GASTROENTJZROLOGY 1990;98:662-666
Effect of Chemically Induced Diabetes Mellitus on Glutamine Transport in Rat Intestine KENT VAN VOORHIS, FAYEZ K. GHISHAN
HAMID
Departments of Pediatrics, Molecular PhysioloRy/Biophyics, Medical School, Nashville, Tennessee
The effect of chemically induced diabetes mellitus on the intestinal transport of glutamine was examined using a brush-border membrane vesicle technique. Diabetes was induced by a single intraperitoneal injection of streptozotocin (100 mg/kg body weight). Control and diabetic rats were studied 5 days following the induction of diabetes. Na+-dependent and Na+-independent glutamine (0.5 mM) transport was found to be significantly higher in the diabetic rats than in the control rats. This increase was found to be caused by a significant increase in the V,, of the Na+-dependent and the Na+-independent glutamine transport processes in the diabetic rats (V,, of 3742 t 487 and 2055 + 279 pmol/mg protein per 7 s, respectively) compared with that of the control rats (2183 t 75 and 1271 + 83 pmol/mg protein per 7 s, respectively). The apparent K, values of glutamine transport systems, on the other hand, were similar in the two rat groups. Insulin treatment of the diabetic rats significantly reduced the V,, of glutamine transport by both the Na+-dependent and the Na+-independent processes to a level similar to that of the control rats (V max in the insulin-treated diabetic rats of 2036 + 123 and 1247 k 105 pmol/mg protein per 7 s, respectively). This study demonstrates that chemically induced diabetes mellitus is associated with an increase in intestinal glutamine transport. This increase is the result of the diabetic condition itself and appears to be mediated through an increase in the number of the transport carriers of glutamine.
n the rat, chemically induced diabetes mellitus results in changes in intestinal morphology and functions including transport of nutrients (l-3). An increase in hexoses (4,5], amino acids [S), and folate transport (7) has been documented in diabetic rats. In
I
and
M. SAID, NAJI ABUMRAD, and Surgery, Vanderbilt
University
contrast, intestinal transport of calcium (81 and strontium (9) has been shown to decrease under this condition. Although the described changes in intestinal transport are more pronounced in chronic diabetes mellitus, alterations have been shown to occur after 4-5 days of induction of diabetes when mucosal growth is similar between diabetic and control rats (lo]. The mechanisms of these changes are still largely unknown. Glutamine (gln) is an important amino acid involved in the synthesis of numerous biologically active compounds in mammalian cells and is a carrier form of ammonia 111).Recent studies have shown that gln, not glucose, is the major substrate for energy production in the small intestinal epithelial cells (12-14). We have recently characterized the transport process of gln across the intestinal brush border membrane of rat and human intestine using membrane vesicles technique (15,161. In both species, our results showed the existence of Na+-dependent and Na+-independent carrier-mediated transport processes (15,16). The effect of diabetes mellitus on gln transport is not known. In this study, we examined this issue in the rat using a well-validated brush-border membrane vesicle (BBMV) technique. Materials and Methods Materials L-(G-3H)-Gln (sp. act., 39 Ci/mmol] and scintillation fluid (ACS) were purchased from Amersham/Searle, Des Plaines, Ill.; unlabeled gln was purchased from Sigma
Abbreviations used in this paper: ACS, scintillatiok 8uid; BBM, brush-border membrane; BBMV, brush-border membrane vesicle; gin, glutamine. @ 1996 by the American Gastroenterological Association 0016-5065/90/$3.00
GLUTAMINE TRANSPORT IN DIABETIC RAT INTESTINE
April 1990
Chemical Company, St. Louis, MO. All other chemicals and reagents were obtained commercially and were of analytical quality.
Animals Male Sprague-Dawley rats (180-220 g) were purchased from Sasco, Omaha, Neb., and were maintained on Purina rat chow and allowed water ad libitum. Rats were made diabetic by intraperitoneal injection of streptozotocin [Upjohn, Kalamazoo, Mich.) (100 mg/kg body weight) dissolved in citrate buffer, pH 4.5. Control rats were given the same injection but without streptozotocin. Three days after the injection of streptozotocin the presence of diabetes was verified by the presence of glucosuria and blood sugar concentration >300 mg/ml. Rats were killed 5 days after the injection of streptozotocin. In some experiments, diabetic rats were given insulin twice daily (1 U/100 g body weight] (Squibb-Novo Inc., Princeton, N.J.) for 2 days and were killed 5 days after streptozotocin injection. This treatment caused total disappearance of glucosuria in our diabetic rats and has been shown to normalize blood glucose level (17).
Preparation
Membrane
863
constituents. The preincubation and incubation media, unless otherwise stated, were always isoosmotic. After incubation for 7 s [a time at which gln transport is linear, i.e., initial rate, see references 15 and 16), the reaction was terminated by the addition of 1 ml of ice-cold stop solution (200 mM NaCl, 100 mM mannitol, 10 mM K,HOP,, pH 7.4). The cold, diluted reaction mixture was immediately pipetted onto a prewetted filter and kept under suction. The filter was rinsed with 5 ml ice-cold stop solution and then dissolved in 5 ml of ACS scintillation cocktail. Radioactivity was counted in a scintillation counter (model LS 3801, Beckman Instruments, Irvine, Calif.). Nonspecific binding of the substrate to the filter [background) was determined by filtering a reaction mixture that contained an identical solution but no vesicles, and was subtracted from the transport data. Triplicate transport determinations were performed on each BBMV preparation. Transport data presented in this report are the results of 3-6 separate experiments performed on different days using freshly prepared BBMV and are expressed as the mean +SEM in picomoles per milligram protein per unit time. Protein concentrations were measured by the method of Lowry et al. (241 using bovine serum albumin as a standard. The specific activity of the brushborder membrane (BBM) marker enzyme sucrase was measured as described previously (25).
