Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Isosorbide dinitrate (ISDN) is widely used in various formulations and by different routes of administration as a basic vasodilator for the management of angina pectoris and congestive heart failure (Parker and Parker, 1998). A number of reports on disposition data of ISDN in human and animals have been published (Bogaert, 1983). There is substantial evidence suggesting that the vasodilating effect of organic nitrates such as ISDN requires a series of reactions, including cellular conversion to nitric oxide, activation of guanylyl cyclase, and intercellular accumulation of cyclic GMP (Ignarro et al., 1981; Kukovetz and Holxmann, 1990). However, in spite of this enormous quantity of knowledge, the relationship between plasma concentration and pharmacological response is still poorly understood (Thadani and Whitsett, 1988).

A major factor complicating the pharmacokinetic (PK)-pharmacodynamic (PD) relationship of ISDN may be the development of vascular tolerance during long-term therapy (Thadani, 1997). Another factor is the greater persistence of active metabolites, isosorbide-2-mononitrate (2-ISMN) and isosorbide-5-mononitrate (5-ISMN), making the interpretation of data difficult even after the initial dosing. The change in pulse pressure of rats following i.v. administration of 2-ISMN and 5-ISMN, which produce no active metabolites, has been successfully related with the plasma concentration (Tzeng and Fung, 1992a). ISDN and its active metabolites have the same mechanism of vasodilating action (Ignarro et al., 1981; Kukovetz and Holxmann, 1990), so the pharmacological effect after a single dose of ISDN can be described as a simple additive effect.

ISDN is a dinitrate having two pharmacologically active diastereomers, isoiodide dinitrate (IIDN) and isomannide dinitrate (IMDN), and their active metabolites, 2-ISMN, 5-ISMN, isoiodide mononitrate (IIMN), and isomannide mononitrate (IMMN) are also isomeric mononitrates. They are structurally different in a functional group, nitrate and hydroxy groups, and in their exo and endo positions (Fig. 1). Some in vitro and in situ studies have suggested that the vasodilating potency of different isomeric organic nitrates depends on the positions of functional groups (Bogaert and Rosseel, 1972; Noack, 1984). The systemic and conjugation clearances of mononitrates were also affected by their stereochemical arrangements (Tzeng and Fung, 1993). Unfortunately, little is known about the influence of stereostructure on the in vivo disposition and vasodilating action for dinitrates, because little information is available on IIDN and IMDN.

The objectives of the present study were to establish the PK-PD relationship of organic nitrates, especially dinitrates, and to comprehensively reevaluate the influence of stereochemistry on the PK and PD properties. Plasma nitrate concentration and blood pressure were simultaneously measured in rats after a single i.v. bolus administration of dinitrates and mononitrates. The PK model incorporating PD was developed to characterize the vasodilating effect.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Chemical structures of stereoisomeric organic nitrates used in this study.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Organic nitrates (IIDN, IMDN, and ISDN) and mononitrates (IIMN, IMMN, 2-ISMN, and 5-ISMN) were synthesized and purified by Toshin Chemical Co. (Tokyo, Japan). Their purities were more than 98%. Other chemicals and solvents were of reagent grade.

Animals. Male Wistar rats (Japan SLC Inc., Hamamatsu, Japan) aged 8 weeks were used. Rats were allowed free access to water and laboratory rat chow and housed in a room with a 12 h light/dark cycle for at least 1 week before the day of experiment. One day prior to the experiment, the right jugular vein was cannulated for nitrate administration and blood sampling, and the left carotid artery was for blood pressure measurement, according to Hatanaka et al. (1998a). After a one-night recovery period, each rat was studied in a conscious and unrestrained state.

Dosing. Twenty-one groups of rats (n = 3 in each group) received a single i.v. bolus dose of 50, 100, or 200 mg/kg of either a dinitrate or a mononitrate via jugular venous cannula. N,N-Dimethylacetamide was as a solvent for dinitrates due to their low aqueous solubilities, and only 0.2 ml of the solution was administered to minimize the toxicity. Mononitrates were dissolved in normal saline, making the injection volume less than 2 ml/kg. The dose was delivered over 5 s followed by a saline flush of the cannula.

