Vascular dysfunction precedes hypertension associated with a blood pressure locus on rat chromosome 12

Abstract

We previously isolated a 6.1-Mb region of SS/Mcwi (Dahl salt-sensitive) rat chromosome 12 (13.4–19.5 Mb) that significantly elevated blood pressure (BP) (Δ+34 mmHg, P < 0.001) compared with the SS-12^BN consomic control. In the present study, we examined the role of vascular dysfunction and remodeling in hypertension risk associated with the 6.1-Mb (13.4–19.5 Mb) locus on rat chromosome 12 by reducing dietary salt, which lowered BP levels so that there were no substantial differences in BP between strains. Consequently, any observed differences in the vasculature were considered BP-independent. We also reduced the candidate region from 6.1 Mb with 133 genes to 2 Mb with 23 genes by congenic mapping. Both the 2 Mb and 6.1 Mb congenic intervals were associated with hypercontractility and decreased elasticity of resistance vasculature prior to elevations of BP, suggesting that the vascular remodeling and dysfunction likely contribute to the pathogenesis of hypertension in these congenic models. Of the 23 genes within the narrowed congenic interval, 12 were differentially expressed between the resistance vasculature of the 2 Mb congenic and SS-12^BN consomic strains. Among these, Grifin was consistently upregulated 2.7 ± 0.6-fold (P < 0.05) and 2.0 ± 0.3-fold (P < 0.01), and Chst12 was consistently downregulated −2.8 ± 0.3-fold (P < 0.01) and −4.4 ± 0.4-fold (P < 0.00001) in the 2 Mb congenic compared with SS-12^BN consomic under normotensive and hypertensive conditions, respectively. A syntenic region on human chromosome 7 has also been associated with BP regulation, suggesting that identification of the genetic mechanism(s) underlying cardiovascular phenotypes in this congenic strain will likely be translated to a better understanding of human hypertension.

vascular health is a strong predictor of hypertension (5, 21, 27, 36) and is highly heritable (24–26, 30), but the mechanisms contributing to alterations in vascular reactivity and elasticity that cause hypertension remain largely undefined (19, 29). One method of localizing hypertension risk genes is by consomic and congenic mapping (i.e., introgression of chromosomal regions from one genetic background onto another). Although our group typically focuses on mapping blood pressure (BP) and renal phenotypes, it is increasingly apparent that a single hypertension locus can impact multiple organ systems (14). As such, we have now taken to systematically examining additional organ systems (e.g., the vasculature) that might also impact the overall hypertension risk at a particular locus. We recently generated several rat chromosome 12 congenic strains where segments of the SS chromosome 12 were transferred back onto the SS-12^BN consomic background (13). In one such locus, we identified a 6.1 Mb region of SS/Mcwi (Dahl salt-sensitive) rat chromosome 12 (13.4–19.5 Mb) that severely increased risk of hypertension, marked by significantly elevated mean arterial pressure (MAP) (Δ+34 mmHg, P < 0.001) compared with the MAP of the SS-12^BN control consomic strain (144 ± 6 mmHg) on 8% NaCl diet (13). Here, our primary goal was to determine whether changes to the vasculature contribute to the hypertension risk associated with the 6.1-Mb (13.4–19.5 Mb) locus on chromosome 12. We specifically examined mesenteric small resistance arteries because small artery elasticity has been shown to be a stronger independent predictor of hypertension than large artery elasticity (36). Second, we narrowed the candidate region from 6.1 Mb (chr12:13.4–19.5 Mb) and 133 genes to 2 Mb (chr12:13.4–15.4 Mb) and 23 genes by congenic mapping. Collectively, we identified 12 differentially expressed genes and 2 (Grifin and Chst12) that were differentially expressed on both low- and high-salt diets that could potentially be associated with decreased elasticity and enhanced sensitivity to vasoconstrictors prior to significant elevations in BP.

MATERIALS AND METHODS

Generation of SS-12^BN congenic rats.

