Describing a Transcription Factor Dependent Regulation of the MicroRNA Transcriptome


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

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



Herein we propose a strategy to study the effect of a transcription factor of interest on the microRNA transcriptome using publically available data, computational resources and high throughput data from microRNA arrays after transfecting cells with small hairpin (sh)RNA targeting a transcription factor of interest.

Cite this Article

Copy Citation | Download Citations

Rozovski, U., Hazan-Halevy, I., Calin, G., Harris, D., Li, P., Liu, Z., Keating, M. J., Estrov, Z. Describing a Transcription Factor Dependent Regulation of the MicroRNA Transcriptome. J. Vis. Exp. (112), e53300, doi:10.3791/53300 (2016).


While the transcription regulation of protein coding genes was extensively studied, little is known on how transcription factors are involved in transcription of non-coding RNAs, specifically of microRNAs. Here, we propose a strategy to study the potential role of transcription factor in regulating transcription of microRNAs using publically available data, computational resources and high throughput data. We use the H3K4me3 epigenetic signature to identify microRNA promoters and chromatin immunoprecipitation (ChIP)-sequencing data from the ENCODE project to identify microRNA promoters that are enriched with transcription factor binding sites. By transfecting cells of interest with shRNA targeting a transcription factor of interest and subjecting the cells to microRNA array, we study the effect of this transcription factor on the microRNA transcriptome. As an illustrative example we use our study on the effect of STAT3 on the microRNA transcriptome of chronic lymphocytic leukemia (CLL) cells.


MicroRNAs are endogenous small non coding regulatory RNAs that typically function as negative regulators of mRNA expression at the posttranscriptional level. Approximately 1,000 non-coding 20 to 25 nucleotide long microRNAs are found in the human genome 1,2. MicroRNAs regulate gene expression through canonical base pairing between the seed sequence of the microRNA and its complementary seed match sequence, which is commonly located at the 3' untranslated region (UTR) of the target mRNAs. Collectively, microRNAs regulate more than 30% of protein coding genes 3, but only little is known about the transcription from DNA of microRNAs. It has been suggested that the regulation of microRNA transcription is similar to that of mRNA 4,5. In particular, similar to its activity in promoting transcription of protein coding genes, transcription factors are thought to activate transcription of microRNAs 6. Transcription factor-microRNA interplay has been reported as a modulator factor of gene expression 7, and may also form feed-back and feed-forward loops. For example, Yamakuchi et al. reported a feedback loop in which p53 induces the expression of microRNA34a, which in turn inhibits translation of the p53 repressor SIRT and thereby increasing p53 activity 8.

Whereas specific examples of transcription factor dependent expression of microRNAs have been reported, an accepted method which provides information on how a transcription factor of interest regulates the expression of the microRNA-transcriptome is lacking. The purpose of the protocol suggested herein is to provide an in-depth description of transcription factor-dependent regulation of the microRNA-transcriptome. By combining publically available data, bioinformatics tools and using microarray technology, researchers who follow this algorithm would be able to capture on a genomic scale how any transcription factor in any cell type of interest regulates the expression of the microRNA-transcriptome and to explore a putative contribution of the transcription factor-mRNA in regulating microRNA expression.

Subscription Required. Please recommend JoVE to your librarian.


1. Identify Transcription Factor Binding Sites in the Promoter of MicroRNA Genes Using Data Mining Approach

  1. Use the University of California Santa Cruz (UCSC) genome browser to extract chromatin immunoprecipitation (ChIP) sequencing data generated as part of the Encyclopedia of DNA element (ENCODE) project.
    1. Open the table browser in the UCSC genome browser.
    2. Use the following specifications to extract the table: Clade: (Mammals), genome: (Human), Assembly: (Feb2009(GRch37/hg18)), group: (regulation), track (TxnFactorChIP), table: (weEncoderesTFbsCo7steredv3), region: (Genome), output format (all genes from selected table).
    3. Save the output from 1.1.2 as a .txt file and into a spreadsheet.
    4. Sort and filter for the transcription factor of interest (e.g., STAT2).
  2. Use the list of microRNA promoters based on H3K4me3 epigenetics signature available at Baeer et al.9. Copy this list into a .txt file.
  3. To match according to coordinates on the human genome, map the data from 1.1 and 1.2 (e.g., STAT3 binding on putative microRNA promoters) and determine the median binding affinity using the code written in C sharp as outlined in supplemental coding file 1.

