Fig 1.
The Ty1 retrotransposon lifecycle and transcriptional regulation.
(A) The Ty1 mobility lifecycle. Following transcription, Ty1 RNA is exported to the cytoplasm where it localizes co-translationally to a microscopically distinct cytoplasmic focus known as the retrosome. The retrosome is the site of assembly of the VLPs, which serve as the sites for Ty1 protein maturation and reverse transcription of Ty1 cDNA. Ty1 cDNA is then transported back into the nucleus and integrated into the host genome. (B) Genetic structure of Ty1. Ty1 contains two ORFs (GAG and POL) flanked by a 5′ and a 3′ LTR. Ty1 is transcribed from the 5′ untranslated region (R-U5) in the 5′ LTR beginning at nt +238, and it terminates within the 3′ LTR. In addition to the primary Ty1 transcript, the Ty1i transcript initiates within the GAG ORF at nt +1000 to produce a 5.0 kb truncated transcript. Antisense (AS) RNA is also transcribed from within the GAG ORF to the 5′ LTR. The Ty1 5′ LTR as well as the first 1 kb of the GAG ORF contain several transcription factor binding sites, as well as a TATA element. Sites for Ste12, Tec1, Tye7, Mcm1, Tea1, and Rap1 occur downstream of the Ty1 TSS. The transcription termination sites in the 3′ LTR (TS1 and TS2) are located in the R-U5 region. Adapted from [18].
Fig 2.
Mediator subunit deletions influence Ty1 mobility in a module-specific manner.
(A) The Mediator transcriptional coactivator complex is composed of head (red), middle (blue), tail (yellow), and kinase (grey) modules (figure based on recent cryo-EM structure [50]). Individual subunits investigated in this study are labeled. (B) Schematic of the retromobility assay [19]. The HIS3 ORF is inserted downstream of POL in the opposite orientation of Ty1 and contains an intron that is in the sense orientation relative to Ty1. Splicing of this intron from the Ty1his3AI transcript, followed by integration, results in His+ colonies. (C) The frequency of retrotransposition, shown on a log scale, of the chromosomal Ty1his3AI-3114 element was measured in congenic WT, spt3Δ and Mediator subunit deletion strains. Error bars represent standard deviation for three biological replicates, except for med18Δ (n = 2), where the range of measurement is indicated (but is too small to be visible). Blue circles denote values that represent upper limit retrotransposition estimates in strains with no retrotransposition events among the total number of cells assayed. Bars are color-coded to match structural organization as shown in (A).
Fig 3.
Mediator subunits influence Ty1 cDNA levels without altering levels of Ty1 RNA or Gag protein.
(A) Quantitative northern blot analysis of sense-strand (Ty1) and antisense-strand (Ty1 AS) RNA. Total RNA was fractioned by gel electrophoresis, blotted, and the membranes probed for Ty1 AS RNA, followed by stripping and probing for Ty1 RNA using strand-specific riboprobes as schematized at the top, and for 18S rRNA as a loading control. The graph on the right shows the quantitation of Ty1 RNA and Ty1 AS RNA levels from two biological replicates, each relative to the 18S subunit rRNA level, normalized to WT levels for Ty1 and to spt3Δ yeast for Ty1 AS. (B) Western blot of total cell lysates probed for Gag using a polyclonal antibody against p18 [24]. Data for med5Δ and med16Δ is derived from two biological replicates; all other values represent data from three biological replicates, and values are normalized to WT. (C) Quantitative Southern blot analysis to determine the level of unintegrated Ty1 cDNA (cDNA) relative to the amount of DNA in bands representing two genomic Ty1 elements (G1 and G2). Image quantification is representative of three biological replicates, only one of which is shown. All experiments were performed using congenic WT, spt3Δ and Mediator subunit deletion strains harboring Ty1his3AI-3114. Bars in (B) and (C) are color coded as in Fig 2, and all error bars represent s.d. except for measurements with n = 2, in which case the range is indicated.
Fig 4.
The Mediator tail acts on Ty1 mobility in an LTR promoter-dependent manner.
(A) Schematic of the PTEF1-Ty1his3AI element relative to a Ty1his3AI element with the standard LTR promoter. PTEF1-Ty1his3AI has a TEF1 promoter in place of Ty1 promoter elements in the U3 region of the 5′ LTR (See Fig 1C), while retaining the Ty1 TSS and R-U5 region of the 5′ LTR. (B) Retrotransposition frequency, shown on a log scale, for a plasmid-based Ty1his3AI (left) or PTEF1-Ty1his3AI (right) element in the WT and spt3Δ negative control strains (grey bars), Mediator head subunit gene deletion strains (red bars), middle subunit gene deletion strains (blue bars) and tail subunit gene deletion strains (yellow bars). (C) Retromobility frequencies, shown on a log scale, of a chromosomal Ty1kanMXAI element and a plasmid-borne PTEF1-Ty1his3AI element contained in the same wild type or med15Δ strain. All error bars represent s.d. Blue circles in (B) and (C) denote values that represent upper limit mobility estimates for strains in which most or all cultures had no His+ prototrophs (LTR-Ty1his3AI or PTEF1-Ty1his3AI) or G418R colonies (Ty1kanMXAI) per total number of Ura+ or Leu+ cells analyzed, respectively.
