Characterization of Slam-mediated Surface Lipoprotein Translocation across the Bacterial Outer Membrane
by
Yogesh Hooda
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Biochemistry University of Toronto
© Copyright by Yogesh Hooda 2019
i
Characterization of Slam-mediated surface lipoprotein translocation across the bacterial outer membrane
Yogesh Hooda
Doctor of Philosophy
Department of Biochemistry University of Toronto
2019 Abstract
Surfaces of many Gram-negative bacteria are decorated by peripheral membrane proteins that are anchored in the membrane by a lipid group, commonly referred to as surface lipoproteins or SLPs.
SLPs play key role in nutrient acquisition, immune evasion and have been proposed as excellent vaccine antigens. Previously our lab had shown that the proper display of host transferrin binding
SLP TbpB in Neisseria meningitidis required an outer membrane protein called Slam. The aim of the present study was to investigate the role Slam plays in SLP biogenesis.
Using bioinformatic analysis, we show that Slams are present in a number of Gram-negative bacteria. Putative Slam genes are often found adjacent to their putative SLP substrates. In N. meningitidis, we discovered two Slam paralogs, Slam1 and Slam2, that are specific for SLPs TbpB and HpuA respectively. All putative Slam-dependent SLPs contain a C-terminal 8-stranded soluble barrel domain. The last two strands of the SLP were found to be essential for TbpB translocation and the C-terminal 8-stranded barrel domain mediated Slam specificity. Using GST-fused TbpB, we showed that the Slam-dependent translocation occurs from the C-terminus to the N-terminus.
To investigate Slam mechanism, we developed an in vitro translocation system. Upon the addition of the periplasmic chaperone LolA, SLPs can be released from spheroplasts into the supernatant.
ii
We discovered that Slam containing proteoliposomes can successfully translocate spheroplast released SLPs into the liposomal lumen. Addition of other outer membrane factors such as the Bam complex did not increase the efficiency of SLP insertion and Slam1&2 retained their specificity in the assay. Interestingly, Slam1 proteoliposomes were also able to translocate purified unfolded
TbpB into the lumen. Collectively, these findings show that Slams are both necessary and sufficient for the translocation of SLPs across the outer membrane, indicating that they act as translocons.
iii
Acknowledgments
I am more interested in research now than I was before starting graduate school. I might be more competent at it too. Being in school for so long, I realize how rare that is and I have a lot of people to thank for it.
First and foremost, I would like to thank my supervisor Dr. Trevor Moraes. Trevor has been an excellent mentor and adviser. He always highlighted the importance of designing sound experiments with proper controls and being respectful to other people and equipment in the lab. I have learnt a lot under his supervision, and I hope to have him as my mentor for the rest of my scientific career.
Apart from Trevor, I would also like to thank a number of people in the Moraes lab who worked on this project with me. Christine identified Slam as one of the hits in a genetic screen 7 years ago now. Since then, she has been instrumental in the development of tools and assays I describe in this thesis. She manages the lab with an iron fist and lots of cat pictures, which kept me going. I am also grateful to Andrew and Sang for being my co-collaborators on the Slam project. For those who know them, they are two very different people to work with and yet they were instrumental in some of the most important results presented in this study.
My favorite part of working in a lab is arguing with other people over why a given experiment has failed, especially given that everyone involved is generally wrong. During my time in the Moraes lab, I have interacted with a lot of intelligent and enthusiastic people who never hesitated to hypothesize. I would personally like to thank current and past Moraes lab members: Ana, Megha, Charles, Esther, Tom, Nick, Chuxi, Epshita, Jaime, Steven and Maciej for all their constructive feedback. During my time here, I have also had the opportunity to interact with and mentor highly intelligent and motivated undergraduates. I am happy to report that most of them still intend to pursue a PhD, or at least that is what they tell me. I would also like to thank our collaborators Drs. Scott Gray-Owen and Tony Schryvvers for their support.
I had a very fruitful time being at the University of Toronto. My committee members Drs. Roman Melnyk and Will Navarre were very helpful, and I always came out of committee meetings smarter. I also interacted with a number of other faculty members in the department of Biochemistry who were always supportive of my work. Apart from academics, I also had a lot of fun being part of the
iv
Biochemistry Graduate Student Union. I helped in organizing departmental social events and the experience was always rewarding and often more successful than scientific experiments. Being part of UofT has provided me with numerous opportunities of growth in my personal and professional life. While I actively tried to avoid as many of these events as possible, the vast size of UofT made sure that I end up attending a few stochastic events, for which I am now very grateful. Starting graduate school in Toronto was one of the best decisions I made, especially given that it was the only choice I had.
None of the work described here would have been possible without the constant support of my family. My parents might not understand what I do, but they always knew in their heart that I could be doing something worse. I also want to thank my brother who has aged considerably waiting in earnest for a PhD brother; his watch is ending soon. During graduate school I acquired two parents and a younger brother through marriage. Having done their PhDs many moons ago, my parent-in- laws always accepted that I will never be rich. Same goes from my brother-in-law, who is on his own path of doing a PhD but in the much poorer social sciences.
This brings me to the most important person I want to acknowledge, my cat’s legal guardian and my wife Dr. Senjuti Saha. A lot has happened in our lives in the past 8 years. We got a cat named Ava Fontaine, started traveling the world, got married, moved into a beautiful apartment and survived graduate school together. I am excited and curious about what the future holds.
v
Table of Contents
Acknowledgments ...... iv
Table of Contents ...... vi
List of Figures ...... x
List of Appendices ...... xiii
List of Abbreviations ...... xiv
Chapter 1 Introduction ...... 1
1.1 Overview ...... 1
1.2 Cell envelope of Gram-negative bacteria ...... 1 1.2.1 Inner membrane proteins...... 3 1.2.2 Outer membrane proteins ...... 3 1.2.3 Lipoproteins ...... 4
1.3 Surface lipoproteins ...... 6 1.3.1 Experimental methods for identifying SLPs...... 7 1.3.2 Prevalence of SLPs ...... 8 1.3.3 Translocation pathways for SLPs ...... 9
1.4 Surface lipoproteins in Neisseria ...... 11 1.4.1 Iron acquisition systems: TbpB, LbpB and HpuA ...... 12 1.4.2 Vaccine antigens: fHbp and NHBA ...... 13 1.4.3 Partially surface-exposed proteins: AniA, NalP ...... 13 1.4.4 Other SLPs: MIP, SliC and FrpD ...... 15
1.5 Study of neisserial SLP biogenesis and discovery of Slam ...... 16 1.5.1 Role of Slam in N. meningitidis pathogenicity ...... 18 1.5.2 Role of beta-barrel domain of Slam in translocation of TbpB ...... 20 1.5.3 E. coli translocation assay ...... 22
1.6 Overview of the thesis ...... 23
Chapter 2 Prevalence of Slam-dependent SLP translocation in Gram-negative bacteria ...... 26
2.1 Overview ...... 26
2.2 Methods ...... 27 2.2.1 Identification of Slam homologs...... 27
vi
2.2.2 Analysis of gene neighbourhoods around putative Slam homologs ...... 27 2.2.3 Bacterial strains and growth condition ...... 28 2.2.4 Generation of plasmids for expression of Slams and SLPs ...... 28 2.2.5 Flow cytometry...... 29 2.2.6 Sucrose density ultracentrifugation ...... 30
2.3 Results ...... 30 2.3.1 Identification of putative Slam homologs in Gram-negative bacteria ...... 30 2.3.2 Identification and functional characterization of Slam-like proteins in Neisseria ...... 32 2.3.3 Identification and characterization of Slam2 in N. meningitidis ...... 33 2.3.4 Slams adjacent to TbpB in M. catarrhalis and H. influenzae translocate their respective TbpBs to the surface in E. coli ...... 35 2.3.5 Predicted SLP genes are found adjacent to Slam genes in a number of Gram-negative bacteria ...... 37 2.3.6 A putative SLP gene in Pasteurella multocida is displayed on the surface of E. coli in a Slam-dependent manner 39 2.3.7 Comparison of putative SLP proteins revealed a conserved structural domain ...... 41 2.3.8 A number of non-SLP genes found adjacent to putative Slam homologs ...... 42
2.4 Discussion ...... 43
Chapter 3 Investigating the molecular details of Slam function using an E. coli reconstitution assay ...... 47
3.1 Overview ...... 47
3.2 Methods ...... 48 3.2.1 Bacterial strains and growth conditions ...... 48 3.2.2 Cloning of Slam, SLP homologs ...... 49 3.2.3 Flow cytometry...... 50 3.2.4 Globomycin assay...... 51 3.2.5 Sucrose density ultracentrifugation ...... 51 3.2.6 Pulldown of TbpB with LolA R43L mutant ...... 52 3.2.7 Spheroplast generation and flow cytometry assay ...... 52 3.2.8 Flow cytometry analysis of TamA and TamB deficient E. coli ...... 53 3.2.9 Purification of GST-C-lobe and Slam1 from outer membrane ...... 53 3.2.10 Crosslinking studies between Slam1 and GST-C-lobe-cys ...... 53
3.3 Results ...... 54 3.3.1 Slam acts at the outer membrane ...... 54 3.3.2 Slam and its membrane domain exhibit specificity ...... 56 3.3.3 Slam is specific for SLPs ...... 58
vii
3.3.4 TbpB C-lobe contains the translocation motif ...... 58 3.3.5 The last two strands within TbpB contains the translocation motif ...... 60 3.3.6 Lumenal facing residues of the two strands of TbpB are important for translocation ...... 63 3.3.7 C-terminal domain dictates specificity of Slam-dependent SLPs ...... 65 3.3.8 Trapping TbpB translocation across the outer membrane ...... 65 3.3.9 Role of Omp85 family members in Slam function ...... 68 3.3.10 Cross-linking experiments with Slam mutants ...... 69
3.4 Discussion ...... 71
Chapter 4 Investigating the molecular details of Slam function using an in vitro reconstitution of translocation...... 75
4.1 Overview ...... 75
4.2 Methods ...... 76 4.2.1 Bacterial strains and growth conditions ...... 76 4.2.2 Cloning of Slam, SLP homologs ...... 77 4.2.3 High-throughput assay for Slam function ...... 77 4.2.4 Purification of Slams ...... 77 4.2.5 Purification of Bam complex...... 78 4.2.6 Liposome and proteoliposome preparation ...... 78 4.2.7 Purification of LolA ...... 79 4.2.8 Spheroplast release assay ...... 79 4.2.9 Bam complex functional assay ...... 80 4.2.10 Co-secretion translocation assay ...... 80 4.2.11 Post-secretion translocation assay ...... 81 4.2.12 Purification of TbpB ...... 81 4.2.13 Translocation assay with purified TbpB ...... 81
4.3 Results ...... 82 4.3.1 Selection of Slam homologs for purification ...... 82 4.3.2 Purification of Slam1 and Slam2 ...... 84 4.3.3 Purification of functional Bam complex ...... 85 4.3.4 Proteoliposomes for Slam1 and the Bam complex ...... 87 4.3.5 Release of TbpB and HpuA from spheroplasts by purified LolA ...... 89 4.3.6 Slam1 is necessary and sufficient for TbpB translocation ...... 90 4.3.7 Post-secretion translocation of Slam proteoliposomes ...... 90 4.3.8 Slam2 is necessary and sufficient for HpuA translocation ...... 93 4.3.9 Slam1 and Slam2 retain their specificity in the in vitro translocation assay ...... 93
viii
4.3.10 Reconstitution of purified TbpB translocation in Slam proteoliposomes ...... 96
4.4 Discussion ...... 97
Chapter 5 Summary and Future directions ...... 100
5.1 Key findings of the study ...... 100
5.2 Slam is a protein translocon in the outer membrane ...... 101
5.3 Model of Slam function ...... 102
5.4 Future directions ...... 104 5.4.1 Chaperones involved in translocation ...... 104 5.4.2 Slam secretion and specificity motif on SLPs ...... 105 5.4.3 Structure of Slams ...... 106 5.4.4 Slam molecular mechanism...... 107
References ...... 108
ix
List of Figures
Figure 1.1 Structure of Gram-negative bacterial cell envelope ...... 2
Figure 1.2: Translocation of lipoproteins in Gram-negative bacteria ...... 5
Figure 1.3: Translocation of SLPs across the bacterial outer membrane ...... 10
Figure 1.4: Surface lipoproteins in Neisseria ...... 14
Figure 1.5: Transposon mutagenesis for discovery of Slam ...... 17
Figure 1.6 Characterization of nmb0313 in Neisseria meningitidis ...... 19
Figure 1.7 Role of beta-barrel domain on Slam translocation ...... 21
Figure 1.8 Reconstitution of Slam-dependent SLP translocation in E. coli ...... 22
Figure 1.9 Potential models for Slam function ...... 24
Figure 2.1: Putative Slam family of proteins in Gram negative bacteria ...... 31
Figure 2.2: Slam and SLP homologs in Neisseria species ...... 33
Figure 2.3: Slam2 characterization and identification of its substrate, HpuA ...... 34
Figure 2.4: Translocation assay with Slam and TbpB pairs from Moraxella catarrhalis and Haemophilus influenzae ...... 36
Figure 2.5: Slam related gene clusters identified in this study with known lipoproteins ...... 38
Figure 2.6: Identification of Slam-dependent surface lipoprotein in Pasteurella multocida ...... 40
Figure 2.7: Structures of known Slam-dependent SLPs ...... 42
Figure 2.8: Putative non-lipoprotein substrate of Slam homolog in Neisseria gonorrhoeae...... 43
Figure 3.1: Translocation pathway for the surface lipoprotein, TbpB, to the cell surface ...... 55
x
Figure 3.2 Slam1 and Slam2 demonstrate substrate specific translocation of SLPs in the E. coli translocation assay ...... 57
Figure 3.3: Localization of PgaB in E. coli cells co-expressing Slams ...... 58
Figure 3.4 Translocation of TbpB mutants and truncations ...... 59
Figure 3.5: Characterization of the mutants in last two strands ...... 61
Figure 3.6: Multiple sequence alignment of the last two strands of TbpB ...... 63
Figure 3.7: Domain swapping experiments between TbpB and HpuA ...... 64
Figure 3.8: Slam1 interacts with TbpB in E. coli ...... 66
Figure 3.9: Validation of α-GST antibodies for flow cytometry ...... 67
Figure 3.10: Analysis of TbpB surface exposure in TamA or TamB-deficient E. coli ...... 69
Figure 3.11: Crosslinking and studies on stalled Slam-GST-C-lobe complex ...... 70
Figure 4.1: Functional test of Slam homologs used in the study ...... 83
Figure 4.2: Purification strategy for Slam1&2 ...... 85
Figure 4.3: Purification and characterization of E. coli Bam complex ...... 86
Figure 4.4: Formation of Slam1 and Bam proteoliposomes...... 88
Figure 4.5: Purification and functional assay with LolA ...... 89
Figure 4.6: Slam1 is necessary and sufficient for TbpB translocation ...... 91
Figure 4.7: Slam2 is necessary and sufficient for HpuA translocation ...... 92
Figure 4.8: Slam1&2 retain their specificity in the in vitro translocation assay ...... 94
Figure 4.9: Reconstitution of TbpB translocation in a defined in vitro system...... 95
Figure 5.1: Outer membrane translocons in Gram-negative bacteria ...... 102
xi
Figure 5.2: Model of Slam-mediated SLP translocation across the outer membrane...... 103
xii
List of Appendices
A: List of Slam homologs identified in this study
B: Gene neighborhood analysis for 353 Slam homologs
C: List of strains, constructs and reagents used in the study
xiii
List of Abbreviations
ABC – ATP binding cassette
ATP – adenosine triphosphate cryoEM – cryo-electron microscopy
DHFR – dihydrofolate reductase
DTT - dithiothreitol
DUF – domain of unknown function
EDTA – ethylenediaminetetraacetic acid
FSC – forward scattering
FPLC – fast protein liquid chromatography
GST – glutathione S-transferase hTf – human transferrin
IM – inner membrane
IMP – inner membrane proteins
IPTG – isopropyl b-D-1-thiogalactopyranoside kDa – kilodalton
LB – Luria-Bertani broth
LCP – lipid cubic phase
LPS – lipopolysaccharide
MFI – mean fluorescence intensity
OM – outer membrane
OMP – outer membrane proteins
OMV – outer membrane vesicles
ORF – open reading frame
PBS – phosphate buffer saline
xiv
PE - phycoerthyrin
PFA - paraformaldehyde
PG – peptidoglycan
PK – proteinase K
SEM – standard error of the mean
SLP – surface lipoproteins
SP – signal peptide
TBDR – TonB-dependent receptor
TM – transmembrane helices
TPR – tetratricopeptide repeats
WT - wildtype
xv Chapter 1 Introduction
Acknowledgements: This chapter is adapted from the following published manuscripts: Hooda and Lai et al. Nat. Microbiology 2016, Hooda et al. FEMS Pathog Dis 2017 and Hooda et al. COSB 2018.
1.1 Overview
The bacterial cell envelope comprises of the cytoplasmic cell membrane and the cell wall [1]. A large number of bacterial species, including all Gram-negative bacteria also possess an additional membrane, referred to as an outer membrane. Embedded in these bacterial membranes are proteins that can either traverse the membranes (integral membrane proteins) or anchor to the membrane through a lipid binding domain or a post-translational modification (peripheral membrane proteins). The appropriate biosynthesis and transport of these membrane proteins to their final cellular destination is an important process and essential for bacterial cell survival.
The current thesis focuses on the specific translocation pathway employed by a class of peripheral membrane proteins referred to as surface lipoproteins [2]. These are soluble proteins that are anchored on the membrane through a lipid group present on their N-terminus. They play important roles in the cellular signaling and nutrient acquisition and have been extensively studied as putative vaccine antigens. Surface lipoproteins can be found in both Gram-positive and Gram-negative bacteria. While the systems required for successful display of SLPs require movement across the cytoplasmic membrane (through the Sec or the Tat translocon), the SLPs in Gram-negative bacteria require two additional steps that involves traversing the periplasmic space and the outer membrane. The specific pathway utilized by different SLPs to complete the final two steps is an active field of research and is the central focus of the current study.
1.2 Cell envelope of Gram-negative bacteria
The cell envelope of the Gram-negative bacteria is composed of a symmetrical inner membrane composed of phospholipids, a periplasmic space that contains a thin peptidoglycan layer (cell wall), and an asymmetric outer membrane [1] (Figure 1.1). The outer membrane is composed of phospholipids on the inner face and lipopolysaccharide that face the extracellular milieu. The
1
composition of lipopolysaccharide differs between different bacterial species and can even differ within strains of given bacterial species. Both these membranes are home to wide diversity of proteins that either span the membrane (integral membrane proteins) or are peripherally attached to one of the faces of these membranes.
Figure 1.1 Structure of Gram-negative bacterial cell envelope. The inner membrane (IM), peptidoglycan cell wall (PG) and the outer membrane (OM) are shown. Alpha-helical and Beta- barrel integral membrane proteins are found in IM and OM respectively. Peripheral membrane proteins such as lipoproteins have also been identified on the periplasmic face of IM and OM as well as the cell surface.
2
1.2.1 Inner membrane proteins
~25% of the genes in Escherichia coli encode for integral inner membrane protein or IMPs [3]. IMPs play roles in important cellular pathways involved in energy generation and transduction in the respiratory chain, signal transduction, biomolecule transport and cell division. IMPs can vary greatly in size and complexity owing to the diversity and number of transmembrane helices (TMs) as well as periplasmic and cytoplasmic domains present on them. Further, these proteins often form complexes with other IMPs and/or peripheral membrane proteins to perform their biological function. The first TM of IMPs upon its translation by ribosomes is recognized by signal recognition particle (SRP), a ribonucleoprotein particle found in all domains of life [3]. Ribosome nascent chain (RNC) - SRP complexes contact the SRP receptor FtsY at the IM, which delivers the RNCs to the Sec translocon, a conserved protein complex that inserts IMPs into the inner membrane. The core Sec complex is composed of SecY, SecE and SecG but accessory proteins (SecD, SecF, YajC and YidC) are important for efficient processing of complex IMPs [3]. Insertion of IMPs is critical for the survival of bacterial cells and a number of cytoplasmic and membrane- bound chaperones are important for the quality control of this important pathway. Interestingly, aside from the Sec-dependent co-translational insertion pathway, some IMPs have been shown to use a post-translational Sec-independent pathway that utilizes the insertase YidC [4].
1.2.2 Outer membrane proteins
Outer membrane proteins are integral membrane proteins that are present in the outer membrane of the Gram-negative bacteria. While most OMPs are known to be beta-barrel proteins, alpha-helical OMPs have been described [5]. OMPs often act as passive or active channels for the movement of ions and small molecules across the OM. Many OMPs also form trans-envelope complexes with IMPs that are involved in secretion of biomolecules across the cell envelope. OMPs also contain an N-terminal signal sequence, however this signal is less hydrophobic than IMP signal sequence, which is not recognized by the SRP. Hence, the movement of OMPs across the IM occurs post- translationally. In E. coli, most nascent OMPs released from the ribosomes bind to the SecA motor protein that then delivers them to the Sec translocon. The signal peptide is processed by signal peptidase I, and the mature OMPs are released in the periplasm. A network of periplasmic chaperones (SurA, Skp, DegP and FkpA) interact with OMPs and delivers them to the OM where they are inserted into the OM by the Bam complex [6]. In E. coli, the Bam complex is composed of
3
five proteins: BamA is an OMP that acts as the insertase and BamB, C, D and E are OM lipoproteins that are important for the assembly of the complex, efficient substrate processing and release [7]. Recent findings have shown that certain large OMPs (such as FimD) requires the Tam complex, which is composed of an OMP, TamA (a distant homolog of BamA) and a single-pass IMP with large and flexible periplasmic domain, TamB [8]. Also, a few Bam independent OMPs have also been identified and these often do not have a typical beta-barrel structure, however the mechanism of their insertion is not fully understood [9].
1.2.3 Lipoproteins
Lipoproteins are peripheral membrane proteins that are soluble and contain a N-terminal cysteine residue. The cysteine residue is post-translationally modified to add three acyl fatty acid chains, which allows lipoproteins to the attach to the membrane [10]. In E. coli, lipoproteins are primarily found on the periplasmic face of the inner and the outer membrane. However, in bacterial genus such as Neisseria, Spirochetes and Bacteroides, lipoproteins are also found on the surface [2]. These lipoproteins are referred to surface lipoproteins and will be described in the next section. Several OM lipoproteins can partially access the surface by forming transmembrane complexes with other lipoproteins (such as CagA) or OMPs (such as Lpp and RscF) [11].
The translocation pathway used by lipoproteins is shown in Figure 1.2. Lipoprotein precursors are synthesized in the cytoplasm with an N-terminal signal peptide. The last four residues at the C terminus of the signal peptide is the conserved region referred to as the lipobox motif ([LVI] [ASTVI] [GAS] C). The cysteine residue at the last position of the lipobox eventually becomes the first residue of the mature lipoprotein [12]. Most lipoprotein precursors contain the Sec-specific secretion signal peptide, but lipoprotein precursors with a twin-arginine translocation (Tat) specific secretion motif have also been reported [13]. Upon translocation across the inner membrane, lipoprotein precursors undergo a series of post-translational modifications by enzymes: diacylglyceryl transferase (Lgt), signal peptidase II (SpII) and apolipoprotein N-acyltransferase (Lnt), resulting in the removal of the signal peptide and attachment of lipid groups to the N- terminus and the cysteine side chain. This modification anchors the mature lipoprotein to the periplasmic leaflet of the inner membrane. The lipoprotein biosynthetic enzymes appear to be essential for viability with the exception of Lnt, which has been shown to be dispensable in certain
4
bacterial species [14]. Recent structural studies have provided insight into the mechanism by which these enzymes function [15–18].
Figure 1.2: Translocation of lipoproteins in Gram-negative bacteria. The key proteins and protein complexes implicated in transport of lipoproteins to the outer membrane are shown. Lipoproteins (green) contain an N-terminal signal peptide (red) that is recognized by the Sec (SecY – yellow, SecE – dark red and secG- blue, PDB ID: 5AWW) or the Tat translocon (TatC – TV red, PDB ID: 4B4A). Once inside the periplasm, the lipoproteins are post-translationally modified by three enzymes: Lgt (cyan, PDB ID: 5AZB), SpII (light green, PDB ID: 5DIR) and Lnt (purple, PDB ID: 5AZB). Most lipoproteins are then transported across the periplasm by the five-member Lol system composed of LolA (orange, PDB ID: 1IWL), LolB (green, PDB ID: 1IWM) and LolCDE (LolC – salmon, LolD dimer – dark blue and LolE – olive green). The model of LolCDE was obtained from the structure of LptBFG (PDB ID: 5UDF) and is shown in surface representation. Lipoprotein pullulanase in Klebsiella pneumoniae bypasses the Lol system by using the Type II secretion system, shown as a trans-envelope complex. PulD secretin model (purple) was obtained from structure of GspD from Vibrio cholerae (PDB ID: 5WQ8). The inner membrane Pul complex model (PulC – gold, PulE –dark yellow, PulF – grey, Pul GHIJK – light blue, PulL – magenta and PulM - green) is shown in surface representation. The structure of type IVa pilus machinery from Myxococcus xanthus (PDB ID: 3JC9) was used as a model for the inner membrane Pul complex. (Adapted from Hooda et al. COSB 2018) [2]
The lipoproteins are chiefly transported from the inner membrane to the periplasmic leaflet of the outer membrane by the Lol pathway [19]. In E. coli, the Lol system consists of five proteins; Lol ABCDE. The LolCDE forms an ATP binding cassette (ABC) transporter in the inner membrane
5
and releases lipoproteins that are destined to the outer membrane in an ATP-dependent manner [20,21]. The periplasmic chaperone LolA [22] then shuttles the lipoprotein from the inner membrane to the outer membrane and delivers it to an outer membrane receptor LolB [23]. While the structures of LolA and LolB have been solved, structures of LolCDE complex remain to be elucidated (structure of the homologous LptB2FG shed some light on this class of ABC transporters [24]). Recent identification of small molecules specific to inhibition of LolCDE, which may stabilize the complex, thereby allowing elucidation of its structure [25]. Residues close to the lipobox motif at the N-terminus of the lipoprotein serve as Lol sorting signals, however these requirements differ between different bacterial species [26–28]. Like the lipoprotein biosynthetic enzymes, Lol components are conserved and essential in E. coli, however this may not be true for other Gram-negative bacteria. LolB is only found in b- and g-proteobacteria [29], and many bacterial species lack LolE. Interestingly, lipoproteins that bypass the Lol system have also been reported, such as pullulanase in Klebsiella pneumoniae, which uses the Type II secretion system to move across the periplasmic space [30]. Further, the essentiality of the Lol system for movement of lipoproteins across the periplasm was recently brought to question by a recent publication by Grabowicz and Silhavy [31]. Collectively, these studies highlight a more complex pathway where different routes are used by different lipoproteins to cross the periplasm. Periplasmic chaperones (such as Skp) or LolA/B-like proteins such as the MlaA [32] and LptA [33] could potentially be involved in the movement of lipoprotein across the periplasm.
1.3 Surface lipoproteins
The lipoproteins present on the surface are referred to as surface lipoproteins or SLPs and have been found to be involved in key cellular pathways for nutrient acquisition, cellular adhesion and stress response [34]. Several SLPs have been identified to date in Gram-negative bacteria and have been reviewed recently [2,11,34–36]. Gram-negative SLPs, such as fHbp and NHBA in Neisseria meningitidis [37,38], TbpB in Haemophilus parasuis [39], OspA in Borrelia burdorferi [40] and Tp0751 in Treponema pallidum [41] to list a few, have been investigated as vaccine antigens. The cellular pathway for the translocation of lipoproteins across the inner membrane and through the periplasm to the inner leaflet of the outer membrane is well characterized, whereas the translocation
6
systems used by SLPs to move across the outer membrane has only been identified for a handful of SLPs.
The first reported SLP was TraT, a protein of the F sex factor in Escherichia coli [42]. Several other SLPs were identified soon after in other Gram-negative bacteria such as Klebsiella [43], Neisseria [44] and Spirochetes [45,46]. There has been an increase in the reports of SLPs with distinct structures and surface topologies from different bacterial species in recent years [11]. Most of the recent SLPs were first identified by functional proteomics screens often looking for vaccine antigens. Upon identification of a putative SLP, a number of experimental assays have been developed to confirm their lipidation and surface localization [47]. SLPs can be further characterized by biochemical and genetic studies to elucidate their biological function [35]. As SLPs contain soluble protein domains, they are also amenable to biophysical and structural characterization.
1.3.1 Experimental methods for identifying SLPs
A number of experimental assays have been developed to identify these surface-exposed lipoproteins or SLPs. First, cell fractionation assays have been used to identify the presence of the lipoprotein in the inner and the outer membrane [10]. Second, in-vivo lipidation of a lipoprotein has been tested by quantifying the amount of radiolabeled lipoprotein in bacterial cells supplemented with 3H-palmitic acid [48]. Third, dot blots or proteinase K shaving assays has been used to test the surface exposure of the lipoprotein and its sensitivity to protease activity respectively [49]. An alternative strategy is to use flow cytometry or high-resolution fluorescence microscopy using fluorescently labeled antibodies against the lipoprotein [27]. Electron microscopy has been used to confirm the surface display of a handful of SLPs [50]. All of these assays help in identifying a bonafide SLP and have been used in a number of bacterial species. However, these techniques are time-consuming and not applicable at a genome-scale.
Most of the current knowledge on neisserial SLPs comes from either examination of iron acquisition systems or vaccine development against neisserial pathogens. As traditional capsule based vaccines have been ineffective against serogroup B, large efforts have been given to the development of a protein or an outer membrane vesicle (OMV) based vaccine. This is illustrated by the extensive study conducted by Pizza et al. in which 350 candidate protein antigens were
7
identified in N. meningitidis. These candidate proteins were used to immunize mice to identify the production of a bactericidal antibody response against the pathogen [51]. Using this approach four potential vaccine antigens were identified, two of which were later confirmed to be SLPs. This work led to the development of 4CMenB, a recent FDA and EMA-approved vaccine, and demonstrates the efficacy of SLPs as vaccine antigens [37].
1.3.2 Prevalence of SLPs
From information available to date, it appears that different bacterial species contain a plethora of different SLPs. On one extreme is E. coli strain K12 that contains only three SLPs (RscF, Lpp and BamC) to date that are only partially surface displayed and the majority of lipoproteins (~100) in E. coli are localized to the periplasmic space. At the other extreme are Sphirochetes such as Borrelia burgdorferi, (causative agent of Lyme disease) where almost all lipoproteins are surface displayed [45,52]. Most bacteria are probably in the middle of the two extremes, such as N. meningitidis. N. meningitidis contains a handful of SLPs that include fully exposed SLP mammalian transferrin binding protein TbpB, autotransporter protease NalP that also contains a membrane spanning domain, and AniA, a partially surface exposed nitrite reductase [53]. Another interesting family of SLPs are found in Bacteroides sp., a family of mutualistic bacteria found in mammalian guts that breakdown complex carbohydrates [35]. Most of the Bacteroides SLPs recognize specific carbohydrates and/or cleave polysaccharide bonds. A recent study by Glenwright et al. provided the first structural snapshot of large complexes formed by Bacteroides SLPs with a membrane- spanning TonB-dependent receptor [54].
Taken together, these studies have established that SLPs are present in most, if not all Gram- negative bacteria. But this brings forward a more central question as to why Gram-negative bacteria would have surface lipoproteins? We can think of several reasons. First, given that SLPs often contain large and flexible anchor peptides on their N-terminus, they can extend much further from the cell surface than OM proteins and capture their interaction partners or substrates. Second, because of their peripheral attachment, SLPs have a different distribution in the outer membrane than OM proteins [55], which tend to cluster in supramolecular complexes within the outer membrane [56]. Third, unlike outer membrane proteins that are not specifically targeted for degradation, SLPs can be removed by a single proteolytic cut as is the case with the SLP LbpB and protease NalP [57]. Fourth, SLPs often contain catalytic subunits (such as proteases, hydrolases)
8
that can detoxify substrates on the surface before their subsequent uptake. Finally, unlike secreted hemophores and siderophores, surface-bound SLPs involved in iron acquisition cannot be hijacked by surrounding bacteria (social cheats). As more SLPs are discovered and studied, we will be able to discern the myriad roles they play in bacterial cell biology.
1.3.3 Translocation pathways for SLPs
Like all lipoproteins, SLPs enter the periplasm through the Sec translocon or the Tat translocon in an unfolded or folded (partially) manner respectively. Some SLPs (e.g. pullulanase that utilizes the Type II secretion system [29]) can be directly transported from the outer leaflet of the inner membrane to the cell surface without passing through a periplasmic intermediate. However, most other SLPs are thought to utilize the Lol system for crossing the periplasmic space. The current literature suggests that the Lol system does not discriminate between folded and unfolded lipoproteins [35,36]. It has been proposed that a subset of SLPs remain unfolded for their subsequent transport across the periplasm. While the role of the periplasmic chaperones in the biogenesis of integral OM proteins is well understood [58], their role in SLP biogenesis is poorly characterized. The requirement for a periplasmic ‘holding’ chaperone has been shown for Borrelia SLPs using a conditionally folding protein domain [59]. However, such studies are currently missing for SLPs in other Gram-negative bacteria.
Once the SLPs reach the inner leaflet of the outer membrane, they need to cross the asymmetric outer membrane which presents numerous challenges [34]. First, SLPs are often stable soluble proteins that can be either unfolded (Borrelia SLPs) or folded (pullulanase) before their transport across the outer membrane, which would dictate the translocon they use. Second, they contain a lipid anchor that in most cases (except SLPs like BamC) needs to be moved from the inner leaflet of the outer membrane to outer leaflet. Third, given that periplasmic space is devoid of ATP, the energy for the transport either comes from chaperones that are involved in the translocation, folding on the surface of the cell or an inner membrane system like the TonB-ExbB-ExbD system. These factors make transport of SLPs a non-trivial endeavor.
