Antibody recognition of bacterial surfaces and extracellular polysaccharides

Available online at www.sciencedirect.com

ScienceDirect Antibody recognition of bacterial surfaces and extracellular polysaccharides Caroline Soliman1, Gerald B Pier2 and Paul A Ramsland1,3,4 Because of the ongoing increase in antibiotic-resistant microbes, new strategies such as therapeutic antibodies and effective vaccines are required. Bacterial carbohydrates are known to be particularly antigenic, and several monoclonal antibodies that target bacterial polysaccharides have been generated, with more in current development. This review examines the known 3D crystal structures of anti-bacterial antibodies and the structural basis for carbohydrate recognition and explores the potential mechanisms for antibody-dependent bacterial cell death. Understanding the key interactions between an antibody and its polysaccharide target on the surface of bacteria or in biofilms can provide essential information for the development of more specific and effective antibody therapeutics as well as carbohydratebased vaccines. Addresses 1 School of Science, RMIT University, Melbourne, Victoria, Australia 2 Division of Infectious Diseases, Department of Medicine, Brigham and Women’s Hospital/Harvard Medical School, Boston, MA, United States 3 Department of Immunology, Central Clinical School. Monash University, Melbourne, Victoria, Australia 4 Department of Surgery Austin Health, University of Melbourne, Melbourne, Victoria, Australia Corresponding author: Ramsland, Paul A ([email protected])

Current Opinion in Structural Biology 2020, 62:48–55 This review comes from a themed issue on Carbohydrates

increasingly difficult to combat, and new therapies are greatly needed [1

,2–4]. Often used for targeted therapies, monoclonal antibodies (mAbs) are produced by single clones of B lineage cells, with each mAb having a unique sequence and the capacity to bind a single epitope on a target antigen. To date, mAbs have seen many successes, particularly in the treatment of cancers and autoimmune diseases [5–8]. More recently, efforts are turning to using mAbs as therapies for infectious diseases where antibiotics, anti-infectives, and vaccines have been the mainstays of clinical management and prevention [9]. Carbohydrates are regarded as promising targets for antibody and vaccine development against infectious diseases because cell surface glycans have been shown to play important roles in pathogenesis, with most pathogens having dense distributions of complex polysaccharides, oligosaccharides and glycans on their cell surface [10]. While most of the approved therapeutic antibodies target protein or glycoprotein antigens, it has been known for many years that carbohydrate-binding antibodies are capable of specifically recognising altered self-cells and are effective in providing protective responses to pathogens [11,12]. Progress has been slow when delivering carbohydrate-binding mAbs as therapeutics for bacterial infections, although a number have been generated and structurally characterised.

Edited by Sony Malhotra and Paul Ramsland

https://doi.org/10.1016/j.sbi.2019.12.001 0959-440X/ã 2019 Elsevier Ltd. All rights reserved.

Introduction Bacterial pathogens can cause different conditions that range in severity, and for the most part these conditions have been treatable through the use of antibiotics. However, antibiotic resistance has become a common occurrence and is a serious problem for the treatment and control of infectious diseases [1

,2–4]. Resistance can occur via natural mechanisms, acquisition of resistance factors via horizontal gene transfer or through mutations but is mainly thought to be promoted by selective pressures from the misuse of antibiotics. As a result, many bacterial infections are becoming Current Opinion in Structural Biology 2020, 62:48–55

Carbohydrate-based antibody targets on the surface of bacteria fall into two major classes: (1) Capsular polysaccharides (CPS), which are generally long repeating saccharide structures that surround and protect many bacteria and contribute to cellular adhesion, and (2) lipopolysaccharides (LPS) that consist of a common core saccharide proximal to the lipid and an O-antigen (O-Ag) that differs between species and strains of bacteria. Depending on the bacterial species both CPS and LPS glycans can vary greatly in saccharide composition and structure. As examples, clinical isolates of Salmonella may have 1 type of CPS (known as Vi antigen; many CPS variants likely remain undefined) but have 1 of greater than 40 different LPS glycans, while Escherichia coli isolates have been characterised with more than 70 CPS and 170 LPS surface glycans [13]. All bacterial cell walls are additionally composed of a layer or matrix of peptidoglycans where the glycan component consists of alternating N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) saccharides. The thickness and location of the peptidoglycan layer distinguishes the two major classes of bacteria as either www.sciencedirect.com

