Methods for assessing synthetically lethal gene pairs

Abstract

The invention relates to the field of synthetically lethal gene pairs and provides inducible gene- inactivation systems and methods useful for screening and analyzing synthetically lethal gene pairs.

Description

[0001] Methods for assessing synthetically lethal gene pairs

[0002] FIELD OF THE INVENTION

[0003] The invention relates to the field of synthetically lethal gene pairs and provides inducible geneinactivation systems and methods useful for screening and analyzing synthetically lethal gene pairs.

[0004] BACKGROUND OF THE INVENTION

[0005] Tumors are primarily dependent on one or more driver mutations, which can be exploited in therapy by inhibiting the synthetically lethal gene pair of the mutated driver gene, resulting in the death of only the cancer cells. Currently, only one synthetic lethality-based cancer therapy is used in clinical practice, which exploits the relationship between the BRCA1-PARP1 gene pair (Huang, A., Garraway, L.A., Ashworth, A. and Weber, B. (2020) Synthetic lethality as an engine for cancer drug target discovery. Nat Rev Drug Discov, 19, 23-38.).

[0006] The majority of synthetic lethality research is currently based on experiments using RNA interference or CRISPR library screens. These methods share the common feature of delivering an RNAi or sgRNA library into parallel wild-type and mutant / knockout cell lines (Guo et al. (2022) SLOAD: a comprehensive database of cancer-specific synthetic lethal interactions for precision cancer therapy via multi -omics analysis. Database (Oxford), 2022.; O'Neil, N.J., Bailey, M.L. and Hieter, P. (2017) Synthetic lethality and cancer. Nat Rev Genet, 18, 613-623.). However, identifying synthetic lethality using these approaches is not always straightforward. Usually, synthetic lethality is classified as either conditional or non-conditional. The primary distinction between the two types is that the conditional type requires an auxiliary, internal or external stimulus - such as DNA damage - to induce cell death, whereas the non-conditional type does not require any additional effect. The aforementioned RNAi and CRISPR-based methods have the disadvantage of detecting only the non-conditional type of synthetic lethality (Kaelin, W.G., Jr. (2005) The concept of synthetic lethality in the context of anticancer therapy. Nat Rev Cancer, 5, 689-698.; O'Neil et al. supra; Liu et al. (2020) SL(2)MF: Predicting Synthetic Lethality in Human Cancers via Logistic Matrix Factorization. IEEE / ACM Trans Com-put Biol Bioinform, 17, 748-757.). Recently, statistical and bioinformatic approaches have become prevalent in predicting synthetic lethal relationships. Although these algorithms are more and more accurate, further experimental validation will always be required (Li et al. (2018) ATM inhibition induces synthetic lethality and enhances sensitivity of PTEN-deficient breast cancer cells to cisplatin. Exp Cell Res, 366, 24-33.).Further research is still needed to identify additional synthetic lethal interactions and refine the selection of those patients who would benefit the most from these therapies.

[0007] SUMMARY OF THE INVENTION

[0008] Method for assessing a gene of interest (gene A) and a candidate synthetically lethal gene pair (gene B) of gene A, comprising

[0009] providing a cell in which gene A is silenced, knocked out or knocked down,

[0010] introducing a gene silencing construct targeting gene B under the regulation of an inducible transcription regulating sequence into the cell.

[0011] A method of screening for a synthetically lethal gene pair (gene B) of a gene of interest (gene A), comprising

[0012] providing a cell in which gene A is silenced, knocked out or knocked down,

[0013] introducing a gene silencing construct targeting gene B under the regulation of an inducible transcription regulating sequence into the cell.

[0014] A method for assessing the mechanism of synthetic lethality caused by a synthetically lethal gene pair, comprising

[0015] providing a cell in which one gene of the pair (gene A) is silenced, knocked out or knocked down,

[0016] introducing a gene silencing construct targeting the other gene of the pair (gene B) under the regulation of an inducible transcription regulating sequence into the cell.

[0017] Use of an expression vector comprising a minimal promoter sequence fused to a TetO element sequence, a DNA sequence coding for a shRNA (shDNA sequence) targeting gene A or B and optionally a sequence coding for an antibiotic resistance gene or a fluorescence marker or both in any of the methods above.

[0018] Use of an expression vector comprising a minimal promoter sequence fused to a TetO element sequence, a DNA sequence coding for a shRNA (shDNA sequence) targeting gene A or B, a sequence coding for TetR and optionally a sequence coding for an antibiotic resistance gene or a fluorescence marker or both in any of the methods above.

[0019] An expression or transfection vector comprising a minimal promoter sequence fused to a TetO element sequence, a DNA sequence coding for a shRNA (shDNA sequence) targeting gene A or B and optionally a sequence coding for an antibiotic resistance gene or a fluorescence marker or both.

[0020] An expression or transfection vector comprising a minimal promoter sequence fused to a TetO element sequence, a DNA sequence coding for a shRNA (shDNA sequence) targeting gene Aor B, a sequence coding for TetR and optionally a sequence coding for an antibiotic resistance gene or a fluorescence marker or both.

[0021] The following are preferred embodiments of the methods and uses above.

[0022] Preferably gene A is inducibly silenced. Preferably gene A is transiently silenced. Preferably gene A is stably silenced.

[0023] Preferably gene A is silenced by introducing a gene silencing construct targeting gene A under the regulation of an inducible transcription regulating sequence into the cell.

[0024] Preferably the method comprises introducing a gene silencing construct targeting gene A under the regulation of an inducible transcription regulating sequence into a cell. Preferably the method further comprises silencing the cell into which the gene silencing construct targeting gene A under the regulation of an inducible transcription regulating sequence has been introduced, thereby providing the cell in which gene A is silenced.

[0025] Preferably introducing the gene silencing construct (targeting gene A or B) is performed by transfecting the cell with the gene silencing construct. Preferably transfection is transient. Preferably the gene silencing construct is stably integrated into the cell.

[0026] Preferably the gene silencing construct targeting gene A or B comprises a a DNA sequence coding for a shRNA (shDNA sequence) capable of silencing gene A or B, respectively. Preferably the gene silencing construct targeting gene A or B comprises a a DNA sequence coding for a shRNA (shDNA sequence) the expression of which is capable of silencing gene A or B, respectively.

[0027] Preferably the DNA sequence coding for a shRNA (shDNA sequence) comprises (synthesized) antisense mRNA coding part(s) targeting gene A or B.

[0028] Preferably the DNA sequence coding for a shRNA (shDNA sequence) targeting gene B comprises variable antisense mRNA coding part(s) capable of silencing (variable silencer) more than one gene (e.g. a gene family comprising a consensus mRNA or amino acid coding sequence; i.e. gene B is more than one gene).

[0029] Preferably the inducible transcription regulating sequence comprises a TetO element and a promoter. Preferably the promoter is a minimal promoter.

[0030] Preferably the inducible transcription regulating sequence comprises a TetO regulated element, preferably a TetO regulated promoter, preferably TetO regulated pHl promoter or TetO regulated CMV promoter.

[0031] Preferably the gene silencing construct is comprised in a transfection (expression) vector. Preferably the vector is a plasmid.Preferably the gene silencing construct or the vector further comprises a coding sequence for the repressor of the Tet resistance element (TetR protein), preferably a Tet repressor gene. Preferably the transfection vector further comprises a sequence coding a marker, preferably an antibiotic resistance gene or a gene coding for a fluorescent protein. Preferably the marker is an antibiotic resistance marker, e.g. neomycin, puromycin, hygromycin resistance gene or a fluorescence marker, e.g. yellow fluorescent protein or mCherry or venus.

[0032] Preferably the method further comprises propagating the cell after transfection. Preferably the method further comprises inducing the silencing of gene A or B or both via the applicable inducible gene silencing construct. Preferably the method further comprises screening the cell clones (resulting from the propagation of the cell) comprising the gene silencing construct(s) for viability after silencing gene A or B or both. Preferably clones showing reduced viability after said silencing (or inducing the silencing) are isolated and analyzed, identifying thereby synthetically lethal pairs of said gene of interest.

[0033] Preferably the method of screening for a synthetically lethal gene pair (gene B) of a gene of interest (gene A) comprises inducing the variable silencer. Preferably cell clones showing reduced viability after the induction are analyzed to establish which gene was silenced in them by the induction of the variable silencer, identifying thereby the synthetically lethal pair of gene A.

[0034] Preferably the cell is a mammalian cell.

[0035] Preferably the cells comprising the gene silencing construct targeting gene A or the gene silencing construct targeting gene B or both are stable cell lines. Preferably the cells in which gene A is silenced, knocked out or knocked down is a stable cell line. Preferably the cells express TetR.

