Intestinal absorption of temocapril, a prodrug of temocaprilat, was evaluated in an in situ rat jejunal perfusion model under various conditions of luminal pH and in the presence and absence of carboxylesterase-mediated hydrolysis. Temocapril was more easily taken up by mucosal cells at a luminal pH of 5.4 than at pH 6.4 or 7.4 and was extensively hydrolyzed to temocaprilat in mucosal cells. The hydrolysis was limited by the intrinsic clearance and the influx rate at luminal perfusate pHs of 5.4 and 7.4, respectively. Temocaprilat, derived from temocapril, was transported into both mesenteric vein and jejunal lumen according to pH partition theory. The net absorption of both temocapril and temocaprilat was highest at a luminal perfusate pH of 5.4. When both the luminal and venous fluid were at pH 7.4, temocaprilat was transported approximately 3-fold faster into the lumen than into the vein, due presumably to the greater surface area of the brush border membrane because of the presence of microvilli. Under carboxylesterase-inhibited conditions, the hydrolysis of temocapril was inhibited by only 50%. It is postulated that serine esterases located on the membranes of the epithelial cells were responsible for the residual hydrolysis. We have confirmed that temocapril is most easily absorbed in the proximal intestine after meals, due to prolongation of the gastric emptying time, the lower intraluminal pH caused by secretion of bile acid, and the interaction between serine esterases and the digesta.
Prodrug modification is a major factor in drug design that can improve, delay, prolong, control, or target the action of the parent drug (Ettmayer et al., 2004; Testa, 2004). The prodrug, itself an inactive compound, is activated by enzymes in the body to allow the pharmacological effect of the parent drug to be delivered at the target sites. The majority of prodrugs are ester derivatives, involving modification of carboxyl and hydroxyl groups in the parent drugs, because of the ubiquitous expression of hydrolase, which is required for their activation, in many organs (Imai, 2006; Imai and Hosokawa, 2010). Most prodrugs used in clinical practice are developed to improve the oral bioavailability of the parent drug, e.g., valacyclovir and prulifloxacin (Liederer and Borchardt, 2006). These prodrugs are more easily taken up into the intestine and liver because of their membrane permeability, which depends on their lipophilicity, and are mostly activated by first-pass metabolism before entering the systemic circulation.
We have previously reported the extensive hydrolysis of an ester prodrug in the intestine in an in situ rat jejunal single-pass perfusion experiment (Masaki et al., 2006). In that report, isovaleryl propranolol was easily taken up into rat jejunal mucosal cells by passive diffusion and completely hydrolyzed at a rate limited by its membrane permeability. Both the propranolol and isovaleric acid derived from isovaleryl propranolol were present at their highest concentrations in epithelial cells, compared with blood vessels and the intestinal lumen. Propranolol (pKa 9.44) and isovaleric acid (pKa 4.77) were predominantly transported into the lumen and vein, respectively, by passive diffusion according to pH partition theory. These results suggested that successful oral prodrugs are stable in the intestine and rapidly hydrolyzed in the liver, and that extensive intestinal hydrolysis would be able to provide adequate bioavailability for an acidic but not a basic parent drug.
In the present study, we examined the possibility of increasing the bioavailability of a prodrug that is hydrolyzed to an acidic parent drug in the intestinal mucosa. Temocapril, an angiotensin-converting enzyme (ACE) inhibitor, was selected as a model prodrug. The parent drug of temocapril is temocaprilat, a dicarboxylic acid compound. It has been reported that the oral bioavailability of temocapril is 81 and 65% in rat and man, respectively (Koike et al., 1992; Püchler et al., 1998). The absorption of temocapril could be achieved by improving its membrane permeability in the intestinal mucosa. We have reported previously that carboxylesterase (CES) is one of the major enzymes involved in the hydrolysis of temocapril in the intestine and liver (Imai et al., 2005). In man, temocapril is hardly hydrolyzed by human CES2 isozyme, which is expressed in the intestine, whereas it is rapidly hydrolyzed by human CES1 isozyme, the major CES isozyme in the liver (Imai et al., 2006); therefore, it may be expected that temocapril is absorbed in the intestine as an intact prodrug and is then rapidly hydrolyzed to temocaprilat in the liver. In contrast, rat intestine contains more than three CES2 isozymes (Sanghani et al., 2002; Furihata et al., 2005) and shows more extensive hydrolytic activity than human intestine (Taketani et al., 2007).
