Altered vascular smooth muscle function in the ApoE knockout mouse during the progression of atherosclerosis

Abstract

Objectives

Relaxation of vascular smooth muscle (VSM) requires re-uptake of cytosolic Ca²⁺ into the sarcoplasmic reticulum (SR) via the Sarco/Endoplasmic Reticulum Ca²⁺ ATPase (SERCA), or extrusion via the Plasma Membrane Ca²⁺ ATPase (PMCA) or sodium Ca²⁺ exchanger (NCX). Peroxynitrite, a reactive species formed in vascular inflammatory diseases, upregulates SERCA activity to induce relaxation but, chronically, can contribute to atherogenesis and altered vascular function by escalating endoplasmic reticulum stress. Our objectives were to determine if peroxynitrite-induced relaxation and Ca²⁺ handling processes within vascular smooth muscle cells were altered as atherosclerosis develops.

Methods

Aortae from control and ApoE^−/− mice were studied histologically, functionally and for protein expression levels of SERCA and PMCA. Ca²⁺ responses were assessed in dissociated aortic smooth muscle cells in the presence and absence of extracellular Ca²⁺.

Results

Relaxation to peroxynitrite was concentration-dependent and endothelium-independent. The abilities of the SERCA blocker thapsigargin and the PMCA inhibitor carboxyeosin to block this relaxation were altered during fat feeding and plaque progression. SERCA levels were progressively reduced, while PMCA expression was upregulated. In ApoE^−/− VSM cells, increases in cytosolic Ca²⁺ [Ca²⁺]_c in response to SERCA blockade were reduced, while SERCA-independent Ca²⁺ clearance was faster compared to control.

Conclusion

As atherosclerosis develops in the ApoE^−/− mouse, expression and function of Ca²⁺ handling proteins are altered. Up-regulation of Ca²⁺ removal via PMCA may offer a potential compensatory mechanism to help normalise the dysfunctional relaxation observed during disease progression.

Highlights

• •Expression and function of SERCA and PMCA are temporally altered in ApoE^−/− VSM.
• •TG-induced increases in [Ca²⁺]_c were reduced in ApoE^−/− aortic SM cells.
• •Ca²⁺ extrusion is upregulated in isolated ApoE^−/− aortic SM cells.

Abbreviations

Ca²⁺

calcium ion

[Ca²⁺]_c

cytosolic calcium concentration

CE

carboxyeosin

MLCK

myosin light chain kinase

NCX

sodium/Ca²⁺ exchanger

NO

nitric oxide

ONOO⁻

peroxynitrite

PLB

phospholamban

PMCA

plasma membrane Ca²⁺ ATPase

RNS

reactive nitrogen species

RyR

ryanodine receptor

SERCA

sarco/endoplasmic reticulum ATPase

SOCE

store operated Ca²⁺ entry

SOCC

store operated Ca²⁺ channel

SR

sarcoplasmic reticulum

TG

thapsigargin

VLDL

very low density lipoprotein

hfd

high fat diet

VSM

vascular smooth muscle

1. Introduction

Increased oxidant stress is implicated in the progression of several diseases affecting the vasculature, including hypertension, diabetes and atherosclerosis [1–4]. In vascular smooth muscle (VSM), reactive oxygen (ROS) and nitrogen species (RNS) can alter core functions [5], including migration, cell growth, vascular reactivity and inflammatory processes [6–8]; all critical factors in the development and progression of atherosclerosis and cardiovascular disease. One such RNS, peroxynitrite (ONOO⁻), is the reaction product of nitric oxide (NO) and superoxide (O₂^.) and has direct and indirect effects on the relaxation of blood vessels [9–12]. However, the precise mechanism by which ONOO⁻ induces relaxation remains unclear.

