PURPOSE: The selection of cosalane (NSC 658586) by the National Cancer Institute for further development as a potential drug candidate for the treatment of AIDS led to the exploration of the solubility behavior of this extremely hydrophobic drug, which has an intrinsic solubility (S0 approaching 1 ng/ml. This study describes attempts to reliably measure the intrinsic solubility of cosalane and examine its pH-solubility behavior. METHODS: S0 was estimated by 5 different strategies: (a) direct determination in an aqueous suspension: (b) facilitated dissolution; (c) estimation from the octanol/water partition coefficient and octanol solubility (d) application of an empirical equation based on melting point and partition coefficient; and (e) estimation from the hydrocarbon solubility and functional group contributions for transfer from hydrocarbon to water. RESULTS: S0 estimates using these five methods varied over a 5 x 107-fold range Method (a) yielded the highest values, two-orders of magnitude greater than those obtained by method (b) (facilitated dissolution. 1.4 +/- 0.5 ng/ml). Method (c) gave a value 20-fold higher while that from method (d) was in fair agreement with that from facilitated dissolution. Method (e) yielded a value several orders-of-magnitude lower than other methods. A molecular dynamics simulation suggests that folded conformations not accounted for by group contributions may reduce cosalane's effective hydrophobicity. Ionic equilibria calculations for this weak diprotic acid suggested a 100-fold increase in solubility per pH unit increase. The pH-solubility profile of cosalane at 25 degrees C agreed closely with theory. CONCLUSIONS: These studies highlight the difficulty in determining solubility of very poorly soluble compounds and the possible advantage of the facilitated dissolution method. The diprotic nature of cosalane enabled a solubility enhancement of > 107-fold by simple pH adjustment.
PURPOSE: The selection of cosalane (NSC 658586) by the National Cancer Institute for further development as a potential drug candidate for the treatment of AIDS led to the exploration of the solubility behavior of this extremely hydrophobic drug, which has an intrinsic solubility (S0 approaching 1 ng/ml. This study describes attempts to reliably measure the intrinsic solubility of cosalane and examine its pH-solubility behavior. METHODS: S0 was estimated by 5 different strategies: (a) direct determination in an aqueous suspension: (b) facilitated dissolution; (c) estimation from the octanol/water partition coefficient and octanol solubility (d) application of an empirical equation based on melting point and partition coefficient; and (e) estimation from the hydrocarbon solubility and functional group contributions for transfer from hydrocarbon to water. RESULTS: S0 estimates using these five methods varied over a 5 x 107-fold range Method (a) yielded the highest values, two-orders of magnitude greater than those obtained by method (b) (facilitated dissolution. 1.4 +/- 0.5 ng/ml). Method (c) gave a value 20-fold higher while that from method (d) was in fair agreement with that from facilitated dissolution. Method (e) yielded a value several orders-of-magnitude lower than other methods. A molecular dynamics simulation suggests that folded conformations not accounted for by group contributions may reduce cosalane's effective hydrophobicity. Ionic equilibria calculations for this weak diprotic acid suggested a 100-fold increase in solubility per pH unit increase. The pH-solubility profile of cosalane at 25 degrees C agreed closely with theory. CONCLUSIONS: These studies highlight the difficulty in determining solubility of very poorly soluble compounds and the possible advantage of the facilitated dissolution method. The diprotic nature of cosalane enabled a solubility enhancement of > 107-fold by simple pH adjustment.
Authors: M Cushman; W M Golebiewski; J B McMahon; R W Buckheit; D J Clanton; O Weislow; R D Haugwitz; J P Bader; L Graham; W G Rice Journal: J Med Chem Date: 1994-09-16 Impact factor: 7.446
Authors: M Cushman; W M Golebiewski; Y Pommier; A Mazumder; D Reymen; E De Clercq; L Graham; W G Rice Journal: J Med Chem Date: 1995-02-03 Impact factor: 7.446