Antibodies Directed against Shiga-Toxin Producing Escherichia coli Serotype O103 Type III Secreted Proteins Block Adherence of Heterologous STEC Serotypes to HEp-2 Cells.

Shiga toxin-producing Escherichia coli (STEC) serotype O103 is a zoonotic pathogen that is capable of causing hemorrhagic colitis and hemolytic uremic syndrome (HUS) in humans. The main animal reservoir for STEC is ruminants and hence reducing the levels of this pathogen in cattle could ultimately lower the risk of STEC infection in humans. During the process of infection, STECO103 uses a Type III Secretion System (T3SS) to secrete effector proteins (T3SPs) that result in the formation of attaching and effacing (A/E) lesions. Vaccination of cattle with STEC serotype O157 T3SPs has previously been shown to be effective in reducing shedding of STECO157 in a serotype-specific manner. In this study, we tested the ability of rabbit polyclonal sera against individual STECO103 T3SPs to block adherence of the organism to HEp-2 cells. Our results demonstrate that pooled sera against EspA, EspB, EspF, NleA and Tir significantly lowered the adherence of STECO103 relative to pre-immune sera. Likewise, pooled anti-STECO103 sera were also able to block adherence by STECO157. Vaccination of mice with STECO103 recombinant proteins induced strong IgG antibody responses against EspA, EspB, NleA and Tir but not against EspF. However, the vaccine did not affect fecal shedding of STECO103 compared to the PBS vaccinated group over the duration of the experiment. Cross reactivity studies using sera against STECO103 recombinant proteins revealed a high degree of cross reactivity with STECO26 and STECO111 proteins implying that sera against STECO103 proteins could potentially provide neutralization of attachment to epithelial cells by heterologous STEC serotypes.

Introduction

Shiga toxin-producing Escherichia coli (STEC) is an enteric pathogen that causes diarroheal illness in humans which can lead to hemorrhagic colitis and haemolytic uremic syndrome (HUS), one of the main causes of renal failure in children [1]. Shiga toxins produced by this pathogen play an important role in causing these clinical manifestations. Currently, there is no treatment available for human STEC infections other than supportive care as the administration of antibiotics can exacerbate the disease. STEC O157:H7 is the predominant serotype associated with human infections in North America, while non-O157:H7 serotypes such as O103, O26, O111 are more prevalent in many European countries, South America and parts of Australia [1,2,3] The main reservoir for STEC is ruminants [4] and therefore intervention strategies aimed at lowering the levels of this pathogen in cattle could ultimately result in improved human health [5].

During the process of infection, STEC uses a Type Three Secretion System (T3SS) to inject virulence factors known as effector proteins directly into host cells, leading to the formation of attaching and effacing lesions (A/E) lesions, which are hallmarks of STEC infections. The major structural components of the STEC T3SS include EspA (filament), EspB and EspD (translocon complex) [6]. The STEC T3SS secretes over 50 effector proteins that are encoded on the LEE Pathogenicity Island or elsewhere on the chromosome (non-LEE effectors) [7]. The translocated intimin receptor, Tir, is an effector protein which enters host cells and forms a receptor that binds to intimin that is expressed on the surface of STEC cells [6]. Many studies have shown that the STEC T3SS is essential for colonization of cattle, implying that this is a major virulence factor employed by this pathogen [8,9,10].

Vaccination of cattle with STECO157 T3SP’s has shown to be an effective strategy in reducing the shedding of STECO157 [11,12,13,14,15,16,17,18,19]. However, this protection appears to be serotype specific [20,21]. Therefore, alternative antigens need to be identified that offer protection against non-O157 STEC serotypes. Recently, it has been shown that anti-sera to an extract of STECO157 T3SPs had the highest degree of cross-reactivity with STECO103 recombinant T3SPs [20], suggesting that STECO103 T3SPs may have cross-protective potential. In this study, we tested the effect of sera against STECO103 recombinant proteins on STECO103 and STECO157 adherence to HEp-2 cells. Moreover, we tested the vaccine potential of the recombinant proteins against STECO103 challenge in mice.

