Search

A generalized screening approach, applying isothermal calorimetry at 37 °C 100% RH, to formulations of spray dried dispersions (SDDs) for two active pharmaceutical ingredients (APIs) (BMS-903452 and BMS-986034) is demonstrated. APIs 452 and 034, with similar chemotypes, were synthesized and promoted during development for oral dosing. Both APIs were formulated as SDDs for animal exposure studies using the polymer hydroxypropylmethlycellulose acetyl succinate M grade (HPMCAS-M). 452 formulated at 30% (wt/wt %) was an extremely robust SDD that was able to withstand 40 °C 75% RH open storage conditions for 6 months with no physical evidence of crystallization or loss of dissolution performance. Though 034 was a chemical analogue with similar physical chemical properties to 452, a physically stable SDD of 034 could not be formulated in HPMCAS-M at any of the drug loads attempted. This study was used to develop experience with specific physical characterization laboratory techniques to evaluate the physical stability of SDDs and to characterize the propensity of SDDs to phase separate and possibly crystallize. The screening strategy adopted was to stress the formulated SDDs with a temperature humidity screen, within the calorimeter, and to apply orthogonal analytical techniques to gain a more informed understanding of why these SDDs formulated with HPMCAS-M demonstrated such different physical stability. Isothermal calorimetry (thermal activity monitor, TAM) was employed as a primary stress screen wherein the SDD formulations were monitored for 3 days at 37 °C 100% RH for signs of phase separation and possible crystallization of API. Powder X-ray diffraction (pXRD), modulated differential scanning calorimetry (mDSC), Fourier transform infrared spectroscopy (FTIR), and solid state nuclear magnetic resonance (ssNMR) were all used to examine formulated SDDs and neat amorphous drug. 452 SDDs formulated at 30% (wt/wt %) or less did not show phase separation behavior upon exposure to 37 °C 100% RH for 3 days. 034 SDD formulations from 10 through 50% (wt/wt %) all demonstrated thermal traces consistent with exothermic phase separation events over 3 days at 37 °C 100% RH in the TAM. However, only the 15, 30, and 50% containing 034 samples showed pXRD patterns consistent with crystalline material in post-TAM samples. Isothermal calorimetry is a useful screening tool to probe robust SDD physical performance and help investigate the level of drug polymer miscibility under a humid stress. Orthogonal analytical techniques such as pXRD, ssNMR, and FTIR were key in this SDD formulation screening to gain physical understanding and confirm or refute whether physical changes occur during the observed thermal events characterized by the calorimetric screening experiments.

Poly (vinyl alcohol), PVA, a commonly used excipient to coat tablets, forms insoluble films in the presence of acids and thermal stress. This may lead to drug products failing to meet dissolution specifications over time. Studies were conducted to understand the effect of acid strength, processing conditions, and storage stress on the mechanism of insoluble film formation using PVA and OpadryⓇ II as model systems. Aqueous cast films, prepared by incorporating hydrochloric acid (HCl) into the coating solutions or exposing pre-cast "as is" films to HCl vapors, were used as surrogates to develop analytical methods. To understand effect of acid and processing on coatings, acidified OpadryⓇ II was spray coated onto inert cores under "wet" or "dry" conditions. Samples stored at 50-60 °C were analyzed for film disintegration to understand physical/chemical changes in the polymer. Rate and extent of insoluble films formation was dependent on the acid concentration and thermal stress. Analysis of the films indicated significant de-acetylation and ether bond formation in insoluble aqueous cast films. In contrast, acidified coated films showed only ether bond formation, which increased on stress, forming insoluble films. The reduction in the time to form insoluble films for "wet" versus "dry" coated films was rationalized by considering effect of coating, drying, and storage on the microstructure of acidified PVA and ether bond propagation. The results highlight the need to develop an in-depth understanding of the design space for PVA coated products and storage conditions in presence of acids.

