Sensors, methods and kits for detecting NADPH based on resonance energy transfer

Abstract

The invention relates to the detection of the cofactor reduced nicotinamide adenine dinucleotide phosphate (NADPH). Provided is a sensor molecule for the resonance energy transfer (RET)-based detection of NADPH, the sensor comprising a segment A connected via a linker to a segment B, wherein each of segment A and segment B comprises a member of a RET pair comprising a donor moiety and an acceptor moiety, further characterized in that (i) segment A comprises a binding protein (BP) for NADPH, the BP being dihydrofolate reductase (DHFR; EC 1.5.1.3) or a functional homolog, fragment, derivative or variant thereof, showing the desired NADPH binding properties, and wherein the BP comprises a heterologous protein domain inserted at or replacing at least part of the region corresponding to positions (20) to (27) of E. coli DHFR, said heterologous protein domain comprising the member of the RET pair; (ii) segment B comprises a ligand (L) capable of intramolecular binding to said BP only in the presence of NADPH; such that the donor moiety and the acceptor moiety are in a suitable juxtaposition to yield a RET signal when L is bound to BP, and wherein NADPH-induced binding of L to BP results in an increase in RET efficiency.

Claims

1. A sensor molecule for the bioluminescence resonance energy transfer (BRET)-based detection of reduced nicotinamide adenine dinucleotide phosphate (NADPH), the sensor comprising a segment A connected via a linker to a segment B, wherein each of segment A and segment B comprises a member of a BRET pair comprising a donor moiety and an acceptor moiety, further characterized in that (i) segment A comprises a binding protein (BP) for NADPH, the BP being dihydrofolate reductase (DHFR; EC 1.5.1.3) of E. coli (UniprotKB accession number POABQ4) having the amino acid sequence of SEQ ID NO:2 or a variant thereof having at least 95% sequence identity to SEQ ID NO:2, or a functional homolog thereof selected from SEQ ID NOS:3-6 and variants thereof having at least 95% sequence identity to any of SEQ ID NOS:3-6, wherein the BP is capable of binding NADPH, and wherein the BP comprises a heterologous protein domain inserted at or replacing at least part of the region corresponding to positions 22 to 25 of E. coli DHFR, said heterologous protein domain comprising as one member of the BRET pair a luciferase enzyme as bioluminescent donor protein (BDP); (ii) segment B comprises a ligand (L) capable of intramolecular binding to said BP only in the presence of NADPH and wherein segment B comprises a fluorescence acceptor moiety whose excitation spectrum at least partially overlaps with an emission spectrum of the luciferase; wherein the donor moiety and the acceptor moiety are in a suitable juxtaposition to yield a BRET signal when L is bound to BP, and wherein NADPH-induced binding of L to BP results in an increase in BRET efficiency.

2. The sensor molecule according to claim 1, wherein said heterologous protein domain further comprises a fluorescent protein, a self-labelling protein tag conjugated to a fluorophore, or a protein domain comprising an unnatural amino acid conjugated to a fluorophore.

3. The sensor molecule according to claim 1, wherein said luciferase enzyme is selected from the group consisting of a circular permutated luciferase, Renilla luciferase, firefly luciferase, Gaussia luciferase, and mutants thereof.

4. The sensor molecule according to claim 1, wherein the linker is a flexible polypeptide of at least 5 Gly residues.

5. The sensor molecule according to claim 1, wherein said L is selected from the group consisting of trimethoprim, methotrexate, aminopterin and 2,4 diamino-N10-methyl-pteroic acid (DAMPA).

6. The sensor molecule according to claim 1, wherein the sensor molecule is immobilized or absorbed to a solid carrier.

7. A method for fluorescence or luminescence-based in vitro detection of a concentration of NADPH in a sample, the method comprising (a) contacting the sample with a sensor molecule according to claim 1 under conditions allowing for an NADPH-dependent binding of said L to BP; and (b) analyzing a change in a signal generated by modulation of the spectroscopic properties of the fluorescence acceptor moiety or emission spectra of the sensor molecule and relating the signal change to the concentration of NADPH in the sample.

