OF MOLECULES

DESCRIPTION

The present invention relates to devices and methods for optical recognition of molecules, in particular for biological applications, through the use of nanopores.

The ability to identify biological molecules, such as nucleic acids and proteins, has become increasingly important in the field of basic biological research and in numerous technical fields, such as diagnostics, biotechnology, microbiology, virology, pharmaceutics and forensic chemistry. For this reason, it is believed that the possibility of recognizing and sequencing long biological molecules will revolutionize many aspects of medical research and practice in the near future.

Several techniques for biomolecule recognition and sequencing have been developed in the last decade. In particular, the so-called latest-generation sequencing methods are considered to be the most promising ones, because they ensure high analysis quality and high speed. Most of such methods are based on the use of nanopores, as described in the article by Feng et al.,“Nanopore-based fourth-generation DNA sequencing technology”, Genomics and Proteomics Bioinformatics 13, 2015, 4-16. Such nanopores are designed to allow for sequential passage of molecules and monitoring the chemical-physical differences among the single units (e.g. the nucleotides of a single DAM filament) that compose the polymer. Such differences are then translated into characteristic and discernible signals, e.g. electric or optical signals.

For example, when a potential difference is applied across the nanopore support membrane, an electric signal can be generated by the modulation of the ionic current generated by the passage of a molecule through the nanopore. Impedance variations are then used in order to determine the sequence of the polymer that has gone through the pore.

One example of technologies exploiting optical signal reading is provided, for example, by the systems developed by Quantapore Inc and described in patent applications WO2016200737, US20180163268A1, US20140335513A9.

These systems are based on the use of fluorescent nanopores and molecules, and read the optical signal generated when the analyte passes through the nanopore. In particular, such systems employ pairs of molecules called FRET (Fluorescence Resonance Energy Transfer) to permit analyte sequencing via specific recognition of marked units. These systems require the presence of fluorescent markers on the single monomers of the polymeric chain to be analyzed and also on the nanopore, so that it is possible to specifically recognize the single molecule units as a function of the fluorescent signal generated by the interaction between the pairs of fluorophores.

The nanopores commonly used for, for example, sequencing nucleic acids can be divided into two categories according to their nature: biologic nanopores and synthetic nanopores.

Biologic nanopores are proteins, common in nature, that spontaneously tend to organize themselves in such a way as to assume a toroidal shape. One example of a protein pore is a-hemolysin, but other pores of this kind are commonly found in cell membranes, in which they act as channels for transporting ions and other molecules in and out of the cell. These pores are normally stable when inserted in a support membrane, e.g. a membrane with a double lipid layer or a membrane of polymeric material.

Synthetic nanopores, also known as solid-state nanopores, are commonly created by using microfabrication techniques, usually by the so-called“ion beam sculpting”, i.e. formation of a thin solid-state film, and are characterized by good chemical, thermal and mechanical stability, freely adjustable dimensions, and integrability. Solid-state nanopores can work under a wide variety of experimental conditions and can be mass-produced by using conventional semiconductor fabrication processes.

Although previous works have demonstrated the potentialities of nanopores, in particular as regards the recognition of DNA sequences, there is still a need for new devices and methods that will make it possible to analyze long polymeric sequences, in particular proteins and other complex molecules, at the level of the single constituent units, while ensuring high speed, resolution, accuracy and volumes at low cost.

In fact, the nanopore-based devices and methods currently available are mostly used for sequencing nucleic acids and, even though they may in principle be adapted also for sequencing more complex molecules, e.g. proteins, the results obtained when sequencing such molecules are not yet reliable.

For molecules like proteins, in fact, the base units to be identified are the twenty aminoacids, some of which are present in different titration states or are altered following post-translation modifications, and this aspect complicates the analysis considerably.

At present, the most reliable analysis for protein sequencing is mass spectrometry. In particular, the most common technique for intact proteins is top-down mass spectrometry, which has recently allowed the identification of a remarkable quantity of proteins. However, although mass spectrometry is a powerful technique in the proteomics field, its use for sequencing new proteins has proven to be costly and technically complex. Moreover, a further problem related to this technique is its poor sensitivity when analyte concentration is low. In fact, mass spectrometry is not sufficiently sensitive when low protein concentrations need to be analyzed. For example, the analysis of the proteom of a single cell, which is an application that is currently of high biomedical interest, cannot be carried out by using this technique because the proteom of a cell cannot be amplified (as it is commonly done for nucleic acids by using a technique called Polymerase Chain Reaction (PCR)).

