The Journal of Nutrition Vol. 129 No. 1 January 1999, pp. 105-108

A Carbohydrate-Rich Diet Stimulates Glucose-6-Phosphate Dehydrogenase Expression in Rat Hepatic Sinusoidal Endothelial Cells^1,2

Zoltán Spolarics

Department of Anatomy, Cell Biology and Injury Sciences, University of Medicine and Dentistry of New Jersey, Newark, NJ 07103

	ABSTRACT

Abstract Introduction Methods Results Discussion References

A carbohydrate-rich diet induces glucose-6-phosphate dehydrogenase (G6PD) in liver parenchymal cells, which supports fatty acid synthesis de novo. Bacterial endotoxins stimulate G6PD expression in hepatic sinusoidal endothelial and Kupffer cells but not in parenchymal cells. This study was designed to elucidate whether G6PD expression is regulated uniformly by dietary carbohydrates in hepatic sinusoidal and parenchymal cells. Freshly isolated cells from five groups of Sprague-Dawley rats were analyzed for G6PD activity and mRNA abundance. The rats were grouped as follows: 1) food deprived for 24 h; 2) food deprived for 24 h followed by consumption of the standard diet for 48 h; 3) food deprived for 24 h followed by consumption of a carbohydrate-rich diet for 48 h; 4) fed standard diet; and 5) fed standard diet followed by consumption of a carbohydrate-rich diet for 48 h. In endothelial cells, G6PD activity was 150% greater in group 3 than in group 1 and 125% greater in group 5 than in group 4. Steady-state G6PD mRNA levels were elevated by 300% in endothelial cells from group 3 compared with those from group 1. In Kupffer cells, G6PD activity and mRNA abundance were not different among the groups. As expected, G6PD expression was 700-1200% greater in parenchymal cells from rats fed a carbohydrate diet (groups 3 and 5) than from controls. Our results indicate that short-term consumption of a carbohydrate-rich diet stimulates G6PD expression in endothelial and parenchymal cells. Because G6PD supports reactive oxygen metabolism, the response may represent a preconditioning of antioxidant pathways in the hepatic cell populations that are targets of sinusoid-born reactive oxygen species during infections.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Glucose-6-phosphate dehydrogenase (G6PD)³ catalyzes the first and rate-limiting step of the hexose monophosphate shunt (HMS). NADPH and five carbon sugars produced in the HMS are required for synthetic processes and play an essential role in the maintenance of cellular redox status (Cramer et al. 1995, Kletzien et al. 1994, Pandolfi et al. 1995, Spolarics 1998). Basal G6PD expression varies widely in tissues, with the lowest activity shown in muscle and the greatest in phagocytic cells (Beutler 1990, Kletzien et al. 1994, Luzzatto and Mehta 1995). G6PD expression can be induced by a variety of physiologic and pathologic stimuli; however, the functional importance of elevated G6PD expression is different in various cell types (Kletzien et al. 1994, Spolarics 1998). In cells with a limited synthetic capacity such as RBC, G6PD plays a primary role in the elimination of reactive oxygen metabolites (Beutler 1990, Luzzatto and Mehta 1995). Recent investigations, however, have emphasized the importance of G6PD in supporting the metabolism of reactive oxygen species in hepatic sinusoidal cells as well as in extrahepatic tissues (Cramer et al. 1995, Kletzien et al. 1994, Pandolfi et al. 1995, Spolarics 1998). NADPH generated by the HMS is required for the production of superoxide anions and nitric oxide, whereas elimination of these species or their metabolites also depends on the HMS via glutathione peroxidases and catalase (Spolarics 1998).

We showed previously that G6PD is under divergent regulation in hepatic cells. Bacterial endotoxin in vivo stimulated G6PD expression in hepatic endothelial and Kupffer cells but not in parenchymal cells. A carbohydrate-rich diet stimulates G6PD expression in parenchymal cells in which the elevated HMS activity provides NADPH for de novo fatty acid synthesis (Morikawa et al. 1984, Prostko et al. 1989, Volpe and Vagelos 1976). However, no information is available on the nutritional regulation of HMS in hepatic sinusoidal cells. Thus, we hypothesized that the single copy G6PD gene is not uniformly regulated by nutritional carbohydrates in the functionally divergent cell types of the hepatic microenvironment. In this study, we tested whether a carbohydrate-rich diet alters G6PD expression in endothelial and Kupffer cells when consumed for a short time (48 h).

