WO2022008720A1 - Novel hybrid optical voltage sensors - Google Patents

Abstract

The invention relates to an optical voltage sensor comprising a plasma membrane-anchored fluorophore and a lipophilic small molecule quencher capable of redistribution within the plasma membrane in response to transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%. The invention further relates to a method to detect a change of transmembrane potential across a plasma membrane of a cell and a kit of parts comprising said plasma membrane-anchored fluorophore and said lipophilic small molecule quencher.

Description

NOVEL HYBRID OPTICAL VOLTAGE SENSORS

The present invention relates to optical voltage sensors comprising a membrane anchored fluorophore and a lipophilic small molecule quencher.

BACKGROUND OF THE INVENTION

Direct optical voltage sensing using genetically expressed probes is highly promising for large scale recordings of neuronal activity. Hybrid voltage sensing based on Forster resonance energy transfer (FRET) between a fluorescent particle anchored to the plasma membrane and a small lipophilic anion that can rapidly translocate in the membrane has been pioneered over 20 years ago (10). The approach has been refined by using the FRET reaction between a stationary fluorescent lipid and a mobile dye (11). The principle was turned into a genuine “hybrid genetically encodable voltage indicator (hGEVI)” approach by using a genetically encodable membrane-targeted fluorescent protein as the membrane anchored fluorophore, and dipicrylamine (DPA) as its FRET pair (12). DPA was known from early charge-pulse relaxation experiments to electrophorese through lipid membranes with a sub-millisecond translocation rate (13). As the DPA absorption and eGFP emission spectra do not greatly overlap, improvements in the method have been attained by using the blue-shifted cerulean fluorescent protein (18) and by developing a membrane localized fluorophore (hybrid voltage sensor (hVOS) 2.0) (19). In general, the hGEVI approach provides good signal-to-noise ratio for the detection of action potentials (APs) and for recording sub threshold synaptic events in various preparations, but in all previous studies the voltage-dependent small molecule quencher remained the same: DPA. While this molecule satisfies many of the original requirements for hybrid voltage sensing, it also has several drawbacks. These include its accumulation on the outer surface of the membrane below approximately -50 mV thus making it difficult to report small membrane hyperpolarisations (21), the considerable capacitive membrane load near the required concentrations for voltage sensing (12, 20, 24) that causes time-dependent deterioration of APs (17), and its interactions with various neurotransmitter systems (25-27). In addition, as it is made up of two trinitrotoluene (TNT) molecules joined together, DPA is highly explosive.

In order to develop more sensitive hybrid voltage sensors and to replace DPA in the hGEVI approach, the inventors carried out a systematic search for different quenching molecules suitable to quench specific fluorophores.

The voltage sensor of the present invention provides inter alia the following advantages: (i) the voltage sensor is more sensitive than known hybrid genetically encoded voltage indicators (hGEVIs); (ii) the voltage sensor is faster than known hybrid genetically encoded voltage indicators (hGEVIs), enabling it to detect APs from rapidly firing cell types such as intemeurons; (iii) the voltage sensor is more specific than known hybrid genetically encoded voltage indicators (hGEVIs); (iv) the voltage sensor shows a linear response to voltages over a wide range (-100 to +40 mV) of membrane potentials, enabling it to detect even small membrane hyperpolarizations; (v) the voltage sensor has a high dynamic range, enabling it to differentiate between APs and failures; (vi) the voltage sensor does not affect cell survival, passive membrane properties and synaptic events, in particular the voltage sensor does not affect membrane capacitance, action potential width, frequency of synaptic excitatory and inhibitory post synaptic currents (sEPSCs and sIPSCs) and peak amplitude of sEPSCs and sIPSCs; (vii) the voltage sensor does not produce photocurrents upon illumination which disturb the membrane potential; (viii) the quenchers of the voltage sensor stay within the membrane once loaded and do not need to be continuously present in the extracellular space; thus, the voltage sensor is suitable for in vivo recordings; (ix) the quenchers of the voltage sensor do not exhibit any auto fluorescence, thereby improving the specificity of the voltage sensor; (x) the fluorophores of the voltage sensor are characterized by high fluorescence, thereby rendering the voltage sensor very strong; (xi) the fluorophores of the voltage sensor do not exhibit excessive bleaching, thereby rendering the voltage sensor very stable.

SUMMARY OF THE INVENTION

In a first aspect the present invention relates to an optical voltage sensor comprising (a) a fluorophore, linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; and (b) a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%.

In a second aspect, the present invention relates to a method to detect a change of transmembrane potential across a plasma membrane of a cell, comprising the steps of effecting within the cell expression of a fluorophore, linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; contacting the cell with a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; illuminating the cell with light suitable for excitation of the fluorophore; detecting a fluorescence intensity emitted from the fluorophore; and detecting a change of transmembrane potential, wherein a change in fluorescence intensity corresponds to a change of transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%.

In a third aspect, the present invention relates to a fluorescent protein comprising or consisting of (a) an amino acid sequence of SEQ ID NO: 1, wherein Xi is selected from K and Q, and wherein amino acids 242 to 249 independently may be present or absent, or (b) an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or at least 99% identity to SEQ ID NO: 1, wherein Xi is present and selected from K and Q, wherein the fluorescent protein emits fluorescence upon excitation, preferably with an emission maximum between 440 550 nm, 480-530 nm, 495-515 nm or 505-512 nm. In a fourth aspect, the present invention relates to a nucleic acid encoding the fluorescent protein of the third aspect. In another aspect, the present invention relates to a vector comprising the nucleic acid of the fourth aspect.

In a fifth aspect, the present invention relates to a kit of parts comprising a first reagent comprising a fluorophore, preferably a fluorescent protein, more preferably the fluorescent protein of the third aspect, or a nucleic acid encoding said fluorescent protein, or a vector comprising said nucleic acid, or a cell comprising said fluorophore, wherein the fluorophore is linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; and a second reagent comprising a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Before the present invention is described in detail below, it is to be understood that this invention is not limited to the particular methodology, protocols and reagents described herein as these may vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to limit the scope of the present invention, which will be limited only by the appended claims. Unless defined otherwise, all technical and scientific terms used herein have the same meanings as commonly understood by one of ordinary skill in the art.

Preferably, the terms used herein are defined as described in "A multilingual glossary of biotechnological terms: (IUPAC Recommendations)", Leuenberger, H.G.W, Nagel, B. and Klbl, H. eds. (1995), Helvetica Chimica Acta, CHAO 10 Basel, Switzerland) and as described in "Pharmaceutical Substances: Syntheses, Patents, Applications" by Axel Kleemann and Jurgen Engel, Thieme Medical Publishing, 1999; the "Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals", edited by Susan Budavari et ah, CRC Press, 1996, and the United States Pharmacopeia-25/National Formulary-20, published by the United States Pharmcopeial Convention, Inc., Rockville Md., 2001.

Several documents are cited throughout the text of this specification. Each of the documents cited herein (including all patents, patent applications, scientific publications, manufacturer's specifications, instructions, etc.), whether supra or infra, are hereby incorporated by reference in their entirety. Nothing herein is to be construed as an admission that the invention is not entitled to antedate such disclosure by virtue of prior invention.

To practice the present invention, unless otherwise indicated, conventional methods of chemistry, biochemistry, and recombinant DNA techniques are employed which are explained in the literature in the field (cf, e.g., Molecular Cloning: A Laboratory Manual, 2^nd Edition, J. Sambrook et al. eds., Cold Spring Harbor Laboratory Press, Cold Spring Harbor 1989).

Throughout this specification and the claims which follow, unless the context requires otherwise, the word “comprise”, and variations such as “comprises” and “comprising”, will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps. As used in this specification and the appended claims, the singular forms “a”, “an”, and “the” include plural referents, unless the content clearly dictates otherwise.

In the following, some definitions of terms frequently used in this specification are provided. These terms will, in each instance of its use, in the remainder of the specification have the respectively defined meaning and preferred meanings.

As used herein, the term “protein”, “peptide”, “polypeptide”, “peptides” and “polypeptides” are used interchangeably throughout. These terms are used in the context of the present invention to refer to both naturally occurring peptides, e.g. naturally occurring proteins and synthesized peptides that may include naturally or non-naturally occurring amino acids.

In the context of the present specification, the name “Disperse Orange 1” refers to a compound of the following formula:

In the context of the present specification, the name “Disperse Orange 3 (D3)” refers to a compound of the following formula:

In the context of the present specification, the name “Disperse Orange 13” refers to a compound of the following formula:

In the context of the present specification, the name “Disperse Orange 25” refers to a compound of the following formula:

In the context of the present specification, the name “Disperse Orange 37” refers to a compound of the following formula:

In the context of the present specification, the name “Disperse Red 1” refers to a compound of the following formula:

In the context of the present specification, the name “Disperse Red 13” refers to a compound of the following formula:

In the context of the present specification, the name “Disperse Blue 124” refers to a compound of the following formula:

In the context of the present specification, the name “Oil Red O” refers to a compound of the following formula:

In the context of the present specification, the name “DABCYL SE” refers to a compound of the following formula:

In the context of the present specification, the name “Dipicrylamine (DPA)” refers to a compound of the following formula:

The terms absorbance and absorption are used interchangeably within the present specification.

In the context of the present specification, the abbreviation “AP” refers to “action potential”.

In the context of the present specification, the abbreviation “RMP” refers to “resting membrane potential”.

In the context of the present specification, the abbreviation “FRET” refers to “Forster resonance energy transfer”.

In the context of the present specification, the expression “fluorophore” is used to refer to a compound with “fluorogenic properties”. The expression “fluorogenic properties” refers to the ability of a compound to form a fluorescence emitting product.

In the context of the present specification, the term “small molecule” refers to compounds having a molecular weight of less than 900 Da.

In the context of the present specification, the term “quencher” refers to a compound capable of decreasing the fluorescence intensity of a given substance.

In the context of the present specification, the term “extinction coefficient” refers to a parameter that specifies the capacity for light absorption at a specific wavelength (unit: cm ¹ M ¹).

In the context of the present specification, the term “hVOS 2.0” relates to the hVOS 2.0 construct described in Wang et al., 2010 (19).

In the following, the elements of the present invention will be described. These elements are listed with specific embodiments, however, it should be understood that they may be combined in any manner and in any number to create additional embodiments. The variously described examples and preferred embodiments should not be construed to limit the present invention to only the explicitly described embodiments. This description should be understood to support and encompass embodiments, which combine the explicitly described embodiments with any number of the disclosed and/or preferred elements. Furthermore, any permutations and combinations of all described elements in this application should be considered disclosed by the description of the present application unless the context indicates otherwise.

In a first aspect the present invention relates to an optical voltage sensor comprising (a) a fluorophore, linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; and (b) a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%. In preferred embodiments of any aspect of the present invention, the spectral overlap of the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher is at least 75%, at least 80%; at least 85%, at least 90% at least 95% (a high degree of spectral overlap is preferred).

The expression “linked to an anchoring moiety” is meant to indicate that the fluorophore and the anchoring moiety are covalently or non-covalently, preferably covalently, directly or indirectly bound to each other.

The anchoring moiety may be an anchoring moiety capable of anchoring the fluorophore to the outer leaflet of the plasma membrane or to the inner leaflet of the plasma membrane. The anchoring moiety preferably is or comprises a lipid moiety. Anchoring is effected by insertion of the anchoring moiety into the plasma membrane (inner or outer leaflet), and thereby determining that the localization of the fluorophore is close to the plasma membrane.

The skilled person is aware that emission and absorption spectra may have several (local) maxima. The expression “absolute maximum” refers to the highest maximum.

The inventors have carried out a systematic search for quencher fluorophore FRET pairs suitable for hybrid genetically encodable voltage indicators.

Quencher

From the plurality of available quenchers, the inventors in a first step focused on disperse dyes, which are mainly azobenzene- or anthraquinone-based molecules. The inventors further selected small hydrophobic azobenzene-based molecules. Depending on the fluorophore used in the voltage indicator, quenchers having a high spectral overlap with the fluorophore were selected. A multitude of small molecule quenchers was tested by recording single fluorescence traces in 10 mM of the respective quencher using an encoded membrane anchored fluorophore. The results are shown in Figure 12. Quenchers using which no signal or only a weak signal was produced were deleted from the list of possible quenchers (e.g. Phenolblue, Reichardt’s dye, Oxonol IV, BHQ-2). In addition, quenchers which affected cell survival, passive membrane properties or synaptic events were also excluded from further analysis (e.g. Merocyanine 540, New methylene blue).

In certain embodiments, the spectral overlap is an absolute spectral overlap.

In certain embodiments, the spectral overlap is a normalized spectral overlap. For the calculation of the normalized spectral overlap, the height of the absorption maximum of the quencher is normalized to the height of the emission maximum of the fluorophore (see Fig. 14A for eGFP and several quenchers). The normalized spectral overlap is then calculated as area under the curve (AUC) of the emission spectrum of the fluorophore for the individual quenchers.

