FUSION PROTEIN COMPRISING STREPTOCOCCAL ANTIGEN

Abstract

The disclosure provides a fusion protein comprising at least one antigenic fragment of a protein from a bacterium from genus Streptococcus, as well as means for its expression. Outer membrane vesicles and vaccines comprising the fusion protein are also disclosed, as well as a method of vaccination using such vaccines.

Claims

1. A fusion protein, comprising: i. a passenger domain comprising a beta stem domain from an autotransporter protein, wherein the beta stem forming sequence of the passenger domain is essentially intact; ii. a translocator domain from an autotransporter protein; iii. a signal peptide that targets the fusion protein to the inner membrane of a Gram negative bacterium; and iv. at least one antigenic fragment; wherein the passenger domain of the autotransporter in its native form comprises at least one side domain, and wherein said antigenic fragment replaces or partly replaces said side domain; and wherein said at least one antigenic fragment is a fragment of a protein from a bacterium of the genus Streptococcus.

2. The fusion protein according to claim 1, wherein said at least one antigenic fragment is a fragment of a protein from Streptococcus pneumoniae.

3. The fusion protein according to claim 2, wherein said at least one antigenic fragment is a fragment of pneumococcal surface protein A from Streptococcus pneumoniae.

4. The fusion protein according to claim 3, wherein said at least one antigenic fragment is a fragment of the α-helical coiled-coil domain of pneumococcal surface protein A from Streptococcus pneumoniae.

5. The fusion protein according to claim 4, wherein said α-helical coiled-coil domain comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 9 and sequences having at least 30% identity thereto.

6. The fusion protein according to claim 1, wherein said at least one antigenic fragment consists of 20-250 amino acids.

7. The fusion protein according to claim 1, comprising at least a first antigenic fragment α1 and a second antigenic fragment α2 which are different, overlapping or non-overlapping, fragments of said α-helical coiled-coil domain of pneumococcal surface protein A from Streptococcus pneumoniae, wherein the passenger domain of the autotransporter in its native form comprises at least two side domains, and wherein each of said antigenic fragments α1 and α2 replaces or partly replaces a separate side domain.

8. The fusion protein according to claim 7, wherein the amino acid sequence of α1 consists of 100-150 amino acid residues, for example from 120-140 amino acid residues, for example 125-135 amino acid residues.

9. The fusion protein according to claim 7, wherein the amino acid sequence of α1 comprises a sequence selected from the group consisting of SEQ ID NO: 10 and sequences having at least 30% identity thereto.

10. The fusion protein according to claim 9, wherein the amino acid sequence of α1 comprises SEQ ID NO: 10.

11. The fusion protein according to claim 9, wherein the amino acid sequence of α1 consists of a sequence selected from the group consisting of SEQ ID NO: 10 and sequences having at least 30% identity thereto.

12. The fusion protein according to claim 11, wherein the amino acid sequence of α1 consists of SEQ ID NO: 10.

13. The fusion protein according to claim 7, wherein the amino acid sequence of α2 consists of 60-110 amino acid residues.

14. The fusion protein according to claim 7, wherein the amino acid sequence of α2 comprises a sequence selected from the group consisting of SEQ ID NO: 11 and sequences having at least 30% identity thereto.

15. The fusion protein according to claim 14, wherein the amino acid sequence of α2 comprises SEQ ID NO: 11.

16. The fusion protein according to claim 14, wherein the amino acid sequence of α2 consists of a sequence selected from the group consisting of SEQ ID NO: 11 and sequences having at least 30% identity thereto.

17. The fusion protein according to claim 16, wherein the amino acid sequence of α2 consists of SEQ ID NO: 11.

18. The fusion protein according to claim 1, wherein the passenger domain (i) and the translocator domain (ii) are derived from a serine protease autotransporter of Enterobacteriaceae (SPATE) protein.

19. The fusion protein according to claim 18, wherein the SPATE protein is selected from the group consisting of hemoglobin-binding protease (Hbp), extracellular serine protease (EspC) and temperature-sensitive hemagglutinin (Tsh) from Escherichia coli.

20. The fusion protein according to claim 19, wherein the SPATE protein comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 2 and homologous sequences having at least 35% identity thereto.

21. The fusion protein according to claim 20, wherein said passenger domain in its native form comprises five side domains, which are defined by amino acids 54-308, 533-608, 657-697, 735-766 and 898-922 of SEQ ID NO: 1 or SEQ ID NO: 2.

