Accumulation of endogenous Muse cells in the myocardium and its pathophysiological role in patients with fulminant myocarditis

Abstract

Multi‐lineage differentiating stress‐enduring (Muse) cells, identified as pluripotent surface marker SSEA‐3(+) cells, are stress tolerant endogenous pluripotent‐like stem cells, and are involved in tissue repair. However, the significance of Muse cells in acute myocarditis has not been evaluated. In the present study, we counted Muse cells/area in biopsied myocardial tissue samples from 17 patients with fulminant myocarditis, and 6 with non‐inflammatory myocardial disease as controls. Compared with controls, patients with fulminant myocarditis had significantly more Muse cells (p = 0.00042). Patients with mechanical circulatory support (p = 0.006) and myocardial degeneration (p = 0.023) had significantly more Muse cells than those without them. The Muse cell number was correlated with acute phase CK‐MB level (ρ = 0.547, p = 0.029), indicating the severity of myocardial injury, and was also correlated with acute/recovery phase ratio of CK‐MB (ρ = 0.585, p = 0.023) and cardiac troponin I (ρ = 0.498, p = 0.047) levels, indicating resilience of myocardial injury. In fulminant myocarditis, the Muse cell number was associated with the severity of clinical features in the acute phase, and also with the recovery from myocardial damage in the chronic phase. Endogenous Muse cells might be mobilized and accumulate to the myocardial tissues in fulminant myocarditis, and might participate in the repair of injured myocardium.

Abbreviations

BNP

brain natriuretic peptide

CK‐MB

creatine kinase‐myocardial band

HLA

human leukocyte antigen

Muse cell

Multi‐lineage differentiating stress‐enduring cells

S1P

sphingosine‐1‐phosphate

SSEA‐3

stage‐specific embryonic antigen‐3

Study Highlights.

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

Fulminant myocarditis exhibits various pathologic features and its prognosis is difficult to predict. Furthermore, the mechanism of fulminant transformation remains unknown. Although recent advances in both medical and mechanical treatments have delivered favorable outcomes for fulminant myocarditis, there is still a need to develop further effective approaches.

• WHAT QUESTION DID THIS STUDY ADDRESS?

Whether endogenous Muse cells, which accumulate at damaged sites in the event of tissue injury to repair the tissues, play some significant role in pathophysiology of fulminant myocarditis?

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

This is the first in‐human study to examine the tissue distribution of endogenous Muse cells in myocardium. From this study, we can envision that Muse cells might be mobilized and accumulate to the myocardial tissues in fulminant myocarditis, and might participate in the repair of injured myocardium.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

This study provided a new approach to elucidating the pathogenesis of fulminant myocarditis from the perspective of stem cell biology. The potential effects of exogenous Muse cell administration for the treatment or prevention of fulminant transformation in acute myocarditis would be promising.

INTRODUCTION

Fulminant myocarditis is a rare form of acute myocarditis with a very high mortality risk that is characterized by rapidly progressive sequences of myocardial inflammation, damage, degeneration, and necrosis, eventually leading to hemodynamic instability, cardiogenic shock, and serious arrhythmias. ¹ , ² , ³ Recent advances in immunosuppression therapy and mechanical circulatory/ventilatory support for maintaining end‐organ function, however, have improved clinical outcomes in previously fatal cases. ⁴ Nevertheless, some patients do not respond to these aggressive treatments and die, and some patients who survive may progress to chronic heart failure. ⁵ Because fulminant myocarditis exhibits various pathophysiological features, the prognosis is difficult to predict. Furthermore, the mechanism of fulminant transformation remains unknown and no practical indicators exist.

Multi‐lineage differentiating stress‐enduring (Muse) cells are reparative endogenous pluripotent‐like stem cells that can be identified as cells positive for the pluripotent surface marker stage‐specific embryonic antigen (SSEA)‐3 in the bone marrow, organ connective tissues, and peripheral blood by immunohistochemistry and cell sorting. ⁶ , ⁷ Muse cells secrete factors involved in stress tolerance, such as serpins and 14‐3‐3 proteins, ⁸ and activate homologous recombination as well as non‐homologous end‐joining repair systems to repair DNA damage within a short time period, compared with other stem cells such as mesenchymal stem cells. ⁹ Due to their stress tolerance, Muse cells are able to survive in a strong genotoxic stress environment.

