Bone marrow deficiency of TRPC3 channel reduces early lesion burden and necrotic core of advanced plaques in a mouse model of atherosclerosis

Abstract

Aims

Macrophage apoptosis plays a determinant role in progression of atherosclerotic lesions. An important goal in atherosclerosis research is to identify new components of macrophage apoptosis that can eventually be exploited as molecular targets in strategies aimed at manipulating macrophage function in the lesion. In the previous work from our laboratory, we have shown that transient receptor potential canonical 3 (TRPC3) channel is an obligatory component of survival mechanisms in human and murine macrophages and that TRPC3-deficient non-polarized bone marrow-derived macrophages exhibit increased apoptosis, suggesting that in vivo TRPC3 might influence lesion development. In the present work, we used a bone marrow transplantation strategy as a first approach to examine the impact of macrophage deficiency of TRPC3 on early and advanced atherosclerotic lesions of Apoe^−/− mice.

Methods and results

After 3 weeks of high-fat diet, lesions in mice transplanted with bone marrow from Trpc3^−/− donors were smaller and with reduced cellularity than controls. Advanced lesions from these mice exhibited reduced necrotic core, less apoptotic macrophages, and increased collagen content and cap thickness. In vitro, TRPC3-deficient macrophages polarized to the M1 phenotype showed reduced apoptosis, whereas both M1 and M2 macrophages had increased efferocytic capacity.

Conclusions

Bone marrow deficiency of TRPC3 has a dual beneficial effect on lesion progression by reducing cellularity at early stages and necrosis in the advanced plaques. Our findings represent the first evidence for a role of a member of the TRPC family of cation channels in mechanisms associated with atherosclerosis.

1. Introduction

In atherosclerosis, macrophage apoptosis is fundamental in determining plaque progression. At early stages, macrophage apoptosis combined with efficient clearance of apoptotic cells—efferocytosis—helps in maintaining reduced cellularity and slows lesion progression^1,2 In advanced plaques, however, when efferocytosis is impaired, accumulation of apoptotic cells contributes to enlargement of necrotic areas and plaque instability, the preamble of acute coronary syndromes.^2,3 Thus, identifying signalling components associated with those mechanisms is of importance to pinpoint candidate targets for molecular or pharmacological manipulation of macrophage function in the lesion.

The previous work from our laboratory showed that transient receptor potential canonical 3 (TRPC3), a member of the TRPC family of Ca²⁺-permeable channels,⁴ is an obligatory component of survival mechanisms in human and murine macrophages.^5,6 In those studies, we found that TRPC3-deficient, non-polarized bone marrow-derived macrophages (BMDMs) exhibit, in vitro, impaired survival signalling and increased apoptosis, suggesting that macrophage deficiency of TRPC3 would be influential in accumulation of lesional apoptotic cells and lesion development in the setting of atherosclerosis.⁵ To address this, in the present work we used a bone marrow transplantation (BMT) strategy in which Apoe^−/− mice were transplanted with bone marrow from Trpc3^−/−Apoe^−/− or Trpc3^+/+ Apoe^−/− animals and lesions were analysed at early and advanced stages of development. After 3 weeks of high-fat diet mice that received TRPC3-deficient bone marrow had smaller lesions with less macrophage than the control group. Contrarily, advanced lesions from these mice showed a reduction in apoptotic cell accumulation and necrosis with an increase in collagen content and cap thickness. In contrast to our previous in vitro observations on non-polarized macrophages,⁵ TRPC3-deficient macrophages polarized to the M1 phenotype showed reduced apoptosis, whereas both M1 and M2 macrophages exhibited increased efferocytic capacity. Our findings indicate that bone marrow deficiency of TRPC3 has a dual beneficial effect on lesion progression by reducing cellularity at early stages and necrosis in the advanced plaques. This represents the first evidence indicating a role of a member of the TRPC family of channel-forming proteins in mechanisms associated with atherosclerotic lesion formation.

