Glypican 1 and syndecan 1 differently regulate noradrenergic hypertension development: focus on IP3R and calcium

Abstract

Background–

Vascular dysfunction is a checkpoint to the development of hypertension. Heparan sulfate proteoglycans (HSPG) participate in nitric oxide (NO) and calcium signaling, key regulators of vascular function. The relationship between HSPG-mediated NO and calcium signaling and vascular dysfunction has not been explored. Likewise, the role of HSPG on the control of systemic blood arterial pressure is unknown. Herein, we sought to determine if the HSPG syndecan 1 and glypican 1 control systemic blood pressure and the progression of hypertension.

Purpose–

To determine the mechanisms whereby glypican 1 and syndecan 1 regulate vascular tone and contribute to the development of noradrenergic hypertension.

Experimental Approach and Key Results–

By assessing systemic arterial blood pressure we observed that syndecan 1 (Sdc1^−/−) and glypican 1 (Gpc1^−/−) knockout mice show a similar phenotype of decreased systolic blood pressure that is presented in a striking manner in the Gpc1^−/− strain. Gpc1^−/− mice are also uniquely protected from a norepinephrine hypertensive challenge failing to become hypertensive. This phenotype was associated with impaired calcium-dependent vasoconstriction and altered expression of calcium-sensitive proteins including SERCA and calmodulin. In addition, Gpc1^−/− distinctively showed decreased IP₃R activity and increased calcium storage in the endoplasmic reticulum.

Conclusions and Implications–

Glypican 1 is a trigger for the development of noradrenergic hypertension that acts via IP₃R- and calcium-dependent signaling pathways. Glypican 1 may be a potential target for the development of new therapies for resistant hypertension or conditions where norepinephrine levels are increased.

Graphical abstract

1. Introduction

Hypertension is a multifactorial syndrome that affects approximately 30% of global adult population (Mills, Stefanescu & He, 2020). It is often presented in its essential form, i.e., with unknown origin. Its progression relies on numerous factors, but a common feature of hypertensive disease is increased sympathetic tone (Cruickshank, 2017; Engelman, Portnoy & Sjoerdsma, 1970; Missouris, Markandu, He, Papavasileiou, Sever & MacGregor, 2016; Watson et al., 2019) which is also observed in animal models of hypertension (Jackson, Head, Gueguen, Stevenson, Lim & Marques, 2019; Kong et al., 2018; Xu et al., 2020).

Another common feature of hypertension is vascular dysfunction (Ghiadoni, Taddei & Virdis, 2012; Taddei, Virdis, Ghiadoni, Versari & Salvetti, 2006) which is intimately associated with impaired nitric oxide (NO) signaling. Heparan sulfate proteoglycans (HSPG) are known regulators of vascular function. HSPGs are proteoglycans that contain a core protein and one or more heparan sulfate (HS) glycosaminoglycan (GAG) chains covalently attached to their extracellular domain. HSPG can be either secreted to the extracellular matrix or physically interacting with cellular membranes (Sarrazin, Lamanna & Esko, 2011). The known HSPGs located at cell membranes are syndecans and glypicans (Mills, Stefanescu & He, 2020). These HSPG differ in their structure and on the content of the associated GAG. Syndecans are transmembrane proteoglycans that contain a large extracellular domain and a highly conserved cytoplasmic domain. They carry both heparan sulfate and chondroitin sulfate attached to a GXXXG motif at the extracellular domain. Glypicans are bound to the cell membrane by a glycosyl-phosphatidylinositol (GPI) anchor located within a hydrophobic domain close to the C-terminus and only carry heparan sulfates at their extracellular domain (Bartlett, Hayashida & Park, 2007; Chignalia et al., 2016; Filmus, Capurro & Rast, 2008).

The effects of HSPG on the vasculature include the activation of NO signaling (Ebong, Lopez-Quintero, Rizzo, Spray & Tarbell, 2014; Mahmoud, Mayer, Cancel, Bartosch, Mathews & Tarbell, 2020), the control of calcium dynamics (Borland et al., 2017) and the regulation of leukocyte adherence (Kowalewska, Patrick & Fox-Robichaud, 2014). The role of HSPG on noradrenergic signaling and its impact on vascular function and disease is yet to be clarified.

Among the different HSPG, syndecan 1 and glypican 1 have been shown to activate endothelial nitric oxide synthase (eNOS) in conditions of increased shear stress (Bartosch, Mathews & Tarbell, 2017; Ebong, Lopez-Quintero, Rizzo, Spray & Tarbell, 2014). Recently, glypican 1 expression has been shown to be decreased in aortas of aged mice (Mahmoud, Mayer, Cancel, Bartosch, Mathews & Tarbell, 2020) and this was associated with decreased eNOS activity. The specific contribution of syndecan 1 and glypican 1 to eNOS-dependent regulation of the vascular tone has not been explored. Moreover, the potential effects of glypican 1 and syndecan 1 on vascular adrenergic signaling are unresolved. A recent report suggested that HSPG may contribute to decreased vascular reactivity in response to a high-fat diet (Kang et al., 2020). In addition, a correlation between elevated serum levels of syndecan 4 and increased blood pressure has been demonstrated (De Luca, Bryan & Hunter, 2021). However, the role of HSPG in controlling systemic blood pressure is incipient.