Materials and Methods

Bacterial strains and growth conditions

The bacterial strains used in this study comprised of E. coli N01-2454 (O103:H2), EDL933 (O157:H7), CL9 (O26:H11) and CL101 (O111:NM) [22,23]. For cloning and protein expression, we used the E. coli K-12 lab strains, JM109 (endA1, recA1, gyrA96, thi, hsdR17 (rk –, mk +), relA1, supE44, Δ (lac-proAB), [F´ traD36, proAB, laqIqZΔM15]) and BL21 (F-, dcm, ompT, hsdS (rB -,mB -), gal, λ(DE3)) obtained from Qiagen and Invitrogen, respectively. The strains were grown in Luria Bertani (LB) medium at 37°C in an orbital shaker (250 rpm), unless otherwise stated. E. coli serotypes O103 and O157 were transformed with a green fluorescent protein expressing plasmid, pNR78, for visualization by flouresence microscopy during the adherence inhibition assays as described [21]. Plasmid pNR78 was constructed in our lab by amplifying the GFP gene from pQBI-25 (Quantum Biotechnologies) which was cloned downstream of the GroEL promoter.

Protein expression and purification

The STEC serotype O103:H2 T3SS genes escC, espA, espB, espF, espG, espR1, nleA, nleE, nleF, nleG2, nleH, sepD, tccp2 and tir were amplified by PCR (Applied Biosystems) based on the sequence provided by GenBank®. Similarly, for cross reactivity studies, espA, espB, espF, nleA and tir from STEC serotypes O26 and O111 were amplified by PCR. The genes were cloned in either pQE-30 (Qiagen), pET-15b (Novagen), pGEX-5X-1 (GE Healthcare) or pGEX-5X-3, of which the first two are 6x His-tagged protein expression vectors while the latter are Glutathione S-transferase (GST)-fusion expression vectors. The constructs were confirmed by PCR and sequencing (Plant Biotechnology Institute, Saskatoon). Proteins were expressed in Escherichia coli K-12 lab strains (JM109 or BL21) and purified using either the method described in the QIAexpressionist™ manual (Qiagen) for His-tagged proteins or the GST Gene Fusion System Handbook (GE Healthcare) for the GST-fusion protein. Purified protein samples were greater than 90% pure as determined by SDS-PAGE followed by Coomassie blue staining as described previously [20].

Raising polyclonal anti-sera to STECO103 T3SS recombinant proteins

Purified recombinant proteins (100 μg each) were formulated with 30% Emulsigen D (MVP Laboratories) and two New Zealand White rabbits (Charles River) per STECO103 recombinant protein were immunized subcutaneously on day 0, followed by booster injections on days 21 and 42. The rabbits were euthanized on day 56 and sera were collected. Antibody titers against STECO103 recombinant proteins were confirmed using ELISA in duplicate wells as previously described [20]. For antibody titer determinations, the cut-off value was considered to be the average of the blank and two standard deviations. All rabbits used in this study were handled and treated in accordance with the guidelines provided by the Canadian Council on Animal Care (CCAC) as administered by the University Committee on Animal Care and Supply (UCACS), protocol 1994–213. This protocol was approved by the UCACS at the University of Saskatchewan for the present study.

Cell culture

HEp-2 cells (ATCC® CCL-23™, CEDARLANE®) were grown in HyClone Dulbecco modified Eagle medium (DMEM; Thermo Scientific) supplemented with 10% fetal bovine serum (FBS; PAA Laboratories) and 1% HEPES Buffer (Invitrogen) at 37°C in a 5% CO2 incubator. One day prior to the adherence inhibition assays, 105 cells per well were seeded onto eight well chamber slides (Nunc) and allowed to incubate overnight.

