Pleiohomeotic interacts with the core transcription elongation factor Spt5 to regulate gene expression in Drosophila

Abstract

The early elongation checkpoint regulated by Positive Transcription Elongation Factor b (P-TEFb) is a critical control point for the expression of many genes. Spt5 interacts directly with RNA polymerase II and has an essential role in establishing this checkpoint, and also for further transcript elongation. Here we demonstrate that Drosophila Spt5 interacts both physically and genetically with the Polycomb Group (PcG) protein Pleiohomeotic (Pho), and the majority of Pho binding sites overlap with Spt5 binding sites across the genome in S2 cells. Our results indicate that Pho can interact with Spt5 to regulate transcription elongation in a gene specific manner.

Figure 1. Pho physically interacts with Spt5.

A) Yeast 2-hybid assay showing binding of full length Pho with the Spt5 DD domain and full length Spt5. Vector (pGBKT7) containing no insert was used as a control to demonstrate that Pho does not activate reporter gene expression in the absence of Spt5. B) Pho binds to immobilized GST-DD. Ten percent of the input Pho is run in left lane, immobilized GST in middle lane incubated with Pho as negative control. C) Western blots of co-immunoprecipitation (co-IP) assays from S2 cell extracts of Flag-tagged Spt5 with Myc-Spt4 (positive control), Myc-Pho, Myc-N-Pho (amino acids 1–351), Myc-C-Pho (351–520), Myc-GFP (negative control) and no protein. D) Western blots of co-IP assays from S2 cell extracts of Flag-tagged W049 variant of Spt5 with Myc-Spt4 (positive control), Myc-Pho, Myc-GFP (negative control) and no protein.

Figure 2. Modification of the extra sex combs phenotype of pho^cv/pho^cv mutants by Spt5 and NELF mutant alleles.

A chart representing the frequency of ectopic sex combs in pho^cv/pho^cv mutants and siblings heterozygous for Spt5^W049, Spt5^{MGE −3} or NELF-A^KG over wild-type chromosomes. p values from two proportion z-tests are shown.

Figure 3. Pho and Spt5 function together in wing maturation.

A wing inflation phenotype is observed in approximately 10% of pho^cv/pho^cv B) and 51% of 386Y-Gal4>UAS-RNAi pho males (n = 136), but not in 765-Gal4>UAS-RNAi-pho. E) Percentage of flies of indicated genotypes displaying wing inflation phenotypes. F) Ventral view of da-Gal4>UAS-RNAi-pho male, red arrow points to ectopic sex comb on middle (mesothoracic) leg. G) Dorsal view of da-Gal4>UAS-RNAi-pho male displaying homeotic transformations in the abdominal segments.

Figure 4. Depletion of Spt5 leads to cell death in vivo.

A) Homozygous clones of the Spt5^MGE null allele are not viable. Attempts were made to make clones of homozygous Spt5^MGE cells using the FLP/FRT technique . Third instar imaginal wing disk (anterior to the left and dorsal to the top) stained for GFP. All cells stain green and are thus either heterozygous or homozygous (bright green) for the FRT42B, GFP chromosome; loss of GFP would mark clones of homozygous FRT42B, Spt5^MGE. Similarly, when we induced homozygous germ-line clones of Spt5^MGE in females using the FLP/FRT/ovoD technique , they did not lay any eggs indicating that homozygous Spt5^MGE clones are cell lethal (data not shown). B) Residual wing stub from fly expressing 765-Gal4>UAS-RNAi-Spt5 at 18°C the portion of the wing expressing 765-Gal4 does not develop as there is a deficit of cells consistent with expression of UAS-RNAi-Spt5 being lethal to cells.

Figure 5. Meta-analysis of ChIP data for Pho , Spt5 , NELF , and GAF binding across the genome of Drosophila S2 cells.

Asterisks denote the NELF data from . A) Venn diagram showing peaks the overlap of Pho binding with Spt5. B) Heat maps shows peaks of Spt5, NELF-B and Pho binding relative to the TSS (centre of each column) for all coding genes annotated in the genome from the Ensembl database (Release 5.48). Plots show 200 bp up and downstream of the TSS. SEQMINER was used to cluster and visualise the data using the default settings (the 'Kmeans raw' clustering normalization method with 10 expected clusters) . C) Venn diagram showing overlap of Pho and Spt5 peaks with NELF-B binding. D) Overlap of Pho peaks with peaks derived from NELF-B* and NELF-E* datasets. E) Venn diagram showing the overlap between the NELF-B peaks from and NELF-B* and NELF-E* datasets from . Differences in the profiles of NELF binding between these two studies may be due to the use of different antibodies and/or experimental conditions to perform ChIP-chip. F) Overlap between Pho and GAF peaks in S2 cells. Note: the total number of peaks for an individual factor can vary due to the merging of several overlapping peaks into a single peak if multiple peaks of one factor overlap with a single peak of another. G) Overlap of ChIP peaks for Spt5 and Pho binding at the Hsp70Aa gene. Note: There are differences in the publicly available track format of Spt5 (bedgraph) and Pho (smoothed wig) data, and differences in the tiling array used for the ChIP-chip experiments (Spt5: Nimblegen Henikoff_Dmel_r52_ChIP tiling design, mean probe length = 53 bp, mean distance between probes = 12 bp; Pho: Affymetrix Drosophila v2.0R tiling array, probe length = 25 bp, mean distance between probes = 38 bp).