Site-specific impairment of perivascular adipose tissue on advanced atherosclerotic plaques using multimodal nonlinear optical imaging
Edited by Junichi Abe, University of Texas MD Anderson Cancer Center, Houston, TX, and accepted by Editorial Board Member Lia Addadi July 24, 2019 (received for review February 2, 2019)
Significance
Perivascular adipose tissue (PVAT), adipose tissue surrounding blood vessels, has been reported to secrete inflammatory cytokines and adipokines in a paracrine/autocrine manner to regulate the vascular system in healthy and diseased states. However, how PVAT locally and spatially controls vascular functions as atherosclerosis progresses remains unknown. Using multimodal nonlinear optical imaging and an ex vivo assay, we found that PVAT changed differentially from the initial to advanced stages. Interestingly, PVAT at the atherosclerotic plaques showed highly agglomerated lipid droplets and increased fibrosis, suggesting spatial impairment. Our results provide a new concept of PVAT as modulating a direct interaction between the adipose tissue and cardiovascular system.
Abstract
Perivascular adipose tissue (PVAT), as a mechanical support, has been reported to systemically regulate vascular physiology by secreting adipokines and cytokines. How PVAT spatially and locally changes as atherosclerosis progresses is not known, however. We aimed to reveal the molecular changes in PVAT in advanced atherosclerosis based on multimodal nonlinear optical (MNLO) imaging. First, using an atherogenic apolipoprotein E knockout mouse model, we precisely assessed the browning level of thoracic PVAT via a correlative analysis between the size and number of lipid droplets (LDs) of label-free MNLO images. We also biochemically demonstrated the increased level of brown fat markers in the PVAT of atherosclerosis. In the initial stage of atherosclerosis, the PVAT showed a highly activated brown fat feature due to the increased energy expenditure; however, in the advanced stage, only the PVAT in the regions of the atherosclerotic plaques, not that in the nonplaque regions, showed site-specific changes. We found that p-smad2/3 and TGF-β signaling enhanced the increase in collagen to penetrate the PVAT and the agglomeration of LDs only at the sites of atherosclerotic plaques. Moreover, atherosclerotic thoracic PVAT (tPVAT) was an increased inflammatory response. Taken together, our findings show that PVAT changes differentially from the initial stages to advanced stages of atherosclerosis and undergoes spatial impairment focused on atherosclerotic plaques. Our study may provide insight into the local control of PVAT as a therapeutic target.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
Atherosclerosis is a chronic inflammatory disease that starts with the infiltration of lipids and monocytes and leads to the development of macrophages from the vascular wall (1–5). As atherosclerotic plaques develop, the proliferation of smooth muscle cells and the accumulation of some typical types of lipids proceed, and plaque ruptures can occur in significantly advanced stages (6, 7). The foregoing is a well-established model of atherosclerosis, focusing mainly on internal factors in the vessel wall. In contrast, recent models focus on changes in the outer layer of the vessels, especially in perivascular adipose tissue (PVAT), which might involve the internal pathophysiological state of the vessel wall, including such diseases as atherosclerosis, aneurysms, and hypertension (8).
As an adipose tissue, PVAT is distinguished by the typical features of brown adipose tissue (BAT) for efficient energy expenditure and of white adipose tissue (WAT) for energy storage (9, 10). Depending on the anatomic location, thoracic PVAT (tPVAT) shows BAT-like features, including multilocular lipid droplets (LDs) and increased mitochondrial content (11). Functionally, tPVAT highly expresses the BAT marker uncoupling protein-1 (UCP-1) and plays a role in thermogenesis (12).
PVAT surrounding the outside of the adventitia plays a dual role, offering mechanical support and controlling endocrine or paracrine secretions. Although adipocytes predominantly constitute PVAT, other cell types, including capillary endothelial cells, fibroblasts, and macrophages, are present and participate in biological responses. PVAT has been reported to secrete inflammatory cytokines and reactive oxygen species in a paracrine/autocrine manner, which is associated with vascular dysfunction and atherosclerosis (13, 14). Specifically, the peroxisome proliferator-activated receptor-γ (PPARγ) in the vascular smooth muscle cells (VSMCs) and brown adipocytes of specific knockout mice inhibit PVAT development and lead to endothelial dysfunction and atherosclerosis development (15, 16). Recently, the browning of tPVAT has been found to be decreased during aging in a hypertensive rat model and to be associated with vascular dysfunction (17). The browning of PVAT has also been suggested to reduce cardiovascular risk (18). These studies support the possibility that browning control to maintain homeostasis may play an important role in the progression of atherosclerosis.