The efficiency of a quencher to quench a fluorophore is also influenced by the peak absorption value of the quencher, i.e. by the absorption value at the absorbance maximum of the quencher. In general, a stronger quencher is characterized by a higher absorption value. The skilled person is well capable of determining an absorption value for a quencher. One way to determine the absorption value is described in the methods section. The determined absorption values of several quenchers (peak absorption value, absorption value at 475-650 nm, absorption value at 500-550 nm) are shown in Table 1.

In preferred embodiments, the quencher is characterized by a peak absorption value of at least 0.06, at least 0.08, at least 0.10, at least 0.12, at least 0.15, or at least 0.17 (a high peak absorption value is preferred).

A further parameter to describe the efficiency of a quencher to quench a fluorophore is the extinction coefficient of the quencher. In the context of the present specification, the parameter “extinction coefficient” (unit: cm ¹ M ¹) specifies the capacity for light absorption at a specific wavelength. The peak extinction coefficient relates to the extinction coefficient at the absolute absorbance maximum of the quencher. In general, a stronger quencher is characterized by a higher extinction coefficient. The skilled person is well capable of determining an extinction coefficient for a quencher. One way to determine the extinction coefficient is described in the methods section. The peak extinction coefficient and the extinction coefficient at 509 nm of several quenchers are shown in Table 1.

In preferred embodiments, the quencher is characterized by a peak extinction coefficient of at least 10,000 cm ¹ M ¹, at least 12,500 cm ¹ M ¹, at least 15,000 cm ¹ M ¹ or at least 17,500 cm ¹ M ¹ (a high peak extinction coefficient is preferred).

Table 1

In preferred embodiments, the quencher is an aryl azo compound. In preferred embodiments, the quencher is characterized by a molecular weight of 200-450 g/mol, particularly 250-430 g/mol, more particularly 310-410 g/mol. This size range allows for an efficient translocation of the quencher into the plasma membrane and for a high degree of mobility within the plasma membrane.

In preferred embodiments, the quencher is an aryl azo compound characterized by a molecular weight of 200-450 g/mol, particularly 250-430 g/mol, more particularly 310-410 g/mol. The plasma membrane of an animal cell comprises an inner leaflet and an outer leaflet. In a neuronal cell, the RMP, i.e. the membrane potential that prevails if the cell is not excited (or inhibited) typically ranges from -70 to -80 mV. Depending on the structure of the quencher, it is envisioned that at RMP, the quencher resides closer to the outer surface of the plasma membrane or closer to the inner surface of the plasma membrane. In response to a change in transmembrane potential, the quencher redistributes within the membrane, in other words changes its localization within the membrane, either from one leaflet to the other leaflet or from the “middle” between the two leaflets into one of the leaflets, or within one leaflet closer towards or further away from the surface of the adjacent plasma membrane.

In preferred embodiments of the optical voltage sensor, the quencher resides neither very close to the outer surface of the plasma membrane nor very close to the inner surface of the plasma membrane but rather “in the middle” between the inner and the outer surface of the plasma membrane, at RMP. In preferred embodiments, it is envisioned that a depolarization of the plasma membrane results in a distribution of the quencher to the outer leaflet of the plasma membrane and a hyperpolarization of the plasma membrane results in a distribution of the quencher to the inner leaflet of the plasma membrane.

The quencher is capable of redistribution within in the plasma membrane in response to a deviation of 3.6 mV. A change of plasma membrane potential of 3.6 mV is the minimal deviation that can be detected. The skilled person is aware that the quencher also redistributes in response to larger changes in plasma membrane potential.

The quencher is capable of redistribution within in the plasma membrane in response to a hyperpolarization of 3.6 mV.

In preferred embodiments, the quencher redistributes within in the plasma membrane in response to a deviation of at least 3.6 mV from resting membrane potential.

In preferred embodiments, the quencher redistributes within in the plasma membrane in response to a hyperpolarization of at least 3.6 mV from resting membrane potential.

In preferred embodiments, the quencher does not affect neuronal cells with regard to membrane capacitance, action potential width, cell survival, frequency of synaptic excitatory and inhibitory postsynaptic currents (sEPSCs and sIPSCs) and peak amplitude of sEPSCs and sIPSCs. In particular, at a concentration of 20 mM or less, the quencher does not affect neuronal cells with regard to membrane capacitance, action potential width, cell survival, frequency of synaptic excitatory and inhibitory post synaptic currents (sEPSCs and sIPSCs) and peak amplitude of sEPSCs and sIPSCs. Accordingly, the voltage sensor of the present invention does not disturb the neuronal properties. The quenchers used in the voltage sensor of the present invention are devoid of the harmful properties of DPA.

The inventors have demonstrated that the quenchers used in the voltage sensor of the present invention stay within the membrane once loaded and do not need to be continuously present in the extracellular space; thus, the voltage sensor is also suitable for in vivo recordings. Furthermore, the quenchers of the voltage sensor do not exhibit any auto fluorescence. This improves the specificity of the voltage sensor. Fluorophore

In preferred embodiments of any aspect of the present specification, the fluorophore is a fluorescent protein. In instances where the fluorophore is a fluorescent protein, said fluorescent protein may comprise additional N- and/or C-terminal amino acid sequences as described below, in particular a membrane translocation sequence, a sequence motif to which the anchoring moiety can be attached and optionally one or more linkers to connect said sequences to the amino acid sequence of the fluorescent protein.

It is advantageous if the fluorophore exhibits a strong fluorescence. The fluorescent proteins used in the voltage sensor of the invention exhibit a particularly strong fluorescence, as indicated by their extinction coefficient and brightness in Table 2.

Table 2

Generally, the fluorophores of the voltage sensor are characterized by high fluorescence, thereby rendering the signal generated by the voltage sensor very strong. In addition, the fluorophores of the voltage sensor do not exhibit excessive bleaching upon prolonged excitation, thereby rendering the voltage sensor very stable (Fig. 21).

It is preferred that the fluorophore, preferably the fluorescent protein, comprises a membrane translocation sequence. In instances where the fluorophore is a fluorescent protein, it is preferred that the membrane translocation sequence is comprised at the N-terminus of the fluorescent protein. Preferably the membrane translocation sequence comprises the proacrosin signal peptide (MVEMLPTVAVLVLAVSVVA - SEQ ID NO: 4), which is cleaved during protein maturation, and the proacrosin N-terminal peptide (KDNTT - SEQ ID NO: 5), which is part of the mature fluorescent protein.

The fluorophore, in particular the fluorescent protein further comprises a motif to which the anchoring moiety can be attached. In instances where the fluorophore is a fluorescent protein, it is preferred that the motif is comprised at the C-terminus of the fluorescent protein. The skilled person is well capable of determining the amino acid sequence of such motifs depending on the preferred anchoring moiety to be attached to the fluorescent protein.

Anchoring moieties capable of anchoring the fluorophore, in particular a fluorescent protein, to the inner leaflet of the plasma membrane are preferably selected from the group consisting of a famesyl moiety, a geranylgeranyl moiety and a fatty acid, in particular palmitic acid, myristic acid, stearic acid, or arachidonic acid. Preferred examples of motifs to which anchoring moieties for anchoring to the inner leaflet of the plasma membrane can be attached are a GAP43 motif or a truncated h-ras motif. The “GAP43 motif’ relates to an N-terminal motif comprised in the neuromodulin protein and comprises two cysteine residues that can be palmitoylated to S-palmitoyl cysteine. The “h-ras motif’ relates to a C-terminal motif comprised in the h-ras protein and comprises two cysteine residues that can be palmitoylated to S-palmitoyl cysteine and one cysteine that can be famesylated to S-famesyl cysteine.

In preferred embodiments, the fluorophore is anchored to the outer leaflet of the plasma membrane. Thus, the fluorophore is located outside the cell, close to the extracellular surface of the plasma membrane. An anchoring moiety capable of anchoring the fluorophore, in particular a fluorescent protein, to the outer leaflet of the plasma membrane is e.g. a glycosylphosphatidylinositol (GPI) moiety. An example of a motif to which a GPI moiety can be attached is the Thy-1 GPI anchoring signal (LENGGISLLVQNTSWMLLLLLSLSLLQALDFISL - SEQ ID NO: 6). Of this sequence, only the first three amino acids (LEN) remain in the mature, processed form of the fluorescent protein, the rest of the sequence is cleaved during protein maturation. Another possibility of anchoring a fluorescent protein to the outer leaflet of the plasma membrane is to fuse the fluorescent protein with a transmembrane domain of Platelet-derived growth factor receptor beta (PDGFR{5), as described in Toffalini et al., 2010, J Biol Chem. 2010 Apr 16; 285(16).

Preferably, the anchoring moiety is a GPI moiety. The use of a GPI moity positively affects the expression level of the fluorescent protein compared to moieties for anchoring the fluorophore to the inner leaflet of the plasma membrane (Fig. 17). It is preferred that the GPI moiety is fused to the C-terminus of the fluorescent protein.

The fluorescent protein can be truncated at the terminus where the motif is located to which the anchoring moiety is attached, preferably the C-terminus. In particular, 1 to 20, 1 to 17, 1 to 15, 1 to 12, 1 to 10, 1 to 9, 3 to 9, 5 to 9, 7 to 9, 8 to 9, or 9 amino acids at the N- or C-terminus, preferably the C-terminus, of the fluorescent protein are deleted. It is envisioned that deletion of amino acids prior to the motif to which the anchoring moiety is attached has the effect to bring the fluorescent protein closer to the plasma membrane and thus closer to the quencher. Without wishing to be bound by theory, the inventors expect that this increases the sensitivity of the voltage sensor according to the invention.

It is envisioned that in certain embodiments, when anchored to the plasma membrane, the mean distance of the fluorophore to the midplane of the plasma membrane is 60 A or less, 50 A or less, 40 A or less, or 30 A or less (a shorter distance is preferred). eGFP

In preferred embodiments, the fluorophore has an emission spectrum characterized by an absolute emission maximum of 440-550 nm. In more preferred embodiments, the fluorophore has an emission spectrum characterized by an absolute emission maximum of 480-530 nm. In even more preferred embodiments, the fluorophore has an emission spectrum characterized by an absolute emission maximum of 495-515 nm. In most preferred embodiments, the fluorophore has an emission spectrum characterized by an absolute emission maximum of 505-512 nm.

Preferably, the fluorophore is or comprises a green fluorescent protein, in particular selected from the group consisting of GFP, enhanced GFP (eGFP), Emerald, Superfblder GFP, Azami Green, mWasabi, TagGFP, Turbo GFP, AcGFP, ZsGreen and T-Sapphire. It is preferred that the green fluorescent protein is eGFP. As described above with respect to fluorescent proteins, it is preferred that the eGFP comprises a membrane translocation sequence, which is preferably located at the N-terminus of eGFP. Preferably the membrane translocation sequence comprises the proacrosin signal peptide (MVEMLPTVAVLVLAVSVVA - SEQ ID NO: 4), which is cleaved during protein maturation, and the proacrosin N-terminal peptide (KDNTT - SEQ ID NO: 5), which is part of the mature fluorescent protein.

In addition, it is preferred that the eGFP comprises a motif to which the anchoring moiety can be attached, which is preferably located at the C-terminus of eGFP. In preferred embodiments, the motif to which the anchoring moiety can be attached is the Thy-1 GPI anchoring signal (LENGGISLLVQNTSWMLLLLLSLSLLQALDFISL - SEQ ID NO: 6). Of this sequence, only the first three amino acids (LEN) remain in the mature, processed form of eGFP.

In preferred embodiments, the most C-terminal amino acid of the eGFP sequence, which in the original eGFP sequence is a K, has been mutated to Q. An advantage of this mutation is that it makes the eGFP less prone to degradation.

As described above, it is preferred that the eGFP is truncated at the terminus where the motif is located to which the anchoring moiety is attached, preferably the C-terminus. Preferably 1-9, more preferably 5-9, even more preferably 9 amino acids at the C-terminus, prior to the motif to which the anchoring moiety is attached, are deleted. In instances where the most C-terminal amino acid of the original eGFP sequence has been mutated from K to Q, this Q is maintained and the 1-9 amino acids N-terminal to the Q are deleted. A comparison between truncated and non-truncated eGFP is shown in Fig. 18.

It is preferred that the unprocessed eGFP has an amino acid sequence of SEQ ID NO: 2 (unprocessed GPI-eGFP construct), wherein Xi is selected from K and Q and preferably is Q and wherein amino acids 261 to 268 independently may be present or absent, preferably absent, or an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or at least 99% identity to SEQ ID NO: 2, wherein XI is present and selected from K and Q, preferably Q, wherein eGFP emits fluorescence upon excitation, preferably with an emission maximum between 440-550 nm, 480-530 nm, 495-515 nm or 505-512 nm.

