Intestinal Calcium Absorption Is Enhanced by n-glucose in Diabetic and Control Rats

Transcription

1 GASTROENTEROLOGY 1985;88:933-8 Intestinal Calcium Absorption Is Enhanced by n-glucose in Diabetic and Control Rats M. KABIR YOUNOSZAI and R. NATHAN, with the technical assistance of LYNN YOKOYAMA.,. Department of Pediatrics, Division of Gastroenterology, University of Iowa Hospitals and Climes, Iowa City, Iowa The effect of glucose on intestinal absorption of calcium was studied in the jejunum and ileum of control, diabetic [streptozotocin-induced), and insulin-treated diabetic rats. Intestinal absorption was determined in vivo using an in situ one-pass perfusion technique. In the jejunum of control and diabetic rats, net absorption and unidirectional lumen to mucosa flux of calcium and net absorption of water were significantly greater during perfusion of an isotonic Nael solution, containing 15 mmolll of glucose, than during perfusion of the same solution containing 15 mmolll of mannitol instead of glucose. To determine if net absorption of calcium and water were related, the jejunum was perfused with a hypotonic solution [260 mosmollkg) in separate groups of rats. Although net absorption of water was equivalent during perfusion of the hypotonic solution to that noted during perfusion of the isotonic glucose-containing solution, rate of absorption of calcium was not enhanced. Thus, it appeared that, if the enhancement in absorption of calcium by glucose was an effect on the passive absorption of calcium, it was through a mechanism not related to passive absorption of water. As expected, jejunal absorption of calcium was lower in diabetic than in control rats. Because in diabetic rats the metabolism of vitamin D to its active metabolite, 1,25-dihydroxy vitamin D, is defective, the enhancement in absorption of calcium by glucose did not appear to be due to a mechanism influenced by vitamin D metabolism. In the ileum, rate of absorption of calcium was lower than in the corresponding jejunum and was not significantly altered by the presence of glucose Received May 3, Accepted October 12, Address requests for reprints to: M. Kabir Younoszai, M.D., Department of Pediatrics, Division of Gastroenterology, University of Iowa Hospitals and Clinics, Iowa City, Iowa by the American Gastroenterological Association /85/$3.30 in the perfusion solution, by perfusion of a hypotonic solution, or by the diabetic state, The mechanism of action of glucose on calcium absorption in the jejunum needs to be studied further. Absorption of calcium from the intestine takes place via carrier-mediated saturable and nonmediated diffusive processes. The carrier-mediated process is under the influence of vitamin D, especially its metabolite 1,25-dihydroxycholetalciferol (1,25 [OH]zD3)' Passive diffusion across the brush border membrane of the epithelial cells, and across the tight junctions and lateral intercellular spaces, is secondary to electrochemical gradients and solvent drag. In adult diabetic (chemically induced) rats, 1-2 wk after induction of diabetes, absorption of calcium from the small intestine was reduced as was the serum levels of 1,25[OH]zD3 (1-3). Absorption of glucose, however, was found to! be enhanced in diabetic rats (4). In healthy hunians, glucose appeared to increase intestinal absorption of calcium, sodium, and water (5). We undertook the present studies to 'determine if glucose wquld also enhance absorption of calcium, sodium, ~nd water in the intestine of young, growing diabetic rats. In the human intestine, the increased absorption of calcium was thought to be only indiirectly related to absorption of water and not caused by solvent drag (5). To further define the mechanism of enhanced calcium absorption, we studied the effect of passive water absorption, induced by imposing an osmotic pressure gradient between lumen and plasma (perfusion of hypotonic solution), to determine if absorption of calcium would be enhanced as the result of solvent drag. One group of diabetic rats was treated' Abbreviations used in this paper: 1,25[OH)zD 3, 1,25-dihydroxycholecalciferol.

