Publications

MdfA is an interesting member of a large group of secondary multidrug (Mdr) transporters. Through genetic, biochemical and biophysical studies of MdfA, many challenging aspects of the multidrug transport phenomenon have been addressed. This includes its ability to interact with chemically unrelated drugs and how it utilizes energy to drive efflux of compounds that are not only structurally, but also electrically, different. Admittedly, however, despite all efforts and a recent pioneering structural contribution, several important mechanistic issues of the promiscuous capabilities of MdfA still seek better molecular and dynamic understanding.

Are integral membrane protein-encoding mRNAs (MPRs) different from other mRNAs such as those encoding cytosolic mRNAs (CPRs)? This is implied from the emerging concept that MPRs are specifically recognized and delivered to membrane-bound ribosomes in a translation-independent manner. MPRs might be recognized through uracil-rich segments that encode hydrophobic transmembrane helices. To investigate this hypothesis, we designed DNA sequences encoding model untranslatable transcripts that mimic MPRs or CPRs. By utilizing in vitro-synthesized biotinylated RNAs mixed with Escherichia coli extracts, we identified a highly specific interaction that takes place between transcripts that mimic MPRs and the cold shock proteins CspE and CspC, which are normally expressed under physiological conditions. Co-purification studies with E. coli expressing 6His-tagged CspE or CspC confirmed that the specific interaction occurs in vivo not only with the model uracil-rich untranslatable transcripts but also with endogenous MPRs. Our results suggest that the evolutionarily conserved cold shock proteins may have a role, possibly as promiscuous chaperons, in the biogenesis of MPRs.

Multidrug transporters are membrane proteins that catalyze efflux of antibiotics and other toxic compounds from cells, thereby conferring drug resistance on various organisms. Unlike most solute transporters that transport a single type of compound or similar analogues, multidrug transporters are extremely promiscuous. They transport a broad spectrum of dissimilar drugs and represent a serious obstacle to antimicrobial or anticancer chemotherapy. Many challenging aspects of multidrug transporters, which are unique, have been studied in detail, including their ability to interact with chemically unrelated drugs, and how they utilize energy to drive efflux of compounds that are not only structurally but electrically different. A new and surprising dimension of the promiscuous nature of multidrug transporters has been described recently: they can move long molecules through the membrane in a processive manner.

Multidrug transporters are ubiquitous efflux pumps that provide cells with defense against various toxic compounds. In bacteria, which typically harbor numerous multidrug transporter genes, the majority function as secondary multidrug/proton antiporters. Proton-coupled secondary transport is a fundamental process that is not fully understood, largely owing to the obscure nature of proton-transporter interactions. Here we analyzed the substrate/proton coupling mechanism in MdfA, a model multidrug/proton antiporter. By measuring the effect of protons on substrate binding and by directly measuring proton binding and release, we show that substrates and protons compete for binding to MdfA. Our studies strongly suggest that competition is an integral feature of secondary multidrug transport. We identified the proton-binding acidic residue and show that, surprisingly, the substrate binds at a different site. Together, the results suggest an interesting mode of indirect competition as a mechanism of multidrug/proton antiport.

The mechanism underlying the interaction of the Escherichia coli signal recognition particle receptor FtsY with the cytoplasmic membrane has been studied in detail. Recently, we proposed that FtsY requires functional interaction with inner membrane lipids at a late stage of the signal recognition particle pathway. In addition, an essential lipid-binding α-helix was identified in FtsY of various origins. Theoretical considerations and in vitro studies have suggested that it interacts with acidic lipids, but this notion is not yet fully supported by in vivo experimental evidence. Here, we present an unbiased genetic clue, obtained by serendipity, supporting the involvement of acidic lipids. Utilizing a dominant negative mutant of FtsY (termed NG), which is defective in its functional interaction with lipids, we screened for E. coli genes that suppress the negative dominant phenotype. In addition to several unrelated phenotype-suppressor genes, we identified pgsA, which encodes the enzyme phosphatidylglycerophosphate synthase (PgsA). PgsA is an integral membrane protein that catalyzes the committed step to acidic phospholipid synthesis, and we show that its overexpression increases the contents of cardiolipin and phosphatidylglycerol. Remarkably, expression of PgsA also stabilizes NG and restores its biological function. Collectively, our results strongly support the notion that FtsY functionally interacts with acidic lipids.

