Article | Open | Published:

Differential DNA methylation of potassium channel KCa3.1 and immune signalling pathways is associated with infant immune responses following BCG vaccination

Scientific Reportsvolume 8, Article number: 13086 (2018) | Download Citation


Bacillus Calmette–Guérin (BCG) is the only licensed vaccine for tuberculosis (TB) and induces highly variable protection against pulmonary disease in different countries. We hypothesised that DNA methylation is one of the molecular mechanisms driving variability in BCG-induced immune responses. DNA methylation in peripheral blood mononuclear cells (PBMC) from BCG vaccinated infants was measured and comparisons made between low and high BCG-specific cytokine responders. We found 318 genes and 67 pathways with distinct patterns of DNA methylation, including immune pathways, e.g. for T cell activation, that are known to directly affect immune responses. We also highlight signalling pathways that could indirectly affect the BCG-induced immune response: potassium and calcium channel, muscarinic acetylcholine receptor, G Protein coupled receptor (GPCR), glutamate signalling and WNT pathways. This study suggests that in addition to immune pathways, cellular processes drive vaccine-induced immune responses. Our results highlight mechanisms that require consideration when designing new TB vaccines.


Mycobacterium bovis Bacillus Calmette–Guérin (BCG) is the only licensed vaccine against tuberculosis (TB). BCG protects against severe childhood cases of miliary TB and tuberculous meningitis1. However, the protection afforded by the vaccine against pulmonary TB varies across geographical regions with studies reporting vaccine efficacy ranging from 0% in India2 to 80% in the United Kingdom (UK)3. Reasons for these variations remain unknown. Possible explanations include comorbidities4, pre-exposure to environmental mycobacteria5, and genetics6. Previous studies in Malawi7, The Gambia8, Indonesia9 and the UK8,10 have shown disparate immune responses across these populations, with a dominant production of IFNγ in the UK and greater production of T helper (Th)2 cytokines in Malawi and The Gambia10. Multiple efforts are being undertaken to identify correlates of protection following BCG vaccination11,12. The importance of IFNγ has been highlighted in numerous studies8,13,14,15,16, however, the Th1 boosting candidate TB vaccine MVA85A (Modified Vaccinia virus Ankara expressing Ag85A from M. tuberculosis) failed to improve on protection afforded by BCG in vaccinated infants17; a large South African study did not find that BCG specific Th1 response correlated with risk of TB disease18; however further analysis of the MVA85A clinical trial showed that the frequencies of cells producing BCG-specific IFNγ was associated with a reduced risk of developing disease16.

Another unknown factor in the immune response to the BCG vaccine is the molecular mechanism that drives these immune responses. Recent studies highlight the role of transcriptomics19,20,21 as means of identifying the key mechanisms and the importance of measuring RNA profiles in vaccine trials. Epigenetics, a mechanism known to play a role in regulation of gene expression, is another mechanism of growing importance. Specifically, DNA methylation of CpG dinucleotides in mammals is known to regulate gene expression and subsequent protein production22,23. In-vitro methylation of herpes thymidine kinase (tk) genes resulted in in-vivo downregulation of gene expression24 and methylation of the O6-methylguanine-DNA-methyltransferase (MGMT) gene was negatively correlated with its protein concentration in humans25. In the context of immune responses, two recent studies have examined the role of epigenetic regulation, gene expression and protein production in responses to Hepatitis B26 and Influenza27 vaccines. Several differentially methylated genes were found between low and high immune responders to Hep B vaccine and DNA methylation was correlated with gene expression and protein levels for multiple genes following influenza vaccination. Lastly, recent evidence shows that DNA methylation of macrophages is correlated with anti-mycobacterial activity in BCG vaccinated participants28.

We have examined whether the DNA methylation profile of peripheral blood mononuclear cells (PBMC) in BCG vaccinated infants is associated with the magnitude of BCG specific immune responses. We found novel pathways and genes that were differentially methylated between high and low BCG responders amongst South African infants who received BCG vaccination at birth. This knowledge will allow us to understand molecular mechanisms that drive vaccine-induced immune responses, paving the way to design better and more effective TB vaccines.


Sample collection and processing

We used 60 archived frozen PBMC samples from a previous study18. 36.7% were female, 83.3% Cape Mixed Ancestry, 13.3% Black African and 3.3% Asian. All participants received the Japanese BCG vaccine at birth and their blood was collected at 10 weeks post vaccination.

An increase in cytokine production after BCG stimulation is not correlated with cell phenotype

To investigate the magnitude of immune responses to the BCG vaccine, we measured the cytokine production after stimulation with BCG (SSI strain, 1.2 × 106 organisms/ml) or Staphylococcus Enterotoxin B (SEB) using an intracellular cytokine staining (ICS) assay. We looked at cytokine responses in the PBMC population (see gating strategy in Fig. 1A). There was a consistent increase in the production of all cytokines after BCG stimulation and IFNγ, TNFα, IL2 but not IL8 and IL4/5/13 after SEB stimulation (Fig. 1B). We examined whether the observed magnitude of immune responses was due to changes in T cell frequency and conducted a Spearman correlation analysis of frequencies of IFNγ-expressing cells following BCG stimulation with other immune parameters. The frequencies of BCG-specific IFNγ cells (hereafter called IFNγ BCG) were correlated with IL4/5/13, IL2 and IL8 secreted following stimulation with BCG but not with any T cell population measured (Table S1). This indicates that in this assay, the BCG induced IFNγ production is not correlated with T cell and lymphocyte composition. All samples were stratified based on their IFNγ BCG production and 15 highest (IFNγ Low) and 15 lowest (IFNγ High) IFNγ BCG producers were selected for the analysis of DNA methylation (Fig. 1C).

