JBC Papers in Press. Published on December 19, 2017 as Manuscript M117.

795955
The latest version is at http://www.jbc.org/cgi/doi/10.1074/jbc.M117.795955
HDAC6 deacetylates ERK1

Histone deacetylase 6 (HDAC6) deacetylates extracellular signal-regulated kinase 1 (ERK1) and

thereby stimulates ERK1 activity *

Jheng-Yu Wua,b, Shengyan Xiangb, Mu Zhanga, Bin Fangc, He Huangd, Oh Kwang Kwond,
Yingming Zhaod, Zhe Yange, Wenlong Baib, Gerold Beplera, Xiaohong Mary Zhanga,1

From the aDepartment of Oncology, Molecular Therapeutics Program, Karmanos Cancer Institute, 4100
John R. St., Detroit, Michigan 48201, bDepartment of Pathology and Cell Biology, Morsani College of
Medicine, University of South Florida, 12901 Bruce B. Downs Blvd. Tampa, FL 33612, cThe Proteomics
Core, H. Lee Moffitt Cancer Center and Research Institute, 12902 USF Magnolia Dr. Tampa, FL 33612,
d
Ben May Department of Cancer Research, The University of Chicago, 929 East 57th St. W410, Chicago,
Illinois 60637, eDepartment of Microbiology, Immunology & Biochemistry, Wayne State University
School of Medicine, 540E Canfield Avenue, Detroit, MI 48201. Department of Microbiology,

*Running Title: HDAC6 deacetylates ERK1

1
To whom correspondence should be addressed: Xiaohong Mary Zhang, Karmanos Cancer Institute, 4100

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

John R. St., Detroit, MI 48201, US, Tel.: (313)576-8672; Fax: (313)576-8928; E-mail:
[email protected].

Key words: extracellular signal-regulated kinase1/2 (ERK1/2); histone deacetylases (HDACs);

acetylation; deacetylation; mitogen-activated protein kinases (MAPKs)
ABSTRACT adjacent to Lys-71, which binds to ATP,
Histone deacetylase 6 (HDAC6), a class suggesting that acetylation status of Lys-72
IIb HDAC, plays an important role in many may affect ERK1 ATP binding.
biological and pathological processes. Interestingly, an acetylation-mimicking ERK1
Previously, we found that ERK1, a mutant (K72Q) exhibited less phosphorylation
downstream kinase in the MAPK signaling than the WT enzyme and a deacetylation-
pathway, phosphorylates HDAC6, thereby mimicking mutant (K72R). Of note, the K72Q
increasing HDAC6-mediated deacetylation of mutant displayed decreased enzymatic activity
α-tubulin. However, whether HDAC6 in an in vitro kinase assay and in a cellular
reciprocally modulates ERK1 activity is luciferase assay compared with the WT and
unknown. Here, we report that both ERK1 K72R mutant. Taken together, our findings
and 2 are acetylated and that HDAC6 suggest that HDAC6 stimulates ERK1 activity.
promotes ERK1 activity via deacetylation. Along with our previous report that ERK1
Briefly, we found that both ERK1 and 2 promotes HDAC6 activity, we propose that
physically interact with HDAC6. Endogenous HDAC6 and ERK1 may form a positive feed-
ERK1/2 acetylation levels increased upon forward loop, which might play a role in
treatment with a pan-HDAC inhibitor, an cancer.
HDAC6-specific inhibitor or depletion of
HDAC6, suggesting that HDAC6 deacetylates Histone deacetylases (HDACs) and histone
ERK1/2. We also noted that the acetyltransferases (HATs) are the enzymes that
acetyltransferases CBP and p300 both can regulate core histones and non-histone proteins
acetylate ERK1/2. Acetylated ERK1 exhibits by deacetylation and acetylation, respectively (1).
reduced enzymatic activity toward the HATs acetylate proteins via adding acetyl groups
transcription factor ELK1, a well-known to lysine residues, while HDACs catalyze a
ERK1 substrate. Furthermore, mass reverse reaction by removing the acetyl group
spectrometry analysis indicated Lys-72 as an from lysine residues. Up to now, a total of 18
acetylation site in the ERK1 N-terminus, HDACs are identified in humans and grouped

Copyright 2017 by The American Society for Biochemistry and Molecular Biology, Inc.
 HDAC6 deacetylates ERK1

into four classes based on their sequence MAPK kinase kinase (MAP3K), a MAPK kinase
similarity to yeast orthologs (2). Class I HDACs (MAP2K), and a MAPK. These MAPK pathways
are homologous to yeast reduced potassium participate in transducing signals from the surface
dependency 3 (Rpd3) and include HDACs 1, 2, 3 to the interior of the cell. Being triggered by
and 8. Class II HDACs are homologous to yeast extracellular stimulus, first tier MAP3Ks are
histone deacetylase 1 (HdaI) and are further activated to phosphorylate MAP2Ks, which
divided into class IIa and class IIb. Class IIa subsequently phosphorylate MAPKs. These
contains HDACs 4, 5, 7 and 9, and class IIb MAPK pathways have their own unique primary
includes HDACs 6 and 10. HDAC11 is the only kinases in different tiers, but they also share some
member in class IV. The deacetylase activity of minor activators (10,11). All MAPKs, except
HDAC classes I, II and IV is zinc-dependent. ERK3/4 and NLK, contain a conserved Thr-X-
Class III HDACs, also knowns as sirtuins, are Tyr motif in their kinase domain.
homologous to yeast silent information regulator Phosphorylation of both Thr and Tyr residues in
2 (Sir2), and the deacetylase activity of this class this motif is a critical step for MAPK activation
is oxidized nicotinamide adenine dinucleotide (10).
(NAD+)-dependent. Human ERK1 (also known as MAPK3 or
HDAC6 belongs to class IIb HDACs. Its p44MAPK) and ERK2 (also known as MAPK1

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

structure is quite unique among all HDACs, in or p42MAPK) are 84% identical in sequence;
that it contains two functional deacetylase they share many functions (12). Thus they are
domains in tandem and a zinc finger domain in often referred to as ERK1/2. Among these
the C-terminus (2,3). HDAC6 participates in MAPKs, ERK1/2 are associated with cell
numerous biological and pathologic processes, proliferation, cell growth, cell mobility, and cell
such as cell migration, DNA damage response survival (13), and the Ras-Raf (MAP3K)-MEK
and oncogenesis, through modulating its (MAP2K)-ERK1/2 (MAPK) signal transduction
substrates (4-7). For example, HDAC6 cascade can be activated by growth factors,
deacetylates cytoskeleton proteins and their osmotic stress, and cytokines (11,14,15). To date,
associated proteins, such as -tubulin and more than 160 substrates of ERK1/2 have been
cortactin, to regulate cell mobility (4,5). HDAC6 discovered from the nucleus and cytosol to the
also deacetylates and ubiquitinates the DNA cell membrane (16).
mismatch repair protein MSH2, in order to Post-translational modifications (PTMs)
regulate MutS homeostasis, DNA mismatch have long been documented as a critical means of
repair, and DNA damage response (6). In regulating ERK1/2 activity. Compared with the
addition, HDAC6 deacetylates cell signaling decades of studies of ERK1/2 phosphorylation,
regulators K-Ras and -catenin, leading to altered especially at the Thr202/Tyr204 sites in ERK1
oncogenic activity and nuclear localization, and Thr185/Tyr187 in ERK2 of the Thr-X-Tyr
respectively (8,9). motif, the studies of other PTMs of ERK1/2,
Mitogen-activated protein kinases (MAPKs) including acetylation, methylation, and
are a conserved family of serine/threonine protein ubiquitination, are just emerging. Recently, two
kinases connected to various essential cellular lysine sites at the ERK1 C-terminus, Lys302 and
processes (10). To date, a total of 14 MAPKs Lys361, have been revealed to be tri-methylated,
have been isolated in humans, all of which fall and methylation of ERK1 enhances its
into the seven classes that follow: the phosphorylation (17). Arg309 of ERK1 has also
extracellular signal-regulated kinase class, p38 been reported as a methylation site via a
class, c-Jun N-terminal kinase (JNKs) class, and proteome-wide analysis, yet the function of this
ERK5 class belong to the conventional group of site is not clear (18). Ubiquitination of ERK1/2
MAPKs, all of which have been studied has been reported from three different proteome-
extensively, whereas the Nemo-like kinase (NLK) wide analyses, but the role of ubiquitination in
class, ERK3/4 class, and ERK7 class belong to ERK1/2 still remains elusive (19-21). Likewise,
the atypical group of MAPKs, all of which have one ERK1 acetylation site has been identified by
been studied inadequately (10,11). In general, the SILAC assays, but the function of this site still
MAPK pathway contains at least three tiers: a remains to be determined (22). Lately, several

2
 HDAC6 deacetylates ERK1

reports have shown that when cell lines including anti-AcK antibodies from Cell Signaling
A549, MB361, BT474, MV4-11, PC-3, SKBR-3, Technology. As shown in Figure 2, both
HN-9 and SQ20B were treated with HDAC6 antibodies specifically recognized acetylated
inhibitors, the level of phosphorylated ERK1/2 BSA but not no-acetyalted BSA. We then used
decreased (23-26), suggesting that acetylation of these two antibodies in the following experiments.
ERK1/2 compromises their activities and that To determine whether ERK1 is acetylated,
HDAC6 inhibition may down-regulate ERK1/2 mammalian expression vector GST-ERK1 was
activities. However, the mechanisms underlying transfected into HEK293T cells. Then the
this observation are not clear. transfected cells were treated with 0, 50, 100, 200,
Our previous study showed that ERK1 400, or 600 ng/ml of pan-HDAC inhibitor,
phosphorylates HDAC6 at its Ser1035 site, and Trichostatin A (TSA) 12 hours prior to harvest.
phosphorylation of this site increases HDAC6’s As shown in Figure 3A, after normalizing with
activity toward -tubulin and stimulates cell the total ERK1, the level of acetylated GST-
migration (27). In this study, we have determined ERK1 was increased as the dosage of TSA was
that ERK1/2 are acetylated proteins, and that increased. 600 ng/ml TSA increased the level of
ERK1/2 are novel substrates of HDAC6. Both acetylated GST-ERK1 by four-fold as compared
ERK1/2 show the ability to physically interact to a vehicle. We then treated the GST-ERK1-

