Biotechnology Journal
Biotechnol. J. 2014, 9
DOI 10.1002/biot.201400026
www.biotechnology-journal.com
Technical Report
Click chemistry for the conservation of cellular structures and fluorescent proteins: ClickOx Anna Löschberger, Thomas Niehörster and Markus Sauer Department of Biotechnology and Biophysics, Julius-Maximilians-Universität Würzburg, Biozentrum, Würzburg, Germany
Reactive oxygen species (ROS), including hydrogen peroxide, are known to cause structural damage not only in living, but also in fixed, cells. Copper-catalyzed azide–alkyne cycloaddition (click chemistry) is known to produce ROS. Therefore, fluorescence imaging of cellular structures, such as the actin cytoskeleton, remains challenging when combined with click chemistry protocols. In addition, the production of ROS substantially weakens the fluorescence signal of fluorescent proteins. This led us to develop ClickOx, which is a new click chemistry protocol for improved conservation of the actin structure and better conservation of the fluorescence signal of green fluorescent protein (GFP)-fusion proteins. Herein we demonstrate that efficient oxygen removal by addition of an enzymatic oxygen scavenger system (ClickOx) considerably reduces ROS-associated damage during labeling of nascent DNA with ATTO 488 azide by Cu(I)-catalyzed click chemistry. Standard confocal and super-resolution fluorescence images of phalloidin-labeled actin filaments and GFP/yellow fluorescent protein-labeled cells verify the conservation of the cytoskeleton microstructure and fluorescence intensity, respectively. Thus, ClickOx can be used advantageously for structure preservation in conventional and most notably in super-resolution microscopy methods.
Received 14 JAN 2014 Revised 07 FEB 2014 Accepted 21 FEB 2014
Supporting information available online
Keywords: Actin cytoskeleton · Click chemistry · dSTORM · Reactive oxygen species · Super-resolution imaging
1 Introduction Due to fast reaction kinetics and bioorthogonality, coppercatalyzed azide–alkyne cycloaddition has evolved as a powerful standard method for conjugation, immobilization, and purification of biomolecules [1–4]. Whereas proteins can be labeled specifically with genetically encoded tags, such as the green fluorescent protein (GFP), click chemistry calls when smaller and less structurally per-
Correspondence: Prof. Markus Sauer, Department of Biotechnology and Biophysics, Julius-Maximilians-Universität Würzburg, Biozentrum, Am Hubland, 97074 Würzburg, Germany E-mail: [email protected] Abbreviations: EdU, 5-ethynyl-2’deoxyuridine; eYFP, enhanced yellow fluorescent protein; GFP, green fluorescent protein; PALM, photoactivated localization microscopy; ROS, reactive oxygen species; dSTORM, direct stochastic optical reconstruction microscopy
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
turbing labels are required. On the other hand, it is known that Cu(I) generates reactive oxygen species (ROS) from O2, causing toxicity in live-cell experiments [5]. Even in fixed-cell experiments, using, for example, the commercially available Click-iT® Edu Imaging Kit from Life Technologies for fluorescence imaging of 5-ethynyl-2´-deoxyuridine (EdU)-labeled DNA [6], Cu(I)-induced structural damage of the cytoskeleton has to be accepted. This is because actin filaments are seriously affected by ROS, such as hydroxyl radicals [7–9]. In addition, fluorescent proteins readily lose their fluorescence upon exposure to ROS [10]. To address these implications, alkynes activated by ring strains were introduced that did not require Cu(I) [11]. However, strained cyclooctynes react much slower with azides than classical terminal alkynes requiring the presence of Cu(I) [12, 13]. Alternatively, click reactions with tris(hydroxypropyltriazolyl) methylamine (THPTA) and tetrazine derivatives were introduced [14, 15]. For
1
Biotechnology Journal
Biotechnol. J. 2014, 9 www.biotecvisions.com
www.biotechnology-journal.com
example, THPTA accelerates cycloaddition and protects cells from damage by oxidative agents produced by the Cu(I)-catalyzed reduction of oxygen by ascorbate, which is required to maintain the metal in the active +1 oxidation state. To minimize ROS generation and associated damage of cellular structures and fluorescent proteins, we developed a new Cu(I)-catalyzed click chemistry protocol, termed ‘ClickOx’, that efficiently removes oxygen. An elegant method for oxygen removal is the addition of enzymatic or chemical oxygen-scavenger systems, which allow sustainable depletion of a solution in the presence of continuous oxygen intake. Such oxygen-depleting systems have found widespread use in many biophysical studies, including single-molecule fluorescence detection [16–20]. Herein, we implemented an enzymatic system that consists of glucose oxidase, β-D-glucose, and catalase [16] in our ClickOx protocol. We demonstrate that ClickOx considerably reduces ROS-associated damage during labeling of nascent DNA with ATTO 488 azide by Cu(I)-catalyzed click chemistry. Standard confocal and super-resolution fluorescence images of phalloidin-labeled actin filaments and GFP/enhanced yellow fluorescent protein (eYFP)-labeled cells verify the conservation of the cytoskeleton microstructure and fluorescence intensity, respectively.
2 Materials and methods 2.1 Preparation of cells C2C12 and PC3-GFP cells were cultured in DMEM (D6546, PAA) with 10% FCS, 1% penicillin/streptomycin, 1 mM L-glutamine (PAA). HEK293 channelrhodopsin 2-eYFP (ChR2-eYFP) cells were grown in the same medium with 10 μg/mL blasticidin and 150 μg zeocin™ (both from Life Technologies) for selection. For the experiments, cells were grown overnight on cover slips or in Lab Tek II Chambered Coverglass (Nunc) at 37°C and 5% CO2. The next day, EdU (Life Technologies) was added (1:1000) for 24 h. The HEK293 ChR2-eYFP cell line was activated with tetracyclin for expression of the eYFP construct.
