- Email: [email protected]
Transport of levovirin prodrugs in the human intestinal Caco-2 cell line
Transport of Levovirin Prodrugs in the Human Intestinal Caco-2 Cell Line FUJUN LI,1 LEI HONG,1 CHENG-I MAU,2 REBECCA CHAN,2 THAN HENDRICKS,3 CHUCK DVORAK,4 CALVIN YEE,3 JASON HARRIS,3 TOM ALFREDSON1 1
Pharmaceutics, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
2
Pharmacokinetics and Drug Metabolism, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
3
Medicinal Chemistry, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
4
Chemical Development, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
Received 9 May 2005; revised 27 May 2005; accepted 31 May 2005 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20434
ABSTRACT: The transport of 10 amino acid ester prodrugs of levovirin (LVV) was investigated in the human intestinal Caco-2 cell line in order to overcome the poor oral bioavailability of LVV, an investigational drug for the treatment of hepatitis C infection. The prodrugs were designed to improve the permeability of LVV across the intestinal epithelium by targeting the di/tri-peptide carrier, PepT1. Caco-2 cell monolayers were employed to study the transport and hydrolysis properties of the prodrugs. Among all mono amino acid ester prodrugs studied, the LVV-50 -(L)-valine prodrug (R1518) exhibited the maximum increase (48-fold) in permeability with nearly complete conversion to LVV within 1 h. Di-amino acid esters did not offer significant enhancement in permeability comparing with mono amino acid esters and exhibited slower conversion to LVV in Caco2 cell monolayers. Pharmacokinetic screening studies of the prodrugs in rats yielded the highest fold increase (6.9-fold) of AUC with R1518 and in general displayed a similar trend to that observed in increases of permeability in Caco-2 cells. Mechanisms involved in the Caco-2 cell transport of R1518 were also investigated. Results of bi-directional transport studies support the involvement of carrier-mediated transport mechanisms for R1518, but not for the LVV-50 -(D)-valine prodrug or LVV. Moreover, the permeability of R1518 was found to be proton dependent. PepT1-mediated transport of R1518 was supported by results of competitive transport studies of R1518 with the PepT1 substrates enalapril, Gly-Sar, valganciclovir, and cephalexin. R1518 was also found to inhibit the permeability of valganciclovir and cephalexin. These results suggest that R1518 is a PepT1 substrate as well as an inhibitor. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 95:1318–1325, 2006
Keywords: levovirin; prodrugs; hepatitis C; Caco-2 cells; transport; permeability; peptide transporters
INTRODUCTION Levovirin (1-b-L-ribofuranosyl-1,2,4-triazole-3carboxamide) is a guanosine nucleoside analog and the L-enantiomer of ribavirin. Levovirin Correspondence to: Fujun Li (Telephone: 650-852-3113; Fax: 650-855-5172; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 95, 1318–1325 (2006) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
1318
(LVV) is an investigational drug for the treatment of hepatitis C virus-mediated diseases. The combination of ribavirin and interferon is the current first line therapy for the treatment of chronic hepatitis C.1 LVV has similar immunomodulatory potency to ribavirin in vitro without accumulating in red blood cells or causing hemolytic anemia, a known side effect of ribavirin.2 LVV exhibits poor oral bioavailability in rat and monkey (15% and 17%, respectively)3 and in
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
TRANSPORT OF LEVOVIRIN PRODRUGS
human clinical studies (estimated apparent oral bioavailability of 10%).4 It has been suggested that the low oral bioavailability of LVV is due to its limited permeability across the intestinal epithelium. Although ribavirin was reported to be actively transported by the human intestinal N1 sodium-dependent concentrative nucleoside transporters,5 LVV is not likely to be recognized by these nucleoside transporters since it contains the nonnatural L-ribose moiety. Caco-2 cell monolayers have been found to lack the concentrative NT1 and NT2 nucleoside transporters.6 Prodrug strategies have been employed to optimize physicochemical properties of poorly absorbed compounds for improving drug delivery.7–9 One successful prodrug approach for improved intestinal absorption exploits active transport systems to move the prodrug across the intestinal membrane. The promoiety portion of the molecule is designed to confer recognition by the active transport system and is cleaved after transport is complete to yield the active compound. The intestinal di/tri-peptide transporter, PepT1 has been investigated as a target for improving the bioavailability of poorly-absorbed drugs due to its broad substrate specificity.10–12 Valine esters of acyclovir (valacyclovir) and ganciclovir (valganciclovir) have been shown to exhibit improved absorption, which is attributed to the uptake via a peptide transporter even though there is no peptide bond in the structures.13–16 For example, valganciclovir exhibited a 10-fold increase in plasma ganciclovir concentrations compared to oral formulations of ganciclovir.17 Valganciclovir was found to be recognized as a substrate by the intestinal transporter PepT1.16 In the present investigation, the permeability of 10 amino acid ester prodrugs of LVV (see Fig. 1) was determined in human intestinal Caco-2 cells. The transport mechanisms of the lead prodrug, R1518 were also studied. Furthermore, the inhibitory effects of known PepT1 inhibitors on the permeability of R1518 and the reverse inhibitory effect of R1518 on PepT1 substrates were investigated. Pharmacokinetic screening studies in rats were carried out for LVV and the prodrugs.
MATERIALS AND METHODS Materials Caco-2 cells (passage 108–120) were obtained from Roche Basel. 12-well Transwell1 inserts DOI 10.1002/jps
H2N
N
O
N
N
O R3O
1319
5' 3'
R 2O
2'
OR1
Figure 1. Levovirin (R1, R2, R3 ¼ H) and amino acid ester prodrugs: 50 -(L)-valinate [R1, R2 ¼ H, R3 ¼ (L)-Val]; 50 -(D)-valinate [R1, R2 ¼ H, R3 ¼ (D)-Val]; 50 -(L)-isoleucinate [R1, R2 ¼ H, R3 ¼ (L)-Ile]; 50 -(L)-alaninate [R1, R2 ¼ H, R3 ¼ (L)-Ala]; 50 -(L)-leucinate [R1, R2 ¼ H, R3 ¼ (L)-Leu]; 50 -(L)-sarcosinate [R1, R2 ¼ H, R3 ¼ (L)Sar]; 50 -(L)-phenylalaninate [R1, R2 ¼ H, R3 ¼ (L)-Phe]; 20 ,30 -(L)-bis-valinate [R1, R2 ¼ (L)-Val, R3 ¼ H]; 50 -(L)valinylprolinate [R1, R2 ¼ H, R3 ¼ (L)-Val-Pro]; 50 -(L)prolinylvalinate [R1, R2 ¼ H, R3 ¼ (L)-Pro-Val].
(diameter: 12 mm) with collagen-coated polytetrafluoroethylene membrane (0.4 mm pores) and T225 cm2 cell culture flask (tissue culture treated) were purchased from Corning Incorporated (Corning, NY). Dulbecco’s Modified Eagle Media with high glucose and L-glutamine (DMEM), Lglutamine, penicillin-streptomycin, nonessential amino acids and fetal bovine serum were obtained from Gibco/Life Technologies (Gaithersburg, MD). Krebs–Henseleit bicarbonate buffer mix, calcium chloride dihydrate, enalapril, glycylsarcosine (Gly-Sar), and cephalexin were purchased from Sigma (St. Louis, MO). Sodium bicarbonate was obtained from Mallinckrodt Chemicals (Phillipsburg, NJ). Nanopure water was used for the buffer preparation. Amino acid ester prodrugs of LVV were generally prepared via a three or more step synthesis from LVV. The 50 -monoester syntheses utilized either a 20 ,30 -LVV cyclopentylidene or isopropylidene ketal intermediate which allowed selective esterification with the N-butoxycarbonyl (N-Boc) protected amino acids valine, isoleucine, alanine, leucine, sacrosine, and phenylalanine. The 20 , 30 bis-valinate prodrug was synthesized using the JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1320
LI ET AL.