Ca²⁺ is the fundamental second messenger controlling VSM contraction and relaxation, and can enter the cytoplasm via membrane depolarisation, non-selective cation channels, store operated Ca²⁺ entry (SOCE) or from the sarcoplasmic reticulum (SR) [13]. Ca²⁺ clearance occurs either via re-uptake into the SR via sarco/endoplasmic reticulum Ca²⁺ ATPase (SERCA) pumps [14], which utilise energy from ATP hydrolysis to maintain a 10000-fold Ca²⁺ gradient between the SR lumen and cell cytoplasm, or extrusion into the extracellular space via the plasma membrane Ca²⁺ ATPase (PMCA) [15] or sodium Ca²⁺ exchanger (NCX) [16]. In healthy vessels, ONOO⁻ has been shown to induce relaxation of rat aorta following constriction to phenylephrine [12]. This relaxation was not reduced by inhibitors of nitric oxide synthase or endothelial denudation, and was thought to occur via elevation of cGMP levels, membrane hyperpolarisation and direct activation of myosin phosphatase activity in the smooth muscle. A further potential mechanism for ONOO⁻-induced vascular relaxation is via increased uptake of Ca²⁺ into the internal SR store via SERCA or extrusion across the plasma membrane. Lower levels of ONOO⁻ (10–50 μM) have been shown, in purified cardiac fractions and aorta homogenates, to increase the activity of SERCA via reversible S-glutathiolation [9]. Conversely, higher ONOO⁻ concentrations (>100 μM) were associated with tyrosine nitration and SERCA inhibition. ONOO⁻ has also been determined to contribute to atherogenesis [17], and chronically elevated levels of RNS in atherosclerotic rabbit arteries impair SERCA activity by irreversible oxidation of relevant cysteine thiols [9]. Oxidative stress has also been shown to target the plasma membrane pump PMCA in disease states; altering Ca²⁺ extrusion in platelets [18] and neuronal tissue [19], but its role in PMCA activity in smooth muscle has not yet been investigated. A role for NCX in VSM remains controversial [20–22], although what is clear is that NCX is expressed in this cell type and is also sensitive to ROS which may disrupt Ca²⁺ homoeostasis under oxidative stress [23]. Therefore, significant evidence exists for the interplay of reactive species and Ca²⁺ signalling mechanisms [24]. In addition, defects in smooth muscle Ca²⁺ handling has been shown to affect responses to vasodilators [25].

In this study, we have utilised pharmacological Ca²⁺ pump inhibitors to examine the effect of up to 4 months high fat feeding of atherosclerosis-prone ApoE^−/− mice on ONOO⁻-mediated relaxation. Mice on a high fat diet are reported to have approximately 5% lesion coverage of the aorta [39] but, even prior to the development of lesions, changes in smooth muscle function are apparent [26]. This indicates the importance of circulating factors and perhaps the influence of other cell types present in the lesion on smooth muscle function. We have determined the expression levels of SERCA and PMCA during disease progression and have explored how alterations in these proteins impacts upon Ca²⁺ handling by examining single cell Ca²⁺ responses in the presence and absence of extracellular Ca²⁺. Changes in ONOO⁻ responses have been correlated with high fat feeding and plaque development.

2. Materials and methods

2.1. Animal model and aorta preparation

Adult male C57/BL-6 or apolipoprotein E deficient mice (ApoE^−/−) [27–29] were used. ApoE^−/− mice were placed on high fat diet (hfd) for 2 or 4 months to accelerate plaque progression. Age-matched chow fed ApoE^−/− mice were also assessed. Procedures conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and Directive 2010/63/EU of the European Parliament. Mice were terminally anaesthetised via intraperitoneal injection of sodium pentobarbital (200 mg/ml). Thoracic aortae were removed to ice cold oxygenated (95% O₂:5% CO₂) Krebs' solution and cleaned of adherent tissue.

2.2. Histological analysis

Cleaned thoracic aortae were fixed in neutral-buffered formalin, embedded in paraffin and 4 μm sections cut on a rotary microtome. Morphological analysis was performed on haematoxylin and eosin stained sections. Nitrotyrosine (Millipore) and α-actin (Abcam) were detected using rabbit primary antibodies (diluted 1 in 100) and visualised using biotin labelled secondary antibody-streptavidin-HRP complexes and DAB (3,3′ diaminobenzidine) chromogenic substrate (Vector Laboratories). Images were analysed using ImmunoRatio analysis software (IBT, University of Tampere), which calculates the percentage of DAB staining over total nuclear area.

