|Home | About | Journals | Submit | Contact Us | Français|
The outer membranes of Gram-negative bacteria, mitochondria, and chloroplasts all contain transmembrane β-barrel proteins. These β-barrel proteins serve essential functions in cargo transport and signaling and are also vital for membrane biogenesis. They have also been adapted to perform a diverse set of important cellular functions including acting as porins, transporters, enzymes, virulence factors and receptors. Recent structures of transmembrane β-barrels include that of a full length autotransporter (EstA), a bacterial heme transporter complex (HasR), a bacterial porin in complex with several ligands (PorB), and the mitochondrial voltage-dependent anion channel (VDAC) from both mouse and human. These represent only a few of the interesting structures of β-barrel membrane proteins recently elucidated. However, they demonstrate many of the advancements made within the field of transmembrane protein structure in the past few years.
Gram-negative bacteria, mitochondria, and chloroplasts contain both an inner and outer membrane. The outer membrane (OM) contains numerous β-barrel proteins commonly called outer membrane proteins (OMPs), which serve essential functions in cargo transport and signaling and are also vital for membrane biogenesis[1–2]. The number of strands observed in all OMPs thus far range from 8 to 24, and almost all OMPs contain an even number of strands (Table 1). Despite the common β-barrel scaffold, OMPs have evolved to perform many different functions including acting as porins, transporters, enzymes, and receptors [3–6].
In Gram-negative bacteria, OMPs are synthesized in the cytoplasm and then transported across the inner membrane and into the periplasm by the SecYEG translocon [5,7] (Figure 1A). Once in the periplasm, chaperones such as SurA and Skp guide nascent OMPs across the periplasm and peptidoglycan layer to the inner leaflet of the OM. Here, the nascent OMPs are recognized by a complex known as the β-barrel assembly machinery (BAM) complex which folds and inserts the new OMPs into the OM [7–8]. The Escherichia coli BAM complex consists of five subunits named BamA (an OMP itself), BamB, BamC, BamD, and BamE. Although we do not yet understand how the BAM complex functions in detail, studies have shown that BamA and BamD are essential for cell viability and OMP biogenesis, while BamB, BamC, and BamE serve regulatory roles .
Similar mechanisms for OMP biogenesis exist for both mitochondria and chloroplasts, providing further evidence of the evolutionary relationships of these organelles [5,9–10]. In mitochondria, nascent OMPs are imported across the OM into the inner membrane space (IMS) via the translocase of the outer mitochondrial membrane (TOM) complex  (Figure 1B). Once in the IMS, the nascent OMPs are escorted by chaperones [12–13] (Tim proteins) to the sorting and assembly machinery (SAM) complex where they are folded and inserted into the mitochondrial OM . In chloroplasts, the translocon of the outer membrane of the chloroplast (TOC) complex serves a similar function to that of the TOM complex in mitochondria [5,10,14] (Figure 1C). However, it is not clear whether the TOC complex has the ability to insert proteins into the OM of the chloroplast on its own. Toc75-V, an essential β-barrel protein present in the OM of chloroplasts , may be involved in inserting and folding β-barrel proteins into the OM similar to the SAM complex in mitochondria .
In this review, we will briefly discuss recent structures of components of the BAM complex and will summarize selected new structures of bacterial and mitochondrial β-barrel proteins, indicating the important findings from each study.
All of the structural information on the folding systems responsible for biogenesis of OMPs has come exclusively from structures from the BAM complex (Figure 1D, 1E, 1F, 1G and 1H). Recently, structures of BamB [17–19], BamE [20–21], and large portions of the periplasmic domain of BamA [22–24] have been reported. In addition, reports that the structures of BamC and BamD have been solved  suggest that these structures will be released soon as well, thereby providing the structures of all known lipoprotein components of the BAM complex. The β-barrel domain of BamA is thought to resemble FhaC (whose structure is known ), which is an OMP involved in two partner secretion. Together, these structures provide the basic framework to begin to visualize what the BAM complex looks like at the cell membrane. However, more structural information, particularly of the core folding machineries for each system, is needed to fully understand how nascent OMPs are recognized and then folded and inserted into OMs.
While there have been many OMP structures solved in recent years, it is beyond the scope of this review to do a comprehensive assessment of them all. We therefore provide a complete list of current known β-barrel structures in Table 1. Here, we have selected the structures of EstA, HasR, and PorB for brief discussion.
