production from carbon 1 of glucose, which reflects glucose oxidation in both the pentose phosphate pathway and Krebs cycle, 13 was much higher in duodenal ulcer patients' body mucosa than in any other tissue (P < 0.001) (table 1). It was also higher in the uninvolved body mucosa of gastric ulcer patients than in the uninvolved antral mucosa of both gastric ulcer (P < 0.001) and gastric cancer patients (P < 0.03) and in gastric cancer tissue (P < 0.02). Similar levels of C0 2 production from carbon 1 were seen in the uninvolved body mucosa of the gastric ulcer and cancer patients (P = 0.15). Finally, the value given by cancer group body mucosa was greater than that of gastric ulcer group antral mucosa (P < 0.05). None of the remammg differences observed between tissues were significant (P = 0.15 to 0.59). C0 2 production from carbon 6 of glucose, used here as a direct measure of Krebs cycle activity, 13 was very high in body mucosa of duodenal ulcer patients, which gave a value greater than that of any of the remaining tissues (P < 0.001) (table 1). The body mucosa of gastric
I
GLUCOSE (PERCENT
ulcer patients was next in activity with a value which exceeded those of antral mucosa from the same patients (P < 0.001), antral mucosa from cancer patients (P < 0.02), and tumor tissue (P < 0.005). Furthermore, glucose carbon 6 oxidation was greater in the body mucosa of gastric ulcer, as compared to gastric cancer patients by a margin which closely approached significance (P = 0.07). In turn, cancer group body mucosa gave a value which exceeded that of gastric ulcer patients' antral mucosa (P < 0.05). No further differences were seen among the tissues compared in table 1 (P = 0.09 to 0.82). Retention of glucose C-6 by the tissues. These results (table 1) provide a measure of the extent to which glucose carbon has been diverted into synthetic pathways such as protein synthesis and proteoglycan (mucus) formation. They may also include a contribution from low molecular weight, perchloric acid-soluble compounds not completely removed from the tissue during the washing steps. Duodenal ulcer patients' body mucosa retained less glucose carbon 6 than the uninvolved antral mucosa of both gastric ulcer (P < 0.03) and gastric cancer (P < 0.01) patients. Otherwise, the tissues were similar with respect to this parameter (P = 0.09 to 0.92).
Proportions of Glucose Metabolized via Various Pathways in Gastric Mucosa and Tumor Tissue These results are illustrated in figure 1, where all values are expressed as percentages of the glucose taken up from the medium by each tissue. It is emphasized that the values given for the pentose phosphate pathway, Krebs cycle, and glucose C-6 retained by the tissues refer to the fate of exogenous glucose, whereas
MfTA.IOUSM IN GASTRIC MUCQ$A
OF GLUCOSE
TAKEN
UP &Y EACH
TISSUE:
MEANS
t
SE )
~~
FIG. 1. Relative contributions to carbohydrate metabolism of the pentose phosphate pathway (PP), the tricarboxylic acid (TCA) cycle, incorporation into the tissue (Tissue) and glycolysis (Lactate and ~yruvate): Uninvol~ed gastric .mucosa from. du.odenal. ulcer (D.U.), gastric ulcer (G.U .) and gastric cancer (G.Ca.) patients, together With gastnc cancer tissue were mcubated as md1cated m table 1. All figures are expressed as percentages of the glucose utilized by the tissues. Means± SE. See Results for method used to calculate Deficit term.
December 1977
GLUCOSE OXIDATION IN STOMACH MUCOSA
those given for lactate and pyruvate represent products formed from exogenous glucose and endogenous substrates. Consequently, it was possible for the sum of the quantities tabulated in figure 1 to exceed 100% and this sum is clearly of no value in calculating the experimental recoveries of exogenous glucose. Nevertheless, because of its usefulness, a deficit term was calculated to represent glucose taken up but which could not be accounted for in terms of the other parameters (fig. 1). Only cancer tissue gave a sum exceeding 100%, so that in this case a surplus, rather than a deficit, was recorded. Pentose phosphate pathway . This pathway made only a small contribution to glucose metabolism, equivalent to 0.5 to 1.2% of glucose utilized (Fig. 1). Its percentage contribution was greater in tumor tissue than in the uninvolved antral mucosa of gastric ulcer patients (P = 0.05), but was otherwise similar among the six tissues studied (P = 0.18 to 0.95). Glycolytic pathway. The percentage corresponding to lactate formation (fig. 1) tended to be lower in body mucosa than in antral mucosa and to be especially high in cancer tissue. The lactate value for cancer tissue exceeded that of body mucosa from both gastric ulcer (P < 0.005) and gastric cancer (P < 0.05) patients. The lactate value was also higher in cancer patients' antral mucosa than in gastric ulcer body mucosa (P < 0.05). Apart from these differences, the percentage equivalent to lactate formation was similar among the tissues studied (P = 0.09 to 0.92). The corresponding figures for pyruvate formation were similar for all six types of tissue (P = 0.10 to 0.85) (fig. 1). Tricarboxylic acid cycle. The percentage contribution of the Krebs cycle to glucose metabolism was greater in duodenal ulcer group body mucosa than in any of the other five tissues (P < 0.05) (fig. 1) . Gastric ulcer patients' body mucosa was next in activity with a value exceeding that of any of the four remaining tissues (P < 0.05). Considering only uninvolved body mucosa, the tricarboxylic acid cycle contributions discriminated between the three patient groups, ranking them in order of decreasing activity as follows (P < 0.05): duodenal ulcer; gastric ulcer; gastric cancer. In gastric ulcer patients, the relative importance of the Krebs cycle was significantly greater in the body than in the antral mucosa (fig. 1) (P < 0.001). In cancer patients, on the other hand, this contrast was not apparent. The body mucosa of these patients gave a relatively small Krebs cycle contribution which, although greater than that of gastric cancer tissue (P < 0.05), resembled that of antral mucosa from either gastric ulcer (P = 0.08) or gastric cancer (P = 0.30) patients. The smallest Krebs cycle contribution to glucose metabolism was seen in tumor tissue (fig. 1). It yielded a value which was less than that of body mucosa from any of the three patient groups (P < 0.05), but which resembled those of antral mucosa from the gastric ulcer (P = 0.30) and cancer (P = 0.18) patients. Pathways leading to incorporation of glucose into gastric tissues. The glucose fraction retained by the
1323
tissue pellet (fig. 1) was relatively small in those tissues showing appreciable tricarboxylic acid cycle activity. Thus it was smaller in duodenal ulcer patients' body mucosa than in any of the other tissues (P < 0.02) . It was also smaller in gastric ulcer group body mucosa than in cancer group antral mucosa by a margin which was close to significance (P = 0.07). Conversely, the glucose fraction retained by the pellet was high in those tissues showing low Krebs cycle activity. This was evident in the similar, relatively high percentages recorded for gastric ulcer antral mucosa, gastric cancer body, and antral mucosa and cancer tissue in figure 1 (P = 0.42 to 0.98). This tendency toward an inverse relationship between Krebs cycle activity and [6- 14 C]glucose counts incorporated into the tissue suggests that in gastric tissues there may be competition for glucose between oxidative and synthetic pathways. Deficit. The deficit calculated in figure 1 was influenced most by the largest of the measured parameters, lactate formation. It was smallest in those tissues where lactate formation was high, notably in antral mucosa and cancer tissue. It is possible that in these tissues, substantial amounts of endogenous substrates, rather than exogenous glucose were utilized for energy production during the experiment. The largest deficit was seen in gastric ulcer group body mucosa which gave a higher value than cancer group antral mucosa (P < 0.04) or cancer tissue (P < 0.02).