[0036] Preferably the method further comprises treating the cells with an anti -tumor agent after the induction of silencing gene A, B or both.

[0037] BRIEF DESCRIPTION OF THE FIGURES

[0038] Figure 1 Experimental setup for the base single- and co-transfection systems. Equal amounts of cells were plated into 6-6 wells per cell line to a 24 well plate. All wells were induced with 1 pg / ml Tet for the rest of the experiment. On the fourth day, half of the samples were treated with a DNA-damaging agent. On the second day after treatment, the cells were stained with Hoechst33342 and propidium iodide. U2OS TetR cells were additionally stained with Cel-lEvent Caspase-3 / 7 Red detection reagent.Figure 2 Experimental setup for the single- and co-transfection systems with fluorescent markers using transient transfections. SL: synthetic lethality. Equal amounts of cells were plated into 4 wells of a 6-well plate. Next day, cells were (co-)transfected with silencing plasmids. After transfection, all samples were induced with Tet until the end of the experiment. The day after transfection, each well was divided into 6-6 wells of a 24-well plate and put on selective media (1 pg / ml puromycin + 100 pg / ml hygromycin for U2OS and 300 pg / ml neomycin for RPE1 cells). On the fourth day after plating, half of the samples were treated with 30 pM CP (cisplatin). On the day after treatment, the cells were stained with Hoechst33342 and propidium-io-dide. U2OS TetR cells were additionally stained with CellEvent Caspase-3 / 7 Red detection reagent.

[0039] Figure 3 Schematic representation of the process of cloning the silencing plasmids

[0040] The original pSCl plasmid contained a neomycin-resi stance gene (purple box) and an EGFP (light green box), which we switched to a TetO-controlled pHl promoter (light blue box). Then, we switched the NeoR cassette to puromycin and hygromycin resistance genes (lighter and darker purple boxes, respectively). In the plasmid in which we kept the neomycin resistance, we incorporated the Tet repressor gene (green box). In the next step, we inserted the sh constructs (blue boxes) into all plasmids. To adapt the plasmids to high throughput-microscopy, we inserted YFP-NLS into the NeoR, TRIB-V2 into the HygroR, and MEK1-V1 into the PuroR-containing plasmids, respectively (all represented by yellow boxes).

[0041] Figure 4 Expected results of the YFP-silencing tests in a U2OS TetR cell line

[0042] Expected result using different plasmid combinations without (A-D) or with tetracycline induction (E-H). Black curved arrows represent transcription, which is blocked if the TetR protein (green semilunar figure) binds to the TetO-regulated pHl promoter, indicated by a light blue semicircle in front of the regulated gene. TetR is released from its binding site when Tet (light green circle) binds to it, enabling transcription. When shYFP is produced, the translation of YFP is silenced.

[0043] Figure 5 Results of the YFP-silencing tests in the U2OS TetR cell line

[0044] (A) We scanned all samples at three wavelengths corresponding to Hoechst33342 staining and the YFP and mCherry fluorescent proteins. The appropriate channels are indicated above the columns. The left block shows the uninduced control, while the right one displays the Tet-induced samples. As expected, differences between the control and the induced samples were visible with all plasmids. The non-induced samples showed YFP signal only, while the Tet-induced ones displayed mCherry only, except for the single-transfected plasmid pIL3565, which showed both YFP and mCherry signals. (B) Percentage of YFP+ cells in the silencingtests compared to their own non-induced controls. (C) Percentage of mCherry+ cells in the silencing tests compared to their own non-induced controls.

[0045] Figure 6. Results of the Tet induction tests

[0046] (A) Long-term induction test in a U2OS TetR cell line. The graph depicts the sum YFP and mCherry signal intensities of Tet-induced wells 17 days after co-transfection. After co-trans-fection, a new well was induced each day and kept on Tet. (B) Investigation of expression rates during cell line isolation. The graph shows the average YFP and mCherry signal intensities in the wells, measured one day after transfection. Every third day after transfection, a new well was induced. (C) Induction test of the U2OS TetR P-(YFP-NLS)-TetO:mCherry H-shYFP stable cell line. The graph displays the ratios of YFP+ and mCherry+ cells in each well induced for 1, 2, and 3 days as well as in the uninduced control.

[0047] Figure 7 The effect of cisplatin (A), ultraviolet light (B), and hydroxyurea (C) treatment on survival, detected in the single-transfection system using the BRCA1-RAD18 gene pair.

[0048] CP, UV, and HU indicate cisplatin, ultraviolet light, and hydroxyurea, respectively. CP was used at 30 pM concentration, UV treatment was applied at 40 J / m2 dosage, and HU was added to the cells at three different concentrations (0.25 mM; 0.5 mM and 1 mM). Columns represent the mean values of the number of dead cells compared to un-treated control with + / -SEM from at least three independent experiments. Statistical analysis was performed by two-way ANOVA (*p < 0.05; **p<0.01).

[0049] Figure 8 Results from the co-transfection system using the BRCA1-RAD18 gene pair in U2OS TetR stable cell lines with cisplatin (A) and ultraviolet light (B) treatment and in U2OS TetR cells with transient transfection with cisplatin treatment (C).

[0050] CP and UV indicate cisplatin and ultraviolet light, respectively. CP was used at 30 pM concentration, UV treatment was applied at 40 J / m2 dosage. Experiments were performed in U2OS TetR stable cell lines (A, B) and a U2OS TetR cell line with transient transfection (C). Columns represent the mean values of the number of dead cells compared to untreated controls with + / -SEM from at least three independent experiments. Statistical analysis was performed by two-way ANOVA (*p < 0.05; **p<0.01; ***p<0.001; ****p<0.0001).

[0051] Figure 9 Results of the DNA fiber assays reveal fork recovery

[0052] Schematic representation and results of the DNA fiber assay used to investigate the recovery of replication forks following a 2-hour treatment with 4 mM HU. Orange bars represent the percentage of the replication forks that were able to resume DNA replication after the HU treatment, while red bars represent the percentage of stalled forks. Statistical analysis was performed by two-way ANOVA (*p < 0.05; **p<0.01; ***p<0.001; ****p<0.0001).Figure 10 Schematic representation (A), results (B) and representative photos (C) of the BrdU alkaline comet assays.

[0053] Quantification of the BrdU alkaline comet assays showing each value of % comet tail and the mean of every sample. Each cell line is represented by a different color. Mean of each group is labeled with a horizontal line (black in the case of untreated and red in the case of HU -treated samples). Statistical analysis was performed by one-way ANOVA (*p < 0.05; **p<0.01; ***p<0.001; ****p<0.0001).

[0054] Figure 11 Quantification (A) and representative images (B) of pRPA immunostainings Quantification of pRPA immunostainings showing the percentage of cells that contain more than five nuclear pRPa foci. Each cell line is represented by a different color. Statistical analysis was performed by two-way ANOVA (*p < 0.05; **p<0.01; ***p<0.001; ****p<0.0001) (A) and representative microscopic images (B).

[0055] Figure 12 Schematic representation of the YFP-silencing test experiments

[0056] On the first day, wild-type U2OS or U2OS TetR cells were plated in 6-well plates. The following day, cells were transfected with the following plasmid combinations: pIL3565; pIL3484 + pIL3565; pIL3519 + pIL3565; and pIL3529 + pIL3565. Two wells were transfected with each combination. After transfection, one of all duplicates was induced with tetracycline. Two days after transfection, cells were fixed with paraformaldehyde, stained with Hoechst33342, and scanned with a high-throughput microscope.

[0057] Figure 13 Expected results of the YFP-silencing tests in a wild-type U2OS cell line Expected result using different plasmid combinations without (A-D) or with tetracycline induction (E-H). Black curved arrows represent transcription, which is blocked, if the TetR protein (green semilunar figure) binds to the TetO-regulated pHl promoter, indicated by a light blue semicircle in front of the regulated gene. TetR is released from its binding site when tetracycline (light green circle) binds to it, enabling transcription. When shYFP is produced, the translation of YFP is silenced.

[0058] Figure 14 Results of the YFP-silencing tests in a wild-type U2OS cell line

[0059] (A) All samples were scanned at three wavelengths corresponding to Hoechst33342 staining and the YFP and mCherry fluorescent proteins; the channels used are indicated above the columns. The left block shows the uninduced control, while the right one displays the tetracycline-induced samples. When only plasmid pIL3565 was used, both the YFP and the mCherry signals were visible in the induced and non-induced samples as well, while when co-transfected with plasmids pIL3484 or pIL3529, the samples displayed red fluorescence only. As expected, difference between the control and the induced samples was only visible in the case of pIL3519:the control sample displayed YFP, while the induced sample showed mCherry signal only. (B) Percentage of YFP+ cells in the silencing tests, compared to their own non-induced controls. (C) Percentage of mCherry+ cells in the silencing tests, compared to their own non-induced controls.