Rats were provided food and water ad libitum and were housed at the Medical College of Wisconsin (MCW) Biomedical Resource Center. Protocols were approved by the MCW Institutional Animal Care and Use Committee (IACUC). Line Ca [SS.BN-(D12Hmgc3-AU047911)/Mcwi; Rat Genome Database (RGD) ID 7248453] was generated by 1) crossing the line C [SS.BN-(D12Hmgc3-D12Hmgc6)/Mcwi; RGD ID 5683890] (13) and SS-12^BN/Mcwi strains; 2) intercrossing the F1 generation; 3) screening the F2 generation for recombinations by marker-assisted selection as described previously (31); 4) backcrossing any animals with recombinations in the candidate region to SS-12^BN/Mcwi; and 5) intercrossing animals with recombinations in the candidate region. Line C contains a 6.1-Mb congenic interval of SS rat chromosome 12 (13.4–19.5 Mb) that was introgressed back into the SS-12^BN consomic background. Line Ca contains a 2-Mb congenic interval of SS rat chromosome 12 (13.4–15.4 Mb) that was introgressed back into the SS-12^BN consomic background.

BP measurements.

Male rats were fed a 0.3% NaCl diet (7034 Teklad, Harlan, Indianapolis, IN). MAP was measured by implanting a telemetry transmitter with a catheter in the abdominal aorta of 9- to 10-wk-old rats, as described previously (13). MAP was recorded for 3 consecutive days and the reported MAP values were averages of the measurements in 3-h intervals (e.g., 9 AM–12 PM).

Examination of vascular reactivity.

Third-order mesenteric arteries were isolated, cleaned of fat and connective tissue, and hung by tungsten wires on a wire myograph (Danish Myo Technology, Aarhus, Denmark), as previously described (46). Force measurements were recorded with the PowerLab data-acquisition system (ADInstruments, Colorado Springs, CO). Mesenteric arteries were bathed at 37°C and 95% O₂-5% CO₂ in a bicarbonate-buffered physiological salt solution (PSS) [119 mM NaCl, 24 mM NaHCO₃, 5.5 mM dextrose, 4.7 mM KCl, 1.6 mM CaCl₂·2H₂O, and 1.17 mM MgSO₄·7H₂O in ddH₂O, pH 7.4]. A passive tension of 2 mN was applied to each vessel and the vessels were allowed to equilibrate for 30 min. Vessel integrity was gauged with a challenge of 10 μM phenylephrine (PE), followed by 10 μM acetylcholine (ACh). Arteries that failed to contract more than 4 mN past baseline and/or relax more than 40% from maximal contraction were excluded from further analysis (23). Following the integrity challenge, the myograph baths were washed with PSS three times, and the baseline level was reestablished. Arteries were preconstricted with 10 μM PE, and once the contraction level had plateaued, concentration-response curves were performed for endothelium-dependent (ACh, 10⁻¹⁰–10⁻⁶ M) and endothelium-independent (sodium nitroprusside, 10⁻⁹–10⁻⁶ M) vasodilators. Concentration-response curves were also completed with the vasoconstrictors PE (10⁻⁸–10⁻⁵ M) and serotonin (5-HT, 10⁻⁹–10⁻⁶ M), with thorough washout of the agonists between curves.

Determination of passive arterial wall mechanics.

Third-order mesenteric arteries were used to assess passive wall mechanics, as described previously (37). Briefly, arteries were bathed in Ca²⁺-free PSS and intraluminal pressure was changed in 20-mmHg increments from 20 to 160 mmHg. The inner and outer diameters were determined at each pressure. Strain values (D/Dref) were calculated by dividing the inner (lumen) diameters at each measurement by the inner diameter at 100 mmHg.

Vessel wall thickness was calculated as described previously (37):

$$\text{WT}=(\text{OD}-\text{ID})/2$$ (1)

where WT is wall thickness (μm) and ID and OD represent the mesenteric artery's inner and outer wall diameters, respectively (μm).

For the calculation of circumferential stress, intraluminal pressure (P) was converted from millimeters Hg to kilopascals, where 1 mmHg = 0.1333 kPa. Circumferential stress (σ) was calculated as described previously (37):

$$\sigma =\frac{\text{P}\cdot \text{ID}}{2\text{WT}}$$ (2)

Raw stress-strain data were fit to a smoothing spline, f, in MATLAB R2012B (MathWorks, Natick, MA) where:

$$\sigma =f\left(\frac{\text{D}}{{\text{D}}_{\text{ref}}}\right)$$ (3)

The elastic modulus, Y, was calculated as the first derivative of equation (3):

$$\text{Y}=f\prime \left(\frac{\text{D}}{{\text{D}}_{\text{ref}}}\right)$$ (4)

Because the measurements from 100 to 160 mmHg are the BP ranges for our rat models (13) and are physiologically relevant, the elastic modulus was calculated from the derivative of the spline at each of these measurements (100, 120, 140, and 160 mmHg). The elastic moduli at these pressures were averaged to determine the mean elastic modulus for each animal, similar to previous reports (22).