PK Study. Blood samples (0.25 ml) were withdrawn from the jugular vein just prior to nitrate administration and at appropriate postdose times, which were predetermined based on the detection limit for each nitrate. The blood samples were immediately transferred to heparinized micro test tubes and centrifuged at 12,000 rpm for 3 min. The separated plasma samples were stored at 20°C until assay.

PD Study. Blood pressure of rats was measured with a biophysiograph (180 system, San-ei Instrument Co., Tokyo) connected to the carotid arterial cannula via a pressure transducer (MPU-0.5, Toyo Baldwin Co., Tokyo) and digitized with an A/D converter (GP-IB, Shoei Densi Laboratories, Nagoya, Japan). Three direct hemodynamic parameters, systolic blood pressure, diastolic blood pressure and heart rate, and the mean arterial pressure calculated by direct parameters were recorded every minute on a personal computer (HP-85, Hewlett-Packard, Corvallis, OR). After about a 30-min stabilization period, the measurement was carried out for 15 min to obtain the baseline level, and continued after nitrate administration until the last blood sampling.

PK-PD Modeling. To relate the vasodilating effect with plasma concentration of organic nitrates, a PK-PD model was developed as shown in Fig. 2. The following assumptions were made: 1) dinitrates distribute into a central compartment (compartment 1) and a peripheral compartment (compartment 2); 2) the elimination of dinitrates occurs only via metabolism to the active metabolites; 3) the disposition of mononitrates is described by a conventional one-compartment model (compartment 3 or 4); 4) the site of action is in the central and metabolite compartments; and 5) both dinitrates and mononitrates have an identical mechanism of action.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   PK-PD model for organic nitrates. C, concentration; V, volume of distribution; D, dose; k, first-order rate constant; subscripts 1 and 2, central and peripheral compartments of dinitrate; subscripts 3 and 4, compartments of mononitrate metabolites.

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

Statistics. Statistical significance of difference was evaluated by one-way ANOVA, and then by Scheffe's multiple comparison method. In all cases, P < 0.05 was considered to be significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

PK-PD of Dinitrates. Figure 3 shows the typical time courses of plasma concentration and mean arterial pressure in rats receiving a single i.v. dose of 100 mg/kg IIDN, IMDN, or ISDN. In all cases, the plasma concentration of dinitrate declined rapidly in a biphasic manner, whereas the mononitrate concentration peaked soon after dosing and then reduced slowly, maintaining higher levels than the dinitrate. The peak concentrations and terminal slopes of mononitrate, and the slopes of dinitrate decline differed among stereoisomers. The mean arterial pressure decreased to a minimum at the earliest recording time (1 min) after dosing and then recovered gradually to the baseline level. The duration to recovery of blood pressure level varied depending on the dinitrate administered.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Typical time courses of plasma concentration and mean arterial pressure in rats receiving a single i.v. dose of 100 mg/kg IIDN, IMDN, or ISDN. The smooth solid lines represent nonlinear least-squares fit of data to the PK-PD model shown in Fig. 2, and the other smooth lines are the contribution of dinitrate and mononitrate to total vasodilating effect, which was calculated based on the sigmoidal Emax model.

PK of Organic Nitrates. Plasma concentration profile after i.v. injection of a mononitrate to a rat showed a monoexponential decline. The average profiles are shown in Fig. 4. The slope of decline was independent of dose but clearly differed among mononitrate isomers. The PK parameters obtained by fitting the data for each rat to a linear one-compartment model are summarized in Table 1. Although the apparent volumes of distribution are almost the same among stereoisomeric mononitrates except for 5-ISMN, the elimination rate constant and systemic clearance were highest for IIMN, followed by IMMN, 2-ISMN, and 5-ISMN.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Average plasma concentration profiles after i.v. administration of mononitrates to rats. Each point represents the mean ± S.D. of three rats.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Average plasma concentration profiles after i.v. administration of ISDN to rats. Each point represents the mean ± S.D. of three rats.