2 934 YOUNOSZAI ET AL. GASTROENTEROLOGY Vol. 88, No.4 with insulin to assess the effect of such treatment on intestinal absorption of calcium. Material and Methods Weanling male albino rats (40-50 g body wt) were obtained from Biolabs (Madison, Wis.) and maintained on a regular laboratory rat diet (Teklad Diets, Madison, Wis.) offered ad libitum. Rats to be made diabetic were injected intraperitoneally with streptozotocin (LaRoche, Kalamazoo, Mich.) dissolved in citrate buffer (ph 4.0) 1 min before injection. Control rats matched in body weight to rats to be rendered diabetic were injected with a similar volume of citrate buffer. Streptozotocin was injected in two doses, 60 mg/kg body wt on the first day, and 50 mg/kg body wt 24 h later. About 40% of the rats injected became diabetic as evidenced by loss of weight or poor weight gain, glucosuria, hyperglycemia, polyuria, and polydypsia. Fifty percent of the diabetic rats were treated with neutral protamine Hagedorn insulin, 2 IU/100 g body wt once daily. Intestinal absorption of calcium, sodium, and water were determined 2 wk after induction of diabetes using a one-pass in vivo perfusion technique. For these studies the experimental procedures used were similar to those described previously (6,7). Rats were anesthetized with intraperitoneal injection of phenobarbital, 0.7 mg/100 g body wt. The entire small intestine was perfused as two segments: the jejunum started just distal to the ligament of Treitz and extended cm distally, and the ileum started ~1 cm beyond the jejunum and extended to just proximal to the ileocecal junction. These segments were rinsed with ml of an isotonic NaCI solution, flushed with 50 ml of air, and then cannulated at both proximal and distal ends. The cannulated segments were inserted back into the abdominal cavity and perfused in situ, at a rate of 0.3 mllmin, with a solution containing per liter (double distilled water): 120 mmol of NaCI, 5 mmol of KCI, 3 mmol of CaCl, radioactive 45CaClz (1 x 10 5 dpm/ml) (2.5 /LCi//Lmol, New England Nuclear, Boston, Mass.), 20 mg of phenol red, and mannitol (Fisher Scientific Company, Fair Lawn, N.J.) in sufficient amounts (40 mmol) to achieve an osmolality of mosmollkg. To determine the effect of glucose on calcium absorption, 15 mmol of D-glucose (Fisher Scientific) was added to the perfusion solution. In these solutions the concentration of mannitol was 25 mmolil. To determine the effect of passive water absorption on calcium transport, the segments of the intestine in control and diabetic rats were perfused with a hypotonic solution (260 mosmollkg) free of glucose. The composition of the solution was identical to that described previously, with the exception that the content of mannitol was decreased to achieve 260 mosmollkg of perfusion solution. The glassware and tubings used for the studies were rinsed with 1 N hydrochloric acid to remove adherent calcium. Six to 8 control, diabetic, and treated diabetic rats were used for study of absorption of calcium, sodium, and water from each of the four solutions (+G and -G; isotonic and hypotonic). During the perfusion period, the body temperature of the rats was maintained at C. The solution perfused through the cannulated segments over the first 60 min of perfusion was discarded, and that over the next 60 min was collected in three consecutive 20-min periods. The concentrations of phenol red, calcium, and sodium were determined in the initial perfusion solutions and in each of the collected solutions (perfusates). At the end of the perfusion period the perfused jejunal and ileal segments were stripped from the mesentery and their lengths were measured. Wet weight of the segments was determined after gently pressing out the luminal contents with the forefingers. The tissues were then dried in a vacuum oven at C for 24 h and their dry weights were determined. Calcium content of the perfusion solution was determined immediately after perfusion using a Perkin Elmer atomic absorption spectrometer (Perkin Elmer Corp., Norwalk, Conn.), 45Ca radioactivity was determined in a Beckman LS300 liquid scintillation spectrometer (Beckman Instruments