Response Gene to Complement 32 Maintains Blood Pressure Homeostasis by Regulating α-Adrenergic Receptor Expression

Abstract

Rationale:

Hypertension prevalence is much higher among children and adolescents with low birth weight and greater postnatal weight gain than in individuals with normal birth weight. However, the etiology and molecular mechanisms underlying this complication remain largely unknown. Our previous studies have shown that response gene to complement 32 deficient (RGC-32−/−) mice are born significantly smaller, but grow faster than their wild type controls, which allows adult RGC32−/− mice to attain body weights similar to those of control mice.

Objective:

The objective of this study is to determine if RGC32−/− mice develop hypertension, and if so, to elucidate the underlying mechanisms.

Methods and Results:

By using a radiotelemetry system, we found that RGC-32−/− mice exhibit higher mean arterial pressure than wild-type mice (101 ± 4 mmHg vs.119 ±5 mmHg), which enabled us to use RGC-32−/− mice to study the mechanisms underlying low birth weight-related hypertension. The increased blood pressure in RGC-32−/− mice was associated with increased vascular tone and decreased distensibility of small resistance arteries. The increased vascular tone was due to an increase in the relative contribution of sympathetic versus parasympathetic activity and was linked to increased expression of angiotensin II type I (AT1R) and α1-adrenergic (AdR) receptors in arterial smooth muscles. Mechanistically, RGC-32 regulated AT1R gene transcription by interacting with Sp1 transcription factor and further blocking its binding to the AT1R promoter, leading to suppression of AT1R expression. The attenuation of AT1R led to reduction in α1-AdR expression, which was critical for the balance of sympathetic vs. parasympathetic control of vascular tone. Of importance, down-regulation of RGC-32 in arterial smooth muscles was also associated with low birth weight and hypertension in humans.

Conclusions:

Our results indicate that RGC-32 is a novel protein factor vital for maintaining blood pressure homeostasis, especially in individuals with low birth weight.

INTRODUCTION

Arterial hypertension is a risk factor for kidney, cerebrovascular, and coronary heart diseases and affects about one-third of the adult population in Western societies.¹ Blood pressure regulation is an integrated complex response involving various organ systems, including the vasculature, heart, kidneys, adrenal glands, and the central nervous system.¹ Recent studies have shown that the prevalence of hypertension and prehypertension has markedly increased among children and adolescents, highlighting the importance of identifying the mechanisms underlying the elevated blood pressure in early life. ²

Accumulating evidence shows that fetal programming plays a significant role in adult hypertension. Low birth weight (LBW) may influence the rate of weight gain during childhood, body weight in adults, and blood pressure during childhood and adulthood.³ In fact, more rapid gain in body mass index during the first six postnatal months or in the preschool years can lead to a higher systolic blood pressure in mid-childhood.² LBW-associated hypertension and cardiovascular diseases may be attributable to the narrowing of arteries.^4–6 Indeed, placental vascular dysfunction causes placental ischemia-initiated hypertension in preeclampsia.⁷ In addition, systemic vascular dysfunction plays an important role in the pathogenesis of essential hypertension.⁸ However, the underlying mechanisms by which LBW and rapid early childhood weight gain lead to the vascular changes causing eventual hypertension remain largely unknown. This is due, at least in part, to the lack of appropriate animal models.

Response gene to complement 32 (RGC-32) was first cloned from rat oligodendrocytes responding to complement activation.⁹ RGC32 is expressed in a number of organs or tissues such as placenta, kidney, brain, liver, heart, adipose tissue, vascular smooth muscle cells (VSMCs), endothelial cells (ECs), macrophages and T cells.^10–16 Functionally, RGC-32 regulates cell proliferation, differentiation, epithelial–mesenchymal transition, and fibroblast activation.^{10, 12, 16, 17} Our previous studies have shown that RGC-32 deficiency (RGC-32−/−) in mice causes fetal growth restriction by interrupting placental angiogenesis, resulting in a lower birth weight followed by accelerated weight gain during the first four weeks of life.¹⁸ Therefore, we undertook the current study to test the hypothesis that blood pressure is elevated in RGC-32−/− mice. We found that RGC-32−/− mice did exhibit higher blood pressure than wild-type mice. Therefore, RGC-32−/− mice may serve as a valuable animal model for studying the mechanisms underlying hypertension associated with LBW and rapid early childhood weight gain.

In the current study, we also found that the increased blood pressure observed in RGC-32−/− mice is associated with increased vascular tone and an increase in the relative influence of sympathetic versus parasympathetic activity on cardiovascular function. Our additional finding that RGC-32 deficiency is associated with increased expression of angiotensin II type I receptor (AT1R) and α1-adrenergic receptor (AdR) in arterial VSMCs led us to postulate a causal link between these events, i.e., that RGC-32 may normally modulate α−1-AdR levels by inhibiting AT1R expression. Our results showed that RGC-32 inhibits AT1R gene transcription by interacting with Sp1, thus blocking its binding to AT1R promoter and attenuating AT1R expression. Decreased AT1R expression, in turn, inhibits α1-AdR expression.

METHODS

The authors declare that all methods and supporting data are available within the article and its online supplementary files.

Animals.

Male RGC-32−/− mice in C57BL6 genetic background and the genotyping primers and procedures have been previously described.¹⁸ Only male mice were included in this study because sex differences are observed in the pathophysiological response to an adverse fetal environment and in fetal programming of hypertension in both mouse and human.¹⁹ Thus, female mice need to be studied in a different project. Experiments were performed on 3 to 6 month-old male C57BL6 and RGC-32−/− mice, and animals were assigned using a simple randomization method. All animal experiments were completed within 2 years. The data collection and analyses were conducted in double-blinded manner. Mice were housed in a conventional environment with a 12-hour light/12-hour dark cycle and allowed free access to water and chow diet. All animal procedures were performed according to the NIH Guide for the Care and Use of Laboratory Animals (publication no. 85–23. Revised 1985) and were approved by the Institutional Animal Care and Use Committee of the University of Georgia.

