And C-Terminal Domains of Src Regulate Distinct Roles Of

Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Acetylation within the N- and C-terminal domains of Src regulate distinct roles of

STAT3-mediated tumorigenesis

Chao Huang1, 3, 6 *, Zhe Zhang2, 6, Lihan Chen3, 6, Hank W. Lee3, Marina K. Ayrapetov4,

Ting C. Zhao4, Yimei Hao3, Jinsong Gao4, Chunzhang Yang5, Gautam U. Mehta5,

Zhengping Zhuang5, Xiaoren Zhang3, Guohong Hu3, and Y. Eugene Chin1, 3 *

1 Translation Medicine Center, Shanghai Chest Hospital, Shanghai Jiao Tong

University, Shanghai 200025, China

2 Department of Surgery, First Affiliated Hospital, Zhejiang University School of

Medicine, Hangzhou, Zhejiang 310003, China

3 Institute of Health Sciences, Chinese Academy of Sciences and Shanghai Jiaotong

University School of Medicine, 320 Yueyang Road, Shanghai 200031, China

4 Departments of Surgery and Medicine, Brown University School of Medicine-Rhode

Island Hospital, 593 Eddy Street, Providence, Rhode Island 02903 USA

5 Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke,

NIH, Bethesda, Maryland 20892, USA

6 Co-first authors

* Correspondence: [email protected] and

[email protected]

Running tittle: Regulation of Src-STAT3 signaling pathway by acetylation Keywords: c-Src; STAT3; acetylation; EGF; tumorigenesis Grant Support: This work was supported by grants from the National Natural Science

Foundation of China (81230059 to Y.E.Chin, 81672804 to C.Huang, 81402097 to

Z.Zhang), Ministry of Science and Technology of China (2015CB910402 to C.Huang).

Disclosure of Potential Conflicts of Interest

Downloaded from cancerres.aacrjournals.org on September1 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

No potential conflicts of interest were disclosed. Significance

CBP-mediated acetylation of lysine clusters in both the N-terminal and C-terminal

regions of c-Src provides additional levels of control over STAT3 transcriptional

activity

Downloaded from cancerres.aacrjournals.org on September2 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Abstract

Post-translational modifications of mammalian c-Src N-terminal and C-terminal

domains regulate distinct functions: myristoylation of G2 controls its cell membrane

association and phosphorylation of Y419/Y527 controls its activation or inactivation,

respectively. We provide evidence that Src-cell membrane association-dissociation and

catalytic activation-inactivation are both regulated by acetylation. In EGF-treated cells,

CREB binding protein (CBP) acetylated an N-terminal lysine cluster (K5, K7, and K9)

of c-Src to promote dissociation from the cell membrane. CBP also acetylated the C-

terminal K401, K423, and K427 of c-Src to activate intrinsic kinase activity for STAT3

recruitment and activation. N-terminal domain phosphorylation (Y14, Y45, and Y68) of

STAT3 by c-Src activated transcriptionally active dimers of STAT3. Moreover, acetyl-

Src translocated into nuclei where it formed the Src-STAT3 enhanceosome for gene

regulation and cancer cell proliferation. Thus, c-Src acetylation in the N-terminal and

C-terminal domains play distinct roles in Src activity and regulation.

Introduction

Src family kinases are cell membrane-associated non-receptor protein tyrosine kinases

that coordinate various cellular events including differentiation, adhesion, and

migration. Aberrant Src kinase activity has been widely implicated in cancer

development, progression and metastasis (1, 2). Functionally, Src has N-terminal and

C-terminal regions of similar size. The human c-Src N-terminal region is comprised of

a unique domain (UD), an SH3 domain, and an SH2 domain while the C-terminal

region contains a catalytic domain (CD) followed by a regulatory tail (2). Src

Downloaded from cancerres.aacrjournals.org on September3 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

associates with the cytoplasmic membrane, which depends on myristoylation of the N-

terminal glycine residue (G2) mediated by N-myristoyltransferase (Nmt) and positively

charged residues within UD that interact with the overall negatively charged cell

membrane (3). Cell membrane association is required for cell transformation, whereas

dissociation permits Src to be targeted to other cellular compartments (4, 5). c-Src

catalytic activity has been detected in the cytoplasm but also in the endoplasmic

reticulum (ER), mitochondria and nucleus (6-8). c-Src is inhibited by the C-terminal

Src kinase (Csk) that phosphorylates Y527 (in chicken) or Y530 (in human) and renders

Src catalytically inactive as a result of intramolecular interactions between the C-

terminal tail and the SH2 domain (9).

c-Src catalytic activation needs conformational changes within the catalytic domain:

the orientation of two lobes of catalytic domain; rearrangement of the activation loop;

and movement of the helical interface between the two lobes into the catalytic cleft.

Global phosphorylation of c-Src may cause this conformational change and augment its

kinase activity (10). In cancer cells, c-Src is activated by growth factors, and leads to

phosphorylation and activation of oncogenic substrates including STAT3 (11).

However, it remains unclear how c-Src regulates down-stream targets in response to

growth factor treatment.

In this study, we report that c-Src global phosphorylation is induced by CBP, an

acetyltransferase in the cells. We observe that after EGF stimulation, unique domain

and catalytic domain of c-Src are acetylated by CBP, which results in c-Src dissociation

from plasma membrane and recruitment and activation of STAT3. Importantly, c-Src

forms a protein kinase-transcription factor enhanceosome with STAT3 in the nuclei

where c-Src promotes STAT3 activity to up-regulate oncogene expression in cancer

cells.

Downloaded from cancerres.aacrjournals.org on September4 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Materials and Methods

Cell Culture and Reagents

NIH3T3, HEK293T and MCF-7 cells were purchased from ATCC (Oct, 2010), Src-/-

SYF and control (Src introduced) SYF cells from DC Flynn (Apr, 2002), U3A cells

from and GR Stark (Feb, 1996), and cbp-/- MEF were cultured in DMEM medium

containing 10% fetal bovine serum, 2 μM glutamine, and 100 U/ml each of penicillin

and streptomycin by Invitrogen (Carlsbad, CA). MCF-7 and U3A cells were

authenticated using short tandem repeat (STR) profiling analysis by Suzhou Genetic

Testing Biotechnology Co. Ltd. HEK293T was STR-authenticated by Shanghai

Biowing Applied Biotechnology Co. Ltd. NIH3T3, SYF and MEF cells were

authenticated as mouse origin by Shanghai Biowing Applied Biotechnology Co. Ltd.

All cells were tested for mycoplasma contamination by using GMyc-PCR Mycoplasma

Test kit (Yeasen BioTech, Shanghai) and were maintained at 37ºC in a humidified 5%

CO2 atmosphere. Antibodies used were anti-STAT3 (cat# sc-8019, sc-482), anti-c-Src

(cat# sc-5266), anti-HA, anti-Myc, anti-GFP from Santa Cruz Biotech (Santa Cruz, CA);

anti-c-Src (cat# 2109) from Cell Signaling; Anti-Src (phospho Y416) antibody from

Abcam (cat# ab185617) and Merck Millipore (cat# 04-857); anti-Na+/K+ ATPase (cat#

ab167390), anti-IkB (cat# ab7547), anti-H3 (cat# ab1791), anti-pan acetylated lysine

(cat# 22550) from Abcam; anti-GST, and anti-FLAG from Sigma-Aldrich (St. Louis,

MO); 4G10 from Millipore (Billerica, MA); anti-pan acetylated lysine (cat# 9441) and

anti-pY705-STAT3 (cat# 9145) from Cell Signaling (Boston, MA). Rabbit polyclonal

antibodies against acetyl-Src (aK7, aK9, aK423, and aK427) and phospho-STAT3

(pY45 and pY68) were custom prepared by Millipore or AB-land, Inc. (Hangzhou,

China). Polyclonal antibodies are produced by immunizing animals with synthetic

Downloaded from cancerres.aacrjournals.org on September5 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

acetyl peptides corresponding to residues surrounding acetylated K7, acetylated K9,

acetylated K423, or acetylated K427 of Src and synthetic phosphopeptides

corresponding to residues surrounding Y45 or Y68 of Stat3. Antibodies were purified

by protein A and peptide affinity chromatography. GST or GST tagged Src protein was

purchased from Abcam. Recombinant human EGF was purchased from Sigma-Aldrich;

recombinant mouse EGF, human OSM, and mouse OSM were purchased from R&D

Systems (Minneapolis, MN). For convenience reason, chicken c-Src protein sequence

was used in the context, i.e., K401, K423, K427, Y416 and Y527 should be K404,

K426, K430, Y419, and Y530 of human Src correspondingly.

