Fundamental increase in pressure-dependent constriction of brain parenchymal arterioles from subarachnoid hemorrhage model rats due to membrane depolarization

Abstract

Intracerebral (parenchymal) arterioles are morphologically and physiologically unique compared with pial arteries and arterioles. The ability of subarachnoid hemorrhage (SAH) to induce vasospasm in large-diameter pial arteries has been extensively studied, although the contribution of this phenomenon to patient outcome is controversial. Currently, little is known regarding the impact of SAH on parenchymal arterioles, which are critical for regulation of local and global cerebral blood flow. Here diameter, smooth muscle intracellular Ca²⁺ concentration ([Ca²⁺]_i), and membrane potential measurements were used to assess the function of intact brain parenchymal arterioles isolated from unoperated (control), sham-operated, and SAH model rats. At low intravascular pressure (5 mmHg), membrane potential and [Ca²⁺]_i were not different in arterioles from control, sham-operated, and SAH animals. However, raising intravascular pressure caused significantly greater membrane potential depolarization, elevation in [Ca²⁺]_i, and constriction in SAH arterioles. This SAH-induced increase in [Ca²⁺]_i and tone occurred in the absence of the vascular endothelium and was abolished by the L-type voltage-dependent calcium channel (VDCC) inhibitor nimodipine. Arteriolar [Ca²⁺]_i and tone were not different between groups when smooth muscle membrane potential was adjusted to the same value. Protein and mRNA levels of the L-type VDCC Ca_V1.2 were similar in parenchymal arterioles isolated from control and SAH animals, suggesting that SAH did not cause VDCC upregulation. We conclude that enhanced parenchymal arteriolar tone after SAH is driven by smooth muscle membrane potential depolarization, leading to increased L-type VDCC-mediated Ca²⁺ influx.

cerebral blood flow is regulated by the diameter of resistance arteries and arterioles both on the surface of the brain and within the brain parenchyma. Parenchymal arterioles, unlike pial arteries and arterioles, lack extrinsic innervation and are encased by astrocytic processes (“endfeet”) (17, 27). The close association of this microvasculature with astrocytic endfeet is essential for functional hyperemia, whereby focal increases in neuronal activity are coupled to vasodilation of nearby arterioles and increased blood flow (6, 14, 27, 45). In addition to their role in neurovascular coupling, parenchymal arterioles also contribute significantly to autoregulation of global cerebral blood flow and account for ∼40% of total cerebral vascular resistance (12). Thus parenchymal arterioles are a unique vascular bed, central in the regulation of both local and global cerebral blood flow.

Pial arteries and arterioles constrict to physiological increases in intravascular pressure via a process involving vascular smooth muscle membrane potential depolarization, activation of voltage-dependent Ca²⁺ channels (VDCCs), and increased global intracellular Ca²⁺ concentration ([Ca²⁺]_i) (1, 13, 19, 32, 44). Recent work has established that pressure-dependent constriction, or myogenic tone, is significantly greater in isolated parenchymal arterioles than in pial arteries (3). The cellular basis for enhanced pressure-dependent myogenic tone in the parenchymal vasculature is unclear. Furthermore, the relationship between intravascular pressure, membrane potential, [Ca²⁺]_i, and diameter of isolated parenchymal arterioles is not known.

Dysfunction of the cerebral microcirculation, including parenchymal arterioles, has been implicated in a number of brain pathologies, including vascular dementia, Alzheimer disease, and the development of delayed neurological deficits associated with cerebral aneurysm rupture and subarachnoid hemorrhage (SAH) (26, 65). A number of studies have suggested that SAH causes enhanced constriction of the microcirculation to limit parenchymal blood flow (34, 51, 58, 59, 65). However, the small size of parenchymal arterioles (maximum luminal diameter: 30–60 μm) poses a significant technical challenge to direct examination of their function.

The objective of the present study was to elucidate the relationship between intravascular pressure, [Ca²⁺]_i, membrane potential, and diameter of isolated parenchymal arterioles from unoperated, sham-operated, and SAH model rats. We report that after SAH, augmented pressure-dependent depolarization of parenchymal arteriolar smooth muscle causes elevated [Ca²⁺]_i and enhanced vasoconstriction. Furthermore, this enhanced constriction of parenchymal arterioles after SAH occurs independently of endothelial dysfunction and is abolished by smooth muscle hyperpolarization or direct inhibition of L-type VDCCs.

METHODS

Animals.

Sprague-Dawley rats (males, 300–350 g, 10–15 wk; Charles River Laboratories, Saint Constant, QC, Canada) were used in this study. All experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Pub. No. 85-23, revised 1996) and followed protocols approved by the Institutional Animal Use and Care Committee of the University of Vermont.

Rat SAH model.