[0060] Figure 15 Validation of the YFP-silencing experiment via Western blot

[0061] (A) Workflow of the experiment: On the first day, U2OS TetR cells were plated in a 6-well plate. The next day, cells were transfected with the following plasmid combinations: pIL3484 + pIL3565; pIL3519 + pIL3565, and pIL3529 + pIL3565. Two wells were transfected with each combination. After transfection, one of each duplicate was induced with tetracycline. Two days after transfection, cells were collected and lysed in 2xSDS buffer. Cell lysates were analyzed with Western blot. (B) Results of the Western blot developed with a-GFP primary and a-rabbit secondary antibodies. (C) Quantification of the Western blot band intensities.

[0062] Figure 16 Validation of the RAD18-silencing experiment via Western blot in RPE1 p53- / - N-TetR-shRAD18 (A) and RPE1 p53- / - BRCA1- / - N-TetR-shRAD18 (B) cell lines.

[0063] Cells were grown from each sample for five days in 6-well plates. For each cell line, there were three different conditions: grown without tetracycline or kept on tetracycline for three or five days.

[0064] Figure 17 Results from the single-transfection system using the BRCA1-Radl8 gene pair with cisplatin (A), UV (B) and hydroxyurea (C) treatment. CP, UV and HU indicate cisplatin, ultraviolet light and hydroxyurea respectively. Cisplatin was used in a 30 pM concentration, UV treatment happened with 40 J / m2 dosage and HU was added to the cells in three different concentrations (0.25 mM; 0.5 mM and 1 mM respectively). Columns represent the mean cell numbers in each well with + / -SEM from at least three independent experiments. Statistical analysis was performed by two-way ANOVA (*p < 0.05; **p<0.01; ***p<0.001; ****p<0.0001). Figure 18 Quantification of the BrdU alkaline comet assays showing each value of % comet tail and the mean of every sample. HU recovery time optimization on RPE1 p53 - / - BRCA1- / -shControl (A) and RPE1 p53- / - shControl (B) cell lines. Statistical analysis was performed by one-way ANOVA (*p < 0.05; **p<0.01; ***p<0.001; ****p<0.0001).

[0065] Figure 19 Quantification of the BrdU alkaline comet assays using double silencing, showing each value of % comet tail and the mean of every sample. Each cell line is represented by a different color. Mean of each group is labeled with a horizontal line (black in the case of untreated and red in the case of HU-treated samples). Statistical analysis was performed by oneway ANOVA (*p < 0.05; **p<0.01; ***p<0.001; ****p<0.0001).DETAILED DESCRIPTION OF THE INVENTION

[0066] The present invention provides novel tools based on inducible gene silencing for isolating and analyzing synthetically lethal gene pairs. The systems provided can be employed to evaluate new, previously unidentified synthetically lethal gene pairs and to investigate synthetically lethal interactions of already identified gene pairs in greater detail. By the use of the novel tools it is possible to construct stable cell lines comprising synthetically lethal gene pairs wherein the cells only die upon the induction of the expression of the silencer sequences.

[0067] More specifically, the invention provides a method for the evaluation of a synthetically lethal pair of a gene of interest known or supposed to have at least one synthetically lethal gene pair. According to this method of the invention, a cell line is provided in which two vectors, each carrying an inducible gene silencer construction comprising shDNA sequences (DNA sequences coding for shRNA molecules) having synthesized antisense mRNA coding parts, are introduced into said cell line. Then cell clones comprising said vectors are screened for viability after silencing both genes. Clones showing reduced viability after said silencing are isolated and analyzed, identifying thereby synthetically lethal pairs of said gene of interest.

[0068] In another aspect, the invention further provides a method for screening for a synthetically lethal gene pair of a desired gene known or supposed to have at least one synthetically lethal gene pair. According to this method of the invention, a cell line is provided in which said desired gene is knocked out or inducibly silenced and a vector comprising an inducible gene silencer construction comprising shDNA sequences having variable anti-sense mRNA coding parts capable of silencing more than one gene (e.g. a gene family comprising a consensus mRNA or amino acid coding sequence) is introduced into said cell line (a variable silencer expression construct). Then cell clones comprising said vector are screened for viability after both silencing the desired gene (unless it is knocked out) and inducing the variable silencer construction. Cell clones showing reduced viability after said silencing (if needed) and induction are analyzed to establish which gene was silenced in them by the induction of the variable silencer expression construct, identifying thereby the said synthetically lethal pair of the desired gene.

[0069] In another aspect, the invention provides a method for studying the conditions of lethality of gene pairs known or supposed to be synthetically lethal. According to this method of the invention, a cell line is provided in which the first one of said gene pair to be studied is knocked out or inducibly silenced and a vector carrying an inducible gene silencer construction comprising shDNA sequences the expression of which is capable of silencing the second one of said gene pair is introduced. Then the viability of cell clones comprising said vector are checked in different conditions after silencing the first gene of said gene pair (if it is not already knocked out)and also inducing the silencer construction to express the shDNA silencing the second one of said gene pair.

[0070] The cell lines applied in the methods of the invention are advantageously mammalian cell lines and the vectors used in the methods of the invention are advantageously plasmids.

[0071] In still another aspect, the invention provides an inducible expression vector usable in the methods of the invention, which enables the inducible expression of shDNA sequences inserted therein. The inducible expression vectors of the invention are advantageously capable of transient gene silencing and are also useful for isolating stable cell lines comprising synthetically lethal gene pairs at least one of which is only silenced upon induction. The inducible expression vectors of the invention are advantageously plasmid vectors. One advantageous inducible expression vector of the invention is the pSCl-TetR-(Neo)-pHl-TetO plasmid. Another advantageous inducible expression vector of the invention is the pSCl-(Puro)-(ApR) -pHl-TetO plasmid. Still another advantageous inducible expression vector of the invention is the pSCl-(Hy-gro)-(ApR) -pHl-TetO plasmid. A further advantageous inducible expression vector of the invention is the pSCl-(Hygro)-(ApR)-(TRIBl-Venus2)-pHl-TetO plasmid. Another advantageous inducible expression vector of the invention is the pSCl-(Puro)-(ApR)-(MEKl-Venus2)-pHl-TetO plasmid. A further advantageous inducible expression vector of the invention is the pSCl-TetR-(Neo)-(YFP-NLS)-pHl-TetO.

[0072] The objective was to establish a standardized system for human cell lines that is able to assess both conditional and non-conditional synthetically lethal gene pairs and is compatible with high-throughput microscopy. Two distinct approaches were developed: a co-transfection and a single-transfection system. Both systems are based on plasmids that silence gene expression by carrying shDNA constructs under the regulation of the TetO operator sequence and are inducible with Tet (tetracycline). This feature allows for the analysis of non-conditional synthetic lethality, since without applying Tet induction cells will not die immediately after transfection. A reversible silencing approach was chosen, that avoids generating DNA damage, unlike CRISPR systems, which depend on DNA repair mechanisms, impeding the study of genes such as BRCA1 / 2 that participate in DNA repair pathways. The single-transfection system is adaptable to any cell line that has one of the potentially synthetically lethal genes knocked out / down, as the shDNA-harboring plasmid carries the TetR gene, which is required for the repression of shDNA expression in the absence of Tet.

[0073] The functionality of the new systems was evaluated in U2OS and U2OS-TetR cell lines via the simultaneous silencing and induction of the YFP (yellow fluorescent protein) and mCherry genes, respectively. The extent of YFP silencing was also confirmed by Western blot analysis.It was found that the plasmids functioned as expected in both systems. Due to the variability in co-transfection efficiency across experiments, antibiotic selection was employed to enrich co-transfectant cells in the subsequent experiments and examined the efficacy of Tet induction and the extent of silencing over an extended period of time. The results showed that after 17 days of antibiotic selection, TetO -regulated mCherry gene expression remained inducible, and YFP silencing was detectable. In addition, it was observed that the expression rates reached their minimum on day 4 - as the unintegrated plasmids were eliminated from the cells - and remained constant thereafter. To demonstrate the feasibility of employing these systems with stable cell lines, a line containing plasmids pIL3565 and pIL3529 was isolated and evaluated its induci-bility. The line exhibited effective induction with Tet.