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

with HDAC6 in vitro. Also, both CBP and p300 transfected HEK293T cells with 600 ng/ml TSA
acetylate ERK1/2. One novel acetylation site, at 0, 2, 4, 8, 12 and 24 hours. As shown in Figure
Lys72, was identified in ERK1 via mass- 3B, the level of acetylated GST-ERK1 increased
spectrometry analysis, and the acetylation- as the TSA treatment time increased. At the 24-
mimicking mutant of ERK1 exhibits reduced hour time point, the longest time point, the level
kinase activity, suggesting that the acetylation of acetylated GST-ERK1 is nearly two-fold as
status of ERK1 plays an important role in compared to that at the 0-hour time point. To
regulating ERK1 enzymatic activity. ensure that TSA is functional (TSA inhibits class
I, II, and IV HDACs including HDAC6) we also
RESULTS detected the level of acetylated -tubulin, which
ERK1/2 interact with HDAC6 directly – We is a well-known HDAC6 substrate (4). Using a
previously showed that ERK1/2 interact with similar approach, we have shown that ERK2 is
HDAC6 endogenously (27). However, how these acetylated upon TSA treatment (Figure 3C and
proteins interact with each other was unknown. 3D). Taken together, TSA increases ERK1/2
To determine whether ERK1/2 interact with acetylation in a dose- and time-dependent manner.
HDAC6 directly or through other proteins, we Next, we set out to determine whether
performed in vitro GST pull-down assays with specific inhibition of HDAC6 would increase
bacterially-purified HDAC6 and ERK1/2. As ERK1/2 acetylation. We treated GST-ERK1- or
shown in Figure 1A, GST-HDAC6, but not GST, GST-ERK2-transfected HEK293T cells with an
efficiently pulled down His-ERK1. Similarly, as HDAC6-specific inhibitor, ACY-1215, at the
shown in Figure 1B, GST-HDAC6 also pulled concentrations of, 0, 0.5, 1, 2, 4, and 6 g/ml, for
down His-ERK2. Therefore, ERK1/2 physically 12 hours. As shown in Figures 4, ACY-1215
interact with HDAC6. increased the level of acetylated GST-ERK1 and
Inhibition or depletion of HDAC6 increases GST-ERK2 by 9-fold and 4-fold, respectively.
ERK1/2 acetylation – Previously, we To determine whether endogenous ERK1/2
demonstrated that ERK1 phosphorylates HDAC6 are acetylated, we treated 293T cells with the
at the Ser1035 site (27). Given the fact that pharmacological inhibitor TSA or ACY-1215
HDAC6 is a deacetylase, we interrogated and examined the level of acetylation of ERK1/2.
HDAC6 to see if it deacetylates ERK1, in other As shown in Figure 5, both inhibitors increased
words, whether ERK1 is a substrate of HDAC6. the acetylation levels of endogenous ERK1/2,
To this end, we set out to determine whether suggesting that inhibition of HDAC6 increases
ERK1 is acetylated. Before using the anti- the acetylation of endogenous ERK1/2. To
acetylated lysine (AcK) antibodies to examine further confirm the role of HDAC6 in regulating
ERK1 acetylation, we tested two commercial ERK1/2 deacetylation, we compared the

3
 HDAC6 deacetylates ERK1

acetylation level of endogenous ERK1/2 in p300 increased acetylation of GST-ERK1,

HDAC6 wild-type and HDAC6 knockout MEFs. whereas overexpression of HDAC6 reduced
As shown in Figure 6A, the acetylation of CBP- or p300-mediated ERK1 acetylation. Then
endogenous ERK1/2 increased significantly in we tested whether HDAC6 is able to deacetylate
HDAC6 knockout MEFs as compared with in vitro acetylated GST-ERK1 purified from
HDAC6 wild-type MEFs. Likewise, the level of bacteria. As shown in Figure 9C, HDAC6
endogenous ERK1/2 increased in HDAC6 purified from 293T cells could efficiently
knockdown A549 cells as compared with that of deacetylate acetylated ERK1 in vitro, suggesting
control A549 cells (Figure 6B). These results that ERK1 is a bona fide substrate of HDAC6.
validate that inhibition or depletion of HDAC6’s Acetylation of ERK1 reduces ERK1’s
enzymatic activity is responsible for the increase enzymatic activity -We next examined whether
of ERK1/2 acetylation. acetylation of ERK1 affects its enzymatic activity.
CBP and p300 acetylate ERK1/2 in vivo and Because HDAC6 deacetylates ERK1, we then
in vitro – To determine which histone hypothesize that in the absence of HDAC6,
acetyltransferase (HAT) acetylates ERK1/2, we ERK1 acetylation would be increased which may
tested five HATs, which belong to the following alter ERK1’s enzymatic activity. To test this
three families: Gcn5-related N-acetyltransferase hypothesis, the ERK1 plasmid was transfected

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

(GNAT) family (PCAF), MYST family (TIP60 into wild-type 293T cells or HDAC6KO 293T
and HBO1), and p300/ CREB-binding protein cells. Then ERK1 was isolated from these two
(CBP) family (p300 and CBP) (28-30). These type of cells and ERK1’s kinase activity was
HATs were co-expressed with GST-ERK1 in examined using recombinant ELK1 as a substrate.
293T cells. When GST-ERK1 was co-expressed ELK1 is a member of Ets transcription factor
with CBP, the acetylation level of GST-ERK1 family, and several serine and threonine sites of
was much higher than that with empty vector and ELK1 can be phosphorylated by ERKs (31).
other HATs (Figure 7A). p300 also weakly Among these sites, phosphorylation status of
acetylated ERK1 (Figure 7A). To further confirm ELK1 Ser383 is pivotal for ELK1 transcriptional
the results, we co-expressed the increasing activation (32). Because of this reason, we used
amounts of CBP or p300 with GST-ERK1. As ELK1 as the substrate to execute non-radioactive
shown in Figures 7B and 7C, CBP and p300 kinase assays to measure the ability of ERK1 to
acetylated GST-ERK1 in a dose-dependent phosphorylate the Ser383 site in ELK1. As shown
manner. To further confirm CBP and p300’s in Figure 10A, ERK1 purified from HDAC6KO
effect on endogenous ERK1, we overexpressed cells displayed significantly lower activity than
p300 in HEK293T cells and tested the acetylation that from wild-type cells. Then we also directly
level of endogenous ERK1. As shown in Figure acetylated ERK1 using CBP and investigated the
7D, exogenous p300 increased the acetylation of enzymatic activity of vehicle-incubated ERK1
endogenous ERK1. To eliminate the potential versus CBP-incubated ERK1. As shown in
influence of endogenous HATs and HDACs in Figure 10B, CBP-acetylated ERK1 harbored
the cells on ERK1 acetylation, we executed in lower enzymatic activity. In summary, we
vitro acetylation assays to confirm that CBP is an concluded that acetylation of ERK1 decreases its
ERK1 acetyltransferase. As shown in Figure 7E, enzymatic activity.
recombinant CBP indeed acetylated bacterially ERK1 Lysine 72 is acetylated – To detect
purified GST-ERK. Similarly, we have shown the acetylation site of ERK1, we prepared
that both CBP and p300 promote ERK2 samples for mass spectrometry analyses. GST-
acetylation in cells and that CBP acetylated ERK1 and CBP were co-expressed in HEK293T
ERK2 in vitro. (Figure 8A-D). cells for 36 hours. Cells were harvested and lysed
HDAC6 deacetylates ERK1 in vivo and in in lysis buffer. GST-ERK1 was then pulled-down
vitro – Because of the low basal acetylation level by glutathione-agarose and resolved on SDS-
of ERK1, to confirm whether HDAC6 PAGE. The SDS-PAGE gel was stained by
deacetylates ERK1, we co-expressed CBP to coomassie blue, and the specific bands were
increase the ERK1 acetylation level. As shown in excised. Samples were further digested with
Figure 9A and 9B, as expected, both CBP and chymotrypsin and Lys-C endoproteinase

4
 HDAC6 deacetylates ERK1

sequentially and subjected for LC-tandem mass used to normalize the Firefly luciferase reading.
spectrometry analysis. Lys72 was identified as a As shown in Figure 12C, the ELK1-dependent
novel acetylation site of ERK1 (Figure 11A). promoter activity is significantly lower in
Lys72 is located on 3-strand of ERK1’s N-lobe acetylation mimetic mutant ERK1(K72Q)
and is very close to the glycine-rich loop (Figure transfected cells than in wild-type or
11B). To show the conservation of Lys72 in deacetylation mimetic mutant ERK1(K72R)
ERK1, ERK1 sequences from human to transfected cells, indicating that acetylation at
nematode were compared by the T-Coffee K72 site reduces ERK1’s activity to activate
alignment program. The alignment results ELK1-mediated transcription.
showed that this mass spectrometry-identified In order to determine how
ERK1 Lys72 was highly conserved among acetylation/deacetylation of Lys72 regulates
mammals and even in Zebrafish (Danio rerio), ERK1 kinase activity, we examined ERK1’s
Drosophila, and C. elegans (Figure 11C), crystal structure. Lys72 is located near the ATP
indicating that Lys72 plays an important role in binding site and stabilizes one wall of the ATP
ERK1 function. binding site via intramolecular contacts. In
Acetylation mimetic mutant of ERK1 particular, Lys72 forms a salt bridge with Asp117
abolishes ERK1 kinase activity toward ELK1 – and links to Tyr119 with a hydrogen bond.

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

To test whether acetylation status of Lys72 in Acetylation mimetic Lys72 would break the
ERK1 affects its kinase activity, Lys72 was contacts to both Asp117 and Tyr119, and might
mutated to glutamine (an acetylation mimetic change the conformation of the ATP binding site
mutant, K72Q) or arginine (a deacetylation leading to reduced enzymatic activity of ERK1
mimetic mutant, K72R), and the resulting (Figures 8C and 8D).
mutants were tested by their phosphorylation
status in cells followed by kinase assays. As DISCUSSION
shown in Figure 12A, K72Q, but not K72R, In this study, we have demonstrated that the
displayed a reduced level of phosphorylation as acetylation/deacetylation status of Lys72 in
compared with wild-type ERK1, implying that ERK1 regulates its enzymatic activity. For the
the K72Q mutant exhibits a diminished first time, we have revealed that ERK1 and ERK2
enzymatic activity. To confirm this notion, the are acetylated proteins and are novel substrates of
kinase assay using the most thoroughly studied the deacetylase HDAC6 and the
ERK1 substrate, ELK1, was performed. As acetyltransferases CBP and p300. We have also
shown in Figure 12B, the acetylation mimetic discovered that Lys72 of ERK1 is a novel
mutant, ERK1 (K72Q) displayed a significantly acetylation site. Furthermore, we have shown that
reduced kinase activity toward ELK1 as the ERK1 Lys72 acetylation-mimicking mutant
compared with the deacetylation mimetic mutant, (K72Q) displayed reduced kinase activity as
ERK1 (K72R) and the wild-type of ERK1. compared with the wild-type and deacetylation-
To further demonstrate the impact of ERK1- mimicking mutant (K72R). Overall, our results
K72 acetylation in vivo, we monitored activity of suggest that HDAC6/CBP and p300 govern
ELK1 in a reporter assay. The luciferase ERK1’s kinase activity via
construct, (ELK1)2-TATA-Luc, being used in deacetylation/acetylation of Lys72.
this assay was a kind gift from Dr. Manohar Although we have provided strong evidence
Ratnam. In this construct, the cis-element that HDAC6 deacetylates ERK1/2, we cannot
preferred by ELK1 was placed as two tandem rule out the possibility that other Class I, II and
repeat elements upstream of a minimal TATA- IV HDACs can deacetylate ERK1/2. In addition,
dependent Firefly luciferase promoter (33). Wild- we tested whether Class III HDACs, also called
type, K72Q and K72R of ERK1 were examined sirtuins, can deacetylate ERK1/2. We found that
by their ability to activate ELK1-mediated the Sirtuin inhibitor, nicotinamide, increased
transcription by luciferase assays in HeLa cells. ERK2, but not ERK1, acetylation (data not
The pRL plasmid encoding Renilla luciferase was shown), suggesting that one or multiple sirtuins
also transfected in HeLa cells together with the may deacetylate ERK2. Further investigations are
above constructs. The Renilla reading was then

5
 HDAC6 deacetylates ERK1

warranted to study whether other HDACs and case of ERK1, K71 and K72), which either bind
sirtuins regulate ERK1/2. to ATP or form salt bridges to affect ATP binding,
In addition, we have identified a conserved is a strategy employed by HATs and HDACs to
lysine, Lys72, which is adjacent to a critical ATP- fine-tune the enzymatic activities of MAPKs.
binding site, Lys71, as a novel acetylation site in As the main moderator in the downstream
ERK1, and the acetylation status of Lys72 of the MAPK pathway, ERK1/2 are emerging as
significantly decreases ERK1’s enzymatic alternative targets, especially when inhibitors of
activity toward a well-known ERK1 substrate, the upstream kinases become resistant to patients
ELK1. According to the structural analysis, the (38). There are several ERK1/2-specific
acetylation mimetic mutant of ERK1(K72Q), but inhibitors being used to combat the resistance to
not the deacetylation mimetic mutant EGFR, Raf or MEK inhibitors in clinical trials.
ERK1(K72R), would block the formation of the HDAC6-specific inhibitors are also being tested
salt bridges to Asp117 and Tyr119, leading to in clinical trials. Most of these trials were
decreased stability of the 3-strand, diminished conducted with other anti-cancer drugs (33,39-
ATP binding, and reduced ERK1 kinase activity. 47). The combination of HDAC inhibitors and
Our study is the first to report that the ERK1/2 pathway inhibitors have shown
acetylation/deacetylation of a conserved lysine in synergistic cell killing (48-52). Here we show

