Potential Roles Of ACK1and Src In The Development Of Intestinal Inflammation

Background and aimsInflammatory bowel diseases (IBD) are chronic inflammatory disorders affecting the gastrointestinal tract. The two main forms of IBD are Crohn’s disease (CD) and ulcerative colitis (UC). These conditions are thought to be genetically determined, overaggressive immune responses of the intestinal mucosa to normal resident bacterial constituents and environmental factors. The prevalence of IBD in the Western Britain and Northern Europe is high and it shows increasing trend in China.TNF-a was first found by its ability to induce hemorrhagic tumor necrosis. However, extensive research has shown conflicting roles of TNF-a as both a growthpromoting and an apoptotic factor. Proapoptotic signals initiated by TNF include stress-activated protein kinase/JNK and p38, whereas antiapoptotic pathways regulated by TNF-a include Akt/protein kinase B, ERK/MAPK, and NF-κB. However, little is known about the molecular switch determining TNF-a regulation of these two different functions. ErbB family is consisted of four tyrosine-kinase receptors:EGF receptor (EGFR)(ErbB1), ErbB2, ErbB3and ErbB4. ErbBs modulate cell proliferation, survival, and differentiation in many kinds of tissue types. Direct interaction with EGF-like ligands can activate EGFR and other ErbB family members and initiate formation of homodimers and/or heterodimers and increase kinase activity. Moreover, various extracellular stimuli have been reported to transactivate EGFR, such as agonists for G protein-coupled receptors (GPCR) and cytokine receptors. TNF-a, IL-1, IL-8and IFN-y can transactivate EGFR inmammary epithelial cells, hepatocytes and cancer-derived cell lines. The Src family members of nonreceptor intracellular tyrosine kinases have been implicated as targets of GPCRs in EGFR transactivation by direct phosphorylating cytoplasmic domains of EGFR or stimulating matrix metalloproteinase (MMP) activity to promote the release of the membrane-bound EGFR ligand. Src and MMPs also have been reported for cytokine-stimulated EGFR transactivation in several cell types.A number of signal transduction pathways activated by TNF-as, such as MAPK and Akt, which overlap with those by EGFR in intestinal epithelial cells. A mechanism by which TNF-a regulates signaling pathways to promote survival of colon epithelial cells is through transactivation of ErbBs. In vitro and in vivo, EGFR and ErbB2transactivation can promote intestinal epithelial cell survival exposed to TNF-a. In addition, ErbBs transactivated by TNF-a requires Src-kinase activity with Akt as a downstream target in this process. These results show an important relationship between ErbBs and TNF-a signal transduction pathway which permits maintenaning the survival ability of intestinal epithelial cells in a cytokine-enriched environment during acute injury.Activated Cdc42kinase (ACK1, also known as ACK or TNK2) an atypical non-receptor tyrosine kinase, is expressed in various organs including the brain, spleen, thymus and liver. ACK1integrates extracellular growth factor stimuli from activated receptor tyrosine kinases, to initiate intracellular signaling cascades. Recent studies have shown that the interplay of ACK1with downstream effecters, for example, AR, AKT and Wwox, is critical for cell survival, cell growth, cell differentiation and cell proliferation. To date, most investigations of ACK1have focused on the respiratory and reproductive system. Although the amplification of ACK1gene has been detected in a number of primary tumors (such as primary prostate, ovarian and lung tumors) and correlated with disease prognosis, little is known about the expression of ACK1protein in colorectal diseases. Our study group recently demonstrated significantly increased ACK1expression in the intestinal mucosa of colonic inflammatory conditions