View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 585 (2011) 2291–2294

journal homepage: www.FEBSLetters.org

Regucalcin (RGN/SMP30) alters agonist- and thapsigargin-induced cytosolic [Ca2+] transients in cells by increasing SERCA Ca2+ATPase levels ⇑ Pei Lai, Nga Chi Yip, Francesco Michelangeli

School of Biosciences, The University of Birmingham, Edgbaston, Birmingham B15 2TT, United Kingdom

article info abstract

Article history: Regucalcin (RGN), also reported as senescence marker -30 (SMP30), plays a role in Ca2+ Received 17 May 2011 homeostasis by modulating a number of Ca2+-dependent . RGN also plays a cyto-protective Accepted 26 May 2011 role and its decrease is linked to age-related diseases and cell death. This study shows that RGN Available online 14 June 2011 reduces agonist (histamine)-induced Ca2+ transients in RGN+ transfected COS-7 cells (RGN+) and also increases their Ca2+ storage capacity. These observations are explained by RGN+ cells having Edited by Vladimir Skulachev increased mRNA and protein expression levels of sarco/ Ca2+-ATPase (SERCA). Therefore down-regulation of RGN expression may contribute to characteristics of age-dependent Keywords: Ca2+ homeostasis dis-regulation, by decreasing SERCA levels. Regucalcin Senescence marker protein-30 Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Calcium homeostasis Sarco/endo-plasmic reticulum Ca2+-ATPase Aging

1. Introduction Ca2+-dependent protein such as camodulin- and PKC [2,3]. Secondly, RGN has been shown to have intrinsic enzy- Regucalcin was first identified in the late 1970s as an abundant matic activity, acting as a gluconolactonase which is involved in Ca2+-binding protein in rat [1], which was involved in regulat- the biosynthesis of in some mammals [14]. It may also ing Ca2+ signalling [2]. It has since been shown to be expressed in a act as a detoxifying enzyme in the liver as it can metabolise toxins wide variety of other tissues [2,3]. Although predominantly found such as diisopropyl-phosphofluoridate [15]. Thirdly, RGN over- in the cytosol, it has also been found in the nucleus [4] and mito- expression in NRK52E cells can repress the expression of L-type chondria [5]. The expression levels of RGN protein in some tissues Ca2+ channel and the Ca2+-sensing receptor [16], suggesting that is age-dependent, and therefore commonly referred to as senes- RGN may also be a transcriptional regulator. All of these roles cence marker protein-30 (SMP30) [6]. RGN increases steadily dur- may therefore work in concert to account for its cyto-protective ing neonatal development rising to a maximum in early adulthood. effects to oxidative stress and toxic insult [2,3,17]. The level of RGN then drops substantially in old age [6]. Changes in Given RGNs ability to stimulate Ca2+-ATPases [2,11–13], the the expression of RGN has been associated with a number of path- main aim of this study was to examine whether RGN expression ological conditions, such as cancer [3], Alzheimer’s disease [7] and can modulate intracellular [Ca2+] transients caused by agonist [3,8]. Furthermore, RGN knockout mice have been stimulation or thapsigargin within cells. shown to have a number of conditions often associated with old age including; hearing loss [9], increased lipid accumulation in 2. Materials and methods the liver [8] and poor glucose tolerance [8,10]. RGN is a multifunctional protein playing several different func- Anti-RGN was a custom-made polyclonal antibody from 2+ tions inside cells. It was first shown to play a regulatory role in Ca Covalab, using the full length rat recombinant RGN protein as the 2+ homeostasis by modulating the activity Ca -binding proteins, antigen. Anti-SERCA antibody (Y1F4) was a gift from Dr. Malcolm 2+ 2+ such as Ca -ATPases/Ca pumps [2,11–13], [2] and East, University of Southampton, UK. Both SMP30-pcDNA3.1(À) (containing human RGN coding insert) and control pcDNA3.1(À) (with no coding sequence insert) plasmids were gifts from Prof. J. Abbreviations: RGN, regucalcin; SMP30, senescence marker protein 30; SERCA, Lee (Pusan National University, Korea). Cells transfected with sarcoplasmic/endoplasmic recticulum Ca2+ ATPase; PCR, polymerase chain reaction ⇑ Corresponding author. Fax: +44 121 414 5925. either the SMP30 containing plasmid or the empty plasmid are + À E-mail address: [email protected] (F. Michelangeli). referred to as RGN or RGN , respectively.