Circulating endogenous or exogenously administered Muse cells selectively accumulate at sites of damage by sensing sphingosine‐1‐phosphate (S1P), a key mediator of inflammation produced by damaged cells. ¹⁰ After homing to the site of damage, Muse cells phagocytose damaged/apoptotic‐differentiated cells, recycle factors such as transcription factors originally functioning in the differentiated cells, and repair the tissue by rapidly differentiating into the same cell type as the damaged/apoptotic cells to replace the damaged cells. ¹¹ In addition to cellular replacement, Muse cells exert pleiotropic effects such as anti‐inflammatory, anti‐fibrotic and trophic effects, that also support tissue repair. ¹⁰ , ¹² , ¹³ Because Muse cells have an immune privilege system represented by the expression of human leukocyte antigen (HLA)‐G, which is relevant to immunotolerance in the placenta, ¹⁰ clinical trials using donor‐derived Muse cells delivered intravenously have been conducted for stroke, acute myocardial infarction, epidermolysis bullosa, spinal cord injury, neonatal hypoxic–ischemic encephalopathy, amyotrophic lateral sclerosis, and COVID‐19 acute respiratory distress syndrome without HLA‐typing or immunosuppressive treatment. ¹⁴ , ¹⁵

Muse cells have been studied in cardiovascular disease. In a rabbit ischemia–reperfusion myocardial infarction model, intravenously injected autogenic‐ and allogenic‐Muse cells selectively recruited to the infarct area, spontaneously differentiated into physiologically active cardiomyocytes as well as into vascular cells, thereby reducing the infarct size and enhancing the recovery of cardiac function. ¹⁰ A clinical study demonstrated that endogenous Muse cells mobilized into the peripheral blood in patients with acute myocardial infarction during the acute phase. An increased number of peripheral blood Muse cells in patients during the acute phase was positively correlated with better recovery of cardiac function and avoidance of heart failure during the chronic phase compared with patients showing no increase in the Muse cell number, suggesting that endogenous Muse cells contribute to tissue repair in acute myocardial infarction. ¹⁶

Acute myocarditis is one of the few diseases that require myocardial biopsy in the acute phase for diagnosis and treatment decisions. In this study, we examined the localization of endogenous Muse cells in myocardial tissues obtained by biopsy and at autopsy from patients with fulminant myocarditis and compared with that in non‐inflammatory myocardial biopsy tissue. We also analyzed the correlation between the myocardial Muse cell number and clinical features. This is the first study examining the localization of endogenous Muse cells in human heart tissue.

METHODS

Subjects and study outline

Patients with acute myocarditis who were hospitalized at Dokkyo Medical University Hospital from May 2016 to July 2022 were screened retrospectively. The diagnosis of acute myocarditis was based on an endomyocardial biopsy performed on the day of or the day after admission (5.5 ± 7.2 days after the onset: median 3 days) in all patients. The endomyocardial biopsy was performed by sampling from the right ventricular septum via venous access using fluoroscopic guidance. From these patients, we selected 17 (51 ± 19 years, 12 men and 5 women) with fulminant myocarditis accompanied by major organ failure caused by hemodynamic disruption, cardiogenic shock, or serious tachy or bradyarrhythmia. As controls, biopsy specimens were also taken from 6 patients (69 ± 5 years, 4 men and 2 women) who underwent myocardial biopsy because of suspected non‐inflammatory myocardial disease such as cardiomyopathy, but whose myocardium was found to have no histopathological abnormalities. Of the 17 patients with fulminant myocarditis, 3 died during hospitalization; 1 of the 3 patients who died of a non‐cardiac cause in the recovery phase of acute myocarditis (31 days after the onset) underwent an autopsy. Samples obtained by biopsy and autopsy were subjected to histopathologic evaluation and Muse cell detection as described below.

Survey for clinical features

We retrospectively surveyed the clinical background of each patient from the patient's chart record to analyze the following items: age; sex; usage of specific medicines such as corticosteroid and/or gamma‐globulin; undergoing mechanical ventilatory support such as ventilator under intratracheal intubation or noninvasive positive pressure ventilation; undergoing blood purification such as continuous hemodiafiltration; undergoing mechanical circulatory support such as intra‐aortic balloon pumping, percutaneous cardiopulmonary support, and/or Impella® heart pump catheter; and in‐hospital mortality. The timing of onset was defined as the time at which the myocarditis symptoms, such as heart failure and arrhythmia, developed.

Echocardiography‐based ejection fraction and biomarkers, such as brain natriuretic peptide (BNP), creatine kinase‐myocardial band (CK‐MB), cardiac troponin I, and C‐reactive protein, were examined at admission (defined as “the acute phase”) and at discharge or just before death (defined as “the recovery phase”), and then the ratio of acute phase and recovery phase values was calculated.

Assessment of histopathologic findings

Tissues obtained by myocardial biopsy were first fixed with 10% buffered formalin and then embedded in paraffin according to the routine hospital procedure. The histopathologic findings were assessed using hematoxylin–eosin staining and Masson trichrome staining. Histopathologic findings specific to acute myocarditis, such as myocardial hypertrophy, myocardial disarray, myocardial degeneration, fibrosis, and inflammatory cell infiltration, were semi‐quantified. Myocardial hypertrophy was considered positive when enlarged cardiomyocytes (>20 μm) were detected. Myocardial degeneration was considered positive when 2 of the following 3 findings were observed: (1) irregular cardiomyocyte size, (2) vacuolar degeneration, and (3) deformed, enlarged, and/or darkly stained nucleus. Myocardial fibrosis was considered positive when fibrosis was observed either in the stromal tissue or the perivascular areas. Inflammatory cell infiltration was classified into 3 grades: grade 0, none; grade 1, focal; grade 2, multifocal or diffuse. Only grade 2 was considered positive for severe inflammation.