It is preferred that the eGFP has an amino acid sequence of SEQ ID NO: 1 (mature, processed GPI- eGFP construct), wherein Xi is selected from K and Q and preferably is Q and wherein amino acids 242 to 249 independently may be present or absent, preferably absent, or an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or at least 99% identity to SEQ ID NO: 1, wherein the eGFP emits fluorescence upon excitation, preferably with an emission maximum between 440-550 nm, 480-530 nm, 495-515 nm or 505-512 nm.

The determination of percent identity between two sequences is accomplished using the mathematical algorithm of Karlin and Altschul, Proc. Natl. Acad. Sci. USA 90, 5873-5877, 1993. Such an algorithm is incorporated into the BLASTN and BLASTP programs of Altschul et al. (1990) J. Mol. Biol. 215, 403-410. To obtain gapped alignments for comparative purposes, Gapped BLAST is utilized as described in Altschul et al. (1997) Nucleic Acids Res. 25, 3389-3402. When utilizing BLAST and Gapped BLAST programs, the default parameters of the respective programs are used. Alternatively, a variant can also be defined as having up to 20, 15, 10, 5, 4, 3, 2, or 1 amino acid substitutions, in particular conservative amino acid substitutions. Conservative substitutions are well known in the art (see for example Creighton (1984) Proteins. W.H. Freeman and Company). An overview of physical and chemical properties of amino acids is given in Table 3 below. In a particular embodiment, conservative substitutions are substitutions made with amino acids having at least one property according to Table 3 in common (i.e. of column 1 and/or 2).

Table 3

In some embodiments, the membrane-anchored eGFP is encoded by a nucleic acid sequence of SEQ ID NO: 3. The skilled person is aware that alternative nucleic acid sequences can be used to encode the same fluorescent protein, based on different codon usage.

Quenchers suitable for eGFP

In preferred embodiments, the quencher is characterized by an absolute absorbance maximum of 440-550 nm. In preferred embodiments, the quencher is characterized by an absolute absorbance maximum of 480-550 nm.

The capacity of a quencher to quench fluorescence in a certain wavelength range can be expressed by an absorption value for a certain wavelength range. One way to perform this calculation is described in the methods section.

In preferred embodiments, the quencher is characterized by an absorption value between 475 nm and 650 nm of at least 5; at least 7.5; at least 10; at least 12.5; at least 15; or at least 17.5.

In preferred embodiments, the quencher is characterized by an absorption value between 500 nm and 550 nm of at least 2.25; at least 3; at least 4; at least 5; at least 6; at least 7; or at least 8.

The capacity of a quencher to quench fluorescence at a certain wavelength can further be expressed by the extinction coefficient at said wavelength. One way to determine the extinction coefficient is described in the methods section.

In preferred embodiments, the quencher is characterized by an extinction coefficient at 509 nm of at least 5,500 cm ¹ M ¹, at least 7,500 cm ¹ M ¹, at least 10,000 cm ¹ M ¹, at least 12,500 cm ¹ M ¹, at least 15,000 cm ¹ M ¹ or at least 17,500 cm ¹ M ¹ (a high extinction coefficient at 509 nm is preferred). In preferred embodiments, the quencher is characterized by formula 1

(1), wherein

R¹ is selected from NO₂, NR and N=NR⁴, wherein R³ is a C₁-C₃ alkyl,

R⁴ is a substituted or non-substituted aryl moiety, and R² is selected from H, OH, NR⁵R⁶, and (CH₂)_niC(=0)0R⁷ wherein R⁵ is selected from H and a C₁-C₅ alkyl,

R⁶ is selected from the group consisting of H and a substituted or non-substituted aryl moiety, in particular a phenyl moiety, and (CH2)_n2R⁸, wherein n2 is 1-3, in particular 2, and R⁸ is CN, OH or 0C(=0)R⁹, wherein R⁹ is a C1-C5 alkyl;

R⁷ is a succinimidyl moiety and nl is 0-5,

aryl moiety; and

X¹, X², X³ and X⁴ are independently selected from H, F, Cl, Br and I.

In preferred embodiments, R¹ is NO2.

In preferred embodiments, R³ is CH₃.

In preferred embodiments, R⁴ is a substituted or non-substituted phenyl or naphthyl moiety. In preferred embodiments, R⁹ is C(=CH2)CH3.

In preferred embodiments,

is a phenyl moiety.

In preferred embodiments, X¹, X², X³ and X⁴ are H.

In preferred embodiments,

is a phenyl moiety, R¹ is NO2 and X¹, X², X³ and X⁴ are H.

In preferred embodiments, the quencher is selected from the group consisting of Disperse Orange 37, Disperse Orange 13, Disperse Orange 3, DABCYL SE, Disperse Orange 25, Disperse Orange 1, Disperse Red 1, Oil Red O, and Disperse Red 13. In more preferred embodiments, the quencher is selected from the group consisting of Disperse Orange 3, DABCYL SE, Disperse Orange 25, Disperse Orange 1, Disperse Red 1, Oil Red O, and Disperse Red 13. In even more preferred embodiments, the quencher is Disperse Orange 3.

In most preferred embodiments of the optical voltage sensor according to the first aspect of the invention, the quencher is Disperse Orange 3 and the fluorophore is eGFP. In alternative embodiments of the optical voltage sensor according to the first aspect of the invention, the fluorophore is selected from the group consisting oftdTomato, Kusabira Orange, mOrange, RFPs (e.g. RFP630, RFP611), and DsRed and the quencher is Disperse Blue 124. Preferably, the fluorophore is tdTomato and the quencher is Disperse Blue 124. A comparison of the sensitivity of known voltage sensor hVOS 2.0 and the voltage sensor according to the first aspect of the invention can be seen in Fig. 13. GPI-eGFP-D3 is capable of resolving membrane potential changes of 3.6 mV at a frequency of 83 Hz or higher with minimal phase lag. In contrast, hVOS 2.0 can only resolve membrane potential changes of 5 mV at a frequency of 72.5 Hz with a considerable phase lag. Thus, GPI-eGFP-D3 shows an increased voltage- and temporal sensitivity. A further comparison of known optical voltage sensors and the optical voltage sensor according to the first aspect of the invention can be seen in Table 4.

Table 4

dFoF (AF/F) refers to a relative change in fluorescence and is the fluorescent trace divided by the mean of 200 ms prior baseline that contained no obvious spontaneous activity in the electrical recording.

The z-score equivalent (SNR) is a parameter that characterizes the signal-to-noise ratio and is measured as the AF/F of the AP divided by the standard deviation (SD) of the 200 ms prior baseline that contained no obvious spontaneous activity in the electrical recording. A high signal to noise ratio is preferred.

A further advantage of the voltage sensor according to the invention is that it does not produce photocurrents upon illumination which disturb the membrane potential. The analysis of photocurrents is described in Abdelfattah et ah, 2019 (41). In a second aspect, the present invention relates to a method to detect a change of transmembrane potential across a plasma membrane of a cell, comprising the steps of effecting within the cell expression of a fluorophore, linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; contacting the cell with a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; illuminating the cell with light suitable for excitation of the fluorophore; detecting a fluorescence intensity emitted from the fluorophore; and detecting a change of transmembrane potential, wherein a change in fluorescence intensity corresponds to a change of transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%.

In preferred embodiments of the second aspect, the cell is a neuronal cell. In even more preferred embodiments of the second aspect, the cell is a neuron.

In preferred embodiments of the second aspect, the method is carried out in vivo.

In preferred embodiments, the optical voltage sensor according to the first aspect of the invention is used in the method according to the second aspect of the invention.

Using the optical voltage sensor according to the first aspect of the invention, a change of 100 mV in the transmembrane potential of the cell generally corresponds to an average change in fluorescence intensity (dFoF) of at least 1%, at least 2%, at least 3%, at least 4%, at least 5%, at least 6%, at least 6.5%.

The method according to the second aspect of the invention allows for a very sensitive detection of changes of transmembrane potential.

Using the optical voltage sensor according to the first aspect of the invention, changes of transmembrane potential over a range from -100 mV to 40 mV generally elicit a linear response of fluorescence intensity. This enables the detection of subthreshold depolarizations and subthreshold hyperpolarizations and the differentiation between APs and failures.

The method according to the second aspect of the invention allows the detection of action potentials elicited at a frequency of 100 Hz. This enables the recording of APs from rapidly firing cell types such as intemeurons.

It is sufficient if the cell is contacted with the quencher only once in 60 minutes. The quencher does not need to be continuously present in the extracellular space. Preferably, the cell is contacted with the quencher at a concentration of 20 mM or less, in particular 10 pM or less.

In a third aspect, the present invention relates to a fluorescent protein comprising or consisting of (a) an amino acid sequence of SEQ ID NO: 1, wherein Xi is selected from K and Q, and wherein amino acids 242 to 249 independently may be present or absent, or (b) an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or at least 99% identity to SEQ ID NO: 1, wherein Xi is present and selected from K and Q, wherein the fluorescent protein emits fluorescence upon excitation, preferably with an emission maximum between 440 550 nm, 480-530 nm, 495-515 nm or 505-512 nm.

In a fourth aspect, the present invention relates to a nucleic acid encoding the fluorescent protein of the third aspect. In another aspect, the present invention relates to a vector comprising the nucleic acid of the fourth aspect. Preferably, the nucleic acid of the fourth aspect encodes an unprocessed fluorescent protein comprising a membrane translocation sequence such the proacrosin signal peptide (SEQ ID NO: 4), which is cleaved during protein maturation, and the proacrosin N-terminal peptide (SEQ ID NO: 5) which is maintained in the mature fluorescent protein, as well as a motif to which an anchoring moiety can be attached, such as the Thy-1 GPI anchoring signal (SEQ ID NO: 6). An example for a nucleic acid sequence encoding an unprocessed GPI-eGFP construct is SEQ ID NO: 3.

In a fifth aspect, the present invention relates to a kit of parts comprising a first reagent comprising a fluorophore, preferably a fluorescent protein, more preferably the fluorescent protein of the third aspect, or a nucleic acid encoding said fluorescent protein, or a vector comprising said nucleic acid, or a cell comprising said fluorophore, wherein the fluorophore is linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; and a second reagent comprising a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%.

Preferably, the fluorophore is a fluorophore described as preferred with respect to the first aspect of the invention and the quencher is a quencher described as preferred with respect to the first aspect of the invention. More preferably, the kit comprised a combination of a fluorophore and a quencher that is described as preferred with respect to the first aspect of the invention.

Brief Description of Drawings

Fig. 1 shows the two hGEVI methods, GPI-eGFP (A) and hVOS 2.0 (B) with their fluorophores and small molecule quenchers, D3 and DPA. The GPI-eGFP sensor is composed of an eGFP tagged to the outer leaflet of the plasma membrane via a GPI anchor. The hVOS 2.0 is composed of a cerulean fluorescent protein (CeFP) tagged to the inner leaflet of the membrane N-terminally with a GAP43 motif and C-terminally with a truncated h-ras motif (Wang at al., Biophys T, 2010). The fluorescence of the fluorophores is quenched by either of the small molecule quenchers, D3 or DPA, which rapidly change position in the membrane in a voltage-dependent manner. The proposed position in the plasma membrane and the direction of the positional change during de- and hyperpolarizations is indicated. (C) Confocal images of two rat cortical neurons in the same culture expressing GPI-eGFP 6 weeks after recombinant adeno-associated virus (rAAV) transduction. Typically, neurons were used for experiments 2-3 weeks after transduction. Insets show the somata 4-fold magnified in reverse color to emphasize the fluorescence in the membrane.

Fig. 2 shows simultaneous optical and electrical recordings of membrane potential changes in a cultured neuron using the hGEVI approach with 10 mM D3. (a) The panels show single continuous traces of optical recordings (red) without image processing or filtering, as sampled at 1.08 kHz with an EM-CCD camera. The patch-clamp recordings in the I-clamp configuration (black) were sampled at 50 kHz. Various current pulses of 300 ms duration were injected into the neuron to produce hyper and depolarizations of the membrane and AP firing. The average AF/F (± SEM) for the 27 APs depicted in this trace was 5.01 ± 0.05 %. Horizontal dashed line indicates the resting membrane potential of -65 mV. (b-e) Parts of the traces labelled with the respective letters on panel (a) are shown in enlarged snapshots as superimposed traces of optical (red) and electrical (black) recordings. Note the highly accurate temporal overlap between the fluorescence and membrane potential changes (f) Overlay of fluorescence (red) and membrane voltage (black) during an AP at higher temporal resolution. Note the superimposition of the two traces during both the pre-AP voltage rising to threshold and during the post-AP hyperpolarization (g) Fluorescence and membrane potential are shown normalized to the peak of the AP. (h) The sampling data points are shown for the two recording modalities to illustrate the rapid change in fluorescence in spite of the relatively low sampling frequency.

