Zyxin is a novel interacting partner for SIRT1
© Fujita et al; licensee BioMed Central Ltd. 2009
Received: 10 July 2008
Accepted: 27 January 2009
Published: 27 January 2009
Skip to main content
© Fujita et al; licensee BioMed Central Ltd. 2009
Received: 10 July 2008
Accepted: 27 January 2009
Published: 27 January 2009
SIRT1 is a mammalian homologue of NAD+-dependent deacetylase sirtuin family. It regulates longevity in several model organisms and is involved with cell survival, differentiation, metabolism among other processes in mammalian cells. SIRT1 modulates functions of various key targets via deacetylation. Recent studies have revealed SIRT1 protects neurons from axonal degeneration or neurodegeneration. Further, SIRT1 null mice exhibit growth retardation and developmental defects, suggesting its critical roles in neurons and development.
To identify novel binding partners for SIRT1 in the central nervous system, we performed yeast two-hybrid screening on human fetal brain cDNA library and found that zyxin is a possible binding partner. SIRT1 and zyxin transcript were both preferentially expressed in developmental mouse brain. Zyxin accumulates in the nucleus where it is co-localized with SIRT1 after treatment with leptomycin B in COS-7 cells. Furthermore, SIRT1 deacetylates zyxin, suggesting SIRT1 could interact with nuclear-accumulated zyxin and modulate its function through deacetylation.
Zyxin could be a novel interacting partner of SIRT1. Zyxin is an adaptor protein at focal adhesion plaque, regulating cytoskeletal dynamics and signal transduction to convey signal from the ECM (extracellular matrix) to the nucleus. Our results raise the possibility that SIRT1 regulates signal transmission from ECM to the nucleus by modulating the functions of zyxin via deacetylation.
SIRT1 is the mammalian homologue closest to yeast NAD+-dependent deacetylase Sir2 (silent information regulation 2). It was originally identified as a lifespan-extending gene when over-expressed in budding yeast, Caenorhabditis elegans, and Drosophila melanogaster [1–3]. There are seven homologues of SIRT (named SIRT1-7) in mammals, and SIRT1 is the homologue closest to Sir2. Post-translational modification of proteins, like acetylation, methylation, and phosphorylation, plays a pivotal role in the biological functions [4, 5]. Recently, several lines of evidence have indicated that SIRT1 regulates various functions such as cellular survival, differentiation, and metabolism by modulating key targets via deacetylation. These targets include FOXO, Ku70, p53, NF-κB, PGC-1α (PPAR-gamma co-activator 1-alpha) and PPARγ (peroxisome proliferator-activated receptor γ). [6–11]. The knockout of SIRT1 is not embryonic lethal however most null mice die in the perinatal period, exhibiting growth retardation and developmental defects in various tissues including eye, lung, pancreas, heart, and reproductive system[12, 13]. This indicates that SIRT1 plays a pivotal role in the development of multiple organ systems. Furthermore, SIRT1 protects neuron from neurodegeneration in cell-based models of Alzheimer's disease, amyotrophic lateral sclerosis (ALS), and Wallerian degeneration [14–16]. However, the molecular mechanism by which SIRT1 functions in the central nervous system (CNS) or in the development remains unclear. In this study, to identify interacting partners for SIRT1, we performed yeast two-hybrid screening on human fetal brain cDNA library. We found that zyxin is one of binding partners for SIRT1.
Zyxin is primarily localized at the focal adhesion plaque complex as an adaptor protein [17, 18]. Zyxin contains a proline-rich domain at the N-terminus and three LIM domains at the C-terminus that are cysteine-rich motifs involved in protein-protein interactions. Zyxin interacts with cytoskeleton-related proteins, including α-actinin, Mena/VASP, CRP, and H-warts/LATS1 [19–22], and signaling molecules, including Vav, CasL, and p130Cas [23, 24]. Focal adhesion plaques, where cells interact with the extracellular matrix (ECM), form a structure where multiple interactions and signal transduction networks crosstalk. This not only regulates cell adhesion, spreading, and motility but also transduces signals into the nucleus to regulate gene expression, cell proliferation, differentiation, and apoptosis [25, 26]. LIM domain proteins, including two related subfamily prototypes zyxin and paxillin, have been implicated in the regulation of cytoskeletal dynamics and signal transduction to convey signals from the ECM to the nucleus at the focal adhesion plaques [17, 18]. It was recently reported that zyxin can shuttle between the cytosol and the nucleus, where it could affect transcriptional activity . Furthermore, nuclear-accumulated zyxin executes anti-apoptotic function in cooperation with Akt in myocardiac cells [28, 29].
