Introduction: Tunica media extracellular matrix (ECM) remodeling is well understood to occur in response to elevated blood pressure, unlike the remodeling of other tunicas. We hypothesize that perivascular adipose tissue (PVAT) is responsive to hypertension and remodels as a protective measure. Methods: The adventitia and PVAT of the thoracic aorta were used in measuring ECM genes from 5 pairs of Dahl SS male rats on 8 or 24 weeks of feeding from weaning on a control (10% Kcal fat) or high-fat (HF; 60%) diet. A PCR array of ECM genes was performed with cDNA from adventitia and PVAT after 8 and 24 weeks. A gene regulatory network of the differentially expressed genes (DEGs) (HF 2-fold > con) was created using Cytoscape. Results: After 8 weeks, 29 adventitia but 0 PVAT DEGs were found. By contrast, at 24 weeks, PVAT possessed 47 DEGs while adventitia had 3. Top DEGs at 8 weeks in adventitia were thrombospondin 1 and collagen 8a1. At 24 weeks, thrombospondin 1 was also a top DEG in PVAT. The transcription factor Adarb1 was identified as a regulator of DEGs in 8-week adventitia and 24-week PVAT. Conclusion: These data support that PVAT responds biologically once blood pressure is elevated.

When exposed to an elevated blood pressure (hypertension), the cells of the artery rearrange, express more collagen to strengthen the vessel wall (extracellular matrix [ECM] changes, fibrosis), and increase cell number (hyperplasia) [1]. This can result, depending on the artery, in an increased medial thickening and reduced lumen/wall ratio among other changes. This process, described as remodeling of the artery, is best recognized in two tunicas of the vessel, the tunica media and to a lesser extent the tunica adventitia [2‒7]. Remodeling is also recognized in many species in which the vasculature has been studied, from the mouse to the human. Missing from this is the consideration that the tunica adiposa, equivalent to the perivascular adipose tissue (PVAT), may also remodel in hypertension.

PVAT is best recognized for its ability to produce anticontractile substances [8, 9]. PVAT responds to hypertension/cardiovascular disease with a loss of anticontractile function due to a lower production or effectiveness of substances that reduce contraction. These include nitric oxide, hydrogen sulfide, and adiponectin to name a few [8, 9]. This knowledge is important but is limited in consideration of the overall functions of PVAT. We have a long-term hypothesis that PVAT has the potential to mechanosense and mechanorespond. One could argue that the loss of anticontractile function in the face of hypertension is just one such example of mechanosensing and responding to a change in pressure. Presently, we turn to a different response of PVAT, that of ECM remodeling in the face of hypertension.

The media [1] and, to some extent, the adventitia [2‒7] remodel with/after development of an elevated blood pressure with elevation in ECM proteins that ultimately increase fibrosis of the artery. Here, we test the specific hypothesis that PVAT remodels its ECM to become more fibrotic, as measured by changes in genes recognized to play a role in this process. In this study, we compare the PVAT to the adventitia directly. This is because the adventitia serves as the basement for PVAT, its closest neighbor. Also, the adventitia is collagen rich, while PVAT is comparatively collagen poor in the normal state [10]. The model used is the Dahl SS male rat made hypertensive by being fed a high-fat (HF) diet (HFD) from weaning [11, 12]. The thoracic aorta, which is exposed to the highest blood pressures in the body, is the arterial model. Use of this specific artery provided three benefits. First, the PVAT of this tissue is discrete to this artery; it is not shared with a vein. Second, the thoracic aorta is large enough such that the adventitia can be dissected off from the media. Third and finally, the individual tunicas of one thoracic aortic media and adventitia provide sufficient RNA such that a full plate of ECM genes can be run for each tissue. This allows for a powerful comparison between adventitia and PVAT within one animal.

Animals and Diet

Dahl SS male rats, just weaned (3-4 weeks of age), were purchased from Charles River Laboratories (RRID:RGD_2308886; Kingston, NY, USA). Only males were used in this study given the specific nature of the question at hand and the primary comparison needed to be made between adventitial versus PVAT remodeling. Females do become equivalently hypertensive on the diet next described (12), so we do not have a scientific reason to believe females would be different from males in these outcomes. Once received at MSU, rats were randomized to control diet or HF diet and fed for either 8 (N = 5 pairs) or 24 weeks (N = 5 pairs). A total of twenty (20) rats were used in this study. The experiments that result from animals on diet at these 2 time points were carried out at independent times. Rats are given a unique group number and ear clipped to distinguish one from another. Animals were group housed (2/3 per cage dedicated to one diet) on a 12-h light/dark cycle at 22–25°C room temperature. The animals were housed with Heat Treated Aspen Hardwood Laboratory Bedding (Northeastern Products Corp., Warrensburg, NY, USA). The control diet was 10% kCal fat (lard) and 0.2% Na (Diet D12450J, Research Diets, New Brunswick, NJ, USA) while the HFD was 60% kCal HF (lard) and 0.3% Na (Diet D12492). Diets were well labelled in a freezer in the same room housing experimental rats. Food and water were available ad libitum. MSU animal care staff and study staff worked together to ensure that food was fresh and feeding of a specific diet was not interrupted during the course of experiments. Procedures using animals complied with National Institutes of Health “Guide for the Care and Use of Laboratory Animals” (2011). Procedures used were approved by the MSU Institutional Animal Care and Use Committee (PROTO202000009). Finally, this study was conducted with “Animal Research: Reporting of In Vivo Experiments ”(ARRIVE) guidelines (essential 10 and recommended) in mind.

