Physicochemical determinants of passive membrane permeability: role of solute hydrogen-bonding potential and volume.

The relationship of solute structure with cellular permeability was probed. Two series of dipeptide mimetics consisting of glycine, alanine, valine, leucine, phenylalanine, and cyclohexylalanine with amino acids in the D-configuration were prepared. Partition coefficients for the peptidemimetics were obtained in the octanol/water (log P(octanol/water)), hydrocarbon/octanol (Delta log P), and heptane/ethylene glycol (log P(heptane/glycol)) systems in order to explore the contributions of solute volume, or surface area, and hydrogen-bond potential to the permeability of the solutes. Permeability coefficients were obtained in Caco-2 cell monolayers as a model of the human intestinal mucosa. The results were interpreted in terms of a partition/diffusion model for solute transport where membrane partitioning into the permeability-limiting membrane microdomain is estimated from the solvent partition coefficients. Neither log P(octanol/water) nor Delta log P alone correlated with cellular permeability for all the solutes. In contrast, log P(heptane/glycol) gave a qualitatively better correlation. With regard to solute properties, log P(octanol/water) is predominantly a measure of solute volume, or surface area, and hydrogen-bond acceptor potential, while Delta log P is principally a measure of hydrogen-bond donor strength. Log P(heptane/glycol) contains contributions from all these solute properties. The results demonstrate that both hydrogen-bond potential and volume of the solutes contribute to permeability and suggests that the nature of the permeability-limiting microenvironment within the cell depends on the properties of a specific solute. Collectively, these findings support the conclusion that a general model of permeability will require consideration of a number of different solute structural properties.

Small molecule peptidomimetics containing a novel phosphotyrosine bioisostere inhibit protein tyrosine phosphatase 1B and augment insulin action.

Protein tyrosine phosphatase 1B (PTP1B) attenuates insulin signaling by catalyzing dephosphorylation of insulin receptors (IR) and is an attractive target of potential new drugs for treating the insulin resistance that is central to type II diabetes. Several analogues of cholecystokinin(26)(-)(33) (CCK-8) were found to be surprisingly potent inhibitors of PTP1B, and a common N-terminal tripeptide, N-acetyl-Asp-Tyr(SO(3)H)-Nle-, was shown to be necessary and sufficient for inhibition. This tripeptide was modified to reduce size and peptide character, and to replace the metabolically unstable sulfotyrosyl group. This led to the discovery of a novel phosphotyrosine bioisostere, 2-carboxymethoxybenzoic acid, and to analogues that were >100-fold more potent than the CCK-8 analogues and >10-fold selective for PTP1B over two other PTP enzymes (LAR and SHP-2), a dual specificity phosphatase (cdc25b), and a serine/threonine phosphatase (calcineurin). These inhibitors disrupted the binding of PTP1B to activated IR in vitro and prevented the loss of tyrosine kinase (IRTK) activity that accompanied PTP1B-catalyzed dephosphorylation of IR. Introduction of these poorly cell permeant inhibitors into insulin-treated cells by microinjection (oocytes) or by esterification to more lipophilic proinhibitors (3T3-L1 adipocytes and L6 myocytes) resulted in increased potency, but not efficacy, of insulin. In some instances, PTP1B inhibitors were insulin-mimetic, suggesting that in unstimulated cells PTP1B may suppress basal IRTK activity. X-ray crystallography of PTP1B-inhibitor complexes revealed that binding of an inhibitor incorporating phenyl-O-malonic acid as a phosphotyrosine bioisostere occurred with the mobile WPD loop in the open conformation, while a closely related inhibitor with a 2-carboxymethoxybenzoic acid bioisostere bound with the WPD loop closed, perhaps accounting for its superior potency. These CCK-derived peptidomimetic inhibitors of PTP1B represent a novel template for further development of potent, selective inhibitors, and their cell activity further justifies the selection of PTP1B as a therapeutic target.

An NMR study of conformations of substituted dipeptides in dodecylphosphocholine micelles: implications for drug transport.

