Asynchronous transcriptional silencing of individual retroviral genomes in embryonic cells
© Schlesinger et al.; licensee BioMed Central Ltd. 2014
Received: 5 February 2014
Accepted: 1 April 2014
Published: 17 April 2014
Retroviral DNAs are profoundly silenced at the transcriptional level in embryonic cell types. The transcriptional profile of pluripotent stem cells has been demonstrated to be extremely heterogeneous from cell to cell, and how the silencing of retroviral DNAs is achieved is not yet well characterized.
In the current study, we investigated the transcriptional silencing dynamics in stem cells by independently monitoring the expression of two Moloney murine leukemia virus (MMLV) retroviral vectors newly introduced into embryonic carcinoma (EC) cells. Although MMLV is efficiently silenced by epigenetic mechanisms in most such cells, a small number of the doubly-transduced EC cells transiently show double-positive proviral expression. These cells were sorted and their expression patterns were studied over time as silencing is established.
Our data suggest that retroviral silencing occurs stochastically, in an individual locus-specific fashion, and often without synchronous silencing of both viruses in the same cells. Surprisingly, the chromatin modifications that mark the silenced proviruses are unchanged even in cells that temporarily escape silencing. This local silencing effect is a feature of stem cell epigenomic regulation that has not previously been revealed.
KeywordsEmbryonic stem cells Transcriptional silencing Chromatin modifications
Developmental programs are executed by the tightly controlled and temporally coordinated transcriptional regulation of large sets of genes. As cells move through developmental stages during embryogenesis, groups of genes are often synchronously turned on and off to induce specific changes in cell phenotype and in the capabilities needed for tissue and organ formation. This gene regulation is mediated by trans-acting transcription factors, and is accompanied by long-lasting alterations in chromatin folding, histone modifications, and DNA methylation at the genomic regions of the genes being regulated. While the sets of coordinately regulated genes may sometimes be genetically linked at a single chromosomal region, they are most often unlinked and dispersed at many disparate locations on many chromosomes. In these cases the trans-acting machinery must find and act on the multiple loci in a coordinated, synchronous manner. The timing of these regulatory events can be tightly synchronous, but there is some inherent noise or variability in the regulatory machinery, and thus there can be considerable cell-to-cell fluctuation in gene expression . As a result, even genetically identical cells in a largely homogeneous environment can display different phenotypes. This variability can arise either from stochastic fluctuations in biochemical reactions that regulate gene expression in “trans” (“extrinsic” variability), or from preexisting epigenetic heterogeneity of the genes being regulated (“intrinsic” variability) . These fluctuations in gene expression can play useful and important roles in development. Fate choice in pluripotent stem cells involves the modulation of networks of transcription factors occurring in an apparently stochastic fashion and resulting in a heterogeneous cell population [1, 3, 4]. The heterogeneity in expression of many components of the transcriptional network was shown to be a key feature of pluripotency [5–7].
Retroviral DNAs provide a unique tool for the analysis of gene expression because they can be introduced into the genome at will by infection, and because they are introduced as “naked”, unmodified, newly synthesized DNA that is assembled into chromatin by the addition of new nucleosomes. Vector genomes that express reporter genes are powerful tools to monitor regulation of transcription in infected cells . Regulatory elements within the newly integrated DNA have to be recognized by trans-acting factors, and the subsequent expression pattern of the DNA has to be established by those factors and by the preexisting state of the chromatin around the site of insertions. Although retroviral expression can be affected by the site of integration, it is most profoundly influenced by the cell type. In differentiated cells, infection by the MLVs typically results in high-level constitutive expression of the provirus, while in mouse embryonic stem cells, embryonic carcinoma cell lines, and other primitive cell types, the viral DNA is heavily silenced at the transcriptional level [9–11]. The silencing of retrovirus DNAs occurs rapidly when the virus contains a conserved sequence element termed the proline primer binding site (PBS), an 18-nucleotide sequence complementary to the 3′ end of proline tRNA, the tRNA primer used for initiation of reverse transcription by MMLV [12, 13]. Silencing at the proline PBS is mediated by the zinc finger DNA binding protein ZFP809, and by a well-characterized silencing protein, Trim28/Kap-1/Tif1b, which interacts with ZFP809 [14, 15]. Retroviral vectors utilizing alternative PBS sequences that are not recognized by this silencing machinery escape the rapid silencing but are still subject to some transcriptional repression [16, 17]. The machinery mediating this silencing is also used to repress or regulate expression of many of the endogenous proviruses resident in the mouse genome, most strikingly the so-called IAP elements , through the DNA binding factor YY1 . We note, however, that not all endogenous proviruses are equally or even similarly regulated in this pattern, and that some retroelements show the inverse behavior, expressing well in ES cells and not in differentiated cells; these include the HERV-H proviruses in human cells  and the virus-related L1td1/Ecat11 gene in mouse . For these elements other regulatory factors must be in play.