[0074] To demonstrate the performance of the novel systems with a known synthetic lethal pair, they were tested with the BRCA1 and RAD18 genes. For each of the novel systems, four stable cell lines were isolated: control, silenced for either BRCA1 or RAD 18 or both genes. By counting the number of dead or dying cells stained with propidium-iodide and / or, depending on the cell line used, a caspase-based apoptosis-specific dye (the RPE1 cells used lack this type of apoptosis due to the knockout of p53), synthetic lethality between the two genes was shown, in line with the literature. Three types of treatment were used as external stimuli for synthetic lethality: CP (cisplatin), UV, and HU. Taglialatela et al. ((2021) REVl-Polzeta maintains the viability of homologous recombination-deficient cancer cells through mutagenic repair of PRIMPOL-de-pendent ssDNA gaps. Mol Cell, 81, 4008-4025 e4007.) and Cybulla et al. ((2024) A RAD18-UBC13-PALB2-RNF168 axis mediates replication fork recovery in BRC Al -deficient cancer cells. Nucleic acids research) demonstrated in multiple cell lines that the combined loss of BRCA1 and RADI 8 expression is synthetically lethal under untreated conditions. In RPE1 cells, the same phenomenon was observed: cells lacking both BRCA1 and RAD 18 expression showed significantly lower viability compared to the controls. In U2OS cells, however, no significant change in viability under untreated conditions was observed. This difference may be explained by the lower efficiency of silencing compared to knockout. While in RPE1 cells one of the genes was stably knocked out, in U2OS cells both genes were silenced, and some residual activity may have protected the cells from death. However, this residual activity was not sufficient to prevent cell death in the presence of the DNA damaging agent CP. With CP treatment, a synthetically lethal relationship in both RPE1 and U2OS cell lines was observed. In the case of UV treatment, it was found that loss of the other gene did not further sensitize BRCA1- or RAD18-silenced cells in either of the cell lines used. In RPE1 cells, the effect of HU treatment at three concentrations was examined. It was found that while HU had an effect in each cellline, the proportion of dead cells increased greatly in the cells lacking both BRCA1 and RAD 18 expression.

[0075] Another aim was to improve our systems to operate effectively with transient transfection, without the need to isolate stable cell lines. To achieve this, plasmids were modified to facilitate the tracking of transfectants / co-transfectants. Fluorescent markers were integrated into the plasmids to enable yellow fluorescence of only the transfected cells in the single-transfection system and the co-transfected cells in the co-transfection system. To prove the functionality of this modification, the synthetically lethal interaction between BRCA1 and RAD 18 was tested using the modified co-transfection system with CP treatment. The evaluation was the same as for the base systems, except that the statistics only included the cells that showed yellow fluorescence. It was demonstrated that viability was significantly lower in the double-silenced cells than in the other samples, but only in the CP -treated samples, in line with the results of the experiment using stable U2OS cell lines.

[0076] To better understand the molecular background of the synthetically lethal relationship between BRCA1 and RAD 18, further functional assays were performed. First, the DNA fiber assay was used to investigate the dynamics of replication forks. A two-hour long 4 mM HU treatment was performed in between the 15-minute-long labeling with Idu and CldU, respectively. It was observed that the knockout of BRCA1 or the silencing of RAD 18 did not increase the percentage of stalled replication forks. However, in the case when cells lacked both BRCA1 and RADI 8, the percentage of stalled forks increased significantly. The results prove that both BRCA1 and RAD 18 participate in the rescue of stalled replication forks; however, only the cells that lack both genes have defects in fork recovery. The BrdU alkaline comet assay was also used to evaluate the difference between the DNA repair capacity of cells lacking one or both genes. It was found that the loss of either gene results in an increased level of DNA damage and, in addition, cells lacking RADI 8 accumulate more DNA damage than cells lacking BRCA1. This difference may be due to the fact that translesion synthesis, which is dependent on RAD 18, is essential during S-phase. However, it is also important to note, that RPE1 cells in which the BRCA1 gene is stably knocked out may have already adapted to the loss of BRCA1 using alternative pathways such as the overexpression of RAD18. This hypothesis is further strengthened by the comet assay in RPE1 cells, where BRCA1 silencing was used instead of knockout, and the results of the viability assays in U2OS cells. Cells lacking both genes accumulated even more damage in the comet assays. These observations were true under both untreated and HU-treated conditions. After six hours of recovery following HU treatment, differences in the DNA repair capacity of the cell lines were observed. Compared to controls, cells lacking BRCA1exhibited an elevated level of unrepaired damage, possibly due to the cells’ inability to repair double-stranded breaks and a reduced ability to recruit RAD18. RAD18-silenced cells were able to repair less damage since RAD 18 has an important role in PRIMPOL-generated gap filling in the G2 phase. Loss of RAD 18 in BRCA1 knockout cells abrogated DNA repair even more, since BRCAl(-) cells are highly dependent on RAD18. The accumulation of DNA damage in these cells is the result of a combination of different factors, including the inability to restore replication forks and fill ssDNA gaps. Furthermore, unrepaired ssDNA breaks and gaps are converted into DSBs that are not amenable to repair due to the absence of functional homologous recombination. To prove these hypotheses, pRPA immunostaining was used on RPE1 cells under conditions identical to those employed in the BrdU alkaline comet assays. It was observed that the number of cells containing more than five pRPA foci was slightly increased compared to the control in the BRCA1 knockout samples but only if the cells were treated with HU. The silencing of RAD 18 had a more pronounced effect, increasing the number of pRPA foci even in absence of treatment. A higher rise in the percentage of pRPA foci -containing cells in BRCA1 KO cells was observed following RAD 18 silencing, resulting in significant differences compared to the control both in HU-treated and untreated samples. These findings reinforce the hypothesis derived from the comet assay, namely, that the primary cause of synthetic lethality between the two genes is the accumulation of unfilled ssDNA gaps, which may subsequently transform into DSBs, enhanced by the cell’s observed deficiency in reinitiating stalled replication forks.

[0077] In the light of these results, it can be concluded that the new systems can be used to investigate synthetic lethality between gene pairs. The systems provide the opportunity to study synthetically lethal interactions in detail since - due to the inducibility of the systems - cell death does not occur prematurely, and thus phenotypic assays can be performed to describe the interaction between the genes.

[0078] EXAMPLES

[0079] 1. Materials and methods

[0080] 2.1 Plasmid constructs

[0081] To construct vectors carrying different antibiotic and fluorescent markers (Table 1), the Gibson assembly strategy was used (Gibson, D.G. (2009) Synthesis of DNA fragments in yeast by one-step assembly of overlapping oligonucleotides. Nucleic acids research, 37, 6984-6990.).

[0082] Table 1A

[0083]

[0084] Table IB

[0085]

[0086] Table 1C

[0087]

[0088]

[0089] Silencing DNAs were synthesized as long oligos (Table 2), annealed, and cloned into the vectors as Bglll-Hindlll fragments. Each step was verified by Sanger sequencing.

[0090] Table 2 List of primers used to generate gene-silencing constructs

[0091]

[0092]

[0093] 2.2 Cell culture, transfection

[0094] U20S cells were purchased from Sigma-Aldrich, while the two strains of the RPE1 cell line (RPE1 p53- / ~ and RPE1 p53- / ~ BRCA1- / -) were provided by Alan D. D’Andrea (Dana-Farber Cancer Institute, Boston). U2OS cells expressing the repressor of the Tet resistance element (TetR) were isolated by the integration of the pCMV-TetR-SV40pA construct into the AAVS1 region of the U2OS cell line via Cas9-aided targeted insertion. U2OS and U2OS TetR cell lines were cultured in DMEM (Dulbecco’s Modified Eagle’s Medium) (Capricorn), while RPE1 cells were grown in DMEM-F12 (1:1 mixture of DMEM and Ham’s F12 media) (Thermo Fisher Scientific), both supplemented with 5% Fetal Bovine Serum (FBS) (Gibco) and 1% pen-icillin-streptomycin solution (Coming).

[0095] U2OS and U2OS TetR cells were co-transfected using JetOPTIMUS (Polyplus-transfection), according to the manufacturer’s instructions, with equal amounts (1-1 pg) of plasmids. All RPE1 cell lines were transfected using Xfect (Takara) following the manufacturer’s instructions. Transfected cells were cultured in their appropriate mediums supplemented with the combination of antibiotics based on the resistance genes of the co-transfected plasmids (Table 3). Table 3 List of isolated stable cell lines

[0096]

[0097]

[0098] 2.3 Silencing tests

[0099] Cells (U20S and U20S TetR) were plated in 6-well plates (containing 22x22 mm glass coverslips (Menzel-Glaser)) in duplicates and were co-transfected at 80% confluence with pIL3565 and one of the three shYFP silencing plasmids (pIL3484, pIL3519, or pIL3529). One of the duplicate samples was induced with Tet (1 pg / ml) right after the transfection. Two days after transfection, cells were fixed in 4% paraformaldehyde (PF A) and stained with Hoechst 33342 (Thermo Fischer Scientific). Samples were scanned with an ImageXpress Micro Confocal high-throughput microscope (Molecular Devices) and analyzed with the MetaXpress 6.7.0.211 software package (Molecular Devices).

