Pre-vaccination inflammation and B-cell signalling predict age-related hyporesponse to hepatitis B vaccination

Slim Fourati, Razvan Cristescu, Andrey Loboda, Aarthi Talla, Ali Filali, Radha Railkar, Andrea K Schaeffer, David Favre, Dominic Gagnon, Yoav Peretz, I-Ming Wang, Chan R Beals, Danilo R Casimiro, Leonidas N Carayannopoulos, Rafick-Pierre Sékaly, Slim Fourati, Razvan Cristescu, Andrey Loboda, Aarthi Talla, Ali Filali, Radha Railkar, Andrea K Schaeffer, David Favre, Dominic Gagnon, Yoav Peretz, I-Ming Wang, Chan R Beals, Danilo R Casimiro, Leonidas N Carayannopoulos, Rafick-Pierre Sékaly

Abstract

Aging is associated with hyporesponse to vaccination, whose mechanisms remain unclear. In this study hepatitis B virus (HBV)-naive older adults received three vaccines, including one against HBV. Here we show, using transcriptional and cytometric profiling of whole blood collected before vaccination, that heightened expression of genes that augment B-cell responses and higher memory B-cell frequencies correlate with stronger responses to HBV vaccine. In contrast, higher levels of inflammatory response transcripts and increased frequencies of pro-inflammatory innate cells correlate with weaker responses to this vaccine. Increased numbers of erythrocytes and the haem-induced response also correlate with poor response to the HBV vaccine. A transcriptomics-based pre-vaccination predictor of response to HBV vaccine is built and validated in distinct sets of older adults. This moderately accurate (area under the curve≈65%) but robust signature is supported by flow cytometry and cytokine profiling. This study is the first that identifies baseline predictors and mechanisms of response to the HBV vaccine.

