Although the click chemistry-based methodology designed to detect DNA replication (44, 49) and RNA synthesis (45) was introduced only very recently there are already several reports of its applications in cytometry. A variant of this methodology described by Capella et al. (50) also utilizes EdU as a DNA precursor incorporated into live cells. However, in their method, the exposed nucleotide alkyne moiety in DNA instead of being derivatized by Cu(I)-catalyzed azide-alkyne cycloaddition directly with azide-tagged fluorochrome is derivatized with BrdU-tagged azide. The attached BrdU can then be detected immunocytochemically with the BrdU Ab as in the classic procedure of DNA labeling with BrdU developed by Gratzner (20) and Dolbeare et al. (22). While this procedure also does not require a DNA denaturation step, and is therefore compatible with immunocytochemical detection of surface and intracellular proteins, it is more complex in comparison with the Salic and Mitchison method (44) since it has an additional step involving the immunocytochemical detection of BrdU. Furthermore, being a much larger molecule than the complex of azide-tagged fluorochrome, the fluorochrome-tagged BrdU-Ab may encounter greater steric hindrance in attaching to the incorporated precursor.
Using the Salic and Mitchison (44) approach, Yu et al. (51) observed a good correlation between the standard assay of proliferation of spleen lymphocytes based on 3H-thymidine incorporation and EdU incorporation detected by flow cytometry. The authors studied the effect of partial immune reconstitution on the response of lymphocytes from purine nucleoside phosphorylase defective mice after their mitogenic stimulation by CD-3 Abs (51). They concluded that the EdU-click methodology can be used as an alternative to 3H-thymidine incorporation. Different methods of cell fixation and permeabilization to optimize the detection of the incorporated EdU were explored by Hamelik and Krishan (52).
Interesting in vivo studies were recently carried out by Jensen et al. (53) who administered EdU to mice and at different time intervals analyzed both the incorporation of the precursor and the immunophenotype of the keratinocytes. The cells were fixed in a Zinc (Zn) salt-based solution which unlike formaldehyde does not crosslink proteins or nucleic acids. As a result, nucleic acids can be extracted and analyzed by gel electrophoresis and/or subjected to PCR. This approach combines the analytical capabilities of cytometry and molecular biology that collectively can be used to assess DNA replication in subpopulations of cells characterized by a particular immunophenotype and upon cell sorting to extract DNA, RNA, or protein, and then use PCR, RT-PCR, or Western blotting to analyze these subpopulations.
The applicability of the EdU-click methodology in cell kinetics or tracking studies was thoroughly explored in elegant studies by Diermeier-Daucher et al. (54). Using several cell lines the authors studied the effect of EdU on DNA synthesis, cell cycle progression and viability, comparing it with BrdU. They observed rather minor differences between BrdU and EdU. They concluded that as with BrdU, EdU could be used not only for pulse-labeling but also for continuous labeling in dynamic studies of cell cycle progression. However, noticing the increased sensitivity of breast cancer lines BT474 and SK-BR-3 to EdU, the authors cautioned that any potential cytostatic/cytotoxic effect of this precursor on a particular cell type should be examined before using this approach to assess cell cycle progression (54, 55). The application of the EdU-click methodology to assess DNA replication in synchronized bacteria (E. coli) was recently reported by Ferullo et al. (56).
We took advantage of the analytical capabilities offered by the click chemistry approach to explore, by flow cytometry, a possible correlation between DNA replication and induction of histone H2AX phosphorylation in cells subjected to DNA damage by treatment with the DNA topoisomerase I inhibitor topotecan (Tpt), an analogue of camptothecin (Cpt) (57). Histone H2AX phosphorylation on Ser139 is a sensitive biomarker of DNA damage response and phosphorylated H2AX, which can be detected with a phospho-specific Ab, has been defined as γH2AX (58). We previously observed that treatment with Tpt induces DNA damage manifesting as histone H2AX phosphorylation predominantly in the S-phase cells (59, 60); subsequently the S-phase cells undergo apoptosis (57, 59). Therefore, it was expected that, upon cell treatment with Cpt or Tpt, histone H2AX phosphorylation would be detected primarily in DNA replicating cells. As it is evident from the data shown in Figure 3 such a correlation indeed was demonstrated. In this experiment, the human lymphoblastoid TK6 cells were incubated with EdU to label DNA replicating cells, and concurrently exposed to Tpt. The incorporated EdU was then detected using azide conjugated to the AlexaFluor® 488 fluorochrome emitting green fluorescence while phosphorylated H2AX was detected immunocytochemically using AlexaFluor 647® tagged Ab which fluoresces red. DNA was counterstained with blue-fluorescing 4′,6-diamidino-2-phenylindole (DAPI). The multiparameter “paint-a-gate” analysis made it possible to “paint” DNA replicating (EdU-incorporating) cells red (Fig. 3 left panels) and on the bivariate EdU versus γH2AX distributions (Fig. 3 right panels) to observe that the cells expressing γH2AX from the Tpt-treated cultures were predominantly the same that incorporated EdU i.e., marked red. These data thus demonstrate that the immunocytochemical detection of γH2AX with phospho-specific Ab is fully compatible with the detection of DNA replication by using the click chemistry.
