Cloning of a human nucleoside transporter implicated in the cellular uptake of adenosine and chemotherapeutic drugs

In most mammalian cells nucleoside uptake occurs primarily via broad-specificity, es (e, equilibrative; 5, sensitive to NBMPR inhibition) transporters that are potently inhibited by nitrobenzylthioinosine (NBMPR). These transporters are essential for nucleotide synthesis by salvage pathways in hemopoietic and other cells that lack de novo pathways and are the route of cellular uptake for many cytotoxic nucleosides used in cancer and viral chemotherapy. They play an important role in adenosine-mediated regulation of many physiological processes, including neurotransmission and platelet aggregation, and are a target for coronary vasodilator drugs. We have previously reported the purification of the prototypic es transporter from human erythrocytes and have shown that this glycoprotein of apparent M, 55,000 is immunologically related to nucleoside transporters from several other species and tissues, including human placenta. Here we report the isolation of a human placental cDNA encoding a 456-residue glycoprotein with functional characteristics typical of an es-type transporter. It is predicted to possess 11 membrane-spanning regions and is homologous to several proteins of unknown function in yeast, nematodes, plants and mammals. Because of its central role in the uptake both of adenosine and of chemotherapeutic nucleosides, study of this protein should not only provide insights into the physiological roles of nucleoside transport but also open the way to improved therapies.

Demonstration of the existence of mRNAs encoding N1/cif and N2/cit sodium/nucleoside cotransporters in rat brain

Nucleoside transport may be involved in the regulation of extracellular levels of adenosine, an inhibitory neuromodulator in the central nervous system. Previous reports have provided functional evidence for Na+-dependent nucleoside transport in rat brain. We isolated total RNA from various regions of rat brain and tested for the presence of mRNA for two recently cloned Na+/nucleoside cotransporters using reverse transcriptase PCR (RT-PCR). Messenger RNA for a pyrimidine-selective Na+/nucleoside cotransporter mRNA (rCNT1) was detected in samples from each brain region tested by RT-PCR amplification of a 309-bp DNA product. Southern blot and sequence analysis confirmed that this product was derived from rCNT1 mRNA. A purine-selective Na+/nucleoside cotransporter mRNA (rCNT2, also termed SPNT) was detected throughout brain by amplifying a 235-bp DNA product, the sequence of which was identical to that published. These experiments demonstrate the presence of both rCNT1 and rCNT2 mRNA in rat brain.

Transport of adenosine by recombinant purine- and pyrimidine-selective sodium/nucleoside cotransporters from rat jejunum expressed in Xenopus laevis oocytes

Two major Na+-dependent nucleoside transporter subtypes implicated in adenosine transport in mammalian cells are distinguished functionally on the basis of substrate specificity: one is selective for pyrimidine nucleosides but also binds adenosine, and the other has selectivity for purine nucleosides but also binds uridine. Transportability of adenosine by the purine-selective system has been established by measurements of [3H]adenosine fluxes, whereas the conclusion that adenosine is permeant of the pyrimidine-selective system is based on inhibition assays. We investigated adenosine transport mediated by a recombinant pyrimidine-selective rat jejunal/kidney Na+/nucleoside cotransporter (rCNT1) expressed in Xenopus laevis oocytes and compared it with that mediated by a recombinant purine-selective rat jejunal/liver Na+/nucleoside cotransporter (rCNT2). Adenosine fluxes mediated by rCNT1 were 1 order of magnitude lower than those mediated by rCNT2. In kinetic studies, rCNT1 transported adenosine with an apparent Km value for influx (26 microM) similar to that for uridine but with a very much lower Vmax value, and the Vmax/Km ratios were 0.003 and 0.57 for adenosine and uridine, respectively. Recombinant rCNT1 mediated efflux of [3H]uridine from preloaded oocytes, demonstrating a capacity for bidirectional transport of nucleoside permeants. Uridine efflux was stimulated by extracellular uridine and inhibited by extracellular adenosine, suggesting that the rate of conversion of rCNT1 from its outward-facing conformation to its inward-facing conformation was increased when the transporter was complexed with uridine and decreased when it was complexed with adenosine. Thus, although rCNT1 binds adenosine and uridine with similar affinities, it kinetically favors transport of uridine.

Transport of the antiviral nucleoside analogs 3'-azido-3'-deoxythymidine and 2',3'-dideoxycytidine by a recombinant nucleoside transporter (rCNT) expressed in Xenopus laevis oocytes

Expression screening in Xenopus oocytes has been used to isolate a cDNA from rat jejunal epithelium encoding an intestinal/kidney Na(+)-dependent nucleoside transporter protein named rCNT1 [J. Biol. Chem. 269:17757-17760 (1994)]. rCNT1 is predicted to have 648 amino acid residues (relative molecular mass, 71,000) with 14 potential transmembrane domains and belongs to a new family of transporter proteins. Recombinant rCNT1 transports physiological pyrimidine nucleosides and adenosine. In the current investigation, functional expression in Xenopus oocytes was used to determine whether recombinant rCNT1 also transports antiviral pyrimidine nucleoside analogs. The recombinant protein mediated Na(+)-dependent transport of both 3'-azido-3'-deoxythymidine (AZT) and 2',3'-dideoxycytidine (ddC). Apparent K(m) values of '0.5 mM were obtained for both [3H]AZT and [3H]ddC influx compared with 37 microM for [3H]uridine influx, with Vmax/Km ratios of 0.048, 0.039, and 0.57 for AZT, ddC, and uridine, respectively. Extracellular AZT and ddC stimulated rCNT1-mediated efflux of [3H]uridine from preloaded oocytes. These experiments provide direct evidence for Na(+)-dependent transport of AZT and ddC and suggest that members of the cNT family may be involved in the intestinal absorption and renal handling of pyrimidine nucleoside analogs used to treat acquired immune deficiency syndrome.

Functional characterization of a recombinant sodium-dependent nucleoside transporter with selectivity for pyrimidine nucleosides (cNT1rat) by transient expression in cultured mammalian cells

We have demonstrated that monkey kidney (COS-1) cells have a single type of nucleoside transport process, which, because it was equilibrative, sodium-independent and could be inhibited by nitrobenzylthioinosine (NBMPR), was identified as the 'equilibrative sensitive' or 'es' transporter. Using NBMPR or dilazep to inhibit the endogenous nucleoside transport activity, we have transiently expressed a cDNA that encodes an inhibitor-insensitive, concentrative nucleoside transporter protein (cNT1rat) of rat intestine in COS-1 cells. The production of recombinant cNT1rat was examined by immunoblotting using an epitope-tagged construct and by analysis of inward fluxes of 3H-labelled nucleosides. Recombinant cNT1rat was sodium-dependent and selective for pyrimidine nucleosides, with approximately Km values of 21 microM, 12.5 microM and 15 microM for uridine, thymidine and adenosine, respectively. Although adenosine exhibited high affinity for the recombinant transporter, its Vmax value was low. A variety of anti-viral and anti-cancer nucleoside drugs inhibited cNT1rat-mediated uptake of uridine by transfected COS-1 cells although to different extents (Floxidine > Idoxuridine > Zidovudine > Zalcitabine > Cytarabine > Gemcitabine), suggesting that the concentrative pyrimidine-selective nucleoside transporters, of which cNT1rat is a representative, may play a role in cellular uptake of these drugs. The cNT1rat/COS-1 expression system is a useful tool for analysis of cNT1rat-mediated transport processes.

Zidovudine and dideoxynucleosides deplete wild-type mitochondrial DNA levels and increase deleted mitochondrial DNA levels in cultured Kearns-Sayre syndrome fibroblasts

Kearns-Sayre syndrome is the most commonly diagnosed mitochondrial cytopathy and produces severe neuromuscular symptoms. The most frequent cause is a mitochondrial DNA deletion that removes a 4977-base pair segment of DNA that includes several genes encoding for respiratory chain subunits. Treatment of AIDS patients with nucleoside analogs has been reported to cause mtDNA depletion and myopathies. Here, we report that azidothymidine, dideoxyguanosine, and dideoxycytidine cause a depletion of wild-type mtDNA while increasing the levels of deleted mitochondria DNA in Kearns-Sayre syndrome fibroblasts. The result of these effects is a large increase in the relative amounts of delta mtDNA in comparison to wild type mtDNA. We found that Kearns-Sayre syndrome fibroblasts are a mixed population of cells with deleted mtDNA comprising from 0 to over 20% of the total mtDNA in individual cells. Treatment of cloned cell lines with dideoxycytidine did not result in increased levels of delta mtDNA. The results suggest that nucleoside analogs may act to increase the average delta mtDNA levels in a mixed population of cells by preferentially inhibiting the proliferation of cells with little or no delta mtDNA. This raises the possibility that modulation of deleted mtDNA levels may occur by similar mechanisms in vivo, in response to the influence of exogenous agents.