Preparation of Subcellular Protein Fractions, Preparation of Flag-STAT3, Myc-

Src and HA-CBP proteins

(i) Subcellular protein fractions: Cell membrane, cytosolic and nuclear fractions were

prepared from CBP-/- or control MEF cells. Membrane fraction was prepared by Mem-

Per Eukaryotic Membrane Protein Extraction Reagent Kit from PIERCE; cytosolic and

nuclear fractions were prepared by Nuclei EZ prep Nuclei Isolation Kit from Sigma. (ii)

Flag-STAT3 and Myc-Src proteins: Flag-STAT3 and Myc-Src with or without HA-CBP

were transiently transfected into HEK293T cells using Lipofectamine 2000 (Invitrogen)

according to the manufacturer's protocol. Whole cell lysates were prepared 36 hrs post

transfection in lysis buffer (50 mM Tris-HCl, pH 7.8, 137 mM NaCl, 1mM NaF, 1mM

NaVO3, 1% Triton X-100, 0.2% Sarkosyl, 1 mM dithiothreitol, and 10% glycerol) and

incubated with anti-FLAG (M2, Sigma-Aldrich Co) or anti-Myc (9E10, Santa Cruz

Biotech) for 2 hrs at 4°C. Protein G Agarose beads were then added and incubated at

4°C O/N. After washing with lysis buffer 3 times and elution buffer (50 mM Tris-HCl

pH7.9, 100mM NaCl, 20% glycerol and 0.1% NP-40) once, STAT3 or Src was eluted

Downloaded from cancerres.aacrjournals.org on September6 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

from the beads by incubating at 4°C for 120 min with 100 l elution buffer containing

FLAG or Myc peptide (100 μg/ml). Purified Flag-tagged STAT3, Myc-tagged Src or

acetyl-Src proteins were then verified by Western Blot with anti-Flag, anti-Myc and

anti-pan-acetylated lysine. Amicon Ultra-0.5 mL Centrifugal Filters for Protein

Purification and Concentration, 100 kDa (Millipore) was used to remove HA-CBP (280

KDa) from acetylated Myc-Src (75 KDa) and confirmed by Western Blot against HA-

tag; HA-CBP protein: HA-CBP was transiently transfected into HEK293T and protein

was purified according to routine procedures.

Constructing Src and STAT3 Mutants

(i) Src constructs: Src was subcloned into the EcoRI-BamHI sites of pcDNA 3.1 vector

with or without Myc or HA, or Flag tag. The Src K R mutants (K5R/N, K7R/N,

K37R/N, K59R/N, K401R/N, K423R/N, and K427R/N) as well as Src-R175A, Src-

K200E, Src-K295M were constructed by performing site-directed mutagenesis. Src

domain deletion constructs were generated by PCR and cloned into the HA-tagged,

Myc-tagged, Flag-tagged or GST-tagged vector. Full-length Src-YFP and Src-UD-YFP

containing the N-terminal unique domain were generated by PCR and cloned into the

EcoRI-BamHI sites of pEYFP-N1 vector. Src-YFP mutants (K5R, K5N, K7R, K7N,

K9R, K9N, K37R, K37N, K59R, or K59N were constructed by site-directed

mutagenesis. (ii) STAT3 constructs: Myc-STAT3 mutants including Y F mutation

(Y14F, Y22F, Y45F, Y68F, Y79F, Y94F, and Y705F) were constructed by site-directed

mutagenesis based on STAT3 full length or STAT3-ND (1-155) backbone. Flag-tagged

STAT3 domain deletion constructs were created by PCR and cloned into pcDNA3.1

vector. GST-STAT3 full length and GST-STAT3 SH2 domain were cloned into PGEX-

4T-3 vector. All mutations were confirmed by DNA sequencing. All other constructs

Downloaded from cancerres.aacrjournals.org on September7 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

used in this work including CBP domain deletion cDNA and 2xSIE (IRF1 promoter

STAT3 binding site)-luciferase reporter were described previously (12).

RNAi Transfection, BrdU Incorporation, Myristoylation Assay, and Luciferase

Reporter Assay

(i) RNAi transfection: The siRNA targeting mouse or human CBP and control siRNA

(Santa Cruz) were transiently transfected into mammalian cells as previously described

using Lipofectamine 2000 according to the manufacturer's protocol (Invitrogen) and our

published protocol (12). (ii) STAT3 in vitro transcription assay: The promoter region of

BCL-XL gene, which contains two STAT3 binding sites, was amplified by PCR with

forward primer 5’-CCTGCCTGCCTTTGCCTAA and reverse primer 5’-

TCTCCCTCCTTCTGGCAC. The amplicon was quantified and verified by sequencing.

(iii) BrdU incorporation assay: SYF cells were transiently transfected with STAT3

along with EV or indicated form of Src (at 1:3 STAT3 to Src ratio) using Lipofectamine

2000 (Invitrogen). Twenty-four hrs post transfection cells were serum deprived (0.5%

FBS in DMEM) for 24 hrs and then cells were cultured in fresh media containing 10%

FBS and 100 ng/ml mouse EGF for additional 20 hrs. Cell proliferation was measured

using BrdU Cell Proliferation kit (Millipore) according to manufacturer’s protocol.

BrdU absorbance was measured in 96-well plates using dual wavelength of 450/540nm.

(iv) Src myristoylation assay: Src was transfected alone or along with CBP into HeLa

cells for 24 hrs followed by metabolic labeling with 3H-myristic acid (54 Ci/mmol, 10

Ci/mL, Perkin Elmer) for 3 hrs. Immunopurified Src was washed extensively prior to

elution and 3H incorporation was measured by scintillation counting (Perkin Elmer).

Numeric reading was plotted using GraphPad Prism 5.0 software. (v) Luciferase

reporter assay: Cells were transfected with 0.5 μg of the STAT3 responsive 2xSIE

Downloaded from cancerres.aacrjournals.org on September8 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

(IRF1 promoter)-luciferase reporter, 10 ng Renilla expression vector (internal control),

and 1.5 μg of STAT3 and EGFR or Src and STAT3 and CBP, or control empty vector

pcDNA3 (Invitrogen). 24 hrs post transfection cells were treated with EGF for 6 hrs.

Cell lysates were prepared using Luciferase Assay kit (Promega) and luciferase activity

was measured using a luminometer. Relative luciferase activity was corrected for

tranfection efficienty using Renilla expression vector.

In vitro Binding Assay, in vitro Tyrosine Kinase Assay, in vitro Transcription

Assay and in vitro acetylation assay

(i) In vitro binding assay: A total of 1 μg GST-STAT3-FL, GST-STAT3-SH2 domain

fusion proteins and GST protein were purified from bacteria and applied to glutathione

Sepharose 4B resins (GE Health) and incubated 1 hr at 4°C. The resin was washed and

mixed with 0.2 mg cellular extracts prepared from HEK293T cells then incubated for 2

hrs at 4°C. After washing, GST-STAT3-FL or GST-STAT3-SH2 domain bound

proteins were eluted and subjected to SDS-PAGE, followed by immunoblotting. (ii) In

vitro tyrosine kinase assay: Myc-Src was transfected with or without CBP in 293T cells.

Anti-Myc immunoprecipitated Src was then eluted with Myc peptide and confirmed for

its purity with Coomassie blue staining. Similarly, STAT3-HA was purified with anti-

HA and eluted with HA peptide. Acetylated or unacetylated Src, were incubated with

purified STAT3-HA in the in vitro phosphorylation system (25 mM Tris-HCl, pH7.5, 5

mM - glycerophosphate, 2 mM DTT, 0.1 mM Na3VO4, 10 mM MgCl2, 0.2 mM ATP)

for 30 min. Reaction was stopped by 5 min incubation at 95°C and equal amount were

analyzed in Western blot. (iii) In vitro transcription assay: 8 units of HeLaScribe

Nuclear Extract, 400 ng BCL-xl PCR product, 100 ng STAT3, and 200 ng Src or

acetylated Src were mixed in the reaction buffer containing 1 l [α-32P]ATP (3,000

Downloaded from cancerres.aacrjournals.org on September9 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Ci/mmol, 10 mCi/ml), 1 μl 25X dNTP Mix, and 1.5μl 50 mM MgCl2 in a total volume

of 25 l. After incubation at 30°C for 60 min, the reaction was abolished by adding 175

μl stop solution. RNA synthesized in the reaction were then extracted and heated at

90°C for 10 min, and then separated in 4% denaturing SDS-PAGE gel, followed by X-

ray film exposure at -80°C for 48 hrs. (iv) In vitro acetylation assay: GST and GST-

Src were purchased from Abcam, HA-CBP was purified from transfected 293T cells.

Ac-CoA was purchased from Sigma. In vitro acetylation reaction was performed

according to previously described (13).