Relative Importance of Aerobic and Anaerobic Pathways Table 2 gives ratios chosen to illustrate the relative importance of the major aerobic and anaerobic pathways in each tissue. Ratio A was derived from the data summarized in figure 1 and used to calculate ratio B on the basis that the complete oxidation of 1 mole of glucose is capable of yielding the equivalent of 38 moles of ATP, whereas its conversion to lactate yields only 2 moles of ATP. 14 Higher ratios were seen in body mucosa from duodenal ulcer and gastric ulcer patients than ih the two types of antral mucosa or in cancer tissue (table 2; P < 0.05). The ratios given by cancer patients' body mucosa resembled those of the other uninvolved mucosal tissues (P = 0.12 to 0.28) but were higher than those of cancer tissue (P < 0.05). · The high ratios in the body mucosa of the noncancer patients indicate that these tissues rely substantially on aerobic pathways for energy production. The dominant role of the Krebs cycle in these tissues is further emphasized when it is considered that the ratios in table 2 are low, possibly minimum estimates of aerobic, as compared to anaerobic energy production. This follows from the fact that oxidations of endogenous substrates were excluded from the estimates of aerobic pathways but were included in the measurements of · anaerobic metabolic pathways. Similar ratios were obtained from duodenal ulcer and gastric ulcer uninvolved body mucosa (P = 0. 72). This contrasts with the difference in Krebs cycle activity observed between these two tissues (fig. 1). Although
1324 TABLE
ORWELL AND PIPER
Vol. 73, No . 6
2. Ratios illustrating the relative importance of aerobic and anaerobic pathways for energy production in gastric tissues from duodenal ulcer (DU), gastric ulcer (GU), and gastric cancer (GCa) patients; means ± SE Tissue type
GCa
GU
GCa
Gastric cancer tissue
Tissue 3
Tissue 4
Tissue 5
Tissue 6
15
13
14
10
0.45 ± 0.10
0.51 ± 0.13
0.26 ± 0.08
0.14 ± 0.05
0.15 ± 0.06
8 0.05 ± 0.014
8.61 ± 1.88
9.61 ± 2.43
4.96 ± 1.44
2.66 ± 0.86
2.83 ± 1.05
0.91 ± 0.27
Uninvolved antral mucosa
Uninvolved body mucosa
No. ofpatients A = (moles of glucose metabolized via Krebs cycle)/(moles of glucose equivalent to lactate formed) B = (ATP equivalent of Krebs cycle activity)/(ATP equivalent of lactate production)
DU
GU
Tissue 1
Tissue 2
12
the tissue from the duodenal ulcer patients possessed the higher Krebs cycle activity, the ratios (table 2) suggest that there is little qualitative difference between these tissues regarding the balance between utilization of aerobic and anaerobic pathways. Nevertheless, there are marked quantitative differences, notably in glucose utilization, evolution of C02 , and lactate formation, as can be seen in table 1. The lower ratios seen in antral mucosa and cancer tissue indicate that these tissues rely much more than body mucosa on anaerobic pathways for energy production. It is interesting to note that in cancer tissue ratio B was less than unity, suggesting that in this tissue, alone of those studied, more energy may be derived from anaerobic than from aerobic pathways.
activity resembles that found in Ehrlich and Krebs ascites tumor cells, 16• 17 but was much less than the activities found in rat small bowel mucosa, 18 and in lipogenic tissues such as rat epididymal adipose tissue' 9 and rat lactating mammary gland.20 It was also smaller in the human tissues studied than in rat gastric mucosa, as reported by Sernka and Harris. 21 This difference was unexpected, inasmuch as the latter authors stressed the role of the pentose phosphate pathway and suggested that it provided most of the energy required for hydrochloric acid secretion by rat gastric mucosa. The difference may be partly attributable to the fact that Sernka and Harris2 1 used a supporting medium containing 25 mM glucose and 1 mM phosphate which stimulated acid secretion, whereas in the present work 5 mM glucose and 11.2 mM phosphate Discussion were used. Nevertheless, there appears to be a signifiTwo contrasting metabolic patterns were seen in the cant species difference between rat and human gastric tissues under investigation. The first, which was most mucosa regarding the contribution of the shunt to pronounced in the uninvolved body mucosa of duodenal glucose metabolism. Among the tissues studied, gastric and gastric ulcer patients, was characterized by a very cancer tissue alone gave a pentose phosphate pathway active tricarboxylic acid cycle, relatively low or moder- contribution which exceeded 1%. ate lactate production, few [I 4 C]glucose counts retained Two tissues warrant special mention. Firstly, of all by the tissue, and a noticeable deficit in the glucose the tissues studied, duodenal ulcer patients' body muwhich could be accounted for in the experiments. This cosa possessed the highest rate of glucose utilization pattern suggests that these tissues are mainly aerobic and the highest tricarboxylic acid cycle activity, both in function and that glucose carbon is utilized more in in absolute terms and when expressed as a proportion energy producing than in synthetic pathways. These of the glucose utilized. It also revealed the smallest findings agree with those of Menguy et al., 15 who proportion of [14 C]glucose counts retained by the tissue. stressed the aerobic character of energy production in It was clearly the most energetic and metabolically rat glandular gastric mucosa. active of all the tissues studied. These features could be The second pattern· was evident in the uninvolved expected, given the known hydrochloric acid secreting antral mucosa and most noticeably in gastric cancer function of body, as compared to antral mucosa, and tissue. The tricarboxylic acid cycle, lactate production, the high gastric acid outputs which are characteristic of duodenal ulcer patients. 22 , 23 [ 14C]glucose counts retained by the tissue, and the deficit all showed tendencies opposite to those described The second noteworthy tissue was the uninvolved above, indicating a relatively anaerobic metabolism, body mucosa of the cancer patients. Compared to the diminished energy production, and the possibility of uninvolved body mucosa of the noncancer patients, this increased use of glucose carbon in synthetic pathways. tissue was deficient in tricarboxylic acid cycle activity The deficit was small and even became a surplus in and showed tendencies toward higher rates of lactate gastric cancer tissue, forcing one to conclude that in production and higher levels of [14 C]glucose counts this tissue, endogenous as well as exogenous substrates retained. It resembled uninvolved antral mucosa in all these respects. These abnormalities provide biochemical were mobilized by the tissue in vitro. Less than 1% of the glucose taken up by the unin- evidence of antralization of the body mucosa in gastric volved gastric mucosal tissues studied was metabolized cancer. They point to deficient energy production in via the pentose phosphate pathway. Thif? level of shunt this tissue and may play a role in the gastric hydrochlo-
