Edited by: Rino Rappuoli, GlaxoSmithKline, Italy
Reviewed by: Thorsten Demberg, Immatics Biotechnologies, Germany; Randy A. Albrecht, Icahn School of Medicine at Mount Sinai, United States
†These authors have contributed equally to this work.
Specialty section: This article was submitted to Vaccines and Molecular Therapeutics, a section of the journal Frontiers in Immunology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Systems biology approaches have recently provided new insights into the mechanisms of action of human vaccines and adjuvants. Here, we investigated early transcriptional signatures induced in whole blood of healthy subjects following vaccination with a recombinant HIV-1 envelope glycoprotein subunit CN54gp140 adjuvanted with the TLR4 agonist glucopyranosyl lipid adjuvant-aqueous formulation (GLA-AF) and correlated signatures to CN54gp140-specific serum antibody responses. Fourteen healthy volunteers aged 18–45 years were immunized intramuscularly three times at 1-month intervals and whole blood samples were collected at baseline, 6 h, and 1, 3, and 7 days post first immunization. Subtle changes in the transcriptomic profiles were observed following immunization, ranging from over 300 differentially expressed genes (DEGs) at day 1 to nearly 100 DEGs at day 7 following immunization. Functional pathway analysis revealed blood transcription modules (BTMs) related to general cell cycle activation, and innate immune cell activation at early time points, as well as BTMs related to T cells and B cell activation at the later time points post-immunization. Diverse CN54gp140-specific serum antibody responses of the subjects enabled their categorization into high or low responders, at early (<1 month) and late (up to 6 months) time points post vaccination. BTM analyses revealed repression of modules enriched in NK cells, and the mitochondrial electron chain, in individuals with high or sustained antigen-specific antibody responses. However, low responders showed an enhancement of BTMs associated with enrichment in myeloid cells and monocytes as well as integrin cell surface interactions. Flow cytometry analysis of peripheral blood mononuclear cells obtained from the subjects revealed an enhanced frequency of CD56dim NK cells in the majority of vaccines 14 days after vaccination as compared with the baseline. These results emphasize the utility of a systems biology approach to enhance our understanding on the mechanisms of action of TLR4 adjuvanted human vaccines.
香京julia种子在线播放
Since the onset of the epidemic, more than 70 million people have been infected with human immunodeficiency virus (HIV), and about 35 million people have died of HIV (
Antibodies directed against the viral
Early transcriptomics profiling has recently provided new insights into the mode of action of human vaccines and adjuvants (
Healthy male (
Whole blood samples were collected in PAXgene RNA stabilization tubes (Qiagen, USA) from each participant before vaccination (day 0, 0 h) as well as at 6 h and 1, 3, and 7 days following first vaccination. RNA was extracted using the PAXgene Blood RNA kit, according to the manufacturer’s protocols (Qiagen, USA). Spectrophotometry was used to measure RNA concentration (ND-1000 spectrophotometer, NanoDrop Technologies Inc., USA) and the RNA integrity number (RIN), an indicator of sample quality, was determined using an Agilent 2200 TapeStation and 2100 Expert Software (Agilent Technologies, USA). The RIN for the mRNA extracted from the remaining 66 whole blood samples ranged from 6.2 to 9.0 with a mean of 7.9 ± 0.6 for all extracted samples. Only samples with good quality mRNA, defined as having a 260/280 ratio approximating 2 and RIN >7, were used in microarray analyses.
Whole genome microarray analysis was performed at the Genomics Core Facility, University of Gothenburg, Sweden using an Illumina platform according to the manufacturer’s protocols. Briefly, the Fluorescent Linear Amplification Kit was used for RNA labeling, and quantity and labeling efficiency were verified prior to hybridization of samples to whole genome 8x60k human expression arrays (HumanHT-12 v4 Expression BeadChip kit; Illumina, USA). Post-hybridization scans were performed using an Agilent scanner at 5 µm and image analysis to generate raw data was completed using Feature Extraction software (version 11.5.1.1, Agilent Technologies). The raw data was pre-processed and normalized using the limma package and corrected for background using the “normexp” method in the statistical program R (
Gene set enrichment analysis (GSEA) using the blood transcription modules identified in Ref. (
Serum samples were obtained during the course of the study and concentrations of CN54gp140-specific serum IgM, IgA, and IgG (including subclasses IgG1, IgG2, IgG3, and IgG4) antibodies were determined as previously described (
For phenotypic analysis of lymphocyte subsets by multi-color flow cytometry, PBMCs were stained with fluorochrome-conjugated antibodies for the surface markers CD3 (OKT3, Biolegend), CD4 (SK3, Biolegend), CD8 (SK1, Biolegend), CD14 (M5E2, Biolegend), CD19 (HIB19, Biolegend), CD56 (NCAM16.2, BD Biosciences), CD94 (DX22, Biolegend), CD161 (HP-3G10, Biolegend), CRACC (235614, R&D Systems) NKG2A (Z199, Beckman Coulter), NKp80 (MA152, Beckman Coulter), and TCRγδ (IMMU510, Beckman Coulter) as well as for intracellular EOMES (WD1928, eBioscience), granulysin (GNLY) (DH2, Biolegend), NKG7 (2G9, Beckman Coulter), and PLZF (R17-809, BD Biosciences). Intracellular HOPX was detected with an unconjugated primary antibody (rabbit polyclonal, Proteintech) followed by a fluorochrome-labeled F(ab)2 anti-rabbit IgG secondary antibody (goat polyclonal, ThermoFisher). Dead cells were identified using a fixable live/dead stain (ThermoFisher).