Figure 1
Figure 1

The immune response to the BCG vaccine was used to select samples for analysis. (A) The FACS gating strategy used to measure all immune responses. Singlet cells and CD3+ populations were chosen to measure intracellular cytokine responses as well as phenotypic markers such as CD3+, CD3+CD4+, CD3+CD8+, CD3+CD4+CCR7+, CD3+CD4+CD45RO+. (B) The magnitude of immune responses of PBMCs after 12 hours stimulation with BCG and SEB (% of PBMCs). Each column represents one donor sample. Stimulated samples were compared to non-stimulated with Wilcoxon paired t test. N = 60; ns = not significant; *p < 0.05; ****p < 0.0001 (C) A magnitude of IFNγ BCG responses when stratified by origin (CD3+ population). Fifteen lowest (non-responders) and fifteen highest (responders) IFNγ BCG producers (red) were selected for the DNA methylation analysis. Samples coloured in grey were not selected. N = 60.

Differential methylation of 318 genes between low and high BCG cytokine respondents

A total of 30 samples were selected for the DNA methylation analysis. These were stratified based on their BCG specific immune responses forming low vs. high groups for five cytokines: IFNγ; IL4/5/13; IL2; TNFα; and IL8. For each cytokine, DNA methylation of low responders was compared to high responders using logistic regression with a minimum difference in methylation of 5%. The analysis showed several probes to be differentially methylated between low and high groups of: IFNγ (70), IL4/5/13 (108), IL2 (50), TNFα (146), and IL8 (122) (Fig. 2). A gene feature report highlighted 318 genes (Table 1) that span the differentially methylated probes. Several novel genes of which levels of methylation can indirectly affect the magnitude of BCG immune responses were identified. Notably, the ZFP57 gene was found to be differentially methylated in all cytokine low and high groups comparisons. ZFP57 is a transcriptional regulator of gene imprinting and it acts by controlling DNA methylation during the earliest stages of multicellular development29,30. Another gene candidate KCa3.1 is a protein that forms the voltage-independent potassium channel and was found to be differentially methylated across 4 cytokine low and high groups (IFNγ, IL2, IL4/5/13 and IL8). Kca3.1 regulates calcium influx and can influence TH1 and TH2 development31. We also found ERICH-1 to be differentially methylated across cytokine low and high groups. ERICH-1 encodes the a glutamate rich 1 protein and glutamate is known to regulate Ca2+ and K+ efflux thereby affecting T cell activation32.

Figure 2
Figure 2

A scatter plot of all probes of which methylation was compared between a high and low cytokine responder group. Each dot represents a probe of which % methylation is measured in a high cytokine group (X axis) and low cytokine group (Y axis). Red dots represent probes which are significantly differentially methylated. BCG specific low and high cytokine groups that were compared included: IFNγ (A), IL2 (B), IL8 (C), IL4/5/13 (D), and TNFα (E).

Table 1 A summary of all genes that were differentially methylated across all low and high cytokine groups.

Hierarchical cluster analysis reveals clusters of differentially methylated genes

We conducted a t-distributed stochastic neighbour embedding (tSNE) analysis of gene methylation for the 30 examined samples, however this did not show a clear separation of IFNγ low and IFNγ high nor the other cytokine low and high groups (data not shown). This indicates subtle differences in DNA methylation between cytokine low and high groups, suggesting that there might also be other mechanisms that drive immune responses to the BCG vaccine. We then proceeded with a hierarchical cluster analysis of the differentially methylated probes when stratified by cytokine respondents. We saw visible differences in cluster formation for all the cytokine groups measured (Fig. 3).

Figure 3
Figure 3

A hierarchical per probe normalised clustering analysis of the DNA methylation profile between high and low IFNγ (A), IL2 (B), IL8 (C), IL4/5/13 (D) and TNFα (E) responder groups. First fifteen columns represent a methylation of probes in cytokine high responder group and last fifteen columns represent a methylation of probes in the low responder group. Each horizontal line represents a differentially methylated probe. n = 30.

Differentially methylated genes are part of immune and cellular processes pathways

We examined biological functions of the differentially methylated genes for all the cytokines using the Panther Pathway Analysis (PPA). Differentially methylated genes are included in 67 pathways that are involved in multiple biological processes including immune functions (2.8%), metabolic (20.8%) and cellular (25.3%) processes and other processes (Table 2). Several pathways of the immune system could explain the polarisation of cytokine responses. Notably, T cell activation, interferon gamma signalling, interleukin and JAK/STAT signalling pathways were all involved in the regulation of cytokine responses. B cell activation, TGF-beta and chemokine and cytokine mediated inflammation were also highlighted as pathways containing genes that were differentially methylated. A large proportion of highlighted pathways regulate cellular and metabolic processes. We found differentially methylated genes that are part of the G-protein signalling pathways, which are known to regulate T cell migration and activation33. We also found 6 differentially methylated genes in the Wnt pathway, which is known to regulate IL12 signalling in antigen presenting cells (APCs)34. Lastly, we highlight the muscarinic acetylcholine receptor signalling pathway (mAChRs), which can mediate cytokine production35. By regulating the immune signalling and cellular processes these pathways can directly and indirectly affect the magnitude of the BCG induced immune response that is observed in our samples.