50 -mono-triisopropylsilyl ether (TIPS) protected intermediate followed by esterification with N-Cbz (benzyloxycarbonyl)-valine and subsequent hydrogenolysis. LVV prodrug esters were characterized by elemental analysis, NMR, and HPLC-MS analyses.
Cell Culture High passage (108–120) Caco-2 cells were cultured in Dulbecco’s Modified Eagle Media with high glucose and L-glutamine (DMEM) supplemented with 10% fetal bovine serum, 1 L-glutamine, 1 penicillin-streptomycin, and 1 nonessential amino acids. Cells were maintained in T225 cm2 cell culture flask (tissue culture treated) at 378C and 5% CO2. For transport experiments, cells were plated at 7.1 104 cells/well onto 12-well collagen-coated PTFE membrane inserts. Cells were fed every 3 days and maintained at 378C and 5% CO2 for 21 days to allow complete formation of a polarized monolayer with tight junctions.
Transport Studies The Krebs–Henseleit bicarbonate buffer solution for transport studies were prepared by following the manufacturer’s instruction. The buffer was adjusted to pH 6.5 or 7.4 prior to use. Stock solutions of test compounds were prepared at 5 mg/mL concentrations in DMSO and stored at 48C. An aliquot of the 5 mg/mL stock solution was diluted with either pH 6.5 or pH 7.4 Krebs– Henseleit bicarbonate buffer to obtain a concentration of 20 mM for use as initial donor dosing solution (D0). Buffers, drug solutions, and Caco-2 cell plates containing buffers were prewarmed to 378C before use. The transepithelial electrical resistance (TEER) values were measured using a millicell1-ERS equipped with ‘‘chopstick’’ electrodes (Millipore, Bedford, MA). The inserts with TEER values above 300 ohm cm2 were used in the study after washing with warm Krebs–Henseleit bicarbonate buffer. For transport studies from apical to basolateral side, 0.5 mL Krebs–Henseleit buffer (pH 6.5) was added to apical side of the cell monolayers and 1.25 mL Krebs–Henseleit buffer (pH 7.4) to the basolateral chamber. The cells were equilibrated at 378C and 5% CO2 in an incubator for at least 30 min. The apical side buffer was removed and replaced with 0.5 mL of 20 mM drug solution. The JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
cells were then incubated at 378C and 5% CO2. At 30, 60, and 90 min time points, the inserts were transferred to new plates with 1.25 mL fresh warm pH 7.4 buffer at the receiver side. The media from all plates were collected as receiver samples. For bi-directional transport studies from basolateral to apical side, 1.25 mL of 20 mM drug solution in pH 7.4 buffer was placed in the basolateral chamber, and 0.5 mL pH 6.5 buffer was put into the apical chamber. At 30, 60, and 90 min time points, the samples in the apical side were collected and then replaced with 0.5 mL fresh warm pH 6.5 buffer. After 60 min of transport studies, Lucifer Yellow (0.05 mL 1000 mM) was added to the apical side of the wells. At the end of the transport studies (90 min), the fluorescence of the receiver side samples was measured. Sample solutions from the donor side were collected at the end of the experiments as D90 samples. For inhibitory studies, the study compound concentration was 20 mM, whereas the concentration of the inhibitors was 20 mM. All prodrug samples were acidified to approximately pH 3.5 with formic acid solution immediately after collection to prevent further chemical hydrolysis. Chemical Stability Study of R1518 Solutions containing 20 mM of R1518 were prepared using Krebs–Henseleit buffers at pH 6.5 and at pH 7.4. The solutions were kept in a 378C/ 5% CO2 incubator for 2 h. Samples were taken at 30, 60, 90, and 120 min to check the conversion of R1518 to LVV. Pharmacokinetic Screening Studies in Rats Oral administration of single 10-mg/kg doses of LVV and equivalent LVV doses of prodrugs were given to nonfasted male Hanover–Wistar rats. Animals were allowed access to water and food on their normal schedule. LVV and prodrugs were administered as solutions in water or buffered saline. Blood samples were collected at 0.25, 0.5, 1, 2, 4, 6, 8, 10, and 24 h after dosing. Plasma concentrations of LVV were determined by a HPLC/MS/MS method as described in the sample analysis after plasma sample extraction. Sample Analysis LVV and its prodrugs were analyzed using highperformance liquid chromatography with tandem mass-spectrometry (HPLC/MS/MS). An Agilent DOI 10.1002/jps
TRANSPORT OF LEVOVIRIN PRODRUGS
1321
Table 1. Caco-2 Permeability Studies of Levovirin and Its Prodrugs % Conversion Ratio at 90 min Levovirin= Levovirin þ Prodrug Compound
Caco-2 Permeability (106 cm/s)a
Permeability Fold Increase
Apical Side
Basolateral Side
0.12 0.06 0.18 0.06 5.70 1.99 0.52 0.26 1.46 1.09 1.17 0.57 0.88 0.17 0.50 0.39 0.14 0.07 8.55 1.75 3.42 0.18 1.98 1.04
1.5 47.9 4.4 12.4 9.9 7.5 4.3 1.2 72.5 29.0 16.8
7.6 11.0 5.9 41.2 17.4 34.9 61.7 53.4 99.1 5.7
93.8 2.7 72.8 100 80.2 100 67.3 12.7 24.1 45.9
Levovirin Ribavirin LVV-50 -(L)-valine (R1518) LVV-50 -(D)-valine LVV-50 -(L)-isoleucine LVV-50 -(L)-alanine LVV-50 -(L)-leucine LVV-50 -(L)-sarcosine LVV-50 -(L)-phenylalanine LVV-20 ,30 -(L)-bis-valine LVV-50 -val-pro LVV-50 -pro-val a
Mean SD, n ¼ 18 for levovirin, n ¼ 16 for R1518, n ¼ 4 for ribavirin, and LVV-50 -(D)-valine, and n ¼ 2 for others.
Zorbax SB-Aq 4.6 50 mm column (5 mM) was used for separation. Electrospray ionization (ESI) was used for the ionization process. The mobile phase A contained 5 mM ammonium formate in water and mobile phase B contained methanol. Elution was performed with 98% A and 2% B for 3 min, then with linear gradient to 10% A and 90% B in 1 min and held at this condition for 2 min with a flow rate of 0.5 mL/min. Data Analysis for Permeability Studies in Caco-2 Cells The dQ/dt of test substance was calculated from sampling data at 30, 60, and 90 min. The apparent permeability coefficient (Papp) was calculated from the following equation, Papp ¼
dQ 1 dt A Co
where A is the surface area (cm2) of the insert, Co is the initial concentration of the drug, and dQ/dt is the change in total amount of the prodrug and its parent drug (LVV) in the receiver solution over the 90 min incubation time, that is, the slope of the drug amount in the receiver solution versus time.
RESULTS AND DISCUSSION Permeability of Levovirin and Its Prodrugs The apical (A) to basolateral (B) permeability of 10 LVV prodrugs was determined in Caco-2 cells and the results were compared with the DOI 10.1002/jps
parent drug, LVV (Table 1). LVV has a very low transport rate through Caco-2 cell monolayers (1.2 107 cm/s). Enhanced permeability was achieved with the majority of LVV prodrugs investigated in this study. R1518 showed the maximum improvement in permeability (48-fold increase) among all mono amino acid ester prodrugs. In contrast, only a slight increase (fourfold) in permeability was observed for LVV-50 -D-valinate prodrug, suggesting that a carrier-mediated mechanism is involved in the intestinal transport of R1518, but not LVV-50 -(D)-valinate.18 The fourfold increase in permeability for LVV-50 -(D)-valinate is likely due to the increase in hydrophobicity via passive diffusion. The transport of the 20 ,30 -Lbis-valine ester prodrug increased 72-fold compared with LVV. However, only approximately 13% prodrug converted to the parent drug in the basolateral side within 90 min. The low permeability of ribavirin through Caco-2 cell monolayers is probably due to the lack of the concentrative N1 transporter and sodium gradient in the transport system.5 Table 2. Chemical Stability of R1518 at 378C
Time (min) 30 60 90 120
% R1518 Remaining in pH 6.5 Buffer
% R1518 Remaining in pH 7.4 Buffer
97.8 96.1 92.7 88.1
96.9 92.8 88.7 83.7
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1322
LI ET AL.