An important step towards understanding critical steps in SLP biogenesis was the identification of a potential secretion signal present on SLPs. The first studies towards this aim were performed by the Zückert group that showed that SLPs in Borrelia sp. contain the secretion motif is present on the N-
9
terminal tether peptide [59,60]. An N-terminal secretion motif has also been recently identified for Bacteroides SLPs by Lauber et al. [27]. However, unlike Borrelia, where no conserved secretion sequence was identified, Bacteroides required charged residues at position +2, +3 and +4. This begs the question if the N-terminal secretion motif is common to all SLPs or do other secretion strategies also exist? Furthermore, it is currently unclear if the specificity of the secretion signal is dictated by periplasmic factors or the outer membrane translocon.
One of the first families of SLPs for which the mechanism of surface translocation was discovered was the pullulanase which uses the Type II secretion system [30]. Another interesting route towards surface display is taken by RcsF, a member of the Rcs signaling system which requires the Bam complex and an OM porin for surface display (Figure 1.3) [61,62].
Figure 1.3: Translocation of SLPs across the bacterial outer membrane. Three systems have been implicated in the translocation of surface lipoproteins across the outer membrane: 1) The Bam complex (blue) is used by the Type Va autotransporter protease, NalP from Neisseria (green) for insertion and translocation of its passenger domain to the surface of the cell; 2) The Bam complex (blue) together with an outer membrane porin (red) is used by RscF (green) in E. coli, to reach the surface and 3) Slam (orange) is required by neisserial SLPs (green) such as TbpB and fHbp and can be used to reconstitute translocation of these SLPs to the surface of E. coli . These represent a small fraction of surface lipoproteins reported in the literature. (Adapted from Hooda et al. COSB 2018) [2]
10
The most well understood SLP translocation pathway is the Type-Va secretion system (autotransporter) used by the protease NalP [63]. A lot of work has been done towards understanding the mechanism of autotransporter function and it has been summarized previously [64–66]. Briefly, after crossing the inner membrane through the Sec translocon, the autotransporters are delivered to the Bam complex by periplasmic chaperones SurA and Skp. The Bam complex assembles the C-terminal 12-stranded transporter domain in the outer membrane. The insertion of the transporter domain leads to formation of a hairpin motif at the C-terminus of the passenger domain that allows the movement across the outer membrane (Figure 1.3). The lateral gate of the Bam complex is predicted to play a role in flipping the lipid anchor of NalP from the periplasmic leaflet to the surface after the transport of the protease domain.
These systems account for the translocation of only a handful of SLPs and for most SLPs, the outer membrane protein or complexes remain to be elucidated.
1.4 Surface lipoproteins in Neisseria
One of the most well studied families of SLPs is present in the genus Neisseria. Members of this genus can be found to reside on the mucosal surfaces of animals, including at least 10 species that can colonize humans [67]. Neisseria contains two human pathogens: Neisseria meningitidis, one of the leading causative agents of sepsis and meningitis, and Neisseria gonorrhoeae that causes the sexually transmitted disease gonorrhea. N. meningitidis and N. gonorrhoeae are both obligate human pathogens exhibiting unique adaptations that allow them to survive inside their host [68]. The importance of these pathogens in relationship to human health promoted fundamental research in these organisms including pioneering work in the understanding of OM protein and lipopolysaccharide (LPS) biogenesis that resulted in the identification of the Bam and Lpt complexes in N. meningitidis [69,70]. Furthermore, N. meningitidis and N. gonorrhoeae are targets for the development of vaccines. Two decades of research has resulted in the development of two protein-based vaccines against N. meningitidis serogroup B [38,71,72]. Interestingly, both vaccines use SLPs as vaccine antigens demonstrating the importance of this class of proteins. This wealth of information makes Neisseria a great model system for understanding the roles of SLPs in Gram- negative bacteria.
11
1.4.1 Iron acquisition systems: TbpB, LbpB and HpuA
Transferrin-binding protein B or TbpB was the first SLP identified from N. meningitidis as a 71 kDa protein expressed under iron limiting conditions [44]. The N-terminal sequence of TbpB showed homology to signal peptides of known lipoproteins and it was shown to undergo acylation [48]. Treatment of intact cells with proteases resulted in a loss of TbpB exposed to the extracellular environment suggesting that it is surface exposed [44]. Schryvvers and Morris demonstrated that TbpB specifically binds to human transferrin, a human serum iron binding protein allowing N. meningitidis to overcome nutritional immunity [73]. The tbpB gene lies immediately upstream of the gene encoding a TonB dependent OM receptor, TbpA [74]. Competitive solid phase binding assays illustrated that TbpA binds apo and holo transferrin indiscriminately, while TbpB preferentially binds iron loaded transferrin [75].
Analogous to TbpBA system, N. meningitidis also contains a bipartite system for acquiring iron from lactoferrin [76] and hemoglobin [77]. The systems are composed of an integral OM transporter, LbpA or HpuB, and a SLP, LbpB and HpuA. While TbpB and HpuA have been found to exclusively work in iron acquisition, recent evidence points to a role of LbpB in pathogen defense [78]. Clusters of negatively charged residues present in the C-lobe of LbpB have been shown to neutralize lactoferricin, a short 49 amino acid cationic antimicrobial peptide that functions as part of the innate immune system [79].
Structures of the apo form of TbpB were solved independently by Calmettes et al. and Noinaj et al. [80,81]. TbpB is a bilobed protein, and each lobe is comprised of a N-terminal β-handle and an 8 stranded C-terminal β-barrel (Figure 1.4). TbpB has a long, disordered anchoring peptide that may allow it to sequester transferrin at a greater distance from the bacterial cell surface. While the structure of full-length LbpB has not been solved, the N-lobe of LbpB shows that like TbpB, the N- terminal β-handle domain of LbpB contains predominantly β-sheets, as well as an α-helix and a short 310 helix [82]. LbpB also has a C-terminal 8 stranded β-barrel (Figure 1.4). Structures of full- length and the C-terminal region of HpuA from Kingella denitrificans (Kd) and N. gonorrhoeae respectively, were recently solved by X-ray crystallography [83]. Despite the low sequence identity with SLPs such as TbpB, HpuA contains an N-terminal β-handle, and C-terminal 8 stranded β- barrel (Figure 1.4). Interestingly, HpuA contains only one lobe composed of the N-terminal β- handle, and C-terminal 8 stranded β-barrel while TbpB and LbpB contain two lobes.
12
1.4.2 Vaccine antigens: fHbp and NHBA fHbp and NHBA were discovered through a search for novel vaccine antigens against N. meningitidis [51]. Both NHBA and fHbp was shown to be lipidated by the incorporation of 3H- palmitic acid, and surface exposed by flow cytometry and electron microscopy [51,84,50]. fHbp was shown to bind specifically to human factor H, which functions to protect host cells by inhibiting the activation of the alternative complement pathway [85]. The ability to bind factor H provides N. meningitidis protection against complement mediated killing. Initial functional studies of NHBA demonstrated that it bound to heparin and enhanced survival in a human serum bactericidal assay [86]. An in vitro adhesion assay using purified recombinant protein showed that NHBA binds to heparan sulfate proteoglycans present on the surface of epithelial cells, and this binding is mediated through an arginine-rich region present on NHBA [87]. NHBA deletion mutants are impaired in their ability to bind to epithelial cells, and sera against NHBA limits adhesion. NHBA has also been implicated in biofilm formation [88]. Furthermore, a peptide fragment called C2 generated by NalP mediated proteolysis of NHBA was found to increase the permeability of an endothelial cell monolayer [89]. C2 led to the internalization of VE-cadherin, a component of adherens junctions and it has been hypothesized to increase vascular permeability during meningococcal sepsis.
Representative structures of fHbp from each of the three variant groups were solved by X-ray crystallography or NMR (Figure 1.4). The N-terminal domain is described as “barrel-like”, and is composed of 6 antiparallel β-strands facing 2 smaller ones and a short α helix [90]. The C-terminal domain is comprised of an 8-stranded antiparallel β-barrel and a short 310 helix. The structure of the C-terminal region of NHBA (Figure 1.4) was solved by NMR since the N-terminal region was predicted to be largely unstructured due to the presence of many small amino acids and low complexity sequence [91]. NHBA contains an 8-stranded β-barrel and is highly structurally similar to that of TbpB and fHbp despite their low sequence identity.
1.4.3 Partially surface-exposed proteins: AniA, NalP
AniA was first identified as a lipoprotein that is expressed when N. gonorrhoeae is grown anaerobically [92,93]. AniA is a nitrite reductase (NirK) that coverts nitrite (NO2-) to nitric oxide (NO.), allowing the bacteria to use nitrite as the terminal electron acceptor to grow anaerobically. AniA is composed of an N-terminal lipobox containing a flexible peptide region, a central copper
13
(Cu)-containing nitrite reductase domain that has both Type I & Type II Cu centers, and a long- glycosylated C-terminal extension [94,95] (Figure 1.4). First evidence for the surface exposure of AniA came from the observation that expression of AniA provided protection against killing by human sera [96]. This was further validated by electron microscopy (ImmunoSEM) and trypsin digestion [97]. Curiously, the trypsin digestion was unable to deplete the entire AniA protein suggesting that AniA may only be partially exposed on the cell surface. Such an orientation has previously been observed for RscF and Lpp in E. coli [11]. The most likely region that is surface exposed is the C-terminal glycosylated region that contains an immunodominant epitope [97].
Figure 1.4: Surface lipoproteins in Neisseria. Structural representation of the 10 known neisserial SLPs are shown. SLPs for which atomic coordinates are available are shown in cartoon representation while modeled structures are shown in surface representation. TbpB (green, PDB ID: 3VE1) and LbpB (orange, PDB ID: 4U9C) possess two copies of the b-handle and b-barrel domain, while HpuA (magenta, PDB ID: 5EC6), fHbp (blue, PDB ID: 4AY1) and NHBA (yellow PDB ID: 2LFU) possess one copy. Only the structure of N-lobe of LbpB (PDB ID: 4U9C) has been solved and C-lobe structure is predicted based on TbpB C-lobe (PDB ID: 3PQU). The membrane topology of AniA (red, PDB ID: 1KBW) is based on the current literature; AniA is predicted to be partially surface exposed with its NirK structural domain predicted to be in the periplasm suggesting that the lipid anchor may also be in the inner leaflet. NalP (blue-green) is an autotransporter protease for which the structure of membrane bound b-barrel domain is available (PDB ID: 1UYN), while the protease domain is modeled based on structures of other serine proteases (PDB ID: 3AFQ). The structure of MIP (cyan) is obtained using Legionella-MIP as the model (PDB ID: 1FD9). FrpD forms a novel fold based on the recently solved structure (black, PDB ID: 5EDJ). The structure of SliC was modeled based on the structure of MliC in complex with hen egg white lysozyme (brown, PDB ID: 3F6Z). The lipid moiety for most neisserial SLPs is shown in the outer leaflet of outer membrane, however experimental evidence for this is currently lacking. (Adapted from Hooda et al. FEMS Pathog. Dis. 2017) [53]
14
Unlike AniA, NalP is an autotransporter SLP that has a distinct domain architecture composed of an N-terminal surface anchored protease domain and a C-terminal beta-barrel domain embedded in the outer membrane (Figure 1.4). NalP (AspA) was first identified as a 112-kDa-autotransporter protein containing a serine protease catalytic triad found on several secreted subtilisin-like proteases [63]. The lipidation and surface exposure of NalP was confirmed using 3H-palmitic acid labeling experiments, cellular fractionation and electron microscopy [63,98]. NalP was found to specifically cleave itself and other autotransporters (IgA protease, App and AusI) from the surface [98]. NalP also cleaves neisserial SLPs NHBA and LbpB [57,86]. The lipidation is important for retention of NalP on the surface and loss of lipidation results in lower cleavage of its substrates [57]. Given its role in processing several important virulence factors, NalP was found to be important for the survival of in human whole blood [99]. Interestingly, NalP can also enhance human serum survival by directly cleaving the a-chain of the human complement factor C3b in a species-specific manner [100]. Surface displayed NalP is antigenic and generates cross-reactive antibodies as confirmed by presence of anti-NalP antibodies in patient sera [63].
1.4.4 Other SLPs: MIP, SliC and FrpD
Macrophage Infectively Potentiator (MIP) was initially identified as a 30-kDa lipoprotein present in N. gonorrhoeae (Ng-MIP) with homology to surface exposed MIP proteins found in several intracellular pathogens [101]. Ng-MIP was shown to be a SLP using western blot analysis on N. gonorrhoeae OMVs, electron microscopy and 3H-palmitic acid labeling experiments [101]. Anti- Ng-MIP antibodies can be identified in sera from patients with uretheritis and disseminated gonoccocal infection [102]. Deletion of Ng-MIP leads to decreased survival of N. gonorrhoeae in macrophages indicating a key role of MIP in intracellular survival. A homolog of Ng-MIP referred to as Nm-MIP was also identified in N. meningitidis [103]. Nm-MIP was found to be upregulated in human sera and induces the production of protective antibodies against N. meningitidis. Both Nm- MIP and Ng-MIP are conserved within different N. meningitidis and N. gonorrhoeae strains [104]. While no structure is available for neisserial MIPs, Humbert et al. used the structure of Legionella MIP proteins to model the structure of neisserial MIP. Based on the model, neisserial MIPs exist as homodimers with each monomer consisting of a N-terminal a-helix involved in dimerization, a long central a-helix and a globular C-terminal domain containing the PPIase activity (Figure 1.4).
15
Truncated Nm-MIP excluding the globular C-terminal domain was found to elicit bactericidal antibodies against different serogroups of N. meningitidis [104].
FrpD is an iron-regulated protein described in N. meningitidis [105]. It is encoded by frpCD operon which also encodes for FrpC, an 1829-residue protein RTX protein that is secreted into the extracellular environment during the early stages of infection. FrpC is secreted by the Type I secretion system found in N. meningitidis. FrpD was confirmed as a lipoprotein through 3H- palmitic acid labeling and found to be present in the outer membrane [105]. It binds to the N- terminus of FrpC, indicating it is present on the cell surface. The structure of FrpD was recently solved by X-ray crystallography [106]. It is composed of 14 β-strands, one 310 helix and 3 α- helices. FrpD did not show structural similarity to any known proteins, indicating it has a novel fold. FrpD-FrpC (1-414) complex was found to be important for the ability of N. meningitidis to bind to the surface of epithelial cells in the nasopharynx [106].
Recently, two studies identified a lipoprotein in N. gonorrhoeae involved in inhibiting the activity of eukaryotic antimicrobial protein lysozyme [107,108]. Proteinase K shaving assays and dot blots were used to validate the surface exposure of the lipoprotein [107]. Hence, the lipoprotein was named surface-exposed lysozyme inhibitor of c-type lysozyme (SliC). SliC-like lysozyme inhibitor named adhesin complex protein (ACP) has been previously described in N. meningitidis, however ACP is not a lipoprotein [109]. SliC shows highest sequence similarity to MliC (also a lysozyme inhibitor) from P. aeruginosa (40% similarity) and E. coli (32% similarity), and the key residues involved in lysozyme binding were found to be conserved between MliC and SliC [106]. Structural prediction software predicts SliC folds into an 8-stranded lipocalin-like β-barrel domain (Figure 1.4). SliC was found to bind and specifically inhibit c-type lysozyme and provide survival advantage for N. gonorrhoeae in human tears and saliva, and in the murine model of gonococcal infection.
1.5 Study of neisserial SLP biogenesis and discovery of Slam
In order to study the factors involved in biogenesis of neisserial SLPs, we focused on the SLP TbpB. The structure of TbpB had been solved [80,110,111] and availability of reagents including a-TbpB antibodies and substrate human transferrin, made TbpB a good candidate. To identify
16
component(s) required for SLP translocation to the N. meningitidis cell surface, a genome-wide random Tn5-based transposon mutant library was constructed within the N. meningitidis strain B16B6 [49]. Using a whole cell solid phase binding assay, 4000 mutants from this library were analyzed for the ability to display TbpB on the cell surface (Figure 1.5A, B). One of the defective mutants had incorporated a Tn5 into a genetic region that aligns with the ORF nmb0313 of the published N. meningitidis MC58 genome sequence [49]. This nmb0313::Tn5 mutant (nmb0313/tn5) has effectively lost TbpB from its cell surface when compared with the wild type strain (Figure 1.5C, D). Complementation with a genomic copy of nmb0313 under the control of a lac promoter (nmb0313/tn5 + Slam1), together with a complete deletion of the entire gene (Δnmb0313) (Figure 1.5E, 1.6A), confirmed the importance of the gene in TbpB surface display [49]. Moraes - Figure S2 A Kan ....
Kan
Kan Sheared genomic DNA with N. meningitidis colonies with ~4000 Tn N. meningitidis mutants kanamycin cassette inserted by Tn5 Tn insertions (40 plates) B Plate: 2/40 D Mutants Gene name Predicted ORF
2E1 nmb1037 Putative glutamate- cysteine ligase 5H3 nmb0313 TPR-containing protein 24C10 nmb0173 LysR family trans- criptional regulator Transcription termin- 28D2 nmb0617 nation factor Rho Amidophospho- 3411F nmb0690 ribosyltransferase Whole cell solid phase binding assay of mutants probed with α TbpB antibodies C E Emp Compl WT Δtbpb 2E1 5H3 PFA PFA α TbpB α TbpB SDS SDS
Figure S2: Discovery of transposon mutants that lack TbpB surface localization from the N. meningitidis transposon library. α TbpB (A) A schematic of N. meningitidis transposon library generation. Sheared genomic DNA was subjected to an in-vitro Figure 1.5:reaction Transposon with transposon mutagenesis elements fanking for kanamycin discovery cassettes ofand Slama transposase,. (A) using A schematic the EZ::TN
on the same membranes as controls. Some mutants had no signal (black circle) or no growth (NG). (C) Example verification of TbpB surface expression mutants by solid phase binding assay and SDS-PAGE. After initial screening, wild type (WT), ∆tbpb, and transposon mutants lacking surface TbpB, such as mutants 2E1 and 5H3, were grown to equal densities, subjected to whole cell crosslinking with PFA for solid phase binding assays, or lysed for SDS-PAGE and α-TbpB western analysis to evaluate whole cell expression levels. (D) Table of unique transposon mutants lacking TbpB on their surfaces. Transposon mutants were sequenced and the ORF affected by transposon insertion was identified. From this list, nmb0313 was an obvious hit to pursue for further testing because it encoded a predicted outer membrane protein. (E) Confirmation of nmb0313 as the causative gene via complementation. nmb0313 was cloned into the expression vector pGCC4. Empty plasmid (Emp) and complementation plasmid (Compl) were transformed into the transposon mutant 5H3, and PFA and SDS solid phase binding assays were used to evaluate surface and total TbpB levels. (Adapted from Hooda and Lai et al. Nat. Microbiology 2016) [49]
Next, the effects of NMB0313 deletion was assessed on other SLPs [49]. Exposure of wild type N. meningitidis or the nmb0313 complemented strains to proteinase K led to degradation of the surface exposed TbpB, LbpB and fHbp, whereas these proteins remained intact within the Δnmb0313 strain due to their retention within the cell (Figure 1.6B). To quantify this effect at the single-cell level, antibodies to TbpB, LbpB and fHbp were used to probe wild type and Δnmb0313 strains of N. meningitidis by flow cytometry (Figure 1.6C-E). The Δnmb0313 strain lacked a display of each SLP on its surface similar to Δtbpb and Δlbpb (Figure 1.6C and D). Together, these data show that NMB0313 is required for N. meningitidis to efficiently display TbpB, fHbp and LbpB on the cell surface. Thus, the protein encoded by nmb0313 was named Slam for Surface Lipoprotein Assembly Modulator [49].
1.5.1 Role of Slam in N. meningitidis pathogenicity
To examine the effects of Slam within a mammalian infection, our lab collaborated with Gray- Owen lab at the University of Toronto, who introduced isogenic wild type or mutant N. meningitidis strains into a mouse sepsis model. While wild type meningococci caused severe sepsis, mice infected with either B16B6 Δnmb0313 or the positive control strain, Δtbpb, progressed well through to the endpoint of the experiment (Figure 1.6F). To confirm that bacterial fitness did not cause this reduction in infectivity for B16B6 Δnmb0313, in vitro growth rates were measured and showed no differences with the WT. While there was no statistical difference between Δnmb0313 and Δtbpb in terms of survival rates, we noticed a difference in their abilities to bind human transferrin, which may contribute to the small difference in mice survivability. Taken together, these results firmly
18
established the essential nature of Slam in the establishment of bacterial colonization and promotion of meningococcal disease.
Moraes - Figure 1 A B + Emp + Slam1
WT ∆nmb0313∆nmb0313∆nmb0313 PK - + - + - + - + - + - + B16B6 α TbpB
B16B6 WT nmb0313nmb0313/tn5 nmb0313/tn5 +∆ Empnmb0313/tn5 + Slam1 α LbpB
PFA B16B6 α FbpA α TbpB SDS MC58 α fHbp
MC58 α FbpA C D E α TbpB α LbpB α fHbp Count Count Count
PE Fluorescence PE Fluorescence PE Fluorescence B16B6 ∆tbpb B16B6 ∆lbpb MC58 ∆nmb0313 B16B6 ∆nmb0313 B16B6 ∆nmb0313 MC58 WT B16B6 WT B16B6 WT
α TbpB α LbpB α fHbp
**** ) ***
) *** %
( 150 %
( 150
e e
**** c *** c n n e e c c s s
r 100
r 100 o o u l u l f
f
n n a a e e 50 50 Relative Relative Relative m m
e e v v i i t t a a l l e e 0 0 R R t wt LbpB- Slam-
Mean Fluorescence (%) Mean Fluorescence w (%) Mean Fluorescence (%) Mean Fluorescence TbpB- Slam-tbpb lbpb
MC58 WT nmb0313 B16B6 WT nmb0313 B16B6 WT nmb0313 B16B6 ∆ B16B6 ∆
B16B6 ∆ B16B6 ∆ MC58 ∆
F 100 B16B6 WB16B6T WT B16B6 nmb0313 l 80 *** B16B6 ∆nmb0313 a v i v r 60 u s
t n
e 40 c r e P
Percent survival Percent 20
0 0 12 24 36 48 HHoursours p postost in infectionfection
Figure 1.6 Characterization of nmb0313 in Neisseria meningitidis. (A) A solid phase binding analysis of TbpB in cells treated with paraformaldehyde (PFA) or lysed with SDS. B16B6 wild type
19
(WT), NMB0313 transposon mutants (nmb0313/tn5), complemented strains with empty vector (Emp) or Slam1, and NMB0313 complete knockout cells (Δnmb0313) were examined as labeled. (B) Whole cell proteinase K (PK) surface shaving assays of B16B6 or MC58 WT, Δnmb0313 and complemented strains as indicated. Samples were analyzed by immunoblotting for TbpB, LbpB and fHbp or periplasmic FbpA. Representative blots of at least three independent experiments are shown in (A,B). (C-E) Flow cytometry profiles and quantification of WT (red) or Δnmb0313 (blue) cells for TbpB, LbpB or fHbp surface display. For TbpB and LbpB surface profiles, Δtbpb (black) and Δlbpb (black) strains are shown as controls. Relative mean fluorescence was calculated by scaling the averaged mean fluorescence intensity of the WT to 100 and applying the same scaling factor to the other samples. Error bars represent the standard error of the mean (SEM) from three biological replicates. Statistical significance was determined by ANOVA, ***p≤0.001; ****p≤0.0001. (F) Survival curve of mice following intraperitoneal injection with 1x106 CFU of B16B6 WT or Δnmb0313. Mice (n=11, combined from two independent experiments) were monitored for survival every 12 hr starting 48 hr pre-infection to 48 hr post-infection and additionally monitored at 3 hr post-infection. Statistical significance was based on the Mantel-Cox log rank test, ***p≤0.001. (Adapted from Hooda and Lai et al. Nat. Microbiology 2016) [49]
1.5.2 Role of beta-barrel domain of Slam in translocation of TbpB
To examine the regions within Slam that were required for SLP surface display, we initially used PSIPRED to define secondary structure elements and domain architecture. Further analysis using SignalP [112], XtalPred [113], InterProScan [114] and Boctopus [115] predicted that Slam encodes an outer membrane protein that consists of a 31 residue signal peptide and a 457 amino acid (52.6 kDa) mature protein with two domains: a soluble N-terminal domain (Ntd) containing two tetratricopeptide (TPR, residues 118-151, 152-185) repeats and a C-terminal b-barrel domain (residues 204-488) consisting of 14 outer membrane-spanning strands annotated as a DUF560 (Figure 1.7A).
To examine the role of each Slam domain, the b-barrel and N-terminal domains were individually expressed in Δnmb0313 strains of N. meningitidis B16B6 and MC58, and the SLP surface display was examined by proteinase K digestion and flow cytometry (Figure 1.7B-D) [49]. The b-barrel domain of Slam is sufficient to rescue TbpB, LbpB and fHbp surface translocation when expressed in N. meningitidis, whereas the N-terminal TPR-containing domain was unable to complement the SLP translocation deficiency of the knockout strain. By flow cytometry, it was confirmed that the β-barrel domain is able to confer approximately half of TbpB translocation activity (Figure 1.7C) and equivalent levels of fHbp surface display (Figure 1.7D) compared to wild type strains. These results suggest that the b-barrel domain is critical for SLP translocation while the TPR-containing
20
N-terminal domain may have an ancillary role in SLP translocation by stabilizing the SLP substrate and/or acting as a plug for the barrel domain. Moraes- Figure 2
A B el td r SP N-terminal domain DUF560 Slam1 + Bar (Ntd) (Barrel) 1 31 203 488 nmb0313 + Emp SP N-terminal domain Ntd ∆ ∆nmb0313 + Slam1∆nmb0313+ N∆nmb0313 (Ntd) 1 31 203 PK - + - + - + - + SP DUF560 (203-488) Barrel (Barrel) 1 31 315 B16B6 α TbpB MVIFYFCGKTFMPARNRWMLLLPLLASAAYAEETPREPDLRSRPEFRLHE 50 AEVKPIDREKVPGQVREKGKVLQIDGETLLKNPELLSRAMYSAVVSNNIA 100 B16B6 α LbpB GIRVILPIYLQQAQQDKMLALYAQGILAQADGRVKEAISHYRELIAAQPD 150 APAVRMRLAAALFENRQNEAAADQFDRLKAENLPPQLMEQVELYRKALRE 200 B16B6 α FbpA RDAWKVNGGFSVTREHNINQAPKRQQYGKWTFPKQVDGTAVNYRLGAEKK 250 WSLKNGWYTTAGGDVSGRVYPGNKKFNDMTAGVSGGIGFADRRKDAGLAV α fHbp 300 MC58 FHERRTYGNDAYSYTNGARLYFNRWQTPKWQTLSSAEWGRLKNTRRARSD 350 NTHLQISNSLVFYRNARQYWMGGLDFYRERNPADRGDNFNRYGLRFAWGQ MC58 α FbpA 400 EWGGSGLSSLLRLGAAKRHYEKPGFFSGFKGERRRDKELNTSLSLWHRAL 450 HFKGITPRLTLSHRETRSNDVFNEYEKNRAFVEFNKTF 488
C D *** n.s. α TbpB α fHbp * *** n.s. e (%) e (%) ***
en c * e en c e ti v ti v es c es c el a el a R luo r R luo r ean F ean F M M T td rel T td rel W W
B16B6 MC58 nmb0313 + Emp ∆nmb0313 + Emp ∆nmb0313 + N ∆ nmb0313 + Slam1∆nmb0313 + N ∆nmb0313 + Slam1 ∆nmb0313 + Bar ∆ ∆nmb0313 + Bar
Figure 1.7 Role of beta-barrel domain on Slam translocation. (A) Structural features of Slam1. Slam1 contains a signal peptide (SP), N-terminal domain (Ntd) and C-terminal DUF560 domain (β- barrel). Predicted helices are represented by cylinders in the Ntd and strands are represented by arrows in the β-barrel domain. (B) Whole cell PK surface shaving assays of B16B6 or MC58 strains complemented with empty vector (Emp), full length (Slam1) or domains (Ntd or Barrel) of Slam1. Samples were analyzed for TbpB, LbpB and fHbp or periplasmic localized FbpA. One representative blot of at least three independent experiments is shown. (C-D) Quantification of TbpB or fHbp surface display in WT and complemented strains, as determined by flow cytometry. Relative mean fluorescence was calculated by scaling the averaged mean fluorescence intensity of the WT to 100 and applying the same scaling factor to the other samples. Error bars represent the SEM from three biological replicates. Statistical significance was determined by ANOVA, *p<0.05;
21
***p<0.001; ns: not significant. (Adapted from Hooda and Lai et al. Nat. Microbiology 2016) [49]
1.5.3 E. coli translocation assay
To decipher the specificity and mechanism of Slam translocation, the cell surface display of neisserial SLPs was examined in laboratory strains of E. coli (BL21–C43(DE3)). Surprisingly, the simple addition of Slam1 (referred to as 5H3 in Judd et al. Msc. Thesis, University of Toronto, 2015 [116]) to E. coli C43 cells co-expressing TbpB, LbpB and fHbp facilitated their display on the bacterial surface, demonstrating a “gain of function” phenotype in E. coli cells (Figure 1.8A-C) [116]. Importantly, the E. coli surface displayed TbpB was able to recognize human transferrin suggesting that these translocated SLPs are fully functional [49]. To ensure that this translocation activity was not due to aberrant expression of SLPs in the presence of Slams, western blots were performed and found no correlation between surface display and expression (Figure 1.8A-C).
Figure 1.8 Reconstitution of Slam-dependent SLP translocation in E. coli. The surface expression of each SLP, TbpB with antibody (A), LbpB with antibody (B), and Fhbp with antibody (C), is shown by the increase in phycoerythrin binding when SLPs are expressed with 5H3 (Slam1). Background fluorescence was measured using cells expressing 5H3 (Slam1) alone. Cells expressing
22
5H3 alone were used to measure background levels of fluorescence. Any cells exhibiting fluorescence above this background level were considered to be displaying an SLP on the surface. The percentages of cells able to do so are plotted above. Representative histograms for each experiment are shown below the bar graphs. Solid black lines represent a sample with both SLP and 5H3 (Slam1) expressed, while grey lines display that of a sample expressing the SLP alone. Grey shaded areas represent samples expressing only 5H3 (Slam1) that were stained with the corresponding antibody to act as a background control. FL2 was the detector used to measure phycoerythrin fluorescence. Western blots were performed for each SLP to confirm expression. (Adapted from Judd et al. Msc. Thesis, University of Toronto, 2015) [116]
1.6 Overview of the thesis
The previous work done in our lab on the discovery and initial characterization of Slam set the groundwork for my PhD thesis. While the discovery of Slam provides insight into the final step of the translocation, there are a number of outstanding questions that remain:
1. How prevalent are Slams in other Gram-negative bacteria? 2. What is the Slam-specific secretion motif present on SLPs? 3. Does Slam function alone or does it require other factors? 4. Is there a role for periplasmic factors in the movement of SLPs across the periplasm? 5. What regions within Slam are important for SLP translocation?
The tools available in our lab provided a great system to answer these questions and in the following chapters, I will describe in detail the progress I have made thus far. In Chapter 2, I will describe the bioinformatics analysis performed to identify Slam homologs in other bacterial species. Selected potential homologs were then rapidly tested using the E. coli translocation assay. The assay also provides a great tool for answering the second question as mutations and truncations of TbpB can be rapidly screened for surface display and translocation efficiency. Hence in Chapter 3, I will describe the work done to narrow down the Slam-specific secretion motif on SLPs. I also used this assay to probe various aspects of SLP translocation and show that the translocation occurs from the C-terminus to the N-terminus.
23
Figure 1.9 Potential models for Slam function. Currently, the mechanism of Slam function is not known. We propose two models for Slam function: I) Chaperone model where the Slam delivers the SLPs to the Bam complex for their surface display and II) Translocon model where the membrane domain of Slam acts a conduit for the movement of the SLP across the OM. (Adapted from Hooda et al. FEMS Pathog. Dis. 2017) [53]
Based on the literature, we can propose two models for Slam translocation (Figure 1.9): 1) A chaperone model where Slam delivers the SLP to an outer membrane translocon such as the Bam complex, or 2) Translocon model where Slam acts as the conduit for the movement of SLPs across the outer membrane. Previous studies have used an in vitro liposome assay to reconstitute the transport of proteins across membrane [160, 161]. No such assay existed for the study of SLPs and hence I decided to develop an assay for studying Slam-dependent SLP translocation. However, the central roadblock towards this assay is the requirement of purified Slam protein. In Chapter 4, I discuss work done to optimize the production of Slam protein for biophysical studies. With purified Slam protein, we were able to develop an in vitro assay to examine SLP translocation. The work done towards this aim is also described in Chapter 4 and using these studies, I have shown that Slam is both necessary and sufficient for the SLP translocation.
Taken together, this study has established that Slam in an outer membrane translocon that is both necessary and sufficient for the movement of SLPs across the OM. Slam specifically recognizes SLPs of a defined structural fold composed of a soluble 8-stranded barrel domain on the C- terminus. Genes encoding Slams are often present adjacent to genes encoding its putative substrates
24
further highlighting a genetic link between the substrate and translocon. The translocation occurs from the C-terminus to the N-terminus and Slam function can be reconstituted in the model organism E. coli as well as liposomes. In Chapter 5, I will compare the Slam mechanism to other known translocons, Further, I will discuss other outstanding questions and future experiments that will help in further uncovering various aspects of this fascinating family of proteins.
25
2 Chapter 2 Prevalence of Slam-dependent SLP translocation in Gram-negative bacteria
Acknowledgements: This chapter is adapted from published manuscripts: Hooda et al. Front Cell Infect Microbiol 2017 and Hooda and Lai et al. Nat Microbiol 2016. Ms. Christine Lai, Moraes lab manager for providing plasmids with the Slam and SLP constructs used in the study. pETDUET HpuA was obtained from Ms. Anastasia Bosc, former undergraduate student, Moraes lab.
