Dual Effects of Intravenous Anesthetics on the Function of Norepinephrine Transporters

Eagle’s minimum essential medium (MEM) was obtained from Nissui Pharmaceuticals (Tokyo, Japan). Fetal calf serum, l-norepinephrine, pargyline hydrochloride, and l-ascorbic acid were obtained from Nacalai Tesque (Kyoto, Japan). Collagenase was obtained from Nitta Zerachin (Osaka, Japan). 2,6-Diisopropylphenol (propofol) was obtained from Tokyo Kasei (Tokyo, Japan). Diazepam hydrochroride was a gift from Takeda Chemical Industries (Osaka, Japan). Thiamylal sodium was obtained from Sankyo Co. (Osaka, Japan). Desipramine hydrochloride was obtained from Sigma (St. Louis, MO), and nisoxetine hydrochloride was obtained from Research Biochemicals International (Natick, MA). l-[7,8-3[³H]Noradrenaline (34.0 Ci/mmol) was obtained from Amersham International (Buckinghamshire, United Kingdom), and [benzene ring, 10,11-[³H]-desmethylimipramine (desipramine) hydrochloride (73.0 Ci/mmol) was obtained from New England Nuclear (Boston, MA). 2,6-Diisopropylphenol was diluted with dimethyl sulfoxide for experiments. Dimethyl sulfoxide at the concentrations used for experiments had no effect on [³H]norepinephrine uptake and [³H]desipramine binding. Thiamylal sodium and diazepam hydrochroride were dissolved with distilled water.

Adrenal medullary cells were isolated from bovine adrenal medulla as described previously. 18The cells were plated at 4 × 10⁶cells per dish (Falcon, 35 mm) in Eagle’s MEM containing 10% fetal calf serum, 60 μg/ml aminobenzylpenicillin, 100 μg/ml streptomycin, 0.3 μg/ml amphotericin B, and 3.0 μm cytarabine. 15The cells were cultured in 5% CO²–95% air at 37°C and used for experiments between 2 and 4 days of culture.

Cultured cells (4 × 10⁶per dish) were incubated at 37°C for 15 min in oxygenated Krebs-Ringer phosphate buffer containing 100 μm pargyline, 100 μm ascorbic acid, and 500 nm [³H]norepinephrine in the presence or absence of propofol (0.1–300 μm), thiamylal (3–1,000 μm), or diazepam (3–1,000 μm). Pargyline is a monoamine oxidase inhibitor that prevents the enzymatic decomposition of norepinephrine in the cells. Ascorbic acid is an antioxidant of norepinephrine. Krebs-Ringer phosphate buffer was composed of 154 mm NaCl, 5.6 mm KCl, 1.1 mm MgSO⁴, 2.2 mm CaCl², 0.85 mm NaH²PO⁴, 2.15 mm Na²HPO⁴, and 10 mm glucose, adjusted to pH 7.4. For the kinetic analysis of [³H]norepinephrine uptake, the cells were incubated with increasing concentrations (1–30 μm) of [³H]norepinephrine in the presence or absence of 100 μm propofol, thiamylal, or diazepam. After incubation, the cells were rapidly washed four times with 1 ml ice-cold buffer and solubilized in 1 ml Triton X-100 (10%; Nacalai Tesque, Kyoto, Japan). The radioactivity in the solubilized cells was counted by a liquid scintillation counter (LSC-3500E; Aloka, Tokyo, Japan). Nonspecific uptake was determined in the presence of 10 μm desipramine, and specific uptake was obtained by subtracting the nonspecific uptake from the total uptake. The desipramine-sensitive uptake was 92 ± 3% (n = 12) of the total uptake.

Plasma membranes isolated from bovine adrenal medulla were prepared as described previously. 19The binding of [³H]desipramine was determined by incubation of membranes (10 μg protein) suspended in buffer B (composition: 135 mm NaCl, 10 mm Tris-HCl (pH 7.4), 5 mm KCl, 1 mm MgSO⁴) for 30 min at 25°C. The incubation medium (final volume, 250 μl) contained [³H]desipramine (2–24 nm), and in some experiments also contained propofol (0.1–300 μm). After incubation, binding was terminated by the addition of 2 ml ice-cold buffer B and rapid filtration of the membrane suspension under vacuum through Whatman GF/C glass fiber filters (Whatman, Maidstone, United Kingdom). The filters were rapidly washed twice with 2 ml ice-cold buffer B and were placed in counting vials containing a scintillation cocktail. The radioactivity was counted in Aloka LSC-3500E. Nonspecific binding was determined in the presence of 10 μm nisoxetine, a selective NET inhibitor, and specific binding was obtained by subtracting the nonspecific binding from the total binding.

