Effects of Cilastatin on the Pharmacokinetics of a New Carbapenem, DA-1131, in Rats, Rabbits, and Dogs

Abstract

DA-1131, a new carbapenem antibiotic, undergoes renal metabolism by renal dehydropeptidase I (DHP-I), located on the brush border of the proximal tubular cell. Species differences with regard to the effects of cilastatin, a renal DHP-I inhibitor, were investigated after a 1-min intravenous infusion of DA-1131, with or without cilastatin, to rats, rabbits, and dogs. After intravenous infusion, the nonrenal clearance (CL_NR) of DA-1131 was significantly slower in rats (3.00 versus 8.01 ml/min/kg) and rabbits (2.41 versus 6.77 ml/min/kg) when the drug was coadministered with cilastatin; this could be due to the slower metabolism of DA-1131 by rat and rabbit kidney DHP-I. This indicated that renal metabolism of DA-1131 by renal DHP-I was inhibited by cilastatin. However, coadministration with cilastatin to dogs did not affect the CL_NR of DA-1131.

(1R,5S,6S)-(2S,4S)-2-[(E)-3-Methansulfonylamino-1-propenyl]pyrrolidine-4-ylthiol-6-[(R)-1-hydroxyethyl]-1-methyl-1-carbapen-2-em-3-carboxylic acid (DA-1131), a new anionic carbapenem antibiotic, has a broad spectrum of activity against both gram-positive and gram-negative organisms (7). DA-1131 is resistant to degradation by various types of β-lactamases (4). DA-1131 is relatively stable against hydrolysis by ICR mouse, Sprague-Dawley rat, New Zealand White rabbit, beagle dog, and human renal dehydropeptidase I (DHP-I), located on the brush border of the proximal tubular cell, compared with imipenem and meropenem (8). Judging from the maximum velocity–to–Michaelis-Menten constant (V_max/K_m) ratios, DA-1131 shows relatively greater resistance to mouse, rat, rabbit, dog, and human DHP-I than imipenem or meropenem; the ratios of DA-1131 for resistance to DHP-I were from 1.3 to 4.6 times higher than those of imipenem and meropenem (unpublished data). The following have been reported on by our laboratory: high-performance liquid chromatographic analysis of DA-1131 in biological fluids (18); the stability of DA-1131, its metabolism and distribution in tissues, and its partitioning in blood (10); the pharmacokinetics of DA-1131 in animals (17); interspecies pharmacokinetic scaling of DA-1131 (12); the mechanism of renal excretion of DA-1131 in rats, rabbits (15), and dogs (14); and the pharmacokinetics of DA-1131 in rats with uranyl nitrate-induced acute renal failure (19), alloxan-induced diabetes mellitus (13), or hypertension (16) and in rabbits with endotoxin-induced pyrexia (11). DA-1131 is now being evaluated in a preclinical study.

The low-level urinary recovery of unchanged imipenem in laboratory animals suggested that this antibiotic was extensively metabolized by renal DHP-I; this resulted in a reduced antimicrobial activity and increased side effects induced by its metabolites (20). Very high doses of imipenem were reported (22) to induce renal tubular toxicity in rabbits, but this effect could be blocked by concomitant administration of cilastatin, a renal DHP-I inhibitor. Therefore, in order to develop new carbapenem antibiotics, it is necessary to establish their stability against DHP-I. The rat kidney showed high metabolic activity for DA-1131 in an in vitro tissue homogenate study (10). The purpose of the present study was to report the effects of cilastatin on the pharmacokinetics of DA-1131 in rats, rabbits, and dogs.

Male Sprague-Dawley rats of 8 weeks of age (weight, 245 to 310 g), male New Zealand White rabbits (weight, 1.8 to 3.1 kg), and male conditioned beagle dogs (weight, 8.5 to 10.5 kg) were purchased from Charles River Company (Atsugi, Japan), Korea Laboratory of Animal Development (Seoul, Korea), and Marshall Farms (New York, N.Y.), respectively. The animals were housed in a light-controlled room (Animal Center for Pharmaceutical Research, College of Pharmacy, Seoul National University, Seoul, Korea, and Research Laboratory, Dong-A Pharmaceutical Company, Yongin, Korea) kept at a temperature of 22 ± 1°C and a humidity of 55% ± 10%, with food (Samyang Company, Seoul, Korea) and tap water provided ad libitum.

