Abstract
In vitro inhibition studies on drug-metabolizing enzyme activity are useful for understanding drug-drug interactions and for drug development. However, the profile of the inhibitory effects of carboxylesterase (CES) activity has not been fully investigated concerning species and tissue differences. In the present study, we measured the inhibitory effects of 15 drugs and 1 compound on CES activity using liver and jejunum microsomes and cytosol in human and rat. In addition, the inhibition constant (Ki values) and patterns were determined for the compounds exhibiting strong inhibition. Hydrolysis of imidapril and irinotecan hydrochloride (CPT-11) is catalyzed mainly by CES1 and CES2, respectively. In the inhibition study, imidaprilat formation from imidapril in human liver was strongly inhibited by nordihydroguaiaretic acid (NDGA) and procainamide. The inhibition profile and pattern were similar in human liver and rat liver. The compounds showing potent inhibition were similar between liver and jejunum. The Ki value of NDGA (Ki = 13.3 ± 1.5 μM) in human liver microsomes was 30-fold higher than that in rat liver microsomes (Ki = 0.4 ± 0.0 μM). On the other hand, 7-ethyl-10-hydroxycamptothecin (SN-38) formation from CPT-11 was not inhibited except by carvedilol, manidipine, and physostigmine. The Ki value of physostigmine (Ki = 0.3 ± 0.0 μM) in human jejunum cytosol was 10-fold lower than that in rat jejunum cytosol (Ki = 3.1 ± 0.4 μM) and was similar to that for manidipine. The present study clarified the species differences in CES inhibition. These results are useful for the development of prodrugs.
Carboxylesterase (CES) belongs to the α/β-hydrolase fold family and plays an important role in the hydrolysis of many esterified drugs such as anticancer and antihypertension drugs. There are species differences in CES isoforms between human and rat. In human, two major CES isoforms designated CES1 and CES2 have been characterized, and CES3 has recently been found in liver at considerably lower expression levels than those for the other CES isoforms (Sanghani et al., 2004). The mRNA expression levels of human CES1 are higher in liver than those in small intestine (Satoh et al., 2002), whereas human CES2 is abundant in small intestine (Schwer et al., 1997). On the other hand, in rat, there are several isoforms in CES1 and CES2 families. In the CES1 family, ES-10 (hydrolase A), ES-4 (hydrolase B/C), and ES-3 are mainly expressed in liver, and the mRNA expression levels of ES-10 are the highest of those of the three isoforms (Linke et al., 2005). ES-2 has been identified as CES (Murakami et al., 1993) and can hardly be detected in liver (Sanghani et al., 2002). Three rat CES2 mRNAs D50580, AB010635, and AY034877 have been found in various tissues such as liver, small intestine, stomach, and kidney (Sanghani et al., 2002).
The coadministration of drugs can affect their efficacy. Alterations in drug metabolism such as oxidation, reduction, and hydrolysis are often important causes of drug interactions. In particular, because inhibition of drug-metabolizing enzymes is recognized as a prevalent factor, it is necessary to understand and clarify the inhibitory effects. The inhibitory effects of some compounds on CES activity have been previously reported for isomalathion (Buratti and Testai, 2005) and trifluoromethylketone-containing compounds (Wadkins et al., 2007). An anticancer drug, tamoxifen, has been found to be an inhibitor of human CES1 (Fleming et al., 2005) and rat ES-10 (Mésange et al., 2002). However, species differences in the inhibitory effects on CES activity have never been investigated comprehensively. In preclinical development of prodrugs catalyzed by esterases, information on differences in the inhibitory properties between rat and human is useful. In the present study, imidapril, an inhibitor of angiotensin-converting enzyme, and irinotecan hydrochloride (CPT-11) were used as typical substrates for CES1 and CES2, respectively (Takai et al., 1997; Humerickhouse et al., 2000; Sanghani et al., 2004). We investigated the species differences in the inhibitory effects of 15 drugs and 1 compound for two representative CES activities.
Materials and Methods
Materials. Imidapril hydrochloride and imidaprilat were kindly supplied by Mitsubishi Tanabe Pharma Corporation (Osaka, Japan). Delapril was kindly provided by Takeda Pharmaceutical Company (Osaka, Japan). Capecitabine was purchased from The United States Pharmacopeial Convention (Rockville, MD). Carvedilol was obtained from LKT Laboratories (St. Paul, MN). CPT-11, 7-ethyl-10-hydroxycamptothecin (SN-38), docetaxel trihydrate, temocapril hydrochloride, and temocaprilat were purchased from Toronto Research Chemicals (North York, ON, Canada). Acemetacin, dexamethasone, nordihydroguaiaretic acid (NDGA), procainamide hydrochloride, and proglumide were purchased from Sigma-Aldrich (St. Louis, MO). Camptothecin, ciprofloxacin hydrochloride monohydrate, manidipine, nifedipine, and physostigmine sulfate were obtained from Wako Pure Chemicals (Osaka, Japan). Pooled human liver microsomes (HLM) and cytosol (HLC) were purchased from BD Gentest (Woburn, MA). Pooled human jejunum microsomes (HJM) and cytosol (HJC) were obtained from Tissue Transformation Technologies (Edison, NJ). All other chemicals and solvents were of analytical or the highest grade commercially available.
