University of Dundee
Deubiquitinating enzyme amino acid profiling reveals a class of ubiquitin esterases
De Cesare, Virginia; Carbajo Lopez, Daniel; Mabbitt, Peter D.; Fletcher, Adam J.; Soetens,
Mathieu; Antico, Odetta
Published in:
Proceedings of the National Academy of Sciences of the United States of America
DOI:
10.1073/pnas.2006947118
Publication date:
2021
Licence:
CC BY
Document Version
Publisher's PDF, also known as Version of record
Link to publication in Discovery Research Portal
Citation for published version (APA):
De Cesare, V., Carbajo Lopez, D., Mabbitt, P. D., Fletcher, A. J., Soetens, M., Antico, O., Wood, N. T., & Virdee,
S. (2021). Deubiquitinating enzyme amino acid profiling reveals a class of ubiquitin esterases. Proceedings of
the National Academy of Sciences of the United States of America, 118(4), [e2006947118].
https://doi.org/10.1073/pnas.2006947118
General rights
Copyright and moral rights for the publications made accessible in Discovery Research Portal are retained by the authors and/or other
copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with
these rights.
• Users may download and print one copy of any publication from Discovery Research Portal for the purpose of private study or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain.
• You may freely distribute the URL identifying the publication in the public portal.
Take down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Download date: 09. Aug. 2022
Deubiquitinating enzyme amino acid profiling reveals
a class of ubiquitin esterases
Virginia De Cesare
a
, Daniel Carbajo Lopez
a
, Peter D. Mabbitt
a
, Adam J. Fletcher
a
, Mathieu Soetens
a
,
Odetta Antico
a
, Nicola T. Wood
a
, and Satpal Virdee
a,1
a
Medical Research Council Protein Phosphorylation and Ubiquitylation Unit, University of Dundee, DD1 5EH, Scotland, United Kingdom
Edited by Brenda A. Schulman, Max Planck Institute of Biochemistry, Martinsried, Germany, and approved December 14, 2020 (received for review April
11, 2020)
The reversibility of ubiquitination by the action of deubiquitinat-
ing enzymes (DUBs) serves as an important regulatory layer within
the ubiquitin system. Approximately 100 DUBs are encoded by the
human genome, and many have been implicated with pathologies,
including neurodegeneration and cancer. Non-lysine ubiquitina-
tion is chemically distinct, and its physiological importance is
emerging. Here, we couple chemically and chemoenzymatically syn-
thesized ubiquitinated lysine and threonine model substrates to a
mass spectrometry-based DUB assay. Using this platform, we profile
two-thirds of known catalytically active DUBs for threonine esterase
and lysine isopeptidase activity and find that most DUBs demon-
strate dual selectivity. However, with two anomalous exceptions,
the ovarian tumor domain DUB class demonstrates specific (iso)pep-
tidase activity. Strikingly, we find the Machado–Joseph disease
(MJD) class to be unappreciated non-lysine DUBs with highly specific
ubiquitin esterase activity rivaling the efficiency of the most active
isopeptidases. Esterase activity is dependent on the canonical cata-
lytic triad, but proximal hydrophobic residues appear to be general
determinants of non-lysine activity. Our findings also suggest that
ubiquitin esters have appreciable cellular stability and that non-
lysine ubiquitination is an integral component of the ubiquitin sys-
tem. Its regulatory sophistication is likely to rival that of canonical
ubiquitination.
ubiquitin
|
DUBs
|
non-lysine ubiquitination
U
biquitination impacts on almost all cellular processes and is
carried out by a multienzyme cascade involving E1 activat-
ing enzymes (E1s), E2 conjugating enzymes (E2s), and E3 ligases
(E3s) (1, 2). E3s confer substrate specificity and can broadly be
classified into two main classes. The largest class consists of
RING E3s which use an adapter-like mechanism to facilitate di-
rect transfer of ubiquitin (Ub) from thioester-linked E2 (E2∼Ub)
to substrate (3). On the other hand, engagement of E2∼Ub by
HECT-like E3s results in formation of a thioester-linked E3 in-
termediate that carries out substrate transfer autonomously (4, 5).