Animals were initially anesthetized by isoflurane (5%) with an induction chamber and then maintained on isoflurane (2–2.5%) anesthesia with the aid of a nose cone. A small (0.5–1.5 cm), longitudinal, midline suboccipital incision was centered over the foramen magnum, and the neck muscles were dissected until dura was visualized. Autologous unheparinized blood (0.5 ml) drawn from the tail artery was injected into the cisterna magna (day 0) with a 25-gauge butterfly needle (37). The animal was then positioned on an incline board at a 45° angle with the head down in neutral position for 30 min. Twenty-four hours later, a second injection of blood was delivered by repeating the above procedure. Sham-operated animals were treated in a similar manner with the exception that injection of 0.5 ml of saline was substituted for injection of 0.5 ml of unheparinized blood. Buprenorphine (0.01 mg/kg) was given every 12 h (for 36 h, then as needed) as an analgesic. On day 4, animals were euthanized under deep pentobarbital anesthesia by decapitation. Parenchymal arterioles (30–60 μm in diameter) were dissected from the middle cerebral artery territory by a method previously described by Dacey and Duling (7). Briefly, the arachnoid mater around the middle cerebral artery was gently removed from the brain such that parenchymal arterioles (2–4 mm in length) were removed from the brain while attached to the middle cerebral artery. Arterioles were dissected in cold (4°C) 3-(N-morpholino)propanesulfonic acid (MOPS)-buffered saline solution of the following composition (in mM): 145 NaCl, 5 KCl, 1 MgSO₄, 2.5 CaCl₂, 1 KH₂PO₄, 0.02 EDTA, 3 MOPS, 2 pyruvate, 5 glucose, 1% bovine serum albumin, pH 7.4. Arteriole segments used for this study were located downstream of the Virchow-Robin space at depths >1 mm from the surface of the cortex.

Measurement of intracellular Ca²⁺ concentration and diameter.

Freshly isolated parenchymal arterioles were cannulated on glass micropipettes mounted in a 5-ml myograph chamber (University of Vermont Instrumentation and Model Facility) as described previously (7, 29). Vascular endothelium was removed from some arterioles by passing an air bubble through the lumen after cannulation of one end (62). After cannulation, arterioles were loaded with the ratiometric Ca²⁺-sensitive dye fura-2 [acetoxymethyl ester (AM) membrane-permeant form] by incubating in MOPS-buffered saline solution containing fura-2 AM (5 μM) (Invitrogen, Carlsbad, CA) with pluronic acid (0.1%) (Invitrogen) at room temperature (∼22°C) for 45 min. Next, the myograph chamber was mounted on a Nikon TE2000-S inverted fluorescence microscope. To allow for equilibration and deesterification of fura-2 AM, arterioles were pressurized to 5 mmHg and continuously superfused (37°C, 30 min) with artificial cerebrospinal fluid (aCSF) of the following composition (in mM): 125 NaCl, 3 KCl, 18 NaHCO₃, 1.25 NaH₂PO₄, 1 MgCl₂, 2 CaCl₂, 5 glucose aerated with 5% CO₂, 20% O₂, 75% N₂. Bath pH was closely monitored and maintained at 7.30–7.35 (9).

After the equilibration period, studies were performed in which intravascular pressure was increased to 10, 20, 40, and 60 mmHg or extracellular K⁺ concentration ([K⁺]_o) was increased from 3 mM to 60 mM with aCSF made by isosmotic replacement of NaCl with KCl. Fluorescence ratio was obtained from the background-corrected ratio of the 510-nm emission from arterioles alternately excited at 340 and 380 nm with hardware and software developed by IonOptix (Milton, MA). [Ca²⁺]_i was estimated with the following equation (16): [Ca²⁺] = K_d × β × (R − R_min)/(R_max − R). R_min and R_max, the ratios of emission signals under Ca²⁺-free and Ca²⁺-saturated conditions, respectively, were measured from a separate set of ionomycin-treated arterioles, and β was determined as the ratio of F₃₈₀ intensities at R_min and R_max. Calibration values were not significantly different between arterioles obtained from healthy and SAH animals (Table 1) and were pooled for calculations of [Ca²⁺]_i using an apparent dissociation constant (K_d) of 282 nM of fura-2 for Ca²⁺ (32). Total fluorescence counts (photomultiplier signal at F₃₄₀ + F₃₈₀) measured immediately after the equilibration period were not significantly different between endothelium-intact (539 ± 66 counts) and denuded (532 ± 40 counts) arterioles, suggesting that endothelial loading with fura-2 was negligible.

Table 1.

Fura-2 calibration values

	n	Rmin	Rmax	β
Control	4	0.371 ± 0.008	9.969 ± 0.565	8.031 ± 0.895
SAH	4	0.378 ± 0.005	10.064 ± 0.795	7.764 ± 0.370
Pooled	8	0.374 ± 0.004	10.017 ± 0.483	7.898 ± 0.451

Values are means ± SE for n arterioles. Ratios of emission signals under Ca²⁺-free and Ca²⁺-saturated conditions (R_min and R_max, respectively) and ratio of F₃₈₀ intensities at R_min and R_max (β) were determined from a separate set of arterioles (see methods). No differences were found among control and subarachnoid hemorrhage (SAH) arterioles. Pooled values shown were used for analysis of intracellular Ca²⁺ concentration measurements.

Luminal diameter of arterioles was simultaneously recorded with [Ca²⁺]_i with a charge-coupled device (CCD) camera and the edge-detection function of IonOptix software. Constriction is presented as a decrease in arterial diameter relative to the maximum diameter at a given pressure obtained at the end of each experiment using Ca²⁺-free aCSF containing nimodipine (300 nM) (29, 66).

Measurement of membrane potential.

Arterioles were cannulated as described above and pressurized to either 5 or 40 mmHg. Smooth muscle membrane potential was measured by insertion of a sharp glass microelectrode (∼100-MΩ resistance) filled with 0.5 M KCl into the vessel wall. The criteria for successful impalement were 1) an abrupt negative potential deflection upon entry, 2) a stable membrane potential for ≥30 s, and 3) an abrupt positive potential deflection upon removal (32, 53). Measurements were made with an electrometer (World Precision Instruments) and recorded via computer with Axotape and Dataq software.

