Myocarditis is an important cause of sudden death and chronic heart failure with an estimated worldwide prevalence of 0.5–4.0% (1–3). We showed previously that the heart failure biomarker soluble ST2 (sST2) (which is also called interleukin-1 receptor-like 1/IL1RL1) was significantly elevated in the sera of patients with clinically suspected and biopsy confirmed myocarditis, but only in men with myocarditis that were under 50 years of age (4). Sera sST2 levels were found to correlate with New York Heart Association (NYHA) classification in young men with myocarditis, but not left ventricular (LV) ejection fraction (LVEF) obtained from echocardiography (4). These findings highlight the importance of examining biomarkers according to sex and age. Note that sST2/IL1RL1, which is located on human chromosome 2 and is a member of the IL-1R/Toll-like receptor (TLR)4 family (sST2, GeneID: 9173), is consistently confused with the protein suppression of tumorigenicity 2 (especially in the cardiovascular literature) which is located on human chromosome 11 (ST2, GeneID: 6761) and is associated with cancer.

The ST2 gene (IL1RL1 GeneID: 9173) encodes 4 isoforms of ST2 by alternative splicing: ST2L, a transmembrane receptor for interleukin (IL)-33; sST2, a secreted soluble form of ST2; ST2V, a variant of ST2 found in the gut; and ST2LV, which is secreted by the eye, heart, lung, and liver (5–7), sST2 mainly acts as a decoy receptor to dampen IL-33 cytokine activity by binding to the cytokine (8) but some recent findings indicate that it might also function as an extracellular ligand by binding to an unknown receptor, suggesting a secondary function as an inflammatory mediator or growth factor instead of only as a decoy receptor (9, 10). sST2 is a cytokine that has been shown to be induced by biomechanical strain in cardiac fibroblasts, cardiomyocytes, and vascular endothelial cells that binds IL-33, thereby inhibiting its effect (11, 12). In translational studies we found that serum sST2 levels are associated with increased myocardial inflammation and are increased by testosterone in male mice with viral myocarditis but unaffected by estradiol in females (4). Serum IL-33 and sST2 have been associated with more severe autoimmune diseases (i.e., Sjögren’s disease, myocarditis), asthma, and cancers, but use of knockout mice and recombinant protein or blocking antibody in animal models have yielded conflicting results (13–15). Almost no animal studies have taken into consideration sex, age, or mouse strain in relation to the effect of these cytokines. Although serum sST2 has been clearly demonstrated to display sex differences during myocarditis and chronic heart failure (HF), there is little data available that examines sex and age differences in sST2 levels for specific CVDs, especially for acute or sub-chronic stages of CVD.

Two recent studies examined sera levels of sST2 in large cohorts of patients to determine whether sex-specific cutoffs would enhance determination of risk for hospitalization or death (16, 17). The study by Harmon et al. examined participants from the Rochester Epidemiology Project for cardiovascular outcomes. They found sex-specific differences in sST2 associated with incident HF, HF with reduced EF (HFrEF), HF with preserved EF (HFpEF), major adverse cardiac events (MACE), and mortality (16). Vergaro et al. examined 13 cohorts of patients with stable chronic HF that were part of the Biomarkers In Heart Failure Outpatient Study (BIOS) consortium. Patients were classified as having HFrEF (LVEF ≤ 40%), HF with mildly reduced LVEF (HFmrEF, LVEF 41–49%) or HFpEF (LVEF ≥ 50%) according to the European Society of Cardiology (ESC) guidelines (18, 19). In 4,540 patients with chronic HF, they found that HFrEF was more prevalent in men (88% vs. 73%, p < 0.001), while HFmrEF and HFpEF occurred more often in women (10% vs. 7% and 17% vs. 5%; both p < 0.001), and men had significantly higher serum levels of sST2 and high sensitivity cardiac troponin T than women but there were no sex differences in NT-proBNP (17). Another large study by Suthahar et al. examined sex differences in primary CVD biomarkers including sST2 from several large cohorts and reported that sST2 was elevated in the sera of men with chronic HF and that there was an interaction between sex and age (13). Although previous studies examined age or sex, none of the studies specifically examined the effect of age on sST2 levels according to sex. The purpose of this study was to examine whether sex and age differences exist in serum sST2 levels in several acute CVDs such as myocarditis, cardiomyopathy, atherosclerosis/coronary artery disease (CAD), and myocardial infarct (MI) as well as congestive HF (CHF). Here we would like to gain insight by examining the effect of sex and age on sST2 levels by comparing a number of CVDs.

Demographic data of controls and CVD patients are shown in Tables 1, 2. The age of control subjects that did not have CVDs was matched closely to CVD patients (average age in 60s) (Table 1). In contrast, patients with myocarditis had an average age in the 40s in this study. Myocarditis is a CVD known to occur predominantly in young adults (4). A sex difference in age in our study was observed only in men with myocarditis (average age 38) who were significantly younger than women (average age 46) (p = 0.0002), but not for other CVDs (Table 1). There were no significant differences between men and women in body mass index (BMI) in controls or CVD groups (Table 1). Men with cardiomyopathy (p = 0.013) or CHF (p < 0.000001) had worse cardiac function based on LVEF than women (Table 1). Race was mainly white for controls and CVDs (Table 2). Studies from Mayo Clinic have been reported to have low ethnic diversity (29). However, all groups had a similar ethnic diversity for this study. Patient numbers according to New York Heart Association (NYHA) class are shown in Table 3 and patients separated according to LVEF categories in Table 4.

