Bacteria exposed excessively to antibiotics may develop resistance. To ensure our future capacity in fighting infectious diseases it is therefore of vital importance to decrease the spread of antibiotics globally. Antibiotics are commonly used in veterinary medicine. Simple and cheap methods for their monitoring in milk and meat are therefore needed and required by regulatory authorities. The objective of this project is to develop a portable micro fabricated plug-in cartridge for flow analysis of b-lactam antibiotic residues in food. Robust synthetic polymers imprinted with b-lactams will serve as the recognition elements using fluorescent detection. The device will be compared to existing methods for the analysis of b-lactam antibiotics and will be critically evaluated in field tests.
Libraries of molecularly imprinted polymers (MIPs) targeted for penicillin G were synthesized, screened, and optimized. MIP candidates were studied in a radioligand competitive binding assay. Some cross-reactivity was seen for 4-ppb antibiotics (ampicillin and amoxicillin), 6-aminopenicillanic acid (6-APA), and penicillin V. Low cross-reactivity was seen for 30-ppb antibiotics (oxacillin, nafcillin, and cloxacillin). No cross-reactivity was observed for other antibiotics (erythromycin, dapsone, tetracycline, chloramphenicol, and cephapirin). Molecular recognition studies in aqueous acetonitrile indicated that the binding has to take place at low water content in order to maintain binding capacity and selectivity. 1% buffer pH 7 in acetonitrile gave selective recognition of both penicillin G and a fluorescent b-lactam analog (PAAP). Novel pyrene- and dansyl-labelled b-lactam analogs and acrylate/acrylamide monomers were synthesized and characterized. Two analogues of penicillin G bearing bromine and dimethylamino groups, respectively, were also synthesized. Most of the new molecules maintained the same absorption and fluorescent features as the parent fluorophore. The labelled analogs were first evaluated by a radioligand competitive binding assay. Dansyl derivatives (DAP, DAM) were poor competitors to penicillin G. MIPs 1 and 6 recognized selectively pyreneacetic aminopenicillanic acid (PAAP) and to a lower extent pyrenebutyric aminopenicillanic acid (PBAP).
From these experiments, PAAP and MIP 1, the pair providing the highest sensitivity for penicillin G analysis, was selected and applied to the development of the assay. The response range of the assay has not the required sensitivity for milk measurements, unless the sample is pre-concentrated. The competitive fluorescent assays showed cross-reactivity for some antibiotics derived from 6-APA, particularly those with an MLR of 4 ppb. Other antibiotics (chloramphenicol, tetracycline and cephapirin) did not compete with PAAP. Several measuring systems were optimized and different assay formats were evaluated for implementation in a cartridge-based assay. The stability of PAAP was demonstrated in concentrated acetonitrile solutions at -20 C for over 6 months and in solid form for over two years. Kinetic studies of PAAP binding and competition indicated a minimum required assay time of 30 min. Two HPLC methods were optimised for b-lactam analysis. Several MIPs prepared with photoluminescent monomers were also synthesized. Preliminary emission studies of the polymers were carried out using different microfabricated reactors. As both the developments of the assay and of the cartridge were concurrent, a number of initial assumptions on the required cartridge functionality were made. A modular approach was followed to facilitate the cartridge development. Stand-alone devices were fabricated and characterized for each of identified key operations, that is sample, mix, precipitate, filtrate, pre-concentrate and detect. Silicon based microfabrication techniques were mostly used. However, with the perspective of a later industrialization, different plastics materials and fabrication techniques were also evaluated. Each device could be interconnected to emulate a complete cartridge system. This allowed a rapid experiment construction and device characterization. Best elements were integrated in the final design and cartridges were fabricated. The cartridge consisted of a sampling unit, a microreactor and an optical cell. Preliminary characterisation showed excellent MIP packing properties, successful sampling and detection of the synthesised reporter.
The entire functionality of the cartridge and its analytical performances could however not be demonstrated within the project time frame. Although thorough characterisation still needs to be carried out and eventual minor corrections brought to the design, there are strong indications that the proposed cartridge format is suitable and performs as expected. Optical and fluidic interfaces for the operation of the cartridge were developed and optimised. With the view of developing a portable and practical instrument, each individual instrument elements were mostly considered against performances, power consumption and robustness. A detection limit of 250 nM of the synthesised probe was achieved. This performance does not meet the project objective for the detection of the low MRL residues, but is the best achievable for the selected configuration. The spectroscopic characteristics of the selected label significantly challenged the development of the optical detection system. Optical and incidentally chemical constraints set on the materials narrowed the selection of these to a very few candidates. Working at higher wavelengths would not only lever the requirements on background fluorescence and transmittance, but also increase the number of available light sources offering high optical output power.
Funding SchemeCSC - Cost-sharing contracts
LE11 3NH Loughborough