Sphygmomanometry

Blood pressure measurements of rats were made by sphygmomanometry in conscious, restrained, and warmed (∼37°C for 3–5 min) rats. The animal was placed in a restraint device designed to maintain body temperature and keep the animal calm. The entire measurement process was automated via a computer program that controls the apparatus (CODA High Throughput System, Kent Scientific, Torrington, CT, USA). The tail cuff was inflated 15 times (to a pressure of ∼250 mm Hg with a slow deflation over a period of ∼15 s) with 30-s intervals between inflations. Blood pressure was obtained during each inflation cycle by a volume pressure recording sensor; the final reading was the average of ten inflations. While this system non-invasively calculates systolic, diastolic, and mean arterial pressure, we report only mean arterial pressure. Once measures were done, animals were returned to their home cage. This measure was made at least once during the time on diet and the day before animals were taken for experimentation. We report mean, systolic, and diastolic blood pressures in Table 1.

Table 1.

Mean arterial blood pressure (MABP), systolic blood pressure (SBP), and diastolic blood pressures (DBP) of Dahl SS rats at 8- and 24-week time points

Each group (N = 5)MABP, mm HgSBP, mm HgDBP, mm Hg
Control 8 weeks 119.0±4.0 152.3±5.2 103.1±3.5 
HF 8 weeks 133.4±1.4 163.7±1.4 117.7±2.5 
Control 24 weeks 115.8±1.6 148.4±8.9 98.0±3.0 
HF 24 weeks 150.9±7.8* 191.4±5.7* 124.9±7.9* 
Each group (N = 5)MABP, mm HgSBP, mm HgDBP, mm Hg
Control 8 weeks 119.0±4.0 152.3±5.2 103.1±3.5 
HF 8 weeks 133.4±1.4 163.7±1.4 117.7±2.5 
Control 24 weeks 115.8±1.6 148.4±8.9 98.0±3.0 
HF 24 weeks 150.9±7.8* 191.4±5.7* 124.9±7.9* 

Values are reported as means + SEM for number of animals in parentheses.

p < 0.05 by one-way ANOVA followed by Sidak’s multiple comparisons with * marking that group statistically different.

Dissection

Rats were given pentobarbital as a deep anesthetic (80 mg kg−1, ip). A bilateral pneumothorax was created prior to vessel dissection from the rat. The thoracic aorta was dissected from the aortic arch to the diaphragm. Tissue dissection took place under a stereomicroscope and in a silastic®-coated dish filled with physiological salt solution containing (Mm) NaCl 130; KCl 4.7; KH2PO4 1.18; MgSO4 • 7H2O 1.17; NaHCO3 14.8; dextrose 5.5; CaNa2EDTA 0.03, CaCl2 1.6 (pH 7.2). The whole aorta was threaded onto a wire mounted in a black silastic-filled dish. One small ring (2 mm wide) containing all arterial tunicas was dissected off for histology (below). The wire was then secured to the silastic and all PVAT dissected off from the whole of the mounted aorta; this became the PVAT sample for the RT2 Profiler Arrays. The aorta was then removed from the wire, cut longitudinally, and pinned lumen-side down. The adventitia was scored with a scalpel and gently teased off the media. This became the adventitia sample for the RT2 Profiler Arrays.

Histology

Immediately upon dissection, the 2 mm rings of thoracic aorta were formalin (10%) fixed and paraffin embedded by MSU Investigative Histopathology. Sections (8 micron thick) were cut and tissues stained standardly with Masson trichrome stain. Sections were photographed on a Nikon TE2000 inverted microscope using a Nikon Digital Sight DS-Qil camera and Nikon NIS Elements BR 4.6 software. Settings were used consistently across all images.

RNA Isolation, Quantification, and Purity

Adventitia (9–34 mg from the 5 aortae) or thoracic aortic PVAT (48–93 mg) was homogenized in 2 mL Bead Ruptor tubes with 1.4 mm ceramic bead media (catalog # 19-645-3, Omni International, Kennesaw, GA, USA) and 900 mL Qiazol Lysis Reagent (from Universal Mini Kit) in an Omni Bead Ruptor (Omni International; speed = 5.65 m/s, time = 30 s, cycles = 2). Following disruption, RNA was isolated with the RNeasy Plus Universal Mini Kit (catalog # 73404, Qiagen, Germantown, MD, USA) according to manufacturer’s recommended protocol. RNA concentration and purity were determined using a Nanodrop 2000c (Thermo Scientific, Waltham, MA, USA).

cDNA Preparation

The RT2 First Strand Kit (catalog # 330421, Qiagen, Germantown, MD, USA) was used according to manufacturer’s recommended protocol to reverse transcribe 0.5 μg RNA into cDNA in a SimpliAmp thermal cycler (Applied Biosystems, Waltham, MA, USA).

ECM and Cell Adhesion RT2 Profiler Arrays

A single array plate (RT2 Profiler PCR Array Rat ECM and Adhesion Molecules; GeneGlobe ID: PARN-013Z, Qiagen, Germantown, MD, USA) was used for comparison of ECM genes between each tissue and types of tissue. Adventitia and PVAT samples for each animal were run on individual plates; these plates were run on the same day. In this way, the adventitia and PVAT of the same aorta could be best compared. The PCR conditions were as follows (run on Applied Biosystems QuantStudio 7 Flex system connected to a Dell Optiplex XE2 computer):

  • 1 cycle of 95 °C 10 min

  • 40 cycles of 95 °C 15 s

  • 60 °C 1 min

RT2 SYBR Green Master Mix (catalog # 330623, Qiagen, Germantown, MD, USA) was used for all SuperArray plates. As stated in the RT2 Profiler PCR Array Handbook, the threshold value is determined when CTPPC is 20 ± 2. All plates in this study were set to the same value of 0.16. Data were exported (described below).

