Volume 47, Issue 1, Pages 133-139 (July 2012) Metabolic Enzyme IMPDH Is Also a Transcription Factor Regulated by Cellular State  Elena N. Kozhevnikova, Jan A. van der Knaap, Alexey V. Pindyurin, Zeliha Ozgur, Wilfred F.J. van Ijcken, Yuri M. Moshkin, C. Peter Verrijzer  Molecular Cell  Volume 47, Issue 1, Pages 133-139 (July 2012) DOI: 10.1016/j.molcel.2012.04.030 Copyright © 2012 Elsevier Inc. Terms and Conditions

Molecular Cell 2012 47, 133-139DOI: (10.1016/j.molcel.2012.04.030) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 1 Regulation of IMPDH Nuclear Localization by Cell Cycle and Stress (A) Subcellular distribution of IMPDH. All S2 cells have cytoplasmic IMPDH, but in ∼60% of the cells IMPDH is excluded from the nucleus. S2 cells were stained with antibodies against IMPDH (green), and nuclei were visualized by DAPI staining of DNA (blue). (B and C) IMPDH nuclear localization is largely restricted to CycA- and CycB-positive cells. CycA (red; B), CycB (red; C), and IMPDH (green) were detected by immunofluorescence. (D) Quantification of the percentage of S2 cells with nuclear IMPDH, shown as bar graphs with standard deviations. Based on three independent experiments in which ∼1,000 cells were analyzed. (E) IMPDH distribution on polytene chromosomes stained with α-IMPDH (green) and α-PCNA (red) antibodies. The dotted white line separates chromosomes originating from two different nuclei. (F) Nuclear IMPDH is mainly restricted to the G2 phase of the cell cycle, as indicated by costaining with CycA, CycB, and PCNA as markers of cell-cycle stage. (G) Effect of H2O2, HU, or replication factor C depletion on IMPDH subcellular localization. The corresponding cell cycle profiles are shown in Figure S1. (H) Quantification of IMPDH subcellular distribution after metabolic stress. Analysis as described above. Molecular Cell 2012 47, 133-139DOI: (10.1016/j.molcel.2012.04.030) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 2 IMPDH Binds Chromatin and Represses Its Target Genes (A) Polytene chromosomes were stained with α-IMPDH (green) and α-RNA Pol II (red) antibodies. IMPDH binding to chromosomal region 39D-E, which harbors the histone gene cluster, is indicated. (B) Determination of IMPDH binding to histone genes in mock- (red) or IMPDH-depleted cells (IMPDH KD; pink) by ChIP-qPCR. (C) The expression of histone genes upon IMPDH knockdown was determined by RT-qPCR. Tub84B and brahma mRNA levels were averaged and used for normalization. Derepression is represented as fold change relative to mock-treated cells. The corresponding cell-cycle profiles are shown in Figure S2A. (D) Western immunoblotting analysis of mock- or IMPDH-depleted cells using antibodies directed against the indicated proteins. (E) ChIP-qPCR on four IMPDH target genes (E2F, Mlc2, CG6923, and JIL1) and two genes that were not bound by IMPDH (CG5160 and CG11777). IMPDH was ChIPed from mock- (red) or IMPDH-depleted (pink) cells. (F) The effect of IMPDH depletion on gene expression determined by RT-qPCR. For all experiments, mean and standard deviation were derived from three independent biological replicates. Molecular Cell 2012 47, 133-139DOI: (10.1016/j.molcel.2012.04.030) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 3 IMPDH Binding Sites Have the Tendency to Unwind and Form ssDNA (A) Heat map displaying the ChIP-chip enrichment of IMPDH at 2,920 binding sites, aligned at the center of the binding peaks. See also Figure S3. (B) IMPDH loci are CT rich. The average values of nucleotide content of aligned IMPDH binding sites in S2 cells are displayed as percentages. (C) Genomic IMPDH-bound DNA sequences within the His3-His4 promoter and the Mlc2 gene, as well as a region from the Tub84B not bound by IMPDH, were cloned in pGEM-T. Supercoiled plasmids were subjected to S1 nuclease treatment (lanes 1–4), or incubated with a buffer control (lanes 5–8), and then digested by ScaI. Products were resolved by agarose gel electrophoresis. (D) Sequencing of the S1/ScaI-generated fragments from the His3-His4 promoter region and Mlc2 gene. Nucleotide detection intensities are plotted, and gray boxes indicate the CT-rich sequences that are S1 nuclease sensitive and correspond to the oligonucleotides used in the band-shift assays in Figure 4C. Molecular Cell 2012 47, 133-139DOI: (10.1016/j.molcel.2012.04.030) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 4 IMPDH Is a Sequence-Specific ssDNA-Binding Protein (A) Position-weight matrices illustrating the overrepresentation of a T(C/T)C(T/C)C consensus motif in IMPDH bound fragments. ssDNA oligonucleotides harboring three copies of the CT-rich consensus motif (TOP), the complementary GA-rich bottom strand, the cognate dsDNA, and two mutant probes were radiolabeled and used in gel-motility-shift assays. (B) IMPDH DNA-binding assays using the indicated probes. As indicated, affinity purified α-IMPDH or irrelevant (α-GMPS) antibodies were added to the binding reaction. (C) IMPDH binding to CT-rich (upper), GA-rich (lower), and dsDNA probes derived from its genomic loci within the His3-His4 promoter, Mlc2, and E2f was tested by gel motility shift assays. His3-His4 and Mlc2 probes correspond to the boxed sequences in Figure 3D. (D) Cartoon of IMPDH and substitution mutations. (E) Western immunoblotting analysis of purified mock-expressed, WT IMPDH, IMPDH-C350S, and IMPDH-R243P-D245N expressed in 293-T cells. (F) Binding of WT or mutant IMPDH to the consensus TOP oligonucleotide tested by gel motility-shift assays. (G) Coomassie-stained gel of purified E. coli IMPDH (GuaB). (H) Binding of recombinant E. coli IMPDH (GuaB) protein to the indicated probes tested by gel mobility shift assays. For sequence conservation between IMPDH homologs, see Figure S4. Molecular Cell 2012 47, 133-139DOI: (10.1016/j.molcel.2012.04.030) Copyright © 2012 Elsevier Inc. Terms and Conditions