Cryopreserved PBMCs from 0 h to 14 days were thawed, washed with cold RPMI-1640 (Gibco) supplemented with 10% FCS (Sigma-Aldrich) and labeled on ice with fluorochrome-conjugated antibodies (Table S1 in Supplementary Material) and live/dead stain in FACS buffer (PBS supplemented with 2% FCS and 2 mM EDTA; all Sigma-Aldrich). Cells were then washed with FACS buffer, fixed at room temperature in PBS containing 2% formaldehyde (Polysciences Inc.), permeabilized at room temperature in PBS with 0.1% Triton X-100 (Sigma-Aldrich) followed by intracellular staining performed in FACS buffer at room temperature. Samples were acquired on an LSRII Fortessa (BD Bioscience) and data analyzed and visualized using FlowJo software (v9.9.4; FlowJo, LLC), and GraphPad PRISM (v5.0). Gating strategy is shown in Figure S3 in Supplementary Material.
Changes in expression at any time point following vaccination compared to baseline (0 h) were identified for 529 genes (adjusted
Transcriptional changes in gene expression over time following vaccination. Subjects were intramuscularly immunized with CN54gp140 + GLA. In the Venn diagram,
The magnitude of observed group level transcriptional changes varied between time points compared to baseline (Figures
We next identified functional pathways perturbed by the vaccination by performing a two-class comparison in GSEA to identify enriched blood transcription modules (BTMs), first presented in Ref. (
Enrichment of blood transcription modules (BTMs) regulated upon vaccination. Spider web chart indicating enriched BTMs at 6 h (red), 1 day (yellow), 3 days (blue), and 7 days (green). BTMs are functionally divided into: cell cycle regulation and signaling
Next, we sought to study whether early transcript signatures correlated with later serum antibody responses. This was achieved by using GSEA to identify enriched BTMs in the transcriptomes of individuals grouped by their CN54gp140-specific antibody responses, at different time point post vaccination relative to the corresponding baseline. Individuals were divided statistically into high and low CN54gp140-specific serum IgM, IgG, or IgA antibody responders based on their serum antibody concentrations and separation into groups was validated using statistical tests as described above (Figure S2 in Supplementary Material).
Individuals categorized as high IgM responders (day 14) exhibited a gene signature marked by a repressed enrichment of NK cell-related genes (M7.2) at 3 and 7 days post-vaccination (Figure
Molecular signatures associated with early
Vaccinees were also categorized into high and low responders at later time points (up to 6 months post vaccination) based on their individual CN54gp140-specific serum antibody concentrations (Figure S2 in Supplementary Material). Using this analysis, no specific BTMs were identified using the gene expression data from individuals classified as high or low serum IgG or IgG1 responders. However, BTM-based gene signatures were identified which differentiated late high serum IgA, IgG3, and IgG4 responders from low responders (Figure
Given tendencies for vaccination-induced enrichment in NK cell-related BTMs and significant associations between such NK cell-related BTMs modules and serum antibody responses, we next examined peripheral blood NK cells by flow cytometry in samples from subjects at 0 h baseline and at 14 days post first vaccination. Samples from intermediate time-points were not available. To determine whether observed changes in NK cell-related BTMs could be attributed to change in NK cell numbers or expression levels of proteins upon vaccination, we stained PBMC for lineage markers, NK cell receptors, and an array of proteins within regulated BTMs. Such proteins included transcription factors EOMES and HOPX, granule constituent’s
Frequency of CD56dim NK cell was enhanced in high antibody responders 14 days post vaccination. FACS plots showing NK cell populations of representative low (no IgM 14 days and low IgG 168 days) and high (IgM response 14 days and high IgG 168 days) responder before (D0) and 14 days following vaccination
Herein, we employed a whole genome transcriptomics analysis combined with a systems biology approach to pinpoint early transcriptional signatures in whole blood of healthy volunteers vaccinated with the TLR4 agonist GLA-AF adjuvanted CN54gp140 HIV-1 vaccine candidate. We identified DEGs and BTM signatures of the whole blood response to the vaccine candidate up to 7 days following vaccination, and the BTMs that showed correlation with early and late CN54gp140-specific antibody responses.
