Abstract
OBJECTIVE
Evaluate the immunogenicity of a vaccine targeting the S protein (Ssee) of Streptococcus equi subsp equi and determine antibody activity against Ssee in horses with strangles.
METHODS
The study was designed as a prospective experiment using 20 university-owned Quarter Horses and a cross-sectional serosurvey of 78 privately owned horses with strangles. Horses were immunized IM with 0 (n = 4), 200 (n = 8), or 400 (n = 8) μg of recombinant Ssee at weeks 0, 4, and 12. Serum and nasal secretions were collected at weeks 0, 4, 6, 12, 16, and 28 and tested by ELISA for immunoglobulin (Ig)-G against Ssee; nasal secretions were also tested for anti-Ssee IgA. The function of anti-Ssee IgG in serum was tested for complement deposition onto Ssee and opsonophagocytic killing of S equi subsp equi. Serum from horses with strangles was tested by ELISA for anti-Ssee IgG activity.
RESULTS
Immunization with Ssee significantly (P < .05) increased serum and nasal IgG (but not nasal IgA) against Ssee for up to 12 weeks after the third immunization, and serum from vaccinated horses mediated significantly (P < .001) greater complement deposition onto Ssee, but not opsonophagocytic killing (P > .05), than controls. Horses with strangles did not develop high levels of serum IgG activity against Ssee.
CONCLUSIONS
Immunizing horses with Ssee resulted in increased activity of functional IgG in serum and nasal secretions, and horses with strangles had very low levels of serum IgG activity against Ssee.
CLINICAL RELEVANCE
S protein has potential as a vaccine to reduce the severity of strangles and differentiate between infected and vaccinated horses.
Strangles is a highly contagious disease that is considered the most frequently diagnosed infectious disease of horses.1–10 Strangles is caused by infection with Streptococcus equi subsp equi (SEE), a group C Streptococcus. Infection with SEE results in fever, profuse nasal discharge, swollen lymph nodes, and subsequent abscess formation in the nodes draining the upper respiratory tract.1–10 During outbreaks, morbidity is generally high, and mortality may occur when the upper airways become obstructed by swollen lymph nodes.5,6,8,9 Infection can spread to other organs (so-called bastard strangles), and sequelae, such as muscle disorders that may be life-threatening, can develop.1–5,9–15 Outbreaks can result in closure of events and venues, leading to lost revenue in addition to direct expenses of disease.
Streptococcus equi subsp equi produces long-lasting immunity after natural infection,1–10 has limited genetic diversity,8,16–19 does not have intermediate or alternative hosts,1–10 and does not persist long term in the environment,20,21 indicating that a vaccine could be effective for controlling strangles. Vaccines currently available in the US have limitations. Vaccines using modified-live strains of SEE can cause clinical disease indistinguishable from strangles16,22,23 and are not licensed for use in horses aged ≤ 10 months. Moreover, intranasal inoculation of live vaccines results in strong avoidance behavior by many horses. Vaccines of killed bacteria or bacterial components are much safer, but available products in the US lack evidence of inducing a protective immune response. Recently, a vaccine made of multiple proteins of SEE has been licensed in the European Union (EU) and the United Kingdom (UK) for IM use,24 indicating that it is possible to develop an effective strangles vaccine based on proteins of SEE. Although ponies immunized with the vaccine available in the EU and UK did not react to a commercial indirect ELISA,24 none of the available vaccines in the US permit differentiation of infected from vaccinated animals (DIVA). A DIVA vaccine would enable serological identification of infected horses to better control and prevent strangles.
We (IHW and DJG) were the first to identify S protein as a virulence factor in streptococcal bacteria belonging to groups A (GAS) and B (GAS).25,26 The S protein maintains GAS cell wall properties, and its absence reduces the survival of GAS in human blood.25 The S protein facilitates GAS coating with lysed RBCs to promote molecular mimicry, increasing virulence in vitro and in vivo.25 Deleting the gene encoding the S protein from GAS alters cellular and extracellular protein landscapes and is accompanied by a decrease in the abundance of several key GAS virulence factors, including M protein.25 In vivo, S protein–deleted GAS shows marked attenuation of virulence, promoting immunity against GAS infection.25
Here, we describe identification of the S protein sequence in SEE (Ssee), its cloning and expression, and serological responses of horses immunized with Ssee and horses naturally infected with SEE. The results indicate that immunizing horses with Ssee induces antibodies that deposit complement specifically onto Ssee and that antibodies to Ssee were not detected among horses naturally infected with SEE. These findings indicate that an Ssee vaccine might have the capacity for DIVA.
Methods
Ethics statement
All methods were performed in accordance with relevant guidelines and regulations for animal use and for laboratory practices (Texas A&M University Infectious Biohazard Committee #2017–105, IACUC Animal Use Protocol [AUP] #2022–0026, IACUC AUP #2023–0302 CA, and IACUC AUP #2023–0227), including informed consent for serum samples collected from client-owned horses.
Cloning, expression, and purification of recombinant Ssee
The genome sequence of the S protein of GAS was aligned with the genome of a reference strain of SEE to identify the ortholog in SEE, which we refer to as Ssee. A total of 670 S equi genome assemblies derived from an extensive global sampling study8 were downloaded in FASTA format and used to create a custom BLAST database.27 Our recombinantly expressed S equi S protein sequence, derived from accession WP_111678006, was subsequently used to query this S equi genome database using the tblastn command line tool,27 the number of amino acid mismatches relative to our query sequence was then plotted, and the sequences were aggregated and visualized using the ggmsa R package (Supplementary Figure S1).28 The sequence for Ssee was found to be present and highly conserved in all strains of SEE.
