Characterization of heart macrophages in rhesus macaques as a model to study cardiovascular disease in humans

In addition to muscle cells, normal heart tissue includes cells, such as macrophages, that participate in immune responses and tissue repair, and these cell functions may also contribute to tissue damage or disease over time. Thus, we initially evaluated the distribution of macrophages in relation to total number of cells (based on number of nuclei) in different areas of the heart from 3 adult rhesus macaques of 5.7, 6.7, and 11.8 years of age with histopathology scores of 0–2 (animals IP62, IB90, and EL85, respectively; Table 1 ). Cellularity ranged from 1374 ± 83 to 1782 ± 423/mm 2 among all the areas examined ( Figs. 1A and B ). There were no statistically significant differences in mean numbers of cells per area between regions of the heart. However, cellularity was slightly higher in the atrium than the ventricles and in the left side than right side (data not shown). We then used 3 antibodies specific for CD163 (haemoglobin– haptoglobin scavenger receptor), HAM56 (macrophage marker), and CD206 (mannose receptor) ( Table 2 ) to enumerate macrophages and their subsets in heart tissue of these 3 animals with relatively lower histopathology scores. Macrophages were readily detected within interstitial spaces between myocytes ( Fig. 1C ). The distribution of nuclei, that is, cellularity ( Fig. 1B ) and average percentages of total macrophages detected with each antibody were similarly distributed throughout all areas of the heart ( Fig. 1D ) despite the low or sub-clinical inflammation scored in the left atrium of all 3 animals. The mean percentages of macrophages relative to total numbers of cells showed that the CD163-expressing subset of macrophages was the highest (1.6 ± 1.3–2.8 ± 1.3 with an overall mean for all areas being 2.4 ± 1.1), followed by HAM56-positive (1.3 ± 0.7–2.2 ± 1.1; mean of all areas = 1.9 ± 0.9) and CD206-positive (0.7 ± 0.8–1.5 ± 2.0; mean of all areas = 1.0 ± 1.0) macrophages in all cardiac anatomic locations. There were no significant differences between macrophage phenotype populations based on expression of combinations of CD163, HAM56, and CD206 in different anatomic locations examined. Most of the macrophages positive for HAM56 (i.e., 1.4 of 1.9%) or CD206 (0.8 of 1.0%) were also positive for CD163, while a smaller fraction of double-positive macrophages expressed HAM56 and CD206 (0.3 of 1.9% and 1.0%, respectively; Fig. 1E ). Next, we analyzed the cellularity in left or right ventricles from 21 animals ranging from fetal stage to 24 years of age with histopathology scores of 0–2. As shown in Fig. 2A (left graph), there was a significant correlation between declining numbers of cell nuclei/mm 2 area with increasing age that showed a dramatic decline early in life. We, thus, compared results by age groups ( Fig. 2A right graph), which showed that there was a significantly higher mean number of cells/mm 2 (4313 ± 951) in heart tissue of fetus to 0.5-year-old macaques ( n = 6) compared to that in heart tissues from macaques of 1.9–12 years of age ( n = 8; 1641 ± 164) or 13–24 years of age ( n = 7; 1325 ± 285). Representative images in Fig. 2B showed that the decrease in cellularity was associated with growth in heart myocyte size, as well as a slight increase of the intermyocyte spaces as animals became older. The percentages of macrophage populations were then examined in ventricle and atrium heart tissue sections from macaques of different ages that exhibited histopathology scores of 0–2. Results shown in Fig. 3A (left graph) demonstrated that there was a significantly higher mean percentage of CD163-staining macrophages in the group of animals aged 13–24 years old ( n = 7; 3.8 ± 0.9) compared to that in animals aged 0–0.5 years old ( n = 6; 1.8 ± 0.6). The percentage of CD163-positive macrophages also was higher in the older group compared to the 2- to 12-year-old animals ( n = 5; 2.6 ± 0.6) but this difference did not reach statistical significance. The mean percent of HAM56-positive macrophages in the group of animals at 0–0.5 years of age was 0.9 ± 0.5 and increased to 2.1 ± 0.6 in the 2- to 12-year-old animals, while the percent of CD206-positive macrophages was 0.8 ± 0.3 in the very young group and similarly increased to 1.2 ± 0.6 in the 2- to 12-year-old