Nature - USA (2020-01-23)

580 | Nature | Vol 577 | 23 January 2020

Article

replication timing was reprogrammed to ensure that the whole genome
was replicated efficiently.
We next investigated whether the R257 and K333 residues of
SUV420H1 are involved in activating early origins in vivo. We found
that neither SUV420H1R257A nor SUV420H1K333A could rescue H4K20me2
after knocking down endogenous SUV420H1 (Extended Data Fig. 7c).
In addition, through real-time PCR we found that these two SUV420H1
mutants could not rescue nascent-strand or BrdU signals (labelled for
10 minutes immediately after release from G1/S arrest) (Extended Data
Fig. 7d). Thus, H2A.Z regulates the selection and activation of early-
replication origins through SUV420H1.

H2A.Z in activated T cells

To study the function of H2A.Z-regulated replication in a more physi-
ological context, we conditionally knocked out (CKO) H2az1/H2az2
in T cells by generating CD4CreH2A.Zf/f mice (Extended Data Fig. 8a).
We found that the number of mature T cells in the spleen decreased
dramatically in H2A.Z CKO mice (Extended Data Fig. 8b, c). H2A.Z
CKO T cells showed reduced BrdU incorporation in the homeostatic
state, and reduced dilution of carboxyfluorescein succinimidylester

(CFSE) in activated T cells on stimulation with CD3 and CD28 antibod-

ies (Extended Data Fig. 8d–f ). In addition, fluorescence-activated cell

sorting (FACS) analysis showed that activated T cells from CKO mice

had a prolonged G1 phase and a shorter S phase than the wild type

(Extended Data Fig. 8g, h). These results suggest an essential role of

H2A.Z in DNA replication and T-cell proliferation.

ChIP-seq analysis showed that the H4K20me2 and nascent-strand

signals were markedly reduced in H2A.Z CKO activated T  cells

(Extended Data Fig. 8i). Western blot analysis confirmed the global

loss of H4K20me2 upon H2A.Z depletion in activated T cells (Extended

Data Fig. 8j). We classified nascent-strand signals into ZD-NS (H2A.Z-

dependent-nascent strand; n = 18,382) and ZI-NS (H2A.Z-independent-

nascent strand; n = 7,901). We found that H4K20me2, as well as H2A.Z

and nascent strands, in ZD-NS regions were higher than those in ZI-NS

regions (Extended Data Fig. 8k), indicating that, as with HeLa cells, the

H2A.Z–SUV420H1–H4K20me2–ORC1 axis has an important regulatory

role in the selection and firing of replication origins in activated T cells.

Discussion

We have shown that nucleosomes comprising H2A.Z histones can

directly bind SUV420H1 to efficiently stimulate the dimethylation of

H4 K20 residues, thereby licensing and activating early-replication

origins (Fig. 3f). However, given that H2A.Z nucleosomes are much

less abundant than canonical H2A nucleosomes in cells, it remains

unknown how H2A.Z regulates the global abundance of H4K20me2

in vivo. It has been shown that H4K20me2 levels are passively diluted

twofold during DNA replication, recovering gradually by the next G1

phase^22. Notably, H2A.Z is lost on nascent chromatin after DNA replica-

tion, and is restored along with H4K20me2 (ref.^22 ), suggesting that it

has an essential role in establishing H4K20me2 on newly synthesized

histones. Recently, a class of cis-regulatory elements called early-rep-

licating control elements was found to be essential for maintaining the

timing of early replication^23. Another study showed that poly(dA:dT)

tracts were associated with efficient replication origins^20. Interest-

ingly, nucleosomes were depleted at the centres of these poly(dA:dT)

tracts, but strongly positioned at flanking regions^20 , akin to the features

of chromatin structure seen at replication origins in yeast^24. We have

found here that active early origins are highly enriched with G/C-rich

and asymmetric A/T-rich motifs. Thus, we speculate that coordination

between genetic determinants and epigenetic features is involved in

fine-tuning the licensing and activation of replication origins during

the cell cycle.

Online content

Any methods, additional references, Nature Research reporting sum-

maries, source data, extended data, supplementary information,

acknowledgements, peer review information; details of author con-

tributions and competing interests; and statements of data and code

availability are available at https://doi.org/10.1038/s41586-019-1877-9.

1. Fragkos, M., Ganier, O., Coulombe, P. & Méchali, M. DNA replication origin activation in
   space and time. Nat. Rev. Mol. Cell Biol. 16 , 360–374 (2015).


