Aging influences the cardiac macrophage phenotype and function during steady state and during inflammation

Abstract Aging‐mediated immune dysregulation affects the normal cardiac immune cell phenotypes and functions, resulting in cardiac distress. During cardiac inflammation, immune activation is critical for mounting the regenerative responses to maintain normal heart function. We investigated the impact of aging on myeloid cell phenotype and function during cardiac inflammation induced by a sub‐lethal dose of LPS. Our data show that hearts of old mice contain more myeloid cells than the hearts of young mice. However, while the number of monocytic‐derived suppressor cells did not differ between young and old mice, monocytic‐derived suppressor cells from old mice were less able to suppress T‐cell proliferation. Since cardiac resident macrophages (CRMs) are important for immune surveillance, clearance of dead cells, and tissue repair, we focused our studies on CRMs phenotype and function during steady state and LPS treatment. In the steady state, we observed significantly more MHC‐IIlow and MHC‐IIhigh CRMs in the hearts of old mice; however, these populations were decreased in both young and aged mice upon LPS treatment and the decrease in CRM populations correlated with defects in cardiac electrical activity. Notably, mice treated with a liver X receptor (LXR) agonist showed an increase in MerTK expression in CRMs of both young and old mice, which resulted in the reversal of cardiac electrical dysfunction caused by lipopolysaccharide (LPS). We conclude that aging alters the phenotype of CRMs, which contributes to the dysregulation of cardiac electrical dysfunction during infection in aged mice.

