Mutually exclusive and cassette exons underlie alternatively spliced isoforms of the Na/Ca exchanger

Alternative splicing of the Na(+)-Ca2+ exchanger gene, NCX1

We describe an analysis of the NCX1 gene and show that various tissues express different alternatively spliced forms of the gene. Alternative splicing has been confirmed by the genomic analysis of the Na(+)-Ca2+ exchanger gene. We also describe the Drosophila Na(+)-Ca2+ exchanger as having many of the same structural characteristics of the mammalian exchangers and this locus as possibly undergoing alternative splicing in the same region that has been described in the NCX1 gene. The general structure of the exchangers is similar to that of the alpha-subunit of the (Na(+)+ K+)-A Pase. Finally, sequence comparison of the various molecules demonstrates that structural characteristics of these molecules are more strongly conserved than the primary sequence of these products.

Aging does not affect steady-state expression of the Na+/Ca2+ exchanger in rat brain

1. Steady-state protein and mRNA levels of the Na+/Ca2+ exchanger were studied in old (24-month) and young (3-month) brains of male Fischer 344 rats by Western and Northern analysis. 2. Northern analysis with a cDNA proble for the Na+/Ca2+ exchanger amplified from human brain RNA indicated the presence of two transcripts for the Na+/Ca2+ exchanger (6 and 16 kb). Both transcripts were present in similar abundance in the cerebrum and hippocampus. In the cerebellum the 6-kb transcript predominated. The cerebellum had the highest overall level of expression. There were no significant age-related effects seen on the level of expression of either transcript in each of the brain areas tested. 3. Western analysis of plasma membrane vesicles purified from cerebral cortex identified a single protein of 116 kDa. Consistent with the Northern analysis, no age-related effect on protein levels was seen. 4. The mechanisms underlying altered Na+/Ca2+ exchange activity in aging rat brain (Michaelis, 1989) most likely do not involve changes in gene expression and are therefore more likely to represent posttranslation modifications of the protein.

Activation of adenylyl cyclase downregulates sodium/calcium exchanger of arterial myocytes

Chronic elevation of adenosine 3',5'-cyclic monophosphate (cAMP) is known to inhibit the proliferation of cultured vascular smooth muscle cells. The present findings show that the activation of adenylyl cyclase with forskolin decreased Na+/Ca2+ exchanger (NCX) mRNA and activity. Fetal bovine serum restored NCX transcript and activity. The changes in NCX transcript preceded the changes in NCX activity. Incubation of low-passage immortalized myocytes with forskolin plus 3-isobutyl-1-methylxanthine (IBMX), which inhibits cAMP phosphodiesterase, decreased NCX mRNA by 60% in 6 h and 80% in 24 h. After a 6-h lag, forskolin plus IBMX decreased NCX activity almost linearly to 20% of control at 40 h. 1,9-Dideoxyforskolin, which does not activate adenylyl cyclase, had no effect on NCX mRNA or activity. Forskolin plus IBMX decreased the c-Myc transcript, an immediate-early gene whose expression correlates with cell proliferation, but had no effect on plasma membrane Ca(2+)-ATPase transcripts. Removal of forskolin plus IBMX and addition of fetal bovine serum increased NCX and c-Myc transcripts seven- to eightfold in 6 h and restored NCX activity in 24 h. Inhibition of protein or RNA synthesis by cycloheximide or actinomycin D, respectively, prevented the increase in NCX mRNA. In contrast to blocking NCX induction, cycloheximide potentiated c-Myc induction by serum. Transcription factors that regulate myocyte growth may mediate the opposing influences of serum and forskolin on NCX mRNA and activity.

