Transcriptome-wide RNA sequencing analysis of rat skeletal muscle feed arteries. II. Impact of exercise training in obesity

Abstract

We employed next-generation RNA sequencing (RNA-Seq) technology to determine the extent to which exercise training alters global gene expression in skeletal muscle feed arteries and aortic endothelial cells of obese Otsuka Long-Evans Tokushima Fatty (OLETF) rats. Transcriptional profiles of the soleus and gastrocnemius muscle feed arteries (SFA and GFA, respectively) and aortic endothelial cell-enriched samples from rats that underwent an endurance exercise training program (EndEx; n = 12) or a interval sprint training program (IST; n = 12) or remained sedentary (Sed; n = 12) were examined. In response to EndEx, there were 39 upregulated (e.g., MANF) and 20 downregulated (e.g., ALOX15) genes in SFA and 1 upregulated (i.e., Wisp2) and 1 downregulated (i.e., Crem) gene in GFA [false discovery rate (FDR) < 10%]. In response to IST, there were 305 upregulated (e.g., MANF, HSPA12B) and 324 downregulated genes in SFA and 101 upregulated and 66 downregulated genes in GFA, with an overlap of 32 genes between arteries. Furthermore, in aortic endothelial cells, there were 183 upregulated (e.g., eNOS, SOD-3) and 141 downregulated (e.g., ATF3, Clec1b, npy, leptin) genes with EndEx and 71 upregulated and 69 downregulated genes with IST, with an overlap of 35 between exercise programs. Expression of only two genes (Tubb2b and Slc9a3r2) was altered (i.e., increased) by exercise in all three arteries. The finding that both EndEx and IST produced greater transcriptional changes in the SFA compared with the GFA is intriguing when considering the fact that treadmill bouts of exercise are associated with greater relative increases in blood flow to the gastrocnemius muscle compared with the soleus muscle.

the increase in skeletal muscle blood flow during treadmill running in rats is not uniform across all hindlimb skeletal muscles (1, 2, 4, 10, 24, 33–35). This has provided a unique and exquisite model to investigate the role of increased blood flow as a signal mediating exercise-induced vascular adaptations (5, 26, 28–32). While the rat is standing, blood flow to the primarily slow-twitch, oxidative soleus muscle is two- to fourfold greater than that to the primarily fast-twitch, glycolytic gastrocnemius muscle (1). However, during treadmill running, recruitment of muscle fibers in the gastrocnemius muscle results in a drastic increase in blood flow such that blood flow to the soleus and gastrocnemius muscles are similar (1). When considering this information, it is not unexpected that the functional vascular adaptations [e.g., enhanced endothelium-dependent dilation (EDD)] to exercise training are largely conferred within the gastrocnemius muscle (5, 28), where the relative increase in muscle blood flow from rest to exercise is highest (1). Reciprocally, functional vascular adaptations are not observed in the soleus muscle (5, 18, 28), likely because this is a muscle largely recruited during standing and its relative increase in blood flow during exercise is substantially less than that of the red portion of the gastrocnemius muscle (1, 2, 33). In short, functional vascular adaptations to exercise appear to be dependent on the relative increase in blood flow, thus supporting the idea that shear stress may play a key role in modulating the phenotype of vascular cells with exercise training (19, 25, 36, 45).

To gain further insights into the molecular events underpinning exercise-induced vascular adaptations, we assessed transcriptional profiles in skeletal muscle feed arteries from sedentary and exercise-trained animals, using the same techniques as those employed in our companion paper (21). Specifically, we performed a transcriptome-wide RNA sequencing (RNA-Seq) analysis in the soleus (SFA) and gastrocnemius (GFA) feed arteries from a group of Otsuka Long-Evans Tokushima Fatty (OLETF) rats that underwent an endurance exercise training program (EndEx), a group that underwent an interval sprint training program (IST), and a group that remained sedentary (Sed). We hypothesized that the greatest effects of exercise training on the transcriptome would be in the GFA compared with the SFA. Furthermore, we reasoned that, in the GFA, IST would produce greater vascular transcriptional changes compared with EndEx because of greater increases in skeletal muscle fiber recruitment of this muscle during sprinting. We also sought to determine the influence of exercise training on the endothelium per se, and therefore tested the hypothesis that aortic endothelial cell-enriched samples from exercise-trained OLETF rats would display a more atheroprotective gene expression profile compared with sedentary OLETF rats.

METHODS

Animals and experimental design.

OLETF rats (n = 36) were obtained at age 4 wk (Japan SLC, Hamamatsu, Shizuoka, Japan). The OLETF rat, characterized by a mutated cholecystokinin-1 receptor that results in a hyperphagic phenotype, is an established model of obesity, insulin resistance, and type 2 diabetes mellitus (22). Animals were individually housed in a temperature-controlled (21°C) environment with light from 0600 to 1800 and dark from 1800 to 0600. All animals were given ad libitum access to standard chow with a macronutrient composition of 56% carbohydrate, 17% fat, and 27% protein (Formulab 5008, Purina Mills, St. Louis, MO). At 20 wk of age, rats were randomly assigned to one of three groups: 1) Sed (n = 12), 2) EndEx (n = 12), and 3) IST (n = 12). For EndEx, treadmill running duration and intensity were increased progressively over the first 4 wk to reach 60 min of treadmill running at 20 m/min at a 15% incline for the remaining 8 wk (1.2 km/day). For IST, six bouts of treadmill running, with 4.5-min rest periods, were progressively increased in duration and intensity over the first 5 wk to reach running speeds of 40 m/min at a 15% incline for 2.5 min/bout for the remaining 7 wk (0.6 km/day). Both EndEx and IST groups exercised for 5 days/wk.

Rats were anesthetized at 30–32 wk of age with an intraperitoneal injection of pentobarbital sodium (50 mg/kg) between 0800 and 0930. Tissues were then harvested, and the animals were killed by exsanguination. The last exercise bout for EndEx and IST animals was performed ∼18 h prior to death. Food was removed from the cages 12 h prior to death. All protocols were approved by the University of Missouri Animal Care and Use Committee.

Body weight, body composition, food intake, and citrate synthase.

Body weights and food intakes were monitored and recorded on a weekly basis. Weekly food intakes were averaged across the period of the intervention (age 20–30 wk). Body composition was assessed by dual-energy X-ray absorptiometry (DXA; Hologic QDR-1000, calibrated for rodents) on the day of death. Omental and retroperitoneal adipose tissue depots were then removed and weighed to the nearest 0.01 g. Citrate synthase activity was measured from whole muscle homogenate of the vastus lateralis with the spectrophotometric method of Srere (43).

Isolation of skeletal muscle feed arteries and aortic endothelial cells.