Efficient transport of intact drug (solute) across the intestinal epithelium is typically a requirement for good oral activity. In general, the membrane permeability of a solute is a complex function of its size, lipophilicity, hydrogen bond potential, charge, and conformation. In conjunction with theoretical/computational and in vitro drug transport studies, seven dipeptide (R(1)-D-Xaa-D-Phe-NHMe) homologues were each dissolved in a micellar d(38)-dodecylphosphocholine solvent system. In this homologous dipeptide series, factors such as size, lipophilicity, hydrogen-bond potential, and charge were either tightly controlled or well-characterized by other methods in order to investigate by nmr how conformational factors relate to transport. Nuclear Overhauser effect spectroscopy experiments and amide-NH-H(2)O chemical exchange rates showed that the five more lipophilic dipeptides were predominately associated with micelle, whereas the two less lipophilic analogues were not. Rotating frame nuclear Overhauser effect spectroscopy derived interproton distance restraints for each analogue, along with (3)J(HH)-derived dihedral restraints, were used in molecular dynamics/simulated annealing computations. Our results suggest that-other factors being equal-flexible dipeptides having a propensity to fold together nonpolar N- and C-terminal moieties allow greater segregation of polar and nonpolar domains and may possess enhanced transport characteristics. Dipeptides that were less flexible or that retained a less amphiphilic conformation did not have comparably enhanced transport characteristics. We suggest that these conformational/transport correlations may hold true for small, highly functionalized solutes (drugs) in general.

Mechanism, structure-activity studies, and potential applications of glutathione S-transferase-catalyzed cleavage of sulfonamides.

The mechanism of sulfonamide cleavage of PNU-109112, a potent HIV-1 protease inhibitor, by glutathione-S-transferase (GST) was investigated in the presence of reduced GSH. GST-catalyzed sulfonamide cleavage takes place via the nucleophilic attack of GSH on the pyridine moiety of the substrate with formation of the GS-para-CN-pyridinyl conjugate, the corresponding amine, and sulfur dioxide. Structure activity studies with a variety of sulfonamides indicate that an electrophilic center alpha to the sulfonyl group is required for cleavage. Substituents that withdraw electron density from the carbon atom alpha- to the sulfonyl group facilitate nucleophilic attack by the GS(-) thiolate bound to GST. The rate of sulfonamide cleavage is markedly affected by the nature of the electrophilic group; replacement of para-CN by para-CF(3) on the pyridine ring of PNU-109112 confers stability against sulfonamide cleavage. On the other hand, stability of sulfonamides is less dependent on the nature of the amine moiety. These principles can be applied to the synthesis of sulfonamides, labile toward cellular GST, that may serve as prodrugs for release of bioactive amines. Tumors are particularly attractive targets for these sulfonamide prodrugs as GST expression is significantly up-regulated in many cancer cells. Another potential application could be in organic synthesis, where protection of amines as the corresponding activated sulfonamides can be reversed by GST/GSH under mild conditions.

Strategies toward predicting peptide cellular permeability from computed molecular descriptors.

The therapeutic efficacy of an orally administered drug is dictated not only by its pharmacological properties such as potency and selectivity, but also its pharmacokinetic properties such as its access to the site of activity. Thorough evaluation of the physicochemical and biological barriers to drug delivery is essential to the selection and successful development of drug candidates. We have demonstrated previously that cellular permeability, as a primary component of drug delivery, is principally dependent upon the desolvation potential of the polar functionalities in the molecule and, secondarily, upon the solute lipophilicity [Conradi, R.A., Hilgers, A.R., Ho, N.F.H., Burton, P.S. (1992). The influence of peptide structure on transport across Caco-2 cells. II. Peptide bond modification which results in improved permeability. Pharm. Res. 9, 473-479]. Increasingly sophisticated computational methods are becoming available for describing molecular structural features proposed to correlate with such molecular physicochemical determinants of permeability. Herein we examine the relationships of various computationally derived molecular geometric descriptors for a set of peptides and peptidomimetics, in the context of experimentally measured hydrogen-bond potentials and lipophilicities, with their cellular permeabilities. These descriptors include molecular volume, polar and non-polar surface areas and projected molecular cross-sectional areas. Particular attention is paid to the roles of solvation treatments and other computational factors in descriptor generation, deconvolution of cellular transport mechanisms and statistical analyses of the resulting data for the development of valid, structure-based and mechanistically meaningful models of cellular permeability. No significant correlation of cellular permeability with computed descriptors was found. This was primarily because of our inability to identify surrogates for hydrogen-bond desolvation potential for the solutes from among these descriptors.

