Two monoclonal antibodies with defined epitopes of P44 major surface proteins neutralize Anaplasma phagocytophilum by distinct mechanisms

Abstract

Anaplasma phagocytophilum is an obligatory intracellular bacterium that causes human granulocytic anaplasmosis. The polymorphic 44-kDa major outer membrane proteins of A. phagocytophilum are dominant antigens recognized by patients and infected animals. However, the ability of anti-P44 antibody to neutralize the infection has been unclear due to a mixture of P44 proteins with diverse hypervariable region amino acid sequences expressed by a given bacterial population and lack of epitope-defined antibodies. Monoclonal antibodies (MAbs) 5C11 and 3E65 are directed to different domains of P44 proteins, the N-terminal conserved region and P44-18 central hypervariable region, respectively. Passive immunization with either MAb 5C11 or 3E65 partially protects mice from infection with A. phagocytophilum. In the present study, we demonstrated that the two monoclonal antibodies recognize bacterial surface-exposed epitopes of naturally folded P44 proteins and mapped these epitopes to specific peptide sequences. The two MAbs almost completely blocked the infection of the A. phagocytophilum population that predominantly expressed P44-18 in HL-60 cells by distinct mechanisms: MAb 5C11 blocked the binding, but MAb 3E65 did not block binding or internalization. Instead, MAb 3E65 inhibited internalized A. phagocytophilum to develop into microcolonies called morulae. Some plasma from experimentally infected horses and mice reacted with these two epitopes. Taken together, these data indicate the presence of at least two distinct bacterial surface-exposed neutralization epitopes in P44 proteins. The results indicate that antibodies directed to certain epitopes of P44 proteins have a critical role in inhibiting A. phagocytophilum infection of host cells.

FIG. 1.

MAb 5C11 recognizes the P44 N-terminal conserved region. (A) Affinity-purified rP44N (3 μg) and rP44N-N (3 μg) were subjected to SDS-PAGE followed by Coomassie blue staining. M, molecular size markers. rP44N is indicated by the arrow. (B) A duplicate gel was subjected to Western blot analysis with MAb 5C11.

FIG. 2.

MAb 5C11 and MAb 3E65 label the surface of A. phagocytophilum. Organisms were prefixed in paraformaldehyde (A to D) or in methanol (E to G) and subjected to immunofluorescence labeling with (A) MAb 5C11, (B) MAb 3E65, (C) MAb 5C11 and NtrX, (D) normal mouse IgG, (E) MAb 5C11 and NtrX, (F) MAb 3E65 and NtrX, or (G) MAb 5C11 and rabbit preimmune serum. Note ring-like surface labeling of individual organisms with two MAbs more clearly seen in the small figures to the right of panels (A and B, red; C and E to G, green). Note red (anti-NtrX) labeling of the cytoplasm of methanol-permeabilized bacteria (E and F). Bar, 5 μm.

FIG.3.

MAb 5C11 and MAb 3E65 inhibit infection of A. phagocytophilum in HL-60 cells by inhibiting binding and intracellular development, respectively. (A and B) Inhibition of infection of A. phagocytophilum or Ehrlichia chaffeensis with MAb 5C11, MAb 3E65, normal mouse IgG, or isotype-matched mouse IgG control. The percent inhibition of infection is expressed as described below. (A) Inhibition of A. phagocytophilum infection. (B) Inhibition of Ehrlichia chaffeensis infection. (C and D) Inhibition of binding of the host cell-free A. phagocytophilum with MAb 5C11, MAb 3E65, normal mouse IgG, isotype-matched mouse IgG control, or RPMI 1640 medium. (C) Immunofluorescence micrographs of A. phagocytophilum bound to HL-60 cells. Bar: 5 μm. (D) Percent inhibition of binding to HL-60 cells. Numbers of bound or internalized A. phagocytophilum were scored in 100 HL-60 cells in triplicate samples. The percent inhibition of binding or internalization is expressed as the number of bacteria per HL-60 cell incubated with RPMI medium minus the number of bacteria per HL-60 cell under the indicated conditions divided by the number of bacteria per HL-60 cell incubated with RPMI medium multiplied by 100. (A and D) Significantly different from the remaining groups by the Tukey honestly significant differences test (*, P < 0.01; **, P < 0.05). (E and F) Inhibition of transformation from individual bacterium to a microcolony (morula) by MAb 3E65 in HL-60 cells. (E) Number of total and extracellular bacteria in HL-60 cells at 4 h postinoculation; (F) size of bacteria or inclusion at 16 h postinoculation. (E and F) Significantly different between the two groups by Student's t test (*, P < 0.002).

FIG. 4.

Two views of the representative P44 protein three-dimensional structure predicted by the Robetta program and the MAb 5C11 epitope. A. Open hinge structure; B. closed hinge structure. The P44N-N and P44hv C-C region are shown in blue (the MAb 5C11 epitope is green) and red, respectively. From the N terminus, approximately 242 amino acids which include the whole N-terminal conserved region and approximately half of the central hypervariable region form an α-helix- and β-turn-rich domain which is expected to be surface exposed on the outer membrane of A. phagocytophilum. Another half of the central hypervariable region and most of the C-terminal conserved region (amino acid positions 243 to 440) are predicted to make a β-barrel structure embedded in the outer membrane of A. phagocytophilum. The P44N-N region (amino acid positions 35 to 98) is characterized by a four-stranded β sheet. The central hypervariable region (amino acid positions 195 to 229) between two absolutely conserved cysteines mainly contains β turns. The outer surface protruded domain and the membrane-embedded β-barrel are connected by the flexible hinge (amino acid positions 236 to 242, yellow).

FIG. 5.

Peptide mapping of MAb 5C11 and MAb 3E65 epitopes. Peptide-bound pins were incubated with either MAb 5C11 (upper panel) or 3E65 (lower panel). The amino acid sequences of overlapping synthetic peptides within P44N-N (A) and P44-18hvC-C (B) are arranged from the N to the C terminus and indicated on the x axis. Relative absorbance as determined by ELISA analysis is shown on the y axis. A representative result of triplicate assays is shown.

FIG. 6.

ELISA analysis for the linear B-cell epitopes within P44N-N and P44-18hvC-C of A. phagocytophilum-infected horses and mice. Preimmune and immune plasma from horse EQ001, EQ005, and EQ006; plasma from seven control horses; pools of plasma from three infected ICR, three infected C3H/HeN, and three infected C3H/HeJ strain mice; and three pooled uninfected plasmas from >5 mice each were allowed to react with the various synthesized peptides. The amino acid sequences of overlapping synthetic peptides are arranged from the N to the C terminus and indicated on the x axis. The y axis shows the difference in the absorbance (OD_415-492) of immune plasma (individual reactions are shown) and that of control plasma (means and standard deviations are shown). The 5C11 and 3E65 epitopes are indicated in square boxes. A reaction was considered to be positive when the immune plasma yielded an OD_415-492 value larger than the mean OD_415-492 + 3 standard deviations of negative control plasma or sera (shown as dash with solid triangle). A representative result of three to five assays is shown.