The coding sequence of Ssee was cloned, sequenced, and purified in the Protein Production, Characterization, and Molecular Interaction directed by WL. Briefly, the coding sequence of Ssee (Sag1386) with a StrepII tag was codon optimized and synthesized by Integrated DNA Technologies as a gene fragment, and then the synthesized Ssee StrepII gene fragment was cloned into an in-house derivative of pET28b vector (Agilent Technologies) encoding an N-terminal His6 tag with a tobacco etch virus (TEV) protease cleavage site by using the In-Fusion HD Cloning system (Takara Bio), producing a construct that expressed His6-TEVsite-Sag1386_StrepII under the driver of T7lac promoter. The integrity of the resulting plasmid was confirmed by DNA sequencing as well as restriction enzyme digestion.
The Escherichia coli strain ClearColi BL21(DE3) (Lucigen) was transformed with the pET28b/His6-TEVsite-Sag1386_StrepII expression plasmid and grown overnight in Terrific Broth media containing kanamycin (50 µg/mL) at 37 °C. The overnight culture was used to inoculate Terrific Broth containing kanamycin (50 µg/mL) and incubated while being shaken at 37 °C until the absorbance at 610 nm reached 0.8. After the culture was chilled to 4 °C, 0.5 mM of isopropyl 1-thio-β-D-galactopyranoside was added, and the culture was shaken for 24 hours at 20 °C. Cells were harvested by centrifugation at 5,000 X g for 15 minutes.
To purify recombinant His6-TEVsite-Sag1386_StrepII, cells were resuspended in 50 mM Tris/HCl (pH 8.0) containing 500 mM NaCl, 5 mM MgCl2, 20 mM imidazole, 0.5% Tween 20, and 10% glycerol. To decrease the viscosity, DNase I was added. The homogeneous suspension was lysed with 2 passes through an M-110 P microfluidizer (Microfluidics) at 20,000 psi and then centrifuged for 45 minutes at 100,000 X g and 4 °C. The supernatant containing recombinant His6-TEVsite-Sag1386_StrepII was then applied onto HisTrap HP columns (GE Healthcare) for Ni2+-based affinity chromatography, followed by a thorough wash using wash buffer (50 mM Tris, 350 mM NaCl, 10% glycerol, 20 mM imidazole, and 0.025% [w/v] n-dodecyl-β-D-maltoside, pH 8.0) and then eluted with the same buffer containing 500 mM imidazole, which (ie, imidazole) was then removed by using a HiPrep 26/10 desalting column (GE Healthcare) that was already preequilibrated with wash buffer. To remove the His6-TEVsite tag from the His6-TEVsite-Sag1386_StrepII protein, His6-tagged TEV protease was added with a ratio of 1:100 (ie, 1 µg of TEV protease:100 µg of His6-TEVsite-Sag1386_StrepII) and incubated overnight at 4 °C. The reaction mixture was then applied onto HisTrap HP columns, and the flow-through was harvested and loaded onto StrepTrap HP columns for StrepTactin-based affinity chromatography. After a thorough wash with wash buffer, the purified Sag1386_StrepII was then eluted with wash buffer containing 2.5 mM desthiobiotin. The eluates were harvested, concentrated, and polished with size exclusion chromatography by using a 120-mL HiLoad 16/600 Superdex 200 column (GE Healthcare) with PBS buffer containing 0.025% (w/v) n-dodecyl-β-D-maltoside as elution buffer. The fractions containing purified Sag1386_StrepII, which behaves as an oligomeric form on size exclusion chromatography, were then pooled and concentrated by using Amicon Ultra-15 Centrifugal Filters with 100,000 nominal molecular weight limit (Merck Millipore). The whole process was conducted on the ÄKTA pure system (GE Healthcare).
The purity of the purified protein was checked via SDS-PAGE and Coomassie blue staining (Supplementary Figure S2). The protein concentration was determined by measuring UV absorbance at 280 nm with a BioPhotometer D30 (Eppendorf) using molar extinction coefficients calculated from the amino acid sequences of Sag1386_StrepII protein (with a calculated molecular weight of 19,098 Da, or 19.1 kDa). Purified protein (which behaves as an oligomeric form) was used to immunize horses and to coat immunoassay plates to determine the activity of immunoglobulins against Ssee and complement deposition by ELISA onto Ssee (as described below).
Study population
A group of 20 Quarter Horse geldings (n = 10) and mares (n = 10) owned by Texas A&M University and ranging in age from 8 to 25 years (median age, 18 years) were used to assess the immunogenicity of the Ssee vaccine (described below). These horses had no known history of strangles, exposure to strangles, or vaccination against strangles for at least the preceding 7 years. All horses were serologically negative (all samples < 20% reactivity relative to positive control by indirect ELISA using plates coated either with lysate of SEE or recombinant SEE M [SeM] protein performed in our laboratory). Use of these horses for this study was approved by the Texas A&M University IACUC (protocol #2022–0226). Horses were group housed in pastures and paddocks at the Veterinary Medical Park at Texas A&M University with ad libitum access to coastal Bermuda grass hay and fed a 14% protein pelleted ration at approximately 1% of body weight daily. Horses were randomly assigned to 1 of 3 groups: (1) control horses immunized with Montanide Gel 01 (Seppic Inc) only (n = 4; 2 geldings and 2 mares); (2) vaccine formulation with 200 µg of recombinant Ssee (n = 8; 4 geldings and 4 mares); or (3) horses with 400 µg of recombinant Ssee (n = 8; 4 geldings and 4 mares). Horses were housed with their usual herd mates in paddocks and pastures and were not segregated by vaccine group during the study. Vital signs (pulse, respiratory rate, and heart rate) and evidence of swelling or painful response to palpation were monitored by veterinarians or veterinary technical staff twice daily for 3 days after each immunization. Horses were immunized at baseline (week 0) and weeks 4 and 12. Blood samples were collected by jugular venipuncture from horses prior to immunization at weeks 0 (prior to immunization), 4 (prior to immunization), 6, 12 (prior to immunization), 16, and 28; blood samples were maintained at approximately 22 °C until serum was separated within 2 hours of collection. At these same timepoints, nasal secretions were collected using nasal sponges attached to string placed in the nasal cavity for 5 minutes prior to retrieval (as previously described29); nasal sponges were maintained at approximately 22 °C in sterile conical tubes until secretions were collected by centrifugation of sponges at 700 g for 10 minutes within 2 hours of collection. Serum samples and nasal secretions were aliquoted as 1-mL volume into cryovials and maintained frozen at −80 °C for 29 weeks until tested by ELISA. Serum samples were tested for activity of immunoglobulin (Ig)-G reactivity to recombinant Ssee by ELISA as described below, and nasal secretions were tested for IgG and IgA reactivity to S protein by ELISA as described below. Aliquots of serum samples collected from 78 horses with strangles for another project were available for testing by ELISA for S protein. These serum samples also were aliquoted as 1-mL volume into cryovials and maintained frozen at −80 °C for between 10 and 51 weeks until tested by ELISA.