Transforming growth factor beta (TGF-β) plays roles both in preventing and in enhancing atherosclerosis progression. Although TGF-β prevents atherosclerosis development by inhibiting inflammation, it also enhances VSMC proliferation and extracellular matrix (ECM) expansion to promote vascular fibrosis in atherosclerosis (19, 20). Fibrosis is also one of the causes of adipose tissue dysfunction. In typical adipose tissue such as WAT and BAT, the TGF-β superfamily regulates multiple layers of adipose tissue physiology, including adipogenesis, adipose tissue fibrosis, and inflammation (21, 22). However, the role of TGF-β in PVAT and how it is related to atherosclerosis development are unknown.
Because PVAT, the outermost layer, is generally exfoliated during experimental procedures and is difficult to access via imaging modalities, the molecular mechanisms of PVAT in vascular diseases are not well understood. In this study, we applied multimodal nonlinear optical (MNLO) imaging of tPVAT-intact atherosclerotic aorta, which we previously developed and proposed as a new methodology for atherosclerosis (23, 24). In brief, the MNLO imaging platform includes multiphoton excitation fluorescence (MPEF), second harmonic generation (SHG), and coherent anti-Stokes Raman scattering (CARS) and provides the simultaneous label-free imaging of elastin, collagen, and lipids based on the inherent nature of the molecules (25).
Here we hypothesized that the submicron phenotypic characterization of tPVAT can be accomplished in a nondestructive way using MNLO molecular imaging. The browning level of tPVAT can be precisely quantified in a label-free manner that correlates with the level of atherosclerosis. We further demonstrated the feasibility of the target molecule-specific MNLO imaging and molecular biological experiments via a study of the morphological and functional changes in the tPVAT extending from the alteration of a specific signaling pathway. The present study offers insight into the specific changes in tPVAT in atherosclerosis development and implications for patient treatment using MNLO imaging.
Results
Whole-Mount MNLO Imaging for tPVAT Based on the Size and Number of LDs.
To characterize tPVAT, we used MNLO microscopy combined with CARS, MPEF, and SHG (schematic in SI Appendix, Fig. S1). MNLO microscopy has been previously used for various types of biological and medical research due to its property of providing a signal intensity based on inherent molecular structures (23–25). Here we performed whole-mount ex vivo CARS imaging of tPVAT to mitigate the tissue processing difficulties of tPVAT, which is a loose and atypical tissue at the outermost layer. Since tPVAT is composed mainly of lipids, and fixatives could affect the preservation of its morphological features, we prepared nonfixed specimens in a lumen side up direction (Fig. 1A). The nonfixed sample of C57BL/6 mice fed a normal chow diet (ND-C57BL/6) provided boundary-clear LD images that were more reliable for the LD analysis than those produced using the fixed and cryosectioned processing methods (Fig. 1B; red indicates C-H2 rich lipids); thus, this method was applied to obtain all subsequent imaging results in this study.
Fig. 1.
Next, to examine the feasibility of MNLO imaging for tPVAT characterization, we compared the en face CARS imaging of tPVAT with that of several types of adipose tissues, including interscapular BAT (intBAT) and epididymal WAT (epiWAT), in ND-C57BL/6 mice (Fig. 1C). CARS imaging showed that tPVAT and intBAT had multilocular LDs, while epiWAT had unilocular LDs with enlarged size. These label-free CARS images of LDs in adipose tissues were plotted to precisely quantify the differences (Fig. 1D). epiWAT showed few large-sized LDs, but intBAT showed a large number of small-sized LDs. tPVAT was characterized as being relatively similar to BAT, exhibiting modest-sized LDs with multilocularity (R2 = 0.9652). This quantitative analysis can be used to determine the levels of adipose tissues based on the features of LDs.
To confirm the distribution of LDs on tPVAT, we performed very-wide-view CARS imaging via horizontal tiling at a length of 867.5 μm (SI Appendix, Fig. S2). Two representative regions of interest (ROIs) of 3D-rendered CARS images showed small LDs and large LDs, implying heterogeneous features of tPVAT between WAT and BAT, as reported previously (10, 11). These results show that MNLO imaging-based LD analysis can be a useful methodology to precisely characterize adipose tissues, including tPVAT.
The Features of LDs in Atherosclerotic tPVAT Are Phenotypically Shifted to Those of BAT.