animals. However, these changes were not statistically significant. Furthermore, the percentages of HAM56-positive (2.0 ± 1.2) and CD206-positive (1.2 ± 0.6) macrophages in the 13- to 24-year-old animals remained similar to that of the 2- to 12-year-old animals but with greater variability. Macrophages in the heart tissues also were quantitated based on tissue section area rather than as a percent of total cell numbers or nuclei ( Fig. 3A , right graph). Interestingly, the mean number of CD163+ macrophages per mm 2 was higher in the youngest group of 0- to 0.5-year-old animals (67.9 ± 23.0) compared to those in heart tissues of the 2- to 12-year-old animals (44.0 ± 13.9) and 13- to 24-year-old animals (52 ± 21.14) despite the increasing mean percentages of macrophages in the older groups. Also, the absolute numbers of macrophages expressing CD163, HAM56, or CD206 per mm 2 were not statistically significantly different between age groups. To better understand the increasing percentage of macrophages in heart tissues of the older group of animals, the phenotype and fluorescence intensity expression for each of the macrophage markers were compared. Among the macrophages in the animals aged 2–12 and 13–24 years old, over 80% of the HAM56-positive cells co-expressed CD163 (i.e., 1.7 of 2.1 = 80.95% and 2.0 of 2.1 = 95.24%, respectively) and over 90% of the CD206-positive macrophages co-expressed CD163 (i.e., 1.1 of 1.2 = 91.67% and 1.2 of 1.2 = 100%, respectively; Fig. 3B ). There also was a higher proportion of CD163 single-positive macrophages in the fetus to 0.5-year-old group of animals compared to the younger and older adult groups of animals. Similar levels of cell fluorescence intensity for each of the 3 macrophage surface biomarkers were observed in heart tissues when comparing between the 2–12 and 13–24 year old groups ( Fig. 3C ). However, the mean fluorescence intensity was significantly lower for CD163-staining macrophages in heart tissue of the fetus to 0.5 year old animals (4.2 ± 2.4) compared to the 2- to 12-year-old animals (23.9 ± 9.3) and 13- to 24-year-old animals (27.3 ± 10.0). The mean fluorescence intensity of HAM56 was significantly higher in the animals aged 2–12 years (31.1 ± 7.8) than the youngest group of animals (10.9 ± 6.3) and was higher but more variable in the 13–24 year old animals (23.6 ± 18.1) that was not statistically significantly different. Mean fluorescence intensity for CD206 also was lower in the youngest group compared to the 2–12 and 13–24 year old animals but these comparisons were not statistically significantly different. Lymphocytes found in heart tissue also may play a role in heart disease and aging. 35 Thus, we examined the distribution of CD3-positive T-cells and CD20-positive B-cells in relation to CD163-positive macrophages in rhesus macaque heart tissue with histopathology scores of 0–2. In heart tissues from animals of the 3 age groups, the mean percentages of CD3-positive and CD20-positive cells were significantly lower than of CD163-positive cells ( Fig. 4B left graph). The average percent of B-cells was below 1% in each age group, but the percent of T-cells increased from 0.1 ± 0.1 in the group of fetus to 0.5 year old animals ( n = 6) to 0.5 ± 0.3 in the 2- to 12-year-old animals ( n = 5) and then to 0.8 ± 0.6 in the older group at 13–24 years of age ( n = 7). There were similar trends in absolute numbers of CD3+ T-cells and CD20+ B-cells that also were lower than numbers of CD163+ macrophages in the 3 age groups ( Fig. 4B right graph). Although the percentages of CD3-positive T-cells were not statistically significantly different between the 3 age groups, there was a statistically significant direct correlation between percentages of CD163-positive macrophages and CD3-positive T-cells ( r = 0.510; P = 0.0362) ( Fig. 4C ) but not between absolute numbers of these 2 cell populations (not shown). Diseased heart tissue often is associated with an accumulation of inflammatory cells, including macrophages, 1 , 4 , 5 so we next examined the macrophage phenotype populations in heart tissue sections of rhesus macaques that were assigned pathology scores of 3–4 ( Table 1 ; Supplementary Fig. 1 ). Heart tissues from 7 rhesus macaques exhibiting severe myocarditis were analyzed and compared with heart tissues from 9 animals of similar ages with histopathology