Histological analysis.

Superior mesenteric arteries were collected from 9- to 10-wk-old SS-12^BN and Ca rats maintained on 0.3% NaCl diets, fixed in 10% buffered formalin, sectioned at 4-μm thickness, and costained with Verhoeff's elastic and Masson's trichrome stains (33). Slides were digitized with a Hamamatsu NanoZoomer. Images of entire mesenteric arteries were taken at 10× with NDP View software (Hamamatsu Photonics, Hamamatsu City, Japan) and analyzed with MetaMorph (Molecular Devices, Sunnyvale, CA). Percent areas of collagen and elastin in the tunica media were quantified by encircling the tunica media in MetaMorph and selecting for blue and black areas, respectively.

RT-qPCR.

Mesenteric vessels were collected from 9- to 10-wk-old SS-12^BN and Ca rats maintained on a 0.3% NaCl diet and from 14- to 15-wk-old rats that were switched to an 8% NaCl diet (AIN-76, Dyets, Bethlehem, PA) for 3 wk. RNA was extracted from the mesenteric vessels with the RNeasy Fibrous Tissue Kit (Qiagen, Germantown, MD) according to the manufacturer's protocol. cDNA was synthesized and RT-qPCR was performed as previously described (20). Data were normalized to GAPDH and relative mRNA expression was determined by the ΔΔCt method (40). Primers were designed and validated as previously described (20). Primer sequences are listed in Table S1, available with the online version of this article.

Statistical analysis.

Statistical analyses for the BP, vascular reactivity, and mechanics data were performed using SigmaPlot 12.0 software. All data are presented as means ± standard error of the mean (SE). MAP and systolic BP data were analyzed by 2-way repeated-measures ANOVA followed by the Holm-Sidak multiple comparison test vs. the control (SS-12^BN) group. Contractile forces in response to 10 μM PE were analyzed by 1-way ANOVA. Percent relaxation data in response to 10 μM ACh were analyzed by 1-way ANOVA on ranks. Curve analyses and calculations of EC₅₀ values for the vascular reactivity data were completed with Prism computing software (GraphPad Software, LaJolla, CA). EC₅₀ values of the vascular reactivity data were analyzed by 1-way ANOVA followed by the Holm-Sidak multiple comparison test vs. the SS-12^BN control group. Since the mean elastic modulus values were not normally distributed, the data were log-transformed prior to analysis by 1-way ANOVA followed by the Holm-Sidak multiple comparison test vs. the SS-12^BN control group. Lumen diameters and wall thickness-to-lumen ratios were analyzed by 2-way repeated-measures ANOVA on ranks followed by the Holm-Sidak multiple comparison test. Gene expression and histology data were analyzed by Student's t-test.

RESULTS

Blood pressure.

We recently identified a 6.1-Mb region of SS chromosome 12 (13.4–19.5 Mb) in the line C congenic interval that significantly increased BP on the SS-12^BN rat background (13). One shortcoming of that study was that BP was higher on both diets (1% and 8% NaCl) that were tested, preventing us from identifying whether impairment of the resistance vasculature (structure/function) preceded BP elevation. As such, we tested whether a lower-salt diet (0.3% NaCl) had any effects on BP in SS-12^BN, line C, or line Ca, a smaller congenic (chr12:13.4–15.4 Mb; Fig. 1).

Fig. 1.

Schematic representation of the two salt-sensitive (SS)-12^BN congenic strains that were generated by introgressing segments of the SS chromosome 12 (black) into the genetic background of the SS-12^BN consomic rat (white) by marker-assisted selection. Line Ca is a smaller congenic derived from line C. Mean arterial pressures from 9 AM to 12 PM while on 0.3% NaCl are shown. The sample size for each group is shown. The thin black bars represent chromosomal regions that could be either Brown Norway (BN) or SS. *P < 0.05 vs. SS-12^BN. All other chromosomes are SS. SSLP, simple sequence length polymorphism.

