Science - USA (2022-04-29)

checkpoint. We confirmed premature mitotic
entry observed in the siRNA-based screen using
individual siRNAs targeting CAD and ICAD
and in the CAD-deficient KO cells (Fig. 4, A
to C). This revealed that the breakdown in G 2

checkpoint control in CAD-deficient cells was

most pronounced 24 hours after IR, which

corresponded with our observed peak of CAD-

inflicted DNA breaks (Fig. 4, B and C). To

further characterize the molecular basis of

the CAD/ICAD-dependent checkpoint reg-

ulation, we noted reduction of the inhibitory

phosphorylation of cyclin-dependent kinase

CDK1 (Tyr^15 ), reduction in active checkpoint

kinase CHK2, and moderate difference in

SCIENCEscience.org 29 APRIL 2022¥VOL 376 ISSUE 6592 481

Fig. 4. CAD/ICAD are required for tumor cell

survival and checkpoint maintenance after

IR.(A) Mitotic (pS10-H3) staining of irradiated

and nocodazole (NZ)–trapped U2OS cells

transfected with control siRNA (UNC) and siRNA

against CAD. Cells were irradiated with 6 Gy

of IR; after a 2-hour recovery, NZ was added for

8 hours before cells were fixed for immuno-

fluorescence. Scale bar, 50mm. (B)G 2

checkpoint maintenance in HCT116 CAD siRNA

knockdown cells upon 8 Gy of IR. Data are

means ± SEM;N=3.*P< 0.05 (unpaired

Student’sttest). (C)G 2 checkpoint mainte-

nance in HCT116 CAD KO cells 24 hours after

8 Gy of IR. Data are means ± SEM;N= 3.

*P< 0.05 (unpaired Student’sttest). (D)G 2

checkpoint maintenance in U2OS ICAD KO cells

24 hours after 6 Gy of IR. Data are means ±

SEM;N=3.*P< 0.05 (unpaired Student’s

ttest). (E) Relative colony outgrowth of

CAD-depleted cells after IR exposures.

Data are means ± SEM;N=4.*P< 0.05

(unpaired Student’sttest). (F) Relative colony

outgrowth of CAD KO HCT116 cells upon

indicated IR exposure. Data are means ± SEM;

N= 4. *P< 0.05 (unpaired Student’sttest).

(G) Genomic instability in U2OS WT and ICAD

KO cells upon 8 Gy of IR. % Genomic instability

represents total number of cells displaying

micronuclei and fragmented nuclei divided by

the total number of cells. Data are means ±

SEM;N= 3,n> 400. **P= 0.0016 (ANOVA).

(H) Immunoblotting of pY701 STAT1 3 days after

6 Gy of IR in WT and ICAD KO cells. CDK1

inhibitor RO-3306 was added 2 hours after IR.

(I) Normalized tumor growth of HCT116 WT and

HCT116 CAD KO tumors after 4 Gy of IR. Data

are means ± SEM;n=6.**P= 0.0055 (two-way

multiple-comparisons ANOVA). (J) Model of

CAD-dependent G 2 phase checkpoint.

IR (8Gy):8h 24h IR (8Gy)24h

0123

0.01

0.1

1

IR (Gy) IR (Gy)

Relative Colony Outgrowth

UNC

siCAD

0123

0.01

0.1

1

Relative Colony Outgrowth

WT

CAD KO

0

5

10

15

20

25

0

5

10

15

Relative Mitotic Entry (%)

0246810

0

2

4

6

Days post IR (4Gy)

Relative tumor growth

WT

CAD KO

UNC siCAD

pS10-H3

pS10-H3

DAPI

AB

I

CD

EG

H

F

Relative Mitotic Entry (%)

0

2

4

6

Relative Mitotic Entry (%)

0.0

0.1

0.2

0.3

0.4

Genomic Instability %

IR (8Gy):

pY701 STAT1

S TAT 1

Vinculin

WT KO WT KO WT KO

NT IR (6Gy)

IR (6Gy)

+CDKi

CAD Proficient Cancer Cell

Repair

Premature

Mitosis

G2 Checkpoint

CAD Deficient Cancer Cell

Endogenous Breaks

No Damage IR Breaks

Proficiency

IR (8Gy)

IR (8Gy)24h

Genomic

Instability

ICAD: WT KOWT KOWT KOWT KO

NT 1d 2d 3d

UNCsiCADUNCsiCAD

WT

CAD KO

WT

ICAD KO 101ICAD KO 102

J

Lack of Endogenous Breaks

RESEARCH | RESEARCH ARTICLES