In this study, the absorption and hydrolysis of temocapril was evaluated in an in situ rat jejunal single-pass perfusion model, under various conditions of luminal pH, to clarify the direction of efflux of temocaprilat produced in epithelial cells. The relative areas of the brush border membrane and the plasma membrane were determined by comparing the efflux rates of temocaprilat from mucosal cells into the intestinal lumen and blood vessels. To predict the human intestinal absorption of temocapril, rat jejunum perfusion experiments were performed in the absence of CES-mediated hydrolysis, after treatment with a specific CES inhibitor. Finally, we examined the contribution of serine proteases other than CES in the intestinal hydrolysis of temocapril.
Materials and Methods
Temocapril and temocaprilat were kindly provided by Daiichi Sankyo Co., Ltd. (Tokyo, Japan). Bovine serum albumin (BSA; fraction V) and fluorescein isothiocyanate dextran 4000 (FD-4) were purchased from Sigma-Aldrich (St. Louis, MO). HEPES and dimethyl sulfoxide (DMSO) were purchased from Wako Pure Chemicals (Tokyo, Japan). Bis-p-nitrophenyl phosphate (BNPP) and pentobarbital sodium salt were purchased from Nacalai Tesque (Kyoto, Japan). All other chemicals were of analytical grade.
Male Wistar rats (230–280 g, 8 weeks of age) were housed in an air-conditioned room with ad libitum access to commercial chow and tap water. Animals were fasted for 18 to 24 h before the experiments.
In Situ Intestinal Single-Pass Perfusion.
The perfusion study was performed according to a previously described method (Masaki et al., 2006). In brief, rats were anesthetized by intraperitoneal injection of 2% pentobarbital sodium salt dissolved in physiological saline (0.5 ml). A segment of approximately 10 cm of upper jejunum was isolated, and both ends of the jejunal loop were cannulated with Teflon tubing (3 mm i.d., 5 mm o.d.) after flushing out the intestinal contents with warmed physiological saline. The superior mesenteric artery and the portal vein were cannulated with polyethylene tubes (PE10 and PE50, respectively) to enable perfusion of the blood vessels. The cannulated intestinal segment was isolated and immersed in Krebs-Henseleit bicarbonate buffer containing 3% BSA and 10 mM d-glucose (pH 7.4), warmed to 37°C.
Krebs-Henseleit bicarbonate buffer was used as the vascular perfusate at a flow rate of 2.5 ml/min. The jejunal loop was perfused with phosphate buffer adjusted to pH 5.4, 6.4, or 7.4 at a flow rate of 0.2 ml/min. The osmolarity of the luminal perfusate containing temocapril or temocaprilat (100 μM) was adjusted to 290 mOsm/l before the addition of FD-4 (0.1 mg/ml), a nonabsorbable marker, to enable monitoring of the luminal volume. The perfusates from the luminal segment and vascular outflow were collected at appropriate points. A CES-inhibited condition was achieved by preperfusion with BNPP (400 μM) for 40 min, in accordance with a previously reported methodology (Masaki et al., 2007). The chemical degradation of temocapril was negligible during the perfusion experiment.
For the determination of temocapril and temocaprilat concentrations, luminal perfusate samples (0.5 ml) were diluted with equal volumes of acetonitrile and shaken. A sample of the supernatant (180 μl) was added to 3.4 M phosphoric acid (20 μl) and injected onto an HPLC column. FD-4 levels (0.1 mg/ml) in the luminal samples were determined using a fluorescence spectrometer (F-4500; Hitachi High-Technologies Co., Tokyo, Japan). Samples (5 ml) of vascular perfusate were treated with 0.67 M phosphoric acid (0.5 ml) and extracted with ethyl acetate (20 ml). The organic phase was separated and evaporated. The resulting residue was redissolved in a 180-μl solution of acetonitrile-water (1:2, v/v) with 3.4 M phosphoric acid (20 μl) and injected onto an HPLC column. After the perfusion period, the intestinal segment was cut open and washed with ice-cold phosphate buffer (pH 7.4). A portion of the intestinal mucosa (350–400 mg) was immediately stripped and homogenized in a solvent (6 ml) consisting of acetonitrile-methanol (1:2, v/v). After centrifugation at 1500g for 10 min, the supernatant (4 ml) was evaporated. An HPLC sample was made from the residue, using the same method as for the vascular perfusate sample.