Cells were preincubated with or without propofol (1–30 μm) for 6–24 h. After preincubation, the cells were washed with 2 ml Eagle’s MEM and stood for another 3 h in a culture chamber to completely wash out propofol. The cells were then incubated with [³H]norepinephrine (500 nm) for 15 min, and [³H]norepinephrine uptake by the cells was evaluated as previously described.

After treatment of cells with 10 μm propofol for 12 h and subsequent washing with 1 ml MEM and standing for 1 h, cells were collected and crashed by a homogenizer (Ultra-Turrax T8; IKA Labortechnik, Staufen, Germany) for 60 s in a lysis buffer (10 mm Tris-HCl [pH 7.4], 2 mm MgCl², 1 mm dithiothreitol, 1 mm phenylmethylsulphonyl fluoride) and centrifuged at 1,000 g  for 10 min at 4°C. The supernatant was centrifuged at 60,000 g  for 30 min. The final pellet containing plasma membranes was suspended in the binding buffer (50 mm Tris-HCl [pH 7.4], 300 mm NaCl, 5 mm KCl). 16The protein of isolated plasma membranes was quantified by the method of Lowry et al.  20The binding of [³H]desipramine was determined by incubation of membranes (100 μg protein) suspended in buffer B for 30 min at 25°C as described previously.

Poly(A)⁺RNA was isolated from control or propofol-treated cells by guanidine hydrochloride, ethanol fractionation, chloroform–isobutanol extraction, and oligo(dT) cellulose column separation as previously described. 21The first-strand cDNA was synthesized from Poly(A)⁺RNA using a first-strand cDNA synthesis kit (Amersham Pharmacia Biotech, Piscaway, NJ). The obtained cDNA was amplified by polymerase chain reaction. Each sample was assayed for NET and β-actin mRNAs using their specific primers. The sense and antisense primers for NET were 5′-CTGACCAGCACCATCAACTGT-3′ and 5′-GTGAAGAGTTTCCGGTGTCGC-3′, and those for β-actin were 5′-TGGAGAAGAGCTATGAGCTGCCTG-3′ and 5′-GTGCCACCAGACAGCACTGTGTTG-3′, respectively. Polymerase chain reaction using primers for NET or β-actin and a Takara Ex Taq kit (Takara, Otsu, Japan) was conducted with an automatic thermal controller (PC-800; Astec, Fukuoka, Japan). The thermocycling conditions were as follows: 28 cycles of 94°C for 1 min, 62°C for 30 s, 72°C for 20 s for NET; and 20 cycles of 94°C for 15 s, 66°C for 15 s, 72°C for 30 s for β-actin. The resultant polymerase chain reaction products were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and visualized by GelStar Nucleic Acid Gel Stain (Takara, Otsu, Japan). Fluorescence intensity of the bands was quantified with a Fuluoroimage Analyzer (BAS 3000; Fujifilm, Tokyo, Japan).

All values are expressed as mean ± SD. Statistical analysis was conducted by one-way analysis of variance followed by a Dunnett t  test for multiple comparisons in figure 1and by paired Student t  test for comparisons of the values of Michaelis constant (K  ^m), maximal velocity (V^max), dissociation constant (K  ^d), and maximal binding (B^max) in figures 2 and 3and table 1. In figures 4 and 5, analysis was performed using one-way analysis of variance followed by a Dunnett t  test, and in figure 6by paired Student t  test for comparisons of the values of Kd and Bmax. In figure 7, one-way analysis of variance followed by a Dunnett t  test was used. Differences were considered as statistically significant at P  less than 0.05.

