Posted in influenza, Resource

Systems Analysis of Immunity to Influenza Vaccination across Multiple Years and in Diverse Populations Reveals Shared Molecular Signatures

Figure showing that the number of DEGs related to innate immune responses are more highly expressed in the young compared to the elderly subjects after receiving the Influenza TIV. Source from Nakaya et al., 2015, Immunity.

Although vaccination is considered the most effective method for preventing influenza, it shows limited efficacy in the elderly. Here, Nakaya et al used a systems vaccinology approach to understand the mechanisms behind poor vaccine efficacy in the inactivated influenza vaccine (TIV) in the elderly.

212 individuals from the current study and 218 individuals from a previously published study (Franco et al., 2013) were included in analyses. 54 of these were elderly (>65 years old). As expected, antibody responses to the Influenza vaccine (TIV) decrease with age.

Blood Transcriptomic Modules related to the induction of interferons and activation of dendritic cells were enriched on days 1 and 3 after TIV vaccination, whereas modules related to T cells at these time points were negatively associated with the antibody response. On day 7, there was a robust enrichment of antibody secreting cells (ASC) and cell cycle-related modules. The enriched modules in young and elderly were similar, particularly those related to the interferon response and activation of dendritic cells on day 1. However, the magnitude of expression of interferon-related genes was significantly higher in young individuals (See top figure).

Combining all datasets, several B-cell- and T-cell-related modules at pre-vaccination was positively correlated with an increased antibody response to vaccination. In contrast, modules related to monocytes were negatively correlated with antibody responses, supporting the concept that inflammatory responses at baseline might be detrimental to the induction of vaccine-induced antibody responses.

Consistent with the neutralizing antibody responses, B-cell and plasmablast modules (BTM S3) showed reduced expression in the elderly compared to the young on day 7. However, Natural killer (NK) cell and monocyte modules were enriched in the elderly at day 3 and day 7 after vaccination.

To examine if the transcriptional changes were due to changes in these specific cell types, flow cytometry was performed. Proportions of total NK cells in elderly subjects were higher than those of young subjects at baseline and all time points studied (days 0–14). The NK-cell activation markers were also more prominent in the elderly. Similarly, increased quantities of monocytes were seen in the elderly, with higher CCR5 expression at all time-points tested.

Differential expression of miRNA is also evident between the elderly and young, which suggests that miRNA could be important regulators of the immune response to influenza vaccination.

Data is deposited in GEO as GSE74817.

Posted in influenza, Resource

High-Resolution Temporal Response Patterns to Influenza Vaccine Reveal a Distinct Human Plasma Cell Gene Signature

Antibody titers, cell subset numbers, and gene transcript expression in different subjects across the different subjects. Source from Henn AD. et al., Scientific Reports, 2013

Influenza vaccines produce highly variable B cell responses among individuals, making it difficult to predict who will achieve protective antibody titers after vaccination.

In this paper by Henn AD et al., 2013, daily sampling of serum, peripheral blood mononuclear cells (PBMC), B cells and plasma cells from 14 human subjects was performed over 11 days post-influenza inactivated vaccine administration. Peripheral blood was drawn during the week prior to vaccination (pre-V), immediately before vaccination (day 0), daily for days 1–10 and on day 21 post-vaccination.

Most differentially expressed genes were detected at days 5-6 post vaccination, and this corresponded with the heightened IgM, IgG, plasmablast and activated ASC responses in most subjects. Many of these transcripts were validated to be B cell differentiation genes.

742 genes were differentially regulated temporally, and the majority of these genes were significantly correlated with CD27hiCD38hiCD138− plasmablasts. These genes are termed as the plasma cell gene signature (PCgs).

Ten of the top 30 categories of functionally related genes in the PCgs involved ER function and protein production. These findings are consistent with involvement of the PCgs in program-level upregulation of antibody production machinery and the unfolded protein response seen during plasma cell development.

Of interest, the other genes involved in the PCgs were expressed by the myeloid/DC lineages, many of which peaked at day 1. This is consistent with the notion that the magnitude of the innate immune response is also associated with antibody responses from influenza vaccination.

Posted in influenza, Resource

Systems biology of immunity to MF59-adjuvanted versus nonadjuvanted trivalent seasonal influenza vaccines in early childhood

Figure showing that the differentially expressed genes (DEGs) were significantly higher in adults compared to children under 2 years of age. Overall, the adjuvanted inactivated influenza vaccine generated more DEGs than the unadjuvanted counterpart. Source from Nakaya et al., PNAS, 2016.