We applied this methodology to investigate how tPVAT features are changed during atherosclerosis progression. First, to exclude the influences of diet and genetic background, the phenotypic differences were imaged by an MNLO system in mice based on the type of diet (normal chow diet [ND] or atherogenic diet [AD]) and the genetic background (C57BL/6 and apolipoprotein E knockout; ApoE−/−). MNLO images were merged to show the inside of the aorta, and atherosclerotic lesions in AD-ApoE−/− mice were observed in the deep intima (Fig. 2 A, Left; green, elastin; red, lipid; magenta, collagen). Fig. 2 A, Right shows the 3D morphologies of LDs in tPVAT, corresponding to the sites of the internal aorta. In a quantitative analysis (Fig. 2B), the morphometric features of LDs in C57BL/6 and ApoE−/− mice were similar based on our initial analysis shown in Fig. 1D. However, after AD feeding, the LDs in C57BL/6 were shifted to a form similar to that of WAT, while the LDs in ApoE−/− mice were shifted to a form similar to that of BAT (R2 = 0.9491) in Fig. 2B. Since this result could be explained by sustained stress in the plasma and vessels disrupting homeostasis, we measured the lipid levels (high-density lipoprotein [HDL], low-density lipoprotein [LDL], and total cholesterol) in plasma acquired from the 4 types of mice (ND- and AD-C57BL/6 and ApoE−/−). As shown in SI Appendix, Fig. S3, the HDL level was lower in AD-ApoE−/− mice than in AD-C57BL/6 mice, while LDL and total cholesterol levels were higher in AD-ApoE−/− mice. It is expected that the increased LDL and total cholesterol levels is one of the causes of tPVAT changes in AD-ApoE−/− mice.
Fig. 2.
MNLO imaging was then applied to a mouse model in the various progression levels using AD-ApoE−/− mice. The morphological changes of tPVATs in AD-ApoE−/− mice (at 8, 12, 16, and 20 wk) were compared with those in ND-ApoE −/− mice (n > 4) (Fig. 2C). Interestingly, at the relatively early stage of 12 wk, the LDs already showed BAT features, as shown in the quantitative graph in Fig. 2D, suggesting that the BAT transition could occur in the early stage of atherosclerosis. The LDs in AD-ApoE−/− mice were smaller than those in ND-ApoE−/− mice from 12 wk of AD feeding; however, after 12 wk, the size of the already small LDs in AD-ApoE−/− was maintained through to 20 wk with no further decrease, but that of the LDs in ND-ApoE−/− increased. Considering this notable shift in LD size at the 12 wk, we also compared the correlation between atherosclerotic lesion formation and tPVAT changes. As shown in SI Appendix, Fig. S4, at 12 wk, atherosclerotic lesions began to form in the deep intima in AD-ApoE−/− mice, whereas ND-ApoE−/− mice did not exhibit lesions. There is a possibility that BAT transition of tPVAT might be associated with the formation of atherosclerotic lesions, and the morphological change occurs at 8 to 12 wk.
After a quantitative assessment of the changes in LDs using MNLO imaging, we used biochemical BAT markers to further analyze whether these phenotypic changes in LDs in atherosclerotic tPVAT were caused by a BAT transition. First, the brown color of the tPVAT in AD-ApoE−/− mice was observed under a bright-field microscope (Fig. 3A). The expression of UCP-1, a representative BAT marker, was significantly increased on Western blot analysis (Fig. 3B), and the mitochondrial content, as measured by MitoTracker staining (Fig. 3C), was also increased compared with that in the tPVAT in AD-C57BL/6 mice. To confirm the BAT transition in the tPVAT over a wider area, we performed en face MNLO imaging of tPVAT with horizontal tiling at a length of 1.2 mm. The results indicate increased UCP-1 expression being dominant in a wider area of tPVAT in AD-ApoE−/− mice relative to that in nonatherosclerotic mice (green in Fig. 3D). These results indicate that the atherosclerotic tPVAT in AD-ApoE−/− mice highly expresses BAT markers.
Fig. 3.
TGF-β Signaling Is Involved in the BAT Transition of tPVAT in Atherosclerosis.
To account for the BAT transition of tPVAT, we performed a microarray analysis for the gene profiling of tPVAT in 4 groups: ND- or AD-fed C57BL/6 and ApoE−/− mice (24 wk). First, we screened 1.5-fold-changed genes (P < 0.05; from 3 independent samples) by diet type in ND-ApoE−/− and AD-ApoE−/mice compared with ND-C57BL/6 and AD-C57BL/6 mice, respectively, to minimize the metabolic effect of the diets and the genetic alterations. Second, we compared the screened ND-ApoE−/− and AD-ApoE−/− genes during the progression of atherosclerosis. The results show that the genes were divided into an AD-specific group (1,808 genes), an ND-specific group (1,241 genes), and a nonregulated group by diet (1,625 genes) in ApoE−/− mice (Fig. 4A). Third, only the AD-specific group (AD-ApoE−/−) was subjected to Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. The results of the pathway categorization using the KEGG pathway database are shown in Fig. 4B. Among the candidates, we focused on the TGF-β signaling pathway, which was up-regulated in AD-ApoE−/− mice, because TGF-β signaling is known to be important for adipogenesis and the immune system (26). It is also known that TGF-β prevents atherosclerosis development by inhibiting inflammation; however, in contrast, TGF-β promotes vascular fibrosis by enhancing VSMC proliferation and ECM expansion (19, 20). Fig. 4C shows the fold changes in the TGF-β signaling pathway due to gene expression in the KEGG pathway database.
Fig. 4.
