Molecularly imprinted nanoparticles (MIN) are synthetic smart nanomaterials which pretend to imitate the binding behaviour of their natural counterparts, like antibodies or enzymes [1]. Even if natural receptors have excellent recognition capabilities, they present inherent limitations due to their biologic nature, showing poor stability under non-physiological conditions. In fact, pH and temperature variations may induce conformational changes on them, thus losing their functionality. MIN, as artificial antibodies, are intended to overcome common drawbacks associated with natural ones, showing improved stability to high temperatures, extreme pH and pressure, which allows them for being stored for long time periods even at room temperature [2]. Furthermore, MIN present higher batch-to-batch reproducibility, and relatively faster and more cost-effective production [3]. In addition, production of imprinted materials does not involve experiments with animals, which may contribute to a significant reduction of their use. Given their benefits and potential applicability, it is expected that, in the future, molecularly imprinted polymers (MIP) may replace many biomolecules, not only antibodies, but also enzymes or DNA [4].

Over the last decades, the fabrication of nano-sized molecularly imprinted polymers (MIP) has experienced swift growth due to their benefits over larger sized imprinted polymers, including a higher surface-to-volume ratio, which, in turn, contributes to an increase of the number of binding sites with better accessibility for the target ligand, thereby reducing mass transfer kinetics. All these attributes make MIN especially interesting for binding macromolecules such as proteins, which suffer from slow binding kinetics upon high cross-link densities [5], being a promising tool for in vivo applications including virus counteracting [3], toxin scavenging [6] and drug delivery [7]. The remarkable binding performance shown by MIN towards proteins has definitely widened MIN applications in biomedicine, being recently focused on targeting surface receptor proteins expressed on cancer cells [8], such as the epidermal growth factor receptors (EGFR) [9] or the vascular endothelial growth factor (VEGF) [10].

Maximising binding affinity and reducing cross reactivity of imprinted nanoparticles plays a key role if real antibody substitutes are to be produced for the application in the biomedical field. In this regard, fundamental research on imprinting technology is crucial to generate the basic knowledge required to achieve those ideally perfect plastic receptors. Thus, the influence of the polymerisation conditions, such as monomer composition [11], cross-linker density [12, 13] or polymerisation time [14], on binding parameters must be determined when MIN with highest binding performance are desired. Within the polymerisation mixture, functional monomers are responsible for establishing non-covalent interactions with the template, which could be a linear peptide [15], a motif [3] or a whole globular protein [16, 17] when MIN targeting proteins are produced [5]. By contrast, cross-linker monomers are the ones providing physical stability to the formed non-covalent complex, having a significant effect on the physical properties of the resulting polymers and little contribution to specific interactions that may occur within the binding site [5]. In addition, the amount of cross-linker directly determines the rigidity of produced nanoparticles, and it is undoubtedly recommendable to determine the right balance of flexibility/rigidity to produce nanoparticles with improved binding performance [13]. High percentages of this constituent give rise to too rigid materials, hindering accessibility of target ligands to imprinted sites and decreasing binding affinity [12, 13]. However, high cross-linker densities contribute to selectivity improvement, as the spatial stability of binding sites increases, which reduces cross-reactivity. In any case, based on the above, more research is still needed in order to determine the optimal cross-linking density for the production of MIN that resemble more to real antibodies [12].

Over the last decade, the imprinting of polymers that undergo conformational and structural changes upon external stimuli (such as temperature, pH or light) has grown considerably [18, 19]. Particularly, polymers made of monomers such as N-isopropyl acrylamide (NIPAm) and/or N-tert-butylacrylamide (TBAm) are known for being thermoresponsive [15, 20], being capable to swell or shrink upon temperature change. Poly(NIPAm) presents a lower critical solution temperature (LCST) of about 32 °C, at which the degree of swelling of the polymer particles changes undergoing a volume phase transition from a random-coil hydrophilic structure to a more hydrophobic globular structure, which repels water and, thus, shrinks [21, 22]. Liu et al. reported that the coil-globule transition process of single chain poly(NIPAm) was divided into three steps named nucleation, adjacent mergence and bead stacking (NAS transition), finding that the final globule was composed of stacked beads rather than being a uniform compact globule [21]. This unique feature makes these materials especially interesting for protein imprinting, as temperature-controlled ligand binding-release may be done, favouring ligand access to or from imprinted sites. In this regard, thermoresponsive MIN have been reported as artificial antibodies for proteins such as lysozyme [23], thyroid peroxidase [24], bovine haemoglobin [25], trypsin [26, 27], human serum albumin [28] or the cannabinoid receptor type 1 (CB₁ receptor) [15]. On the other hand, MIN ability to respond to thermal stimuli also helps in conducting temperature-based nanoparticle purification by affinity chromatography after MIN production, which allows for discarding low affinity nanoparticles and collecting the ones with highest affinity [29]. This purification has been used after solid-phase synthesis of imprinted nanoparticles targeting proteins such as trypsin and/or kallikrein [27, 30, 31], or the CB₁ receptor [15]. Here, MIN are produced by solid-phase synthesis using the epitope imprinting approach. Thereafter, an affinity-based washing is conducted at temperatures above the LCST first, to remove low affinity nanoparticles, and later, below LCST, to dissociate the MIN-peptide complex obtaining a suspension of high affinity nanoparticles.