Immunofluorescent Microscopy

Cells were fixed and stained with fluorescein isothiocyanate (FITC)-conjugated

monoclonal anti-Src or anti-CBP (Santa Cruz Biotech) as described previously (12).

Image acquisition was conducted on an epifluorescence microscope (TE2000-U

microscope; Nikon), using a 60× oil objective coupled to a CoolSNAP HQ CCD (Roper

Scientific).

Immunohistochemical staining

Src acetylation and STAT3 phosphorylation were evaluated in sections from formalin-

fixed and paraffin-embedded breast cancer tissue using a standardized method. In brief,

tissue samples were fixed in 4% paraformaldehyde (24 h), processed and embedded in

paraffin. Sections (5 μm) were prepared and mounted on coverslips. Sections were

stained with various anti-acetyl Src or anti-phospho STAT3 antibody with or without

competition peptieds and counterstained with hematoxylin.

Tandem Mass Spectrometry Analysis of Src

Downloaded from cancerres.aacrjournals.org on September10 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Src was prepared from 293T co-expressing CBP or treated with EGF was

immunoprecipited and resolved by 10% SDS-PAGE. A Coomassie blue-stained band

corresponding to Src was excised, divided in half, reduced with dithiothreitol, alkylated

with iodoacetamide, and digested with either trypsin or chymotrypsin for mass

spectrometry analysis performed by Taplin Biological Mass Spectrometry Facility

(Boston, MA). Peptide mixtures were separated by microcapillary reverse-phase

chromatography and analyzed online in a hybrid linear ion trap-Orbitrap (LTQOrbitrap;

Thermo Electron) mass spectrometer . Mass spectra were data base-searched using the

SEQUEST algorithm. All peptide matches were initially filtered based on enzyme

specificity, mass measurement error, XCorr and DCorr scores and further manually

validated for peptide identification and site localization.

Circular dischroism (CD) analysis

Src-Flag full length and Src-CD-Flag were purified from transfected 293T cells and

were diluted to 200 g/ml in HEPES buffer, pH7.2. An aliquot (1.5 mL) of this stock

was centrifuged at 16,000 g for 30 min to remove any particulate matter. We applied

circular dichroismina JascoJ-715 spectropolarimeter at 22°C and protein secondary

structure determination was performed by scanning purified Src FL and Src CD from

260 to 190 nm in a 2 mm quartz cell.

3D alignment of acitive and inactive c-Src

Current alignment tool is RCSB PDB-Structure Alignment using JFATCAT-flexible

comparison method. RCSB PDB Comparison and Algorithm has been described

previously (14, 15)

Downloaded from cancerres.aacrjournals.org on September11 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Tumor Xenografts

Six-week-old male Balb/c nu/nu mice were injected with 2×106 Src-transformed

NIH3T3 cells. Tumor volume was measured every 3 days and calculated as 0.5×a2×b,

where a is the width and b is the length of the tumor. Tumors were excised and

weighted when the mice were sacrificed 24 days after injection. Mice under endpoint

criteria were excluded from analysis in long-term experiments.

Colony Forming Assay

1×103 NIH 3T3 cells transfected with v-Src-HA or v-Src-HA KR variants were

suspended in 0.5 ml of 0.35% agar in DMEM containing 10% FBS, and seeded in each

well of 24-well plate pre-solidified 0.6% agar. DMEM with 10% FBS was gently

layered over the cultures every 4 days for 14 days. The colonies were counted and

photographed under an inverted microscope.

For experiment using v-Src-HA KN variants, 1×103 NIH 3T3 cells transfected with v-

Src-HA or v-Src-HA KN variants were seeded in plates. After incubation for 10–14

days, cells were washed with PBS twice, fixed with methanol for 15 min, and stained

with 0.5% crystal violet for 15 min at room temperature. Visible colonies were counted.

Colony formation rate = (number of colonies/number of seeded cells) × 100%.

Statistical analyses

Student’s t-test (paired, two-tailed) and Mann–Whitney test were used as indicated in

the figure legends.

Study approval

Downloaded from cancerres.aacrjournals.org on September12 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

All mice experiments were handled according to the guidelines for the care and use of

laboratory animals and were approved by the institutional biomedical research ethics

committee of Shanghai Institutes for Biological Sciences.

Results

EGF-induced Src acetylation by CBP

To study Src activation by growth factors, we examined EGF-dependent c-Src

posttranslational modifications. Among all those putative posttranslational

modifications (16, 17), we noticed that human c-Src acetylation was induced by EGF in

a time-dependent manner in MCF-7 cells, which was further enhanced by pretreatment

of the cells with histone deacetylase (HDAC) inhibitor trichostatin A (TSA) or Sirt

deacetylase inhibitor nicotinamide (NAM) (Fig. 1A). c-Src Tyr 419 phosphorylation

was also increased upon EGF and deacetylase inhibitor treatment (Fig. 1A). EGF can

trigger cAMP-response element-binding protein (CREB)-binding protein (CBP), which

initially was found to localize in nucleoplasm, to shuttle from the nucleus to the

cytoplasm to acetylate EGF receptor (18). We and others have also shown that CBP

can shuttle between cytoplasm and nuclear in quiescent cells or in stimulated cells to

interact with cytoplasmic proteins or membrane receptors (19, 20). Therefore, we set

out to investigate the feasibility of Src acetylation by CBP. To do this, we used small

interfering RNA (siRNA) to deplete endogenous CBP and found that EGF-induced c-

Src acetylation was largely abolished by CBP depletion (Fig. 1A). We performed in

vitro acetylation assay by incubating purified GST-Src protein with purified CBP

protein in the presence of acetyl-coA and c-Src was found to be acetylated by CBP

(Supplementary Fig. S1A, left panel).

Downloaded from cancerres.aacrjournals.org on September13 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

In MCF-7 cells, CBP and c-Src formed a complex upon EGF stimulation (Fig. 1B).

Immunostaining and cell fraction assay further showed that CBP-Src complex was

formed at plasma membrane upon EGF treatment (Supplementary Fig. S1B and S1C).

Coimmunoprecipitation (co-IP) experiments using domain deletion constructs of c-Src

and CBP (Supplementary Fig. S1D and S1E) revealed that deletion of either the c-Src

UD (Unique Domain) or the CD (Catalytic Domain) reduced Src binding with CBP

(Fig. 1C), suggesting both UD and CD contribute to the Src-CBP complex formation.

It appeared that the interaction was independent of the c-Src catalytic activity as Src

with K295M mutation (kinase inactive) (21) retained a high binding affinity for CBP

(Supplementary Fig. S1F). c-Src K200E mutation, which can functionally cause the

Src SH2 defect (22), largely impaired Src-CBP complex formation (Supplementary Fig.

S1F). The region from 1099 to 1620 of CBP on the other hand, was important for CBP

to form complex with c-Src (Fig. 1D). However, CBP 1-1799 did not bind to Src,

which may due to CBP 1-777 that interfere with binding between CBP and Src, as

another truncate CBP 775-1779 that do not contain this segment (1-777) binds to Src

strongly (Fig. 1D). These results suggest that the CBP Bromo-CH2 domain (1099-1620)

interact with unique domain and catalytic domain of c-Src to form Src-CBP complex in

cells.

Src acetylation occurs within both unique domain and catalytic domain

c-Src protein sequence analysis revealed multiple histone-type lysine motifs or

lysine clusters within both unique domain and the catalytic domain as potential CBP

acetylation sites (Fig. 1E and Supplementary Table S1). The catalytic domain truncated

c-Src with lysine to arginine substitution of residue K5R, K7R, K9R, K37R, or K59R,

individually or collectively, reduced the acetylation intensity when co-transfected with

CBP (Fig. 1F and Supplementary Fig. S1G). c-Src catalytic domain with K401R or

Downloaded from cancerres.aacrjournals.org on September14 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

K427R substitution markedly abolished the acetylation intensity (Fig. 1G), which was

consistent with a previous report that some acetylation sites can be mutually affected

(12). We also performed an in vitro acetylation assay by comparing wild type Src with

all acetyl-lysine (K5,7,9,37,59,401,423,427R) mutated Src (Src-KR mutant) for CBP-

mediated acetylation. The results in Supplementary Fig. S1A suggested that these are

the major acetylation sites of c-Src (Supplementary Fig. S1A, right panel). To clarify

Src acetylation in vivo, we generated polyclonal antibodies that specifically recognize

individual acetylation sites of c-Src (Supplementary Fig. S2A and S2B). While K7

acetylation was CBP-dependent, K427 acetylation was only slightly affected by CBP in

293T as detected by specific antibodies (Fig. 1H). In MCF-7 cells, c-Src acetylation on

K7, K9, K423 and K427 was indeed induced by EGF in a time-dependent manner (Fig.

