This study comprises the development of an enzyme-linked immunosorbent assay (ELISA) for the quantification of caffeine in complex aqueous matrices without any sample clean-up procedure. Salinity and dissolved organic matter were selected as potential interfering agents. The addition of a sample buffer containing bovine serum albumin (BSA) prior to the sample was found to decrease the influence of those interfering agents. The working range of the developed method was 0.1–100 µg L- 1. Quantification of caffeine was possible in 43 out of 51 real aqueous samples, at values between < LOD and 15 µg L- 1. Results correlate well with those obtained by LCMS/MS. To the best of our knowledge this is the first study dealing with the quantification of caffeine in Portugal's surface waters.
Estrone (E1), a metabolite of the estrogenic hormones 17β-estradiol (β-E2) and 17α-estradiol (α-E2), is itself a potent estrogen which can have a significant impact on the hormonal balance. Due to ist high potential for adverse effects on human health and aquatic life even at pg/L to ng/L levels, its appearance in water should be monitored. E1 has also been considered a marker substance for the presence of other estrogens. This study presents a newly developed direct competitive enzymelinked immunosorbent assay (ELISA) for quantification of E1 in environmental water samples using new monoclonal antibodies.
The quantification range of the ELISA is 0.15 μg/L to 8.7 μg/L E1, and the limit of detection is around 60 ng/L for not pre-concentrated water samples. A pre-concentration step after careful selection of suitable phases for SPE was developed, too. The influence of organic solvents and natural organic matter on the ELISA was assessed. The high selectivity of the monoclonal antibody was demonstrated by determining the cross-reactivity against 20 structurally related compounds.
For the assessment of matrix effects, a concept (“LC–ELISA”) is thoroughly exploited, i.e., separating complex samples by HPLC into 0.3 min fractions and determination of the apparent E1 concentration. Furthermore, fractions with interferences for nontarget/suspected-target analysis can be assigned. A dilution approach was applied to distinguish between specific interferences (cross-reactants) and non-specific interferences (matrix effects). In the determination of 18 environmental samples, a good agreement of the E1 concentration in the respective fractions was obtained with mean recoveries of 103 % to 132 % comparing ELISA to LC–MS/MS.