S protein sequence in SEE vaccine
To immunize horses, 200 or 400 μg of recombinant Ssee was diluted in sterile 0.9% NaCl (saline) solution with a commercial adjuvant (Montanide Gel 01; Seppic Inc) at a final concentration of 10% in a total volume of 1 mL for injection. The vaccine was freshly prepared on the day of each immunization. Each horse was vaccinated IM in the neck on 3 occasions (weeks 0, 4, and 12), alternating sides between immunizations (left side of the horse for immunizations 1 and 3, right side for immunization 2).
Immunofluorescence demonstrating surface expression of Ssee
Eight-well, tissue culture–treated, chambered slides (Ibidi USA Inc) were coated with approximately 2.5 X 107 CFUs of either a strain of SEE obtained from a horse with strangles from Texas or a strain of Rhodococcus equi obtained from a foal with pneumonia from Texas; the bacteria were fixed with ethanol and acetic acid. Slides coated with SEE and R equi were incubated overnight at 4 °C with 250 μL each of the following primary antibodies diluted 1:8,000 in 1% bovine serum albumin: (1) serum from a horse hyperimmunized against either SEE (SEE positive control, generously provided by Mg Biologics) or R equi (R equi positive control containing high titer to the surface-expressed virulence-associated protein A of R equi, generously provided by Mg Biologics); (2) serum from a horse immunized against Ssee (Ssee positive control); (3) serum from a 28-day-old foal that had neither been exposed to SEE nor immunized with Ssee (negative control primary antibody); and (4) PBS only (negative control with no primary antibody). Bacterial cells were not permeabilized to restrict staining to the outer surface of either SEE or R equi. After overnight incubation, the chambered wells were washed 5 times with PBS, and 250 μL of a 1:500 dilution of goat anti-horse IgG (heavy and light chains) conjugated to a fluorophore (DyLight; Bethyl Laboratories) was added to wells. After incubation at 4 °C for 1.5 hours, Hoechst 33342 stain in 1% bovine serum albumin (3 μg in 250 μL) was added to each well and incubated at 4 °C for 0.5 hours. Stain was then removed by aspiration, and wells were washed 3 times in PBS and imaged with a fluorescent microscope (EVOS FL AutoImaging System; ThermoFisher Scientific).
S protein sequence in SEE ELISA testing
Indirect ELISAs were used to determine activities against Ssee of IgG in serum and IgG and IgA in nasal secretions. Immunoassay plates (Nunc Maxisorp immunoassay plates; ThermoFisher) were coated overnight with recombinant Ssee (2 µg/mL) diluted in coating buffer (BioLegend; catalog #421701). After blocking of excess protein binding sites with 1% skim milk in PBS, plates were incubated with the serum samples from horses diluted 1:5,000 in blocking buffer or nasal secretions diluted at 1:500 for IgA detection and 1:5,000 for IgG detection. After washing the plates, the primary antibodies were detected using horseradish peroxidase–conjugated isotype-specific secondary antibodies (goat anti-equine IgG [Jackson ImmunoResearch; catalog #108–035-003] and goat anti-equine IgA [Bethyl Laboratories; catalog #A70–103P]) and a substrate system containing tetramethylbenzidine and peroxide as described in previous studies.29–31 Serum from a horse that had been immunized with Ssee in preliminary studies that had high reactivity was included as a positive control. Optical densities (ODs) were determined colorimetrically at 450 nm using a plate reader (Cytation 5; Agilent Technologies). Samples from all timepoints for each individual horse were tested on the same plate. The results were represented as the ratio of the OD for the sample (minus the blank) to the positive control (minus the blank). The same methods were used for SEE and SeM ELISAs to test horses for evidence of exposure to SEE except that plates were coated overnight with either SEE lysate or recombinant SeM (2 µg/mL for each antigen).
Complement deposition assay
Complement component 1 q (C1q) deposition assays were conducted to compare the ability of serum from vaccinated and unvaccinated horses to deposit complement onto Ssee as previously described.29,30,32 Maxisorp 96-well flat-bottom immunoassay plates (Thermo Fisher Scientific) were coated with Ssee as described for ELISA and then stored overnight at 4 °C. After washing the plates 3 times with PBS plus 0.05% Tween, they were blocked for 60 minutes at 37 °C with 150 μL of PBS plus 1% skim milk (1 g/100 mL). All C1q deposition assays were performed in duplicate using serum samples that were heated to inactivate complement (56 °C for 30 minutes) and diluted to 1:20 in gelatin veronal buffer (Boston Bioproducts). Immunoglobulin G/IgM/IgA-depleted human serum (Pel Freeze Biologicals; catalog #34041-5) was diluted to 1:20 in gelatin veronal buffer and served as a complement source. In each well, 50 μL of each sample was combined with 50 μL of complement source (volume totaling 100 μL/well) for 60 minutes at 37 °C with gentle shaking at 200 rpm. After washing, 100 μL of affinity-purified goat anti-complement C1q primary antibody (Cedarlane Labs) at a dilution of 1:10,000 was placed in each well and incubated for 60 minutes at room temperature (approx 22 °C). This was followed by washing, after which 100 uL of a rabbit anti-goat IgG alkaline phosphatase secondary antibody (Sigma-Aldrich) was placed in each well at a dilution of 1:2,000. Plates were incubated for 60 minutes at room temperature. After another 3 wash cycles, the alkaline phosphatase–enzymatic production of color was elicited using 100 L/well of 1-Step PNPP Substrate (Thermo Fisher Scientific) for 15 minutes at 37 °C. The ODs for each plate were recorded at a wavelength of 405 nm. An OD ratio calculated as described for Ssee ELISA was used as the outcome for data analysis.
