CELL CYCLE-REGULATION OF THE FANCD2 PROTEIN VIA CDK-MEDIATED S592 PHOSPHORYLATION

Fanconi anemia (FA) is a rare genetic disease characterized by increased risk for bone marrow failure and cancer. The FA proteins function together to repair damaged DNA. A central step in the activation of the FA pathway is the monoubiquitination of the FANCD2 and FANCI proteins under conditions of cellular stress and during Sphase of the cell cycle. The regulatory mechanisms governing S-phase monoubiquitination, in particular, are poorly understood. In this study, we have identified a CDK regulatory phospho-site (S592) proximal to the site of FANCD2 monoubiquitination. FANCD2 S592 phosphorylation during S-phase was detected by LC-MS/MS and by immunoblotting with a S592 phospho-specific antibody. Mutation of S592 disrupts S-phase and DNA damage-inducible monoubiquitination. In addition, FA-D2 (FANCD2-/-) patient cells expressing S592 mutants display reduced proliferation under conditions of replication stress and increased mitotic aberrations, including nucleoplasmic bridges and multinucleated cells. Our findings describe a novel cell cycle-specific regulatory mechanism for the FANCD2 protein.

Monoubiquitination occurs following exposure to DNA damaging agents and during S-phase of the cell cycle (I Taniguchi et al. 2002).
Recently, it was discovered that the monoubiquitination of FANCD2 and FANCI leads to the formation of a closed ID2 clamp that encircles DNA (Alcón et al. 2020; R. Wang et al. 2019). Moreover, ubiquitinated ID2 (ID2-Ub) assembles into nucleoprotein filament arrays on double-stranded DNA (Tan et al. 2020). The monoubiquitination of FANCD2 is necessary for efficient interstrand crosslink repair, the maintenance of common fragile site stability and faithful chromosome segregation, events that are all crucial for genomic stability (Meetei et al. 2003;Nakanishi et al. 2005;Vinciguerra et al. 2010).
A DNA damage-independent role for the FA pathway during the cell cycle has been implicated in several studies. For example, FANCD2 promotes replication fork protection during S-phase and ensures the timely and faithful replication of common chromosome fragile sites (Howlett et al. 2005;Madireddy et al. 2016). FANCD2 has also been shown to be involved in mitotic DNA synthesis (MiDAS) during prophase and is present on the terminals of anaphase ultrafine bridges (Chan et al. 2009;Garribba et al. 2018;Vinciguerra et al. 2010). During the cell cycle FANCD2 monoubiquitination is maximal during S-phase and minimal during M-phase (Taniguchi et al. 2002). FANCD2 and FANCI are phosphorylated by the ATR and ATM kinases following exposure to DNA damaging agents, promoting their monoubiquitination (Chen et al. 2015;G. P. H. Ho et al. 2006). However, in general, the function and regulation of FANCD2 and FANCI during the cell cycle remains poorly understood.
Cyclin-dependent kinases (CDKs) play a major role in regulating cell cycle progression, with CDK-mediated hyperphosphorylation of the retinoblastoma protein (pRb) being the primary mechanism of cell cycle regulation (Adams et al. 1999;Ezhevsky et al. 2001

FANCD2 is phosphorylated under unperturbed conditions and during S-phase of the cell cycle
To study the phosphorylation of the FANCD2 protein in the absence and presence of DNA damaging agents, we performed a lambda-phosphatase assay with several cell lines following incubation in the absence or presence of the DNA crosslinking agent mitomycin C (MMC). Notably, we observed a large increase in FANCD2 mobility following incubation of lysates with lambda-phosphatase even in the absence of MMC in all cell lines examined (Fig. 1A). These results suggest that FANCD2 is subject to extensive phosphorylation even in the absence of an exogenous DNA damaging agent.
A similar change in protein mobility was not observed for FANCI (Fig. 1A). To determine if FANCD2 is subject to phosphorylation during the cell cycle, we performed a double thymidine block experiment causing an early S-phase arrest and analyzed phosphorylation at regular time points following release using the lambdaphosphatase assay. We observed maximal phosphorylation of FANCD2 during Sphase of the cell cycle, with much less phosphorylation observed as cells progressed through G2/M-and G1-phases of the cell cycle (Fig. 1B). Again, we observed no appreciable change in FANCI mobility upon lambda-phosphatase treatment, suggesting that FANCI is not subject to the same level of phosphorylation as FANCD2 during the cell cycle (Fig. 1B). Similar findings were observed with U2OS cells ( Fig. S1A and B). We also performed a M-phase arrest using nocodazole and again observed maximal levels of FANCD2 phosphorylation during S-phase (~15 h after release) of the cell cycle ( Fig. S1C

FANCD2 is phosphorylated on S592 during S-phase of the cell cycle
To map the in vivo sites of FANCD2 phosphorylation, we immunoprecipitated FANCD2 from asynchronous U2OS cells stably expressing 3xFLAG-FANCD2 under stringent conditions. Immunoprecipitated FANCD2 bands were combined and subjected to phosphoproteomic analysis using LC-MS/MS ( Fig. 3A and B). Under these non-DNA damaging conditions, we observed the phosphorylation of multiple sites including the previously detected ATM/ATR phosphorylation sites S1401 and S1404 (Table 1) (Taniguchi et al. 2002). We also detected phosphorylation of the putative CDK site S592 (Table 1) Table 1). In cells expressing FANCD2-WT, we observed a robust increase in CDC2 pY15 levels following MMC exposure. However, this increase was attenuated in the absence of FANCD2 and in cells expressing the FANCD2 S592 variants (Fig. 4B), suggestive of a defect in the mitotic checkpoint or mitotic progression.