Array Analysis

The RT2 Profiler Array uses a Microsoft Excel-based file that allows for import of CT values from the real-time PCR machine. Data were gathered in separate Excel sheets for each of the following four comparison: 8-week adventitia HF versus control; 8-week PVAT HF versus control; 24-week adventitia HF versus control; and 24-week PVAT HF versus control. In each sheet, CT values were compared to the average of five (5) different reference genes: β-actin (Actb), beta-2-microglobulin (B2M), hypoxanthine phosphoribosyltransferase 1 (Hprt1), lactate dehydrogenase A (Ldha), and ribosomal protein, large, P1 (Rplp1). The 2−ΔCT were calculated for the gene of interest where ΔCT is CT value of the gene of interest – CT value of average house-keeping gene. From this, the fold change (FC) of HF/control was calculated. These are the values reported in online supplementary Figure 1 (for all online suppl. material, see https://doi.org/10.1159/000535513). A FC of 2.00 or above was considered biologically meaningful. Statistically, the p values were calculated based on a Student’s t test of the replicate 2−ΔCT values for each gene in the control group and treatment (HF) groups.

Generation of Volcano Plots of Differential Gene Expression at 8 and 24 Weeks on Diet

Volcano plots were used to highlight the upregulated and downregulated genes in each tissue. Plots were generated using the RT2 Profiler Array analysis Excel document. The plots were modified to highlight the gene names of the top differentially expressed genes (DEGs) in each tissue and time point, denoted by black dots.

Construction of Color Plots to Visualize FC

The Cytoscape application [13] was leveraged to visualize the genes that were upregulated with a FC ≥2 with a color-intensity plot. A FC of 2.00 is represented by the lightest color, and as the FC increases the color darkens in turn. These values were used to create a quantitative Venn diagram of DEGs at 24 weeks where the shape fill represents PVAT gene FC, and the shape outline is indicative of adventitia gene FC.

Construction of Gene Regulatory Network

The Cytoscape plugin iRegulon was employed to understand the regulatory interactions among the DEGs within PVAT and adventitia at week 8 and 24 on HF versus control diet through the construction of a gene regulatory network. iRegulon predicts transcription factors (TFs) involved in regulating co-expressed gene sets through the ranking of enriched motifs using position weight matrices [14]. Mus musculus was used in analysis as species for both analyses as there was no rat species offered in the program. The minimum identity between orthologous genes was set to 0.05, while the maximum false discovery rate on motif similarity was defined as 0.001. The gene regulatory network was constructed using a threshold of >4 normalized enrichment score to select TFs for analysis in both tissues at 8 and 24 weeks.

Data Presentation

Weights and blood pressures of rats are reported as the means ± SEM for N = 5 for each group. Histological images shown are those originally taken microscopically or brightened/contrasted comparatively between the control and HF condition. Images were brightened/contrasted on the whole, never in part. One RT2 Profiler Array plate containing a sample of adventitia from a HF-fed rat did not amplify and was not included in subsequent analyses, meaning that the HF adventitia samples represent N = 4 while all other samples represent N = 5. The volcano plot presents the average of the log base 2 FC in HF/control in all five pairs on x-axis with log base 10 of the p value generated within the RT2 Profiler Array Excel file on the y-axis. Genes 2-fold upregulated at both the 8- and 24-week time point in PVAT and adventitia are shown as color plots, bar scaled for FC. Cytoscape generated the gene regulatory plots for TFs involved in regulating gene sets, again using the mean values for the HF/con value for all five pairs.

Statistical Analyses

A one-way ANOVA followed by Sidak’s multiple comparison was used to determine statistical differences between the 8- and 24-week weights and the 8- and 24-week blood pressures. Here, a p value of 0.05 or less was considered significant. Analysis of the RT2 Profiler PCR Array results were complete using an Excel template distributed by Qiagen. A Student’s t test was used to calculate the p values of the replicate 2−ΔCT values for each gene in the Profiler Array for adventitia and PVAT at 8 and 24 weeks.

Blood Pressure and Body Weights of Experimental Rats at 8 and 24 Weeks on Diets

Figure 1 shares the final weights (Fig. 1a) and mean arterial blood pressures (MABPs; Fig. 1b) of animals after 8 and 24 weeks of being on control or HF diet. The body mass of both control and HF diet-fed rats was greater at 24 weeks compared to their 8-week counterpart group. Moreover, the mass of the HF rat was greater than the control at the 24-week but not the 8-week time. Similarly, the MABP of the 24-week HF diet-fed rat was significantly greater than the paired control diet-fed rat. This was not the case at the 8-week time point in which the blood pressures were statistically similar (p = 0.082, Sidak’s multiple comparison test). Table 1 includes the specific values of MABPs as well as systolic and diastolic pressures. Thus, aortic samples came from animals without (8 weeks) and with (24 weeks) elevated mass and MABP.

Fig. 1.

Final body weights and MABP of male Dahl SS rats after 8 or 24 weeks on diet. a Final weights of animals (grams) on control or HF diet, where the HF diet group had elevated body mass compared to control after 24 weeks on diet. b MABP was determined before euthanasia of animals, demonstrating that animals after 8 weeks on either diet were normotensive. Animals on 24 weeks of HF diet had statistically elevated blood pressure compared to control. N = 5 for all groups. HF, high fat; MABP, mean arterial blood pressure. *statistical difference (p < 0.05) between control and high-fat-fed animals.

Fig. 1.

Final body weights and MABP of male Dahl SS rats after 8 or 24 weeks on diet. a Final weights of animals (grams) on control or HF diet, where the HF diet group had elevated body mass compared to control after 24 weeks on diet. b MABP was determined before euthanasia of animals, demonstrating that animals after 8 weeks on either diet were normotensive. Animals on 24 weeks of HF diet had statistically elevated blood pressure compared to control. N = 5 for all groups. HF, high fat; MABP, mean arterial blood pressure. *statistical difference (p < 0.05) between control and high-fat-fed animals.

