FIGURE SUMMARY
Title

SPRTN protease and checkpoint kinase 1 cross-activation loop safeguards DNA replication

Authors
Halder, S., Torrecilla, I., Burkhalter, M.D., Popović, M., Fielden, J., Vaz, B., Oehler, J., Pilger, D., Lessel, D., Wiseman, K., Singh, A.N., Vendrell, I., Fischer, R., Philipp, M., Ramadan, K.
Source
Full text @ Nat. Commun.

SPRTN depleted cells endure severe DNA replication stress but fail to activate a CHK1 response. a Schematic representation of DNA fiber assay. See also Methods. b DNA fibers obtained from HEK293 cells that have been treated with the indicated siRNAs against SPRTN (siSPRTN) or with hydroxyurea (HU). Scale bar: 10 µm. Data shown are representative images of three independent experiments. ce DNA fiber assay analysis showing replication fork length (c), stalled replication forks (d) and newly fired origins (e). c siSPRTN cells exhibit decreased fork velocity. >100 individual IdU tracts were measured per experiment per condition. Data are shown as mean with 25–75% percentile range (box) and 10–90% percentile (whiskers); n = 3 independent experiments, two-tailed Student's t-test. d siSPRTN cells exhibit an increased frequency of stalled replication forks. >400 forks were scored per condition per experiment. Mean ± SEM; n = 3 independent experiments, two-tailed Student's t-test. e siSPRTN cells exhibit increased newly firing of dormant origin, contrary to cells treated with HU. >400 forks were scored per condition per experiment. Mean ± SEM; n = 3 independent experiments, two-tailed Student's t-test. f Knock-down of SPRTN in HEK293 cells diminishes the phosphorylation status of CHK1 residues S296 and S345, and enhances Cdc25A stability, contrary to treatment with HU. Immunoblots were run with whole cell lysates and represent three independent experiments. g Quantifications for (f) of CHK1 phosphorylation signal at residues S345 and S296, normalised to vinculin. Mean ± SEM; n = 3 independent experiments, two-tailed Student's t-test. hj Treatment of cells with the CHK1 inhibitor UCN-01 induces severe replication stress to a similar extent as SPRTN-inactivation, detected by DNA fiber assay analysis. Schematic representation of experimental layout showing the addition of UCN-01 with IdU is also shown in (h, top); d: day. k Cell growth of wt and SPRTN-deficient (ΔSPRTN) HeLa cells in response to the CHK1 inhibitor UCN-01 (300 nM) on day 4 after seeding cells at same density (day 0). Mean ± SEM; n = 3 independent experiments, two-tailed Student's t-test. Source data for (ck) are provided as a Source Data file

Ectopic CHK1 overexpression rescues SPRTN-deficiency phenotypes. a Experimental strategy followed to assess the effect of different CHK1 variants on DNA replication by DNA fiber assay (d: day, EV: empty vector). b Rescue of DNA replication defects in SPRTN-depleted HEK293 cells with ectopic expression of CHK1-wt, but not with the phospho-deficient CHK1 variants. Graphs show quantification data from DNA fibers for replication fork velocity (left panel; mean ± 25–75 percentile range (box) ± 10–90 percentile range (whiskers)), newly fired origins (centre panel; mean ± SEM) and stalled replication forks (right panel; mean ± SEM); >100 DNA fibers were analysed per condition and experiment; n = 3 independent replicates, two-tailed Student's t-test. c Overexpression of CHK1-wt diminishes the number of chromosomal aberrations caused by SPRTN knock-down visualised by metaphase spreads from HEK293 cells. >30 cells were scored per condition per experiments; mean ± SEM; n = 3 individual experiments, two-tailed Student's t-test. d Representative images of zebrafish embryos non-injected or injected with either a control or a previously characterised morpholino (MO) against SPRTN at 9–10 h post fertilization (hpf). To assess CHK1 function, capped RNA encoding different mutants of GFP-tagged human CHK1 were co-injected with the SPRTN MO. Upper row: live images. Middle row: pictures of γH2AX to show DNA damage. Lower row: Pictures of GFP to show expression of CHK1 protein variants. e Representative images for normal, retarded and dead zebrafish embryos at 24 hpf. Scale bar = 500 µm. fh Overexpression of CHK1-wt, but not of the phospho-deficient CHK1 variants, rescues developmental defects and DNA damage induced by SPRTN depletion in zebrafish embryos. Percentage of zebrafish embryos that show developmental retardation relative to living embryo are represented in (f) (mean ± SEM; >70 embryo were scored in n = 4 independent experiments, two-tailed Student's t-test), whereas data scoring normal, retarded and dead zebrafish embryos are represented in (h), Fisher’s exact test. γH2AX, marker of DNA damage, is represented in (g) (mean ± SEM; at least 30 GFP-positive embryos were counted per condition; n = 2 replicates, two-tailed Student's t-test). Source data for (b, c, fh) are provided as a Source Data file