The trivalent inactivated influenza vaccine (TIV) is poorly immunogenic and has low effectiveness under 2 years of age. In this study by Nakaya et al., 2016 , the investigators studied the innate, adaptive and molecular responses to the seasonal TIV and MF59-adjuvanted TIV (ATIV) in 90 children from 14- to 24-months of age.

At day 28 post-boost, HAI geometric mean titers were higher in ATIV vaccinees compared with TIV. However, the magnitude of the plasmablast response was much lower in children than in adults. TIV and ATIV induced a similar magnitude of IgM- and IgG-secreting plasmablast cells in children.

MF59-adjuvanted TIV vaccine induced a higher expansion of multicytokine-producing vaccine-specific CD4+ T cells, mostly producing TNF-α and IL-2.

ATIV induced more alterations in gene expression at days 1, 3, and 7 post-boost compared with TIV. However, the numbers of DEGs were much smaller than in an adult cohort. There was high heterogeneity in the individual host responses, which could have accounted for the fewer DEGs detected.

Gene set enrichment analysis (GSEA) on individual subject’s responses revealed that at day 1 post vaccination, the positively enriched modules are M75 “antiviral interferon signature,” S5 “dendritic cell signature,” M16 “Toll-like receptor (TLR) and inflammatory signaling.” Among the negatively enriched modules, several modules related to T-cell function, NK cells, and cell cycle were found, including M7.1 “T cell activation,” M7.2 “enriched in NK cells,” and M4.0 “cell cycle and transcription”

GSEA was also applied to rank genes based on correlation with HAI titers. These modules include M75 “antiviral interferon signature,” M165 “enriched in activated dendritic cells,” and several others. These modules were positively correlated with HAI response at days 1, 7, and 28 following the booster shot.

The kinetics of enrichment of two blood transcriptomic modules associated with antibody-secreting cells (M156.1 and S3) show that enrichment is higher on days 7 and 28 for ATIV. The M156.1 module “plasma cell, immunoglobulins” was only significant at day 28 postboost, suggesting that unlike in adults, the expansion of antibody secreting cells may occur after day 7.

Overall, these findings highlight the differences in host immune responses between children and adults.

Posted in influenza, Resource

Innate gene signature distinguishes humoral versus cytotoxic responses to influenza vaccination

Hierarchical clustering showing how the changes in transcript levels at day 1 differ between patients. Two distinct clusters, C1 and C2 are seen. Source from Gonçalves et al., JCI, 2020.

Does the vaccine administration route affect host responses to vaccines? In this paper, Gonçalves et al investigates how vaccine administration via the intramuscular (i.m.), intradermal (i.d.) and transcutaneous (t.c.) routes affects innate and adaptive responses to the inactivated influenza vaccine.

60 healthy subjects, 18–45 years old recruited to compare the immunogenicity of the 2012–2013 seasonal trivalent inactivated influenza vaccine (TIV) administered by the t.c. (20 subjects), i.d. (20 subjects) and i.m. (20 subjects) routes.

The i.m. and i.d. route generated higher neutralising antibodies compared to the t.c. route. Higher vaccine-specific CD8+GRZ+ T cells was seen after t.c. and i.d. vaccination compared to i.m. However, vaccine-specific CD8+ T-cell responses were not significantly different between the conditions.

More differentially expressed genes were detected in i.m. and i.d., as compared to t.c.. Two distinct gene signature clusters (C1 and C2; see above figure) were observed, but the clusters did not segregate by vaccine administration route. The C1-C2 dichotomy is attributed to genes involved in multiple pathways, such as those for antigen-presentation, DC maturation, and IFN signaling, where subjects in the C1 cluster expressed higher levels of these transcripts (see above figure).

Instead, C1 and C2 clusters had significant differences in humoral and CD8+GRZ+ T cell responses. Of note, C1 individuals had significantly higher influenza-specific MN antibody titers, but lower frequency of TIV-specific CD8+GRZ+ T cell responders.

80 transcripts related to interferon signaling and antigen presentation pathways are correlated with neutralising antibody responses. On the other hand, 31 transcripts related to metabolic pathways were correlated with TIV-specific CD8+GRZ T cell responses.

The top positively correlated genes with antibody responses are CXCR2P1, C2, and CKS1B and the top negatively correlated genes are PRKAA1 and TMEM8B. The top genes involved in correlation with CD8+GRZ+ T cell response were MAP2K5, PVRL1, SARM1, and CXCR4.

Data deposited in ArrayExpress with the accession code E-MTAB-7741.

Posted in influenza, Resource

Temporal Dynamics of Host Molecular Responses Differentiate Symptomatic and Asymptomatic Influenza A Infection

Figure describing the 8 distinct clusters of genes that are temporally modulated (0-108hrs) in the symptomatic (Sx) and the asymptomatic (Asx) volunteers after challenge with influenza virus. Source: Yongsheng H et al., PLOS Genetics, 2011.