To investigate the effect of TGF-β on tPVAT, we performed an ex vivo adipocyte differentiation of tPVAT in ApoE−/− mice. Explanted tPVATs on Matrigel were treated with recombinant TGF-β1, BMP7, IL-6, and IL-1β proteins, and Oil red O staining and MNLO imaging were performed 2 wk after tissue culture. As shown in Fig. 5A, ex vivo differentiated adipocytes were reduced in the presence of TGF-β1 or IL-1β, as highlighted by Oil Red O staining. In Fig. 5B, the SHG image in MNLO shows increased collagen synthesis in TGF-β1; however, the CARS image shows decreased differentiated adipocytes in the presence of TGF-β1 or IL-1β compared with the control. Interestingly, the areas of lipids and collagen did not overlap, meaning that the decrease in differentiated adipocytes and the increase in collagen were complementarily controlled (Fig. 5 C and D). The ex vivo adipocyte differentiation was confirmed by the mRNA expression of UCP-1 for a BAT marker and PPARγ for adipocyte differentiation (Fig. 5E). In TGF-β1– and IL-1β–treated tPVATs, the levels of UCP-1 and PPARγ mRNA were decreased, in good agreement with previous reports (27, 28).
Fig. 5.
To determine whether TGF-β signaling is the cause of PVAT changes, ex vivo adipocyte differentiation of explanted tPVAT from ApoE−/− mice was performed by combining TGF-β1 and SB431542, a TGF-β signaling inhibitor (29). The presence of SB431542 with TGF-β1 increased ex vivo differentiated adipocytes by Oil Red O staining and CARS imaging (Fig. 6 A and B in red) and decreased collagen synthesis by SHG imaging (Fig. 6B in magenta) compared with TGF-β1 only. The quantitative graphs in Fig. 6 C and D show that the effect of SB431542 on adipocyte differentiation and collagen synthesis in tPVAT was significant (P < 0.01). Moreover, mRNA expression of UCP-1 and PPARγ was increased in the presence of SB431542 with TGF-β1 compared with control (nontreated) and TGF-β1 only (Fig. 6E). These results suggest that TGF-β1 might mediate PVAT dysfunction through a decrease in adipocyte differentiation and an increase in collagen synthesis. These changes in ex vivo differentiated adipocyte and collagen synthesis by TGF-β1 and SB431542 were confirmed using MNLO imaging of tPVAT derived from LDL receptor-deficient (LDLR−/−) mice, another common murine model for atherosclerosis studies (SI Appendix, Fig. S5). These results imply that TGF-β1 might concomitantly involve inhibition of adipocyte differentiation and activation of fibrosis in tPVAT.
Fig. 6.
The TGF-β Signaling/Fibrosis Axis Is Closely Involved in tPVAT Impairment Depending on the Sites of Atherosclerotic Plaques.
To study whether the TGF-β signaling pathway inhibits adipocyte differentiation and induces fibrosis associated with atherosclerosis progression in a mouse model, we performed MNLO imaging from both cross-sectioned and longitudinally sectioned atherosclerotic aortas with the tPVAT preserved. In the early stage (AD for 8 wk), there was little atherosclerotic plaque formation, with no significant changes in tPVAT (Fig. 7A). As we previously reported, the tPVATs in ROI1 and ROI2 showed UCP-1 expression. Meanwhile, in the advanced stage (AD for 20 wk), depending on the sites of the atherosclerotic plaques, the tPVAT at the atherosclerotic plaque sites (ROI3) showed developed collagen and decreased UCP-1 compared with that at the nonplaque sites (ROI4) (Fig. 7B). This site specificity is quantified in Fig. 7 C and D. More interestingly, at the atherosclerotic plaque sites, p-smad2/3, a signal transducer of the TGF-β signaling pathway, was significantly activated (Fig. 7E), implying that the TGF-β mechanism is a mediated mechanism. These data indicate that fibrosis and adipocyte changes were observed simultaneously, depending on the site of the atherosclerotic plaque formation.
Fig. 7.
The site specificity in the advanced stage was demonstrated in longitudinal sections, such as that shown in Fig. 7F. First, UCP-1 was decreased, especially at the sites of atherosclerotic plaques in the longitudinal MNLO imaging. In one atherosclerotic vessel, as shown in Fig. 8A and SI Appendix, Fig. S6A (AD for 24 wk) and Fig. 8 B and C (AD for 32 wk), the size of the LDs on adjacent plaques was increased compared with that of the LDs at nonplaque sites. Collagen was also increased at the sites of atherosclerotic plaques compared with the nonplaque sites. Interestingly, at the atherosclerotic sites, the agglomeration of LDs and fibrosis showed a highly correlated coincrease (R2 = 0.7631; SI Appendix, Fig. S6B). Therefore, the tPVATs outside the sites of atherosclerotic plaques showed increased lipid agglomeration and collagen synthesis. These results support the hypothesis that the activation of the TGF-β signaling/fibrosis axis is closely involved in tPVAT impairment and results in decreased UCP-1 expression.