Preparation of Homogenate and S9 Fraction of Intestinal Mucosa.
Rats were anesthetized with ether and sacrificed by exsanguination from the external jugular vein. The duodenum and jejunum were removed and washed with ice-cold 1.15% KCl. The mucosa was stripped, minced, and homogenized with 3 volumes of 50 mM HEPES buffer (pH 7.4) containing 1.15% KCl using a Potter-Elvehjem Teflon pestle under ice-cold conditions. The homogenates were centrifuged at 9000g for 20 min at 4°C to obtain the supernatant (S9 fraction). Protein contents were determined by the method described by Bradford (1976) using BSA as the standard. The homogenates and S9 fractions were stored at −80°C until use.
Hydrolysis of Temocapril in the Rat Duodenal and Jejunal Mucosa.
The duodenal and jejunal S9 fractions and homogenates were diluted with 50 mM HEPES buffer (pH 7.4) to the required concentrations. The S9 solutions (200 μl) were preincubated at 37°C for 5 min, and the reaction was started by adding 2 μl of temocapril (final concentration, 10–500 μM) dissolved in DMSO. DMSO inhibits hydrolase activity at higher concentrations but has no effect at 1%. The concentration of DMSO was therefore maintained at 1%. After a 20-min incubation, reactions were terminated by adding 200 μl of ice-cold acetonitrile. The reaction mixture was centrifuged at 7200g for 3 min. Phosphoric acid (40 μl, 2.9 M) was added to a 200-μl sample of the supernatant, and the resultant mixture was analyzed by HPLC. Hydrolytic activity was evaluated by measuring the formation of temocaprilat. Kinetic parameters, Km and Vmax, were calculated by fitting the data to the Michaelis-Menten equation by nonlinear least-squares analysis, using the MULTI program (Yamaoka et al., 1981).
The hydrolase activities of intestinal homogenate and S9 fractions were inhibited using BNPP, a specific CES inhibitor, as well as the serine protease inhibitors diisopropyl fluorophosphonate (DFP) and paraoxon, by preincubation for 5 min before the addition of temocapril. The degree of inhibition was calculated as a percentage of control activity.
Temocapril and temocaprilat concentrations were determined by HPLC. The HPLC system (Jasco, Tokyo, Japan) consisted of a pump (PU-980), an UV detector (UV-970), an autosampler (AS-950), a column oven (CO-965), and a ChromNAV data application apparatus (version 1.5). An aliquot of the sample was injected onto a Mightysil RP-18 column (5 μm, 250 × 4.6 mm i.d.; Cica-Merck Co., Inc., Tokyo, Japan) and eluted at a flow rate of 0.8 ml/min with methanol (solvent A) and 10 mM phosphoric acid (solvent B) according to the following gradient schedule: 40% solvent A for the first 20 min, a linear gradient from 40 to 55% solvent A over the next 2 min, 55% solvent A for 15 min, a linear gradient from 55 to 70% solvent A over the next 3 min, and 70% solvent A for 5 min. The temperature of the column was maintained at 40°C. The elution times of temocaprilat and temocapril were 23.1 and 41.0 min, respectively. UV detection was performed at 258 nm, and the detection limits for both temocapril and temocaprilat were 10 pmol.