We have previously shown that noradrenergic signaling contributes to the activation of eNOS by a canonical heparan sulfate-dependent pathway (Chignalia et al., 2018). In the present study, we show that syndecan 1 and glypican 1 are regulators of noradrenergic signaling and vascular tone. These HSPG act by distinct pathways with glypican 1 having a pivotal role on IP₃R activation, calcium dynamics and vasoconstriction. Herein, we expose glypican 1 as a trigger for the development of noradrenergic hypertension.

2. Methods

2.1. Animals

Male glypican 1 knockout (Gpc1^−/−), syndecan 1 knockout (Sdc1^−/−), CD1 and C57BL/6 mice with 6 to 8 weeks of age were used in this study. CD1 and C57BL/6 are the genetic background for Gpc1^−/− and Sdc1^−/− respectively and were used as the controls for each knockout strain. Gpc1^−/− and Sdc1^−/− breeding pairs were kindly donated by Dr. Arthur Lander and Dr. Pyong Park, respectively. CD1 and C57BL/6 were purchased from Charles River Laboratories (Wilmington, MA). Animal experiments were approved by the Animal Committee of The University of Illinois at Chicago (18-101) and the Animal Committee from The University of Arizona (18-491) as well as were according to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.

2.2. Genotyping

Genomic DNA was isolated from mouse-tail tissue following the protocol provided by Jacks Lab (Koch Institute for Integrative Cancer Research at MIT). Tail tissue was cut into small pieces and incubated overnight at 50-60°C in lysis buffer supplemented with proteinase K (20 mg/mL, code 25530-049, Invitrogen). In the next day, 250 μL of NaCl (6 mol/L) was added to the samples and strongly shaken in a vortex. The mixture was centrifuged (10.000 rpm) for 10 min at 4°C. Supernatant was placed in a clean tube and 650 μL of isopropanol was added to the samples for 15 min at room temperature. The samples were centrifuged (13.5000 rpm) for 10 min at room temperature to recover DNA. Then, tubes were inverted on bench to allow air dry for 5 min. After that, 150 μL of TAE buffer (Tris-acetate-EDTA, pH 7.5) was added in each sample, incubated at 50-60°C for 10 min. Concentration of total DNA was measured in a Nanodrop ND-1000 Spectrophotometer (Thermo Scientific). Polymerase Chain Reactions. Primers were synthesized by Integrated DNA Technologies. Endpoint PCR was performed using Taq Platinum (Invitrogen) Kit according to manufacturer’s instructions. Reactions were carried out in a 2720 thermocycler (Applied Biosystems). Samples were kept at −20°C until submitted to electrophoresis in agarose gel for visualization of the bands. PCR primers are listed in Table 1.

Table 1. Sequence of primers used for genotyping Sdc1^−/− and Gpc1^−/− mice.

Primers Syndecan 1 forward and reverse and neomycin were used to distinguish knockout and wild type animals for the Sdc1^−/− strain. The primers Gpc1 Mutant (forward and reverse) and Gpc1 wild type (forward and reverse) were used to distinguish knockout from wild type mice from the Gpc1^−/− strain.

Primer name	Forward	Reverse
Syndecan 1 (Sdc1)	CGC CGA AAC CTA CAG CCC TC	GCA TCG GCG AGT GGC GAG TC
Neomycin	CGAGACTAGTGAGACGTGCTACTTCC
Glypican 1 Mutant	AGCCGGCTTTTGTTGTCTC	CACGAGTGTGCTAGGATAGGG
Glypican 1 Wild Type	CAGCGAAGTCCGCCAGAT	CAGACCTCCCGAGTGCTAGG

2.3. Model of noradrenergic hypertension

For the implantation of osmotic minipumps, mice were anesthetized via intraperitoneal injection of ketamine (90 mg/kg) and xylazine (10 mg/kg). When the loss of the pedal pain withdrawal reflex, slowing of the heart and breathing was set, Buprenorphine SR (1.0 mg/Kg) was administered subcutaneously for analgesia and surgery was started. Briefly, the abdominal region was shaved, and a 1-cm incision was done in the abdominal wall. An osmotic minipump (Alzet model 2002; Alza Corp) that contained norepinephrine was inserted to allow norepinephrine infusion (5.6 mg · kg⁻¹ · d⁻¹) for fourteen days. Sham-operated animals underwent an identical surgical procedure, except that an osmotic minipump filled with PBS was implanted. The dose of norepinephrine was based on previously published studies (Fukai, Siegfried, Ushio-Fukai, Griendling & Harrison, 1999; Qin et al., 2008).

2.4. Systolic blood pressure measurement in mice

Mice were anaesthetized via intraperitoneal injection of ketamine (90 mg/kg) and xylazine (10 mg/kg). When the loss of the pedal pain withdrawal reflex and slowing of breathing was set, blood pressure measurements were performed. Readings were replicate in non-anesthetized, trained, mice to exclude potential effects of anesthetics. Briefly, the mouse was placed in platform with the tail exposed to a 15 min warming period (39 ± 2 °C) to stimulate vasodilation and then, the values of systolic blood pressure (SBP) and heart rate (HR) were measured using an indirect mouse-tail blood pressure system (Coda^™). These data were obtained at days 0 (before and after minipump implantation), 3, 7 and 14 of the treatment.