1I). Thus, EGF induces CBP-dependent c-Src acetylation within both the N-terminal

unique domain and the C-terminal catalytic domain.

Acetylation triggers Src global phosphorylation modification

It is well-established that c-Src catalytic activity is tightly regulated by global

phosphorylation and autophosphorylation modifications within all four domains (23).

To determine if CBP-mediated c-Src acetylation affects Src phosphorylation and/or

autophosphorylation, chicken c-Src was cotransfected with CBP in the cells. Over-

expression of CBP increased Src global tyrosine phosphorylation including

phosphorylation at Y416 in 293T cells (Fig. 2A). In contrast, CBP deficiency abolished

both acetylation and tyrosine phosphorylation of c-Src (Fig. 2B). To identify c-Src

phosphorylation sites affected by CBP, we purified chicken c-Src from 293T cells

cotransfected with or without CBP for differential mass spectrometry analysis.

Strikingly, 15 tryptic phosphopeptides (9 pTyr, 5 pSer and 1 pThr) were identified from

c-Src purified from the cells cotransfected with CBP, whereas only 3 tryptic

Downloaded from cancerres.aacrjournals.org on September15 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

phosphotyrosine peptides were identified from only c-Src transfected cells (Fig. 2C;

Supplementary Fig. S3A, S3B and Supplementary Table S2). Tryptic peptides

containing acetylated lysine residues including K7, K9, K37, K59, and K423 were all

recovered from c-Src purified from c-Src and CBP co-transfected cells (Fig. 2C and

Supplementary Table S2).

Interestingly, the phosphoserine and phosphothreonine sites are clustered in all

three regulatory domains, whereas the phosphotyrosine sites are enriched only in the

catalytic and the adjacent SH2 domain in chicken c-Src (Fig. 2C and Supplementary

Table S1). Therefore, overexpressed CBP may not only affect Src activity in Src

autophosphorylation but also other protein kinases that responsible for Src

phopshorylation.

Acetylation of Src unique domain dissociates Src from cell membrane

Dynamic association with plasma membrane is an important property of Src (1).

To delineate acetylation on Src-plasma membrane association regulation, we examined

c-Src acetylation in NIH3T3 cells. c-Src acetylation was induced either by CBP

overexpression or treatment with TSA or NAM (Fig. 3A and Supplemental Fig. S4A).

We then compared c-Src-plasma membrane association in wild type versus CBP-/-

MEF cells. In wild-type MEF cells, EGF induced c-Src-plasma membrane association

at 15 min (Fig. 3B), which is reasonable as Src can be recruited by EGFR upon EGF

stimulation (24, 25). Src-plasma membrane association was then gradually decreased

in a time-dependent manner up to 120 min of EGF treatment (Fig. 3B). Pretreatment of

the MEF cells with deacetylase inhibitors decreased c-Src-plasma membrane

association steadily (Fig. 3B). Plasma membrane marker Na+/K+ ATPase was used as

control. Indeed, in CBP-/- MEF cells, c-Src was constitutively associated with the

Downloaded from cancerres.aacrjournals.org on September16 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

plasma membrane and no dissociation of Src occurred in response to EGF and

deacetylase inhibitors treatment (Fig. 3B).

To confirm the role of c-Src unique domain lysine cluster acetylation on Src-

plasma membrane dissociation, we fractionated c-Src from three cell compartments:

plasma membrane, cytoplasm and nuclei. While c-Src from both cytoplasmic and

nuclear fractions was strong in K7 acetylation, Src from plasma membrane fraction was

very weak in K7 acetylation (Fig. 3C). A quantification analysis revealed that acetyl-

K7 c-Src accounted for only 20% of the total Src in the plasma membrane fraction but

over 80% in either the cytoplasmic or nuclear fraction (Supplementary Fig. S4B).

We also constructed YFP-tagged wild type full-length chicken c-Src (Src-YFP) and

c-Src with N-terminal amino acids 1-83 (Src-UD-YFP) and various c-Src unique

domain lysine to arginine or lysine to asparagine (KR or KN) substitution mutants to

visualize localization of Src at plasma membrane or cytoplasmic. Lysine to neutral

asparagine can mimic acetylation and reduce electrostatic interaction between amino-

terminal of Src and cell membrane (3). Like endogenous c-Src, wild type Src-YFP

fusion protein responded normally to EGF stimulation, suggesting YFP fusion protein

did not affect Src activity (Supplementary Fig. S4C and S4D). While full length Src-

YFP and Src-UD-YFP responded to EGF for plasma membrane dissociation as

observed in 293T cells or wild type MEF cells (Fig. 3D and Supplementary Fig. S4D),

Src-UD-YFP failed to respond to EGF for plasma membrane dissociation in CBP-/-

MEF cells (Supplementary Fig. S4D). We then examined the effects of c-Src unique

domain lysine cluster acetylation on Src membrane dissociation. Src-UD-YFP with KR

substitution of individual acetyl-lysines (K5R, K7R, K9R, K37R, or K59R) showed no

apparent effects on plasma membrane association and dissociation (Fig. 3D). In

contrast, Src-UD-YFP with K7N mutation became cytoplasmic orientated whereas Src-

Downloaded from cancerres.aacrjournals.org on September17 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

UD-YFP with K5N or K9N mutation became both cellular and nuclear membrane

orientated (Fig. 3D) (26). Similar results were obtained by using full length Src-YFP

(Supplementary Fig. S4E). Interestingly, EGF treatment induced nuclear accumulation

of Src mutants (K5N, K9N), but not Src mutants (K5R, K7R, K9R) (Supplementary Fig.

S4E). These results indicate that while all three acetyl-lysine residues of the lysine

cluster (K5, K7, K9) contribute to the Src-plasma membrane association and

dissociation, acetylation on K7 plays the most important role in Src dissociation from

plasma membrane.

Since N-terminal G2 myristoylation is critical for signaling Src to plasma

membrane (27), we questioned whether K7 acetylation will affect G2 myristoylation.

Myristoylation prediction and the 3H-incorporation experiment indicated that c-Src G2

myristoylation was not affected by c-Src K7N substitution or overexpression of CBP in

Hela cells (Fig. 3E and Supplementary Table S3).

We next delineated CBP subcellular distribution upon growth factor treatment. In

NIH3T3 fibroblasts treated with either EGF or oncostatin M (OSM), cytoplasmic CBP

signal was greatly enhanced (Fig. 3F, left panel), which may acetylate c-Src upon EGF

stimulation. Hence, cytoplasmic CBP is likely responsible for Src acetylation in

cytoplasm.

When expressed in Src-/- null SYF cells, c-Src with K7R, K9R, or K5R/K7R/K9R

substitution failed to translocate to the nucleus in response to EGF (Fig. 3F, right panel).

On the other hand, c-Src lacking the unique domain (Src-ΔUD) was nuclear localized

(Fig. 3F, right panel). To address whether nuclear accumulation of Src was CBP-

dependent, we examined c-Src in wild type and CBP-/- MEF cells. In wild type MEF

cells, Src nuclear accumulation increased upon EGF treatment (Fig. 3G). Treatment of

wild type MEF cells with EGF, OSM or EGF plus TSA or NAM for 60 min led to

Downloaded from cancerres.aacrjournals.org on September18 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

increased Src nuclear accumulation, although Src-plasma membrane dissociation was

not observed by Western blot (Fig. 3G). In contrast, Src failed to respond to similar

treatment in CBP-/- MEF cells (Fig. 3G). Next, CBP lacking nuclear localization signal

(CBP-ΔNLS) was transfected into CBP-/- MEF cells. By using CBP ΔNLS we can

exclude Src acetylation in the nucleoplasm and enhance the chance for CBP-Src

interaction in the cytoplasm/plasma membrane (19). We found that active form of c-

Src was accumulated in the nucleus in response to EGF treatment (Fig. 3H). These

results revealed that nuclear accumulation of Src was dependent on CBP-mediated Src

acetylation.

