Identification of Transcription Factor Regulators using Medium-Throughput Screening of Arrayed Libraries and a Dual-Luciferase-Based Reporter

Cancer Research

Your institution must subscribe to JoVE's Cancer Research section to access this content.

Fill out the form below to receive a free trial or learn more about access:



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.

Cite this Article

Copy Citation | Download Citations | Reprints and Permissions

Xiao, Y., Lamar, J. M. Identification of Transcription Factor Regulators using Medium-Throughput Screening of Arrayed Libraries and a Dual-Luciferase-Based Reporter. J. Vis. Exp. (157), e60582, doi:10.3791/60582 (2020).


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.

Subscription Required. Please recommend JoVE to your librarian.


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.

  1. Add 1.3 mL of Luria Broth (LB) (1% bacto-trypton, 0.5% yeast extract, 1% NaCl, pH 7.5) containing 100 µg/mL of ampicillin to each well of a 96-well deep well plate. Inoculate each well with 2 µL of glycerol stock and grow at 37 °C overnight with constant agitation at 225 rpm.
  2. Transfer each bacterial culture into a 1.5 mL centrifuge tube and pellet the bacteria by centrifugation at 21,000 x g at 4 °C for 10 min.
  3. Purify each vector using a bacterial mini-prep kit by following the manufacturer's protocol.
  4. Determine the concentration of each vector using a spectrophotometer.
  5. Store the plasmids at -20 °C.
    NOTE: The protocol can be paused here.

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.

  1. Expand the 293FT cells using complete growth media (Dulbecco's modified Eagle medium (DMEM) containing 4 mM L-glutamine, 4,500 mg/L glucose and sodium pyruvate, supplemented with 10% fetal bovine serum (FBS), 100 units/mL of penicillin, 100 µg/mL streptomycin antibiotic and 2 mM L-glutamine).
  2. For each vector in the library from step 1.4, seed one 24-well with 1 x 105 293FT cells.
    NOTE: It is recommended that some extra wells of a control viral vector be packaged and used to test the titer of the virus prior to proceeding to step 3 (see below).
  3. Incubate the cells at 37 °C with 5% CO2 for 24 h.
    NOTE: A general protocol for packaging lentivirus that was described previously14 has been scaled down to 24 wells for this protocol. It uses psPAX2 for lentiviral packaging and VSVG as a coat protein. If ectropic virus is desired, a vector delivering Eco can be used instead of VSVG. It is highly recommended that this protocol is optimized to achieve a viral titer that gives between 30%-70% infection efficiency of the target cells (see Discussion). See Supplemental Table 1 for a list of all vectors used.
  4. Set up a transfection mixture for each viral vector from Step 1.4 as described below. Each transfection should contain 250 ng of the viral vector, 125 ng of psPAX2, 125 ng of VSVG, 1.25 µL of transfection reagent 1, and 23.75 µL of transfection buffer (see Table of Materials).
    1. Dilute each lentiviral vector to 50 ng/µL with nuclease-free water, and then transfer 5 µL (250 ng) into a well of a 96-well PCR plate.
    2. Make the transfection super mix by mixing 1.25 µL * X of transfection reagent 1 and 23.75 µL * X of pre-warmed transfection buffer where "X" is the total number of transfections plus several extra to account for volume loss during pipetting.
    3. Incubate the transfection super mix at room temperature for 5 min.
    4. Add 125 ng * X of psPAX2 and 125 ng * X of VSVG to the tube of transfection super mix from Step 2.4.3 and gently pipet up and down to mix. Rapidly proceed to Step 2.4.5.
    5. Immediately aliquot the mixture from Step 2.4.4 into each tube of a PCR strip, and then use multi-channel pipette to transfer 25 µL the mixture into each well containing viral vector from Step 2.4.1.
    6. Incubate at room temperature for 20 min.
    7. Transfer all 30 µLs from each 96-well from Step 2.4.6 into a well of the 24-well containing 293 FT cells from Step 2.3.
  5. Incubate the cells at 37 °C with 5% CO2 for 24 h and then replace the media in every well of the 24-well plate with 500 µL of fresh complete growth media. Incubate the cells at 37 °C with 5% CO2 for another 24 h.
  6. Using a multi-channel pipette, collect the viral supernatant from each well and aliquot 220 µL (enough for 1 infection in Step 3 plus some extra volume) into two 96-wells each. These are the arrayed viral supernatant plates.
  7. Store the arrayed viral supernatant plates at -80 °C.
    NOTE: The protocol can be paused here. It is also recommended to test some of the extra control virus that was packaged (see above) on the cells to be infected before proceeding to Step 3. This is to ensure that the titer is sufficient to achieve at least 30% infection efficiency.

