In the regulation of vascular tone, the dilatory nitric oxide (NO)/cGMP pathway balances vasoconstriction induced by the renin-angiotensin and sympathetic nervous systems. NO-induced cGMP formation is catalyzed by two guanylyl cyclases (GC), NO-sensitive guanylyl cyclase 1 (NO-GC1) and NO-GC2, with indistinguishable enzymatic properties. In vascular smooth muscle cells, NO-GC1 is the major isoform and is responsible for more than 90% of cGMP formation. Despite reduced vasorelaxation, NO-GC1–deficient mice are not hypertensive. Here, the role of NO-GC1 in hypertension provoked by contractile agonists angiotensin II (Ang II) and norepinephrine (NE) was evaluated in NO-GC1–deficient mice. Hypertension induced by chronic Ang II treatment did not differ between wild-type (WT) and NO-GC1 knockout mice (KO). Also, attenuation of NO-dependent aortic relaxation induced by the Ang II treatment was similar in both genotypes and was most probably attributable to an increase of phosphodiesterase 1 expression. Analysis of plasma NE content—known to be influenced by Ang II—revealed lower NE in the NO-GC1 KO under Ang II-treated- and nontreated conditions. The finding indicates reduced sympathetic output and is underlined by the lower heart rate in the NO-GC1 KO. To find out whether the lack of higher blood pressure in the NO-GC1 KO is a result of reduced sympathetic activity counterbalancing the reduced vascular relaxation, mice were challenged with chronic NE application. As the resulting blood pressure was higher in the NO-GC1 KO than in WT, we conclude that the reduced sympathetic activity in the NO-GC1 KO prevents hypertension and postulate a possible sympatho-excitatory action of NO-GC1 counteracting NO-GC1’s dilatory effect in the vasculature.
Hypertension, a major risk factor for cardiovascular diseases, is associated with increased vascular tone. Within the regulation of the vascular tone, the dilatory nitric oxide (NO)/cGMP pathway balances the contractile action of: angiotensin II, the major biologically active metabolite of the renin-angiotensin system (RAS); and norepinephrine (NE) released from sympathetic nerve terminals. The cGMP-forming NO-sensitive guanylyl cyclase (NO-GC) holds the key position in the NO/cGMP pathway; in addition, cGMP is formed by the membrane guanylyl cyclase receptors (Kuhn, 2003). The NO-GC is a heterodimeric enzyme composed of an α and a β subunit and contains a prosthetic heme group that binds NO and causes an up to 200-fold stimulation of the cGMP production (Friebe and Koesling, 2003). Two isoforms of the NO-GC exist, NO-GC1 and NO-GC2, which contain the same beta1 subunit but different alpha subunits (α1 or α2) (Russwurm and Koesling, 2002). Although they differ in subcellular localization, the isoforms do not differ in regulatory or catalytic properties (Russwurm et al., 2001). To unravel their physiologic function, we generated knockout mice lacking either the NO-GC1 or NO-GC2 isoform (Mergia et al., 2006). First characterization revealed NO-GC1 as the major isoform in the vascular system, as shown by the pronounced reduction of cGMP-forming activity (>90%) and vascular relaxation. Yet, despite the reduced vascular relaxation, blood pressure of NO-GC1 knockout (KO) mice was not increased.
In addition to cGMP formation, cGMP signals are controlled by cGMP-hydrolyzing phosphodiesterases (PDEs). In vascular smooth muscle cells, PDE1 and PDE5 play important roles in regulation of the intracellular cGMP concentration (Rybalkin et al., 2003). PDE1 degrades cAMP and cGMP, whereas PDE5 is specific for cGMP degradation. PDE1 is activated by Ca2+ and calmodulin (CaM); thus, in the presence of a contractile agonist increasing intracellular Ca2+, PDE1 is activated and facilitates contraction by lowering cyclic nucleotide levels (Sonnenburg et al., 1995). An important contractile agonist that acts via a rise of Ca2+ is angiotensin II (Ang II) with an established role in the regulation of blood pressure (Stegbauer and Coffman, 2011). Ang II is targeted by the antihypertensive ACE inhibitors and AT1 receptor antagonists (Yusuf et al., 2008), and its effects are counteracted by NO/cGMP (Yan et al., 2003).
