Structure of a human monoclonal antibody in complex with Outer surface protein C (OspC) of the Lyme disease spirochete, Borreliella burgdorferi

Lyme disease is a tick-borne, multisystem infection caused by the spirochete, Borreliella burgdorferi. Although antibodies have been implicated in the resolution of Lyme disease, the specific B cell epitopes targeted during human infections remain largely unknown. In this study, we characterized and defined the structural epitope of a patient-derived bactericidal monoclonal IgG (“B11”) against Outer surface protein C (OspC), a homodimeric lipoprotein necessary for B. burgdorferi tick-mediated transmission and early-stage colonization of vertebrate hosts. High-resolution epitope mapping was accomplished through hydrogen deuterium exchange-mass spectrometry (HDX-MS) and X-ray crystallography. Structural analysis of B11 Fab-OspCA complexes revealed the B11 Fabs associated in a 1:1 stoichiometry with the lateral faces of OspCA homodimers such that the antibodies are essentially positioned perpendicular to the spirochete’s outer surface. B11’s primary contacts reside within the membrane proximal regions of α-helices 1 and 6 and adjacent loops 5 and 6 in one OspCA monomer. In addition, B11 spans the OspCA dimer interface, engaging opposing α-helix 1’, α-helix 2’, and loop 2-3’ in the second OspCA monomer. The B11-OspCA structure is reminiscent of the recently solved mouse transmission blocking monoclonal IgG B5 in complex with OspCA, indicating a mode of engagement with OspC that is conserved across species. In conclusion, we provide the first detailed insight into the interaction between a functional human antibody and an immunodominant Lyme disease antigen long considered an important vaccine target.


Introduction
Lyme borreliosis or Lyme disease is the most common vector-borne infection in the United States, with an estimated 450,000 cases per year [1].The primary etiologic agent of Lyme disease is the spirochete bacterium, Borreliella burgdorferi sensu latu (herein referred to as simply B. burgdorferi).In North America, the spirochete is transmitted to humans by black legged ticks, Ixodes scapularis and Ixodes pacificus, during the course of a blood meal.The spirochete proliferates at the site of the tick bite, typically resulting in an expanding skin lesion commonly referred to as a bull's eye rash or erythema migrans [2][3][4].In the absence of antibiotic intervention, B. burgdorferi disseminates to peripheral tissues, organs, large joints, and the central nervous system, potentially resulting in severe complications including neuroborreliosis, carditis and/or Lyme arthritis [2,5].A fraction of Lyme disease patients who receive a full regimen of antibiotics will report persistent health issues (e.g., fatigue, cognitive issues, musculoskeletal pain), a syndrome referred to as post-treatment Lyme disease (PTLD) [6][7][8].
In an effort to define the immunologic factors that drive resolution of Lyme disease, Blum and colleagues characterized B cell responses in patients across the course of infection [9].
Blood plasmablasts from Lyme disease patients and a healthy cohort were subjected to bulk BCR and single cell paired VH and VL sequencing [9].The analysis revealed that robust plasmablast responses correlated with a more rapid resolution of disease symptoms.Moreover, expression of a panel of recombinant human monoclonal antibodies (MAbs) using paired VH and VL sequences from representative clones were screened for reactivity with B. burgdorferi antigens.The "hits" included a number of immunodominant B. burgdorferi surface lipoproteins previously known to be reactive in convalescent human Lyme disease sera, including VlsE, DbpA, DbpB and OspC [9][10][11].The recombinant MAbs were further screened for the ability to inhibit growth of B. burgdorferi strain B31 in culture.Of the six MAbs with borreliastatic activity (and therefore implicated in disease resolution), two (B2 and B11) were directed against OspC.
Outer surface protein C (OspC) is a member of the small variable surface protein (Vsp) family of immunodominant antigens unique to Borrelia and Borreliella including the relapsing fever spirochetes (B.turicatae) [12].In B. burgdorferi, OspC is expressed during tick transmission and in the early stages of mammalian infection [13][14][15].During this period, OspC plays a multifaceted role in pathogenesis, including facilitating spirochete egress from the tick during the course of a blood meal, enabling survival in the early stages of mammalian skin infection, plasminogen interactions, and modulating transmigration across vascular walls [14][15][16][17][18][19][20].The importance of OspC in B. burgdorferi pathogenesis is underscored by the fact that active vaccination with recombinant OspC or passive transfer of OspC antibodies completely prevent tick-mediated B. burgdorferi infections in mouse models [21][22][23][24][25].However, while only a single copy of ospC is encoded on any B. burgdorferi genome, there are >26 ospC types and subtypes within B. burgdorferi isolates in North America, with multiple types often found in the same geographical proximity [26][27][28].The polymorphic nature of OspC is proposed to constitute a means of B. burgdorferi immune evasion, as antibodies to one OspC type have limited crossreactivity and cross-protection with other OspC types [28][29][30][31].
Structurally, OspC is ~21 kDa helical bundle that forms homodimers on the spirochete outer surface [32][33][34][35][36].Despite a high degree of amino acid variability across the different OspC types, the tertiary and quaternary structures of OspC are virtually identical.As OspC's N-termini are anchored in the outer membrane via a lipid moiety, the molecule is generally depicted with the stem tethered to the spirochete surface and the dome, with more variable residues, projecting outward [35].As such it is reasonable to expect that type-specific antibodies target the outward face of OspC, occluding the putative ligand binding cavities [35].However, the X-ray crystal structure of OspC type A (OspCA) complexed with the mouse MAb, B5, reveals a more complicated picture.B5 is the most well-characterized OspC monoclonal antibody originally isolated from B. burgdorferi-infected mice [23,37].In passive immunization studies, B5 blocks tick-mediated transmission of B. burgdorferi, presumably by entrapping the spirochetes in the tick midgut [23].The X-ray crystal structure revealed that B5 Fabs associates at a nearly perpendicular angle with the lateral sides of homodimeric OspCA [38].The side-on orientation of B5 with OspCA raises questions about the mechanisms by which the antibody interferes with spirochetal transmission as the putative ligand binding pockets on the outer most face of OspC are unobstructed.A limited insight into the nature of OspC-antibody interactions in mice (and none in humans) represents a gap in our understanding of a critical immune interaction with implications for both Lyme disease resolution and vaccine design.In this study, we report the structure of the human MAb B11 in complex with OspC and its impact on B. burgdorferi viability in vitro.