0014-5793/$36.00 Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2011.05.058 2292 P. Lai et al. / FEBS Letters 585 (2011) 2291–2294

COS-7 cells were cultured in DMEM media (containing 5.5 mM glucose) (Lonza), supplemented with 10% v/v FBS (Lonza), 2 mM L-glutamine and 1% v/v penicillin-streptomycin (PAA). For trans- fection, they were seeded at a density of 1 Â 105 cells per well of a 12-well plate (or onto coverslips) and Genejuice transfection reagent (Novagen) with plasmid DNA was used following manu- facturer’s instructions. Transfection was done 24 h after seeding cells and transfected cells were used 48 h after transfection. mRNA from RGN+ and RGNÀ cells were extracted and converted into cDNA as described in [18]. The level of cDNA (corresponding to mRNA for specific products) was quantified via PCR amplifica- tion using appropriate primers. In all PCR reactions; BIOTAQTM (2.5U/ reactions), 2 mM dNTP, 2 mM MgCl and between 10 and 30 pmol of the primers were used. The primer sequences for: RGN were (Forward: AATGATGGGAAGGTGGATC, Reverse: GTCAT AGTCAAAGGCATCC); for SERCA2B were (Forward: CCCTTGG TTGTACTTCTGTTA, Reverse: CCTTCACAAACATCTTGCTC); and for b-Actin were (Forward: ATGGATGATGATATCGCCGC; Reverse: TGATCTGGGTCATCTTCTCG). The annealing temperatures were Fig. 1. Over-expression and localisation of Regucalcin in COS-7 cells. (A) Immuno- 47 °C, 49 °C and 52 °C, respectively and the number of cycles was blot of lysates (20 lg) of RGN+ or RGNÀ cells, where the proteins were immuno- 37 for SERCA2B and b-Actin and 35 for RGN. PCR products were stained with the anti-RGN antibody. (B) PCR product of cDNA prepared from RGN+ then separated and visualised as described in [18]. and RGNÀ mRNA and probed with RGN primers. (C) Transfected cells immuno- Western blotting was as described previously [18] with the stained with the anti-RGN antibody. The blots, gels and immuno-micrographs are typical of three replicated experiments. following modifications. Lysates were prepared by adding lysis buffer (50 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 1% Triton X100, 1% sodium deoxycholate, 0.1% SDS and 1 mM PMSF, pH7.4) À to transfected cells attached to the surface of wells in a 12-well and RGN cells showed that the height of the response peaks plate. The antibody dilution ratios used were 1:50 for anti-RGN (from basal to maximum fluorescence intensity levels) were and 1:800 for anti-SERCA (YIF4). Immunohistochemical labelling statistically different (P < 0.05) (n = 5 and 125 cells). Fig. 2C of cells was as described previously [18] using a 1:100 dilution of shows that the mean change in peak heights in response to thapsi- + the anti-RGN. gargin for RGN cells were on average, 40% greater than for the À Intracellular Ca2+ imaging was as described previously [18] but RGN cells. Whereas mean peak heights in response to 10 lM + with the following modifications: Fluo-4 AM (Invitrogen) was used histamine for RGN cells were on average 27% lower compared to À at a final concentration of 7.5 lM. 1 lM thapsigargin or 10 lM his- RGN cells. 2+ + tamine (an agonist which mobilizes Ca through IP3 generation) One possible explanation for these findings is that RGN cells 2+ was added to cells 1 minute after initiating recording and each could have increased levels of SERCA Ca pump. Fig. 3 shows that + À recording lasted for 5 minutes in total. In order to focus on changes when the RGN and RGN cells where probed with an anti-SERCA in cytosolic Ca2+ concentrations occurring from internal stores, antibody, the SERCA protein levels were found to be 21% higher in + À prior to initiating the recordings, the Hank’s buffered saline RGN cells than RGN cells, which was statistically significant solution was supplemented with 1.26 mM EGTA such that the cells (P < 0.05) (n = 3). In order to ascertain whether this increase in were in the absence of extracellular Ca2+. SERCA was due to increased transcription rather than protein turnover, the cDNA for SERCA2B (the isoform found in these cells) was measured using PCR in both RGN+ and RGNÀ cells. Again from 3. Results Fig. 3 it is clear that the transcript is greatly enhanced (an increase of 65%) in RGN+ compared to RGNÀ cells, while the levels of b-actin Over-expression of RGN at both the protein and mRNA level transcript (a control) was very similar in both cells. were demonstrated by the presence of an anti-RGN antibody + immuno-reactive band at 33 kDa from RGN cell lysates (Fig. 1A) 4. Discussion and the elevated level of the PCR product at 237 bp, in the cDNA + generated from RGN transfected COS-7 cells (Fig. 1B). Little or This study has also shown that cells over-expressing RGN have no RGN was detected by either method when the cells were trans- increased levels of SERCA both at the mRNA and protein level. This À fected with the empty vector (RGN ). Immunofluorescence also study also showed that RGN+ cells had increased Ca2+ storage demonstrated RGN over-expression was greatly increased in cells capacity (as determined by the increase in thapsigargin-induced + À transfected with the RGN plasmid compared to RGN cells, with Ca2+ mobilization) and decreased levels agonist-stimulated predominant labelling occurring in and around the nuclear region cytosolic [Ca2+] transients in intact cells. These observations are (Fig. 1C). Transfection efficiency was >70%. consistent with other studies in which SERCA expression was 2+ Intracellular Ca transients, which are represented by plots of altered within cells. In a study by Brini et al. [19] in which SERCA2B fluorescence intensity change over time measured from transfec- was over-expressed in CHO cells, an increase in the ER Ca2+ pool + ted cells loaded with Fluo-4 AM, were visible for both RGN and was also observed. In another study by Green et al. [20] using fibro- À RGN control cells. In order to achieve statistical relevance, 5 blasts in which SERCA2B was ‘knocked-down’, the level of thapsi- individual transfections were undertaken and in each experiment gargin-induced Ca2+ release was reduced compared to control cells. 25 cells from the field-of-view were analysed (individual traces), All these observations therefore indicate that SERCA levels directly 2+ such that each graph represented about 125 cells. Ca -mediated control the levels of ER luminal [Ca2+]. Furthermore, the Brini et al. + À responses were seen in both RGN and RGN cells when stimulated [19], study also showed that SERCA2B over-expression decreased with either 1 lM thapsigargin (Fig. 2A), or 10 lM histamine the amount of agonist-induced Ca2+ release into the cytoplasm. 2+ + (Fig. 2B). Analysis of the Ca transients when comparing RGN This was explained by the fact that the opening probability of P. Lai et al. / FEBS Letters 585 (2011) 2291–2294 2293