The pretreatment and surgical procedures were similar to previously reported methods for rats and rabbits (15) as well as dogs (14). DA-1131 (obtained from Dong-A Pharmaceutical Company, Yongin, Korea, as an HCl salt powder and dissolved in an injectable normal saline solution; treatment I [n = 13]) or a DA-1131–cilastatin (the latter donated by Merck, Sharp & Dohme Research Laboratories, Rahway, N.J.) mixture (1:1 ratio [23], dissolved in an injectable normal saline solution; treatment II [n = 10]), both at 200 mg of DA-1131/kg of body weight, was administered intravenously to rats over a 1-min period via the jugular vein. The total injection volume was approximately 1 ml. Approximately 0.12-ml volumes of blood were collected via the carotid artery at 0 (to serve as a control), 1 (at the end of the infusion), 5, 15, 30, 45, 60, 90, 120, 180, 240, and 360 min after intravenous administration of the drug. Urine was collected over an 8-h period. Other procedures were similar to previously reported methods (15).

DA-1131 (treatment III, n = 10) or DA-1131–cilastatin (1:1 mixture [23]; treatment IV, n = 10), at 50 mg of DA-1131/kg, was administered intravenously to rabbits over a 1-min period via the jugular vein. The total injection volume was approximately 1 ml. Approximately 0.25-ml volumes of blood were collected via the carotid artery at 0 (to serve as a control), 1 (at the end of the infusion), 5, 15, 30, 45, 60, 90, 120, 180, 240, and 360 min after intravenous administration of the drug. Urine was collected over an 8-h period. Other procedures were similar to previously reported methods (15).

DA-1131 (treatment V) or DA-1131–cilastatin (1:1 ratio [23]; treatment VI), at 50 mg of DA-1131/kg, was administered intravenously to male dogs (n = 6) over a 1-min period via the cephalic vein (the total injection volume was approximately 10 ml) by parallel design. Approximately 2.5-ml volumes of blood were collected via the other cephalic vein at 0 (to serve as a control), 1 (at the end of the infusion), 5, 15, 30, 45, 60, 90, 120, 180, 240, 300, 360, 420, and 480 min after intravenous administration of the drug. Urine was collected over an 8-h period. Other procedures were similar to those reported previously (21). The DA-1131 in the biological samples described above was analyzed within 7 days by the previously reported high-performance liquid chromatographic method developed by our laboratory (18).

Standard methods (6) were used to calculate the following pharmacokinetic parameters: the total area under the plasma concentration-time curve from time zero to infinity (AUC_0–∞) (1), the time-averaged total body clearance (CL), the area under the first moment of the plasma concentration-time curve (AUMC_0–∞), the mean residence time (MRT), the apparent volume of distribution at steady state (V_ss), and the time-averaged renal (CL_R) and nonrenal (CL_NR) clearances (9). The mean values of CL_R and CL_NR (3), V_ss (2), and the terminal half-life (t_1/2) (5) were calculated by the harmonic mean method.

Levels of statistical significance were assessed by the t test between two means for unpaired (rat and rabbit studies) and paired (dog study) data. Significant differences were judged as a P value of less than 0.05. All results are expressed as means ± standard deviations.