PHENOTYPE:
Fish:
Knockdown Reagent:
Observed In:
Stage Range: 90%-epiboly to Prim-5

SPRTN does not contribute to the activation of CHK1 after DSB formation. a Phosphorylation status of the ATR-CHK1 and ATM-CHK2 pathways in control (siNS) and siRNA SPRTN-depleted HEK293 cells under unchallenged condition or when cells were treated with hydroxyurea (HU). Whole cell extracts were used for the immunoblots. b DNA fiber assay analysis comparing HeLa-wt (wild type, parental) and HeLa-ΔSPRTN cells: quantification of replication fork velocity (top; mean ± 25–75 percentile range (box) ± 10–90 percentile range (whiskers)), newly fired origins (bottom left; mean ± SEM) and stalled replication forks (bottom right; mean ± SEM). >100 DNA fibers were analysed per condition and experiment; n = 3 experimental replicates, two-tailed Student's t-test. c Analysis of the phosphorylation status of the ATR-CHK1 and ATM-CHK2 pathways comparing HeLa -wt and -ΔSPRTN under unchallenged or HU-treated conditions. Whole cell extracts were used for the immunoblots. SE: short exposure; LE: long exposure. d, e SPRTN-deficient (siSPRTN) or wt cells (siNS) treated with formaldehyde (FA; 50 μM, 1 h) do not exhibit robust single-stranded DNA foci in S-phase cells (CldU positive). HU (500 μM, 1 h) was used as a positive control to generate replication stress-induced ssDNA formation. Camptothecin (CPT, 1 μM, 1 h followed by 1 h recovery) was used as a positive control for ssDNA formation induced by DSBs. Data are shown as representative immunofluorescent microscopy images of BrdU foci (ssDNA) in S-phase (CldU positive) HeLa cells (d), and as quantification of cells with more than 15 BrdU foci (e). Scale bar = 10 µm. >70 individual CldU positive cells were scored per condition per experiment. Mean ± SEM; n = 3 experimental replicates, two-tailed Student's t-test. f FA treatment and SPRTN-depletion fail to induce phosphorylation of CHK1 at Ser345 due to the absence of ssDNA formation. HU or CPT were used at the same conditions as in (d) as positive controls for ssDNA-induced CHK1 and CHK2 activation. Immunoblots of HeLa cells whole cell extracts shown here represent three experimental replicates. Source data for (ac, e, f) are provided as a Source Data file

SPRTN protease evicts CHK1 from replicative chromatin. a Diagram of cellular fractionation into cytosolic, nuclear and chromatin fractions. The chromatin fraction was thoroughly washed with 1% NP-40 and 250 mM NaCl to isolate clean chromatin. b The cleanliness of the chromatin fraction was confirmed by fractionation protein markers. c, d SPRTN deficiency leads to accumulation of tightly bound CHK1 on chromatin during the S-phase progression. HeLa-wt and HeLa-ΔSPRTN cells were arrested at the G1/S boundary by a double thymidine block and then released to progress through S-phase. c Proteins tightly bound to DNA were isolated by a modified RADAR assay and the total content of CHK1 tested by immunoblotting. Double-stranded DNA (dsDNA) was used as loading control. Cyclin E and cyclin B from whole cell extracts (WCE) were used as cell cycle markers. d Quantification of CHK1 in (c). Mean ± SEM; n = 2, two-tailed Student's t-test. e SPRTN and CHK1 interact physically in vitro. Purified CHK1 co-precipitated with recombinant His6-tagged SPRTN on Ni-NTA beads. Immunoblots represent three replicates. f SPRTN and CHK1 interact physically in vivo. CHK1 co-precipitated with SPRTN from lysates of Flp-In HEK293 T-REx cells expressing either SPRTN-wt or SPRTN-Y117C fused to a cSSH (Strep-strep-HA) tag. Immunoblots represent three replicates. EV: empty vector, PD: pull down. g, h iPOND analysis revealed the presence of SPRTN and CHK1 proteins at nascent DNA (0 min) but not on mature chromatin (5–12 min after a thymidine chase). Cells were treated with EdU for 10 min to label nascent DNA, and then chased or not with thymidine. PCNA and histone H3 were used as replication fork and loading controls, respectively, n = 3. i HeLa-ΔSPRTN cells exhibit increased retention of CHK1 on mature DNA compared with SPRTN-proficient (wt) cells. j, k SPRTN protease is necessary for the efficient eviction of CHK1 from replicative chromatin. Expression of SPRTN-wt, but not of the protease inactive variant SPRTN-E112A, in SPRTN knock-down cells reduced the amount of CHK1 on replicative chromatin and reversed the accumulation of CHK1 in chased chromatin. Mean ± SEM; n = 3, two-tailed Student's t-test. Source data for (bf, hk) are provided as a Source Data file