Close modal

Histology of the Thoracic Aorta from Rats on Control or HF Diet for 8 and 24 Weeks

The small ring of thoracic aorta that gave rise to the adventitia and PVAT samples used in the RT2 Profiler Array were saved for histology. Figure 2 shares the Masson trichrome staining of an exemplar of each of the five pairs. Of specific importance are the two (2) tunicas that were dissected for measure of ECM genes. First, the adventitia stained a brilliant blue, indicative of collagen, in the aorta from animals 8 weeks on diet. Adventitia became more diffuse and less brilliant/dense with age (24 weeks on diet). This staining was more vivid in the adventitia versus the PVAT at both 8 and 24 weeks of diet. Second, PVAT was composed primarily of fascia with densely packed/small brown adipocytes in the control-fed rat. Comparatively, the HF diet increased the size and placement of white adipocytes within brown pads. This was readily apparent in the aorta from the rat on the HF diet for 24 weeks. Moreover, collagen was clearly present in the PVAT of the control and PVAT from the HF rats but less dense than in the adventitia.

Fig. 2.

Remodeling of adventitia and PVAT with progression of HFD-induced hypertension. Representative images (of N = 5) at ×10 objective of male Dahl SS rat thoracic aorta and PVAT stained with Masson trichrome to visualize collagen in blue after 8 (left) or 24 weeks (right) on either control (top) or high-fat (bottom) diet (scale bar, 100 µm). P, perivascular adipose tissue; A, tunica adventitia; M, tunica media; L, lumen.

Fig. 2.

Remodeling of adventitia and PVAT with progression of HFD-induced hypertension. Representative images (of N = 5) at ×10 objective of male Dahl SS rat thoracic aorta and PVAT stained with Masson trichrome to visualize collagen in blue after 8 (left) or 24 weeks (right) on either control (top) or high-fat (bottom) diet (scale bar, 100 µm). P, perivascular adipose tissue; A, tunica adventitia; M, tunica media; L, lumen.

Close modal

Comparison of PVAT to Adventitial Genes at 8 and 24 Weeks of Control or HF Diet

The absolute and relative levels of expression of all genes and their associated p values are shared in online supplementary Table 1. At the 8-week time point, changes in the adventitia far outnumbered those observed in PVAT. Specifically, no genes reached the 2-fold threshold for change in the PVAT while 29 did so in the adventitia of the same vessel. Figure 3a details these genes in volcano plot form with the top genes that changed to the greatest magnitude labelled. Thrombospondin 1 (Thbs1) and collagen 8a1 (Col8a1) were the genes with the greatest magnitude of change in the adventitia at these 8 weeks on diet time point.

Fig. 3.

Visualization of results of RT2 Profiler Array with volcano plots. Volcano plots depicting the differential expression of ECM-related genes in thoracic aortic PVAT and adventitia of male Dahl SS rats fed a HF diet. Black line represents a p value of 0.05, any points above the line having p value <0.05 and below a p value >0.05. Purple lines representing FC or the magnitude of change from control to HF diet animal gene expression. N = 5 for all PVAT groups, N = 5 for control adventitia group, N = 5 for HF adventitia at 8 weeks, and N = 4 for HF adventitia group at 24 weeks.

Fig. 3.

Visualization of results of RT2 Profiler Array with volcano plots. Volcano plots depicting the differential expression of ECM-related genes in thoracic aortic PVAT and adventitia of male Dahl SS rats fed a HF diet. Black line represents a p value of 0.05, any points above the line having p value <0.05 and below a p value >0.05. Purple lines representing FC or the magnitude of change from control to HF diet animal gene expression. N = 5 for all PVAT groups, N = 5 for control adventitia group, N = 5 for HF adventitia at 8 weeks, and N = 4 for HF adventitia group at 24 weeks.

Close modal

By contrast, PVAT had the greatest number of changes at the 24-week time point: 47 genes upregulated 2-fold versus the 3 observed in the adventitia (Fig. 3a). Here, too, the thrombospondin 1 gene Thbs1 was one of the top three most upregulated genes. It was accompanied by ADAM metallopeptidase with thrombospondin type 1 motif, 2 (Adamts2) and platelet/endothelial cell adhesion molecule 1 (PECAM1). Figure 4 depicts the relative magnitude of changes of gene transcription at the 8 (Fig. 4a)-, 24 (Fig. 4b)-week time point. Only the adventitial results are represented in Figure 4a because of the lack of genes that changed over 2-fold in PVAT at the 8-week time point. Online supplementary Figure 1 depicts the significant staining for thrombospondin 1 protein in the adventitia and PVAT at the 8-week and 24-week time points.

Fig. 4.

Visualization of results of RT2 Profiler Array with color plots. FC for each DEG is visualized using Cytoscape through the utilization of color to represent FC. a After 8 weeks on HFD, there were 29 genes with FC >2 in the adventitia and no DEGs in PVAT, where shape fill color is representative of FC. b Quantitative Venn diagram depicting DEGs of ECM-associated genes from control to HF diet after 24-week HFD in thoracic PVAT (47 genes) and adventitia (3 genes) tissues. FC for PVAT is represented as shape fill color, while adventitia FC is represented as border color. N = 5 for all PVAT groups, N = 5 for control adventitia group, N = 5 for HF adventitia at 8 weeks, and N = 4 for HF adventitia group at 24 weeks.

Fig. 4.

