GASTROENTEROLOGY

1992;102:886-894

Vitamin D and Mineral Deficiencies Increase the Plasma Membrane Calcium Pump of Chicken Intestine ROBERT H. WASSERMAN, CHRISTINA A. SMITH, NOR1 DE TALAMONI, CURTIS S. FULLMER, JOHN and RAJIV KUMAR

MARIE E. BRINDAK, T. PENNISTON,

Departments of Physiology and Pathology, College of Veterinary Medicine, Cornell University, York; and Department of Biochemistry and Molecular Biology and Department of Medicine, Division of Nephrology, Mayo Clinic and Foundation, Rochester, Minnesota

The hasolateral membrane of the enterocyte was previously shown to contain an adenosine triphosphate-dependent calcium pump. Using immunological procedures, the localization of the Ca2+ pump in chick intestine, and the effect of dietary variables on the concentration of the pump, were studied. A monoclonal antibody produced against the human erythrocyte calcium pump was shown to cross-react with a chick intestinal Ca’+ pump epitape. The most intense staining of intestinal tissue, as determined immunohistochemically, occurred at the basolateral membrane of the duodenum, jejunum, ileum, and colon, with minor staining elsewhere. By the Western blotting procedure, vitamin D repletion of vitamin D-deficient chicks was shown to significantly increase the concentration of the Ca2+ pump epitope of duodenal, jejunal, and ileal mucosa by a factor of 2-3.Chicks were also fed diets deficient in calcium or phosphorus, a situation known to result in the stimulation of the synthesis of calbindin-Dml and an enhancement of the efficiency of Ca2+ absorption. Adaptation of the chicks to these deficient diets was verified by an increase in intestinal levels of calbindin-Dal,, and is now shown to increase the Ca2+ pump epitope. From these immunological studies, it seems apparent that dietary variables that enhance intestinal Ca2+ absorption also increase the amount of the intestinal basolateral Ca’+ pump.

C

alcium, during the course of intestinal absorption, enters the enterocyte across the microvillar membrane by a diffusion-type process down a steep electrochemical gradient. The Ca2+ ion is extruded from the enterocyte, to the blood, across the basolatera1 membrane against a similar gradient of opposite polarity by an active transport process.‘*’ Consistent with an active extrusion process, an adenosine triphosphate (ATP)-dependent Ca2+ pump in basolat-

Ithaca, New

era1 membranes of rat intestine was initially identified by Nellans and Popovitch,3 Hildeman et a1.,4 and Ghijsen et al.’ The vitamin D dependency of the overall absorption of calcium is well documented,‘~2~6Lg and the stimulatory effect of vitamin D and its hormonal D [1,25(OH),D,], on form, 1,25-dihydroxyvitamin the ATP-dependent uptake of Ca2+ by basolateral membrane vesicles has been reported by several groups. *O-l4However, the validity of the data showing a stimulatory effect of vitamin D on Ca2+ pump activity, as assessed with isolated basolateral membrane vesicles, has been challenged by Van Corven et a1.15 and Van OS.’ They suggested that the Ca2+ pump activity of basolateral membranes isolated from vitamin D-deficient animals was more prone to inactivation by proteases and lipases than that from vitamin D-replete animals. The greater susceptibility of the basolateral membrane Ca2+ pump isolated from vitamin D-deficient animals, in their view, would account for the differences due to vitamin D repletion. In addition to vitamin D status, the capacity of the intestine to absorb Ca2+ is influenced by the dietary intake of calcium and phosphorus.16-27 Animals fed diets deficient in either calcium or phosphorus are known to respond to these stresses by increasing their efficiency of calcium absorption. As part of this adaptative mechanism, the production of 1,25(OH),D, is increased, the synthesis of intestinal calbindin-D,,k is stimulated, and the calcium-transport capacity of the intestine is enhanced. Somewhat contrary to expectations, Favus et al.” report that the ATP-dependent uptake of calcium by isolated intestinal basolateral membrane vesicles was diminished when the vesicles were derived from rats fed a calcium-deficient diet. 0

1992 by the American Gastroenterological 0016-5085/92/$3.00

Association

INTESTINAL CALCIUM PUMP 887

March 1992

Monoclonal antibodies produced against the human erythrocyte plasma membrane Ca2+ pump have been shown to cross-react with Ca’+-pump epitopes in mammalian intestine,28 choroid plexus,2g plaand kidney.31-33 In the present study, one of centa, these monoclonal antibodies, designated 5Fl0, was shown to cross-react with a Ca2+-pump epitope in chick intestine. This permitted an approach other than the use of isolated basolateral membrane vesicles to assess the response of the intestinal Ca2+ pump to various factors and experimental conditions. By an immunohistochemical technique, the localization of the Ca2+ pump in different segments of the intestinal tract was visualized. By Western blot methodology, the effects of vitamin D, and dietary mineral deficiencies on the relative concentrations of the intestinal Ca2+ pump epitope were also determined. Materials and Methods Experimental