MAb B11 promotes B. burgdorferi agglutination and alterations in outer membrane permeability.
We recently reported that antibody-mediated agglutination of live B. burgdorferi spirochetes results in flow cytometric events with increased size (FSC) and granularity (SSC) [41].
Furthermore, the addition of propidium iodide (PI) to the spirochete preparations just prior to flow cytometry serves as indicator of antibody-induced changes to OM permeability [41].We have postulated that antibody-mediated agglutination explains, at least in part, how OspA and possibly OspC antibodies compromise B. burgdorferi within the tick midgut and limit their transmission to vertebrate hosts [23,30,38].We therefore examined what impact that B11 had on B. burgdorferi agglutination.Following B11 treatment, we observed significant B. burgdorferi B313 agglutination (20-25% of the population) with a fraction of aggregated cells also PI + .This level of agglutination and PI + staining is similar to what we have observed previously for MAb B5 (Figure 1B).Neither MAb B11 nor B5 influenced agglutination or PI positivity of the B. burgdorferi DospC mutant or B. burgdorferi strains ZS7 or 297 (Figure 1B; data not shown).Thus, B11 recognizes OspCA on the spirochete surface and has the propensity to promote both spirochete agglutination and alterations in outer membrane permeability in the absence of human complement.

Complement-dependent and independent impact of MAb B11 on B. burgdorferi motility
Blum and colleagues reported that MAb B11 (20 µg/mL) had growth inhibitory (bacteriostatic) activity on B. burgdorferi in standard culture conditions (BSK II) [9].To confirm this observation, we examined B. burgdorferi B31 motility (as a proxy for cell viability) in the absence and presence of 20% human complement across a range of B11 doses (1-30 μg/mL).To circumvent issues associated with intrinsically low OspC expression by B. burgdorferi B31 in culture, we utilized a strain with an IPTG inducible rpoS allele, thereby activating native ospC expression in trans (see Methods).In the absence of complement, B11 had a modest effect on bacterial motility that but did not reach statistical significance as compared to an the isotype control, PB10 (Figure 2).The addition of human complement resulted in significant reduction in bacterial motility at B11 concentrations at 3 µg/ml and above.The transmission blocking MAb B5 was significantly more potent than B11 at arresting B. burgdorferi motility in absence of human complement and was similar to B11 in the presence of complement (Figure 2).Thus, in our hands, B11's effects on spirochete motility arrest are complement-dependent.