Fig. 1 shows the absorption parameters in the in situ perfusion experiment, calculated as normalized values for the 10-cm intestine, as described in a previous report (Masaki et al., 2006). The rates of appearance of temocapril (v2) and temocaprilat (v4) in the mesenteric vein were calculated by v2 = Qb × Cb, temocapril and v4 = Qb × Cb, temocaprilat, respectively, where Qb is the flow rate of vascular perfusion and Cb is the concentration in the mesenteric vein. The rate of disappearance (v1) of administered compound and the rate of appearance (v3) of temocaprilat in the intestinal lumen were calculated by the following: where Ql is the flow rate of intestinal perfusion, Cin and Cout are the concentrations of administered compound at the entrance and exit of the jejunal segment, respectively, and Cl, temocaprilat is the concentration of temocaprilat at the exit of the jejunal segment.
The apparent absorption clearance into the mesenteric vein (CLabs) was calculated by the following: where AUCb and AUCl are the areas under the curve of the administered compound in the mesenteric vein and the intestinal lumen, respectively, at steady state. The degradation clearance of temocapril to temocaprilat in the jejunal mucosa (CLdeg) was calculated by the following: where AUCl, temocaprilat and AUCl, temocapril are the areas under the curve of temocaprilat and temocapril in the intestinal lumen, respectively. It was assumed that the concentration of administered compound in the intestinal loop decreases according to first-order kinetics when calculating the AUC in the intestinal lumen. AUCb, temocaprilat is the area under the curve of temocaprilat in the mesenteric vein.
The permeability rate constant (Peff) was calculated by the following: where R is the radius of the segment, assumed to be 0.178 cm (Yamashita et al., 1997), and L is the length of the segment (i.e., 10 cm).
Results are expressed as the mean ± S.D. Data were analyzed using the Student's t test with p < 0.05 as the minimal level of significance.
Absorption of Temocaprilat by Rat Jejunal Single-Pass Perfusion.
When 100 μM temocaprilat was perfused in the rat jejunal lumen at pH 6.4, steady state was achieved after 10 min. The rate of appearance of temocaprilat in the blood vessel was 0.193 nmol/min; the concentration of temocaprilat in the luminal perfusate was nearly constant during the experiment. The absorption clearance (CLabs) was calculated as 1.87 μl/min, as shown in Table 1.
Absorption of Temocapril by Rat Jejunal Single-Pass Perfusion.
Temocapril (100 μM) was perfused into the jejunal lumen at pH 6.4. The results are shown in Fig. 2 and Table 1. Steady state was achieved after 70 min. The rate of disappearance of temocapril from the jejunal lumen (v1) was 3.89 nmol/min, and the effective permeability coefficient (Peff) was 3.43 × 10−3 cm/min; however, the rate of appearance of intact temocapril in the mesenteric vein (v2) was 0.147 nmol/min, only 3.8% of the rate of disappearance in the jejunum lumen. Therefore, temocapril molecules taken up into jejunal mucosa were mostly hydrolyzed to temocaprilat in the cell, and temocaprilat was then transported into both the jejunal lumen and the mesenteric vein. The rate of appearance of temocaprilat in the blood vessel (v4) and in the jejunal lumen (v3) was 1.22 and 2.25 nmol/min, respectively, approximately 31 and 58% of the disappearance rate of temocapril in the jejunum lumen, respectively (Fig. 2). CLdeg was calculated as 35.4 μl/min; this value is 22-fold greater than the CLabs of temocapril (1.59 μl/min). The ratio of CLdeg to the sum of CLdeg and CLabs is 0.957, indicating nearly complete hydrolysis of temocapril in the mucosal membrane.
Effect of the pH of the Luminal Perfusates on the Absorption of Temocapril.