2.3. Small vessel wire myography

Descending thoracic aorta was systematically divided into rings of 2 mm, and endothelium was removed. Rings were mounted on a small vessel wire myograph (Danish Myotech), placed under a resting tension of 1 g and allowed to equilibrate. Reproducible responses were obtained to 40 mM KCl and 30 nM 9,11-Dideoxy-9α,11α-methanoepoxy prostaglandin F_2α (U46619, Tocris) and endothelial function was assessed by the addition of 10 μM acetylcholine before commencing experiments. Rings were pre-contracted to U46619 (30 nM) and then cumulative doses of ONOO⁻ (Calbiochem, 1 × 10⁻⁶ – 5x10⁻⁴ M) added at 10 min intervals. ONOO⁻ was diluted in argon-purged dH₂O and kept in the dark at 4 °C in order to protect activity. 3 μM thapsigargin (TG, Sigma) or 10 μM carboxyeosin (CE, Marker Gene Technologies) were added 30 min prior to ONOO⁻.

2.4. Protein expression/immunoblotting

Denuded aortae were pulverised in liquid nitrogen and incubated in lysis buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM DTT, 0.25% (w/v) Na-deoxycholate, 1% (v/v) TX-100, 1× protease inhibitor cocktail (Roche)) for 30 min at 4 °C. Lysates were spun to remove debris, and protein concentrations determined using Coomassie Plus Protein Assay Reagent (Perbio, USA). Denatured samples were subjected to electrophoresis (NuPAGE, 3–8% tris–acetate gels (Invitrogen)) and western blotting. SERCA was detected using rabbit polyclonal anti-SERCA2b (a kind gift from Prof. F Wuytack, Leuven). SERCA2b was identified as a dimer, with a fainter monomer band detected at 110 kDa. Protein abundance was quantified from a range of protein loads using Quantity One software (BioRad). PMCA (Affinity Bioreagents) was detected and quantified similarly. Expression was normalised to GAPDH (Cell Signaling Technology).

2.5. Single cell isolation and Ca²⁺ measurement in the presence and absence of extracellular Ca²⁺

Single aortic smooth muscle cells were enzymatically isolated [21] and changes in Ca²⁺ levels measured as fluorescence using the membrane permeable dye Fluo-3 AM (10 μM, Molecular Probes/Invitrogen). Cells were loaded with Fluo-3 for 30 min prior to the beginning of the experiment. Cells were allowed to settle prior to perfusion with bathing solution containing (in mM): 80 Na glutamate, 40 NaCl, 20 tetraethylammonium chloride, 1.1 MgCl₂, 3 CaCl₂, 10 HEPES, and 30 glucose; adjusted to pH 7.4 with NaOH. The Ca²⁺⁻free extracellular solution additionally contained (mM): MgCl₂, 3 (substituted for Ca²⁺); and EGTA, 1. Fluorescence was quantified using a microfluorimeter, which consisted of an inverted microscope (Olympus IX81) and a photomultiplier tube with a bi-alkali photocathode. Fluo-3 was excited at 488 nm (bandpass 9 nm) from a PTI Delta Scan (Photon Technology International Inc., London, UK) through the epi-illumination port of the microscope. Excitation light was passed through a field stop diaphragm to reduce background fluorescence and reflected off a 505 nm long-pass dichroic mirror. Emitted light was guided through a 535 nm barrier filter (bandpass 35 nm) to a photomultiplier in photon counting mode. Caffeine (10 mM) was applied by hydrostatic pressure ejection using a pneumatic pump (PicoPump PV 830, World Precision Instruments). TG (1 μM) was perfused into the solution bathing the cells.