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).

  1. Expand the cells to be infected in complete growth media.
  2. Seed 24-well plates with 1 x 105 cells in 0.5 ml of complete growth media per well. Seed one well for each viral vector to be tested (including controls) and include an extra well that will not be infected which will serve as a control for drug selection in Step 3.7.
  3. Incubate the cells at 37 °C with 5% CO2 for 24 h.
  4. Infect each well from Step 3.2 with a different viral supernatant from the frozen arrayed lentiviral supernatant as follows.
    1. Prepare complete growth media that contains 20 µg/mL polybrene.
    2. Thaw the arrayed lentiviral library supernatants from Step 2.7 to room temperature.
    3. Aspirate the growth media from the 24-well plates from Step 3.2 and immediately add 200 µL of polybrene-containing growth media to each well.
    4. Using a multi-channel pipette, transfer 200 µL of viral supernatant from each 96-well from Step 3.4.2 to the 24 wells from Step 3.4.3.
  5. Incubate the cells at 37 °C with 5% CO2 for 24 - 48 h.
    NOTE: Some cell lines may require longer than 24 h to express the shRNAs and become puromycin resistant. The viral vector used here delivers a Turbo-GFP-IRES-puroR with a miR30-based shRNA in the 3'UTR of puroR (Figure 1). Infection efficiency and expression of the puromycin resistance gene and the shRNA were monitored by green fluorescent protein.
  6. Prepare complete growth media that contains 2.5 µg/mL puromycin.
    NOTE: The selection concentration of puromycin varies between cell lines. It is recommended that an antibiotic kill curve be performed for each cell line to be assayed prior to the screen.
  7. Aspirate the media from each well and replace with 500 µL of puromycin-containing complete growth media.
    NOTE: Be sure to also add puromycin to a control non-infected well that can be used in subsequent steps to ensure the puromycin selection is complete.
  8. Incubate the cells at 37 °C with 5% CO2 for 48 h.
    NOTE: It is best to select for 48 h, so plating density and viral titer should be optimized so that the cells are not overconfluent prior to 48 h.
  9. Ensure that the infected cells are green under the fluorescent microscope, and that cells on a control non-infected well treated with puromycin are all dead before proceeding to Step 4.

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.

  1. Trypsinize each well from Step 3.9 and transfer roughly 1 x 105 cells into wells at the corresponding position on the new 24-well plate as follows.
    NOTE: This protocol is designed for the screening of libraries with hundreds of shRNAs so it is not feasible to count every well of infected cells. Therefore, the steps below were used to estimate cell numbers in each well to help ensure roughly equal plating density.
    1. Group the wells from Step 3.9 into 3 - 4 groups such that all wells in a group have a similar cell density.
    2. Trypsinize 1 representative well from each group with 200 µL of trypsin-EDTA (1x PBS supplemented with 0.5 mM EDTA and 0.1% trypsin) for 5 min at 37 °C. Then neutralize the trypsin-EDTA by adding 400 µL of puromycin containing complete growth media.
    3. Count each representative well to determine the total cell number and dilute the cell suspension from each representative well to 2 x 105 cells/mL of using complete growth media.
    4. Seed 0.5 mL (1 x 105 cells) of each well from Step 4.1.3 into the corresponding position on a new 24-well plate and incubate this new 24-well plate at 37 °C with 5% CO2 for 24 h.
    5. For each group from Step 4.1.1, use the total cell number determined in Step 4.1.3 to calculate the volume of trypsin-EDTA to add to each well so that the resulting cell suspension will be 1 x 106 cells/mL.
    6. Add the appropriate volume of trypsin-EDTA to each well and incubate for 5 min at 37 °C.
      NOTE: For larger screen it is recommended that the 24-well plates be trypsinized and replated in groups rather than all at once to ensure the viability of the cells.
    7. During the above incubation, use a multi-channel pipette to add 400 µL of puromycin-containing complete growth media to the corresponding wells on a new 24-well plate.
    8. Transfer 100 µL of cell suspension (approximately 1 x 105 cells) from each well to the corresponding position on the new 24-well plates prepared in Step 4.1.7.
    9. Repeat Steps 4.1.6 through 4.1.8 for all plates of grouped wells from Step 4.1.1, one plate at a time.
  2. Incubate the cells at 37 °C with 5% CO2 for 24 h.

