Nature - USA (2020-01-02)

Nature | Vol 577 | 2 January 2020 | 129

that had abundances ranging between 30 and 75%, including several
isoforms of HSP90 and six HSP70 isoforms (Fig. 2a; see Supplementary
Information for details).

NMR spectroscopy in cells
Next, we carried out NMR experiments in cells to study the interaction
between α-synuclein and chaperones inside living mammalian cells
at atomic resolution. [U-^15 N]-α-Synuclein was delivered into HEK293
cells at concentrations of 3–10 μM, yielding intensity patterns that are

characteristic for mammalian cell lines^9 (Fig. 2b, c), such as the canoni-

cal chaperone-interaction signature. Multiple molecular chaperones

are present in the cell that have mutually overlapping functions and

‘clientomes’^20. To complement the in vitro chaperone analyses, we

investigated two of the most abundant chaperones found in mammalian

cells, HSC70 and HSP90β. When [U-^15 N]-α-synuclein was delivered into

HEK293 cells with reduced HSC70 levels (Extended Data Fig. 7c, d), the

NMR intensity profile resembled the one observed for untreated cells,

suggesting that there is functional redundancy between HSC70 and

other chaperones in these cells (Fig. 2d, e). Next, we treated HEK293

IHEK

/I buffer

I/I

HEK

I/IHEK

I/IHEK

I/I

HEK

1.2

0.4

0.8

0.8

1.2

0.4

1.5

2.0

1.0

0.5

20 40 60 80 100 120 140

α-Synuclein residue number

α-Synuclein residue number

HSC70-depleted HEK293 cells

HEK293 cells + 24 h HSP90 inhibitors

HSC70-depleted HEK293 cells + 24 h HSP90 inhibitors

HSP90

`

HSP90

α

HSP60

HSP90 family HSP70 family HSP60 family 14-3-3 family

CCT

b

CCT

e

CCT

ε

CCT

β

CCT

ζ

CCT

γ

CCT

α

HSP105 14-3-3

γ

14-3-3

ζ/δ

14-3-3

ε

Cyclophilin A

14-3-3

θ

14-3-3

β/α

HSP70

1A

HSP70

4

HSC70 GRP-78GRP-75

Endoplasmin

d

a

ΔN-

α-Synuclein/α-synuclein

0

0.4

0.8

0

0.2

0.4

0.6

0.8

1.0

G86 G31

G93

T92

T81

D121

E130

E110

E83

K96 L38

A89

L113

A27

A53

N103

Q109

V95

V15V48

V77

V118

D115K12Y39

Y136

V37

A107

δ 2 (^1 H) (ppm)

δ 2 (^1 H) (ppm)

8.50 8.25 8.00

110

115

120

125

130

δ^1

(^15) (
N) (ppm)
δ^1
(^15) (
N) (ppm)
HEK293 cells
fHSC70-depleted HEK293 cells + 24 h HSP90 inhibitors
8.50 8.25 8.00
HEK293 cells
e HSC70-depleted HEK293 cells
0
0.8
1.2
0.4
G51 G41
G31
M5
D2
D121
A124
A17
E137
A140
E126
V40
D119
S129 S9
N122
E130H50
AcM1
K6
K21 F4
D135 L8
V3
8.50 8.25 8.00
In vitro
HEK293 cells
b
c
20 40 60 80 100 120 140
110
115
120
125
130
HSC70-depleted HEK293 cells + 4 h HSP90 inhibitors
Fig. 2 | The interaction between α-synuclein and chaperones is dominant in
living cells. a, Abundance ratios of proteins bound to ΔN-α-synuclein versus
wild-type full-length α-synuclein determined by relative quantitative mass
spectrometry (mean values, n = 2). b, Overlay of two-dimensional [^15 N, ^1 H]-NMR
spectra of [U-^15 N]-α-synuclein in NMR buffer (black) and inside living HEK293
cells (blue-green). Representative spectrum from n > 5. c, Residue-resolved
backbone amide NMR signal attenuation (IHEK/Ibuffer) of α-synuclein in
mammalian cells. d, NMR signal attenuation in treated cells, relative to
untreated cells (I/IHEK). Different combinations of HSC70 depletion and HSP90
inhibition were applied, as indicated. e, f, Overlay of two-dimensional [^15 N, ^1 H]-
NMR spectra of [U-^15 N]-α-synuclein in untreated HEK293 cells (black) and in
HSC70-depleted HEK293 cells (green) (e) or in HSC70-depleted HEK293 cells
after 24 h of HSP90 inhibition (green) (f). Representative data (d–f) for three
technical replicates, with similar results.