cardiac aging is due to left ventricular hypertrophy and fibrosis, resulting in diastolic dysregulation (Costantino et al., 2016;Kovacic et al., 2011). The elderly have attenuated immune function due to senescence of the adaptive immune system, which is accompanied by lowgrade chronic inflammation termed "inflammaging" (Cevenini et al., 2013;Franceschi et al., 2000). Inflammaging is linked to increased risk of chronic diseases, such as cancer, cardiovascular diseases, infectious diseases, and premature death (Bruunsgaard & Pedersen, 2003;Pawelec et al., 2014;Singh & Newman, 2011). Cardiac aging is characterized by cardiomyocyte hypertrophy, inflammation, and the gradual development of cardiac fibrosis (Meschiari et al., 2017). In addition, the aging heart shows decreased elasticity and increased stiffness (Cieslik et al., 2011;Oh et al., 2017). Emerging evidence from studies in experimental animals and in humans has emphasized that excessive reactive oxygen species (ROS) and superoxide generated by oxidative stress and low-grade inflammation (inflammaging) are the major causes for age-related cardiovascular diseases (CVD) (de Almeida et al., 2020;Panth et al., 2016;Papaconstantinou, 2019;Sies, 2015;Wu et al., 2014). However, the mechanism of infection-mediated cardiac dysfunction during aging and the role of aged cardiac immune cells is not well studied. Recently, we demonstrated that Mycobacterium avium infection in old mice resulted in dissemination of mycobacteria into the heart tissue and caused cardiac dysfunction (Headley et al., 2019). M. avium infected aged mice had significant premature atrial contractions and cardiac dysrhythmia, increased inflammation, and cardiac fibrosis.
Infection with pathogens can also lead to sepsis, which is characterized by excessive inflammatory response to the pathogen (van der Poll et al., 2017). Sepsis is a disease of the elderly with more than half of sepsis cases occurring in patients older than 65 years (Martin et al., 2006;Mayr et al., 2010). Sepsis caused by gram-negative bacteria is largely due to release of lipopolysaccharide (endoxin), which triggers the inflammatory response by binding to the Toll-like receptor TLR4. In the heart, the inflammatory response induced by sepsis causes decreased ventricular function and myocardial damage, which may progress to decreased cardiac output and death (Parker et al., 1984). While studies have defined the damage to the heart caused by severe sepsis, less is known about the damage to the heart caused by low-level inflammation. Thus, in the current study, we investigated the effect of aging on the composition of myeloid cells in the heart during steady state and after an inflammatory event induced by a sub-lethal dose of LPS.
The heart is immunologically active even during normal functioning and contains all major leukocyte populations. The majority of leukocytes present in the steady-state functioning adult mice heart are F4/80 + CD11b + macrophages, whereas other immune cells, including mast cells, dendritic cells (DCs), B cells, and regulatory T cells, are found minimally in cardiac tissue (Epelman et al., 2014(Epelman et al., , 2015Ramos et al., 2017). Monocytes and neutrophils are not normally seen in myocardial tissue of the adult steady-state heart, but are seen during an inflammatory stage (Epelman et al., 2014(Epelman et al., , 2015Ramos et al., 2017). Macrophage numbers in the heart of mice have been reported to either decrease or increase with age. A study by Ramos et al. (2017) showed that mice aged 12-15 months have decreased numbers of monocytes/macrophages and increased numbers of neutrophils.
Other studies (Ma et al., 2018;Salminen et al., 2019) have reported that cardiac macrophages increase in number beginning at 18 months of age and that the number positively correlates with age.
The mouse heart at steady state contains macrophages subsets with distinct functions and origins. The cardiac macrophages can be grouped into three subsets based on their origin and their expression of CC chemokine receptor 2 (CCR2) and MHC class II. The two CCR2 − macrophages subsets (CCR2 − MHC-II high , CCR2 − MHC-II low ) are of embryonic origin and maintain their numbers through self-renewal rather than through infiltration of blood monocytes.
The third subset is CCR2 + MHC-II high macrophages, which are maintained by blood monocyte recruitment. Transcriptional profile analysis showed that the CCR2 + macrophages are enriched for genes involved in inflammatory pathways, suggesting that these macrophages are inflammatory in nature (Lafuse et al., 2020). All CRM subsets were efficient in taking up dead cardiomyocytes, indicating the critical role of CRMs in clearing dead cells and cell debris. The cells that most efficiently engulfed dead cells in the cardiac environment were the CCR2 − MHC-II low macrophages, which can clear dead cells without triggering an immune response (Epelman et al., 2014). Also, MerTK, a phagocytic receptor involved in phagocytosis of apoptotic cells, is highly expressed in cardiac resident macrophages (Ma et al., 2018;Salminen et al., 2019). Thus, each of the CRM subsets has a distinct role in the response to cardiac injury, triggering an inflammatory response, clearing dead cells and debris, or acting as antigen-presenting cells. However, the cardiac immune cell phenotypic changes during infection in aged mice are not known.
In the current study, we examined the effect of aging and inflammation on the phenotypic changes of myeloid cells and macrophage functions in the heart. We observed that aging increased the number of neutrophils and monocytes in steady-state hearts, but did not significantly increase total macrophage numbers. After low-dose LPS injection of mice, the numbers of total CRMs, MHC-II high CRMs, and MHC-II low CRMs were significantly reduced in both young and old mice. CRMs expressing MerTK were also reduced in young and old mice. Further, we report that low-dose LPS significantly altered cardiac electrical activity in both young and old mice; however, the effect of LPS was more prominent in old mice. We also report that treatment of bone marrow-derived macrophages with a LXR agonist increases MerTK expression, and in vivo treatment with the LXR agonist increased MerTK expression in all three CRM populations in both young and old mice. Furthermore, we found that the LXR treatment reversed the LPS-induced cardiac electrical dysfunction in old mice.

| Aging enhances the accumulation of myeloid cells in the heart
Myeloid immune cells play a critical role in steady-state heart function, immune surveillance, cardiac inflammation, and fibrosis (Epelman et al., 2014(Epelman et al., , 2015Ramos et al., 2017). Aging-mediated chronic inflammation is associated with many diseases, including cardiovascular diseases (Bruunsgaard & Pedersen, 2003;Costantino et al., 2016;Pawelec et al., 2014;Singh & Newman, 2011). Therefore, we examined whether aging alters the myeloid cell population in steady-state hearts. We harvested hearts from young (6-8 weeks old) and old (18 months old) C57BL/6 mice. Single-cell suspensions were prepared by enzymatic digestion (Meeson et al., 2013) and analyzed by flow cytometry. The gating strategy and representative flow cytometry plots are shown in Figure 1 for old mice (a) and young mice (b). The fluorescence minus ones (FMO's) for all myeloid cell markers are shown in Figure S1. We identified leukocytes, myeloid cells, neutrophils, monocytes (Ly6C high and Ly6C low ), and macrophages. Total leukocytes (CD45 + cells) were considerably more prevalent in the hearts of old mice compared with those of young mice (Figure 1c).
This finding is consistent with our previously reported immunohistochemistry data, which showed a larger CD45 + leukocyte population in the cardiac muscle of old mice compared with that of young mice (Headley et al., 2019). Furthermore, lymphocyte-negative CD11b + myeloid cells were significantly more prevalent in the hearts of old mice compared with those of young mice ( Figure 1d). To determine the specific phenotypes of myeloid cells that contribute to the difference in CD11b + cell numbers, we analyzed the number of neutrophils (Ly6G + ), monocytes (Ly6C + ), and macrophages (F4/80 + ) in heart tissue of old and young mice. Among the CD11b + myeloid cells, neutrophils and both Ly6C high/low monocyte subsets were significantly more prevalent in old mice compared with young mice (Figure 1eh). However, there were no significant differences in the total CRM population ( Figure 1i) in the steady-state hearts of old and young mice. Macrophages efferocytose apoptotic cells via phosphatidylserine receptors, such as MerTK. Thus, we determined the mRNA expression of MerTK in the CRMs isolated from the hearts of young and old mice. Interestingly, we found that MerTK mRNA expression in CRMs of old mice at steady state was slightly lower than in young mice ( Figure S2). Thus, we conclude that hearts of old mice contain a significantly increased number of myeloid cells, which suggests that aging-mediated chronic inflammation persists in the heart tissue.