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

subdomain II of ERK1 could influence its that inhibition of HDAC6 down-regulates
enzymatic activity. It would be interesting to ERK1’s enzymatic activity, suggesting that the
speculate whether the acetylation/deacetylation combination of HDAC6 inhibitors and ERK1/2
of Lys53 in ERK2, which is equivalent to Lys72 inhibitors may be a promising strategy to
in ERK1, regulates ERK2’s kinase activity, overcome the resistance to EGFR, Raf or MEK
although Lys53 has not been identified as an inhibitors.
acetylation site yet.
More than two decades ago, it was EXPERIMENTAL PROCEDURES
demonstrated that Lys71 within subdomain II is Antibodies-Anti-acetylated lysine mouse mAb
critical for ATP binding (34,35). Substitution of (Ac-K-103) (#9681), anti-acetylated lysine rabbit
Lys to Arg at this site therefore abolishes ERK1 polyclonal antibody (#9441), anti-acetyl-p53
kinase activity (35). Interestingly, we found that (Lys382) (#2525), anti-p44/42 (ERK1/2) (#9102),
Lys71 can be acetylated by mass spectrometry anti-phospho-p44/42 (ERK1/2) (Thr202/Tyr204)
analysis (data not shown). It was expected that the (#9101), anti-ELK-1 (#9182), anti--tubulin
replacement of Lys with any other amino acid (#2125), anti--tubulin (#2146), and anti--actin
would ablate ERK1 kinase activity. Because of (#4967) antibodies were purchased from Cell
this reason, Lys to Arg (the deacetylation mimetic Signaling Technology, Inc. The anti-Flag M2
mutation) or Lys to Glu (the acetylation mimetic antibody (F1804), anti-Flag® M2 affinity gel
mutation) substitution of Lys71 would not tell us (A2220) and anti-Myc antibody (C3956) were
how deacetylation/acetylation regulates ATP purchased from Sigma. Anti-p53 (sc-126), anti-p-
binding and ERK1 enzymatic activity. Future ELK-1(B4) (sc-8406) and anti-HDAC6 (sc-
studies using a special t-RNA synthetase capable 11420) antibodies were purchased from Santa
of binding N-acetyl lysine to synthesize ERK1 Cruz Biotechnology. The Anti-HA antibody
with acetylated Lys72 may elucidate the role of (16B12) was purchased from Covance.
the acetylation of this site in ERK1 function. Chemicals and reagents-Q5® High-Fidelity
However, it is intriguing that acetylation of DNA Polymerase (#M0491S), BamHI
Lys53, a homologous site of ERK1’s Lys71, in (#R0136S), NotI (#R0189S), XhoI (#R0146S),
p38 augments p38’s kinase activity (36). SalI (#R0138S), and SpeI (#R0133S) restriction
Moreover, Lys52 in ERK2 and Lys55 in JNK1 enzymes were purchased from New England
and JNK2 are also homologous to ERK1’s Lys71 Biolabs. Protease inhibitor cocktail
(36,37), but whether these sites are acetylated (#11836170001) was purchased from Roche.
remains to be determined. It is tempting to Protein G agarose (#15920-010) and Western
hypothesize that the acetylation/deacetylation of blotting substrates (#32106) were purchased from
the two conserved lysines in subdomain II (in the ThermoFisher Scientific. Glutathione agarose

6
 HDAC6 deacetylates ERK1

(#G4510) and Trichostatin A (#T8552) were GCTGGAATC-3’ (NotI). pGEX-4T-1-GST-

purchased from Sigma. Ni2+-NTA agarose ERK2 was then used as a template for PCR using
(#635659) was purchased from Clontech. ELK-1 the following primers: 5’-
fusion protein (#9184) was purchased from Cell CCCACTAGTATGTCCCCTATACTAGGTTA
Signaling Technology. Recombinant CBP protein TTG-3’ (SpeI) and 5’-
(#BML-SE452-0100) was purchased from Enzo CGCGCGGCCGCTCAAGATCTGTATCCTG
Life Sciences. ACY-1215 was purchased from GCTGGAATC-3’ (NotI) to generate GST-ERK2
Selleckchem. Dual-Luciferase® Reporter Assay cDNA, which was further subcloned into the
System was purchased from Promega. pLEX-MCS vector (Thermo Scientific, Catalog #
Plasmids construction-pGEX-4T-1-HDAC6 OHS4735) between SpeI and NotI sites to
was generated by PCR using HA-HDAC6-F (5) generate pLEX-GST-ERK2.
as the template and the following primers: GST- Then we generated the mammalian
HD6-F 5’- expression K72Q and K72R mutants of ERK1 in
CCCGTCGACTCATGACCTCAACCGGCCA the pLEX-MCS vector as described below. The
GGA-3’ (SalI) and GST-HD6-R 5’- GST-ERK1 cDNA was generated by PCR using
TGCGGCCGCTTAGTGTGGGTGGGGCATA pGEX-4T-1-ERK1 as a template and the
TC-3’ (NotI). The PCR product was inserted into following primers: 5’-

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

the SalI and NotI site of the pGEX-4T-1 vector to CCCACTAGTATGTCCCCT
generate pGEX-4T-1-HDAC6. pLEX-GST- ATACTAGGTTATTG-3’ (SpeI) and 5’-
ERK1 was generated from the pGEX-ERK1 GGGCTCGAGCTAGGGGGCCTCCAGCACT
plasmids described in Williams et al.(27). Briefly, CC-3’ (XhoI). Then the PCR product was
ERK1 cDNA was isolated from the pDONR223- inserted between SpeI and XhoI sites to generate
MAPK3 (Addgene Plasmid 23509) vector by pcDNA3.1/Hygro-GST-ERK1. The
PCR using the following primers: ERK1-F 5’- pcDNA3.1/Hygro-GST-ERK1 (K72Q) and
CCCGGATCCATGGCGGCGGCGGCGGCTC pcDNA3.1/Hygro-GST-ERK1 (K72R) plasmids
AG-3’ (BamHI) and ERK1-R 5’- were made by site-directed mutagenesis using the
GGGCTCGAGCTAGGGGGCCTCCAGCACT following primers: for K72Q, 5’-
CC-3’ (XhoI). The PCR product was then GTGGCCATCAAGCAGATCAGCCCCTTC-3’
inserted into the BamH1 and Xho I sites into the and 5’- GATGGCCACGCGAGTCTTGCGCA-
pGEX-4T-1 vector to generate pGEX-ERK1. 3’; for K72R, 5’-
GST-ERK1 cDNA was isolated by PCR using GTGGCCATCAAGAGGATCAGCCCCTTC-3’
pGEX-ERK1 as the template and the following and 5’-GATGGCCACGCGAGTCTTGCGCA-
primers: GST-SpeI 5’- 3’. The PCR cycle for site-directed mutagenesis
CCCACTAGTATGTCCCCTATACTAGGTTA was as follows: 95 °C 5 min, 95 °C 3 min, 55 °C
TTG-3’ (SpeI) and ERK1-R 5’- 1 min, 72 °C 6 min for 16 cycles; and finally 72°C
GGGCTCGAGCTAGGGGGCCTCCAGCACT 10 min. Then the pcDNA3.1/Hygro-GST-ERK1
CC-3’ (XhoI). The PCR product was then (K72Q) and pcDNA3.1/Hygro-GST-ERK1
inserted into the SpeI and XhoI sites of the pLEX- (K72R) plasmids were cut with SpeI and XhoI to
MCS vector to generate pLEX-GST-ERK1. The isolate the cDNA fragments of GST-
pLEX-GST-DN-ERK1 plasmid was described in ERK1(K72Q) and GST-ERK1(K72R), which
Williams et al. (27). The mammalian expression were subcloned into SpeI and XhoI sites of the
GST-tagged ERK2 construct pLEX-GST-ERK2 pLEX-MCS vector to generate pLEX-GST-
was generated as follows. Human wild-type ERK1(K72Q) and pLEX-GST-ERK1(K72R),
ERK2 cDNA was amplified from pDONR223- respectively. To generate the 6xHis-tagged ERK1,
MAPK1 (Addgene plasmid 23498) by PCR and pET-ERK1, the cDNA of ERK1 was transferred
the PCR product was inserted into pGEX-4T-1 from pGEX-4T-1-ERK1(27) to the pET28a
between SalI and NotI sites to construct pGEX- vector by digesting with BamHI and XhoI to
4T-1-GST-ERK2. The primers used were 5’- generate pET28a-ERK1. To generate pET28a-
GCGGTCGACTTATGGCGGCGGCGGCGGC ERK2, the cDNA of ERK2 was excised from the
GGGC-3’ (SalI) and 5’- pGEX-4T-1-GST-ERK2 vector as described
CGCGCGGCCGCTCAAGATCTGTATCCTG above. Then the insert was subcloned into the

7
 HDAC6 deacetylates ERK1

pET28a vector via SalI and NotI sites. pCMV- immunoprecipitation assays, cells were lysed in
ELK1 purchased from Origene (SC116858) lysis buffer (25 mM Tris-HCl pH7.4, 150 mM
contains untagged human ELK1 (member of ETS NaCl, 1 mM EDTA, 1% NP-40, 5% glycerol and
oncogene family (ELK1), transcript variant protease inhibitor cocktail (11836170001,
2, NM_005229.2). The construct was made by ROCHE)). Then the lysates were first pre-cleared
inserting the cDNA of ELK1 into the Not I site of with protein G agarose for 30 minutes at 4 °C
the pCMV6-XL4 vector. with rotation, then incubated with interested
Cell culture and cell lines-HEK293T cells were antibodies overnight at 4 °C on a rocker, and
grown in Dulbecco’s modified Eagles’s medium followed by protein G agarose incubation for 4
(DMEM) with 10% bovine calf serum. HDAC6 hours at 4 °C with rotation. The samples were
wild-type and HDAC6 knockout mouse washed by washing buffer (TBS with 0.5% Triton
embryonic fibroblasts (MEFs) were cultured in X-100) 4 times and subject to further analyses.
DMEM with 10% fetal bovine serum (FBS). For immunoblotting, the samples were resolved
HDAC6 control and HDAC6 knockdown A549 on SDS-PAGE and transferred to nitrocellulose
cells were cultured in RPMI-1640 with 10% FBS. membranes. The membranes were blocked by 5%
All cell lines were cultured in the medium with non-fat milk in TBST (0.1% Tween-20 in TBS
penicillin (100 U/ml), and streptomycin (100 (20 mM Tris pH 7.5, 150 mM NaCl)) for 1 hour