Joo et al. have described, in patent application WO2014014347A1, a protein recognition platform based on the use of nanopores and FRET molecules. In particular, they have developed a device for identifying a protein in a liquid, which utilizes a molecular machine for guiding the protein, functionalized with fluorescent markers, through a detector configured to detect a signal as a function of the fluorescent markers bound to the single aminoacids, and thus determine the sequence of the protein. However, such a device does not ensure an accurate recognition of the single units in the case of complex molecules, nor does it permit analyzing proteins of any kind. In fact, the molecular machine employed belongs to a specific enzyme class and can only be used for analyzing specific protein classes. In addition, the described system does not allow processing high volumes, and is therefore incompatible with the specific application.

It follows that there are still a number of challenges that need to be addressed within the field of molecule analysis and sequencing, such as reducing the total time necessary for sequencing, simplifying the preparation of the sample to be analyzed, possibility of analyzing very long sequences, improving the accuracy of the analysis, etc. In light of this examination, it can be stated that a technical problem at the basis of the invention is to provide a method and a device for optical recognition of molecules, the characteristics of which are such as to overcome the limitations of the above-described prior art.

Within the scope of said technical problem, it is a specific object of the invention to provide a device for nanopore-based optical recognition of molecules which permits the analysis of complex polymers at the level of the single monomeric constituent units.

It is a further object of the invention to simplify the procedure for preparing the molecules of the sample to be recognized. In particular, the objects of the invention include not requiring the amplification of the sample prior to the analysis and eliminating the need for marking the molecules to be analyzed.

The idea which solves the above-mentioned technical problem, while also achieving the aforementioned objects, is to execute an optical recognition of molecules by means of nanopores in which there is at least one flexible and/or deformable element or appendix, extending from the inner wall or surface of the nanopore and capable of switching, in a reversible manner, from an idle condition to a deformed condition upon interaction with a molecule passing through the nanopore; in accordance with a preferred solution, said flexible element can interact with light.

According to the invention, a nanopore-based optical recognition device is therefore proposed which can be used for molecular analysis purposes. In particular, one possible application of the device concerns the biological field, for the analysis of molecules such as proteins and nucleic acids.

The nanopores according to the invention are toroidal, or ring-like, structures, the nature of which is biologic (e.g. membrane proteins and other proteins capable of self-assembling to form a ring-like structure), synthetic (e.g. Si3N4, Si02, MoS2, A1203 and other dielectrics; graphene and other conductors) or hybrid (e.g. DNA origami) inserted in a substrate or matrix; preferably, the inside diameter of the nanopores ranges from 0.5 to 10 nm, and their thickness ranges from 0.5 to 20 nm. Such nanopores allow the molecule to be analyzed to be translocated as a single chain, even for small biologic molecules, and the single constituent units dwell within the nanopore for sufficient time to complete the analysis.

The nanopores according to the invention comprise flexible elements or appendices constrained to the inner surface of the nanopore and protruding inwards. The nature of such flexible elements may be biologic (e.g. peptides and/or oligonucleotides) or synthetic (e.g. poly ethers, polymethacrylates, polycaprolactones, cyclodextrins, etc.) and have a length shorter than or equal to approximately half the diameter of the nanopore. Such elements, in the absence of a molecule within the nanopore, preferably lie in an idle position and leave a gap, for the passage of the molecule to be analyzed, having a diameter preferably ranging from 1.5 to 3.2 nm. When a molecule passes through the nanopore, the flexible elements interact with it and modify their own conformation from the idle one, depending on the molecule involved, thus assuming a deformed condition.

The nanopores according to the invention comprise anchor points arranged on their inner surface, which allow the flexible elements to be connected to the nanopore. Said anchor points can be connected to the nanopore surface by using known surface modification and/or molecular biology techniques, as described, for example, by Marie et al. (Marie, R., Dahlin, A. B., Tegenfeldt, J. O., Hook, F. (2007). Generic surface modification strategy for sensing applications based on Au/Si02 nanostructures. Biointerphases, 2(1), 49-55.) and by Ardini et al. (Ardini, M., Golia, G, Passaretti, P., Cimini, A., Pitari, G, Giansanti, F., Morandi, V. (2016). Supramolecular self-assembly of graphene oxide and metal nanoparticles into stacked multilayers by means of a multitasking protein ring. Nanoscale, 5(12), 6739-6753.).