F9 cells infected with two reporter viruses rapidly and efficiently silence both proviruses
Transiently-expressing PBSpro viruses are rapidly silenced, but not globally and synchronously
Isolation of two double-positive populations – one stochastically silencing and one permanently expressing
Stochastically expressed proviruses are still marked by repressive epigenetic marks
Different histone modifications on the proviral alleles
The experiments above show that embryonic cell populations that have silenced most proviruses still contain some cells that transiently transcribe the viral DNAs at low frequencies, and then quickly shut them off. We can recover such expressing cells by sorting, and we can watch the kinetics of the shutoff on a cell-by-cell basis. The results suggest that the silencing of retroviruses in these cells is most often a stochastic event mediated through local changes occurring independently at each provirus, and is not mediated by a global change in the state of the cell that acts simultaneously in a coordinated way on all the proviruses in a cell. This characteristic seems to be true whether the cells utilize the highly efficient PBSpro-directed machinery, or the less efficient PBS-independent machinery. The local, cis-acting nature of the suppression seen in these cells is reminiscent of the cis-acting control of expression seen in the course of silencing of MLV vectors introduced in successive infections [26, 27].
EC cells have long been known for their ability to silence incoming viral DNAs, first documented for the silencing of the papova viruses . This behavior is manifested by many mouse EC lines, including the F9 studied here and the PCC4 line , as well as authentic embryonic stem cell populations. The silencing of retroviral DNAs in our EC cells is moderately rapid, with expression decreasing sharply over the course of a few days at most. We note that the true time of transcriptional shutoff may be even faster than the apparent time course, because the GFP and mCherry proteins are relatively stable and the observed loss of fluorescent signal cannot be faster than the rate of decay of the accumulated proteins, likely occurring over several hours. Similarly, the appearance of a single-positive cell from a previously double-positive cell requires that the difference in time between the shutoff of the two reporters must be at least several hours. Thus, the asynchrony in the shutoff that we see in our experiments is very significant.
It was previously shown that genes expressed at low levels tend to be bound by fewer transcriptional regulators, both activators and repressors, which results in high cell-to-cell variability . This variability has been proposed to be the result of the transcriptional regulators fluctuations , suggesting that as the number of different regulators increases, target sensitivity and variability of expression decreases . In our experiments, the variability in timing of the establishment of silencing does not strongly depend on the level of expression. Both the efficiently silenced PBSpro and the less efficiently silenced PBSgln viruses are stochastically silenced. Thus, both the potent silencing mediated by the PBSpro-dependent ZFP809/TRIM28 system, and the silencing of the less potent non-PBS system, show similar asynchronous silencing. The behavior of both these systems may result from local fluctuations in the regulators acting at each provirus.
The status of the histone modifications and DNA methylation at the proviruses suggests that our EC cells typically mark the incoming viral DNAs for silencing soon after infection. Even those cells selected as transiently double-positive and single-positive, as well as double-negative, contain proviral DNAs exhibiting the chromatin and DNA marks of silent genes. Thus, the brief expression of the proviruses in these cells occurs in spite of these marks, and relatively quickly responds to the marks by shutting down. It is also possible that because most cells contain more than one copy of each virus, and because only one copy in the double-positive population may be expressing at any given moment, the silencing chromatin marks that we observe are present on the silent alleles. The expressed alleles thus may be at least partially depleted of the silencing marks, and this loss may be obscured by the silent alleles.
The rare population of cells that are stable expressors of both GFP and mCherry, recovered as high expressers after two sorts separated by many days in culture, are unusual escapees of the silencing process. These cells may have been selected to have proviral integrations at special loci. They may be similar to an EC line described by the Jaenisch lab, isolated after selection for expression of a marker gene delivered by a vector with a proline PBS . It is noteworthy that the proviruses in these cells show the histone marks, and DNA methylation status, of completely open chromatin and highly expressed genes. But the fact that these cells also show altered chromatin marks and DNA methylation status for their endogenous retrovirus DNAs strongly suggests that they have undergone a global change in their silencing machinery acting in trans, and that the expression of both GFP and mCherry vectors is not solely a consequence of their integration sites. We surmise that the cell sorting regimen has selected for a cell population that has stably lost some components of the silencing system. Further analysis of the repertoire of expressed genes in these cells may ultimately allow for identification of the basis of their permissivity.