Animals

One-day-old white leghorn cockerels were obtained from Clock and DeCloux, Ithaca, NY, and used at about 4 weeks of age. In the vitamin D-repletion study, l-day-old chicks were fed a vitamin D-deficient diet34 for 4 weeks. At that time, half of the chicks were given intramuscular injections of 500 IU of vitamin D, in propylene glycol 72 hours before the experiment; the other half were given vehicle alone. The dietary calcium- and phosphorus-deficiency study was initiated by feeding the l-day-old chicks a commercial chick starter diet (Agway, Ithaca, NY) for 3 weeks. At that time the animals were divided into three groups and fed one of the following diets varying in calcium and phosphorus contents for 10 additional days: (a) control diet (1.20% Ca, 0.86% P); (b) low-Ca diet (0.06% Ca, 0.86% P); or [c) low-P diet (1.20% Ca, 0.32% P). The basal diet was as previously described,35 and all diets contained 1200 IU of vitamin D, per kilogram. All chicks were fasted overnight before experiment. Blood samples were taken by cardiac puncture for determination of plasma calcium and phosphorus levels, and the animals were killed by decapitation. The excised duodena were rinsed with cold 0.9% NaCl and the mucosa was further processed for Western blot analysis.

Tissue Preparation

for Western Blot Analysis

Direct analysis. Mucosa from the vitamin D-deficient and vitamin D-replete chicks was analyzed for the presence of the Ca2+ epitope by this procedure. The mucosa was removed immediately from the underlying muscle layers by scraping with a microscope slide. Approximately 1 g of mucosa was homogenized (setting 7; Polytron, Kinematica G.m.b.H., Brinkmann Instruments, Westbury, NY) in 5 mL of a solution containing 10 mmol/L sodium ethylenediaminetetraacetate (NaEDTA), pH 7.5, for 30 seconds. Aliquots of homogenate were retained for total protein determination. Other aliquots were added di-

rectly to dissociation buffer (1% Tris, 10% j3-mercaptoethanol, 5% sodium dodecyl sulfate (SDS), and 20% glycerol; pH 6.8) and subjected to polyacrylamide gel electrophoresis (PAGE) in the presence of SDS followed by Western blotting. Partial membrane purification. Intestinal mucosa from chicks fed the low-calcium, low-phosphorus, or normal diets were homogenized as above in 5 mL of buffer A (150 mmol/L NaCl, 10 mmol/L NaEDTA, and 5 mmol/L HEPES; pH 7.4) and centrifuged at 700g for 10 minutes at 4°C (Sorvall, model RCB-B; Du Pont Co., Newton, CT). An aliquot of supernate (3.3 mL) was added to 18.7 mL of buffer A. After the addition of 12 mL of Percoll (Sigma Chemical Co., St. Louis, MO) to yield a Percoll concentration of 35%, the preparation was centrifuged at 58,600g for 1 hour (SW-28 rotor and L5-50 centrifuge; Beckman, Palo Alto, CA). The top layers of the Percoll gradient were removed, diluted with buffer A (25 mL) and centrifuged at 31,OOOgfor 30 minutes. The supernate was discarded and the pellet suspended in 2 mL of 10 mmol/NaEDTA, pH 7.5, to a final concentration of approximately 3.4 g/mL. An aliquot of this suspension was added to the dissociation buffer and further processed for Western blotting. Basolateral membrane isolation. Basolateral membranes were isolated from the chicks fed the diets varying in calcium and phosphorus contents. Modifications of the procedures of Mircheff and Wright36 and Steck,37 as described by Meyer,38 were used. Isolated duodenal enterocytes were obtained mechanically after filling the rinsed duodenal loops with isolation medium (118 mmol/L NaCl, 1.5 mmol/L KCl, 5.3 mmol/L KH,PO,, and 8.3 mmol/L Na,HPO,; pH 7.4; plus 0.2 mmol/L phenylmethylsulfonyl fluoride, 1 mL/L aprotinin, and 1 mg/mL hyaluronidase) and incubating at 37’C for 30 minutes. Cells were collected and washed two times by suspending in isolation medium minus hyaluronidase, followed by centrifugation at 200g for 4 minutes. Enterocytes were homogenized in 10 mmol/Na HEPES, pH 7.4, with 10 strokes of a Dounce homogenizer. Tonicity was restored by the addition of sorbito1 plus NaCl stock solution to yield final concentrations of 250 mmol/L sorbitol, 12.5 mmol/L NaCl, and 6.7 mmol/L Na HEPES. Unbroken cells, nuclei, and large aggregates were removed by centrifugation at 200g for 4 minutes. Supernates were centrifuged at 750g for 10 minutes, and the pellets (crude basolateral sheets) were resuspended in fractionation buffer (250 mmol/L sorbitol, 12.5 mmol/L NaCl, and 5 mmol/L Na HEPES; pH 7.4) and homogenized with a Dounce homogenizer. Further enrichment was achieved by sorbitol density-gradient centrifugation. Purified basolateral membranes were collected at the interface between the 36% and 47.5% sorbitol layers. Na+,K+-adenosine triphosphatase (Na+,K+-ATPase), a marker enzyme for plasma membrane, was enriched about 12-fold compared with the homogenates. The purification factors were not significantly affected by the different diets. Western blot procedure. Samples of intestinal homogenate, partially purified membranes, or isolated basolateral membranes of known protein concentration were diluted with an equal volume of dissociation buffer and heated in a boiling water bath for 5 minutes. Five microliters of 0.1% bromophenol blue tracking dye was added. The proteins

