Fig 1.
Schematic representation of the positions of genes and predicted 36 GQs associated with the putative promoter or regulatory regions in the HCMV genome.
Among 38 GQs (GQ1~GQ38), which were initially annotated between -500 and +100 with respect to the translation initiation sites of genes (S1 Table), 36 GQs that were analyzed in this study are presented in the diagram. The x-axis represents the genomic scale or base positions. Two strands of HCMV DNA are separately shown as parallel lines. Approximate locations and strand positions of GQs are shown with open circles. Location and orientation of the associated genes are indicated as black bar pointers between the two strands. The types of G4s are differentiated by different font colors: type-I (conventional), blue; type-II (long-loop), black; and type-III (bulged), red. The unique long (UL) and unique short (US) regions of the HCMV genome and the terminal and internal repeat regions (open boxes) are indicated at the top of the figure.
Fig 2.
CD spectra of HCMV G4 oligos in the absence and presence of G4 stabilizers TMPyP4 and NMM.
Comparison of CD spectra of 36 G4-oligonucleotides in the presence of G4 ligands, TMPyP4 and NMM. Fifteen μM DNA oligos were annealed in the presence of 10 mM Tris-HCl [pH 7.5] and 100 mM KCl buffer with or without 30 μM TMPyP4 and NMM (DNA to chemical ratio 1:2). Each spectrum was an average of 3 accumulations in the wavelength range between 230–320 nm. The spectra were blanked with buffer only. Note that the sequences of GQ28 and GQ37 were identical, as GQ29 and GQ36.
Fig 3.
The CD spectra of the representative HCMV G4 oligonucleotides.
(A to D) CD spectra of GQ2 (A), GQ18 (B), GQ12 (C), and GQ33 (D) represent the parallel, antiparallel, mixed, and weak G4 sequences, respectively. A parallel G4 (GQ2) showed a peak at 260 nm and a trough at 240 nm (A), whereas an antiparallel G4 (GQ18) was characterized by a peak at 290 nm and trough at 260 nm (B). GQ12 showed a mixed structure with a shoulder around the 290 nm region in addition to the parallel G4 peaks (C). In a weak G4 (GQ33), a broad shoulder was observed around the 230–280 nm region (D). (E and F) CD spectra of the control oligonucleotides. CD spectra of a G4-forming oligonucleotide from the promoter region of C-MYC (CMYC22) (as a positive control) (E) and a single-stranded 24mer-poly(T) [Poly(T)] (as a negative control) (F) are shown.
Fig 4.
CD melting temperature (Tm) graph of HCMV G4 oligonucleotides in the absence and presence of G4 stabilizer TMPyP4 and NMM.
(A) Comparison of CD Tm spectra of HCMV G4-forming oligonucleotides in the presence of G4 ligands, NMM and TMPyP4. Fifteen μM DNA oligos were annealed in the presence of 10 mM Tris-HCl [pH 7.5] and 100 mM KCl buffer with or without 30 μM TMPyP4 and NMM (DNA to chemical ratio 1:2). CD melting graph was calculated at 290 nm wavelength for GQ18 and at 262 nm for rest of the G4s. The data were normalized using the maximum ellipticity and smoothed using 12-point Savitzky-Golay algorithm. Note that the data of 27 GQs, which showed melting curves within the temperature range indicated, are shown. (B) Comparison between mean CD melting temperatures of each type of G4s in the absence or presence of TMPyP4 and NMM.
Table 1.
CD melting temperature (Tm) of HCMV G4-containing oligos in the absence and presence of G4 stabilizers TMPyP4 and NMM (DNA to chemical ratio 1:2).
Fig 5.
Effect of G4-binding ligands on the activity of HCMV promoters.