Table 3. Rat Pharmacokinetic Screening Studies at 10 mg/kg Levovirin Equivalent of Levovirin Prodrugs
Compound
AUC0-inf of Levovirin (mg h/mL)a
AUC Fold Increase
1.70 0.62 11.70 0.61 9.70 1.35 6.58 3.06 2.41 0.60 1.01 0.20 1.75 0.50 5.22 0.42 1.38 0.21 3.85 1.11
— 6.9 5.7 3.9 1.4 0.6 1.0 3.1 0.8 2.3
Levovirin LVV-50 -(L)-valine (R1518) LVV-50 -(L)-isoleucine LVV-50 -(L)-alanine LVV-50 -(L)-leucine LVV-50 -(L)-sarcosine LVV-50 -(L)-phenylalanine LVV-20 ,30 -(L)-bis-valine LVV-50 -val-pro LVV-50 -pro-val
a Mean SD, n ¼ 3; AUC0-inf: area under the curve from time zero to infinity; Nonfasted male Wistar rats; Dose administration: oral gavage.
R1518 converted to LVV almost completely in basolateral side at the end of the transport study, whereas less than 3% of LVV-50 -(D)-valine converted to LVV. Chemical stability study showed that more than 88% of R1518 remained intact in either apical or basolateral side solution after 90 min of incubation at 378C (Table 2), indicating that ester hydrolysis occurred within the epithelial cells.13 These results support that both carriermediated transport and ester hydrolysis in Caco-2 cell monolayers are stereoselective towards Lamino acid esters.18,19 R1518 was selected as the lead prodrug candidate due to its high permeability and fast hydrolysis to the parent drug. It has been evaluated in clinical trials for the treatment of hepatitis C. Consistent enhancement in oral bioavailability (AUC) for most of the LVV prodrugs was observed in rat pharmacokinetic screening studies (Table 3) with the highest increase (6.9-fold) observed for R1518. In general, the pharmacokinetic screening studies in rats displayed a similar trend to that observed among the mono amino acid prodrugs for increases in permeability in Caco-2 cells. For di-
amino acid prodrugs, there was only a moderate improvement in AUC in rats compared to corresponding increase in Caco-2 permeability, which is probably due to slow conversion of prodrugs to LVV in rats. The pharmacokinetic results in rats are also consistent with that from human single- and multiple-dose pharmacokinetics studies of R1518 in healthy volunteers which yielded significantly improved oral absorption of LVV upon oral administration.4 Plasma exposures (AUC) of LVV dosed as R1518 in human exhibited an eightfold increase versus LVV administered orally.4
Bi-Directional Transport Studies of LVV and Its Prodrugs A to B and B to A permeability of LVV and its prodrugs were determined in Caco-2 cells with a pH gradient of apical side pH 6.5 and basolateral side pH 7.4. The results are shown in Table 4. There are no significant differences between A to B and B to A permeability values for LVV and
Table 4. Bi-Directional Transport Study of Levovirin and Valine Ester Prodrugs with pH Gradient (Apical pH 6.5, Basolateral pH 7.4) Permeability (106 cm/s)a Compound
A to B
Levovirin 0.12 0.06 (n ¼ 18) LVV-50 -(L)-valine (R1518) 5.70 1.99 (n ¼ 16) LVV-50 -(D)-valine 0.52 0.26 (n ¼ 4) a
B to A
A to B= B to A Ratio
0.21 0.01 (n ¼ 2) 0.14 0.06 (n ¼ 4) 0.44 0.01 (n ¼ 2)
0.56 40.7 1.18
Mean SD.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
DOI 10.1002/jps
TRANSPORT OF LEVOVIRIN PRODRUGS
Table 5. Bi-Directional Transport Study of R1518 without pH Gradient (Apical and Basolateral pH 7.4) Permeability (106 cm/s)a A to B 4.13 0.82 (n ¼ 4) a
A to B= B to A Ratio
B to A 0.16 0.04 (n ¼ 2)
26.5
Mean SD.
LVV-50 -(D)-valine, respectively, indicating that their transport through intestinal epithelium is mainly by passive diffusion. A to B transport of R1518 is approximately 40-fold higher than that of B to A, suggesting that a carrier-mediated transport is involved in the intestinal absorption.15 In addition, a bidirectional transport study of R1518 was performed without pH gradient (pH 7.4 at both apical and basolateral sides). As shown in Table 5, the B to A permeability of R1518 did not change significantly. However, the A to B permeability decreased approximately 30% resulting in lower A to B=B to A ratio. The pH appears to have more effect on the transport from A to B than that from B to A. pH Effect on Permeability of R1518
1323
Inhibitory Effect of Gly-Sar, Enalapril, Cephalexin and Valganciclovir on the Permeability of R1518 Human intestinal Caco-2 cell line has been shown to express PepT1 in the brush-border membranes,21,22 and it has been used as a model for studying the functions of PepT1 in the transport of peptide-like compounds.15,19 We examined the transport of R1518 across Caco-2 cell monolayers in the presence or absence of excess Gly-Sar, enalapril, cephalexin, or valganciclovir which are well known PepT1 substrates and inhibitors.16,19,23–25 As shown in Figure 3, the permeability of R1518 (Lev-Val prodrug) was reduced 88% by enalapril, 80% by valganciclovir, 80% by Gly-Sar, and 52% by cephalexin, indicating that R1518 competes with enalapril, valganciclovir, Gly-Sar, or cephalexin ( p < 0.01) for the same carrier-mediated transport pathway. These results suggest that R1518 is a substrate of PepT1. Cephalexin appears to be a weaker inhibitor than enalapril, valganciclovir, and Gly-Sar for PepT1.
Inhibitory Effect of R1518 on the Permeability of Valganciclovir, Enalapril, and Cephalexin
A to B permeability of R1518 was determined at apical side pH 5.0, 6.5, 7.4, and fixed basolateral side pH 7.4. The results in Figure 2 show that as the apical side pH increases, the A to B permeability decreases, suggesting that the transport of R1518 is pH dependent. This finding is consistent with the previous report that transport via the PepT1 transporter is coupled to a downhill movement of Hþ across brush border membrane.20
Since valganciclovir, enalapril, and cephalexin are known PepT1 substrates, their permeability through Caco-2 cell monolayers was determined in the presence or absence of excess R1518 to examine the inhibitory effect of R1518. The results in Figure 4 show that R1518 reduced the transport of valganciclovir and cephalexin by over 50% ( p < 0.01), indicating that R1518 is an inhibitor of PepT1. Due to the low permeability of enalapril through Caco-2 cell monolayers, the
Figure 2. A to B permeability of R1518 across Caco-2 cell monolayers as a function of apical pH (basolateral pH 7.4).
Figure 3. Permeability of R1518 (20 mM) in the presence or absence of enalapril, valganciclovir, Gly-Sar and cephalexin (20 mM). Mean SD, n ¼ 4 except with valganciclovir (n ¼ 2).
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1324
LI ET AL.
the cell culture facility at Roche Palo Alto for preparation of Caco-2 cell transwell plates used for these studies. Members of the LVV/prodrug Research and Development Team are gratefully acknowledged for their help and support. Colleagues at Roche Carolina, Inc., are thanked for their collaboration in LVV and R1518 synthesis and scale up.