of Intestinal Brush-Border Vesicles and Transport Studies Results
Rats were killed with an overdose of ether. The jejunum [the 50 cm of the intestine that follows the ligament of Tritz) was removed, washed, and everted, and the mucosa was scraped with a glass slide. Intestinal BBMVs were isolated by a modification of Kessler’s divalent cations (Mg”) precipitation technique (18) as described in detail by us previously (19-22). All preparation steps were conducted at 4°C. By use of a Waring blender-type homogenizer at maximum speed, the mucosal scrapings were homogenized for 3 min in 60 ml of 300 mM mannitol, 5 mM ethyleneglycolbis (@aminoethyl ether)-N,N’-tetraacetic acid (EGTA), and 12 mM Tris aminomethane [(Tris)-HCL, pH 7.11; 240 ml of ice-cold distilled water was then added. The homogenate was treated with 3 ml of 1 M MgCl, and centrifuged at 3000g for 15 min (centrifuge model J2-21, Beckman Instruments, Fullerton, Calif.). The supernatant was then centrifuged at 27,000g for 30 min. The resulting pellet was resuspended in a Potter-Elvehjem tube for 10 strokes at the highest speed. The homogenate was treated with 0.6 ml of 1 M MgCl, and centrifuged at 3000g for 15 min. The pellet was resuspended in 30 ml of 250 mM mannitol and 20 mM HEPES Tris buffer and centrifuged at 50,OOOgfor 30 min. Using a tuberculin syringe with a 25-gauge needle, the pellet was resuspended in the desired volume of the transport (intravesicular) buffer [280 mM mannitol and 20 mM HEPES/Tris or 2-(Nmorpholino) ethane sulfonic acid, pH 7.41. Transport studies were performed by a rapid-filtration technique (23) as described by us previously (19-22). All incubations were done at room temperature to decrease possible metabolism of gin. The reaction was inhibited by adding a 20-~1 aliquot of membrane vesicle suspension to 80 ~1 of incubation buffer (final concentrations, 100 mM NaCl or KCl, 80 mM mannitol. and 20 mM HEPES/Mes, pH 6.5) containing various amounts of radiolabeled and unlabeled substrate plus other
To insure that the BBM fractions in the final BBMV preparations of the control and the diabetic rats were enriched to the same extent, we measured and compared the activity of the BBM marker enzyme sucrase in the final BBMV preparations and in the initial mucosal homogenates. The results are shown in Table 1. Enrichment of sucrase activity of 22- and 24.4fold were found in the control and the diabetic rats, respectively. Transport of 0.5 mM gln was examined in intestinal BBMV in the presence and absence of a Na + gradient (outside, 100 mM; inside, 0 mM) (in the absence of Na+ gradient, K+ gradient was used) in diabetic rats, and results were compared with transport in control rats. Incubation was performed for 7 s (i.e., initial rate)
Table 1. Specific Activity of Sucrose in Jejunal Brush-Border Membrane Vesicle Preparations and Initial Mucosal Homogenates of Control and Diabetic Rats Sucrase (mmol/mg protein/min) Control BBMV Homogenate Diabetic BBMV Homogenate Data are expressed as the mean + SEM. Number in parentheses.
747 i 84(7]
33.7 r 617) 1036 f 142(4) 42.5 F 1.2(5) of experiments
is
864
VAN VOORHIS
ET AL.
Vol. 98, No. 4
GASTROENTEROLOGY
because our previous studies in rat (and human) intestinal BBMV have shown that uptake of gln is linear at this time with no metabolism (l&16). The results obtained showed significantly higher gln transport in the diabetic rats than in control rats both in the presence (493.4 & 44.1 and 359.8 + 22.91 pmol/mg protein per 7 s, respectively; p -C0.025) and the absence (288.12 + 17.70 and 167.8 t 17.17 pmol/mg protein per 7 s, respectively; p < 0.011 of a Na+ gradient. To determine whether this increase in gln transport is caused by an increase in the number, activity, and/or affinity of the gln transport carriers, and to determine whether insulin treatment of the diabetic rats would normalize gln transport, we examined and compared gln transport in control, diabetic, and insulin-treated diabetic rats as a function of concentration (0.01-10 mM) in both the presence and absence of a Na+ gradient (outside, 100 mM; inside, 0 mM]. In all rat groups studied, gln transport was found, as before (15,16), to be saturable in the presence and absence of a Na+ gradient and was higher in the presence of a Na+ gradient than in its absence. Figure 1 shows the result of a single representative experiment of this type in the control rat. Kinetic parameters (i.e., V,,, and the apparent K,) of the Na+-dependent and the Na+-independent transport processes were calculated using the Woolf-Augustinisson-Hofstee plot (i.e., v against v/S). Figure 2A and B shows the transformation of the data in Figure 1 into this plot as a representative example. The results (tabulated in Table 2) showed that chemically induced diabetes mellitus is associated with a significant increase in the V,,, of gln transport by both the Na+-dependent (p < 0.01)
4 Glutamine
R = 0.96 2ooo
. 1000
0
A0
V/S 1600
I R = 0.99
1200
??
800
400
0
B0
200
100
300
400
v/s Figure 2. Woolf-Augustinsson-Hofstee in Figure 1.
plot of the data presented
A. Transport of gln by the Na+-dependent B. Transport of gln by the Na+-independent
8 concentration
1200
800
400
12 (mM)
system. system.
Figure 1. Transport of gln in the presence of a Na+ and a K+ gradient as a function of concentration. Jejunal BBMVs were preloaded with a buffer of 280 mM mannitol and 20 mM HEPES/Tris, pH 7.4. Incubation was performed for 7 s (i.e., initial rate) in an incubation buffer of 100 mM NaCl (0) or KC1 (O), 80 mM mannitol, and 20 n&l HEPES/Mes, pH 8.5. These results are from a representative experiment in control rats. Triplicate transport measurements were performed at each point.