Antisense inhibition of Na+/Ca2+ exchange in primary cultured arterial myocytes

The effects of chimeric phosphorothioated antisense oligodeoxynucleotides (AS-oligos) targeted against the Na+/Ca2+ exchanger (NCX) were tested in primary cultured rat mesenteric artery myocytes. In parallel cultures, myocytes proliferated and were morphologically normal in the presence of scrambled nonsense (NS-) or AS-oligos or no oligos (controls). NCX function was examined with digital imaging, using fura 2 to estimate the cytosolic free Ca2+ concentration ([Ca2+]cyt). Resting [Ca2+]cyt was higher (145 +/- 4 nM; P < 0.05) in AS-oligo-treated cells than in controls (125 +/- 5 nM) or NS-oligo-treated cells (131 +/- 4 nM). Lowering external Na+, to promote Ca2+ entry via NCX, increased [Ca2+]cyt transiently in controls and NS-oligo-treated cells but not in AS-oligo-treated cells. Raising the cytosolic free Na+ concentration with ouabain augmented the low-Na(+)-induced rise in [Ca2+]cyt in controls and NS-oligo-treated cells, but AS-oligo-treated cells still did not respond. Nevertheless, serotonin (5-HT) increased [Ca2+]cyt in all three groups. Thus AS-oligos selectively blocked NCX activity but not the 5-HT response. To determine the effect of NCX knockdown on the modulation of stored Ca2+, the 5-HT response was tested immediately after removal of external Ca2+. Ouabain augmented the 5-HT-induced rise in [Ca2+]cyt in control and NS-oligo-treated cells but not AS-oligo-treated cells. This indicates that the NCX can modulate intracellular Ca2+ stores. We conclude that AS-oligos are useful for investigating the physiological role of NCX in vascular smooth muscle.

Post-transcriptional regulation of the Na+/Ca2+ exchanger in aging rat heart

Altered calcium homeostasis in the senescent heart appears to be the result, at least in part, of decreased Na+/Ca2+ exchange activity. To further investigate the basis of the decrease in Na+/Ca2+ exchange activity, Na+/Ca2+ exchanger gene expression in the heart was compared in 3 and 24 month old male Fischer 344 rats. Sarcolemmal vesicles prepared from left ventricle and septum showed reduced Na(+)-dependent Ca2+ uptake in 24 month old animals when compared to 3 month old animals (0.156 +/- 0.005 and 0.135 +/- 0.008 nmol Ca2+/mg/10 s; mean +/- S.E. for 3 month and 24 month old animals, respectively). Western analysis showed immunodetectable Na+/Ca2+ exchanger protein levels were decreased by 19% in 24 month old animals when compared to 3 month old animals. Poly(A+) RNA was purified from left and right ventricle and left and right atria and subjected to Northern analysis using digoxin labeled cDNA probes for the Na+/Ca2+ exchanger and actin. The Na+/Ca2+ exchanger probe labeled a 7 kb message in both ventricle and atria, while the actin probe labeled both beta-actin (2.2 kb) and alpha-actin (1.4 kb). The steady state level of expression of Na+/Ca2+ exchanger Poly(A+) RNA when normalized to beta-actin, was similar when ventricle and atria were compared. There were no observable differences in Na+/Ca2+ exchanger or alpha-actin Poly(A+) RNA steady state levels when comparing 3 and 24 month old animals. The results suggest that reduced Na+/Ca2+ exchange activity in the left ventricle of 24 month old animals was most likely the result of post-transcriptional modification of the protein that was detectable by Western analysis.

Expression of a functionally active human renal sodium-calcium exchanger lacking a signal sequence

