Abnormal Intracellular Calcium Homeostasis in Sympathetic Neurons From Young Prehypertensive Rats
Hypertension is associated with cardiac noradrenergic hyperactivity, although it is not clear whether this precedes or follows the development of hypertension itself. We hypothesized that Ca2+ homeostasis in postganglionic sympathetic neurons is impaired in spontaneously hypertensive rats (SHRs) and may occur before the development of hypertension. The depolarization-induced rise in intracellular free calcium concentration ([Ca2+]i; measured using fura-2-acetoxymethyl ester) was significantly larger in cultured sympathetic neurons from prehypertensive SHRs than in age matched normotensive Wistar-Kyoto rats. The decay of the [Ca2+]i transient was also faster in SHRs. The endoplasmic reticulum Ca2+ content and caffeine-induced [Ca2+]i amplitude were significantly greater in the young SHRs. Lower protein levels of phospholamban and more copies of ryanodine receptor mRNA were also observed in the young SHRs. Depleting the endoplasmic reticulum Ca2+ store did not alter the difference of the evoked [Ca2+]i transient and decay time between young SHRs and Wistar-Kyoto rats. However, removing mitochondrial Ca2+ buffering abolished these differences. A lower mitochondrial membrane potential was also observed in young SHR sympathetic neurons. This resulted in impaired mitochondrial Ca2+ uptake and release, which might partly be responsible for the increased [Ca2+]i transient and faster decay in SHR sympathetic neurons. This Ca2+ phenotype seen in early development in cardiac stellate and superior cervical ganglion neurons may contribute to the sympathetic hyperresponsiveness that precedes the onset of hypertension.
Hypertension is a multiorgan disease involving the kidney, vasculature, and autonomic nervous system. In particular, abnormal neurohumoral activation is a hallmark of hypertension and is a negative prognostic indicator for sudden cardiac death and a strong independent predictor of mortality.1,2 Much evidence supports the observation that sympathetic hyperresponsiveness is involved in the pathophysiology of human and animal primary hypertension.3–5 Increases in muscle sympathetic nerve activity in response to mental stress have been documented in normotensive offspring linked to a family history of hypertension.6 This suggests that dysregulated sympathohumoral activation may be an early marker of hypertension in those who are genetically predisposed to the disease.
Intracellular calcium concentration ([Ca2+]i) plays a pivotal role in triggering neurotransmitter release from sympathetic neurons.7 In superior cervical ganglion (SCG) neurons, Ca2+ signals govern the release of norepinephrine (NE).8,9 In turn, NE release plays a critical role in the regulation of blood pressure and cerebral blood flow distribution.10,11 We have shown in the spontaneously hypertensive rat (SHR), enhanced heart rate responses and evoked NE release from the postganglionic neurons innervating the heart when compared with normotensive Wistar-Kyoto (WKY) rats.12,13 Postsynaptically, basal and NE-stimulated l-type calcium current in single sinoatrial node cells was also enhanced in the SHR, as was the heart rate response to bath-applied NE.14 Taken together, these results suggest that a significant component of the sympathetic hyperresponsiveness in the SHR occurs at the end-organ level.
We tested the hypothesis that intracellular Ca2+homeostasis is dysregulated in the SHR and that this is genetically programmed and precedes the subsequent development of hypertension. To investigate the mechanisms involved in the potential differences between the SHR and WKY rat, we targeted calcium handling by the endoplasmic reticulum (ER) and mitochondria in an attempt to delineate whether disruption of Ca2+ handling proteins in intracellular stores might explain the enhanced exocytotic response in the SHR.12,15
Neonatal (4–7 days), prehypertensive young (4–6 weeks), and hypertensive adult (15–17 weeks) male SHRs and WKY rats were used in this study. Cells were isolated from the SCG and cardiac stellate ganglion for phenotyping. The investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (publication No. 85–23, revised 1996) and the Animals (Scientific Procedures) Act 1986 (United Kingdom).
An expanded Materials and Methods section is available in the online-only Data Supplement.