Human zyxin encodes 572 amino acids protein with two major domains: the N-terminal half domain including the proline-rich region and nuclear export signal or nuclear exclusive signal (NES) and three consecutive C-terminal LIM domains (LIM1-3) (Figure 1B). Previous reports have shown the binding regions for α-actinin, Ena/VASP, CRP, and h-warts/LATS1 [19–22], as described in Figure 1B. Since all identified zyxin clones in our screening contained a C-terminal region and since the LIM domain was implicated in protein-protein interaction, we conducted mapping assays concentrating on the LIM domain of zyxin. To map the precise region for zyxin to bind to SIRT1, a yeast two-hybrid assay was performed using the SIRT1 catalytic domain (aa 185–505) as bait and various shorter zyxin fragments as prey. As shown in Figure 1C, colonies appeared only when the yeast reporter line (AH109) containing the SIRT1 catalytic domain was transformed with the plasmid encoding the zyxin aa 176–572 region on a 4-DO (drop out) medium lacking adenine, histidine, leucin, and tryptophan. Furthermore, these colonies exhibit LacZ+ phenotype on β-Gal assay, indicating that the LIM3 domain (aa 501–572) is required for binding to the SIRT1 catalytic domain. Interestingly, we could not detect any colonies where the yeast reporter line (AH109) containing the SIRT1 catalytic domain was transformed with plasmid encoding full-length zyxin on a 4-DO medium. A previous study showed that full-length zyxin interacts with h-warts/LATS1 in vivo but not in vitro, raising the possibility that LIM1/2 domains are masked in full-length zyxin and some modification might be critical for their interaction . Based on these data, the LIM3 domain is essential for the interaction between SIRT1 and zyxin in yeast; however, we cannot rule out the possibility that some modification might play a pivotal role for the interaction in physiological conditions.
Next, to confirm the interaction between endogenous SIRT1 and GFP-zyxin, we performed a co-immunoprecipitation assay using HEK292T cells expressing SIRT1 endogenously. HEK293T cells were transfected with GFP-zyxin expressing plasmid, and cell lysates were immunoprecipitated with anti-SIRT1 antibody or control IgG, followed by Western blot analysis with anti-GFP antibody (Figure 4C). GFP-zyxin was detected in the immunoprecipitates with anti-SIRT1 antibody, but not in those with control IgG, indicating the interaction between endogenous SIRT1 and exogenously overexpressed GFP-zyxin. Taken together, these results suggest that SIRT1 and zyxin, both shuttling between cytosol and nucleus, could interact primarily in the nucleus in HEK293T cells.
Then, to investigate the expression profile of SIRT1 and zyxin transcript in the developmental brain, we performed real-time PCR assay using total RNAs derived from mouse brains at various developmental stages. The expression pattern of zyxin transcript in the developmental brain is also similar to that of SIRT1, with preferential expression at the E13-P1 stage (Figure 5B).
We next examined whether SIRT1 mediates the deacetylation of zyxin in vivo. COS-7 cells were co-transfected with expressing plasmids encoding Myc-tagged SIRT1 and GFP-zyxin in the presence or absence of LMB, and cell lysates were immunoprecipitated with anti-GFP antibody followed by Western blot analysis using anti-Ac-Lys antibody to monitor acetylation levels. As shown in Figure 6B, the signals for acetylated GFP-zyxin were remarkably reduced in the presence of Myc-tagged SIRT (first panel, lanes 2 and 4) as compared with the control (first panel, lanes 1 and 3), suggesting that SIRT1 can deacetylate zyxin in vivo. The signals for acetylated GFP-zyxin without LMB treatment are stronger as compared to those with LMB treatment (Figure 6B, first panel), while the amounts of total GFP-zyxin protein in immunoprecipitates are comparable (Figure 6B, second panel). This suggests that SIRT1 could deacetylate nuclear-accumulated zyxin. To confirm the specificity of deacetylation by SIRT1, we performed an in vivo deacetylation assay using SIRT1 H363Y, a loss-of-function mutant [6, 9]. The signals for acetylated GFP-zyxin were remarkably enhanced by SIRT1 H363Y overexpression (Figure 6C), indicating the specificity for deacetylation by SIRT1. To strengthen this result, we then examined the effect of SIRT1 inhibitor, nicotinamide (NAm), in an in vivo deacetylation assay. HEK293T cells were transfected with plasmids expressing GFP-zyxin, with or without NAm treatment for 24 h, and GFP-zyxin was immunoprecipitated with anti-GFP antibody. The deacetylation levels were monitored by a Western blot analysis with anti-Ac-Lys antibody. The signals for acetylated GFP-zyxin were remarkably enhanced in the presence of NAm (Figure 6D), indicating the specificity of deacetylation by SIRT1. Taken together, these results clearly show that in mammalian cells, SIRT1 can modulate the deacetylation levels of zyxin, most likely in the nucleus.
