To identify novel regulators of transcription factors, we developed an approach to screen arrayed lentiviral or retroviral RNAi libraries using a dual-luciferase-based transcriptional reporter assay. This approach offers a quick and relatively inexpensive way to screen hundreds of candidates in a single experiment.
Transcription factors can alter the expression of numerous target genes that influence a variety of downstream processes making them good targets for anti-cancer therapies. However, directly targeting transcription factors is often difficult and can cause adverse side effects if the transcription factor is necessary in one or more adult tissues. Identifying upstream regulators that aberrantly activate transcription factors in cancer cells offers a more feasible alternative, particularly if these proteins are easy to drug. Here, we describe a protocol that can be used to combine arrayed medium-scale lentiviral libraries and a dual-luciferase-based transcriptional reporter assay to identify novel regulators of transcription factors in cancer cells. Our approach offers a quick, easy, and inexpensive way to test hundreds of genes in a single experiment. To demonstrate the use of this approach, we performed a screen of an arrayed lentiviral RNAi library containing several regulators of Yes-associated protein (YAP) and transcriptional co-activator with PDZ-binding motif (TAZ), two transcriptional co-activators that are the downstream effectors of the Hippo pathway. However, this approach could be modified to screen for regulators of virtually any transcription factor or co-factor and could also be used to screen CRISPR/CAS9, cDNA, or ORF libraries.
The purpose of this assay is to use viral libraries to identify regulators of transcription factors in a relatively quick and inexpensive manner. Aberrant transcriptional activity is associated with cancer and metastasis1,2,3,4,5,6, so targeting transcription factors in cancer cells is a promising therapeutic approach. However, transcription factors are often difficult to target pharmacologically7 and many are required for normal cellular function in adult tissues8,9,10. Targeting the cancer-associated pathways that aberrantly activate transcription factors to drive disease is a more feasible approach with the potential to have less severe side effects. The commercial availability of arrayed lentiviral and retroviral RNAi, CRISPR/CAS9, cDNA, or ORF libraries allows researchers to test the importance of numerous genes in a single experiment. However, a reliable readout for altered transcriptional activity is required.
Here, we describe the use of a dual-luciferase-based transcriptional reporter assay and arrayed lentiviral libraries to identify proteins that regulate transcription factors in cancer cells. In this assay, shRNAs that target cancer-associated genes are delivered to mammalian cancer cells via lentiviral transduction and cells are selected for stable integration using puromycin. The cells are next transfected with a reporter construct that expresses firefly luciferase driven by a promoter specific to the transcription factor that is being investigated and a control construct that expresses Renilla luciferase from a constitutively active promoter that is not responsive to the transcription factor being investigated. We demonstrate this approach with a proof-of-concept screen for regulators of YAP and TAZ, the critical downstream effectors of the Hippo pathway8,10,11. Abnormal activity of YAP and TAZ promotes several steps of the metastatic cascade11 and is observed in many cancers11,12,13. However, how YAP and TAZ become aberrantly activated in some cancer cells is not yet fully understood. YAP and TAZ do not bind DNA, but instead are recruited to promoters by other transcription factors. Members of the TEA domain (TEAD) family of transcription factors are the major binding partners for YAP and TAZ, and are critical for most YAP and TAZ-dependent functions. Our reporter construct expresses firefly luciferase from a YAP/TAZ-TEAD-responsive promoter and previous studies have demonstrated that it faithfully detects changes in YAP-TEAD and TAZ-TEAD transcriptional activity2,14,15.
Our approach is rapid, medium-throughput, and does not require screening facilities, automated robots, or deep sequencing of pooled libraries. The costs are relatively low and there are numerous commercially available libraries to choose from. The required equipment and reagents are also relatively standard in most laboratories. It can be used to screen for regulators of virtually any transcription factor if a luciferase-based reporter exists or is generated. We use this approach to screen shRNAs in cancer cells, but any cell line that can be transfected with reasonable efficiency could be used with any type of arrayed library.
NOTE: A schematic summary of this protocol is shown in Figure 1.
1. Lentiviral vector library preparation
NOTE: The demonstrated screen used an arrayed shRNA library purchased as glycerol stocks in 96-well plates, but libraries can also be assembled manually based on a list of candidates. Appropriate controls should be considered and included in any library. This includes a non-targeting control shRNA (shNTC), a control shRNA targeting the transcription factor being investigated, and if possible, an shRNA targeting firefly luciferase.