2.2 Fixation and staining of cells For conventional microscopy, cells were rinsed in PBS (37°C), fixed for 15 min at 37°C with 4% formaldehyde/ PBS, rinsed in PBS and permeabilized for 8 min in 0.1% Triton X-100/PBS (PC3-GFP: 0.05% Triton X-100/PBS for 2–3 min) at room temperature. Cells were washed again for at least 5 min in PBS and incubated with 100 nM ATTO 655 phalloidin or 25 nM ATTO 520 phalloidin in PBS. After a washing step in PBS (at least 10 min), cells were clicked with Alexa Fluor 647 or ATTO 488 azide (protocols see below), washed for 10 min in PBS, post-fixed in 4%
2
formaldehyde/PBS for 5 min, washed again in PBS, and embedded in Mowiol (Roth). For direct stochastic optical reconstruction microscopy (dSTORM) experiments, C2C12 cells were prepared as previously described [21]. Cells were rinsed in PBS (37°C), fixed, and permeabilized for 1–2 min in fixation buffer (0.3% glutaraldehyde and 0.25% Triton X-100 in CB-Buffer). A fixation step of 10 min with 2% glutaraldehyde and a short washing step in PBS followed. Then cells were incubated for 7 min in 0.1% sodium borohydride/ PBS and washed again in PBS. They were incubated at 4°C for 48 h with 66 nM (0.2 U) Alexa 647 phalloidin (Life Technologies) in PBS.
2.3 Click reactions Conventional click reactions were performed using click reagents delivered by the provider, following the protocol given in [6], using 100 mM Tris/HCl (pH 8.5), 100 mM ascorbic acid, 1 mM CuSO4, and Alexa Fluor 647 azide (1:2000, Life Technologies) or ATTO 488 azide (2 μM, ATTO-TEC) for 10 min. The improved ClickOx reaction contained an additional oxygen-scavenger system with 1% glucose, 1% glycerol (stock solution: 10% glucose, 10% glycerol), 2 U glucose oxidase, and 40 U catalase (1:150 from stock solution). The oxygen-scavenger enzyme stock solution contained 25 mM KCl, 22 mM Tris/HCl pH 7, 4 mM TCEP (used optionally for storage), 2000 U/mL glucose oxidase (Sigma: G2133-50KU), 40000 U/mL catalase (Sigma: C1000), and 50% glycerol. After a short washing step in PBS (rinsing) a 5 min incubation in 10 mM EDTA/ PBS followed (only for ClickOx). The commercial click reaction was performed according to the data sheet of the supplier (Life Technologies: C10340), but with the same dye concentration and incubation time as the other protocols.
2.4 dSTORM and wide-field fluorescence microscopy C2C12 cells were grown in LabTek II Chambered Coverglass and stained as described above. LabTek wells were washed with PBS for 5 min and filled with switching buffer (PBS, pH 7.4 containing enzymatic oxygen scavenger and 100 mM mercaptoethylamine (Sigma)) [16]. The dSTORM setup was previously described [22]. Briefly, it used an inverted microscope (Olympus IX-71) equipped with an oil-immersion objective (×60, NA 1.45; Olympus) and a 639 nm laser Genesis MX 639-1000 (Coherent) for excitation of Alexa 647 and ATTO 655. 30 000 frames were measured with a frame rate of 105 Hz at excitation intensities of 1–3 kW cm-2 using total internal reflection fluorescence (TIRF) microscopy. Fluorescence light was separated and cleaned from excitation light by a beam splitter (FF-425/532/656-Di01) and emission filter (Em01R442/514/647-25, Semrock), respectively, and imaged on
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Biotechnology Journal
Biotechnol. J. 2014, 9
www.biotechnology-journal.com
www.biotecvisions.com
an electron-multiplying charge-coupled device (EMCCD) camera (iXON Ultra, Andor). For conventional wide-field microscopy, the same setup was used at lower irradiation intensities. The laser Genesis MX 514-500 STM (Coherent) was used for excitation of ATTO 520 and eYFP. The laser iBeam smart 488-S (Toptica Photonics) was used for excitation of GFP. Here another beam splitter (FF403/497/574-Di01) and emission filter (FF01-433/517/613, Semrock) were used.
2.5 Confocal fluorescence microscopy For confocal images, a LSM 780 microscope from Zeiss was used. The settings were not changed during a series of measurements to guarantee comparable results. The argon multiline laser was used for excitation of ATTO 488 (488 nm) and eYFP (514 nm), the helium–neon laser (633 nm) was used for excitation of ATTO 655 and Alexa 647. The following objectives were used: a Plan-Apochromat 20×/0.8 M27 for overviews and a C-Apochromat 63×/1.20 W Korr M27 for detailed images.
2.6 Intensity analysis For intensity comparison of wide-field images, the maximum fluorescence intensity of thick actin fibers (brightest pixel) was determined for each image. For most click protocols, 60–70 images were used to calculate the mean of the maximum fluorescence intensity and the standard error of the mean. Due to the clearly visible fluorescence breakdown after the commercial click reaction, only 10 images were taken from these samples. The average eYFP intensities were calculated from several confocal overview images, each of which contained 60–200 cells. ImageJ was used to discriminate cells from cover glass areas by threshold selection. Hence, the whole intensity information of each cell could be used to calculate the mean intensity of eYFP.
3 Results To demonstrate the effectiveness of ClickOx, we incorporated EdU into nascent DNA in proliferating C2C12 cells and labeled the DNA with ATTO 488 azide using Cu(I)catalyzed click chemistry in the presence and absence of oxygen [4, 23]. EdU incorporation for DNA labeling reduces the assay time and improves the work flow compared with the traditional method using antibody-based bromodeoxyuridine detection and was already successfully used for super-resolution fluorescence imaging of chromosomal DNA by dSTORM [22, 24, 25]. To investigate ROS-induced damage of the standard click chemistry protocol on cellular structures, we labeled the actin skeleton with phalloidin ATTO 655 before application of the click reaction buffer containing 100 mM Tris/HCl
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. Fluorescence imaging of nascent DNA (green) and actin filaments (red) using ClickOx. (A) Average maximum fluorescence intensities of about 60 images per protocol measured for thick actin fibers labeled with phalloidin ATTO 655 using different click chemistry protocols. Error bars indicate the standard error of the mean (n = 60 (b–d, f), n = 30 (e), n = 10 (–)). (B–F) Confocal fluorescence images of C2C12 cells labeled with phalloidin ATTO 655. DNA has been labeled with EdU and ATTO 488 azide. (B) Actin labeling without click reaction (control). (C, D) Actin and DNA labeling. Click reaction with 1 mM CuSO4 under standard and ClickOx conditions. (E, F) Application of 5 mM CuSO4 emphasizes the positive effect of ClickOx on preservation of the actin skeleton compared with the standard protocol. Data in (A) are also given for a commercially available labeling kit. Scale bar, 25 μm.
