α-helical membrane proteins comprise about a quarter of all proteins in a cell and carry out a wide variety of essential cellular functions. This thesis is focused on topology analyses of bacterial membrane proteins. The topology describes the two-dimensional structural arrangement of a protein relative to the membrane. By combining large-scale experimental and bioinformatics techniques we have produced experimentally constrained topology models for the major part of the Escherichia coli membrane proteome. This represents a substantial increase in available topology information for bacterial membrane proteins. Many membrane protein structures show signs of internal duplication and approximate two-fold in-plane symmetry. We propose a step-wise pathway to explain how proteins with such internal inverted repeats have evolved. The pathway is based on the ‘positive-inside’ rule and starts with a protein that can adopt two topologies in the membrane, i.e. a “dual” topology protein. The gene encoding the dual topology protein is duplicated and eventually, through re-distribution of positively charge residues, the two resulting homologous proteins become fixed in opposite orientations in the membrane. Finally, the two proteins may fuse into one single polypeptide with an internal inverted repeat structure. Finally, we re-create the proposed step-wise evolutionary pathway in the laboratory by showing that only a small number of mutations are required in order to transform the homo-dimeric, dual topology protein EmrE into a hetero-dimeric complex composed of two oppositely oriented proteins.

Bacteriophage holins are a diverse group of proteins that are responsible for the spontaneous and specifically-timed triggering of host cell lysis. The best-studied holin, S105 of phage lambda, is known to form lesions, or “holes”, in the inner membrane of E. coli which are large enough to allow the endolysin through to the periplasm. S105 has been studied extensively by both genetic and biochemical approaches; however, little is known about the mechanism of hole formation or the structure of the lambda holin and its inner membrane lesion. An in vitro system for reconstituting hole formation by S105 was developed in which liposomes containing a self-quenched fluorophore served as artificial cell membranes (1-2). Upon delivery of solubilized S105 to the liposomes, an increase in fluorescence was observed, indicating that the fluorophore within the liposomes had escaped into the surrounding media via an S105-mediated hole in the membrane. This in vitro system, which has been optimized in this work, has been a valuable biochemical tool for analysis and reconstitution of the pathway to S105 hole formation in the cell membrane. Due to the difficulty associated with over-expression and purification of toxic membrane proteins, there are no solved structures of bacteriophage holins. Sample preparation and experimental conditions for NMR spectroscopy were optimized and structural information about a lambda holin mutant protein was obtained. Specifically, micellar contacts of transmembrane domain regions versus water contacts of the C-terminal region, secondary structure, and backbone dynamics were determined. Cryo-electron microscopy was used to visualize the inner membrane lesions formed by phage holins [lambda] S105, P2 Y, and T4 T. Therefore, the large holes initially seen in cells expressing S105 are not specific to the lambda holin, nor to class I holins. The S105 holes average ~340 nm (3), and are the largest membrane lesions ever observed in biology. They are stable at their original size, and are not localized to a specific region of the membrane. In addition, missense mutants of S105 were used to correlate hole size, protein accumulation, and lysis timing in a current model for the S105 hole formation pathway.

During co-translational integration, the transmembrane (TM) sequence of a nascent membrane protein moves laterally into the ER lipid bilayer upon reaching the translocon. Our lab has previously shown that this movement is a multistep process, but it was not clear whether the observed photocrosslinking of the TM segment to translocon proteins resulted from specific interactions or simply from TM-translocon proximity. If the latter, the TM α-helix will be oriented randomly with respect to translocon proteins, whereas, if the former, a specific TM helix surface would face TRAM and/or Sec61α. Integration intermediates were prepared by in vitro translation of truncated mRNAs in the presence of a Lys-tRNA analog with a photoreactive moiety attached to the lysine side-chain. When photoadduct formation was monitored as a function of probe location within the TM α-helix, we found that the extent of photocrosslinking to TRAM and Sec61α was non-random. Thus, the TM sequence occupies a distinct location within the translocon, a result that can only be achieved through protein-protein interactions that mediate the lateral movement, positioning, and integration of the TM sequence. In the case of multi-spanning membrane proteins, it was unknown how multiple hydrophobic regions integrated into the ER membrane. By placing photoprobes within each of several TM domains of a multi-spanning membrane protein, we were able to determine at what stage of integration each TM segment was no longer adjacent to translocon proteins. Using this approach we were able to establish a mechanism of integration for multi-spanning membrane proteins co-translationally inserted into the ER membrane.

