Skip main navigation

Diversity in the Expressed Genomic Host Response to Myocardial Infarction

Originally publishedhttps://doi.org/10.1161/CIRCRESAHA.121.318391Circulation Research. 2022;131:106–108

Meet the First Author, see p 3

Sepsis is a syndrome of life-threatening organ dysfunction caused by a dysregulated host response to infection. Leukocyte gene expression suggests that although most septic patients upregulate inflammatory pathways in response to infection, as many as 40% shift to an immunosuppressed state.1 This heterogeneity may underlie neutral results from sepsis immunomodulation trials.2

The genomic diversity in the host response to infection may parallel that of myocardial infarction (MI). Cardiomyocyte necrosis activates similar mediators as sepsis and these responses may similarly be maladaptive. The systemic inflammatory response to MI can further propagate myocardial injury and is associated with incident heart failure; it may also accelerate coronary inflammation causing recurrent MI. Similar patient-to-patient heterogeneity in host response to MI may exist and may have contributed to unsuccessful efforts to target inflammation during acute MI. The modest correlation between post-MI CRP (C-reactive protein) and troponin suggests that host inflammatory response may have origins beyond infarct size. We hypothesized that leukocyte gene expression patterns could cluster patients with ST-segment–elevation MI into biologically informative host response endotypes, paralleling sepsis.1

In this exploratory study, we examined mRNA expression data from Affymetrix Human Gene 1.0 ST microarrays performed on peripheral blood mononuclear cells isolated using FICOLL liquid density gradient medium from 107 patients with ST-segment–elevation MI (25.2% women; mean age 58.5 years) at a single center and 46 controls (34.8% women; mean age 58.0 years) with stable coronary artery disease.3

Using the 471 probes with the most variable expression (SD≥0.5), we agglomerated patients into endotypes using hierarchical clustering (Pearson centering with Ward’s method). A silhouette analysis suggested that the optimal number of clusters was 2. We identified genes most differentially expressed between the 2 endotypes using moderated t tests (Benjamini-Hochberg false discovery rate [FDR] corrected P<0.05) performed on the entire postfiltered 25 030 probe set. Analyses were performed in GeneSpring (Agilent Technologies, CA), JMP (SAS, NC), and R (r-project.org).

Variable expression was observed between the clusters (Figure [A]). Among 3044 probes differentially expressed between clusters, 5 of the 10 most differentially upregulated genes in cluster 2 (relative to cluster 1) coded for inflammatory proteins, including interleukin-1β, HLA-DQA1 (major histocompatibility complex, class II, DQ alpha), HLA-DQB1 (major histocompatibility complex, class II, DQ beta), and EGR-1 and -2 (early growth response gene 1 and 2; Figure [B]). Cluster 2 had higher expression levels of these transcripts, although in both clusters they were upregulated relative to stable coronary artery disease (eg, interleukin-1β; Figure [C and D]). Conversely, 5 of the 10 genes most differentially upregulated in cluster 1 (relative to cluster 2) coded for proteins related to platelet function and coagulation, including platelet factor 4, integrins IIb and β3, proplatelet basic protein, and TREML1 (triggering receptor expressed on myeloid cells like 1; Figure [B]). In cluster 1, many of these genes were upregulated relative to stable coronary artery disease (eg, integrin, α2b; Figure [E and F]). Platelet activation/coagulation and immune function differed in pathway analysis between the endotypes.

Figure.

Figure. Leukocyte expression profiling to define 2 endotypes of host response in ST-segment–elevation myocardial infarction (STEMI) using agnostic, agglomerative hierarchical clustering based on leukocyte genomic expression at time of clinical presentation.* A, Expression intensities stratified by cluster (red and blue); (B) volcano plot of differentially expressed genes between the 2 clusters (10 most differentially expressed genes labeled); (C) density curves of distributions of normalized IL1B (coding for interleukin-1β) expression intensities stratified by endotype (green: cluster 1; blue: cluster 2; red: control [stable coronary artery disease]); (D)scatter plot and overlaid nonparametric densities for normalized IL1B expression by cluster; (E) density curves; and (F) matrix for ITGA2B (coding for integrin, α2b [platelet glycoprotein IIb of IIb/IIIa complex]) stratified by cluster, as in C and D. *Clinical traits did not segregate by endotype, including: age (cluster 1 vs 2: 60.3 vs 57.6 y; P=0.20), sex (23.8% vs 20.2%; P=0.87), diabetes (26.3% vs 21.7%; P=0.77), Thrombolysis in Myocardial Infarction (TIMI) STEMI risk score (2.3 vs 2.5; P=0.25), left ventricular ejection fraction (48.7% vs 49.7%; P=0.59), and the presence of multivessel coronary artery disease (39.5% vs 46.5%; P=0.63). ATP5E indicates ATP synthase F1 subunit epsilon; ELOVL7, ELOVL fatty acid elongase 7; FOSB, FosB proto-oncogene, AP-1 transcription factor subunit; ITGB3, integrin subunit beta 3; MOP-1, mediator of paramutation1; NR4A2, nuclear receptor 4A2; PF4, platelet factor 4; PPBP, pro-platelet basic protein; PTGS2, prostaglandin-endoperoxide synthase 2; RASGEF1B, RasGEF domain family member 1B; SDPR, serum deprivation response; SH3BGRL2, SH3 domain binding glutamate rich protein like 2; SPARC, secreted protein acidic and cysteine rich; and TUBB1, tubulin beta 1 class VI.