REFERENCES Figure 4. Inhibitory effect of R1518 (20 mM) on the permeability of valganciclovir, enalapril and cephalexin (20 mM). Mean SD, n ¼ 4 except enalapril alone (n ¼ 6) and valganciclovir with R1518 (n ¼ 2).
inhibitory effect of R1518 on enalapril is not statistically significant.
CONCLUSIONS The low permeability of LVV in Caco-2 cell monolayers suggests that the low oral bioavailability in human is due to its limited absorption in the gastrointestinal tract. This study demonstrated that several amino acid ester prodrugs of LVV significantly increased permeability in Caco-2 cell monolayers via a peptide transport mechanism. The study also demonstrated that the lead prodrug, R1518 is both a substrate and an inhibitor of the PepT1 transporter. The transport process via PepT1 is pH dependent. In addition, R1518 is rapidly converted to LVV in the basolateral side of Caco2 cell monolayers. Pharmacokinetic screening studies of the prodrugs in rats displayed a similar trend of improvement in AUC compared with that observed in permeability in Caco-2 cells with the highest increase (6.9-fold) for R1518. The pharmacokinetic results in rats are also consistent with that from pharmacokinetic studies of R1518 in human healthy volunteers which yielded an eightfold increase in plasma exposures (AUC) versus LVV administered orally.4,26 The results of this investigation on LVV prodrugs provide further evidence of the feasibility of enhancing oral absorption by design of peptidelike prodrugs targeting peptide transporters.
ACKNOWLEDGMENTS The authors thank L. He and H. Liu for their support on the sample analysis by LC/MS/MS and JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1. Keating GM, Curran MP. 2003. Peginterferonalpha-2a (40 kD) plus ribavirin: A review of its use in the management of chronic hepatitis C. Drugs 63:701–730. 2. Tam RC, Ramasamy K, Bard J, Pai B, Lim C, Averett DR. 2000. The ribavirin analogue ICN 17261 demonstrates reduced toxicity and antiviral effects with retention of both immunomodulatory activity and reduction of hepatitis-induced serum alanine aminotransferase levels. Antimicrob Agents Chemother 44:1276–1283. 3. Lin CC, Luu T, Lourenco D, Yeh LT, Lau JYN. 2003. Absorption, pharmacokinetics and excretion of levovirin in rats, dogs and cynomolgus monkeys. J Antimicrob Chemother 51:93–99. 4. Huang Y, Ostrowitzki S, Hill G, Navarro M, Berger N, Kopeck P, Mau CI, Alfredson T, Lal R. 2005. Single- and multiple-pharmacokinetics of levovirin valinate hydrochloride (R1518) in healthy volunteers. J Clin Pharmacol 45:578–588. 5. Patil SD, Ngo LY, Glue P, Unadkat JD. 1998. Intestinal absorption of ribavirin is preferentially mediated by the Naþ-nucleoside purine (N1) transporter. Pharm Res 15:950–952. 6. Ward JL, Tse CM. 1999. Nucleoside transport in human colonic epithelial cell lines: Evidence for two Naþ-independent transport systems in T84 and Caco-2 cells. Biochim Biophys Acta 1419:15–22. 7. Charman WN, Porter CJH. 1996. Lipophilic prodrugs designed for intestinal lymphatic transport. Adv Drug Deliv Rev 19:149–169. 8. Fleisher D, Bong R, Stewart BH. 1996. Improved oral drug delivery: Solubility limitations overcome by the use of prodrugs. Adv Drug Deliv Rev 19:115– 130. 9. Ettmayer P, Amidon GL, Clement B, Testa B. 2004. Lessons learned from marketed and investigational prodrugs. J Med Chem 47:1–12. 10. Balimane PV, Tamai I, Guo A, Nakanishi T, Kitada H, Leibach FH, Tsuji A, Sinko PJ. 1998. Direct evidence for peptide transporter (PepT1)-mediated uptake of a nonpeptide prodrug, valacyclovir. Biochem Biophys Res Commun 250:246–251. 11. Sawada K, Terada T, Saito H, Hashimoto Y, Inui KI. 1999. Recognition of L-amino acid ester DOI 10.1002/jps
TRANSPORT OF LEVOVIRIN PRODRUGS
12.
13.
14.
15.
16.
17.
18.
19.
compounds by rat peptide transporters PEPT1 and PEPT2. J Pharmacol Exp Ther 291:705–709. Rubio-Aliaga I, Daniel H. 2002. Mammalian peptide transporters as targets for drug delivery. Trends Pharmacol Sci 23:434–440. Ganapathy ME, Huang W, Wang H, Ganapathy V, Leibach FH. 1998. Valacyclovir: A substrate for the intestinal and renal peptide transporters PEPT1 and PEPT2. Biochem Biophys Res Commun 246: 470–475. Sinko PJ, Balimane PV. 1998. Carrier-mediated intestinal absorption of valacyclovir, the L-valyl ester prodrug of acyclovir. 1. Interactions with peptides, organic anions and organic cations in rats. Biopharm Drug Dispos 19:209–217. de Vrueh RL, Smith PL, Lee CP. 1998. Transport of L-valine-acyclovir via the oligopeptide transporter in the human intestinal cell line, caco-2. J Pharmacol Exp Ther 286:1166–1170. Sugawara M, Huang W, Fei YJ, Leibach FH, Ganapathy V, Ganapathy ME. 2000. Transport of valganciclovir, a ganciclovir prodrug, via peptide transporters PEPT1 and PEPT2. J Pharm Sci 89: 781–789. Brown F, Banken L, Saywell K, Arum I. 1999. Pharmacokinetics of valganciclovir and ganciclovir following multiple oral dosages of valganciclovir in HIV- and CMV-seropositive volunteers. Clin Pharmacokinet 37:167–176. Han H, de Vrueh RL, Rhie JK, Covitz KM, Smith PL, Lee CP, Oh DM, Sadee W, Amidon GL. 1998. 50 amino acid esters of antiviral nucleosides, acyclovir, and AZT are absorbed by the intestinal PEPT1 peptide transporter. Pharmaceutical Res 15:1154– 1159. Han HK, Oh DM, Amidon GL. 1998. Cellular uptake mechanism of amino acid ester prodrugs
DOI 10.1002/jps
20.
21.
22.
23.
24.
25.
26.