In the present study, NO-GC1 KO mice with greatly reduced vascular cGMP content were challenged with Ang II (high dose, 1.44 mg/kg per day, 2 weeks) to find out if the resulting hypertension is aggravated. Yet, Ang II-induced hypertension was not exacerbated by the NO-GC1 deficiency. Also, relaxation of aortic rings was likewise reduced by the Ang II treatment in wild-type (WT) and NO-GC1 KO and shown to result from an increased expression of PDE1. Further analysis revealed a reduced sympathetic activity in NO-GC1 KO. Thus, we hypothesized that normotension in the NO-GC1 KO is a balance between the lower sympathetic output and the reduced vascular cGMP. Consistent with this, hypertension induced by NE treatment was higher in NO-GC1 KO than in WT.
Materials and Methods
Experiments were performed with male NO-GC1 KO mice lacking the α1 subunit of the heterodimeric NO-GC1 (α1β1) and WT littermates backcrossed to C57Bl/6Rj background for more than 12 times (>N12 generation). The KO mice were generated and genotyped as described previously (Mergia et al., 2006). Two- to three-month-old mice were used. All animal experiments were performed according to the Guide for the Care and Use of Laboratory Animal published by the US National Institutes of Health (NIH Publication, 8th ed., 2011) and were also approved by the local animal care committee (license no. 87-51.04.2010.A039 and 8.87-50.10.34.08.216).
Angiotensin II, Norepinephrine, and Sildenafil Administration.
osmotic minipump (model 1002; ALZET/DURECT Corporation, Cupertino, CA) was implanted subcutaneously in the interscapular region to deliver Ang II or NE at a constant rate of 0.25 μl/hour in saline solution. For implantation, mice were anesthetized (100 mg/kg ketamine, 10 mg/kg xylazine) and thereafter treated with analgesics (0.2 mg/kg meloxicam) for three days. Ang II was infused at a dose of 1.44 mg/kg per day for 2 weeks, NE at a dose of 2.88 mg/kg per day for 2 weeks. Sildenafil citrate (Viagra; Pfizer, New York, NY) was dissolved in drinking water acidified with citric acid (pH ∼3) to a final concentration of 800 mg/l and given ad libitum resulting in a free plasma concentration of 21 ± 5 nmol/l sildenafil determined as described in Stegbauer et al. (2013); on the basis of an average water consumption of about 3 ml per day and a body weight of ∼25 g for an adult mouse, the daily sildenafil dose corresponded to 100 mg/kg. Mice were treated with sildenafil during the second week of Ang II treatment.
Two weeks after the minipump implantation, the mice were anesthetized by CO2 inhalation and decapitated. Blood was collected and hearts and aortas were harvested, weighed, and subjected to further analysis.
Blood Pressure Measurement.
Systolic blood pressure and heart rate were measured in conscious mice by noninvasive tail-cuff photoplethysmography (BP-98A; Softron Co., Tokyo, Japan). Because this device does not directly measure diastolic pressure but rather supplies an estimation calculated by a software algorithm, we do not report diastolic blood pressure values here.
For habituation the mice were measured daily for 7 days. After training, 10 measurements per mouse were taken daily for at least 5 days. Mice were measured before and during the second week after implantation of the osmotic minipumps.
Chronic Measurement of Intra-Arterial Pressure by Radiotelemetry.
Blood pressures were measured in conscious, unrestrained WT and NO-GC1 KO mice using radiotelemetry (PA-C10) as described previously (Butz and Davisson, 2001). In brief, for implantation of the telemetry catheters, the left carotid artery was cannulated and the catheter was advanced to the point where the small notch on the tubing resided at the vessel opening. Inserting the catheter up to this landmark notch assures the critical placement of the pressure-sensing tip just inside of the thoracic aorta. For the implantation of the telemetry catheters, mice were anesthetized with a single intraperitoneal injection of ketamine and xylazine (100 and 10 mg/kg, respectively). The adequacy of anesthesia was determined by the loss of a pedal withdrawal reflex. For analgesia, all mice received meloxicam (0.2 mg/kg). After transmitter implantation, mice were allowed to recover for 7 days to reestablished normal circadian rhythms. Blood pressure levels were recorded continuously with sampling every 5 minutes for 10-second intervals. Baseline measurements were recorded for 5 consecutive days. On day 6, osmotic minipumps (model 1002; ALZET) were implanted to infuse Ang II (1.44 mg/kg per day) for 14 days.