Identification of a novel membrane transporter associated with intracellular membranes by phenotypic complementation in the yeast Saccharomyces cerevisiae

A partial mouse cDNA was isolated by its ability to functionally complement a thymidine transport deficiency in plasma membranes of the yeast, Saccharomyces cerevisiae. The full-length cDNA encoded a previously unidentified 27-kDa protein (mouse transporter protein (MTP)) with four predicted transmembrane-spanning domains. MTP mRNA was detected in cells of several mammalian species, and its predicted protein sequence exhibited near identity (98%) with that of a human cDNA (HUMORF13). MTP and its homologs evidently reside in an intracellular membrane compartment because a protein (about 24 kDa) that was recognized by MTP-specific antibodies was observed in a subcellular fraction of rat hepatocytes enriched for Golgi membranes. Deletion of the hydrophilic C terminus of MTP, which encompassed two putative signal motifs for intracellular localization (Tyr-X-X-hydrophobic amino acid), allowed expression of recombinant protein (MTP deltaC) in plasma membranes of Xenopus laevis oocytes. MTP deltaC-expressing oocytes exhibited greater fragility than nonexpressing oocytes, and those that survived the experimental manipulations were capable of mediated uptake of thymidine, uridine, and adenosine. Thymidine uptake by MTP deltaC-expressing oocytes was inhibited by thymine and dTMP. MTP may function in the transport of nucleosides and/or nucleoside derivatives between the cytosol and the lumen of an intracellular membrane-bound compartment.

Multidrug transporter P-glycoprotein 170 as a differentiation antigen on normal human lymphocytes and thymocytes: modulation with differentiation stage and during aging

P-glycoprotein 170 (P-gp), the multidrug transport pump, excludes drugs from the interior of cells and is inhibited by agents such as cyclosporin A (CsA), verapamil, and FK-506, which are also substrates for the P-gp pump. This work documents the age- and differentiation-related changes in P-gp on T and B lymphocytes from human blood or spleen, and its absence on most thymus and bone marrow cells. Analysis of rhodamine 123 (Rh123) dye efflux, and its inhibition by cyclosporin A, was used as a quantitative measure of functional P-gp, and reactivity with MRK-16 was used as a measure of P-gp surface expression. The dye efflux and phenotypic expression of P-gp+ PBMC appeared equivalent to that of a moderately drug-resistant cell line, although efflux is prolonged. The sensitivity to inhibition by CsA, cyclosporin G (CsG), and PSC833 of P-gp on PBMC, thymocytes, or T-cell lines varied with apparent cell-type specificity. Normal blood and splenic T- or B-cells included 50-80% of cells with surface P-gp (MRK-16+), which mediated CsA-sensitive dye export. The proportion of P-gp+ T- and B-cells was lowest among children under age 10 years, increased in adulthood, and decreased after age 60. Thymus included 30% of P-gp+ cells mediating CsA-sensitive dye export, including CD3-4-8- progenitors and mature CD3hi CD4+8- or CD4-8+ thymocytes. Mature T-cells in cord or adult blood, spleen, and bone marrow included a large proportion (50-60%) with efficient CsA-sensitive dye export, preferentially among the CD45RA+ subset. Monocytes from all tissue sources, and most bone marrow cells, expressed surface P-gp but retained Rh123, suggesting the absence of a functional dye export mechanism. In vitro mitogen-stimulated PBMC T and B lymphocytes lost P-gp function within 4-24 hr, consistent with the observation that P-gp was reduced on antigen-experienced CD45R0+ T-cells in vivo. Drug export by P-gp may protect lymphocytes from toxic effects of CsA, and may contribute to the immunosuppressive effects of such drugs. The developmentally regulated expression of P-gp function on lymphocytes, and its modulation on activated T- or B-cells, suggest an important role in normal immune development.

Cloning and functional expression of a complementary DNA encoding a mammalian nucleoside transport protein

Expression screening in Xenopus oocytes was used to isolate a cDNA from rat jejunal epithelium encoding a Na(+)-dependent nucleoside transport protein (named cNT1). The cDNA sequence of cNT1 predicts a protein of 648 amino acids (relative molecular mass 71,000) with 14 potential transmembrane domains. Data base searches indicate significant sequence similarity to the NUPC proton/nucleoside symporter of Escherichia coli. There is no sequence similarity between cNT1 and proteins of mammalian origin. Functionally, cNT1 exhibited the transport characteristics of the nucleoside transport system cit (selective for pyrimidine nucleosides and adenosine) and accepted both 3'-azido-3'-deoxythymidine (AZT) and 2',3'-dideoxycytidine (ddC) as permeants (Km = 0.49 and 0.51 mM, respectively). The demonstration of transport of AZT by cNT1 expressed in Xenopus oocytes provides the first direct evidence that AZT enters cells by transporter-mediated processes, as well as by passive diffusion. Consistent with the tissue distribution of system cit transport activity, transcripts for cNT1 were detected in kidney as well as jejunum. cNT1 therefore belongs to a potential new gene family and may be involved in the intestinal absorption and renal handling of pyrimidine nucleoside analogs used to treat acquired immunodeficiency syndrome (AIDS).

Quantification of mitochondrial DNA in heteroplasmic fibroblasts with competitive PCR

Kearns-Sayre syndrome (KSS) is a disease with severe clinical symptoms that often arises from a mitochondrial DNA deletion of 4977 bp. Quantification of defective mitochondrial DNA is important since the severity of symptoms in KSS is thought to be related to increased content of abnormal mitochondrial DNA. We developed a rapid, quantitative and competitive PCR assay to measure both wild-type and mutant forms of mitochondrial DNA in cells from KSS patients. The assay can accurately measure absolute numbers of mitochondrial DNA per cell by normalizing to a single copy nuclear gene.

Functional expression of the nitrobenzylthioinosine-sensitive nucleoside transporter of human choriocarcinoma (BeWo) cells in isolated oocytes of Xenopus laevis

Cultured human choriocarcinoma (BeWo) cells have previously been shown to exhibit, in comparison with other cultured cell types, elevated nitrobenzylthioinosine (NBMPR)-sensitive transport activity and large numbers (> 10(7)/cell) of high-affinity NBMPR-binding sites [Boumah, Hogue and Cass (1992) Biochem. J. 288, 987-996]. The present study investigates whether NBMPR-sensitive nucleoside transport activity could be induced in Xenopus laevis oocytes by microinjection of poly(A)+ RNA isolated from proliferating cultures of BeWo cells. Expression of uridine transport activity was assayed by comparing rates of uptake (22 degrees C) of 100 microM [3H]uridine by RNA-injected oocytes with uptake by water-injected or uninjected oocytes. A 4-fold stimulation of uridine uptake (2.0 versus 0.5 pmol/90 min per oocyte) was seen when oocytes were injected with 50 ng of BeWo poly(A)+ RNA, and this stimulation was abolished when the RNA-injected oocytes were assayed in the presence of 10 microM NBMPR. The expressed uridine transport activity in oocytes was highly sensitive to NBMPR, with a 50% reduction seen at 1.1 nM NBMPR (IC50 value). The IC50 value for NBMPR inhibition of uptake of 100 microM [3H]uridine by intact BeWo cells was 1.4 nM. Inward fluxes of [3H]uridine in the RNA-injected oocytes were greatly reduced in the presence of high concentrations (2 mM) of non-radioactive nucleosides (adenosine, thymidine, inosine) that are known permeants of NBMPR-sensitive nucleoside transport processes. These results establish that the abundance of NBMPR-sensitive nucleoside transporter mRNA in poly(A)+ RNA preparations from BeWo cells is sufficient to achieve production of functionally active transporter protein in Xenopus oocytes and that, when expressed in Xenopus oocytes, the transporters exhibit NBMPR sensitivity and permeant selectively similar to that of the native transporters.

Functional expression of Na(+)-dependent nucleoside transport systems of rat intestine in isolated oocytes of Xenopus laevis. Demonstration that rat jejunum expresses the purine-selective system N1 (cif) and a second, novel system N3 having broad specificity for purine and pyrimidine nucleosides

Isolated stage VI oocytes from Xenopus laevis expressed uridine transport activity after microinjection of mRNA from rat jejunum. Uridine uptake during 30 min (10 microM, 20 degrees C) by mRNA-injected oocytes reached 2.5 pmol/oocyte, compared with endogenous uptake by water-injected oocytes of about 0.05 pmol/oocyte. The expressed transport activity was 96% Na(+)-dependent, saturable (apparent Km = 15 microM) and inhibited by phloridzin (IC50 = 100 microM). Nucleoside inhibition studies resolved the expressed transport activity into two components: 1) a novel Na(+)-dependent system of broad purine and pyrimidine specificity that was inhibited by low concentrations of guanosine, inosine, adenosine, uridine, thymidine, and cytidine and 2) a Na(+)-dependent system of narrower specificity that was inhibited by low concentrations of guanosine, inosine, adenosine, and uridine and by high concentrations of thymidine and cytidine. The characteristics of the latter system are consistent with those of the Na(+)-dependent nucleoside transport system N1 (cif), previously identified in a number of cell types and tissues, including intestinal epithelia and cultured cells of intestinal origin. The broad specificity system, which was also detected in mRNA-injected oocytes using thymidine as permeant, has been given the provisional designation N3 to distinguish it from the previously described N1 (purine-selective) and N2 (pyrimidine-selective) Na(+)-linked nucleoside transporters. Rat jejunal transporters N1 and N3 were both expressed maximally by the same mRNA size fraction (1.6-3.0 kb, peak 2.3 kb).