2. Packaging of the arrayed lentiviral library
NOTE: All work involving lentivirus, including packaging, infection, and subsequent culturing of infected cells should strictly follow the institutional biosafety rules and regulations.
3. Infection of the cells for the screen
NOTE: Human melanoma cells (A375) were used to demonstrate this approach, but this method can be applied to any adherent cells that infect with lentivirus. However, cell culture and plating conditions should be optimized for each cell line (see Discussion).
4. Seeding cells for transfection of dual-luciferase reporter
NOTE: A test transfection should be done to determine the optimal seeding density for each new cell line.
5. Transfection of dual-luciferase reporter
6. Quantification of dual-luciferase activity
Our YAP/TAZ-TEAD reporter construct (pGL3-5xMCAT (SV)-492,14,15) contains a minimal SV-49 promoter with 5 repeats of the canonical TEAD binding element (MCAT)15 driving the firefly luciferase gene (Figure 1). It is co-transfected into cells along with the PRL-TK control vector (Promega), which expresses Renilla luciferase from the constitutively active HSV TK promoter (Figure 1). It is critical to ensure that the YAP/TAZ-TEAD reporter construct is behaving as expected in the cell line(s) being tested. Therefore, we first co-transfected the PRL-TK and pGL3-5xMCAT (SV)-49 constructs into A375 cells stably expressing either a control vector, a highly active LATS-insensitive mutant of YAP (YAP2SA), or a mutant of YAP2SA unable to bind TEADs (YAP2SA,S94A). As expected, firefly luciferase activity was significantly increased by YAP2SA, but not the control vector or the YAP2SA,S94A (Figure 2A). This suggests that YAP increases the activity of the YAP/TAZ-TEAD reporter in a TEAD-dependent manner. Importantly, YAP2SA did not significantly alter the levels of the Renilla luciferase. As an additional control, we also confirmed that the YAP2SA does not alter the activity of a minimal promoter construct that lacks the MCAT TEAD binding elements (Figure 2B).
A second control experiment was also done in which A375 cells were infected with viral constructs that deliver either a non-targeting control shRNA (shNTC) or a construct with tandem YAP and TAZ shRNAs. After stable selection, the cells were co-transfected with the PRL-TK construct and either the YAP/TAZ-TEAD reporter construct or the minimal promoter construct. In cells transfected with the YAP/TAZ-TEAD reporter construct, the YAP/TAZ shRNA significantly reduced firefly luciferase levels, but the control shNTC did not (Figure 2C). Firefly luciferase levels were not changed by either the shNTC or the YAP/TAZ shRNA in cells transfected with the minimal promoter (Figure 2C). Consistent with the results above, the Renilla signal in each well was also not significantly altered by the control shNTC or the YAP/TAZ shRNA (Figure 2C), further indicating that the PRL-TK control construct is not responsive to YAP or TAZ. Collectively, these experiments show that the reporter system is behaving as expected in A375 cells, which is consistent with previous studies using these vectors2,14.
As a proof of concept experiment, we screened a small lentiviral shRNA library in A375 cells. Our test library contained shRNAs targeting several genes that were previously shown to regulate YAP/TAZ function in other cell types. However, the role of several of these genes in the regulation of YAP and TAZ in melanoma is unknown. Our library included genes that are required for YAP/TAZ activity, such as RAF1, MDM2, PIK3CA, MAPK8, PAK4, EZH2, PDK1, ERBB4, and CCNE216,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38, and genes predicted to inhibit YAP/TAZ activity, such as CSK, ERBB2, ATM, CDH1, Gelsolin (GSN), PTEN, ATR, and RB139,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54. As controls, we included a tandem YAP/TAZ shRNA14, a non-targeting control shRNA (shNTC), and an shRNA targeting Src, which we previously showed was required for maximal YAP/TAZ activity in A375 cells14.