Quantitative real-time polymerase chain reaction.

Total RNA was extracted with RNA STAT-60 total RNA/mRNA isolation reagent (Tel-test, Friendswood, TX) and reverse transcribed to cDNA with the SuperScript First-Strand synthesis system (Invitrogen) (49). Polymerase chain reaction was performed with primers detecting Ca_V1.2 (GenBank accession no. AY974797; forward: 5′-CAGCAGGTCCTATGTCAGCA-3′, reverse: 5′-CCGACAGCAGTGAATGAAGA-3′) and 18S (GenBank accession no. M11188; forward: 5′-AGTCGCCGTGCCTACCAT-3′, reverse: 5′-GCCTGCTGCCTTCCTTG-3′). Amplification was performed with SYBR Green JumpStart Taq ReadyMix (Sigma, St. Louis, MO) and a real-time PCR system (Applied Biosystems, Carlsbad, CA). Quantification was performed by comparison of threshold cycles (C_T) relative to a standard curve constructed from serially diluted plasmids containing target sequences. Expression levels were normalized to 18S rRNA, which was not statistically different between arterioles isolated from SAH animals and controls [C_T = 15.5 ± 0.3 cycles (n = 5) vs. C_T = 16.1 ± 0.6 cycles (n = 6), respectively; 500 ng total RNA].

Western blot analysis.

Parenchymal arterioles were pooled from two animals per sample (n). Whole parenchymal arteriole lysates were prepared by sonication (20 min, 4°C) in a Triton lysis buffer solution containing (in mM) 50 HEPES, 150 NaCl, 1 EGTA, and 1.5 MgCl₂, with 1% Triton X, 10% glycerol, and protease inhibitors (10 μg/ml of leupeptin, aprotinin, and antipain and 10 mM phenylmethylsulfonyl fluoride). Tissue debris and nuclear fragments were removed by centrifugation at 7,000 rpm (10 min, 4°C). The whole cell lysate was obtained as the supernatant, and protein concentration was determined by Bradford assay. Molecular weight marker (SeeBluePlus2 prestained standard, Invitrogen) was loaded as a size standard, and protein samples (15 μg total protein) were separated under reducing conditions on a 7.5% polyacrylamide gel (Bio-Rad, Hercules, CA) by electrophoresis. Samples were run at 120 V for 1.5 h on an 8 × 10-cm electrophoresis cell (Bio-Rad) and electrophoretically transferred to a nitrocellulose membrane at 80 V for 2 h (4°C). The membranes were washed in Tris-buffered saline with 0.1% Tween 20 (TBS-T) and blocked with 10% nonfat dried milk in TBS-T (1 h, room temperature). Membranes were then incubated with the specific polyclonal antibodies against Ca_V1.2 (1:200, overnight, 4°C; Alomone, Jerusalem, Israel) or monoclonal antibody against GAPDH (1:100,000, 30 min, room temperature; Sigma) prepared in 5% nonfat dried milk in TBS-T. Membranes were then incubated (1 h, room temperature) with horseradish peroxidase-labeled donkey anti-rabbit IgG (Ca_V1.2, 1:5,000; GAPDH, 1:10,000, GE Healthcare, Waukesha, WI) in TBS-T containing 5% nonfat dried milk. Bands were identified by chemiluminescence and exposed to X-ray films. Densitometry for immunoreactive bands was performed with National Institutes of Health software (ImageJ), and density was expressed as percentage of GAPDH density for each lane. The quantity of GAPDH protein, normalized to total protein, was similar between control and SAH groups.

Statistical analysis.

Values are presented as means ± SE. One-way analysis of variance followed by Tukey multiple comparison test was used in the comparison of multiple groups. Student's t-test was used in the comparison of two groups. Statistical significance was considered at the level of P < 0.05 or P < 0.01.

RESULTS

Parenchymal arterioles isolated from SAH model animals exhibit elevated cytosolic Ca²⁺ and enhanced tone at physiological intravascular pressures.

To assess the role of cytosolic Ca²⁺ in pressure-induced parenchymal arteriolar constriction, simultaneous measurements of [Ca²⁺]_i and diameter were obtained from freshly isolated parenchymal arterioles loaded with the ratiometric Ca²⁺ indicator fura-2. In response to stepwise increases in intravascular pressure, arterioles from unoperated (control), sham-operated, and SAH animals exhibited graded increases in cytosolic Ca²⁺ ([Ca²⁺]_i) and active constriction. At low intravascular pressure (5–10 mmHg), similar levels of [Ca²⁺]_i and tone were observed in vessels isolated from all groups. However, as intravascular pressure was increased to physiological levels (i.e., 20–60 mmHg) (12, 21), arterioles from SAH animals exhibited markedly enhanced levels of [Ca²⁺]_i and constriction compared with arterioles from control and sham-operated animals (Fig. 1, A–E). For example, at 40 mmHg, [Ca²⁺]_i was ∼33% higher in arterioles from SAH animals (349 ± 11 nM, n = 5) compared with arterioles isolated from control animals (261 ± 8 nM, n = 6). Corresponding to this SAH-induced increase in [Ca²⁺]_i, at 40 mmHg arterioles from SAH animals constricted 29 ± 2 μm (53 ± 4% decrease in diameter, n = 5) compared with a constriction of 14 ± 1 μm in arterioles from control animals (29 ± 2% decrease in diameter, n = 6) (Table 2). Although [Ca²⁺]_i and tone at a given pressure were greater in arterioles from SAH model animals, the relationships between [Ca²⁺]_i and constriction were remarkably similar between arterioles isolated from control and SAH animals (Fig. 1F). These data suggest that elevated [Ca²⁺]_i is sufficient to account for enhanced pressure-induced constriction of parenchymal arterioles from SAH model animals.