Inhibitory effects of propentofylline on [3H]adenosine influx. A study of three nucleoside transport systems

The neuroprotective effects of adenosine are well-recognized. Recently, propentofylline, a xanthine derivative, has been shown to increase extracellular concentrations of adenosine in ischemic brain and to limit the extent of neuronal damage in experimental models of cerebral ischemia. Since the concentration of adenosine in brain is controlled, in part, by nucleoside transporter proteins, the action of propentofylline was proposed to be due to inhibition of mediated transfer of adenosine across cell membranes. To determine the likelihood of this mechanism, we examined the inhibitory effects of propentofylline on [3H]adenosine transport by the three best-characterized nucleoside transport processes, es, ei, and cif in cultured cell lines under conditions where only a single transporter type was operative. Propentofylline inhibited [3H]adenosine uptake by each of the three transport processes in a concentration-dependent manner. The greatest inhibitory potency was for es transporters (L1210/B23.1 cells), with an IC50 value of 9 microM, followed by ei transporters, with IC50 values of 170 microM (L1210/C2 cells) and 166 microM (Walker 256 cells). Propentofylline was a weak inhibitor of cif transporter, with an IC50 value of 6 mM. These results demonstrate that propentofylline is an inhibitor of adenosine transport processes and suggest that its neuroprotective effects may be due to an increase in extracellular concentrations of adenosine by virtue of inhibition of es transporter function.

CSF-1 stimulates nucleoside transport in S1 macrophages

We have examined nucleoside transport (NT) in a cell line derived from primary day 7 murine bone marrow macrophages (S1 macrophages) in response to the macrophage growth factor, colony-stimulating factor 1 (CSF-1). Adenosine and uridine transport in quiescent S1 macrophages occurred primarily by two facilitated diffusional routes, one that was sensitive and one that was relatively resistant to the inhibitor nitrobenzylthioinosine (NBMPR). Addition of CSF-1 to quiescent cultures resulted in increased adenosine and uridine transport with biphasic kinetics with respect to the cell cycle. Basal NT activity was elevated (about twofold) within 15 min of CSF-1 addition, returned to near basal levels by 1 h, and then increased again (three- to fourfold) 8-12 h later, returning again to basal levels by 48 h post CSF-1 stimulation. We propose that the large increase in NT activity at 8-12 h corresponded with the time when cultures synchronously began to enter the S phase of the cell cycle. In addition to these changes in the absolute rates, the proportions of NBMPR-sensitive and NBMPR-insensitive transport also change after CSF-1 addition. Quiescent cultures exhibited primarily NBMPR-insensitive transport while logrithmically growing cultures exhibited primarily NBMPR-sensitive nucleoside transport activity. The increase in the NBMPR-sensitive component of the transport process paralleled a similar increase in the number of high-affinity NBMPR binding sites, suggesting that the mechanism for upregulating NBMPR-sensitive NT activity involves increases in the number of NBMPR-sensitive transporter sites. Interestingly, we were unable to detect Na(+)-dependent concentrative uptake of adenosine, uridine, or formycin-B either in the S1 macrophage cell line or in primary (day 7) murine macrophages. Thus these bone marrow derived macrophages did not display the characteristically large Na(+)-dependent transport systems observed by others in peritoneal macrophages, implying that these two populations of macrophages are, indeed, functionally distinct.

Expression of high levels of nitrobenzylthioinosine-sensitive nucleoside transport in cultured human choriocarcinoma (BeWo) cells

We have examined binding of [3H]nitrobenzylthioinosine (NBMPR) and influx of [3H]thymidine in adherent cultures of human choriocarcinoma (BeWo) cells and, for comparison, cervical-carcinoma (HeLa) cells. Specific association of NBMPR with BeWo cells at 22 degrees C required 1.5 h to reach an equilibrium between free and bound ligand, whereas association with HeLa cells required 20-30 min. Scatchard analysis of NBMPR binding to low-density cultures of BeWo cells revealed a total of 27 x 10(6) sites per cell, consisting of two distinct populations that differed in their affinities for NBMPR. One population bound NBMPR with 'high' affinity (Bmax.1 15.0 pmol/10(6) cells; Kd1 0.6 nM) and the other, larger, population bound NBMPR with 'low' affinity (Bmax.2 29.0 pmol/10(6) cells; Kd2 14.5 nM). By contrast, HeLa cells possessed only 4.1 x 10(5) sites per cell, and these sites all bound NBMPR with the same affinity (Bmax. 0.7 pmol/10(6) cells; Kd 0.5 nM). Interaction of NBMPR with both populations of sites in BeWo cells could be blocked by nitrobenzylthioguanosine (NBTGR), dilazep or dipyridamole. Concentration-effect relationships for dilazep inhibition of binding of 1 nM- and 25 nM-NBMPR to BeWo cells were monophasic, with virtually complete inhibition achieved at 0.1 microM and 1 microM respectively. Plasma-membrane preparations from BeWo cells also had high numbers of NBMPR-binding sites, and u.v. irradiation of site-bound [3H]NBMPR in such preparations labelled polypeptides that migrated in electrophoretograms as a broad band with a peak M(r) of 60,000. The concentration-effect relationship for NBMPR inhibition of thymidine transport by BeWo cells was biphasic, with an IC50 for inhibition of the 'NBMPR-sensitive' component of 1.6 nM and a substantial (15-20%) component of flux that was not inhibited by 10 microM-NBMPR and was thus 'NBMPR-insensitive'. Vmax. values for thymidine transport by BeWo cells were 20-30-fold larger than the corresponding values for transport by HeLa cells. Elimination of the Na+ gradient had no effect on initial rates of thymidine fluxes measured in either the presence or the absence of NBMPR. Our results demonstrate that BeWo cells have an unusually large capacity for NBMPR-sensitive nucleoside transport, apparently resulting from high levels of expression of 'erythrocyte-like' transport elements, identified by their high-affinity interaction with NBMPR. The relationship of the low-affinity binding sites to NBMPR-sensitive transporter elements is uncertain.

Nucleobase transporter-mediated permeation of 2',3'-dideoxyguanosine in human erythrocytes and human T-lymphoblastoid CCRF-CEM cells

Several 2',3'-dideoxynucleosides (ddNs), agents that inhibit the replication of human immunodeficiency virus and hepatitis B virus, enter mammalian cells by simple diffusion. In this report, we show that the membrane permeation of 2',3'-dideoxyguanosine (ddG) in human erythrocytes and CCRF-CEM cells, in contrast with that of other ddNs, is transporter-mediated. Inward fluxes of ddG in both cell types were inhibited by adenine, hypoxanthine, and acyclovir, but not by inhibitors of nucleoside transport (nitrobenzylthioinosine, dipyridamole, dilazep). Fluxes of ddG in human erythrocytes were attributable to a single, rate-saturable process (Km, 380 +/- 90 microM and Vmax, 7.9 +/- 0.8 pmol/s/microliter cell water) that was competitively inhibited by adenine (Ki, 16 microM). These results showed that ddG entered human erythrocytes and CCRF-CEM cells by a transporter-mediated process that was also the basis for entry of purine nucleobases. In contrast, inward fluxes of 2,6-diaminopurine-2',3'-dideoxyriboside (ddDAPR), a prodrug of ddG, were not affected by purine nucleobases or nucleoside transport inhibitors in either cell type. Thus, the permeation properties of ddDAPR resembled those of 2',3'-dideoxyadenosine, a diffusional permeant (cell uptake is transporter-independent), and contrasted with those of ddG, the deamination product of ddDAPR. This study demonstrated that the nucleobase moiety of ddNs is an important determinant of membrane permeation.

A comparison of the abilities of nitrobenzylthioinosine, dilazep, and dipyridamole to protect human hematopoietic cells from 7-deazaadenosine (tubercidin)