Human samples.

The study was performed by strictly following international ethical guidelines for biomedical research involving human subjects published by CIOMS and was approved by the Institutional Review Board of Shiyan Renmin Hospital, Hubei University of Medicine. Written informed consent was obtained from all participating individuals. Human umbilical cords of were collected before disposal after babies were born in Shiyan Renmin Hospital. The thoracic aorta specimens were collected during the post-mortem examination with informed consent from patients or family members.

Reagents and chemicals.

Phenylephrine (PE, P616), atropine (Atrop, A0132), prazosin (Praz, A7791), hexamethonium (Hexa, H2138), angiotensin II (Ang II, A9525), norepinephrine (NE, A0937), sodium nitroprusside (SNP, 71778) azilsartan (Azis, SML0432) and propranolol (Prop, P8688) were purchased from Sigma-Aldrich and reconstituted in their corresponding vehicle according to the manufacturer’s instructions. RSV-Sp1 plasmid was obtained from Addgene.com (plasmid No. 12098).

Blood pressure measurements.

Blood pressure in conscious, unrestrained male mice (n=6 each, wild-type (WT) and RGC-32−/−) was measured using a radiotelemetry system (Data Science International), as described previously.²⁰ To measure acute responses to various substances, Blood pressure and HR were collected continuously in 5-second intervals, and the values were averaged over an interval of 1 minute following a 10 minutes recovery after injection of each substance. The following drugs were administered as single intraperitoneal injections: atropine (Atrop, 1 mg/kg i.p.), propranolol (Prop, 1 mg/kg IP), prazosin (Praz, 1 mg/kg i.p.), phenylephrine (PE, 1 mg/kg i.p.), hexamethonium (Hexm, 20 mg/kg i.p.), and azilsartan (3.2 mg/kg i.p.), with a 48 hours washout period allowed between administration of each substance.^20,21

Vessel distensibility and vascular tone measurement.

To measure vascular distensibility, the second-order branches of the mesenteric artery (MRA) ²² from 5–8 mice of each genotype were isolated and mounted onto the glass cannulae in the chamber of Culture Myograph System (DMT). The vessel wall thickness, passive inner and outer diameters over a range of intralumenal pressures (0 to 160 mm Hg) were measured as described previously.²³ Distensibility (D20) was defined as percentage change in luminal volume per mm Hg during stepwise changes in intravascular pressure, as previously described.²⁴ Myogenic reactivity was measured over a pressure range of 10–160 mm Hg in both Ca²⁺-containing PSS (active tone) and Ca²⁺-free PSS (passive tone). Vascular tone and myogenic index (MI) were calculated as previously described.^{23, 25}

Vessel contraction and resistance measurement.

Vessel contraction in response to KCl (0.01–0.1 M) or NE (10⁻⁴-10⁻⁷ M) was measured by recording the lumen diameter change in pressurized (70 mmHg) MRA and was calculated as previously described.²² The relative vessel resistance (RR) per unit length was calculated as previously described. ²⁶

In vivo vascular reactivity assays.

Six each RGC32−/− and WT mice (3 month old) were anesthetized, and an osmotic mini-pump (Alzet model 2002), filled with either Ang II or NE, was implanted subcutaneously in the midscapular region of each mouse. The infusion rate for Ang II and NE was 0.75 mg/kg/day and 4.2 mg/kg/day, respectively. These dosage rates have been shown to induce modest systolic arterial hypertension.^{27, 28} Saline infused mice served as s controls. Indirect blood pressures were measured with a noninvasive computerized tail-cuff system (CODA-MNTR, KENT), as described previously, at baseline and following 14 days of infusion.²⁹

Histomorphometric analysis, immunohistochemistry (IHC), and immunofluorescent staining (IF).

Elastin contents of mesenteric resistance arteries (MRA) from 3 month old mice were analyzed as previously described.²³ Smooth muscle α-actin (α-SMA) expression or myosin light chain phosphorylation (pMLC) in MRA, calponin expression in kidney artery, or RGC-32 expression in human umbilical arteries or aorta specimens were examined by IHC or IF as previously described.^14,16

Whole-mount double-staining of retinal vessels.

Mouse retinas were obtained and labeled with anti-α-SMA-Cy3 (5 μg/ml, C6198, Sigma-Aldrich) and lectin-FITC (5 μg/ml, L2895, Sigma-Aldrich) antibodies as previously described.³⁰ The diameter of each vessel was measured using the Nikon Elements software. Retinal arteriole-to-venule diameter ratios were calculated and averaged for each image.

Quantitative RT-PCR.

Total RNA was extracted from arteries or kidneys of 3-month-old RGC32−/− mice and WT control littermates using TRIzol reagents (Life Technologies). Transcript levels of genes of interest in WT and RGC32−/− kidneys were analyzed by qPCR and compared after normalization to cyclophilin levels as described previously.¹² The primers used are listed in Online Table I.

Western blot.

Mouse MRA, human umbilical arteries, or rat primary aortic VSMCs were homogenized on ice in 0.1% Tween-20 homogenization buffer containing protease inhibitors. Western blot was performed as described previously.¹⁷ Primary antibodies used were: goat anti-AT1R (ab59018, Abcam); rabbit anti-α1A-AR (ab137123, Abcam) rabbit anti-α1B-AR (ab169523, Abcam), rabbit anti-α1D-AR antibody (ab84402, Abcam), α-tubulin (T9026, Sigma), and RGC-32, ¹⁷ respectively.

