Arterial α₂-Na⁺ pump expression influences blood pressure: lessons from novel, genetically engineered smooth muscle-specific α₂ mice

Genetically decreased or increased, mouse smooth muscle-specific α₂-Na⁺ pump expression effects on blood pressure are attenuated by compensatory changes in arterial Ca²⁺ transporter expression. This implies that the increased Ca²⁺ transporter expression observed in many hypertension models is not compensatory but, rather, contributes significantly to the blood pressure elevation.

Abstract

Arterial myocytes express α₁-catalytic subunit isoform Na⁺ pumps (75–80% of total), which are ouabain resistant in rodents, and high ouabain affinity α₂-Na⁺ pumps. Mice with globally reduced α₂-pumps (but not α₁-pumps), mice with mutant ouabain-resistant α₂-pumps, and mice with a smooth muscle (SM)-specific α₂-transgene (α₂^SM-Tg) that induces overexpression all have altered blood pressure (BP) phenotypes. We generated α₂^SM-DN mice with SM-specific α₂ (not α₁) reduction (>50%) using nonfunctional dominant negative (DN) α₂. We compared α₂^SM-DN and α₂^SM-Tg mice to controls to determine how arterial SM α₂-pumps affect vasoconstriction and BP. α₂^SM-DN mice had elevated basal mean BP (mean BP by telemetry: 117 ± 4 vs. 106 ± 1 mmHg, n = 7/7, P < 0.01) and enhanced BP responses to chronic ANG II infusion (240 ng·kg⁻¹·min⁻¹) and high (6%) NaCl. Several arterial Ca²⁺ transporters, including Na⁺/Ca²⁺ exchanger 1 (NCX1) and sarcoplasmic reticulum and plasma membrane Ca²⁺ pumps [sarco(endo)plasmic reticulum Ca²⁺-ATPase 2 (SERCA2) and plasma membrane Ca²⁺-ATPase 1 (PMCA1)], were also reduced (>50%). α₂^SM-DN mouse isolated small arteries had reduced myogenic reactivity, perhaps because of reduced Ca²⁺ transporter expression. In contrast, α₂^SM-Tg mouse aortas overexpressed α₂ (>2-fold), NCX1, SERCA2, and PMCA1 (43). α₂^SM-Tg mice had reduced basal mean BP (104 ± 1 vs. 109 ± 2 mmHg, n = 15/9, P < 0.02) and attenuated BP responses to chronic ANG II (300–400 ng·kg⁻¹·min⁻¹) with or without 2% NaCl but normal myogenic reactivity. NCX1 expression was inversely related to basal BP in SM-α₂ engineered mice but was directly related in SM-NCX1 engineered mice. NCX1, which usually mediates arterial Ca²⁺ entry, and α₂-Na⁺ pumps colocalize at plasma membrane-sarcoplasmic reticulum junctions and functionally couple via the local Na⁺ gradient to help regulate cell Ca²⁺. Altered Ca²⁺ transporter expression in SM-α₂ engineered mice apparently compensates to minimize Ca²⁺ overload (α₂^SM-DN) or depletion (α₂^SM-Tg) and attenuate BP changes. In contrast, Ca²⁺ transporter upregulation, observed in many rodent hypertension models, should enhance Ca²⁺ entry and signaling and contribute significantly to BP elevation.

NEW & NOTEWORTHY

Genetically decreased or increased, mouse smooth muscle-specific α₂-Na⁺ pump expression effects on blood pressure are attenuated by compensatory changes in arterial Ca²⁺ transporter expression. This implies that the increased Ca²⁺ transporter expression observed in many hypertension models is not compensatory but, rather, contributes significantly to the blood pressure elevation.

arterial smooth muscle (SM) cells (ASMCs) from rodents and humans express two types of Na⁺ pumps (11, 28, 57). The majority of pumps (∼75–80%) have an α₁-catalytic subunit, which, in rodents, has unusually low ouabain affinity (3, 36); the remainder are high ouabain affinity α₂-Na⁺ pumps. α₁-Pumps (“housekeepers”) maintain the low intracellular Na⁺ concentration in most of the cytoplasm (1). The more sparse α₂-pumps colocalize with plasma membrane (PM) Na⁺/Ca²⁺ exchanger 1 (NCX1) and sarcoplasmic reticulum (SR) Ca²⁺ pumps [sarco(endo)plasmic reticulum Ca²⁺-ATPase 2 (SERCA2)] at PM-SR junctions in ASMCs (21, 28, 42, 51). As a result, the functional coupling of these transporters helps regulate local cytosolic Na⁺ and Ca²⁺ concentrations and Ca²⁺ signaling (1, 25, 31, 40, 42).

A number of studies have linked the α₂-Na⁺ pump and its high-affinity ligand, endogenous ouabain (EO), to blood pressure (BP) regulation. In mice with one null mutant α₂-allele (α₂^+/−) (20), arterial α₂-Na⁺ pump expression is reduced by ∼50%, but the total Na⁺ pump number (mostly α₁) is still ∼90% of normal (50, 57); these mice have modestly elevated basal BP (57). In contrast, mice with one null α₁-allele (α₁^+/−) express many fewer total Na⁺ pumps in ASMCs (∼60% of normal) but have normal basal BP (57). In addition, plasma EO is elevated in many humans with essential hypertension or mineralocorticoid hypertension (33, 39, 49), in several rodent hypertension models (14, 15, 24, 27), and in pregnant rats (19). Indeed, prolonged ouabain administration induces hypertension in rodents (11, 12, 34). Furthermore, genetically engineered mice with ouabain-resistant α₂-Na⁺ pumps (α₂^R/R) are resistant to adrenocorticotropic hormone-induced hypertension (10, 30) and ouabain-induced hypertension (11). In addition, α₂^R/R dams have unusually low BP during late pregnancy (38). Thus, there is ample reason to characterize the influence of arterial α₂-Na⁺ pumps on BP. To this end, we studied mice that were genetically engineered to either decrease or increase expression of α₂-Na⁺ pumps in SM.