According to a preferred embodiment, said flexible elements comprise chiral molecules and/or fluorescent molecules (e.g. fluorophores, aromatic aminoacids, etc.), which make it possible to follow the movements of the flexible elements, induced by the passage of the molecules through the nanopore, by means of optical methods. In particular, the presence of chiral molecules permits highlighting the conformational variations of the flexible elements by using the circular dichroism technique.

An apparatus for optical recognition of molecules according to the invention comprises at least one nanopore, with one or more flexible elements inside of it, as previously described herein. The nanopore is inserted in a biologic or synthetic substrate and/or matrix, and translocation of the molecule within the nanopore is induced by applying a potential difference across the support matrix. Suitable optical means for detecting signals associated with the passage of molecules through the nanopore make it possible to analyze the molecule.

Although the nanopore-based device has been conceived mainly for biological and biotechnological applications, such as, for example, protein sequencing, it may also prove useful in other applications where analysis of small molecules is required. In particular, the possible fields of application also include the pharmaceutical and toxicological fields, wherein the nanopores can be used for recognizing small molecules such as antibiotics, glucose, mycotoxins, neurotoxins, organophosphorus compounds, etc.

The features and advantages of the product proposed herein will be presented in the detailed description that follows with reference to the annexed drawings, supplied merely by way of non-limiting example, wherein:

- Figs. 1 and 2 show a schematic representation of a nanopore according to the invention;

- Fig. 3 is a schematic sectional representation of a nanopore during the interaction with a molecule to be analyzed;

- Figs. 4, 5 and 6 are graphs that show the results concerning the classification of aminoacids obtained by simulating an apparatus according to the invention;

- Fig. 7 is a table containing data relating to the percent deformation undergone by the flexible elements that are present within the nanopore as a molecule to be analyzed passes through the nanopore.

Example 1

According to one possible embodiment, the nanopore 1 is a modified version of the Prxl protein of the Schistosoma Mansoni (SmPrxI) (schematically represented in Figs. 1 and 2). In physiological conditions, ten monomeric sub-units of Prxl assemble to form a ring-like closed structure having a thickness of 5 nm, an inside diameter of 6 nm, and an outside diameter of 13 nm (PDB code: 3ZTL).

The modified Prxl protein was produced as described by Ardini et al. in " Supramolecular self-assembly of graphene oxide and metal nanoparticles into stacked multilayers by means of a multitasking protein ring" Nanoscale, 2016, 8, 6739-6753. In particular, the SmPrxI was produced from the recombinant bacterium BL21-DE3 pLysS containing a plasmid suitably modified to obtain the modified leucine zipper pattern and the hystidine tag at the N-terminal. The BL21-DE3 recombinant cells were cultivated in an LB selective culture medium, and expression of the protein was induced with isopropyl-P-D- 1 -thiogalactopyranoside (IPTG). A clarified cellular extract was collected after a sonication process followed by ultracentrifugation, and then loaded on a column balanced with Ni2+ for purification. The SmPrxI monomers were then eluted from the column by imidazole gradient for further characterization and appropriate functionalization. Under the conditions indicated by Ardini et al. in the above-mentioned article, the protein monomers self-assemble to form a ring-like structure.

Proteins other than Prxl may be used as well, such as, for example, other proteins belonging to the family of 2-Cys peroxiredoxins (PRXs) and, more, in general, all transmembrane proteins capable of forming ring-like structures.

The inner surface of the nanopore 1 was modified by attaching ten flexible elements 10 (one flexible element per monomer constituting the nanopore) connected to corresponding anchor points 2 located at a distance of 9 A from each other. In particular, said anchor points 2 are represented by the N-terminal ends of each monomer. Preferably, said flexible elements 10 are disposed in an initial position, also referred to as idle position, and extend inwardly from the anchor points 2 on the inner surface of the nanopore 1.