The high affinity sodium/glucose cotransporter (SGLT1) couples transport of Na+ and glucose. Investigation of the structure/function relationships of the sodium/glucose transporter (SGLT1) is crucial to understanding co-transporter mechanism. In the first project, we used cysteine-scanning mutagenesis and chemical modification by methanethiosulphonate (MTS) derivatives to test whether predicted TM IV participates in sugar binding. Charged and polar residues and glucose/galactose malabsorption (GGM) missense mutations in TM IV were replaced with cysteine. Mutants exhibited sufficient expression to be studied in detail using the two-electrode voltage-clamp method in Xenopus laevis oocytes and COS-7 cells. The results from mutants T156C and K157C suggest that TM IV participates in sugar interaction with SGLT1. This work has been published in Am J Physiol Cell Physiol 295 (1), C64-72, 2008. The crystal structure of Vibrio parahaemolyticus SGLT (vSGLT) was recently published (1) and showed discrepancy with the predicted topology of mammalian SGLT1 in the region surrounding transmembrane segments IV-V. Therefore, in the second project, we investigated the topology in this region, thirty-eight residues from I143 to A180 in the N-terminal half of rabbit SGLT1 were individually replaced with cysteine and then expressed in COS-7 cells or Xenopus laevis oocytes. Based on the results from biotinylation of mutants in intact COS-7 cells, MTSES accessibility of cysteine mutants expressed in COS-7 cells, effect of substrate on the accessibility of mutant T156C in TM IV expressed in COS-7 cells, and characterization of cysteine mutants in TM V expressed in Xenopus laevis oocytes, we suggest that the region including residues 143-180 forms part of the Na+- and sugar substrate-binding cavity. Our results also suggest that TM IV of mammalian SGLT1 extends from residue 143-171 and support the crystal structure of vSGLT. This work has been published in Biochem Biophys Res Commun 378 (1), 133-138, 2009 Previous studies established that mutant Q457C human SGLT1 retains full activity, and sugar translocation is abolished in mutant Q457R or in mutant Q457C following reaction with methanethiosulfonate derivatives, but Na+ and sugar binding remain intact. Therefore, in the third project, we explored the mechanism by which modulation of Q457 abolishes transport, Q457C and Q457R of rabbit SGLT1 expressed in Xenopus laevis oocytes were studied using chemical modification, the two-electrode voltage-clamp technique and computer model simulations. Our results suggest that glutamine 457, in addition to being involved in sugar binding, is a residue that is sensitive to conformational changes of the carrier. This work has been published in Biophysical Journal 96 (2), 748-760, 2009. Taken together our study along with previous biochemical characterization of SGLT1 and crystal structure of vSGLT, we propose a limited structural model that attempts to bring together the functions of substrate binding (Na+ and sugar), coupling, and translocation. We propose that both Na+ and sugar enter a hydrophilic cavity formed by multiple transmembrane helices from both N-terminal half of SGLT1 and C-terminal half of SGLT1, analogous to all of the known crystal structures of ion-coupled transporters (the Na+/leucine transporter, Na+/aspartate transporter and lactose permease). The functionally important residues in SGLT1 (T156 and K157 in TM 4, D454 and Q457 in TM 11) are close to sugar binding sites.

Legionella spp. can cause a life threatening form of pneumonia, which is observed world-wide. Outbreaks of the disease are, unfortunately, not a rare event, despite the introduction of government regulations which enforce the mandatory testing of cooling towers to ensure that they contain levels of the organism which are regarded as being within safe limits. Therefore, cooling towers should be monitored for Legionella spp. by using a biosensor. These could potentially save the community from a great deal of morbidity and mortality due to legionellosis. This study identified and investigated novel outer membrane proteins in L. pneumophila, and analysed their potential for use in a Legionella biosensor. A combination of bioinformatics and laboratory investigations was used to identify the Omp87, an outer membrane protein of L. pneumophila which had not been previously described in this organism. Sequence analysis of the protein showed that it shares similarity with various other members of the Omp85 protein family, including the D15 antigen of Haemophilus influenzae and the Oma87 of Pseudomonas aeruginosa. The omp87 gene of L. pneumophila was amplified and cloned, and was found to encode a protein of 786 amino acids, with a molecular weight of 87 kDa. Distribution studies revealed that the gene is present in most, but not all species and serogroups of Legionella. To investigate the function of the Omp87 protein in L. pneumophila, the omp87 gene was insertionally inactivated with the use of a kanamycin resistance gene. Amplicons of this disrupted gene were then introduced into L. pneumophila, and a double-cross over event occurred, integrating the inactivated gene into the genome of the organism. This resulted in non-viable cells, indicating that the gene is essential in L. pneumophila. The expression vector pRSETA was used to express the Omp87 protein in E. coli, and four truncates of varying sizes were designed, through the use of different PCR primers. Two of the protein truncates were then expressed and purified by gravity flow chromatography using columns packed with Ni-NTA sepharose resin. Following analysis of the proteins by SDS-PAGE and Western blotting, polyclonal antibodies were raised against the truncates. Distribution studies were then performed using the antiserum with different strains and species of Legionella. This study demonstrated that most serogroups of L. pneumophila, and most other Legionella species reacted with the polyclonal anti-Omp87 L. pneumophila antisera. Cross-reactivity was also observed with most other Legionella related organisms tested. The results presented in this thesis demonstrated that the Omp87 protein or the omp87 gene can be used to construct a biosensor. In addition other novel outer membrane proteins were identified which could also serve as potential targets for a biosensor. These biosensors will be able to identify Legionella spp. in water reservoirs and in clinical samples and hopefully reduce the number of infections and deaths caused by this organism.

