![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
School of Pharmacy, Texas Tech University Health Sciences Center, Amarillo, Texas
(Received January 27, 2005; Accepted March 16, 2005)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionConclusionsReferences |
|---|
30 min), concentration dependently, and enzyme-selectively in the rat liver. Interestingly, among all the studied P450 enzymes, only CYP2D1 was not affected by NO donors. However, these studies were conducted using only a single concentration of the substrates, thus lacking information about the possible simultaneous changes in both maximum velocity (Vmax) and affinity (Km) of the enzymes. In the present study, we systematically evaluated the effects of NO on the enzyme kinetic parameters of marker substrates for a range of P450 enzymes, including 2D1. Livers were perfused (1 h) in the absence (control) or presence of two NO donors with different mechanisms of NO release. At the end of the perfusion, microsomes were prepared and used for kinetic analysis. Except for 2D1, NO reduced the Vmax of all the model reactions studied, although to a varying degree. However, the effects of NO donors on Km were more diverse. Whereas the Km values for testosterone 6ß-hydroxylation (3A2) and 16
-hydroxylation (2C11) significantly decreased, the values for chlorzoxazone 6-hydroxylation (2E1), dextromethorphan N-demethylation (3A2), and high affinity ethoxyresorufin O-dealkylation (1A1/2) significantly increased in the presence of NO donors. Furthermore, the Km values for the high-affinity component of dextromethorphan O-demethylation and benzyloxyresorufin O-dealkylation remained unchanged. These results indicate that NO can potentially change both the Vmax and Km of various substrates selectively and confirm our previous findings that the activity of CYP2D1 is not affected by NO donors.
, 2001; Renton, 2004; Riddick et al., 2004). Despite wide acceptance that the stimulation of the immune system can inhibit drug metabolism, the exact mechanisms responsible for this down-regulation are still not well understood. Large quantities of cytokines, produced during inflammation, are believed to be responsible, at least in part, for mediating suppression of P450 enzymes at the transcriptional level (Morgan, 1997). Indeed, several reports (Ghezzi et al., 1986; Warren et al., 1999) have shown that the administration of inflammatory cytokines depresses P450 activities and contents both in vivo and in vitro.
In addition to cytokines, inflammatory states induce the expression of inducible nitric-oxide synthase and result in excessive production of nitric oxide (NO) in both hepatocytes and Kupffer cells (Nathan, 1992). Although some reports (Sewer et al., 1998; Sewer and Morgan, 1998; Takemura et al., 1999) suggest that P450 enzyme down-regulation occurs independently of NO, others (Stadler et al., 1994; Khatsenko and Kikkawa, 1997; Khatsenko, 1998) have shown that NO may also play a role in this event. Besides this indirect evidence for an NO role in inflammation-induced P450 depression, attempts have also been made at delineating the effects of NO directly, using NO donors in the microsomes or purified enzyme systems (Khatsenko et al., 1993; Wink et al., 1993; Minamiyama et al., 1997). Supporting the above findings, very recently (Vuppugalla and Mehvar, 2004a), we showed the inhibitory effects of NO on P450 to be rapid, concentration-dependent, and enzyme-selective in an isolated perfused rat liver (IPRL) model. Furthermore, we also conducted time course studies to assess the mechanisms responsible for the NO-mediated P450 inhibition and demonstrated that the short-term inhibitory effects of NO are time-dependent, consisting of both reversible and irreversible components (Vuppugalla and Mehvar, 2004b).
One of the novel findings of our IPRL studies (Vuppugalla and Mehvar, 2004a,b) was the apparent lack of effect of NO donors on CYP2D1 enzyme, along with different degrees and time courses of inhibition or reversal of activities for other P450 enzymes. These studies were carried out only with a single substrate concentration. Therefore, one may argue that the observed enzyme-selective effects of NO may have been due to different positions of substrate concentrations along their respective Michaelis-Menten kinetic curves. This is especially true if NO alters both Vmax and Km of P450 enzymes. Although the interaction of NO with P450 heme (Kim et al., 1995) suggests a reduction in Vmax, the possibility of alterations in Km cannot be ruled out based on the available data. Additionally, the significance of NO-induced inhibition of P450 enzymes to the overall pharmacokinetics of a drug can be better appraised if the effects of NO on the P450 enzyme kinetic parameters (Km, Vmax, and CLint) are known. However, to the best of our knowledge, this information is not available in the literature. Thus, in the present study, we investigated the direct effects of NO donors sodium nitroprusside (SNP) and isosorbide dinitrate (ISDN) on the kinetic parameters of various P450 probe substrates using an IPRL model.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
-hydroxytestosterone (16-HT) were purchased from Steraloids (Newport, RI). All the other reagents were of analytical grade and obtained from commercial sources. Animals. Male Sprague-Dawley rats weighing between 230 and 320 g were purchased from Charles River Laboratories, Inc. (Wilmington, MA). Animals were housed in a temperature- and humidity-controlled room under a 12-h light/dark cycle, with free access to food and water. The Texas Tech Health Sciences Center Animal Care and Use Committee approved the study protocol, and all the animals received humane care in compliance with guidelines set by the National Institutes of Health (publication 85-23, revised 1985, Bethesda, MD).