Determination of plasma catecholamines.

Plasma concentrations of epinephrine (EPI) and NE were determined by ELISA (CatCombi, IBL Hamburg) as described previously.²¹ To keep the sampling conditions consistent for all animals, extreme care was taken when collecting the samples. Blood samples were taken by submandibular bleeding within 10 seconds of capture of the mouse.

Promoter reporter Luciferase Assay.

Primary rat VSMCs were transduced with Ad-GFP or Ad-RGC32 and transfected with AT1R promoter construct for 48 h followed by vehicle or Ang II treatment (200 nM) for 24 h. Luciferase assay was performed using the Dual-Luciferase Reporter Assay System (Promega).³¹ Independent experiments were performed in triplicate and repeated for three times.

Co-immunoprecipitation (Co-IP) Assay.

Cells were lysed with ice-cold mild lysis buffer containing protease inhibitor mix provided in a Co-IP kit (Thermo Scientific). The lysates were incubated with IgG (negative control), Sp1, or RGC-32 antibodies. The immunoprecipitates were pelleted, washed, and subjected to immunoblotting according to the manufacturer’s instructions.

Chromatin Immunoprecipitation Assay (ChIP).

ChIP assays were performed using a ChIP kit (Millipore).³² Growth-arrested primary rat VSMCs were treated with Ang II for 24 h. Chromatin complexes were immunoprecipitated with 3 μg of Sp1 antibody or IgG (negative control) according to the manufacturer’s instructions. Semiquantitative PCR and qPCR were performed to amplify the AT1R promoter region containing GC box sequence (Sp1 binding site) using the following primer set: 5′-AGC TGA GCT TGG ATC TGG AA-3′ (forward) and 5′- GAG GTC AGC AGC TAG GCA CT-3′ (reverse).

Statistical analysis.

All values are presented as mean ± SEM. All data were evaluated for normality by D’Agostino-Pearson Omnibus test. For data with a normal distribution, statistical difference was calculated by the Student’s t-test or one-way analysis of variance (ANOVA). The Bonferroni multiple comparisons test was used after the two-way ANOVA for post hoc analysis. For data were not normally distributed, the Mann-Whitney U test was used. Two-sided P < 0.05 was considered as statistically significant. All statistical analyses were performed using Prism software (GraphPad Software Inc., InStat version 3.06).

RESULTS

RGC32−/− mice exhibited elevated blood pressure.

RGC32−/− mice had LBW compared to their WT littermates. However, at 5 weeks of age, no body weight difference was observed between the WT and RGC-32−/− mice, suggesting a greater rate of weight gain in RGC32−/− mice during the immediate postnatal period.¹⁸ Because of this difference in growth rate, we measured blood pressure at 3–6 months of age in both strains. Mean arterial pressure (MBP) was significantly greater in RGC32−/− mice compared to WT littermates (101 ± 4 mmHg vs.119 ±5 mmHg), as measured by radiotelemetry (Figure 1A-1B), which allowed the measurement of blood pressure and heart rates over a 24-hour period. The higher values for MBP and heart rates were particularly pronounced during the awake hours in RGC32−/− mice (Figure 1A-1D).

Figure 1: RGC32 deficiency caused hypertension in mice.

A-D, Mean arterial pressure (MBP, A-B) and heart rate (HR, C-D) in 3 month old wild type (WT) and RGC-32 knockout (RGC-32−/−) mice were measured simultaneously and continuously for a 24-hour period. The MBP (B) and HR (D) were calculated by averaging the 24 hour recording for all mice in each group. Shown are the means ± SEM. *P<0.05 vs. WT, n=6. E, Whole-mount double-staining of lectin (green-ECs) and α-SMA (red-VSMCs) in retinal vessels of WT and RGC-32 mice. Arrows indicate veins; arrowheads indicate α-SMA-positive arteries. Scale bar=50μm. F, The ratios of artery to vein diameters (A:V) were calculated by averaging the diameters of all retinal arteries and veins in the same microscope fields. *P<0.05 compared to the WT, n=5.

Generalized narrowing of arterioles in the retina has been found in lower birth weight children and is recognized as an early characteristic of hypertension in adulthood.^4,5,33 In the normal retina, the ratio of arteriolar diameter to venular diameter is typically 2:3, and myogenic constriction of retinal arterioles in response to the increased blood pressure results in a reduction of this ratio.³⁴ Thus, determining relative changes in the size of small vessels of the retina is an effective way to monitor the in vivo response of these vessels to high blood pressure. As shown in Figure 1E, the loss of RGC-32 was associated with a significant decrease in the retinal arteriolar/venular diameter ratio (from 2:3 to 1:2, Figure 1F). Consistent with this finding, retinal arteriolar diameters in RGC32−/− mice were reduced by 12.7%, as compared with those in WT mice. Moreover, analysis of cardiac hypertrophy, an adaptive response to a pressure-dependent increase in mechanical load of cardiomyocytes, revealed an increase in left ventricle-to-body weight (LVW/BW) ratios in RGC-32−/− mice (Online Figure I, A). Notably, although no obvious difference for LVW/BW ratios were observed between 3 and 6 month old RGC32−/− mice, the blood pressure was further increased in 6-month- vs. 3-month-old RGC32−/− mice (Online Figure I, B), suggesting that a cardioprotective mechanism might be present to prevent the left ventricle from further damage in these mice. These results indicate that RGC-32 deficiency is associated with a significant increase in systemic arterial blood pressure and the related complications, such as cardiac hypertrophy and retinal arteriolar constriction.

