Midostaurin: a magic bullet that blocks mast cell expansion and activation.

Clinically relevant features in patients with systemic mastocytosis (SM) include the cosmetic burden of lesional skin, mediator-related symptoms, and organ damage resulting from mast cell (MC) infiltration in advanced forms of SM. Regardless of the SM variant, expansion of neoplastic MC in the skin and other organs is triggered by mutant forms of KIT, the most prevalent being D816V. Activation of MC with subsequent release of chemical mediators is often caused by IgE-dependent mechanisms in these patients. Midostaurin, also known as PKC412, blocks the kinase activity of wild-type KIT and KIT D816V, counteracts KIT-dependent growth of neoplastic MC, and inhibits IgE-dependent mediator secretion. Based on this activity-profile, the drug has been used for treatment of patients with advanced SM. Indeed, encouraging results have been obtained with the drug in a recent multi-center phase II trial in patients with advanced SM, with an overall response rate of 60% and a substantial decrease in the burden of neoplastic MC in various organs. Moreover, midostaurin improved the overall survival and relapse-free survival in patients with advanced SM compared with historical controls. In addition, midostaurin was found to improve mediator-related symptoms and quality of life, suggesting that the drug may also be useful in patients with indolent SM suffering from mediator-related symptoms resistant to conventional therapies or those with MC activation syndromes. Ongoing and future studies will determine the actual value of midostaurin-induced MC depletion and MC deactivation in these additional indications.

Target interaction profiling of midostaurin and its metabolites in neoplastic mast cells predicts distinct effects on activation and growth.

Proteomic-based drug testing is an emerging approach to establish the clinical value and anti-neoplastic potential of multikinase inhibitors. The multikinase inhibitor midostaurin (PKC412) is a promising new agent used to treat patients with advanced systemic mastocytosis (SM). We examined the target interaction profiles and the mast cell (MC)-targeting effects of two pharmacologically relevant midostaurin metabolites, CGP52421 and CGP62221. All three compounds, midostaurin and the two metabolites, suppressed IgE-dependent histamine secretion in basophils and MC with reasonable IC(50) values. Midostaurin and CGP62221 also produced growth inhibition and dephosphorylation of KIT in the MC leukemia cell line HMC-1.2, whereas the second metabolite, CGP52421, which accumulates in vivo, showed no substantial effects. Chemical proteomic profiling and drug competition experiments revealed that midostaurin interacts with KIT and several additional kinase targets. The key downstream regulator FES was recognized by midostaurin and CGP62221, but not by CGP52421 in MC lysates, whereas the IgE receptor downstream target SYK was recognized by both metabolites. Together, our data show that the clinically relevant midostaurin metabolite CGP52421 inhibits IgE-dependent histamine release, but is a weak inhibitor of MC proliferation, which may have clinical implications and may explain why mediator-related symptoms improve in SM patients even when disease progression occurs.

Inhibition of CXCR4 in CML cells disrupts their interaction with the bone marrow microenvironment and sensitizes them to nilotinib.

Drug resistance is a growing area of concern. It has been shown that a small, residual pool of leukemic CD34+ progenitor cells can survive in the marrow microenvironment of chronic myeloid leukemia (CML) patients after years of kinase inhibitor treatment. Bone marrow (BM) stroma has been implicated in the long-term survival of leukemic cells, and contributes to the expansion and proliferation of both transformed and normal hematopoietic cells. Mechanistically, we found that CML cells expressed CXCR4, and that plerixafor diminished BCR-ABL-positive cell migration and reduced adhesion of these cells to extra cellular-matrix components and to BM stromal cells in vitro. Moreover, plerixafor decreased the drug resistance of CML cells induced by co-culture with BM stromal cells in vitro. Using a functional mouse model of progressive and residual disease, we demonstrated the ability of the CXCR4 inhibitor, plerixafor, to mobilize leukemic cells in vivo, such that a plerixafor-nilotinib combination reduced the leukemia burden in mice significantly below the baseline level suppression exhibited by a moderate-to-high dose of nilotinib as single agent. These results support the idea of using CXCR4 inhibition in conjunction with targeted tyrosine kinase inhibition to override drug resistance in CML and suppress or eradicate residual disease.

Imatinib: a selective tyrosine kinase inhibitor.