GLUTAMINE
April1990
Table 2. Kinetic Parameters of Glutamine Transport Na+-Dependent and the Na+-Independent Processes in Control, Diabetic, and Treated Diabetic Flats
by the
TRANSPORT
IN DIABETIC RAT INTESTINE
865
partment with minimal binding to the membrane surface. Furthermore, in both species, gln transport was found to involve an electrogenic Na+-dependent carrier-mediated process that is probably occurring Na+-Dependent System Na+-Independent System through a gln-Na+ cotransport and an electroneutral Na+-independent carrier-mediated process (15,16). Vmax Vmax Our previous results indicated that the rat is an (pmol/mg (pmol/mg protein protein excellent animal model to be used in the study of the J$lJ [rkl] per 7 s) per 7 s) characteristics of gln transport under different conditions. Control 2183+ 75(5) 2.84+ 0.23(5] 1271r 83(5) 3.21+ O.lO(5) Diabetic 3742t 482(S)3.37+ 0.27(5] 2055+ 279(5]3.13k 0.35(5] Our present study on gln transport showed that (p < 0.01) (p < 0.025) (NSI JNSI chemically induced diabetes mellitus is associated Treated with a significant increase in gln transport by both the diabetic 2306f 132(5)2.82e 0.40(5) 1247+ 105(5) 2.96+ 0.86(5) Na+-dependent and the Na+-independent processes. INS) JNSI JNSI JNSJ This increase in gln transport is not caused by differJejunal BBMV were preloaded with a buffer of 280mM mannitol ences in the enrichment of BBM fraction in the final Incubation was performed for 7 s and 20 mM HEPES/Tris, pH 7.4. vesicular preparation in the two rat groups as shown [initial rate) in an incubation buffer of 100 mM NaCl or KCI, 80mM by the study with the BBM marker enzyme sucrase mannitol, and 20 mM HEPES/Mes, pH 6.5. Transport kinetic parameters were determined by the Woolf-Augustinsson-Hofstee (see Results). The increase in gln transport in the plot. Number of separate experiments performed on freshly prediabetic rats compared with the control rats was found pared BBMV on different days is indicated in parentheses. Statistito be caused by an increase in the V,,, of gln transport cal analysis was performed using the Student’s t-test. Comparison by both the Na+-dependent and the Na+-independent was performed relative to control. processes with no changes in the apparent K, value of the transport systems. These findings suggest that diabetes mellitus is associated with an increase in the and the Na+-independent (p < 0.025) transport pronumbers (and/or the activity] of the gln transport cesses compared with control rats. On the other hand, carriers (i.e., the Na+-dependent and the Natthe apparent K, values of the gln transport processes independent carriers] without changes in the affinity were similar in the diabetic and the control groups. of these carriers. Treatment of the diabetic rats with insulin caused The increase in the gln transport in the diabetic rats significant reduction in the increase in gln transport cannot be attributed to a decrease in Na+ permeability observed in the untreated diabetic rats by both the (261, because Na+ permeation is not affected by the Na+-dependent (p < 0.025) and the Na+ -independent induction of diabetes in the rat (27) and because not all (p < 0.025) transport processes (Table 2). In fact, the Na+ -dependent transport processes are affected by of gln transport in the insulin-treated V max values the induction of diabetes mellitus (28). Furthermore, in diabetic rats became similar to those of the control the present study the increase in gln transport was also rats. Insulin treatment, on the other hand, showed no observed in the absence of a Na+ gradient (see effect on the apparent K, values of the gln transport Results). This latter finding also suggests a common systems. mechanism responsible for the increase in gln transport by the Na+-dependent and the Na’-independent Discussion carrier-mediated processes. The increase in gln transport observed in the present study with acute diabetes The present study was designed to examine the effect of chemically induced diabetes mellitus on the mellitus cannot be attributed to changes in BBM composition or fluidity parameters that can ultimately transport of gln across rat intestinal BBM. This study is important because gln is the major substrate for affect the transport by both the Na+-dependent and energy production in the intestine (12-14) and because the Nat-independent transport systems. This is bediabetes mellitus is known to cause alterations in cause no changes occur in these parameters 4-5 days intestinal functions including nutrients absorption (lafter induction of diabetes in the rat (29). The increase 9). Transport of gln across the rat and human intestinal in gln transport observed in the diabetic rats was BBM has been recently characterized in our laboradiminished following treatment of the diabetic rats tory using BBMV technique (15,16). The results showed with insulin. This finding indicates that the increase in that the characteristics and mechanism of gln transgln transport in streptozotocin-induced diabetes melliport to be similar in the two species. Uptake of gln by tus is related to the diabetic condition itself, i.e., intestinal BBMV was found to be mostly the result of related to the lack of insulin and its consequences and transport of the substrate into the intravesicular comnot due to other nonspecific effects of streptozotocin.
866
VAN VOORHIS
ET AL.
As mentioned earlier, gln is the major fuel source for the intestine (12-141. The intestine obtains gin from two sources: an endogenous source (by uptake across the BLM) in which gln is produced as a product of protein catabolism by different tissues, particularly the skeletal muscles, and dietary sources through uptake from the lumen across the BBM (11-14). It has been shown that intestinal removal of endogenous gln is decreased in diabetic rats (30). From this finding and our present finding of increased gln transport across the BBM of the diabetic rat, an interesting picture appears. It seems that while the intestine of the diabetic rat increases its uptake of dietary gln across the BBM, it decreases gln removal from the endogenous source. This adaptation might be designed to allow other tissues (e.g., the liver), which under this condition might need more gln, to extract the substrate from the circulation. In summary, the present study demonstrates that chemically induced diabetes mellitus is associated with an increase in intestinal gln transport. The increase is caused by the diabetic condition itself and appears to be mediated through an increase in the number of the transport carriers of gln. References 1. Jarvis EL, Levin RJ. Anatomic
2. 3. 4.