Immediately after the gastrocnemius-plantaris-soleus muscle complex was harvested, the muscles were pinned down in a petri dish containing a cold RNA-stabilizing agent (RNAlater; Ambion, Austin, TX). Under the microscope, the SFA and GFA were then dissected clean of perivascular adipose tissue and excess adventitia as described previously (5, 20, 26, 28, 29, 46). The single GFA supplying the medial head of the gastrocnemius muscle was used for the present study. In our experience, OLETF rats typically have one to three SFA. All SFA were dissected and pooled for the present RNA-Seq analysis. The reader is referred to our recent publication for a visual of the anatomic location and structure of GFA and SFA (20). Aortic endothelial cells were isolated by gentle scraping of longitudinally opened aortas as described previously (8, 38, 40, 42). This method of scraping the luminal surface yields an endothelial cell-enriched sample (38). The aorta was chosen to provide sufficient yield of endothelial cells. Indeed, we have established that we recover ∼200 ng of total RNA per rat aortic endothelial scrape, which is above the minimum required for RNA-Seq experiments. Samples were kept in RNAlater for 48 h at 4°C and then removed from the RNAlater solution and stored at −80°C until analysis.

Blood parameters.

Whole blood was collected on the day of euthanasia for analysis of glycosylated hemoglobin (HbA1c) by the boronate-affinity high-performance liquid chromatography method (Primus Diagnostics, Kansas City, MO) in the Diabetes Diagnostics Laboratory at the University of Missouri. Serum samples were prepared by centrifugation and stored at −80°C until analysis. Glucose, triglyceride (TG), and nonesterified fatty acid (NEFA) assays were performed by a commercial laboratory (Comparative Clinical Pathology Services, Columbia, MO) on an Olympus AU680 automated chemistry analyzer (Beckman-Coulter, Brea, CA) using commercially available assays according to manufacturers' guidelines. Plasma insulin concentrations were determined with a commercially available rat-specific enzyme-linked immunosorbent assay (Alpco Diagnostics, Salem, NH). Samples were run in duplicate, and manufacturers' controls and calibrators were used according to assay instructions.

RNA extraction.

Total RNA isolations were performed with the NucleoMag 96 RNA kit procedure (Clontech part no. 744350.1), which is a single-tube method based on reactive magnetic bead technology designed for automated small- or large-scale preparation of highly pure total RNA from tissue or cell samples. All liquid handling was optimized for use on a Beckman3000 robotic liquid handler housed within a laminar flow hood (with UV decontamination) designed to ensure a clean room environment for working with microtissues, which yield low (pg to ng) amounts of RNA. Briefly, groups of 24 sample vessels were removed from −80°C and immediately homogenized for 60–120 s in their own 2-ml microtube with the Mini-Beadbeater-24 (BioSpec Products) in the presence of NucleoMag lysis buffer and several miniature chrome-steel (RNase treated) BBs. Care was taken to get complete microvessel disruption without extending grinding times to prevent the generation of excess heat. The resulting homogenate was then loaded onto the robot deck, and a digital photo was taken before the sample was transferred into 96-deep well microplates. The photo allowed us to have a physical record of each sample ID prior to loading into the microplate for accurate tracking purposes. This process was repeated four times to completely fill a 96-well plate within 10–20 min. The combination of using stabilized tissue and immediate homogenization in chaotropic salt-containing lysis buffer ensured that the RNA was protected from RNase degradation during tissue disruption. After homogenization, the RNA was bound to RNA beads in the presence of alcohol, and a magnet was used to perform several wash and elution steps in a completely automated fashion. This method included a DNase digestion step ultimately yielding RNA of similar yield and quality from column-based procedures. Immediately after RNA isolation, pure RNA was transferred to a new 96-well plate and a 5-μl aliquot was taken into a second plate for RNA quality control. Both plates were stored at −80°C with cryogenic plate seals and placed in secondary containment to prevent frost buildup on the plates during storage.

RNA quality control (concentration and integrity).

For assessing total RNA yield and integrity, tandem Agilent Bioanalyzer 2100 instruments were used in combination with the Total RNA 6000 Pico Assay. At the time of this study, the RNA Pico LabChip Kit was the only platform to give unbiased assessment of RNA integrity (via RIN) and accurate results with extremely low RNA concentrations such as those provided by microvessels. The qualitative range for the total RNA assay is 200–5,000 pg/μl, and the most important advantage of this system is the small amount of sample used (1 μl), leaving the rest of the RNA for other applications. Typical yields from rat microvessels were ∼500–1,000 pg/μl. RNA quality control was performed with only the aliquot from each isolation plate.

Illumina library preparation (SMARTer amplification and RNA-Seq).

Because of the low yields of total RNA from microvessels, total RNA could not be used directly in traditional Illumina gene expression profiling methods (RNA-Seq) because of the low concentration of the samples (standard RNA-Seq kits during this project required 0.1–1 μg of total RNA). Thus the SMARTer Ultra Low RNA Kit for Illumina Sequencing (Clontech, catalog no. 634935) was utilized for generating full-length cDNA transcripts prior to Illumina RNA-Seq library preparation. Briefly, the technology involves SMARTer first-strand cDNA synthesis and purification, utilizing the SMARTer anchor sequence and poly(A) sequence that serve as universal priming sites for end-to-end generation of single-stranded cDNA, followed by cDNA amplification with LongDistance PCR (LD-PCR) using the manufacturer's recommended Advantage 2 PCR system (Clontech, catalog no. PT3281-1) containing a novel polymerase and ultrapure dNTPs specifically for Illumina sequencing. Using the concentration values from the Bioanalyzer RNA Pico Assay, we sought to use 100–1,000 pg of total RNA as input to the SMARTer 1st cDNA reaction.

After cDNA generation, validation was performed with thes Bioanalyzer 2100 High Sensitivity DNA Assay (Agilent, catalog no. 5067-4626) for select samples from each plate of 96 samples in order to accurately size and quantitate DNA up to 12 kb in length, again consuming minimal sample volumes (1 μl). After 14 cycles of LD-PCR amplification the final cDNA yields were estimated at ∼1–10 ng for each microvessel, which is a suitable input amount for library preparation for cDNA/ChIP Seq library preparation. To generate Illumina Paired-End RNAseq libraries, cDNA was fragmented to ∼200 bp with the Q700 DNA fragmentation system (QSonica) and then used directly with the NextFlex DNA preparation kit (Bioo Scientific, catalog no. 5140-02) with some modifications. Briefly, fragment cDNA was end-repaired and purified with 1.8x SPRI beads to remove reaction components (Agencourt). The resulting blunt ends were A-tailed in preparation for cohesive ligation to the Illumina specific sequencing adapters diluted to 0.6 μM working concentration (NextFlex DNA Adapters, Bioo Scientific, catalog no. 514104). Ligated DNA was purified twice with 1.0x SPRI to remove adapter dimers and perform gel-free size selection and then amplified through 14 cycles of PCR. The final sequencing construct was purified with a 1.0x SPRI to remove low-molecular-weight adapter dimer artifacts (if any), and libraries were validated to contain ∼330-bp fragments with the Bioanalyzer 2100 High Sensitivity DNA Assay. Library quantitation was performed with the Qubit 2.0 fluorometer and the High Sensitivity DNA assay (Life Tech, catalog no. Q32851).

RNA sequencing.

By utilizing 48 unique adapter indexes during library preparation across each plate of 96 libraries created, we were able to overcome several common technical mistakes. First, it allowed us to account for technical biases through randomization of samples by vessel type (group) and treatment across each plate of samples. Second, by having a priori knowledge of the sequencing index used to identify each sample from the sequencing, we were able to use a single manufacturing lot of adapters that were uniformly diluted and preplated to ensure similar ligation efficiencies across several plates (hundreds of samples) used in the study. Third, this indexing scheme allowed us to standardize the pooling of several libraries by row within each plate, where equimolar volumes of each sample in a plate row were pooled to a final concentration of 5–10 nM. Altogether, this approach prevented inadvertent use of the wrong adapters during preparation, randomized the sample and index combinations, and allowed for reduced mixing up of libraries within each sequencing pool. The final pools (>50 total) were each loaded on a single lane of single-read 50-base sequencing on the Illumina HiSeq2000 and ultimately yielded ∼175–200 million useable reads per lane (14–17 million reads per RNAseq sample). It should be noted that the combination of the 48 adapters that resulted in 4 pools of 12 indexes was carefully designed and wet lab tested to be compatible with the HiSeq to ensure maximum sequence yields and to ensure that each sample was correctly identified by the HiSeq during the index identification steps.

Statistical analysis.

The analysis of the RNA-Seq data was carried out for the SFA and GFA samples as described in our companion paper (21). There were minor differences in some analysis parameters for the aortic endothelial cell samples, as follows. Nonspecific filtering of genes prior to statistical testing was carried out to increase detection power (7), based on the requirement that a gene have an expression level greater than 32 counts per million reads mapped (CPM) for at least 12 libraries (12 was chosen because it was the size of each of the groups involving aortic endothelial cells). This CPM cutoff was established empirically based on the point at which the ERCC Spike-Ins at different concentrations were no longer distinguishable in the aortic endothelial samples. Adjustment to the P values was made to account for multiple testing using the false discovery rate (FDR) method of Benjamini and Hochberg (6). For all comparisons involving aortic endothelial cells (e.g., EndEx vs. Sed within aortic endothelial cells), we chose 10% as our FDR threshold for statistical significance. As an empirical measure of the FDR, we evaluated what proportion of the identical ERCC probe/concentration combinations (“Set B”) appeared in our list of differentially expressed genes. Similarly, we looked at a set of 13 putative housekeeping genes derived from a study of more than 13,000 rat samples (11) to have another estimate of our FDR. The set of genes was Actb, B2m, Gapdh, Gusb, Hprt1, Hmbs, Hsp90b1, Rpl13a, Rps29, Rplp0, Ppia, Tbp, and Tuba1. For both of these sets of controls, we also estimated the fold change of each of the genes as a measure of the accuracy of the fold change estimates.

Based on our list of differentially expressed genes, networks were generated through the use of Ingenuity Pathway Analysis (Ingenuity Systems, www.ingenuity.com), henceforth IPA, as previously described (21, 40).

Finally, for the remaining data (i.e., non-RNA-Seq) between-group differences for all descriptive variables were determined by using an independent t-test, for which statistical significance was declared at P ≤ 0.05.

RESULTS

Animal characteristics are summarized in Table 1. For the four subsequent group comparisons reported involving SFA and GFA, the average ERCC Spike-In (Set B, n = 10 probes) empirical FDR was 7.5% at the nominal FDR cutoff of 10% (mean fold = 1.08), while for the putative housekeeping genes the average was 18% (fold = 0.97). For the two subsequent group comparisons reported involving aortic endothelial cells, the average ERCC Spike-In (Set B, n = 5 probes) empirical FDR was 15% at the nominal FDR cutoff of 10% (mean fold = 1.07), while for the putative housekeeping genes the average was 0% (fold = 1.55). It should be noted that the statistics reported for the putative housekeeping genes are only based on 11 (SFA and GFA) or 10 (aortic endothelial cells) of the 13 planned housekeeping genes, because some genes were expressed at low levels and did not pass the nonspecific filtering criteria. In their entirety these findings strongly support the methodology used because, on average, the fold changes for these controls are approximately equal to 1 and the empirical FDR is approximately equal to the target FDR (10%).

Table 1.

Animal characteristics

Variable	Sed	EndEx	IST
BW, g	687 ± 12	592 ± 11*	589 ± 13*
Food intake, g/day	31.9 ± 0.5	28.6 ± 0.5*	27.3 ± 0.8*
Food intake, g·day−1·g BW−1	0.330 ± 0.003	0.322 ± 0.005	0.314 ± 0.005*
Body fat, %	35.0 ± 1.2	25.0 ± 1.4*	27.3 ± 1.3*
Retroperitoneal adipose tissue mass, g	47.0 ± 2.7	28.5 ± 2.7*	30.6 ± 2.5*
Omental adipose tissue mass, g	3.2 ± 0.4	1.5 ± 0.2*	1.8 ± 0.1*
Epididymal adipose tissue mass, g	19.5 ± 0.9	13.3 ± 0.8*	14.6 ± 0.6*
HW, g	1.92 ± 0.04	1.89 ± 0.04	1.90 ± 0.05
HW-to-BW ratio (× 103)	2.8 ± 0.1	3.2 ± 0.1*	3.2 ± 0.1*
Citrate synthase VL-red, μmol·min−1·g−1	28.3 ± 1.6	35.9 ± 1.3*	35.6 ± 1.5*
Citrate synthase VL-white, μmol·min−1·g−1	8.0 ± 0.3	11.0 ± 0.8*	16.8 ± 0.9*†
Total cholesterol, mg/dl	147.4 ± 6.5	95.0 ± 5.0*	100.6 ± 8.4*
LDL-cholesterol, mg/dl	43.0 ± 6.5	43.3 ± 3.7	51.8 ± 5.2
HDL-cholesterol, mg/dl	33.0 ± 1.8	27.8 ± 1.1*	26.3 ± 1.8*
Triglycerides, mg/dl	357.1 ± 43.8	119.8 ± 18.2*	112.6 ± 14.2*
NEFA, mmol/l	0.97 ± 0.06	0.57 ± 0.06*	0.61 ± 0.05*
Insulin, ng/ml	8.1 ± 1.4	4.1 ± 0.5*	3.9 ± 0.6*
Glucose, mg/dl	297.0 ± 14.3	246.5 ± 26.6	208.7 ± 9.9*
HOMA-IR index	6.1 ± 1.2	2.6 ± 0.5*	2.0 ± 0.2*
HbA1c, %	7.2 ± 0.3	5.4 ± 0.1*	5.5 ± 0.1*

Values are means ± SE.

Sed, sedentary; EndEx, endurance exercise trained; IST, interval sprint trained; BW, body weight; HW, heart weight; NEFA, nonesterified fatty acids; HbA1c, glycosylated hemoglobin.

^*Significant difference from Sed (P < 0.05);

^†significant difference from EndEx (P < 0.05).

Impact of endurance exercise on SFA and GFA.

Figure 1 displays the number of genes in SFA and GFA differentially expressed between EndEx and Sed rats. There were 39 upregulated and 20 downregulated genes in SFA and 1 upregulated and 1 downregulated gene in GFA (FDR < 10%). Expression of both of these genes (i.e., Wisp2 and Crem) was also altered in the SFA. Table 2 provides a list of the top 20 genes differentially expressed between EndEx and Sed groups in the SFA (sorted by magnitude of fold change), and the full list of differentially expressed genes is provided in the Supplemental Table S1.¹ Table 3 provides information on the two genes that were differentially expressed between EndEx and Sed groups in the GFA. Figure 2 and Figure 3 illustrate the top-scoring gene networks influenced by EndEx in the SFA and the GFA, respectively. The names and scores for these gene networks were “Amino Acid Metabolism, Post-Translational Modification, Small Molecule Biochemistry” (score = 52) in the SFA and “Neurological Disease, Cardiovascular System Development and Function, Hematological System Development and Function” (score = 6) in the GFA. Overall, there was no overlap in the top gene networks influenced by EndEx between arteries (Supplemental Table S2).

Fig. 1.

Number of genes altered by endurance exercise (EndEx) in soleus and gastrocnemius muscle feed arteries (SFA and GFA, respectively); ↑ indicates upregulation in EndEx relative to sedentary (Sed), and ↓ indicates downregulation in EndEx relative to Sed. Circle sizes and overlapping area are proportional to the number of genes altered.

Table 2.

Top 20 genes differentially expressed between EndEx and Sed in SFA, sorted by magnitude of fold change

EntrezID	Symbol	Name	FDR	Fold
EndEx > Sed in SFA
65206	Kcnn4	Potassium intermediate/small conductance calcium-activated channel, subfamily N, member 4	0.026	3.8
363133	Pcbp4	Poly(rC) binding protein 4	0.078	3.2
308408	Gpr4	G protein-coupled receptor 4	0.004	2.8
361473	Mtrf1l	Mitochondrial translational release factor 1-like	0.095	2.5
171445	Akr7a2	Aldo-keto reductase family 7, member A2 (aflatoxin aldehyde reductase)	0.095	2.5
171440	Abtb2	Ankyrin repeat and BTB (POZ) domain containing 2	0.078	2.5
500795	RGD1562114	RGD1562114	0.004	2.4
315989	Manf	Mesencephalic astrocyte-derived neurotrophic factor	0.021	2.4
362514	Tstd2	Thiosulfate sulfurtransferase (rhodanese)-like domain containing 2	0.003	2.3
252934	Cox5a	Cytochrome c oxidase, subunit Va	0.095	2.3
362978	Creld2	Cysteine-rich with EGF-like domains 2	0.056	2.1
312638	Creld1	Cysteine-rich with EGF-like domains 1	0.055	2.1
361436	Afg3l1	AFG3(ATPase family gene 3)-like 1 (S. cerevisiae)	0.038	2.0
289707	Lyar	Ly1 antibody reactive	0.095	1.9
29576	Wisp2	WNT1 inducible signaling pathway protein 2	0.012	1.9
EndEx < Sed in SFA
287554	Tefm	Transcription elongation factor, mitochondrial	0.033	−2.3
81639	Alox15	Arachidonate 15-lipoxygenase	0.033	−2.1
360733	Rtp4	Receptor (chemosensory) transporter protein 4	0.021	−2.1
295037	Mgst2	Microsomal glutathione S-transferase 2	0.043	−2.0
94188	Zfp423	Zinc finger protein 423	0.097	−1.9

SFA, soleus feed artery.

Table 3.

Genes differentially expressed between EndEx and Sed in GFA, sorted by magnitude of fold change

EntrezID	Symbol	Name	FDR	Fold
EndEx > Sed in GFA
29576	Wisp2	WNT1 inducible signaling pathway protein 2	0.025	2.0
EndEx < Sed in GFA
25620	Crem	cAMP responsive element modulator	0.025	−1.9

GFA, gastrocnemius feed artery.

Fig. 2.

Top-scoring gene network influenced by EndEx in the SFA (score = 52). Nodes represent genes/molecules. Shading is in proportion to the size of the fold change (red, upregulation; green, downregulation). Gray nodes denote network members that did not reach false discovery rate (FDR) < 10%.

Fig. 3.

Top-scoring gene network influenced by EndEx in the GFA (score = 6). Nodes represent genes/molecules. Shading is in proportion to the size of the fold change (red, upregulation; green, downregulation). White nodes denote network members that were not represented in the RNA-seq database. Gray nodes denote network members that did not reach FDR < 10%.

Impact of interval sprint training on SFA and GFA.

Figure 4 displays the number of genes in SFA and GFA differentially expressed between IST and Sed rats. There were 305 upregulated and 324 downregulated genes in SFA and 101 upregulated and 66 downregulated genes in GFA (FDR < 10%). Of the 32 genes altered in both SFA and GFA (overlapping portion of Venn diagram in Fig. 4), 13 were upregulated and 19 were downregulated in both vessels. Table 4 (SFA) and Table 5 (GFA) provide lists of the top 20 genes differentially expressed between IST and Sed groups, and the full lists of differentially expressed genes are provided in Supplemental Table S1. Figure 5 and Figure 6 illustrate the top-scoring gene networks influenced by IST in the SFA and GFA, respectively. The names and scores for these gene networks were “Gene Expression, Post-Translational Modification, Cancer” (score = 46) in the SFA and “Organ Morphology, Skeletal and Muscular System Development and Function, Cancer” (score = 41) in the GFA. Overall, there was very little or no overlap in the top gene networks influenced by IST between arteries (Supplemental Table S2).

Fig. 4.

Top: number of genes altered by interval sprint training (IST) in SFA and GFA; ↑ indicates upregulation in IST relative to Sed, and ↓ indicates downregulation in IST relative to Sed. Circle sizes and overlapping area are proportional to the number of genes altered. Bottom: between-artery correlation in changes of gene expression induced by IST. There was an overlap of 32 genes between SFA and GFA. Each dot represents a gene. Dashed line of identity indicates perfect agreement between the arteries.

Table 4.

Top 20 genes differentially expressed between IST and Sed in SFA, sorted by magnitude of fold change

EntrezID	Symbol	Name	FDR	Fold
IST > Sed in SFA
140942	Ddit4	DNA-damage-inducible transcript 4	0.025	4.8
406169	Atf6b	Activating transcription factor 6 beta	0.002	3.7
65206	Kcnn4	Potassium intermediate/small conductance calcium-activated channel, subfamily N, member 4	0.004	3.5
362514	Tstd2	Thiosulfate sulfurtransferase (rhodanese)-like domain containing 2	<0.001	3.4
361810	Fkbp5	FK506 binding protein 5	0.003	3.4
66026	Trpv4	Transient receptor potential cation channel, subfamily V, member 4	<0.001	3.2
363133	Pcbp4	Poly(rC) binding protein 4	0.016	3.1
266687	Slc35e4	Solute carrier family 35, member E4	0.002	3.0
308408	Gpr4	G protein-coupled receptor 4	<0.001	3.0
84489	Fgfr3	Fibroblast growth factor receptor 3	0.022	2.8
362613	Zdhhc18	Zinc finger, DHHC-type containing 18	0.001	2.8
304021	Col8a1	Collagen, type VIII, alpha 1	0.012	2.7
690617	LOC690617	Hypothetical protein LOC690617	0.005	2.6
362111	Sh2d3c	SH2 domain containing 3C	0.002	2.6
311427	Hspa12b	Heat shock protein 12B	<0.001	2.5
IST < Sed in SFA
500037	Foxp2	Forkhead box P2	<0.001	−5.6
25512	Phex	Phosphate regulating endopeptidase homolog, X-linked	0.026	−3.8
29157	Ccng2	Cyclin G2	0.002	−2.9
365748	Bhlhe22	Basic helix-loop-helix family, member e22	0.037	−2.9
309922	Rhobtb3	Rho-related BTB domain containing 3	0.002	−2.6

Table 5.

Top 20 genes differentially expressed between IST and Sed in GFA, sorted by magnitude of fold change

EntrezID	Symbol	Name	FDR	Fold
IST > Sed in GFA
29389	Tnni2	Troponin I type 2 (skeletal, fast)	<0.001	38.7
691644	Myh2	Myosin, heavy chain 2, skeletal muscle, adult	<0.001	31.3
287408	Myh1	Myosin, heavy chain 1, skeletal muscle, adult	<0.001	29.6
296369	Tnnc2	Troponin C type 2 (fast)	0.015	25.4
292879	Mybpc2	Myosin binding protein C, fast-type	<0.001	25.3
25269	Pvalb	Parvalbumin	<0.001	24.2
301073	Hhatl	Hedgehog acyltransferase-like	0.001	21.3
500377	Tuba8	Tubulin, alpha 8	0.015	20.3
171009	Actn3	Actinin alpha 3	<0.001	19.8
503446	Asb12	Ankyrin repeat and SOCS box-containing 12	<0.001	19.7
24837	Tnnt2	Troponin T type 2 (cardiac)	0.001	19.1
114508	Fbp2	Fructose-1,6-bisphosphatase 2	0.034	17.7
24584	Mylpf	Myosin light chain, phosphorylatable, fast skeletal muscle	<0.001	17.3
25028	Ampd1	Adenosine monophosphate deaminase 1	0.001	17.2
291605	Myot	Myotilin	0.064	16.5
360543	Myh4	Myosin, heavy chain 4, skeletal muscle	0.001	14.4
498440	Myoz1	Myozenin 1	<0.001	13.7
361002	Dusp13	Dual specificity phosphatase 13	0.003	12.8
287881	Ppp1r27	Protein phosphatase 1, regulatory subunit 27	0.076	12.0
290223	Fitm1	Fat storage-inducing transmembrane protein 1	0.05	10.9

Fig. 5.

Top-scoring gene network influenced by IST in the SFA (score = 46). Nodes represent genes/molecules. Shading is in proportion to the size of the fold change (red, upregulation; green, downregulation). Gray nodes denote network members that did not reach FDR < 10%.

Fig. 6.

Top-scoring gene network influenced by IST in the GFA (score = 41). Nodes represent genes/molecules. Shading is in proportion to the size of the fold change (red, upregulation; green, downregulation). Gray nodes denote network members that did not reach FDR < 10%.

Impact of endurance exercise and interval sprint training on aortic endothelial cells.

Figure 7 displays the number of genes in aortic endothelial cells differentially expressed between EndEx and Sed rats and between IST and Sed rats. There were 183 upregulated and 141 downregulated genes with EndEx and 71 upregulated and 69 downregulated genes with IST (FDR < 10%). Of the 35 genes altered by both EndEx and IST (overlapping portion of Venn diagram in Fig. 7), 16 were upregulated and 19 were downregulated. Expression of only two genes (i.e., Tubb2b and Slc9a3r2) was altered (i.e., increased) by exercise in the SFA and the GFA as well as in aortic endothelial cells. Table 6 (EndEx) and Table 7 (IST) provide lists of the top 20 genes differentially expressed in aortic endothelial cells, and the full lists of differentially expressed genes are provided in Supplemental Table S1. Figure 8 and Figure 9 illustrate the top-scoring gene networks influenced by EndEx and IST in aortic endothelial cells, respectively. The names and scores for these gene networks were “Cellular Function and Maintenance, Nervous System Development and Function, Organismal Injury and Abnormalities” (score = 42) for EndEx and “Cell-to-Cell Signaling and Interaction, Cellular Function and Maintenance, Connective Tissue Disorders” (score = 45) for IST. Overall, there was very little or no overlap in the top gene networks influenced by exercise between aortic endothelial cells and feed arteries (Supplemental Table S2).

Fig. 7.

Top: number of genes altered by EndEx and IST in aortic endothelial cells (AECs); ↑ indicates upregulation relative to Sed, and ↓ indicates downregulation relative to Sed. Circle sizes and overlapping area are proportional to the number of genes altered. Bottom: correlation in changes of gene expression induced by EndEx and IST. There was an overlap of 35 genes between the modes of exercise. Each dot represents a gene. Dashed line of identity indicates perfect agreement between the modes of exercise.

Table 6.

Top 20 genes differentially expressed between EndEx and Sed in aortic ECs, sorted by magnitude of fold change

EntrezID	Symbol	Name	FDR	Fold
EndEx > Sed in Aortic ECs
259247	Obp3	Alpha-2u globulin PGCL4	0.014	2.5
29687	C1qb	Complement component 1, q subcomponent, B chain	0.006	2.1
300664	Abcg4	ATP-binding cassette, subfamily G (WHITE), member 4	0.051	2.0
362634	C1qc	Complement component 1, q subcomponent, C chain	0.011	2.0
83801	Ptms	Parathymosin	0.053	2.0
499088	Dact3	Dapper, antagonist of beta-catenin, homolog 3 (Xenopus laevis)	0.023	1.9
24695	Pthlh	Parathyroid hormone-like hormone	0.014	1.9
EndEx < Sed in Aortic ECs
500336	Clec1b	C-type lectin domain family 1, member B	<0.001	−6.3
294254	Hspa1b	Heat shock 70 kDa protein 1B (mapped)	<0.001	−5.0
24604	Npy	Neuropeptide Y	0.003	−4.9
25608	Lep	Leptin	0.017	−4.7
25389	Atf3	Activating transcription factor 3	0.071	−2.7
293524	Bag3	BclII-associated athanogene 3	0.014	−2.6
291863	LOC291863	Carboxylesterase-like	0.062	−2.6
361384	Dnajb1	DnaJ (Hsp40) homolog, subfamily B, member 1	0.011	−2.5
83574	Rxrg	Retinoid X receptor gamma	0.057	−2.4
50681	Acox1	Acyl-CoA oxidase 1, palmitoyl	0.014	−2.3
89813	Pdk4	Pyruvate dehydrogenase kinase, isozyme 4	0.029	−2.1
24468	Hspa8	Heat shock 70 kDa protein 8	<0.001	−2.1
29436	Tfpi	Tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor)	0.029	−2.0

EC, endothelial cell.

Table 7.

Top 20 genes differentially expressed between IST and Sed in aortic ECs, sorted by magnitude of fold change

EntrezID	Symbol	Name	FDR	Fold
IST > Sed in Aortic ECs
361619	Hbb-b1	Hemoglobin, beta adult major chain	0.084	3.6
290636	Pgls	6-Phosphogluconolactonase	0.029	2.7
24517	Junb	Jun B proto-oncogene	0.021	2.6
305897	Nfatc4	Nuclear factor of activated T cells, cytoplasmic, calcineurin-dependent 4	0.021	2.5
309170	Fam89b	Family with sequence similarity 89, member B	0.017	2.4
24723	Rn45s	45S preribosomal RNA	0.017	2.2
288916	Gadd45 gip1	Growth arrest and DNA-damage-inducible, gamma interacting protein 1	0.036	2.2
362634	C1qc	Complement component 1, q subcomponent, C chain	0.009	2.1
291327	Mrc1	Mannose receptor, C type 1	0.009	2.1
288704	RGD1311899	Similar to RIKEN cDNA 2210016L21 gene	0.034	2.0
64303	Pfn1	Profilin 1	0.03	2.0
368001	Bcl7b	B-cell CLL/lymphoma 7B	0.084	2.0
361545	Rbm42	RNA binding motif protein 42	0.036	1.9
499913	Abhd12	Abhydrolase domain containing 12	0.036	1.8
300791	Spg21	Spastic paraplegia 21 homolog (human)	0.072	1.7
298566	C1qa	Complement component 1, q subcomponent, A chain	0.073	1.7
IST < Sed in Aortic ECs
25420	Cryab	Crystallin, alpha B	0.027	−2.2
295588	Rnd3	Rho family GTPase 3	0.043	−2.0
287382	Mfap4	Microfibrillar-associated protein 4	0.062	−1.9
295549	Dnajb4	DnaJ (Hsp40) homolog, subfamily B, member 4	0.048	−1.9

Fig. 8.

Top-scoring gene network influenced by EndEx in aortic endothelial cells (ECs) (score = 42). Nodes represent genes/molecules. Shading is in proportion to the size of the fold change (red, upregulation; green, downregulation). Gray nodes denote network members that did not reach FDR < 10%.

Fig. 9.

Top-scoring gene network influenced by IST in aortic ECs (score = 45). Nodes represent genes/molecules. Shading is in proportion to the size of the fold change (red, upregulation; green, downregulation). Gray nodes denote network members that did not reach FDR < 10%.

Changes induced by obesity that were partially or entirely restored by exercise training.

Figure 10 summarizes the number of genes whose expression was significantly changed by hyperphagia-induced obesity (see Ref. 21) and for which exercise training either partially or entirely restored their expression (toward the levels of the lean sedentary animal) (Supplemental Table S2). As shown, of the 415 genes altered by obesity in the SFA, 10 genes were reversed by both EndEx and IST, 3 genes were reversed by EndEx only (i.e., Alox15, Gbp2, and Lsm1), and 51 genes were reversed by IST only. Of the 240 genes altered by obesity in the GFA, no genes were reversed by both EndEx and IST, 1 gene was reversed by EndEx only (i.e., Crem), and 13 genes were reversed by IST only. Of the 396 genes altered by obesity in aortic endothelial cells, 8 genes were reversed by both EndEx and IST, 26 genes were reversed by EndEx only, and 13 genes were reversed by IST only. There were no genes affected by obesity that were restored by exercise in all three arteries (i.e., SFA, GFA, aortic endothelial cells). However, there were two genes (i.e., Crem and Ifit2) in which obesity increased expression in both the SFA and GFA, and exercise training (EndEx, IST, or both) reduced their expression in both arteries. In addition, there was one gene (i.e., ISG15) in which obesity increased expression in both the SFA and aortic endothelial cells and IST reduced expression in both the SFA and aortic endothelial cells. In addition, overall, there was very little or no overlap in the top gene networks influenced by obesity vs. exercise in all three arteries (Supplemental Table S2).

Fig. 10.

Changes induced by obesity that were partially or entirely restored by exercise training. Lists of genes whose expression was significantly changed by hyperphagia-induced obesity were obtained from our companion paper (21).

Furthermore, it should be noted that there were a few genes for which the effect of obesity was further amplified by exercise. That is, the change induced by exercise was in the same direction as that induced by obesity. In the SFA, both obesity and EndEx increased expression of Wisp2. Also in the SFA, both obesity and IST increased expression of Wisp2, F2rl1, and Cct6a and decreased expression of Slc3a1 and Uri1. In the GFA, both obesity and IST decreased expression of Frzb. In the aortic endothelial cells, both obesity and EndEx increased expression of RT1-Db1, RT1-Ba, Fn1, and RT1-Db2 and decreased expression of Hmgcs2, Olfml1, Ndrg2, LOC30413, and Mterfd1. Also in the aortic endothelial cells, both obesity and IST increased expression of RT1-A2, Pip4k2a, and Nod1 and decreased expression of LOC304131, LOC501110, and Slc37a4.

DISCUSSION

We evaluated the impact of exercise training on vascular gene expression profiles, using a transcriptome-wide RNA-Seq analysis in obese rats. In particular, the analysis was performed on the SFA and GFA harvested from three groups of OLETF rats: EndEx, IST, and Sed. Recognizing that the GFA and SFA analyses were performed on whole vessel homogenates, we also sought to determine the influence of exercise training on endothelial cells obtained from aortic scrapes. We found that the number of genes whose expression was altered in response to both EndEx and IST was greater in the SFA compared with the GFA. Considering the fact that exercise produces greater relative increases in blood flow to the gastrocnemius muscle compared with the soleus muscle (1), these data suggest an interesting disassociation between the magnitude of exercise-induced vascular transcriptional changes and the magnitude of exercise-induced blood flow stimulus to which the arteries are exposed. Furthermore, we found that the magnitude of transcriptional changes produced by IST in the SFA and GFA was greater than that produced by EndEx, whereas in aortic endothelial cells EndEx evoked greater transcriptional changes than IST. These results suggest that the signals produced by bouts of exercise that result in altered gene expression are not totally driven by hemodynamic forces associated with increased blood flow, suggesting that other signals produced by exercise may also be involved (19, 39).

While reductionist approaches have represented a crucial forward step toward our understanding of the influence of exercise training on vascular cells, it is important to recognize that there is a need to move toward more holistic and integrative molecular approaches (23). In this regard, the utilization of RNA-Seq analysis allowed us to examine the regulatory effects of exercise on the entire transcriptome of arteries instead of focusing on expression of a targeted set of genes. Our finding that exercise training produced marked transcriptional changes in the SFA, relative to the GFA, is certainly intriguing. If increased blood flow during exercise is a primary signal for vascular adaptations, one would expect greater changes in gene expression in the artery that is exposed to greater increases in blood flow (i.e., the GFA) and not vice versa. Indeed, vasomotor function of the GFA appears to be more amenable to exercise training than that of the SFA (5, 18, 28). In particular, previous studies from our group have shown that OLETF rats given access to voluntary running wheels exhibit improved EDD in the GFA but not the SFA (5). Earlier work in lean healthy rats also demonstrated a lack of exercise training-induced change in EDD in the SFA (18). Likewise, our recently published data from the present group of rats indicates that treadmill EndEx and IST slightly improve EDD in the GFA, whereas exercise-induced changes were not observed in the SFA (28). In the present study, we identified no alterations in gene expression unique to the GFA that would explain previously observed alterations in EDD. Taken together, our data suggest that alterations in vascular gene expression and vasomotor function induced by exercise may not always be associated. That is, the GFA is more amenable to exercise-induced changes in EDD and less amenable to exercise-induced changes in vascular gene expression, whereas the opposite is true for the SFA. Perhaps at the transcriptional level, the SFA exhibits greater responsiveness to small changes in blood flow stimulus. Alternatively, it is possible to speculate that the SFA is more sensitive than the GFA to other exercise-induced signals, beyond shear stress forces, including changes in circulating factors (19). In support of this idea, in our companion paper (21) in which we evaluated the impact of hyperphagia-induced obesity on transcriptional profiles in SFA vs. GFA, we found that obesity produced greater changes in gene expression in the SFA (415 genes) compared with the GFA (240 genes), again suggesting that, at the transcriptional level, the SFA is a more responsive artery. Perhaps this heightened plasticity of the SFA reflected in its capacity to modulate gene expression in response to different physiological stimuli including exercise training and/or obesity may contribute to the inherent resistance of the SFA to developing vascular dysfunction (5, 20). On the other hand, it is also possible that exercise-induced changes in EDD are not mediated at the level of transcription.

Another important finding of the present study is that IST appears to produce greater transcriptional changes than EndEx in both SFA and GFA. Of note, the number of genes whose expression was altered with IST in the GFA (167 genes; Fig. 4) was ∼80-fold greater than the number of genes whose expression was altered with EndEx (2 genes; Fig. 1). In comparison, the number of genes whose expression was altered with IST in the SFA (629 genes; Fig. 4) was ∼10-fold greater than the number of genes altered with EndEx (59 genes; Fig. 1). Thus, while it is clear that in absolute terms the SFA is a more responsive artery relative to the number of genes with altered transcripts, the GFA is particularly more responsive to IST than to EndEx. This observation may be explained by the fact that, compared with endurance training, sprinting exercise is associated with a greater recruitment of the white portions (i.e., fast-twitch, low oxidative muscle fibers) of the gastrocnemius muscle, thus presenting a greater blood flow stimulus to the GFA artery wall. Alternatively, it is possible that the intermittent/cyclic nature of the flow stimulus is more important for exercise-mediated alterations in gene expression.

Also intriguing is the observation that IST evoked greater transcriptional effects than EndEx in skeletal muscle feed arteries but not in aortic endothelial cells, where the opposite was the case (i.e., EndEx produced greater effects than IST; Fig. 7). Thus it appears that aortic endothelial cells are more responsive to low-intensity, longer-duration bouts of exercise, whereas the skeletal muscle feed arteries seem more responsive to high-intensity, short-duration exercise bouts. In this regard, future research should attempt to separate and study gene expression in endothelial cells vs. smooth muscle cells and determine whether these differential effects of exercise mode are specific to a vascular cell type and/or specific to the anatomic location of the vascular tissue studied.

A number of genes whose expression was altered with exercise training may provide insights regarding underlying molecular mechanisms by which exercise affords vascular protection. For example, we found that both EndEx and IST increased expression of MANF (mesencephalic astrocyte-derived neurotrophic factor) mRNA in the SFA (Table 2 and Supplemental Table S1). The protein encoded by this gene is localized in the endoplasmic reticulum and Golgi complex. Increased expression of MANF has been shown to protect cells from endoplasmic reticulum stress-induced apoptosis (13, 14). This protection has been observed in multiple cell and tissue types, including the ischemic heart and brain, as well as in animal models of neurodegenerative disorders (13, 14). In fact, it has recently been proposed that a “MANF-based therapeutic is predicted to be of high impact” (13). Importantly, here we show that exercise training may be an effective therapeutic strategy to induce expression of MANF in vascular cells.

In addition, we found that IST increased expression of HSPA12B (heat shock 70-kDa protein 12B) mRNA in the SFA (Table 4). HSPA12B is predominantly expressed in endothelial cells and required for angiogenesis (17, 44). In a recent study, Zhou et al. (48) showed that overexpression of HSPA12B significantly attenuated cardiac dysfunction in endotoxin-septic mice compared with wild-type control mice. The mechanisms by which HSPA12B attenuated cardiac dysfunction involved the preserved activation of the PI3K/Akt signaling pathway, resulting in reduced LPS-induced expression of VCAM-1, ICAM-1, and iNOS and leukocyte infiltration into the myocardium (48). Whether exercise-induced vascular expression of HSPA12B decreases the susceptibility of artery wall to inflammation warrants future research. Also in the SFA, we found that IST increased expression of TRPV4 (transient receptor potential vanilloid subtype 4) mRNA (Table 4). TRPV4 channels are expressed in both endothelium and smooth muscle cells, are activated by numerous stimuli including shear stress, and play an important role in the regulation of vascular tone (12). Indeed, TRPV4 channel activation leads to smooth muscle cell hyperpolarization and vasodilation (12). Current evidence also indicates that vascular expression of TRPV4 is reduced in streptozotocin-induced diabetic rats, which could be an underlying cause for impaired endothelium-dependent hyperpolarization in these animals (27). Furthermore, we found that EndEx decreased expression of ALOX15 (arachidonate 15-lipoxygenase) mRNA in the SFA (Table 2). Endothelial 15-lipoxygenase produces arachidonic acid metabolites that mediate vascular relaxation by activating smooth muscle cell SK_Ca-like channels (9). Of note, vascular expression of ALOX15 is increased by cytokines, hypoxia, and hypercholesterolemia (9) and enhanced in atherosclerotic lesions of arteries from rabbits and humans (16, 47). In light of this, our observation that EndEx animals exhibited reduced vascular expression of ALOX15 suggests an atheroprotective effect of exercise.

