This protocol presents a new reporter gene system and the experimental setup to detect transcription at DNA double-strand breaks with single-molecule sensitivity.
DNA double-strand breaks (DSB) are the most severe type of DNA damage. Despite the catastrophic consequences on genome integrity, it remains so far elusive how DSBs affect transcription. A reason for this was the lack of suitable tools to simultaneously monitor transcription and the induction of a genic DSB with sufficient temporal and spatial resolution. This work describes a set of new reporters that directly visualize transcription in live cells immediately after the induction of a DSB in the DNA template. Bacteriophage RNA stem-loops are employed to monitor the transcription with single-molecule sensitivity. For targetting the DSB to a specific gene region, the reporter genes are engineered to contain a single recognition sequence of the homing endonuclease I-SceI, otherwise absent from the human genome. A single copy of each reporter gene was integrated into the genome of human cell lines. This experimental system allows the detection of single RNA molecules generated by the canonical gene transcription or by DNA break-induced transcription initiation. These reporters provide an unprecedented opportunity for interpreting the reciprocal interactions between transcription and DNA damage and to disclose hitherto unappreciated aspects of DNA break-induced transcription.
DNA double-strand breaks (DSBs) are toxic DNA lesions that disrupt cell function and contribute to the insurgence of several diseases and ageing1. Mutations resulting from the inaccurate repair of DSBs impact gene expression and set the basis for the functional decline of the cell. The emergent view that DSBs drive de novo break-induced transcription at the lesion site2,3,4,5,6,7 suggests that DSBs may also affect cellular function through break-induced RNAs. Several recent studies indicate that DSBs are sufficient to initiate programmed (e.g., at stimulus-inducible genes) and unscheduled (e.g., at non-canonical promoters) transcription4,5,7. However, despite several studies exploring the links between DNA damage and transcription, the field still lagged in its capacity to deliver a precise (i.e., single-molecule) characterization of the transcriptional events at DNA break sites. One important reason for this was the lack of appropriate experimental tools. Cell irradiation (γ-rays, X-rays, heavy ions) and drug treatments (e.g., topoisomerase inhibitors or intercalating agents) lack spatial precision and induce DNA lesions other than DSBs, including single-strand breaks and DNA adducts8. Endonucleases, such as I-PpoI and AsiSI, generate locus-specific DSBs but have not been combined with a system that allows simultaneous live-cell visualization of transcription at a single locus with high temporal precision8. To bypass this limitation, our lab spearheaded developing a set of cutting-edge reporters that directly visualize transcription with single-molecule resolution upon the controlled induction of a unique DSB4. Here, we describe these reporters, provide a detailed protocol for live-cell imaging of transcription at DSBs and show data revealing transcription initiation at a single DSB.
The reporter gene systems used in this protocol are based on the well-characterized mouse IgM reporter gene and contain the exons M1 and M2 of the membrane-bound form (µm) of the IgM µ heavy chain9,10,11. A hybrid intron separates the two exons with the strong adenovirus major late transcript (AdML) PY tract12. The expression of the reporter genes is controlled by the human cytomegalovirus promoter (CMV), into which two tandem copies of the Tet operator (TetO) sequence have been inserted. The reporter genes are each inserted into a plasmid vector containing an Flp Recombination Target (FRT) site and inserted into a specific FRT target site in the genome of a HEK293 host cell line. This cell line also constitutively expresses the Tet repressor protein to regulate the expression of the reporter gene via the presence or absence of tetracycline/doxycycline. To allow visualization of the reporter gene transcription, 24 tandem repeats of the MS2 stem loop sequence and 24 tandem repeats of the PP7 stem loop sequence were inserted at different positions in respect to transcription start site and exon/intron structure of the reporter gene. The MS2/PP7 RNA stem loops form upon transcription and are specifically bound by ectopically expressed MS2/PP7 coat proteins tagged with green and red fluorescent proteins, a strategy widely used before to image transcription13,14,15. In addition, a single copy of the 18 bp recognition sequence was inserted for the homing endonuclease I-SceI that is directly flanked by the RNA stem-loop sequence arrays in the reporter genes. Standard cloning techniques were used to generate all plasmids, the fragment containing the I-SceI-24xMS2 stem-loop of the PROP reporter gene was synthesized by a commercial gene synthesizing service.