In vitro and in vivo experimental models are frequently used to assess a new chemical entity’s (NCE) biopharmaceutical performance risk for food effect (FE) in humans. Their ability to predict human FE hinges on replicating key features of clinical FE studies and building an in vitro-in vivo relationship (IVIVR). In this study, 22 compounds that span a wide range of physicochemical properties, Biopharmaceutics Classification System (BCS) classes, and food sensitivity were evaluated for biorelevant dissolution in fasted- and fed-state intestinal media and the dog fed/fasted-state pharmacokinetic model. Using the area under the curve (AUC) as a performance measure, the ratio of the fed-to-fasted AUC (FE ratio) was used to correlate each experimental model to FE ratio in humans. A linear correlation was observed for the in vitro dissolution-human IVIVR (R 2 = 0.66, % mean square error 20.7%). Similarly, the dog FE ratio correlated linearly with the FE ratio in humans (R 2 = 0.74, % mean square error 16.25%) for 15 compounds. Data points near the correlation line indicate dissolution-driven mechanism for food effect, while deviations from the correlation line shed light on unique mechanisms that can come into play such as GI physiology or unusual physicochemical properties. In summary, fed/fasted dissolution studies and dog PK studies show a reasonable correlation to human FE, hence are useful tools to flag high-risk NCEs entering clinical development. Combining kinetic dissolution, dog FE model and in silico modeling one can study FE mechanism and formulation strategies to mitigate the FE risk.

The aim of this study was to investigate novel approaches to determine spray dried dispersion (SDD) specific particle characteristics through the use of imaging based technologies. The work demonstrates approaches that can be applied in order to access quantitative approximations for powder characteristics for hollow particles, such as SDD. Cryo-SEM has been used to measure the solid volume fraction and/or particle density of SDD particles. Application of this data to understand the impact of spray drying process conditions on SDD powder properties, and their impact on processability and final dosage form quality were investigated. The use of data from a Morphologi G3 image based particle characterisation system was also examined in order to explain both the propensity and extent of attrition within a series of SDD samples, and also demonstrate the use of light transmission data to assess the relative wall thickness of SDD particles. Such approaches demonstrate a means to access potentially useful information that can be linked to important particle characteristics for SDD materials which, in addition to the standard bulk powder measurements such as bulk density, may enable a better understanding of such materials and their impact on downstream processability and final dosage form acceptability.

Prodrugs are biologically inactive agents that upon biotransformation in vivo result in active drug molecules. Since prodrugs might alter the tissue distribution, efficacy and the toxicity of the parent drug, prodrug design should be considered at the early stages of preclinical development. In this regard, natural and synthetic amino acids offer wide structural diversity and physicochemical properties. This review covers the use of amino acid prodrugs to improve poor solubility, poor permeability, sustained release, intravenous delivery, drug targeting, and metabolic stability of the parent drug. In addition, practical considerations and challenges associated with the development of amino acid prodrugs are also covered.

The expression levels of genes involved in drug and nutrient absorption were evaluated in the Madin-Darby Canine Kidney (MDCK) in vitro drug absorption model. MDCK cells were grown on plastic surfaces (for 3 days) or on Transwell® membranes (for 3, 5, 7, and 9 days). The expression profile of genes including ABC transporters, SLC transporters, and cytochrome P450 (CYP) enzymes was determined using the Affymetrix® Canine GeneChip®. Expression of genes whose probe sets passed a stringent confirmation process was examined. Expression of a few transporter (MDR1, PEPT1 and PEPT2) genes in MDCK cells was confirmed by RT-PCR. The overall gene expression profile was strongly influenced by the type of support the cells were grown on. After 3 days of growth, expression of 28% of the genes was statistically different (1.5-fold cutoff, p < 0.05) between the cells grown on plastic and Transwell® membranes. When cells were differentiated on Transwell® membranes, large changes in gene expression profile were observed during the early stages, which then stabilized after 5–7 days. Only a small number of genes encoding drug absorption related SLC, ABC, and CYP were detected in MDCK cells, and most of them exhibited low hybridization signals. Results from this study provide valuable reference information on endogenous gene expression in MDCK cells that could assist in design of drug-transporter and/or drug-enzyme interaction studies, and help interpret the contributions of various transporters and metabolic enzymes in studies with MDCK cells.