5.
6. 7.
8.
9.
10. 11. 12.
13.
adaption of the alimentary tract of the rat to the hyperphagia of chronic alloxan diabetes. Nature 1966;210:391-393. Schedl HP, Wilson HD. Effect of diabetes on intestinal growth in the rat. J Exp Biol1971;176:187-496. Thompson AB. Experimental diabetes and intestinal barriers to absorption. Am J Physiol1982;244:G151-G157. Flares P, Schedle HP. Intestinal transport of 3-O-methyl-Oglucose in normal and alloxan diabetic rat. Am J Physiol 1968;214:725-729. Caspary WF. Increase of active transport of conjugated bile salts in streptozotocin-diabetic rat small intestine. Gut 1973;14:949955. Olsen WA, Rosenberg IH. Intestinal transport of sugars and amino acids in diabetic rats. J Clin Invest 1970;49:96-105. Said HM, Ghishan FK, Murrell J. Intestinal transport of 5methyltetrahydrofolate in diabetes mellitus. Diabetes Res 1986; 3:363-367. Schenider LE. Onedahl J. Schedle HP. Effect of vitamin D and its metabolites on calcium transport in the diabetic rat. Endocrinology 1976;99:793-796, Miller DL, Schedle HP. Effects of diabetes on duodenal and iteal strontium absorption in the rat. Gastroenterology 1975;68: 956-960. Schedl HP, Wilson HD. Effect of diabetes on intestinal growth in the rat. J Exp Biol1971;176:487-496, Haussinger D, Sies H. Glutamine metabolism in mammalian tissues. New York: Springer Verlag, 1984. Windmuller HG. Metabolism of vesicular and luminal glutamine by intestinal mucosa in vivo. In: Glutamine metabolism in mammalian tissues. New York: Springer Verlag, 1984:61-75. Windmuller HG, Spaeth AE. Identification of ketone bodies and glutamine as the major substrate for respiratory fuel in vivo for post absorptive rat small intestine. J Biol Chem 1978:253:69-76.
GASTROENTEROLOGY
Vol. 98, No. 4
14. Windmuller HG, Spaeth AE. Respiratory fuels and nitrogen metabolism in vivo in small intestine of fed rats. J Biol Chem 1980;255:197-112. 15. Said HM. Van Voorhis K, Ghishan FK, Abumrad N, Nylander W, Redha R. Transport characteristics of glutamine in human intestinal brush border membrane vesicles. Am J Physiol 1989: 256:G240-G245. 16. Van Voorhis K, Said HM, Ghishan FK, Abumrad N. Transport of glutamine in rat intestinal brush border membrane vesicles. Biochim Biophys Acta 1989;978:51-55. 17. Ghishan FK, Greene HL. Intestinal transport of zinc in the diabetic rat. Life Sci 1983;32:1735-1743. 18. Kessler M, Acto 0, Strolli C, Murer H, Muller M, Senenze G. A modified procedure for rapid preparation of efficient transporting vesicles from the small intestinal brush border membrane. Biochim Biophys Acta 1978;506:136-154. 19. Said HM, Ghishan FK. Redha R. Folate transport by human intestinal brush border membrane vesicles. Am J Physiol 1987; 252:G229-G236. 20. Said HM, Redha R. Nylander W. A carrier-mediated, Na. gradient-dependent transport system for biotin in human intestinal brush border membrane vesicles. Am J Physiol 1987;253: G631-G636. 21. Said HM, Redha R, Tipton W, Nylander W. Regulation of intestinal biotin transport in the rat: effect of biotin deficiency and supplementation. Am J Physiol256:G306-G311. 22. Said HM. Ghishan FK, Redha R. Transport of glycyl-L-proline in intestinal brush border membrane vesicles of the suckling rats: Characteristics and maturation. Biochim Biophys Acta 1988;941:232-240. 23. Hopfer U, Nelson K. Prevetto J, Isselbacher KG. Glucose transport in isolated brush border membrane from rat intestine. J Biol Chem 1973;248:253. 24. Lowry DH, Rosenbrough NJ, Farr AL, Randall RL. Protein measurements with folin phenol reagent. J Biol Chem 1951;193: 269.-275. A. Assay of intestinal disaccharidases. Anal Chem 25. Dahlquist 1984;7:18-25. of 26. Hopfer U. Diabetes mellitus: Changes in transport properties isolated intestinal microvillus membranes. Proc Nat1 Acad Sci USA 2975;2027-2031. 27. Ghishan FK, Borowitz S, Mulberg A. Phosphate transport by jejunal brush border membrane vesicles of the streptozotocin diabetic rat. Diabetes 1985;34:723-727. 28. Fedorak RN, Chang EB, Madara JL, Field M. Intestinal adaptation to diabetes: Altered Na +-dependent nutrient absorption in streptozotocin-treated clinically diabetic rats. J Clin Invest 1987;792:1571-1578. GR, Loismo HA, Olsen WA. Intestinal mucosa in 29. Gourley diabetic rats: Studies of microvillus membrane composition and microviscosity. Metabolism 1983;30:1053-1058. JT, Maer K. Hall DE, Colbourne SA, Bvosnan ME. 30. Brosnan Interorgan metabolism of amino acids in streptozotocin diabetic ketoacidotic rat. Am J Physiol 1983:244:E151-E158,
___--Received June 13,1989. Accepted September 22,1989. Address requests for reprints to: Hamid M. Said, M.D., Department of Medicine, Room C 352, Medical Science Building 1. University of California College of Medicine, Irvine, California 92717. This study was supported by NIH grants DK39501 and AM.26657. Kent Van Voorhis, 3rd-year medical student, is a recipient of the Diabetes Association of Tennessee Medical Student Research Fellowship.