2.1 Overview
In Chapter 1, I outlined the various systems known for translocation of peripheral membrane proteins found on the surface of Gram-negative bacteria referred to as surface lipoproteins or SLPs. However, these systems are only for a subset of known SLPs (which itself is a small percentage of all SLPs). A central question in the field is the mechanism by which SLPs are transported across the outer membrane. The discovery and initial characterization of Slam in Neisseria meningitidis highlighted that Slam is a unique system for the translocation of SLPs and insights into its mechanism would shed light on this molecular process.
In this chapter, I will outline the bioinformatics analysis performed for the identification Slam-like proteins in Gram-negative bacteria. Using this analysis, I was able to identify ~ 800 Slam-like proteins throughout the phylum of Proteobacteria. Many bacterial species, genes encoding transporter and substrate pairs are adjacent to each other. Based on this observation, we searched and annotated genes upstream and downstream of 353 Slam homologs found in proteobacterial species. This dataset showed that a large number of Slam related sequences are located adjacent to genes that encode putative lipoproteins with TbpB-like folds suggesting a genetic linkage between these two families of proteins. Taken together, the analysis highlighted that Slams and their SLP pairs form a “plug-and-play” play system for delivery of effectors to the surface of Gram-negative bacteria.
The bioinformatics analysis presented here also provided a computational approach towards the identification of putative SLPs. I was able identify TbpB-like SLPs in many bacterial species that
26
were previously not known to possess SLPs, including the human pathogens Acinetobacter baumannii and Salmonella enterica subsp. arizonae. Although limited to Slam-dependent SLPs, similar studies can be used to accurately predict SLPs present in different Gram-negative bacteria. This provides further evidence to the recent articles that have hypothesized that most Gram- negative bacteria, if not all, possess SLPs [2,34,35].
2.2 Methods
2.2.1 Identification of Slam homologs
To generate a database of Slam related sequences, iterative psi-blast searches were performed (March 4, 2016) against a non-redundant database containing all partial and complete bacterial genome sequences using the sequence of Slam1 protein (NMB0313) from Neisseria meningitidis strain MC58 as the query. Four independent psi-blast searches were performed for different clades of proteobacteria (alpha-, beta-, gamma- and epsilon/zeta-proteobacteria). The lists of putative Slam genes obtained from these four psi-blast searches were pooled and only unique representative Slam sequences were kept from a given bacterial species. The list was manually checked to remove the following: (i) partial sequences (containing premature stop codons or with partial gene sequence coverage), and (ii) sequences coding for only the N-terminal domain (Ntd) of Slam. This gave a final list (Appendix A) of 832 Slam sequences from 638 bacterial species.
To understand the distribution of Slam related sequences, a phylogenetic tree of different proteobacterial species was made using the 16S-RNA sequences obtained from the database Greengenes [117]. One representative member was kept from each family of bacteria. In total 52 species (8: alpha-, 10: beta-, 23: gamma-, 5: delta-, 5: epsilon- and 1: zeta-proteobacteria) were selected for the final tree. The tree was made using PhyML plugin in the software Geneious [118] with 100 bootstraps. The nodes were kept if they appeared in 60% of the bootstrap runs. The presence of Slams related sequences was mapped on the phylogenetic tree.
2.2.2 Analysis of gene neighbourhoods around putative Slam homologs
The list of 353 Slam related sequences generated in our previous study [119] was used to further investigate the neighbouring genes in the Slam gene clusters. This number is more than a third of
27
the total Slam related sequences and covers all major bacterial phyla that possess Slam related sequences, except for epsilon- and zeta-proteobacteria (Appendix A). In selecting genomes for a given bacterial species, fully sequenced reference genomes were given preference. For each of the Slam related genes present in these species, the corresponding genomic record (NCBI genome) was used to identify genes upstream and downstream along with their corresponding functional annotations (NCBI protein database, Ensembl bacteria). In a few of the cases, no genes were predicted upstream or downstream as the Slam related genes were close to the beginning or the end of the contig respectively and these sequences were ignored.
Within the Slam-related gene clusters, a number of the neighbouring genes were predicted to encode lipoproteins (predicted by an N-terminal lipobox motif using LipoP and/or SignalP) and we also found many examples of genes encoding TonB dependent transporters (IPR000531). The Slam-adjacent proteins contained were curated if they contained either GNA1870-related lipoproteins, TBP-like solute-binding proteins or pagP-beta barrel protein (InterPro signature; IPR01490; IPR001677; IPR011250 respectively). All the genes with one of the above-mentioned annotations are included in Appendix B.
2.2.3 Bacterial strains and growth condition
Strains used in this study are summarized in Appendix C. E. coli were grown in LB media containing antibiotics when necessary (50 µg/mL kanamycin and 100 µg/mL ampicillin). Cloning procedures were carried out using E. coli MM294 competent cells. Protein expression was performed using E. coli C43 (DE3) cells for all the flow-cytometry and western blot analysis.
2.2.4 Generation of plasmids for expression of Slams and SLPs
For flow cytometry experiments, the four SLPs (H. influenzae TbpB, M. catarrhalis TbpB and Pasteurella multocida PM1514) were cloned into pET52b (to make pET52b Hinf TbpB, Mcat TbpB or Pmul SLP) using the restriction-free (RF) cloning strategy by Ms. Christine Lai [120]. pETDUET Nmen HpuA was obtained from Ms. Anastasia Bosc. The tbpb genes were amplified from the genomes of H. influenzae strain 86-028NP and M. catarrhalis strain O35E, and the pm1514 gene was amplified from P. multocida strain h48. A FLAG tag was inserted on the C- terminus of Mcat tbpb using FastCloning [121] to make pET52b Mcat TbpB-flag. pET52b PmSLP-
28
flag and pET52b PmSLP-flag-Slam was cloned by replacing the Mcat tbpb gene with pm1514 and pm1515-pm1514 respectively in frame with the FLAG tag using RF cloning by Ms. Christine Lai.
The corresponding Slams were inserted into pET26b (pET26 Hinf Slam1, Mcat Slam1 and Pmul Slam) using RF cloning [120] by Ms. Christine Lai. A 6xHis-tag was inserted between the pelB and the mature Slam sequences.
2.2.5 Flow cytometry
For the E. coli translocation assays, the display of an SLP was determined using flow cytometry. Pairs of SLP and Slam plasmids (shown in Appendix C were transformed into C43 (DE3) cells and grown in 1 mL of auto-induction media [122] for 18 hours at 37oC. H. inf Slam showed poor expression when grown overnight in autoinduction media. Hence, for H. inf TbpB flow cytometry o assays were performed by growing cells at 37 C to an OD600 ~0.6 and then inducing protein expression by the addition of 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG). Upon induction, cells were grown at 18 oC for 16-18 hrs. Cells were harvested, washed twice in PBS containing 1 mM MgCl2, and incubated with α-Flag antibodies (1:200, Sigma), or biotinylated human transferrin (0.05 mg/ml, Sigma) for 1 hr at 4 oC. The cells were then washed twice with PBS containing 1 mM MgCl2 and then labelled with R-phycoerythrin (R-PE) conjugated Streptavidin (0.5 mg/ml, Cedarlane) or R-PE conjugated α-mouse IgG (25 µg/mL, Thermo Fisher Scientific) for 1 hr at 4 oC. Following staining, cells were fixed in 2% formaldehyde for 20 minutes and further washed with PBS containing 1 mM MgCl2. Flow cytometry was performed with a Becton Dickinson FACSCalibur and the results were analyzed using FLOWJO software. Mean fluorescence intensity (MFI) from at least three biological replicates were used to compare surface exposure of a given SLP between different samples. Statistical significance was calculated by comparing MFI between different samples using the one-way ANOVA test available in the software Prism 6.
Western blots were used to test the expression levels of each of the constructs used for the flow cytometry experiments. α-Flag (1:5000, Sigma) and α-His (1:5000, Thermo Fisher Scientific) antibodies were used to test expression of the SLP and Slam constructs respectively. α-GroEL (1:10,000) antibodies were used as loading controls.
29
2.2.6 Sucrose density ultracentrifugation
E. coli C43(DE3) cells expressing pET52b PmSLP-flag and pET26b empty or pET26 PmSlam were grown overnight and then used to inoculate 50 ml LB with the appropriate antibiotic. The cells were grown to an OD600 ~ 0.6, induced with 1 mM IPTG and then grown for an additional 18 hrs. The cells were pelleted, resuspended in 20 mM Tris pH 8.0, 200 mM NaCl with fresh lysozyme (1 mg/ml), 2 mM PMSF and DNase I (0.05 mg/ml), lysed by sonication and then centrifuged at 10,000 r.c.f. to remove cell debris. The supernatant was centrifuged at 125,000 r.c.f. for 1 hr to collect the cellular membranes. The membrane pellet was resuspended in 1 ml of 20 mM Tris pH 8.0, 200 mM NaCl using a micro-glass homogenizer.
The inner and the outer membrane of E. coli were separated using a modified sucrose density ultracentrifugation protocol that was previously described [119]. For this assay, 100 µl of the membrane pellet was applied on top of a 13.2 ml thin-wall polypropylene tube containing step gradients of 3 ml of 2.02 M, 6 ml of 1.44 M and 3 ml of 0.77 M sucrose. The tubes were centrifuged at 83,000 r.c.f. for 16 h. The outer membrane and inner membranes partitioned to the interface of the 2 M and 1.44 M sucrose cushions and 1.44 M and 0.77 M sucrose layers, respectively. Twelve 1 ml fractions were collected and subjected to SDS–PAGE followed by western blotting with α-Flag (1:10,000), α-LepB (1:10,000) and α-OmpA (1:40,000) antibodies.
2.3 Results
2.3.1 Identification of putative Slam homologs in Gram-negative bacteria
To identify putative Slam homologs, I performed updated psi-blast searches and was able to identify 832 Slam related sequences in 638 Gram-negative bacteria (Appendix A). The Slam1 gene (nmb0313) was used as the search template and I manually analyzed the list to remove genes for which only partial sequences were available. I also removed several hits that contained a large single domain with TPR repeats that are similar to the TPR repeats found in Slam-Ntd. All Slam- like sequences obtained in our dataset contained both the Ntd and DUF560 domains (Figure 2.1A) and no sequences containing only the DUF560 were obtained.
30
Figure 2.1: Putative Slam family of proteins in Gram negative bacteria. (A) Domain architecture of N. meningitidis Slam1. Slams possess two domains: a periplasmic N-terminal domain (Ntd) containing tetratricopeptide repeats and a predicted membrane bound 14-stranded barrel domain referred to as DUF560. (B) Distribution of Slam related sequences in Proteobacteria. A family tree of Proteobacteria was made with PhyML with 100 bootstraps using 16S-RNA sequences from 52 species representing the major bacterial families within Proteobacteria. The families containing at least one species with a Slam related sequence containing both the Ntd and DUF560 domains are highlighted by black circles. The size of the circle represents the number of
31
Slams identified in a given bacterial family. Slams were found within all clades of Protebacteria. (Adapted from Hooda et al. Front Cell Infect Microbiol 2017)
With these additional genomic sequences, I was able to identify Slam related sequences in all clades of the phylum Proteobacteria (Figure 2.1B). No Slam related sequence was found in other phylums including Sphirochetes, Bacteroides etc. that are known to possess SLPs. Slam related sequences were identified in bacterial species living in diverse environments including free-living (such as Desulfovibrio and many Pseudomonas), commensal (several Neisseria species) and/or pathogenic bacteria. Slam-like proteins were found in many human pathogens such as Vibrio cholerae, Salmonella enterica subsp. arizonae and Acinetobacter baumannii. While many species have reported to contain SLPs such as Haemophilus influenzae and Moraxella catarrhalis, many of the species such as Salmonella enterica subsp. arizonae and Acinetobacter baumannii were previously not known to possess any SLPs. This suggests that SLPs are more widespread and many of these bacterial species may also possess SLPs that are yet to be identified.
2.3.2 Identification and functional characterization of Slam-like proteins in Neisseria
A large number of Slam-related sequences were found in the Neisseria. Slam related sequences were identified in all sequenced Neisseria species to date (Figure 2.2). In N. meningitidis two Slams were identified, while in N. gonorrhoeae, three Slam homologs were found and in N. lactamica four Slams were predicted. While one Slam homolog in N. gonorrhoeae and N. lactamica shared high sequence similarity (>70% sequence identity) to Slam1 of N. meningitidis, other Slams were different from Slam1 (~ 20-30% sequence identity). The putative N. meningitidis Slam paralog could also be identified in N. gonorrhoeae and N. lactamica. Interestingly, the third Slam paralog in N. gonorrhoeae was present in N. lactamica but not N. meningitidis. Finally, N. lactamica contained one additional unique putative Slam paralog. Taken together, these findings suggest that different Slam paralogs might transport different SLPs to the surface. Interestingly, many commensal neisserial species such as N. bacilliformis and N. elongata possess multiple Slam related sequences but no Slam dependent SLPs have been reported in these organisms. This suggests that many Slam-dependent SLPs in Neisseria have yet to be identified. Even in pathogenic Neisseria sp. (N. meningitidis and N. gonorrhoeae), existence of multiple Slams indicates that not all Slam- dependent SLPs have been identified in these species.
32
Slam TbpB LbpB HpuA fHbp NHBA Ng-MIPAniA NalP
(2)
(4)
(5)
(2)
(3)
(2)
(3)
(4)
(2)
(2)
(4)
(4)
Figure 2.2: Slam and SLP homologs in Neisseria species. A phylogenetic tree of sequenced Neisseria genomes is shown using rplF sequences as described previously by Weyand et al. 2016 (the tree was generated using PhyML with 100 bootstraps). The number of Slam homologs found in each species is shown in parenthesis next to the species name. In each neisserial species the presence of a homolog of a known neisserial SLPs is denoted Ö: SLP homolog found, : SLP homolog not found. SLPs were identified using blastp with the Nme or Ngo SLP as query sequences. No Slams or SLPs could be identified in three neisserial species as their genomes are not currently available in the NCBI. (Adapted from Hooda et al. FEMS Pathog. Dis. 2017)
2.3.3 Identification and characterization of Slam2 in N. meningitidis
As mentioned previously, we discovered an additional Slam homolog in N. meningitidis. The second Slam paralog (nmb1971) also contains a DUF560 and a TPR domain (Figure 2.3A), and is located directly upstream of another known SLP, HpuA (Figure 2.3B). This co-localization is conserved in multiple Neisseria species. Hence, we refer to NMB1971 as Slam2 and NMB0313 as Slam1.
33
Figure 2.3: Slam2 characterization and identification of its substrate, HpuA. (A) Domain comparisons of Slam1 and Slam2. Slam2 was identified from psi-blast as a paralog of Slam1 in N. meningitidis. Slam1 and Slam2 share 26% overall sequence identity but have the same domain architecture: N-terminal soluble domain containing TPR repeats (Ntd) and C-terminal membrane spanning β-barrel domain (DUF560). (B) Gene neighbor analysis of Slam2. The gene encoding Slam2, nmb1971, is adjacent to HpuA and HpuB genes. Further, blastp searches for Slam2 homologs in other neisserial species reveal that Slam2 is found adjacent to HpuA and HpuB-like genes. The percentage sequence identity of each homolog to its N. meningitidis counterpart is shown. This suggests a co-evolutionary link between HpuA and Slam2 and that HpuA is the cognate substrate for Slam2. (C) Quantification of surface display of HpuA in E.coli, as analyzed by flow cytometry. E.coli was transformed as labeled and stained with α-HpuA. Mean fluorescence intensity is shown on the Y-axis. Error bars represent the standard error of the mean (SEM) from three experiments. Western blot using α-HpuA antibodies showing that the difference in surface exposure of HpuA in the presence of Slam2 and other samples is not due to HpuA expression. (Adapted from Hooda and Lai et al. Nat. Microbiology 2016)
To confirm that HpuA depends on its adjacent Slam2 for translocation to the cell surface, we introduced the gene encoding HpuA in N. meningitidis in laboratory strains of E. coli with and without its neighbouring Slam2. I labelled the cells using a-HpuA antibodies. Flow cytometry was used to quantify the amount of N. meningitidis HpuA on the surface of the cell. An increase in N. meningitidis HpuA was observed only in the presence of its neighbouring Slam (Slam2) and not Slam1, as quantified using mean fluorescence intensity (MFI) (Figure 2.3C, upper panel). Western
34
blots with a-HpuA were used to test the expression of N. meningitidis HpuA (Figure 2.3C, lower panel). N. meningitidis HpuA were robustly displayed on the surface of E. coli in the presence of Slam2 confirming that Slam2 is a bonafide Slam paralog.
2.3.4 Slams adjacent to TbpB in M. catarrhalis and H. influenzae translocate their respective TbpBs to the surface in E. coli
A number of potential Slam homologs identified in this study were found in bacterial species that colonize the upper respiratory tract of mammals. Human respiratory tract bacteria that contain Slam sequences include N. meningitidis, M. catarrhalis and the HACEK (Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, Kingella) group of bacteria [123]. Slam sequences were also identified in bacteria that colonize the upper respiratory tract of cattle (Moraxella bovis, Mannheimia haemolytica and Histophilus somni) and pigs (Actinobacillus pleuropneumoniae). Many of these species have been reported to contain transferrin-binding surface lipoproteins that are homologs of TbpB in N. meningitidis [124]. Not surprisingly, two TbpB homologs in Moraxella catarrhalis [125] and Haemophilus influenzae [126] were found to be adjacent to putative Slam genes (Figure 2.4A). Both M. catarrhalis and H. influenzae are human pathogens and their TbpBs have been previously shown to bind to human transferrin. The presence of TbpB genes adjacent to a Slam gene in their genome strongly suggests that these bacteria use a Slam-dependent translocation system to deliver TbpBs to the bacterial cell surface.
To confirm that these TbpBs depend on their adjacent Slam for translocation to the cell surface, we introduced the gene encoding TbpB in M. catarrhalis in laboratory strains of E. coli with and without its neighbouring Slam. To monitor the expression of M. catarrhalis TbpB and Slam, we introduced a flag-tag at the C-terminus of TbpB and 6xHis tag on the N-terminus of Slam respectively. We labelled the cells using with biotinylated human transferrin and a-Flag antibodies (Figure 2.4B). Flow cytometry was used to quantify the amount of M. catarrhalis TbpB on the surface of the cell. An increase in M. catarrhalis TbpB was observed only in the presence of its neighbouring Slam and was quantified using mean fluorescence intensity (MFI) (Figure 2.4C). Western blots with a-Flag & a-His antibodies were used to test the expression of M. catarrhalis TbpB & Slam respectively (Figure 2.4C, lower panel). M. catarrhalis TbpB were robustly displayed on the surface of E. coli in the presence of Slam confirming that this TbpB homolog also uses a Slam-dependent system to reach the cell surface.
35
Figure 2.4: Translocation assay with Slam and TbpB pairs from Moraxella catarrhalis and Haemophilus influenzae. (A) Slam and TbpB gene cluster in M. catarrhalis and H. influenzae. From the bioinformatics analysis, Slam genes were found adjacent to the known transferrin binding surface lipoprotein TbpB in the human pathogens M. catarrhalis and H. influenzae. (B) E. coli translocation assay used in this study. The Slam (with a N-terminal 6xHis tag) and TbpB genes (with a C-terminal flag-tag) were expressed in E. coli C43 (DE3) cells. The cells were labelled with either biotinylated human transferrin and streptavidin linked to R-phycoerthyrin (PE) or mouse a- flag antibody and a-mouse linked to R-phycoerthyrin (PE). Surface display of TbpB was quantified using flow cytometry. (C) Flow cytometry profile of M. catarrhalis TbpB-flag obtained with Slam (shown in red) or without Slam (shown in blue) is shown using a-flag (left panel) and biotinylated human transferrin (right panel). An increase in fluorescence intensity was observed in the presence of Slam for both a-flag and biotinylated human transferrin suggesting that the flag does not affect translocation or function of surface-displayed TbpB. Surface TbpB was quantified using mean
36
fluorescence intensity. (D) Flow cytometry profiles of TbpB homologs. A scatter plot of cell counts (Counts) vs PE fluorescence (FL2-H) for M. catarrhalis TbpB (McTbpB) and H. influenzae TbpB (HiTbpB) in the presence (shown in red) or absence (shown in blue) of their cognate Slam is illustrated. A higher PE fluorescence signal was observed in the presence of Slam. (E) Mean fluorescence intensity blots for TbpB homologs. Surface TbpB was quantified using mean fluorescence intensity. Statistical significance was determined using one-way ANOVA. *** p£0.001. (Adapted from Hooda et al. Front Cell Infect Microbiol 2017)
To verify that the flag-tag did not affect the Slam-dependent display of M. catarrhalis TbpB, I also tested the surface display of M. catarrhalis TbpB without the flag-tag in laboratory strains of E. coli. I labelled the cells using biotinylated human transferrin to test the functional display of the TbpB. Flow cytometry was used to quantify the amount of TbpB located on the surface of the cell. Similar to the results obtained for the flag-tagged construct and an increase in signal was obtained only in the presence of the neighbouring Slam gene suggesting that the flag-tag does not affect translocation (Figure 2.4D). Additionally, I also tested the surface display of H. influenzae TbpB with or without its neighbouring Slam using biotinylated human transferrin. In this case, a signal increase was obtained on the surface of E. coli in the presence of Slam, however the signal was weaker compared to M. catarrhalis TbpB. Westerns blot analysis using a-His antibody suggested that M. catarrhalis Slam (Figure 2.4D, lower panel) is expressed much more strongly in comparison to H. influenzae Slam (Figure 2.4E, lower panel), which may contribute to the lower signal obtained for H. influenzae TbpB.
2.3.5 Predicted SLP genes are found adjacent to Slam genes in a number of Gram-negative bacteria
For 353 of the 832 Slam related genes identified, a list of neighbouring genes was analyzed using InterProScan [127] to identify Slam related gene clusters that also contain lipoprotein-encoding genes. From our analysis, 185 of the 353 (~52% of the clusters examined) Slam related genes contained lipoprotein-encoding genes in their gene clusters. This list of putative lipoproteins contained three experimentally confirmed SLPs (Figure 2.5A). These include two HpuA homologs in Neisseria gonorrhoeae and Kingella denitrificans that were functionally and structurally characterized by Wong et al. [83]. I also identified a putative human factor H binding protein from H. influenzae, referred to as protein H (PH) [128] that has sequence homology to the Slam- dependent SLPs in N. meningitidis.
37
Figure 2.5: Slam related gene clusters identified in this study with known lipoproteins. (A) Three SLP-containing Slam gene clusters were found in Neisseria gonorrhoeae, Kingella denitrificans and Haemophilus influenzae. (B) Lipoprotein-containing Slam gene cluster in Xenorhabdus nematophilia. (Adapted from Hooda et al. Front Cell Infect Microbiol 2017)
The 185 predicted lipoproteins were found in 129 different species distributed throughout Proteobacteria. One such lipoprotein was identified in the nematode pathogen Xenorhabdus nematophila that has not been previously shown to contain SLPs (Figure 2.5B). This lipoprotein, named NilC has been previously shown to be lipidated in vivo, present in the outer membrane and is important in host colonization [129]. The gene encoding NilB, a putative Slam homolog, is present next to the gene encoding NilC, and is shown to be required for host colonization [130]. Our analysis suggests that NilC is a surface lipoprotein (SLP) that is dependent on the Slam homolog, NilB, for surface display.
38
Studying other genes in the Slam related gene clusters, I also uncovered that 120 of the 185 clusters also possessed genes that encode proteins annotated as TonB-dependent receptors (TonBDR). This is not surprising as in Neisseria sp., three Slam-dependent SLPs (TbpB, LbpB and HpuA) work in conjunction with a TonBDR to acquire iron [53]. The presence of TonBDRs gene in proximity to these putative SLPs supports their potential roles in nutrient acquisition. Furthermore, a small subset of Slam gene clusters (nine) contained multiple lipoprotein-encoding genes suggesting that they may be responsible for the display of multiple target SLPs.
2.3.6 A putative SLP gene in Pasteurella multocida is displayed on the surface of E. coli in a Slam-dependent manner
To further confirm the hypothesis that I have identified a large family of Slam-dependent SLPs, I sought to characterize the surface display of some of the remaining lipoproteins that were identified in this study for which no other functional data could be found. One such predicted lipoprotein was found in Pasteurella multocida, a zoonotic pathogen that resides in the normal respiratory microbiota of mammals. I identified the gene of the putative Slam (PM1515) adjacent to the predicted lipoprotein (PM1514) in all the sequenced strains of P. multocida (Figure 2.6A). The Slam displayed 32% identity to N. meningitidis Slam1 while the putative SLP showed no sequence similarity to any of the known Slam-dependent neisserial SLPs. Interestingly, the Slam gene cluster also included two other SLPs, PM1517 (PlpD) and PM1518 (PlpE) that have been investigated as potential vaccine antigens against Pasteurella infections [131,132].
To test if this pm1514 gene encoded a Slam-dependent SLP, we cloned the predicted Slam and lipoprotein genes into E. coli expression vectors (Figure 2.6B). I transformed E. coli cells with flag- tagged P. multocida lipoprotein and 6xHis tagged Slam and used the flag-epitope to detect the SLP on the surface. As predicted, I saw an increase in flag signal upon expression of Slam as seen in flow cytometry profiles (Figure 2.6C) and quantified in mean fluorescence intensity (MFI) plots (Figure 2.6D). Furthermore, to confirm the translocation pathway used by the P. multocida SLP, I illustrated its outer membrane localization in the presence or absence of Slam by sucrose density ultracentrifugation (Figure 2.6E). Collectively, these findings suggest that PM1514 is a putative Slam-dependent SLP and that many other Slam-dependent SLPs on our list (Appendix B) likely use a similar translocation pathway as N. meningitidis TbpB to reach the surface [119].
39
Figure 2.6: Identification of Slam-dependent surface lipoprotein in Pasteurella multocida. (A) Slam gene cluster in P. multocida strain Pm70. PM1515 (shown in blue) was identified as a Slam homolog in our bioinformatics search. PM1514 (shown in green) was annotated as a hypothetical protein with a predicted signal peptidase II cleavage site ending with a putative lipobox motif. (B) P. multocida constructs made for the translocation assay. To investigate if PM1514 is a Slam-
40
dependent SLP, we cloned the following: PM1514 with a C-terminal flag-tag (PM1514-flag), PM1515 with an N-terminal His-tag and pelB signal sequence, and PM1515-PM1514-flag construct with both PM1515 and PM1514 region. (C) Flow cytometry profiles of P. multocida constructs. The three constructs were expressed in E. coli C43 (DE3) cells and labelled with a-flag antibody and a mouse secondary antibody linked to R-phycoerthyrin (PE). Flow cytometry profiles of PM1514-flag (blue), PM1515+PM1514-flag (red) and PM1515-PM1514-flag (green) are shown. (D) Mean fluorescence intensity plots for P. multocida constructs. Surface PM1514 was quantified using mean fluorescence intensity (MFI). Statistical significance was determined using one-way ANOVA. *** p£0.001. (E) Localization of P. multocida SLP using sucrose density ultracentrifugation. To test the localization of P. multocida SLP in E. coli, cells expressing PM1514-flag with or without PM1515 were harvested. Cell membranes were then isolated and layered on a sucrose gradient to separate the inner and outer membrane. Westerns blots were performed on different fractions with α-Flag, α-LepB and α-OmpA antibodies to detect PmSLP, LepB (inner membrane control) and OmpA (outer membrane control) respectively. (Adapted from Hooda et al. Front Cell Infect Microbiol 2017)
2.3.7 Comparison of putative SLP proteins revealed a conserved structural domain
With this database of putative Slam-dependent SLPs, I was interested in identifying structural features that are shared by this family of proteins. Structures for four neisserial Slam-dependent SLPs have been solved by X-ray crystallography and NMR (Figure 2.7). While these proteins share no sequence similarity, they do share a protein domain composed of a flexible handle domain followed by an eight-stranded barrel domain. N. meningitidis TbpB and LbpB contain two lobes of the conserved domain, while fHbp and HpuA contain only one lobe.
To gain insight into the structure of Slam-dependent SLPs, I compared the predicted domains (InterProScan) on our list of putative Slam-dependent SLPs. I found that 58 contain either a lipoprotein GNA1970-related (InterPro signature: IPR014902) or a solute-binding protein, TBP-like (InterPro signature: IPR001677) domain. These domains are found on fHbp and TbpB respectively, both which are Slam-dependent SLPs, and contain an eight-stranded beta barrel at their C-terminus. Further, 127 were predicted to encode a pagP-beta barrel (InterPro signature: IPR011250). PagP is an outer membrane enzyme that forms an eight-stranded barrel and is involved in catalyzing palmitate transfer from a phospholipid to a glucosamine unit of lipid A [133]. While PagP is an integral outer membrane protein, several soluble proteins are also predicted to contain an eight- stranded PagP-beta barrel. Most of the predicted Slam-adjacent lipoproteins also had a variable N-
41
terminal region preceding the eight-stranded barrel that could form a handle-like domain seen in the representative three-dimensional structures of Slam-dependent SLPs.
human N-lobe C-lobe Factor H human Transferrin
human Hemoglobin
N-lobe α β β α
C-lobe
N. meningitidis TbpB K. denitrifcans HpuA N. meningitidis fHbp PDB: 4QQ1 PDB: 5EC6 PDB: 4AYI
Species PDB ID Species PDB ID Species PDB ID Neisseria meningtidis 4QQ1, 3VE1, 3VE2, Neisseria gonorrhoeae 5EE2 Neisseria meningitidis 4Z3T, 4AYD, 4AYE, 3V8U Kingella denitrifcans 5EC6, 5EE4 4AYI, 4AYM, 4AYN, Actinobacillus suis 4O3W 2Y7S, 3KVD, 2KC0, Actinobacillus pleuropneumoniae 4O3X, 4O3Y, 4O3Z, 2W80, 2W81 4O49, 3PQU, 3PQS, 3HOE, 3HOL Hemophilus parasuis 4O4U, 4O4X
Figure 2.7: Structures of known Slam-dependent SLPs. Structures of TbpB (green), HpuA (cyan) and fHbp (orange) are shown. The binding partner for each of the SLPs (human transferrin (red), hemoglobin (black) and Factor H (blue)) is also shown. Table below lists the accession numbers for all Slam-dependent SLP structures deposited to the Protein Data Bank (PDB). Primary references for TbpB [80,81,110,111,134,135], HpuA [83] and fHbp [90,136–140] structures are also included. (Adapted from Hooda et al. Front Cell Infect Microbiol 2017)
2.3.8 A number of non-SLP genes found adjacent to putative Slam homologs
From my analysis, ~52% of genes encoding Slam-like proteins had an adjacent gene encoding a putative SLP. To examine the domain structure of the remaining ~48% of the genes, I used InterProScan to predict domains on all Slam-adjacent proteins. Interestingly, similar to all putative Slam-adjacent SLPs, 116 (32.9% of 353 of Slam clusters) also had a PagP domain (InterPro signature: IPR011250) predicted. 89 of these also had a TonB-dependent receptor present adjacent to them (Appendix B).
42
Figure 2.8: Putative non-lipoprotein substrate of Slam homolog in Neisseria gonorrhoeae. Slam gene cluster in N. gonorrhoeae strain FA1090. NGO0555 (shown in blue) was identified as a Slam homolog in our bioinformatics search. NGO0554 (shown in green) was annotated as a hypothetical protein with an 8-stranded pagP barrel domain (InterPro signature: IPR011250). It contained a signal peptide (shown in yellow) but no lipobox motif. Rather it contained a signal peptidase I site found on periplasmic and several secreted proteins. NGO0553 is a predicted TonB- dependent receptor also referred as TdfG in the literature.
One such example was found in N. gonorrhoeae NGO0553-0555 gene cluster. NGO0553 is a predicted TonB-dependent receptor, NGO0554 is a PagP-containing protein (InterPro signature: IPR011250) with signal sequence but no lipobox motif and NGO0555 is a putative Slam (Figure 2.8). To ensure that NGO554 is not a lipoprotein, I reannotated the ORF to ensure there are no start codons that can result in the presence of a lipobox motif. Previous study has shown that both NGO0554 and NGO0555 are upregulated upon peroxide stress indicating that they are functional and play a role in antioxidative stress response [141]. Additionally, NGO0553 is also referred to as TdfJ, a TonB dependent receptor whose function is not known [142]. Further experiments are required to tease apart the functional role of this Slam gene cluster.
2.4 Discussion
With the increase in the number of bacterial genomic sequences, it has become evident that surface lipoproteins (or SLPs) are widespread in Gram-negative bacteria. One family of SLPs is characterized by a common structural architecture composed of an eight-stranded beta-barrel
43
domain and beta rich handle domain (TbpB, LbpB, HpuA, and fHbp). These proteins require a member of a unique family of proteins to traverse the outer membrane named Slam [119]. While we still do not know the exact role played by Slams, these proteins are specific for TbpB-like SLPs and may represent a novel class of outer membrane translocon or chaperone dedicated to the transport of SLPs from the inner leaflet of the outer membrane to the surface.
In this study, we performed a bioinformatic analysis of putative Slam homologs and identified a number of Slam-dependent SLPs in many different species of Proteobacteria. To our knowledge, this is the first systematic study to look at the distribution of SLPs in different Gram-negative bacteria. Previous attempts to look at TbpB-like lipoproteins have been stymied by the degree of variation that is found in these proteins. Since Slam sequences are more conserved, we were able to identify a large number of SLP homologs owing to the genetic linkage between Slams and TbpB- like SLPs. We have extended the number of bacteria that are now predicted to possess SLPs but also provide a framework to systematically search for these proteins. Based on this work, Slam- dependent SLPs represent the largest sub-family of SLPs reported thus far. However, our approach also has some limitations. We were only identified genes for putative SLPs located in the immediate vicinity of genes for Slams. Using this method, we would have overlooked many known Slam-dependent SLPs in pathogenic Neisseria sp., which are transported in a Slam1-dependent manner but are encoded by genes not located in the vicinity of the gene for Slam1 (eg. NmTbpB and NmLbpB). Therefore, our list of Slam-dependent SLPs is certainly incomplete; and the bacterial species with Slam homologs but no Slam-adjacent lipoprotein may also contain SLPs.
