Moazur Rahman

Topology-informed strategies for the overexpression and purification of membrane proteins

Membrane proteins represent a significant fraction of all genomes and play key roles in many aspects of biology, but their structural analysis has been hampered by difficulties in large-scale production and crystallisation. To overcome the first of these hurdles, we present here a systematic approach for expression and affinity-tagging which takes into account transmembrane topology. Using a set of bacterial transporters with known topologies, we tested the efficacy of a panel of conventional and Gateway™ recombinational cloning vectors designed for protein expression under the control of the tac promoter, and for the addition of differing N- and C-terminal affinity tags. For transporters in which both termini are cytoplasmic, C-terminal oligohistidine tagging by recombinational cloning typically yielded functional protein at levels equivalent to or greater than those achieved by conventional cloning. In contrast, it was not effective for examples of the substantial minority of proteins that have one or both termini located on the periplasmic side of the membrane, possibly because of impairment of membrane insertion by the tag and/or att-site-encoded sequences. However, fusion either of an oligohistidine tag to cytoplasmic (but not periplasmic) termini, or of a Strep-tag II peptide to periplasmic termini using conventional cloning vectors did not interfere with membrane insertion, enabling high-level expression of such proteins. In conjunction with use of a C-terminal Lumio™ fluorescence tag, which we found to be compatible with both periplasmic and cytoplasmic locations, these findings offer a system for strategic planning of construct design for high throughput expression of membrane proteins for structural genomics projects.

Ligand binding analyses of the putative peptide transporter YjdL from E. coli display a significant selectivity towards dipeptides

Proton-dependent oligopeptide transporters (POTs) are secondary active transporters that couple the inwards translocation of di- and tripeptides to inwards proton translocation. Escherichia coli contains four genes encoding the putative POT proteins YhiP, YdgR, YjdL and YbgH. We have over-expressed the previously uncharacterized YjdL and investigated the peptide specificity by means of uptake inhibition. The IC(50) value for the dipeptide Ala-Ala was measured to 22 mM while Ala-Ala-Ala was not able to inhibit uptake. In addition, IC(50) values of 0.3 mM and 1.5 mM were observed for Ala-Lys and Tyr-Ala, respectively, while the alanyl-extended tripeptides Ala-Lys-Ala, Ala-Ala-Lys, Ala-Tyr-Ala and Tyr-Ala-Ala displayed values of 8, >50, 31 and 31 mM, respectively. These results clearly indicate that unlike most POT members characterized to date, including YdgR and YhiP, YjdL shows significantly higher specificity towards dipeptides.

Investigation of the structure and function of a Shewanella oneidensis arsenical-resistance family transporter

The toxic metalloid arsenic is an abundant element and most organisms possess transport systems involved in its detoxification. One such family of arsenite transporters, the ACR3 family, is widespread in fungi and bacteria. To gain a better understanding of the molecular mechanism of arsenic transport, we report here the expression and characterization of a family member, So_ACR3, from the bacterium Shewanella oneidensis MR-1. Surprisingly, expression of this transporter in the arsenic-hypersensitive Escherichia coli strain AW3110 conferred resistance to arsenate, but not to arsenite. Purification of a C-terminally His-tagged form of the protein allowed the binding of putative permeants to be directly tested: arsenate but not arsenite quenched its intrinsic fluorescence in a concentration-dependent fashion. Fourier transform infrared spectroscopy showed that the purified protein was predominantly α-helical. A mutant bearing a single cysteine residue at position 3 retained the ability to confer arsenate resistance, and was accessible to membrane impermeant thiol reagents in intact cells. In conjunction with successful C-terminal tagging with oligohistidine, this finding is consistent with the experimentally-determined topology of the homologous human apical sodium-dependent bile acid transporter, namely 7 transmembrane helices and a periplasmic N-terminus, although the presence of additional transmembrane segments cannot be excluded. Mutation to alanine of the conserved residue proline 190, in the fourth putative transmembrane region, abrogated the ability of the transporter to confer arsenic resistance, but did not prevent arsenate binding. An apparently increased thermal stability is consistent with the mutant being unable to undergo the conformational transitions required for permeant translocation.

New insights into the substrate specificities of proton-coupled oligopeptide transporters from E. coli by a pH sensitive assay

Proton‐coupled oligopeptide transporters (POTs) are secondary active transporters that facilitate di‐ and tripeptide uptake by coupling it to an inward directed proton electrochemical gradient. Here the substrate specificities of Escherichia coli POTs YdgR, YhiP and YjdL were investigated by means of a label free transport assay using the hydrophilic pH sensitive dye pyranine and POT overexpressing E. coli cells. The results confirm and extend the functional knowledge on E. coli POTs. In contrast to previous assumptions, alanine and trialanine appears to be substrates of YjdL, albeit poor compared to dipeptides. Similarly tetraalanine apparently is a substrate of both YdgR and YhiP.