Effect of Chemically Induced Diabetes Mellitus on Glutamine Transport in Rat Intestine KENT VAN VOORHIS, FAYEZ K. GHISHAN
HAMID
Departments of Pediatrics, Molecular PhysioloRy/Biophyics, Medical School, Nashville, Tennessee
The effect of chemically induced diabetes mellitus on the intestinal transport of glutamine was examined using a brush-border membrane vesicle technique. Diabetes was induced by a single intraperitoneal injection of streptozotocin (100 mg/kg body weight). Control and diabetic rats were studied 5 days following the induction of diabetes. Na+-dependent and Na+-independent glutamine (0.5 mM) transport was found to be significantly higher in the diabetic rats than in the control rats. This increase was found to be caused by a significant increase in the V,, of the Na+-dependent and the Na+-independent glutamine transport processes in the diabetic rats (V,, of 3742 t 487 and 2055 + 279 pmol/mg protein per 7 s, respectively) compared with that of the control rats (2183 t 75 and 1271 + 83 pmol/mg protein per 7 s, respectively). The apparent K, values of glutamine transport systems, on the other hand, were similar in the two rat groups. Insulin treatment of the diabetic rats significantly reduced the V,, of glutamine transport by both the Na+-dependent and the Na+-independent processes to a level similar to that of the control rats (V max in the insulin-treated diabetic rats of 2036 + 123 and 1247 k 105 pmol/mg protein per 7 s, respectively). This study demonstrates that chemically induced diabetes mellitus is associated with an increase in intestinal glutamine transport. This increase is the result of the diabetic condition itself and appears to be mediated through an increase in the number of the transport carriers of glutamine.
n the rat, chemically induced diabetes mellitus results in changes in intestinal morphology and functions including transport of nutrients (l-3). An increase in hexoses (4,5], amino acids [S), and folate transport (7) has been documented in diabetic rats. In
I
and
M. SAID, NAJI ABUMRAD, and Surgery, Vanderbilt
University
contrast, intestinal transport of calcium (81 and strontium (9) has been shown to decrease under this condition. Although the described changes in intestinal transport are more pronounced in chronic diabetes mellitus, alterations have been shown to occur after 4-5 days of induction of diabetes when mucosal growth is similar between diabetic and control rats (lo]. The mechanisms of these changes are still largely unknown. Glutamine (gln) is an important amino acid involved in the synthesis of numerous biologically active compounds in mammalian cells and is a carrier form of ammonia 111).Recent studies have shown that gln, not glucose, is the major substrate for energy production in the small intestinal epithelial cells (12-14). We have recently characterized the transport process of gln across the intestinal brush border membrane of rat and human intestine using membrane vesicles technique (15,161. In both species, our results showed the existence of Na+-dependent and Na+-independent carrier-mediated transport processes (15,16). The effect of diabetes mellitus on gln transport is not known. In this study, we examined this issue in the rat using a well-validated brush-border membrane vesicle (BBMV) technique. Materials and Methods Materials L-(G-3H)-Gln (sp. act., 39 Ci/mmol] and scintillation fluid (ACS) were purchased from Amersham/Searle, Des Plaines, Ill.; unlabeled gln was purchased from Sigma
Abbreviations used in this paper: ACS, scintillatiok 8uid; BBM, brush-border membrane; BBMV, brush-border membrane vesicle; gin, glutamine. @ 1996 by the American Gastroenterological Association 0016-5065/90/$3.00
GLUTAMINE TRANSPORT IN DIABETIC RAT INTESTINE
April 1990
Chemical Company, St. Louis, MO. All other chemicals and reagents were obtained commercially and were of analytical quality.
Animals Male Sprague-Dawley rats (180-220 g) were purchased from Sasco, Omaha, Neb., and were maintained on Purina rat chow and allowed water ad libitum. Rats were made diabetic by intraperitoneal injection of streptozotocin [Upjohn, Kalamazoo, Mich.) (100 mg/kg body weight) dissolved in citrate buffer, pH 4.5. Control rats were given the same injection but without streptozotocin. Three days after the injection of streptozotocin the presence of diabetes was verified by the presence of glucosuria and blood sugar concentration >300 mg/ml. Rats were killed 5 days after the injection of streptozotocin. In some experiments, diabetic rats were given insulin twice daily (1 U/100 g body weight] (Squibb-Novo Inc., Princeton, N.J.) for 2 days and were killed 5 days after streptozotocin injection. This treatment caused total disappearance of glucosuria in our diabetic rats and has been shown to normalize blood glucose level (17).
Preparation
Membrane
863
constituents. The preincubation and incubation media, unless otherwise stated, were always isoosmotic. After incubation for 7 s [a time at which gln transport is linear, i.e., initial rate, see references 15 and 16), the reaction was terminated by the addition of 1 ml of ice-cold stop solution (200 mM NaCl, 100 mM mannitol, 10 mM K,HOP,, pH 7.4). The cold, diluted reaction mixture was immediately pipetted onto a prewetted filter and kept under suction. The filter was rinsed with 5 ml ice-cold stop solution and then dissolved in 5 ml of ACS scintillation cocktail. Radioactivity was counted in a scintillation counter (model LS 3801, Beckman Instruments, Irvine, Calif.). Nonspecific binding of the substrate to the filter [background) was determined by filtering a reaction mixture that contained an identical solution but no vesicles, and was subtracted from the transport data. Triplicate transport determinations were performed on each BBMV preparation. Transport data presented in this report are the results of 3-6 separate experiments performed on different days using freshly prepared BBMV and are expressed as the mean +SEM in picomoles per milligram protein per unit time. Protein concentrations were measured by the method of Lowry et al. (241 using bovine serum albumin as a standard. The specific activity of the brushborder membrane (BBM) marker enzyme sucrase was measured as described previously (25).
