Transgelin: An actin-binding protein and tumour suppressor

Available online at www.sciencedirect.com

The International Journal of Biochemistry & Cell Biology 41 (2009) 482–486

Molecules in focus

Transgelin: An actin-binding protein and tumour suppressor Stephen J. Assinder a,∗ , Jo-Ann L. Stanton b , Priya D. Prasad b a

Discipline of Physiology, School of Medical Sciences & Bosch Institute, University of Sydney, NSW 2006, Australia b Department of Anatomy & Structural Biology, School of Medical Sciences, University of Otago, P.O. Box 913, Dunedin, New Zealand Received 5 February 2008; received in revised form 14 February 2008; accepted 14 February 2008 Available online 10 March 2008

Abstract Transgelin is a shape change sensitive 22 kDa actin-binding protein of the calponin family. It contains a C-terminal calponinlike module (CLIK23 ) and an upstream positively charged amino acid region required for actin binding. Transgelin is ubiquitous to vascular and visceral smooth muscle and is an early marker of smooth muscle differentiation, where its expression is driven by CArG box, smooth muscle gene promoter. It is also present in fibroblasts, and some epithelium where expression is likely driven by TGF-␤1. Transgelin null mice reveal that, whilst it is not required for smooth muscle development, transgelin may be involved in calcium-independent smooth muscle contraction. Recent evidence suggests that transgelin acts as a tumour suppressor. Its expression is lost in prostate, breast and colon cancers. This is consistent with suppression of the metallo matrix protease-9 (MMP-9) by transgelin, where MMP-9 is upregulated in these common cancers. © 2008 Elsevier Ltd. All rights reserved. Keywords: Transgelin; SM22; TGF-␤; MMP-9; Cancer

1. Introduction Transgelin (TAGLN) was first identified in 1987 (Lees-Miller, Heeley, Smillie, & Kay, 1987) as a 22 kDa protein of unknown function and named as SM22. It has since been identified in many species and studied under several different monikers of mouse p27, WS3-10, transgelin and SM22 (Lawson, Harris, & Shapland, 1997). Both gene and protein sequences are highly conserved across species and homologies have been identified in Drosophila melanogaster (mp20; Ayme-Southgate, Lasko, French, & Purdue, 1989), and in Caenorhabditis elegans (unc-87; Goetnik & Waterson, 1994). It is a member of the calponin family of actin-binding proteins

∗

Corresponding author. Tel.: +61 2 93512514; fax: +61 2 93518400. E-mail address: [email protected] (S.J. Assinder).

1357-2725/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2008.02.011

and is localised to the cytoskeletal apparatus (Lawson et al., 1997). 2. Structure The human transgelin is transcribed from a 5.4 kb gene located on chromosome 11q23.2 (CamorettiMercado et al., 1998). It consists of 5 exons, a large first intron and 3 short introns (Fig. 1a). An approximately 800 bp upstream region contains several putative promoter sites (Camoretti-Mercado et al., 1998). TAGLN generates a 1556 bp transcript (Ensembl ID: ENST00000278968). The whole of exon 1 and first 12 bp of exon 2 represent a 5 -untranslated region, whilst the final 432 bp of exon 5 account for the 3 -untranslated region (Fig. 1a). The translated protein derived from this transcript is a 201 amino acid peptide. Members of the calponin family are characterised by two specific motifs,

S.J. Assinder et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 482–486

483

Fig. 1. Human transgelin gene and protein structures. (a) Gene structure showing exonic (numbered boxes) and intronic (line) regions and size. The expanded promoter sequence provides relative positions of promoter elements upstream (−) and downstream (+) of the transcription start site (0). Serum response factor binding CArG boxes are shown, as are the transforming growth factor ␤1 control element (TCE) and Smad-binding element (SBE). The resulting transcript is represented showing the translated region (solid line) and the five prime and three prime untranslated regions (5 -UTR and 3 -UTR, respectively; broken line). (b) Primary protein structure gives the amino acid (aa) positions corresponding to the calponin homology (CH) domain containing a putative Ca2+ -binding site (EF Hand) and a calponin-like module (CLIK23 ). Starred sites indicate potential caseine kinase II (CKII) and protein kinase C (PKC) target sites.

