- Email: [email protected]
High-Performance Liquid Chromatographic Determination of Anandamide Amidase Activity in Rat Brain Microsomes
ANALYTICAL BIOCHEMISTRY ARTICLE NO.
238, 40–45 (1996)
0247
High-Performance Liquid Chromatographic Determination of Anandamide Amidase Activity in Rat Brain Microsomes Wensheng Lang,* Ce Qin,* W. Adam G. Hill,* Sonyuan Lin,* Atmaram D. Khanolkar,* and Alexandros Makriyannis*,†,‡,1 *School of Pharmacy, †Department of Molecular and Cell Biology, and ‡Institute of Materials Science, University of Connecticut, Storrs, Connecticut 06269
Received February 12, 1996
A rapid, sensitive, and reliable method for measuring anandamide amidase activity in rat brain microsomes by reversed-phase high-performance liquid chromatography (RP-HPLC) and its applications are described. Enzymatic activity was assayed by the determination of the rates of hydrolysis of anandamide or its analogs at 377C. The reaction products were separated using an ODS guard column eluted with aqueous phosphoric acid–acetonitrile and quantitated with uv detection at 204 nm and an external standard method. Baseline separation of the acid products from their substrates was completed in less than 2 min. The detection limits were 1.4 pmol for arachidonic acid and 0.22 pmol for anandamide at a signal to noise ratio of 4:1. The stability of anandamide in the acidic mobile phase was tested, and no significant decomposition was observed up to 1 h. The method was successfully applied to the examination of substrate specificity as well as for testing the ability of amidase inhibitors to block its hydrolysis. Kinetic constants obtained for (S)-methanandamide were an apparent Km of 8.6 { 1.3 mM and a Vmax of 362 { 16 pmol/min/mg of protein. A highly potent inhibitor, palmitylsulfonyl fluoride (PSF), was found to have an IC50 of 50 nM. PSF is 210 times as potent as phenylmethylsulfonyl fluoride. The method offers several advantages over existing methodology using radioisotopes or a solvent extraction procedure. q 1996 Academic Press, Inc.
An amidohydrolase (anandamide amidase) present in rat brain microsomes catalyzes the hydrolysis of ara-
1
To whom correspondence should be addressed at School of Pharmacy, University of Connecticut, Storrs, CT 06269. Fax: (860) 4863089.
chidonylethanolamide (anandamide, AEA2), an endogenous brain cannabinoid receptor agonist (1–3), into arachidonic acid (AA) and ethanolamine (EA) (4). Previous studies have demonstrated that the enzyme, which breaks down AEA, is abundantly distributed in the regions where a high density of cannabinoid receptor exists in the central nervous system (5). Recently, the enzyme was partially purified and characterized as a highly hydrophobic protein with Mr ca. 60 kDa (6). There is much interest at present in the search for AEA analogs which have high binding affinity for the cannabinoid receptor and are resistant to hydrolysis by the enzyme (7, 8). Moreover, a lot of effort has been made to find selective AEA amidase inhibitors which do not bind to the receptor (9). Previous assays for AEA amidase activity commonly rely on the measurement of radioactivity of acid products formed from 3H- or 14C-labeled AEA or its analogs. A normal-phase thin-layer chromatographic (TLC) method for the separation of 3H-labeled arachidonates was described by Deutsch et al. (4, 9). A reversed-phase TLC separation coupled with phosphor-imaging quantitation was also described by Devane and Axelrod (10). Non-TLC methods, including minicolumn chromatography (5) or a solvent extraction approach using [1,214 C]-labeled ethanolamide (11), were recently reported. Thus far, nonradioactive assays for the amidase activity have not been reported. In the present paper we describe a reversed-phase HPLC method for measuring substrate specificity and inhibitor potency of AEA amidase using direct uv detection. The method does not require radioisotopes. 2 Abbreviations used: AEA, arachidonylethanolamide; AA, arachidonic acid; BSA, bovine serum albumin; PSF, palmitylsulfonyl fluoride; LEA, linoleyl ethanolamide; DMSO, dimethyl sulfoxide; MAEA, arachidonyl-1*-hydroxy-2*-propylamide; HBA, N-(4-hydroxy-benzyl)arachidonamide; PMSF, phenylmethylsulfonyl fluoride.
40
AID
0003-2697/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AB 9600
/
6m18$$$241
05-29-96 02:11:26
aba
AP: Anal Bio
CHROMATOGRAPHIC DETERMINATION OF ANANDAMIDE AMIDASE
MATERIALS AND METHODS
Materials Arachidonic acid, linoleic acid, phenylmethylsulfonyl fluoride, and essentially fatty acid-free BSA were purchased from Sigma Chemical Co. (St. Louis, MO). Palmityl alcohol, 4-hydroxybenzonitrile, ethanolamine, and other amino alcohols were obtained from Aldrich Chemical Co. (Allentown, PA). HPLC grade acetonitrile was purchased from Fisher Scientific Co. (Pittsburgh, PA). Anandamide and its analogs were synthesized from appropriate fatty acid chlorides and amino alcohols as previous described (7). Palmitylsulfonyl fluoride (PSF) was essentially prepared from palmitylsulfonyl chloride and ammonium hydrofluoride. Preparation of Rat Brain Microsomal Fraction Adult Sprague–Dawley rat brains obtained from PelFreez Co. (Rogers, AK) were homogenized with a Potter–Elvehjem tissue grinder in 5 vol of ice-cold buffer containing 20 mM TrisrHCl and 1 mM EGTA (pH 7.4). The homogenate was centrifuged at 1000g for 10 min, and the supernatant was further centrifuged at 10,700g for 30 min and 105,000g for 60 min at 47C, sequentially. The pellet from the last centrifugation step (microsomal fraction) was resuspended in 40 ml of Tris buffer. Aliquots (1.5 ml) of the suspension were stored at 0807C until used. No loss of amidase activity was observed for at least 2 months. Protein concentration was determined using a modification of Lowry’s method (12).