2. Methods

2.1. Bone marrow transplantation (BMT)

Animal studies conform to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health and have been approved by University of Toledo IACUC. Briefly, recipient mice (Apoe^−/− females, 6-week-old) were irradiated (10 Gy, 3 min; ¹³⁷Cs-Gammacell 40 Exactor, Nordion Int. Inc.) and 4 h later injected via tail vein with bone marrow cells (∼5 × 10⁶ cells) from Trpc3^+/+Apoe^−/− or Trpc3^−/−Apoe^−/− mice. Details are provided in Supplementary material online.

2.2. Aortic root sectioning

Aortic root sectioning, immunohistochemistry and lesion analysis were performed as described in Smedlund et al.⁷ and in Supplementary material online.

2.3. Preparation of BMDMs

BMDMs were obtained as described in Tano et al.⁵ (see Supplementary material online). The evaluation of apoptosis and immunoblotting of whole-cell proteins from BMDMs were as we described in Tano et al.⁸

2.4. Statistical analysis

Values are shown as means ± SEM. Comparisons between groups were performed by the Mann–Whitney U test using Prism Graph Pad version 6 for Windows 2007 (San Diego, CA). P-values >0.05 were considered significant.

3. Results

We generated ApoE^−/− mice with bone marrow deficiency of TRPC3 by means of BMT as a first approach to examine the impact of macrophage TRPC3 deficiency on atherosclerotic lesion characteristics. Six-week-old Apoe^−/− female mice received myeloablative irradiation and received bone marrow from Trpc3^−/−Apoe^−/− (Trpc3^−/−Apoe^−/−BM→Apoe^−/−, for simplicity, ‘Trpc3^−/−Apoe^−/−’) or Trpc3^+/+Apoe^−/− (Trpc3^+/+Apoe^−/−BM→Apoe^−/−, for simplicity, ‘Trpc3^+/+Apoe^−/−’) mice. Phenotype conversion was confirmed 4 weeks after BMT by PCR of gDNA from the peripheral blood (see Supplementary material online, Figure S1). At this time, mice were placed on a Western-type diet (D12079B, Research Diets) for 3 or 8 weeks. At the end of each diet period, qRT–PCR of cDNA from BMDMs confirmed the lack of the expression of Trpc3 in macrophages from mice receiving Trpc3^−/−Apoe^−/− bone marrow but not in macrophages from mice transplanted with Trpc3^+/+Apoe^−/− bone marrow (see Supplementary material online, Figure S1).