Piperazinyl oxazolidinone antibacterial agents containing a pyridine, diazene, or triazene heteroaromatic ring.

Oxazolidinones are a novel class of synthetic antibacterial agents active against gram-positive organisms including methicillin-resistant Staphylococcus aureus as well as selected anaerobic organisms. Important representatives of this class include the morpholine derivative linezolid 2, which is currently in phase III clinical trials, and the piperazine derivative eperezolid 3. As part of an investigation of the structure-activity relationships of structurally related oxazolidinones, we have prepared and evaluated the antibacterial properties of a series of piperazinyl oxazolidinones in which the distal nitrogen of the piperazinyl ring is substituted with a six-membered heteroaromatic ring. Compounds having MIC values

How structural features influence the biomembrane permeability of peptides.

Successful drug development requires not only optimization of specific and potent pharmacological activity at the target site, but also efficient delivery to that site. Many promising new peptides with novel therapeutic potential for the treatment of AIDS, cardiovascular diseases, and CNS disorders have been identified, yet their clinical utility has been limited by delivery problems. Along with metabolism, a major factor contributing to the poor bioavailability of peptides is thought to be inefficient transport across cell membranes. At the present time, the reasons for this poor transport are poorly understood. To explore this problem, we have designed experiments focused on determining the relationship between peptide structure and peptide transport across various biological membranes both in vitro and in vivo. Briefly, peptides that varied systematically in chain length, lipophilicity, and amide bond number were prepared. Permeability results with these solutes support a model in which the principal determinant of peptide transport is the energy required to desolvate the polar amides in the peptide for the peptide to enter and diffuse across the cell membrane. Further impacting on peptide permeability is the presence of active, secretory transport systems present in the apical membrane of intestinal epithelial and brain endothelial cells. In Caco-2 cell monolayers, a model of the human intestinal mucosa, this pathway displayed substrate specificity, saturation, and inhibition. Similar results have been shown in vivo in both rat intestinal and blood-brain barrier absorption models. The presence of such systems serves as an additional transport barrier by returning a fraction of absorbed peptide back to the lumen.

A biophysical model of passive and polarized active transport processes in Caco-2 cells: approaches to uncoupling apical and basolateral membrane events in the intact cell.

This report is aimed at the biophysical modeling of transmembrane events involving a passive diffusion and directional pumplike mechanism at the apical (AP) and basolateral (BL) membranes of cultured cell monolayers. The essence of the model is based on experimental evidences for the existence of a saturable, apically polarized transport system in Caco-2 cells for peptides which hindered apical to basolateral flux, enhanced basolateral to apical flux, and showed substrate specificity. This system was further inhibited by verapamil, suggesting some homology with P-glycoprotein, the principal mediator of drug resistance in multidrug resistant cancer cells. Preliminary evidence was also obtained suggesting an additional polarized uptake system for the same peptides in the basolateral membrane. Upon saturation and/or inhibition of the active transport mechanisms with verapamil, the peptide fluxes in apical-to-basolateral direction and the basolateral-to-apical direction converged and became controlled by the passive mechanism. Since the intent of the modeling was to provide useful templates for the design of probing experiments and to delineate and quantify mass transfer mechanisms at the AP and BL membranes and their interrelationships, theoretical equations were developed for a host of kinetic boundary conditions: (a) AP-->BL and BL-->AP transfluxes, (b) bidirectional effluxes from substrate-preloaded cells, (c) undirectional efflux across the AP or BL membrane from preloaded cells, and (d) uptake kinetics via the AP or BL membrane leading to equilibrium. Furthermore, flux expressions were reduced to membrane permeability coefficients to accommodate passive diffusion, saturation, inhibition, and directionality. The diffusional mass transport resistances of the aqueous boundary layers and microporous filter support of the cell monolayer were necessarily included.

Epithelial cell permeability of a series of peptidic HIV protease inhibitors: aminoterminal substituent effects.