2.5. Vascular reactivity

Vessel preparation

Mice were anesthetized with isoflurane (1-5% in 100% O₂) by inhalation. When the loss of the pedal pain withdrawal reflex and slowing of the heart and breathing was set, mice were exsanguinated. Thoracic aortas were collected and placed in a cylindrical glass dish containing cold Krebs-Henseleit solution [mmol/L] (NaCl 130.00, NaHCO₃ 14.9, C₆H₁₂O₆ 5.5, KCl 4.7, KH₂PO₄ 1.18, MgSO₄ 1.17, CaCl₂ 1.6, HEPES 10.0; pH 7.4). The thoracic aortas were cleaned of connective tissue with the assistance of a stereomicroscope and 2 mm long aortic rings were cut. The aortic rings were placed in a myograph chamber (Radnoti LLC) and then connected to a transducer system for recording isometric tension (Power Lab 8/35, ADInstruments). To develop active basal tension, the aortic rings were stretched until they reached 0.75 g. Active basal tension was chosen based on preliminary tension-response curve experiments (data not shown). The rings were then left under basal resting tension in Krebs-Henseleit solution (pH 7.4, 37°C) for 45 min. Next, the vasoactive response to different agonists were tested as described in the vascular reactivity protocols section below.

Protocols

In non-treated groups assigned as Sdc1^−/−, Gpc1^−/−, C57BL/6 and CD1, cumulative concentration–response curves to NE (1 nmol/L to 10 μmol/L), ACh (0.1 μmol/L to 0.1 mmol/L), SNP (1 nmol/L to 0.1 mmol/L) and KCl (0, 20, 40, 60, 80 and 100 mmol/L) in the presence or absence of endothelium were performed. Moreover, cumulative concentration–response curves were done to NE (1 nmol/L to 10 μmol/L) and ACh (0.1 μmol/L to 0.1 mmol/L) in groups treated with NE or PBS. The aortic rings were contracted with phenylephrine (PE, 1 μmol/L), the relaxation (70%) in response to acetylcholine (ACh, 10 μmol/L) confirmed endothelium integrity (E+), while the absence of relaxation (10%) confirmed the efficiency of endothelium removal (E−) by friction of aortic ring in a metal wire. Aortic ring integrity was also tested at the end of each experiment by the assessment of contractile response to KCl. Aortic rings were considered viable and with integrity if the second response to KCl 60mmol/L, at the end of the experiment, was at least 95% of the initial response.

In another sequence of experiments, using aortas from animals chronically treated with NE or PBS, in the presence of nominal zero calcium Krebs solution, aortic rings were incubated with Nifedipine (1 μmol/L) and a cumulative concentration–response curves to PE (1 nmol/L to 0.1 mmol/L) was performed. To assess if calcium storage is altered in the knockout strains, the maximal contractile response to one single concentration of caffeine (20 mmol/L) was analyzed. In addition, to indirectly assess if the entry of calcium via L-type calcium channels is altered in the knockout strains, the maximal response to one single concentration of KCl (60mmol/L) was analyzed.

2.6. Western Blot

The thoracic aortas of mice treated with NE or PBS were cleaned, frozen in liquid nitrogen, and stored at −80°C. The samples were macerated in RIPA buffer supplemented with a cocktail of protease and phosphatase inhibitors (Sigma Aldrich), NaF (1 mmol/L), Na₃VO₄ (1 mmol/L) and PMSF (10 mmol/L). The homogenates were centrifuged (4 °C, 12000 rpm, 20 min) and the supernatants were used for protein quantification by the Bradford assay. A total of 30 μg of protein obtained from each sample was submitted to electrophoresis on polyacrylamide gel (8 to 15%) and transferred to a nitrocellulose membrane. The membranes were incubated with primary antibodies (overnight) against alpha-1D adrenoceptor, alpha-2B adrenoceptor, endothelial nitric oxide synthase (eNOS), p-eNOS Ser¹¹⁷⁷, sarcoplasmic reticulum Ca2+-ATPase (SERCA 1/2/3), TRPC3, P2X4, calmodulin (CaM), phospho inositol-triphosphate receptor (p-IP₃R) and IP₃R. The membranes were then incubated with specific secondary antibody (anti-rabbit, anti-mouse or anti-goat) at room temperature for 60 min. The bands were detected by chemiluminescent system (LI-COR, Odyssey Fc, Gene Company), β-actin was used to normalize the results. The bands were quantified with the ImageJ Software (NIH Image).

2.7. Measurement of norepinephrine by ELISA

Whole-blood samples were collected from mice with treated with NE or PBS and placed in polystyrene tubes with 0.1 mL heparin (1000U/mL). The samples were centrifuged (4 °C, 3000 rpm, 10 min) to separate the plasma. The measurement of plasma NE was performed using an ELISA kit (Abnova, KA1891), according to the manufacturer’s protocol.

2.8. Colorimetric Griess reaction

The mice thoracic aortic rings (n = 5 per group) were incubated in a bath chamber containing Krebs-Henseleit solution (pH 7.4, 37 °C) for 30 min. The rings were stimulated with PE (10 μM) until they reached the plateau (~ 10 min). The aortic rings were then stimulated with ACh (10 μM). When relaxation reached maximum response, 50 μL of the bath solution of each sample was collected and added to 50 μL of Griess reagent (a 1:1 dilution of N-(1-Naphthyl)ethylenediamine dihydrochloride 1% in deionized H₂O and sulfanilamide 1% in H3PO4 5%) in a 96-well plate. A standard curve to Sodium nitrite (3 μmol/L to 200 μmol/L) was used as standard. The absorbance was read at 540 nm.

2.9. Statistical analysis

The maximum effect (ME) values are considered as the maximum response reached in the cumulative concentration-curves for the relaxation or contractile agents. For the vascular reactivity studies, the values were determined after logarithmic transformation of the normalized concentration–response curves and are reported as the negative logarithm pD₂, that are calculated as −log of the EC₅₀. In addition, the delta area under the curve of (AUC) was calculated to all concentration-effect curves.