22. The fusion protein according to claim 21, wherein one of said α1 and α2, when present, is inserted into, replaces or partly replaces the side domain defined by amino acids 54-308, and the other one of said α1 and α2, when present, is inserted into, replaces or partly replaces the side domain defined by amino acids 533-608.

23. The fusion protein according to claim 1, which does not comprise a cleavage site, or comprises a disrupted cleavage site, such that the fusion protein is displayed on the surface of a cell in which it is expressed.

24. A polynucleotide encoding a fusion protein according to claim 1.

25. An expression vector comprising a polynucleotide according to claim 24.

26. A gram-negative bacterial host cell comprising an expression vector according to claim 25.

27. The host cell according to claim 26, which belongs to the family Enterobacteriaceae.

28. The host cell according to claim 27, which belongs to the species Salmonella enterica.

29. The host cell according to claim 28, which is a ΔtoIRA derivative of S. Typhimurium strain SL3261.

30. A method of expressing a fusion protein comprising the steps of i. providing a host cell according to claim 26; and ii. culturing said host cell under conditions suitable for expression of said fusion protein.

31. A method of producing outer membrane vesicles displaying a fusion protein on their surface, comprising: expressing a fusion protein using the method according to claim 30; and shedding of vesicles from the outer membrane of the host cell to obtain outer membrane vesicles displaying the fusion protein on their surface.

32. An outer membrane vesicle, displaying at least one fusion protein according to claim 1 on its surface.

33. A vaccine comprising an outer membrane vesicle according to claim 32.

34. A method of inducing protective immunity against Streptococcus, comprising the step of administering a vaccine according to claim 33 to a subject in need thereof.

35. The method according to claim 34, wherein said protective immunity comprises protection against streptococcal colonization.

36. The method according to claim 34, wherein said protective immunity is characterized by a high expression of IL-17A.

37. The method according to claim 36, wherein said expression of IL-17A is localized in nasopharyngeal tissue.

38. The method according to claim 34, wherein said vaccine is mucosally administered.

39. The method according to claim 38, wherein said vaccine is intranasally administered.

40. The method according to claim 34, which induces protective immunity against a streptococcal disease.

41. The method according to claim 34, wherein said protective immunity comprises protection against colonization by Streptococcus pneumoniae.

42. The method according to claim 41, which induces protective immunity against a pneumococcal disease selected from the group consisting of pneumonia, meningitis, otitis media, bacteremia, sepsis and acute exacerbations of chronic bronchitis, sinusitis, arthritis and conjunctivitis.

43. The fusion protein according to claim 7, wherein the amino acid sequence of α2 consists of 70-100 amino acid residues.

44. The fusion protein according to claim 7, wherein the amino acid sequence of α2 consists of 75-95 amino acid residues.

45. The host cell according to claim 27, which is selected from the group consisting of Escherichia coli, Salmonella spp., Vibrio spp., Shigella spp., Pseudomonas spp., Burkholderia spp. and Bordetella spp.

46. The host cell according to claim 45, which belongs to the subspecies Salmonella enterica subsp enterica.

47. The host cell according to claim 46, which belongs to serovar Typhimurium.

48. The host cell according to claim 47, which is strain SL3261.

49. The method according to claim 40, wherein the streptococcal disease is selected from the group consisting of pneumonia, endocarditis, meningitis, otitis media, bacteremia, sepsis, pharyngitis, respiratory infections, dental caries and acute exacerbations of chronic bronchitis, sinusitis, arthritis and conjunctivitis.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0091] FIG. 1 is a schematic illustration of the vaccine design described in the Example. A: schematic representations of PspA and pneumolysin. The slightly overlapping fragments that were selected for fusion to HbpD cover the α-helical coiled coil domain (α1 and α2), the lactoferrin-binding domain (LFBD) and the Pro-rich region (PRR) but not the cell wall-anchoring choline-binding domain (CBD) of PspA, and the entire Ply sequence. Numbers above the diagrams and boundaries next to the fragments correspond to aa positions calculated from the N termini of full-length PspA and Ply. B: Wild-type Hbp (included for reference) comprises an N-terminal cleavable signal sequence (ss), a secreted passenger domain, and a linker (grey) and a C-terminal domain that inserts into the OM as a β-barrel, which together facilitate translocation of the passenger. Side domains d1-d5, which are dispensable for secretion and can be replaced by heterologous polypeptides, are indicated, while the remainder of the passenger domain is black. Point mutations D1100G and D1101G (denoted X) prevent autocatalytic cleavage after translocation across the OM, creating a surface-exposed (display; D) version of Hbp (Jong et al (2007) Mol Microbiol 63(5):1524-36). Numbers above the diagrams correspond to the aa positions of the wt Hbp precursor, calculated from the N terminus. Insertion of the pneumococcal PspA fragments, defined in A, are highlighted. Each insert is flanked by short flexible Gly/Ser linkers (FL).