Table 2 An overview of biological processes and selected pathways of the differentially methylated genes when stratified by low and high cytokine groups.


BCG vaccination induces variable immunity across populations36,37 translating to an inconsistent protection against pulmonary tuberculosis1. This study aimed to answer whether epigenetic modifications, specifically DNA methylation, could be one of the molecular mechanisms that drive the observed disparate immune responses.

We first measured the magnitude of immune responses induced after the BCG vaccination. We chose to include all cells in the PBMC population as the DNA used for the methylation analysis was extracted from matching vials of PBMC. Similar to the previous study18, we saw an increase of all the measured cytokines following BCG stimulation in vitro. The magnitude of immune responses for IFNγ ranged from 0% to 2.61% measured as the frequency of cytokine producing PBMCs. As expected, BCG and SEB stimulation decreased the expression of CD3 protein and had no effect on the expression of CD4 and CD8 proteins (data not shown) and SEB, an inflammatory superantigen38, failed to increase the production of TH2 cytokines and IL8. CD4+ and CD8+ T cells are known to be the main producers of IFNγ39, however, we saw that none of the T cell phenotypes measured correlated with levels of BCG induced IFNγ (Table S1). This indicates that the magnitude of immune responses is independent of cellular T cell composition and is likely driven by molecular mechanisms within individual cells. The frequencies of BCG-induced IFNγ had a strong positive correlation with IL2 and IL4/5/13, and a weak negative correlation with IL8. IL2 is known to activate T cells, which are one of the main IFNγ producers40, and the BCG vaccine was shown to previously induce both TH1 and TH2 cytokines7,9. We cannot exclude that changes in cell numbers of Natural Killer (NK) cells and monocytes between groups could explain observed differences in cytokine levels. This could be addressed by analysing DNA methylation patterns in purified cell populations, which due to a limited sample size was beyond the scope of the project.

Having measured the immune responses and selected samples for further analysis, we measured DNA methylation and stratified samples by high and low cytokine responses. We found probes that were differentially methylated between high and low cytokine responders and identified the 318 corresponding genes. Interestingly, genes that were differentially methylated in many of the cytokine groups were involved in potassium, calcium and neurotransmitter signalling.

KCNN4 or Kca3.1, Potassium Calcium-Activated Channel Subfamily N Member 4, was found to be differentially methylated in IFNγ, IL2, IL4/5/13, and IL8 high and low groups. The Kca3.1 channel is a voltage-independent potassium channel, which is activated by an increase in intracellular calcium, resulting in membrane hyperpolarization and continuous calcium influx31. Calcium regulates multiple cellular processes. A prolonged contact between a CD4+ T cell and an antigen presenting cell (APC) results in an increase in intracellular Ca2+ levels. The main sources of Ca2+ in T cells are store-operated calcium channels (SOCE) and calcium-release-activated calcium (CRAC)41,42,43,44. A SOCE and CRAC dependent release of calcium inside a CD4+ T cells changes gene expression, and thus cytokine production, drives the differentiation of naïve T cells to TH1 or TH2 cells, and the development of immature T cells41. A strong Ca2+ signal favours TH1 differentiation45 whereas weak Ca2+ release skews cells to a TH2 phenotype46,47,48. Kca3.1 can thus affect T cell activation by maintaining a negative membrane potential and continuing Ca2+ release via CRAC channels31.

ERICH1, which encodes the glutamate rich 1 protein, was found to be differentially methylated in 4 out of 5 low and high cytokine groups. Glutamate is an excitory neurotransmitter of the nervous system, which is critical for brain development and function and plays a signalling role in peripheral organs, including T lymphocytes32,49. It can be produced by several cells of the immune system including T cells, dendritic cells and others. Glutamate mediates intracellular Ca2+ fluxes and outward K+ flows thereby affecting adhesion, migration, proliferation, survival, activation and metabolism of T cells32,49,50. A differential methylation of ERICH1 could therefore lead to varied intracellular glutamate levels affecting the immune response to the BCG vaccine.

To look more broadly at differentially methylated genes, we investigated the array of pathways which contained differentially methylated genes. As expected, we found several immune pathways: T cell activation, JAK/STAT signalling, Interleukin signalling, Interferon-gamma signalling, inflammation mediated by chemokine and cytokine signalling, integrin signalling, TGF-beta signalling, B cell activation. All these pathways could contribute to the underlying differences observed in the cytokine response to the BCG vaccine. Interestingly, we also found a large proportion of pathways that were involved in metabolic and cellular processes. Several neurotransmitter mediating pathways were highlighted.