2. Use shRNA to Down-regulate the Expression of a Transcription Factor of Interest

  1. Plate 1.5 x 106 cells from 293 human embryonic kidney cell line in 10 cm plate at approximately 50% confluence (DMEM with 10% FBS).
  2. Transfect cells from 293 human embryonic kidney cell line with 5 µg of green fluorescence protein (GFP) lentivirus containing shRNA directed to the transcription factor of interest and with 5 µg of packaging vectors using transfection reagent for adherent cells according to manufacturer's protocol.
  3. As a control, transfect the cells from 293 human embryonic kidney cell line with scrambled shRNA and the packaging vectors according to manufacturer's protocol.
  4. Keep transfection mix on cells for 16 hr (at 37 °C, CO2 incubator), then change media to 10 ml fresh 10% DMEM media with 10% FBS.
  5. Wait 48 hr post transfection, centrifuge the cell culture (300 x g, 5 min) and collect infectious supernatant. Filter the supernatant through a 0.45 µm syringe filter (25-mm surfactant free cellulose acetate membrane) to remove any floating cells.
  6. Concentrate the supernatant and collect the lentivirus using an ultracentrifugal filter device with threshold of 100 kDa. Spin at 950 x g for 30 - 60 min until the volume has been concentrate to less than 250 µl. Store the concentrated virus at -80 °C.
  7. Transfect the cells with the lentivirus. Remove frozen lentivirus from -80 °C freezer and thaw to room temperature. Transfer 100 µl of viral supernatant to a fresh 1.5 ml microfuge tube.
    1. Bring up the volume in the tube to 1 ml with reduced serum medium. Add hexadimethrine bromide to 1 ml virus suspension for final concentration of 10 ng/ml, mix gently and let the mixture stand for 5 min.
  8. Centrifuge 5 x 106 cells for each transduction and gently resuspend the cell pellet in 0.5 ml of media containing virus. Let the cells stay in the incubator for 4 - 24 hr, then add 0.5 ml of medium with 20% FBS to a final concentration of 10% FBS.
  9. Wait 48 to 72 hr and stain the cells with propidium iodide (PI) and green florescent protein (GFP) according to the manufacturer's instructions. Protect the cells from light and use a FACS sorter to measure the rates of GFP+ / PI- cells. Since PI stains only dead cell, this rate is an estimate of transfection efficiency in living cells.
  10. Sort the positive cell population to GFP expression (GFP+) by FACS sorter as previously described 10.
  11. Use Western immune blotting as previously described 10 to determine the levels of a transcription factor of interest before and after infecting the cells of interest (for example, CLL cells) with designated shRNA.

3. Determine the Expression Level of MicroRNA Transcriptome in Cells Transfected with Transcription Factor-shRNA

  1. Isolate RNA using a commercial kit according to manufacturer's protocol.
  2. Label the RNA and hybridize it to microRNA microarray11.
  3. Determine the differentially expressed microRNA in cells transfected with transcription factor-shRNA or with empty vector controls5.
  4. Validate the microarray results for the most differentially expressed microRNAs using real-time PCR 5.

4. Determine the Overlap between the Bioinformatics and the shRNA Approach in Describing the Transcription Factor Dependent Transcriptome

  1. To determine the expected and observed ratios of microRNA genes that harbor transcription factor binding sites in their promoters and were downregulated in transcription factor-shRNA transfected cells, do the following:
    1. Obtain the ratio of genes that harbor the transcription factor of interest in their promoter / total microRNA from the list generated in 1.3. This list is the expected ratio (e.g., microRNA genes with STAT3 binding sites / total microRNA genes tested = 0.25).
    2. Determine the number of genes that were downregulated in transcription factor-shRNA transfected cells from the list generated in 3.3 AND has transcription factor of interest binding sites from the list generated in 1.3. This number/Total number of downregulated genes is the observed ratio (e.g., microRNA genes with STAT3 binding sites/total number of downregulated microRNAs = 0.6).
  2. Use χ2 statistics to compare the observed and expected ratios that were generated above and determine whether the list of genes that were downregulated in transcription factor-shRNA transfected cells are enriched with transcription factor binding sites in their promoter.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

STAT3 is a transcription factor which typically induces the transcription of genes that have anti apoptotic and proliferative effects 12 . Whether STAT3 also affect the non-coding RNA transcriptome is currently unknown. In all CLL cells STAT3 is constitutively phosphorylated on serine 707 residues 10,13. Phosphoserine STAT3 shuttles to the nucleus, binds to DNA, and activates genes known to be activated by tyrosine pSTAT3 in other cell types 10. Because CLL is characterized by global deregulation of the microRNA network 14, we hypothesized that the presence of serine pSTAT3 affects the expression of microRNAs in CLL cells.