All intramembrane proteases are known to cleave membrane proteins with a single transmembrane helix. Such cleavages often release anchored soluble domains, which play a role in physiologically important inter- and intracellular processes. However, in many cases the physiological roles/substrates of intramembrane proteases are not known. It is interesting that no multispanning substrates were identified so far, despite the fact that intramembrane proteases have promiscuous substrate recognition and cleavage capabilities. Here we determined whether, in a synthetic experimental system, intramembrane proteases have the capability to interact with and cleave multispanning membrane proteins. We utilized the Escherichia coli rhomboid GlpG, an intramembrane serine protease, and truncated versions of the E. coli multidrug transporter MdfA as model multispanning membrane proteins. On the basis of in vivo and in vitro studies on the association of GlpG with various MdfA constructs and their cleavage, we conclude that GlpG is able to recognize and cleave truncated forms of MdfA but not the intact protein. We propose that GlpG has the capacity to act on unfolded multispanning membrane proteins, thus providing an incentive for investigating possible physiological consequences.

Escherichia coli membrane protein biogenesis is mediated by a signal recognition particle and its membrane-associated receptor (FtsY). Although crucial for its function, it is still not clear how FtsY interacts with the membrane. Analysis of the structure/function differences between severely truncated active (NG+1) and inactive (NG) mutants of FtsY enabled us to identify an essential membrane-interacting determinant. Comparison of the three-dimensional structures of the mutants, combined with site-directed mutagenesis, modeling, and liposome-binding assays, revealed that FtsY contains a conserved autonomous lipid-binding amphipathic α-helix at the N-terminal end of theNdomain. Deletion experiments showed that this helix is essential for FtsY function in vivo, thus offering, for the first time, clear evidence for the functionally important, physiologically relevant interaction of FtsY with lipids.

MdfA is a prototypic secondary multidrug transporter from Escherichia coli, which recognizes and exports a broad spectrum of structurally and electrically dissimilar toxic compounds. Here we review recent studies of MdfA, which, on the one hand, provide advanced understanding of certain aspects of secondary multidrug transport, and, on the other, address major mechanistic questions, some of which remain to be elucidated. Using biochemical, genetic, and physiological approaches, we have revealed several surprisingly promiscuous properties of MdfA including its multidrug recognition capacity, proton recognition determinants, aspects of energy utilization, and physiological role.

MdfA is an Escherichia coli multidrug transporter of the major facilitator superfamily (MFS) of secondary transporters. Although several aspects of multidrug recognition by MdfA have been characterized, better understanding the detailed mechanism of its function requires structural information. Previous studies have modeled the 3D structures of MFS proteins, based on the X-ray structure of LacY and GlpT. However, because of poor sequence homology, between LacY, GlpT, and MdfA additional constraints were required for a reliable homology modeling. Using an algorithm that predicts the angular orientation of each transmembrane helix (TM) (kPROT), we obtained a remarkably similar pattern for the 12 TMs of MdfA and those of GlpT and LacY, suggesting that they all have similar helix packing. Consequently, a 3D model was constructed for MdfA by structural alignment with LacY and GlpT, using the kPROT results as an additional constraint. Further refinement and a preliminary evaluation of the model were achieved by correlated mutation analysis and the available experimental data. Surprisingly, in addition to the previously characterized membrane-embedded glutamate at position 26, the model suggests that Asp34 and Arg112 are located within the membrane, on the same face of the cavity as Glu26. Importantly, Arg112 is evolutionarily conserved in secondary drug transporters, and here we show that a positive charge at this position is absolutely essential for multidrug transport by MdfA.

MdfA is an Escherichia coli multidrug-resistance transporter. Cells expressing MdfA from a multicopy plasmid exhibit multidrug resistance against a diverse group of toxic compounds. In this article, we show that, in addition to its role in multidrug resistance, MdfA confers extreme alkaline pH resistance and allows the growth of transformed cells under conditions that are close to those used normally by alkaliphiles (up to pH 10) by maintaining a physiological internal pH. MdfA-deleted E. coli cells are sensitive even to mild alkaline conditions, and the wild-type phenotype is restored fully by MdfA expressed from a plasmid. This activity of MdfA requires Na⁺ or K⁺. Fluorescence studies with inverted membrane vesicles demonstrate that MdfA catalyzes Na⁺- or K⁺-dependent proton transport, and experiments with reconstituted proteoliposomes confirm that MdfA is solely responsible for this phenomenon. Studies with multidrug resistance-defective MdfA mutants and competitive transport assays suggest that these activities of MdfA are related. Together, the results demonstrate that a single protein has an unprecedented capacity to turn E. coli from an obligatory neutrophile into an alkalitolerant bacterium, and they suggest a previously uncharacterized physiological role for MdfA in pH homeostasis.