Conventionally, Ub is linked to the e-amino group of lysine side
chains by an isopeptide bond, or, less frequently, it can be
appended to the α-amino group of proteins via a regular peptide
bond (1). Multiple residues within Ub itself can also become
ubiquitinated, allowing the formation of Ub polymers with distinct
linkage topologies that can mediate different cellular processes
(6). Ubiquitination is a dynamic modification and is reversed by
the action of deubiquitinating enzymes (DUBs). Approximately
100 DUBs have been identified in humans and are assigned to
seven distinct classes (7). For the majority of DUBs, substrate
specificity is poorly understood, with most biochemical insights
gained thus far coming from studies toward isopeptide-linked Ub
polymers. Alterations in substrate ubiquitination are often the
molecular basis for pathology, and DUBs have become attractive
therapeutic targets (7).
Although ubiquitination is typically a lysine-specific post-
translational modification, the RING E3 MIR1 encoded by
Kaposi’s sarcoma associated-herpes virus can evade host im-
mune responses by carrying out ubiquitination of cysteine within
major histocompatibility complex class I (MHC-I) molecules
(8, 9). This promotes their endocytosis and lysosomal degradation.
It was subsequently shown that MIR1 from murine γ-herpes virus
also ubiquitinates MHC-I molecules but targets serine, threonine,
and lysine residues and promotes their degradation by endoplas-
mic reticulum-associated degradation (ERAD) (10). However, the
adapter-like mechanism demonstrated by RING E3s relies upon
the active site of the E2 to mediate transfer chemistry. This grants
E2s with the important ability to direct ubiquitination to specific
sites within a substrate (1, 11). It was subsequently demonstrated
that murine MIR1 functions with the poorly studied mammalian
E2 UBE2J2, which was shown to possess cellular serine/threonine
esterification activity (12).
In further support of the physiological importance of non-lysine
ubiquitination, HECT-like E3s also possess intrinsic esterification
activity. MYCBP2/Phr1 has important roles in neural develop-
ment and programmed axon degeneration (13) and has highly
selective threonine esterification activity (5). Furthermore, the
E3 HOIL-1 has fundamental roles in immune signaling (14) and
forms ester linkages with serine/threonine residues within Ub
polymers and protein substrates (15). However, the cellular sta-
bility of Ub esters is unknown so it is not clear whether
Significance
Ubiquitination involves the covalent attachment of the protein
ubiquitin to substrates. It can be reversed by the action of
deubiquitinating enzymes (DUBs), thereby providing an im-
portant layer of regulation. Originally believed to be restricted
to lysine residues, it is emerging that additional amino acids,
including serine, threonine and cysteine, are also modified. It
remains unknown which DUBs might target these unusual sites
for deubiquitination. Herein, we develop representative model
substrates and screen 53 DUBs for non-lysine activity, thereby
providing important insights into DUB function. Strikingly, we
find that a poorly studied DUB class has potent and highly
selective serine/threonine activity. These findings suggest that
non-lysine ubiquitination rivals the regulatory sophistication of
its conventional counterpart and might serve distinct cellular
functions.
Author contributions: V.D.C., D.C.L., P.D.M., A.J.F., M.S., O.A., N.T.W., and S.V. designed
research; V.D.C., D.C.L., P.D.M., A.J.F., M.S., O.A., and N.T.W. performed research; V.D.C.,
D.C.L., P.D.M., A.J.F., M.S., and N.T.W. contributed new reagents/analytic tools; V.D.C.,
D.C.L., P.D.M., A.J.F., O.A., N.T.W., and S.V. analyzed data; and V.D.C., A.J.F., and S.V.
wrote the paper.
The authors declare no competing interest.
This article is a PNAS Direct Submission.
This open access article is distributed under Creative Commons Attribution License 4.0
(CC BY).
See online for related content such as Commentaries.
1
To whom correspondence may be addressed. Email: s.s.virdee@dundee.ac.uk.