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

g/ml) and in the incubators with 5% CO2 at 37 °C. at room temperature. Then the membranes were
Establishment of HDAC6 stable knock-down incubated with a first antibody overnight at 4 °C
A549 cells and HDAC6 knock-out 293T cells- and washed with TBST three times. Membranes
A549 scramble and HDAC6 knockdown stable were further incubated with an appropriate
cell lines were clonally selected by 0.5 µg/ml secondary antibody that was conjugated with
puromycin. First, A549 cells were transiently horseradish peroxidase for 2-4 hours at room
transfected with control vector pRS (Cat.# temperature and washed with TBST three times.
TR20003) or shRNA vector against HDAC6 Proteins on the membranes were detected by
(recognize sequence 5- Western blotting substrates and exposed to the X-
AGGTCTACTGTGGTCGTTACATCAATGGC ray films. To re-blot the same membrane with a
-3′, tube ID:TI349960, ORIGENE). Twenty-four different antibody, the membrane would be
hours after transfection, cells were split to stripped with stripping buffer (2% SDS, 62.5 mM
duplicate plates of 1∶20 in RPMI1640 medium Tris-HCl pH6.8, and 0.8% -mercaptoethanol)
containing 0.5 µg/ml puromycin. Puromycin was for 45 minutes at 50°C. The membrane would
replenished every 2 days to maintain sufficient then be re-blocked by 5% non-fat milk in TBST
level of selection pressure. The well-isolated and subject to the above immunoblotting
single clones were transferred into 24-well plates. procedures.
The knockdown effect was verified by Western Purification of GST-tagged ERK proteins from
Blotting analysis using anti-HDAC6 antibodies. 293T cells-pLEX-GST-ERK1, pLEX-GST-
HDAC6 knockout 293T cells were created ERK2, pLEX-GST-ERK1(K72Q), or pLEX-
using CRISPR/Cas9 (Clustered regularly GST-ERK1 (K72R) was transfected into 293T
interspaced short palindromic repeats) method. cells. Cells were lysed in lysis buffer (25 mM
Briefly, the guide RNA targeting HDAC6 exon 5 Tris-HCl pH7.4, 150 mM NaCl, 1 mM EDTA,
(5’-GAAAGGACACGCAGCGATCT-3’) was 1% NP-40, 5% glycerol) followed by incubation
selected and inserted into the LentiCRISPRv2 with glutathione agarose at 4°C overnight. The
vector (Addgene plasmid 52961) to generate the agarose beads were spun down and washed four
HDAC6KO vector which also express the codon- times with cold wash buffer (TBS with 0.5%
optimized Cas9 protein as well as puromycin Triton X-100). The agarose-bound GST proteins
resistance gene. The 293T cells infected with the were subject to assays that follow.
HDAC6-KO virus were selected for stable clones Purification of His-ERK1/2 proteins-BL21
using puromycin at 1 µg/ml. The HDAC6- bacteria harboring His-ERK1 or His-ERK2 were
Knockout clones were screened by anti-HDAC6 grown in the log phase and induced with
(H-300) Western blot analysis. Isopropyl -D-1-thiogalactopyranoside (IPTG) at
Immunoprecipitation and immunoblotting-For 4 hr. The cell pellets were lysed in bacteria lysis

8
 HDAC6 deacetylates ERK1

buffer (10 mM Tris-HCl pH 8.0, 150 mM NaCl, reactions were stopped by adding 5x SDS sample
1 mM EDTA, 5 mM DTT, 1.5% N- loading buffer and heating for 5 minutes at 100°C.
lauroylsarcosine and 1% Triton X-100). The The samples were subject to Western blotting
mixtures were further sonicated in appropriate analyses, and the anti-phospho-Ser383-ELK1
conditions until the mixtures were clear. The antibody was used for phospho-ELK1 detection.
mixtures were centrifuged at 10,000xg at 4°C for In vitro acetylation assay-Bacterial GST-ERK1
5 minutes, and the supernatants were incubated or GST-ERK2 protein was purified by
with Ni2+-NTA agarose (#635659, Clontech) at glutathione agarose as described in the GST pull-
4°C overnight. The agarose was washed four down assay. The glutathione agarose-bound
times with cold washing buffer (TBS with 0.5% GST-ERK1 or GST-ERK2 protein was mixed
Triton X-100) after incubation. His-ERK1 and with 2 g recombinant CBP proteins (BML-
His-ERK2 were further eluted by 250 mM SE452-0100, Enzo Life Sciences), 100 nM
imidazole from Ni2+-NTA agarose for next assays. acetyl-CoA in 1x acetylation buffer (50 mM Tris-
GST pull-down assay- BL21 harboring GST or HCl pH8.0, 10% glycerol, 0.1 mM EDTA, 1 mM
GST-HDAC6 were grown in the log phase and DTT, 10 mM sodium butyrate, and protease
induced with Isopropyl -D-1- inhibitor cocktail), and the mixtures were
thiogalactopyranoside (IPTG) at 4 hr. The cell incubated at 30°C for 60 minutes. Then the

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

pellets were re-suspended in bacteria lysis buffer agarose beads were washed with washing buffer
(10 mM Tris-HCl pH 8.0, 150 mM NaCl, 1 mM (TBS with 0.5% Triton X-100) three times and
EDTA, 5 mM DTT, 1.5% N-lauroylsarcosine and the reactions were terminated by adding 5x SDS
1% Triton X-100) and sonicated in appropriate sample loading buffer and boiling at 100°C for 5
conditions until the mixtures were clear. The minutes. The samples were then subject to
mixtures were centrifuged at 10,000xg at 4°C for Western blotting analyses.
5 minutes, and the supernatants were incubated Samples preparation for Mass-spectrometry-
with glutathione-agarose to isolate beads-bound GST-ERK1 and HA-CBP were co-expressed in
GST or GST-HDAC6. The purified His-ERK1 or 293T cells for 36 hours. Cells were harvested and
His-ERK2, described in the “Experimental lysed in lysis buffer. GST-ERK1 was then pulled-
Procedures” was incubated with either down by glutathione agarose and analyzed by
glutathione agarose-bound GST or GST-HDAC6 SDS-PAGE. The gel was stained by coomassie
for 30 minutes at 4°C on a rotator. After blue, and the specific bands were excised. The
incubation, the glutathione-agarose beads were excised gels were further digested with
spun down and washed four times with cold chymotrypsin and Lys-C endoproteinase
washing buffer (TBS with 0.5% Triton X-100). sequentially and subject for mass spectrometry
The samples were then subject to Western- analysis.
blotting analyses. LC-MS/MS Analysis- A nanoflow ultra high
Non-radioactive in vitro kinase assay- Wild-type, performance liquid chromatograph (RSLC,
dominant-negative, K72Q or K72R mutant ERK1 Dionex, Sunnyvale, CA) coupled to an
proteins were overexpressed in 293T cells, and electrospray bench top orbitrap mass
the cells were lysed in lysis buffer (25 mM Tris- spectrometer (Q-Exactive plus, Thermo, San
HCl pH7.4, 150 mM NaCl, 1 mM EDTA, 1% NP- Jose, CA) was used for tandem mass
40, 5% glycerol and protease inhibitor cocktail spectrometry peptide sequencing experiments.
(11836170001, ROCHE)). These GST fusion The sample was first loaded onto a pre-column (2
ERK1 proteins were purified as described in cm x 100 µm ID packed with C18 reversed-phase
“Experimental Procedures.” GST fusion proteins resin, 5µm, 100Å) and washed for 8 minutes with
were washed one time by 1x the kinase buffer (5 aqueous 2% acetonitrile and 0.04%
mM Tris pH 7.5, 5 mM -Glycerolphosphate, 2 trifluoroacetic acid. The trapped peptides were
mM DTT, 0.1 mM Na3VO4, 10 mM MgCl2) eluted onto the analytical column, (C18, 75 µm
before the reaction. All glutathione agarose- ID x 50 cm, 2 µm, 100Å, Dionex, Sunnyvale,
bound proteins were incubated with 250 ng ELK- CA). The 90-minute gradient was programmed
1 (9184, Cell Signaling), 200 M ATP and 1x as: 95% solvent A (2% acetonitrile + 0.1% formic
kinase assay buffer for 30 minutes at 30°C. The acid) for 8 minutes, solvent B (90% acetonitrile +

9
 HDAC6 deacetylates ERK1

0.1% formic acid) from 5% to 38.5% in 60 ERK1-WT, GST-ERK1-K72Q, or GST-ERK1-

minutes, then solvent B from 50% to 90% B in 7 K72R as shown in Figure 12C. The pRL plasmid
minutes and held at 90% for 5 minutes, followed encoding Renilla luciferase gene was also
by solvent B from 90% to 5% in 1 minute and re- transfected. 24 hours post transfection, cells
equilibrate for 10 minutes. The flow rate on were lyzed with 100 l 1X lysis buffer (Dual-
analytical column was 300 nl/min. Sixteen luciferase reporter, E1910/Promega). 10 l of
tandem mass spectra were collected in a data- lysates was taken out to mix with 50 l LARII
dependent manner following each survey scan. and 50 l “STOP & Glo” for firefly and renilla
Both MS and MS/MS scans were performed in reading, sequentially, in a white opaque 96-well
Orbitrap to obtain accurate mass measurement plate using a BioTek Synergy 2 Multi-Mode
using 60 second exclusion for previously sampled reader.
peptide peaks. In vitro deacetylation assay- F-HDAC6 was
Data Analysis-Sequest (53) and Mascot (54) overexpressed in 293T cells and purified with
searches were performed against the Swiss-Prot ANTI-FLAG® M2 affinity gel by following
human database. Two trypsin missed cleavages vendor’s instruction. Purified F-HDAC6 was
were allowed, the precursor mass tolerance was resolved on SDS-PAGE and stained by
10 ppm. MS/MS mass tolerance was 0.05 Da. coomassie blue to verify the amount and purity.

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

Dynamic modifications included Purified GST-ERK1 was first acetylated by CBP
carbamidomethylation (Cys), oxidation (Met) in vitro, then acetylated GST-ERK1 was
and acetylation (Lys). Both MASCOT and incubated with or without F-HDAC6 in the 1x
SEQUEST search results were summarized in deacetylation buffer (20 mM Tris [pH 8.0],
Scaffold 4.4. 150 mM NaCl, and 10% glycerol) at 30°C for 2
Luciferase Reporter Assay- Briefly, HeLa cells hr. The reaction was stopped by adding 5x SDS-
were seeded in 24-well plates overnight prior to PAGE sample buffer and boiled in 95°C for 10
transfection. The cells were transfected with minutes.
plasmids of (ELK1)2-TATA-Luc (33), pCMV-
ELK1 (Origene), HA-MEK(S218/222D), GST-

ACKNOWLEDGEMENTS
We thank Victoria Izumi at the H. Lee Moffitt Cancer Center Proteomics facility and Dr. Paul Stemmer at
the Wayne State University Proteomics Core Facility for the mass spectrometry analysis, Dr. Edward Seto
for the HBO1 plasmid, Dr. Tso-Pang Yao for the CBP construct, Dr. Manohar Ratnam, Dr. Rayna Rosati
for the plasmids of pCMV-ELK1 and (ELK1)2-TATA-Luc, and assistance of the luciferase reporter assays,
Dr. Jie Wu for HA-MEK(S218/222D) plasmid and discussion, and Mr. Joshua Haakenson for proofreading
the manuscript.

CONFLICT OF INSTEREST
We have no conflict of interest to declare.

AUTHOR CONTRIBUTIONS
J-Y W performed most of the experiments. SX helped make some constructs. MZ did the luciferase reporter
assays. BF, HH, OKK, and YZ did the mass spectrometry analyses. ZY did the structural analysis of ERK1.
WB and GB provided the critical reagents and helped design the experiments. J-Y W and XZ wrote the
manuscript. All authors reviewed the results and approved the final version of the manuscript.