The results are expressed as the mean ± standard error of the mean (SEM) of the obtained values; n indicates the number of animals or the number of aortic rings that were used in the experiments. Statistical analysis was performed using one-way ANOVA with Tukey post-hoc test using GraphPad Prism 5.0 software or unpaired, two-tailed, student t-test when appropriate. Differences were considered statistically significant when p < 0.05.

3. Results

3.1. Gpc1^−/− mice do not develop hypertension with norepinephrine challenge.

Basal values of systolic blood pressure (SBP) were lower is Sdc1^−/− and Gpc1^−/− mice when compared to with their respective controls (C57BL/6 and CD1 mice, respectively), while heart rate (HR) was similar among groups (Figures 1A–B). Treatment with PBS (Sham) did not alter SBP in all strains (Figures 1C–D). When mice were chronically treated with NE (5.6mg/kg, 14 days) (Fukai, Siegfried, Ushio-Fukai, Griendling & Harrison, 1999) the values of SBP were significantly increased in C57BL/6, CD1 and Sdc1^−/− after 3 days, when a plateau was reached and sustained until day 14 (Figures 1 C–D). The SBP values fit the hypertensive range of other mouse models of hypertension. In an exceptional manner, Gpc1^−/− mice did not show an increase in systolic blood pressure when challenged with NE (Figure 1D). The delta (Δ) increase in SBP induced by NE treatment was lower in Sdc1^−/− than in C57BL/6 mice (Figure 1E). Treatment with PBS or NE had no effect on HR (Figure 1F).

Figure 1: Blood Pressure Measurements.

(A) Gpc1^−/− and Sdc1^−/− have lower systolic blood pressure (SBP, mmHg) than respective controls; (B) Gpc1^−/− and Sdc1^−/− do not show differences in heart rate (HR, bpm). (C, D) Treatment with norepinephrine (NE) for fourteen days leads to hypertension in Sdc1^−/−, C57BL/6 and CD1 but not Gpc1^−/− mice. (E) The delta (Δ) increase on systolic blood pressure (SBP, mmHg) after treatment with NE is reduced in Sdc1^−/− and Gpc1^−/−. (F) Heart rate (HR, bpm) was not changed after treatment with NE or PBS in all groups. Data represent the mean ± SEM of the experiments, and n=8 represents the number of animals used in the experiments. *^{, #}p < 0.05 statistical difference between Sdc1^−/− vs C57BL/6, Gpc1^−/− vs CD1 or strain-NE vs strain Sham (One-way ANOVA followed by Tukey post-hoc test [Fig1A, B, E, F]; Two-way ANOVA [Figure 1C–D]).

3.2. Gpc1^−/− and Sdc1^−/− show increased vascular expression of adrenergic receptors without changes in norepinephrine metabolism.

Basal expression of alpha-1D adrenoceptor (α1D-AR) and alpha-2B adrenoceptor (α2B-AR) subtypes was increased in Sdc1^−/− (Figure 2A) and Gpc1^−/− (Figure 2B) aortas when compared with their respective control groups. Increased α1D-AR and α2B-AR expression was maintained after NE treatment (Figures 2C–D). Plasma levels of norepinephrine were similar in basal conditions across studied strains. Norepinephrine challenge led to a similar increase in norepinephrine plasma levels in all studied strains (Figure 2E).

Figure 2: Increased Adrenoceptor expression in Gpc1^−/− and Sdc1^−/−.

The expression of alpha1D and alpha 2B adrenoceptors is increased in aortic rings of sham (PBS treated) (A) Sdc1^−/− and (B) Gpc1^−/− when compared to respective wild-type controls (C57BL/6 and CD1). (C, D) Treatment with norepinephrine (NE) does not alter increased expression of alpha1D and alpha 2B adrenoceptors in aortic rings. Changes in brightness and contrast were equally applied over the entire representative western blot images. (E) Norepinephrine plasma levels increased with NE treatment in Sdc1^−/−, Gpc1^−/−, C57BL/6 and CD1 mice. Data represent the mean ± SEM of the experiments, and n=5-6 represents the number of sample or aortic rings used in the experiments. * indicates p < 0.05 statistical difference between Sdc1^−/− vs C57BL/6, Gpc1^−/− vs CD1 or NE versus Sham (PBS) treatments (One-way ANOVA followed by Tukey post-hoc test [Fig 2E] and Student’s t-test [Fig 2A–D]).

3.3. Gpc1^−/− mice uniquely show impaired vascular contractile response phenotype

The vascular contractile response to norepinephrine was assessed in aortic rings of all studied strains. The maximum contractile effect induced by norepinephrine was increased in Sdc1^−/− endothelium-intact aortas when compared to C57BL/6 aortas (Figure 3A). This difference was abolished in endothelium-denuded aortic rings (Figure 3B). On the other hand, norepinephrine-induced maximum effect was significantly reduced in Gpc1^−/− aortic rings with endothelium when compared to aortic rings from CD1 mice. This response was not altered in endothelium-denuded aortic rings (Figures 3A–B).

Figure 3: Impaired vasoreactivity in Gpc1^−/− and Sdc1^−/−.