Mutation of S592 disrupts
To analyze the effects of S592 mutation on S-phase FANCD2 monoubiquitination (Taniguchi et al. 2002), cells were subject to a double thymidine block, and FANCD2 monoubiquitination was analyzed upon release. Compared to FANCD2-WT, S-phase monoubiquitination was markedly attenuated for both S592 variants (Fig. 4C). We also analyzed levels of the mitotic marker H3 pS10 in these cells. Compared to cells expressing FANCD2-WT, we observed persistent levels of  Specifically, FANCL binding to UBE2T exposes a basic triad of the UBE2T active site, promoting favorable interactions with a conserved acidic patch proximal to K561, the site of ubiquitination (Chaugule et al. 2020). We speculate that S592 phosphorylation may also augment interaction with the basic active site of UBE2T.

DISCUSSION
Alternatively, S592 phosphorylation may inhibit FANCD2 de-ubiquitination by USP1, in a manner similar to that previously reported for the FANCI S/TQ cluster (Cheung et al. 2017).
Our studies further emphasize the critical nature of coordinated posttranslational modification of FANCD2. Several studies have previously established intricate dependent and independent relationships between FANCD2 monoubiquitination and phosphorylation. For example, the ATM kinase phosphorylates FANCD2 on several S/TQ motifs following exposure to ionizing radiation, e.g. S222, S1401, S1404, and S1418 -phosphorylation of S1401 and S1404 were also detected under unperturbed conditions in this study. While phosphorylation of S222 promotes the establishment of the IR-inducible S-phase checkpoint, S222 phosphorylation and K561 monoubiquitination appear to function as independent events (Taniguchi et al. 2002

Immunoblotting
For immunoblotting analysis, cell pellets were washed in PBS and lysed in 2% w/v SDS, 50 mM Tris-HCl, 10 mM EDTA followed by sonication for 10 s at 10% amplitude. Proteins were resolved on NuPage 3-8% w/v Tris-Acetate or 4-12% w/v Bis-Tris gels (Invitrogen) and transferred to polyvinylidene difluoride (PVDF) membranes. The following antibodies were used: rabbit polyclonal antisera against

Plasmids
The S592A

Generation of Lentiviral
Particles 5 x 10 6 293FT cells were seeded in 10 cm 2 dishes in DMEM + 12% v/v FBS + 1% v/v L-glutamine + 1% v/v sodium pyruvate +2% v/v non-essential amino acids without antibiotics. 5mL of a mix of Opti MEM and Fugene 6 was added dropwise to the cells and then cells were incubated. A mixture of 9 µg of Virapower packaging mix and 3 µg pLenti-6.2-vector in OptiMEM was added to the cells and incubated for 6-8 h. The supernatant was collected and filtered through a 45 µm filter.

In-vitro CDK Phosphorylation Assay
Purified FANCD2, pRb and CDK2/Cyclin A proteins were a generous gift from Andrew Deans at the University of Melbourne. In order to remove any previous phosphorylation 2 µg of protein were incubated with 100 U of lambda phosphatase, 10 mM protein metallophosphatases (PMP) and 10 mM MnCl2 at 30 o C for 30 min.
Lambda phosphatase was then removed from the reaction using 30K columns. Protein

Micronucleus Assay
For micronuclei analysis, cells were seeded at a density of 2 x 10 4 in chamber slides

Cell Proliferation Assay
For cell proliferation assays we used electrical impedance with the xCELLigence RTCA DP system from Acea Biosciences. Cells were seeded at a density of 5.0 x 10 3 in polyethylene terephthalate (PET) E-plates (300600890, Acea Biosciences). Cells were treated with 20 nM MMC or 0.4 µM APH. Electric impedance measurements were taken every 15 min for 120 h. We tested whether electric impedance measurements differed across timepoints by using the base package R-function t.test to perform Student's t-test. All statistical analyses were performed in R version 3.6.1 (R Core team 2015).

G2/M Accumulation Assay
Cells were seeded at a density of 4.0   S8 P S126 P S1401 S1404 S1407 S1412 S1435 S592 Nucleoplasmic bridges, micronuclei, nuclear buds and multinucleated cells were scored. We tested whether frequency of mitotic defects differed across groups by using the base package R-function t.test to perform Student's t-test. All statistical analyses were performed in R version 3.6.1 (R Core team 2015). (A-D).

Figure S6. Mutation in S592 causes increased mitotic aberrations during nonstressed conditions.
FA-D2 cells were incubated without the presence of DNA damaging agents. Then cells were fixed in ice cold methanol and stained with DAPI. Binucleated, micronuclei, nuclear buds and multinucleated cells were scored. We tested whether frequency of mitotic defects (A-Binucleated, B-Nuclear Buds, C-Micronuclei, D-Multinucleated) differed across groups by using the base package R-function t.test to perform Student's t-test. All statistical analyses were performed in R version 3.6.1 (R Core team 2015).