REFERENCES

1. Yang, X. J., and Seto, E. (2007) HATs and HDACs: from structure, function and regulation to novel
strategies for therapy and prevention. Oncogene 26, 5310-5318
2. Yang, X. J., and Seto, E. (2008) The Rpd3/Hda1 family of lysine deacetylases: from bacteria and yeast to
mice and men. Nat Rev Mol Cell Biol 9, 206-218

10
 HDAC6 deacetylates ERK1

3. Seto, E., and Yoshida, M. (2014) Erasers of histone acetylation: the histone deacetylase enzymes. Cold
Spring Harb Perspect Biol 6, a018713
4. Hubbert, C., Guardiola, A., Shao, R., Kawaguchi, Y., Ito, A., Nixon, A., Yoshida, M., Wang, X. F., and
Yao, T. P. (2002) HDAC6 is a microtubule-associated deacetylase. Nature 417, 455-458
5. Zhang, X., Yuan, Z., Zhang, Y., Yong, S., Salas-Burgos, A., Koomen, J., Olashaw, N., Parsons, J. T.,
Yang, X. J., Dent, S. R., Yao, T. P., Lane, W. S., and Seto, E. (2007) HDAC6 modulates cell motility by
altering the acetylation level of cortactin. Mol Cell 27, 197-213
6. Zhang, M., Xiang, S., Joo, H. Y., Wang, L., Williams, K. A., Liu, W., Hu, C., Tong, D., Haakenson, J.,
Wang, C., Zhang, S., Pavlovicz, R. E., Jones, A., Schmidt, K. H., Tang, J., Dong, H., Shan, B., Fang, B.,
Radhakrishnan, R., Glazer, P. M., Matthias, P., Koomen, J., Seto, E., Bepler, G., Nicosia, S. V., Chen, J.,
Li, C., Gu, L., Li, G. M., Bai, W., Wang, H., and Zhang, X. (2014) HDAC6 Deacetylates and Ubiquitinates
MSH2 to Maintain Proper Levels of MutSalpha. Mol Cell 55, 31-46
7. Lee, Y. S., Lim, K. H., Guo, X., Kawaguchi, Y., Gao, Y., Barrientos, T., Ordentlich, P., Wang, X. F.,
Counter, C. M., and Yao, T. P. (2008) The cytoplasmic deacetylase HDAC6 is required for efficient
oncogenic tumorigenesis. Cancer Res 68, 7561-7569
8. Yang, M. H., Laurent, G., Bause, A. S., Spang, R., German, N., Haigis, M. C., and Haigis, K. M. (2013)
HDAC6 and SIRT2 regulate the acetylation state and oncogenic activity of mutant K-RAS. Mol Cancer
Res 11, 1072-1077
9. Li, Y., Zhang, X., Polakiewicz, R. D., Yao, T. P., and Comb, M. J. (2008) HDAC6 is required for

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

epidermal growth factor-induced beta-catenin nuclear localization. J Biol Chem 283, 12686-12690
10. Cargnello, M., and Roux, P. P. (2011) Activation and function of the MAPKs and their substrates, the
MAPK-activated protein kinases. Microbiol Mol Biol Rev 75, 50-83
11. Johnson, G. L., and Lapadat, R. (2002) Mitogen-activated protein kinase pathways mediated by ERK, JNK,
and p38 protein kinases. Science 298, 1911-1912
12. Lloyd, A. C. (2006) Distinct functions for ERKs? J Biol 5, 13
13. Roskoski, R. (2012) ERK1/2 MAP kinases: Structure, function, and regulation. Pharmacological Research
66, 105-143
14. Raman, M., Chen, W., and Cobb, M. H. (2007) Differential regulation and properties of MAPKs.
Oncogene 26, 3100-3112
15. Boulton, T. G., Yancopoulos, G. D., Gregory, J. S., Slaughter, C., Moomaw, C., Hsu, J., and Cobb, M. H.
(1990) An insulin-stimulated protein kinase similar to yeast kinases involved in cell cycle control. Science
249, 64-67
16. Yoon, S., and Seger, R. (2006) The extracellular signal-regulated kinase: multiple substrates regulate
diverse cellular functions. Growth Factors 24, 21-44
17. Vougiouklakis, T., Sone, K., Saloura, V., Cho, H. S., Suzuki, T., Dohmae, N., Alachkar, H., Nakamura, Y.,
and Hamamoto, R. (2015) SUV420H1 enhances the phosphorylation and transcription of ERK1 in cancer
cells. Oncotarget 6, 43162-43171
18. Larsen, S. C., Sylvestersen, K. B., Mund, A., Lyon, D., Mullari, M., Madsen, M. V., Daniel, J. A., Jensen,
L. J., and Nielsen, M. L. (2016) Proteome-wide analysis of arginine monomethylation reveals widespread
occurrence in human cells. Science Signaling 9, rs9
19. Kim, W., Bennett, E. J., Huttlin, E. L., Guo, A., Li, J., Possemato, A., Sowa, M. E., Rad, R., Rush, J.,
Comb, M. J., Harper, J. W., and Gygi, S. P. (2011) Systematic and quantitative assessment of the ubiquitin-
modified proteome. Mol Cell 44, 325-340
20. Mertins, P., Qiao, J. W., Patel, J., Udeshi, N. D., Clauser, K. R., Mani, D. R., Burgess, M. W., Gillette, M.
A., Jaffe, J. D., and Carr, S. A. (2013) Integrated proteomic analysis of post-translational modifications by
serial enrichment. Nat Methods 10, 634-637
21. Wagner, S. A., Beli, P., Weinert, B. T., Nielsen, M. L., Cox, J., Mann, M., and Choudhary, C. (2011) A
proteome-wide, quantitative survey of in vivo ubiquitylation sites reveals widespread regulatory roles.
Molecular & Cellular Proteomics 10, M111 013284
22. Hornbeck, P. V., Zhang, B., Murray, B., Kornhauser, J. M., Latham, V., and Skrzypek, E. (2015)
PhosphoSitePlus, 2014: mutations, PTMs and recalibrations. Nucleic Acids Res 43, D512-520
23. Chuang, M. J., Wu, S. T., Tang, S. H., Lai, X. M., Lai, H. C., Hsu, K. H., Sun, K. H., Sun, G. H., Chang, S.
Y., Yu, D. S., Hsiao, P. W., Huang, S. M., and Cha, T. L. (2013) The HDAC inhibitor LBH589 induces
ERK-dependent prometaphase arrest in prostate cancer via HDAC6 inactivation and down-regulation. Plos
One 8, e73401
24. Inks, E. S., Josey, B. J., Jesinkey, S. R., and Chou, C. J. (2012) A novel class of small molecule inhibitors

2
 HDAC6 deacetylates ERK1

of HDAC6. ACS Chem Biol 7, 331-339

25. Kim, I. A., No, M., Lee, J. M., Shin, J. H., Oh, J. S., Choi, E. J., Kim, I. H., Atadja, P., and Bernhard, E. J.
(2009) Epigenetic modulation of radiation response in human cancer cells with activated EGFR or HER-2
signaling: potential role of histone deacetylase 6. Radiother Oncol 92, 125-132
26. Tien, S. C., and Chang, Z. F. (2014) Oncogenic Shp2 disturbs microtubule regulation to cause HDAC6-
dependent ERK hyperactivation. Oncogene 33, 2938-2946
27. Williams, K. A., Zhang, M., Xiang, S., Hu, C., Wu, J. Y., Zhang, S., Ryan, M., Cox, A. D., Der, C. J.,
Fang, B., Koomen, J., Haura, E., Bepler, G., Nicosia, S. V., Matthias, P., Wang, C., Bai, W., and Zhang, X.
(2013) Extracellular signal-regulated kinase (ERK) phosphorylates histone deacetylase 6 (HDAC6) at
serine 1035 to stimulate cell migration. J Biol Chem 288, 33156-33170
28. Marks, P., Rifkind, R. A., Richon, V. M., Breslow, R., Miller, T., and Kelly, W. K. (2001) Histone
deacetylases and cancer: causes and therapies. Nature reviews. Cancer 1, 194-202
29. Lee, K. K., and Workman, J. L. (2007) Histone acetyltransferase complexes: one size doesn't fit all. Nature
reviews. Molecular cell biology 8, 284-295
30. Roth, S. Y., Denu, J. M., and Allis, C. D. (2001) Histone acetyltransferases. Annual review of biochemistry
70, 81-120
31. Gille, H., Kortenjann, M., Thomae, O., Moomaw, C., Slaughter, C., Cobb, M. H., and Shaw, P. E. (1995)
ERK phosphorylation potentiates Elk-1-mediated ternary complex formation and transactivation. EMBO J
14, 951-962

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

32. Marais, R., Wynne, J., and Treisman, R. (1993) The SRF accessory protein Elk-1 contains a growth factor-
regulated transcriptional activation domain. Cell 73, 381-393
33. Patki, M., Chari, V., Sivakumaran, S., Gonit, M., Trumbly, R., and Ratnam, M. (2013) The ETS domain
transcription factor ELK1 directs a critical component of growth signaling by the androgen receptor in
prostate cancer cells. J Biol Chem 288, 11047-11065
34. Hanks, S. K., Quinn, A. M., and Hunter, T. (1988) The protein kinase family: conserved features and
deduced phylogeny of the catalytic domains. Science 241, 42-52
35. Robbins, D. J., Zhen, E., Owaki, H., Vanderbilt, C. A., Ebert, D., Geppert, T. D., and Cobb, M. H. (1993)
Regulation and properties of extracellular signal-regulated protein kinases 1 and 2 in vitro. J Biol Chem
268, 5097-5106
36. Pillai, V. B., Sundaresan, N. R., Samant, S. A., Wolfgeher, D., Trivedi, C. M., and Gupta, M. P. (2011)
Acetylation of a conserved lysine residue in the ATP binding pocket of p38 augments its kinase activity
during hypertrophy of cardiomyocytes. Mol Cell Biol 31, 2349-2363
37. Robinson, M. J., Harkins, P. C., Zhang, J., Baer, R., Haycock, J. W., Cobb, M. H., and Goldsmith, E. J.
(1996) Mutation of position 52 in ERK2 creates a nonproductive binding mode for adenosine 5'-
triphosphate. Biochemistry 35, 5641-5646
38. Samatar, A. A., and Poulikakos, P. I. (2014) Targeting RAS-ERK signalling in cancer: promises and
challenges. Nature Reviews Drug Discovery 13, 928-942
39. Mevissen, T. E. T., and Komander, D. (2017) Mechanisms of Deubiquitinase Specificity and Regulation.
Annu Rev Biochem 86, 159-192
40. A Phase 1b Study of Paclitaxel And Ricolinostat For The Treatment Of Gynecological Cancer.
41. ACY-1215 + Nab-paclitaxel in Metastatic Breast Cancer.
42. Phase 1b Study Evaluating ACY-1215 (Ricolinostat) in Combination With Pomalidomide and
Dexamethasone in Relapsed or Relapsed-and-Refractory Multiple Myeloma.
43. ACY-1215 for Relapsed/Refractory Lymphoid Malignancies.
44. Alternative 10 mg/mL Liquid Formulation of ACY 1215 (Ricolinostat) in Healthy Subjects.
45. ACY-1215 (Ricolinostat) in Combination With Pomalidomide and Low-dose Dex in Relapsed-and-
Refractory Multiple Myeloma.
46. Study of ACY-1215 in Combination With Lenalidomide, and Dexamethasone in Multiple Myeloma.
47. Study of ACY-1215 Alone and in Combination With Bortezomib and Dexamethasone in Multiple
Myeloma.
48. Zhang, L., Zhang, Y., Mehta, A., Boufraqech, M., Davis, S., Wang, J., Tian, Z., Yu, Z., Boxer, M. B.,
Kiefer, J. A., Copland, J. A., Smallridge, R. C., Li, Z., Shen, M., and Kebebew, E. (2015) Dual inhibition of
HDAC and EGFR signaling with CUDC-101 induces potent suppression of tumor growth and metastasis in
anaplastic thyroid cancer. Oncotarget 6, 9073-9085
49. Carson, R., Celtikci, B., Fenning, C., Javadi, A., Crawford, N., Perez-Carbonell, L., Lawler, M., Longley,
D. B., Johnston, P. G., and Van Schaeybroeck, S. (2015) HDAC Inhibition Overcomes Acute Resistance to