888 WASSERMAN ET AL.

were separated by SDS PAGE (7% acrylamide, 0.1% SDS) by the method of Laemellis3’Protein standards of known

molecular weight (prestained standards; Bio-Rad Laboratories, Richmond, CA) were run on the same gel. The separated proteins were transferred to nitrocellulose sheets (Millipore Corp., Bedford, MA) using the procedure of Towbin et aL4’The blots were incubated for z hours at 37°C in 100 mL of blocking solution containing 10% newborn calf serum and 3% bovine serum albumin in 0.12mol/L NaCl, 4.7 mmol/L KCl, and 0.0137 mol/L Tris, pH 7.4. After rinsing twice in Tris-buffered saline (TBS) containing 0.02 mol/L Tris and 0.5 mol/L NaCl, pH 7.4, the blots were incubated in 70 mL of TBS containing mouse anti-Cazf-pump antibody (1:500 dilution) at room temperature for 2 hours. The blots were rinsed five times for 5 minutes each, incubated in 70 mL of the above-described blocking solution containing peroxidase-conjugated goat anti-mouse immunoglobulin G (diluted 1:5000; horseradish peroxidase conjugate substrate kit, Bio-Rad) and again washed five times for 5 minutes each with TBS. The peroxidase substrate for color development was 4chloro-l-napthol in the presence of hydrogen peroxide. Densitometric analysis was performed using the UltraScan XL enhanced laser densitometer (Pharmacia LKB, Piscataway, NJ) and an image-analysis computer program (ImagePro, Media Cybernetics, Silver Spring, MD). Immunohistochemical localization of the Caz+ pump. Ca’+ pump localization was performed by indirect immunofluorescent staining. Intestinal tissue was rapidly excised from animals killed by decapitation and immediately snap-frozen. Two-micrometer cryostat sections were obtained, air dried for 10 minutes, and fixed in 95% ethanol for 10 minutes followed by washing twice with 0.01 mol/L phosphate-buffered saline (PBS), pH 7.2. The sections were incubated for 30 minutes at room temperature with the anti-Ca’+-pump monoclonal antibody 5F10 diluted 1:50. The sections were then washed three times with PBS for 5 minutes each and incubated for 30 minutes at room temperature with goat fluorescein isothiocyanateconjugated anti-mouse immunoglobulins (Organon Teknika Corp., Durham, NC). The sections were washed three times with PBS and mounted in polyvinyl alcohol (Gelvatol-Monsanto, Springfield, MA) containing triethylenediamine to retard fading of fluorescence. The localization of fluorescence was observed using a Leitz (Rockleigh, NJ) ortholux incident light fluorescent microscope. All incubations were carried out in a humid atmosphere. Control sections were obtained by incubating corresponding sections with BALB/c ascitic fluid in place of the primary antibody. The control sections showed no detectable fluorescence. Other analytical procedures. Protein concentrations were determined using the procedure of Lowry et al.*l adapted to an autoanalyzer. Plasma calcium levels were quantified by atomic absorption spectrometry (model 360; Perkin-Elmer, Norwalk, CT), and plasma phosphorus levels were measured by an automated calorimetric method described by the analyzer manufacturer (Technicon, Tarrytown, NY). Duodenal levels of calbindin-DZBk were determined using a radial immunodiffusion technique.42 Statistical analysis. Student’s t test or, when appro-

GASTROENTEROLOGY Vol. 102. No. 3

priate, analysis of variance was performed using a computer program (Minitab, Inc., State College, PA). Duncan’s multiple range test43 was performed for multiple comparisons of the means.

Results Immunohistochemical

localization

of the Ca2+

in various segments of the chick intestine using the monoclonal antibody 5FlO is shown in Figure 1. Pronounced staining in each segment occurs at the basolateral membrane with lighter, diffuse staining in other parts of the intestinal cell. This pattern of localization is essentially the same as previously reported for rat intestine. ” In their study Borke et a1.28 readily noted differences in staining intensity among the three segments of the small intestine, the most intense staining occurring in the duodenum and the least in the ileum. Such readily observable differences in staining intensities among the various intestinal segments were not as apparent in the chick small intestine.

pump

Figure 1. Localization of Ca’+-pump epitope in different segments of the intestinal tract. Immunohistochemical localization of the Ca*+-pump epitope in chick intestine: (A)duodenum; (B) jejunum; (C) ileum: (D] colon. Note heavy staining on the hasolatera1 membranes (single arrows) and the staining of a jejunal villus (B)cut in cross section (double arrow) (bar = 30 pm).

INTESTINAL

March 1992

Western blot analysis of duodenal mucosal homogenates (direct analysis procedure) from vitamin Ddeficient and vitamin D-replete chicks is shown in Figure 2. The major polypeptides reacting with the anti-Ca’+-pump antibody are those primarily associated with a relatively diffuse band at about 175 kilodaltons and a relatively sharp band at about 141 kilodaltons. Minor bands associated with smaller polypeptides can also be seen. For comparison, the pattern of purified human red cell Ca2+ pump44 is also shown in Figure 2 where a diffuse band corresponding to 154 kilodaltons, a sharp band at 130 kilodaltons, and several minor bands are evident. The chick intestinal pattern and the erythrocyte pattern appear similar, but differences in the electrophoretic mobility of the prominent bands are evident. The sharp bands in each pattern presumably represent the monomeric Ca’+-pump polypeptides. The diffuse bands and the minor bands are probably aggregation products and/or proteolytic degradation products of the pump, as previously suggested for the human red cell Ca2+ pump by Niggli et a1.45The ability of monomers of the Ca2+ pump to dimerize was recently reported46,47 and the intermolecular binding region of the Ca2+ pump was suggested to be associated with calmodulin-binding sites.47 Multiple bands of the Ca2+ pump have also been observed in Western blots of rat intestine,” pig stomach muscle,46 human placenta,4g and human and rat kidney.3*,32 Vitamin D, repletion of vitamin D-deficient chicks, as shown in Figure 2, clearly increases the intensity of the duodenal Ca2+-pump epitope bands. Densitometric measurements (Figure 3) of the diffuse band at 175 kilodaltons and the sharper band at