The raw data and analysis for this screen is shown in Supplemental Table 2. Several shRNAs (shPDK-1, shMDM2-1, shMDM2-2, shPAK4, shMAPK8-1 and shMAPK8-2) significantly reduced Renilla luciferase signal relative to the mean Renilla signal for all wells, suggesting that these shRNAs were causing cytotoxicity in A375 cells. Indeed these wells had significantly fewer cells at the time of the assay (not shown). This makes it very difficult to distinguish candidates that may regulate YAP/TAZ from the ones that are essential for cell survival. Therefore, the results from these samples are excluded from further data analysis. As expected, shRNAs targeting YAP and TAZ or Src significantly reduced YAP/TAZ-TEAD activity (Figure 3). Consistent with published work showing that PIK3CA promotes YAP and TAZ activity21,26,31, we found that shRNAs targeting PIK3CA reduced normalized firefly luciferase levels. shRNAs targeting ATM, CDH1, CSK, ERBB2, GSN each increased normalized firefly luciferase levels, which is consistent with published studies showing that these proteins repress YAP and/or TAZ39,41,42,43,45,46,47,48,50,51,53,54. Despite established roles for ATR, CCNE2, and ERBB4 in other cell types27,28,35,36,38,49, shRNAs targeting these genes did not significantly change normalized firefly luciferase levels in A375 cells. shRNAs targeting EZH2 and RB1 showed an effect on firefly luciferase levels that was opposite of what was expected based on studies in other cell types32,34,40,52. Both PTEN and RAF1 were targeted by two distinct shRNAs that had opposite effects on luciferase activity. Results inconsistent with previous studies could indicate that the influence of these proteins on YAP/TAZ-TEAD function is cell type-dependent; however, it may also be due to poor knockdown efficiency or off-target effects by some shRNAs. As an n = 1 "blind" screen, some false positives and false negatives would also be expected. As discussed in more detail in the Discussion, additional validation experiments would need to be done using other assays such as qPCR for genes that are regulated by YAP and TAZ and western blots for post translational modifications that influence YAP/TAZ function. This validation would also include confirming which shRNAs effectively knocked down the protein of interest. It is also important to note that since our reporter is TEAD responsive, any genes identified by our screen could be influencing TEADs but not YAP and TAZ directly. Follow-up mechanistic studies would be required to determine whether it was YAP and/or TAZ or the TEADs that the identified protein is regulating. However, YAP and TAZ are the major drivers of TEAD-mediated transcription, so it is likely that most of these genes are regulating YAP and TAZ. Indeed, as stated above, most of the shRNAs that hit in the screen target proteins that are known to regulate YAP and/or TAZ directly.
Figure 1: Schematic diagram of the workflow. The critical steps of this protocol are summarized with step numbers corresponding to the numbers in the protocol. Please click here to view a larger version of this figure.
Figure 2: Optimization of the YAP/TAZ-TEAD dual-luciferase transcriptional reporter system. (A) A375 cells stably expressing control vector (MSCV-IRES-Hygro), LATS-insensitive YAP (YAP2SA), or LATS-insensitive YAP unable to bind TEADs (YAP2SA,S94A) were co-transfected with the PRL-TK (Renilla) and pGL3-5xMCAT (SV)-49 (YAP/TAZ-TEAD reporter) constructs and luciferase signal was measured 24 hours later. n = 7 independent experiments. (B) A375 cells were co-transfected with either YAP2SA or control vector, the PRL-TK construct, and either the pGL3-5xMCAT (SV)-49 construct or the pGL3 (SV)-49 construct (minimal promoter) and luciferase signal was measured 24 hours later. n = 1 experiment transfected in quadruplicate. (C) A375 cells were infected with retrovirus encoding a non-targeting control shRNA (shNTC) or tandem YAP and TAZ shRNAs (shYAP/TAZ). After stable selection, cells were co-transfected with the PRL-TK construct and either the pGL3-5xMCAT (SV)-49 construct or the pGL3 (SV)-49 construct and luciferase signal was measured 24 hours later. n = 4 independent infections; Statistical significance was determined using one-way ANOVA followed by Tukey's multiple comparisons test; n.s. = p>0.05, * p<0.05, ** p<0.01, **** p<0.0001. Please click here to view a larger version of this figure.
Figure 3: Results of representative screen. The result from each knock down vector is compared with shNTC and presented as fold difference. The bar graph shows the average of firefly luciferase/Renilla luciferase ratio ± standard deviation. n = 1 experiment. Please click here to view a larger version of this figure.