Localization of MAb B11's epitope on OspCA by HX-MS.
We employed HX-MS as a means of identifying B11's epitope on OspCA.A series of preliminary quench and digestion experiments revealed that proteolysis of recombinant dimeric OspCA with Nepenthesin II (without the addition of urea) generated the largest set of observable peptides.Filtering out weak and overlapping signals resulted in 74 unique peptides with a sequence coverage of 98.8% and redundancy of 5.3.The addition of MAb B11 IgG resulted in strong protection at residues 49-57 and 186-196, corresponding to the N-and C-termini of OspCA, respectively (Figure 3; Supplemental Excel file).Moderate protection was also observed elsewhere (e.g., peptide 157-162).The B11 IgG HX-MS protection profile is reminiscent of the profile we reported for B5 IgG using slightly different HX-MS conditions [38].To enable a direct comparison between B11 and B5, we subjected OspCA to HX-MS with B5 IgG under the same conditions as B11.The results revealed that HX-MS profiles were indeed distinct with B5 causing notably stronger protection at residues 72-80, 157-176, and 186-196, further from the N/C termini (Figure 3).These results suggest that B5 and B11 recognize overlapping but distinct epitopes primarily focused on OspCA's a-helix 5 and 6.To test this experimentally, we performed competitive binding assays between B5 and B11 using the BLI platform in which OspCA was first saturated with B5 then probed with B11.The results confirmed that B11's ability to associate with OspCA is abrogated by B5 occupancy (Figure S1C).

Structural analysis of the Fab B11-OspCA complex.
To resolve B11's epitope on OspCA in greater detail, we solved the X-ray crystal structure of B11 Fabs in complex with OspCA at 3.1 Å resolution in the P1 space group.The structure revealed two B11 Fabs bound to a single OspCA homodimer (1:1 Fab:OspCA stoichiometry) in a side-on fashion reminiscent (at first glance) of the B5 Fab-OspCA complex (Figure 4) .The B11 Fab fragments (Fab, Fab') made identical contacts on opposite sides of the OspCA homodimer (OspCA-OspCA'), as described in detail below.Each B11 Fab assumed the canonical antibody structure with two heavy chain immunoglobulin domains (VH, CH1) and two light immunoglobulin domains (VL, CL) each containing 7-10 β-strands arranged in two β-sheets that fold into a two-layer β-sandwich.The six CDRs (LCDR1-3, HCDR1-3) were situated on one face of the molecule.The four B11 FAb-OspCA complexes within the asymmetric unit were structurally similar, with Root-Mean-Square Deviations (RMSD) in the range of 0.8-1.8Å upon Ca-superpositioning. OspCA homodimers in the absence and presence of B11 Fabs were also structurally nearly identical [RMSD of 0.6 Å], demonstrating that the B11 Fabs do not induce any significant conformational changes OspCA.
Each B11 Fab contacted a-helix 1, a-helix 6, loop 5 (located between a-helices 5 and 6) and loop 6 (after a-helix 6) of their respective OspCA monomers.In addition, the Fabs each bridged the OspCA dimer interface with residues from HCDR1, HCRD3, and LCDR2 engageing with a-helix 1', a-helix 2', and loop 2' on the opposing OspCA molecule (Figure 5A, B), thereby demonstrating that B11's epitope is quaternary in nature.The B11 Fab and OspCA interface buried a total surface area ranging from 2,406 Å 2 to 2,772 Å 2 (Table 2) Shape complementarity (SC) scores ranged between 0.44-0.59within the eight Fab-OspCA interfaces in the asymmetric unit, while H-bonds ranged from 3 to 13 and salt bridges from 2 to 4.

