Solid-Phase Extraction and Reverse-Phase HPLC: Application to Study the Urinary Excretion Pattern of Benzophenone-3 and its Metabolite 2,4-Dihydroxybenzophenone in Human Urine

BACKGROUND Benzophenone-3 (BZ-3) is a common ultraviolet (UV) absorbing compound in sunscreens. It is the most bioavailable species of all UV-absorbing compounds after topical application and can be found in plasma and urine. OBJECTIVES The aim of this study was to develop a reverse-phase high performance liquid chromatography (HPLC) method for determining the amounts BZ-3 and its metabolite 2,4-dihydroxybenzophenone (DHB) in human urine. The method had to be suitable for handling a large number of samples. It also had to be rapid and simple, but still sensitive, accurate and reproducible. The assay was applied to study the urinary excretion pattern after repeated whole-body applications of a commercial sunscreen, containing 4% BZ-3, to 25 healthy volunteers. METHODS Each sample was analyzed with regard to both conjugated/non-conjugated BZ-3 and conjugated/non-conjugated DHB, since both BZ-3 and DHB are extensively conjugated in the body. Solid-phase extraction (SPE) with C8 columns was followed by reverse-phase HPLC. For separation a Genesis C18 column was used with an acethonitrile-water mobile phase and the UV-detector was set at 287 nm. RESULTS The assay was linear r(2) > 0.99, with detection limits for BZ-3 and DHB of 0.01 micromol L(-1) and 0.16 micromol L(-1) respectively. Relative standard deviation (RSD) was less than 10% for BZ-3 and less than 13% for DHB. The excretion pattern varied among the human volunteers; we discerned different patterns among the individuals. CONCLUSIONS The reverse-phase HPLC assay and extraction procedures developed are suitable for use when a large number of samples need to be analyzed and the method fulfilled our objectives. The differences in excretion pattern may be due to differences in enzyme activity but further studies, especially about genetic polymorphism, need to be performed to verify this finding.