Organ Bath Experiments with Isolated Aortic Rings.
Thoracic aorta placed in Krebs-Henseleit buffer (118 mmol/l NaCl, 4.7 mmol/l KCl, 1.2 mmol/l MgSO4, 1.2 mmol/l KH2PO4, 25 mmol/l NaHCO3, 2.55 mmol/l CaCl2, and 7.5 mmol/l d-glucose, oxygenated with 5% CO2 in O2) were cut in four rings of similar size. The aortic rings were mounted on fixed segment support pins in two four-chamber myographs (Multi Wire Myograph 610 M; DMT, Aarhus, Denmark) containing 5 ml Krebs-Henseleit buffer. Resting tension was set to 5 mN. After equilibration in the presence of diclofenac (3 μmol/l), aortic rings were contracted (1 μmol/l phenylephrin), and subsequently vasodilation to carbachol, DEA-NO, or forskolin was recorded. Reponses to DEA-NO and forskolin were determined in the presence of the NO synthase inhibitor L-NAME (200 μmol/l).
Aortic rings derived from nontreated and Ang II-treated WT or NO-GC1 KO mice were examined in parallel. Effects of sildenafil were examined in parallel with aortic rings derived from Ang II-treated and Ang II + sildenafil-treated WT mice, and nontreated animals.
Western Blot Analysis.
To obtain aortic homogenates, aortas (from aortic arch to abdominal bifurcation) were homogenized in 300 μl of buffer (50 mmol/l TEA/HCl, 50 mmol/l NaCl, 2 mmol/l dithiothreitol, 0.2 mmol/l benzamidine, 0.5 mmol/l phenylmethylsulfonyl fluoride, 1 μmol/l pepstatin A; pH 7.4, 4°C) using a glass/glass homogenizer (900 rpm). After centrifugation (800g, 5 minutes, 4°C), supernatants were removed and subjected to further analysis. Protein concentrations were determined using a Bradford assay (Bio-Rad, Munich, Germany). The following Western blot analyses were performed as described previously (Mergia et al., 2003). Anti-PDE1A (Santa Cruz Biotechnology, Heidelberg, Germany) was used in a 1:500 and anti-PDE5 (New England Biolabs, Frankfurt, Germany) in a 1:1,000 dilution. The PDE signals were normalized to the alpha smooth muscle actin signal in the same lane (1:5000; ab5694; Abcam, Cambridge, UK).
Determination of cGMP Content in Intact Aortic Rings.
Aortic rings (eight per aorta) were allowed to equilibrate for 15 minutes in Krebs-Henseleit buffer (37°C, oxygenated with 5% CO2 in O2) and were then stimulated by carbachol (30 μmol/l, 5 minutes) or DEA-NO (100 μmol/l, 2.5 minutes). The equilibrated, nonstimulated, aortic rings were used as controls. After the reaction, aortic rings were snap frozen in liquid nitrogen and homogenized in 70% (v/v) ice-cold ethanol using a glass/glass homogenizer (900 rpm). After centrifugation (20,000g, 15 minutes, 4°C), supernatants were dried at 95°C and cGMP contents were measured by radioimmunoassay (Mergia et al., 2006). To standardize the different samples, protein pellets were dissolved (0.1 M NaOH/0.1% SDS) and protein content was determined using the bicinchoninic acid method (Thermo Scientific, Sunnyvale, CA).
Measurement of cGMP-Hydrolyzing PDE Activity.
PDE activity was measured by the conversion of [32P]cGMP (synthesized from [α-32P]GTP using purified NO-GC) to guanosine and [32P]phosphate in the presence of alkaline phosphatase (Sigma-Aldrich, St. Louis, MO) at 37°C for 5 minutes.
Reaction mixtures (0.1 ml) contained aortic homogenate (2 μl, ∼5 μg protein), [32P]cGMP (∼2 kBq), 1 μmol/l cGMP, 12 mmol/l MgCl2, 3 mmol/l dithiothreitol, 0.5 mg/ml bovine serum albumin, 2 IU of alkaline phosphatase, and 50 mmol/l TEA/HCl, pH 7.4. Reactions were stopped by the addition of 900 μl ice-cold charcoal suspension (30% activated charcoal in 50 mmol/l KH2PO4, pH 2.3). After pelleting of the charcoal by centrifugation (12,000g, 4 minutes), [32P]phosphate was measured in supernatant.