Liver Perfusion and Microsomal Preparation. The procedures for isolation of the livers, their ex vivo perfusion with NO donors, and subsequent preparation of microsomes have been reported in detail in our previous study (Vuppugalla and Mehvar, 2004a). Briefly, after isolation from untreated animals, livers were perfused for 1 h either in the absence (control) or presence of 200 µM SNP or ISDN (n = 4/group) in a single-pass mode. At the end of the perfusion, the livers were blotted dry, weighed, and subjected to preparation of microsomes using well established ultracentrifugation methods. Protein concentrations were measured by the Bradford (1976) method.
Measurement of P450 Enzyme Activities. In preliminary studies, all the assays were checked for linearity of the rate of metabolism with respect to the incubation time and microsomal protein concentrations. Additionally, substrate consumption was determined at the lowest and highest substrate concentrations included in the study.
The activities of CYP3A2 and 2C11 were analyzed based on the formation of 6ß- and 16-hydroxytestosterone, respectively, from testosterone, based on our modification (Vuppugalla and Mehvar, 2004a) of a previously published method (Purdon and Lehman-McKeeman, 1997). The microsomal protein concentration and incubation time were 200 µg/ml and 10 min, respectively. The substrate (testosterone) concentrations were 10, 20, 50, 100, 200, and 300 µM.
The activity of CYP2E1 was measured based on the formation of 6-HCLZ from CLZ. A detailed description of the incubation conditions and high-performance liquid chromatographic analysis of the drug and the metabolite has been reported by us before (Vuppugalla and Mehvar, 2004a). The microsomal protein concentration and incubation time were 800 µg/ml and 15 min, respectively. The substrate (CLZ) concentrations were 5, 10, 25, 50, 100, 250, and 500 µM.
The activities of CYP2D1 and 3A2 were measured based on the formation of DT and MXM, respectively, from DM. The details of the procedure have been described by us before (Vuppugalla and Mehvar, 2004a). The microsomal protein concentration and incubation time were 500 µg/ml and 5 min, respectively. The substrate (DM) concentrations were 0.5, 2.5, 5, 10, 50, 100, 200, and 500 µM.
The activities of CYP2B1/2 and 1A1/2 were assessed based on the formation of resorufin from BR (Burke et al., 1985) and ER (Rutten et al., 1992), respectively. Rate of formation of resorufin was monitored fluorometrically based on a kinetic analysis over 150 to 300 s of incubation, and the microsomal protein concentration was 200 µg/ml. The substrate concentrations were 20, 40, 80, 160, 320, 1250, 2500, and 5000 nM for BR and 10, 20, 40, 80, 320, 625, 1250, and 2500 nM for ER.
Estimation of Microsomal Protein Binding. Binding of various substrates to microsomes was determined by an ultrafiltration technique. Samples contained the same concentration of microsomes and the lowest and highest substrate concentrations used in the kinetic study, described above. After incubation at 37°C for the indicated times, an aliquot was transferred into 10,000-molecular weight cutoff filters (Centricon 10; Millipore Corporation, Billerica, MA) and centrifuged at 2000g for 20 min. Subsequently, the substrate concentrations were determined in the filtrates and the initial samples.
Analysis of Nitrate/Nitrite (NOx) in the Perfusate. The NOx concentration in the outlet perfusate was measured using the Griess reagent (Active Motif Inc., Carlsbad, CA) at the end of the 1-h perfusion with SNP or ISDN.
Data Analysis. The kinetic parameters were estimated based on nonlinear regression analysis using the WinNonlin program (Pharsight, Mountain View, CA). Biotransformation of all the substrates was modeled using a subset of the following general equation:
 | (1) |
 | (2) |
Model selection was based on the examination of Eadie-Hofstee transformed plots (rate versus rate/[s]), which suggested a subset of eq. 1 (e.g., single-versus dual-enzyme models or presence or absence of sigmoidicity). Further model selection from the possible subsets was based on criteria that included Akaike's information criterion, visual inspection of the fit, the random distribution of residuals, and the standard error of the parameter estimates. All fittings were carried out with a weight of 1/v.