Adherence inhibition assays

Adherence of STECO103 and STECO157 to HEp-2 cells was assessed using an assay described elsewhere [21]. Briefly, an overnight culture of STEC grown in LB media was subcultured (1:100) into DMEM containing 10% FBS and 1% HEPES Buffer for 2 hours (until the OD600 was 0.2) at 37°C and 5% CO2 without shaking. For testing the effect of pooled sera against STECO103 T3SPs on adherence, HEp-2 cells were infected with 25 μl of STEC (1.7 x 106 colony forming units), 10 μl of each serum and 225 μl fresh DMEM. The effect of individual anti-serum was tested by infecting HEp-2 cells with 25 μl of STEC (1.7 x 106 CFU), 20 μl of anti-serum and 225 μl fresh DMEM (anti-O103 antibodies were prepared as described previously [21]). The chamber slides were incubated at 37°C and 5% CO2 for 3 hours (STECO157) or 3.5 hours (STECO103). The slides were washed six times with 200 μl Phosphate Buffered Saline (PBS, 0.1M pH 7.2) and fixed with 200 μl PBS containing 3.7% Formaldehyde. This was followed by two washes with PBS after which the slides were allowed to air dry. Coverslips were mounted with Vectashield® (Vector) containing DAPI and sealed. The slides were visualized under the fluorescent microscope (Axiovert 200 inverted microscope–Zeiss). Bacteria were observed under FITC, while HEp-2 cells were observed under DAPI. Each experimental group was first tested using 2 replicate wells in an 8 well chamber slide and 4 random grids were examined per well under the fluorescent microscope as described below. After observing clear differences in STEC adherence to HEp-2 cells between the different treatments, the experiments were repeated independently on a separate occasion using 8 replicates per test group as previously published [21] with 4 random grids per well used for enumerating the number of STEC per HEp-2 cell. The resulting pictures (4 under FITC and 4 under DAPI) per well (total of 8 pictures per well) were used to enumerate the number of STEC and HEp-2 cells per well. The total numbers of STEC per grid were then divided by the total numbers of HEp-2 cells per grid to determine the number of STEC per HEp-2 cell in one grid. This was repeated for the 8 duplicate wells per group, resulting in a total of 64 pictures per test group. Each data point in Figs 1 and 2 represent the average number of STEC per HEp-2 cell from 4 counts (4 random grids) per well. For statistical analysis, the median STEC per HEp-2 cell across the different test groups were compared using a non-parametric analysis as described below.

Fig 1

Inhibition of STECO103 adherence to HEp-2 cells by either (A) pooled sera against recombinant STECO103 T3SPs or (B and C) individual serum samples specific for STECO103 recombinant proteins EspB, EspF, EspG or NleA.

Anti-O103 refers to antibodies against a secreted fraction of T3SPs from STECO103. Values are expressed as median bacteria per cell from 8 replicates. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Fig 2

Inhibition of STECO157 adherence to HEp-2 cells by pooled sera against recombinant STECO103 T3SPs.

(A) STECO157 adherence to HEp-2 cells was significantly lower in the presence of antibodies against STECO103 EspA, EspB, EspF, NleA and Tir. (B) STECO157 adherence was partially lower in the presence of STECO103 anti-EspA and anti-Tir or anti-EspB, anti-EspF and anti-NleA. Values are expressed as median bacteria per cell from 8 replicates. **, P < 0.01; ****, P < 0.0001.