To determine the bioactivation and uptake of prolidase-targeted proline prodrugs of melphalan in six cancer cell lines with variable prolidase expression and to evaluate prolidase-dependence of prodrug cytotoxicity in the cell lines compared to that of the parent drug, melphalan.Hydrolysis, cell uptake, and cell proliferation studies of melphalan and the L: - and D: -proline prodrugs of melphalan, prophalan-L: and prophalan-D: , respectively, were conducted in the cancer cell lines using established procedures.The bioactivation of prophalan-L: in the cancer cell lines exhibited high correlation with their prolidase expression levels (r (2) = 0.86). There were no significant differences in uptake of melphalan and its prodrugs. The cytotoxicity of prophalan-L: (GI(50)) in cancer cells also showed high correlation with prolidase expression (r (2) = 0.88), while prophalan-D: was ineffective at comparable concentrations. A prolidase targeting index (ratio of melphalan to prophalan-L: cytotoxicity normalized to their uptake) was computed and showed high correlation with prolidase expression (r (2) = 0.82).The data corroborates the specificity of prophalan-L: activation by prolidase as well as prolidase-targeted cytotoxicity of prophalan-L: in cancer cell lines. Hence, prophalan-L: , a stable prodrug of melphalan, exhibits potential for efficiently targeting melanoma with reduced systemic toxicity.

Floxuridine is a clinically proven anticancer agent in the treatment of metastatic colon carcinomas and hepatic metastases. However, prodrug strategies may be necessary to improve its physiochemical properties and selectivity and to reduce undesirable toxicity effects. Previous studies with amino acid ester prodrugs of nucleoside drugs targeted to the PEPT1 transporter coupled with recent findings of the functional expression of the PEPT1 oligopeptide transporter in pancreatic adenocarcinoma cell lines suggest the potential of PEPT1 as therapeutic targets for cancer treatment. In this report, we show the feasibility of achieving enhanced transport and selective antiproliferative action of amino acid ester prodrugs of floxuridine in cell systems overexpressing PEPT1. All prodrugs exhibited affinity for PEPT1 (IC50, 1.1–2.3 mmol/L). However, only the prolyl and lysyl prodrugs exhibited enhanced uptake (2- to 8-fold) with HeLa/PEPT1 cells compared with HeLa cells, suggesting that the aspartyl prodrugs are PEPT1 inhibitors. The selective growth inhibition of Madine-Darby canine kidney (MDCK)/PEPT1 cells over MDCK cells by the prodrugs was consistent with the extent of their PEPT1-mediated transport. All ester prodrugs hydrolyzed to floxuridine fastest in Caco-2 cell and MDCK homogenates and slower in human plasma and were most chemically stable in pH 6.0 buffer. Prolyl and lysyl prodrugs were relatively less stable compared with aspartyl prodrugs in buffers and in cell homogenates. The results suggest that optimal design for targeted delivery would be possible by combining both stability and transport characteristics afforded by the promoiety.

Gemcitabine, a clinically effective nucleoside anticancer agent, is a polar drug with low membrane permeability and is administered intravenously. Further, extensive degradation of gemcitabine by cytidine deaminase to an inactive metabolite in the liver affects its activity adversely. Thus, strategies that provide both enhanced transport and high metabolic bioevasion would potentially lead to oral alternatives that may be clinically useful. The objective of this study was to evaluate whether amino acid ester prodrugs of gemcitabine would (a) facilitate transport across intestinal membranes or across cells that express hPEPT1 and (b) provide resistance to deamination by cytidine deaminase. 3′-Monoester, 5′-monoester, and 3′,5′-diester prodrugs of gemcitabine utilizing aliphatic (L-valine, D-valine, and L-isoleucine) and aromatic (L-phenylalanine and D-phenylalanine) amino acids as promoieties were synthesized and evaluated for their affinity and direct hPEPT1-mediated transport in HeLa/hPEPT1 cells. All prodrugs exhibited enhanced affinity (IC 50 : 0.14-0.16 mM) for the transporter. However, only the 5′-L-valyl and 5′-L-isoleucyl monoester prodrugs exhibited (a) increased uptake (11.25- and 5.64-fold, respectively) in Hel_a/hPEPT1 cells compared to HeLa cells and (b) chemical stability in buffers, that were comparable to valacyclovir, a commercially marketed oral amino acid ester prodrug. The widely disparate enzymatic bioconversion profiles of the 5′-L-valyl and 5′-L-isoleucyl prodrugs in Caco-2 cell homogenates along with their significant resistance to deamination by cytidine deaminase suggest that the disposition of gemcitabine following oral administration would be controlled by the rate of bioconversion following transport across the intestinal epithelial membrane. The combined results also suggest that it may be possible to modulate these characteristics by the choice of the amino acid promoiety.