The present report describes a new mouse model: SM-specific knockdown of α₂-Na⁺ pumps, by ∼60% in ASMCs, generated with a dominant negative (DN), nonfunctional α₂-transgene (α₂^SM-DN mice). We then compared basal BP and BP responses to chronic ANG II infusion and high dietary salt (HS) in these α₂^SM-DN mice and in mice that markedly overexpress α₂-Na⁺ pumps in their ASMCs due to a SM-specific normal α₂-transgene (α₂^SM-Tg) (43). The results reveal that basal BP and the responses to ANG II and HS are significantly, albeit modestly, increased in α₂^SM-DN mice and modestly decreased when α₂-Na⁺ pump expression is increased in ASMCs.

We attribute the relatively small magnitude of these BP changes to observed concurrent modifications in the expression of ASMC NCX1 and other Ca²⁺ transporters that likely compensate for the alterations in α₂ expression by minimizing changes in cell Ca²⁺ homeostasis. This view is bolstered by reports that basal BP is inversely correlated with ASMC NCX1 expression in mice with SM-specific genetic modification of NCX1 expression (18, 53, 58). We discuss how this helps explain the finding that the BP phenotypes in SM-α₂ engineered mice are only modestly different from those of wild-type (WT) control mice even though α₂-Na⁺ pump activity apparently has a major influence on Ca²⁺ homeostasis and contractile regulation. This illustrates why attempts to deduce details about the significance of individual proteins in BP regulation from single gene-engineered mouse models may be misleading unless related compensatory changes are also considered. Finally, we describe how an understanding of these compensatory responses to primary arterial defects in α₂-Na⁺ pumps provides insights into the significance of arterial myocyte Ca²⁺ transporter protein reprogramming that is observed in many hypertension models.

MATERIALS AND METHODS

Experimental Animals

Ethical approval.

All mouse protocols were approved by the Institutional Animal Care and Use Committee of the University of Maryland School of Medicine. The protocol used for the generation of α₂^SM-DN mice was approved by the Institutional Animal Care and Use Committee of Cornell University College of Veterinary Medicine.

Mice were housed in an Association for Assessment and Accreditation of Laboratory Animal Care-approved facility with a 12:12-h light-dark cycle. Mice were unrestrained and had free access to food and water.

Generation of Na⁺ pump α₂-subunit NH₂-terminal constructs.

We generated the DNA sequence corresponding to the NH₂-terminal 1–120 amino acids (including the cytoplasmic domain and first transmembrane helix; see Fig. 1A) of the α₂-isoform of the Na⁺ pump catalytic subunit with a COOH-terminal Flag epitope tag (Fig. 1B). This sequence was cloned into pFlag-CMV-5a vector (Sigma Chemical, St. Louis, MO) by a PCR-based approach. A COOH-terminal Flag tag was then also cloned in using primer pairs (5′-TTAAAGCTTCGCCACCATGGGTCGTGGGGCA-3′ and 5′-CGCCGAATTCCCTACTTGTCATCGTCGTCCT-3′) to create a DN construct with a NH₂-terminal Kozak sequence [α₂(1–120)flag; Fig. 1]. Expression of this nonfunctional construct, which contains the normal α₂ sorting sequence, replaces native, functional α₂-Na⁺ pumps in arterial myocytes and is therefore DN (51).

Fig. 1.

A: Na⁺ pump α- and β-subunit organization in the plasma membrane (PM). The locations of the antibody epitope amino acid sequences used for the generation of isoform-specific antibodies are indicated on the α-subunit. They are all located within a short segment on the large cytoplasmic loop between transmembrane segments TM4 and TM5 (41). B: structure of the Flag-tagged Na⁺ pump α₂-subunit NH₂-terminal segment used to generate α₂ dominant negative (DN) mice. This construct is expressed in the PM and causes greatly reduced PM expression of native α₂-Na⁺ pumps (the DN effect) in arterial smooth muscle (SM) cells (ASMCs) (51).

Gene targeting strategy and generation of α₂^SM-DN mice.

To generate α₂^SM-DN mice, the α₂(1–120)flag transgene was engineered with a HindIII site and Kozak sequence at the 5′-end and an EcoRI site at the 3′-end (Fig. 2A). The PCR product was digested with HindIII and EcoRI and then subcloned into pBluescript PA plasmid upstream of two polyA sites (Fig. 2B). XhoI digestion of the plasmid yielded a 1-kb α₂(1–120)flag-pA fragment, and the fragment was then introduced into the pGEX4T-SM-myosin heavy chain (MHC) plasmid at the SalI site immediately downstream of the SM-MHC promoter (Fig. 2C) (32). The transgene sequence was verified by sequencing the 3′- and 5′-ends of the cDNA. The successful recombinant plasmid was linearized by digestion with NotI, purified, and injected into the pronucleus of B6/D2F2 fertilized mouse eggs, which were then implanted into pseudopregnant female mice (at the Cornell Core Transgenic Mouse Facility, Cornell University, Ithaca, NY). Founder α₂^SM-DN mice generated by this procedure were identified by PCR and Southern blot analyses and then subsequently bred to WT C57Bl/6 mice. The presence of Na⁺ pump α₂-DN was verified by genotyping of α₂^SM-DN progeny using PCR primer pairs (SM-MHC forward 5′-ATCTTCTAACTGGGTGGTGGTGGT-3′ and Flag reverse 5′-TGTAATCGGATCCTTCGTCCTCCA-3′) to yield a 500-bp product.

Fig. 2.

Diagram of the gene strategy used to generate mice that express the truncated NH₂-terminal DN segment of the Na⁺ pump α₂-subunit. Expression of the DN segment was under the control of the SM-specific myosin heavy chain (MHC) promoter. See text for details.

