Biosensors - Approaches

Introduction

Localized, transient signaling is intrinsic to cell migration. To understand these spatial and temporally regulated signals will require probes that report the locations of molecular players in their activated or functional forms. While some signaling molecules localize in very discrete locations because of their specific, constitutive interactions with adaptor molecules, others only become localized upon activation, which in turn leads to specific interactions with target or effector molecules. Fluorescent probes are being developed to elucidate these activation states. They include fluorescently-labeled antibodies against the activated state of the molecule of interest and reagents that sense interactions between two different proteins. The later type of probe requires the introduction of labeled molecules into cells. One approach is to fluorescently label the expressed molecule or a domain of interest and introduce it into cells by injection or some other approach. A more robust approach is to produce a cDNA encoding the molecule of interest fused to a derivative of the green fluorescent protein (GFP). When expressed in the cell this produces a fluorescent fusion protein of the molecule of interest with GFP. A large number of GFP derivatives have been developed with differing fluorescence properties including color, e.g., green, cyan, yellow, and stability (see Tsien, 1998; Terskikh et al., 2000; Zang et al., 2002; Matz et al., 2002; Gaits & Hahn, 2003).

return to top

Biosensor Strategies

The following is a brief outline of some of the major strategies used to detect activated states that have been used for migration studies.

Tagged Domains

When a particular molecule or domain interacts robustly with a signaling partner, its localization can often be seen upon expression or introduction of the tagged domain into the cell. Using this approach a GFP-tagged AKT, a PH domain-containing protein that binds to phosphatidyl inositols was used to show that phosphatidylinositol-3,4,-trisphosphate (PIP2) and phosphatidylinositol-3,4,5-trisphosphate (PIP3), accumulates at the cell front in response to chemotactic stimulation in Dictyostelium discoideum (Dormann et al, 2002). As another example, GFP-tagged SH2 domains have been used to show that PLC-gamma translocates from a uniform cytosolic distribution to a punctate plasma membrane microdomain in response to activation of tumor mast cells by IGE (Stauffer & Meyer, 1997).

Proximity Sensors Using FRET

When the binding interactions between an activated signaling molecule and its target are not robust, they can be difficult to detect since the localized interaction can be hard to see over background fluorescence. For these interactions, fluorescence energy transfer can be used since it generates a unique fluorescence signal. In a typical fluorescent measurement, a fluorophore is excited by a specific wavelength of light (excitation wavelength) and emits light at a different wavelength (emission wavelength). However, when two fluorophores are paired in such a way that the emission wavelength of one overlaps with the excitation wavelength of the other, excitation of the former will stimulate fluorescence of the latter when they reside within about 60 Å of each other. This approach has tremendous utility because the unique fluorescence signal generated under these circumstances can be used to visualize and quantify the position and concentration of interacting fluorophores. The details behind FRET and the microscopy approaches used to visualize it have been reviewed extensively (see Pollok & Heim, 1999; Selvin, 2000 as examples).

return to top

Two major strategies have been used to develop FRET biosensors: (i) two chain probes in which the fluorophores are on two different molecules resulting in intermolecular FRET when the two molecules come into proximity, or (ii) single chain probes in which different regions of a single molecule are tagged and undergo FRET due to intramolecular, conformational changes. Using these strategies a number of different FRET biosensors have been developed (Gaits & Hahn, 2003; Dormann et al., 2002; Chamberlain et al., 2000; Kraynov et al., 2000).

FRET-Dual Chain : fluorophores are engineered into two different proteins or domains of proteins which are known to interact with one another. Interaction between the proteins brings them close enough together to permit FRET, which is an indicator of the interaction Click here for an example animation

The diagram below shows how this two chain probe FRET approach has be used to demonstrate the localized activity of the small GTPase, Rac, which regulates the formation of protrusions and adhesions in migrating cells. The measurement senses the interaction of GFP-labeled Rac with the protein binding domain (PDB) from PAK, a Rac effector, that is labeled with an Alexa dye. When Rac is loaded with GTP, it binds the PBD. When the GFP is excited by light, fluorescence is observed at the wavelength of the Alexa dye, whose excitation wavelength coincides with the emission wavelength of the GFP. The Alexa fluorescence is readily distinguishable from that of GFP alone. This provides a demonstration of their interaction (Kraynov et al., 2000).

FRET - Single Chain: using this approach both fluorophores are engineered into a single molecule which undergoes a conformational change upon activation or interaction with a specific ligand. The fluorophores are integrated into the protein so that the conformational transition induces a change in proximity or orientation of the fluorophores that alters FRET efficiency. Click here for example animations 1 & 2

This approach overcomes some of the stoichiometry issues associated with the two chain probe approach, since a one to one relationship of the fluorophores is built into the probe. They are also simpler to use since only one probe must be introduced into cells, and it can be introduced as a cDNAs that is expressed in the cell. However, the conformational transition must be sufficient to generate a measurable change in FRET. Many proteins have "closed", inactive conformations that are converted to an "open" active form upon binding appropriate regulatory molecules, making this a particularly useful approach for these molecules. In other cases an effector domain can be engineered onto the molecule of interest. When the regulatory molecule is activated the effector domain binds inducing a change in the proximity of the fluorophores that are fused to the signaling molecule and its effector. This single chain approach has been used successfully for monitoring Rac and Cdc42 activation.

return to top

Mechanical Tension Sensor

return to top

Environment-Sensitive Fluorophores

Polarity sensitive fluorophores can also be used to sense activation. In this approach, a fluorophore is introduced at a site in a protein that undergoes a change in its environment upon activation. For example a phosphorylation event could be detected if an environmentally sensitive fluorophore is engineered into a protein close to a phosphorylation site. Upon phosphorylation, the polarity around the fluorophore would change, and this would be reflected in altered fluorescence properties of the fluorophore. Alternatively, the fluorophores may sense phosphorylation-dependent binding of specific proteins to the site (see figure below). It is also possible that environment sensitive dyes could sense global changes in conformation if the tagged site changes from buried within the protein core to being exposed. To detect these kinds of phosphorylation and binding events, environment sensitive fluorescent dyes are being developed (Wouters et al., 2001; Zang et al., 2002).

This schematic illustrates how fluorescent probes can detect a phosphorylation event through specific interactions between a fluorophore and the incipient phosphate.

This schematic illustrates how fluorescent probes can be integrated into the kinase target and signal the phosphorylation indirectly due to changes when the phosphoprotein binds to a cognate protein or ligand such as an SH2 domain.

The introduction of environment sensitive probes have proved to be powerful tools for the study of complex biological systems, but since many biologically important species do not fluoresce, chemical modification of existing proteins or chemical synthesis and semi-synthesis of fluorescent analogs of native peptides and proteins has to be carried out. Of particular interest are the environment-sensitive probes that change their spectral properties upon changes in the environment.

PRODAN and Derivatives: The 6-propionyl-2-(dimethylamino)naphthalene (PRODAN) fluorophore, first introduced by (Weber & Farris, 1979, is one of the best studied environment-sensitive fluorophores. Upon transfer of the fluorophore from water to a more hydrophobic environment there is a marked blue shift in the emission maximum wavelength, and an increase in the quantum yield of the PRODAN. Some examples of the use of PRODAN-based fluorophores in biologically relevant systems include:Nitz et al., 2002; Cohen et al., 2002 & Vazquez et al., 2003.

Structure of PRODAN and its derivatives. Structure of DANA, the PRODAN amino acid derivative developed for Fmoc solid phase peptide synthesis. Fluorescence spectra of PRODAN in cyclohexane (1), chlorobenzene (2), dimethylformamide (3), ethanol (4) and water (5). Adapted from Valeur, 2002

We have successfully used the PRODAN amino acid DANA in the synthesis of peptides that monitor protein-protein interactions (Nitz et al., 2002), bioactive peptides that interact with 14-3-3 proteins involved in cell cycle regulation (Vazquez et al., 2003), and peptides that selectively bind to SH2 domains involved in cell signaling.

Structure of an SH2 domain bound to a phosphotyrosine peptide. Fluorescence emission spectra of the peptide TEE-F-PpYSYPT in buffer and in the presence of two SH2 domains from Crk and Src showing selective binding to Crk-SH2 domain.

4-AP and 4-DMAP: Phthalimide derivatives 4-aminophthalimide (AP) and 4-N,N-dimethylamino phthalimide (DMAP) are structurally related compounds with outstanding solvatochromic properties, with a change in the quantum yield by a factor of 70 and fluorescent spectral shifts as large as 100 nm on changing the solvent from 1,4-dioxane to water. They have been extensively used in studies of membrane dynamics and structure.