A deep understanding of the molecular underpinnings underlying severe viral disease outcome in humans is critical for the development of drugs and therapeutics. Controlled human infection studies, in which volunteers are intentionally infected with a pathogen, can advance our understanding of disease pathogenesis as the incubation time, time-course of disease progression, symptomatic rates and immune responses can be closely monitored. In this manuscript published by Yongsheng H et al., PLOS Genetics, 2011, the authors investigated how the transcripts are differentially modulated in the symptomatic and asymptomatic subjects after challenge with the live influenza (H3N2/Wisconsin) strain

17 healthy adults were inoculated with live influenza (H3N2/Wisconsin) strain at 3 different doses (1∶10, 1∶100, 1∶1000, 1∶10000). 9 subjects were symptomatic whereas 8 were asymptomatic. Changes in host peripheral blood gene expression measured at -12, 0, 12, 21, 29, 36, 45, 53, 60, 69, 77, 84, 93, 101 and 108 hrs. 

Increasing doses of virus does not correlate with increased symptomatic outcome. This finding is congruent with our previous findings showing vaccine viremia does not influence symptomatic rates (Chan et al., JCI insight, 2017). In contrast, gene signature patterns were strongly associated with disease severity.

Using EDGE with false discovery rate (FDR) significance level (q-value)<0.01, 5,076 genes were temporally changed, comparing between symptomatic and asymptomatic phenotypes. Self-organizing maps (SOM) identified eight distinct classes with differential expression dynamics (See figure above).

Cluster 3 reveal genes that are uniquely increased in symptomatic subjects. These include PRR genes such as Toll-like receptor 7 (TLR7), the RNA helicases (RIG-I), and interferon induced with helicase C domain 1 (IFIH1). In addition, 11 genes from the TLR signaling pathway, including MyD88, TRAF6, and STAT1. NOD1, RIPK2, NOD2, NLPR3, and CASP5 and CASP1 and IL1b were increased in symptomatic subjects but not in the asymptomatic subjects.

Cluster 6 genes identified genes that are uniquely increased in the asymptomatic subjects. Enriched pathways were enriched are related to cellular response to oxidative stress. These include superoxide dismutase (SOD1) and serine/threonine kinase 25 (STK25 or SOK1), which have been linked to anti-oxidant/stress response and reduced concentration of ROS. This cluster also contain genes related to ribosomal synthesis, suggesting that protein biosynthesis could be protective against severe disease.

Both raw and normalized gene expression data are available at GEO (GSE30550).

Posted in influenza, Resource

Systems biology of vaccination for seasonal influenza in humans

Windrose plots showing the immune cell subsets that are affected by influenza LAIV and TIV. Figure obtained from Nakaya et al., Nature Immunology, 2011

Does live-attenuated vaccines trigger different host responses compared to inactivated vaccines? In this study, Nakaya et al., compared the transcriptomic responses of the influenza live-attenuated vaccine (LAIV) against the inactivated vaccine (TIV). Interestingly, despite both vaccines exhibiting similar clinical efficacy, the host responses that are induced by LAIV and TIV are vastly different. The summary of their findings are as follows:

Antibody responses of 56 healthy young adults vaccinated with either LAIV (n = 28) or TIV (n = 28).

Despite similar efficacy between the two vaccines, mean neutralising antibody response of subjects vaccinated with TIV was 6-fold higher than that of those vaccinated with LAIV. Of note, the magnitude of the antibody response can differ by >100-fold between subjects.

Only modest positive correlation was seen with IgG-secreting plasmablasts (day 7) and the HAI response (day 28). Among the different cytokines detected by the Luminex, only CXCL10 was significantly induced by TIV. However, CXCL10 does not correlate with neutralising antibody responses.

Interferon-related genes were induced by LAIV but not TIV. The transcriptomic data demonstrated that vaccination with TIV or LAIV induced distinct molecular signatures in the blood. The LAIV induce more genes related to T-cells whereas the TIV trigger more genes related to B-cells (see figure at top)

Correlation analysis reveal that XBP-1 related genes (Unfolded Protein Response), innate immune genes and reactive oxygen species response in macrophages were positively correlated with neutralising antibody responses. Interestingly, five members of the leukocyte immunoglobulin-like receptor family (LILRB1, LILRB2, LILRA1, LILRA3, LILRA6) were also correlated with neutralising antibody responses.

Based on their predictive network, the authors validated in mice that CAMPK4 is responsible for vaccine immunogenicity .

Data deposited as GSE29619.