Fig. 2: Phosphate recognition and transport by XPR1.

a Channel-like architecture. Gray surface represents the channel path within XPR1 transmembrane ___domain. Transmembrane α-helices forming the channel are indicated, and channel-lining residues are shown as sticks. b Pore radius along the channel. The gray dashed line indicates a radius of 1 Å for water access limit. c Structures superposition. The transmembrane ___domain structures of Pi-unbound XPR1, Pi-bound XPR1^Pi1, Pi-bound XPR1^Pi2, and Pi-bound XPR1^Pi3 are colored in gray, cyan, pink, and green, respectively. Only the channel-forming transmembrane α-helices are represented for clarity. Pi ion in each structure is shown as sticks. d–f Pi recognition sties 1, 2, and 3, observed in the structures of XPR1^Pi1, XPR1^Pi2, XPR1^Pi3, respectively. The black dashed lines indicate plausible interactions between the Pi ion and the protein, while the red dashed lines represent the interactions that may indirectly contribute to Pi coordination. The Pi-recognizing residues at sites 1 and 2 largely overlap but adopt different binding models. g, h Activity assay. The Pi transport activity is determined by measuring phosphate uptake into liposomes containing wild-type and variant forms of XPR1. Activity of PFBC families variants, R459C, R570C and R570L, are highlighted in red. The time-course activity (g) was collected at external Pi concentration of 500 μM, with transport reaction times of 3, 6, 9, 12, 18, and 24 min. The data presented of R570C and R570L are the average of 3 independent assays, while the others are averaged from 4 independent assays, each with technical triplicates. The error bars indicates the SD. The single-point activity (h) was collected at the external Pi concentration of 500 μM, with a transport reaction time of 6 min. The data presented are the average of 3–4 independent assays, each with technical triplicates. The error bars indicates the SD.