December 1977
GLUCOSE OXIDATION IN STOMACH MUCOSA
ric acid hyposecretion frequently found in gastric cancer subjects. 22• 23 However, the changes in lactate production seen in cancer patients' uninvolved body mucosa appear to be too small to account for the in vivo lactate hypersecretion seen in gastric cancer patients. 24 A more likely explanation of this secretory abnormality is provided by the exceptionally high rates of lactate production maintained by tumor tissue, particularly when compared to those of uninvolved mucosa from the cancer patients. On this basis, the lactate hypersecretion is probably caused by the excessive glycolysis of the tumor and the degree of lactic acidosis would vary more or less directly with the size of the tumor. This is in agreement with the present authors' previous findings. 25 In gastric cancer, the glucose content of the mucus fraction of the gastric juice is increased.26• 27 A finding consistent with this observation is the shift, in the uninvolved body mucosa of the cancer patients, from oxidative pathways to those leading to the incorporation of glucose into the tissue. The excess contribution the mucosa must make to gastric mucus formation in this disease would partially account for these changes in the metabolic pattern. The gastric mucosa is a mixture of several groups of cells: parietal cells, mucus-secreting cells, chief cells, and cells of intestinal metaplasia. The gastric mucus cells are quantitatively the most important, although probably not as metabolically active as the parietal cells, which secrete hydrogen ions against an enormous concentration gradient. The observations in this study correlate well with expectation based on parietal and chief cell function, the fundic mucosa of duodenal ulcer patients being exceptional among the tissues studied in glucose utilization and in the activity of the energetically important Krebs cycle. Nevertheless, here our expectations are extrapolations from data based on acid and pepsin secretion, and we have no evidence that the metabolic activity of the parietal and chief cells dominates the total mucosal metabolic activity. Clearly, the striking metabolic differences observed in this study justify further work on the above groups of patients using centrifugal fractionation of the cells comprising the tissue. Only in this way can a clearer interpretation of the data presented be achieved. REFERENCES 1. Oi M, Oshida K, Sugimura S: The location of gastric ulcer. Gastroenterology 36:45-56, 1959 2. Card WI, Marks IN: The relationship between acid output of the stomach following "maximal" histamine stimulation and parietal cell mass. Clin Sci 19:147-163, 1960 3. Taylor KB, Fischer JM: Gastritis. Prog Gastroenterol 1:1-21, 1968 4. Rohrer GU, Welsh JD: Correlative study: gastric secretion and histology. Gastroenterology 52:185-191, 1967 5. Morson BC: Carcinoma arising from areas of intestinal metaplasia in the gastric mucosa. Br J Cancer 9:377-385, 1955 6. Prochaska B, Jirasek V, Barta V, et al: Differences in lactate dehydrogenase isoenzyme patterns in various parts of the human stomach and in peptic ulcer and gastric carcinoma. Gastroenterology 54:65-71, 1968
1325
7. Woollams R, Barratt PJ, Orwell RL, et al: LDH isoenzyme patterns of uninvolved gastric mucosa of patients with gastric carcinoma and benign gastric disease. Digestion 14:20-28, 1976 8. Umbreit WW, Burris RH , Stauffer JF: Manometric techniques and tissue metabolism. Second edition. Minneapolis, Burgess Publishing Co, p 119, 1949 9. Hansen DL, Bush ET: Improved solubilization procedures for liquid scintillation counting of biological materials. Anal Biochem 18:320-332, 1967 10. Lowry OH, Passonneau JV, Hasselberger FX, et al: Effect of ischaemia on known substrates and cofactors of the glycolytic pathway in brain. J Bioi Chern 239:18-30, 1964 11. Hohorst HJ: L-+-Lactate determination with lactate dehydrogenase and DPN. In Methods of Enzymatic Analysis. Edited by HV Bergmeyer. New York, London, Academic Press, p 266, 1962 12. Husain S, Paradise RR: Pitfalls in measurement of 14 C0 2 activity from glucose-6- 14 C and two corrective procedures. Exp Bioi Med 142:316-320, 1973 13. Katz J, Wood HG: The use of 14 C02 yields from glucose-1- and6- 14 C for the evaluation of the pathways of glucose metabolism. J Bioi Chern 238:517-523, 1963 14. Conn EE, StumpfPK: Electron transport and oxidativephosphorylation. In Outlines of Biochemistry. Second edition. New York, John Wiley & Sons, p 271, 1967 15. Menguy R, Desbaillets L, Masters YF: Mechanism of stress ulcer: influence of hypovolemic shock on energy metabolism in the gastric mucosa. Gastroenterology 66:46-55, 1974 16. Brand K, Deckner K: Quantitative relationship between the pentose phosphate pathway and the nucleotide synthesis in ascites tumor cells. Hoppe-Seylers Z Physiol Chern 351:711-717, 1970 17. Gumaa KA, McLean P: The pentose phosphate pathway of glucose metabolism; enzyme profiles and transient and steadystate content of intermediates of alternative pathways of glucose metabolism in Krebs ascites cells. Biochem J 115:1009-1029, 1969 18. Nakayama H, Weser E: Adaptation of small bowel after intestinal resection: increase in the pentose phosphate pathway. Biochim Biophys Acta 279:416-423, 1972 19. Kather H, Rivera M, Brand K: Interrelationship and control of glucose metabolism and lipogenesis in isolated fat cells. Biochem J 128:1089-1096, 1972 20. Katz J, Wals PA: Pentose cycle and reducing equivalents in rat mammary gland slices. Biochem J 128:879-899, 1972 21. Sernka TJ, Harris JB: Pentose phosphate shunt and gastric acid secretion in the rat. Am J Physiol 222:25-32, 1972 22 . Marks IN: The augmented histamine test. Gastroenterology 41:599-603, 1961 23. Grossman MI, Kirsner JB, Gillespie IE: Basal ·and histalogstimulated gastric secretion in control subjects and in patients . with peptic ulcer or gastric cancer. Gastroenterology 45:14-26, 1963 24. Piper DW, Kemp ML, Fenton BH, et al: Gastric juice lactic acidosis in the presence of gastric carcinoma. Gastroenterology 58:766-771, 1970 25. Kasugai T, Orwell RL, Clarke AD, et al: Gastric juice lactic acidosis: the incidence of gastric juice lactic acidosis in gastric cancer and benign gastric disease in communities of high and low gastric cancer incidence. Am J Dig Dis 19:599-602, 1974 26. Schrager J: Chromatographic studies of ·carbohydrate components of gastric and salivary mucopolysaccharides. Gut 5:166169, 1964 27. Piper DW, Griffith EM, Fenton BH: Gastric mucus secretion in normal subjects and patients with gastric ulcer, gastric carcinoma and pernicious anaemia. Am J Dig Dis 10:411-418, 1965
Vol. 73, No. 6 Printed in U.S A .