Fig. 2. Ca2+ transients following thapsigargin and histamine stimulation of RGN+ and RGNÀ transfected cells. Each panel (in A and B) shows five separate experiments with each individual trace representing the mean data from 25 cells (125 cells in total). (A) Effects of 1 lM thapsigargin on RGN+ and RGNÀ cells. (B) Effects of 10 lM histamine on these cells. The stimulants were added about 60 s following the initiation of the recording. (C) Means of change in peak heights measured from the traces shown in A and B. ⁄ Each change in peak height was determined as the distance from basal to maximum F/F0 values reached for each trace. Indicates that a statistically significant difference in peak heights between RGN+ and RGNÀ cells were observed (⁄P < 0.05).

Fig. 3. Increased SERCA expression in RGN+ cells. The figure shows the differences in SERCA expression levels (at the protein level and mRNA levels) between RGN+ and RGNÀ cells. The figure shows the immuno-stained SERCA blot and the SERCA2B-specific PCR product from the cells. The PCR product for b-actin was also determined for the RGN+ and RGNÀ cells as a control. Each experiment was repeated three times and the mean ratios ± SD, between RGN+ and RGNÀ cells are presented. ⁄ and ⁄⁄ indicates a statistical significance between RGN+ and RGNÀ cells (⁄P < 0.05 and ⁄⁄P = P < 0.01).