Immunization of mice with STECO103 T3SPs

Twenty four BALB/C mice were obtained from Charles River Canada. Mice were housed at the VIDO Animal Care Facility (University of Saskatchewan) and handled in accordance with the guidelines provided by the CCAC as administered by the UCACS, protocol 1998–0003. This protocol was approved by the UCACS at the University of Saskatchewan for the present study. Mice were randomly divided into two groups with 12 mice per group. The mice were immunized subcutaneously on Day 0 with 100 μl of either PBS (0.1M, pH 7.2) or a pool of STECO103 recombinant proteins EspA, EspB, EspF, NleA and Tir (1 μg of each protein) followed by a second immunization at Day 21. The vaccines were formulated with 30% Emulsigen D (MVP Laboratories). Sera were collected prior to both immunizations on day 0 and day 21 as well as prior to challenge with STECO103 on day 35. Antibody titers were determined using ELISA in duplicate wells as described previously [20]. For antibody titer determinations, the cut-off value was considered to be the average of the blank and two standard deviations. The mice were challenged as described below.

STEC mouse colonization model

For colonization of mice, we used the streptomycin-treated model as previously described [24,25]. Briefly, mice were given water containing Streptomycin Sulfate (5 g/L) on day 32 for two days. Subsequently, mice were deprived of food and water for 24 hours. On Day 35, mice were orally challenged with 100 μl of 109 cfu of STECO103 Nalr (resuspended in 20% sucrose). The mice were permitted access to food and water containing Streptomycin for the rest of the experiment. Fecal pellets were collected every 3 days for 21 days post challenge. Shedding of STECO103 was monitored by incubating the fecal samples in 1 ml LB broth for 2 hours at room temperature to allow the pellets to soften. The samples were vortexed, serially diluted in PBS and plated on MacConkey Sorbitol Agar containing Nalidixic Acid (15 μg/ml), Cefixime (5 μg/ml) and Potassium Tellurite (2.5 μg/ml). The plates were incubated overnight at 37°C and STEC colonies were enumerated the following day. Bacterial counts were expressed as cfu per gram of fecal content.

Cross-reactivity of STECO103 T3SS recombinant protein specific sera

Purified STEC026 and STEC0111 EspA, EspB, EspF, NleA and Tir recombinant proteins were separated by SDS-PAGE and transferred to a nitrocellulose membrane using a Mini Trans-Blot Electrophoretic Cell (Bio-Rad) as per the manufacturer’s instructions. The membranes were probed with either polyclonal sera (1:2500) from mice vaccinated with a pool of the corresponding STECO103 recombinant proteins or with rabbit polyclonal sera (1:2500) raised against STECO103 EspA, EspB, EspF, NleA and Tir. Alkaline phosphatase labeled goat anti-mouse or goat anti-rabbit IgG (KPL) antibodies were used as secondary antibodies (1:2000). The membranes were developed using 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitroblue tetrazolium (NBT) salt according to the manufacturer’s instructions (Sigma).

Statistical analyses

Statistical Analyses were performed using GraphPad Prism 6.02. Adherence inhibition assays were analyzed using a non-parametric analysis (Kruskal-Wallis test) and individual groups were tested using Dunn’s multiple comparison test. Mouse antibody titers were presented as medians plus/minus the 25th and 75th percentile ranges. Differences in immune responses between the vaccine and control groups were tested using non-parametric repeated measures ANOVA. A P value of < 0.05 was considered significant.

Results

Immune responses against STECO103 recombinant proteins in rabbits

Polyclonal sera were raised against 14 STECO103 recombinant proteins in rabbits in order to test the adherence inhibition effect of the sera in vitro and for cross reactivity studies. All the recombinant proteins induced a significant IgG specific antibody response as determined by ELISA (Table 1). The mean IgG titer across all the proteins was 911,801, while NleE had the lowest antibody titer (151,399) and EspR1 had the highest antibody titer (2,771,000).

Table 1

IgG antibody titers specific for STECO103 recombinant proteins in rabbits.