abstract: Coronaviruses are medically important viruses that cause respiratory and enteric infections in humans and animals. The recent emergence through interspecies transmission of severe acute respiratory syndrome coronavirus (SARS-CoV) strongly supports the need for development of vaccines and antiviral reagents. Understanding the molecular details of virus assembly is an attractive target for development of such therapeutics. Coronavirus membrane (M) proteins constitute the bulk of the viral envelope and play key roles in assembly, through M-M, M-spike (S) and M-nucleocapsid (N) interactions. M proteins have three transmembrane domains, flanked by a short amino-terminal domain and a long carboxy-terminal tail located outside and inside the virions, respectively. Two domains are apparent in the long tail - a conserved region (CD) at the amino end and a hydrophilic, charged carboxy-terminus (HD). We hypothesized that both domains play functionally important roles during assembly. A series of changes were introduced in the domains and the functional impacts were studied in the context of the virus and during virus-like particle (VLP) assembly. Positive charges in the CD gave rise to viruses with neutral residue replacements that exhibited a wild-type phenotype. Expression of the mutant proteins showed that neutral, but not positive, charges formed VLPs and coexpression with N increased output. Alanine substitutions resulted in viruses with crippled phenotypes and proteins that failed to assemble VLPs or to be rescued into the envelope. These viruses had partially compensating changes in M. Changes in the HD identified a cluster of three key positive charges. Viruses could not be recovered with negatively charged amino acid substitutions at two of the positions. While viruses were recovered with a negative charge substitution at one of the positions, these exhibited a severely crippled phenotype. Crippled mutants displayed a reduction in infectivity. Results overall provide new insight into the importance of the M tail in virus assembly. The CD is involved in fundamental M-M interactions required for envelope formation. These interactions appear to be stabilized through interactions with the N protein. Positive charges in the HD also play an important role in assembly of infectious particles. / Dissertation/Thesis / Ph.D. Microbiology 2010

Abstract All neurons and macroglial cells of vertebrates derive from the neuroepithelium. Neuroepithelial (NE) cells first proliferate and, after closure of the neural tube, some cells start generating neurons. It is still unclear what triggers differentiation but apparently there is interplay between extrinsic (secreted or transmembrane signals) and intrinsic factors. Diriving from the embryonic ectoderm, the NE cells inherit epithelial characteristics. It has been shown in other developmental systems that epithelial determinants, such as cell-cell contacts and contact to basal laminar components can guide differentiation. The key epithelial features include cell polarity, and tight junctions. We studied these in the NE at two developmental stages, the neural plate, a proliferative stage and the neural tube, a differentiative stage. The polarity of membrane proteins in NE cells was studied with polarly budding viruses. Mouse embryos were infected with Fowl plague- and vesicular stomatitis viruses and cultured in a whole embryo culture system. Viral envelope proteins (HA and G-protein) were localized by indirect immunofluorescence and immunoelectron microscopy. HA was polarized in the plate stage neuroepithelial cells, whereas in the tube it was not polarized anymore. It is also shown by penetrance of apically injected horseradish peroxidase that in the neural plate, NE cells have functional tight junctions. At this stage, they also express occludin, a transmembrane protein of tight junctions, as shown by indirect immunofluorescence. In the neural tube, the paracellular barrier is lost and there is no occludin expression. In contrast, expression of ZO-1, a cytoplasmic protein binding to occiudin, is upregulated. The downregulation of these epithelial features occurs in all NE cells, irrespective of their mode of division and before any neurons are generated in the NE. The change is initiated already at the plate stage and coincides with the switch from E- to N-cadherin. Later, with birth of neurons, the proliferative cell layer also looses contact to basal lamina. This is probably an important step in the regulation of neurogenesis. Furthermore, lack of apico-basolateral polarity of non-anchored membrane proteins may contribute to the mechanism of rapid neuron generation. Until now, it has been impossible to distinguish a neuroepithelial cell preparing for neuron generation from the surrounding cells that give rise to two precursor cells. In this study, the immediate neuron precursors are shown to express the antiproliferative gene TIS2 1. Using this new marker and ISH in serial sections, we show that the switch to differentiation is initiated in single NE cells.