To test this hypothesis promoters of microRNAs that harbor STAT3 binding sites had to be identified. By crossing data generated by Baer et al. 9 of regions with the H2K4me3 histone modification which characterize promoter sites, with STAT3 binding sites identified by ChIP-seq experiment 15, putative STAT3 binding sites were identified. Using this approach 160 putative promoters were detected in nearly 25% of the microRNA genes examined (N=200) with binding scores ranging from 100 (the lowest score) to 1,000 (the highest score) (Table 1).

Subsequently CLL cells were transfected with STAT3-shRNA or with an empty vector and using microRNA array and identified 63 microRNAs that were down regulated following the transfection (Figure 1) suggesting that STAT3 promotes the transcription of these microRNAs. For 60% of the 63 downregulated microRNA genes (n = 38) ChIP-seq data confirmed STAT3 binding in a putative promoter upstream of the gene location, significantly more than expected by chance (p<0.0001). Nine microRNAs that were down-regulated after the transfection, suggesting that STAT3 negatively regulate its levels of transcription.

Figure 1
Figure 1. Transfection of CLL cells with STAT3-shRNA reduces the expression levels of STAT3 protein and STAT3 mRNA. A: After transfection of CLL cells with STAT3-shRNA levels of STAT3 mRNA (left panel, detected by Quantitative RT-PCR) and levels of the STAT3 protein (right panel, detected by western immunoblotting) significantly decreased. B: microRNA array of CLL cells depicted 23 microRNAs whose expression differed significantly between CLL cells transfected with STAT3-shRNA and CLL cells transfected with an empty vector. P value of less than 0.01 was considered statistically significant. C: The mean expression quantified by RT-PCR of 7 microRNAs who had differential expression after STAT3-shRNA treatment using the microRNA array. Bars represent the standard error of the mean. Please click here to view a larger version of this figure.

Supplemental Coding File 1. Please click here to download this file.

Micro RNA gene Chromosome Promoter start coordinates Promoter end coordinates Median (range) STAT3 binding score*
miR-1205, miR-1206,miR-1207 8 q24.21 128961454 128962791 1000 (1000-1000)
miR-1537 1 q42.3 236045425 236047415 1000 (1000-1000)
miR-21 17 q23.1 57901872 57921277 1000 (112-1000)
miR-3124 1 q44 249115404 249123965 1000 (1000-1000)
miR-451 17q11.2 27222251 27224114 1000 (1000-1000)
miR-92b 1 q22 155162340 155168439 1000 (1000-1000)
miR-3197 21 q22.2 42537544 42543023 943 (943-943)
miR-646 20 q13.33 58712550 58715320 789 (789-789)
miR-629 15 q23 70383751 70394586 773 (661-885)
miR-30e, miR-30c-1 1 p34.2 41173077 41177703 759 (759-759)
miR-3125 2 p24.3 12855381 12862915 756 (756-756)
miR-3145 6 q23.3 138776942 138779365 743 (487-1000)
miR-645 20 q13.13 49199911 49201187 743 (743-743)
miR-1256 1 p36.12 21346830 21350211 725 (725-725)
miR-619 12 q24.11 109248263 109253306 719 (719-719)
miR-181a-2, miR-181b-2 9 q33.3 127418928 127426139 710 (710-710)
miR-29a, miR-29b-1 7 q32.3 130583383 130597803 697 (482-1000)
miR-202 10 q26.3 135069499 135077337 696 (393-1000)
miR-3142, miR-146a 5 q34 159890882 159899475 671 (671-671)
miR-548c 12 q14.2 65000968 65011503 660 (660-660)
miR-630 15 q24.1 72764289 72769197 627 (255-1000)
miR-135b 1 q32.1 205416952 205452990 622 (245-1000)
miR-29c, miR-29b-2 1 q32.2 207991044 208002382 608 (608-608)
miR-1825 20 q11.21 30791020 30798310 604 (209-1000)
miR-548h-1 14 q23.2 64578834 64581657 587 (174-1000)
miR-612 11 q13.1 65183633 65198528 581 (157-1000)
miR-148b 12 q13.13 54717640 54721204 578 (578-578)
miR-3174 15 q26.1 90543381 90549092 576 (152-1000)
let7a-3, let7b 22 46480680 46481826 573 (146-1000)
miR-1255a 4 q24 102263848 102272541 557 (557-557)