Na⁺,K⁺-ATPase (pig α1,β1) has been expressed in the methylotrophic yeast Pichia pastoris. A protease-deficient strain was used, recombinant clones were screened for multicopy genomic integrants, and protein expression, and time and temperature of methanol induction were optimized. A 3-liter culture provides 300-500 mg of membrane protein with ouabain binding capacity of 30-50 pmol mg^-1. Turnover numbers of recombinant and renal Na⁺,K⁺-ATPase are similar, as are specific chymotryptic cleavages. Wild type (WT) and a D369N mutant have been analyzed by Fe²⁺- and ATP-Fe²⁺-catalyzed oxidative cleavage, described for renal Na⁺,K⁺-ATPase. Cleavage of the D369N mutant provides strong evidence for two Fe²⁺ sites: site 1 composed of residues in P and A cytoplasmic domains, and site 2 near trans-membrane segments M3/M1. The D369N mutation suppresses cleavages at site 1, which appears to be a normal Mg²⁺ site in E₂ conformations. The results suggest a possible role of the charge of Asp ³⁶⁹ on the E₁ ↔ E₂ conformational equilibrium. 5-Adenylyl-β,γ-imidodiphosphate(AMP-PNP)-Fe ²⁺-catalyzed cleavage of the D369N mutant produces fragments in P (⁷¹²VNDS) and N (near ⁴⁴⁰VAGDA) domains, described for WT, but only at high AMP-PNP-Fe²⁺ concentrations, and a new fragment in the P domain (near ³⁶⁷CSDKTGT) resulting from cleavage. Thus, the mutation distorts the active site. A molecular dynamic simulation of ATP-Mg ²⁺ binding to WT and D351N structures of Ca²⁺-ATPase (analogous to Asp³⁶⁹ of Na⁺,K⁺-ATPase) supplies possible explanations for the new cleavage and for a high ATP affinity, which was observed previously for the mutant. The Asn³⁵¹ structure with bound ATP-Mg²⁺ may resemble the transition state of the WT poised for phosphorylation.

In Escherichia coli, ribosomes must interact with translocons on the membrane for the proper integration of newly synthesized membrane proteins, cotranslationally. Previous in vivo studies indicated that unlike the E. coli signal recognition particle (SRP), the SRP receptor FtsY is required for membrane targeting of ribosomes. Accordingly, a putative SRP-independent, FtsY-mediated ribosomal targeting pathway has been suggested (Herskovits, A.A., E.S. Bochkareva, and E. Bibi. 2000. Mol. Microbiol. 38:927-939). However, the nature of the early contact of ribosomes with the membrane, and the involvement of FtsY in this interaction are unknown. Here we show that in cells depleted of the SRP protein, Ffh or the translocon component SecE, the ribosomal targeting pathway is blocked downstream and unprecedented, membrane-bound FtsY-ribosomal complexes are captured. Concurrently, under these conditions, novel, ribosome-loaded intracellular membrane structures are formed. We propose that in the absence of a functional SRP or translocon, ribosomes remain jammed at their primary membrane docking site, whereas FtsY-dependent ribosomal targeting to the membrane continues. The accumulation of FtsY-ribosome complexes induces the formation of intracellular membranes needed for their quantitative accommodation. Our results with E. coli, in conjunction with recent observations made with the yeast Saccharomyces cerevisiae, raise the possibility that the SRP receptor-mediated formation of intracellular membrane networks is governed by evolutionarily conserved principles.