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Male Sprague-Dawley rats (300-340g, Charles River, Wilmington, MA) were used in the study; the diet is similar to that employed in previous investigations (Morikawa et al. 1984, Prostko et al. 1989). Animals were grouped according to the following dietary protocols: 1) food-deprived for 24 h; 2) food-deprived for 24 h then fed the standard diet for 48 h; 3) food-deprived for 24 h then fed a carbohydrate-rich diet for 48 h; 4) fed the standard diet; and 5) fed the standard diet followed by a carbohydrate-rich diet for 48 h. The carbohydrate-rich diet (AIN 1977 and 1980) consisted of 74.5% sucrose, 20.5% protein (casein), 3.5% salt mix, 1.0% vitamin mix, 0.3% DL-methionine and 0.2% choline bitartrate (wt/wt), (Dyets, Bethlehem, PA, diet #112144, vitamin mix #300050, salt mix #200000). Standard nonpurified rodent diet (Purina Mills, diet #5001) contained (wt/wt) 23.4% protein, 1.7 % lactose, 0.3% fructose, 0.2% glucose, 3.7% sucrose, 32% starch, 5.5% fat and 2.5% salt mix; the remaining part was ash and fibers. The standard and carbohydrate-rich diets contain the same vitamins in similar quantities. Metabolizable energy value of standard and carbohydrate-rich pellets was 12.7 and 15.8 kJ/g, respectively. Rats had free access to water and were kept in a climate-controlled animal facility with a 12-h dark:light cycle. The experiments were performed in accordance with NIH guidelines for the use of laboratory animals (NRC 1985).

Hepatic parenchymal and sinusoidal cells were isolated as described previously (Spolarics 1996). Purity of sinusoidal and parenchymal cells was >94 and 99%, respectively; cell viability, assessed by trypan blue exclusion, was >95 and 90%, respectively. Total cellular RNA from freshly prepared cells was subjected to Northern blot analysis as described previously (Spolarics and Navarro 1994). Sprague-Dawley rat G6PD cDNA, a gift from Susan Stapleton (Western Michigan University, Kalamazoo, MI), was labeled by random priming labeling. Hybridization signals were quantified using the Phosphorimager SI analyzer (Molecular Dynamics, Sunnyvale, CA). Values were normalized to optical densities of 28S rRNA stained by methylene blue before hybridizations (Spolarics 1996).

Freshly isolated hepatic cells were suspended in 50 mmol/L Tris buffer, pH 8.3, containing 100 mmol/L KCl, 0.2 mg/mL Triton X-100, 0.01 mmol/L NADP⁺ and a cocktail of proteinase inhibitors (Spolarics and Navarro 1994). The suspension was sonicated and samples of the 14,000 × g supernatant were analyzed for G6PD and 6-phosphogluconate dehydrogenase (6PGD) activities as described earlier (Spolarics and Navarro 1994).

ANOVA, followed by the Newman-Keuls test, was used to evaluate the data. A P-value 0.05 was considered significant. Values are means ± SEM.

	RESULTS

Abstract Introduction Methods Results Discussion References

We tested the effects of the various dietary regimens on the G6PD activity in endothelial and Kupffer cells (Fig. 1A). The activity of 6PGD, the second NADPH-generating enzyme in the HMS, was also determined (Fig. 1B). The enzyme activities were also determined in parenchymal cells prepared from the same livers, which served as a biological control (Fig. 1C, D). G6PD activity [nmol NADPH/(min·mg cell protein)] in cells from food-deprived rats was 21.1 ± 2.2 in endothelial, 8.9 ± 2.3 in parenchymal and 135.9 ± 6.1 in Kupffer cells. Mean 6PGD activity [nmol NADPH/(min·mg cell protein)] in cells from food-deprived rats was 24.5 ± 3.9 in endothelial, 42.6 ± 3.4 in parenchymal and 136.5 ± 5.6 in Kupffer cells. Enzyme activities in the other groups were expressed relative to those of the food-deprived rats.

View larger version (46K): [in this window] [in a new window]   Fig 1. The effect of a carbohydrate-rich diet on (A) glucose-6-phosphate dehydrogenase and (B) 6-phosphogluconate dehydrogenase activities in rat sinusoidal endothelial and Kupffer cells. For comparison, results obtained in parenchymal cells isolated from the same livers are also shown (C, D). Enzyme activities were measured in the cytosol of freshly isolated hepatic cells obtained from rats kept on various dietary schedules. Results are expressed relative to activities in cells from food-deprived rats. Bars represent means ± SEM, n = 6-8 independent cell preparations. Different letters above bars indicate significantly different means, P < 0.05.