Visualization of results of RT2 Profiler Array with color plots. FC for each DEG is visualized using Cytoscape through the utilization of color to represent FC. a After 8 weeks on HFD, there were 29 genes with FC >2 in the adventitia and no DEGs in PVAT, where shape fill color is representative of FC. b Quantitative Venn diagram depicting DEGs of ECM-associated genes from control to HF diet after 24-week HFD in thoracic PVAT (47 genes) and adventitia (3 genes) tissues. FC for PVAT is represented as shape fill color, while adventitia FC is represented as border color. N = 5 for all PVAT groups, N = 5 for control adventitia group, N = 5 for HF adventitia at 8 weeks, and N = 4 for HF adventitia group at 24 weeks.

Close modal

Transcriptional Drivers of Adventitial and PVAT Gene Changes

The Cystoscape plugin iRegulon allows for tracking of gene changes to curated relationships with TFs. Figure 5 shares these results. For 22 of the genes 2-fold upregulated in the adventitia from the 8-week HF-fed Dahl SS rat, the TFs Adarb1, Fos, Gm10323, Smad3, E2f1, Bcl11a, Rreb1, Jazf1, and MAPK1 were identified. For thirty five (35) of the genes 2-fold upregulated in the PVAT from the 24-week HF-fed Dahl SS rat, Adarb1 was again identified as a potential TF as well as Stat1, Bcl6b, Tsnax, Pax8, and Jun.

Fig. 5.

Gene regulatory network may identify TFs responsible for remodeling. The Cytoscape plugin iRegulon was used to predict regulatory TFs and construct a gene regulatory network. TFs were included in the analysis with NES >4. TFs are identified by light blue shape fill and a shape outline color matching arrow color. Arrows are directed from TFs to their respective predicted target. a The gene regulatory network of the DEGs of the adventitia at 8 weeks predicted 8 TFs regulating 22 out of 29 DEGs. b In the DEGs of PVAT at 24 weeks, 35 genes out of 47 were predicted to be regulated by 6 TFs. The TF Adarb1 was identified in both gene regulatory networks. N = 5 for all PVAT groups, N = 5 for control adventitia group, N = 5 for HF adventitia at 8 weeks, and N = 4 for HF adventitia group at 24 weeks. DEG, differentially expressed gene; NES, normalized enrichment score; TF, transcription factor.

Fig. 5.

Gene regulatory network may identify TFs responsible for remodeling. The Cytoscape plugin iRegulon was used to predict regulatory TFs and construct a gene regulatory network. TFs were included in the analysis with NES >4. TFs are identified by light blue shape fill and a shape outline color matching arrow color. Arrows are directed from TFs to their respective predicted target. a The gene regulatory network of the DEGs of the adventitia at 8 weeks predicted 8 TFs regulating 22 out of 29 DEGs. b In the DEGs of PVAT at 24 weeks, 35 genes out of 47 were predicted to be regulated by 6 TFs. The TF Adarb1 was identified in both gene regulatory networks. N = 5 for all PVAT groups, N = 5 for control adventitia group, N = 5 for HF adventitia at 8 weeks, and N = 4 for HF adventitia group at 24 weeks. DEG, differentially expressed gene; NES, normalized enrichment score; TF, transcription factor.

Close modal

This study was designed to test the hypothesis that PVAT is responsive to hypertension and likely remodels to assist the blood vessel in protection against higher pressures. We investigated both the PVAT and tunica adventitia of the thoracic aorta. This was done to compare directly how neighboring tissues respond to HFD-induced hypertension through changes in expression of ECM-related genes, as well as compare PVAT to a tissue better understood in its remodeling in response to hypertension.

Tissue Remodeling Occurs before and after Hypertension

Our data support that ECM remodeling occurs in the thoracic aortic adventitia of the Dahl SS male rat prior to the onset of hypertension and in the PVAT after hypertension is established. This general finding is important because it supports the idea that PVAT may be significantly responsive to elevation of blood pressures as opposed to a change in diet. Similarly, the adventitia may be an exquisitely sensitive barometer of local environmental change while PVAT might restrain/retard overall vascular remodeling.

Collagen Genes Are Upregulated in Both Adventitia and PVAT but at Different Times

Collagen is a protein well recognized to be involved in tissue remodeling. The genes for several collagen isoforms changed notably. Specifically, the gene for collagen isoform 4a2, important to the basement membrane of the cell, was upregulated in the adventitia at 8 weeks. Col2a1, a fibrillar collagen [15], was upregulated with the HF diet in both tissues but at different times. It was upregulated at 8 weeks of diet in the adventitia but at 24 weeks in PVAT. Finally, Col8a1, the gene for the non-fibrillar short-chain collagen, was the most upregulated gene in the adventitia at 8 weeks and upregulated in the PVAT at 24 weeks [15, 16]. This finding is consistent with the upregulation of type 8 (VIII) collagen in diseases or injury involving vascular remodeling [17, 18]. Upregulation of Col8a1 during the early remodeling response in the adventitia highlights the sensitivity of this vascular layer. The adventitia may preemptively adapt its ECM to meet changes in diet or before the onset of hypertension. The finding that the adventitia did not exhibit significant further changes at the 24 weeks of diet (comparing HF to control) while PVAT had marked changes to gene expression is consistent with the idea that remodeling may be complete in the adventitia at this time.

Thrombospondin 1 Is Upregulated in Adventitia and PVAT

The gene Thbs1, which encodes for the matricellular ECM protein thrombospondin 1 (TSP1) [19], was one of the most upregulated genes in the adventitia at 8 weeks and PVAT at 24 weeks on HF diet. Indeed, TSP1 staining was significantly higher in the adventitia and PVAT versus the media regardless of diet or time point (online suppl. Fig. 1). Immunohistochemical staining, considered a qualitative measure, was carried out not to determine whether TSP1 staining became higher in magnitude in PVAT at 24 weeks after the HF versus the control diet but rather to provide evidence that the change in Thbs1 gene expression was relevant because these tissues ultimately make TSP1.