1325
in Caco-2/hPEPT1 cells overexpressing a human peptide transporter. Pharm Res 15:1382–1386. Boll M, Herget M, Wagener M, Weber WM, Markovich D, Biber J, Clauss W, Murer H, Daniel H. 1996. Expression cloning and functional characterization of the kidney cortex high-affinity proton-coupled peptide transporter. Proc Natl Acad Sci USA 93:284–289. Saito H, Inui K. 1993. Dipeptide transporters in apical and basolateral membranes of the human intestinal cell line Caco-2. Ameri Physio Society 265:G289–G294. Walker D, Thwaites DT, Simmons NL, Gilbert HJ, Hirst BH. 1998. Substrate upregulation of the human small intestinal peptide transporter, hPepT1. J Physiol 507:697–706. Abe H, Satoh M, Miyauchi S, Shuto S, Matsuda A, Kamo N. 1999. Conjugation of dipeptide to fluorescent dyes enhances its affinity for a dipeptide transporter (PEPT1) in human intestinal Caco-2 cells. Bioconjugate Chem 10:24–31. Temple CS, Boyd CA. 1998. Proton-coupled oligopeptide transport by rat renal cortical brush border membrane vesicles: A functional analysis using ACE inhibitors to determine the isoform of the transporter. Biochim Biophys Acta 1373:277– 281. Chu XY, Sanchez-Castano GP, Higaki K, Oh DM, Hsu CP, Amidon GL. 2001. Correlation between epithelial cell permeability of cephalexin and expression of intestinal oligopeptide transporter. J Pharmacol ExpTher 299:575–582. Mau C, Liu H, Tran T, Young M, Ahene A, Tarnowski T, Alfredson T, Huang Y, Lal R. 2004. Pharmacokinetic studies of levovirin valinate hydrochloride (R1518) in mice, rats, monkeys, and humans. 7th International ISSX Meeting. August.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
Pharmaceutics, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
2
Pharmacokinetics and Drug Metabolism, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
3
Medicinal Chemistry, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
4
Chemical Development, Roche Palo Alto LLC, 3431 Hillview Avenue, Palo Alto, California 94304
Received 9 May 2005; revised 27 May 2005; accepted 31 May 2005 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20434
ABSTRACT: The transport of 10 amino acid ester prodrugs of levovirin (LVV) was investigated in the human intestinal Caco-2 cell line in order to overcome the poor oral bioavailability of LVV, an investigational drug for the treatment of hepatitis C infection. The prodrugs were designed to improve the permeability of LVV across the intestinal epithelium by targeting the di/tri-peptide carrier, PepT1. Caco-2 cell monolayers were employed to study the transport and hydrolysis properties of the prodrugs. Among all mono amino acid ester prodrugs studied, the LVV-50 -(L)-valine prodrug (R1518) exhibited the maximum increase (48-fold) in permeability with nearly complete conversion to LVV within 1 h. Di-amino acid esters did not offer significant enhancement in permeability comparing with mono amino acid esters and exhibited slower conversion to LVV in Caco2 cell monolayers. Pharmacokinetic screening studies of the prodrugs in rats yielded the highest fold increase (6.9-fold) of AUC with R1518 and in general displayed a similar trend to that observed in increases of permeability in Caco-2 cells. Mechanisms involved in the Caco-2 cell transport of R1518 were also investigated. Results of bi-directional transport studies support the involvement of carrier-mediated transport mechanisms for R1518, but not for the LVV-50 -(D)-valine prodrug or LVV. Moreover, the permeability of R1518 was found to be proton dependent. PepT1-mediated transport of R1518 was supported by results of competitive transport studies of R1518 with the PepT1 substrates enalapril, Gly-Sar, valganciclovir, and cephalexin. R1518 was also found to inhibit the permeability of valganciclovir and cephalexin. These results suggest that R1518 is a PepT1 substrate as well as an inhibitor. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 95:1318–1325, 2006
Keywords: levovirin; prodrugs; hepatitis C; Caco-2 cells; transport; permeability; peptide transporters
INTRODUCTION Levovirin (1-b-L-ribofuranosyl-1,2,4-triazole-3carboxamide) is a guanosine nucleoside analog and the L-enantiomer of ribavirin. Levovirin Correspondence to: Fujun Li (Telephone: 650-852-3113; Fax: 650-855-5172; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 95, 1318–1325 (2006) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
1318
(LVV) is an investigational drug for the treatment of hepatitis C virus-mediated diseases. The combination of ribavirin and interferon is the current first line therapy for the treatment of chronic hepatitis C.1 LVV has similar immunomodulatory potency to ribavirin in vitro without accumulating in red blood cells or causing hemolytic anemia, a known side effect of ribavirin.2 LVV exhibits poor oral bioavailability in rat and monkey (15% and 17%, respectively)3 and in
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
TRANSPORT OF LEVOVIRIN PRODRUGS
human clinical studies (estimated apparent oral bioavailability of 10%).4 It has been suggested that the low oral bioavailability of LVV is due to its limited permeability across the intestinal epithelium. Although ribavirin was reported to be actively transported by the human intestinal N1 sodium-dependent concentrative nucleoside transporters,5 LVV is not likely to be recognized by these nucleoside transporters since it contains the nonnatural L-ribose moiety. Caco-2 cell monolayers have been found to lack the concentrative NT1 and NT2 nucleoside transporters.6 Prodrug strategies have been employed to optimize physicochemical properties of poorly absorbed compounds for improving drug delivery.7–9 One successful prodrug approach for improved intestinal absorption exploits active transport systems to move the prodrug across the intestinal membrane. The promoiety portion of the molecule is designed to confer recognition by the active transport system and is cleaved after transport is complete to yield the active compound. The intestinal di/tri-peptide transporter, PepT1 has been investigated as a target for improving the bioavailability of poorly-absorbed drugs due to its broad substrate specificity.10–12 Valine esters of acyclovir (valacyclovir) and ganciclovir (valganciclovir) have been shown to exhibit improved absorption, which is attributed to the uptake via a peptide transporter even though there is no peptide bond in the structures.13–16 For example, valganciclovir exhibited a 10-fold increase in plasma ganciclovir concentrations compared to oral formulations of ganciclovir.17 Valganciclovir was found to be recognized as a substrate by the intestinal transporter PepT1.16 In the present investigation, the permeability of 10 amino acid ester prodrugs of LVV (see Fig. 1) was determined in human intestinal Caco-2 cells. The transport mechanisms of the lead prodrug, R1518 were also studied. Furthermore, the inhibitory effects of known PepT1 inhibitors on the permeability of R1518 and the reverse inhibitory effect of R1518 on PepT1 substrates were investigated. Pharmacokinetic screening studies in rats were carried out for LVV and the prodrugs.
MATERIALS AND METHODS Materials Caco-2 cells (passage 108–120) were obtained from Roche Basel. 12-well Transwell1 inserts DOI 10.1002/jps
H2N
N
O
N
N
O R3O
1319
5' 3'
R 2O
2'
OR1
Figure 1. Levovirin (R1, R2, R3 ¼ H) and amino acid ester prodrugs: 50 -(L)-valinate [R1, R2 ¼ H, R3 ¼ (L)-Val]; 50 -(D)-valinate [R1, R2 ¼ H, R3 ¼ (D)-Val]; 50 -(L)-isoleucinate [R1, R2 ¼ H, R3 ¼ (L)-Ile]; 50 -(L)-alaninate [R1, R2 ¼ H, R3 ¼ (L)-Ala]; 50 -(L)-leucinate [R1, R2 ¼ H, R3 ¼ (L)-Leu]; 50 -(L)-sarcosinate [R1, R2 ¼ H, R3 ¼ (L)Sar]; 50 -(L)-phenylalaninate [R1, R2 ¼ H, R3 ¼ (L)-Phe]; 20 ,30 -(L)-bis-valinate [R1, R2 ¼ (L)-Val, R3 ¼ H]; 50 -(L)valinylprolinate [R1, R2 ¼ H, R3 ¼ (L)-Val-Pro]; 50 -(L)prolinylvalinate [R1, R2 ¼ H, R3 ¼ (L)-Pro-Val].
(diameter: 12 mm) with collagen-coated polytetrafluoroethylene membrane (0.4 mm pores) and T225 cm2 cell culture flask (tissue culture treated) were purchased from Corning Incorporated (Corning, NY). Dulbecco’s Modified Eagle Media with high glucose and L-glutamine (DMEM), Lglutamine, penicillin-streptomycin, nonessential amino acids and fetal bovine serum were obtained from Gibco/Life Technologies (Gaithersburg, MD). Krebs–Henseleit bicarbonate buffer mix, calcium chloride dihydrate, enalapril, glycylsarcosine (Gly-Sar), and cephalexin were purchased from Sigma (St. Louis, MO). Sodium bicarbonate was obtained from Mallinckrodt Chemicals (Phillipsburg, NJ). Nanopure water was used for the buffer preparation. Amino acid ester prodrugs of LVV were generally prepared via a three or more step synthesis from LVV. The 50 -monoester syntheses utilized either a 20 ,30 -LVV cyclopentylidene or isopropylidene ketal intermediate which allowed selective esterification with the N-butoxycarbonyl (N-Boc) protected amino acids valine, isoleucine, alanine, leucine, sacrosine, and phenylalanine. The 20 , 30 bis-valinate prodrug was synthesized using the JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1320
LI ET AL.
50 -mono-triisopropylsilyl ether (TIPS) protected intermediate followed by esterification with N-Cbz (benzyloxycarbonyl)-valine and subsequent hydrogenolysis. LVV prodrug esters were characterized by elemental analysis, NMR, and HPLC-MS analyses.