At sacrifice, body weight, plasma cholesterol, triglycerides, and lipoprotein profiles were examined, parameters which did not show significant differences between groups at either diet period (Supplementary material online, Table I and Figure S2). After 3 weeks of diet morphometric analysis of aortic root lesions showed a reduction in the total lesion area and neutral lipid content in Trpc3^−/−Apoe^−/− mice compared with the Trpc3^+/+Apoe^−/− group (Figure 1). At this time, macrophage content in lesions from Trpc3^−/−Apoe^−/− mice was also reduced compared with Trpc3^+/+Apoe^−/− mice (see Supplementary material online, Figure S3; respective normalized MOMA2 area/lesion area, 0.78 ± 0.08 vs. 1.19 ± 0.15, P = 0.022). The number of total (CD11b⁺) or inflammatory (Ly6C^high) circulating monocytes or adhesion of sorted CD11b⁺Ly6C^high monocytes to endothelial cells in vitro was not different between Trpc3^+/+Apoe^−/− and Trpc3^−/−Apoe^−/− mice (not shown), indicating that monocyte trafficking, availability, and adhesion were unlikely to account for such differences. At this stage, acellular necrotic areas were not observed. Contrary to Trpc3^+/+Apoe^−/− mice, localization of TRPC3 immunostaining to macrophage-rich areas was not observed in lesions from Trpc3^−/−Apoe^−/− animals, indicating a majority of Trpc3^−/− cells in lesions of mice with Trpc3^−/− bone marrow (see Supplementary material online, Figure S1D). As shown previously by others,^3,9 feeding transplanted Apoe^−/− mice with high-fat diet during 8 weeks was sufficient to induce advanced lesions in the proximal aorta. At this time, the total lesion size and the neutral lipid content were not different between groups (Figure 2), neither was macrophage content (see Supplementary material online, Figure S3). However, advanced lesions from Trpc3^−/−Apoe^−/− mice showed a reduction of ∼40% in the necrotic core area (138 273 ± 22 595 µm² compared with 232 300 ± 33 440 µm² in control mice; P = 0.03; Figure 3), corresponding to a 1.7-fold decrease in the per cent necrotic area (18 ± 3 vs. 30 ± 8%, for Trpc3^−/−Apoe^−/− vs. Trpc3^+/+Apoe^−/−, respectively; P = 0.02). In addition, compared with Trpc3^+/+Apoe^−/− mice, advanced plaques in Trpc3^−/−Apoe^−/− animals had a higher collagen content (Figure 3B), larger fibrous cap thickness (∼35%, P = 0.001; Figure 3C; see also Supplementary material online, Figure S4), and higher cap/core ratio (0.25 ± 0.06, compared with 0.09 ± 0.02 in control animals, P = 0.002). Trichrome staining did not show obvious differences in the smooth muscle content between Trpc3^+/+Apoe^−/− and Trpc3^−/−Apoe^−/− sections. Of note, mRNA levels and activity of the matrix metalloproteinase-9 (MMP9) were markedly reduced in Trpc3^−/−Apoe^−/− BMDMs compared with control cells, either in the control medium or in the presence of oxidized-LDL (oxLDL) (see Supplementary material online, Figure S5); levels of the MMP inhibitors TIMP-1, TIMP-4, or those of TGFβ and IL1β, known to modulate fibroblast-dependent collagen synthesis, were not different between Trpc3^+/+ and Trpc3^−/− macrophages (not shown).

Figure 1.

Aortic root sections from Trpc3^+/+Apoe^−/−BM→Apoe^−/− or Trpc3^−/−Apoe^−/−BM→Apoe^−/− mice on a 3-week high-fat diet were stained with haematoxylin–eosin (A), or Oil-Red-O (B) to evaluate lesion area and lipid content, respectively. Quantifications of mean stained areas are shown. Trpc3^+/+Apoe^−/−BM→Apoe^−/− (n = 11–12), Trpc3^−/−Apoe^−/− BM→Apoe^−/− (n = 13–14).

Figure 2.

Aortic root sections from Trpc3^+/+Apoe^−/−BM→Apoe^−/− or Trpc3^−/−Apoe^−/−BM→Apoe^−/− mice on an 8-week high-fat diet were stained with haematoxylin–eosin (A), or Oil-Red-O (B). Quantifications of mean stained areas are shown. Trpc3^+/+Apoe^−/−BM→Apoe^−/− (n = 13), Trpc3^−/−Apoe^−/− BM→Apoe^−/− (n = 10–11).

Figure 3.

Aortic root sections from Trpc3^+/+Apoe^−/−BM→Apoe^−/− or Trpc3^−/−Apoe^−/−BM→Apoe^−/− mice maintained on an 8-week high-fat diet were stained with Masson's trichrome to evaluate collagen content (blue) and necrosis. Representative sections are shown in (A). (B) The mean per cent collagen content relative to the lesion area. (C) Cap thickness, measured from largest necrotic cores (n = 32 for Trpc3^+/+Apoe^−/−; n = 21 for Trpc3^−/−Apoe^−/−) from 10 to 13 mice per group.

