Oligosaccharide ligand tuning in design of third generation carbohydrate pneumococcal vaccines

Introduction

Prevention of pneumococcal diseases with polysaccharide and conjugate vaccines based on capsular polysaccharides (CPs) of Streptococcus pneumoniae has led to a significant reduction of the morbidity rate of otitis media, pneumonia, meningitis and mortality in all age groups [1], [2], [3], [4], [5]. The efficacy of polysaccharide-based vaccines containing CPs of clinically significant serotypes of S. pneumoniae is due to the fact that the polysaccharide capsule surrounding the bacterial cell is one of the major protective antigens of pneumococci [6].

The currently used first generation polysaccharide-based vaccines consist of a mixture of CPs. Their main disadvantage is a lack of efficiency in children of younger age groups because purified CPs are unable to induce T-dependent immune response with the following production of CP-specific IgG antibodies and memory cells [7], [8]. For the induction of the T-dependent immune response, the second generation of pneumococcal vaccines has been developed in which CPs are covalently bound to a protein carrier (diphtheria or tetanus toxoids or protein D of Haemophilus influenzae type b) [3], [9], [10].

The main disadvantages of the first and second generation carbohydrate vaccines are expensive and time-consuming processes of production and purification of CPs, the inevitable presence of bacterial impurities in the final product, insufficient immunological activity of some CPs of S. pneumoniae [11], the necessity to work with living bacteria cells and, additionally for conjugate vaccines, not always successful conjugation of CPs to the protein carrier [12]. Despite the relative efficacy of the first and second generation carbohydrate vaccines, their disadvantages determine the necessity of further improvement of the design of pneumococcal vaccines.

Currently, third generation carbohydrate pneumococcal vaccines [13] based on comprising protective epitopes synthetic analogues of fragments of the CPs of S. pneumoniae are under development. These studies include several technological approaches: development of neoglycoconjugate vaccines with adjuvant [14]; peptide-free, liposome-based oligosaccharide vaccine adjuvanted with a natural killer T cell antigen [15]; lipid-carbohydrate conjugate vaccine without protein carrier which stimulates invariant natural killer T (iNKT) cells that possess the ability to stimulate the production of high-affinity IgG antibodies specific for pneumococcal polysaccharides and long-lived memory B cells [16]; the use of gold nanoparticles as carriers for synthetic oligosaccharides [17]. Advantages of neoglycoconjugate vaccines include the well-defined structure of the protective epitope, the possibility to control the binding of ligands to the protein carrier and appropriate conformational structure of the ligands.

The main problem in the development of such vaccines is the absence of convenient rules for correct choice of efficient oligosaccharide ligands to be used in vaccines. The search for them includes the synthesis of oligosaccharides corresponding to different fragments of the polysaccharide chain, their conjugation to a protein carrier for induction of T-dependent immune response, evaluation of their immunogenicity by measuring the level of specific IgG antibodies in sera of immunized animals and the ability of sera to promote opsonophagocytosis, as well as the study of the protective activity of the obtained neoglycoconjugates by challenging animals with the relevant serotype of S. pneumoniae. In the 2000s, such an approach was applied to the search for an optimal oligosaccharide ligand for S. pneumoniae type 14 [18]; however, one of the most important characteristics of a vaccine candidate, namely, the protective activity in vivo was not determined.

In this review, we summarize the results of our studies on the determination of the most appropriate oligosaccharide ligand related to S. pneumoniae type 14 for the design of candidate pneumococcal vaccine with the use of synthetic oligosaccharides representing one (1a), “one and a half” (2a) and two (3a) repeating units of the CP (Fig. 1); their conjugates 1b, 2b and 3b with the model protein carrier BSA; as well as biotinylated oligosaccharides 1c, 2c and 3c used as coating antigens in ELISA assays. The choice of S. pneumoniae type 14 is explained by its high prevalence in the population, a high degree of invasiveness and the ability to cause severe pneumococcal diseases in children [19], [20].

Fig. 1:

Structures of synthetic oligosaccharides and their BSA and biotin conjugates used to determine an optimal ligand for a synthetic glycoconjugate vaccine.