α₂^SM-DN mice were viable and exhibited no obvious behavioral or anatomic abnormalities. They bred normally and were inbred to generate mice with a maximum number of copies of the transgene. On two back crosses with WT C57Bl/6 mice, all offspring expressed the transgene (as verified by PCR).

Transgenic α₂^SM-Tg mice.

α₂^SM-Tg mice (on a FVB/n background) were generated at the University of Cincinnati (43). Heterozygote breeders (α₂^SM-Tg/−) were used by the University of Maryland investigators to generate two lines: one designated WT FVB/n, which had no transgene, and an inbred α₂^SM-Tg line to maximize transgene expression. On two back crosses with WT FVB/n mice, >90% of the offspring expressed the transgene (as identified by PCR).

BP Measurements

Telemetry.

In most experiments, BP was monitored continuously in awake, free-moving mice using implanted BP transmitters (TA11PA-C10, Data Science, St. Paul, MN). WT and transgenic mice (16–18 wk old) were anesthetized with 2–2.5% isoflurane supplemented with 100% O₂. The right carotid artery was isolated through a midline neck incision and was ligated near its bifurcation onto internal and external branches. A small hole was made with a 25-gauge needle, and the transmitter catheter was inserted and threaded forward to the aortic arch. The transmitter body was passed subcutaneously to a subcutaneous pocket in the left abdominal flank. Seven to ten days after surgery, 24-h BPs were recorded by telemetry.

Tail cuff.

In a preliminary experiment, systolic arterial BP and heart rate were screened in conscious, restrained α₂^SM-DN mice with a commercial tail-cuff pulse BP system (model MC4000, Hatteras Instruments, Cary, NC). Many of these “first generation” α₂^SM-DN mice were then used as breeders; others were euthanized for tissue harvesting and immunoblot analysis. Ten BP readings were obtained, and the measurements were repeated on 3 consecutive days. These data were averaged to obtain a single basal systolic BP value for each mouse. The tail-cuff technique was validated by simultaneously recording BP by telemetry in three mice implanted with BP transmitters. The correlation power coefficient (r²) from the linear regression was 0.87, indicating good agreement between the two methods.

The BP data collection was performed “double blind”: the individual measuring the BP did not know the genotype. Animal code numbers were matched with genotype and BP after the data were collected. Male mice were used for telemetric BP data collection to simplify comparison with a previous study of α₂^SM-Tg male mice (43).

ANG II and Dietary Protocols

Once basal BPs were obtained, mice were anesthetized with 1.5% isofluorane inhalation and an ANG II-filled (acetate salt, Bachem Americas, Torrance, CA) or vehicle-filled osmotic minipump (Alzet model 2004, Durect, Cupertino, CA) was then implanted. It was inserted into a subcutaneous pocket in the right ventral abdominal wall via a small right groin incision.

The standard diet contained 0.4% NaCl; high-salt (HS) diets contained either 2% or 6% NaCl, as indicated in the results. All diets were purchased from Dyets (Bethlehem, PA).

Myogenic Tone and Reactivity

Details of the methods have been previously published (58). In brief, segments of fourth-order arteries (mean passive diameter ≈ 140 μm) were dissected from the mesenteric artery arcade on the external surface of the small intestine. Segments were cannulated at both ends, intraluminal pressurize was raised to 70 mmHg (no internal flow), and segments were superfused with physiological salt solution (PSS) at 35–37°C. When myogenic tone (MT) developed, intraluminal pressure was lowered to 10 mmHg and raised stepwise in 20 mmHg increments to 130 mmHg to determine myogenic reactivity (MR). Segments were then equilibrated in Ca²⁺-free (0Ca) PSS, and the sequence was repeated to determine the passive diameter at each pressure level. During this process, external diameter was continuously monitored with a custom online edge detector based on a Nikon TMS microscope (Nikon, Melville, NY). Artery segments that developed significant leaks were discarded.

Data are presented as graphs of artery diameter versus intraluminal pressure in normal Krebs solution and 0Ca PSS (passive diameter). The difference between the two curves is the MR.

Immunoblot Analysis

Preparation of tissue extracts.

The aorta, urinary bladder, heart, and brain from α₂^SM-DN mice and the aorta from α₂^SM-Tg mice were minced in homogenization buffer (HB1) at 4°C with a Polytron homogenizer (Brinkmann Instruments, Westbury, NY). The suspension was then centrifuged for 20 min at 16,000 g and 4°C. The pellet (cell membranes) from this centrifugation step was resuspended in HB2 buffer [similar to HB1 but with the addition of the nonionic detergent IGEPAL CA-630 (1%), Sigma-Aldrich] and incubated for 1 h on ice. The lysate was then centrifuged for 20 min at 16,000 g and 4°C. The supernatant was collected and stored at −80°C until use. The protein concentration was determined with a bicinchoninic acid assay (Bio-Rad, Hercules, CA) with BSA as a standard.

Immunoblot analysis.

One volume of protein preparation was mixed with an equal volume of 2× Laemmli Sample Buffer (Bio-Rad) containing 5% 2-mercaptoethanol, and proteins were separated on SDS-PAGE minigels (Bio-Rad). The protein on the gel was then transferred to a polyvinylidene difluoride membrane with an iBlot Dry Blotting System (Invitrogen, Carlsbad, CA). The membrane was blocked with 5% nonfat dry milk in Tris-buffered saline with Tween 20 for 2 h and subjected to immunoblot analysis using isoform-specific monoclonal or polyclonal antibodies. Details of our methods have been previously published (26). Immunoblot band densities were normalized to the respective loading controls (α-actin in the heart or β-actin in the aorta, bladder, and brain) and then compared with expression in the respective WT tissues (equal to 100%).

Solutions and Antibodies

PSS contained (in mM) 112 NaCl, 25.7 NaHCO₃, 4.9 KCl, 2.0 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, 11.5 d-glucose, and 10 HEPES (pH 7.4; gas composition: 5% O₂-5% CO₂-90% N₂). In 0Ca PSS, CaCl₂ was omitted and 0.5 mM EGTA was added.