Autotransporters (AT) are OMPs that contain both a β-barrel transmembrane domain (β-domain) and a passenger domain . Once the protein is folded and inserted into the OM, passenger domains of ATs either remain tethered to the β-domain or a cleavage event within the AT releases the passenger into the extracellular space. Current structures of ATs that have been reported are either of the β-barrel domain or the passenger domain alone. However, the structure of EstA, an esterase enzyme from Pseudomonas aeruginosa, contains both the β-domain and an intact passenger domain . EstA is important for various cellular processes including cell motility and biofilm formation and interestingly, has been used for the presentation of various proteins on the surface of bacteria [29–30]. A majority of passenger domains from ATs are thought to have a β-solenoid fold abundant in β-strands . By contrast, the EstA passenger domain was found to contain a more globular fold consisting primarily of α-helices which likely contributed to the crystallization of the intact AT.
Iron is essential for the survival of Gram-negative bacteria and therefore various iron import mechanisms have evolved to obtain iron from the environment . Most pathogenic bacteria have complex mechanisms for scavenging iron from their host for survival, highlighting the importance of fully understanding these processes . Recently, the crystal structures of the hemophore complex HasA-HasR from Serratia marcescens was reported in the heme-bound and unbound forms . HasR is an OMP that is responsible for importing heme across the OM and can accomplish this with or without its hemophore binding partner HasA. These structures reveal that the transfer of the high affinity heme from HasA to HasR results from a steric clash resulting from a conformational change in HasA, likely induced by binding HasR. Similar mechanisms are thought to exist for other bacterial iron scavenging systems which utilize transferrin, lactoferrin, or hemopexin as iron sources.
Recently the structure of PorB, the second most common OMP in Neisseria meningitidis, was reported . PorB is required for pathogenesis and acts as a voltage-gated semi-selective porin responsible for transporting sugars across the OM, with smaller sugars transported more quickly than their larger counterparts. In addition, PorB has been reported to act in a number of ways at different stages of infection primarily through targeting mitochondria. Here, PorB has been reported to interact with the mitochondrial voltage-dependent anion channel (VDAC) at the mitochondrial OM and even to be inserted into the mitochondrial inner membrane, where it causes disruption of the mitochondrial membrane potential [36–37]. Structures of PorB were reported in apo form and in complex with sucrose, galatose, and the ATP analogue, AMP-PNP. Analysis of the structures indicates that PorB exploits various regions of the pore surface in order to mediate substrate transport; ATP binding may regulate this process by leading to a constriction of the pore diameter.
The transport of small molecules and metabolites between the cytoplasm and mitochondria of all eukaryotic organisms is mediated by VDAC, the most abundant protein in the mitochondrial outer membrane [38–40]. In addition to providing a lifeline between the mitochondria and the cell, VDAC is also involved in the release of signaling molecules that lead to apoptotic cell death [41–42]. The structures of VDAC-1 from both Homo sapiens and Mus musculus were recently solved by three independent labs using three different methods: (1) NMR of LDAO solubilized protein , (2) a hybrid X-ray/NMR approach on Cymal-5 and LDAO solubilized protein [44–45] and (3) X-ray crystallography of DMPC/CHAPSO bicelle solubilized protein . All three methods utilized refolding techniques to obtain large quantities of purified VDAC from E. coli inclusion bodies. Although these structures were obtained in different detergent and lipid environments, they all share remarkable similarity with RMSDs for Cα carbons of less than 1.7 Å.
The structures of all bacterial OM β-barrels solved to date are composed of an even number of strands arranged in an anti-parallel fashion. However, VDAC is composed of an odd-numbered 19-standed β-barrel in which strands β1 and β19 form a parallel β-sheet, constituting a new class of β-barrel transmembrane proteins (Figure 3A). Preceding the β-barrel domain, the N-terminus of VDAC loops down through the interior of the pore. While no secondary structure appears to be present in the N-terminus of the NMR structure, in both the X-ray/NMR hybrid and X-ray structures the N-terminus forms an α-helix that is perpendicular to and lines the inside of the β-barrel wall (Figure 3A). In the X-ray lipid bicelle structure, this N-terminal helix reduces the minimal diameter of the pore to 14 Å, which is large enough to accommodate many metabolites. Both the N-terminal α-helix and the N-terminal portion of the β-barrel wall are believed to be involved in the gating function of the channel, allowing it to be turned on and off to control the passage of larger metabolites such as nucleotides [44,47]. Along with the NMR structural study of VDAC, the binding sites for cholesterol, NADH, and Bcl-xL were elucidated . In addition, VDAC has sequence similarity and is believed to share a structural resemblance to Tom40, which led to a homology model of Tom40 based on the VDAC structure  (Figure 3B).