Previously, on 1% NaCl, line C (146 ± 6 mmHg, n = 11, P < 0.001) had significantly increased MAP compared with SS (121 ± 3 mmHg; n = 12) and SS-12^BN (127 ± 1 mmHg; n = 10) (13). Additionally, on 8% NaCl, line C (178 ± 7 mmHg; n = 11; P < 0.001) had significantly elevated MAP compared with SS (137 ± 5 mmHg; n = 12) and SS-12^BN (144 ± 6 mmHg; n = 10) (13). After lowering the salt intake in the present study, the MAP of line C (105 ± 2 mmHg; n = 9; P < 0.05) and line Ca (106 ± 2 mmHg; n = 8; P < 0.05) from 9 AM to 12 PM was slightly, albeit significantly, lower on the 0.3% NaCl diet compared with the MAP of SS-12^BN from 9 AM to 12 PM (111 ± 1 mmHg; n = 11) (Fig. 1). Because MAP was not elevated in line C and line Ca compared with SS-12^BN on the 0.3% NaCl diet at any time during the day or night (Fig. 2), we postulated that any observed differences in vascular reactivity between these strains would be independent of BP elevation. Unfortunately, historical MAP data for SS and BN measured via telemetry on 0.3% NaCl diets were not available.

Fig. 2.

Mean arterial pressures averaged over 3-h intervals for 3 consecutive days. Data are presented as means ± SE. Data were analyzed by 2-way repeated-measures ANOVA followed by the Holm-Sidak multiple comparison test vs. the control SS-12^BN group. n = 11 SS-12^BN, n = 9 C, and n = 8 Ca. *P < 0.05 vs. SS-12^BN.

Vascular reactivity responses of resistance arteries.

Since enhanced vasoconstriction and impaired vasodilation increase vascular resistance and consequently elevate BP (3), we determined whether there were any differences in vasoconstrictor and vasodilator responses between lines C, Ca, and SS-12^BN. Contractions to the initial PE challenge were not significantly different between line C (12.88 ± 1.61 mN; n = 14), line Ca (17.84 ± 2.26 mN; n = 10), and SS-12^BN (13.42 ± 1.63 mN; n = 15), suggesting that there were no differences in maximal contractile force between lines C, Ca, and SS-12^BN. In addition, there were no differences in percent relaxation in response to 10 μM ACh between line C (69.82 ± 6.15%; n = 14), line Ca (65.48 ± 10.48%; n = 9), and SS-12^BN (67.72 ± 6.09%; n = 15), demonstrating that the endothelium was intact in all isolated vessels. However, lines C (logEC₅₀ value of −6.38 ± 0.08 M; P < 0.05) and Ca (logEC₅₀ value of −6.58 ± 0.03 M; P < 0.05) mesenteric arteries were significantly more sensitive to PE compared with SS-12^BN mesenteric arteries (logEC₅₀ value of −6.18 ± 0.02 M) (Fig. 3A). Similarly, lines C (logEC₅₀ value of −7.82 ± 0.08 M; P < 0.05) and Ca (logEC₅₀ value of −7.64 ± 0.02 M; P < 0.05) mesenteric arteries were significantly more sensitive to 5-HT compared with SS-12^BN mesenteric arteries (logEC₅₀ value of −7.34 ± 0.03 M) (Fig. 3B). However, there were no differences in response to the endothelium-dependent vasodilator ACh in line C (logEC₅₀ value of −8.66 ± 1.03 M) and line Ca (logEC₅₀ value of −7.91 ± 0.04 M) mesenteric arteries compared with SS-12^BN mesenteric arteries (logEC₅₀ value of −8.20 ± 0.04 M) (Fig. 3C). There were also no differences in the response of line C (logEC₅₀ values of −8.52 ± 0.34 M) and line Ca (logEC₅₀ value of −7.60 ± 0.04 M) mesenteric arteries to the endothelium-independent vasodilator sodium nitroprusside, compared with SS-12^BN mesenteric arteries (logEC₅₀ value of −7.98 ± 0.19 M) (Fig. 3D). Collectively, these data show that while line C and Ca mesenteric arteries have increased sensitivity to the vasoconstrictors PE and 5-HT compared with SS-12^BN, they have no differences in response to the endothelium-dependent and endothelium-independent vasodilators, ACh and sodium nitroprusside, respectively.

Fig. 3.