cyclo[d-Asp(2),Dap(5)]Dyn A-(1-13)NH(2) (Dap, 2,3-diaminopropionic acid; Dyn A, dynorphin A), synthesized previously in our laboratory, showed sub-nanomolar affinity for kappa opioid receptors and potent agonist activity in the guinea pig ileum assay (Arttamangkul et al., J. Med. Chem. 1995, 38, 2410-2417). Various modifications were made in position 3 of cyclo[d-Asp(2),Dap(5)]Dyn A-(1-11)NH(2) that could influence the opioid receptor affinity, selectivity, and/or efficacy of this peptide. An optimized orthogonal synthetic strategy was developed for the synthesis of these cyclic peptides in which the final peptides could be cleaved from the solid support with trifluoroacetic acid. Substitutions of Gly(3) by Ala, d-Ala, Trp, and d-Trp in cyclo[d-Asp(2),Dap(5)]Dyn A-(1-11)NH(2) and its linear counterpart [d-Asp(2),Dap(5)]Dyn A-(1-11)NH(2) were generally well tolerated by both kappa and micro opioid receptors. Despite differences in the size and stereochemistry of the substitutions, most of the peptides (except for cyclo[d-Asp(2),Pro(3),Dap(5)]Dyn A-(1-11)NH(2) and [d-Asp(2),d-Ala(3), Dap(5)]Dyn A-(1-11)NH(2)) exhibited low nanomolar affinity for both kappa (K(i) = 0.21 to 2.2 nM) and micro (K(i) = 0.22 to 7.27 nM) opioid receptors. All of the 3-substituted cyclic and linear analogues synthesized showed reduced affinity for delta opioid receptors. Incorporation of d-Ala at position 3 of cyclo[d-Asp(2),Dap(5)]Dyn A-(1-11)NH(2) exhibited 2-fold higher kappa opioid receptor affinity and 16-fold higher selectivity for kappa over micro opioid receptors than the parent cyclic peptide. In contrast, substitution of Ala at position 3 resulted in an analogue with 2.4-fold lower affinity and very low preference for kappa over micro opioid receptors. The Trp and d-Trp cyclic and linear analogues exhibited similar nanomolar affinities for kappa opioid receptors. cyclo[d-Asp(2),Pro(3),Dap(5)]Dyn A-(1-11)NH(2) showed the largest decreases in affinity for all three opioid receptors compared to the parent cyclic peptide. Except for cyclo[d-Asp(2), Pro(3),Dap(5)]Dyn A-(1-11)NH(2), which was a partial agonist, all of the cyclic peptides exhibited full agonist activity in the adenylyl cyclase assay using cloned kappa opioid receptors.

