The expression and function of nicotinic ACh receptors (nAChRs) in rat coronary microvascular endothelial cells (CMECs) were examined using RT-PCR and whole cell patch-clamp recording methods. RT-PCR revealed expression of mRNA encoding for the subunits α2, α3, α4, α5, α7, β2, and β4 but not β3. Focal application of ACh evoked an inward current in isolated CMECs voltage clamped at negative membrane potentials. The current-voltage relationship of the ACh-induced current exhibited marked inward rectification and a reversal potential (Erev) close to 0 mV. The cholinergic agonists nicotine, epibatidine, and cytisine activated membrane currents similar to those evoked by ACh. The nicotine-induced current was abolished by the neuronal nAChR antagonist mecamylamine. The direction and magnitude of the shift in Erev of nicotine-induced current as a function of extracellular Na+ concentration indicate that the nAChR channel is cation selective and follows that predicted by the Goldman-Hodgkin-Katz equation assuming K+/Na+ permeability ratio of 1.11. In fura-2-loaded CMECs, application of ACh, but not of nicotine, elicited a transient increase in intracellular free Ca2+ concentration. Taken together, these results demonstrate that neuronal nAChR activation by cholinergic agonists evokes an inward current in CMECs carried primarily by Na+, which may contribute to the plasma nicotine-induced changes in microvascular permeability and reactivity induced by elevations in plasma nicotine.
- microvascular endothelium
- mRNA expression
- acetylcholine-evoked current
- nicotinic receptor subunits
- intracellular calcium
nicotine, a major component of cigarette smoke, has been shown to induce morphological and functional changes in vascular endothelium. Nicotine stimulates DNA synthesis and proliferation in endothelial cells, which may contribute to tumor angiogenesis and cardiovascular disease (8, 28, 31). Furthermore, nicotine increases vascular permeability (2) and glucose transport (4) across the endothelium of cerebral vessels, increases the production of prostacyclin (1) and nitric oxide (27), and impairs agonist-induced endothelium-dependent relaxation in resistance arterioles (16, 17) and in coronary arteries (19).
The presence of [3H]nicotine binding sites in mammalian cerebral vessels (10) and, more recently, the expression of neuronal nicotinic ACh receptor (nAChR) subunits in human aortic and umbilical vein endothelial cells have been reported (9, 15, 29). These observations suggest that the effects of nicotine on endothelium-dependent responses may be mediated by nAChRs on the endothelial cells lining the blood vessels. However, given that the endothelium displays considerable heterogeneity between macro- and microvascular endothelial cells due to the local environment (6, 26), it is important to characterize the nAChR subtypes in microvascular endothelial cells. The aim of the present study was to investigate the expression and characterize the biophysical and pharmacological properties of nAChRs in coronary microvascular endothelial cells. A preliminary report of some of these results has been presented in abstract form (20).
MATERIALS AND METHODS
Isolation and identification of CMECs. CMECs were isolated enzymatically from the left ventricle removed from adult rat hearts according to procedures described previously (21, 24). Adult Wistar rats (250–350 g) were killed by pentobarbitone overdose in accordance with the guidelines of the University of Queensland Animal Experimentation Ethics Committee. Cells reached confluence after 7–9 days and displayed an almost uniform “cobblestone” morphology. Primary cultures of CMECs were positively identified as described previously (21, 22).
Detection of nAChR subunit mRNA using RT-PCR. To eliminate nonspecific amplification products, nested PCR was performed on all the α-subunits and the β3-subunit. The external primer sequences were essentially those used by Lena et al. (14). Internal (nested) primer sequences were synthesized based on published rat sequence data and were located within the same exon. In addition, each primer pair contained a unique restriction site to allow confirmation of the identity of the amplification products. The internal primer pair sequences for each subunit with the corresponding restriction enzyme and predicted product sizes are presented in Table 1.
TriPure reagent (Roche Diagnostics) was used according to the manufacturer's protocol to simultaneously extract RNA and DNA from the samples. To remove potential DNA contamination from the RNA samples, DNase I (Roche Diagnostics) treatment was performed and further purified on an RNeasy spin column (Qiagen) according to the manufacturer's instructions.