	Pre-immune	Day 56
Anti-EscC	87.5	241547
Anti-EspA	882.5	838090
Anti-EspB	783.5	1256000
Anti-EspF	711	618333
Anti-EspG	773.5	846100
Anti-EspR1	789.5	2771000
Anti-NleA	2778	1332000
Anti-NleE	555.5	151399
Anti-NleF	2748.5	1036000
Anti-NleG2	652	1358000
Anti-NleH	499.5	614552
Anti-SepD	536.5	483176
Anti-Tccp2	252.5	402873
Anti-Tir	122.5	816244

Antibody titers were determined by setting the cut-off value as the average of the blank and two standard deviations. All serum samples were tested in duplicate wells.

Antibodies against STECO103 T3SP’s inhibit adherence of STECO103

To test the effect of rabbit polyclonal sera against recombinant STECO103 T3SPs on adherence, we used a functional assay where we measured the level of STECO103 adherence to HEp-2 cells. Our results demonstrate that pooled sera against STECO103 recombinant proteins significantly reduced adherence of STECO103 to HEp-2 cells relative to the group incubated with pre-immune sera (Fig 1A). In order to determine which serum samples were involved in this adherence inhibition effect, we tested specific anti-sera to EspA, EspB, EspF, EspG, EscC, EspR1, NleA, NleE, NleF, NleG2, NleH, SepD, Tir and Tccp2 individually in duplicate. We observed that sera against EspB, EspG, EspF and NleA were involved in blocking adherence (data not shown). To confirm this observation, we performed an adherence inhibition assay where sera against EspB, EspG, EspF and NleA were tested individually with 8 replicates. The data clearly suggest that anti-sera to these four proteins were also highly effective in blocking STECO103 adherence to HEp-2 cells compared to the group treated with pre-immune serum (Fig 1B and 1C).

Anti- STECO103 T3SP sera have cross-protective potential

In order to determine if antibodies against STECO103 recombinant proteins can block adherence of other STEC serotypes, we evaluated the effect of pooled sera on STECO157 adherence to HEp-2 cells. Interestingly, our results indicate that incubation of STECO157 with anti-sera to STECO103 EspA, EspB, EspF, NleA and Tir significantly lowered adherence to HEp-2 cells, while anti-sera to the other proteins did not have a major effect (Fig 2A). We further investigated this adherence inhibition effect by testing pooled sera against STECO103 EspA and Tir in one group and sera against EspB, EspF and NleA in another group. Adherence of STECO157 to HEp-2 cells was lower in both groups relative to the control group (Fig 2B) but not to the same level as in the pooled group (Fig 2A). These results suggest that antibodies to STECO103 EspA, EspB, EspF, NleA and Tir proteins have a combined effect on blocking STECO157 adherence and that they have cross-protective potential.

Immunization of mice with T3SP’s from STECO103 induces a strong humoral response but does not affect fecal shedding

In order to test the protective capacity of STECO103 effectors, mice were vaccinated subcutaneously with a pool of recombinant proteins and were subsequently infected with STECO103 by oral challenge. Two weeks following the booster immunization, significant EspA-, EspB-, NleA- and Tir–specific IgG titers were detected in the sera relative to the control group (Table 2). In contrast, immunization with EspF elicited a weak IgG specific serum response. To assess the protective capacity of our vaccine, fecal shedding of STECO103 was monitored over 21 days post challenge. The levels of STECO103 were similar in both vaccinates and non-vaccinates throughout the duration of the study, suggesting that antibodies against the antigens used for immunization did not prevent STECO103 from persisting in the intestine (Fig 3), or that the response was not of sufficient magnitude.

Table 2

Median IgG antibody titers specific for STECO103 EspA, EspB, EspF, NleA or Tir in mice vaccinated with either PBS or a pool of STECO103 recombinant proteins.