[0100] 2.4 Validation of YFP silencing by Western blot

[0101] The first steps of the experiment are identical to those of the silencing tests. On the second day after transfection, cells were collected and lysed in 30 pl 2X SDS buffer, then boiled at 100 °C for 10 min. Samples were loaded on 10% denaturing SDS gel and electrophorized for 45 min. Proteins were transferred to nitrocellulose membrane (Amersham) overnight at 150 mA. Nextday, the membrane was blocked in 5% BSA solution for 2 h then incubated with the primary antibody (anti-GFP) (Invitrogen) in 500x dilution for 1.5 h, followed by a washing step with lx TBS-Tween solution three times for 5 min each. After the washing steps, the membrane was incubated with the secondary antibody (anti-rabbit HRP conjugated) (Invitrogen) in 3000x dilution for 45 min, then washed with lx TBS-Tween five times for 5 min each. For labeling, WesternB right ECL HRP Substrate (Advansta) was used, and the signal was detected with an Alliance Q9 Advanced chemiluminescence imager (Uvitec). Signal intensities were quantified with the Image Studio Lite 5.2 software package (LLCOR Biosciences).

[0102] 2.5 Validation of long-term inducibility and detection of changes in the expression rate via livecell imaging

[0103] For the validation of long-term inducibility, U2OS TetR cells were transfected with a 1 : 1 mixture of the pIL3529 and pIL3565 plasmids in a 6-well plate at 80% confluence. Next day, the transfected cells were distributed equally between all wells of a 24-well plate. The first two wells served as absolute negative controls. The remaining wells were induced with Tet on each following day until day 17 following transfection. All induced wells were kept on Tet until the end of the experiment (day 18). All wells were stained with Hoechst33342 and scanned on the last day with an ImageXpress Micro Confocal high-throughput microscope (Molecular Devices).

[0104] The experiment for detecting changes in the expression rates was carried out similarly to the validation of long-term inducibility with the only difference that every well was stained and scanned the day after induction.

[0105] 2.6 Isolation of stable cell lines for the co- and single-transfection systems

[0106] U2OS TetR cells were plated in 6-well plates then co-transfected at 80% confluence with two plasmids carrying the appropriate silencing constructs. Selection with combined puromycin (1 pg / ml) and hygromycin (100 pg / ml) was started the day after transfection and was maintained until the end of the experiments. The medium on the cells was changed every second day, and the cells were grown until 95-100% confluence in four 10-cm petri dishes.

[0107] RPE1 p53- / ~ and RPE p53- / -BRCAl- / ~ cells were plated in 6-well plates and transfected at 80% confluence with the silencing plasmid carrying the YFP-NLS fluorescent marker gene. The following day, the medium was changed to selective medium (300 pg / ml neomycin). Cells were grown until 100% confluence in four 10-cm Petri-dishes. Medium was changed every second day.

[0108] The isolated stable cell lines are shown in Table 3.

[0109] 2.7. Induction test on a control stable cell lineU2OS TetR P-(YFP-NLS)-TetO:mCherry H-shYFP cells were plated on four coverslips in a 6-well plate. After plating, the first well served as a negative control, and each day one new well was induced with and kept on Tet, until the end of the experiment (day 4). On the last day, all samples were fixed with 4% PF A, stained with Hoechst33342, and scanned.

[0110] 2.8. Validation of RAD18 silencing in stable cell lines

[0111] Cells were grown in wells of a 6-well plate for five days. For each cell line, there were three different growing conditions: 1. Grown without Tet; 2. Kept on Tet-containing medium on the last three days of growing; 3. Kept on Tet-containing medium for five days). On the fifth day, 6xl05cells were collected from each well and lysed in 60 pl 2x SDS buffer then boiled at 100 °C for 10 min. The following steps of the western blot were the same as mentioned for the validation of YFP silencing by western blot. The only difference was the primary antibody used. In this experiment, we used anti-RAD18 (18333-1-AP; Proteintech) as the primary antibody in lOOOx dilution.

[0112] 2.9 Viability assays with the co- and single-transfection systems with stable cell lines Three thousand cells from each stable cell line harboring the different combinations of silencing plasmids were plated in 6 wells of a 24-well plate. All wells were filled with induction medium containing Tet (1 pg / ml). Tet induction was maintained from the plating of the cells until the end of the experiment. Cells were grown for 3 days, then half of the samples were treated with the DNA-damaging agent CP (30pM) (Adooq Bioscience), UV (40 J / m2), or with HU (0.125; 0.25; 0.5 and ImM). Two days after treatment, cells were stained with Hoechst33342 (20 pM), propidium-iodide (1 pg / ml), and / or CellEvent Caspase-3 / 7 Red detection reagent (Invitrogen) for 30 min and subsequently scanned. This experiment was executed with both RPE1 and U2OS TetR cell lines (Figure 1).

[0113] 2.10 Viability assays with the high-throughput microscopy-compatible transient systems U2OS TetR or RPE1 p53- / - and RPE1 p53- / - BRCA1- / - cells were co-transfected or transfected, respectively, one day after plating into a 6-well plate with various combinations of silencing plasmids carrying fluorescent markers. After transfection, they were induced with 1 pg / ml of Tet until the end of the experiment and plated into 6-6 wells of a 24-well plate. Trans-fectants were then enriched by a one-day selection process (1 pg / ml puromycin + 100 pg / ml hygromycin for U2OS and 300 pg / ml neomycin for RPE1 cells). Four days after plating, cells were exposed to the DNA-damaging agent CP (30 pM). The day after the treatment, the samples were stained with Hoechst33342 and - depending on the cell line - CellEvent Caspase-3 / 7 Detection Reagent (Thermo Fisher) (U2OS TetR) and / or propidium iodide (RPE1) and subsequently scanned (Figure 2).2.11 Alkaline BrdU comet assay

[0114] A modified version of the previously published method was employed (Morocz, M., Gali, H., Rasko, I., Downes, C.S. and Haracska, L. (2013) Single cell analysis of human RAD18-de-pendent DNA post-replication repair by alkaline bromodeoxyuridine comet assay. PloS one, 8, e70391.). Exponentially growing RPE1 cells (stable p53- / ~ and p53- / ~ BRCA1- / - lines carrying pIL3517 or pIL3534) were plated at a density of 3xl04cells per well of a six-well plate. The cells were grown for three days in induction medium (containing 1 pg / ml Tet in DMEM-F12 medium). Following a three-day incubation period, the cells were labeled in a medium containing 25 pM BrdU for 30 min. After two thorough washes with PBS, half of the samples were treated with 4 mM HU for 3 h. Then the HU solution was removed from the samples, and the cells were washed twice with PBS. Half of the HU-treated and untreated cells were collected and embedded in 70 pl 0.75% low-melting agarose in duplicates on comet slides. The remaining cells were collected and embedded after a 6-hour recovery period in induction medium. To facilitate cell lysis, the previously published lysis solution (2.5 M NaCl, 100 mM EDTA, 10 mM Tris [pH 10], 1% Triton X-100, and 0.5% N-laurylsarcosine sodium salt) was used in Coplin jars overnight. For electrophoresis, the slides were placed into a tank containing cold alkaline electrophoresis buffer (0.3 M NaOH, 1 mM EDTA, pH 13) and were incubated for 40 min to let the DNA unwind. Subsequently, slides were electrophorized for 30 min at 1 V / cm and washed three times with neutralization buffer (0.4 M Tris-HCl, pH 7.4) and two times with PBS on a paper tray. Subsequently, the samples were blocked for 20 min in a solution of 1% horse serum in PBS. All steps of the immunostaining were carried out at room temperature, in the dark, for 2 h. Both the primary and secondary antibodies were diluted in blocking solution. First, the slides were incubated with 50 pl mouse monoclonal anti-BrdU anti-body in a 1:300 dilution. After a thorough washing with PBS, the slides were incubated with 50 pl goat polyclonal anti-mouse antibody labelled with Alexa Fluor 488 in a 1:400 dilution. Samples were counterstained with propidium-iodide and kept in a humidified box at 4 °C until microscopy. Microscopic scanning was performed with a Zeiss Axioscope Z2 fluorescent microscope using the automated scanning platform of Metasystem, while the evaluation was done with the Metasystems Neon Metafer4 software. In all three independent experiments, 100-300 comets / slides were analyzed.