| Aging affects myeloid-derived suppressor cell populations and their ability to suppress T-cell proliferation
Since myeloid-derived suppressor cells (MDSCs) function as a regulators of inflammation and T-cell activation (Salminen et al., 2019), we examined the MDSC population in the hearts of young and old mice during steady state. We quantified the monocytic (mMDSCs) and granulocytic (gMDSCs) subsets of MDSCs by CD11b and GR1 expression levels. GR1 binds to both the Ly6C and Ly6G antigens, but at an epitope distinct from the epitopes recognized by the Ly6C and Ly6G antibodies. Plots of CD11b and GR1 expression identified F I G U R E 1 Phenotypic analysis of myeloid cells in the hearts of young and old mice. The myeloid cell abundance in the heart of old and young was examined by flow cytometry. Old (18 months) and young (6-8 weeks) C57BL/6 mice were euthanized and hearts digested to obtain single-cell suspensions. Cells were labeled with different myeloid cells markers and analyzed by flow cytometry. Representative flow cytometry plots and gating strategy are shown for heart myeloid cells from old (a) and young (b) mice. The total number of cells per heart from old and young mice was determined for (c) leukocytes (CD45 + ), (d) myeloid cells (CD11b + ), (e) neutrophils (CD11b + Ly6G + ), (f) monocytes (CD11b + F4/80 − ), (g and h) Ly6C high and low monocytes (CD11b + F4/80 − Ly6C high/low ), and (i) macrophages (CD11b + F4/80 + ). Data shown in the graphs are from a representative experiment of three independent experiments (4 mice per group). A two-tailed Student's t test was used to analyze the data (*p < 0.05 and ** p < 0.005) These studies suggest that while the numbers of mMDSCs in the hearts of old mice at steady-state do not differ from young mice, mMDSCs from old mice are deficient in suppressing T-cell proliferation, which may contribute to chronic inflammation.

| MerTK-positive CRM population is decreased in the hearts of old and young mice upon endotoxin (LPS) treatment
Myocardial dysfunction is a common complication of severe sepsis and increased mortality in ICU patients (Deutschman & Tracey, 2014;Romero-Bermejo et al., 2011). The intraperitoneal injection of LPS has been extensively used to model many of the clinical features of sepsis, including hyper-inflammation and cardiac dysfunction (Drosatos et al., 2013;Fallach et al., 2010). To test whether low-dose LPS treatment alters the cardiac myeloid cell phenotypes, we infused a sub-lethal dose of LPS (2.5 mg/kg) via intraperitoneal injection and examined the cardiac immune cell phenotypes (flow cytometry) and monitored the cardiac electrical activity (ECG) of young and old mice. We followed the gating strategy used in

| LPS treatment causes cardiac electrical dysfunction
Since we found that a significant loss of CRMs in low-dose LPSinfused young and old mice, we then examined the cardiac electrical activity in LPS-injected young and old mice. We found that LPS injection significantly altered the cardiac electrical activity. LPS increased the RR and QT intervals and decreased the heart rate in both young and old mice (Figure 4a-c). Also, we found variability in RR intervals and in heart rate among individual mice. However, the effect of LPS on cardiac electrical dysfunction was more prominent (increased RR, QT intervals, and decreased heart rate) in old mice, which indicates that these mice were experiencing the cardiac arrhythmia. The graph in Figure S3 shows the variability in RR and QT intervals from representative mice. Together the data suggest that LPS-induced cardiac electrical dysfunction may be due to the decreased CRMs in the heart tissue.