Fig. 8.
Inflammation of tPVAT Is Activated in the Late Stage of Atherosclerosis.
The inflammation of tPVAT was demonstrated using immunohistochemical MNLO imaging because such inflammation is closely related to PVAT dysfunction (8). SI Appendix, Fig. S7A shows that CD68+ macrophages were observed in the tPVAT of AD-ApoE−/− mice at 24 wk. In addition, the mRNA levels of proatherosclerotic cytokines, such as IL-6 and IL-1β, were increased in advanced atherosclerosis compared with the initial stage of atherosclerosis (SI Appendix, Fig. S7B). Adiponectin, as an adipogenesis marker, was decreased compared with the initial stage (AD for 8 wk) (SI Appendix, Fig. S7C). These data are consistent with our previous data, showing that tPVAT impairment is induced during the progression of atherosclerosis.
Discussion
We report that the dysfunction of outside PVAT occurred only inside atherosclerotic plaques as atherosclerosis progressed. As the atherosclerosis progressed, both the aggregation of LDs and fibrosis developed site-specifically around the atherosclerotic plaques. The loose tissue feature of PVAT has been an obstacle to studying spatial tissue impairment using conventional fixed and cryosectioning methods. However, MNLO is suitable for imaging how PVAT could interact with the internal vascular system, because it provides label-free and nondestructive imaging based on inherent molecular vibrations. Using this method, we report the possibility that the site-specific PVAT dysfunction is mediated by TGF-β signaling as an interplay response between the inside and outside of the vessel. It is noteworthy that site-specific morphometric changes rather than systemic changes in PVAT in atherosclerosis (11, 18).
The role of PVAT in atherosclerosis has been reported as either preventing or inducing progression (12). The study that showed that PVAT promotes atherosclerosis was focused mainly on inflammatory reactions. Öhman et al. (30) reported that transplanted proinflammatory visceral WAT resulted in atherosclerotic plaque development. In contrast, the antiatherogenic role of PVAT has been studied mainly from the perspective of energy metabolism. Because the energy metabolism is affected by temperature changes, the exposure of mice to cold has been shown to reduce the development of atherosclerosis by activating PVAT thermogenesis (15). That report suggested that the increased metabolic activity of PVAT in cold-exposed mice might reduce the lipid content of vessels, thereby reducing atherosclerosis. In our study, AD-ApoE−/− mice, in which the energy expenditure was high, showed significantly increased BAT in tPVAT. Although the phenotype of tPVAT was heterogeneous between BAT and WAT, the level of BAT in tPVAT was significantly increased in the atherosclerotic model (Figs. 2 and 3). This might be due to homeostatic responses for energy expenditure due to an AD. We also confirmed whether the level of BAT was increased with a mimetic chemical for cold exposure, CL316243 (31, 32). Under treatment with CL316243 in AD-ApoE−/− mice for 8 wk, the morphology of the LDs in tPVAT as imaged by CARS changed to that of BAT, resulting in reduced atherosclerosis compared with the nontreated group (data in SI Appendix, Fig. S8). Therefore, PVAT might play a protective role by altering the energy metabolism during the initial stage of atherosclerosis (AD-ApoE−/− mice, 8 wk).
However, this morphometric change in LDs in tPVAT paints a different picture of the advanced atherosclerotic model. By applying MNLO imaging, which provides a concomitant analysis of lipids, elastin, and collagen, we found that the tPVAT at atherosclerotic plaque sites typically showed increased collagen synthesis and agglomerated LDs but decreased UCP-1 in the advanced stage of atherosclerosis (AD-ApoE−/− mice, 20 wk). Interestingly, other regions without atherosclerotic plaques still maintained high BAT features, even in the same vessel (Fig. 8). In the early stage of arteriosclerosis, browning was promoted to increase energy expenditure, but in the advanced stage of atherosclerosis, tPVAT impairment was spatially localized on the atherosclerotic plaques (Fig. 9). Notably, spatial tPVAT impairment occurs only on atherosclerotic plaques; this analysis was possible due to the combination of longitudinal sample preparation and MNLO imaging.
Fig. 9.
Regarding the differential changes in tPVAT mentioned above, TGF-β is perceived to be a mediator in tPVAT impairment. We found that the synthesis of collagen was enhanced in ex vivo PVAT treated with recombinant TGF-β1 (Figs. 5 and 6), in good agreement with previous reports (20). In addition, in the tPVAT adjacent to plaques, the presence of p-smad2/3, a TGF-β activating factor, was significantly increased (Fig. 7), supporting the mediation of increased fibrosis by TGF-β (26, 33). Although TGF-β–modified animal models are not common due to lethality, in vivo experiments in future studies may be able to provide further TGF-β–mediated mechanistic insights.