2.6. Statistical analysis

Myography data was analysed via Graphpad Prism software and significance determined using 2-way ANOVA. EC₅₀ and E_max values were determined using non-linear fit curve analysis of individual data sets followed by ANOVA. Student's t-tests (unpaired) were performed on protein expression data. Changes in [Ca²⁺]_c were expressed as ratios (F/F₀ or ΔF/F₀) of fluorescence counts (F) relative to baseline (control) values before stimulation (F₀), and were analysed via Origin (OriginLab) and WinFluor (University of Strathclyde) data analysis software. Data are expressed as means ± SEM and p < 0.05 was taken to be indicative of statistical significance. Confidence intervals are also shown where appropriate.

3. Results

3.1. Protein nitrosylation is increased in atherosclerotic ApoE^−/− mouse aortae

Nitrotyrosine expression was higher in ApoE^−/− thoracic aorta in comparison to C57/BL-6 and particularly marked in medial and adventitial areas around atherosclerotic plaques (Fig. 1a and b). Some diffuse staining also was found within the plaque. There was significantly more nitrotyrosine detected (Fig. 1b) in the 4 month high fat fed ApoE^−/− aorta (7.50 ± 3.01% vs. 24.3 ± 2.15% expressed as percentage of nuclear area, Fig. 1e ***p < 0.001). The increased ONOO⁻ generation during atherosclerosis could lead to nitrosylation and modified function of vascular Ca²⁺ handling proteins. Spleen size in atherosclerotic mice was also significantly and progressively increased with time spent on high fat diet (0.27 ± 0.03, 0.40 ± 0.04* and 0.52 ± 0.07** % of total body weight for C57/BL-6, 2 month and 4 month high fat fed ApoE^−/− respectively, *p < 0.05 and **p < 0.01 vs. C57/BL-6); an indicator of progressive inflammatory responses in these mice which is known to be correlated to peroxynitrite bioavailability. Smooth muscle α-actin staining revealed similar numbers of smooth muscle cells in C57/BL-6 and ApoE^−/− aortic sections (Fig. 1 c, d and e, 49.83 ± 10.71% vs. 58.14  ±  5.03%), with smooth muscle staining also being observed in the fibrous cap of plaques in ApoE^−/− aortae (Fig. 1d). Haematoxylin and eosin staining revealed typical plaque morphology in ApoE^−/− aortic cross-sections from mice which have been fed a high fat diet for 4 months (Fig. 1f).

Fig. 1.

Protein nitrosylation is increased in atherosclerotic ApoE^−/− mouse thoracic aortae. (a, b) Representative cross-sectional images of nitrotyrosine stained C57/BL-6 and 4 month hfd ApoE^−/− mouse thoracic aortae (200× magnification). (c, d) Smooth muscle actin staining. (e) Mean DAB staining/nuclear area (%) for nitrotyrosine and actin. (f) H&E stained aorta from 4 month hfd ApoE^−/−. L = lumen, SM = smooth muscle, P = plaque, FC = fibrous cap, +ve = areas of positive staining ***p < 0.001 vs. C57/BL-6, n = 5 mice per group.

3.2. Mouse thoracic aorta relaxes to ONOO⁻ in a dose-dependent manner

Endothelium denuded aortic rings relaxed to ONOO⁻ in a concentration dependent manner (Fig. 2a). In 2 month ApoE^−/− mice sensitivity to ONOO⁻ was increased (Fig. 2b, ***p < 0.001 vs. C57/BL-6) but this had normalised by four months on high fat diet. EC₅₀ values were 3.9, 1.9, 3.3 × 10⁻⁵ M (95% confidence intervals of 3.2–4.8, 1.6 to 2.3 and 2.7 to 3.9 × 10⁻⁵ M) and E_max values were 87.8 ± 4.7, 89.6 ± 2.9, 82.5 ± 3.7% for C57/BL-6, 2 month and 4 month high fat diet ApoE^−/−, respectively. An overview of EC₅₀ and E_max values is given on Table 1. No significant difference was observed between endothelial denuded or intact vessels (data not shown).