G6PD activity was 125-150% greater in endothelial cells from rats fed the high carbohydrate diet compared with cells from those deprived of food and those fed the standard diet. G6PD activities were not different in endothelial cells when the rats were food deprived, fed the standard diet or refed the standard diet after food deprivation. In contrast, G6PD activities were not affected by the dietary protocols in Kupffer cells. As expected, the carbohydrate-rich diet resulted in a markedly greater G6PD activity in parenchymal cells compared with activity in cells from food-deprived rats or those fed the standard diet (Fig. 1C, D) (Morikawa et al. 1984, Prostko et al. 1989, Volpe 1976). Differences in 6PGD activity due to diet treatments were similar to those in G6PD in endothelial and parenchymal cells.

High carbohydrate feeding after food deprivation resulted in 300% greater steady-state G6PD mRNA concentrations in endothelial cells compared with cells from food-deprived rats (Fig. 2). Refeeding food-deprived rats with the standard diet did not result in higher levels of G6PD mRNA in endothelial cells. In Kupffer cells, the carbohydrate-rich diet did not affect steady-state levels of G6PD mRNA, consistent with the unaffected enzyme activity. A representative blot of parenchymal cells, also depicted in Figure 2, indicates that G6PD mRNA was not detectable in parenchymal cells from food-deprived rats, whereas refeeding with carbohydrate or standard diet increased G6PD mRNA level, in agreement with previously published observations (Kletzien et al. 1994, Morikawa et al. 1984, Prostko et al. 1989, Volpe 1976).

View larger version (45K): [in this window] [in a new window]   Fig 2. A carbohydrate-rich diet stimulates glucose-6-phosphate dehydrogenase mRNA abundance in rat sinusoidal endothelial cells. Total RNA from freshly isolated hepatic cells was subjected to Northern blot analyses. Signals on individual membranes were quantified by a phosphoimage analyzer and were normalized to optical densities of 28S rRNA stained before hybridizations. Results are expressed relative to signals in cells from food-deprived rats. Bars represent means ± SEM, n = 4-7 independent cell preparations. The upper panel depicts one representative Northern blot finding. For comparison, a typical finding from parenchymal cells is also shown. Different letters above bars indicate significant differences, P < 0.05.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

This study demonstrates for the first time that G6PD is under divergent dietary regulation in various cell populations of the hepatic microenvironment. Short-term carbohydrate diet consumption stimulates expression of G6PD in nonphagocytic sinusoidal endothelial and parenchymal cells, whereas it does not alter G6PD expression in resident liver macrophages. The high constitutive G6PD expression and the lack of response to diet in Kupffer cells do not represent the upper limit of gene expression because bacterial endotoxin or cytokine treatment can markedly increase G6PD expression in this cell type (Spolarics and Navarro 1994, Spolarics and Wu 1997). The elevated activity of G6PD in endothelial cells upon sucrose challenge is accompanied by increased levels of mRNA, suggesting that the responsible mechanism is gene activation and/or increased stability of the mRNA in endothelial cells similar to that observed in parenchymal cells (Fritz et al. 1986, Manos et al. 1991).

There is no evidence showing that sinusoidal endothelial cells contribute to hepatic de novo fatty acid synthesis, which suggests that the elevations in G6PD and 6PGD are not related to stimulated fatty acid anabolism in these cells. We propose that the responsiveness of the endothelium to dietary sucrose is associated with regulatory mechanisms that are retained in endothelial cells but lost in Kupffer cells during their differentiation. A difference in the insulin sensitivity of these cells is not a plausible explanation for the divergent responses because both sinusoidal cell types showed elevated glucose uptake after insulin administration in vivo (Spolarics et al. 1992). The exact mechanism of action of the sucrose diet in activating the G6PD gene is also not yet known in parenchymal cells (Kletzien et al. 1994). Therefore, the mechanism responsible for the divergent responses of hepatic cells remains to be elucidated.