The understanding of the pathophysiology of a large number of cancer types provides a strategy to target cancer cells with minimal effect on normal cells. Protein phosphorylation and dephosphorylation play a pivotal role in intracellular signaling; to regulate signal transduction pathways, there are approximately 700 protein kinases and 100 protein phosphatases encoded within the human genome. In cancer, as well as in other proliferative diseases, unregulated cell proliferation, differentiation and survival frequently results from abnormal protein phosphorylation. Although it is often possible to identify a single kinase that plays a pivotal role in a given disease, the development of drugs based upon protein kinase inhibition has been hampered by unacceptable side effects resulting from a lack of target selectivity. With the growing understanding of the molecular biology of protein tyrosine kinases and the use of structural information, the design of potential drugs directed towards the bind adenosine triphosphate (ATP)-binding site of a single target has become possible. These advances have transferred emphasis away from the identification of potent kinase inhibitors and more towards issues of target selectivity, cellular efficacy, therapeutic effectiveness and tolerability. In this paper, the relationship between molecular biology and drug discovery methods, as utilized for the identification of anticancer drugs, will be illustrated.

Synthesis and characterization of a novel tritiated KATP channel opener with a benzopyran structure.

The synthesis of a tritiated benzopyran-type opener of the ATP-dependent K+ channel (KATP channel), [3H]-PKF217 - 744 (3S,4R)-N-[3,4-dihydro-2,2-dimethyl-3-hydroxy-6-(2-methyl-4-pyridinyl)-2H-1-benzopyran-4-yl]-3-[2,6-3H]pyridinecarboxamide with a specific activity of 50 Ci mmol(-1) is described. Binding of the ligand was studied in membranes from human embryonic kidney cells transfected with the sulphonylurea receptor isoforms, SUR2B and SUR2A, respectively. PKF217 - 744 was confirmed as being a KATP channel opener by its ability to open the Kir6.1/SUR2B channel, the recombinant form of the vascular KATP channel, and to inhibit binding of the pinacidil analogue, [3H]-P1075, to SUR2B (Ki=26 nM). The kinetics of [3H]-PKF217 - 744 binding to SUR2B was described by rate constants of association and dissociation of 6.9x10(6) M(-1) min(-1) and 0.09 min(-1), respectively. Binding of [3H]-PKF217 - 744 to SUR2B/2A was activated by MgATP (EC50 approximately 3 microM) and inhibited (SUR2B) or enhanced (SUR2A) by MGADP: Binding of [3H]-PKF217 - 744 to SUR2B was inhibited by representatives of the different structural classes of openers and sulphonylureas. Ki values were identical with those obtained using the opener [3H]-P1075 as the radioligand. Glibenclamide accelerated dissociation of the SUR2B-[3H]-PKF217 - 744 complex. The data show that the affinity of [3H]-PKF217 - 744 binding to SUR2B is approximately 6 times lower than that of [3H]-P1075. This is due to a surprisingly slow association rate of the benzopyran-type ligand, suggesting a complex mechanism of opener binding to SUR. The other pharmacological properties of the two opener radioligands are identical.

Su-6668. SUGEN.

SUGEN is developing SU-6668, a tyrosine kinase inhibitor that inhibits three distinct growth factor receptor targets, for the potential treatment of cancer [304530]. The compound is in phase I trials in the UK and US [321260], [374505]. A report in January 2001 stated that phase I/II trials for hematological and solid tumors were expected to commence shortly thereafter [395657]. In May 2001, phase I data from a dose-escalation study conducted at UCLA were presented at the 37th ASCO meeting in San Francisco, CA. By that time, 74 patients had been enrolled in this study which aimed to determine the toxicities of SU-6668 when delivered to fed and fasting patients. SU-6668 was administered orally either once or twice-daily at doses of 100 to 2400 mg/m2 to patients diagnosed as having advanced malignancies. Accrual in phase I is continuing to define the toxicities of doses > 200 mg/m2 twice-daily [409984], [411418]. In December 1998, SUGEN filed an IND with the FDA for the clinical testing of this compound with oral and iv formulations [310237]. In November 1998, SUGEN entered into a collaboration with the Cancer Research Campaign (CRC) to conduct a phase I trial of SU-6668 at the Royal Marsden Hospital, UK using an iv formulation [304530]. After the first six patients had been treated, the trial was halted owing to problems with the iv formulation. Some toxicities, including nausea, vomiting, fatigue and tumor pain were observed [413538]. No licensing agreement as involved between the CRC and SUGEN for this trial [408572].