Muscarinic acetylcholine receptor signalling pathway signals via muscarinic acetylcholine receptors (mAChRs), which is made up of five types (M1–M5) of class A G protein-coupled receptors (GPCRs)51. Activation of T cells with PHA and PMA upregulates the expression of mAChRs52,53; a stimulation via CD11a upregulates M5 gene expression52 suggesting an increase in cholinergic transmission following T cell activation; a stimulation of mAChRs in PBMC upregulated the IL2 production35; and M1-M5 knockout mice produced lower levels of OVA Specific IgG1, TNFα and IL-654.

Seven genes in the GPCRs signalling pathway were differentially methylated across low and high cytokine groups. G proteins act as molecular switches to control downstream effector molecules including chemokines and chemokine receptors33. In immune cells, G proteins impact signal transduction and affect survival, proliferation, differentiation and cell migration and recent evidence suggests non-canonical GPCR signalling in immune cells33,55.

The WNT pathway, of which 6 genes were differentially methylated, is known to indirectly influence the immune system. There are 19 WNT genes in the human genome and they all encode lipid-modified secreted glycoproteins. There is numerous evidence that show how the WNT proteins affect the immune system: WNT1 and WNT4 deficient mice have impaired T and B cell development56,57; WNT proteins are crucial for the initial proliferation of thymocytes before β-selection58; the overexpression of WNT signalling increases the survival of CD4+CD25+ regulatory T cells (Tregs)59; and WNT signalling in APCs results in increased IL12 production and subsequent TH1 differentiation34. Variation in methylation of WNT genes could therefore lead to changes in WNT proteins, ultimately leading to mediating the immune response to the BCG vaccine.

A report by Verma et al.28 has recently investigated the effect of BCG vaccination on the DNA methylation of PBMCs in vivo. They found an enrichment of the DNA methylation in genes of immune pathways, including T cell activation, after the BCG vaccine vs. before vaccination. Their study shows that several non-immune pathways such as regulation of biological processes are enriched however these processes were not discussed in the report. It is important to highlight notable differences between the current study and that of Verma et al. We investigated the preceding methylation patterns induced by BCG vaccination and how they regulate the immune response to BCG stimulation in infants whereas the Verma et al. study investigated how BCG vaccination modulates the DNA methylation of PBMCs in adults.

Hierarchical cluster analysis of DNA methylation revealed visible differences in cluster formation between high and low IFNγ groups as well as other cytokines but tSNE analysis failed to separate infants based on their DNA methylation patterns. These findings suggest that DNA methylation is probably not the main mechanism to regulate the immune response to the BCG vaccine and other epigenetic mechanisms might play a role. Histone modifications and microRNA are other types of epigenetic modifications that could explain differences in observed immune responses. Histone modifications are known to control transcriptional profile of memory lymphocytes thereby shaping their function60. They also regulate mechanisms of adaptive features of trained immunity61. microRNAs have shown to play a role in modulation of inflammatory responses62.

We acknowledge that the observed differences in DNA methylation patterns may not be specific to the BCG vaccine responses and may in fact, represent a regulation of general vaccine induced immune responses. However, it is important to highlight that in a context of this study, cytokine levels were measured after BCG stimulation, hence they are BCG specific. Thus, the differences in cytokine levels and the subsequent stratification to low and high cytokine responding groups is BCG specific. The DNA methylation differences observed between these groups are associated with BCG specific differences in cytokine responses, however we cannot exclude that they may also be associated with immune responses to other vaccines.

Taken together, the data reported here identified genes and pathways that were differentially methylated in PBMC from infants with high or low cytokine responses to BCG vaccination. As expected we observed a differential methylation of multiple immune pathways that could directly influence the disparate immune responses to the BCG vaccine. Unexpectedly, we identified several genes and pathways that could indirectly affect the BCG specific cytokine production. These findings suggest that in addition to immune pathways, vaccine induced immune responses could be modulated by molecular mechanisms and mediators that regulate cellular processes such as glutamate, flux of potassium and calcium through membrane channels, GPCRs and mACHRs signalling and the WNT pathway. Studies should focus on investigating the role of methylation as well as other epigenetic marks in the regulation of the genes and pathways in question as well as measuring differences in gene expression when stratified by high and low cytokine responses. This knowledge will allow us to understand the molecular mechanisms that drive vaccine-induced immune responses, paving a way to design better and more effective TB vaccines.


The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Study participants

We retrieved blood samples collected from a subset of 10 week old infants who were enrolled into a large study of BCG vaccination at the South African Vaccine Initiative (SATVI) field site in the Worcester area, near Cape Town, South Africa (Hawkridge et al., BMJ 200818. Parents All participants were vaccinated with BCG at birth and samples were collected 10 weeks after birth. Participants were excluded from the study if any of the exclusion criteria were met: infant not immunised with BCG within 24 hours; mother infected with human immunodeficiency virus (HIV); chronic and acute disease in the infant at the time of enrolment; clinical anaemia in the infant; significant perinatal complications in the infant; and contact with any person with TB disease or anyone who was coughing. Participants were followed for two years to observe the development of TB disease. Only healthy infants who did not develop TB disease were selected for this study. Parents gave consent for their infants to participate in the study. The study was conducted according to the U.S. Department of Health and Human Services and Good Clinical Practice guidelines, and included protocol approval by the University of Cape Town Research Ethics Committee and written informed consent from the parent or legal guardian. The study also received ethical approval from the Ethics Committee of the London School of Hygiene and Tropical Medicine (LSHTM, #8720).