HB1 contained 140 mM NaCl, 10 mM NaH₂PO₄, 2 mM EDTA, 10 mM NaN₃, and protease inhibitor cocktail tablets (2 tablets/50 ml, Roche Diagnostics, Mannheim, Germany). Tris-buffered saline with Tween 20 solution contained 50 mM Tris, 150 mM NaCl, and 0.05% Tween 20 at pH 7.6 (adjusted with HCl).

The following antibodies were used for immunoblot analysis: 1) polyclonal anti-Na⁺ pump α₁-subunit raised against the NASE epitope (gift from T. Pressley, Texas Tech University, Lubbock, TX; see Fig. 1A); 2) polyclonal anti-Na⁺ pump α₂-subunit raised against the HERED epitope (Millipore, Billerica, MA; see Fig. 1A); 3) monoclonal anti-NCX1 antibody (R3F1, gift from K. D. Philipson, University of California-Los Angeles, Los Angeles, CA); 4) monoclonal anti-SERCA2 (Affinity BioReagents, Golden, CO); 5) polyclonal anti-PMCA1 (Thermo Scientific, Rockford, IL); 6) monoclonal anti-Flag (Sigma-Aldrich); 7) polyclonal anti-transient receptor potential C-type channel protein 6 (TRPC6; Sigma-Aldrich); 8) monoclonal anti-cardiac α-actin (Sigma-Aldrich); 9) monoclonal anti-SM β-actin (Sigma-Aldrich); and 10) monoclonal anti-GAPDH (Abcam, Cambridge, MA).

Statistical Analyses

Data are expressed as means ± SE; n values denote numbers of mice. Comparisons of BP data were made using ANOVA or Student's paired or unpaired t-tests, as appropriate. Differences were considered significant at P < 0.05.

RESULTS

Genotype Verification of α₂-Na⁺ Pump Expression and Basal BP in Genetically Engineered Mice

α₂^SM-DN mice.

Immunoblot data (Fig. 3) demonstrated that α₂^SM-DN mice expressed the α₂(1–120)flag DN construct (identified with anti-Flag antibodies) in the aorta and urinary bladder, in which SM is prevalent. The DN construct was not detected in the heart or brain. Native α₂ was identified with antibodies to the HERED epitope because this epitope is present in the native protein (41) but not in the DN construct (Fig. 1A). Native α₂ expression was reduced by ∼50% in the urinary bladder and by ∼70% in the aorta of α₂^SM-DN mice, but normal levels were expressed in the heart and brain (Fig. 3). Thus, expression of the DN construct was SM specific. Although the heart and brain are highly vascularized, Na⁺ pumps (and NCX1) are ∼5- to 15-fold more prevalent in cardiac myocytes, glia, and neurons than in arteries (see, e.g., Refs. 16, 17, 35, and 57)¹. Therefore, downregulation of α₂ in cardiac and cerebral blood vessels would not have been detected in these heart and brain whole tissue immunoblots.

Fig. 3.

Several Ca²⁺ transporters are markedly downregulated in SM of α₂^SM-DN mice. A: representative immunoblots of aorta, urinary bladder, brain, and heart membrane proteins from wild-type (WT) and α₂^SM-DN mice. Na⁺ pump α₁- and α₂-isoforms were detected with antibodies raised against the amino acid sequences (NASE and HERED, respectively); see Fig. 1A and Ref. 41. Expressed DN peptide was detected with anti-Flag antibodies (see Fig. 1B). Expression of α₂-Na⁺ pumps but not α₁-Na⁺ pumps, Na⁺/Ca²⁺ exchanger 1 (NCX1), sarco(endo)plasmic Ca²⁺-ATPase 2 (SERCA2), PM Ca²⁺-ATPase 1 (PMCA1) and transient receptor potential C-type channel protein 6 (TRPC6) proteins were all reduced in the aorta and urinary bladder of α₂^SM-DN mice. Expression of all these proteins in the brain and heart was comparable in WT and α₂^SM-DN mice. Loading controls were α-actin (heart) or β-actin (aorta, bladder, and brain). B: summarized aorta and urinary bladder data from 3 to 8 mice. Values were normalized to the loading control (α-actin in the heart and β-actin in the aorta, bladder and brain) and then compared with expression in the respective WT tissues (equal to 100%). Note that α₁ in the α₂^SM-DN mouse aorta and bladder and PMCA1 in the bladder were not markedly different from the WT group. Bar graphs for the brain and heart are not shown because expression levels for all tested proteins in α₂^SM-DN mice were within 100 ± 10% of WT mice (A).

Note that the DN construct has only one transmembrane segment and is missing the cation binding and translocation sites as well as the α₂ catalytic machinery (Fig. 1) and is thus nonfunctional. Nevertheless, this construct contains the α₂ sorting signal and is expressed in the PM (51). The reduced expression of native α₂ in the aorta and bladder indicates that in vivo, as well as in vitro (51), expression of the DN construct downregulates native α₂ in SMs.

Basal BP was modestly but significantly higher in α₂^SM-DN mice than in WT (C57Bl/6) control mice. Figure 4A shows systolic BPs (tail cuff) from the first male and female offspring of the founder mice. Figure 4B shows mean BP (MBP) data (telemetry) from a subsequent α₂^SM-DN generation of male mice.

Fig. 4.

Basal blood pressure (BP) is elevated in α₂^SM-DN mice. A: systolic BP (tail cuff) data from the first male and female offspring of the α₂^SM-DN colony founder mice and their WT C57Bl/6 controls. BPs determined on 3 consecutive days were averaged to obtain a single value for each mouse. Male mice: n = 10 WT mice and 11 DN mice, **P < 0.01; female mice: n = 4 WT mice and 14 DN mice, *P < 0.03. B: basal 24-h mean BPs (MBPs) of male WT mice and a later generation of male α₂^SM-DN mice recorded by telemetry. MBPs from 3 consecutive days were averaged to obtain a single value for each mouse. n = 7 WT mice and 7 DN mice. **P < 0.01.