The phenotypic characteristics of both young and adult SHRs and WKY rats are summarized in Table S1 (please see the online-only Data Supplement). No differences in ventricular weight:body weight ratio (an indication of left ventricular hypertrophy), mean arterial blood pressure, or heart rate measured in vivo were observed in the young SHRs, whereas these measures were significantly increased in the adult SHRs when compared with the age-matched WKY rats.
Intracellular Free Calcium Transients in Sympathetic Neurons
Sympathetic neurons were confirmed by the catecholamine neuronal marker antityrosine hydroxylase stain (Figure 1A). Intracellular calcium concentration was performed by ratiometric recordings (using fura-2-acetoxymethyl ester, fura-2/AM) in single sympathetic neurons (Figure 1B). A typical calcium intensity profile of a sympathetic neuron responding to 100 mmol/L of KCl was shown in Figure 1C. In the SHR, high K+ evoked [Ca2+]i was significantly enhanced in both stellate ganglion (Figure 2) and SCG neurons (Figure 3A and 3B) when compared with age-matched WKY rats. In-depth profiling of the neural phenotype was performed on SCG cells because of the higher yield of neurons and ease of dissection.
Baseline [Ca2+]i was significantly higher in SCG neurons from the SHRs when compared with age-matched WKY rats in all 3 of the age groups (Table S2). The baseline [Ca2+]i in the SHR decreased with age, whereas this was not seen in the WKY rat. During depolarization with high K+, SCG neurons from the neonatal SHR had a greater increase in [Ca2+]i when compared with those from the WKY rat (Figure 3A and 3B). This effect was present in animals of all age groups, from neonatal to adult rats. There was a significant increase in evoked [Ca2+]i between the young and adult SHRs (Figure 3B). SCG neurons from the WKY rats also showed an increase in depolarization evoked [Ca2+]i with age, but the response remained smaller than the SHRs (eg, at adult, SHR: +32.10%; WKY: +13.92% compared with young rats; Figure 3B). We also calculated the area under the curve of the Ca2+ response to high K+ (Figure S1 in the online-only Data Supplement), and observed that this was significantly greater in the SHRs from neonatal to the fully developed hypertensive rat when compared with the age-matched WKY rats. Analyzing the responses as a percentage of change from baseline also resulted in greater [Ca2+]i responses in the young SHRs, although this was not observed in the neonatal or adult rats (Table S3 in the online-only Data Supplement).
Increased Rate and Decay Time of the Calcium Transient in SCG Neurons
In both SHRs and WKY rats, the [Ca2+]i increased rapidly in response to high K+ and fell slowly on high K+ removal (Figure 3A). The rate of rise (Δratio/Δtime) increased with age (Table S4), but there was no significant difference between SHRs and WKY rats in any age group. The WKY group demonstrated a significantly longer (50% and 90%) decay time in comparison with the SHR group from young and adult rats (Figure 3C). Because clear differences in both the amplitude and decay time of the [Ca2+]i were apparent in young prehypertensive SHRs and WKY rats, further experiments to investigate the role of the ER and mitochondria focused on young rats (4 weeks old).
ER Calcium Handling in SCG Neurons
Monitoring of ER Ca2+ Concentration
Ca2+ concentration in the ER can be directly measured by monitoring the Ca2+ within the organelles after loading with a low-affinity Ca2+ indicator mag-fura-2-acetoxymethyl ester (mag-fura-2/AM; Figure S2). Baseline mag-fura-2/AM fluorescence ratios were significantly higher in sympathetic neurons of young SHRs when compared with WKY rats (Figure 4A, left group bar). This difference persisted when the suffusate had 0 calcium present (Figure 4A, right group bar, and 4B). Activation of ryanodine receptors (RyRs) and depletion of Ca2+ from ER stores with 10 mmol/L of caffeine produced a larger drop in the mag-fura-2/AM fluorescence ratio in the SHR (−7.05±1.14%) compared with the WKY rat (−2.58±0.68%; P<0.01; Figure 4B). Subsequent introduction of the cell-permeant intraluminal Ca2+/heavy metal chelator, N,N,N′,N′-tetrakis(2-pyridylmethyl)-ethylenediamine (TPEN), caused a further reduction in the mag-fura-2/AM fluorescence ratio in both groups (Figure 4B). Fluorescence ratio normalized by baseline showed that the caffeine-evoked changes in ER Ca2+ were significantly greater in the SHRs than the WKY rats (to 0.92±0.01 versus 0.96±0.01; P<0.01; Figure 4C), whereas the TPEN response was similar (to 0.82±0.03 versus 0.85±0.03; P=0.51; Figure 4C). Releasable ER Ca2+ load was also estimated based on the [Ca2+]i responses to caffeine (10 mmol/L; 30 seconds; Figure 4D).