Tube A – Transfection Dilution | |
Reagent | Volume |
Transfection Buffer | 25 µl per reaction |
Transfection Reagent 2 (add second) | 1.5 µl per reaction |
Tube B – Reporter Dilution | |
Reagent | Volume |
Transfection Buffer | 25 µl per reaction |
20:1 Firefly luciferase reporter: control Renilla mix (add second) | 400 ng per reaction |
Transfection Reagent 3 (add third) | 1 µl per reaction |
Total: 26.5 µl per reaction |
Table 1: Preparation of transfection mix. For the transfection in each well of the 24-well plates, 1.5 µL of transfection reagent 2 is diluted into 25 µL of transfection buffer to produce the transfection dilution (tube A); while 400 ng of 20:1 firefly luciferase reporter: control Renilla mix and 1 µL of transfection reagent 3 is diluted into 25 µL of Transfection Buffer to produce the reporter dilution (tube B).
Existing/Purchased/Gift Vectors | |
Vector | Source |
GIPZ human shRNA lentiviral vectors | GE Healthcare |
VSVG | Hynes Lab [2] |
psPAX2 | Addgene (#12260) |
gag/pol | Addgene (#14887) |
pGL3-(SV)-49 | Iain Farrance [15] |
pGL3-5xMCAT(SV)-49 | Iain Farrance [15] |
PRL-TK | Promega |
MSCV-IRES-Hygro | LAMAR LAB [2] |
MSCV-YAP-S127A,S381A-IRES-Hygro | LAMAR LAB [2] |
MSCV-ZSGreen-2A-Puro-hYAP7/hTAZ3 | LAMAR LAB [14] |
New Vectors | |
Vector | Source Backbone |
MSCV-YAP-S94A,S127A,S381A-IRES-Hygro | MSCV-IRES-Hygro |
Supplemental Table 1: Table of vectors. All vectors used are listed with new vectors described. Standard molecular biology techniques were used to generate new vectors.
Supplemental Table 2: Luciferase assay data analysis. The arrangement of the shRNA library is shown in the first table. The second and third tables show the raw firefly and Renilla luciferase signals, respectively. The mean and standard deviation for the Renilla luciferase signal for all wells are indicated in the yellow boxes. Wells with a Renilla luciferase signal more than 1 standard deviation below the mean are highlighted with red text and excluded from further analysis. The firefly/Renilla ratio of every well was obtained by dividing the raw firefly luciferase signal by the raw Renilla luciferase signal (fourth table). The firefly/Renilla ratio of each well was then normalized to the average of firefly/Renilla ratio of the control (shNTC) wells (fifth table). The duplicate wells were next averaged for each construct and the standard deviation was calculated (sixth table). Please click here to view this table (Right click to download).
In this study, we demonstrate an approach for medium-throughput screening of arrayed viral libraries in combination with a dual-luciferase-based transcriptional reporter assay that can be used to identify and test novel regulators of transcription factors. It is critical to characterize and optimize the reporter system for each cell line prior to any screen. Experiments should be done to confirm that the reporter is responsive to altered activity of the transcription factor being investigated and the magnitude of change in activity should be tested relative to control vectors. Co-transfection of the PRL-TK construct along with the reporter construct is important because it helps control for the number of cells transfected with the reporter and the copy number of the reporter construct, both of which can alter the magnitude of the luciferase signal. Optimization experiments should also ensure that the transcription factor does not influence the activity of the constitutive Renilla construct or a minimal promoter construct as this could complicate the interpretation of the results. Controls to be included in the library should also be carefully considered. This protocol is designed for a "blind screen" of libraries with hundreds of shRNAs, where it is not be feasible to confirm effective knockdown by each shRNA. Therefore, some false positives and negatives are likely. Increasing the number of technical and/or biological replicates in the screen could help reduce the number of false positives and negatives. However, this will also significantly increase the number of wells to culture and assay so care should be taken to ensure this does not result in suboptimal culture conditions (see below). Another approach to reduce false positives and negatives would be to repeat the entire screen in the same cell line or additional cell lines, or to screen a smaller library including only the "hits" using 3 biological replicates. In all cases, any hits should be validated using readouts besides the reporter, such as qPCR for known target genes. Effective knockdown of the targeted gene should also be confirmed in these validation steps. Conclusions should not be made about shRNAs that do not alter activity unless effective knockdown by the shRNA is confirmed. Validation experiments should also ensure that the transcription factor being investigated is what is regulated by the targeted gene(s) and that these genes do not influence the Renilla or minimal promoter constructs.