The influence of the aminoterminal substituent in a homologous series of tetrapeptide analogs on transport across Caco-2 cell monolayers was studied. In a series of pyridylcarboxamide regioisomers, the 2-pyridyl isomer was significantly more permeable than either the 3- or 4-congeners. The uniqueness of this peptide was further suggested by examining the partitioning behavior between heptane and ethylene glycol, a system which has been developed as a simple estimate of the desolvation energy or hydrogen bonding potential of a peptide. In this model, the 2-isomer has a much larger partition coefficient than either the 3- or 4-analogs, consistent with its being less solvated than expected based on simple structural considerations. Factors possibly contributing to this decreased effective polarity could be steric interactions or intramolecular hydrogen bonding.

Evidence for a polarized efflux system for peptides in the apical membrane of Caco-2 cells.

The transport of two model peptides across confluent monolayers of human colon adenocarcinoma (Caco-2) cells was studied. In the case of AcPhe(NMePhe)2NH2, transport in the apical to basolateral direction was increased with increasing peptide concentration in the apical compartment. Transport was also increased in the presence of verapamil. In contrast, the flux of AcPheNH2 was neither concentration dependent nor affected by verapamil. Further, in the presence of verapamil, transport in the basolateral to apical direction was showed for AcPhe(NMePhe)2NH2 and again unchanged for AcPheNH2. These results are consistent with the presence of a saturable, apically polarized transport system in Caco-2 cells which serves to hinder transport in the apical to basolateral direction, increase flux in the basolateral to apical direction and shows substrate specificity for these model peptides.

The influence of peptide structure on transport across Caco-2 cells. II. Peptide bond modification which results in improved permeability.

In order to study the influence of hydrogen bonding in the amide backbone of a peptide on permeability across a cell membrane, a series of tetrapeptide analogues was prepared from D-phenylalanine. The amide nitrogens in the parent oligomer were sequentially methylated to give a series containing from one to four methyl groups. The transport of these peptides was examined across confluent monolayers of Caco-2 cells as a model of the intestinal mucosa. The results of these studies showed a substantial increase in transport with each methyl group added. Only slight difference in the octanol-water partition coefficient accompanied this alkylation, suggesting that the increase in permeability is not due to lipophilicity considerations. These observations are, however, consistent with a model in which hydrogen bonding in the backbone is a principal determinant of transport. Methylation is seen to reduce the overall hydrogen bond potential of the peptide and increases flux by this mechanism. These results suggest that alkylation of the amides in the peptide chain is an effective way to improve the passive absorption potential for this class of compounds.

The influence of peptide structure on transport across Caco-2 cells.

The relationship between structure and permeability of peptides across epithelial cells was studied. Using confluent monolayers of Caco-2 cells as a model of the intestinal epithelium, permeability coefficients were obtained from the steady-state flux of a series of neutral and zwitterionic peptides prepared from D-phenylalanine and glycine. Although these peptides ranged in lipophilicity (log octanol/water partition coefficient) from -2.2 to +2.8, no correlation was found between the observed flux and the apparent lipophilicity. However, a strong correlation was found for the flux of the neutral series and the total number of hydrogen bonds the peptide could potentially make with water. These results suggest that a major impediment to peptide passive absorption is the energy required to break water-peptide hydrogen bonds in order for the solute to enter the cell membrane. This energy appears not to be offset by the favorable introduction of lipophilic side chains in the amino acid residues.

Caco-2 cell monolayers as a model for drug transport across the intestinal mucosa.

Human colon adenocarcinoma (Caco-2) cells, when grown on semipermeable filters, spontaneously differentiate in culture to form confluent monolayers which both structurally and functionally resemble the small intestinal epithelium. Because of this property they show promise as a simple, in vitro model for the study of drug absorption and metabolism during absorption in the intestinal mucosa. In the present study, the transport of several model solutes across Caco-2 cell monolayers grown in the Transwell diffusion cell system was examined. Maximum transport rates were found for the actively transported substance glucose and the lipophilic solutes testosterone and salicyclic acid. Slower rates were observed for urea, hippurate, and saliylate anions and were correlated with the apparent partition coefficient of the solute. These results are similar to what is found with the same compounds in other, in vivo absorption model systems. It is concluded that the Caco-2 cell system may give useful predictions concerning the oral absorption potential of new drug substances.

