1. Molecularly Imprinted Polymers
1.1. Structural Levels Target Analytes
As biomacromolecules and viruses have complex, flexible, and fragile structures, in addition to milder preparation conditions, alternative techniques comprising different structural levels of the target analyte have been utilized. Yarman and Scheller recently summarized these levels exemplifying proteins
[1][2][3][4]. Similar approaches have also been applied to viruses (
Figure 1). These approaches include:
Figure 1. Schematic representation of structural levels of target analyte in the virus imprinting process.
(1)
Whole virus imprinting: In this approach, a whole virus is used as a template
[5][6][7][8]. In contrast to low-molecular-weight analytes, there are some obstacles when imprinting viruses. For imprinting, a high amount of pure virus is needed. Moreover, virus sample preparation requires appropriate laboratory, equipment, and experienced personnel
[9]. Moreover, due to their large number of potential interaction sites and functional groups on their surfaces, heterogeneous binding sites and higher cross-reactivities can be obtained
[6][9][10][11][12][13][14].
(2)
Functional viral protein imprinting: Viruses consist of various proteins, which have different functions. Utilizing glycoprotein gp51 of bovine leukemia virus, the S or N protein of SARS-CoV-2 as templates for MIP preparation can be examples of this approach
[15][16][17]. In addition to whole protein, subunits or peptide fragments have been applied as templates.
(i)
Subunit imprinting: Subunit imprinting is based on using the fragments of the viral protein as a template. Denizli’s group utilized the antigen-binding fragment (Fab) as a template for the determination of immunoglobulin G (IgG) on a surface plasmon resonance (SPR) chip. The MIP sensor could recognize both the target, Fab, and the whole IgG MIP synthesis on a SPR-chip. This Fab-imprinted polymer layer binds both the Fab fragment and the whole IgG molecule. Scheller’s group further extended this approach to oxidase (BMO) and reductase domains (BMR), which could recognize their targets or holo Cytochrome P450 BM3. This aspect has also been presented for SARS-CoV-2, in which the receptor-binding domain was used as a template
[18][19].
(ii)
Epitope Imprinting: To overcome the limitations in biomacromolecule and virus imprinting, exposed peptides of the analyte have been used as templates, which could recognize both the template and its holoprotein/whole virus
[10][20][21][22][23][24][25][26][27][28][29]. This concept was introduced by Rachkov and Minoura and was termed epitope imprinting, as it is similar to the immunological determinant recognized by an antibody
[10][30][31]. The first “epitope-MIP” was constructed via bulk imprinting, whereas Shea’s group immobilized the target on the support that was removed following the polymer formation generating the complementary cavities
[10][32]. Later, Scheller’s group developed a fully electrochemical approach (including template removal: anodic potential pulses) on gold surfaces
[33]. Epitope imprinting was also exploited for virus sensing for various viruses, comprising recently SARS-CoV-2
[9][12][26][34][35][36][37]. It is important to note that in the epitope approach, the area of specific interaction of the protein with the MIP is restricted to the epitope cavities. “Out-of-pocket interaction” of the protein/virus with the polymer surface can cause pronounced nonspecific binding
[38]. On the other hand, interaction with the underlying support, e.g., metal electrodes, has to be taken into consideration for whole protein/virus MIPs
[39].
1.2. Steps of MIP Preparation
(1) Formation of the pre-polymerization complex: In the first step, functional monomer(s) and the template molecules interact with each other to form the pre-polymerization complex. Two main approaches, namely covalent and noncovalent, have been described for the preparation of the pre-polymerization complex.
(i) Covalent approach: The covalent approach, which was introduced by Wulff and Sarhan
[40], and Shea
[41], is based on the formation of reversible covalent bonds between the template molecules and the functional monomer(s), followed by crosslinking. To remove the template from the polymer matrix, these chemical bonds must be cleaved, and rebinding occurs via the same covalent bonds
[42][43]. This method results in the formation of the stable and stoichiometric pre-polymerization complex. Moreover, in contrast to the non-covalent approach more homogenous binding sites can be obtained. Nevertheless, this approach has some obstacles, especially a narrow template spectrum and slower binding kinetics as compared to the noncovalent approach.
