1 Introduction Wagner et al., 2004 IGHG in vivo Wagner et al., 2004 Wagner, 2006 Wagner et al., 2002, 2004 Sheoran et al., 2000 Butler, 1998 Sheoran et al., 2000 Nelson et al., 1998; Breathnach et al., 2006 Streptococcus equi Galan and Timoney, 1985; Galan et al., 1986; Sheoran et al., 1997 Kydd et al., 2006 Although a role for equine IgG antibodies in protection against disease has long been recognised, the structures and functions of the individual IgG subclasses are not well characterised. Identification and cloning of the full complement of IgG H chain genes has provided a fresh resource for the study of equine IgG proteins. Here, we describe the first expression of recombinant versions of all seven equine IgG subclasses and present an analysis of their individual physical and biological properties. 2 Materials and methods 2.1 Construction of equine γ H chain expression vectors H Morton et al., 1993 IGHG3, IGHG4 IGHG7 IGHG1 IGHG2 IGHG5 IGHG6 Wagner et al., 2002, 2004 IGHG3 H IGHG3 H 2.2 Expression of reqIgGs in CHO-K1 cells Morton et al., 1993 Morton et al., 1993 2.3 Purification and analysis of reqIgGs Morton et al., 1993 2 2 Pleass et al., 1999 2.4 Reactivity of commercially available anti-horse IgG antibodies with the reqIgGs The following antibodies were tested for reactivity against the reqIgGs in ELISA: goat HRP-conjugated polyclonal antibodies specific for horse IgGa, IgGb, IgGc or IgG(T), (kindly provided by Serotec, Oxford, UK and Bethyl Laboratories, Montgomery, TX, USA), and mouse monoclonal antibodies (mAb) against horse IgGa (CVS48), IgGb (CVS39), IgGc (CVS53) and IgG(T) (CVS38) (kindly provided by Serotec). Secondary antibody used to detect binding of the mAb was a goat anti-mouse IgG1-HRP conjugate diluted 1/10,000 (kindly provided by Serotec). In addition, HRP-conjugated goat polyclonal anti-horse IgG H + L chain (kindly provided by Bethyl Laboratories) and goat anti-horse IgG Fc (Rockland) were tested in ELISA. Antigen-capture ELISAs were performed as above using 250 ng/ml of each purified reqIgG subclass. Detection antibodies were diluted 1/10,000 (goat polyclonals) or 1/100 (anti-IgGa mAb), 1/200 (anti-IgGb mAb), 1/10 (anti-IgGc mAb) and 1/500 (anti-IgG(T) mAb). The goat polyclonal anti-horse IgGa, IgGb, IgGc and IgG(T) antibodies were also tested in immunoblotting. Purified reqIgGs (1 μg) were electrophoresed under reducing and non-reducing conditions and the detection antibodies were diluted as for ELISA. 2.5 Analysis of reqIgG glycosylation N Pleass et al., 1999 2.6 Interaction with staphylococcal protein A and streptococcal protein G Interaction of reqIgGs with protein A and protein G was analysed by ELISA. NIP-BSA coated wells were incubated with 100 μl (0.25 μg/ml) of one of the purified reqIgGs or recombinant human IgG1 for comparison. Binding to protein A or protein G was detected by incubation with 100 μl of serial dilutions of either HRP-conjugated protein A (0–50 μg/ml in PBS-T) or HRP-conjugated protein G (0–5 μg/ml in PBS-T) (both Sigma, Poole, UK). Plates were developed with ABTS and absorbance read at 405 nm. 2.7 Functional assays Pleass et al., 1999 Bruggemann et al., 1987 2 2 3 Results 3.1 Analysis of purified antibody by SDS-PAGE and Western blotting Transfection of each of the equine γ1–γ7 H chain expression vectors into CHO cells stably expressing a compatible mouse λ light (L) chain allowed expression of all seven subclasses. Expressed Ig was purified by antigen affinity chromatography followed by FPLC. All seven IgGs eluted from FPLC as a single major peak representing monomer. Minor peaks eluting earlier were attributed to antibody that had aggregated during purification. For each subclass, fractions corresponding to the major peak were pooled and used for all subsequent experiments. Fig. 1 Fig. 1 Fig. 1 2 3.2 Reactivity of anti-horse IgG antibodies with the reqIgG subclasses Table 1 Table 1 Morton et al., 1993 Fig. 2 Fig. 2 Sheoran et al., 1998 Fig. 2 2 3.3 Analysis of the glycosylation of reqIgGs N N N N Fig. 3 N O O Fig. 3 N Fig. 4 N N N N 3.4 Interaction with staphylococcal protein A and streptococcal protein G Fig. 5 Sheoran and Holmes, 1996; Burton and Woof, 1992 Sheoran and Holmes (1996) Sugiura et al. (2000) Sheoran and Holmes, 1996; Sugiura et al., 2000 Sheoran and Holmes (1996) 3.5 Functional assays Fig. 6 Burton and Woof, 1992 Fig. 7 Fig. 7 4 Discussion Morton et al., 1993 2 2 O N O O Fanger and Smyth, 1972a, 1972b; Kim et al., 1994; Kabir et al., 1995 O O Wilson et al., 1991 O O Fanger and Smyth, 1972b; Kabir et al., 1995; Parham, 1983 2 2 2 2 Burton and Woof, 1992 Aalberse and Schuurman, 2002 Angal et al., 1993; Bloom et al., 1997 2 2 Sheoran and Holmes, 1996; Sugiura et al., 2000 Deisenhofer, 1981; Sauer-Eriksson et al., 1995 Fig. 8 Fig. 8 Burton and Woof, 1992 Fig. 8 Burton and Woof, 1992 Banks and McGuire, 1975 N Fig. 8 Burton and Woof, 1992; Woof and Burton, 2004 Fig. 8 Burton and Woof, 1992 Klinman et al., 1966 Duncan and Winter, 1988; Idusogie et al., 2000; Thomessen et al., 2000 Fig. 8 Our data on the effector function capabilities of the IgG subclasses have important new implications for the design of effective horse vaccines. It is clear that to achieve maximal protection via FcγR- and complement-mediated elimination mechanisms, vaccines should seek to elicit IgG antibodies of the IgG1, IgG3, IgG4 and IgG7 subclasses. Vaccines that trigger only IgG2, IgG5 or IgG6 antibodies are predicted to offer less effective protection. Since IgG plays key roles in both serum and mucosal compartments in the horse, these considerations are applicable to both systemic and mucosal vaccination strategies. Nelson et al., 1998; Breathnach et al., 2006; Soboll et al., 2003 Streptococcus equi Sheoran et al., 1997 Rhodococcus equi Lopez et al., 2002 Nelson et al., 1998; Lopez et al., 2002; Goodman et al., 2006 Sheoran et al., 2000 Balasuriya and MacLachlan, 2004 Snyder et al., 1981 Rhodococcus equi Escherichia coli Actinobacillus equuli Hietala and Ardans, 1987; Grondahl et al., 2001; Cauchard et al., 2004 Lunn et al., 1998