Table 1. Putative microRNA's promoters with STAT3 binding sites. The protocol provides a method to identify transcription factor binding on putative promoters of microRNAs using data mining of published data. As an example, we present here a table depicting the binding of STAT3 to putative microRNA promoters. The binding score is given at a 0 to 1,000 scale from whole genome ChIP seq data published as part of the ENCODE project 15. The promoters are identified by the presence of the H3K4me3 epigenetic signature 9.

Subscription Required. Please recommend JoVE to your librarian.


The mechanism underlying the RNA polymerase II- dependent transcription of protein coding genes has been extensively studied. While these elements make up only 1% - 2% of the human genome, evidence from the ENCODE project suggest that over 80% of the human genome may undergo transcription 17 and what regulates the transcription of the non-coding DNA elements remains largely unknown 6.

Several studies, which indicated that Pol II is also responsible for the transcription of some non-protein-coding genes including microRNAs 6, led us to develop a strategy that combine data from publically available resources, computational algorithms and in vitro studies to decipher the potential function of a transcription factor of interest in regulating transcription of microRNAs. The strategy suggested herein includes 2 critical steps. First we use publically available data to identify transcription factor binding in microRNA promoters. To that end we use Chip-seq data published by the ENCODE consortium to identify transcription factor binding and epigenetic signature that typifies promoter as an indirect marker of microRNA-promoters. Crossing the genetic coordinates from these datasets provides a gross estimation of how frequent a transcription factor of interest binds to microRNA-promoter. Second by using shRNA technology to silence the expression of a transcription factor and subjecting the cells to microRNA-array, it is possible to explore the functional significance of a transcription factor on the microRNA-transcriptome.

The variability in microRNA expression is only partially explained by transcription factor dependent regulation. Stoichiometric variability and other cellular or extracellular factors play an important role and are not simulated in the proposed algorithm. Other limitations include the following: The promoter analysis based on the H3K4me3 signature was done on 939 annotated microRNAs. Since then the genomic locations of many more microRNAs have been identified. However to the best of our knowledge a more comprehensive list that is based on an updated database has not been published yet. Of the 939 microRNA genes, the H3K4me3 which typifies the promoter region was identified in 781 microRNA genes (83%). Hence, while this analysis is clearly based on an incomplete dataset it captures a significant fraction of the microRNA-transcriptome.

Moreover, epigenetics signature is in part imprinted and in part cell-specific. Therefore, the generability of putative promoters that were defined by epigenetics markers may be questioned. Because H3K4me3 persists independent of transcription 18 it is generally considered a marker of imprinted promoters. The analytic algorithm we propose herein may therefore miss cell-specific microRNA promoters if these promoters were identified at a different cell-type. Finally, any conclusion should be tested and confirmed empirically. Most notably, the association between down regulation of a transcription factor (using shRNA approach) and microRNA expression (identified by microRNA array) can only suggest a direct transcriptional role that should be confirmed by acceptable assays such as chromatin immunoprecipitation (ChIP) or electrophoretic mobility shift assay (EMSA). Modifications to the method suggested herein include different ways of identifying microRNA promoter and different ways of knocking down the expression of a transcription factor, for example, small interfering RNA instead of shRNA. The transcription regulation of microRNAs may be substantially different for microRNAs that reside within protein coding genes (intragenic microRNAs) and those that do not (intergenic microRNAs). Because intragenic microRNAs are commonly transcribed in conjunction with their host genes 19 the promoter is usually found immediately upstream the transcription start site. However for many intergenic microRNAs the transcription start site is poorly annotated, and prediction tools that are commonly used for protein coding genes perform poorly 20.

Subscription Required. Please recommend JoVE to your librarian.


The authors declare no competing financial interests.


This study was supported by a grant from the CLL Global Research Foundation. The University of Texas MD Anderson Cancer Center is supported in part by the National Institutes of Health through a Cancer Center Support Grant (P30CA16672).