HPLC was performed using a Beckman System Gold (Fullerton, CA) consisting of a 128 pump, a 166 uv detector, and a Rheodyne 7725i injector with a 20-ml loop. An IBM 466DX2/Si computer was interfaced to the system, and Beckman System Gold software was used for system control and data processing. Separations were carried out on a Beckman Ultrasphere ODS guard column (5 mm, 45 1 4.6 mm i.d.). The mobile phase was 8.5% aqueous phosphoric acid/acetonitrile (10:90, v/v) at a flow rate of 1 ml/min. Quantitation was based on the integration of peak areas at 204 nm. The amount of product formed was calculated from calibration curves of appropriate standards. Enzyme Assay For the standard assay of the amidase activity, 50 nmol of substrate (AEA) was incubated with 150 mg rat brain microsomal protein in a final volume of 500 ml buffer containing 50 mM TrisrHCl, 1 mM EDTA, and 0.1% BSA (pH 7.4) at 377C for 15 min with shaking. The reaction was terminated by pipetting 200 ml of the
AB 9600
suspension into a microcentrifuge tube containing 800 ml acetonitrile. The tube was vortexed, then centrifuged at room temperature for 4 min to remove the proteins. Twenty microliters of the supernatant was injected into the HPLC system for quantitation. An identical incubation was carried out in the absence of substrate to measure the amount of arachidonic acid existing naturally in the brain microsomal preparation. The rate of hydrolysis was calculated from the average of four experiments with results determined in duplicate. The amount of AA in the blank was subtracted from that in the sample. For the amidase inhibition assay, various amounts of test compound, 50 nmol of anandamide, and 150 mg rat brain microsomal protein were incubated as above. An identical incubation was carried out in the absence of inhibitor as a control. The same amount of protein was also incubated in the absence of both anandamide and inhibitor as a blank. The assays were performed in quadruplicate. The percentage of inhibition was calculated directly as % inhibition Å
(Ac 0 As ) 1 100, (Ac 0 Ab )
where As refers to the peak area of AA from a sample, Ac is the peak area of AA from a control, and Ab is the peak area from a blank. RESULTS
Fundamental Conditions for the Measurement of AEA Amidase Activity
HPLC Analysis
AID
41
/
6m18$$$242
05-29-96 02:11:26
AEA analogs and their acid products are highly lipophilic and strongly retained on the ODS column. Accordingly, a short guard column and a high percentage (90% v/v) of organic modifier in the mobile phase were used. The major advantage of the use of RP-HPLC is that the polar impurities from microsomes can be easily eluted from the column. Representative chromatograms showing the separation of AA and AEA are shown in Fig. 1. Sharp, symmetrical, and well-resolved AEA and AA peaks were obtained under the acidic conditions. The retention times for AEA and AA were 1.0 and 1.5 min, respectively. One unidentified peak (Rt 1.32 min) was recognized from microsomes, and no other metabolites were detected. The AA peak is very susceptible to the pH of the mobile phase. When the pH was adjusted to 2.5 or above, a broad tailing peak of AA was obtained, which led to a remarkable loss of sensitivity. A low pH is necessary to prevent ionization of AA. Under such chromatographic conditions, the retention times for linoleyl ethanolamide (LEA) and linoleic acid (LA) are 1.06 and 1.75 min, respectively. Preliminary experiments showed that AA occurs nat-
aba
AP: Anal Bio
42
LANG ET AL.
creasing enzyme activity in the presence of methanol and ethanol was detected at concentrations above 0.5%. In our enzymatic assays, DMSO was chosen as the solvent and the final DMSO concentrations were less than 1%. Stability of AEA and AA
FIG. 1. Chromatogram (A) of a standard solution containing anandamide (1.00 min) and arachidonic acid (1.50 min); (B) representative chromatogram of a sample of 100 mM anandamide incubated with 0.3 mg/ml microsomal proteins for 15 min. The unidentified peak at 1.32 min was from the microsomal preparation. Chromatographic conditions and methods for sample preparation are given under Materials and Methods.
urally in the microsomal preparations in a concentration range of 23–33 nmol/ml. This portion of AA interferes with the determination of that produced by enzymatic hydrolysis. In order to remove the interference, an identical experiment was performed in the presence of the same amount of enzyme incubated at 377C. The net amount of AA hydrolyzed by the enzyme could be obtained by subtracting the amount of AA in a blank from the total amount in the sample. The effect of organic solvents, e.g., dimethyl sulfoxide (DMSO), methanol, or ethanol, on the enzymatic activity was also studied because a solution of the test compounds had to be made in the solvents prior to dissolving them in aqueous medium. No significant effect of DMSO was found at concentrations below 5%, while slightly de-
AID
AB 9600
/
6m18$$$242
05-29-96 02:11:26
The stability of AEA in the acidic mobile phase was studied by the incubation of AEA in 8.5% aqueous phosphoric acid/acetonitrile (10:90) at ambient temperature. Twenty microliters of the solution was injected into the HPLC system after the first minute and at 5-min intervals for 1 h. The results showed that no significant change of AEA concentration and no formation of AA were detected during this time (data not shown). Thus, the hydrolysis of AEA during a chromatographic run is negligible. In addition, the stability of AEA in 50 mM Tris buffer (pH 7.4) at 377C was also tested, and a similar result was obtained. In contrast, AEA breakdown by AEA amidase is much faster than its natural hydrolysis in the biological buffer. The stability of AA in the same Tris buffer in the absence or presence of 0.3 mg/ml microsomal protein was also studied using the above method. There was a small, but significant fall in AA concentration with increasing incubation time. The data were fitted with both firstorder rate reaction and linear equations, and the same rate constants were obtained in both cases. The slopes (concentration decrease) were 0.12 and 0.22% per minute in the presence and absence of microsomal protein, respectively. The results indicate that AA is twice as stable in the presence of microsomal protein as in its absence. Less than 2% AA decomposed during the 15-min incubation used in the activity assay. Reproducibility The factors which affect reproducibility could be chromatographic, due to the precipitation of protein with acetonitrile and variations in centrifugation. The reproducibility of the chromatographic assay was evaluated by the injection of 20 ml of standard AA solution, at seven different concentrations, in quintuplicate. In Table 1, the reproducibility of the HPLC is represented by the coefficients of variation for AA, which lie between 0.5 and 3.4%. The overall reproducibility of the method is represented by the standard deviations (SDs) of the relative rates of hydrolysis for different AEA analogs and is shown in Table 2. The experiments were carried out for each compound in quadruplicate. The results indicate that the variation of SDs of the relative rates of hydrolysis for five compounds is between 1.2 and 4.2%, while the coefficients of variation for the relative rates of hydrolysis depend on the absolute rate values. Because a high percentage of acetonitrile (90%, v/v)
aba
AP: Anal Bio
43
CHROMATOGRAPHIC DETERMINATION OF ANANDAMIDE AMIDASE TABLE 1
TABLE 3
Reproducibility of HPLC Determination of AA in Standard Solutions (n Å 5)
Recovery (%) of AA Incubated in 0.3 mg/ml Microsomal Protein (n Å 4)
AA (nmol/ml)
Peak area (AU)
0.305 0.611 1.22 2.44 4.89 9.75 19.5
0.059 0.271 0.711 1.28 2.63 4.84 10.1
{ { { { { { {
Coefficient of variation (%)
0.002 0.009 0.015 0.006 0.028 0.035 0.17
3.4 3.3 2.1 0.5 1.1 0.7 1.7
in the mobile phase was used, most contaminants from microsomes are eluted together with the solvent front. The HPLC method is very reliable, and after 5000 runs and 6 months, no changes in retention time and peak resolution were observed.