Preparation of Microsomes and Cytosol from Rat Liver or Jejunum. Male Wistar rats, 7 weeks old, were obtained from Japan SLC (Shizuoka, Japan). Pooled microsomes and cytosol from five rat livers (RLM and RLC) and jejunums (RJM and RJC) were prepared according to the method of Emoto et al. (2001).
Imidaprilat Formation. Imidaprilat formation from imidapril was determined according to the method of Takahashi et al. (2008) with slight modifications. A typical incubation mixture (200-μl total volume) contained the enzyme source, 100 mM Tris-HCl buffer (pH 7.4), and inhibitors. After a 2-min preincubation at 37°C, the reaction was initiated by the addition of imidapril, and then the mixture was incubated at 37°C for 30 min except for RLM and RLC (10 min). The reaction was terminated by adding 100 μlof ice-cold acetonitrile. After centrifugation at 9000g for 5 min, 10 μlofthe supernatant was subjected to a liquid chromatography-tandem mass spectrometry system. The formed imidaprilat was quantified by the liquid chromatography-tandem mass spectrometry peak area of an authentic standard. The linearity of the standard curve of imidaprilat was confirmed (r = 0.99), and the imidaprilat in the incubation mixture was determined within the range. In the preliminary study, bis(p-nitrophenyl)phosphate, a CES specific inhibitor, (300 μM) inhibited the imidaprilat formation potently with all enzyme sources at 150 μM imidapril.
SN-38 Formation. SN-38 formation from CPT-11 was determined according to the method of Tabata et al. (2004) with slight modifications. A typical incubation mixture (200-μl total volume) contained the enzyme source, 100 mM potassium phosphate buffer (pH 7.4), and inhibitors. CPT-11 was dissolved in dimethyl sulfoxide (DMSO). The final concentration of DMSO in the reaction mixture was <1.0%. After a 2-min preincubation at 37°C, the reaction was initiated by the addition of CPT-11, and then the mixture was incubated at 37°C for 3 min except for RJM and RJC (2 min). The reaction was terminated by adding 200 μl of ice-cold acetonitrile, and 10 μl of 1 M HCl was then added. Camptothecin (20 pmol) was added as an internal standard. After centrifugation at 6500g for 5 min, 50 μl of the supernatant was subjected to high-performance liquid chromatography on an Inertsil ODS-3 analytical column (4.6 × 250 mm; GL Sciences, Inc., Tokyo, Japan). The eluent was fluorometrically monitored at an excitation of 380 nm and emission of 556 nm with a noise-base clean Uni-3 (Union, Gunma, Japan). The column temperature was 35°C, and the flow rate was 0.8 ml/min. The mobile phase consisted of 30% acetonitrile and 70% of 10 mM KH2PO4 containing 30 mM 1-heptanesulfonic acid sodium salt. The retention times of SN-38 and camptothecin were 13.3 and 16.3 min, respectively. The SN-38 formed was quantified by the high-performance liquid chromatography peak height of an authentic standard. The linearity of the standard curve of SN-38 was confirmed (r = 0.99), and the SN-38 in the incubation mixture was determined within the range. In the preliminary study, bis(p-nitrophenyl)phosphate (10 μM) inhibited the SN-38 formation potently with all enzyme sources at 5 μM CPT-11.
Inhibition Analysis of CES Activities. Inhibitory effects on imidaprilat and SN-38 formation were investigated using 15 drugs and 1 compound. Acemetacin, capecitabine, carvedilol, CPT-11, dexamethasone, docetaxel, manidipine, NDGA, nifedipine, temocapril, and temocaprilat were dissolved in DMSO. Ciprofloxacin, delapril, imidapril, physostigmine, procainamide, and proglumide were dissolved in distilled water. Capecitabine, delapril, and temocapril are CES substrates, whereas the other drugs and one compound are not. These compounds were added to the incubation mixtures described above to investigate their inhibitory effects on the imidaprilat and SN-38 formations. The final concentration of DMSO in the incubation mixture was <1% except for manidipine (2%) in the imidaprilat formation. All data were analyzed using the average of duplicate determinations.