| LXR agonist treatment enhances MerTK expression in macrophages in vitro and in vivo
Given our observation that cardiac macrophages from LPS-infused MerTK expression can be enhanced by LXR activation or peroxisome proliferator-activated receptor (PPAR)γ inhibition in human macrophages (Zizzo & Cohen, 2015). To confirm whether MerTK in mice is also regulated by LXR activation or PPARγ inhibition, we treated BMDMs from young mice with the PPARγ antagonist (GW9662-10 μM) or the LXR agonist (T0901317-1 μM) for 24 h.
MerTK mRNA levels were determined by qRT-PCR and protein levels by Western blotting. Our results show that the treatment of BMDMs with LXR agonist (T0901317) increased MerTK mRNA compared to DMSO control treatment, in contrast, we found that the F I G U R E 2 Quantification of monocyte-derived suppressor cells in the hearts of young and old mice. Old mice contain fewer monocytic MDSCs in the heart than young mice and are less efficient in suppressing T-cell proliferation in vitro. Heart single-cell suspensions from old and young mice were labeled with antibodies specific for myeloid-derived suppressor cells (MDSCs) and analyzed by flow cytometry. (a and b) representative gating strategies for heart myeloid-derived suppressor cells in the heart of old and young mice, respectively. The graphs show the total number of cells per heart from old and young mice. (c) Neutrophils (CD11b + Gr1 high -P2), (d) Total MDSCs (CD11b + Gr1 int+ -P1), (e) Granulocytic MDSCs (CD11b + Gr1 int Ly6G + Ly6C + ), and (f) Monocytic-MDSCs (CD11b + Gr1 low Ly6G − Ly6C + ). Monocytic and granulocytic MDSCs were sorted from heart single-cell suspensions of old and young C57BL/6 mice and co-cultured ex vivo with Naïve CD4 + T cells isolated from the spleens of young C57BL/6 mice and in vitro stimulated with anti-CD3/CD28. (g and h) T-cell proliferation index in the presence of granulocytic MDSCs and monocytic MDSCs, respectively, from the heart of old and young mice. The data shown in the graphs are cumulative from two independent experiments (3 mice per group). A two-tailed Student's t test was used to analyze the data (*p < 0.05 and **p < 0.005) PPARγ antagonist (GW9662) treatment did not increase the expression of MerTK ( Figure S4a). In agreement with the qRT-PCR data, MerTK protein expression levels were higher after treatment with the LXR agonist (T0901317) but not with the PPARγ antagonist, as shown in Figure S4b,c. These results provide evidence that treatment with LXR agonist (T0901317) induces MerTK expression in mouse macrophages, but PPARγ is not involved. Based on these in vitro observations, we next determined whether treatment of mice with the LXR agonist increases MerTK in vivo.

| LXR agonist treatment enhances the accumulation of MerTK-positive macrophages and reverses the cardiac electrical dysfunction caused by LPS in the mice
Having determined that administration of LXR in young and old mice enhances the number of MerTK + CRMs in the hearts, we repeated the experiment and tested whether LXR treatment protects the mice from LPS-mediated cardiac electrical dysfunction. We administered LXR agonist or vehicle control to young and old mice for 12 days and subsequently delivered LPS via intraperitoneal route. The CRM phenotype, population, and heart function were examined as described above. Our flow cytometry data show that LXR treatment increased the number of MerTK + CRMs in the hearts of young and old mice by 1.5-fold compared with vehicle-treated control mice (Figure 6a

QRS peaks with intermittent increase in RR intervals) and increased
QT intervals ( Figure S6) in both young and old mice, which was decreased by LXR agonist treatment. In conclusion, our data suggest that the LXR agonist is a potential target for cardio-protection in old mice during cardiac infection or inflammation.