In MNLO imaging, molecular structure-based CARS imaging has been applied to lipids containing C-H bonds, thereby demonstrating its many applications in cardiovascular diseases, such as atherosclerosis. In fact, we have reported the classification of 4 types of atherosclerotic lipids using CARS (23). However, it proved difficult to study the interaction between lipids and the ECM with CARS alone, and the integration of SHG and MPEF led to the development of label-free MNLO imaging for collagen and elastin. Using MNLO imaging, we have also reported that atherosclerotic LDs physically affect the environmental ECM (24). In the present study, this experimental platform was expanded for PVAT-preserved atherosclerotic vessels, providing a useful 3D and label-free methodology for PVAT. Although MNLO imaging primarily enables label-free imaging due to its property of detecting molecular interactions, targeted label-based imaging is also possible with suitable MPEF-based fluorescent markers.
In summary, MNLO imaging is used to assess PVAT by concomitantly visualizing lipids and the ECM and the spatial impairment of PVAT outside atherosclerotic plaques. In addition to applications in cardiovascular diseases, this methodology may expand the potential of PVAT as a target for therapeutic purposes.
Materials and Methods
Animal Experiments.
Male 8-wk-old ApoE−/− mice were obtained from The Jackson Laboratory. Prior to the experiments, the mice were acclimatized to a 12-h light/dark cycle at 22 ± 2 °C for 2 wk with unlimited food and water in a specific pathogen-free facility at the Korea Research Institute of Bioscience and Biotechnology. For the atherogenic progression experiments, the mice were fed an ND (Harlan 2018S; Teklad Premier Laboratory Diets) or an AD (102571; Dyets Inc.) for 8, 12, 16, 20, and 24 wk (n = 4 to 5 for each group). For brown fat activation, the mice were fed an AD and injected daily with the β3-AR agonist CL316243 (1 mg/kg i.p.; C5976, Sigma-Aldrich) or vehicle (distilled water) for 8 wk (n = 5–6 for each group). The thoracic aortas, including their tPVAT, intBAT, and epiWAT, were isolated from the anesthetized mice. All mice were euthanized by CO2 asphyxiation. All animal experiments were approved by the Institutional Animal Care and Use Committee of the Korea Research Institute of Bioscience and Biotechnology (KRIBB-AEC-15133 and KRIBB-AEC-18112) and were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals.
Whole-Mount MNLO Imaging.
We previously developed and reported on the MNLO microscopy and imaging method (23, 24). In brief, we used 2 separate laser systems, a picosecond (ps)-pulsed laser system (pico-EMERALD, HighQ Laser; APE) and a femtosecond (fs)-pulsed laser system (Chameleon Vision S; Coherent), to obtain CARS images and multiphoton images, respectively. To obtain CARS images of lipids in tissues, a signal beam was used as a pump beam (ωp) and a Stokes beam (ωs) to set the center wavelengths of the ps-pulsed laser to 817 nm and 1,064 nm, respectively. In addition, for the MPEF and SHG images, we set the wavelength of the fs-pulsed laser to 810 nm. A schematic diagram and the optical beam paths for MNLO microscopy are presented in SI Appendix, Fig. S1.
For en face label-free MNLO imaging, tPVAT samples were longitudinally opened and mounted lumen side down on a glass-bottom chamber and then covered with a coverslip. Z-depth images were obtained to measure depths of 0 to 100 μm. For the tPVAT imaging of sectioned samples, tPVATs were isolated from the mice and immediately frozen at −80 °C. The tPVAT samples were either cross-sectioned or longitudinally cryosectioned in 10-μm-thick slices. Each sectioning method is displayed in the corresponding figures.
Quantification of the LD and Fibrosis Areas.
The LD and fibrosis areas in the tPVATs were selected at random from at least 5 different CARS images and SHG images, respectively, at the same magnification. After removal of the background from each image, the images were removed at high spatial frequencies (blurring the image; down to 40 pixels) and low spatial frequencies (similar to subtracting a blurred image; up to 3 pixels) using the “bandpass filter” feature of ImageJ version 1.51s. The number and average size of the LDs were determined and fixed to set up as 5 to infinity for size (pixel2) and as 0 to 1 for circularity on the “analyze particles” tab. To determine the identity of the fibrosis features, the “analyze particles” tab was set to 0 to infinity for size (pixel2) and to 0 to 1 for circularity. The fibrosis was quantified as a percentage of the particle area versus the tPVAT area.
Ex Vivo Adipocyte Differentiation of Explanted tPVAT.