Functional Investigation of Conserved Membrane-Embedded Glutamate Residues in the Proton-Coupled Peptide Transporter YjdL

Proton-dependent oligopeptide transporters (POTs) are secondary active symporters that utilize the proton gradient to drive the inward translocation of di- and tripeptides. We have mutated two highly conserved membraneembedded glutamate residues (Glu20 and Glu388) in the E. coli POT YjdL to probe their possible functional roles, in particular if they were involved/implicated in recognition of the substrate N-terminus. The mutants (Glu20Asp, Glu20Gln, Glu388Asp, and Glu388Gln) were tested for substrate uptake, which indicated that both the negative charge and the side chain length were important for function. The IC50 values of dipeptides with lack of or varying N-terminus (Ac-Lys, Gly- Lys, β-Ala-Lys, and 4-GABA-Lys), showed that Gly-Lys and β-Ala-Lys ranged between ~0.1 to ~1.0 mM for wild type and Glu20 mutants. However, for Glu388Gln the IC50 increased to ~2.0 and > 10 mM for Gly-Lys and β-Ala-Lys, respectively, suggesting that Glu388, and not Glu20, is able to sense the position of the N-terminus and important for the interaction. Furthermore, uptake as a function of pH showed that the optimum at around pH 6.5 for wild type YjdL shifted to 7.0-7.5 for the Glu388Asp/Gln mutants while the Glu20Asp retained the wild type optimum. Uptake by the Glu20Gln on the other hand was completely unaffected by the bulk pH in the range tested, which indicated a possible role of Glu20 in proton translocation.

Probing the Putative Active Site of YjdL: An Unusual Proton-Coupled Oligopeptide Transporter from E. coli

YjdL from E. coli is an unusual proton-coupled oligopeptide transporter (POT). Unlike prototypical POTs, dipeptides are preferred over tripeptides, in particular dipeptides with a positively charged C-terminal residue. To further understand this difference in peptide specificity, the sequences of YjdL and YdgR, a prototypical E. coli POT, were compared in light of the crystal structure of a POT from Shewanella oneidensis. Several residues found in the putative active site were mutated and the activities of the mutated variants were assessed in terms of substrate uptake assays, and changes in specificity in terms of uptake inhibition. Most strikingly, changing the YjdL specific Asp392 to the conserved Ser in YjdL obliterated the preference for a positively charged C-terminal residue. Based on this unique finding and previously published results indicating that the dipeptide N-terminus may interact with Glu388, a preliminary orientation model of a dipeptide in the YjdL cavity is presented. Single site mutations of particularly Ala281 and Trp278 support the presented orientation. A dipeptide bound in the cavity of YjdL appears to be oriented such that the N-terminal side chain protrudes into a sub pocket that opens towards the extracellular space. The C-terminal side chain faces in the opposite direction into a sub pocket that faces the cytoplasm. These data indicated a stabilizing effect on a bulky N-terminal residue by an Ala281Phe variant and on the dipeptide backbone by Trp278. In the presented orientation model, Tyr25 and Tyr58 both appear to be in proximity of the dipeptide backbone while Lys117 appears to be in proximity of the peptide C-terminus. Mutational studies of these conserved residues highlight their functional importance.

Promoter hypermethylation of Wnt pathway inhibitors in hepatitis C virus - induced multistep hepatocarcinogenesis

BackgroundAberrant DNA methylation profiles are a characteristic feature of almost all types of cancers including hepatocellular carcinoma (HCC) and play an important role in carcinogenesis. In spite of the accumulating evidence that suggests appearance of such aberrations at precancerous stages, very little effort has been invested to investigate such possible methylation events in patients at risk of developing HCC i.e. those suffering from chronic hepatitis C virus (HCV) infection and liver cirrhosis (LC). We reasoned that such an analysis could lead to the identification of novel predictive biomarkers as well as potential drug targets.MethodsPromoter methylation status of two Wnt inhibitors SFRP2 and DKK1 was quantitatively analyzed by bisulfite pyrosequencing in a series of liver biopsy samples. These biopsies were collected from HCV-infected individuals suffering from chronic hepatitis (CH; n = 15), liver cirrhosis (LC; n = 13) and hepatocellular carcinoma (HCC; n = 41). DNA isolated from infection free normal livers (N; n =10) was used as control.ResultsOur analysis revealed that both of the genomic loci were significantly hypermethylated in CH patients’ livers as compared to normal controls (p = 0.0136 & 0.0084 for SFRP2 and DKK1, respectively; Mann–Whitney U test). DNA methylation levels for both loci were also significantly higher in all the diseased cohorts as compared to normal controls (p < 0.0001 and = 0.0011 for SFRP2 and DKK1, respectively; Kruskal-Wallis test). However, a comparison between three disease cohorts (CH, LC & HCC) revealed no significant difference in levels of DNA methylation at DKK1 promoter. In contrast, a progressive increase in DNA methylation levels was observed at the SFRP2 promoter (i.e. N < CH & LC < HCC).ConclusionThis study demonstrated that in HCV infected liver tissues hypermethylation at promoter regions of key cancer related genes SFRP2 and DKK1, appears early at CH and LC stages, long before the appearance of HCC.