[0115] 2.12 DNA fiber assay

[0116] For DNA fiber assays, cells were first induced with Tet for three days. On the fourth day, cells were pulse-labeled for 15 min with 5-chloro-2’-deoxyuridine (IdU, 20 pM) and treated with HU (4 mM) for 2 h after washing twice with PBS. HU was washed out with PBS twice thenlabeled with 5-chl oro-2’ -deoxyuridine (CldU, 200 pM) for 15 min. Cells were collected, centrifuged, then resuspended in 40 pl PBS. 3 pl of cell suspensions were dropped onto a glass slide and left to dry at RT for 5 min then mixed with 7 pl lysis solution (0.5% SDS, 50 mM EDTA, 200 mM Tris-HCl (pH 7.5)) and left to dry at RT for 5 min again. After the drying steps, the slides were tilted at a 30° angle for fiber spreading. Samples were dried at RT for 15 min then fixed in a 3:1 methanol :acetic acid solution for 30 min, washed with MQ water, dried and stored at -20 °C until immunostaining, but at least for one day. Before immunostaining, the samples were rehydrated with MQ water for 3 min then denatured in HC1 (2.5 mM) at RT for 60 min. After washing the slides three times with MQ water and then PBS, the slides were blocked with 5% BSA (in PBS solution) at RT for 1 h. Immunostaining of the samples was done with the primary antibodies anti-BrdU (mouse, 347580, BD Biosciences) and anti-BrdU (rat, ab6326, Abeam) in 150x and 200x dilutions, respectively, for 2 h at RT. After washing with PBS three times, the slides were incubated with the secondary antibodies AlexaFluor546 conjugated anti-mouse (goat, A21123, ThermoFisher Scientific) and CF488A conjugated antirat (goat, 20023, Biotium) in 300x dilution for 2 h at RT. Slides were washed three times with PBS then mounted with Fluoromount-G Mounting Medium (00-4958-02, Invitrogen). For image acquisition, a Zeiss Axioscope Z2 fluorescent microscope was used, with the Zen Blue 2.6 software. Images were taken with a 63x oil immersion objective. At least ten images were acquired from each slide; areas were selected according to the AlexaFluor546 signal (labeling before the HU treatment). At least 100 individual fibers were counted for each sample.

[0117] 2.13 Phospho-RPA immunostaining

[0118] For immunostaining, cells were plated on coverslips in 6-well plates (two wells from each cell line) then induced with Tet for the rest of the experiments. After three days of induction, half of the samples were treated with HU (4 mM) for 3 h. Six hours after treatment, cells were lysed in pre-extraction buffer (0.5% TritonX-100, 20 mM HEPES [pH 7.4], 3 mM MgC12, 50 mM NaCl, 300 mM sucrose) for 5 min at 4 °C. Lysed samples were fixed first with 4% PF A for 20 min at 4 °C, washed with PBS, then fixed with methanol for 20 min at -20 °C. Cells were blocked overnight at RT with 5% FBS in 0.1% TritonX-100 in PBS solution. Samples were immunostained with the anti-pRPA primary antibody (rabbit, Abl09394, Abeam) in 500x dilution for 1 h at RT. Cells were washed four times with 0.1% TritonX-100 in PBS, then incubated with the AlexaFluor568-conjugated anti-rabbit secondary antibody (goat, Al 1011, Invitrogen) for 1 h at RT in 500x dilution. Samples were washed four times with 0.1% TritonX-100 in PBS, then stained with Hoechst33342 (20 pM) for 10 min, then subsequently scanned. 2.14 Microscopic scanning and image analysisSamples were scanned with an ImageXpress Micro Confocal microscope (Molecular Devices) with Plan Fluor ELWD 40X and Plan Apo X 4X objectives. For scanning, four filters were used: TL100: white light; DAPI: 377 / 447 nm; FITC: 475 / 536 nm; and Texas Red: 560 / 624 nm. Excitation times were 100 ms for all filters. For the coverslips, at least 49 photos were taken using the 40X objective, while for the 24-well plates nine photos per well were taken with the 4X objective.

[0119] All microscopic photos were analyzed with the MetaXpress 6.7.0.211 software package (Molecular Device) using optimized custom modules. Settings for the used modules are shown in Table 4.

[0120] Table 4 Parameters for microscopic scanning

[0121]

[0122]

[0123] 2. Results

[0124] 2.1 Construction of silencing plasmids

[0125] Our gene-silencing vectors (Figure 3; Table 1) are based on the pSCl plasmid backbone, in which the pCMV-EGFP fragment was replaced by a Tet-inducible pHl promoter carrying the TetO operator sequence (pIL3305). To create a plasmid in which we can control the expression of the silencing shDNA from this promoter (pIL3430), we inserted the TetR gene into pIL3305, which provides the constitutive repression of pHl in the absence of Tet. For double silencing, we designed a co-transfection system from pIL3305, consisting of the pair of plasmids pIL3476 and pIL3527, carrying the puromycin and hygromycin selection markers, respectively. Finally, the shDNA constructs were integrated into the plasmids as Bglll-Hindlll fragments annealed from complementary oligo sequences.

[0126] To track transfected cells and adapt the system to high-throughput microscopy, the plasmids were further supplemented with fluorescent marker genes. We incorporated an NLS-tagged YFP into the single-transfection system plasmid pIL3430 carrying the TetR repressor gene, resulting in pIL3569. To track co-transfected cells, our silencer plasmids pIL3476 and pIL3527 were combined with the MEK1-V1 and TRIB1-V2 split-Venus constructs, respectively, resulting in the pIL3564 and pIL3568 plasmids (Table 1). Since the Mekl-Vl / Tribl-V2 complex shows strong, considerably homogenous nuclear localization, the co-transfected cells can be followed throughout the transfection and antibiotic selection processes, and co-transfection efficiency can be calculated from the ratio of YFP-positive / total cell number.

[0127] 2.2 Construction of the test plasmids

[0128] For the evaluation of the silencing vectors, we cloned a YFP -targeting shDNA into the plasmids pIL3430, pIL3476, and pIL3527, creating pIL3519, pIL3484, and pIL3529, respectively (Table 1). As a test vector, we created a plasmid harboring a pCMV-TetO-mCherry-NLS construct, inwhich the expression of the mCherry gene was under the control of the TetO operator. Additionally, the plasmid included a constitutively expressed YFP-NLS construct (pIL3565) (Table 1).

[0129] 2.3 Functional testing of silencing in wild-type U2OS and U2OS TetR cell lines

[0130] All silencing plasmids (harboring different resistance genes) and the fluorescent-marker plasmid carrying the YFP and the mCherry genes (pIL3565) were tested in wild-type U2OS and U2OS TetR cell lines (Figure 12).

[0131] Without Tet induction, both the YFP and the mCherry genes are expressed in wild-type U2OS cells transfected with the marker plasmid only (Figure 13 A), while U2OS TetR cells only express YFP (Figure 4A). In wild-type U2OS cells in which pIL3565 is co-transfected with shYFP-harboring plasmids that lack the TetR gene (pIL3484 or pIL3529), YFP is silenced, but the mCherry gene is expressed (Figure 13B,C). Conversely, when we use pIL3519 - due to the presence of the TetR protein encoded in this vector - both the shYFP and the mCherry expressions are downregulated, resulting in YFP production only (Figure 13D). In the case of U2OS TetR cells when we use any of the pIL3484, pIL3519, or pIL3529 plasmids, due to the constant expression of the TetR protein, both the shYFP-silencing construct and mCherry are repressed, resulting in yellow fluorescence only (Figure 4B,C,D). When we induce our samples with Tet, we observe no difference compared to the uninduced samples in wild-type U2OS cells, expect in the case of plasmid pIL3519, where the YFP signal disappears due to the silencing, and the mCherry signal becomes visible (Figure 13E,F,G,H). In the case of the U2OS TetR cell line, the shYFP and mCherry constructs are expressed upon induction, resulting in the disappearance of yellow and the appearance of red fluorescence (Figure 4E,F,G,H).

[0132] All experiments were repeated three times. Although we could see some variation in the transfection efficiencies between experiments and between the different plasmid combinations, we observed that silencing was clearly visible with all three plasmids tested, and the inductions were effective (Figure 14 and Figure 5A-C). In addition, we validated the silencing in U2OS TetR cells via Western blot. We compared the Tet-induced samples with their non-induced counterparts. By quantifying the pixel intensities of the bands, we observed a significant decrease in the YFP signal in the Tet-induced samples as compared to their controls: pIL3484: 84%; pIL3519: 92%; and pIL3529: 86% (Figure 15).

