Immunization with the Haemophilus ducreyi hemoglobin receptor HgbA with adjuvant monophosphoryl lipid A protects swine from a homologous but not a heterologous challenge

Abstract

Haemophilus ducreyi, the etiological agent of chancroid, has a strict requirement for heme, which it acquires from its only natural host, humans. Previously, we showed that a vaccine preparation containing the native hemoglobin receptor HgbA purified from H. ducreyi class I strain 35000HP (nHgbAI) and administered with Freund's adjuvant provided complete protection against a homologous challenge. In the current study, we investigated whether nHgbAI dispensed with monophosphoryl lipid A (MPL), an adjuvant approved for use in humans, offered protection against a challenge with H. ducreyi strain 35000HP expressing either class I or class II HgbA (35000HPhgbAI and 35000HPhgbAII, respectively). Pigs immunized with the nHgbAI/MPL vaccine were protected against a challenge from homologous H. ducreyi strain 35000HPhgbAI but not heterologous strain 35000HPhgbAII, as evidenced by the isolation of only strain 35000HPhgbAII from nHgbAI-immunized pigs. Furthermore, histological analysis of the lesions showed striking differences between mock-immunized and nHgbAI-immunized animals challenged with strains 35000HPhgbAI but not those challenged with strain 35000HPhgbAII. Mock-immunized pigs were not protected from a challenge by either strain. The enzyme-linked immunosorbent assay (ELISA) activity of the nHgbAI/MPL antiserum was lower than the activity of antiserum from animals immunized with the nHgbAI/Freund's vaccine; however, anti-nHgbAI from both studies bound whole cells of 35000HPhgbAI better than 35000HPhgbAII and partially blocked hemoglobin binding to nHgbAI. In conclusion, despite eliciting lower antibody ELISA activity than the nHgbAI/Freund's, the nHgbAI/MPL vaccine provided protection against a challenge with homologous but not heterologous H. ducreyi, suggesting that a bivalent HgbA vaccine may be needed.

FIG. 1.

nHgbA_I/MPL vaccine reduces lesion severity of experimental chancroid at the macroscopic and microscopic levels. (A) Photographs of pig ear lesions. Pigs were either mock immunized with MPL adjuvant only (2 and 4) or immunized with nHgbA_I/MPL (1 and 3) and challenged with either H. ducreyi strain 35000HPhgbA_I (1 and 2) or 35000HPhgbA_II (3 and 4). Photos were taken 1 week after challenge and immediately prior to biopsy. (B) H&E-stained biopsy sections from pig ears 1 week after infection. Pigs were either mock immunized with MPL adjuvant only (6 and 8) or immunized with nHgbA_I/MPL (5 and 7) and challenged with either H. ducreyi strain 35000HPhgbA_I (5 and 6) or 35000HPhgbA_II (7 and 8) (magnification, ×50).

FIG. 2.

The activity of the nHgbA_I/MPL antisera is lower than that of the antisera from nHgbA_I/Freund's-immunized animals. Sera were collected preimmunization (prebleed), before the second immunization, and prior to challenge infection (three weeks after the third immunization). Data are expressed as OD₄₀₅ readings and given as median ± variance. Solid lines represent data from each pig immunized with nHgbA_I/Freund's adjuvant (1F, 4F, 5F, and 6F) (1), whereas dotted lines represent data from pigs immunized with nHgbA_I/MPL adjuvant (3M, 4M, 7M, and 8M). ELISA data for week 6 (prior to 3rd immunization) are not shown due to an incomplete data set for nHgbA_I/Freund's antiserum samples. P values comparing anti-nHgbA_I antibodies using Freund's and MPL adjuvants after one and after three immunizations (just prior to challenge) were 0.029 and 0.057, respectively, and were determined using the Mann-Whitney rank sum test (n = 3).

FIG. 3.

nHgbA_I/MPL antisera bind HgbA_I but not HgbA_II in the context of whole H. ducreyi cells. Antisera from individual pigs (pigs 3, 4, 7, and 8), obtained after three immunizations with nHgbA_I/MPL, and pooled antiserum from pigs immunized with nHgbA_I/Freund's (Fp; obtained from a previous study) (1) were tested for binding to whole cells of H. ducreyi strains 35000HPhgbA_I, 35000HPhgbA_II, 35000HPΔhgbA, and DMC111. The antisera used in these assays were from preinfection bleeds. Experiments were performed in triplicate on at least three different days. Data are expressed as relative light units and given as median ± variance. A P value of 0.005 was found for the difference between the binding of nHgbA_I/MPL and nHgbA_I/Freund's antisera to whole cells of H. ducreyi strain 35000HPhgbA_I (Mann-Whitney rank sum test).

FIG. 4.

nHgbA_I/MPL antiserum immunoprecipitates only HgbA_I from whole cells of H. ducreyi. The ability of anti-nHgbA_I antisera to immunoprecipitate HgbA from different H. ducreyi strains is shown in a Coomassie blue-stained 10% SDS-PAGE gel (A) and a Western blot probed with an antiserum to full-length rHgbA_I (19) (B). The H. ducreyi strains used in these immunoprecipitation assays were 35000HPhgbA_I (1), 35000HPhgbA_II (2), and DMC111 (3). Antisera used in these experiments were either pooled anti-nHgbA_I/Freund's (Fp) purified from antisera obtained in a previous study (1) or pooled anti-nHgbA_I/MPL (Mp). Molecular size markers are shown in kilodaltons.

FIG. 5.

IgGs from nHgbA_I/MPL-immunized animals partially inhibit Hb binding to nHgbA. The ability of pooled IgGs from pigs immunized with either nHgbA_I/MPL or nHgbA_I/Freund's to inhibit the binding of Hb to immobilized nHgbA_I or nHgbA_II was measured. P values comparing the ability of anti-nHgbA_I IgG to block Hb binding to nHgbA_I and nHgbA_II were determined using the Mann-Whitney rank sum test. Pooled nHgbA_I/Freund's IgG was purified from antisera obtained in a previous study (1) (n = 3).