| DISCUSS ION
Since the elderly population is increasing, there is a greater need to understand age-related changes and stressors (Franceschi et al., 2000). Although acute inflammation is important and helpful in protecting the body from infection, constant tissue damage, repair, and adaptation of tissues can lead to chronic, low-grade inflammation.
Such inflammation can be harmful by offsetting the normal functioning of tissues, ultimately leading to a decline in immune system function of elderly people (Calder et al., 2017). The elderly also commonly suffer from cardiovascular diseases, which are linked to dysregulated immune function (Bruunsgaard & Pedersen, 2003;Pawelec et al., 2014;Singh & Newman, 2011). Recently, the cardiac immune system has emerged as a critical player in maintaining steady-state heart function, but the cardiac immune system

F I G U R E 4
Aging augments the LPS-mediated cardiac electrical dysfunction. Surface electrocardiogram (ECG) recordings were obtained from old and young mice injected with LPS (2.5 mg/kg) or mock-treated control old and young mice at baseline and at 24 and 48 h posttreatment by Powerlab 4/30 (AD Instruments). ECG traces were analyzed using LabChart 8 Pro (AD Instruments). Shown are accumulative graphs of RR intervals (a), QT intervals (b), and heart rate (c) of young and old mice (N = 8). A one-way ANOVA followed by the Bonferroni test was used to analyze the data (*p < 0.05; **p < 0.005; and ***p < 0.0005) is dysregulated by aging (Cieslik et al., 2011;Loffredo et al., 2014;Ramos et al., 2017). Myocardial aging is also associated with increased interstitial fibrosis, hypertrophy, inflammation, stiffness, and mild contractile dysfunction (Loffredo et al., 2014). Increased infiltration and differentiation of CD45 + cells of myeloid origin is the major driver for aging-mediated cardiac dysfunction (Cieslik et al., 2011). Furthermore, we showed in prior studies that M. aviuminfected old mice had significant dysrhythmia, cardiac hypertrophy, increased recruitment of CD45 + leukocytes, cardiac fibrosis, and increased expression of inflammatory genes in heart tissue (Headley et al., 2019). To develop therapeutic strategies that ameliorate the inflammation that leads to cardiac dysfunction, it is crucial to understand the pathogenesis and mechanism of myocarditis induced by systemic inflammation.
In this study, we determined the effects of aging on the phenotype and function of cardiac immune cells and began to define the mechanisms by which aging causes cardiac inflammation in either the healthy condition or after low-dose LPS treatment. We used a refined imaging approach to obtain accurate cell numbers and tissue distribution patterns for cardiac CD45 + cells in the steadystate condition in old and young mice. We expanded the findings of Campos Ramos et al. (2017), which showed that age-related myocardial impairment occurs in parallel with shifts in the composition of tissue-resident leukocytes and with an accumulation of activated CD4 + T cells in the heart and draining lymph nodes.
In line with those findings (Ramos et al., 2017) There is convincing evidence that aging is associated with increased production of myeloid cells, including MDSCs, in order to suppress inflammaging, which results in inimmunosenescence (Salminen et al., 2019). However, the infiltration of these specific cell types and their subsets (mMDSCs and gMDSCs) into the heart tissue of aged mice and their involvement in cardiac dysfunction has not been studied. Thus, in the current study, we compared total MDSC populations and their subsets in the hearts of old and young mice at steady-state condition. In contrast to previous findings of higher frequencies of MDSCs in circulation (Salminen et al., 2019), we found that total MDSCs and more specifically, gMDSCs, were lower in the heart tissue of aged mice. While mMDSCs numbers did not differ between young and old mice, mMDSCs from old mice compared with mMDSCs from young mice were less able to suppress T-cell F I G U R E 6 LXR agonist treatment enhances the number of MerTK + CRMS and preserves the cardiac electrical function in old mice. Young and old female C57BL/6 mice were intraperitoneal injected either with T0901317 (50 mg/kg/day) or vehicle for 9 days in separate experiments. At day 10, mice were injected intraperitoneally with LPS (2.5 mg/kg) for 48 h and then euthanized and single-cell suspensions were analyzed by flow cytometry: old (a) and young (b). Graphs shown in (c and d) are the total number of macrophages (CD45 + CD11b + Ly6G -F4/80 + ) and MerTK + macrophages in the heart tissue of vehicle and LXR agonist (T0901317) treated young and old mice followed by LPS injection, respectively. Surface electrocardiogram (ECG) recordings were obtained from control and LXR-treated mice with and without LPS injection (2 mg/kg) of young and old mice at baseline and at 24 h post-treatment by Powerlab 4/30 (AD Instruments). ECG traces were analyzed using LabChart 8 Pro (AD Instruments). Shown are representative graphs of RR intervals in young (e) and old (f) mice, QT intervals of young (g) and old (h) mice and heart rate of young (i) and old (j) mice. Data shown are representative of 5 mice (N = 5). A one-way ANOVA followed by the Bonferroni test was used to analyze the data (*p < 0.05; **p < 0.005; and ***p < 0.0005) proliferation ex vivo (Figure 2). In contrast, gMDSCs from young and old mice were equal in suppressing T-cell proliferation. Thus, our data indicate that mMDSCs are deficient in suppressing T-cell activation. This impairment becomes more critical in the case of HF or infection, which can lead to cardiac dysfunction. However, the literature concerning the role of MDSCs during HF, myocarditis, and