Modular interaction between the acetylated Src catalytic domain and the STAT3

SH2 domain

K401, K423, and K427 of the catalytic loop, activation loop, and p+1 loop of the

chicken c-Src catalytic domain form an isosceles triangle according to c-Src structural

analysis (28, 29). A 3D alignment analysis reveals the side chain of c-Src K423 will

alter its orientation between active and inactive Src (Fig. 4A). c-Src primarily

associated with STAT3 C-terminal region (585-770) (Fig. 4B). Reciprocally we

observed that interaction between STAT3 and c-Src gradually increased as the c-Src

SH2 domain and the SH3 domain were sequentially added back (Fig. 4C).

c-Src with K427R rather than K401R or K423R mutation failed to recruit STAT3

(Fig. 4D). Consistently, the interaction between Src and STAT3 was disrupted by

incubation with Src K427 acetyl-peptide but not with the control peptide (Fig. 4E),

indicating that acetyl-K427 of Src catalytic domain (CD) p+1 loop can directly bind to

the STAT3 SH2-domain (30). Therefore, Src SH2-CD interacted with both CBP

Bromo-CH2 domain (Fig. 1C) and STAT3 SH2-TA domain (Fig. 4B and 4C). Our in

vitro GST-pull down assay confirmed a direct protein-protein interaction between Src

Downloaded from cancerres.aacrjournals.org on September19 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

and the STAT3-SH2 domain (Fig. 4F), agreed with the notion that the extended

STAT3 SH2 domain is necessary for STAT3 homodimerization or interaction with

other proteins (31).

Catalytic domain-acetylated Src phosphorylates STAT3 on multiple N-terminal

tyrosines

Acetylation on c-Src catalytic activity regulation was further delineated by

performing an in vitro kinase assay. Acetyl-Src, prepared from 293T cells transfected

with chicken c-Src and CBP, showed higher activity and was much more efficient than

Src prepared from 293T cells transfected with Src alone in STAT3 phosphorylation

induction in a dose-dependent manner in the in vitro system (Fig. 5A). c-Src, either

full length or the catalytic domain only, became more active in phosphorylation of

STAT3 when CBP was cotransfected (Fig. 5B). c-Src catalytic domain and Src full

length yielded a nearly identical structural pattern in circular dichroism analysis

(Supplementary Fig. S4F), suggesting that the catalytic domain of c-Src may represent

Src full length in function.

STAT3 Y705 phosphorylation has long been considered to play a central role in

STAT3 dimerization and STAT3 enhanceosome activity (32). However, STAT N-

terminal domain also plays a critical role in active STAT dimer formation (33). STAT3

N-terminal helix cluster bears 6 tyrosine residues (Fig. 5C). Tyrosine to phenylalanine

substitution of these N-terminal tyrosines in STAT3-Y705F plasmid reduced STAT3

phosphorylation upon c-Src cotransfection (Fig. 5D). Antibodies recognizing STAT3

phospho-Y45 and phospho-Y68 were prepared for confirming STAT3 phosphorylation

by Src (Fig. 5E and Supplementary Fig. S5A, S5B) (34). Interestingly, STAT3

phosphorylation on Y45, Y68, and Y705 showed differential responses toward Src

K401R, K423R, and K427R mutants (Fig. 5F). Src-K401R reduced STAT3

Downloaded from cancerres.aacrjournals.org on September20 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

phosphorylation at Y45 and Y68, Src-K423R mainly affected STAT3 Y45

phosphorylation, whereas Src-K427R affected STAT3 Y45 and Y705 phosphorylation

(Fig. 5F). These results clearly demonstrate that Src phosphorylates STAT3 not only

on Y705 of C-terminal region but also on tyrosine residues of the N-terminal helix.

Treatment with different cytokines/growth factors or deacetylase inhibitors

induced global STAT3 tyrosine phosphorylation in MCF-7 cells (Fig. 5G). Unlike

OSM, which showed the strongest effect on STAT3 Y45 and Y705 phosphorylation

induction, EGF preferentially induced STAT3 N-terminal Y45 and Y68

phosphorylation (Fig. 5G). Interestingly, deacetylase inhibitors induced STAT3

phosphorylation in a manner quite similar to the one induced by EGF (Fig. 5G).

Furthermore, STAT3 Y45 and Y68 phosphorylation in response to the above

conditions was largely abolished in Src-null SYF cells but was recovered in these cells

after c-Src was reintroduced (Fig. 5H). In SYF cells with c-Src reintroduction, K427R

mutation reduced c-Src activity in phosphorylation of STAT3 in response to cytokine

or EGF treatment (Fig. 5I). Collectively, these results indicate that acetylation at c-Src

catalytic domain upon EGF or cytokines treatment is required for Src to catalyze

STAT3 phosphorylation and activation.

STAT3 transcriptional activity relies on STAT3/acetyl-Src enhanceosome

formation

STAT3 nuclear translocation and transcriptional competency are induced by EGF

stimulation (35-37). A luciferase reporter assay in SYF cells demonstrated that full

length c-Src or Src-CD-induced STAT3 transcription activity was enhanced by CBP

cotransfection (Fig. 6A). However, Src-CD K401R and Src-CD K427R failed to

respond to CBP for STAT3 transcription activity promotion (Fig. 6A). Consistently,

Downloaded from cancerres.aacrjournals.org on September21 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Src-CD K427R mutation failed to induce STAT3 Y705 phosphorylation

(Supplementary Fig. S6A).

We further tested STAT3 with N-terminal YF (tyrosine to phenylalanine) mutation

on gene regulation. STAT3 with either Y14F or Y45F mutation reduced its

transcriptional activity upon EGF stimulation (Fig. 6B), suggesting phosphorylation of

these N-terminal tyrosine residues also play a role in STAT3 transcriptional activity

regulation. STAT N-terminal domain (ND) plays several distinct roles including

forming inactive dimer prior to cytokine-dependent STAT3 activation (38-41). When

cotransfected in HEK293T cells, STAT3 FL and STAT3 ND were constitutively

associated together (Fig. 6C). Src cotransfection disrupted this interaction between two

STAT3 protein molecules within their N-terminal domain (Fig. 6C). However, STAT3

N-terminal Y45F mutation failed to respond to Src for dedimerization (Fig. 6C). Thus,

STAT3 N-terminal domain dimerization is a reversible process regulated by Y45

phosphorylation.

Because both c-Src and STAT3 translocate to nuclei upon EGF treatment and can

form complex in the nuclei (5, 8, 37, 42), we analyzed their DNA binding activity by

incubating SIE oligo-agarose beads with nuclear fractions prepared from HEK293T

cells transfected with STAT3, c-Src and CBP. STAT3 bound to SIE sequence in the

presence of CBP (Fig. 6D), agreeing with previous finding that STAT3 can bind to

DNA independent of Y705 phosphorylation (43). c-Src was recovered from STAT3-

DNA-oligo beads, suggesting STAT3-Src complex can bind to DNA, which was

enhanced by CBP (Fig. 6D). Src K427R mutant failed to be recovered from STAT3-

DNA complex (Fig. 6D, last lane). Consistently, our chromatin immunoprecipitation

(ChIP) assay results showed that both STAT3 and c-Src were specifically present at the

promoters of STAT3-regulated genes including bcl-xl, cyclin D1, and muc1 (Fig. 6E),

Downloaded from cancerres.aacrjournals.org on September22 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

We next performed an in vitro transcription assay and the results revealed that purified

acetylated Src carried a stronger activity than Src in initiating Bcl-xL transcription (Fig.

6F). In STAT3-null cells, introducing STAT3 Y45F mutant or Y45/Y68F double

mutant reduced STAT3-dependent gene expression (Fig. 6G). All these results suggest

that c-Src/STAT3 co-ordinates to initiate gene transcription.

Src requires acetylation in both unique domain and catalytic domain for its

tumorigenic activity

Src plays an important role in tumorigenesis (6, 7). In SYF cells, we compared

different forms of c-Src in cell growth. Src-ΔUD or Src-CD was more active than full

length Src in cell growth promotion (Fig. 7A), and Src-ΔUD was hyper phosphorylated

on Y416 (Supplementary Fig. S6B). c-Src with K7N or K427N mutation was more

active than Src with K7R or K427R mutation in cell growth, respectively (Fig. 7A),

further supporting our conclusion that acetylation of these residues is necessary for full

activity of Src in its cellular substrate recruitment and activation. In NIH 3T3

fibroblasts, different forms of v-Src were tested for their colony formation activity.

While wild type v-Src dramatically induced more colonies formation in larger size; v-

Src KR mutants reduced its activity in colony formation (Fig. 7B and 7C). On the

contrary, v-Src with KN mutations showed greater activity in colony formation

(Supplementary Fig. S6C). Our subsequent in vivo tumor formation experiments

revealed that v-Src with KR mutation of these sites (K5, 7, 9, 423, and 427) all

markedly reduced tumorigenic activity of Src (Fig. 7D-F).

By using acetyl-Src or phospho-STAT3 antibodies whose specificity for

immunostaining was validated in Supplementary Fig. S7A, we revealed that in breast

carcinoma, but not in the adjacent normal tissues, K7-acetylated c-Src was

predominantly in nuclei, K9- or K423-acetylated Src was detected mainly in nuclei but

Downloaded from cancerres.aacrjournals.org on September23 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

cytoplasmic Src was also detected, and K427-acetylated Src was found in both

cytoplasm and nuclei (Fig. 7G and Supplementary Fig. S7B). N-terminal phospho-

STAT3 detected with anti-pY45 or anti-pY68 was found in both cytoplasm and nuclei

but was enriched in nuclei of breast cancer cells (Fig. 7G and Supplementary Fig. S7B).