Nitrobenzylthioinosine, dilazep, and dipyridamole are potent inhibitors of equilibrative transport of nucleosides that may have pharmacological applications in modulating the therapeutic index of nucleoside antimetabolites used in cancer chemotherapy. We have compared the relative abilities of these inhibitors to reduce the toxicity of in vitro exposures to tubercidin against clonogenic progenitor cells of normal human bone marrow (CFU-GEMM, BFU-E, CFU-GM) and of two leukemic human cell lines (HL-60/C1, CCRF-CEM) that differ in their expression of transporter subtypes. Short (1-h) exposures to 1 microM tubercidin alone inhibited colony formation (a) of normal human hematopoietic progenitors (CFU-GEMM, BFU-E, CFU-GM) by 100%, and (b) of HL-60/C1 and CCRF-CEM cells by > 90%. Pretreatment (30 min) with nitrobenzylthioinosine, dilazep, or dipyridamole followed by simultaneous treatment (1 h) with these transport inhibitors during tubercidin exposures reduced toxicity against hematopoietic progenitors and cell lines. Greater reductions of toxicity were consistently seen with bone marrow progenitors and CCRF-CEM cells than with HL-60/C1 cells. For CFU-GEMM, BFU-E, and CFU-GM cells, reductions in tubercidin toxicity of 50-100% were achieved at these concentrations: > or = 0.1 microM (nitrobenzylthioinosine); > or = 0.1 microM (dilazep); and > or = 3.0 microM (dipyridamole). Pretreatment (30 min) followed by simultaneous treatment (1 h) with any of the transport inhibitors (> or = 0.1 microM) and 0.1 microM [3H]-tubercidin blocked the uptake of radioactivity completely in CCRF-CEM cells and only partially in HL-60/C1 cells. These effects, which were consistent with the nucleoside transport phenotypes of CCRF-CEM cells (inhibitor-sensitive) and HL-60/C1 cells (inhibitor-sensitive and inhibitor-resistant), suggested that protection was due to the inhibition of tubercidin uptake via equilibrative nucleoside transport system(s). Light-density mononuclear cells from human bone marrow, of which the clonogenic progenitors represented only a minor (< 0.01%) subpopulation, possessed far fewer nitrobenzylthioinosine-binding sites (2 x 10(4) sites/cell, Kd = 0.7 nM) than either HL-60/C1 cells (1.7 x 10(5) sites/cell, Kd = 0.9 nM) or CCRF-CEM cells (3.3 x 10(5) sites/cell, Kd = 0.5 nM). Initial rates of uptake of 1 microM [3H]adenosine (0-6 s, 20 degrees C) by human bone marrow mononuclear cells were reduced partially by 0.1 microM inhibitor (nitrobenzylthioinosine > dipyridamole > dilazep) and completely by 10 microM inhibitor.(ABSTRACT TRUNCATED AT 400 WORDS)

L1210/B23.1 cells express equilibrative, inhibitor-sensitive nucleoside transport activity and lack two parental nucleoside transport activities

Cultured mouse leukemia L1210 cells express the nucleoside-specific membrane transport processes designated es, ei, and cif. The es and ei processes are equilibrative, but may be distinguished by the high sensitivity of the former to 6-[(4-nitrobenzyl)thio]-9-beta-D-ribofuranosylpurine (NBMPR); the cif process is mediated by a Na+/nucleoside cotransporter of low sensitivity to NBMPR. Cells of an ei-deficient clonal line, L1210/MC5-1, were mutagenized, and clones were selected in soft agar medium that contained (i) NBMPR (an inhibitor of es processes), (ii) erythro-9-(2-hydorxy-3-nonyl)adenine (an inhibitor of adenosine deaminase), and (iii) arabinofuranosyladenine (a cytotoxic substrate for the three nucleotide transporters). The selection medium did not allow es activity and selected against cells that expressed the Na(+)-linked cif process. Cells of the L1210/B23.1 clonal isolate were deficient in cif transport activity, and inward fluxes of formycin B, a poorly metabolized analog of inosine, were virtually abolished by NBMPR in these cells. In the mutant cells, nonisotopic formycin B behaved as a countertransport substrate during influx of [3H]formycin B, and inward fluxes of the latter were competitively inhibited by purine and pyrimidine nucleosides. The transport behavior of L1210/B23.1 cells indicates that (i) the mutation/selection procedure impaired or deleted the Na(+)-linked cif process and (ii) es nucleoside transport activity is expressed in the mutant cells.

Transport function and subcellular distribution of purified human erythrocyte glucose transporter reconstituted into rat adipocytes

In order to delineate the insulin-independent (constitutive) and insulin-dependent regulations of the plasma membrane glucose transporter concentrations in rat adipocytes, we introduced purified human erythrocyte GLUT-1 (HEGT) into rat adipocytes by poly(ethylene glycol)-induced vesicle-cell fusion and its transport function and subcellular distribution in the host cell were measured. HEGT in adipocytes catalysed 3-O-methylglucose equilibrium exchange with a turnover number that is indistinguishable from that of the basal adipocyte transporters. However, insulin did not stimulate significantly the HEGT function in adipocytes where it stimulated the native transporter function by 7-8-fold. The steady state distribution and the transmembrane orientation assays revealed that more than 85% of the HEGT that were inserted in the physiological, cytoplasmic side-in orientation at the adipocytes plasma membrane were moved into low-density microsomes (LDM), while 90% of the HEGT that were inserted in the wrong, cytoplasmic side-out orientation were retained in the plasma membrane. Furthermore, more than 70% of the LDM-associated HEGT were found in a small subset of LDM that also contained 80% of the LDM-associated GLUT-4, the insulin-regulatable, native adipocyte glucose transporter. However, insulin did not cause redistribution of HEGT from LDM to the plasma membrane under the condition where it recruited GLUT-4 from LDM to increase the plasma membrane GLUT-4 content 4-5-fold. These results demonstrate that the erythrocyte GLUT-1 introduced in adipocytes transports glucose with an intrinsic activity similar to that of the adipocyte GLUT-1 and/or GLUT-4, and enters the constitutive GLUT-4 translocation pathway of the host cell provided it is in physiological transmembrane orientation, but fails to enter the insulin-dependent GLUT-4 recruitment pathway. We suggested that the adipocyte plasma membrane glucose transporter concentration is constitutively kept low by a mechanism where a cell-specific constituent interacts with a cytoplasmic domain common to GLUT-1 and GLUT-4, while the insulin-dependent recruitment requires a cytoplasmic domain specific to GLUT-4.

Effects of transformation by v-fps on nucleoside transport in Rat-2 fibroblasts

Important cellular nutrients, including nucleosides and hexose sugars, are rapidly taken up by cells, largely through mediated carrier systems. The present study examined nucleoside and hexose transport activity in normal Rat-2 fibroblasts and clonal derivatives that expressed either the wild-type (C10) or a temperature-sensitive mutant (NA9) form of v-fps, a transforming protein-tyrosine kinase. Initial uptake rates (transport) of adenosine, thymidine, 3-O-methylglucose and 2-deoxyglucose were greater in v-fps-transformed cells than in normal cells. Elevated transport rates were seen in cells that expressed the temperature-sensitive mutant v-fps only after growth at a temperature that was permissive for protein-tyrosine kinase activity. Nucleoside transport rates declined with increasing cell density in both normal and v-fps transformed cells. Analysis of the sensitivity of adenosine transport to inhibition by nitrobenzylthioinosine (NBMPR) indicated that Rat-2 fibroblasts, like many other rat cell types, possess at least two nucleoside transport systems, which can be distinguished by differences in sensitivity to NBMPR. Although both transport activities were elevated in v-fps-transformed cells, a greater increase was seen in the NBMPR-sensitive component than in the NBMPR-insensitive component. Mass law analysis of the binding of [3H]NBMPR indicated that transformed cells had either the same number (NA9) or a smaller number (C10) of NBMPR-binding sites than normal cells, and photolabelling of membrane proteins with [3H]NBMPR identified polypeptides with similar electrophoretic mobilities (55-75 kDa) in both normal and transformed cells. Thus transformation by v-fps resulted in an increase in NBMPR-sensitive transport activity which was not related to either the number of NBMPR-binding sites or the apparent molecular mass of NBMPR-binding polypeptides.

Glucose and nucleoside transporters of human erythrocytes: effects of detergents on immunoadsorption of a membrane protein to its monoclonal antibody

Immunoadsorption of membrane proteins solubilized in detergents has been used widely for identification, purification and quantitation of transporters and receptors. In an effort to separate the glucose and nucleoside nucleoside transporters of human erythrocytes (GT and NT, respectively) that copurify in a membrane protein fraction band 4.5, we examined in the present study the effects of seven different detergents on the immunoadsorption of GT to its monoclonal antibody, 65D4 (Craik, et al. (1988) Biochem. Cell Biol. 66, 839-852). The following results were obtained. (1) The maximum extent of the immunoadsorption of GT by 65D4 varied between 52 to 98% in different detergents. For non-ionic detergents, there was an apparent inverse correlation between the maximum immunoreactivity of GT and the aggregation number or micellar size of detergents. (2) The immunoprecipitate of GT by 65D4 was contaminated with nucleoside transporter to an extent that varied from 2 to 35 mol% in different detergents. There is an inverse correlation between the extent of the contamination and the detergent aggregation number. However, this contamination was quantitatively accounted for by a time-dependent, non-specific aggregation of NT with GT in detergents. (3) A high degree of purification of NT in band 4.5 by immunoadsorptive removal of GT with 65D4 was achieved in C12E8 as predicted by the observed low NT-GT aggregation and the relatively high epitope-accessibility of GT in this detergent. Based on these findings, we conclude that certain detergents can reduce the immunoreactivity of membrane proteins significantly by modulating epitope accessibility, and may also produce a false immuno-cross-reactivity by inducing nonspecific protein aggregation.

