Differential expression of human nucleoside transporters in normal and tumor tissue.

Responses to nucleoside analog drugs used in the treatment of cancers and viral infections can vary considerably between individuals. Genetic variability between individuals in their ability to transport drugs may be a contributory factor. Nucleoside transporters (NTs) move nucleosides and analog drugs across cell membranes. Four human NTs have been cloned: hENT1, hENT2, hCNT1, and hCNT2. Human NT expression profiles are not well defined; therefore, we undertook a comprehensive quantitative analysis of the differential expression of NTs within normal and tumor tissue. Results show tissue specific expression of the different NTs in normal tissue while matched normal/tumor tissue cDNA array data show considerable variability in all NT expression profiles from different individuals, in particular decreased expression in tumor tissue. Decreased NT expression in tumor tissue may contribute to reduced drug uptake and the development of resistance. These data suggest that nucleoside analog drug therapies may be optimized by determining individual NT expression profiles.

Differential transport of cytosine-containing nucleosides by recombinant human concentrative nucleoside transporter protein hCNT1.

The transportability of cytosine-containing nucleosides by recombinant hCNT1 was investigated in transfected mammalian cells. Apparent K(m) values for hCNT1-mediated transport of uridine, cytidine and deoxycytidine were, respectively, 59, 140 and 150 microM. Uridine transport was inhibited 89, 32 and 11%, respectively, by 500 microM gemcitabine, cytarabine and lamivudine, demonstrating that, unlike gemcitabine (a high-affinity hCNT1 permeant), cytarabine and lamivudine are poor hCNT1 permeants.

Molecular identification of the equilibrative NBMPR-sensitive (es) nucleoside transporter and demonstration of an equilibrative NBMPR-insensitive (ei) transport activity in human erythroleukemia (K562) cells.

Equilibrative nucleoside transport processes in mammalian cells are categorized as either nitrobenzylthioinosine (NBMPR)-sensitive (es) or NBMPR-insensitive (ei). Inhibition of the es process arises from binding of NBMPR to a high-affinity site(s) on the es transporter that can be identified by photoaffinity labeling with [3H]NBMPR. This study examined the equilibrative nucleoside transport processes of cultured human erythroleukemia (K562) cells. The presence of NBMPR binding sites (4.8 +/- 0.9 x 10(5)/cell, Kd = 0.3 nM), together with the identification of polypeptides by specific photolabeling of membranes with [3H]NBMPR, indicated that K562 cells possess es nucleoside transporters (ca 500,000 copies/cell). The photolabeled polypeptides of K562 cells migrated with lower relative mobility (peak M(r) value, 63,000) than did those of human erythrocytes (peak M(r) value, 53,000). This difference in apparent M(r) was abolished by prolonged treatment of membrane proteins with N-glycosidase F, suggesting that equilibrative nucleoside transport in K562 cells and erythrocytes is mediated by the same, or a closely related, es isoform. A cDNA encoding the es nucleoside transporter of human placenta (termed hENT1) was recently isolated by a strategy based on the N-terminal sequence of the es transporter of human erythrocytes. hENT-like mRNA species were detected in K562 cells, as well as in several other human cell lines of neoplastic origin (A459, G361, HeLa, HL-60, Molt-4, Raji, SW480), by high-stringency northern analysis with a placental hENT1 probe. A cDNA that encoded a protein identical to hENT1 was isolated by reverse transcriptase polymerase chain reaction with primers specific for hENT1. NBMPR inhibited zero-trans influx of 3H-labeled adenosine, uridine and thymidine by 50% (IC50 values) at 0.4-1.0 nM, confirming the presence of an NBMPR-sensitive (es) transport process, which accounted for 80-90% of total transport activity. The remaining component was identified as the equilibrative NBMPR-insensitive (ei) transport process since it: (i) exhibited low (IC50 > 1.0 microM) sensitivity to NBMPR; (ii) was not concentrative; and (iii) was unchanged by elimination of the sodium gradient. The kinetic parameters (determined at 37 degrees C) for the es- and ei-mediated processes differed markedly. Values for transport of uridine by the es- and ei-mediated processes were, respectively: K(m) = 229 +/- 39 and 1077 +/- 220 microM; Vmax, 186 +/- 31 and 40 +/- 5 pmol/microliter cell water/sec. Values for transport of adenosine by the es and ei-mediated processes were, respectively, 61 +/- 9 and 133 +/- 17 microM; Vmax, 70 +/- 5 and 23 +/- 8 pmol/microlitere cell water/sec. The ei-mediated process, although small, was of pharmacologic importance since K562 cells could not be protected by NBMPR (10 microM) from the cytotoxic effects of tubercidin (7-deazaadenosine).

Ethanol alters the subcellular localization of delta- and epsilon protein kinase C in NG108-15 cells.

