On the Evolution of Specificity in Members of the Yeast Amino Acid Transporter Family as Parts of Specific Metabolic Pathways

Abstract

In the recent years, molecular modeling and substrate docking, coupled with biochemical and genetic analyses have identified the substrate-binding residues of several amino acid transporters of the yeast amino acid transporter (YAT) family. These consist of (a) residues conserved across YATs that interact with the invariable part of amino acid substrates and (b) variable residues that interact with the side chain of the amino acid substrate and thus define specificity. Secondary structure sequence alignments showed that the positions of these residues are conserved across YATs and could thus be used to predict the specificity of YATs. Here, we discuss the potential of combining molecular modeling and structural alignments with intra-species phylogenetic comparisons of transporters, in order to predict the function of uncharacterized members of the family. We additionally define some orphan branches which include transporters with potentially novel, and to be characterized specificities. In addition, we discuss the particular case of the highly specific l-proline transporter, PrnB, of Aspergillus nidulans, whose gene is part of a cluster of genes required for the utilization of proline as a carbon and/or nitrogen source. This clustering correlates with transcriptional regulation of these genes, potentially leading to the efficient coordination of the uptake of externally provided l-Pro via PrnB and its enzymatic degradation in the cell.

Keywords: APC; Area; CreA; YAT; amino acids; l-proline; prn cluster; transcriptional regulation.

Figure 1

The important residues for substrate binding and determination of the specificity in the YAT family. A multiple alignment of the TMs1, 3, 6, 8 and 10 sequences from selected YAT members—for most of which structure-function studies have been performed. Residues in red font are highly conserved, while residues in red filled columns are absolutely conserved in the sequences of the selected YATs. Blue frames define residues that are conserved at least 70% of the sequences. Red triangles indicate the positions of residues that are highly conserved in YATs and are predicted to interact with the backbone of the amino acid substrates. Black triangles indicate specificity-determining residues that are poorly conserved in YATs and predicted to interact with the side chain of the amino acid substrates. Empty triangles indicate residues that indirectly affect the specificity of YATs. Color-coding next to the names of YATs shows which clade of the phylogenetic tree of Figure 2 the transporters belong to. The annotation of the alignment was done with ESPript 3.0 [45].

Figure 2

A phylogenetic tree of YATs. All YATs from A. nidulans and S. cerevisiae are included, together with other functionally characterized fungal transporters. The S. cerevisiae proteins TAT1, TAT2, TAT3, GAP1, HIP1, GNP1, AGP1, AGP2, AGP3, BAP2, BAP3, SAM3, MMP1, LYP1, ALP1, CAN1, DIP5, PUT4 and SSY1 were obtained from the Saccharomyces Genome Database (http://www.yeastgenome.org). The sequences of the YATs of A. nidulans were collected and manually corrected for errors occurring by predictions of exon splicing [18]. The previously detected pseudogenes [18], ANID_4604, ANID_8556, ANID_8659 and ANID_5024, are excluded from the tree. Other proteins included are two characterized broad specificity amino acid permeases from Neurospora crassa (NAAP1) [47] and Trichoderma harzianum (INDA1) [48], one Aromatic and one Aromatic and aliphatic amino acid transporter from Penicillium crysogenum (MTR, ArlP) [49,50], a general amino acid permease from Hebeloma cylindrosporum (HcGap1) [51] and a cystine-specific transporter from Candida glabrata (CgCyn1) [52]. Alignment was performed with MAFFT G-INS-I with default parameters [53], alignment curation with BMGE with default parameters [54]. A maximum likelihood tree was obtained with PhyML (http://www.atgc-montpellier.fr/phyml/) with automatic selection substitution model (LG) [55]. The tree was drawn with FigTree (FigTree v1.4.2. http://tree.bio.ed.ac.uk/software/figtree/). The tree shown is in a simplified cartoon form in which several branches are collapsed into triangles, only when related to a functionally characterized homologue. Color codes. Red: basic amino acid transporters; green: dicarboxylic amino acid transporters; yellow: putative cystine transporters; dark grey: putative branched amino acid transporters; dark purple: aromatic amino acid transporters; Dark blue: proline transporters; orange: transporters related to a functionally diverse group of structurally related S. cerevisiæ transporters. Within this latter group, there is a subclade of characterized general amino acid transporters from different species. A more comprehensive tree comprising the YATs of all available Aspergilli can be found at Additional file 23 of [18].

Figure 3

Schematic representation of the prn gene cluster organization in chromosome VII of A. nidulans. prnB (AN1732) encodes the specific proline transporter, prnD (AN1731) codes for proline oxidase, prnC (AN1733) for Δ1-pyrroline-5-carboxylate dehydrogenase, prnA (AN1729) for the pathway specific transcriptional activator and prnX (AN1730) is a proline-inducible gene of unknown function (see text). Genes are presented in scale with their standard names followed by their systematic names in the AspGD bank (http://www.aspgd.org/). Introns are shown red and the direction of transcription is indicated by the arrow-like gene boxes. The inset represents a zoom in the boxed area, the prnB–prnD intergenic region, where the relative position of the two physiologically relevant PrnA- (green oval), CreA-, corresponding to prn^d20 and prn^d22 mutations, (purple rectangles) and AreA- binding sites (orange rectangles) are shown.