Current research

Structure-function relationships in selected transporters

Very little is known on transporter structure-function relationships due to the difficulties in obtaining X-ray diffraction-quality crystals and in applying NMR methods for the study of polytopic membrane proteins. Furthermore, the few currently known structures come exclusively from bacteria and archea, and in the majority of cases are not associated with functional studies. An alternative approach to understand structure-function relationships of transporters and channels is through functional analysis of mutated or molecularly modified versions of these proteins in easily manipulated systems. The magnificent work of R. Kaback and his group with the LacY lactose permease of Escherichia coli constitutes a unique paradigm in biology. In this case, solely through the analysis of mutations, biochemical and biophysical approaches and the development of Cys-scanning mutagenesis, a model was proposed on how a transporter works, even before the eventual crystallization of LacY (Guan L, Kaback HR. 2006. Annu Rev Biophys Biomol Struct, 35:67-91).

The contribution of our lab in this direction concerns the use of classical and reverse genetics, direct biochemical transport assays, in vivo fluorescent microscopy and kinetic modeling to understand how several purine, pyrimidine or amino acid transporters work (for reviews see Diallinas, 2008; Pantazopoulou and Diallinas, 2008; Gournas et al, 2008; Diallinas & Gournas 2008).

By far our favorite molecular is the UapA uric acid-xanthine/H+ symporter. UapA is the prototype and founding member of an important and ubiquitous transporter family, called Nucleobase Ascorbate Transporters (NAT). At present, the function and specificity of nearly 20 NAT proteins from bacteria, fungi, plants and mammals is known. Seven of these NATs (six fungal and one plant) have been characterized and studied in our lab. All non-mammalian homologues of known function are specific for nucleobases, namely xanthine, uric acid or uracil. The mammalian NATs transport either L-ascorbic acid (SVCT1 and SVCT2) or nucleobases, namely uracil, xanthine or hypoxanthine (rSNBT17).

In the last 15 years, hundreds of UapA mutations, obtained by classical or reverse genetic approaches, as well as chimeric constructs, have been analyzed at the molecular, cellular and functional level, giving rise to unprecedented knowledge of the molecular elements underlying the function of this eukaryotic carrier. Based on our results, we have proposed models on how the UapA recognizes and translocates its substrates, and provided data on which amino residues are involved in substrate selection, binding and transport, and on which amino acid residues are key elements for protein stability and trafficking to the plasma membrane. In fact, it is through our studies that channel-like gating has been proposed to exist and be critical in determining UapA transporter specificity. This is an entirely novel concept that breaks the dogmatic distinction of transporters and channels. Supporting evidence for this idea has recently come form direct structural studies on other transporters.

At present, we try to understand in more detail how the binding and release of substrates leads to the opening and closing of the substrate translocation trajectory, how the gates synergize with the major substrate binding site, how our findings are related to the generally accepted model of outward- and inward-facing alternating transporter conformers (rocker-switch mechanism), and how ions drive solute symport. We also ask whether we can engineer new UapA versions transporting potential antifungal drugs and how domains external to the substrate biding site act as filters or gates affecting UapA substrate specificity.

Our current efforts also include sophisticated molecular dynamics aimed at understanding the molecular changes associated with the transitions between functional states, together with efforts to obtain the three-dimensional X-ray structure of the protein. For these approaches, we actively collaborate with the groups of Emmanuel Mikros (University of Athens, Pharmacy Department and Bernadette Byrne and Alex Cameron (Imperial College, London, UK). In particular, with the group at Imperial College, we are implementing a high-throughput platform for expression, screening and purification of wild-type or stable mutant forms of UapA and other NAT homologues. Up to date, substantial progress has been made towards the determination of UapA structure with a dataset processed to a maximum resolution of 3.5-3.8 Å so far.

Similar approaches of those used to study UapA, are currently followed to study other NAT members (AzgA, UapD) and selected members of an evolutionary and functionally district purine-pyrimidines transporter family (NCS1 transporters).