Figure 3. Assessment of DNA damage response revealed by histone H2AX phosphorylation induced by treatment of TK6 cells with DNA topoisomerase I inhibitor topotecan (Tpt) vis-à-vis DNA replication (EdU incorporation). The three top panels represent control cultures that were exposed to 5′-ethynyl-2-deoxyuridine (EdU) for 70 min, and then harvested. The bottom three panels show cultures exposed to 0.2 μM of topotecan (Tpt) for the final 40 min of the 70 min exposure to EdU. The incorporated EdU was subsequently detected by “click chemistry” using a click-IT™ reagents kit provided by Invitrogen-Molecular Probes (Carlsbad, CA) with AlexaFluor® 488 tagged azide. Histone H2AX phosphorylation was detected immunocytochemically using AlexaFluor® 647 secondary Ab (far red fluorescence) and DNA was counterstained with DAPI (blue fluorescence), as described elsewhere (61). Cellular fluorescence was measured using a MoFlo XDP (Beckman-Coulter) high speed flow cytometer/sorter. Using “paint-a-gate” multivariate analysis the EdU incorporating cells (above the skewed line, left panels) were electronically colored red, EdU negative cells blue. When re-plotted as bivariate distributions on the DNA content (DAPI) versus γH2AX coordinates (mid-panels) the expression of phosphorylated H2AX (γH2AX) can be related to EdU incorporation. It is quite evident that in the culture treated with Tpt, phosphorylation of H2AX (expression of γH2AX) occurred predominantly in the DNA replicating cells (bottom mid-panel). Also, the extent of EdU incorporation was distinctly decreased in Tpt-treated cells compared to control (bottom left panel). Right panels show the correlation between H2AX phosphorylation and EdU incorporation. The insets in mid-panels show the DNA content histograms with EdU incorporating cells marked red. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Download figure to PowerPoint
The successful application of EdU in vivo has been described by Kaiser et al.(62) who assessed DNA replication in proliferating cells in the regenerating cochlea of two-week-old chicks. Using the EdU-click approach these authors were able to correlate EdU incorporation with the expression of several proteins related to cochlear regeneration detected immunocytochemically.
Click Chemistry by Laser Scanning Cytometry (LSC)
We have tested the applicability of EdU- and EU-click methodology using laser scanning cytometry (LSC), an instrumentation that combines the analytical capabilities of flow- and image cytometry (63, 64). The latest versions of the LSC (iGeneration: iCyte®, iCys®, and iColor®) provide fluorescence excitation with up to four laser wavelengths (selected from 405, 488, 532, 561, 594, and 633 nm) and four photomultipliers allowing fluorescence measurements in wavelength bands appropriate for the respective excitation wavelengths. Its current software offers highly sophisticated multivariate data collection and analysis (65, 66). In the first set of experiments utilizing the EdU-click methodology we explored whether propidium iodide (PI), a fluorochrome commonly used as a viability probe has the potential to affect DNA replication (67). Figure 4 illustrates the results of the experiment in which the untreated and PI-treated cells were pulse labeled for 60 min with EdU which subsequently was detected using azide tagged with AlexaFluor 633. Cellular DNA was counterstained with DAPI and the various fluorescence emissions measured by LSC. The bivariate DNA content versus EdU incorporation distributions reveal excellent signal to noise ratio in discriminating the EdU labeled versus unlabeled cells. In addition, the high resolution of the DNA content frequency histograms (Fig. 4) demonstrates, as was also the case with flow cytometry (Fig. 1), that the procedure of cell labeling with EdU and its detection by fluorochrome tagged azide has no deleterious effect on subsequent DNA stainability with DAPI.