Fig. 3 shows a calculation of the correspondence between AF/F and membrane potential (a) A similar current injection protocol as used in Fig. 2 in another cultured neuron expressing GPI-eGFP recorded in the presence of 10 mM D3. Superimposed traces of non-image processed fluorescence sampled at 1.08 kHz (red) and the I-clamped membrane voltage (black) down-sampled to the same frequency from the original 50 kHz. The horizontal dashed line indicates the resting membrane potential of -60 mV. (b) A portion of the traces is shown from the part enclosed in the dashed lined box of panel (a). Note the excellent overlay between the optical and electrical recordings as indicated by the accurate reporting of the decreasing AP amplitudes during the train, and the subthreshold depolarizing events (probably spontaneous EPSPs; arrowhead) of <10 mV amplitudes (c) The relationship between AF/F and membrane potential was calculated by plotting the two traces pointby- point against each other. The slope of the linear regression (black line) yields the relationship for this cell as indicated in the inset.

Fig. 4 shows (Figure 4) the speed of the hGEVI approach as measured during the decay of APs. (a) Rapidly decaying APs (usually those early on during a current pulse injection) were normalized to their peaks (grey: electrophysiology; black: fluorescence) and single exponentials were fitted to their decay phases (dotted red lines) (b) The same was done for slower APs (usually recorded during the late phases of the depolarizing current injections) . (c) Comparison of the fast AP decays show a remarkable one-to-one correspondence between the optical and the electrical measurements (dotted line with a slope=l for comparison, red line is the linear fit to the data; slope = 1.0334 ± 0.0366; R2=0.746). (d) Similar to (c) but for the APs with slower decays (slope = 1.0495 ± 0.026; R2=0.939). Values expressed as mean ± SD. Fig. 5 shows amplitude and phase correspondence between AF/F and Vm during subthreshold stimuli of increasing frequencies (a) Superimposed plots of the AF/F (red) and Vm (black) during a 4 s chirp pulse (10-100 Hz). Morlet wavelet transforms of the AF/F trace (top) and of the Vm trace (bottom) showing the linearly increasing frequency responses. Note how the AF/F response faithfully follows the linear change in frequency (b-c) Zoomed images of the AF/F and Vm traces from the shaded boxes labelled on panel (a). The top panels show the smoothed AF/F (dark red) and raw AF/F (light red) traces, superimposed with the Vm trace (black). Bottom panels depict the phases of the AF/F (red) and Vm (black) with the smoothed AF/F and Vm traces in the background shown in fainter colors (d) Plot of AF/F phase vs Vm phase. Individual points are represented in red, while the greyscale represents the binned histogram values with a bin width of 0.04p rad. Pearson’s R=0.61; p= 0. (e) Normalized cross-correlation between AF/F phase and Vm phase. The RMS normalization of the cross-correlogram was done as described in the Materials and Methods section (f) Histogram of point-by-point differences between AF/F phase and Vm phase. The mean of the Gaussian fit (red) is at -0.011 rad. i.e., -311 ps. The mean ± SEM lag betwen AF/F phase and Vm phase determined in this manner in 8 experiments was -276 ± 240 ps.

Fig. 6 shows the accuracy of AP detection at 50 Hz and 100 Hz and its ROC analysis (a) Detection of APs elicited with 4 ms current pulses at 50 Hz. Upper panel: raw (light red) and smoothed (red) AF/F signal of example recording. Threshold (dashed line) for detection of fluorescent APs (fAPs) was set at 75% peak amplitude of the first fAP, determined as the peak AF/F in a ±3 ms time window of the first electrophysiological (Vm) AP relative to a 180 ms baseline period. Crosses indicate threshold crossings, peaks of detected fAPs are indicated in blue. Lower panel: Corresponding electrophysiological trace. Threshold for detection of Vm APs was set at 0 mV. (b) Superimposed raw fluorescent and electrophysiological traces from (a) (R=0.98). (c) Same as (a) at 100 Hz. (d) Superimposed raw fluorescent and electrophysiological traces from (c) (R=0.95).

Fig. 7 shows optical recordings of synchronous activity (a) Two fluorescent cultured neurons on the same coverslip. Cell 1 was recorded in the I-clamp whole-cell mode. Cell 2 was only optically monitored (b) When treated with 4-AP (50 mM) neurons developed epileptiform bursting. The simultaneous recordings show the fluorescent signals in the two cells (red and green traces) and the electrical recording in Cell 1 (black trace) (c) The area shaded in grey on the left of panel (b) magnified to show suprathreshold activity in Cell 2 and subthreshold activity in Cell 1. (d) The area shaded in grey on the right of panel (b) magnified to show suprathreshold activity in both cells. Note the alignment of the start of the synchronous discharges in both (c) and (d). Fig. 8 shows long-lasting optical measurements of membrane voltage following removal of the acceptor/quencher D3 from the extracellular space. Individual traces of raw unfiltered and unprocessed fluorescent signals (sampled at 1.08 kHz) of membrane potential changes following the indicated times after washout of 10 mM D3 from the recording chamber. The current injection protocol is the same as that shown in Figs 2 and 3. Each protocol required a 6 s continuous illumination and was repeated 4-times every 10 min. In spite of the multiple exposures to light, the SNR (z-score) of the first AP in the train was remarkably constant over time (at 0 min: 13.8 and at 60 min: 12.2).

Fig. 9 shows lack of effects of D3 on passive membrane and AP properties. Whole-cell recordings were carried out in cultured neurons (n=12) without the expression of GPI-eGFP. Each measurement was taken before (Pre) and 10 min after (Post) the perfusion of 20 pM D3 (in 0.2% DMSO), a concentration 2-fold higher than that used in the hGEVI experiments (a) Membrane capacitance, (b) input resistance, (c) AP width measured at 50% amplitude, also referred to as FWHM, and (d) AP threshold measured as the voltage at the first peak of the third derivative of the membrane potential. All statistical comparisons are done using Wilcoxon matched-pairs signed rank tests.

Fig. 10 shows comparison of the effects of DMSO, DPA and D3 on passive membrane and AP properties. Whole-cell recordings were carried out in cultured neurons without the expression of GPI-eGFP. (a) Comparisons of input resistance (Rn), membrane capacitance (Cm), AP width measured at 50% amplitude (FWHM), and AP threshold (APth) measured as the voltage at the first peak of the third derivative of the membrane potential. Measurements for Rn and Cm were taken before (Left bars) and 5 min after (Right bars) the perfusion of DPA (2.5 mM, n=7; 3 pM, n=6; and 5 pM, n=4 cells; total pooled n = 17 cells), D3 (10 pM, n=6; and 20 pM, n=l l cells; total pooled n = 17 cells), and DMSO (0.2%, n = 9 cells). The only significant effect was the increase in Cm by DPA (from 68.31 ± 6.46 to 118.63 ± 23.33 pF; p=0.0032). The values for FWHM and APth were assessed before (Left bars) and 10 min after (Right bars) the perfusion of DPA (2.5 pM, n=7 cells), D3 (10 pM, n=6; and 20 pM, n=12 cells; total pooled n = 18 cells), and DMSO (0.2%, n = 8 cells). The only significant effect was the increase in FWHM by DPA (from 1.537 ± 0.140 to 3.911 ± 0.737 ms; pO.OOOl). Significance was assessed by post two-way ANOVA Sidak’s multiple comparisons tests with p-values adjusted for multiple comparisons (b) Representative APs during three experimental conditions before (pre, black) and 10 min after (post, red) perfusion of the compounds indicated above the traces. Dashed vertical lines mark AP threshold (c) Evolution of Cm changes and cell survival during the time following perfusion of 3 pM DPA or 20 pM D3. Left axis represents normalized mean Cm (± SEM) calculated as the ratio to the averaged value of the pre-perfusion period. Right axis shows the number of cells. The number of cells (dotted to solid lines) surviving the perfusion of DPA gradually diminished over the period of 10 min, whereas no cells were lost to D3 perfusion. Fig. 11 shows a cross-comparison between hVOS 2.0, GPI-eGFP, and GPI-CeFP with the quenchers DPA and D3 as indicated. (A) Recording from HEK293T cells transfected with the respective construct as indicated. Voltage-dependent quenching of the fluorescence depends on the combination of quencher and fluorophore. Values are averages of GPI-eGFP D3: n=9 cells, GPI- CeFP D3: n=5 cells; hVOS 2.0 D3: n=7cells, GPI-eGFP DPA: n=6 cells; hVOS 2.0 DPA: n=7 cells. (B) superimposed traces showing the differences in AP kinetics between groups.

Fig. 12 shows the summary of the screening for alternative small molecule quenchers. Left panels:

Absorbance spectra of identified alternative small molecule quenchers: Dabcyl SB, Oil Red O, and the disperse dyes Redl, Orange 1, Orange25, Orange37 and blue 124. Middle panels: Single fluorescence traces recorded in 10 mM of the respective quencher. Optical recordings (red) are shown without image processing or filtering, sampled at 1.08 kHz with an EM-CCD camera. Right: the Fluorophore used for the measurement is indicated.

Fig. 13 shows superimposed plots (a, c) of the AF/F (red) and Vm (black) during a 4 s chirp pulse with increasing frequency (20-83 Hz) and increasing amplitude (3.6 mV - 19.6 mV from resting membrane potential (RMP)). (b, d) Zoomed images of the AF/F and Vm traces from the shaded boxes labelled on panels (a). The panels show the smoothed AF/F (dark coloured) and raw AF/F (light coloured) traces, superimposed with the Vm trace (black).

Fig. 14 shows the normalized spectral overlap of the indicated quenchers with the indicated fluorophores.

The height of the absorption maximum of the quenchers is normalized to the height of the emission maximum of the fluorophore. (A) Normalized spectral overlap between eGFP and the indicated quenchers. (B) Normalized spectral overlap between eGFP and the indicated quenchers calculated as area under the curve (AUC) of the eGFP emission spectrum for the individual quenchers. (C) Normalized spectral overlap between fluorescent proteins eGFP and CeFP and Disperse Orange 3 and DPA. The normalized spectral overlap between CeFP and DPA calculated as AUC is 66%.

Fig. 15 shows the determined absolute absorption values of the respective quenchers. The tables show a calculation of absolute absorption values for certain ranges of wavelengths calculated as the area under the curve (AUC) for the respective quenchers in the 475-650nm and 500-550nm range.

Fig. 16 (A) shows the normalized spectral overlap of the quencher blue 124 with tdTomato. The height of the absorption maximum of the quencher is normalized to the height of the emission maximum of the fluorophore. (B) Single fluorescence traces recorded with tdTomato in 10 mM blue 124. Simultaneous voltage (black, right axis) and optical recordings (gray, left axis) are shown without image processing or filtering, sampled at 1.08 kHz with an EM-CCD camera. Fig. 17 shows HEK293T transfected with hVOS 2.0 and GPI-eGFP constructs using the transfection agent Lipofectamine 3000. Pictures were taken 48 h after the transfection.

Fig. 18 shows a comparison of signal strength and quality using a voltage sensor with a truncated and a non-truncated eGFP. Bar graphs show the average SNRs. Dots show the single events. Error Bars indicate the Standard error of the mean.

Fig. 19 shows the absorption spectra determined for the quencher molecules of Table 7.

Fig. 20 shows a comparison of the kinetics of action potentials waveforms recorded with a non-hybrid voltage indicator, ASAP2s, and GPI-eGFP D3. (a) Action potential waveforms recorded in HEK293T cells transfected with ASAP2s and GPI-eGFP D3 in voltage clamp mode. Optical traces (black) show the average of 12 traces from 12 cells. The voltage trace is indicated in gray.

(b) Comparison of AP width of the optical traces measured at 50% amplitude. Significance was assessed by Wilcoxon rank-sum test.

Fig. 21 shows that GPI-eGFP D3 has a high photostability. (a) Normalized fluorescence intensity of GPI- eGFP D3 over 15 minutes continuous illumination with an intensity of 4.3mW/mm². Datapoints show averages of n=5 cells. Error bars show s.e.m. Graph is superimposed with the photostability curves for ASAP2f, Ace2N-mNeon, Voltron525, and Voltron549 in primary neuronal cultures (shaded), illumination intensity ~23 mW/mm2) (Abdelfattah AS et ah, (2018) Bright and photostable chemigenetic indicators for extended in vivo voltage imaging. bioRxiv, preprint) (b) Normalized SNRs of GPI-eGFP D3 over 15 minutes continues illumination of the same 5 neurons.

(c) example traces from a single cell showing spontaneous action potentials at BL and after 15minutes illumination.

Fig. 22 shows that GPI-eGFP D3 resolves dendritic signal propagation (a) 128 x 128 pixel image of GPI- eGFP-expressing cultured cortical primary mouse neuron. Spontaneous activity was elicited by incubation with 50mM 4 aminopyridin. (b) Superimposed single traces of the dendritic signals at a proximal location (orange) and at a more distal location (green). The somatic signal (black) precedes the dendritic signals (c) Quantification of the soma-dendritic delay of a single neuron measured 12.6 pm (orange) and 40.62 pm (green) from the soma, n = 13 APs) error bars indicate s.e.m.