Our analysis allowed to identify a previously uncharacterized protein PM1514 in Pasteurella multocida as a putative SLP. Using localization assays we confirmed PM1514 is present in the outer membrane but is only detected on the surface of E. coli when co-expressed with the putative Slam PM1515. Taken together, these findings suggest that PM1515 is a Slam homolog and PM1514 is a Slam-dependent SLP. Additional work is required to further investigate this putative SLP. 3H-palmitoyl labelling assay should be used to confirm the lipidation of PM1514 in both E. coli and P. multocida. Flow cytometry and proteinase K shaving assay should be performed in P. multocida to test the surface display of PM1514 in its endogenous host. PM1514 is found in most sequenced strains of P. multocida, indicating that it may play an important role in survival of P. multocida in its host organisms.
44
In our search, we identified that Slam related sequences are found only in Proteobacteria. and not found in Bacteroides, Borrelia and Campylobacter. Interestingly, we investigated the gene neighbourhoods of four other known SLPs namely, JlpA in Campylobacter [143], HmuY [144] and SusD [145] in Bacteroides and OspA/C in Borrelia [60]. Upon inspection none of these SLPs showed genetic linkage with any known/predicted outer membrane protein (data not shown) nor were any Slam related sequences found in Bacteroides or Borrelia. Coupled with the mechanistic studies completed on the SLP translocation pathways for Bacteroides [27] and Borrelia [59], it is likely other translocation systems are used by Bacteroides and Borrelia SLPs.
Using the list of putative Slam-dependent SLPs, we have found that all these proteins contain a predicted eight-stranded barrel domain. All Slam-dependent SLPs contain either one or two copies of this barrel domain. Hence, we predict that the barrel domain may interact with Slam and contribute to Slam-dependent translocation. However further experiments are required to tease apart the translocation motif present on these SLPs. Towards this end, the dataset generated by this study will be a valuable tool to identify secretion motif that is common amongst Slam-dependent SLPs.
The ability of Slam proteins to robustly potentiate the display of TbpB-like SLPs from a diverse set of bacteria in E. coli provides further evidence of their direct involvement in translocation. While more work is required to understand the mechanism of Slam function, the work done so far shows the efficacy of using Slam as a system to deliver proteins to the surface of Gram-negative bacteria and has potential applications in development of bacterial surface display technology [146]. Taken together, our work suggests that TbpB-like SLPs and Slams form a “plug-and-play” cassette, reminiscent of the two-partner secretion systems or Type Vb secretion system [147,148]. Many parallels can be drawn between Slams and the transporter proteins of two partner secretion systems, commonly referred to as TpsBs. First, both Slams and TpsBs often exist in the same gene clusters as their substrates. Second, multiple Slams and TpsBs can be present in a single bacterial species with each paralog specific for a distinct set of substrates [149]. Third, based on our preliminary studies, both TpsB and Slam recognize substrates of defined structural domains (soluble 8-stranded C-terminal barrel for Slams and right-handed beta helix for TpsB). However, TpsBs and Slams have different predicted structures and potentially different molecular mechanisms of action. TpsB belong to the Omp85 family of proteins that include outer membrane translocons such as TamA and BamA [150]. In contrast, Slam possesses a unique domain architecture and no structure of any
45
protein of this family is currently available. Additionally, while TpsBs are specific to secreted proteins, Slams are responsible for transport of SLPs that possess a lipid anchor that needs to be flipped from the inner leaflet to the outer leaflet of the outer membrane. Further, through in vitro reconstitution and cross-linking studies, TpsBs have been shown to act as outer membrane translocons for their substrates [151,152]. This is critical as such studies are currently lacking for Slams and hence the role of other outer membrane proteins such as the Bam complex in Slam- mediated SLP translocation cannot be ruled out. We will revisit these questions in Chapter 4.
To date, experimental evidence shows that Slams are specific for the delivery of lipidated proteins to the surface of Gram-negative bacteria. However, I have also identified many Slam gene clusters with adjacent non-lipidated proteins with an 8-stranded barrel domain. This suggests that Slams may also secrete substrates indicating that Slams represent a more generalized secretion system. Finally, this study furthers the argument that SLPs are an important and yet under-appreciated family of proteins and their study may lead to identification of novel mechanisms utilized by bacteria to interact with their environmental surroundings and/or their hosts.
46
3 Chapter 3 Investigating the molecular details of Slam function using an E. coli reconstitution assay
Acknowledgements: Parts of this chapter were published in Hooda and Lai et al. Nat. Microbiology 2016. Ms. Christine Lai cloned most of the constructs used in the study. Mr. Andrew Judd cloned Slam single point mutant constructs for cross-linking studies and also helped on flow cytometry studies for studying specificity of Slam1 and Slam2. Ms. Esther Shin helped with the globomycin assay and westerns for the sucrose-flotation assays. LepB and OmpA antibodies were obtained from Dr. Jan Willem deGier, Stockholm University. PgaB construct and antibodies were obtained from Dr. Lynne Howell, SickKids, Toronto. TamA and TamB knockouts were obtained from Dr. Karen Maxwell, University of Toronto. fHbp1 plasmid and antibody was obtained from Drs. Dan Granoff and Rolando Pajon at CHORI, Oakland. LbpB construct and antibodies were obtained from Dr. Joanne Lemieux, University of Calgary.
3.1 Overview
In the previous chapter, I described the bioinformatics analysis performed to obtain a database of putative SLPs and Slams. We predicted that the database would be a great asset for understanding Slam function in detail. Hence, I took a number of putative SLPs and Slams and performed multiple sequence alignments to identify features that are conserved amongst these families of protein. However, owing to the high sequence variability between different Slams and SLPs, I was unable to identify regions that might be important for function. This suggested to us that we would need to use alternate strategies for teasing apart the regions on Slams and SLPs that are important for the movement of SLPs across the outer membrane.
In Chapter 2, I also described the E. coli translocation assay that was used to reconstitute surface display of diverse set of Slam-dependent SLPs, demonstrating its versatility. By mutating various positions on Slams and SLPs, I reasoned that we could examine the regions on Slam and SLPs that are important for function. To this end, I focused on N. meningitidis SLPs TbpB and HpuA as these SLPs are delivered to the surface by Slam1 and Slam2 respectively. First, I showed that Slam1 and Slam2 are specific for their substrates. Then, by performing mutational analysis and domain swaps,
47
I was able to show that the Slam translocation motif lies on the last two strands of the 8-stranded beta-barrel. The C-terminal domain also dictated specificity as swapping domains between TbpB with HpuA resulted in the ability to switch the specificity of Slams.
The E. coli assay also provided some clues about the regions on Slams that are important for its function. I was able to show that Slam barrel domain is essential for function in E. coli (similar to results obtained by Christine Lai and Andrew Judd in N. meningitidis as described in Chapter 1). Further, the barrel domain retained its specificity indicating it contains the residues that discriminate between SLP substrates. Given that the structure of Slam homolog or any DUF560- containing proteins is not available, Andrew and I made mutations on residues predicted to face of the lumen of the membrane spanning barrel domain and performed functional and immunoprecipitation assays with them. However, no residues could be identified on Slams that are critical for Slam function.
In this chapter, I will also describe other insights that could be inferred regarding Slam function. First, I was able to demonstrate that Slams act downstream of the Signal peptide II and Lol confirming that they act in the outer membrane. Second, by using a GST-TbpB Clobe construct and domain swaps between TbpB and HpuA, I was able to show that translocation of SLPs occurs from the C-terminus to the N-terminus and that Slam1 and TbpB interact in the outer membrane. Third, I was able to investigate the role of the Tam complex and to a lesser extent, the Bam complex and show that they are not involved in translocation of Slam-dependent SLPs. Taken together, these studies have helped develop a more detailed model of Slam function.
3.2 Methods
3.2.1 Bacterial strains and growth conditions
Strains used in this study are summarized in Appendix C. E. coli were grown in LB media containing antibiotics when necessary – 50 µg/mL kanamycin and 100 µg/mL ampicillin. Cloning procedures were carried out using E. coli MM294 competent cells. Protein expression was performed using E. coli C43 (DE3) cells for all the flow-cytometry and western blot analysis.
48
3.2.2 Cloning of Slam, SLP homologs Genes were cloned into expression vectors by RF cloning [120] and signal peptides and tags were inserted using round the horn cloning by Ms. Christine Lai [121].
For flow cytometry experiments performed in E. coli C43(DE3), the four SLPs (TbpB, LbpB, fHbp, and HpuA) were cloned into pETDuet (to make pETDUET TbpB, LbpB, fHbp or HpuA) and Slams were inserted into pET26b (pET26 Slam1, pET26 Slam1barrel and pET26 Slam2). Genes tbpb and hpua were amplified from the genome of N. meningitidis strain B16B6 where as lbpb and fhbp were amplified from N. meningitidis strain MC58 (due to the availability of genetic information or antibodies for these proteins) and inserted into pETDuet by RF cloning. The E. coli pelB signal peptide was inserted to replace the endogenous neisserial signal peptides. The mature slam1 gene was cloned from B16B6 and slam2 from MC58 to be in frame with the pelB signal peptide of pET26b. A 7xHis-tag was inserted between the pelB and the mature Slam sequences. The pelB Slam1 was also subcloned into the MCS site 1 of pETDuet for immunoprecipitation assays. The pET26 Slam1 barrel construct was cloned by amplifying the b-barrel domain (coding region from amino acid 204 to the end) and inserting the amplicon in frame with the 7xHis-tag of pET26 Slam1 while replacing the mature full-length sequence.
Minimal domain and domain swap constructs used were pETDUET HpuA, TbpB N-lobe, TbpB c- lobe, TbpB N-lobe-HpuA. The mature HpuA was amplified and inserted in frame with the signal peptide of TbpB and simultaneously replaced the rest of the TbpB gene in pETDUET TbpB to create pETDUET HpuA using RF cloning. The anchor peptide (1-40), N lobe of TbpB (41-331) and C lobe of TbpB (332-579) were parsed based on the N. meningitidis strain B16B6 TbpB structure (PDB ID: 4QQ1). The N-lobe construct was cloned using inverse cloning to insert a stop codon between the N and C lobes of TbpB using pETDUET TbpB as template. The C-lobe construct of TbpB was cloned using RF cloning to insert and simultaneously replace TbpB N-lobe in pETDUET TbpB N-lobe. The C-lobe inserted in frame with the anchoring peptide. The fusion construct pETDUET TbpB N-lobe-HpuA was cloned by inserting HpuA (3’-end) in frame with the N-lobe of pETDUET TbpB-N-lobe by RF cloning. Poly-Ala mutants were created by FastCloning using pETDUET TbpB as template.
49
For flow cytometry experiments performed in E. coli Keio strains, Slam1 and TbpB were expressed from an arabinose and a pT5-lac promoter, respectively, since Keio strains do not carry the lambda DE3 construct. slam1 from pET26 was subcloned into pHERD30T to make pHERD30T pelB Slam1, and the pT5-lac promoter from pQE30 (Qiagen) was inserted by assembly PCR[134] upstream of TbpB in pETDUET TbpB to make pETDUET pT5 TbpB.
To clone the pETDuet GST-C-lobe construct for the spheroplast assays, the GST tag was amplified from pGEX-6p-1 with primer ends complementary to the anchor peptide (forward) or C-lobe (reverse) of TbpB so that the amplicon could replace the N-lobe by RF cloning. For pETDUET GST-C-lobe-cys, Cys was introduced in the linker region using site-directed mutagenesis.
For the Slam1 mutants, site-directed mutagenesis was used to obtain Slam single-point mutants listed in Appendix C. The mutants were tested in the E. coli translocation assay and then used for cross-linking studies.
3.2.3 Flow cytometry
For the E. coli translocation assays, the display of SLP was determined using flow cytometry. Pairs of SLP and Slam plasmids (listed in Appendix C) were transformed into C43 (DE3) cells and grown in 1 mL of auto-induction media for 18 hrs at 37 oC. Cells were harvested, washed twice in
PBS containing 1 mM MgCl2, and labelled with the antibodies with α-TbpB (1:200, rabbit sera), α- LbpB (1:200, rabbit sera), α-fHbp (1:200, monoclonal antibody), α-HpuA (1:200, rabbit sera) and α-flag (1:200, Sigma), or biotinylated human transferrin (0.05 mg/ml, Sigma) for 1 hr at 4 ˚C. The cells were then washed twice with PBS containing 1mM MgCl2 and then labelled with R-FITC conjugated α-rabbit IgG (25 µg/mL, Thermo Fisher Scientific) R-PE conjugated Streptavidin (0.5 mg/ml, Cedarlane) or R-PE conjugated α-mouse IgG (25 µg/mL, Thermo Fisher Scientific) for 1 hour at 4 ˚C. Following staining, cells were fixed in 2% formaldehyde for 20 mins and further washed with PBS containing 1 mM MgCl2. Flow cytometry was performed with a Becton Dickinson FACSCalibur and the results were analyzed using FLOWJO software. Mean fluorescence intensity (MFI) was calculated using at least three replicates was used to compare surface exposure of a given SLP between different strains. Statistical significance was calculated by comparing MFI between different samples using the one-way ANOVA test.
50
Western blots were used to test the expression levels of each of the constructs used for the flow cytometry experiments. α-TbpB (1:5000, rabbit sera), α-LbpB (1:5000, rabbit sera), α-fHbp (1:5000, monoclonal antibody), α-HpuA (1:5000, rabbit sera), α-flag (1:5000, Sigma) and α-His (1:5000, Thermo Fisher Scientific) antibodies were used to test expression of the SLP and Slam constructs respectively.
3.2.4 Globomycin assay N. meningitidis strain B16B6 WT and Δnmb0313 were grown on BHI plates overnight and then resuspended in BHI media containing 0.1 mM of the iron chelator deferoxamine mesylate (Sigma) for 1 hour. Cells were dosed with 10 µg/mL of globomycin (a gift from L. Howell) for 2 hours, harvested, lysed in SDS loading buffer and analyzed for TbpB buildup by western blot using α- TbpB antibodies.
E. coli C43(DE3) cells transformed with pairs of SLPs (Nme TbpB WT and mutants) and their respective Slams were used to inoculate 1 ml of LB supplemented with appropriate antibiotics. The starter cultures were diluted 1:50 with LB to start fresh 200 µl cultures. After reaching an OD600 ~ 0.8, cells were dosed with 1 mM IPTG and 10 µg/ml of globomycin and further incubated for 3 hr at 37 °C. Cells were then collected by centrifugation, lysed in SDS loading buffer and analyzed using western blots probed with α-TbpB and α-flag antibodies.
3.2.5 Sucrose density ultracentrifugation
E. coli C43(DE3) cells expressing pET26 Nme Slam1 and pETDUET TbpB WT and poly-alanine mutants were grown overnight and then used to inoculate 200 ml LB with the appropriate antibiotic.
The cells were grown to an OD600 ~ 0.6, induced with 1 mM IPTG and then grown for an additional 18 h. The cells were pelleted, resuspended in 20 mM Tris pH 8.0, 200 mM NaCl with fresh lysozyme (1 mg/ml), 2 mM PMSF and DNase I (0.05 mg/ml), lysed by sonication and then centrifuged at 10,000 r.c.f. to remove cell debris. The supernatant was centrifuged at 125,000 r.c.f. for 1 h to collect the cellular membranes. The membrane pellet was resuspended in 1 ml of 20 mM Tris pH 8.0, 200 mM NaCl using a micro-glass homogenizer.
The inner and the outer membrane of E. coli were separated using the sucrose density ultracentrifugation using a modified protocol from the one previously described [119]. For this assay, 500 ul of the membrane pellet was applied on top of a 13.2 ml thin-wall polyproplyene tube
51
containing step gradients of 2.5 ml of 2 M, 6 ml of 1.44 M and 3 ml of 0.77 M sucrose. The tubes were centrifuged at 83,000 r.c.f. for 16 hours. The outer membrane and inner membranes collect at the interface of the 2 M and 1.44 M sucrose cushions and 1.44 M and 0.77 M sucrose layers, respectively. Twelve 1 ml fractions were collected and subjected to SDS–PAGE followed by western blotting with α-TbpB (1:10,000), α-SecY (1: 2,000) and α-OmpA (1:40,000) antibodies.
3.2.6 Pulldown of TbpB with LolA R43L mutant E. coli lolA was cloned into pETDuet MCS site 1 of pETDuet TbpB (TbpB in site 2). Further, a 6xHis-tag was inserted at the C-terminal end of LolA. Site-directed mutagenesis was performed to generate the R43L mutation, producing pETDuet LolA R43L_TbpB. The pETDuet TbpB and pETDuet LolA R43L_TbpB plasmids were separately transformed into E. coli C43(DE3) and a swab of at least 10 colonies was used to inoculate 5 mL of auto-induction media [122] with appropriate antibiotics. Cells were grown for 18 hours, harvested and lysed in 1 mL of lysis buffer (50 mM Tris pH 8.0, 200 mM NaCl, 1 % Triton X-100, 2 mM EDTA, 1 mg/mL lysozyme, 0.05 mg/mL of DNase I, and 2 mM PMSF) for 15 mins at room temperature. After cell debris was removed by centrifugation, 400 µl of lysate was diluted ten times in resuspension buffer (50 mM Tris pH 8.0, 200 mM NaCl, 10 mM imidazole, 0.1 % Triton X-100) and then incubated with 25 µl of Ni-NTA resin (pre-equilibrated with resuspension buffer) for 1 hour of batch binding at 4oC. Ni- NTA resin was washed three times with resuspension buffer before mixing with SDS loading buffer for analysis by SDS-PAGE and western blots using α-His (1:5,000) or α-TbpB antibodies.
3.2.7 Spheroplast generation and flow cytometry assay
E. coli C43(DE3) cells transformed with pET26b Nme Slam1 and pETDUET Nme TbpB wildtype and polyalanine mutants were grown overnight in 1 ml auto-induction media with appropriate antibiotics. Cells were collected and resuspended in 500 µl of 100 mM Tris pH 7.5 and 0.5 M sucrose. Spheroplast formation was initiated by the addition of 500 µl of 8 mM EDTA pH 8.0 and 0.2 mg/ml lysozyme to the samples and incubating for 30 min on ice. The spheroplasts were then collected by centrifugation at 10,000 r.c.f. for 5 min at 25 °C, washed once with 50 mM Tris pH 7.5, 0.25 M sucrose and used for flow cytometry experiments.
Sample preparation and flow cytometry analysis were optimized for spheroplasts as described previously [119]. During the downstream analysis of cells in FlowJo, an increase in size forward
52
scattering (FSC) was observed upon spheroplast formation. The cell population was gated to exclude intact bacterial cells.
3.2.8 Flow cytometry analysis of TamA and TamB deficient E. coli E. coli Keio collection [12] strains BW25113 WT, JW4179-1 (ΔytfM) and JW4180-1 (ΔytfN) were transformed with plasmids expressing pETDUET pT5 TbpB and pHERD30T pelB Slam1. Transformants were grown in 0.01% arabinose overnight to induce expression of Slam1 and then in 1 mM IPTG for 3 hrs for TbpB expression before cells were harvested and prepared for flow cytometry, as described in “Flow cytometry assay” listed in Section 2.3.
3.2.9 Purification of GST-C-lobe and Slam1 from outer membrane
To purify the stalled GST-C-lobe and Slam1 construct, 6L culture of E. coli C43(DE3) cells with pETDUET GST-C-lobe and pET26b Slam1 construct were grown to OD ~ 0.8, induced with IPTG and grown at 18 °C. After 18hrs of growth, cells were harvested and resuspended in 50 mM Tris pH 8, 200 mM NaCl. Following solubilization, cells were lysed through sonication and the lysate obtained was spun at 14,000 rpm for 30 mins. The supernatant obtained was spun in Beckman ultracentrifuge at 40,000 rpm for 1 hr at 4 °C to obtain the membrane pellet. The pellet was solubilized in 3% Elugent for 16 hrs at 4 °C and then respun at 40,000 rpm for 1 hr at 4 °C to remove debris. The supernatant was incubated with Glutathione Sepharose 4B beads (GE Healthcare) and incubated for 2 hrs at 4 °C. The GST beads were then collected in a column and washed with 25 ml of 50mM Tris pH 8, 200 mM NaCl, 0.06% DDM three times. The protein was eluted in 10 ml of 50mM Tris pH 8, 200 mM NaCl, 0.06% DDM, 10 mM reduced glutathione. The eluted samples were run on an SDS-PAGE gel to check purity and then concentrated with a 100 kDa filter and then run on a Sephadex S200 gel filtration column for further purification. Different fractions were collected and run on an SDS gel to check purity and estimate yield.
3.2.10 Crosslinking studies between Slam1 and GST-C-lobe-cys
E. coli C43(DE3) cells transformed with pET26b Nme Slam1 mutants and pETDUET GST-C-lobe- cys were grown overnight in 25 ml auto-induction media with appropriate antibiotics. Before harvesting, cells were treated with 100 µM CuSO4 for 10 mins for promoting disulphide bond formation [153]. The cells were resuspended in 5 ml of 50 mM Tris pH 7.5, 200 mM NaCl and lysed using sonication. The membrane pellet was obtained by spinning the cells using
53
ultracentifugation. The membrane pellet was reconstituted in SDS sample buffer (50 mM Tris-Hcl, pH 6.8, 2% SDS, 10% glycerol, 1% b-mercaptoethanol, 12.5 mM EDTA, 0.02% bromophenol blue) in the presence and absence of reducing agent DTT and boiled at 95 °C for 5 mins. Western blots were used to determine the presence of GST-C-lobe, Slam1 and Slam-GST-C-lobe complexes.
3.3 Results
3.3.1 Slam acts at the outer membrane
In order to properly traffic to the cell surface, the neisserial SLPs require signaling motifs to translocate across the bacterial cell envelope (Figure 3.1A). SLPs contain the canonical signal peptide and consensus lipobox motif that are necessary for translocation through the Sec secretion machinery and lipidation of their mature amino-terminal cysteine [14]. We confirmed that neisserial SLPs use this pathway for translocation by testing the proper cleavage of TbpB by signal peptidase II (SpII). Treatment of globomycin, an inhibitor of SpII, led to detection of unprocessed full length TbpB buildup in both N. meningitidis and E. coli [154] (Figure 3.1B). This confirms that neisserial SLPs follow the translocation pathways used by all lipoproteins. Most lipoproteins traverse the periplasm through the Lol system. Previous studies have shown that the E. coli LolA R43L mutation inhibits the transfer of lipoproteins to LolB, effectively trapping LolA-lipoprotein complexes [155]. Expression of TbpB together with LolA R43L in E. coli stalls lipoprotein translocation and the complex of TbpB bound to LolA R43L can be detected in pulldowns (Figure 3.1C). This confirms that TbpB uses the Lol system by neisserial SLPs. Interestingly, TbpB expressed in laboratory strains of E. coli is able to reach the outer membrane (Figure 3.1D), but it is not displayed on the surface. Expression of Slam with TbpB did not affect the membrane localization of TbpB (Figure 3.1D) suggesting that Slams act in the outer membrane and play a role in the final translocation step where the SLPs are ‘flipped’ across the outer membrane to the surface of the cell.
54
A C
D
B
E Inner Outer membrane membrane
Fractions: 12 3 4 5 6 7 8 9 10 11 12
α TbpB
α LepB
α OmpA
α Slam1
Figure 3.1: Translocation pathway for the surface lipoprotein, TbpB, to the cell surface. (A) Proposed model of TbpB translocation to the cell surface of Neisseria. TbpB is directed to SecYEG by its signal peptide sequence. Lgt, SPII and Lnt are responsible for processing lipoproteins in the inner membrane (IM). TbpB transits through the Lol system, consisting of IM LolCDE, periplasmic LolA and outer membrane (OM) LolB, and is consequently inserted into the inner leaflet of the OM. An unidentified protein or complex is responsible for flipping TbpB onto the cell surface. Two parts of this pathway can be inhibited: chemically, by globomycin or genetically, by mutations to LolA at R43L, as indicated. (B) Western blot analysis of globomycin treated cells. N. meningitidis WT and Δnmb0313 (above) and E. coli (below) cells (transformed as labeled) were treated with globomycin, an SPII inhibitor, leading to the accumulation of TbpB with uncleaved signal peptide (SP). This confirms that TbpB requires SPII after being transported by the Sec machinery in the IM. (C) LolA-R43L pulldown assay. E. coli cells expressing TbpB alone or TbpB with His-LolA R43L, a Lol translocation deficient mutant, were lysed and batch bound to Ni-NTA resin. TbpB specifically interacts with His-LolA-R43L suggesting that it is transported from the inner to the outer membrane by the Lol machinery. (D) Western blot analysis of sucrose gradient ultracentrifugation fractions. E. coli cells expressing TbpB were lysed and the cellular membrane fraction was subjected to sucrose gradient ultracentrifugation to separate the IM and OM. IM
55
(fractions 5, 6, 7) and OM (fractions 9, 10, 11) were identified by the enrichment of resident proteins, LepB and OmpA, respectively. TbpB can be found throughout the IM and OM, with the majority in the OM fractions. (E) Localization of TbpB using sucrose gradient ultracentrifugation in the presence of Slam1. To test if expression of Slam1 changes the cellular localization of TbpB, sucrose gradient ultra-centrifugation was used to separate the inner (IM) and the outer membranes (OM) of E. coli cells expressing both. LepB, an IM protein and OmpA, an OM protein, were used as controls to highlight the IM (fractions 5,6 and 7) and OM (fractions 9, 10 and 11). TbpB is present in both IM and OM, with an enrichment in OM fractions, regardless of the presence or absence of Slam1. One representative blot from two biological replicates is shown. (Adapted from Hooda and Lai et al. Nat. Microbiology 2016)
3.3.2 Slam and its membrane domain exhibit specificity In N. meningitidis, our bioinformatic investigation revealed two Slam paralogs: Slam1 that could translocate TbpB to the surface of E. coli, and Slam2 that was able to translocate HpuA. Thus, to decipher if the paralogs exhibit specificity for their substrate, we examined the cell surface display of neisserial SLPs in laboratory strains of E. coli (BL21–C43(DE3)). As seen previously in N. meningitidis, the simple addition of Slam1 to E. coli C43 cells co-expressing TbpB, LbpB or fHbp facilitated their display on the bacterial surface. To ensure that this translocation activity was not due to the aberrant expression of SLPs in the presence of Slams, we performed western blots and found no correlation between surface display and expression (Figure 3.2A-C). This demonstrates a “gain of function” phenotype in E. coli and indicates that Slam1 plays a direct role in surface display of TbpB, LbpB and fHbp.
In contrast to Slam1, Slam2 was ineffective at translocating TbpB, LbpB and fHbp to the E. coli surface (Figure 3.2A-C). In considering that Slams have cargo specificity, we also tested the surface translocation potential of HpuA that is translocated by Slam2. Consistent with this model, HpuA was surface-exposed only in E. coli co-expressing Slam2 but not in cells co-expressing Slam1 (Figure 3.2D). Furthermore, this cargo specificity resides primarily within the b-barrel domains as the Slam1 b-barrel alone can translocate TbpB, LbpB and fHbp but not HpuA to the surface of E. coli (Figure 3.2), similar to the results obtained for full-length Slam1. Taken together, this establishes that Slams translocate specific neisserial SLPs across the outer membrane to the cell surface and the barrel domain is important for specificity.
56
Moraes - Figure 3 n.s. A n.s. B a bio-hTf b *** α LbpB *** 0 150 *** 150 *** 0 e (%) e (%) 100 100 en c en c 0 e e ti v ti v es c es c el a el a 0 50 R
R 50 luo r luo r ean F ean F M M 0 0
LbpB TbpB Barrel Barrel
TbpB + Slam1 TbpB + Slam2 LbpB + Slam1 LbpB + Slam2 + Slam1 + Slam1
TbpB LbpB
α TbpB α LbpB
*** C * D c d n.s. 150 **** α fHbp 150 α HpuA **** n.s. e (%) e (%) 100 100 en c en c e e ti v ti v es c es c el a el a R R 50 50 luo r luo r ean F ean F M M 0 0
fHbp Barrel HpuA Barrel
fHbp + Slam1 fHbp + Slam2 HpuA + Slam1 + Slam1 HpuA + Slam2 puA fHbp + Slam1 H
α fHbp α HpuA
Figure 3.2 Slam1 and Slam2 demonstrate substrate specific translocation of SLPs in the E. coli translocation assay. (A-D) Quantification of surface TbpB, LbpB, fHbp or HpuA in E. coli, as analyzed by flow cytometry. E. coli was transformed as labeled and grown in auto-induction media. Cells were labeled with biotinylated human transferrin (bio-hTf) to detect TbpB surface expression, or antibodies to detect LbpB, fHbp or HpuA. Relative mean fluorescence was calculated by scaling the averaged mean fluorescence intensity of cells reconstituted with full length Slam1 (A-C) or Slam2 (D) to 100 and applying the same scaling factor to the other samples. Error bars represent the SEM from three biological replicates. Statistical significance was determined by ANOVA, *p<0.05; ***p<0.001; ****p<0.0001; ns: not significant. (A-D) lower panel, Western blot analysis of TbpB, LbpB, fHbp or HpuA on lysates prepared from cells of equal density that were used for labeling. Blots are representative of three independent experiments. (Adapted from Hooda and Lai et al Nat. Microbiology 2016)
57
3.3.3 Slam is specific for SLPs To confirm that Slams are specific to neisserial SLPs, we also tested for the surface display of E. coli PgaB [156], an inner-leaflet outer membrane lipoprotein. The PgaB plasmid and antibodies were obtained from Dr. Lynne Howell lab at the University of Toronto. Similar to the E. coli reconstitution assay in Section 3.3.2, Slam1 and PgaB were co-expressed in E. coli and flow cytometry was used to quantify the amount of PgaB on the surface (Figure 3.3). No surface PgaB was observed, illustrating that Slams have specificity for SLPs and do not translocate periplasmic- facing E. coli lipoproteins. However, this result should be taken cautiously as I did not test PgaB fluorescence upon permeabilizing the outer membrane (which should lead to increase in PgaB fluoresence) and hence can be attributed to PgaB antibodies themselves. Nonetheless, these findings indicate that Slam1 is selective in transporting neisserial SLPs across the outer membrane and is not generic to all lipoproteins.
α PgaB
400
300 PgaB PgaB + Slam1 200 PgaB + Slam2 Counts
100
0 0 1 2 3 4 10 10 10 10 10 PE Fluorescence
Figure 3.3: Localization of PgaB in E. coli cells co-expressing Slams. Analysis of surface PgaB, an inner-leaflet OM lipoprotein in E. coli, by flow cytometry. E. coli was transformed as indicated and grown in auto-induction media. Cells were harvested and labeled with antibodies to detect PgaB. Plots of cell counts against PE fluorescence are shown and is representative of three independent experiments. No surface PgaB was detected when co-expressed with Slam1 or Slam2 suggesting that Slams do not promote surface display of E. coli lipoproteins and are specific for neisserial SLPs. (Adapted from Hooda and Lai et al Nat. Microbiology 2016)
3.3.4 TbpB C-lobe contains the translocation motif
Motivated by the observation that Slams are specific for neisserial SLPs, we examined the role of the different TbpB domains in translocation. From my bioinformatics analysis (Chapter 2), I found
58
that all Slam-dependent SLPs share the same domain architecture with the beta-handle and 8- stranded beta-barrel domains. Interestingly, while some Slam-dependent SLPs have two copies of these domains (TbpB and LbpB, both which are Slam1 substrates), others have a single copy
Figure 3.4 Translocation of TbpB mutants and truncations. (A) TbpB domain architecture. TbpB is composed of a signal peptide (SP), anchor peptide (AP) and two structural lobes composed of a barrel and a handle domain. The TbpB N-lobe is more variable amongst TbpB homologs and binds to its substrate human transferrin, while the TbpB C-lobe is more conserved, but its function is not known. (B) The secondary structure of the TbpB C-lobe is shown. The poly-alanine mutations were introduced in strands B22, B23, B30 and B31 (shown in red). Mutants in residues facing inside and outside the barrel on strand B30 and B31 are also shown. TbpB mutants with
59
poly-alanine mutations on the inner and outer face of the last strands were also made (B30in, B30out, B31in, B31out) (C) Translocation of the individual domains of TbpB. TbpB, TbpB N-lobe and TbpB C-lobe were expressed with Slam1 and surface display was tested using flow cytometry. Mean fluorescence intensity (MFI) plots and western blots with α-TbpB antibodies are shown. (D) Mean fluorescence intestity (MFI) plots and western blots using α-TbpB antibodies of different TbpB mutant constructs are shown.
(HpuA, fHbp that are substrates for Slam2 and Slam1 respectively). To test which of the two copies of TbpB play a role in translocation, we focused on the N. meningitidis TbpB (Figure 3.4A). I tested the surface display of each of the two lobes of TbpB individually in the E. coli translocation assay. From flow cytometry profiles and mean fluorescence intensity (Figure 3.4C), I confirmed that the C-lobe of TbpB is both necessary and sufficient for Slam-dependent translocation. This is very interesting, as previous studies have shown that C-lobe of TbpB is conserved between different homologs while the more variable N-lobe binds to the substrate transferrin [80,134].
3.3.5 The last two strands within TbpB contains the translocation motif
With the information that TbpB C-lobe contains the translocation motif, we further investigated the structure of C-lobe of TbpB (Figure 3.4A). Using an analysis of the C-lobe structure, we hypothesized that the translocation motif might lie on the conserved eight-stranded barrel. To test this hypothesis, we made poly-alanine mutations in the first two (B22 and B23) and last two strands (B30 and B31) of the 8-stranded barrel domain within the C-lobe of TbpB (Figure 3.4B).