Protein kinase C (PKC) has been shown to regulate the ethanol sensitivity of membrane-bound receptors and transporters, but little is known about the molecular mechanisms underlying this regulation. PKC is a family of isozymes that translocate to new intracellular sites on activation. Here we present immunochemical data showing that ethanol causes translocation of delta- and epsilon-PKC to new intracellular sites. Ethanol causes translocation of delta-PKC from the Golgi to the perinucleus; this translocation is similar to that induced by activation of PKC with phorbol esters. In contrast, epsilon-PKC translocation caused by ethanol is different from that induced by phorbol esters; ethanol causes translocation of epsilon-PKC from the perinucleus to the cytoplasm, whereas phorbol ester activation causes translocation of epsilon-PKC to the nucleus. Because the substrate specificity of these kinases is determined by their site of localization, ethanol-induced translocation of delta- and epsilon-PKC to new intracellular sites may explain some of the pleiotropic effects of ethanol on cellular functions.

The role of protein kinase C in cellular tolerance to ethanol.

We have shown that ethanol inhibits uptake of adenosine by a specific nucleoside transporter in NG108-15 neuroblastoma x glioma cells and that cAMP-dependent protein kinase (PKA) activity is required for this inhibition. After chronic exposure to ethanol, adenosine uptake is no longer inhibited on rechallenge with ethanol, i.e. transport has become tolerant to ethanol. Here we show that protein kinase C (PKC) contributes to ethanol-induced tolerance of adenosine transport. Activation of PKC by phorbol esters in control cells results in an ethanol-tolerant phenotype, similar to that produced by chronic ethanol exposure. In addition, chronic exposure to ethanol increases the amounts of alpha, delta, and epsilon PKC. However, reducing PKC activity by inhibition with chelerythrine during chronic exposure to ethanol or down-regulation by phorbol esters prevents the development of ethanol-induced tolerance of adenosine transport. By contrast, the inhibition of PKA activity produces tolerance to ethanol inhibition of adenosine uptake. When protein phosphatase inhibitors are present, inhibiting PKA activity has no effect on ethanol sensitivity of adenosine uptake, suggesting a role for protein phosphatases in the regulation of ethanol sensitivity of uptake. Taken together, our results suggest that PKA and PKC have opposing effects on the ethanol sensitivity of adenosine transport; PKA activity is required for ethanol sensitivity, and PKC activation produces tolerance. Based on these data, we propose that chronic ethanol exposure increases PKC activity, leading to the activation of a protein phosphatase (1 or 2A). This phosphatase then dephosphorylates a PKA-phosphorylated site, which is required for ethanol to inhibit adenosine uptake. Therefore, the sensitivity of adenosine transport to ethanol appears to be maintained by a balance of PKA and protein phosphatase activities, and PKC may regulate phosphatase activity.

Activation of cyclic AMP-dependent protein kinase reverses tolerance of a nucleoside transporter to ethanol.

Adenosine mediates some of the acute and chronic effects of ethanol in neural cells. In cultured NG108-15 cells, ethanol inhibits adenosine uptake via a specific facilitative nucleoside transporter leading to an increase in extracellular adenosine, activation of adenosine A2 receptors and increases in intracellular cyclic AMP (cAMP). After chronic ethanol exposure, an adaptive decrease in receptor-stimulated cAMP levels occurs. Additionally, the transporter becomes insensitive to rechallenge with ethanol and adenosine uptake is not inhibited. cAMP levels are decreased in cells chronically exposed to ethanol and we show here that cAMP-dependent kinase (PKA) activity in cellular homogenates also is decreased. Therefore, decreased cAMP-dependent phosphorylation may be responsible for loss of ethanol sensitivity. To test this hypothesis, NG108-15 cells were treated with agents that alter PKA activity and the ethanol sensitivity of adenosine transport was measured. In naive cells, decreasing PKA activity with the cAMP antagonist, Rp-adenosine-3',5'-cyclic phosphorothioate, resulted in ethanol-insensitive adenosine uptake. This effect was blocked by the phosphatase inhibitor, okadaic acid. These results suggest that loss of ethanol sensitivity is correlated with decreased PKA activity. Therefore, stimulating PKA activity in chronically treated cells should restore sensitivity of adenosine uptake to inhibition by ethanol. Indeed, the cAMP agonist, Sp-adenosine-3',5'-cyclic phosphorothioate, restored ethanol sensitivity of transport in cells treated chronically with ethanol. Our results suggest that ethanol sensitivity of adenosine transport is regulated by PKA and protein phosphatase activities in NG108-15 cells. Moreover, the effects of chronic ethanol exposure on adenosine transport can be reversed by activating PKA.

Characterization of the Pacific salmon gonadotropin-releasing hormone gene, copy number and transcription start site.

Multiple forms of gonadotropin-releasing hormone (GnRH) have been shown to exist in all vertebrates examined except recently-evolved placental mammals. To study the origin and regulation of the GnRH genes in a Pacific salmon (Oncorhynchus nerka), we isolated and sequenced the salmon form of GnRH. The Southern blot shows a single band that strongly hybridizes to a probe for the gene reported here and weaker bands that may represent genes for related forms of GnRH. There is strong conservation of sequence in the hormone coding region and of the gene organization between fish and mammals. However, the GnRH-associated peptide (GAP) shows very little sequence identity with the mammalian GAPs, questioning its physiological role. We also show for the first time the transcriptional start site for a GnRH gene in a non-mammalian species. Interestingly, a large segment of 1152 nucleotides in the promoter region of the Pacific salmon GnRH gene is missing compared with the Atlantic salmon (Salmo salar) gene. These gene rearrangements suggest that these two salmonid species, which have been geographically separated for 8-15 million years, have evolved promoters with different mechanisms for control and transcription of GnRH.