2+ 2+ the IP3-sensitive Ca channel is regulated by luminal ER [Ca ], the IP3 receptor is also regulated by kinase-dependent phosphory- such that if the ER Ca2+ stores contain higher [Ca2+] this will de- lation [22] and since RGN can also alter kinase-dependent activity crease the open probability of the channel causing less Ca2+ to be [2,3] it could be envisaged that altered phosphorylation states of 2+ released through the IP3-sensitive Ca channel [21]. An alternative the IP3 receptor, caused by RGN over-expression, may also contrib- explanation in our study could be that as the opening probability of ute to reduced channel opening. The role of SERCA2B in regulating 2294 P. Lai et al. / FEBS Letters 585 (2011) 2291–2294

2+ 2+ Ca release through the IP3-sensitive Ca channel was also [7] Sun, L.W., Wang, L., Sun, Y., Tang, S.W. and Hu, Y.H. (2006) Protective effects of confirmed in xenopus oocytes studies where over-expressing EUK4010 on beta-amyloid(1–42) induced degeneration of neuronal cells. Eur. 2+ J. Neurosci. 24, 1011–1019. SERCA2B caused a diminished amount of Ca release induced by [8] Yamaguchi, M. (2010) Regucalcin and metabolic disorders: osteoporosis and low levels of photo-released IP3 compared to controls [20]. hyperlipidemia are induced in regucalcin transgenic rats. Mol. Cell. Biochem. Our studies show that RGN predominantly localises in and 341, 119–133. + [9] Kashio, A., Amano, A., Kondo, Y., Sakamoto, T., Iwamura, H., Suzuki, M., around the nucleus of RGN cells. Earlier studies have also shown Ishigami, A. and Yamasoba, T. (2009) Effect of vitamin C depletion on age- RGN to translocate from the cytosol to the nucleus and to bind to related hearing loss in SMP30/GNL knockout mice. Biochem. Biophys. Res. nuclear proteins and DNA [16,23]. In addition, RGN can alter the Commun. 390, 394–398. [10] Hasegawa, G., Yamasaki, M., Kadono, M., Tanaka, M., Asano, M., Senmaru, T., expression of a number of including those associated with Kondo, Y., Fukui, M., Obayashi, H., Maruyama, N., Nakamura, N. and Ishigami, tumours. The expression of oncogene c-src mRNA was significantly A. (2010) Senescence marker protein-30/gluconolactonase deletion worsens suppressed in hepatoma cells over-expressing regucalcin, while, glucose tolerance through impairment of acute insulin secretion. Endocrinology 151, 529–536. the expression of the tumor suppressor gene p53, was significantly [11] Yamaguchi, M. and Mori, S. (1989) Activation of hepatic-microsomal Ca2+- enhanced [23]. In these studies it was suggested that RGN is acting adenosine triphosphatase by calcium-binding protein regucalcin. Chem. as a transcriptional regulator, possibly interacting with transcrip- Pharmaceut. Bull. 37, 1031–1034. 2+ tion factors or repressors. Our study now shows that RGN can also [12] Karota, H. and Yamaguchi, M. (1997) Regucalcin increases Ca -ATPase activity and ATP-dependent calcium uptake in the microsomes of rat increase the expression of SERCA2B, by directly or indirectly cortex. Mol. Cell. Biochem. 177, 201–207. enhancing transcription, possibly via similar mechanisms. [13] Yamaguchi, M., Morooka, Y., Misawa, H., Tsurusaki, Y. and Nakajima, R. (2002) However, detailed studies to investigate the effects of RGN on Role of endogenous regucalcin in transgenic rats: suppression of kidney cortex cytosolic activity and enhancement of heart muscle the region of the SERCA2B gene would be required to microsomal Ca2+-ATPase activity. J. Cell. Biochem. 86, 184–193. more fully elucidate this possibility. [14] Kondo, Y., Inai, Y., Sato, Y., Handa, S., Kubo, S., Shimokado, K., Goto, S., It is well established that aging and cell death is associated with Nishikimi, M., Maruyama, N. and Ishigami, A. (2006) Senescence marker 2+ protein-30 functions as gluconolactonase in L-ascorbic acid biosynthesis, and defective Ca homeostasis, particularly when cells experience its knockout mice are prone to scurvy. Proc. Natl. Acad. Sci. USA 103, aberrant elevated levels of cytosolic Ca2+ levels [24]. SERCA plays 5723–5728. a fundamental role in minimising these abnormal changes [25]. [15] Kondo, Y., Ishigami, A., Kubo, S., Handa, S., Gomi, K., Hirokawa, K., Kajiyama, N., Chiba, T., Shimokado, K. and Maruyama, N. (2004) Scenescence marker Previous studies have suggested that RGN can directly bind to protein-30 is aunique enzymethat hydrolyzes diisopropylphosphofluoridate 2+ and stimulate Ca pump activity [2,11–13] and now from this in the liver. FEBS Lett. 570, 57–62. study we also show evidence that the expression of SERCA Ca2+ [16] Nakagawa, T. and Yamaguchi, M. (2006) Overexpression of regucalcin enhances its nuclear localization and suppresses L-type Ca2+ channel and pumps has increased. Therefore during the aging process when calcium-sensing receptor mRNA expressions in cloned normal rat kidney RGN levels fall this would concomitantly reduce SERCA levels as proximal tubular epithelial NRK52E cells. J. Cell. Biochem. 99, well as directly reducing SERCA activity, exposing cells to abnor- 1064–1077. mal [Ca2+] fluctuations, leading to cell death. [17] Son, T.G., Kim, S.J., Kim, K., Kim, M.S., Chung, H.Y. and Lee, J. (2008) Cytoprotective roles of senescence marker protein 30 against intracellular calcium elevation and oxidative stress. Arch. Pharm. Res. 31, Acknowledgements 872–877. [18] Lai, P. and Michelangeli, F. (2009) Changes in expression and activity of the secretory pathway Ca2+ ATPase1 (SPCA1) in A7r5 vascular smooth muscle cells The authors would like to thank Prof. J. Lee, Dr. N. Hotchin and cultured at different glucose concentrations. Biosci. Rep. 29, 397–404. Dr. J.M. East for supplying materials. The BBSRC is also thanked for [19] Brini, M., Bano, D., Manni, S., Rizzuto, R. and Carafoli, E. (2000) Effects of PMCA 2+ the Ph.D. studentship given to P. Lai. and SERCA pump overexpression on the kinetics of cell Ca signalling. EMBO J. 19, 4926–4935. [20] Green, K.N., Demuro, A., Akbari, Y., Hitt, B.D., Smith, I.F., Parker, I. and LeFerla, References F.M. (2008) SERCA pump activity is physiologically regulated by presenilin and regulates amyloid b production. J. Cell Biol. 181, 1107–1116. [1] Yamaguchi, M. and Yamamoto, T. (1978) Purification of calcium-binding [21] Bezprovanny, I. and Ehrlich, B.E. (1994) Inositol (1,4,5)-trisphosphate (InsP3)- substance from soluble fraction of normal rat liver. Chem. Pharmaceut. Bull. gated Ca channels from cerebellum: conduction properties fro divalent cations 26, 1915–1918. and regulation by intraluminal calcium. J. Gen. Physiol. 104, 821–856. [2] Yamaguchi, M. (2000) Role of regucalcin in . Life Sci. 66, [22] Vanderheyden, V., Devogelaere, B., Missiaen, L., Humbert De Smedt, H., 1769–1780. Bultynck, G. and Parys, J.B. (2009) Regulation of inositol 1, 4, 5-trisphosphate- 2+ [3] Yamaguchi, M. (2005) Role of regucalin in mechanisms of cell homeostasis and induced Ca release by reversible phosphorylation and dephosphorylation, function. Int. J. Mol. Med. 15, 371–389. Biochim. Biophys. Acta 1793, 959–970. [4] Nakagawa, T. and Yamaguchi, M. (2008) Nuclear localization of regucalcin is [23] Tsurusaki, Y. and Yamaguchi, M. (2004) Role of regucalcin in liver nuclear enhanced in culture with C activation in cloned normal rat function: Binding of regucalcin to nuclear protein or DNA and modulation of kidney proximal tubular epithelial NRK52E cells. Int. J. Mol. Med. 21, 605–610. tumor-related . Int. J. Mol. Med. 14, 277–281. [5] Yamaguchi, M., Takakura, Y. and Nakagawa, T. (2008) Regucalcin increases [24] Toescu, E.C. and Verkhratsky, A. (2007) The importance of being subtle: small Ca2+-ATPase activity in the mitochondria of brain tissues of normal and changes in calcium homeostasis control cognitive decline in normal aging. transgenic rats. J. Cell. Biochem. 104, 795–804. Aging Cell 6, 267–273. 2+ [6] Fujita, T., Shirasawa, T., Uchida, K. and Maruyama, N. (1996) Gene regulation of [25] Michelangeli, F. and East, J.M. (2011) A diversity of SERCA Ca pump senescence marker protein-30 (SMP30): coordinated up-regulation with inhibitors. Biochem. Soc. Trans. 39, 789–797. tissue maturation and gradual down-regulation with aging. Mech. Ageing Develop. 87, 219–229.