Fig. 2.

Mouse thoracic aorta relaxes to ONOO⁻ in a dose-dependent manner. (a) Representative trace of 3x10^-8 M U46619-induced contraction of mouse aorta and subsequent relaxation to increasing doses of ONOO⁻ (1 × 10⁻⁶ – 5x10⁻⁴ M). (b) ONOO⁻-induced relaxation in C57/BL-6, 2 and 4 month hfd ApoE^−/− aortae. All data presented are from endothelial denuded vessels. ***p < 0.001 for 2 month hfd ApoE^−/− vs. C57/BL-6, minimum of 7 mice per group.

Table 1.

EC₅₀ and E_max values for ONOO⁻-induced relaxations.

		EC50 (M)	Change in EC50 (M) (compared to control)	Emax(%)	Change in Emax( %) (compared to control)
C57/BL-6	Control	3.9 × 10−5 (CI: 3.2–4.8 × 10−5)	N/A	87.8 (SEM: 4.7)	N/A
	3 μM TG	4.9 × 10−5 (CI: 3.5–6.9 × 10−5)	+1.0 × 10−5	22.3 (SEM: 4.3)	−65.6
	10 μM CE	3.4 × 10−5 (CI: 2.1–5.5 × 10−5)	−0.5 × 10−5	97.5 (SEM: 9.1)	+9.6
2mo hfd ApoE−/−	Control	1.9 × 10−5 (CI: 1.6–2.3 × 10−5)	N/A	89.6 (SEM: 2.9)	N/A
	3 μM TG	8.5 × 10−5 (CI: 6.8–10.6 × 10−5)	+6.6 × 10−5	83.6 (SEM: 4.9)	−6.6
	10 μM CE	4.6 × 10−5 (CI: 2.8–7.8 × 10−5)	+2.7 × 10−5	82.6 (SEM: 9.5)	−7.0
4mo hfd ApoE−/−	Control	3.3 × 10−5 (CI: 2.7–3.9 × 10−5)	N/A	82.5 (SEM: 3.7)	N/A
	3 μM TG	No relaxation	No relaxation	No relaxation	No relaxation
	10 μM CE	2.5 × 10−5 (CI: 1.9–3.3 × 10−5)	−0.8 × 10−5	82.2 (SEM: 4.0)	−0.2

3.3. Expression of SERCA2b is progressively reduced while PMCA levels are increased in aortic smooth muscle from ApoE^−/− mice

Expression of SERCA2b (Fig. 3a) was significantly down-regulated in aortic smooth muscle from ApoE^−/− (45% reduction, *p < 0.05 vs. C57/BL-6 after 2 months fat feeding and 75%, ***p < 0.001 vs. C57/BL-6 after 4 months). In contrast, expression of PMCA (Fig. 3b) was significantly upregulated (by 45%, *p < 0.05 vs. C57/BL-6 following 2 months high fat feeding and by **137%, p < 0.01 vs. C57/BL-6 following 4 months). We found no significant changes in NCX expression at 2 months and a significant reduction at 4 months (data not shown).

Fig. 3.

Expression levels of SERCA2b are progressively reduced while PMCA levels are increased in aortic smooth muscle from ApoE^−/− mice. (a, b) Representative immunoblots and averaged protein expression data for SERCA2b, pan-PMCA and GAPDH. Protein loads were 2.5, 5, 7.5 μg for SERCA and 10, 15, 20 μg for PMCA. Slopes of plotted band densities were standardised to GAPDH. *p < 0.05, **p < 0.01 and ***p < 0.001 vs. C57/BL-6, minimum of 5 independent experiments.