Sildenafil at 100 nmol/l was added to the measurements to determine the PDE activity ascribed to PDE5. EGTA (2 mmol/l; Sigma-Aldrich) or calcium (1 mmol/l CaCl2) and calmodulin (10 μmol/l; Merck, Darmstadt, Germany) was added to inhibit or stimulate PDE1, respectively. PDE assays were carried out in triplicates.
As the Ang II treatment yielded higher protein concentrations in the aortas, which can mostly be attributed to extracellular matrix proteins, enzyme activities (PDE, GC) were normalized to milliliters, with each aorta homogenized in 300 μl homogenization buffer.
Determination of NO-Stimulated GC activity.
The NO-stimulated GC activity (100 μmol/l DEA-NO, 10 minutes, 37°C) was determined in aortic homogenates of WT mice (3 μg) in the presence of GTP. The forming cGMP was quantified by radioimmunoassay as described previously (Mergia et al., 2006).
Determination of Norepinephrine in Plasma and Urine Samples.
To obtain plasma, blood from decapitated mice was collected in vials containing 0.3 mol/l EDTA, pH 8.0, and centrifuged (10,000g, 15 minutes, 4°C). Urine was collected from mice housed for 24 hours in metabolic cages (Tecniplast Deutschland, GmbH, Hohenpeißenberg Germany). Norepinephrine and 3,4-dihydoxybenzylamine hydrobromide (added as an internal standard; Chromsystems, München, Germany) were adsorbed onto alumina from EDTA plasma (100 μl) or urine (50 μl) and subsequently eluted with 0.1 mol/l HClO4 (250 μl). The eluate (100 μl) was injected into a high-performance liquid chromatography system equipped with a reverse-phase column (Guard-Pak Resolve C18; Waters Corporation, Milford, MA) in 15 mmol/l Na2HPO4, 30 mmol/l citric acid, 2 mmol/l Na2EDTA, 2.77 mmol/l (–)-octanesulfonic acid, and 12% methanol (v/v) (pH 6.5). Norepinephrine was detected with an electrochemical detector (Waters 460).
All data were expressed as means ± S.E.M. (n = number of mice). The experiments were compared statistically by unpaired Students’ t test. Concentration-response curves were evaluated by analysis of variance (ANOVA) for repeated measurements. Experiments were regarded as significant at a P value of less then 0.05. Blood pressure control values of all WT and NO-GC1–deficient mice used in the current study were averaged to circumvent minor differences owing to the small number of animals in a given group.
Deletion of the major NO-GC isoform, NO-GC1, resulted in a pronounced reduction of NO-induced cGMP formation and attenuated vascular relaxation. Yet, blood pressure in these mice was not increased (WT 107 ± 1 versus KO 108 ± 1 mmHg, on a C57BL/6 background). Normotension of the NO-GC1–deficient mice did not result from an upregulation of NO-GC2, as the expression NO-GC2 in the KO mice was found to be unaltered (Mergia et al., 2006).
In this study, we challenged mice with a high dose of Ang II (1.44 mg/kg per day, 2 weeks) to examine whether the NO-GC1 deficiency, and thus the reduced cGMP, exacerbates hypertension.
Ang II-Induced Hypertension and Hypertrophy Did Not Differ between WT and NO-GC1 KO Mice.
Chronic Ang II infusion is known to induce hypertension and heart hypertrophy in mice. Here, the systolic blood pressure was measured in conscious mice by noninvasive tail-cuff plethysmography. The two-week Ang II treatment led to an increase in systolic blood pressure by 42 ± 2 mmHg, which was similar in WT and NO-GC1 KO mice (Fig. 1A). The result indicates that the NO-GC1 deficiency with the reduced cGMP did not aggravate hypertension induced by this treatment. In support, blood pressure responses induced by lower, acute Ang II doses (10- and 100-fold, corresponding to 0.144 and 0.0144 mg/kg, i.p.) were comparable (30 and 10 mmHg, respectively), underlining that responsiveness of NO-GC1 KO toward Ang II is as in WT.
To validate the finding of the noninvasive blood pressure measurements, blood pressure was also measured directly using radiotelemetry. As shown in Fig. 1B, recordings of mean arterial pressure (MAP) were similar in WT and NO-GC1 KO mice under basal conditions and increased to the similar levels after Ang II infusion.
