Abstract
Single-molecule localization and tracking techniques have contributed towards observing the spatiotemporal organization of proteins in the plasma membrane. Individual proteins labeled with fluorescent nanoparticles (FNPs) can be imaged over long time with ultrahigh spatial and temporal resolution. A key challenge for the biophysical application of FNPs, however, is to site-specifically target the nanoparticles to proteins in living cells. For selective labeling of cell surface proteins with FNPs, biomolecules such as antibodies and streptavidin have been employed as well as chemical recognition based on immobilized transition-metal ions. These recognition units, however, are not compatible with FNP targeting to proteins inside living cells, because the structural integrity of antibodies is often affected by the reducing conditions in the cytoplasm, streptavidin is blocked with endogenous biotin, and transition-metal ions are coordinated by cysteine-rich proteins. Intracellular FNP targeting is furthermore challenging, because blocking of nonspecific binding and washing out of nonbound FNPs is not possible in intact cells. Here, we aimed to establish a highly specific and efficient approach for FNP targeting inside live cells, which overcomes these particular challenges. As a biochemical recognition system compatible with the cytoplasm, we employed an enzymatic covalent labeling approach based on the HaloTag. This engineered dehalogenase irreversibly reacts with a chlorohexane moiety (HaloTag ligand, HTL) attached to fluorescent dyes and other probes. This highly specific reaction has been exploited for protein labeling in live cells. We attempted functionalization of FNPs with HTL through maleimide/thiol-chemistry as well as by using different biotin derivatives. However, neither significant specific binding to the immobilized HaloTag was observed in vitro, nor efficient targeting upon microinjection into live cells expressing HaloTag fusion proteins using a microcapillary. For this reason, we characterized the association kinetics of different derivatives of the HTL in more detail using realtime surface-sensitive detection by simultaneous reflectance interference (RIF) and total internal reflection fluorescence spectroscopy (TIRFS) detection. For this purpose, purified HaloTag with a His-tag (His= histidine, HaloTag-H12) was site-specifically immobilized on a polyethylene glycol (PEG) polymer brush functionalized with tris(nitrilotriacetic acid), tris-NTA, and binding of fluorescent substrates was monitored in real time (see Figure S1 in the Supporting Information). Rapid binding of HTL conjugated with the fluorescent dye AlexaFluor 488 (HTL) was detected by TIRFS (see Figure S1 in the Supporting Information), yielding a reaction rate constant of 1 10m 1 s . To directly compare the reaction rate constants of fluorescent and nonfluorescent HTL conjugates (see Scheme S1 in the Supporting Information), a competition assay was established with HTL as a fluorescent tracer (see Figure S2 in the Supporting Information). These rate constants are summarized in Table S1 in the Supporting Information. Strikingly, a substantially faster reaction rate constant of 1 10m 1 s 1 was observed for HTL conjugated with tetramethylrhodamine (HTL), which is similar to the published rate constant of this reaction measured in solution (2.7 10m 1 s ). Surprisingly, an elongated ethylene glycol linker substantially reduced the reaction rate constant. Even slower rate constants were obtained for biotinylated and for unmodified HTL (around 10m 1 s ). These results suggested that the conjugated fluorescence dyes play a critical role for the association kinetics, probably by stabilizing the noncovalent enzyme–substrate complex, by hydrophobic interactions, prior to the ester formation by reaction with D106 in the binding pocket of the HaloTag. Strikingly, engineering of the HaloTag from the original dehalogenase involved incorporation of hydrophobic residues in the proximity of the reactive site. Based on the observation that hydrophobic as well as positively charged residues increase the reaction rate constant of HTL derivatives, we implemented a novel approach for surface functionalization with the HTL based on click chemistry using commercially available dibenzocyclooctynelike (DBCO) derivatives and azide-functionalized HTL (Figure 1a and Scheme S2 in the Supporting Information). Thus a hydrophobic moiety was integrated to HTL similar to the HTL-dye conjugates. The reaction kinetics of the HTL derivative obtained by this reaction (clickHTL) was compared with other HTL-derivatives of the competitive binding [*] D. Lise, Dr. C. You, Prof. Dr. J. Piehler Division of Biophysics, Department of Biology Universit t Osnabr ck, Barbarastrasse 11 49076 Osnabr ck (Germany) E-mail: piehler@uos.de Homepage: http://www.biologie.uni-osnabrueck.de/Biophysik/ Piehler/
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have