Interactions of cytosolic termini of the Jen1 monocarboxylate transporter are critical for trafficking, transport activity and endocytosis

Plasma membrane (PM) transporters of the major facilitator superfamily (MFS) are essential for cell metabolism and growth, as well as for survival in response to stress or cytotoxic drugs, in both prokaryotes and eukaryotes. In the yeast Saccharomyces cerevisiae, Jen1 is a monocarboxylate/H+ symporter that has been used to dissect the molecular details underlying control of cellular expression, transport mechanism and turnover of MFS transporters. Here, we present evidence supporting previously non-described roles of the cytosolic N- and C- termini in Jen1 biogenesis, PM stability and activity, through functional analyses of rationally designed truncations and chimeric constructs with UapA, a S. cerevisiae endocytosis-insensitive purine transporter from Aspergillus nidulans. Our results reveal a cryptic role of the N-terminal region and thus show that both cytosolic N- and C-termini are critical for Jen1 trafficking to the PM, transport activity and endocytosis. In particular, we provide evidence that the N- and the C-cytosolic termini of Jen1 undergo transport-dependent dynamic intra-molecular interactions, which critically affect the mechanism of transport and turnover of Jen1. Our results support an emerging concept where the cytosolic tails of PM transporters control transporter expression and function, through flexible intra-molecular interactions with each other and the transmembrane core of the protein. This idea may be extended to other MFS members providing a deeper understanding of conserved, but also evolving, mechanisms underlying MFS transporter structure-function relationships.


Introduction
terminus is likely to undergo dynamic conformational changes, in response to excess 97 of substrate or stress, enhancing its endocytic down-regulation (31). 98 In the filamentous fungus A. nidulans, a C-terminus region of the uric acid 99 transporter UapA is essential for ArtA-mediated ubiquitylation, endocytosis and 100 vacuolar degradation in response to ammonium or excess of substrate (11,14). A. 101 nidulans Fur4 homologues have also been shown to possess elements in their N-102 and C-terminus that are critical for endocytosis and surprisingly substrate specificity. 103 In this case, the authors provided evidence that the N-and the C-terminus interact 104 physically and promote proper transporter function and turnover (32, 33).

105
Whether long-range regulatory effects of cytosolic N-and C-termini extend to 106 transporters other than Fur-like proteins and a handful of other members of the 107 amino acid-polyamine-organocation (APC) superfamily (34), remains to be formally 108 shown. 109 Here, we address this issue by using Jen1, a well-studied yeast transporter  Prior to these constructions, it was essential to define the limits of the N-and  Table S1). These predictions were used to construct three Jen1 152 truncated versions by deleting the longest predicted N-terminal region (133 residues) 153 and the two versions of the putative C-terminus (62 or 33 residues) of Jen1. The   Figure 2C, Figure S1B). This suggested that truncating the 33 last residues of Jen1 234 is not just epistatic to the instability conferred by deleting the N-terminal 94 residues, 235 but also pointed to the idea that the two termini of Jen1 interact functionally.      UapA/Jen1CT33 to glucose-elicited endocytosis, we measured the steady state 429 protein levels of these proteins by western blotting. As shown in Figure 5E, solely 430 UapA/Jen1CT33 protein levels were significantly reduced in the presence of glucose.

431
This suggests that the C-terminus of Jen1 is sufficient to promote glucose-elicited 432 turnover of UapA, in a context-independent manner. 433 We, subsequently, analysed the localization and the protein steady state highlighted when comparing western blots in Figure 5E and 5F, showed that  (Figure 5C, 5D, S5B and S5C). Based solely on the western blot analysis ( Figure   456 5E, 5F and 5G), we may conclude that none of the two chimeras responds to    termini conformation during a substrate transport cycle (i.e., outward-facing, occluded, 527 inward-facing). In the absence of substrate, Jen1 is in an outward-facing conformation with 528 the N-and C-termini in close contact or interacting with each other. Upon substrate addition, 529 Jen1 moves to an inward conformation with consequent loss of termini interaction/proximity. 530 The topological changes in Jen1 termini seem to be crucial for Jen1 endocytic turnover, 531 biogenesis/folding, transport activity and trafficking. The aforementioned scheme represents 532  When considering the fact that Jen1ΔCT33 is stable and fully functional while 589 Jen1ΔCT62 is non-functional due to retention in the ER, we can also conclude that 590 the middle part of the C-terminal segment, between amino acids residues 554-583 591 (the areas missing in Jen1ΔCT62 but present in Jen1ΔCT33, Figure S2), might 592 contain elements critical for ER-exit or proper folding. Such ER-exit motifs have been 593 identified at the cytosolic termini of other transporters but, to our knowledge, none 594 has been rigorously shown to act in a context-independent manner (see the recent 595 review (34). A preliminary in silico analysis of the sequence of Jen1 between amino 596 acids residues 554-583 showed that a short di-acidic motif, 577 EYE 579 might be an 597 interesting candidate as an ER-exit motif (see Figure S2B). 598 Role of the N-tail. Jen1ΔΝT94 was shown to be normally produced at basal 599 levels, but proved to be a rather unstable version of Jen1, exhibiting rapid turnover (e.g., 2.5-fold increased substrate affinity), which points to a positive 'distant' effect 604 on the transport mechanism, albeit weaker than that of the C-terminal segment.

605
Notably also, the N-terminal part proved critical for endocytic down-regulation in 606 response to prolonged growth on lactate or glucose, because when ROD1 gene was 607 knocked-out or the C-tail of Jen1 was deleted (i.e., no Rod1 binding), the presence 608 of the N-terminal conferred partial endocytosis, while its absence led to an increased 609 stability. Our data further support the conclusion that glucose triggered endocytosis  When the NT94 segment of Jen1 was transferred to UapA, it led to a chimera 614 that showed significant ER-retention, despite being transport-competent. As a result, 615 the functional analysis of this chimera did not provide us with additional evidence on 616 the role the Jen1 N-tail. Finally, comparing the effect of deleting the entire N-terminal 617 cytosolic region of Jen1 (residues 1-133), which led to ER-retention, to Jen1ΔΝT94, 618 which led to turnover after translocation to the PM, we conclude that the segment 619 between 94-133 might also include motifs driving ER-exit or necessary for proper 620 folding. Interestingly, this Jen1 segment contains a well-conserved motif, 621 126 NPIPE 133 , that is worthy to be studied by mutational analysis, in the future (see 622 Figure S2B).

668
The present work on Jen1 also shows that rather cryptic roles of transporter 669 cytosolic tails can be exploited to rationally modify transporter function, which will be   Table S1.  Table S4) using 714 yeast genomic DNA (unless it is clearly specified). The resulting PCR products were 715 co-transformed with a linearized plasmid (digested with a specific restriction enzyme) 716 in S. cerevisiae cells. All plasmids used and constructed are listed in Table S5. 717 Specifically, for construction of JEN1 termini truncated versions pGPDJEN1ΔNT133,   Table S3. 829 To study the possible interaction of the Jen1 N-with C-terminus, jen1∆ cells