Responses of SS-12^BN, line C, and line Ca mesenteric arteries to phenylephrine (A), serotonin (B), acetylcholine (C), and sodium nitroprusside (D). Data are presented as means ± SE and as a percentage of the maximal contraction to phenylephrine. n = 7–15 SS-12^BN, n = 9–14 C, and n = 6–10 Ca at each measurement, except n = 3 C at the −10 and −9.5 acetylcholine measurements. *P < 0.05 vs. SS-12^BN for EC₅₀ values.

Passive wall mechanics.

To determine whether there were any inherent structural or passive mechanical differences between SS-12^BN, line C, and line Ca mesenteric arteries, we examined passive stress-strain relationships in Ca²⁺-free PSS. There was a left shift in the passive stress-strain curves of line C and line Ca arteries compared with the SS-12^BN curve (Fig. 4A). The mean elastic modulus values for line C (1,730 ± 303 kPa; P < 0.05) and Ca (1,574 ± 193 kPa; P < 0.05) mesenteric arteries were significantly elevated compared with SS-12^BN (942 ± 109 kPa) mesenteric arteries (Fig. 4B). These results suggest that line C and Ca resistance arteries are intrinsically stiffer than SS-12^BN arteries, which may play a role in the elevated BP levels observed in lines C and Ca when on high salt diets.

Fig. 4.

Passive circumferential wall stress vs. strain relationships of SS-12^BN, line C, and line Ca mesenteric arteries (A). Strain (D/Dref) was calculated by dividing the inner diameters at each measurement by the inner diameter at 100 mmHg. Mean elastic modulus values (B) were determined by averaging the derivatives at 100, 120, 140, and 160 mmHg. The data were log-transformed and analyzed by 1-way ANOVA followed by the Holm-Sidak multiple comparison test vs. the control (SS-12^BN) group. Data are presented as means ± SE. n = 7 SS-12^BN, n = 5 C, and n = 12 Ca. *P < 0.05 vs. SS-12^BN.

We also investigated whether there were any differences in wall thickness-to-lumen diameter ratios, which can lead to enhanced vasoconstriction (15). From 20 to 80 mmHg, there were no significant differences in the wall thickness-to-lumen ratios between lines C, Ca, and SS-12^BN (Table 1). At 120 mmHg, lines C (0.070 ± 0.003; P < 0.01) and Ca (0.080 ± 0.003; P < 0.05) mesenteric arteries had significantly lower wall thickness-to-lumen ratios compared with those of SS-12^BN rats (0.087 ± 0.004). Line C arteries also had significantly lower wall thickness-to-lumen ratios at 100 (0.076 ± 0.002 vs. 0.094 ± 0.005; P < 0.01), 140 (0.065 ± 0.003 vs. 0.078 ± 0.004; P < 0.05), and 160 mmHg (0.063 ± 0.003 vs. 0.076 ± 0.004; P < 0.05) compared with SS-12^BN arteries (Table 1), suggesting that the wall thickness, which often increases in response to elevated BP (15), does not contribute to enhanced vasoconstriction and/or decreased elasticity observed in lines C and Ca. Finally, we observed no differences in the lumen diameters of maximally dilated arteries from line C, line Ca, and SS-12^BN rats (data not shown), suggesting that the passive diameters of the third-order mesenteric arteries are unlikely to contribute to the elevated BP levels observed in our congenic animals when exposed to high-salt diets (13). Collectively, our data indicate that the line C and Ca mesenteric arteries are stiffer and have an enhanced sensitivity to the vasoconstrictors PE and 5-HT, independent of arterial wall hypertrophy or reduced lumen diameter.

Table 1.

Mesenteric artery wall thickness-to-lumen diameter ratios at different intraluminal pressures

Pressure, mmHg	SS-12BN	C	Ca
20	0.227 ± 0.004	0.205 ± 0.008	0.210 ± 0.004
40	0.172 ± 0.006	0.156 ± 0.005	0.157 ± 0.005
60	0.138 ± 0.008	0.121 ± 0.005	0.122 ± 0.005
80	0.108 ± 0.006	0.090 ± 0.002	0.095 ± 0.004
100	0.094 ± 0.005	0.076 ± 0.002†	0.086 ± 0.003
120	0.087 ± 0.004	0.070 ± 0.003†	0.080 ± 0.003*
140	0.078 ± 0.004	0.065 ± 0.003*	0.075 ± 0.003
160	0.076 ± 0.004	0.063 ± 0.003*	0.073 ± 0.003

Data are presented as mean wall thickness-to-lumen diameter ratio ± SE; n = 7 SS-12^BN, n = 5 C and n = 12 Ca. Statistical significance was determined by 2-way repeated-measures ANOVA on ranks followed by the Holm-Sidak multiple comparison test.