AA added (nmol)
AA measured (nmol)
Recovery (%)
1.15 2.88 17.3 34.5
1.20 2.62 17.1 33.1
104 91 99 96
{ { { {
0.10 0.04 0.4 0.6
{ { { {
8.7 1.4 2.3 1.7
centrations. Good linearity for AA was obtained over a concentration range of 0.3 to 20 nmol/ml. The regression line was Y (peak area) Å 0.515X (concentration, nmol/ml) 0 0.007 with a correlation coefficient of 0.9996. The curve was used for the calibration of the concentrations in samples. The detection limits of 0.22 pmol for anandamide and 1.4 pmol for arachidonic acid were obtained at a signal to noise ratio of 4:1.
Recovery of AA The recovery of AA after precipitation of proteins with acetonitrile and centrifugation was investigated using a standard addition method. An exact amount of a standard AA solution was added to a microsomal suspension containing 150 mg protein and a known amount of AA in a final volume of 500 ml. The concentration of the standard AA solution prior to the addition was confirmed by HPLC analysis. The suspension was incubated, mixed with acetonitrile, and centrifuged as previously described under Materials and Methods. The supernatant was analyzed by HPLC. The results are listed in Table 3. The average recovery of AA obtained was 97.5% in the working concentration range.
AEA Amidase Activity Assay Optimization The enzymatic activity assays were conducted at pH 7.4, which is in the optimal pH range as previously reported (5). The effect of enzyme concentration on the hydrolysis of AEA was investigated as shown in Fig. 2, by altering the amount of rat brain microsomes added to the incubation mixture. AEA (100 mM) was incubated with the microsomal protein in a range of
Linearity and Detection Limits Linear relationships between peak area and concentration of AA and AEA were observed at different con-
TABLE 2
Anandamide Analog Specificity for Anandamide Amidase and CB1 Receptor (n Å 4) Rate of hydrolysis (pmol/min/mg) AEA (R)-MAEA (S)-MAEA HBA LEA
1300 29.7 297 300 982
{ { { { {
23.0 16.0 22.0 32.5 55.0
Relative rate of hydrolysis (%) 100 2.3 23 23 75.5
{ { { { {
1.8 1.2 1.7 2.5 4.2
Affinity for CB1 receptor (7) Ki (nM) 78 20 172 217
{ { { {
2 1.6 26 3
Note. AEA, anandamide; MAEA, methanandamide; HBA, N-(4hydroxybenzyl)-arachidonamide; LEA, linoleyl ethanolamide.
AID
AB 9600
/
6m18$$$242
05-29-96 02:11:26
FIG. 2. Dependence of initial rate of hydrolysis of AEA by anandamide amidase in rat brain microsomes on enzyme concentration. Assays were performed in 50 mM Tris buffer (pH 7.4) as described under Materials and Methods. The activity increases linearly with increasing protein concentration up to 0.8 mg/ml. Data shown are means { SD of four independent experiments.
aba
AP: Anal Bio
44
LANG ET AL.
FIG. 3. Progression of AEA hydrolysis by anandamide amidase in rat brain microsomes. AA formed from hydrolysis was determined by HPLC with uv detection at 204 nm. The chromatographic conditions are described under Materials and Methods. Data shown are means { SD of three separate measurements.
0.1 to 1.0 mg/ml for 15 min at 377C. The results obtained were linear up to at least 0.8 mg/ml of microsomal protein. The dependence of the enzymatic hydrolysis observed at 100 mM anandamide and 0.3 mg/ml microsomal protein on incubation time was also investigated. As shown in Fig. 3, the results indicate complete linearity over the entire range of 60 min. Fifteen minutes was selected as routine incubation time for convenience, and approximately 10% of AEA was hydrolyzed during this time. Enzymatic activity as a function of the substrate concentration was examined. The results of a representative experiment to determine the initial velocity of (S)-methanandamide (S-MAEA) hydrolysis at various substrate concentrations is shown in Fig. 4. Kinetic analysis was performed by fitting a hyperbola to the data using nonlinear least-squares regression and yielded an apparent Km of 8.6 { 1.3 mM and a Vmax of 362 { 16 pmol/min/mg of protein.