Fig 5.
Deletion of Mediator tail module triad subunits increases levels of polyA+ Ty1i RNA and p22-Gag.
(A) Northern blot probed for Ty1, Ty1i and PYK1 RNA, the latter as a loading control that has been shown not to be altered in med2Δ, med3Δ, or med15Δ yeast [57, 60]. Quantifications of Ty1 and Ty1i RNA relative to PYK1 RNA are averages of three biological replicates and are normalized to WT levels. (B) Pol II occupancy averaged over all 31 genomic Ty1 elements in wild type and med3Δ med15Δ yeast using ChIP-seq data from [61]. Ty1 elements begin at 0 kb on the x-axis, and the Ty1 TSS at +238 and Ty1i TSS at +1000 are marked by green bars on the x-axis. (C) Western blot of total cell lysate measuring levels of p22- and p18-Gag relative to the loading control, GAPDH. Quantitation, shown on a log scale, is the average ratio of p22-Gag or p18-Gag to GAPDH signal from three biological replicates of each strain. In panels (A) and (C), quantitation was performed on RNA or protein samples from the WT strain and congenic spt3Δ strain as a negative control (grey bars), Mediator head subunit gene deletion strains (red bars), middle subunit gene deletion strains (blue bars) and tail subunit gene deletion strains (yellow bars). All error bars represent s.d.
Fig 6.
Mediator occupancy at the Ty1 and Ty1i proximal promoters.
Occupancy of Med17-myc and Med15-myc from the Mediator head and tail modules, respectively, was determined by ChIP-seq and summed over all Ty1 elements in kin28-AA yeast that were otherwise wild type (WT), med20Δ, or carried the triple deletion med2Δ med3Δ med15Δ. Ty1 elements begin at 0 kb on the x-axis, and the Ty1 TSS at +238 and Ty1i TSS at +1000 are marked by green bars on the x-axis. ChIP-seq signals were normalized to an untagged kin28-AA control (S3 Fig). Reads deriving from the Ty1 TSS, and therefore in the LTR, are unavoidably also assigned to the 3’ LTR, leading to the observed signal in that region.
Fig 7.
Ty1i situated downstream of a repressed Ty1 element is repressed by Mediator tail module triad subunits.
(A) Schematic of the GAL1:Ty1ΔPOL cassette in pGTy1ΔPOL showing forward primer locations for detection of Ty1 RNA (blue) versus Ty1i RNA (red). Note that because Ty1i is contained within Ty1, the Ty1i primer reports both Ty1i and Ty1 transcripts. Both amplifications utilized the same reverse primer (purple), that crosses the deletion junction and contains sequences unique to the pGTy1ΔPOL element. A reverse primer specific for ACT1 mRNA was also used to synthesize cDNA used as a template for the PCR amplification. No PCR product was detected using the ΔPOL reverse primer when RNA from yeast lacking pGTy1ΔPOL was used as template. (B) Quantitation of the products of Reverse Transciption-PCR reactions using polyA+ RNA isolated from wild-type yeast bearing plasmid pGTy1ΔPOL, and grown in glucose-containing broth, and from a genomic DNA control. Aliquots were taken from reactions at the indicated number of cycles and analyzed by agarose gel electrophoresis (S6 Fig). Levels of RT-PCR amplification products using Ty1, Ty1i, and ACT1 primers are indicated. (C) Northern blot probed with a single-stranded riboprobe that hybridizes to Ty1, Ty1i, Ty1ΔPOL and Ty1iΔPOL RNA. Note that the full-length Ty1ΔPOL transcript cannot be distinguished from the rRNA band. Image is representative of three biological replicates. (D) Quantification for Ty1, Ty1i, and Ty1iΔPOL RNA from three biological replicate northern analyses. Graph bar colors correspond to WT strain and the congenic spt3Δ strain as a negative control (grey bars), Mediator head subunit gene deletion strain (red bars), middle subunit gene deletion strains (blue bars) and tail subunit gene deletion strains (yellow bars). Error bars represent s.d.
Fig 8.
Model for altered balance of utilization of Ty1 and Ty1i promoters in Mediator deletion mutants.
In wild type yeast, the Mediator complex acts at both the Ty1 and Ty1i proximal promoters, stimulating robust transcription of Ty1 and a small amount of Ty1i expression that is sufficient to restrict mobility. Deletion of subunits from the tail module triad increases Mediator activity at the Ty1i promoter, thus increasing Ty1i production and reducing retromobility. Conversely, when a subunit from the head module is deleted, the association between Mediator and the Ty1i promoter is perturbed, permitting increases in retromobility. Finally, when the SAGA-dependent Ty1 promoter is swapped for the strong, TFIID-dependent TEF1 promoter, the complex preferentially associates with the Ty1 promoter, irrespective of the influence of the tail module.