^*P < 0.05,

^†P < 0.01 vs. SS-12^BN.

Histology.

To determine whether there was any histological evidence of structural differences, superior mesenteric arteries from SS-12^BN and Ca animals were collected and costained with Verhoeff's elastic and Masson's trichrome stains. While there were no differences in the relative amount of collagen in the tunica media between Ca and SS-12^BN (15.13 ± 1.97 vs. 14.88 ± 0.93%; P = 0.92), superior mesenteric arteries from line Ca rats had significantly less elastin (14.10 ± 1.33 vs. 35.29 ± 3.32%; P < 0.01), and consequently, significantly increased collagen-to-elastin ratios (1.11 ± 0.20 vs. 0.43 ± 0.05%; P < 0.05) compared with SS-12^BN (Fig. 5), suggesting that decreased elastin levels or elevated collagen to elastin ratios may contribute to the development of stiffer resistance arteries in line Ca.

Fig. 5.

Analysis of combined Verhoeff's elastic and Masson's trichrome staining of SS-12^BN and Ca superior mesenteric arteries. Percent area of collagen (A) and elastin (B) in the tunica media and collagen to elastin ratios (C) of superior mesenteric arteries. Data are presented as means ± SE. Statistical significance was determined by t-test. n = 3 SS-12^BN and n = 4 Ca. *P < 0.05, †P < 0.01 vs. SS-12^BN.

Gene expression.

We also examined whether any of the 23 genes within the line Ca region were differentially expressed between SS-12^BN and Ca mesenteric vessels on 0.3% NaCl diets and 3 wk of 8% NaCl diets (Table 2). Only two genes (Grifin and Chst12) were differentially expressed in the same direction on both 0.3% and 8% NaCl diet comparisons. Grifin was upregulated 2.7 ± 0.6-fold (P < 0.05) and 2.0 ± 0.3-fold (P < 0.01) in line Ca on 0.3% and 8% NaCl diets, respectively. Chst12 was downregulated −2.8 ± 0.3-fold (P < 0.01) and −4.4 ± 0.4-fold (P < 0.00001) in line Ca on 0.3% and 8% NaCl diets, respectively. Ftsj2 was differentially expressed on both 0.3% and 8% NaCl diets, but in opposite directions (Table 2). Nine genes (Sdk1, LOC100363385, Brat1, Iqce, Lfng, Elfn1, Mafk, Ints1, and Micall2) were differentially expressed on either 0.3% or 8% NaCl diets, but not both. Collectively, these data prioritize Grifin and Chst12 as positional candidate genes for vascular reactivity and elasticity but do not preclude the potential role(s) of the other differentially expressed genes.

Table 2.

Candidate gene expression in the mesenteric vessels of SS-12^BN and Ca rats on 0.3% and 8% NaCl diets