3.4. Efficacy of Ca²⁺ pump inhibitors is temporally altered as atherosclerosis progresses

As ONOO⁻ is known to activate SERCA [9], the expression of which is progressively reduced in atherosclerotic VSM (Fig. 3a), we next evaluated the function of SERCA using the irreversible SERCA inhibitor thapsigargin (TG). Contraction of endothelium denuded aortic rings to U46619 did not differ among groups or between controls and TG pre-treated tissues. TG (3 μM) was found to inhibit ONOO⁻-induced relaxation in control and atherosclerotic aorta [Fig. 4a–c]. A 65.5% reduction in E_max was observed in TG treated C57/BL-6 preparations, while a complete abolition of relaxation was observed in 4 month high fat diet ApoE^−/− preparations (98.5% reduction in E_max). Increased SOCE may influence the function of TG at this latter time-point, and this will be discussed in detail later. 2 month high fat diet ApoE^−/− aortae were less sensitive to TG (only 6% reduction in E_max), which implies that other Ca²⁺ removal mechanisms can compensate to mediate the relaxation to ONOO⁻ at this stage of disease progression.

Fig. 4.

Efficacy of TG and CE is temporally altered as atherosclerosis progresses. (a–c) ONOO⁻-induced relaxation following 3 μM TG pre-incubation. ***p < 0.001 vs. control, †††p < 0.001 vs. TG treated C57/BL-6, ‡‡‡p < 0.001 vs. TG treated 2mo hfd ApoE^−/−. (d–f) ONOO⁻-relaxation following pre-incubation with 10 μM CE. ***p < 0.001 vs. control, ††p < 0.01 vs. CE treated C57/BL-6, ‡‡‡p < 0.001 vs. CE treated 2mo hfd ApoE^−/−. All experiments performed following endothelial denudation. Minimum of 5 mice per group.

To investigate the contribution of PMCA, vessels were pre-treated with 10 μM of the PMCA inhibitor CE (Fig. 4d–f). Contraction to U46619 did not differ between non-CE and CE treated tissues. Significant inhibition of ONOO⁻-induced relaxation was only found in 2 month fat fed aortae (***p < 0.001 vs. no CE control). A higher concentration of CE (30 μM) was ineffective in blocking relaxation in 4 month high fat diet ApoE^−/− aortae. An overview of EC₅₀ and E_max values following preincubation with TG or CE is also shown on Table 1

We hypothesised that PMCA activity may be responsible for increased relaxation to ONOO⁻ in 2 months fat fed ApoE^−/− and so attempted to study intracellular calcium regulation in isolated aortic smooth muscle cells.

3.5. Ca²⁺ extrusion is upregulated in atherosclerotic mouse aortae

Changes in [Ca²⁺]_c in response to the ryanodine receptor (RyR) agonist caffeine and the SERCA inhibitor TG were measured in freshly isolated VSM cells. TG induced a transient [Ca²⁺]_c increase in all sample groups and completely inhibited any further caffeine induced responses, confirming store emptying following SERCA inhibition (Fig. 5a). Caffeine responses were similar across control and atherosclerotic groups (Fig. 5b and e), but the decay times of the TG-induced Ca²⁺ transients (after SERCA blockade) were found to be significantly faster in cells from ApoE^−/− mice (Fig. 5c and f, decay times (s): 264 ± 25, 149 ± 29* and 135 ± 23** for C57/BL-6, 2 month and 4 month high fat fed ApoE^−/− respectively, *p < 0.05 vs. C57/BL-6). This effect was also observed in the absence of extracellular Ca²⁺ (Fig. 5 d and f, decay times (s): 228 ± 31, 138 ± 26** and 100 ± 26*** for C57/BL-6, 2 month and 4 month high fat fed ApoE^−/− respectively) *p < 0.05, **P < 0.01, ***p < 0.001 vs. C57/BL-6. The peak Ca²⁺ response to TG was reduced in the absence of extracellular Ca²⁺ and there was a significantly greater reduction in cells from ApoE^−/− mice (Fig. 5d and e, ΔF/F₀: 0.85 ± 0.16, 0.36 ± 0.07*** and 0.33 ± 0.12*** for C57/BL-6, 2 month and 4 month high fat fed ApoE^−/− respectively, ***p < 0.0001 vs. C57/BL-6). This suggests a difference in contribution of SERCA, as well as SOCE, between cells from C57/BL-6 and ApoE^−/− mice. Since SERCA was blocked with TG in these experiments, the increased expression of PMCA in atherosclerotic VSMs (Fig. 3c), may be responsible for the more rapid sarcolemmal Ca²⁺ extrusion observed in ApoE^−/− cells in Fig. 5c and f.