TSP1 is a regulator of diverse cellular processes. These include ECM organization, matrix metalloproteinase activity, cell adhesion and migration, and apoptosis to name a few [20]. TSP1 also regulates the cellular functions implicated in the ECM homeostasis that are disturbed in remodeling. More specifically, TSP1 regulates the TGF-β pathway by binding and activating latent TGF-β, where activated TGF-β increased myofibroblast differentiation, recruited inflammatory cells, stimulated new matrix deposition, and promoted angiogenesis [21‒26]. In addition, fibroblast homeostasis is regulated through the C-terminal domain of TSP1 binding to collagen 1 [22]. The upregulation of TSP-1 in the adventitia and PVAT supports that both are remodeling their ECM at different times but through at least one similar pathway. Interference with this pathway would thus potentially reduce remodeling of both layers.

Identifying Potential Drivers of Gene Changes in Adventitia and PVAT during Remodeling

Identification of TFs and their respective gene network gives mechanistic insight into what drives disease progression. Using the DEGs (>2FC) of the adventitia and PVAT, a gene regulatory network was computed to identify potential upstream regulators responsible for the observed remodeling in the present study. Adarb1 (adenosine deaminase RNA-specific B1) was the only TF predicted to influence gene expression in both the adventitia at 8 weeks (6 target genes) and PVAT at 24 weeks (12 target genes). Interestingly, the genes Ncam2, Ctnna2, Syt1, Hapln1, and Col8a1 were identified as targets for Adarb1 in both tissues but at different times. Adarb1 was not, however, connected to Thbs1 in either the adventitia or PVAT. Nonetheless, Adarb1 may be responsible for mediating the ECM remodeling responses of both tissue types in this model, potentially identifying a target to ameliorate the observed changes.

PVAT Is Biologically Responsive to Hypertension

The present study supports that the thoracic aortic PVAT, a unique fat depot surrounding the vasculature, is sensitive to changes in blood pressure and responds through differential expression of ECM-related genes and in morphology. This finding is consistent with our hypothesis that PVAT remodels its ECM in response to an elevated (sustained) pressure. Here, remodeling may serve as an adaptive response during hypertension to protect the vasculature it surrounds. The impact of this remodeling would be reflected by increased arterial stiffness. PVAT, in a healthy aorta, reduces the overall stiffness of the aorta [27]. PVAT itself has a measurably lower stiffness than the artery it surrounds. Collagen content of PVAT was linearly and positively correlated with the low-stress stiffness of the tissue [27]. PVAT might restrain the overall remodeling of the vessel, though unavoidably remodeling itself in the face of persistently elevated blood pressure. To this point, the PVAT of the isolated thoracic aorta of the HF versus con diet-fed male Dahl SS rats lost the ability to assist in arterial stress relaxation and displayed a modest increase in PVAT fibrosis [10].

Limitations

We recognize limitations to the present study. By lieu of the experiments done, we have limited consideration of adventitial/PVAT remodeling to genes of the ECM. First, only male Dahl SS rats were used in the current study. In future studies, examining the response of female rats to HF diet-induced hypertension in the present context is of interest, though females become similarly hypertensive to males on the HF diet (12). Second, study of the media was omitted in this study. Classic measures were made in this vascular layer in response to HF diet, and modest medial thickening was observed (10). Another limitation is that samples from HF or control animals were only collected at week 8 (before hypertension onset) and 24 (after hypertension onset) for analysis. Other time points may lend insight into the progression of the ECM remodeling.

The RT2 Profiler Array Kit that was used to determine gene expression has the innate limitation of being a finite, curated set of 84 genes related to the ECM. Those genes, however, are well curated and meaningful to the hypothesis at hand. We acknowledge the lack of statistical significance in the DEGs from control to HF diet animals in both tissue types and time points. Despite having a relatively low number of genes reaching statistical significance, the magnitude of genes that had FC >2 at both time points may support that there are overall cumulative changes in the gene expression that could harmoniously/collaboratively contribute to ECM remodeling in the face of elevated blood pressure. A final point relative to these analyses is that some of the DEGs had Ct values “undetermined.” When the amplification data were input into the RT2 Profiler Array analysis template distributed by Qiagen, the genes with Ct values “undetermined” had an arbitrary Avg Ct value of 35 assigned by the data analysis spreadsheet. This is why some of the values for different genes are equivalent.

While we discuss PVAT and adventitia as independent tunicas, growing evidence strengthens the existence of communication between PVAT and adventitia. For example, the PVAT-derived secretory protein complement 3 induced adventitial fibroblast migration and differentiation to contribute to the adventitial thickening and remodeling in the deoxycorticosterone acetate-salt hypertensive rat model [28]. Similarly, adventitial remodeling was promoted by dysfunctional PVAT in the obese mini pig model through the nod-like receptor protein 3 (NLRP3)/interleukin-1 pathway [29]. Communication may occur/mediators may be passed between adventitia and PVAT during onset of hypertension that could mediate the changes in ECM gene expression that were observed in the current study. Tunica cross talk is not something we could measure in this specific study. Finally, the PVAT studied here is recognized as brown fat. White fat that is PVAT, as is found around mesenteric resistance vessels, may not remodel in the same manner [30].

PVAT of the thoracic aorta remodeled its ECM as a response to elevated blood pressure but not directly to diet. PVAT did not demonstrate biologically relevant changes in ECM genes until an elevated blood pressure was established. This contrasted with the remodeling of the adventitia evident after 8 weeks on the HF diet without an elevation in pressure. The gene Thbs1 was identified as elevated in the HF versus control diet in both the adventitia and PVAT though at different time points. Collectively, these findings strengthen the foundation that PVAT is a functionally important layer of the blood vessel in sensing and responding to mechanical challenges.