[0092] FIG. 2 are gel photographs showing the result of vaccine production. A: SDS-PAGE/Coomassie analysis of the vaccine stocks used. Equivalent volumes of vaccine stock containing equivalent amounts of OMVs, based on OD units of the original cultures, were isolated from S. Typhimurium SL3261 ΔtoIRA expressing HbpD or the indicated HbpD-antigen chimera and loaded. All constructs were expressed in the presence of 100 μM IPTG (unlabeled or H; high), except for HbpD-PspA[α1-α2] L (low), which was expressed in the presence of 1 μM IPTG, as determined by densitometric analysis. Below the panel, relative amounts of Hbp chimeras and OmpA in OMV samples are displayed as determined by densitometric analysis. Highest measured densities were put at 100%. B-C: Equal amounts of intact (−tx) and Triton X-100-permeabilized (+tx) OMVs described in A were incubated with Proteinase K (+pk) or mock treated (−pk), separated by SDS-PAGE and detected with B: immunoblotting using indicated antibodies, and C: Coomassie staining. An amount of 0.5 OD units of OMV material was loaded in each lane. The Hbp (chimeras) are marked by asterisks, and the OMV marker proteins OmpF/C and OmpA, and Proteinase K are indicated with arrowheads.

[0093] FIG. 3 demonstrates that OMV/Hbp platform-induced protection is influenced by choice of antigen fragment, antigen amount, and immunization number. Bacterial recovery of S. pneumoniae from nasal tissue three days post intranasal challenge of C57BL/6 mice that received A: three intranasal immunizations with OMVs displaying HbpD, HbpD-PspA[α1-α2], HbpD-PspA[LFBD-PRR] or HbpD-Ply[F1-F3], or B: three, two or one intranasal immunization(s) (3×; 2×; 1×) with OMVs expressing HbpD or high (H) or 7-fold lower (L) levels of HbpD-PspA[α1-α2]. Symbols indicate individual mice (n=5-15 per group), bars represent group mean and the dotted line indicates the lower limit of detection. *, p<0.05; **, p<0.01; ***, p<0.001.

[0094] FIG. 4 shows that protective immunity correlates with intranasal IL-17A levels. A-B: Nasopharyngeal A: IFNγ and B: IL-17A three days post-infection in mice immunized three times with OMVs expressing HbpD or HbpD-PspA[α1-α2]. C: Nasopharyngeal IL-17A three days post-infection in mice immunized three, two or one times (3×; 2×; 1×) with OMVs expressing HbpD or high (H) or 7-fold lower (L) levels of HbpD-PspA[α1-α2]. A-C, *, p<0.05; **, p<0.01; ***, p<0.001. D, Pooled data from study 1 and 2 for nasopharyngeal IL-17A and number of CFU three days post-infection of mice. Symbols represent individual mice (n=5-15 per group) immunized three times with OMV-HbpD (filled symbols) or with OMV-HbpD-PspA[α1-α2] (open symbols), and bars represent group mean. Spearman's correlation coefficient (ρ) and p-value are indicated.

[0095] FIG. 5 shows that intranasal antigen-specific IgG is influenced by antigen amount. Nasal antigen-specific IgG responses directed against the whole protein, i.e. PspA and pneumolysin, in mice immunized A: three times with OMVs displaying the indicated HbpD (chimera), or B: three, two or one times (3×; 2×; 1×) with OMVs displaying HbpD, or HbpD-PspA[α1-α2] with varying antigen amount (H; high or L; 7-fold lower). Symbols represent individual mice, and bars represent mean of groups of 5-15 mice. *, p<0.05; **, p<0.01; ***, p<0.001.

[0096] FIG. 6 illustrates bacterial recovery upon high expression of PRR. Bacterial recovery of S. pneumoniae from nasal tissue three days post intranasal challenge of C57BL/6 mice that received three (3×) intranasal immunizations of OMVs displaying HbpD or high (H) levels of antigen fragment PRR. Each symbol represents an individual mouse, and bars represent mean per group. NB: The control samples (HbpD) are the same as shown in FIG. 2B.