2. MacAlpine, D. M. & Almouzni, G. Chromatin and DNA replication. Cold Spring Harb.
   Perspect. Biol. 5 , a010207 (2013).


3. O’Donnell, M., Langston, L. & Stillman, B. Principles and concepts of DNA replication in
   bacteria, archaea, and eukarya. Cold Spring Harb. Perspect. Biol. 5 , a010108 (2013).


4. Dhar, M. K., Sehgal, S. & Kaul, S. Structure, replication efficiency and fragility of yeast ARS
   elements. Res. Microbiol. 163 , 243–253 (2012).


5. Kuo, A. J. et al. The BAH domain of ORC1 links H4K20me2 to DNA replication licensing
   and Meier-Gorlin syndrome. Nature 484 , 115–119 (2012).


6. Pesavento, J. J., Bullock, C. R., LeDuc, R. D., Mizzen, C. A. & Kelleher, N. L. Combinatorial
   modification of human histone H4 quantitated by two-dimensional liquid
   chromatography coupled with top down mass spectrometry. J. Biol. Chem. 283 ,
   14927–14937 (2008).


7. Cayrou, C. et al. The chromatin environment shapes DNA replication origin organization
   and defines origin classes. Genome Res. 25 , 1873–1885 (2015).


8. Costas, C. et al. Genome-wide mapping of Arabidopsis thaliana origins of DNA replication
   and their associated epigenetic marks. Nat. Struct. Mol. Biol. 18 , 395–400 (2011).

0

0.2

0.4

0.6 Plus-ZD-ORC 1

Minus-ZD-ORC 1

ZI-ORC1

Early Late

Percentage

of ORC

sites

a

Early

Late

Replication timing

siNC siH2A.ZsiNCsiH2A.Z

4.9 × 10 –323

1.3 × 10 –23

d

siNC 10 min

siH2A.Z 10 min

siNC 1 h

siH2A.Z 1 h

siNC 3 h

siH2A.Z 3 h

siNC RT

siH2A.Z RT

siNC 6 h

siH2A.Z 6 h

BrdU

[0–100]

[0–100]

[0–65]

[0–65]

[0–50]

[0–50]

[–2.5–5]

[–2.5–5]

[0–50]

[0–50]

500 kb

MEX3B RPS17 C15orf4 0

Chromosome15:

e 79,685,601–81,514, 775

c Target 1

(early origin)

Target 2

(late origin)

siNC siH2A.Z

Input (%) Input (%)

Input (%) Input (%)

Input (%) Input (%)

Input (%) Input (%)

1.2

0.8

0.4

0

0.6

0.4

0.2

0

3

2

1

0 0

4

3

2

1

0.5

0.4

0.3

0.1

0.2

0

15

10

5

0 0

0.8

0.6

0.4

0.2

5

4

3

12

0

H2A.Z H2A.Z

H4K20me2 H4K20me2

ORC1 ORC1

Nascent strandsNascent strands

P<0.0001 P<0.0001

P=0.0014 P=0.0080

P=0.0014 P=0.0398

P=0.0009 P=0.0212

RPKM

b

Plus-ZD-ORCMinus-ZD-ORC1MPliunsu-s-ZDZD-O-ORCR (^11) C 1
0
ZI-ORC1
1.2 × 10 –187
<2.2 × 10 –16
<2.2 × 10 –16
2
4
6
8
10
10 min11 h3 h6 h10 min h3 h6 h
siNC siH2A.Z
Fig. 4 | H2A.Z regulates early replication origins and replication timing.
a, Graph showing the replication timing of plus-ZD-ORC1 (n = 20,978), minus-
ZD-ORC1 (n = 52,655) and ZI-ORC1 (n = 27,284) peaks. b, Box plot showing the
BrdU signal in siNC or siH2A.Z cells. c, Real-time PCR analysis of the ChIP signal
from H2A.Z, H4K20me2 and ORC1, and nascent-strand signals in siNC or
siH2A.Z cells. d, Box plot showing the dynamics of replication timing of the
2,000 earliest and 300 latest replication origins (nascent-strand peaks).
e, Genome tracks of a relatively late replication domain (red shaded area) show
the increased BrdU signal at 10 min and advanced replication timing after
H2AFZ knockdown. Numbers in square brackets indicate the data range of
corresponding track. MEX3B, C15orf40 and RPS17 are RefSeq gene names.
Data in c are mean ± s.e.m.; n = 3 biological replicates; two-tailed unpaired
t-test. Data in b, d were analysed by two-tailed Wilcoxon test. P-values are
indicated within b–d. The experiment in e was independently repeated twice
with similar results.