3
 HDAC6 deacetylates ERK1

MEK Inhibition in BRAF-Mutant Colorectal Cancer by Downregulation of c-FLIPL. Clinical Cancer

Research 21, 3230-3240
50. Ahn, M. Y., Ahn, J. W., Kim, H. S., Lee, J., and Yoon, J. H. (2015) Apicidin inhibits cell growth by
downregulating IGF-1R in salivary mucoepidermoid carcinoma cells. Oncol Rep 33, 1899-1907
51. Yu, C., Friday, B. B., Lai, J. P., McCollum, A., Atadja, P., Roberts, L. R., and Adjei, A. A. (2007)
Abrogation of MAPK and Akt signaling by AEE788 synergistically potentiates histone deacetylase
inhibitor-induced apoptosis through reactive oxygen species generation. Clinical Cancer Research 13,
1140-1148
52. Bahr, J. C., Robey, R. W., Luchenko, V., Basseville, A., Chakraborty, A. R., Kozlowski, H., Pauly, G. T.,
Patel, P., Schneider, J. P., Gottesman, M. M., and Bates, S. E. (2016) Blocking downstream signaling
pathways in the context of HDAC inhibition promotes apoptosis preferentially in cells harboring mutant
Ras. Oncotarget 7, 69804-69815
53. Eng, J. K., McCormack, A. L., and Yates, J. R. (1994) An approach to correlate tandem mass spectral data
of peptides with amino acid sequences in a protein database. J Am Soc Mass Spectrom 5, 976-989
54. Perkins, D. N., Pappin, D. J., Creasy, D. M., and Cottrell, J. S. (1999) Probability-based protein
identification by searching sequence databases using mass spectrometry data. Electrophoresis 20, 3551-
3567

FOOTNOTES

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

*This work was supported in part by an NIH grant, R01CA164147, and the Karmanos Cancer Institute
start-up funds to X.Z. J-Y W was supported in part by the Department of Pathology & Cell Biology, USF
Morsani College of Medicine.

ABBREVIATIONS
HDAC6, histone deacetylase 6; ERK1/2, extracellular signal-regulated kinase 1/2; MAPKs, mitogen-
activated protein kinases; JNK, c-Jun N-terminal kinases; NLK, Nemo-like kinase; PTMs, post-
translational modifications; MEFs, mouse embryonic fibroblasts; HATs, histone acetyltransferases; Rpd3,
reduced potassium dependency 3; MAP3K, MAPK kinase kinase; MAP2K, MAPK kinase; TSA,
Trichostatin A; NAD+, oxidized nicotinamide adenine dinucleotide; EGFR, epidermal growth factor
receptor; GNAT, Gcn5-related N-acetyltransferase.

FIGURE LEGENDS

FIGURE 1. ERK1/2 interact with HDAC6 physically. A, His-ERK1 binds to GST-HDAC6. The GST
pull-down assays were performed with indicated proteins as described in the “Experimental Procedures.”
The proteins pulled down by glutathione agarose were resolved on SDS-PAGE followed by Western
blotting analyses using the anti-ERK1/2 antibody (upper panel). Bacterially purified proteins, GST, GST-
HDAC6, and His-ERK1, were stained by coomassie blue (middle and lower panels). B, His-ERK2 binds
to GST-HDAC6. The GST pull-down assays and coomassie blue staining were conducted as described in
A.

FIGURE 2. Characterization of the anti-acetylated-lysine antibodies. Increasing doses of non-

acetylated BSA and acetylated BSA were resolved on SDS-PAGE. Two commercial anti-acetylated lysine
antibodies as indicated were used for Western blotting analyses. Both short exposures and long exposures
were shown.

FIGURE 3. Inhibition of HDACs by TSA increases ERK1 and ERK2 acetylation. A, TSA increases
ERK1 acetylation in a dose-dependent manner. Left panels: HEK293T cells were transfected with GST-
ERK1 followed by treatment with TSA for 12 hours with indicated concentrations. GST-ERK1 proteins
were then isolated from HEK293T cells as described in “Experimental Procedures.” Western blotting
analyses were performed with the indicated antibodies. Right panel: The acetylated GST-ERK1 and total
GST-ERK1 bands were quantified by densitometry using software Image LabTM from Bio-Rad. The value

4
 HDAC6 deacetylates ERK1

of untreated acetylated GST-ERK1 band normalized against the total GST-ERK1 band was designated as
1, and the value of other normalized treated acetylated GST-ERK1 bands were indicated as a fold change
relative to the untreated one. A bar graph was used to show relative intensity of untreated and treated
acetylated GST-ERK1. Four independent experiments were performed. B, TSA increases ERK1 acetylation
in a time-dependent manner. Left panels: HEK293T cells were transfected with GST-ERK1 followed by
treatment with 600 ng/ml TSA at indicated time points. GST-ERK1 proteins were then isolated from
HEK293T cells as described in “Experimental Procedures.” Western blotting analyses were performed with
the indicated antibodies. Right panel: A bar graph was drawn as described in A. Five independent
experiments were performed. C and D, TSA increases ERK2 acetylation in a dose- and time-dependent
manner. For C and D, the experiments were performed the same as described in A and B, respectively. For
C, four independent experiments were performed. For D, three independent experiments were performed.
Student’s t tests were performed with *, p<0.05, **, p<0.01. Error bars, S.D.

FIGURE 4. Inhibition of HDAC6 by ACY-1215 increases ERK1 and ERK2 acetylation. A, ACY-1215
increases ERK1 acetylation in a dose-dependent manner. The experiments were performed the same as
described in 3A, except that ACY-1215 was used for the treatment instead of TSA. B, ACY-1215 increases
ERK2 acetylation in a dose-dependent manner. The experiments were performed the same as described in

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

A, except that GST-ERK2 was used for transfection instead of GST-ERK1. For A, four independent
experiments were performed. For B, six independent experiments were performed. Student’s t tests were
performed with *, p<0.05, **, p<0.01, ***, p<0.001. Error bars, S.D.

FIGURE 5. Inhibition of HDACs by TSA and inhibition of HDAC6 by ACY-1215 increase

endogenous ERK1 and ERK2 acetylation. A, TSA increases endogenous ERK1/2 acetylation. Left panels:
HEK293T cells were treated with a vehicle or 600 ng/ml TSA for 24 hours. The anti-AcK antibody was
used to immunoprecipitate acetylated ERK1/2. The immunoprecipitates were resolved on SDS-PAGE and
the anti-ERK1/2 Western blot analysis was performed. 0.5% of whole cell lysate was used as the input.
Western blotting analyses were performed using anti-ERK1/2, anti-acetyl--tubulin, or -tubulin
antibodies as indicated. Right panel: A bar graph was used to show relative intensity of endogenous ERK1
and ERK2. Three independent experiments were performed. B, ACY-1215 increases endogenous ERK1/2
acetylation. The experiments were performed as described in A except that ACY-1215 was used for the
treatment instead of TSA. Four independent experiments were performed. Student’s t tests were performed
with *, p<0.05, **, p<0.01. Error bars, S.D.

FIGURE 6. Depletion or knockdown of HDAC6 increases endogenous ERK1/2 acetylation. A,

Knockout of HDAC6 increases ERK1/2 acetylation. Left panels: The anti-AcK antibody was used to
immunoprecipitate acetylated ERK1/2 in HDAC6 wild-type or HDAC6 knockout MEFs. The
immunoprecipitates were resolved on SDS-PAGE and the anti-ERK1/2 Western blotting analysis was
performed. 0.5% of whole cell lysate was used as the input. Western blotting analyses were performed
using anti-ERK1/2, anti-HDAC6, anti-acetyl--tubulin, or -tubulin antibodies as indicated. Right panel:
A bar graph was used to show relative intensity of endogenous ERK1 and ERK2. Five independent
experiments were performed. B, Knockdown of HDAC6 increases ERK1/2 acetylation. The experiments
were conducted the same as A except that A549 control and A549 HDAC6-KD pair was used to replace the
wild-type and HDAC6KO MEFs pair. Four independent experiments were performed. Student’s t tests
were performed with *, p<0.05, **, p<0.01. Error bars, S.D.

FIGURE 7. ERK1 is acetylated by CBP and p300 in vivo and in vitro. A, CBP and p300 acetylate ERK1.
GST-ERK1 was co-transfected with each of the indicated HAT plasmids into HEK293T cells. GST-ERK1
was pulled-down by glutathione-agarose, then the beads-bound proteins were resolved on SDS-PAGE
followed by Western blotting analyses with anti-acetyl-lysine antibodies. The membrane was then stripped
and reblotted with anti-ERK1/2 antibodies. The input was subject to Western blotting analyses with
indicated antibodies. B, CBP acetylates ERK1 in a dosage-dependent manner. GST-ERK1 was co-

5
 HDAC6 deacetylates ERK1

transfected with an increasing amount of HA-CBP plasmids as indicated into HEK293T cells. GST-ERK1
was further pulled down by glutathione agarose, then the beads-bound ERK1 was subject to the anti-acetyl-
lysine Western blotting analysis. The membrane was then stripped and reprobed with anti-ERK1/2
antibodies. The input was subject to Western blotting analyses with indicated antibodies. C, p300 acetylates
ERK1. GST-ERK1 was co-transfected with an increasing amount of HA-p300 plasmids. The experiments
were performed as described in B. D, p300 acetylates endogenous ERK1/2. HEK293T cells were
transfected with empty vector or HA-p300. The immunoprecipitation assays were carried out with anti-
AcK antibodies as described in the “Experimental Procedures.” The immunoprecipitates were subject to
the anti-AcK Western blotting analysis. 0.5% of the whole cell lysate was used as input, which was subject
to Western blotting analyses with indicated antibodies. E, Recombinant CBP acetylates ERK1 in vitro.
Bacterially expressed GST-ERK1 and the catalytic domain of CBP were subject to in vitro acetylation
assays as described in “Experimental Procedures.” An equal amount of GST-ERK1 was incubated with or
without 2 g of the recombinant CBP catalytic domain, and the reactions were analyzed by the anti-AcK
Western blotting analysis (upper panel). After transfer, the membrane was stained with Ponceau S to
confirm the amount and purity of GST-ERK1 (lower panel). For panels A-C, the acetylated GST-ERK1
bands were quantified by the software Image LabTM from Bio-Rad and normalized by the total GST-ERK1
bands. The values of the empty vector-transfected Ac-GST-ERK1 bands normalized by total GST-ERK1

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

were designated as 1, and the density of other HATs-transfected Ac-GST-ERK1 bands was indicated as a
fold change relative to the corresponding empty vector transfected bands. For the panel D upper panel, the
ERK1/2 bands were quantified by densitometry, and the empty vector-transfected ERK1/2 bands were
designated as 1. The density of the HA-p300-transfected ERK1/2 bands were quantified relative to the
empty vector-transfected ones.