For screening of the inhibitory effects, the imidaprilat formation at 150 μM imidapril was examined in the presence of 15 drugs and 1 compound (300 μM) except manidipine (50 μM). The SN-38 formation at 5 μM CPT-11 was determined in the presence of 15 drugs and 1 compound (10 μM).
For determination of the Ki (inhibition constant) values, the concentrations of imidapril ranged from 100 to 500 μM for HLM and HLC, from 25 to 150 μM for RLM, from 15 to 90 μM for RLC, from 20 to 80 μM for RJM, and from 25 to 100 μM for RJC, respectively. The concentrations of the inhibitors ranged as follows: for carvedilol, 1 to 6 μM for RLM, 1 to 6 μM for RLC, 4 to 24 μM for RJM, and 5 to 30 μM for RJC; for NDGA, 10 to 40 μM for HLM, 2 to 7 μM for HLC, 0.2 to 2 μM for RLM, 0.2 to 1.5 μM for RLC, 3 to 15 μM for RJM, and 2 to 12 μM for RJC; and for procainamide, 25 to 150 μM for HLM and 25 to 100 μM for HLC. The protein concentrations of HLM, HLC, RLM, RLC, RJM, and RJC were 0.2, 1.0, 0.01, 0.05, 0.2, and 0.5 mg/ml, respectively. In the preliminary study, the rate of imidaprilat formation was linear with respect to all protein concentrations and incubation times.
For determination of the Ki values, the concentrations of CPT-11 ranged from 2.5 to 15 μM for all enzyme sources and the concentrations of inhibitors ranged as follows: for carvedilol, 2 to 8 μM for HLM and 1 to 12 μM for HLC; for manidipine, 0.1 to 0.5 μM for HJC, 0.5 to 5 μM for RJM, and 0.5 to 4 μM for RJC; and for physostigmine, 0.1 to 0.6 μM for HLM, 0.2 to 5 μM for HLC, 1 to 6 μM for HJM, 0.2 to 1 μM for HJC, 1 to 8 μM for RJM, and 2 to 10 μM for RJC. The protein concentrations of microsomes and cytosol were 0.2 and 1.0 mg/ml except for RJM and RJC (0.1 and 0.5 mg/ml), respectively. In the preliminary study, the rate of SN-38 formation was linear with respect to all protein concentrations and incubation times. The Ki values and inhibition types were determined by fitting the kinetic data to a competitive, noncompetitive, uncompetitive, or mixed inhibition model by nonlinear regression analysis using GraphPad Prism 5 (GraphPad Software Inc., San Diego, CA).
Results
Inhibitory Effects of 15 Drugs and 1 Compound on Imidaprilat Formation. The inhibitory effects on the imidaprilat formation were investigated using 15 drugs and 1 compound (Fig. 1). Imidaprilat formation in both HLM and HLC was moderately inhibited by acemetacin, carvedilol (only HLC), nifedipine, and procainamide (20–50% of control) and strongly inhibited by NDGA (<20% of control). On the other hand, all compounds except ciprofloxacin, dexamethasone (only RLC), and proglumide inhibited the imidaprilat formation by more than 50% in both RLM and RLC. In particular, delapril, temocapril, and temocaprilat, which are structurally similar to imidapril, showed stronger inhibitions in RLM and RLC than in HLM and HLC. The compounds exhibiting strong inhibition in RJM and RJC were similar to those in RLM and RLC. Imidaprilat formation in HJM and HJC was not detected at 200 μM imidapril; therefore, inhibition studies were not performed.
Inhibitory Effects of 15 Drugs and 1 Compound on SN-38 Formation. Fifteen drugs and one compound were screened for their inhibitory effects on SN-38 formation (Fig. 2). Screening of the inhibitory effects in HJM and HJC were performed only for six drugs and one compound because the lots of HJM and HJC available were limited. In both HLM and HLC, carvedilol, physostigmine, and manidipine (only HLC) inhibited SN-38 formation moderately (20–50% of control). In both RLM and RLC, SN-38 formation was not inhibited except by manidipine and NDGA, exhibiting weak inhibition (50–70% of control) only in RLC. These formations in both RJM and RJC were inhibited moderately by manidipine and physostigmine. SN-38 formation in HJC was inhibited moderately by carvedilol (20–50% of control) and inhibited strongly by manidipine and physostigmine, whereas this formation in HJM was inhibited strongly only by physostigmine.