The silencing of newly introduced retroviral genomes in embryonic cells is achieved by the recognition of the viral DNAs by transcription factors and chromatin modifying machinery. This machinery acts on the independently integrated proviruses in a temporally asynchronous manner, such that each provirus can be silenced without strict coordination with another provirus in the same cell. The findings suggest that there is a stochastic aspect to the determination of the chromatin modifications that are imposed on each provirus. Rare clones in which more profound escape from silencing arise by more dramatic global changes in the embryonic cell state.
Cells and viruses
F9 and NIH3T3 cells were cultured in DMEM with 10% FBS, 2 mM glutamine, 1000 U/mL penicillin, 100 mg/mL Streptomycin. F9 cells were cultured on gelatinized tissue culture plates. All cells were cultured at 37°C in 5% CO2. Viruses were prepared as previously described using pNCA-GFP/mCherry vectors  pseudotyped with VSV-G amphotropic envelope glycoprotein. At 48 hr posttransfection, culture supernatants were harvested, filtered through a 0.45 μm filter, concentrated by ultracentrifugation, and resuspended in growth medium.
Virus copy number determination
The difference in threshold cycle (CT) values (DCT) between Q-PCR assays using two of the primers on the pro-virus, 40NT and PBS, was used to monitor proviral DNA copy number. The Gapdh primer set for the genomic DNA was used to normalize the amount of genomic DNA. A one copy number standard was established by infecting NIH 3 T3 cells at a very low multiplicity of infection (MOI) with the MLV-GFP vector and sorting the GFP(+) cell population (10% of the total), ensuring that a single copy of GFP virus was present . The ratio of GFP to genomic DNA (Gapdh) in this sample is normalized to 1 (one copy of provirus per cell genome), and this sample is taken as the calibrator. The differences in DCTs (DDCT) for the samples of interest and the calibrator are used to estimate the relative quantity (RQ) of provirus by using the formula RQ = 2 -DDCT. The copy number values given were obtained by averaging results from three PCR reactions. Uninfected NIH 3 T3 cells were used as negative control. In the flow analysis results presented, all numbers were normalized to the infection efficiency as seen by the % of NIH3T3 expressing cells. The numbers are given as GFP-positive F9 cells/GFP-positive NIH3T3 cells x100; this is to correct for variation in multiplicity with different viruses.
Flow cytometry and sorting
GFP-positive cells were isolated on a cell sorter (FACSAria Cell Sorter; BD Biosciences). Data were acquired on an automated cell analyzer (LSR II; BD Biosciencs) and analyzed with FlowJo software (Treestar).
Bisulfite analysis of methylation
Bisulfite conversion of genomic DNA was carried out using the Zymo EZ DNA Methylation-Gold™ Kit. PCR primers were designed using Methyl Primer Express software version 1.0 (https://www2.appliedbiosystems.com). See Additional file 4: Table S1.
Chromatin immunoprecipitation (ChIP)
107 cells were crosslinked with 1% formaldehyde at room temperature for 10 min. Chromatin was extracted and then sonicated to an average size of 300–1,000 bp. Immunoprecipitation was carried out by using Magna ChIP™ kit as recommended by the manufacturer (Millipore) and then purified using QIAquick PCR purification kit (Qiagen). Antibodies used (about 5 μg per 10–30 μg of DNA) were: Anti YY1 (H-414, Santa Cruz), Anti Trim28 (Anti tif1b- MAB3662, Millipore), Anti-trimethyl-Histone H3 (Lys9) (07–442, Millipore), Anti-trimethyl-Histone H3 (Lys27) (07–449, Millipore) and Anti-trimethyl-Histone H3 (Lys4) (07–473, Millipore). IP with IgG antibody (sc-2027, Santa Cruz) resulted in enrichment level < 1. Amplification was carried out by real-time PCR, and the bound/input values were then normalized by setting the negative control gene results to 1. Multiple assays of the same sample or the same gene sequence were analyzed in separate immunoprecipitations. All immunoprecipitations were repeated at least 3 times. Primer sequences used for qPCR are listed in Additional file 4: Table S1. In Figure 6D,E bound/input ratios were normalized to the U5-PBS primers value of the same sample and results are presented relative to the value in the negative sorted cells fraction.
This work was supported by NCI grant R37 CA 30488 from the National Cancer Institute and the National Institutes of Health. S.P.G. is an Investigator of the Howard Hughes Medical Institute.