Studies have shown that advancing age can selectively change the genetic expression and protein levels of RyR3 but not RyR2 and RyR1 isoforms in the SCG aged 6-, 12-, and 24-month rats.16 Therefore, we investigated whether the RyR3 receptor is altered in young SHRs. Western blots were not sensitive enough to detect RyR3 receptor protein in the SCG dissected from young rats. However, SHRs were found to have significantly more RyR3 mRNA copies than WKY rats using RT-PCR (Figure 4E).
Protein Expression of Sarco/ER Ca2+-ATPase and Phospholamban
To understand the mechanism underlying the faster [Ca2+]i decay in the SHR, Western blot analyses were performed to assess the expression levels of sarco/ER Ca2+-ATPase (SERCA) 2a and phospholamban (PLN) in SCG homogenates from young WKY rats and SHRs (Figure 5). SERCA pump is under the regulatory control of the phosphoprotein phospholamban, which inhibits the apparent affinity of SERCA for Ca2+ in its nonphosphorylated form.17 There was no difference in the protein level of SERCA2a from young SHRs and WKY rats (P=0.63; Figure 5B). However, both the total and the phosphorylated portions (PLN-Ser16) of the PLN were significantly lower in the SHR group (Figure 5C and 5D). There was no difference in the PLN-Thr17 between the 2 groups (data not shown).
ER Contribution to the Changes [Ca2+]i in the SHR
To evaluate the contribution of the ER to the differences in peak [Ca2+]i rise and recovery time, ER Ca2+ stores were depleted by using caffeine (10 mmol/L; 30 seconds), and Ca2+ reuptake from SERCA pumps was blocked with 1 μmol/L of thapsigargin. Caffeine was reintroduced at 8.5 minutes and 13.5 minutes to confirm ER Ca2+ depletion (Figure 6A). Under these conditions, the SHR group continued to have a significantly higher elevation of [Ca2+]i during high K+ depolarization when compared with the WKY group (Figure 6B). The baseline [Ca2+]i was increased by 0.157±0.028 μmol/L in the SHR and 0.136±0.021 μmol/L in the WKY rat after ER Ca2+ depletion (P=0.56, t test). Because 90% decay time for the second high K+ depolarization was not reached in some experiments, only 50% decay time was measured, and this remained significantly shorter in the SHR when compared with the WKY group (P<0.01; Figure 6C).
Mitochondrial Calcium in SCG Neurons
Mitochondria Contribution to the [Ca2+]i
The proton uncoupler carbonylcyanide-p-trifluoromethoxyphenylhydrazone (FCCP; 1 μmol/L) causes an immediate depolarization of the inner mitochondrial membrane, resulting in depletion of stored mitochondrial Ca2+ and inhibition of any further mitochondrial Ca2+ uptake.18 Application of FCCP produced a small transient rise in [Ca2+]i (Figure 7A, center). This increase was not different between the 2 strains (SHR: 189±10 nmol/L versus WKY: 198±10 nmol/L; P=0.55, t test; n=17 and 23, respectively). Subsequent exposure to high K+ in the continued presence of FCCP resulted in an increase in the amplitude of the [Ca2+]i transient of 23.15±1.78% in the WKY rat and 12.74±2.48% in the SHR (P<0.01, t test; Figure 7B), abolishing the difference in depolarization-induced [Ca2+]i increase between the 2 groups. The 50% and 90% decay times of the [Ca2+]i transient were significantly shortened by FCCP in both WKY and SHR neurons (P<0.001, ANOVA), and the differences in 50% and 90% decay time between SHRs and WKY rats were abolished (Figure 7C).
