Elsevier

Vaccine

Volume 26, Issue 35, 18 August 2008, Pages 4506-4512
Vaccine

Protection of mice against H. somni septicemia by vaccination with recombinant immunoglobulin binding protein subunits

https://doi.org/10.1016/j.vaccine.2008.06.046Get rights and content

Abstract

Histophilus somni causes bovine pneumonia as well as septicemia and its sequelae but mechanisms of virulence and protective immunity are poorly understood. Since surface immunoglobulin binding proteins are virulence factors, we addressed their role as protective antigens in a mouse model of H. somni septicemia. Immunoglobulin binding protein A (IbpA), has homology to Bordetella pertussis filamentous hemagglutinin and other large bacterial exoproteins. IbpA is a major surface antigen encoded by the ibpA gene with many domains that may be important in pathogenesis and immune protection. Three IbpA recombinant protein subunits, IbpA3, IbpA5 and IbpADR2 were chosen for study because of putative functional domains and motifs. These recombinant GST fusion subunit proteins were compared with GST (negative control), formalin-killed H. somni (commercial vaccine control), live H. somni (to induce convalescent immunity) and H. somni culture supernatant (containing IbpA shed from the bacterial surface). In vaccination/challenge studies, both live H. somni (convalescent immunity) and supernatant protected equally but formalin-killed H. somni and GST did not protect against septicemia. The DR2 and A3 subunits protected moderately well and induced antibody responses against supernatant antigen and the homologous subunit in ELISA but not against whole cell antigens. Supernatant immunization protected better than the IbpA subunit antigens and induced high antibody activity against both whole cells and supernatant antigens. The results indicate that culture supernatant antigens or perhaps recombinant IbpA subunits may be useful in H. somni vaccines. These studies also provide insight into the contribution of IbpA domains to pathogenesis of H. somni septicemia.

Introduction

Histophilus somni, also called Haemophilus somnus[1], is a major etiologic agent of the bovine respiratory disease complex [2], [3]. It also causes septicemia, thrombotic meningoencephalitis, myocarditis, arthritis, infertility, and abortion [4], [5], [6], [7], [8], [9], [10]. Infection is associated with vasculitis, thrombosis and macrophage degeneration [2]. In addition, infected cattle are often H. somni carriers without clinical signs [11]. Some strains from carriers (carrier or avirulent strains) do not express surface immunoglobulin binding proteins (IgBPs) [12] whereas all tested pathogenic invasive strains of H. somni have IgBPs, which bind bovine IgG2 by the Fc portion [13], [14]. These carrier strains were serum sensitive but the virulent strains were serum resistant [13], [15], [16]. The IgBPs are associated with resistance to complement-mediated killing of H. somni[12], [13]. This indicates that serum resistance and IgBPs may be related factors. Furthermore, cattle with disease due to H. somni develop high antibody titers to IgBPs [17]. Thus, IgBPs may be candidates for subunit vaccines. There is a need for new, more effective vaccines for H. somni disease because the efficacy of available vaccines is variable and controversial [18]. Even today most vaccines are composed of killed whole bacteria, which may be deficient in IgBPs because they are largely shed into the culture supernatant [14].

The IgBPs consist of a series of high molecular weight (HMW) proteins and a 76-kDa surface protein (p76) detected by SDS-PAGE and Western blotting [16]. Our original cloning studies showed that the HMW IgBPs and p76 were genetically linked [19]. When the whole DNA insert encoding the HMW IgBPs and the linked p76 was sequenced, only one open reading frame (ORF) was found [13], [19], [20]. This large (12.2 kb) gene, immunoglobulin binding protein A (ibpA) has multiple initiation sites, many of which are functional, partially accounting for the proteins of different sizes [21], [22]. The HMW IgBPs consist of a fibrillar network on the bacterial surface [16] and are shed in culture supernatant as well as being detected in the whole cell pellet [23]. The p76 IgBP is a surface exposed, peripheral outer membrane protein [16]. Sequencing of clones expressing p76 revealed a single 3.9-kb ORF containing tandem 1.2-kb direct repeats (DR1 and DR2) [21]. Later, several upstream domains in IbpA were defined, including IbpA3 and IbpA5 [20].

Current commercially available vaccines for H. somni utilize killed whole H. somni cells and have been only moderately successful [24], [25], [26]. Whole cell vaccines also have a reputation for occasionally causing severe adverse reactions in cattle [27]. Most vaccines still consist of whole killed H. somni, so better vaccines are needed. We showed that H. somni convalescent phase calf serum passively protected calves against experimental H. somni pneumonia [28]. Antigens recognized by this protective antiserum may be critical for immune protection. Our recent studies of a series of recombinant truncated IbpA subunits showed that glutathione S-transferase (GST)-fused recombinant subunits IbpA3 (aa 972–1515) and IbpA5 (aa 2071–2730) reacted strongly with protective calf convalescent phase serum [20], so it was hypothesized that these protein subunits may be protective. In addition, preliminary studies showed that bovine H. somni convalescent phase serum or rabbit antiserum to IbpA DR2 passively protected mice against H. somni septicemia [R. Kruger, JE Dixon, LB Corbeil, unpublished data]. Therefore, the purpose of this study was to evaluate the protective ability of the IbpA3, IbpA5, and IbpA DR2 recombinant subunits in comparison with crude native IbpA in culture supernatant. A mouse model of H. somni septicemia was used because bovine H. somni disease is mostly due to septicemia and its sequelae [4], [5], [6], [7], [8], [9], [10], [29]. The model includes preincubation of H. somni in fetal calf serum for 5 min to bind bovine transferrin [29]. This enhances virulence of H. somni for mice and simulates bovine septicemia more closely [29]. The level of protection was compared with vaccines composed of killed whole cells, live virulent organisms, and native IgBPs found in culture supernatant.