KREBS CYCLE, PENTOSE PHOSPHATE PATHWAY, AND GLYCOLYSIS IN THE UNINVOLVED GASTRIC MUCOSA OF PEPTIC ULCER AND GASTRIC CANCER PATIENTS RALPH L. ORWELL, PH.D. , AND DOUGLAS w. PIPER, M.D., F.R.C.P. , F.R.A.C.P. The Department of Medicine, University of Sydney at the R oyal North Shore Hospital , Sydney, New South Wales, Australia
Uninvolved gastric mucosa from duodenal ulcer, gastric ulcer, and gastric cancer patients was incubated with [l-14C]glucose and [6- 14C]glucose in order to assess the relative contributions of the pentose phosphate pathway and Krebs cycle to glucose metabolism. [14C]Glucose counts retained by the tissue, glycolysis, and pyruvate formation were also measured. Tumor tissue from the cancer patients was included in the study. Less than 1.2% of the glucose entering the tissues was metabolized via the pentose phosphate pathway, suggesting that this pathway plays a minor role in energy production from glucose. The major determinant of energy production was the Krebs cycle. Its contribution to glucose rr. 3tabolism was greatest in the body mucosa of duodenal ulcer patients, less in the uninvolved body mucosa of gastric ulcer patients, and lower still in the corresponding body mucosa of gastric cancer patients. The low levels of Krebs cycle activity seen in the latter tissue resembled those of uninvolved antral mucosa. The smallest Krebs cycle contribution was seen in tumor tissue. [ 14C]Glucose counts retained by the tissue and glycolysis both tended to vary inversely with Krebs cycle activity among the tissues studied. Thus, both were small in the body mucosa of noncancer patients and somewhat larger in the body mucosa of cancer patients, in uninvolved antral mucosa and in tumor tissue. Disease states in gastric mucosa have been related to several criteria, for example, the extent to which body and antral mucosa are involved, 1 the number of parietal cells present, 2 and the extent of gastritis and intestinal metaplasia.3-5 More recently, evidence has been presented that the processes involved in glucose metabolism may be a relevant variable. 6 • 7 In the present study, the body and antral mucosa of chronic peptic ulcer patients, together with the body mucosa, antral mucosa, and tumor tissue of gastric cancer patients,
were compared in in vitro experiments as regards rates of glucose utilization, the relative contributions to glucose metabolism of the pentose phosphate pathway and Krebs cycle, and the rates of glycolysis within the tissues. Methods
Tissue procurement. Samples of gastric t issue were obtained from gastric ulcer and gastric cancer patients undergoing partial gastrectomy. These were sampled for uninvolved body and antral mucosa at least 3.0 em from any lesion. With Received April4, 1977. Accepted June 29, 1977. duodenal ulcer patients, the surgical procedure (vagotomy) limited sampling to uninvolved body mucosa; the latter was Address requests for reprints to: Professor D. W. Piper, Department of Medicine, Royal North Shore Hospital of Sydney, St. resected before the vagotomy step. Gastric cancer tissue was Leonards, New South Wales 2065, Australia. obtained from the outer sections of tumors, avoiding necrotic This study was supported by grants-in-aid from the National areas. Mucosa representative of every sample was preserved for Health and Medical Research Council of Australia, the New South Wales State Cancer Council, and the University of Sydney Cancer histological examination; the pathologist's findings were used to classify samples as body or antral mucosa, and to ensure Research Fund. Dr. Orwell is Research Officer, National Health and Medical that uninvolved mucosa from cancer patients was free from Research Council of Australia. His present address is: School of Life malignant cells. The cancer patients all possessed histologiSciences, The New South Wales Institute of Technology, West- cally confirmed adenocarcinomas. Biochemical methods . Gastric mucosa was placed, within bourne Street, Gore Hill, New South Wales 2065, Australia. The authors acknowledge with thanks the expert assistance of 20 min of resection, in ice-cold Krebs-Ringer phosphate methe staff of the Department of Pathology, Royal North Shore Hospi- dium, pH 7.4,8 and dissected into intact mucosal segments of tal, Sydney, who performed the histological examinations, and the area approximately 4 mm. 2 Tumor tissue was sliced into cooperation of our surgical colleagues, Dr. V. H. Cumberland, Dr. segments of similar dimensions. Portions (0.1 g) of prepared tissue were preincubated 15 min in Warburg flasks without R. M. Hollings, and Associate Professor G. A. E. Coupland in labeled substrate and then incubated for 3 hr at 37°C with providing gastric tissue. 1320
December 1977
1321
GLUCOSE OXIDATION IN STOMACH MUCOSA
agitation in 2.0 ml of the same medium containing 5.0 mM unlabeled glucose and 0.20 1-LCi of 14C-labeled glucose per flask. All flasks were gassed with 100% 0 2 ; the center wells contained filter paper wicks impregnated with 0.1 ml of 10% KOH (w/v). Incubation was stopped by chilling flasks to 4°C and adding 0.50 ml of 1 M perchloric acid via the sidearm. After 1.5 hr with occasional agitation to collect 14 C0 2 , the suspension was centrifuged for 10 min at 600 x g and the pellet was washed by resuspending in 1.0 ml of water and recentrifuging. The pellet was then dried over P 2 0 5 , weighed, rehydrated with 0.3 ml of water, and finally digested in 2.0 ml of 1 M hyamine hydroxide in methanol (50°C, 1 hr). 9 A 0.40-ml aliquot was analyzed for radioactivity. The supernatant and washing were combined, neutralized with 0.60 ml of 1.0 M potassium bicarbonate and, after removal of potassium perchlorate, stored at -30°C. Glucose and pyruvate were assayed in extracts from duplicate flasks by the methods of Lowry et al. 10 Lactate was similarly assayed using Hohorst's method. 11 The paper wicks, together with 0.10 ml of water used to rinse each center well, were transferred to counting vials containing 0.40 g of Cab-O-SiP 2 and 10 ml of scintillation fluid (naphthalene 120 g; 2,5-diphenyloxazole (PPO) 4.0 g; 1,4-bis[2-(5-phenyloxazolyl)]benzene (POPOP) 0.05 g; toluene 400 ml; 1,4-dioxan 300 ml; 2-methoxyethanol300 ml). Counting was performed in a Packard Tri-Carb model 3375 counter with 45 to 50% efficiency. Hyamine hydroxide digests were counted in the same system without Cab-0-Sil at 50 to 55% efficiency. In all cases quenching was corrected for by means of external standards. Net changes in glucose, lactate, and pyruvate were obtained by subtracting the values given by duplicate tissue samples which received only the preincubation step. Yields of 14 C0 2 from [1- 14 C]glucose and [6- 14 C]glucose were used firstly to calculate the pentose phosphate pathway contributions to glucose metabolism, using specific 14 C0 2 yields as defined by Katz and Wood. 13 Secondly, the 14 C0 2 yield from [6- 14 C]glucose was used as a direct measure of over-all Krebs cycle activity, although it is more accurately an estimate of glucose carbon entering the cycle. Figures given for 14 C label retained by the tissue refer to experiments in which [6- 14 C]glucose was used. All results were calculated on the basis of the dry weight of tissue recovered after the experiment and expressed as micromoles per gram of dry tissue per hour. These data were further analyzed by expressing the glucose equivalent of each metabolite as a percentage of the glucose utilized.