Western blot was also performed to show that Src-aK7, -aK9, -aK423 and -aK427 and

STAT3-pY45, -pY68, and pY705 were highly expressed in breast cancer tissue

(Supplementary Fig. S7C and S7D). These findings clearly demonstrate that

acetylation plays an essential role in regulation of Src subcellular localization and

catalytic activity that eventually lead to Src-STAT3 enhanceosome formation and

STAT3 activation.

Discussion

Binding of Src to phospholipid membranes requires both hydrophobic insertion of

myristate into the hydrocarbon interior of the membrane and nonspecific electrostatic

interaction of its N-terminal positively charged cluster(s) of basic residues with acidic

phospholipids (27). Since N-terminal Ser 6 is important for Src myristoylation by Nmt

(44), the positively charged Src K7 cluster plays a de facto role in dynamic shuttling of

Src on and off from cell membrane when it is activated by EGF. Phosphorylation of the

C-terminal S181 of k-Ras adds a negative charge to the farnesylated C-terminal CAAX

motif and promotes dissociation from cell membrane for mitochondria (45). Similarly,

side-chain neutralization of the K7-cluster dissociates Src from the plasma membrane

and targets it to other subcellular compartments. K7N Src mutant attenuates its

transforming ability due to the lack of global posttranslational modification induced by

EGF (26). Src acetylated at K7-cluster does not bind to cell membrane, which allows it

Downloaded from cancerres.aacrjournals.org on September24 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

to reach substrates such as nuclear STAT3 that do not have an easily access to the cell

membrane.

The three lysine residues, K401, K423, and K427, form a triangle in the catalytic

domain and help to link the N-lobe and C-lobe of kinase domain (29). The protruding

feature of the p+1 loop allows K427 to be easily accessed by both CBP and STAT3

(30). The K427 residue appears to be a strong determinant for the selection of basic

residues at the kinase enzymatic core for substrate docking. Hence, the SH2-catalytic

domain of Src and the SH2-dimerization domain of STAT3 undergo a modular enzyme-

substrate heterodimerization. While STAT3 C-terminal region and Src-catalytic

domain may undergo “SH2 domain-aK427” modular protein-protein interaction,

STAT3 N-terminal domain and Src may also undergo “pY45-SH2 domain” modular

interaction. Formation of STAT3-Src complex will be necessary for an enhanced

STAT3 transcriptional activity. DNA-associated STAT3 enhanceosome contains not

only histone acetyltransferase such as CBP/p300 but also a protein tyrosine kinase like

Src (39). STAT3 enhanceosome activity can be therefore maximally warranted.

In order to form a transcriptionally active dimer, STAT3 requires phospho-Y705-

dependent C-terminal dimerization and N-terminal dedimerization to disrupt the

transcriptionally inactive STAT3 dimer (33, 46). Other group also reported that in the

case of STAT1, N-terminal dimerization is required for efficient presentation of STAT1

pY to phosphatases, allowing dephosphorylation to complete the activation–inactivation

cycle (40, 46). However, how N-terminal domain dimer dissociates remains largely

unknown. We found that phosphorylation at STAT3 Y45 by Src regulates STAT3 N-

terminal dimerization and Y45F mutation reduced STAT3 transcriptional activity,

which may result from prolonged dimerization and/or accelerated dephosphorylation of

activated STAT3. Our results suggest that STAT3 N-terminal Y45 phosphorylation and

Downloaded from cancerres.aacrjournals.org on September25 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

C-terminal Y705 phosphorylation work together to swtich N-terminal dimerization into

C-terminal dimerization which is critical for STAT3 switch from inactive dimer into

active dimer. Structural analyses reveal significant involvement of charged and polar

groups in protein–protein association (47). While the negative charged phospho-Y705

interacts with those positive charged lysine and arginine residues of the SH2 domain

within the C-terminal region, phospho-Y45 as well as other phosphotyrosine residues in

the N-terminal domain reduce local protein-protein interaction presumably via

increasing negative charge in the interaction interface (48). Phosphorylation on

multiple sites within the N-terminal domain may have other roles in addition to N-

terminal dedimerization (38-40, 49).

Acetylation as a regulatory mechanism of enzymatic activity has been highly

conserved during evolution. Our findings have established a close link between

acetylation and protein tyrosine kinase activity as central step in signal transduction and

transcriptional regulation in mammalian cells in response to environmental changes.

Acknowledgments

We are grateful to JS Brugge for chicken c-Src and v-Src cDNA constructs; SK Pierce

for Src-UD1-10-YFP; DC Flynn for Src-/- SYF and control (Src introduced) SYF cells;

and GR Stark for U3A cells. We thank H Zhang for technical assistance and our

colleagues RA Altura and WT Yang for critical reading of this manuscript.

References

Downloaded from cancerres.aacrjournals.org on September26 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

1. Buss JE, Kamps MP, Sefton BM. Myristic acid is attached to the transforming protein of Rous sarcoma virus during or immediately after synthesis and is present in both soluble and membrane-bound forms of the protein. Molecular and cellular biology. 1984;4:2697-704. 2. Martin GS. The hunting of the Src. Nature reviews Molecular cell biology. 2001;2:467-75. 3. Sigal CT, Zhou W, Buser CA, McLaughlin S, Resh MD. Amino-terminal basic residues of Src mediate membrane binding through electrostatic interaction with acidic phospholipids. Proceedings of the National Academy of Sciences of the United States of America. 1994;91:12253-7. 4. Liebl EC, Martin GS. Intracellular targeting of pp60src expression: localization of v-src to adhesion plaques is sufficient to transform chicken embryo fibroblasts. Oncogene. 1992;7:2417-28. 5. Jaganathan S, Yue P, Paladino DC, Bogdanovic J, Huo Q, Turkson J. A functional nuclear epidermal growth factor receptor, SRC and Stat3 heteromeric complex in pancreatic cancer cells. PloS one. 2011;6:e19605. 6. David-Pfeuty T, Bagrodia S, Shalloway D. Differential localization patterns of myristoylated and nonmyristoylated c-Src proteins in interphase and mitotic c-Src overexpresser cells. Journal of cell science. 1993;105 ( Pt 3):613-28. 7. Miyazaki T, Neff L, Tanaka S, Horne WC, Baron R. Regulation of cytochrome c oxidase activity by c-Src in osteoclasts. The Journal of cell biology. 2003;160:709-18. 8. Yang W, Xia Y, Ji H, Zheng Y, Liang J, Huang W, et al. Nuclear PKM2 regulates beta-catenin transactivation upon EGFR activation. Nature. 2011;480:118-22. 9. Thomas JE, Soriano P, Brugge JS. Phosphorylation of c-Src on tyrosine 527 by another protein tyrosine kinase. Science. 1991;254:568-71. 10. Pu M, Akhand AA, Kato M, Hamaguchi M, Koike T, Iwata H, et al. Evidence of a novel redox-linked activation mechanism for the Src kinase which is independent of tyrosine 527-mediated regulation. Oncogene. 1996;13:2615-22. 11. Song L, Turkson J, Karras JG, Jove R, Haura EB. Activation of Stat3 by receptor tyrosine kinases and cytokines regulates survival in human non-small cell carcinoma cells. Oncogene. 2003;22:4150-65. 12. Tang X, Gao JS, Guan YJ, McLane KE, Yuan ZL, Ramratnam B, et al. Acetylation-dependent signal transduction for type I interferon receptor. Cell. 2007;131:93-105. 13. Wang W, Pan K, Chen Y, Huang C, Zhang X. The acetylation of transcription factor HBP1 by p300/CBP enhances p16INK4A expression. Nucleic acids research. 2012;40:981-95. 14. Ye Y, Godzik A. Flexible structure alignment by chaining aligned fragment pairs allowing twists. Bioinformatics. 2003;19 Suppl 2:ii246-55. 15. Prlic A, Bliven S, Rose PW, Bluhm WF, Bizon C, Godzik A, et al. Pre- calculated protein structure alignments at the RCSB PDB website. Bioinformatics. 2010;26:2983-5. 16. Harris KF, Shoji I, Cooper EM, Kumar S, Oda H, Howley PM. Ubiquitin- mediated degradation of active Src tyrosine kinase. Proceedings of the National Academy of Sciences of the United States of America. 1999;96:13738-43. 17. Liang X, Lu Y, Wilkes M, Neubert TA, Resh MD. The N-terminal SH4 region of the Src family kinase Fyn is modified by methylation and heterogeneous fatty acylation: role in membrane targeting, cell adhesion, and spreading. The Journal of biological chemistry. 2004;279:8133-9.