(ii) Noncovalent approach: The noncovalent approach was developed by Arshady and Mosbach
[44]. By contrast, in this approach, a pre-polymerization complex is formed via the noncovalent interactions, such as hydrogen bonds, ionic bonds, van der Waals forces, and hydrophobic interactions, between the template and the functional monomer(s)
[43]. As it resembles molecular recognition in nature, it is also called the biochemists’ approach. Template molecules can be removed by simple solvent extraction, and rebinding of the analyte is again obtained by the same noncovalent interactions. Furthermore, the template spectrum is broad. However, the yield of binding sites is low compared to the covalent approach.
To overcome the drawbacks of both approaches described above, the semi-covalent approach was developed by Whitcombe et al., which is a hybrid of two approaches
[45]. In this approach, a pre-polymerization complex is formed via a covalent bond, and rebinding of the analyte is achieved by noncovalent interactions between the polymer and the analyte
[43].
(2)
Polymerization: The second step of MIP preparation is polymerization. Bulk polymerization is most frequently exploited for the preparation of MIPs among the different formats. With this technique, monolithic structures are produced, which must be then grounded and sieved. The disadvantages of this method are that it is time-consuming, and slow binding kinetics are obtained. To overcome these drawbacks, different methods have been introduced including suspension, emulsion, or precipitation polymerization, which result in the formation of micro- or nanobeads; MIP nanomaterials such as nanoparticles and nanospheres; MIP nanomaterial composites; self-assembled monolayers of thiols; the spreader-bar technique; stamping; and electropolymerization
[6][46][47][48][49][50][51][52][53][54][55][56]. It is worth mentioning that apart from the last four formats, MIPs have to be immobilized on a transducer surface following the preparation.
(3) Template removal
The template removal step is as crucial as polymerization. Incomplete removal can result in reduced binding efficiency due to the smaller number of free binding cavities, while complete removal trials may cause partial or complete destruction of the polymeric network
[10][57][58]. Unfortunately, there is no general removal procedure for MIPs such as that described for the regeneration of aptamers
[10]. For decades, different strategies such as Soxhlet extraction, changing the pH or ionic strength, detergents, electrochemical methods, proteolytic digestion, elevated temperature, ultrasound, microwave-assisted extraction, and supercritical CO
2 have been demonstrated. Extensive information about this topic can be found elsewhere
[57][58].
2. MIP-Based Biomimetic Sensors for SARS-CoV-2 Detection
Basically, two different procedures have been described for the MIP-based biomimetic sensors against low-molecular-weight substances, biomacromolecules, and viruses. The first procedure is based on two steps of MIP preparation: (i) synthesis of MIPs separately and (ii) integration of the synthesized MIPs on a transducer
[10].
Alternatively, diverse surface imprinting methods including the aforementioned methods such as soft lithography, electropolymerization, and self-polymerization have been developed, which allows direct preparation of the MIPs on the surface of the transducer.
2.1. Electrochemical Detection of SARS-CoV-2
Electrochemical approaches are easy to apply and, due to the smaller size of instruments, experiments can be performed without the need for professional personnel or well-equipped laboratories.
Among the diverse approaches for the detection of analytes with MIP-based electrochemical sensors, voltammetric methods are widely utilized. By contrast, the number of potentiometric transducers, capacitors, or field-effect transistors is lower
[1]. It should be noted that the potential window of voltammetric sensors is restricted by the cathodic hydrogen generation and the anodic oxygen evolution. Hence, the electrode material should be considered.
Three main electrochemical readout methods have been utilized for MIP-based sensors
[1]: (i) direct measurement of electroactive analytes (low-molecular-weight analytes, proteins); (ii) measurement of the signal generated by catalytically active analytes (enzymes); (iii) indirect measurement using a redox marker such as ferricyanide, ferrocene, and ruthenium (low-molecular-weight molecules, biomacromolecules, viruses, bacteria, cells), which is based on the gate effect
[59][60]. This effect was, for the first time, described by Yoshimi et al.
[59]. However, the mechanism is still under discussion
[60]. Among the described readout methods, the last one is appropriate for virus sensing. Differential pulse voltammetry (DPV), square-wave voltammetry (SWV), or cyclic voltammetry (CV) play an important role in virus sensing. In comparison to CV, DPV and SWV are more sensitive as they allow the elimination of the charging current. In addition, electrochemical impedance spectroscopy (EIS) has been utilized in MIP-based virus sensing.
(1) Whole virus imprinting: Despite the challenges of whole virus imprinting, some successful examples have been presented for SARS-CoV-2 sensing.