Predictive relationships in the water solubility of salts of a nonsteroidal anti-inflammatory drug.

A number of metal and amine salts of the nonsteroidal anti-inflammatory drug 2-(2-fluoro-4-biphenylyl)propionic acid (flurbiprofen) (1) were synthesized, and the water solubilities of these materials were investigated. The solubility of flurbiprofen versus pH is in excellent agreement with the theoretical profile which assumes an intrinsic solubility of 5.0 X 10(-5) M for the free acid and a pKa of 4.22. The solubility determination of 1:1 amine salts of slightly soluble carboxylic acids is complicated by possible precipitation of the free acid when the salt is dissolved in water. The lower than expected apparent salt solubility seen when this phenomenon occurs is referred to as the "stoichiometric solubility." A theoretical treatment delineating three distinct regions in the phase solubility diagram for the tromethamine salt of flurbiprofen illustrates the care which must be taken in characterizing the system for salt solubility determinations. A multiple linear regression analysis was carried out to determine the relationship between the log Ksp of six amine salts and their melting points and cation hydrophilicities. While a strong dependence of Ksp on melting point was found, there was not a significant correlation between Ksp and amine cation hydrophilicity.

Strategies in the design of solution-stable, water-soluble prodrugs III: influence of the pro-moiety on the bioconversion of 21-esters of corticosteroids.

Succinate esters and many other carboxylic acid esters utilized as water-soluble prodrugs have limited utility due to their aqueous solution instability. In an earlier study, a strategy for the design of solution-stable 21-carboxylic acid esters of corticosteroids was developed from a consideration of various physical organic factors which influence ester hydrolysis. Several 21-esters of methylprednisolone were synthesized, and their stability in aqueous solution was monitored to demonstrate the feasibility of the approach. In this study, the bioconversion of representative examples of 21-esters of methylprednisolone exhibiting shelf lives of greater than or equal to 2 years at 25 degrees C was monitored to evaluate their utility as prodrugs in comparison to a commercially marketed sodium succinate ester. Ester hydrolysis studies conducted in human and rhesus monkey serum suggest that derivatives having an anionic solubilizing moiety (sulfonate or carboxylate) are not hydrolyzed in serum, while compounds having a cationic (tertiary amino) solubilizing moiety are hydrolyzed rapidly by serum esterases. In vivo pharmacokinetic studies in rhesus monkeys were also conducted to compare the bioconversion rates and overall bioavailability of several solution-stable prodrugs with the 21-succinate ester. Derivatives having solution stability exceeding 2 years at 25 degrees C with a faster bioconversion rate and an overall bioavailability equal to or higher than that of the succinate ester have been identified. Relative bioavailability appears to be highly sensitive to the charge of the solubilizing pro-moiety and pro-moiety chain length.

Strategies in the design of solution-stable, water-soluble prodrugs I: a physical-organic approach to pro-moiety selection for 21-esters of corticosteroids.

The ideal water-soluble prodrug should exhibit sufficient aqueous solution stability to allow long-term storage of its solutions (i.e., 2 years at room temperature) and yet should be converted rapidly in vivo to the active parent drug--two severe and seemingly conflicting demands which limit the utility of many common solubilizing pro-moieties. For example, succinate esters, which are commonly utilized as water-soluble prodrugs, are unstable in solution and may undergo slow and incomplete bioconversion in vivo. In this study, the solution stability problems associated with 21-esters of corticosteroids are reviewed. It is concluded that the most important reaction limiting shelf life is ester hydrolysis. From a consideration of the influence of molecular structure on ester reactivity, a strategy for the design of solution-stable, water-soluble prodrugs of corticosteroids has been developed. Two key requirements for dilute solution stability are high solubility at the pH of optimum stability and appropriate design of the pH-rate profile. Several 21-esters of methylprednisolone have been synthesized, and the rates of their aqueous solution hydrolysis have been determined to test the strategy. Compounds exhibiting estimated shelf lives in dilute solution of greater than 2 years at 25 degrees C have been identified.

Strategies in the design of solution-stable, water-soluble prodrugs II: properties of micellar prodrugs of methylprednisolone.