β-barrel proteins have traditionally been crystallized in their detergent solubilized form. However, recent advances and improvements of technology in the membrane protein crystallography field are allowing β-barrels to be crystallized in more native-like lipid environments. Bicelles, discs of lipids surrounded by a belt of detergent, are a newly-developed tool that was used in the crystallization of M. musculus VDAC . Screening kits for bicelle crystallization are now commercially obtainable from MemX Biosciences, and several references on this new membrane protein crystallization technique are also available [49–50]. The use of lipidic cubic phase (LCP) has also recently come back into the limelight [51–52]. Traditionally, the monoolein lipid-protein mixture used in LCP crystallization set-ups has been difficult to work with and pipette (it has a toothpaste-like consistency). This made the task of performing crystallization screens with LCP extremely tedious and time-consuming. The latest progress in pipetting robotics allows for the precision dispensing of nanoliter volumes of LCP onto 96-well substrate plates. Robots with the ability to pipette LCP are available from TTP Labtech and Art Robbins Instruments at the time of the writing of this article.
There are many online resources available for gathering further information on β-barrel membrane proteins and their structures. Similar to the Protein Data Bank, there are databases dedicated specifically to membrane proteins such as The Membrane Protein Data Bank (http://www.mpdb.tcd.ie) , the Membrane Proteins of Known 3D Structure database (http://blanco.biomol.uci.edu/Membrane_Proteins_xtal.html) hosted by Stephen White’s lab at UC Irvine, and the Protein Data Bank of Transmembrane Proteins (http://pdbtm.enzim.hu). While each website has its unique features, all of these resources provide current statistics on deposited membrane protein structures, educational resources for further information, tools for structure analysis, and searchable databases. Another useful website is the PRED-TMBB: Prediction of Transmembrane Beta-barrel Proteins server (http://bioinformatics2.biol.uoa.gr/PRED-TMBB/) which can be used to predict both the number of β-strands and topology of putative β-barrel membrane protein sequences. In addition, the Cherezov lab website (http://cherezov.scripps.edu/resources.htm) has a wealth of information on the LCP crystallization method.
The structure determination of β-barrel membrane proteins remains an exciting field of research and has benefited from both an increase in interest in the field and by the advancement of new crystallization and sample preparation techniques. While we have briefly summarized four structures here, there have been many interesting OMP structures published recently and all OMP structures have been organized in Table 1 for quick reference. Still, many important structures remain elusive primarily due to the limited availability of functional, homogeneous samples. However, recent reports by a number of groups have shown that co-expression of a particular chaperone or binding partner increased the expression levels of their target protein [8,54]. This approach is likely to increase the yields of samples needed for structural studies and should lead to more new structures of β-barrels in the near future. Much remains to be learned: the structures of many hypothetical or unidentified bacterial OM proteins with new or unknown functions are still unsolved, and very few structures of OMP complexes have been determined to date.
The mechanism by which β-barrels are properly folded and inserted into the OM of bacteria, mitochondria, and chloroplasts is poorly understood at present. Future structures of BamA, Tom40, Sam50, Toc75-III and Toc75-V, both alone and in complex with BAM, TOM, SAM, and TOC accessory proteins, will further illuminate our knowledge of this pathway. Once we understand how these complexes function in OMP biogenesis, we can design better systems to improve heterologous expression of OMPs targeted for structural analysis.
>We provide an introduction to β-barrel proteins. >We review the structures of three bacterial outer membrane proteins. >We review the recent structures of VDAC. >We provide links to useful online databases for β-barrel proteins. >We summarize the current field of b-barrel transmembrane proteins and provide our opinion about the future.
We thank K. Zeth for providing a homology model for Tom40. JWF, NN, and SKB are supported by the Intramural Research Program of the NIH, National Institute of Diabetes and Digestive and Kidney Diseases.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.