Fluorescence Resonance Energy Transfer FRET

FRET-based readouts are the most prominent methods used for endopeptidase and carboxypepti-dase activity assays. In general, a huge number of biological assays have been developed based on the FRET principle (Van der Meer et al., 1994; Andrews and Demidov, 1999). FRET substrates extend on both sides of the scissile bond. This is important for proteases, where the S' binding site significantly contributes to the binding affinity of the substrate. For example, the activity of human

FIGURE 2.4 FRET readout principle. In the intact peptidic substrate (amino acids symbolized by X, Y and Z) labeled with two fluorophores (dye 1 and dye 2) at the opposite sites of the scissile bond. Dye 1 serves as fluorescence donor and dye 2 as fluorescence acceptor. Through an energy transfer (ET) from the donor to the acceptor, the fluorescence emission of dye 2 is observed (dark gray arrow). Thus the emission is significantly shifted to longer wavelengths compared to the emission of dye 1 (light gray arrow). After proteolytic cleavage of the substrate between amino acids X and Y by a protease, the energy transfer is disrupted and only the fluorescence emission of dye 1 is observed. An increase of fluorescence intensity of dye 1 and a decrease of fluorescence intensity of dye 2 over time dependent on the enzymatic velocity is recorded.

FIGURE 2.4 FRET readout principle. In the intact peptidic substrate (amino acids symbolized by X, Y and Z) labeled with two fluorophores (dye 1 and dye 2) at the opposite sites of the scissile bond. Dye 1 serves as fluorescence donor and dye 2 as fluorescence acceptor. Through an energy transfer (ET) from the donor to the acceptor, the fluorescence emission of dye 2 is observed (dark gray arrow). Thus the emission is significantly shifted to longer wavelengths compared to the emission of dye 1 (light gray arrow). After proteolytic cleavage of the substrate between amino acids X and Y by a protease, the energy transfer is disrupted and only the fluorescence emission of dye 1 is observed. An increase of fluorescence intensity of dye 1 and a decrease of fluorescence intensity of dye 2 over time dependent on the enzymatic velocity is recorded.

neutrophil elastase (HNE) depends greatly on the lengths of the synthetic substrates (Lestienne and Bieth, 1980). The S1' site of this protease in particular plays an important role in substrate binding (Stein and Strimpler, 1987).

The FRET readout depends on the presence of fluorescence donor and fluorescence acceptor molecules. In protease assays, these two dyes are incorporated into the cleavable substrate. The donor and the acceptor molecule must be attached on either side of the scissile bond but must remain in close proximity to each other to enable an efficient energy transfer from donor to acceptor (Figure 2.4). The mechanism of resonance energy transfer is formulated by classic and quantum mechanical theory. The Förster theory (Förster, 1948; Turro, 1965) describes the transfer mechanism as a weak dipole-dipole resonance coupling between donor and acceptor. According to this theory, the efficiency of the energy transfer is dependent on the quantum yield of the donor in the absence of transfer and the spectral overlap between donor and acceptor. This means that the maximum of the fluorescence emission spectrum of the donor must be at or near the excitation maximum of the fluorescence acceptor to enable a significant energy transfer. The efficiency of energy transfer decreases with the sixth power of the distance.

2.3.2.1 FRET Quench

In most cases of FRET dye pairs for protease assays, the acceptor is a fluorescence dark quencher (Figure 2.5). The fluorescence quencher does not emit fluorescence and the transferred energy is lost through, for example, singlet-triplet transitions or internal conversions. The energy can also be transferred by a collision mechanism arising from direct contact between donor and acceptor when electron shares of donor and acceptor overlap. An effective quenching by the collision mechanism requires a short distance between donor and acceptor. The peptide substrates carrying the two fluorophores are usually flexible molecules for which the predictions of the distance between donor and acceptor and thus of the energy transfer mechanism are difficult.

FIGURE 2.5 FRET quench readout principle. In the intact peptidic substrate (amino acids symbolized by X, Y and Z) labeled with a fluorophore (dye) and a quencher (Q) at the opposite sites of the scissile bond, the fluorescence emission is quenched through an energy transfer (ET) from the fluorophore to the quencher. After cleavage of the substrate between amino acids X and Y by a protease, the energy transfer is disrupted and an increase in fluorescence emission is observed (light gray arrow). An increase of fluorescence intensity over time dependent on the enzymatic velocity is recorded.