Concentration-response curves to (A, B) norepinephrine (NE, 1 nmol/L to 10 μmol/L), to (C, D) potassium chloride (KCl, 10 to 100 mmol/L), to (E) acetylcholine (ACh, 0.1 μmol/L to 0.1 mmol/L) and to (F) sodium nitroprusside (SNP, 1 nmol/L to 10 μmol/L) in aortic rings with (E+) or without (E−) endothelium of Sdc1^−/−, Gpc1^−/− C57BL/6 and CD1 mice. The delta area under the curve (ΔAUCs) were calculated from the individual concentration-response curve plots. Gpc1^−/− mice show decreased vasoconstriction to NE and KCl and reduced relaxation to ACh. Data represent the mean ± SEM of the experiments, and n=3-7 represents the number of aortic rings used in the experiments. *p < 0.05 statistical difference in maximum effect values between Sdc1^−/− vs C57BL/6. ^#p < 0.05 statistical difference in maximum effect values for Gpc1^−/− vs CD1 (One-way ANOVA followed by Tukey post-hoc test).

To further investigate the contractile properties of aortic rings across mice strains we assessed the vasoconstrictive response to potassium chloride (KCl) to bypass the components of the noradrenergic signaling cascade. Increased K+ concentrations in the extracellular space directly depolarize the membrane and activate voltage-gated calcium channels that contributes to increase intracellular Ca²⁺ and induce contraction. This mechanism called electromechanic coupling is an extremely valuable tool for investigation, especially in our case that the pharmacomechanical activation mechanism is implied in the overall hypothesis. Sdc1^−/− showed similar vasoconstriction in response to KCl when compared to C57BL/6 mice in the presence and absence of endothelium (Figures 3C–D). KCl-induced vasoconstriction was decreased in Gpc1^−/− compared to CD1 mice, in aortic rings with or without endothelium (Figures 3C–D). The delta AUC analysis showed a similar effect as the maximum contractile effect observed in all concentration-response curves stimulated by NE or KCl (Figures 3A–D).

3.4. Glypican 1 or Syndecan 1 genetic deletion results in similar vasodilator profile

The endothelium-dependent vasodilation, as assessed by ACh-induced relaxation, was decreased in Sdc1^−/− and Gpc1^−/− mice when compared with their respective control groups (Figure 3E). Delta AUC analysis did not indicate a difference in the endothelium-dependent response between knockout strains. In the absence of the endothelium, the relaxation induced by sodium nitroprusside (SNP), a classic NO donor, was similar in all strains (Figure 3F) as well as the delta AUC. eNOS activity, as assessed by the ratio of phosphorylation at Serine¹¹⁷⁷/total eNOS expression, and the production of nitrogen oxides (NO_x) were decreased in aortic rings of Sdc1^−/− and Gpc1^−/− mice (Figure 4A–C). Total eNOS expression was decreased in aortic rings of Sdc1^−/−, but unaltered between Gpc1^−/− and CD1 mice (Figure 4B).

Figure 4: Altered NO signaling in Gpc1^−/− and Sdc1^−/−.

(A, B) eNOS activity is reduced in Gpc1^−/− and Sdc1^−/−; eNOS expression is reduced in Sdc1^−/− but not Gpc1^−/−. Figures A and B contain representative images of western blots. β-Actin representative images are from different exposure times and were combined for illustrative purposes. Edited images are indicated by dashed lines. Changes in brightness and contrast were equally applied over the entire representative western blot images. (C) NO_x levels measured in aorta’s supernatant from Sdc1^−/−, Gpc1^−/−, C57BL/6 and CD1 mice. Concentration-response curves to acetylcholine (ACh, 0.1 μmol/L to 0.1 mmol/L) in aortic rings with endothelium (E+) of (D, E) Sdc1^−/−, Gpc1^−/−, C57BL/6 and CD1 mice treated with norepinephrine (NE) or PBS (Sham). The delta area under the curve (ΔAUC) was calculated from the individual concentration-response curve plots. Data represent the mean ± SEM of the experiments and n=3-5 represents the number of samples or aortic rings used in the experiments. *p < 0.05 statistical difference between Sdc-1^−/− vs C57BL/6, Gpc-1^−/− vs CD-1, maximum effect values between C57BL/6 NE vs C57BL/6 Sham or CD1 NE vs CD1 Sham (One-way ANOVA followed by Tukey post-hoc test [Fig. 4D–E] and Student’s t-test [Fig 4D–E]).

3.5. Decreased endothelium-dependent relaxation is maintained after NE treatment in Sdc1^−/− and Gpc1^−/− mice

Endothelium-dependent relaxation was assessed in aortic rings isolated from mice of all studied strains treated with PBS or norepinephrine. ACh-induced vasorelaxation was decreased in aortic rings from wild type mice (C57BL/6 and CD1) treated with norepinephrine compared to PBS-treated mice. Norepinephrine challenge did not further alter the decreased ACh-induced relaxation or AUC in Sdc1^−/− or Gpc1^−/− mice (Figure 4D–E).

3.6. Desensitization to norepinephrine occurs in Sdc1^−/− but not Gpc1^−/− aortic rings

Vascular contractility was assessed in aortic rings isolated from mice treated with PBS (Sham) or norepinephrine for fourteen days. The maximum effect of norepinephrine was decreased in aortic rings of C57BL/6 (Figure 5A), Sdc1^−/− (Figure 5A) and CD1 (Figure 5B) treated with NE when compared to respective vehicle-treated mice. This phenomenon was not observed in Gpc1^−/− mice (Figure 5B). Delta AUC values corroborated the maximum effect response observed in concentration-response curves to NE.

Figure 5: Altered vasocontractility in norepinephrine-treated Gpc1^−/− and Sdc1^−/−.