SPRTN cleaves CHK1 and releases kinase-active CHK1 fragments. a Purified Flag-CHK1 was incubated with recombinant SPRTN-wt or SPRTN-E112A (a protease-dead variant). Anti-Flag immunoblotting detected N-terminal CHK1 fragments released by the SPRTN proteolytic activity, referred to here as CHK1 Cleavage Products (CPs) -1, -2 and -3 and indicated by blue arrows. Representative immunoblots from three replicates. b Graphical representation of the three CPs released by SPRTN activity. The approximate size of these CPs was estimated from the immunoblots; CHK1 amino acids 1–338, 1–293 and 1–237, respectively. Red ticks represent the locations of Ser296, Ser317 and Ser345. Inh: autoinhibitory domain. c, d Endogenous CHK1 from T24 cells synchronised in G0- or S-phase (c) or from HEK293 cells ectopically expressing SPRTN-wt or SPRTN-E112A (d) was immuno-purified under denaturing conditions using a CHK1 antibody against an N-terminal CHK1 epitope. SE: short exposure, LE: long exposure. Quantification of the CHK1 fragments for (d). Mean ± SEM; n = 3, two-tailed Student's t-test. e Lysates from HEK293 cells ectopically expressing Flag-CHK1-wt or Flag-CHK1-S317A were denatured and CHK1 was then immuno-purified. Representative immunoblots from three replicates. f Ectopic expression of CHK1-CPs or CHK1-FL (full-length) in SPRTN-inactivated HEK293 cells. Left: mean ± 25–75 percentile range (box) ± 10–90 percentile range (whiskers); centre and right: mean ± SEM. >100 DNA fibers were analysed per condition and experiment; n = 3 experiments, two-tailed Student's t-test. g Ectopic expression of CHK1-CP3 in zebrafish embryos. Ref: reference value for statistics. Mean ± SEM; n = 3, two-tailed Student's t-test. h Graphical description of experimental setting. In the first reaction, GST-CHK1 was mixed with SPRTN-E112A or SPRTN-wt to generate CHK1 fragments (o/n, 37 °C); in the subsequent reaction, the product of the first reaction was mixed with Cdc25A in a kinase buffer to induce its phosphorylation (30 min, 37 °C). i Top panel: cleavage of CHK1 by SPRTN-wt after the first reaction. Bottom panel: phosphorylation of Cdc25A by the products of the first reaction. j quantification of phospho-Cdc25A in (i). Experiment was repeated 3 times with similar results. Source data for (a, cg, i, j) are provided as a Source Data file

PHENOTYPE:
Fish:
Knockdown Reagent:
Observed In:
Stage Range: 90%-epiboly to Prim-5

CHK1 phosphorylates SPRTN and regulates its recruitment to chromatin. a, b siRNA SPRTN-depleted HEK293 or haploinsufficient SPRTN HeLa cells (ΔSPRTN) still contain a residual amount of SPRTN (10, 30%, respectively). Data are representative immunoblots (a) and quantification of SPRTN normalised to vinculin (b). Mean ± SEM; n = 3, two-tailed Student's t-test. c, d CHK1 promotes SPRTN recruitment on chromatin. HeLa-wt or -ΔSPRTN cells ectopically expressing CHK1-wt, or treated with formaldehyde (FA) as a positive control, were fractionated and endogenous SPRTN protein level was assessed in the chromatin fraction. Residual SPRTN in ΔSPRTN cells were also recruited to chromatin. Representative immunoblots (c) and quantification of SPRTN on chromatin (d). Mean ± SEM; n = 3, two-tailed Student's t-test. e, f Overexpression of CHK1-full length (FL) or N-terminal products (CP1, 2 or 3) promote SPRTN retention on chromatin. Doxycycline-inducible stable cells expressing SPRTN-wt-cSSH cells were fractionated and SPRTN was assessed in chromatin fractions. Data are representative immunoblots (e) and quantification of HA (SPRTN) on chromatin (f). Mean ± SEM; n = 3, two-tailed Student's t-test. g Schematics to identify the CHK1 phosphorylation sites on SPRTN by Mass-Spectrometry. h Diagram depicting human SPRTN with the CHK1 phosphorylation sites on Ser 373, 374 and 383 identified in this study. SPRTN domain and motifs; DBS: DNA binding site; SHP box: p97 interacting region; PIP box: PCNA interacting peptide; UBZ: ubiquitin binding domain. Sequence below (from SwissProt: Q9H040) shows the location of these sites highlighted in red and amino acids surrounding the region. i, j The degree of phosphorylation at a CHK1 consensus target sequence is diminished in the phospho-deficient SPRTN variants. Different SPRTN variants were expressed in HEK293 cells and purified under denaturing conditions. An antibody recognizing the epitope Rxxp (S/T) (phospho-Ser/Thr preceded by Arg at position -3) was used to visualise CHK1 phosphorylated targets. The CHK1 inhibitor UCN-01 was used as a control. Data are representative immunoblots (i) and quantification of CHK1-mediated phosphorylation of SPRTN normalised to total SPRTN protein level in denatured sample (j). Mean ± SEM; n = 3, two-tailed Student's t-test. Source data for (af, i, j) are provided as a Source Data file