5'-S-(2-aminoethyl)-N6-(4-nitrobenzyl)-5'-thioadenosine (SAENTA), a novel ligand with high affinity for polypeptides associated with nucleoside transport. Partial purification of the nitrobenzylthioinosine-binding protein of pig erythrocytes by affinity chromatography

Derivatives of N6-(4-aminobenzyl)adenosine (substituted at the aminobenzyl group) and 5'-linked derivatives of N6-(4-nitrobenzyl)adenosine (NBAdo) were evaluated as inhibitors of site-specific binding of [3H]nitrobenzylthioinosine (NBMPR) to pig erythrocyte membranes. Potent inhibitors were SAENTA [5'-S-(2-aminoethyl)-N6-(4-nitrobenzyl)-5'-thioadenosine] and acetyl-SAENTA (the 2-acetamidoethyl derivative of SAENTA). SAENTA was coupled to derivatized agarose-gel beads (Affi-Gel 10) to form an affinity matrix for chromatographic purification of NBMPR-binding polypeptides, which in pig erythrocytes are part of, or are associated with, the equilibrative nucleoside transporter. When pig erythrocyte membranes were solubilized with octyl glucoside (n-octyl beta-D-glucopyranoside) and applied to SAENTA-Affi-Gel 10 (SAENTA-AG10), polypeptides that migrated as a broad band on SDS/PAGE with an apparent molecular mass of 58-60 kDa were selectively retained by the affinity gel. These polypeptides were identified as components of the nucleoside transporter of pig erythrocytes by reactivity with a monoclonal antibody (mAb 11C4) that recognizes the NBMPR-binding protein of pig erythrocytes. Retention of the immunoreactive polypeptides by SAENTA-AG10 was blocked by NBAdo. The immunoreactive polypeptides were released from SAENTA-AG10 by elution under denaturing conditions with 1% SDS or by elution with detergent solutions containing competitive ligands (NBAdo or NBMPR). A 72-fold enrichment of the immunoreactive polypeptides was achieved by a single passage of solubilized, protein-depleted membranes through a column of SAENTA-AG10, followed by elution with detergent solutions containing NBAdo. These results demonstrate that polypeptide components of NBMPR-sensitive nucleoside-transport systems may be partly purified by affinity chromatography using gel media bearing SAENTA groups.

Effects of inhibition of N-linked glycosylation by tunicamycin on nucleoside transport polypeptides of L1210 leukemia cells

Membrane polypeptides (relative mass (Mr) 48,000--55,000) associated with the equilibrative transport of nucleosides were identified in cultured murine leukemia (L1210/C2) cells by site-specific photolabeling with [3H]nitrobenzylthioinosine ([3H]NBMPR). Growth of cells in the presence of tunicamycin resulted in the gradual conversion of 3H-labeled polypeptides to a form that migrated more rapidly (Mr 42,000--47,000) during sodium dodecyl sulfate (SDS)--polyacrylamide gel electrophoresis. When plasma membrane fractions were photolabeled and incubated with O-glycanase or endoglycosidase F, the [3H]NBMPR-labeled polypeptides migrated in SDS-polyacrylamide gels with the same mobility as native NBMPR-binding polypeptides, whereas incubation with either N-glycanase or trifluoromethane sulfonic acid converted [3H]NBMPR-labeled polypeptides to the more rapidly migrating form (Mr 41,000--48,000). These observations are consistent with the presence of N-linked oligosaccharides of the complex type on the NBMPR-binding polypeptides of L1210/C2 cells. Tunicamycin exposures that reduced incorporation of [3H]mannose into plasma membrane fractions by greater than 95% had little, if any, effect on either the affinity (Kd values, 0.1-0.2 nM) or abundance (Bmax values, 200,000--220,000 sites/cell) of NBMPR-binding sites, whereas uridine transport kinetics at 37 degrees C were altered in a complex way. Thus, although N-linked glycosylation is not required for insertion of the NBMPR-binding protein into the plasma membrane or for interaction of NBMPR with the high-affinity binding sites, it is important for function of at least one of the three nucleoside transporters expressed by L1210/C2 cells.

Effect of duration of exposure to verapamil on vincristine activity against multidrug-resistant human leukemic cell lines

Verapamil sensitizes multidrug-resistant cell lines to various heterocyclic anticancer drugs by inhibition of energy-dependent release of drug, presumably by interaction with membrane glycoproteins involved in drug efflux. This work assessed verapamil sensitization of human multidrug-resistant lymphocytic and myeloid leukemic cell lines (CEM/VLB100, HL-60/AR) to vincristine during exposures of short duration (4 h). When cells were transferred to drug-free medium immediately after simultaneous 4-h exposures to vincristine and verapamil, the antiproliferative activity of vincristine was not altered in CEM/VLB100 cells and was only moderately increased in HL-60/AR cells. In contrast, when cells were transferred to verapamil-containing medium, vincristine activity was greatly increased against both CEM/VLB100 and HL-60/AR cells. Verapamil enhanced accumulation and inhibited release of [3H]vincristine by CEM/VLB100 and HL-60/AR cells, indicating that the sensitization was due to an increase in cell-associated vincristine after transfer of cells to vincristine-free medium. Slot blot analysis of cellular RNA with the pMDR1 probe revealed high levels of expression of the mdr1 gene in CEM/VLB100 cells but no detectable expression in HL-60/AR cells. Consistent with this finding, polypeptides (Mr 170,000 to 180,000) that were recognized by a monoclonal antibody (C219) against P-glycoprotein were greatly overexpressed in CEM/VLB100 cells, but were expressed at low levels, if at all, in HL-60/AR cells. These results demonstrate the importance of duration of exposure to verapamil in reversing multidrug resistance, not only in cells that overexpress P-glycoprotein but also in cells, such as HL-60/AR, that express little, if any, P-glycoprotein.

Comparison of Coulter volumes with radiometrically determined intracellular water volumes for cultured cells

During methotrexate-induced differentiation of cultured human choriocarcinoma (BeWo) cells, proliferation is inhibited, morphologic and biochemical changes occur, and giant, often multinucleated, cells form. We have used the increase in cell volume as a marker of the mature syncytiotrophoblastlike phenotype. Uninduced and differentiated BeWo cells are not spherical, and theoretical considerations suggested that deviations in shape could result in significant errors in Coulter volume. To determine if the values obtained by electrical pulse sizing reflected the actual mass of BeWo cells, we have evaluated the relationship between Coulter volumes and intracellular water volumes obtained using a shape-independent estimate for eight cell types. A close correlation (r2 = 0.97) was found, indicating that cell volume changes in populations of irregularly shaped cells can be accurately measured using a Coulter instrument.

Some pharmacological activities of novel adenine-related compounds isolated from a marine sponge Agelas mauritiana

Agelasimine A and agelasimine B, two novel compounds related to adenine, have been isolated from the orange sponge, Agelas mauritiana, and have been tested for a variety of biological activities. Both compounds inhibited proliferation of cultured L1210 leukemia cells at nanomolar concentrations with accumulation in the G1 stage of the cell cycle. However, no prolongation of life was observed in mice bearing P388 leukemia treated with these compounds. In the rat isolated aorta, micromolar concentrations of agelasimines were very effective in inhibiting contractions elicited by potassium chloride but had little or no effect on responses for prostaglandin F2 alpha and had modest effects on the responses to noradrenaline and significant effects on 5-hydroxytryptamine. Agelsamines A and B appeared to be equipotent in causing relaxation in rabbit jejunum and bovine coronary artery, and they also inhibited nucleoside transport into rabbit erythrocytes in micromolar concentrations.

Identification of glucose and nucleoside transport proteins in neonatal pig erythrocytes using monoclonal antibodies against band 4.5 polypeptides of adult human and pig erythrocytes

Cytochalasin B and nitrobenzylthioinosine (NBMPR), which inhibit membrane transport of glucose and nucleosides, respectively, have served as photoaffinity ligands that become covalently linked at inhibitor binding sites on transporter-associated proteins. Thus, when membranes from erythrocytes of neonatal pigs with site-bound [3H]cytochalasin B or [3H]NBMPR were irradiated with uv light, two labeled membrane polypeptides (peak Mr values: 55,000 and 64,000, respectively) were identified. Treatment of the photolabeled membranes with endoglycosidase F increased the mobility of [3H]cytochalasin B- and [3H]NBMPR-labeled material (peak Mr values: 44,000 and 57,000, respectively) and limited digestion with trypsin yielded different polypeptide fragments (Mr values: 18,000-23,000 and 43,000, respectively). Identification of the photolabeled polypeptides as transporter components was established using monoclonal antibodies (MAbs) raised against partially purified preparations of band 4.5 from erythrocytes of adult pigs and humans. MAbs 65D4 and 64C7 (anti-human band 4.5), raised in this study, reacted with [3H]cytochalasin B-labeled material from membranes of human erythrocytes and bound to permeabilized erythrocytes but not to intact cells. MAb 65D4 also bound to erythrocytes of mice and neonatal pigs and to a variety of cultured cells (mouse, human, rat), including AE1 mouse lymphoma cells, which lack an NBMPR-sensitive nucleoside transporter. Also employed was MAb 11C4 (anti-pig band 4.5), which recognizes the NBMPR-binding protein of erythrocyte membranes from adult pigs. When membrane proteins from neonatal and adult pigs were subjected to electrophoretic analysis and blots were probed with different MAbs, MAb 65D4 (anti-human band 4.5) bound to material that comigrated with [3H]cytochalasin B-labeled polypeptides (band 4.5) from neonatal, but not adult, pig erythrocytes, whereas MAb 11C4 (anti-pig band 4.5) bound to material that comigrated with [3H]NBMPR-labeled band 4.5 polypeptides of erythrocytes from both neonatal and adult pigs. These results, which indicate structural differences in the cytochalasin B- and NBMPR-binding proteins of pig erythrocytes, establish the presence of both proteins in erythrocytes of neonatal pigs and suggest that only the NBMPR-binding protein is present in erythrocytes of adult pigs.