Phenylalanine at position 4 of the peptide dynorphin A (Dyn A) is an important residue for opioid receptor affinity and activity, but there is very little information available on the structure-activity relationships or conformational preference of this residue for interaction with kappa-opioid receptors. Based on the hypothesis that the spatial orientation of the aromatic ring at position 4 of Dyn A is important for opioid receptor affinity and selectivity, a series of Dyn A analogues with various Phe derivatives substituted at position 4 were synthesized and evaluated for their opioid receptor affinity and activity. The L- and D-Homophe4 (homophenylalanine) analogues of [D-Ala8]Dyn A-(1-11)NH2 were compared to the (R)- and (S)-Atc4 (2-aminotetralin-2-carboxylic acid) derivatives (Aldrich et al. Chirality 2001, 13, 125-129). [l-Homophe4,D-Ala8]Dyn A-(1-11)NH2 exhibited higher kappa-opioid receptor affinity than the D-Homophe4 isomer, while [(R)-Atc4,D-Ala8]Dyn A-(1-11)NH2 exhibited higher kappa-opioid receptor affinity than the (S)-Atc4 isomer. Comparing the structure of Atc to those of Phe and Homophe, these results suggest that the Atc isomers are functioning more as constrained Homophe rather than Phe analogues in these Dyn A derivatives. The higher kappa-opioid receptor affinity of the (R)-Atc4 analogue suggests that Phe4 of Dyn A most likely adopts a gauche (-) or trans conformation in the kappa-opioid receptor binding site. Comparison of [D-Ala8]Dyn A-(1-11)NH2 derivatives containing Aic4 (2-aminoindan-2-carboxylic acid) and Tic4 (1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid) with the peptides containing their acyclic counterparts alpha-MePhe4 and N-MePhe4, respectively, suggest that the loss in opioid receptor affinity seen for the Aic4 and Tic4 analogues is probably due to an improper orientation of the aromatic ring in these residues. Most of the analogues in this series showed much lower affinity for delta-opioid receptors than the parent peptide, suggesting that kappa- and delta-opioid receptors have distinct binding pockets for the residue at position 4 of Dyn A. All of the analogues with high affinity for kappa-opioid receptors exhibited full agonist activity in the adenylyl cyclase assay using cloned kappa-opioid receptors, indicating that changes in the position or orientation of the phenyl ring in this residue did not alter the ability of the peptides to activate the receptor.

Purpose. To synthesize amino acid ester prodrugs of floxuridine (FUdR) and to investigate the effects of structure, stereochemistry, and site of esterification of promoiety on the rates of hydrolysis of these prodrugs in Caco-2 cell homogenates. Methods. Amino acid ester prodrugs of FUdR were synthesized using established procedures. The kinetics of hydrolysis of prodrugs was evaluated in human adenocarcinoma cell line (Caco-2) homogenates and pH 7.4 phosphate buffer. Results. 3′-Monoester, 5′-monoester, and 3′,5′-diester prodrugs of FUdR utilizing proline, L-valine, D-valine, L-phenylalanine, and D-phenylalanine as promoieties were synthesized and characterized. In Caco-2 cell homogenates, the L-amino acid ester prodrugs hydrolyzed 10 to 75 times faster than the corresponding D-amino acid ester prodrugs. Pro and Phe ester prodrugs hydrolyzed much faster (3- to 30-fold) than the corresponding Val ester prodrugs. Further, the 5′-monoester prodrugs hydrolyzed significantly faster (3-fold) than the 3′,5′-diester prodrugs. Conclusions. Novel amino acid ester prodrugs of FUdR were successfully synthesized. The results presented here clearly demonstrate that the rate of FUdR prodrug activation in Caco-2 cell homogenates is affected by the structure, stereochemistry, and site of esterification of the promoiety. Finally, the 5′-Val and 5′-Phe monoesters exhibited desirable characteristics such as good solution stability and relatively fast enzymatic conversion rates.

Novel N-terminus-to-side-chain cyclic analogs of the opioid peptide dynorphin (Dyn) A-(1-11)NH(2) were prepared that retain the basicity of the N-terminal amine and restrict the backbone conformation around the important Tyr(1) residue. Cyclic peptides were synthesized in which the N-terminal amine and the N(epsilon)-amine of a Lys at position 3 or 5 were attached to the alpha-carbon and carbonyl of an acetyl group, respectively. Several synthetic strategies were explored with detailed analysis of the side reactions in order to obtain the desired cyclic peptides. One of the side reactions observed involved premature loss of the N-terminal 9-fluorenylmethoxycarbonyl (Fmoc) group during the neutralization step following deprotection of the Mtt (4-methyltrityl) protecting group from the side chain of Lys. The successful strategy involved the synthesis of the linear peptide up through Gly(2) and functionalization through the N(epsilon)-amine of Lys. A linear N-terminal alkylated analog was prepared by alkylation of the peptide on the resin with an equimolar amount of bromoacetamide, followed by treatment of the peptide with Fmoc-OSu prior to cleavage from the resin to facilitate separation by reversed phase high performance liquid chromatography of unreacted peptide from the desired alkylated product. The novel N-terminal cyclic Dyn A analogs and the linear analog were evaluated for their opioid receptor affinities. These peptides exhibited large losses in affinity for opioid receptors; the low affinity of the linear N-terminal alkylated peptide suggested that the alpha-acetamide group on the N-terminal amine resulted in unfavorable interactions with opioid receptors.