- Balázsi G, van Oudenaarden A, Collins J: Cellular decision making and biological noise: from microbes to mammals. Cell. 2011, 144: 910-925.PubMed CentralView ArticlePubMedGoogle Scholar
- Raser J, O’Shea E: Control of stochasticity in eukaryotic gene expression. Science (New York, NY). 2004, 304: 1811-1814.View ArticleGoogle Scholar
- Martinez Arias A, Brickman J: Gene expression heterogeneities in embryonic stem cell populations: origin and function. Curr Opin Cell Biol. 2011, 23: 650-656.View ArticlePubMedGoogle Scholar
- Buchanan FQ, Rochette-Egly C, Asson-Batres MA: Detection of variable levels of RARalpha and RARgamma proteins in pluripotent and differentiating mouse embryonal carcinoma and mouse embryonic stem cells. Cell Tissue Res. 2011, 346: 43-51.View ArticlePubMedGoogle Scholar
- Hayashi K, Lopes S, Tang F, Surani M: Dynamic equilibrium and heterogeneity of mouse pluripotent stem cells with distinct functional and epigenetic states. Cell Stem Cell. 2008, 3: 391-401.View ArticlePubMedGoogle Scholar
- Kobayashi T, Mizuno H, Imayoshi I, Furusawa C, Shirahige K, Kageyama R: The cyclic gene Hes1 contributes to diverse differentiation responses of embryonic stem cells. Gene Dev. 2009, 23: 1870-1875.PubMed CentralView ArticlePubMedGoogle Scholar
- Chambers I, Silva J, Colby D, Nichols J, Nijmeijer B, Robertson M, Vrana J, Jones K, Grotewold L, Smith A: Nanog safeguards pluripotency and mediates germline development. Nature. 2007, 450: 1230-1234.View ArticlePubMedGoogle Scholar
- Franz K, Singh A, Weinberger LS: Lentiviral vectors to study stochastic noise in gene expression. "Synthetic Biology, Part A", Methods in Enzymology, Volume 497. Edited by: Voight C. 2011, Amsterdam NL/Academic Press, NY NY: Elsevier, 603-622.Google Scholar
- Teich NM, Weiss RA, Martin GR, Lowy DR: Virus infection of murine teratocarcinoma stem cell lines. Cell. 1977, 12: 973-982.View ArticlePubMedGoogle Scholar
- Bogerd HP, Wiegand HL, Hulme AE, Garcia-Perez JL, O’Shea KS, Moran JV, Cullen BR: Cellular inhibitors of long interspersed element 1 and Alu retrotransposition. Proc Natl Acad Sci U S A. 2006, 103: 8780-8785.PubMed CentralView ArticlePubMedGoogle Scholar
- Rowe HM, Jakobsson J, Mesnard D, Rougemont J, Reynard S, Aktas T, Maillard PV, Layard-Liesching H, Verp S, Marquis J, Spitz F, Constam DB, Trono D: KAP1 controls endogenous retroviruses in embryonic stem cells. Nature. 2010, 463: 237-240.View ArticlePubMedGoogle Scholar
- Harada F, Peters GG, Dahlberg JE: The primer tRNA for Moloney murine leukemia virus DNA synthesis. Nucleotide sequence and aminoacylation of tRNAPro. J Biol Chem. 1979, 254: 10979-10985.PubMedGoogle Scholar
- Petersen R, Kempler G, Barklis E: A stem cell-specific silencer in the primer-binding site of a retrovirus. Mol Cell Biol. 1991, 11: 1214-1221.PubMed CentralView ArticlePubMedGoogle Scholar
- Wolf D, Goff SP: TRIM28 mediates primer binding site-targeted silencing of murine leukemia virus in embryonic cells. Cell. 2007, 131: 46-57.View ArticlePubMedGoogle Scholar
- Wolf D, Goff SP: Embryonic stem cells use ZFP809 to silence retroviral DNAs. Nature. 2009, 458: 1201-1204.PubMed CentralView ArticlePubMedGoogle Scholar
- Pannell D, Ellis J: Silencing of gene expression: implications for design of retrovirus vectors. Rev Med Virol. 2001, 11: 205-217.View ArticlePubMedGoogle Scholar
- Cherry SR, Biniszkiewicz D, van Parijs L, Baltimore D, Jaenisch R: Retroviral expression in embryonic stem cells and hematopoietic stem cells. Mol Cell Biol. 2000, 20: 7419-7426.PubMed CentralView ArticlePubMedGoogle Scholar