FIG. 4. Dependence of anandamide amidase activity on substrate concentration. Rectangular hyperbolic plot of initial rate of AA formed from AEA amidase hydrolysis vs AEA concentration. Apparent Km and Vmax were determined by fitting hyperbola to the data. The rate of hydrolysis is in units of nmols AA/min/mg of protein.
fluoride derivative, PSF, was synthesized and tested for its inhibitory activity for AEA amidase. For the inhibitor assays, AEA was incubated with microsomal protein in the presence of a series of concentrations of inhibitor at 377C. The results are shown in Fig. 5. The inhibitory ability was expressed as IC50 , the concentration of the inhibitor which results in 50% inhibition of
Applications The method was first applied to the determination of differences in substrate specificity. The test substrates include AEA, a pair of chiral anandamide congeners [(R)- and (S)-arachidonyl-1*-hydroxy-2*-propylamide (MAEA)], a phenolic derivative of anandamide [N-(4hydroxy-benzyl)-arachidonamide, HBA], and LEA. The substrate specificity is expressed as the relative rate of hydrolysis to that of AEA. Data shown in Table 2 were the means { SD for four separate experiments. Previous studies showed that phenylmethylsulfonyl fluoride (PMSF) inhibits the breakdown of AEA by the amidase in brain homogenates (4, 5). A new sulfonyl
AID
AB 9600
/
6m18$$$242
05-29-96 02:11:26
FIG. 5. Mean concentration–inhibition curves of inhibitors, PSF (s) and PMSF (l), for anandamide amidase activity in rat brain microsomes. Each value represents the mean { SD of inhibition from four independent experiments. Data analyzed by nonlinear leastsquares regression. The IC50 values are 50 nM for PSF and 10.5 mM for PMSF.
aba
AP: Anal Bio
CHROMATOGRAPHIC DETERMINATION OF ANANDAMIDE AMIDASE
the enzymatic reaction. Each IC50 value was calculated based on the curve fitting of inhibition percentage versus logarithm of inhibitor concentration. The IC50s were 50 nM for PSF and 10.5 mM for PMSF, indicating that the inhibitory activity of PSF is 210 times as high as that of PMSF. The details of AEA amidase inhibition by PSF will be published elsewhere.
45
hydrolysis of (R)-methanandamide is about 10-fold less than that of its (S)-isomer. Interestingly, when tested for their affinities for the cannabinoid receptor, this pair of enantiomers exhibited a stereoselectivity opposite that of their enzyme activity, with (R)-methanandamide having the higher affinity (7). ACKNOWLEDGMENTS
DISCUSSION
The previous methods (4–6, 9–11) for AEA amidase assays limit testing to radiolabeled substrates only. A method for the determination of AEA or its analogs by HPLC with a prefluorogenic derivatization described by Koga et al. is not suitable for the measurement of acid products (13). Our method using RP-HPLC with direct uv detection can generally be applied to uv active substrates and has sufficient sensitivity at a picomolar level for their acid products. Using this method, we overcame the disadvantage of using radioactivity. Also, HPLC provides much better resolution and reproducibility than do TLC and column chromatography (4– 6). The solvent extraction approach described by Omeir et al. (11) lacks specificity. Our method using solvent precipitation of protein and centrifugation offers acceptable recovery and does not require a complicated sample clean-up procedure. The use of a short guard column for fast separation and determination allows significant time and reagent saving. The simple, fast, and sensitive method is suitable not only for the assay of AEA amidase activity, but also for that of AEA synthase, with minor modifications. We applied the method to the determination of substrate specificity and inhibitor activity for AEA amidase. The results of hydrolysis of different substrates indicated that anandamide was hydrolyzed at the highest rate among the five substrates tested. We found that the chiral pair of AEA analogs, (R)- and (S)-methanandamide, in which a methyl group was introduced in the a-position of the ethanolamine head group showed great resistance to enzymatic hydrolysis. The rate of
AID
AB 9600
/
6m18$$$243
05-29-96 02:11:26
This work was supported by Grants DA3801 and DA7215 from the National Institute on Drug Abuse. We thank Dr. Pusheng Fan for his excellent technical assistance with the protein assays.
REFERENCES 1. Devane, W. A., Hanus, L., Breuer, A., Pertwee, R. G., Stevenson, L. A., Griffin, G., Gibson, D., Mandelbaum, A., Etinger, A., and Mechoulam, R. (1992) Science 258, 1946–1949. 2. Felder, C. C., Briley, E. M., Axelrod, J., Simpson, J. T., Mackie, K., and Devane, W. A. (1993) Proc. Natl. Acad. Sci. USA 90, 7656–7660. 3. Mackie, K., Devane, W. A., and Hille, B. (1993) Mol. Pharmacol. 44, 498–503. 4. Deutsch, D. G., and Chin, S. A. (1993) Biochem. Pharmacol. 46, 791–796. 5. Desarnaud, F., Cadas, H., and Piomelli, D. (1995) J. Biol. Chem. 270, 6030–6035. 6. Ueda, N., Kurahashi, Y., Yamamoto, S., and Tokunaga, T. (1995) J. Biol. Chem. 270, 23823–23827. 7. Abadji, V., Lin, S., Taha, G., Griffin, G., Stevenson, L. A., Pertwee, R. G., and Makriyannis, A. (1994) J. Med. Chem. 37, 1889– 1893. 8. Pertwee, R. G., Fernando, S. R., Griffin, G., Abadji, V., and Makriyannis, A. (1995) Eur. J. Pharmacol. 272, 73–78. 9. Koutek, B., Prestwich, G. D., Howlett, A. C., Chin, S. A., Salehani, D., Akhavan, N., and Deutsch, D. G. (1994) J. Biol. Chem. 269, 22937–22940. 10. Devane, W. A., and Axelrod, J. (1994) Proc. Natl. Acad. Sci. USA 91, 6698–6701. 11. Omeir, R. L., Chin, S., Hong, Y., Ahern, D. G., and Deutsch, D. G. (1995) Life Sci. 56, 1999–2005. 12. Markwell, M. A. K., Haas, S. M., Bieber, L. L., and Tolbert, N. E. (1978) Anal. Biochem. 87, 206–210. 13. Koga, D., Santa, T., Hagiwara, K., Imai, K., Takizawa, H., Nagano, T., Hirobe, M., Ogawa, M., Sato, T., and Inoue, K. (1994) Biomed. Chromatogr. 9, 56–57.