Reagents and chemicals . [V 4 C]Glucose and [6- 14 C]glucose were obtained from The Radiochemical Centre, Amersham, United Kingdom. PPO, POPOP, and Cab-0-Sil were purchased from Consolidated Nucleonics, Sydney, Australia, and hyamine hydroxide was from Calbiochem (Aust.) Ltd., Sydney, Australia. Enzymes, cofactors, and substrates were from Boehringer Mannheim (Aust. ) Ltd., Melbourne, Australia. All other reagents were analytical reagent grade. Statistical analysis. Student's t-test was used for all comparisons.
Results The results for the six types of tissue studied will first be expressed in terms of the molar quantities of glucose utilized and metabolites formed, and secondly, in terms of the proportions of the glucose metabolized by various pathways. Finally, data on the relative importance of aerobic and anaerobic pathways for energy production in the mucosa is presented.
Molar Quantities of Glucose Utilized and Metabolites Formed in Gastric Mucosa and Tumor Tissue These data are given in table 1, expressed as micromoles per gram of dry tissue per hour. Noteworthy features are: Glucose utilization. The body mucosa of duodenal ulcer patients utilized glucose more rapidly than any of . the other tissues studied (table 1; P < 0.05). Among the latter, similar rates of glucose utilization were observed (P = 0.15 to 0.89). Lactate and pyruvate production. Lactate production was higher in duodenal ulcer patients' body mucosa than in gastric ulcer patients' body (P < 0.005) and antral (P < 0.02) mucosa, and in gastric cancer patients' body mucosa (P < 0.02) (table 1). It was also higher in · tumor tissue than in gastric ulcer group body mucosa (P < 0.01), gastric ulcer antral mucosa (P < 0.05), and in cancer group body mucosa (P < 0.03). The cancer patients' antral mucosa gave a relatively high level of lactate formation, which exceeded that found in gastric ulcer patients' body mucosa (P < 0.05). Apart from these differences, similar amounts of lactate were formed by the tissues studied (P = 0.12 to 0.80).
TABLE 1. Utilization of 14C-labeled glucose by segments of gastric mucosa and slices of gastric cancer tissue from duodenal ulcer (DUJ, gastric ulcer (GUJ, and gastric cancer (GCaJ patients; means ± BE. Segments or slices (1 00 mg) were incubated for 3 hr at 3 7oC in 2 .0 ml of Krebs-Ringer phosphate medium containing 5 .0 mM glucose and 0 .20 JLCi of {1-'
Uninvolved antral mucosa
Uninvolved body mucosa
No. ofpatients Glucose utilized C0 2 from C-1 [1- 14 C]glucose C0 2 from C-6 [6-' 4 C]glucose Lactate" Pyruvate" C-6 retained by tissue [6-' 4 C]glucose) a
DU
GU
GCa
GU
GCa
Gastric cancer tissue
Tissue 1
Tissue 2
Tissue 3
Tissue 4
Tissue 5
Tissue 6
12 ± ± ± ± ± ±
15 46.0 ± 3.6 6.29 ± 0.81 5.91 ± 0.88 38.6 ± 7.1 0.11 ± 0.05 7.94 ± 0.87
13 43.1 ± 4.2 4.48 ± 0.90 3.75 ± 0.77 44.6 ± 7.6 O.o7 ± 0.03 7.78 ± 0.92
72.9 13.63 13.07 81.3 0.22 6.48
7.4 1.56 1.60 11.6 O.o7 0.74
Includes contribution from endogenous substrates. _
14 39.0 ± 2.41 ± 1.92 ± 47.9 ± 0.30 ± 10.21 ±
3.7 0.47 0.37 6.9 0.18 1.27
10 45.2 ± 4.1 3.58 ± 0.62 2.73± 0.54 67.8 ± 13.1 0.09 ± 0.04 10.38 ± 1.15
47.1 3.06 1.78 77.0 0.29 7.56
8 ± 8.0 ± 0.69 ± 0.56 ± 11.3 ± 0.12 ± 1.43
1322
Vol . 73, No.6
ORWELL AND PIPER
Duodenal ulcer patients' body mucosa and tumor tissue both yielded significantly more pyruvate than cancer patients' body mucosa (P = 0.05). The high pyruvate yields in the first-mentioned tissues parallel the lactate results and point to the coupling of lactate and pyruvate levels in the tissue via lactate dehydrogenase. A high level of pyruvate production was also seen in gastric ulcer group antral mucosa, although this result was not significantly different from values obtained for the other tissues (P = 0.22 to 0.94). Elsewhere, no differences in pyruvate formation were seen (P = 0.12 to 0.94).