Independent of the mechanism, the fact that dietary carbohydrates stimulate G6PD in endothelial cells may have an important effect on the hepatic detoxification of reactive oxygen species because the activity of the HMS in endothelial and Kupffer cells plays an important role in the maintenance of oxidant balance in the hepatic sinusoid (Spolarics 1998). The endotoxin-induced elevated G6PD parallels the conversion of Kupffer cells to the prooxidant state, whereas it aids the transition of endothelial cells to a more antioxidant condition. Induced G6PD in endotoxin-activated hepatic endothelial cells is associated with efficient maintenance of reduced glutathione and accelerated H₂O₂ detoxification (Spolarics et al. 1996, Spolarics and Wu 1997a and 1997b, Spolarics 1998). Because endothelial cells are the interfacing cellular layer between hepatic phagocytes and parenchymal cells, their intrinsic antioxidant activity is important in the maintenance of hepatic intercellular oxidant balance. The importance of G6PD as an antioxidant enzyme is supported by the following facts: cells deficient of G6PD can survive only under low oxygen tension (Pandolfi et al. 1995); red blood cells deficient of G6PD are sensitive to oxidant stress (Beutler 1990, Luzzatto and Mehta 1995); and oxidants stimulate G6PD expression in parenchymal cells (Cramer et al. 1995). Thus, the diet-induced alterations in the nonphagocytic cell population of the liver may represent a preconditioning of antioxidant pathways.

Diets containing varying concentrations of saturated and polyunsaturated fatty acids alter G6PD expression (Stabile et al. 1996, Taniguchi et al.1994, Tomlinson et al. 1988) and modulate the oxidative burst and oxidant injury in macrophages and pulmonary endothelial cells (Eicher and McVey 1995, Guimaraes et al. 1992, Hart et al.1991). Because the diet employed in this study causes lipid deposition in the liver, it remains to be elucidated if hepatic lipid accumulation, or dietary or circulating fatty acids, are involved in the induction of G6PD expression in sinusoidal endothelial cells.

Taken together, these studies indicate that, in liver, the single copy G6PD gene is under cell-specific regulation by nutritional carbohydrates. The responsiveness of these cells is not related to their endodermal or mesenchymal derivation or to their intrinsic capacity of de novo fatty acid synthesis. Differences in signaling pathways acting on the G6PD promoter, uniqueness of the cytoplasmic environment, the degree of cell differentiation or the accompanying level of oxidative stress can all potentially be responsible for the distinct cellular responses. The diet-induced elevated G6PD expression in endothelial and parenchymal cells may modulate hepatic responses during endotoxemia, sepsis or ischemia-reperfusion when oxidants from recruited neutrophils or resident macrophages target the sinusoidal endothelium and subsequently the parenchyma.

	FOOTNOTES

Manuscript received 28 July 1998. Initial reviews completed 10 September 1998. Revision accepted 21 October 1998.