Opsonophagocytic killing assay
To determine opsonic killing of SEE, bacterial cultures were routinely grown overnight at 37 °C on blood agar plates, and then killing was assessed using a previously described protocol30,32 modified for SEE. Briefly, polymorphonuclear cells (PMNs) for the opsonophagocytic killing (OPK) assays were prepared from fresh equine blood collected from a single Quarter Horse mare from the herd of origin of the study horses using a double-layer Histopaque (Sigma-Aldrich; isolation densities, 1.077 and 1.191). Each isolation was confirmed to be composed of > 95% PMNs using microscopy. The concentration of PMNs was adjusted to 1 X 107 cells/mL in modified RPMI 1640 medium free of phenol red (Lonza Bioscience) plus 5% heat-inactivated (ie, 56 °C for 30 minutes) fetal bovine serum (Thermo Fisher Scientific). Sera were diluted 1:10 in RPMI for testing and heat-inactivated (56 °C for 30 minutes) to eliminate differences between horses in endogenous complement. Exogenous complement (Pel-Freez Biologicals; catalog #34041–5) was added to a final concentration of 25%. Overnight cultures of SEE were grown in Todd-Hewitt broth at 37 °C. Bacterial suspensions were made at an OD of 650 nm of approximately 0.95 (corresponding to approx 1 X 108 CFUs/mL), and the final concentration was adjusted to 1 X 107 CFUs/mL in RPMI for use in the OPK assay. The assay was performed by mixing 100 μL of each of the test sera (heat inactivated) with complement, the PMN suspension, and the SEE solution. The reference negative control was composed of 100 μL of bacterial solution added to 300 μL of RPMI to maintain a consistent volume among wells. A negative control with only PMNs was included to account for the effects of PMNs without serum or complement. The assay tubes were incubated on a rotator rack at 37 °C for 4 hours; samples were taken at time 0 and after 4 hours. Samples were plated onto blood agar plates in duplicate at 50 μL/plate. The CFUs were quantified after a 24-hour incubation at 37 °C. For each horse, the ratios of the CFUs from assays using sera from week 16 (4 weeks after the third immunization) to the CFUs from baseline (week 0) were calculated and used as the analysis. Samples were tested in duplicate, and the ratio of the mean of the CFUs of SEE at week 16 (4 weeks after the third immunization) divided by the mean of the CFUs of SEE at week 0 (prior to immunizations) was calculated; this ratio representing proportional OPK mediated by serum was compared between controls and vaccinates in the low- and high-dose groups.
Data analysis
Analysis was performed using R statistical software (version 4.4.2; R Foundation for Statistical Computing). Data were displayed in plots, and linear mixed-effects modeling with the nlme package in R was used to examine the effects on ELISA OD ratios of vaccine group and study week and the interaction of group and week, with individual horse modeled as a random effect to account for repeated measures on individual horses over time. Post hoc pairwise differences between pairs of groups and times were made using the method of Tukey with the multcomp package in R. Plots of residuals indicated that log10 transformation of the serum IgG improved model fit. Correlations between nasal and serum IgG concentrations were assessed using a Spearman correlation coefficient (Spearman ρ) and significance test with the cor.test function in R. Criteria for interpreting correlation based on the absolute value of the Spearman ρ were as follows: (1) > 0.9 to 1.0 is very strong; (2) > 0.70 to 0.90 is strong; (3) > 0.4 to 0.7 is moderate; (4) > 0.2 to 0.4 is weak; and (5) ≤ 0.2 is very weak.33 Comparisons of Ssee OD ratios between serum from horses with strangles and serum from the 16 Ssee-vaccinated horses 4 weeks after the third immunization were compared using a generalized linear model with Gaussian link using the glm function in R. Generalized linear modeling was also used to compare proportional killing (ratios of CFUs after immunization to before immunization from OPK assays). Significance for all analyses was set at P < .05.
Results
Surface expression of Ssee
The results of immunofluorescence indicated that Ssee was expressed on the surface of SEE (Supplementary Figure S3). Specificity of the polyclonal anti-Ssee serum was indicated by the absence of fluorescence detected either using serum from a horse not immunized against Ssee as primary antibody or with the secondary antibody alone of slides coated with SEE, and by absence of fluorescence using serum from a horse immunized against Ssee of slides coated with R equi (Supplementary Figure S3).
Immunogenicity of Ssee vaccine
Serum IgG activity against Ssee varied significantly by study group in a time-dependent manner (Figure 1; Supplementary Table S1). Anti-Ssee activity increased significantly by 4 weeks and 6 weeks following the initial immunizations in both groups of vaccinated horses; anti-Ssee activity decreased significantly by week 12 relative to the values at weeks 4 and 6 but remained significantly greater than values at baseline (week 0), rose significantly at week 16 (4 weeks after the third immunization at week 12), and remained significantly higher than baseline at week 28 (16 weeks after booster). The results for nasal secretion samples followed a similar pattern (Figure 2; Supplementary Table S2). The values of serum and nasal IgG activity levels were strongly and significantly correlated (ρ = 0.860 and 0.818 for the low- and high-dose vaccine groups, respectively; P < .001, respectively; Supplementary Figure S4) but only weakly correlated for the unvaccinated control horses (ρ = −0.200; P = .361).