RGC-32 deficiency decreased vascular distensibility and increased responsiveness to vasoconstrictor stimuli.

Small arteries and arterioles are the main determinants of peripheral resistance and thus play a central role in blood pressure homeostasis.³⁵ We thus analyzed passive mechanical characteristics and active responsiveness to vasoconstrictor stimuli in MRA under defined pressures. As shown in Figure 2A-2B, the passive diameter of RGC-32−/− MRA was smaller than that of WT MRA. For each applied pressure, the passive diameter of RGC-32−/− MRA was also smaller (Figure 2B), leading to an increased wall thickness-to-lumen ratio compared to the WT controls at most pressures (Figure 2C). Moreover, compared to the WT MRA, MRA from RGC-32−/− mice showed significantly higher relative resistances and significantly lower distensibilities over the range of intraluminal pressures tested (Figure 2D-2E). Similar differences between groups were observed when MRAs were exposed to buffer containing Ca2+ (Online Figure II). Furthermore, the magnitude of spontaneous myogenic tone at a given pressure was increased in RGC-32−/− MRA, compared to the WT MRA (Figure 2F). Structural analyses revealed that elastin and VSMC contents were not altered in RGC-32−/− MRA (Online Figure III, A-D), suggesting that RGC-32 deficiency increased the vascular tone of resistance arteries independent of vascular structural change. Consistent with this interpretation, RGC-32−/− MRA exhibited increased level of myosin light chain phosphorylation (Ser19) and upregulation of the genes involved in renin-angiotensin system (RAS, Online Figure III, E-G).

Figure 2: RGC-32 deficiency decreased arterial distensibility and increased arterial vascular tone.

A-C, RGC-32 deficiency (RGC32−/−) reduced the distension of mesenteric arteries (MRA) in response to intraluminal pressure in Ca²⁺-free and EGTA-containing PSS passive condition as measured by myograph. A, Vessel images with or without 70 mmHg intraluminal pressure; Scale bar=100 μm. B-C, Quantification of the vessel lumen diameters and artery wall to lumen ratios. *P<0.05 vs wild type (WT) vessels (n=6). D, RGC32−/− MRA showed greater resistance to intraluminal pressure than the WT vessels.*P<0.05 vs. WT vessels (n=5). E, Distensibility (D20) of the MRA from 3-month-old WT and RGC32−/− mice over a range of intraluminal pressures as indicated was measured by cultured myograph system and calculated as described in Methods. *P<0.05 vs. WT MRA (n=6). F, Vessel tone of MRA at the given pressures was calculated as: [1-(active diameter/passive diameter)]*100. *P<0.05 vs. WT vessel (n=6). G-H, Dose-dependent contractile response of mesenteric arteries (MRA) to KCl or norepinephrine (NE) at 70 mmHg intraluminal pressure. *P<0.01 vs. WT MRA at the same dosage of the stimuli (n=6). KCl or NE caused greater contraction responses in RGC-32−/− MRA compared to the WT vessels.

To further characterize the effect of RGC-32 deficiency on smooth muscle contractility in MRA, we applied the depolarization agent KCl and vessel constrictor NE to the WT and RGC-32−/− MRA under 70 mmHg intraluminal pressures. As shown in Figure 2G and 2H, RGC-32−/− MRA responses to KCl or NE stimulation were greater than those of WT MRA, as shown by the increased vessel contraction. These data suggest that RGC-32−/− mice developed high blood pressure due, at least in part, to the increased responsiveness of small resistance arteries to vasoconstrictor stimuli.

RGC-32 regulated blood pressure through effects on the α-adrenergic receptor.

Sympathetic regulation of cardiac function and vascular tone is central to blood pressure homeostasis and tissue perfusion, and an increase in the relative influence of the sympathetic versus parasympathetic systems on cardiovascular function can contribute to the pathogenesis of hypertension.³⁶ Therefore, we tested whether RGC-32 deficiency causes sympathetic-parasympathetic imbalance, or dysautonomia, in conscious mice. The increase in heart rate observed following the administration of atropine in WT mice was significantly attenuated in RGC-32−/− mice (Figure 3A-3B), indicating less parasympathetic contribution to resting heart rate in the latter. Conversely, the decrease in heart rate in response to propranolol administration in RGC-32−/− mice was more than 2-fold greater than in WT mice (Figure 3C-3D), indicating exaggerated sympathetic control in RGC-32−/− mice.

Figure 3: RGC32 deficiency altered the balance of sympathetic vs. parasympathetic influences on cardiovascular function.

A-B,RGC32 deficiency (RGC32−/−) suppressed the tachycardic response to atropine (Atrop), indicating a decreased parasympathetic tone in 3 month old RGC32−/− mice as compared to the wild type (WT) mice. C-D, RGC32−/− increased the bradycardic response to propranolol (Prop), suggesting an increased sympathetic tone in RGC32−/− mice. Changes in heart rates in A-D were obtained by subtracting the resting heart rates measured prior to each treatment. E-F, The resting systolic blood pressure (SBP) was 117±6 mmHg in WT (n=6) and 144±10 mmHg in RGC32−/− mice (n=6). RGC32−/− mice exhibited a much greater reduction in SBP with a ganglion blockade by hexamethonium (Hexa) than WT mice, suggesting that RGC32−/− increased the neurogenic sympathetic control of vascular resistance. G-H, The resting SBP was 121±5 mmHg in WT (n=6) and 146±9 mmHg RGC32−/− mice (n=6). RGC32−/− mice exhibited a much greater reduction in SBP with the α1-adrenergic blockade by prazosin (Praz) compared to the WT mice, further demonstrating that RGC32−/− increased the neurogenic sympathetic control of vascular resistance. The changes of SBP were obtained by subtracting the resting SBP measured prior to each treatment. *P<0.05 compared to the corresponding WT mice in each treatment (unpaired t-test, n=6).