Cell Culture High passage (108–120) Caco-2 cells were cultured in Dulbecco’s Modified Eagle Media with high glucose and L-glutamine (DMEM) supplemented with 10% fetal bovine serum, 1 L-glutamine, 1 penicillin-streptomycin, and 1 nonessential amino acids. Cells were maintained in T225 cm2 cell culture flask (tissue culture treated) at 378C and 5% CO2. For transport experiments, cells were plated at 7.1 104 cells/well onto 12-well collagen-coated PTFE membrane inserts. Cells were fed every 3 days and maintained at 378C and 5% CO2 for 21 days to allow complete formation of a polarized monolayer with tight junctions.
Transport Studies The Krebs–Henseleit bicarbonate buffer solution for transport studies were prepared by following the manufacturer’s instruction. The buffer was adjusted to pH 6.5 or 7.4 prior to use. Stock solutions of test compounds were prepared at 5 mg/mL concentrations in DMSO and stored at 48C. An aliquot of the 5 mg/mL stock solution was diluted with either pH 6.5 or pH 7.4 Krebs– Henseleit bicarbonate buffer to obtain a concentration of 20 mM for use as initial donor dosing solution (D0). Buffers, drug solutions, and Caco-2 cell plates containing buffers were prewarmed to 378C before use. The transepithelial electrical resistance (TEER) values were measured using a millicell1-ERS equipped with ‘‘chopstick’’ electrodes (Millipore, Bedford, MA). The inserts with TEER values above 300 ohm cm2 were used in the study after washing with warm Krebs–Henseleit bicarbonate buffer. For transport studies from apical to basolateral side, 0.5 mL Krebs–Henseleit buffer (pH 6.5) was added to apical side of the cell monolayers and 1.25 mL Krebs–Henseleit buffer (pH 7.4) to the basolateral chamber. The cells were equilibrated at 378C and 5% CO2 in an incubator for at least 30 min. The apical side buffer was removed and replaced with 0.5 mL of 20 mM drug solution. The JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
cells were then incubated at 378C and 5% CO2. At 30, 60, and 90 min time points, the inserts were transferred to new plates with 1.25 mL fresh warm pH 7.4 buffer at the receiver side. The media from all plates were collected as receiver samples. For bi-directional transport studies from basolateral to apical side, 1.25 mL of 20 mM drug solution in pH 7.4 buffer was placed in the basolateral chamber, and 0.5 mL pH 6.5 buffer was put into the apical chamber. At 30, 60, and 90 min time points, the samples in the apical side were collected and then replaced with 0.5 mL fresh warm pH 6.5 buffer. After 60 min of transport studies, Lucifer Yellow (0.05 mL 1000 mM) was added to the apical side of the wells. At the end of the transport studies (90 min), the fluorescence of the receiver side samples was measured. Sample solutions from the donor side were collected at the end of the experiments as D90 samples. For inhibitory studies, the study compound concentration was 20 mM, whereas the concentration of the inhibitors was 20 mM. All prodrug samples were acidified to approximately pH 3.5 with formic acid solution immediately after collection to prevent further chemical hydrolysis. Chemical Stability Study of R1518 Solutions containing 20 mM of R1518 were prepared using Krebs–Henseleit buffers at pH 6.5 and at pH 7.4. The solutions were kept in a 378C/ 5% CO2 incubator for 2 h. Samples were taken at 30, 60, 90, and 120 min to check the conversion of R1518 to LVV. Pharmacokinetic Screening Studies in Rats Oral administration of single 10-mg/kg doses of LVV and equivalent LVV doses of prodrugs were given to nonfasted male Hanover–Wistar rats. Animals were allowed access to water and food on their normal schedule. LVV and prodrugs were administered as solutions in water or buffered saline. Blood samples were collected at 0.25, 0.5, 1, 2, 4, 6, 8, 10, and 24 h after dosing. Plasma concentrations of LVV were determined by a HPLC/MS/MS method as described in the sample analysis after plasma sample extraction. Sample Analysis LVV and its prodrugs were analyzed using highperformance liquid chromatography with tandem mass-spectrometry (HPLC/MS/MS). An Agilent DOI 10.1002/jps
TRANSPORT OF LEVOVIRIN PRODRUGS
1321
Table 1. Caco-2 Permeability Studies of Levovirin and Its Prodrugs % Conversion Ratio at 90 min Levovirin= Levovirin þ Prodrug Compound
Caco-2 Permeability (106 cm/s)a
Permeability Fold Increase
Apical Side
Basolateral Side
0.12 0.06 0.18 0.06 5.70 1.99 0.52 0.26 1.46 1.09 1.17 0.57 0.88 0.17 0.50 0.39 0.14 0.07 8.55 1.75 3.42 0.18 1.98 1.04
1.5 47.9 4.4 12.4 9.9 7.5 4.3 1.2 72.5 29.0 16.8
7.6 11.0 5.9 41.2 17.4 34.9 61.7 53.4 99.1 5.7
93.8 2.7 72.8 100 80.2 100 67.3 12.7 24.1 45.9
Levovirin Ribavirin LVV-50 -(L)-valine (R1518) LVV-50 -(D)-valine LVV-50 -(L)-isoleucine LVV-50 -(L)-alanine LVV-50 -(L)-leucine LVV-50 -(L)-sarcosine LVV-50 -(L)-phenylalanine LVV-20 ,30 -(L)-bis-valine LVV-50 -val-pro LVV-50 -pro-val a
Mean SD, n ¼ 18 for levovirin, n ¼ 16 for R1518, n ¼ 4 for ribavirin, and LVV-50 -(D)-valine, and n ¼ 2 for others.
Zorbax SB-Aq 4.6 50 mm column (5 mM) was used for separation. Electrospray ionization (ESI) was used for the ionization process. The mobile phase A contained 5 mM ammonium formate in water and mobile phase B contained methanol. Elution was performed with 98% A and 2% B for 3 min, then with linear gradient to 10% A and 90% B in 1 min and held at this condition for 2 min with a flow rate of 0.5 mL/min. Data Analysis for Permeability Studies in Caco-2 Cells The dQ/dt of test substance was calculated from sampling data at 30, 60, and 90 min. The apparent permeability coefficient (Papp) was calculated from the following equation, Papp ¼
dQ 1 dt A Co
where A is the surface area (cm2) of the insert, Co is the initial concentration of the drug, and dQ/dt is the change in total amount of the prodrug and its parent drug (LVV) in the receiver solution over the 90 min incubation time, that is, the slope of the drug amount in the receiver solution versus time.
RESULTS AND DISCUSSION Permeability of Levovirin and Its Prodrugs The apical (A) to basolateral (B) permeability of 10 LVV prodrugs was determined in Caco-2 cells and the results were compared with the DOI 10.1002/jps
parent drug, LVV (Table 1). LVV has a very low transport rate through Caco-2 cell monolayers (1.2 107 cm/s). Enhanced permeability was achieved with the majority of LVV prodrugs investigated in this study. R1518 showed the maximum improvement in permeability (48-fold increase) among all mono amino acid ester prodrugs. In contrast, only a slight increase (fourfold) in permeability was observed for LVV-50 -D-valinate prodrug, suggesting that a carrier-mediated mechanism is involved in the intestinal transport of R1518, but not LVV-50 -(D)-valinate.18 The fourfold increase in permeability for LVV-50 -(D)-valinate is likely due to the increase in hydrophobicity via passive diffusion. The transport of the 20 ,30 -Lbis-valine ester prodrug increased 72-fold compared with LVV. However, only approximately 13% prodrug converted to the parent drug in the basolateral side within 90 min. The low permeability of ribavirin through Caco-2 cell monolayers is probably due to the lack of the concentrative N1 transporter and sodium gradient in the transport system.5 Table 2. Chemical Stability of R1518 at 378C
Time (min) 30 60 90 120
% R1518 Remaining in pH 6.5 Buffer
% R1518 Remaining in pH 7.4 Buffer
97.8 96.1 92.7 88.1
96.9 92.8 88.7 83.7
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1322
LI ET AL.