Name Company Catalog Number Comments
Lipofectamin 2000 Life Technologies 11668027
0.45 µm syringe filter Thermo Scientific (Nalgene) 190-2545
Amicon ultracentrifugal filter device with threshold of 100 kDa Merck Millipore
Polybrene Merck Millipore TR-1003-G
TRIzol reagent Life Tachnologies (Invitrogen) 15596-026
293 Cell line human Sigma-Aldrich 85120602



  1. Bentwich, I., et al. Identification of hundreds of conserved and nonconserved human microRNAs. Nat Genet. 37, 766-770 (2005).
  2. Hata, A. Functions of microRNAs in cardiovascular biology and disease. Annu Rev Physiol. 75, 69-93 (2013).
  3. Bartel, D. P. MicroRNAs: target recognition and regulatory functions. Cell. 136, 215-233 (2009).
  4. Piriyapongsa, J., Jordan, I. K., Conley, A. B., Ronan, T., Smalheiser, N. R. Transcription factor binding sites are highly enriched within microRNA precursor sequences. Biol Direct. 6, 61 (2011).
  5. Rozovski, U., et al. Signal transducer and activator of transcription (STAT)-3 regulates microRNA gene expression in chronic lymphocytic leukemia cells. Mol Cancer. 12, 50 (2013).
  6. Turner, M. J., Slack, F. J. Transcriptional control of microRNA expression in C. elegans: promoting better understanding. RNA Biol. 6, 49-53 (2009).
  7. Cui, Q., Yu, Z., Pan, Y., Purisima, E. O., Wang, E. MicroRNAs preferentially target the genes with high transcriptional regulation complexity. Biochem Biophys Res Commun. 352, 733-738 (2007).
  8. Yamakuchi, M., Lowenstein, C. J. MiR-34, SIRT1 and p53: the feedback loop. Cell Cycle. 8, 712-715 (2009).
  9. Baer, C., et al. Extensive promoter DNA hypermethylation and hypomethylation is associated with aberrant microRNA expression in chronic lymphocytic leukemia. Cancer Res. 72, 3775-3785 (2012).
  10. Hazan-Halevy, I., et al. STAT3 is constitutively phosphorylated on serine 727 residues, binds DNA, and activates transcription in CLL cells. Blood. 115, 2852-2863 (2010).
  11. Melo, S. A., et al. A TARBP2 mutation in human cancer impairs microRNA processing and DICER1 function. Nat Genet. 41, 365-370 (2009).
  12. Akira, S. Functional roles of STAT family proteins: lessons from knockout mice. Stem Cells. 17, 138-146 (1999).
  13. Frank, D. A., Mahajan, S., Ritz, J. B lymphocytes from patients with chronic lymphocytic leukemia contain signal transducer and activator of transcription (STAT) 1 and STAT3 constitutively phosphorylated on serine residues. J Clin Invest. 100, 3140-3148 (1997).
  14. Calin, G. A., et al. MicroRNA profiling reveals distinct signatures in B cell chronic lymphocytic leukemias. Proc Natl Acad Sci U S A. 101, 11755-11760 (2004).
  15. Consortium, E. P. A user's guide to the encyclopedia of DNA elements (ENCODE). PLoS biology. 9, e1001046 (2011).
  16. Miranda, K. C., et al. A pattern-based method for the identification of MicroRNA binding sites and their corresponding heteroduplexes. Cell. 126, 1203-1217 (2006).
  17. Consortium, E. P. An integrated encyclopedia of DNA elements in the human genome. Nature. 489, 57-74 (2012).
  18. Lau, J. C., Hanel, M. L., Wevrick, R. Tissue-specific and imprinted epigenetic modifications of the human NDN gene. Nucl Acids Res. 32, 3376-3382 (2004).
  19. Corcoran, D. L., et al. Features of mammalian microRNA promoters emerge from polymerase II chromatin immunoprecipitation data. PLoS One. 4, e5279 (2009).
  20. Bhattacharyya, M., Feuerbach, L., Bhadra, T., Lengauer, T., Bandyopadhyay, S. MicroRNA transcription start site prediction with multi-objective feature selection. Stat Appl Genet Mol Biol. 11, Article 6 (2012).



    Post a Question / Comment / Request

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

    Usage Statistics