a N-terminal calponin homology domain (CH) and one or more copies of a C-terminal calponin-like module (CLIK23 ). Transgelin features a single CLIK23 module and N-terminal CH domain that contains a potential Ca2+ -binding EF-hand (Fig. 1b; Fu et al., 2000; Lawson et al., 1997). As with the CLIK23 repeats of calponin, the C-terminal CLIK23 domain of transgelin is required for actin binding (Danninger & Gimona, 2000; Fu et al., 2000). C-terminal truncation of transgelin demonstrates that multiple regions within the final quarter of the protein are required for full actin binding activity (Fu et al., 2000). This includes an upstream externalCLIK23 sequence of 154-KKAQEHKR-161 where both positively charged amino acids KK (154/155) and KR (160/161) are involved in actin binding. Replacement of these positively charged residues with uncharged leucine significantly reduces actin binding but does not totally abolish it (Fu et al., 2000). Only the absence of this sequence and the CLIK23 module results in an ablation of actin binding (Fu et al., 2000). The putative EF-hand calcium-binding site at amino acids 108–119 (Fig. 1b) is non-functional (Fu et al., 2000) with transgelin-actin binding unaffected by presence or absence of calcium.

Three possible protein kinase target sites are also present (Fig. 1b). Transgelin is only sensitive to protein kinase C phosphorylation in vitro which attenuates actin binding (Fu et al., 2000). However, the physiological significance of this is unclear as phosphorylated transgelin has yet to be identified in vivo (Gimona, Sparrow, Strasser, Herzog, & Small, 1992). 3. Expression Tissue distribution of transgelin is contentious. It is found throughout smooth muscle tissues of normal adult vertebrates (Lees-Miller, Heeley, & Smillie, 1987). Indeed its mRNA levels are highest in adult human tissues containing large amounts of smooth muscle such as uterus, bladder, stomach and prostate (CamorettiMercado et al., 1998; Lawson et al., 1997). However, it is also expressed in several adult tissues containing little or no smooth muscle including spinal chord, adrenal gland (Lawson et al., 1997) and notably in the heart (Camoretti-Mercado et al., 1998; Lawson et al., 1997). Transgelin expression is one of the earliest markers of smooth muscle differentiation. Indeed, transgenic mice

484

S.J. Assinder et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 482–486

studies distinguish both temporal and spatial patterns of expression. Expression of the ␤-galactosidase reporter, driven by the TAGLN promoter, demonstrates that the promoter is active at embryonic (E) day 9.5 (Lepore et al., 2005). Expression is found in smooth muscle cells of the developing vasculature and cardiac myocytes but not somites as previously reported by Xu, Ho, Ritchie, & Li (2003). Promoter activity in the myotomal component of somites described by Xu et al. (2003) is diminished by E16.5, whilst activity in the heart and arterial system is maintained throughout embryogenesis and in the neonate, after which activity decreases. No reporter expression is evident in venous or visceral smooth muscle (Xu et al., 2003). In contrast, the transgelin promoter reporter mouse of Lepore et al. (2005) exhibits promoter activity during later stages of development in smooth muscle throughout the arterial and venous system, as well as in visceral smooth muscle. This pattern of expression is maintained in adulthood with distinct expression in smooth muscle of the bladder, bronchi, stomach and intestine, and in cardiac myocytes (Lepore et al., 2005). Discrepancies between these studies most likely arise from the promoter constructs used to drive the ␤-galactosidase reporter expression. The human (and mouse) promoter region of the transgelin gene contains two CArG boxes at bp −283 and −156 (Fig. 1a; Camoretti-Mercado et al., 1998). The CArG box is a cis-acting regulatory element with a consensus sequence of CC(A/T)6 GG that provides a binding site for serum response factor (Shore & Sharrocks, 1995). The −283 CArG box appears to be unnecessary as deletion of this region in the human promoter does not affect constitutive or serum-induced promoter activity (Camoretti-Mercado et al., 1998). Several other promoter regions are also present including AP-1 and SP1 transcription factor sites (CamorettiMercado et al., 1998). Indeed, the promoter construct fragment, consisting of the 445 bp upstream region of the transgelin transcription start site, used by Xu et al. (2003) is sufficient to drive arterial, but not visceral expression. However, in transgenic mice containing a bacterial artificial chromosome clone (150 kb genomic DNA fragment) spanning the human transgelin gene, expression was seen in arterial, venous and visceral smooth muscle at E13.5 (Xu et al., 2003). Similarly expression of this transgene was expressed in smooth muscle rich tissues, with weak expression in the heart. This demonstrates, as with the mouse construct of Lepore et al. (2005) that regulatory elements other than CArG are required for complete temporal and spatial transgelin expression in vascular and visceral smooth muscle.