(pH 8.5), 100 mM ascorbic acid, and 1 mM CuSO4 in the absence (standard) and presence of additional enzymatic oxygen scavenger (ClickOx). Immediately after the click reaction (10 min), Cu(I) was quenched by addition of 10 mM EDTA to avoid further sample damage. Two-color confocal fluorescence images clearly show that ClickOx preserves the actin skeleton compared with a standard click chemistry labeling protocol [6] (Fig. 1B–D and Supporting information, Figs. S1–S3). In particular, the fine actin meshwork of C2C12 mouse myoblasts becomes more distinct. This effect is aggravated when higher copper concentrations, such as 5 mM CuSO4, are
3
Biotechnology Journal
Biotechnol. J. 2014, 9 www.biotecvisions.com
www.biotechnology-journal.com
Figure 2. dSTORM images of actin filaments with ClickOx. C2C12 cells were labeled with phalloidin Alexa 647. (A) Control dSTORM image of actin filaments without click reaction. (B) Conserved actin cytoskeleton after ClickOx reaction with oxygen removal (10 min labeling time, 1 mM CuSO4). (C) Remaining actin fibers after standard click chemistry in the presence of oxygen (10 min labeling time, 1 mM CuSO4). Scale bar, 2 μm
used, resulting in severe destruction of fine actin filaments (Fig. 1E and F). With spatial resolution provided by laser scanning microscopy, thick actin fibers appear to be less affected by lower CuSO4 concentrations (compare Fig. 1C and D). On the other hand, the difference in fluorescence intensity measured under standard and ClickOx reaction conditions becomes clearly visible, measuring the average maximum fluorescence intensities of thick actin fibers by wide-field fluorescence microscopy (Fig. 1A). ClickOx reaction conditions increase the average fluorescence intensity of thick actin fibers 4–21-fold in the presence of 1–5 mM CuSO4. These findings demonstrate that ClickOx substantially decreases ROS-induced damaging effects and enables high-quality multicolor fluorescence imaging of molecules labeled by click chemistry together with actin filaments. To confirm these results, we performed experiments with phalloidin Alexa 647 and phalloidin ATTO 520 and observed similar results (Supporting information, Figs. S4 and S5). In addition, we labeled actin filaments after the click reaction to exclude any direct damaging effect of Cu(I) on the organic fluorophore and obtained identical results (Supporting information, Fig. S6).
Figure 3. Fluorescence intensities of HEK 293-Ch2-eYFP cells. (A) Average fluorescence intensities measured for eYFP after a standard click reaction and ClickOx protocol (10 min labeling time, 1 mM CuSO4). Error bars indicate the standard error of the mean. (B, C) Confocal images of eYFP in HEK 293-Ch2-eYFP cells with color-coded fluorescence intensities using standard and ClickOx reaction conditions. Scale bar, 50 μm.
4
To illustrate the conservation effect of ClickOx on the microstructure of the cellular actin skeleton, we performed super-resolution imaging experiments on phalloidin Alexa 647 labeled cells (Fig. 2). The dSTORM images clearly demonstrate that ClickOx preserves the cellular microstructure and enables imaging of the fine actin meshwork that is destroyed under standard click chemistry conditions in the presence of oxygen. We also observed a positive effect of ClickOx on the preservation of GFP and eYFP fluorescence. Herein, we used PC3 and HEK293 cells stably expressing GFP and channelrhodopsin 2-eYFP, respectively, for comparison of fluorescence intensities measured after click reactions in the presence and absence of oxygen. In these experiments, ClickOx-treated cells show a substantially higher GFP and eYFP fluorescence intensity than that of standard samples (Supporting information, Figs. S7 and S8). For example, ClickOx improved the brightness of eYFP in HEK293 cells by up to 67% (Fig. 3).
4 Discussion Our data clearly show that ClickOx conditions preserve ROS-sensitive cellular structures, such as actin filaments, and the fluorescence of fluorescent proteins. This is especially relevant for super-resolution imaging methods, for which extractable structural information is not only determined by the optical resolution of the instrument, but, equally, by the labeling density. Here click chemistry holds an important position because it provides the required high-density labeling with small fluorescent probes [26, 27]. In addition, oxygen removal is also often used in dSTORM experiments to improve the reliability of photoswitching of carbocyanine and other organic dyes [22, 24, 28]. Furthermore, ClickOx is also likely to preserve the fluorescence of photoactivatable fluorescent proteins, such as mEos2, and thus, enable photoactivated localization microscopy [29] in combination with click chemistry labeling protocols. In addition, ClickOx can be used advantageously for structure preservation in single-mole-
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Biotechnology Journal
Biotechnol. J. 2014, 9
www.biotechnology-journal.com
cule localization microscopy methods that induce stochastic photoswitching by reversible coordination of Cu2+ ions to a fluorescent probe [30]. To conclude, our data demonstrate that ClickOx can play an important role in the conservation of high-density labeling for fluorescence imaging with high spatiotemporal resolution. We foresee that other ROS-sensitive structures or proteins may also benefit from ClickOx.
Acknowledgement: We thank L. Pließ and P. Geßner for their support with cell culture, B. Vogel for assistance with the confocal microscope, and U. Terpitz for providing us with the HEK293-ChR2–eYFP cells. We gratefully acknowledge financial support from the Biophotonics Initiative of the Bundesministerium für Bildung und Forschung (BMBF grants 13N11019 and 13N12507). The authors declare no financial or commercial conflict of interest.