[0133] 3.4 Testing long-term inducibility of expression from transiently transfected plasmids and stable cell lines

[0134] Although we were able to successfully suppress YFP expression, transfection efficiencies varied between experiments, and the frequency of co-transfection of the marker and the silencingvectors was quite low. In order to circumvent this problem, we isolated stable cell lines instead of employing transient transfection. First, to monitor how the duration of the selection and the timing of Tet treatment affect the inducibility and efficiency of silencing, two experiments were performed. We monitored the expression rate changes over the time course of the induction by detecting the appearance of the mCherry signal and evaluated the time and strength of silencing via the disappearance of YFP fluorescence. U2OS TetR cells were transfected with plasmids pIL3565 and pIL3529 and kept on medium containing puromycin and hygromycin in two 24-well plates. In the first plate, all wells were scanned for mCherry and YFP fluorescent signals at the end of the experiment, while in the other one we scanned the wells the day after induction. In the first experiment, we induced a new well every day, starting from the day after transfection. At the end of the experiment, all samples were scanned. We observed that the inducibility of the TetO-regulated constructs does not change significantly over time, and - even after 17 days of growth - when induced with Tet, the cells maintained their capacity to express mCherry as well as the shDNA (Figure 6A). In the case of the second plate, we added Tet to one new well each third day, starting from the day of transfection. Samples were scanned one day after induction. This way, we could monitor changes in fluorescence intensity levels and thus in the fluorescent protein expression rates over the course of time. We observed that the mCherry and YFP signal intensities reached a minimum level around the fourth day and remained constant until the end of the experiment (Figure 6B). Both experiments demonstrated that cells cultured under appropriate selection can maintain the plasmids for at least 17 days without significant loss of function.

[0135] Furthermore, we isolated a cell line by inserting the pIL3565 and pIL3529 plasmids into the AAVS1 region of the cells’ genome. Similarly to the previous long-term selection experiments, all cells of the stable cell line showed only yellow fluorescence in the absence of induction. However, two-three days after the addition of Tet, mCherry expression and YFP-silencing reached their approximate maximum (Figure 6C). In view of these results, we isolated stable cell lines carrying pairs of silencing constructs (Table 3).

[0136] 3.5 Validating the systems with the BRCA1-RAD18 gene pair

[0137] To validate our systems, the BRCA1-RAD18, a previously published synthetically lethal gene pair was used. We decided to use both the single-transfection and the co-transfection version of the base system, and the co-transfection type of the modified, fluorescently tagged system. For the base system, we first established stable cell lines carrying the silencing constructs (Table 3). RAD 18 silencing was checked with western blots (Figure 16). Then, equal numbers of cells were plated from each cell line into 6-6 wells. The expression of the shDNAs was induced bythe addition of Tet. Half of the samples were subjected to a DNA-damaging treatment (CP or UV) or the replication stress-inducing agent HU, and a cell survival assay was performed using a caspase-based apoptosis-specific dye and / or propidium-iodide, depending on the cell line used. RPE1 cells do not have the specific caspase-based apoptotic pathway due to the lack of p53.

[0138] The experiments were evaluated as follows: if there was no difference between the viability of the cells carrying the single silencing constructs and that of the cells with both genes silenced, then the gene pair is not synthetically lethal, at least under the conditions of the DNA-damaging agent used. The genes are non-conditionally synthetically lethal if the double-silenced cells display lower viability than those containing one or two shControl constructs, in both the treated and untreated samples. If viability differs only between the treated samples, then the gene pair is considered to be conditionally synthetically lethal (Figure 1). For the evaluation, the averages of the parallel triplicates were used.

[0139] In the single-transfection system, we used the RPE I( / U3- -) and the RPE1 (p53- / / ~; BRCA1- / -) stable cell lines carrying either shRAD18 or shControl constructs (Table 3). As we detected a signal in less than 1% of the cells using the caspase-based dye - probably due to the lack of p53, a key regulator of apoptosis - we used propidium-iodide staining to visualize dead cells. In the untreated samples, the percentage of dead cells was four times higher in the p53- / ~ BRCA1- / - shRAD18 cells than in the other three cell lines. The effect of CP treatment (30 pM) was visible in all four cell lines: the number of dead cells increased by approximately 30% in the p53- / ~ BRCA1- / - shRAD18 cell line and by approximately 15% in the others (Figure 7A). We conclude that the BRCA1-RAD18 gene pair is non-conditionally synthetically lethal, and the use of CP enhances this phenomenon. When we changed the treatment to UV (40 J / m2), we could still see the synthetic lethality under untreated conditions; however, upon UV treatment we saw no significant difference between cells lacking both BRCA1 and RAD 18 and the other cell lines (Figure 7B). We repeated the experiment, keeping the same experimental setup but employing a different treatment, HU, at three different concentrations (0.25 mM, 0.5 mM, and 1 mM) instead of CP or UV. The non-conditional nature of the synthetic lethality between the BRCA1 and the RAD18 genes was still visible, as well as the dose dependence curves, and HU showed the greatest effect in cells lacking both BRCA1 and RAD18 (Figure 7C). We also counted the overall cell numbers in each well (Figure 17).

[0140] We repeated the experiments with CP and UV treatment in U2OS TetR cells with four cell lines (double-control, 7?7?G47-silenced, 7MD7S-silenced and double-silenced). Interestingly, we did not observe a significant effect, just a slight increase in the ratio of apoptotic cells compared tothe controls under untreated conditions, just as in the case of RPE1 cells. This difference may arise from some residual activity left in the cells, as silencing does not have a hundred percent efficiency, unlike knockout. However, in the case of CP treatment (30 pM), we observed a significant increase in the number of dead cells in BRCA1 and RAD18 co-silenced samples compared to the controls (Figure 8A). When we treated the cells with UV, we did not observe an additional effect, when both RAD18 and BRCA1 were silenced, just as with the RPE1 cells (Figure 8B).

[0141] Furthermore, we intended to evaluate the modified versions of the silencing constructs that express marker genes enabling the detection of fluorescence in transfected cells via high-throughput microscopy, thus making the system suitable for transient transfection and eliminating the need to isolate stable cell lines (Figure 2). We used the same set of confirmed pairs of silencing constructs as in the previous experiments with U2OS TetR cells to validate this modified version of our system as well. To enrich the transfected cells, a short-term one-day selection step was included. During the evaluation process, we counted only those cells that exhibited yellow fluorescence and thus were co-transfected. This enabled us to exclude the untransfected cells from our statistical analysis, making the evaluation of the results more accurate.

[0142] The evaluation criteria for synthetic lethality were the same as in the previous experiments, taking into account the ratio of either apoptotic or necrotic cells compared to living cells in treated and control cells. To stain the dead cells, both propidium-iodide and caspase-based dye were used.

[0143] We observed no significant difference between any of the cell lines in the case of the untreated samples just like in the case of stable cell lines. CP treatment caused no significant increase in the percentage of dead cells in the control and single-silenced samples; however, it resulted in a more than sixfold increase in cell death in the double-silenced sample (Figure 8C). With this transient system, the effect of CP seemed smaller, but this difference could be due to the low sample number.