Antibody	PBS				Vaccine				P
	Day 0		Day 35		Day 0		Day 35
Anti-EspA	693	(590–853)	1089	(736–1462)	772	(581–860)	6979	(3109–11618)	0.0007
Anti-EspB	235	(157–304)	600	(353–814)	243	(221–477)	294644	(195617–427747)	<0.0001
Anti-EspF	882	(742–1240)	1232	(973–5538)	1239	(753–1930)	3822	(1067–13092)	0.052
Anti-NleA	481	(161–839)	2734	(422–7809)	552	(250–4314)	1019000	(5364–1385000)	0.0004
Anti-Tir	570	(470–1154)	845	(711–3461)	517	(387–652)	280528	(151906–821988)	<0.0001

Values are expressed as median titers. Numbers in parentheses represent the 25th– 75th percentile. Antibody titers were determined by setting the cut-off value as the average of the blank and two standard deviations. All serum samples were tested in duplicate wells.

Fig 3

STECO103 shedding in feces following oral challenge in mice.

Mice were immunized subcutaneously with either PBS (control) or a pool of STECO103 EspA, EspB, EspF, NleA and Tir followed by a booster immunization three weeks later. Two weeks after the second immunization, mice were orally challenged with 109 cfu of STECO103. N = 12 for both groups. Values are expressed as median cfu per gram of feces.

STECO26 and STECO111 T3SS proteins display significant cross-reactivity with anti-sera to the corresponding STECO103 proteins

The cross-reactivity of sera against STECO103 T3SS proteins with other STEC serotypes including STECO26 and STECO111, was first tested by western blotting using rabbit polyclonal sera. Our results indicate that EspBO111, EspFO111 and NleAO111 reacted strongly with anti-sera to the corresponding STECO103 proteins, while TirO111 displayed a weaker reaction and EspAO111 did not react (Fig 4A). The western blot profile for STECO26 proteins was similar with respect to EspBO26 and EspFO26. However, EspAO26 also reacted strongly, unlike EspAO111, while NleAO26 did not react (Fig 4B). Subsequently, sera from mice immunized with a pool of STECO103 recombinant proteins was used to study the cross-reactivity with the equivalent STECO26 and STECO111 proteins. The results indicate that EspBO26 reacted strongly with the anti-sera while EspAO26, EspFO26 and NleAO26 did not (Fig 5A). In contrast, the western blot profile for the STECO111 recombinant proteins showed a significant degree of cross reactivity for EspBO111, NleAO111 and TirO111 (Fig 5B). Taken together, the results suggest that EspBO103, NleAO103, and TirO103 possess significant cross-reactive properties with the corresponding proteins from STECO26 and STECO111. Hence, these proteins may form the basis of a cross-protective vaccine that confers protection against multiple STEC serotypes.

Fig 4

Cross-reactivity of STECO103 EspA, EspB, EspF, NleA and Tir specific rabbit polyclonal sera with the corresponding STECO26 and STECO111 recombinant proteins.

(A) Western blot using anti-EspA. Lane 1, marker; Lane 2, EspAO103 (20.5 kDa); Lane 3, EspAO26 (20.5 kDa); Lane 4, EspAO111(20.5 kDa). (B) Western blot using anti-EspB. Lane 1, marker; Lane 2, EspBO103 (33.1 kDa); Lane 3, EspBO26 (33.2 kDa); Lane 4, EspBO111 (32.8 kDa). (C) Western blot using anti-EspF. Lane 1, marker; Lane 2, EspFO103 (57 kDa); Lane 3, EspFO26 (40 kDa); Lane 4, EspFO111 (27 kDa). (D) Western blot using anti-NleA. Lane 1, marker; Lane 2, NleA O103 (44 kDa); Lane 3, NleA O26 (11.7 kDa); Lane 4, NleA O111 (47.5 kDa). (E) Western blot using anti- Tir. Lane 1, marker; Lane 2, Tir O103 (56 kDa); Lane 3, Tir O111 (56.9 kDa).

Fig 5

Cross-reactivity of STECO103 EspA, EspB, EspF, NleA and Tir specific mouse polyclonal sera with the corresponding STECO26 and STECO111 recombinant proteins.