To further test the neurogenic control of blood pressure in RGC-32−/− mice, we inhibited ganglionic transmission with the nicotinic ACh receptor blocker, hexamethonium (Figure 3E-3F) and the α1-adrenergic blocker, prazosin (Figure 3G-3H). Both substances induced greater reductions in blood pressure in RGC-32−/− mice as compared to the WT controls, with prazosin abolishing the blood pressure increase associated with RGC-32 deficiency. The new blood pressure levels were attained within minutes, indicating that chronic volume retention was unlikely to be a major pathogenic factor in the maintenance of hypertension in RGC-32−/− mice.

Increased sympathetic nerve activity often leads to increased systemic catecholamine levels.²⁰ Therefore, we measured plasma EPI and NE levels in both WT and RGC-32−/− mice. RGC-32 deficiency was not associated with significantly altered plasma EPI or NE, compared to WT controls (Figure 4A). Since sympathetic nerves regulate vascular tone through binding of the neurotransmitter NE to α-adrenergic receptors in arterial VSMCs,^10,35 we measured the expression of both the α1- and α2-adrenergic receptors in MRA. As shown in Figure 4B, RGC-32 deficiency was associated with a significant increase in mRNA transcript levels of all α1-, but not every α2-, adrenergic receptor. RGC-32 deficiency also increased the protein expression of α1-adrenergic receptor by nearly 1.8 fold (Figure 4C-4D), suggesting that RGC-32 deficiency may enhance the sensitivity of resistance arteries to sympathetic nerve activity by increasing α1-adrenergic receptor density. Indeed, RGC32−/− mice exhibited a greater blood pressure response to PE stimulation (61.8 ± 6.6 vs. 49.2 ± 3.6 mmHg, Figure 4E-4F). Moreover, infusion of NE via osmotic mini-pump caused a larger increase in blood pressure in RGC-32−/− mice as compared to the WT mice (47.3 ± 5.6 vs. 35.3 ± 4.0 mmHg, Figure 4G) while causing narrower retinal arterioles and a lower retinal arteriolar-to-venular diameter ratio in RGC-32−/− mice than in WT mice (2:5 vs. 1:2, Online Figure IV, A-B).

Figure 4: RGC-32 deficiency increased the expression of α-adrenergic receptor in small/resistance arteries.

A, RGC-32 deficiency (RGC-32−/−, 3 month old) did not alter plasma epinephrine (E) and norepinephrine (NE) levels. B, The mRNA expression of different α-adrenergic receptors (α1A, α1B, α1D, α2, A-C) in mesenteric arteries (MRA) was measured by qPCR. *P<0.05 vs wild type (WT) mouse MRA (n=6). C, RGC-32−/− increased α1-adrenergic receptor (α1-AdR) protein expression in MRA as measured by Western blot and quantified by normalizing to the α-Tubulin level. *P<0.05 vs WT (n=6). E, Time course systolic blood pressure (SBP) changes following phenylephrine (PE, 1mg/kg, i.p.) stimulation as determined by a radiotelemetry system. *P<0.05 vs. WT mice (n=6). F, Blood pressure elevation in RGC32−/− mice in response to PE was significantly greater than in WT mice. *P<0.05 vs. WT mice (n=6). G, Osmotic pump administration of NE (4.2 mg.day⁻¹ kg⁻¹) for 14 days caused a more blood pressure increase in RGC32−/− mice than in WT mice, as measured by noninvasive CODA blood pressure monitor system, *P<0.05 vs. NE-treated WT mice (n=6).

RGC-32 regulated blood pressure through effects on the AT1R.

The renin-angiotensin-aldosterone system (RAAS) plays an important role in blood pressure regulation.^{21, 37} Although RGC-32−/− mice exhibited slightly higher plasma and urine volumes associated with higher water intake (Online Figure V, A-D), renal function of RGC-32−/− mice appeared to be normal as shown by the normal hematocrit and fractional excretion of sodium (FENa), as well as the normal levels of plasma aldosterone, Na⁺, and K⁺ (Online Figure V, E-H), suggesting that RGC-32 deficiency caused the significant increase in BP through mechanisms independent of the alteration in kidney functions. However, the vasoconstrictor response to Ang II was greater in RGC-32−/− mice, as shown by a larger increase in blood pressure than in WT mice (39.3 ± 3.1 vs. 28.3 ± 1.8 mmHg, Figure 5A). In addition, the AT1R inhibitor azilsartan caused a larger decrease in blood pressure in RGC-32−/− mice (41.1 ± 1.6 vs. 28.6 ± 2.6 mmHg, Figure 5B). Furthermore, Ang II infusion by mini-pump caused a greater blood pressure elevation in RGC-32−/− mice (44.9 ± 5.9 vs. 37.0 ± 2.0 mmHg, Figure 5C) and contributed to the further narrowing of retinal arterioles, resulting in a smaller retinal arteriolar-to-venular diameter ratio (2:5 vs. 1:2, Online Figure IV, C-D). Renin expression was increased in the kidneys and arterial VSMC while angiotensinogen (AGT) and (ACE1) mRNA levels were elevated in livers and lungs of RGC-32−/− mice, respectively (Online Figure V, I-L). These results further suggested that RGC-32 deficiency may be associated with augmentation of the RAS.

Figure 5: RGC32 deficiency increased blood pressure through angiotensin II (Ang II) receptor (AT1R)-mediated upregulation of α-adrenergic receptor.