Figures

Figure 1. Study design and antibody titres…
Figure 1. Study design and antibody titres for the three vaccines used in the EM131 study.
(a) Schematic representation of the study design indicating blood collections and assays performed. All analysed participants were HBV-naive at the time of recruitment and received vaccines for hepatitis A/B, Cholera and Tetanus/Diphtheria. (b) Response plots showing antibody titres for HBsAg (hep. B), Diphtheria toxin (dip.), Tetanus toxin (tet.) and Cholera toxin (chol.) for the 174 study participants as a function of the vaccination status (x-axis: before and after vaccination). Red horizontal lines indicate standard titre thresholds and the percentage of participants above the protective thresholds are indicated above each plot (Protect. Ab.).
Figure 2. Development of the BioAge signature…
Figure 2. Development of the BioAge signature and application to the EM131 cohort.
(a) Expression of the 2,285 age-related transcripts derived from SAFHS data set (n=1,240); all transcripts correlating with chronological age (moderated t-test: adjusted P≤0.05) were clustered in 20 modules using k-means clustering. In the heat map, transcripts (columns) were ordered by their membership to the 20 modules; the 1,240 samples/participants (row) were ordered by their BioAge score (signed average expression of the age-related transcripts). Transcript expression was transformed to z-score and is depicted in blue to white to red colour scale. The chronological age is given in the bar plot at the right. (b) Expression of the 2,285 transcripts of the BioAge signature in the EM131 data set (n=174). The chronological age of the 174 participants is given in the bar plot at the right. (c) Scatter plot showing the chronological age as a function of the BioAge. The vertical line indicates a BioAge score of 0. We observe that young participants (red circles) have significantly younger BioAge; and among elderly (black circles) about half of participants have a young BioAge (<0) while the other half have an old BioAge (oAge).
Figure 3. BioAge predicts HBV vaccine response.
Figure 3. BioAge predicts HBV vaccine response.
(a) BioAge classification of the 135 elderly patients based on gene expression prior to vaccination is significantly associated with the response to HBV vaccination. Each bar of the bar plot represents one of the elderly participants in the EM131 cohort. The height of the bar indicates the BioAge score of that donor. Bars were ordered by increasing level of the BioAge, separately for the HBV poor-responders (in red) and HBV responders (in black). Fisher's exact test P values are given on the plot. (b) Forward selection among the modules composing the BioAge signature resulted in the selection of modules M1 and M16 as optimal signatures, predicting HBV vaccine response in the EM131 elderly cohort (age≥65). (c) Pathway enrichment analysis on the genes included in modules M1 and M16 using the IPA canonical pathway database. Fisher exact test was performed to assess statistical enrichment and gene sets with FDR-corrected P value≤0.05 are presented.
Figure 4. Identification of gene-expression signature predicting…
Figure 4. Identification of gene-expression signature predicting the HBV vaccine response.
(a) Expression of 15 genes identified as predictors of the response to the HBV vaccine in the EM131 training set (n=95). The mean-centred gene expression is represented using a blue to white to red colour scale. Rows and columns correspond to the genes and the profiled samples, respectively. Samples were ordered by increasing levels of their predicted probability of responding to the vaccine (that is, posterior probability). Antibody response to the HBV vaccine (log(HepBDifV4/5-V2)) and the response group predicted by the 15-gene signature (nbClass) are presented in coloured squares above each sample. (b) Expression of the 15-gene signature on the EM131 test set (n=49). (c) ROC curve and area under the curve (AUC) for the prediction of the HBV vaccine response using the 15-gene signature on the EM131 test set (n=49). (d) Network inference based on the 15 markers identified as predictors of the response to the HBV vaccine. Red and blue nodes represent genes induced or repressed in HBV vaccine responders (R) compared with poor-responders (PR), respectively. (e) Networks were inferred for the BioAge module 1 combined to the 15-gene signature and the BioAge module 16 combined to the 15-gene signature, respectively. Nodes included in the BioAge or the 15-gene signature are coloured by their fold-change between R versus PR to the HBV vaccine in the EM131 training set.
Figure 5. B-cell subset as well as…
Figure 5. B-cell subset as well as innate immune cell subsets and RBC counts are predictive of the response to the HBV vaccine.
Polychromatic FCM was used to identify cell surface markers of the response to the HBV vaccine. (a) Box plots presenting the frequencies of the four FCM markers selected in the forward selection model in responders and poor-responders to the HBV vaccine on the EM131 training set (n=95). (b) ROC curves for the prediction of the HBV vaccine response using the FCM data on the EM131 test set. (c) Box plots presenting the RBC counts (measured in a 28-day window prior to HBV vaccination) in responders and poor-responders to the HBV vaccine on the EM131 training set (n=95). (d) ROC curves illustrating the prediction of the HBV vaccine response using RBC counts on the EM131 test set. (e) Heat map representation of the genes differentially expressed between HBV vaccine R and PR overlapping the HIF-1a canonical pathway in the training set. The mean-centred gene expression is represented using a blue–red colour scale. Rows and columns correspond to the genes and the profiled samples, respectively. Samples were ordered by increasing level of their expression of the genes associated to the vaccine response (mean-rank ordering). Antibody response to the HBV vaccine (logHepBDifV4/5-V2) and HBV vaccine response group (HBV vaccine response) and red blood cells counts (RBC in 1012 per l) are presented in coloured squares above each sample. The P value of t-test between RBC and the ordering of the samples is 0.0462.
Figure 6. Integrative analysis reveals positive correlations…
Figure 6. Integrative analysis reveals positive correlations between biomarkers of HBV vaccine response.
A projection-based multivariate approach was adopted to assess the correlation between transcriptomic, FCM, haematologic and cytokine/chemokine expression data sets. Least-square regressions between pairs of data sets was performed using the R package ‘mixOmics'. The resulting regression coefficients were converted to Pearson's correlations between pairs of features of the different data sets (presented at each quadrant of the figure). Significant (t-test: r≥0.188, P≤0.05) positive correlations are presented as edges and the features as vertices. The vertices are coloured by fold-change between HBV vaccine responders versus poor-responders of the EM131 training set.