Detection of endogenous Muse cells in immunohistochemistry

Muse cells are detected in tissue specimens, including human umbilical code tissue, as SSEA‐3(+) cells, as reported previously. ⁷ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ The human umbilical cord tissue was used to indicate positive and negative controls (Figure S4). Formalin‐fixed paraffin‐embedded sections of human umbilical cord were purchased from Pantomics (#UMB01; Richmond, CA). Tissues obtained by biopsy and autopsy were fixed overnight with 4% paraformaldehyde in phosphate‐buffered saline, embedded in paraffin, and cut into 3‐μm thick sections. After deparaffinization and rehydration, the sections were treated with 0.3% hydrogen peroxide in methanol for 30 min to inactivate intrinsic peroxidase activity and incubated sequentially with anti‐SSEA‐3 antibody (1:200; Thermo Fisher Scientific, MA1‐020) and goat anti‐rat IgM antibody conjugated with horseradish peroxidase (1:200; Jackson ImmunoResearch, 112‐035‐075). Anti‐SSEA‐3 was visualized by 3,3′‐diaminobenzidine tetrahydrochloride and nuclear counterstaining was done by hematoxylin staining. Muse cells were counted as follows: 5 regions were randomly selected at a ×400 magnification and the number of SSEA‐3(+) Muse cells was counted; the mean of the cell numbers per unit area for 5 locations was calculated. Such a quantification was performed by an independent investigator unaware of each patient's clinical features.

Analysis overview

First, we compared the number of Muse cells per unit area in the myocardial tissues between 17 patients with fulminant myocarditis and 6 control subjects. Next, based on clinical background and histopathologic findings, 17 fulminant myocarditis patients were stratified into “Yes” and “No” groups. The Yes group included age over 54 years (median value); male sex; use of specific medicines; undergoing mechanical ventilatory support, blood purification, and mechanical circulatory support; in‐hospital death; time from onset to biopsy within 3 days (median value), and histopathologic findings positive for myocardial hypertrophy, myocardial disarray, myocardial degeneration, fibrosis, and severe inflammation. Then, for each item, the number of Muse cells per unit area in the myocardial tissues was compared between the Yes and No groups. Finally, in 17 fulminant myocarditis patients, we assessed correlations of the number of Muse cells per unit area in the myocardial tissues with acute and recovery phases echocardiography‐based ejection fraction and values of BNP, CK‐MB, cardiac troponin I and C‐reactive protein. In addition, correlations of the number of Muse cells/area with a recovery/acute phase ratio of ejection fraction and an acute/recovery ratio of BNP, CK‐MB, cardiac troponin I, and C‐reactive protein values, which would represent resilience for each item, were also assessed.

Statistical analysis

The normality of the distribution of continuous variables was assessed using the Kolmogorov–Smirnov test with Lilliefors' correlation. Values are expressed as the mean ± SD for normally distributed data, and median values and interquartile ranges for data with a skewed distribution. Categorical variables are expressed as the number and percentage. An intergroup comparison of the number of Muse cells localized in myocardial tissues was performed using an unpaired t‐test as the data were normally distributed. Correlations between Muse cell number and other continuous variables were assessed using the Spearman's rank correlation coefficient since the data showed skewed distribution. p < 0.05 was considered statistically significant.

Ethics statement

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all patients for being included in the study. The study protocol was approved by the Dokkyo Medical University ethics committee (approval number: R‐35‐8J, date of approval: 16th June/2020). All of the patients provided written informed consent for this study when we obtained consent for endomyocardial biopsy. In an autopsy case, the written informed consent was again given from family members when consent for the autopsy was obtained.

RESULTS

Clinical features of patients with fulminant myocarditis

The main clinical features of the 17 patients with fulminant myocarditis are shown in Table S1. For the 3 patients who died during hospitalization, the cause of death was aggravation of heart failure (cardiac cause) in 2 patients and stroke (non‐cardiac cause) in the remaining patient.