Probing metal ion substrate-binding to the E. coli ZitB exporter in native membranes by solid state NMR

Metal ion homeostasis is important for healthy cell function and is regulated by metal ion transporters and chaperones. To explore metal ion binding to membrane transport proteins we have used cadmium-113 as a solid state NMR probe of the Escherichia coli zinc exporter ZitB present in native membrane preparations. Competition experiments with other metal ions indicated that nickel and copper are also able to bind to this protein. Metal ion uptake studies were also performed using ZitB-reconstituted into proteoliposomes for a well established fluorescence assay. The results of both the solid state NMR and the uptake studies demonstrate that ZitB is potentially capable of transporting not only zinc but also cadmium, nickel and copper. The solid state NMR approach therefore offers great potential for defining the substrate spectrum of metal ion transporter proteins in their native membrane environments. Further, it should be useful for functional dissection of transporter mechanisms by facilitating the identification of functional residues by mutational studies.

Comparative molecular dynamics simulation of Hepatitis C Virus NS3/4A protease (Genotypes 1b, 3a and 4b) predicts conformational instability of the catalytic triad in drug resistant strains.

The protease domain of the Hepatitis C Virus (HCV) nonstructural protein 3 (NS3) has been targeted for inhibition by several direct-acting antiviral drugs. This approach has had marked success to treat infections caused by HCV genotype 1 predominant in the USA, Europe, and Japan. However, genotypes 3 and 4, dominant in developing countries, are resistant to a number of these drugs and little progress has been made towards understanding the structural basis of their drug resistivity. We have previously developed a 4D computational methodology, based on 3D structure modeling and molecular dynamics simulation, to analyze the active sites of the NS3 proteases of HCV-1b and 4a in relation to their catalytic activity and drug susceptibility. Here, we improved the methodology, extended the analysis to include genotype 3a (predominant in South Asia including Pakistan), and compared the results of the three genotypes (1b, 3a and 4a). The 4D analyses of the interactions between the catalytic triad residues (His57, Asp81, and Ser139) indicate conformational instability of the catalytic site in HCV-3a and 4a compared to that of HCV-1b NS3 protease. The divergence is gradual and genotype-dependent, with HCV-1b being the most stable, HCV-4a being the most unstable and HCV-3a representing an intermediate state. These results suggest that the structural dynamics behavior, more than the rigid structure, could be related to the altered catalytic activity and drug susceptibility seen in NS3 proteases of HCV-3a and 4a.

LC–MS/MS based method development for the analysis of florfenicol and its application to estimate relative distribution in various tissues of broiler chicken

Florfenicol, a broad spectrum bacteriostatic antibiotic belonging to amphenicol class, is widely used in poultry and livestock for the treatment of various infections. The major metabolite of florfenicol in different animal species is florfenicol amine which is exploited as the marker residue for the determination of florfenicol. Analysis of florfenicol merely by solvent extraction cannot determine the accurate amount of the drug present in incurred tissues (muscle, liver and kidney) of treated birds, as indicated by this study. Thus the methods solely based on solvent extraction may lead to false negative results. A reliable LC-MS/MS based confirmatory method for the analysis of florfenicol and its metabolites in chicken muscle was developed and validated according to the European Union Commission Decision 2002/657/EC. The method was based on acid hydrolysis to liberate non-extractable residues having presumably been covalently bound to tissues, and to convert all the florfenicol residues as well as its metabolites into florfenicol amine. The amine was subsequently recovered with ethyl acetate at pH 10.5, defatted and further cleaned up with dispersive solid phase extraction (dSPE). The LC separation was achieved on reverse phase C-18 column with isocratic elution using acetonitrile/water mobile phase and the analysis was performed on linear ion trap mass spectrometer. Calibration curve was obtained over a concentration range of 25-600μg/kg for chicken muscles. The accuracy values ranged from 84 to 101.4% and the precision values for within day and between days ranged from 1.2-11.7%, respectively. Limit of detection (LOD), limit of quantification (LOD), CCα and CCβ values were 0.98, 3.2, 113 and 126μg/kg, respectively. The developed method was highly robust and was further applied to estimate the relative distribution of solvent-extractable against solvent-non-extractable florfenicol drug residues in muscle, liver and kidney samples of broiler chicken after 5days of oral dosing.