Precipitation of weakly basic drugs in intestinal fluids can affect oral drug absorption. In this study, the implications of self-association of brivanib alaninate in acidic aqueous solution, leading to supersaturation at basic pH condition, on its solubility and oral absorption were investigated. Self-association of brivanib alaninate was investigated by proton NMR spectroscopy, surface tension measurement, dynamic light scattering, isothermal titration calorimetry, and molecular modeling. Drug solubility was determined in various pH media, and its tendency to supersaturate upon pH shift was investigated in buffered and biorelevant aqueous solutions. Pharmacokinetic modeling of human oral drug absorption was utilized for parameter sensitivity analyses of input variables. Brivanib alaninate exhibited continuous, and pH- and concentration-dependent self-association. This phenomenon resulted in positive deviation of drug solubility at acidic pH and the formation of a stable supersaturated drug solution in pH-shift assays. Consistent with the supersaturation phenomenon observed in vitro, oral absorption simulations necessitated invoking long precipitation time in the intestine to successfully predict in vivo data. Self-association of a weakly basic drug in acidic aqueous solution can increase its oral absorption by supersaturation and precipitation resistance at the intestinal pH. This consideration is important to the selection of parameters for oral absorption simulation.

Pro drugs are biologically inactive drug molecules that undergo an enzymatic and/or chemical transformation in vivo to release the active parent drug. Pro drugs may be developed serendipitously or through rationale drug design with an objective to improve a drug’s pharmaceutical properties (solubility, permeability, stability, taste masking and so on), pharmacokinetic properties (absorption, distribution, metabolism and excretion) or site-specific delivery [1–3]. The pro drug approach has been very successful in improving developability of the molecules, with almost 15 of the top 100 best selling small- molecule drugs in 2009 being classified as pro drugs. One interesting example of these blockbuster pro drugs is lisdexamfetamine dimesylate, the oral pro drug of the psychostimulant dextroamphetamine containing the amino acid lysine as a promoiety. This amino acid pro drug was designed to have less abuse potential than other amphetamines due to the slower release of the active parent drug if inhaled or injected. Also valacyclovir belongs to this group of the best-selling drugs. The pro drug approach is commonly used to improve oral bioavailability of drugs whose absorption is limited due to low aqueous solubility and/or permeability. As such, oral pro drugs should provide improved solubility/permeability, have high chemical stability, and rapidly and quantitatively convert to parent drug to maximize drug exposure and minimize unwanted metabolism. This could be achieved by careful selection of the promoiety and in vitro and in vivo evaluation of the pro drug until a desired profile is achieved. Amino acids as promoieties

The purpose of this study was to investigate the utility of stably transfected MDCK-hPepT1 cells for identifying peptide transporter substrates in early drug discovery and compare the characteristics of this cell line with Caco-2 cells. MDCK-hPepT1, MDCK-mock, and Caco-2 cells grown to confluence on 24-well Transwell were used for this study. Expression levels of different transporter proteins (PepT1, PepT2, P-gp) in these cell lines were assessed by qRT-PCR. Permeability studies were conducted in parallel in all the cells with a diverse set of peptide substrates using the optimized experimental condition: 100 microM, apical pH 6.0, basolateral pH 7.4, 2 hr incubation at 37 degrees C. Permeability studies were also conducted with classical P-gp substrates (tested in bi-directional mode) and paracellularly absorbed probes to investigate the differences between the cell lines. As expected, MDCK-hPepT1 cells express significantly higher level of PepT1 mRNA compared to both Caco-2 and MDCK-mock cells. Efflux transporter, P-gp, was expressed adequately in all the cell lines. Permeability studies demonstrated that classical peptide substrates had significantly higher permeability in stably transfected MDCK-hPepT1 cells compared to MDCK-mock and Caco-2 cells. The transfected MDCK-hPepT1 cells were qualitatively similar to Caco-2 cells with respect to functional P-gp efflux activity and paracellular pore activity. Stably transfected MDCK-hPepT1 cells have been demonstrated as a viable alternative to Caco-2 cells for estimating the human absorption potential of peptide transporter substrates. These cells behave similar to Caco-2 cells with regards to P-gp efflux and paracellular pore activity but demonstrate greater predictability of absorption values for classical peptide substrates (for which Caco-2 cells under-estimate oral absorption).