scores of 0–2. Representative images of immunofluorescence-stained tissues showed the presence of more CD163-positive macrophages in the ventricles of heart tissue with higher histopathology scores of 3–4 than in heart tissue with a lower histopathology scores of 0–2 ( Fig. 5A ). A mean of 26.7% (± 14.6) CD163-positive macrophages was detected in ventricles of animals with pathology scores of 3–4, which was significantly higher than the mean of 2.9% (± 0.6) of CD163-positive macrophages detected in heart tissues with lower histopathology scores of 0–2 ( Fig. 5B left graph). Similarly, there was a significantly higher mean absolute number of CD163+ macrophages per mm 2 area in animals with higher histopathology scores (618.5 ± 586.3) compared to those with lower histopathology scores (54.1 ± 21.7), as shown in Fig. 5B (right graph). Most of the CD163-positive macrophages co-expressed HAM56 in heart tissues with histopathology scores of 0–2 whereas CD163 single-positive macrophages predominated in heart tissues of animals with histopathology scores of 3–4 ( Fig. 5C ). Moreover, macrophages from heart tissues with histopathology scores of 3–4 exhibited lower cell fluorescence intensities for CD163 (13.4 ± 11.2) and HAM56 (12.1 ± 12.5) than those for CD163 (24.2 ± 11.1) and HAM56 (27.3 ± 9.0) from heart tissues with histopathology scores of 0–2 ( Fig. 5D ). This suggested that diseased heart tissue is comprised of more CD163 single-positive macrophages compared to a higher presence of macrophages that express both CD163 and HAM56 in relatively healthier heart tissue. Heart muscle and skeletal muscle are controlled by the autonomic and somatic nervous systems, respectively, but both are categorized as striated muscle. 36 So, we compared the cellularity and macrophage composition in tissue sections of heart ventricle ( n = 5) and skeletal muscle (middle rectus femoris , n = 3) from animals aged 2–12 years with histopathology scores of 0–2 that were stained with H&E and DAPI. Representative images are shown in Fig. 6A , and quantification results plotted in Fig. 6B indicated that cellularity, or mean number of nuclei/mm 2 , was significantly lower in skeletal muscle (586 ± 31) than in heart (1573 ± 139). Interestingly, among total macrophages, there were no significant differences between percentages of CD163-positive cells in skeletal muscle (2.7 ± 2.0) versus heart (2.6 ± 0.6) or HAM56-positive macrophages in skeletal muscle (1.7 ± 1.4) versus heart (2.4 ± 0.8) ( Fig. 6C ). In addition, the proportion of CD163-HAM56 double-positive cells was similar in both muscle tissues ( Fig. 6D ). We next examined heart and skeletal muscle for the presence of long-lived macrophages that incorporated and retained dextran. Dextran was inoculated 15–43 days prior to obtaining the samples from 3 macaques aged 1.9, 2.9, and 5.7 years of age (animals KR74, KG16, and IM47, respectively; Table 1 ) with histopathology scores of 0–2. Heart and skeletal muscle tissues were stained with antibodies specific for dextran and the macrophage biomarkers CD163, HAM56, and CD206, as well as with DAPI to identify cell nuclei ( Fig. 7 ). In heart tissue, means of 2.3% (± 1.3) macrophages stained positive for only CD163 and 2.0% (± 1.3) stained positive for both CD163 and dextran indicating that 46.5% of all CD163-positive macrophages in heart were long lived. In skeletal muscle, means of 0.8% (± 0.5) of macrophages stained positive for only CD163 and 2.7% (± 1.1) stained positive for both CD163 and dextran resulting in a higher percentage of 77.1% of CD163-positive macrophages in skeletal muscle that were long lived. Similarly, means of 1.6% (± 1.1) macrophages in heart tissue sections stained for only HAM56 and 1.2% (± 0.9) macrophages stained for both HAM56 and dextran, reflecting that 42.9% of this population of macrophages were long lived. In skeletal muscle, means of 2.0% (± 0.1) macrophages stained with HAM56 only and 2.2% (± 1.7) macrophages were positive for HAM56 and dextran, reflecting a majority or 53.0% of these macrophages that were longer lived. CD206-staining macrophages were not determined in skeletal muscle but in heart muscle, means of 0.3% (± 0.2) macrophages stained for CD206 only and 0.7% (± 0.6) macrophages stained for CD206 and dextran, suggesting that 70% of these CD206-positive macrophages appeared to be longer lived.