Fig. 23 shows that D3 persists in the membrane for 48 hours (a) left: superimposed average trace of 5 action potential waveforms induced in a HEK293T cell expressing GPI-eGFP measured 24 and 48 hours after 10 minutes of incubation with D3. Right: bar graph showing SNRs 24 and 48 hours after incubation compared to Ctrl (b) simultaneous optical (red) and electrical (black) recording from a mouse cultured neuron recorded 24 hours after 10 minutes of incubation in 2mM D3. The patch-clamp recordings in the I-clamp configuration (black) were sampled at 50 kHz. Upper panel: Whole trace of current pulses of 300 ms duration injected into the neuron to produce hyper- and depolarizations of the membrane and AP firing. Enlarged snapshots of hyperpolarizing (lower left) and depolarizing (lower right) current injections.

Examples Section

Methods

Calculation absolute absorption values:

Absorption values of the quencher were acquired using a Varian Cary® 50 UV-Vis Spectrophotometer at a concentration of 10 mM using a cuvette with a diameter of 1 cm. Absolute peak absorption values correspond to the absorption measured at the absolute absorption maximum of the respective quencher. Absolute absorption values for certain ranges of wavelengths were calculated as the area under the curve (AUC) for the respective quenchers in the 475-650nm and 500-550nm range, respectively.

Calculation extinction coefficient:

The extinction coefficient was calculated using Lambert Beer law for a 1 cm cuvette:

EC (wavelength x (nm)) = absorbance (wavelength x (nm)) / concentration (M)

Animals

Timed-pregnant Wistar rats and C57BL/6N mice were used to prepare cell cultures between days 16 and 18 of embryonic development. Adult (>11 weeks old) C57BL/6N mice of both sexes were used to prepare brain slices. All animal storage, handling, and experiments were conducted in accordance with the guidelines of the Animal Care and Use Committee of the University of Bonn.

Cell cultures

For cell culturing, molecular biological and virus preparation procedures, all reagents were purchased from Thermo Scientific (Dreieich, Germany), unless indicated otherwise. Cortical neurons were obtained from Wistar rats or C57BL/6N mice mice between the days 16 and 18 of embryonic development. Pregnant rats or mice were anesthetized with isoflurane, decapitated, and the embryos were removed from the uterus. After decapitation of the embryos, the cortices were isolated in HBSS buffer and digested with trypsin (0.25%) and DNAse I (1 mg/ml, purchased form Sigma (St. Louis, MO, USA)). Cells were cultured in the mixture of Basal Medium Eagle (BME) enriched with Fetal Bovine Serum (FBS, 1%), B-27® Supplement (2%), Glucose (1%) and LGlutamine (0.23%). Cells were plated at a density of 20,000 - 35,000 cells per well onto poly-D-Lysine coated coverslips (d=12 mm) in a 24-well plate. Twenty-four hours post preparation the plating medium was changed to 1 ml of fresh medium per well. As a second approach, we also used frozen cell stocks of rat cortical neurons (AMS Biotechnology, Abingdon, UK). These were revived and put into culture according to the manufacturer's instructions with a plating density of 35,000 cells per well. All culturing procedures were similar to those described for the primary mouse neuronal cultures above.

Molecular biology

The pCAG:GPI-eGFP plasmid (Rhee et al., 2006) encoding a fusion protein of proacrosin signal peptide, enhanced GFP and Thy-1 GPI anchoring signal (GPI-eGFP) was purchased from Addgene (Plasmid # 32601; deposited by Anna-Katerina Hadjantonakis). The hVOS 2.0 plasmid was also purchased from Addgene (Plasmid # 45282; deposited by Meyer B. Jackson). All other reagents were purchased from Thermo Scientific (Dreieich, Germany), if not indicated otherwise. To minimize non neural expression, the open reading frame for GPI-eGFP was cloned to an AAV plasmid backbone (pAAV Synl :MCS) under the human synapsin-1 promoter (van Loo et al., 2015) with the Xbal and Hindlll restriction sites (primers: F Xbal GPI KOZ and RHindlll GPI). Virus preparation and transduction procedures. Recombinant adeno- associated virus production and preparation were carried out as previously described (Hauck et al., 2003; McClure et al., 2011). Briefly, plasmids for viral vector production were grown in Stbl2 bacteria while HEK293T cells (ATCC® CRL-3216TM) were transfected using the CaP04 method. Virus particles were harvested 4 or 5 days after transfection and subsequently purified with HiTrap Heparin HP columns (GE Healthcare Life Sciences, Chicago, IL, USA). After concentration and sterile filtration, viral proteins were separated using denaturizing polyacrylamide gel electrophoresis to detect major viral proteins VP1, VP2 and VP3 for quality control. A serial dilution of each virus stock was applied to determinate the appropriate virus titer. For transduction of cultured neurons, virus particles were diluted in sterile PBS, and were added directly to the bathing medium.

Confocal imaging

To demonstrate plasma membrane targeting of the GPI-eGFP fusion protein, images of transgene expressing neurons were taken at 40X magnification on a TCS SP5 confocal platform (Wetzlar, Germany) and on an Olympus 1X81 confocal microscope (Olympus Corporation, Tokyo, Japan). Cells were fixed with 4% PFA for ~3 min and then coated with Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA).

Combined optical and electrophvsiological recordings

Reagents for electrophysiological experiments were purchased from Sigma (St. Louis, MO, USA) unless indicated otherwise. Experiments were conducted using an Olympus BX61WI microscope (Olympus Corporation, Tokyo, Japan) equipped with epifluorescence and DIC. An electronmultiplying charge- coupled device (EM-CCD) camera (Evolve 512 Delta with LightSpeedTM, Photometries, Tucson, AZ, USA) was used to visualize neurons and to verify fluorescence. Whole-cell patch clamp recordings were amplified using a Multiclamp 700B amplifier (Molecular Devices, Sunnyvale, USA), low-pass filtered at 10 kHz and digitized at 50 kHz with a NI USB-6341 (National Instruments, Austin, TX) controlled by Strathclyde Electrophysiology Software WinWCP (John Dempster, University of Strathclyde, Glasgow, UK). Data were stored on a hard disk for offline analyses. Pipettes were pulled from borosilicate glass (King Precision Glass, Inc., Claremont, CA, USA) using a DMZ Zeitz puller (Zeitz-Instruments, Martinsried, Germany). Patch pipettes had resistances of 3-5 MW and contained (in mM): 135 K-Gluconate, 5 KC1, 10 HEPES, 0.1 ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid (EGTA), 1 MgC , 3 MgATP, 0.2 Na2ATP at pH 7.2. GPI-eGFP expressing cultured neurons 2-3 weeks after viral transduction were transferred to a modified submerged chamber (Hill & Greenfield, 2011) and perfused with HEPES-buffered ACSF (3 ml/min, at 32 ± 1°C, in mM): 145 NaCl, 5 KC1, 1.5 CaCF. 2 MgCF. 26 2- (4-(2-Hydroxyethyl)- 1- piperazinyl)-ethanesulfonic acid (HEPES), 10 Glucose at pH 7.4. Recordings were started after a ~ 5 min preincubation period with ACSF containing of the diazobenzene dye D3 (10 mM) in 0.1% DMSO. The EM-CCD camera was controlled by pManager software. Expression of GPI-eGFP was verified using epifluorescence. Excitation illumination (470 nm) at 3-5 mWcm-2 was provided with a custom made light source (parts from Thorlabs Inc, Newton, NJ, USA with a Luxeon Rebel 470 nm LED LXML-PB01-0040) that was driven by a custom made TTL switched stable current source. The excitation light as well as the collected fluorescence were filtered using a FITC filter set (Ex: HQ480/40x; Di: Q505LP; Em: HQ535/50m, Chroma Technology Corp., Bellow Falls, VT, USA). For optical voltage imaging two different settings of the EM-CCD camera were used. Frame rates of -1.08 kHz were achieved by measuring a 100 x 100 pixel region of interest (ROI) without binning. In order to gain more precise timing of the optically recorded peak of electrical events the inventors also applied 4 x 4 pixel binning in a 100 x 80 pixel ROI resulting in 25 x 20 pixel ROI and frame rates of -2.225 kHz. The timing of illumination and image acquisition were controlled by a digital stimulator (PG4000A, Cygnus Technology Inc., Delaware Water Gap, PA, USA) that was triggered by the acquisition software WinWCP. No image processing was applied. Imaging sequences were analyzed using ImageJ and Igor Pro (Wavemetrics Inc., Lake Oswego, OR, USA). Average grey values were extracted from the image sequences using the Z profiler-plugin for ImageJ. The precise timing of the images was obtained by digitally recording the "Exposure Out" TTL signal given off by the camera. This signal was then imported into Igor Pro and the rising and falling edges of the TTL pulses were detected. The midpoints between the detected edges constituted the precise time points for the image acquisition which was then lined up with the digitized electrophysiology traces. The optically recorded AP peak and the decay time constant of the AP were analyzed in recordings with 2.225 kHz sampling. Exponential fits of the electrophysiological and optical signals were made using IgorPro. The SNR (z-score equivalent) for APs was measured as the AF/F of the AP divided by the SD of the 200 ms prior baseline (Ghitani et ah, 2015) that contained no obvious spontaneous activity in the electrical recording. To analyse the ability of the hGEVI approach to faithfully report high frequency AP firing, trains of 16 or 21 high amplitude (800-1500 pA) 4 ms current pulses were applied at a frequency of 50 or 100 Hz in the I-clamp configuration. Prior to AP detection, fluorescent traces were detrended, smoothed using the Savitzky-Golay method by a 17 point (for 2.225 kHz sampling) and 7 point (for 1.08 kHz sampling) fourth order polynomial and aligned with the electrophysiological traces as described above. Detection threshold for optical APs was set as 75% peak amplitude of the first fluorescent AP in the train relative to the mean 180 ms prior baseline period. AP threshold in the electrophysiological (Vm) trace was set at 0 mV. In some of the cultured neurons APs and failures were not clearly distinguishable in the electrophysiological recording. Therefore, traces with events that exceeded -20 mV but did not reach 0 mV were omitted from analysis. To elicit spontaneous activity in a synaptically connected network of cultured neurons and to record from multiple neurons simultaneously, the K⁺-channel antagonist 4-aminopyridine (4-AP) was added to the ACSF. In some of the experiments simultaneous electrophysiological recordings from one of the imaged neuron was also performed.

Chirp function and phase determination

The 10-100 Hz chirp function was generated by the following equation: A x sin[2n(Fo+((F_max-Fo) x t/2T) x t)] where A is 50% of the peak-to-peak amplitude, Fo is the starting frequency (10 Hz), F_max is the frequency at the end of the pulse (100 Hz), t is time, and T is the duration of the chirp pulse (4 s). This pulse generated at a sampling interval of 1000 Hz was fed into the D/A converter, and subsequently low-pass filtered at 300 Hz before feeding it into the amplifier to circumvent step-like changes in Vm. For determination of the phases of the equivalently sampled and mean subtracted AF/F and Vm signals, the inventors first used the HilbertTransform function built into IgorPro. The phase was then determined by the value of the atari function of the point-by-point division between the HilbertTransform/signal.

Assessment of quencher effects on passive and AP properties

Non-fluorescent similarly aged cultured neurons as those used for combined optical and electrophysiological recordings were used for these experiments. The procedures for whole-cell I-clamp recordings were identical to those described for the electrophysiological recordings above. Data were analyzed using Igor Pro. For passive membrane properties, a 300 ms hyperpolarizing voltage pulse was elicited by current injection every 5 s for a 5 min baseline period followed by the bath perfusion of the vehicle (0.2% DMSO) or the quencher, i.e., 3 mM DPA (Biotium; Fremont, CA, USA) (in 0.03% DMSO), 10 or 20 pM D3 (in 0.1 and 0.2% DMSO, respectively) for 10 minutes. Membrane time constants were assessed by fitting an exponential to the hyperpolarization. APs were elicited by 200 ms long depolarizing current injections. AP threshold was determined from the first peak of the third derivative of the voltage signal (Henze & Buzsaki, 2001). AP width at half amplitude (FWHA) was determined as the time difference between the two points between the rising and decaying phases of the APs both at 50% of AP amplitude. Slice recordings

Wild-type C57BL/6N mice (aged 11-14 weeks) were anaesthetized with isoflurane and decapitated. Brains were quickly removed and transferred into ice-cold cutting solution containing: (in mM): 60 NaCl, 100 sucrose, 2.5 KC1, 1.25 NaH₂P0₄, 26 NaHCCfi, 1 CaCl₂, 5 MgCl₂, 20 Glucose (pH 7.3). 300 pm coronal slices were prepared with a vibratome (Leica VT1200S, Wetzlar, Germany) and gradually warmed to 37°C. For electrophysiology, slices were transferred into a submerged chamber (Hill & Greenfield, 2011) and superfused with ACSF containing (in mM): 125 NaCl, 3.5 KC1, 1.25 NaH₂P0₄, 26 NaHCCfi, 2 CaCl₂, 2 MgCl₂, 15 D-glucose (pH 7.3) and allowed to equilibrate for at least 20 min at room temperature. All solutions were bubbled with 95% 02-5% C0₂. Wholecell voltage clamp recordings were performed on visually identified cortical layer 2/3 pyramidal neurons. Patch pipettes (3-6 MW) were filled with (in mM): 135 Csmethansulfonate, 5 KC1, 10 HEPES, 0.16 EGTA, 2 MgCl₂, 3 NaCl, 4 Na₂-phosphocreatine, 2 MgATP, 0.2 NaGTP (pH adjusted to 7.3 with CsOH, osmolarity 290 mOsm/kg). Signals were low-pass filtered at 3 kHz and sampled at 10 kHz. Series resistance was monitored before and after the recording. Experiments with series resistances >20 MW or a change >20% during the recording were excluded. Spontaneous excitatory postsynaptic currents (sEPSCs) were recorded at a holding potential (Vh) of -60 mV. Spontaneous inhibitory postsynaptic currents (sIPSCs) were recorded at a Vh of 0 mV. sEPSCs and sIPSCs were recorded starting during a 2-3 min baseline period before bath application of 10 pM D3 in 0.1% DMSO and then continuously during D3 perfusion for >10 min. Events were analyzed during 1 min epochs collected during the baseline period and at least 10 min after bath application of D3. Data were analysed using a custom-written LabView software (EVAN) which provided peak amplitudes, 20-80% rise times, weighted decay time constants, and averaged traces.