External location of sites on pig erythrocyte membranes that bind nitrobenzylthioinosine

Nucleoside transport in erythrocytes of various species is inhibited by the binding of nitrobenzylthioinosine (NBMPR) to high affinity sites associated with nucleoside transport elements of the plasma membrane. The present study examined binding of [3H]NBMPR to unsealed ghosts and to sealed right-side-out vesicles (ROVs) and inside-out vesicles (IOVs) prepared from pig erythrocytes. Kd values for NBMPR dissociation from the ligand-site complex in unsealed ghosts, ROVs and IOVs were similar (1.6-2.4 nM), and Bmax values (mean +/- SD) were, respectively, 22.2 +/- 5.5, 25.8 +/- 6.4, and 37.3 +/- 4.0 molecules/fg of protein, reflecting differences in the protein content of the membrane preparations. When temperatures were decreased from 22 degrees to 4 degrees, NBMPR binding to erythrocyte membrane preparations was reduced in IOVs relative to that in unsealed ghosts and ROVs. At 22 degrees, the association of NBMPR molecules with IOVs was slower than with ROVs and unsealed ghosts, differences that were virtually eliminated by permeabilization of the membrane preparations with saponin. Thus, the binding sites were more accessible to external NBMPR in sealed ROVs and unsealed ghosts than in sealed IOVs, indicating that the NBMPR sites are located on the extracellular aspect of the membrane.

Inhibition of methotrexate-induced differentiation of cultured human choriocarcinoma (BeWo) cells by thymidine

During exposure to methotrexate, cultured human choriocarcinoma (BeWo) cells stop proliferating, enlarge, and undergo a complex differentiative response that resembles in utero development of quiescent syncytiotrophoblasts. In the present work, complete inhibition of proliferation and maximal cell enlargement required exposure to 1 microM methotrexate, whereas colony-forming ability, determined after transfer of cells to drug-free medium, was unaffected over a wide range of concentrations (10(-12)-10(-5) M). BeWo cells were sensitive to the antifolate effects of methotrexate since thymidylate synthase activity and incorporation of [14C]formate into DNA, RNA, and protein were reduced by greater than 90% after short drug exposures, and progression of cells through S phase of the cell cycle was blocked by prolonged drug exposures. When methotrexate was coadministered with hypoxanthine and thymidine or leucovorin, its antiproliferative and differentiative effects were blocked. When methotrexate was coadministered with either hypoxanthine or thymidine, its antiproliferative activity was unaffected, whereas expression of syncytiotrophoblastic markers was blocked in the presence of thymidine but not in the presence of hypoxanthine. Exposure of BeWo cells to fluorodeoxyuridine also stimulated cell enlargement and expression of syncytiotrophoblastic markers, and these effects were blocked by coadministration of thymidine. Thus BeWo cells, which were sensitive to the antifolate effects of methotrexate, were not killed during cytostasis but instead entered a reversible differentiated state, apparently resulting from thymidylate starvation and consequent inhibition of DNA synthesis.

Characterization of monoclonal antibodies that recognize band 4.5 polypeptides associated with nucleoside transport in pig erythrocytes

Three monoclonal antibodies have been raised against partially purified band 4.5 polypeptides [Steck (1974) J. Cell Biol. 62, 1-19] from pig erythrocyte membranes. The antibodies were capable of binding to both intact pig erythrocytes and protein-depleted membrane preparations and recognized detergent-solubilized polypeptides from adult and neonatal pig erythrocytes that were photolabelled with [G-3H]nitrobenzylthioinosine (NBMPR), a potent specific inhibitor of nucleoside transport. The antibodies did not recognize polypeptides from neonatal pig erythrocytes that were photolabelled with the glucose-transport inhibitor [3H]cytochalasin B. Reactivity with polypeptides of apparent Mr 64,000 [10% (w/v) acrylamide gels] was demonstrated by Western-blot analysis. The antibodies recognized pig band 4.5 polypeptides after prolonged treatment with endoglycosidase F, a finding consistent with reactivity against polypeptide, rather than carbohydrate, determinants. Trypsin digestion of NBMPR-labelled protein-depleted pig erythrocyte membranes generated two labelled polypeptide fragments (Mr 43,000 and 26,000). Two of the antibodies recognized both fragments on Western blots, whereas the third bound to the larger, but not to the smaller, fragment. The antibodies had no significant effect on reversible binding of NBMPR to protein-depleted pig erythrocyte membranes and did not bind to NBMPR-labelled polypeptides in human, rabbit or mouse erythrocytes.

Density-dependent inhibition of expression of syncytiotrophoblastic markers by cultured human choriocarcinoma (BeWo) cells

In the presence of methotrexate, cultured human choriocarcinoma (BeWo) cells undergo a differentiative response that resembles normal trophoblastic development. In the current study, the effects of cell number and population density on drug-induced conversion of BeWo cells from the cytotrophoblastlike to the syncytiotrophoblastlike phenotype were investigated using as markers of differentiation formation of "giant" cells, a process shown to require exogenous purines, and expression of placental (heat-stable) alkaline phosphatase. Giant cell formation, assessed by determination of cell volumes, was reduced in crowded cultures, and addition of hypoxanthine to growth media partially restored methotrexate-induced cell enlargement. Cellular uptake of methotrexate, assessed by following the loss of methotrexate from cell culture fluids during drug exposures, was two-threefold greater in sparsely populated than in densely populated cultures. Although the concentration of methotrexate in culture fluids of crowded cultures declined during exposures of 48 hr, the amount of extracellular drug remaining at 48 hr was well above the threshold for induction of the differentiative response. When culture population was held constant and population density was manipulated by varying the substratum available to cells, methotrexate-induced cell enlargement was inversely related to population density. Expression of placental alkaline phosphatase, salvage of exogenous hypoxanthine, and synthesis of RNA were also reduced at high population densities. These results indicate that expression of markers of methotrexate-induced differentiation of BeWo cells was inhibited in a density-dependent manner that may have been related to reduced cellular uptake of the inducing agent and of exogenous nutrients (purines) from culture fluids.

The effects of hypoxanthine on methotrexate-induced differentiation of cultured human choriocarcinoma (BeWo) cells

When cultured human choriocarcinoma (BeWo) cells are exposed to methotrexate, proliferation ceases and cells undergo a complex differentiative response that resembles development of normal trophoblast. Although thymidylate starvation has been shown to be causative in methotrexate-induced expression of syncytiotrophoblastic markers by BeWo cells, the role of purine deprivation is uncertain since previous studies utilized growth media containing exogenous purines. This work investigated the effects of hypoxanthine on methotrexate-induced cell enlargement, expression of placental alkaline phosphatase, and morphological differentiation to the syncytiotrophoblast-like phenotype. When methotrexate exposures (1 microM, 48 h) were conducted in a purine-free basal medium supplemented with dialyzed fetal bovine serum, RNA synthesis was greatly reduced and cell enlargement did not occur. Specific methods for removing purines (charcoal extraction and xanthine oxidase treatment) decreased the ability of serum to support cell enlargement during methotrexate exposures, whereas addition of hypoxanthine to culture fluids restored its ability to support maximal increases in cell mass, confirming that purines were the factors lost during dialysis. In contrast, morphologically differentiation to the syncytiotrophoblast-like phenotype and increased expression of placental alkaline phosphatase were unaffected by the availability of purines during exposure to methotrexate.

Differential cellular retention of vincristine and vinblastine by cultured human promyelocytic leukemia HL-60/Cl cells: the basis of differential toxicity

Differential toxicity of vincristine and vinblastine against cells of a cloned subline of human promyelocytic leukemia (HL-60/Cl) was dependent on exposure conditions. During continuous exposures of 48 h, vincristine and vinblastine were equitoxic with drug concentrations that inhibited proliferation rates by 50% of 7.6 and 8.1 nM, respectively. When cells were subjected to 4-h exposures and transferred to drug-free medium, the drug concentration of vinblastine that inhibited proliferation rates by 50% (1.1 microM) was significantly greater than that of vincristine (41 nM). Analysis by flow cytometry of the effects of equitoxic drug exposures on cell-cycle progression suggested that vincristine and vinblastine acted by the same mechanism (G2-M phase inhibition). [3H]Vincristine and [3H]vinblastine were bound to serum proteins in growth medium to the same extent (25%) over a wide range of concentrations, and the amounts of "free" extracellular drug did not decrease during prolonged exposures. Analysis by high-performance liquid chromatography of extracts of cultures incubated with growth-inhibitory concentrations of [3H]vincristine or [3H]vinblastine indicated little, if any, metabolism of either drug by cells or culture fluids; after 24 h, 85-95% of radioactivity was recovered from cells or growth medium as unchanged vincristine or vinblastine. At concentrations from 6 nM to 6 microM, vinblastine entered cells rapidly, reaching maximum levels within 0.5-2 h, and the relationship between maximal cell-associated drug and extracellular free vinblastine was linear. Although uptake of vincristine was slower than that of vinblastine, the cellular content of vincristine reached that of vinblastine during prolonged (12-24 h) exposures, and the amounts of cell-associated drug, relative to extracellular drug concentrations, indicated considerable "concentrative" accumulation (intra: extracellular ratios, greater than 100). When drug exposures were ended by transfer of cells to drug-free medium, vinblastine was released from cells more rapidly and to a greater extent than vincristine, independent of whether exposures were 4 or 24 h. Rates of uptake and release of vinblastine (50 nM) were unaffected by depletion of cellular adenosine triphosphate, suggesting that rapid release was not mediated by an energy-dependent efflux system.