Table 3. Rat Pharmacokinetic Screening Studies at 10 mg/kg Levovirin Equivalent of Levovirin Prodrugs
Compound
AUC0-inf of Levovirin (mg h/mL)a
AUC Fold Increase
1.70 0.62 11.70 0.61 9.70 1.35 6.58 3.06 2.41 0.60 1.01 0.20 1.75 0.50 5.22 0.42 1.38 0.21 3.85 1.11
— 6.9 5.7 3.9 1.4 0.6 1.0 3.1 0.8 2.3
Levovirin LVV-50 -(L)-valine (R1518) LVV-50 -(L)-isoleucine LVV-50 -(L)-alanine LVV-50 -(L)-leucine LVV-50 -(L)-sarcosine LVV-50 -(L)-phenylalanine LVV-20 ,30 -(L)-bis-valine LVV-50 -val-pro LVV-50 -pro-val
a Mean SD, n ¼ 3; AUC0-inf: area under the curve from time zero to infinity; Nonfasted male Wistar rats; Dose administration: oral gavage.
R1518 converted to LVV almost completely in basolateral side at the end of the transport study, whereas less than 3% of LVV-50 -(D)-valine converted to LVV. Chemical stability study showed that more than 88% of R1518 remained intact in either apical or basolateral side solution after 90 min of incubation at 378C (Table 2), indicating that ester hydrolysis occurred within the epithelial cells.13 These results support that both carriermediated transport and ester hydrolysis in Caco-2 cell monolayers are stereoselective towards Lamino acid esters.18,19 R1518 was selected as the lead prodrug candidate due to its high permeability and fast hydrolysis to the parent drug. It has been evaluated in clinical trials for the treatment of hepatitis C. Consistent enhancement in oral bioavailability (AUC) for most of the LVV prodrugs was observed in rat pharmacokinetic screening studies (Table 3) with the highest increase (6.9-fold) observed for R1518. In general, the pharmacokinetic screening studies in rats displayed a similar trend to that observed among the mono amino acid prodrugs for increases in permeability in Caco-2 cells. For di-
amino acid prodrugs, there was only a moderate improvement in AUC in rats compared to corresponding increase in Caco-2 permeability, which is probably due to slow conversion of prodrugs to LVV in rats. The pharmacokinetic results in rats are also consistent with that from human single- and multiple-dose pharmacokinetics studies of R1518 in healthy volunteers which yielded significantly improved oral absorption of LVV upon oral administration.4 Plasma exposures (AUC) of LVV dosed as R1518 in human exhibited an eightfold increase versus LVV administered orally.4
Bi-Directional Transport Studies of LVV and Its Prodrugs A to B and B to A permeability of LVV and its prodrugs were determined in Caco-2 cells with a pH gradient of apical side pH 6.5 and basolateral side pH 7.4. The results are shown in Table 4. There are no significant differences between A to B and B to A permeability values for LVV and
Table 4. Bi-Directional Transport Study of Levovirin and Valine Ester Prodrugs with pH Gradient (Apical pH 6.5, Basolateral pH 7.4) Permeability (106 cm/s)a Compound
A to B
Levovirin 0.12 0.06 (n ¼ 18) LVV-50 -(L)-valine (R1518) 5.70 1.99 (n ¼ 16) LVV-50 -(D)-valine 0.52 0.26 (n ¼ 4) a
B to A
A to B= B to A Ratio
0.21 0.01 (n ¼ 2) 0.14 0.06 (n ¼ 4) 0.44 0.01 (n ¼ 2)
0.56 40.7 1.18
Mean SD.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
DOI 10.1002/jps
TRANSPORT OF LEVOVIRIN PRODRUGS
Table 5. Bi-Directional Transport Study of R1518 without pH Gradient (Apical and Basolateral pH 7.4) Permeability (106 cm/s)a A to B 4.13 0.82 (n ¼ 4) a
A to B= B to A Ratio
B to A 0.16 0.04 (n ¼ 2)
26.5
Mean SD.
LVV-50 -(D)-valine, respectively, indicating that their transport through intestinal epithelium is mainly by passive diffusion. A to B transport of R1518 is approximately 40-fold higher than that of B to A, suggesting that a carrier-mediated transport is involved in the intestinal absorption.15 In addition, a bidirectional transport study of R1518 was performed without pH gradient (pH 7.4 at both apical and basolateral sides). As shown in Table 5, the B to A permeability of R1518 did not change significantly. However, the A to B permeability decreased approximately 30% resulting in lower A to B=B to A ratio. The pH appears to have more effect on the transport from A to B than that from B to A. pH Effect on Permeability of R1518
1323
Inhibitory Effect of Gly-Sar, Enalapril, Cephalexin and Valganciclovir on the Permeability of R1518 Human intestinal Caco-2 cell line has been shown to express PepT1 in the brush-border membranes,21,22 and it has been used as a model for studying the functions of PepT1 in the transport of peptide-like compounds.15,19 We examined the transport of R1518 across Caco-2 cell monolayers in the presence or absence of excess Gly-Sar, enalapril, cephalexin, or valganciclovir which are well known PepT1 substrates and inhibitors.16,19,23–25 As shown in Figure 3, the permeability of R1518 (Lev-Val prodrug) was reduced 88% by enalapril, 80% by valganciclovir, 80% by Gly-Sar, and 52% by cephalexin, indicating that R1518 competes with enalapril, valganciclovir, Gly-Sar, or cephalexin ( p < 0.01) for the same carrier-mediated transport pathway. These results suggest that R1518 is a substrate of PepT1. Cephalexin appears to be a weaker inhibitor than enalapril, valganciclovir, and Gly-Sar for PepT1.
Inhibitory Effect of R1518 on the Permeability of Valganciclovir, Enalapril, and Cephalexin
A to B permeability of R1518 was determined at apical side pH 5.0, 6.5, 7.4, and fixed basolateral side pH 7.4. The results in Figure 2 show that as the apical side pH increases, the A to B permeability decreases, suggesting that the transport of R1518 is pH dependent. This finding is consistent with the previous report that transport via the PepT1 transporter is coupled to a downhill movement of Hþ across brush border membrane.20
Since valganciclovir, enalapril, and cephalexin are known PepT1 substrates, their permeability through Caco-2 cell monolayers was determined in the presence or absence of excess R1518 to examine the inhibitory effect of R1518. The results in Figure 4 show that R1518 reduced the transport of valganciclovir and cephalexin by over 50% ( p < 0.01), indicating that R1518 is an inhibitor of PepT1. Due to the low permeability of enalapril through Caco-2 cell monolayers, the
Figure 2. A to B permeability of R1518 across Caco-2 cell monolayers as a function of apical pH (basolateral pH 7.4).
Figure 3. Permeability of R1518 (20 mM) in the presence or absence of enalapril, valganciclovir, Gly-Sar and cephalexin (20 mM). Mean SD, n ¼ 4 except with valganciclovir (n ¼ 2).
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1324
LI ET AL.
the cell culture facility at Roche Palo Alto for preparation of Caco-2 cell transwell plates used for these studies. Members of the LVV/prodrug Research and Development Team are gratefully acknowledged for their help and support. Colleagues at Roche Carolina, Inc., are thanked for their collaboration in LVV and R1518 synthesis and scale up.
REFERENCES Figure 4. Inhibitory effect of R1518 (20 mM) on the permeability of valganciclovir, enalapril and cephalexin (20 mM). Mean SD, n ¼ 4 except enalapril alone (n ¼ 6) and valganciclovir with R1518 (n ¼ 2).
inhibitory effect of R1518 on enalapril is not statistically significant.