Uptake of [³H]norepinephrine was linearly related to cell density (0.2–4 × 10⁶cells) and incubation time (10–40 min). 17Thus, the assay condition (4 × 10⁶cells, 15 min) selected for the subsequent experiments was well within the linear range of [³H]norepinephrine uptake. All anesthetics, including propofol, thiamylal, and diazepam, inhibited [³H]norepinephrine uptake in a concentration-dependent manner (fig. 1). Control values (absent of anesthetics) were 1.12 ± 0.04 pmol · 4 × 10⁶cells⁻¹· min⁻¹. From the results of figure 1, each the half-maximal inhibition concentration (IC⁵⁰) for [³H]norepinephrine uptake by intravenous anesthetics was calculated. Because the inhibition curve by propofol was biphasic, the data of propofol were examined using a modified Scatchard analysis 22(data not shown), and it was resolved in two components (table 1). Incubation of the cells with increasing concentrations of [³H]norepinephrine (1–30 μm) showed that [³H]norepinephrine uptake was a saturable process (fig. 2A). From Eadie-Hofstee analysis (fig. 2B), all anesthetics produced a significant reduction in the V^maxwithout altering the K^mvalues (table 1). The kinetic analysis of additional experiments in clinical relevant concentration (5 μm of propofol) showed that low concentration of propofol significantly (P < 0.05) changed V^max(control, 10.5 ± 0.4 pmol · 4 × 10⁶cells⁻¹· min⁻¹; propofol, 8.7 ± 0.3 pmol · 4 × 10⁶cells⁻¹· min⁻¹) but did not change K^m(control, 4.9 ± 0.4 μm; propofol, 5.1 ± 0.3 μm; data not shown), suggesting noncompetitive inhibition similar to that observed with a high concentration of propofol.

A specific binding of [³H]desipramine was found to be saturable (fig. 3A). Scatchard analysis in control [³H]desipramine binding showed a single population of binding site with an apparent K  ^dof 6.05 ± 0.23 nm and B^maxof 3.11 ± 0.16 pmol/mg protein (fig. 3B). Propofol (60 μm) inhibited [³H]desipramine binding by increasing the K^d  value to 29.0 ± 4.0 μm (P < 0.05) without any change in the B^maxvalue (2.84 ± 0.26 pmol/mg protein;fig. 3B). We also performed the experiment with a clinically relevant concentration. Low concentration (5 μm) of propofol changed K^d  without altering B^max(data not shown), suggesting competitive inhibition as observed with a high concentration of propofol. As shown in figure 4, propofol inhibited the binding of [³H]desipramine in a concentration-dependent manner (1–300 μm). The inhibition curve appeared to be biphasic. When the data were examined by a modified Scatchard analysis 22(data not shown), it showed two components with IC⁵⁰values of 2.8 ± 0.3 μm and 50 ± 6 μm. Furthermore, inhibitory constant (K  ⁱ) values of 0.9 ± 0.1 μm and 16 ± 2 μm for inhibition of [³H]desipramine binding by propofol were calculated, 23respectively.

Long-term treatment of cells with propofol caused time- (6–24 h) and concentration (1–10 μm)-dependent increases in [³H]norepinephrine uptake (figs. 5A and 5B). In the assay of [³H]desipramine binding, specific binding of [³H]desipramine to the plasma membranes prepared from control and propofol-treated cells was saturable with increasing concentrations of [³H]desipramine (2–24 nm;fig. 6A). Scatchard analysis showed that 10 μm propofol produced a significant (P < 0.05) increase in the B^max(control, 230 ± 18 fmol/mg protein; propofol, 270 ± 13 fmol/mg protein) without any change in the K^d  (control, 6.6 ± 0.3 nm; propofol, 6.3 ± 0.4 nm;fig. 6B). In the assay of mRNA expression, polymerase chain reaction with NET primers and β-actin primers yielded single bands corresponding to approximately 0.3-kb and 0.2-kb fragments, respectively (fig. 7). The band for NET mRNA was sequenced and found to be identical to the reported bovine NET 13(data not shown). Furthermore, propofol (30 μm) increased the NET mRNA level by 2.2- and 2.0-fold at 12 and 24 h, respectively (fig. 7A). The increase in NET mRNA level produced by propofol was concentration-dependent (10–30 μm;fig. 7B). Each result of NET mRNA was normalized with β-actin mRNA.