Figure 2. Effect of vitamin D, on duodenal Ca’+-pump epitope. Western blot analysis of monoclonal antibody binding to purified human erythrocyte Ca’+ pump and to duodenal mucosal homogenates from vitamin D-deficient and vitamin D-repleted chicks. The homogenates were from three -D chicks and three +D chicks. The human erythrocyte Ca’+ pump was purified by the procedure Penniston et a1.44The amount of protein applied to the gels was 0.85 f 0.04 mg and 0.94 rtr0.12 mg for the -D and +D samples, respectively. The +D chicks were injected intramuscularly with 500 IU vitamin D3 72 hours before experiment.

205 175 141 116.5 77

46.5

CALCIUM

PUMP

889

141 kilodaltons, shown in Figure 2, provide an estimate of the approximate differences in band intensities. On average, there was an approximately twofold to threefold increase in the density of the Ca2+-pump bands by vitamin D, repletion. The Western blot patterns shown in Figure 4 and its corresponding densitometric analysis (Figure 5) show that vitamin D, repletion increases the Ca2+pump epitope in each segment of the chick small intestine, i.e., the duodenum, jejunum, and ileum. Again, there was a twofold to threefold increase due to vitamin D, repletion. The plasma calcium and phosphorus levels of the chicks adapted to the mineral-deficient diets are given in Table 1. Those chicks fed the low-calcium diet showed a significantly diminished plasma calcium level compared with the control animals, whereas those fed the low-phosphorus diet were normocalcemic and hypophosphatemic. The considerable and significant increase in intestinal calbindin-D26k concentration (Table 1) verifies that the chicks had indeed adapted to the two mineral-deficient diets.‘g,26 The process of adaptation is accompanied by an increase in the concentration of the duodenal Ca2+pump epitope as shown by Western blot analysis (Figures 6 and 7). The apparent molecular weights of the diffuse and sharper band of the purified basolateral membranes were 151 and 125 kilodaltons, respectively, which closely correspond to the apparent molecular weights of the purified red cell epitope as observed in the present study (Figure 2). The apparent molecular weights of the diffuse and sharper bands of the partially purified membranes were approxi-

890

WASSERMAN

ET AL.

GASTROENTEROLOGY

-I-

ox 2.00

m 4 m +o

6 f 1.50 $ 9 P) 1.00 .$ 0 z

0.50

0.00 141 kDa

175 kDa

Figure 3. Effect of vitamin D, on the duodenal Ca’+-pump epitape. Densitometric assessment of the 141- and 175-kilodalton bands of Figure 2. For the 14l-kilodalton band, the mean f SE of the -D and +D bands are 0.37 f 0.09 and 0.97 f 0.13, respectively, with a P value of 0.03. For the 175-kilodalton band, the mean +SE of the -D and tD bands are 0.56 f 0.11 and 1.79 + 0.13, respectively, with a P value of 0.01.

mately 175 and 141 kilodaltons, respectively, corresponding closely to those determined by the direct analysis procedure (Figures 2 and 4). Western blots from mucosal tissue prepared by the direct analysis procedure (two experiments in duplicate) also showed that each mineral-deficient diet increased the density of the Ca’+-pump epitope by a factor of approximately 2 (not shown). Discussion The present observations clearly show that the avian intestine contains an epitope of the human

+D

-D

+D -D

+D

Vol. 102. No. 3

red cell plasma membrane Ca2+ pump. Immunohistochemically, the epitope was primarily associated with the basolateral membrane of intestinal cells of the duodenum, jejunum, ileum, and colon. Western blot analysis of crude preparations of the small intestinal segments revealed major epitopes associated with a diffuse band with an average apparent molecular weight of approximately 175 kilodaltons and a sharper band with an average molecular weight of approximately 141 kilodaltons, the latter corresponding to the approximate molecular weight of the human erythrocyte plasma membrane CaATPase pump as reported by Graf et aL5’ Of interest, the apparent molecular weights of the diffuse and sharper bands of the epitopes of the purified basolateral membranes were similar to those of the purified erythrocyte Ca’+-pump epitope as determined in the present study. It was further shown that the concentration of the Ca’+-pump epitope in all segments of the small intestine was increased by vitamin D repletion of vitamin D-deficient chicks. These observations are consistent with previous reports indicating that vitamin D or 1,25(OH),D, increases the ATP-dependent uptake of Ca’+ by isolated basolateral membrane vesicles.*“-*4 In contrast to these studies, Van Corven et a1.15 and Van 02 have proposed that the observed stimulatory effect of vitamin D on Ca’+-pump activity, using isolated membrane vesicles, is an artifact. It was their contention that the Ca2+ pump associated with the intestinal basolateral membranes from rachitic animals was more prone to degradation and

-D

c-

175 141

DUOD

ILE

Figure 4. Effect of vitamin D, on the Ca’+pump epitope in segments of small intestine. Western blot analysis of monoclonal antibody binding to mucosal homogenates of different segments of small intestine from vitamin D-deficient and vitamin D-repleted chicks. The amount of protein, in milligrams, applied to the gels for the -D group was 0.75 (duodenum), 0.60 (jejunum), and 0.76 (ileum), and, for the +D group, 0.60 (duodenum), 0.78 (jejunum), and 0.85 (ileum]. A duplicate experiment yieldedsimilar electrophoretic and Western blot patterns. The +D chicks were injected with 500 IU vitamin D, 72 hours before the experiment.