α₂^SM-Tg mice.

Overexpression of α₂-Na⁺ pumps in the aorta and gastric antrum, but not in the heart, was described in the initial study (43) of α₂^SM-Tg male mice, which were modestly hypotensive (tail cuff). We confirmed that α₂^SM-Tg mice overexpress the α₂-Na⁺ pump in the aorta (Fig. 5A) and have lower BP (telemetry) than WT-FVB/n control mice (Fig. 5B).

Fig. 5.

Expression of α₂-Na⁺ pumps is increased and basal MBP is reduced in mice with a SM-specific α₂-transgene (α₂^SM-Tg mice). A: immunoblots of aorta membrane proteins from WT FVB/n control and α₂^SM-Tg mice. B: basal 24-h MBPs of male WT and α₂^SM-Tg mice were recorded by telemetry. MBPs from 3 consecutive days were averaged to obtain a single value for each mouse. n = 9 WT mice and 15 Tg mice. *P < 0.02.

Ca²⁺ Transporter Expression Correlates With α₂-Na⁺ Pump Expression

Prior immunoblot studies (42, 43) of the aorta and gastric antrum in α₂^SM-Tg mice revealed that not only the α₂-Na⁺ pumps but also other Na⁺ and Ca²⁺ transporters were affected in these SM-containing tissues. α₁-Na⁺ pump, NCX1, SERCA2, and PMCA proteins were all upregulated in both the aorta and gastric antrum. Contraction-related proteins (actin and 20-kDa myosin light chain) were, however, unaffected (42, 43).

Expression of α₁-Na⁺ pumps is negligibly affected in α₂^SM-DN mice. However, a number of Ca²⁺ transporter proteins are markedly downregulated in both the aorta and urinary bladder (Fig. 3): NCX1, SERCA2, PMCA1, and TRPC6 (a component of receptor-operated Ca²⁺ channels). Of these, PMCA1 appears to be least affected, especially in the urinary bladder. Again, expression of all these proteins in the heart and brain was normal (Fig. 3A). These α₂^SM-DN and α₂^SM-Tg mouse data indicate that the expression of the several Ca²⁺ transporters in SM is directly correlated with SM-specific α₂-Na⁺ pump expression. The fact that similar changes in Ca²⁺ transporter expression are seen in multiple SMs in the two models suggests a cause-effect relationship between α₂-Na⁺ pump expression and these Ca²⁺ transporter changes. We attribute this to a compensatory effect on Ca²⁺ transport (see discussion).

Effects of ANG II and a HS Diet on BP in α₂^SM-DN and α₂^SM-Tg Mice

α₂^SM-DN mice.

The high basal BP in α₂^SM-DN mice (Fig. 4) raised the possibility that these mice might be “salt sensitive” and that their arteries might be hypercontractile, despite the downregulation of Ca²⁺ transporter proteins in the arteries. To explore this possibility, we measured the changes in BP during a 10-day period on a 6% NaCl (HS) diet (Fig. 6A) and during an 11-day subcutaneous infusion of low-dose ANG II (240 ng·kg⁻¹·min⁻¹; Fig. 6B). Both treatments increased BP more rapidly in α₂^SM-DN mice than in their WT counterparts (Fig. 6). Moreover, both treatments caused a sustained increase in BP in α₂^SM-DN mice but not in WT control mice (Fig. 6). At days 8–10 of treatment, the HS diet increased MBP by 6.1 mmHg in DN mice versus only 0.4 mmHg (i.e., no change from baseline) in WT mice, and ANG II increased MBP by 9.3 mmHg in DN mice vs 6.1 mmHg in WT mice. Thus, expression of the DN construct in SM augmented the responses to both HS and ANG II.

Fig. 6.

α₂^SM-DN mice are more sensitive to high dietary salt (HS) and low-dose ANG II than are WT mice. A: effect of a high-salt (HS; 6% NaCl) diet on 24-h MBP in WT and α₂^SM-DN mice. B: effect of 240 ng·kg⁻¹·min⁻¹ ANG II (subcutaneous infusion by osmotic minipump) on 24-h MBP in WT and α₂^SM-DN mice. n = 5 mice/group for both A and B. *P < 0.05 by ANOVA.

α₂^SM-Tg mice.

Since α₂^SM-Tg mice have reduced basal BP (Fig. 5) (43), we hypothesized that these mice, in contrast to α₂^SM-DN mice, might be “salt resistant” and relatively insensitive to the BP elevating effect of ANG II. Indeed, subcutaneous infusion of a modest dose of ANG II (400 ng·kg⁻¹·min⁻¹) for 12 days elevated MBP in WT control mice by ∼30 mmHg but by only ∼20 mmHg in α₂^SM-Tg mice (Fig. 7A). In other words, the ANG II-induced MBP elevation from the respective baselines was significantly attenuated in Tg mice. Furthermore, the ANG II-induced rise in MBP was slow in these mice. MBP did not peak until 96 h after the ANG II infusion was started in α₂^SM-Tg mice, whereas MBP peaked by 48 h in WT mice.

Fig. 7.

α₂^SM-Tg mice have reduced sensitivity to ANG II and ANG II + HS. A: effects of a moderate dose of ANG II (400 ng·kg⁻¹·min⁻¹ sc) on 24-h MBP in 5 WT mice and 5 α₂^SM-Tg mice. *P < 0.05 by ANOVA. B: effects of 12 days of a low dose of ANG II (300 ng·kg⁻¹·min⁻¹ sc) plus HS (2% NaCl diet) on systolic BP in 4 WT mice and 9 α₂^SM-Tg mice. Systolic BPs in each mouse were the means of three 12-h light period recordings. NS, not significant. **P < 0.01 by ANOVA.