compared to HC by using integrated strategy in proteomics. Therefore, we aimed to study the mechanizms of ACK1and Src involved in intestinal inflammation and tunors.Materials and MethodsAntibodies and ChemicalsAnti-ACK1and anti-COX-2were purchased from Santa Cruz; Anti-MAPK, Anti-pMAPK, Anti-AKT, Anti-pAKT, Anti-pACKl, Anti-Src, Anti-pSrc, Anti-pErbB family, Anti-p65were purchased from Cell Signaling; Human recombined TNF-a, EGF, IL-1α anf IFN-γ were purchased from R&D. Src inhibitor PP1, TACE inhibitor TAPI-1and EGFR inhibitor AG1478were purchased from Calbiochem, MMP inhibitor BB94were purchased from Tocris, ErbB2inhibitor AG879, TGF-a nurolization were purchased from R&D; Anti-1κBαTyr42was purchased from ECM Biosciences; ACK1inhibitor AIM-100and Tyr176-AKT were synethesd by SAB and Invitrogen company.Cell cultureIECs were grown in RPMI1640medium with5%FBS and5U/ml murine IFN-y and maintained in5%CO2at37℃. IECs were plated onto1μg/ml fibronectin-coated surfaces. Then cells were transferred to37℃under nonpermissive conditions with0.5%FBS serum-free medium for16-24h before all experiment threatment. siRNA, plasmids and site-directed mutagenesisACK1constructs, kdACK and caACK, which consisted the kinase domain, SH3domain, CRIB domain, and proline-rich domain were generated. The full-length Ackl construct have been observed to be expressed poorly. Thus, the truncated constructs and both kdACK and caACK lacked the COOH-terminal region (788-1036amino acids) were used. A point mutation, K158R, was possessed in the ATP acceptor site of the kdAck construct which didn’t autophosphorylate. Mutagenesis was done using GeneEditor system (Promega, Madison, WI).ACK1primers:5’-CTAGGATCCTGCCCGCCCTCCCTGGCGCAG-3’, BamHI site;5’-CAGAAGCTTTTAAGGCACAGGCAGGGGGGT--3’,HindⅢ site.ErbB4primers:5’-ATGGCGATCGCATGAAGCCGGCGACAGGACTTTG-3’, Sgfl site;5’-TTGGGCCGGACCGGCCTTACACCACAGTATTCCGGTG-3’, SfiⅠ site. cDNA was subcloned into pcDNA3.1vector (Invitrogen, Carlsbad, CA). All plasmid constructs were sequenced.The short interfering RNA (siRNA) oligos were chemically synthesized. ACK1siRNA ACK siA:5’-AAGAUGGUGACAGAGCUGGCA-3’; ACK siA2:5’-GAAGAUGGUGACAGAGCUGGCACTT-3’The negative control was set up using the21-nucleotide RNA oligos (AAGUUCAGGUCGAUAUGUGCA), which does not match any DNA sequence in GenBank, as determined by NCBI Blast search.Cellular transfectionsIECs were maintained under permissive conditions at33℃with5%CO2. For transfection, IECs were cultured overnight to90%confluency. The Lipofect2000transfection kit was used for the transfection according to the manufacturer’s instructions (Invitrogen). the cells were lysed with precold mammlian cell lysis buffer (40mM HEPES, pH7.4,100mM NaCl,1%Triton X-100,25mM glycerol phosphate,1mM sodium orthovanadate,1mM EDTA,10μg/ml aprotinin, and10 μg/ml leupeptin) by rocking the plates at4℃for30min after24h transfection. The IECs lysates were cleared by centrifugation at14000g in a microfuge at the tempreture of4℃for4min before use.Migration assays1.5×105confluent IECs were trypsinized and plated on35mm culture dishes which were coated with fibronectin. After cell attachment for2h, the medium was changed to medium with0.5%FBS IFN-γ-free and then ncubated overnight at37℃. The medium was aspirated. Nine circular1mm cell-free areas were then created with a stabilized rotating silicone tip. This method was similar to that described by Watanabe and colleagues. IECs were washed twice with PBS immediately after the wound creation. Culture medium was replaced with the indicated factors in the presence or absence of siRNA or pharmacological inhibitors and cultured up to24h. The surface areas of wound were recorded by time-lapse video microscopy at the end of treatment. The surface areas of wound were measured by BioQuant Image Analysis software.Proliferation assayIECs cultured on fibronectin-coated chamber slides were untreated or