In WT, the Ang II treatment induced a significant heart hypertrophy as measured by heart-to–body-weight ratio. Again, Ang II-induced heart hypertrophy did not differ between WT and NO-GC1–deficient mice (Fig. 1C).
The Ang II Treatment Reduced Vascular Relaxation.
The functional consequences of long term Ang II treatment on blood vessel properties were analyzed by measuring relaxation of aortic rings in organ bath experiments. In line with previous results, endothelium-dependent relaxation in response to carbachol was reduced in nontreated NO-GC1 KO mice (54 ± 4% of WT maximal response). Ang II treatment attenuated endothelium-dependent relaxation in both WT and NO-GC1 KO mice with the extent of reduction being similar (WT 21 ± 3% and KO 17 ± 3%; Fig. 2A).
Next, aortic relaxation induced by an NO donor (DEA-NO) was studied. As shown by the rightward shift of the DEA-NO concentration-response curves, the Ang II treatment caused a reduction of NO-induced relaxation in both WT and NO-GC1–deficient mice; again the reduction was similar in WT and NO-GC1 KO as reflected by the 2-fold higher EC50 values (Fig. 2B). Thus, the Ang II treatment comparably affected NO/cGMP-dependent vascular relaxation in WT and NO-GC1 KO mice.
In addition to cGMP, cAMP also causes smooth muscle relaxation using similar cellular mechanisms. So, we asked whether cAMP-mediated relaxation is also affected by the Ang II treatment and studied aortic relaxation induced by the adenylyl cyclase (AC) stimulator forskolin (Fig. 2C). As with the NO-response, Ang II treatment caused a rightward shift of the forskolin concentration-response curve in both WT and NO-GC1–deficient aortic rings, reflected by 2-fold higher EC50 values for forskolin (see Fig. 2C).
Together, these results demonstrate: 1) alterations in vascular relaxation induced by Ang II treatment, which 2) were not aggravated by deletion of NO-GC1, the major vascular NO-GC.
Ang II Reduced the cGMP Content in Aortic Rings.
To test whether the Ang II treatment affected the aortic cGMP content, cGMP was measured in aortas in the absence or presence of a stimulator (carbachol, 30 μmol/l, 5 minutes; DEA-NO, 100 μmol/l, 2.5 minutes) by radioimmunoassay. In WT, the aortic cGMP content and the cGMP increased by carbachol was found to be reduced in the rings of the Ang II-treated mice (Fig. 3). Yet, aortic cGMP increases induced by high NO concentrations were unaffected by the Ang II-treatment (see Fig. 3).
In the aortas of NO-GC1–deficient mice, cGMP levels under nonstimulated conditions were only 10% of WT, which is in accord with a 90% reduction of the cGMP-forming activity. Also here, aortas of Ang II-treated mice showed reduced cGMP levels (see Fig. 3). Neither carbachol nor DEA-NO caused measurable cGMP increases in the NO-GC1 KO rings and the Ang II treatment had no effect (see Fig. 3).
The finding of decreased cGMP levels under Ang II treatment is in accord with the reduction of aortic relaxation and points to alterations of either cGMP-forming or -degrading activities.
Ang II Treatment Increased Aortic cGMP–Degrading Activity by Enhancing PDE1 Expression.
The intracellular cGMP concentration depends on the cGMP-degrading phosphodiesterases and the cGMP-forming guanylyl cyclases. Changes in expression and/or activity of one of these enzymes affect intracellular cGMP levels.
At first, the expression of PDE1A and PDE5, the major PDEs responsible for cGMP degradation in smooth muscle cells, was measured in aortic homogenates of nontreated and Ang II-treated WT mice. Quantitative analysis of Western blot results normalized to smooth muscle α-actin shows a 2-fold increase in expression of the PDE1A isoform in the Ang II-treated group, whereas the PDE5 content was unaffected by the Ang II treatment (Fig. 4A).
