Métodos indirectos

Fluorescent proteins (FPs) are widely used as in vivo reporter molecules and are

available in multiple colors spanning almost the entire visible light spectrum (Tsien, 2009). Genetically fused to any target protein, FPs offer a powerful tool to study

protein localization, interactions and dynamics in vivo using photobleaching

techniques such as fluorescence recovery after photobleaching (FRAP), fluorescence loss in photobleaching (FLIP) or fluorescence resonance energy transfer (FRET) (Belmont, 2001; Misteli, 2001; Lippincott-Schwartz et al., 2003; van Royen et al., 2009b). The diversity of FPs has dramatically expanded from the prototypical green fluorescent protein (GFP) (Shimomura et al., 1962; Prasher et al., 1992; Chalfie et al., 1994) isolated from the jellyfish Aequorea Victoria, to engineered mutants (enhanced eGFP) and additional spectral variants (cyan CFP, yellow YFP, etc.) thereof (reviewed in (Tsien, 2005)). The discovery of a red fluorescent protein

(DsRed) in the coral Discosoma sp. provided a new red-shifted reporter tool with

better spectral separation from cellular autofluorescence and allowed multicolor tracking of fusion proteins in one cell (Matz et al., 1999; Wall et al., 2000). However, the obligate tetramerization of DsRed caused serious problems for its use in live-cell imaging. Subsequent mutagenesis of the red progenitor has resulted in several monomeric red FPs (mRFP1, mCherry, mOrange, mPlum, etc) (Campbell et al., 2002; Shaner et al., 2004; Wang et al., 2004; Tsien, 2005; Shu et al., 2006). These improved red FPs are characterized by higher brightness and photostability, complete chromophore maturation and promise a wide variety of features for biological imaging and multicolor labeling (Mizuno et al., 2001; Zhang et al., 2002; Shaner et al., 2005). Very recently, an infrared-fluorescent protein (IFP) with excitation and emission maxima of 684 and 708 nm, respectively, has been engineered, that allows penetration of excitation light in tissues and thus is suitable for whole-body imaging (Shu et al., 2009).

INTRODUCTION

- 28 -

Fluorescence recovery after photobleaching (FRAP)-technique to measure

protein kinetics in vivo

Photobleaching techniques, such as fluorescence recovery after photobleaching (FRAP), fluorescence loss in photobleaching (FLIP) or fluorescence resonance energy transfer (FRET), are powerful methods to explore protein localization, interactions and dynamics in vivo (Periasamy, 2001; Lippincott-Schwartz et al., 2003; van Royen et al., 2009b). Two decades ago photobleaching methods were developed to detect diffusion characteristics in living cells (Axelrod et al., 1976). Later on, the experimental approach was extended to study binding of lipid proteins in the plasma membrane, cytoskeleton dynamics and nucleocytoplasmic protein shuttling (Edidin et al., 1976; Amato and Taylor, 1986; Koster et al., 2005). Today, fluorescence imaging is commonly used and is expanding rapidly because of synergistic advancements in fluorescent protein tags, instrumentation and data analysis, enabling single-molecule detection, multi-protein and live-cell imaging (Giepmans et al., 2006). Especially, advanced confocal laser scanning systems enable to measure protein dynamics via FRAP or FLIP in vivo (Stephens and Allan, 2003; van Royen et al., 2009a).

A FRAP experiment is based on the irreversible photobleaching of fluorescent fusion proteins in a selected compartment within the cell nucleus by a high-powered focused laser beam (Figure 11). Subsequent diffusion of surrounding non-bleached fluorescent molecules into the bleached area leads to recovery of fluorescence over time by displacing the bleached molecules until an equilibrium is reached (steady-state) (Figure 11 A). Complete recovery of fluorescence will only occur, if the underlying entire protein population is dynamically exchanged and no immobile population is present (Phair et al., 2004a; Beaudouin et al., 2006). Comparable kinetic parameters can be extracted from the normalized recovery curve by the relative measures of the half time of recovery (t1/2) and the mobile fraction (MF)

(Figure 11 C) (Lippincott-Schwartz et al., 2001; Reits and Neefjes, 2001).