Fig. 5.

Ca²⁺ extrusion is upregulated in atherosclerotic mouse aortae. (a) Typical experimental protocol. (b–d) Representative responses to 10 mM caffeine and 1  μM TG (in the presence and absence extracellular Ca²⁺). (e) Mean peak Ca²⁺ responses (ΔF/F₀). (f) Decay times (s) of Ca²⁺ transients. *p < 0.05, **p < 0.01 and ***p < 0.001 vs C57/BL-6. The average of multiple cells from a single mouse was used as 1 data point, n = 5–6 mice per group.

4. Discussion

In this study, we have presented evidence showing novel time-dependent and progressive changes in Ca²⁺ handling within VSM cells during high fat feeding in ApoE^−/− mice. In agreement with previous investigations in canine coronary and cerebral arteries and rabbit and rat aortae [12,30–33], the pro-oxidant ONOO⁻ relaxed pre-contracted C57/BL-6 and ApoE^−/− mouse aorta. However, we additionally report for the first time that this vasodilator response (Fig. 2b), and the expression and function of VSM SERCA and PMCA (Figs. 3 and 4), are altered with time on high fat diet and reveal modifications and compensatory mechanisms of Ca²⁺ handling in the disease state (Fig. 5).

Facilitation of cellular Ca²⁺ removal requires a combination of SR Ca²⁺ uptake via SERCA and plasma membrane extrusion [34] via PMCA and NCX, although the extent of the involvement of the latter protein is controversial in aortic smooth muscle [20–22,35]. Importantly though, the majority of Ca²⁺ clearance in mouse aortic VSM (60–70%) has been attributed to the combined function of PMCA and SERCA and not NCX and SERCA [36]. We therefore focussed this study on the impact of SERCA and PMCA. The changes we observed in relaxation to ONOO⁻ over time in vessels from fat fed animals will be influenced not only by our observed changes in SERCA function and SOCE, but also by the compensatory changes we have identified in the expression and function of PMCA. We suggest the altered balance and influence of these processes accounts for the altered sensitivity to ONOO⁻ observed in this study.

Surprisingly, ONOO⁻-induced relaxation was enhanced in ApoE^−/− mice which had been high fat fed for 2 months while, after 4 months on diet, this enhancement was lost (Fig. 2b). Improved smooth muscle cell function in ApoE deficient mice has not been reported previously, however a recent study by Beleznai et al. [40] did reveal increased endothelial cell function and sensitivity to dilation in the mesenteric arteries of ApoE^−/− of 9–14 weeks of age. This increased sensitivity was reduced in older ApoE^−/− [40], which correlates with our findings presented here on aortic smooth muscle. Paradoxically, although we observed greater relaxation of aortae from 2 month fat fed ApoE^−/− mice, SERCA expression was significantly reduced in comparison to controls. We would suggest that a compensatory upregulation of calcium extrusion occurs in the smooth muscle of ApoE^−/− mice, which allows for augmented relaxation at the earlier 2 month time-point but is “over-ridden” by smooth muscle dysfunction as age and time on high fat diet increases. Our data suggests that PMCA is likely to be responsible for the increased calcium extrusion and enhanced relaxation to ONOO⁻ at 2 months high fat diet since expression was increased (Fig. 3b) and the PMCA inhibitor carboxyeosin normalised the enhanced relaxation to ONOO⁻ (Fig. 4e).

