Vaccine 33 (2015) 1614–1619

Contents lists available at ScienceDirect

Vaccine journal homepage: www.elsevier.com/locate/vaccine

Preparation and testing of a Haemophilus influenzae Type b/Hepatitis B surface antigen conjugate vaccine So Jung An, Joo Sung Woo, Myung Hwa Chae, Sudeep Kothari, Rodney Carbis ∗ Vaccine Development Section, International Vaccine Institute, SNU Research Park, 1 Gwanak-ro, Gwanak-gu, Seoul 151-742, Republic of Korea

a r t i c l e

i n f o

Article history: Received 3 December 2014 Received in revised form 6 January 2015 Accepted 26 January 2015 Available online 7 February 2015 Keywords: Haemophilus influenzae Type b Hepatitis B Polyribosylribitol phosphate (PRP) Hepatitis B surface antigen (HBsAg) Conjugate vaccine

a b s t r a c t The majority of conjugate vaccines focus on inducing an antibody response to the polysaccharide antigen and the carrier protein is present primarily to induce a T-cell dependent response. In this study conjugates consisting of poly(ribosylribitolphosphate) (PRP) purified from Haemophilus influenzae Type b bound to Hepatitis B virus surface antigen (HBsAg) virus like particles were prepared with the aim of inducing an antibody response to not only the PRP but also the HBsAg. A conjugate consisting of PRP bound to HBsAg via an adipic acid dihydrazide (ADH) spacer induced strong IgG antibodies to both the PRP and HBsAg. When conjugation was performed without the ADH spacer the induction of an anti-PRP response was equivalent to that seen by conjugate with the ADH spacer, however, a negligible anti-HBsAg response was induced. For comparison, PRP was conjugated to diphtheria toxoid (DT) and Vi polysaccharide purified from Salmonella Typhi conjugated to HBsAg both using an ADH spacer. The PRPAH –DT conjugate induced strong anti-PRP and anti-DT responses, the Vi–AH HBsAg conjugate induced a good anti-HBsAg response but not as strong as that induced by the PRPAH –HBsAg conjugate. This study demonstrated that in mice it was possible to induce robust antibody responses to both polysaccharide and carrier protein provided the conjugate has certain physico-chemical properties. A PRPAH –HBsAg conjugate with the capacity to induce anti-PRP and anti-HBsAg responses could be incorporated into a multivalent pediatric vaccine and simplify formulation of such a vaccine. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Combination vaccines are an important contribution to reducing the number of injections that children receive during the course of the recommended immunization schedules and are now the preferred option in pediatric and adult immunization programs. Pediatric combination vaccines started with three components, diphtheria, tetanus and pertussis, since then Hepatitis B, Haemophilus influenzae Type b (Hib) and inactivated polio antigens have been added in various combinations [1]. Formulating combination vaccines is not as simple as just mixing the various single component vaccines together; a number of technical challenges need to be overcome. The components in the combination should induce at least an equivalent antibody response to that of the individual vaccine and the safety of the combination must be no worse than that of the individual vaccine components. Combining vaccines into one formulation can result in immune interference, combinations containing Hib

∗ Corresponding author. Tel.: +82 2 881 1169. E-mail address: [email protected] (R. Carbis). http://dx.doi.org/10.1016/j.vaccine.2015.01.061 0264-410X/© 2015 Elsevier Ltd. All rights reserved.

conjugate and acellular pertussis have been shown to induce lower antibody responses to Hib PRP than those obtained with Hib conjugate administered separately [2,3]. Interference was not obvious when the pertussis component consisted of whole cells, it was suggested that the whole cells provided an adjuvant effect [2]. A further difficulty in combination vaccine formulation is the presence of aluminum based adjuvants which have been used by some manufacturers to adsorb diphtheria, tetanus and acellular pertussis antigens. Animal experiments have demonstrated reduced anti-PRP responses when Hib conjugate is adsorbed on aluminum hydroxide [4]. Pentavalent vaccines (DTwP–HepB–Hib) containing whole cell pertussis and produced for immunization in developing countries are mainly liquid formulations containing all five components [5]. Multivalent vaccines that contain acellular pertussis usually supply the Hib component as a separate powder which is reconstituted with the liquid components immediately prior to vaccination thus adding to the complexity of delivering vaccine. A number of Hib conjugate vaccines have now been licensed and they differ from each other in a number of ways including the type of carrier protein, the size of the polysaccharide, the conjugation chemistry and the manufacturing processes [6]. Carrier proteins used include diphtheria and tetanus toxoids, CRM197 (non