Although there have been significant advances in the diagnosis and treatment of HF, the condition remains associated with high rates of hospitalization and mortality worldwide (35–37). Sera sST2 has been found to be a powerful predictor of mortality where a change in sera sST2 levels over time predicts NYHA class III or IV HF, subsequent mortality, and the need for transplantation (38–41). Elevated sera sST2 values have been shown to significantly correlate with both LVEF and NYHA class, although only weakly with LVEF (4, 39, 42). Many clinical studies have confirmed that sera sST2 is a powerful predictor of mortality independently and when combined with N-terminal pro-brain natriuretic peptide (NT-proBNP) in patients with HF and LV systolic dysfunction with a LVEF ≤ 45% (38–41). Because many past studies used a LVEF ≤ 45% as a cutoff and to correspond to our previous publication on myocarditis (4), we examined the relationship of sST2 to heart failure using LVEF ≤ 45%. We also analyzed sST2 levels according to sex using the refined ESC definition of HF (18, 19).

Echocardiography is an important tool used to measure LVEF and HF for most hospitals and centers (43, 44). A recent assessment of nearly 5,000 patients with HF found that men had lower LVEF values that were associated with more hospitalization for heart failure and HF-related mortality (45). In this study we show that elevated serum sST2 levels correlate with lower LVEF and higher NYHA class and HF in patients with clinically suspected or biopsy confirmed myocarditis, but only in males (Figures 8A, 10A). Traditional echocardiography measurements are known to be an insensitive measure of cardiac function and future studies should examine the relationship of serum sST2 to echocardiography-derived global longitudinal strain (GLS) (46), which is likely to be a better measure of early and mild cardiac dysfunction (i.e., HFpEF, LVEF ≥ 50%). We published previously that elevated serum sST2 levels correlate with worse disease in myocarditis patients with NYHA class III-IV HF, but only in males (4). These data indicate the importance of analyzing sST2 and other HF biomarkers according to sex. Recent studies have demonstrated that sST2 and other HF biomarkers like high sensitivity cardiac troponin T (hs-cTnT) differ by sex and that sex-specific cutoffs are needed to enhance determination of risk for hospitalization or death from HF (16, 17).

Most studies examining sST2 and HF have not reported data according to sex and age. In this study we found sex and age differences in serum sST2 levels for myocarditis, cardiomyopathy, and MI. Males with myocarditis and cardiomyopathy had higher levels of sST2 and worse HF. However, after age 50 sST2 levels increased in females with myocarditis, while males over 50 years of age with cardiomyopathy or MI had higher levels of sST2. These data indicate that there are important differences that occur during aging in men and women related to sST2. Changes in sex hormone levels or ratios are one important contributor to sex differences in sST2. We showed previously that testosterone increases serum sST2 levels in male mice with coxsackievirus-induced myocarditis, but estrogen (17β-estradiol) did not alter sST2 levels (4). A study that examined the sera of healthy patients found a similar result that sera estrogen levels did not correlate to sera sST2 levels in females while sST2 levels were affected by testosterone in males, similar to our findings in the mouse model (47). Evidence that estrogen/s may be protective comes from our finding that women with cardiomyopathy and early menopause had higher sST2 levels (Figure 5) indicating that estrogen may reduce sST2 levels. A recent study found that women with early or surgical menopause had higher Framingham Risk Scores for CVD than those with normal menopause (48). But other factors may contribute to age-related changes in sST2 in men and women such as shifts over time in mitochondrial function, which has been associated with HF (49).

A number of studies have found that individuals without known CVDs have elevated levels of sST2 and that levels are higher in men than women (4, 13, 20, 47, 50, 51). In a large reference population (n = 1136) from the Framingham Heart Study, sera sST2 levels were found to be significantly elevated in healthy men compared with healthy women where male sex (p < 0.0001) and older age (p = 0.004) predicted higher serum sST2 levels (20). In another study of 528 healthy men and women, sST2 was found to be significantly elevated in men compared with women (47). In this study we also found that sST2 levels from apparently healthy individuals from the Mayo Clinic Biobank also had elevated sST2 levels and that males had higher levels than females (p < 0.0001).