A-B, Compared to WT, RGC-32 deficiency (RGC-32−/−) exhibited a much greater angiotensin II (Ang II, 1 ug/kg, i.p)-induced increase of systolic blood pressure (SBP) (A) or Azilsartan (1 ug/kg, i.p)-induced decrease of SBP (B). The SBP was measured by radiotelemetry. *P<0.05 vs WT mice (n=6). C, SBP increase in response to Ang-II (0.75mg.kg⁻¹ day⁻¹) stimulation by minipump infusion for 14 days. *P<0.05 vs. Ang-II-treated WT mice, n=6. D, RGC-32 deficiency increased ATR mRNA expression in mesenteric arteries (MRA) as measured by qRT-PCR and normalized to the cyclophilin level. *P<0.05 vs WT MRA for each receptor (n=5). E, RGC-32 deficiency increased ATR1 protein expression in MRA as measured by Western blot and normalized to α-tubulin level. *P<0.05 vs. WT MRA (n=5). F, Forced expression of AT1R rescued RGC-32-suppressed α-adrenergic receptor (α1A-AdR) expression in primary cultured rat VSMCs. Adenoviral vectors expressing GFP (Ad-GFP), RGC-32 (Ad-RGC32), or ATR1 (Ad-ATR1) were transduced individually or in combination into vehicle (−) or Ang II-treated VSMCs as indicated. α1-AdR protein expression was assessed by Western blot. G, α1-AdR protein expression shown in E was normalized to α-Tubulin. *^,&P<0.05 vs. Ad-GFP group; ^#P<0.05 vs. Ad-RGC32 group; ^$P<0.05 vs. Ad-GFP+Ang II group; ^@P<0.05 vs. Ad-RGC32+Ang II group (n=3).

Since RGC-32 is expressed only in VSMC of kidney resistance arteries in the normal mouse kidney, ¹² and VSMCs are the effector cells for vessel contraction, we determined if RGC-32 deficiency affects the expression of angiotensin II receptors in MRA. As shown in Figure 5D, RGC-32 deficiency increased mRNA transcript levels for both AT1R and the Ang II type 2 receptor (AT2R). However, the increase in expression for the AT1R was much more dramatic than the AT2R. Indeed, RGC-32 deficiency also caused a significant increase (2.4 fold) in AT1R protein expression in MRA (Figure 5E). These data suggest that RGC-32 may confine AT1R expression in normal resistance arteries to maintain blood pressure homeostasis.

Previous studies have shown that AT1R not only mediates the vasoconstrictor function of Ang II, but also regulates α1-AdR expression in VSMCs. ³⁸ Indeed, Ang II stimulation or overexpression of AT1R up-regulated α1-AdR expression in VSMCs (Figure 5F-5G). RGC-32 overexpression, however, inhibited Ang II-induced α1-AdR expression, which was rescued by AT1R overexpression (Figure 5F-5G). Since AT1R is negatively regulated by RGC-32 (Figure 5D-5E), RGC-32 may modulate α1-AdR levels by inhibiting AT1R expression.

RGC32 interacted with Sp1 and blocked Sp1-mediated AT1R promoter activity.

To explore the mechanism by which RGC-32 restrains AT1R expression, we tested whether RGC-32 regulates AT1R promoter activity. As shown in Figure 6A, Ang II increased AT1R promoter activity. However, forced expression of RGC-32 significantly attenuated the Ang II-activated AT1R promoter activity, suggesting that RGC-32 is involved in AT1R gene transcription. Previous studies have shown that the transcription factor Sp1 plays a critical role in Ang II-induced AT1R gene transcription in VSMCs.³⁹ Indeed, Sp1 enhanced AT1R promoter activity in both basal and Ang II-induced states (Figure 6B). However, forced expression of RGC-32 significantly attenuated Sp1- and Ang II-mediated increases in AT1R promoter activity (Figure 6B), suggesting that Sp1 activity in AT1R transcription was regulated by RGC-32.

Figure 6: RGC-32 inhibited AT1R gene transcription by interacting with Sp1 and thus blocking its binding to ATR1 promoter.

A, RGC-32 inhibited Angiotensin II (Ang II)-induced AT1R promoter activity. Primary rat VSMCs were transduced with adenovirus expressing GFP (AdGFP) or RGC-32 (AdRGC32) and transfected with AT1R promoter reporter construct for 48 h followed by vehicle or Ang II treatment (200 nM) for 24 h. Luciferase assay was performed. *P<0.01 compared with AdGFP group with vehicle treatment. ^#P<0.01 compared with AdGFP group with Ang II treatment (n=3). B, RGC32 inhibited Sp1-mediated AT1R promoter activity. Primary rat VSMCs were transduced with AdGFP or AdRGC32 and cotransfected with AT1R promoter reporter construct and control (−) or Sp1 plasmid as indicated for 48 h followed by vehicle or Ang II treatment for 24 h. Luciferase assay was performed. *P<0.01 compared with AdGFP alone group with vehicle treatment. ^#P<0.01 compared with Sp1/AdGFP group with vehicle treatment. ^@P<0.01 compared with AdGFP group with Ang II treatment. ^&P<0.01 compared with Sp1/AdGFP group with Ang II treatment (n=3). C-F, Co-IP with endogenous proteins indicated that RGC32 physically interacted with Sp1. Rat VSMCs were treated with vehicle (−) or Ang II (+, 200 nM) for 24 h. Cell lysates were immunoprecipitated with normal IgG, Sp1 (C), or RGC32 (E) antibody. The immunoprecipitates were immunoblotted (IB) with RGC32 and Sp1 antibodies (C & E). RGC32 pulled down by Sp1 (D) or Sp1 pulled down by RGC32 (F) as shown in C and E was quantified by normalizing to the Sp1 level in the input. *P<0.01 compared with IgG pulldown group. ^#P<0.01 compared with Sp1- (D) and RGC32- (F) pulldown group with vehicle treatment (n=3). The interaction between RGC32 and Sp1 was attenuated by Ang II induction. G-H, RGC32 suppressed Ang II-enhanced Sp1 binding to the AT1R promoter in a chromatin setting. Primary rat VSMCs were transduced with AdGFP or AdRGC32 followed by vehicle or Ang II treatment for 24 h. ChIP assays were performed, and the Sp1 binding to the GC box in AT1R promoter was detected by semi-quantitative PCR (G) and qPCR (H). *P<0.01 compared with vehicle-treated group; ^#P<0.01 compared with Ad-GFP group treated with Ang II (n=3).