treated with siACK1for72hours. The proliferation of IECs was detected by BrdU labeling at the end of treatment. IECs were incubated in cell culture medium for30min at the tempreture of37℃with10μM BrdU. And then mouse anti-BrdU antibody and fluorescein-conjugated anti-mouse IgG was used to dectect IECs without corporation of BrdU. Fluorescence microscopy was used to observe IECs. At least200IECs were counted to determine the proliferation rate in randomly chosen fields. IECs expressing BrdU positive cells were counted as a percentage of the total number of cells.Apoptosis assay IECs were cultured on fibronectin-coated chamber slides and then wounded or not and treated with TNF or EGF for24h. cells were fixed with1%paraformaldehyde in PBS at the end of treatment. Terminal deoxynucleotidyl transferase was used to detecte apoptotic cells [TdT; also termed TdT-mediated dUTP nick end labeling (TUNEL)] and labeled with FITC-conjugated anti-digoxigenin. Fluorescence microscopy was used to observe IECs. The percentage of IECs undergoing apoptosis was then determined similar to proliferation assays. At least500IECs were counted to determine the apoptosis rate in randomly chosen fields.Preparation of cellular lysatesIECs were rinsed twice with ice-cold PBS and then scraped into cell lysis buffer containinglmM orthovanadate,1mM sodium pyrophosphate,20mM HEPES (pH7.5),50mM glycerol phosphate,1%Triton X-100,10mg/ml leupeptin,10mg/ml aprotinin,18mg/ml PMSF, and incubated for30min on ice after treatment with EGF or TNF in the presence or absence of inhibitors or siRNA for several indicated times. The lysates were centrifuged at14000g rpm for15min under the tempreture of4℃. A DC protein assay was used to determine the protein concentration of the Triton-soluble fraction.TGF-a measurementTGF-a levels were measured by enzyme-linked immunosorbent assay in conditioned media. Briefly,96-well Nunc immunoplates were coated with0.4μg/ml TGF-a capture antibody diluted in phosphate-buffered saline overnight at room temperature, washed with phosphate-buffered saline (PBS) containing0.1%Tween20(PBST) three times, and then incubated with PBS which contained2%bovine serum albumin for1h. Each well was incubated with100μl conditioned media or TGF-a standard serially diluted in M3D media for2h at room temperature after washing once with PBST. Wells were washed, folowed by incubated with0.3μg/ml biotin-conjugated anti-TGF-a for60min, then washed three times, and finally incubated with streptavidin-horseradish peroxidase for30min. TMB peroxidase substrate reagents were used to detecte peroxidase activity.StatisticsStatistical significance in each study was determined by Student’s t test or factorial analysis with a confidence level of0.05. All data presented were representative of at least three repeat experiments and presented as mean±SEM.Results1. ACK1expression and activation was required for IEC survival in the presence of TNF. Apoptosis rate of ACK1siRNA group was significantly higher than control (t=-2.982, P=0.041). Apoptosis rate of group treated with a combination of ACK1siRNA and TNF-a was significantly higher than that with ACK1siRNA alone0=-13.877, P=0.000). Apoptosis rate of group treated with AIM-100was significantly higher than control (t=-2.915,P=0.043). Apoptosis rate group treated with AIM-100combined with TNF-α was significantly higher than that with AIM-100alone (t=-11.534, P=0.000). No significant difference were observed between group treated with a combination of AIM-100with IL-la (t=-1.281,P=0.269) or IFN-γ (t=-1.215, P=0.291) and group with AIM-100alone.2. ACK1suppression decreased IEC wound closure. Wound healing rate was significantly lower in ACK1siRNA group than control (t=13.231, P=0.000). Wound healing rate was significantly lower in group treated with a combination of ACK1siRNA and TNF-a than that with TNF-a alone (t=-25.414, P=0.000). Wound healing rate was significantly lower in EGF group than control (t=43.528, P=0.000). Wound healing rate was significantly lower in group treated with a combination of ACK1siRNA and EGF than that with EGF alone (t=11.045, P=0.000). 