A mechanism for glycoconjugate vaccine activation of the adaptive immune system and its implications for vaccine design

Transcription

1 A mechanism for glycoconjugate vaccine activation of the adaptive immune system and its implications for vaccine design Fikri Y. Avci 1,2, Xiangming Li 3, Moriya Tsuji 3, Dennis L. Kasper 1,2*

2 Supplementary Figure 1 Glycoconjugate vaccine Protein BCR Polysaccharide B cell CD8/86 CD28 T cell MHC II Peptide TCR CD4 CD4L IL-4 Receptor IL-4 IL-2 Receptor IL-2 Figure S1. Schematic representation of steps in the original hypothesis for processing and presentation of glycoconjugate vaccine epitopes and consequent CD4+ helper T-cell induction of B-cell production of IgG antibodies to the polysaccharide. (1) Polysaccharide-protein conjugates bind to the B-cell receptor (BCR) of polysaccharide-specific pre-b cells. (2) Conjugates are taken up into the endosome of B cells. (3) nce a conjugate is inside a cell, the protein portion is digested in the endolysosome by acid proteases to release peptides. (4) Peptides bind to MHCII molecules with the help of HLA-DM to replace the CLIP portion of the self-peptide. (5) The peptide from the vaccine carrier protein in the context of MHCII is presented to the CD4+ T cell (αβtcr). (6) Along with co-stimulatory molecules, peptide/mhcii activates the αβtcr. (7) Cytokines like IL-4 and IL-2 are secreted by the CD4+ T cell. The cognate interactions of the B cell with the T cell, in combination with the action of cytokines, stimulate B-cell maturation and induce both immunoglobulin class switching (from IgM to polysaccharide-specific IgG) and B-cell memory.

3 Supplementary Figure 2 a -IgG (day 14) b -IgM Antibody concentration (µg/ ml) Day : ND* PBS VA + VA III-VA Antibody titer* Day : ND* PBS Day 14 Day 21 ND* VA + VA III-VA Day 14: III-VA III-VA III-VA III-VA III-VA c Anti-CD4+ treated mouse Isotype control treated mouse.1% 18.6% CD4 CD19 Figure S2. -specific IgG and IgM antibody responses to immunization. (a) -specific IgG titers on day 14 after immunization of BALB/c mice (from Fig. 1a) with one dose of the specified antigens. (b) -specific IgM levels in sera obtained from immunized BALB/c mice (from Fig. 1a) on day 14 (after priming) or on day 21 (after boosting) with the specified antigens. *Antibody titers are reported as the reciprocal dilution that resulted in an A45 of.5. (c) Flow cytometry analysis of CD4- and CD19-stained splenic lymphocytes obtained on day 14 from mice previously treated with antibody to CD4 or an isotype control. Results shown are representative of eight spleens obtained on day 14 or 21.

4 Supplementary Figure 3 H H H H HC H H H H NHAc H H H H NHAc H H 1. NaI 4 (7%) 2. ( 3 H)NaBH 4, ph 8 3. NaI 4 (3%) 4. VA (or TT, or VAp) Na(CN)BH 4, ph 7.2 H H H H H HC 7% H 3 H H H NHAc H H H H NHAc H H H H H H H H H H NHAc H H HC Protein N H H H NHAc Figure S3. Chemistry for radiolabeling of followed by conjugation to carrier protein/peptide. In brief, has a side-chain sialic acid residue on each repeating unit. A selected number of these sialic acids can be oxidized with sodium periodate to create an aldehyde at C7 or C8. Radioactive labeling takes place by reduction of the aldehyde with [3H]sodium borohydride. A selected additional number of the nontritiated sialic acids can then be oxidized, and the aldehyde that is formed can be linked as a secondary amine to lysine groups in the protein by means of reductive amination.

5 Supplementary Figure 4 Endolysosomes Cell Surface HLA-DR LAMP-1 RAB5 CD19 HLA-DR LAMP-1 RAB5 CD kda 82 kda 64 kda 26 kda Figure S4. Fractionation of Raji B cells by differential centrifugation yielded endolysosomes and cell surface fractions. Western blot analysis of the solubilized endolysosomes showed endosomal (Rab5) and lysosomal (LAMP-1) proteins, and cell surface fractions contained membrane proteins (CD19). None of the isotype controls for the antibodies used in the assay labeled proteins in the membranes (not shown). Serial dilutions of endolysosomal content were run on the gels along with undiluted cell-surface fraction. After transfer and labeling of the membranes, both fractions were exposed to films simultaneously until bands for LAMP-1 and Rab5 were observed on the cell surface fractions. These bands on the cell surface fractions had the same density as 25X-diluted LAMP-1 and Rab5 bands from endolysosomal fractions. This result demonstrated that the cell surface fraction was 95% free of endosomes and lysosomes (not shown).