PD of Organic Nitrates. A rapid decrease immediately after administration and then a slower increase to the baseline level in mean arterial pressure were found for doses of mononitrates as true for dinitrates. The hemodynamic effect after dosing of 100 mg/kg IMMN and 5-ISMN, and 50 mg/kg of all mononitrates, however, was too weak and brief to analyze the PK-PD relationship. Figure 6 shows representative relationships between plasma concentration and MAP in rats from 200 mg/kg mononitrate administration groups. Although no distinct maximum effect was observed, there was a log-linear relationship in data below 10% of mean arterial pressure at the baseline (above about 10 mm Hg of MAP) for each diastereomer. The best-fit lines to the log-linear model are shown in this figure and the estimated PD parameters are summarized in Table 3.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Relationships between plasma concentration and MAP in representative rats from 200 mg/kg mononitrate administration groups. The straight lines represent nonlinear least-squares fit of data to a log-linear model.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

PK of mononitrates, particularly 5-ISMN, has been well investigated in human and animals (Chasseand, 1987; Tzeng and Fung, 1993). The PK properties obtained here were essentially consistent with those reported previously. The volumes of distribution of all mononitrates were of the same order of magnitude as the volume (0.9 liters/kg) of total body water (Rothwell and Stock, 1979), although the value of 5-ISMN was significantly higher than that of other isomers (Table 1). The systemic clearance was highest for IIMN followed by IMMN, 2-ISMN, and 5-ISMN. No remarkable nonlinear PK behavior was observed for any mononitrates in the dose range tested (Fig. 4).

PK properties of ISDN have also been extensively studied (Bogaert, 1983). In the present study, a two-compartment model with elimination only from the central compartment via metabolism to mononitrates, as shown in Fig. 2, was used in the PK analysis of ISDN based on the plasma concentration profiles (Fig. 5). It is already known that ISDN distributes to various tissues (Reed et al., 1977) and eliminates mainly as denitrated metabolites and their conjugates (Rosseel and Bogaert, 1973). The steady-state volume of distribution calculated based on the model was comparable with the reported values, and the sum of metabolic clearances from ISDN to 2-ISMN and 5-ISMN (109 ml min¹ kg¹) was almost equal to the reported systemic clearance of ISDN (Table 2). The PK of IIDN and IMDN could be described based on the same model as ISDN except for a difference in the number of mononitrate metabolites (Fig. 3). This is not surprising because of their similarity to ISDN in physicochemical properties (Hatanaka et al., 1998b) and renal excretion aspects (Rosseel and Bogaert, 1973).

Difference in the estimated volumes of distribution was about 2-fold among dinitrate isomers (Table 2). A possible explanation for the difference is stereoselective plasma protein binding, which was confirmed for many racemic drugs (Jamali et al., 1989). The metabolic rate constants had the same order of magnitude as the corresponding elimination constants of mononitrates except for 5-ISMN (Tables 1 and 2). Therefore, the terminal slope in plasma concentration profile of 5-ISMN after ISDN administration equaled its elimination slope, whereas the terminal phase of other mononitrates was controlled by both production and elimination rates (Fig. 3). The metabolic clearances of all dinitrates (Table 2) exceeded hepatic plasma flow in rat (Bischoff et al., 1971). The extensive metabolism of nitroglycerin and ISDN by extrahepatic tissues including blood vessels has been confirmed in vitro (Fung et al., 1984). The clearances were always higher than systemic clearances of mononitrates capable of being denitrated at the same position (Tables 1 and 2). The faster denitration in dinitrates might be due to their larger distribution volume and thus better access to enzymes catalyzing denitration. Another explanation could be higher binding affinity, by analogy with experimental results for nitroglycerin and the corresponding dinitrates (Needleman and Hunter, 1965). Compared among dinitrate diastereomers, the metabolic clearance had about 7-fold variation, and the values were significantly different from one another (Table 2).