5 References [1] Kolb, H. C., Finn, M. G., Sharpless, K. B., Click chemistry: Diverse chemical function from a few good reactions. Angew. Chem. Int. Ed. 2001, 40, 2004–2021. [2] Wang, Q., Chan, T. R., Hilgraf R., Fokin, V. V. et al., Bioconjugation by copper(I)-catalyzed azide–alkyne [3 + 2] cycloaddition. J. Am. Chem. Soc. 2003, 125, 3192–3193. [3] Rostovtsev, V. V., Green, L. G., Fokin, V. V., Sharpless, K. B., A stepwise Huisgen cycloaddition process: Copper(I)-catalyzed regioselective “ligation” of azides and terminal alkynes. Angew. Chem. Int. Ed. 2002, 41, 2596–2599. [4] Palomo, J. M., Click reactions in protein chemistry: From the preparation of semisynthetic enzymes to new click enzymes. Org. Biomol. Chem. 2012, 10, 9309–9318. [5] Hong, V., Steinmetz, N. F., Manchester, M., Finn, M. G., Labeling live cells by copper-catalyzed alkyne–azide click chemistry. Bioconjugate Chem. 2010, 21, 1912–1916. [6] Salic, A., Mitchison, T. J., A chemical method for fast and sensitive detection of DNA synthesis in vivo. Proc. Natl. Acad. Sci. USA 2008, 105, 2415–2420 [7] Lassing, I., Schmitzberger, F., Björnstedt M., Holmgren, A. et al., Molecular and structural basis for redox regulation of beta-actin. J. Mol. Biol. 2007, 370, 331–48. [8] Gaetke, L. M., Chow, C. K., Copper toxicity, oxidative stress, and antioxidant nutrients. Toxicology 2003, 189, 147–163. [9] Fagotti, A., Di Rosa, I., Simoncelli, F., Pipe, R. K. et al., The effects of copper on actin and fibronectin organization in Mytilus galloprovincialis haemocytes. Dev. Comp. Immunol. 1996, 20, 383–391. [10] Alnuami, A. A., Zeedi, B., Qadri, S.M., Ashraf, S. S., Oxyradicalinduced GFP damage and loss of fluorescence. Int. J. Biol. Macromol. 2008, 43,182–186.
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.biotecvisions.com
[11] Jewett, J. C., Bertozzi, C. R., Cu-free click cycloaddition reactions in chemical biology. Chem. Soc. Rev. 2010, 39,1272–1279. [12] Jewett, J. C., Sletten, E. M., Bertozzi, C. R., Rapid Cu-free click chemistry with readily synthesized biarylazacyclooctynones. J. Am. Chem. Soc. 2010, 132, 3688–3690. [13] Uttamapinant, C., Grecian, S., Clarke, S., Singh, U. et al., Fast, cellcompatible click chemistry with copper-chelating azides for biomolecular labeling. Angew. Chem. Int. Ed. 2012, 51, 5852–5856. [14] Hong, V., Presolski, S. I., Ma, C., Finn, M. G., Analysis and optimization of copper-catalyzed azide–alkyne cycloaddition for bioconjugation. Angew. Chem. Int. Ed. 2009, 48, 9879–9883. [15] Devaraj, N. K., Weissleder, R., Hildebrand, S. A. Tetrazine-based cycloadditions: Application to pretargeted live cell labeling. Bioconjugate Chem. 2008, 19, 2297–2299. [16] Benesch, R. E., Benesch, R., Enzymatic removal of oxygen for polarography and related methods. Science 1953, 118, 447–448. [17] Englander, S. W., Calhoun, D. B., Englander, J. J., Biochemistry without oxygen. Biochemistry 1987, 161, 300–306. [18] Harada, Y., Sakurada, K., Aoki, T., Thomas, D. D., Yanagida, T., Mechanochemical coupling in actomyosin energy transduction studied by in vitro movement assay. J. Mol. Biol. 1990, 216, 49–68. [19] Shi, X. H., Lim, J., Ha, T., Acidification of the oxygen scavenging system in single-molecule fluorescence studies: In situ sensing with a ratiometric dual-emission probe. Anal. Chem. 2010, 82, 6132–6138. [20] Vogelsang, J., Kasper, R., Steinhauer, C., Person, B. et al., A reducing and oxidizing system minimizes photobleaching and blinking of fluorescent dyes. Angew. Chem. Int. Ed. 2008, 47, 5465–5469. [21] Small, J. V., Rottner, K., Hahne, P., Anderson, K. I., Visualising the actin cytoskeleton. Microscopy Res. Tec. 1999, 47, 3–17. [22] van de Linde, S., Löschberger, A., Klein, T., Heidbreder, M. et al., Direct stochastic optical reconstruction microscopy with standard fluorescent probes. Nat. Protoc. 2011, 6, 991–1009. [23] Buck, S. B., Bradford, J., Gee, K. R., Agnew, B. J. et al., Detection of S-phase cell cycle progression using 5-ethynyl-2’-deoxyuridine incorporation with click chemistry, an alternative to using 5-bromo2’-deoxyuridine antibodies. Biotechniques 2008, 44, 927–929. [24] Heilemann, M., van de Linde, S., Schüttpelz, M., Kasper, R. et al., Subdiffraction-resolution fluorescence imaging with conventional fluorescent probes. Angew. Chem. Int. Ed. 2008, 47, 6172–6178. [25] Zessin, P., Finan, K., Heilemann, M., Super-resolution fluorescence imaging of chromosomal DNA. J. Struct. Biol. 2012, 177, 344–348 [26] van de Linde, S., Aufmkolk, S., Franke, C., Holm, T. et al., Investigating cellular structures at the nanoscale with organic fluorophores. Chem. Biol. 2013, 20, 8–18. [27] Sauer, M., Localization microscopy coming of age: From concepts to biological impact. J. Cell Sci. 2013, 126, 3505–3513. [28] Leleka, M., Di Nunzio, F., Henriques, R., Charneau, P. et al., Superresolution imaging of HIV in infected cells with FlAsH-PALM. Proc. Natl. Acad. Sci. USA 2012, 109, 8564–8569. [29] Betzig, E., Patterson, G. H., Sougrat, R., Lindwasser, O. W. et al., Imaging intracellular fluorescent proteins at nanometer resolution. Science 2006, 313, 1642–1645. [30] Schwering, M., Kiel, A., Kurz, A., Lymperopoulos, K. et al., Far-field nanoscopy with reversible chemical reactions. Angew. Chem. Int. Ed. 2011, 50, 2940–2945.