The present study shows that zyxin is a novel interacting partner for SIRT1. SIRT1 and zyxin are co-localized in the nucleus after LMB treatment, suggesting that nuclear-accumulated zyxin interacts with SIRT1. SIRT1 and zyxin transcript are both preferentially expressed in developing mouse brain and various adult tissues, including lungs, spleen, and testis. Furthermore, SIRT1 deacetylates zyxin, especially after LMB treatment, raising the possibility that SIRT1 could modulate zyxin's functions in the nucleus via deacetylation. Zyxin-null mice exhibit almost no abnormal phenotype, probably due to genetic redundancy . On the other hand, SIRT1-null mice show growth retardation and developmental defects in various tissues [12, 13], which suggests a critical role of SIRT1 in the developmental stage. Based on these reports, our data could raise the possibility that zyxin is one of the downstream effectors necessary for SIRT1 to execute some biological functions in the developing brain and various adult tissues.
There is an apparent contradiction between the yeast and mammalian results. Apparently SIRT1 does not bind to full length zyxin in yeast because of some hypothesized masking of the LIM3. However, the mammalian cell experiments do show apparent association between full-length zyxin and SIRT1. As mentioned above, a previous study showed that full-length zyxin interacts with h-warts/LATS1 in vivo but not in vitro, raising the possibility that LIM1/2 domains are masked in full-length zyxin and some modification might be critical for their interaction . Therefore we also hypothesize some modification would be required for full-length zyxin to bind to SIRT1 in mammalian cells.
Then what is the biological function of the interaction between SIRT1 and zyxin? Cell survival effect is one of the features shared by these proteins. SIRT1 can protect neurons from oxidative stress in mammalian cells , neurotoxicity in cell-based models for AD/tauopathies, ALS , and Wallerian degeneration . SIRT1 expression levels are induced in mouse models of neurodegeneration , and moderate heart-specific overexpression of SIRT1 in mice delays aging of the heart by conferring resistance to oxidative stress . On the other hand, nuclear-accumulated zyxin can protect cardiomyctes from oxidative stress . These reports raise the possibility that the interaction of these proteins could be implicated in cellular survival, especially in the brain and heart, during physiological senescence. Since zyxin could affect transcriptional activity  and since nuclear-accumulated zyxin executes anti-apoptotic function cooperating with Akt in the nucleus , it is possible SIRT1 could modulate these functions of zyxin via deacetylation. Considering that SIRT1-null mice exhibit developmental defects with frequent exencephaly and retinal phenotype with abnormal proliferation [12, 13], it is possible that the interaction of these proteins could be implicated in cellular survival pathways in the developmental stage.
In addition, shuttling between the cytosol and the nucleus is a characteristic feature shared by SIRT1 and zyxin. In this study, it was not determined whether the acetylation status of zyxin affects its cellular localization or whether deacetylated zyxin is retained in the nucleus. Recently, it was reported that nuclear zyxin, phosphorylated by Akt, interacts with acinus-S to prevent apoptosis and 14-3-3γ plays a pivotal role in the nuclear translocation of zyxin . The morphological studies in the present study show that zyxin is localized primarily in the cytosol with puncta and it accumulates in the nucleus only with the LMB treatment, but not with insulin-like growth factor (IGF-1) or LY294002 treatment, an inhibitor of PI 3-kinase, in COS-7 cells (data not shown). This suggests that some unknown stimuli other than the IGF-1-Akt-related pathway could play a critical role for zyxin to translocate into the nucleus. The nuclear accumulation of zyxin is not induced by co-expression with SIRT1, suggesting that it is unlikely SIRT1 enhances the accumulation and/or translocation of zyxin into the nucleus. Since the sub-cellular localization of SIRT1 differs based on cell type and differentiation [12, 32–34], it is possible that SIRT1 could interact with zyxin in the cytosol depending on the cell type and differentiation.
ECM, beyond scaffolding functions, is responsible for transmitting environmental signals into cells, thereby essentially affecting all aspects of cell life, including its proliferation, differentiation, and death . Since zyxin is considered to convey signals from ECM into the nucleus at the focal adhesion plaques, we now assume that SIRT1 could regulate the signal transmission from ECM into the nucleus by modulating zyxin's functions via deacetylation, thereby reflecting its biological functions.