I tested the translocation of these mutants of TbpB using flow cytometry with a-TbpB antibodies. Poly-alanine mutation to the residues in the first two strands of C-lobe do not change the translocation of TbpB to the surface (Figure 3.4D) that is observed for wildtype TbpB. Compared to the mutations in the first two strands of the barrel, poly-alanine mutations of residues in the last two strands abrogated the translocation of TbpB to the surface (Figure 3.4D).
To ensure that mutations did not affect the expression levels of both the TbpB full length and TbpB C-lobe mutants, I also performed western blot analysis on whole cells (Figure 3.4D, lower panel). While there were no differences for TbpB mutants B22, B23, TbpB mutants B30 and B31
60
150 n.s. A α-TbpB ) I F M (
y
t n.s. i
s 100 - Intact cells n e t n I n.s. e Intensity (MFI) e - Spheroplasts c n e
c *** s e r o
u 50
l ** * F
n a e M
Mean Fluorescenc 0
TbpB sphTbpB
TbpB TbpB23+Slam1 TbpB30+Slam1 TbpB31+Slam1 TbpBwt + SlamTbpB1TbpB22 + Slam1 TbpB22sphTbpB23+SlamTbpB23sphTbpB30+Slam1 TbpB30sphTbpB31+Slam1 TbpB31 1 sphTbpBwt + SlamsphTbp1 B22 + Slam1 + Slam + Slam + Slam + Slam + Slam
B TbpB TbpB TbpB22 TbpB23 TbpB30 TbpB31 + Slam + Slam + Slam + Slam + Slam
Globomycin - + - + - + - + - + - +
TbpB + SP α-TbpB TbpB
Inner Membrane Outer Membrane C 1 2 3 4 5 6 7 8 9 10 11 12
TbpBwt + Slam1 α-SecY
TbpBwt + Slam1 α-OmpA
TbpBwt + Slam1 α-TbpB
TbpB22 + Slam1 α-TbpB
TbpB23 + Slam1 α-TbpB
TbpB30 + Slam1 α-TbpB
TbpB31 + Slam1 α-TbpB
Figure 3.5: Characterization of the mutants in last two strands. (A) Quantification of surface TbpB in intact and spheroplast E. coli cells as analyzed by flow cytometry. Intact and spheroplast cells were labeled with a-TbpB to detect TbpB surface expression. Mean fluorescence was calculated using FlowJo software. Error bars represent the SEM from three biological replicates between spheroplast and intact cells. Statistical significance was determined by ANOVA, *p<0.05; **p<0.01; ****p<0.0001; ns: not significant. (B) Western blot analysis of globomycin treated cells. E. coli cells (transformed as labeled) were treated with globomycin, an SPII inhibitor, leading to the
61
accumulation of TbpB with uncleaved signal peptide (SP). This confirms that all TbpB mutants are effectively processed by Signal peptidase II. (C) Western blot analysis of sucrose gradient ultracentrifugation fractions. E. coli cells expressing different TbpB mutants were lysed and the cellular membrane fraction was subjected to sucrose gradient ultracentrifugation to separate the IM and OM. IM (fractions 3, 4, 5) and OM (fractions 9, 10, 11) were identified by the enrichment of resident proteins, SecY and OmpA, respectively. All TbpB mutants can be found throughout the IM and OM, with the majority in the OM fractions indicating that they all reach the outer membrane, irrespective of lower expression.
exhibited ~50% expression compared to wildtype TbpB. To confirm that TbpB B30 and B31 could still bind to TbpB antibodies, we repeated the flow cytometry experiments with spheroplasts expressing TbpB wildtype and polyalanine mutants (Figure 3.5A). Upon spheroplast formation, a modest increase in signal is obtained for both TbpB B30 and B31 mutants, suggesting that the mutations in the last two strands of the TbpB C-lobe barrel domain do not prevent the antibody from binding to TbpB. To ensure that reduction of TbpB B30 and B31 mutant expression do not affect their transport to the outer membrane in E. coli, we treated the cells expressing TbpB wildtype and mutants with the signal peptidase II-specific inhibitor globomycin. Globomycin treatment leads to formation of a higher molecular weight TbpB band corresponding to unprocessed TbpB in western blots with a-TbpB antibodies [119]. A higher molecular weight band is obtained for all the TbpB mutants including TbpB B30 and B31 mutants suggesting that they are processed by signal peptidase II (Figure 3.5B). Additionally, to test if the TbpB mutants can reach the outer membrane, we used sucrose density ultracentrifugation to separate the inner and the outer membrane. a-OmpA and a-SecY antibodies were used as controls for the outer and the inner membrane proteins respectively. Both TbpB B30 and B31 were observed in the outer membrane suggesting that the mutations do not prevent the translocation of TbpB to the outer membrane (Figure 3.5C). Taken together, this data suggests that the TbpB B30 and B31 are successfully translocated to the outer membrane but are defective in their movement across the outer membrane. This observation suggests that the last two strands may play a role in Slam-mediated SLP translocation.
62
Figure 3.6: Multiple sequence alignment of the last two strands of TbpB. Multiple sequence alignment of experimentally-validated TbpB, LbpB and HpuA homologs from different bacterial species. The strand definition is obtained from A. pleuronemoniae TpbB structure (PDB ID: 4QQ1). The blue highlighted region represents residues conversed between all proteins; red residues are only found on TbpB and LbpB homologs (Slam1 substrates) but not HpuA homologs (Slam2 substrate).
3.3.6 Lumenal facing residues of the two strands of TbpB are important for translocation
Multiple sequence alignment of the last two-strands of the eight-stranded barrel domains were performed and suggested that the specific residues present on the last two strands are conserved amongst all confirmed Slam-dependent surface lipoproteins (Figure 3.6). The conserved residues are [L/M]GGx[F/I/V] on strand B30 and fx[A/T/V][V/S]FG[A/G] on strand B31 with conserved mutations present on the residues facing inside of the barrel. To test this, we made mutations on residues either on inner or outer face of strand B30 (B30in, B30out) and B31 (B31in, B31out). I co- expressed these mutants with Slam1 and performed flow cytometry using a-TbpB antibodies. As predicted, mutations on the inner face of the strand B30 and B31 prevented translocation, while residues on the outer face did not affect the translocation efficiency (Figure 3.4D). Intriguingly, similar to the results obtained for polyAla mutations in B30 and B31, B30in and B31in mutants had lower expression compared to B30out and B31out mutants. Together with the sequence alignment, this data suggests that the last two strands may contain the translocation motif used by the Slam-
63
dependent surface lipoproteins to interact with Slam. However, given that mutations in these strands also affect TbpB expression, the affect may also be due to improper folding of TbpB C-lobe.
A TbpB SP AP N-lobe C-lobe
HpuA SP AP
Nlobe_HpuA SP AP C-lobeN-lobe AP
Clobe_HpuA SP AP C-lobeC-lobeN-lobe AP
HpuA_Clobe SP AP C-lobeN-lobe APC-lobe
B
200 α-TbpB α-HpuA
150 150
100 100
50 50 Mean Fluorescence Intensity (MFI) FluorescenceMeanIntensity Mean Fluorescence Intensity (MFI) FluorescenceMeanIntensity 0 0 Slam - 1 2 - 1 2 - 1 2 - 1 2 Slam - 1 2 - 1 2 - 1 2
TbpB
Nlobe_HpuA Clobe_HpuA HpuA_Clobe Nlobe_HpuA Clobe_HpuA HpuA_Clobe
C
100- 75- α-TbpB 63-
α-His
Figure 2: Domain swapping experiments between TbpB and HpuA. A) The following constructs were Figureused 3 in.7 the: Domain assay: TbpB, swapping HpuA, TbpBexperiments N-lobe with between HpuA (Nlobe_HpuA),TbpB and HpuA. TbpB C-lobe(A) The with following HpuA constructs(Clobe_HpuA) were used and HpuAin the with assay: Clobe TbpB, (HpuA_Clobe). HpuA, TbpB B) The N- lobediferent with TbpB HpuA and (Nlobe_HpuA), HpuA domain swaps TbpB C- lobewere with cloned HpuA and (Clobe_HpuA) expressed either and alone HpuA or with Slam1Clobe or (HpuA_Clobe). Slam2 in E. coli. The(B) cellsThe weredifferent then TbpBlabeled and with α-TbpB and α-HpuA and analyzed by fow cytometry. Mean fuoresence intensity (MFI) plots HpuA domain swaps were cloned and expressed either alone or with Slam1 or Slam2 in E. coli. The obtained for each sample are shown. C) Westerns blots with α-TbpB and α-His antibodies were done cellsby were identifying then labeled the expression with α- TbpBfor each and of theα-HpuA constructs. and analyzed by ow cytometry. Mean fluorescence intensity (MFI) plots obtained from 2 biological replicates for each sample are shown.
64
(C) Westerns blots with α-TbpB and α-His antibodies were done by identifying the expression for each of the SLP and Slam constructs respectively.
3.3.7 C-terminal domain dictates specificity of Slam-dependent SLPs
Intrigued by the role played by C-lobe in dictating translocation of TbpB, we next tested the role of this domain in dictating transporter specificity. To this end, we designed constructs where we swapped the domain of TbpB and HpuA resulting in the following constructs: TbpB N-lobe-HpuA, HpuA-TbpB Clobe and TbpB Clobe-HpuA (Figure 3.7A). We co-expressed each of these constructs in E. coli strain C43 in presence and absence of Slam1 or Slam2 and quantified their surface display using flow cytometry with α-TbpB and α-HpuA antibodies. As predicted from the individual domain experiments, the domain at the C-terminus dictated the specificity (Figure 3.7B) as Slam by each of the construct tested was solely dependent on the C-terminal domain it possessed. Western blots were performed to test the expression of the constructs using α-TbpB and α-His antibodies and demonstrated that the difference in surface display is not due to differences in expression (Figure 3.7C). Taken together with the observation that the residues on the last two strands are important for translocation, our results suggest that the Slam-dependent SLPs are recognized at their C-terminus and the translocation occurs from the C-terminus to the N-terminus.
3.3.8 Trapping TbpB translocation across the outer membrane
As a C-terminally positioned TbpB could potentiate the translocation of HpuA to the surface of the cell in by Slam1, we hypothesized that recognition and translocation must occur from C to N terminus. To validate this hypothesis, we tested the ability of Slam to deliver a rapidly folding soluble domain like GST to the surface. As such, we constructed a GST fusion of TbpB where the GST domain replaced the N-lobe of TbpB (Figure 8A) and transformed the plasmid into E. coli strain C43 with or without Slam1. Interestingly, while flow cytometry with α-TbpB antibodies showed that the GST-C-lobe could still be detected on the cell surface (Figure 3.8B) similar to full length TbpB, flow cytometry with α-GST antibodies revealed that the GST-tag is not present on the surface of the cell (Figure 3.8B).
To confirm that lack of GST tag on surface is not due to inability of α-GST antibodies to detect GST on cells, we re-performed flow cytometry assay on E. coli spheroplast cells that contain a
65
permeabilized outer membrane (Figure 3.9A-C). Upon permeabilization, an increase signal is obtained for GST suggesting α-GST antibodies are functional. Additionally, to confirm the stalled transport of GST-C-lobe in Neisseria, Christine complemented N. meningitidis strain
A SP AP N-lobe C-lobe TbpB
1 21 61 352 599
SP AP GST C-lobe GST-C-lobe
1 21 61 305 552
SP AP GST GST
1 21 61 305
B C GST GST-C-lobe GST GST-C-lobe + Slam1 + Slam1 + Slam1 + Slam1 GST-C-lobe GST-C-lobe + Slam1
α TbpB Input Pulldown Input Pulldown Input Pulldown Input Pulldown α Slam1 α Slam1
BamA Count α BamA α
α TbpB α TbpB
PE Fluorescence GST- GST- C-lobe α GST C-lobe
α GST α GST Count
AP-GST AP-GST
PE Fluorescence GST IP Slam1 IP
Figure 3.8: Slam1 interacts with TbpB in E. coli. (A) Schematic representation of TbpB constructs used in the translocation assays. TbpB has a signal peptide (SP), anchoring peptide (AP), N- and C-lobes. GST-TbpB chimeric fusion constructs were designed as illustrated. (B) Flow cytometry profiles of E. coli cells expressing GST-C-lobe with or without Slam1 using α-TbpB or α-GST antibodies. Cell counts are shown against PE fluorescence for each antibody. One representative histogram of three independent experiments is shown. (C) Interaction assay of GST- C-lobe with Slam1. E. coli cells transformed with GST-C-lobe or GST and Slam1 were immunoprecipitated with glutathione resin (left) or α-Slam1 antibodies conjugated to sepharose beads (right). Whole cell lysate (Input) and eluted fractions (Pulldown) were subjected to western blot analysis probed with α-Slam1, α-BamA, α-TbpB and α-GST antibodies. Blots are representative of at least three independent experiments. (Adapted from Hooda et al Nat Microbiology 2016)
66
B16B6 Dtbpb with full length TbpB, GST-C-lobe, or GST fused to the TbpB signal and anchor peptide. Confirming the E. coli experiments, the C-lobe was surface exposed while the GST was not [49]. Collectively, these findings demonstrate GST-TbpB is not fully delivered to the surface and we have obtained a GST-TbpB chimera that is partially stalled at the outer membrane translocation step. Moraes - Figure S7 A
Intact E. coli cells
Spheroplast E. coli cells
A Inner AOuter Inner A Outer Inner Outer Inner membrane membrane membrane membrane membrane membranemembrane
tions: 1 2 3 4 5Fra6ctions:7 8 19 210311412 513Fra6ctions:7 8 19 210 311 412F135ractions:6 7 81 92 103 114 125 136 7 8 B n.t. αTbpB αTbpB GST-C-lobe αTbpB Intact cells Spheroplasts Intact cells Spheroplasts α TbpB α TbpBαLepB α GSTαLepB α GST αLepB
αOmpA αOmpA αOmpA Counts Counts Counts Counts
αSlam1 αSlam1 αSlam1
PE Fluorescence PE Fluorescence PE Fluorescence PE Fluorescence anti-TbpB anti-TbpB anti-TbpB anti-GST anti-GST anti-GST
CB B B α TbpB α TbpB α GST α GST
y y y y y
α PgaB y t t t α PgaB α PgaB t e e e e t i 80 i 80 i 80 t
i 20
i 20
i 20 s s s s s s n n n n n n e e e e e e t t t t t t en c en c en c en c n n n n n n I I I I I
60 60 60 I 15
15
15
e e e e e e c c c c c c es c es c es c es c n n n n n n r r r r e e e e e e c c c c c 40 40 40 c 10 10 10 s s s s s s e e e e e luo e luo luo luo r r r r r r F F F F o o o o o o u u u u u u l l l l l 20 20 20 l 5 5 5 F F F F F F
n n n n n n ean a a ean ean ean a a a a e e e e e e 0 M Intensity (MFI) 0 0 M M 0 0 Intensity (MFI) 0 Intensity (MFI) M M Intensity (MFI) M M M M M T T T T T T W W W W W W n.t. n.t. n.t. n.t. sph-WT sph-WT sph-WT sph-WT sph-WT sph-WT
WT+pGST-ClobGST-C-lobee WT+pGST-Clobe WT+pGST-ClobGST-C-lobee GST-C-lobeWT+pGST-Clobe WT+pGST-ClobGST-C-lobee WT+pGST-Clobe sph-WT+pGST-Clobe sph-WT+pGST-Clobe sph-WT+pGST-Clobe TamB KO sph-WT+pGST-Clobe TamBsph-WT+pGST-Clob KO e sph-WT+pGST-Clobe FigureC S7: Validation of α-GST antibodiesC for fow cytometry.C (A) Characterization Wof TE. coli spheroplastsTamA KO using fowWT cytometry.TamA ToKO confrm theW activityT ofTamA polyclonal KO α-GST activity inTamA KO Figure 3fow.9: cytometry, Validation E. coli strain of α C43-GST non-transformed antibodies (n.t.) for cells flowand cells cytometry. expressing GST-C-lobe (A) Characterization were treated with of E. coli lysozyme and EDTA to obtain spheroplasts. The fow cytometry plots (forward scatter (FSC) vs side scatter (SSC)) of intact spheroplastsand spheroplast using flow E. coli cellscytometry. are shown. AnTo increase confirm in forward the activityscatter was ofobserved polyclonal upon formation α-GST of spheroplasts. activity in (B) flow cytometry,Detection E. coli of TbpB strain and GST C43 in intactnon -andtransforme spheroplast dcells. (n.t.) Intact cells and spheroplasts and cells cells, expressing transformed GSTor not,- asC indicated,-lobe were ts ts ts ts ts ts ts ts ts were labeled with α-TbpB or α-GST. Plots of cell counts against PE fuorescence are shown. Upon disruption of thets ou n
treated with lysozyme and EDTA to obtainou n spheroplasts. The flow cytometryou n plots (forward scatter ou n ou n ou n ou n C C C outer membrane, increase in fuorescence signals for both TbpB and GSTC were observed in spheroplasts expressing C C C ou n ou n ou n C C (FSC) vsGST-C-lobe side scatter compared (SSC)) to n.t. cells. of intact(C) Quanti andfcation spheroplast of accessible GSTE. coliand TbpB cells in GST-C-lobe are shown. expressing An increaseand n.t.C intact in forwardcells scat andter spheroplasts. was observed Flow cytometry upon resultsformation are shown of asspheroplasts. mean fuorescence (B) intensity Detection (MFI) and of error TbpB bars represent and GST in standard error of the mean (SEM) from three experiments. These results verify the activity of the polyclonal α-GST intact andantibodies spheroplast for fow cytometry cells. Intact. and spheroplasts cells, transformed or not, as indicated, were labeled with α-TbpB or α-GST. Plots of cell counts against PE fluorescence are shown. Upon
67
disruption of the outer membrane, increase in fluorescence signals for both TbpB and GST were observed in spheroplasts expressing GST-C-lobe compared to n.t. cells. (C) Quantification of accessible GST and TbpB in GST-C-lobe expressing and n.t. intact cells and spheroplasts. Flow cytometry results are shown as mean fluorescence intensity (MFI) and error bars represent standard error of the mean (SEM) from three experiments. These results verify the activity of the polyclonal α-GST antibodies for flow cytometry. (Adapted from Hooda and Lai et al. Nat Microbiology 2016)
A number of key conclusions can be derived from this result. First, given that TbpB C-lobe could be detected on the surface while the GST could not, confirms that the Slam-dependent translocation of SLPs occurs from the C-terminus to the N-terminus. Second, given that GST folds in the periplasm and is unable to get to the surface suggests that Slam substrates are most likely kept unfolded in periplasm for their translocation across the outer membrane. Certain periplasmic chaperones may help in keep Slam-dependent SLPs unfolded similar to ones (Skp, SurA and DegP) that are used by outer membrane proteins, however no such chaperones have been studied in the context of SLPs to date. Third, the GST-C-lobe chimera provide a great model for investigating the interactions between Slams and SLPs. Evident from the lack of remaining SLPs after proteinase K shaving in Neisseria, we suspected that the SLPs spend minimal time transiting through the periplasm and interacting with their translocon. In fact, co-immunoprecipitations using α-Slam antibodies in N. meningitidis or E. coli were not fruitful in revealing a direct interaction with the SLPs. We hypothesized that the GST-C-lobe construct that is stalled in the outer membrane would be trapped with its transporter in a similar manner to the stalling of the outer membrane translocon FhaC [152]. Hence, to determine if Slam1 interacts with the stalled GST-C-lobe chimera in the outer membrane, immunoprecipitations were performed with α-Slam1 and α-GST antibodies in E. coli (Figure 3.8C). The interaction between the C-lobe of TbpB and Slam1 was observed and is specific, as lysates from cells expressing lipidated-GST and Slam1 do not immunoprecipitate together. Collectively, this illustrates that TbpB interacts with Slam1 at some point during its transit across outer membrane
3.3.9 Role of Omp85 family members in Slam function
Trapping the GST-C-lobe construct also allows us to test if other outer membrane proteins interact with the SLP during its transit across the OM. To this end, we wanted to test if Omp85 family members BamA and TamA play a role in SLP translocation [8,157]. Both of these proteins are
68
involved in biogenesis of outer membrane proteins and BamA has also been found to be important for the biogenesis of the SLP, RcsF [157]. To examine the role of BamA, we used BamA antibodies to test if it co-immunoprecipitates Slam1 and GST-C-lobe (Figure 3.8C). No BamA was observed suggesting that BamA is not part of the trapped complex with GST-C-lobe. To examine the TamA, E. coli strains lacking either TamA or TamB (ytfM and ytfN) were tested for surface display of TbpB using flow cytometry, which demonstrated that neither Tam components are required to translocate TbpB to the surface in a Slam-dependent manner (Figure 3.10).
WT TamA KO TamB KO
α TbpB α TbpB α TbpB Counts Counts Counts
PE Fluorescence PE Fluorescence PE Fluorescence
TbpB TbpB + Slam1
Supplementary Figure 9: Analysis of TbpB surface exposure in TamA or TamB-defcient E. coli. FigureTransport 3.10 of TbpB: Analysis to the surface of TbpB of E. coli surface is quanti exposurefed with fow in cytometry TamA inor E. TamB coli strains-deficient BW25113 WT,E. coliΔytfm. Transport(TamA ofKO) TbpB or Δ ytfnto the(TamB surface KO). For of each E. sample,coli is cellquantified counts are with shown flow against cytometry PE fuorescence. in E. Results coli strains demonstrate BW 25113that WT, TbpB is transported to the surface of E. coli at similar levels in a Slam1-dependent manner even in the absence Δytfmof TamA (TamA or TamB. KO) This orsuggests Δytfn that (TamB the translocation KO). For of each TbpB sample, to the surface cell of counts E. coli cells are does shown not require against the PE Tam- fluorescence.translocation system. Results demonstrate that TbpB is transported to the surface of E. coli at similar levels in a Slam1-dependent manner even in the absence of TamA or TamB. This suggests that the translocation of TbpB to the surface of E. coli cells does not require the Tam-translocation system. (Adapted from Hooda and Lai et al. Nat Microbiology 2016)
3.3.10 Cross-linking experiments with Slam mutants
Of the outer membrane proteins that we have analyzed so far, Slams are the only proteins that interact with SLPs in the outer membrane. Hence, we hypothesize that the Slam membrane domain forms a conduit in the outer membrane for the transport of SLPs across the outer membrane. Based on this hypothesis, we predicted that the GST-C-lobe traverses the outer membrane through the
69
Slam1 membrane domain. However, our attempts to purify the Slam and GST-C-Lobe complex from the outer membrane were unsuccessful and the complex fell apart during size-exclusion chromatography (Figure 3.11A). This suggests a weak interaction between GST-C-Lobe and Slam1.
Figure 3.11: Crosslinking and studies on stalled Slam-GST-C-lobe complex. (A) Purification of Slam1 and GST-C-lobe complex. Glutathione beads were used to purify the complex between GST- C-lobe and Slam1. The complex only contained GST-C-lobe and Slam1, however upon running on
70
the Sepharose S200 column, the GST-C-lobe and Slam1 ran separately, indicating a weak complex. (B) Mutations on GST-C-lobe and Slam1. Cysteine residues were introduced at the interdomain linker of the GST-C-lobe to make construct GST-C-lobe-cys. Cysteine mutations were also introduced at various position on the membrane domain of Slam1 at residues predicted to be present inside the lumen of the barrel. (C) Crosslinking studies between Slam1 mutants and GST-C-lobe- cys. The different combination of constructs as listed were co-expressed in E. coli strain C43 and treated with CuSO4, an oxidation agent to promote disulphide bond formation. Different mutants were screen on western blots in the presence and absence of reducing agent DTT to look for a formation of a large molecular weight complex. Results are shown for Slam1 D224C and Slam1 S180C and the corresponding controls are shown. No DTT sensitive complex was observed.
In order to stabilize the complex, we decided to introduce cysteine residues on GST-C-lobe and Slam1 such that they form an intermolecular disulphide bridge[152]. From our topology experiments, we predicted that linker between the GST and C-lobe will form the transmembrane region and hence we introduced a cysteine residue in that region (GST-C-lobe-cys) (Figure 3.11B). For Slam1, we selected specific residues in the membrane domain that were predicted to face the inside of the barrel based on secondary structure prediction tools (Figure 3.11B). We confirmed that the Slam1 mutations did not cause any effect on surface display of TbpB using the E. coli translocation assay. Next, we co-expressed the Slam1 mutants with GST-C-lobe-cys and looked for the formation of DTT-resistant higher molecular weight bands that would correspond to GST-C- Lobe-cys and Slam1 complex formed through a disulphide bridge. However, we were unable to detect an intermolecular disulphide bond between Slam1 and GST-C-lobe-cys (Figure 3.11C). This suggests that either the GST-C-lobe-cys is not positioned in the outer membrane or different residues should be selected on Slams, that would permit the formation of the disulphide bond. Alternatively, different chemical crosslinkers can also be used to probe this further [158].
3.4 Discussion
Reconstitution of Slam-dependent SLP translocation in E. coli provided a simplified system for probing the mechanism of translocation in greater detail. E. coli is the most well-studied model system for Gram-negative bacteria with a number of plasmids and reagents available that were critical for this study. These included plasmids with different promoters and antibiotic resistance cassettes, Keio collection for studying the effect of different non-essential genes, which were extensively used in this study.
71
The first question we attempted to answer in the chapter was regarding the secretion motif present on Slam-dependent SLPs. To identify this region, we decided to focus on SLPs TbpB and HpuA. Using these SLPs, I was able to show that Slams exhibit specificity, where Slam1 translocates TbpB, LbpB and HpuA, while Slam2 specifically potentiates the translocation of HpuA. In the western blots, we observed variable levels for SLP upon co-expression with Slam constructs (especially for fHbp and HpuA). No correlation between surface display and SLP expression was observed, suggesting that this change in expression does not affect the results. This variation could be due to the competition for the protein synthesis and the translocation machinery that are used by both Slams and SLPs for their biogenesis. Next, using domain truncations of TbpB, I was able to show that the translocation motif lies on the C-lobe of TbpB. Introduction of poly-alanine mutations in different strands demonstrated that the last two strands of TbpB are crucial for translocation. The last two strands were found to contain conserved motifs: [L/M]GGx[F/I/V] on the second last strand and fx[A/T/V]FG[A/G] on the last strand. Interestingly, the most conserved residues were present on the inner face of the barrel and these were also found to be important for Slam-dependent translocation. While, I was able to show that the poly-alanine mutants in the last two strands are effectively processed by globomycin and can reach the outer membrane, their expression is reduced compared to TbpB wildtype. The lower expression of mutants indicate that results obtained could also be due to improper folding of TbpB C-lobe in the mutants. Hence, to rule out that the expression difference explains their limited surface display profile, additional mutations should be made to identify mutants whose expression is similar to wildtype, but they are inefficient in their surface display. One effective strategy may be to make point mutations in the last two strands, specifically at the residues facing the inner face that are conserved amongst Slam-dependent SLPs. Upon identification, it is also important to ensure that the motif is indeed required for the interaction with Slams. To this end, photoactivable non-canonical amino acids can be introduced at different positions on the last two strands and cross-linking experiments can be performed to show a direct interaction between Slams and the last two strands of the SLPs [158].
Using domain swaps between TbpB and HpuA, we were also able to identify regions that are important for mediating substrate specificity. Swapping the TbpB- C-lobe with HpuA changed the specificity of the construct to be translocated by HpuA. On the other hand, when the chimeric protein constructs HpuA-C-lobe and C-lobe-HpuA were tested, the translocation and surface display of these proteins was dictated by the domain that resided at the very C-terminus. Combined
72
with the mutational analysis, we predict that the last two strands also mediate transporter specificity exhibited by Slam-dependent SLPs. To investigate this further, swaps between the last two strands between TbpB and HpuA may be performed to see if their transporter specificity can be switched. Hence, careful analysis and swap of specific residues between TbpB and HpuA could help in identification of residues that are important for substrate specificity.
The ability of Slams to deliver proteins onto the surface of E. coli can have applications in development of an alternative bacterial cell surface display system [146]. Currently most bacterial cell surface display systems rely on outer membrane proteins such as porins or autotransporters. Surface lipoproteins owing to their peripheral attachment and higher mobility on the surface provides an attractive target for the display of cargo proteins. Motivated by the observation that the C-lobe is sufficient for recognition by Slam1, we examined its ability deliver different protein domains to the surface of E. coli. To this end, we replaced the TbpB N-lobe with a GST tag and performed flow cytometry to test the surface display of GST tag. Interestingly, while TbpB C-lobe could be identified on the cell surface, the GST was retained inside. This may be due to the fact that GST rapidly folds in the periplasm, preventing its movement across the outer membrane indicating a restriction on the cargo that can be displayed on the surface in a Slam-dependent manner. This also suggests the presence of chaperones may prevent the premature folding of SLPs in the periplasm. Further understanding of the mechanism of Slam will allow us to design an efficient system for the delivery of a wider range of cargo to the surface of bacteria.
Obtaining a construct that was stalled in the outer membrane provided a system to test proteins that interact with SLPs in the outer membrane. Not surprisingly, we could identify that Slam1 interacts specifically with GST-C-lobe. We also tested for other protein complexes such as the Bam complex however, to date Slam1 is the only outer membrane protein that we have been able to identify in trapped intermediate complex. To test the role of the Tam complex, we utilized the Keio collection, a library of mutants where each of the non-essential genes in E. coli have been removed. By co- expressing TbpB and Slam1 in E. coli lacking TamA or TamB, we could show that the Tam system does not play a role in SLP transport. Similar studies can also be used to study the role of periplasmic chaperones such as Skp, SurA and DegP that are involved in OM protein biogenesis and might play a role in SLP biogenesis [6]. We can also test the role of non-essential components of the Bam complex (Bam B, C and E) and test if they contribute to SLP biogenesis [7].
73
On the Slam side, domain truncations showed that membrane domain is sufficient for Slam function. The membrane domain of Slam1 also retained its specificity for TbpB indicating that residues dictating substrate specificity on Slams lie in the membrane domain. To identify residues, we next made point mutations in the Slam membrane domains. Charged residues that potentially face the lumen of the Slam barrel were chosen for mutagenesis. The Slam mutants and GST-C- lobe-cys were co-expressed in E. coli and cross-linked complexes were tested using western blots. No cross-linked complex was observed. Additional mutations should be screened to identify putative regions. Another key limitation might be the position of the cys residue on the GST-C- lobe-cys. Chemical cross-linkers can also be used to increase the efficiency of crosslinked products. A final alternative methods might be to use photoactivable non-canonical amino acids that form cross-links upon irradiation with UV [159].
Finally, while we have clearly demonstrated through the E. coli translocation assay that Slams are necessary for SLP translocation. Thus, the question remains: Are Slams the outer membrane translocons for SLPs? While we have ruled out the role of Tam complex in SLP biogenesis, the Bam complex cannot be ruled out as BamA and D are essential for E. coli survival and hence cannot be deleted. This suggests that alternative assays would be required to flesh out this question and these approaches will be discussed in the next chapter.
74
4 Chapter 4 Investigating the molecular details of Slam function using an in vitro reconstitution of translocation
Acknowledgements: Mr. Sang Huynh performed the fluorescent plate reader assay for testing Slam specificity. He also helped in the development of the translocation assay as well in the use of assays to demonstrate specificity and Slam sufficiency. Ms. Christine Lai cloned many of the constructs used in the study. Mr. Ashutosh Gupta helped in purification of Mcat Slam1. OmpA antibodies were obtained from Dr. Jan Willem deGier, Stockholm University. Plasmid for expression of the Bam complex (BamA-E) was obtained from Dr. Harris Bernstein at the NIH.
4.1 Overview
One of the central remaining question in SLP biogenesis is what happens after the interaction with Slams i.e. how does the translocation proceed once the SLPs comes in contact with Slams in the outer membrane? In the introduction, I proposed two models that explain how SLP translocation may happen: 1) Translocon model where the Slam membrane domain acts as the translocating channel, or 2) Chaperone model where SLPs are transferred from Slam to another protein/complex that forms the channel for transport. However, the E. coli translocation assay could not provide a complete answer to this question as there are many homologous proteins between Neisseria and E. coli that could potentially contribute to the translocation process.
In the literature, in vitro translocation assays have been effective for investigating the translocation of proteins across cell membranes [151,160]. Hence, we wanted to test if such an assay could be developed for investigating Slam-dependent SLP translocation. To this end, we purified Slam1 protein and inserted it into synthetic lipid vesicles or liposomes. As controls we also purified and inserted the Bam complex into liposomes [161]. Then we incubated the Slam1 and Bam proteoliposomes with purified LolA and E.coli spheroplasts (an E. coli cell bound by its plasma membrane lacking the outer membrane) expressing TbpB [162]. We hypothesized that addition of LolA to spheroplasts should release TbpB into the supernatant from the LolCDE inner membrane
75
complex [20], where it will get transported into liposomes containing Slams, if Slams are the translocons. To test TbpB insertion into liposomes, we removed the spheroplasts through centrifugation and then added proteinase K to degrade all proteins, leaving those present and protected within the liposomes. Indeed, we found that TbpB was transported into the Slam1 proteoliposomes confirming that Slams are necessary and sufficient of translocation. Additionally, we found that Slam2 was also able to translocate its substrate HpuA into proteoliposomes. Finally, Slam1 and Slam2 retained the specificity for their substrates in the assay indicating that substrate specificity is inherent in the Slams.