(A) Schematic representation of the experimental design. HF cells were transfected via electroporation with luciferase reporter plasmid containing the viral G4-containing promoter region for 24 h. To measure the activity of promoter regions from immediate-early genes, transfected cells were infected with HCMV (Towne) at a multiplicity of infection (MOI) of 2 with treatment of DMSO (as a control) or 5 μM of NMM or TMPyP2 for 24 h. To measure the promoter activity of E and L genes, transfected cells were infected with HCMV for 8 h (for early genes) or 24 h (for late genes), and treated with DMSO or 5 μM of NMM/TMPyP2 for 24 h prior to cell harvest and luciferase assay. (B) Expression profiles for luciferase reporter in response to G4 ligands. The repression folds of luciferase activity by NMM or TMPyP4 treatment versus DMSO treatment obtained from triplicate samples are shown. The promoter regions of the UL112 and UL99 genes without any apparent G4 were used as controls. P values calculated between UL37 and IRS1/TRS1 samples, between the UL112 control gene and early genes, or between the UL99 control gene and late genes are indicated.
Fig 6.
Mutational analysis of GQ8 and GQ18 for G4 stability and reporter gene suppression.
(A and B) Biophysical characterization of G4 mutant oligonucleotides for GQ8 and GQ18. CD spectroscopy (A) and melting analysis (B) of GQ8 and GQ18 oligos (wild-type and G4-disrupting mutant sequences) associated with UL35 and UL75/UL76 genes, respectively, are shown. (C to E) Luciferase reporter assays demonstrating the effects of GQ8 and GQ18 mutations on promoter activities. HF cells were transfected via electroporation with reporter plasmids that expressed luciferase from the promoter regions of UL35, UL75, and UL76 containing wild-type or mutant G4 sequences. At 24 h after transfection, cells were infected with HCMV (Towne) with treatment of DMSO (as a control) or 5 μM NMM or TMPyP2 for 24 h prior to luciferase assays, which were performed at 32 h (for UL35) or 48 h (for UL75 and UL76) after virus infection. The resulting luminescence values are plotted as luciferase units.
Fig 7.
Effects of G4-binding ligands on viral gene expression, DNA replication, and production of progeny virions.
(A) Relative mRNA levels in response to G4-binding ligands. HF cells were infected with HCMV (Toledo) at an MOI of 1, and the mRNA levels were measured by qRT-PCR at 24 h (for immediate-early genes), 32 h (for early genes) or 48 h (for late genes). To examine the effect of G4-binding ligands, cells were treated with 5 μM of NMM and TMPyP2 for 24 h prior to cell harvest. The relative mRNA levels normalized by those of β-actin are shown. (B and C) Comparison of viral DNA levels and viral titers in response to G4 ligands. HF cells were infected with HCMV (Toledo) at an MOI of 1 for 72 h. Cells and supernatants were separately collected. The viral DNA levels from the cells (intracellular) and the supernatants (extracellular) were determined by qPCR. The relative amounts of viral DNA levels normalized by those of β-actin are shown (B). The cell-associated and cell-free virus titers were also determined using infectious center assays (C).
Fig 8.
Analysis of the GQ8 mutant virus.
(A) Schematic representation of the regulatory region of the UL35 gene. Positions of the GQ8 (solid box) and putative TATA box (dashed box) for UL35 are indicated. The predicted polyadenylation signal for UL34 is underlined. Mutations introduced into GQ8 to disrupt G4 formation are indicated in red. The parts of UL34 and UL35 ORFs are indicated with amino acid sequences. (B) Graphical representation of various mutants and revertant viruses generated for the study of GQ8. The HCMV (Toledo) bacmid containing mutant GQ8 (mGQ8) and its revertant were produced using a counter-selection BAC modification kit (Gene Bridges) (see Materials and Methods). (C) Gel profiles showing restriction patterns of HCMV bacmid constructs. Wild-type (Wt), GQ8 mutant (mGQ8), and revertant (R) bacmids were digested with BamH1 and subjected to pulse-field gel electrophoresis. No apparent alteration of restriction fragment patterns was found in mGQ8 and revertant bacmids. (D) Effect of G4 ligands on UL35 transcription in wild-type, GQ8 mutant, and revertant viruses. HF cells were infected with recombinant HCMV [wild-type (Wt), GQ8 mutant (mGQ8), or revertant (R)] at an MOI of 1 and treated with DMSO, 5 μM of NMM, or TMPyP2 for 24 h prior to cell harvest at 32 h after HCMV infection. The mRNA levels of UL112 and UL35 were measured by qRT-PCR. The relative mRNA levels normalized by those of β-actin and IE1 are shown as graphs.