We adapted a previously reported method with some modifications (34). In brief, tPVATs were isolated from 8-wk-old ApoE−/− mice or 6-wk-old LDLR−/− mice. tPVAT fragments were washed with an EGM-2 medium (Lonza) and cut into 1-mm-thick ring segments. Each ring segment was embedded in a 35-mm confocal dish (SPL Life Sciences) containing 50 μL of Matrigel GFR (BD Biosciences). The Matrigel-embedded ring segments were incubated with EGM-2 medium containing 1 μM rosiglitazone, 0.5 μg/mL insulin, and 1 nM 3,3′,5-triiodo-l-thyronine sodium salt (T3) as a control medium. For the treatment of recombinant proteins, the EGM-2 control medium was added to 2 ng/mL hTGF-β1 (240-B002; R&D Systems), 100 ng/mL hBMP7 (354-BP-010; R&D Systems), 20 ng/mL mIL-6 (PMC0064; Gibco), or mIL-1β (PMC0814; Gibco). For the inhibition of TGF-β1 signaling, the EGM-2 control medium was added to hTGF-β1 and SB431542 (S4317; Sigma-Aldrich). The medium was replaced every 2 d until the experiments were performed on day 14.
Additional experimental conditions are described in detail in SI Appendix.
Acknowledgments
This work was supported by the Development of Platform Technology for Innovative Medical Measurements Program (KRISS-2019-GP2019-0013) from the Korea Research Institute of Standards and Science and by the Nano Material Technology Development Program (2014M3A7B6020163) and Creative Materials Discovery Program (2018M3D1A1058814) from the Ministry of Science and ICT of Korea.
Supporting Information
Appendix (PDF)
- Download
- 1.04 MB
References
1
J. Frostegård, Immunity, atherosclerosis and cardiovascular disease. BMC Med. 11, 117 (2013).
2
G. K. Hansson, A. Hermansson, The immune system in atherosclerosis. Nat. Immunol. 12, 204–212 (2011).
3
G. S. Hotamisligil, Inflammation and metabolic disorders. Nature 444, 860–867 (2006).
4
K. J. Moore, I. Tabas, Macrophages in the pathogenesis of atherosclerosis. Cell 145, 341–355 (2011).
5
B. G. Nordestgaard, Triglyceride-rich lipoproteins and atherosclerotic cardiovascular disease: New insights from epidemiology, genetics, and biology. Circ. Res. 118, 547–563 (2016).
6
P. Libby, Inflammation in atherosclerosis. Nature 420, 868–874 (2002).
7
P. Libby, A. H. Lichtman, G. K. Hansson, Immune effector mechanisms implicated in atherosclerosis: From mice to humans. Immunity 38, 1092–1104 (2013).
8
R. Nosalski, T. J. Guzik, Perivascular adipose tissue inflammation in vascular disease. Br. J. Pharmacol. 174, 3496–3513 (2017).
9
M. Gil-Ortega, B. Somoza, Y. Huang, M. Gollasch, M. S. Fernández-Alfonso, Regional differences in perivascular adipose tissue impacting vascular homeostasis. Trends Endocrinol. Metab. 26, 367–375 (2015).
10
J. Padilla, N. T. Jenkins, V. J. Vieira-Potter, M. H. Laughlin, Divergent phenotype of rat thoracic and abdominal perivascular adipose tissues. Am. J. Physiol. Regul. Integr. Comp. Physiol. 304, R543–R552 (2013).
11
T. P. Fitzgibbons et al., Similarity of mouse perivascular and brown adipose tissues and their resistance to diet-induced inflammation. Am. J. Physiol. Heart Circ. Physiol. 301, H1425–H1437 (2011).
12
N. K. Brown et al., Perivascular adipose tissue in vascular function and disease: A review of current research and animal models. Arterioscler. Thromb. Vasc. Biol. 34, 1621–1630 (2014).
13
A. Omar, T. K. Chatterjee, Y. Tang, D. Y. Hui, N. L. Weintraub, Proinflammatory phenotype of perivascular adipocytes. Arterioscler. Thromb. Vasc. Biol. 34, 1631–1636 (2014).
14
R. M. da Costa et al., Increased mitochondrial ROS generation mediates the loss of the anti-contractile effects of perivascular adipose tissue in high-fat diet obese mice. Br. J. Pharmacol. 174, 3527–3541 (2017).
15
L. Chang et al., Loss of perivascular adipose tissue on peroxisome proliferator-activated receptor-γ deletion in smooth muscle cells impairs intravascular thermoregulation and enhances atherosclerosis. Circulation 126, 1067–1078 (2012).
16
W. Xiong et al., Brown adipocyte-specific PPARγ (peroxisome proliferator-activated receptor γ) deletion impairs perivascular adipose tissue development and enhances atherosclerosis in mice. Arterioscler. Thromb. Vasc. Biol. 38, 1738–1747 (2018).
17
L. R. Kong, Y. P. Zhou, D. R. Chen, C. C. Ruan, P. J. Gao, Decrease of perivascular adipose tissue browning is associated with vascular dysfunction in spontaneous hypertensive rats during aging. Front. Physiol. 9, 400 (2018).