In the absence of sigmoidicity, the intrinsic clearance (CLint) value for each enzyme was calculated by dividing the corresponding enzyme Vmax by its Km. The total intrinsic clearance (CLtotal) was then determined by the summation of the individual CLint values:
 | (3) |
), was used instead of CLint or CLtotal. In protein binding studies, the free fraction of substrate was estimated by dividing the substrate concentration in the filtrate to that in the unfiltered suspension. All the statistical comparisons were conducted using analysis of variance with subsequent Fisher's test for comparison of three groups or using an unpaired t test for comparison of two groups. All tests were conducted at a significance level of 0.05. Data are presented as mean ± S.E.M.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
0.85) under the conditions used in this study.
|
|
Similar to the formation of 6ß-hydroxytestosterone (3A2), the formation of 16-hydroxytestosterone from testosterone by 2C11 was best described by a simple, single-enzyme model without any sigmoidicity (Fig. 2; Table 1). Similarly, both SNP and ISDN treatments resulted in a decline in the Km (
80%) and Vmax (60%) values (Table 1). In contrast to data for 3A2, however, the CLint increased after treatment with ISDN (P < 0.05) (Table 1).
|
|
|
|
70%. Consequently, the CLtotal value was decreased by a factor of 3 by ISDN. The Vmax(2) and Km(2) values for the SNP group could not be compared with those of the control group because of the use of different models. However, the CLmax value for the SNP group was significantly (P < 0.05, t test) lower than the CLtotal for the control group (Table 1).
|
3-fold increase in Km(1), resulting in a proportional decrease in CLint(1) (Table 1). On the other hand, both Km(2) and Vmax(2) significantly decreased, whereas CLint(2) remained unaffected after ISDN treatment (Table 1). Nevertheless, ISDN caused a significant reduction in the CLtotal value for the formation of resorufin from ER (Table 1).
|
The NOx concentrations in the outlet perfusate at the end of the 1-h perfusion with 200 µM SNP and ISDN were 10.2 ± 0.4 and 30.2 ± 2.5 µM, respectively.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
; Wink et al., 1993; Minamiyama et al., 1997), hepatocytes (Stadler et al., 1994), and the intact liver (Vuppugalla and Mehvar, 2004a,b). Several mechanisms including the reaction of NO with P450 heme, leading to its loss and degradation, and with apoprotein thiols were held responsible for the direct inhibitory effects of NO (Minamiyama et al., 1997; Vuppugalla and Mehvar, 2004b). Whereas the former mechanism is expected to reduce the Vmax of the enzyme, the latter may affect the binding of the substrates to the enzyme, thus changing the Km value. Indeed, our present findings show, for the first time, that NO may alter both the Km and Vmax values of various P450 enzymes selectively (Table 1).
Whereas NO generally caused a reduction in Vmax, it resulted in a decrease, increase, or no change in the Km values of different enzymes (Table 1). The enzyme-selective magnitude of decreases in Vmax may be due to differential accessibility of heme and/or cysteine thiolate of various enzymes (Gergel et al., 1997). NO can also form reversible nitrosyl-heme complexes in the intact liver, which can potentially decrease the Vmax of different enzymes selectively. However, as we demonstrated recently (Vuppugalla and Mehvar, 2004b), the ferric-NO complex could not be detected in our experimental design, which requires preparation of microsomes. As for Km, the reaction of NO with thiol groups of amino acid residues observed by us (Vuppugalla and Mehvar, 2004b) and others (Minamiyama et al., 1997) has been proposed as one of the mechanisms for the decrease in the enzyme activities by NO donors. Although not shown before experimentally, this suggests a decrease in the binding of the substrate to the enzyme, and thus an increase in Km. In agreement with this postulate, we also observed enzyme-selective increases in the Km values for some enzymes (Table 1). Surprisingly, however, we also noticed a decrease in Km for some reactions such as the formation of 6ß-hydroxytestosterone (3A2) and 16-hydroxytestosterone (2C11) from testosterone and the low-affinity component for the formation of resorufin from ER (1A1/2) (Table 1). This finding suggests that the previously reported (Minamiyama et al., 1997; Vuppugalla and Mehvar, 2004b) reaction of NO with the thiol groups of P450 apoprotein may facilitate the binding of substrates to some enzymes, thus expediting their metabolism. However, in all cases of the observed decrease in Km, the Vmax values were also decreased, such that the CLint values remained mostly unchanged in the presence of NO donors (Table 1).