5
Biotechnol. J. 2014, 9
DOI 10.1002/biot.201400026
www.biotechnology-journal.com
Technical Report
Click chemistry for the conservation of cellular structures and fluorescent proteins: ClickOx Anna Löschberger, Thomas Niehörster and Markus Sauer Department of Biotechnology and Biophysics, Julius-Maximilians-Universität Würzburg, Biozentrum, Würzburg, Germany
Reactive oxygen species (ROS), including hydrogen peroxide, are known to cause structural damage not only in living, but also in fixed, cells. Copper-catalyzed azide–alkyne cycloaddition (click chemistry) is known to produce ROS. Therefore, fluorescence imaging of cellular structures, such as the actin cytoskeleton, remains challenging when combined with click chemistry protocols. In addition, the production of ROS substantially weakens the fluorescence signal of fluorescent proteins. This led us to develop ClickOx, which is a new click chemistry protocol for improved conservation of the actin structure and better conservation of the fluorescence signal of green fluorescent protein (GFP)-fusion proteins. Herein we demonstrate that efficient oxygen removal by addition of an enzymatic oxygen scavenger system (ClickOx) considerably reduces ROS-associated damage during labeling of nascent DNA with ATTO 488 azide by Cu(I)-catalyzed click chemistry. Standard confocal and super-resolution fluorescence images of phalloidin-labeled actin filaments and GFP/yellow fluorescent protein-labeled cells verify the conservation of the cytoskeleton microstructure and fluorescence intensity, respectively. Thus, ClickOx can be used advantageously for structure preservation in conventional and most notably in super-resolution microscopy methods.
Received 14 JAN 2014 Revised 07 FEB 2014 Accepted 21 FEB 2014
Supporting information available online
Keywords: Actin cytoskeleton · Click chemistry · dSTORM · Reactive oxygen species · Super-resolution imaging
1 Introduction Due to fast reaction kinetics and bioorthogonality, coppercatalyzed azide–alkyne cycloaddition has evolved as a powerful standard method for conjugation, immobilization, and purification of biomolecules [1–4]. Whereas proteins can be labeled specifically with genetically encoded tags, such as the green fluorescent protein (GFP), click chemistry calls when smaller and less structurally per-
Correspondence: Prof. Markus Sauer, Department of Biotechnology and Biophysics, Julius-Maximilians-Universität Würzburg, Biozentrum, Am Hubland, 97074 Würzburg, Germany E-mail: [email protected] Abbreviations: EdU, 5-ethynyl-2’deoxyuridine; eYFP, enhanced yellow fluorescent protein; GFP, green fluorescent protein; PALM, photoactivated localization microscopy; ROS, reactive oxygen species; dSTORM, direct stochastic optical reconstruction microscopy
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
turbing labels are required. On the other hand, it is known that Cu(I) generates reactive oxygen species (ROS) from O2, causing toxicity in live-cell experiments [5]. Even in fixed-cell experiments, using, for example, the commercially available Click-iT® Edu Imaging Kit from Life Technologies for fluorescence imaging of 5-ethynyl-2´-deoxyuridine (EdU)-labeled DNA [6], Cu(I)-induced structural damage of the cytoskeleton has to be accepted. This is because actin filaments are seriously affected by ROS, such as hydroxyl radicals [7–9]. In addition, fluorescent proteins readily lose their fluorescence upon exposure to ROS [10]. To address these implications, alkynes activated by ring strains were introduced that did not require Cu(I) [11]. However, strained cyclooctynes react much slower with azides than classical terminal alkynes requiring the presence of Cu(I) [12, 13]. Alternatively, click reactions with tris(hydroxypropyltriazolyl) methylamine (THPTA) and tetrazine derivatives were introduced [14, 15]. For
1
Biotechnology Journal
Biotechnol. J. 2014, 9 www.biotecvisions.com
www.biotechnology-journal.com
example, THPTA accelerates cycloaddition and protects cells from damage by oxidative agents produced by the Cu(I)-catalyzed reduction of oxygen by ascorbate, which is required to maintain the metal in the active +1 oxidation state. To minimize ROS generation and associated damage of cellular structures and fluorescent proteins, we developed a new Cu(I)-catalyzed click chemistry protocol, termed ‘ClickOx’, that efficiently removes oxygen. An elegant method for oxygen removal is the addition of enzymatic or chemical oxygen-scavenger systems, which allow sustainable depletion of a solution in the presence of continuous oxygen intake. Such oxygen-depleting systems have found widespread use in many biophysical studies, including single-molecule fluorescence detection [16–20]. Herein, we implemented an enzymatic system that consists of glucose oxidase, β-D-glucose, and catalase [16] in our ClickOx protocol. We demonstrate that ClickOx considerably reduces ROS-associated damage during labeling of nascent DNA with ATTO 488 azide by Cu(I)-catalyzed click chemistry. Standard confocal and super-resolution fluorescence images of phalloidin-labeled actin filaments and GFP/enhanced yellow fluorescent protein (eYFP)-labeled cells verify the conservation of the cytoskeleton microstructure and fluorescence intensity, respectively.
2 Materials and methods 2.1 Preparation of cells C2C12 and PC3-GFP cells were cultured in DMEM (D6546, PAA) with 10% FCS, 1% penicillin/streptomycin, 1 mM L-glutamine (PAA). HEK293 channelrhodopsin 2-eYFP (ChR2-eYFP) cells were grown in the same medium with 10 μg/mL blasticidin and 150 μg zeocin™ (both from Life Technologies) for selection. For the experiments, cells were grown overnight on cover slips or in Lab Tek II Chambered Coverglass (Nunc) at 37°C and 5% CO2. The next day, EdU (Life Technologies) was added (1:1000) for 24 h. The HEK293 ChR2-eYFP cell line was activated with tetracyclin for expression of the eYFP construct.