The promoter-proximal DSB reporter gene (PROP) was constructed by inserting the I-SceI cutting site 45 base pairs (bp) downstream of the putative transcription start site in exon I, followed by 149 bp until the start of the 24x MS2 stem-loop cassette, which was de novo designed with two alternating non-identical stem-loop sequences16 and additional five non-repetitive 20 bp spacer sequences to reduce redundancy. The MS2 stem-loop array is followed by 72 bp until the beginning of the 1844 bp intron and the 1085 bp exon II until the cleavage and polyadenylation site. The exon II encodes a cyan fluorescent protein (CFP) fused to a C-terminal peroxisomal targeting sequence (PTS) from the human peroxisomal acyl CoA oxidase to allow an independent screening of the reporter gene expression (Figure 1A).
The exon II DSB reporter gene (EX2) consists of a 167 bp exon I followed by the intron and exon II encoding the CFP-PTS. Further downstream at a distance of 169 bp, a cassette containing 24x MS2 stem-loops was inserted, followed by an 84 bp linker sequence with an I-SceI site in the center, followed by 24x PP7 stem-loops and 221 bp until the cleavage and polyadenylation site17 (Figure 1B).
Lastly, the exon II DSB reporter gene with antisense transcription labeling (EX2AS) is based on the human ubiquitin B (UBB) gene transcript UBB-201 and contains two exons and one intron. The exon I have a total length of 1534 bp with an inverse insertion of the 24x MS2 stem-loop sequence. Therefore, the correct MS2 stem-loop RNA sequence will be transcribed in an antisense direction with respect to the sense transcription of the reporter gene from the CMV promoter. The intron has a length of 490 bp, followed by exon II with the I-SceI site, and a coding region was inserted for two in-frame ubiquitin subunits. Downstream of the UBB gene is a sequence that forms a 24x PP7 stem-loop upon transcription of the reporter gene in a sense direction (Figure 1C).
The transient transfection of an inducible construct of the homing endonuclease I-SceI allows for the controlled creation of a DSB at the inserted recognition site within each reporter gene18. The I-SceI endonuclease is fused in frame with the ligand-binding domain of the glucocorticoid receptor and a far-red fluorescent protein iRFP713. This construct is cytoplasmic in the absence of triamcinolone acetonide (TA) but migrates rapidly into the nucleus upon the addition of TA to the growth medium of the cells (Figure 1D). The induction of DSBs by the I-SceI system is robust, as demonstrated before18,19,20. The reporter gene transcription can be monitored in parallel by visualizing the fluorescently tagged RNA stem-loop systems MS2 and PP7.
Conflicts between essential biological processes such as replication, transcription, DNA damage, and DNA repair have been identified as a critical source of genome instability22. These studies also have led to the discovery of transcription at sites of DNA damage and attributed a functional role to the break-induced transcripts in regulating DNA damage repair processes23. The new tools and the protocol described here allow further investigation of RNA Pol II transcription dynamics at DSBs. A critical point in this protocol is the generation of cell lines that contain a single copy of the reporter gene integrated into the genome. This key feature eliminates the noise created by the transcription of several reporter genes integrated with multiple copies within a single genomic locus and allows the collection of kinetic parameters of transcription dynamics and individual RNA transcripts. A crucial technical requirement to observe transcription of single reporter gene integrations is the availability of a microscope system that allows the detection of single RNA transcripts labeled with the MS2 or PP7 system in live cells4,12. Here, live-cell microscopy is performed on a Confocal Spinning Disk system mounted on an inverted microscope, equipped with 100 mW solid-state Lasers coupled to an acoustic-optic tunable filter as described elsewhere24. Furthermore, to study transcription at a single DSB using the reporters, individual cells must be carefully monitored to achieve the highest time resolution, which requires imaging cells for several hours, making this a low throughput assay. Still, we observe several cells in parallel with the positioning controlled by a piezo-driven microscope stage. For ensuring optimal environmental conditions for live cell observation over hours, the microscope body, including the sample stage, is placed inside a plexiglass environmental chamber. In addition, a closed stage incubation chamber is mounted on the microscope stage and connected to CO2 and humidity supply controllers.