We observed an increase in the number of DEGs compared to the baseline as early as 6 h post-vaccination, which peaked at 1 and 3 days followed by a decline to the expression levels comparable to that of baseline by 7 days. The observed kinetics of the transcriptional changes in whole blood is in line with previous studies (
The functional pathway analysis of transcriptomics data with GSEA revealed BTMs related to general cell cycle activation, and innate immune cell activation at early time points, as well as BTMSs related to T cells, and B cell activation at the later time points. BTM offers an unprecedented possibility to functionally assess transcriptomic changes related to immunological responses in human blood (
Notwithstanding, the flow cytometry analyses of the subjects’ PBMCs showed an increase in the frequency of CD3–CD56dim NK cells for 14 days post vaccination relative to the 0 h baseline. Unfortunately, the paucity of cellular samples from the subjects for 1–7 days post vaccination prevented our further study of this phenomenon. It is tempting to speculate that the repression of BTMs related to NK cells observed in the first 7 days post-vaccination reflects NK cells leaving the circulation early in the response. Given that NK cells are short lived, the enhanced frequency of NK cells for 14 days post vaccination is presumably attributed to secondary induction of NK cell differentiation processes in response to vaccination. NK cells have been shown to inhibit generation of long-lived memory B cells by suppressing follicular helper T cells during the first days of infection in mice (
In this respect, it is intriguing that low early serum IgG and IgA responders were associated with the BTMs “Enrichment in myeloid cells and monocytes (M81).” The observed enhancement of the BTMs for myeloid cells over those of NK cells in low responders may suggest potential differences in antigen handling and processing between low and high responders. Further, the enhancement in BTMs related to plasma cell enrichment in the low responders warrants further studies to pinpoint whether short-lived plasmablasts are being generated in preference to long-lived plasma cells in the low antibody responders.
In line with the enriched upregulation of early plasma cell signature at 6 h and 1 day observed in the IgM antibody low responders in our study, the yellow fever vaccine showed an inverse correlation of antibody responses with plasma cell and B cell BTMs. Conversely, the immune responses to the trivalent influenza vaccine (TIV) and the quadrivalent meningococcal conjugate vaccine showed a positive correlation to those BTMs. Recently, significant correlations between the vaccine-induced antibody responses and plasmablast signatures following influenza vaccination (
Altogether, we herein report potential transcript and BTM signatures of human whole blood in response to vaccination with the TLR4 agonist-adjuvanted CN54gp140 anti-HIV vaccine candidate. However, a few caveats should be applied to our findings. First, the small sample size suggests interesting observations that would require the execution of larger clinical studies to provide a greater statistical power. Second, PBMC samples from days 1–7 were not available for detailed analysis of changes in the number and phenotype of circulating cells, thus observed transcriptomic changes may reflect differences in mRNA in circulating cells and/or influx/efflux of cellular populations in the systemic circulation. Nevertheless, these results provide new information on the candidate transcript signatures that may have potential as early biomarkers of antibody responses induced by TLR4 agonist-adjuvanted HIV-1 Env vaccine candidates, and support the potential of systems approaches in providing new insights into underlying mechanisms of individual variation in response to vaccines in humans.
This work was performed in compliance with UK Clinical Trial Regulations and all procedures were approved by the recognized Research Ethics Committee and by the Medicines and Healthcare Products Regulatory Authority. The clinical trial study is registered at
AH and RS conceived the project and designed the experiments. JA, TO, and JP conducted the whole blood RNA sample preparation and quality control, and together with MÖ and AH analyzed the transcriptomics data, and composed the manuscript. SK and PM executed the serology experiments and interpreted the data. All the authors provided critical feedback on the manuscript prior to publication and have agreed to the final content.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We would like to thank Heinrich Schlums and Yenan T. Bryceson of Department of Medicine, Center for Hematology and Regenerative Medicine, Karolinska Institute and Karolinska University Hospital, Sweden for the execution of flow cytometry experiments and data analysis. The authors are grateful to the Genomics Core Facility at Sahlgrenska Academy, University of Gothenburg for running the samples for microarray analysis.
The Supplementary Material for this article can be found online at