The absorption and hydrolysis of temocapril and the efflux of its hydrolysate, temocaprilat, from the mucosal cell were examined in intestinal fluid with the pH adjusted to 5.4, 6.4, or 7.4. The results are shown in Fig. 3 and Table 1. The values of v1 and Peff were increased at pH 5.4 but were nearly identical at pH 6.4 and pH 7.4, because of nonionic fractions of temocapril with pI values of 3.69 [nonionic form: 85.8% at pH 5.4, 98.4% at pH 6.4, and 99.8% at pH 7.4 (Vistoli et al., 2009)]. The greater influx of temocapril into the mucosal cells at a luminal perfusate pH of 5.4 resulted in higher intracellular concentrations of temocapril at this pH compared with pH 6.4 or 7.4 (Table 1); therefore, the highest rate of appearance of temocapril in the blood vessel (v2) was observed at a luminal fluid pH of 5.4. CLdeg increased with increasing pH, and the hydrolysis fraction calculated from CLdeg/(CLdeg + CLabs,temocapril) was smallest at pH 5.4. The intracellular concentration of temocaprilat was approximately 2-fold higher at pH 5.4 than at pH 6.4 and 7.4, and the rate of appearance of temocaprilat (v4) in the blood vessel was 1.6- to 2.2-fold faster at pH 5.4 than at pH 6.4 or 7.4. In contrast, the rate of appearance of temocaprilat in the jejunal lumen (v3) was slowest at pH 5.4.
When the luminal and venous pH were both 7.4, the rate of transport of temocaprilat into the lumen was approximately 3-fold faster than that into the blood, as shown in Fig. 3. Under the same pH conditions in the vascular and luminal perfusates, temocaprilat permeates through brush border membrane and plasma membrane according to the characteristics of each membrane (e.g., surface area and expression of efflux transporters). It seems likely that one reason for the greater permeability of temocaprilat into the lumen is the larger surface area of the brush border membrane, due to the presence of microvilli.
Hydrolysis of Temocapril in Rat Intestinal S9.
Having demonstrated that temocapril was well absorbed in the proximal segment of the intestine, where the luminal pH is approximately 5, the hydrolysis of temocapril was also evaluated in rat duodenal and jejunal S9. Km and Vmax values and the intrinsic hydrolytic clearance, calculated as the Vmax/Km, are listed in Table 2. The enzyme kinetic parameters were nearly the same in duodenal and jejunal S9. Hydrolysis in the duodenal and jejunal S9 was inhibited by 89 and 88%, respectively, after the addition of 1 mM BNPP. These data indicate that CES is predominantly responsible for the intestinal hydrolysis of temocapril in the rat.
Intestinal Absorption of Temocapril by Rat Jejunal Single-Pass Perfusion in the Absence of CES-Mediated Hydrolysis.
It has been reported that temocapril is hardly hydrolyzed to temocaprilat by human intestinal CES (Takai et al., 1997; Imai et al., 2005). Because temocapril is rapidly hydrolyzed by rat intestinal CES, human intestinal absorption of temocapril cannot be predicted by rat single-pass perfusion experiments; therefore, this single-pass perfusion experiment was performed under inhibition of CES-mediated hydrolysis using BNPP, according to a previously reported methodology (Masaki et al., 2007). Masaki et al. (2007) proposed a methodology whereby rat intestinal CES was inhibited by BNPP; BNPP irreversibly inhibits all CES isozymes with Ki values of 44.9 ± 4.95 nM in rat jejunal S9 but has no effect on other hydrolases such as aminopeptidase. Masaki et al. (2007) also established that nearly complete inhibition of CES is obtained by preperfusion with BNPP (400 μM for 40 min). More recently, Ohura et al. (2010) reported that BNPP affects neither active transport, e.g., p-glycoprotein, peptide, and organic anion transporters, nor passive diffusion in Caco-2 cell monolayers.
The perfusion experiment under inhibition of CES-mediated hydrolysis was performed at pH 5.4 and 6.4, and the results are presented in Table 3. Under CES-inhibited conditions, the disappearance rate of temocapril from jejunal lumen (v1) is approximately 60% of control values at both pH 5.4 and 6.4 (2.98 nmol/min versus 4.82 nmol/min at pH 5.4; 2.44 nmol/min versus 3.89 nmol/min at pH 6.4). Peff was also decreased to 2.99 × 10−3 and 2.32 × 10−3 cm/min at pH 5.4 and 6.4, respectively. On the other hand, the intramucosal concentration of temocapril was increased to 20.6 and 15.4 nmol/g tissue at pH 5.4 and 6.4, respectively, and the rate of appearance of temocapril in the mesenteric vein (v2) was increased from 0.353 to 0.862 nmol/min at pH 5.4 and from 0.147 to 0.696 nmol/min at pH 6.4. CLabs, temocapril also increased with increasing v2.