This article contains supporting information online at https://www.pnas.org/lookup/suppl/
doi:10.1073/pnas.2006947118/-/DCSupplemental.
Published January 21, 2021.
PNAS 2021 Vol. 118 No. 4 e2006947118 https://doi.org/10.1073/pnas.2006947118
|
1of12
BIOCHEMISTRY
Downloaded by guest on April 8, 2021
ubiquitination of hydroxy amino acids can serve as a sustained
cellular signal or if it is restricted to transient roles.
The emerging evidence that non-lysine ubiquitination has
important roles across a range of fundamental cellular process,
such as viral infection, ERAD, axon degeneration, and immune
signaling, places urgent emphasis on establishing which of the
∼100 DUBs might confer Ub esterase activity and serve as
negative regulators of this distinct form of ubiquitination. A
small panel of DUBs have been tested for activity against an
ester-linked substrate which indicated that certain DUBs do
possess esterase activity and this need not be mutually exclusive
with isopeptidase activity (16). However, comprehensive, DUB
profiling, across multiple classes, remains to be carried out.
Here, we synthesize model substrates consisting of threonine/
serine that are ester-linked to Ub. Using a high-throughput matrix-
assisted laser desorption/ionization time-of-flight (MALDI-TOF)
DUB assay (17), we profile two-thirds of known active Ub DUBs
for selectivity toward linkage chemistry (lysine isopeptide versus
threonine ester). Our findings show that the vast majority of
DUBs demonstrate isopeptidase and esterase activity with com-
parable kinetics. Isopeptidase versus esterase activity is largely
inherent to DUB class as ubiquitin-specific protease (USP) and
ubiquitin C-terminal hydrolase (UCH) DUBs displayed little
preference for linkage chemistry whereas ovarian tumor domain
(OTU) DUBs were largely dedicated isopeptidases. Two excep-
tions were TRABID and the virally encoded DUB, vOTU.
Strikingly, the Machado–Joseph disease (MJD) class demon-
strated selective threonine and serine esterase activity. We show
that esterase selectivity is maintained toward model peptide sub-
strates. Importantly, we also establish that Ub esters can have an
intracellular half-life of at least 1 h. We also demonstrate that,
in vitro, the E2 UBE2J2 possesses selective esterification activity,
as inferred by its auto modification profile, which is specifically
reversed by the MJD member JOSD1. Using chemically synthe-
sized fluorescent substrates, we quantify the catalytic efficiency of
JOSD1 and find it to be a highly efficient Ub esterase (k
cat
/K
M
=
3.5 × 10
4
M
−1
·s
−1
), an efficiency comparable to that of the most
efficient isopeptidases. While we could only quantify catalytic ef-
ficiency for threonine esterase activity, complementary assays
suggest JOSD1 serine esterase activity is considerably higher.
Taken together, our findings further support the biological sig-
nificance of non-lysine ubiquitination and demonstrate that its
regulatory sophistication is comparable to that of canonical
ubiquitination. The complementary activity profiles of certain
OTU DUBs with that of JOSD1 might also allow them to be used
as research tools for dissecting the emerging prevalence of non-
lysine substrate ubiquitination. Our ester-linked model substrates
should also facilitate the development of robust assays for inhib-
itor screening against MJD members.
Results
DUB Esterase and Isopeptidase Activity Profiling. To determine
DUB activity and specificity toward either ester or isopeptide
bonds, we employed a previously developed MALDI-TOF DUB
assay (17) and the model substrates Ub-Lysine (Ub-Lys) and Ub-
Threonine (Ub-Thr) (Fig. 1 A and B). However, as ubiquitination
site context is typically variable, lysine vs. threonine specificity
could be mediated by extended DUB–substrate interactions,
which our model substrates would not be able to determine.