infection is controversial (Cuervo et al., 2011;Fresno & Girones, 2018;Zhou et al., 2018). Zhou et al. reported that the number of MDSCs was significantly elevated during HF and correlated with increased levels of proinflammatory cytokines, given their compensatory role in resolving inflammation (Zhou et al., 2018). Furthermore, depletion of MDSCs aggravated cardiac hypertrophy, cardiac dysfunction, and inflammation, and increased T-lymphocyte accumulation in the heart (Zhou et al., 2018). However, the adoptive transfer of MDSCs mitigated the overloaded pressure of cardiac hypertrophy and dysfunction (Zhou et al., 2018). Collectively, these findings suggest that MDSCs play a protective role during HF. In the case of infection, after pathogen recognition the myeloid cells generate an inflammatory response to clear the pathogen and develop adaptive immunity; then, MDSCs help repair tissue damage by suppressing inflammation. The expansion and/or activation of MDSC subsets that have regulatory activities play a critical role in these processes (Dorhoi et al., 2019). In a study using a mouse model of Trypanosoma cruzi infection, infiltrated Ly6G + MDSCs (gMDSCs) into heart tissue triggered the recruitment of Ly6G − MDSCs (mMDSCs) to the heart through the expression of S100A8/9 proteins and suppressed T-cell proliferation by inducing arginase I and inducible nitric oxide synthase (Cuervo et al., 2011). In our model of aging, mMDSCs in the hearts of old mice are deficient in ability in inhibiting T-cell proliferation. Thus, the inefficiency of mMDSCs function in old mice might contribute to the detrimental impact of chronic inflammation in the hearts of old mice.
Since the CRM phenotype and function are essential for the maintenance and homeostasis of myocardium under a steady-state condition and during tissue-repair processes (Epelman et al., 2014;Ma et al., 2018), we explored the phenotype of CRMs and its function in young and old mice in the context of infection. Previously, Campos Ramos et al. reported a clear decline in absolute numbers of macrophages in the heart due to aging (Ramos et al., 2017); moreover, we found that the numbers of total CRMs, regardless of their specific subset, did not decline with age in the steady-state heart. However, the CRM population was significantly reduced in both young and old mice after LPS treatment. Furthermore, in-depth analysis of CRM population in old and young mice, we found that MerTK + , MCH II high and low populations were significantly reduced in both old and young mice upon LPS infusion, which correlated with cardiac dysfunction (Figures 3 and 4). These observations raises questions as to whether macrophages are functionally able to resolve the excessive inflammation in the aged heart in steady state or after LPS infusion and by what mechanism they help regulate heart homeostasis. In homeostasis, it was suggested that embryonic precursor-derived macrophage subsets are more practical for internalizing debris and for phagocytosis of apoptotic cardiomyocytes (Epelman et al., 2014;Lafuse et al., 2020). Considering that mechanism, Wan et al.
reported that MerTK, a phagocytosis receptor, was upregulated on tissue macrophages during myocardial infarction (Wan et al., 2013).
Conversely, a lack of MerTK expression can lead to accumulation of apoptotic cardiomyocytes, increased neutrophil persistence, and reduced levels of the anti-inflammatory cytokine IL-10 in the myocardium (Gautier et al., 2012). Here, we found that the number of CRMs  (Zizzo & Cohen, 2015). In another study, it was demonstrated that LXR signaling is not only critical for clearing apoptotic cells but is also pivotal for maintaining immune tolerance (Gonzalez et al., 2009). In our current in vitro study, we found that LXR activation but not PPARγ inhibition resulted in increased MerTK expression in mouse BMDMs. Therefore, we examined the in vivo effects of LXR activation, in young and old mice, on expansion of the different CRM subsets as well as MerTK surface expression. We found that in vivo activation did not alter CRM subsets but substantially increased MerTK expression in MHC-II high/low CRMs ( Figure 5). Since reduced phagocytosis and engulfment of apoptotic bodies occurs in macrophages from elderly humans and aged rodents in vitro and in vivo (Aprahamian et al., 2008;Izgut-Uysal et al., 2004;Plowden et al., 2004), our studies suggest that LXR activation can be used to increase MerTK expression. In support of this notion, our data show that LXR pretreatment reversed the LPS-mediated cardiac dysfunction in old and young mice, which correlates with an increased presence of MerTK-expressing CRMs in the hearts of old and young mice ( Figure 6). Thus, this approach to increase MerTK in CRMs and its function of clearance of apoptotic bodies from cardiac tissue might be a novel treatment for elderly people with chronic inflammatory diseases. The idea of using pharmacological LXR activation to prevent the progression of a wide range of inflammatory disorders has been well documented in preclinical animal models (Gonzalez et al., 2009;Fessler, 2018;Han et al., 2018). Additional research is needed so that this novel drug therapy could be used as a safe and effective treatment to prevent cardiac dysfunction due to infectious diseases, such as infection or chronic inflammation in the elderly.
In summary, our studies show that the hearts of old mice in the steady state have an inflammatory environment, with increased numbers of monocytes and neutrophils that makes them vulnerable to injuries caused by secondary factors, such as infections. We found that mMDSCs in old mice lack the ability to regulate immune T cells, which may also contribute to an inflammatory environment.