In a previous study, a physical-organic approach to the design of solution-stable, water-soluble prodrugs of the corticosteroid methylprednisolone was outlined, and several 21-esters were synthesized to test the approach. Compounds exhibiting dilute solution stabilities approaching 2 years at 25 degrees C were reported. A complicating factor in more concentrated aqueous solutions of water-soluble prodrugs, however, is the limited extent to which hydrolysis can occur before the solution becomes saturated with respect to the relatively insoluble parent drug. In this study the advantages of micellar prodrugs as water-soluble delivery systems for parenteral administration of relatively insoluble parent drugs are explored. Micellar prodrugs, besides being highly water soluble, have additional advantages in that their micelles solubilize poorly soluble degradation products which may otherwise precipitate and may act as a self-stabilizing influence due to protection of the hydrolytically labile prodrug linkage within the micelle interior. Two 21-esters of methylprednisolone previously identified as having promising dilute solution stability have now been shown to self-associate in aqueous solution at higher concentrations, as determined by solubility, kinetic, and light-scattering measurements. One consequence of self-association is that free methylprednisolone, the product of prodrug hydrolysis, is solubilized in concentrated prodrug formulations. In addition, acid- and base-catalyzed hydrolysis rate constants are altered in the micelles, resulting in further prolongation of shelf life in concentrated solutions. Due to the added benefits of self-micellization, the water-soluble 21-esters investigated exhibit shelf lives exceeding 2 years at 30 degrees C, the upper limit of the controlled room temperature range.

Carboxyl group catalysis of acyl transfer reactions in corticosteroid 17- and 21-monoesters.

Succinate esters, although frequently employed as water-soluble prodrugs of poorly soluble parent drugs, are not sufficiently stable to allow long-term storage in solution. Intramolecular catalysis of ester hydrolysis by the terminal succinate carboxyl group is a contributing factor to this instability. Methylprednisolone 21-succinate has recently been reported to undergo both hydrolysis and 21 in equilibrium 17 acyl migration in aqueous solutions. Intramolecular catalysis by the terminal carboxyl group is seen in both reactions, but the catalytic mechanisms are not well understood. While acyl migration can only be catalyzed via the carboxyl group acting as a general acid or general base, hydrolysis may undergo either nucleophilic or general acid-base catalysis. To gain further insight into the catalytic mechanism, hydrolysis of methylprednisolone 21-succinate was carried out in aniline buffers to trap any succinic anhydride (as the anilide) that would form if the catalysis were nucleophilic. The nucleophilic mechanism was shown to account for only 15-20% of the overall catalysis. Comparisons of the rates of the intramolecularly catalyzed reactions of methylprednisolone 21- and 17-succinate were made with the same reactions of methylprednisolone 21- and 17-acetate catalyzed intermolecularly by acetate ion. Interestingly, intramolecular catalysis appears to favor acyl migration over hydrolysis. Hence, the hydrolysis of methylprednisolone 21-succinate is faster in basic solutions (pH greater than 7.4), while acyl migration becomes the dominant reaction in the catalyzed region of the pH profile between pH 3.6 and 7.4. Arguments are presented to account for these differences in catalytic efficiency in terms of the transition-state structures for the two reactions.

Influence of premicellar and micellar association on the reactivity of methylprednisolone 21-hemiesters in aqueous solution.

Self-association of drug molecules at formulation concentrations can have a major impact on formulation properties. In this study a homologous series of methylprednisolone 21-hemiesters were found to undergo self-association in aqueous solution. The effect of aggregate formation on the solution degradation of these compounds was examined. To determine the nature and extent of association of these steroidal esters, partition coefficients between butyronitrile and aqueous buffer (pH 8.5) were measured as a function of ester concentration. The partitioning data were found to be consistent with dimer formation at low concentration followed by true micelle formation at higher concentration. Chain length increases favored micelle formation, but appeared to have little effect on dimerization. The first-order rate constants for ester hydrolysis and 21 leads to 17 acyl migration in aqueous buffer (pH 8.5) were also found to be dependent on ester concentration. The kinetic data are consistent with a model which assumes stabilization by both dimer and micelle formation, the limiting factor at high concentration being the reactivity of the ester in the micelles. The degree of stabilization due to self-association was found to increase with chain length.