Epigenetic Susceptibility to Inflammatory Complications After Cardiac Surgery

Introduction

Hypothesis

1. Enrichment of the epigenetic markers H3K4me3 and H3K27me3 in CD14+ monocytes increases the incidence and severity of organ dysfunction after major cardiac surgery through an exaggerated postoperative inflammatory response.

To test this hypothesis, a prospective observational study recruiting 30 patients undergoing major cardiac surgery (coronary artery bypass grafting (CABG)) will be conducted to assess for association between abundance of 2 key epigenetic markers within CD14+ monocytes (H3K4me3 and H3K27me3) associated with dysregulated innate immune responses and validated clinical and biochemical measures of organ dysfunction and complications. This study will add knowledge to the recruiting multicentre prospective observational study IMPRoVE (Incidence, impact and Mechanisms of Perioperative Right VEntricular dysfunction), as a sub-study of IMPRoVE assesses epigenetic susceptibility to organ dysfunction, via the same mechanism, in patients undergoing major thoracic and vascular surgery. Epi-SICCS will therefore provide a novel opportunity to extend investigation of patients undergoing major thoracic and vascular surgery (via IMPRoVE) by studying patients undergoing major cardiac surgery (Epi-SICCS).

Background and Rationale Background

Postoperative complications are a major cause of morbidity, mortality and increased health care cost in patients undergoing elective surgery and are particularly prevalent after cardiac (57%) surgery. Further, whilst mortality and significant morbidity are easily recognised sequelae of major surgery, it is increasingly recognised that overt complications are the 'tip of the iceberg' and a significant burden of covert postoperative complications exist and have a significant long-term impact. Seeking to better understand and ameliorate this burden of complications remains a research priority.

The 'stress response' to surgical trauma results from well-choreographed activation of autonomic, neuroendocrine, metabolic, and inflammatory responses which are necessary to maintain host homeostasis and facilitate tissue repair. The hypothesis that dysregulation of this response can lead to a non-specific whole-body response (commonly referred to as systemic inflammatory response syndromes (SIRS)) is not new but its mechanisms and pathogenesis are incompletely understood.

Common to the host response to a wide variety of acute tissue injuries e.g., burns, sepsis and trauma, the concept of a 'genomic storm' occurring within hours of major surgery is increasingly recognised. This initial response affects more than 80% of cellular pathways and dynamically alters the leukocyte transcriptome with upregulation of the innate immune system and down regulation of the adaptive immune response. Whilst such a response in evolutionary terms might be considered adaptive, it can have deleterious effects on capillary permeability, immune function and wound healing and predispose to organ dysfunction.

Though simplistic, 'inflammatory markers' have historically been used to describe the magnitude of the postoperative immune response and are consistently associated with post-operative complications following cardiac, thoracic and vascular surgery. Inflammatory injury can be identified in many organs postoperatively (even those remote from the site of surgery - for example, our work demonstrating cardiac inflammation following thoracic surgery) and the magnitude of the postoperative inflammatory response has been associated with multi-organ morbidities including myocardial injury, lung injury, post operative cognitive dysfunction and kidney injury.

Transcriptomic analyses aid understanding of this 'genomic storm'; in patients undergoing major thoracoabdominal surgery Allen et al demonstrated that 522 genes were differentially expressed (in leucocytes) 24-hours after surgery; 248 (48%) were upregulated (reflecting innate immunity and inflammation) and 274 (52%) were downregulated (reflecting adaptive immunity). In this study, within-patient gene expression change was positively associated with postoperative infection and hospital length of stay; in patients where gene expression was most radically altered postoperatively, outcomes were worse.

Host susceptibility to complications

Prediction of perioperative risk is a cornerstone of perioperative medicine. Whilst individual patient and procedural risk factors for complications and mortality exist, and are widely integrated into risk prediction scores, risk prediction remains an imprecise science. Clinicians recognise the heterogeneity in the clinical response to major surgery; apparently similar patients undergoing the same surgical procedure can behave very differently during the post-operative period. This highlights the importance of differing host susceptibility to post-operative organ dysfunction.

The potential for host genomics to impact perioperative outcome has received some study, and whilst it has been possible to identify single nucleotide polymormphisms which might lead to specific perioperative morbidity syndromes (e.g. butyrylcholinesterase deficiency), it is unlikely that any single or combination of polymorphisms is responsible for the wide heterogeneity seen in patient responses to surgery. The potential value of studying genomic susceptibility to complications is further limited by the static nature of the genome; should predictors of outcome be identified, these changes are fixed with no potential for therapeutic manipulation. Because the epigenome is dynamic and as such, may be amenable to therapeutic manipulation or upstream modification, its study provides the potential to not just risk-predict, but risk-modify.

Epigenomics

Epigenomics concerns the study of how cells control gene activity without changing the DNA sequence (i.e. without altering genomic makeup). Cellular DNA is bound in chromatin as a complex of DNA and proteins. Modification of these proteins (classically by DNA methylation or histone modification) will alter the way these proteins interact with DNA and can regulate gene expression changing the way a cell responds to a stimulus.