Quantification and Statistical Analyses

Cohen’s d (difference between two means Ml and M2) was calculated as d = ABS(M1-M2)/SD, where SD is the pooled standard deviation given by SD = SQRT((SDl^A2+SD2^A2)/(Nl+N2-2)). Here SD1 and SD2 are the sample standard deviations of the two distributions, respectively and N1 and N2 are the two corresponding N’s. The cross-correlation coefficients between the AF/F and Vm were normalized by dividing the value by RMS(AF/F) x SQRT[N(AF/F)] x RMS(Vm) x SQRT[N(Vm)]. Statistical significance levels were set at p<0.05. For analysis of the fidelity of high frequency AP firing, the inventors performed a receiver operating characteristic (ROC) analysis. The diagnostic odds ratio (DOR) was calculated (Glas et ak, 2003) as (a/b)/(c/d), with a = true positives, b = false negatives, c = false positives and d = true negatives, with the respective values indicated in Fig. 6. The area under the curve (AUC) was calculated as:

with dx=0.01. All formulas for calculating the parameters of the ROC analysis are given in Fig. 6. All results are expressed as mean ± standard error of the mean (SEM) unless otherwise stated. Using variance estimates from the literature or from our own preliminary data, a power analysis was done to estimate the required sample size to obtain a power of 0.80, for a moderate effect size (0.25), and a significance level of 0.05. Statistical tests used for data comparisons and significance calculations are indicated for each comparison. Significance level was set at p<0.05.

Results

Fast high-fidelity reporting of AP firing and membrane hvperpolarizations

The inventors have recorded from mouse and rat cortical and hippocampal neurons in culture that were transduced with the GPI-eGFP construct carrying recombinant adeno-associated viruses (rAAVs). The neurons expressed the fluorophore in their membranes including their somata, dendrites and axons even as long as six weeks after viral transduction (Fig. lc). In the recordings, typically neurons after 2-3 weeks of viral transduction were used. Simultaneous whole-cell patch clamp recordings and optical recordings were done at 32 ± 1°C starting at ~ 5 min following addition of 10 mM D3. Several recordings persisted for > 60 min that allowed the inventors to determine the lasting presence of D3 in the membrane following its washout (see below). Most recorded neurons were subjected to a standardized current injection protocol consisting of eight 300 ms duration current injections (staircased as 3 hyperpolarizing and 5 depolarizing pulses, usually ranging between -200 to +300 pA, in steps set to provide -30 to -40 mV initial hyperpolarizations from the resting membrane potential (RMP) of -60 to -68 mV and AP firing upon depolarizations). The result of such an experiment is shown in Fig. 2 with the optical recording sampled at 1.08 kHz and the electrical recording sampled at 50 kHz. Note that in addition to the accurate detection of APs (Fig. 2a, d-h), small hyperpolarizations were also truthfully detected by the new hGEVI (Fig. 2b-d, f). In this particular cell, the signal -to-noise ratio (SNR, z-score equivalent) (21) of the first AP in the train was 31.36. The mean (± SEM) AF/F of the 27 action potentials in this recording was 5.01 ± 0.05%. Cohen’s d statistic for the same 27 APs was 288.3 when the SD of a 200 ms baseline was considered for the analysis. However, since the pooled SD (see Materials and Methods) is reduced by the large number of points present in a long baseline, the inventors also calculated the d values using 27 baseline points to match that of the APs. In this case, the d values ranged between 103.0 and 112.3 for 12 randomly selected baselines, an outstanding statistical effect size (30). Remarkably, in a total of 24 neurons the average SNR (z-score equivalent), calculated from unsmoothed fluorescence traces of the first APs evoked by the injected current, was 17.95 ± 1.12 (mean ± SEM; n=24). The standardized current protocol experiments also allowed the inventors to correlate the membrane potential (Vm) changes measured by electrophysiology with the fluorescence changes (AF/F) as measured by the mean grey levels in the ROIs (usually the somatic membrane). For these measurements the electrophysiology traces were down sampled to match the sampling of the fluorescence (-1.08 or 2.225 kHz) (Fig. 3a&b). In this manner, the two traces could be plotted against each other to obtain the graph of AF/F as a function of the membrane potential through a point-by-point correspondence. The slope of the line fitted to the -3,000-6,000 point pairs in each cell yielded the AF/F for a Vm change of 100 mV (Fig. 3c). In n=10 cells, under matching experimental conditions, except for the sampling rate of the fluorescence, the average AF/F for a 100 mV change was 6.604 ± 0.421% (± SEM). In all cases, the responses were consistently linear in the Vm range of -100 mV to ±40 mV.

Fluorescence signals using D3 hGEVI detect rapid changes in Vm

The relatively low maximal sampling rate of the fluorescent signals with a EM-CCD camera (-2.2 kHz) precluded accurate measurements of the rising phase of APs or fitting exponential functions to the rising phases of square-shaped voltage pulses. Therefore, the inventors decided to compare fluorescent and electrical measurements of fast changes in Vm by measuring the decay time constants (t) of APs where more sample points were available in the fluorescence traces. The inventors divided the APs into two groups: the rapidly decaying APs, usually those that were elicited first in a train of spikes, and slowly decaying APs, i.e., those that were the last in the train when due to various biophysical factors the recovery of the membrane potential following an AP is hindered. In these experiments the APs recorded by fluorescence (sampled at -2.2 kHz) or through electrophysiology (sampled at 50 kHz) were normalized to their peaks and single exponentials were fitted to the decay phases (Fig. 4a&b). Comparison of the fast AP decays shown nearly one-to-one correspondence between the two measurements. In 6 cells a total of 27 fast APs (t range: 1-3 ms; Fig. 4c) and 20 slow APs (t range: 3- 22 ms; Fig. 4d) were compared. The linear fit to the fast AP data has a slope=l 0334± 0.0366 and R2=0.746 (Fig. 4c) while the linear fit to the slower decaying APs has a slope=1.0495±0.026 and R2=0.939 (Fig. 4d). The high correlation between the rs indicates the accurate and fast rendition of membrane potential changes by the D3-GPI-eGFP hGEVF Recordings of subthreshold changes in Vm

The inventors wanted to know how accurately the new hGEVI approach reflects subthreshold changes in Vm. Therefore, the inventors created a stimulus protocol that systematically and reproducibly altered the Vm in voltage-clamped cultured neurons expressing GPI-eGFP. The protocol changed the Vm according to a chirp function (see Materials and Methods) that increased from 10 Hz to 100 Hz over 4 s, and had a peak-to-peak amplitude of 40 mV (± 20 mV from the RMP). The inventors then carried out extensive analyses of the correlations between the Vm (down-sampled to the sampling rate of the fluorescence) and the AF/F over the 4 s of the chirp pulse of increasing frequency. Fig. 5a shows such an experiment together with the Morlet wavelet transforms of the two signals (AF/F and Vm). Using the Hilbert transform (see Materials and Methods) the inventors also calculated the phases of the two responses (Fig. 5b&c) during the duration of the 4 s chirp wave. For each experiment the inventors plotted the point-by-point correlation between the phase of the AF/F and the phase of the Vm signals. These plots were binned in 3D histograms (Fig. 5d), and Pearson’s R value was calculated, together with its significance based on the t-distribution of R/SQRT[(l-R^A2)/(N-2)], where N is the number of point pairs. Furthermore, the inventors calculated the cross-correlation between the phase of AF/F and that of Vm (Tab. 1 and 2). The values of the cross correlations were normalized using the RMS values of each signal (see Materials and Methods). Finally, for each experiment the inventors subtracted the phase of the Vm signal from the AF/F phase in a point-by point manner. The values of the subtracted points were binned at 0.1 rad, and a histogram was generated for each experiment. The inventors then fitted a Gaussian to the histogram (Fig. 5f) that provided the mean difference between the two phases (in rad). A negative value of the difference indicates that the phase of the AF/F lags behind that of the Vm signal. The value of the phase difference expressed in rad was converted to ms using the average frequency over the entire duration of the chirp function (55 Hz, i.e., 2p rad = 18.182 ms). From 8 chirp sweeps recorded in 6 different neurons at their RMP of -68 to -72 mV, the mean (± SEM) values were as follows: R (AF/F phase, Vm phase correlation) 0.476 ± 0.056; t value 40.34 ± 6.90; Df (range) 4330-8872; p (range) 0.00 - 1.92E-67; Normalized R (AF/F, Vm cross-correlation) 0.596 ± 0.063; Df (range) 4330-8872; p (range) 0.00 - 5.27E-122; average (AF/F phase - Vm phase) in rad: -0.0097 ± 0.0084, in ps: -276 ± 240. Taken together, the measurements indicate that the novel hGEVI approach highly accurately reflects slow (10 Hz) and fast (100 Hz) subthreshold changes in Vm.

Receiver operating characteristic (ROC) analyses of induced AP firing at 50 Hz and 100 Hz During 300 ms long depolarizing current pulse injections, cultured cortical or hippocampal neurons under the inventors’ recording conditions did not fire at frequencies of >30 Hz (e.g., Fig. 2e). As previously reported (9) for other GEVI approaches, at these relatively low frequencies, there was a 1-to-l correspondence between the APs detected in the optical voltage traces and the APs recorded by the patch- clamp method. However, the inventors wanted to know whether the new hGEVI method is also capable of detecting APs elicited at higher frequencies. The inventors injected short (4 ms) high-amplitude (800-1500 pA) current pulses to elicit APs at 50 Hz (Fig. 6a&b) and at 100 Hz (Fig. 6c&d) in a highly controlled manner. For these experiments, the AF/F traces were smoothed according to the Savitzky-Golay method by a 17 point (for 2.225 kHz sampling) and 7 point (for 1.08 kHz sampling) fourth-order polynomial. The threshold for AP detection in the smoothed AF/F traces was set at 75% of the peak amplitude of the first fluorescent AP. Point-to-point correlations between smoothed AF/F traces and the Vm traces down- sampled to match the fluorescence sampling rates were >0.9 (Fig. 6b&d). In 6 neurons each, the inventors performed a binary ROC analysis of several sweeps with a total 579 APs and 102 failures elicited at 50 Hz and 396 APs and 296 failures elicited at 100 Hz (Table 5). Such an analysis, primarily used for determining the reliability of a diagnostic test, reveals the accuracy of the applied detection method. However, in contrast to a diagnostic test, where the precise prevalence of the disease in the population is rarely known, in the inventors’ experiments the Vm recordings provided the true rates of APs and failures elicited by the current pulses, thereby making this a very powerful analytical tool. It should also be noted that the failures are not simply absences of action potentials but represent quite large subthreshold depolarizations elicited by the short current pulses, thus making the distinction between APs and failures more difficult. The ROC analysis indicates very high levels of sensitivity (50 Hz: 98.8%; 100 Hz: 97.7%) and specificity (50 Hz: 98.0%; 100 Hz: 92.9%) for discriminating between APs and failures at these two frequencies. A valuable statistic is the diagnostic odds ratio (DOR) of the test (31) that represents the ratio of the odds of AF/F positivity when APs are present in the Vm trace relative to the odds of AF/F positivity when there are AP failures in the Vm. The DOR values were >4000 (50 Hz) and >500 (100 Hz) while the calculated values for the area under the curve (AUC) were 0.999 (50 Hz) and 0.993 (100 Hz). Such large values of DOR and AUC are indicative of a test of extremely high diagnostic value (31). In addition to the ROC analysis that does not account for consecutive APs, the inventors also calculated the rates of detection of 2 or more successive APs at 50 Hz and 100 Hz. Of the total of 579 APs elicited at 50 Hz stimulation, 561 were part of bursts of > 2 APs. Of these, 554 (98.75%) were also detected optically. Of the 396 APs elicited by 100 Hz stimulation, 197 occurred in bursts of > 2 APs. Of these, 193 (97.97%) were also detected optically. As previous studies did not make an effort to induce high frequency APs but mainly relied on the intrinsic firing rates of the recorded neurons (<40 Hz), the sizeable detection rates by the new hGEVI approach are encouraging for its usefulness to detect APs from rapidly firing cell types such as interneurons. Accordingly, the new hGEVI method is sufficiently sensitive to detect small Vm deflections, yet its dynamic range is adequately large to enable a very simple threshold detection to differentiate between APs and failures. Table 5

ROC analysis of the indicated number of cells and traces with 579 APs and 102 failures elicited at 50 Hz and 396 APs and 296 failures elicited at 100 Hz.