[0097] FIG. 7 shows systemic antigen-specific IgG levels. Systemic antigen-specific IgG responses directed against PspA or pneumolysin, respectively, in mice immunized with OMVs displaying A: indicated Hbp (derivatives) or B: HbpD or HbpD-PspA[α1-α2] at high (H) or 5-fold lower (L) expression levels, and with varying immunization number (3×; 2×; 1×). Bars represent mean per group, and symbols represent samples from individual mice. *, p<0.05; **, p<0.01; *** p<0.001.

[0098] FIG. 8 shows nasal antigen-specific IgA levels. Nasal antigen-specific IgA responses directed against PspA or pneumolysin, respectively, in mice immunized with OMVs displaying, A: indicated Hbp (derivatives) or B: HbpD or HbpD-PspA[α1α2] at high (H) or 5-fold lower (L) expression levels, and with varying immunization number (3×; 2×; 1×). Bars represent mean per group, and symbols represent samples from individual mice. *, p<0.05; **<0.01; *** p<0.001.

EXAMPLE

Summary

[0099] This Example demonstrates the feasibility of using the autotransporter Hbp platform, designed to efficiently and simultaneously display multiple antigens at the surface of bacterial OMVs, for vaccine development. Using two Streptococcus pneumoniae proteins as antigens, it was shown that mucosally administered Salmonella OMVs displaying high levels of antigens at the surface induced strong protection in a murine model of streptococcal colonization, without the need for a mucosal adjuvant. Importantly, reduction in bacterial recovery from the nasal cavity was correlated with local production of antigen-specific IL-17A. Furthermore, the protective efficacy and the production of antigen-specific IL-17A, and local and systemic IgGs, were all improved at increased concentrations of the displayed antigen.

Materials and Methods

[0100] Bacterial Strains and Growth Conditions:

[0101] S. Typhimurium SL3261 ΔtoIRA [21], and E. coli TOP10F′ and BL21(DE3) were grown at 37° C. in LB medium containing 0.2% glucose. When appropriate, kanamycin was used at a concentration of 25 μg/ml and chloramphenicol at 30 μg/ml. S. pneumoniae TIGR4 (Tettelin et al (2001) Science 293(5529):498-506) was grown and vaccination stocks containing 10.sup.6 colony forming units (CFU)/10 μl in phosphate buffered saline (PBS) were prepared as described (Cron et al (2011) Infect Immun 79(9):3697-710).

[0102] Plasmid Construction:

[0103] All reagents were purchased from Roche, except for Phusion DNA polymerase (Finnzymes) and SacI (New England Biolabs).

[0104] All plasmids used for expression of Hbp (derivatives) have a pEH3 backbone (Hashemzadeh-Bonehi et al (1998) Mol Microbiol 30(3):676-8). Plasmids pHbpD(Δd1), pHbpD(Δd2) and pHbpD(d4in), in which sequences coding for side domains d1, d2 and d4 of the Hbp passenger were substituted for Gly/Ser-encoding linkers containing SacI and BamHI restriction sites have been described (Jong et al (2012), supra).

[0105] Fragments of pspA encoding the N-terminal and the C-terminal portions of the α-helical domain, the LFB domain and the Pro-rich region, and of ply encoding an N-terminal (F1; aa 1-156), a central (F2; aa 145-421) and a C-terminal fragment (F3; aa 357-471) were amplified with flanking SacI/BamHI sites using chromosomal DNA of S. pneumoniae TIGR4 as template (GenBank accession no. AE005672) and the primers listed in Table 2. The resulting PCR amplicons were digested with SacI and BamHI and inserted into the hbp orfs of plasmids pHbpD(Δd1), pHbpD(Δd2) or pHbpD(d4in), which had been digested with the same restriction enzymes. This approach resulted in plasmids pHbpD(Δd1)-PspA(α1), pHbpD(Δd1)-PspA(LFBD), pHbpD(Δd1)-Ply(F1), pHbpD(Δd2)-PspA(α2), pHbpD(Δd2)-Ply(F3) and pHbpD(Δd4)-PspA(PRR).