C0 2 production from C-1 and C-6 of glucose. C0 2 production from carbon 1 of glucose, which reflects glucose oxidation in both the pentose phosphate pathway and Krebs cycle, 13 was much higher in duodenal ulcer patients' body mucosa than in any other tissue (P < 0.001) (table 1). It was also higher in the uninvolved body mucosa of gastric ulcer patients than in the uninvolved antral mucosa of both gastric ulcer (P < 0.001) and gastric cancer patients (P < 0.03) and in gastric cancer tissue (P < 0.02). Similar levels of C0 2 production from carbon 1 were seen in the uninvolved body mucosa of the gastric ulcer and cancer patients (P = 0.15). Finally, the value given by cancer group body mucosa was greater than that of gastric ulcer group antral mucosa (P < 0.05). None of the remammg differences observed between tissues were significant (P = 0.15 to 0.59). C0 2 production from carbon 6 of glucose, used here as a direct measure of Krebs cycle activity, 13 was very high in body mucosa of duodenal ulcer patients, which gave a value greater than that of any of the remaining tissues (P < 0.001) (table 1). The body mucosa of gastric
I
GLUCOSE (PERCENT
ulcer patients was next in activity with a value which exceeded those of antral mucosa from the same patients (P < 0.001), antral mucosa from cancer patients (P < 0.02), and tumor tissue (P < 0.005). Furthermore, glucose carbon 6 oxidation was greater in the body mucosa of gastric ulcer, as compared to gastric cancer patients by a margin which closely approached significance (P = 0.07). In turn, cancer group body mucosa gave a value which exceeded that of gastric ulcer patients' antral mucosa (P < 0.05). No further differences were seen among the tissues compared in table 1 (P = 0.09 to 0.82). Retention of glucose C-6 by the tissues. These results (table 1) provide a measure of the extent to which glucose carbon has been diverted into synthetic pathways such as protein synthesis and proteoglycan (mucus) formation. They may also include a contribution from low molecular weight, perchloric acid-soluble compounds not completely removed from the tissue during the washing steps. Duodenal ulcer patients' body mucosa retained less glucose carbon 6 than the uninvolved antral mucosa of both gastric ulcer (P < 0.03) and gastric cancer (P < 0.01) patients. Otherwise, the tissues were similar with respect to this parameter (P = 0.09 to 0.92).
Proportions of Glucose Metabolized via Various Pathways in Gastric Mucosa and Tumor Tissue These results are illustrated in figure 1, where all values are expressed as percentages of the glucose taken up from the medium by each tissue. It is emphasized that the values given for the pentose phosphate pathway, Krebs cycle, and glucose C-6 retained by the tissues refer to the fate of exogenous glucose, whereas
MfTA.IOUSM IN GASTRIC MUCQ$A
OF GLUCOSE
TAKEN
UP &Y EACH
TISSUE:
MEANS
t
SE )
~~
FIG. 1. Relative contributions to carbohydrate metabolism of the pentose phosphate pathway (PP), the tricarboxylic acid (TCA) cycle, incorporation into the tissue (Tissue) and glycolysis (Lactate and ~yruvate): Uninvol~ed gastric .mucosa from. du.odenal. ulcer (D.U.), gastric ulcer (G.U .) and gastric cancer (G.Ca.) patients, together With gastnc cancer tissue were mcubated as md1cated m table 1. All figures are expressed as percentages of the glucose utilized by the tissues. Means± SE. See Results for method used to calculate Deficit term.
December 1977
GLUCOSE OXIDATION IN STOMACH MUCOSA
those given for lactate and pyruvate represent products formed from exogenous glucose and endogenous substrates. Consequently, it was possible for the sum of the quantities tabulated in figure 1 to exceed 100% and this sum is clearly of no value in calculating the experimental recoveries of exogenous glucose. Nevertheless, because of its usefulness, a deficit term was calculated to represent glucose taken up but which could not be accounted for in terms of the other parameters (fig. 1). Only cancer tissue gave a sum exceeding 100%, so that in this case a surplus, rather than a deficit, was recorded. Pentose phosphate pathway . This pathway made only a small contribution to glucose metabolism, equivalent to 0.5 to 1.2% of glucose utilized (Fig. 1). Its percentage contribution was greater in tumor tissue than in the uninvolved antral mucosa of gastric ulcer patients (P = 0.05), but was otherwise similar among the six tissues studied (P = 0.18 to 0.95). Glycolytic pathway. The percentage corresponding to lactate formation (fig. 1) tended to be lower in body mucosa than in antral mucosa and to be especially high in cancer tissue. The lactate value for cancer tissue exceeded that of body mucosa from both gastric ulcer (P < 0.005) and gastric cancer (P < 0.05) patients. The lactate value was also higher in cancer patients' antral mucosa than in gastric ulcer body mucosa (P < 0.05). Apart from these differences, the percentage equivalent to lactate formation was similar among the tissues studied (P = 0.09 to 0.92). The corresponding figures for pyruvate formation were similar for all six types of tissue (P = 0.10 to 0.85) (fig. 1). Tricarboxylic acid cycle. The percentage contribution of the Krebs cycle to glucose metabolism was greater in duodenal ulcer group body mucosa than in any of the other five tissues (P < 0.05) (fig. 1) . Gastric ulcer patients' body mucosa was next in activity with a value exceeding that of any of the four remaining tissues (P < 0.05). Considering only uninvolved body mucosa, the tricarboxylic acid cycle contributions discriminated between the three patient groups, ranking them in order of decreasing activity as follows (P < 0.05): duodenal ulcer; gastric ulcer; gastric cancer. In gastric ulcer patients, the relative importance of the Krebs cycle was significantly greater in the body than in the antral mucosa (fig. 1) (P < 0.001). In cancer patients, on the other hand, this contrast was not apparent. The body mucosa of these patients gave a relatively small Krebs cycle contribution which, although greater than that of gastric cancer tissue (P < 0.05), resembled that of antral mucosa from either gastric ulcer (P = 0.08) or gastric cancer (P = 0.30) patients. The smallest Krebs cycle contribution to glucose metabolism was seen in tumor tissue (fig. 1). It yielded a value which was less than that of body mucosa from any of the three patient groups (P < 0.05), but which resembled those of antral mucosa from the gastric ulcer (P = 0.30) and cancer (P = 0.18) patients. Pathways leading to incorporation of glucose into gastric tissues. The glucose fraction retained by the
1323
tissue pellet (fig. 1) was relatively small in those tissues showing appreciable tricarboxylic acid cycle activity. Thus it was smaller in duodenal ulcer patients' body mucosa than in any of the other tissues (P < 0.02) . It was also smaller in gastric ulcer group body mucosa than in cancer group antral mucosa by a margin which was close to significance (P = 0.07). Conversely, the glucose fraction retained by the pellet was high in those tissues showing low Krebs cycle activity. This was evident in the similar, relatively high percentages recorded for gastric ulcer antral mucosa, gastric cancer body, and antral mucosa and cancer tissue in figure 1 (P = 0.42 to 0.98). This tendency toward an inverse relationship between Krebs cycle activity and [6- 14 C]glucose counts incorporated into the tissue suggests that in gastric tissues there may be competition for glucose between oxidative and synthetic pathways. Deficit. The deficit calculated in figure 1 was influenced most by the largest of the measured parameters, lactate formation. It was smallest in those tissues where lactate formation was high, notably in antral mucosa and cancer tissue. It is possible that in these tissues, substantial amounts of endogenous substrates, rather than exogenous glucose were utilized for energy production during the experiment. The largest deficit was seen in gastric ulcer group body mucosa which gave a higher value than cancer group antral mucosa (P < 0.04) or cancer tissue (P < 0.02).