Localization of Muse cells in myocardial tissues

The localization of Muse cells was examined in the myocardial tissues obtained by biopsy from the 17 patients with fulminant myocarditis and 6 control patients. Figure 1 shows the mean number of SSEA‐3(+) Muse cells/area measured in 5 randomly selected locations in the myocardial tissues in all 23 patients. In the control group, the number of Muse cells ranged from 6.9 ± 3.6 to 15.9 ± 7.5 cells/mm² (Case 18–23), indicating the presence of SSEA‐3(+) Muse cells in normal myocardial tissue, similar to bone marrow, dermis, adipose tissue, umbilical cord, and other tissues. ⁷ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ The number of Muse cells ranged from 8.0 ± 3.8 to 250.9 ± 68.6 cells/mm² in the patients with fulminant myocarditis (Case 1–17), which was higher than that in the control patients. Cases with in‐hospital death were Cases 6 and 10 (both died of heart failure) and Case 17 (non‐cardiac death). A comparison of the Muse cell number between the patients with fulminant myocarditis and control patients is shown in Figure 2, demonstrating that the Muse cell number was significantly higher in the group of patients with fulminant myocarditis (97.1 ± 77.3 cells/mm²) than in the control group (13.6 ± 3.7 cells/mm²; p = 0.00042). In the 17 patients with myocarditis, no significant correlation was detected between the time from onset to biopsy and the Muse cell number (ρ = 0.100, p = 0.688) (Figure S1 ).

FIGURE 1.

The number of SSEA‐3 positive Muse cells/area measured in 5 randomly selected locations in the myocardial tissues of 17 patients with fulminant myocarditis (Cases 1–17), and 6 control patients with non‐inflammatory myocardium (Cases 18–23). Case 6, Case 10, and Case 17 are patients who died in the hospital. In the 17 fulminant myocarditis patients, the time from onset to biopsy was 5.5 ± 7.2 days (0–23 days, median 3 days). The number of Muse cells ranged from 8.0 ± 3.8 to 250.9 ± 68.6 cells/mm² in the patients with fulminant myocarditis, and from 6.9 ± 3.6 to 15.9 ± 7.5 cells/mm² in control subjects (data are mean ± SD).

FIGURE 2.

Comparison of the number of Muse cells (mean value of measured cell number in 5 locations) in the myocardial tissues between 17 patients with fulminant myocarditis and 6 control patients. The Muse cell number was significantly higher in the myocardial tissues of patients with fulminant myocarditis compared with control subjects.

Figure 3 shows 3 representative cases of Muse cell localization observed in this study. Similar to Muse cells in the bone marrow, dermis, adipose tissue, umbilical cord and other tissues, ⁷ Muse cells in the myocardial tissue were SSEA‐3(+) and had a round shape with a distinct nucleus. The first case was a 54‐year‐old female (Case 9 in Figure 1) with fulminant myocarditis, who was admitted to the hospital on day 4 after onset. This patient developed a critical illness requiring mechanical ventilatory and circulatory support in addition to corticosteroid treatment. Histopathologic findings of biopsy tissue obtained on the day of admission showed diffuse inflammatory cell infiltration. Aggressive treatment led to successful recovery. Her ejection fraction increased remarkably from 15% at admission to 65% at discharge on hospital day 23 and her BNP and CK‐MB levels decreased considerably from 1477 to 64 pg/mL and from 109 to 3 U/L, respectively. In this patient, high accumulation of Muse cells was observed at the site of intensive inflammatory cell infiltration (Figure 3a). The number of Muse cells (250.9 ± 68.6 cells/mm²) was the highest among the 17 patients with fulminant myocarditis (Figure 1).

FIGURE 3.

Representative findings of Muse cell localization in the biopsied myocardial tissues in 3 cases. (a) A 54‐year‐old female (Case 9 in Figure 1) with fulminant myocarditis. The patient developed a critical illness requiring mechanical ventilatory and circulatory support in addition to corticosteroid treatment. Histopathologic findings showed an inflammatory reaction with diffuse inflammatory cell infiltration. Muse cells highly accumulated at the site where inflammatory cell infiltration was strong. The number of Muse cells was 250.9 ± 68.6 cells/mm². (b) A 54‐year‐old male (Case 10 in Figure 1) with fulminant myocarditis. This patient also required mechanical ventilatory and circulatory support. Although histopathologic findings of myocardial tissues showed diffuse inflammatory cell infiltration, only a small number of Muse cells was detected at the site of inflammation. Notably, the number of Muse cells (8.0 ± 3.0 cells/mm²) was the lowest among all 17 patients with fulminant myocarditis. (c) A 59‐year‐old male (Case 21 in Figure 1), who underwent a myocardial biopsy for suspected dilated cardiomyopathy. Histopathologic findings of the myocardial tissues showed a normal myocardium, however, and the patient was not diagnosed with myocarditis. The Muse cell number (6.9 ± 3.6 cells/mm²) was low in this patient.

Case 10 patient was a 54‐year‐old male (Figure 1) with fulminant myocarditis, who also developed a critical condition requiring mechanical ventilatory and circulatory support in addition to corticosteroid treatment. The patient was admitted to the hospital on the day of onset. Histopathologic findings obtained on the day of admission showed diffuse inflammatory cell infiltration. His ejection fraction was 25% at admission. Despite aggressive treatment, he died of heart failure and sepsis on hospital day 30. The CK‐MB level decreased from 115 at admission to 41 U/L just before death, but the BNP level increased from 455 to 1121 pg/mL. Only a small number of Muse cells localized at the site of inflammation in the cardiac tissue (Figure 3b). The number of Muse cells (8.0 ± 3.0 cells/mm²) was the smallest among all 17 patients with fulminant myocarditis (Figure 1).