[0106] To create a plasmid for expression of Hbp fused to both the α1 and the α2 fragments of PspA, the NdeI/NsiI fragment of pHbpD(Δd1)-PspA(α1) was substituted for that of pHbpD(Δd2)-PspA(α2), resulting in pHbpD-PspA(α1α2). Similarly, the NdeI/NsiI fragment of pHbpD(Δd1)-Ply(F1) was substituted for that of pHbpD(Δd2)-Ply(F3), yielding plasmid pHbpD-Ply(F1F3) for expression of a chimera containing both fragments F1 and F3 of pneumolysin. Finally, a plasmid was created for expression of Hbp fused to the LFBD and PRR fragments of PspA by replacing the NsiI/KpnI fragment of pHbpD(Δd1)-PspA(LFBD) for that of pHbpD(Δd4)-PspA(PRR).

TABLE-US-00002 TABLE 2 Primer sequences Name Sequence (5′ .fwdarw. 3′) SEQ ID NO PspA(PRR)-fw cggggagctccgagttaggccctgatggag 23 PspA(PRR)-rv tgccggatccttgtttccagcctgtttttgg 24 PspA(LFBD)-fw cggggagctcccttgctggtgcagatcctgatgatg 25 PspA(LFBD)-rv tgccggatccagtttcttcttcatctccatcag 26 PspA(α2) fw cggggagctccgtaagagcagttgtagttcc 27 PspA(α2) rv tgccggatcctgtgccatcatcaggatctgcaccagc 28 PspA(α1) fw cggggagctccgaagaatctccacaagttgtc 29 PspA(α1) rv tgccggatccatttggttcaggaactacaactg 30 Ply(F1)-fw cggggagctccatggcaaataaagcagtaaatgac 31 Ply(F1)-rv tgccggatccgtgagccgtgattttttcatac 32 Ply(F3)-fw cggggagctccgcttacagaaacggagatttactg 33 Ply(F3)-rv tgccggatccgtcattttctaccttatcttctac 34 nHis-pspA(dN31)-fw aaaccatgggccatcatcatcatcatcatcatcacag 35 cagcggcgaagaatctccacaagttgtcg pspA-rv tttcatatgttaaacccattcaccattgg 36

[0107] Constructs for expression and purification of N-terminally His-tagged PspA and PdT were created using the pET16b(+) vector (Novagen). For cloning of pspA, a fragment corresponding to aa 31 to 744 was amplified PCR using S. pneumoniae TIGR4 chromosomal DNA as template. The forward primer nHis-pspA(dN31)-fw was designed to contain a His.sub.8-epitope encoding sequence and an NcoI restriction site, while the reverse primer pspA-rv contained an NdeI site. A synthetic E. coli-codon-optimized DNA fragment encoding PdT flanked with NcoI and NdeI restriction sites was ordered from Life Technologies. The fragments were digested with NcoI and NdeI and ligated into pET16b(+), digested with the same enzymes, resulting in pET16b(+)::PspA(31-744) and pET16b(+)::PdT. The nucleotide sequences of all constructs were confirmed by DNA sequencing.

[0108] Hbp Expression, OMV Isolation and Protein Analysis:

[0109] S. Typhimurium SL3261 ΔtoIRA harboring pEH3 vectors was grown until an OD.sub.660 of ˜0.6, at which expression of Hbp derivatives was induced from the lacUV5 promoter in the presence of 1 or 100 μM IPTG for 1 hour.

[0110] To isolate OMVs, culture supernatants obtained by low-speed centrifugation were passed through 0.45-μm-pore-size filters (Millipore) and centrifuged at 208 000 g for 60 min, separating OMVs from soluble proteins. Pelleted OMVs were washed by re-suspension in PBS containing 500 mM NaCl (1 OD unit of OMVs per μl) and centrifugation at 440 000 g for 2 h, after which they were taken up in PBS containing 15% glycerol (1 OD unit of OMVs per μl). An amount of 1 OD unit of OMVs is derived from 1 OD.sub.660 unit of cells.

[0111] Proteinase K accessibility of OMV proteins was analyzed as described (Daleke-Schermerhorn et al (2014), supra), and proteins were analyzed by SDS-PAGE and Coomassie G-250 (BioRad) staining or immunoblotting using antisera recognizing the β-domain of Hbp (SN477), PspA or pneumolysoid, the detoxified derivative of Ply (own lab collection). Densitometric analysis on Coomassie-stained gels was carried out using a Molecular Imager GS-800 Calibrated Densitometer (Biorad) and Quantity One software (Biorad).