Structural comparisons between B11 and B5 Fab-OspCA complexes.
As noted in the Introduction, we recently reported the structure of recombinant homodimeric OspCA complexed with the Fab fragments of the transmission blocking murine MAb, B5 [PDB ID 7UIJ] 26 .The B11 Fab-OspCA complex is reminiscent of that structure in that B5 and B11 Fabs engage with the lateral face of OspCA in a side-on orientation (Figure 6A).However, the B5 and B11 Fabs are offset relative to each other, with B5 membrane-distal and B11 membraneproximal (Figure 6A).Moreover, while B5 Fabs are virtually perpendicular to OspCA (Figure 6A), B11 Fabs associate with OspCA at a slight angle, thereby positioning B11 Fabs closer to the spirochete outer surface (Figure 6A).Nevertheless, there is sufficient distance between B11's epitope on with OspCA and the spirochete membrane to accommodate an antibody as shown when we super positioned the HIV-1 IgG b12 (PDB ID: 1HZH) onto the B11 Fab (Figure 6B).Understanding the molecular basis of OspC type-immunospecificity is a longstanding challenge with relevance to vaccine design [42,43].We previously speculated that B5 IgG reactivity is restricted to just three OspC types (A, C3 and I3), based the variability of key paratope-epitope contacts [38].By the same token, multiple sequence alignment (MSA) of 23 prominent B. burgdorferi OspC types suggest that B11 is similarly restricted.The primary sequences of four of the 23 OspC Types (C3, I3, J, and M) examined contain the residues ostensibly associated with B11 recognition on Type A (Figure S5B).OspC types C3, I3, J, and M share 72-80% amino acid identity with OspCA.More importantly, sequences within key B11binding residues within each of these four OspC types are identical OspCA.The main-chain structure around residue 74 is conserved (whereas a deletion exists in other types, as described above) and OspC types C3, I3, J, and M contain a glycine or nonbulky amino acid at position 114 alleviating potential steric clash with B11 that appears in other OspC types (as described above).Each of the four selected potential B11-binders possess a phenylalanine at position 177 which buries much of its bulky side chain in the interface with B11.The loss of this bulky side chain would diminish the interaction with B11 by ~22 Å 2 based on buried surface area calculations with Phe-177 Å 2 as a serine residue as found in many of the 23 OspC genotypes.
Finally, these four putative B11-binding OspC types also bear a lysine or arginine at residue 161, which putatively H-bonds with B11.From this information, we speculate that B11 likely recognizes four additional OspC types: C3, I3, J, and M (Figure S5B).This contrasts with B5 IgG, which we proposed is restricted to OspC types A, C3 and I3 [38].Cross reactivity with I3 is not surprising given that is a naturally occurring chimera between OspCF and OspCA with the Cterminus (residues 128-199) derived from type A [44].

Discussion
OspC plays a pivotal role in B. burgdorferi tick-to-mammalian transmission and in early stages of dissemination.Although OspC may not be the sole determinant of spirochete dissemination, its importance is underscored by the fact that certain ospC genotypes are associated with more invasive B. burgdorferi clinical isolates [45][46][47][48].From the host vantage point, B. burgdorferi infection is accompanied by a robust OspC-specific antibody response, which subjects the spirochetes to an enormous immune pressure that may contribute to bacterial clearance [42,49].Despite the centrality of OspC in both infection and immunity, a highresolution molecular and structural picture OspC-antibody interactions is lacking in humans.
In this study, we solved the first structure of OspC complexed with Fabs from a human monoclonal antibody.The antibody, B11, was originally isolated by Blum and colleagues as a recombinant IgG1 with paired VH and VL sequences from single-cell sorted, blood-derived plasmablasts from a Lyme disease patient [9].Recombinant B11 was shown to be bacteriostatic (without addition of human complement) and capable of promoting B. burgdorferi opsonophagocytosis in culture [9,50].We have extended those observations by confirming that B11 recognizes native OspC on the surface of B. burgdorferi strain B31 (but not strains expressing OspC type B or K) and has both complement-independent and -dependent activities against the spirochetes in culture.Based on these observations we predict that "B11-like" antibodies contribute to the reduction and/or clearance of an active B. burgdorferi infection in humans.
The crystal structure of B11 Fabs complexed with dimeric OspCA is remarkable in that it demonstrates antibody association with the lateral faces ("stem") of OspCA, rather than the more surface exposed "dome".Considering that OspC is tethered to the spirochete outer membrane via its lipidated N-terminus, the molecule is generally depicted as projecting perpendicularly from the spirochete surface [33][34][35].In such an orientation, the predominantly variable residues face outwards, while conserved amino acids are membrane proximal [34,35].Thus, it is natural to assume that type-specific antibodies engage the dome region of OspC via their Fab elements in a top-down orientation [35].That assumption was called into question by the structure of B5 Fab-OspCA [38], and further undermined by the structure of B11 Fab-OspCA.
Perhaps it is worth turning things on their side when rethinking the nature of OspCantibody interactions.Indeed, Lawson and colleagues pondered this exact possibility when scrutinizing the structures of OspC and related Vsps from B. turicatae [35].They cite evidence to suggest that the N-and C-termini of OspC and Vsp1 dimers "…either adopt a mostly 'random coil' conformation or are disordered."If correct, they argue, then OspC is essentially bound to the spirochete by a flexible tail, thereby enabling the bulk of OspC to sit sideways or even "upside down" on the cell membrane.Moreover, they note that such an unconventional orientation would explain how IgM antibodies are able access the very C-terminal decapeptide of OspC, a linear epitope known as C10 [51].In such a configuration, the Fc regions of antibodies like B11 IgG would protrude from the bacterial surface and be prime targets for components of the complement system and Fc receptors on phagocytes [50].
The structural elucidation of the of B11 Fab-OspC complex has implications when considering the design of OspC-based vaccine antigens.First, the B11 Fabs bridge the OspC-OspC' dimer interface.This observation indicates that B11 underwent affinity maturation in the context of dimeric (not monomeric) OspC.From the standpoint of vaccine design, preserving the dimeric interface of OspC antigens is paramount for eliciting potent B11-and B5-like antibodies, an observation that has been alluded to by others [52].Second, a detailed view of B11 and B5's epitopes reveal the molecular basis of OspC type-specificity, a factor that has limited OspC's utility as a Lyme disease vaccine antigen [28,42,53].However, considering the current exponential advances in protein modeling and design, the availability of structural B cell epitopes will prove invaluable.Finally, tt is worth noting that in separate efforts we have identified human monoclonal and camelid-derived single-domain antibodies that have pan-OspC reactivity (L.Cavacini, D. Vance, M. Rudolph, N. Mantis, manuscripts in preparation).Highresolution epitope mapping studies of those pan-reactive OspC antibodies are ongoing, alongside passive protection studies in mouse models of B. burgdoferi by tick-mediated infection.