Membrane trafficking and endocytosis of transporters

Eukaryotic polytopic membrane proteins, such as transporters and channels, or receptors, are co-translationally inserted into the ER membrane through the action of the so-called translocase complex. This mechanistic step seems to be driven by the energy of polypeptide synthesis in the ribosome and several sequence-independent, amphipathicity-dependent, cis-acting elements on the protein cargoes. Once a membrane protein is properly folded within the ER membrane, an important check-point by itself, it then follows a vesicular or tubular trafficking pathway, initially towards the Golgi, and subsequently towards the endosomal pathway, the vacuole or the plasma membrane. This long sub-cellular journey (exocytosis) is dynamically controlled in response to multiple and overlapping developmental and physiological signals, rather than being a default process. Such signals not only promote or arrest exocytosis of a protein towards its target membrane, but can also promote its rapid endocytosis from the plasma membrane, which can lead to degradation in the vacuole or recycling back to the cell surface. The sum of complex processes underlying membrane protein exocytosis and endocytosis is called membrane protein trafficking. In this process, both cis-acting elements on the cargo proteins and trans-acting factors need to be orchestrated in a sequential and flexible manner to achieve proper trafficking. Moreover, it has been recently shown that the lipid composition of membranes, which by itself is dynamically controlled in response to various signals, plays a pivotal role in protein trafficking. The dynamic control of the trafficking of proteins such as transporters, channels and receptors, proteins, which play essential roles in the uptake or efflux of metabolites and drugs or the transmission of molecular signals, constitutes a rapid, flexible and very efficient regulatory mechanism critical for cell homeostasis and for the communication of cells with their environment.

In the yeast S. cerevisiae several plasma membrane transporters have been used extensively in order to understand membrane cargo trafficking, sorting and endocytosis. In most cases, transporter trafficking is controlled in response to substrate availability, general nutrient supply conditions and/or stresses. Strikingly, the mechanisms controlling transporter trafficking are essentially conserved from fungi to mammals. In fact, studies on the intracellular trafficking of yeast permeases have contributed to revealing the central role played by the small protein ubiquitin (Ub), a sorting signal of eukaryotic membrane proteins. In yeasts and A. nidulans the covalent attachment of Ub on cargoes depends on Rsp5/HulA, a Ub ligase of the Nedd4 HECT family. Recent studies suggested a general model in which different Rsp5/HulA adaptor proteins recognize different transporters, or the same transporter in response to different stimuli. Finally, lipid rafts, formed by the lateral association of sphingolipids and cholesterol (mammals) or ergosterol (fungi) in the external membrane leaflet have been implicated in transporter traffic and cell signaling in mammalian cells and yeast.

Several genetic and molecular tools for specific sub-cellular organelles/compartments have also been developed for A. nidulans (mostly in the lab of M.A. Penalva) and several transporters of purines, pyrimidines and amino acids, belonging to evolutionary discrete families, have been used as protein cargoes to study endocytosis in response to a shift in nitrogen source or excess substrate (our lab). As a consequence, important aspects of trafficking mechanisms have been revealed in this model fungus. The primary contribution of our lab in this direction is the identification of two distinct mechanisms controlling transporter down-regulation by endocytic internalization. The first occurs in response to a shift from poor to rich nitrogen media (ammonium ions) and the second in response to substrate excess (Valdez-Taubas et al. 2000; Pantazopoulou et al. 2007; Gournas et al. 2010). In the case of the UapA transporter, both mechanisms are dependent on HulA-dependent ubiquitination of a single Lys residue (K572) at the C-tail of the transporter. However, substrate-induced endocytosis, unlike ammonium-induced internalization, takes place only for active molecules. The use of specific functional mutations of UapA has shown that protein movements associated with the transport process constitute the upstream signal for substrate-induced endocytosis. Interestingly, active transporter molecules can elicit the endocytosis of inactive molecules in trans, suggesting that transporters oligomerize or are organized in membrane microdomains.


A, UapA-GFP endocytosis elicited by NH4 and substrates (UA) showing the requirement of HulA & K572R. B-C, UapA-GFP turnover and ubiquitination by NH4 or UA






Some of our current specific questions are the following: