Mechanisms of cardioprotection resulting from Brown Norway chromosome 16 substitution in the salt-sensitive Dahl rat

Abstract

The SS-16^BN/Mcwi consomic rat was produced by the introgression of chromosome 16 from the Brown Norway (BN/NHsdMcwi) rat onto the genetic background of the Dahl salt-sensitive (SS/Mcwi) rat by marker-assisted breeding. We have previously shown that the normotensive SS-16^BN/Mcwi consomic strain is better protected from developing left ventricular dysfunction and fibrosis with aging than the hypertensive SS/Mcwi parental strain; however, the mechanism of this protection was not clear since the SS-16^BN/Mcwi had both lowered blood pressure and an altered genetic background compared with SS/Mcwi. Microarray analysis of SS-16^BN/Mcwi and SS/Mcwi left ventricle tissue and subsequent protein pathway analysis were used to identify alterations in gene expression in signaling pathways involved with the observed cardioprotection on the SS background. The SS-16^BN/Mcwi rats exhibited much higher mRNA levels of expression of transcription factor JunD, a gene found on chromosome 16. Additionally, high levels of differential gene expression were found in pathways involved with angiogenesis, oxidative stress, and growth factor signaling. We tested the physiological relevance of these pathways by experimentally determining the responsiveness of neonatal cardiomyocytes to factors from identified pathways and found that cells isolated from SS-16^BN/Mcwi rats had a greater growth response to epidermal growth factor and endothelin-1 than those from parental SS/Mcwi. We also demonstrate that the SS-16^BN/Mcwi is better protected from developing fibrosis with surgically elevated afterload than other normotensive strains, indicating that gene-gene interactions resulting from BN chromosomal substitution confer specific cardioprotection. When combined with our previous findings, these data suggest that that SS-16^BN/Mcwi may have an increased angiogenic potential and better protection from oxidative stress than the parental SS/Mcwi strain. Additionally, the early transient idiopathic left ventricular hypertrophy in the SS-16^BN/Mcwi may be related to altered myocyte sensitivity to growth factors.

pathological left ventricular hypertrophy (LVH) is a very strong predictor of abnormal cardiovascular events and heart failure (14). In LVH an increase in chronic left ventricle stress triggers the growth of excessive cardiac tissue that increases myocardial strain, increases metabolic demand, and leads to diastolic dysfunction. Interstitial fibrosis and fibroblast proliferation also frequently accompany LVH as the disease state progresses to heart failure (11). Despite decades of cardiovascular research focused on uncovering molecular mechanisms that underlie the development of pathological cardiac hypertrophy, and related loss of function, all of the factors contributing to disease pathology remain incompletely understood (6).

We have used a consomic rat model to study the genes regulating LVH in a controlled genetic environment (1). The SS-16^BN/Mcwi consomic rat was produced by the introgression of chromosome 16 from the Brown Norway (BN/NHsdMcwi) rat onto the genetic background of the Dahl salt-sensitive (SS/Mcwi) rat by marker-assisted breeding (4). Our previous characterization of this model indicated that despite early transient idiopathic LVH, aging SS-16^BN/Mcwi rats did not experience the impaired left ventricle function or pathological tissue remodeling characteristic of the SS/Mcwi parental strain. SS-16^BN/Mcwi were also better protected from developing left ventricular fibrosis by 36 wk of age and had a higher capillary-to-fiber ratio at 18 wk of age (18).

Despite the striking phenotypic differences between the SS-16^BN/Mcwi and their parental strains, the results of our previous study were not able to determine if BN/NHsdMcwi chromosome 16 substitution in the SS/Mcwi genome provided cardioprotection directly through altered gene-gene interaction or was secondary due to the lower blood pressure in the SS-16^BN/Mcwi. In the current study, we hypothesized that the gene expression pattern resulting from the unique combination of BN chromosome 16 and SS/Mcwi genomes in the SS-16^BN/Mcwi provides cardioprotection to these animals even in the presence of increased afterload. Additionally, we aimed to identify genes that were differentially expressed in left ventricle tissue from naïve SS/Mcwi and SS-16^BN/Mcwi rats to better understand the molecular mechanisms conferring the cardioprotection in the genetically protected model.

METHODS

Animals.

Male SS/Mcwi (n = 46), SS-16^BN/Mcwi (n = 46), and SS-13^BN/Mcwi (n = 24) rats were obtained from colonies at the Medical College of Wisconsin (MCW) and housed in the MCW Biomedical Resource Center on a 12 h/12 h light-dark cycle. SS-16^BN/Mcwi and SS-13^BN/Mcwi rats are normotensive consomic rats resulting from introgression of Brown Norway chromosome 16 or 13 into the SS/Mcwi background, respectively. (23) Rats were given a standard rat chow (Purina) containing 1% salt and water ad libitum.

All animal protocols were approved by the MCW Institutional Animal Care and Use Committee. For all procedures rats were anesthetized by an intramuscular injection of ketamine (100 mg/kg), xylazine (50 mg/kg), and acepromazine (2 mg/kg).

Surgical models of LVH.

The 1-kidney, 1-clip (1K1C) surgery was performed with silver clips as previously described by Brooks et al. (2) to induce cardiac hypertrophy in 9 wk old rats SS-16^BN/Mcwi and SS-13^BN/Mcwi. (n = 5/strain). Sham operation was also performed (n = 5–7/strain). The contralateral kidney was then excised after a 3-0 silk suture was tied around the renal artery and vein. Animals were studied and tissues were harvested 2 wk after surgery. The placement clips was verified post mortem.

Wall thickness and blood pressure measurements.

Transthoracic echocardiography (Vivid 7, GE) was performed on anesthetized SS-16^BN/Mcwi and SS-13^BN/Mcwi and rats prior to 1K1C surgery and at 9 wk of age. Short-axis views of left ventricles were acquired midpapillary. Wall thickness was measured from the M-mode view. Sham surgery or 1K1C surgery was completed on animals, and they were allowed to recover. Two weeks following surgery animals were anesthetized and ultrasound measurements of the left ventricle were repeated. At this time blood pressure was measured in the carotid artery using a SPR-838 Millar Mikro Tip catheter (Millar Instruments). Blood pressure analysis was completed with ADInstruments Chart software.

Tissue collection and fibrosis measurement.

The heart was excised rapidly, placed in 500 mM KCl, and weighed. Atria were removed, and the wet weight of the ventricles was taken. The ventricles were then cut in cross section midventricle, fixed in 10% formalin, mounted in wax, sliced, and rehydrated before trichrome staining was performed. Images of the slide-mounted sections were visualized with a Nikon E-400 microscope (Nikon Instruments) and acquired using a SPOT Insight digital camera (Diagnostic Instruments). Fibrotic tissue was quantified as a percent of left ventricle tissue area with Metamorph software (Molecular Devices).

RNA isolation and microarray.

Left ventricle tissue was collected from 8 wk old male rats (n = 6 SS-16^BN/Mcwi and n = 6 SS/Mcwi) following decapitation and placed immediately in ice-cold RNAlater (Qiagen). Left ventricle tissue from each animal was cut into pieces ∼1 mm³ and stored in 10 ml of RNAlater for 24 h. Tissue pieces were then isolated, snap-frozen in liquid nitrogen, and stored at −80°C.

Total RNA was extracted from the homogenized (Polytron) cardiac tissue of each heart using TRIzol (1 ml per 100 mg of tissue). Pools were created by combining equal amounts of RNA from all six animals from each strain. Each pool was then divided into two technical replicates. These samples were prepared for hybridization to the microarray chips by the Microarray Service Center at the Medical College of Wisconsin according to recommendations by Affymetrix. Briefly, double-stranded cDNA was generated from the single-stranded RNA and purified by according to the Affymetrix protocol. Biotin-labeled cRNA was then synthesized by in vitro transcription, purified, and quantified. The cRNA was hybridized to Affymetrix 230 2.0 Array microarray chips (containing probes for 31,099 genes) and analyzed with GeneChip Expression Analysis.

Direct microarray analysis.

Direct chip-to-chip expression analysis was performed between all four microarray chips. The technical replicates were compared with each other (i.e., SS/Mcwi vs. SS/Mcwi). Affymetrix GeneChip Expression Analysis software determined that there was no difference (significant if P < 0.05) in gene expression in any gene between technical replicates, ensuring that SS/Mcwi to SS-16^BN/Mcwi comparisons from all chips would be valid. Each of the SS/Mcwi chips was also compared with each of the SS-16^BN/Mcwi chips and a signal log ratio was generated for gene expression between each chip-to-chip comparison. Chip-to-chip comparisons for each gene were sorted by signal log ratio.

Pathway analysis of microarray results.

Microarray expression data was analyzed by the Significance Analysis of Microarrays (SAM) approach (32). Genes with expression differences having a delta value of 1 or less were defined as differentially expressed genes. Pathway analysis of these genes was completed by entering their expression values into Ingenuity Pathway Analysis (IPA) software (Ingenuity).

Real-time PCR analysis.

Real-time PCR was performed for JunD mRNA (forward primer 5′-CTCCTCCTCCCGACACCATC-3′; reverse primer 5′-TCGTAGCAAAACAAAACCAAACAA-3′) using SYBR Green chemistry and normalized to 18S rRNA expression, as previously described (19).

JunD sequencing.

Total genomic DNA was extracted from tissue obtained from a tail tip as described previously (27). The following primer pairs F1–5′-TCAACGTGGGTTACATCTTTTG, R1–5′-CTACTTAGCGCCCTGTCAGTTT; F2–5′-ACACTTGGGGAGATGAAAACAG R2–5′-GCGCGCACTCTTATAGCC; F3–5′-CATGACGTCAACCCACAATG, R3–5′-AACTGCTCAGGTTGGCGTAG; F4–5′-ATCTTGGGCTGCTCAAACTC, R4–5′-GGGTCCAGCTTGTCGAGTC, F5–5′-GGAGAAAGTCAAGACCCTCAAA, R5–5′-GAGCACTTCGCTCTACTCCTTC; F6–5′-GCCGGTTTTGTGTTTTCAGTA, R6–5′-AACTGGGACTTACCATGTGACC; F7–5′-CTGACCCGGAACTCAGAGA, R7–5′-AGTGAAGACACTGGTGTATGTGG were used to amplify a 4,793 kb genomic region of JunD including 1,582 kb of upstream untranslated region. The product sizes of the resulting overlapping PCR amplicons ranged from 949 to 999 bp. The unpurified PCR products were employed in sequencing reactions performed on a Bio-Rad DNA Engine Tetrad thermocycler under the following conditions: 96°C for 10 s, 50°C for 5 s, 60°C for 4 min for a total of 25 cycles. The sequencing reaction products were purified using the Millipore Montage clean-up kit and a Beckman Coulter Biomek FX. DNA sequencing was performed using an Applied Biosystems 3730xl DNA Analyzer. All fragments were sequenced from both strands. Sequencing analysis was performed as previously described (27).

Western blots.

Left ventricles were isolated from anesthetized 8 wk old male rats (n = 14 SS-16^BN/Mcwi, n = 11 SS/Mcwi) and snap-frozen in liquid nitrogen until time of homogenization. The frozen hearts were minced and then homogenized in a 1:3 volume of solubilization buffer [0.32 M sucrose, 10 mM Tris·HCl (pH 7.4), 1 mM DTT, 1 mM EGTA, 1 mM EDTA, 1 protease inhibitor tablet (Roche) per 100 ml of buffer] on ice for 30 s with a Polytron homogenizer. For SOD-2, catalase, Flt-1, and vascular endothelial growth factor (VEGF) Western blots, the whole tissue homogenate was stored at −80°C until Western blots were run. To enrich samples for nuclear protein for the JunD Western blots, whole tissue homogenate was centrifuged at 4°C and 1,000 g for 10 min. The supernatant was removed and the nuclear protein-enriched pellet was solubilized in Triton buffer [1% Triton X-100, 150 mM NaCl, 10 mM Tris·HCl (pH 7.4), 1 mM EGTA, 1 protease inhibitor tablet (Roche) per 100 ml of buffer] and centrifuged again at 4°C and 15,000 g for 30 min. The nuclear protein-enriched samples were stored at −80°C until Western blots were run.

Protein concentration was determined using the DC Assay method (Bio-Rad). All Western blots were performed with 40 μg total protein per lane with protein from a single SS-16^BN/Mcwi rat to normalize protein expression in each blot. Equivalent protein loading was first assessed by Ponceau S (Sigma) staining of the membrane prior to blotting. Primary antibodies (VEGF 1:200, BD Pharmingen 55439; Flt-1 1:1,000, Santa Cruz Biotechnology, sc-316; catalase 1:2,000, Sigma C-0979; 1:2,000 SOD-2, Santa Cruz Biotechnology, sc-18503; JunD 1:200 Santa Cruz Biotechnology, sc-74; and Nrf-2 1:200 Santa Cruz Biotechnology, sc-722) in blocking solution followed by appropriate horseradish peroxidase-conjugated secondary antibodies and SuperSignal West Dura or Femto (JunD and Nrf2 only) Chemiluminescent Substrate (Thermo Scientific: #34076 or #34094, respectively) were used for protein detection. Due to variability in expression of traditional housekeeping genes in under pathological and physiological conditions (9) membranes were subjected to Coomassie blue staining to confirm equivalent protein loading per lane, as previously described (19). Protein expression was determined from scanned films by subtracting the background integrated optical densitometry from that of the protein band specific signal (Metamorph Software, Molecular Devices).

Cardiomyocyte isolation and culture.

A total of four separate isolations of hearts from neonatal SS/Mcwi and SS-16^BN/Mcwi strains were completed. Rat pups (1–3 days old) were anesthetized on ice and decapitated before the heart was excised. The heart was placed in room-temperature PBS (without Ca²⁺ and Mg²⁺) pH 7.4 and cleared of blood. Minced tissue from two or three hearts was combined, rinsed in PBS, digested with 1.5 ml of 0.25% Trypsin (Invitrogen) for 15 min at 37°C, and then rinsed with 37.5 U/ml collagenase (Worthington Biochemical). Tissue was then digested in 1 ml collagenase by shaking while incubating at 37°C for ∼4 min. This digestion was repeated 11 times for each sample. Following each digestion the collagenase solution was removed and placed in a culture dish filled with 10% fetal bovine serum (FBS, Sigma) culture medium and incubated for 1 h. The medium was then collected, filtered through a 70 μm cell strainer, and pelleted at 100 g for 15 min at room temperature. The pellet, consisting of a mixed population of primarily cardiomyocytes and fibroblasts, was resuspended in 300 μl of 10% FBS in DMEM (Fisher, MT10013CM), and cells were seeded at a density of 50,000 cells per well on a 96-well plate.

The adherent cells were switched to a low-serum (1% FBS in DMEM) medium 24 h after seeding. After another 24 h the medium in the treated cells was supplemented with 1, 10, or 100 nM of angiotensin II (AII, Sigma), epidermal growth factor (EGF, R&D Systems), endothelin-1 (ET-1, R&D Systems), fibroblast growth factor (FGF, R&D Systems), insulin-like growth factor (IGF, R&D Systems), or 10% FBS for an additional 48 h. All treatments were completed in triplicate with each isolation. Cells were then fixed in 4% paraformaldehyde, and fluorescence immunocytochemistry was performed for detection of the cardiomyocyte marker tropomyosin [1:10 CH1-s, Developmental Studies Hybridoma Bank (DSHB), Iowa City, IA; 1;200 goat anti-mouse Alexa-488 antibody, Invitrogen] and the fibroblast marker phalloidin (1:40, AlexaFluor 568 phalloidin, Invitrogen). Cardiomyoctes were visualized and photographed at ×20 magnification using a TS-100F (Nikon Instruments) microscope and a FLEX (Diagnostic Instruments) camera. The adherent cells were consisted of a mixed population of cardiomyocytes and fibroblasts. The average cardiomyocyte area was calculated from measurements of individual myocytes (6 fields/per replicate, 3 replicates per treatment for each animal) using Metamorph software (Molecular Devices) to determine the growth response to each treatment condition.

Data analysis.

Results were expressed as the means ± SE. One-way analysis of variance was used to evaluate all data, except cardiac functional data (see Fig. 5E) and cardiomyocyte spreading with growth factor treatment (see Fig. 4C) where Student's t-tests were used to compare groups. Statistical significance was considered P < 0.05.

Fig. 5.

Cardioprotection in the SS-16BN/Mcwi with pressure overloading. A: basal systolic blood pressure (BP) was significantly lower in SS-16^BN/Mcwi (n = 5) and SS-13^BN/Mcwi (n = 5) sham operated animals than naïve SS/Mcwi (n = 7) rats. 1-Kidney 1-clip (1K1C) surgery elevated systolic BP to the level of the naïve SS/Mcwi in both consomic strains (n = 5 SS-16^BN/Mcwi and n = 4 SS-13^BN/Mcwi). B: the heart-to-body weight ratio was significantly lower in the 1K1C SS-16^BN/Mcwi (n = 5) than the naïve SS/Mcwi (n = 7) or 1K1C SS-13^BN/Mcwi (n = 5). C: 1K1C also significantly increased LV fibrotic content in SS-13^BN/Mcwi, but not SS-16^BN/Mcwi (n = 6 SS/Mcwi, n = 5 SS-16^BN/Mcwi and n = 5 SS-13^BN/Mcwi). D: representative trichrome-stained left ventricle tissues from naïve SS/Mcwi and 1K1C rats. E: LV dimension and function data of 1K1C animals (n = 5 SS-16^BN/Mcwi, n = 5 SS-13^BN/Mcwi) expressed as percent of sham control (n = 5 SS-16^BN/Mcwi and n = 4 SS-13^BN/Mcwi). The differences in diastolic left ventricle wall thickness (dLVWT), systolic left ventricle wall thickness (sLVWT), stroke volume (SV), fractional shortening (%FS), and heart rate (HR), as adaptive responses to increased afterload, were significantly different between the SS-16^BN/Mcwi and SS-13^BN/Mcwi rats despite similar systolic BPs (*P < 0.05 vs. sham). Results represent means ± SE. *P < 0.05 vs. naïve SS/Mcwi; #P < 0.05 vs. SS-16^BN/Mcwi; and ‡0.05 vs. SS-16^BN/Mcwi.

Fig. 4.

Cardiomyocyte spreading with hypertrophy inducing ET-1 and EGF. A: cultured cardiomyocytes from SS (top) and SS-16^BN/Mcwi (bottom) fixed and stained with tropomyosin. Scale bar = 200 μm. B: quantification of cardiomyocytes revealed that SS-16^BN/Mcwi (n = 4) myocytes (shown in gray) have a smaller area than those from SS/Mcwi (n = 7) rats (shown in white) in low and high serum conditions. C: EGF and ET-1 (10 nM in 1% FBS) induce significantly greater cell spreading in SS-16^BN/Mcwi (n = 4) myocytes compared with SS/Mcwi (A and C). Results represent means ± SE. *P < 0.05 vs. SS/Mcwi. #P < 0.05 vs. 1% FBS control treated cells.

RESULTS

Chip-to-chip analysis, qPCR.

Through chip-to-chip analysis the most striking example of differential expression was in transcription factor JunD, a gene that resides on chromosome 16 in rats. We found SS-16^BN/Mcwi to SS/Mcwi signal log ratios of 5.1, 3.8, 4.6, and 3.7, which suggested JunD expression in the left ventricle was ∼74 times higher in SS-16^BN/Mcwi rats compared with SS/Mcwi rats. Real-time PCR of the original RNA samples used in the microarray verified a higher expression of the JunD in the SS-16^BN rats (100 ± 5.15% in SS vs. 255.92 ± 60.24% in SS-16^BN/Mcwi, n = 6 and 5, respectively). Western blots from tissues from age-matched animals indicated there was no difference in JunD protein expression between the two strains (n = 5 SS/Mcwi and n = 6 SS-16^BN/Mcwi, data not shown). Sequencing of the SS/Mcwi and BN/NHsdMcwi JunD gene did not identify any single nucleotide polymorphisms in the coding region; however, several sequence variants were found in the putative promoter region including a two-nucleotide deletion in the SS/Mcwi sequence (See Table 1).

Table 1.

Sequence variants in noncoding 5′ JunD sequence

Distance from ATG start codon	BN Sequence	SS Variant
−1157	GTGCTCGCTCTCT	T
−1117	TCTCTCTCACACA	A
−635	GCCCAGCCCCCGA	T
−438	AGAGGGTAGAGAGA	deletion

BN, Brown Norway; SS, Dahl salt sensitive. Underlined nucleotides vary between SS/Mcwi and BN/NHsdMcwi rat strains.

Pathway analysis.

SAM analysis identified 2,515 differentially expressed genes between SS/Mcwi and SS-16^BN/Mcwi with a positive false discovery rate of 8.575% (Supplemental Table S1).¹ To identify genetically driven alterations in signaling pathways that may be involved with cardioprotection in the SS-16^BN/Mcwi the 2,515 differentially expressed genes were investigated with IPA software (Ingenuity). The cellular pathways identified with the most differentially expressed genes were ranked by −log (P value) in the IPA software (Fig. 1). Using the known phenotypic differences between the strains (capillary density, fibrosis, hypertrophy, and cardiac function) we selected the most relevant of the highly ranked pathways for further analysis. These pathways included: “hypoxic signaling in the cardiovascular system,” “VEGF signaling,” “PI3-Akt signaling,” and “NRF-2 mediated oxidative stress response.” The cellular pathways containing the most differentially expressed genes are shown in Fig. 2 ranked by −log (P value). Several of these pathways were related to cardiac hypertrophy and cardiac function. Select pathways (marked in Fig. 1) were investigated in great detail in an effort to relate differential gene expression to the cardiac phenotypes that were observed in the functional and morphological studies. Expression of select proteins within these pathways was quantified by Western blot if the transcript showed strong gene expression differences by microarray and had a known function.

Fig. 1.

Ingenuity Pathway Analysis identified pathways containing several differentially expressed genes ranked by −log (P value). The total number of genes found within each pathway are shown above each bar graph. The percentage of genes that were upregulated in the SS-16^BN/Mcwi are shown in red and percentage genes that were downregulated in the SS-16^BN/Mcwi are shown in green. *Pathways containing proteins investigated by Western blot; #pathways involving growth factors tested in cell spreading assay.

Fig. 2.

Elevated VEGF in the SS-16^BN/Mcwi. A: “VEGF signaling” pathway. The SS-16^BN/Mcwi upregulated genes are shown in red and downregulated genes are shown in green. VEGF-A and Flt-1 were both differentially expressed by microarray and selected for protein quantification. B and C: protein expression levels measured by Western blot from whole cell homogenate from 8 wk old rat left ventricles (LV). SS/Mcwi (n = 6) and SS-16^BN/Mcwi (n = 8). Representative bands from each individual Western blot are shown, along with the control sample used to normalize abundance between sample groups. Expression levels reported as percent of a single SS-16^BN/Mcwi (+ control) sample expression. Results represent ± SE. *P < 0.05 vs. SS/Mcwi.

The pathway for hypoxic signaling in the cardiovascular system contained a total of 71 genes. Upregulated genes from our datasets comprised 12.7% of genes within the pathway and the downregulated comprised another 12.7% of the genes. VEGF-A message was upregulated in the SS-16^BN/Mcwi by 1.66-fold compared with SS/Mcwi. The VEGF signaling pathway (Fig. 2A) was also analyzed. In addition to VEGF-A, the VEGF receptor Flt-1 message was upregulated in the SS-16^BN/Mcwi by 1.61-fold. Expression of these two proteins measured by Western blot (Fig. 2, B and C, respectively) showed upregulation for VEGF and no change for Flt-1.

The NRF-2 oxidative stress response pathway (Fig. 3A) was also investigated. Of the 180 genes in the pathway 8.9% were upregulated and 8.9% were downregulated in the SS-16^BN/Mcwi. Notably, several antioxidant proteins had higher expression levels in the SS-16^BN/Mcwi relative to SS/Mcwi. Catalase and SOD-2 were two such antioxidant proteins exhibiting mRNA fold changes of 1.43 and 1.49, respectively. Western blot analysis indicated left ventricular catalase protein levels were higher in both the BN/NHsdMcwi and SS-16^BN/Mcwi (Fig. 3C), while we saw no differences in SOD-2 protein expression between the three strains (Fig. 3D).

Fig. 3.

Upregulation of antioxidant genes in the SS-16^BN/Mcwi. A: nuclear portion of IPA “NRF-2 oxidative stress response” pathway. The SS-16^BN/Mcwi upregulated genes are shown in red and downregulated genes are shown in green. Note the large number of antioxidant proteins with upregulated RNA in the SS-16^BN/Mcwi. B: Nrf2 protein expression was measured in whole cell homogenate enriched for nuclear proteins. Catalase (C) and SOD-2 (D) protein expression was measured in whole cell homogenate. SS/Mcwi (n = 6) and SS-16^BN/Mcwi (n = 7). Representative bands from each individual Western blot are shown, along with the control sample used to normalize abundance between sample groups. Expression levels reported as percent of a single SS-16^BN/Mcwi (+ control) sample expression. Results represent ± SE. *P < 0.05 vs. SS/Mcwi.

Cardiomyocyte hypertrophy.

Pathway analysis (Fig. 1) identified several hypertrophic growth factor regulated pathways containing many differentially expressed genes. Strain specific hypertrophic responses of neonatal cardiomyocytes were tested using hypertrophic growth factors AII, EGF, ET-1, FGF, and IGF in a cell spreading assay. Cultured SS-16^BN/Mcwi myocytes had a smaller area than those from SS/Mcwi rats in both low and high serum conditions (Fig. 4, A and B). EGF and ET-1 (10 nM in 1% FBS) induced significantly greater cell spreading in SS-16^BN/Mcwi than SS/Mcwi cardiomyocytes (Fig. 4C). There was no difference in cell area between SS-16^BN/Mcwi and SS/Mcwi cardiomyocytes treated with AII, FGF, or IGF. Strain-specific differences in cell area were also not observed in response to 1 or 100 nM concentrations of the growth factors.

Left ventricle response to elevated afterload.

The 1K1C model elevated systolic blood pressure in the SS-16^BN/Mcwi and SS-13^BN/Mcwi to the level of the naïve hypertensive SS/Mcwi rats (Fig. 5A). Heart-to-body weight ratio calculations indicate that the SS-13^BN/Mcwi rats alone had a hypertrophic response to the increased afterload, while the SS-16^BN/Mcwi did not (Fig. 5B). Interestingly, the 1K1C SS-13^BN/Mcwi rats also had significantly more fibrosis than either the SS/Mcwi or 1K1C SS-16^BN/Mcwi (Fig. 5, C and D). Cardiac dimension and functional data were obtained by left ventricular echocardiography (Fig. 5E). The adaptive changes in diastolic wall thickness, systolic wall thickness, stroke volume, fractional shortening, and heart rate that occurred with 1K1C were all significantly different between the SS-16^BN/Mcwi and SS-13^BN/Mcwi rats despite similar systolic blood pressures (*P < 0.05, t-test).

DISCUSSION

The goal of the current studies was to explore and characterize pathways involved in cardioprotection in the SS-16^BN/Mcwi at both the molecular and phenotypic levels. To achieve this we used microarray-based gene expression analysis of left ventricle tissue from naïve animals to identify cellular signaling pathways associated with the distinct phenotypic differences in the SS-16^BN/Mcwi and SS/Mcwi rats and functionally confirmed relevant hypertrophic pathways with cell spreading assays. We then examined cardiac phenotypes including fibrosis and hypertrophy in the genetically normotensive SS-16^BN/Mcwi and SS-13^BN/Mcwi rat strains exposed to a surgically elevated afterload to evaluate cardioprotection under stress in these strains. This comprehensive approach has provided insight into the mechanism by which substitution of chromosome 16 from the BN/NHsdMcwi rat results in early LVH with preserved cardiac function in the SS-16^BN/Mcwi rat. In an effort to determine the pathways responsible for the phenotypes observed in the SS-16^BN/Mcwi, including early idiopathic hypertrophy and preserved cardiac function with age, gene expression data were gathered by microarray. The mRNA expression of JunD, which is located on chromosome 16, showed a dramatic upregulation in SS-16^BN/Mcwi rats compared with SS/Mcwi; however, a difference in protein expression was not observed by Western blot. Pathway analysis of expression differences also indicated that expression of several antioxidant genes was increased in the SS-16^BN/Mcwi left ventricular tissue, including catalase, which was confirmed by Western blot. We determined that expression of an upstream regulator of antioxidant genes, nuclear factor erythroid-2 related factor (Nrf2), was also significantly higher in SS-16^BN/Mcwi than SS/Mcwi left ventricles. Oxidative stress has emerged as an important determinant of cardiac phenotypes associated with hypertrophy, heart failure, and aging. It has been shown to play a role in the development of several phenotypes associated with heart failure such as remodeling (8, 16), attenuated calcium sensitivity of the myofibers (17, 31), and reduced contractile efficiency (3, 6). This concept is supported by the observed decreases in antioxidant levels (5, 28, 30) during the transition from compensated hypertrophy to failure. A recent study by Qin et al. (24) suggests that elevated hydrogen peroxide (H₂O₂) levels play a role in this transition process. Pathway analysis of our microarray data suggests that SS-16^BN/Mcwi rats may be better able to protect themselves from oxidative stress through elevations in proteins in antioxidant proteins, including catalase and Nrf2.

Ingenuity analysis also identified several pathways involving cellular growth factors as potentially important in the phenotypic differences between SS/Mcwi and SS-16^BN/Mcwi strains. VEGF-A was found to be upregulated in the SS-16^BN/Mcwi by microarray and Western blot analysis. VEGF is an angiogenic factor that has been shown to control blood vessel growth during embryonic development (9, 33). VEGF also stimulates endothelial cells to migrate and proliferate during the growth of new blood vessels after embryonic development (28). The higher levels of VEGF-A in the SS-16^BN/Mcwi is consistent with the increased capillary-to-fiber ratio and elevated capillary density observed in the hearts of SS-16^BN/Mcwi (16) and our previous studies showing impaired VEGF-mediated angiogenesis in the SS/Mcwi rat compared with BN/NHsdMcwi (25, 26).

While the results of this study cannot directly attribute the cardioprotection observed in the SS-16^BN/Mcwi rat to higher protein expression of the chromosome 16 gene JunD, we did find higher protein expression of the functionally linked transcription factor Nrf2 (on rat chromosome 3). JunD and Nrf2 are tightly linked and can help protect the heart from pathological remodeling and fibrosis through suppression of oxidative stress and other mechanisms (13, 22). Both transcription factors have both been shown to activate antioxidant response elements within the promoter regions of antioxidant responsive genes (14, 34) to induce transcription, with Nrf2 physically interacting with JunD prior to binding (36). We observed higher expression of antioxidant-related genes, as well as VEGF-A, in the left ventricle of SS-16^BN/Mcwi, relative to the parental SS/Mcwi. VEGF-A can induce Nrf2 expression, resulting in increased antioxidant protein expression, which can stimulate additional VEGF expression via HIF-1α (20). Similarly, cultured fibroblasts lacking JunD exhibit increased levels of H₂O₂, which in turn stimulate HIF-1α induced upregulation of VEGF-A (12). Our microarray analysis revealed increased HIF-1α expression in the SS-16^BN/Mcwi, which may result from elevated oxygen demand of hypertrophic cardiac tissue rather than oxidative stress. It is also possible that other BN/NHsdMcwi alleles on chromosome 16 may be impacting the angiogenic pathway when introduced into the SS/Mcwi genomic background. The proangiogenic angiopoietin 2 (Angpt2) gene is also located on chromosome 16, as well as the fibroblast growth factor 1 (Fgfr1); however, these genes were not represented on our microarray. Fibroblast growth factor signaling has been shown to importantly mediate the expression of both VEGF-A and Angpt2 during myocardial angiogenesis (21). Interestingly, the cardiac-specific overexpression of VEGF-A or Angpt-2 results in the growth and formation of new blood vessels; however, the overexpression of both factors simultaneously has been shown to result in augmented angiogenesis (35).

To physiologically validate our gene expression pathway based analysis we experimentally tested the effect of implicated growth factors on cardiomyocyte hypertrophy in the SS/Mcwi and SS-16^BN/Mcwi to determine if they may have a role in the transient idiopathic hypertrophy previously observed (18). Isolated SS-16^BN/Mcwi cardiomyocytes showed in increased cell area compared with SS/Mcwi with both EGF and ET-1 treatment. Temporarily high expression of these growth factors or sensitivity to them during postnatal development may account for the transient hypertrophy observed in the SS-16^BN/Mcwi around 8 wk of age. It is also important to note that SS/Mcwi have intrinsically larger cardiomyocytes when cultured in 1% or 10% FBS. In culture these cells are unloaded and isolated from any humoral signaling. These results suggest that gene expression secondary to the introgression of BN/NHsdMcwi chromosome 16 into the SS/Mcwi background appears to alter cellular mechanisms that regulate growth in cardiomyocytes.

We also examined cardiac phenotypes including fibrosis and hypertrophy in the genetically normotensive SS-16^BN/Mcwi and SS-13^BN/Mcwi rat strains exposed to a surgically elevated afterload. The goal of this experiment was to determine if the both of SS/Mcwi derived strains would demonstrate similar pathology in response to afterload similar to that of the SS/Mcwi or if the SS-16^BN/Mcwi rats would be uniquely protected because of their genetic background. A similar response in the two consomic strains with intrinsically lower blood pressures would suggest that the age-related cardioprotection observed in the SS-16^BN/Mcwi (18) resulted from chronically lower blood pressure. Elevation in left ventricular afterload induced by 1K1C hypertension resulted in the development of significant hypertrophy and interstitial fibrosis in the SS-13^BN/Mcwi rat; however, the SS-16^BN/Mcwi rat was protected despite a similarly elevated blood pressure. Data obtained by left ventricular echocardiography showed dramatically different phenotypic responses to the elevated afterload between the SS-16^BN/Mcwi and the SS-13^BN/Mcwi. The 1K1C and sham SS-16^BN/Mcwi rats exhibited no significant difference in these phenotypes, while the 1K1C SS-13^BN/Mcwi rats experienced characteristic afterload-induced left ventricular changes, including hypertrophy of the myocardium, reduced stroke volume, increased contractility, and a decline in heart rate. This study clearly shows that the protection in the SS-16^BN/Mcwi is not solely the result of lower blood pressure relative to the SS/Mcwi.

Our comprehensive analysis of cardiac phenotypes, function, stress adaptation, and gene expression suggests that the SS-16^BN/Mcwi cardioprotection is the result of modulated gene expression in signaling pathways that ultimately reduce fibrosis and oxidative stress. The BN/NHsdMcwi alleles from chromosome 16 that mediate this response could not be identified in the current study, as the elevation of chromosome 16 gene JunD mRNA expression was not preserved at the protein level. However, the differential expression of VEGF-A in the SS-16^BN/Mcwi may provide a beneficial enhancement of angiogenic capability to the left ventricle.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-66579, HL-29598, and N01-HV-28182.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

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