aba
AP: Anal Bio
238, 40–45 (1996)
0247
High-Performance Liquid Chromatographic Determination of Anandamide Amidase Activity in Rat Brain Microsomes Wensheng Lang,* Ce Qin,* W. Adam G. Hill,* Sonyuan Lin,* Atmaram D. Khanolkar,* and Alexandros Makriyannis*,†,‡,1 *School of Pharmacy, †Department of Molecular and Cell Biology, and ‡Institute of Materials Science, University of Connecticut, Storrs, Connecticut 06269
Received February 12, 1996
A rapid, sensitive, and reliable method for measuring anandamide amidase activity in rat brain microsomes by reversed-phase high-performance liquid chromatography (RP-HPLC) and its applications are described. Enzymatic activity was assayed by the determination of the rates of hydrolysis of anandamide or its analogs at 377C. The reaction products were separated using an ODS guard column eluted with aqueous phosphoric acid–acetonitrile and quantitated with uv detection at 204 nm and an external standard method. Baseline separation of the acid products from their substrates was completed in less than 2 min. The detection limits were 1.4 pmol for arachidonic acid and 0.22 pmol for anandamide at a signal to noise ratio of 4:1. The stability of anandamide in the acidic mobile phase was tested, and no significant decomposition was observed up to 1 h. The method was successfully applied to the examination of substrate specificity as well as for testing the ability of amidase inhibitors to block its hydrolysis. Kinetic constants obtained for (S)-methanandamide were an apparent Km of 8.6 { 1.3 mM and a Vmax of 362 { 16 pmol/min/mg of protein. A highly potent inhibitor, palmitylsulfonyl fluoride (PSF), was found to have an IC50 of 50 nM. PSF is 210 times as potent as phenylmethylsulfonyl fluoride. The method offers several advantages over existing methodology using radioisotopes or a solvent extraction procedure. q 1996 Academic Press, Inc.
An amidohydrolase (anandamide amidase) present in rat brain microsomes catalyzes the hydrolysis of ara-
1
To whom correspondence should be addressed at School of Pharmacy, University of Connecticut, Storrs, CT 06269. Fax: (860) 4863089.
chidonylethanolamide (anandamide, AEA2), an endogenous brain cannabinoid receptor agonist (1–3), into arachidonic acid (AA) and ethanolamine (EA) (4). Previous studies have demonstrated that the enzyme, which breaks down AEA, is abundantly distributed in the regions where a high density of cannabinoid receptor exists in the central nervous system (5). Recently, the enzyme was partially purified and characterized as a highly hydrophobic protein with Mr ca. 60 kDa (6). There is much interest at present in the search for AEA analogs which have high binding affinity for the cannabinoid receptor and are resistant to hydrolysis by the enzyme (7, 8). Moreover, a lot of effort has been made to find selective AEA amidase inhibitors which do not bind to the receptor (9). Previous assays for AEA amidase activity commonly rely on the measurement of radioactivity of acid products formed from 3H- or 14C-labeled AEA or its analogs. A normal-phase thin-layer chromatographic (TLC) method for the separation of 3H-labeled arachidonates was described by Deutsch et al. (4, 9). A reversed-phase TLC separation coupled with phosphor-imaging quantitation was also described by Devane and Axelrod (10). Non-TLC methods, including minicolumn chromatography (5) or a solvent extraction approach using [1,214 C]-labeled ethanolamide (11), were recently reported. Thus far, nonradioactive assays for the amidase activity have not been reported. In the present paper we describe a reversed-phase HPLC method for measuring substrate specificity and inhibitor potency of AEA amidase using direct uv detection. The method does not require radioisotopes. 2 Abbreviations used: AEA, arachidonylethanolamide; AA, arachidonic acid; BSA, bovine serum albumin; PSF, palmitylsulfonyl fluoride; LEA, linoleyl ethanolamide; DMSO, dimethyl sulfoxide; MAEA, arachidonyl-1*-hydroxy-2*-propylamide; HBA, N-(4-hydroxy-benzyl)arachidonamide; PMSF, phenylmethylsulfonyl fluoride.
40
AID
0003-2697/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AB 9600
/
6m18$$$241
05-29-96 02:11:26
aba
AP: Anal Bio
CHROMATOGRAPHIC DETERMINATION OF ANANDAMIDE AMIDASE
MATERIALS AND METHODS
Materials Arachidonic acid, linoleic acid, phenylmethylsulfonyl fluoride, and essentially fatty acid-free BSA were purchased from Sigma Chemical Co. (St. Louis, MO). Palmityl alcohol, 4-hydroxybenzonitrile, ethanolamine, and other amino alcohols were obtained from Aldrich Chemical Co. (Allentown, PA). HPLC grade acetonitrile was purchased from Fisher Scientific Co. (Pittsburgh, PA). Anandamide and its analogs were synthesized from appropriate fatty acid chlorides and amino alcohols as previous described (7). Palmitylsulfonyl fluoride (PSF) was essentially prepared from palmitylsulfonyl chloride and ammonium hydrofluoride. Preparation of Rat Brain Microsomal Fraction Adult Sprague–Dawley rat brains obtained from PelFreez Co. (Rogers, AK) were homogenized with a Potter–Elvehjem tissue grinder in 5 vol of ice-cold buffer containing 20 mM TrisrHCl and 1 mM EGTA (pH 7.4). The homogenate was centrifuged at 1000g for 10 min, and the supernatant was further centrifuged at 10,700g for 30 min and 105,000g for 60 min at 47C, sequentially. The pellet from the last centrifugation step (microsomal fraction) was resuspended in 40 ml of Tris buffer. Aliquots (1.5 ml) of the suspension were stored at 0807C until used. No loss of amidase activity was observed for at least 2 months. Protein concentration was determined using a modification of Lowry’s method (12).