In single-substrate concentration studies, we have recently (Vuppugalla and Mehvar, 2004b) shown that the activities of all the studied P450 enzymes, except for 2D1, were reduced by pretreatment of IPRLs with NO donors. The lack of effect of NO on 2D1 in these studies may have been due to a simultaneous change in both the Vmax and Km values not manifested as a change in activity at the particular substrate concentration selected. In agreement with earlier studies (Kerry et al., 1993; Witherow and Houston, 1999), the kinetics of DM O-demethylation in our study was best described by a two-site Michaelis-Menten model consisting of a high-affinity, low-capacity and a low-affinity, high-capacity site. The high- and low-affinity Km and Vmax values of CYP2D1 activity in our control microsomes (Table 1) were also similar to those reported previously by these authors (Witherow and Houston, 1999). In our studies, the kinetic parameters of the high-affinity component (Km(1), Vmax(1), and Clint(1)), whose CLint contributed to 93% of the total intrinsic clearance, were not substantially affected by NO donors (Table 1). Only the high-affinity component of the formation of DT from DM is attributed to the 2D1 enzyme (Kerry et al., 1993). Therefore, the apparent lack of NO effect on this component confirms the validity of the previous single-substrate concentration studies. It should be noted, however, that because of the small number of samples (n = 4) and substantial variability in the CLint(1) data for DT formation in the control livers (Table 1), the power of our study to detect a small difference between control and ISDN or SNP groups (Table 1), if indeed true, is low. Addition of three IPRL experiments to the control group resulted in a CLint(1) value (µl/min/mg protein) of 620 ± 200 (n = 7), which is within 10% of the average values for the SNP (690 ± 230) and ISDN (630 ± 60) groups (P > 0.05). Although not conclusive, the closeness of the mean values in all three groups suggests that the lack of effect of NO donors on CLint(1) for DT formation is likely real and not due to a low statistical power.
The apparent lack of effects of NO donors on 2D1 Vmax is likely due to inaccessibility of heme and/or cysteine thiolate of this enzyme (Gergel et al., 1997). Additionally, the lack of effects of NO donors on the Km value of this enzyme may be due to the lack of importance of thiol-containing amino acid (i.e., cysteine) residues in binding to the substrate. Indeed, recent reports (Paine et al., 2003) have clearly shown that the critical amino acids for the interaction of 2D6 with basic nitrogen-containing ligands are negatively charged carboxylate-containing amino acids, such as aspartate 301 and glutamate 216, and not cysteine residues. Nevertheless, these studies confirm that CYP2D1 is unique among the studied P450 enzymes in that it is not affected by NO donors.
In our present studies, the metabolic kinetics of 3A2 activity were determined by using both the 6ß-hydroxylation of testosterone (Bogaards et al., 2000) and formation of MXM from DM (Witherow and Houston, 1999) (Table 1). Whereas the former did not show any sigmoidicity (Fig. 1), the latter demonstrated sigmoidal kinetics (Fig. 4; Table 1). This sigmoidal pattern is indicative of autoactivation of the metabolism (Ekins et al., 1997), suggesting that binding of one substrate molecule facilitates the binding of the next molecule (Houston and Kenworthy, 2000). Other investigations of MXM formation (Witherow and Houston, 1999) have also shown sigmoidal kinetics, with a Km value (81 µM) close to that observed in our controls (53 µM; Table 1). The absence of autoactivation kinetics for 6ß-hydroxylation of testosterone would indicate that the autoactivation phenomenon depends on the substrate and/or substrate-enzyme interaction, as demonstrated before (Ekins et al., 1998).
It is interesting to note that although formation of both 6ß-hydroxytestosterone (Bogaards et al., 2000) and MXM (Witherow and Houston, 1999) is catalyzed by the same CYP3A2 enzyme, NO decreased the Km for 6ß-hydroxylation, whereas it increased the Km for MXM formation (Table 1). The opposing effects of NO donors on the two reactions catalyzed by the same enzyme could arise because NO may elicit a conformational, electronic, and/or steric change in the enzyme that increases the testosterone's affinity for the catalytic site, but decreases the metabolism of DM to MXM. This effect has also been demonstrated earlier (Schwab et al., 1988) for the ability of -naphthoflavone to directly inhibit or stimulate rabbit CYP3C enzyme, depending on the substrate used.