Cell separation, processing and stimulation

The methods below have been previously published in other studies18,63. Heparinized blood samples were collected from infants. Peripheral blood mononuclear cells were isolated using the density gradient centrifugation. The remaining 1 ml of blood was incubated with BCG (SSI, 1.2 × 106 organisms/ml), medium alone, and staphylococcal enterotoxin B (10 μg/ml; Sigma-Aldrich) (SEB). The co-stimulatory antibodies, anti-CD28 and anti-CD49d antibodies (1 μg/ml each; BD Biosciences, San Jose, CA) were added to all conditions. Samples were incubated for 7 hours at 37 °C. Later, Brefeldin-A was added and samples were incubated for another 5 hours. Red blood cells were lysed and white cells fixed using FACS Lysing Solution (BD Biosciences). Cells were collected, fixed and cryopreserved as described elsewhere63. Thus, for each participant non-stimulated PBMC sample and three stimulated and fixed whole blood samples were available.

Intracellular cytokine staining

Previously stimulated samples were thawed and washed in FACS buffer (PBS, 5% FBS v/v, 0.05% sodium azide w/v) and stained for 30 minutes at 4 °C in the dark, with a cocktail of surface marker antibodies: CD3-BV650 (clone SK7), CD4-BV605 (clone S3.5), CCR7-PECF594 (clone 150503), CD45RO-APC-H7 (clone UCHL1), (all BD Bioscience) and CD8-BV570 (clone RPA-T8; BioLegend). Samples were then permeabilised using the Cytoperm/Cytofix (BD Pharmingen) solution for 20 mins. After permeabilisation, cells were stained for 30 minutes at room temperature in the dark with a cocktail of intracellular cytokine antibodies: TNFα-PE-Cy7 (clone Mab11), IFNγ-V450 (clone B27), IL-2-FITC (clone MQ1-17H12), (BD Bioscience), IL-4-PE (clone 3010.211; FastImmune), IL-5-PE (clone TRFK5), IL-13-PE (JES10-5A2) and IL-8-APC (clone E8A1), (BioLegend). IL4, 5 and 13 were collected in one PE channel. After staining, samples were washed and stored at 4 °C in the dark to be acquired within 24 hours using BD LSRII Flow Cytometer and acquiring a minimum of 200,00 events.

Sample selection and DNA isolation

Samples were stratified based on the results of the ICS assay and the BCG specific IFNγ response. Corresponding PBMC samples of the 15 lowest (hereafter called IFNγ Low) and 15 highest IFNγ (hereafter called IFNγ High) respondents were selected for the DNA methylation analysis. The selected 30 samples were later stratified based on their IL2, IL8, IL4/5/13 and TNFα responses effectively forming 5 low and high cytokine groups. DNA was extracted using the Dneasy Blood & Tissue Kit (Qiagen). The quality of the DNA was checked with NanoDrop (Thermo Fisher Scientific) and the concentration measured with Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific).

Measurement of the DNA methylation

The DNA methylation was examined using a reduced representation bisulphite sequencing (RRBS) method and with the Premium Reduced Representation Bisulphite Sequencing Kit (Diagenode) according to manufacturer’s instructions.

DNA quality control and MiSeq sequencing

DNA concentration was measured using the Qubit dsDNA HS Assay Kit and size examined with the High Sensitivity DNA Analysis Kit (Genomics Agilent) using Bioanalyzer 2100 (Genomics Agilent). DNA was denatured with 0.2 M NaOH and its concentration was adjusted to 16pM. Samples were sequenced with the MiSeq Reagent Kit v3 (Illumina) at 51 cycles per run and single end reads. A 12.5 pM PhiX v3 library (Illumina) was added for a positive control.

Data processing

FASTA files were generated as the DNA sequencing output. Sequences were trimmed using the TrimGalore (v0.4.3) and Cutadapt software discarding all reads below 30 Phred quality score. The quality of samples was measured using the FastQC software (v0.11.5). A bisulphite treated human genome reference sequence was prepared using the GrCh38p7 assembly and Bismark Genome Preparation (v0.16.3) and Bowtie2 (v2.2.9) software. Samples were mapped to the reference genome using Bismark (v0.16.3) and Bowtie2. The DNA methylation was extracted using Bismark Methylation Extractor (v0.16.3).

Data analysis: DNA methylation

The methylation of DNA sequences was analysed using SeqMonk (v1.38.2) and Rstudio (1.0.44) software. Contig methylation probes were generated with a 10-fold depth cut off, ignoring duplicate reads and merging probes closer that 500 bp. Probes were quantified using read count quantitation and filtered to include probes measured at least 10 times. Chromosomes X, Y and any mitochondrial DNA were removed and probes were quantitated using bisulphite methylation pipeline embedded in the SeqMonk software. Based on the previous ICS responses of whole blood samples, DNA samples were then separated into two groups: IFNγ low & IFNγ high and a logistic regression with a P-value cut-off of 0.05 and multiple testing correction was run. The logistic regression test was then performed on other cytokine groups: IL4/5/13 low & IL4/5/13 high; IL2 low & IL2 high; TNFα low & TNFα high; and IL8 low and IL8 high. Genes that were differentially methylated were visualised using SeqMonk, Panther Pathway Analysis and Cytospace (v3.5.1).