Case 21 was a 59‐year‐old male in the control group (Figure 1), who underwent myocardial biopsy due to a suspicion of dilated cardiomyopathy, though the histopathologic findings showed normal tissue and the patient was not diagnosed with any myocardial diseases. A lower number of Muse cells (6.9 ± 3.6 cells/mm²) was detected in this case (Figure 3c).

Of the 3 cases in which the patient died in the hospital, Case 17 in Figure 1 (58‐year‐old male) died of non‐cardiac causes at 31 days after onset and underwent autopsy. The patient was admitted to the hospital the day after onset due to the progression of cardiogenic shock following sudden dyspnea. At the time of hospitalization, the patient's cardiac pump function was severely impaired (ejection fraction 28%) and the hemodynamics were disrupted, but aggressive treatment with mechanical circulatory and ventilatory support combined with corticosteroid therapy improved the patient's condition and his cardiac function was normalized within a week of admission. Nevertheless, the patient developed an alveolar hemorrhage and stroke and died after 31 days, even though the ejection fraction increased to 58% the day before death. An autopsy was performed the day after death. Figure 4 shows the localization of Muse cells in the right ventricular myocardium, comparing biopsy (taken on day 3 after onset) (Figure 4a) and autopsy (Figure 4b) specimens. In the biopsy specimen, Muse cells accumulated mainly at the site of inflammation where many lymphocytes infiltrated (Figure 4a). The mean number of Muse cells was 86.1 ± 11.4 cells/mm² (Figure 1). In the autopsy specimen, Muse cells were detected, similar to the biopsy specimen, and accumulated in the myocardium near the endocardium with and without inflammation (Figure 4b). The mean Muse cell number was 89.8 ± 52.0 cells/mm², and Muse cells were located sparsely, including in the pericardium. These findings suggest that the number of Muse cells in the myocardium did not decrease or diminish after death, but in contrast to the biopsy samples, the location of the Muse cells did not associate with inflammatory cell infiltration.

FIGURE 4.

Comparison of Muse cell localization in the right ventricular myocardium between the acute phase biopsy specimen and the late‐phase autopsy specimen in a patient with fulminant myocarditis (58‐year‐old male; Case 17 in Figure 1), who died from a non‐cardiac cause. (a) In the acute phase biopsy specimen (3 days after onset), Muse cells accumulated mainly at the site of inflammation where many lymphocytes infiltrated. The mean number of Muse cells was 86.1 ± 11.4 cells/mm². (b) In the autopsy heart (31 days after onset), Muse cells accumulated in the myocardium near the endocardium with and without inflammation. The mean Muse cell number was 89.8 ± 52.0 cells/mm². The Muse cells are also sparsely located in the pericardium.

The localization of Muse cells in the remaining 19 cases, that is, 14 patients with fulminant myocarditis (Cases 1–8, 11–16) and 5 patients with non‐inflammatory myocardium (Cases 18–20, 22, 23) is shown in Figures S2 and S3, respectively.

Correlation between the myocardial Muse cell number and clinical features

First, we stratified 17 patients with fulminant myocarditis into “Yes” and “No” groups according to each item for background and histopathologic findings and compared the number of Muse cells per unit area. The number of Muse cells/area in the time from onset to biopsy was similar between the Yes group (<median value [3 days]) and No group (≥median value; p = 0.925), and between the patients who died (Yes group) and patients who survived (No group) during hospitalization (p = 0.366). The number of Muse cells was greater in patients who underwent treatment with mechanical circulatory support (Yes group) than in patients who did not (No group; p = 0.006), as well as in patients who showed histopathologic findings of myocardial degeneration (Yes group) compared to patients without such findings (No group; p = 0.023). The number of Muse cells tended to be higher in patients with severe inflammation (grade 2 described in the Methods; Yes group) than in patients without severe inflammation (No group), but the difference was not statistically significant (p = 0.082) (Table 1). These results suggest that Muse cell accumulation might be stronger in more critically ill patients with myocardial damage and inflammation.

TABLE 1.

Accumulated Muse cell number versus clinical background and histopathologic findings.

