A novel chimpanzee adenovirus vector with low human seroprevalence: improved systems for vector derivation and comparative immunogenicity

Abstract

Recombinant adenoviruses are among the most promising tools for vaccine antigen delivery. Recently, the development of new vectors has focused on serotypes to which the human population is less exposed in order to circumvent pre-existing anti vector immunity. This study describes the derivation of a new vaccine vector based on a chimpanzee adenovirus, Y25, together with a comparative assessment of its potential to elicit transgene product specific immune responses in mice. The vector was constructed in a bacterial artificial chromosome to facilitate genetic manipulation of genomic clones. In order to conduct a fair head-to-head immunological comparison of multiple adenoviral vectors, we optimised a method for accurate determination of infectious titre, since this parameter exhibits profound natural variability and can confound immunogenicity studies when doses are based on viral particle estimation. Cellular immunogenicity of recombinant E1 E3-deleted vector ChAdY25 was comparable to that of other species E derived chimpanzee adenovirus vectors including ChAd63, the first simian adenovirus vector to enter clinical trials in humans. Furthermore, the prevalence of virus neutralizing antibodies (titre >1:200) against ChAdY25 in serum samples collected from two human populations in the UK and Gambia was particularly low compared to published data for other chimpanzee adenoviruses. These findings support the continued development of new chimpanzee adenovirus vectors, including ChAdY25, for clinical use.

Figure 1. Phylogenetic trees based on alignment of nucleotide sequences of (A) the hexon protein and (B) the fiber protein of different human (HAdV-) and chimpanzee (SAdV-) adenovirus serotypes including Y25.

Note that SAdV-3 (species Simian adenovirus A) is included here as an outgroup, having been derived from a Rhesus Macaque. Clustering of sequences into each of the species Human adenovirus A-F is indicated below. Bootstrap values for branches indicated.

Figure 2. Generation of a molecular clone of Y25 by gap repair insertion of Y25 genomic DNA into the pBAC ‘rescue vector’.

Recombination between BAC vector and the left side of the viral genome can either occur at LFI such that the E1 region is included in the resulting BAC clone, or more desirably at LFII such that E1 is deleted.

Figure 3. Modification of the Y25 adenovirus E4 region to increase yield and hexon expression for titration.

(A) Modified Y25 vectors labelled ChAdY25-A to –E. (B) Virus yield of Y25 based vectors expressing GFP before and after E4Orf6 replacement. HEK293 cells were infected at a multiplicity of infection of 10 ifu/cell and incubated at 37°C for 48hours before harvesting. Infectious titer of the harvested material was measured by quantifying GFP positive foci 48 hrs post infection. (C) Ratio of GFP titer to anti-hexon titer for different Y25 E4 modified vectors expressing the TIPeGFP antigen. Titers assessed 48 hrs post infection. (D) Ratio of transgene: hexon expression for ChAdY25-E based vectors expressing TIPeGFP and Influenza Matrix protein 1 (M1) recombinant transgenes. All data is representative of at least two independent experiments. Error bars show mean and SEM.

Figure 4. Infectivity of recombinant adenovirus vector preparations can affect immunogenicity.

(A–C) IFN-γ Spleen ELISpot responses after immunisation with HAdV-5 TIPeGFP doses based on viral particles (VP) or infectious units (IU). Balb/c mice were immunised intramuscularly with either (A) 10⁷ viral particles or (B) 10⁷ infectious units of one of two independent preparations of HAdV-5 TIPeGFP. Responses to CD8⁺ epitope Pb9 were assayed two weeks post immunisation. Asterisk indicates p<0.05. (C) Titers per ml and particle to infectious unit (P:I) ratios of the two CsCl-band purified vector preparations. (D) Immunogenicity of TIPeGFP expressing Y25 based vectors with different E4 modifications is comparable despite the vectors having different P:I ratios. Balb/c mice were immunised intramuscularly with 10⁸ infectious units of Y25 vectors with either i) a native E4 locus (ChAdY25) ii) Ad5E4Orf6 expressed at E4 (ChAdY25-A) or iii) the clinical ChAdY25-E vector. IFN-γ Spleen ELISpot was performed as in A-C. P:I ratios of vector preparations were as follows; ChAdY25 131, ChAdY25-A 17.4, ChAdY25-E 13.4.

Figure 5. Cellular immunogenicity of ChAdY25 is robust and comparable to current chimpanzee adenovirus vectors AdC68 and AdC63.

Balb/c mice were immunised with 10⁹ infectious units of AdC68TIPeGFP, AdC63TIPeGFP or ChAdY25-A TIPeGFP. Two weeks post immunisation, spleen immunogenicity against (A) dominant CD8⁺ epitope (Pb9) and (B) CD4⁺ epitope (P15) was assessed by IFNγ ELISpot. P:I ratios of vector preparations were as follows; ChAd63 24.9, ChAdY25-A 17.41, AdC68 31.6.

Figure 6. Immunogenicity of SAdV vectors boosted by MVA.

Balb/c mice were immunised with 10⁶ infectious units of AdC68-TIPeGFP, AdvY25TIPeGFP, or ChAd63 TIPeGFP. After 56 days post prime, mice were boosted with 10⁶ pfu MVA-TIPeGFP. Blood was collected A, 55 post prime and B, 7 days post boost. Blood IFNγ⁺ CD8⁺ T cell responses were measured by intracellular cytokine staining (ICS) after stimulation with Pb9 peptide. No statistical significance was found between any of the groups by one way ANOVA.

Figure 7. Anti-vector neutralising antibody titer in human sera from the UK (100 samples) and The Gambia (57 samples) against Y25.