Development of an affinity sensor for the detection of Aflatoxin M1 in milk

Much research has been done on aflatoxins since their discovery in the 1960’s where it was concluded that aflatoxins have carcinogenic, mutagenic, teratogenic and immunosuppressive properties. Aflatoxin M1 exists in milk and since milk is a major component of the diet of infants, the maximum permissible limit set by the EU is 50 parts per trillion (ng L -1 ). Current methods of analysis for aflatoxin M1 is primarily based around techniques such as HPLC and TLC which require extensively trained operators and equipped laboratories. Using antibodies as receptors in an enzyme linked immunosorbent assay (ELISA), the analysis costs can be reduced and simplified, however, an equipped laboratory is still required. Hence there is a need for a low cost, rapid, portable instrument which is easy to use at the point of source for the detection of aflatoxin M1. This thesis describes the development of affinity sensors to meet these requirements. Firstly the design and optimisation of an ELISA method was carried out, utilising a commercially sourced monoclonal antibody. Once the antibodies suitability for sensing aflatoxin M1 was determined the antibody was successfully employed as the receptor for a screen printed HRP/TMB based immunosensor. Upon the analysis of milk it was observed that milk caused extensive interference and through a series of chemical extractions the interference was attributed to whey proteins in the milk with suspicion towards a- lactalbumin. A simple pre-treatment technique of adding calcium chloride was performed and the interference from the whey proteins was removed. The resulting immunosensor achieved a sensitivity of 39 ng L -1 (Figure 3.26), however, poor reproducibility was observed due to the screen printed electrode production (%CV = 21% variance for screen printed electrode production). Gold cell on a chip microelectrode arrays were used to replace the screen printed electrodes and the successful covalent attachment of the antibody to the microelectrode array through PDITC cross linking compound was monitored using atomic force microscopy and scanning electron microscopy. It was shown that the majority of the antibodies during immobilisation orientate in a ‘side on’ orientation and therefore a cheap capture polyclonal antibody was first immobilised before the addition of the sensing anti-aflatoxin M1 monoclonal antibody. Using the microelectrode array an improvement of the sensitivity as well as a reduction of the milk interference was shown. Sensitivity was improved to 8 ng L -1 in milk (Figure 4.23). Further work was performed to substitute the fragile antibody used in the sensing layer for a robust synthetic peptide receptor. Initially a virtual library of synthetic peptides was created using de novo design techniques in silico. Further computational techniques were performed to determine the best peptide from the library. This peptide had a sequence of PVGPRP. From literature a peptide (LLAR) was reported with affinity for aflatoxin B1. This sequence along with the de novo design peptide was synthesised and tested using a host of techniques and immobilisation chemistries such as optical waveguide lightmode spectroscopy (OWLS), BIAcore and enzymatic techniques using EDC/NHS, glutaraldehyde and BS 3 cross linking methods. The affinity of both peptides to aflatoxin M1 was demonstrated however further work is required to quantify the affinity and to incorporate the peptides into the microelectrode array.