2.2 Fixation and staining of cells For conventional microscopy, cells were rinsed in PBS (37°C), fixed for 15 min at 37°C with 4% formaldehyde/ PBS, rinsed in PBS and permeabilized for 8 min in 0.1% Triton X-100/PBS (PC3-GFP: 0.05% Triton X-100/PBS for 2–3 min) at room temperature. Cells were washed again for at least 5 min in PBS and incubated with 100 nM ATTO 655 phalloidin or 25 nM ATTO 520 phalloidin in PBS. After a washing step in PBS (at least 10 min), cells were clicked with Alexa Fluor 647 or ATTO 488 azide (protocols see below), washed for 10 min in PBS, post-fixed in 4%
2
formaldehyde/PBS for 5 min, washed again in PBS, and embedded in Mowiol (Roth). For direct stochastic optical reconstruction microscopy (dSTORM) experiments, C2C12 cells were prepared as previously described [21]. Cells were rinsed in PBS (37°C), fixed, and permeabilized for 1–2 min in fixation buffer (0.3% glutaraldehyde and 0.25% Triton X-100 in CB-Buffer). A fixation step of 10 min with 2% glutaraldehyde and a short washing step in PBS followed. Then cells were incubated for 7 min in 0.1% sodium borohydride/ PBS and washed again in PBS. They were incubated at 4°C for 48 h with 66 nM (0.2 U) Alexa 647 phalloidin (Life Technologies) in PBS.
2.3 Click reactions Conventional click reactions were performed using click reagents delivered by the provider, following the protocol given in [6], using 100 mM Tris/HCl (pH 8.5), 100 mM ascorbic acid, 1 mM CuSO4, and Alexa Fluor 647 azide (1:2000, Life Technologies) or ATTO 488 azide (2 μM, ATTO-TEC) for 10 min. The improved ClickOx reaction contained an additional oxygen-scavenger system with 1% glucose, 1% glycerol (stock solution: 10% glucose, 10% glycerol), 2 U glucose oxidase, and 40 U catalase (1:150 from stock solution). The oxygen-scavenger enzyme stock solution contained 25 mM KCl, 22 mM Tris/HCl pH 7, 4 mM TCEP (used optionally for storage), 2000 U/mL glucose oxidase (Sigma: G2133-50KU), 40000 U/mL catalase (Sigma: C1000), and 50% glycerol. After a short washing step in PBS (rinsing) a 5 min incubation in 10 mM EDTA/ PBS followed (only for ClickOx). The commercial click reaction was performed according to the data sheet of the supplier (Life Technologies: C10340), but with the same dye concentration and incubation time as the other protocols.
2.4 dSTORM and wide-field fluorescence microscopy C2C12 cells were grown in LabTek II Chambered Coverglass and stained as described above. LabTek wells were washed with PBS for 5 min and filled with switching buffer (PBS, pH 7.4 containing enzymatic oxygen scavenger and 100 mM mercaptoethylamine (Sigma)) [16]. The dSTORM setup was previously described [22]. Briefly, it used an inverted microscope (Olympus IX-71) equipped with an oil-immersion objective (×60, NA 1.45; Olympus) and a 639 nm laser Genesis MX 639-1000 (Coherent) for excitation of Alexa 647 and ATTO 655. 30 000 frames were measured with a frame rate of 105 Hz at excitation intensities of 1–3 kW cm-2 using total internal reflection fluorescence (TIRF) microscopy. Fluorescence light was separated and cleaned from excitation light by a beam splitter (FF-425/532/656-Di01) and emission filter (Em01R442/514/647-25, Semrock), respectively, and imaged on
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Biotechnology Journal
Biotechnol. J. 2014, 9
www.biotechnology-journal.com
www.biotecvisions.com
an electron-multiplying charge-coupled device (EMCCD) camera (iXON Ultra, Andor). For conventional wide-field microscopy, the same setup was used at lower irradiation intensities. The laser Genesis MX 514-500 STM (Coherent) was used for excitation of ATTO 520 and eYFP. The laser iBeam smart 488-S (Toptica Photonics) was used for excitation of GFP. Here another beam splitter (FF403/497/574-Di01) and emission filter (FF01-433/517/613, Semrock) were used.
2.5 Confocal fluorescence microscopy For confocal images, a LSM 780 microscope from Zeiss was used. The settings were not changed during a series of measurements to guarantee comparable results. The argon multiline laser was used for excitation of ATTO 488 (488 nm) and eYFP (514 nm), the helium–neon laser (633 nm) was used for excitation of ATTO 655 and Alexa 647. The following objectives were used: a Plan-Apochromat 20×/0.8 M27 for overviews and a C-Apochromat 63×/1.20 W Korr M27 for detailed images.
2.6 Intensity analysis For intensity comparison of wide-field images, the maximum fluorescence intensity of thick actin fibers (brightest pixel) was determined for each image. For most click protocols, 60–70 images were used to calculate the mean of the maximum fluorescence intensity and the standard error of the mean. Due to the clearly visible fluorescence breakdown after the commercial click reaction, only 10 images were taken from these samples. The average eYFP intensities were calculated from several confocal overview images, each of which contained 60–200 cells. ImageJ was used to discriminate cells from cover glass areas by threshold selection. Hence, the whole intensity information of each cell could be used to calculate the mean intensity of eYFP.
3 Results To demonstrate the effectiveness of ClickOx, we incorporated EdU into nascent DNA in proliferating C2C12 cells and labeled the DNA with ATTO 488 azide using Cu(I)catalyzed click chemistry in the presence and absence of oxygen [4, 23]. EdU incorporation for DNA labeling reduces the assay time and improves the work flow compared with the traditional method using antibody-based bromodeoxyuridine detection and was already successfully used for super-resolution fluorescence imaging of chromosomal DNA by dSTORM [22, 24, 25]. To investigate ROS-induced damage of the standard click chemistry protocol on cellular structures, we labeled the actin skeleton with phalloidin ATTO 655 before application of the click reaction buffer containing 100 mM Tris/HCl
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. Fluorescence imaging of nascent DNA (green) and actin filaments (red) using ClickOx. (A) Average maximum fluorescence intensities of about 60 images per protocol measured for thick actin fibers labeled with phalloidin ATTO 655 using different click chemistry protocols. Error bars indicate the standard error of the mean (n = 60 (b–d, f), n = 30 (e), n = 10 (–)). (B–F) Confocal fluorescence images of C2C12 cells labeled with phalloidin ATTO 655. DNA has been labeled with EdU and ATTO 488 azide. (B) Actin labeling without click reaction (control). (C, D) Actin and DNA labeling. Click reaction with 1 mM CuSO4 under standard and ClickOx conditions. (E, F) Application of 5 mM CuSO4 emphasizes the positive effect of ClickOx on preservation of the actin skeleton compared with the standard protocol. Data in (A) are also given for a commercially available labeling kit. Scale bar, 25 μm.