of Intestinal Brush-Border Vesicles and Transport Studies Results
Rats were killed with an overdose of ether. The jejunum [the 50 cm of the intestine that follows the ligament of Tritz) was removed, washed, and everted, and the mucosa was scraped with a glass slide. Intestinal BBMVs were isolated by a modification of Kessler’s divalent cations (Mg”) precipitation technique (18) as described in detail by us previously (19-22). All preparation steps were conducted at 4°C. By use of a Waring blender-type homogenizer at maximum speed, the mucosal scrapings were homogenized for 3 min in 60 ml of 300 mM mannitol, 5 mM ethyleneglycolbis (@aminoethyl ether)-N,N’-tetraacetic acid (EGTA), and 12 mM Tris aminomethane [(Tris)-HCL, pH 7.11; 240 ml of ice-cold distilled water was then added. The homogenate was treated with 3 ml of 1 M MgCl, and centrifuged at 3000g for 15 min (centrifuge model J2-21, Beckman Instruments, Fullerton, Calif.). The supernatant was then centrifuged at 27,000g for 30 min. The resulting pellet was resuspended in a Potter-Elvehjem tube for 10 strokes at the highest speed. The homogenate was treated with 0.6 ml of 1 M MgCl, and centrifuged at 3000g for 15 min. The pellet was resuspended in 30 ml of 250 mM mannitol and 20 mM HEPES Tris buffer and centrifuged at 50,OOOgfor 30 min. Using a tuberculin syringe with a 25-gauge needle, the pellet was resuspended in the desired volume of the transport (intravesicular) buffer [280 mM mannitol and 20 mM HEPES/Tris or 2-(Nmorpholino) ethane sulfonic acid, pH 7.41. Transport studies were performed by a rapid-filtration technique (23) as described by us previously (19-22). All incubations were done at room temperature to decrease possible metabolism of gin. The reaction was inhibited by adding a 20-~1 aliquot of membrane vesicle suspension to 80 ~1 of incubation buffer (final concentrations, 100 mM NaCl or KCl, 80 mM mannitol. and 20 mM HEPES/Mes, pH 6.5) containing various amounts of radiolabeled and unlabeled substrate plus other
To insure that the BBM fractions in the final BBMV preparations of the control and the diabetic rats were enriched to the same extent, we measured and compared the activity of the BBM marker enzyme sucrase in the final BBMV preparations and in the initial mucosal homogenates. The results are shown in Table 1. Enrichment of sucrase activity of 22- and 24.4fold were found in the control and the diabetic rats, respectively. Transport of 0.5 mM gln was examined in intestinal BBMV in the presence and absence of a Na + gradient (outside, 100 mM; inside, 0 mM) (in the absence of Na+ gradient, K+ gradient was used) in diabetic rats, and results were compared with transport in control rats. Incubation was performed for 7 s (i.e., initial rate)
Table 1. Specific Activity of Sucrose in Jejunal Brush-Border Membrane Vesicle Preparations and Initial Mucosal Homogenates of Control and Diabetic Rats Sucrase (mmol/mg protein/min) Control BBMV Homogenate Diabetic BBMV Homogenate Data are expressed as the mean + SEM. Number in parentheses.
747 i 84(7]
33.7 r 617) 1036 f 142(4) 42.5 F 1.2(5) of experiments
is
864
VAN VOORHIS
ET AL.
Vol. 98, No. 4
GASTROENTEROLOGY
because our previous studies in rat (and human) intestinal BBMV have shown that uptake of gln is linear at this time with no metabolism (l&16). The results obtained showed significantly higher gln transport in the diabetic rats than in control rats both in the presence (493.4 & 44.1 and 359.8 + 22.91 pmol/mg protein per 7 s, respectively; p -C0.025) and the absence (288.12 + 17.70 and 167.8 t 17.17 pmol/mg protein per 7 s, respectively; p < 0.011 of a Na+ gradient. To determine whether this increase in gln transport is caused by an increase in the number, activity, and/or affinity of the gln transport carriers, and to determine whether insulin treatment of the diabetic rats would normalize gln transport, we examined and compared gln transport in control, diabetic, and insulin-treated diabetic rats as a function of concentration (0.01-10 mM) in both the presence and absence of a Na+ gradient (outside, 100 mM; inside, 0 mM]. In all rat groups studied, gln transport was found, as before (15,16), to be saturable in the presence and absence of a Na+ gradient and was higher in the presence of a Na+ gradient than in its absence. Figure 1 shows the result of a single representative experiment of this type in the control rat. Kinetic parameters (i.e., V,,, and the apparent K,) of the Na+-dependent and the Na+-independent transport processes were calculated using the Woolf-Augustinisson-Hofstee plot (i.e., v against v/S). Figure 2A and B shows the transformation of the data in Figure 1 into this plot as a representative example. The results (tabulated in Table 2) showed that chemically induced diabetes mellitus is associated with a significant increase in the V,,, of gln transport by both the Na+-dependent (p < 0.01)
4 Glutamine
R = 0.96 2ooo
. 1000
0
A0
V/S 1600
I R = 0.99
1200
??
800
400
0
B0
200
100
300
400
v/s Figure 2. Woolf-Augustinsson-Hofstee in Figure 1.
plot of the data presented
A. Transport of gln by the Na+-dependent B. Transport of gln by the Na+-independent
8 concentration
1200
800
400
12 (mM)
system. system.
Figure 1. Transport of gln in the presence of a Na+ and a K+ gradient as a function of concentration. Jejunal BBMVs were preloaded with a buffer of 280 mM mannitol and 20 mM HEPES/Tris, pH 7.4. Incubation was performed for 7 s (i.e., initial rate) in an incubation buffer of 100 mM NaCl (0) or KC1 (O), 80 mM mannitol, and 20 n&l HEPES/Mes, pH 8.5. These results are from a representative experiment in control rats. Triplicate transport measurements were performed at each point.