Antibody recognition of bacterial glycans Soliman, Pier and Ramsland 49

Gram-positive (thick external peptidoglycan layer) or Gram-negative (thin peptidoglycan layer located beneath a protein and LPS containing outer membrane). Both Gram-positive and Gram-negative bacteria can have CPS as capsules or slime layers. Since many bacteria express and excrete polysaccharides, carbohydrates also represent a major component of biofilms. These extracellular polysaccharides (exopolysaccharides) are also targets for therapeutic antibodies [14–17,18
]. Vaccines based on purified CPS from pneumococcal bacteria were first tested in the 1930s and used clinically shortly after World War II, but with the dawn of the antibiotic era their use was stopped [19]. Both pneumococcal and meningococcal purified capsules have been licensed for use since the 1970s, but most of these purified polysaccharides were not immunogenic in infants and children under two years of age [20
]. Conjugate vaccines (carbohydrate linked to a carrier protein) usually induce protective immunity in all age groups and were first shown to be effective in infants with the introduction of the Haemophilus influenzae type b (Hib) conjugate vaccine in 1990 [20
]. Conjugate vaccines for meningococcal and pneumococcal bacteria have now replaced the CPS vaccines in infant and toddler groups [20
]. Additional conjugate vaccines have been explored, yet a number of factors have limited the potential of carbohydrate-based vaccines. One of the issues is that the cost of production of a carbohydrate-based conjugate vaccine is high compared with antibiotics and anti-infectives. However, with the widespread emergence of antimicrobial resistance there is renewed interest in developing vaccines and therapeutic antibodies to treat bacterial infections [1

,9,18
]. Structural diversity of bacterial polysaccharides results in multiple pathogenic serotypes for each genus and species that require inclusion in a vaccine candidate, which further complicates the complexity of production and associated costs [13,21]. An alternative approach is to drive the

discovery of conserved strain-transcending epitopes and antibodies as has been the goal in the ongoing search for a cure for HIV-1 and malaria [22–25]. Additionally, not all carbohydrate epitopes induce a protective immune response and there is a need to define the structural and molecular basis for protective versus non-protective anticarbohydrate antibodies to accelerate vaccine and therapeutic antibody development for bacterial infections. This review examines a number of known 3D anti-bacterial antibody crystal structures and their features of carbohydrate recognition and explores potential mechanisms around antibody-dependent bacterial cell death. Understanding the key interactions between an antibody and its polysaccharide target on the surface of bacteria or in biofilms can provide essential information for the development of more selective and specific antibody therapeutics.

Overview of known structures of bacterial carbohydrate-binding mAbs Herein we focus on 12 known crystal structures of antibodies in complex with a variety of bacterial antigen targets. The key details of each anti-bacterial antibody structure are presented in Table 1. The crystal structures were determined at resolutions between 1.5 and 2.7 A˚ and are of the fragment antigen binding (Fab) or single-chain fragment variable (scFv) in complex with glycans that range in size from monosaccharides to decasaccharides. This range of carbohydrate sizes gives an idea of how large saccharide epitopes are recognised by existing antibodies, and how much of the cognate polysaccharide chains are in contact or ordered by the interaction with the antibody binding site. In addition to those structures described in Table 1, several additional structures of anti-Chlamydia antibodies related to S25-2 (PDB ID 3SY0) and S25-26 (PDB ID 4M7J) have been determined, including S45-18 (PDB ID 1Q9W), S54-10 (PDB ID 3I02) and S73-2 (PDB IDs 3HZK, 3HZM, 3HZV, 3HZY) each in complex with chlamydial

Table 1 Examples of previously determined crystal structures of bacterial saccharide-antibody complexes PDB ID 1M7I 3BZ4 1MFC 3HNS 3SY0 4M7J 4C83 3WBD 5M63 6DW2 6BE3 6BE4

Target antigen Shigella flexneri Y lipopolysaccharide Decasaccharide from Shigella flexneri 2a O-Oligosaccharide from Salmonella typhimurium serogroup B Hexasaccharide oligoarabinofuranosyl from mycobacteria Kdoa2-8Kdoa2-4Kdo trisaccharide from Chlamydia Kdoa2-8Kdoa2-4Kdo trisaccharide from Chlamydia Neisseria meningitidis lipooligosaccharide Group B meningitidis polysaccharide DP2 oligosaccharide from group B streptococcus type III Teichoic acid (b-WTA) from MRSA Staphylococcus aureus Microbial N-acetyl-D-glucosamine Microbial Nona-N-acetyl-D-glucosamine

Antibody name SYA/J-6 F22-4 Se155-4 CS-35 S25-2 b S25-26 LPT3-1 mAb735 NVS-1-19-5 6078 c F598 F598

Fragment a

Murine Fab (IgG3/l) Murine Fab (IgG1/k) Murine Fab (IgG1/l) Murine Fab (IgG3/k) Murine Fab (IgG1/k) Murine Fab (IgG1/k) Murine Fab (IgG2a/k) Murine scFv (IgG2a/k) Rabbit Fab (IgG/k) Human Fab (IgG1/k) Human Fab (IgG1/l) Human Fab (IgG1/l)

Resolution (A˚)

Reference

2.5 1.8 2.1 2.0 1.5 2.0 2.7 1.8 2.7 1.7 1.6 1.9

[35] [36] [39] [37] [26] [38] [40] [41] [42

] [28

] [45

] [45

]

a

Abbreviations used: Fab, fragment antigen binding; scFv, single chain variable fragment. There are multiple related structures to S25-2 (PDB ID 3SY0) and S25-26 (PDB ID 4M7J), including S45-18 (PDB ID 1Q9W), S54-10 (PDB ID 3I02) and S73-2 (PDB IDs 3HZK, 3HZM, 3HZV and 3HZY). c There are also related mAb structures to 6078 (PDB ID 6DW2), being 4462 (PDB ID 6DWI) and 4497 (PDB IDs 5D6C and 6DWA). b

www.sciencedirect.com

Current Opinion in Structural Biology 2020, 62:48–55

50 Carbohydrates

lipopolysaccharide epitopes [26,27]. For anti-Staphylococcus aureus antibody 6078 (PDB ID 6DW2), structures have been determined of mAbs 4462 (PDB ID 6DWI) and 4497 (PDB IDs 5D6C and 6DWA) in complex with the same carbohydrate ligand [28