The Na+-Ca2+ exchanger is an unusual membrane transport protein as it contains an NH2-terminal signal sequence which is co-translationally removed in the endoplasmic reticulum during synthesis. To determine if the signal sequence was essential for biosynthesis, mutations were introduced in the NH2 terminus of the cDNA coding for the human renal Na+-Ca2+ exchanger in order to alter processing of the protein. To prevent cleavage of the signal sequence during biosynthesis, the last residue of the consensus signal sequence, Ala-1, was changed to Phe. Deletion mutants were also constructed to encode for exchangers which lacked the signal sequence, the signal sequence and the first extracellular loop, or all of the NH2 terminus including the first transmembrane segment of the mature protein. These mutants were expressed in HEK 293 cells and assayed for Na+-Ca2+ exchange activity. Mutants lacking either a signal sequence or containing a noncleavable signal sequence were still targeted to the plasma membrane, where they exhibited Na+-Ca2+ exchange activity. By contrast, the mutants which had more than the signal sequence deleted did not demonstrate any exchange activity. These mutants were, however, still integrated into the membrane and were resistant to alkali extraction. These results show that the signal sequence is not essential for biogenesis of the Na+-Ca2+ exchanger and suggests that the molecule contains one or more internal signal sequences for insertion into the membrane during biosynthesis.

The putative amino-terminal signal peptide of the cloned rat brain Na(+)-Ca2+ exchanger gene (RBE-1) is not mandatory for functional expression

The rat brain Na(+)-Ca2+ exchanger (RBE) gene, as well as other isoforms of this protein family, can be organized into 12 transmembrane alpha helices, the first of which was proposed by Durkin et al. (14) to constitute a cleavable signal peptide. We have prepared three amino-terminal mutants, in which 21, 26, and 31 amino acids beyond the initiating methionine were deleted. The deletions include the hydrophobic core of the putative signal peptide (N21), the entire putative signal peptide and parts of the putative signal peptidase cleavage site (N26), and the entire putative signal peptide and putative signal peptidase cleavage site (N31). All three mutant clones were transiently expressed in HeLa cells. The average Na+ gradient-dependent Ca2+ transport activity of the mutant exchangers was 108% (N21), 37.2% (N26), and 60.06% (N31) of the wild-type clone. Mutation of the putative cleavage site by an exchange of Ala-32 --> Asp, resulted in a decrease in Na(+)-Ca2+ exchange activity to 7.7%, relative to the wild-type exchanger. Functional reconstitution of the proteins that were expressed in the transfected cells, resulted in transport activities of: 60.1% (N21), 26.75% (N26), 85.36% (N31), and 31% (Ala-32 --> Asp) relative to the wild-type exchanger. Western blot analysis of the protein profile of RBE-1, N21, N26, N31 and Ala-32 --> Asp-transfected HeLa cells was carried out by using an antipeptide antibody directed against a pentadecapeptide segment derived from the large putative cytoplasmic loop of the cloned rat exchanger gene. In the total cell extract and in the plasma membrane-enriched fraction, in addition to a major protein band of about 125 kDa, which corresponds to the molecular mass of the mature fully processed Na(+)-Ca2+ exchanger, an additional protein of about 135 kDa is revealed in the profile of N21- and N26-transfected cells. This band is not detected in the protein profile of RBE-1, N31, or Ala-32 -->Asp. The amino-terminal truncated mutants of the cloned Na(+)-Ca2+ exchanger could be expressed and processed also in a reticulocyte lysate supplemented with dog microsomes. Our results suggest that the putative signal peptide of the cloned Na(+)-Ca2+ exchanger gene does not play a mandatory role in functional expression of the protein in HeLa cells.

Phorbol esters downregulate expression of the sodium/calcium exchanger in renal epithelial cells

The Na+/Ca2+ exchanger (NCE) contributes to Ca2+ reabsorption by connecting tubules of the nephron. A line of renal epithelial cells from monkey kidney (LLC-MK2) was used to investigate the regulation of NCE expression. After the activation of protein kinase C (PKC) by phorbol myristate acetate (PMA), NCE activity decreased exponentially by 75% in 48 h (half time approximately 19 h). PMA decreased NCE mRNA by 85% in 24 h. The decrease in NCE transcript preceded the downregulation of NCE activity. NCE protein was quantified with a monoclonal antibody to cardiac NCE. PMA decreased the binding of 3H-labeled antibody to cell sonicates by 40% in 24 h. Immunoblots show that PMA produced a marked and extended increase in membrane-associated PKC-alpha, although PMA depleted total PKC-alpha by 65% in 24 h. In vivo 32P labeling of myristolated alanine-rich C kinase substrate, a specific PKC substrate, confirmed that PMA produced a rapid and extended activation of PKC. 4 alpha-PMA, a stereoisomer of PMA that neither binds nor activates PKC, had no effect on NCE activity or transcript. These findings indicate that activation of PKC with phorbol esters downregulates NCE mRNA, protein, and activity in renal epithelial cells.

