Cell Cycle-Dependent TICRR/TRESLIN and MTBP Chromatin Binding Mechanisms and Patterns

Abstract

The selection of replication origins is a defining characteristic of DNA replication in eukaryotes, yet its mechanism in humans has not been well-defined. In this study, we use Cut&Run to examine genomic binding locations for TICRR/TRESLIN and MTBP, the human orthologs for the yeast DNA replication initiation factors Sld3 and Sld7. We mapped TRESLIN and MTBP binding in HCT116 colorectal cancer cells using asynchronous and G1 synchronized populations. Our data show that TRESLIN and MTBP binding patterns are more defined in a G1 synchronized population compared to asynchronously cycling cells. We also examined whether TRESLIN and MTBP are dependent on one another for binding. Our data suggest MTBP is dependent on TRESLIN for proper association with chromatin during G1 but not S phase. Finally, we asked whether TRESLIN and MTBP binding to chromatin requires licensed origins. Using cell lines with a non-degradable inducible Geminin to inhibit licensing, we show TRESLIN and MTBP binding does not require loaded MCMs. Altogether, our Cut&Run data provides evidence for a chromatin binding mechanism of TRESLIN-MTBP during G1 that is dependent on TRESLIN and does not require interactions with licensed origins.

Fig 1.

TRESLIN and MTBP co-occupy replication origins and early-replicating regions in HCT116 cells. (A). Cut&Run experiments were conducted on HCT116 cells expressing MTBP or TRESLIN genes tagged with mClover. Peaks were called with MACS2 and were designated highly reproducible if found in at least 6 Cut&Run experiments. Euler plot illustrates the extensive overlap of highly reproducible TRESLIN and MTBP peaks. (B) Euler plot shows that MTBP/TRESLIN peaks in HCT116 cells significant overlap peaks from published MTBP Cut&Run data, in which MTBP in DLD1 cells was endogenously FLAG-tagged(19). (C) Heat plots display MTBP Cut&Run signal (Log2 fold-change over background) across high-confidence peaks. (D) As in C, heat plots show TRESLIN signal across high-confidence peaks. (E) Genome track set for human chromosome 1 shows that the istribution of MTBP and TRESLIN binding sites is associated with early replicating domains and DNA replication origins. The following are displayed: Log2Fold Change signal for MTBP (light blue) and TRESLIN (dark blue), along with locations of peaks, DNA replication origins (Rep. ORIs), and high-resolution Repli-Seq data(24, 50, 51). (F) Left highlighted region in E shows a replication timing peak with abundant TRESLIN/MTBP signal. (G) Central region in E depicts late replicating areas with decreased TRESLIN/MTBP signal. (H) Right region in E. A portion of the late-replicating region that replicates earlier and has increase TRESLIN/MTBP binding is highlighted. (I) Euler plot illustrates the overlap of highly reproducible TRESLIN/MTBP peaks with Ini-Seq and Sns-Seq origins(50, 51). (J) Counts of TRESLIN/MTBP peaks in 50 kb genome segments across 16 replication timing fractions, ranging from early to late replication. (K) Mean Cut&Run Log2FC signal over background for MTBP (K) or TRESLIN (L) plotted for each replication timing fraction reveals that MTBP and TRESLIN binding correlates with replication timing genome-wide. (M) Analysis of TRESLIN/MTBP peak locations relative to gene features reveals frequent presence in gene promoters, especially in peaks also found in DLD1 cells(19).

Fig 2.

TRESLIN and MTBP enriched in genomic peaks during the G1 phase of the cell cycle. (A) Schematic of the experimental design comparing TRESLIN and MTBP localization in freely cycling cells and G1-phase cells. Freely cycling cells underwent natural cell cycle progression, while G1 cells were collected post-Nocodazole-induced mitotic arrest. (B) DNA content histograms from a single replicate, illustrating cell cycle distribution in cycling and G1-enriched cells. Histograms highlight cell cycle phases (color-shaded) as estimated by the Dean-Jett-Fox algorithm, with parental HCT-116 cells serving as background controls. (C) Flow cytometry-based quantification of cell cycle phase percentages in two independent replicates. (D) Heat plot of background-normalized Cut&Run signals (Log2FC) across MACS2-identified peaks in at least two samples. (E) Aggregate plot of Log2-fold changes in background-normalized MTBP or TRESLIN signals in consensus peaks relative to HCT-116 cells across three replicates. Lines represent mean values, with ribbons indicating the range for these replicates. (F) Volcano plot of DESeq2 analysis showing FDR versus changes in MTBP and TRESLIN signals at consensus peaks, comparing G1 to cycling cells. Signals were predominantly higher in G1 cells. (G) Background-normalized signals across three representative ranges, indicating reduced peaks in cycling compared to G1 cells. Data for TRESLIN and MTBP are shown separately for each of the three replicates.

Fig 3.

Binding of MTBP to chromatin requires TRESLIN in G1 but not in S. (A) Protein levels of MTBP-mClover and TRESLIN-mClover in HCT116 cells were measured via flow cytometry using an anti-GFP antibody, alongside DNA content. This analysis was conducted 24 hours after siRNA transfection, representing a single experimental replicate. (B) Quantification of the data in panel A involved background-subtracting individual cell anti-GFP immunofluorescence signals, averaging signals for cells in 2N, early S, or 4N DNA content gates, and normalizing values from siTRESLIN or siMTBP transfected cells to siCNTRL. The means of these normalized values across three replicates are shown with error bars indicating 95% confidence intervals. Asterisks indicate significance levels from t-tests comparing each sample to its respective siCNTRL. (C) and (D) replicate (A) and (B), respectively, with the exception that soluble protein was pre-extracted before fixation. Consequently, immunofluorescence signals represent insoluble TRESLIN-mClover or MTBP-mClover protein.

Fig 4.

Licensing is not required for TRESLIN and MTBP binding to genomic loci and early replicating regions. A) Diagram of cell synchronization strategy and induction of Geminin transgene expression to inhibit loading of MCMs. B) Insoluble flow cytometry showing level of chromatin bound MCM7 and TRESLIN-mClover or MTBP-mClover in tagged (TRESLIN or MTBP) or untagged (HCT-116) cells. Anti-MCM7, anti-GFP, or anti-IgG Immunofluorescence signal vs DNA content is plotted to show the level of each protein across the cell cycle. Treatment with doxycycline for 16 hours was sufficient to reduce levels of chromatin bound MCM7 in 2N cells, but it did not affect TRESLIN or MTBP binding. C) Heat plots showing Cut&Run signal over peaks called with MACS2 with or without doxycycline treatment over reproducible TRESLIN or MTBP peaks. D) Aggregate plots of Log2fold signal over HCT116 parental untagged background. Plots show two replicates for each. E) Log2fold signal of Cut&Run for TRESLIN and MTBP plotted against replication timing from high resolution Repli-Seq data(24) shows that licensing inhibition does not affect TRESLIN/MTBP enrichment in early replicating regions F) Example genomic range of peaks present in samples treated with or without doxycycline. Scale = 50kb.