5. Transfection of dual-luciferase reporter

  1. Transfect each well from Step 4.2 with the dual-luciferase reporter constructs as follows.
    NOTE: The total amount of DNA and the optimal ratio of firefly luciferase reporter vector to control Renilla luciferase vector should be determined prior to starting this assay. Here, 400 ng of a DNA mixture that contains 20 parts firefly luciferase reporter and 1 part control Renilla luciferase was used.
    1. Make the transfection dilution mixture (Tube A) and the reporter dilution mixture (Tube B) by mixing the indicated volumes of each reagent (Table 1) multiplied by the total number of transfections (plus several extra).
      NOTE: This protocol is optimized for transfection reagent 2 (see Table of Materials). If a different transfection reagent is used, the transfection should be optimized prior to this step.
    2. Mix the transfection dilution mixture (Tube A) with reporter dilution mixture (Tube B) and incubate at room temperature for 15 min to produce transfection mixture.
    3. During the above incubation, rinse each 24-well from Step 4.2 with 0.25 mL of phosphate buffer saline (PBS), and add 447 µL of complete growth media to each well.
    4. After the 15 min incubation, use a multi-channel pipette to distribute 53 µL of transfection mix to each well of the 24-well plates.
  2. Incubate the cells at 37 °C with 5% CO2 for 24 h.

6. Quantification of dual-luciferase activity

  1. Measure luciferase activity using a plate reader and a dual-luciferase reporter assay kit as described below.
    NOTE: This protocol is optimized for the indicated reporter assay kit (see Table of Materials) and follows the manufacturer's recommended protocol.
    1. Prepare enough 1x passive lysis buffer for all wells plus several extra (75 µL is needed per well) by diluting 5x passive lysis buffer (provided in kit) 1 to 5 with deionized water. Also thaw reagent A and reagent B Buffer (provided in kit, 100 µL of each is needed for each well).
    2. Aspirate the media from each well of the 24-well plate from Step 5.2.
    3. Add 75 µL of 1x passive lysis buffer to each well and incubate at room temperature for 30 min with occasionally shaking.
    4. Prepare reagent B by diluting 50x reagent B substrate (provided in kit) 1:50 with thawed reagent B buffer.
    5. Add 30 µL of 1x passive lysis buffer to 4 wells for blanking (see Supplemental Table 2).
    6. Transfer 30 µL of lysate from Step 6.1.3 into duplicate wells of a 96-well flat bottom white assay plate.
    7. Use a multi-channel pipette to add 50 µL of reagent A to each well and read the firefly luciferase signal with a plate reader.
    8. Use a multi-channel pipette to add 50 µL of reagent B from Step 6.1.4 to each well and read the Renilla luciferase signal with a plate reader.
  2. Process the raw data as follows (for a detailed description, see Supplemental Table 2).
    1. Exclude samples with very low Renilla luciferase signal as low values indicate that the viral construct was toxic or that too few transfected cells were assayed.
      NOTE: As explained in the Discussion, significantly "low" Renilla luciferase signal can result in anomalous results. Here, wells in which the Renilla luciferase signal was more than 1 standard deviation below the mean were excluded (see Supplemental Table 2). This was based on previous studies performed using this reporter system in these cells14, but may differ in other cell lines.
    2. Normalize the raw firefly luciferase value of each well to the raw Renilla luciferase value of the same well to obtain the firefly/Renilla ratio.
    3. Average the firefly/Renilla ratios of all replicate control wells and then divide the firefly/Renilla ratio of every other well by that number to get a fold change.
    4. Assign the control sample and set its firefly/Renilla ratio value to 1.
    5. Average the firefly/Renilla ratios of the duplicate wells and plot with the standard deviation.
      NOTE: The standard deviation is not used for statistical analysis, but instead as a means to identify wells where the replicates differ significantly.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

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
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
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
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-ZSGreen-2A-Puro-hYAP7/hTAZ3 LAMAR LAB [14]
New Vectors
Vector Source Backbone

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).