INTESTINAL CALCIUM PUMP

March 1992

phosphorus also results in an increased net synthesis of the duodenal Ca2+ pump. These results further implicate 1,25(OH),D, as the ultimate effector. Favus et al.‘l recently confirmed that the administration of 1,25(OH),D, to vitamin D-deficient rats does result in an enhanced ATP-dependent uptake of Ca2+ by isolated duodenal basolateral membranes. However, they also report that the Ca2+ pump activity in rats adapted to a low-calcium diet is decreased, contrary to the present observations. Unlike the several studies showing a stimulatory effect of vitamin D on the ATP-dependent Ca2+ uptake by isolated basolateral membrane vesicles, the report of Favus et al.” is the only other study that came to our attention in which the response of Ca2+ pump activity to dietary calcium deficiency was examined. Because of the paucity of information on this point and the discrepancy between the present immunological data and the biochemical data of Favus et al.,” it is apparent that additional experimentation is required to sort out these differences. In addition, these studies show that the basolateral membrane purification is accompanied by changes in the electrophoretic mobility and, thus, the apparent molecular size of pump polypeptides. The underlying reason for these changes and the relationship of these changes to Ca2+-pump activity are not known at present. In addition, the nature of the multiple bands requires further study. As mentioned, some are thought to be degradation products and others aggregation products. The latter suggestion is supported by the ability of monomers of the Ca2+ pump, and probably large degradative fragments, to dimerize.47 The possibility that the larger-molecularweight bands might represent alternate splicing of the Ca2+ pump messenger RNA (mRNA) should also be considered. The mechanism by which vitamin D, stimulates the net synthesis of the Ca2+ pump is a significant problem to address, The vitamin D, hormone 1,25(OH),D, is known to elicit genomic effects on the intestine, and a prominent genomic response is the induction of the de novo synthesis of avian intestinal

c

1.50 2 1.00 : 9 & 0.50 a 2 0.00 : ‘g

891

141 kDa

1.60

z! 1.20

0.60

0.40

0.00 DUODENUM

JEJUNUM

ILEUM

Figure 5. Effect of vitamin D, on the Ca’+-pump epitope in segments of small intestine. Densitometric assessment of the 141and l%kilodalton hands in Figure 4.

inactivation by endogenous proteases and lipases during preparation than those from vitamin D-replete animals. This argument, based on their experimental evidence, is difficult to put aside when isolated basolateral membranes were used to assess Ca’+-pump activity in vitro. In the present immunological experiments, however, intestinal homogenates were directly analyzed for Ca2+-pump epitopes without lengthy membrane purification. From these observations, it is concluded that vitamin D does indeed increase the intestinal concentration of the basolateral Ca2+ pump in vitamin D-deficient animals. The present experiments further show that adaptation of chicks to diets deficient in either calcium or

Table 1. Plasma Calcium and Phosphorus and intestinal Calbindin-D,,, Levels on Chicks Fed Diets Deficient in Calcium and Phosphorus Diet (%) Group

Ca

P

Plasma Ca (mg/dLl

Plasma P (mg/dL)

Intestinal calbindin-D),,, (pg/mg supernatant protein)

Control Low Ca Low P

1.2 0.06 1.2

0.86 0.86 0.32

11.49 f 0.13 10.57 + 0.15” 11.33 f 0.30

6.01 + 0.09 5.18 f 0.13b 4.39 * 0.14b

55.0 + 4.0 146.6 + 7.5b 144.8 + 7.1b

NOTE. Values are means + SE. For plasma Ca and P, n = 15; for calbindin-D,,, , n = 16. “The value of the low-Ca group differs from the control and low P group values at P < 0.01. bThe values of the low-Ca and low-P group differ from the control group values at P < 0.01.

892

WASSERMAN

ET AL.

GASTROENTEROLOGY

Vol. 102, No. 3

-205

4-151 '-125

-116.5 -77

e46.5

N

Lo

Ca

N

Lo P

Lo Ca Lo P

Figure 6. Effect of Ca and P deficiencies on Ca’+-pump epitope. Western blot analysis of monoclonal antibody binding to duodenal mucosal preparations from chicks adapted to a low-calcium or low-phosphorus diet. The partially purified membranes (B) and the isolated basolateral membranes (A) were prepared as described in Materials and Methods. The amounts of protein of the partially purified membranes applied to the gel were &50,0.48, and 0.53 mg for the normal, low-Ca’+, and low-P groups, respectively. For the isolated basolateral membranes, the protein applied was uniformly 26 pg. Dietary protocol is described in Materials and Methods.

