Kinetic studies assessing the transport affinities of both hENT1 and hENT2, the first SLC29 family members identified, have been quite extensive. Earlier research identified hENT1 as the es transport system, while hENT2 was identified as the ei system, where the es system bound NMBPR with high affinity (Kd = 1-10 nM) and the ei system was not affected at nanomolar concentrations, but was affected at higher concentrations (>10 |M) (Kong et al., 2004). In fact, since ENT transport is mediated via a bidirectional facilitated diffusion mechanism, which is dependant upon substrate concentration gradient, the difference in binding NMBPR was used quite successfully for the early biochemical characterization of these two transporters.
The substrate affinities for endogenous nucleosides and some therapeutic nucle-oside analogs have been described for hENT1 and hENT2 and recently for hENT3. Both hENT1 and hENT2 have been shown to be broadly selective for both purine and pyrimidine nucleosides; however, hENT2 exhibits 7.7- and 19.3-fold lower affinities for guanosine and cytidine, respectively, but a fourfold higher affinity for inosine (Kong et al., 2004). Further, hENT2 transports nucleobases, whereas hENT1 does not (Yao et al., 2002). In terms of therapeutic nucleoside analogs, hENT1 has been shown to interact with a number of compounds widely used in the treatment of cancer, such as cladribine, gemcitabine, fludarabine, cytarabine, tiazofurin, and benzamide riboside (Griffiths et al., 1997a; Mackey et al., 1998; 1999; Vickers et al., 2002; Damaraju et al., 2005). However, hENT1 only poorly transports the antivirals 2', 3'-dideoxycytidine (ddC) and 2', 3'-dideoxyinosine (ddI) and does not transport 3'-azido-3'-deoxythymidine (AZT), suggesting an important role of the 3'-hydroxyl group on the ribose moiety for substrate recognition (Yao et al., 2001a). Moreover, Vickers et al. (2002) recently demonstrated the importance of the 3'-hydroxyl group with respect to uridine analogs for both hENT1 and hENT2. Nonetheless, hENT2 demonstrated a much broader substrate selectivity, transporting nucleobases (which lack the ribose moiety), a number of the above compounds, as well as ddC, ddI, and AZT (Yao et al., 2001a; Vickers et al., 2002). However the selectivity of hENT2 for cytidine and its analogs is comparatively lower than that of hENT1, suggesting a lack of tolerance for a 4'-amino moiety on the base (Vickers et al., 2002). Both transporters tolerated halogen substitution at the 5' position on the base, as well as the 2' and 5' positions on the ribose moiety of uridine, suggesting these positions are not essential for uridine-like substrate recognition (Vickers et al., 2002).
Interestingly, recent evidence indicates the antidiabetic compound troglitazone, but not the related thiazolidinediones pioglitazone and ciglitazone, inhibits hENT1 transport of adenosine and uridine via a competitive inhibition mechanism (Leung et al., 2005b). Troglitazone had minimal inhibitory effect on hENT2 in this study (Leung et al., 2005b). Interestingly, inhibition of ENT1 in vascular smooth muscle is suggested to increase extracellular adenosine, which causes vasodilation and inhibits vascular smooth muscle cell proliferation (Rubin et al., 2000; Kim et al., 2004; Masaki et al., 2004). In concert, these data suggest a novel therapeutic approach to modulate extracellular adenosine concentrations. To our knowledge, the transport characteristics of other diabetic drugs via nucleoside transporters have not been studied.
Similar to both hENTl and hENT2, hENT3 has demonstrated broad selectivity for nucleosides (Baldwin et al., 2005). Additionally, hENT3 has also been shown to transport the nucleobase adenine; however, in contrast to hENT2 it does not transport hypoxanthine (Baldwin et al., 2005). Interestingly, the transport activity of hENT3 was relatively insensitive to many of the classical ENT inhibitors, such as NBMPR, dipyridamole, and dilazep although transport was found to be highly pH dependent, indicative of its intracellular localization (Baldwin et al., 2005). HENT3 has also demonstrated good transport efficiency for a wide range of pharmaceutical purine and pyrimidine analogs, including cladribine, cordycepin, tubercidin, zebularine, 5-fluoro-2'-deoxyuridine, ddC, ddl, and AZT (Baldwin et al., 2005). Other compounds, such as gemcitabine and gancyclovir were also substrates for hENT3, although with a much lower transport efficiency (Baldwin et al., 2005). Especially intriguing is the high transport efficiency of AZT, especially compared with hENTl, which does not transport AZT, and hENT2, which has been shown to transport AZT, but at one third its capacity of uridine (Yao etal., 2001a). Comparatively, hENT3 exhibits AZT selectivity comparable to adenosine and inosine, which are higher than that for thymidine and uridine, indicating that the 3'-hydroxy group is not important for substrate recognition (Baldwin et al., 2005). Given the high prevalence of 3'-hydroxyl moieties on many antiviral compounds, it is unfortunate that studies exploring structure-function relationships for hENT3 have not been performed to date. Such studies comparative to either hENTl, or especially hENT2 could prove critical to elucidate the mechanism by which these compounds enter the cell to elicit their cytotoxic effects.