First-strand cDNA synthesis was performed according to manufacturer's instructions using Superscript II Reverse Transcriptase (GIBCO-BRL Life Technologies). To exclude the possibility of genomic DNA contamination in the cDNA, a reverse transcriptase-minus negative control was used with each primer set. The first cycle of PCR was performed with cDNA or genomic DNA (positive control) as the template using Taq DNA polymerase (GIBCO-BRL Life Technologies) and 1.5 mM MgCl2, with 35 cycles at an annealing temperature of 60°C. All RT samples were tested with a β-actin primer pair to test for efficient amplification of the target sequence. With the exception of β2 and β4 PCR, each of the resultant PCR samples was used as the template for secondary (nested) PCR amplification using the internal primer pairs described, and 30 PCR cycles were performed as described above. The specificity of each primer pair was confirmed using the positive control genomic DNA extracted at the same time as the RNA. Sequence identity of the amplified products was also confirmed by restriction enzyme digestion with the appropriate restriction enzymes.
Electrophysiological recordings. Membrane voltage and current were recorded from isolated endothelial cells (3–5 days in culture) using the conventional whole cell patch-clamp recording configuration (7) and a L/M EPC-7 patch-clamp amplifier (List-Electronic; Darmstadt, Germany). Pipettes were pulled from thin-walled borosilicate glass using a Sutter Instruments P-87 pipette puller and, after being fire polished, had resistances of 3–5 MΩ when filled with high-K+ internal solution. Access resistances were ≤2 MΩ following series resistance compensation. The recording chamber had a volume of 0.5 ml and was superfused with a constant flow of 2 ml/min. No flow- or stretch-activated currents could be detected under these conditions. Membrane capacitance (Cm), measured by zeroing the capacitive transients evoked by 10-mV hyperpolarizing steps with the amplifier's built-in compensation section and reading out the corresponding Cm value, was 4.7 ± 0.9 pF (n = 20). Whole cell currents were sampled at 1 kHz, and current-voltage (I-V) curves were obtained by applying, every 1 s, 1.5-s voltage ramps ranging from –100 to +40 mV.
Experiments were carried out at room temperature (20–23°C). Pooled data are given as means ± SE.
Microfluorimetry. Measurements of intracellular calcium concentration ([Ca2+]i) in CMECs were carried out as previously described (21, 22). Briefly, CMECs were incubated for 30 min at room temperature in physiological saline solution (PSS) containing 5 μM fura-2 AM (1 mM fura-2 AM in DMSO stock solution), 0.02% Pluronic F-127, and 0.5% BSA. The coverslip was washed with PSS and fixed to the bottom of a petri dish with the use of silicon grease. The petri dish was then mounted on an upright epifluorescence microscope (Zeiss, Axiolab) equipped with a 100-W Hg lamp. A Zeiss ×40 Achroplan (water immersion objective, 0.75 numerical aperture) was used to visualize the cells. The exciting light was passed through an infrared filter, a neutral density filter and, alternately, a 340- or 380-nm band-pass excitation filter (Chroma Technology; Brattleboro, VT) mounted on a filterwheel equipped with a shutter (Lambda 10, Sutter Instruments; Novato, CA). The neutral density filter was coupled to the 380-nm filter to balance the intensity of the exciting light. Emitted light was passed through a 510-nm filter before being collected by a high-sensitivity camera (Extended-ISIS Camera, Photonic Science; Millham, UK), which was interfaced with a frame grabber (CX100, ImageNation; Beaverton, OR) to a personal computer. Custom software was used to drive the camera, the filterwheel, and the shutter and to measure and plot the fluorescence on-line from a number of rectangular regions of interest. [Ca2+]i was monitored by evaluating the ratio of fluorescence signals emitted at 510 nm when samples were excited at 340 and 380 nm, respectively (F340/F380). Experiments were carried out at room temperature (20–23°C), and ratio measurements were performed every 1.5 s.
Solutions and drugs. The standard extracellular solution was PSS containing (in mM) 150 NaCl, 6 KCl, 1 MgCl2, 1.5 CaCl2, 10 HEPES-KOH, and 10 glucose, titrated to pH 7.4 with NaOH. The relative ionic permeability of the nAChR channel was investigated by replacement of 75 mM NaCl with equimolar N-methyl-d-glucamine or HCl or by doubling the extracellular NaCl concentration. Cholinergic agonists were applied to isolated endothelial cells by pressure ejection (Picospritzer II, General Valve; Fairfield, NJ) from a micropipette (∼3 μm diameter) positioned 50–100 μm from the cell. nAChR activation was obtained by focal application of maximally effective concentrations of the cholinergic agonists ACh, nicotine, epibatidine, and cytisine, and a delay of at least 5 min between agonist applications was maintained. The ganglionic nicotinic receptor antagonist mecamylamine and the muscarinic receptor antagonist atropine were bath applied at the concentrations indicated. The composition of the intracellular solution was (in mM) 140 mM KCl, 5 NaCl, 2 MgCl2, 4 Na2ATP, 0.1 EGTA, and 10 HEPES-KOH, titrated to pH 7.2 with KOH. The osmolarity of the extracellular and intracellular solutions, as measured with an osmometer (Wescor 5500, Logan, UT), was 290–300 mmol/kg.