Meanwhile, CLdeg was decreased by CES inhibition from 34.4 to 24.9 μl/min at pH 5.4 and from 35.4 to 19.5 μl/min at pH 6.4; the intracellular concentration of temocaprilat was therefore decreased. However, the intracellular concentration of temocaprilat under CES-inhibited conditions remained at 50 to 81% of the values in control intestine (Tables 1 and 3). The rates of appearance of temocaprilat into the jejunal lumen (v3) and mesenteric vein (v4) were relatively constant, and the overall rate of appearance of temocaprilat (v3 + v4) was more than 50% of control values. These data suggest that enzymes other than CES contribute to the hydrolysis of temocapril in rat intestine.
Hydrolysis of Temocapril in Homogenate and S9 Fraction of Rat Jejunum Preperfused with BNPP.
To elucidate the non-CES enzyme activity in the rat jejunum, a hydrolysis experiment was performed using the homogenate and S9 fraction prepared from three kinds of jejunum: intact jejunum (without any perfusion), jejunum perfused by buffer (without BNPP), and jejunum preperfused with 400 μM BNPP for 40 min. The results are shown in Table 4. The hydrolysis of temocapril was nearly identical in the intact jejunum, and the jejunum was perfused by buffer (4.17 and 3.96 nmol · min−1 · mg protein−1 in the homogenate and 4.46 and 4.28 nmol · min−1 · mg protein−1 in the S9 fraction, respectively). This confirmed that the single-pass perfusion procedure itself had no effect on hydrolytic activity. It is interesting to note that the activity in the homogenate was nearly the same as that in the S9 sample. The homogenate of whole jejunal mucosa contains cell membrane fragments and nuclei that are absent from the S9 fraction. Because CES is present in the endoplasmic reticulum, CES-mediated hydrolase activity (per milligram of protein) should be higher in the S9 fraction than in the homogenate. Our results, which show similar activity in the homogenate and S9 fraction, indicate the presence of proteins in the homogenate that possess comparable hydrolytic activity to CES. Jejunum preperfused with BNPP exhibited 15 and 52% of the hydrolysis of temocapril in the S9 fraction and homogenate, respectively, compared with the jejunum perfused by buffer. These values are in good agreement with the percentages of activity remaining after the addition of 1 mM BNPP (10.9% in S9 fraction and 47% in homogenate; Table 4). These data show that enzymes other than CES are partly responsible for the hydrolysis of temocapril in rat intestine.
To determine the properties of the enzymes that are involved in the hydrolysis of temocapril, an inhibition experiment was performed. As shown in Fig. 4, hydrolase activity in homogenates was not inhibited by the further addition of BNPP; this presumably is due to the fact that CES activity had been almost completely inhibited by preperfusion with BNPP. The remaining hydrolase activity was nearly completely inhibited by the potent serine protease inhibitors DFP and paraoxon, at only 10 μM. These data suggest that temocapril is hydrolyzed not only by CES but also by other serine esterases in rat intestine.
The prodrug approach has been used successfully to improve the absorption of the ACE inhibitor temocaprilat. The partition coefficients (log P) between n-octanol and phosphate buffer (pH 7.4) are −2.5 and −0.1 for temocaprilat and its prodrug, temocapril, respectively (Shionoiri et al., 2001). The greater membrane permeability of temocapril is due to its higher hydrophobicity. Although the majority of ACE inhibitors are substrates for peptide transporters (Boll et al., 1994), our preliminary findings suggest that temocapril is not a substrate for PEPT1, but rather an inhibitor thereof (data not shown). We have confirmed that temocapril is partially transported by organic anion-transporting polypeptides in Caco-2 cell monolayers (data not shown). The fact that a decrease in the pH of the intestinal lumen led to an increase in the membrane permeability of temocapril indicates its absorption by passive diffusion, due to an increase in the nonionic fraction at lower pH (Vistoli et al., 2009). From the relationship between the in vivo fraction absorbed in man and the Peff obtained from the rat in situ experiment, complete absorption might be predicted in human intestine (Fagerholm et al., 1996).
