Miniaturization of Food Analytical Techniques — with relevance to mycotoxins

Mycotoxins belong to the group of naturally occurring food contaminants that are produced by various species of several fungi genera. Occurring in a wide range of food products, mycotoxins demonstrate high toxicity, with some even being potent carcinogens. This means they pose severe health threat to humans and livestock even in low concentrations.

Contamination with mycotoxin may happen at any stage of the food production process, even when good agricultural, storage, and processing practices are conducted. Therefore, it is of vital importance that the contamination is identified at an early stage to avoid its further spreading around and up the food chain, with the aim of improving food safety and significantly reducing food loss.

To achieve this, efficient analytical tools are essential to detect and give warnings of the contamination as soon as possible. The majority of the monitoring for mycotoxins is performed in routine and reference laboratories following standard and official methods. However, these methods not only rely on complex and expensive instruments and specialized laboratories requiring well-trained personnel, but they are also in general rather slow.

This prevents the traditional analyses from being performed in field conditions or for real time monitoring. These shortcomings have inspired studies aimed at developing analytical techniques which are robust, easier to operate and portable, enabling mycotoxin monitoring in field conditions for direct crop and food testing in an early stage.

The hypothesis of the PhD project is that a miniaturization of mycotoxin analytical techniques can be achieved by adopting novel extraction and sensing strategies, which will be more time and cost-efficient, easier to operate, provide adequate sensitivity, and also hold the potential to be developed into a portable and automated device.

The overall aim has been to produce a proof of concept as a foundation for the development of analytical platforms which integrate real food sample extraction, pre-treatment, and detection steps. Considering the broad scope of techniques proposed for miniaturization of mycotoxin analysis, a review summarizing the current state of the art mycotoxin analytical techniques was drafted as a part of the PhD project, discussing the opportunities and challenges presented for method and technology development.

In this project, supported liquid membrane extraction (SLME) was chosen as the sample pre-treatment and separation method, while fluorescence spectroscopy and surface enhanced Raman spectroscopy (SERS) were studied as detection methods. Ochratoxin A (OTA) and aflatoxin B1 (AFB1) were chosen as targets for analysis due to their high occurrence and the health risks they present, including carcinogenicity, nephrotoxicity and immunotoxicity.

A SLME platform was developed for OTA extraction and pre-concentration from wine samples. It was established on 96-well micro filter plates with polymeric membranes at the bottom. Using optimized experimental parameters, an OTA recovery of about 80 % was achieved within 1 hour of extraction with a standard deviation of 4 %.

The validated linear quantification range was from 0.63 to 10 μg L–1 with a limit of detection of 0.20 μg L–1 using traditional LC-MS analysis, which is well below the maximum tolerance level set by European Commission. The SLME platform has the advantages of low solvent consumption, excellent time-efficiency and high throughput, which was also tested on beer and coffee samples, showing the potential to be applied to other types of food samples.

An independent fluorescence detection method was developed to couple with the SLME, which was carried out with a laboratory set-up consisting of a Ti-sapphire laser, a harmonic generating unit, a sample holder and a spectrum analyzer. The fluorescence spectroscopy setup allows direct and immediate measurement of extracted samples on the SLME acceptor plate.

This does not require any additional manual liquid handling steps after the extraction, takes microseconds for each measurement and consumes zero chemicals. A limit of detection of 51 μg L–1 was achieved, which is promising for application after further optimization. With the advancement of laser diodes, the technique has great potential to be designed into a portable measurement device using SLME coupled with fluorescence detection.

Furthermore, the potential to utilize SERS for OTA and AFB1 determination was investigated. Fingerprint spectra of OTA and AFB1 were obtained on silver and gold SERS substrate respectively. Successful SERS quantification of OTA extracted from red wine samples by SLME was achieved, with good linear correlation in the range from 50 to 1000 μg L–1 between Raman intensity in the representative peak and OTA concentration.

Besides the abovementioned techniques, other techniques were explored in the project, including the fabrication of polymeric extraction discs and the functionalization of SERS substrates using aptamers. None of these techniques produced satisfactory outcomes and were not pursued further.