Recombinant production platform for Group A Streptococcus glycoconjugate vaccines

Abstract

Group A Streptococcus (Strep A) is a human-exclusive bacterial pathogen killing annually more than 500,000 patients, and no current licensed vaccine exists. Strep A bacteria are highly diverse, but all produce an essential, abundant, and conserved surface carbohydrate, the Group A Carbohydrate, which contains a rhamnose polysaccharide (RhaPS) backbone. RhaPS is a validated universal vaccine candidate in a glycoconjugate prepared by chemical conjugation of the native carbohydrate to a carrier protein. We engineered the Group A Carbohydrate biosynthesis pathway to enable recombinant production using the industry standard route to couple RhaPS to selected carrier proteins within Escherichia coli cells. The structural integrity of the produced recombinant glycoconjugate vaccines was confirmed by Nuclear Magnetic Resonance (NMR) spectroscopy and mass spectrometry. Purified RhaPS glycoconjugates elicited carbohydrate-specific antibodies in mice and rabbits and bound to the surface of multiple Strep A strains of diverse M-types, confirming the recombinantly produced RhaPS glycoconjugates as valuable vaccine candidates.

Fig. 1. Schematic showing GAC biosynthesis pathway (left) and the revised GAC biosynthesis pathway (right) that synthesises OTase compatible reducing end sugars.

The GAC pathway synthesises two major cell wall components, (i) the undecorated immunogenic repeating unit (red box) and (ii) the final GAC. Our engineered pathway includes the genes wbbPRQ from the Shigella dysenteriae serotype 1 O-antigen pathway. These glycosyltransferases introduce a different end group, α-l-Rhap-(1→3)-α-l-Rhap-(1→2)-α-d-Galp-(1→3)-d-GlcpNAc that can be extended by GacCFG to produce the polyrhamnose backbone containing the immunogenic →2-α-l-Rhap-(1→3)-α-l-Rhap-1→ repeating unit (red box). Once the lipid linked product has been translocated across the cell membrane via GacDE (ABC-transporter), the OTase PglB recognises the reducing end group disaccharides, cleaves the glycan off the lipid carrier and transfers it onto the OTase sequon on the acceptor protein of choice. This completes the synthesis of the first recombinant Strep A glycoconjugate vaccine candidates. Schematic created in Adobe Illustrator.

Fig. 2. Western blot and spot blot analysis investigating minimal gene cluster requirements for recombinant RhaPS production in E. coli.

A E. coli rfaS deletion cells transformed with two plasmids containing either the Shi1/2a gene cluster fragment, GacB control or empty plasmid (negative control) as well as the RhaPS clusters (ΔB) were grown overnight and whole cell lysates subjected to SDS-PAGE and western blotting with the commercially available anti-GAC antibodies. Plasmids indicated: ΔB = gacACDEFG; B = gacB; Shi1 = wbbPQR, Shi2a = rfbFG, and empty plasmid controls=-. B E. coli rfaS deletion cells transformed with combinations of RhaPS cluster, and Shi1 gene cluster fragments were grown overnight and subjected to dot blot analysis using the GAC antibodies. RhaPS is synthesised by the WbbPQR proteins when co-expressed with the ΔB RhaPS proteins. Omitting one of the Shi1 proteins results in weaker or no RhaPS production.

Fig. 3. Western blot analysis of recombinant glycoconjugate production in E. coli under different media conditions.

E. coli cells (ΔwaaL with chromosomally integrated OTase pglB gene) were transformed with A the Shi1 gene cluster, the RhaPS cluster (ΔB) and NanA carrier protein and B Shi1, ΔB and ExoA-10 tag carrier protein plasmids. Cells were tested for glycoconjugate production in different growth conditions. Total cell lysates were run over SDS-PAGE and analysed via His and GAC-antibodies (red and green, respectively). Glycoconjugate production levels vary between the media conditions. Un uninduced, -pglB cells lack the chromosomal pglB. Arrows indicate glycosylated protein bands.

Fig. 4. Biochemical analysis of glycoconjugate product using chromatography, NMR and mass-spectrometry.

A NTA column purified NanA-RhaPS was purified via two rounds of size exclusion chromatography to isolate the higher molecular weight glycosylated NanA-RhaPS species. B Fractions of the second round were analysed via SDS-PAGE and reveal sequential elution of NanA-RhaPS glycoconjugate and free-protein species. The fractions boxed in blue were pooled for immunisation studies. C NMR analysis confirms that the glycoconjugate vaccine candidates carry the native backbone repeating unit with the structure [→2-α-l-Rhap-(1→3)-α-l-Rhap-1→]_n. Selected region of overlaid ¹H,¹³C-HSQC (black) and ¹H,¹³C-HMBC (red) NMR spectra of the engineered glycoconjugate. The repeating unit of the rhamnan polysaccharide structure is given in SNFG format. D HCD spectra of a tryptic peptide derived from NanA protein. The monoisotopic mass of the precursor ion [M+3H]³⁺ m/z 1428.63096 corresponds to the peptide AQGGDQNATGGEQPLANETQLSGESSTLTDTEK with Δmass 949 Da corresponding to HexNAc-Hex-(dHex)₄. The lists of the b and y ions assigned from MS/MS spectra are shown in this figure. Summary glycosylation features indicate observed site-specific glycosylation for NanA and IdeS. We observed that the glycosylation site close to N-terminal for both proteins could carry up to 36 Rha residues. For the protein NanA, the second glycosylation site was mostly detected, carrying 30 to 40 Rha residues. The C-terminal glycosylation site of IdeS protein exhibited diverse glycan heterogeneity carrying 9 to 41 Rha residues.

Fig. 5. Assessment of antibody titers and cytokine levels in mice post-Immunisation studies.

A, B Mouse immunisation study (n = 10) (70 days). Mouse sera were investigated for IgG antibodies triggered post immunisation with the low RhaPS-containing NanA-RhaPS, NanA alone and with low RhaPS containing IdeS-RhaPS. Evaluation of vaccine candidate specific antibodies in mice sera using an ELISA based assay with respective glycoconjugate vaccine candidates showed that IgG levels are elevated in all immunised mice. Y-axis: 450 nm. Statistical analysis was performed using ANOVA followed by a Dunn’s post hoc test. *p < 0.01; **P < 0.001; ****P < 0.0001; ns- not significant. C Pooled mice sera (n = 10) from IdeS-RhaPS immunisation studies were tested in an IdeS neutralisation assay. A series of dilutions from 1:6 to 1:180 was tested in triplicate and the neutralisation activity is displayed in % compared with a no-sera control. * indicates duplicate data obtained for D35. Dashed line, positive control sera; solid line, negative control sera. D, E Immunised mice splenocytes were investigated for IL-17A and IFN-γ levels after restimulation with the mitogens NanA-RhaPS. Statistical analysis was performed using ANOVA followed by a Bonferroni’s post hoc test. *P < 0.05; **P < 0.001; ****P < 0.0001.

Fig. 6. Representative histograms for antibody deposition on Strep A M serotypes and S. dysgalactiae subsp. equisimilis (SDSE) strain expressing GAC cluster (Stg485 and Stg652).

All strains were stained in rabbit antiserum (1:1000) for except M2 (1:100) from pre-immune (red shading) and post-immune (NanA-RhaPS; blue shading) vaccinated groups. The number on the top shows the geometric mean fluorescence intensity (gMFI) of the pre-immune sera (red) and NanA-RhaPS immunised serum (blue) for the displayed histograms.