New anilinophthalazines as potent and orally well absorbed inhibitors of the VEGF receptor tyrosine kinases useful as antagonists of tumor-driven angiogenesis.

The sprouting of new blood vessels, or angiogenesis, is necessary for any solid tumor to grow large enough to cause life-threatening disease. Vascular endothelial growth factor (VEGF) is one of the key promoters of tumor induced angiogenesis. VEGF receptors, the tyrosine kinases Flt-1 and KDR, are expressed on vascular endothelial cells and initiate angiogenesis upon activation by VEGF. 1-Anilino-(4-pyridylmethyl)-phthalazines, such as CGP 79787D (or PTK787 / ZK222584), reversibly inhibit Flt-1 and KDR with IC(50) values < 0.1 microM. CGP 79787D also blocks the VEGF-induced receptor autophosphorylation in CHO cells ectopically expressing the KDR receptor (ED(50) = 34 nM). Modification of the 1-anilino moiety afforded derivatives with higher selectivity for the VEGF receptor tyrosine kinases Flt-1 and KDR compared to the related receptor tyrosine kinases PDGF-R and c-Kit. Since these 1-anilino-(4-pyridylmethyl)phthalazines are orally well absorbed, these compounds qualify for further profiling and as candidates for clinical evaluation.

Synthesis of and radioligand binding studies with a tritiated pinacidil analogue: receptor interactions of structurally different classes of potassium channel openers and blockers.

The synthesis of N-cyano-N'-[1,1-dimethyl-[2,2,3,3-3H]propyl]-N"-(3- pyridinyl)guanidine, [3H]-15, is described. The utility of this tritiated radioligand in characterizing the interactions of potassium channel openers and blockers with their receptors is demonstrated. Potassium channel openers of the pinacidil, cromakalim, aprikalim, diazoxide, and minoxidil types, as well as KATP channel blockers of the glibenclamide and eosine types, are all capable of displacing [3H]-15 from its receptor. The results indicate that all of these compounds interact with the same target protein, but that several different allosterically coupled receptor binding sites are probably involved. The highly significant correlation between the ability of the structurally diverse potassium channel openers to inhibit [3H]-15 binding and to relax vascular smooth muscle is consistent with their receptor binding sites being closely associated with the potassium channel protein which is the functional target of this class of drugs.

Binding of the K+ channel opener [3H]P1075 in rat isolated aorta: relationship to functional effects of openers and blockers.

P1075 [N-cyano-N'-(1,1-dimethylpropyl)-N"-3-pyridylguanidine], an analogue of the K+ channel opener pinacidil, was shown to be a K+ channel opener in rat aorta, based on (i) its ability to stimulate 86Rb+ efflux, (ii) its ability to relax contractions in response to noradrenaline under normal conditions (5 mM KCl) but not under depolarized conditions (55 mM KCl), and (iii) the sensitivity of these effects to inhibition by the sulfonylurea glibenclamide. In these assays, P1075 was approximately 20 times more potent than cromakalim. Using a tritiated derivative, [3H] P1075, specific binding could not be detected in microsomal preparations from various tissues. However, in rat aortic strips specific binding of [3H]P1075 has been observed and was reduced by lowering the temperature or by decreasing intracellular ATP levels via metabolic inhibition. Specific [3H]P1075 binding was influenced neither by depolarization (55 mM KCl) nor by lowering the pH from 7.4 to 6.0. Binding was inhibited by representatives from all major families of K+ channel openers, with potencies that correlated well with the potencies obtained in 86Rb+ efflux and relaxation studies. However, stimulation of 86Rb+ efflux occurred at 40 times higher concentrations than did binding (and vasorelaxation). Of the various inhibitors of the K+ channel openers tested, only the sulfonylureas inhibited [3H] P1075 binding with the same rank order of potencies as that required for inhibition of P1075-induced 86Rb+ efflux, although at higher concentrations. The results show that binding of [3H] P1075 is independent of membrane potential but decreases concomitantly with the intracellular ATP level. The excellent correlation between the potencies of the openers and sulfonylurea blockers in binding assays and functional studies suggests that the 'drug receptor' labeled by [3H]P1075 in rat isolated aorta is of functional relevance. However, the fact that binding of the openers occurred at concentrations considerably lower than those required for K+ channel opening and that binding of the sulfonylureas was only reflected at concentrations higher than those needed to block the channel requires complex models to link binding and effect, possibly involving two agonist binding sites coupled by negative cooperativity.