Inc., Fullerton, Calif.), phenol red as described by Schedl and Clifton (8). Sodium was determined using a Beckman flame photometer (Beckman Instruments). Net water transport was determined from change in concentration of phenol red as follows: Net water tra~sport (m1/20 min) = V;(l - PR;lPRJ), where Vi is the volume of fluid perfused per milliliter per 20 min and PH is the concentration of phenol red in micrograms per milliliter, in the perfusion fluids. The subscripts i and f are the initial and final values in the perfusion solutions before perfusion (i), and in the perfusates collected (f). Lumen to mucosa flux of calcium and net absorption of calcium or sodium were determined from rates of disappearance of 45Ca, calcium, and sodium, respectively, from the perfusion solution during perfusion. Lumen to mucosa flux of calcium and net absorption of calcium and sodium were calculated using the following formulas: Lumen to mucosa flux of Ca (J-tmoIl20 min) and 4SCa PR.) = y. 4SCa. - J ' -0- sp act,(, PRJ Net ab'sorption (/LmoI/20 min) CJPF;) = Y; C; - --, ( PRJ where 45Ca is the radioactivity of 45Ca in disintegrations per minute per milliliter of perfusion solution; sp act is the specific activity of 45Ca in the perfusion solution in disintegrations per minute per micromole of calcium; and C is the concentration of calcium or sodium in the perfusion fluid in micromoles per milliliter. The other symbols have the same meaning as in the previous equation. Because absorbed 45Ca dilutes in a relatively large pool of 40C a in the intestinal tissues and is carried away by blood, very little 45Ca will reenter the intestinal lumen. Thus, the rate of absorption of 45Ca may be taken to approximate the undirectionallumen to mucosa flux of calcium. During the 60-min collection period, steady-state conditions had been achieved as net absorption of water, calcium, and sodium during each of the three 20-min collection periods did not vary from each other by >20%. The

3 April 1985 GLUCOSE ENHANCES Ca ABSORPTION 935 amounts of calcium, sodium, and water absorbed during these three periods were added to obtain the amount absorbed in 1 h. To normalize for differences in the size of segments between control and diabetic rats, the amount of calcium, sodium, and water absorbed were expressed on the basis of weight of the intestine. Statistical Analysis The paired t-test was used to compare mean values between corresponding segments of the same rats. Tukey's test for multiple-comparison was used to compare mean values between correspondihg segments of the control and the diabetic rats. A value of p < 0.05 was taken as indicating a statistically significant difference between the corresponding mean values (9). Results Urine glucose, tested by clinitest, was negative in the control rats, 2+ to 4+ in the diabetic rats, and occasionally 2 + or more in the treated diabetic nits. Daily water intake and urine volume were <10 ml in the control rats, whereas they were >100 ml in the diabetic and >60 ml in the treated diabetic rats. Serum glucose values were normal in the control tats, usually >400 mg/dl in the diabetic rats, and intermediate in treated diabetic rats (Table 1). The diabetic rats were hypocalcemic iii comparison with the control rats. Treatment of diabetes only partially corrected serum calcium concentration (Table 1). Control rats almost doubled body weight in the 2- wk period. Although the diabetic and treated diabetic rats gained weight, the average gain per day was significantly lower than that in the control (p < 0.01) (Table 1). Treatment of diabetes only partially corrected weight gain. The poor weight gain in the treated diabetic rats occurred because good diabetic control had not been achieved. Length and weight of the small intestine were similar in control, diabetic, and treated diabetic rats (Table 1). However, weight of the intestine expressed as percent body weight was more than 25% greater in diabetic than in control rats. Measurements of the jejunum and ilellm were also not significantly different in the control and diabetic rats. Data pertaining to the effect of glucose on absorption of calcium, sodium, and water