Section snippets

Animals and experimental design

Female, 5–6 weeks old, NIH Swiss Webster mice obtained from the National Cancer Institute (Fredrick, MD) through Charles River Labs were housed in groups of four to five in individual ventilated cages. Immunization experiments were conducted in three separate trials. In each experiment, four or five animals per group were immunized with antigens and adjuvant twice at a 3-week interval and challenged intraperitoneally 2–2.5 weeks later, as described below.

Bacterial strains and culture

The virulent H. somni strain, 2336 (from

Immunization of mice with H. somni culture supernatant or killed cells and challenge with H. somni

To determine if mice could be protected from H. somni septicemia, groups of five mice were immunized twice, 3 weeks apart, with culture supernatant or formalin-killed whole cells. Mice that had been challenged twice with live organisms but no FCS (convalescent mice) were used as a positive controls and mice that received the Quil A adjuvant with PBS alone served as a negative control. The mice were challenged IP 2 weeks after the second immunization, with 8.5 × 107 virulent H. somni organisms

Discussion

This study shows that culture supernatant, enriched for shed IbpA [23], protects mice against H. somni septicemia comparable to convalescent immunity but the formalin-killed H. somni vaccine did not protect. Since the killed cell vaccine group had high serum antibody values against H. somni whole cell antigen but relatively little antibody against supernatant or shed antigens, it can be concluded that the most protective antigens were shed in the 6 h (log phase) culture supernatant. This

Acknowledgements

We thank Dr. Robert Corbeil for statistical analysis and Jason Henry for technical assistance. This work was supported in part by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, grant number 2005-35204-16257 (LBC) and NIH RO1 AI 60662 (JED).

References (34)

  • F.W. Harris et al.

    The Haemophilus somnus disease complex (hemophilosis): a review

    Can Vet J

    (1989)
  • J.M. Kwiecien et al.

    Haemophilus somnus and reproductive disease in the cow

    Can Vet J

    (1991)
  • L.B. Corbeil et al.

    Haemophilus somnus: antigen analysis and immune responses

  • J.D. Humphrey et al.

    Prevalence and distribution of Haemophilus somnus in the male bovine reproductive tract

    Am J Vet Res

    (1982)
  • P.R. Widders et al.

    Immunoglobulin-binding activity among pathogenic and carrier isolates of Haemophilus somnus

    Infect Immun

    (1989)
  • S.P. Cole et al.

    Two linked genes for outer membrane proteins absent in four non-disease strains of Haemophilus somnus

    Mol Microbiol

    (1992)
  • M. Yarnall et al.

    Isolation and characterization of Fc receptors from Haemophilus somnus

    Scand J Immunol

    (1988)
  • Cited by (17)

    • Reverse vaccinology as an approach for developing Histophilus somni vaccine candidates

      2015, Biologicals
      Citation Excerpt :

      Globally it has affected countries such as Hungary, the UK, Argentina, Nigeria, and South Africa and more recently in the Paraná State of southern Brazil [6–8]. Virulence factors for H. somni include immunoglobulin binding proteins that can evade antibody defences, lipooligosaccharide (LOS) that may contribute to vasculitis, antigenic phase variation of phosphorylcholine (ChoP) in LOS epitopes, adherence to host cells, inhibition of phagocytic cell function, and the ability to obtain iron and other essential nutrients from the host [9–20]. Proteins such as high molecular weight immunoglobulin binding protein A (IbpA) encoded by a single large open reading frame of 12,285 bp; 76 kDa and 41 kDa proteins annotated as p76 and p41 respectively; lipoprotein B (LppB) identified as a 40 kDa protein annotated as p40; 31 kDa protein annotated as p31, which is homologous to lipoprotein Plp4 of Mannheimia haemolytica A1, and iron regulated transferrin-binding proteins (Tbps) are outer membrane proteins (OMPs) localized on the surface of H. somni that may be involved in virulence and host immunity [10,11,15,20–23].

    • Antibody responses of calves to Histophilus somni recombinant IbpA subunits

      2012, Comparative Immunology, Microbiology and Infectious Diseases
      Citation Excerpt :

      Protein concentrations were determined using the BioRad Protein Assay (BioRad Laboratory Inc., Hercules, CA). Purity determined by SDS PAGE and Western blotting was previously published [11,12]. For whole cell antigen, H. somni strain 2336 was grown in Bacto brain heart infusion (Becton, Dickinson and Company, Sparks, MD) broth with 1% Trisma base and 0.01% thiamine monophosphate (Sigma–Aldrich, St. Louis, MO) from BHI blood agar plate cultures.

    • IbpA DR2 subunit immunization protects calves against Histophilus somni pneumonia

      2011, Vaccine
      Citation Excerpt :

      Three IbpA domains (A3, A5 and DR2) were cloned and expressed as GST fusion proteins for purification (Fig. 1). Previously described plasmids pHS140 and pHS134 were used for cloning IbpA3 and IbpA5 [13,19] and pHS139 for the DR2 domain [15,17]. The forward and reverse primers were as follows:

    View all citing articles on Scopus
    1

    Present address: ULAR, University of California, Irvine, CA 92697, United States.

    2

    Deceased.

    View full text