The cystic fibrosis transmembrane conductance regulator (CFTR) is an ABC transporter family protein that acts as an ion channel. Mutations in CFTR cause the most common genetic disease in Caucasian populations, cystic fibrosis (CF). The high-resolution X-ray crystal structure of CFTR is now needed to aid the design of CFTR-targeted drugs for CF treatment and also to elucidate the molecular mechanisms behind its unique function as an ATP-ligand gated ion channel. However, until now, such structural studies have been severely limited by the lack of sufficient quantities of purified full-length CFTR protein. This thesis reports the novel over-expression and purification of milligram quantities of the chicken orthologue of CFTR protein from a Saccharomyces cerevisiae (yeast) expression system. A green fluorescent protein (GFP) tag fused to the CFTR C-terminus allowed rapid detection of the protein throughout the purification procedure. CFTR was expressed under an inducible promoter and appeared localised at, or near to, the plasma membrane, where it represented around 1 % of total protein after isolation in yeast microsomes. CFTR was solubilised from microsomes and purified using the detergents dodecylmaltoside (DDM) and lyso-phosphatidyl glycerol (LPG), by nickel affinity and size exclusion chromatography (SEC) to yield 1-2 mg of CFTR protein per 18 L fermentation culture. CFTR thermal stability was probed using fluorescent measurements to reveal a two-state cooperative unfolding transition around 40 °C for the DDM-purified protein, but no such transition was observed for the LPG-purified material. Light scattering and electron microscopy techniques revealed that, in LPG, CFTR was a homogenous population of monomeric particles around 60-Å in length that were soluble up to 8 mg/ml protein concentration. In DDM, CFTR was only soluble below 0.4 mg/ml protein concentration where is existed as a very heterogenous population of different sized amorphous particles, including dimeric particles around 180-Å in length. The DDM-purified CFTR protein could be crystallised as monomers in two-dimensional (2D) crystals with similar lattice parameters to 2D crystals of CFTR purified from mammalian cells. The ATPase activity of DDM-purified and reconstituted CFTR was similar to already published rates, at around 13 nmol Pi/min/mg integrated over a reaction time of 60 min, with an apparent affinity Km for ATP of 0.14 mM. Such a low ATPase rate compared to other ABC transporters may be due to the observed rapid run-down of activity with time and correlation with published CFTR channel gating kinetics. CFTR showed reduced ATPase activity after purification in LPG, suggesting a structural destabilisation in this detergent. The protocols presented here can now be used to provide sufficient quantities of purified CFTR protein for novel biochemical and biophysical studies. The tendency of CFTR to aggregate in a mild detergent remains a major obstacle towards 3D crystallisation trials and a high-resolution structure.

This study sought to determine whether NaCl altered the plasma membrane proteins in alfalfa exhibiting differential NaCl concentrations, and whether caso4 modified the responses. Two alfalfa cultivars, Centurion and Condor, were grown in 0.5 strength Hoagland solution in a greenhouse. The cultivars were exposed to 0, 88, and 132 mM of NaCl alone and mixed with caso4 .H20 at 7 and 14 Mm caso4 for 3, 9, and 60 days. In experiment 1, roots were dried to determine their Na, Ca, K, and Mg concentration. The results were similar to previous reports in which CaS04 alleviated the salt stress by increasing K and Mg levels and reducing Na. In experiment 2, after proteins of the plasma membrane were isolated and their purity was determined by vanadate, ATPase activity showed a significant increase in the presence of calcium. In addition, total plasma membrane protein was analyzed by sodium dodecylsulfate-polyacrylamide gel electrophoresis. Salt treatments induced both quantitative and qualitative changes in proteins. These changes were affected by the length of exposure to treatment solution or the ability of the plants to adapt to the salt stress.

No description available.