	No (cells/mm2)	Yes (cells/mm2)	p value
Age > 54 years	77.9 ± 34.0 n = 10	110.5 ± 96.8 n = 7	0.410
Male sex	113.3 ± 85.3 n = 5	90.4 ± 76.7 n = 12	0.593
Specific medications	85.4 ± 82.4 n = 8	107.4 ± 76.0 n = 9	0.576
Mechanical ventilatory support	98.4 ± 98.3 n = 7	96.2 ± 64.7 n = 10	0.956
Blood purification	92.8 ± 82.6 n = 14	117.1 ± 52.7 n = 3	0.637
Mechanical circulatory support	40.2 ± 40.9 n = 7	137.0 ± 72.3 n = 10	0.006
In‐hospital death	105.2 ± 81.6 n = 14	59.2 ± 44.3 n = 3	0.366
Time from onset to biopsy <3 days	98.9 ± 86.0 n = 9	95.1 ± 72.2 n = 8	0.925
Histopathologic findings
Myocardial hypertrophy	112.1 ± 86.9 n = 12	61.2 ± 30.0 n = 5	0.228
Myocardial disarray	96.2 ± 82.5 n = 12	99.4 ± 72.1 n = 5	0.940
Myocardial degeneration	62.7 ± 56.1 n = 10	146.2 ± 80.2 n = 7	0.023
Fibrosis	97.1 ± 89.5 n = 8	97.1 ± 70.4 n = 9	0.998
Severe inflammation	58.3 ± 56.4 n = 7	124.3 ± 80.7 n = 10	0.082

Note: The number of Muse cells was calculated as the mean cell number per unit area measured at 5 locations. Data are shown as mean ± SD.

Next, in a total of 17 fulminant myocarditis patients, we assessed correlations of the number of Muse cells in the myocardial tissues with echocardiography‐based ejection fraction and biomarkers using the Spearman's rank correlation coefficient. The number of Muse cells/area tended to be positively correlated with acute phase (at admission: 2.9 ± 3.8 days from onset) BNP level (ρ = 0.461, p = 0.065) level and showed a significantly positive correlation with acute phase CK‐MB level (ρ = 0.547, p = 0.029). These results suggest that Muse cell accumulation was stronger in patients who developed more serious myocardial injury and severe heart failure in the acute phase. Also, the number of Muse cells/area tended to positively correlate to the recovery (at discharge or just before death: 49.4 ± 30.2 days from onset)/acute phase ratio of ejection fraction (ρ = 0.528, p = 0.080) and showed a significant positive correlation to acute/recovery phase ratio of CK‐MB (ρ = 0.585, p = 0.023) and cardiac troponin I (ρ = 0.498, p = 0.047) levels. These results suggest that acute phase Muse cell accumulation was associated also with resilience of myocardial injury and potentially with cardiac function resilience. In addition, the number of Muse cells/area correlated negatively with the recovery phase C‐reactive protein level (ρ = −0.506, p = 0.043) and positively with the acute/recovery phase ratio of the C‐reactive protein level (ρ = 0.583, p = 0.020), suggesting that acute phase Muse cell accumulation might be associated also with recovery from inflammatory reaction (Table 2). Namely, in fulminant myocarditis, the Muse cell accumulation in myocardial tissues in fulminant myocarditis might be associated not only with its severity but also with its resilience.

TABLE 2.

Echocardiographic and biomarker values vs. Muse cell number in biopsy tissues (correlation analysis).

	ρ	p value
Ejection fraction
Acute phase	−0.352	0.205
Recovery phase	0.129	0.628
Recovery/acute	0.528	0.080
Brain natriuretic peptide
Acute phase	0.461	0.065
Recovery phase	0.006	0.982
Acute/recovery	0.277	0.268
Creatine kinase‐MB
Acute phase	0.547	0.029
Recovery phase	0.126	0.624
Acute/recovery	0.585	0.023
Cardiac troponin I
Acute phase	0.336	0.179
Recovery phase	−0.374	0.135
Acute/recovery	0.498	0.047
C‐reactive protein
Acute phase	0.375	0.228
Recovery phase	−0.506	0.043
Acute/recovery	0.583	0.020

Note: Correlations were analyzed using a Spearman's rank correlation coefficient, based on Muse cell number as independent variables and echocardiographic and biomarker values as dependent variables.

DISCUSSION

In the present study, we observed the localization of Muse cells in the myocardial tissues of patients with fulminant myocarditis using biopsy specimens obtained 5.5 ± 7.2 days after the onset (median 3 days). The findings of the present study suggested that: (1) endogenous Muse cells significantly increased in number during the acute phase of fulminant myocarditis compared with that in normal cardiac tissue, (2) Muse cells distributed mainly at sites of immune cell infiltration in many cases, (3) the severity of fulminant myocarditis correlated to the increase in endogenous Muse cells accumulated in the cardiac tissue, and (4) resilience against fulminant myocarditis also correlated to an increased number of Muse cells. These results suggest that endogenous Muse cells might play some significant role in pathophysiology of fulminant myocarditis.