Figure 4. Detection of DNA replication by EdU click chemistry in human pulmonary adenocarcinoma A549 cell untreated and treated for 48 h with 1.5 μM propidium iodide (PI). Exponentially growing cells either untreated (A, B, and D) or treated with PI (1.5 μM) for 48 h (C) were incubated with 10 μM EdU for 60 min and then fixed. The incorporated EdU was detected with AlexaFluor® 488 tagged azide using the Click-IT™ kit (Invitrogen-Molecular Probes). Cellular blue (DAPI) and green (AlexaFluor 488) fluorescence was measured by laser scanning cytometry (67). Cells from the control culture not incubated with EdU but processed identically including treatment with AlexaFluor 488 azide are shown in Panel A. Bivariate distributions illustrate DNA content (DAPI fluorescence) versus incorporation of EdU [fluorescence intensity integrated over the nucleus (B,C) or maximal pixel (D)]. The insets show DNA content frequency histograms from the respective cultures. The upper threshold of the control, EdU unlabeled cells is marked by the dashed skewed line. The cells with variable levels of EdU incorporation marked enS and exS represent the cells entering to- and exiting from- S phase during the time of the EdU pulse; such cells were exposed to the precursor for variable time intervals (up to 60 min) while replicating DNA. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Download figure to PowerPoint
The variability in intensity of EdU labeling makes it also possible to identify and count the cells that were entering (enS) and exiting S (exS) phase during the 60 min-pulse with the precursor (Fig. 4). In fact, the apparent differences in the frequency of enS versus exS cells reveals the different rates of cell cycle progression during the initial- and the last- 60 min of the S phase interval. Specifically, in asynchronous cultures the probability of detection of the cells traversing S phase increments of the same time duration (“time windows”) is inversely proportional to the rate of traverse through these increments (68). As it is evident from the data shown in Figure 4, fewer cells were exiting S (exS) than entering S phase (enS). This indicates that the rate of progression through the initial 60-min increment of S phase (initiation of DNA replication) was distinctly slower than through the last 60-min increment.
Also of interest is the variability of EdU incorporation by cells in mid-S phase expressed as the intensity of maximal pixel (Panel D). The cells with the highest intensity of incorporated EdU (marked with oval dashed outline, D) likely represent cells which during the exposure to EdU maintained the DNA replication foci (“factories”), known to contain high local concentrations of enzymes required for chromatin replication thereby reflecting a high rate of DNA replication (37). As it is evident, growth of cells in the presence of 1.5 μM PI for 48 h had little or no effect on either the cell cycle distribution or DNA replication (see Ref. 67 for further details).
In another set of experiments, we explored the relationship between DNA replication and DNA damage caused by UV-B light (61). In these experiments, click chemistry offered the possibility of combining the analysis of DNA replication with phosphorylation of histone H2AX. We had previously observed that illumination with UV triggers DNA damage response revealed by H2AX phosphorylation (69). The experiment shown in Figure 5 was designed to explore a possible correlation between H2AX phosphorylation induced by exposure to UV and DNA replication. Toward this end exponentially growing A549 cells were pulse-labeled with EdU for 60 min, and then were exposed to 50 J/m2 of UV light, and subsequently incubated for 30 min before culture termination. The incorporated EdU was detected using Alexa Fluor® 488 azide (green fluorescence), expression γH2AX was detected immunocytochemically using Alexa Fluor 633 secondary Ab (far red fluorescence), while DNA was counterstained with DAPI (blue fluorescence) and cellular fluorescence measured by LSC (Fig. 5). Using “paint-a-gate” multivariate analysis, the EdU incorporating cells (Fig. 5A) above the skewed dashed line were electronically colored red. As in Figure 4 the cells with variable level of EdU incorporation could be identified as entering (enS) or exiting (exS) S phase, respectively, during the pulse exposure to EdU. It is evident that when the data were re-plotted as bivariate distributions on DNA content (DAPI) versus γH2AX coordinates (Panel B) expression of γH2AX is essentially limited to the red colored cells (EdU-positive) cells. Panel C shows the bivariate analysis of EdU incorporation versus expression of γH2AX. A strong correlation between the degree of H2AX phosphorylation (intensity of γH2AX immunofluorescence) and the level of EdU incorporation is seen among the EdU incorporating (red) cells; the regression plot is marked with a dashed line. Thus, the data reveals that the UV-induced DNA damage response manifesting as H2AX phosphorylation occurs essentially only in DNA replicating cells and is strongly proportional to the extent of DNA replication at the time of UV illumination.