Synthesis of oligosaccharides related to the capsular polysaccharide of S. pneumoniae type 14

The capsular polysaccharide of S. pneumoniae type 14 is built of tetrasaccharide repeating units 4 (Fig. 2), which in turn consist of lactose and N-acetyllactosamine blocks connected through a β-(1→3′)-linkage. In the late 1980s, Kochetkov and coworkers synthesized polysaccharide 4 with an average degree of polymerization of 10 (n=10) by polycondensation of a tetrasaccharide monomer bearing a donor 1,2-O-(1-cyanoethylidene) function in the glucose residue and an acceptor 6-trityloxy group in the glucosamine unit [21]. A large set of oligosaccharides related to the capsular polysaccharide of S. pneumoniae type 14 was synthesized in the late 1990s – early 2000s [22], [23]. This set included oligosaccharides representing different regions of the polysaccharide chain from trisaccharides up to the dodecasaccharide comprising three repeating units.

Fig. 2:

Repeating unit of capsular polysaccharide of S. pneumoniae type 14.

According to the structure of the polysaccharide, disaccharide synthetic blocks derived from lactose (5, 8) and lactosamine (6, 7, 9) were employed for the assembling of the target oligosaccharides (Fig. 3) [24].

Fig. 3:

Lactose and lactosamine building blocks for assembling target oligosaccharides.

The synthesis of the oligosaccharides has been carried out in a straightforward manner using regioselective glycosylation of 3-OH in galactose and 6-OH in glucosamine. The protected precursor of tetrasaccharide 1a was obtained in 87% yield by TMSOTf-catalyzed glycosylation of lactosamine acceptor 9 with lactose trichloroacetimidate 8 (Fig. 4). Deprotection of tetrasaccharide 10 included acidic removal of the isopropylidene group, removal of all O- and N-acyl groups by treatment with hydrazine hydrate, exhaustive N,O-acetylation with Ac₂O in pyridine and O-deacetylation with sodium methoxide. In the final step, the azido group was reduced by hydrogenation over PdO/C to produce tetrasaccharide 1a.

Fig. 4:

Synthesis of tetrasaccharide corresponding to one repeating unit of capsular polysaccharide of S. pneumoniae type 14.

The synthesis of hexa- and octasaccharides is outlined in Fig 5. Lactoside acceptor 5 was subjected to NIS–TfOH-promoted regioselective 3-O-glycosylation with lactosamine thioglycoside 6 followed by removal of the TBS protection to afford tetrasaccharide diol 11 in 76% yield. Further glycosylation of 11 with lactose imidate 8 proceeded highly regioselectively at the primary hydroxyl group and furnished the corresponding hexasaccharide (89%) that provided triol 12 after removal of the isopropylidene group from the terminal galactose residue. Deprotection of 12 followed by reduction of the spacer azido group as described above produced free hexasaccharide 2a.

Fig. 5:

Synthesis of hexa- (2a) and octasaccharide 3a representing to “one and a half” and two repeating units of capsular polysaccharide of S. pneumoniae type 14.

Final NIS–TfOH-promoted glycosylation of 12 with lactosamine thioglycoside 7 also demonstrated high regioselectivity and afforded octasaccharide 13 in 80% yield. The latter was converted into free oligosaccharide 3a using the standard procedure described above.

The synthesized oligosaccharides were converted to immunogens 1b–3b [24] by conjugation to the model protein carrier BSA via a squarate linker [25] (Fig. 6). Acylation of the oligosaccharides with pentafluorophenyl ester 15 [26] derived from biotin equipped with a flexible and hydrophilic hexaethylene glycol linker provided biotin conjugates 1c–3c [27] applied as coating antigens in the ELISA assay.

Fig. 6:

Conjugation of oligosaccharides with BSA and biotin.

Search for an optimal oligosaccharide ligand of S. pneumoniae type 14

For a correct determination of the structure of the ligand that could be suitable for the construction in the future of a multicomponent third generation carbohydrate pneumococcal vaccine, a series of immunological tests was conducted including the ability of conjugates to induce the production of IgG antibodies specific to homologous oligosaccharides (OSs) and capsular polysaccharides (CP), the ability of sera obtained after immunization with the conjugates to promote opsonophagocytosis of bacterial cells and the protective activity of the conjugates.