(pH 8.5), 100 mM ascorbic acid, and 1 mM CuSO4 in the absence (standard) and presence of additional enzymatic oxygen scavenger (ClickOx). Immediately after the click reaction (10 min), Cu(I) was quenched by addition of 10 mM EDTA to avoid further sample damage. Two-color confocal fluorescence images clearly show that ClickOx preserves the actin skeleton compared with a standard click chemistry labeling protocol [6] (Fig. 1B–D and Supporting information, Figs. S1–S3). In particular, the fine actin meshwork of C2C12 mouse myoblasts becomes more distinct. This effect is aggravated when higher copper concentrations, such as 5 mM CuSO4, are
3
Biotechnology Journal
Biotechnol. J. 2014, 9 www.biotecvisions.com
www.biotechnology-journal.com
Figure 2. dSTORM images of actin filaments with ClickOx. C2C12 cells were labeled with phalloidin Alexa 647. (A) Control dSTORM image of actin filaments without click reaction. (B) Conserved actin cytoskeleton after ClickOx reaction with oxygen removal (10 min labeling time, 1 mM CuSO4). (C) Remaining actin fibers after standard click chemistry in the presence of oxygen (10 min labeling time, 1 mM CuSO4). Scale bar, 2 μm
used, resulting in severe destruction of fine actin filaments (Fig. 1E and F). With spatial resolution provided by laser scanning microscopy, thick actin fibers appear to be less affected by lower CuSO4 concentrations (compare Fig. 1C and D). On the other hand, the difference in fluorescence intensity measured under standard and ClickOx reaction conditions becomes clearly visible, measuring the average maximum fluorescence intensities of thick actin fibers by wide-field fluorescence microscopy (Fig. 1A). ClickOx reaction conditions increase the average fluorescence intensity of thick actin fibers 4–21-fold in the presence of 1–5 mM CuSO4. These findings demonstrate that ClickOx substantially decreases ROS-induced damaging effects and enables high-quality multicolor fluorescence imaging of molecules labeled by click chemistry together with actin filaments. To confirm these results, we performed experiments with phalloidin Alexa 647 and phalloidin ATTO 520 and observed similar results (Supporting information, Figs. S4 and S5). In addition, we labeled actin filaments after the click reaction to exclude any direct damaging effect of Cu(I) on the organic fluorophore and obtained identical results (Supporting information, Fig. S6).
Figure 3. Fluorescence intensities of HEK 293-Ch2-eYFP cells. (A) Average fluorescence intensities measured for eYFP after a standard click reaction and ClickOx protocol (10 min labeling time, 1 mM CuSO4). Error bars indicate the standard error of the mean. (B, C) Confocal images of eYFP in HEK 293-Ch2-eYFP cells with color-coded fluorescence intensities using standard and ClickOx reaction conditions. Scale bar, 50 μm.
4
To illustrate the conservation effect of ClickOx on the microstructure of the cellular actin skeleton, we performed super-resolution imaging experiments on phalloidin Alexa 647 labeled cells (Fig. 2). The dSTORM images clearly demonstrate that ClickOx preserves the cellular microstructure and enables imaging of the fine actin meshwork that is destroyed under standard click chemistry conditions in the presence of oxygen. We also observed a positive effect of ClickOx on the preservation of GFP and eYFP fluorescence. Herein, we used PC3 and HEK293 cells stably expressing GFP and channelrhodopsin 2-eYFP, respectively, for comparison of fluorescence intensities measured after click reactions in the presence and absence of oxygen. In these experiments, ClickOx-treated cells show a substantially higher GFP and eYFP fluorescence intensity than that of standard samples (Supporting information, Figs. S7 and S8). For example, ClickOx improved the brightness of eYFP in HEK293 cells by up to 67% (Fig. 3).
4 Discussion Our data clearly show that ClickOx conditions preserve ROS-sensitive cellular structures, such as actin filaments, and the fluorescence of fluorescent proteins. This is especially relevant for super-resolution imaging methods, for which extractable structural information is not only determined by the optical resolution of the instrument, but, equally, by the labeling density. Here click chemistry holds an important position because it provides the required high-density labeling with small fluorescent probes [26, 27]. In addition, oxygen removal is also often used in dSTORM experiments to improve the reliability of photoswitching of carbocyanine and other organic dyes [22, 24, 28]. Furthermore, ClickOx is also likely to preserve the fluorescence of photoactivatable fluorescent proteins, such as mEos2, and thus, enable photoactivated localization microscopy [29] in combination with click chemistry labeling protocols. In addition, ClickOx can be used advantageously for structure preservation in single-mole-
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Biotechnology Journal
Biotechnol. J. 2014, 9
www.biotechnology-journal.com
cule localization microscopy methods that induce stochastic photoswitching by reversible coordination of Cu2+ ions to a fluorescent probe [30]. To conclude, our data demonstrate that ClickOx can play an important role in the conservation of high-density labeling for fluorescence imaging with high spatiotemporal resolution. We foresee that other ROS-sensitive structures or proteins may also benefit from ClickOx.
Acknowledgement: We thank L. Pließ and P. Geßner for their support with cell culture, B. Vogel for assistance with the confocal microscope, and U. Terpitz for providing us with the HEK293-ChR2–eYFP cells. We gratefully acknowledge financial support from the Biophotonics Initiative of the Bundesministerium für Bildung und Forschung (BMBF grants 13N11019 and 13N12507). The authors declare no financial or commercial conflict of interest.