Two salmon neuropeptides encoded by one brain cDNA are structurally related to members of the glucagon superfamily.

A cDNA that codes for two peptides in the glucagon superfamily has been isolated from sockeye salmon brain. The first peptide is related to growth hormone-releasing hormone (GHRH), which has high sequence similarity with PACAP-related peptide. The second peptide is structurally related to vasoactive intestinal peptide, which is also related to a newly identified peptide in mannals, pituitary adenylate-cyclase-activating polypeptide (PACAP). The salmon precursor contains 173 amino acids and has dibasic and monobasic enzyme-processing sites for cleavage of a 45-amino-acid GHRH-like peptide with a free C-terminus and a 38-amino-acid PACAP with an amidated C-terminus. The salmon GHRH-like peptide has 40% amino acid sequence identity with a human GHRH and 56% identity with human PACAP-related peptide. The 38-amino-acid salmon PACAP is highly conserved (89-92% identity) with only three or four amino acid substitutions compared with the human, ovine and rat 38-amino-acid PACAP. Not previously reported for mammalian species, a short precursor coding for only one peptide exists in salmon in addition to the long precursor coding for two peptides. In the short precursor, the coding region for GHRH is deleted leaving the PACAP-coding region in a correct reading frame. This provides one possible control mechanism for an increased expression of one peptide (PACAP) without the concomitant increase in the other peptide (GHRH) as occurs in a double-peptide precursor. The importance of the 3' non-translated region of the salmon GHRH/PACAP precursor in the regulation of translation is suggested by its 70% nucleotide sequence identity to the 3' non-translated regions of the mammalian PACAP precursors. The structural organization of the salmon GHRH/PACAP precursor provides a possible evolutionary scheme for precursors that contain tandem peptides in the glucagon superfamily.

Isolation of different brain-specific isotypes of alpha-tubulins from chum salmon (Oncorhynchus keta).

An alpha-tubulin cDNA clone, pTUB5 (1496 bp), encoding a protein of 444 amino acids (mol. wt. 48,840), has been isolated from the brain of Pacific salmon, Oncorhynchus keta. Partial sequence data were also obtained for two other alpha-tubulin isotypes, pTUB6 and pTUB9, which are similar in sequence to pTUB5 except in the carboxy region of the protein. This region of alpha-tubulin has previously been shown to be important for the binding of microtubule associated proteins suggesting that the isotypes described in this study may represent differentially controlled elements of the neural tubulin population. The isotypes show brain-specific expression and are the first tubulins to be reported for this tissue in fish.

Phylogeny and ontogeny of gonadotropin-releasing hormone: comparison of guinea pig, rat, and a protochordate.

Immunoreactive gonadotropin-releasing hormone (ir-GnRH) was detected in brain extracts of newborn and 10-day-old rats and in adult guinea pigs; it was also present in extracts of the neural ganglion and gland of a protochordate. Radioimmunoassay (RIA) using different GnRH antisera after high-performance liquid chromatography (HPLC) revealed that the dominant form of GnRH is the mammalian form (pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH2) both during ontogenesis in the rat and in the adult guinea pig known to have variant forms of other peptide hormones. None of the other forms of GnRH identified in nonmammalian species to date appear to be present in the rat or guinea pig. A small amount of an unidentified HPLC early eluting form of GnRH is present, but detection by antiserum B-6 implies that it is also mammalian GnRH, with the possibility of changes in positions 2-4. The molecular form of GnRH in a protochordate, the sea squirt Chelyosoma productum, is distinct from salmon and mammalian GnRHs. Cross-reactivity with the sea squirt GnRH-like molecule was highest with an antiserum made against lamprey GnRH; the same antiserum was used to stain nerve fibers in the neural ganglion and some of its roots. This is the first report using RIA, HPLC, and immunocytochemistry to show that protochordates have GnRH-like material. The results suggest that GnRH may have been present at the transition between the invertebrates and vertebrates.

Molecular cloning of two distinct vasotocin precursor cDNAs from chum salmon (Oncorhynchus keta) suggests an ancient gene duplication.

The structures of two different vasotocin precursors from chum salmon brain have been elucidated through the molecular cloning of their corresponding cDNAs. Although the predicted precursors, consisting respectively of 153 and 158 amino acids, have the same structural organisation, they show 35% amino acid sequence divergence, of which only approximately half are isofunctional substitutions. Remarkably, while the C terminal segments of both precursors resemble the glycopeptide moiety of the related mammalian vasopressin precursor, both salmon precursors lack consensus sequences for N-glycosylation.