Transforming growth factor beta (TGF-␤) also plays an important role in regulating smooth muscle cell gene expression and differentiation. TGF-␤ is known to induce transgelin expression and is partly under the control of a TGF-␤ control element (TCE) in the promoter region (Fig. 1a; Adam, Regan, Hautmann, & Owens, 2000). Also present are several Smad-binding sites (Chen, Kulik, & Lechleider, 2003). TGF-␤ signalling occurs via phosphorylation of receptor associated signal transduction proteins, R-Smads. Phosphorylated R-Smads heterodimerise with cytoplasmic Smads, translocate to the nucleus where they can induce or repress gene expression via Smad-binding elements (SBE). A SBE is present in the 5 -UTR of exon 1 of the transgelin gene (+2 to +25; Chen et al., 2003; Qiu et al., 2005). This binding element is necessary for TGF-␤ induction of transgelin expression via Smad3 binding (Qiu et al., 2005). Interestingly a CArG box-dependent serum response co-activator, myocardin, enhances Smad-3 activation of transgelin expression independently of the CArG box. It is likely that expression of transgelin outside of smooth muscle cells is driven by TGF-␤. Many groups have failed to demonstrate transgelin expression outside of smooth muscle (reviewed in Lawson et al., 1997). Indeed, this is supported by transgenic reporter mice studies described above. In these mice, however, the SBE of exon 1 was excluded from the promoter constructs. Therefore, expression driven by TGF-␤ would not be seen. TGF-␤ is known to increase transgelin mRNA in mesenchymal cells of the adult human prostate (Untergasser et al., 2005) and protein is present in normal mesenchymal cells and fibroblasts (Lawson et al., 1997). Expression has also been identified in intestinal epithelial cells (Shields, Rogers-Graham, & Der, 2002), glomerular epithelial cells (Ogawa et al., 2007), breast ductal epithelium (Shields et al., 2002; Wulfkuhle et al., 2002) and prostate epithelium (Yang et al., 2007). We have also found transgelin expression in prostate epithelial stem cells and some prostate cancer cell lines (unpublished data). 4. Biological function Transgelin is an actin stress fibre-associated protein (Lawson et al., 1997) that acts to gel and stabilise actin gels in vitro. It is also associated with actin filaments in smooth muscle (reviewed in Gimona, Kaverina, Resch, Vignal, & Burgstaller, 2003). Given that transgelin is one of the earliest markers of smooth muscle cell differentiation, and that it can associate with actin, it has long been hypothesised that it has functions

S.J. Assinder et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 482–486

485

allowing for cancer cells to invade surrounding tissues and migrate. 5. Role in pathology

Fig. 2. Proposed biological functions of transgelin. Potential roles for transgelin in both embryogenesis and the adult comprise functions modulated by actin binding (promoter) and those that are not (suppressor).

in regulating development and contractile function of these cells. However, transgelin (SM22␣−/− ) null mice develop normally, are fertile and do not exhibit altered blood pressure, heart rate, histology or morphology of tissues (Zhang et al., 2001). However, closer examination of vascular smooth muscle from transgelin null mice shows that there is increased contractile response to agonist-induced, calcium-independent contraction (Zeidan et al., 2004; Je & Sohn, 2007) (Fig. 2). Transgelin may also be involved in cell migration. It has been shown to associated with a specific sub-population of actin filament bundles that form podosomes (Gimona et al., 2003). It is suggested that a loss of calponin causes a reduction of calponin:transgelin ratio thus favouring the formation of podosomes and predisposing cells to an invasive, malignant nature. Recent evidence indicates that transgelin has tumour suppressive functions in certain cells that are unrelated to the cytoskeleton per se. In prostate carcinoma cells transgelin has been shown to block androgen stimulated cell growth by preventing binding of an androgen receptor co-activator with androgen receptor and subsequent translocation to the nucleus (Yang et al., 2007). Transgelin also acts to suppress expression of the metallo-matrix proteinase (MMP)-9 (Nair, Solway, & Boyd, 2006). MMP-9 is involved in tissue remodelling,