Concentration-response curves to (A, B) norepinephrine (NE, 1 nmol/L to 10 μmol/L) in normal Krebs buffer, (C, D) phenylephrine (PE, 1 nmol/L to 10 nmol/L) in the presence of nifedipine (1 μmol/L) in nominal Calcium Krebs buffer performed in aortic rings of Sdc1^−/−, Gpc1^−/−, C57BL/6 and CD1 mice treated with norepinephrine (NE) or PBS (Sham). The delta area under the curve (ΔAUC) were calculated from the individual concentration-response curve plots. Data represent the mean ± SEM of the experiments, and n=4-7 represents the number of aortic rings used in the experiments. *p < 0.05 statistical difference in maximum effect values between Sdc1^−/− vs C57BL/6 or Gpc1^−/− vs CD1 (One-way ANOVA followed by Tukey post-hoc test)

3.7. Blunted response to Norepinephrine in Gpc1^−/− mice is associated with altered calcium mobility in aortic rings

The effects of caffeine and KCl on aortic rings contraction were also explored. The maximum contraction induced by a single concentration of caffeine was similar between aortic rings of Sdc1^−/− and C57BL/6 mice from Sham and NE-treated groups. However, caffeine-induced contraction was increased in Gpc1^−/− mice compared to CD1 in aortic rings treated with vehicle or NE (Figures 6A and 6B). No differences in the response to a single concentration of KCl were observed among the strains, before or after treatment with NE (Figures 6C and 6D).

Figure 6: Effects of Gpc1 and Sdc1 on calcium-sensitive pathways.

(A, B) Aortic rings from glypican 1 but not syndecan show increased calcium storage in control and challenged mice stimulated by Caffeine (20 mmol/L). (C, D) No differences are found in KCl-induced maximum contraction (60 mmol/L). (E, F) Calmodulin (CaM) expression is increased in Sdc1^−/− and Gpc1^−/− aortas. Changes in brightness and contrast were equally applied over the entire representative western blot images. Data represent the mean ± SEM of the experiments, and n=4-7 represents the number of aortic rings used in the experiments. *p < 0.05 statistical difference between Sdc1^−/− vs C57BL/6 or Gpc1^−/− vs CD1 (One-way ANOVA followed by Tukey post-hoc test [Fig 6A–D] and Student’s t-test [Fig 6E–F]).

The role of extracellular calcium on NE-induced contraction was assessed using nifedipine, a L-type Ca²⁺-channel blocker. While nifedipine partly reduced phenylephrine-induced contraction in aortic rings of Sdc1^−/− mice compared to those of C57BL/6 mice (Figure 5C), it completely abolished phenylephrine-induced contraction in Gpc1^−/− aortas compared to aortas from CD1 mice (Figure 5C). This effect was not altered by chronic treatment with NE (Figure 5D).

3.8. Altered calcium mobility in Gpc1^−/− and Sdc1^−/− is associated with increased Calmodulin content

The expression of calcium-sensitive proteins was assessed in aortic rings of all studied mouse strains. Sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) 1/2/3 (Sup. Fig. 1A and 1B) and Calmodulin (CaM) (Figures 6E and 6F) expression was higher in aortic rings of Sdc1^−/− and Gpc1^−/− mice compared to their respective control groups. Treatment with NE did not change the expression of SERCA (1/2/3) (Sup. Fig. 1C and 1D) or CaM (Figures 6E and 6F). No differences in TRPC3 (Sup. Fig. 2A–D) or P2X4 (Sup. Fig. 2E–H) were observed before or after treatment with NE.

3.9. Opposite IP₃R activity in Sdc1^−/− and Gpc1^−/− mice

In basal conditions, vascular IP₃R phosphorylation was increased in Sdc1^−/− mice compared to C57BL/6 mice (Figure 7A). In contrast, IP₃R phosphorylation was decreased in aorta from Gpc1^−/− compared to CD1 mice (Figure 7E). NE challenge led to a non-significant increase in IP₃R activity in both strains without changing the basal pattern of lower IP₃R activity in Gpc1^−/− (Figures 7B and 7F). Total IP₃R expression was similar between aortas of Sdc1^−/− and C57BL/6 mice and aortas of Gpc1^−/− and CD1 (Figures 7C and 7G). After NE challenge a decrease in total IP₃R expression was observed only in Gpc1^−/− (Figures 7D and 7H).

Figure 7: Contrasting IP₃R activity in Gpc1^−/− and Sdc1^−/− aortic rings.

IP₃R phosphorylation is (A, B) increased in aortic rings isolated from Sdc1^−/− treated with norepinephrine (NE) or PBS (Sham) for fourteen days. IP₃R phosphorylation is (E, F) decreased in aortic rings isolated from Gpc1^−/− Sham or NE-treated. Total IP₃R expression is not altered in (C, D) aortic rings of Sdc1^−/− (Sham or NE-treated) and is (G, H) only decreased in in aortic rings isolated from Gpc1^−/− NE-treated. Changes in brightness and contrast were equally applied over the entire representative western blot images. Data represent the mean ± SEM of the experiments, and n=5 represents the number of aortic rings used in the experiments. *p < 0.05 statistical difference between Sdc1^−/− vs C57BL/6 or Gpc1^−/− vs CD1 (Student’s t-test).