To determine how RGC-32 regulated Sp1 function, we tested if RGC-32 physically interacts with Sp1. Coimmunoprecipitation assays (Co-IP) showed that RGC-32 had a strong interaction with Sp1 in the basal VSMCs (Figure 6C-6D) because the protein complex pulled down by Sp1 antibody contained a high level of RGC-32. However, Ang II treatment caused a 2.4 fold reduction in RGC-32 interaction with Sp1, consistent with the higher blood pressure in RGC-32−/−, compared to WT, mice. RGC-32-Sp1 interaction was confirmed by a reverse Co-IP using RGC-32 antibody to pull down Sp1 (Figure 6E-6F).

To further determine how RGC-32 regulates Sp1 activity, we tested whether RGC-32 affects Sp1 binding to AT1R promoter. Thus, we performed chromatin immunoprecipitation assay to detect the enrichment of Sp1 binding to AT1R promoter in a chromatin setting. As shown in Figure 6G-6H, Sp1 bound to the AT1R promoter, and Ang II treatment enhanced the binding by 2.3 fold. However, forced expression of RGC-32 diminished the Sp1 binding (Figure 6G-6H), demonstrating that RGC-32 inhibited AT1R gene transcription by blocking Sp1 binding to AT1R promoter. Collectively, our results suggest that RGC-32 deficiency promotes Sp1 activity in AT1R expression, leading to an enhanced AT1R level in VSMC and thus a disruption of normal vascular tone.

RGC-32 reduction was associated with LBW and hypertension in human.

To determine if RGC-32 deficiency-caused LBW and hypertension in mice is relevant to human conditions, we collected umbilical arteries (UA) of human individuals with LBW and normal birth weights (NBW). Since UA originates from internal iliac artery of the fetus, these arteries are ideal to observe the potential association between RGC-32 level and birth weight in humans. As shown in Figure 7A-7C, RGC-32 was abundantly expressed in UA media layer, especially in α-SMA-positive VSMCs from individuals with NBW, but its expression level was reduced by 10 fold in UA media VSMCs from individuals with LBW, suggesting that RGC-32 reduction is indeed associated with LBW in humans. Moreover, although RGC-32 was expressed at a low level in aortic media of human individuals with normal blood pressure, its expression was dramatically reduced in aortic media of hypertensive patients (Figure 7D-7E). These data indicated that RGC-32 deficiency is associated with both LBW and hypertension in humans.

Figure 7: RGC-32 was down-regulated in arteries of human low birth weight individuals and hypertensive patients.

A, The umbilical arteries of human individuals with normal (NBW) or low birth weight (LBW) were immunostained with α-SMA (red) and RGC-32 (green) antibodies or normal IgG (negative control). DAPI stains nuclei. B-C, RGC32 expression in the umbilical arteries were examined by Western blot and quantified by normalization to α-Tubulin. *P<0.05 vs. NBW, n=3. D-E, RGC-32 expression in thoracic aorta of healthy individuals (Normotensive, NT) and hypertensive patients (Hypertensive, HT) was detected by immunohistochemistry staining using RGC-32 antibody. The RGC-32-positive cells were counted and quantified relative to total VSMC numbers in the artery media. *P<0.01 compared to NT, n=5.

DISCUSSION

In the present study, we have identified RGC-32 as a novel factor important for blood pressure homeostasis. RGC-32 deficiency disrupts normal blood pressure homeostasis, leading to significant increases in systemic arterial pressure. Since RGC-32−/− mice exhibit LBW with an accelerated growth rate in later life to gain normal adult weight while developing hypertension, RGC-32 deficiency may be an essential factor causing LBW-related hypertension. Thus, RGC-32−/− mice may serve as a valuable model for dissecting molecular mechanisms underlying LBW-related hypertension. Indeed, we have identified a novel mechanism underlying LBW-linked hypertension; i.e., increased vascular tone due to increased expression of AT1R and α1-AdR in VSMC by RGC-32 deficiency. The elevated α1-AdR in VSMC, coupled with the exaggerated sympathetic control of the cardiovascular system, further disrupts the blood pressure homeostasis.

RGC-32 appears to inhibit α1-AdR expression in VSMCs by regulating AT1R gene transcription. RGC-32 inhibits AT1R promoter activity, which explains the increased AT1R expression observed in RGC-32−/− VSMCs. RGC-32 mitigates AT1R transcription not by directly binding to AT1R promoter, but by inhibiting the activity of key transcription factor Sp1. RGC-32 physically interacts with Sp1 to limit its binding to AT1R promoter, resulting in the controlled AT1R levels.