We also found that EndEx reduced expression of ATF3 (activating transcription factor 3) mRNA in aortic endothelial cells (Table 6). This may represent another atheroprotective effect of exercise, as Aung et al. (3) recently showed that expression of ATF3 was enhanced in cultured human aortic endothelial cells exposed to triglyceride-rich lipoproteins (TGRL), which are known to contribute to increased risk of atherosclerotic cardiovascular disease. Importantly, siRNA-mediated inhibition of ATF3 in cultured endothelial cells blocked TGRL-induced inflammation. The authors also showed that ATF3−/− mice were resistant to endothelial apoptosis precipitated by treatment with TGRL (3). Our finding that endurance exercise reduces expression of ATF3 in aortic endothelial cells may be important in light of the above-mentioned evidence that ATF3 is a mediator of vascular apoptosis and inflammation.

While obesity and exercise training, in the background of obesity, consistently produced greater transcriptional changes in the SFA compared with the GFA, it is also worth noting the discrepancies in vascular gene expression profiles caused by obesity vs. exercise training. Indeed, the nature of these striking differences can be well appreciated when assessing the top-scoring gene networks identified by IPA. The SFA and GFA gene networks most affected by obesity [Figs. 3 and 2 in our companion paper (21)] appear to present limited interconnectivity between molecules and are centered on the ubiquitin family gene Ubc. In contrast, the SFA and GFA gene networks most influenced by exercise training present much greater interconnectivity and interactions among molecules, conceivably suggesting that the effects of exercise on vascular molecular biology are highly complex and multifaceted.

The observation that the gene networks affected by obesity and exercise were markedly distinctive (Supplemental Table S2) prompted us to examine the extent to which the effects of hyperphagia-induced obesity on gene expression [reported in our companion paper (21)] were reversed by exercise training. We found that exercise training was effective in partially or totally restoring expression of ∼6–15% of the genes (depending on the artery) affected by obesity (Fig. 10). This finding suggests that only a relatively small fraction of the vascular transcriptional changes produced by obesity are counteracted by the exercise training programs used here. Indeed, these results support the notion that regulation of vascular gene expression with exercise is not simply the reversal of increased body weight. In this regard, identification of vascular genes that are responsive to exercise per se vs. genes that are responsive to consequent reductions in adiposity requires further investigation. Also, the genes whose expression was altered with obesity but not reversed by exercise may provide some clues regarding markers/pathways to target through other nonexercise treatments (e.g., diet, pharmacological interventions, etc.).

It is important to note that there were no genes affected by obesity that were restored by exercise in all three arteries (i.e., SFA, GFA, aortic endothelial cells). However, there were two genes (i.e., Crem and Ifit2) in which obesity increased expression in both the SFA and the GFA and exercise training (EndEx, IST, or both) reduced expression in both arteries. In addition, there was one gene (i.e., ISG15) in which obesity increased expression in both the SFA and aortic endothelial cells and IST reduced expression in both the SFA and aortic endothelial cells. Of interest, Ifit2 (Interferon-induced protein with tetratricopeptide repeats 2) and ISG15 (Ubiquitin-like modifier) are two interferon-stimulated genes responsible for induction of cytokine production and chemotactic activity (15, 37, 41). Further research is needed to test whether exercise-mediated downregulation of Ifit2 and ISG15 is a mechanism by which exercise confers vascular protection against obesity.

As highlighted in our companion paper (21), there are a few issues that deserve attention. A number of exercise-induced signals for vascular adaptations have been proposed in the literature (19, 36). One mechanism that we discuss in this report is increased blood flow-induced shear stress; however, many other factors interact with hemodynamic forces in the regulation of vascular cell phenotype in response to exercise. In addition, exercise training in the hyperphagic OLETF rat model results in a robust reduction of body weight and improvement of a number of obesity-related comorbidities including insulin resistance, hyperlipidemia, hyperleptinemia, fatty liver disease, hypertension, etc. Therefore, the vascular effects of exercise training documented here are the result of a constellation of factors and cannot be attributable to any single event. Furthermore, findings from aortic endothelial cells should not be extrapolated to all large and small arteries. The aorta was chosen because it is the largest artery and therefore the most convenient source of arterial endothelial cells.

In summary, this is the first study to provide an RNA-Seq-based characterization of the effects of exercise on the transcriptome of arteries from obese rats. A significant finding of the study is that treadmill exercise in rats, both EndEx and IST, produced greater transcriptional changes in the SFA compared with the GFA. This finding is puzzling given the knowledge that bouts of treadmill exercise are associated with greater relative increases in blood flow to the gastrocnemius muscle compared with the soleus muscle. Another interesting finding is that, at the transcriptional level, IST had greater effect on gene expression in skeletal muscle feed arteries, whereas EndEx had greater effects on gene expression in aortic endothelial cells. Furthermore, while the vascular transcriptional effects of exercise were extensive, our data indicate that chronic exercise training restores expression of a relatively small fraction of the genes altered by obesity. Notably, the present study offers an important resource for generating new hypotheses that may shed some light into potential molecular mechanisms governing the regulation of vascular cell phenotype by exercise training. For example, future research should explore the role of novel exercise-responsive genes identified in this study, including MANF, ALOX15, HSPA12B, TRPV4, ATF3, Ifit2, and ISG15, as potential mechanisms through which exercise may exert salutary effects in the vasculature.

GRANTS

This work was supported by National Institutes of Health (NIH) Grants RO1-HL-036088 (M. H. Laughlin and J. W. Davis) and T32-AR-048523 (N. T. Jenkins and J. S. Martin) and Department of Veterans Affairs Grant VHA-CDA2 1299-02 (R. S. Rector). This work was also supported in part with resources and the use of facilities at the Harry S Truman Memorial Veterans Hospital in Columbia, MO.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: J.P., N.T.J., R.S.R., J.W.D., and M.H.L. conception and design of research; J.P., N.T.J., P.K.T., J.S.M., J.W.D., and M.H.L. performed experiments; J.P., N.T.J., and J.W.D. analyzed data; J.P., N.T.J., R.S.R., J.W.D., and M.H.L. interpreted results of experiments; J.P. and J.W.D. prepared figures; J.P. drafted manuscript; J.P., N.T.J., P.K.T., J.S.M., R.S.R., J.W.D., and M.H.L. edited and revised manuscript; J.P., N.T.J., P.K.T., J.S.M., R.S.R., J.W.D., and M.H.L. approved final version of manuscript.