Human Rev1 relies on insert-2 to promote selective binding and accurate replication of stabilized G-quadruplex motifs

Abstract We previously reported that human Rev1 (hRev1) bound to a parallel-stranded G-quadruplex (G4) from the c-MYC promoter with high affinity. We have extended those results to include other G4 motifs, finding that hRev1 exhibited stronger affinity for parallel-stranded G4 than either anti-parallel or hybrid folds. Amino acids in the αE helix of insert-2 were identified as being important for G4 binding. Mutating E466 and Y470 to alanine selectively perturbed G4 binding affinity. The E466K mutant restored wild-type G4 binding properties. Using a forward mutagenesis assay, we discovered that loss of hRev1 increased G4 mutation frequency >200-fold compared to the control sequence. Base substitutions and deletions occurred around and within the G4 motif. Pyridostatin (PDS) exacerbated this effect, as the mutation frequency increased >700-fold over control and deletions upstream of the G4 site more than doubled. Mutagenic replication of G4 DNA (±PDS) was partially rescued by wild-type and E466K hRev1. The E466A or Y470A mutants failed to suppress the PDS-induced increase in G4 mutation frequency. These findings have implications for the role of insert-2, a motif conserved in vertebrates but not yeast or plants, in Rev1-mediated suppression of mutagenesis during G4 replication.


CONTENTS Supplementary materials and methods
. CD melting temperatures of the ss-DNA G4 oligos used in this study. Table S2. Equilibrium dissociation constants for hRev1  binding to ss-G4 and non-G4 DNA substrates in a buffer containing 100 mM LiCl. Table S3. Equilibrium dissociation constants for hRev1  binding to ds-G4 and non-G4 DNA substrates in a buffer containing 100 mM KCl. Table S4. hRev1 1-1251 -derived peptides (trypsin-digested) identified by mass spectrometry.

Supplementary materials and methods
Circulr dichroism spectroscopy -Circular dichroism (CD) spectroscopy for all DNA oligonucleotide substrates was performed on a Jasco J-1100 spectropolarimeter (Jasco, Easton, MD, USA). The ss-DNA G4 or non G4 oligos, or the corresponding primer-template ds-DNA substrates were prepared by annealing, as described in Methods. Solutions were prepared in 10 mM Tris-HCl, pH 7.5, containing either 100 mM KCl or LiCl.
For measuring spectra, DNA was taken at a concentration of 5 μM in a total volume of 200 μL of buffer in a quartz cuvette with pathlength of 1 mm. Spectra were measured between the wavelength range of 230 nm to 330 nm, with a band-width of 1 nm, data pitch of 0.1 nm, at a scanning speed of 100 nm/min. Spectra were reported as an average of three scans. A similar scan of the buffer alone was used as blank to subtract and obtain the corrected spectra for oligo substrates. Thermal stability for the ss-G4 oligo substrates was measured by monitoring the change in CD signal at the corresponding peak wavelength, across a temperature range of 4°C to 95°C, by applying a melting temperature gradient of 1°C/min. An average of two scans was reported. Data analysis was performed using Graphpad Prism (San Diego, CA, USA), and the melting temperature values were obtained by fitting the data to a four-parameter logistic model allowing a variable slope. For the HAP-1 REV1 KO cells transiently transfected to express the SFB-tagged hRev1  wild-type or mutant proteins, whole cell lysates were prepared 48 hrs post-transfection. REV1 KO cells that were sham-transfected (no plasmid) were used as untransfected control. Protein concentration of the lysates was estimated using the Pierce BCA assay kit (Thermo-Fisher Cat#23225). 50 μg of each lysate sample was loaded and separated by gel electrophoresis followed by transfer and blocking as       . 1150-1251). This plasmid was used to transfect HAP-1 mammalian cells for the forward mutagenesis assay. All the mutant hRev1 constructs described in the mutagenic assay were subsequently made using this construct. (B) HEK293T cells stably expressing SFB-tagged hRev1 1-1251 were generated using the construct described in A, and the overexpressed protein was purified using affinity chromatography with streptavidin-sepharose beads. Aliquots from different stages of the protein purification protocol were separated using SDS-PAGE gel electrophoresis.  plasmid sequence is shown in the top panel, with the supF-tRNA coding region marked in yellow. The numbering is according to Seidman et al (1). In the lower panel, the region marked in green indicates the Myc-G4 sequence inserted into the pSP189 plasmid. The color scheme is identical to that used in Figure 6.  gene locus. The transcription start site (marked by position '0') is indicated, as well as the region +100 bp into the coding region (blue). DNA sequence was retrieved from the Eukaryotic promoter database and the G4-forming prediction was done by analyzing the sequence using the G4-prediction tool on the QGRS web server. The three top-scoring sequences (based on G-score>20) are marked as I (G-score = 42), II (G-score = 35) and III (Gscore = 21) respectively, and the DNA sequences are shown, with the guanine bases involved in putative quadruplex formation marked in bold and underlined. Sequence 'I' was used for designing the Rev1-prom oligonucleotide used in this study. The position of each sequence is marked by the numbers on the 5¢-and 3¢-end. (B) Circular dichroism spectra of the Rev1prom sequence ( Table 1 in main text) were measured in buffer containing 100 mM KCl (blue curve) or 100 mM LiCl (red curve). The position of the signature peak at 265 nm for parallel Figure S9. Binding affinity of hRev1 330-833 to primer-template G4 DNA substrates.
Oligonucleotides forming the different types of G4 DNA folds (described in main text and Table 1) were annealed to a common 11-mer primer to obtain primer-template substrates.
The hRev1 330-833 protein was then titrated into a solution containing each of these ds-G4-DNA