Boxplots of serum immunoglobulin (Ig)-G optical density (OD) ratios ([ODsample – ODblank]/[ODpositive control – ODblank]) faceted by study group. Control = unvaccinated controls, low = 200 μg S protein (Ssee; low dose), and high = 400 μg Ssee (high dose). Arrows denote each of the 3 timepoints of immunization. The tops and bottoms of the boxes represent the 25th and 75th percentiles of the data, and horizontal lines bisecting boxes are the medians; vertical lines extend by multiples of 1.75 of the respective interquartile distance. Different superscript numbers indicate significant (P < .05) pairwise differences between times within and among groups after adjusting for multiple comparisons. Control = Unvaccinated controls. Low = 200 μg S protein (Ssee; low dose). High = 400 μg Ssee (high dose).
Citation: American Journal of Veterinary Research 86, 2; 10.2460/ajvr.24.08.0228
Boxplots of nasal IgG OD ratios ([ODsample – ODblank]/[ODpositive control – ODblank]) faceted by study group. Arrows denote each of the 3 timepoints of immunization. The tops and bottoms of the boxes represent the 25th and 75th percentiles of the data, and horizontal lines bisecting boxes are the medians; vertical lines extend by multiples of 1.75 of the respective interquartile distance. Different superscript numbers indicate significant (P < .05) pairwise differences after adjusting for multiple comparisons.
Citation: American Journal of Veterinary Research 86, 2; 10.2460/ajvr.24.08.0228
The anti-Ssee activity of nasal IgA did not change over time for control horses; temporal changes for vaccinated horses were less consistent and less strong (Figure 3; Supplementary Table S3) than was observed for IgG in serum or nasal secretions. At weeks 4 (P = .009) and 6 (P = .032), values were higher than baseline (week 0) for the low-dose vaccine group; values at weeks 4 and 6 were also significantly higher than those at week 12 (P < .010), but no other pairwise differences between times were significant for this group. For the high-dose vaccine group, although nasal IgA activities against Ssee appeared to be higher at week 4 than at week 0, this difference was not significant (P = .065); however, values at week 4 were significantly higher than those at week 12 (P = .046). No other pairwise differences between times were significant for the high-dose vaccine group. The activities of anti-Ssee IgG and IgA in nasal secretions were moderately and significantly correlated for the control group (ρ = 0.604; P = .002) but only weakly, albeit significantly, correlated for horses in the low- (ρ = 0.398; P = .006) and high-dose (ρ = 0.319; P = .027) groups (Supplementary Figure S5).
Boxplot of nasal IgA OD ratios ([ODsample – ODblank]/[ODpositive control – ODblank]) faceted by study group. Arrows denote each of the 3 timepoints of immunization. The tops and bottoms of the boxes represent the 25th and 75th percentiles of the data, and horizontal lines bisecting boxes are the medians; vertical lines extend by multiples of 1.75 of the respective interquartile distance. Different superscript numbers indicate significant (P < .05) pairwise differences after adjusting for multiple comparisons.
Citation: American Journal of Veterinary Research 86, 2; 10.2460/ajvr.24.08.0228
Complement deposition
Deposition of C1q onto Ssee did not differ significantly between controls and vaccinated horses at baseline (Figure 4; Supplementary Table S4). At week 16, deposition of C1q for horses in the low- or high-dose groups was significantly (P < .001) greater than baseline (week 0) or either week 0 or week 16 in controls; however, C1q deposition did not differ significantly at week 16 between the 2 vaccinated groups (Supplementary Table S4). Serum C1q deposition onto Ssee was significantly correlated with serum IgG activity (Supplementary Figure S6).
Scatterplot of complement component 1q (C1q) deposition onto Ssee OD ratios ([ODsample – ODblank]/[ODpositive control – ODblank]) faceted by study group. Different superscript numbers indicate significant (P < .001) pairwise differences.
Citation: American Journal of Veterinary Research 86, 2; 10.2460/ajvr.24.08.0228
Opsonophagocytic killing assay
No significant difference was detected between controls and either the low- or high-dose vaccinates (P = .599 and 0.716, respectively; Supplementary Figure S7). Although the proportion of controls with OPK ratios ≤ 0.75 (1 of 4 [25%]) was less than that of the vaccinates (9 of 16 [56%]), this difference in proportions was not significant.
Horses with strangles do not have IgG activity against Ssee
To determine whether immunization with Ssee enabled differentiation of vaccinated horses from those infected naturally, serum from horses with strangles was tested for activity against Ssee. Serum was available from 78 horses with strangles. The horses with strangles ranged in age from 4 months to 26 years (mean, 4.5 years; SD, 6.79 years). Most horses with strangles (72 of 78 [92%]) were from Texas; 3 horses were from California, 2 from Oregon, and 1 from Idaho. Breed was reported for 74 of the horses with strangles: 57 were Quarter Horses (77%), 7 were European Warmbloods (9%), 5 were Arabians (7%), 2 were Thoroughbreds, 1 was a Tennessee Walking Horse, 1 was a pony, and 1 horse was crossbred. Of the 74 horses with strangles, 64 had clinical signs of strangles for at least 10 days at the time of sample collection, and 14 horses had recovered from clinical signs within 4 weeks of sample collection. The mean OD ratio of horses with strangles (0.035; 95% CI, −0.137 to 0.207) was significantly (P < .001) less than that of vaccinated horses (mean, 1.998; 95% CI, 1.841 to 2.155; Figure 5). Values of OD ratios for strangles horses were similar to those of study horses prior to vaccination (Supplementary Tables S1 and S2). The results were similar when comparing the vaccinated horses to horses with strangles using samples collected at week 4 (4 weeks after the first immunization) or week 6 (2 weeks after the second immunization; data not shown).
Comparison of the distribution of activity of IgG against recombinant Ssee of horses vaccinated with Ssee (n = 16) compared with serum samples from horses with strangles (n = 78). The tops and bottoms of the boxes represent the 25th and 75th percentiles of the data, and horizontal lines bisecting boxes are the medians; vertical lines extend by multiples of 1.75 of the respective interquartile distance. The OD ratios for vaccinated horses were significantly (P < .001) greater than those of horses with strangles.