Classical descriptions of the immune system describe the innate and adaptive responses with only the adaptive immune response being able to build immunological memory. In recent years, the potential for epigenetic reprogramming of innate immune cells in response to stimulation has been recognised. Following an initial challenge (from a pathogen or inflammatory insult), cells return to a non-activated state, but through modification of the epigenome are primed for response to a subsequent stimulus allowing rapid and enhanced recruitment of transcription factors after a secondary challenge. While much of this study has taken place in-vitro, the in-vivo corollary is that an individual's environment - the accumulated effects of aging, prior toxin ingestion, pathogen exposure and the influence of comorbidities - cumulatively influences the epigenome with the potential to fundamentally alter a cells response to a stimulus.

The CD14+ monocyte (the predominant human monocyte sub-species), whose immunological role is to 'sense and respond' is a key player in the innate immune response, reacting early to a host of stimuli. This fundamental role as trigger of the acute inflammatory response makes it a natural cell population for investigation in this setting. Moreover, due to their innate nature (i.e., immediate and prolonged response to stimuli) CD14+ monocytes are unique circulating sensors of physiological state, and thus their epigenome can provide a surrogate readout of how the wider environment has influenced the cellular state.

H3K4me3 and H3K27me3 are modifications to the DNA-packaging histone H3 which have received wide study as two key epigenetic marks accompanying trained immunity. Tri-methylation of H3K4 (the 4th lysine residue of the histone H3 protein) makes the DNA in chromatin more accessible to transcription factors, promoting gene transcription (conferring a so called 'on-state'). Conversely tri-methylation of H3K27 results in downregulation of nearby genes ('off-state'). With simultaneous presence of both H3 marks (known as bivalency), genes are referred to as 'poised'. Such a poised state confers the ability to rapidly become activated in response to a stimulus. Analogously, the engine is revving (H3K4me3), but the hand brake is on (H3K27me3). Release of the 'brake' in response to tissue injury has the potential to result in a rapid and robust immune response. Previous work by our group demonstrates wide variety in the epigenetic signature of circulating CD14+ monocytes in patients with chronic inflammatory diseases.

Whilst on an evolutionary basis, immune training should confer a survival benefit when faced with a potential pathogen, there are examples where such adaptation is deleterious, particularly where it could trigger enhanced tissue damage15. In addition to chronic inflammatory co-morbidities, common indications for cardiac surgery such as ischemic heart disease, alongside patient factors such as cigarette smoking and alcohol consumption, have the potential to serve as primary stimuli for epigenetic modification altering a patient's susceptibility to a secondary response (i.e. surgery).

The role of H3K4me3/H3K27me3 modification in the acute inflammatory response has seen limited study and the influence of such immune priming on CD14+ monocytes in the perioperative immune system response has not been explored. What data does exist, predominantly examining the effects of prior pathogen exposure (e.g. P.falciparum malaria) or inflammatory comorbidity (e.g. IHD) on the ex-vivo response to experimental sepsis speaks of a fundamental influence. Bekkering et al for example demonstrated monocytes from patients with severe symptomatic atherosclerosis (i.e. surgical candidates) to have an approximately 2-fold increase in production of pro-inflammatory cytokines upon lipo-polysaccharide stimulation compared to healthy controls, an effect which appears to be mediated by altered H3K4me3 and H3K27me3 levels on the TNFα, Interleukin-6 and -1ß promotors. Sadahiro and colleagues investigated 30 patients undergoing either major colorectal or orthopaedic surgery, performing genome-wide DNA methylation analysis perioperatively on patient peripheral blood mononuclear cells and identified distinct methylation changes associated with inflammatory gene activation triggered by surgery which persist to the day of discharge, correlating with increased inflammation. The authors discuss a limitation of their study is the diverse populations of leukocytes (monocytes, plasmablasts, CD8+ lymphocytes, CD4+ lymphocytes, Natural Killer lymphocytes) present in their analysis and caution that future studies should analyse specific cell populations (as proposed in the Epi-SCCS study) to accurately determine the inflammatory contribution of distinct leukocyte populations to major surgery. They further acknowledge that they have not performed gene expression analysis and therefore the link between DNA methylation patterns and inflammatory protein abundance is not causally explained. Laudanski et al later investigated the presence of free histones H3 and H3K4me3 and H3K27ac at several time-points (including 3-months post-operatively) in serum of 59 patients undergoing a variety of major cardiac surgeries, demonstrating the presence of "smouldering inflammation" present at 3-months after major cardiac surgery.

In summary Major surgery triggers inflammation as part of the physiological response to tissue injury, however, excessive inflammation in the post-operative period is associated with organ dysfunction, complications and morbidity. Epigenetic markers can dictate the magnitude and duration of powerful inflammatory responses. The investigators aim to study monocyte epigenetic expression, determine subsequent inflammatory gene activation/deactivation patterns (transcriptomics) and quantify inflammation (proteomics) to describe the epigenetic regulation of inflammation and its role in perioperative morbidity. Due to the modifiable nature of the epigenome, future intervention trials can then study existing or new therapies prior to surgery, to recondition the epigenome in 'at risk' patient groups as part of perioperative prehabilitation programmes, to reduce the morbidity associated with major surgery.