	ACKNOWLEDGMENTS

We thank Jun-Xi Wu for his excellent technical assistance, and Susan Stapleton for providing the original G6PD cDNA for the studies.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• American Institute of Nutrition Report of the American Institute of Nutrition ad hoc committee on standards for nutritional studies. J. Nutr. 1977; 107:1340-1348
• American Institute of Nutrition Second report of the ad hoc committee on standards for nutritional studies. J. Nutr. 1980; 110:1726
• Beutler, E. (1990) Erythrocyte disorders: anemias due to increased destruction of erythrocytes with enzyme deficiencies. In: Hematology (Williams, W. J., Beutler, E., Erslev, A. & Lichtman, M. A., eds.), pp. 591-1726. McGraw-Hill, New York, NY.
• Cramer C. T., Cooke S., Ginsberg L. C., Kletzein R. F., Stapleton S. R., Ulrich R. G. Upregulation of glucose-6-phosphate dehydrogenase in response to hepatocellular oxidative stress: studies with diquat. J. Biochem. Toxicol. 1995; 10:293-298[Medline]
• Eicher S. D., McVey D. S. Dietary modulation of Kupffer cell and splenocyte function during a Salmonella typhimurium challenge in mice. J. Leukoc. Biol. 1995; 58:32-39[Abstract]
• Fritz R. S., Stumpo D. J., Kletzien R. F. Glucose-6-phosphate dehydrogenase mRNA abundance in primary cultures of rat hepatocytes. Effect of insulin and dexamethasone. Biochem. J. 1986; 237:617-619[Medline]
• Guimaraes A.R.P., Costa Rosa L.F.B.P., Safi D. A., Curi R. Metabolic and functional changes in macrophages of rats fed polyunsaturated or saturated fatty acid rich-diets during ageing. Biochem. Int. 1992; 27:9-16[Medline]
• Hart C. M., Tolson J. K., Block E. R. Supplemental fatty acids alter lipid peroxidation and oxidant injury in endothelial cells. Am J. Physiol. 1991; 260:L481-L488[Abstract/Free Full Text]
• Kletzien R. F., Harris P.K.W., Foellmi L. A. Glucose-6-phosphate dehydrogenase: a "housekeeping" enzyme subject to tissue-specific regulation by hormones, nutrients, and oxidant stress. FASEB J. 1994; 8:174-181[Abstract]
• Luzzatto, L. & Mehta, A. (1995) Glucose 6-phosphate dehydrogenase deficiency. In: The Metabolic and Molecular Bases of Inherited Disease (Scriver, C. R., Beaudet, A. L., Sly, W. S. & Valle, D., eds.), vol. 3, pp. 3367-3398. McGraw-Hill, New York, NY.
• Manos P., Nakayama R., Holten D. Regulation of glucose-6-phosphate dehydrogenase synthesis and mRNA abundance in cultured rat hepatocytes. Biochem. J. 1991; 276:245-250
• Morikawa N., Nakayama R., Holten D. Dietary induction of glucose-6-phosphate dehydrogenase synthesis. Biochem. Biophys. Res. Commun. 1984; 120:1022-1029[Medline]
• National Research Counil (1985) Guide for the Care and Use of Laboratory Animals. Publichation no. 85-23 (rev.), National Instututes of Health, Bethesda, MD.
• Pandolfi P. P., Sonati F., Rivi R., Mason P., Grosveld F., Luzzatto L. Targeted disruption of the housekeeping gene encoding glucose 6-phosphate dehydrogenase (G6PD): G6PD is dispensable for pentose synthesis but essential for defense against oxidative stress. EMBO J. 1995; 14:5209-5215[Medline]
• Prostko C. R., Fritz R. S., Kletzien R. F. Nutritional regulation of hepatic glucose-6-phosphate dehydrogenase. Biochem. J. 1989; 258:295-299[Medline]
• Spolarics, Z. (1996) Endotoxin stimulates the gene expression of reactive oxygen eliminating pathways in rat hepatic endothelial and Kupffer cells. Am. J. Physiol. 270: G660--G666.
• Spolarics Z. Endotoxemia, pentose cycle, and the oxidant/antioxidant balance in the hepatic sinusoid. J. Leukoc. Biol. 1998; 63:534-541[Abstract]
• Spolarics Z., Navarro L. Endotoxin stimulates the expression of glucose-6-phosphate dehydrogenase in Kupffer and hepatic endothelial cells. J. Leukoc. Biol. 1994; 56:453-457[Abstract]
• Spolarics Z., Ottlakan A., Lang C. H., Spitzer J. J. Kupffer cells play a major role in the insulin-mediated hepatic glucose uptake. Biochem. Biophys. Res. Commun. 1992; 186:455-459[Medline]
• Spolarics Z., Stein D. S., Garcia Z. C. Endotoxin stimulates hydrogen peroxide detoxifying activity in rat hepatic endothelial cells. Hepatology 1996; 24:691-696[Medline]
• Spolarics Z., Wu J. X. Role of glutathione and catalase in H₂O₂ detoxification in LPS-activated hepatic endothelial and Kupffer cells. Am. J. Physiol. 1997a; 273:G1304-G1311[Abstract/Free Full Text]
• Spolarics Z., Wu J. X. TNF- augments the gene expression of glucose-6-phosphate dehydrogense in rat Kupffer and endothelial cells. Life Sci. 1997b; 60:565-571[Medline]
• Stabile L. P., Hodge D. L., Klautky S. A., Salati L. M. Posttranscriptional regulation of glucose-6-phosphate dehydrogenase by dietary polyunsaturated fat. Arch. Biochem. Biophys. 1996; 332:269-279[Medline]
• Taniguchi M., Kinoshita C., Shoji Y., Inoue K. Insulin response to glucose-6-phosphate dehydrogenase activity is elevated in rats fed diets low in polyunsaturated fatty acids. J. Nutr. Sci. Vitaminol. 1994; 40:115-125
• Tomlinson J. E., Nakayama R., Holten D. Repression of pentose phosphate pathway dehydrogenase synthesis and mRNA by dietary fat in rats. J. Nutr. 1988; 118:408-415
• Volpe J. J., Vagelos P. R. Mechanism and regulation of biosynthesis of saturated fatty acids. Physiol. Rev. 1976; 56:339-417[Free Full Text]

0022-3166/99 $3.00 ©1999 American Society for Nutritional Sciences