Macrophage polarization markers in subcutaneous, pericardial, and
epicardial adipose tissue are altered in patients with coronary heart
- 1Department of Cardiology, Center for Clinical Heart Research, Oslo University Hospital Ullevål, Oslo, Norway
- 2Department of Food and Drug, University of Parma, Parma, Italy
- 3Faculty of Medicine, University of Oslo, Oslo, Norway
- 4Department of Cardiothoracic Surgery, Oslo University Hospital, Oslo, Norway
Background: Epicardial and pericardial adipose
tissue (EAT and PAT) surround and protect the heart, with EAT directly
sharing the microcirculation with the myocardium, possibly presenting a
distinct macrophage phenotype that might affect the inflammatory
environment in coronary heart disease (CHD). This study aims to
investigate the expression of genes in different AT compartments driving
the polarization of AT macrophages toward an anti-inflammatory
(L-Galectin 9; CD206) or pro-inflammatory (NOS2) phenotype.
Methods: EAT, PAT, and subcutaneous (SAT)
biopsies were collected from 52 CHD patients undergoing coronary artery
bypass grafting, and from 22 CTRLs undergoing aortic valve replacement.
L-Galectin9 (L-Gal9), CD206, and NOS2 AT gene expression and circulating
levels were analyzed through RT-PCR and ELISA, respectively.
Results: L-Gal9, CD206, and NOS2 gene expression
was similar in all AT compartments in CHD and CTRLs, as were also L-Gal9
and CD206 circulating levels, while NOS2 serum levels were higher in
CHD (p = 0.012 vs. CTRLs). In CTRLs, NOS2 expression was lower in EAT vs. SAT (p = 0.007), while in CHD patients CD206 expression was lower in both SAT and EAT as compared to PAT (p = 0.003, p = 0.006,
respectively), suggestive of a possible macrophage reprogramming toward
a pro-inflammatory phenotype in EAT. In CHD patients, NOS2 expression
in SAT correlated to that in PAT and EAT (p = 0.007, both), CD206 expression correlated positively to L-Gal9 (p < 0.001) only in EAT, and CD206 expression associated with that of macrophage identifying markers in all AT compartments (p < 0.001,
all). In CHD patients, subjects with LDL-C above 1.8 mmol/L showed
significantly higher NOS2 expression in PAT and EAT as compared to
subjects with LDL-C levels below (p < 0.05), possibly
reflecting increased cardiac AT pro-inflammatory activation. In SAT and
PAT, CD206 expression associated with BMI in both CHD and CTRLs (p < 0.05, all), and with L-Gal9 in EAT, however only in CTRLs (p = 0.002).
Conclusion: CHD seems to be accompanied by an
altered cardiac, and especially epicardial AT macrophage polarization.
This may represent an important pathophysiological mechanism and a
promising field of therapy targeting the excessive AT inflammation, in
need of further investigation...
When comparing gene expression levels of the
investigated molecules in the different AT compartments, we did not find
differences in L-Gal 9 expression between SAT, PAT and EAT in either
CTRLs or CHD patients. Several in vitro studies have demonstrated
that L-Gal 9 is involved in the modulation of a variety of biological
processes, including cell aggregation and adhesion, regulation of T cell
pools, as well as the modulation of macrophages polarization (23, 24).
Only a few studies have evaluated L-Gal 9 expression in AT, and
demonstrated that it is predominantly expressed in the stromal vascular
fraction (SVF), composed of various cell types. It is possible that,
overall, L-Gal 9 expression in the various AT compartments is similarly
and tightly controlled. We observed lower CD206 expression in SAT and
EAT compared to PAT in both groups, statistically significant in CHD
patients only, possibly due to the lower number of CTRLs. This is in
line with the different origin and vascularization of PAT (2, 7).
Lower CD206 expression in EAT and SAT might be important because in
concomitance with atherosclerosis progression the number of M2
macrophages in the plaques decreases (32)
in various tissues including AT, with consequent formation of
crown-like structures, a typical hallmark of chronic fat tissue
inflammation that possibly hints CHD (33).
We found NOS2 to be less expressed in EAT with respect to the other
compartments, although statistical significance was reached only
compared to SAT in the control group. Very recently, it was reported
that adipocytes in EAT generated exosomes containing NOS2 (34).
Thus, it is possible that the NOS2 we measured was adipocyte-derived,
rather than of macrophage origin. Accordingly, its expression did not
correlate with any macrophage, T cell or endothelial cell marker. The
lower NOS2 expression in EAT in controls points to a specific, stricter
regulation of this molecule in EAT, which might be lost in CHD (Graphical abstract, point a).
Moreover, the EAT and SAT expression of NOS2 was positively associated
only in the CHD group, in which also L-Gal 9 and CD206 were positively
inter-correlated in EAT, suggesting activation of a compensatory
anti-inflammatory mechanism specifically occurring in EAT in presence of
CHD (Graphical abstract, point b).
In CHD patients, specifically in EAT, L-Gal 9 expression
positively and strongly correlated to CD68 expression, a cellular
marker identifying macrophages (35),
thus the L-Gal 9 findings described above seems to be mainly related to
its macrophage expression. As a consequence, the positive association
observed specifically in EAT between L-Gal 9 and CD206 expression in CHD
patients might confirm our hypothesis of a L-Gal 9 signaling toward an
anti-inflammatory M2 profile in CHD patients, to compensate a local
cardiac AT low-grade inflammation (36) (Graphical abstract, point b).
Furthermore, specifically in PAT and EAT, a positive, albeit milder
association was observed between L-Gal 9 and the expression of T cell
receptor CD3. This is particularly of interest as several studies have
demonstrated that L-Gal 9 acts also by binding and promoting the
activity of death receptor 3, involved in the expansion of regulatory T
cells number and activity (37), known to exert a fundamental role in AT homeostasis (38).
CD206 gene expression was strongly associated with that of CD163 and
CD68 in all AT analyzed. This could be expected as both CD206 and CD163
are commonly expressed by AT resident macrophages, mostly characterized
by an M2-like signature (39, 40), while CD68 is commonly used as a macrophage-specific cell marker (35),
regardless of the cell phenotype. Interestingly, specifically in PAT,
we observed a positive association between CD31 expression, representing
endothelial cells, and CD206. Notably, it has previously been
demonstrated in humans that the vascular density and amount of
endothelial cells in visceral AT is higher as compared to SAT, with a
higher angiogenic and inflammatory profile (41).
Furthermore, a pre-clinical model of infarcted rats showed that the
administration of stem cells isolated from PAT promoted myogenic
differentiation, with consequent efficient cardiac repair (42).
Hence, the positive association that we observed between CD206 and CD31
expression selectively in PAT in CHD patients may be due to a possible