Specificity profile of NAT/NCS2 purine transporters in Sinorhizobium (Ensifer) meliloti.

Sinorhizobium (Ensifer) meliloti is a model example of a soil alpha-proteobacterium which induces the formation of nitrogen-fixing symbiotic nodules on the legume roots. In contrast to all other rhizobacterial species, S. meliloti contains multiple homologs of nucleobase transporter genes that belong to NAT/NCS2 family (Nucleobase-Ascorbate Transporter/Nucleobase-Cation Symporter-2). We analyzed functionally all (six) relevant homologs of S. meliloti 1,021 using Escherichia coli K-12 as a host and found that five of them are high-affinity transporters for xanthine (SmLL9), uric acid (SmLL8, SmLL9, SmX28), adenine (SmVC3, SmYE1), guanine (SmVC3), or hypoxanthine (SmVC3). Detailed analysis of substrate profiles showed that two of these transporters display enlarged specificity (SmLL9, SmVC3). SmLL9 is closely related in sequence with the xanthine-specific XanQ of E. coli. We subjected SmLL9 to rationally designed site-directed mutagenesis and found that the role of key binding-site residues of XanQ is conserved in SmLL9, whereas a single amino-acid change (S93N) converts the xanthine/uric-acid transporter SmLL9 to a xanthine-preferring variant, due to disruption of an essential hydrogen bond with the C8 oxygen of uric acid. The results highlight the presence of several different purine nucleobase transporters in S. meliloti and imply that the purine transport might be important in the nodule symbiosis involving S. meliloti.

Insight on specificity of uracil permeases of the NAT/NCS2 family from analysis of the transporter encoded in the pyrimidine utilization operon of Escherichia coli.

The uracil permease UraA of Escherichia coli is a structurally known prototype for the ubiquitous Nucleobase-Ascorbate Transporter (NAT) or Nucleobase-Cation Symporter-2 (NCS2) family and represents a well-defined subgroup of bacterial homologs that remain functionally unstudied. Here, we analyze four of these homologs, including RutG of E. coli which shares 35% identity with UraA and is encoded in the catabolic rut (pyrimidine utilization) operon. Using amplified expression in E. coli K-12, we show that RutG is a high-affinity permease for uracil, thymine and, at low efficiency, xanthine and recognizes also 5-fluorouracil and oxypurinol. In contrast, UraA and the homologs from Acinetobacter calcoaceticus and Aeromonas veronii are permeases specific for uracil and 5-fluorouracil. Molecular docking indicates that thymine is hindered from binding to UraA by a highly conserved Phe residue which is absent in RutG. Site-directed replacement of this Phe with Ala in the three uracil-specific homologs allows high-affinity recognition and/or transport of thymine, emulating the RutG profile. Furthermore, all RutG orthologs from enterobacteria retain an Ala at this position, implying that they can use both uracil and thymine and, possibly, xanthine as substrates and provide the bacterial cell with a range of catabolizable nucleobases.

Identification of New Specificity Determinants in Bacterial Purine Nucleobase Transporters based on an Ancestral Sequence Reconstruction Approach.

The relation of sequence with specificity in membrane transporters is challenging to explore. Most relevant studies until now rely on comparisons of present-day homologs. In this work, we study a set of closely related transporters by employing an evolutionary, ancestral-reconstruction approach and reveal unexpected new specificity determinants. We analyze a monophyletic group represented by the xanthine-specific XanQ of Escherichia coli in the Nucleobase-Ascorbate Transporter/Nucleobase-Cation Symporter-2 (NAT/NCS2) family. We reconstructed AncXanQ, the putative common ancestor of this clade, expressed it in E. coli K-12, and found that, in contrast to XanQ, it encodes a high-affinity permease for both xanthine and guanine, which also recognizes adenine, hypoxanthine, and a range of analogs. AncXanQ conserves all binding-site residues of XanQ and differs substantially in only five intramembrane residues outside the binding site. We subjected both homologs to rationally designed mutagenesis and present evidence that these five residues are linked with the specificity change. In particular, we reveal Ser377 of XanQ (Gly in AncXanQ) as a major determinant. Replacement of this Ser with Gly enlarges the specificity of XanQ towards an AncXanQ-phenotype. The ortholog from Neisseria meningitidis retaining Gly at this position is also a xanthine/guanine transporter with extended substrate profile like AncXanQ. Molecular Dynamics shows that the S377G replacement tilts transmembrane helix 12 resulting in rearrangement of Phe376 relative to Phe94 in the XanQ binding pocket. This effect may rationalize the enlarged specificity. On the other hand, the specificity effect of S377G can be masked by G27S or other mutations through epistatic interactions.