It is becoming increasingly evident that transgelin may act as a tumour suppressor. Disorganisation of cytoskeletal actin filaments is a fundamental event in the development of a cancer cell phenotype. Several proteins bind actin to cross-link and increase filament rigidity, protect against depolarisation, and bundle filaments into stress fibres. Prolonged incubation of cells with TGF-␤ induces formation of stress fibres, inhibiting cell migration and invasion. Activation of the Ras-MEK-ERK-myc signalling pathway antagonises TGF-␤ inhibition of cell motility, and suppresses transgelin expression (Shields et al., 2002), as does immortalisation and transformation by RNA and DNA viruses (Kaplan-Alberquerque, Garat, Van Putten, & Nemenoff, 2003; Lawson et al., 1997). Expression of transgelin is decreased in breast and colon cancers (Shields et al., 2002; Wulfkuhle et al., 2002) and in prostate cancer (Yang et al., 2007; ourselves, unpublished data). It is well known that prostate cancer becomes insensitive to TGF-␤, and that in prostate, breast and colon cancers expression of MMP9 is increased. Therefore, loss of expression of transgelin could account in part for development of some of the hallmarks of cancer cells in these neoplasias. As such, transgelin may provide a target for the development of interventions against these common cancers. References Adam, P. J., Regan, C. P., Hautmann, M. B., & Owens, G. K. (2000). Positive and negative-acting Kruppel-like transcription factor bind a transcription factor ␤ control element required for expression of the smooth muscle cell differentiation marker SM22␣ in vivo. J. Biol. Chem., 275, 37798–37806. Ayme-Southgate, A., Lasko, P., French, C., & Purdue, M. L. (1989). Characterization of the gene for mp20: A Drosophila muscle protein that is not found in asynchronous flight muscle. J. Cell Biol., 108, 521–531. Camoretti-Mercado, B., Forsythe, S. M., LeBeau, M. M., Espinosa, R., Vieira, J. E., Halayko, A. J., et al. (1998). Expression and cytogenetic localization of the human SM22 gene (TAGLN). Genomics, 49, 452–457. Chen, S., Kulik, M., & Lechleider, R. J. (2003). Smad proteins regulate transcriptional induction of the SM22␣ gene by TGF-␤. Nucleic Acids Res., 31, 1302–1310. Danninger, C., & Gimona, M. (2000). Live dynamics of GFP-calponin: Isoform-specific modulation of the actin cytoskeleton and autoregulation by C-terminal sequences. J. Cell Sci., 113, 3725–3736. Fu, Y., Liu, H. W., Forsythe, S. M., Kogut, P., McConville, J. F., Halayko, A. J., et al. (2000). Mutagenesis analysis of human SM22: Characterisation of actin binding. J. Appl. Physiol., 89, 1985–1990.