FIGURE 8. ERK2 is acetylated by CBP and p300 in vivo and in vitro. A, CBP and p300 acetylate ERK2.
GST-ERK2 was co-transfected with each of the indicated HAT plasmids into HEK293T cells. GST-ERK2
was pulled-down by glutathione-agarose, then the beads-bound proteins were resolved on SDS-PAGE
followed by Western blotting analyses with anti-acetyl-lysine antibodies. The membrane was then stripped
and reblotted with anti-ERK1/2 antibodies. The input was subject to Western blotting analyses with
indicated antibodies. B, CBP acetylates ERK2 in a dosage-dependent manner. GST-ERK2 was co-
transfected with an increasing amount of HA-CBP plasmids as indicated into HEK293T cells. GST-ERK2
was further pulled down by glutathione agarose, then the beads-bound ERK2 was subject to anti-acetyl-
lysine Western blotting analysis. The membrane was then stripped and reprobed with anti-ERK1/2
antibodies. The input was subject to Western blotting analyses with indicated antibodies. C, p300 acetylates
ERK2. GST-ERK2 was co-transfected with an increasing amount of HA-p300 plasmids in HEK293T cells.
The experiments were performed as described in B. D, Recombinant CBP acetylates ERK2 in vitro.
Bacterially expressed GST-ERK2 and catalytic domain of CBP were subject to in vitro acetylation assays
as described in the “Experimental Procedures.” Equal amount of GST-ERK2 was incubated with or without
2 mg of the recombinant CBP catalytic domain, and the in vitro acetylation reactions were analyzed by the
anti-AcK Western blotting analysis (upper panel). For the above Western blotting analysis, after transfer,
the membrane was stained with Ponceau S to confirm the amount and purity of GST-ERK2 (lower panel).
For panels A-C, the acetylated GST-ERK2 bands were quantified as described in Figure 7.

FIGURE 9. HDAC6 deacetylates ERK1 in vivo and in vitro. A, HDAC6 deacreases CBP-induced ERK1
acetylation in vivo. GST-ERK1 was transfected into HEK293T cells alone, with CBP, or with CBP and
HDAC6. GST-ERK1 proteins were purified as described in “Experimental Procedures” followed by the
anti-AcK Western blotting analysis. The membrane was then stripped and then reprobed with anti-ERK1/2
antibodies. The input was subject to Western blotting analyses with indicated antibodies. B, HDAC6
deacreases p300-induced ERK1 acetylation in vivo. The experiments were performed as described in A,
except that HA-p300 was used to replace HA-CBP. C. HDAC6 deacetylates ERK1 in vitro. Left panels:
Acetylated bacterially purified GST-ERK1 protein was incubated with or without F-HDAC6 purified from
293T cells as described in “Experimental Procedures” followed by the anti-AcK Western blotting analysis.

6
 HDAC6 deacetylates ERK1

The procedure for generating acetylated GST-ERK1 was described in “Experimental Procedures.” The
membrane was stripped and reblotted with the anti-ERK1/2 antibody. Right panel: The relative intensity of
acetylated GST-ERK1 band was quantified by densitometry against the total ERK1 band showing in a bar
graph. This experiment was repeated three times. Student’s t tests were performed with ***, p<0.001. Error
bars, S.D.

FIGURE 10. Acetylated ERK1 exhibits decreased enzymatic activity. A, GST-ERK1 exhibits lower
enzymatic activity in 293T-HDAC6KO cells than in 293T wild-type cells. Left panels: GST-ERK1 was
transfected into 293T wild-type cells and 293T-HDAC6KO cells. Then the GST-ERK1 protein was purified
by GST pull-down followed by kinase assays using recombinant ELK1 as a substrate as described in
“Experimental Procedures.” Anti-p-ELK(S383) Western blot was performed. The membrane was then
stripped and reprobed with the anti-ELK1 antibody. Anti-ERK1/2, anti-HDAC6, anti-ac--tubulin, and
anti--tubulin Western blots were also performed. Right panel: The pELK1 bands were quantified by
densitometry against total ELK1 bands. The results were shown in a bar graph. The same experiments were
repeated five times. B, Acetylated ERK1 exhibits decreased enzymatic activity towards ELK1. Left panels:
In vitro kinase assays were performed with either vehicle incubated GST-ERK1 or CBP incubated GST-
ERK1. GST-ERK1 protein was purified from 293T cells. Recombinant ELK1 fusion protein purified from

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

E. Coli containing ELK1 residue 307-428 was used as a substrate. The kinase assays were performed as
described in “Experimental Procedures.” Anti-pELK1(S383) Western blot was used as a readout for
ERK1’s activity (upper panel). The membrane was stripped and reprobed with the anti-ELK1 antibody.
The anti-AcK Western blot was performed to examine the acetylation status of ERK1 (third panel). The
membrane was stripped and reprobed with the anti-ERK1/2 antibody. Right panel: The relative intensity of
ELK1(S383) phosphorylation was plotted as a bar graph. The experiments were repeated five times.
Student’s t tests were performed with **, p<0.01, ***, p<0.001. Error bars, S.D.

FIGURE 11. The Lysine 72 site of ERK1 is acetylated. A, The Lys72 site of ERK1 is acetylated. The
doubly charged peptide was detected with a mass-to-charge ratio 646.3179, which represents an error of -
3.8 ppm. The tandem mass spectrum matched the following sequence, KISPFEHQTY, indicating that the
Lys72 (highlighted in red) was acetylated; the detection of b2 is consistent with this localization. The
assignment was made with Sequest with Xcorr score 2.1 and ΔCN score 0.19. B, A diagram of ERK1’s
domain structure. The numbers showed the amino acid residues of ERK1. The lysine 72, detected by mass
spectrometry as an acetylated site, is indicated in a red circle. Thr202 and Tyr204 of the Thr-X-Tyr motif,
whose phosphorylation status is critical for ERK1 activity, are shown in blue triangles. The kinase domain
is highlighted as light gray. Other important motifs and regions are indicated as well. C, A stretch of ERK1
amino acids shows the conservation of Lys72 in different species from mammals to nematode. The mass
spectrometry-detected acetylation site Lys72 is highlighted in red.

FIGURE 12. The acetylation-mimicking mutant of ERK1(K72Q) decreases the ERK1 kinase activity.
A, The acetylation-mimicking mutant of ERK1(K72Q) displays decreased ERK1 phosphorylation in
HEK293T cells. The GST-tagged dominant-negative, wild-type, K72Q, or K72R ERK1 was transfected
into HEK293T cells, and GST-tagged proteins were purified as described in the “Experimental Procedures.”
Glutathione-agarose bound proteins were analyzed by the anti-phospho-ERK1/2 Western Blotting analysis.
The membrane was then stripped and reprobed with anti-ERK1/2 antibodies. B, The K72Q mutant of ERK1
exhibits a decreased kinase activity toward ELK1. Dominant-negative, wild-type, K72Q, or K72R GST-
ERK1 was transfected into HEK293T cells. The GST-tagged proteins were purified as described in the
“Experimental Procedures.” The glutathione agarose bound proteins were subject to non-radioactive in vitro
kinase assays using recombinant ELK1 as a substrate. The phosphorylation of ELK1 was measured by anti-
phospho-ELK1 (Ser383) Western blotting analysis. The anti-ERK1/2 Western blotting analysis was also
performed. C, HeLa cells were transfected with indicated plasmids. Luciferase reporter assays were
performed with the Dual-luciferase reporter kit (Promega) as described in the “Experimental Procedures.”
The expression of those plasmids was examined by anti-ELK1, anti-HA, anti-ERK1/2, and anti--actin

7
 HDAC6 deacetylates ERK1

Western blotting analyses as indicated. This experiment was repeated three times. Student’s t tests were
performed with *, p<0.05, ***, p<0.001, ns, not significant. Error bars, S.D. D, Potential structural effects
of K72 mutation and acetylation. Ribbon diagram of ERK1 structure (PDB code: 2ZOQ) and a close-up
view of K72 interaction (right panel). K72 and its interacting residues (D117 and Y119) are depicted by
sticks with their carbon atoms colored in gray and cyan, respectively. Residues T202 and phosphorylated
Y204 indicate the location of the activation loop. The ATP binding site is indicated by binding of the
inhibitor 5ID (5-Iodotubercidin). Hydrogen bonds are illustrated as red broken lines.

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

8
 Wu et al. Figure 1

A B
His-ERK1 His-ERK2

GST-HDAC6

GST-HDAC6
5% input

5% input
GST

GST
GST pull-down GST pull-down
His-ERK1 His-ERK2
IB anti-ERK1/2 IB anti-ERK1/2
250 -
Mr (Kd) 150 -
GST-HDAC6 Mr (Kd) 250 -
GST-HDAC6
150 -
100 - 100 -
75 - 75 -

Coomassie blue 50 - Coomassie blue

50 -
Staining Staining
37 -
37 -

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

25 - GST
25 - GST
Mr (Kd) 250 - Mr (Kd) 250 -
150 - 150 -
100 - 100 -
75 - 75 -

Coomassie blue 50 - His-ERK1 Coomassie blue 50 -

Staining Staining His-ERK2
37 - 37 -

25 - 25 -
1 2 3 1 2 3
 Wu et al. Figure 2

Non-acetylated acetylated
BSA BSA
Mr (kd)
0.2 1.0 5.0 25 0.2 1.0 5.0 25 (ng)
100 -
75 -
IB: anti-AcK103 Short exposure
50 -

Acetylated- 37 -
lysine mouse
mAb (Ac-K-103) 100 -
#9681 75 -
IB: anti-AcK103 Long exposure
50 -

37 -

100 -

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

75 -
IB: anti-AcK (9441) Short exposure
50 -

Acetylated 37 -
lysine antibody
#9441 100 -
75 -
IB: anti-AcK (9441) Long exposure
50 -

37 -
 Wu et al. Figure 3

A TSA (12 hours)

Relative intensity of Ac-GST-ERK1

7
0 50 100 200 400 600 (ng/ml)
* *
IB anti-AcK Ac-GST-ERK1 6
GST 5
Pull-down
IB anti-ERK1/2 GST-ERK1 4 *
IB anti-Ac- 3
Ac--tubulin
-tubulin 2
Input
IB anti--tubulin -tubulin 1
1 2 3 4 5 6 0
0 50 100 200 400 600
TSA concentrations (ng/ml) (n=4)
B
TSA (600 ng/ml)

Relative intensity of Ac-GST-ERK1

3
0 2 4 8 12 24 (Hour) *
IB anti-AcK Ac-GST-ERK1 2.5 **
GST **

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

Pull-down 2
IB anti-ERK1/2 GST-ERK1
1.5
IB anti-Ac-
-tubulin Ac--tubulin 1
Input
IB anti--tubulin 0.5
-tubulin
1 2 3 4 5 6 0
0 2 4 8 12 24
C Time (hours) (n=5)
Relative intensity of Ac-GST-ERK2

TSA (12 hours) 3.5

0 50 100 200 400 600 (ng/ml) *
3
IB anti-AcK Ac-GST-ERK2
GST 2.5 *
Pull-down *
IB anti-ERK1/2 GST-ERK2 2
* *
1.5
IB anti-Ac- Ac--tubulin
Input -tubulin 1

IB anti--tubulin -tubulin 0.5

1 2 3 4 5 6 0
0 50 100 200 400 600
TSA concentrations (ng/ml) (n=4)

D
Relative intensity of Ac-GST-ERK2

TSA (600 ng/ml)

3
0 2 4 8 12 24 (Hour)
**
Ac-GST-ERK2 2.5
IB anti-AcK **
GST ** **
2
Pull-down
IB anti-ERK1/2 GST-ERK2 1.5

IB anti-Ac- 1
Ac--tubulin
Input -tubulin 0.5
IB anti--tubulin -tubulin 0
1 2 3 4 5 6 0 2 4 6 12 24
Time (hours) (n=3)
 Wu et al. Figure 4

Relative intensity of Ac-GST-ERK1

16
ACY-1215 (12 hours) 14 * **
0 0.5 1 2 4 6 g/ml)
12
IB anti-AcK Ac-GST-ERK1
GST 10
*
Pull-down 8
IB anti-ERK1/2 GST-ERK1
6
IB anti-Ac- 4
Ac--tubulin
Input -tubulin
2
IB anti--tubulin -tubulin 0
1 2 3 4 5 6 0 0.5 1 2 4 6
Concentration (g/ml) (n=4)