Fig. 1.

Elevated cytosolic Ca²⁺ and enhanced myogenic tone in parenchymal arterioles from subarachnoid hemorrhage (SAH) animals. A–C: representative simultaneous intracellular Ca²⁺ concentration ([Ca²⁺]_i) and diameter measurements obtained from intact arterioles isolated from unoperated (control; A), sham-operated (Sham; B), and SAH (C) animals. Recordings were obtained during stepwise increases in intravascular pressure and subsequent nimodipine application (Nim, 300 nM) at 60 mmHg. Dashed traces represent diameters in Ca²⁺-free artificial cerebrospinal fluid (aCSF) containing nimodipine (300 nM). D and E: summary of [Ca²⁺]_i (D) and constriction (E) obtained from unoperated (control; n = 6), Sham (n = 6), and SAH (n = 5) animals in the absence and presence of 300 nM nimodipine. Mean nimodipine data were obtained from experiments in which nimodipine was applied before increasing pressure in a separate set of arterioles. Myogenic tone is expressed as % decrease from the passive diameter (obtained in Ca²⁺-free aCSF with 300 nM nimodipine) of arterioles at a given intravascular pressure. Arteriolar [Ca²⁺]_i or diameter was not statistically different between groups at any intravascular pressure tested in Ca²⁺-free aCSF with nimodipine. *P < 0.05, **P < 0.01 vs. control unoperated and sham operated. F: relationship between [Ca²⁺]_i and constriction for arterioles isolated from control, Sham, and SAH animals derived from summary data depicted in D and E. The concentrations of Ca²⁺ causing half-maximal arteriolar constriction, obtained from Boltzmann fit of data [using 60 mM extracellular K⁺ concentration ([K⁺]_o) as maximum; see Fig. 3], were similar between groups (Control: 226 ± 26 nM, n = 5; Sham: 240 ± 6 nM, n = 6; SAH: 240 ± 10 nM, n = 5).

Table 2.

Active and passive diameters of parenchymal arterioles isolated from control, sham-operated, and SAH animals

	Control		Sham Operated		SAH
Intravascular Pressure	Active	Passive	Active	Passive	Active	Passive
5 mmHg	39 ± 2	41 ± 2	41 ± 5	45 ± 6	43 ± 3	46 ± 3
10 mmHg	39 ± 2	43 ± 2	44 ± 5	49 ± 6	40 ± 2	48 ± 1
20 mmHg	38 ± 2	45 ± 3	44 ± 6	53 ± 7	36 ± 3	51 ± 1
40 mmHg	36 ± 3	49 ± 3	42 ± 5	57 ± 8	26 ± 2*	55 ± 2
60 mmHg	33 ± 3	50 ± 3	37 ± 5	58 ± 9	25 ± 2*	56 ± 2
+Nimodipine	49 ± 4		57 ± 8		55 ± 2

Values (in μm) are means ± SE.

^*P < 0.05 vs. control active and sham operated active.

At 60 mmHg, the dihydropyridine L-type VDCC antagonist nimodipine (300 nM) reduced arteriolar [Ca²⁺]_i to basal levels and abolished pressure-induced constrictions in arterioles from control, sham-operated, and SAH animals (Fig. 1 A–C). To further assess the role of L-type VDCCs in SAH-induced elevation of [Ca²⁺]_i and arteriolar tone, these parameters were examined by applying nimodipine (300 nM) before increasing intravascular pressure. Pressure-induced increases in [Ca²⁺]_i and constriction were abolished by nimodipine in control and SAH groups (Fig. 1, D and E, Table 2), indicating that Ca²⁺ influx in parenchymal myocytes from both control and SAH animals is mediated by L-type VDCCs. These results indicate that elevation of intravascular pressure leads to greater L-type VDCC-mediated Ca²⁺ influx, increased [Ca²⁺]_i, and enhanced constriction of parenchymal arterioles from SAH animals.

Previous reports have indicated an increase in the expression of R-type (Ca_V2.3) and T-type (Ca_V3.1 and Ca_V3.3) VDCCs in pial cerebral artery myocytes after experimental SAH (30, 38, 47, 61). To determine whether these channels also contribute to elevated [Ca²⁺]_i and enhanced constriction in parenchymal arterioles after SAH, we examined the effect of the R-type VDCC inhibitor SNX-482 and the T-type VDCC inhibitor mibefradil on arteriolar [Ca²⁺]_i and tone. When arterioles were pressurized to 40 mmHg, application of SNX-482 (200 nM) had no significant effects on arteriolar [Ca²⁺]_i or tone in either the control or the SAH group (Fig. 2). Considering that mibefradil inhibits L-type and T-type channels at similar concentrations (36, 48), this agent was applied subsequent to selective inhibition of L-type channels with 100 nM nimodipine (48). No further reduction in [Ca²⁺]_i or dilation was observed when mibefradil was applied after application of nimodipine (Fig. 2). These results suggest that the elevated [Ca²⁺]_i and enhanced parenchymal arteriolar tone following SAH are due to increased activity of L-type VDCCs rather than R-type or T-type VDCCs.