Simultaneous recordings from two neurons

GEVI offer the possibility to record simultaneously from a large number of neurons thus allowing the monitoring of both subthreshold activities in some cells and AP firing in others. The inventors induced synchronous activity in neuronal cultures using the K⁺ channel blocker 4-amino-pyridine (4-AP; 50 mM), a compound known for its epileptiform activity inducing properties (32). Usually the inventors recorded from two neurons within the same field of view in the culture (Fig. 7a). One of the two cells also underwent whole-cell recording the I-clamp configuration. Recordings from such an experiment are shown in Fig. 7b- d. Both electrophysiological and optical recordings of the membrane voltage indicated the presence of subthreshold and suprathreshold activities (Fig. 7b-d). Similar recordings were obtained in 5 other cell pairs in different cultures, indicating that simultaneous optical recordings of membrane voltage using the D3 hGEVI method will be a valuable tool to detect neuronal synchrony and the temporal activation in a synaptically interconnected network.

Recordings after washout of D3

The GEVI that uses a single molecule for voltage sensing or the FRET between two fluorescent proteins differ from the hGEVI approach in that the latter uses a small molecule that first has to be added to the extracellular compartment to eventually partition itself into the membrane. An open question remains whether the small quenching molecule needs to be continuously present in the extracellular space, or whether it is sufficient to load the membrane only once. This question has not be addressed in previous voltage sensing experiments with DPA, and therefore it is not known if DPA can reside long enough in the cell membrane to allow the optical recordings to persist following its wash-out from the environment. The inventors have addressed this question with the D3-GPI-eGFP hGEVI approach. Cells were exposed to 10 mM D3 for 10 min. Subsequently, the cells were perfused with a solution containing no D3 (and no DMSO) at a flow rate (3 ml/min) that exchanged the recording chamber volume several times a minute. Both optical and electrophysiological recordings were undertaken immediately after wash-out was started and were continued for as long as 70 min. Fig. 8 illustrates such an experiment where the standard current clamp approach was applied to a cell every 10 min for 60 min after the start of the D3 wash-out. Presumably due to the high lipophilicity of D3, the optical recordings had a slow run down in the prolonged absence of D3 from the extracellular space. The SNR (z-score equivalent) for the first AP in the train remained constant: 13.83 at 0 min, and 12.18 after 60 min (Fig 8). In a total of 6 neurons the average (± SEM) SNR (z-score equivalent) of the first AP in the train was 15.4 ± 3.74 after 30 min and 12.21 ± 3.36 after 60 min of D3 washout. Considering the overall average (± SEM) value of AP SNR in the presence of 10 pM D3 of 17.95 ± 1.12 (n=24), the slight loss of SNR upon D3 washout appears to be linear at a rate of 0.095 min-1. Therefore, once loaded in the membrane, D3 does not need to be continuously present in the extracellular space to yield a decent SNR for AP detection even 60 min after its washout. A single membrane loading for a few minutes would provide ample amount of time for subsequent recordings from the GPI-eGFP expressing neurons. This may be highly relevant for the future potential in vivo use of the D3 -GPI-eGFP hGEVI approach where continuous administration of the small quenching molecule may be impractical. Comparison of DPA and D3 on passive membrane properties and AP firing

One of the major concerns with the hGEVI approach is that the small molecule quencher accumulates in the membrane sufficiently to cause considerable changes in capacitance and impede AP firing. The exact concentration threshold for DPA to cause such changes has not been adequately investigated but reports exist that at 5 mM DPA has deleterious effects on evoked responses in hippocampal slices (12). However, no such effects have been reported after incubation of slices with 4 pM DPA when it was used in conjunction with a new membrane targeting approach for fluorescent proteins (21). First we established that there was no difference between the AP width at half amplitude between the cells expressing GPI- eGFP (mean ± SEM: 1.63 ± 0.26 ms, n = 16) and those that did not express the fluorescent protein (1.32 ± 0.07 ms, n = 36; p = 0.4878, Mann-Whitney U-test). Next the inventors compared the effects of DPA and D3 on passive membrane properties and AP firing of cultured neurons without the expression of eGFP. The inventors started by measuring the effects of 20 mM D3 (in 0.2% DMSO), a concentration 2-fold higher than the inventors normally used for optical measurements, on whole-cell capacitance, input resistance, AP width at half amplitude, and AP threshold. None of these parameters were affected by 20 mM D3 (Fig. 9). In another series of experiments the inventors systematically compared the effects of 0.2 % DMSO, 2.5, 3, or 5 mM DPA (dissolved in 0.025, 0.03, 0.05 % DMSO, respectively), and 10 or 20 mM D3 (dissolved in 0.1, and 0.2% DMSO, respectively) on the same membrane parameters. The starting values for each of the properties were not different between the cells. In contrast to D3 and DMSO, DPA significantly increased membrane capacitance and AP width at half maximal amplitude (Fig. lOa&b). The values prior to perfusion were compared to those measured at 5 or 10 min after the perfusion of the compounds. The concentrations of DPA used was commensurate with that customarily employed in hybrid voltage sensing (12, 15, 21). In addition to increasing the membrane capacitance, 3 mM DPA had a toxic effect on the cells, as gradually less and less neurons survived for the entire duration of the 10 min perfusion (Fig. 10c). In a small number of cells (n=3) that expressed GPI-eGPF and lasted sufficiently long in 3 mM DPA the inventors wanted to compare the optical recordings to those they obtained with D3. Unfortunately, in none of the recorded neurons did the inventors obtain any fluorescent signals that correspond to evoked APs. It is possible that D3 and DPA have different energy requirements for voltage-dependent movements or changes in orientation in the membrane, and therefore, do not have similar charging effects. In the case of D3 such energy requirements may be sufficiently low not to perturb the passive membrane properties.

Lack of D3 effects on synaptic responses in slices

Another considerable concern with the use of the two-component voltage measurement approach with DPA is its effect on ligand-gated ion channels. DPA and other hydrophobic anions have been reported to antagonize GABAA (25, 26) and NMDA receptors (27). Therefore, the inventors wanted to test the effects of 10 mM D3 on excitatory and inhibitory synaptic responses recorded in cortical slices. The inventors recorded sEPSCs (at Vh = -60 mV) and sIPSCs (at Vh = 0 mV) in L2/3 pyramidal cells of mouse cortical slices at 32 ± 1°C. In 8 neurons values are given as follows: (mean ± SEM before D3 perfusion; at the end of a 10 min 10 mM D3 perfusion; the respective p-values obtained by a Wilcoxon matched-pairs signed rank test) the frequencies of sEPSCs (6.55 ± 2.34 Hz; 5.90 ± 1.46 Hz; p = 0.8438) and sIPSCs (8.83 ± 1.91 Hz; 9.33 ± 1.61; p = 0.5469), 20-80% rise times of sEPSCs (1.29 ± 0.38 ms; 1.44 ± 0.43 ms; p = 0.4609) and sIPSCs (1.53 ± 0.38 ms; 1.32 ± 0.28 ms; p = 0.8438), weighted decay time constants for sEPSCs (6.93 ± 1.07 ms; 8.22 ± 1.12 ms; p = 0.1484) and sIPSCs (11.25 ± 0.65 ms; 10.64 ± 0.51 ms; p = 0.6406), and peak amplitudes of sEPSCs (13.47 ± 3.16 pA; 11.46 ± 2.09 pA; p = 0.3125) and sIPSCs (19.14 ± 1.07 pA; 20.96 ± 1.65 pA; p = 0.3828) were all unchanged by D3.

Selection of quenchers

To identify quenchers suitable for hybrid genetically encodable voltage indicators, the inventors systematically screened a plurality of available quenchers. The screening included determination of the absorption spectra (Fig. 19) of the quenchers listed in Table 6 to identify quenchers showing a high spectral overlap with a specific fluorophore. Table 6

References

I. Hamel EJ, Grewe BF, Parker JG, & Schnitzer MJ (2015) Cellular level brain imaging in behaving mammals: an engineering approach. Neuron 86(1): 140-159. 2. Knopfel T (2012) Genetically encoded optical indicators for the analysis of neuronal circuits. Nature reviews. Neuroscience 13(10):687-700.

3. Antic SD, Empson RM, & Knopfel T (2016) Voltage imaging to understand connections and functions of neuronal circuits. J Neurophysiol 116(1): 135-152.

4. Inagaki S & Nagai T (2016) Current progress in genetically encoded voltage indicators for neural activity recording. Curr Opin Chem Biol 33:95-100.

5. Knopfel T, Gallero-Salas Y, & Song C (2015) Genetically encoded voltage indicators for large scale cortical imaging come of age. Curr Opin Chem Biol 27:75-83.

6. Lin MZ & Schnitzer MJ (2016) Genetically encoded indicators of neuronal activity. Nat Neurosci 19(9): 1142-1153. 7. Storace D, et al. (2016) Toward Better Genetically Encoded Sensors of Membrane Potential. Trends

Neurosci 39(5):277-289.

8. St-Pierre F, Chavarha M, & Lin MZ (2015) Designs and sensing mechanisms of genetically encoded fluorescent voltage indicators. Curr Opin Chem Biol 27:31-38.

9. Gong Y, et al. (2015) High-speed recording of neural spikes in awake mice and flies with a fluorescent voltage sensor. Science 350(6266): 1361-1366.

10. Gonzalez JE & Tsien RY (1995) Voltage sensing by fluorescence resonance energy transfer in single cells. Biophys J 69(4): 1272-1280.

I I. Gonzalez JE & Tsien RY (1997) Improved indicators of cell membrane potential that use fluorescence resonance energy transfer. Chem Biol 4(4):269-277. 12. Chanda B, et al. (2005) A hybrid approach to measuring electrical activity in genetically specified neurons. Nat.Neurosci.

13. Benz R, Lauger P, & Janko K (1976) Transport kinetics of hydrophobic ions in lipid bilayer membranes. Charge-pulse relaxation studies. Biochimica et biophysica acta 455(3):701-720. 14. Fernandez JM, Taylor RE, & Bezanilla F (1983) Induced capacitance in the squid giant axon. Lipophilic ion displacement currents. J Gen Physiol 82(3):331-346.

15. Bradley J, Luo R, Otis TS, & DiGregorio DA (2009) Submillisecond optical reporting of membrane potential in situ using a neuronal tracer dye. J Neurosci 29(29): 9197-9209.

16. Fink AE, Bender KJ, Trussell LO, Otis TS, & DiGregorio DA (2012) Two-photon compatibility and single-voxel, single-trial detection of subthreshold neuronal activity by a two-component optical voltage sensor. PloS one 7(8):e41434.

17. Shafeghat N, Heidarinejad M, Murata N, Nakamura H, & Inoue T (2016) Optical detection of neuron connectivity by random access two-photon microscopy. J Neurosci Methods 263:48- 56.

18. DiFranco M, Capote J, Quinonez M, & Vergara JL (2007) Voltage-dependent dynamic FRET signals from the transverse tubules in mammalian skeletal muscle fibers. J Gen Physiol 130(6):581-600.

19. Wang D, Zhang Z, Chanda B, & Jackson MB (2010) Improved probes for hybrid voltage sensor imaging. Biophys.J. 99(7):2355-2365.

20. Sjulson L & Miesenbock G (2008) Rational optimization and imaging in vivo of a genetically encoded optical voltage reporter. J Neurosci 28(21):5582-5593.

21. Ghitani N, Bayguinov PO, Ma Y, & Jackson MB (2015) Single-trial imaging of spikes and synaptic potentials in single neurons in brain slices with genetically encoded hybrid voltage sensor. J Neurophysiol 113(4): 1249-1259.

22. Wang D, McMahon S, Zhang Z, & Jackson MB (2012) Hybrid voltage sensor imaging of electrical activity from neurons in hippocampal slices from transgenic mice. J Neurophysiol 108(11):3147-3160.