Citation: American Journal of Veterinary Research 86, 2; 10.2460/ajvr.24.08.0228
Discussion
The Ssee vaccine elicited strong IgG responses in serum and nasal secretions of horses, and these antibodies deposited complement onto Ssee, indicating they were functional. Given evidence that anti–S protein IgG is protective against murine models of GAS, further evaluation of S protein as a vaccine antigen for SEE is warranted. Activities were highest at 16 weeks, 4 weeks after the third immunization, administered to horses 12 weeks after the initial immunization. The vaccine induced significantly increased serum and nasal anti-Ssee IgG activity after a single immunization, and activities were not significantly higher 2 weeks after the second immunization, administered 4 weeks after the initial immunization. The latter finding was unexpected. Because most horses did have a small increase in activity level, it is possible that we lacked adequate statistical power to detect a small but significant boost in IgG activity levels following the second immunization. Variation in vaccination technique or conditions (eg, the horse jumps away during injection, or a hemorrhage in the needle hub indicates the need to reposition the needle) or individual responses to immunization could account for lower responses to immunization in some horses. An anamnestic response would be expected to be detected 2 weeks after the second immunization, so it is unlikely that collecting serum sooner (eg, at 10 days) or later (eg, 4 weeks) after the second immunization would have changed the results.
Vaccinated horses deposited complement onto recombinant Ssee significantly more than control horses. Complement deposition reflects a functional response that could contribute to killing SEE. The results of OPK assays, however, did not demonstrate significant differences between Ssee-vaccinated horses and control horses. The reason for this discrepancy between these 2 functional assays is unclear. Although positive and negative controls for the OPK assays were included, considerable biological variability was noted for this assay, indicating that the reliability of the results was weak. Ultimately, the efficacy of immunization using Ssee to protect horses against infection with SEE will need to be determined in vivo because functional assays, such as complement deposition or OPK, do not always correlate with immunity.29
No significant or qualitative difference was noted between the 200- and 400-μg doses of Ssee, indicating that the lower dose would be suitable for future studies. No adverse reactions were observed in any horses, indicating that the combination of antigen and adjuvant were well tolerated. Larger-scale studies are needed to better characterize the immunogenicity and safety of this Ssee vaccine.
Although some evidence of increased anti-Ssee activity of IgA in nasal secretions (indicative of mucosal immune responses) was detected, these changes were small and inconsistent in magnitude and statistical significance. Immunization via the IM route is not expected to produce strong mucosal responses. Although mucosal immunity is considered important for immunity to SEE,3,5 an IM vaccine licensed in Europe and the UK has been effective for delaying the onset and reducing the severity of strangles in ponies.24 Nevertheless, intranasal administration of Ssee might improve mucosal immune responses.
The finding that horses with strangles did not have reactivity to Ssee was surprising. The reasons for this are unknown. Our laboratory (DJG, IHW, and CB) has demonstrated that S protein of group A Streptococcus (which has extensive homology with Ssee) binds RBC membranes, thereby evading immune responses.25 If Ssee acts similarly, it is possible that this masks immune responses. Numerous other mechanisms exist by which bacterial surface proteins can evade immune stimulation, such as molecular mimicry, secretion of effector proteins, and inhibition of phagocytosis by immune cells. Assuming our results are valid, further investigations are warranted to determine why Ssee is not immunogenic after natural infection. Nevertheless, weak antigens can be valuable targets for vaccine antigens.34,35
Although the sequence of Ssee was highly conserved across the 670 strains of SEE that we evaluated, the sequence in the strain we used to clone and express Ssee differed by 2 amino acids from the sequence found in 668 of the other SEE. Unfortunately, we only compared sequences after we had produced recombinant Ssee. In the future, it would be appropriate to use the more common sequence for protein synthesis and to compare immunogenicity between proteins based on the 2 different sequences; however, we do not expect that the small difference in sequence would have a large impact on antigens targeted by immune responses of vaccinated horses.
Our study has important limitations. Although no horses had a known history of exposure to SEE, including vaccines dating at least to 2017, lifelong medical histories were not available for all horses, and it is possible that some horses had previous exposure to SEE or prior vaccination with SEE prior to 2017. The finding that natural infection does not appear to stimulate immune responses to Ssee somewhat diminishes the importance of this limitation. Nevertheless, evaluating the vaccine in a population of horses naïve to SEE would be helpful to exclude the potential influence of humoral immune responses to vaccination with Ssee. We used in-house indirect ELISAs to test for serological evidence of recent exposure to SEE using plates coated either with lysate of whole SEE or recombinant SeM. The accuracy of these tests has not been reported; however, to our knowledge, a diagnostic accuracy study has not been reported for the ELISA test for SeM available in the US. Unfortunately, we cannot assure readers that all horses included in our study were immunologically naïve to SEE. We evaluated the function of anti-Ssee antibodies with 2 assays: complement deposition onto purified Ssee and OPK. The complement deposition assay allows us to demonstrate both specificity and functionality of the antibodies, whereas the OPK is a bactericidal assay. The reason that the results of complement deposition and OPK did not concur is unknown. One explanation is that while anti-Ssee IgG can deposit complement onto purified protein, it was ineffective for doing this on whole bacteria, thereby preventing OPK. Although resolving this discrepancy was beyond the resources and scope of our study, using either a serum bactericidal assay or complementing our assay for C1q deposition onto purified Ssee with an assay for deposition onto whole SEE might help; however, the latter method would be complicated by the ability of SEE to nonspecifically bind36 and degrade equine IgG.37 This point is important because if complement-mediated killing of SEE cannot be achieved by SEE-specific antibodies, it is possible that a vaccine targeting Ssee might not be effective. To our knowledge, however, the correlate(s) of protective immunity for SEE is unknown, and, ultimately, the clinical efficacy of a strangles vaccine will need to be established in experimentally and naturally infected horses. Our samples were stored prior to testing at −80 °C in cryovials as 1-mL aliquots for 29 weeks for study horses and for between 10 and 51 weeks for horses with strangles. We did not examine the extent to which the degradation of antibodies might have occurred in these samples. Thus, we cannot exclude the possibility of antibody degradation influencing our results. We consider this an unlikely explanation for the fact that anti-Ssee activity levels were lower in horses with strangles because freezing samples −80 °C for less than 1 year would not be expected to result in a dramatic reduction of antibody activity, and graphically (data not shown) we did not observe any pattern of association between the anti-Ssee antibody activity level and duration of freezing. Moreover, these samples were used for another project and showed strong reactivity against SEE.