Immunogenicity of neoglycoconjugates was assessed by the formation of antibodies in mice. We determined the level of antibodies in the sera to homologous OSs 1c–3c and CP of S. pneumoniae type 14. All conjugates 1b–3b were immunogenic but to a different extent. The highest titer of antibodies specific to CP induced conjugated octasaccharide 3b (Fig 7b), whereas production of the highest level of antibodies to the homologous OS evoked conjugated tetrasaccharide 1b (Fig 7a). Hexasaccharide conjugate 2b possessed the lowest immunogenicity giving only a low antibody production to CP and to homologous biotinylated OS 2c. The titers of antibodies to OSs 1c–3c in the same sera of mice immunized with conjugates 1b–3b absorbed on aluminum hydroxide were 66, 32 and 16 times higher than the titers of antibodies detected by CP (Fig. 7) [28]. It was shown previously that some oligosaccharides conjugated to a protein induce the formation of antibodies to capsular polysaccharides at a higher level than traditional polysaccharide conjugate vaccines [29]. We revealed a higher titer of antibodies to the OSs compared to the level of antibodies to CP in the same immune sera probably due to the fact that antibodies to conjugated OSs do not precisely correspond to the conformational structure of the CP. For conjugated OSs 1b–3b related to S. pneumoniae type 14, this fact has been observed for the first time.

Fig. 7:

Titer of IgG1 antibodies to biotinylated OSs 1c–3c as coating antigens (a) and to CP as coating antigen (b).

Thus, in the ELISA assay for the evaluation of the level of antibodies specific to biotinylated OSs 1c–3c, conjugate 1b based on tetrasaccharide 1a corresponding to the repeating unit of CP of S. pneumoniae type 14 possessed the highest immunogenic activity. At the same time, it was revealed that antibodies to conjugate 1b have relatively low affinity to bacterial CP.

The specificity of antibody binding in sera to conjugates 1b–3b and CRM₁₉₇-CP with ligands 1a–3a in ELISA inhibition assay demonstrated that tetrasaccharide 1а possessed the higher capacity to inhibit the binding between anti-OSs antisera and biotinylated OSs or CP also used as a coating material as compared to ligands 2a and 3a (data are not presented) [30]. The functional activity of IgG antibodies to tetrasaccharide conjugate 1b was confirmed in the test of slide agglutination of living bacteria S. pneumoniae type 14 on adding to them the immune serum (Table 1).

The data obtained confirmed efficient agglutination (++++) caused by the serum to tetrasaccharide conjugate 1b related to the repeated unit of CP of S. pneumoniae type 14 and the fact that the serum to octasaccharide conjugate 3b was less active, whereas the serum to hexasaccharide conjugate 2b was found to be inactive in this test [28].

The ability of the serum to glycoconjugates 1b–3b to promote phagocytosis of inactivated bacteria confirmed their high immunological activity (Table 2). The difference in opsonizing activity of the serum to glycoconjugates 1b–3b could not be detected, because the mice were immunized with the most effective dose (10 μg/mouse of carbohydrate). Also no difference was found in the experiments on the passive protection of mice treated with immune serum to glycoconjugates 1b–3b and challenged with S. pneumoniae type 14 (data are not presented) [30].

The selection of protective oligosaccharide related to CP of S. pneumoniae type 14 is initially based on the protocols of Safari et al. [18]. The authors presented the data for the comparative evaluation of the antibody response and the avidity and opsonizing activity of antibodies to conjugated synthetic oligosaccharides (from tri – to dodecasaccharides) and made the conclusion that tetrasaccharide is a serious candidate for a synthetic oligosaccharide conjugate vaccine against infections caused by S. pneumoniae type 14 [18]. However, the studies have not been confirmed by experiments on active protection of immunized animals against infection caused by S. pneumoniae type 14.

