The three different formulations tested in this study were (a) formulation “Gelatin”: 260 mg Sunphenon 90DCF-T in gelatin capsules, (b) formulation “HPMCgell”: 260 mg Sunphenon-90DCF-T in HPMC capsules with Gemcitabine molecular weight gellan gum as the gelling agent and (c) formulation

“HPMC”: 260 mg Sunphenon-90DCF-T in HPMC capsules without gelling agent. All capsules were size 0 and transparent. The formulations were prepared manually using a capsule filling machine (Capsunorm 2000, Tecnyfarma, Barcelona, Spain). The compendial media – 0.1 mol/l hydrochloric acid (pH 1.2), acetate buffer (pH 4.5) and phosphate buffer (pH 6.8) – were prepared according to USP 32. Fasted state simulated intestinal fluid (FaSSIF) and fed state simulated intestinal fluid (FeSSIF) were prepared from simulated intestinal fluid (SIF) powder (Biorelevant.com, Croydon, Surrey, UK) [11]. Capsule disintegration was tested according to USP 32 chapter <2040> with a disintegration tester ZT120 and tube/rack assembly Apparatus B (Erweka GmbH, Heusenstamm, Germany). Chapter <2040> also discusses the acceptance criteria for dietary supplements. The test for hard shell capsules was applied and as the USP advises to omit the

use of discs for botanical dosage forms; capsules were placed in a metal spiral capsule sinker (ProSense BV Dissolution Accessories, Oosterhout, LY294002 cell line the Netherlands) to prevent floating which is a slight modification of the description in <2040>. This modification avoids the mechanical impact discs during each stroke

and at the same time keeps the capsules submerged to ensure ample fluid contact. The recorded capsule disintegration time is the time at which the capsule was visually observed to be completely disintegrated, even if some pieces of the capsule shell remained on the mesh of the test basket. Disintegration of the formulations was assessed in two immersion fluids: the USP recommended 0.05 mol/l acetate buffer as well as in demineralised water, both preheated to 37 °C. All experiments were performed as n = 6 and the mean and standard deviation were calculated. A calibrated dissolution tester “VK 1700” (Varian Inc., Cary NC, USA) was used for GPX6 all dissolution studies. The formulations were tested using the paddle method plus sinker (USP Apparatus 2), employing 900 ml of dissolution medium equilibrated to 37 ± 0.5 °C and a rotational speed of 75 rpm. Spiral capsule sinkers (ProSense BV Dissolution Accessories, Oosterhout, the Netherlands) were used to prevent capsules from floating. Samples were taken after 5, 10, 20, 30, 45, 60 and 120 min by withdrawal of 2 ml at each sampling point and the volume withdrawn was replaced with fresh pre-warmed medium. Each sample was immediately filtered through a 0.2 μm PVDF filter (Type Acrodisc LC, Pall Life Sciences, Uithoorn, The Netherlands) and directly analyzed by HPLC and/or appropriately diluted with buffer media prior to UV–vis spectrophotometer analysis.