8. The method according to claim 7, wherein the sample is a biological sample or a fraction thereof.

9. The method according to claim 7, wherein the sample is contacted with the sensor molecule while the sensor molecule is immobilized or absorbed onto a solid carrier.

10. A kit of parts comprising the sensor molecule according to claim 1 and a solid carrier.

11. The kit according to claim 10, further comprising a luciferase substrate.

12. The kit according to claim 10 further comprising instructions or at least one further component suitable for one or more of the following: (i) ex vivo clinical or diagnostic testing performed on a serum or bodily fluid sample; (ii) an ex vivo enzymatic assay that involves formation or consumption of NADPH; (iii) ex vivo high-throughput screening for compounds that can modulate NADPH in cells or for validation of a toxicity profile of a therapeutic drug; (iv) live cell measurements comprising use of a widefield fluorescence microscope, confocal fluorescence microscope or a Fluorescence Lifetime Imaging Microscopy (FLIM) system with appropriate excitation and emission filters.

13. The sensor molecule according to claim 1, wherein said DHFR is a variant of SEQ ID NO:2 having at least 95% sequence identity to SEQ ID NO:2 and comprising one or more mutations selected from the group consisting of R98A, R44A and H45Q compared to SEQ ID NO:2.

14. The sensor molecule according to claim 6, wherein the solid carrier is selected from the group consisting of glass, a transparent plastic, a gold surface, a membrane, paper and a gel.

15. The method according to claim 8, wherein the sample is a cell lysate or a bodily fluid.

16. The kit of parts according to claim 10, wherein said solid carrier is paper or a transparent object.

17. The sensor molecule according to claim 1, wherein the acceptor moiety is a fluorophore.

18. The sensor molecule according to claim 1, wherein the acceptor moiety is a tethered fluorophore.

19. The sensor molecule according the claim 1, wherein the acceptor moiety is selected from the group consisting of cyanine dyes, fluorescent proteins, rhodamine dyes, fluorescein derivatives, triarylmethane dyes, naphthalimide dyes, xanthene dyes, acridine dyes, and coumarins.

20. The sensor molecule according to claim 1, wherein said heterologous protein domain is inserted at or replacing at least part of the region corresponding to positions 23 to 24 of E. coli DHFR.

Description

LEGEND TO THE FIGURES

(1) FIG. 1. (A) Pictorial description of the general sensing mechanism of a FRET-based NADPH sensor. A functional sensor is formed by tethering a DHFR ligand and a FRET donor fluorophore through a self-labeling protein tag (SLP1) to the DHFR inserted/interrupted by another orthogonal self-labeling protein (SLP2) which is labeled by a FRET acceptor. NADPH closes the sensor and increases the FRET efficiency between the FRET pair. P30 indicates a poly-L-proline linker moiety. (B) Pictorial description of the general sensing mechanism of a BRET-based NADPH sensor comprising a bioluminescent donor protein (BDP). A functional sensor is formed by tethering a DHFR ligand and a BRET acceptor fluorophore to the BDP-inserted DHFR through a SNAP-tag. NADPH closes the sensor and increases the BRET efficiency between fluorophore and BDP. P30 indicates a poly-L-proline linker moiety.

(2) FIG. 2. (A) Schematic representation of cpNLuc (PBD ID 5BOU) and its insertion in E. coli DHFR (PDB ID 4PDJ) (B) Sequence alignment of DHFR from E. coli (POABQ4), human (P00374), chicken (P00378), Drosophila (P17719) and Lactobacillus casei (P00381). The highlighted areas in yellow correspond to the residues at positions 20-27 of E. coli DHFR.