Graphical abstract created using Biorender.com.

Procedures used in this study were approved by the MSU IACUC (PROTO202000009). Animal procedures complied with National Institutes of Health Guide for the Care and Use of Laboratory Animals (2011). Finally, this study was also conducted with ARRIVE guidelines (essential 10 and recommended) in mind.

The authors have conflicts as noted below:

Caitlin Wilson and Leah Terrian: MSU graduate student. Janice M. Thompson, Emma D. Flood, and Lisa Sather: MSU Research Associate. Adam D. Lauver, Gregory D. Fink, Sudin Bhattacharya, and Andres Contreras: MSU Faculty. Stephanie W. Watts: MSU Faculty; funded by NIH P01HL152951; Keystone Scientific Advisory Board Member.

This work was funded by the National Heart, Lung, and Blood Institute through P01 HL152951.

Caitlin Wilson performed dissection, RNA isolation, RNA SuperArray experiments, and Cytoscape analysis for 24-week diet-fed Dahl SS rats, contributed to the writing and revision of manuscript, and read final manuscript. Janice M. Thompson performed dissection, RNA isolation, and RNA SuperArray experiments for 8-week diet-fed Dahl S rats, contributed to the writing and revision of the manuscript, and read final manuscript. Leah Terrian performed Cytoscape analysis for 8-week diet-fed Dahl SS rats and helped revise manuscript and read final manuscript. Adam Lauver ordered Dahl SS rats and kept on appropriate diet for the 8- and 24-week groups and helped revise manuscript and read final manuscript. Emma D. Flood performed image analysis of the Trichrome-stained thoracic aorta for the 8- and 24-week diet-fed rats that were used in the RNA SuperArray experiments and helped revise manuscript and read final manuscript. Gregory D. Fink assisted in care of Dahl SS rats throughout diet, provided statistical advice, and helped revise manuscript and read final manuscript. Lisa Sather performed tail-cuff sphygmomanometry on all Dahl SS rats and helped revise manuscript and read final manuscript. Sudin Bhattacharya assisted Leah Terrian in Cytoscape analysis and helped revise manuscript and read final manuscript. Andres Contreras aided SWW in determining staining protocols for tissues, provided feedback on study and on analysis, and helped revise manuscript and read final manuscript. Stephanie W. Watts created original idea, organized experiments and experimental outcomes, performed and analyzed immunohistochemical results for TSP-1 staining, wrote initial draft of manuscript, helped revise manuscript and read final manuscript, and procured funding.

Additional Information

Caitlin Wilson, Janice M. Thompson, and Leah Terrian contributed equally as first authors to this work.

The majority of data generated during this study is included in this article. The original RT2 Profiler Array spreadsheets will be made available in the repository Figshare at https://doi.org/10.6084/m9.figshare.23899818.v1. Further inquiries upon request to Stephanie Watts (wattss@msu.edu).

