Penniston, John T.

The variable of age was used to study macromolecules in bone that may mediate in part the in vivo readily exchangeable calcium-binding capacity (VCa2+D). Organic components were extracted from nonmineralized bone with 4 M guanidine-HCl and from both nonmineralized and mineralized bone with 0.1 M EDTA. The composition of pup bone extracts demonstrated an enrichment in protein, hexuronate, sialic acid, organic phosphorus, and bound sulfate when compared with other age groups. In vitro calcium-binding studies identified low-affinity (Kd congruent to 10(-3) M) sites in both types of extracts; high-affinity sites (Kd congruent to 10(-5) M) were only evident in EDTA extracts of bone. Readily exchangeable calcium-binding capacity in vivo was found to decrease from pup (40.7 mM) to adolescent (11.1 mM) to the mature/old groups (2.6/1.2 mM); however, a large difference in low-affinity site number was only observed between pup and adolescent bone extracts. The overall organic composition of EDTA and guanidine-HCl extracts generally reflected the composition of total bone, which dropped dramatically on a dry weight basis from pup to adolescent groups. A similar pattern was observed with the number of low-affinity binding sites measured in vitro. In vitro binding data indicate that nonmineralized matrix of pup bone, extractable by 4 M guanidine-HCl, possesses enough capacity to accommodate approximately 40% of the readily exchangeable pool. As age progresses, other components of the blood-bone exchange process, such as vascularity, may reduce the readily exchangeable calcium pool size below the amount of low-affinity sites measured in vitro.

The syncytiotrophoblast represents the primary cellular barrier between maternal and fetal circulations in the placenta. Large amounts of Ca2+ are transported across this barrier by mechanisms that are not clearly understood. To further understand this phenomenon, we examined rat and human placenta by immunohistochemical and protein blotting techniques with a monoclonal antibody raised against the human erythrocyte plasma membrane Ca2+ pump. Immunohistochemistry with this antibody showed specific staining in the human placenta of the basal (fetal facing) surface of the syncytiotrophoblast. In the rat placenta, immunohistochemistry also showed specific staining of the innermost (fetal facing) layer of the trophoblast and the basal surface of the endoderm of the intraplacental yolk sac. In Western blots of placental homogenates and membranes, the monoclonal antibody bound to a 140,000-mol wt band, characteristic of Ca2+ pumps in other tissues. Western blots of isolated basal membranes showed more intense staining than isolated microvillous membranes, confirming the results of the immunohistochemistry. In addition, Ca2+ transport in basal membrane vesicles from human placenta was inhibited by polyclonal antibodies prepared against the erythrocyte Ca2+ pump. We conclude that basal (fetal facing) layers of human and rat placentas contain a high-affinity Ca2+ pump situated to transport Ca2+ from the maternal to the fetal circulation. calcium-magnesium-adenosinetriphosphatase; calcium transport; immunohistochemistry; rat and human placenta Submitted on November 14, 1988 Accepted on March 27, 1989

We have studied localization of mRNAs coding isozymes of rat plasma membrane Ca(2+)-adenosinetriphosphatase pump (rPMCA) in the rat kidney, with use of reverse transcription (RT) with subsequent amplification by polymerase chain reaction (PCR). When zones of the kidney were separated by macrodissection, a large amount of mRNA coding isozyme rPMCA1 was found in all zones; mRNA for isozyme rPMCA2 was abundant in cortex and in outer medulla, and mRNA for isozyme rPMCA3 was prominent in outer medulla. The mRNAs were analyzed in microdissected cortical nephron segments by use of RT-PCR approach described previously [T. Moriyama, H. R. Murphy, B. M. Martin, and A. Garcia-Perez. Am. J. Physiol. 258 (Renal Fluid Electrolyte Physiol. 27): F1470-F1474, 1990]. We detected mRNA for isozyme rPMCA2 in microdissected distal convoluted tubules (DCT) and in cortical thick ascending limbs (CTAL) and, less consistently, also in proximal convoluted tubule and in glomeruli. The mRNA for isozyme rPMCA1 was abundant in glomeruli but was absent in all examined cortical tubular segments. Our results document that mRNAs for all three major isozymes of rPMCA are present and show a unique distribution in the three major zones of rat renal parenchyma. Specific mRNA coding for rPMCA2 was detected in cortical tubules, namely in CTAL and DCT, whereas mRNA coding isozyme rPMCA1 was found in glomeruli. We suggest that isozyme rPMCA2 might be specifically related to epithelial cells and their function, whereas rPMCA1 is probably a component of nonepithelial cells including these in glomeruli.