Temocaprilat, formed from temocapril in epithelial cells, is transported both into blood and into the intestinal lumen. It has been reported that temocaprilat is carried by multidrug-resistance-associated protein 2 (Ishizuka et al., 1997) and organic anion-transporting polypeptides (Ishizuka et al., 1998). It is certainly transported across the double-transfected Madin-Darby canine kidney II monolayers that express rat Oatp4 and Mrp2 on the basal and apical membranes (Sasaki et al., 2004). Indeed, low concentrations of temocaprilat (1 μM) are transported by the carriers mentioned in these reports, but higher concentrations saturate these transporters. When 100 μM temocaprilat was applied to Caco-2 cell monolayers, it permeated equally well in apical-to-basolateral and basolateral-to-apical directions (data not shown), indicating that, at high concentrations, temocaprilat will cross the membrane by passive diffusion. As shown in Table 1, the intracellular concentration of temocaprilat in rat intestinal epithelial cells is 17.6 to 34.2 nmol/g tissue. Temocaprilat may passively diffuse out of epithelial cells, its dicarboxylic acid groups allowing diffusion from an acidic condition to one at a higher pH, according to pH partition theory.
When only the pH in the jejunum was varied (Fig. 3; Table 1), the transport of temocaprilat from epithelial cells into the intestinal lumen was shown to be affected by the pH of the luminal fluid. The fastest rate of transport of temocaprilat from epithelial cells into blood vessels was observed at a luminal pH of 5.4. The ratio of transport of temocaprilat into blood vessel (absorption), to transport into the lumen (secretion), was approximately 1.67 at a luminal perfusate pH of 5.4, 0.54 at pH 6.4, and 0.33 at pH 7.4. The greatest net absorption of both temocapril and temocaprilat was therefore observed at a luminal pH of 5.4.
When the pH of both the vascular and luminal perfusates was 7.4, temocaprilat formed in the mucosa was transported into the lumen at a 3-fold faster rate than into the mesenteric vein. Because there is no pH gradient, the efflux of temocaprilat from the mucosal cell must be dependent on membrane characteristics. At high intracellular concentrations, temocaprilat is transported by passive diffusion, so the extent of the available surface area affects the amount transported. The brush border membrane has a far greater surface area than the plasma membrane because of the presence of microvilli. With propranolol, a passively diffused compound formed from isovaleryl propranolol in the intestinal mucosa, appearance in the lumen is 4-fold greater than in the mesenteric vein (Ohura and Imai, 2010). Hydrolysis in the intestinal mucosa has a negative effect on the absorption of prodrug, because of the 3- to 4-fold greater efflux of hydrolysate into the lumen.
The highest value (0.972) of the hydrolytic ratio of temocapril, calculated from CLdeg/(CLdeg + CLabs,temocapril), was observed at a luminal fluid pH of 7.4, and the lowest value (0.900) was observed at pH 5.4. As this value approaches 1.0, intracellular hydrolysis is limited almost exclusively by the uptake of prodrug; lower values indicate that intracellular hydrolysis is limited by the intrinsic mucosal hydrolysis. We previously reported that the hydrolytic ratio of isovaleryl propranolol was 0.983 (Masaki et al., 2006). Isovaleryl propranolol is rapidly hydrolyzed in mucosal S9, with 2110 μl · min−1 · mg protein−1 of intrinsic clearance; this is 20-fold greater than that of temocapril (108 μl/ · min−1 · mg protein−1; Table 2). The hydrolysis of temocapril to temocaprilat is limited by the uptake and the intrinsic mucosal hydrolysis of temocapril at luminal fluid pHs of 7.4 and 5.4, respectively. The present data indicate that prodrugs with a greater intrinsic hydrolytic clearance than temocapril (approximately 110 μl · min−1 · mg protein−1) may be completely hydrolyzed during intestinal absorption.