References

    1. Gavazzi G. & Krause K. H. Ageing and infection. Lancet Infect. Dis. 2, 659–666 (2002) .
    1. Department of Economic and Social Affairs Population Division. World Population Ageing 2013. Report No. ST/ESA/SER.A/348 (United Nations, New York, NY, USA, 2013) .
    1. Clements M. L. et al. Effect of age on the immunogenicity of yeast recombinant hepatitis B vaccines containing surface antigen (S) or PreS2+S antigens. J. Infect. Dis. 170, 510–516 (1994) .
    1. Hohler T. et al. Differential genetic determination of immune responsiveness to hepatitis B surface antigen and to hepatitis A virus: a vaccination study in twins. Lancet 360, 991–995 (2002) .
    1. Weihrauch M. R. et al. T cell responses to hepatitis B surface antigen are detectable in non-vaccinated individuals. World J. Gastroenterol. 14, 2529–2533 (2008) .
    1. Goronzy J. J. & Weyand C. M. Understanding immunosenescence to improve responses to vaccines. Nat. Immunol. 14, 428–436 (2013) .
    1. Seok J. et al. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc. Natl Acad. Sci. USA 110, 3507–3512 (2013) .
    1. Pulendran B., Li S. & Nakaya H. I. Systems vaccinology. Immunity 33, 516–529 (2010) .
    1. Gaucher D. et al. Yellow fever vaccine induces integrated multilineage and polyfunctional immune responses. J. Exp. Med. 205, 3119–3131 (2008) .
    1. Querec T. D. et al. Systems biology approach predicts immunogenicity of the yellow fever vaccine in humans. Nat. Immunol. 10, 116–125 (2009) .
    1. Obermoser G. et al. Systems scale interactive exploration reveals quantitative and qualitative differences in response to influenza and pneumococcal vaccines. Immunity 38, 831–844 (2013) .
    1. Tan Y. et al. Gene signatures related to B-cell proliferation predict influenza vaccine-induced antibody response. Eur. J. Immunol. 44, 285–295 (2014) .
    1. Li S. et al. Molecular signatures of antibody responses derived from a systems biology study of five human vaccines. Nat. Immunol. 15, 195–204 (2014) .
    1. Furman D. et al. Systems analysis of sex differences reveals an immunosuppressive role for testosterone in the response to influenza vaccination. Proc. Natl Acad. Sci. USA 111, 869–874 (2014) .
    1. Tsang J. S. et al. Global analyses of human immune variation reveal baseline predictors of postvaccination responses. Cell 157, 499–513 (2014) .
    1. Atkinson W. L. et al. General recommendations on immunization. Recommendations of the advisory committee on immunization practices (ACIP) and the American academy of family physicians (AAFP). MMWR Recomm. Rep. 51, 1–35 (2002) .
    1. Scerpella E. G. et al. Serum and intestinal antitoxin antibody responses after immunization with the whole-cell/recombinant B subunit (WC/rBS) oral cholera vaccine in North American and Mexican volunteers. J. Travel Med. 3, 143–147 (1996) .
    1. Van Buren R. C. & Schaffner W. Hepatitis B virus: a comprehensive strategy for eliminating transmission in the United States through universal childhood vaccination: Recommendations of the immunization practices advisory committee (ACIP). MMWR Recomm. Rep. 40, 1–19 (1991) .
    1. Goring H. H. et al. Discovery of expression QTLs using large-scale transcriptional profiling in human lymphocytes. Nat. Genet. 39, 1208–1216 (2007) .
    1. Ioannidis V., Beermann F., Clevers H. & Held W. The beta-catenin--TCF-1 pathway ensures CD4(+)CD8(+) thymocyte survival. Nat. Immunol. 2, 691–697 (2001) .
    1. Panda A. et al. Human innate immunosenescence: causes and consequences for immunity in old age. Trends Immunol. 30, 325–333 (2009) .
    1. Teitell M. A. OCA-B regulation of B-cell development and function. Trends Immunol. 24, 546–553 (2003) .
    1. Anelli T. & van Anken E. Missing links in antibody assembly control. Int. J. Cell Biol. 2013, 606703 (2013) .
    1. Parish S. T., Wu J. E. & Effros R. B. Sustained CD28 expression delays multiple features of replicative senescence in human CD8 T lymphocytes. J. Clin. Immunol. 30, 798–805 (2010) .