are shown in Table 2. Values are expressed on the basis of wet weight of the segments perfused. Values expressed on the basis of dry weight or length of the segments had the same relationships as those shown in Table 2. In the jejunum, in the absence of glucose from the lumen, the lumen to mucosa flux and net absorption of calcium were significantly (p < 0.05) lower in the diabetic than control rats and rate of absorption of sodium and water were not different in diabetic and Table 1. Parameters Measured in the Control, Diabetic, and Treated Diabetic Rats Studied Controls Serum Glucose (mg/dj) 165 ± 23 Calcium (mg/dj) 10.9 ± 0.3 Body weight Final (g) 94 ± 7 Gain (g/doy) 4.3 ± 0.7 Sinall intestine Wet weight g 4.5 ± 0.4 mg/cm 59 ± 2 as % body wt 5.1 ± 1.1 Jejunum Length (ern) 32 ± 2 Weight (mg/em) 65 ± 6 Ileum Length (em) 45 ± 2 Weight (rng/em) 53 ± 4 Diabetic rats 470 ± 26 a B.9 ± ± 5 a 1.8 ± ± ± ± ± 3 71 ± 5 42 ± 1 59 ± 4 Treated diabetic rats 260 ± 60 a 9.5 ± ± 3 a 2.5 ± ± ± ± ± 2 72 ± 5 45 ± 2 5B ± 4 Values represent mean ± SE. a Values in diabetic and treated diabetic rats significantly different than corresponding values in control rats (p < 0.05). control rats. The presence of glucose in the perfusion solution enhanced both lumen to mucosa flux and net absorption of calcium by -40% and absorption of sodium and water by more than twofold in all three groups of rats. Even though rate of absorption of calcium was enhanced by glucose in diabetic rats, rate of absorption remained significantly below the correspohding values in the control rats. In the ileum, absorption of calcium, sodium, and water were not significantly different in the control, diabetic, and treated diabetic rats, and presence of glucose in the lumen did not appear to have significantly enhanced rates of absorption of calcium, sodium, or water. Data pertaining to the effect of perfusion of hypotonic solution on absorption ofcaicium, sodium, and water are shown in Table 3. Values are based on wet weight of the perfused segments. Values based on dry weight or length of the segments had the same relationships shown in Table 3. In the jejunum, perfusion with the hypotonic solution appeared to have suppressed lumen to mucosa flux and net absorption of calcium in controi rats. In diabetic and treated diabetic rats, lumen to mucosa flux and net absorption of calcium Were similar in the isotonic and hypotonic perfused segments. As expected, net absorption of water and sodium were greater during perfusion of the hypotonic than isotonic solution. In the ileum, perfusion with the hypotonic solution had no influence on absorption of calcium, sodium, or water.

4 936 YOUNOSZAI ET AL. GASTROENTEROLOGY Vol. 88, No.4 Table 2. Lumen to Mucosa Flux and Net Absorption of Calcium, Sodium, and Water From Jejunum and Ileum Perfused With Physiologic Solution not Containing (-G) or Containing (+G) Glucose in Control, Diabetic, and Treated Diabetic Rats Controls Diabetic rats Treated diabetic rats Perfusion solution -G +G -G +G -G +G Jejunal absorption per hour Calcium (pmoug LM flux 5.4 ± ± ± 0.6 b 5.5 ± 0.8 o. b 4.6 ± ± Net 4.1 ± ± ± 0.6 b 4.4 ± 0.7 o. b 3.5 ± ± Sodium (JLmoUg 61 ± ± ± ± ± ± 50 0 Water (mug 0.5 ± ± ± ± ± ± 0.20 Ileal absorption per hour Calcium (pmoug LM flux 1.8 ± ± ± ± ± ± 0.5 Net 0.3 ± ± ± ± ± ± 0.5 Sodium (JLmoUg 9 ± ± ± ± ± ± 18 Water (mug -0.6 ± ± ± ± ± ± 0.2 Values represent mean ± SE. LM, lumen to mucosa. 0 Values in +G perfusion significantly greater than corresponding values in -G perfusion (p < 0.05). b Values in diabetic rats significantly lower than corresponding values in control rats (p < 0.05). Discussion In young postweaning rats as in adult rats (10) the carrier-mediated absorption of calcium is under the influence of 1,25[OH]zD 3, the most active metabolite of vitamin D. In adult diabetic (chemically induced) rats the carrier-mediated absorption of calcium from the intestine is reduced apparently because of low levels of 1,25[OH)zD3 and intestinal mucosal content of calcium-binding protein (2,11). The present investigations indicated that similar to findings in adult diabetic rats, the absorption of calcium was reduced in