Structure-activity studies of potassium channel opening in pinacidil-type cyanoguanidines, nitroethenediamines, thioureas, and ureas.

Potassium channel opening activity for pinacidil-type cyanoguanidines, nitroethenediamines, thioureas, and ureas, has been assessed through simultaneous measurement of spontaneous contractile activity and stimulation of 86Rb+ efflux from rat portal veins loaded with 86Rb+. The good correlation between these two effects suggests that the vasodilator activity of the compounds is directly attributable to an increased opening of potassium channels. The resulting quantitative in vitro data has been used to analyze the structure-activity relationships for potassium channel opening, allowing the biological activity to be rationalized in terms of a pharmacophore involving a hydrogen-bond-acceptor element, a hydrogen-bond-donor element, and a lipophilic binding group. A model for the binding of pinacidil-related compounds to their potassium channel receptor has been developed, and compounds designed to test this model have been synthesized and tested. Prototropic equilibria are implicated as playing a fundamental role in determining the hydrogen-bonding ability of the compounds, and conformational changes in the receptor are invoked to explain disparities in the chiral recognition of lipophilic groups in different compounds.

Identification of a specific binding site for K+ channel openers in rat aorta.

The K+ channel openers (KCOs) form a structurally heterogeneous group of compounds which relax smooth muscle by opening K+ channels in the plasmalemma. At present it is not known whether these drugs open the same K+ channel in smooth muscle and, if so, whether they bind to the same site of this channel. To address these questions we present the first binding study with KCOs in a smooth muscle preparation. In intact rat aortic strips, the novel tritiated KCO, 3H-P1075 (N-cyano-N'-(1,1-dimethyl[2,2,3,3(3)H]propyl)-N"-3-pyridylguanidine++ +), a potent pinacidil analogue, showed saturable specific binding of high affinity (KD = 6 +/- 1 nM, Bmax = 23 +/- 3 fmol/mg tissue wet weight). Specific binding of 3H-P1075 was inhibited stereospecifically by representatives from all major families of K+ channel openers, with a rank order of potency that correlated well with the potencies for vasorelaxation in rat aorta. The sulfonylurea glibenclamide, a relatively specific blocker of ATP-sensitive K+ channels and an inhibitor of the effects of KCOs, also inhibited 3H-P1075 binding and increased the rate of dissociation of 3H-P1075 from the tissue in a concentration-dependent manner. Lowering temperature (from 37 degrees C to 2 degrees C) and decreasing intracellular ATP levels by metabolic poisoning, diminished specific 3H-P1075 binding by reducing Bmax. However, depolarization (KCl = 55 mM) or lowering pH from 7.4 to 6.0 did not influence binding. The data demonstrate the existence of a specific binding site for the KCO 3H-P10756 in rat isolated aorta, which seems to be of functional relevance.(ABSTRACT TRUNCATED AT 250 WORDS)

5-[6-1 -(Cyclohexyl-1 H-tetrazol-5-YL)hexyl]-1,8-naphthyridin-2-(1H)-one, SC-44368, a Potent Anti-aggregatory Agent which Selectively Inhibits Platelet Cyclic AMP Phosphodiesterase.

SC-44368 (5-[6-(1-cyclohexyl-1H-tetrazol-5-y)hexyl]-1,8-naphthyridin-2(1H)-one) is a potent and selective competitive inhibitor of platelet cyclic AMP-dependent phosphodiesterase (cAMP-PDE) (Ki: 1.65 µM). For the phosphodiesterase isoenzyms from human platelets SC-44368 shows a 26-fold selectivity (IC50 ratio) for the inhibition of the cAMP-PDE over the cyclic GMP-dependent phosphodiesterase (cGMP-PDE). By comparison, 3-isobutyl-1-methyl-xanthine (IBMX) inhibited the cAMP-PDE and cGMP-PDE from human platelets with approximately equal efficacy. Broad inhibitory activity was evident against human platelet aggregatory responses in vitro. IC50 values of 18.1 ± 5.3 µM (25 nM platelet activating factor, PAF), 17.3 ± 3.0 µM (1.0 µg/ml collagen) and 24.2 ± 10.3 µM (1µM ADP) were obtained against maximum increases in platelet-rich plasma (PRP) light transmission achieved by each agonist. SC-44368 potentiated the prostacyclin-induced increase of intra-platelet cAMP levels but did not potentiate the sodium nitroprusside-induced increase of intraplatelet cGMP levels. In an ex vivo model of platelet aggregation SC-44368 (3 mg/kg, i.v.) produced a potent inhibition of collagen-induced platelet aggregation. SC-44368 produced only weak hypotensive activity in the rat. Thus, SC-44368 is a novel cAMP-PDE inhibitor which possesses potent, broad spectrum anti-aggregatory properties.