Rhesus macaques ( M. mulatta ) of Indian origin used in this study were housed at the Tulane National Primate Research Center, had not been experimentally inoculated with infectious agents, and were SPF for Macacine herpesvirus 1, simian immunodeficiency virus, simian retrovirus, and simian T-lymphotropic virus 1, as well as negative for measles virus and Mycobacterium tuberculosis . All procedures were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals 31 and the Association of the Assessment and Accreditation of Laboratory Animal Care, and were approved by the Institutional Animal Care and Use Committee of Tulane University. Heart and skeletal muscle tissues were examined from a total of 27 rhesus macaques ranging from fetal through 24 years of age, including 10 males, 16 females, and 1 fetus of undetermined sex ( Table 1 ). Of these, 20 animals died from causes unrelated to cardiovascular disease or pathology including placental infarction, endometriosis, arthritis, amyloidosis, chronic enterocolitis, gastritis, gastroenteritis, colitis, or ear infection. The remaining 7 animals had histologic evidence of myocarditis and were selected to compare heart tissues with lower and higher histopathology scores (i.e., not randomly selected). Heart tissues from 3 of the animals with no indications of cardiovascular disease or lesions were examined after being administered dextran to identify long-lived macrophages by immunofluorescence staining. 32 Dextran (Dextran, Amino, 10,000 MW; Thermo Fisher Scientific, Waltham, MA; D1860) was inoculated by intravenous (75–150 mg/kg), aerosol (100–300 μ;g/kg), and intrathecal (1–2.5 mg/kg) routes. Tissues from heart (apex, interventricular septum, left and right atria, papillary muscles, and ventricles) and skeletal muscle (middle section of rectus femoris ) were collected at necropsy, fixed in Z-Fix (buffered zinc formalin fixative; Anatech Ltd., Battle Creek, MI), and embedded in paraffin. Longitudinal tissue sections of 5-μ;m thickness were stained with H&E on a Leica AutoStainer XL (Leica Biosystems, Buffalo Grove, IL). For immunofluorescence staining, 5-μ;m thick tissue sections were deparaffinized with xylene, rehydrated in graded ethanol and deionized water, microwaved for 20 min with 0.1% Tween 20 in 10% high pH solution (Vector Laboratories; Burlingame, CA; cat. # H3301), and transferred to a hot 2% antigen unmasking solution (Vector Laboratories; cat. # H3300) for 30 min. Tissue sections were then incubated for 5 min in 0.05% Sudan Black B, blocked with 10% normal goat serum (Thermo Fisher Scientific; cat. # 16210–064) for 40 min, and incubated for 1 h at room temperature with primary antibody followed by incubation with secondary antibody ( Table 2 ) for 30 min. Antibodies were diluted with 0.2% cold water fish skin gelatin (FSG; Millipore-Sigma, St. Louis, MO; cat. # G-7765) in PBS and slides were washed with PBSFSG-Triton 100 (Millipore-Sigma; cat. # X-100) twice for 10 min after each antibody incubation. Then, nuclei were stained with 0.2 μ;g/mL DAPI dilactate (Thermo Fisher Scientific; cat. # D3571) in PBS for 10 min at room temperature. Slides were washed again and cover-slipped using a fluorescence mounting medium prepared by mixing 2.4 g Mowiol 4–88 (Calbiochem, Darmstadt, Germany, 475904) in 6 g glycerol (Millipore-Sigma; cat. # G6279) followed by addition of 6 mL double-distilled H 2 O. After incubation on a shaker for 4–5 h at room temp, 12 mL of 0.2 M Tris buffer (pH 8.5; Thermo Fisher Scientific; cat. # BP152) was added followed