Downloaded from cancerres.aacrjournals.org on September27 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

18. Song H, Li CW, Labaff AM, Lim SO, Li LY, Kan SF, et al. Acetylation of EGF receptor contributes to tumor cell resistance to histone deacetylase inhibitors. Biochemical and biophysical research communications. 2011;404:68-73. 19. Ma L, Gao JS, Guan Y, Shi X, Zhang H, Ayrapetov MK, et al. Acetylation modulates prolactin receptor dimerization. Proceedings of the National Academy of Sciences of the United States of America. 2010;107:19314-9. 20. Shi D, Pop MS, Kulikov R, Love IM, Kung AL, Grossman SR. CBP and p300 are cytoplasmic E4 polyubiquitin ligases for p53. Proceedings of the National Academy of Sciences of the United States of America. 2009;106:16275-80. 21. Twamley-Stein GM, Pepperkok R, Ansorge W, Courtneidge SA. The Src family tyrosine kinases are required for platelet-derived growth factor-mediated signal transduction in NIH 3T3 cells. Proceedings of the National Academy of Sciences of the United States of America. 1993;90:7696-700. 22. Ayrapetov MK, Nam NH, Ye G, Kumar A, Parang K, Sun G. Functional diversity of Csk, Chk, and Src SH2 domains due to a single residue variation. The Journal of biological chemistry. 2005;280:25780-7. 23. Roskoski R, Jr. Src kinase regulation by phosphorylation and dephosphorylation. Biochemical and biophysical research communications. 2005;331:1-14. 24. Luttrell DK, Lee A, Lansing TJ, Crosby RM, Jung KD, Willard D, et al. Involvement of pp60c-src with two major signaling pathways in human breast cancer. Proceedings of the National Academy of Sciences of the United States of America. 1994;91:83-7. 25. Wilson LK, Luttrell DK, Parsons JT, Parsons SJ. pp60c-src tyrosine kinase, myristylation, and modulatory domains are required for enhanced mitogenic responsiveness to epidermal growth factor seen in cells overexpressing c-src. Molecular and cellular biology. 1989;9:1536-44. 26. Kaplan JM, Mardon G, Bishop JM, Varmus HE. The first seven amino acids encoded by the v-src oncogene act as a myristylation signal: lysine 7 is a critical determinant. Molecular and cellular biology. 1988;8:2435-41. 27. Resh MD. Myristylation and palmitylation of Src family members: the fats of the matter. Cell. 1994;76:411-3. 28. Xu W, Harrison SC, Eck MJ. Three-dimensional structure of the tyrosine kinase c-Src. Nature. 1997;385:595-602. 29. Xu W, Doshi A, Lei M, Eck MJ, Harrison SC. Crystal structures of c-Src reveal features of its autoinhibitory mechanism. Molecular cell. 1999;3:629-38. 30. Yuan ZL, Guan YJ, Wang L, Wei W, Kane AB, Chin YE. Central role of the threonine residue within the p+1 loop of receptor tyrosine kinase in STAT3 constitutive phosphorylation in metastatic cancer cells. Molecular and cellular biology. 2004;24:9390-400. 31. Thomas SM, Brugge JS. Cellular functions regulated by Src family kinases. Annual review of cell and developmental biology. 1997;13:513-609. 32. Levy DE, Darnell JE, Jr. Stats: transcriptional control and biological impact. Nature reviews Molecular cell biology. 2002;3:651-62. 33. Mertens C, Zhong M, Krishnaraj R, Zou W, Chen X, Darnell JE, Jr. Dephosphorylation of phosphotyrosine on STAT1 dimers requires extensive spatial reorientation of the monomers facilitated by the N-terminal domain. Genes & development. 2006;20:3372-81. 34. Xu L, Ji JJ, Le W, Xu YS, Dou D, Pan J, et al. The STAT3 HIES mutation is a gain-of-function mutation that activates genes via AGG-element carrying promoters. Nucleic acids research. 2015;43:8898-912.

Downloaded from cancerres.aacrjournals.org on September28 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

35. de la Iglesia N, Konopka G, Puram SV, Chan JA, Bachoo RM, You MJ, et al. Identification of a PTEN-regulated STAT3 brain tumor suppressor pathway. Genes & development. 2008;22:449-62. 36. Andl CD, Mizushima T, Oyama K, Bowser M, Nakagawa H, Rustgi AK. EGFR-induced cell migration is mediated predominantly by the JAK-STAT pathway in primary esophageal keratinocytes. American journal of physiology Gastrointestinal and liver physiology. 2004;287:G1227-37. 37. Lo HW, Hsu SC, Ali-Seyed M, Gunduz M, Xia W, Wei Y, et al. Nuclear interaction of EGFR and STAT3 in the activation of the iNOS/NO pathway. Cancer cell. 2005;7:575-89. 38. Xu X, Sun YL, Hoey T. Cooperative DNA binding and sequence-selective recognition conferred by the STAT amino-terminal domain. Science. 1996;273:794-7. 39. Zhang JJ, Vinkemeier U, Gu W, Chakravarti D, Horvath CM, Darnell JE, Jr. Two contact regions between Stat1 and CBP/p300 in interferon gamma signaling. Proceedings of the National Academy of Sciences of the United States of America. 1996;93:15092-6. 40. Zhong M, Henriksen MA, Takeuchi K, Schaefer O, Liu B, ten Hoeve J, et al. Implications of an antiparallel dimeric structure of nonphosphorylated STAT1 for the activation-inactivation cycle. Proceedings of the National Academy of Sciences of the United States of America. 2005;102:3966-71. 41. Vogt M, Domoszlai T, Kleshchanok D, Lehmann S, Schmitt A, Poli V, et al. The role of the N-terminal domain in dimerization and nucleocytoplasmic shuttling of latent STAT3. Journal of cell science. 2011;124:900-9. 42. Paladino D, Yue P, Furuya H, Acoba J, Rosser CJ, Turkson J. A novel nuclear Src and p300 signaling axis controls migratory and invasive behavior in pancreatic cancer. Oncotarget. 2016;7:7253-67. 43. Yuan ZL, Guan YJ, Chatterjee D, Chin YE. Stat3 dimerization regulated by reversible acetylation of a single lysine residue. Science. 2005;307:269-73. 44. Yasuda K, Kosugi A, Hayashi F, Saitoh S, Nagafuku M, Mori Y, et al. Serine 6 of Lck tyrosine kinase: a critical site for Lck myristoylation, membrane localization, and function in T lymphocytes. Journal of immunology. 2000;165:3226-31. 45. Bivona TG, Quatela SE, Bodemann BO, Ahearn IM, Soskis MJ, Mor A, et al. PKC regulates a farnesyl-electrostatic switch on K-Ras that promotes its association with Bcl-XL on mitochondria and induces apoptosis. Molecular cell. 2006;21:481-93. 46. Mao X, Ren Z, Parker GN, Sondermann H, Pastorello MA, Wang W, et al. Structural bases of unphosphorylated STAT1 association and receptor binding. Molecular cell. 2005;17:761-71. 47. Jones S, Thornton JM. Principles of protein-protein interactions. Proceedings of the National Academy of Sciences of the United States of America. 1996;93:13-20. 48. Jia J, Tong C, Wang B, Luo L, Jiang J. Hedgehog signalling activity of Smoothened requires phosphorylation by protein kinase A and casein kinase I. Nature. 2004;432:1045-50. 49. Becker S, Groner B, Muller CW. Three-dimensional structure of the Stat3beta homodimer bound to DNA. Nature. 1998;394:145-51.

Figure Legends Figure 1.

Downloaded from cancerres.aacrjournals.org on September29 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Src is acetylated within both the UD-loop and the CD-loop by CBP. A, MCF-7 cells were treated with EGF (100 ng/ml) for indicated time with or without deacetylase inhibitors. CBP was depleted with siRNA. IP and westernblot were performed with Src antibody and pan acetyl-lysine antibody. Whole cell lysates were blotted with Src pY416 antibody. B, Endogenous c-Src-CBP complex formation in MCF7 cells upon EGF treatment. C, Chicken c-Src full length or c-Src with different domain deletions were cotransfected with CBP in 293T cells. D, CBP in full length or different regions as indicated were cotransfected with c-Src in 293T cells. IP and westernblot were performed as shown. E, Alignment of c-Src and Lyn in their UD-loop and CD-loop. F, Acetylation of c-Src-ΔCD and c-Src-ΔCD with K5R, K7R, K9R, K37R or K59R substitutions was analyzed by co-transfecting with CBP in 293T cells. G, Acetylation of c-Src-CD and c-Src-CD with K401R, K423R, or K427R substitutions was analyzed by co-transfecting with CBP in 293T cells. H, CBP-induced acetylation of c-Src of different domain variants was analyzed in 293T cells. I, In MCF-7 cells, IP and westernblot were performed to show EGF induced c-Src acetylation on K7, K9, K423, and K427, and phosphorylation at Y419. Data shown are representative of two independent experiments. Data shown in A, C, F, G were representative of three independent experiments.