the jejunum of young postweaning diabetic rats. Treatment of diabetes, as in previous studies of the older rats (2), improved the diabetic condition and enhanced serum levels and intestinal absorption of calcium, probably by improving the metabolism of vitamin D (2). The results of the present study also confirmed the finding in humans, where during perfusion of the jejunum, glucose enhanced the rate of absorption of calcium (5). In a recent study of the effect of meal Table 3. Lumen to Mucosa Flux and Net Absorption of Calcium, Sodium, and Water From Jejunum and Ileum Perfused With Hypotonic or Isotonic Solutions, in Control, Diabetic, and Treated Diabetic Rats Perfusion solution Jejunal absorption per hour Calcium (JLmoUg LM flux Net Sodium (JLmoUg Water (mug ileal absorption per hour Calcium (JLmoUg LM flux Net Sodium (JLmoUg Water (mug Controls Diabetic rats Treated diabetic rats Iso Hypo Iso Hypo Iso Hypo (295 mosmollkg) (260 mosmollkg) (295 mosmol/kg) (260 mosmollkg) (295 mosmollkg) (260 mosmollkg) 4.9 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.3 Values represent mean ± SE. LM, lumen to mucosa; Iso, isotonic; Hypo, hypotonic. a Values in hypotonic perfusion significantly different from corresponding values in isotonic perfusion p < 0.05.

5 April 1985 GLUCOSE ENHANCES Ca ABSORPTION 937 composition on calcium absorption in humans, feeding of a solution containing glucose polymer caused a marked increase in absorption of calcium (12). These findings strongly suggest that glucose and its polymer have an enhancing effect on intestinal absorption of calcium. However, in an in vivo study of absorption of calcium from the duodenum, midgut, and ileum of rats, using a recirculation perfusion (1 h) technique, absorption of calcium was not altered by inclusion of glucose (10 mm) in the perfusion solution (13). The jejunum was not studied. The basis for the discrepancy between findings of the latter and of the present studies is not known but could be due to the marked differences in rate of perfusion. In the latter study perfusion rate was 2 mllmin in contrast to 0.3 mllmin in the present study. It is possible that glucose acts by increasing the permeability of calcium in the unstirred water layers and thereby its rate of absorption. The relatively high perfusion rate (2 mllmin) had reduced the thickness of the unstirred water layers and hence minimized the enhancing effect of glucose on calcium absorption in the duodenum. Whatever the mechanism by which glucose increased absorption of calcium, it seemed to have been the same in the control and the diabetic rats, as the magnitude of the enhancement in rate of absorption was the same in the control and the diabetic groups of rats. Solvent drag did not appear to be the mechanism responsible for the increased jejunal absorption of calcium induced by glucose. This was also noticed in the perfusion studies of human jejunum (5). Although rate of absorption of water increased over threefold during the hypotonic perfusions, there was no apparent enhancement in absorption of calcium. Thus the passive movement of water did not increase the absorption of calcium via a solvent drag mechanism. In fact, although jejunal absorption of water increased severalfold during the hypotonic perfusion, absorption of calcium was suppressed in control rats, and remained unchanged in the diabetic rats (Table 3). In the present study during perfusion of the glucose-containing solution, the concentration of calcium increased by only 12%. This modest increase in calcium concentration was not large enough to account for the 40% higher rate of absorption of calcium from the glucose-containing than from the glucose-free perfusion solution. In contrast, in the human study concentration of calcium during perfusion (of the glucose-containing solution) increased almost twofold. This marked rise in luminal calcium concentration was thought to have been responsible for the enhanced absorption of calcium (5). It is possible that the presence of glucose in the perfusion solution had altered the electric potential difference across the cell membrane or the villus epithelium and thus had