Inhibition Constant and Inhibitory Patterns of Imidaprilat Formation.Ki values and inhibition patterns of the compounds showing strong inhibition for the imidaprilat formation were determined (Table 1), and representative Lineweaver-Burk plots are shown in Fig. 3, A–C. The Ki value of NDGA in HLM was 30-fold higher than that in RLM, suggesting species differences. The Ki value of NDGA in RLM was also much lower than that in RJM. Likewise, NDGA in RLC showed a lower Ki value than that in RJC. The inhibitory potency of procainamide in HLM was similar to that in HLC. The Ki value of carvedilol in RJM was approximately 5-fold higher than that in RLM. Compared with microsomes and cytosol from rat liver or jejunum, the Ki values of carvedilol were approximately 2-fold different.
Inhibition Constant and Inhibitory Patterns of SN-38 Formation. The Ki values and inhibition patterns for SN-38 formation were determined as shown in Table 2 and representative Lineweaver-Burk plots are shown in Fig. 3, D–F. The Ki values in RLM and RLC were not determined because the activity was not inhibited potently by any drugs or compound. The Ki value of carvedilol in HLM was lower than that in HLC. The Ki value of physostigmine in HJM was similar to that in RJM. On the other hand, the Ki value of physostigmine in HJC was 10-fold lower than that in RJC or HJM. The inhibitory potency of physostigmine in HLM was stronger than that in HJM. The Ki value of manidipine in HJC was lower than that in RJM or RJC.
Discussion
In multiple drug therapy, drug interactions are important issues that must be taken into consideration. It is possible that coadministration of several drugs can change the efficacy of a drug due to the inhibition of drug-metabolizing enzymes. Investigations of inhibitory effects for CES activity provide useful information for the prediction of in vivo drug interactions and for drug development. In the present study, we investigated the inhibitory effects on hydrolysis of two prodrugs in human and rat to clarify the species differences in CES inhibition. Imidapril and CPT-11 were used as representative CES1 and CES2 substrates.
In imidaprilat formation from imidapril by CES1, delapril and temocapril were inhibited weakly (>50% of control) in human liver, whereas they showed more than 90% inhibition in rat liver (Fig. 1). Human CES1 was shown to be less efficient at catalysis of bulky substrates than rabbit CES because of the size-limited access of substrates to the active site (Wadkins et al., 2001). Because delapril and temocapril are substrates of human CES1 (Takai et al., 1997) and are more bulky than imidapril, size differences of the active site between human and rat may have contributed to the present results.
Dexamethasone and docetaxel, which are not CES substrates, inhibited the imidaprilat formation weakly in human liver (Fig. 1). The two drugs exhibited weak inhibitions for the hydrolysis of 4-methylumbelliferyl acetate catalyzed by both CES1A1 and CES2 (IC50 >0.7 mM) (Quinney et al., 2005), consistent with our results. As shown in Table 1, imidaprilat formation in HLM was inhibited competitively by procainamide. Bailey and Briggs (2003) suggested that procainamide inhibits human CES1. Procainamide is also known as a choline binding pocket-specific inhibitor (Jaganathan and Boopathy, 1998) and has been reported to inhibit human butyrylcholinesterase competitively (Ki = 9 μM) (Rush et al., 1981). Because the amino acid sequences at the active site were highly conserved among esterase B, which is inhibited by organophosphorus compounds such as CES and butyrylcholinesterase (Satoh and Hosokawa, 1995), it is reasonable to assume that procainamide inhibits CES1 activity. Imidaprilat formations in rat liver were inhibited more strongly by NDGA than those in human liver (Table 1). NDGA is known as a potent CES inhibitor (Schegg and Welch, 1984). These inhibition patterns showed uncompetitive inhibition except for RJM, suggesting that NDGA could bind to a site different from the active site. In fact, human CES1 has ligand binding sites other than the active site, and each site is called a side door and a Z-site (Bencharit et al., 2003b). Rat CES isoforms may contain other sites such as human CES1. The inhibitory potencies of carvedilol were stronger in rat liver and jejunum than those in human liver (Table 1). Although the structures of rat CES isoforms have not been revealed, structural differences between human and rat CES may be involved in the species differences of the inhibition potency.
The inhibitory effects in both microsomes and cytosol showed similar tendencies in imidaprilat formations in human and rat in the present study. The activities in microsomes were more than 2-fold higher than those in cytosol. However, because the expression levels of cytosolic proteins are approximately 5-fold higher than those of microsomal proteins, the contribution of cytosolic CES is as important as that of microsomal CES (Tabata et al., 2004). Thus, CES inhibition in the cytosol as well as in microsomes should be taken into consideration.