CONCLUSIONS The low permeability of LVV in Caco-2 cell monolayers suggests that the low oral bioavailability in human is due to its limited absorption in the gastrointestinal tract. This study demonstrated that several amino acid ester prodrugs of LVV significantly increased permeability in Caco-2 cell monolayers via a peptide transport mechanism. The study also demonstrated that the lead prodrug, R1518 is both a substrate and an inhibitor of the PepT1 transporter. The transport process via PepT1 is pH dependent. In addition, R1518 is rapidly converted to LVV in the basolateral side of Caco2 cell monolayers. Pharmacokinetic screening studies of the prodrugs in rats displayed a similar trend of improvement in AUC compared with that observed in permeability in Caco-2 cells with the highest increase (6.9-fold) for R1518. The pharmacokinetic results in rats are also consistent with that from pharmacokinetic studies of R1518 in human healthy volunteers which yielded an eightfold increase in plasma exposures (AUC) versus LVV administered orally.4,26 The results of this investigation on LVV prodrugs provide further evidence of the feasibility of enhancing oral absorption by design of peptidelike prodrugs targeting peptide transporters.
ACKNOWLEDGMENTS The authors thank L. He and H. Liu for their support on the sample analysis by LC/MS/MS and JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006
1. Keating GM, Curran MP. 2003. Peginterferonalpha-2a (40 kD) plus ribavirin: A review of its use in the management of chronic hepatitis C. Drugs 63:701–730. 2. Tam RC, Ramasamy K, Bard J, Pai B, Lim C, Averett DR. 2000. The ribavirin analogue ICN 17261 demonstrates reduced toxicity and antiviral effects with retention of both immunomodulatory activity and reduction of hepatitis-induced serum alanine aminotransferase levels. Antimicrob Agents Chemother 44:1276–1283. 3. Lin CC, Luu T, Lourenco D, Yeh LT, Lau JYN. 2003. Absorption, pharmacokinetics and excretion of levovirin in rats, dogs and cynomolgus monkeys. J Antimicrob Chemother 51:93–99. 4. Huang Y, Ostrowitzki S, Hill G, Navarro M, Berger N, Kopeck P, Mau CI, Alfredson T, Lal R. 2005. Single- and multiple-pharmacokinetics of levovirin valinate hydrochloride (R1518) in healthy volunteers. J Clin Pharmacol 45:578–588. 5. Patil SD, Ngo LY, Glue P, Unadkat JD. 1998. Intestinal absorption of ribavirin is preferentially mediated by the Naþ-nucleoside purine (N1) transporter. Pharm Res 15:950–952. 6. Ward JL, Tse CM. 1999. Nucleoside transport in human colonic epithelial cell lines: Evidence for two Naþ-independent transport systems in T84 and Caco-2 cells. Biochim Biophys Acta 1419:15–22. 7. Charman WN, Porter CJH. 1996. Lipophilic prodrugs designed for intestinal lymphatic transport. Adv Drug Deliv Rev 19:149–169. 8. Fleisher D, Bong R, Stewart BH. 1996. Improved oral drug delivery: Solubility limitations overcome by the use of prodrugs. Adv Drug Deliv Rev 19:115– 130. 9. Ettmayer P, Amidon GL, Clement B, Testa B. 2004. Lessons learned from marketed and investigational prodrugs. J Med Chem 47:1–12. 10. Balimane PV, Tamai I, Guo A, Nakanishi T, Kitada H, Leibach FH, Tsuji A, Sinko PJ. 1998. Direct evidence for peptide transporter (PepT1)-mediated uptake of a nonpeptide prodrug, valacyclovir. Biochem Biophys Res Commun 250:246–251. 11. Sawada K, Terada T, Saito H, Hashimoto Y, Inui KI. 1999. Recognition of L-amino acid ester DOI 10.1002/jps
TRANSPORT OF LEVOVIRIN PRODRUGS
12.
13.
14.
15.
16.
17.
18.
19.
compounds by rat peptide transporters PEPT1 and PEPT2. J Pharmacol Exp Ther 291:705–709. Rubio-Aliaga I, Daniel H. 2002. Mammalian peptide transporters as targets for drug delivery. Trends Pharmacol Sci 23:434–440. Ganapathy ME, Huang W, Wang H, Ganapathy V, Leibach FH. 1998. Valacyclovir: A substrate for the intestinal and renal peptide transporters PEPT1 and PEPT2. Biochem Biophys Res Commun 246: 470–475. Sinko PJ, Balimane PV. 1998. Carrier-mediated intestinal absorption of valacyclovir, the L-valyl ester prodrug of acyclovir. 1. Interactions with peptides, organic anions and organic cations in rats. Biopharm Drug Dispos 19:209–217. de Vrueh RL, Smith PL, Lee CP. 1998. Transport of L-valine-acyclovir via the oligopeptide transporter in the human intestinal cell line, caco-2. J Pharmacol Exp Ther 286:1166–1170. Sugawara M, Huang W, Fei YJ, Leibach FH, Ganapathy V, Ganapathy ME. 2000. Transport of valganciclovir, a ganciclovir prodrug, via peptide transporters PEPT1 and PEPT2. J Pharm Sci 89: 781–789. Brown F, Banken L, Saywell K, Arum I. 1999. Pharmacokinetics of valganciclovir and ganciclovir following multiple oral dosages of valganciclovir in HIV- and CMV-seropositive volunteers. Clin Pharmacokinet 37:167–176. Han H, de Vrueh RL, Rhie JK, Covitz KM, Smith PL, Lee CP, Oh DM, Sadee W, Amidon GL. 1998. 50 amino acid esters of antiviral nucleosides, acyclovir, and AZT are absorbed by the intestinal PEPT1 peptide transporter. Pharmaceutical Res 15:1154– 1159. Han HK, Oh DM, Amidon GL. 1998. Cellular uptake mechanism of amino acid ester prodrugs
DOI 10.1002/jps
20.
21.
22.
23.
24.
25.
26.
1325
in Caco-2/hPEPT1 cells overexpressing a human peptide transporter. Pharm Res 15:1382–1386. Boll M, Herget M, Wagener M, Weber WM, Markovich D, Biber J, Clauss W, Murer H, Daniel H. 1996. Expression cloning and functional characterization of the kidney cortex high-affinity proton-coupled peptide transporter. Proc Natl Acad Sci USA 93:284–289. Saito H, Inui K. 1993. Dipeptide transporters in apical and basolateral membranes of the human intestinal cell line Caco-2. Ameri Physio Society 265:G289–G294. Walker D, Thwaites DT, Simmons NL, Gilbert HJ, Hirst BH. 1998. Substrate upregulation of the human small intestinal peptide transporter, hPepT1. J Physiol 507:697–706. Abe H, Satoh M, Miyauchi S, Shuto S, Matsuda A, Kamo N. 1999. Conjugation of dipeptide to fluorescent dyes enhances its affinity for a dipeptide transporter (PEPT1) in human intestinal Caco-2 cells. Bioconjugate Chem 10:24–31. Temple CS, Boyd CA. 1998. Proton-coupled oligopeptide transport by rat renal cortical brush border membrane vesicles: A functional analysis using ACE inhibitors to determine the isoform of the transporter. Biochim Biophys Acta 1373:277– 281. Chu XY, Sanchez-Castano GP, Higaki K, Oh DM, Hsu CP, Amidon GL. 2001. Correlation between epithelial cell permeability of cephalexin and expression of intestinal oligopeptide transporter. J Pharmacol ExpTher 299:575–582. Mau C, Liu H, Tran T, Young M, Ahene A, Tarnowski T, Alfredson T, Huang Y, Lal R. 2004. Pharmacokinetic studies of levovirin valinate hydrochloride (R1518) in mice, rats, monkeys, and humans. 7th International ISSX Meeting. August.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 6, JUNE 2006