We demonstrated that all the intravenous anesthetics used in this study inhibited [³H]norepinephrine uptake by cultured adrenal medullary cells. The rank order of the potency to suppress [³H]norepinephrine uptake was propofol, diazepam, and thiamylal. The peak plasma concentration of propofol was reported to be approximately 50 μm after bolus administration. 24However, the steady state free plasma concentration of propofol may not exceed 2 μm because 98% binds to plasma proteins. In addition, taking protein binding into account, the clinically relevant concentrations of diazepam and thiopental are approximately 1 μm 25and 25 μm, 3respectively, and the plasma concentration of the latter is similar to that of thiamylal. 26At these clinically relevant concentrations, propofol and probably thiamylal seem to inhibit NET function to a small but significant degree. As much as 80% of norepinephrine released from presynaptic terminals is believed to be physiologically reuptaken by the neurons, 27terminating neurotransmission. Therefore, even a slight inhibition of NET activity by clinical concentrations of anesthetics may enhance neurotransmission.

To address the site of action of propofol on NET, we examined the effects of propofol on the kinetic parameters of [³H]norepinephrine uptake and on [³H]desipramine binding. All of the intravenous anesthetics significantly lowered the V^maxvalue of [³H]norepinephrine uptake without changing the K^m  value, indicating noncompetitive inhibition. Propofol inhibited the specific binding of [³H]desipramine with a potency similar to that of [³H]norepinephrine uptake. Scatchard analysis revealed that propofol significantly increased the K^d  value without affecting the value of B^max, indicating competitive inhibition. The present results were also confirmed by a clinical concentration of propofol (see Results). Propofol seems to have biphasic effects on [³H]norepinephrine uptake and [³H]desipramine binding (figs. 1 and 4), suggesting that propofol affects the NET at two sites of action, such as high- and a low-affinity sites. Further study is required to confirm this possibility.

Over the past 10 yr, there has been controversy over whether the site of substrate recognition is identical to that for tricyclic antidepressant binding on monoamine transporters. 28,29From recent molecular cloning and chimeric dopamine–norepinephrine transporter studies, the current prevailing hypothesis is that there are distinct regions within NET molecules that determine substrate recognition and translocation and antagonist affinity, but these regions may overlap each other. 9,30,31In the present study, the noncompetitive kinetics of [³H]norepinephrine uptake suggest that propofol interacts with NET at a different site from the norepinephrine recognition site. The competitive inhibition of [³H]desipramine binding by propofol suggests that propofol acts directly on the desipramine binding site. Alternatively, propofol and desipramine may act at different sites on the NET that are allosterically coupled. Thus, a simple interpretation of these results is that propofol interacts with NET, which, in turn, may allosterically lead to a conformational change in the transporter that inhibits transporter function. We previously proposed that ketamine inhibits the transport of norepinephrine by interacting with a site that partly overlaps the desipramine binding site on NET. 17Propofol inhibited [³H]norepinephrine uptake in a manner very similar to that of ketamine. Taken together, our present results provide further evidence to support the hypothesis that in NET molecules there is a common region or areas in close proximity that are susceptible to some intravenous anesthetics. Further studies using various NET mutants produced by molecular techniques are required to determine the precise site of intravenous anesthetic action on NET.

Our findings explain some of the pharmacologic effects of intravenous anesthetics. For instance, intravenous anesthetics may enhance the action of exogenous or endogenous catecholamines. Indeed, propofol and thiamylal are reported to enhance epinephrine-induced arrhythmias in dogs. 32–34Furthermore, evidence has emerged that the descending inhibitory system consists of noradrenergic neurons. 35Tricyclic antidepressants, including desipramine, that selectively antagonize NET are used to treat the chronic pain that accompanies postherpetic neuralgia, diabetic neuropathy, cancer, and complex regional pain syndrome. 36–38Their antinociceptive effect is considered to arise partly from enhancing noradrenergic neurotransmission by inhibiting NET in the descending inhibitory system in the brain and spinal cord. 39Our results raise the possibility that the inhibitory effects of intravenous anesthetics on NET activity during anesthesia have an antinociceptive action.

Propofol is sometimes administered for prolonged periods during surgery. We found that treatment of cells with clinically relevant concentrations of propofol increased the functional activity of NET and [³H]desipramine binding sites. Presently, clinical significance of the phenomenon is not clear. It may be regarded as a compensatory action of propofol that normalizes the norepinephrine level at noradrenergic synapses during anesthesia.

In conclusion, intravenous anesthetics have a dual effect on NET function: short-term treatment produces inhibition, whereas long-term treatment causes up-regulation. Our findings indicate that NET is one of the target proteins of intravenous anesthetics and help to unveil the pharmacologic basis of this interaction for the better understanding of the various actions of intravenous anesthetics.