It is also important to optimize the cell culture conditions for each cell line to avoid suboptimal conditions such as over-confluence, too few cells, poor cell viability, or variable proliferative capacity. Cell seeding density is particularly important at the time of the transfection of dual-luciferase reporter construct (Step 5) and when the reporter activity is measured (Step 6). Transfection efficiency can vary significantly with cell density and poor transfection efficiency can result in anomalous results if only a small fraction of the cell population is being assayed. Most cell lines tested show the highest transfection efficiency when at 40 – 60% confluence, but it is recommended that transfection conditions and seeding density are optimized for a given cell line using a vector that delivers a fluorescent protein. Cell lines that are difficult to infect and/or transfect, such as primary cells may not be suitable for this screen. Poor reporter transfection efficiency or poor cell viability typically result in very low Renilla luciferase signal. However, a pilot experiment should be performed to determine the range of Renilla luciferase signal for a cell line. It is important that the titer of the virus produced from the arrayed library is high enough to give an infection efficiency of at least 30% in the cell line being assayed. Infection efficiency can vary greatly and several factors can influence viral titer. The amount of viral supernatant used should be optimized for each cell line and if necessary, Step 2 can be optimized further to improve viral titers.
Cell density and cell viability can also influence the activity of cellular pathways that regulate transcription factors, so it is critical to seed the cells for the reporter transfection so that they are all at a similar density on the day the dual-luciferase assay is read (Step 6). It is also important to ensure that the cells adhere uniformly across the well rather than clustering in a dense patch in the center. As described above, significant variation in the Renilla luciferase signal from well to well may result in data that is difficult to interpret. It is best to exclude wells that show a significant reduction in Renilla luciferase signal when compared to all other wells as this suggests that either the viral vector is reducing cell viability or the transfection efficiency for that well was very low. Here we excluded wells greater than 1 standard deviation from the mean Renilla luciferase signal, but optimization experiments should be done to determine appropriate cutoffs for each cell line. It is also possible that shRNAs that reduce cell viability could be doing so by altering the activity of the transcription factor being assayed. If this is a concern, shRNAs that reduce Renilla luciferase signal could be re-tested using inducible shRNAs or other assays. It is also critical to limit the amount of time that adherent cells are in suspension following trypsinization. Use a multi-channel pipette for most steps during the trypsinization and seeding of cells, and for larger screens, work with the plates in batches. In addition, it is best to limit the time between the infection of the cells and the reporter assay. This is particularly important if the transcription factor being assayed regulates cell proliferation or survival. In this case, cells with effective knockdown would be outcompeted by cells with less efficient knockdown.
Several modifications of the described protocol are feasible. This approach can be used effectively to identify regulators of any transcription factor if a luciferase-based reporter is generated, and for many of the most common transcription factors reporter constructs already exist. This protocol can be modified for the use of a wide variety commercially available or manually-assembled libraries, including RNAi, CRISPR/CAS9, ORF, or cDNA. Indeed, we previously used a similar strategy to test a small cDNA library for YAP/TAZ regulators14. Libraries can be delivered using retrovirus, lentivirus, adenovirus, or transient transfection. The viral coat protein used here (VSVG) generates lentivirus that is human-infectious. If rodent cells are being used and non-human infectious ecotropic virus is desired, a vector delivering the Eco coat protein can be used instead of VSVG.
Other methods can be used to screen libraries for regulators of a transcription factor. Access to a high-throughput screening facility would allow for the screening of much larger libraries in an automated manner. Likewise, genome-wide screens can be done using pooled libraries with deep sequencing as a means to identify the "hits". However, both of these approaches require equipment that is not accessible to many researchers, and the fees for high-throughput screening or deep sequencing can be prohibitively high. In addition, pooled screens would require sorting of cells using a fluorescent reporter or another way to enrich the cells that show the desired changes in transcriptional activity. Screening of large arrayed expression libraries using a dual-luciferase-based transcriptional reporter assay has been demonstrated previously using a benchtop robot55, but this is not commonly available for all researchers. In contrast, the method described here is rapid, medium-throughput, relatively inexpensive, and uses equipment and reagents that are likely accessible to most investigators. This method can reveal multiple regulators in a single experiment, which is a cost-effective and convenient way to discover potential regulatory pathways upstream of a transcription factor.
