Preprint Human Immune Cell Epigenomic Signatures in Response to Infectious Diseases and Chemical Exposures, 2023, Wang et al.

[SNT Gatchaman]

Human Immune Cell Epigenomic Signatures in Response to Infectious Diseases and Chemical Exposures
Wenliang Wang; Manoj Hariharan; Anna Bartlett; Cesar Barragan; Rosa Castanon; Vince Rothenberg; Haili Song; Joseph Nery; Andrew Aldridge; Jordan Altshul; Mia Kenworthy; Wubin Ding; Hanqing Liu; Wei Tian; Jingtian Zhou; Huaming Chen; Bei Wei; Irem B Gunduz; Todd Norell; Timothy J Broderick; Micah McClain; Lisa Satterwhite; Thomas Burke; Elizabeth Petzold; Xiling Shen; Chris Woods; Vance G Fowler; Felicia Ruffin; Parinya Panuwet; Dana B Barr; Jennifer L Beare; Anthony K Smith; Rachel R Spurbeck; Sindhu Vangeti; Irene Ramos; German Nudelman; Stuart C Sealfon; Flora Castellino; Anna Maria Walley; Tom Evans; Fabian Muller; William J Greenleaf; Joseph R Ecker

Variations in DNA methylation patterns in human tissues have been linked to various environmental exposures and infections. Here, we identified the DNA methylation signatures associated with multiple exposures in nine major immune cell types derived from peripheral blood mononuclear cells (PBMCs) at single-cell resolution. We performed methylome sequencing on 111,180 immune cells obtained from 112 individuals who were exposed to different viruses, bacteria, or chemicals.

Our analysis revealed 790,662 differentially methylated regions (DMRs) associated with these exposures, which are mostly individual CpG sites. Additionally, we integrated methylation and ATAC-seq data from same samples and found strong correlations between the two modalities. However, the epigenomic remodeling in these two modalities are complementary. Finally, we identified the minimum set of DMRs that can predict exposures.

Overall, our study provides the first comprehensive dataset of single immune cell methylation profiles, along with unique methylation biomarkers for various biological and chemical exposures.

Link | PDF (Preprint: BioRxiv)

 

[Hutan]

Overall, our study provides the first comprehensive dataset of single immune cell methylation profiles, along with unique methylation biomarkers for various biological and chemical exposures.

If that's true now, or can be in the future, that will be a major advance in environmental medicine. It would be amazing to be able to correlate exposure with illness (along with the underlying inherited genetics).

 

[Hutan]

Nine immune cell types

To investigate the DNA methylation alterations during the immune response to pathogens and toxic chemicals in major innate and adaptive immune cell types, we isolated seven cell types (B cells, Monocytes, NK cells, CD8 memory T cells, CD8 naïve T cells, CD4 memory T cells, and CD4 naïve T cells) from PBMCs of patients and healthy controls, and performed single-nucleus methylation sequencing (snmC-seq2) (Luo et al. 2017, 2018). We characterized the cell types based on fluorescence-activated cell sorting (FACS) and genome-wide methylation profiles in the CG context of each cell. This analysis revealed two sub-clusters within B cells and NK cells. We then identified differentially methylated regions (DMRs) within each of these nine cell types.

Samples of immune cells from 112 donors

173 PBMC samples collected from 112 donors, resulting in 111,180 PBMC methylation profiles. We examined the impact of each exposure on the methylome of the immune cell types

Donors covered specific known viral, bacterial and chemical exposures

In this study, our main focus was on three categories of exposures: viral, bacterial, and chemical (Figure 1A). To examine viral exposure, we evaluated the methylation patterns of individuals who were part of a prospective study on HIV-1 infection prevention. Additionally, we analyzed a cohort of volunteers who participated in a vaccine trial against the H3N2 flu virus and a group of patients who experienced SARS-CoV-2 infection with varying degrees of disease severity. For bacterial exposures, we examined the methylation profiles of patients infected with MRSA and MSSA, as well as vaccinated technicians who handled Bacillus anthracis. In terms of chemical exposure, we obtained samples from individuals with high levels of 3,5,6-trichloro-2-pyridinol (TCPY) resulting from exposure to chlorpyrifos—an organophosphate insecticide known to be neurotoxic and shares a mode of action with nerve agents.

Methylomes, methylomics of the DNA in the nucleus of each cell

Subsequently, we profiled the methylomes of each single cell

Cell types also differed by exposure group (although the exposure groups probably differed from each other in many ways e.g by age/sex, not just by specific exposure)

While immune cell types were initially sorted using cell surface markers, the global mCG levels enabled us to identify unique sub-clusters within B cells and NK cells (Figure 1C). Importantly, cells from different exposures were not evenly distributed across these sub-clusters, indicating the presence of distinct epigenomic diversity associated with exposure types (Figure 1C).

Figure 1 explains the study in pictures. Below is Fig Cii and Ciii.
Cii - Clustering of cells by methylene, coloured by cell surface markers
Ciii - Clustering of cells by methlyome, coloured by exposure type

It is interesting to see areas of specific colours in Ciii. For pretty much all of the specific exposure groups, it is possible to identify some clustering of specific cell types with specific methylomic profiles.