HPLC was performed using a Beckman System Gold (Fullerton, CA) consisting of a 128 pump, a 166 uv detector, and a Rheodyne 7725i injector with a 20-ml loop. An IBM 466DX2/Si computer was interfaced to the system, and Beckman System Gold software was used for system control and data processing. Separations were carried out on a Beckman Ultrasphere ODS guard column (5 mm, 45 1 4.6 mm i.d.). The mobile phase was 8.5% aqueous phosphoric acid/acetonitrile (10:90, v/v) at a flow rate of 1 ml/min. Quantitation was based on the integration of peak areas at 204 nm. The amount of product formed was calculated from calibration curves of appropriate standards. Enzyme Assay For the standard assay of the amidase activity, 50 nmol of substrate (AEA) was incubated with 150 mg rat brain microsomal protein in a final volume of 500 ml buffer containing 50 mM TrisrHCl, 1 mM EDTA, and 0.1% BSA (pH 7.4) at 377C for 15 min with shaking. The reaction was terminated by pipetting 200 ml of the
AB 9600
suspension into a microcentrifuge tube containing 800 ml acetonitrile. The tube was vortexed, then centrifuged at room temperature for 4 min to remove the proteins. Twenty microliters of the supernatant was injected into the HPLC system for quantitation. An identical incubation was carried out in the absence of substrate to measure the amount of arachidonic acid existing naturally in the brain microsomal preparation. The rate of hydrolysis was calculated from the average of four experiments with results determined in duplicate. The amount of AA in the blank was subtracted from that in the sample. For the amidase inhibition assay, various amounts of test compound, 50 nmol of anandamide, and 150 mg rat brain microsomal protein were incubated as above. An identical incubation was carried out in the absence of inhibitor as a control. The same amount of protein was also incubated in the absence of both anandamide and inhibitor as a blank. The assays were performed in quadruplicate. The percentage of inhibition was calculated directly as % inhibition Å
(Ac 0 As ) 1 100, (Ac 0 Ab )
where As refers to the peak area of AA from a sample, Ac is the peak area of AA from a control, and Ab is the peak area from a blank. RESULTS
Fundamental Conditions for the Measurement of AEA Amidase Activity
HPLC Analysis
AID
41
/
6m18$$$242
05-29-96 02:11:26
AEA analogs and their acid products are highly lipophilic and strongly retained on the ODS column. Accordingly, a short guard column and a high percentage (90% v/v) of organic modifier in the mobile phase were used. The major advantage of the use of RP-HPLC is that the polar impurities from microsomes can be easily eluted from the column. Representative chromatograms showing the separation of AA and AEA are shown in Fig. 1. Sharp, symmetrical, and well-resolved AEA and AA peaks were obtained under the acidic conditions. The retention times for AEA and AA were 1.0 and 1.5 min, respectively. One unidentified peak (Rt 1.32 min) was recognized from microsomes, and no other metabolites were detected. The AA peak is very susceptible to the pH of the mobile phase. When the pH was adjusted to 2.5 or above, a broad tailing peak of AA was obtained, which led to a remarkable loss of sensitivity. A low pH is necessary to prevent ionization of AA. Under such chromatographic conditions, the retention times for linoleyl ethanolamide (LEA) and linoleic acid (LA) are 1.06 and 1.75 min, respectively. Preliminary experiments showed that AA occurs nat-
aba
AP: Anal Bio
42
LANG ET AL.
creasing enzyme activity in the presence of methanol and ethanol was detected at concentrations above 0.5%. In our enzymatic assays, DMSO was chosen as the solvent and the final DMSO concentrations were less than 1%. Stability of AEA and AA
FIG. 1. Chromatogram (A) of a standard solution containing anandamide (1.00 min) and arachidonic acid (1.50 min); (B) representative chromatogram of a sample of 100 mM anandamide incubated with 0.3 mg/ml microsomal proteins for 15 min. The unidentified peak at 1.32 min was from the microsomal preparation. Chromatographic conditions and methods for sample preparation are given under Materials and Methods.
urally in the microsomal preparations in a concentration range of 23–33 nmol/ml. This portion of AA interferes with the determination of that produced by enzymatic hydrolysis. In order to remove the interference, an identical experiment was performed in the presence of the same amount of enzyme incubated at 377C. The net amount of AA hydrolyzed by the enzyme could be obtained by subtracting the amount of AA in a blank from the total amount in the sample. The effect of organic solvents, e.g., dimethyl sulfoxide (DMSO), methanol, or ethanol, on the enzymatic activity was also studied because a solution of the test compounds had to be made in the solvents prior to dissolving them in aqueous medium. No significant effect of DMSO was found at concentrations below 5%, while slightly de-
AID
AB 9600
/
6m18$$$242
05-29-96 02:11:26
The stability of AEA in the acidic mobile phase was studied by the incubation of AEA in 8.5% aqueous phosphoric acid/acetonitrile (10:90) at ambient temperature. Twenty microliters of the solution was injected into the HPLC system after the first minute and at 5-min intervals for 1 h. The results showed that no significant change of AEA concentration and no formation of AA were detected during this time (data not shown). Thus, the hydrolysis of AEA during a chromatographic run is negligible. In addition, the stability of AEA in 50 mM Tris buffer (pH 7.4) at 377C was also tested, and a similar result was obtained. In contrast, AEA breakdown by AEA amidase is much faster than its natural hydrolysis in the biological buffer. The stability of AA in the same Tris buffer in the absence or presence of 0.3 mg/ml microsomal protein was also studied using the above method. There was a small, but significant fall in AA concentration with increasing incubation time. The data were fitted with both firstorder rate reaction and linear equations, and the same rate constants were obtained in both cases. The slopes (concentration decrease) were 0.12 and 0.22% per minute in the presence and absence of microsomal protein, respectively. The results indicate that AA is twice as stable in the presence of microsomal protein as in its absence. Less than 2% AA decomposed during the 15-min incubation used in the activity assay. Reproducibility The factors which affect reproducibility could be chromatographic, due to the precipitation of protein with acetonitrile and variations in centrifugation. The reproducibility of the chromatographic assay was evaluated by the injection of 20 ml of standard AA solution, at seven different concentrations, in quintuplicate. In Table 1, the reproducibility of the HPLC is represented by the coefficients of variation for AA, which lie between 0.5 and 3.4%. The overall reproducibility of the method is represented by the standard deviations (SDs) of the relative rates of hydrolysis for different AEA analogs and is shown in Table 2. The experiments were carried out for each compound in quadruplicate. The results indicate that the variation of SDs of the relative rates of hydrolysis for five compounds is between 1.2 and 4.2%, while the coefficients of variation for the relative rates of hydrolysis depend on the absolute rate values. Because a high percentage of acetonitrile (90%, v/v)
aba
AP: Anal Bio
43
CHROMATOGRAPHIC DETERMINATION OF ANANDAMIDE AMIDASE TABLE 1
TABLE 3
Reproducibility of HPLC Determination of AA in Standard Solutions (n Å 5)
Recovery (%) of AA Incubated in 0.3 mg/ml Microsomal Protein (n Å 4)
AA (nmol/ml)
Peak area (AU)
0.305 0.611 1.22 2.44 4.89 9.75 19.5
0.059 0.271 0.711 1.28 2.63 4.84 10.1
{ { { { { { {
Coefficient of variation (%)
0.002 0.009 0.015 0.006 0.028 0.035 0.17
3.4 3.3 2.1 0.5 1.1 0.7 1.7
in the mobile phase was used, most contaminants from microsomes are eluted together with the solvent front. The HPLC method is very reliable, and after 5000 runs and 6 months, no changes in retention time and peak resolution were observed.