Growth factor-induced phosphorylation and activation of aortic smooth muscle Na+/Ca2+ exchanger

Although the Na+/Ca2+ exchanger is one of the major Ca2+ extrusion systems in excitable tissues, little is known about its regulation via protein phosphorylation. We now present evidence that the Na+/Ca2+ exchanger is phosphorylated in quiescent and growth factor-stimulated cultured aortic smooth muscle cells. The Na+/Ca2+ exchanger was isolated from 32P-labeled cells by immunoprecipitation with a specific polyclonal antibody. Phosphorylation of the exchanger was increased by up to 1.7-fold in response to platelet-derived growth factor-BB (PDGF-BB), alpha-thrombin, or phorbol 12-myristate 13-acetate (PMA). However, angiotensin II did not enhance the phosphorylation significantly. The extent of phosphorylation appeared to correlate with the growth factor-induced increase in cell 1,2-diacylglycerol. At least four phosphopeptides (P1 to P4) were detected by tryptic phosphopeptide map analysis of the phosphorylated exchanger, suggesting that phosphorylation occurred at multiple sites. PDGF-BB and PMA increased phosphorylation of the same phosphopeptides (in particular P1). Phosphorylated amino acids were exclusively serine residues in both quiescent and stimulated cells. We found that growth factors enhanced Na+/Ca2+ exchange activity and that there was a good correlation between the growth factor-induced stimulations of phosphorylation and exchange activity. PDGF-BB-induced activation of the exchanger was abolished by prior long treatment of cells with PMA. These results suggest that the Na+/Ca2+ exchanger is activated by protein kinase C-dependent phosphorylation in response to growth factors in vascular smooth muscle cells.

An alternative splicing site modifies the carboxyl-terminal trans-membrane domains of the Na+/Ca2+ exchanger

The 6-kilobase (kb) cDNA of pTB11 clone and its 5' fragment of 3.7 kb encoding the canine heart Na+/Ca2+ exchanger (Nicoll, D.A., Longoni, S., and Philipson, K.D. (1990) Science 250, 562-565) were transiently expressed in 293 cells to investigate the role of the 3'-"untranslated" region. Both fragments yielded high levels of expressed protein that were well incorporated in the membranes. Cells expressing the 6-kb cDNA produced rearranged transcripts of smaller than expected size. A 120-kDa polypeptide was produced in cells expressing the modified exchanger, and Ca2+ uptake was higher in this type of transfected cells. A constant stretch of nucleotides located at the 3' end of the 6 kb cDNA was found to be connected, by alternative RNA splicing, to four different upstream sequence positions. The deduced hydrophobic sequence of the spliced-in exon could replace the IX or the XI trans-membrane domain of the exchanger protein in two spliced isoforms. The new exon sequence was not completely included in the pTB11 insert, i.e. these two products were artificially truncated. The RNA processing of these two alternative 5'-splicing sites also occurred in tissues, as shown by RNase protection analysis. In a third type of isoform the splicing took place downstream of the originally proposed stop codon, whereas in a fourth type a stop codon was introduced after the V hydrophobic segment in the large intracellular loop.

Mammalian exchangers and co-transporters

In the past year, novel mammalian exchanger and co-transporter isoforms have been characterized. Specialized subdomains within these oligomeric transporters have been shown to be involved in biosynthesis, targeting, transport and regulation. Progress on the structural front has been limited due to the lack of high-resolution structures, but transport mutants responsible for disease states continue to be identified.