Our results with regard to the enzyme models and the magnitude of kinetic parameters in control microsomes (Table 1) are in agreement with the literature data for the formation of 6ß-hydroxylation from testosterone via 3A2 (Bogaards et al., 2000), MXM from DM via 3A2 (Witherow and Houston, 1999), and DT from DM via 2D1 (Witherow and Houston, 1999). However, formation of 6-HCLZ from CLZ via 2E1 showed sigmoidicity in our control microsomes (Fig. 5), a pattern not reported before (Jayyosi et al., 1995; Court et al., 1997; Bogaards et al., 2000; Howard et al., 2001). Despite this difference, the Km values reported in these studies are close to that observed in our control group (Table 1). As for resorufin formation from BR via 2B1/2, our two-enzyme kinetic model (Table 1) is different from the previously reported (Mayer et al., 1990) single-enzyme model. This discrepancy is most likely due to a narrower substrate concentration range used by these investigators. Overall, most of our data and selected metabolic models for untreated microsomes are in agreement with the literature data.
Despite some quantitative differences between the effects of SNP and ISDN on the kinetic parameters of different substrates, qualitatively, the two NO donors were similar in most cases (Table 1). One exception was the minor increase in the Vmax(1) for the formation of DT from DM in the presence of ISDN and not SNP (Table 1). Similar to the SNP group, however, this minor increase, which might be due to experimental variability, did not result in a change in the CLint(1) of the metabolic pathway.
In our previous (Vuppugalla and Mehvar, 2004a) and current studies, we used both SNP and ISDN, two NO donors with different mechanisms, sites, and modes of NO release (Feelisch, 1998), to assure that the observed P450 inhibitory effects are indeed due to the generation of NO and not related to nonspecific effects of these drugs. Whereas for SNP, a membrane-bound enzyme is likely involved in the generation of NO in biological tissues, an NADPH-dependent cytochrome P450 pathway and an isoenzyme of the glutathione S-transferase family have been proposed for the bioactivation of organic nitrates. Additionally, the metabolism of ISDN, but not SNP, may generate NOx independent of the formation of NO (Feelisch, 1998). Therefore, the perfusate level of NOx after ISDN is likely an overestimation of the liver exposure to NO. In agreement with these observations, our recent study showed that despite producing higher perfusate NOx concentrations at doses of 200 and 500 µM, the inhibitory effects of ISDN were less pronounced than those of SNP for most enzymes (Vuppugalla and Mehvar, 2004a). Nevertheless, the perfusate NOx concentrations produced in our studies for ISDN (30.2 ± 2.4 µM) and SNP (10.2 ± 0.4 µM) are pathophysiologically relevant because they are well below the values reported in the plasma of liver transplant patients (Ioannidis et al., 1995) and endotoxemic animals (Li-Masters and Morgan, 2002) (250 and 800 µM, respectively).
|
Conclusions |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
|
Footnotes |
|---|
ABBREVIATIONS: P450, cytochrome P450; NO, nitric oxide; IPRL, isolated perfused rat liver; Vmax, maximum velocity of metabolism; Km, Michaelis-Menten constant; CLint, intrinsic clearance; SNP, sodium nitroprusside; ISDN, isosorbide dinitrate; CLZ, chlorzoxazone; 6-HCLZ, 6-hydroxychlorzoxazone; DM, dextromethorphan; MXM, methoxymorphinan; DT, dextrorphan; R, resorufin; BR, benzyloxyresorufin; ER, ethoxyresorufin; T, testosterone; 6ß-HT, 6ß-hydroxytestosterone; 16-HT, 16-hydroxytestosterone; NOx, nitrate/nitrite; v, rate of metabolism; [s], substrate concentration; n, Hill coefficient; CLtotal, total intrinsic clearance; CLmax, maximum clearance.
Address correspondence to: Dr. Reza Mehvar, School of Pharmacy, Texas Tech University Health Sciences Center, 1300 S. Coulter, Amarillo, TX 79106. E-mail: reza.mehvar{at}ttuhsc.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Vuppugalla and R. Mehvar SELECTIVE EFFECTS OF NITRIC OXIDE ON THE DISPOSITION OF CHLORZOXAZONE AND DEXTROMETHORPHAN IN ISOLATED PERFUSED RAT LIVERS Drug Metab. Dispos., July 1, 2006; 34(7): 1160 - 1166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. C. Yohe, K. A. O'Hara, J. A. Hunt, T. J. Kitzmiller, S. G. Wood, J. L. Bement, W. J. Bement, J. G. Szakacs, S. A. Wrighton, J. M. Jacobs, et al. Involvement of Toll-like receptor 4 in acetaminophen hepatotoxicity Am J Physiol Gastrointest Liver Physiol, June 1, 2006; 290(6): G1269 - G1279. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