In our previous study of plasma membrane Ca2+ pump (PMCA) isozymes in the rat kidney, we found that the mRNA coding isoform PMCA3 was detected primarily in the outer medullary zone of rat kidney tissue. We now investigated the location of the mRNAs coding for isoforms PMCA3 and PMCA4 of Ca2+ pump in the nephron segments that are present in outer medullary parenchyma using the method of reverse transcription and polymerase chain reaction. In mRNA extracted from whole dissected outer medulla we detected mRNA encoding three splice variants (a, b, and c) of isoform PMCA3; isoform PMCA4 was found in the outer medulla almost exclusively as variant b. Analysis of mRNA from microdissected tubule segments show that proximal straight tubules (PST), medullary thick ascending limbs, outer medullary collecting ducts, and descending thin limb of Henle's loop (DTL) all contained mRNA for isoform 4b. In contrast, the mRNA encoding isoform 3 was detected exclusively in DTL and not in other nephron segments. The unique presence of isoform 3 in DTL is rather surprising, since the specific role of this nephron segment in vectorial Ca2+ transport or in intracellular Ca2+ signaling is not known. The data suggest that isoform 3 in cells of the DTL may have a hitherto unrecognized specific role in Ca2+ signaling or transport of Ca2+, which is distinct from the role of the isoforms of the PMCA in all other nephron segments.

In an effort to extend our studies on Ca2+ pumps to animal models, we developed a new monoclonal antibody (5F10) prepared against the human erythrocyte Ca2+-Mg2+-adenosinetriphosphatase (ATPase) that recognizes a protein of approximately 140 kDa in rat kidney homogenates. Enzyme-linked immunosorbent assays show that monoclonal antibody 5F10 binds purified Ca2+-Mg2+-ATPase and rat kidney membrane extracts in a concentration-dependent manner. In paraffin-embedded tissue sections, antibody 5F10 binds to an epitope in the basolateral membranes of rat kidney distal convoluted tubule principal cells. The antibody does not bind to intercalated cells. The latter cells were characterized by the presence of large amounts of carbonic anhydrase C. Polyclonal antibodies directed against chick intestinal 28-kDa vitamin D-dependent calcium binding protein (28-kDa CaBP) also bind epitopes in distal convoluted tubule cells, connecting tubules, and portions of collecting duct but not intercalated cells. Western blot and 45Ca blot analysis of renal cytosolic proteins showed that the polyclonal 28-kDa CaBP-directed antibody detects a protein which also binds calcium. Western blot analysis with monoclonal antibody 5F10 shows binding to both the authentic purified erythrocyte Ca2+ pump (approximately 138 kDa) and to tryptic fragments of this pump. Antibody JA3, previously used for staining of human kidney tubules, reacts with a different set of tryptic fragments, showing that the two antibodies are directed against different regions or conformational determinants on the pump molecule. We show that Ca2+-Mg2+-ATPase and 28-kDa CaBP are present in the principal cells of the distal convoluted tubule of the rat and are absent in intercalated cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Alternative splices capable of generating proteins with altered functions were found (by PCR) in isoform 2 of the rat plasma membrane Ca2+ pump. These splices were concentrated in two hypervariable regions. One of these regions, near the N-terminus and the lipid-binding region, could be altered by the insertion of either or both of inserts x and y. Insertion of both x and y would add 45 amino acids to the molecule. The y insert causes the appearance of a rather hydrophobic stretch of amino acids in the middle of a highly polar region. The second variable region begins in the middle of the calmodulin-binding domain. Insertion of 229 nucleotides at this point of the message converts the b form to the a form, which has an altered (and shorter) C-terminus. The calmodulin-binding domain of this shortened form has a less basic character, which would decrease the affinity for calmodulin. The b form of isoenzyme 2 contains relatively weak protein kinase A substrate sequences, such as KQNSS and KNNS. These sequences are eliminated in form a, and a strongly activated kinase substrate sequence, RRQSS, appears in a different place. Different tissues use different combinations of alternative splices, with heart and brain showing the greatest diversity.

Inactivation of the wild-type human plasma membrane Ca2+ pump (isoform 4b) by fluorescein isothiocyanate is accompanied by covalent modification of Lys591. The mutation of Lys591 to arginine reduced the Ca2+ transport activity to 35% of the wild-type, and diminished the amount of acylphosphate formed from ATP by a corresponding amount. When this mutant was treated with fluorescein isothiocyanate, the enzyme was still irreversibly inactivated, even though no reactive residue was available at position 591. The results show that, although Ca2+ pump function is sensitive to the residue at position 591, Lys591 is not essential for enzyme activity. They also demonstrate that irreversible inhibition of the plasma membrane Ca2+ pump by fluorescein isothiocyanate does not require the covalent modification of Lys591. This indicates that fluorescein isothiocyanate reacts with lysine residues at other positions in addition to Lys591.