Finally as described in above, several functional significances could be assumed for the interaction, including the cell survival effect especially in development and/or regulation of signal transmission from ECM into the nucleus. Actually we had performed experiments to investigate whether the interaction of these proteins could affect the apoptosis under stress condition or the transcriptional activity of zyxin in luciferase reporter gene assay (data not shown). Thus far, we could not obtain the result to indicate the biological significance of the interaction in such simple systems. Additional experiments are required to clarify the functional significance of the interaction in the future.
In conclusion, zyxin could be a novel interacting partner of SIRT1. Zyxin is an adaptor protein at focal adhesion plaque, regulating cytoskeletal dynamics and signal transduction to convey signal from the ECM (extracellular matrix) to the nucleus. Our results raise the possibility that SIRT1 regulates signal transmission from ECM to the nucleus by modulating the functions of zyxin via deacetylation.
Human embryonic kidney cell line HEK 293T cells and African green monkey kidney fibroblast like cell line COS-7 cells were cultured in DMEM (Gibco) supplemented with 10% fetal bovine serum (FBS) and antibiotics (50 U penicillin-streptomycin) at 37°C in a 5% CO2 atmosphere inside a humidified incubator. Transient transfection was performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions.
The FLAG-tagged mammalian expression vectors encoding wild-type (wt) human SIRT1 or a mutant catalytically inactive SIRT1 , SIRT1 H363Y, were obtained from Addgene (USA). The mammalian expression vectors encoding the Myc-tagged form of human SIRT1 were generated by sub-cloning from SIRT1 wt or SIRT1 H363Y cDNA fragments into the Sac1-Sal1 site of the pMyc-C1 mammalian expression vector containing the Myc epitope tag . SIRT1 wt cDNA were sub-cloned into pGEX-6P-3 (GE Healthcare) for bacterial GST fusion expression.
pEGFP-C1-zyxin was provided by Dr. Saya (Kumamoto University School of Medicine, Japan). For the pull-down assay, cDNA fragments encoding full-length and deletion mutants of zyxin (as illustrated in Figure 2A) were amplified by PCR from pEGFP-zyxin and sub-cloned into pHA-C1 mammalian expression vector containing the HA epitope tag . All constructs were verified by DNA sequencing.
Monoclonal anti-HA (HA-7; Sigma-Aldrich), anti-c-Myc (9E10; Santa Cruz Biotechnology, Inc.), and anti-SIRT1 (2G1/F7; Upstate), polyclonal anti-Acetylated lysine (Cell Signaling technology) were used.
Yeast two-hybrid screening was conducted using Matchmaker GAL4 two-hybrid system 3 (Clontech). The region containing the deacetylation domain of SIRT1 (amino acid (aa) 185–505) was generated by PCR, subcloned downstream of the GAL4 DNA-binding domain in pGBKT7, and introduced into the yeast strain AH109 as bait. The human fetal brain cDNA library in pGADT7 was introduced into the yeast strain Y187. These two yeast strains were combined according to the yeast mating protocol recommended by the manufacturer. Diploid cells were screened for growth on SD agar plates lacking adenine, histidine, leucine, and tryptophan (4 drop out).
To confirm the yeast two-hybrid assay results, the yeast strain AH109 was retransformed with pGBKT-SIRT1/bait (aa 185–505 of SIRT1) and histidine and adenine positive clones. Then, the positive clones were further evaluated for beta-GAL expression by colony-lift filter assay.
The plasmids were isolated from a colony showing beta-GAL activity using an alkaline solution (3% SDS, 0.2 M NaOH), transformed into bacteria by heatshock or electroporation, and then DNA sequenced. The clones encoding the partial protein fragments of zyxin were determined.
Transiently transfected HEK 293T cells were lysed with a lysis buffer (0.5% NP-40, 20 mM HEPES, 150 mM NaCl, 1 mM EDTA, protease inhibitor cocktail (Roche)). Cell lysates were pre-cleared with GST immobilized on glutathione-sepharose 4B (GE Healthcare) for 1 h. GST fusion protein (10 μg) was immobilized on glutathione-sepharose beads at 4°C for 4 h. Then the glutathione-sepharose beads were incubated with cell lysate in 500 μL of the lysis buffer for 4 h at 4°C. After washing with a wash buffer (0.1% NP-40, 20 mM HEPES, 150 mM NaCl, 1 mM EDTA), the bound proteins were eluted by boiling for 5 min in a sample buffer (50 mM Tris-HCl (pH 6.8), 2% SDS, 8% glycerol, 4% mercaptoethanol, 0.04% bromophenol blue), and analyzed by western blot analysis.