    1. Takagi H. et al. Plasmacytoid dendritic cells are crucial for the initiation of inflammation and T cell immunity in vivo. Immunity 35, 958–971 (2011) .
    1. Liquet B., Le Cao K. A., Hocini H. & Thiebaut R. A novel approach for biomarker selection and the integration of repeated measures experiments from two assays. BMC Bioinformatics 13, 325 (2012) .
    1. Shi L. et al. The microarray quality control (MAQC)-II study of common practices for the development and validation of microarray-based predictive models. Nat. Biotechnol. 28, 827–838 (2010) .
    1. Vieira P. & Rajewsky K. Persistence of memory B cells in mice deprived of T cell help. Int. Immunol. 2, 487–494 (1990) .
    1. Goubau D. et al. Transcriptional re-programming of primary macrophages reveals distinct apoptotic and anti-tumoral functions of IRF-3 and IRF-7. Eur. J. Immunol. 39, 527–540 (2009) .
    1. Lazear H. M. et al. IRF-3, IRF-5, and IRF-7 coordinately regulate the type I IFN response in myeloid dendritic cells downstream of MAVS signalling. PLoS Pathog. 9, e1003118 (2013) .
    1. Monaco C., Andreakos E., Kiriakidis S., Feldmann M. & Paleolog E. T-cell-mediated signalling in immune, inflammatory and angiogenic processes: The cascade of events leading to inflammatory diseases. Curr. Drug Targets Inflamm. Allergy 3, 35–42 (2004) .
    1. Gorczynski R. M. CD200: CD200R-mediated regulation of immunity. ISRN Immunol. 2012, 1–18 (2012) .
    1. Iacobelli M., Wachsman W. & McGuire K. L. Repression of IL-2 promoter activity by the novel basic leucine zipper p21SNFT protein. J. Immunol. 165, 860–868 (2000) .
    1. Glass G. A., Gershon D. & Gershon H. Some characteristics of the human erythrocyte as a function of donor and cell age. Exp. Hematol. 13, 1122–1126 (1985) .
    1. Shaw A. C., Goldstein D. R. & Montgomery R. R. Age-dependent dysregulation of innate immunity. Nat. Rev. Immunol. 13, 875–887 (2013) .
    1. Yolima C. G. et al. Immunogenicity of hepatitis B vaccine in patients with inflammatory bowel disease and the benefits of revaccination. J. Gastroenterol. Hepatol. 30, 92–98 (2014) .
    1. Janssen R. S. et al. Immunogenicity and safety of an investigational hepatitis B vaccine with a toll-like receptor 9 agonist adjuvant (HBsAg-1018) compared with a licensed hepatitis B vaccine in patients with chronic kidney disease. Vaccine 31, 5306–5313 (2013) .
    1. Mannick J. B. et al. mTOR inhibition improves immune function in the elderly. Sci. Transl. Med. 6, 268ra179 (2014) .
    1. Treadwell T. L. et al. Immunogenicity of two recombinant hepatitis B vaccines in older individuals. Am. J. Med. 95, 584–588 (1993) .
    1. Thoelen S. et al. The first combined vaccine against hepatitis A and B: an overview. Vaccine 17, 1657–1662 (1999) .
    1. Svennerholm A. M., Holmgren J., Black R., Levine M. & Merson M. Serologic differentiation between antitoxin responses to infection with vibrio cholerae and enterotoxin-producing escherichia coli. J. Infect. Dis. 147, 514–522 (1983) .
    1. Gentleman R. C. et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 5, R80 (2004) .
    1. Hastie T. et al. Imputing missing data for gene expression arrays Technical report, Division of biostatistics, Stanford University (1999) .
    1. Smyth G. K. in Bioinformatics and Computational Biology Solutions Using R and Bioconductor 397–420Springer (2005) .
    1. Benjamini Y. & Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57, 289–300 (1995) .
    1. Tibshirani R., Walther G. & Hastie T. Estimating the number of clusters in a data set via the gap statistic. J. R. Stat. Soc. Ser. B 63, 411–423 (2001) .
    1. Irizarry R. A. et al. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4, 249–264 (2003) .
    1. Merico D., Isserlin R., Stueker O., Emili A. & Bader G. D. Enrichment map: A network-based method for gene-set enrichment visualization and interpretation. PLoS ONE 5, e13984 (2010) .

Source: PubMed

Sponsors and Collaborators

Medical Conditions

Drug Interventions

3
Subscribe