Figure 5. Correlation between DNA replication detected by EdU incorporation and the UV-induced H2AX phosphorylation detected immunocytochemically in A549 cells. Exponentially growing cells were incubated for 60 min with EdU then exposed to 50 J/m2 of UV-B light, incubated for 30 min and fixed. The incorporated EdU was detected using Alexa Fluor 488 azide (green fluorescence), γH2AX was detected immunocytochemically with Alexa Fluor 633 secondary Ab (far red fluorescence), DNA was stained with DAPI (blue fluorescence) and fluorescence was measured by LSC (61). Using the “paint-a-gate” multiparameter analysis, the EdU-incorporating cells were marked as red (A).Note that expression of γH2AX is predominantly in the EdU incorporating cells (B) and is proportional to the intensity of EdU labeling (C). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Download figure to PowerPoint
It should be noted that the subpopulation of cells with the increased expression of γH2AX but not incorporating EdU (61) was gated out from Panel C in Figure 5. These cells were characterized by high intensity of maximal pixel of DAPI fluorescence, the marker of strongly condensed chromatin. Imaging by iCys “CompuSort” and measuring their cellular DNA content revealed that these cells were mitotic (including prophase) cells that had a DNA index (DI) of 2.0 and condensed chromatin. This observation is consistent with our prior findings that the level of the constitutive H2AX phosphorylation reporting DNA damage by endogenously generated oxidants in control, untreated cell populations, is maximal in G2 and M cells (70, 71). Our data demonstrate that the immunocytochemical detection of γH2AX with phospho-specific Ab is fully compatible with the concurrent labeling and the detection of incorporated EdU using the EdU-click methodology regardless of whether fluorescence is measured by flow cytometry (Fig. 3) or LSC (Fig. 5).
Figure 6 illustrates cytometric detection of EU incorporated into RNA. In this experiment we have estimated the degree of suppression of RNA synthesis in A549 cells exposed to different concentrations of the DNA topoisomerase II inhibitor mitoxantrone (Mxt) in relation to the induction of cell senescence (66). The data shows that cell senescence could be induced only at low (2 nM) Mxt concentrations, at which the rate of transcription was not markedly (<50%) inhibited. Higher Mxt concentrations led to strong inhibition of the rate of RNA synthesis that resulted in delayed apoptosis rater than senescence. These data were compatible with the mechanism of induction of cell senescence when the arrest in cell cycle progression is not accompanied by total suppression of cell growth (72).
Figure 6. Effect of treatment of A549 cells with the DNA topoisomerase II inhibitor mitoxantrone (Mxt) on the rate of transcription as measured by 5-ethynyluridine (EU) incorporation during 30-min incubation with the precursor. The cells were untreated (Ctrl) or treated with 2 and 20 nM Mxt for 4 h, then the RNA precursor 5-ethynyluridine (EU) was added for 30 min. The incorporation of EU was detected using the “click chemistry” (45) with AlexaFluor 633 tagged azide. Note significant, Mxt-concentration dependent, decrease in incorporation of EU in all phases of the cell cycle in the drug treated cells (see Ref. 67 for further details). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Download figure to PowerPoint
It should be noted the detection of RNA synthesis by analysis of EU incorporation mediated by the click chemistry is particularly advantageous compared with the incorporation of BrU. As mentioned before, large sections of RNA have double stranded conformation and some sections are in complexes with proteins. This impedes accessibility of the incorporated precursor to large molecules such as Abs. The steric hindrance in the accessibility of the EU incorporated into RNA to the fluorochrome-tagged azide is expected to be much less compared to the accessibility of the Ab to incorporated BrU.
Caution should be exercised, however, in interpreting data on EU (or BrU) incorporation as representing the rate of RNA synthesis. This pertains also to the DNA precursors EdU (or BrdU). There is an uncertainty about the rates of uptake of labeled precursors, which may differ from that of natural, unlabeled precursors and may differ from one cell type to the next. Furthermore, incorporation of labeled precursors is affected by the size of the pool of endogenous natural precursors, which is unknown and expected to vary in different cell types. Thus, any observed differences in the rate of incorporation of 5′ethynyl or halogenated nucleotides (typically measured as the amount of fluorescence) between different cells, should not be assumed to reflect actual differences in the rates of DNA replication or transcription.
Taking into account the simplicity of the staining protocol, excellent resolution of DNA histograms (Figs. 3 and 4), high sensitivity, accessibility of the incorporated precursor (EdU or EU) to the fluorochrome-tagged azides, and compatibility with immunocytochemical procedures, the click chemistry approach opens entirely new analytical capabilities in cell biology and medicine. It can be expected that this approach, in conjunction with flow or imaging cytometry, will be widely used to correlate DNA replication and/or transcription with a variety of other cell attributes such as cell cycle progression, growth, metabolic activity, responses to drug treatment, apoptosis, or senescence.