Intrigued by robustness and self-sufficiency of Slams, we next examined the substrate requirements of Slams i.e. are SLPs sufficient for translocation by Slams? The Slam membrane barrel is predicted to be composed of 14-strands and is not sufficient to accommodate a folded SLPs. Hence, we predicted that the SLPs are kept unfolded by periplasmic factors before its translocation through the outer membrane. To test this question, we purified TbpB (without the anchor peptide and the lipid anchor [111]) and incubated with 8M Urea to unfold the polypeptide. We then added the unfolded TbpB to the Slam containing liposomes and tested its translocation by determining its protection from proteinase K. TbpB could be detected inside the Slam liposomes indicating that the minimal requirement of Slams is an unfolded TbpB-like protein. These findings also suggest that the lipid anchor and anchor peptide are not essential for movement through the Slam translocon. Addition of periplasmic chaperones may increase the efficiency of insertion, but these factors remain unknown. Taken together, our study has shown that Slams are both necessary and sufficient for translocation of TbpB-like proteins across the outer membrane.
4.2 Methods
4.2.1 Bacterial strains and growth conditions
Strains used in this study are summarized in Appendix C. E. coli were grown in LB media containing antibiotics when necessary – 50 µg/mL kanamycin and 100 µg/mL ampicillin. Cloning procedures were carried out using E. coli MM294 competent cells. Protein expression was performed using E. coli C43 (DE3) cells for all Slam homologs, Bam complex and the translocation experiments. E. coli BL21 (DE3) cells were used for purification of LolA and TbpB.
76
4.2.2 Cloning of Slam, SLP homologs Genes were cloned into expression vectors by RF cloning [120]. Signal peptides and tags were inserted using round the horn cloning [121]. pET52 Nme HpuA was made by amplifying hpua from N. meningitidis strain B16B6 and inserting it into an empty pET52b vector. pET52 Nme HpuA-flag was made by addition of a flag-tag at the C-terminus of the hpua gene in pET52 Nme HpuA. pET26 Ngo Slam2 construct was obtained by cloning the mature N. gonorrhoeae strain MS11 gene ngfg_00064 and inserting into empty pET26b vector. The lola gene was cloned from E. coli strain C43 (DE3) genome into an empty pET28a vector with an N-terminal 6xHis tag to make pET28 LolA. The constructs used in this study are summarized in Appendix C.
4.2.3 High-throughput assay for Slam function
Pairs of Slams and SLPs were co-transformed into E. coli C43(DE3) cells. Cells were grown overnight in autoinduction media [122] with ampicillin and kanamycin. Cells were harvested from the overnight culture at 1500×g, 3 mins. Cell pellets were washed gently with PBS + 1mM MgCl2 before incubating with the a-Flag antibody (1:500 dilution). After 1-hour incubation with the primary antibody, cells were harvested and washed with PBS + 1mM MgCl2. The cells were then incubated with secondary goat anti-rabbit PE antibody for 1 hr. Cells were again harvested, washed and resuspended in PBS + 1 mM MgCl2. The samples were aliquoted in a 96-well plate and read on a Synergy 2 (BioTek) plate reader at 488 nm and 575 nm. OD600 was also recorded for data normalization.
4.2.4 Purification of Slams E. coli strain C43 (DE3) with pET26 Mcat Slam1 were grown overnight at 37°C in LB + ampicillin. The cells used to inoculate (1:1000) into 6L of autoinduction media + kanamycin. Cells were grown at 20°C for 48 hours. Cell pellets were resuspended in 20 ml/L of 50 mM Tris–HCl pH 8, 200 mM NaCl and cells were lysed using an EmulsiFlex C3 (Avestin). Lysates were centrifuged at 35000×g at 4°C for 10 min. The supernatants were centrifuged at 185,000×g for 1 hour at 4 °C to isolate cell membranes. Membrane pellets were homogenized and incubated in 15 ml/L of 50 mM Tris pH 8, 200 mM NaCl, 3% Elugent overnight at 4 °C. The centrifugation step was repeated to remove insoluble pellets. Supernatants containing the soluble membrane proteins were then incubated with 1 ml Ni-NTA agarose O/N at 4 °C. Ni-NTA beads were washed three times with 10 column volumes of buffer A (20 mM Tris pH 8, 100 mM NaCl, 0.03% DDM) containing increasing
77
concentration of imidazole. Mcat Slam1 was then eluted in buffer A containing 200 mM imidazole. The protein sample was exchanged into low salt buffer (20 mM Tris pH 8, 20 mM NaCl, 0.03% DDM) using a PD-10 column (GE Healthcare) and then injected onto a MonoQ column (GE Healthcare) equilibrated with low salt buffer. The column was washed with increasing concentration of salt using a high salt buffer (20 mM Tris pH 8, 2 M NaCl, 0.03% DDM). Fractions that contained pure Mcat Slam1 were identified using SDS-PAGE gels, pooled, concentrated and stored at -80oC.
For Ngo Slam2 purification, the protocol described expression and purification for Mcat Slam1 was followed. Fractions that contained pure Ngo Slam2 were identified using SDS-PAGE gels, pooled, concentrated and stored at -80 °C.
4.2.5 Purification of Bam complex E. coli strain C43 (DE3) cells with pJH114 were grown overnight at 37 °C in LB + ampicillin. The cells used to inoculate (1:1000) into 6L of autoinduction media + ampicillin. Cells were grown at 20 °C for 48 hours. Cell pellets were resuspended in 20 ml/L of 50 mM Tris–HCl pH 8, 200 mM NaCl and cells were lysed using an EmulsiFlex C3 (Avestin). Lysates were centrifuged at 35000×g at 4 °C for 10 min. The supernatants were centrifuged at 185,000×g for 1 hour at 4 °C to isolate total membranes. Membrane pellets were homogenized and incubated in 15 ml/L of 50 mM Tris pH 8, 200 mM NaCl, 3% Elugent overnight at 4 °C the centrifugation step was repeated. Supernatants containing the soluble membrane proteins were then rotated in the presence of 1 ml Ni-NTA agarose overnight at 4 °C. Ni-NTA beads were washed with one column volume with buffer A containing increasing concentrations of imidazole. BamABCDE was then eluted in buffer A containing 200 mM imidazole. The protein sample was concentrated and injected onto a S-200 column equilibrated with buffer A. Fractions that contained complete BamABCDE complexes were identified using SDS-PAGE gels, pooled, concentrated and stored at -80 °C.
4.2.6 Liposome and proteoliposome preparation
100 mg of E. coli polar lipid extract (Avanti) was resuspended in chloroform (Sigma). The lipid solution was then dried off under N2 gas and resuspended in 10 mL of buffer B (50 mM Tris-HCl pH7, 200 mM NaCl). The solution was flash frozen and thawed at least 5 times and stored at -80 °C as 10 mg/mL stock. 1 mL of the liposome solution (10mg/mL) was extruded through a 0.2 µm filter
78
(Whatman) to make unilamellar liposomes. The extruded solution was split and purified outer- membrane proteins (Bam complex and Slam1&2) were added into the liposomes solutions at 1.5 µM for Bam and 15 µM for Slam1&2 such that they were diluted 1:5. 100 µg of Biobeads SM-2 (BioRad) were added to remove detergent and promote protein insertion into liposomes. Tubes were sealed with parafilm and kept at room temperature with gently end-to-end rotation for ~ 2 hours. The proteoliposomes were then incubated at 4 °C overnight with end-to-end rotation, separated from Biobeads and spun down at 18000×g at 4 °C for 5 mins. The supernatant was kept at 4 °C and used for the experiments within the week. The insertion of Slam1&2 and the Bam complex was assessed by SDS-PAGE gels, silver stain and western blots with a-His antibody.
4.2.7 Purification of LolA
E. coli BL21 (DE3) cells expressing pET28 LolA were grown overnight in 20 mL of LB + ampicillin at 37 °C and used for inoculating 2 L of 2YT media. The cells were grown at 37 °C to an
OD600 ~ 0.6, induced with 1mM IPTG and then incubated overnight at 18 °C. The cells were harvested the next day by centrifugation at 12200×g for 20 mins at 4 °C. The pellets were resuspended in buffer B (50 mM Tris-HCl pH 7, 200 mM NaCl). Cell lysis was performed using EmulsiFlex C3 (Avestin). The cell lysates were spun down at 35000×g at 4 °C for 50 mins to remove cell debris. Supernatant was filtered through 0.22 µm filter and incubated with 1mL Ni- NTA beads for 2 hrs at 4°C with gentle stirring. Solution was applied to a column and the Ni-NTA beads were subsequently washed 3 times in buffer A with increasing concentration of imidazole (10 mM, 20 mM and 40 mM). Proteins were eluted from Ni beads by adding buffer B, 200 mM imidazole. The purified proteins were dialyzed overnight in buffer B at 4 °C. The purity of LolA was accessed on SDS-PAGE and pure LolA fractions were concentrated and stored at -80 °C.
4.2.8 Spheroplast release assay
The protocol was adapted from Fan et al. 2012 with a few modifications [151]. Briefly, spheroplasts were obtained from E. coli C43(DE3) cells transformed with either pET52 Mcat TbpB-flag or pET52 Nme HpuA-flag. The cells were grown overnight in (LB or autoinduction media with ampicillin). E. coli cells were adjusted to have OD600 ~ 1.0. The cells were harvested by spinning at 6800×g for 2 mins at 4 °C. The pellets were then resuspended in 100µL of sucrose buffer containing 50 mM buffer B and 0.5 M sucrose. The resuspended solutions were kept on ice and converted to spheroplasts by adding 100µL of buffer containing 0.2 mg/mL lysozyme and 8
79
mM EDTA with gently inversion for mixing. The solutions were incubated on ice for at least 20 mins. The spheroplasts were collected by spinning at 10,000×g for 10 min and resuspended in 100 µL of M9 minimum salt media containing 1×M9 salts, 2% glucose and 0.25 µM sucrose. Expression of SLPs was induced by addition of 1 mM IPTG and spheroplasts were incubated at 37 °C. 10 µM of E. coli LolA was added in selected samples to promote the release of SLPs from the spheroplasts. Samples were collected at different time points and spun down at 18000×g for 10 minutes at 4 °C to remove spheroplasts. Supernatant at different time points were mixed with SDS loading buffer and run on an SDS-PAGE gel. Western blot analysis using a-Flag antibody to estimate the quantity of TbpB and HpuA released by spheroplasts upon the addition LolA.
4.2.9 Bam complex functional assay
To test the activity of the Bam complex, we examined the ability of Bam proteoliposomes to potentiate the insertion of spheroplast released OmpA. E. coli strain C43 (DE3) cells were grown overnight in LB. Cells were adjusted to have OD600 ~ 1.0 and converted to spheroplast using the protocol described in Section 2.8. Spheroplasts were incubated with Bam or empty proteoliposomes and incubated at 37 °C for 15 mins. Samples were spun down at 18000×g for 10 mins at 4 °C to remove spheroplasts. Supernatants were mixed SDS loading buffer and run on an SDS-PAGE gel. Western blot analysis was performed using a-OmpA antibody to estimate the amount of OmpA inserted into the Bam/empty proteoliposomes.
4.2.10 Co-secretion translocation assay
To develop the co-secretion translocation assay, we followed the protocol described in Section 2.8 for the generation of spheroplasts. Upon the formation, spheroplasts were collected by spinning at 10,000×g for 10 minutes and resuspended in 100 µL of M9 minimum salt media with 1×M9 salts, 2% glucose, 0.25 µM sucrose, 10 µM of E. coli LolA, and 50 µL of Empty, Bam, Slam1&2 and Bam+Slam1&2 proteoliposomes. Expression of SLPs was induced by the addition of 1mM IPTG and incubation at 37 °C for 15 mins. Spheroplasts were spun down at 18000×g for 10 mins at 4 °C. Supernatants were collected and treated with the final concentration of 0.5 mg/mL proteinase K in the presence/absence of 1% TritonX-100 and incubated at 37oC for 1 hr. 4 mM PMSF was added to inactivate proteinase K and samples were loaded on SDS-PAGE gels followed by western blots with a-Flag antibodies.
80
4.2.11 Post-secretion translocation assay
To develop the post-secretion translocation assay, we followed the protocol described in Section 2.8 with a minor modification. Upon the formation, spheroplasts were collected by spinning at 10,000×g for 10 mins and resuspended in 100 µL of M9 minimum salt media with 1×M9 salts, 2% glucose, 0.25 µM sucrose, 10 µM of E. coli LolA. Expression of SLPs was induced by the addition of 1mM IPTG and incubation at 37 °C for 15 mins. Spheroplasts were spun down at 18000×g for 10 minutes at 4 °C. 450 µL of supernatant obtained was incubated 50 µL of Empty, Bam, Slam1&2 and Bam+Slam1&2 proteoliposomes for additional 15 mins at 37 °C. The samples were then used for proteinase K shaving assay as described in the previous section.
4.2.12 Purification of TbpB
E. coli expressing His-MBP-B16B6 TbpB was grown in 60 mL of LB + ampicillin as the starter cultures overnight at 37 °C and used for inoculation of 6 L of 2YT media. The cells were grown at
37 °C till the OD600 reached 0.6 and then induced with 1 mM IPTG and left for expression overnight at 18 °C. The cells were harvested the next day by centrifugation at 12200×g for 20 mins at 4 °C. The pellets were resuspended in buffer B. Cell lysis was performed using EmulsiFlex C3 (Avestin). The cell lysates were spun down at 35000×g at 4 °C for 50 mins to remove cell debris. Supernatant was filtered through a 0.22 µm filter and incubated with 3 mL Ni-NTA beads for 2 hrs at 4 °C with gently stirring. The filtered solution was passed through a gravity column and the Ni- NTA beads were washed 3 times in buffer B with increasing concentrations of imidazole. Proteins were eluted from Ni-NTA beads by adding buffer B and 200mM imidazole. The purified proteins were further incubated with 500 µL of 4.5 mg/mL TEV and dialyzed overnight in 50 mM Tris pH 7, 20 mM NaCl at 4 °C. Cleavage was confirmed by Coomassie stained SDS-PAGE. The protein sample was run through MonoQ (GE Healthcare) column and TbpB was collected from the flow through fraction. This fraction was concentrated using 30,000 kDa MWCO concentrator and injected on an S200 Increase size exclusion column equilibrated with buffer B. The pure TbpB samples were pooled, concentrated and stored at -80 °C.
4.2.13 Translocation assay with purified TbpB
To develop the translocation assay with purified TbpB, we diluted TbpB to 6 µM in buffer A or 8 M urea. The TbpB samples were rapidly diluted 1/12 into 50 µL of Empty, Bam, Slam1&2 and
81
Bam+Slam1&2 proteoliposomes to bring the final concentration of TbpB to 0.5 µM and urea to 0.66 M. The samples were incubated for 15 min at 37 °C. The samples were then used for proteinase K shaving assay as described in the previous sections.
4.3 Results
4.3.1 Selection of Slam homologs for purification
To establish an in vitro translocation assay, we first required pure Slam protein. Since, all the work has previously been done on N. meningitidis Slam1 and Slam2, we tried to purify them using standard outer membrane protein purification protocols [116]. However, we were unable to produce these proteins at reasonable purity and yields. Hence, we decided to take advantage of the database of Slam homologs that I described in Chapter 3 [163]. We selected 14 different putative Slam homologs and cloned them into a pET26b expression vector. The different Slams were then transformed in E. coli strain C43(DE3) and tested for expression using a-His western blots (data not shown). From this analysis, we identified 2 putative Slam homologs that expressed well: M. catarrhalis Slam1 and N. gonorrhoeae Slam2.
82
Figure 4.1: Functional test of Slam homologs used in the study. (A) Functional plate reader assay used for studying Slam function. Slams and SLPs were co-expressed in E. coli strain C43(DE3) and labeled with anti-flag antibodies followed by labeling with secondary antibody conjugated with fluorescent probe phycoerthyrin (PE). The fluorescence was quantified using a plate reader. (B) & (C) Quantification of surface display of Mcat TbpB and Nme HpuA respectively by different Slam1 and Slam2 homologs. Normalized fluorescence values obtained for each of the Slam homologs is shown. The results represent at least three biological replicates and demonstrate that Mcat Slam1 and Ngo Slam2 are functional.
To ensure that Mcat Slam1 and Ngo Slam2 are functional Slam homologs, we tested them in the E. coli translocation assay (Figure 4.1A). We co-expressed Nme Slam1, Mcat Slam1, Nme Slam2 and Ngo Slam2 with Mcat TbpB-flag and Nme HpuA-flag respectively. The cells were grown overnight and in autoinduction media and then labelled with anti-flag antibodies followed with a-mouse:PE
83
secondary antibodies. An increase in fluorescence indicates the display of SLP on the bacterial cell surface. We observed that Mcat Slam1 can potentiate the display of TbpB but not HpuA (Figure 4.1B). Similarly, Ngo Slam2 can potentiate the display of HpuA but not TbpB, retaining the specificity observed for Nme Slam2 (Figure 4.1C). Taken together, these results confirm that Mcat Slam1 and Ngo Slam2 are bonafide Slam homologs and can be used for the assay.
4.3.2 Purification of Slam1 and Slam2
Next, to purify Mcat Slam1, we developed a large-scale protein purification protocol shown in Figure 4.2A. Different growth conditions were tested, and cells grown in 12 L of autoinduction media at 20oC for 48 hrs was found to give the greatest cell mass (~70g) and Slam expression. Upon harvesting, cells were lysed using an Emulsiflex C3 and membranes isolated through ultracentrifugation. Membrane extraction was performed overnight in 3% Elugent detergent. His- tag purification was used as an initial affinity purification step for Slam1. During the His-tag purification, NiNTA beads were washed with buffer containing different detergents. n-dodecyl-B- D-maltoside (DDM concentration: 0.03%) was selected as the detergent of choice as it leads to highest protein yield. His-tag purification was followed up by ion-exchange chromatography using a MonoQ column. The purest Mcat Slam1 protein was found to elute in the 60 mM NaCl fractions and was further concentrated using an Amicon Ultra centrifugal filters (Millipore Sigma, 100 kDa MWCO) (Figure 4.2B). For each purification, a starting mass of ~70 g of cells yielded ~ 0.5-2 mg of pure Mcat Slam1. The protein was immediately flash frozen in liquid N2 for subsequent use in the translocation assays.
Ngo Slam2 was purified using a similar strategy as described above. ~70 g of cells was obtained from a 12 L culture of E. coli C43 (DE3) expressing pET26b Ngo Slam2. Purification was performed in two steps: Affinity tag purification (His-tag) followed by Ion exchange chromatography (MonoQ). DDM was used the detergent of choice (DDM concentration: 0.03%, buffer described in the previous section). The purest Ngo Slam2 was found in the 60 mM NaCl fraction. Each purification yielded ~ 2-3 mg of pure Ngo Slam2 (Figure 4.2C), which was flash frozen for subsequent use.
84
A
Cell lysis through Obtain membrane Detergent extraction Cell expression Emulsifex pellet of bacterial membrane
B C
Mcat Slam1 Ngo Slam2 kDa kDa
MonoQ His-tag purifcation 75- 75- 63- - Slam1 63- - Slam2 48- 48-
35- 35-
25- 25-
Figure 4.2: Purification strategy for Slam1&2. (A) Overall membrane protein expression and purification pipeline. (B) & (C) Pure MonoQ fractions from Mcat Slam1 and Ngo Slam2 purification. Pure Mcat Slam1 and Ngo Slam2 fractions were obtained in 60 mM NaCl. These were used for the downstream functional assays.
4.3.3 Purification of functional Bam complex
Previous studies have implicated the Bam complex in translocation of subset of SLPs across the outer membrane [62,157]. Hence, when thinking about the Slam chaperone model of SLP translocation, we hypothesized that the Bam complex may play the role of translocon for TbpB translocation. Therefore, we decided to express and purify E. coli BamABCDE. BamABCDE expression plasmid was obtained from Dr. Harris Bernstein and was purified using previously described methods [161]. 6 L culture of E. coli strain C43(DE3) expressing Bam complex lead to ~ 40 g of cells and ~ 6 mg of pure Bam complex. SDS-PAGE showed that the all five components of Bam complex are present in the purified complex (Figure 4.3A). The protein was flash frozen and used along with purified Slam in the translocation assay.
85
Figure 4.3: Purification and characterization of E. coli Bam complex. (A) BamABDCDE fractions obtained from a S200 gel filtration column. The BamABCDE complex was obtained using previously described protocols. (B) Design of an in vitro translocation assay for testing the function of the Bam complex. E. coli spheroplasts secrete porins such as OmpA into the supernatant. When incubated with Bam complex proteoliposomes, spheroplast released OmpA is successfully inserted in Bam proteoliposomes. (C) Insertion of OmpA in Bam proteoliposomes. Empty and Bam proteoliposomes were incubated with E. coli spheroplasts. The spheroplasts were removed and a- OmpA western blot was performed to quantify the OmpA present in the liposomes. Higher levels OmpA protein was observed at 15 mins in Bam proteoliposomes, suggesting that the Bam complex is functional and increases the efficiency of insertion of its substrate OmpA into liposomes.
In order to confirm that the Bam complex is functional, we used an in vitro reconstitution assay (Figure 4.3B). Previous studies have shown that the Bam complex can been successfully incorporated into liposomes and these proteoliposomes can be used to insert purified or spheroplast-
86
released substrate OMPs into the proteoliposomes [162]. The Bam complex was inserted into liposomes using previously described methods [161]. Bam proteo-liposomes were subsequently incubated with E. coli spheroplasts. Spheroplasts are cells whose outer membrane and peptidoglycan has been permeabilized, thus exposing the plasma membrane to the extracellular milieu. These include outer membrane proteins such as OmpA that are released in the supernatant bound to periplasmic chaperones Skp and SurA [164]. Upon recognition by the Bam complex, the OmpA protein is inserted to the proteoliposomes. To quantify the OmpA insertion, the spheroplasts were removed through centrifugation and western blot was performed using a-OmpA antibodies. Time-course experiments illustrate an increasing amount of OmpA present in Bam complex proteoliposomes in comparison empty liposomes (Figure 4.3C). The increase in OmpA incorporation confirms that the purified Bam complex is functional.
4.3.4 Proteoliposomes for Slam1 and the Bam complex
After obtaining purified Mcat Slam1 and Ecol BamABCDE, we tested the insertion of these proteins in liposomes (Figure 4.4A). E. coli polar lipids were extruded using a 200 nm filter to obtain unilamellar vesicles. The detergent stabilized Bam complex and Slam1 were incubated with liposomes. Detergent removal through Biobeads allowed for successful insertion of Mcat Slam1 and Bam complex as estimated by Silver stain, SDS-PAGE and Western blot analysis using a-His antibodies (Figure 4.4B-D). Similar to previous studies, BamABCDE was efficiently inserted into the liposomes (~70-80%). Slam1 was much less efficient (~2-5%). Hence, to get a similar level of insertion, 10 times more Slam1 was used for insertion to obtain similar amounts of Slam1 and Bam complex insertion.
While a number of protocols are available for making proteoliposomes, one advantage of pre- forming liposomes and then adding protein for outer membrane proteins with periplasmic domains (such as Slam, BamA) leads to the directional insertion of the liposomes. Given that the membrane domain will mediate insertion into the liposomes, the periplasmic domain most likely remains on the outer face of the liposomes. This is especially important for our assay as the translocation of substrate that in bacterial cells occurs from the periplasm to the surface can now be reconstituted as translocation of substrate inside the liposomes. To test this “inside-out” confirmation, we predicted that addition of proteinase K should lead to successful cleavage of the periplasmic domains/subunits. In the Slam1, this would lead to loss of the 6xHis tag on the N-terminus.
87
A
+
Slam Slam-detergent proteoliposomes complex
B Bam Slam1 Bam + Slam1 C Bam Slam1 Bam + Slam1 BSA BSA Slam (mg/mL) Slam Slam (mg/mL) Slam
0.125 0.25 0.5 I P I P I P 0.125 0.25 0.5 I P I P I P kDa kDa
100- 100- 75- - BamA 75- - BamA 63- 63- - Slam1 - Slam1 48- 48- 35- 35- 25- 25- 20- 17- 11-
Bam 72 - 42 Bam 45 - 22 % insertion % insertion Slam1 - 4 9 Slam1 - 0.9 2.8
Bam Slam1 Bam + Slam1 D E
Slam Slam
- + + - + + - + + PK Empty Bam + Slam1 Bam Slam1 - - + - - + - - + Triton kDa kDa
75- 63- - Slam1 48- 75- 63- 35- 48- 25- 35-
17- 25- 17- 11- - BamE 11-
α-His α-His Figure 4.4: Formation of Slam1 and Bam proteoliposomes. (A) Protocol used for insertion of protein into liposomes. Slam-DDM complexes were diluted (below the critical micellar concentration (CMC) of DDM) into preformed liposomes. Detergent was removed using BioBeads. (B) & (C) Insertion of Slam1 and Bam using silver stain and SDS-PAGE respectively. (D) Confirmation of Slam1 and Bam using Western blots using a-His antibodies. (E) Proteinase K shaving assay on Slam1 and Bam proteoliposomes. Proteoliposomes were incubated with 0.5 mg/ml proteinase K for 1 hr (in the presence or absence of 1% TritonX-100). Western blots using a-His antibodies were used to quantify the tag left upon proteinase K digestion. All the tag is lost indicating that the protein is inserted in an “inside-out” conformation.
88
In the case of the Bam complex, the 8xHis tag is present on BamE, which also faces the periplasm in the Bam complex structure [164]. Hence, treatment with proteinase K should lead to the loss of Bam complex signal in an a-His western blot (Figure 4.4E). As expected, we see complete loss of the signal upon treatment with proteinase K, indicating that the majority of Slam1 and Bam complex proteins have been inserted in the correct orientation.
Figure 4.5: Purification and functional assay with LolA. (A) Schematic of spheroplast release assay used to quantify LolA-dependent secretion of TbpB. (B) Purification of LolA. SDS-PAGE gel of the pure LolA fraction obtained after S75 run. The LolA was used for subsequent spheroplast release assay. (C) & (D) LolA dependent release of TbpB and HpuA respectively from E. coli spheroplasts. LolA was incubated with spheroplasts expressing TbpB or HpuA for different time period (as labelled). a-Flag antibodies were used to quantify the TbpB and HpuA released at the time points.
4.3.5 Release of TbpB and HpuA from spheroplasts by purified LolA
To deliver the substrate SLP in a functionally relevant manner, we utilized E. coli spheroplasts that lack an intact outer membrane. The SLPs expressed in spheroplasts would be displayed on the inner membrane. Previous studies have shown that addition of periplasmic chaperone LolA leads to
89
release of SLPs from spheroplasts in the culture supernatant [20] (Figure 4.5A). Hence, 5 µM of purified LolA was incubated with spheroplasts and LolA-dependent release of TbpB and HpuA was tested using western blots with a-Flag antibodies (Figure 4.5C and D). Addition of LolA to spheroplasts expressing TbpB or HpuA leads to an increase in TbpB and HpuA release at 10, 15, 20- and 30-min incubation.
4.3.6 Slam1 is necessary and sufficient for TbpB translocation
To establish an in vitro translocation assay for SLPs, we incubated the TbpB expressing spheroplasts with Slam or Bam proteoliposomes for 15 mins (Figure 4.6A). To test the successful translocation of TbpB inside proteoliposomes, we performed a proteinase K shaving assay. From this assay, we found that only proteoliposomes containing Slam1 showed robust protection (~40%), while empty and Bam proteoliposmes did not (Figure 4.6B, upper panel). The protection of TbpB was lost upon the addition of Triton-X100 suggesting that TbpB protection is coming from liposomes. Interestingly, proteoliposomes containing both Bam complex and Slam did not improve efficiency. This is unlike the TPSS system [151] where Bam increases the efficiency of TPSS substrate secretion, and suggests that translocation of SLPs is likely independent of the Bam complex. Taken together, this result confirms that Slam is necessary for the transport of SLPs.
4.3.7 Post-secretion translocation of Slam proteoliposomes
To investigate if the in vitro Slam dependent translocation assay of SLPs across an outer membrane is independent of its transport across the inner membrane, we repeated the experiments but this time, instead of incubating with spheroplasts, we let the spheroplasts secrete TbpB for 15 mins, removed the spheroplasts and then incubated the supernatant with Empty, Bam, Slam1 and Bam+Slam1 liposomes for another 15 mins. As seen previously, TbpB protection from proteinase- K shaving was observed in Slam1 and Bam+Slam1 proteoliposomes, but not Empty and Bam proteoliposomes (Figure 4.6B, lower panel). The efficiency of TbpB insertion (~40%) was similar to previously for co-secretion assay (~38%) (Figure 4.6C). This suggests that that transport of Slam-dependent SLP translocation is independent of translocation across the inner membrane via the Sec translocon or energy derived from the proton-motive force (through the TonB complex [165]) from the inner membrane.
90
Figure 4.6: Slam1 is necessary and sufficient for TbpB translocation. (A) Design of the in vitro assay used to study SLP translocation. (B) Results of the proteinase K shaving assay are shown as westerns with a-Flag antibodies for co-secretion (upper panel) and post-secretion (lower panel) assay. TbpB protection is observed in Slam1 and Bam+Slam1 proteoliposomes. A lower molecular weight band (~52 kDa) is also seen is predicted to be a partial TbpB construct. (C) Quantification of TbpB protection in proteoliposomes. Densitometric analysis with ImageLab (BioRad) was used calculate intensity of the full-length TbpB band (~75kDa). To calculate % TbpB protection, the intensity of band obtained for +PK lane was divided by -PK lane for each sample. The results are shown as a histogram and represent results obtained from four biological replicates. Results from co-secretion (grey) and post-secretion (black) assay are shown in the same graph.
91
A B Empty Bam Slam2 Bam + Slam2
Slam
Slam Slam Slam Slam
Slam
Empty Bam Slam2 Bam+Slam2 - + + - + + - + + - + + PK - - + - - + - - + - - + Triton 63- - Slam2 48- 48- 35- HpuA 35- 25- 20- 25- 17- 20- * * Partial HpuA 11- - BamE α-Flag α-His C
40
30
20
% HpuA insertion 10
0
Bam Empty Slam2 Bam + Slam2
Figure 4.7: Slam2 is necessary and sufficient for HpuA translocation. (A) Insertion of Slam2 and Bam complex into liposomes. Proteoliposomes containing empty, Bam, Slam2 and Bam + Slam2 were run on a SDS-PAGE gel and western blots using a-His antibodies were used to test the insertion of each protein in the liposomes. The results confirm the successful insertion of Slam2 into proteoliposomes, similar to Slam1. (B) Proteinase K shaving assay to detect HpuA translocation. HpuA protection is observed in Slam2 and Bam+Slam2 proteoliposomes. A lower molecular weight band (~20 kDa) is also visible and is predicted to be a partial HpuA construct. For clarity, only 10% of the input was loaded on the SDS gel. A similar partially protected fragment was also observed in the TbpB translocation assay shown in Figure 4.6. (C) Quantification of HpuA protection in proteoliposomes. Densitometric analysis with ImageLab (BioRad) was used calculate intensity of the full-length HpuA band (~40kDa). To calculate % HpuA protection, the intensity of band obtained for +PK lane was divided by -PK lane for each sample. The results are shown as a histogram and represent results obtained from three biological replicates.
92
4.3.8 Slam2 is necessary and sufficient for HpuA translocation
In Chapter 2, I mentioned that two Slam paralogs were found in N. meningitidis: Slam1 that is necessary for translocation of SLP TbpB, and Slam2 that is necessary of translocation of another SLP, HpuA [119]. Using the E. coli translocation assay, we showed that these Slam paralogs exhibit specificity towards their natural substrate. To test if the different Slams homologs function in a similar fashion in the in vitro translocation assay, we generated N. gonorrhoeae Slam2 proteo- liposomes using the methodology described previously. Similar to Slam1, liposome insertion was observed for Slam2 (Figure 4.7A). Bam, Empty and Bam + Slam2 proteoliposomes were also tested in the assay. Empty, Bam, Slam2 and Bam+Slam2 proteoliposomes were incubated with spheroplasts expressing N. meningitidis HpuA (co-secretion) and proteinase K shaving assay was performed (Figure 4.7B). Excitedly, similar to Slam1, Slam2 was sufficient to transport its substrate HpuA into the proteoliposome (~25% protection, Figure 4.7C). The presence of the Bam complex did not increase efficiency indicating that the transport of HpuA is also independent of the Bam complex. These findings suggest that different Slam homologs are active in the in vitro translocation assay, indicating that they function in analogous manner.
4.3.9 Slam1 and Slam2 retain their specificity in the in vitro translocation assay
Next, to examine whether Slam1 and Slam2 retained their substrate preference in the in-vitro translocation assay, we performed a competition assay. Slam1 proteoliposomes or Slam2 proteoliposomes were incubated with mixture of E. coli spheroplasts expressing both TbpB and HpuA (Figure 4.8A, B). A Proteinase K shaving assay was used to identify the proportion of each of the two SLPs protected by the different proteoliposomes. As seen previously in the E. coli translocation assay, Slam1 and Slam2 retained their specificity towards TbpB and HpuA respectively as indicated by the substrate specific protection seen from proteinase K (Figure 4.8C). Three biological replicates were used to quantify the protection by each of the liposomes (Figure 4.8D). These findings confirm that the specificity towards their substrate is inherent in the Slam.
93
A C Empty Bam Slam1 Slam2
TbpB HpuA
Slam SlamBam Slam1 Slam2 complex
Slam1 Slam2 PK - + + - + + - + + - + + Triton - - + - - + - - + - - + kDa 75- - TbpB 63- LolA-HpuA - * partial TbpB 48- LolA-TbpB - HpuA 35- α-Flag
B D kDa Empty Bam Slam1 Slam2
75- 63- - Slam1 48- - Slam2
35-
25- 20- 11- - BamE
α-His
Figure 4.8: Slam1&2 retain their specificity in the in vitro translocation assay. (A) Design of the in vitro assay used to investigate Slam substrate specificity. (B) Insertion of Slam1 and Slam2 into liposomes. Western blots using a-His antibodies were used to estimate insertion efficiency of each protein (Slam1, Slam2 and BamE). (C) Westerns blots after the proteinase K shaving assay with a-Flag antibodies. TbpB protection is observed in Slam1 proteoliposomes, while HpuA protection is seen in Slam2 proteoliposomes. A lower molecular weight band (~52 kDa) is also seen is predicted to be a partial TbpB construct. (D) Quantification of TbpB and HpuA protection in proteoliposomes. Densitometric analysis with ImageLab (BioRad) was used calculate intensity of the full-length TbpB (~75kDa) and HpuA (~ 40kDa) band. To calculate % SLP protection, the intensity of band obtained for +PK lane was divided by -PK lane for TbpB and HpuA band. The results are shown as a histogram and represent results obtained from two biological replicates.