Consequently, in the context of a DUBs physiological substrate,
amino acid specificity might be different to that observed with
these model substrates. For similar reasons, these substrates would
not recapitulate the diverse architectures of polyUb species and
would not be able to evaluate DUB-mediated exo- or endo-
cleavage (7). Ub-Lys was chemically prepared using a modified
implementation of genetically encod ed o rthogon al p rotection
and activated ligation (GOPAL) technology (18) (SI Appendix,
Fig. S1). Ub-Thr was chemoenzymatically prepared using a
reconstituted E1-E2-E3 cascade based on the RING-Cys-Relay
E3 machinery from MYCBP2 (5) (SI Appendix,Fig.S2). For both
amino acid substrates, the α-amino group was acetylated, which
helped mirror the peptide context the model substrates were re-
flective of, and also prevented potential O-N acyl transfer of Ub-
Thr to a peptide-linked species (19).
We screened a panel of 53 recombinant DUBs belonging to all
seven known DUB families (Fig. 2 A and B) (7): USP, OTU,
UCH, JAB1/MPN/Mov34 metalloenzyme (JAMM), MJD, motif
interacting with Ub-containing novel DUB family (MINDY),
and zinc finger with UFM1-specific peptidase domain protein
(ZUFSP). DUBs were incubated with Ub-Lys and Ub-Thr, and,
as positive controls, they were also incubated with an alternative
Ub-derived substrate (either an isopeptide-linked diUb or Ub
with a C-terminal peptide-linked adduct) known to be processed
by the DUB under investigation (SI Appendix, Fig. S3 and Table
S1). However, for the DUBs JOSD1, OTU1, OTUD6A, and
OTUD6B, a readily accessible substrate which is cleaved hitherto
remains to be i dentified. For quantification and normalization
purposes, the ratio of the area of the substrate ion intensity
signal (Ub-Lys or Ub-Thr) and the area of the p roduct signal
(Ub) were recorded and extrapolated to a standard curve,
based on defined substrate/product ratios, enabling calculation
of percent substrate cleavage (SI Appendix,Fig.S4). By com-
paring the percent of cleavage of Ub-Lys versus Ub-Thr as a
function of time, we determined activity and specifici ty toward
the two model subst rates (Fig. 2B). In practice, fractional
cleavage was sometimes found to exceed 100%. This is because
DUB, whi ch is absent when producing the stand ard curve, can
suppress the mass spectrometry (MS) signal of low abundance
Ub species (e.g., Ub-Lys or Ub-Thr after appreciable cleavage),
and data were not normalized for this phenomenon.
Esterase and Isopeptidase Selectivity Is Largely Inherent to DUB
Classification. We find that DUBs belonging to the USP and
UCH family in general cleave both Ub-Lys and Ub-Thr sub-
strates with comparable kinetics (Fig. 2B). The USPs are the
largest class, consisting of ∼60 members, whereas UCHs are a
smaller classification, consisting of only 4 members (7). The only
USPs with negligible activity toward either substrate are USP5
and USP12. However, lack of USP5 activity is anticipated as it is
only functional toward polyUb with an intact free Ub C terminus
(20). In the case of the UCH class, UCHL1 is found to only
cleave the isopeptide-linked substrate while UCHL3, UCHL5,
and BAP1 show no appreciable preference between the two
model substrates (Fig. 2B).
In contrast to USP and UCH classes, OTU family members
invariably demonstrate efficient lysine isopeptidase activity to-
ward our model substrate, but negligible threonine esterase ac-
tivity (Fig. 2B). Two notable exceptions are vOTU and TRABID.
In the context of this assay, TRABID demonstrates selective
threonine esterase activity whereas vOTU demonstrates robust
activity toward both substrates (Fig. 2B). The DUB vOTU is
encoded by the deadly human pathogen, Crimean Congo hem-
orrhagic fever virus. In addition to its ability to hydrolyze four
out of six tested isopeptide-linked Ub polymer types (21), it has
also been shown to have relaxed substrate scope as it removes
the ubiquitin-like modifier ISG15 (21). Thus, the observation
that vOTU demonstrates high isopeptidase and threonine es-
terase activity implies that relaxation of its substrate scope ex-
tends to Ub linkage chemistry and that non-lysine ubiquitination
may promote mammalian antiviral responses more broadly.