5 References [1] Kolb, H. C., Finn, M. G., Sharpless, K. B., Click chemistry: Diverse chemical function from a few good reactions. Angew. Chem. Int. Ed. 2001, 40, 2004–2021. [2] Wang, Q., Chan, T. R., Hilgraf R., Fokin, V. V. et al., Bioconjugation by copper(I)-catalyzed azide–alkyne [3 + 2] cycloaddition. J. Am. Chem. Soc. 2003, 125, 3192–3193. [3] Rostovtsev, V. V., Green, L. G., Fokin, V. V., Sharpless, K. B., A stepwise Huisgen cycloaddition process: Copper(I)-catalyzed regioselective “ligation” of azides and terminal alkynes. Angew. Chem. Int. Ed. 2002, 41, 2596–2599. [4] Palomo, J. M., Click reactions in protein chemistry: From the preparation of semisynthetic enzymes to new click enzymes. Org. Biomol. Chem. 2012, 10, 9309–9318. [5] Hong, V., Steinmetz, N. F., Manchester, M., Finn, M. G., Labeling live cells by copper-catalyzed alkyne–azide click chemistry. Bioconjugate Chem. 2010, 21, 1912–1916. [6] Salic, A., Mitchison, T. J., A chemical method for fast and sensitive detection of DNA synthesis in vivo. Proc. Natl. Acad. Sci. USA 2008, 105, 2415–2420 [7] Lassing, I., Schmitzberger, F., Björnstedt M., Holmgren, A. et al., Molecular and structural basis for redox regulation of beta-actin. J. Mol. Biol. 2007, 370, 331–48. [8] Gaetke, L. M., Chow, C. K., Copper toxicity, oxidative stress, and antioxidant nutrients. Toxicology 2003, 189, 147–163. [9] Fagotti, A., Di Rosa, I., Simoncelli, F., Pipe, R. K. et al., The effects of copper on actin and fibronectin organization in Mytilus galloprovincialis haemocytes. Dev. Comp. Immunol. 1996, 20, 383–391. [10] Alnuami, A. A., Zeedi, B., Qadri, S.M., Ashraf, S. S., Oxyradicalinduced GFP damage and loss of fluorescence. Int. J. Biol. Macromol. 2008, 43,182–186.
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.biotecvisions.com
[11] Jewett, J. C., Bertozzi, C. R., Cu-free click cycloaddition reactions in chemical biology. Chem. Soc. Rev. 2010, 39,1272–1279. [12] Jewett, J. C., Sletten, E. M., Bertozzi, C. R., Rapid Cu-free click chemistry with readily synthesized biarylazacyclooctynones. J. Am. Chem. Soc. 2010, 132, 3688–3690. [13] Uttamapinant, C., Grecian, S., Clarke, S., Singh, U. et al., Fast, cellcompatible click chemistry with copper-chelating azides for biomolecular labeling. Angew. Chem. Int. Ed. 2012, 51, 5852–5856. [14] Hong, V., Presolski, S. I., Ma, C., Finn, M. G., Analysis and optimization of copper-catalyzed azide–alkyne cycloaddition for bioconjugation. Angew. Chem. Int. Ed. 2009, 48, 9879–9883. [15] Devaraj, N. K., Weissleder, R., Hildebrand, S. A. Tetrazine-based cycloadditions: Application to pretargeted live cell labeling. Bioconjugate Chem. 2008, 19, 2297–2299. [16] Benesch, R. E., Benesch, R., Enzymatic removal of oxygen for polarography and related methods. Science 1953, 118, 447–448. [17] Englander, S. W., Calhoun, D. B., Englander, J. J., Biochemistry without oxygen. Biochemistry 1987, 161, 300–306. [18] Harada, Y., Sakurada, K., Aoki, T., Thomas, D. D., Yanagida, T., Mechanochemical coupling in actomyosin energy transduction studied by in vitro movement assay. J. Mol. Biol. 1990, 216, 49–68. [19] Shi, X. H., Lim, J., Ha, T., Acidification of the oxygen scavenging system in single-molecule fluorescence studies: In situ sensing with a ratiometric dual-emission probe. Anal. Chem. 2010, 82, 6132–6138. [20] Vogelsang, J., Kasper, R., Steinhauer, C., Person, B. et al., A reducing and oxidizing system minimizes photobleaching and blinking of fluorescent dyes. Angew. Chem. Int. Ed. 2008, 47, 5465–5469. [21] Small, J. V., Rottner, K., Hahne, P., Anderson, K. I., Visualising the actin cytoskeleton. Microscopy Res. Tec. 1999, 47, 3–17. [22] van de Linde, S., Löschberger, A., Klein, T., Heidbreder, M. et al., Direct stochastic optical reconstruction microscopy with standard fluorescent probes. Nat. Protoc. 2011, 6, 991–1009. [23] Buck, S. B., Bradford, J., Gee, K. R., Agnew, B. J. et al., Detection of S-phase cell cycle progression using 5-ethynyl-2’-deoxyuridine incorporation with click chemistry, an alternative to using 5-bromo2’-deoxyuridine antibodies. Biotechniques 2008, 44, 927–929. [24] Heilemann, M., van de Linde, S., Schüttpelz, M., Kasper, R. et al., Subdiffraction-resolution fluorescence imaging with conventional fluorescent probes. Angew. Chem. Int. Ed. 2008, 47, 6172–6178. [25] Zessin, P., Finan, K., Heilemann, M., Super-resolution fluorescence imaging of chromosomal DNA. J. Struct. Biol. 2012, 177, 344–348 [26] van de Linde, S., Aufmkolk, S., Franke, C., Holm, T. et al., Investigating cellular structures at the nanoscale with organic fluorophores. Chem. Biol. 2013, 20, 8–18. [27] Sauer, M., Localization microscopy coming of age: From concepts to biological impact. J. Cell Sci. 2013, 126, 3505–3513. [28] Leleka, M., Di Nunzio, F., Henriques, R., Charneau, P. et al., Superresolution imaging of HIV in infected cells with FlAsH-PALM. Proc. Natl. Acad. Sci. USA 2012, 109, 8564–8569. [29] Betzig, E., Patterson, G. H., Sougrat, R., Lindwasser, O. W. et al., Imaging intracellular fluorescent proteins at nanometer resolution. Science 2006, 313, 1642–1645. [30] Schwering, M., Kiel, A., Kurz, A., Lymperopoulos, K. et al., Far-field nanoscopy with reversible chemical reactions. Angew. Chem. Int. Ed. 2011, 50, 2940–2945.
5