Western blot using sera from mice immunized with a combination of STECO103 EspA, EspB, EspF, NleA and Tir to test the cross-reactivity with: (A) STECO26 proteins. Lane 1, marker; Lane 2, EspAO26 (20.5 kDa); Lane 3, EspBO26 (33.2 kDa); Lane 4, EspFO26 (40 kDa); and Lane 5, NleAO26 (11.7 kDa). (B) STECO111 proteins. Lane 1, marker; Lane 2, EspAO111 (20.5 kDa); Lane 3, EspBO111 (32.8 kDa); Lane 4, EspF O111 (27 kDa); Lane 5, NleA O111 (47.5 kDa); and Lane 6, TirO111 (56.9 kDa).

Discussion

Many efforts have been made to develop STECO157 vaccines using the T3SS proteins as targets in order to reduce the levels of the pathogen in cattle [5,12]. However, these vaccination strategies provide limited protection as they are directed only against STECO157 and they are limited in their benefit [20,21]. Non-O157 STEC serotypes are more prevalent in other parts of the world [1] and with the rise in non-O157 STEC infections in humans [26] as well as the increase in the prevalence of these serotypes in cattle [27], a vaccine that can confer protection against multiple serotypes is more desirable. The aim of this study was to determine if STECO103 T3SS proteins could provide protection against STECO103 as well as other heterologous serotypes using adherence-inhibition assays and the streptomycin-treated mouse model.

We used STECO103 T3SS proteins as targets for a potentially cross protective vaccine since the T3SS proteins encoded by this serotype have previously been shown to have the highest degree of cross reactivity with STECO157, relative to STECO26 and STECO111 [20,21]. Based on this, we over-expressed and purified STECO103 EscC, EspA, EspB, EspF, EspG, EspR1, NleA, NleE, NleF, NleG2, NleH, SepD, Tccp2 and Tir recombinant proteins. In order to test the protective capacity of these proteins, we first examined the effect of rabbit polyclonal sera against the candidate proteins in vitro using a HEp-2 cell adherence inhibition assay which has been successfully used as a functional assay to study STEC adherence [21,28]. This, in turn, may reflect the effect of antibodies on intestinal colonization. Our results demonstrate that pooled sera against STECO103 recombinant proteins significantly inhibited STECO103 adherence. This is in agreement with the findings of Asper et al, where anti-sera to all STECO103 secreted proteins blocked adherence by this serotype [21]. We also show for the first time that sera against individual STECO103 recombinant proteins including, EspB, EspF, EspG and NleA, inhibited adherence of the bacteria to HEp-2 cells. Interestingly, pooled anti-sera to STECO103 recombinant proteins EspA, EspB, EspF, NleA and Tir were able to block adherence of STECO157, suggesting that these candidate proteins may provide protection against multiple STEC serotypes. However, it appears that the inhibition of STECO157, unlike that of STECO103, was due to a combination of STECO103 anti-sera since there was reduced inhibition of STECO157 once the pooled anti-sera were divided into two groups. Taken together, this is the first report which illustrates that sera against STECO103 T3SS recombinant proteins can block adherence of STECO157 to HEp-2 cells.

The streptomycin-treated mouse model [24,29] was used to test the efficacy of the identified candidate STECO103 recombinant proteins as antigens for protection against STECO103. This model was chosen since it has been widely used by various groups to test their STEC vaccines prior to conducting studies in cattle [25,30,31,32,33]. The mice developed strong serum IgG specific titers against EspAO103, EspBO103, NleAO103 and TirO103 following immunization, while the response to the corresponding EspF recombinant protein was weak. The weak response to EspFO103 is in line with what was observed for EspFO157 in a previous study published by our group [20]. Immunization with STECO103 recombinant proteins did not affect STECO103 fecal shedding over the duration of the experiment relative to the control group. This was unexpected since similar STECO157 based vaccines have been highly effective in mice [30,33]. In addition, a recent vaccination study by our group illustrated that a combination of nine STECO157 recombinant proteins was highly effective in controlling STECO157 fecal shedding in mice (data not shown). It is possible that our STECO103 vaccine may have been more effective against intestinal colonization had it been administered intranasaly. However, both subcutaneous and intranasal immunization of mice with an extract of STECO157 secreted proteins as well as individual recombinant proteins have proven to be highly effective in controlling STECO157 shedding [30]. In addition, the lack of a robust immune response against EspA may have contributed to the persistence of STECO103 in the intestines. Alternatively, since very little work has been done on STECO103 in mice, we speculate that the T3SS may play a different role in STECO103 infection in mice. Therefore, further analysis of the STECO103 T3SS may be required in mice, while a similar vaccine study should be performed in cattle with STECO103.