TRABID has been implicated with Wnt and immune signaling
(22, 23) and has efficient isopeptidase activity in the context of
Lys29- and Lys33-linked Ub polymers (18, 24). However, in our
assay, TRABID has negligible isopeptidase activity toward Ub-
Lys, consistent with that observed for other, albeit peptide-
linked, small molecule substrates (25). Thus, our observation
2of12
|
PNAS De Cesare et al.
https://doi.org/10.1073/pnas.2006947118 Deubiquitinating enzyme amino acid profiling reveals a class of ubiquitin esterases
Downloaded by guest on April 8, 2021
that TRABID has high activity toward Ub-Thr implies that its
esterase activity is more promiscuous than its isopeptidase ac-
tivity, and hence a significant proportion of its physiological
substrates may in fact be non-lysine ubiquitination sites.
The MJD family is a small class of DUBs consisting of four
members (26). Unlike the other DUB classifications which are
found in all eukaryotes, MJD DUBs are absent in yeast and
might be reflective of a specific demand of higher eukaryotes.
Strikingly, with the exception of ATXN3L, all MJD DUBs dem-
onstrate preferential threonine esterase activity (Fig. 2B). This is
particularly notable for JOSD1 where isopeptidase activity is
negligible but quantitative cleavage of the Ub-Thr substrate is
observed after the first time point. Similarly, Josephin-2 (JOSD2)
cleaves both substrates but has a significant preference for the Ub-
Thr substrate over the lysine counterpart. ATXN3 also cleaves
Ub-Thr more efficiently than Ub-Lys whereas ATXN3L does not
demonstrate any notable substrate preference (Fig. 2B).
We also tested the recently discovered MINDY and ZUFSP
classes of DUB. The MINDY class consists of four members,
which demonstrate exo (cleaving from the distal end) activity
toward extended Lys48 linked Ub polymers (27). ZUFSP con-
sists of a single founding member (ZUFSP/ZUP1) and is specific
for Lys63-linked Ub polymers and is involved in DNA repair
(28–31). Consistent with activity of these DUBs being dependent
on polyUb linkage context, negligible activity was observed to-
ward either of our model isopeptide or ester-linked substrates
(Fig. 2B).
Unlike the other DUB classes identified thus far which are
cysteine isopeptidases/peptidases, the JAMM class of DUBs are
metalloproteases (26). Two of the six functional JAMM class
DUBs included in our panel are AMSH and AMSH-LP, which
have both been shown to have specific activity toward isopeptide-
linked Lys63 Ub polymers with comparable efficiency. Interest-
ingly, under the enzyme concentrations employed, AMSH displays
no detectable esterase nor isopeptidase activity toward the
model substrates whereas AMSH-LP is active against both
model substrates (Fig. 2B) (32).
Cellular Stability of Ubiquitin Esters. The cellular stability of a Ub
ester would determine whether posttranslational modification
via this linkage would be restricted to transient roles or whether
it might also serve as a sustained cellular signal. Intrinsic sus-
ceptibility to hydrolysis and the action of the highly efficient and
promiscuous esterase activity found in mammalian cells might
limit its cellular function to transient roles (33). To assess the
cellular stability of a serine residue esterified with Ub, we de-
vised an experiment. Here, we prepared an engineered E2∼Ub
conjugate (E2
Ser
∼Ub), linked via a serine ester (Fig. 2C), and
delivered it into cells via electroporation (Fig. 2D) (34, 35).
Strikingly, we found that the conjugate had a cellular half-life of
1 to 2 h and could still be detected after 8 h, thereby suggesting
that ubiquitination of hydroxy amino acids can serve as a sus-
tained cellular signal.
Validation of Selective USP and OTU Isopeptidase Activity. To validate the
activity profiles determined by the MALDI-TOF assay format, we
initially prepared a fluorescent model substrate where the α-amino
group of threonine is labeled with 5/6-carboxytetramethylrhod amine
(TAMRA) (Ub-Thr-TAMRA) (Fig. 3 A and B). The fluorescent
amino acid was linked to Ub via an ester bond using the chemo-
enzymatic strategy adopted earlier. For comparison, we used com-
mercially available isopeptide-linked Ub-Lys-TAMRA-Gly (Fig. 3C).