Thrombin receptor antagonists. Structure-activity relationships for the platelet thrombin receptor and effects on prostacyclin synthesis by human umbilical vein endothelial cells.

Structure-activity studies on a series of analogues of N-(3-methyl-S-(1-pyrrolidinyl carbonyl) butyl)-D-alanine ethyl ester hydrochloride (SC42619) have defined the features of this dipeptide analogue required for observation of thrombin receptor antagonist activity on the human platelet. The affinity for SC42619, and for its structural analogue SC43583 is enhanced by pretreatment of the platelets with chymotrypsin. Endothelial cell prostacyclin (PGI2) synthesis induced by thrombin and trypsin is selectively inhibited by SC42619 provided that prolonged exposure to this antagonist is avoided. However inhibition of PGI2 synthesis by SC42619 is not overcome by increasing the thrombin concentration. The data provide further support for identification of SC42619 and certain of its analogues as selective antagonists at the platelet thrombin receptor but suggest that these compounds may have more complex, and possibly non-selective effects on the endothelial cell.

Identification of small peptide analogues having agonist and antagonist activity at the platelet thrombin receptor.

Two tripeptide analogues (N-[3-methyl-1-S[[2-S [(methyl-amino)carbonyl]-1-pyrrolidinyl] carbonyl]butyl-D-analine) (SC40476) and N-[3-methyl-S-(1-pyrrolidinylcarbonyl)butyl]-D-alanine, ethyl ester, hydrochloride (SC42619], inhibit aggregation of, and secretion from, human platelets induced by thrombin but cause no significant inhibition of esterolysis or fibrin formation catalysed by this enzyme. Inhibition by SC40476 of the aggregatory response induced by thrombin is incomplete. Neither peptide analogue inhibits aggregation induced by ADP, collagen, vasopressin or 11,9-epoxymethanoprostaglandin H2 (U-46619). Enhancement of the response is observed when nonsaturating concentrations of these agonists are employed. SC42619 causes a parallel shift to the right in the concentration-response curve describing aggregation induced by thrombin. The Schild plot of these data has a slope of 1.05 and the pA2 is 2.9 +/- 0.1. Both SC40476 and SC42619 induced a small but significant decrease in the single platelet content of platelet suspensions. Neither peptide analogue increases platelet cytosolic [Ca2+] measured using quin 2 or Fura 2. Both analogues cause inhibition of the increase in cytosolic [Ca2+] induced by thrombin. Inhibition by SC42619 is competitive with respect to thrombin when the extracellular [Ca2+] is reduced to less than 0.1 microM but is non-competitive in the presence of 1 mM Ca2+. SC42619 also inhibits the increase in cytosolic [Ca2+]induced by ADP in the presence of 1 mM Ca2+ but not the smaller increase caused by this agonist when the medium contains less than 0.1 microM Ca2+. SC42619 inhibits Mn2+ influx induced by thrombin and ADP. SC40476 and SC42619 inhibit the enhanced incorporation of [32P] into phosphatidic acid observed on stimulation by thrombin of platelets pre-labelled with [32P]-phosphate. Addition of the peptide analogues alone fails to increase significantly the 32P content of phosphatidate, phosphatidylcholine, phosphatidylserine or phosphatidylethanolamine. SC40476 causes no detectable hydrolysis of glycoprotein V as detected by release of the proteolytic product (glycoprotein VFR). The results indicate that SC40476 and SC42619 interact selectively with the platelet thrombin receptor. Both peptide analogues act as effective antagonists for this receptor but also possess weak agonist activity which may also result from interaction with the thrombin receptor. The molecular basis for this latter activity has not been defined. SC42619 non-selectively inhibits Ca2+ influx induced by several agonists but this effect does not appear to contribute to the observed inhibition of the aggregatory and secretory responses.