S1P is a general mediator relevant to tissue destruction and inflammation. S1P production is increased in the injured brain after ischemic challenge, as well as in bronchopulmonary dysplasia. ²¹ , ²² In acute myocardial infarction, S1P is elevated in the blood serum as well as in the border and infarct areas of the cardiac tissue. Similar to inflammatory cells, Muse cells migrate to S1P in a dose‐dependent manner in vitro, and are selectively recruited to damaged tissue by the S1P‐S1P receptor 2 axis. In a rabbit model of acute myocardial infarction, Muse cell homing to the infarct area was largely impeded by treatment with an S1P antagonist or introduction of S1P receptor 2‐siRNA to inhibit S1P receptor activity. ¹⁰ Therefore, endogenous Muse cells are assumed to have increased their number also in the fulminant myocarditis tissue by the same mechanisms. To gain insight into these mechanisms, we can envision the future approaches as follows: (1) an experiment using acute myocarditis animal models to investigate whether injection of donor Muse cells accumulate in myocardium via S1P and whether the injection repair injured myocardium. (2) An observation of co‐localization of S1P with Muse cells in biopsied myocardium of patients with fulminant myocarditis. (3) Simultaneous measurement of circulating Muse cell number and amount of S1P in peripheral blood of fulminant myocarditis patients.

In the present study, the number of Muse cells in the myocardium did not correlate with the time from onset of myocarditis to biopsy, but was greater in patients who underwent mechanical circulatory support, and in patients who showed myocardial degeneration in histopathology. Higher Muse cell numbers also tended to be observed in patients with histopathologic findings of severe inflammation. Also, the Muse cell number tended to be positively correlated with acute phase BNP level and was positively correlated with acute phase CK‐MB level. That is, patients with more serious illness with stronger myocardial damage and inflammation exhibited a higher number of myocardial Muse cells than patients with less critical illness and less myocardial damage/inflammation. Based on these results, Muse cell might accumulate in the myocardium during the acute phase, in association with the severity of fulminant myocarditis. In addition, the number of Muse cells also correlated positively to acute/recovery phase ratio of CK‐MB, cardiac troponin I and C‐reactive protein levels, and negatively to recovery phase C‐reactive protein level. Namely, acute phase Muse cell accumulation might be associated also with resilience of myocardial injury and inflammatory reaction. Also, the Muse cell number tended to positively correlate to the recovery/acute phase ratio of ejection fraction, suggesting potential resilience of cardiac function. The correlation of Muse cell number with acute/recovery phase ratio of CK‐MB was significant, but that with acute/recovery phase ratio of ejection fraction was only a trend, suggesting that Muse cells might first be directly involved in the repair of myocardial injury and then indirectly involved in improving cardiac function. Taken together, in fulminant myocarditis, the Muse cell accumulation in myocardial tissues in fulminant myocarditis might be associated not only with its severity but also with its resilience.

Muse cells are unique in that they exhibit the characteristics of both stem cells and macrophages. Muse cells and macrophages migrate to sites of damage by sensing S1P. ²³ Muse cells express some macrophage‐related markers such as C–C motif chemokine ligand 2, interleukin‐10, and toll‐like receptor 2, and exhibit phagocytotic activity. Muse cells phagocytose apoptotic‐differentiated cells, directly utilize the machinery (such as transcription factors) of the originally functioning differentiated cells, and differentiate into the same cell type as the phagocytosed cells to replace the apoptotic cells. In this manner, Muse cells are suggested to repair damaged tissues. Indeed, Muse cells differentiate into cardiac‐lineage cells after phagocytosing apoptotic cardiomyocytes. ¹¹ In a rabbit model of ischemia–reperfusion acute myocardial infarction, engrafted intravenously injected exogenous Muse cells spontaneously differentiated into physiologically active cardiomyocytes in vivo, which demonstrated synchronous activity in electrocardiogram recordings as well as calcium influx and efflux activity revealed by GCaMP GFP‐based Ca calmodulin probe fluorescence. The injection of autologous‐, allogenic‐ and xenogenic (human)‐Muse cells substantially reduced the infarct size and increases the ejection fraction, compared with injection of mesenchymal stem cells. ¹⁰ Similar effects were also observed in a swine acute myocardial infarction model receiving an intravenous injection of xenograft (human) Muse cells. ²⁴ These previous findings suggest that endogenous Muse cells that accumulated to fulminant myocarditis tissue might have spontaneously differentiated into cardiomyocytes and replaced the damaged cardiomyocytes to repair the injured myocardium; this could explain the correlation of the Muse cell number in the cardiac tissue with recovery indicators (the acute/recovery ratios of CK‐MB and cardiac troponin I).

Acute myocarditis and acute myocardial infarction are similar in that they exhibit pathologic features such as inflammatory response, myocardial damage, degeneration, necrosis, and fibrosis, and occasionally lead to serious states, including death. ¹ , ² , ³ , ⁴ Inflammation located upstream in acute myocarditis, however, but downstream in acute myocardial infarction. In humans, intravenous injection of allograft donor‐derived Muse cells improved left ventricular function in a small number of patients with acute myocardial infarction who underwent emergent coronary revascularization. ¹⁴ Furthermore, in patients with acute myocardial infarction, the number of peripheral blood‐endogenous Muse cells increased in the acute phase (within 1 week after onset) and the increase was more prominent in patients with a higher CK‐MB level. Patients with a higher increase in the number of Muse cells during the acute phase showed a substantial improvement in left ventricular function and suppressed remodeling in the chronic phase (at 6 months) compared with patients who showed no increase in the peripheral blood‐endogenous Muse cell number during the acute phase. ¹⁶ Therefore, in patients with acute myocardial infarction, the severity of myocardial infarction in the acute phase and cardiac functional recovery in the chronic phase significantly was correlated with the increased number of peripheral blood‐endogenous Muse cells in the acute phase, similar to our findings in patients with fulminant myocarditis.