Relative Importance of Aerobic and Anaerobic Pathways Table 2 gives ratios chosen to illustrate the relative importance of the major aerobic and anaerobic pathways in each tissue. Ratio A was derived from the data summarized in figure 1 and used to calculate ratio B on the basis that the complete oxidation of 1 mole of glucose is capable of yielding the equivalent of 38 moles of ATP, whereas its conversion to lactate yields only 2 moles of ATP. 14 Higher ratios were seen in body mucosa from duodenal ulcer and gastric ulcer patients than ih the two types of antral mucosa or in cancer tissue (table 2; P < 0.05). The ratios given by cancer patients' body mucosa resembled those of the other uninvolved mucosal tissues (P = 0.12 to 0.28) but were higher than those of cancer tissue (P < 0.05). · The high ratios in the body mucosa of the noncancer patients indicate that these tissues rely substantially on aerobic pathways for energy production. The dominant role of the Krebs cycle in these tissues is further emphasized when it is considered that the ratios in table 2 are low, possibly minimum estimates of aerobic, as compared to anaerobic energy production. This follows from the fact that oxidations of endogenous substrates were excluded from the estimates of aerobic pathways but were included in the measurements of · anaerobic metabolic pathways. Similar ratios were obtained from duodenal ulcer and gastric ulcer uninvolved body mucosa (P = 0. 72). This contrasts with the difference in Krebs cycle activity observed between these two tissues (fig. 1). Although
1324 TABLE
ORWELL AND PIPER
Vol. 73, No . 6
2. Ratios illustrating the relative importance of aerobic and anaerobic pathways for energy production in gastric tissues from duodenal ulcer (DU), gastric ulcer (GU), and gastric cancer (GCa) patients; means ± SE Tissue type
GCa
GU
GCa
Gastric cancer tissue
Tissue 3
Tissue 4
Tissue 5
Tissue 6
15
13
14
10
0.45 ± 0.10
0.51 ± 0.13
0.26 ± 0.08
0.14 ± 0.05
0.15 ± 0.06
8 0.05 ± 0.014
8.61 ± 1.88
9.61 ± 2.43
4.96 ± 1.44
2.66 ± 0.86
2.83 ± 1.05
0.91 ± 0.27
Uninvolved antral mucosa
Uninvolved body mucosa
No. ofpatients A = (moles of glucose metabolized via Krebs cycle)/(moles of glucose equivalent to lactate formed) B = (ATP equivalent of Krebs cycle activity)/(ATP equivalent of lactate production)
DU
GU
Tissue 1
Tissue 2
12
the tissue from the duodenal ulcer patients possessed the higher Krebs cycle activity, the ratios (table 2) suggest that there is little qualitative difference between these tissues regarding the balance between utilization of aerobic and anaerobic pathways. Nevertheless, there are marked quantitative differences, notably in glucose utilization, evolution of C02 , and lactate formation, as can be seen in table 1. The lower ratios seen in antral mucosa and cancer tissue indicate that these tissues rely much more than body mucosa on anaerobic pathways for energy production. It is interesting to note that in cancer tissue ratio B was less than unity, suggesting that in this tissue, alone of those studied, more energy may be derived from anaerobic than from aerobic pathways.
activity resembles that found in Ehrlich and Krebs ascites tumor cells, 16• 17 but was much less than the activities found in rat small bowel mucosa, 18 and in lipogenic tissues such as rat epididymal adipose tissue' 9 and rat lactating mammary gland.20 It was also smaller in the human tissues studied than in rat gastric mucosa, as reported by Sernka and Harris. 21 This difference was unexpected, inasmuch as the latter authors stressed the role of the pentose phosphate pathway and suggested that it provided most of the energy required for hydrochloric acid secretion by rat gastric mucosa. The difference may be partly attributable to the fact that Sernka and Harris2 1 used a supporting medium containing 25 mM glucose and 1 mM phosphate which stimulated acid secretion, whereas in the present work 5 mM glucose and 11.2 mM phosphate Discussion were used. Nevertheless, there appears to be a signifiTwo contrasting metabolic patterns were seen in the cant species difference between rat and human gastric tissues under investigation. The first, which was most mucosa regarding the contribution of the shunt to pronounced in the uninvolved body mucosa of duodenal glucose metabolism. Among the tissues studied, gastric and gastric ulcer patients, was characterized by a very cancer tissue alone gave a pentose phosphate pathway active tricarboxylic acid cycle, relatively low or moder- contribution which exceeded 1%. ate lactate production, few [I 4 C]glucose counts retained Two tissues warrant special mention. Firstly, of all by the tissue, and a noticeable deficit in the glucose the tissues studied, duodenal ulcer patients' body muwhich could be accounted for in the experiments. This cosa possessed the highest rate of glucose utilization pattern suggests that these tissues are mainly aerobic and the highest tricarboxylic acid cycle activity, both in function and that glucose carbon is utilized more in in absolute terms and when expressed as a proportion energy producing than in synthetic pathways. These of the glucose utilized. It also revealed the smallest findings agree with those of Menguy et al., 15 who proportion of [14 C]glucose counts retained by the tissue. stressed the aerobic character of energy production in It was clearly the most energetic and metabolically rat glandular gastric mucosa. active of all the tissues studied. These features could be The second pattern· was evident in the uninvolved expected, given the known hydrochloric acid secreting antral mucosa and most noticeably in gastric cancer function of body, as compared to antral mucosa, and tissue. The tricarboxylic acid cycle, lactate production, the high gastric acid outputs which are characteristic of duodenal ulcer patients. 22 , 23 [ 14C]glucose counts retained by the tissue, and the deficit all showed tendencies opposite to those described The second noteworthy tissue was the uninvolved above, indicating a relatively anaerobic metabolism, body mucosa of the cancer patients. Compared to the diminished energy production, and the possibility of uninvolved body mucosa of the noncancer patients, this increased use of glucose carbon in synthetic pathways. tissue was deficient in tricarboxylic acid cycle activity The deficit was small and even became a surplus in and showed tendencies toward higher rates of lactate gastric cancer tissue, forcing one to conclude that in production and higher levels of [14 C]glucose counts this tissue, endogenous as well as exogenous substrates retained. It resembled uninvolved antral mucosa in all these respects. These abnormalities provide biochemical were mobilized by the tissue in vitro. Less than 1% of the glucose taken up by the unin- evidence of antralization of the body mucosa in gastric volved gastric mucosal tissues studied was metabolized cancer. They point to deficient energy production in via the pentose phosphate pathway. Thif? level of shunt this tissue and may play a role in the gastric hydrochlo-