[0112] Expression and Purification of his-Tagged PspA and PdT:

[0113] Overnight cultures of E. coli BL21(DE3) harboring pET16b(+) plasmids for expression of PspA and PdT were grown at 37° C. in LB medium supplemented with ampicillin (100 μg/ml) and glucose (0.4%). The next morning, cultures were diluted to an OD.sub.660 of 0.05 in fresh medium and growth was continued under the same conditions. When the cultures had reached an OD.sub.660 of 0.6, recombinant protein production was induced by the addition of 100 μM IPTG for 2 h. Cells were harvested by low-speed centrifugation, re-suspended in ice cold PBS (pH 7.4) containing 5 mM imidazole and 300 mM NaCl, and lysed in a One Shot cell disrupter (Constant Systems Ltd) at 1.72 kbar. Unbroken cells were removed by low-speed centrifugation. Thereafter, soluble proteins were obtained by high-speed centrifugation at 208 000 g at 4° C. for 90 min. For PdT, His-tagged protein present in the high-speed supernatant was purified by affinity chromatography using HiTrap TALON crude columns (GE Healthcare). Bound PdT was eluted over a gradient of 5-500 mM imidazole in PBS (pH 7.4) containing 300 mM NaCl. Fractions containing His-tagged PdT were pooled and dialyzed against PBS containing 15% glycerol. The same procedure was followed to isolate His-tagged PspA, except that the high-speed supernatant and the buffers used during HiTrap TALON affinity chromatography contained 8 M urea.

[0114] Mouse Immunizations and Challenge:

[0115] Seven week-old female C57BL/6 mice (Charles River Laboratories) were intranasally (i.n.) immunized three, two or one time(s) with 8 OD units of OMVs (corresponding to ˜4 μg total protein) in a volume of 10 μl, at two-week intervals, under anesthesia (2.5% v/v isoflurane, AU Veterinary Services). Three weeks after the final immunization, mice were challenged i.n. with 106 CFU of S. pneumoniae TIGR4 (Cron et al (2011), supra). Three days after infection, mice were euthanized, and blood and mucosal nasal tissue were harvested. Nasal tissue was homogenized using an IKA T10 basic blender, and serially diluted samples were plated on Gentamicin Blood Agar (Mediaproducts BV) to determine bacterial recovery (log CFU/organ). All animal work was performed with approval of the Radboud University Medical Center Committee for Animal Ethics. Consequently, three mice were euthanized and excluded from further experimental analysis after reaching a humane endpoint, i.e. over 20% weight loss, potentially caused by a reaction to lipopolysaccharide present in the OMVs.

[0116] Detection of Antibody Responses by Enzyme-Linked Immunosorbant Assay Analysis:

[0117] Maxisorp high binding affinity plates (Nunc) were coated with 2 μg/ml purified PspA or PdT in carbonate coating buffer (0.1 M carbonate/bicarbonate pH 9.6) at 4° C. overnight. The next day, wells were blocked with 1% BSA (Sigma) and subsequently incubated for 1 h at 37° C. with serum or nasal samples from individual mice. Thereafter, the wells were incubated with primary anti-mouse IgG-alkaline phosphatase (Sigma) or primary anti-mouse IgA-alkaline phosphatase (Southern Biotech) for 1 h at 37° C. Between and after the incubations steps, all wells were washed with PBS containing 0.05% Tween-20 (Merck). Samples were developed using 1 mg/ml p-nitrophenylposphate in substrate buffer (1 M diethanolamine, 0.5 mM MgCl.sub.2 pH 9.8) (Calbiochem, VWR) and the optical density was measured at 405 nm 10 and 30 minutes after substrate addition.

[0118] Measurement of Local IFNγ and IL-17A:

[0119] Cytokine production in mouse nasal samples were determined with Cytometric Bead Array (Becton Dickinson) according to manufacturer's instructions, using the Mouse Enhanced Sensitivity buffer kit in combination with the Enhanced Sensitivity Flex set for IFNγ and IL-17A (Becton Dickinson). Concentrations were calculated using Soft Flow FCAP Array v1.0 (Becton Dickinson).

[0120] Statistical Analyses:

[0121] All statistical analyses were performed using GraphPad Prism version 5.0 (Graphpad Software). For bacterial recovery data, the Grubbs outlier test was used to test for significant outliers. The One-Way Anova Kruskall Wallis with Bonferroni post-test for multiple groups or Mann-Whitney t test for two groups were used for comparisons of protection and immune responses. To determine the relation between IL-17A and protection, a Spearman Correlation test was applied.