]. We discuss here the overall binding modes and summarise the interactions with antibody. For additional molecular details on each system, we refer readers to the primary citations (Table 1) and an excellent review on antibody recognition of carbohydrates [29]. While the majority of known bacterial-saccharide antibody structures describe murine antibodies and one rabbit antibody, only five structures are of human antibodies (three structures are listed in Table 1). It has previously been established that there are significant differences in the complementarity-determining region (CDR) length of mouse antibodies when compared to human antibodies [30]. In particular, the mean length of murine CDR H3 is significantly shorter than that of human CDR H3, and the few antibodies that are identical in length instead differ in amino acid composition [31]. Therefore, future research should concentrate on defining human antibodies to each of the different types of bacterial saccharides, which will provide a better understanding of the structural nature of the human antibody response to glycans.

Carbohydrate-binding modes displayed by bacterial targeting mAbs An association exists between the overall shape of the antigen binding site and the nature of the antigen. For example, antibodies that bind to proteins have a relatively

flat binding site, while those that bind to peptides frequently have groove-type binding sites [32]. Additionally, it was previously proposed that anti-bacterial antibodies recognise their carbohydrate ligands in a relatively conserved manner similar to haptens in cavity or groove shaped binding sites [29,33]. However, after inspection of the available structures of antibody-carbohydrate complexes to bacterial targets, the diversity of recognition modes is clearly higher than previously assumed. Carbohydrate recognition by antibodies can be categorised into at least four main types: (1) end-on insertion, (2) bifurcated binding, (3) groove-type binding, and (4) combination binding (Figure 1). End-on insertion involves binding of terminal portions of longer saccharide chains into a deep binding pocket that is often centrally located in the antigen-binding site. The remaining proportion of the ligand tends to protrude from the binding site and does not participate in direct contacts with the antibody (Figure 1a). Similarly, bifurcated binding involves the branched ends of carbohydrate ligands interacting with the antibody in two binding pockets. Again, the tail portion of the ligand is often situated to protrude from the binding pocket (Figure 1b). Comparatively, groove-type binding occurs when the carbohydrate chain binds laterally in an extended valley in the antigen-binding site. In groovetype binding, the carbohydrate often sits flat against the antibody (i.e. not protruding from the binding site) and curves around any CDRs and side-chains of residues that project into the groove. Furthermore, groove-type binding can result in the greatest number of direct interactions with the antibody (Figure 1c). Combination binding modes involve contributions from each of the other binding modes, incorporating both binding pockets (suited for end-on

Figure 1

(a)

(b)

(c)

(d)

Current Opinion in Structural Biology

Structural comparison of bacterial carbohydrate-binding antibodies showing a range of antigen binding modes. (a) End-on insertion binding of S25-2 mAb to a Kdo trisaccharide from Chlamydia (PDB ID: 3SY0). (b) Bifurcated binding of CS-35 mAb to the hexasaccharide oligoarabinofuranosyl from mycobacteria (PDB ID: 3HNS). (c) Groove-type binding of F598 mAb to a pentasaccharide epitope of the microbial polysaccharide PNAG (PDB ID: 6BE4). (d) Combination-type binding of NVS-1-19-5 to the DP2 oligosaccharide from group B Streptococcus type III (PDB ID: 5M63). The variable regions (VL and VH) of each antibody are represented as transparent solvent-accessible surfaces (side views) with the light chain in blue and heavy chain in grey. All ligands are represented as sticks coloured by atom type (C, green; O, red; N, blue). Current Opinion in Structural Biology 2020, 62:48–55

www.sciencedirect.com

Antibody recognition of bacterial glycans Soliman, Pier and Ramsland 51

insertion) and grooves (lateral binding). As two separate sections of a branched carbohydrate are involved this also represents a form of bifurcated ligand interaction. Often in longer oligosaccharide chains there are short branches emerging from extended polysaccharide chains that are suited for combination binding modes (Figure 1d). Similar modes of binding have also been observed for non-bacterial targeting anticarbohydrate antibodies [29,34].