18
P. Aldiss et al., “Browning” the cardiac and peri-vascular adipose tissues to modulate cardiovascular risk. Int. J. Cardiol. 228, 265–274 (2017).
19
D. J. Grainger, Transforming growth factor beta and atherosclerosis: So far, so good for the protective cytokine hypothesis. Arterioscler. Thromb. Vasc. Biol. 24, 399–404 (2004).
20
M. Ruiz-Ortega, J. Rodríguez-Vita, E. Sanchez-Lopez, G. Carvajal, J. Egido, TGF-beta signaling in vascular fibrosis. Cardiovasc. Res. 74, 196–206 (2007).
21
C. Crewe, Y. A. An, P. E. Scherer, The ominous triad of adipose tissue dysfunction: Inflammation, fibrosis, and impaired angiogenesis. J. Clin. Invest. 127, 74–82 (2017).
22
M. J. Lee, Transforming growth factor beta superfamily regulation of adipose tissue biology in obesity. Biochim Biophys Acta Mol Basis Dis 1864, 1160–1171 (2018).
23
S. H. Kim et al., Multiplex coherent anti-Stokes Raman spectroscopy images intact atheromatous lesions and concomitantly identifies distinct chemical profiles of atherosclerotic lipids. Circ. Res. 106, 1332–1341 (2010).
24
E. S. Lee et al., Lipid crystals mechanically stimulate adjacent extracellular matrix in advanced atherosclerotic plaques. Atherosclerosis 237, 769–776 (2014).
25
J. Y. Lee, S. H. Kim, D. W. Moon, E. S. Lee, Three-color multiplex CARS for fast imaging and microspectroscopy in the entire CHn stretching vibrational region. Opt. Express 17, 22281–22295 (2009).
26
H. Yadav, S. G. Rane, TGF-β/Smad3 signaling regulates brown adipocyte induction in white adipose tissue. Front. Endocrinol. (Lausanne) 3, 35 (2012).
27
L. Choy, R. Derynck, Transforming growth factor-beta inhibits adipocyte differentiation by Smad3 interacting with CCAAT/enhancer-binding protein (C/EBP) and repressing C/EBP transactivation function. J. Biol. Chem. 278, 9609–9619 (2003).
28
A. Gagnon, C. Foster, A. Landry, A. Sorisky, The role of interleukin-1β in the anti-adipogenic action of macrophages on human preadipocytes. J. Endocrinol. 217, 197–206 (2013).
29
B. Chignon-Sicard et al., Platelet-rich plasma respectively reduces and promotes adipogenic and myofibroblastic differentiation of human adipose-derived stromal cells via the TGFβ signalling pathway. Sci. Rep. 7, 2954 (2017).
30
M. K. Öhman et al., Perivascular visceral adipose tissue induces atherosclerosis in apolipoprotein E-deficient mice. Atherosclerosis 219, 33–39 (2011).
31
G. Barbatelli et al., The emergence of cold-induced brown adipocytes in mouse white fat depots is determined predominantly by white to brown adipocyte transdifferentiation. Am. J. Physiol. Endocrinol. Metab. 298, E1244–E1253 (2010).
32
J. F. Berbée et al., Brown fat activation reduces hypercholesterolaemia and protects from atherosclerosis development. Nat. Commun. 6, 6356 (2015).
33
T. H. Lan, X. Q. Huang, H. M. Tan, Vascular fibrosis in atherosclerosis. Cardiovasc. Pathol. 22, 401–407 (2013).
34
K. V. Tran, T. Fitzgibbons, S. Y. Min, T. DeSouza, S. Corvera, Distinct adipocyte progenitor cells are associated with regional phenotypes of perivascular aortic fat in mice. Mol. Metab. 9, 199–206 (2018).
Information & Authors
Information
Published in
Classifications
Copyright
© 2019. Published under the PNAS license.
Submission history
Published online: August 19, 2019
Published in issue: September 3, 2019
Keywords
Acknowledgments
This work was supported by the Development of Platform Technology for Innovative Medical Measurements Program (KRISS-2019-GP2019-0013) from the Korea Research Institute of Standards and Science and by the Nano Material Technology Development Program (2014M3A7B6020163) and Creative Materials Discovery Program (2018M3D1A1058814) from the Ministry of Science and ICT of Korea.
Notes
This article is a PNAS Direct Submission. J.A. is a guest editor invited by the Editorial Board.
Authors
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Altmetrics
Citations
Cite this article
Site-specific impairment of perivascular adipose tissue on advanced atherosclerotic plaques using multimodal nonlinear optical imaging, Proc. Natl. Acad. Sci. U.S.A.
116 (36) 17765-17774,
https://doi.org/10.1073/pnas.1902007116
(2019).
Copied!
Copying failed.
Export the article citation data by selecting a format from the list below and clicking Export.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.