GLUTAMINE
April1990
Table 2. Kinetic Parameters of Glutamine Transport Na+-Dependent and the Na+-Independent Processes in Control, Diabetic, and Treated Diabetic Flats
by the
TRANSPORT
IN DIABETIC RAT INTESTINE
865
partment with minimal binding to the membrane surface. Furthermore, in both species, gln transport was found to involve an electrogenic Na+-dependent carrier-mediated process that is probably occurring Na+-Dependent System Na+-Independent System through a gln-Na+ cotransport and an electroneutral Na+-independent carrier-mediated process (15,16). Vmax Vmax Our previous results indicated that the rat is an (pmol/mg (pmol/mg protein protein excellent animal model to be used in the study of the J$lJ [rkl] per 7 s) per 7 s) characteristics of gln transport under different conditions. Control 2183+ 75(5) 2.84+ 0.23(5] 1271r 83(5) 3.21+ O.lO(5) Diabetic 3742t 482(S)3.37+ 0.27(5] 2055+ 279(5]3.13k 0.35(5] Our present study on gln transport showed that (p < 0.01) (p < 0.025) (NSI JNSI chemically induced diabetes mellitus is associated Treated with a significant increase in gln transport by both the diabetic 2306f 132(5)2.82e 0.40(5) 1247+ 105(5) 2.96+ 0.86(5) Na+-dependent and the Na+-independent processes. INS) JNSI JNSI JNSJ This increase in gln transport is not caused by differJejunal BBMV were preloaded with a buffer of 280mM mannitol ences in the enrichment of BBM fraction in the final Incubation was performed for 7 s and 20 mM HEPES/Tris, pH 7.4. vesicular preparation in the two rat groups as shown [initial rate) in an incubation buffer of 100 mM NaCl or KCI, 80mM by the study with the BBM marker enzyme sucrase mannitol, and 20 mM HEPES/Mes, pH 6.5. Transport kinetic parameters were determined by the Woolf-Augustinsson-Hofstee (see Results). The increase in gln transport in the plot. Number of separate experiments performed on freshly prediabetic rats compared with the control rats was found pared BBMV on different days is indicated in parentheses. Statistito be caused by an increase in the V,,, of gln transport cal analysis was performed using the Student’s t-test. Comparison by both the Na+-dependent and the Na+-independent was performed relative to control. processes with no changes in the apparent K, value of the transport systems. These findings suggest that diabetes mellitus is associated with an increase in the and the Na+-independent (p < 0.025) transport pronumbers (and/or the activity] of the gln transport cesses compared with control rats. On the other hand, carriers (i.e., the Na+-dependent and the Natthe apparent K, values of the gln transport processes independent carriers] without changes in the affinity were similar in the diabetic and the control groups. of these carriers. Treatment of the diabetic rats with insulin caused The increase in the gln transport in the diabetic rats significant reduction in the increase in gln transport cannot be attributed to a decrease in Na+ permeability observed in the untreated diabetic rats by both the (261, because Na+ permeation is not affected by the Na+-dependent (p < 0.025) and the Na+ -independent induction of diabetes in the rat (27) and because not all (p < 0.025) transport processes (Table 2). In fact, the Na+ -dependent transport processes are affected by of gln transport in the insulin-treated V max values the induction of diabetes mellitus (28). Furthermore, in diabetic rats became similar to those of the control the present study the increase in gln transport was also rats. Insulin treatment, on the other hand, showed no observed in the absence of a Na+ gradient (see effect on the apparent K, values of the gln transport Results). This latter finding also suggests a common systems. mechanism responsible for the increase in gln transport by the Na+-dependent and the Na’-independent Discussion carrier-mediated processes. The increase in gln transport observed in the present study with acute diabetes The present study was designed to examine the effect of chemically induced diabetes mellitus on the mellitus cannot be attributed to changes in BBM composition or fluidity parameters that can ultimately transport of gln across rat intestinal BBM. This study is important because gln is the major substrate for affect the transport by both the Na+-dependent and energy production in the intestine (12-14) and because the Nat-independent transport systems. This is bediabetes mellitus is known to cause alterations in cause no changes occur in these parameters 4-5 days intestinal functions including nutrients absorption (lafter induction of diabetes in the rat (29). The increase 9). Transport of gln across the rat and human intestinal in gln transport observed in the diabetic rats was BBM has been recently characterized in our laboradiminished following treatment of the diabetic rats tory using BBMV technique (15,16). The results showed with insulin. This finding indicates that the increase in that the characteristics and mechanism of gln transgln transport in streptozotocin-induced diabetes melliport to be similar in the two species. Uptake of gln by tus is related to the diabetic condition itself, i.e., intestinal BBMV was found to be mostly the result of related to the lack of insulin and its consequences and transport of the substrate into the intravesicular comnot due to other nonspecific effects of streptozotocin.
866
VAN VOORHIS
ET AL.
As mentioned earlier, gln is the major fuel source for the intestine (12-141. The intestine obtains gin from two sources: an endogenous source (by uptake across the BLM) in which gln is produced as a product of protein catabolism by different tissues, particularly the skeletal muscles, and dietary sources through uptake from the lumen across the BBM (11-14). It has been shown that intestinal removal of endogenous gln is decreased in diabetic rats (30). From this finding and our present finding of increased gln transport across the BBM of the diabetic rat, an interesting picture appears. It seems that while the intestine of the diabetic rat increases its uptake of dietary gln across the BBM, it decreases gln removal from the endogenous source. This adaptation might be designed to allow other tissues (e.g., the liver), which under this condition might need more gln, to extract the substrate from the circulation. In summary, the present study demonstrates that chemically induced diabetes mellitus is associated with an increase in intestinal gln transport. The increase is caused by the diabetic condition itself and appears to be mediated through an increase in the number of the transport carriers of gln. References 1. Jarvis EL, Levin RJ. Anatomic
2. 3. 4.