The mean arterial plasma concentration-time profiles of DA-1131 in rats (treatments I and II) are shown in Fig. 1A, and some relevant pharmacokinetic parameters are listed in Table 1. After intravenous administration of the drug to rats, the plasma concentrations of DA-1131 declined in a polyexponential fashion for both treatment groups and were significantly higher for treatment II than for treatment I (Fig. 1A). The significantly higher plasma concentrations (Fig. 1A) and the significantly greater AUC_0–∞ (32% increase) of DA-1131 in rats when given in combination with cilastatin (treatment II) could be due to a significantly slower CL of DA-1131 (24% decrease) in treatment II (Table 1). The significantly slower CL in treatment II was due to a significantly slower CL_NR (63% decrease) with this treatment, because the CL_R values for the two rat groups were not significantly different (Table 1). The significantly slower CL_NR in treatment II could be due to a considerably slower metabolism of DA-1131 by DHP-I in the rat kidney in the presence of cilastatin. The above data indicated that the metabolism of DA-1131 by rat renal DHP-I was inhibited by cilastatin. This was supported by a significant increase in the percentage of the intravenous dose of DA-1131 excreted in urine over an 8-h period as unchanged drug (56% increase) in treatment II (Table 1). The CL_R values for the two treatments were not significantly different (Table 1), indicating that the CL_R of DA-1131 was not affected by cilastatin. The V_ss of DA-1131 was significantly larger (32% increase) in treatment II. The exact reason for this is not clear; however, it was not due to an increase in the unbound fraction of DA-1131 in plasma due to cilastatin, since the level of plasma protein binding of DA-1131 in the rat was less than 10% (17). The significantly slower CL and significantly larger V_ss of DA-1131 in treatment II resulted in a significantly longer terminal t_1/2 and MRT (Table 1).

FIG. 1.

Mean arterial plasma concentration-time profiles of DA-1131 after a 1-min intravenous infusion of the drug, at 200 mg/kg, without (○) (n = 13) or with (●) (n = 10) cilastatin (at 200 mg/kg) to rats (A) or, at 50 mg/kg, without (○) (n = 10) or with (●) (n = 10) cilastatin (at 50 mg/kg) to rabbits (B), and corresponding venous profiles of DA-1131, at 50 mg/kg, without (○) or with (●) cilastatin (at 50 mg/kg) (n = 6, by parallel design) to dogs (C). Bars represent standard deviations. ∗, P < 0.05; ∗∗, P < 0.01; ∗∗∗, P < 0.001.

TABLE 1.

Pharmacokinetic parameters of DA-1131

Treatment group	Pharmacokinetic parametera
	Terminal t1/2 (min)	AUC0–∞ (μg · min/ml)	AUMC0–∞ (μg · min2/ml)	MRT (min)	CL (ml/min/kg)	CLR (ml/min/kg)	CLRN (ml/min/kg)	Vss (ml/kg)	Xug (% of i.v. dose)
Ratsb
Treatment I	15.2 ± 3.15	13,900 ± 2,480	167,000 ± 43,300	12.0 ± 2.08	14.3 ± 2.79	6.07 ± 1.60	8.01 ± 1.64	167 ± 51.5	43.8 ± 6.06
Treatment II	27.6 ± 2.94d	18,300 ± 1,930d	381,000 ± 94,700d	20.8 ± 4.14d	10.9 ± 1.20f	7.19 ± 1.34	3.00 ± 1.92d	221 ± 44.7e	68.2 ± 13.7d
Rabbitsc
Treatment III	19.1 ± 6.95	3,880 ± 1,240	42,700 ± 23,000	10.8 ± 3.16	12.9 ± 5.17	6.34 ± 2.77	6.77 ± 3.78	128 ± 59.2	45.4 ± 12.4
Treatment IV	15.9 ± 3.09	10,100 ± 1,160d	141,000 ± 31,400d	13.9 ± 2.08e	4.96 ± 0.567d	2.11 ± 0.574d	2.41 ± 0.970f	68.1 ± 10.9f	47.5 ± 14.5
Dogsc
Treatment V	47.2 ± 5.70	7,680 ± 1,170	338,000 ± 74,400	43.8 ± 3.88	6.51 ± 0.945	3.78 ±0.492	2.61 ± 0.780	285 ± 35.3	58.3 ± 7.35
Treatment VI	48.2 ± 3.75	9,320 ± 817e	465,000 ± 58,700d	49.8 ± 3.96f	5.37 ± 0.508e	2.60 ± 1.00	2.28 ± 0.783	266 ± 28.6	54.8 ± 15.1

^aValues are means ± standard deviations. 