SN-38 formations in jejunum were inhibited more strongly by manidipine than those in liver in the present study. Manidipine is similar in structure to nifedipine and has larger alcohol groups that are hydrolyzed well by CES2. Because nifedipine did not inhibit SN-38 formation in HJC, manidipine could be a potent CES2 inhibitor in human jejunum. Carvedilol also showed inhibitions for the SN-38 formations in both human liver and jejunum except HJM, whereas there were no inhibitions in rat liver and jejunum, indicating species differences in CES2 activity. As shown in Figs. 1 and 2, the SN-38 formations in all enzyme sources were not inhibited by procainamide, whereas the imidaprilat formations were inhibited strongly. The major isoforms involved in the hydrolysis of CPT-11 and imidapril are CES2 and CES1, respectively (Schwer et al., 1997; Takai et al., 1997). Human CES1 is also expressed in liver at much higher levels than in small intestine (Imai, 2006), whereas CES2 is abundant in small intestine (Schwer et al., 1997). Thus, procainamide exhibited more potent inhibitory effects for CES1 than for CES2.
Because there were no drugs and compounds that inhibited the SN-38 formations potently in rat liver, the Ki values in RLM and RLC were not determined. On the other hand, the SN-38 formations in RJM and RJC were inhibited competitively by manidipine with low Ki values, suggesting that tissue differences would exist in the inhibition of CPT-11 hydrolase activity. Satoh et al. (1994) demonstrated that ES-4 and ES-10 purified from RLM, which both belong to CES1 family, have the ability to hydrolyze CPT-11 more effectively than human CES1. By Northern blot analysis, both CES1 and CES2 were expressed in rat liver (Sanghani et al., 2002), whereas CES2 isoforms were mainly expressed in Wistar rat jejunum, but the mRNA expression levels of the CES1 isoforms were much lower than those of CES2 (Masaki et al., 2007). From these studies, it could be speculated that CES2 isoforms are responsible for CPT-11 hydrolysis; thus, the potent inhibitory effects by manidipine may be the result of CES2 inhibition in rat jejunum. Although the clinical serum concentration of manidipine is on the order of nanomolar concentrations, intestinal mucosa may be exposed to high concentrations of a given drug. Therefore, we have to keep the CES2 inhibition in mind in clinical practice. Manidipine also exhibited a potent competitive inhibition in HJC, whereas imidaprilat formations in human liver were not inhibited. The entrance to the active site of human CES1 is smaller than that for CES2 (Wadkins et al., 2001), leading to the difference in inhibition. Although the crystal structure of human CES2 is not obvious, human CES2 contains N-glycosylation sites at two positions, whereas CES1 has one N-glycosylation site (Bencharit et al., 2003a).
In the present study, the Ki values of physostigmine, a specific cholinesterase inhibitor, were approximately 10-fold lower in HLM (Ki = 0.2 ± 0.0 μM) and HJC (Ki = 0.3 ± 0.0 μM) than the other enzyme sources. These Ki values were close to the previous result showing that physostigmine inhibited the hydrolysis of 4-methylum-belliferyl acetate by purified human CES2 (Ki = 0.10 ± 0.01 μM) (Pindel et al., 1997). The CES1 and CES2 family share 40 to 50% amino acid sequence identity, but the substrate specificities of CES1 and CES2 in human are different; CES1 hydrolyzes compounds with larger acyl groups and CES2 preferentially hydrolyzes large alcohol groups (Satoh et al., 2002; Imai, 2006).
The present comprehensive study clarified that species differences existed in the inhibition of imidapril and CPT-11 hydrolysis and should provide useful information on the differences in CES activity in human and rat.
Acknowledgments
We acknowledge Mitsubishi Tanabe Pharma Corporation for kindly providing imidapril and imidaprilat and Takeda Pharmaceutical Company for kindly supplying delapril. We thank Brent Bell for reviewing the manuscript.
Footnotes
-
S.T. and M.K. contributed equally to this work.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.108.024331.
-
ABBREVIATIONS: CES, carboxylesterase; CPT-11, irinotecan hydrochloride; SN-38, 7-ethyl-10-hydroxycamptothecin; NDGA, nordihydroguaiaretic acid; HLM, human liver microsomes; HLC, human liver cytosol; HJM, human jejunum microsomes; HJC, human jejunum cytosol; RLM, rat liver microsomes; RLC, rat liver cytosol; RJM, rat jejunum microsomes; RJC, rat jejunum cytosol; DMSO, dimethyl sulfoxide.
-
↵1 Current affiliation: Faculty of Pharmacy, Meijo University, Tempaku-ku, Nagoya, Japan.
- Received September 2, 2008.
- Accepted February 12, 2009.
- The American Society for Pharmacology and Experimental Therapeutics