Measurement of the Mitochondria Membrane Potential in Sympathetic Neurons
We then measured the membrane potential of mitochondria (ΔΨm) using the fluorescent dye tetra-methylrhodamine ethyl ester (TMRE; 20 nmol/L; 5 minutes). We confirmed that TMRE uptake was indeed attributed to accumulation within the mitochondria by demonstrating that it was rapidly released on mitochondrial depolarization with FCCP (Figure 8B). TMRE fluorescence intensity was significantly higher in the WKY rat than in the SHR (Figure 8A and 8B; P<0.001, t test).
There are 3 novel findings from this study. First, sympathetic neurons from the SHR have a significantly greater depolarization-evoked [Ca2+]i transient when compared with the age-matched WKY rat. This difference persisted from neonates through to young prehypertensive and fully developed adult hypertensive animals. Furthermore, the rate of decay of [Ca2+]i was faster in young and adult SHRs. Second, increased ER Ca2+ content, upregulated SERCA (because of reduced PLN inhibition), and RyR activity were seen in the young SHR. Third, mitochondrial membrane potential was reduced in the young SHR sympathetic neurons, and this may contribute to the differences in depolarization-evoked [Ca2+]i transients and the rate of decay between the 2 groups. When all of the results are taken together, they suggest that alteration of [Ca2+]i in sympathetic neurons from the SCG cells (and, by extrapolation, stellate ganglion cells), precedes the actual onset of hypertension and thereby maybe responsible for the enhanced sympathetic responsiveness seen at this developmental stage.
Intracellular Free Calcium Transients in the SHR
Calcium is an important ubiquitous secondary signaling messenger that is involved in both short-term and long-term regulation of cell function, metabolism and growth.19,20 In neurons, an increase in [Ca2+]i forms the pivotal link between the action potential and neurotransmitter release.7,21 It is well established that hypertension is strongly associated with noradrenergic hyperactivity, with increased central sympathetic output, and with elevated plasma epinephrine and NE levels.22,23 This dysregulation occurs at several levels of the cardiovascular neural axis, where peripheral sympathetic hyperresponsiveness in the adult hypertensive rat results in enhanced evoked cardiac NE release12,15 and also increased β-adrenergic responses in myocytes.24 Abnormal calcium handling properties have been reported in SHR vascular smooth muscle cells25 and endothelial cells,26 and it is conceivable that this also occurs in sympathetic neurons, thus providing a possible molecular link to enhanced noradrenergic neurotransmission.
The resting [Ca2+]i level in the SCG neurons from the neonates through to fully developed hypertensive animals was higher than that observed in the age-matched WKY rat. Similarly, the [Ca2+]i handling expressed as a percentage change from baseline was also greater in the young SHRs. However, this was not evident in the neonatal or adult cells from the SHRs and may be related to poor statistical power. Nevertheless, when taken together with the absolute [Ca2+]i responses and the area under the curve measurement for [Ca2+]i (Figure S1), the data are consistent with the hypothesis that prehypertensive SHRs have abnormal [Ca2+]i handling properties compared with age-matched WKY animals. The overall magnitude of the evoked [Ca2+]i transient increased with age in both groups, and the increase in the SHR group was larger, thus maintaining a difference between the 2 groups. This supports the idea that abnormal calcium regulation occurs early before hypertension takes place but that age alone does not fully explain the development of the calcium phenotype in the SHR when compared with the WKY rat.
The higher depolarization-evoked [Ca2+]i was not selective for SCGs, which predominately provide sympathetic innervation to the neck and cranial tissue,27 but it was also seen in neurons from the stellate ganglion that have a significant innervation into the heart. These data may at least partially explain why there is an increased local cardiac noradrenergic neurotransmission in the adult SHR when compared with the WKY rat.12,15 Moreover, emerging evidence has demonstrated enhanced NE release and heart rate responses to cardiac sympathetic activation in young prehypertensive SHRs in vitro.28
Evoked [Ca2+]i responses are influenced by multiple factors, including Ca2+ entry, extrusion across the plasma membrane, Ca2+ uptake and release from internal stores, and endogenous and exogenous Ca2+ buffering. From our results, there was no significant difference in the rate of rise of the [Ca2+]i transient in response to K+-induced depolarization between SHRs and WKY rats in all of the age groups. This suggests that the depolarization-induced Ca2+ entry through the plasma membrane was not different in sympathetic neurons from the SHR compared with the WKY rat at any age. However, we did not measure the membrane calcium current directly, so we cannot completely exclude some involvement of plasma membrane-bound Ca2+ channels.