Fig. 2.

Elevated [Ca²⁺]_i and enhanced constriction of parenchymal arterioles after SAH are due to increased activity of L-type voltage-dependent Ca²⁺ channels (VDCCs). Summary of [Ca²⁺]_i (A) and constriction (B) data obtained from arterioles isolated from control (n = 4) and SAH (n = 4) animals. Arterioles were pressurized to 40 mmHg and allowed to develop stable myogenic tone before application of the R-type VDCC inhibitor SNX-482 (200 nM). After SNX-482 was washed from the bath (∼15 min), nimodipine (100 nM) was applied, followed by subsequent addition of the T-type VDCC inhibitor mibefradil (10 μM). **P < 0.01 vs. control; ^NSP > 0.05 vs. control.

Increased L-type VDCC activity, rather than increased channel expression, mediates enhanced parenchymal arteriolar constriction after SAH.

SAH-induced Ca²⁺ influx via L-type VDCCs could reflect either an increase in the density of functional channels or a more depolarized membrane potential leading to an increase in the open-state probability (P_open) of existing channels. To explore whether upregulation of L-type VDCC expression occurs after SAH, quantitative real-time PCR was used to examine Ca_V1.2 mRNA, encoding the predominantly expressed L-type VDCC pore-forming α₁-subunit in vascular smooth muscle (2, 43). Expression of Ca_V1.2 mRNA, normalized to 18S rRNA levels, was similar in parenchymal arterioles isolated from control and SAH animals (Fig. 3A). In addition, protein levels of Ca_V1.2, determined by Western blot, were not different between groups (Fig. 3, B and C). These data suggest that upregulation of Ca_V1.2 does not contribute to elevated [Ca²⁺]_i after SAH.

Fig. 3.

Ca_V1.2 mRNA and protein expression is unchanged after SAH. Quantitative real-time PCR and Western blot were used to measure Ca_V1.2 mRNA and protein, respectively. A: summary of quantitative real-time PCR data obtained from arterioles isolated from control (n = 6) and SAH (n = 5) animals. Ca_V1.2 mRNA is expressed relative to 18S rRNA. ^NSP > 0.05 vs. control. B: representative Western blot images for Ca_V1.2 (∼200–240 kDa) and GAPDH (∼36 kDa) using whole cell lysates of parenchymal arterioles (each lane represents pooled arterioles from 2 animals). GAPDH was used as an internal standard. C: summary of Western blot data obtained by densitometry using ImageJ software (NIH). Ca_V1.2 protein levels were normalized to GAPDH and expressed as fold change compared with corresponding control in each experiment (n = 5). ^NSP > 0.05 vs. control.

On the basis of the above data, we hypothesized that greater smooth muscle membrane potential depolarization, rather than increased VDCC expression, underlies enhanced pressure-induced constriction of parenchymal arterioles after SAH. Thus, at a given membrane potential, similar [Ca²⁺]_i and constriction would be predicted for arterioles from control and SAH animals. To test this possibility, [Ca²⁺]_i and constriction were measured when [K⁺]_o was elevated from 3 mM to 60 mM by isosmotic replacement of NaCl with KCl in the bath solution. In pial arteries, membrane potential closely follows the K⁺ equilibrium potential (E_K) at [K⁺]_o > 16 mM (24, 32). This also appears to be the case with parenchymal arterioles, as smooth muscle membrane potential in arterioles bathed in 60 mM [K⁺]_o was very close to theoretical E_K (control: −22 ± 1 mV, n = 4; SAH: −21 ± 1 mV, n = 4), predicted by the Nernst equation (−22 mV; assuming intracellular [K⁺] = 140 mM) (Fig. 4A). When smooth muscle membrane potential was “clamped” with 60 mM [K⁺]_o, [Ca²⁺]_i was not different in parenchymal arterioles isolated from healthy (393 ± 18 nM, n = 7) and SAH (388 ± 14 nM, n = 5) animals (Fig. 4B). Constriction was also similar for control and SAH arterioles in the presence of 60 mM [K⁺]_o (Fig. 4C). Elevations in [Ca²⁺]_i and constrictions in response to 60 mM [K⁺]_o were abolished by nimodipine (300 nM) in arterioles from both groups (Fig. 4, B and C). These findings suggest that enhanced pressure-induced constrictions following experimental SAH likely reflect differences in smooth muscle membrane potential rather than increased VDCC expression.

Fig. 4.

Elevated arteriolar Ca²⁺ and enhanced constriction after SAH depend on membrane potential depolarization. A: summary membrane potential data obtained from isolated pressurized (5 mmHg) parenchymal arterioles bathed in 60 mM [K⁺]_o. Membrane potential closely matched the theoretical equilibrium potential (E_K) of −22 mV in arterioles from both control (n = 4) and SAH (n = 4) animals. B and C: summary data showing [Ca²⁺]_i (B) and constriction (C; % decrease in diameter) in control (n = 7) and SAH (n = 5) arterioles exposed to aCSF containing either 3 mM or 60 mM [K⁺]_o. Nimodipine (300 nM) reduced [Ca²⁺]_i to basal levels and dilated arterioles. ^NSP > 0.05 vs. control.

Elevation of intravascular pressure leads to greater smooth muscle membrane potential depolarization after experimental subarachnoid hemorrhage.

To directly assess the impact of SAH on smooth muscle membrane potential, myocytes of intact parenchymal arterioles were impaled from the adventitial surface with intracellular microelectrodes. At low intravascular pressure (5 mmHg), membrane potential was similar in arterioles isolated from control (−58 ± 2 mV, n = 4) and SAH (−59 ± 2 mV, n = 4) animals (Fig. 5A). Raising intravascular pressure from 5 to 40 mmHg resulted in membrane potential depolarization in arterioles from both groups. However, parenchymal arteriolar myocytes from SAH animals were significantly more depolarized at 40 mmHg (−28 ± 1 mV, n = 4) compared with arterioles from control animals (−35 ± 1 mV, n = 5) (Fig. 5, B and C). These data suggest that greater pressure-dependent membrane potential depolarization underlies enhanced parenchymal arteriolar tone after SAH.

Fig. 5.

Myocytes of pressurized parenchymal arterioles are depolarized after SAH. A and B: representative smooth muscle membrane potential measurements in arterioles isolated from control and SAH animals at 5 mmHg (A) and 40 mmHg (B) intravascular pressure. Scale bars represent 10 s. C: summary of membrane potential measurements obtained at 5 and 40 mmHg. At 40 mmHg, myocytes of arterioles isolated from SAH animals were significantly more depolarized (−28 ± 1 mV, n = 4) compared with myocytes of arterioles isolated from control animals (−35 ± 1 mV, n = 5). **P < 0.01 vs. control. ^NSP > 0.05 vs. control.

Endothelial function promotes vasodilation in parenchymal arterioles after SAH.

Membrane potential of arteriolar myocytes can be modulated by the vascular endothelium (13). Previous work has indicated that endothelial small (SK_Ca)- and intermediate (IK_Ca)-conductance Ca²⁺-activated K⁺ channel activity promotes smooth muscle membrane potential hyperpolarization and plays an important role in the regulation of parenchymal arteriolar tone and cortical blood flow (4, 18). To assess endothelial function in parenchymal arterioles after SAH, we examined the effect of the inhibitors of SK_Ca and IK_Ca channels, apamin and TRAM-34, respectively, on parenchymal arteriolar tone. At 40 mmHg, application of apamin (300 nM) and TRAM-34 (1 μM) caused a similar level of constriction in arterioles isolated from control (11 ± 2%, n = 4) and SAH (10 ± 1%, n = 6) animals, suggesting that these channels are functional and their basal activity is unchanged after SAH. In addition, we observed that an opener of SK_Ca and IK_Ca channels, NS-309 (1 μM) (25, 57), hyperpolarized smooth muscle membrane potential to similar levels in endothelium-intact arterioles from control (−67 ± 2 mV, n = 5) and SAH (−67 ± 2 mV, n = 4) animals. Consistent with membrane potential hyperpolarization, NS-309 significantly decreased [Ca²⁺]_i in arterioles from control (242 ± 7 nM to 141 ± 15 nM, n = 4) and SAH (330 ± 11 nM to 151 ± 14 nM, n = 4) animals and induced >90% vasodilation in both groups (Fig. 6, A and C). These results demonstrate that activation of endothelial SK_Ca and IK_Ca channels causes a similar marked membrane potential hyperpolarization and vasodilation of arterioles from control and SAH animals.

Fig. 6.

Elevated [Ca²⁺]_i and enhanced myogenic tone after SAH is independent of endothelial function. A and B: summary of [Ca²⁺]_i in the absence and presence of NS-309 (1 μM) for pressurized (40 mmHg) endothelium-intact (+endothelium; A) and endothelium-denuded (−endothelium; B) parenchymal arterioles from control (+endothelium: n = 4; −endothelium: n = 5) and SAH (+endothelium: n = 4; −endothelium: n = 5) animals. Endothelium was removed by passing an air bubble through the vessel lumen for ∼30 s. C and D: summary of myogenic tone data (% decrease in diameter) in the absence and presence of NS-309 obtained from endothelium-intact and endothelium-denuded arterioles from control and SAH animals. **P < 0.01 vs. control. ^NSP > 0.05 vs. control.

To further examine whether decreased basal endothelial function contributes to elevated [Ca²⁺]_i and enhanced constriction of parenchymal arterioles from SAH animals, [Ca²⁺]_i and diameter were measured in endothelium-denuded vessels. The endothelium was removed from parenchymal arterioles by passing an air bubble through the vessel lumen during the cannulation procedure. Although endothelium-denuded arterioles displayed increased [Ca²⁺]_i and constriction compared with corresponding endothelium-intact vessels in both groups, these parameters remained significantly higher in parenchymal arterioles from SAH animals (Fig. 6, B and D). Application of NS-309 was used to confirm absence of endothelial function in denuded vessels. Consistent with expression of SK_Ca and IK_Ca channels in endothelial cells but not smooth muscle, NS-309 had no effect on [Ca²⁺]_i or diameter in denuded arterioles from either control or SAH animals (Fig. 6, B and D). These results suggest that SAH-induced elevated [Ca²⁺]_i and enhanced arteriolar tone result from a direct modulation of smooth muscle membrane potential rather than decreased endothelial vasodilatory influence.