Results

[0122] Selection and Fusion of PspA and Pneumolysin Fragments to Hbp:

[0123] To facilitate efficient expression of the model antigens PspA and Ply via the Hbp display system, these two complex, multi-domain proteins were split into shorter fragments (FIG. 1A). To avoid potential destruction of immune epitopes, fragments were designed to partially overlap with adjacent sequences. PspA was divided into four fragments that cover the N-terminal, surface-exposed part of the protein (FIG. 1A). Of these, two fragments derived from the N-terminal portion of α-helical coiled coil domain (“α1” and “α2”) were fused to a single Hbp carrier, yielding HbpD-PspA[α1-α2] (SEQ ID NO:4; FIG. 1B). A second construct, HbpD-PspA[LFBD-PRR] (SEQ ID NO:6; FIG. 1B), was created by combining Hbp with two fragments corresponding to the lactoferrin-binding domain (‘LFBD’) and the conserved Pro-rich region (‘PRR’) of PspA, respectively. Similarly, pneumolysin was divided into three fragments (FIG. 1A), of which an N-terminal (′F1′) and a C-terminal fragment (‘F3’) were fused to a single Hbp carrier to create HbpD-Ply[F1-F3] (SEQ ID NO:7; FIG. 1B). Unfortunately, the central pneumolysin fragment (‘F2’; FIG. 1A), that includes the membrane-penetrating D3 domain, caused lysis upon expression in the context of Hbp and was therefore excluded from further studies.

[0124] Efficient Display of PspA and Pneumolysin Fragments at the Surface of Salmonella OMVs:

[0125] The three Hbp chimeras were expressed under control of a lacUV5 promoter in a previously described ΔtoIRA derivative of the attenuated Salmonella Typhimurium strain SL3261 that produces large amounts of OMVs (Daleke-Schermerhorn et al (2014), supra). A display derivative of Hbp lacking domain d1, HbpD (SEQ ID NO:3; FIG. 1B; Jong et al (2012), supra), was produced in the same strain and the resulting vesicles served as a negative control in subsequent studies. OMVs were isolated from cell-free culture supernatants by ultracentrifugation, after which they were washed with PBS containing a high concentration of NaCl to remove peripherally attached soluble contaminants. The absence of live bacteria was verified by plating, and OMVs were confirmed to contain the Hbp-antigen chimera by SDS-PAGE (FIG. 2A) and immunoblotting (FIG. 2B). Importantly, and similar to the HbpD control, all three chimeras were exposed at the OMV surface as judged by their sensitivity to externally added Proteinase K (FIGS. 2B and C). In contrast, degradation of the protease-sensitive C-terminal periplasmic domain of OmpA occurred only after permeabilization of the OMVs with Triton X-100, confirming the integrity and membrane orientation of the vesicles (FIG. 2C). Differences in expression levels of the various HbpD-antigen fusion proteins are likely determined by the difference in complexity of the inserted fragments. Nevertheless, all Hbp-antigen fusions were visible after Coomassie staining indicating that the expression levels are substantial.

[0126] Salmonella OMVs Displaying PspA Fragments Protect Against Pneumococcal Colonization:

[0127] To investigate whether Salmonella OMVs displaying PspA and Ply fragments at the surface can confer protection against pneumococcal colonization, we made use of a previously established mouse model (Wu et al (1997) Microb Pathog 23(3):127-37). Mice were intranasally immunized three times with two-week intervals, and three weeks after the final immunization, they were intranasally inoculated with the S. pneumoniae serotype 4 TIGR4 strain. Recovery of live pneumococci from nasal tissue three days post-infection revealed that mice immunized with OMVs displaying HbpD-PspA[α1-α2] were significantly protected compared to mice that received OMVs displaying HbpD (FIG. 3A). Remarkably, over 50% of the mice that received OMVs/HbpD-PspA[α1-α2] completely cleared the pneumococci within three days post-infection. In contrast, mice immunized with OMVs displaying either HbpD-PspA[LFBD-PRR] or HbpD-Ply[F1-F3] showed little, and in the case of HbpD-PspA[LFBD-PRR] rather variable protection, which did not significantly differ from the negative control. To investigate whether the reduced protection may be due to insufficient expression levels of HbpD-PspA[LFBD-PRR] and HbpD-Ply[F1-F3] (FIG. 2A), immunization was repeated with OMVs displaying one of the antigen fragments, i.e. PspA[PRR] (FIG. 2A, lane 10; SEQ ID NO:5), at even ˜3-fold higher levels than PspA[α1-α2] (FIG. 2A, lane 8). However, no reduction in bacterial load was observed (FIG. 6). Together these data show that the OMV/Hbp platform is suitable for intranasal antigen delivery. Furthermore, the N-terminal half of the α-helical coiled coil domain of PspA is able to induce protection against pneumococcal colonization.

