Catalytic properties of the immobilized Aspergillus tamarii xylanase

Xylanase from Aspergillus tamarii was covalently immobilized on Duolite A147 pretreated with the bifunctional agent glutaraldehyde. The bound enzyme retained 54.2% of the original specific activity exhibited by the free enzyme (120 U/mg protein). Compared to the free enzyme, the immobilized enzyme exhibited lower optimum pH, higher optimum reaction temperature, lower energy of activation, higher K_m (Michaelis constant), lower V_max (maximal reaction rate). The half-life for the free enzyme was 186.0, 93.0, and 50.0 min for 40, 50, and 60°C, respectively, whereas the immobilized form at the same temperatures had half-life of 320, 136, and 65 min. The deactivation rate constant at 60°C for the immobilized enzyme is about 6.0 × 10⁻³, which is lower than that of the free enzyme (7.77 × 10⁻³ min). The energy of thermal deactivation was 15.22 and 20.72 kcal/mol, respectively for the free and immobilized enzyme, confirming stabilization by immobilization. An external mass transfer resistance was identified with the immobilization carrier (Duolite A147). The effect of some metal ions on the activity of the free and immobilized xylanase has been investigated. The immobilized enzyme retained about 73.0% of the initial catalytic activity even after being used 8 cycles.     

Entrapment of lipase into K-carrageenan beads and its use in hydrolysis of olive oil in biphasic system

The porcine pancrease lipase was immobilized by entrapment in the beads of K-carrageenan and cured by treatment with polyethyleneimine (PEI) in the phosphate buffer. The retention of hydrolytic activity of lipase and compressive strength of the beads were examined. The activity of free and immobilized lipase was assessed by using olive oil as the substrate. The immobilized enzyme exhibited a little shift towards acidic pH for its optimal activity and retained 50% of its activity after 5 cycles. When the enzyme concentration was kept constant and substrate concentration was varied the K_m and V_max were observed to be 0.18 × 10⁻² and 0.10, and 0.10 × 10⁻² and 0.09 respectively, for free and for entrapped enzymes. When the substrate concentration was kept constant and enzyme concentration was varied, the values of K_m and V_max were observed to be 0.19 × 10⁻⁷ and 0.41, and 0.18 × 10⁻⁷ and 0.41 for free and entrapped enzymes. Though this indicates that there is no conformational change during immobilization, it also shows that the reaction velocity depends on the concentration. Immobilized enzyme showed improved thermal and storage stability. Hydrolysis of olive oil in organic–aqueous two-phase system using fixed bed reactor was carried out and conditions were optimized. The enzyme in reactor retained 30% of its initial activity after 480 min (12 cycles).     

Immobilization of catalase into chemically crosslinked chitosan beads

Bovine liver catalase was immobilized into chitosan beads prepared in crosslinking solution. Various characteristics of immobilized catalase such as the pH–activity curve, the temperature–activity curve, thermal stability, operational stability, and storage stability were evaluated. Among them the pH optimum and temperature optimum of free and immobilized catalase were found to be pH 7.0 and 35 °C. The K_m value of immobilized catalase (77.5 mM) was higher than that of free enzyme (35 mM). Immobilization decreased in V_max value from 32,000 to 122 μmol (min mg protein)⁻¹. It was observed that operational, thermal and storage stabilities of the enzyme were increased with immobilization.     

Glutathione-proline activation of feeding behavior in the zoanthid Palythoa psammophilia Walsh & Bowers

Palythoa psammophilia Walsh & Bowers has a well coordinated, stereotyped feeding response, the culminating step of which is ingestion; this may be elicited by the synergistic effect of the tripeptide glutathione and the -imino acid, proline. Either activator acting separately causes responses only at high concentrations (above 10⁻⁵ M for glutathione; above 10⁻⁴ M for proline) in a reduced number of animals and at a low rate (5.00 ± 1.73 min in 5 × 10⁻³ M solutions of glutathione; 11.10±3.74 min in 5 × 10⁻³ M solutions of proline). Highest percentages of response were obtained in combinations where glutathione was at a concentration of 5 × 10⁻³ M and proline at 5 × 10⁻⁴ M or in combinations of glutathione at concentrations 5 × 10⁻⁶ M and proline at 5 × 10⁻⁵ M. The speed of ingestion is considerably enhanced when these activators are combined (1.17±1.18 min).     