All chemical reagents used were of analytic grade. ACh chloride, ATP, atropine sulfate, cytisine, mecamylamine hydrochloride, and l-nicotine hydrogen tartrate were obtained from Sigma (St. Louis, MO), and d-epibatidine hydrochloride was from RBI (Natick, MA).
Detection of nAChR subunit mRNA expression using RT-PCR. RT-PCR analysis of CMECs revealed that these cells express mRNA for the subunits α2, α3, α4, α5, α7, β2, and β4 of the nAChR (Fig. 1A), whereas the β3-subunit was not detected. Control samples in which the reverse transcriptase had been omitted showed no PCR products, whereas positive controls using genomic DNA confirmed that all the primer sets detected the nAChR subunit sequence (Fig. 1A). The identity of the PCR products was confirmed by restriction analysis (Fig. 1B).
Whole cell recording of nAChR-activated current. The mean resting membrane potential of isolated CMECs was –33.3 ± 1.8 mV (n = 24), a value similar to that previously reported (–31.4 ± 2.9 mV) (22). Brief application (100 ms) of 300 μM ACh plus 1 μM atropine evoked an inward current at negative holding potentials, and a depolarizing response (26.3 ± 2.3 mV, n = 3) under current clamp, in 53% (18 of 34) of the cells studied (Fig. 2A). ACh-evoked currents obtained at –40, –20, 0, and +20 mV are shown in Fig. 2B, and the I-V relationship obtained in response to a voltage ramp from –100 to +40 mV is shown in Fig. 2C. The I-V relationship for the ACh-evoked current exhibited marked inward rectification whereby the current density was –110 ± 19 pA/pF at –20 mV, +30 ± 3 pA/pF at +20 mV, and reversed at +0.9 ± 1.4 mV (n = 6). The half-time of decay of the ACh-evoked current was 6.5 ± 0.9 s (n = 6) and independent of the membrane potential.
Maximally effective doses of the cholinergic agonists nicotine (300 μM), epibatidine (10 nM), and cytisine (100 μM) evoked inward currents in 71%, 38%, and 62% of cells, respectively (Fig. 3, A, C, and D). Although the agonist concentrations used in these experiments were maximally effective concentrations, lower doses of nicotine (10–100 μM) also evoked excitatory responses and inward currents in isolated CMECs (not shown). The neuronal nAChR antagonist mecamylamine (1 μM) reversibly abolished the nicotine-induced current in all cells examined (n = 5; Fig. 3B).
The ionic basis of the nicotine-evoked currents was examined using voltage ramps and measurement of the reversal (zero current) potential (Erev) in the presence of different extracellular Na+ concentrations ([Na+]o). The Erev determined from I-V relations for the nicotine-evoked current obtained in normal [Na+]o was +2.3 ± 3.3 mV (n = 3) and shifted (ΔErev) by –20.8 ± 0.2 mV in 0.5× [Na+]o and by +23.9 ± 2.9 mV in 2× [Na+]o (Fig. 4A). The shift in Erev is similar to that predicted by the Goldman-Hodgkin-Katz (GHK) voltage equation assuming that only Na+ and K+ are permeant. Erev is plotted as a function of [Na+]o in Fig. 4B and the line fit to the data by the GHK voltage equation assuming PK/PNa = 1.11.
Measurement of [Ca2+]i in fura-2-loaded CMECs. Fura-2 fluorescence ratio imaging was used to determine changes in [Ca2+]i in response to ACh receptor activation in CMECs. Brief application of nicotine (300 μM) had no effect on resting [Ca2+]i (n = 55). In contrast, both ACh (100 μM) and ATP (20 μM) evoked characteristic increases in [Ca2+]i (n = 23) in the same cells (Fig. 5, A–C). The failure of nicotine, but not ACh, to induce changes in [Ca2+]i indicates that ACh is most likely acting via muscarinic receptors. Furthermore, ACh-induced increases in [Ca2+]i were inhibited by bath application of 1 μM atropine (not shown).