Structures of bacterially derived saccharides in complex with antibodies Two antibodies have been developed to Shigella flexneri, SYA/J-6 mAb binds to Y lipopolysaccharide and F22-4 mAb to serotype 2a O-antigen. Interestingly, both antibodies participate in groove-type binding of their respective ligands

utilising 5 of the 6 CDRs. The SYA/J-6 Fab makes contacts with all sugar residues of the pentasaccharide [35], while F22-4 makes contacts with six sugar residues of a longer branched decasaccharide ligand (contacts derived from Fab with chains labelled A and B) [36] (Figure 2a,b). In total, SYA/J-6 forms 8 hydrogen bonds and 2 water-mediated hydrogen bonds with the bound carbohydrate out of a total of 74 contacts (55% of the ligand is involved in the interaction; based on carbohydrate atoms within 4 A˚ of an antibody atom). The F22-4 mAb participates in 11 hydrogen bonds and 14 water-mediated hydrogen bonds (29% of the ligand is involved in the interaction). Comparatively, CS-35 mAb, which recognises surface glycolipid lipoarabinomannans (LAMs) on mycobacteria, displays bifurcated binding with the antigen-binding site forming three binding pockets

Figure 2

(a)

(b)

(c)

(d)

(e)

(f)

Current Opinion in Structural Biology

Structural comparison of bacterial carbohydrate-binding antibodies. (a) SYA/J-6 mAb with a pentasaccharide from Shigella flexneri Y lipopolysaccharide (PDB ID: 1M7I). (b) F22-4 mAb in complex with a decasaccharide from Shigella flexneri 2a (PDB ID: 3BZ4). (c) Se155-4 mAb in complex with a cell-surface O-oligosaccharide from Salmonella typhimurium serogroup B (PDB ID: 1MFC). (d) CS-35 mAb in complex with the hexasaccharide oligoarabinofuranosyl from mycobacteria (PDB ID: 3HNS). (e) S25-2 mAb in complex with a Kdo trisaccharide from Chlamydia (PDB ID: 3SY0). (f) S25-26 mAb in complex with a Kdo trisaccharide from Chlamydia (PDB ID: 4M7J). The variable regions (VL and VH) of each antibody are depicted as surfaces (end-on views) with the light chain in blue and heavy chain in grey. All ligands are represented as light grey sticks with contact atoms in orange using a 4-A˚ cutoff from any antibody atom. www.sciencedirect.com

Current Opinion in Structural Biology 2020, 62:48–55

52 Carbohydrates

(a broad groove and two ‘specificity’ pockets) involving all 6 CDRs. Despite the binding conformation, all residues of the Y-shaped hexasaccharide contact the antibody, with the hexasaccharide binding with the strongest affinity of all ligands tested (63% of the ligand is involved in the interaction) [37] (Figure 2d). Two antibodies have been developed to Chlamydia, S25-2 and S25-26 mAbs, both bound to a trisaccharide of 3-deoxyD-manno-oct-2-ulosonic acid (Kdo). While S25-2 mAb binds the Kdo trisaccharide by end-on insertion and only makes contacts with two of the sugar residues (45% of the ligand is involved in the interaction) [26], S25-26 mAb recognises the entire trisaccharide by groove-type binding (53% of the ligand is involved in the interaction) [38] (Figure 2e,f).

Interestingly, in both structures only 4 of the 6 CDRs are involved in contacts, but S25-26 mAb interacts with its carbohydrate ligand using 13 hydrogen bonds and 16 bridging water interactions. Comparatively, Se155-4 recognises the O-chain from Salmonella typhimurium serogroup B by end-on insertion, with only a small portion of the sugar making antibody contacts involving 5 of the 6 CDRs and a number of water molecules (50% of the ligand is involved in the interaction) [39] (Figure 2c). Similarly, LPT3-1 mAb recognises the lipooligosaccharide inner cores from Neisseria meningitidis by end-on insertion with minimal antibody contact. Antibody-antigen interactions are mediated almost entirely through the heavy chain with a total of 9 hydrogen bonds (34% of the ligand is involved in the interaction) [40] (Figure 3a).

Figure 3

(a)

(b)

(c)

(d)

(e)

(f)

Current Opinion in Structural Biology

Structural comparison of bacterial carbohydrate-binding antibodies. (a) LPT3-1 mAb in complex with Neisseria meningitidis lipooligosaccharide (PDB ID: 4C83). (b) mAb735 with half of octasialic acid, a homopolymer of group B meningitidis polysaccharide (PDB ID: 3WBD). (c) NVS-1-19-5 mAb in complex with the DP2 oligosaccharide from group B Streptococcus type III (PDB ID: 5M63). (d) Antibody 6078 in complex with teichoic acid from Staphylococcus aureus (PDB ID: 6DW2). (e) F598 mAb in complex with the N-acetyl-D-glucosamine monosaccharide (PDB ID: 6BE3). (f) F598 in complex with a pentameric epitope of microbial PNAG (PDB ID: 6BE4). The variable regions (VL and VH) of each antibody are depicted as surfaces (end-on views) with the light chain in blue and heavy chain in grey. All ligands are represented as light grey sticks with contact atoms in orange using a 4-A˚ cutoff from any antibody atom. Current Opinion in Structural Biology 2020, 62:48–55

www.sciencedirect.com

Antibody recognition of bacterial glycans Soliman, Pier and Ramsland 53

The mAb735 antibody targets the main polysialic acid component of the CPS of N. meningitides. Interestingly, one octasaccharide ligand, folded up on itself, is simultaneously recognised by two antibodies in a paired manner. Direct interactions between mAb735 and the long polysialic acid chain are relatively few and involve 5 of the 6 CDRs, although several ordered water molecules participate in bridging interactions with antibody and ligand (46% of the ligand is involved in the interaction) [41] (Figure 3b). In contrast, NVS-1-19-5 mAb recognises a very long chain of Group B Streptococcus type III CPS by a combination binding mode. Only a small fraction of the long polysaccharide chain contacts the antibody through 4 of the 6 CDRs, with the terminal disaccharide (sialic acid and galactose) binding by end-on-insertion where sialic acid establishes most of the direct binding interactions (contacts derived from Fab with chains H and L). The longer oligosaccharide chain is bound in a shallow groove partially formed by CDR residues with the majority of the glycan chain not making contacts with the Fab (24% of the ligand is involved in the interaction) [42

