Fig 1.
Surface binding profiles of anti-OspAST1 Bin 1 mAbs to B. burgdorferi OspA ST1-7.
(A) Representative flow cytometric histograms of live B. burgdorferi HB19-R1 reporter strains expressing OspA ST 1-7, each probed with 10 µg/ml of mAbs indicated in bold text above each column, followed by Alexa 647-labeled goat anti-human IgG secondary antibody. HB19-R1 harboring the empty vector (EV) was used as a negative control. The horizontal bracket represents the region of events positive for fluorescence labeling (647+) on the subsequent plots. The percent (%) of events positive for Alexa 647 fluorescence labeling and the geometric mean fluorescence intensity (gMFI) are shown in the top left corner of each box. Compared to the empty vector (EV) controls, the gMFI values were significantly greater for binding of LA-2 to ST1 (p < 0.0001); 857-2 to ST1-7 (p < 0.0001); 221-5 to ST1-7 (p < 0.0001); 221-20 to ST1-7 (p < 0.0001); 221-11 to ST1 (p < 0.001), ST2, 4-7 (p < 0.0001), and ST3 (p < 0.05); 227-2 to ST1-2, 4-7 (p < 0.0001) and ST3 (p < 0.05); 227-1 to ST1-2, 4-6 (p ≤ 0.0001). N = 3. Statistical comparisons were performed using two-way ANOVA with Dunnet’s multiple-comparison test. Panels and plots were assembled using FlowJo. (B) Heat map summarizing the average gMFIs (with scale shown on right) of B. burgdorferi HB19-R1 reporter strains probed with each of the different mAbs depicted in Panel A. The results per box are the average of three biological replicates.
Table 1.
Characteristics of OspA mAbs used in this study.
Fig 2.
OspA Bin1 mAbs promote complement-dependent killing of recombinant B. burgdorferi strains expressing OspA ST1-7.
(A) Borreliacidal mAb titration curves and (B) heat map comparing the susceptibilities of an infectious B. burgdorferi B31-5A4 mScarlet-I viability reporter strain and an HB19-R1 reporter strain expressing the same OspAST1 (B31) allele to complement-dependent killing mediated by OspAST1 mAbs or an isotype IgG1 (PB10) control. Experiments were performed under conditions optimized for each strain as described within the Materials and Methods. Ec50 values represent the mean minimum concentration (nM) of mAb resulting in >50% reduction in MFI relative to controls, as determined from 3-5 independent experiments per strain. Statistical differences in mAb susceptibility were assessed via two-way ANOVA followed by Tukey’s multiple comparison test. No significant differences were observed. (C) Heat map depicting the mean Ec50 (nM) of Bin 1 mAbs against recombinant B. burgdorferi strains expressing OspA ST 1-7. Complement-dependent borreliacidal assays were performed with anti-OspAST1 Bin 1 mAbs and a panel of B. burgdorferi HB19-R1 (ospA negative) strains harboring an IPTG-inducible mScarlet-I viability reporter plasmid and expressing ospA ST1-7 under conditions described in the Materials and Methods. Controls included an HB19-R1 strain carrying the IPTG-inducible viability reporter plasmid alone (empty vector, ospA negative) and a mAb with borreliacidal activity restricted to OspAST1 (LA-2). EC50 values shown represent the mean lowest concentration of antibody (nM) resulting in >50% reduction in MFI relative to controls. Reporter strains that exhibited resistance to complement-mediated killing by mAbs at 10 nM were later retested at a single higher dose (66.6 nM). The data shown encompasses 3-5 independent experiments per strain with data normalized as described within the Materials and Methods section. Corresponding antibody titration curves and statistical comparisons are shown in S3 Fig.
Table 2.
Summary of OspA-Fab interactions.
Fig 3.
Crystral structures of Fab-OspA complexes reveal conserved epitopes on OspAST1.
(top row) Ribbon diagrams of OspAST1 (green) in complex with Fabs from (A) 857-2 (class I), (B) 221-5 (class I), (C) 221-11 (class II), and (D) 227-1 (class III). Fab heavy chains (VH and CH1) are colored cyan; light chains (VL and CL) are colored magenta. The OspA N- and C-termini are labelled accordingly, along with selected strand numbers. (middle row) The four OspA (green) ribbon representations aligned in the same orientation with respective Fab contract points colored in blue. (bottom row) The ribbon and surface representations of OspA highlight the Fab-interacting residues, shown in blue. The N and C-termini of OspA are labelled N and C, respectively. PDB ID for each structure are provided in Table 2.
Fig 4.
Key interactions between each Fab and OspA.
Close-up of notable salt bridges between OspA Lys-107 (green) and Fab residues are shown as yellow dashed lines. Fab heavy chains (VH/CH1) are colored cyan, and light chains (VL/CL) are colored magenta. (A) Fab 857-2 engages Lys-107 through Glu-55 and Asp-110. (B) Fab 221-5 interacts with Lys-107 via Asp-55 and Asp-109. (C) Fab 221-11 contacts Lys-107 through Asp-109. (D) Fab 227-1 engages Lys-107 through Glu-55. All side chains are drawn as sticks and color coordinated to the main chain color with nitrogen atoms blue and oxygen atoms red.