ANG II and HS are “convergent signals” that act on the brain to elevate BP (9, 37); low-dose ANG II was therefore used to sensitize the mice to HS (9), as shown in Fig. 7B. WT and α₂^SM-Tg mice were fed a 2% NaCl (HS) diet during subcutaneous infusion with ANG II (300 ng·kg⁻¹·min⁻¹) for 12 days. This treatment elevated BP modestly in both groups of mice, but the increase in systolic BP in α₂^SM-Tg mice was only ∼60% of that in WT mice. Hence, the responses of α₂^SM-Tg mice to ANG II and to ANG II + HS diet were attenuated, in contrast to the augmented responses in α₂^SM-DN mice. This was reflected in both the rates of the response to ANG II and the magnitudes of the changes in MBP.

MR in Isolated Arteries From α₂^SM-DN and α₂^SM-Tg Mice

Davis and Hill (8) noted that the level of tone in isolated arteries “is often comparable to that observed in the same vessels in vivo.” With this concept in mind, we measured MT as a function of the intralumenal pressure (i.e., MR) in mesenteric small arteries from α₂^SM-DN and α₂^SM-Tg mice. As shown in Fig. 8A, the mean passive diameter of α₂^SM-Tg mouse arteries was slightly smaller than that of their WT FVB/n counterparts. When this was taken into account, MR was virtually identical in α₂^SM-Tg and WT FVB/n mouse arteries (Fig. 8B).

Fig. 8.

Expression of a Na⁺ pump α₂-transgene in SMs does not affect myogenic reactivity (MR) in mouse mesenteric small arteries. A: pressure-diameter curves for WT (FVB/n) and α₂^SM-Tg mouse arteries. B: MR of WT and α₂^SM-Tg mouse arteries. MR is the difference between the passive diameter (PD) and the diameter measured in normal physiological salt solution (PSS). MR data are presented as percent decreases in arterial diameter relative to the PD at 130 mmHg intraluminal pressure (PD₁₃₀), graphed as a function of intraluminal pressure.

The mean passive diameter of arteries from α₂^SM-DN mice was identical to that of their WT C57Bl/6 control counterparts (Fig. 9A). Interestingly, not only was the MR of α₂^SM-DN arteries significantly less than that of control arteries (Fig. 9B), but α₂^SM-DN arteries developed MT much more slowly than did WT arteries. All WT C57Bl/6 arteries developed steady tone ∼25–35 min after arteries were pressurized to 70 mmHg and warmed to ∼36°C, similar to both α₂^SM-Tg and WT FVB/n arteries. In contrast, it took at least 60 min for all six α₂^SM-DN arteries to develop tone, and several arteries required priming with one to two doses of 1 μM phenylephrine for 5 min before MT developed. As discussed below, we attribute this slow tone generation to the greatly reduced NCX1 and SERCA2 expression (Fig. 3).

Fig. 9.

Expression of a Na⁺ pump α₂ DN construct in SMs reduces MR in mouse mesenteric small arteries. A: pressure-diameter curves for WT (C57Bl/6) and α₂^SM-DN mouse arteries. B: MR of WT and α₂^SM-DN mouse arteries (see Fig. 8 for further information). *P < 0.05 by ANOVA.

DISCUSSSION

Basal BP and BP Sensitivity to ANG II and Salt Correlate Inversely With α₂-Na⁺ Pump Expression

Previously, we reported that basal BP was elevated in α₂^+/− mice, in which α₂-Na⁺ pump expression in ASMCs was reduced by ∼50% (57). α₂^+/− mice are global heterozygotes; global homozygous null mutants (α₂^−/−) are born, but the neonates do not survive (20). To investigate further the role of arterial α₂-Na⁺ pumps in BP modulation, we generated mice with a SM-specific reduction of expressed α₂ based on a DN strategy (51). As described in the results, in vivo SM-specific expression of the NH₂-terminal 120-amino acid segment of α₂, a DN construct (51), reduced functional, full-length α₂ expression in arteries by >50% (Fig. 3) and modestly elevated basal BP (Fig. 4). This is consistent with α₂^+/− mouse data (57), albeit seemingly not with the evidence that mice with cardiovascular (CV)-specific α₂ knockout (KO; α₂^CV-KO mice) have normal basal BP (48) (but see How Can We Reconcile the Elevated Basal BP in α₂^SM-DN and α₂^+/− Mice With the Normal Basal BP in α₂^CV-KO Mice? below).

BP may be directly correlated with Ca²⁺ levels and Ca²⁺ signaling in the arteries (13, 57, 58). If this is the case, we reasoned that mice with higher basal BP (α₂^SM-DN) and, presumably, higher arterial Ca²⁺ levels, should exhibit increased sensitivity to ANG II and HS, factors that foster vasoconstriction and hypertension (9, 37). Our observations demonstrated that the α₂^SM-DN mice were, indeed, more sensitive to ANG II and HS than WT mice (Fig. 6).

In contrast, α₂-Na⁺ pump transgenic mice (α₂^SM-Tg), which overexpress the α₂-subunit specifically in SM, including ASMCs, have low basal BP, as measured by tail cuff (43) and by telemetry (Fig. 5). Based on the aforementioned reasoning, these mice may be presumed to have reduced arterial Ca²⁺ levels and, therefore, may be expected to exhibit attenuated responses to ANG II and HS, as observed (Fig. 7). Thus, SM-specific α₂ expression correlated inversely with both basal BP and sensitivity to ANG II and HS in these mouse lines.

This speculation about Ca²⁺-dependent mechanisms in both α₂^SM-DN and α₂^SM-Tg mice obviously requires verification by direct measurements of cytosolic Ca²⁺ concentration. Nevertheless, the results, as shown in Figs. 6 and 7, are consistent with these expectations.