The first critical step in the protocol is the selection of regions of interest with cells for imaging. Each XY position marked for imaging must contain one or more cells showing transfection with the fluorescent RNA stem-loop binding proteins according to the reporter gene and transfection scheme described in Section 1, Table 1, as well as in Figure 2D, E. Furthermore, the cells must exhibit bright labeled transcription sites must be co-transfected with the I-SceI-GR-iRFP713 construct, and the protein must be localized initially in the cytoplasm (Figure 2D and E).
The cells should show a fluorescence intensity level of unbound fluorescently tagged MS2 and/or PP7 coat protein low enough to detect single labeled transcripts over the background fluorescence level. At the same time, a robust fluorescence intensity level of the fluorescently tagged MS2 and/or PP7 coat proteins is necessary to allow imaging over at least 60 min without losing too much fluorescence due to some bleaching that occurs. The "Scale image display" with a fixed range as described in Section 3.7 is used to allow a standardized selection of cells according to their fluorescence intensity level.
A second critical protocol step is adding the TA to the cells on pre-determined XY positions through the small hole in the lid of the glass bottom dish. Any manipulation of the glass bottom dish would cause a shift from the marked XY position of the cells and must be avoided. Therefore, carefully handling the micropipette while adding the TA diluted in the cellular growth medium is vital for pre-selected cells' successful observation, as demonstrated in Figure 2F. The adaptation of different systems to add drugs to cells mounted on a microscope stage, such as a perfusion system, would require a separate stage incubation chamber with tube entrance and exit openings and a pump or injection system to administer drugs. Other methods such as channel slides with coverslip-like bottom surfaces result in a slow diffusion of administered drugs into the channel and cause an additional delay between drug addition and effect. Finally, incautious pipetting into a channel slide opening may shift the sample position as well. Therefore, the present system with a custom drilled hole in the lid of a glass-bottom dish is straightforward to adapt, low cost, and suitable for administering different growth media, drugs, and components. The small diameter of the hole and the humidified atmosphere in the stage incubation chamber also prevents drying out of the cell medium.
A third critical step in this protocol is the data analysis, which requires a manual inspection of the time points when transcription ceases due to the induction of a DSB. The time point of terminating transcription is indicated by releasing the last transcripts from the previously bright labeled site of the reporter gene transcription. Similarly, the events of break-induced transcription initiation must be inspected with care to detect individual transcription events with the relatively low signal-to-noise ratio of single fluorescently labeled mRNAs.
The dynamics of repair of the induced DSB adds an extra layer of complexity to the analyses of data generated using these reporters, limiting them to the first minutes immediately after induction of the DSB. The transgenic nature of reporter genes and the repeat-rich nature of the MS2 and PP7 stem-loop arrays may assemble a unique chromatin landscape, interfering with establishing putative stable break-induced transcription programs. Nonetheless, compared to ionizing- or UV-irradiation, the I-SceI mediated induction of a DSB in reporter genes is a much more robust system to investigate transcription at individual DSBs.
Different endonuclease systems such as I-CreI, I-PpoI, or AsiSI that have or don't have additional recognition sites within the human genome can be combined with the present reporter gene systems for a possible higher efficiency of generating DSBs. However, they require first introducing the endonuclease recognition site into the reporter genes. Secondly, they may have a similar variability on the timing and efficiency induction of a DSB in individual cells. On the other hand, inserting tandem copies of endonuclease recognition sites may increase the efficiency of DSB induction. Moreover, testing the presented reporter gene systems in different cell lines would allow the comparison of transcription dynamics at DSB sites between different cellular backgrounds and the availability of different DNA damage repair pathways such as in cancer cells, primary cells, and differentiated non-cycling cells. However, the construction of the reporter genes to be compatible with the Flp/FRT system is currently limiting the integration into available Flp/FRT host cell lines.
In addition to microscopy-based applications, the current reporter genes may also be combined with biochemical assays, such as chromatin immunoprecipitation, to study the recruitment of DNA repair or transcription factors to a single DSB or to assess nucleosome occupancy, histone modifications, and chromatin state around the DSB site. Furthermore, a combination with different reporter systems would allow the study of functional links between DNA damage and processes like genome organization or DNA replication.