enhanced passive absorption of calcium across the villus cell membrane. In this case passive absorption of calcium and water should have occurred through different pathways. It is also possible that calcium could have played a role in the carrier-mediated transport of glucose probably through a carrier other than the carrier for cotransport of glucose and sodium. It is interesting to note that the enhancing effect of glucose on calcium absorption was mainly observed in the jejunum. In the ileum there was minimal net absorption of calcium, sodium, and water in both the control and the diabetic rats, and absorption of calcium was not influenced by glucose or by perfusion of the hypotonic solution. In previous studies of calcium absorption in the ileum of rats, glucose as well as other sugars appeared to have enhanced calcium absorption during the first hour after instillation into ligated loops (14). In another study of calcium absorption determined indirectly from dietary balance studies and from the activity of 45Ca in the femur 24 h' after ingestion of 45Ca, glucose, galactose, fructose, and sucrose had no effect on absorption of calcium and strontium, whereas carbohydrates like cellobiose, sorbose, and ribose appeared to enhance absorption of calcium, which appeared to be dependent on the residence time of the carbohydrate in the gut, or the absorption of carbohydrate by a passive mechanism (15). The methods used for determinat~on of calcium absorption in the latter studies are not directly comparable to those in the present study. The promoting effect of glucose on calcium absorption may have clinical implications. The hypocalcemia noted in the diabetic rats could have resulted from the reduced intestinal absorption of calcium and the disturbed metabolism of vitamin D. It is possible, however, that if an abundant amount of glucose was present in the diet, it could have improved intestinal absorption of calcium. The diets of the rats in the present study contained animal and vegetable products and thus very little free glucose. References 1. Schneider LE, Schedl HP. Diabetes and intestinal calcium absorption in the rat. Am J PhysioI1972;223: Schneider LE. Schedl HP. McCain T. Haussler MR. Experimental diabetes reduces circulating 1.25-dihydroxyvitamin D in the rat. Science 1977;196: Hough S. Fausto A. Sonn Y. Dong Jo OK. Birge S1. Avioli LV. Vitamin D metabolism in the chronic streptozotocin-induced diabetic rat. Endocrinology 1983;113: Thomson ABR. Uptake of glucose into the intestine of diabetic rats: effects of variations in the effective resistance of the unstirred water layer. Diabetes 1981;30:

6 938 YOUNQSZAI ET AL. GASTROENTEROLOGY Vol. 88. No Norman OA, Morawski SG, Fordtran JS. Influence of glucose, fructose, and water movement on calcium absorption in the jejunum. Gastroenterology 1980;78: YounQszai MK, Sch~dl HP. Intestinal Ca transport: comparison of duodenum and ileum in vivo in the rat. Gastroenterology 1972;62: Younoszai MK. Ghishan FK. In vivo intestinal calcium transport in normal and growth retarded infant rats. J Nutr 1979;109: Schedl HP. Clifton JA. Small intestinal absorption of steroids. Gastroenterology 1961;41: Huntsberger DV, Leaverton PE. Statistical inference in the biomedical sciences. Boston: Allyn & Bacon. 1970; Ambrecht HI. Zenser TV. E1urns MEH. Davis BB. Effect of age on intestinal absorption and adaptation to dietary calcium. Am J PhysioI1979;236:E Schneider LE, Wilson HD. Schedl HP. Effects of alloxan diabetes on duodenal calcium binding protein in the rat. Am J PhysioI1974;227: Kelly SE. Chawla-Singh K, Sellin JH. Yasillo NJ, Rosenberg IH. Effect of meal composition on calcium absorption: enhancing effect of carbohydrate polymers. Gastroenterology 1984;87: Urban E. Pena M. Failure of lactose and glucose to influence in vivo intestinal calcium transport in normal rats. Digestion 1977;15: Vaughan OW, Filer LJ. The enhancing action of certain carbohydrates on the intestinal absorption of calcium in the rat. J Nutr 1960;71: Wasserman RH. Comar CL. Carbohydrates and gastrointestinal absorption of radiostrpntium and radiocalcium in the rat. Proc Soc Exp Bioi Med 1959;101:314-7.