Absorption and emission spectra of 4-AP derivative DAPL in different solvents, emission is greatly enhanced in more hydrophobic solvents, and also the maximum emission wavelength is shifted towards shorter wavelengths. Taken from Saroja et al., 1999.

Benzofurazan Derivatives: Benzofurazan derivatives are commonly used derivatization reagents that are non-fluorescent themselves, but become fluorescent upon reaction with the target molecule. The main advantages of benzofurazan derivatives are the high quantum yield of the tagged species, the long excitation and emission wavelengths and its fluorescent properties are highly sensitive to the polarity of the medium.

Structure of NDB-Cl and NDB derivatives: Fmoc-Dap(NDB)-OH for peptide synthesis and RTI-233. Photophysical properties of NDB derivatives in solvent mixtures of different polarities showing the increase in fluorescence upon decrease of polarity and binding of RTI-233 to serotonin transporter, taken from Rasmussen et al., 2001

return to top

Chelation-Enhanced Fluorescence

We have developed a versatile and sensitive fluorescent probe design for monitoring protein kinase activity. The design is flexible and contains kinase recognition elements. The sensing motif is small and does not perturb reactivity with the target kinase. In addition, the robust fluorescence allows these reactions to be monitored continuously. This approach has been used to develop Sox-based kinase assays for several different kinases in collaboration with Biosource International.  The assay is marketed under the name OMNIA.

General Sensing Mechanism. Fluorescence sensing of kinase activity is detected by a specific interaction between the introduced phosphate, Mg2+ and a chelation-sensitive fluorophore.

The sensor consists of a small sensing motif appended to an optimized peptide substrate. The fluorescence signal is generated when the non-natural amino acid Sox (Shults, et al.,2003), undergoes chelation-enhanced fluorescence in the presence of divalent magnesium. The sensing motif contains a beta-turn element to pre-organize Mg2+-binding between Sox and the newly introduced phosphate (Shults & Imperiali, 2003).

Relation of Mg2+-binding to fluorescence increase. Titration curves show that Mg2+ affinity of the product phosphopeptide is much greater than the substrate peptide. At 10 mM MgCl2, spectra collected with 360 nm excitation illustrate the large fluorescence difference between the peptide and phosphopeptide, which results in a large fluorescence increase upon phosphorylation by the kinase.

Reporter peptides for select serine/threonine and tyrosine kinases, as well as their corresponding phosphorylated product peptides, were synthesized by standard Fmoc solid phase peptide synthesis. The reporter substrates show comparable efficacy against the target kinase as the kinase recognition motifs they contain, indicating that the sensing motif does not affect reactivity.

return to top

Activation-Specific Antibodies

These antibodies bind specifically to the activated form of the molecule of interest. It may be that the epitope with which the antibody interactions is buried deep within the molecule and only becomes exposed through a conformational change that accompanies activation. Many proteins are regulated by phosphorylation, which can be assayed by antibodies that recognize phosphorylated epitopes. With these antibodies, the location of activated molecules can be imaged by traditional immunofluorescence techniques. A wide array of these phospho-specific antibodies are now available that include many key migration related proteins as their targets. While the antibody approach is not generally used for live cell imaging, it can be used in conjunction with video imaging to determine the history of the cell immediately prior to staining.

return to top

Text References

Dormann D, Weijer G, Parent CA, Devreotes PN, Weijer CJ. 2002. Visualizing PI3 kinase-mediated cell-cell signaling during Dictyostelium development. Curr Biol 12(14):1178-88 PubMed.

Chamberlain CE, Kraynov VS, Hahn KM. 2000. Imaging spatiotemporal dynamics of Rac activation in vivo with FLAIR. Methods Enzymol 325:389-400. PubMed

Cohen BE, McAnaney TB, Park ES, Jan YN, Boxer SG, Jan LY. 2002.Probing protein electrostatics with a synthetic fluorescent amino acid. Science. 296(5573):1700-3. PubMed

Gaits F, Hahn K. Shedding light on Cell Signaling: Interpretation of FRET Biosensors. 2003. Sci STKE 2003 Jan 14;2003(165):PE3, download PDF. PubMed

Kraynov VS, Chamberlain C, Bokoch GM, Schwartz MA, Slabaugh S, Hahn KM. 2000. Localized Rac activation dynamics visualized in living cells. Science 290(5490):333-7. PubMed

Matz MV, Lukyanov KA, Lukyanov SA. 2002. Family of the green fluorescent protein: journey to the end of the rainbow. Bioessays (10):953-9. PubMed

Nitz M, Mezo AR, Ali MH, Imperiali B. 2002. Enantioselective synthesis and application of the highly fluorescent and environment-sensitive amino acid 6-(2-dimethylaminonaphthoyl) alanine (DANA). Chem Commun (Camb). (17):1912-3. PubMed

Pollok BA, Heim R. 1999. Using GFP in FRET-based applications .Trends Cell Biol. 9(2):57-60. PubMed

Rasmussen SG, Carroll FI, Maresch MJ, Jensen AD, Tate CG, Gether U. 2001. Biophysical characterization of the cocaine binding pocket in the serotonin transporter using a fluorescent cocaine analogue as a molecular reporter. J. Biol. Chem.276(7):4717-23. PubMed

Saroja G, Ramachandram B, Saha S, Samanta A. 1999. J. Phys. Chem. B 103, 2906.

Selvin PR. 2000. The renaissance of fluorescence resonance energy transfer. Nat Struct Biol. 7(9):730-4. PubMed.

Shults MD, Imperiali B. Versatile fluorescence probes of protein kinase activity. J Am Chem Soc. 2003 Nov 26;125(47):14248-9. PubMed

Shults, M.D., Pearce, D.A., and Imperiali, B. 2003. A modular and tunable chemosensing scaffold for divalent zinc. J. Am. Chem. Soc. 125, 10591-10597. PubMed

Stauffer TP, Meyer T. 1997. Compartmentalized IgE receptor-mediated signal transduction in living cells. J Cell Biol. 139(6):1447-54. PubMed

Terskikh A, Fradkov A, Ermakova G, Zaraisky A, Tan P, Kajava AV, Zhao X, Lukyanov S, Matz M, Kim S, Weissman I, Siebert P.2000. "Fluorescent timer": protein that changes color with time. Science. 290(5496):1585-8. PubMed

Tsien RY. 1998. The green fluorescent protein. Annu Rev Biochem. 67:509-44. Review. PubMed

Valeur, B. Molecular Fluorescence, Principles and Applications; Wiley-VCH; Weinheim, Federal Republic of Germany, 2002.

Vazquez ME, Nitz M, Stehn J, Yaffe MB, Imperiali B.2003.Fluorescent caged phosphoserine peptides as probes to investigate phosphorylation-dependent protein associations. J Am Chem Soc.125(34):10150-1. PubMed

Weber G, Farris FJ. 1979. Synthesis and spectral properties of a hydrophobic fluorescent probe: 6-propionyl-2-(dimethylamino)naphthalene. Biochemistry. 18(14):3075-8. PubMed

Wouters FS, Verveer PJ, Bastiaens PI. 2001. Imaging biochemistry inside cells. Trends Cell Biol. (5):203-11. PubMed

Zhang J, Campbell RE, Ting AY, Tsien RY. 2002. Creating new fluorescent probes for cell biology. Nat Rev Mol Cell Biol (12):906-18. PubMed

return to top

Additional Reading

Bazzoni G, Hemler ME.1998. Are changes in integrin affinity and conformation overemphasized? Trends Biochem Sci. 23(1):30-4. PubMed

Dufau, I., Mazarguil, H. Tetrahedron Lett., 2000, 41, 6063. Synthesis of a NDB fluorescent amino acid for peptide synthesis.

Hadrich, D., Berthold, F., Steckan, E., Bönish, H. J. Med. Chem. 1999, 42, 3101. Biophysical studies using NDB-based reporter molecules.

G. Saroja, T. Soujanya, B. Ramachandram, A. Samanta J. Fluorescence 1998, 8, 405. Review about the applications of 4-AP and its derivatives as probes.

Soujanya, T.; Fessenden, R. W.; Samanta, A. J. Phys. Chem. 1996, 100, 3507. Photophysical study of the behavior of 4-aminophthalimide and derivatives.

Turcatti, G., Nemeth, K., Edgerton, M. D., Meseth, U., Talabot, F., Petisch, M., Knowles, J., Vogel, H., Chollet, A. J. Biol. Chem., 1996, 271, 19991.

Uchiyama, S., Santa, T., Imai, K. J. Chem. Soc. Perkin Trans. 2, 1999, 2525. Photophysical study of the behavior of different NDB derivatives.

return to top