S.J. An et al. / Vaccine 33 (2015) 1614–1619

toxic mutant form of diphtheria toxin) and OMPC (outer membrane protein complex of Neisseria meningitides B). The primary purpose of the carrier protein is to improve the immunogenicity of the PRP polysaccharide by induction of a T-cell dependent response [7]. Some vaccine developers have used carrier proteins derived from the same bacteria as the polysaccharide to enhance antibacterial immunity [8,9], thus augmenting the role of the carrier protein. In this study Hepatitis B surface antigen (HBsAg) was used as the carrier protein for a Hib (PRP) conjugate. This conjugate has the potential to simplify manufacture and formulation of multivalent pediatric vaccines. 2. Materials and methods 2.1. Bacterial strain and cultivation in a bioreactor Haemophilus influenzae Type b (Hib) strain ATCC 31441 was grown in a Biostat B-DCU bioreactor (Sartorius-Stedim) using fed batch conditions using media developed by Instituto Butantan [10]. The starting media (10 g/L soy peptone (Becton Dickinson and Co.), 5.0 g/L dialyzed yeast extract (Oxoid), 5.0 g/L glucose (Sigma), 13.1 g/L Na2 HPO4 (USB), 3.3 g/L NaH2 PO4 (USB), 2.5 g/L K2 HPO4 (USB), 15 mg/L nicotinamide adenine dinucleotide (NAD) (Sigma) and 30 mg/L hemin (Fluka)) pH 7.5 was sterilized by filtration (Sartopore 2300 0.2 ␮m filter (Sartorius-Stedim)). Feeding (100 g/L soy peptone, 50 g/L dialyzed yeast extract, 100 g/L glucose, 150 mg/L NAD and 300 mg/L hemin) was initiated at 6 h, at a constant rate of 54 mL/h, and fermentation was allowed to continue to the end of the growth phase, 12 h from the start of fermentation. Feeding was continued and the cells held in stationary phase for an additional 2 h. pH was controlled at 7.5, dissolved oxygen at 30% and the temperature at 37 ◦ C. 3.36 L of fermentation broth was harvested (2.9 L starting media plus seed culture and 0.46 L of added feed solution, base and antifoam). Cells were inactivated by addition of formaldehyde to a final concentration of 0.15% and held overnight at 4 ◦ C before processing. Cell growth was monitored by optical density at 600 nm (OD600 ) (UV–vis spectrophotometer (Beckman model DU 530)). 2.2. Purification of PRP PRP was purified using the method developed for the purification of Vi from Salmonella Typhi [11] with the following modifications: Precipitation of PRP with cetavlon (Sigma) was at 0.67% cetavlon instead of 0.5% and the PRP was dissolved in 50% ethanol (Duksan) in 50 mM sodium acetate (USB) rather than 60% ethanol. 2.3. Preparation of conjugates 2.3.1. Preparation of PRP–HBsAg and PRP–DT conjugates Purified PRP was activated with 1-cyano-4-dimethylaminopyridinium tetrafluoroborate (CDAP) (Sigma) and derivatized with adipic acid dihydrazide (ADH) (Sigma) using a modification of the method described previously [9] and briefly was performed as follows: PRP in purified water was adjusted to pH 5.2, CDAP (100 mg/mL in acetonitrile) was added to a final concentration of 2.5 mg/mL and held at room temperature (RT) for 2 min. The pH was increased to 8 by addition of 0.2 M triethylamine (TEA) (Sigma), ADH (90 mg/mL in 0.1 M NaHCO3 ) (Sigma) was added to the reaction mixture to a final concentration of 12.5 mg/mL giving a ratio of PRP:ADH of 1:5 (w/w). The final concentration of PRP in the reaction mixture was 2.5 mg/mL. The reaction was allowed to proceed for 2 h at RT, maintaining the pH between 8.0 and 8.5, the mixture