Soluble ST2 (sST2) levels have been found to consistently be elevated in disease states such as infections, CVDs, asthma and autoimmune diseases, and where studies have been conducted, to display sex differences (4, 13, 14). Sex differences in sST2 may be explained at least in part by the role of ST2 in inflammatory responses. IL-33 and ST2/sST2 are members of the IL-1 receptor family that include Toll-like receptor (TLR)4. IL-33 binds the ST2 receptor (ST2L) found on many cell types including innate immune cells like mast cells and macrophages that can drive T helper (Th)1, Th2, and Th17-type immune responses, although the pathway most associated with IL-33/ST2 signaling are Th2 responses and remodeling (52–57). In the context of CVD, Th1- and Th17-associated cytokines increase acute inflammation while Th2- and Th17-associated cytokines promote remodeling and fibrosis, all of which are known to influence the pathogenesis of disease in myocarditis leading to DCM as well as CAD/MI (13, 14, 58–62). IL-33 signaling via the ST2 receptor increases inflammation in many auto/inflammatory diseases such as rheumatoid arthritis, inflammatory bowel disease, Sjögren’s disease, and asthma (52–54). In animal models, sST2 has been found to be released predominantly from CD11b+ and CD4+ immune cells (52), which are the primary cardiac immune cells observed during acute myocarditis in mice and humans (32, 58, 63). IL-33/ST2 signaling has been found to upregulate CD11b, TLR4, caspase-1, and IL-1β in a number of human and animal studies of infection and autoimmune disease (52–54). Previously we reported that CD11b, TLR4, caspase-1, and IL-1β were increased by testosterone in male mice with myocarditis (32, 58, 63). We previously showed that treatment of male mice with recombinant IL-33 during initiation of myocarditis significantly increased perimyocarditis resulting in eosinophilic myocarditis (55). We also showed that serum sST2 levels are increased by IL-1β and that there is a direct association between cardiac IL-1β levels and sST2 levels in the heart during myocarditis in male BALB/c mice (4). TLR4 which leads to IL-1β production is a critical pathway in the pathogenesis of myocarditis, CAD, and MI (59, 61–63), and it may be this interaction between TLR4/IL-1β signaling on CD11b + immune cells (i.e., mast cells, macrophages) that elevates sST2 levels in males, where the ST2 receptor is cleaved to form sST2 in an effort to counteract IL-33. Evidence for this hypothesis comes from previous studies that showed that circulating IL-1β levels correlate with NYHA classification and predict mortality in heart failure patients (20). Inflammation was determined to be the best predictor of death or heart transplantation following acute myocarditis based on endomyocardial biopsies (30). Bartunek et al. found that ST2 production did not correlate with wall stress but rather that endothelial cells and/or inflammation contributed to the release of sST2 (64). These authors further showed that IL-1β induced sST2 secretion from human venous and arterial endothelial cells and suggested that the distinct ability of sST2 to predict HF aside from BNP, which is released by stress, indicates that inflammation may contribute to elevated sST2 levels in HF patients (64). Thus, the consistent finding of elevated sST2 levels in males with various forms of HF suggest that inflammation, and the TLR4-IL-1β pathway in particular, drive the pathogenesis of disease that leads to HF and death.

Limitations of this study include a relatively small sample size, especially for some sub-analyses. For CVDs other than myocarditis obtained from the Mayo Clinic Biobank, we were not able to confirm diagnosis because patients self-reported their disease. However, we selected patients that had echocardiography data available indicating that they had been seen by a cardiologist. Additionally, NYHA class status was available for a significant proportion of the patients further indicating that the Biobank samples were from patients with HF. Another limitation is that we do not know when the initiation of the CVD condition occurred in relation to when the sera was acquired for the Biobank. Future studies should examine whether sex and age differences are found in sST2 levels for these and other CVDs where cases can be confirmed. Age 50 years was used as a cut-off to assess the effect of aging on both men and women, but it remains unclear whether this is the best age cut-off to use to assess the effects of aging in patients with CVD. Another limitation of the study is that we obtained myocarditis samples from various sites mainly in Europe while other CVDs were obtained from the Mayo Clinic Biobank in the US. We also do not have information about medication use or history for the patients and it is possible that certain medications like glucocorticoids could alter sST2 levels. Additionally, myocarditis samples were analyzed using an R&D Systems ELISA kit and converted to Presage kit levels, as previously (4). However, we found that men with myocarditis consistently had significantly higher sST2 levels than women at each study site (data not shown). Additionally, this study provided data on a primarily white population and so may not represent findings in other races and ethnicities.

In conclusion, in this study of a number of mainly self-reported CVDs, we show that serum sST2 levels are elevated according to sex for myocarditis, cardiomyopathy, and MI, especially in particular age groups. These data indicate that there are important sex and age differences that occur during CVD related to sST2. These findings should be followed up using larger well-defined myocarditis, cardiomyopathy, CAD, MI, and CHF cohorts.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Approval was obtained from the local IRB or ethics board at each institution for studies involving patients. Informed written consent or IRB approved waiver of consent was obtained from all patients.

KB and DF: conceptualization and project administration. DB, KB, DD, ED, and SS: data acquisition. CT, MC, JK, JB-V, SP, DM, E-SJ, SL, LB, LC, and DF: sample acquisition. DB, KB, and DF: methodology and data curation. DB, KB, DD, and DF: data analysis. DB and DF: writing—original draft. DB, KB, DD, ED, SS, CT, MC, JK, JB-V, SP, DM, E-SJ, SL, LB, LC, and DF: writing—review and editing. All authors have read and agreed to the published version of the manuscript.