In one possible embodiment, said flexible elements 10 are protein molecules modelled after the helical leucine zipper pattern. In particular, said elements are composed of a sequence of aminoacids modified, compared to that of the leucine zipper pattern, through the addition of two cysteine residues. Furthermore, said flexible elements are composed of a sequence of 16 aminoacids modified, compared to the leucine pattern sequence, in positions 1 and 10; in position 1 of the chain, a cysteine molecule replaces a leucine molecule, and in position 10 a cysteine molecule replaces a serine molecule, so as to give the following sequence: 1CYS- 2GLU-3ASP-4LYS-5VAL-6GLU-7GLU-8LEU-9LEU-10CYS 1LYS-12ASN-13TYR- 14HIS-15LEU-16GLU (PDB: 2ZTA). Said sequence has a total length of approximately 25 A and has a strong inclination towards the helical structure.

Preferably, the glutamic acid (GLU) in position 16 is connected to the inner surface of the nanopore 1 through the anchor point 2, and the cysteine (CYS) in position 1 constitutes the free end of the flexible element 10 that extends towards the inside of the nanopore 1, i.e. that portion of the flexible element which is free to oscillate during the passage of the molecule to be analyzed 4 (the distances between the free ends of the flexible elements 10 within the nanopore are indicated in Fig. 1, expressed in nanometers).

In a preferred solution, the flexible elements 10 are mutually connected in pairs by formation of a disulfide bridge between the cysteines in position 1 of adjacent flexible elements 10. This solution allows reducing the undesired oscillations of the flexible elements 10 in the absence of the molecule to be analyzed 4 within the nanopore 1, and ensures better stability of the flexible elements 10 during the interaction with the molecule 4 (Fig. 3 shows the flexible elements 10 during the interaction with the molecule to be analyzed 4, and the arrows indicate possible directions of motion of said flexible elements 10).

Furthermore, the cysteine in position 10, through interaction with a maleimide group, can be used as an element for attaching molecules to the flexible element 10. In particular, fluorescent molecules 3 can be connected to the flexible elements. Such molecules can be used for tracing the conformational modifications undergone by the flexible elements.

In one specific embodiment, said fluorescent molecules 3 are connected to the flexible elements 10 in order to make it possible to trace the variations occurring in the distance between adjacent flexible elements 10 during the translocation of the molecule to be analyzed 4 within the nanopore 1. In particular, said fluorescent molecules 3 are a pair of so-called FRET molecules. More in particular, the pair of molecules ATTO 425 and ATTO 647N was chosen as a FRET pair. The Forster critical distance of the FRET pairs, 3.6 nm and 4.3 nm, was chosen on the basis of the inside diameter of the nanopore (approx. 6 nm). In order to ensure FRET signal maximization, 5 fluorescent molecules (2 donor molecules and 3 acceptor molecules) were used.

The signal emitted by the FRET molecules is sensitive to the movements of the flexible elements 10 induced by the passage of the molecule to be analyzed 4 through the nanopore 1

The apparatus for optical recognition of molecules to be analyzed 4, comprising at least one nanopore 1 according to the above description, comprises in this specific example, as optical detection means, a light source with a wavelength such that it excites the donor molecule of the FRET pair, and a detector capable of detecting the luminous emission of the acceptor molecule of the FRET pair. In the specific embodiment wherein the FRET molecule pair consists of the ATTO 425 and ATTO 647N molecules, a light source with a wavelength of 425 nm is used in order to excite the donor molecule ATTO 425.

The analysis of the luminous signal generated by interaction between FRET pairs allows discerning the single units of the molecule to be analyzed 4 that has passed through the nanopore 1 at single-unit level (e.g. in the case of a protein, at the level of a single aminoacid).

Identification of the molecule 4 at single-unit level is attained by implementing a machine learning data analysis platform that allows correlating the luminous signal, generated by interaction between FRET pairs, with the single units constituting the molecule to be analyzed 4. Some examples of this type of platforms are known in the art (see scientific publication Schiitte, M, Risch, T., Abdavi-Azar, N., Boehnke, K., Schumacher, D., Keil, M, ... & Worth, C. L. (2017). Molecular dissection of colorectal cancer in pre-clinical models identifies biomarkers predicting sensitivity to EGFR inhibitors. Nature Communications, 8, 14262.)

Compared to the sequencing systems already available on the market or described in the literature, the newly developed system ensures overall stability in terms of structure solidity and accurate recognition of a wide range of elements. For example, the twenty aminoacids that constitute proteins can be identified as a function of the FRET signal generated by the movements made by the flexible elements 10 upon the passage of the molecule to be analyzed 4. In particular, the stability of adjacent flexible elements 10 marked with donor/acceptor FRET pairs promotes maximization of the signal-to-noise ratio during the passage of the molecule 4 through the nanopore 1.