(3) FIG. 3. (A) Pictorial description of the sensing mechanism of a BRET-based NADPH sensor where BDP is cpNLuc. A functional sensor is formed by tethering a DHFR ligand TMP and a BRET acceptor Cy3 to the cpNLuc-inserted DHFR through a SNAP-tag. NADPH closes the sensor and increases the BRET efficiency between the Cy3 and cpNLuc. (B) Chemical structure of the tethered compound. Benzylguanine (BG) tethers covalently the Cy3 and TMP to the SNAP-tag. A peg11 linker is placed between the BG and Cy3. (C) Response curve of the sensor titrated with NADPH. Various mutations are introduced on the cpNLuc-inserted DHFR to tune the C50 of the sensor. The resulting variants have C50s ranging from 5.6 nM to 6.2 ?M. (D) NADPH response curve of sensors generated by inserting cpNLuc into DHFR (red) and by fusing NLuc to the N-terminus of either DHFR (white), circular permutated DHFR (grey), or circular permutated DHFR with a mutation (black). (E) Fluorescent polarization assay to measure affinity between TMP and cpNLuc-inserted DHFR as BP. (F) Fluorescent polarization assay to measure affinity between TMP and wild-type DHFR as BP.

(4) FIG. 4. (A) Quantification of glucose using NADPH BRET sensor. (B) Quantification of glutamate using NADPH BRET sensor.

(5) FIG. 5. (A) Response curve of the test paper titrated with NADPH. The sensor emission ratios are obtained by analyzing the photo. The ratios are plotted with the corresponding NADPH concentration in each spot. (B) Quantification of phenylalanine spiked in blood. (C) Quantification of glucose spiked in serum sample. (D) Quantification of glutamate spiked in serum sample. The plot B, C and D show the analyte concentrations after the sample dilution in the reaction buffer.

(6) FIG. 6. (A) Fluorescent polarization assay to measure affinity between TMP and cpSNAP-inserted DHFR as BP. (B) Fluorescent polarization assay to measure affinity between TMP and cpHalo-inserted DHFR as BP. (C) Pictorial description of the sensing mechanism of a FRET-based NADPH sensor where cpHalo is inserted in DHFR. (D) Emission spectra of FRET-based NADPH sensor with presence of NADPH at various concentrations. (E) Response curve of the FRET-based NADPH sensor titrated with NADPH.

EXPERIMENTAL SECTION

Example 1: NADPH Sensor Based on Bioluminescent Resonance Energy Transfer (BRET)

(7) This example describes the development of a BRET-based sensor capable of sensing NADPH. Circular permuted NLuc (cpNLuc) is used as the heterologous protein domain inserted in E. coli DHFR. The cpNLuc also forms the BRET partner with a fluorophore Cy3. The Cy3 is tethered with a DHFR ligand trimethoprim in the synthetic part of the sensor. See FIG. 3A for a schematic representation of the NADPH sensing mechanism.

(8) The proteinaceous moiety of the sensor comprises the cpNLuc-inserted DHFR, a poly-L-proline linker and a SNAP-tag. The fusion proteins were cloned in pET51b(+) plasmids and expressed in E. coli strain BL21. The DNA sequence of the plasmids was verified through Sanger sequencing. The fusion protein was purified from bacterial lysate using Ni-NTA and Strep-Tactin columns.

(9) The synthetic moiety of the sensor contains an O.sup.6-benzylguanine (BG) group for the SNAP-tag labeling, a peg11 linker, a fluorophore Cy3 and trimethoprim. The moiety as shown in FIG. 3B was prepared according to a previously described method (Griss et al. Nature Chemical Biology, 10(7), 598-603. doi: 10.1038/Nchembio.1554). The functional sensor was formed by labeling the synthetic moiety to the fusion protein. 8 ?M synthetic moiety and 2 ?M fusion protein were mixed in HEPES buffer containing 50 mM HEPES and 50 mM NaCl at pH 7.2. The mixture was incubated at room temperature for 1 h. The resulting sensor shown in FIG. 3A was used without further purification.