Figure 2. Src global post-translational modifications are triggered by CBP. A, c-Src protein was immunoprecipitated from 293T cells transfected with chicken c-Src with or without CBP. Src tyrosine phosphorylation level was detected with anti-pTyr (4G10) antibody and Src-pY416 antibody. Representative result was shown. B, c-Src acetylation and tyrosine phosphorylation were detected in WT or CBP-/- MEF cells with or without EGF or OSM treatment. Representative result was shown. C, Illustration of c-Src global posttranslational modifications in the absence or presence of CBP.

Figure 3. Src UD-loop acetylation dissociates Src from plasma membrane for nuclear translocation. A, In NIH3T3 cells, endogenous c-Src acetylation, as detected with either anti-acetyl-lysine (pan) or anti-Src-aK7 antibody, was induced by overexpression of CBP or by NAM or TSA treatment. B, Plasma membrane associated c-Src proteins were recovered by IP from CBP-/- or control MEF cells treated with EGF for indicated

Downloaded from cancerres.aacrjournals.org on September30 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

times. Na+/K+ ATPase was blotted as control. Representative result was shown. C, c- Src recovered from membrane, cytoplasm, and nuclei fractions were analyzed for K7 acetylation. Representative result was shown. D, Representative 293T cells expressing various forms of Src-UD-YFP as indicated were visualized with confocal fluorescent microscope. 293T cells expressing wild type Src-UD-YFP was treated with EGF for 60 min. Bar charts showed the percentage of cells with nucleoplasmic YFP signal in 100 cells. E, Src G2 myristoylation levels were measured. HeLa cells expressing indicated proteins were metabolic labeled with 3H-myristic acid followed by 3H incorporation detection by scintillation. Data are mean ± S.D. (n = 3). F, Cellular distribution of CBP was examined by confocal immunofluorescence in NIH3T3 cells before and after EGF or OSM treatment (left panel. Bar charts showed the percentage of cells with cytoplasmic CBP staining in 100 cells). Cellular distribution of c-Src and its variants in SYF cells was examined by confocal immunofluorescence before and after EGF treatment (right panel). Scale bar: 10uM. G, CBP-/- and control MEF cells were treated as indicated. Src from plasma membrane, cytoplasmic, and nuclear fractions was pull down and detected with anti-Src antibody. Representative result was shown. H, CBP-/- MEF cells were transfected with EV or CBP- ΔNLS followed by EGF treatment for indicated times. Nuclear extracts were analyzed with indicated antibodies.

Figure 4. The modular interaction between the Src and STAT3. A, 3D alignment of K401, K423, and K427 in the CD-loop between active and inactive chicken c-Src. 1FMK.PDB- inactive Src and 1Y57.PBD-active Src were used for alignment. B, STAT3 FL or various domain deletion mutants were co-transfected with Src in 293T cells to analyze their interactions (upper panel). Arrangement of STAT3 domains and phosphotyrosine sites were illustrated (bottom panel). C, Src FL or domain deletion mutants were cotransfected with STAT3 in 293T cells to analyze their interactions. D, Src WT or Src KR mutants were co-transfected with STAT3 to analyze their interactions. E, Control or acetyl-Src-K427 peptide were incubated with STAT3-Src complex recovered from 293T transfactants. F, Full-length Myc-Src interaction with GST-STAT3-FL or GST- STAT3-SH2 domain was performed as indicated.

Figure 5.

Downloaded from cancerres.aacrjournals.org on September31 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Acetyl-Src induces STAT3 N-terminal phosphorylation. A, Purified STAT3 was incubated with purified Src or acetyl-Src (CBP acetylated) in the in vitro phosphorylation assay. B, STAT3/ Src-FL or STAT3/Src-CD was transfected with or without CBP in 293T cells. Cell lysates were blotted with indicated antibodies. C, Alignment of the N-terminal helical domain among STAT family members. Helical regions are printed in blue color. D, STAT3 N-terminal YF substitutions based upon Y705F as the backbone were transfected with Src. STAT3 immunoprecipitates were analyzed for tyrosine phosphorylation with 4G10. E, Src induced STAT3 phopsphorylation on Y45, Y68, and Y705 in 293T cells cotransfected with STAT3 and c-Src. F, Various Src mutants (K401R, K423R, and K427R) were tested for STAT3 phopsphorylation on Y45, Y68, and Y705 in 293T cells cotransfected with STAT3 and Src. G, MCF-7 cells were treated with indicated reagents for 30 min and STAT3 pY45, pY68 and pY705 were detected with specific antibodies. H, STAT3 phosphorylation on Y45, Y68, and Y705 was assessed with indicated antibodies in SYF cells or SYF cells stably re-introduced with Src upon treatment with indicated reagents for 30 min. I, STAT3 phosphorylation level on Y705 was assessed in SYF cells stably re-introduced with wild type c-Src or Src K427R mutant upon treatment with indicated reagents for 30 min. Data shown in D, F, G, H, I were representative of two independent experiments.

Figure 6. The CD-acetylated Src is a critical component of STAT3 enhanceosome in gene regulation. A, Luciferase activity of IRF1 promoter 2xSIE-luciferase was measured in SYF cells expressing Src FL or Src-CD variants and cotransfected with or without CBP. Data are mean ± S.D. (n = 3). Student’s t-test was used to calculate P value. **, P < 0.01. B, Luciferase activity of indicated STAT3 constructs co-transfected with EGFR was analyzed in STAT1-null U3A cells with EGF treatment. Data were mean ± S.D. (n = 3). Student’s t-test was used to calculate P value. *, P < 0.05; **, P < 0.01. C, STAT3-FL (Y705F) and STAT3-ND (1-155) with or without Src were cotransfected into 293T cells. Anti-Myc immunoprecipitates were analyzed with anti-HA antibody. D, Myc-STAT3 and Src-HA variants were co-transfected with or without HA-CBP in 293T cells. STAT3 and Src complex was purified with STAT3 SIE DNA oligo-agarose beads and analyzed by westernblot with anti-Myc or anti-HA antibodies. Mutated SIE DNA oligo-agarose was used as control. E, Both STAT3 and Src were recovered from

Downloaded from cancerres.aacrjournals.org on September32 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

promoters of bcl-xl cyclin D1 and muc1 genes by ChIP assay in MCF-7 cells with or without EGF, OSM, or NAM+TSA. F, In vitro transcription of bcl-xl gene in presence of purified STAT3 or in combination with purified Src or acetylated Src. G, STAT3 WT, Y45F, or Y45Y68F was stably introduced into Stat3-/- MEF cells. Expression of various proteins as indicated was detected.

Figure 7. Src acetylation is critical for its tumorogenesis property. A, In Src-null cells expressing STAT3 and Src of indicated forms, cell proliferation was measured by BrdU incorporation after EGF treatment. Data were mean ± S.D. (n = 3). Student’s t-test was used to calculate P value. *, P < 0.05; **, P < 0.01. B-C, v-Src WT or KR mutants transformed NIH3T3 cells in colony formation (B) and counted in each well (C). Results shown were mean ± S.D., n=4. Student’s t-test was used to calculate P value. **, P < 0.01. D-F, v-Src-transformed NIH3T3 cells were injected into nude mice. Tumors from these mice were shown at the bottom panels (D). Tumor weight and numbers were plotted in (E) and (F) respectively. The data are presented as mean tumour volume + S.D. (n=5 mice per group); Mann–Whitney test, *, P < 0.05; **, P < 0.01. G, Immunostaining of human breast carcinoma tissues with indicated antibodies. The arrows indicate either cytoplasmic CBP or the nuclear associated human c-Src. Bar charts showed the percentage of cells with indicated protein localization. Nuc, staining of protein only in nucleoplasm; Nuc/Cyt, staining of protein in both nucleoplasm and cytoplasm.

Downloaded from cancerres.aacrjournals.org on September33 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 12, 2018; DOI: 10.1158/0008-5472.CAN-17-2314 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Acetylation within the N- and C-terminal domains of Src regulate distinct roles of STAT3-mediated tumorigenesis

Chao Huang, Zhe zhang, Lihan chen, et al.

Cancer Res Published OnlineFirst March 12, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-17-2314

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2018/03/10/0008-5472.CAN-17-2314.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet Manuscript been edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2018/03/10/0008-5472.CAN-17-2314. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.