The insertion of the flexible elements 10 modelled after leucine zippers makes the nanopore 1 more sensitive to the structural adjustments induced by the passage of the molecule 4. In particular, it induces an intensification of the FRET signal through a change in the mutual distance and orientation of the donor and the acceptor of the FRET pair. Unlike the other systems available on the market and in the literature, this newly implemented system requires no tag on the molecule to be analyzed 4.

Molecular dynamics simulations were carried out in order to study the behaviour of the nanopore 1 (in the absence of fluorescent molecules 3) for 1 ps, using the GROMACS (GROningen MAchine for Chemical Simulations) simulation software. The system was inserted into explicit solvent using the ΉR3R explicit water model, including also 60 Na ⁺ ions (for the purpose of neutralizing the net charge). The system was prepared for the simulation by using a standard minimization and balancing protocol starting from a constrained system and progressively removing the constraints as temperature increased. The production simulation took place in the canonical ensemble (NVT). The production trajectories were analyzed by using the K-Means clustering algorithm. The medoid of the most populated cluster, i.e. the cluster element whose mean dissimilarity with respect to all other elements is lowest, was used as being representative of the conformation and as input for the subsequent calculations of the so-called“molecular docking”.

The molecular docking approach was used in order to identify the most stable positions where the fluorescent molecules 3 could be attached. In the simulation, the fluorescent molecules 3 were positioned near the flexible elements 10, at a distance compatible with the distance that would be obtained in the presence of an anchor element (linker) for connecting the fluorescent molecule 3 to the flexible element 10. The molecular docking simulations were carried out for all of the selected fluorescent molecules. Firstly, the fluorescent molecules 3 were bound, without a linker, to the nanopore 1 in a manner such as to identify a number of possible preferred positions and orientations. Each configuration was evaluated by means of short simulations of“smoothed potential” molecular dynamics, wherein scaling the potential energy reduces the energetic barriers and facilitates the identification of the most stable configurations, as in the work by Mollica et al.“Kinetics of protein-ligand unbinding via smoothed potential molecular dynamics simulations” (Sci. Reports, Nature Publishing Group, 5, Article number: 11539, 2015).

Subsequently, each fluorescent molecule 3 in the chosen configuration was connected to the flexible elements 10 by using a maleimide molecule as an attachment molecule.

The molecular dynamics simulations demonstrated the stability of the structure over a time of 1 ps and the effectiveness of the selected design in reducing the undesired effects related to the oscillations of the flexible elements within the pore, thus maximizing the signal-to- noise ratio.

In order to evaluate the accuracy of discrimination of the units of a molecule to be analyzed 4, a molecular dynamics simulation was carried out, wherein the FRET signal generated upon the passage of said molecule through the nanopore 1 was estimated by using a simulation of the system in the presence of an endecapeptide at the center of the nanopore for a duration of 125 ns and 8 different angles of rotation of the endecapeptide.

The efficiency parameter ( E ) of the FRET pair and the Forster distance ( Ro ) were extrapolated from the trajectory files by extrapolating the positions of key atomic markers during the simulation.

The efficiency of the FRET pair, E, depends on the distance between donor and acceptor according to a law that varies with the sixth power of the radius, r, due to the dipole-dipole coupling mechanism, as described by the equation:

Ro is the distance between donor and acceptor at which energy transfer efficiency is 50%, and depends on the superimposition of the integral of the emission spectrum of the donor on the emission spectrum of the acceptor and on their mutual molecular orientation, as expressed by the formula:

Nine endecapeptides with a generic sequence GGGGGXGGGGG were studied, wherein G represents the glycine aminoacid and X represents one of the TRP, TYR, ARG, HIS, GLU, GLN, SER, ILE and GLY aminoacids. In every simulation, lasting 125 ns, a different relative initial orientation of the side chains of X relative to the structure was taken into account. The cumulative time refers, therefore, to the reference plus the 8 replicas for the 9 different endecapeptides, for a total of 10 ps.

Finally, in order to obtain an estimate of the aliasing effect occurring during the translocation of a real polypeptide through the nanopore, a simulation was made of a realistic sequence extracted from a protein and its translocation through the nanopore at a speed variable from 0.2 to 7.2 nm/ns.

