Nature - USA (2020-01-16)

428 | Nature | Vol 577 | 16 January 2020

Article

Our map does not allow us to assign amino acids in FACT that directly
participate in subnucleosome binding. To validate our structures,
we used hydrogen–deuterium exchange (HDX) coupled to mass spec-
trometry, comparing deuterium uptake in FACT alone and in FACT
bound to subnucleosomes (Extended Data Figs. 4, 5). On the basis of
volcano plot analysis^16 , we find a significant decrease in deuterium
uptake (with a ΔHDX of more than 0.3 Da; P < 0.01) in regions of the
dimerization and middle domains of SPT16 and SSRP1. These regions
map primarily to the inner face of the fork that interfaces directly with

DNA and histones (Fig. 2b–d). A decrease in deuterium is also observed

at the interface between the SSRP1 middle domain and both dimeriza-

tion domains (the hinge region) (Fig. 2b), suggesting decreased domain

flexibility upon subnucleosome binding. These solution-based results

are entirely consistent with the structures, and almost no changes in

deuterium uptake are observed in regions outside those present in

the structure.

Despite the extensive DNA interface, phosphorylated FACT purified

from insect cells is unable to bind free DNA (Fig. 2e). We hypothesized

a

Histone

contact

90°

90°

DNA binding hub

DNA

contact

Fork tips

Complex 1

DNA contact

Complex

5% native PAGE; Atto-647 label

SPT1 6

SSRP 1

–FL ΔCFL ΔC

–FL FLΔC ΔC

Free DNA

(30 bp)

1 2 3 4 5

c

d

e

f

b

0.001 0.01 0.1 110 100 1,00010,000

100

120

140

160

180

200

Fluor

escence polarization (mP)

ΔCΔC

FACT (nM)

560–572

901–915

864–870

807–815

619–620

622–636

105–108

135–139

180–186

201–215

256–272

343–345

377–390

406–407

SPT16 SSRP1

Hinge

region

–1

10 –^2

10 –^3

10 –^4

10 –5

–0.5 0.0 0.5 1.0 1.5

ΔHDX (Da)

P value

486–500

18–27

560–567

622–639

628–639623–639

SPT16 DD

622–636

560–568

606–620567–573

11–27

1,016–1,030

901–916

807–815807–816

902–916

858–870

901–915859–870

857–870

909–917909–916

SPT16 MD

FACT versus FACT + subnucleosome

10 –^1

10 –^2

10 –^3

10 –^4

10 –^5

–0.5 0.0 0.5 1.0 1.5

ΔHDX (Da)

P value

SSRP1 DD

FACT versus FACT + subnucleosome

180–186

180–188

130–139

98–118104–114

131–138

128–139

130–138130–136

105–114

203–214

343–352344–352257–272

208–222

256–272255–272

SSRP1 MD

202–214

377–391

393–407380–392404–414

201–207377–392

132–138

180–186

Fig. 2 | FACT makes extensive contacts with tetrasome DNA. a, Surface
potential of the FACT–subnucleosome complex. Colouring changes from red
for −5 kT e−1 to blue for +5 kT e−1, generated with APBS Tools 2.1 (PyMOL 2.3.2).
Histones are omitted from the top centre and right panels. b, Regions of
significant (P < 0.01) changes in average HDX (ΔHDX, from c, d) are mapped
onto complex 1. Notable regions are coloured according to domain location
(Fig. 1a). Regions with no detectable change are in white; regions for which no
peptides were recovered are in grey. The grey arrow denotes protected regions
at the intersubunit interface. The subnucleosome is coloured wheat.
Representative uptake plots are listed in Extended Data Fig. 5. c, Volcano plot
comparing the average ΔHDX of SPT16 peptides in FACT and in FACT bound to
subnucleosomes at all time points. Each time point was collected in triplicate
(n = 3); the Welch’s t-test was one-sided. d, Volcano plot comparing average
ΔHDX of SSRP1 peptides in FACT and in FACT bound to subnucleosomes at all

time points. Each time point was collected in triplicate (n = 3) and the Welch’s t-

test was one-sided. For c, d, dotted lines show significance cut-offs of Δ average

HDX > 0.3 Da and P < 0.01 from a Welch’s t-test. Insignificant peptides are in

light grey, and significant peptides are coloured according to their domain

location and listed. e, DNA binding by FACT CTD deletions. We incubated

400 nM FACT or deletion constructs with 100 nM Atto-647-labelled 30-bp DNA,

and analysed the results by native PAGE, visualized by DNA f luorescence. This

was repeated twice with similar results, and multiple times with DNA of

different lengths. ΔC, ΔCTD; FL, full-length. f, Fluorescence polarization assay

of FACT (0–2, 500 nM) with 20 nM Alexa-488-labelled tetrasome. Dissociation

constants (Kd) (in nM): full-length FACT (magenta) 204.2 ± 19 (R^2  = 0.9723);

SPT16∆C–SSRP1 (blue) 184.8 ± 14 (R^2  = 0.9807); SPT16–SSRP1∆C (orange)

201.0 ± 50.6 (R^2  = 0.8266); and ∆C∆C (black) 28.15 ± 1.9 (R^2  = 0.9843). Kd and

standard deviations were derived from four biological replicates.