^bRats were given a 1-min intravenous infusion of DA-1131 (200 mg/kg) without (treatment I) or with (treatment II) cilastatin (200 mg/kg). 

^cRabbits and dogs were given a 1-min intravenous infusion of DA-1131 (50 mg/kg) without (treatments III and V, respectively) or with (treatments IV and VI, respectively) cilastatin (50 mg/kg). 

^dP < 0.001. 

^eP < 0.05. 

^fP < 0.01. 

^gX_u, drug amount in urine. 

The mean arterial plasma concentration-time profiles of DA-1131 in rabbits (treatments III and IV) are shown in Fig. 1B, and, again, some relevant pharmacokinetic parameters are listed in Table 1. After intravenous administration of the drug to rabbits, the plasma concentrations of DA-1131 declined in a polyexponential fashion for both treatment groups and were significantly higher in treatment IV than in treatment III (Fig. 1B). The significantly higher plasma concentrations (Fig. 1B) and the significantly greater AUC_0–∞ (160% increase) for DA-1131 when given in combination with cilastatin (treatment IV) to rabbits could be due to a significantly slower CL of DA-1131 (62% decrease) in treatment IV (Table 1). The significantly slower CL in treatment IV was due to a significantly slower CL_R (67% decrease) and CL_NR (64% decrease) in treatment IV (Table 1). In the previous rabbit studies (15, 17), it was found that DA-1131 was excreted in urine via glomerular filtration and active secretion. The CL_R of DA-1131 was significantly slower when given in combination with probenecid (15), indicating that active renal secretion of DA-1131 was inhibited by probenecid in rabbits. The significantly slower CL_R of DA-1131 in treatment IV could be due to inhibition of active renal secretion by cilastatin. It was also reported (22) that cilastatin inhibited active renal secretion of imipenem by competitively inhibiting the penetration of imipenem into the proximal tubular cells. The significantly slower CL_NR in treatment IV could be due to a considerably slower metabolism of DA-1131 in the rabbit kidney, indicating that the metabolism of DA-1131 by rabbit renal DHP-I was at least partly inhibited by cilastatin.

The mean venous plasma concentration-time profiles of DA-1131 in dogs (treatments V and VI) are shown in Fig. 1C, and some relevant pharmacokinetic parameters are also listed in Table 1. After intravenous administration of the drug to dogs, the plasma concentrations of DA-1131 declined in a polyexponential fashion for both treatment groups and were significantly higher in treatment VI than in treatment V (Fig. 1C). The significantly higher plasma concentrations (Fig. 1C) and the significantly greater AUC_0–∞ (21% increase) of DA-1131 when given in combination with cilastatin (treatment VI) to dogs could be due to a significantly slower CL of DA-1131 (18% decrease) in treatment VI (Table 1). Unlike those in rats and rabbits (Table 1), CL_NR values of DA-1131 for the two treatments were not significantly different (Table 1). In the previous dog studies (14), it was found that DA-1131 was excreted in urine via glomerular filtration and tubular reabsorption. The CL_R values for DA-1131 in the two treatment groups were comparable, suggesting that the tubular reabsorption of DA-1131 in dogs was not affected by cilastatin. The tubular reabsorption of DA-1131 in dogs was also not affected by probenecid (14).

In conclusion, species differences in the metabolism of DA-1131 by renal DHP-I were observed. Coadministration with cilastatin caused a significantly slower CL_NR of DA-1131 in rats by inhibition of renal DHP-I, and this resulted in a significant increase in urinary excretion of DA-1131 in these animals. Cilastatin also caused a significantly slower CL_NR of DA-1131 by inhibition of renal DHP-I and a significantly slower CL_R by inhibition of the tubular secretion of DA-1131 in rabbits. However, coadministration with cilastatin did not affect the CL_NR of DA-1131 in dogs. The above data suggested that the stability of DA-1131 to renal DHP-I varied widely in the three animal species studied.