AA added (nmol)
AA measured (nmol)
Recovery (%)
1.15 2.88 17.3 34.5
1.20 2.62 17.1 33.1
104 91 99 96
{ { { {
0.10 0.04 0.4 0.6
{ { { {
8.7 1.4 2.3 1.7
centrations. Good linearity for AA was obtained over a concentration range of 0.3 to 20 nmol/ml. The regression line was Y (peak area) Å 0.515X (concentration, nmol/ml) 0 0.007 with a correlation coefficient of 0.9996. The curve was used for the calibration of the concentrations in samples. The detection limits of 0.22 pmol for anandamide and 1.4 pmol for arachidonic acid were obtained at a signal to noise ratio of 4:1.
Recovery of AA The recovery of AA after precipitation of proteins with acetonitrile and centrifugation was investigated using a standard addition method. An exact amount of a standard AA solution was added to a microsomal suspension containing 150 mg protein and a known amount of AA in a final volume of 500 ml. The concentration of the standard AA solution prior to the addition was confirmed by HPLC analysis. The suspension was incubated, mixed with acetonitrile, and centrifuged as previously described under Materials and Methods. The supernatant was analyzed by HPLC. The results are listed in Table 3. The average recovery of AA obtained was 97.5% in the working concentration range.
AEA Amidase Activity Assay Optimization The enzymatic activity assays were conducted at pH 7.4, which is in the optimal pH range as previously reported (5). The effect of enzyme concentration on the hydrolysis of AEA was investigated as shown in Fig. 2, by altering the amount of rat brain microsomes added to the incubation mixture. AEA (100 mM) was incubated with the microsomal protein in a range of
Linearity and Detection Limits Linear relationships between peak area and concentration of AA and AEA were observed at different con-
TABLE 2
Anandamide Analog Specificity for Anandamide Amidase and CB1 Receptor (n Å 4) Rate of hydrolysis (pmol/min/mg) AEA (R)-MAEA (S)-MAEA HBA LEA
1300 29.7 297 300 982
{ { { { {
23.0 16.0 22.0 32.5 55.0
Relative rate of hydrolysis (%) 100 2.3 23 23 75.5
{ { { { {
1.8 1.2 1.7 2.5 4.2
Affinity for CB1 receptor (7) Ki (nM) 78 20 172 217
{ { { {
2 1.6 26 3
Note. AEA, anandamide; MAEA, methanandamide; HBA, N-(4hydroxybenzyl)-arachidonamide; LEA, linoleyl ethanolamide.
AID
AB 9600
/
6m18$$$242
05-29-96 02:11:26
FIG. 2. Dependence of initial rate of hydrolysis of AEA by anandamide amidase in rat brain microsomes on enzyme concentration. Assays were performed in 50 mM Tris buffer (pH 7.4) as described under Materials and Methods. The activity increases linearly with increasing protein concentration up to 0.8 mg/ml. Data shown are means { SD of four independent experiments.
aba
AP: Anal Bio
44
LANG ET AL.
FIG. 3. Progression of AEA hydrolysis by anandamide amidase in rat brain microsomes. AA formed from hydrolysis was determined by HPLC with uv detection at 204 nm. The chromatographic conditions are described under Materials and Methods. Data shown are means { SD of three separate measurements.
0.1 to 1.0 mg/ml for 15 min at 377C. The results obtained were linear up to at least 0.8 mg/ml of microsomal protein. The dependence of the enzymatic hydrolysis observed at 100 mM anandamide and 0.3 mg/ml microsomal protein on incubation time was also investigated. As shown in Fig. 3, the results indicate complete linearity over the entire range of 60 min. Fifteen minutes was selected as routine incubation time for convenience, and approximately 10% of AEA was hydrolyzed during this time. Enzymatic activity as a function of the substrate concentration was examined. The results of a representative experiment to determine the initial velocity of (S)-methanandamide (S-MAEA) hydrolysis at various substrate concentrations is shown in Fig. 4. Kinetic analysis was performed by fitting a hyperbola to the data using nonlinear least-squares regression and yielded an apparent Km of 8.6 { 1.3 mM and a Vmax of 362 { 16 pmol/min/mg of protein.