	0.3% NaCl		8% NaCl
Gene	SS-12BN	Ca	SS-12BN	Ca
Sdk1	1.00 ± 0.10	1.08 ± 0.23	1.26 ± 0.22	2.24 ± 0.38*
Card11	1.00 ± 0.57	1.01 ± 0.48	0.49 ± 0.14	0.65 ± 0.17
Gna12	1.00 ± 0.14	1.17 ± 0.07	1.27 ± 0.06	1.32 ± 0.07
LOC100363385	1.00 ± 0.08	1.45 ± 0.13*	1.51 ± 0.30	1.17 ± 0.06
Amz1	1.00 ± 0.17	1.16 ± 0.11	0.51 ± 0.12	0.51 ± 0.03
Brat1	1.00 ± 0.40	0.95 ± 0.05	0.32 ± 0.04	0.45 ± 0.03*
Iqce	1.00 ± 0.10	1.76 ± 0.14†	0.76 ± 0.08	0.83 ± 0.04
Ttyh3	1.00 ± 0.31	1.26 ± 0.32	0.76 ± 0.08	1.05 ± 0.16
Lfng	1.00 ± 0.13	1.52 ± 0.18*	0.62 ± 0.09	0.85 ± 0.09
Grifin	1.00 ± 0.18	2.65 ± 0.55*	0.31 ± 0.05	0.61 ± 0.08†
Chst12	1.00 ± 0.13	0.37 ± 0.03†	0.79 ± 0.06	0.19 ± 0.02‡
LOC679924	1.00 ± 0.17	1.27 ± 0.15	0.98 ± 0.16	0.69 ± 0.04
Eif3b	1.00 ± 0.13	1.29 ± 0.08	0.54 ± 0.03	0.49 ± 0.02
Snx8	1.00 ± 0.22	1.27 ± 0.17	0.37 ± 0.07	0.34 ± 0.04
Nudt1	1.00 ± 0.20	1.99 ± 0.40	2.41 ± 1.00	1.15 ± 0.22
Ftsj2	1.00 ± 0.08	1.24 ± 0.03*	1.59 ± 0.24	1.01 ± 0.10*
Mad1l1	1.00 ± 0.18	1.45 ± 0.17	0.90 ± 0.16	0.75 ± 0.03
Elfn1	1.00 ± 0.14	2.21 ± 0.41*	2.89 ± 1.22	2.57 ± 0.31
Psmg3	1.00 ± 0.10	1.10 ± 0.07	2.08 ± 0.47	1.32 ± 0.10
Tmem184a	1.00 ± 0.25	1.17 ± 0.18	1.21 ± 0.57	1.06 ± 0.33
Mafk	1.00 ± 0.11	1.32 ± 0.27	1.26 ± 0.17	0.69 ± 0.04*
Ints1	1.00 ± 0.15	1.03 ± 0.04	0.58 ± 0.04	0.77 ± 0.04†
Micall2	1.00 ± 0.04	1.35 ± 0.08†	1.81 ± 0.29	1.50 ± 0.17

Gene expression of 9- to 10-wk-old rats on 0.3% NaCl and 14- to 15-wk-old rats on 8% NaCl diets for 3 wk is shown. Data are presented as mean fold-expression ± SE normalized to SS-12^BN expression on 0.3% NaCl; n = 7–8 SS-12^BN and n = 9 Ca on 8% NaCl diets. Statistical significance was determined by Student's t-test between SS-12^BN and Ca on the same diet.

^*P < 0.05,

^†P < 0.01,

^‡P < 0.00001 vs. SS-12^BN.

DISCUSSION

We previously isolated a 6.1-Mb region on rat chromosome 12 (line C) with 133 genes that significantly increased BP when exposed to elevated salt diets (13). We have since reduced this candidate interval to a 2-Mb region (line Ca) with 23 genes. The goal of the present study was to determine whether there were any vascular alterations in lines C and Ca prior to elevations of BP. In this study, we lowered the salt content of the diets so that there were no substantial differences in BP between the congenic and consomic strains. Therefore, any vascular differences observed between strains would be occurring prior to any elevations in BP. Before developing hypertension, mesenteric arteries from lines C and Ca had an enhanced sensitivity to PE and 5-HT and were stiffer compared with mesenteric arteries from SS-12^BN, suggesting that the resistance vasculature is a major contributor to the development of hypertension in our congenic models. Since this congenic region overlaps with a syntenic human locus that was previously implicated in human hypertension (1, 42), discovering the causative gene(s) involved in these vascular phenotypes could provide novel insights into the pathogenesis of human hypertension as well.

Vascular dysfunction in hypertension: disease driver, responder, or both?

Vascular dysfunction is prevalent in essential hypertension (39), but its causal and temporal role in disease initiation and pathogenesis remains unclear. Historically, vascular dysfunction detected weeks after the establishment of hypertension has largely been considered a secondary response to elevated BP due to remodeling induced by increased hemodynamic stress (39). As a result of elevated BP, vascular remodeling and smooth muscle hyperplasia can lead to increased vascular stiffening and hypercontractility (15, 39). Our data do not dispute the traditional view of hypertension-induced vascular remodeling, but rather also suggest that in some cases, vascular dysfunction (Fig. 3) and remodeling (Figs. 4 and 5) precede hypertension (Fig. 1) and therefore should also be considered a “disease-driver,” not just a response to high BP. This is consistent with the BP-independent vascular dysfunction that occurs in the SS rat (6–10) and in the elastin haploinsufficient mouse (12, 35), which, along with our data, collectively suggest that several vascular-dependent genetic mechanisms can lead to essential hypertension. Moreover, similar risk profiles exist in human hypertension, whereby vascular dysfunction and remodeling are strong predictors of hypertension risk (5, 32, 36, 45). As such, identifying the genetic components that contribute to vascular dysfunction and remodeling may help to better stratify hypertensive individuals and nominate additional targets for novel antihypertensive therapies.