MR Does Not Correlate With SM-Specific Expression of α₂-Na⁺ Pumps

In earlier studies, we observed that MT and MR were correlated directly with basal BP in α₂^+/− mice (57) and in mice with SM-specific KO (53, 58) and SM-specific overexpression (18) of NCX1 (NCX1^SM-KO and NCX1^SM-Tg, respectively). This, plus the aforementioned reasoning about Ca²⁺ homeostasis and the reported association between arterial tone in vivo and in isolated arteries (8), led us to anticipate that MR might be inversely correlated with arterial α₂-Na⁺ pump expression. This was not the case, however. MR was virtually identical in arteries from α₂^SM-Tg mice and their WT control counterparts (Fig. 8B). Furthermore, contrary to expectations, despite the higher basal BP, isolated small arteries from α₂^SM-DN mice had less MR than did control arteries (Fig. 9B).

These findings indicate that the relationship between arterial tone in vivo and in isolated arteries in vitro is more complicated than previously anticipated (8). One factor is that the in vitro tone is “purely” myogenic, whereas in vivo tone normally includes neural, humoral, and endothelial (shear stress) contributions. The myogenic component likely depends directly on myocyte Ca²⁺ homeostasis. MT is generated when an isolated arterial segment is pressurized and warmed to 36°C, and the cytosolic Ca²⁺ concentration rises (57). Because the expression of multiple Ca²⁺ transporters is altered in engineered mice (e.g., Fig. 3) (43), it is difficult to predict the net effect on Ca²⁺ homeostasis of these changes (and, likely, others that have not yet been identified). Also, membrane potential plays an important role in the generation of MT (23), and transporters that affect membrane potential might also be altered in engineered mice. Some of these factors, for example, the greatly reduced expression of NCX1, which mainly mediates Ca²⁺ entry in ASMCs (18, 53, 58), and SERCA2 (Fig. 3), may contribute to the delayed generation of MT (22, 46) in α₂^SM-DN mouse arteries.

Arterial Ca²⁺ Transporter Expression and Function in Mice With Genetically Altered SM-Specific Expression of α₂-Na⁺ Pumps

It is instructive to compare the phenotypes in these mice with primary (targeted) genetic manipulation of α₂-Na⁺ pumps and secondary (untargeted) changes in NCX1 and in mice with primary genetic alteration of NCX1. In the latter, basal BP correlated directly with arterial NCX1 expression (18, 53, 58): engineered downregulation of arterial NCX1 in NCX1^SM-KO mice lowered cytosolic Ca²⁺ concentration and reduced basal MBP by 8–10 mmHg (53, 58). Conversely, SM-specific upregulation of NCX1 in NCX1^SM-Tg mice raised cytosolic Ca²⁺ concentration and increased basal systolic BP by ∼10 mmHg (18). This was attributed to the higher cytosoli Ca²⁺ concentration and enhanced signaling in mice expressing more NCX1, thereby favoring Ca²⁺ entry in ASMCs (18, 58).

In mice with engineered α₂-Na⁺ pumps (α₂^SM-Tg, α₂^SM-DN, and α₂^−/−), however, the expression and function of arterial NCX1 and other Ca²⁺ transporters (e.g., SERCA2, TRPC6, and PMCA1) are also greatly altered, i.e., there are secondary, compensatory changes [Fig. 3 (31, 43)]. In α₂^SM-Tg and α₂^SM-DN models, basal BP correlates inversely with NCX1 expression: in α₂^SM-DN mice, despite the reduced NCX1 expression, basal BP is elevated (Figs. 3 and 4), whereas basal BP is reduced in α₂^SM-Tg mice even though NCX1 is overexpressed (Fig. 5) (42). Also, in α₂^+/− mice, which have elevated basal BP, the fractional SEA0400 (NCX1 blocker)-induced decrease in MT was ∼35% less than in WT control mice (Fig. 9C in Ref. 57), which implies reduced NCX1 activity. Ca²⁺ homeostasis and signaling and, thus, arterial tone and BP are obviously influenced by the expression, in arterial myocytes, of α₂-Na⁺ pumps as well as NCX1 and other Ca²⁺ transporters. These relationships may be complex, but the monogenic mutations of SM α₂ and NCX1 suggest a common thread based on an understanding of the distribution and function of these transporters.

As a result of the functional coupling of α₂ and NCX1 (see Introduction), decreased α₂-Na⁺ pump-mediated Na⁺ extrusion, either because of reduced expression or pump inhibition, raises cytosolic Na⁺ concentration and, via NCX1, cytosolic Ca²⁺ concentration and enhances Ca²⁺ signals and contraction (1, 57). Therefore, NCX1 (and other Ca²⁺ transporters) may be downregulated to compensate for the reduced α₂ expression and tendency to Ca²⁺ overload in α₂^SM-DN mice. Although protein expression was not measured, reduced NCX1 activity in α₂^−/− mouse ASMCs (31) and in α₂^+/− arteries (57) is consistent with this concept. Conversely, NCX1 (and other Ca²⁺ transporters) may be upregulated to compensate for the enhanced α₂ expression and, presumably, reduce local cytosolic Na⁺ concentration and the tendency, therefore, to deplete Ca²⁺ in α₂^SM-Tg mice.

Primary Versus Secondary (Compensatory) Mechanisms That Influence BP in Mice With Genetically Engineered SM-α₂ and SM-NCX1

What is the explanation for this apparently paradoxical relationship between NCX1 expression and basal BP? In NCX1^SM-Tg and NCX1^SM-KO mice, the primary defects are in one Ca²⁺ transport mechanism. Other compensatory mechanisms are therefore required to attenuate Ca²⁺ overload or Ca²⁺ depletion, respectively. In α₂^SM-DN and α₂^SM-Tg mice, the changes in α₂-Na⁺ pump expression/activity alter the local Na⁺ electrochemical gradient driving force on NCX1 (1), thereby favoring NCX1-mediated Ca²⁺ entry and overload or Ca²⁺ extrusion and depletion, respectively. In both α₂ engineered mouse lines, the modifications in NCX1 expression reflect, in part, the myocyte compensatory mechanisms needed to minimize the Ca²⁺ overload or depletion.