FIG. 4. Dependence of anandamide amidase activity on substrate concentration. Rectangular hyperbolic plot of initial rate of AA formed from AEA amidase hydrolysis vs AEA concentration. Apparent Km and Vmax were determined by fitting hyperbola to the data. The rate of hydrolysis is in units of nmols AA/min/mg of protein.
fluoride derivative, PSF, was synthesized and tested for its inhibitory activity for AEA amidase. For the inhibitor assays, AEA was incubated with microsomal protein in the presence of a series of concentrations of inhibitor at 377C. The results are shown in Fig. 5. The inhibitory ability was expressed as IC50 , the concentration of the inhibitor which results in 50% inhibition of
Applications The method was first applied to the determination of differences in substrate specificity. The test substrates include AEA, a pair of chiral anandamide congeners [(R)- and (S)-arachidonyl-1*-hydroxy-2*-propylamide (MAEA)], a phenolic derivative of anandamide [N-(4hydroxy-benzyl)-arachidonamide, HBA], and LEA. The substrate specificity is expressed as the relative rate of hydrolysis to that of AEA. Data shown in Table 2 were the means { SD for four separate experiments. Previous studies showed that phenylmethylsulfonyl fluoride (PMSF) inhibits the breakdown of AEA by the amidase in brain homogenates (4, 5). A new sulfonyl
AID
AB 9600
/
6m18$$$242
05-29-96 02:11:26
FIG. 5. Mean concentration–inhibition curves of inhibitors, PSF (s) and PMSF (l), for anandamide amidase activity in rat brain microsomes. Each value represents the mean { SD of inhibition from four independent experiments. Data analyzed by nonlinear leastsquares regression. The IC50 values are 50 nM for PSF and 10.5 mM for PMSF.
aba
AP: Anal Bio
CHROMATOGRAPHIC DETERMINATION OF ANANDAMIDE AMIDASE
the enzymatic reaction. Each IC50 value was calculated based on the curve fitting of inhibition percentage versus logarithm of inhibitor concentration. The IC50s were 50 nM for PSF and 10.5 mM for PMSF, indicating that the inhibitory activity of PSF is 210 times as high as that of PMSF. The details of AEA amidase inhibition by PSF will be published elsewhere.
45
hydrolysis of (R)-methanandamide is about 10-fold less than that of its (S)-isomer. Interestingly, when tested for their affinities for the cannabinoid receptor, this pair of enantiomers exhibited a stereoselectivity opposite that of their enzyme activity, with (R)-methanandamide having the higher affinity (7). ACKNOWLEDGMENTS
DISCUSSION
The previous methods (4–6, 9–11) for AEA amidase assays limit testing to radiolabeled substrates only. A method for the determination of AEA or its analogs by HPLC with a prefluorogenic derivatization described by Koga et al. is not suitable for the measurement of acid products (13). Our method using RP-HPLC with direct uv detection can generally be applied to uv active substrates and has sufficient sensitivity at a picomolar level for their acid products. Using this method, we overcame the disadvantage of using radioactivity. Also, HPLC provides much better resolution and reproducibility than do TLC and column chromatography (4– 6). The solvent extraction approach described by Omeir et al. (11) lacks specificity. Our method using solvent precipitation of protein and centrifugation offers acceptable recovery and does not require a complicated sample clean-up procedure. The use of a short guard column for fast separation and determination allows significant time and reagent saving. The simple, fast, and sensitive method is suitable not only for the assay of AEA amidase activity, but also for that of AEA synthase, with minor modifications. We applied the method to the determination of substrate specificity and inhibitor activity for AEA amidase. The results of hydrolysis of different substrates indicated that anandamide was hydrolyzed at the highest rate among the five substrates tested. We found that the chiral pair of AEA analogs, (R)- and (S)-methanandamide, in which a methyl group was introduced in the a-position of the ethanolamine head group showed great resistance to enzymatic hydrolysis. The rate of
AID
AB 9600
/
6m18$$$243
05-29-96 02:11:26
This work was supported by Grants DA3801 and DA7215 from the National Institute on Drug Abuse. We thank Dr. Pusheng Fan for his excellent technical assistance with the protein assays.
REFERENCES 1. Devane, W. A., Hanus, L., Breuer, A., Pertwee, R. G., Stevenson, L. A., Griffin, G., Gibson, D., Mandelbaum, A., Etinger, A., and Mechoulam, R. (1992) Science 258, 1946–1949. 2. Felder, C. C., Briley, E. M., Axelrod, J., Simpson, J. T., Mackie, K., and Devane, W. A. (1993) Proc. Natl. Acad. Sci. USA 90, 7656–7660. 3. Mackie, K., Devane, W. A., and Hille, B. (1993) Mol. Pharmacol. 44, 498–503. 4. Deutsch, D. G., and Chin, S. A. (1993) Biochem. Pharmacol. 46, 791–796. 5. Desarnaud, F., Cadas, H., and Piomelli, D. (1995) J. Biol. Chem. 270, 6030–6035. 6. Ueda, N., Kurahashi, Y., Yamamoto, S., and Tokunaga, T. (1995) J. Biol. Chem. 270, 23823–23827. 7. Abadji, V., Lin, S., Taha, G., Griffin, G., Stevenson, L. A., Pertwee, R. G., and Makriyannis, A. (1994) J. Med. Chem. 37, 1889– 1893. 8. Pertwee, R. G., Fernando, S. R., Griffin, G., Abadji, V., and Makriyannis, A. (1995) Eur. J. Pharmacol. 272, 73–78. 9. Koutek, B., Prestwich, G. D., Howlett, A. C., Chin, S. A., Salehani, D., Akhavan, N., and Deutsch, D. G. (1994) J. Biol. Chem. 269, 22937–22940. 10. Devane, W. A., and Axelrod, J. (1994) Proc. Natl. Acad. Sci. USA 91, 6698–6701. 11. Omeir, R. L., Chin, S., Hong, Y., Ahern, D. G., and Deutsch, D. G. (1995) Life Sci. 56, 1999–2005. 12. Markwell, M. A. K., Haas, S. M., Bieber, L. L., and Tolbert, N. E. (1978) Anal. Biochem. 87, 206–210. 13. Koga, D., Santa, T., Hagiwara, K., Imai, K., Takizawa, H., Nagano, T., Hirobe, M., Ogawa, M., Sato, T., and Inoue, K. (1994) Biomed. Chromatogr. 9, 56–57.
aba
AP: Anal Bio