FIGURE 2.5 FRET quench readout principle. In the intact peptidic substrate (amino acids symbolized by X, Y and Z) labeled with a fluorophore (dye) and a quencher (Q) at the opposite sites of the scissile bond, the fluorescence emission is quenched through an energy transfer (ET) from the fluorophore to the quencher. After cleavage of the substrate between amino acids X and Y by a protease, the energy transfer is disrupted and an increase in fluorescence emission is observed (light gray arrow). An increase of fluorescence intensity over time dependent on the enzymatic velocity is recorded.

In the substrate, the fluorescence of the donor is quenched by the quenching label in close proximity. Upon cleavage of the substrate by the protease, this proximity is lost and a strong increase in fluorescence can be recorded. Table 2.3 summarizes a selection of fluorescence donor and acceptor pairs frequently used for protease activity assays based on a FRET quench readout. In principle, the FRET quench effect can be achieved by labeling the substrate with two different dyes with overlapping spectra (hetero-double labeling).

Historically, the oriho-aminobenzoyl group (Abz) was frequently used as fluorescence donor because of its small size, high hydrophilicity, and high quantum yield (Gershkovich and Kholodovych,

TABLE 2.3

Examples of Fluorescence Donor and Acceptor Pairs Frequently Used for Protease Assays Based on FRET Quench Readout Principle

Excitation Emission Excitation

Donor Wavelength (nm) Wavelength (nm) Acceptor Wavelength (nm) Reference

TABLE 2.3

Examples of Fluorescence Donor and Acceptor Pairs Frequently Used for Protease Assays Based on FRET Quench Readout Principle

Excitation Emission Excitation

Donor Wavelength (nm) Wavelength (nm) Acceptor Wavelength (nm) Reference

Abz

340

415

pNA

315

Stambolieva et al., 1992

Abz

340

415

Phe(NO2)

280

Nishino et al., 1992

Abz

340

415

EDDnp

360

Carmel et al., 1977

Abz

340

415

Nitrotyrosine

428

Angliker et al., 1995

Abz

340

415

Dnp

360

Mao et al., 2008

EDANS

336

495

DABCYL

472

Wang et al., 1990

Mca

328

393

Dnp

360

Xia et al., 1999

DANSYL

330

520

Phe(NO2)

280

Knight et al., 1992; Chen, 1968

Cy5.5

675

695

NIRQ820

790

Malfroy and Burnier, 1987

Cy3

530

570

Cy5Q

640

www.gehealthcare.com

TAMRA

547

573

QSY7

560

www.invitrogen.com

1996). Abz was combined with a broad variety of non-fluorescent acceptors such as p-nitrobenzyl for leucine aminopeptidase (Carmel et al., 1977), pNA for trypsin (Bratanova and Petkov, 1987), 4-ni-trophenylalanine [Phe(NO2)] for HIV protease (Toth and Marshall, 1990), and N-(2,4-dinitrophenyl) ethylenediamine (EDDnp) for thermolysin and trypsin (Nishino et al., 1992). Lecaille et al. (2003) described a FRET quench assay based on a specific substrate for cathepsin K labeled with Abz and EDDnp. This substrate is not cleaved by the other C1 cysteine cathepsins and serine proteases in contrast to methoxycoumarin (Mca)-based substrates described earlier (Aibe et al., 1996; Xia et al., 1999) and merely covered the non-primed site of the scissile bond. The 5-[(2-aminoethyl)amino] naphthalene-1-sulfonic acid (EDANS) compound is a second example of a fluorescence donor historically used for many FRET quench-based protease assays, e.g., in combination with tryptophan as a quencher in an ECE activity assay (Von Geldren et al., 1991). The FRET-1 example in Table 2.2 shows the typical dynamic range that can be achieved with an EDANS/DABCYL-based assay.