The authors have nothing to disclose.
We thank RH Singer, J.A. Chao, T. Misteli, M. Carmo-Fonseca for gifts of plasmids and reagents. We are also indebted to the iMM Bioimaging Facility staff, A. Temudo, A. Nascimento, and J. Rino, for critical reading of the manuscript. This work was funded by PTDC/MED-OUT/32271/2017, PTDC/BIA-MOL/30438/2017 and PTDC/MED-OUT/4301/2020 from Fundação para a Ciência e a Tecnologia (FCT), Portugal and by LISBOA-01-0145-FEDER-007391, project cofunded by FEDER through POR Lisboa, Portugal 2020-Programa Operacional Regional de Lisboa, and FCT. Funding was also received from EU Horizon 2020 Research and Innovation Programme (RiboMed 857119). M.A. is the recipient of the FCT Ph.D. fellowship 2020.05899.BD.
100 mW solid-state Lasers | Coherent Inc., Santa Clara, CA, USA | ||
3i Marianas SDC Confocal Spinning Disk system | Intelligent Imaging Innovations Inc. | ||
Air-cooled EMCCD Camera Evolve 512 | Photometrics, Tucson, AZ USA | ||
Axio Observer Z1 inverted microscope | Carl Zeiss MicroImaging, Germany | ||
Blasticidin | InvivoGen | ant-bl-1 | |
charcoal-stripped fetal bovine serum | Sigma-Aldrich | F6765-500 ML | |
CO2 module S | PeCon GmbH, Erbach, Germany | ||
CSU-X1 confocal spinning disk unit | Yokogawa Electric, Tokyo, Japan | ||
DMEM | Gibco | 41966029 | |
DMEM with Hepes no PhenolRed | Gibco | 21063-029 | |
Doxicyclin | Sigma-Aldrich | D9891 | for induction of reporter gene expression; stock solution of 0.5 mg/ml was used at 1:1000 dillution in cell growth medium |
FBS | Gibco | 10270106 | |
Flp-In T-REx 293 cell line | Thermo Fischer Scientific Invitrogen | R75007 | |
I-SceI-24x MS2 stem loop sequence | GeneArt, Thermo Fischer Scientific | custom synthesized DNA fragment containing a single I-SceI recognition sequence and 24 tandem MS2 stem loop sequences | |
Heating Device Humidity 2000 | PeCon GmbH, Erbach, Germany | ||
Hygromycin B | Roche | 10843555001 | |
Immersion oil Immersol 518 F | Carl Zeiss MicroImaging Inc.) | 444960-0000-000 | |
L-glutamine | Thermo Fisher Scientific | 25030081 | |
Lipofectamin 3000 helper reagent P3000 | Thermo Fisher Scientific | L3000001 | transfection helper reagent |
Lipofectamine 3000 reagent | Thermo Fisher Scientific | L3000001 | lipid-based transfection reagent |
MatTek 35 mm dish, Glass bottom No. 1.5 | MatTek Corporation, Ashland, MA, USA | P35G-1.5-10-C | |
microscope incubation chamber | PeCon GmbH, Erbach, Germany | ||
pcDNA5/FRT/TO | Thermo Fischer Scientific Invitrogen | V652020 | |
pOG44 plasmid | Thermo Fischer Scientific Invitrogen | V600520 | |
SlideBook 6.0 Software | Intelligent Imaging Innovations Inc. | ||
stage incubation chamber PeCon P-Set 2000 | PeCon GmbH, Erbach, Germany | ||
StaQtool Software | iMM-JLA Lisbon, Portugal | available at: https://imm.medicina.ulisboa.pt/facility/bioimaging/lib/exe/fetch.php?media=STaQTool_setup.zip | |
triamcinolone acetonide (TA) | Sigma-Aldrich | T6501 | synthetic glucocorticoid; induces the glucocorticoid receptor to migrate from the cytoplasm to the nucleus |
Trypsin/EDTA Solution (TE) | Thermo Fisher Scientific | R001100 |