Consider, for example, the behavior of small arteries (e.g., passive diameter = 140 μm, with MT at 70 mmHg = 15% of passive diameter; see Figs. 8 and 9). When α₂-Na⁺ pumps are acutely inhibited by ∼50%, e.g., by low-dose ouabain (equivalent to a 50% decrease in α₂ expression), local cytosolic Na⁺ concentration will rise and promote NCX1-mediated Ca²⁺ gain. MT at 70 mmHg will increase by ∼10% (Fig. 3D in Ref. 57) as the diameter declines from 119 to 114 μm. According to Poiseuille's law, at constant cardiac output, this corresponds to an 18% increase in systemic vascular resistance and, thus, an increase in MBP from 105 to 125 mmHg–a large increase! In vivo, of course, rapidly recruited baroreflex mechanisms would be expected to attenuate the rise in MBP, but this is not likely to be sustained. When α₂-Na⁺ pump-mediated transport is congenitally reduced by ∼50%, however, as in α₂^SM-DN mice, slower compensatory mechanisms such as a decrease in NCX1 expression may (should) come into play, thereby chronically minimizing the MBP elevation–as we observed. This scenario illustrates how potentially large changes in Ca²⁺ homeostasis and, thus, arterial constriction, induced here by monogenic changes in ASMC Na⁺ or Ca²⁺ transporters may be masked by compensatory “Ca²⁺ transporter reprogramming.”

How Can We Reconcile the Elevated Basal BP in α₂^SM-DN and α₂^+/− Mice With the Normal Basal BP in α₂^CV-KO Mice?

The fact that α₂^SM-DN and α₂^+/− mice have modestly elevated basal BP (Fig. 4) (57), whereas α₂^CV-KO mice do not (48), appears to conflict with the general idea proposed above. Consideration of model-specific differences, however, suggests an explanation. First, α₂^SM-DN and α₂^+/− mouse arteries express 30–50% of the normal α₂ protein (Fig. 3) (56), whereas the level in α₂^CV-KO mice, 11 ± 7% (48), is not significantly greater than 0%. [A very low level of α₂ expression is expected, however, because endothelial cells also express α₂-Na⁺ pumps (54) and this should not be knocked out in α₂^CV-KO mice.] Second, NCX1 is not downregulated in α₂^CV-KO mice (48), which indicates that the normal basal BP cannot be explained by the type of compensatory Ca²⁺ transporter regulation that apparently attenuates BP elevation in α₂^SM-DN mice. The implication is that the PM-sarco(endo)thelial reticulum junctional complexes, where NCX1 normally functionally couples to α₂-Na⁺ pumps (25, 51, 57), are evidently disrupted by the virtually complete KO of α₂ in α₂^CV-KO mice. As a result, NCX1-mediated Ca²⁺ transport becomes linked to the Na⁺ concentration gradient established by α₁-Na⁺ pumps. Because α₁-Na⁺ pumps have higher affinity for cytosolic Na⁺ concentration than do α₂-Na⁺ pumps [α₂ apparent K_m = 12 mM and α₂ K_m = 22 mM cytosolic Na⁺ concentration (55)], they help maintain a steady, low cytosolic Ca²⁺ concentration in quiescent myocytes when α₂ is absent.

Furthermore, α₂^CV-KO mice are resistant to adrenocorticotropic hormone-induced hypertension, an EO-dependent form of hypertension (10), as expected for mice with no arterial ouabain-sensitive α₂-Na⁺ pumps (10, 30). In contrast, the BP in mice with a reduced, but finite, number of these pumps (vs. WT control mice) should be elevated under conditions that raise plasma EO, such as ANG II and HS (5, 6, 15), as observed in α₂^SM-DN mice (Fig. 6). Thus, in summary, the contrast between α₂^SM-DN and α₂^CV-KO mice, rather than undermining our view about the relationship between α₂-Na⁺ pumps and BP, actually supports it.

Implications for the Pathogenesis of Hypertension

Our findings provide insights that appear relevant to the pathogenesis of hypertension. In numerous rodent hypertension models, including the Dahl, spontaneously hypertensive rat, ANG II, DOCA-salt, Milan, and ouabain-induced models, these same Ca²⁺ transporters (e.g., NCX1, SERCA2, and/or TRPC6) are markedly upregulated (2, 15, 44, 45, 52, 59, 60). In both the α₂^SM-Tg and α₂^SM-DN models, the Ca²⁺ transporter reprogramming appears to be “protective”: it either minimizes Ca²⁺ depletion and attenuates BP decline (α₂^SM-Tg mice) or minimizes Ca²⁺ overload and attenuates BP elevation (α₂^SM-DN mice). In the aforementioned hypertension models, however, the Ca²⁺ transporter protein reprogramming can be expected to enhance Ca²⁺ gain and signaling by myocytes (29, 45, 59) and thereby promote the rise in BP. The implication is that this protein reprogramming, which we attribute to a centrally regulated, sustained increase in circulating EO (7, 15) that acts directly on myocytes (28, 45, 60), contributes significantly to the elevated BP in all those hypertension models.

Conclusions

This report reinforces the long-standing idea (4, 47) that Na⁺ pumps and NCX are important factors in the regulation of arterial Ca²⁺ homeostasis and constriction. Although SM-specific monogenic alteration of expression of either protein might be expected to modify, profoundly, Ca²⁺ homeostasis and, thus, BP, compensatory changes in other proteins may markedly attenuate the responses and confound the interpretation. The resulting modest net changes in basal BP may be misleading and, therefore, overlooked, unless the functional significance of all these factors is taken into account.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants R01-HL-107555 and P01-HL-78870 (to M. P. Blaustein) and R01-HL-107654 (to J. Zhang).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

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