To the authors’ knowledge, although a vaccine licensed in the EU and UK appears to have DIVA capacity,24 no vaccine licensed in the US has DIVA properties. Our finding that IgG activity against Ssee was not detected in the serum of horses with strangles indicates that a vaccine with Ssee as an antigen could distinguish vaccinates from naturally infected horses. Evidence that immunization against S protein can protect against infection with group A streptococcal infections suggests the possibility that immunization with Ssee could protect horses against strangles. Thus, our results are of clinical importance because they indicate that vaccination with Ssee is immunogenic and has DIVA properties; thus, evaluation of Ssee alone or in combination with other antigens as a strangles vaccine for horses is warranted.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
The authors acknowledge Erin Lester and Krissy Schroeder for providing access to horses owned by the Departments of Large Animal Clinical Sciences and Animal Sciences, respectively. The authors thank Dr. Marie-Eve Koziol from the Société d’exploitation pour les produits de l’industrie chimique for providing Montanide Gel 01 for use in this study.
The authors thank those veterinarians who generously provided serum from horses with strangles for this research project.
Disclosures
The University of California-San Diego has filed patents with Dr. David J. Gonzalez as inventor related to S protein work in Streptococcus equi (patent number: WO 2022/256310 A1).
No AI-assisted technologies were used in the generation of this manuscript.
Funding
Funding was provided by the Morris Animal Foundation (grant #D23EQ-014). Additional support was provided by the Link Equine Research Endowment, Texas A&M University. Dr. Cohen is supported by the Glenn Blodgett Chair in Equine Studies at the College of Veterinary Medicine and Biomedical Sciences, Texas A&M University.
ORCID
N. D. Cohen https://orcid.org/0000-0002-0384-2903
References
- 1.↑
Paillot R, Lopez-Alvarez R, Newton JR, Waller AS. Strangles: a modern clinical view from the 17th century. Equine Vet J. 2017;49(2):141–145. doi:10.1111/evj.12659
- 2.
Waller AS, Paillot R, Timoney JF. Streptococcus equi: a pathogen restricted to one host. J Med Microbiol. 2011;60(Pt 9):1231–1240. doi:10.1099/jmm.0.028233-0
- 3.↑
Sweeney CR, Timoney JF, Newton JR, Hines MT. Streptococcus equi infections in the horses: guidelines for treatment, control, and prevention of strangles. J Vet Intern Med. 2005;19(1):123–134.
- 4.
Slater JD. Strangles, bastard strangles, vives, and glanders: archeological relics in a genomic age. Equine Vet J. 2003;35(2):118–120. doi:10.2746/042516403776114252
- 5.↑
Boyle AG, Timoney JF, Newton JR, Hines MT, Waller AS, Buchanan BR. Streptococcus equi infections in horses: guidelines for treatment, control, and prevention of strangles-revised consensus statement. J Vet Intern Med. 2018;32(2):633–647.
- 6.↑
Ivens PA, Matthews D, Webb K, et al. Molecular characterization of ‘strangles’ outbreaks in the UK: the use of M-protein typing of Streptococcus equi ssp. equi. Equine Vet J. 2011;43(3):359–363. doi:10.1111/j.2042-3306.2010.00177.x
- 7.
Newton JR, Verhayen K, Talbot NC, et al. Control of strangles outbreaks by isolation of guttural pouch carriers identified using PCR and culture of Streptococcus equi. Equine Vet J. 2000;32(6):515–526. doi:10.2746/042516400777584721
- 8.↑
Mitchell C, Steward KF, Charbonneau ARL, et al. Globetrotting strangles: the unbridled national and international transmission of Streptococcus equi between horses. Microbial Genom. 2021;7(3):mgen000528. doi:10.1099/mgen.0.000528
- 9.↑
Waller A. Unsaddling Streptococcus equi infection of horses. UK Vet Equine 2022;6(2):61–67. doi:10.12968/ukve.2022.6.2.61
- 10.↑
Rendle D, de Brauwere N, Hallowell G, et al. Streptococcus equi infections: current best practice in the diagnosis and management of ‘strangles.’ UK Vet Equine. 2021;5(suppl 2):S3–S15. doi:10.12968/ukve.2021.5.2.S.3
- 11.
Sweeney CR, Whitlock RH, Meirs DA, Whitehead SC, Barningham SO. Complications associated with Streptococcus equi infection on a horse farm. J Am Vet Med Assoc. 1987;191(11):1446–1448.
- 12.
Pusterla N, Watson JL, Affolter VK, Magdesian KG, Carlson GP. Purpura haemorrhagica in 53 horses. Vet Rec. 2003;153(4):118–121. doi:10.1136/vr.153.4.118
- 13.
Delph KM, Beard LA, Trimble AC, Sutter ME, Timoney JF, Morrow JK. Strangles, convalescent Streptococcus equi subspecies equi M antibody titers, and presence of complications. J Vet Intern Med. 2019;33(1):275–279. doi:10.1111/jvim.15388
- 14.
Sponseller BT, Valberg SJ, Tennent-Brown BS, Foreman JH, Kumar P, Timoney JF. Severe acute rhabdomyolysis associated with Streptococcus equi infection in four horses. J Am Vet Med Assoc. 2005;227(11):1800–1807. doi:10.2460/javma.2005.227.1800
- 15.↑
Kaese HJ, Valberg SJ, Hayden DW, et al. Infarctive purpura hemorrhagica in five horses. J Am Vet Med Assoc. 2005;226(11):1893–1898. doi:10.2460/javma.2005.226.1893
- 16.↑
Morris ERA, Hillhouse AE, Konganti K, et al. Comparison of whole genome sequences of Streptococcus equi subsp. equi from an outbreak in Texas with isolates from within the region, Kentucky, USA, and other countries. Vet Microbiol. 2020;243:108638. doi:10.1016/j.vetmic.2020.108638
- 17.