ER Calcium Signaling in the SHR
Increasing evidence suggests that altered ER Ca2+ uptake, storage, and release play a central role in cardiac hypertrophy and heart failure.29 In the present study, before neurons were re-exposed to high KCl, ER Ca2+ stores were depleted by caffeine, and Ca2+ reuptake from SERCA pumps were prevented by thapsigargin. Under these conditions, the increased [Ca2+]i should represent a non-ER contribution. We found that the SHR group continued to have a higher increase in depolarization-evoked [Ca2+]i even after ER Ca2+ depletion. This result suggests that the non-ER sources may underlie the significantly higher depolarization-evoked [Ca2+]i transients seen in the SHR.
The larger ER Ca2+ store in the SHR was confirmed by directly measuring ER-free Ca2+ concentration using mag-fura-2/AM, followed by ER Ca2+ release/depletion with caffeine and the ER-Ca2+ chelator, TPEN. This finding itself does not necessarily indicate a larger ER contribution toward depolarization-evoked [Ca2+]i in the SHR, because the net ER contribution depends on the relative release and reuptake of Ca2+ by the ER.30,31 It does, however, imply a larger ER Ca2+ store that is potentially available for release on appropriate stimulation.
The rate at which SERCA moves Ca2+ across the ER membrane can be reduced by PLN, whereas phosphorylation of the PLN relieves its inhibition.32,33 Our results showed that the difference in total PLN was much larger than the difference in PLN-Ser16, suggesting that there was less nonphosphorylated PLN in the SHR than in the WKY rat and, therefore, less PLN-dependent inhibition of SERCA activity in the prehypertensive SHR. This is in keeping with the functional data showing shorter 50% and 90% decay times in [Ca2+]i.
RyRs are Ca2+ permeable channels that open in response to increase in [Ca2+]i34. We introduced the RyR activator caffeine to deplete Ca2+ from ER stores. Prehypertensive SHRs had a significantly higher caffeine-induced [Ca2+]i amplitude when compared with the age-matched WKY rat, indicating that RyRs in sympathetic neurons were upregulated in the young SHR. This was confirmed by more copies of RyR3 mRNA in the SHR. These data are consistent with observations by others in cardiac myocytes.35 Taken together, these findings suggest that sympathetic neurons from prehypertensive SHRs have more active/dynamic ER Ca2+ handling machinery.
Mitochondrial Calcium Signaling in the SHR
Mitochondria take up Ca2+ primarily through the mitochondrial calcium uniporter,36 which is modulated by both [Ca2+]i and the mitochondrial membrane potential (ΔΨm).37 The uniporter transports Ca2+ down the electrochemical gradient, and this gradient is maintained across the mitochondrial inner membrane without direct coupling to ATP hydrolysis or transport of other ions. In this study we found that TMRE uptake was reduced in the SHR neurons, which suggests that the membrane potential of mitochondria (ΔΨm) was more depolarized in the SHR. Depolarization of ΔΨm could lead to a reduced Ca2+ uptake by the mitochondria in the SHR.
A significant association of hypertension with mitochondrial uncoupling proteins has been reported both in experimental38 and human hypertensive states.39 Moreover, in experimental hypertension, a decreased activity of complex IV has been observed in the hypertrophied myocardium from the SHRs.40 However, we could not find any differences in protein levels of mitochondrial uncoupling protein 2, citrate synthase (used as a quantitative enzyme marker for the presence of intact mitochondria), and complexes I to V in SCG homogenates from young SHRs and WKY rats (Figure S3). This indicates that mitochondrial number and the electron transport chain were not changed in young SHRs. The differences in depolarization-evoked [Ca2+]i transient and the 50% and 90% decay times between SHRs and WKY rats were abolished by application of the proton uncoupler FCCP. These results suggest that mitochondria play a major role in the depolarization-evoked [Ca2+]i difference observed between the SHR and the WKY groups. Although we have no direct proof to show whether the ER and mitochondria contributed to the differences with age in [Ca2+]i transients in the SHR, our results are supportive of this idea when taken together with others41,42 that showed an age-related decline in SERCA function with a subsequent increased reliance on mitochondria to control high K+-evoked [Ca2+]i transients.
