(Invited) Molecularly Imprinted Polymer “Antibodies” to Recognize Biological Species and Bioactive Molecules

Abstract

Diagnostics, healthcare, and lifestyle are key drivers of sensors and assays that allow for measuring at home or at the point of care. Lateral-flow assays relying on antibodies are probably the most widely known examples for this. Despite their undisputed qualities, there is a strong drive to replace antibodies by artificial mimics that come at lower cost and are more stable. Molecularly imprinted polymers (MIPs) tick those boxes: they result from template-directed polymerization and, thus, from straightforward synthesis, and have appreciable binding properties [1]. Traditional imprinting approaches, however, suffer from limited reproducibility [2]. Solid-phase synthesis of MIP nanoparticles followed by enriching high-affinity fractions through sequential elution from the synthesis column can overcome such limitations [3].The approach has turned out very versatile for addressing analyte species almost any size. Originally, it has been focusing on proteins, probably because it follows the logic of interactions between antibodies and antigens. There are two different strategies for achieving that goal: using the entire protein molecule as the template for imprinting, or only a part of it. The former strategy, for instance, allows for establishing both direct and competitive sensing assays for human serum albumin (HSA) and insulin, respectively. Both aim at mass-sensitive sensing with quartz crystal microbalances (QCMs) as the transducers. Hence it is interesting to replace and/or complement the organic polymers with silica or titania particles to increase sensitivity: the number of available binding sites on MIP surface is limited by film and sensor dimensions. Using heavier particles hence increases the frequency shift per binding event. In the same manner, it turned out possible to access cardiac biomarkers, such as Gallectin3 (Gal3).Considering that protein molecules are large and structurally flexible makes it interesting to use only epitopes on their surfaces for imprinting. This leads to (surprisingly) high selectivity: In the case of a Dengue virus structural protein, imprinting with an epitope comprising 11 amino acids indeed leads to binding between the MIP nanoparticles and the respective protein. Changing only one amino acid reduces the sensor responses by a factor of more than two orders of magnitude thus revealing excellent selectivity bordering to specificity. Following that logic further, it is even possible to synthesize MIP nanobodies for small molecules, such as salbutamol. In such cases one usually applies a linker molecule to ensure synthesis. The resulting particles again highly selectively bind their target compound and are also useful for competitive assay formats.Going the other direction in terms of size, it is possible to use entire bacteria, such as S. epidermidis, as the template [4]. In contrast to the previous MIP nanobody approaches, this yields a variety of different nanoparticles that bind to different positions on the bacteria surface. The nanoparticle ensemble selectively interacts with its target species, even though the exact binding sites are unknown. Thus, when using them to incubate a different bacterium, such as E. coli, this reduces the fluorescence intensity by more than 80%. In other words: this means one can synthesize artificial antibodies for species without knowing the exact chemical details.[1] Bui BTS, Haupt K, Anal Bioanal Chem 398 (2010) 2481-92, doi: 10.1007/s00216-010-4158-x[2] Unger C, Lieberzeit PA, Reactive Funct Polym 161 (2021) 104855, doi: 10.1016/j.reactfunctpolym.2021.104855[3] a) Canfarotta F, Poma A, Guerreiro A, Piletsky S (2016) Solid-phase synthesis of molecularly imprinted nanoparticles. Nat Protoc 11:443–455. https://doi.org/10.1038/nprot.2016.030; b) Xu J, et al., Biomacromolecules 17 (2016) 345-353, doi: 1021/acs.biomac.5b01454 [4] Hlaoperm C et al. Sensors 23 (2023) 3526, doi: 10.3390/s23073526