Data analysis: correlation of immune responses

We used StataSE (v15.0) software to measure whether the magnitude of the BCG induced immune response is cell population dependent. Frequencies of BCG specific cytokines (TNFα, IFNγ, IL2, IL8 and IL4/5/13) and cell phenotypes (CD3 + CD4 + CCR7 + , CD3 + CD4 + CD45RO + , CD3 + , CD3 + CD4 + , CD3 + CD8 + , % of Lymphocytes) were correlated with the BCG specific IFNγ response using spearman correlation.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    Mangtani, P. et al. Protection by BCG vaccine against tuberculosis: A systematic review of randomized controlled trials. Clin. Infect. Dis. 58, 470–480 (2014).

  2. 2.

    Trial of BCG vaccines in south India for tuberculosis prevention: First report. Bull. World Health Organ. 57 819–827 (1979).

  3. 3.

    BCG and vole bacillus vaccines in the prevention of tuberculosis in adolescence and early adult life. Bull. World Health Organ. 46 371–385 (1972).

  4. 4.

    Webb, E. L. et al. Effect of single-dose anthelmintic treatment during pregnancy on an infant’s response to immunisation and on susceptibility to infectious diseases in infancy: A randomised, double-blind, placebo-controlled trial. Lancet 377, 52–62 (2011).

  5. 5.

    Weir, R. E. et al. The influence of previous exposure to environmental mycobacteria on the interferon-gamma response to bacille Calmette-Gu?in vaccination in southern England and northern Malawi. Clin. Exp. Immunol. 146, 390–399 (2006).

  6. 6.

    Newport, M. J. et al. Genetic regulation of immune responses to vaccines in early life. Genes Immun. 5, 122–129 (2004).

  7. 7.

    Lalor, M. K. et al. BCG vaccination induces different cytokine profiles following infant BCG vaccination in the UK and Malawi. J. Infect. Dis. 204, 1075–1085 (2011).

  8. 8.

    Hur, Y.-G. et al. Factors affecting immunogenicity of BCG in infants, a study in Malawi, The Gambia and the UK. BMC Infect. Dis. 14, 184 (2014).

  9. 9.

    Djuardi, Y., Sartono, E., Wibowo, H., Supali, T. & Yazdanbakhsh, M. A longitudinal study of BCG vaccination in early childhood: The development of innate and adaptive immune responses. PLoS One 5 (2010).

  10. 10.

    Lalor, M. K. et al. Population differences in immune responses to Bacille Calmette-Guerin vaccination in infancy. J Infect Dis 199 (2009).

  11. 11.

    Kaufmann, S. H. E. et al. TB biomarkers, TB correlates and human challenge models: New tools for improving assessment of new TBvaccines. Tuberculosis 99, S8–S11 (2016).

  12. 12.

    Bhatt, K., Verma, S., Ellner, J. J. & Salgame, P. Quest for correlates of protection against tuberculosis. Clinical and Vaccine Immunology 22, 258–266 (2015).

  13. 13.

    Marchant, A. et al. Newborns develop a Th1-type immune response to mycobacterium bovis bacillus Calmette-Guerin vaccination. J Immunol 163 (1999).

  14. 14.

    Lalor, M. K. Infant immune responses following BCG vaccination in the UK and Malawi (2009).

  15. 15.

    Lammas, D. A., Casanova, J. L. & Kumararatne, D. S. Clinical consequences of defects in the IL-12-dependent interferon-gamma (IFN-γ) pathway. Clinical and Experimental Immunology 121, 417–425 (2000).

  16. 16.

    Fletcher, H. A. et al. T-cell activation is an immune correlate of risk in BCG vaccinated infants. Nat. Commun. 7, 11290 (2016).

  17. 17.

    Tameris, M. D. et al. Safety and efficacy of MVA85A, a new tuberculosis vaccine, in infants previously vaccinated with BCG: A randomised, placebo-controlled phase 2b trial. Lancet 381, 1021–1028 (2013).

  18. 18.

    Kagina, B. M. N. et al. Specific T cell frequency and cytokine expression profile do not correlate with protection against tuberculosis after bacillus Calmette-Guérin vaccination of newborns. Am. J. Respir. Crit. Care Med. 182, 1073–1079 (2010).

  19. 19.

    Fletcher, H. A. Profiling the host immune response to tuberculosis vaccines. Vaccine 33, 5313–5315 (2015).

  20. 20.

    Matsumiya, M. et al. Inflammatory and myeloid-associated gene expression before and one day after infant vaccination with MVA85A correlates with induction of a T cell response. BMC Infect. Dis. 14, 314 (2014).

  21. 21.

    Matsumiya, M. et al. Roles for Treg Expansion and HMGB1 Signaling through the TLR1-2-6 Axis in Determining the Magnitude of the Antigen-Specific Immune Response to MVA85A. PLoS One 8 (2013).

  22. 22.

    Allis, C. D. & Jenuwein, T. The molecular hallmarks of epigenetic control. Nat. Rev. Genet. 17, 487–500 (2016).

  23. 23.

    Jones, P. A. Functions of DNA methylation: islands, start sites, gene bodies and beyond. Nat. Rev. Genet. 13, 484–492 (2012).

  24. 24.