Results from this study suggest that the difference in Ca2+ homeostasis between sympathetic neurons of SHRs and WKY rats occurs early in the development and before the actual onset of hypertension itself. Impairment of [Ca2+]i handling was observed at 2 neural sites in the sympathetic nervous system, suggesting that this impairment may be widespread. A close link between faulty mitochondria Ca2+ release and reuptake appears to be central to the enhanced [Ca2+]i transients observed in prehypertensive SHRs. The precise molecular pathway underpinning this is not firmly established but warrants further investigation. Moreover, it would be desirable to see whether the changes that we report here are also seen in other animal models of hypertension. When all of the observations are viewed together with the current data, there is compelling evidence to suggest that alterations of Ca2+ homeostasis are central to sympathetic hyperactivity in the SHR, resulting in enhanced sympathetic neurotransmission at the end organ. The resultant chronotropic and inotropic actions of NE will contribute to raising cardiac output and arterial blood pressure. These pathways may be important targets to prevent sympathetic dysregulation that occur before the onset of hypertension.
Sources of Funding
This work was supported by a project grant from the British Heart Foundation (PG/08/061) and the Wellcome Trust OXION initiative. N.H. and D.J.P. acknowledge additional support from the British Heart Foundation Centre of Research Excellence, Oxford.
We thank Prof Sergey Kasparov and Dr Haibo Xu for providing neonatal SHRs and Dr Lisa Heather for her kind donation of primary antibody. We also thank Dr Mary McMenamin, Julia Shanks, Chieh-ju Lu, and Kate Wannop for technical assistance.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.111.186460/-/DC1.
- Received October 20, 2011.
- Revision received November 4, 2011.
- Accepted December 21, 2011.
- © 2012 American Heart Association, Inc.
- Ramchandra R,
- Hood SG,
- Denton DA,
- Woods RL,
- McKinley MJ,
- McAllen RM,
- May CN
- Grassi G,
- Dell'Oro R,
- Seravalle G,
- Foglia G,
- Trevano FQ,
- Mancia G
- Velez-Roa S,
- Ciarka A,
- Najem B,
- Vachiery JL,
- Naeije R,
- van de Borne P
- Flaa A,
- Mundal HH,
- Eide I,
- Kjeldsen S,
- Rostrup M
- Noll G,
- Wenzel RR,
- Schneider M,
- Oesch V,
- Binggeli C,
- Shaw S,
- Weidmann P,
- Luscher TF
- Cassaglia PA,
- Griffiths RI,
- Walker AM
- Li D,
- Wang L,
- Lee CW,
- Dawson TA,
- Paterson DJ
- Heaton DA,
- Lei M,
- Li D,
- Golding S,
- Dawson TA,
- Mohan RM,
- Paterson DJ
- Vanterpool CK,
- Vanterpool EA,
- Pearce WJ,
- Buchholz JN
- Greenwood JP,
- Stoker JB,
- Mary DA
- Goldstein DS
- Harzheim D,
- Movassagh M,
- Foo RS,
- Ritter O,
- Tashfeen A,
- Conway SJ,
- Bootman MD,
- Roderick HL
- Shanks J,
- Manou-Stathopoulou S,
- Lu CJ,
- Li D,
- Paterson DJ,
- Herring N
- Albrecht MA,
- Colegrove SL,
- Friel DD
- Simmerman HK,
- Jones LR
- MacDonnell SM,
- Kubo H,
- Crabbe DL,
- Renna BF,
- Reger PO,
- Mohara J,
- Smithwick LA,
- Koch WJ,
- Houser SR,
- Libonati JR
- Chalmers S,
- McCarron JG