5.
6. 7.
8.
9.
10. 11. 12.
13.
adaption of the alimentary tract of the rat to the hyperphagia of chronic alloxan diabetes. Nature 1966;210:391-393. Schedl HP, Wilson HD. Effect of diabetes on intestinal growth in the rat. J Exp Biol1971;176:187-496. Thompson AB. Experimental diabetes and intestinal barriers to absorption. Am J Physiol1982;244:G151-G157. Flares P, Schedle HP. Intestinal transport of 3-O-methyl-Oglucose in normal and alloxan diabetic rat. Am J Physiol 1968;214:725-729. Caspary WF. Increase of active transport of conjugated bile salts in streptozotocin-diabetic rat small intestine. Gut 1973;14:949955. Olsen WA, Rosenberg IH. Intestinal transport of sugars and amino acids in diabetic rats. J Clin Invest 1970;49:96-105. Said HM, Ghishan FK, Murrell J. Intestinal transport of 5methyltetrahydrofolate in diabetes mellitus. Diabetes Res 1986; 3:363-367. Schenider LE. Onedahl J. Schedle HP. Effect of vitamin D and its metabolites on calcium transport in the diabetic rat. Endocrinology 1976;99:793-796, Miller DL, Schedle HP. Effects of diabetes on duodenal and iteal strontium absorption in the rat. Gastroenterology 1975;68: 956-960. Schedl HP, Wilson HD. Effect of diabetes on intestinal growth in the rat. J Exp Biol1971;176:487-496, Haussinger D, Sies H. Glutamine metabolism in mammalian tissues. New York: Springer Verlag, 1984. Windmuller HG. Metabolism of vesicular and luminal glutamine by intestinal mucosa in vivo. In: Glutamine metabolism in mammalian tissues. New York: Springer Verlag, 1984:61-75. Windmuller HG, Spaeth AE. Identification of ketone bodies and glutamine as the major substrate for respiratory fuel in vivo for post absorptive rat small intestine. J Biol Chem 1978:253:69-76.
GASTROENTEROLOGY
Vol. 98, No. 4
14. Windmuller HG, Spaeth AE. Respiratory fuels and nitrogen metabolism in vivo in small intestine of fed rats. J Biol Chem 1980;255:197-112. 15. Said HM. Van Voorhis K, Ghishan FK, Abumrad N, Nylander W, Redha R. Transport characteristics of glutamine in human intestinal brush border membrane vesicles. Am J Physiol 1989: 256:G240-G245. 16. Van Voorhis K, Said HM, Ghishan FK, Abumrad N. Transport of glutamine in rat intestinal brush border membrane vesicles. Biochim Biophys Acta 1989;978:51-55. 17. Ghishan FK, Greene HL. Intestinal transport of zinc in the diabetic rat. Life Sci 1983;32:1735-1743. 18. Kessler M, Acto 0, Strolli C, Murer H, Muller M, Senenze G. A modified procedure for rapid preparation of efficient transporting vesicles from the small intestinal brush border membrane. Biochim Biophys Acta 1978;506:136-154. 19. Said HM, Ghishan FK. Redha R. Folate transport by human intestinal brush border membrane vesicles. Am J Physiol 1987; 252:G229-G236. 20. Said HM, Redha R. Nylander W. A carrier-mediated, Na. gradient-dependent transport system for biotin in human intestinal brush border membrane vesicles. Am J Physiol 1987;253: G631-G636. 21. Said HM, Redha R, Tipton W, Nylander W. Regulation of intestinal biotin transport in the rat: effect of biotin deficiency and supplementation. Am J Physiol256:G306-G311. 22. Said HM. Ghishan FK, Redha R. Transport of glycyl-L-proline in intestinal brush border membrane vesicles of the suckling rats: Characteristics and maturation. Biochim Biophys Acta 1988;941:232-240. 23. Hopfer U, Nelson K. Prevetto J, Isselbacher KG. Glucose transport in isolated brush border membrane from rat intestine. J Biol Chem 1973;248:253. 24. Lowry DH, Rosenbrough NJ, Farr AL, Randall RL. Protein measurements with folin phenol reagent. J Biol Chem 1951;193: 269.-275. A. Assay of intestinal disaccharidases. Anal Chem 25. Dahlquist 1984;7:18-25. of 26. Hopfer U. Diabetes mellitus: Changes in transport properties isolated intestinal microvillus membranes. Proc Nat1 Acad Sci USA 2975;2027-2031. 27. Ghishan FK, Borowitz S, Mulberg A. Phosphate transport by jejunal brush border membrane vesicles of the streptozotocin diabetic rat. Diabetes 1985;34:723-727. 28. Fedorak RN, Chang EB, Madara JL, Field M. Intestinal adaptation to diabetes: Altered Na +-dependent nutrient absorption in streptozotocin-treated clinically diabetic rats. J Clin Invest 1987;792:1571-1578. GR, Loismo HA, Olsen WA. Intestinal mucosa in 29. Gourley diabetic rats: Studies of microvillus membrane composition and microviscosity. Metabolism 1983;30:1053-1058. JT, Maer K. Hall DE, Colbourne SA, Bvosnan ME. 30. Brosnan Interorgan metabolism of amino acids in streptozotocin diabetic ketoacidotic rat. Am J Physiol 1983:244:E151-E158,
___--Received June 13,1989. Accepted September 22,1989. Address requests for reprints to: Hamid M. Said, M.D., Department of Medicine, Room C 352, Medical Science Building 1. University of California College of Medicine, Irvine, California 92717. This study was supported by NIH grants DK39501 and AM.26657. Kent Van Voorhis, 3rd-year medical student, is a recipient of the Diabetes Association of Tennessee Medical Student Research Fellowship.