] (Figure 3c). Interestingly, antibody 6078 binds to wall teichoic acid (b-WTA) from S. aureus in a binding pocket formed primarily by the heavy chain. The relatively small ligand (N-acetyl glucosamine, GlcNAc, b-linked to ribitol-1-phosphate) is shielded in the binding site by the extended CDR3 loop of the H chain and makes numerous contacts with the antibody involving 4 of the 6 CDRs (82% of ligand is involved in the interaction) [28

] (Figure 3d). Two other S. aureus targeting antibodies (4462 and 4497) bind the GlcNAc residue of b-WTA by end-on insertion in cavities formed between the LCDR and HCDR loops [28

,43]. One antibody, F598, uniquely targets a broadly expressed microbial polysaccharide called poly-N-acetylglucosamine (PNAG), which is found in bacteria (Gram negative and positive), fungi, and protozoans, and as an exopolysaccharide in many biofilms [44]. We recently determined high resolution structures of the F598 Fab in complex with a GlcNAc monosaccharide and a synthetic nonasaccharide version of PNAG, where a core pentasaccharide epitope completely fills the binding site. F598 displays groove-type binding, with the GlcNAc monosaccharide sitting in the centre of the binding site (93% of the ligand is involved in the interaction). However, the large binding groove of F598 accommodates and makes contacts with all sugar residues of the pentasaccharide epitope in an extended conformation involving all 6 CDRs (63% of the ligand is involved in the interaction). The monosaccharide is almost perfectly aligned with one of the GlcNAc residues from the pentasaccharide, which suggests an anchored binding mechanism is employed for recognition of PNAG by the F598 mAb [45

] (Figure 3e,f).

Structural considerations of protective immunity by therapeutic antibodies Currently there is an incomplete understanding of what makes the difference between a protective antibody response (i.e. protects against infectious disease) and a non-protective www.sciencedirect.com

antibody response. However, all existing successful antibacterial vaccines for humans have the in vitro correlate of being able to kill the target organism by either bactericidal or opsonophagocytic activity, but not all antibodies with these properties are protective. One of the most structurally characterized systems is the murine immune response to Chlamydia LPS where multiple related murine mAbs have been structurally characterised with the structural role of germline antibody sequences, and mutations responsible for high affinity and selective recognition have been elucidated [46]. Similarly, the other available structures all use Fab or scFv rather than the intact antibody, which due to its inherent flexibility is difficult to crystallise. Thus, while we have gained a detailed appreciation of the structural basis for carbohydrate recognition it is not yet clear why certain mAbs are protective while others of similar specificity are non-protective. Even when the fragment crystallisable (Fc) has the same sequence, antibodies capable of binding the same bacterial saccharide can display a range of antibody-dependent effector functions. For example, three antibodies (F628, F630 and F598) isolated from an individual after S. aureus bacteraemia and engineered as IgG1 (g1 heavy chains) all bind PNAG but show different correlates of protective immunity, with F598 mAb displaying superior complement deposition and opsonophagocytic activity compared to F628 and F630. The F598 mAb also binds strongly to chemically deacetylated PNAG (dPNAG) while F628 and F630 bind weakly or show no binding to dPNAG [44,47]. Interestingly, in conjugate vaccine studies it was found that dPNAG was superior to PNAG in eliciting protective antibody responses in animal models [48]. Thus, antibody recognition of both PNAG and dPNAG likely contributes to optimal effector functions and protection similar to observed for the F598 mAb. Whether F598 has a unique interaction with PNAG compared to F628 and F630 is not yet known. However, it is possible that interaction with both acetylated and partially acetylated epitopes facilitates the clustering of the F598 Fc to optimally engage with complement component C1q, which is known to bind a hexamer of IgG molecules [49]. Antibody-bacterial cell complexes are also taken up into cells by Fc receptor-mediated phagocytosis and occasionally bacteria such as S. aureus can survive inside phagocytic cells evading immune clearance and antibiotic treatment. An antibiotic conjugate of the b-WTA binding antibody 4497 has been demonstrated to effectively kill intracellular S. aureus [43]. Binding of the GlcNAc portion of b-WTA is critical for the interaction with 4497 and the 2 other antibodies (4462 and 6078) isolated from patients after recovery from S. aureus infections [28

]. The high levels of b-WTA on the surface of bacteria should enable the Fc of bound mAbs to aggregate or cluster to more effectively engage C1q and Fc receptors on the surface of phagocytic cells. Another way of enhancing bacterial killing by carbohydrate-binding antibodies will be to engineer the Fc of the respective mAbs Current Opinion in Structural Biology 2020, 62:48–55

54 Carbohydrates

for enhanced effector functions, which is an active area of therapeutic antibody development [50].