The epitope location and specificity of monoclonal antibodies JA9, 5F10 and JA3, raised against the human plasma membrane Ca2+ pump (hPMCA), were analysed by using synthetic peptides of the corresponding epitopes as well as the complete isoforms, hPMCA4b, hPMCA4a and hPMCA1b, expressed in COS-1 cells. The experiments with the peptides showed that JA9 reacted specifically with a region containing residues 51–75 of hPMCA4 (a or b), but not with the same region of isoforms 1, 2 or 3. JA3 reacted with residues 1156–1180, a region unique to hPMCA4b. 5F10 reacted in the region of residues 719–738, which is highly conserved in all PMCA isoforms. Indeed, 5F10 recognized all three isoforms expressed in COS-1 cells. JA9, in contrast, reacted with both variants a and b of hPMCA4 but not with hPMCA1, and JA3 recognized exclusively hPMCA4b. We used these antibodies to discern the distribution of hPMCA4a and hPMCA4b in human brain, heart, kidney and lung. In Western blots of human brain samples, we could identify both hPMCA4a and hPMCA4b. Heart tissue also showed isoform 4b, and probably 4a. In contrast, kidney and lung showed primarily hPMCA4b. In brain, overlapping bands that did not correspond to either variant of hPMCA4 were detected, and in kidney a band migrating in the same position as hPMCA1b was observed. The distribution of the a and b forms of hPMCA4 at the protein level, as analysed by these antibodies, is consistent with the available data about the abundance of mRNAs for the hPMCA isoforms. The presence of hPMCA4b in all the samples supports the proposed role of this isoenzyme as a constitutive form of the pump.

1. A monoclonal antibody (1G4) was raised against the red-cell Ca2+ pump, and it reacted with the pump, as verified by Western blot analysis and by the e.l.i.s.a. method. 2. At 1 mM-ATP and 10 microM-Ca2+, 1G4 inhibited the activity of the purified Ca2+ pump by 40%. 3. Ca2+ pump inhibition by the antibody was non-competitive with regard to Ca2+, calmodulin and the high-affinity portion of the ATP curve. Thus its mechanism was quite different from that of the antibody previously reported [Verbist, Wuytack, Raemaekers, VanLeuven, Cassiman & Casteels (1986) Biochem. J. 240, 633-640], which partially caused inhibition by competition at the ATP site. 4. Antibody 1G4 reduced the steady-state level of phosphorylated intermediate and increased by 50% the calmodulin-activated p-nitrophenyl phosphatase activity of the pump. 5. The experimental results are consistent with the hypothesis that 1G4 inhibits the Ca2+ pump by decreasing the rate of the transition from the E2 form to the E1 form, causing a higher concentration of E2. 6. Analysis by Western blot of the pattern of cross-reaction of 1G4 after tryptic digestion of the pump showed that this antibody reacts with bands of Mr 90,000, 85,000, 50,000 and 33,000. After chymotryptic digestion, the antibody reacts almost exclusively with a fragment of Mr 105,000 that is fully active but is not responsive to calmodulin. Altogether, the results indicate that 1G4 binds to an epitope involved in the functional properties of the enzyme but which is not related to the calmodulin-binding domain.

Development of myometrium in young female rats was stimulated by administration of diethylstilboestrol. Plasma membrane and sarcoplasmic reticulum from rat myometrium were separated by a new and rapid method using a Percoll gradient. Calcium uptake was inhibited in plasma membrane vesicles isolated from oxytocin-treated myometrium, while no consistent effect of oxytocin was found on the Ca2+ uptake in the sarcoplasmic reticulum. Oxytocin regulated the plasma membrane Ca2+ pump by decreasing its apparent affinity for Ca2+ without affecting its maximal velocity. The K1/2 for Ca2+ in the absence of calmodulin was 0.41 +/- 0.04 microM in normal membranes; this was increased to 0.93 +/- 0.12 microM in oxytocin-treated membranes. Calmodulin decreased the K1/2 for Ca2+ to 0.27 +/- 0.027 microM and oxytocin also increased this, to 0.46 +/- 0.061 microM. The effect of oxytocin on the plasma membrane Ca2+ pump was highly dependent on the hormonal status of the animals. When the diethylstilboestrol was administered together with progesterone, the inhibitory action of oxytocin was totally suppressed, consistent with the expected action of this agent. The results suggest that regulation of the plasma membrane Ca2+ pump may be important in the prolonged elevation of intracellular Ca2+ caused by oxytocin.