DISCUSSION

Here we report that cytosolic Ca²⁺ is elevated and pressure-dependent constriction is enhanced in parenchymal arterioles from experimental SAH animals. Our data demonstrate that elevations in intravascular pressure cause greater membrane potential depolarization after SAH, leading to increased activity of L-type VDCCs and enhanced constriction. Furthermore, this study suggests that SAH-induced differences in parenchymal arteriolar [Ca²⁺]_i and tone were independent of vascular endothelial dysfunction.

This work also provides novel information regarding the physiological function of arterioles within the brain parenchyma in the absence of SAH. Our data, consistent with other recent studies (3, 4), demonstrate that parenchymal arterioles develop significantly greater pressure-induced myogenic tone at lower intravascular pressures than pial surface arteries, such as middle cerebral arteries (Fig. 7A). Further comparison of our present data with those previously published by Knot and Nelson (32) from rat middle cerebral arteries indicates that at a given intravascular pressure parenchymal arteriolar myocytes are more depolarized and exhibit greater levels of [Ca²⁺]_i than pial artery myocytes. For example, at 40 mmHg, membrane potential of parenchymal myocytes was approximately −35 mV compared with −50 mV in myocytes of middle cerebral arteries (32). At a given level of intravascular pressure, parenchymal arterioles also display greater [Ca²⁺]_i compared with middle cerebral arteries (Fig. 7B) (i.e., 260 nM vs. ∼160 nM, respectively, at 40 mmHg intravascular pressure). However, the relationships between [Ca²⁺]_i and constriction are similar in these vascular beds (Fig. 7C), suggesting that in healthy animals enhanced pressure-dependent constriction observed in parenchymal arterioles results from a greater elevation in [Ca²⁺]_i rather than altered Ca²⁺ sensitivity of smooth muscle contractility. Thus the relationship between intravascular pressure and membrane potential appears to be shifted toward lower pressures in parenchymal arterioles. We propose that greater membrane potential depolarization at lower intravascular pressures may be critical for development of parenchymal arteriolar tone at the lower physiological pressures experienced in vivo.

Fig. 7.

Relationship between intravascular pressure, cytosolic Ca²⁺, and diameter of parenchymal arterioles vs. pial arteries from control animals. A: relationship between intravascular pressure and constriction for parenchymal arterioles and middle cerebral arteries [MCA; obtained from results published by Knot and Nelson (32)]. Constriction is expressed as % of maximal constriction obtained in 60 mM [K⁺]_o (see Fig. 4). B: relationship between intravascular pressure and [Ca²⁺]_i for parenchymal arterioles and middle cerebral arteries (32). C: relationship between [Ca²⁺]_i and constriction for parenchymal arterioles and middle cerebral arteries (32).

We also demonstrate that the relationship between intravascular pressure and [Ca²⁺]_i is further shifted to the left in parenchymal arterioles after SAH. This effect is correlated with enhanced pressure-induced depolarization at physiological intravascular pressures in arterioles from SAH animals. The membrane potentials reported here for parenchymal arteriolar myocytes (±SAH) are within the range in which VDCC P_open is steeply voltage dependent (44). It is estimated that a depolarization of 7 mV observed after SAH (−35 to −28 mV; Fig. 5C) could more than double VDCC P_open from 0.016 to 0.035 (44). Therefore, the depolarization observed after SAH in this study is consistent with a substantial increase in VDCC activity, leading to a marked increase in Ca²⁺ entry in parenchymal arteriolar myocytes. It remains possible that VDCCs may also be more active at a given membrane potential after SAH, for example, due to channel phosphorylation via increased protein kinase activity. However, the observation that intracellular Ca²⁺ in smooth muscle cells from healthy and SAH animals was similar at the same membrane potentials argues against a significant change in VDCC open probability at a given membrane potential.

Vascular smooth muscle membrane potential can be modulated by various vasoactive factors released by the endothelium. A number of studies suggest that nitric oxide (NO)-mediated vasodilation is impaired in large-diameter cerebral arteries in experimental models of SAH (5, 13). Similar findings have been reported in basilar arteries of SAH patients (23). However, the contribution of endothelial dysfunction to cerebral artery constriction is controversial, and others have reported that endothelium-dependent relaxation is unchanged after SAH (8, 39). The relative contribution of endothelium-dependent vasodilators to vascular tone may depend on vessel size, and the effects of SAH on endothelial function may differ between large- and small-diameter vessels. Whereas basal NO production seems to control tone in both cerebral arteries and arterioles, the contribution of endothelium-derived hyperpolarizing factor (EDHF) to blood flow regulation may be greater at the level of the microcirculation (64). Previous studies suggest that EDHF-mediated relaxation requires the activity of endothelial SK_Ca and IK_Ca channels (10, 67), and these channels are involved in modulating basal tone in parenchymal arterioles but not in middle cerebral arteries (4). Here we show that constrictions following SK_Ca and IK_Ca inhibition are similar between control and SAH arterioles. Thus, while EDHF activation following endothelial stimulation (i.e., ATP) may be impaired in parenchymal arterioles after experimental SAH (52), our data suggest that the basal activity of these channels in the cerebral microcirculation is unaltered. Furthermore, activation of endothelial SK_Ca and IK_Ca channels with NS-309 caused a reduction of [Ca²⁺]_i to basal levels and profound dilation of parenchymal arterioles from both control and SAH animals. These results are consistent with activation of endothelial SK_Ca and IK_Ca channels by NS-309 hyperpolarizing electrically coupled smooth muscle cells toward E_K. In addition, endothelium removal did not ablate SAH-induced pressure-induced increases in [Ca²⁺]_i and myogenic tone. Therefore, reduced endothelial vasodilator influence cannot account for enhanced vasoconstriction of parenchymal arterioles after SAH. These results suggest that a direct alteration of smooth muscle membrane potential regulation contributes to dysfunction of intracerebral arterioles from SAH animals.