94
A B Empty Bam Slam1 Bam + Slam1
Slam Bam Slam SlamBam Slam complex TbpB complex Slam
- + + - + + - + + - + + PK Slam1 - - + - - + - - + - - + Triton kDa 75- 63- α-TbpB 48- purifed TbpB (-40) kDa 75- Urea 63- α-TbpB denaturation 48- Urea-denatured TbpB (-40)
Purifed TbpB (-40) C
Figure 4.9: Reconstitution of TbpB translocation in a defined in vitro system. (A) Design of a defined assay for investigating TbpB translocation. (B) Proteinase K shaving assay with a-Flag antibodies. TbpB in buffer (folded) or 8M Urea (unfolded) was incubated with Empty, Bam, Slam1 and Bam+Slam1 proteoliposomes. TbpB protection is observed in Slam1 and Bam+Slam1 proteoliposomes only upon unfolding with 8M Urea. (C) Quantification of TbpB protection in proteoliposomes. Densitometric analysis with ImageLab (BioRad) was used calculate intensity of
95
the full-length TbpB band (75kDa). To calculate % HpuA protection, the intensity of band obtained for +PK lane was divided by -PK lane for each sample. The results are shown as a histogram and represent results obtained from two biological replicates. Protection seen for folded TbpB (grey) and unfolded TbpB (black) is shown.
4.3.10 Reconstitution of purified TbpB translocation in Slam proteoliposomes
Slams are predicted to be composed of two domains: C-terminal 14-stranded membrane domain and a N-terminal domain composed to tetratricopeptide repeats (TPR) [2]. Given the size of Slam (~ 50kDa), it cannot structurally accommodate a fully folded SLP as a monomeric OMP. Not surprisingly, the movement of SLPs (such as those found in Borrelia sp. [59]) have been proposed to also require periplasmic chaperones to prevent premature folding of SLPs. Given this, it was surprising that our in vitro translocation assay did not require additional chaperones. Intrigued by this, we attempted to reconstitute the transport of SLPs in a defined system. To this end, we purified N. meningitidis TbpB (40-599) in E. coli using previously described protocols [111]. The purified TbpB was then added directly or first unfolded in 8M urea and then added to Slam proteoliposomes (Figure 4.9A). Translocation of TbpB was tested using a proteinase K shaving assay. Similar to the results obtained for TbpB delivered from spheroplasts, only unfolded purified TbpB could be successfully translocated in Slam proteoliposomes (~3% insertion), but not in empty liposomes or Bam complex proteoliposomes (Figure 4.9B, C). Combination of Bam and Slam liposomes did not increase efficiency of insertion. Interestingly, the efficiency of translocation was much lower compared to the co- or post-translocation assay. This could be due to the fact that a higher amount of TbpB (0.5 uM) was used in the purified assay compared to co/post-secretion assays. Alternatively, the lower efficiency of TbpB translocation and protection could be due to the rapid folding of TbpB upon the dilution of 8M urea to 0.66 M urea, which would allow a very short time for the unfolded TbpB to interact with its translocon Slam.
A number of key observations can be derived from these results. First, this result confirms that Slam1 is both necessary and sufficient for the movement of TbpB across the outer membrane. Second, the translocation is only successful upon addition of 8M urea indicating that Slam- dependent translocation requires an unfolded substrate. This suggests that there are factors present in the periplasm of E. coli that prevent TbpB from folding during its transit to the outer membrane.
96
However, no such chaperones/factors have been described to date for SLPs. Third, Slam-dependent translocation does not require any small molecule/protein effectors that may be potentially secreted by the spheroplasts. Fourth, these results also suggest that the lipid anchor and the anchor peptide are not essential for SLP translocation. The lipid anchor has previously been shown not to be necessary for the secretion of the TbpB homolog in N. gonorrhoeae [166], where the absence of an N-terminal cysteine residue still leads to secretion from cells. However, mutations in the lipid anchor have been shown to affect expression and secretion of TbpB [166]. This suggests that the lipid anchor may be important for transit across the inner membrane (through the Sec translocon) or transit through the periplasm (by binding to periplasmic factors/chaperones).
4.4 Discussion
The molecular function of Slam and the role it plays in SLP translocation is the central aim of the current study. In this chapter, I described our work towards developing an in vitro functional assay to answer this question. Similar assays have been developed for many other protein transport system including outer membrane translocons of the Omp85 family members: the Bam complex [160,161] and two-partner secretion systems [151,162]. One of the key advantages of this reductionist approach is that it is able to identify the minimal components required for a given biological function. The assay also allows us to examine essential factors such as the Bam complex that are harder to analyze using genetic screens. Bam A and D are essential for survival and removing them in E. coli requires the presence of an additional copy of these genes on an inducible plasmid [167]. The introduction and appropriate control of Slam1 and TbpB in this system is difficult. Additionally, Slam is an outer membrane protein that presumably requires the Bam complex for its own efficient insertion into the outer membrane. This makes it more difficult to evaluate if the Bam complex plays a direct or an indirect role in Slam-dependent TbpB translocation. Hence, a biochemical approach of in vitro reconstitution was chosen to answer this question.
As a first step, many Slam homologs were screened and Slam1 from M. catarrhalis and Slam2 from N. gonorrhoeae were selected for the study owing to their purification yield and stability in detergents. Detergent removal through Biobeads was used to insert Slams into preformed
97
liposomes. The Bam complex was used as a control. Compared to the Bam complex, lower amount of Mcat Slam1 and Ngo Slam2 insertion was observed. Hence, to get similar levels of insertion as quantified by SDS-PAGE gel and western blot, 10 times more Slam1 was required. Further optimization of the insertion procedure may result in higher levels of Slam in the proteoliposomes. Lipid composition, incubation strategy, ratio of protein:lipid can be further tested to improve Slam insertion efficiency [168]. Further work on the purification of Slam (more stable protein) may also increase Slam insertion into liposomes.
Next, to deliver the substrate in a functionally relevant manner, we utilized E. coli spheroplasts that lack an intact outer membrane. We incubated Mcat Slam1 and Bam proteoliposomes. We found that Mcat Slam proteoliposomes protected ~ 40% of TbpB released from the spheroplasts. This protection was also shown by Ngo Slam2 proteoliposomes that were able to successfully protect the SLP HpuA from proteinase K degradation. Mcat Slam1 and Ngo Slam2 retained their specificity in a competition assay, indicating that substrate specificity is inherent to Slams. Taken together, these findings suggest that Slam is an outer membrane translocon that is specific for TbpB-like surface lipoproteins.
Next, to identify the minimal components required for Slam translocon function, we purified N. meningitidis TbpB. The TbpB in folded or unfolded (with 8M Urea) were incubated with Slam proteoliposomes. Proteinase K shaving assay was done to quantify the amount of TbpB inserted into the proteoliposomes. From this analysis, we found that Mcat Slam1 specifically transported only unfolded TbpB into proteoliposomes. No protection was seen for folded TbpB or proteoliposomes with Bam or no protein. This result conclusively proves that Slam is a translocon and it transports an unfolded substrate. Interestingly, while we have so far reported that Slams transport SLPs, the lipid anchor was found not to be required for the transport of TbpB by Mcat Slam1 proteoliposomes. This suggests that Slams can also secrete proteins from bacteria. In Chapter 2, I mentioned that many Slam-adjacent genes (such as ngo554 in N. gonorrhoeae) are predicted to encode an 8-stranded barrel domain and a Sec-dependent signal peptide but lack a lipobox motif. The Slam-dependent secretion of these putative proteins can be tested in the E. coli translocation assay. Alternatively, to validate this prediction, a TbpB construct that lacks the lipobox motif can also tested in the E. coli and in vitro translocation assay.
98
Another key observation from this assay is that TbpB needs to be unfolded for translocation by Slam. TbpB is a well-folded soluble protein with a high-melting temperature and stable 3- dimensional structure [110]. This suggests that TbpB is kept unfolded in the periplasm and the factor responsible for this is conserved between N. meningitidis and E. coli. A number of periplasmic chaperones, such as SurA, Skp and DegP are conserved between these two species and have been studied extensively in the context of outer membrane protein biogenesis [6]. However, no information is available regarding the role of these or other periplasmic chaperones in SLP biogenesis. The in vitro translocation assay described in this study will be great tool to identify such periplasmic factors that bind to SLPs and aid in translocation.
Availability of the in vitro translocation assay also allows us to test some of the observations made in Chapter 3 regarding Slam mechanism and substrate specificity. By incubating Slam proteoliposomes with spheroplasts expressing different TbpB-HpuA domain swaps, we may be able to identify the residues that are important for translocation. The translocation of GST fusion with TbpB C-lobe could help to validate that translocation of SLPs occurs from C-terminus to the N- terminus. While we have developed this study for studying TbpB-like SLPs, a similar approach can also be applied to investigate potential SLPs translocons in other Gram-negative bacteria especially Borrelia and Bacteroides sp. [2].
99
5 Chapter 5 Summary and Future directions
5.1 Key findings of the study
In Chapter 1, I introduced surface lipoproteins and the myriad of roles they play in bacterial cell biology. With current literature and recent review articles, it is clear that SLPs are found in most, if not all Gram-negative bacteria. As more and more SLPs are found, questions regarding how different bacteria deliver SLPs to the surface remain to be discovered for most reported SLPs. Recently, our lab identified an outer membrane protein in N. meningitidis that is necessary for surface display of the SLP TbpB [49]. Hence, we named this protein Surface lipoprotein assembly modulator or Slam. Slams contain a predicted N-terminal periplasmic domain containing tetratricopeptide repeats and a C-terminal membrane bound 14-stranded beta-barrel domain, which is also essential for Slam function. Addition of Slam to E. coli, the model lab organism that contains no Slam or TbpB homologs allows for successful display of TbpB on the surface of the cell[49]. These findings indicate that Slam can potentiate TbpB translocation in a hitherto unknown mechanism. The aim of the current thesis was to shed more light onto the molecular mechanism of Slam function.
In Chapter 2, I talked about the results obtained from bioinformatic analysis on Slams. Using blast searches, we showed that putative Slam homologs are found in a large of number of Gram-negative bacteria from the phylum Proteobacteria [163]. Interestingly, while curating the genes found adjacent to putative Slam genes, we discovered that many Slam adjacent genes encode predicted lipoproteins. Many of these included known SLPs, and we tested their Slam-dependence by reconstituting their surface display in E. coli. For one putative Slam-adjacent SLP from Pasteurella multocida, we were able to elucidate Slam dependence and surface display by reconstituting its transport in E. coli. By comparing the domain architecture of these putative Slam-dependent SLPs, we were able to find that they all share a predicted 8-stranded beta-barrel domain at their C-terminal end [163]. We hypothesize that this domain contains the secretion signal required for Slam- dependent surface display.
100
In Chapter 3, I described the results regarding the key residues on SLPs from the E. coli translocation assay. A Slam paralog in N. meningitidis was identified and it specifically recognizes the SLP HpuA [49]. Slam1 and Slam2 exhibit different substrate preference and their specificity depends on the 8-stranded beta-barrel domain present on the C-terminus. Using mutational analysis, we showed the last two strands of the barrel are important for translocation. Additionally, using a construct of GST fused to the TbpB C-lobe, we found that the translocation occurs from C-terminus to the N-terminus and that TbpB interacts with Slam1 in the outer membrane. Slam1 acts in the outer membrane and potentiates translocation of TbpB from the inner leaflet of the outer membrane to the surface.
In Chapter 4, I discussed the work done in developing an in vitro translocation assay for further investigating Slam function. E. coli cells lacking an intact outer membrane (spheroplasts) release TbpB into the supernatant when incubated with periplasmic lipoprotein chaperone LolA. When incubated with liposomes containing Slam1, the spheroplast secreted TbpB was transported inside the liposomes containing Slam1, but not empty or Bam complex containing liposomes. This ability was not restricted to Slam1 as Slam2 could successfully deliver spheroplast secreted HpuA into liposomes. Interestingly, Slam1&2 retained their substrate specificity in the in vitro translocation assay. Extending this work to a defined system, we showed that purified non-lipidated TbpB upon unfolding with 8 M Urea can be successfully inserted into Slam1 containing liposomes. Taken together, these findings showed the Slams are both necessary and sufficient for the translocation of TbpB-like SLPs and they recognize an unfolded substrate. Additionally, our results also indicate that the lipid anchor is not essential for Slam-mediated transport across the outer membrane.
5.2 Slam is a protein translocon in the outer membrane
The work described in this thesis answers key questions regarding the role of Slam in SLP translocation. From our work, we conclude that Slam is a protein translocon in the outer membrane and the Slam membrane domain forms the conduit for SLP translocation. A number of translocons have been found in the outer membrane of Gram-negative bacteria (Figure 5.1) including members of the Omp85 family (BamA, TamA, TpsB), porins (OmpF) and secretins (Type II & Type III secretion system) [169]. Given the molecular size of Slam (~50 kDa) and predicted 14-stranded
101
membrane barrel domain, we postulate that Slam is one of the smallest known translocons found in Gram-negative bacteria, if it functions as a monomer[169]. Additional experiments are required to investigate if Slam is indeed a monomer. While the structure of Slam could provide evidence on the multimeric state of Slam, biophysical assays including electrophysiology [170] and cross-linking experiments [171] can be used to investigate the oligomeric state of Slam. Regardless, investigating the molecular mechanism of Slams will bring insights into the mechanism of protein transport across the outer membrane in general.
Figure 5.1: Outer membrane translocons in Gram-negative bacteria. A schematic representation of different outer membrane translocons present in Gram-negative bacteria. Slam structure (blue – membrane domain, red – TPR) was predicted using the Phyre2 prediction software. OmpF (green, PDB: 3O0E), BamA (cyan, PDB: 4K3B), FimD (silver, PDB: 3RFZ), CsgG (yellow, PDB: 4UV3), GspD (dark blue, PDB: 5WQ7) and InvG (orange, PDB: 5TCQ) are other outer membrane translocons that have been described in the literature. The number of beta- strands present in each of the protein is also shown.
5.3 Model of Slam function
The work described in this thesis allows us to propose a more complete model of Slam function (Figure 5.2). First, we have shown that Slams act in the outer membrane after the insertion of the SLP in the outer membrane by the Lol system (Step 1). Slams require an unfolded substrate (as evidenced by the GST-fusion studies and the defined in vitro translocation assays) indicating that periplasmic chaperones likely play a role to keep the SLP unfolded. Upon insertion, the Slam recognizes the “secretion motif” present on the C-terminal end of the SLP (Step 2). Mutational
102
analysis and domain swap experiments have shown that the secretion motif lies on the last two strands of the C-terminal barrel domain of Slam-dependent SLPs. Upon interaction, we predict that conformational changes in the Slam opens the channel for translocation of the SLP across the outer membrane (Step 3). The membrane domain forms the conduit for transport, whereas the Ntd domain plays a role in controlling the entrance to the barrel and may dictate specificity or interactions with the periplasmic chaperones. Domain swap and GST-fusion experiments have shown that the translocation proceeds from the C-terminus to the N-terminus through the membrane domain of Slam. The folding of the translocated SLP on the extracellular side of the outer membrane or lumen of the Slam liposomes (especially the 8-stranded beta barrel domain) provides a “Brownian-ratchet” to prevent back movement of the SLP (Step 4). The folding may also provide the energy to pull the remaining SLP across the outer membrane. In the final step, Slam flips the lipid anchor from inner leaflet to the outer leaflet potentially through a lateral gate (Step 5). This is based on the molecular structure of LptD, which flips lipopolysaccharide molecules that also contain a lipid anchor across the outer membrane [172]. Once completed, the SLP is free to move laterally in the outer membrane and Slam is ready for another round of translocation.
Figure 5.2: Model of Slam-mediated SLP translocation across the outer membrane. A schematic representation of various stages of SLP transport across the outer membrane is shown. Briefly, the unfolded SLPs are inserted in the outer membrane by the Lol system and kept unfolded by a potential chaperone (Step 1). The “specificity motif” present at the C-terminus of the unfolded SLP is identified by Slam1 (Step 2). The interaction of Slam1 and SLP leads to movement of the
103
Ntd domain and opening of the conduit in the barrel domain (Step 3). The SLP is then transported across the outer membrane through the Slam-barrel and folds rapidly (Step 4). The folding of SLP pulls the rest of the protein across the outer membrane. Once, the entire length of the SLP is transported, the Slam lateral gate allows the lipid anchor to “flip” from the inner leaflet to the outer leaflet of the outer membrane (Step 5).
While more details of Slam function will be identified as additional molecular information on these proteins becomes available, the model serves as tool for designing future experiments. Of key importance are the identity of the periplasmic chaperone that interacts with the SLPs as they transit across the periplasm in an unfolded state and of course the atomic structure of Slam itself. In the next section, I will describe some key experiments that may shed more light on these proteins.
5.4 Future directions
The current study opens up a number of questions regarding the molecular mechanism of Slam function. The E. coli and the in vitro reconstitution assays developed in this study can now be used to answer some of these questions.
5.4.1 Chaperones involved in translocation
One of the key remaining questions is regarding the chaperones that assist in keeping the SLPs unfolded in the periplasm for recognition by Slams. As described in the Chapter 2, several chaperones have been studied in the context of the outer membrane proteins including SurA, Skp, DegP and Fkbp [6,173]. Interestingly, one study has tried to investigate the role of a SurA, Skp and DegP in surface display of fHbp (a Slam substrate) in N. meningitidis and showed that the loss of SurA, Skp or DegQ (homolog of DegP) did not affect fHbp surface display [174]. This suggests that either multiple chaperones are involved in SLP translocation and the loss of one can be compensated by others, or a different chaperone is involved in this process. Given that we were able to successfully reconstitute TbpB and HpuA translocation in E. coli suggests that the periplasmic chaperone involved in SLP biogenesis is conserved between E. coli and N. meningitidis. Additionally, TbpB is a stable protein indicating that the interaction between SLP and the chaperone needs to happen directly after its movement across the inner membrane via the Sec translocon.
104
To answer this question, a number of experiments can be designed. First, we have found that post- secretion translocation works as efficiently as co-secretion translocation, indicating that interaction between SLP and the chaperone is a stable one. Pull down experiments +/- cross-linking followed by mass-spectrometry should be able to identify the factors bound to SLPs secreted by spheroplasts. Once identified, the factors can be rapidly examined in the defined or the spheroplast based translocation assays. Alternative approaches include using the Keio collection in E. coli [175]. SurA, Skp, DegP and Fkbp are not essential in E. coli, hence deletion strains of these genes are available in Keio collection. We have previously used the E. coli translocation assay in the Keio collection strains to show that TamA and TamB are not required for Slam function [49]. Similar experiments can also be performed for mutants in each of the chaperone genes. Spheroplasts made from these strains may also be used in the in vitro translocation assay. Together, these studies should be able to identify the chaperone and further investigate the molecular basis of SLP- chaperone interaction.
5.4.2 Slam secretion and specificity motif on SLPs
Another key question that remains to be answered is regarding the secretion motif present on Slam- dependent SLPs. In Chapter 3, I described the work done on this key question using the E. coli reconstitution assay. Single point mutations should be carried out to examine the role of conserved residues on the last two strands of the barrel, which were found to be important for TbpB surface display. The specificity exhibited by Slam1 and Slam2 can be also used to further identify the residues that dictate substrate preference. Using domain swap experiments, we have shown that the specificity motif also lies on the last C-terminal domain of Slam-dependent SLPs. Either by swapping specific beta-strands or key residues, preferences of Slams could be switched. This could help in identifying residues important for translocation. The interesting SLP mutants can be rapidly tested in the spheroplast-based or the defined in vitro translocation assay. These studies should help in narrowing down the secretion motif present on Slam-dependent SLPs. Additionally, the secretion motif can also be used to organize the database of Slams and SLPs and could be used to classify Slams on the basis of their substrate preferences.
While the E. coli translocation assay has been very informative, it also has certain limitations. First, most of the above studies were done with pET-based vectors that often lead to overproduction of client proteins. We have tried to rectify this by testing the key findings in Neisseria, however this is
105
difficult for most of the assays owing to the fact that N. meningitidis is a Level 2 pathogen. Second, the lack of a molecular structure of Slam has stymied the design of mutants that could be used to enhance cross-linking studies between a substrate and Slam which have not been successful to date thus making it harder to declare a direct interaction between Slams and SLPs. To this end, a direct evolution approach can also be used as an unbiased approach to identify residues involved in Slam- dependent SLP translocation [176]. We have shown that TbpB and HpuA can be successfully recognized on the surface of Slam expressing E. coli by flow cytometry. Mutations on the TbpB that leads the loss of surface display or more interestingly can switch the specificity from Slam1 to Slam2 could be rapidly screened using the flow cytometry and sorting. Such an approach can also be used to investigate residues on the Slams as well.
5.4.3 Structure of Slams
One of the key hurdles in designing experiments for investigating Slam function, especially towards the identification of key residues on Slam, is the lack of atomic resolution structure of Slams. Given that proteins with a DUF560 domain (the Slam membrane domain) have not been studied in detail, no structural information is available on these family of proteins. A large chunk of my PhD work has been focused on developing a robust protocol for large scale expression and purification of Slams. A number of Slam homologs were tested and Mcat Slam1 and Ngo Slam2 were chosen as Slams for in vitro translocation studies described in Chapter 4. Work done by various members (including myself) in the Moraes lab have resulted in protein crystals of different Slam homologs for X-ray crystallography studies. These crystals were subsequently screened at different synchrotron facilities. However, to date the datasets obtained have not permitted determination of Slam structure. This work will continue and as more Slam homologs are screened, and the purification protocol further refined, structure of Slam should be determined in the near future.
In recent years, there has been tremendous progress in structural characterization of membrane proteins. Two technologies have been of great importance for this progress: development of membrane mimetics (nanodiscs, LCP, bicelles, steroid/lipid-based detergents) [177] and cryo- electron microscopy (cryoEM) [178]. Our lab has begun working on different membrane mimetics for improving Slam stability and purification yield and in the future, we hope this will help in Slam structure determination. On the cryo-EM side, so far, the size of Slam (~50 kDa) and access to cryoEM facility have precluded any work. However, the latest advances in direct electron detector
106
as well as particle selection and analysis algorithms have pushed the size limit of cryoEM. Hopefully, these breakthroughs will bring Slam structure determination by cryo-EM within reach in the near future.
5.4.4 Slam molecular mechanism
Based on the work done so far, Slams represent a unique solution to the problem of delivering of polypeptides across a lipid bilayer. Many translocons have been studied to date but given their relative size and minimal substrate requirements, Slams provide a unique test case to study this essential and fundamental process in greater detail. Based on the work done on other translocons, Slams appear to be most analogous to the translocation protein (TpsB) of the Type V secretion system [148]. TpsB protein has been structurally characterized [170], a liposome-based translocation assay is available [151] and cross-linking studies have also been done to investigate the molecular mechanism of TpsB function [152]. With these studies, a model has been proposed where the TpsB acts as a funnel in the outer membrane for the translocation of its substrate [148]. As the first step, the substrate interacts with weak affinity to the residues at the periplasmic face of the TpsB. These interactions allow for the movement of substrate towards the outer membrane within the lumen of TpsB barrel. As the substrate moves, the TpsB lumen residues act as a funnel, guiding the substrate to the cell surface. The folding of substrate on the surface acts as “Brownian- ratchet” that prevents that back movement of the substrate allowing for unidirectional transport of the substrate to the extracellular milieu.
Similar work on Slams can shed light on their molecular mechanism. To this end, the in vitro translocation assay provides a great experimental setup. First, TbpB fusion with GST or other tags (such as DHFR [179]), should be tested in this translocation assay. We hypothesize that like the E. coli translocation assay, while the TbpB will be translocated inside the liposomes, the GST tag should be remain outside. In this stalled complex, the GST-TbpB and Slam1 will be present together. By introducing various cross-linkers (photoactivable, chemical etc.) at different positions in the GST-TbpB construct, we should be able to obtain a stable complex between GST-SLP and Slam1. This can either be investigated through western blots, mass-spectrometry or structurally characterized using cryoEM or X-ray crystallography studies [180,181]. These studies would provide an unprecedented look at this process and will hopefully shed more light on these fascinating family of proteins.
107
References
1. Silhavy TJ, Kahne D, Walker S: The Bacterial Cell Envelope. Cold Spring Harb Perspect Biol 2010, 2.
2. Hooda Y, Moraes TF: Translocation of lipoproteins to the surface of gram negative bacteria. Curr Opin Struct Biol 2018, 51:73–79.
3. Luirink J, Yu Z, Wagner S, de Gier J-W: Biogenesis of inner membrane proteins in Escherichia coli. Biochim Biophys Acta BBA - Bioenerg 2012, 1817:965–976.
4. Peschke M, Le Goff M, Koningstein GM, Karyolaimos A, de Gier J-W, van Ulsen P, Luirink J: SRP, FtsY, DnaK and YidC Are Required for the Biogenesis of the E. coli Tail- Anchored Membrane Proteins DjlC and Flk. J Mol Biol 2018, 430:389–403.
5. Dong C, Beis K, Nesper J, Brunkan-LaMontagne AL, Clarke BR, Whitfield C, Naismith JH: Wza the translocon for E. coli capsular polysaccharides defines a new class of membrane protein. Nature 2006, 444:226–229.
6. Plummer AM, Fleming KG: From Chaperones to the Membrane with a BAM! Trends Biochem Sci 2016, 41:872–882.
7. Noinaj N, Gumbart JC, Buchanan SK: The β-barrel assembly machinery in motion. Nat Rev Microbiol 2017, 15:197–204.
8. Stubenrauch C, Belousoff MJ, Hay ID, Shen H-H, Lillington J, Tuck KL, Peters KM, Phan M-D, Lo AW, Schembri MA, et al.: Effective assembly of fimbriae in Escherichia coli depends on the translocation assembly module nanomachine. Nat Microbiol 2016, 1:16064.
9. Dunstan RA, Hay ID, Wilksch JJ, Schittenhelm RB, Purcell AW, Clark J, Costin A, Ramm G, Strugnell RA, Lithgow T: Assembly of the secretion pores GspD, Wza and CsgG into bacterial outer membranes does not require the Omp85 proteins BamA or TamA. Mol Microbiol [date unknown], 97:616–629.
10. Narita S, Tokuda H: Sorting of bacterial lipoproteins to the outer membrane by the Lol system. Methods Mol Biol Clifton NJ 2010, 619:117–129.
11. Konovalova A, Silhavy TJ: Outer membrane lipoprotein biogenesis: Lol is not the end. Phil Trans R Soc B 2015, 370:20150030.
12. Babu MM, Priya ML, Selvan AT, Madera M, Gough J, Aravind L, Sankaran K: A database of bacterial lipoproteins (DOLOP) with functional assignments to predicted lipoproteins. J Bacteriol 2006, 188:2761–2773.
13. Arambula D, Wong W, Medhekar BA, Guo H, Gingery M, Czornyj E, Liu M, Dey S, Ghosh P, Miller JF: Surface display of a massively variable lipoprotein by a Legionella diversity-generating retroelement. Proc Natl Acad Sci 2013, 110:8212–8217.
108
14. Kovacs-Simon A, Titball RW, Michell SL: Lipoproteins of Bacterial Pathogens. Infect Immun 2011, 79:548–561.
15. Vogeley L, Arnaout TE, Bailey J, Stansfeld PJ, Boland C, Caffrey M: Structural basis of lipoprotein signal peptidase II action and inhibition by the antibiotic globomycin. Science 2016, 351:876–880.
16. Mao G, Zhao Y, Kang X, Li Z, Zhang Y, Wang X, Sun F, Sankaran K, Zhang XC: Crystal structure of E. coli lipoprotein diacylglyceryl transferase. Nat Commun 2016, 7:ncomms10198.
17. Wiktor M, Weichert D, Howe N, Huang C-Y, Olieric V, Boland C, Bailey J, Vogeley L, Stansfeld PJ, Buddelmeijer N, et al.: Structural insights into the mechanism of the membrane integral N-acyltransferase step in bacterial lipoprotein synthesis. Nat Commun 2017, 8:ncomms15952.
18. Noland CL, Kattke MD, Diao J, Gloor SL, Pantua H, Reichelt M, Katakam AK, Yan D, Kang J, Zilberleyb I, et al.: Structural insights into lipoprotein N-acylation by Escherichia coli apolipoprotein N-acyltransferase. Proc Natl Acad Sci 2017, 114:E6044– E6053.
19. Okuda S, Tokuda H: Lipoprotein Sorting in Bacteria. Annu Rev Microbiol 2011, 65:239– 259.
20. Yakushi T, Yokota N, Matsuyama S, Tokuda H: LolA-dependent Release of a Lipid- modified Protein from the Inner Membrane of Escherichia coli Requires Nucleoside Triphosphate. J Biol Chem 1998, 273:32576–32581.
21. Yakushi T, Masuda K, Narita S, Matsuyama S, Tokuda H: A new ABC transporter mediating the detachment of lipid-modified proteins from membranes. Nat Cell Biol 2000, 2:212–218.
22. Matsuyama S, Tajima T, Tokuda H: A novel periplasmic carrier protein involved in the sorting and transport of Escherichia coli lipoproteins destined for the outer membrane. EMBO J 1995, 14:3365–3372.
23. Matsuyama S, Yokota N, Tokuda H: A novel outer membrane lipoprotein, LolB (HemM), involved in the LolA (p20)-dependent localization of lipoproteins to the outer membrane of Escherichia coli. EMBO J 1997, 16:6947–6955.
24. Dong H, Zhang Z, Tang X, Paterson NG, Dong C: Structural and functional insights into the lipopolysaccharide ABC transporter LptB 2 FG. Nat Commun 2017, 8:222.
25. McLeod SM, Fleming PR, MacCormack K, McLaughlin RE, Whiteaker JD, Narita S, Mori M, Tokuda H, Miller AA: Small-Molecule Inhibitors of Gram-Negative Lipoprotein Trafficking Discovered by Phenotypic Screening. J Bacteriol 2015, 197:1075–1082.
26. Lewenza S, Mhlanga MM, Pugsley AP: Novel Inner Membrane Retention Signals in Pseudomonas aeruginosa Lipoproteins. J Bacteriol 2008, 190:6119–6125.
109
27. Lauber F, Cornelis GR, Renzi F: Identification of a New Lipoprotein Export Signal in Gram-Negative Bacteria. mBio 2016, 7:e01232-16.
28. Schulze RJ, Zückert WR: Borrelia burgdorferi lipoproteins are secreted to the outer surface by default. Mol Microbiol 2006, 59:1473–1484.
29. Sutcliffe IC, Harrington DJ, Hutchings MI: A phylum level analysis reveals lipoprotein biosynthesis to be a fundamental property of bacteria. Protein Cell 2012, 3:163–170.
30. D’Enfert C, Chapon C, Pugsley AP: Export and secretion of the lipoprotein Pullulanase by Klebsiella pneumoniae. Mol Microbiol 1987, 1:107–116.
31. Grabowicz M, Silhavy TJ: Redefining the essential trafficking pathway for outer membrane lipoproteins. Proc Natl Acad Sci 2017, 114:4769–4774.
32. Ekiert DC, Bhabha G, Isom GL, Greenan G, Ovchinnikov S, Henderson IR, Cox JS, Vale RD: Architectures of Lipid Transport Systems for the Bacterial Outer Membrane. Cell 2017, 169:273-285.e17.
33. Suits MDL, Sperandeo P, Dehò G, Polissi A, Jia Z: Novel Structure of the Conserved Gram-Negative Lipopolysaccharide Transport Protein A and Mutagenesis Analysis. J Mol Biol 2008, 380:476–488.
34. Wilson MM, Bernstein HD: Surface-Exposed Lipoproteins: An Emerging Secretion Phenomenon in Gram-Negative Bacteria. Trends Microbiol 2016, 24:198–208.
35. Szewczyk J, Collet J-F: Chapter One - The Journey of Lipoproteins Through the Cell: One Birthplace, Multiple Destinations. In Advances in Microbial Physiology. Edited by Poole RK. Academic Press; 2016:1–50.
36. Zückert WR: Secretion of Bacterial Lipoproteins: Through the Cytoplasmic Membrane, the Periplasm and Beyond. Biochim Biophys Acta 2014, 1843:1509–1516.
37. Vesikari T, Esposito S, Prymula R, Ypma E, Kohl I, Toneatto D, Dull P, Kimura A, EU Meningococcal B Infant Vaccine Study group: Immunogenicity and safety of an investigational multicomponent, recombinant, meningococcal serogroup B vaccine (4CMenB) administered concomitantly with routine infant and child vaccinations: results of two randomised trials. Lancet Lond Engl 2013, 381:825–835.
38. Jiang H-Q, Hoiseth SK, Harris SL, McNeil LK, Zhu D, Tan C, Scott AA, Alexander K, Mason K, Miller L, et al.: Broad vaccine coverage predicted for a bivalent recombinant factor H binding protein based vaccine to prevent serogroup B meningococcal disease. Vaccine 2010, 28:6086–6093.
39. Barasuol BM, Guizzo JA, Fegan JE, Martínez-Martínez S, Rodríguez-Ferri EF, Gutiérrez- Martín CB, Kreutz LC, Schryvers AB, Frandoloso R: New insights about functional and cross-reactive properties of antibodies generated against recombinant TbpBs of Haemophilus parasuis. Sci Rep 2017, 7.