The present study is the first report of the expression of mRNA encoding for neuronal nAChR subunits in CMECs and confirms the hypothesis that the response to ACh in vascular endothelium may also be mediated by nAChRs (3). The proliferative effect of nicotine on calf pulmonary artery endothelial cells has recently been suggested to depend on the activation of nicotinic receptors (28). The presence of [3H]nicotine binding sites in mammalian cerebral vessels (10) and mRNA expression of neuronal nAChR α- and β-subunits in human and bovine aortic endothelial cells (15, 29) have also been reported. The RT-PCR data shown here now suggest that the same nAChR subunits are expressed in both macro- and microvascular endothelial cells. Indeed, mRNA encoding for α3-, α5-, α7-, β2-, and β4-subunits has been detected in both CMECs and aortic endothelium (15, 29), whereas only the α4-subunit is not expressed in macrovessels (29). Immunohistochemical staining for nAChR isoforms using nAChR subunit-specific antibodies demonstrated that α3, α4, α7, β2, and β4 were expressed in human umbilical vein endothelial cells, whereas α2, α5, and β3 were undetectable (9).
The electrophysiological response upon ACh application to isolated CMECs demonstrates that the neuronal nAChR subunits are expressed as functional ionic channels, which upon activation depolarize CMECs. The I-V relationship of the ACh-induced current displays marked inward rectification and an Erev close to 0 mV, similar to that exhibited by nAChRs in the central and peripheral nervous systems (18). The cholinergic agonists nicotine, epibatidine, and cytisine activate membrane currents similar to those evoked by ACh, and the nicotine-induced current is abolished by the ganglionic nAChR antagonist mecamylamine (5). The direction and magnitude of the shift in Erev of the nicotine-induced current as a function of [Na+]o indicate that the nAChR channel is cation selective and follows that predicted by the GHK voltage equation assuming PK/PNa = 1.11. These data indicate that Na+ is the major permeant cation through the open nAChR channel, as also observed in rat parasympathetic ganglion neurons (5).
Nicotine, in contrast to ACh or ATP, did not cause any change in the resting [Ca2+]i in CMECs. The lack of a significant rise in [Ca2+]i upon nAChR stimulation by nicotine could be explained by the relatively low resting membrane potential of CMECs (present study; 22). Indeed, neuronal nAChRs are possible sources of Ca2+ entry at hyperpolarized membrane potentials (23, 30). In CMECs, the weaker driving force for Ca2+ influx could therefore only induce a slight nAChR-dependent [Ca2+]i increase, which may occur in the subplasmalemmal region (see 30). A localized [Ca2+]i increase in the cell periphery is beyond the resolution of our imaging system.
The physiological ligand of neuronal nAChRs in vascular endothelium might be ACh present in the blood or ACh synthesized by endothelial cells themselves (3). In the blood of smokers, however, nicotine may reach relatively high concentrations (up to 70 ng/ml). Acute exposure and hyperstimulation of nAChRs by nicotine could upregulate nAChR-dependent functions, thus inducing the changes in vascular permeability and reactivity observed after elevations in plasma nicotine, such as tumor angiogenesis and cardiovascular disease (8, 16, 17, 19, 28, 31). The impairment of endothelial-dependent arteriolar dilatation, for instance, could be due to the Na+ influx into the endothelium after nAChRs activation. Indeed, Na+ entry can activate or contribute to the production of oxygen-derived free radicals (12), which may mediate the damage of endothelial-dependent vasoreactivity induced by nicotine (16, 17). Furthermore, nicotine-induced contraction of the rat coronary artery in the presence of N-nitro-l-arginine methyl ester has been shown to be endothelium dependent and involve reactive oxygen species and cyclooxygenase 1 metabolites of arachidonic acid (13). The mitogenic effect of nicotine could also be stimulated by the Na+ flux mediated by nAChRs, as Na+ entry has been suggested to be a necessary event to elicit DNA synthesis in several cell types (11, 25).
In conclusion, our findings further support the hypothesis that, in addition to muscarinic receptor activation, exposure of CMECs to ACh also activates neuronal nAChRs. This suggests an important role for nAChRs in mediating changes in microvascular function induced by elevated levels of nicotine in the circulation.
We thank Mark Stafford for technical support.
Present address of F. Moccia: Laboratory of Cell Biology, Stazione Zoologica “A. Dohrn,” 80121, Naples, Italy.
This research was supported by a National Heart Foundation of Australia grant (to D. J. Adams).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 by the American Physiological Society