CHK1 targeted phosphorylation of SPRTN is essential for its biological function. ac DNA replication analysis based on DNA fibers showing that SPRTN phospho-defective mutants (S to A) are functionally impaired and unable to rescue DNA replication fork velocity (a), suppress new origin firing (b) and decrease fork stalling (c) in SPRTN-depleted HEK293 cells, whereas SPRTN phospho-mimetic mutants (S to E) are functional and restore normal DNA replication phenotype. Left: mean ± 25–75 percentile range (box) ± 10–90 percentile range (whiskers); right: mean ± SEM. >100 DNA fibers were analysed per condition and experiment; n = 3 experiments, two-tailed Student's t-test. d SPRTN phospho-defective mutants failed to promote the eviction of CHK1 from chromatin. Chromatin fractions from HEK293 cells were immunoblotted to show the effect of SPRTN variants on CHK1 release from chromatin. Data are representative immunoblots (top) and quantification of CHK1 on chromatin fraction normalised to histone H2B (bottom). Mean ± SEM; n = 3 experimental replicates, two-tailed Student's t-test. e, f SPRTN phospho-defective mutants were unable to rescue the zebrafish embryo developmental retardation (e) and DNA damage (f, γH2AX positive signal) induced by SPRTN deficiency. SPRTN-depleted embryo cells (SPRTN MO) were co-injected with plasmids for the expression of SPRTN variants. >70 embryos were scored per experiment. Mean ± SEM; n = 3 independent experiments, two-tailed Student's t-test. Source data for (af) are provided as a Source Data file

PHENOTYPE:
Fish:
Knockdown Reagent:
Observed In:
Stage Range: 90%-epiboly to Prim-5

CHK1 and SPRTN promote replication fork progression by removing DPCs. a Ectopic expression of CHK1-wt or of SPRTN-wt promotes replication fork progression over FA-induced DPCs in HEK293 cells. Upper panel shows the strategy for DNA fiber assay. Lower graph plots the quantification of IdU tract length. >100 DNA fibers were analysed per condition and experiment. Mean ± 25–75 percentile range (box) ± 10–90 percentile range (whiskers), n = 3 experiments, two-tailed Student's t-test. b, c Overexpression of CHK1-wt, but not of CHK1-S345A, reduces the amount of DPCs in HeLa-ΔSPRTN cells. DNA protein crosslink extracts were visualised by silver staining (b). ds-DNA blot is shown as a loading control. The immunoblots for Flag, SPRTN and actin were prepared from whole cell lysates. Data are representative immunoblots (b) and quantification of CHK1 on chromatin fraction normalised to histone H2B (c). Mean ± SEM; n = 3 experimental replicates, two-tailed Student's t-test. d Model for the SPRTN-CHK1 cross-activation loop. During steady-state DNA replication, SPRTN evicts CHK1 from replicative chromatin and stimulates CHK1 activity. In turn, CHK1 phosphorylates SPRTN on C-terminal serines. CHK1 activity enhances the recruitment of SPRTN to chromatin and its activity towards DPC proteolysis. This steady-state SPRTN-CHK1 cross-activation loop enables normal physiological replication fork velocity, inhibits unscheduled replication origins firing, stabilises replication forks and removes DPCs to safeguard genome stability. Source data for (ac) are provided as a Source Data file

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ Nat. Commun.