Organic nitrates introduced into systemic circulation partition into vascular cells where they undergo transformation involving denitration with the subsequent liberation of nitric oxide (Ignarro et al., 1981). Nitric oxide simulates guanylyl cyclase, which leads to the conversion of GTP to cyclic GMP, which in turn causes vasodilation (Kukovetz and Holxmann, 1990). The series of reactions following generation of nitric oxide is very rapid (Keith et al., 1982; Kelm et al., 1988). These facts allow us to assume that the vasodilating effect after dinitrate administration is a simple additive effect of both the nitrate and its mononitrate metabolites with an identical mechanism of action. In the vasodilation by mononitrates, the experimental E_max was not obtained even at the maximum dose of 200 mg/kg (Fig. 6), nor was E_max observed in the reduction of rat pulse pressure (Tzeng and Fung, 1992a) or relaxation of aortic ring (Tzeng and Fung, 1992b). Although the hemodynamic and antianginal actions of organic nitrates are mediated through vasodilation of veins and arteries, the venodilation predominates and dilation of arteries and arterioles occurs to a lesser extent (Imhof et al., 1982). A log-linear model was therefore used to describe the PK-PD relationship of mononitrates (Fig. 6). In contrast, the pharmacological effect after dinitrate dosing was so strong that it could be analyzed by a sigmoidal E_max model (Fig. 3).

Regardless of stereoisomers, the estimated E_max and  values were about 50 mm Hg and 1 (Table 4). The slopes of log-linear regression line for the vasodilation after mononitrate dosing, which is a function of E_max and  (eq. 6), were identical among isomers (Table 3). These results support that the vasodilating effect of all organic nitrates is caused by a common mediator, cyclic GMP. The EC₅₀ values of dinitrates were remarkably lower than those of mononitrates (Table 4). The higher vasodilator potency of dinitrates might be due to their higher lipophilicity and thus greater partition into the vascular cells, as pointed out in previous in situ and in vitro studies (Bogaert and Rosseel, 1972; Noack, 1984). Although EC₅₀ showed a large interindividual difference, the vasodilating potency also seemed to depend on the stereostructure of organic nitrates.

Based on the PK and PD parameters obtained here, the influence of stereochemistry on the PK and PD of organic nitrates is discussed. The metabolic clearance from dinitrate to mononitrate showed that two nitrate groups rather than just one at the exo position rather than endo position are easily denitrated (Table 2). The EC₅₀ values of dinitrates had a tendency to decrease with increase in the number of exo nitrate groups, and thus another advantage of exo position to denitration was found (Table 4). The nitrate group at the endo position interacts with the lone-pair electrons of the oxygen atom in the adjacent ring and thus lends itself less easily to enzymatic attack than the same group at the exo position (Hayward et al., 1967). Comparison of EC₅₀ among mononitrate isomers showed that exo hydroxy groups are helpful to denitration (Table 4). This can be explained by the hypothesis that enzymes metabolizing nitrates to nitric oxide contain two attachment sites that can bind to the oxygen atoms and are separated by a distance of 5.8 Å (Tzeng and Fung, 1992c). The effect of stereostructure on the elimination of mononitrates is not simple, because the compounds are cleared from the body via not only denitration but also glucuronidation (Wood et al., 1984). The urinary conjugation clearances were around 10% except for 43% of IMMN (Tzeng and Fung, 1993). The hydroxy groups at the endo position are also involved in intermolecular interactions via hydrogen bonding to an oxygen atom in the same ring (Anteunis and Verhegghe, 1971), but the interaction is favorable to glucuronidation, which requires a backside attack (the so-called S_N2 mechanism). The intermolecular interaction of both nitrate and hydroxy groups might cause the exceptionally extensive conjugation and then systemic elimination of IMMN (Table 1).

In conclusion, the PK-PD relationship of organic mononitrates and dinitrates was established, and the stereoselective clearances and vasodilating effects were demonstrated. Stereoisomers having exo nitrate groups generally showed high vasodilating potency, but their duration of action was short due to high systemic and metabolic clearances. It is understandable that ISDN and 5-ISMN, whose potency and clearance are in the middle among isomers, are used clinically. However, the duration of action can be extended by formulations and administration routes (e.g., transdermal patches), so that other isomers can also be clinically useful vasodilators. The vasodilating effect of organic nitrates is not always parallel to the anti-ischemic effect and nitrate tolerance shifts the PK-PD relationship depending on time (Thadani, 1997; Parker and Parker, 1998). Some information obtained here may not be directly utilized in the improvement of nitrate treatment of angina pectoris and congestive heart failure. However, the present study is a thorough PK-PD study of several organic dinitrates and mononitrates in conscious rats and thus a worthwhile step to establish the clinical PK-PD relationship.