Muse cells were observed not only in the acute phase biopsy specimen, but also in the recovery phase autopsy specimen (31 days after the onset) from the patient who died of a non‐cardiac cause. While the number of Muse cells was quantitively similar in each phase, the location of the Muse cells was different. Muse cells were found mainly at the sites of inflammation in the biopsy specimen, but were also observed in tissue with no inflammation in the autopsy specimen (Figure 4). It is possible that the Muse cells at the sites of inflammation in the acute phase remained in place even after the inflammatory response was over in the recovery phase, although in‐depth analysis with a larger number of specimens is required.

Potential limitations/clinical implications

The present study has a lot of potential limitations. This was a preliminary study with a small sample size. In this study, we examined correlation between the number of Muse cells in the biopsied myocardium and various parameters for clinical features. However, each parameter is influenced by the others, and thus we cannot draw definitive conclusions without considering confounding factors. To draw clinically established conclusions, a multivariate analysis or adjusted analyses would be required, based on a larger sample size. Due to the preliminary nature of this study, however, we did not intend to draw strong conclusions based on statistically significant findings from this study, as we would base on findings from a large clinical trial. The results obtained from each case in the present study, however, suggest that the accumulation of myocardial Muse cells in fulminant myocarditis is clinically relevant. In the present study, myocarditis was diagnosed based on endomyocardial biopsy from the right ventricular septum in all of 17 patients recruited. Recently, however, left ventricular endomyocardial biopsy has been frequently performed under establishing its safety and efficacy. Although right ventricular biopsy is not necessarily ineffective for diagnosing myocarditis, evaluation of Muse cell accumulation in left ventricular biopsy specimens might have revealed a clearer association with the clinical features. However, in any case, it is impossible to discuss the pathology of the entire myocardium based only on very small tissue samples taken from a local area, and the evaluation of myocardial biopsy itself might have limitations.

Fulminant myocarditis is a less frequent disease, but it is a serious disease that can be lethal even with aggressive treatment. Although recent advances in both medical and mechanical treatments delivered favorable outcomes, we need to challenge the development of further effective approaches for diagnosis, examination, and treatments. In fulminant myocarditis, Muse cells are suggested to be mobilized into circulating blood, and then migrate into and accumulate to damaged myocardium, so measuring the number of Muse cells using flow cytometric analysis would help understanding its pathophysiological features. Also, since these Muse cell kinetics are promoted via sensing S1P, measurement of the amount of S1P in the circulating blood would be a possible surrogate marker for Muse cells. Finally, we believe that Muse cell therapy would be promising for the future treatment strategy not only to improve the prognosis of fulminant myocarditis but also to prevent fulminant transformation of acute myocarditis.

CONCLUSION

This is the first in‐human study demonstrating the localization of SSEA‐3‐positive Muse cells in biopsied myocardium in acute phase fulminant myocarditis. The localized Muse cell number was greater in the myocardium of fulminant myocarditis, compared with the control myocardium. In fulminant myocarditis, the localized Muse cell number was related to the severity of clinical features in the acute phase, and also to the resilience of myocardial damage in the chronic phase. Endogenous Muse cells might be mobilized and accumulate to the myocardial tissues in fulminant myocarditis, and might participate in the repair of injured myocardium.

AUTHOR CONTRIBUTIONS

T.I., S.T., M.S., K.I., R.S., and M.D. wrote the manuscript. T.I., S.T., M.S., and M.D. designed the research. S.T., M.S., K.I., Y.K., R.S., H.T., and K.A. performed the research. T.I., S.T., M.S., K.I., and Y.K. analyzed the data. All authors take responsibility for the integrity of the data and the accuracy of the data analysis. All authors read and approved the final manuscript.

FUNDING INFORMATION

This work was supported in part by the Grant‐in‐Aid for Scientific Research from the Japan Society for the Promotion of Science (Grant Number 20K08496).

CONFLICT OF INTEREST STATEMENT

All authors declared no competing interests for this work.

Supporting information

Figure S1‐S4

Table S1

Toyoda S, Sakuma M, Ishida K, et al. Accumulation of endogenous Muse cells in the myocardium and its pathophysiological role in patients with fulminant myocarditis. Clin Transl Sci. 2024;17:e70067. doi: 10.1111/cts.70067

DATA AVAILABILITY STATEMENT

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.