    Keshet, I., Yisraeli, J. & Cedar, H. Effect of regional DNA methylation on gene expression. Proc Natl Acad Sci USA 82, 2560–2564 (1985).

  25. 25.

    Uno, M. et al. Correlation of MGMT promoter methylation status with gene and protein expression levels in glioblastoma. Clinics 66, 1747–1755 (2011).

  26. 26.

    Lu, Y., Cheng, Y., Yan, W. & Nardini, C. Exploring the molecular causes of hepatitis B virus vaccination response: an approach with epigenomic and transcriptomic data. BMC Med. Genomics 7, 12 (2014).

  27. 27.

    Zimmermann, M. T. et al. System-wide associations between DNA-methylation, gene expression, and humoral immune response to influenza vaccination. Plos One 11 (2016).

  28. 28.

    Verma, D. et al. Anti-mycobacterial activity correlates with altered DNA methylation pattern in immune cells from BCG-vaccinated subjects. Sci. Rep. 7, 12305 (2017).

  29. 29.

    Takikawa, S. et al. Human and mouse ZFP57 proteins are functionally interchangeable in maintaining genomic imprinting at multiple imprinted regions in mouse ES cells. Epigenetics 8, 1268–1279 (2013).

  30. 30.

    Li, X. et al. A Maternal-Zygotic Effect Gene, Zfp57, Maintains Both Maternal and Paternal Imprints. Dev. Cell 15, 547–557 (2008).

  31. 31.

    Lam, J. & Wulff, H. The lymphocyte potassium channels Kv1.3 and KCa3.1 as targets for immunosuppression. Drug Development Research 72, 573–584 (2011).

  32. 32.

    Ganor, Y. & Levite, M. The neurotransmitter glutamate and human T cells: Glutamate receptors and glutamate-induced direct and potent effects on normal human T cells, cancerous human leukemia and lymphoma T cells, and autoimmune human T cells. J. Neural Transm. 121, 983–1006 (2014).

  33. 33.

    Boularan, C. & Kehrl, J. H. Implications of non-canonical G-protein signaling for the immune system. Cellular Signalling 26, 1269–1282 (2014).

  34. 34.

    Blumenthal, A. et al. The Wingless homolog WNT5A and its receptor Frizzled-5 regulate inflammatory responses of human mononuclear cells induced by microbial stimulation. Blood 108, 965–973 (2006).

  35. 35.

    Fujino, H., Kitamura, Y., Yada, T., Uehara, T. & Nomura, Y. Stimulatory roles of muscarinic acetylcholine receptors on T cell antigen receptor/CD3 complex-mediated interleukin-2 production in human peripheral blood lymphocytes. Mol. Pharmacol. 51, 1007–1014 (1997).

  36. 36.

    Fine, P. E. Variation in protection by BCG: implications of and for heterologous immunity. Lancet 346 (1995).

  37. 37.

    Colditz, G. A. et al. Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. JAMA 271, 698–702 (1994).

  38. 38.

    Levy, R. et al. Superantigens hyperinduce inflammatory cytokines by enhancing the B7-2/CD28 costimulatory receptor interaction. Proc. Natl. Acad. Sci. USA 113, E6437–E6446 (2016).

  39. 39.

    Schoenborn, J. R. & Wilson, C. B. Regulation of interferon-gamma during innate and adaptive immune responses. Advances in immunology 96, 41–101 (2007).

  40. 40.

    Boyman, O. & Sprent, J. The role of interleukin-2 during homeostasis and activation of the immune system. Nat. Rev. Immunol. (2012).

  41. 41.

    Feske, S. Calcium signalling in lymphocyte activation and disease. Nat. Rev. Immunol. 7, 690–702 (2007).

  42. 42.

    Lewis, R. S. Calcium signaling mechanisms in T lymphocytes. Annu. Rev. Immunol. 19, 497–521 (2001).

  43. 43.

    Parekh, A. B. & Putney, J. W. Store-operated calcium channels. Physiol. Rev. 85, 757–810 (2005).

  44. 44.

    Prakriya, M. & Lewis, R. S. CRAC channels: Activation, permeation, and the search for a molecular identity. Cell Calcium 33, 311–321 (2003).

  45. 45.

    Rogers, P. R., Huston, G. & Swain, S. L. High antigen density and IL-2 are required for generation of CD4 effectors secreting Th1 rather than Th0 cytokines. J. Immunol. 161, 3844–52 (1998).

  46. 46.

    Constant, S., Pfeiffer, C., Woodard, A., Pasqualini, T. & Bottomly, K. Extent of T cell receptor ligation can determine the functional differentiation of naive CD4+ T cells. J. Exp. Med. 182, 1591–1596 (1995).

  47. 47.

    Leitenberg, D. & Bottomly, K. Regulation of naive T cell differentiation by varying the potency of TCR signal transduction. Semin Immunol 11, 283–292 (1999).

  48. 48.

    Sloan-Lancaster, J., Steinberg, T. H. & Allen, P. M. Selective loss of the calcium ion signaling pathway in T cells maturing toward a T helper 2 phenotype. J. Immunol. 159, 1160–8 (1997).

  49. 49.

    Fallarino, F. et al. Metabotropic glutamate receptor-4 modulates adaptive immunity and restrains neuroinflammation. Nat. Med. 16, 897–902 (2010).