by heating at 50°C for 10 min and addition of 0.45 g 1,4-Diazabicyclo[2.2.2]octane (DABCO; Millipore-Sigma, cat. # D27802 ) to reduce fading. Prior to use, the mounting medium was centrifuged at 5,000 × g for 15 min and supernatant medium was supplied at room temperature to avoid formation of air bubbles. Tissue sections from each animal processed for H&E and immunofluorescent antibody staining were examined by 2 pathologists and scored for lesions as previously described with slight modification 33 and as represented in Supplementary Fig. 1 . Tissue scores were: 0 = normal tissue with no inflammatory infiltrate ( Supplementary Fig. 1A ), 1 = minimal with 1–5 mononuclear cells/HPF (40× objective lens) and/or 1–2 foci/per section ( Supplementary Fig. 1B ), 2 = mild with 6–20 mononuclear cells/HPF and/or 3–5 foci/section ( Supplementary Fig. 1C ), 3 = moderate with more than 20 mononuclear cells/HPF and/or 6–20 foci/section ( Supplementary Fig. 1D ), and 4 = severe inflammation with multifocal to coalescent inflammatory infiltrates ( Supplementary Fig. 1E – H ). Occasionally, the histolopathology scores were adjusted to accommodate lesions of hemorrhage ( Supplementary Fig. 1E ), necrosis ( Supplementary Fig. 1G ), myocardial degeneration ( Supplementary Fig. 1I ), or bacterial abscess ( Supplementary Fig. 1H ). Photomicrographs were taken from 11 random HPFs/tissue section (i.e., 1 HPF = 0.09048 mm 2 so 11 HPFs = 1 mm 2 ) from H&E slides used to complete histopathological evaluations and fluorescence-stained slides to perform macrophage phenotyping and detection of T- and B-cells, using a 40× objective of a Leica DMRE microscope (Leica Microsystems; Wetzlar, Germany) and Nuance FX camera (PerkinElmer; Waltham, MA). Images containing large blood vessels were excluded from the study. The percentage of positive cells for each biomarker was calculated by using the total number of DAPI-stained nucleated cells as the denominator. Mean fluorescence intensity of cells stained with individual fluorochromes was measured and compared using normalized values for each fluorochrome calculated as counts/gain*binning2*exposure time*2bit depth. Images were analyzed with InForm (PerkinElmer) and Fiji software. 34 Mean values of results were assessed by unpaired t test for comparing 2 groups or 1-way ANOVA for comparing more than 2 groups that followed by Kruskal-Wallis post-test for pairwise comparisons. Spearman analysis was used for measuring correlations. P < 0.05 was considered statistically significant. Analyses and graphs were prepared using GraphPad Prism version 8.1.2 for Windows, GraphPad Software, La Jolla, CA ( www.graphpad.com ). Venn diagrams were generated with online software ( https://omics.pnl.gov/software/venn-diagram-plotter ; Pacific Northwest National Laboratory, United States Department of Energy) to illustrate the percent of macrophages expressing 2 biomarkers singly and in combination (i.e., CD163, HAM56, and/or CD206) relative to total heart cell nuclei.

These studies corroborate the presence of at least 2 distinct populations of macrophages in heart tissue of rhesus macaques and demonstrated that macrophages increase as a percent of total heart cells with aging and severity of (histo)pathologic condition. This rhesus macaque model is expected to contribute to continued studies focusing on functional roles of macrophages in heart tissues to address mechanisms of homeostasis, as well as pathogenesis in cardiovascular disease for developing intervention strategies to ameliorate disease.

Immune cells of the heart participate in resistance to pathogens and tissue homeostasis but also play roles in pathogenesis of HIV infection, aging, and other conditions of risk for cardiovascular disease in humans. Rhesus macaques are similar in physiology to humans and, thus, serve as useful experimental models to study human disease, as well as to perform experimental procedures and repeated tissue samplings that may not be readily accomplished in humans. The purpose of this study was to characterize the macrophages in heart tissues of rhesus macaques with relatively normal to severe histopathology scores and ranging from fetal to older age. The rationale to focus on macrophages was that these cells function in innate immunity and homeostasis, but also may contribute to disease pathogenesis via inflammation and fibrosis. 15 Published reports primarily evaluated numbers of macrophages per microscopic field of view 35 , 37 , 38 and here, we also included enumerations of macrophages relative to total cell numbers (cellularity) in heart tissues from animals in different age groups. These results demonstrated that the overall cellularity (i.e., cell nuclei/mm 2 ) was similar throughout different areas of the heart in young adult animals with lower histopathology lesion scores of 0–2. Heart tissue cellularity was significantly higher in animals of fetal to 0.5 years (i.e., 6 months) of age compared to adult and older macaques. The increased cellularity, likely due to the smaller size of myocytes, facilitated the presence of more cells per area of tissue. With increasing age, myocyte size appeared to increase leading to reduce overall cellularity per field of view. Studies in mice indicated that numbers of cardiac myocytes are established by birth, further suggesting that the higher cellularity in the infant rhesus macaques related to smaller size rather than from proliferation of myocytes. 39 Since heart tissue cellularity may affect macrophage analyses, the studies presented here included approaches to normalize the numbers of macrophages relative to the total numbers of cell nuclei per field of view. We first examined relatively healthy heart tissues based on the lower histopathology scores of 0–2 ( Supplementary Fig. 1 ) to determine the basal distribution of macrophages. The distribution of macrophages in different regions of the heart was homogeneous, suggesting that examination of limited heart samples appears to reflect macrophage populations throughout the heart with the exception of specialized structures such as valves. As a result, the subsequent studies performed here relied on tissue samples available from atria and/or ventricles to compare the distribution of macrophage subpopulations from animals of different ages. Overall, the highest percentages of macrophages in heart were positive for CD163, followed by expression of HAM56 and then CD206. This higher percentage of CD163-positive macrophages relative to total heart cells in the animals over 2 years of age was consistent with reports by others. 35 However, we did not observe a decrease of HAM56-expressing macrophages that was reported previously in older animals. 35 Instead, we found that while HAM56 cells were less than 1% of all cells in heart tissue from animals of prenatal to 6 months of age, these cells increased and remained stable in animals over 6 months of age. Cell fluorescence intensities of CD163 and HAM56 in heart tissue were lower in the younger animals and higher in the older animals. Thus, the lower percentage of macrophages observed in the youngest animals could be due to a lower number of positively staining macrophages relative to total cell nuclei or to a lower level of biomarker expression that may have been undetected by the immunofluorescence staining techniques used here. The increased percentage of macrophages in heart tissue of animals over 6 months of age suggested that blood monocytes may traffic to heart tissue where they differentiate into tissue macrophages or that macrophages proliferate or self-renew within the heart. 40 In vivo thymidine analogue labeling using BrdU in macaques has been used to evaluate the kinetics, turnover, and phenotype of short-lived macrophages originating from the bone marrow. 11 , 41 , 42 Unincorporated BrdU has a short half-life in vivo and cells in the process of division incorporate this thymidine analogue which then can be detected by immunostaining. This approach determined that in lung, a high proportion of the interstitial macrophages divided recently and, thus, were considered to be shorter-lived. 11 To our knowledge, however, no studies have demonstrated that local proliferation was responsible for the higher percentage of macrophages in heart during aging, and thymidine analogue staining has not yet been performed to determine if shorter-lived or recently dividing macrophages exist in heart tissue. Results published by Bajpai et al, however, support the presence of recently dividing tissue macrophages based on gene expression data on left ventricular heart tissue specimens from patients with dilated and ischemic cardiomyopathies. 18 The investigators reported that the CCR2-positive rather than CCR-negative macrophages were more similar to blood monocytes and, thus, appeared to represent a recently recruited and shorter-lived population that would be analogous to the shorter-lived BrdU-labeled lung macrophages of our published studies. 11 Conversely, longer-lived macrophages can be identified with preferential uptake and retention of in vivo administered amino dextran. Using this approach in a previous study, alveolar macrophages, as opposed to interstitial macrophages of the lung, were characterized as being longer-lived and, thus, resident tissue macrophage. 11 In the present study, we observed that while cellularity was lower in skeletal muscle than in heart muscle, a predominant population of macrophages in both muscle tissues was longer-lived based on detection of internalized dextran. This was consistent with the reports by Epelman and colleagues who suggested that the origin of most heart macrophages in mice is embryonic, and under homeostasis, these macrophages are maintained through local proliferation without significant monocyte recruitment. 14 In the report by Bajpai et al comparing gene expression in macrophages from human heart failure patients, the CCR2-negative population appeared to represent such tissue-resident macrophages 18 and may resemble or be analogous to the dextran-retaining longer-lived macrophages reported here. In the present study, macrophages also were observed at a higher percentage and number of total immune response cells compared to CD3 T-cells and CD20 B-cells in relatively healthy rhesus macaque heart tissue. While the percent of macrophages was higher in heart tissues of the older groups of animals, the percentages of B-cells remained fairly stable and in the older 13–24 year old monkeys, the percentage of T-cells increased slightly but at greater variability within this group. The lower percentage of T-cells measured in this report was consistent with a previous human study reporting 0–4 T-cells/mm 2 area of normal myocardium. 43 While T-cells increased significantly with aging in another study using macaques that may also have exhibited myocarditis, 35 our studies did suggest some increase with age although this was not statistically significant, possibly because the heart tissues selected for examination exhibited lower or relatively healthy histopathology scores. However, there was a significant direct correlation between percent of CD163-positive macrophages that increased with age, and percent of CD3 T-cells, suggesting a relationship between these cells during aging. Analyses on additional specimens from more animals may be needed to better understand shifts of T-cells in heart tissues of aging rhesus macaques. Since macrophages regulate inflammation and may contribute to cardiovascular disease, we also examined heart tissues with higher histopathology scores for comparison to healthy heart tissues with relatively lower histopathology scores of 0–2 in animals over 0.5 years of age ( Table 1 ). The percent of CD163-positive macrophages was significantly higher in the heart tissues with histopathology scores of 3–4 compared to those with lower scores, but was associated with lower mean fluorescence intensity levels. Under inflammatory conditions or heart tissue damage, monocytes appeared to be recruited to the heart to become tissue macrophages, 14 , 44 which may have accounted for the lower fluorescence intensity during cell differentiation.

Myocarditis is often reported in cases of sudden cardiac death or unexplained cardiomyopathy in humans and rhesus macaques ( Macaca mulatta ). 1 – 3 An accurate characterization of the different levels of severity in myocarditis in relation to disease outcome, however, is still challenging. 4 – 6 Cells of the mononuclear phagocyte system, specifically monocytes and macrophages, participate in tissue regeneration, innate inflammatory and adaptive immune responses, and tissue repair to help re-establish homeostasis. 7 – 10 Heterogeneous populations of macrophages exist in different organs, such as lung 11 – 13 and heart 14 – 18 of mammals, including humans, rhesus macaques, and mice 19 – 21 that are necessary for regulating immune responses, as well as tissue repair. Macrophage population imbalances and cell dysfunction, however, may adversely impact the host through tissue damage including reduced cardiac function. 22 Recent work in mice demonstrated that tissue-resident macrophages originate from progenitor cells within the yolk sac that developed during embryogenesis and “transitory myeloid cells” from hematopoietic stem cells within bone marrow. 14 , 23 – 25 The self-renewal capacity of embryo-derived cardiac macrophages in normal heart declines with age and these macrophages are slowly replaced by monocytes from blood. 17 Phenotypic characteristics of cardiac macrophages at varying ages and health conditions are difficult to study in humans because most samples are obtained postmortem, and there are limited numbers of antemortem biopsies available for research. One such recent report examining left ventricular myocardial specimens from human patients with dilated and ischemic cardiomyopathies demonstrated the presence of discernible macrophage subpopulations based on CCR2 expression. 18 From gene expression profiles, the authors suggested that the CCR2-negative population was from tissue resident self-renewing macrophages, while the CCR2-positive population resembled monocyte-derived tissue macrophages. 15 , 18 , 26 Little information exists about macrophage characteristics in normal or relatively healthy human heart tissue. Rhesus macaques provide a useful model to study the roles and characteristics of macrophages in both normal and diseased heart tissue due to their similar physiology to humans. 27 , 28 Thus, studies using rhesus macaques are expected to translate to humans for developing intervention strategies to improve heart disease conditions. In previous studies from our laboratory on lung tissues in healthy rhesus macaques, we identified unique macrophage populations on the basis of CD163 and CD206 expression among other features, 11 and subsequently demonstrated that increased monocyte turnover in blood reflected a direct relationship to macrophage tissue destruction during SIV infection and progression to terminal AIDS. 29 , 30 The purpose of this current report was to characterize macrophages in heart tissues of rhesus macaques with and without disease. In addition to examining macrophage distribution and surface phenotype, we applied in vivo dextran inoculation to identify longer-lived, presumably resident, heart tissue macrophages. The results further support the presence of distinct macrophage populations in heart and help establish a rhesus macaque model to study the roles of macrophages in cardiovascular diseases that can then be translated to developing treatment strategies in humans.