Subscription Required. Please recommend JoVE to your librarian.


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.

Subscription Required. Please recommend JoVE to your librarian.


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).


Name Company Catalog Number Comments
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



  1. Chen, K. S., Lim, J. W. C., Richards, L. J., Bunt, J. The convergent roles of the nuclear factor I transcription factors in development and cancer. Cancer Letters. 410, 124-138 (2017).
  2. Lamar, J. M., et al. The Hippo pathway target, YAP, promotes metastasis through its TEAD-interaction domain. Proceedings of the National Academy of Sciences of the United States of America. 109, (37), E2441-E2450 (2012).
  3. Liu, C. Y., Yu, T., Huang, Y., Cui, L., Hong, W. ETS (E26 transformation-specific) up-regulation of the transcriptional co-activator TAZ promotes cell migration and metastasis in prostate cancer. Journal of Biological Chemistry. 292, (22), 9420-9430 (2017).
  4. Semenza, G. L. Hypoxia-inducible factor 1: oxygen homeostasis and disease pathophysiology. Trends in Molecular Medicine. 7, (8), 345-350 (2001).
  5. Willmer, T., Cooper, A., Peres, J., Omar, R., Prince, S. The T-Box transcription factor 3 in development and cancer. Bioscience Trends. 11, (3), 254-266 (2017).
  6. Zhu, C., Li, L., Zhao, B. The regulation and function of YAP transcription co-activator. Acta Biochim Biophys Sin (Shanghai). 47, (1), 16-28 (2015).
  7. Dang, C. V., Reddy, E. P., Shokat, K. M., Soucek, L. Drugging the 'undruggable' cancer targets. Nature Reviews: Cancer. 17, (8), 502-508 (2017).
  8. Fu, V., Plouffe, S. W., Guan, K. L. The Hippo pathway in organ development, homeostasis, and regeneration. Current Opinion in Cell Biology. 49, 99-107 (2017).
  9. Hansen, C. G., Moroishi, T., Guan, K. L. YAP and TAZ: a nexus for Hippo signaling and beyond. Trends in Cell Biology. 25, (9), 499-513 (2015).
  10. Yu, F. X., Zhao, B., Guan, K. L. Hippo Pathway in Organ Size Control, Tissue Homeostasis, and Cancer. Cell. 163, (4), 811-828 (2015).
  11. Warren, J. S. A., Xiao, Y., Lamar, J. M. YAP/TAZ Activation as a Target for Treating Metastatic Cancer. Cancers. 10, (4), (2018).
  12. Janse van Rensburg, H. J., Yang, X. The roles of the Hippo pathway in cancer metastasis. Cellular Signalling. 28, (11), 1761-1772 (2016).
  13. Zanconato, F., Cordenonsi, M., Piccolo, S. YAP/TAZ at the Roots of Cancer. Cancer Cell. 29, (6), 783-803 (2016).
  14. Lamar, J. M., et al. SRC tyrosine kinase activates the YAP/TAZ axis and thereby drives tumor growth and metastasis. Journal of Biological Chemistry. 294, (7), 2302-2317 (2019).
  15. Mahoney, W. M., Hong, J. H., Yaffe, M. B., Farrance, I. K. The transcriptional co-activator TAZ interacts differentially with transcriptional enhancer factor-1 (TEF-1) family members. Biochemical Journal. 388, (Pt 1), 217-225 (2005).
  16. Codelia, V. A., Sun, G., Irvine, K. D. Regulation of YAP by mechanical strain through Jnk and Hippo signaling. Current Biology. 24, (17), 2012-2017 (2014).
  17. Cosset, E., et al. Glut3 Addiction Is a Druggable Vulnerability for a Molecularly Defined Subpopulation of Glioblastoma. Cancer Cell. 32, (6), 856-868 (2017).
  18. de Cristofaro, T., et al. TAZ/WWTR1 is overexpressed in papillary thyroid carcinoma. European Journal of Cancer. 47, (6), 926-933 (2011).
  19. Densham, R. M., et al. MST kinases monitor actin cytoskeletal integrity and signal via c-Jun N-terminal kinase stress-activated kinase to regulate p21Waf1/Cip1 stability. Molecular and Cellular Biology. 29, (24), 6380-6390 (2009).