The human intestinal oligopeptide transporter (PEPT1) facilitates the absorption of dipeptides, tripeptides, and many peptidomimetic drugs. In this study, a large number of peptides were selected to investigate the structural features required for PEPT1 transport. Binding affinity was determined in a Gly-Sar uptake inhibition assay, whereas functional transport was ranked in a membrane depolarization assay. Although most of the peptides tested could bind to PEPT1, not all were substrates. As expected, single amino acids and tetrapeptides could not bind to or be transported by PEPT1. Dipeptide transport was influenced by charge, hydrophobicity, size, and side chain flexibility. The extent of transport was variable, and unexpectedly, some dipeptides were not substrates of PEPT1. These included dipeptides with two positive charges or extreme bulk in either position 1 or 2. Our results identify key features required for PEPT1 transport in contrast to most previously described pharmacophores, which are based on the inhibition of transport of a known substrate.

Amino acid ester prodrugs of antiviral and anticancer nucleoside drugs were developed to improve oral bioavailability or to reduce systemic toxicity. We studied the interaction of human concentrative nucleoside transporter (hCNT2) cloned from intestine with various amino acid ester prodrugs of floxuridine (FUdR) and 5,6-dichloro-2-bromo-1-beta-D-ribofuranosylbenzimidazole (BDCRB). Na(+)-dependent uptakes of [(3)H]-inosine and [(3)H]-adenosine were measured in U251 cells transiently expressing intestinal hCNT2. FUdR significantly inhibited the uptake of both [(3)H]-inosine and [(3)H]-adenosine (60-70% of control), while its amino acid ester prodrugs including Val, Phe, Pro, Asp, and Lys esters exhibited markedly decreased inhibition potency (10-30% of control). On the other hand, BDCRB and its amino acid prodrugs markedly inhibited the uptake of both [(3)H]-inosine and [(3)H]-adenosine. Val, Phe, and Pro ester prodrugs of BDCRB showed similar inhibition capacities as parent compound BDCRB (80-90% for adenosine and 60-80% for inosine). The amino acid site of attachment (3'- and 5'-monoesters) and stereochemistry (L- and D-amino acid esters), did not significantly affect the uptake of [(3)H]-inosine and [(3)H]-adenosine. These results demonstrate that the hCNT2 favorably interacts with BDCRB and its amino acid prodrugs, compared to those of FUdR, and that neutral amino acid esters of BDCRB have a high affinity toward this transporter. Therefore, the intestinal hCNT2 may be a target transporter as a factor for modulating oral pharmacokinetics of BDCRB prodrugs.

Amino acid ester prodrugs of 2-bromo-5,6-dichloro-1-(beta-d-ribofuranosyl)benzimidazole (BDCRB) were synthesized and evaluated for their affinity for hPEPT1, an intestinal oligopeptide transporter. Assays of competitive inhibition of [(3)H]glycylsarcosine (Gly-Sar) uptake in HeLa/hPEPT1 cells by the amino acid ester prodrugs of BDCRB suggested their 2- to 4-fold higher affinity for hPEPT1 compared to BDCRB. Further, promoieties with hydrophobic side chains and l-configuration were preferred by the hPEPT1 transporter.