4. Discussion

Increased adrenergic tone, circulating norepinephrine levels and vascular dysfunction are hallmarks of essential hypertension (Engelman, Portnoy & Sjoerdsma, 1970; Ghiadoni, Taddei & Virdis, 2012; Grassi, Mark & Esler, 2015; Missouris, Markandu, He, Papavasileiou, Sever & MacGregor, 2016; Taddei, Virdis, Ghiadoni, Versari & Salvetti, 2006). The underlying mechanisms that result in exacerbated adrenergic signaling during hypertensive disease are unresolved. Herein, we reveal glypican 1 as a unique piece of this complex puzzle that determines noradrenergic hypertension progression. In addition, the role of HSPG on systemic blood pressure had not been explored. We show that the global deletion of two different HSPG, syndecan 1 and glypican 1, results in a similar phenotype of lower basal systemic arterial pressure.

In the endothelium, Sdc1^−/− and Gpc1^−/− have decreased eNOS activity and NOx production. These findings are supported by studies that characterized heparan sulfates as an upstream regulator of Kir channels on eNOS-dependent flow-mediated dilation, (Fancher et al., 2020) and corroborate studies showing that Gpc1^−/− have decreased eNOS activity (Mahmoud, Mayer, Cancel, Bartosch, Mathews & Tarbell, 2020). The reduction on eNOS activity is associated with decreased eNOS expression only in Sdc1^−/− mice. This suggests that the mechanisms whereby syndecan 1 and glypican 1 regulate eNOS may differ. Although the global genetic deletion of Sdc1 and Gpc1 results in a localized decrease in eNOS signaling and NO availability, this does not seem to contribute to the regulation of systemic arterial pressure as both knockout strains show decreased basal MAP, SBP and DBP. This could be associated to altered sensitivity to NO which was ruled out by a similar vasodilatory response to sodium nitroprusside (exogenous NO donor). Altogether, these data indicate that the systemic phenotype of lower systemic MAP does not derive from altered NO signaling.

To further explore the role of syndecan 1 and glypican 1 on the regulation of the vascular tone, we performed concentration-response curves to norepinephrine. The maximal responses to NE where changed in opposite directions in aortic rings from Sdc1^−/− as compared to Gpc1^−/− mice. Whereas Sdc1^−/− mice showed increased vasoconstriction to norepinephrine, Gpc1^−/− showed decreased vasocontractile response. The vasoconstrictive response is partly dependent on the endothelium only in Sdc1^−/− as the maximum effect of norepinephrine was decreased after endothelium removal. This suggests that the cell-to-cell communication between the endothelium and the vascular smooth muscle (VSM) layers differs in these knockout strains.

The deletion of Gpc1 results in a decrease in the vascular response to norepinephrine. The increase in plasma norepinephrine levels caused by the hypertensive challenge used in this study did not translate in rescue of the vasoconstrictive response observed in wild-type mice. Interestingly, the expression of the adrenoceptors α1D-AR and α2B-AR, the main isoforms expressed in aorta, were not altered by norepinephrine challenge, and it was similar in both knockout strains. Altogether, these data suggest that glypican 1 may interfere with noradrenergic signaling downstream of the adrenoceptors involved.

As norepinephrine induces VSM contraction by the release of intracellular calcium (Lagaud, Randriamboavonjy, Roul, Stoclet & Andriantsitohaina, 1999; Leijten, Saida & van Breemen, 1985), we assessed the contractile response to KCl, known to cause calcium release from the sarcoplasmic reticulum via activation of voltage-gated calcium channels in the VSM layer by depolarization of the cell membrane (Ratz, Berg, Urban & Miner, 2005). We observed that only Gpc1^−/− had a decreased response to KCl, suggesting that the decreased response to norepinephrine in Gpc1^−/− is associated to altered calcium signaling. This effect was not altered by endothelium removal, suggesting that decreased contractile response is associated to a VSM mechanism/phenotype. Contrarily, endothelium-intact aortic rings from Sdc1^−/− mice had increased response to KCl in intact aortic rings, which was decreased by endothelium removal. These data support the hypothesis that Syndecan 1, but not Glypican 1, signals via a potential EDCF.

Classical pharmacological tools were used to check calcium entry and release from the sarcoplasmic reticulum (SR) in the VSM layer. To assess the contribution of extracellular calcium to the vasocontractile response in Sdc1^−/− and Gpc1^−/− we assessed the vasocontractile response to phenylephrine in the presence of nifedipine (voltage-gated L-type calcium channels blocker). Nifedipine reduced phenylephrine-induced contraction in Sdc1^−/− and abolished in Gpc1^−/− before and after norepinephrine challenge. These data suggest that the absence of syndecan 1 and glypican 1 impaired Ca²⁺ content release from the SR, that is more evident in Gpc1^−/−, consequently the vasoconstriction in vessels from these mice will be blunted or reduced when stimulated with adrenergic agonist. This was observed in parallel to the modulation of blood pressure: Sdc1^−/− have partial increase in blood pressure when chronically treated with NE and Gpc1^−/− do not show any increase in blood pressure with NE treatment. Remarkably, data from Gpc1^−/− aortas show the cascade of events that release intracellular Ca2+ from the SR in these mice is disrupted.