mRNA and calbindin-D,,, .2,17,27,51-54 calbindin-D,,, in Calbindin-D,,, , which is virtually undetectable the intestine of vitamin D deficient animals, appears some 3-4 hours after repletion with 1,25(OH),D,.55 Unlike calbindin-D,,, , the Ca’+-pump epitope is readily detectable in the vitamin D-deficient chick. Whether the increase in the concentration of the Ca2+ pump is controlled at the transcriptional or the translational level, or both, requires an assessment of Ca2+-pump mRNA concentrations under conditions of vitamin D deficiency and repletion. There is the possibility that the additional pump units might result from the transfer of pre-existing

cytosolic pump units to the plasma membrane surface. This recruitment suggestion derives from the reported insulin-dependent translocation of cytosolic transporters to the cell surface of fat cellP and the vasopressin-dependent translocation of water channels to the plasma membrane of toad urinary bladdera5’ Recently, the movement of Ca2+ across the intestinal epithelium was visualized by ion-microscopic imaging of a stable isotope of calcium, 44Ca2f.58 In vitamin D-deficient animals, Ca2+ appeared to readily cross the brush border membrane, binding to intracellular components that lie subjacent to the

I

NORMAL

Lo Co

Lo P

NORMAL

Lo Co

Lo P

--I

Figure 7. Effect of Ca and P deficiencies on the duodenal Ca*+-pump epitope. Densitometric assessment of the 141- and 176-kilodalton bands in Figure 6.

INTESTINAL CALCIUM PUMP

March 1992

membrane, and only diffusing slowly away from this region. Events occurring in this region may very well represent a critical step in transcellular Ca’+ transport. After vitamin D, treatment of the deficient animal, the transfer of Ca2+ away from the brush border region occurs rapidly, possibly as a result of the appearance in the cytosol of the soluble vitamin D,-induced high-affinity calcium-binding protein calbindin-Dzak. As proposed by Kretsinger et aL5’ and supported by the in vitro modeling experiments of Feher et a1.,60*61 cytosolic calbindin-D,,, might serve to facilitate the movement of Ca2+ from the brush border region to the site of the extruding Ca2+ pump on the basolateral membrane. The Ca2+ pump is now unequivocally shown to increase in concentration by vitamin D, repletion of vitamin D-deficient animals and in chicks adapted to calcium- or phosphorus-deficient diets. This seems physiologically reasonable because of the stimulating effect of vitamin D, on processes that occur in the transport path before extrusion of Ca2+ by the Ca2+ pump. Addendum In a recent publication, Zelinski et al. report that 1,25(OH),D, increases intestinal Ca2+ pump mRNA concentration in vitamin D-deficient rats. Our preliminary studies suggest that vitamin D likewise increases Ca2+ pump mRNA concentration in rachitic chick duodenum.

11.

12.

Van OS CH. Transcellular calcium transport in intestinal and renal epithelial cells. Biochim Biophys Acta 1987;906:195-

14.

15.

16.

17.

18.

19.

20.

21.

222. 2.

3.

4.

5.

6.

7.

8. 9. 10.

Wasserman RH, Fullmer CS. On the molecular mechanism of intestinal calcium transport. In: Dintzis FR, Laszlo FR, eds. Mineral absorption in the monogastric GI tract. Advances in experimental medical biology. Volume 249. New York: Plenum, 1989:46-65. Nellans HN, Popovitch JE. Calmodulin-regulated, ATPdriven calcium transport by basolateral membranes of rat small intestine. J Biol Chem 1981;256:9932-9936. Hildemann BA, Schmidt A, Murer H. Ca++-transport across basal-lateral plasma membranes from rat small intestinal epithelial cells. J Membran Biol 1982;65:55-62. Ghijsen WEJM, DeJong, MD, Van OS CH. ATP-dependent calcium transport and its correlation with Ca’+-ATPase activity in basolateral membranes of rat duodenum. Biochem Biophys Acta 1982;689:327-336. Bronner F, Pansu D, Stein WD. An analysis of intestinal calcium transport across the rat intestine. Am J Physiol 1986;250:G561-G569. Favus MJ. Factors that influence absorption and secretion of calcium in the small intestine and colon. Am J Physiol 1985;248:G147-G157. Norman AW. Studies on the vitamin D endocrine system in the avian. J Nutr 1987;117:797-807. DeLuca HF. Vitamin D-dependent calcium transport. Sot Gen Physiol Ser 1985;39:159-176. Chandler JS, Meyer SA, Wasserman RH. Vitamin D-dependent, ATP-driven Ca*+-transport in chick duodenal basal-lateral membrane vesicles. In: Norman AW, Schaefer K, Grigo-

leit HG, Herrath D, eds. Vitamin D. Chemical, biochemical and clinical update. Berlin: Walter de Gruyter, 1985:408-409. Favus MJ, Tenbe V, Ambrosic KA, Nellans HN. Effects of 1,25(OH),D, on enterocyte basolateral membrane calcium transport in rats. Am J Physiol 1989;256:G613-G617. Ghijsen WEJM, Van OS CH. 1,25-Dihydroxyvitamin D, regulates ATP-dependent calcium transport in basolateral membranes of rat enterocytes. Biochim Biophys Acta 1982;689: 170-172.

13.

References 1.

893

22.

23.

24.

25.

26.

27.

28.

