Development Of Novel Matrices For Biomolecule Immobilisation On Sensor Surfaces

Abstract

The development of a novel protocol for the covalent immobilisation of biomolecules containing primary amines using either polythiol compounds or novel, inexpensive and simple polymers is presented in this thesis. When developing biosensors, the method used for the immobilisation of the sensing elements is very important. The immobilisation needs to be fast, cheap and most importantly should not affect the biorecognition activity of the immobilised receptor. The chemistry used for the immobilisation is based on the well known reaction between primary amines and thioacetal groups, formed upon reaction of o-phthaldialdehyde (OPA) and thiol compounds. Initially the possibility to use this chemistry to immobilise receptors and develop biosensors was proved using commercially available polythiol compounds. Such compounds can be irreversibly adsorbed, creating self-assembling monolayers (SAMs), on noble metal transducer surfaces. These SAMs were immobilised on Biacore surface plasmon resonance (SPR) gold chips and then used to study kinetic of biomolecules interactions and to detect cells. A general protocol suitable for the immobilisation of enzymes and antibodies such as anti-prostate specific antigen (anti-PSA) and anti-Salmonella typhimurium antibody was optimised. Kinetic data were obtained for PSA binding to anti-PSA antibody and they were compared to the results obtained using commercially available Biacore chips, CM1. For Salmonella typhimurium cells, a detection limit of 5 × 106 cells ml-1 with minimal non-specific binding of other biomolecules was obtained. An interesting capability shown by these SAMs, in contrast with commercially available chips, was the opportunity to immobilise any proteins, even those with very low or high isoelectric points, pI. In addition protein immobilisation was achieved with a simple step, without requirement of any activation. These findings make this immobilisation technique a very promising alternative to peptide bond formation for amine coupling. Even though, the developed SAMs showed to be useful for certain type of applications (kinetic study and detection of very large analyte), it was clear that due to a combination of factors (e.g. limited and steric hindrance), they were not suitable for the development of biosensors good enough for practical applications. Therefore to overcome the drawbacks shown by polythiol SAMs, a novel 3-D polymer was developed. The main advantage of this polymer is the tridimensional (3D) network, which, after

immobilisation, ensures the availability of a high percentage of receptor binding sites. As the polythiol SAMs, also the 3-D polymer contains thioacetal groups, which do not need any activation to react with primary amines in proteins. The novel 3-D polymer also contains thiol derivative groups (disulphide groups or thioethers) that promote self-assembling on metal surfaces. As before, the polymer was immobilised on SPR gold chips and the resulting layer was characterised using contact angle meter, atomic force microscopy (AFM) and ellipsometry. Contact angle demonstrated that the immobilisation of polymer on sensor surface produced a relatively hydrophobic surface. The thickness of polymer layer was determined by applying ellipsometry, whereas AFM showed the change of surface roughness after polymer attachment. A general protocol suitable for the immobilisation of BSA, enzymes and antibodies such as polyclonal anti-microcystin-LR and monoclonal anti-prostate specific antigen (anti-PSA) antibody was then optimised. The affinity characteristics of developed immunosensors were investigated in reaction with microcystin-LR, and PSA. The calculated detection limit for analytes depended on the properties of the antibodies. The detection limit for microcystin-LR was 10 ng ml and for PSA 0.05 ng ml. The 3-D polymer chips were stored for up to 2 months without any noticeable deterioration in their ability to react with proteins. The performance of 3-D polymer chips were also compared with commercially available Biacore chips, as CM5. The main advantages were found to be the low cost, the possibility to immobilise biomolecules at physiological pH (pH 7.4), the lack of any activation step for biomolecules immobilisation and the opportunity to immobilise proteins with very different pI (also very low pI). Despite the successful detection of PSA achieved in buffer (detection limit 0.05 ng ml-1) using 3-D polymer chips, the detection of proteins in serum resulted to be very challenging due to the complex nature of the matrix, which contains a high content of many different compounds. Different techniques were applied in order to reduce the non specific adsorption of serum on 3-D polymer sensors with antibodies immobilised on the surface. Satisfactory results were finally obtained by including the surfactant P20 into the measuring system. The detection of PSA in serum using 3-D polymer sensors, however, became possible only by switching from a direct detection to a ‘sandwich detection’. In this sandwich format, after injecting samples of PSA (prepared both in buffer or 20%

serum) onto a specific antibody (capture-Ab, C-Ab) immobilised on the 3-D polymer surface, the analytical signal is recorded by injecting a second specific Ab (detection-Ab, prepared in PBS), which recognises a different epitope of the antigen. With this format, the analytical signal is recorded in absence of any complex matrix, avoiding interference from non specific adsorption. The detection limit for PSA, obtained using the sandwich immunosensor (developed on 3-D polymer chips) was 0.1 ng ml-1 in buffer and 5 ng ml-1 in 20% serum, which is very close to the sensitivity necessary for detection of the prostate biomarker in real samples. Therefore this study has demonstrated the opportunity to apply the novel 3-D polymer for development of biosensors suitable for applications in real samples.