In the described approach, candidate genes were stably knocked down and then after selection, the reporter constructs were transiently transfected into the cells. However, transfection efficiency is poor in many cell lines and can introduce variability and cause cellular toxicity. An improved alternative approach would be to stably integrate the reporter construct into the genome of the cell line of interest and then infect the reporter-expressing cells with the viral libraries for screening. This improvement would reduce the variability introduced by the transient transfection and prevent any potential changes in transcription factor activity due to prolonged post-infection culture. As mentioned above, the use of a small benchtop robot would greatly streamline the process and increase the size of the libraries that could be screened. Another considerable alteration is to use arrayed inducible vector libraries, which would reduce the likelihood of selection against cells with more effective knockdown and reduce variability caused by changes in cell viability.
A significant challenge that prevents the effective treatment of many cancers is that tumor cells acquire resistance to even the most effective targeted therapies. The identification of proteins that are required for this therapeutic resistance is necessary to improve patient outcome. The described approach could easily be used in combination with targeted therapeutics to reveal genes that cause increased sensitivity or resistance to these compounds. Another potential future application of this approach would be to use this arrayed screening to test candidate therapeutic targets in combination. For this, cells would be infected with two viral vectors each with the goal of identifying proteins that have a synergistic effect on transcription factor activity.
The authors have nothing to disclose.
We would like to thank Emily Norton and Mikaelan Cucciarre-Stuligross for assisting in the preparation of shRNA vectors. This work was supported in part by a Susan G. Komen Career Catalyst Grant that awarded to J.M.L. (#CCR17477184).
2.0 ml 96-well deep well polypropylene plate | USA Scientific | 1896-2000 | For bacterial mini-prep |
Trypsin – 2.50% | Gibco | 15090-046 | Component of trypsin-EDTA |
96 well flat bottom white assay plate | Corning | 3922 | For dual-luciferase assay |
Ampicillin – 100 mg/ml | Sigma-Aldrich | 45-10835242001-EA | For bacterial mini-prep |
Bacto-tryptone – powder | Sigma-Aldrich | 95039 | Component of LB broth |
Dual-luciferase reporter assay system, which include LAR II reagent (reagent A), Stop & Glo substrate (reagent B substrate) and Stop & Glo buffer (reagent B buffer) – Kit | Promega | E1960 | For dual-luciferase assay |
Dulbecco's phosphate buffered saline w/o calcium, magnesium and phenol red – 9.6 g/L | Himedia | TS1006 | For PBS |
EDTA – 0.5 M | VWR | 97061-406 | Component of trypsin-EDTA |
Ethanol – 100% | Pharmco-AAPER | 111000200 | For bacterial mini-prep |
Foetal Bovine Serum – 100% | VWR | 97068-085 | Component of complete growth media |
Hexadimethrine bromide (Polybrene) – 8 mg/ml | Sigma-Aldrich | 45-H9268 | For virus infection |
HyClone DMEM/High glucose – 4 mM L-Glutamine; 4500 mg/L glucose; sodium pyruvate | GE Healthcare life sciences | SH30243.01 | Component of complete growth media |
I3-P/i3 Multi-Mode Microplate/EA | Molecular devices | For dual-luciferase assay | |
L-Glutamine – 200 mM | Gibco | 25030-081 | Component of complete growth media |
Lipofectamine 3000 (Transfection Reagent 2) – 100% | Life technologies | L3000008 | For transfections |
Molecular Biology Water – 100% | VWR | 02-0201-0500 | For dilution of shRNA vector for virus packaging |
NaCl – powder | BDH | BDH9286 | Component of LB broth |
NanoDrop One Microvolume UV-Vis Spectrophotometer | Thermo scientific | For measuring vector DNA concentration | |
Opti-MEM (Transfection Buffer) – 100% | Gibco | 31985-062 | For transfections |
Penicillin Streptomycin – 10,000 Unit/ml (Penicillin); 10,000 µg/ml (Streptomycin) | Gibco | 15140-122 | Component of complete growth media |
PureLink Quick Plasmid Miniprep Kit – Kit | Thermo Fisher Scientific | K210010 | For bacterial mini-prep |
Puromycin – 2.5 mg/ml | Sigma-Aldrich | 45-P7255 | For antibiotic selection after infection |
TC20 automated cell counter | Bio-Rad | For cell counting | |
X-tremeGENE 9 DNA transfection reagent (Transfection Reagent 1) – 100% | Roche | 6365787001 | For virus packaging |
Yeast extract – powder | VWR | J850 | Component of LB broth |
P3000 (Transfection Reagent 3) – 100% | Life technologies | L3000008 | For transfections |