The effect of pH on interaction of nitrobenzylthioinosine and hydroxynitrobenzylthioinosine with the nucleoside transporter of human erythrocyte membranes

Site-specific binding to human erythrocyte membranes of nitrobenzylthioinosine (NBMPR), un-ionized at physiological pH, was compared with that of hydroxynitrobenzylthioinosine (HNBMPR), pKa 6.4, at graded pH values. Binding of [3H]NBMPR was measured directly, and that of HNBMPR was assayed by competitive inhibition by HNBMPR of [3H]NBMPR binding. Kd and Bmax values for binding of [3H]NBMPR to erythrocyte membranes were independent of pH. Kd values for the competing ligand were determined by mass law analysis of equilibrium binding data using either (a) apparent ligand concentration (dissociated plus undissociated forms of HNBMPR) or (b) the concentration of undissociated HNBMPR. Kd values for HNBMPR calculated with the apparent ligand concentration increased 10-fold as the fraction of HNBMPR molecules present in the dissociated form was increased (by pH changes) from 14 to 88%, whereas Kd values for the undissociated form of HNBMPR were independent of pH. The results presented here demonstrate that the undissociated form of HNBMPR binds more tightly to the transport-inhibitory sites of erythrocytes than NBMPR and suggest that ionization of S6-substituted thiopurine ribonucleosides eliminates or greatly decreases their ability to interact with the binding sites.

Inward fluxes of adenosine in erythrocytes and cultured cells measured by a quenched-flow method

Dilazep, a vasodilator previously recognized as an inhibitor of adenosine permeation, very rapidly blocked the uptake of adenosine by cultured L5178Y cells, and accordingly was used as a quencher in a simple quenched-flow system for measuring cellular uptake of nucleosides during very short intervals. Time courses of cellular uptake of adenosine, assayed during intervals between 0.05 and 0.5s with the quenched-flow system, were linear and defined initial rates of adenosine uptake. The latter are rates of inward transport of adenosine. Kinetic constants for that process in cultured S49 cells determined with the quenched-flow procedure were similar to those determined with an assay dependent on manual timing. In studies of adenosine uptake kinetics in human erythrocytes at 22 degrees C and 37 degrees C in which the quenched-flow procedure was used, time courses of adenosine uptake were linear at both temperatures and defined initial uptake rates; kinetic constants (means +/- S.E.M.) at 22 degrees C (n = 8) were Km 25 +/- 14 microM and Vmax. 15 +/- 5 pmol/s per microliter of cell water and at 37 degrees C (n = 3) were Km 98 +/- 17 microM and Vmax. 80 +/- 9 pmol/s per microliter of cell water.

Separation of nitrobenzylthioinosine from high affinity binding sites on human erythrocytes by gel chromatography

The potent inhibitor of nucleoside transport, 6-[(4-nitrobenzyl) mercapto]-9-beta-D-ribofuranosylpurine (NBMPR), binds reversibly, but with high affinity (KD 0.3-1.4 nM) to plasma membrane sites on human erythrocytes, and occupancy of those sites by NBMPR correlates with inhibition of transport. An earlier study from this laboratory showed that, upon photoactivation of site-bound [3H]NBMPR on erythrocyte membranes by exposure to UV light, isotopic ligand molecules became covalently linked to membrane polypeptides, which migrated as a discrete band on gel electrophoretograms, thereby identifying nucleoside transport polypeptides. The present study showed that erythrocytes on which the high affinity sites were occupied by reversibly bound [3H]NBMPR were freed of the latter by slow passage at 37 degrees C through 9 X 300 mm columns of Sephadex G-200 gel. Photoactivation of site-bound [3H]NBMPR on the intact cells caused apparent covalent attachment of the ligand because cells so treated retained the isotopic label during passage through the gel columns. Apparent covalent binding to erythrocytes also resulted from photoactivation of site-bound [3H]nitrobenzylmercaptopurine arabinoside.

Differential activity of vincristine and vinblastine against cultured cells

Vincristine and vinblastine exhibit differential activity against tumors and normal tissues. In this work, a number of cultured cell lines were assayed for their sensitivity to the antiproliferative and cytotoxic effects of the two drugs following short-term (4 hr) or during continuous exposures. Differential activity was not seen when cells were subjected to continuous exposures. The concentrations of vincristine and vinblastine, respectively, that inhibited growth rates by 50% were: mouse leukemia L1210 cells, 4.4 and 4.0 nM; mouse lymphoma S49 cells, 5 and 3.5 nM; mouse neuroblastoma cells, 33 and 15 nM; HeLa cells, 1.4 and 2.6 nM; and human leukemia HL-60 cells, 4.1 and 5.3 nM. In contrast, differential toxicity was seen when cells were subjected to 4-hr exposures and transferred to drug-free medium: the 50% growth-inhibitory concentrations for vincristine and vinblastine, respectively, for inhibition (a) of proliferation of L1210 cells were 100 and 380 nM and of HL-60 cells were 23 and 900 nM and (b) of colony formation of L1210 cells were 6 and greater than 600 nM and of HeLa cells were 33 and 62 nM. Uptake and release of [3H]-vincristine and [3H]vinblastine were examined in L1210 cells under the conditions of growth experiments. Uptake of both drugs was dependent on the pH of culture media, and significantly greater amounts of [3H]vinblastine than of [3H]vincristine were associated with cells after 4-hr exposures to equal concentrations of either drug. When cells were transferred to drug-free medium after 4-hr exposures, vinblastine was released much more rapidly from cells than was vincristine, and by 0.5 hr after resuspension of cells, the amount of vincristine associated with the cells was greater than the amount of vinblastine and remained so for up to at least 6 hr.

Transport of uridine and 3-deazauridine in cultured human lymphoblastoid cells

The transport of uridine and 3-deazauridine was compared in two lines of cultured human lymphoblastoid cells that differ in their sensitivity to 3-deazauridine apparently because of reduced uridine-cytidine kinase activity in the resistant line. The kinetic parameters (+/- S.E.) of uridine transport were similar in the two cell lines: Km, 0.23 +/- 0.02 and 0.25 +/- 0.07 mM; and Vmax, 35 +/- 2 and 57 +/- 10 pmol/microliter of cell water per sec, respectively, for 6410/MP (parental) and 6410/MP/DU (resistant) cells. 3-Deazauridine, while transported with similar kinetic characteristics in both cell lines, was not as good a substrate for the nucleoside transporter as was uridine, and its transport was dependent on pH. Kinetic parameters, determined using calculated concentrations of the undissociated form of 3-deazauridine (pKa, 6.5), were: Km, 0.52 +/- 0.01 and 0.51 +/- 0.03 mM; and Vmax, 28 +/- 0.5 and 24 +/- 0.9 pmol/microliter of cell water per sec, respectively, for 6410/MP and 6410/MP/DU cells. At pH 8, a condition in which 97% of 3-deazauridine molecules are ionized, rates of transport were almost zero. It is concluded that the undissociated form of 3-deazauridine is the substrate for the nucleoside transporter.

The kinetics of dissociation of the inhibitor of nucleoside transport, nitrobenzylthioinosine, from the high-affinity binding sites of cultured hamster cells

Nucleoside transport in various types of animal cells is inhibited by the binding of nitrobenzylthioinosine (NBMPR) to a set of high-affinity sites on the plasma membrane. This work examined the binding of [3H]NBMPR to the nucleoside transporters of cultured Nil 8 hamster fibroblasts and of cells of a virus-transformed clone (Nil SV) derived from Nil 8. Experiments conducted with intact Nil 8 and Nil SV cells and with membrane preparations indicated that the two lines differed significantly in the cellular content of binding sites and only slightly in the affinities of these sites for NBMPR. Nil 8 and Nil SV cells possessed (4.2-8.0) X 10(5) and (2.0-4.0) X 10(6) sites per cell respectively, whereas the dissociation constants of site-bound NBMPR obtained with intact cells and with membrane preparations were similar, ranging from 0.29 to 1.5 nM. Dilazep, a potent inhibitor of nucleoside transport that is structurally unrelated to NBMPR, appeared to compete with NBMPR for binding to the high-affinity sites when tested under equilibrium conditions with Ki values for inhibition of NBMPR binding to Nil 8 and Nil SV cells respectively of 15 +/- 4 and 32 +/- 4 nM. The dissociation of NBMPR from the binding site--NBMPR complex of Nil SV membrane preparations was a first-order decay process with a rate constant of 0.68 +/- 0.26 min-1. The rate of dissociation of NBMPR from the binding-site complex of membrane preparations and intact cells was decreased significantly in the presence of dilazep and increased in the presence of the permeant uridine. These results suggest that the apparent competitive-inhibition kinetics obtained for dilazep under equilibrium conditions should not be interpreted as binding of dilazep to the same site as NBMPR but rather as binding of the two inhibitors to closely associated sites on the nucleoside transporter. Similarly, uridine also appears to bind to a site separate from the NBMPR-binding site.