Enzyme-entrapping membranes for biosensors obtained by radiation-induced polymerization

A new method of physically immobilizing enzymes in poly(2-hydroxyethyl methacrylate) membranes was developed in order to obtain suitable biosensors. It was possible to prepare an enzyme sensor based on an oxygen Clark electrode and on glucose oxidase immobilized by low-temperature gamma radiation-induced polymerization. Temperature and pH effects on the activity of immobilized enzyme are described and the response characteristics of the resulting biosensor are summarized. The determination of glucose in standard solutions was carried out and a linear calibration curve, with an R² value of 0·9993, from the detection limit 5 × 10⁻⁵ to 1·2 × 10⁻³ was obtained. The biosensor was employed to analysis of control sera and the results were compared to those obtained by enzymatic-spectrophotometric detection.     

Hydrodynamic properties of the fractions of mannan formed by Rhodotorula rubra yeast

Aqueous solutions of fractions of an extracellular linear mannan formed by Rhodotorula rubra yeast have been investigated by hydrodynamic methods (high-speed sedimentation, translation isothermic diffusion and viscometry). The molecular weight was determined according to Svedberg ( ) and the polydispersity parameters of the initial sample were also determined (M_w/M_n = 1·20 and M_z/M_w = 1·21). Relationships between the molecular weight (M) and s_o, D_o and [η] in the range were: [η] = 2·33 × 10⁻² M^0.75, D_o = 1·65 × 10⁻⁴ M^0·58, s_o = 2·24 × 10⁻¹⁵ M^0·43. The equilibrium rigidity and hydrodynamic diameter of chains representing mannan molecules were evaluated.     

Production of cyclodextrin by poly-lysine fused Bacillus macerans cyclodextrin glycosyltransferase immobilized on cation exchanger

Bacillus macerans cyclodextrin glycosyltransferase (CGTase) fused with 10 lysine residues at its C-terminus (CGTK10ase) was immobilized onto a cation exchanger by ionic interaction and used to produce -cyclodextrin (CD) from soluble starch. Poly-lysine fused immobilization increased the V_m of the immobilized CGTase by 40% without a change in K_m. The activation energies of thermal deactivation (E_a) were 41.4, 28.1, and 25.9 kcal mol⁻¹, respectively, for soluble wild-type (WT) CGTase, soluble CGTK10ase, and immobilized CGTK10ase, suggesting destabilization of CGTase by poly-lysine fusion and immobilization onto a cation exchanger. Maximum -CD productivity of 539.4 g l⁻¹ h⁻¹ was obtained with 2% soluble starch solution which was constantly fed at a flow rate of 4.0 ml min⁻¹ (D = 240 h⁻¹) in a continuous operation mode of a packed-bed reactor. The operational half-life of the packed-bed enzyme reactor was estimated 12 days at 25 °C and pH 6.0.     

Immobilization of soybean seed coat peroxidase on polyaniline: Synthesis optimization and catalytic properties

Soybean seed coat peroxidase (SBP) was immobilized on various polyaniline-based polymers (PANI), activated with glutaraldehyde. The most reduced polymer (PANIG₂) showed the highest immobilization capacity (8.2 mg SBP g^-1 PANIG₂). The optimum pH for immobilization was 6.0 and the maximum retention was achieved after a 6-h reaction period. The efficiency of enzyme activity retention was 82%. When stored at 4°C, the immobilized enzyme retained 80% of its activity for 15 weeks as evidenced by tests performed at 2-week intervals. The immobilized SBP showed the same pH-activity profile as that of the free SBP for pyrogallol oxidation but the optimum temperature (55°C) was 10°C below that of the free enzyme. Kinetic analysis show that the K_m was conserved while the specific V_max dropped from 14.6 to 11.4 µmol min^-1 µg^-1, in agreement with the immobilization efficiency. Substrate specificity was practically the same for both enzymes. Immobilized SBP showed a greatly improved tolerance to different organic solvents; while free SBP lost around 90% of its activity at a 50% organic solvent concentration, immobilized SBP underwent only 30% inactivation at a concentration of 70% acetonitrile. Taking into account that immobilized HRP loses more than 40% of its activity at a 20% organic solvent concentration, immobilized SBP performed much better than its widely used counterpart HRP.     

Immobilization of invertase onto poly(3-methylthienyl methacrylate)/poly(3-thiopheneacetic acid) matrix

A novel immobilization matrix, poly(3-methylthienyl methacrylate)–poly(3-thiopheneacetic acid) (PMTM–PTAA), was synthesized and used for the covalent immobilization of Saccharomyces cerevisiae invertase to produce invert sugar. The immobilization resulted in 87% immobilization efficiency. Optimum conditions for activity were not affected by immobilization and the optimum pH and temperature for both free and immobilized enzyme were found to be 4.5 and 55 °C, respectively. However, immobilized invertase was more stable at high pH and temperatures. The kinetic parameters for free and immobilized invertase were also determined using the Lineweaver–Burk plot. The K_m values were 35 and 38 mM for free and immobilized enzyme, respectively. The V_max values were 29 and 24 mg glucose/mg enzyme min for free and immobilized enzyme, respectively. Immobilized enzyme could be used for the production of glucose and fructose from sucrose since it retained almost all the initial activity for a month in storage and retained the whole activity in repeated 50 batch reactions.     

Biocatalysis of tyrosinase using catechin as substrate in selected organic solvent media

The enzymatic activity of mushroom tyrosinase was investigated using catechin as substrate in selected organic solvent media. The results showed that optimal tyrosinase activity was obtained at pH 6.2, 6.6, 6.0 and 6.2 in the organic solvent media of heptane, toluene, dichloromethane, and dichloroethane, respectively, and at a temperature between 25°C and 27.5°C. In addition, the kinetic studies showed that the K_m values were 5.38, 1.03, 2.52 and 4.03 mM, for the tyrosinase-catechin biocatalysis in the reaction media of heptane, toluene, dichloromethane, and dichloroethane, respectively, while the corresponding V_max values were 1.22×10⁻³, 0.33×10⁻³, 1.47×10⁻³ and 1.20×10⁻³ δA per μg protein per second, respectively. The use of acetone as co-solvent for the tyrosinase-catechin biocatalysis showed that acetone concentrations ranging from 5% to 30% (v/v) in the heptane reaction medium produced a decrease of 4.3% to 96.7% in tyrosinase activity. The results also indicated that the presence of 12.5% acetone in the reaction medium of dichloromethane, and 22.0% in those of toluene and dichloroethane produced a maximal increase of 42.6%, 92.1% and 71.8%, respectively, in tyrosinase activity. However, the overall findings indicated that additional increases in acetone concentration resulted in an inhibition of tyrosinase activity.     

Preparation and characterization of immobilized phospholipase A2 on chitosan beads for lowering serum cholesterol concentration

Phospholipase A₂ (PLA₂) from cobra venom, which can hydrolyze the S_N2 ester bond of 1,2-diacylphosphatides, was immobilized by covalent binding to porous chitosan beads. Immobilization has to be carried out by using the carboxylic groups instead of the amine groups of the enzyme to get reasonable activity retention (higher than 50%). The effects of amount of activating reagent EDC and enzyme loading during the immobilization step were investigated. Since EDC could modify important Asp groups in the enzyme, the EDC/enzyme weight ratio should be less than 10. Although the activity retention of immobilized enzyme increased with enzyme/bead weight ratio, this ratio should be kept to a minimum at 1×10⁻³ to optimize coupling yield of enzyme activity and reduce internal diffusion resistance. The kinetic properties and stability of the immobilized enzyme were determined. The immobilized PLA₂ was packed into a column to hydrolyze phospholipid in a circulating packed-bed reactor. The flow rate of the substrate solution should be set at 37.5 cm/min (superficial velocity) to eliminate external diffusion resistance, under which condition the column reactor could be reused up to 10 times with less than 20% loss of activity. Since enzymatic hydrolysis of phospholipid on low density lipoprotein (LDL) particle surface with PLA₂ could result in faster plasma clearance of the modified LDL particles, an in vitro bioreactor containing immobilized PLA₂ should be able to lower serum cholesterol concentration. A significant decrease in total serum cholesterol concentration in hypercholesterolemic rabbits was observed after 90-min treatment.     