Hassan’s group proposed a MIP-based sensor against the whole SARS-CoV-2 particles
[61]. The sensor was fabricated by electropolymerization of a mixture containing 3-aminophenol and virus particles on a carbon nanotube (CNT)/WO
3-modified screen-printed carbon electrode. Steps of the MIP preparation were characterized by EIS in a solution of double redox mediators ferricyanide and DCIP. LOD and LOQ values were determined to be 57 and 175 pg/mL, respectively. Furthermore, almost no cross-reactivity was observed toward H1N1, H5N1, and H3N2 influenza A viruses, whereas MERS-CoV and the other human coronaviruses resulted in about 2 and 36% of cross-reactivity, respectively. Moreover, the virus-imprinted sensor can rapidly quantify the SARS-CoV-2 concentration in clinical samples and differentiate between the healthy and infected samples. By comparing the LODs, it was claimed to obtain an almost 27-fold higher sensitivity compared to RT-PCR.
(2) Functional viral protein (N- or S-Protein) imprinting: Raziq et al. described the first biomimetic sensor for the diagnosis of SARS-CoV-2 since the outbreak of COVID-19, which employed an electrochemical transducer. As a template, a functional viral protein, namely nucleoprotein (ncovNP), was used rather than the whole virus
[17]. The MIP was prepared after covalent immobilization of ncovNP on a 4-aminothiophenol (4-ATP)-modified gold electrode. It was described that the oriented immobilization of the target prior to polymerization via site-specific anchors allows the formation of more uniform binding cavities (grafted target imprinting)
[10][62]. It can also prevent the denaturation of viral proteins due to the absorption on the metal surface. Moreover, thiol groups on the virus’ surface may lead to nonspecific interactions with a metal electrode. After molecular docking and quantum chemical calculations, m-phenylenediamine was chosen as the functional monomer. All the steps of MIP preparation were characterized with a redox marker, ferri-/ferrocyanide mixture. Limit of detection (LOD) and limit of quantification (LOQ) were calculated from DPVs to be 15 fM and 50 fM, respectively, which lies in the clinical range. Further, the selectivity ones with various proteins possessing different sizes, molecular weight, and isoelectric points demonstrated that the highest response was obtained for the target analyte ncovNP. Moreover, the performance of the sensor was exploited in nasopharyngeal swab specimens and a correlation with RT-PCR was found.
Another electrochemical MIP-based sensor, which explored ncovNP as a template, was constructed on a gold/graphene nanohybrid-modified screen-printed electrode by electropolymerization of arginine. DPV was applied to characterize the sensor. Under the optimized conditions, the peak currents of the redox marker decreased with the increase in the ncovNP concentration from 10.0 and 200.0 fM with a very small LOD value of 3.0 fM, which was fivefold less than the previous example (15 fM). Furthermore, the sensor was applied to artificial nasal and saliva samples spiked with ncovNP, and detection of ncovNP was achieved with acceptable recovery values.
2.2. Optical Detection of SARS-CoV-2
In addition to the electrochemical readout, the optical readout was exploited for the detection of SARS-CoV-2 with MIP-based sensors. Optical sensors allow the direct detection (label-free) of analytes by measuring the changes such as refractive index and fluorescence. Compared to electrochemical MIP-based sensors, the number of optical MIP-based sensors against SARS-CoV-2 is limited.
Cennamo et al. reported for the first time an acrylamide-based MIP on a POF-covered gold SPR chip addressing the specific recognition of the S1 subunit of the SARS-CoV-2 spike protein
[63]. This first prototype was exploited to detect the S1 subunit. The LOD and affinity constants were determined to be 0.058 µM and 2.318 µM
−1, respectively. Moreover, preliminary tests on SARS-CoV-2 virions were performed on samples of nasopharyngeal swabs in the universal transport medium and physiological solution (0.9% NaCl), and the results were compared with RT-PCR. They obtained a higher sensitivity and faster response. However, it was expressed that the method should be validated.
2.3. Commercial MIP for SARS-CoV-2
Several successful examples of nanoMIPs have been exploited in bioanalysis for the development of optical and electrochemical sensors
[64][65][66]. The company MIP Diagnostics has developed the first commercial nanoMIPs against different SARS-CoV-2 variants, addressing the RBD of SARS-CoV-2. The particle size varies from 40 nm to 80 nm and the affinity constants are ≤18. As the nanoMIPs are amino-functionalized, they could be immobilized on an electrode. The developed thermal resistance sensor allowed the measuring of concentrations of <5 fg/mL for the RBD from spike protein.
This entry is adapted from the peer-reviewed paper 10.3390/biomimetics7020058