Furthermore, FRAP analyses and subsequent kinetic modeling yield qualitative and quantitative information about the number of binding states and the binding strength of molecular interactions (Carrero et al., 2003; Sprague and McNally, 2005; McNally, 2008; Dobay et al., in preparation). Hence, protein mobility reflects the current function in the cell.

INTRODUCTION

- 29 -

Figure 11: Representative FRAP experiment from image acquisition to data evaluation.

(A) Half nucleus FRAP series of a C2C12 cell, transiently expressing a GFP-fusion protein (Dnmt1). Selected pre-and postbleach frames show time-dependent fluorescence recovery of the fusion protein until a steady state equilibrium is reached after 80 seconds (s). Rectangle indicates the bleached region of interest (ROI). Bar: 5 µm. (B) ROI and set of compartments necessary for subsequent normalization and data extraction as shown in C. Quantitative evaluation of the FRAP experiment (C). Fluorescence intensities in the bleached and unbleached region were measured over time and normalized to correct for bleaching during image acquisition, nuclear import, bleaching depth and background. Triple normalized (*) data take into account the loss of fluorescence recovery occurring during the last bleach and the first postbleach time point (Dobay et al., in preparation). F∞ indicates the plateau level of fluorescence reached after complete equilibration and indicates the immobile / mobile fraction; t1/2 marks the halftime of fluorescence recovery. Notably, as the half of the nucleus was bleached, convergence of the bleached and unbleached curves occurs at 0.5 of relative intensity and indicates only a little immobile, stably bound fraction (<6%).

Within the eukaryotic cell, protein mobility and dynamic exchange is of key importance for many cellular functions and essential to provide cellular plasticity and efficient responses to external signals (Gorski et al., 2006). Indeed, several nuclear factors display a high degree of mobility (Phair and Misteli, 2000; Phair et al., 2004b; Misteli, 2005). Transient binding to rather immobile structures such as chromatin, or stationary enzyme complexes like the replication machinery, decreases protein mobility (Dundr and Misteli, 2001; Misteli, 2001). Notably, the mobility of one and the same protein can differ drastically dependent on its current function or binding partners. One example is PCNA that forms a trimeric ring and stably slides along the DNA double strand, acting as central loading platform for interacting factors in DNA replication and repair (Mortusewicz et al., 2005; Sporbert et al., 2005; Moldovan et al., 2007). While one fraction of nucleoplasmic PCNA molecules reveals high mobility in its unengaged state, a second fraction becomes temporarily immobilized at DNA

INTRODUCTION

- 30 -

replication or repair sites, the sites of action (Sporbert et al., 2002; Essers et al., 2005). Hence, the dynamic properties of a protein are crucial for determining the function within the cell. However, photobleaching techniques have still some drawbacks since bleaching and imaging of the fluorescent fusion proteins by intensive laser radiation might induce DNA damage and could affect cell viability (Dobrucki et al., 2007). In addition, the significant size of the fused fluorescent protein tag might interfere with protein function, localization or interactions.

Direct visualization of Dnmt1 enzyme activity in living cells

The catalytic activity of Dnmt1 can be monitored directly in vivo via a trapping assay (Schermelleh et al., 2005; Schermelleh et al., 2008). This method is based on the application of the nucleoside analog 5-aza-2′-deoxycytidine (5-aza-dC), which gets incorporated into the newly synthesized DNA during replication and serves as mechanism-based inhibitor. When Dnmt1 undertakes the methyl group transfer reaction at the 5-aza-dC residue, an irreversible covalent complex is formed and Dnmt1 gets immobilized (trapped) at replication foci (RF), the site of action. This time-dependent immobilization of GFP-Dnmt1 (trapping rate) can be measured by FRAP analyses and reflects the enzymatic activity of the fusion protein. Applying this method, one can study Dnmt1 activity and binding preference to its natural substrate in the physiological environment, a living cell. Furthermore, this assay allows the comparison of wt GFP-Dnmt1 and mutants thereof concerning subnuclear targeting and postreplicative methylation efficiency.

INTRODUCTION

- 31 -