Cloning and expression of B11 IgG
The B11 VH and VL coding regions were extracted from NCBI's BioProject database under accession PRJNA470931 (GSE114310) [9].The B11 VH and VL coding regions were synthesized as gBlockÔ dsDNA fragments (GenScript, Piscataway, NJ) and cloned in-frame into pcDNA3.1-basedhuman IgG1 Fc and kappa light chain expression plasmids, respectively [54,55].The two plasmids were co-transfected into Expi293 cells using ExpiFectamine293™ transfection reagents (Thermo Fisher Scientific, Waltham, MA), following manufacturer's instructions.Five to six days later, supernatants containing the secreted antibodies were harvested and clarified before being subjected to protein A chromatography.Affinity purified IgG1 B11 was then subjected to buffer exchange into PBS and stored at 4 °C.B11 Fabs were generated as described [38].Purification and specificity of MAbs B5 [38]and hPB10 [56]have been previously described.
A strain of B. burgdorferi B31 was engineered to deliver consistent ectopic expression of ospC under in vitro culturing conditions (BSK-II at 37℃).An IPTG-inducible rpoS expression plasmid was created through modification of the pBSV2G-derived mCherry plasmid, pGW163 [58].Briefly, the rpoS open reading frame (ORF) was amplified from B. burgdorferi B31 [NCBI:txid224326] genomic DNA using Q5 DNA polymerase (NEB, Beverly, MA) and the b31_rpoS_gibson_F (5'-agaattcattaaagaggagaaattacccatgaacatatttagtaatgaggatttaaacat -3') and b31_rpoS_gibson_R (5'-gtaaaacgacggccagtgccaagcttaatttatttcttcttttaattttttaagaactc-3') tailed primer set.The mCherry ORF was excised from pGW163 using NcoI and HindIII (NEB) and subsequently replaced with the tailed rpoS DNA fragment using the HiFi DNA assembly kit (NEB).Following the assembly reaction, the DNA was transformed into NEB Ò 5-alpha F' I q chemically competent cells (NEB).Recombinants were selected on LB agar supplemented with 10 µg/ml of gentamicin.Transformants that arose following overnight incubation at 30℃ were screened via PCR using the b31_rpoS_gibson_F and b31_rpoS_gibson_R primer set.The IPTGinducible rpoS expression plasmid, pGW181, was isolated from a single PCR-positive colony.
For routine culture, frozen aliquots of B. burgdorferi were thawed and used to inoculate modified BSK-II lacking gelatin, supplemented with 6% rabbit serum (Pel-Freeze Biologicals, Rogers, AR) [60].Base BSK-II medium was prepared by the Wadsworth Center's Tissue and Media Core Facility and filter sterilized (0.2 μm) prior to use.Cultures were maintained at 37°C with 5% CO2 and passaged by dilution (1:10,000) into fresh BSK-II medium.B. burgdorferi cultures were routinely inspected for culture viability and motility during in vitro culture maintenance prior to the initiation of any experiments.
To induce RpoS expression, GGW941 culture medium was spiked with 250 µM IPTG at 24 h before use.For flow cytometry and microscopy studies (as described below), B. burgdorferi cells were collected in mid-logarithmic phase (~1-3 × 10 7 spirochetes per mL).As controls, spirochetes were heat killed by incubation at 56°C for at least 30 min, as described [61].