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

7

Relative intensity of Ac-GST-ERK2

**
ACY-1215 (12 hours) 6
0 0.5 1 2 4 6 g/ml) **
5
IB anti-AcK Ac-GST-ERK2
GST 4
Pull-down * ***
IB anti-ERK1/2 GST-ERK2 **
3
IB anti-Ac-
Ac--tubulin 2
Input -tubulin
IB anti--tubulin -tubulin 1

1 2 3 4 5 6 0
0 0.5 1 2 4 6
Concentrations (g/ml) (n=6)
 Wu et al. Figure 5

6 **
A
293T 5

Relative intensity of ERK1/2

IP anti-AcK 0.5% input
TSA (600 ng/ml) - + - + 4 *
ERK1
IB: anti-ERK1/2 ERK2
3

IB anti-Ac--tubulin Ac--tubulin
Input 2
IB anti--tubulin -tubulin

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

1 2 3 4 1

0
ERK1 ERK2
Vehicle TSA (600 ng/ml)

(n=3)

6 *
B

293T 5
Relative intensity of ERK1/2

IP anti-AcK 0.5% input

ACY-1215 (6 g/ml) - + - + 4
ERK1
IB: anti-ERK1/2 ERK2
3 **
IB anti-Ac--tubulin Ac--tubulin
Input
IB anti--tubulin -tubulin 2

1 2 3 4
1

0
ERK1 ERK2

Vehicle ACY-1215 ( 6 μg/ml)

(n=4)
 Wu et al. Figure 6

A
MEFs 5 **
IP anti-AcK 0.5% input
WT KO WT KO 4.5

IB: anti-ERK1/2 ERK1 4

Relative intensity of ERK1/2

ERK2
3.5
IB anti-HDAC6 HDAC6 3

Input 2.5
IB anti-Ac--tubulin Ac--tubulin **
2
IB anti--tubulin -tubulin 1.5
1 2 3 4
1
0.5

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

0
ERK1 ERK2

MEFs WT MEFs HDAC6 KO

(n=5)
B
A549 5
IP anti-AcK 0.5% input
*
4.5
Ctrl KD Ctrl KD
ERK1 4
Relative intensity of ERK1/2

IB: anti-ERK1/2 ERK2

3.5

IB anti-HDAC6 HDAC6 3
2.5 *
IB anti-Ac--tubulin Ac--tubulin
Input 2
IB anti--tubulin -tubulin 1.5
1 2 3 4 1
0.5
0
ERK1 ERK2
A549 Ctrl A549 HDAC6-KD

(n=4)
 Wu et al. Figure 7
GST-ERK1
A

Empty vector

Flag-HBO1
Flag-PCAF

Myc-TIP60
HA-p300
HA-CBP
IB anti-AcK Ac-GST-ERK1
1.00 5.05 90.1 1.11 3.67 0.77
GST pull-down
IB anti-ERK1/2 GST-ERK1

IB anti-HA HA-CBP/p300
IB anti-Flag Flag-PCAF
IB anti-Flag Flag-HBO1

Input IB anti-Myc Myc-Tip60

IB anti-Ac-p53-K382 Ac-p53-K382
IB anti-p53 p53
IB anti--tubulin -tubulin
1 2 3 4 5 6

B
GST-ERK1

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

HA-CBP -
IB anti-AcK Ac-GST-ERK1
GST pull-down 1.00 2.01 3.77 5.83

IB anti-ERK1/2 GST-ERK1

IB anti-HA HA-CBP

IB anti-Ac-p53-K382 Ac-p53-K382
Input
IB anti-p53 p53

IB anti--tubulin -tubulin
1 2 3 4

C GST-ERK1
HA-p300 -
IB anti-AcK Ac-GST-ERK1
GST 1.00 1.97 2.43 3.69
Pull-down IB anti-ERK1/2 GST-ERK1

IB anti-HA HA-p300
IB anti-Ac-p53-K382 Ac-p53-K382
Input
IB anti-p53 p53
IB anti--tubulin -tubulin
1 2 3 4

D 293T
E GST-ERK1
IP anti-AcK 0.5% input CBP - +
Mr (Kd) 75 -
Empty HA- Empty HA-
vector p300 vector p300 Ac-GST-ERK1

IB: anti-ERK1/2 ERK1 IB Anti-AcK 50 - Auto-Ac-CBP

ERK2
1.00 2.36
1.00 16.1
37 -
IB anti-HA HA-p300 75 -

IB anti-Ac-p53-K382 Ac-p53-K382 GST-ERK1

Input Ponceau S
IB anti-p53 p53 50 -
staining
IB anti--tubulin -tubulin
1 2 3 4 37 -

1 2
 Wu et al. Figure 8
GST-ERK2
A

Empty vector

Flag-HBO1
Flag-PCAF

Myc-TIP60
HA-p300
HA-CBP
IB anti-AcK Ac-GST-ERK2
1.00 6.32 10.3 2.02 1.79 0.68
GST pull-down
IB anti-ERK1/2 GST-ERK2

IB anti-HA HA-CBP/p300
IB anti-Flag Flag-PCAF
IB anti-Flag Flag-HBO1
IB anti-Myc Myc-Tip60
Input
IB anti-Ac-p53-K382 Ac-p53-K382
IB anti-p53 p53
IB anti--tubulin -tubulin
1 2 3 4 5 6

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

GST-ERK2
HA-CBP -
IB anti-AcK Ac-GST-ERK2
GST pull-down 1.00 1.01 1.68 3.21

IB anti-ERK1/2 GST-ERK2

IB anti-HA HA-CBP

IB anti-Ac-p53-K382 Ac-p53-K382
Input
IB anti-p53 p53

IB anti--tubulin -tubulin
1 2 3 4
C
GST-ERK2
HA-p300 -
IB anti-AcK Ac-GST-ERK2
GST 1.00 2.16 3.16 3.69
Pull-down IB anti-ERK1/2 GST-ERK2

IB anti-HA HA-p300

IB anti-Ac-p53-K382 Ac-p53-K382
Input
IB anti-p53 p53

IB anti--tubulin -tubulin
1 2 3 4

D GST-ERK2
CBP - +
Mr (Kd) 75 -
Ac-GST-ERK2

50 - Auto-Ac-CBP
IB Anti-AcK

37 -
75 -

GST-ERK2
Ponceau S 50 -
staining

37 -
1 2
 Wu et al. Figure 9

B
A
GST-ERK1 + + + GST-ERK1 + + +
HA-CBP + + HA-p300 + +
HA-HDAC6 + HA-HDAC6 +

IB anti-AcK Ac-GST-ERK1 IB anti-AcK Ac-GST-ERK1

GST pull-down GST pull-down
IB anti-ERK1/2 GST-ERK1 IB anti-ERK1/2 GST-ERK1

IB anti-HA HA-CBP IB anti-HA HA-p300

Input IB anti-HA HA-HDAC6 Input IB anti-HA HA-HDAC6

IB anti--tubulin -tubulin IB anti--tubulin -tubulin

1 2 3 1 2 3

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

1.2
C ***
Relative intensity of Ac-GST-ERK1

GST-ERK1 1
F-HDAC6

0.8
Ctrl

0.6
IB anti-AcK Acetyl-GST-ERK1
0.4
IB anti-ERK1/2 GST-ERK1

0.2

0
Control F-HDAC6
(n=3)
 Wu et al. Figure 10

A GST-ERK1

HDAC6 KO
1.2
***
WT
1

ELK1 Ser383 phosphorylation

IB anti-pELK1Ser383 pELK1S383

Relative intensity of
0.8

IB anti-ELK1 ELK1 In vitro kinase assay 0.6

IB anti-ERK1/2 GST-ERK1 0.4

IB anti-HDAC6 HDAC6 0.2

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

IB anti-Ac--tubulin Ac--tubulin 0
Whole cell lysate
WT HDAC6
KO
IB anti-α-tubulin -tubulin
(n=5)
1 2

B 1.2
GST-ERK1 **
1
ELK1 Ser383 phosphorylation

Rec CBP ‐ +
Relative intensity of

IB anti-pELK1Ser383 pELK1S383 0.8

In vitro kinase assay
IB anti-ELK1 ELK1 0.6

0.4
IB anti-AcK Acetyl-GST-ERK1
In vitro acetylation
IB anti-ERK1/2 GST-ERK1 0.2

1 2 0
Control CBP
(n=5)
 Wu et al. Figure 11

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

B

C K72
H.sapiens YGMVSSAYDHVRKTRVAIKKISPFEHQTYCQRTLRE 88
P.troglodytes YGMVSSAYDHVRKTRVAIKKISPFEHQTYCQRTLRE 88
M.mulatta YGMVSSAYDHVRKTRVAIKKISPFEHQTYCQRTLRE 74
B.taurus YGMVSSAYDHVRKTRVAIKKISPFEHQTYCQRTLRE 71
S.scrofa YGMVSSAYDHVRKTRVAIKKISPFEHQTYCQRTLRE 89
R.norvegicus YGMVSSAYDHVRKTRVAIKKISPFEHQTYCQRTLRE 89
M.musculus YGMVSSAYDHVRKTRVAIKKISPFEHQTYCQRTLRE 88
C.lupus YGMVSSAYDHVRKVRVAIKKISPFEHQTYCQRTLRE 89
M.putorius YGMVSSAYDHVRKVRVAIKKISPFEHQTYCQRTLRE 80
O.hannah YGMVCSAYDHVNKIRAAIKKISPFEHQTYCQRTLRE 134
D.rerio YGMVCSAFDNVNKIRVAIKKISPFEHQTYCQRTLRE 102
X.laevis YGMVCSAHDNVNKVRVAIKKISPFEHQTYCQRTLRE 74
D.melanogaster YGMVVSADDTLTNQRVAIKKISPFEHQTYCQRTLRE 84
C.elegans YGMVASALDTITRDRVAIKKISPFEHQTFCQRTLRE 74
cons **** ** * : . *.************:*******
 Wu et al. Figure 12

C ns
A 4
*

Relative luciferase activity

3.5
GST-ERK1 3 ***
DN WT K72Q K72R 2.5
IB Anti-pERK pGST-ERK1 2
1.5
IB anti-ERK1/2 GST-ERK1 1

1 2 3 4 0.5
0
1 2 3 4
IB anti-ELK1 ELK1

B IB anti-HA HA-MEK (S218/222D)

GST-ERK1-K72Q

GST-ERK1-K72R
GST-ERK1-WT
GST-DN ERK1

IB anti-ERK1/2 GST-ERK1WT/K72Q/K72R

Downloaded from http://www.jbc.org/ by guest on December 21, 2017

IB anti--actin -actin

(ELK1)2-TATA-Luc + + + +

IB anti-pELK1S383 pELK1S383 pCMV-ELK1 + + + +

HA-MEK(S218/222D) - + + +

IB anti-ERK1/2 GST-ERK1 GST-ERK1-WT - + - -

1 2 3 4 GST-ERK1-K72Q - - + -

GST-ERK1-K72R - - - +

D
 Histone deacetylase 6 (HDAC6) deacetylates extracellular signal-regulated kinase 1
(ERK1) and thereby stimulates ERK1 activity
Jheng-Yu Wu, Shengyan Xiang, Mu Zhang, Bin Fang, He Huang, Oh Kwang Kwon,
Yingming Zhao, Zhe Yang, Wenlong Bai, Gerold Bepler and Xiaohong Mary Zhang
J. Biol. Chem. published online December 19, 2017

Access the most updated version of this article at doi: 10.1074/jbc.M117.795955

Alerts:
• When this article is cited
• When a correction for this article is posted

Click here to choose from all of JBC's e-mail alerts

Downloaded from http://www.jbc.org/ by guest on December 21, 2017