[0128] Immunogenicity of Salmonella OMVs Displaying Pneumococcal Antigens:

[0129] To shed light on the underlying mechanism behind the observed protection, we quantified the levels of IFNγ and IL-17A in nasal tissue of mice immunized with OMVs displaying HbpD-PspA[α1-α2] or the control protein HbpD. Of note, because the vaccinated mice were subjected to subsequent pneumococcal infection, the cytokine levels measured at three days post-challenge reflect recall responses of immunization-induced memory towards the pneumococcal antigen fragments. Interestingly, while both groups produced similar levels of IFNγ (FIG. 4A), mice immunized with OMVs/HbpD-PspA[α1-α2] produced significantly higher levels of IL-17A levels compared to control mice (FIG. 4B). This strongly suggests that local production of IL-17A, but not IFNγ, is important for protection against pneumococcal colonization.

[0130] Because IgGs are important for bacterial opsonization and initiation of robust immune responses, we determined the levels of PspA- and Ply-specific IgGs and IgAs in nasal tissue homogenates and sera of immunized mice (FIGS. 5A, 7A and 8A). Interestingly, significant yet variable local and systemic antigen-specific IgG production was detected upon immunization with OMVs/HbpD-PspA[α1-α2]. In contrast, with the exception of apparently lower local IgGs induced by OMVs/HbpD-Ply[F1-F3], no antibody responses were detected in the other groups. Moreover, there were no differences observed for local IgA production between the treatment groups.

[0131] Together these results show that the OMV/Hbp platform induces local antigen-specific IL-17A responses. Additionally, the platform can induce local and systemic humoral responses.

[0132] Level of Protection is Influenced by the Vaccine-Associated Antigen Load and the Number of Immunizations:

[0133] Although intrinsic properties of the α1/α2 fragments of PspA are apparently crucial for protection, the strikingly high expression levels of HbpD-PspA[α1-α2] (FIG. 2A) prompted us to investigate whether antigen levels represent an important determinant for immune responses and protection. To this end, mice were immunized with two new batches of OMVs displaying HbpD-PspA[α1-α2] at high levels similar to the first batch or at approximately ˜7-fold lower levels (FIG. 2A). Similar to the previous experiment, mice immunized with OMVs expressing high levels of HbpD-PspA[α1-α2] produced significantly elevated levels of antigen-specific IL-17A in nasal tissue (FIG. 4C), and local and systemic antigen-specific IgG, but not local antigen-specific IgA (FIGS. 5B, 7B and 8B). In contrast, IL-17A and IgG levels were low or undetectable in samples from mice immunized with OMVs displaying low levels of HbpD-PspA[α1-α2], and did not significantly differ from those of control mice that received three doses of OMVs without pneumococcal protein fragments (FIGS. 4C, 5B and 7B). Furthermore, although low PspA[α1-α2] induced some protection against pneumococcal colonization, a clear trend suggested that enhanced expression of PspA[α1-α2] improved protection (FIG. 3B).

[0134] To investigate whether similar levels of protection could be reached with fewer immunizations, two additional groups of mice were included that received one or two immunizations with OMVs expressing high levels of HbpD-PspA[α1-α2]. Antibody analysis revealed significantly elevated local and systemic IgG, but not local IgA levels independently of the number of immunizations (FIGS. 5B, 7B and 8B). However, although slightly more IL-17A was detected after two vaccinations, three immunizations were required to induce significantly higher IL-17A production compared to control mice (FIG. 4C). Moreover, three immunizations offered significantly better protection than two immunizations, while one immunization was not sufficient to induce protection as compared with the control (FIG. 3B).

[0135] In conclusion, the antigen abundance and the number of immunizations directly influence the levels of nasal IL-17A and protection against S. pneumoniae colonization. Importantly, production of nasopharyngeal IL-17A significantly correlated with reduced bacterial recovery from the nasal tissue (p=0.0032) (FIG. 4D), suggesting that local IL-17A responses are crucial for protective memory responses induced by our OMV/Hbp antigen display platform.