  20. Eda, H., Aoki, K., Marumo, K., Fujii, K., Ohkawa, K. FGF-2 signaling induces downregulation of TAZ protein in osteoblastic MC3T3-E1 cells. Biochemical and Biophysical Research Communications. 366, (2), 471-475 (2008).
  21. Elbediwy, A., et al. Integrin signalling regulates YAP and TAZ to control skin homeostasis. Development. 143, (10), 1674-1687 (2016).
  22. Enomoto, M., Igaki, T. Src controls tumorigenesis via JNK-dependent regulation of the Hippo pathway in Drosophila. EMBO Reports. 14, (1), 65-72 (2013).
  23. Enomoto, M., Kizawa, D., Ohsawa, S., Igaki, T. JNK signaling is converted from anti- to pro-tumor pathway by Ras-mediated switch of Warts activity. Developmental Biology. 403, (2), 162-171 (2015).
  24. Fan, R., Kim, N. G., Gumbiner, B. M. Regulation of Hippo pathway by mitogenic growth factors via phosphoinositide 3-kinase and phosphoinositide-dependent kinase-1. Proceedings of the National Academy of Sciences of the United States of America. 110, (7), 2569-2574 (2013).
  25. Feng, R., et al. MAPK and Hippo signaling pathways crosstalk via the RAF-1/MST-2 interaction in malignant melanoma. Oncology Reports. 38, (2), 1199-1205 (2017).
  26. Fisher, M. L., et al. Transglutaminase Interaction with alpha6/beta4-Integrin Stimulates YAP1-Dependent DeltaNp63alpha Stabilization and Leads to Enhanced Cancer Stem Cell Survival and Tumor Formation. Cancer Research. 76, (24), 7265-7276 (2016).
  27. Haskins, J. W., Nguyen, D. X., Stern, D. F. Neuregulin 1-activated ERBB4 interacts with YAP to induce Hippo pathway target genes and promote cell migration. Science Signaling. 7, (355), (2014).
  28. Hoeing, K., et al. Presenilin-1 processing of ErbB4 in fetal type II cells is necessary for control of fetal lung maturation. Biochimica et Biophysica Acta. 1813, (3), 480-491 (2011).
  29. Hwang, J. H., et al. Extracellular Matrix Stiffness Regulates Osteogenic Differentiation through MAPK Activation. PloS One. 10, (8), e0135519 (2015).
  30. Kaneko, K., Ito, M., Naoe, Y., Lacy-Hulbert, A., Ikeda, K. Integrin alphav in the mechanical response of osteoblast lineage cells. Biochemical and Biophysical Research Communications. 447, (2), 352-357 (2014).
  31. Kim, N. G., Gumbiner, B. M. Adhesion to fibronectin regulates Hippo signaling via the FAK-Src-PI3K pathway. Journal of Cell Biology. 210, (3), 503-515 (2015).
  32. Kuser-Abali, G., Alptekin, A., Cinar, B. Overexpression of MYC and EZH2 cooperates to epigenetically silence MST1 expression. Epigenetics. 9, (4), 634-643 (2014).
  33. Liu, N., et al. HDM2 Promotes NEDDylation of Hepatitis B Virus HBx To Enhance Its Stability and Function. Journal of Virology. 91, (16), (2017).
  34. Liu, X., et al. The EZH2- H3K27me3-DNMT1 complex orchestrates epigenetic silencing of the wwc1 gene, a Hippo/YAP pathway upstream effector, in breast cancer epithelial cells. Cellular Signalling. 51, 243-256 (2018).
  35. Omerovic, J., et al. Ligand-regulated association of ErbB-4 to the transcriptional co-activator YAP65 controls transcription at the nuclear level. Experimental Cell Research. 294, (2), 469-479 (2004).
  36. Pegoraro, S., et al. A novel HMGA1-CCNE2-YAP axis regulates breast cancer aggressiveness. Oncotarget. 6, (22), 19087-19101 (2015).
  37. Xia, H., et al. EGFR-PI3K-PDK1 pathway regulates YAP signaling in hepatocellular carcinoma: the mechanism and its implications in targeted therapy. Cell Death & Disease. 9, (3), 269 (2018).