To verify enhanced expression of PDE1, PDE activity was determined in the respective aortic homogenates (1 μmol/l cGMP as substrate). Total PDE activity, as well as the PDE activity in the presence of Ca2+/CaM known to stimulate PDE1, was increased in Ang II-treated animals (Fig. 4B). Compared with nonstimulated conditions (EGTA), the Ca2+/CaM-induced increase in PDE activity was 3-fold in the Ang II-treated versus only 2-fold in the nontreated group, which is in accord with the enhanced expression of PDE1A in Western blot analysis. In contrast, PDE5 activity as measured in the absence and presence of 100 nmol/l sildenafil was not altered by the Ang II treatment.
To see whether the Ang II treatment also altered cGMP-forming activity, NO-stimulated GC activity was measured in aortic homogenate of nontreated and Ang II-treated WT mice. Yet, NO-stimulated cGMP-forming activities (100 μmol/l DEA-NO, 10 minutes) were not altered by the Ang II treatment.
In sum, Ang II treatment increased expression of PDE1, which most likely accounts for the Ang II-induced decrease of cGMP levels.
Sildenafil Treatment Reversed Ang II-Induced Reduction of the cGMP Content but Did Not Cause Functional Improvement.
To study whether inhibition of PDE1 is able to reverse the Ang II effects, the mice were treated with high sildenafil doses—considered to inhibit PDE1 in addition to PDE5—in the second week of Ang II treatment (800 mg of sildenafil per liter of drinking water).
Sildenafil was able to reverse the reductions in the cGMP content of aortic rings caused by the Ang II treatment in WT and NO-GC1 KO mice (see Fig. 3). Yet, despite the increased cGMP content in the aortic rings, the Ang II-induced impairment of endothelium-dependent and NO-induced aortic relaxation was unaffected (Fig. 5). Additionally, the sildenafil treatment did not affect the Ang II-induced hypertension in WT and NO-GC1 KO mice (WT: 142 ± 4 versus 147 ± 1 mmHg; KO: 155 ± 2 versus 152 ± 2 mmHg, sildenafil+Ang II- and Ang II-treated mice, respectively).
In sum, although sildenafil increased the cGMP content, the functional impairment caused by the Ang II treatment was not improved.
Norepinephrine Content Is Lower in the NO-GC1 KO Mice.
Ang II is known to increase activity of the sympathetic nervous system by increasing norepinephrine (NE) release. Thus, the activity status of the sympathetic system was monitored by determining NE in plasma and urine of WT and NO-GC1 KO mice. It is noteworthy that plasma NE of the NO-GC1 KO was found to be lower than in WT (Fig. 6A); a result further substantiated by the reduced NE content in the urine (Fig. 6B). As expected, Ang II treatment increased plasma NE levels in both NO-GC1 KO and WT mice (KO 1.4 ± 0.1-fold and WT 1.9 ± 0.05-fold). However, Ang II-induced plasma NE was still lower in NO-GC1 KO than in WT mice, indicating a reduction of sympathetic activity (see Fig. 6A). This finding prompted us to take a closer look at the heart rate, an important parameter regulated by the vegetative nervous system, and we found the heart rate in the NO-GC1 KO to be reduced (584 ± 10 versus 625 ± 7 beats/min, KO and WT, respectively; P < 0.001). In line with a reduced sympathetic activity, the ganglion blocker hexamethonium (30 mg/kg, i.p.) only increased the heart rate in the NO-GC1 KO to WT level (672 to 740 bpm in NO-GC1 KO, P < 0.05; 707 to 728 bpm in WT, n = 12 WT and 12 NO-GC1 KO) indicating that the parasympathetic control of the heart rate in the NO-GC1 KO is higher than in WT.
These results reveal a reduced sympathetic activity in the NO-GC1 KO mice that may be able to counterbalance the reduced cGMP in the vasculature, thereby explaining why Ang II-induced blood pressure increases are similar in WT and NO-GC1 KO mice.
NE-Induced Blood Pressure Increase Is Higher in the NO-GC1 KO Mice.
To verify the assumption that a reduced sympathetic activity is required for normotension in the NO-GC1 KO, the mice were treated with NE (2.88 mg/kg per day, 2 weeks).
As anticipated, NE treatment increased blood pressure in WT and NO-GC1 KO mice (Fig. 7). Yet, the NE-induced blood pressure increase in NO-GC1 KO mice was far more pronounced than in WT mice (ΔSBP: WT 22 ± 2 versus KO 37 ± 2 mmHg; P < 0.002). This finding supports our hypothesis that reduced sympathetic activity in the NO-GC1–deficient mice counteracts the reduced vascular cGMP content, thereby preventing hypertension. Thus, upon application of NE, the blood pressure increase in the NO-GC1 KO is higher than in WT. The results demonstrate for the first time the relevance of the NO-GC1–formed cGMP for the sympathetic tone in a KO model.