110
40. Wressnigg N, Pöllabauer E-M, Aichinger G, Portsmouth D, Löw-Baselli A, Fritsch S, Livey I, Crowe BA, Schwendinger M, Brühl P, et al.: Safety and immunogenicity of a novel multivalent OspA vaccine against Lyme borreliosis in healthy adults: a double-blind, randomised, dose-escalation phase 1/2 trial. Lancet Infect Dis 2013, 13:680–689.
41. Lithgow KV, Hof R, Wetherell C, Phillips D, Houston S, Cameron CE: A defined syphilis vaccine candidate inhibits dissemination of Treponema pallidum subspecies pallidum. Nat Commun 2017, 8:14273.
42. Manning PA, Beutin L, Achtman M: Outer membrane of Escherichia coli: properties of the F sex factor traT protein which is involved in surface exclusion. J Bacteriol 1980, 142:285–294.
43. Pugsley AP, Chapon C, Schwartz M: Extracellular pullulanase of Klebsiella pneumoniae is a lipoprotein. J Bacteriol 1986, 166:1083–1088.
44. Schryvers AB, Morris LJ: Identification and characterization of the transferrin receptor from Neisseria meningitidis. Mol Microbiol 1988, 2:281–288.
45. Brandt ME, Riley BS, Radolf JD, Norgard MV: Immunogenic integral membrane proteins of Borrelia burgdorferi are lipoproteins. Infect Immun 1990, 58:983–991.
46. Chamberlain NR, Brandt ME, Erwin AL, Radolf JD, Norgard MV: Major integral membrane protein immunogens of Treponema pallidum are proteolipids. Infect Immun 1989, 57:2872–2877.
47. Konovalova A: Cell Surface Exposure. In Bacterial Protein Secretion Systems. . Humana Press, New York, NY; 2017:87–95.
48. Anderson JE, Sparling PF, Cornelissen CN: Gonococcal transferrin-binding protein 2 facilitates but is not essential for transferrin utilization. J Bacteriol 1994, 176:3162–3170.
49. Hooda Y, Lai CC-L, Judd A, Buckwalter CM, Shin HE, Gray-Owen SD, Moraes TF: Slam is an outer membrane protein that is required for the surface display of lipidated virulence factors in Neisseria. Nat Microbiol 2016, 1:16009.
50. Fletcher LD, Bernfield L, Barniak V, Farley JE, Howell A, Knauf M, Ooi P, Smith RP, Weise P, Wetherell M, et al.: Vaccine Potential of the Neisseria meningitidis 2086 Lipoprotein. Infect Immun 2004, 72:2088–2100.
51. Pizza M, Scarlato V, Masignani V, Giuliani MM, Aricò B, Comanducci M, Jennings GT, Baldi L, Bartolini E, Capecchi B, et al.: Identification of vaccine candidates against serogroup B meningococcus by whole-genome sequencing. Science 2000, 287:1816–1820.
52. Dowdell AS, Murphy MD, Azodi C, Swanson SK, Florens L, Chen S, Zückert WR: Comprehensive Spatial Analysis of the Borrelia burgdorferi Lipoproteome Reveals a Compartmentalization Bias Toward the Bacterial Surface. J Bacteriol 2017, doi:10.1128/JB.00658-16.
111
53. Hooda Y, Shin HE, Bateman TJ, Moraes TF: Neisserial surface lipoproteins: structure, function and biogenesis. Pathog Dis 2017, 75.
54. Glenwright AJ, Pothula KR, Bhamidimarri SP, Chorev DS, Baslé A, Firbank SJ, Zheng H, Robinson CV, Winterhalter M, Kleinekathöfer U, et al.: Structural basis for nutrient acquisition by dominant members of the human gut microbiota. Nature 2017, 541:407– 411.
55. Tuson HH, Foley MH, Koropatkin NM, Biteen JS: The Starch Utilization System Assembles around Stationary Starch-Binding Proteins. Biophys J 2018, 0.
56. Rassam P, Copeland NA, Birkholz O, Tóth C, Chavent M, Duncan AL, Cross SJ, Housden NG, Kaminska R, Seger U, et al.: Supramolecular assemblies underpin turnover of outer membrane proteins in bacteria. Nature 2015, 523:333–336.
57. Roussel-Jazédé V, Jongerius I, Bos MP, Tommassen J, Ulsen P van: NalP-Mediated Proteolytic Release of Lactoferrin-Binding Protein B from the Meningococcal Cell Surface. Infect Immun 2010, 78:3083–3089.
58. Ruiz N, Kahne D, Silhavy TJ: Advances in understanding bacterial outer-membrane biogenesis. Nat Rev Microbiol 2006, 4:57–66.
59. Chen S, Zückert WR: Probing the Borrelia burgdorferi Surface Lipoprotein Secretion Pathway Using a Conditionally Folding Protein Domain. J Bacteriol 2011, 193:6724– 6732.
60. Schulze RJ, Chen S, Kumru OS, Zückert WR: Translocation of Borrelia burgdorferi Surface Lipoprotein OspA Through the Outer Membrane Requires an Unfolded Conformation and Can Initiate at the C-terminus. Mol Microbiol 2010, 76:1266–1278.
61. Konovalova A, Perlman DH, Cowles CE, Silhavy TJ: Transmembrane domain of surface- exposed outer membrane lipoprotein RcsF is threaded through the lumen of β-barrel proteins. Proc Natl Acad Sci U S A 2014, 111:E4350–E4358.
62. Cho S-H, Szewczyk J, Pesavento C, Zietek M, Banzhaf M, Roszczenko P, Asmar A, Laloux G, Hov A-K, Leverrier P, et al.: Detecting envelope stress by monitoring β-barrel assembly. Cell 2014, 159:1652–1664.
63. Turner DPJ, Wooldridge KG, Ala’Aldeen DAA: Autotransported Serine Protease A of Neisseria meningitidis: an Immunogenic, Surface-Exposed Outer Membrane, and Secreted Protein. Infect Immun 2002, 70:4447–4461.
64. Leyton DL, Rossiter AE, Henderson IR: From self sufficiency to dependence: mechanisms and factors important for autotransporter biogenesis. Nat Rev Microbiol 2012, 10:213– 225.
65. Grijpstra J, Arenas J, Rutten L, Tommassen J: Autotransporter secretion: varying on a theme. Res Microbiol 2013, 164:562–582.
112
66. Bernstein HD: Looks can be deceiving: recent insights into the mechanism of protein secretion by the autotransporter pathway. Mol Microbiol 2015, 97:205–215.
67. Weyand NJ, Ma M, Phifer-Rixey M, Taku NA, Rendón MA, Hockenberry AM, Kim WJ, Agellon AB, Biais N, Suzuki TA, et al.: Isolation and characterization of Neisseria musculi sp. nov., from the wild house mouse. Int J Syst Evol Microbiol 2016, 66:3585– 3593.
68. Virji M: Pathogenic neisseriae: surface modulation, pathogenesis and infection control. Nat Rev Microbiol 2009, 7:274–286.
69. Voulhoux R, Bos MP, Geurtsen J, Mols M, Tommassen J: Role of a Highly Conserved Bacterial Protein in Outer Membrane Protein Assembly. Science 2003, 299:262–265.
70. Bos MP, Tefsen B, Geurtsen J, Tommassen J: Identification of an outer membrane protein required for the transport of lipopolysaccharide to the bacterial cell surface. Proc Natl Acad Sci U S A 2004, 101:9417–9422.
71. Giuliani MM, Adu-Bobie J, Comanducci M, Aricò B, Savino S, Santini L, Brunelli B, Bambini S, Biolchi A, Capecchi B, et al.: A universal vaccine for serogroup B meningococcus. Proc Natl Acad Sci 2006, 103:10834–10839.
72. Granoff DM: Review of meningococcal group B vaccines. Clin Infect Dis Off Publ Infect Dis Soc Am 2010, 50 Suppl 2:S54-65.
73. Hood MI, Skaar EP: Nutritional immunity: transition metals at the pathogen–host interface. Nat Rev Microbiol 2012, 10:525–537.
74. Legrain M, Mazarin V, Irwin SW, Bouchon B, Quentin-Millet MJ, Jacobs E, Schryvers AB: Cloning and characterization of Neisseria meningitidis genes encoding the transferrin- binding proteins Tbp1 and Tbp2. Gene 1993, 130:73–80.
75. Retzer MD, Yu R, Schryvers AB: Identification of sequences in human transferrin that bind to the bacterial receptor protein, transferrin-binding protein B. Mol Microbiol [date unknown], 32:111–121.
76. Bonnah RA, Yu R, Schryvers AB: Biochemical analysis of lactoferrin receptors in the Neisseriaceae: identification of a second bacterial lactoferrin receptor protein. Microb Pathog 1996, 19:285–297.
77. Lewis LA, Gray E, Wang YP, Roe BA, Dyer DW: Molecular characterization of hpuAB, the haemoglobin-haptoglobin-utilization operon of Neisseria meningitidis. Mol Microbiol 1997, 23:737–749.
78. Ostan NKH, Yu R-H, Ng D, Lai CC-L, Pogoutse AK, Sarpe V, Hepburn M, Sheff J, Raval S, Schriemer DC, et al.: Lactoferrin binding protein B – a bi-functional bacterial receptor protein. PLOS Pathog 2017, 13:e1006244.
113
79. Brooks CL, Arutyunova E, Lemieux MJ: The structure of lactoferrin-binding protein B from Neisseria meningitidis suggests roles in iron acquisition and neutralization of host defences. Acta Crystallogr Sect F Struct Biol Commun 2014, 70:1312–1317.
80. Calmettes C, Alcantara J, Yu R-H, Schryvers AB, Moraes TF: The structural basis of transferrin sequestration by transferrin-binding protein B. Nat Struct Mol Biol 2012, 19:358–360.
81. Noinaj N, Easley NC, Oke M, Mizuno N, Gumbart J, Boura E, Steere AN, Zak O, Aisen P, Tajkhorshid E, et al.: Structural basis for iron piracy by pathogenic Neisseria. Nature 2012, 483:53–58.
82. Arutyunova E, Brooks CL, Beddek A, Mak MW, Schryvers AB, Lemieux MJ: Crystal structure of the N-lobe of lactoferrin binding protein B from Moraxella bovis. Biochem Cell Biol Biochim Biol Cell 2012, 90:351–361.
83. Wong CT, Xu Y, Gupta A, Garnett JA, Matthews SJ, Hare SA: Structural analysis of haemoglobin binding by HpuA from the Neisseriaceae family. Nat Commun 2015, 6.
84. Masignani V, Comanducci M, Giuliani MM, Bambini S, Adu-Bobie J, Arico B, Brunelli B, Pieri A, Santini L, Savino S, et al.: Vaccination against Neisseria meningitidis using three variants of the lipoprotein GNA1870. J Exp Med 2003, 197:789–799.
85. Madico G, Welsch JA, Lewis LA, McNaughton A, Perlman DH, Costello CE, Ngampasutadol J, Vogel U, Granoff DM, Ram S: The Meningococcal Vaccine Candidate GNA1870 Binds the Complement Regulatory Protein Factor H and Enhances Serum Resistance. J Immunol Baltim Md 1950 2006, 177:501–510.
86. Serruto D, Spadafina T, Ciucchi L, Lewis LA, Ram S, Tontini M, Santini L, Biolchi A, Seib KL, Giuliani MM, et al.: Neisseria meningitidis GNA2132, a heparin-binding protein that induces protective immunity in humans. Proc Natl Acad Sci 2010, 107:3770–3775.
87. Vacca I, Tordello ED, Gasperini G, Pezzicoli A, Fede MD, Paccani SR, Marchi S, Mubaiwa TD, Hartley-Tassell LE, Jennings MP, et al.: Neisserial Heparin Binding Antigen (NHBA) Contributes to the Adhesion of Neisseria meningitidis to Human Epithelial Cells. PLOS ONE 2016, 11:e0162878.
88. Arenas J, Nijland R, Rodriguez FJ, Bosma TNP, Tommassen J: Involvement of three meningococcal surface-exposed proteins, the heparin-binding protein NhbA, the α- peptide of IgA protease and the autotransporter protease NalP, in initiation of biofilm formation. Mol Microbiol 2013, 87:254–268.
89. Casellato A, Paccani SR, Barrile R, Bossi F, Ciucchi L, Codolo G, Pizza M, Aricò B, Bernard M de: The C2 fragment from Neisseria meningitidis antigen NHBA increases endothelial permeability by destabilizing adherens junctions. Cell Microbiol 2014, 16:925–937.
114
90. Cantini F, Veggi D, Dragonetti S, Savino S, Scarselli M, Romagnoli G, Pizza M, Banci L, Rappuoli R: Solution Structure of the Factor H-binding Protein, a Survival Factor and Protective Antigen of Neisseria meningitidis. J Biol Chem 2009, 284:9022–9026.
91. Esposito V, Musi V, Chiara C de, Veggi D, Serruto D, Scarselli M, Kelly G, Pizza M, Pastore A: Structure of the C-terminal Domain of Neisseria Heparin Binding Antigen (NHBA), One of the Main Antigens of a Novel Vaccine against Neisseria meningitidis. J Biol Chem 2011, 286:41767–41775.
92. Hoehn GT, Clark VL: The major anaerobically induced outer membrane protein of Neisseria gonorrhoeae, Pan 1, is a lipoprotein. Infect Immun 1992, 60:4704–4708.
93. Hoehn GT, Clark VL: Isolation and nucleotide sequence of the gene (aniA) encoding the major anaerobically induced outer membrane protein of Neisseria gonorrhoeae. Infect Immun 1992, 60:4695–4703.
94. Boulanger MJ, Murphy MEP: Crystal structure of the soluble domain of the major anaerobically induced outer membrane protein (AniA) from pathogenic Neisseria: a new class of copper-containing nitrite reductases1. J Mol Biol 2002, 315:1111–1127.
95. Ku SC, Schulz BL, Power PM, Jennings MP: The pilin O-glycosylation pathway of pathogenic Neisseria is a general system that glycosylates AniA, an outer membrane nitrite reductase. Biochem Biophys Res Commun 2009, 378:84–89.
96. Cardinale JA, Clark VL: Expression of AniA, the Major Anaerobically Induced Outer Membrane Protein of Neisseria gonorrhoeae, Provides Protection against Killing by Normal Human Sera. Infect Immun 2000, 68:4368–4369.
97. Shewell LK, Ku SC, Schulz BL, Jen FE-C, Mubaiwa TD, Ketterer MR, Apicella MA, Jennings MP: Recombinant truncated AniA of pathogenic Neisseria elicits a non-native immune response and functional blocking antibodies. Biochem Biophys Res Commun 2013, 431:215–220.
98. Van Ulsen P, Van Alphen L, Ten Hove J, Fransen F, Van Der Ley P, Tommassen J: A Neisserial autotransporter NalP modulating the processing of other autotransporters. Mol Microbiol 2003, 50:1017–1030.
99. Echenique-Rivera H, Muzzi A, Tordello ED, Seib KL, Francois P, Rappuoli R, Pizza M, Serruto D: Transcriptome Analysis of Neisseria meningitidis in Human Whole Blood and Mutagenesis Studies Identify Virulence Factors Involved in Blood Survival. PLOS Pathog 2011, 7:e1002027.
100. Del Tordello E, Vacca I, Ram S, Rappuoli R, Serruto D: Neisseria meningitidis NalP cleaves human complement C3, facilitating degradation of C3b and survival in human serum. Proc Natl Acad Sci U S A 2014, 111:427–432.
101. Leuzzi R, Serino L, Scarselli M, Savino S, Fontana MR, Monaci E, Taddei A, Fischer G, Rappuoli R, Pizza M: Ng-MIP, a surface-exposed lipoprotein of Neisseria gonorrhoeae,
115
has a peptidyl-prolyl cis/trans isomerase (PPIase) activity and is involved in persistence in macrophages. Mol Microbiol 2005, 58:669–681.
102. Starnino S, Leuzzi R, Ghisetti V, De Francesco MA, Cusini M, Impara G, Galluppi E, Pizza M, Stefanelli P: Molecular analysis of two novel Neisseria gonorrhoeae virulent components: the macrophage infectivity potentiator and the outer membrane protein A. New Microbiol 2010, 33:167–170.
103. Hung M-C, Salim O, Williams JN, Heckels JE, Christodoulides M: The Neisseria meningitidis Macrophage Infectivity Potentiator Protein Induces Cross-Strain Serum Bactericidal Activity and Is a Potential Serogroup B Vaccine Candidate. Infect Immun 2011, 79:3784–3791.
104. Humbert MV, Mendoza HLA, Jackson AC, Hung M-C, Bielecka MK, Heckels JE, Christodoulides M: Vaccine potential of bacterial macrophage infectivity potentiator (MIP)-like peptidyl prolyl cis/trans isomerase (PPIase) proteins. Expert Rev Vaccines 2015, 14:1633–1649.
105. Prochazkova K, Osicka R, Linhartova I, Halada P, Sulc M, Sebo P: The Neisseria meningitidis Outer Membrane Lipoprotein FrpD Binds the RTX Protein FrpC. J Biol Chem 2005, 280:3251–3258.
106. Sviridova E, Rezacova P, Bondar A, Veverka V, Novak P, Schenk G, Svergun DI, Kuta Smatanova I, Bumba L: Structural basis of the interaction between the putative adhesion-involved and iron-regulated FrpD and FrpC proteins of Neisseria meningitidis. Sci Rep 2017, 7.
107. Ragland SA, Humbert MV, Christodoulides M, Criss AK: Neisseria gonorrhoeae employs two protein inhibitors to evade killing by human lysozyme. PLOS Pathog 2018, 14:e1007080.
108. Zielke RA, Van AL, Baarda BI, Herrera MF, Acosta CJ, Jerse AE, Sikora AE: SliC is a surface-displayed lipoprotein that is required for the anti-lysozyme strategy during Neisseria gonorrhoeae infection. PLOS Pathog 2018, 14:e1007081.
109. Humbert MV, Awanye AM, Lian L-Y, Derrick JP, Christodoulides M: Structure of the Neisseria Adhesin Complex Protein (ACP) and its role as a novel lysozyme inhibitor. PLOS Pathog 2017, 13:e1006448.
110. Moraes TF, Yu R, Strynadka NCJ, Schryvers AB: Insights into the Bacterial Transferrin Receptor: The Structure of Transferrin-Binding Protein B from Actinobacillus pleuropneumoniae. Mol Cell 2009, 35:523–533.
111. Adamiak P, Calmettes C, Moraes TF, Schryvers AB: Patterns of structural and sequence variation within isotype lineages of the Neisseria meningitidis transferrin receptor system. MicrobiologyOpen 2015, 4:491–504.
116
112. Petersen TN, Brunak S, Heijne G von, Nielsen H: SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat Methods 2011, 8:785–786.
113. Slabinski L, Jaroszewski L, Rychlewski L, Wilson IA, Lesley SA, Godzik A: XtalPred: a web server for prediction of protein crystallizability. Bioinformatics 2007, 23:3403–3405.
114. Jones P, Binns D, Chang H-Y, Fraser M, Li W, McAnulla C, McWilliam H, Maslen J, Mitchell A, Nuka G, et al.: InterProScan 5: genome-scale protein function classification. Bioinformatics 2014, 30:1236–1240.
115. Hayat S, Elofsson A: BOCTOPUS: improved topology prediction of transmembrane β barrel proteins. Bioinformatics 2012, 28:516–522.
116. Judd A: Investigating Surface Lipoprotein Translocation in Neisseria meningitidis. 2014,
117. DeSantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, Keller K, Huber T, Dalevi D, Hu P, Andersen GL: Greengenes, a Chimera-Checked 16S rRNA Gene Database and Workbench Compatible with ARB. Appl Environ Microbiol 2006, 72:5069–5072.
118. Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, et al.: Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 2012, 28:1647–1649.
119. Hooda Y, Lai CC-L, Judd A, Buckwalter CM, Shin HE, Gray-Owen SD, Moraes TF: Slam is an outer membrane protein that is required for the surface display of lipidated virulence factors in Neisseria. Nat Microbiol 2016, 1:16009.
120. van den Ent F, Löwe J: RF cloning: A restriction-free method for inserting target genes into plasmids. J Biochem Biophys Methods 2006, 67:67–74.
121. Li C, Wen A, Shen B, Lu J, Huang Y, Chang Y: FastCloning: a highly simplified, purification-free, sequence- and ligation-independent PCR cloning method. BMC Biotechnol 2011, 11:92.
122. Studier FW: Protein production by auto-induction in high density shaking cultures. Protein Expr Purif 2005, 41:207–234.
123. Nørskov-Lauritsen N: Classification, Identification, and Clinical Significance of Haemophilus and Aggregatibacter Species with Host Specificity for Humans. Clin Microbiol Rev 2014, 27:214–240.
124. Gray-Owen SD, Schyvers AB: Bacterial transferrin and lactoferrin receptors. Trends Microbiol 1996, 4:185–191.
125. Yu R, Schryvers AB: The interaction between human transferrin and transferrin binding protein 2 from Moraxella (Branhamella) catarrhalis differs from that of other human pathogens. Microb Pathog 1993, 15:433–445.
117
126. Gray-Owen SD, Loosmore S, Schryvers AB: Identification and characterization of genes encoding the human transferrin-binding proteins from Haemophilus influenzae. Infect Immun 1995, 63:1201–1210.
127. Jones P, Binns D, Chang H-Y, Fraser M, Li W, McAnulla C, McWilliam H, Maslen J, Mitchell A, Nuka G, et al.: InterProScan 5: genome-scale protein function classification. Bioinformatics 2014, 30:1236–1240.
128. Fleury C, Su Y-C, Hallström T, Sandblad L, Zipfel PF, Riesbeck K: Identification of a Haemophilus influenzae factor H-Binding lipoprotein involved in serum resistance. J Immunol Baltim Md 1950 2014, 192:5913–5923.
129. Cowles CE, Goodrich-Blair H: Characterization of a lipoprotein, NilC, required by Xenorhabdus nematophila for mutualism with its nematode host. Mol Microbiol 2004, 54:464–477.
130. Bhasin A, Chaston JM, Goodrich-Blair H: Mutational Analyses Reveal Overall Topology and Functional Regions of NilB, a Bacterial Outer Membrane Protein Required for Host Association in a Model of Animal-Microbe Mutualism. J Bacteriol 2012, 194:1763– 1776.
131. Nardini PM, Mellors A, Lo RY: Characterization of a fourth lipoprotein from Pasteurella haemolytica A1 and its homology to the OmpA family of outer membrane proteins. FEMS Microbiol Lett 1998, 165:71–77.
132. Wu J-R, Shien J-H, Shieh HK, Chen C-F, Chang P-C: Protective immunity conferred by recombinant Pasteurella multocida lipoprotein E (PlpE). Vaccine 2007, 25:4140–4148.
133. Cuesta-Seijo JA, Neale C, Khan MA, Moktar J, Tran CD, Bishop RE, Pomès R, Privé GG: PagP Crystallized from SDS/Cosolvent Reveals the Route for Phospholipid Access to the Hydrocarbon Ruler. Structure 2010, 18:1210–1219.
134. Calmettes C, Yu R, Silva LP, Curran D, Schriemer DC, Schryvers AB, Moraes TF: Structural Variations within the Transferrin Binding Site on Transferrin-binding Protein B, TbpB. J Biol Chem 2011, 286:12683–12692.
135. Frandoloso R, Martínez-Martínez S, Calmettes C, Fegan J, Costa E, Curran D, Yu R, Gutiérrez-Martín CB, Rodríguez-Ferri EF, Moraes TF, et al.: Nonbinding Site-Directed Mutants of Transferrin Binding Protein B Exhibit Enhanced Immunogenicity and Protective Capabilities. Infect Immun 2015, 83:1030–1038.
136. Cendron L, Veggi D, Girardi E, Zanotti G: Structure of the uncomplexed Neisseria meningitidis factor H-binding protein fHbp (rLP2086). Acta Crystallograph Sect F Struct Biol Cryst Commun 2011, 67:531–535.
137. Johnson S, Tan L, Veen S van der, Caesar J, Jorge EGD, Harding RJ, Bai X, Exley RM, Ward PN, Ruivo N, et al.: Design and Evaluation of Meningococcal Vaccines through
118
Structure-Based Modification of Host and Pathogen Molecules. PLOS Pathog 2012, 8:e1002981.
138. Konar M, Pajon R, Beernink PT: A meningococcal vaccine antigen engineered to increase thermal stability and stabilize protective epitopes. Proc Natl Acad Sci 2015, 112:14823– 14828.
139. Scarselli M, Aricò B, Brunelli B, Savino S, Marcello FD, Palumbo E, Veggi D, Ciucchi L, Cartocci E, Bottomley MJ, et al.: Rational Design of a Meningococcal Antigen Inducing Broad Protective Immunity. Sci Transl Med 2011, 3:91ra62-91ra62.
140. Schneider MC, Prosser BE, Caesar JJE, Kugelberg E, Li S, Zhang Q, Quoraishi S, Lovett JE, Deane JE, Sim RB, et al.: Neisseria meningitidis recruits factor H using protein mimicry of host carbohydrates. Nature 2009, 458:890–893.
141. Stohl EA, Criss AK, Seifert HS: The transcriptome response of Neisseria gonorrhoeae to hydrogen peroxide reveals genes with previously uncharacterized roles in oxidative damage protection. Mol Microbiol 2005, 58:520–532.
142. Cornelissen CN: Subversion of nutritional immunity by the pathogenic Neisseriae. Pathog Dis 2018, 76.
143. Jin S, Joe A, Lynett J, Hani EK, Sherman P, Chan VL: JlpA, a novel surface-exposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol Microbiol 2001, 39:1225–1236.
144. Wójtowicz H, Guevara T, Tallant C, Olczak M, Sroka A, Potempa J, Solà M, Olczak T, Gomis-Rüth FX: Unique Structure and Stability of HmuY, a Novel Heme-Binding Protein of Porphyromonas gingivalis. PLOS Pathog 2009, 5:e1000419.
145. Shipman JA, Berleman JE, Salyers AA: Characterization of Four Outer Membrane Proteins Involved in Binding Starch to the Cell Surface of Bacteroides thetaiotaomicron. J Bacteriol 2000, 182:5365–5372.
146. Ståhl S, Uhlén M: Bacterial surface display: trends and progress. Trends Biotechnol 1997, 15:185–192.
147. Jacob-Dubuisson F, Guérin J, Baelen S, Clantin B: Two-partner secretion: as simple as it sounds? Res Microbiol 2013, 164:583–595.
148. Guérin J, Bigot S, Schneider R, Buchanan SK, Jacob-Dubuisson F: Two-Partner Secretion: Combining Efficiency and Simplicity in the Secretion of Large Proteins for Bacteria- Host and Bacteria-Bacteria Interactions. Front Cell Infect Microbiol 2017, 7.
149. Rahman S ur, Ulsen P van: System Specificity of the TpsB Transporters of Coexpressed Two-Partner Secretion Systems of Neisseria meningitidis. J Bacteriol 2013, 195:788–797.
119
150. Maier T, Clantin B, Gruss F, Dewitte F, Delattre A-S, Jacob-Dubuisson F, Hiller S, Villeret V: Conserved Omp85 lid-lock structure and substrate recognition in FhaC. Nat Commun 2015, 6:7452.
151. Fan E, Fiedler S, Jacob-Dubuisson F, Müller M: Two-partner Secretion of Gram-negative Bacteria A SINGLE β-BARREL PROTEIN ENABLES TRANSPORT ACROSS THE OUTER MEMBRANE. J Biol Chem 2012, 287:2591–2599.
152. Baud C, Guérin J, Petit E, Lesne E, Dupré E, Locht C, Jacob-Dubuisson F: Translocation path of a substrate protein through its Omp85 transporter. Nat Commun 2014, 5.
153. Noinaj N, Kuszak AJ, Balusek C, Gumbart JC, Buchanan SK: Lateral Opening and Exit Pore Formation Are Required for BamA Function. Structure 2014, 22:1055–1062.
154. Inukai M, Takeuchi M, Shimizu K, Arai M: Mechanism of action of globomycin. J Antibiot (Tokyo) 1978, 31:1203–1205.
155. Taniguchi N, Matsuyama S, Tokuda H: Mechanisms Underlying Energy-independent Transfer of Lipoproteins from LolA to LolB, Which Have Similar Unclosed β-Barrel Structures. J Biol Chem 2005, 280:34481–34488.
156. Little DJ, Bamford NC, Pokrovskaya V, Robinson H, Nitz M, Howell PL: Structural basis for the De-N-acetylation of Poly-β-1,6-N-acetyl-D-glucosamine in Gram-positive bacteria. J Biol Chem 2014, 289:35907–35917.
157. Konovalova A, Perlman DH, Cowles CE, Silhavy TJ: Transmembrane domain of surface- exposed outer membrane lipoprotein RcsF is threaded through the lumen of β-barrel proteins. Proc Natl Acad Sci U S A 2014, 111:E4350–E4358.
158. Chin JW, Martin AB, King DS, Wang L, Schultz PG: Addition of a photocrosslinking amino acid to the genetic code of Escherichia coli. Proc Natl Acad Sci U S A 2002, 99:11020–11024.
159. Chin JW, Schultz PG: In Vivo Photocrosslinking with Unnatural Amino Acid Mutagenesis. ChemBioChem 2002, 3:1135–1137.
160. Hagan CL, Silhavy TJ, Kahne D: β-Barrel Membrane Protein Assembly by the Bam Complex. Annu Rev Biochem 2011, 80:189–210.
161. Roman-Hernandez G, Peterson JH, Bernstein HD: Reconstitution of bacterial autotransporter assembly using purified components. eLife 2014, 3:e04234.
162. Norell D, Heuck A, Tran-Thi T-A, Götzke H, Jacob-Dubuisson F, Clausen T, Daley DO, Braun V, Müller M, Fan E: Versatile in vitro system to study translocation and functional integration of bacterial outer membrane proteins. Nat Commun 2014, 5.
163. Hooda Y, Lai CCL, Moraes TF: Identification of a Large Family of Slam-Dependent Surface Lipoproteins in Gram-Negative Bacteria. Front Cell Infect Microbiol 2017, 7.
120
164. Noinaj N, Gumbart JC, Buchanan SK: The β-barrel assembly machinery in motion. Nat Rev Microbiol 2017, 15:197–204.
165. Celia H, Noinaj N, Zakharov SD, Bordignon E, Botos I, Santamaria M, Barnard TJ, Cramer WA, Lloubes R, Buchanan SK: Structural insight into the role of the Ton complex in energy transduction. Nature 2016, 538:60–65.
166. Ostberg KL, DeRocco AJ, Mistry SD, Dickinson MK, Cornelissen CN: Conserved regions of gonococcal TbpB are critical for surface exposure and transferrin iron utilization. Infect Immun 2013, 81:3442–3450.
167. Noinaj N, Kuszak AJ, Gumbart JC, Lukacik P, Chang H, Easley NC, Lithgow T, Buchanan SK: Structural insight into the biogenesis of β-barrel membrane proteins. Nature 2013, 501:385–390.
168. Gessmann D, Chung YH, Danoff EJ, Plummer AM, Sandlin CW, Zaccai NR, Fleming KG: Outer membrane β-barrel protein folding is physically controlled by periplasmic lipid head groups and BamA. Proc Natl Acad Sci 2014, doi:10.1073/pnas.1322473111.
169. Karuppiah V, Berry J-L, Derrick JP: Outer membrane translocons: structural insights into channel formation. Trends Microbiol 2011, 19:40–48.
170. Clantin B, Delattre A-S, Rucktooa P, Saint N, Méli AC, Locht C, Jacob-Dubuisson F, Villeret V: Structure of the membrane protein FhaC: a member of the Omp85-TpsB transporter superfamily. Science 2007, 317:957–961.
171. Henrich E, Peetz O, Hein C, Laguerre A, Hoffmann B, Hoffmann J, Dötsch V, Bernhard F, Morgner N: Analyzing native membrane protein assembly in nanodiscs by combined non-covalent mass spectrometry and synthetic biology. eLife [date unknown], 6.
172. Dong H, Xiang Q, Gu Y, Wang Z, Paterson NG, Stansfeld PJ, He C, Zhang Y, Wang W, Dong C: Structural basis for outer membrane lipopolysaccharide insertion. Nature 2014, 511:52–56.
173. Costello SM, Plummer AM, Fleming PJ, Fleming KG: Dynamic periplasmic chaperone reservoir facilitates biogenesis of outer membrane proteins. Proc Natl Acad Sci 2016, 113:E4794–E4800.
174. Volokhina EB, Grijpstra J, Stork M, Schilders I, Tommassen J, Bos MP: Role of the Periplasmic Chaperones Skp, SurA, and DegQ in Outer Membrane Protein Biogenesis in Neisseria meningitidis. J Bacteriol 2011, 193:1612–1621.
175. Baba T, Ara T, Hasegawa M, Takai Y, Okumura Y, Baba M, Datsenko KA, Tomita M, Wanner BL, Mori H: Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol Syst Biol 2006, 2:2006.0008.
176. Podgornaia AI, Laub MT: Pervasive degeneracy and epistasis in a protein-protein interface. Science 2015, 347:673–677.
121
177. Moraes I, Evans G, Sanchez-Weatherby J, Newstead S, Stewart PDS: Membrane protein structure determination — The next generation. Biochim Biophys Acta 2014, 1838:78–87.
178. Smith MTJ, Rubinstein JL: Beyond blob-ology. Science 2014, 345:617–619.
179. Schneider A: Dihydrofolate reductase and membrane translocation: evolution of a classic experiment. EMBO Rep 2018, 19:e45692.
180. Li L, Park E, Ling J, Ingram J, Ploegh H, Rapoport TA: Crystal structure of a substrate- engaged SecY protein-translocation channel. Nature 2016, 531:395–399.
181. Voorhees RM, Hegde RS: Structure of the Sec61 channel opened by a signal sequence. Science 2016, 351:88–91.
122