Release of Ca²⁺ from the SR will also activate SOCE, which is upregulated in ApoE knockout mice [26]. We have confirmed this finding in our single cell experiments through the addition of the SERCA blocker (and, consequently, SOCC activator) TG in extracellular Ca²⁺ -containing and Ca²⁺- free/EGTA conditions (Fig. 5). We also report here for the first time that the TG-induced increase in [Ca²⁺]c (in extracellular Ca²⁺-free conditions) is reduced in aortic smooth muscle from ApoE^−/− (Fig. 5e) and this is in line with our finding that SERCA protein expression is reduced in these samples (Fig. 3a). SERCA plays an important role in ONOO⁻-induced relaxation, since pre-treatment with TG caused significant reduction in relaxation in aortae from C57/BL-6 and ApoE^−/− mice (Fig. 4a–c). SERCA is highly sensitive to oxidative damage and, in ApoE^−/− mice with demonstrable atherosclerotic plaques, we found nitrotyrosine staining around the plaque correlated with a progressive reduction in SERCA protein expression and altered effectiveness of TG in inhibiting ONOO-induced relaxation. TG reduced ONOO⁻-induced relaxation most notably in C57/BL-6 and 4 month ApoE^−/− preparations while, in 2 month ApoE^−/− aortae, inhibition by TG was much less marked. The lack of sensitivity to SERCA blockade at 2 months is likely due to reduced SERCA expression (Fig. 3a) but this does not correlate with data at 4 months where SERCA expression was further reduced but the relaxation to ONOO⁻ was more sensitive to TG. The observed increase in SOCC function in VSM from ApoE^−/− mice goes some way to explaining the increased efficacy of TG in blocking ONOO⁻-induced relaxation in aorta from 4 month high fat fed ApoE^−/− mice (Fig. 4c), as TG will cause indirect SOCC activation as a consequence of SR Ca²⁺ store depletion. Therefore, increased Ca²⁺ entry via SOCCs potentially precludes relaxation following incubation with TG at this later time point.

Interestingly, a recent paper by Barriga et al. [41] reported that SERCA2 protein levels are reduced in cultured atrial myocytes by the addition of LDL. Data from our laboratory has indicated there is no significant decrease in SERCA2b expression between age-matched chow-fed C57/BL-6 and ApoE^−/− aortae (data not shown). Other studies have shown raised LDL levels with time spent on high fat diet [42–44], with approximately a two-fold increase in comparison to chow-fed animals. Since we have shown reduced SERCA expression following 2 and 4 months high fat feeding, it is probable that in addition to the impact of atherosclerosis, increased LDL levels may play a role in altered SERCA expression and function.

No other investigators to date have specifically investigated the temporal relationship between atherosclerotic development, high fat feeding and the relationship between smooth muscle cell SERCA and PMCA, although other studies have observed a compensatory relationship between the two in other cell types. Over-expression of PMCA has been shown to lead to down-regulation of the SERCA pump in rat aortic endothelial cells by Liu et al. [37] and work by Wang et al. [38] indicates that PMCA functions co-operatively with SERCA and NCX in rabbit aortic endothelium. We therefore believe that this study represents an important step forward in the understanding of the temporal and compensatory alterations observed in VSM function between healthy and hypercholesterolaemic ApoE^−/− mice.

5. Conclusion

For the first time we have demonstrated time-dependent changes in the expression and activity of Ca²⁺ regulatory mechanisms in VSM during progressive high fat feeding in atherosclerosis-prone mice. Compensatory changes, comprising a rise in PMCA expression, coupled with a fall in SERCA expression and function act as important mechanisms to sustain ‘normal’ relaxation to some vasodilator agents as a pathological condition is developing. Data from single cells suggest enhanced Ca²⁺ extrusion, likely via PMCA may be responsible for maintenance of close to normal relaxation responses for a period of time and perhaps augmented vasodilator response at early stages of fat feeding. This study highlights the importance of examining multiple time points during disease development and the interdependence and complexity of SERCA, SOCE and PMCA function during atherosclerosis.

Source of funding

This work was supported by project grants from the British Heart Foundation (BHF-PG/07/012/22342), (BHF-PG/10/79/28603), The Royal Society (RG110356) and Tenovus Scotland (S12/4).

Disclosures

No conflict of interest declared.