Our results indicate that elevated parenchymal arteriolar [Ca²⁺]_i and tone after SAH depend on intravascular pressure. Accordingly, the effects of SAH on parenchymal arteriolar constriction observed in the present study may be exacerbated by elevated blood pressures commonly observed in SAH (15, 40). At this time, it is unclear how SAH causes enhanced pressure-dependent depolarization in parenchymal arterioles. Activation of mechanosensitive nonselective cation channels belonging to the transient receptor potential family (i.e., TRPC6, TRPM4) is reported to contribute to pressure-dependent membrane potential depolarization of cerebral artery myocytes (11, 63). Thus increased expression or activity of these channels after SAH may lead to membrane depolarization and enhanced VDCC activity. On the other hand, SAH-induced depolarization may also result from suppressed hyperpolarizing influence. For example, significant voltage-dependent K⁺ (K_V) channel-mediated current is observed at physiological membrane potentials (54), and activation of these channels represents an important feedback mechanism in response to pressure-induced depolarization. Consistent with this hypothesis, inhibition of K_V channels with 4-aminopyridine causes membrane potential depolarization and vasoconstriction (33, 56). Numerous studies have suggested that substances within blood may lead to membrane potential depolarization of cerebral artery myocytes via suppression of K⁺ channel activity (13, 20, 55). For example, using rabbit pial arteries, our laboratory has shown (28, 35) that acute exposure of vascular smooth muscle to the blood component oxyhemoglobin can cause internalization of K_V1.5 via a mechanism involving protein tyrosine kinase activation. Others have additionally demonstrated that mRNA and protein levels of K_V2.2 channels were decreased in basilar arteries from a canine SAH model (31). Further studies are required to determine whether K_V channel suppression or additional mechanisms contribute to enhanced pressure-dependent depolarization of parenchymal arteriolar myocytes in SAH animals.

Our present findings indicate a dominant role for L-type VDCC-mediated Ca²⁺ influx in enhanced parenchymal arteriolar constriction following SAH. Previous reports have demonstrated upregulated expression of R-type (Ca_V2.3) and T-type (Ca_V3.1 and Ca_v3.3) channels in pial cerebral arteries after experimental SAH (30, 47, 61). In contrast to these findings, we observed no effect on arteriolar [Ca²⁺]_i or diameter of the R-type VDCC inhibitor SNX-482 or the T-type VDCC inhibitor mibefradil (after L-type VDCC inhibition) in the present study. There are several potential explanations that could account for the observed differences between our present study and those suggesting functional involvement of R-type and T-type VDCCs after SAH, including type of blood vessel studied (pial artery vs. parenchymal arteriole), cellular environment, and proximity to the induced blood clot. It is also of interest that two previous studies have reported no difference in spontaneous tone development in parenchymal arterioles and brain stem-penetrating arterioles after experimental SAH (52, 60). The basis for the observed differences between those studies and our present work is not readily apparent. However, in support of our present findings, histological studies have found evidence of increased constriction of intracerebral arterioles from SAH model animals (22, 37, 50).

Although rat models of SAH have been used widely to study the impact of subarachnoid blood on the brain and cerebral vasculature (37, 46), caution should be used in extrapolating data obtained from rodents to SAH-induced pathologies observed in humans after cerebral aneurysm rupture. For example, pial artery vasospasm in SAH model rats is less severe and of shorter time course than reported in canine and primate SAH models or in many cases of SAH-induced vasospasm in humans (41, 42). It should also be noted that in vivo cerebral blood flow measurements were not performed in the present study and represent an important future direction to establish whether SAH-induced decreased cortical blood flow parallels the enhanced parenchymal arteriolar constriction that we observed in vitro.

To summarize, we provide evidence that myocytes of parenchymal arterioles obtained from SAH model rats exhibit enhanced membrane potential depolarization at physiological intravascular pressures leading to increased L-type VDCC activity, elevated [Ca²⁺]_i, and enhanced constriction. Furthermore, we demonstrate that inhibitors of L-type VDCCs (i.e., nimodipine) as well as agents that cause membrane potential hyperpolarization (i.e., NS-309) are effective in reducing [Ca²⁺]_i and dilating parenchymal vessels from SAH animals. We propose that therapeutic strategies targeting enhanced vasoconstriction of the cerebral microcirculation may be beneficial in reducing neurological deficits associated with SAH.

GRANTS

This work was supported by the Totman Medical Research Trust Fund, the Peter Martin Brain Aneurysm Endowment, National Institutes of Health Grants P20-RR-16435, R01-HL-078983, R01-HL-078983-0551, P01-HL-095488, R01-HL-44455, R01-HL-098243, and R01-HL-58231, and American Heart Association Founders Affiliate predoctoral fellowship award (0815736D) and an award from the American Heart Association Founders Affiliate and the Pulmonary Hypertension Association (10POST3690006).

DISCLOSURES

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