6 Supplementary Figure 5 a CPM (normalized) void volume 7 kda Elution volume (ml) No RS Scavenger TEMP (1 mm) TEMP (1 mm) intact *-VA 1 kda b CPM (normalized) c CPM (normalized) void volume 7 kda void volume Elution volume (ml) 7 kda 1 kda Elution volume (ml) No RS Scavenger D-Mannitol (1mM) D-Mannitol (1 mm) intact *-VA No RS Scavenger Sodium Pyruvate (1mM) Sodium Pyruvate (1 mm) intact *-VA 1 kda Figure S5. Endosomal processing of [3H]III-VA in Raji B cells. [3H]III-VA (2 µg, 18 cpm/ng) was incubated in Raji cells (3x16/ml per well) for 12 h, either alone or in the presence of a RS scavenger: (a) 4-H TEMP (superoxide inhibitor); (b) D-mannitol (hydroxyl radical inhibitor); or (c) sodium pyruvate (hydroperoxide inhibitor). The molecular size distribution of the end products from whole-cell lysates was analyzed by Superose 12 chromatography. *The portion of the glycoconjugate was selectively radiolabeled with tritium.

7 Supplementary Figure 6 a CPM *** b CPM void volume 7 kda 1 kda Intact -TT Raji cell surface RJ225 cell surface 1 HLA-DR LAMP-1 CD19 HLA-DR Elution volume (ml) Raji RJ225 Figure S6. Carbohydrate epitopes of III-VA and III-TT glycoconjugates are presented on the surface of APCs in the context of HLA-DR. (a) Raji or RJ2.2.5 B cells (the latter not expressing MHCII) were incubated (18 h) with [3H]III-VA, and cell membrane extracts were co-immunoprecipitated with mab to HLA-DR; mabs to LAMP-1 and CD19 served as controls. Each co-ip was performed with cell membrane extract obtained from 17 Raji or RJ2.2.5 cells. Data are expressed as mean ± SD CPM values. ***, p <.1 (b) [3H]III-TT incubation with Raji B cells (18 h) and Superose 12 molecular sizing column analysis of cell surface content immunoprecipitated with mab to HLA-DR revealed that low-molecular-weight (~1 kda) was bound to MHCII molecules at the Raji cell surface. As a negative control, this experiment was repeated with MHCII-deficient Raji cells (RJ2.2.5), and no tritiated oligosaccharide was detected at the cell surface (pink). As a control for the mab to MHCII used for co-ip, an antibody to LAMP-1 was used; no counts were detected in this immunoprecipitate (not shown).

8 Supplementary Figure 7 a HLA-DR Ab VAp Ab Ab b HLA-DR Ab VAp Ab Ab III-VAp III-VAp III-VAp VAp VAp VAp 82 kda 64 kda 82 kda 64 kda c Cell Count Mean Fluorescence Intensity (MFI) RJ2.2.5 VAp RJ2.2.5 VAp-Biotin Raji VAp Raji VAp -Biotin Figure S7. Characterization of a cell surface complex containing MHCII, VAp, or. Cell membrane extracts of Raji B cells incubated with III-VAp or (a) and with VAp or (b) were run on SDS gels and stained by western blotting with antibodies to HLA-DR, VAp, and. A band at ~82 kda was detected by all three antibodies in the III-VAp-incubated cell membrane extracts, a result suggesting that a complex of MHCII and a glycanp-peptide exists on the surface of these B cells. As control antibodies, a mouse IgG2a isotype (a control for HLA-DR and antibodies) or normal mouse serum (a control for VAp antibodies) were used in both a and b. None of the PVDF membranes showed a band when incubated with control antibodies (data not shown). Raji cells were incubated with a control non-hla-binding peptide (i.e., scrambled VAp) and the PVDF membrane didn t show a band after labeling with antibody to VAp. (c) Flow cytometric analysis of Raji B cells or RJ2.2.5 cells after incubation with VAp or an VAp-biotin conjugate followed by surface staining of the cells with fluorescein-conjugated NeutrAvidin at 4 C.