23. Weigel S, Flisikowska T, Schnieke A, & Luksch H (2014) Hybrid voltage sensor imaging of eGFP-F expressing neurons in chicken midbrain slices. J Neurosci Methods 233:28-33.

24. Oberhauser AF & Fernandez JM (1995) Hydrophobic ions amplify the capacitive currents used to measure exocytotic fusion. Biophys J 69(2):451-459.

25. Chisari M, Wu K, Zorumski CF, & Mennerick S (2011) Hydrophobic anions potently and uncompetitively antagonize GABA(A) receptor function in the absence of a conventional binding site. Br J Pharmacol 164(2b):667-680.

26. Limon A, Estrada-Mondragon A, Ruiz JM, & Miledi R (2016) Dipicrylamine Modulates GABArhol Receptors through Interactions with Residues in the TM4 and Cys-Loop Domains. Mol Pharmacol 89(4):446-456.

27. Linsenbardt AJ, et al. (2013) Noncompetitive, voltage-dependent NMDA receptor antagonism by hydrophobic anions. Mol Pharmacol 83(2):354-366.

28. Ikezawa H (2002) Glycosylphosphatidylinositol (GPI)-anchored proteins. Biol Pharm Bull 25(4):409- 417.

29. Brinks D, Klein AJ, & Cohen AE (2015) Two-Photon Lifetime Imaging of Voltage Indicating Proteins as a Probe of Absolute Membrane Voltage. Biophys J 109(5):914-921.

30. Sawilowsky S (2009) New effect size rules of thumb. J. Modem Appl. Stat. Meth. 8:467-474.

31. Glas AS, Lijmer JG, Prins MH, Bonsel GJ, & Bossuyt PM (2003) The diagnostic odds ratio: a single indicator of test performance. J Clin Epidemiol 56(11): 1129-1135.

32. Avoli M & Williamson A (1996) Functional and pharmacological properties of human neocortical neurons maintained in vitro. Prog.Neurobiol. 48(6):519-554.

33. Shaner NC, Steinbach PA, & Tsien RY (2005) A guide to choosing fluorescent proteins. Nature methods 2(12):905-909.

34. Latt SA, Cheung HT, & Blout ER (1965) Energy Transfer. A System with Relatively Fixed Donor- Acceptor Separation. Journal of the American Chemical Society 87:995-1003.

35. McClure C, Cole LH, Wulff P, Klugmann M, Murray AJ (2011) Production and Titering of Recombinant Adeno-Associated Viral Vectors. J Vis Exp. (57), e3348

36. Hauch B et al., Chen L and Xiao W (2003) Generation and Characterization of Chimeric Recombinant AAV Vectors. Mol Ther 7(3): 419-425

37. van Loo KMJ et al., (2015) Zinc Regulates a Key Transcriptional Pathway for Epileptogenesis via Metal-Regulatory Transcription Factor 1. Nat Commun. 6:8688

38. Cao et al., (2016) Pharmacological reversal of a pain phenotype in iPSC-derived sensory neurons and patients with inherited erythromelalgia. Sci Transl Med. 20;8(335)

39. Sakaguchi R, Leiwe MN, Imai T (2018) Bright multicolor labeling of neuronal circuits with fluorescent proteins and chemical tags. eLife 7:e40350

40. Goparaju SK et al., (2017) Rapid differentiation of human pluripotent stem cells into functional neurons by mRNAs encoding transcription factors. Scientific Reports 7; 42367

41. Abdelfattah AS, Farhi SL, Zhao Y, Brinks D, Zou P, Ruangkittisakul A, et al. (2016) A bright and fast red fluorescent protein voltage indicator that reports neuronal activity in organotypic brain slices. J Neurosci ;36:2458-72.

42. Kannan M, Vasan G, Huang C, Haziza S, Li JZ, Inan H, et al. (2018) Fast, in vivo voltage imaging using a red fluorescent indicator. Nat Methods.15: 1108-16.

43. Piatkevich KD, Jung EE, Straub C, Linghu C, Park D, Suk HJ, et al. (2018) A robotic multidimensional directed evolution approach applied to fluorescent voltage reporters. Nat Chem Biol.14:352-60.

44. Yang HH, St-Pierre F, Sun X, Ding X, Lin MZ, Clandinin TR. (2016) Subcellular imaging of voltage and calcium signals reveals neural processing in vivo. Cell.166:245-57.

45. Chamberland S, Yang HH, Pan MM, Evans SW, Guan S, Chavarha M, et al. (2017) Fast two-photon imaging of subcellular voltage dynamics in neuronal tissue with genetically encoded indicators. Elife. 6:e25690.

46. St-Pierre F, Marshall JD, Yang Y, Gong Y, Schnitzer MJ, Lin MZ. (2014) High-fidelity optical reporting of neuronal electrical activity with an ultrafast fluorescent voltage sensor. Nat Neurosci.17:884- 9

47. Hochbaum DR, Zhao Y, Farhi SL, Klapoetke N, Werley CA, Kapoor V, Zou P, Kralj JM, Maclaurin D, Smedemark-Margulies N, Saulnier JL, Boulting GL, Straub C, Cho YK, Melkonian M, Wong GK, Harrison DJ, Murthy VN, Sabatini BL, Boyden ES, Campbell RE, Cohen AE. (2014) All-optical electrophysiology in mammalian neurons using engineered microbial rhodopsins. Nat Methods.11(8): 825- 33.

48. Adam Y, Kim J, Lou S, Yongxin Z, Brinks D, Wu H, et al. (2018) All-optical electrophysiology reveals brain-state dependent changes in hippocampal subthreshold dynamics and excitability. bioRxiv..

49. Rhee MR, Pirity MK, Lackan CS, Long JZ, Kondoh G, Takeda J, Hadjantonakis A (2006) In Vivo Imaging and Differential Localization of Lipid-Modified GFP-variant Fusions in Embryonic Stem Cells and Mice. Genesis. 44(4):202-18.

50. Hill MRH, Greenfield SA (2011) The Membrane Chamber: A New Type of in Vitro Recording Chamber. J Neurosci Methods: 195(1): 15-23

51. Ghitani N, Bayguinov PO, Ma Y, Meyer B Jackson, Single-trial Imaging of Spikes and Synaptic Potentials in Single Neurons in Brain Slices With Genetically Encoded Hybrid Voltage Sensor

52. Henze DA, Buzsaki G (2001) Action Potential Threshold of Hippocampal Pyramidal Cells in Vivo Is Increased by Recent Spiking Activity. Neuroscience 105(1): 121-30

53. Glas AS, Lijmer JG, Prins MH, Bonsel GJ, Bossuyt PMM (2003) The Diagnostic Odds Ratio: A Single Indicator of Test Performance. J Clin Epidemiol 56(11): 1129-35 54. Zhang, Y., Pak, C., Han, Y., Ahlenius, H., Zhang, Z., Chanda, S., ... Su, T. C. (2013) Rapid Single-

Step Induction of Functional Neurons from Human Pluripotent Stem Cells, Neuron 78(5): 785-798.

Claims

Claims

1. An optical voltage sensor comprising a) a fluorophore, in particular a fluorescent protein, linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; and b) a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%.

2. The optical voltage sensor according to claim 1, wherein the spectral overlap corresponds to a normalized spectral overlap, wherein the height of the absorption maximum of the quencher is normalized to the height of the emission maximum of the fluorophore.

3. The optical voltage sensor according to claim 1 or 2, wherein the quencher is characterized by a) a peak absorption value of at least 0.06, at least 0.08, at least 0.10, at least 0.12, at least 0.15 or at least 0.17; and/or b) a peak extinction coefficient of at least 10000 cm ¹ M ¹, at least 12500 cm ¹ M ¹, at least 15000 cm ¹ M ¹ or at least 17500 cm ¹ M ¹.

4. The optical voltage sensor according to any of claims 1-3, wherein the quencher is an aryl azo compound characterized by a molecular weight of 200-450 g/mol, particularly 250-430 g/mol, more particularly 310-410 g/mol.

5. The optical voltage sensor according to any of claims 1-4, wherein the fluorophore a. has an emission spectrum characterized by an absolute emission maximum of 440-550 nm, particularly 480-530 nm, more particularly 495-515 nm; and/or b. is or comprises a green fluorescent protein, particularly selected from the group consisting of GFP, enhanced GFP (eGFP), Emerald, Superfblder GFP, Azami Green, mWasabi, TagGFP, Turbo GFP, AcGFP, ZsGreen and T-Sapphire, more particularly eGFP.

6 The optical voltage sensor according to claim 5, wherein the fluorophore is an eGFP comprising or consisting of a. an amino acid sequence of SEQ ID NO: 1, wherein Xi is selected from K and Q and preferably is Q and wherein amino acids 242 to 249 independently may be present or absent, preferably absent, or b. an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or at least 99% identity to SEQ ID NO: 1, wherein Xi is present and selected from K and Q, preferably Q, wherein the eGFP emits fluorescence upon excitation, preferably with an emission maximum between 440-550 nm, 480-530 nm, 495-515 nm or 505-512 nm.

7. The optical voltage sensor according to any of claims 1-6, wherein the quencher is characterized by an absolute absorbance maximum of 440-550 nm, in particular 480-550 nm.

8. The optical voltage sensor according to any of claims 1-7, in particular claims 5-7, wherein the quencher is characterized by formula 1

(1), wherein

R¹ is selected from NO2, NR and N=NR⁴, wherein R³ is a C₁-C₃ alkyl, in particular CEE,

R⁴ is a substituted or non-substituted aryl moiety, in particular a substituted or non-substituted phenyl or naphthyl moiety, and

R² is selected from H, OH, NR⁵R⁶, and (CH₂)_niC(=0)0R⁷ wherein R⁵ is selected from H and a C₁-C₅ alkyl,

R⁶ is selected from the group consisting of H and a substituted or non-substituted aryl moiety, in particular a phenyl moiety, and (CH2)_n2R⁸, wherein n2 is 1-3, in particular 2, and

R⁸ is CN, OH or OC(=0)R⁹, wherein

R⁹ is a C₁-C₅ alkyl, in particular R⁹ is C(=CH₂)CH₃; R⁷ is a succinimidyl moiety and nl is 0-5,

is an aryl moiety, in particular a phenyl moiety; and X¹, X², X³ and X⁴ are independently selected from H, F, Cl, Br and I.

9. The optical voltage sensor according to claim 8, wherein the quencher is selected from the group consisting of Disperse Orange 37, Disperse Orange 13, Disperse Orange 3, DABCYL SE, Disperse Orange 25, Disperse Orange 1, Disperse Red 1, Oil Red O, and Disperse Red 13, in particular Disperse Orange 3, DABCYL SE, Disperse Orange 25, Disperse Orange 1, Disperse Red 1, Oil Red O, and Disperse Red 13.

10. The optical voltage sensor according to any of claims 1-5, wherein the fluorophore is tdTomato and the quencher is Disperse Blue 124.

11. The optical voltage sensor according any of claims 1-5, wherein the fluorophore is anchored to the outer leaflet of the plasma membrane, wherein preferably the anchoring moiety is a glycosylphosphatidylinositol (GPI) moiety.

12. A method to detect a change of transmembrane potential across a plasma membrane of a cell, particularly a neuronal cell, more particularly a neuron, comprising the steps of a) effecting within the cell expression of a fluorophore, in particular a fluorescent protein, linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; b) contacting the cell with a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; c) illuminating the cell with light suitable for excitation of the fluorophore; d) detecting a fluorescence intensity emitted from the fluorophore; and e) detecting a change of transmembrane potential, wherein a change in fluorescence intensity corresponds to a change of transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%.

13. A fluorescent protein comprising or consisting of a. an amino acid sequence of SEQ ID NO: 1, wherein Xi is selected from K and Q, preferably Q and wherein amino acids 242 to 249 independently may be present or absent, preferably absent, or b. an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or at least 99% identity to SEQ ID NO: 1, wherein

Xi is present and selected from K and Q, preferably Q, wherein the fluorescent protein emits fluorescence upon excitation, preferably with an emission maximum between 440-550 nm, 480-530 nm, 495-515 nm or 505-512 nm.

14. A nucleic acid encoding the fluorescent protein of claim 13 or a vector comprising said nucleic acid.

15. A kit of parts comprising a) a first reagent comprising a fluorophore, preferably a fluorescent protein, more preferably the fluorescent protein of claim 13, or a nucleic acid encoding said fluorescent protein, or a vector comprising said nucleic acid, or a cell comprising said fluorophore, said nucleic acid or said vector, wherein the fluorophore is linked to an anchoring moiety for anchoring of the fluorophore to a plasma membrane of a cell; and b) a second reagent comprising a lipophilic small molecule quencher, capable of redistribution within the plasma membrane in response to a change in transmembrane potential; wherein the absolute maximum of the emission spectrum of the fluorophore and the absolute maximum of the absorption spectrum of the quencher exhibit a spectral overlap of at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%.