The role of HSPG on calcium dynamics is unresolved. The mechanically-induced calcium transient in endothelial cells is significantly reduced when heparan sulfate chains are digested by heparinase (Klinger, Pichette, Sobolewski & Eckmann, 2011). This effect was replicated by shedding of syndecan 1 ectodomain by phorbol 12-myristate 13-acetate (PMA). The syndecan family were associated to the activation of ion channels by mechanical stretch (Gopal et al., 2015; Mitsou, Multhaupt & Couchman, 2017), including TRPC4 and TRPC7 (calcium-permeable channels). Particularly, syndecan 4 has a specific region in its variable intracellular domain that forms a twisted-clamp dimeric structure (Lee, Oh, Woods, Couchman & Lee, 1998) that binds both the lipid phosphatidylinositol 4,5-bisphosphate and also protein kinase Cα (Oh, Woods, Lim, Theibert & Couchman, 1998). The role of glypican 1 on the regulation of calcium mobility in vascular cells has barely been investigated (Brucato, Bocquet & Villers, 2002). Our data show that Syndecan 1 and Glypican 1 play a key role on calcium dynamics in the VSM layer. When aortic rings were stimulated with one single dose of caffeine, we detected an increased response exclusively in Gpc1^−/− rings, revealing that Gpc1^−/− have increased intracellular Ca²⁺ storage. This suggests that the overall contractile response to exogenous calcium is impaired in both knockout strains. We hypothesize that this could result from reduced Ca2+ transport (by the decreased expression or activity of the many proteins that control calcium mobility at the membrane level, e.g., ion channels and transporters). This would provide a smaller macroscopic conductance to Ca²⁺ that could trigger a positive feedback response to try to restore homeostatic calcium levels. This may be the reason why we see increased alpha adrenoceptor, SERCA and CaM expression.

To further explore the pathway regulated by HSPG, we assessed the contractile response to phenylephrine in the presence of nifedipine in Krebs nominal zero calcium. Gpc1^−/− uniquely showed a blunt response to phenylephrine suggesting that glypican 1 is a positive regulator of phenylephrine-induced calcium-release in the VSM layer. To pinpoint the downstream effector of glypican 1 on the VSM layer we assessed whether glypican 1 regulates IP₃R activity, a receptor that mediates Ca2+ release from the SR after adrenergic stimulation. Notably, Gpc1^−/− exclusively showed decreased IP₃R activity that is not changed with norepinephrine challenge, supporting our findings of glypican 1 being a positive regulator of phenylephrine-induced calcium-release. Remarkably, Sdc1^−/− showed an opposite control of phenylephrine-induced calcium-release and IP₃R signaling. Aortic rings isolated from Sdc1^−/− partly contracted in response to phenylephrine and showed increased IP₃R activity in control and hypertensive conditions, suggesting that syndecan 1 can be a negative regulator of phenylephrine-induced calcium-release.

Altogether, these data indicate that glypican 1, but not syndecan 1, is required for IP₃ signaling in the VSM. Through this pathway, glypican 1 function as a trigger for vasoconstriction and development of noradrenergic hypertension. Our findings support previous reports that show that IP₃R activity is required for VSM contractility playing an essential role in blood pressure regulation and contributing to hypertensive disease (Abou-Saleh et al., 2013; Lin et al., 2016; Lin et al., 2019; Yuan et al., 2016).

In summary, syndecan 1 and glypican 1 regulate vascular tone by distinct pathways. This is associated with altered eNOS signaling in the endothelium and altered IP₃R signaling and calcium dynamics in the VSM. Glypican 1 stands out as a signaling checkpoint of noradrenaline effects on vasocontractility and systemic blood pressure in mice. These findings may translate to human essential hypertension, known to be linked to increased adrenergic tone (Cruickshank, 2017; Engelman, Portnoy & Sjoerdsma, 1970; Missouris, Markandu, He, Papavasileiou, Sever & MacGregor, 2016; Watson et al., 2019). The findings presented in this study have the technical limitations of having used indirect measurement of blood pressure and to study vasoreactivity in conductance vessels due to equipment needs. A deeper mechanistic and molecular analysis using primarily isolated cells from our knockout animals could strengthen our findings.

5. Conclusions

Glypican 1 is a trigger for the development of noradrenergic hypertension that acts via IP₃R- and calcium-dependent signaling pathways. Glypican 1 may be a potential target for the development of new therapies for resistant hypertension or conditions where norepinephrine levels are increased.

Highlights.

• Glypican 1 is a trigger for the development of noradrenergic hypertension that acts via IP₃R- dependent signaling pathways.

• Syndecan 1 and glypican 1 regulate vascular tone in basal conditions.

• Syndecan 1 and glypican 1 activate different signaling pathways with syndecan 1 acting on the endothelium and media layer and glypican 1 exclusively in the vascular smooth muscle layer.

• Glypican 1 may be a potential target for the development of new therapies for resistant hypertension or conditions where norepinephrine levels are increased.

Non-standard Abbreviations and Acronyms

ACh

Acetylcholine

CaCl₂

Calcium Chloride

CICR

Calcium Induced Calcium Release

eNOS

Endothelial nitric oxide synthase

Gpc1

Glypican 1

Gpc1 ^−/−

Glypican 1 knockout

HSPG

Heparan sulfate proteoglycans

NO

Nitric Oxide

NOx

Nitrogen Oxides

NE

Norepinephrine

PMA

Phorbol 12-myristate 13-acetate

PBS

Phosphate Buffer Saline

PE

Phenylephrine

KCl

Potassium Chloride

SERCA

Sarco/endoplasmic reticulum Ca²⁺-ATPase

SR

Sarcoplasmic Reticulum

SNP

Sodium nitroprusside

Sdc1

Syndecan 1

Sdc1 ^−/−

Syndecan 1 knockout

7.1. Data availability statement:

The data that support the findings of this study are available from the corresponding author upon reasonable request.