We studied the protective activity of the conjugates in experiments on active immunization of mice [30]. The highest protective activity after immunization of mice with glycoconjugates 1b–3b in a single dose of 2.5 μg of carbohydrate was revealed for tetrasaccharide conjugate 1b and CRM₁₉₇-CP of S. pneumoniae type 14 conjugate (positive control) that is a part of the commercial pneumococcal conjugate vaccine Prevenar-13 (Fig. 8). Protective activity of octasaccharide conjugate 3b and, in particular, hexasaccharide conjugate 2b was significantly lower. Notably, immunization of mice with glycoconjugates 1b–3b without adjuvant did not protect animals against infection caused by S. pneumoniae type 14.

Fig. 8:

Protective activity of glycoconjugates absorbed on aluminum hydroxide.

Some authors indicate that conjugated OSs with different chemical structures related to CP of S. pneumoniae type 14, including hexasaccharide conjugate, may be immunologically active [18], [31]. Synthetic hexasaccharide conjugate 2b was characterized by serotype specificity in the ELISA assay, was used as coating antigen recognizing the antibodies only to the antimicrobial serum to S. pneumoniae type 14 and did not interact with sera to serotypes 19A and 19F [32]. We showed that hexasaccharide conjugate 2b also possessed immunological activity but less pronounced than that of conjugates 1b and 3b. The presence of two extra monosaccharides in addition to the tetrasaccharide repeating unit of the CP in its structure imparts it greater similarity to CP of S. pneumoniae type 14 compared to OSs with a shorter chain length. This was the basis for the in-depth immunological study of conjugated hexasaccharide 2b to evaluate its action on the stimulation of innate immunity with the following development of cell-mediated and antibody immune response.

A single immunization of mice with hexasaccharide conjugate 2b resulted in an increase in bactericidal activity of the peripheral blood leukocytes against the heterologous pathogen, Staphylococcus aureus [33]. When administered to mice, hexasaccharide conjugate 2b, non-absorbed or absorbed on the aluminum hydroxide, increased the number of cells expressing Toll-like receptor 2 (TLR2) on mononuclear leukocytes in the spleen of mice. It was proved that the activation of TLR2 was not the result of a direct ligand-receptor interaction. The addition of conjugated hexasaccharide 2b to the culture of cells generated from the mice bone marrow led to maturation of dendritic cells (CD11c⁺, CD80⁺ and MHCII⁺), which produced cytokines IL-1β, IL-6, and TNFα into the culture medium [34].

After a single administration of hexasaccharide conjugate 2b to mice, the levels of cytokines IL-1β, IL-6, IL-10, IFNγ and TNF α increased in the interval from 2 to 24 h. Immunization with hexasaccharide conjugate 2b absorbed on aluminum hydroxide stimulated the production of a broader spectrum of cytokines GM-CSF, IL-1β, IL-5, IL-6, IL-10, IL-17, IFNγ and TNFα [34].

The number of CD3⁺ T lymphocytes increased after the second immunization with conjugated hexasaccharide 2b absorbed on aluminum hydroxide, whereas that of CD4⁺ T lymphocytes remained within normal values and the number of СD8⁺ T lymphocytes decreased. The amount of B cells (CD5⁺, CD19⁺), NK⁺ cells and molecules of the antigen presentation MHCII⁺ increased during this period, thus indicating the influence of hexasaccharide conjugate 2b on the activation of cell-mediated immune response [34]. After two-fold immunization of mice with conjugated hexasaccharide 2b absorbed on aluminum hydroxide, the level of antibodies to CP diminished on the 92^nd day. Booster immunization with hexasaccharide conjugate 2b with the adjuvant stimulated the production of IgG memory antibodies, which were determined within 97 days. Immunization with conjugated hexasaccharide 2b absorbed on aluminum hydroxide elicited the formation of predominantly IgG1 antibodies [34]. The obtained data characterizing the immunological activity of the synthetic analog of CP (2b) of S. pneumoniae type 14 on the stimulation of innate and adaptive immunity can be used to improve the quality of the second and third generation carbohydrate pneumococcal vaccines.

Thus, the set of the obtained results indicated the advisability of the use of tetrasaccharide 1a as a carbohydrate ligand specific to CP of S. pneumoniae type 14 for the development of a synthetic third generation carbohydrate pneumococcal vaccine. Moreover, tetrasaccharide 1a can be applied for the development of diagnostic assays and for obtaining pneumococcal antisera for serotyping S. pneumoniae type 14.