Resistance to 9-beta-D-arabinofuranosyladenine in cultured leukemia L 1210 cells

A cultured line of L1210 leukemia cells, designated L1210/ara-A, was selected for resistance to 9-beta-D-arabinofuranosyladenine (ara-A) by a series of 72-hr exposures to increasing concentrations of ara-A in the presence of 1 microM deoxycoformycin. Cells of the resistant line were about one-tenth as sensitive as were cells of the parent line to the effects of ara-A on proliferation, viability, and tumorigenicity. Cross-resistance, as determined by comparison of drug effects on rates of proliferation of L1210/C2 and L1210/ara-A cells, was seen with adenosine, deoxyadenosine, methylmercaptopurine ribonucleoside, tubercidin, and cordycepin but not with 1-beta-D-arabinofuranosylcytosine or with 9-beta-D-arabinofuranosyl-2-fluoroadenine. The levels of resistance to methylmercaptopurine ribonucleoside, cordycepin, and tubercidin were considerably greater than that seen with ara-A itself. L1210/C2 and L1210/ara-A cells were compared with respect to the effects of ara-A on cell size distributions, DNA distributions, labeling indices, and apparent rates of DNA synthesis, and the differences seen were consistent with inhibition of DNA synthesis and unbalanced growth as the major mechanism of ara-A cytotoxicity. The decreased sensitivity of DNA synthesis in L1210/ara-A cells treated with ara-A, relative to L1210/C2 cells, was due to reduced intracellular accumulation of ara-A phosphates in the resistant line. Phosphorylation of ara-A, adenosine, and tubercidin, but not deoxyadenosine or deoxycytidine, was greatly reduced in intact L1210/ara-A cells, relative to L1210/C2 cells, and adenosine kinase activity in extracts of L1210/ara-A cells was negligible. Resistance to ara-A, and cross-resistance to tubercidin, methylmercaptopurine ribonucleoside, and cordycepin is attributed to loss of adenosine kinase activity.

Initial rate kinetics of the transport of adenosine and 4-amino-7-(beta-D-ribofuranosyl)pyrrolo[2,3-d]pyrimidine (tubercidin) in cultured cells

A procedure is described for determining early time courses of nucleoside uptake by cultured cells in suspension. Replicate samples of cell suspensions were exposed to medium containing 3H-nucleosides for brief intervals (sec) ended by addition of nitrobenzylthioinosine, a potent inhibitor of nucleoside transport that terminated nucleoside uptake virtually instantaneously. Time courses of nucleoside uptake were constructed from the cellular content of nucleoside acquired by the replicate samples during graded intervals of exposure to the labeled permeant. Such time courses were definitive of cellular uptake of nucleosides during the first few sec of exposure to permeant and yielded initial rates of uptake of adenosine and 4-amino-7-(beta-d-ribofuranosyl)pyrrolo[2,3-d]pyrimidine (tubercidin). Defining initial rates of nucleoside uptake as rates of inward transport, relationships between transport rates and extracellular concentrations of these permeants were evaluated in HeLa cells and in two cultured lines of mouse lymphoma L5178Y cells that differ in their abilities to phosphorylate adenosine and tubercidin. Transport rates for these permeants were similar in the two L5178Y cell types and were saturable in the 3 cell lines with Km values between 14 and 38 microM. Adenosine and tubercidin were mutually competitive permeants in L5178Y cells, indicating that they are substrates for the same transport mechanism.

Comparison of the effects on cultured L1210 leukemia cells of the ribosyl, 2'-deoxyribosyl, and xylosyl homologs of tubercidin and adenosine alone or in combination with 2'-deoxycoformycin

The biologic effects of a series of sugar-substituted analogs of tubercidin were evaluated and compared with the effects of the homologous series of adenosine analogs in combination with 2'-deoxycoformycin. The greatest cytotoxicity against cultured mouse L1210 leukemia cells was exhibited by tubercidin and by 3'-deoxyadenosine or xylosyladenine in combination with 2'-deoxycoformycin. Somewhat less active were xylotubercidin and the combination of arabinosyladenine (araA) plus 2'-deoxycoformycin. The arabinosyl and 2'- and 3'-deoxyribosyl derivatives of tubercidin were relatively ineffective in their ability to inhibit proliferation of L1210 cells. The major biochemical effects of the most active agents were inhibition of RNA synthesis (3'-deoxyadenosine and xylosyladenine) and depletion of cellular ATP plus general inhibition of macromolecular synthesis (tubercidin). The in vitro activities of xylosyladenine and 3'-deoxyadenosine (in combination with 2'-deoxycoformycin) and xylotubercidin (as a single agent) were greater than or equivalent to that of araA (in combination with 2'-deoxycoformycin).

Absence of binding sites for the transport inhibitor nitrobenzylthioinosine on nucleoside transport-deficient mouse lymphoma cells

Cells of an adenosine-resistant clone (AE1) of S49 mouse lymphoma cells were compared with cells of the parental line with respect to (a) characteristics of nucleoside transport, (b) high affinity binding of the inhibitor of nucleoside transport, nitrobenzylthioinosine (NBMPR), and (c) the antiproliferative effects of the nucleoside antibiotics, tubercidin, arabinosyladenine and showdomycin. Rates of inward transport of uridine, thymidine, adenosine, 2'-deoxyadenosine, tubercidin, showdomycin, and arabinosyladenine in AE1 cells were less than 1% of those in cells of the parental S49 line. The inhibitor of nucleoside transport, NBMPR, reduced rates of inward nucleoside transport in S49 cells to levels comparable to those seen in the transport-defective mutant. S49 cells possessed high affinity sites that bound NBMPR (6.6 X 10(4) sites/cell, Kd = 0.2 nM), whereas site-specific binding of NBMPR to AE1 cells was not demonstrable, indicating that loss of nucleoside transport activity in AE1 cells was accompanied by loss of the high affinity NBMPR binding sites. Relative to S49 cells, AE1 cells were resistant to the antiproliferative effects of tubercidin and showdomycin, but differences between the two cell lines in sensitivity toward arabinosyladenine were minor, suggesting that nucleoside transport activity was required for cytotoxicity of tubercidin and showdomycin, but not for that of arabinosyladenine.

Binding of nitrobenzylthioinosine to high-affinity sites on the nucleoside-transport mechanism of HeLa cells

Nitrobenzylthioinosine (NBMPR) binds reversibly, but with high affinity (Kd 0.1--1.2 nM), to inhibitory sites on nucleoside-transport elements of the plasma membrane in a variety of animal cells. The present study explored relationships in HeLa cells between NBMPR binding and inhibition of uridine transport. The Km value for inward transport of uridine by HeLa cells in both suspension and monolayer culture was about 0.1 mM. The affinity of the transport-inhibitory sites for uridine (Kd 1.7 mM), inosine (Kd 0.4 mM) and other nucleoside permeants was low relative to that for NBMPR. The pyrimidine homologue of NBMPR, nitrobenzylthiouridine, also exhibited low affinity for the NBMPR-binding sites. Pretreatment of HeLa cells with p-chloromercuribenzene sulphonate (p-CMBS) or N-ethylmaleimide (NEM) decreased binding of NBMPR to its high-affinity sites and inhibited uridine transport, indicating the presence of thiol groups essential to both processes. NEM, a more penetrable reagent than p-CMBS, inhibited binding and transport at much lower concentrations than the latter compound. Pretreatment of cells with concentrations of p-CMBS that alone had no effect on either NBMPR binding or uridine transport increased the sensitivity of transport to NBMPR inhibition and changed the shape of the NBMPR concentration-effect curve, suggesting synergistic inhibiton of uridine-transport activity by these two agents.

Effect of lithium on the myelosuppressive and chemotherapeutic activities of vinblastine

Serum concentrations of lithium after i.p. administration to normal mice declined with biphasic kinetics. Treatment of mice bearing ascitic L1210 leukemia with lithium resulted in a modest protection against the myelosuppressive effects of vinblastine. Lithium as a single agent was without effect on the survival times of mice bearing either the L1210 or P388 leukemia. Similarly, there was no evidence of significant antiproliferative or cytotoxic activity when cultured L1210 leukemia and murine neuroblastoma cells were exposed to lithium at levels comparable to those observed in the serum of lithium-treated mice. The therapeutic activity of vinblastine against the L1210 and P388 leukemias was not significantly altered by either simultaneous or subsequent administration of lithium. Lithium did not antagonize the antiproliferative or cytotoxic action of vinblastine against L1210 leukemia or murine neuroblastoma cells and was without effect in experiments with neuroblastoma cells that assessed vinblastine inhibition of a biological function dependent on formation of cytoplasmic microtubules (neurite formation induced by serum deprivation). The results obtained suggest that administration of lithium to reduce myelosuppression is not likely to counteract the tumoricidal activity of vinblastine.