Both rat and human intestine express only CES2 isozymes (Taketani et al., 2007); in the rat intestine, three major isozymes are present, AB010635, AY034877, and D50580, which have, respectively, 70, 67, and 70% homology with the amino acid sequence of human CES2 isozyme (Sanghani et al., 2002). Although temocapril is hydrolyzed in rat jejunum mucosal S9 at an intrinsic clearance of 108 μl · min−1 · mg protein−1 (Table 2), its intrinsic hydrolytic clearance is only 0.324 μl · min−1 · mg protein−1 in human intestinal S9 (Imai et al., 2005); this value is 330-fold smaller than that in rat jejunum S9. Therefore, rat intestinal absorption of temocapril does not accurately reflect human intestinal absorption.
To predict human intestinal absorption in the rat model, we used the previously reported methodology for inhibition of CES-mediated hydrolysis by preperfusion with BNPP (Masaki et al., 2007). The absence of CES-mediated hydrolysis led to an increase in the intramucosal concentration of temocapril, followed by an increase in CLabs and decrease in Peff (Table 3). It is interesting to note that, under CES-inhibited conditions, the CLdeg is maintained at 55 to 72% of control, whereas the intracellular concentration of temocaprilat is kept at 14.6 to 17.0 nmol/g tissue. In contrast, the CLdeg of isovaleryl propranolol, which is mainly hydrolyzed by CES, was shown to be inhibited by 77% after treatment with BNPP (Masaki et al., 2007). This supports our contention that enzymes other than CES are involved in the hydrolysis of temocapril in rat intestine. The enzyme involved in the residual hydrolysis of temocapril after treatment with BNPP was mainly present in the cell components in the homogenate but not the S9 fraction, and this activity was inhibited by organophosphates. These data suggest that temocapril is not only hydrolyzed by CES, but also by other hydrolases, containing serine residues in their catalytic centers and present in cell membranes.
It is also possible that temocapril is hydrolyzed in the cell membrane before entering the mucosal cell. The CLdeg in a luminal perfusate at pH 5.4 was only inhibited by 30% after treatment with BNPP, suggesting that the membrane enzyme involved is more active at pH 5.4 than at pH 6.4. However, in the homogenates of jejunum preperfused with BNPP, temocapril was hydrolyzed at rates of 0.535 and 1.19 nmol · min−1 · mg protein−1 at pH 5.4 and 6.4, respectively. These results suggest that the membrane-bound enzymes may also catalyze the hydrolysis of temocapril inside the cell, depending on its intracellular concentration. Consequently, CES must be responsible for approximately 50% of the hydrolysis of temocapril in rat intestine. This is the first time that esterases other than CES have been shown to be involved in the intestinal hydrolysis of ACE inhibitors. It is expected that membrane enzymes are also present in human intestinal cells, and their contribution may be larger than CES-mediated hydrolysis, as might be expected from the extremely low levels of hydrolysis of temocapril in human intestinal S9.
The present data suggest that temocapril is absorbed in the proximal intestine. Its absorption will be increased by administration after meals, due to the decrease of pH caused by secretion of bile acids, the prolongation of gastric emptying times, and the interaction of serine proteases on the cell membrane with several compounds contained in food. Under these conditions, temocapril may even be absorbed in an intact form, without hydrolysis, in man. In the present study, we were able to mimic the conditions of the human intestine in a rat perfusion model by inhibiting CES activity. In addition, we demonstrated the involvement of serine-containing hydrolases on the cell membrane in the hydrolysis of prodrugs. We will need to characterize these enzymes further to optimize the use of this model in predicting the intestinal absorption properties of prodrugs in man.
Participated in research design: Imai.
Conducted experiments: Nozawa.
Contributed new reagents or analytic tools: Imai and Nozawa.
Performed data analysis: Nozawa.
Wrote or contributed to the writing of the manuscript: Imai and Nozawa.
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
- angiotensin-converting enzyme
- area under the curve
- bis-p-nitrophenyl phosphate
- bovine serum albumin
- diisopropyl fluorophosphonate
- dimethyl sulfoxide
- fluorescein isothiocyanate dextran 4000
- high-performance liquid chromatography.
- Received December 24, 2010.
- Accepted April 7, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics