2Institute of Chemical Sciences and Engineering, école Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
O6-alkylguanine-DNA alkyltransferase (AGT) fusion proteins can be specifically and covalently labeled with fluorescent O6-benzylguanine (O6-BG) derivatives for multicolor live cell imaging approaches. Here, we characterize several new BG fluorophores suitable for in vivo AGT labeling that display fluorescence emission maxima covering the visible spectrum from 472 to 673 nm, thereby extending the spectral limits set by fluorescent proteins. We show that the photostability of the cell-permeable dyes BG Rhodamine Green (BG505) and CP tetra-methylrhodamine (CP-TMR) is in the range of enhanced green fluorescent protein (EGFP) and monomeric red fluorescent protein (mRFP), and that BG diethylaminomethyl coumarin (BGDEAC), a derivative of coumarin, is even more stable than enhanced cyan fluorescent protein (ECFP). Due to the increasing number of new BG derivatives with interesting fluorescence properties, such as far-red emission, fluorescence labeling of AGT fusion proteins is becoming a versatile alternative to existing live cell imaging approaches.
Multicolor imaging of proteins to study their function in the context of the living cell has become a fundamental technique in cell biology during the last decade. Besides the well-known genetically encoded fluorescent proteins, new techniques to transfer fluorescence properties on a protein of interest have been developed, such as the biarsenical-tetracysteine system, the nitrilotriacetate (NTA)-chromophore oligohistidine approach, the use of labeling through posttranslational modifications, or the labeling of O6-alkylguanine-DNA alkyltransferase (AGT) fusion proteins (1,2,3,4). These techniques are based on reporter peptides or enzymes that allow for the site-specific incorporation of small synthetic compounds such as fluorophores into the fusion protein.
For labeling AGT fusion proteins, the specific attachment of the synthetic probe is based on the irreversible transfer of the benzyl group from O6-benzylguanine (O6-BG) to a cysteine residue of the human DNA repair protein AGT resulting in a stable thioether bond (5,6). O6-BG derivatives bearing a fluorescent dye at the para-position of the benzyl ring have been used for various imaging applications such as in vivo imaging of dynamic processes, fluorescence pulse chase experiments, as well as fluorescence resonance energy transfer (FRET) applications (4,5,7,8,9). A major advantage of this approach is the availability of a multitude of fluorescent O6-BG derivatives that is constantly increasing due to the ease of their chemical synthesis. Because of the possibility to choose the fluorescent dye that bests suits the experiment, AGT labeling has the potential to overcome some of the spectral limitations and complex photophysics of the fluorescent proteins (10,11).
So far the possibility of the unwanted labeling of endogenous wild-type AGT (wtAGT) present as a DNA repair protein in mammalian cells has been the main drawback of the technique. However, we found that many mammalian cell lines such as PtK2 (used this study) as well as NRK cells (unpublished data) did not show detectable labeling of wtAGT above background (see Supplementary Figure S7 in the supplementary material available online at www.BioTechniques.com). In addition, the recent development of an AGT mutant, MAGT, that is resistant towards the inhibitor CG [N9-cyclopentyl-O6-(4-bromothenyl)guanine] allows researchers to efficiently block labeling of the endogenous, but not the mutant, AGT using CG (12). Besides the resistance toward CG, other advantages of MAGT are that it is smaller in size than wtAGT (182 amino acids instead of 207 amino acids), its affinity toward alkylated DNA has been depleted, and all nonessential cysteine residues have been replaced. This mutant was used therefore throughout this study.
Major limitations for live cell imaging are rapid photobleaching and relatively short emission wavelengths of the available fluorophores. Longer wavelength dyes with emission maxima up to the near-infrared could become valuable tools as these dyes expand the range of options for multicolor detection, especially because fluorescent proteins such as mPlum only go up to an emission maximum of 649 nm until now (13). Furthermore, the red and far-red dyes can be excited with longer less cytotoxic wavelengths and emit at wavelengths longer than the usual sources of cell autofluorescence, resulting in a higher signal-to-noise ratio (14).Materials and Methods MAGT-H2B Construct/Cell Culture
The photostability measurement of the various dyes was performed on living PtK2 cells expressing histone 2B (H2B) tagged at the N terminus with MAGT. After removing the AgeI site in MAGT by a silent point mutation, this AGT mutant was cloned into the pEGFP-C1 vector (BD Biosciences, Palo Alto, CA, USA) by AgeI/BsrGI, resulting in pMAGT-C1. H2B was transferred from pEGFP-H2B (15) into pMAGT-C1 using XhoI/BamHI, resulting in pMAGT-H2B. A monoclonal stable PtK2 cell line expressing MAGT-H2B was produced and maintained as described (16). To generate pH2B-mRFP, the vector backbone and the entire coding sequence for H2B were obtained from pH2B-ECFP and ligated to monomeric red fluorescent protein (mRFP) (17,18). The fusion generated a proline between the two proteins. The photostability measurements of enhanced cyan fluorescent protein (ECFP), enhanced green fluorescent protein (EGFP), and mRFP were performed on PtK2 cells transiently expressing H2B-ECFP, H2B-EGFP, and H2B-mRFP, respectively. For live cell imaging, cells were cultured in Lab Tek™ II chambered coverglasses (No. 1; NalgeNunc International, Rochester, NY, USA) and maintained at 37°C in imaging medium [CO2-independent Dulbecco's modified Eagle medium (DMEM; Invitrogen, Carlsbad, CA, USA) without phenol red, and supplemented with 20% fetal calf serum (FCS), 10 mM glutamine, penicillin/streptomycin] as described (16).