Macrophage apoptosis modulates cellularity of early lesions and necrosis of advanced plaques.^1,2,10,11 Using a modified in situ TUNEL technique, we examined if bone marrow deficiency of TRPC3 affected accumulation of apoptotic cells in lesions after either diet period. After 3 weeks of high-fat diet, few to none TUNEL⁺ cells were found, a common observation in early stages,^1,12 in line with the notion that an efficient removal of apoptotic cells exists at this stage.¹ Contrarily, in advanced plaques from both groups of animals TUNEL⁺ cells were clearly detectable (Figure 4A), with TUNEL staining at times observed near or inside areas of necrosis. Notably, lesions from Trpc3^−/−Apoe^−/− mice showed a two-fold reduction in the number of apoptotic cells compared with controls (Figure 4A) and this correlated with a reduced number of apoptotic cells that co-localized with macrophages (Figure 4B).

Figure 4.

(A) Aortic root sections from Trpc3^+/+Apoe^−/−BM→Apoe^−/− or Trpc3^−/−Apoe^−/−BM→Apoe^−/− mice maintained on a high-fat diet for 8 weeks were stained for in situ TUNEL to detect apoptotic cells (arrows). In (B) representative merged images are shown of sections co-stained for macrophage (green, AIA31240), TUNEL, and DAPI. P-values were determined using the Mann–Whitney U test.

In a recent study, we found that Trpc3^−/− non-polarized BMDMs were more susceptible to apoptosis than Trpc3^+/+ cells.⁵ Besides the non-polarized, a number of macrophage phenotypes can be found in lesions among which the M1 and M2 classes are well characterized.^13,14 To examine the impact of TRPC3 deficiency in polarized macrophages, we prepared BMDMs from Trpc3^+/+Apoe^−/− and Trpc3^−/−Apoe^−/− mice and induced them to differentiate in vitro into M1 and M2 types. TRPC3 deficiency did not affect the expression of specific M1 and M2 markers (see Supplementary material online, Figure S6A and B). In addition, the expression of SRA, CD36, or TLR-4 was comparable between Trpc3^+/+ and Trpc3^−/− cells both at mRNA and protein levels (see Supplementary material online, Figure S6C–E). BMDMs were serum starved or treated with oxLDL (50 µg/mL) or the endoplasmic reticulum (ER) stressor thapsigargin (1 µM) for 24 h and apoptosis was evaluated by the in vitro TUNEL assay. Whereas TRPC3 deficiency did not affect apoptosis of M2 macrophages (not shown), apoptosis induced by serum starvation, oxLDL, or thapsigargin was reduced in M1-Trpc3^−/−Apoe^−/− macrophages compared with controls (Figure 5A). In macrophages, the Ca²⁺/calmodulin-dependent protein kinase II (CAMKII) and signal transducer and activator of transcription 1 (STAT1) are chief mediators of ER stress-induced apoptosis.¹⁵ Prompted by this and by our previous observations in non-polarized macrophages indicating that TRPC3-mediated constitutive Ca²⁺ influx modulates CAMKII function,^6,8 we compared the activation status of CAMKII and STAT1 in M1-Trpc3^−/−Apoe^−/− vs. M1-Trpc3^+/+Apoe^−/− macrophages under conditions identical to those used to induce apoptosis. As indicated by the extent of phosphorylation of Thr286 on CAMKII and of Ser727 on STAT1, their activation was markedly impaired in TRPC3-deficient M1 macrophages compared with control cells (see Supplementary material online, Figure S7). Using iNOS as an M1 marker and its co-localization with macrophage staining, significant M1 immunoreactivity was observed in advanced lesions from Trpc3^+/+Apoe^−/− and Trpc3^−/−Apoe^−/− mice, whereas it was either minimal or not present in early lesions from both groups of animals (see Supplementary material online, Figure S8).

Figure 5.

(A) Bone marrow-derived Trpc3^+/+Apoe^−/− or Trpc3^−/−Apoe^−/− macrophages differentiated to the M1 phenotype were incubated (24 h) in serum-free RPMI (RPMI) or RPMI containing oxidized-LDL (oxLDL, 50 µg/mL) or thapsigargin (1 µM), and processed for the TUNEL assay. ‘ns’: not statistically significant. Values are means ± SEM (n = 3). (B) Apoptotic non-polarized (M0) BMDMs from Trpc3^+/+Apoe^−/− mice were labelled with Calcein green and added to Trpc3^+/+Apoe^−/− or Trpc3^−/−Apoe^−/− M1 or M2 macrophages to evaluate efferocytosis. Shown are means ± SEM values (n = 3).