B. burgdorferi motility determinations by dark field microscopy
Mid-log-phase cultures of GGW941 treated 24 h prior with IPTG to induce rpoS expression, were adjusted to 5 x 10 6 bacteria per 50 μL in PBS then treated with indicated OspC MAbs (1, 2, 3, 10, 20, 30 μg/mL) in the presence or absence of 20% human complement (Sigma-Aldrich) for 16 or 24 h.Cultures were examined in a double-blind fashion by dark-field microscopy for motile spirochetes.1:5 dilutions of the reactions were prepared by aspirating from the center of the reaction and diluted into fresh BSK-II medium.Dark-field microscopy was performed utilizing a Trinocular DF microscope (AmScope) equipped with a camera with reduction lens (AmScope SKU: MU1603) using a 40× dry darkfield condenser (AmScope; DK-DRY200).
Spirochetes were considered dead when complete loss of motility and refractivity was observed.
Spirochetes were enumerated in 4 visual fields, and the percent viability was calculated as the ratio of live spirochetes (mean of 4 fields) in treated samples to spirochetes in the untreated control samples (mean of 4 fields).Polyclonal serum from B. burgdorferi-infected mice and MAb B5 were used as positive controls; naive serum and the PB10 isotype were used as negative controls.This experimental set up was conducted over the course of three independent sessions and data is plotted as the means for the three days of counting.Statistical analysis was determined using 2-way ANOVA with Dunnett's multiple comparisons test in which experimental means (B11, B5) were compared to control mean (PB10) for each antibody concentration.

Flow Cytometry
To examine the ability of B11 and B5 IgG1 to bind the spirochete surface and promote bacterial agglutination, and membrane permeability, B. burgdorferi strains expressing OspC types A (B313), B (ZS7), and K (297), were cultured, treated with 10 µg/ml IgG1, and analyzed by flow cytometry as described [38].Briefly, an Alexa Fluor 647-labeled goat anti-human IgG [H+L] secondary antibody (Invitrogen) was used to detect bound IgG1 to the bacterial surface, and propidium iodide staining was used to indicate bacterial membrane permeability.The ospC deletion strain B31-A3ospCK1was used as a negative strain control, and the ricin toxin MAb, PB10, was used as an IgG1 isotype control.Bacteria were analyzed on a BD FACSCalibur (BD Biosciences).Voltage was set with untreated bacteria as a reference point (gMFI~ 5.0), and bacteria were gated on forward scatter (FSC) and side scatter (SSC) to exclude debris and assess aggregate size and granularity.Agglutination was calculated as the sum of events in the upperleft, upper-right, and lower-right quadrants relative to the total event counted (20,000).

Epitope mapping of B11 by HX-MS
Stock concentrations of OspCA (8.5 µM) in PBS alone or in a complex with a 2-fold molar excess of antibody B11 were diluted into 90 µL of deuterated PBS buffer (20 mM phosphate, 150 mM NaCl, 0.02% sodium azide, 1 mM EDTA pH* 7.54, 85%D final) containing 0.2 nM bradykinin and incubated 3 sec on ice, or either 3 seconds, 1 minute, 30 minutes, or 20 hours at 21°C.Each starting stock also included a mixture of imidazolium compounds to serve as exchange reference standards [62].At the desired time point the sample was rapidly mixed with an equal volume of ice cold 0.2% formic acid and 0.1% trifluoroacetic acid (TFA) for a final pH of 2.5.Samples were then immediately frozen on ethanol/dry ice and stored at -80°C until LC-MS analysis.Undeuterated samples were prepared the same way but with undeuterated buffer for each step.
Samples were thawed at 5°C for 8 minutes and injected using a custom LEAP robot integrated with an LC-MS system [63].The protein was first passed over a Nepenthesin II column (2.1 x 30 mm; AffiPro) at 400 µL/min for inline digestion with the protease column held at 20°C.Peptides were then trapped on a Waters XSelect CSH C18 trap cartridge column (2.1 x 5 mm 2.5 µm) and resolved over a CSH C18 column (1 x 50 mm 1.7 µm 130Å) using linear gradient of 5 to 35% B (A: 0.1% FA, 0.025% TFA, 5% ACN; B: ACN with 0.1% FA) over 10 minutes and analyzed on a Thermo Orbitrap Ascend mass spectrometer at a resolution setting of 120,000.A series of washes over the trap and pepsin columns was used between injections to minimize carry-over as described [63].Data dependent MS/MS acquisition was performed on an undeuterated sample using rapid CID and HCD scans and processed in Byonic (Protein Metrics) with a score cutoff of 150 to identify peptides.Deuterium incorporation was analyzed using HDExaminer v3 (Sierra Analytics).