1615

was then dialyzed (MWCO 6–8 kD, Spectrum Laboratories) against phosphate buffered saline (PBS) overnight at 4 ◦ C. HBsAg (Commercially available) was conjugated to PRP as follows: 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) (20 mg/mL in 80 mM 2-(N-morpholino)ethanesulfonic acid (MES) (Sigma) pH 5.6) was added to HBsAg in 80 mM MES, the mixture was added to derivatized PRP so that final concentrations of PRPAH :HBsAg:EDC were 1:1:2 mg/mL, respectively. The reaction proceeded for 3 h at RT maintaining the pH at 5.7. Residual EDC was removed by dialysis against PBS overnight at 4 ◦ C. Conjugate without a spacer was prepared using the following method: The PRP was activated with CDAP as described above but instead of adding ADH, HBsAg was added to a final concentration of 1 mg/mL with the final concentrations in the reaction mixture being 1:1:7.5 mg/mL (w/w) for PRP:HBsAg:CDAP, respectively. The conjugation reaction was 2 h at RT with pH maintained at 8.0, the mixture was dialyzed against PBS. PRPAH –DT (with spacer) and PRP–DT (without spacer) conjugates using Diphtheria Toxoid (DT) as the carrier protein were prepared in the same way as conjugates using HBsAg carrier protein. The final step for all conjugates was diafiltration against 10 volume changes of PBS pH 7.2 using a 300 kD polyethersulfone membrane (Sartorius-Stedim). 2.3.2. Preparation of Vi–AH HBsAg and Vi–AH DT conjugates HBsAg and DT were derivatized as previously described [12]. Briefly the HBsAg or DT was derivatized by binding ADH to the protein in the presence of EDC, final concentrations of protein, ADH and EDC were 3.0:10.5:1.2 mg/mL for HBsAg and 10.0:35.0:4.0 mg/mL for DT. Reaction time was 60 min at RT and pH was maintained at 5.7. Residual EDC was removed by dialysis against PBS. Conjugates were prepared as previously described [12] with the reaction mixture containing final concentrations of 1 mg/mL derivatized protein (DT or HBsAg), 1 mg/mL of Vi and 2 mg/mL of EDC in MES at pH 5.7. Reaction time was 3 h at RT, and residual EDC was removed by dialysis against PBS. 2.4. Chemical and physical analysis of conjugates Vi and Vi conjugates were assayed for O-acetyl by Hestrin assay [13] and converted to mg/mL using a Vi standard (1 mg/mL), the polysaccharide content of PRP and PRP conjugates was assayed by ribose assay [14], and compared to a standard PRP purchased from the National Institute for Biological Standards and Control (NIBSC UK), ADH concentration by TNBS assay [15], protein content by Lowry assay [16], nucleic acid by ultraviolet spectroscopy [17] and molecular size by size exclusion chromatography using Sephacryl S-1000 (GE Healthcare) in 0.2 M NaCl pH 7.0. NMR was performed by the Inter-University Research Facilities (NCIRF) of Seoul National University, Republic of Korea using a method previously described [18]. 2.5. Immunogenicity of conjugates Groups of 10 ICR 6 week old female mice were injected subcutaneously with each of the conjugate preparations or un-conjugated controls. The dose of polysaccharide per injection was 1.0 ␮g for PRP and 2.5 ␮g for Vi, equivalent to 1/10 of a human dose. The dose of protein varied from conjugate to conjugate and is presented in Table 2. Mice received 3 doses at 0, 4 and 8 weeks and were bled by retro-orbital puncture at 2, 6 and 10 weeks. Titers of serum IgG were assayed by enzyme-linked immunosorbent assay (ELISA). A Mouse Anti-Hib-PRP IgG ELISA kit (Alpha Diagnostic International) was used to determine anti-PRP IgG

1616

S.J. An et al. / Vaccine 33 (2015) 1614–1619

Fig. 1. Partial 500 MHz 1 H NMR spectra of PRP polysaccharide in D2 O. (a) NIBSC (b) IVI.

levels which were calculated against a standard supplied with the kit. For PRP only, ELISAs were performed in triplicate on a pool of sera (equal volumes for each mouse) for each group, the small volumes of sera and the low titers (limiting dilution) preventing testing of individual serum samples, for the other antigens individual mouse sera were tested. The ELISAs used to measure anti-Vi and anti-DT IgG were as described previously [19] and the measurement of anti HBsAg used a modified Vi ELISA. Briefly 100 ␮l of either HBsAg (0.5 ␮g/mL), Vi (2 ␮g/mL) or DT (5 ␮g/mL) in PBS, pH 7.4 was used to coat 96-well plates and incubated overnight at RT. Plates were then blocked with PBS containing 1% BSA for 4 h at 37 ◦ C. Sera were 2 fold serially diluted across the plates. Alkaline phosphatase-conjugated rat anti-mouse IgG in PBS with 1% BSA and 0.1% Brij35 was added to each well (100 ␮l/well). pNitrophenyl phosphate (Sigma) 1 mg/mL in 1 M Tris–HCl buffer, pH 9.8 was used to develop the color. The anti-Vi, anti-DT and antiHBsAg titers were calculated using Program ELISA for Windows (Centers for Disease Control and Prevention) [20] and expressed as the GM of ELISA unit (EU). Hyperimmune mouse serum pools against Salmonella Typhi or DT was used as standards for the Vi and DT ELISAs and monoclonal mouse anti-HBsAg was used as the reference for the HBsAg ELISA. The reference sera were all assigned a value of 100 EU and OD405 lower than 0.15 was assigned a value of 0.02 EU.

2.6. Statistical analysis Student’s t-test, Welch’s t-test or Kolmogorov–Smirnov test was used for the comparison of the anti-HBsAg, anti-DT and anti-Vi IgG immune responses between the conjugates and controls depending on whether the variance of data was equal or not and the distribution of data. Nonparametric and median test (two-sided) was used for analysis of anti-PRP IgG data. P values