2.

Banin E, Hughes D, Kuipers OP: Editorial: bacterial pathogens, antibiotics and antibiotic resistance. FEMS Microbiol Rev 2017, 41:450-452.

3.

Coleman JP, Smith CJ: Microbial resistance. In xPharm: The Comprehensive Pharmacology Reference. Edited by Enna SJ, Bylund DB. Elsevier; 2011:1-3.

4.

Roca I, Akova M, Baquero F, Carlet J, Cavaleri M, Coenen S, Cohen J, Findlay D, Gyssens I, Heure OE et al.: The global threat of antimicrobial resistance: science for intervention. New Microbes New Infect 2015, 6:22-29.

5.

Adler MJ, Dimitrov DS: Therapeutic antibodies against cancer. Hematol Oncol Clin North Am 2012, 26:447-481.

6.

Gura T: Therapeutic antibodies: magic bullets hit the target. Nature 2002, 417:584-586.

7.

Kaplon H, Reichert JM: Antibodies to watch in 2019. mAbs 2019, 11:219-238.

8.

Grilo AL, Mantalaris A: The increasingly human and profitable monoclonal antibody market. Trends Biotechnol 2019, 37:9-16.

9.

Irani V, Guy AJ, Andrew D, Beeson JG, Ramsland PA, Richards JS: Molecular properties of human IgG subclasses and their implications for designing therapeutic monoclonal antibodies against infectious diseases. Mol Immunol 2015, 67:171-182.

Concluding remarks While murine antibodies have provided an excellent foundation for antibody development, the focus is now moving towards patient derived antibodies as fully human antibodies have many advantages over those produced in other species. The known structures of bacterial anti-carbohydrate antibodies have been developed to a range of bacterial pathogens, including Shigella, Salmonella, Mycobacterium, Chlamydia, N. meningitidis, Streptococcus and Staphylococcus, as well as one broadly expressed microbial polysaccharide antigen. Binding can be categorised into four main types, being end-on insertion, bifurcated binding, groove-type binding and combination binding. While there are many differences in ligand size, positioning and number of interactions, all structures utilise 4 or more of the 6 CDRs for antibody recognition. Of these antibody structures, only two systems involve human antibodies. Given the structural diversity of bacterially derived carbohydrate recognition by antibodies further studies are needed to define the basis for targeting of any newly developed antibodies.

Conflict of interest statement Gerald B. Pier is an inventor of intellectual properties [human monoclonal antibody to PNAG and PNAG vaccines] that are licensed by Brigham and Women’s Hospital to OneBiopharma Inc., and Alopexx Vaccines LLC, entities in which GBP also holds equity. As an inventor of intellectual properties, GBP also has the right to receive a share of licensing-related income (royalties, fees) through Brigham and Women’s Hospital from OneBiopharma Inc. and Alopexx Vaccine, LLC. GBP’s interests were reviewed and are managed by the Brigham and Women’s Hospital and Partners Healthcare in accordance with their conflict of interest policies.

Acknowledgements

10. Nishat S, Andreana PR: Entirely carbohydrate-based vaccines: an emerging field for specific and selective immune responses. Vaccines 2016, 4. 11. Anish C, Schumann B, Pereira CL, Seeberger PH: Chemical biology approaches to designing defined carbohydrate vaccines. Chem Biol 2014, 21:38-50. 12. Dingjan T, Spendlove I, Durrant LG, Scott AM, Yuriev E, Ramsland PA: Structural biology of antibody recognition of carbohydrate epitopes and potential uses for targeted cancer immunotherapies. Mol Immunol 2015, 67:75-88. 13. Weintraub A: Immunology of bacterial polysaccharide antigens. Carbohydr Res 2003, 338:2539-2547. 14. Scott JR, Barnett TC: Surface proteins of gram-positive bacteria and how they get there. Annu Rev Microbiol 2006, 60:397-423. 15. Siegel SD, Liu J, Ton-That H: Biogenesis of the gram-positive bacterial cell envelope. Curr Opin Microbiol 2016, 34:31-37. 16. Bazaka K, Crawford RJ, Nazarenko EL, Ivanova EP: In Bacterial Extracellular Polysaccharides, , vol 715. Edited by Linke D, Goldman A. Dordrecht, Netherlands: Springer; 2011. 17. Limoli DH, Jones CJ, Wozniak DJ: Bacterial extracellular polysaccharides in biofilm formation and function. Microbiol Spectr 2015, 3.

CS is supported by a research training program stipend scholarship from the Australian Government, Department of Education and Training. PAR is supported by a Vice Chancellor’s senior research fellowship from RMIT University. Research studies on the F598 mAb in GBP’s laboratory have been supported by the National Institutes of Health, including the National Institute of Allergy and Infectious Diseases and the National Eye Institute.