EDANS is still a prominent fluorescence donor in protease assay development today, mainly in combination with 4-(4-dimethyl-aminophenylazo)-benzoic acid (DABCYL) as a non-fluorescence quencher. Zou et al. (2005) describe a recent application of this dye pair in an ADAM33 assay. The combination of Mca as a fluorophore and dinitrophenyl (Dnp) as a quencher shows similar high levels of fluorescence intensity and high quenching efficiency of >90% as EDANS/DABCYL. In contrast to many other FRET dye pairs, both EDANS/DABCYL and Mca/Dnp generate a dynamic range between cleaved and uncleaved substrate, sufficient for the application of such protease assays in HTS and high-throughput compound profiling under initial velocity conditions.

One recent example for the application of the FRET quench principle with Mca and Dnp as a dye pair for the development of robust assays for ADAMTS-4 and ADAMTS-5 was described by Lauer-Fields et al. (2007). High quenching efficiencies of >90% were also observed with DANSYL as a donor and tryptophan and Phe(NO2) as a quencher, respectively, in assays for neutral endopeptidase (Malfroy and Burnier, 1987; Florentin et al., 1984). One disadvantage of the EDANS/DABCYL and Mca/Dnp dye pairs is the excitation maximum in the ultraviolet range. As a result, the potential for the interference of the fluorescence signal with that of the chemical compounds tested in protease assays is high, leading to enhanced false negative rates or misinterpretation of IC50 data.

Many drug-like compounds display fluorescence characteristics when excited at wavelengths in the ultraviolet range. In addition, poor solubility of Mca/Dnp can limit the application of this dye pair for the development of robust protease assays in an HTS environment. Other FRET dye pairs with red-shifted fluorescence signals diminish the avoidance of fluorescence artifacts caused by interference with the compounds tested. The near-infrared (NIR) probe pair consisting of Cy5.5 a as fluorescence emitter (excitation and emission at 667 and 690 nm, respectively) and NIRQ820 as a fluorescence quenchner is an example. The successful application of this dye pair in an MMP7 assay was demonstrated (Pham et al., 2004).

Other red-shifted dye pairs include tetramethylrhodamine (TAMRA), MR121 or europium chelates (Truepoint™ by Perkin Elmer) in combination with Cy3/Cy5Q or members of the QSY dark quencher family (QSY7 or QSY21). Mao et al. (2008) recently described an assay based on the Truepoint principle to measure the inhibition of the NS3-4A hepatitis C virus protease. Konstantinidis et al. (2007) compared the interferences of compounds with substrates labeled with EDANS/DABCYL and TAMRA/QSY7 (excitation and emission at 547 and 573 nm), respectively, in a hepatitis C virus NS3-4A assay. As expected, they observed higher compound interference with EDANS/DABCYL. George et al. (2003) compared a Cy3/Cy5Q-labeled substrate with the respective Mca/Dnp-labeled analogue in an MMP3 assay and observed higher robustness and less compound interference with the Cy dye substrate.

The major disadvantages of these alternatives to EDANS/DABCYL and Mca/Dnp are the much greater expense and the frequent generation of dynamic signal ranges that are not sufficient for measurement under initial velocity conditions in our hands. Moreover, a limitation of the Truepoint™ is the minimum length of the applicable peptide substrates of 8 to 15 amino acids (personal communication). A disadvantage of double labeling is that the quenching mechanism often affects the conformational flexibility within the substrate in a way that compensation by increasing enzyme concentrations is required to obtain a sufficiently high fluorescence signal. This increase in enzyme concentration causes a reduction of sensitivity.

In general, a FRET quench readout is simple. A broad range of available fluorescence donors and acceptors allows cost-efficient operations in an industrialized HTS and automated compound profiling environment. On the other hand, the readout can suffer from inner filter effects due to high absorption coefficients of the dyes and fluorescence artifacts by the tested compounds, resulting in enhanced false positive and false negative rates. Moreover, the readout is limited to substrates in which short distances between donor and acceptor dye can be realized without disturbing the interaction of enzyme and substrate. The flexibility of the peptide conformation makes the prediction of the effective distance between the dyes and consequently the prediction of the FRET effect difficult. The distance between donor and acceptor cannot be easily approximated by the mean hydrodynamic radii of the dyes.

0 0

Post a comment