By means of an in vitro efferocytosis assay, we also examined the effect of TRPC3 deficiency on the efferocytic capacity of M1 and M2 BMDMs (Figure 5B). Remarkably, M1-Trpc3^−/−Apoe^−/− and M2-Trpc3^−/−Apoe^−/− phagocytes exhibited higher efferocytic capacity than corresponding controls. mRNA levels of LRP-1 and MERTK, both recognized to influence efferocytosis,^2,9 were similar regardless of the TRPC3 expression status; however, protein levels of MERTK were 60% higher (after background subtraction and normalizing by GAPDH) in TRPC3-deficient macrophages compared with control cells (see Supplementary material online, Figure S6E).

4. Discussion

The present studies were aimed at examining the impact of bone marrow deficiency of TRPC3 on the characteristics of atherosclerotic lesions in the aortic sinus of Apoe^−/− mice. A number of salient findings were made. Bone marrow deficiency of TRPC3 resulted in a beneficial effect on early stages, evidenced by a reduction in size and cellularity of lesions in mice maintained on a 3-week high-fat diet. In advanced lesions, bone marrow deficiency of TRPC3 was associated with reduced necrosis, increased collagen content, and increased cap thickness. Compared with control animals, advanced plaques from mice with Trpc3^−/− bone marrow had a reduced number of apoptotic macrophages.

As reported previously by others,^3,16 our studies also show that the reduced necrosis in advanced plaques can occur in the absence of changes in lipid and macrophage content. This is not surprising considering that necrotic cores are rich in free cholesterol, a poor substrate for Oil-Red-O¹² and that the macrophage antibody used also stains pre-necrotic macrophages, rendering the resultant staining independent of necrosis content. Also, the reduced necrosis in lesions from Trpc3^−/−Apoe^−/− mice was not paralleled by changes in the total lesion area. These findings support the notion put forward by Manning-Tobin et al.¹⁷ that necrosis results from a number of events occurring over the entire timeline of the experimental studies and reflects the integrated effect of macrophage apoptosis, deficient efferocytosis, and bidirectional trafficking of macrophages through the intima, not necessarily paralleling lesion growth.

The pool of lesional macrophages consists of a number of phenotypes which includes non-polarized and the polarized M1 and M2 types.^13,18 Studies in human and murine lesions indicate that as lesion progresses M1 macrophages become dominant.^13,19,20 The in vitro results obtained here with Trpc3^+/+Apoe^−/− and Trpc3^−/−Apoe^−/− BMDMs differentiated to M1 and M2 phenotypes indicate that TRPC3 deficiency has quite opposite effects in polarized macrophages compared with our previous observations on non-polarized cells.⁵ While TRPC3 deficiency increases apoptosis of non-polarized BMDMs,⁵ TRPC3-deficient M1, but not M2 macrophages, exhibited reduced susceptibility to atherorelevant pro-apoptotic stimuli. Activated CAMKII and STAT1 are key mediators of ER stress-induced apoptosis in non-polarized macrophages from murine and human origin.¹⁵ Our findings in TRPC3-deficient M1 macrophages showing a correlation between reduced apoptosis and impaired activation of CAMKII and STAT1 suggest that also in polarized cells CAMKII and STAT1 seem associated with ER stress-induced apoptosis, in line with the proposed universality of these pathways in macrophages.¹⁵ Furthermore, our data suggest that TRPC3 expression and/or constitutive function is required, to some extent, for proper operation of these signalling pathways. In addition, M1 and M2 Trpc3^−/− macrophages showed, in vitro, higher efferocytic capacity compared with Trpc3^+/+ cells. This may be explained, at least in part, by the higher expression of MERTK in TRPC3-deficient macrophages compared with Trpc3^+/+ cells. Indeed, increased expression of MERTK has been shown to be sufficient for augmented efferocytic function.²¹