Fig 5.
Influence of OspA primary sequence deviations on antibody function.
(A) Structural basis for Fabs 221-11 and 227-1 cross-reactivity loss in OspA serotypes 3 and 7. Distinct hydrogen bond networks between Fab Arg-59 (cyan) and OspA residues (green) contribute to serotype specificity. (A) In Fab 221-11, Arg-59 forms a salt bridge with OspA Glu-131. (B) The same Glu-131 interaction is observed in Fab 227-1. (C) In 221-11, Arg-59 hydrogen bonds with OspA Ser-152. (D) Similarly, Fab 227-1 also engages Ser-152 through Arg-59. The primary sequence difference distinguishing serotypes (STs) 1, 2, 4, 5, and 6 from STs 3 and 7 involves a Glu-Lys substitution at position 131 and a Ser-Asn at position 152. These two primary sequence differences conceivably disrupt the Arg-59–Glu-131 salt bridge and the Arg-59 hydrogen bond with Ser-152, consequently reducing the cross-reactivity of Fabs 221-11 and 227-1 with ST3 and ST7. All side chains are drawn as sticks and color coordinated to the main chain color with nitrogen atoms blue and oxygen atoms red. Salt bridges are represented as yellow dashes and hydrogen bonds are red dashes.
Fig 6.
Phylogenic analysis of OspA within the Borrelia burgdorferi sensu lato (Bbsl) genospecies complex.
(A) Unrooted phylogenetic tree comprised of 135 OspA protein sequences from 23 Bbsl genospecies. Phylogenetic analysis identified 33 distinct OspA clades corresponding to serotypes (ST) 1-8, in silico types (IST) 9-17, and 16 other unclassified OspA IST’s. Clades representing each OspA variant/type are distinguished by color. Enlarged and bolded strain names denote OspA alleles selected for HB19-R1 viability reporter strain construction. Bootstrap values representing approximately maximum likelihood analysis are distinguished by color: > 0.50 = green, > 0.75 = blue, 1.00 = red. Nodes with bootstrap values under 0.50 are not depicted in the figure (gray) (B) Simplified multiple sequence alignment of the Bin1 epitope region of 33 unique OspA types. Representative sequences from each OspA clade were aligned using OspA from B. burgdorferi strain B31 as a reference. Bolded strain names indicate OspA variants expressed by HB19-R1 viability reporter strains. Amino acid residues numbered in white denote conservation within at least two Bin 1 mAb epitopes with solved Fab-OspA crystal structures (857-2, 221-5, 221-11, 221-7, and 227-1). Numbered amino acid residues with colored background form critical interactions with residues within the paratopes of Bin 1 mAbs with solved crystal structures (Lys-107, Glu-31, Ser-152).
Fig 7.
Genetically diverse OspA types are susceptible to complement-mediated killing by anti-OspAST1 Bin 1 mAbs.
Complement-dependent bactericidal assays were performed as described in Materials and Methods using anti-OspA Bin 1 mAbs and a panel of B. burgdorferi HB19-R1 mScarlet-I viability reporter strains expressing 19 OspA in silico types (ISTs). The heat map summarizes mean EC50 values for complement-mediated killing by Bin 1 mAbs from Classes 1-3. EC50 values represent the lowest mean antibody concentration (nM) producing >50% reduction in mScarlet-I fluorescence relative to untreated controls and were derived from ≥3 independent experiments. Susceptibility profiles of reporter strains expressing OspA ST1-7 are included for comparison. Strains that exhibited resistance to mAbs at 10 nM were retested at 66.6 nM. Corresponding antibody titration curves and statistical comparisons between the HB19-R1 OspA ST1 reporter strain and strains expressing OspA ST2-7 are shown in S8 Fig.
Fig 8.
Lys-107 polymorphisms are a determinant of susceptibility to nearly all Bin 1 mAbs Complement-dependent killing assays were performed as described in the Materials and Methods with anti-OspA Bin1 mAbs and B. burgdorferi HB19-R1 viability reporter strains expressing “WT” OspA variants from B. burgdorferi B31 (ST1), B. valaisiana VS116 (IST15), and B. turdi TPT2017 (IST16), or mutated derivatives containing deletions, insertions, or substitutions at residue 107.
Heat maps (A) and radar plots (B) depicting the Ec50 profiles of WT and mutated OspA variants. Ec50 values denote the lowest antibody concentration (nM) capable of reducing mScarlet-I fluorescence by >50% relative to untreated controls and represent the mean of > 3 independent experiments. Reporter strains exhibiting resistance to killing by mAbs at 20 nM were re-examined at 40 nM and 66.6 nM doses. (C) Heat maps graphically depicting significant differences in mAb susceptibility between HB19-R1 reporter strains expressing WT and mutated OspA variants as determined via two-way ANOVA followed by Tukey’s multiple comparisons test. Adjusted P-values reflecting significant differences between strains are shown in shades of purple or pink. Antibody titration curves summarizing borreliacidal activity against each strain are provided in S9 Fig.