Isolated mouse brains at various ages and various normal tissues were flash-frozen in liquid nitrogen and stored at -80°C until RNA purification. Total RNA was extracted using Trizol (Invitrogen) according to the manufacturer's instructions. RNA (2 μg) was reverse transcribed to produce cDNA using Ready-To-Go RT-PCR beads (GE Healthcare).
Real-time PCR was performed using SYBR GreeenER qPCR SuperMis for ABI PRISM (Invitrogen). All real-time PCR assays were performed in triplicate using a 7300 Real-Time PCR System (Applied Biosystems) as follows: 50°C for 2 min, 95°C for 10 min, 40 cycles at 95°C for 20 s and at 60°C for 1 min. Primers for the mouse Sir2α (SIRT1) mRNA were designed as described previously . Primers for mouse zyxin were designed using Primer Express version 3.0 (Applied Biosystems). For standardization of relative mRNA expression, rodent GAPDH primers were used. The specificity of each primer set was determined with a pre-test showing the specific amplification for a specific gene by gel visualization and sequencing. The results of cycle threshold values (Ct values) were calculated by the ΔΔCt method to obtain the fold differences.
COS-7 cells transfected with plasmids encoding GFP-zyxin and/or HA-SIRT1 were cultured for 42 h and treated with or without Leptomycin B (LMB) for the next 6 h. Then, the cells were fixed with 4% paraformaldehyde for 1 h and then rehydrated by PBS. Cells were permeabilized and non-specific sites were blocked by incubating with PBS containing 0.5% Triton X-100 and 5% bovine serum albumin (BSA). The cells were incubated with antibodies diluted in a blocking solution for 1 h at 4°C. Then, the cells were washed in PBS and incubated with secondary antibodies, Alexa fluor, for 1 h at room temperature. After immuno staining, the slides were mounted with a fluorescent mounting medium (DakoCytomation). Confocal images were obtained using a Fluoview FV1000 (Olympus) laser scanning microscope. XZ and YZ sections were created when all Z-series sections at 0.50 μm intervals were merged.
HEK293T cells were transfected with plasmids encoding GFP-zyxin and/or HA-SIRT1 using Lipofectamine 2000 reagent as previously described. The transfected cells were cultured for 42 h and treated with or without leptomycin B (LMB) for the next 6 h. To immunoprecipitate the expressed GFP-tagged proteins, the cells were lysed in a lysis buffer (0.5% NP-40, 20 mM HEPES (pH 8.0), 150 mM NaCl, 1 mM EDTA, protease inhibitor cocktail (Roche)). The whole-cell lysates were pre-cleared with rProtein A sepharose Fast Flow (GE Healthcare) at 4°C for 1 h. The supernatants were incubated at 4°C for 4 h with anti-GFP antibody or anti-HA antibody and rProtein A sepharose. After washing three times in a wash buffer, the immunoprecipitates were boiled in a sample buffer containing 12% beta-mercaptoethanol for 5 min and subjected to western blot analysis.
Western blot analysis was modified as described previously . In brief, cell lysates or immunoprecipitated proteins were separated on SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Millipore). The membrane was blocked with 5% nonfat dry milk in PBS containing 0.05% Tween 20 (PBS-T) and incubated for 2 h at room temperature with the first antibody diluted in PBS-T containing 1% nonfat dry milk. After washing with PBS-T, the membrane was incubated with horseradish peroxidase-linked anti-mouse IgG antibody or anti-rabbit IgG antibody (Cell Signaling Technology). For detection, an ECL chemiluminescence system (GE Healthcare) was used.
For the in vitro deacetylation assay, zyxin was immunoprecipitated and incubated with GST-SIRT1. The deacetylation reaction was performed as reported previously . COS-7 or HEK293T cells were transfected with plasmids encoding GFP-zyxin and/or Myc-SIRT1. The cells were cultured for 48 h in the presence or absence of LMB during the last 6 h. The cells were lysed with an RIPA buffer (20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 0.1% SDS, 1% Triton X-100, 1% sodium deoxycholate, protease inhibitor cocktail (Roche)). Lysates were immunoprecipitated with anti-GFP antibody and the acetylation levels of zyxin were detected by anti-acetylated lysine antibody.
We are grateful to Dr. Yuji Nakayama at the Dept. of Molecular Cell Biology, Graduate School of Pharmaceutical Sciences, Chiba University, and all Yamashita's lab members for critical advice and discussions. We also thank Professor Hideyuki Saya (Kumamoto University School of Medicine) for providing the zyxin expression vector.
This work was supported by a Research Grant from the National Institute of Biomedical Innovation (05-12) and a Grant-in-Aid for Young Scientists (S) from JSPS.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.