(10) The sensor was titrated with NADPH to assess its response. NADPH at various concentrations were mixed with 100 ?M sensor in 50 ?L buffer (50 mM HEPES, 50 mM NaCl, pH 8.5) in a white microtiter plate (Greiner Bio-One). 50 ?L of the same buffer containing 500-fold diluted furimazine (Nano-Glo Luciferase Assay Substrate, Promega) was subsequently added in each well. An EnVision Multilabel Reader (PerkinElmer) was used to measure the bioluminescent signal from the wells. The NLuc emission was measured at 460 nm (bandwidth 25 nm) while the Cy3 emission was measured at 595 nm (bandwidth 60 nm). To record the bioluminescent emission spectra, the same titration solutions were measured by an infinite M1000 spectrofluorometer (Tecan) with 2 nm step size, 20 nm bandwidth and 100 ms integration time.

(11) To cover a wider concentration range, several mutations (R98A, R44A and H45Q) were introduced on the DHFR inserted by cpNluc to tune the C50 of the sensor. FIG. 3C shows the response of different sensors to NADPH. The sensors demonstrated a maximum ratio change between 1485% and 3027% with C50s ranging from 5.6 nM to 6.2 ?M

Example 2: Relevance of Inserting a Heterologous Protein Domain in the Targeting Region of DHFR

(12) This example compares the performance of BRET sensors generated by either the insertion of cpNLuc in between positions 23 and 24 of E. coli DHFR according to the invention, or by fusing NLuc to the N terminus of either E. coli DHFR, circular permutated E. coli DHFR (cpDHFR), or circular permutated E. coli DHFR with mutations.

(13) Two proteinaceous moieties with NLuc that is not inserted into E. coli DHFR are designed for this comparison: SNAP-p30-NLuc-DHFR and SNAP-p30-NLuc-cpDHFR (Griss et al. Nature Chemical Biology, 10(7), 598-603. doi: 10.1038/Nchembio.1554). The fusion proteins were obtained according to the methods described in Example 1 and were further labeled with the synthetic moiety containing an O6-benzylguanine (BG) group for the SNAP-tag labeling, a peg11 linker, a fluorophore Cy3 and trimethoprim.

(14) The labeled constructs were titrated with NADPH to assess its response according to the methods described in Example 1. However, both constructs only showed modest response (<50%) to NADPH. By contrast, the sensor with cpNLuc inserted into E. coli DHFR as BP showed a response of 1485%, see FIG. 3D.

(15) In the absence of NADPH, the construct SNAP-p30-NLuc-cpDHFR remains in a closed conformation when TMP is used as a ligand, indicating a too strong affinity of the tethered ligand. Hence, a mutation L54G was introduced in the cpDHFR to reduce its affinity. Furthermore, a ligand with weaker affinity than TMP, namely DAMPA, was used as the tethered ligand. However, even with such improvement, the resulting construct only showed 87% change of the emission ratio in response to NADPH, see FIG. 3D.

(16) As demonstrated in Example 1 and the red plot of FIG. 3D, the insertion of cpNLuc in DHFR significantly improves the NADPH-dependent affinity of the BP towards the tethered ligand.

(17) This example further demonstrates how NADPH can affect the affinity between the binding proteins and the ligand. The apparent K.sub.D between the binding proteins and the ligand was determined by fluorescent polarization assays in the absence and presence of 100 ?M NADPH. In accordance with the titration results (FIG. 3D), the cpNLuc-inserted DHFR showed a 1400-fold increase of affinity towards TMP with the presence of NADPH (FIG. 3E). By contrast, NADPH only increased the affinity of the wild-type DHFR towards TMP by 23 folds (FIG. 3F).

(18) The fluorescent polarization assays were performed by mixing various concentrations of BP with 1 nM TMP derivatized with fluorophore tetramethylrhodamine (TMP-TMR) in 100 ?L of buffer (50 mM HEPES, 50 mM NaCl, pH 8.5) in a black 96-well plate (Greiner Bio-One). The fluorescent polarization values were measured using a Spark? 20M microplate reader (Tecan) with excitation wavelength at 535 nm (bandwidth 25 nm) and emission wavelength at 595 nm (bandwidth 35 nm). For receptors with an apparent K.sub.D lower than 1 nM towards TMP, 0.5 nM TMP-TMR was used in the assay.