18. Raafat D, Otto M, Reppschlager K, Iqbal J, Holtfreter S: Fighting
Staphylococcus aureus biofilms with monoclonal antibodies. Trends Microbiol 2019, 27:303-322. Discusses the advantages of using monoclonal antibodies to fight S. aureus biofilms, including preventing bacterial attachment and biofilm maturation. This review also looks at the bacterial and biofilm targets at the centre of current research as well as ongoing efforts in passive vaccination.

References and recommended reading

19. Ma¨kela¨ PH, Butler JC: History of pneumococcal immunization. In Pneumococcal Vaccines. Edited by Siber GR, Klugman KP, Ma¨kela¨ PH. ASM Press; 2008:19-29.

Papers of particular interest, published within the period of review, have been highlighted as:
of special interest

of outstanding interest 1.



McConnell MJ: Where are we with monoclonal antibodies for multidrug-resistant infections? Drug Discov Today 2019, 24:1132-1138. Summary of the current status of anti-bacterial monoclonal antibodies, particularly for antibiotic resistant microbes and multidrug-resistant Gram-negative infections, and the key considerations for further development.

Current Opinion in Structural Biology 2020, 62:48–55

20. Rappuoli R, De Gregorio E, Costantino P: On the mechanisms of
conjugate vaccines. Proc Natl Acad Sci U S A 2019, 116:14-16. This commentary focusses on mechanistic considerations for conjugate vaccines, looking at the interactions of polysaccharides with conjugate vaccines. 21. Mostowy RJ, Holt KE: Diversity-generating machines: genetics of bacterial sugar-coating. Trends Microbiol 2018, 26:10081021. 22. Richards JS, Ramsland PA: Human antibodies against DBP. Nat Microbiol 2019, 4:1428-1429. www.sciencedirect.com

Antibody recognition of bacterial glycans Soliman, Pier and Ramsland 55

23. Dutta S, Tewari A, Balaji C, Verma R, Moitra A, Yadav M, Agrawal P, Sahal D, Jarori GK: Strain-transcending neutralization of malaria parasite by antibodies against Plasmodium falciparum enolase. Malaria J 2018, 17. 24. Huang Y, Yu J, Lanzi A, Yao X, Andrews CD, Tsai L, Gajjar MR, Sun M, Seaman MS, Padte NN et al.: Engineered bispecific antibodies with exquisite HIV-1-neutralizing activity. Cell 2016, 165:1621-1631. 25. Calarese DA, Scanlan CN, Zwick MB, Deechongkit S, Mimura Y, Kunert R, Zhu P, Wormald MR, Stanfield RL, Roux KH et al.: Antibody domain exchange is an immunological solution to carbohydrate cluster recognition. Science 2003, 300:20652071. 26. Nguyen HP, Seto NO, MacKenzie CR, Brade L, Kosma P, Brade H, Evans SV: Germline antibody recognition of distinct carbohydrate epitopes. Nat Struct Biol 2003, 10:1019-1025. 27. Brooks CL, Muller-Loennies S, Borisova SN, Brade L, Kosma P, Hirama T, Mackenzie CR, Brade H, Evans SV: Antibodies raised against chlamydial lipopolysaccharide antigens reveal convergence in germline gene usage and differential epitope recognition. Biochemistry 2010, 49:570-581. 28. Fong R, Kajihara K, Chen M, Hotzel I, Mariathasan S,

Hazenbos WLW, Lupardus PJ: Structural investigation of human S. aureus-targeting antibodies that bind wall teichoic acid. mAbs 2018, 10:979-991. Analysis of the crystal structures of three human antibodies that target a cell surface epitope on MRSA S. aureus. This paper highlights the importance of human patient-based antibody discovery for development of new therapeutics. 29. Haji-Ghassemi O, Blackler RJ, Martin Young N, Evans SV: Antibody recognition of carbohydrate epitopes. Glycobiology 2015, 25:920-952. 30. Ramsland PA, Kaushik A, Marchalonis JJ, Edmundson AB: Incorporation of long CDR3s into V domains: implications for the structural evolution of the antibody-combining site. Exp Clin Immunogenet 2001, 18:176-198. 31. Zemlin M, Klinger M, Link J, Zemlin C, Bauer K, Engler JA, Schroeder HW, Kirkham PM: Expressed murine and human CDR-H3 intervals of equal length exhibit distinct repertoires that differ in their amino acid composition and predicted range of structures. J Mol Biol 2003, 334:733-749. 32. Collis AVJ, Brouwer AP, Martin ACR: Analysis of the antigen combining site: correlations between length and sequence composition of the hypervariable loops and the nature of the antigen. J Mol Biol 2003, 325:337-354. 33. Ramsland PA, Farrugia W, Yuriev E, Edmundson AB, Sandrin MS: Evidence for structurally conserved recognition of the major carbohydrate xenoantigen by natural antibodies. Cell Mol Biol (Noisy-Le-Grand, France) 2003, 49:307-317. 34. Soliman C, Yuriev E, Ramsland P: Antibody recognition of aberrant glycosylation on the surface of cancer cells. Curr Opin Struct Biol 2017, 44:1-8. 35. Vyas NK, Vyas MN, Chervenak MC, Johnson MA, Pinto BM, Bundle DR, Quiocho FA: Molecular recognition of oligosaccharide epitopes by a monoclonal Fab specific for Shigella flexneri Y lipopolysaccharide: X-ray structures and thermodynamics. Biochemistry 2002, 41:13575-13586. 36. Vulliez-Le Normand B, Saul FA, Phalipon A, Belot F, Guerreiro C, Mulard LA, Bentley GA: Structures of synthetic O-antigen fragments from serotype 2a Shigella flexneri in complex with a protective monoclonal antibody. Proc Natl Acad Sci U S A 2008, 105:9976-9981. 37. Murase T, Zheng RB, Joe M, Bai Y, Marcus SL, Lowary TL, Ng KK: Structural insights into antibody recognition of mycobacterial polysaccharides. J Mol Biol 2009, 392:381-392.