9 Supplementary Figure 8 a VAp-IgG b VAp-IgM c -IgM P >.5 (NS) Antibody titer* 2 1 WT D11.1 Antibody titer* WT D11.1 Antibody titer* 4 2 <5 WT D11.1 Figure S8. Response of D11.1 mice to immunization with glycoconjugate vaccine. (a) VAp-specific IgG titers were measured on day 21 after immunization (2 doses) with III-VAp in the D11.1 and wild-type (WT) BALB/c mice from Figs. 3a and 3b. (b, c) VAp- and -specific IgM levels in sera from III-VAp-immunized WT and D11.1 mice (same mice as Figs. 3a and 3b). *Antibody titers are reported as the reciprocal dilution that resulted in an A45 of.5.

10 Supplementary Figure 9 IL-2 secreting cells per 1 cells Anti-TCR blocking of mil-2 ELISpot of CD4+ T-cell clone # µg/ml (-) (-) (-) Antigen III-VA III-TT III-HEL None No antibody Anti-αβTCR Anti-γδTCR IL-2 secreting cells per 1 cells Anti-TCR blocking of mil-2 ELISpot of CD4+ T-cell clone # µg/ml (-) (-) (-) Antigen III-VA III-TT III-HEL None No antibody Anti-αβTCR Anti-γδTCR Figure S9. Two CD4+ T-cell clones specifically recognize the carbohydrate portion of a glycoconjugate vaccine in the context of MHCII on APCs. ELISpot assays for the detection of CD4+ T-cell clones secreting IL-2 were conducted with several antigens, first in the absence and then in the presence of various concentrations of mab to αβtcr or γδtcr. Irradiated naïve mouse splenocytes were used as APCs.

11 Supplementary Figure 1 a b Key: VA VAp C terminus extension (ESGK) N-Acetylation of N terminus Figure S1. Cartoon representations of III-VA (a) and III-VAp (b). III-VA has a bulky matrix-like structure with one or two antigenic glycanp-peptides per polysaccharide chain (with an assumption that the peptide portion of each glycan p -peptide is an MHCII-binding peptide). III-VAp has a uniform structure containing ~8 glycanp-peptide epitopes per carbohydrate chain based on the carbohydrate-to-peptide mass ratios in the conjugate (3:1). The single-chain presentation of III-VAp is based on the coupling chemistry. VAp can react with only one aldehyde group on the sugar chain through its free amino group at the C-terminus lysine residue; the N-terminus is N-acetylated.

12 Supplemental Methods In vivo CD4+ T-cell depletion. We treated mice i.p. with a mab to CD4 (3 µg per dose; clone GK1.5, BioLegend) on days 11 and 13 between priming (day ) and boosting (day 14). In a control experiment, we treated mice with an isotype control (rat IgG2b, BioLegend). We analyzed CD4+ T-cell depletion by flow cytometry with a noncompeting FITC-conjugated mab to CD4 (clone RM4-4, BioLegend). Cell fractionation and in vitro processing assays. We cultured Raji B cells (1 8 ) in the presence of.4 mg of [ 3 H]III-VA or.2 mg of [ 3 H] for 3 18 h at 37 C to allow uptake and processing. After removal of unreacted [ 3 H], we performed cell lysis by passage through a 27-gauge needle in 25 mm sucrose with 1 mm Tris-HCl, ph 7.5 (fractionation buffer). We then performed differential fractionation of the lysed cells into endolysosome and cell-membrane fractions with an adaptation of previously published methods of differential centrifugation 1-3. Endolysosomal products were solubilized by boiling for 2 min in 1% SDS 3M NaCl and were analyzed by molecular sieve chromatography on a Superose 12 column in PBS with a BioRad FPLC system (BioLogic HR, BioRad) as described 3. Western blot. MHCII-glycan p -peptide complexes from the cell membrane fractions of Raji B cells (1 8 ) incubated with.4 mg of either III-VAp or were run on a Novex Tris-glycine polyacrylamide gel, transferred to an Immobilon-P SQ transfer membrane (Millipore), and labeled with antibodies to HLA-DR (clone L243, BioLegend), [mouse IgG2a (clone 1B1), a gift from Harold Jennings], or VAp (antiserum from VAp-immunized mice) to reveal the association of HLA-DR with the glycan p -peptide. 1. Schroter, C.J. et al. A rapid method to separate endosomes from lysosomal contents using differential centrifugation and hypotonic lysis of lysosomes. Journal of Immunological Methods 227, (1999). 2. Cobb, B.A., Wang, Q., Tzianabos, A.. & Kasper, D.L. Polysaccharide processing and presentation by the MHCII pathway. Cell 117, (24). 3. Duan, J., Avci, F.A. & Kasper, D.L. Microbial carbohydrate depolymerization by antigen-presenting cells: Deamination prior to presentation by the MHCII pathway. Proc Natl Acad Sci USA 15, 5183 (28).