Characterization of oligosorbents and application to the purification of ochratoxin A from wheat extracts.

The aim of this work was to optimize the preparation of an anti-ochratoxin A (OTA) oligosorbent (OS), a solid-phase extraction sorbent based on OTA aptamers covalently immobilized on sepharose. Different syntheses were carried out by modifying the side of the oligonucleotide chain bound to the sepharose, the length of the spacer arm between the aptamer and the sepharose and the amount of the aptamers introduced during the covalent grafting. Indeed, the capacity of OSs prepared using 3'- or 5'-amino-modified sequences with a C6 or a C12 was studied. In the best conditions, the concentration of aptamers sequence used during their grafting was increased and a capacity close to 40 nmol g(-1) of OS was reached. The potential of the resulting OSs was also studied in pure media. For this, their selectivity was checked by comparing them to a control sorbent prepared without immobilizing aptamers. Extraction recoveries close to 100% were obtained on all OSs, while no retention was observed on the control sorbent. OS does not demonstrate any cross-reactivity towards OTA metabolites, i.e., ochratoxin B and ochratoxin hydroquinone. The oligosorbent was finally applied to the clean-up of OTA from wheat sample extracts. Extraction recoveries were not affected by matrix interferences and the resulting chromatogram clearly highlights the selectivity of the sorbent that allows the removal of matrix components thus improving the reliability of the quantitation of OTA in real samples.

Solid-phase extraction using molecularly imprinted polymers for selective extraction of a mycotoxin in cereals.

The aim of this work was to develop a method for the clean-up of a mycotoxin, i.e. Ochratoxin A (OTA), from cereal extracts employing a new molecularly imprinted polymer (MIP) as selective sorbent for solid-phase extraction (SPE) and to compare with an immunoaffinity column. A first series of experiments was carried out in pure solvents to estimate the potential of the imprinted sorbent in terms of selectivity studying the retention of OTA on the MIP and on a non-imprinted polymer using conventional crushed monolith. The selectivity of the MIP was also checked by its application to wheat extracts. Then, after this feasibility study, two different formats of MIP: crushed monolith and micro-beads were evaluated and compared. Therefore an optimization procedure was applied to the selective extraction from wheat using the MIP beads. The whole procedure was validated by applying it to wheat extract spiked by OTA at different concentration levels and then to a certified contaminated wheat sample. Recoveries close to 100% were obtained. The high selectivity brought by the MIP was compared to the selectivity by an immunoaffinity cartridge for the clean-up of the same wheat sample. The study of capacity of both showed a significant higher capacity of the MIP.

Selective sample pretreatment by molecularly imprinted polymer for the determination of LSD in biological fluids.

For the first time, a molecularly imprinted polymer (MIP) was synthesized by a noncovalent imprinting approach for the selective extraction of an illicit drug, LSD, from hair and urine samples. For the synthesis of MIP, an analog of LSD, was taken as a dummy template, methacrylic acid as a functional monomer, and ACN as a porogen solvent. The MIP was used for offline extraction before HPLC-MS analysis. By studying the interactions taking place between the LSD and the MIP, a selective procedure was established in organic media and applied to hair samples. By this way, 0.1 ng/mg of LSD was successfully detected in hair with 82% of extraction recovery. A low retention was also obtained on the control polymer (only 9%). This procedure was then modified to obtain a selective extraction in aqueous media for the determination of LSD in urine samples. The comparison with a conventional C18 clearly demonstrated the selectivity brought by the MIP to the determination of LSD in urine. LSD was easily detected in urine at only 0.5 ng/mL with 83% of extraction recovery on the MIP and 11% on the NIP. An LOQ of 0.2 pg/mL was estimated in urine samples.