The mechanism by which multidrug transporters interact with structurally unrelated substrates remains enigmatic. Based on transport competition experiments, photoaffinity labeling, and effects on enzymatic activities, it was proposed in the past that multidrug transporters can interact simultaneously with a number of dissimilar substrate molecules. To study this phenomenon, we applied a direct binding approach and transport assays using the Escherichia coli multidrug transporter MdfA, which exports both positively charged (e.g., tetraphenylphosphonium, TPP⁺), zwitterionic (e.g., ciprofloxacin), and neutral (e.g., chloramphenicol) drugs. The interaction of MdfA with various substrates was examined by direct binding assays with the purified transporter. The immobilized MdfA binds TPP⁺ in a specific manner, and all the tested positively charged substrates inhibit TPP⁺ binding. Surprisingly, although TPP⁺ binding is not affected by zwitterionic substrates, the neutral substrate chloramphenicol stimulates TPP⁺ binding by enhancing its affinity to MdfA. In contrast, transport competition assays show inhibition of TPP⁺ transport by chloramphenicol. We suggest that MdfA binds TPP⁺ and chloramphenicol simultaneously to distinct but interacting binding sites, and the interaction between these two substrates during transport is discussed.

In vivo and in vitro studies have suggested that the bacterial version of the mammalian signal recognition particle (SRP) system plays an essential and selective role in protein biogenesis. The bacterial SRP system consists of at least two proteins and an RNA molecule (termed Ffh, FtsY and 4.5S RNA, respectively, in Escherichia coli). Recent evidence suggests that other putative bacterial-specific SRP components may also exist. In vitro experiments confirmed the expected basic features of the bacterial SRP system by demonstrating interactions among the SRP components themselves, between them and ribosomes, ribosome-linked hydrophobic nascent polypeptides or inner membranes. The availability of a conserved (and essential) bacterial SRP version has facilitated the implementation of powerful genetic and biochemical approaches for studying the cascade of events during the SRP-mediated targeting process in vivo and in vitro as well as the three-dimensional structures and the properties of each SRP component and complex.

A small open reading frame from the Escherichia coli chromosome, bcrC(EC), encodes a homologue to the BcrC subunit of the bacitracin permease from Bacillus licheniformis. We show that disruption of the chromosomal bcrC(EC) gene causes bacitracin sensitivity and, conversely, that BcrC(EC) confers bacitracin resistance when expressed from a multicopy plasmid.

Newly synthesized polytopic membrane proteins and secretory proteins often share the same target membrane as their primary destination, and in some cases, the cellular machinery that targets and transfers them into or across the membrane. Unlike secretory proteins, which are localized to the external compartment, each polytopic membrane protein molecule must be partitioned among the cytoplasm, the membrane and the external milieu. How does the ribosome-translocon complex cope with the different domains of polytopic membrane proteins?.

Recent studies have revealed that Escherichia coli possesses an essential targeting system for integral membrane proteins, similar to the mammalian signal recognition particle (SRP) machinery. One essential protein in this system is FtsY, a homologue of the α-subunit of the mammalian SRP- receptor (SR-α). However, E. coli does not possess a close homologue of the integral membrane protein SR-β, which anchors SR-α to the membrane. Moreover, although FtsY can be found as a peripheral membrane protein, the majority is found soluble in the cytoplasm. In this study, we obtained genetic and biochemical evidence that FtsY must be targeted to the membrane for proper function. We demonstrate that the essential membrane targeting activity of FtsY is mediated by a 198-residue-long acidic N-terminal domain. This domain can be functionally replaced by unrelated integral membrane polypeptides, thus avoiding the need for specific FtsY membrane targeting factors. Therefore, the N terminus of FtsY constitutes an independent domain, which is required only for the targeting of the C-terminal NG domain of FtsY to the membrane.

The assembly of functional proteins from fragments in vivo has been recently described for several proteins, including the secreted maltose binding protein in Escherichia coli. Here we demonstrate for the first time that split gene products can function within the eukaryotic secretory system. Saccharomyces cerevisiae strains able to use sucrose produce the enzyme invertase, which is targeted by a signal peptide to the central secretory pathway and the periplasmic space. Using this enzyme as a model we find the following: (i) Polypeptide fragments of invertase, each containing a signal peptide, are independently translocated into the endoplasmic reticulum (ER) are modified by glycosylation, and travel the entire secretory pathway reaching the yeast periplasm. (ii) Simultaneous expression of independently translated and translocated overlapping fragments of invertase leads to the formation of an enzymatically active complex, whereas individually expressed fragments exhibit no activity. (iii) An active invertase complex is assembled in the ER, is targeted to the yeast periplasm, and is biologically functional, as judged by its ability to facilitate growth on sucrose as a single carbon source. These observations are discussed in relation to protein ridding and assembly in the ER and to the trafficking of proteins through the secretory pathway.