| Heart tissue digestion and single-cell preparation
Mice were euthanized by CO 2 inhalation. Heart single cells were prepared according to the protocol of Nahrendorf et al. (2007) with modification as described below.
Hearts were perfused with 20 ml of cold PBS, weighed, minced, and then digested using a Multi Tissue Dissociation Kit 2 and gentleMACS™ Dissociator (MiltenyiBiotec) according to the manufacturer's instructions. Briefly, perfused hearts were cut into small pieces (1-2 mm 2 ), placed in gentleMACS C-tubes containing 2.5 ml of enzyme mix, incubated at 37℃ for 15 min without agitation, and then dissociated using the Multi-G program of a gentleMACS Dissociator (Meeson et al., 2013). These steps were then repeated twice. After the program was terminated, the C-tube was detached from the gentleMACS Dissociator, 7.5 ml of cell culture medium with 20% FBS was added, and the cells passed through a MACS Smart Strainer (70 µm) and centrifuged to collect single cells. Red blood cells were removed by resuspending cells in red blood cell lysis buffer and washed with 3ml of cell culture medium with 20% FBS. The single-cell suspensions were resuspended with appropriate buffer or medium, counted, and used for further applications. Detailed gating strategies, antibodies used and data analysis can be found in the Supplemental Information section.

| Statistical analysis and Study approval
The data are presented as mean ± standard error of mean. A Student's two-tailed t test was used to compare two groups. A one-way ANOVA followed by the Bonferroni test was used to compare three or more groups. All statistical analyses were performed using Graph Pad Prism software. p < 0.05 was considered statistically significant. Animal use in this study was approved by the Institutional Biosafety Committee and IACUC of the Ohio State University and adhered to NIH guidelines for the use of experimental animals.

ACK N OWLED G M ENTS
The authors take the opportunity to thank Ohio State University Comprehensive Cancer Center Shared Resources, Flow cytometry core, College of Medicine. We also thank Elizabeth Whetstone and Meera Nair for their help in proofreading. The work was supported by a grant from the American Association of Immunologist Careers in Immunology Award 2019, AHA-17GRNT33420101, and NIH grants AI 146252, AI 146690.

CO N FLI C T O F I NTE R E S T
The authors have declared that no conflict of interests exists.

AUTH O R CO NTR I B UTI O N S
NS and MVSR conceived the ideas and designed the experiments.
NS, QW, NK, LPG, WPL, and MVSR performed most of the experiments, acquired, and analyzed the data. NS, WPL, and MVSR wrote the manuscript. LPG, WPL, and MVSR edited the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The authors do not include any large data sets in this manuscript.