Bioinformatics tools such as Perl, Visual Basic, Cluster, and TreeView were used to analyze public gene expression databases in order to identify potential enzyme targets for prodrug strategies. The analyses indicated that prolidase might be a desirable enzyme target based on its differential expression in melanoma cancer cell lines and its high substrate specificity for dipeptides containing proline at the carboxy terminus. RT-PCR expression of prolidase and hydrolytic activity against N-glycyl-L-proline (GLY-PRO), a standard substrate of prolidase, determined in tumor cell lines, exhibited a high correlation (r 2 = 0.95). These results suggest the possibility of targeting prolidase with prodrugs of anticancer agents for enhanced selectivity. The feasibility of such a scenario was tested by (a) synthesizing prodrugs of melphalan that comprised linkage of the carboxy terminus of the L-phenylalanine moiety of melphalan to the N-terminus of L and D stereoisomers of proline and (b) determining their bioconversion and antiproliferative activities in SK-MEL-5 cells, a melanoma cancer cell line with high expression levels of prolidase. The results of hydrolysis studies of the L- and D-proline prodrugs of melphalan, designated as prophalan-L and prophalan-D, respectively, indicated a ∼7-fold higher rate of activation of prophalan-L compared to prophalan-D in SK-MEL-5 cell homogenates. Prophalan-L exhibited cytotoxicity (Gl 50 = 74.8 μM) comparable to that of melphalan (Gl 50 = 57.0 μM) in SK-MEL-5 cells while prophalan-D was ineffective, suggesting that prolidase-specific activation to the parent drug may be essential for cytotoxic action. Thus, melphalan prodrugs such as prophalan-L that are cleavable by prolidase offer the potential for enhanced selectivity by facilitating cytotoxic activity only in cells overexpressing prolidase.

PepT1 is a transporter of proven pharmaceutical utility for enhancing oral absorption. A high-throughput, robust functional assay, capable of distinguishing PepT1 binders from substrates, allowing identification and/or prediction of drug candidate activation was developed. An MDCK epithelial cell line was transfected with rPepT1. The high level of stable rPepT1 expression that was achieved enabled development of a miniaturized PepT1 assay in a 96-well format, which could be scaled to 384 wells. The assay is based on measurement of membrane depolarization resulting from the cotransport of protons and PepT1 substrates. Membrane potential changes are tracked with a voltage-sensitive fluorescent indicator. Control (mock-transfected) cells are used to determine nonspecific membrane potential changes. A variety of fluorescent dyes were tested during initial assay design, including intracellular pH and membrane potential indicators. A membrane potential indicator was chosen because of its superior performance. Upon PepT1 activation with glycylsarcosine, dose-dependent membrane depolarization was observed with an EC50 of 0.49 mM. Maximum depolarization was dependent on the level of PepT1 expression. Testing of 38 known PepT1 substrates, binders, and nonbinders demonstrated that this assay accurately distinguished substrates from binders and from nonbinders. Initial validation of this novel assay indicates that it is sensitive and robust, and can distinguish between transporter substrates and antagonists. This important distinction has been previously achieved only with lower-throughput assays. This assay might also be used to determine substrate potency and establish a high-quality data set for PepT1 SAR modeling.

A novel N-terminal-to-side chain cyclic dynorphin A analogue lacking the basic N-terminus was designed based on Ac[Lys(2),Trp(3),Trp(4),d-Ala(8)]dynorphin A-(1-11)NH(2) (Wan et al. J. Med. Chem. 1999, 42, 3011-3013). cyclo(N,5)[Trp(3),Trp(4),Glu(5)]dynorphin A-(1-11)NH(2) showed similar kappa opioid receptor affinity (K(i) = 27 nM) and selectivity (K(i) ratio (kappa/mu/delta) = 1/12/330) to the linear peptide and antagonized dynorphin A-(1-13)NH(2) at kappa opioid receptors. This is the first opioid peptide cyclized through the N-terminus that retains high opioid receptor affinity.

Although dynorphin A (Dyn A) is thought to be an endogenous ligand for kappa opioid receptors, it also exhibits high affinity for mu and delta opioid receptors [1]. The flexibility of Dyn A may be responsible for its lack of selectivity. A large number of side chain-to-side chain cyclic analogs of Dyn A have been synthesized in order to reduce its flexibility and increase its affinity and selectivity for k receptors [see ref. 2 for a review]. Tyr1 is one of the key pharmacophoric groups in Dyn A [3], but none of the cyclic analogs of Dyn A synthesized to date restrict the conformation of this important residue. Here we describe the synthesis of novel head-to-side chain cyclic analogs of Dyn A that constrain this important residue and retain a basic N-terminal amine (Figure 1).