  38. Yan, F., et al. ErbB4 protects against neuronal apoptosis via activation of YAP/PIK3CB signaling pathway in a rat model of subarachnoid hemorrhage. Experimental Neurology. 297, 92-100 (2017).
  39. Aragona, M., et al. A mechanical checkpoint controls multicellular growth through YAP/TAZ regulation by actin-processing factors. Cell. 154, (5), 1047-1059 (2013).
  40. Bonilla, X., et al. Genomic analysis identifies new drivers and progression pathways in skin basal cell carcinoma. Nature Genetics. 48, (4), 398-406 (2016).
  41. Enger, T. B., et al. The Hippo signaling pathway is required for salivary gland development and its dysregulation is associated with Sjogren's syndrome. Laboratory Investigation. 93, (11), 1203-1218 (2013).
  42. Fausti, F., et al. ATM kinase enables the functional axis of YAP, PML and p53 to ameliorate loss of Werner protein-mediated oncogenic senescence. Cell Death and Differentiation. 20, (11), 1498-1509 (2013).
  43. He, J., et al. Positive regulation of TAZ expression by EBV-LMP1 contributes to cell proliferation and epithelial-mesenchymal transition in nasopharyngeal carcinoma. Oncotarget. 8, (32), 52333-52344 (2017).
  44. Huang, W., et al. The N-terminal phosphodegron targets TAZ/WWTR1 protein for SCFbeta-TrCP-dependent degradation in response to phosphatidylinositol 3-kinase inhibition. Journal of Biological Chemistry. 287, (31), 26245-26253 (2012).
  45. Imada, S., et al. Role of Src Family Kinases in Regulation of Intestinal Epithelial Homeostasis. Molecular and Cellular Biology. 36, (22), 2811-2823 (2016).
  46. Kim, N. G., Koh, E., Chen, X., Gumbiner, B. M. E-cadherin mediates contact inhibition of proliferation through Hippo signaling-pathway components. Proceedings of the National Academy of Sciences of the United States of America. 108, (29), 11930-11935 (2011).
  47. Lai, J. K. H., et al. The Hippo pathway effector Wwtr1 regulates cardiac wall maturation in zebrafish. Development. 145, (10), (2018).
  48. Li, H., Gumbiner, B. M. Deregulation of the Hippo pathway in mouse mammary stem cells promotes mammary tumorigenesis. Mammalian Genome. 27, (11-12), 556-564 (2016).
  49. Pefani, D. E., O'Neill, E. Hippo pathway and protection of genome stability in response to DNA damage. The FEBS Journal. 283, (8), 1392-1403 (2016).
  50. Serrano, I., McDonald, P. C., Lock, F., Muller, W. J., Dedhar, S. Inactivation of the Hippo tumour suppressor pathway by integrin-linked kinase. Nature Communications. 4, 2976 (2013).
  51. Vlug, E. J., et al. Nuclear localization of the transcriptional coactivator YAP is associated with invasive lobular breast cancer. Cellular Oncology (Dordrecht). 36, (5), 375-384 (2013).
  52. Xie, Q., et al. YAP/TEAD-mediated transcription controls cellular senescence. Cancer Research. 73, (12), 3615-3624 (2013).
  53. Yee, K. S., et al. A RASSF1A polymorphism restricts p53/p73 activation and associates with poor survival and accelerated age of onset of soft tissue sarcoma. Cancer Research. 72, (9), 2206-2217 (2012).
  54. Zhou, Z., et al. Oncogenic Kinase-Induced PKM2 Tyrosine 105 Phosphorylation Converts Nononcogenic PKM2 to a Tumor Promoter and Induces Cancer Stem-like Cells. Cancer Research. 78, (9), 2248-2261 (2018).
  55. Baker, J. M., Boyce, F. M. High-throughput functional screening using a homemade dual-glow luciferase assay. Journal of Visualized Experiments. (88), (2014).



    Post a Question / Comment / Request

    You must be signed in to post a comment. Please or create an account.

    Usage Statistics