  50. 50.

    Pacheco, R., Gallart, T., Lluis, C. & Franco, R. Role of glutamate on T-cell mediated immunity. J. Neuroimmunol. 185, 9–19 (2007).

  51. 51.

    Kruse, A. C. et al. Muscarinic acetylcholine receptors: novel opportunities for drug development. Nat. Rev. Drug Discov. 13, 549–560 (2014).

  52. 52.

    Fujii, T., Watanabe, Y., Inoue, T. & Kawashima, K. Upregulation of mRNA encoding the M5 muscarinic acetylcholine receptor in human T- and B-lymphocytes during immunological responses. Neurochem. Res. 28, 423–429 (2003).

  53. 53.

    Watanabe, Y., Kawashima, K., Fujii, T. & Fujimoto, K. Expression of acetylcholine in lymphocytes and modulation of an independent lymphocytic cholinergic activity by immunological stimulation. Biogenic Amines 17, 373–386 (2002).

  54. 54.

    Fujii, Y. X. et al. Diminished antigen-specific IgG1 and interleukin-6 production and acetylcholinesterase expression in combined M1 and M5 muscarinic acetylcholine receptor knockout mice. J. Neuroimmunol. 188, 80–85 (2007).

  55. 55.

    Wang, Y., Li, Y. & Shi, G. The regulating function of heterotrimeric G proteins in the immune system. Archivum Immunologiae et Therapiae Experimentalis 61, 309–319 (2013).

  56. 56.

    Mulroy, T., McMahon, J. A., Burakoff, S. J., McMahon, A. P. & Sen, J. Wnt-1 and Wnt-4 regulate thymic cellularity. Eur. J. Immunol. 32, 967–971 (2002).

  57. 57.

    Staal, F. J. T., Luis, T. C. & Tiemessen, M. M. WNT signalling in the immune system: WNT is spreading its wings. Nat. Rev. Immunol. 8, 581–593 (2008).

  58. 58.

    Weerkamp, F. et al. Wnt signaling in the thymus is regulated by differential expression of intracellular signaling molecules. Proc. Natl. Acad. Sci. USA 103, 3322–3326 (2006).

  59. 59.

    Ding, Y., Shen, S., Lino, A. C., Curotto de Lafaille, M. A. & Lafaille, J. J. Beta-catenin stabilization extends regulatory T cell survival and induces anergy in nonregulatory T cells. Nat. Med. 14, 162–169 (2008).

  60. 60.

    Weng, N., Araki, Y. & Subedi, K. The molecular basis of the memory T cell response: differential gene expression and its epigenetic regulation. Nat. Rev. Immunol. 12, 306–315 (2012).

  61. 61.

    Kleinnijenhuis, J. et al. Bacille Calmette-Guerin induces NOD2-dependent nonspecific protection from reinfection via epigenetic reprogramming of monocytes. Proc. Natl. Acad. Sci. 109, 17537–17542 (2012).

  62. 62.

    O’Connell, R. M., Rao, D. S. & Baltimore, D. microRNA Regulation of Inflammatory Responses. Annu. Rev. Immunol. 30, 295–312 (2012).

  63. 63.

    Hanekom, W. A. et al. Novel application of a whole blood intracellular cytokine detection assay to quantitate specific T-cell frequency in field studies. J. Immunol. Methods 291, 185–195 (2004).

Download references


We would like to thank all the infants and parents who participated in this study. We acknowledge the funding from: the European Commission within Horizon2020 TBVAC2020 (Grant No. H2020 PHC-643381); National Institutes of Health grant RO1-AI065653, European and Developing Countries Clinical Trial Partnership, Aeras, and the Bill and Melinda Gates Foundation through Grand Challenges in Global Health grant 37772 (“Biomarkers of Protective Immunity against TB in the context of HIV/AIDS in Africa”).

Author information


  1. Department of Immunology and Infection, Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, Keppel Street, London, WC1E 7HT, United Kingdom

    • Mateusz Hasso-Agopsowicz
    • , Hazel M. Dockrell
    •  & Steven G. Smith
  2. South African Tuberculosis Vaccine Initiative and Institute of Infectious Disease and Molecular Medicine, Division of Immunology, Department of Pathology, University of Cape Town, Institute of Infectious Diseases and Molecular Medicine, Rondebosch, 7701, Cape Town, South Africa

    • Thomas J. Scriba
    •  & Willem A. Hanekom


  1. Search for Mateusz Hasso-Agopsowicz in:

  2. Search for Thomas J. Scriba in:

  3. Search for Willem A. Hanekom in:

  4. Search for Hazel M. Dockrell in:

  5. Search for Steven G. Smith in:


Conceptualization M.H.A., S.S. and H.M.D.; Methodology M.H.A. and S.S.; Investigation and Formal Analysis M.H.A.; Resources T.S., W.A.H.; Writing – Original Draft M.H.A.; Writing – Review & Editing S.S., T.S. and H.M.D.; Visualization M.H.A., Supervision S.S.; Funding Acquisition H.M.D. and W.H.

Competing Interests

The authors declare no competing interests.

Corresponding authors

Correspondence to Mateusz Hasso-Agopsowicz or Steven G. Smith.

Electronic supplementary material

About this article

Publication history






By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.