Designs and applications of fluorescent protein-based biosensors

https://doi.org/10.1016/j.cbpa.2009.09.033Get rights and content

Genetically encoded biosensors allow the noninvasive imaging of specific biochemical or biorecognition processes with the preservation of subcellular spatial and temporal information. Aequorea green fluorescent protein (FP) and its engineered variants are a critical component of genetically encoded biosensors, as they serve to provide a ‘read-out’ of the biorecognition event under investigation. The family of FP-based biosensors includes a diverse array of designs that utilize various photophysical characteristics of the FPs. In this review, we will discuss these designs and their read-outs through reviewing some of the recent works in this area.

Introduction

Biosensing encompasses a diverse array of techniques for the generation of an experimentally accessible ‘read-out’ of a molecular interaction between a biomolecule-derived molecular recognition element (MRE) (e.g. a protein domain) and an analyte of interest (e.g. a small molecule, another protein, or an enzymatic activity). Molecular entities or devices that enable biosensing are generally referred to as biosensors. The primary challenge of creating biosensors is transducing the nanometer-scale event of a biorecognition process into an observable change in a macroscopic property such as color or fluorescence hue. One of the nanometer-scale changes that typically accompany biorecognition events is the change in molecular ‘geometry’ of the MRE. This change could be a distance change between the MRE and its analyte as in the case of protein–protein interaction, or a conformational change of the MRE as in the case of allosteric proteins. As will be discussed in this review, researchers have now devised a variety of strategies by which changes in the molecular geometry of an MRE can modulate the fluorescence hue or intensity of an intrinsically fluorescent protein (FP) belonging to the superfamily of Aequorea green FP-like proteins.

As described elsewhere in this issue [1], engineered FPs have revolutionized the ability of researchers to study protein localization and dynamics in live cells. FPs have also enabled the construction of genetically encoded FP-based biosensors that have numerous advantages relative to alternative technologies such as dye-based probes. Specifically, FP-based biosensors are relatively easy to construct using standard molecular biology techniques; able to be noninvasively introduced into living cells where they are produced using the cellular transcriptional and translational machinery; able to yield information about a biorecognition process in the natural habitat of the protein thus preserving spatial and temporal information of this interaction; able to be targeted to most cellular compartments using specific signal sequence tags.

Practically all genetically encoded FP-based biosensors can be classified into five groups depending on their structure. We define Group I as those biosensors based on intramolecular Förster Resonance Energy Transfer (FRET). Such biosensors have all of their components in a single polypeptide chain, and the analyte brings about a change in the structure or conformation of the MRE unit. This change is detected by ratiometric intensity measurements of the two FPs. Group II includes biosensors based on intermolecular FRET. In contrast to Group I, the two FPs are in two different polypeptide chains and are brought into proximity by a protein–protein interaction. Group III includes those biosensors based on bimolecular fluorescence complementation (BiFC). In this biosensing strategy, a biorecognition event is used to bring two fragments of a split FP suitable proximity for the reconstitution of an intact (and fluorescent) FP.

Groups IV and V are both based on single FPs encoded by a single polypeptide chain. The difference between these two groups is whether or not the MRE element of the biosensor is exogenous (Group IV) or endogenous (Group V) with respect to the FP. In the case of an exogenous MRE, the binding of the analyte causes conformational changes that are relayed to the chromophore environment and alter its spectral properties. In the case of an endogenous MRE, the FP plays a dual role: it is responsible for both the molecular recognition and the fluorescence read-out.

In this review, we will provide examples of the different designs of genetically encoded FP-based biosensors belonging to the aforementioned groups and describe recent progress in their development and application.

Section snippets

Group I: intramolecular FRET-based biosensors

FRET is the phenomenon of nonradiative energy transfer observed between an excited blue-shifted fluorescent chromophore (donor) and a chromophore with a red-shifted absorption spectrum (acceptor) through dipole–dipole coupling. FRET has proven to be extremely useful in the design of genetically encoded biosensors. The canonical structure of biosensors belonging to this group consists of two FPs flanking an MRE (Figure 1a). Changes in the MRE conformation alter the distance between the two FPs

Group II: intermolecular FRET-based biosensors

Biosensors belonging to this group are necessarily split constructs, in which the MRE is fused to one of the FPs and the analyte protein is fused to the other (Figure 1e). This design of biosensors is particularly useful for the study of protein–protein interactions. Intermolecular FRET has been applied to study the oligomerization state of different member of the G-protein-coupled-receptor (GPCR) superfamily [18, 19]. However, the versatility of this design of biosensors does not end at merely

Group III: BiFC-based biosensors

BiFC is dependent on the intrinsic ability of some FP variants, when expressed in a split form tagged to a pair of interacting proteins, to refold properly into the β-barrel structure and thus reconstitute the fluorescent form of the protein. BiFC-based biosensors are necessarily split constructs in which the MRE is genetically fused to one fragment of the FP and the analyte protein is fused to the other (Figure 2a). Several recent reviews provide a thorough treatment of the practical aspects

Group IV: single FP-based biosensors with an exogenous MRE

This class of genetically encoded single FP-based biosensors depends on the ability of some of the variants of FPs to tolerate protein insertion and circular permutations at certain locations. This property has allowed researchers to construct ligand sensitive single FP-based biosensors. The biorecognition event is carried out by an exogenous MRE and information about this event is relayed to the chromophore changing its spectral properties (Figure 2b). Some examples of a biosensor with this

Group V: single FP-based biosensors with an endogenous MRE

Most FP variants show pH-dependent change in their spectral properties [34]. For example, the engineered avGFP variants known as EGFP, ECFP, and EYFP have pKas for fluorescence quenching of 6.15, 6.4, and 7.1, respectively [35]. Recently an engineered variant of Discosoma RFP, known as mNectarine, was shown to exhibit a useful pH-dependency [36]. To demonstrate its potential, the authors fused mNectarine to the cytoplasmic amino acid terminus of human concentrative nucleoside transporter

Hybrid strategies

Owing to continuous innovation in the development of biosensor designs and experimental techniques to detect protein–protein interaction, some designs do not fit in any of the aforementioned categories. For example, some FRET-based biosensors do not depend on conformational changes in the FRET construct but rather on spectral changes in the acceptor FP. Esposito et al. designed a FRET construct that consists of the pH-insensitive donor cyan FP and a pH-sensitive variant yellow FP variant (

Outlook

Ongoing protein engineering efforts will eventually provide researchers with a complete repertoire of genetically encoded biosensors, each with specific properties that are ‘tuned’ to the conditions of the event under investigation. It is apparent that a substantial amount of progress has already been made toward this goal. For example, genetically encoded Ca2+ biosensors with a range of different affinities to Ca2+ have been developed. It is likely that other classes of biosensor will see

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

References (47)

  • R.E. Campbell

    Fluorescent-protein-based biosensors: modulation of energy transfer as a design principle

    Anal Chem

    (2009)
  • E.A. Jares-Erijman et al.

    FRET imaging

    Nat Biotechnol

    (2003)
  • Y.X. Wang et al.

    Fluorescence proteins, live-cell imaging, and mechanobiology: seeing is believing

    Annu Rev Biomed Eng

    (2008)
  • M.T. Tsai et al.

    Real-time monitoring of human enterovirus (HEV)-infected cells and anti-HEV 3C protease potency by fluorescence resonance energy transfer

    Antimicrob Agents Chemother

    (2009)
  • R. Sabariegos et al.

    Fluorescence resonance energy transfer-based assay for characterization of hepatitis C virus NS3-4A protease activity in live cells

    Antimicrob Agents Chemother

    (2009)
  • C.D. Harvey et al.

    A genetically encoded fluorescent sensor of ERK activity

    Proc Natl Acad Sci U S A

    (2008)
  • N.-N. Aye-Han et al.

    Fluorescent biosensors for real-time tracking of post-translational modification dynamics

    Curr Opin Chem Biol

    (2009)
  • M.A. Dwyer et al.

    Periplasmic binding proteins: a versatile superfamily for protein engineering

    Curr Opin Struct Biol

    (2004)
  • J.S. Ha et al.

    Design and application of highly responsive fluorescence resonance energy transfer biosensors for detection of sugar in living Saccharomyces cerevisiae cells

    Appl Environ Microbiol

    (2007)
  • J.E. McCombs et al.

    Measuring calcium dynamics in living cells with genetically encodable calcium indicators

    Methods

    (2008)
  • M. Mank et al.

    Genetically encoded calcium indicators

    Chem Rev

    (2008)
  • P. Vincent et al.

    Real-time monitoring of cyclic nucleotide signaling in neurons using genetically encoded FRET probes

    Brain Cell Biol

    (2008)
  • M.L. Rives et al.

    Crosstalk between GABA(B) and mGlu1a receptors reveals new insight into GPCR signal integration

    EMBO J

    (2009)
  • Cited by (152)

    • Fluorescent proteins

      2023, Theoretical and Computational Photochemistry: Fundamentals, Methods, Applications and Synergy with Experimental Approaches
    View all citing articles on Scopus
    View full text