In the present study, we analyzed the role of the major NO-GC isoform (NO-GC1) in the vascular system and development of hypertension with the help of NO-GC1–deficient mice. In these mice, the NO-induced cGMP content in the vascular system is greatly reduced (by about 90%) as is vascular relaxation. Yet, these mice do not show any increase in blood pressure (on a C57BL/6 background; Buys et al., 2012; Stegbauer et al., 2013). Deletion of the β subunit, causing deficiency in both NO-GC isoforms, NO-GC1 and NO-GC2, is known to cause hypertension, suggesting that minor cGMP generation catalyzed by the NO-GC2 is able to prevent the increase of blood pressure (Friebe at al., 2007).
Chronic Ang II Treatment Impairs Vascular Relaxation.
In our study, chronic Ang II infusion was used to evaluate whether the resulting hypertension is exacerbated by the greatly reduced cGMP response caused by deletion of the NO-GC1. Ang II as a contractile agonist increases Ca2+ influx in the vascular smooth muscle cell causing smooth muscle contraction (Wynne et al., 2009). As expected, the treatment with Ang II caused a pronounced increase in blood pressure (∼40 mmHg) and heart weight. However, both parameters did not differ between WT and NO-GC1 KO mice.
Vascular relaxation analyzed in isolated aortic rings revealed that the Ang II treatment caused a reduction of endothelium-dependent relaxation, which is in accord with the results of others (Rajagopalan et al., 1996; Mollnau et al., 2002). Notably, WT and NO-GC1–deficient aortic rings were likewise affected by the Ang II treatment. A reduction of endothelium-dependent vascular relaxation has also been reported in the 2-kidney 1-clip model, another method known to increase the renin-angiotensin-system (Heitzer et al., 1999; Jung et al., 2003, 2004; Stegbauer et al., 2013). Increased superoxide production reducing bioavailability of NO is postulated to be responsible for the reduced endothelium-dependent vascular relaxation and is summarized in the term endothelial dysfunction (Förstermann and Sessa, 2012).
In addition to its effect on endothelium-dependent relaxation, Ang II treatment also reduced NO-induced relaxation in WT and NO-GC1 KO mice. The impairment of NO-mediated relaxation demonstrates that the effects of Ang II on the NO/cGMP cascade are not limited to the endothelium but also occur in smooth muscle cells, most likely on the level of the cGMP-forming/degrading enzymes. Respective Ang II-induced alterations like reduced expression of NO-GC (Mollnau et al., 2002), S-nitrosation of sGC (Crassous et al., 2012), or enhanced expression of PDEs (Giachini et al., 2011) have been reported.
Chronic Ang II Treatment Enhances PDE1 Expression.
In line with a reduced NO responsiveness, cGMP-levels of aortic rings of Ang II-treated WT and NO-GC1 KO mice were reduced under nonstimulated and carbachol-stimulated conditions. Our analysis of cGMP-forming and -degrading activities showed unaltered activity of NO-GCs, whereas cGMP degradation was enhanced in Ang II-treated WT mice. Closer analysis revealed an increase in Ca2+/CaM-stimulated cGMP-degrading activity, which can be ascribed to PDE1, the Ca2+-sensitive PDE. Increased expression of PDE1 was verified in Western blots. As PDE1 is known to also degrade cAMP, the enhanced PDE1 expression is in accord with the decreased vascular relaxation toward forskolin, the AC stimulator found in the present study.
Our results are in agreement with the study of Giachini et al. (2011), who also showed increased PDE1 expression in rat arteries induced by Ang II treatment. On the other hand, increased expression of PDE1 has been reported in vessels of rats treated with nitroglycerin to induce nitrate tolerance (Kim et al., 2001). Apparently, expression of PDE1 is an important mechanism in the control of cellular cGMP levels under quite diverse pathophysiological conditions (Miller et al., 2009). Yet, high Ang II doses are required for the pronounced increase of PDE1 expression, as a moderate increase of the renin-angiotensin-system induced in the 2K1C model was not paralleled by an increase of PDE1 expression (Stegbauer et al., 2013).