www.sciencedirect.com

38. Haji-Ghassemi O, Muller-Loennies S, Saldova R, Muniyappa M, Brade L, Rudd PM, Harvey DJ, Kosma P, Brade H, Evans SV: Groove-type recognition of chlamydiaceae-specific lipopolysaccharide antigen by a family of antibodies possessing an unusual variable heavy chain N-linked glycan. J Biol Chem 2014, 289:16644-16661. 39. Cygler M, Wu S, Zdanov A, Bundle DR, Rose DR: Recognition of a carbohydrate antigenic determinant of Salmonella by an antibody. Biochem Soc Trans 1993, 21:437-441. 40. Parker MJ, Gomery K, Richard G, MacKenzie CR, Cox AD, Richards JC, Evans SV: Structural basis for selective crossreactivity in a bactericidal antibody against inner core lipooligosaccharide from Neisseria meningitidis. Glycobiology 2014, 24:442-449. 41. Nagae M, Ikeda A, Hane M, Hanashima S, Kitajima K, Sato C, Yamaguchi Y: Crystal structure of anti-polysialic acid antibody single chain Fv fragment complexed with octasialic acid: insight into the binding preference for polysialic acid. J Biol Chem 2013, 288:33784-33796. 42. Carboni F, Adamo R, Fabbrini M, De Ricco R, Cattaneo V,

Brogioni B, Veggi D, Pinto V, Passalacqua I, Oldrini D et al.: Structure of a protective epitope of group B Streptococcus type III capsular polysaccharide. Proc Natl Acad Sci U S A 2017, 114:5017-5022. Crystal structure of an antibody to Group B Streptococcus type III capsular polysaccharide. Binding involves a combination of end-on insertion and groove binding of a long oligosaccharide chain and gives further insight into the molecular basis for antibody-carbohydrate interactions. 43. Lehar SM, Pillow T, Xu M, Staben L, Kajihara KK, Vandlen R, DePalatis L, Raab H, Hazenbos WL, Morisaki JH et al.: Novel antibody-antibiotic conjugate eliminates intracellular S. aureus. Nature 2015, 527:323-328. 44. Cywes-Bentley C, Skurnik D, Zaidi T, Roux D, Deoliveira RB, Garrett WS, Lu X, O’Malley J, Kinzel K, Zaidi T et al.: Antibody to a conserved antigenic target is protective against diverse prokaryotic and eukaryotic pathogens. Proc Natl Acad Sci U S A 2013, 110:E2209. 45. Soliman C, Walduck AK, Yuriev E, Richards JS, Cywes-Bentley C,

Pier GB, Ramsland PA: Structural basis for antibody targeting of the broadly expressed microbial polysaccharide poly-Nacetyl glucosamine. J Biol Chem 2018, 293:5079-5089. Crystal structures of a fully human antibody that targets a broadly expressed microbial polysaccharide. This paper compares the structures of a monosaccharide and oligosaccharide in complex with the Fab showing an anchored groove-type binding mechanism. 46. Blackler RJ, Mu¨ller-Loennies S, Brade L, Kosma P, Brade H, Evans SV: Antibody recognition of chlamydia LPS: structural insights of inherited immune responses. In Anticarbohydrate Antibodies. Edited by Kosma P, Mu¨ller-Loennies S. Springer; 2012. 47. Kelly-Quintos C, Cavacini LA, Posner MR, Goldmann D, Pier GB: Characterization of the opsonic and protective activity against Staphylococcus aureus of fully human monoclonal antibodies specific for the bacterial surface polysaccharide poly-Nacetylglucosamine. Infect Immun 2006, 74:2742. 48. Gening ML, Maira-Litran T, Kropec A, Skurnik D, Grout M, Tsvetkov YE, Nifantiev NE, Pier GB: Synthetic b-(1!6)-linked Nacetylated and nonacetylated oligoglucosamines used to produce conjugate vaccines for bacterial pathogens. Infect Immunity 2010, 78:764-772. 49. Diebolder CA, Beurskens FJ, de Jong RN, Koning RI, Strumane K, Lindorfer MA, Voorhorst M, Ugurlar D, Rosati S, Heck AJ et al.: Complement is activated by IgG hexamers assembled at the cell surface. Science 2014, 343:1260-1263. 50. Wang X, Mathieu M, Brezski RJ: IgG Fc engineering to modulate antibody effector functions. Protein Cell 2018, 9:63-73.

Current Opinion in Structural Biology 2020, 62:48–55