Effect of SC 38249, a novel substituted imidazole, on platelet aggregation in vitro and in vivo.

SC 38249 [RS)-1-(2,3-bis-[(4-methoxyphenyl)methoxy]propyl)-1H-imidazole) caused dose-related inhibition of collagen-induced thromboxane A2 formation in human platelet rich plasma (IC50: 9.9 +/- 1.0 microM) accompanied by a dose-dependent increase in plasma PGE2. Broad inhibitory activity was evident against human platelet aggregatory and secretory responses in vitro. IC50 values of 11.9 +/- 1.9 microM (0.64 mM arachidonic acid), 18.3 +/- 3.8 microM (0.5 microgram ml-1 collagen) and 37.6 +/- 6.1 microM (25 nM Paf-acether) were obtained against maximum increase in PRP light transmission achieved by each agonist. Although less potent, SC 38249 retained significant inhibitory activity against PRP responses induced by a higher (3.0 micrograms ml-1) concentration of collagen (IC50: 272.5 +/- 24.6 microM), and against Paf-acether-induced responses in PRP pre-treated with 10 microM indomethacin (I.C.50: 192.0 +/- 16.1 microM). Experimental animal studies confirmed the in vitro anti-aggregatory efficacy of SC 38249, since significant inhibitory activity was observed against Paf-acether and ADP-induced responses in dog PRP ex vivo, anti-Forssman antibody-induced thrombocytopenia in anaesthetized guinea pigs, and collagen-induced intravascular aggregation in anaesthetized rabbits. Thus, SC 38249 is a novel thromboxane synthase inhibitor which possesses interesting anti-aggregatory properties which cannot wholly be attributed to prevention of platelet thromboxane A2 formation.

Thromboxane synthase inhibitors. Synthesis and pharmacological activity of (R)-, (S)-, and (+/-)-2,2-dimethyl-6-[2-(1H-imidazol-1-yl)-1-[[(4-methoxyphenyl)- methoxy]methyl]ethoxy]hexanoic acids.

A series of substituted omega-[2-(1H-imidazol-1-yl)ethoxy]alkanoic acid derivatives were synthesized and evaluated for their ability to inhibit thromboxane synthase both in vitro and in vivo. Compound 13 was identified as a potent and selective competitive inhibitor of human platelet thromboxane synthase having a Ki value of 9.6 X 10(-8) M. In collagen-treated human whole blood, 13 potentiated levels of 6-keto PGF1 alpha. Enantiospecific syntheses afforded the R and S enantiomers of 13, of which the S enantiomer 13b was the more potent. Compounds 13 and 13b were potent in vivo inhibitors of thromboxane synthase with good oral activity and duration of action.

Dibenzoquinazoline diones as antihypertensive cyclic guanosine monophosphate phosphodiesterase inhibitors.

The discovery and structure-activity of a new class of renal artery phosphodiesterase inhibitors is reported, some of which are highly selective for the guanosine cyclic 3',5'-monophosphate phosphodiesterase. One of these compounds, 5,6-dihydro-8,9,11,12-tetramethoxy-1,3-dioxo-1H-benz[f]- isoquino [8,1,2- hij]quinazoline-2(3H)-carboxylic acid, ethyl ester (9), is amongst the most potent and selective compounds of this class yet identified. Furthermore, this compound demonstrates an antihypertensive effect in vivo which is presumably mediated through vascular smooth muscle relaxation.

Structure-activity relationships in an imidazole-based series of thromboxane synthase inhibitors.

Analogues of 4-[[2-(1H-imidazol-1-yl)-1-[[(4-methoxyphenyl)methoxy]methyl] ethoxy]methyl]benzoic acid (5m) were prepared and evaluated as thromboxane synthase inhibitors. A series of esters of 5m showed a parabolic relationship between lipophilicity and inhibition of TxB2 generation in intact platelets, with activities up to 50 times greater than that of dazoxiben. However, on administration to rabbits the ethyl ester 5d had a short duration of action, due to rapid metabolism and excretion via deesterification and beta-glucuronidation. Attempts at replacing the carboxylate group with other potential pharmacophores were unsuccessful.