Morris ERA, Wu J, Bordin AI, Lawhon SD, Cohen ND. Differences in the accessory genomes and methylomes of strains of Streptococcus equi subsp. equi and of Streptococcus equi subsp. zooepidemicus obtained from the respiratory tract of horses from Texas. Microbiol Spectr. 2022;10(1):e0076421. doi:10.1128/spectrum.00764-21
- 18.
Harris SR, Robinson C, Steward KF, et al. Genome specialization and decay of the strangles pathogen, Streptococcus equi, is driven by persistent infection. Genome Res. 2015;25(9):1360–1371. doi:10.1101/gr.189803.115
- 19.↑
Holden MTG, Heather Z, Paillot R, et al. Genomic evidence for the evolution of Streptococcus equi: host restriction, increased virulence, and genetic exchange with human pathogens. PLoS Pathog. 2009;5(3):e1000346. doi:10.1371/journal.ppat.1000346
- 20.↑
Weese JS, Jarlot C, Morley PS. Survival of Streptococcus equi in an outdoor environment. Can Vet J. 2009;50(9):968–970.
- 21.↑
Durham AE, Hall YS, Kulp L, Underwood C. A study of the environmental survival of Streptococcus equi subspecies equi. Equine Vet J. 2018;50(6):861–864.
- 22.↑
Borst LB, Patterson SK, Lanka S, Barger AM, Fredrickson RL, Maddox C. Evaluation of a commercially available modified-live Streptococcus equi subsp equi vaccine in ponies. Am J Vet Res. 2011;72(8):1130–1138. doi:10.2460/ajvr.72.8.1130
- 23.↑
Kemp-Symonds J, Kemble T, Waller A. Modified live Streptococcus equi (‘strangles’) vaccination followed by clinically adverse reactions associated with bacterial replication. Equine Vet J. 2007;39(3):284–286. doi:10.2746/042516407X195961
- 24.↑
Robinson C, Waller AS, Frykberg L, et al. Intramuscular vaccination with Strangvac is safe and induces protection against equine strangles caused by Streptococcus equi. Vaccine 2020;38(31):4861–4868. doi:10.1016/j.vaccine.2020.05.046
- 25.↑
Wierzbicki IH, Campeau A, Dehaini, et al. Group a streptococcal S protein utilizes red blood cells as immune camouflage and is a critical determinant for immune evasion. Cell Rep. 2019;29(10):2979–2989.e15. doi:10.1016/j.celrep.2019.11.001
- 26.↑
Campeau A, Uchiyama S, Sanchez C, Sauceda C, Nizet V, Gonazalez DJ. The S protein of group B Streptococcus is a critical virulence determinant that impacts the cell surface virulome. Front Microbiol. 2021;12:729308. doi:10.3389/fmicb.2021.729308
- 27.↑
Camacho C, Coulouris G, Avagyan V, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10:421. doi:10.1186/1471-2105-10-421
- 28.↑
Zhou L, Feng T, Xu S, et al. ggmsa: a visual exploration tool for multiple sequence alignment and associated data. Brief Bioinform. 2022;23(4):bbac222. doi:10.1093/bib/bbac222
- 29.↑
Cohen ND, Cywes-Bentley C, Kahn SM, et al. Vaccination of yearling horses against poly-N-acetyl glucosamine fails to protect against infection with Streptococcus equi subspecies equi. PLoS One. 2020;15(10):e0240479. doi:10.1371/journal.pone.0240479
- 30.↑
Cywes-Bentley C, Rocha JN, Bordin AI, et al. Antibody to poly-N-acetyl glucosamine provides protection against intracellular pathogens: mechanism of action and validation in horse foals challenged with Rhodococcus equi. PLoS Pathog. 2018;14(7):31007160. doi:10.1371/journal.ppat.1007160
- 31.↑
Bordin AI, Pillai SD, Brake C, et al. Immunogenicity of an electron beam inactivated Rhodococcus equi vaccine in neonatal foals. PLoS One. 2014;9(8):e105367. doi:10.1371/journal.pone.0105367
- 32.↑
Folmar CN, Cywes-Bentley C, Bordin AI, et al. In vitro evaluation of complement deposition and opsonophagocytic killing of Rhodococcus equi mediated by poly-n-acetyl glucosamine hyperimmune plasma compared to commercial plasma products. J Vet Intern Med. 2019;33(3):1493–1499. doi:10.1111/jvim.15511
- 33.↑
Overholser BR, Sowinski KM. Biostatistics primer: part 2. Nutr Clin Pract. 2008;23(1):76–84. doi:10.1177/011542650802300176
- 34.↑
Orr MT, Ireton GC, Beebe EA, et al. Immune subdominant antigens as vaccine candidates against Mycobacterium tuberculosis. J Immunol. 2014;193(6):2911–2918. doi:10.4049/jimmunol.1401103
- 35.↑
Lindahl G. Subdominance in antibody responses: implications for vaccine development. Microbiol Mol Biol Rev. 2020;85(1):e00078-20. doi:10.1128/MMBR.00078-20
- 36.↑
Meehan M, Lynagh Y, Woods C, Owen P. The fibrinogen-binding protein (FgBP) of Streptococcus equi subsp. equi additionally binds IgG and contributes to virulence in a mouse model. Microbiology (Reading). 2001;147(Pt 12):3311–3322. doi:10.1099/00221287-147-12-3311
- 37.↑
Lannergård J, Guss B. IdeE, an IgG-endopeptidase of Streptococcus equi ssp. equi. FEMS Microbiol Lett. 2006;262(2):230–235. doi:10.1111/j.1574-6968.2006.00404.x