Takito J, Shinki T, Sasaki T, Suda T. Calcium uptake by brush-border and basolateral membrane vesicles in chick duodenum. Am J Physiol 1990;258:G16-G23. Walters JR, Weiser MM. Calcium transport by rat duodenal villus and crypt basolateral membranes. Am J Physiol 1987;252:G170-G177. Van Corven EJM, De Jong MD, Van OS CH. The adenosine triphosphate-dependent calcium pump in rat small intestine: effects of vitamin D deficiency and cell isolation methods. Endocrinology 1987;120:868-873. Armbrecht HJ, Zenser TV, Bruns ME, Davis BB. Effect of age on intestinal calcium absorption and adaptation to dietary calcium. Am J Physiol 1979;236:E769-E774. Bar A, Hurwitz S. The interaction between dietary calcium and gonadal hormones in their effect on plasma calcium, bone, 25-hydroxycholecalciferol-l-hydroxylase, and duodenal calcium-binding protein measured by a radio-immunoassay in chicks. Endocrinology 1979;104:1455-1460. Bar A, Wasserman RH. Control of calcium absorption and intestinal calcium-binding protein synthesis. Biochem Biophys Res Commun 1973;54:191-196. Fox J, Pickard DW, Care AD, Murray TM. Effect of low phosphorus diets on intestinal calcium absorption and the concentration of calcium-binding protein in intact and parathyroidectomized pigs. J Endocrinol 1978;78:379-387. Morrissey RL, Wasserman RH. Calcium absorption and calcium-binding protein in chicks on differing calcium and phosphorus intakes. Am J Physiol1971;220:1509-1515. Pansu D, Bellaton C, Bronner F. Effect of Ca intake on saturable and nonsaturable components of duodenal Ca transport. Am J Physiol 1981;240:G32-G37. Rader JI, Baylink DJ. Hughes M, Safilian EF, Haussler MR. Calcium and phosphorus deficiency in rats: Effects of PTH and 1,25dihydroxyvitamin D,. Am J Physiol 1979;236:E118E122. Ribovich ML, DeLuca HF. The influence of dietary calcium and phosphorus on intestinal calcium transport in rats given vitamin D metbolites. Arch Biochem Biophys 1975;170:529535. Ribovich ML, DeLuca HF. Effect of dietary calcium and phosphorus on intestinal calcium absorption and vitamin D metabolism. Arch Biochem Biophys 1978;188:145-156. Swaminathan R, Sommerville BS, Care AD. The effect of dietary calcium on the activity of 25-hydroxycholecalciferol-l-hydroxylase and Ca absorption in vitamin D-replete chicks. Br J Nutr 1977;38:47-54. Thomasset M, Cuisinier-Gleizes P, Mathieu H. Differences in duodenal calcium-binding protein (CaBP) in response to a low-calcium or a low-phosphorus intake. Calcif Tissue Res 1977;22s:45-50. Friedlander EJ, Henry HL, Norman AW. Studies on the mode of action of calciferol. Effects of dietary calcium and phosphorus on the relationship between the 25-hydroxyvitamin D,-l-hydroxylase and production of chick intestinal calcium binding protein. J Biol Chem 1977;252:8677-8683. Borke JL, Caride A, Verma AK, Penniston JT, Kumar R. Cellular and segmental distribution of Ca’+-pump epitopes in rat intestine. Pflugers Arch 1990;417:120-122.

894

WASSERMAN

ET AL.

29. Borke JL, Caride AJ, Yaksh TL, Penniston JT, Kumar R. Cerebrospinal fluid calcium homeostasis: evidence for a plasma membrane Ca*+-pump in mammalian choroid plexus. Brain Research 1989;489:355-360. 30. Borke JL, Caride A, Verma AK, Kelley LK, Smith CH, Penniston JT, Kumar R. Calcium pump epitopes in placental trophoblast basal plasma membranes. Am J Physiol 1989;257:C341C346. 31. Borke JL, Minami J, Verma A, Penniston JT, Kumar R. Monoclonal antibodies to human erythrocyte membrane Ca++Mg++ adenosine triphosphatase pump recognize an epitope in the basolateral membrane of human kidney distal tubule cells, J Clin Invest 1987;80:1225-1231, 32. Borke JL, Minami J, Verma AK, Penniston JT, Kumar R. Co-localization of erythrocyte Ca++-Mg++ ATPase and vitamin Ddependent 28-kilodalton-calcium binding protein in the cells of human kidney distal tubules. Kidney Int 1988;34:262-267. 33. Borke JL, Caride A, Verma AK, Penniston JT, Kumar R. Plasma membrane calcium pump and 28-kDa calcium binding protein in cells of rat kidney distal tubules. Am J Physiol 1989;257:F842-F849. 34. Mykkanen HM, Wasserman RH. Effect of vitamin D on the intestinal absorption of *“Pb and 47Ca in chicks. J Nutr 1982;112:520-527. 35. Mykkanen HM, Wasserman RH. Gastrointestinal absorption of lead (*03Pb)in chicks. Influence of lead, calcium and age. J Nutr 1981;111:1757-1765. 36. Mircheff AK, Wright EM. Analytical isolation of plasma membranes of intestinal epithelial cells: Identification of Na,K-ATPase rich membranes and the distribution of enzyme activities. J Membr Biol 1976;28:309-333. 37.Steck TL. Preparation of impermeable inside-out and rightside-out vesicles from erythrocyte membranes. Methods Membrane Biol 1974;2:245-281. 38. Meyer SA. Calcium transport by isolated basolateral membranes from chick duodenum. Ph.D. thesis, Cornell University, Ithaca, NY, 1984. 39. Laemelli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage R.,. Nature 1970;227: 680-685. 40. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc Nat1 Acad Sci USA 1979;76:4350-4354. 41. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-275. 42. Taylor AN. Chick brain calcium-binding protein: comparison with intestinal vitamin D-induced calcium-binding protein. Arch Biochem Biophys 1974;161:100-108. 43. Duncan DB. Multiple range and multiple F tests. Biometrics 1955;82:70-77. 44. Penniston JT, Filoteo AG, McDonough CS, Carafoli E. Purification, reconstitution and regulation of plasma membrane Ca’+ pumps. Meth Enzymol1988;157:340-351. 45. Niggli V, Penniston JT, Carafoli, E. Purification of the Ca’+MgZ+ ATPase from human erythrocyte membranes using a calmodulin affinity column. J Biol Chem 1979;254:9955-9958. 46. Kosk-Kosicka D, Bzdega T, Wawrzynow A. Fluorescence energy transfer studies of purified erythrocyte Ca’+-ATPase. Ca’+-regulated activation by oligomerization. J Biol Chem 1989;264:19495-19499. 47. Vorherr T, Kessler T, Hofmann F, Carafoli E. The calmodulinbinding domain mediates the self-association of the plasma membrane Ca*+ pump. J Biol Chem 1991;266:22-27. 48. de Jaegere S, Wuytack F, Eggermont JA, Herboomen H, Casteels R. Molecular cloning and sequencing of the plasma-