Whereas the marked in vivo differences somewhat contrast with the less drastic in vitro effects, it is conceivable that the reduction in necrotic areas in Trpc3^−/−Apoe^−/− lesions results from the combined effect of reduced apoptosis and enhanced efferocytosis of TRPC3-deficient macrophages, as any effect on apoptosis and efferocytosis, even if minor, is likely to exert a cumulative impact on plaque necrosis. The in vivo relevance of our in vitro findings is in part supported by the observation that advanced lesions from both groups of animals showed prominent immunoreactivity for M1 macrophages compared with early stages, in line with previous findings on Apoe^−/− mice.¹³ This provides an explanation to the impact of TRPC3 deficiency on accumulation of apoptotic cells in advanced plaques: as the number of M1 macrophages increases along lesion development, the anti-apoptotic effect of TRPC3 deficiency on M1 cells becomes evident. Advanced lesions from Trpc3^−/−Apoe^−/− mice also showed increased collagen content. No obvious differences were found in myofibroblast or smooth muscle content between recipients of Trpc3^+/+ or Trpc3^−/− bone marrow; notably, Trpc3^−/−Apoe^−/− macrophages exhibited a marked reduction in MMP9 expression and in both released and cell associated activity, suggesting that reduced collagen degradation may contribute to the increased fibrosis observed in vivo. The reduced apoptosis and necrosis altogether with the increase in collagen content and cap thickness in advanced plaques of mice with Trpc3-deficient bone marrow suggest that Trpc3 acts as a modulating factor in mechanisms underlying necrotic core growth. Of importance, lymphocytes, neutrophils, and dendritic cells from mouse origin do not express TRPC3,^22–24 supporting the notion that the lesion phenotype in Trpc3^−/−Apoe^−/− mice results from the impact of TRPC3 deficiency on macrophage functions.

At early stages, macrophage apoptosis is coupled to efficient efferocytosis, which explains the common observation that apoptotic cells in early lesions are found in very low numbers or none at all.^1,12 In agreement with this, our in situ TUNEL staining showed that after 3 weeks of diet apoptotic cells were a rare find in both groups of animals. Considering our previous findings that TRPC3 deficiency increases apoptosis of non-polarized macrophages and that a significant proportion of macrophages in early lesions of Apoe^−/− mice are non-polarized cells,¹³ it is probable that at this time point increased apoptotic rates of Trpc3^−/− non-polarized cells contributed to the reduced plaque cellularity and size observed in these mice.

The present studies illustrate the not rare observation that signalling proteins involved in macrophage apoptosis can differentially affect lesion characteristics depending upon lesion stage and underscore the importance of changes in relative abundance of macrophage phenotypes throughout lesion development in determining the actual impact of a particular signalling protein on plaque characteristics through macrophage apoptosis.

A great deal of experimental evidence has accumulated documenting the participation of TRPCs in the pathogenesis of cardiovascular diseases (reviewed by us in Tano et al.²⁵ and Abramowitz and Birnbaumer²⁶). TRPC3, in particular, has been implicated in essential hypertension, cardiac hypertrophy, and endothelial dysfunction, but its potential role in atherosclerosis has remained largely unexplored. As of this writing, there is no evidence for a role of TRPC3 in murine or human atherosclerosis. Despite the existence of several single nucleotide polymorphisms in the human Trpc3 gene, clinically associated human variations have not been reported. Interestingly, in expression profile studies using human genomic microarrays Trpc3 mRNA appears up-regulated in plaques from patients with atherosclerotic lesions in the left anterior descendent coronary artery.²⁷ The present work thus pioneers the field by providing for the first time experimental evidence underscoring a potential role of TRPC3 at both early and advanced stages of lesion development.

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Funding

This study was supported by NIH grant R01HL111877-01 (to G.V.), theIntramural Research Program of the NIH Project Z01-ES-101864 (to L.B.), and University of Toledo College of Medicine.