Esterase or isopeptidase activity, respectively, would cleave the fluo-
rescent amino acid from Ub, allowing continuous and quantitative
+
MALDI-TOF Analysis
O
O
H
N
O
HN
O
OH
O
NH
O
H
N
O
HN
O
OH
O
Ub
NH
O
H
N
O
HN
O
OH
O
Ub
DUB
NH
2
HN
O
OH
O
DUB
Ub
O
O
H
N
O
HN
O
OH
O
Ub
DUB
Ub
O
H
N
O
OH
HO
HN
O
OH
O
Acetyl-lysine
Ub
O
H
N
O
OH
Ub
8565.76 Da
Acetyl-threonine
time
Ub-lysine (Ub-Lys)
8731.8 Da
Ub-Threonine (Ub-Thr)
8709.6 Da
A
B
+
DUB
ion intensity
ion intensity
ion intensity
DUB esterase activity
DUB isopeptidase activity
Ub
8565.76 Da
Ub
8565.76 Da
Ub
8565.76 Da
Ub-Threonine (Ub-Thr)
8709.6 Da
Ub-Threonine (Ub-Thr)
8709.6 Da
m/z m/z m/z
Fig. 1. MALDI-TOF–based methodology for profiling DUB isopeptidase/esterase activity. (A) DUBs are incubated either with ubiquitin-lysine (Ub-Lys) or
ubiquitin-threonine (Ub-Thr). (B) Reactions are quenched by addition of TFA (2%), spotted on an AnchorChip 1536 target and analy zed by MALDI-TOF mass
spectrometry. DUB isopeptidase or esterase activity releases free lysine or threonine and generates native Ub (8565.76 Da) in a time-dependent manner. Ion
intensity ratios for peaks corresponding to Ub-Lys (8731.8 Da)/Ub-Thr (8709.6 Da) versus Ub (8565.76 Da) are used for qualitative assessment of % cleavage.
De Cesare et al. PNAS
|
3of12
Deubiquitinating enzyme amino acid profiling reveals a class of ubiquitin esterases https://doi.org/10.1073/pnas.2006947118
BIOCHEMISTRY
Downloaded by guest on April 8, 2021
0 153045607590120
USP1
USP2
USP4
USP5
USP6
USP7
USP8
USP9x
USP10
USP11
USP12
USP15
USP16
USP20
USP21
USP25
USP27x
USP28
USP30
USP36
USP45
USP46
USP47
CYLD
Ub-Lysine
0 153045607590120
USP
USP2
USP4
USP5
USP6
USP7
USP8
USP9x
USP10
USP11
USP12
USP15
USP16
USP20
USP21
USP25
USP27x
USP28
USP30
USP36
USP45
USP46
USP47
CYLD
Ub-Threonine
OTU
0 153045607590120
OTULIN
OTU1
OTUD
1
OTUD
3
OTUD5
OTUD6A
OTUD6B
OTUB
1
OTUB
2
VCPIP1
vOTU
TRABID
A2
0
CEZANN
E
0 153045607590120
OTULIN
OTU1
OTUD1
OTUD3
OTUD5
OTUD6A
OTUD6B
OTUB1
OTUB2
VCPIP1
vOTU
TRABID
A20
CEZANNE
USP
UCHL1
UCHL3
UCHL5
BAP1
UCH
0 153045607590120
JOSD1
JOSD2
ATXN3
ATXN3-L
0 153045607590120
JOSD1
JOSD2
ATXN3
ATXN3-L
MJD
0 153045607590120
AMSH
AMSH-LP
0 153045607590120
AMSH
AMSH-LP
JAMM
0 50 100
% Cleavage
0 153045607590120
ZUFSP
0 153045607590120
MiNDY1
MiNDY2
MiNDY3
MiNDY4
MINDY
ZUFSP
0 153045607590120
ZUFSP
0 153045607590120
MiNDY1
MiNDY2
MiNDY3
MiNDY4
0 153045607590120
0 153045607590120
UCHL1
UCHL3
UCHL5
BAP1
Ub-Lysine
Ub-Threonine
profiled for esterase and
isopeptidase activity
USP
1
3
USP
4
4
USP4
8
USP22
USP21
USP31
USP4
USP
37
USP
2
8
USP
2
0
USP6
CYLD
USP43
USP
7