The use of lactose permease-alkaline phosphatase fusions (lacY-phoA) demonstrates that the lactose permease of Escherichia coli contains 12 transmembrane domains and that approximately half of a transmembrane domain is required to translocate alkaline phosphatase to the periplasmic surface of the membrane [Calamia, J., & Manoil, C. (1990) Proc. Natl. Acad. Sci. U.S.A. 87, 4937-4941], We have now used fusion analysis in combination with site-directed spectroscopy to examine more precisely the topology of putative helices VII and XI which contain the interacting residues Asp237 and Lys358, respectively. For this purpose, alkaline phosphatase was fused to alternate amino acid residues in transmembrane domains VII and XI. A sharp increase in alkaline phosphatase activity is observed as the fusion junction proceeds from Tyr228 to Ile230 in helix VII and from Phe354 to Phe356 in helix XI, suggesting that these residues approximate the middle of the corresponding transmembrane helices. Analysis of fluorescence quenching of the pyrene-labeled single-Cys mutants Asp237→Cys or Lys358→Cys, as well as measurement of collision frequencies between freely diffusing paramagnetic probes and a nitroxide spin-label at these sites, also indicates that Asp237 and also Asp240, which interacts with Lys319 (helix X), are located in transmembrane domains. However, Asp237 and Asp240 are accessible both from the aqueous phase and from within the membrane. The results provide more direct evidence that the three residues are located within transmembrane helices and suggest that Asp237 and Asp240 are either located near the periplasmic surface of the membrane or exposed within a solventfilled cleft in the permease.

Deletion of putative transmembrane helix III from the lactose permease of Escherichia coli results in complete loss of transport activity. Similarly, replacement of this region en bloc with 23 contiguous Ala, Leu, or Phe residues abolishes active lactose transport. The observations suggest that helix III may contain functionally important residues; therefore, this region was subjected to Cysscanning mutagenesis. Using a functional mutant devoid of Cys residues (Cless permease) each residue from Tyr 75 to Leu 99 was individually replaced with Cys. Twentyone of the 25 mutants accumulate lactose to >70% of the steadystate exhibited by Cless permease, and an additional 3 mutants transport to lower, but significant levels (4060% of Cless). Cys replacement for Leu 76 results in low transport activity (18% of Cless). However, when placed in the wildtype background, mutant Leu 76 → Cys exhibits highly significant rates of transport (55% of wild type) and steadystate levels of lactose accumulation (65% of wild type). Immunoblots reveal that the mutants are inserted into the membrane at concentrations comparable to wild type. Studies with Nethylmaleimide show that mutant Gly 96 → Cys is rapidly inactivated, whereas the other singleCys mutants are not altered significantly by the alkylating agent. Moreover, the rate of inactivation of Gly 96 → Cys permease is enhanced at least 2fold in the presence of βgalactopyranosyl 1thioβ,Dgalactopyranoside. The observations demonstrate that although no residue per se appears to be essential, structural properties of helix III are important for active lactose transport.

The possibility that simple lipophilic cations such as tetraphenylphosphonium (TPP+), tetraphenylarsonium (TPA+), triphenylmethylphosphonium (TPMP+), and diphenyldimethylphosphonium (DDP+) are substrates for the multidrug-resistance transport protein, P-glycoprotein, was tested. Hamster cells transfected with and overexpressing mouse mdr1 or mouse mdr3 exhibit high levels of resistance to TPP+ and TPA+ (20-fold) and somewhat lower levels of resistance to TPMP+ and DDP+ (3-12-fold). Transfected cell clones expressing mdr1 or mdr3 mutants with decreased activity against drugs of the MDR spectrum (e.g., Vinca alkaloids and anthracyclines) also show reduced resistance to lipophilic cations. Studies with radiolabeled TPP+ and TPA+ demonstrate that increased resistance to cytotoxic concentrations of these lipophilic cations is correlated quantitatively with a decrease in intracellular accumulation in mdr1- and mdr3-transfected cells. This decreased intracellular accumulation is shown to be strictly dependent on intact intracellular nucleotide triphosphate pools and is reversed by verapamil, a known competitive inhibitor of P-glycoprotein. Taken together, these results demonstrate that lipophilic cations are a new class of substrates for P-glycoprotein and can be used to study its mechanism of action in homologous and heterologous systems.