December 1977
GLUCOSE OXIDATION IN STOMACH MUCOSA
ric acid hyposecretion frequently found in gastric cancer subjects. 22• 23 However, the changes in lactate production seen in cancer patients' uninvolved body mucosa appear to be too small to account for the in vivo lactate hypersecretion seen in gastric cancer patients. 24 A more likely explanation of this secretory abnormality is provided by the exceptionally high rates of lactate production maintained by tumor tissue, particularly when compared to those of uninvolved mucosa from the cancer patients. On this basis, the lactate hypersecretion is probably caused by the excessive glycolysis of the tumor and the degree of lactic acidosis would vary more or less directly with the size of the tumor. This is in agreement with the present authors' previous findings. 25 In gastric cancer, the glucose content of the mucus fraction of the gastric juice is increased.26• 27 A finding consistent with this observation is the shift, in the uninvolved body mucosa of the cancer patients, from oxidative pathways to those leading to the incorporation of glucose into the tissue. The excess contribution the mucosa must make to gastric mucus formation in this disease would partially account for these changes in the metabolic pattern. The gastric mucosa is a mixture of several groups of cells: parietal cells, mucus-secreting cells, chief cells, and cells of intestinal metaplasia. The gastric mucus cells are quantitatively the most important, although probably not as metabolically active as the parietal cells, which secrete hydrogen ions against an enormous concentration gradient. The observations in this study correlate well with expectation based on parietal and chief cell function, the fundic mucosa of duodenal ulcer patients being exceptional among the tissues studied in glucose utilization and in the activity of the energetically important Krebs cycle. Nevertheless, here our expectations are extrapolations from data based on acid and pepsin secretion, and we have no evidence that the metabolic activity of the parietal and chief cells dominates the total mucosal metabolic activity. Clearly, the striking metabolic differences observed in this study justify further work on the above groups of patients using centrifugal fractionation of the cells comprising the tissue. Only in this way can a clearer interpretation of the data presented be achieved. REFERENCES 1. Oi M, Oshida K, Sugimura S: The location of gastric ulcer. Gastroenterology 36:45-56, 1959 2. Card WI, Marks IN: The relationship between acid output of the stomach following "maximal" histamine stimulation and parietal cell mass. Clin Sci 19:147-163, 1960 3. Taylor KB, Fischer JM: Gastritis. Prog Gastroenterol 1:1-21, 1968 4. Rohrer GU, Welsh JD: Correlative study: gastric secretion and histology. Gastroenterology 52:185-191, 1967 5. Morson BC: Carcinoma arising from areas of intestinal metaplasia in the gastric mucosa. Br J Cancer 9:377-385, 1955 6. Prochaska B, Jirasek V, Barta V, et al: Differences in lactate dehydrogenase isoenzyme patterns in various parts of the human stomach and in peptic ulcer and gastric carcinoma. Gastroenterology 54:65-71, 1968
1325
7. Woollams R, Barratt PJ, Orwell RL, et al: LDH isoenzyme patterns of uninvolved gastric mucosa of patients with gastric carcinoma and benign gastric disease. Digestion 14:20-28, 1976 8. Umbreit WW, Burris RH , Stauffer JF: Manometric techniques and tissue metabolism. Second edition. Minneapolis, Burgess Publishing Co, p 119, 1949 9. Hansen DL, Bush ET: Improved solubilization procedures for liquid scintillation counting of biological materials. Anal Biochem 18:320-332, 1967 10. Lowry OH, Passonneau JV, Hasselberger FX, et al: Effect of ischaemia on known substrates and cofactors of the glycolytic pathway in brain. J Bioi Chern 239:18-30, 1964 11. Hohorst HJ: L-+-Lactate determination with lactate dehydrogenase and DPN. In Methods of Enzymatic Analysis. Edited by HV Bergmeyer. New York, London, Academic Press, p 266, 1962 12. Husain S, Paradise RR: Pitfalls in measurement of 14 C0 2 activity from glucose-6- 14 C and two corrective procedures. Exp Bioi Med 142:316-320, 1973 13. Katz J, Wood HG: The use of 14 C02 yields from glucose-1- and6- 14 C for the evaluation of the pathways of glucose metabolism. J Bioi Chern 238:517-523, 1963 14. Conn EE, StumpfPK: Electron transport and oxidativephosphorylation. In Outlines of Biochemistry. Second edition. New York, John Wiley & Sons, p 271, 1967 15. Menguy R, Desbaillets L, Masters YF: Mechanism of stress ulcer: influence of hypovolemic shock on energy metabolism in the gastric mucosa. Gastroenterology 66:46-55, 1974 16. Brand K, Deckner K: Quantitative relationship between the pentose phosphate pathway and the nucleotide synthesis in ascites tumor cells. Hoppe-Seylers Z Physiol Chern 351:711-717, 1970 17. Gumaa KA, McLean P: The pentose phosphate pathway of glucose metabolism; enzyme profiles and transient and steadystate content of intermediates of alternative pathways of glucose metabolism in Krebs ascites cells. Biochem J 115:1009-1029, 1969 18. Nakayama H, Weser E: Adaptation of small bowel after intestinal resection: increase in the pentose phosphate pathway. Biochim Biophys Acta 279:416-423, 1972 19. Kather H, Rivera M, Brand K: Interrelationship and control of glucose metabolism and lipogenesis in isolated fat cells. Biochem J 128:1089-1096, 1972 20. Katz J, Wals PA: Pentose cycle and reducing equivalents in rat mammary gland slices. Biochem J 128:879-899, 1972 21. Sernka TJ, Harris JB: Pentose phosphate shunt and gastric acid secretion in the rat. Am J Physiol 222:25-32, 1972 22 . Marks IN: The augmented histamine test. Gastroenterology 41:599-603, 1961 23. Grossman MI, Kirsner JB, Gillespie IE: Basal ·and histalogstimulated gastric secretion in control subjects and in patients . with peptic ulcer or gastric cancer. Gastroenterology 45:14-26, 1963 24. Piper DW, Kemp ML, Fenton BH, et al: Gastric juice lactic acidosis in the presence of gastric carcinoma. Gastroenterology 58:766-771, 1970 25. Kasugai T, Orwell RL, Clarke AD, et al: Gastric juice lactic acidosis: the incidence of gastric juice lactic acidosis in gastric cancer and benign gastric disease in communities of high and low gastric cancer incidence. Am J Dig Dis 19:599-602, 1974 26. Schrager J: Chromatographic studies of ·carbohydrate components of gastric and salivary mucopolysaccharides. Gut 5:166169, 1964 27. Piper DW, Griffith EM, Fenton BH: Gastric mucus secretion in normal subjects and patients with gastric ulcer, gastric carcinoma and pernicious anaemia. Am J Dig Dis 10:411-418, 1965