 

[Hutan]

The issue exposed and unexposed groups potentially being different in much more than just exposure is not a problem in this analysis, where they used pre-exposure/post-exposure PBMC samples:

Impact of HIV-1 Infection on the Methylome

To investigate the impact of HIV-1 infection on the methylome of immune cells, we conducted a comprehensive analysis using immune cells obtained from the same individuals at different stages: "pre" (before infection), "acute" (after diagnosis), and "chronic" (after treatment)

Significant changes in the genome-wide mCG levels of certain cell types were observed following HIV-1 infection. For instance, memory CD8 T cells exhibited an increase in global mCG levels from "pre" to "chronic" and "acute" stages (Figure S3E). Other cell types such as Memory B cells, Naive B cells, and Naive NK cells also demonstrated significant global mCG changes between the "pre" and "acute" stages (Figure S3E). Subsequently, we identified differentially methylated regions (DMRs) between the "pre," "acute," and "chronic" stages for these cell types. Interestingly, highly distinctive methylation patterns were observed between the three stages across all samples (Figure 3B), indicating that these DMRs are highly consistent with disease progression.

We could do this with pre and post Covid-19 infection samples, for Long Covid and non-Long Covid people. We could do this with ME/CFS samples, tracking changes pre and post exercise. I think we may have seen some limited methylation studies in ME/CFS - we could definitely do with some more.

Super interesting, thanks for posting it @SNT Gatchaman.

 

[Trish]

I haven't read the whole paper, just the bits quoted here. Does it explain how recent the exposure has to be and how long it lasts? Any individual in their lifetime will be exposed to hundreds of infectious agents and chemicals, so how do they single out the effect of one specific agent?

 

[Hutan]

Change in the methylation of circadian-related genes with HIV infection

Interestingly, motifs of transcription factors associated with circadian function were enriched in hypo-DMRs from the "pre" stage (Figure 3C). These circadian transcription factors include CLOCK, BMAL1, bHLHE41, and NPAS2, suggesting that circadian regulation may be altered in CD8 T cells after HIV-1 infection. Circadian transcription motifs were also enriched in hypo-DMRs of B cells in the "acute" stage.

The authors talk about two different methods of determining methylation - I haven't understood the methods. But they report that the methods can find some different things 'highlighting the necessity of using both technologies to profile the epigenomic remodeling'.

As described above, we performed integration between our methylation data and single-cell ATAC-seq data. In order to validate our integration approach and data quality, we calculated the genome-wide correlation between these two modalities (Figure 4G). We observed that the two modalities (methylation and open chromatin) were strongly correlated across all cell types, with the strongest correlation observed in monocytes and the lowest correlation in different types of T cells (Figure 4G). Surprisingly, only ~5% of the DMRs in each cell type overlapped with a peak in the corresponding cell type in ATAC-seq data, indicating that most of the methylation changes associated with COVID-19 are in inaccessible chromatin.

I haven't read the whole paper, just the bits quoted here. Does it explain how recent the exposure has to be and how long it lasts? Any individual in their lifetime will be exposed to hundreds of infectious agents and chemicals, so how do they single out the effect of one specific agent?

Re HIV, they have samples before and after becoming infected as well as later during treatment, and of course that's a chronic infection. The participants were part of a study that enrolled high risk uninfected individuals and tested them regularly. Around 250 days before infection and 200 days after infection.

For the flu cohort, they were part of a vaccination study that involved infecting participants with the flu. the post sample was 20 days after infection.

For the Covid-19 cohort, these participants were all hospitalised, with samples taken during their hospitalisation. I'm not sure when the samples were taken, sounds like within maybe a couple of weeks of infection?

For the MRSA/MSSA cohort, testing was done when the infection was identified and then 2 or three times after.

The anthrax sample might indeed have been a bit of a mixed bag. They had been immunised against anthrax but

The donors also handled chemical weapon agents, precursor molecules to chemical agents, biological warfare agents, explosives, precursors to explosives, agricultural chemicals, or radioactive materials in a controlled environment.

I suspect that if I knew more of what goes on in the world around biological warfare agents, I would sleep quite a lot worse.

For the organophosphate, they had participants with known levels in urine.

One commonly used form of this pesticide, Chlorpyrifos, has extensive use in the United States. Exposure to chlorpyrifos, the most widely used OP in the US, was estimated based on levels of its urinary metabolite 3,5,6-trichloro-2-pyridinol (TCPY) and classified as high, moderate, or low (with undetectable levels of TCPy). In this study, we analyzed DNA methylation patterns in 18 high-exposure, six moderate-exposure, and three low-exposure samples (see Supplementary Table S1).

 

[SNT Gatchaman]

ME/LC/Covid studies relating to DNA methylation changes —

Persistent blood DNA methylation changes one year after SARS-CoV-2 infection (2022, Clinical Epigenetics)
EWAS of post-COVID-19 patients shows methylation differences in the immune-response associated gene, IFI44L, three months after COVID-19 infection (2022, Nature Scientific Reports)
Epigenome-wide DNA methylation profiling of healthy COVID-19 recoverees reveals a unique signature in circulating immune cells (2021, Preprint: MedRxiv)
The SARS-CoV-2 receptor angiotensin-converting enzyme 2 ACE2 in myalgic encephalomyelitis/chronic fatigue syndrome: A meta-analysis of public DNA methylation and gene expression data (2021, Heliyon)
Blood DNA methylation and COVID-19 outcomes (2021, Clinical Epigenetics)
Changes in DNA methylation profiles of myalgic encephalomyelitis/chronic fatigue syndrome patients reflect systemic dysfunctions (2020, Clinical Epigenetics)
Alterations in DNA Methylation Status Associated with Gulf War Illness (2019, DNA and Cell Biology)

Also spaceflight (ish) —

DNA methylation dynamics associated with long-term isolation of simulated space travel (2022, iScience)