Crystallization and data collection.
Fab B11 was complexed with dimeric OspCA in a 1:1 stoichiometry, then concentrated to 10 mg/ml final for all crystallization trials.Crystals were grown by sitting drop vapor diffusion using a protein to reservoir volume ratio of 1:1 with total drop volumes of 0.2 μl.Crystals of the B11 Fab-OspCA complex were produced at 22°C using a crystallization solution containing 100 mM sodium HEPES [pH 6.5], 18.6% PEG 6K, and 6.6 mM prasseodymium acetate.All crystals were flash frozen in liquid nitrogen after a short soak in the appropriate crystallization buffers supplemented with 10% ethylene glycol.Data were collected at the 24-ID-E beamline at the Advanced Photon Source, Argonne National Labs.All data was indexed, merged, and scaled using HKL2000 [64] then converted to structure factor amplitudes using CCP4 [65].
discussions.As always, we are grateful to Elizabeth Cavosie (Wadsworth Center) and Donna DeLuca (Health Research, Inc) for administrative assistance and program overview.(B) Corresponding FSC/SSC dot plots.Events that are increased in size (FSC) and granularity (SSC) represent multi-bacteria aggregates as the result of antibody-mediated agglutination of live spirochetes.The percentage of events that are agglutinated is indicated (black) and was calculated from the sum of events with increased FSC and SSC, (UL+UR+LR quadrants), relative to the total events counted (20,000).The percent of events positive for propidium iodide staining, indicating membrane permeability, is labeled, and highlighted in red.a , primary and secondary (in parentheses) interfaces; b ,hydrogen bonds; c , salt bridges; d , shape complementarity score BioProject database indicate that one other B11 clone exists, alongside three closely related clones all derived from the same patient (D.Vance, unpublished observations).The B11 VH and VL coding regions were cloned in-frame into pcDNA3.1-basedhuman IgG1 Fc and kappa light chain expression plasmids, respectively, and then co-transfected into Expi293 cells.Purified B11 IgG1 was assessed for the ability to recognize native OspC.Live B.

Figure 1 .
Figure 1.Reactivity of B11 and B5 with OspCA on live spirochetes.Flow cytometric analysis of B11 and B5 MAb reactivity with B. burgdorferi strains expressing OspC types A (B313), B (ZS7), or K (297).A B. burgdorferi B31 OspC mutant (B31-A3ΔospC ) was included as a control.(A) Representative fluorescence histograms where an Alexa Fluor 647-labeled anti-human IgG secondary antibody was used to detect bacterial surface bound IgG.Panels show B11 (blue) or B5 (gray) compared to the isotype control (PB10, black).Events positive for 647 fluorescence fall to the right of the isotype control.The percent and geometric mean fluorescence intensity (gMFI) of positive events for bacteria incubated with B11 (blue) and B5 (gray) are indicated on each panel.

Figure 2 .
Figure 2. Complement-dependent and -independent effects of B5 and B11.Mid-log-phase cultures of B. burgdorferi strain GGW941 treated for 24 h with IPTG induce rpoS expression, were adjusted to 5 x 10 6 bacteria per 50 μL in BSK II without (left columns) and with (right columns) 20% human complement.Cultures were then treated with (A) B11, (B) B5 or (C) PB10 at indicated concentrations (x-axis).Sixteen hours later, the cultures were examined in a double-blind fashion by dark-field microscopy for motile spirochetes.Spirochetes were enumerated in 4 visual fields, and the percent viability was calculated as the ratio of live spirochetes (mean of 4 fields) in treated samples to spirochetes in the untreated control samples (mean of 4 fields).The experimental set up was conducted over the course of three independent sessions and data is plotted as the means for the three days of counting.Statistical analysis to determined significance (*, p<0.05) was done with 2-way ANOVA with Dunnett's multiple comparisons test for each B11 and B5 versus PB10 (isotype control) at each dose.