[0144] 3.6 Describing the synthetic lethal interaction between BRCA1 and RAD 18 Subsequently, functional assays were employed to elucidate the underlying mechanism of the synthetic lethality observed between BRCA1 and RAD18. Firstly, the dynamics of the replication fork progression upon HU treatment was examined using a DNA fiber assay. Cells were pulse-labeled with 20 pM IdU (red) for 15 min, followed by a 2-hour treatment with 4 mM HU, and a 15-minute pulse of 200 pM CldU (green). In the case of replication forks being unable to resume DNA synthesis following HU treatment, only IdU is incorporated, resulting in fiberscomprising a single continuous red tract. Those forks that were able to restart DNA synthesis and incorporated both thymidine analogs, resulted in fibers that appeared as contiguous red-green lines. We observed that following HU treatment a proportion of replication forks were unable to restart DNA synthesis even in the control cell line. The absence of BRCA1 did not result in a significant increase in the number of stalled forks compared to the control, and unexpectedly, the silencing of RAD 18 did not have a notable impact. However, in cells lacking both, a significant reduction in the percentage of restarted forks was observed (Figure 9). Next, alkaline comet assays were performed on cell lines treated with HU, which depletes the nucleotide pool, resulting in the arrest of replication forks, potentially leading to single- and double-stranded DNA breaks and ssDNA gap formation. Firstly, the appropriate recovery time following a three-hour treatment was defined in the RPE1 p53- / ~ BRCA1- / - shControl cell line. Samples were taken immediately post-treatment and 3, 6, 9, and 24 hours after washout of HU. A significant difference was observed between the HU-treated and untreated samples at 0, 3, and 6 hours of recovery. In the samples that underwent a 9- or 24-hour recovery period, the treated samples exhibited no statistically significant difference from their respective untreated controls (Figure 18 A). The experiment was repeated with the RPE1 p53- / ~ shControl cell line with only 0, 3, and 6 hours of recovery (Figure 18B). A significant difference was observed at all three recovery times. However, as anticipated, the differences between treated and untreated samples were less pronounced than in the case of the BRCA1 KO cell line, likely due to the presence of the intact BRCA1 gene. Based on the aforementioned results, the 6-hour recovery time was selected for the subsequent experiments including all four cell lines from the preceding viability studies. This way, we could observe the repair efficiency of the various cell lines during and shortly after S-phase. For the experiment, the cells were labeled with BrdU for 30 min, and then they were treated with HU for an additional 3 hours. Samples were collected at two timepoints: immediately following HU treatment and after a six-hour recovery period (Figure 10A). In the absence of recovery time, considerable differences were observed between treated and un-treated samples, which decreased across all cell lines after 6 hours of recovery, albeit to varying degrees. Upon HU treatment without recovery time, only the BRCA1 KO cells did not accumulate significantly more DNA damage compared to the double-control line. However, in the absence of treatment, we observed that the BRC Al -deficient cells accumulated significantly more DNA damage than the BRCA1 -proficient cells. In both the untreated and the HU-treated samples, RAD 18 silencing had a significant effect compared to the previously mentioned cell lines. However, the accumulated amount of DNA damage was slightly higher in the BRCA1 KO cells that were also RAD 18 silenced, which difference was more prominent uponHU treatment. Following a six-hour recovery period, the differences between the four cell lines became more evident. In the case of the untreated samples, all cell lines exhibited a reduction in DNA damage after the six hours recovery period, in comparison to those samples that did not undergo this recovery process. This phenomenon may be explained by the fact that BrdU labeling by itself causes replication stress. Furthermore, it was evident that the knockout of BRCA1 resulted in a slower repair process compared to the double-control cell line (RPE1 p53- / - shControT), leading to an increased number of unrepaired DNA breaks and gaps. We observed that the repair process was even slower in the single 7 D7S-silenced cells. Nevertheless, the highest degree of unrepaired DNA damage was observed in the BRC Al -deficient and RAD 18-silenced cells (Figure 10B and C). Our findings suggest that RAD 18 may play a more significant role in reducing or preventing the increase of DNA damage during the stalling of the replication forks caused by HU treatment. However, the lack of BRCA1 appears to further enhance the accumulation of gaps and DNA breaks. Following a six-hour recovery period, which coincides with the end of the S-phase or the subsequent G2 phase, both RAD 18 and BRC Al are necessary for the repair of DNA damage. However, in the case of the simultaneous loss of both, the quantity of unrepaired DNA will be significantly higher, likely to result in an elevated level of cell death in comparison to the loss of either BRC Al or RAD 18 alone. This suggests that BRCA1 and RAD 18 fulfil distinct roles in the repair of DNA damage occurring during the G2 phase. We can also hypothesize that as a consequence of HU treatment the ratio of gaps and single-stranded breaks to double-stranded breaks is higher, resulting in a more pronounced effect of RAD 18 silencing than that observed in BRC Al knockout cells. This could also be due to the lack of RAD18-mediated translesion synthesis. It is possible that the further increase in DNA damage in cells lacking both functional genes could be due to unrepaired single-stranded breaks and unfilled gaps that are converted into DSBs, which cannot be repaired due to the absence of BRC Al and the inability to recover stalled replication forks.

[0145] Next, the experiment was repeated with only six hours of recovery time. However, in this case, the 7?7?G47-silenced cell lines (RPE1 p53- / - P-shBRCAl lA-shControl and RPE1 p53- / - P-shBRCAl N-shRadl8) were used in place of the BRCA1 knockout RPE1 cell lines. No statistically significant differences were observed between the knockout and silencing of BRCA1 in the alkaline BrdU comet assay experiments. However, slight differences were noted. The silencing approach appeared to enhance the accumulation of DNA in the comet tails to a greater extent than the knockout, compared to the control samples (Figure 19).

[0146] Furthermore, we sought to validate our comet assay-derived hypothesis that the primary forms of DNA damage responsible for the synthetically lethal interaction between BRCA1 and RAD 18are not DSBs, but rather ssDNA gaps (which may subsequently evolve into DSBs). To verify this hypothesis, phospho-RPA (pRPA) immunostaining was employed. The experimental setup was identical to that used in the comet assays, comprising a three-hour-long treatment with 4 mM HU, followed by six hours of recovery. We observed that the knockout of BRCA1 did not result in an increase in the number of pRPA foci -containing cells under normal conditions. However, upon HU treatment, a slight, statistically non-significant increase was noted in comparison to the HU-treated control cells. RAD 18 silencing resulted in a further elevation in the amount of pRPA foci, both in the presence and absence of HU treatment. Following the silencing of RAD 18 in BRCA1 knockout cells, a significant increase in the number of cells containing more than five pRPA foci was observed, both under untreated and HU-treated conditions, indicating the presence of an elevated number of ssDNA gaps (Figure 11 A and B).

Claims

CLAIMS1. A method for assessing a gene of interest (gene A) and a candidate synthetically lethal gene pair (gene B) of gene A, comprisingproviding a cell in which gene A is silenced, knocked out or knocked down,introducing a gene silencing construct targeting gene B under the regulation of an inducible transcription regulating sequence into the cell.

2. A method of screening for a synthetically lethal gene pair (gene B) of a gene of interest (gene A), comprisingproviding a cell in which gene A is silenced, knocked out or knocked down,introducing a gene silencing construct targeting gene B under the regulation of an inducible transcription regulating sequence into the cell.

3. A method for assessing the mechanism of synthetic lethality caused by a synthetically lethal gene pair, comprisingproviding a cell in which one gene of the pair (gene A) is silenced, knocked out or knocked down,introducing a gene silencing construct targeting the other gene of the pair (gene B) under the regulation of an inducible transcription regulating sequence into the cell.

4. The method according to any one of the preceding claims, wherein gene A is inducibly silenced by the introduction of a gene silencing construct targeting gene A under the regulation of an inducible transcription regulating sequence into the cell and inducing the silencing of gene A.

5. The method according to any one of the preceding claims, wherein the gene silencing construct targeting gene A or B comprises a DNA sequence coding for a shRNA molecule (shDNA sequence) the expression of which is capable of silencing gene A or B, respectively or the gene silencing construct targeting gene A comprises a shDNA sequence the expression of which is capable of silencing gene A and the gene silencing construct targeting gene B comprises a shDNA sequence the expression of which is capable of silencing gene B.

6. The method according to any one of the preceding claims, wherein the inducible transcription regulating sequence comprises a TetO regulated promoter, preferably TetO regulated pHl promoter or TetO regulated CMV promoter.

7. The method according to any one of the preceding claims, wherein the inducible transcription regulating sequence comprises a coding sequence for the repressor of the Tet resistance element, preferably a Tet repressor gene.

8. The method according to any one of the preceding claims, wherein the gene silencing construct is comprised in a transfecting vector further comprising a sequence coding a marker, preferably an antibiotic resistance gene or a gene coding for a fluorescent protein.

9. The method according to any one of the preceding claims, wherein the method further comprisesa) propagating the cell after the introduction of the gene silencing construct targeting gene B or the gene silencing construct targeting gene B and the gene silencing construct targeting gene A,b) inducing the silencing of gene A or B or both,c) screening the cell clones resulting from a) for viability after silencing gene A or B or both.

10. The method according to any of the preceding claims, wherein gene A is knocked out.

11. The pSCl-TetR-(Neo)-pHl-TetO, the pSCl-(Puro)-(ApR)-pHl-TetO, the pSCl-(Hygro)-(ApR)-pHl-TetO, the pSCl-(Hygro)-(ApR)-(TRIBl-Venus2)-pHl-TetO, the pSCl-(Puro)-(ApR)-(MEKl-Venus2)-pHl-TetO or the pSCl-TetR-(Neo)-(YFP-NLS)-pHl-TetO plazmid.

12. Use of a plasmid according to claim 11 for assessing a gene of interest (gene A) and a candidate synthetically lethal gene pair (gene B) of gene A.

Images