A potential limitation of the present study is that BP was only measured for 3 days at 9–10 wk of age. Thus it is possible that BP levels or the rate of increase in BP varied between strains at a younger age, which could contribute to the vascular differences. Additionally, while there were no differences in MAP, there could have been differences in amplitude/frequency of BP spikes or any other BP parameters (e.g., variability) which were not measured in the present study.

Candidate genes.

Our expression analysis nominated two genes, Grifin and Chst12 (Table 2), as the top candidate genes for the vascular differences because these genes were differentially expressed in the same direction under both normotensive and hypertensive conditions. Grifin is a galectin-related extracellular matrix protein that is highly expressed in the lens and may be structurally important for lens development (2, 34). Grifin is also expressed in vascular smooth muscle cells (4), but its role(s) in vascular remodeling and dysfunction have not been elucidated. Chst12 (carbohydrate chondroitin 4 sulfotransferase 12) regulates chondroitin and dermatan sulfate synthesis, which was implicated in vascular dysfunction and remodeling. Removal of chondroitin-dermatan sulfate-containing glycosaminoglycans from the arterial wall increases mesenteric vessel stiffness (17). Additionally, two human genomewide association studies in individuals of European ancestry (11, 43) have suggestively associated a single-nucleotide polymorphism (rs2969070 [G]) that is intergenic of CHST12 and GRIFIN to hypertension. Interestingly, previous whole genome sequencing analysis of overlapping rat blood pressure loci within our candidate region found an ∼86-kb region (chr12:14,541,567–14,627,442 bp) containing single nucleotide variants near Chst12 and Grifin that were unique to the hypertensive SS strain compared with the normotensive BN, Dahl salt-resistant, and Wistar-Kyoto strains, suggesting that the SS-derived variant(s) within this region may be involved in BP regulation (38). Thus our studies may provide insights into the potential role(s) of CHST12 and GRIFIN in the development of vascular-dependent human hypertension.

Several other genes were also differentially expressed in the line Ca congenic compared with the SS-12^BN consomic (Table 2) and therefore cannot be disregarded as potential candidates. Ftsj2 was differentially expressed, but in different directions, on both 0.3% and 8% NaCl diets. This gene has not been directly or indirectly implicated in hypertension. LOC100363385, Iqce, Lfng, Elfn1, and Micall2 were differentially expressed on 0.3% NaCl only, while Sdk1, Brat1, Mafk, and Ints1 were differentially expressed on 8% NaCl only, suggesting that these four genes may be differentially expressed in response to elevated BP levels and/or elevated salt intake. Of the genes that were differentially expressed on 0.3% NaCl only, Elfn1 is the next likely candidate based on previously published evidence. Elfn1 (PPP1R28) inhibits the phosphatase activity of protein phosphatase 1 (18), which has been implicated in hypertension (41). None of the remaining candidates (LOC100363385, Iqce, Lfng, Ints1, and Micall2) have been directly or indirectly implicated in hypertension. Narrowing the congenic interval and gene-targeted approaches will be required to include/exclude these candidate genes and determine which genes are causative or secondary to hypertension. In conclusion, our mapping studies strongly suggest that a 2-Mb region of rat chromosome 12 can lead to the development of hypertension through primary changes in the vasculature.

Perspectives.

In the present study, we observed alterations in the resistance vasculature on a low-salt (0.3% NaCl) diet prior to the development of the hypertension that occurs when the animals are exposed to high-salt diets. High-salt diets have also been shown to alter vascular function (16, 28, 44, 46). However, the vascular differences that we observed in the present study may not be the same as the vascular dysfunction occurring in response to high-salt diets. Future studies will be needed to determine whether our congenic lines have additional vascular alterations when exposed to high-salt diet and to identify the nature of any salt-induced changes that occur in the vasculature prior to any increase in BP.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute (NHLBI) Grant 5R01-HL-089930 and National Institute of General Medical Sciences (NIGMS) Grant 5P50-GM-094503. S. Z. Prisco is a member of the Medical Scientist Training Program at the Medical College of Wisconsin, which is partially supported by NIGMS Grant 5T32-GM-080202. M. J. Flister is supported by NHLBI Training Grant 5T32-HL-007792.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

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