- Lueders KK, Kuff EL: Sequences associated with intracisternal A particles are reiterated in the mouse genome. Cell. 1977, 12: 963-972.View ArticlePubMedGoogle Scholar
- Schlesinger S, Goff S: Silencing of proviruses in embryonic cells: efficiency, stability and chromatin modifications. EMBO Rep. 2013, 14: 73-79.PubMed CentralView ArticlePubMedGoogle Scholar
- Santoni FA, Guerra J, Luban J: HERV-H RNA is abundant in human embryonic stem cells and a precise marker for pluripotency. Retrovirology. 2012, 9: 111-PubMed CentralView ArticlePubMedGoogle Scholar
- Wong RC, Ibrahim A, Fong H, Thompson N, Lock LF, Donovan PJ: L1TD1 is a marker for undifferentiated human embryonic stem cells. PLoS One. 2011, 6: e19355-PubMed CentralView ArticlePubMedGoogle Scholar
- Berstine EG, Hooper ML, Grandchamp S, Ephrussi B: Alkaline phosphatase activity in mouse teratoma. Proc Natl Acad Sci U S A. 1973, 70: 3899-3903.View ArticlePubMedGoogle Scholar
- Artzt K, Dubois P, Bennett D, Condamine H, Babinet C, Jacob F: Surface antigens common to mouse cleavage embryos and primitive teratocarcinoma cells in culture. Proc Natl Acad Sci U S A. 1973, 70: 2988-2992.PubMed CentralView ArticlePubMedGoogle Scholar
- Macfarlan T, Gifford W, Driscoll S, Lettieri K, Rowe H, Bonanomi D, Firth A, Singer O, Trono D, Pfaff S: Embryonic stem cell potency fluctuates with endogenous retrovirus activity. Nature. 2012, 487: 57-63.PubMed CentralPubMedGoogle Scholar
- Katz R, Jack-Scott E, Narezkina A, Palagin I, Boimel P, Kulkosky J, Nicolas E, Greger J, Skalka A: High-frequency epigenetic repression and silencing of retroviruses can be antagonized by histone deacetylase inhibitors and transcriptional activators, but uniform reactivation in cell clones is restricted by additional mechanisms. J Virol. 2007, 81: 2592-2604.PubMed CentralView ArticlePubMedGoogle Scholar
- Taketo M, Gilboa E, Sherman MI: Isolation of embryonal carcinoma cell lines that express integrated recombinant genes flanked by the Moloney murine leukemia virus long terminal repeat. Proc Natl Acad Sci U S A. 1985, 82: 2422-2426.PubMed CentralView ArticlePubMedGoogle Scholar
- Sorge J, Cutting AE, Erdman VD, Gautsch JW: Integration-specific retrovirus expression in embryonal carcinoma cells. Proc Natl Acad Sci U S A. 1984, 81: 6627-6631.PubMed CentralView ArticlePubMedGoogle Scholar
- Swartzendruber DE, Lehman JM: Neoplastic differentiation: interaction of simian virus 40 and polyoma virus with murine teratocarcinoma cells in vitro. J Cell Physiol. 1975, 85: 179-187.View ArticlePubMedGoogle Scholar
- Venters BJ, Wachi S, Mavrich TN, Andersen BE, Jena P, Sinnamon AJ, Jain P, Rolleri NS, Jiang C, Hemeryck-Walsh C, Pugh BF: A comprehensive genomic binding map of gene and chromatin regulatory proteins in Saccharomyces. Mol Cell. 2011, 41: 480-492.PubMed CentralView ArticlePubMedGoogle Scholar
- Maamar H, Raj A, Dubnau D: Noise in gene expression determines cell fate in Bacillus subtilis. Science. 2007, 317: 526-529.View ArticlePubMedGoogle Scholar
- Carey LB, van Dijk D, Sloot PM, Kaandorp JA, Segal E: Promoter sequence determines the relationship between expression level and noise. PLoS Biol. 2013, 11: e1001528-PubMed CentralView ArticlePubMedGoogle Scholar
- Barklis E, Mulligan RC, Jaenisch R: Chromosomal position or virus mutation permits retrovirus expression in embryonal carcinoma cells. Cell. 1986, 47: 391-399.View ArticlePubMedGoogle Scholar
- Ooi SK, Wolf D, Hartung O, Agarwal S, Daley GQ, Goff SP, Bestor TH: Dynamic instability of genomic methylation patterns in pluripotent stem cells. Epigenetics Chromatin. 2010, 3: 17-PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.