Example 3: NADPH Sensor for Screening Enzymatic Activity in Solution

(19) The example describes the use of an NADPH sensor to screen for a desired enzymatic activity in solution. As a proof of principle, mutants of wild-type phenylalanine dehydrogenase (PDH) were screened for NADPH-dependent enzymatic activity. The wild-type PDH takes only NADH as cofactor. A NADPH-dependent variant is needed for developing a paper-based bioluminescent assay for phenylalanine.

(20) Wild-type PDH and 2 PDH mutants were expressed and purified in E. coli strain BL21. Ni-NTA and Strep-Tactin columns were used to purify the enzymes from the bacterial lysate. 1 ?M purified enzymes were added into 50 ?L buffer containing 1 nM NADPH cpNLuc-based BRET sensor (C50=1.36 ?M), 1 mM NADP.sup.+, 20 ?M phenylalanine, 100 mM glycine, 100 mM KCl and 100 mM KOH at pH 10.4. After a 5 min incubation, 50 ?L buffer containing NLuc substrate furimazine, 100 mM glycine, 100 mM KCl and 100 mM KOH at pH 10.4 was added into the previous buffer. The bioluminescent signal from the solution was measured using an EnVision Multilabel Reader (PerkinElmer). The NLuc emission was measured at 460 nm (bandwidth 25 nm) while the Cy3 emission was measured at 595 nm (bandwidth 60 nm). The intensity ratio between the NLuc and Cy3 emission was used to assess the NADPH-dependent activity of the variants. A lower NLuc/Cy3 intensity ratio indicates a higher concentration of the produced NADPH, and thus a higher desired activity.

Example 4: NADPH Sensor for Solution-Based Enzymatic Assay with Nanomolar Sensitivity

(21) The NADPH BRET sensor of Example 1 was used to develop enzymatic assays with nanomolar sensitivity. A glucose assay based on the following reaction was developed as a proof of principle:
Glucose+ATP<=>glucose-6-phosphate+ADP(Hexokinase)
G6P+NADP.sup.+<=>6-phosphogluconate+NADPH+H.sup.+(G6P dehydrogenase)

(22) 20 nM to 800 nM glucose were added in a buffer containing 1 nM NADPH BRET sensor (C50=27 nM), 5 ?M NADP.sup.+, 250 ?M ATP, 2 mM MgCl.sub.2, >1 ku/L hexokinase, >1 ku/L G6P dehydrogenase, NLuc substrate furimazine, 50 mM HEPES and 50 mM NaCl at pH 8.5.

(23) After a 5 min incubation, the bioluminescent signal from the solution was measured using an EnVision Multilabel Reader (PerkinElmer). The NLuc emission was measured at 460 nm (bandwidth 25 nm) while the Cy3 emission was measured at 595 nm (bandwidth 60 nm). The glucose concentration was obtained based on the NLuc/Cy3 emission ratio and a titration curve performed under the same condition prior to the assay (FIG. 4A).

(24) Similarly, an enzymatic assay for glutamate was performed based on the following reaction:
Glutamate+NADP.sup.+<=>ketoglutarate+NADPH (Glutamate dehydrogenase)

(25) Samples with 20 nM to 800 nM glutamate were added in a buffer containing 1 nM NADPH BRET sensor (C50=27 nM), 10 ?M NADP.sup.+, 100 ?M ADP, 1 mM EDTA, >1 ku/L glutamate dehydrogenase, NLuc substrate furimazine, 50 mM HEPES and 50 mM NaCl at pH 8.5.

(26) After a 5 min incubation, the bioluminescent signal from the solution was measured and the glutamate concentration in the solution was obtained using the same procedure described above (FIG. 4B).

Example 5: Bioluminescent Test Device for Phenylalanine

(27) The example demonstrates the use of immobilized NADPH BRET sensors to develop a quantitative point-of-care test device for the measurement of phenylalanine using a low volume of non-separated blood.