3. Src promoted ACK1activation in IEC. In the presence of TNF-a, Apoptosis rate in group threated with a combination of PP1and AIM-100is significantly lower than that with PP1(PP1+TNF vs PP1+AIM-100+TNF, P=0.000; PP1vs PP1+TNF, P=0.000; PP1+AIM-100vs PP1+AIM-100+TNF, P=0.000) or AIM-100alone (AIM-100+TNF vs PP1+AIM-100+TNF, P=0.000; PP1+AIM-100vs PP1+AIM-100+TNF, P=0.000).4. Src interacted with ErbBs. TNF-a-induced phosphorylation of EGFR, ErbB2and ErbB4was suppressed in the presence of Src but not ACK1inhibitor. TNF-a induced phosphorylation of ACK1was singnificantly down-regulated in in the presence of EGFR, ErbB2and ErbB4inhibitors, while Src activity was only slightly affected.5. TNF-a-induced phosphorylation of Src, ACK1, AKT, ERK and PI3K were significantly lower in groups treated with a combination of EGFR and ErbB2inhibitor were significantly higher than that with either EGFR or ErbB2alone. EGFR siRNA significantly suppressed TNF-a-induced phosphorylation of ACK1in IECs with ErbB4-overexprssion. ErbBs inhibitor didn’t affect TNF-a-induced p65activity.6. TNF induced of ACK1phosphorylation via MMP and TACE. TNF-a-induced phosphorylation of ACK1and EGFR were significantly reduced in the presence of MMP than that of TACE.7. TNF-a induced phosphorylation of ACK1via TGF-a. TNF-a-induced phosphorylation of ACK1, AKT, ERK, EGFR, ErbB2and ErbB4were significantly downregulated in the presence of TGF-a nutilization.8. Both TNF-a and EGF increased phosphorylation of Src, ACK1, AKT, ERK, p38and p65. Src inhibitor significantly suppressed EGF-induced phosphorylation of ACK1, AKT, ERK and p38, but only slightly downregulated that of p65. ACK1inhibitor significantly suppressed EGF-induced phosphorylation of AKT and ERK with phosphorylation of Src, p65and p38unchanged. TNF-α-induced phosphorylation of ACK1, PI3K and AKT but not ERK, p38or p65was downregulated in the presence of Src inhibitor. TNF-a-induced phosphorylation of AKT but not Src, PI3K and ERK, p38or p65was downregulated in the presence of ACK1inhibitor.9. Both Src and ACK1mediated EGF or TNF-α-induced phosphorylation of AKT and ERK via. Inhibition of ACK1activation by AIM-100not only suppressed AKT Tyr176-phosphorylation but also inhibited AKT activation, as noted by a significant decrease in AKT Ser473-and Thr308-phosphorylations. ACK1mediated EGF and TNF-α-induced phosphorylation of AKT at Tyrl76. EGF-induced IκBα phosphorylation at Tyr42residual was affected by neither ACK1nor Src inhibitor. TNF promoted the expression of COX-2and cPLA2via Src, ACK1and ErbB. TNF promoted the expression of IL-8via Src (t=9.250, P=0.001), but not ACK1(t=2.222, P=0.090).DiscussionACK1may play more important anti-apoptosis roles in the presenc of TNF-α than that in either IFN-γ or IL-1. The ability of ACK1in wound healing may be associated with its anti-apoptosis role and migration prompted by TGF-α transactivetion of EGFR activation. Src, MMP, TGF-a and ErbB may mediate TNF-α induced ACK1activation. Src and ACK1may activate AKT, and influence NF-κB activity indirectly via AKT. Neither Src nor ACK1may be associated with ERK. Src and ACK1may mediate the synesis of TNF-α-induced COX-2and cPLA2. Src may promote the release of TNF-α-induced IL-8. We conclude that ACK1and Src may promote wound healing as well as contribute to the development of intestinal inflammation via activating inflammory singnor passway, promoting the release of inflammation mediators, and envn IEC thansformation. Our findings of ACKl and Src involving inflammiton, cell survival and apoptosis provide new ideas in the diagnosis and treatment of IBD. ACKl and Src may be target in intestinal diseases.