1.
Humphrey
JD
.
Mechanisms of vascular remodeling in Hypertension
.
Am J Hypertens
.
2021
;
34
(
5
):
432
41
.
2.
Majesky
MW
,
Weiser-Evans
MCM
.
The adventitia in arterial development, remodeling, and hypertension
.
Biochem Pharmacol
.
2022
;
205
:
115259
.
3.
Majesky
MW
,
Dong
XR
,
Hoglund
V
,
Daum
G
,
Mahoney
WM
.
The adventitia: a progenitor cell niche for the vessel wall
.
Cells Tissues Organs
.
2012
195
1–2
73
81
.
4.
Wang
A
,
Cao
S
,
Stowe
JC
,
Valdez-Jasso
D
,
Caligiuri
G
,
Nicoletti
A
.
Substrate stiffness and stretch regulate profibrotic mechanosignaling in pulmonary arterial adventitial fibroblasts
.
Cells
.
2021
;
10
(
5
):
1000
.
5.
McGrath
JC
,
Deighan
C
,
Briones
AM
,
Shafaroudi
MM
,
McBride
M
,
Adler
J
.
New aspects of vascular remodelling: the involvement of all vascular cell types
.
Exp Physiol
.
2005
;
90
(
4
):
469
75
.
6.
Siow
RCM
,
Churchman
AT
.
Adventitial growth factor signalling and vascular remodelling: potential of perivascular gene transfer from the outside-in
.
Cardiovasc Res
.
2007
;
75
(
4
):
659
68
.
7.
Coen
M
,
Gabbiani
G
,
Bochaton-Piallat
M-L
.
Myofibroblast-mediated adventitial remodeling: an underestimated player in arterial pathology
.
Arterioscler Thomb Vasc Biol
.
2011
;
31
(
11
):
2391
6
.
8.
Xia
N
,
Li
H
.
The role of perivascular adipose tissue in obesity-induced vascular dysfunction
.
Br J Pharmacol
.
2017
;
174
(
20
):
3425
42
.
9.
Gollasch
M
.
Adipose-vascular coupling and potential therapeutics
.
Annu Rev Pharmacol Toxicol
.
2017
57
Jan 6
417
36
.
10.
Watts
SW
,
Darios
ES
,
Contreras
GA
,
Garver
H
,
Fink
GD
.
Male and female high-fat diet-fed Dahl SS rats are largely protected from vascular dysfunctions: PVAT contributions reveal sex differences
.
Am J Physiol Heart Circ Physiol
.
2021
321
1
H15
H28
.
11.
Beyer
AM
,
Raffai
G
,
Weinberg
B
,
Fredrich
K
,
Lombard
JH
.
Dahl salt-sensitive rats are protected against vascular defects related to diet-induced obesity
.
Hypertension
.
2012
;
60
(
2
):
404
10
.
12.
Fernandes
R
,
Garver
H
,
Harkema
JR
,
Galligan
JJ
,
Fink
GD
,
Xu
H
.
Sex differences in renal inflammation and injury in high–fat diet fed Dahl Salt-sensitive rats
.
Hypertension
.
2018
;
72
(
5
):
e43
e52
.
13.
Shannon
P
,
Markiel
A
,
Ozier
O
,
Baliga
NS
,
Wang
JT
,
Ramage
D
.
Cytoscape: a software environment for integrated models of biomolecular interaction networks
.
Genome Res
.
2003
;
13
(
11
):
2498
504
.
14.
Janky
R
,
Verfaillie
A
,
Imrichová
H
,
Van de Sande
B
,
Standaert
L
,
Christiaens
V
.
iRegulon: from a gene list to a gene regulatory network using large motif and track collections
.
PLoS Comput Biol
.
2014
;
10
(
7
):
e1003731
.
15.
Bella
J
,
Hulmes
DJ
.
Fibrillar collagens
.
Subcell Biochem
.
2017
;
82
:
457
90
.
16.
Shuttleworth
CA
.
Type VIII collagen
.
Int J Biochem Cell Biol
.
1997
;
29
(
10
):
1145
8
.
17.
Hansen
NU
,
Willumsen
N
,
Sand
JM
,
Larsen
L
,
Karsdal
MA
,
Leeming
DJ
.
Type VIII collagen is elevated in diseases associated with angiogenesis and vascular remodeling
.
Clin Biochem
.
2016
;
49
(
12
):
903
8
.
18.
Lopes
J
,
Adiguzel
E
,
Gu
S
,
Liu
SL
,
Hou
G
,
Heximer
S
.
Type VIII collagen mediates vessel wall remodeling after arterial injury and fibrous cap formation in atherosclerosis
.
Am J Pathol
.
2013
;
182
(
6
):
2241
53
.
19.
Bornstein
P
.
Diversity of function is inherent in matricellular proteins: an appraisal of thrombospondin 1
.
J Cell Biol
.
1995
;
130
(
3
):
503
6
.
20.
Murphy-Ullrich
JE
.
Thrombospondin 1 and its diverse roles as a regulator of extracellular matrix in fibrotic disease
.
J Histochem Cytochem
.
2019
;
67
(
9
):
683
99
.
21.
Zhang
K
,
Li
M
,
Yin
L
,
Fu
G
,
Liu
Z
.
Role of thrombospondin‑1 and thrombospondin‑2 in cardiovascular diseases (Review)
.
Int J Mol Med
.
2020
;
45
(
5
):
1275
93
.
22.
Rosini
S
,
Pugh
N
,
Bonna
AM
,
Hulmes
DJS
,
Farndale
RW
,
Adams
JC
.
Thrombospondin-1 promotes matrix homeostasis by interacting with collagen and lysyl oxidase precursors and collagen cross-linking sites
.
Sci Signal
.
2018
11
532
eaar2566
.
23.
Schultz-Cherry
S
,
Chen
H
,
Mosher
DF
,
Misenheimer
TM
,
Krutzsch
HC
,
Roberts
DD
.
Regulation of transforming growth factor-beta activation by discrete sequences of thrombospondin 1
.
J Biol Chem
.
1995
;
270
(
13
):
7304
10
.
24.
Kellouche
S
,
Mourah
S
,
Bonnefoy
A
,
Schoëvaert
D
,
Podgorniak
MP
,
Calvo
F
.
Platelets, thrombospondin-1 and human dermal fibroblasts cooperate for stimulation of endothelial cell tubulogenesis through VEGF and PAI-1 regulation
.
Exp Cell Res
.
2007
;
313
(
3
):
486
99
.
25.
Murphy-Ullrich
JE
,
Poczatek
M
.
Activation of latent TGF-beta by thrombospondin-1: mechanisms and physiology
.
Cytokine Growth Factor Rev
.
2000
11
1–2
59
69
.
26.
Schultz-Cherry
S
,
Lawler
J
,
Murphy-Ullrich
JE
.
The type 1 repeats of thrombospondin 1 activate latent transforming growth factor-beta
.
J Biol Chem
.
1994
;
269
(
43
):
26783
8
.
27.
Tuttle
T
,
Darios
E
,
Watts
SW
,
Roccabianca
S
.
Aortic stiffness is lower when PVAT is included: a novel ex vivo mechanics study
.
Am J Physiol Heart Circ Physiol
.
2022
322
6
H1003
H1013
.
28.
Ruan
CC
,
Zhu
DL
,
Chen
QZ
,
Chen
J
,
Guo
SJ
,
Li
XD
.
Perivascular adipose tissue-derived complement 3 is required for adventitial fibroblast functions and adventitial remodeling in deoxycorticosterone acetate-salt hypertensive rats
.
Arterioscler Thromb Vasc Biol
.
2010
;
30
(
12
):
2568
74
.
29.
Zhu
X
,
Zhang
H-W
,
Chen
H-N
,
Deng
X-J
,
Tu
Y-X
,
Jackson
AO
.
Perivascular adipose tissue dysfunction aggravates adventitial remodeling in obese mini pigs via NLRP3 inflammasome/IL-1 signaling pathway
.
Acta Pharmacol Sin
.
2019
;
40
(
1
):
46
54
.
30.
Contreras
GA
,
Thelen
K
,
Ayala-Lopez
N
,
Watts
SW
.
The distribution and adipogenic potential of perivascular adipose tissue adipocyte progenitors is dependent on sexual dimorphism and vessel location
.
Physiol Rep
.
2016
;
4
(
19
):
e12993
.