(28) The test paper was developed based on the following enzymatic reaction:
phenylalanine+NADP.sup.+<=>phenylpyruvate+NADPH(engineered phenylalanine dehydrogenase)

(29) Test device was made by freeze-drying 0.5 pmol of NADPH sensor (C50=27 nM) on a small paper disk. The paper disk has a 6 mm diameter and the area of the disk was defined by wax-based ink serving as a hydrophobic barrier. To perform the assay, 0.5 ?L of whole blood was incubated for 5 min in 24.5 ?L buffer containing 2 ?M engineered NADP.sup.+-dependent phenylalanine dehydrogenase (see Example 3), 1 mM NADP.sup.+, NLuc substrate furimazine, 100 mM glycine, 100 mM KCl and 100 mM KOH at pH 10.4. The engineered dehydrogenase consumes phenylalanine and produces NADPH stoichiometrically, which was further quantified by the test paper.

(30) To measure the produced NADPH, 5 ?L of the mixture was added onto the paper disk. The light produced by the sensor was measured using a digital camera mounted on a homemade cardboard box. The photo of the test paper was analyzed by a software that calculates the intensity ratio of the blue (NLuc) and red (Cy3) light emitted from each spot (Nature Chemical Biology, 10(7), 598-603). Phenylalanine concentrations in each spot were further calculated based on the emission ratios and a titration curve obtained under the same condition (FIGS. 5 A and B).

Example 6: Bioluminescent Test Device for Glucose

(31) The NADPH BRET sensor was employed to develop a quantitative point-of-care test paper for glucose. The test paper was developed based on the following enzymatic reaction:
Glucose+ATP<=>glucose-6-phosphate+ADP (Hexokinase)
G6P+NADP<=>6-phosphogluconate+NADPH+H.sup.+(G6P dehydrogenase)

(32) The test paper was made by freeze-drying 0.5 pmol of immobilized NADPH sensor (C50=1.36 ?M) on a small paper disk. 0.5 ?L of sample was diluted 500 folds in a buffer containing 100 ?M NADP.sup.+, 1 mM ATP, 2 mM MgCl.sub.2, >1 ku/L hexokinase, >1 ku/L G6P dehydrogenase, NLuc substrate furimazine, 50 mM HEPES and 50 mM NaCl at pH 8.5.

(33) After a 5 min incubation, glucose concentration in the sample was quantified by measuring the stoichiometrically produced NADPH using the test paper and camera (FIG. 5C).

Example 7: Bioluminescent Test Device for Glutamate

(34) The example shows the use of NADPH BRET sensor to develop a quantitative point-of-care test paper for glutamate. The assay was developed based on the following enzymatic reaction:
glutamate+NADP.sup.+<=>ketoglutarate+NADPH (Glutamate dehydrogenase)

(35) The test paper was made by freeze-drying 0.5 pmol of NADPH sensor (C50=1.36 ?M) on a small paper disk. 0.5 ?L of sample was diluted 500 folds in a buffer containing 100 ?M NADP.sup.+, 1 mM ADP, >1 ku/L glutamate dehydrogenase, NLuc substrate furimazine, 50 mM HEPES and 50 mM NaCl at pH 8.5.

(36) After a 5 min incubation, glutamate concentration in the sample was quantified by measuring the stoichiometrically produced NADPH using the test paper and camera (FIG. 5D).

Example 8: FRET-Based NADPH Sensors

(37) Example 1 described a BRET-based sensor for NADPH with the BP developed by inserting a circular permuted Nanoluc (cpNLuc) as the heterologous protein domain in DHFR. The affinity of this BP towards the ligand TMP demonstrates a dramatically improved NADPH-dependency compared to the wild-type DHFR (FIGS. 3E&F). The presence of NADPH decreased the K.sub.D between the BP and TMP from 54 ?M to 39 nM, representing a 1,400-fold increase of affinity, while the wild-type DHFR only demonstrate a 23-fold increase of affinity towards TMP when NADPH is present.