The development of an amperometric microbial biosensor using Acetobacter pasteurianus for lactic acid

A microbial biosensor, using Acetobacter pasteurianus cells and an oxygen electrode, was developed for the determination of lactic acid. The bacterial cells were retained on a nylon membrane and attached to the surface of the oxygen electrode. In view of response time, stability and sensitivity, the biosensor performed best at 26°C and in pH 6 phthalate buffer containing magnesium sulfate. The activity of the retained cells was stable for approximately 170 h and was regenerable. The biosensor exhibited a hyperbolic response to both D- and L-lactic acid in the range of 10⁻⁴ M to 25 × 10⁻³ M. However, in the range 10⁻⁴ M to 15 × 10⁻⁴ M the response was linear. The microbial biosensor was applicable for detecting lactate concentration in yogurt and milk, since it was not sensitive to lactose, sucrose and glucose — three major components of such dairy products.     

A performance comparison of choline biosensors: anodic or cathodic detections of H2O2 generated by enzyme immobilized on a conducting polymer

Amperometric choline biosensors were fabricated by the covalent immobilization of an enzyme of choline oxidase (ChO) and a bi-enzyme of ChO/horseradish peroxidase (ChO/HRP) onto poly-5,2′:5′,2″-terthiophene-3′-carboxylic acid (poly-TTCA) modified electrodes (CPMEs). A sensor modified with ChO utilized the oxidation process of enzymatically generated H₂O₂ in a choline solution at +0.6 V. The other one modified with ChO/HRP utilized the reduction process of H₂O₂ in a choline solution at −0.2 V. Experimental parameters affecting the sensitivity of sensors, such as pH, applied potential, and temperature were optimized. A performance comparison of two sensors showed that one based on ChO/HRP/CPME had a linear range from 1.0×10⁻⁶ to 8.0×10⁻⁵ M and the other based on ChO/CPME from 1.0×10⁻⁶ to 5.0×10⁻⁵ M. The detection limits for choline employing ChO/HRP/CPME and ChO/CPME were determined to be about 1.0×10⁻⁷ and 4.0×10⁻⁷ M, respectively. The response time of sensors was less than 5 s. Sensors showed good selectivity to interfering species. The long-term storage stability of the sensor based on ChO/HRP/CPME was longer than that based on ChO/CPME.     

Activity and storage stability of immobilized glucose oxidase onto magnesium silicate

Glucose oxidase (GOD) was covalently immobilized onto florisil (magnesium silicate) carrier via glutaraldehyde. Immobilization conditions were optimized: the amount of initial GOD per grams of carrier as 5 mg, pH as 5.5, immobilization time as 120 min and temperature as 10 °C. Under the optimized reaction conditions activities of free and immobilized GOD were measured. Free and immobilized GOD samples were characterized with their kinetic parameters, and thermal and storage stabilities. K_M and V_max values were 68.2 mM and 435 U mg GOD⁻¹ for free and 259 mM and 217 U mg GOD⁻¹ for immobilized enzymes, respectively. Operational stability of the immobilized enzyme was also determined by using a stirred batch type column reactor. Immobilized GOD was retained 40% of its initial activity after 50 reuses. Storage stabilities of the immobilized GOD samples stored in the mediums with different relative humidity in the range of 0–100% were investigated during 2 months. The highest storage stability was determined for the samples stored in the medium of 60% relative humidity. Increased relative humidity from 0% to 60% caused increased storage stability of immobilized GODs, however, further increase in relative humidity from 80% to 100% caused a significant decrease in storage stability of samples.     

Biochemical and biophysical studies on cytochrome c oxidase XV. Reaction with fluoride

1. Fluoride is a mixed-type inhibitor of the cytochrome c oxidase activity with a K_i for the free enzyme of 10 mM and a K_i for the cytochrome c-complexed enzyme of 35 mM.

2. Fluoride shifts the γ-band of the enzyme from 423 to 421 nm and the -band from 597 to 598 nm. The difference spectrum (oxidized enzyme in the presence of fluoride minus oxidized enzyme) has peaks at 400, 453, 482, 605 and 638 nm and troughs at 430, 520, 552 and 674 nm. The changes in absorbance are small (about 3% at absorbance maxima) with respect to those of other hemoproteins.

3. On addition of fluoride to isolated cytochrome c oxidase 3 reactions can be distinguished: (I) a bimolecular binding reaction (K_on = 4 M⁻¹ · s⁻¹ and k_off = 2.9 · 10⁻²s⁻¹ at 25 °C, pH 7.4) contributing at 638 nm and 430 nm; (II) a first-order reaction (k = 2.4 · 10⁻²) s⁻¹ at 22 °C, pH 7.2) visible mainly at 430 nm and (III) a very slow reaction with a half-time in the order of 10 min.

4. The spectroscopic dissociation constants for the fluoride binding, determined from Hill plots using the absorbance changes at 638 and 430 nm, are similar (7 and 10 mM, respectively, at 22 °C, pH 7.2).

5. A mechanism for the reaction is discussed in which the bimolecular binding reaction is followed by a conformational change of the enzyme-fluoride complex.     

The respiratory chain of Azotobacter vinelandii II. The effect of cyanide on cytochrome d

1. Cyanide causes a slow disappearance of the oxidized band (648 nm) of cytochrome d in particles of Azotobacter vinelandii and inhibits the appearance of the reduced band (631 nm). No effect of cyanide is found on the reduced band of cytochrome d.

2. The kinetics of the disappearance of the 648-nm band of cytochrome d with excess cyanide deviates from first-order kinetics at lower temperatures (22 °C) indicating that at least two conformations of the enzyme are involved. At higher temperatures (32 °C) the observed kinetics of the cyanide reaction are first order with a k_on = 0.7 M⁻¹·s⁻¹ and with an estimated k_off of approximately 5·10⁻⁵ s⁻¹.

3. The value of the k_off (7·10⁻⁴−14·10⁻⁴ s⁻¹ at 32 °C) determined from the rate of reduction of cyanocytochrome d by Na₂S₂O₄ or NADH is one order of magnitude larger than the k_off value found when the enzyme is in its oxidized state.

4. No effect of cyanide is found on the spectrum of cytochrome a₁.     

Bradykinin counteracts the stimulatory effect of angiotensin-(1-7) on the proximal tubule Na+ -ATPase activity through B2 receptor

Recently, we demonstrated that angiotensin-(1–7) (Ang-(1–7)) stimulates the Na⁺-ATPase activity through a losartan-sensitive angiotensin receptor, whereas bradykinin inhibits the enzyme activity through the B₂ receptor [Regul. Pept. 91 (2000) 45; Pharmacol. Rev. 32 (1980) 1]. In the present paper, the effect of bradykinin (BK) on Ang-(1–7)-stimulated Na⁺-ATPase activity was evaluated. Preincubation of Na⁺-ATPase with 10⁻⁹ M Ang-(1–7) increases enzyme activity from 7.9±0.9 to 14.1±1.5 nmol Pi mg⁻¹ min⁻¹, corresponding to an increase of 79% (p<0.05). This effect is reverted by bradykinin in a dose-dependent manner (10⁻¹⁴–10⁻⁸ M), reaching maximal inhibitory effect at 10⁻⁹ M. Des-Arg⁹ bradykinin (DABK), an agonist of B₁ receptor, at the concentrations of 10⁻⁹–10⁻⁷ M, does not mimic the BK inhibitory effect, and des-Arg⁹-[Leu⁸]-BK (DALBK), a B₁ receptor antagonist, at the concentrations of 10⁻¹⁰–10⁻⁷ M, does not prevent the inhibitory effect of BK on Ang-(1–7)-stimulated enzyme. On the other hand, HOE 140, an antagonist of B₂ receptor, abolishes the inhibitory effect of BK on the Ang-(1–7)-stimulated enzyme in a dose-dependent manner, reaching maximal effect at 10⁻⁷ M. Taken together, these data indicate that stimulation of B₂ receptors by BK can counteract the stimulatory effect of Ang-(1–7) on the proximal tubule Na⁺-ATPase activity.     

Estradiol inhibits the estrone sulfatase activity in normal and cancerous human breast tissues

It is well accepted that estradiol (E₂) plays an important role in the genesis and evolution of breast cancer. Quantitative evaluation indicates that in human breast tumor, estrone sulfate (E₁S) ‘via sulfatase’ is a much more likely precursor for E₂ than is androstenedione ‘via aromatase’. In previous studies, it was demonstrated that in isolated MCF-7 and T-47D breast cancer cell lines, estradiol can block estrone sulfatase activity. In the present study, the effect of E₂ was explored using total normal and cancerous breast tissues. This study was carried out with post-menopausal patients with breast cancer. None of the patients had a history of endocrine, metabolic or hepatic diseases or had received treatment in the previous 2 months. Each patient received local anaesthetic (lidocaine 1%) and two regions of the mammary tissue were selected: (A) the tumoral tissue and (B) the distant zone (glandular tissue) which was considered as normal. Samples were placed in liquid nitrogen and stored at –80 °C until enzyme activity analysis. Breast cancer histotypes were ductal and post-menopausal stages were T2. Homogenates of tumoral or normal breast tissues (45–75 mg) were incubated in 20 mM Tris–HCl, pH 7.2 with physiological concentrations of [³H]-E₁S (5 × 10⁻⁹ M) alone or in the presence of E₂ (5 × 10⁻⁵ to 5 × 10⁻⁷ M) during 30 min or 3 h. E₁S, E₁ and E₂ were characterized by thin layer chromatography and quantified using the corresponding standard. The sulfatase activity is significantly more intense with the breast cancer tissue than normal tissue, since the concentration of E₁ was 3.20 ± 0.15 and 0.42 ± 0.07 pmol/mg protein, respectively after 30 min incubation. The values were 27.8 ± 1.8 and 3.5 ± 0.21 pmol/mg protein, respectively after 3 h incubation. Estradiol at the concentration of 5 × 10⁻⁷ M inhibits this conversion by 33% and 31% in cancerous and normal breast tissues, respectively and by 53% and 88% at the concentration of 5 × 10⁻⁵ M after 30 min incubation. The values were 24% and 18% for 5 × 10⁻⁷ M and 49% and 42% for 5 × 10⁻⁵ M, respectively after 3 h incubation. It was observed that [³H]-E₁S is only converted to [³H]-E₁ and not to [³H]-E₂ in normal or cancerous breast tissues, which suggests a low or no 17β-hydroxysteroid dehydrogenase (17β-HSD) Type 1 reductive activity in these experimental conditions. In conclusion, estradiol is a strong anti-sulfatase agent in cancerous and normal breast tissues. This data can open attractive perspectives in clinical trials using this hormone.     

Cholesterol sulphate sulphohydrolase from human placenta microsomes--purification and properties of the dephosphorylated form of enzyme

The procedure for purification of cholesterol sulphate sulphohydrolase (ChS-ase) from human placenta microsomes was elaborated. The highy purified enzyme preparation (specific activity 2000 nmol×min⁻¹×mg protein⁻¹) exhibited optimal activity at pH 9.0. The K_m value was established to be 1.5±0.85×10⁻⁵ M. The high molecular weight form (200 kDa) and the low molecular weight form (20 kDa) of the enzyme were separated. The interconversion of the high molecular weight variant into the low one occurs under the influence of dephosphorylation. Both forms exhibited typical Michaelis–Menten saturation kinetics. The effect of different compounds on the enzyme activity was tested.     

Bradykinin dilates rat middle cerebral artery and its large branches via endothelial B2 receptors and release of nitric oxide

Ring segments of rat middle cerebral artery (MCA) were prepared for measurement of isometric force and precontracted with 10⁻⁴ M uridine triphosphate (UTP). Concentration-effect curves (CEC) were constructed for bradykinin (BK, 10⁻⁸–10⁻⁵ M) in segments with functionally intect (E+) or denuded (E−) endothelium. E− segments did not dilate to BK. The BK receptor was characterized by application of specific B₁ or B₂ antagonists [des-Arg⁹-Leu⁸] BK (10⁻⁵ M) and [ -Arg⁰-Hyp³-Thi⁵- -Tic⁷-Oic⁸] BK (HOE140,3 × 10⁻⁷ M), respectively, or B₁ agonist [des-Arg⁹] BK (10⁻⁸–10⁻⁴ M). Involvement of nitric oxide (NO) was tested with N^G-nitro- -arginine (LNNA, 10⁻⁴ M). BK induced concentration-dependent relaxation with a maximal effect (E_max) of 40.86 ± 1.50% at 10⁻⁶ M and a pD₂ (−log₁₀ EC₅₀) of 6.818 ± 0.044. This relaxation could be prevented with HOE140 or LNNA, but was not influenced by [des-Arg⁹-Leu⁸] BK. [des-Arg⁹] BK did not induce any effect. These results demonstrate that BK induced relaxation via endothelial B₂ receptors and release of NO in isolated rat MCA.