Visualizing Biochemical Activities in Living Cells Through Chemistry
Total Page:16
File Type:pdf, Size:1020Kb
868 CHIMIA 2011, 65, No. 11 NCCR CHEMICAL BIOLOGY doi:10.2533/chimia.2011.868 Chimia 65 (2011) 868–871 © Schweizerische Chemische Gesellschaft Visualizing Biochemical Activities in Living Cells through Chemistry Luc Reymond, Gražvydas Lukinavicˇius, Keitaro Umezawa, Damien Maurel, Matthias A. Brun, Anastasiya Masharina, Karolina Bojkowska, Birgit Mollwitz, Alberto Schena, Rudolf Griss, and Kai Johnsson* Abstract: The development of molecular probes to visualize cellular processes is an important challenge in chemi- cal biology. One possibility to create such cellular indicators is based on the selective labeling of proteins with synthetic probes in living cells. Over the last years, our laboratory has developed different labeling approaches for monitoring protein activity and for localizing synthetic probes inside living cells. In this article, we review two of these labeling approaches, the SNAP-tag and CLIP-tag technologies, and their use for studying cellular pro- cesses. Keywords: Cell biology · Imaging · Protein chemistry · SNAP-tag · Synthetic probes Introduction utility for applications in living cells, such ria was introduced by the group of Roger an approach must fulfill a number of crite- Tsien in 1998.[3] The approach is based on Autofluorescent proteins (AFPs) such as ria: First, the rate of reaction between the the selective chelation of biarsenical fluo- green fluorescent protein (GFP) are only tag and its substrate must be sufficiently rophores by a polypeptide containing four mediocre fluorophores compared to syn- fast at low micromolar concentrations of appropriately spaced cysteine residues, thetic fluorescent dyes. Yet they have be- both reaction partners so that quantitative the so-called tetracysteine tag. In 2003, come the fluorophores of choice for most labeling can be achieved within minutes. our group introduced SNAP-tag, an engi- applications in biology.[1] The reason for Second, the reaction must be of sufficient neered O6-alkylguanine-DNA alkyltrans- this is the possibility to genetically encode selectivity so that in a cellular environment ferase (AGT) that specifically reacts with AFPs and thereby through simple genetic only the tag interacts with the substrate. At O6-benzylguanine (BG) derivatives (Fig. engineering fuse them to other proteins in the same time, the derivatization of the 1).[4] Wild-type AGT repairs mutagenic and living cells. This enables an almost unlim- substrate with different synthetic probes cytotoxic DNA lesions that result from O6- ited number of applications, ranging from should be possible without significantly alkylation of guanine. Repair is achieved the specific fluorescent labeling of proteins affecting speed and selectivity of the re- by transferring the alkyl group to a reac- of interest to the generation of sophisti- action. Furthermore, the substrate should tive cysteine; the alkylated AGT is subse- cated fluorescent sensors for biochemical possess good cell permeability and mini- quently not regenerated. Taking advantage activities. Inspired by the success of AFPs, mal toxicity. The first labeling approach of this unusual DNA repair mechanism, we chemists developed approaches that com- that fulfilled at least partially these crite- generated SNAP-tag by engineering wild- bine the advantages of genetic engineer- ing with the potential offered by synthetic [2] probes. These hybrid approaches are Fig. 1. Labeling based on the specific reaction of a polypep- mechanism of SNAP- tide with a synthetic substrate; derivatizing tag (a) and CLIP-tag such a substrate with a synthetic probe and (b) fusion proteins. expressing the polypeptide (often referred to as a tag) in fusion with a protein of inter- est results in a specific targeting of the probe to the protein of interest. To be of practical *Correspondence: Prof. Dr. K. Johnsson Institute of Chemical Sciences and Engineering NCCR Chemical Biology École Polytechnique Fédérale de Lausanne CH-1015 Lausanne E-mail: [email protected] NCCR CHEMICAL BIOLOGY CHIMIA 2011, 65, No. 11 869 type AGT for an efficient reaction with O O BG derivatives carrying synthetic probes, (a) O O 2- O O leading to a covalently labeled tag. SNAP- CO CO 2+ 2 2 O Ca O tag is a monomeric protein of 182 residues O2C N N CO2 N N (20 kDa)[5] that can be efficiently labeled O O O O in living cells and even in vivo. SNAP-tag Localized Ca2+ Localized Protein Protein has become a widely accepted tool in life Fluorophore Fluorophore sciences, with recent applications ranging S SNAP S SNAP O N Linker O N Linker from measurement of protein half-lives in H Change in H animals to superresolution microscopy and Fluorescence single molecule spectroscopy.[6] To enable the specific and simultane- (b) - - (c) - - ous labeling of two different proteins of CO2 CO2 CO2 CO2 - - - - interest with different probes, our group O2C NNCO2 O2C NNCO2 introduced in 2008 the CLIP-tag.[7] CLIP- O O O O tag was generated by engineering the sub- strate specificity of SNAP-tag so that it NH O H - N N 2 N N O OOC N N specifically reacts with O -benzlycytosine H B N N F F O N N N N (BC) derivatives. SNAP-tag and CLIP-tag H2N N NH O H H N N N fusion proteins can be specifically and si- O 2 H multaneously labeled with BG and BC de- BG-Indo-1 BOCA rivatives in cells. One of the applications of such a double labeling is the charac- Fig. 2. (a) Localizable calcium indicators. (b) Structure of BG-Indo-1. (c) Structure of BODIPY- terization of protein–protein interactions based calcium indicator (BOCA). through Förster resonance energy transfer (FRET).[8] Acceptance of chemistry-based tech- SNAP-tag fusion proteins in living cells formed with UV light from a dark to a highly nologies in the life science community (Fig. 2a).[12] Indo-1 is a useful synthetic fluorescent state, and variants of Eos, whose depends on the commercial availability of Ca2+ indicator whose emission spectrum emission and excitation wavelength can be the necessary reagents. Fortunately, a large depends on [Ca2+] and that therefore can be red-shifted with a light pulse. pFPs have be- number of different substrates for labeling used for ratiometric measurements.[13] To come important tools in biological research of SNAP-tag and CLIP-tag fusion proteins localize Indo-1 in living cells we have syn- as they permit to highlight subpopulations of are distributed by New England Biolabs, thesized a BG-Indo-1 derivative that can be a given protein with optical methods in liv- enabling biologists with no expertise in coupled to SNAP-tag fusion proteins and ing cells. An alternative approach to generate synthetic chemistry to apply these tech- retains its Ca2+-sensing ability (Fig 2b).[12a] photosensitive proteins is the specific label- nologies to their research. By discussing Local Ca2+ sensing was demonstrated in ing of proteins with photosensitive synthetic recent work from our laboratory, we will cultured primary muscle cells of mice fluorophores. We have recently introduced a highlight how SNAP-tag and CLIP-tag can expressing a nucleus-localized SNAP-tag generally applicable strategy for the genera- be exploited to study a wide variety of dif- fusion. While ratiometric Ca2+ indicators tion of photoactivatable and photoconvert- ferent problems in biology. are useful for determining absolute [Ca2+], ible fluorescent probes that can be selective- they are less sensitive than intensity-based ly coupled to SNAP-tag fusion proteins in synthetic fluorescent indicators. We there- living cells (Fig. 3a).[15] The photosensitivity Localizable Calcium Indicators fore introduced a synthetic Ca2+ indicator of the probes is based on photocleavage of a based on a BODIPY fluorophore that linker between a quencher or a second fluo- Critical for the role of calcium (Ca2+) shows a large, 180-fold fluorescence in- rophore (the acceptor) and the fluorophore as a second messenger is a precise spatial crease upon Ca2+ binding when coupled to of interest (the donor). Prior to photocleav- and temporal control of its concentration in SNAP-tag fusion proteins (Fig. 2c).[12b] We age, FRET from the donor to the acceptor cells.[9] Local Ca2+ concentrations ([Ca2+]) further demonstrated how the SNAP-tag- quenches the fluorescence of the donor; af- in the cell can rise within milliseconds by bound sensor can be used to sense changes ter photocleavage the donor becomes highly orders of magnitudes. There are two main in [Ca2+] in the nuclei and in the cytosol fluorescent and remains attached to SNAP- classes of fluorescent probes to measure of live CHO-K1 cells. Our hybrid indi- tag (Fig. 3a). Following this design princi- fluctuations in [Ca2+]: synthetic Ca2+ indi- cators combine the spatial specificity of ple, photoactivatable versions of fluorescein cators such as Fura-2 and Fluo-4 or AFP- biosensors with the fast kinetics and high and Cy3 as well as a photoconvertible Cy5- based Ca2+ indicators such as cameleon and dynamic range of small synthetic indica- Cy3 probe were synthesized and coupled GCamP2.[10] AFP-based sensors can be tors. These features should make them a to selected proteins on the cell surface, in genetically targeted to an area of interest valuable addition to the existing methods the cytosol and in the nucleus of cells. We in cells, but have slower response time and to sense [Ca2+] in living cells. employed the photoactivatable Cy3 probe to lower sensitivity than synthetic indicators. measure the mobility of cell surface recep- In contrast, synthetic indicators lack the tors. The approach can be extended to a large ability to be specifically localized in cells. Photoactivatable Probes for variety of fluorophores and thereby estab- The targeting of synthetic Ca2+ indicators Protein Labeling lishes a generally applicable strategy for the through coupling to localized protein tags generation of photosensitive and localizable potentially combines the advantages from Photosensitive fluorescent proteins fluorophores with tailor-made properties.