(38) The present example demonstrates that similar advantageous results are achieved when other heterologous protein domains are inserted in the sequence of DHFR. Circular permutated SNAP-tag (cpSNAP) or Halo-tag (cpHalo) were inserted between the amino acid residue 23N and 24L of DHFR to afford BPs with strong NADPH-dependent affinity towards the ligand TMP. The BP with cpHalo was further used to afford a FRET-based sensor for NADPH.

(39) The cpSNAP was developed by creating new N- and C-termini at 91Q and 93S and by linking the original N- and C-termini by a GGTGGSGGTGGSGGS linker (SEQ ID NO:1). The cpHalo was developed by creating new N- and C-termini at 141W and 144F and by linking the original N- and C-termini by a GGTGGSGGTGGSGGS linker (SEQ ID NO:1). The fusion proteins were produced as described in Example 1.

(40) The BPs developed by inserting either one of the heterologous protein domains (cpSNAP or cpHalo) showed a dramatically improved NADPH-dependent affinity towards the ligand. This affinity was measured by K.sub.D through fluorescent polarization assays as described in Example 2. By inserting cpSNAP, NADPH increased 2,000-fold the affinity between the BP and TMP (FIG. 6A). Similarly, by inserting cpHalo, NADPH increased 600-fold the affinity between the BP and TMP (FIG. 6B).

(41) A FRET-based sensor for NADPH was developed by fusing the BP comprising the inserted cpHalo to a poly-L-proline linker and a SNAP-tag. The cpHalo was labeled with a fluorophore Halo-SiR (Lukinavicius et al., 2013. Nat Chem, 5(2), 132-139) and the SNAP-tag was labeled with the synthetic molecule containing an O.sup.6-benzylguanine (BG) group for SNAP-tag labeling, a peg11 linker, a fluorophore Cy3 and TMP. The tethered fluorophore Cy3 is known to be a good FRET pair of Halo-SiR. See FIG. 6B for a schematic representation of the NADPH sensing mechanism.

(42) The functional sensor was formed by labeling Halo-SiR and the synthetic molecule to the fusion protein. 8 ?M Halo-SiR and 8 ?M synthetic molecule were mixed with 2 ?M fusion protein in HEPES buffer containing 50 mM HEPES and 50 mM NaCl at pH 7.2. The mixture was incubated at room temperature for 1 h. The labeled protein was then washed three times by HEPES buffer containing 50 mM HEPES and 50 mM NaCl at pH 7.2 using a protein centrifugal filter (Amicon Ultra) with a cut-off of 50 KDa.

(43) The sensor was titrated with NADPH to assess its response. NADPH at various concentrations were mixed with 10 nM sensor in 100 ?L buffer (50 mM HEPES, 50 mM NaCl, pH 8.5) in a black microtiter plate (Greiner Bio-One). A Spark? 20M microplate reader (Tecan) was used to record the emission spectra (FIG. 6D) of the sensor from 550 nm to 740 nm with excitation at 520 nm, 1 nm step size, 10 nm bandwidth and 40 ?s integration time.

(44) FIG. 6E shows the sensor's response to NADPH. The sensor demonstrated a maximum ratio change of 700% and a C50 of 158 nM.

REFERENCES

(45) Brun, M. A. et al. (2011). Journal of the American Chemical Society, 133(40), 16235-16242. doi: 10.1021/ja206915m Brun, M. A. et al. (2009). Journal of the American Chemical Society, 131(16), 5873-5884. doi: 10.1021/ja900149e Griss, R. et al., (2014). Nature Chemical Biology, 10(7), 598-603. doi: 10.1038/Nchembio.1554 Schena, A. et al., (2015). Nature Communications, 6. doi: Artn 7830 10.1038/Ncomms8830 Xue, L. et al., (2016). Journal of the American Chemical Society, 138(16), 5258-5261. doi: 10.1021/jacs.6b03034