Thermoresponsiveness of MIN depends not only on the type of monomers used for polymerisation, but also on the cross-linker amount over functional monomers. Ambrosini et al. produced thermoresponsive MIN using a 95/5 molar ratio of NIPAm and N,N-ethylenebis(acrylamide) (BisAm) [27]. Higher percentages of BisAm (10–20%) produced nanoparticles which did not change size upon temperature change, thereby not showing thermoresponsivity [27, 30]. Navarro et al. reported the influence of the cross-linking density on thermoresponsive nanogels, and they did not observe any clear change on cloud point temperature (i.e. the temperature of the inflection point of the normalised transmittance at 500 nm vs temperature curve) of nanogels prepared with different cross-linker densities, which were synthesised using cross-linker:NIPAm ratios that varied between 25/75 and 50/50. However, the nanogels with the highest cross-linking degree suffered much less size change upon heating than nanogels prepared with less cross-linker. This may happen due to an increase of the rigidity of the system on more cross-linked materials, which also provide a lower swelling ratio [32]. Based on all the above, it is undoubtedly crucial to determine how the cross-linking amount in a particular monomer mixture influences thermoresponsive character on the resulting polymer nanogel. Besides, research is still needed to determine how all this temperature-based rearrangement and coil-to-globule transition phenomena contribute to binding behaviour of imprinted nanomaterials. Going deeper into this knowledge may pave the way for the creation of MIN with faster kinetics and better binding affinities, approaching even more to the behaviour of natural receptors.

This work aims to explore how the cross-linker density influences physicochemical characteristics and binding behaviour of peptide imprinted nanogels. Based on previous experience of our research group on solid-phase synthesis of thermoresponsive anti-CB₁ artificial antibodies [15], we wanted to go further on basic knowledge concerning this methodology to determine how temperature and cross-linking densities of produced MIN may compromise specific binding of the recombinant fusion GST-CB1 protein. Increasing binding affinity and selectivity may widen the applicability of these anti-CB₁ antibodies to other (bio)analytical assays where complex biological matrices are targeted. To this end, a series of nanoparticles have been produced here using variable cross-linker ratios versus the total monomer moles added to the polymerisation mixture. MIN have been fabricated by solid-phase synthesis using a 15 amino acid sequence (458-KVTMSVSTDTSAEAL-472) as template. This sequence matches with the extreme end of the cytosolic carboxy-terminal (C-terminal) tail of the CB₁ receptor, which is the mostly expressed G-protein coupled receptor (GPCR) in the mammalian central nervous system, being especially abundant in the human brain [33]. Dysregulation of CB₁ receptor-mediated signalling may contribute to several pathological conditions, including depression, anxiety, epilepsy, schizophrenia or neuropathic pain [34, 35]. Given the physiopathological significance of the neuronal CB₁ receptor signature and the low availability of effective and high affinity natural antibodies for this purpose [36], the production of artificial antibodies is presented here as a tool to broaden the knowledge regarding this receptor. All the findings of this research have been exploited to widen the applicability of artificial MIP-based antibodies, tested here as natural antibody substitutes for immunoprecipitation assays using streptavidin-magnetic beads, which are a versatile tool that enables the pulling down compounds labelled with biotin. Produced fit-for-purpose biotin-labelled MIN have proven effective for immunoprecipitating a recombinant GST-CB₁ fusion protein, which may be exploited in the future for the purification, enrichment and determination of the concentration of this target receptor on heterologous cell expression systems and brain tissue homogenates.