Interestingly, 6 month old RGC-32−/− mice exhibits an increased blood pressure with similar left ventricle weight as compared to 3 month old RGC-32−/− mice. Previous studies using an elegant intrauterine growth restriction (IUGR) rat model show that male IUGR offspring remain hypertensive in adulthood largely due to testosterone with the participation of the RAS.^40,41 The augmented RAS in male RGC-32−/− mice is consistent with the inverse relationship between birth weight and blood pressure in the IUGR rat model. ⁴⁰ The increase in blood pressure without exacerbation in left ventricle hypertrophy in 6 month vs. 3 month old RGC32−/− mice is likely due to the function of testosterone as testosterone is reported to protect the heart from hypertrophy or remodeling.^42,43 However, extensive studies would be required to clearly establish the potential diverse roles of testosterone in the hypertension and cardiac hypertrophy observed in RGC-32−/− mice.

Although the kidney is critical in blood pressure regulation, RGC-32 deficiency does not alter renal function. Our previous study shows that RGC-32 knockdown protects the kidney from ureter obstruction-induced renal fibrosis although the pathological processes as well as the mechanism underlying this protection are different from those involved in blood pressure regulation.¹² Since RGC-32 is only expressed in renal arterial VSMC under physiological conditions,¹² RGC-32 deletion may only impair VSMC function in the renal artery. Indeed, RGC-32−/− mouse renal arteries exhibit an increased expression of VSMC marker calponin (Online Figure VI), consistent with the increased vascular tone in RGC-32−/− MRA. Since both renal artery and MRA are small arteries, they may undergo similar pathological alterations when RGC-32 is deleted. Therefore, lack of RGC-32 in resistance artery VSMCs may be the major factor causing increased vascular tone and hypertension in RGC-32−/− mice. In addition, since RGC-32 deletion increases renin expression in kidney and VSMCs along with increased AGT expression in liver and ACE1 in lung, it is likely that the circulating Ang II level is elevated, which may contribute to the increased vascular tone. Moreover, renal nerve activity is known to play an important role in mediating LBW-linked hypertension,⁴⁴ RGC-32 may maintain blood pressure homeostasis also by regulating the renal sympathetic nerve activity, which may be studied in the future.

Of importance, RGC-32 deficiency is related to LBW and hypertension in humans. RGC-32 is abundantly expressed in arterial medial VSMCs of umbilical arteries of fetuses who are born with normal weights. However, RGC-32 levels are significantly reduced in arteries of fetus with LBW. These data are consistent with previous studies showing the RGC-32 is significantly down-regulated in human preeclamptic placentas compared with normal controls.^45,46 Preeclampsia is known to be a major contributor to preterm and low birth weight babies in humans.⁴⁷ Furthermore, RGC-32 is also down-regulated in human hypertensive subjects, suggesting that restoring RGC-32 expression may be a potential effective strategy for treating preeclampsia or LBW-associated complications including hypertension.

Taken together, our current findings suggest that RGC-32 deficiency disrupts blood pressure homeostasis through multiple mechanisms, specifically by augmenting the RAS, increasing vascular tone, and interrupting the balance between sympathetic and parasympathetic influences on cardiovascular function (Online Figure VII). Whether one or more of these mechanisms play a dominant role requires future investigation. Nevertheless, RGC-32 deficient mice may be a valuable model for studying LBW-related hypertension, and our results reveal a novel mechanism underlying this pathological process.

What Is Known?

• Hypertension is more prevalent among children and adolescents with low birth weight.

• Fetal programming plays a significant role in adult hypertension.

• RGC-32 deficiency in mice causes fetal growth restriction by interrupting placental angiogenesis.

What New Information Does This Article Contribute?

• RGC-32 deficiency causes low birth weight and high blood pressure in adult mice..

• RGC-32 regulates blood pressure homeostasis by altering vascular tone and distensibility of small resistance arteries.

• RGC-32 modulates α1-adrenergic receptor level by inhibiting AT1R expression.

Previous studies have shown that low birth weight influences the rate of weight gain during childhood and blood pressure during childhood and adulthood. However, the underlying mechanisms by which LBW and rapid early childhood weight gain lead to the vascular changes causing eventual hypertension remain largely unknown. This is due, in part, to the lack of appropriate animal models. Here, we use an RGC-32-deficient mouse model to identify the mechanism underlying the low birth weight-related hypertension. Our studies show that RGC-32 is a novel protein factor vital for maintaining blood pressure homeostasis, especially in individuals with low birth weight. RGC-32 regulates blood pressure by maintaining vascular tone and distensibility of small resistance arteries. RGC-32 deficiency increases vascular tone by enhancing the relative contribution of sympathetic versus parasympathetic activity and the expression of angiotensin II type I (AT1R) and α1-adrenergic (AdR) receptors in arterial smooth muscles. In the molecular level, RGC-32 regulates AT1R gene transcription by interacting with transcription factor Sp1 and further blocking its binding to AT1R promoter, leading to suppression of AT1R expression. Decrease in AT1R expression causes down-regulation of AdR, which leads to balanced control of vascular tone.

Nonstandard Abbreviations and Acronyms:

RGC-32

response gene to complement 32

AT1R

angiotensin II type I receptor

AdR

adrenergic receptors

VSMC

vascular smooth muscle cell

MRA

mesenteric resistant artery

SBP

systolic blood pressure

DBP

diastolic blood pressure

MBP

mean arterial pressure

HR

heart rate

WT

wild type

Atrop

atropine

Prop

propranolol

Praz

prazosin

PE

phenylephrine

SNP

sodium nitroprusside

Hexm

hexamethonium

Ang II

angiotensin II

Ach

acetylcholine

MRA

mesenteric resistance artery

EPI

epinephrine

NE

nonepinephrine

ChIP

Chromatin Immunoprecipitation

Co-IP

Co-immunoprecipitation