486

S.J. Assinder et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 482–486

Gimona, M., Kaverina, I., Resch, G. P., Vignal, E., & Burgstaller, G. (2003). Calponin repeats regulate actin filament stability and formation of podosomes in smooth muscle cells. Mol. Biol. Cell, 14, 2482–2491. Gimona, M., Sparrow, M. P., Strasser, P., Herzog, M., & Small, J. V. (1992). Calponin and SM22 isoforms in avian and mammalian smooth muscle-absence of phosphorylation in vivo. Eur. J. Biochem., 205, 1067–1075. Goetnik, S., & Waterson, R. H. (1994). The Caenorhabditis elegans muscle-affecting gene unc-87 encodes a novel thin filamentassociated protein. J. Cell Biol., 127, 79–93. Je, H. D., & Sohn, U. D. (2007). SM22alpha is required for agonist induced regulation of contractility: Evidence from SM22alph knockout mice. Mol. Cells, 23, 175–181. Kaplan-Alberquerque, N., Garat, C., Van Putten, V., & Nemenoff, R. A. (2003). Regulation of SM22␣ expression by argenine vasopressin and PDGF-BB in vascular smooth muscle cells. Am. J. Physiol. Heart Circ. Physiol., 285, H1444–H1452. Lawson, D., Harris, M., & Shapland, C. (1997). Fibroblast transgelin and smooth muscle SM22␣ are the same protein, the expression of which is down regulated in many cell lines. Cell Motil. Cytoskel., 38, 250–257. Lepore, J. J., Cheng, L., Lu, M. M., Mericko, P. A., Morrisey, E. E., & Parmacek, M. S. (2005). High-efficiency somatic mutagenesis in smooth muscle cells and cardiac myocytes in SM22␣-Cre transgenic mice. Genesis, 41, 179–184. Lees-Miller, J. P., Heeley, D. H., & Smillie, L. B. (1987). An abundant and novel protein of 22 kDa (SM22) is widely distributed in smooth muscles. Purification from bovine aorta. Biochem. J., 244, 705–709. Lees-Miller, J. P., Heeley, D. H., Smillie, L. B., & Kay, C. H. (1987). Isolation and characterisation of an abundant and novel 22 kDa protein (SM22) from chicken gizzard smooth muscle. J. Biol. Chem., 262, 2988–2993. Nair, R. J., Solway, J., & Boyd, D. D. (2006). Expression cloning identifies transgelin (SM22) as a novel repressor of 92-kDa type IV collagenase (MMP-9) expression. J. Biol. Chem., 281, 26424–26436.

Ogawa, A., Sakatsume, M., Wang, X. Z., Sakamaki, Y., Tsubata, Y., Alchi, B., et al. (2007). SM22 alpha: The novel phenotype marker of injured epithelial cells in anti-glomerular basement membrane nephritis. Nephron Exp. Nephrol., 106, 77–87. Qiu, P., Ritchie, R. P., Fu, Z., Cao, D. S., Cumming, J., Miano, J. M., et al. (2005). Myocardin enhances Smad3-mediated transforming growth factor-␤1 signaling in a CArG box-independent manner. Circ. Res., 97, 983–991. Shields, J. M., Rogers-Graham, K., & Der, C. J. (2002). Loss of transgelin in breast and colon tumours and in RIE Cells by ras deregulation of gene expression through raf-independent pathways. J. Biol. Chem., 277, 9790–9799. Shore, P., & Sharrocks, A. D. (1995). The MADS-box family of transcription factors. Eur. J. Biochem., 229, 1–13. Untergasser, G., Gander, R., Lilg, C., Lepperdinger, G., Plas, E., & Berger, P. (2005). Profiling molecular targets of TGF-␤1 in prostate fibroblast-to-myofibroblast transdifferentiation. Mech. Age Dev., 126, 59–69. Wulfkuhle, J. D., Sgroi, D. C., Krutzsch, H., Maclean, K., McGarvey, K., Knowlton, M., et al. (2002). Proteomics of human ductal carcinoma in situ. Cancer Res., 62, 6740–6749. Xu, R., Ho, Y.-S., Ritchie, R. P., & Li, L. (2003). Human SM22␣ BAC encompasses regulatory sequences for expression in vascular and visceral smooth muscles at fetal and adult stages. Am. J. Heart Circ. Physiol., 284, H1398– H1407. Yang, Z., Chang, Y.-J., Miyamoto, H., Ni, J., Niu, Y. J., Chen, Z. D., et al. (2007). Transgelin functions as a suppressor via inhibition of ARA54-enhanced androgen receptor transactivation and prostate cancer cell growth. Mol. Endocrinol., 21, 343–358. Zhang, J. C. L., Kim, S., Helmke, B. P., Yu, W. W., Du, K. L., Lu, M. M., et al. (2001). Analysis of SM22␣-deficient mice reveals unanticipated insights into smooth muscle cell differentiation and function. Mol. Cell Biol., 21, 1336–1344. Zeidan, A., Sward, K., Nordstrom, I., Ekblad, E., Zhang, J. C. L., Parmacek, M. S., et al. (2004). Ablation of SM22 alpha decreases contractility and actin contents of mouse vascular smooth muscle. FEBS Lett., 562, 141–146.