GASTROENTEROLOGY

Vol. 102,No. 3

membrane Ca’+ pump of pig smooth muscle. Biochem J 1990;271:655-660. 49. Kelley LK, Borke JL, Verma AK, Kumar R, Penniston JT, Smith CH. The calcium-transporting ATPase and the calcium- or magnesium-dependent nucleotide phosphatase activities of human placental trophoblast basal plasma membrane are separate enzyme activities. J Biol Chem 1990;265:5453-5459. 50. Graf E, Verma AK, Gorski JP, Lopaschuk G, Niggli V, Zurini M, Carafoli E, Penniston JT. Molecular properties of calcium pumping ATPase from human erythrocytes. Biochemistry 1982;21:4511-4516. 51. Bishop CW, Kendrick NC, DeLuca HF. The early time course of calcium-binding protein induction by 1,25-dihydroxyvitamin D, as determined by computer analysis of two-dimensional electrophoresis gels. J Biol Chem 1984;259:3355-3360. 52. Corradino RA. Embryonic chick intestine in organ culture. A unique system for the study of the intestinal calcium absorptive mechanism. J Cell Biol 1973;58:64-78. 53. Fullmer CS. Regulation of intestinal calbindin-D,,k gene exstudy. Arch Biochem pression: a solution hybridization Biophys 1990;283:193-199. 54. Meyer J, Fullmer CS, Wasserman RH, Komm BS, Haussler MR. Dietary restriction of calcium and phosphorus elicit distinctly different patterns of regulation of the mRNAs for avian intestinal calcium binding protein and 1,25-(OH),D receptor (abstr). J Bone Miner Res 1989;48:263. 55. Wasserman RH, Brindak ME, Meyer SA, Fullmer CS. Evidence for multiple effects of vitamin D, on calcium absorption: Response of rachitic chicks with or without partial vitamin D, repletion, to 1,25_dihydroxyvitamin D,. Proc Nat1 Acad Sci 1982;79:7939-7943. 56. Suzuki S, Kono T. Evidence that insulin causes translocation of glucose transport activity to the plasma membrane from an intracellular storage site. Proc Nat1 Acad Sci USA 1980;77: 2542-2545. 57. Shi LB, Verkan AS. Very high water permeability in vasopressin-induced endocytic vesicles of toad urinary bladder. J Gen Physiol 1989;94:1101-1115. 58. Chandra S, Fullmer CS, Smith CA, Wasserman RH, Morrison GH. Ion microscopic imaging of calcium transport in the intestinal tissue of vitamin D-deficient and vitamin D-replete chickens: a %a stable isotope study. Proc Nat Acad Sci USA 1990;87:5715-5719. 59. Kretsinger RH, Mann JE, Simmonds JG. Model of facilitated diffusion of calcium by the intestinal calcium binding protein. In: Norman AW, Schaefer K, von Herrath D, Grigoleit H-G, eds. Vitamin D, chemical, biochemical and clinical endocrinology of calcium metabolism. Berlin: de Gruyter, 1982:233248. 60. Feher JJ. Facilitated calcium diffusion by intestinal calciumbinding protein. Am J Physiol 1983;244:C303-C307. 61. Feher JJ, Fullmer CS, Fritzsch GK. Comparison of the enhanced steady-state diffusion of calcium by calbindin-D,, and calmodulin. Possible importance in intestinal calcium absorption. Cell Calcium 1989;10:189-203. 62. Zelinski JM, et al. Biochem Biophys Res Commun 1991; 179:749-755.

Received May 7,1991.Accepted July 30,1991. Address requests for reprints to: Robert H. Wasserman, Ph.D., Department of Physiology, VRT717, Cornell University, Ithaca, New York 14853. Supported by National Institutes of Health grants DK-04652 (Cornell) and GM-28835 (Mayo Clinic). The authors thank N. Jayne for manuscript preparation and Francis Davis for technical assistance.