Increasing cGMP Levels Does Not Improve Ang II-Induced Dysfunctions.
The increased PDE1 most probably accounts for the Ang II-mediated reduction of NO/cGMP-signaling. To find out whether inhibition of PDEs and the subsequent cGMP increase can attenuate Ang II-induced effects, sildenafil in a high concentration (100 mg/kg per day) was applied during the second week of the Ang II treatment. Although cGMP levels were restored or even increased by the sildenafil treatment, confirming the effectiveness of the drug, vascular relaxation remained reduced and the Ang II-induced hypertension was not reversed.
We conclude that alterations already induced by Ang II (first week) are not susceptible to further increases by cGMP. For closer analysis of the contractility of the NO-GC1 KO mice, we plan to challenge the mice in the hypertensive state with relaxing agents that exert the dilatory function outside the NO/cGMP signaling cascade, e.g., calcium channel blockers.
Sympathetic Activity Is Reduced in the NO-GC1 KO.
In the vasculature, contractile agonists like Ang II and NE balance vasodilators like NO to maintain smooth muscle tone. By treating NO-GC1 KO mice with Ang II, we expected to see a blood pressure increase higher than in WT. Intriguingly, neither the blood pressure increase nor the impairment of vascular relaxation was aggravated by the NO-GC1 deficiency. The lack of an effect of the NO-GC1 deficiency can be explained by the Ang II-induced increase of PDE1 that reduces cGMP and thereby transforms WT into a NO-GC1–deficient–like phenotype. Still, why does NO-GC1 deficiency with an almost complete loss of vascular cGMP (90%) not cause hypertension?
Our analysis of the NE content revealed that sympathetic activity is reduced by the NO-GC1 deficiency as indicated by lower plasma and urine NE levels. NE levels, although increased by the Ang II treatment, still remained lower in the KO than in the Ang II-treated WT. In support of a reduction of sympathetic activity, the heart rate, a cardiovascular parameter highly regulated by the sympathetic tone, was found to be reduced in the NO-GC1 KO. The results underline that the sympathetic activity is reduced in the NO-GC1 KO and suggest that the reduction of sympathetic activity counterbalances the reduced vascular cGMP in the NO-GC1 KO mice. In support of the thesis, hypertension induced by NE treatment was more severe in NO-GC1 KO than in WT. The findings show that: 1) the reduced sympathetic activity has an important impact in the control of the blood pressure in the NO-GC1 KO; and 2) assign a sympatho-stimulatory role to NO-GC1. Indeed, stimulatory and inhibitory effects of NO/cGMP in the regulation of sympathetic activity in the blood pressure-controlling areas of brain stem have been shown before (Harada et al., 1993; Hirooka et al., 1996; Martins-Pinge et al., 1997; Sander and Victor, 1999; Morimoto et al., 2000; Mayorov DN 2005; Wang et al., 2007), and occurrence of NO-GC1 and NO-GC2 in the medulla oblongata has been reported (Mergia et al., 2003). Yet, although our data provide evidence for the notion that NO via NO-GC1 increases sympathetic tone, more direct measurements such as recordings of renal sympathetic activity are necessary to definitely establish this relationship.
Moreover, the relative impact of NO-GC1’s direct vasodilatory effect versus its sympatho-excitatory blood pressure–increasing effect has to be unraveled with the help of genetically modified mice with distinct neuronal or vascular NO-GC1 deficiency (Fig. 8).
We thank Medah Özcan, Arkadius Pacha, Christine Schwandt, and Caroline Vollmers for technical assistance.
Participated in research design: Koesling, Mergia.
Conducted experiments: Broekmans, Mergia, Stegbauer.
Contributed new reagents or analytic tools: Potthoff, Stegbauer, Russwurm.
Performed data analysis: Broekmans, Mergia, Russwurm, Stegbauer.
Wrote or contributed to the writing of the manuscript: Broekmans, Koesling, Mergia, Russwurm.
- Received July 15, 2015.
- Accepted November 3, 2015.
This work was supported by the Deutsche Forschungsgemeinschaft (DFG FOR 2060/1). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
- Ang II
- Angiotensin II
- nitric oxide
- NO-sensitive guanylyl cyclase
- Copyright © 2015 by The American Society for Pharmacology and Experimental Therapeutics