USP3
4
USP
2
6
USP25
USP1
6
USP9
Y
USP47
USP3
6
USP
24
USP2
USP19
USP32
USP
3
3
USP3
USP15
USP45
USP
4
9
USP
27X
USP
1
0
USP1
2
USP5
1
USP
4
1
USP
1
7
USP
4
0
USP9
X
USP
1
8
USP
1
USP
8
USP46
USP29
USP35
USP
5
USP42
USP3
8
USP1
1
USP1
4
USP3
0
USP
JOSD1
ATXN3
JOSD2
A
TXN3
L
MJD
OTU
UCH
L
1
UCHL3
B
AP
1
UCHL
5
UCH
MINDY3
MINDY2
MINDY1
MINDY4
MINDY
MY
S
M1
MPN
D
BRCC
3
S
TAM
B
PL
S
T
AMBP
PSMD14
JAMM
ZUFSP
YOD1
OTUB2
OTUD5
OTUD1
OTUD6B
OTUD7B
OTUD3
OTUD7A
ZRAN
B
1
OTUD6A
VCPIP1
OTULIN
A20
vOTU
OTUB1
ZUFSP
25 –
15 –
20 –
Electroporation:
0.5
1
8
IB: His
Time post electroporation (h):
Recombinant serine ester: + +++++
Electroporation buffer: –
+
+
2
4
+++++
–––––
0.5
1
8
+++++
2
4
+++++
+++++
IB: Tubulin
50 –
––––––––––MG-132 (20 µM):
0.5
1
8
+++++
2
4
+++++
–––––
0.5
1
8
+++++
2
4
+++++
+++++
++++++++++
CD
kDa
- E2
- E2~Ub este
r
B
A
Fig. 2. DUB esterase and isopeptidase screen by MALDI-TOF mass spectrometry and assessment of cellular ubiquitin ester stability. (A) Phylogenetic clas-
sification of deubiquitinating enzymes based on their catalytic domains. Only DUBs that are active and recognize ubiquitin are displayed. The vOTU (red)
catalytic domain exists within Protein L, which is encoded by Crimean–Congo hemorrhagic fever virus. DUBs annotated with a solid circle correspond to those
profiled for activity in this study. (B) A panel of 53 DUBs were tested for their activity toward model substrates Ub-Lys and Ub-Thr. Reactions were then
quenched by addition of TFA (2%) at the relevant time points and spotted onto a 1536 AnchorChip target plate, followed by MALDI-TOF analy sis (17). Results
are reported as percent of cleavage in a scale from white (no activity) to dark blue (100% substrate consumption). Employed DUB concentrations are specified
in SI Appendix, Table S1.(C) An engineered E2∼ Ub conjugate linked to the Ub C terminus via a serine ester (E2
Ser
∼Ub) was used as model ester-linked
substrate. Resistance to BME, but sensitivity to NaOH, is consistent with the conjugate being ester-linked. (D)E2
Ser
∼Ub was delivered into HEK293 cells by
electroporation and then transferred directly into prewarmed media and then incubated at 37 °C, 5% CO
2
. Samples were taken at the indicated time points
postelectroporation. The experiment was replicated in parallel, using HEK293 pretreated with 20 μM MG132 for 30 min, and cells incubated in media
containing MG-132 throughout.
4of12
|
PNAS De Cesare et al.
https://doi.org/10.1073/pnas.2006947118 Deubiquitinating enzyme amino acid profiling reveals a class of ubiquitin esterases
Downloaded by guest on April 8, 2021