Admission CRP was greater in endotype 2 compared with endotype 1 (median [interquartile range], 6.95 [4.31–14.02] versus 3.68 [1.35–12.64]; P=0.049), with a trend persisting at discharge (9.78 [5.61–17.22] versus 5.61 [2.03–27.3]; P=0.24). A gene expression signature of endotoxin tolerance—reflecting an immunosuppressed response to infection—was upregulated in cluster 1 (P<0.01). Clinical traits (age, sex, diabetes, Thrombolysis in Myocardial Infarction, ST-segment–elevation MI risk score, ejection fraction, and multivessel coronary artery disease) were similar by endotype.

External validation was performed by separately hierarchically clustering 934 patients with ST-segment–elevation MI in an independent cohort4 into 2 groups (232 and 702 individuals) based on Illumina HT12v4-profiled peripheral blood mononuclear cell expression (median time 21-hour between cardiac catheterization and blood sampling). Probes with most variable expression (SD≥0.5, 216 probes, excluding ribosomal genes) were used. From the 20 most differentially expressed genes (Figure [B]) in the discovery cohort, 19 were available in the validation cohort of which 14 (74%) were significantly differentially expressed (FDR≤0.05).

These exploratory findings highlight heterogeneity in host response to MI. Among thousands of transcripts examined, agnostic clustering grouped patients based on variability in mediators of inflammation and coagulation or platelet activation. Profiling heterogeneity in host response to MI could provide opportunities to reduce incident heart failure and recurrent MI, possibly through more personalized application of immune-modulating treatments.5

This exploratory analysis is limited by uncertain relative leukocyte subclass abundances, limited infarct size measures, and low post-MI event rates that prevented linking endotypes with outcomes—larger studies are needed. Expression of key transcripts overlapped between endotypes, highlighting that although clustering may categorize patients into discrete subgroups, variability occurs on continuous, overlapping scales—potentially reflecting links between inflammation and thrombosis—and may be challenging to categorize. These hypothesis-generating observations support characterizing heterogeneity in host response to MI and potential overlap with infection. Future, larger studies should consider cluster number stability and, importantly, recruitment of context-specific control subjects such as those undergoing elective coronary interventions to better understand host response timing.

Article Information

Acknowledgments

Data from the discovery cohort are publicly available (E-GEOD-59867); data from the validation cohort, as well as code to recreate the analysis, are available upon request to the corresponding author. Some data were deposited on Zenodo (https://zenodo.org/record/6460228#.YnFQznbMI2w).

Nonstandard Abbreviations and Acronyms

CRP

C-reactive protein

MI

myocardial infarction

Disclosures Dr Lawler has received unrelated research funding from the Canadian Institutes of Health Research, the National Institutes of Health (National Heart, Lung, and Blood Institute), the Peter Munk Cardiac Centre, the LifeArc Foundation, the Thistledown Foundation, the Ted Rogers Centre for Heart Research, the Medicine by Design Fund, the University of Toronto, and the Government of Ontario. Dr Lawler has received unrelated consulting honoraria from Novartis, CorEvitas, and Brigham and Women’s Hospital, as well as unrelated royalties from McGraw-Hill Publishing. The other authors report no conflicts.

Footnotes

For Sources of Funding and Disclosures, see page 108.

Correspondence to: Patrick R. Lawler, MD, MPH, FRCPC, Peter Munk Cardiac Centre at Toronto General Hospital, RFE3-410, 200 Elizabeth St, Toronto, ON, Canada M5G 2C4. Email

References

  • 1. Davenport EE, Burnham KL, Radhakrishnan J, Humburg P, Hutton P, Mills TC, Rautanen A, Gordon AC, Garrard C, Hill AV, et al. Genomic landscape of the individual host response and outcomes in sepsis: a prospective cohort study.Lancet Respir Med. 2016; 4:259–271. doi: 10.1016/S2213-2600(16)00046-1CrossrefMedlineGoogle Scholar
  • 2. Lawler PR, Fan E. Heterogeneity and phenotypic stratification in acute respiratory distress syndrome.Lancet Respir Med. 2018; 6:651–653. doi: 10.1016/S2213-2600(18)30287-XCrossrefMedlineGoogle Scholar
  • 3. Maciejak A, Kiliszek M, Michalak M, Tulacz D, Opolski G, Matlak K, Dobrzycki S, Segiet A, Gora M, Burzynska B. Gene expression profiling reveals potential prognostic biomarkers associated with the progression of heart failure.Genome Med. 2015; 7:26. doi: 10.1186/s13073-015-0149-zCrossrefMedlineGoogle Scholar
  • 4. Teren A, Kirsten H, Beutner F, Scholz M, Holdt LM, Teupser D, Gutberlet M, Thiery J, Schuler G, Eitel I. Alteration of multiple leukocyte gene expression networks is linked with magnetic resonance markers of prognosis after acute ST-elevation myocardial infarction.Sci Rep. 2017; 7:41705. doi: 10.1038/srep41705CrossrefMedlineGoogle Scholar
  • 5. Lawler PR, Bhatt DL, Godoy LC, Lüscher TF, Bonow RO, Verma S, Ridker PM. Targeting cardiovascular inflammation: next steps in clinical translation.Eur Heart J. 2021; 42:113–131. doi: 10.1093/eurheartj/ehaa099CrossrefMedlineGoogle Scholar

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.