Figure 3 .
Figure 3. Localization of the B11 epitope on OspCA using HDX-MS.HDX changes upon the addition of (A) B11 IgG and B5 IgG are plotted on the structure of OspCA [PDB ID 1GGQ].Regions with reduced deuterium exchange (more protected) are colored blue, while more deuterium exchange are colored red.(B-G) Deuterium uptake plots for unbound OspCA (black liunes), B5-OspCA (purple lines) and B11-OspCA (green lines) are shown for selected regions, with the panel letter corresponding to the labels in panel A. Error bars represent standard deviations from triplicate measurements.Comprehensive exchange data is provided as an Excel file as Supplemental material.

Figure 4 .
Figure 4. Structure of B11 Fab-OspCA.(A) Side-on and (B) top-down ribbon diagrams of OspCA homodimer (OspCA, OspCA') in complex with B11 Fabs (B11, B11').OspCA is colored in cyan and OspCA' in green.The B11 Fab VH and CH1 elements are colored in salmon and VL and CL in light gray.The B11' Fab VH and CH1 elements are colored in magenta and VL and CL in yellow.The VH and VL domains (VH,VL) along with the N-and C-termini of OspCA and OspCA' are labelled accordingly.

Figure 5 .
Figure 5. Detailed interactions between B11 and OspCA revealed from the co-crystal structure.(A) Ribbon structure (top-down view) of OspCA homodimer (OspCA, cyan; OspCA', green) in complex with a single B11 Fab (VH and CH1 elements, magenta; VL and CL, yellow).The OspCA residues that engage with B11 are colored blue.Key secondary structures (a-helices 1, 2, 3, 5, and 6) and all six CDRs are labeled; (B) Ribbon (left) and surface (right) depiction of an OspCA homodimer (OspCA, cyan; OspCA', green) with B11-interacting residues shaded in dark blue.OspCA N and C-termini are labelled N and C, respectively.Representations of key Hbonds (red dashes) and salt bridges (yellow dashes) between (C) OspCA (green) and B11' VH domain (magenta) and (D) OspCA (green) and B11' VL domain (yellow).Side chains are drawn as sticks and color coordinated to the main chain color, with nitrogen atoms shaded blue and oxygen atoms shaded red.CDR elements are labelled per convention: CDR-L1, -L3, -L3; CDR-H1, -H2, -H3.OspCA N and C-termini are labelled N and C, respectively.All drawings are depicted as ribbon diagrams except the molecular surface representation in panel B.

Figure 6 .
Figure 6.Structural comparison of B11-OspCA with and B5-OspCA.(A) Superposition of B11-OspCA and B5-OspCA drawn as Ca-traces depicting the similar side-on approach of each Fab to OspCA while highlighting the greater distance of B5 from the N and C-termini of OspCA relative to the B11 position.OspCA is colored cyan and green while the B11 Fab VH and CH1 elements are colored in salmon red and VL and CL in light gray.The B11' Fab VH and CH1 elements are colored in magenta and VL and CL in yellow.B5 Fabs are colored dark gray.(B) Superposition of the anti-HIV IgG1 mAb (PDB ID 1HZH) drawn as a gray Ca-trace onto B11 also depicted as a Ca-trace revealed no significant steric clash between the OspCA-bound IgG and the spirochete membrane.The OspCA-B11 structure is colored identically to panel A in this figure.

Table 1 .a,
Functional activity associated with MAbs B11 and B5 B. burgdorferi surface binding; b , Agglutination; c , propidium iodide positive events.Flow cytometric analysis of OspC IgG mAbs B11 and B5 and their reactivity with live B. burgdorferi strains expressing different OspC types.An anti-human IgG Alexa 647-labeled secondary antibody was used to detect mAb binding and is displayed as percentage of events positive for 647 fluorescence, and geometric mean fluorescence intensity (gMFI).The percentage of events with increased size and granularity (FSC/SSC) demonstrating agglutination, and percentage of events with propidium iodide uptake are shown.Displayed are mean values, with background subtraction, +/-standard deviation, n=4.