The serological cross reactivity of STECO103 recombinant proteins EspA, EspB, EspF, NleA and Tir with the corresponding STECO26 and STECO111 proteins was analyzed by western bloting. Overall our results indicate that there was significant cross reactivity of the STECO26 recombinant proteins, EspAO26, EspBO26 and EspFO26 when rabbit polyclonal sera were used. These observations are supported by the protein sequence homology of the STECO26 proteins to STECO103: EspAO26 (92%), EspBO26 (99%) and EspFO26 (91%). The fact that NleAO26 did not cross react was not unexpected since the STECO26 genome contains an NleA-like gene which encodes for an 11 kDa protein, while the actual size of NleAO103 is 44 kDa. Therefore, sequence homology between NleAO103 and NleAO26 is expected to be low (58%) with few shared epitopes, if any. We did not show the results for TirO26 since we were unable to express or purify this protein despite numerous attempts. This may be explained by the fact that TirO26 may require co-expression and co-purification with a chaperone [34]. The western blot profile for STECO111 recombinant proteins EspBO111, EspFO111, NleAO111 showed a high degree of cross reactivity with the corresponding sera, while there was lower cross reactivity with TirO111. This is consistent with the observed sequence homologies between the STECO103 and STECO111 proteins: EspBO111 (71%), EspFO111 (70%), NleA (83%) and Tir (65%). The fact that EspAO111 did not react to sera against EspAO103 was surprising since EspAO111 shares greater than 81% sequence homology to EspAO103. The serological cross reactivity of the STEC serotypes O26 and O111 recombinant proteins was remarkably lower when mouse polyclonal sera were used. The difference in the results may be due to differences in recognition of epitopes by the mouse and rabbit immune systems. Overall, the data from both cross reactivity studies suggests that EspBO103, NleAO103 and TirO103 are highly cross reactive and have the potential to form an efficacious recombinant vaccine that protects cattle not only against STECO103 but other STEC serotypes as well. This finding is supported by two recent studies which demonstrate that EspBO157 and TirO157 are immunogenic and protective in cattle against STECO157 [35,36]. Although these studies provide important information about STECO157, this can be used as a basis for conducting similar studies with STECO103 to test for cross serotype protection.

Vaccination with a commercially available STECO157 T3SS vaccine (Econiche™) is an effective strategy to control STEC shedding in cattle [37]. Many recent studies have proven that this preslaughter intervention does lead to reduced levels of this pathogen in cattle [38,39]. Moreover, Mathews et al have recently predicted that vaccination of cattle against STECO157 will have a significant impact on public health by lowering human STEC infections by 85% [40]. Our in vitro results are the first steps towards a vaccine that may provide protection against multiple STEC serotypes, which is highly desirable for both North America as well as other regions where non-O157 STEC serotypes are more prevalent. The STECO157 SRP® vaccine (contains siderophore and porin proteins) has also shown to be effective in reducing fecal shedding in cattle [41,42]. However, this vaccine also confers limited serotype protection like the Econiche™ vaccine [43,44]. Taken together, the need for an STEC vaccine that provides protection against more than one serotype is required and our in vitro results suggest that STECO103 may be a likely candidate, though further testing is required in cattle.