Although limited, some literature does explore the structure-function relationships of both hENTl and hENT2. Using chimeric constructs, researchers have shown the primary sites of substrate binding for hENTl reside in TMD three through six, while TMD one to two confer a secondary contribution (Sundaram et al., 200l). Molecularly, studies have shown that a single substitution of M33I was sufficient to alter sensitivity to a number of substrates (Visser et al., 2002), while Glyl79 is important to NBMPR binding (SenGupta et al., 2002). With respect to ENT2, chimeric studies using the rat homologues of ENTl and ENT2 have shown that TMD five and six form at least part of the translocation pathway (Yao et al., 2002). Modification of the corresponding Cysl40 (transmembrane domain four) inhibits transport, suggesting that this residue lies within, or is adjacent to the substrate translocation pathway (Yao et al., 200lb). However, additional studies are required to fully characterize and elucidate the structure-function relationships of each of these transporters.
Tissue expression of CNTl has been shown in epithelial tissues including small intestine, kidney, and liver. In contrast, CNT2 has been shown to be widely distributed in kidney, liver, heart, brain, placenta, pancreas, skeletal muscle, colon, rectum, and throughout the small intestine, while CNT3 has been shown in pancreas, trachea, bone marrow, and mammary gland, with lower levels in intestine, lung, placenta, prostrate, testis, and liver (Gray et al., 2004). Interestingly, only CNT1 and CNT2 were widely expressed on the brush-border membranes of enterocytes along the length of both the fetal and adult small intestines, as determined via nucleoside uptake by intestinal brush-border membrane vesicles (Ngo et al., 2001). Using double transfected MDCK cells, which simultaneously expressed YFP-tagged CNT1 and ENT1, Lai et al. (2002) confirmed the apical cellular localization of CNT1. While the rat homologue of CNT2 has been shown to be apically localized in polarized MDCK and LLC-PK cells, in liver parenchymal cells it is highly expressed in the basolateral (sinusoidal) membrane, suggesting tissue-specific sorting (Mangravite et al., 2001; Duflot et al., 2002). Duflot et al. (2002) also demonstrated the apical localization of CNT1 in liver parenchymal cells, suggesting transcytotic membrane insertion. The subcellular localization of CNT3 has not been reported as of preparation of this chapter.
Tissue distribution studies of hENT1, hENT2, and hENT3 suggest these transporters are ubiquitously expressed, at least on the mRNA level, although there does exist appreciable intertissue and interindividual expression differences (Griffiths and Jarvis, 1996; Crawford et al., 1998; Pennycooke et al., 2001; Jennings et al., 2001; Baldwin et al., 2005). On the cellular level, it is thought that ENTs and CNTs asymmetrically distribute between the apical and basolateral membranes to mediate the vectorial transepithelial flux of nucleosides (Lai et al., 2002; Kong et al., 2004). However, no studies directly examine ENT1, or ENT2 protein localization in intestinal epithelial tissues. Nevertheless, hENT1 has been shown to be predominantly basolaterally localized in YFP-tagged ENT1 transfected MDCK cells, with some apical localization (Lai et al., 2002; Mangravite et al., 2003), whereas hENT2 exhibits only basolateral localization (Mangravite et al., 2003). Interestingly, some functional data indicate that ENT1 may also exhibit some intracellular localization (Pisoni and Thoene, 1989; Mani et al., 1998; Jimenez et al., 2000), although further studies are required to determine the functional relevance of intracellular nucleoside transport due to ENT1 (Kong et al., 2004). HENT3 is now known to be an intracellular transporter that colocalized with lyso-somal markers, but not with golgi, early endosomes, or endoplasmic reticulum (Baldwin et al., 2005). As expected, truncation of the N-terminus or mutation of its dileucine motif caused relocation of the transporter to the cell surface (Baldwin et al., 2005). Given its lysosomal localization, it should be of no surprise that its transport kinetics were optimum at pH 5.5 (Baldwin et al., 2005).
Due to the inherent importance of nucleosides to overall cellular and tissue function, it is of no surprise that the regulation mechanisms of both CNTs and ENTs are quite complex and difficult to effectively study. In short, expression of both CNT1 and CNT2 has been shown to be tissue specific, dependent on cell cycle, certain hormones and cytokines, as well as the presence of substrate (Gomez-Angelats et al., 1996; Del Santo et al., 1998; Soler et al., 1998, 2001; Valdes et al., 2000, 2002). However, with the exception of nutritional regulation studies (Valdes et al., 2000), virtually all of the regulatory studies have been conducted in hepatocytes, or immune cells. Given tissue-specific regulation dependence, one should question the applicability of these data to the intestinal model.
In short, not much is known concerning either the transcriptional, or transla-tional regulation of either the SLC28, or SLC29 families. Regulation of hENT1 has been shown to be dependent on deoxynucleotide levels in cultured cancer cells (Pressacco et al., 1995). Moreover, hENT1 expression has also been demonstrated to be a function of cell cycle, with expression doubling between G1 and G2-M phases (Pressacco et al., 1995). Cultured cells also exhibited an upregula-tion of hENT1 due to phorbol ester treatment, which appears to be due to either PKC 6, or PKC e; however, this remains to be elucidated (Coe et al., 2002). Interestingly, glucose treatment has been shown to upregulate ENT1 protein activity, as well as its protein and mRNA expression in human aortic smooth muscle cells (Leung et al., 2005a). The functional significance of this finding is still uknown; however, it has been shown that ENT1 expression is modulated due to diabetes (Pawelczyk et al., 2003), and that it is not transcriptionally controlled via peroxisome proliferator-activated receptor gamma (PPARy) (Leung et al., 2005b). Still less is known concerning the mechanisms regulating the expression of hENT2 and hENT3.
Was this article helpful?
Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.