Based on the modifications of the FRET signals, caused by the interaction of the nanopore and the flexible elements with each aminoacid, the reduction in the T-SNE dimension of 5,000 randomly chosen data was analyzed (as shown in Fig. 4).

A machine-learning classification model was then used in order to identify the single aminoacids. In particular, the“Support Vector Machine” (SVM) and“Random Forest” algorithms were selected for the classification.

The processing of the data demonstrated that the translocation of each aminoacid implies a characteristic modification of the FRET signal compared to the idle signal, and allows discerning the single aminoacids with a high level of accuracy. Fig. 5 shows the sharp separation among the data clusters relating to the passage of the single aminoacids during a simulation, wherein the analyzed aminoacids were TRP, TYR, ARG, HIS, GLU, GLN, SER, ILE, GLY.

Furthermore, the reconfiguration of the protein-based elastic elements 10 modelled after the leucine zipper pattern was demonstrated by evaluating the helicality variations undergone by the flexible elements 10. Helicality was defined as the mean number of residues in alpha- helix form throughout the duration of the simulation. The results obtained for each molecular dynamics simulation carried out are shown in the graph of Fig. 6.

The difference between the idle state and the deformed state of the flexible elements 10, which is dependent on the specific aminoacid translocating through the nanopore 1, was calculated in terms of percent deformation of the flexible elements 10 as the value of the mean distance between the main atom chain of the protein, calculated as: where N is the number of atoms of the helical elements, ro is the idle position, and rs is the deformed position.

The table in Fig. 7 shows the values of the mean distances, in percentage terms, of the atoms facing towards the center of the nanopore 1, compared to a reference structure.

Example 2

In another possible embodiment of the invention, the nanopore 1 is a synthetic nanopore of dielectric material (e.g. S1 ₃ N ₄ , S1O ₂ , M0S ₂ , AI ₂ O ₃ and other dielectrics; graphene and other conductors).

The geometry of the synthetic nanopore and its dimensions are controlled in a precise manner and in accordance with the design requirements and the chosen fabrication technique.

In the case of synthetic nanopores, the flexible elements 10, realized as previously explained, must comprise a terminal cluster that permits interaction with the inner surface of the pore. Some possible functionalizations are the following: functionalization with a thiol group in the case ofMoS ₂ nanopores; functionalization with siloxanes containing COOH/NH ₂ groups in the case of S13N4, S1O2 or AI2O3 nanopores.

Hybrid nanopores (DNA origami) offer another possible alternative structure that may be used.

Both solid-state nanopores and hybrid nanopores are functionalized with flexible elements 10, similarly to protein nanopores. They may host flexible elements 10 like those previously described or made of different materials.

Some examples of polymers that can be used for creating the flexible elements 10 are: peptides, PEG, PEG in combination with peptides, PEG-methacrylate, poly(8-caprolactone), polyisoprene, as well as cyclodextrins and oligonucleotides.

Solid-state nanopores modified with polymeric flexible elements may gain additional physical-chemical properties, such as, for example, sufficient stability and balance between hydrophobic and hydrophilic properties, which can ensure a fast readaptation of the structure after the translocation of the molecule. Moreover, they offer the possibility of making modifications with different tag molecules in order to follow the structural readaptation of the structure. Example 3

In a further embodiment of the invention, the nanopore 1 is a protein or synthetic nanopore, as described in the preceding examples, wherein the flexible elements 10 comprise one or more chiral molecules (e.g. aminoacids). The presence of said chiral molecules makes the flexible elements 10 sensitive to polarized light, so that they can be analyzed by using the known circular dichroism technique. In other words, in the presence of chiral molecules the conformation changes undergone by the flexible elements 10 during the passage of the molecule to be analyzed 4 are traced as a function of the differences in the absorption of the circularly polarized light by the chiral substance, without the necessity of using fluorescent molecules 3 connected to the flexible elements 10.

The apparatus for optical recognition of molecules to be analyzed 4, based on nanopores 1 according to the invention, comprises in this specific example, as optical detection means, a source of linearly polarized light, a monochromator, a frequency modulator, which lets in the right-hand circularly polarized light first and then the left-hand circularly polarized light (or vice versa), and a detector, which can detect the optical signal determined by the structural rearrangement of the flexible elements 10 caused by the passage of the molecule 4 through the nanopore 1.