Androgen-inducible gene 1 increases the ER Ca2 + content and cell death susceptibility against oxidative stress

    Androgen-inducible gene 1 increases the ER Ca 2 + content and cell death susceptibility against oxidative stress Nadine Nickel, Astrid Cleven, Vitalij Enders, Dmitrij Lisak, Lars Schneider, Axel Methner PII: DOI: Reference:

S0378-1119(16)30245-1 doi: 10.1016/j.gene.2016.03.055 GENE 41261

To appear in:

Gene

Received date: Revised date: Accepted date:

7 December 2015 29 March 2016 30 March 2016

Please cite this article as: Nickel, Nadine, Cleven, Astrid, Enders, Vitalij, Lisak, Dmitrij, Schneider, Lars, Methner, Axel, Androgen-inducible gene 1 increases the ER Ca2 + content and cell death susceptibility against oxidative stress, Gene (2016), doi: 10.1016/j.gene.2016.03.055

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Androgen-inducible gene 1 increases the ER Ca2+ content and cell death susceptibility against oxidative stress

PT

Nadine Nickel*, Astrid Cleven, Vitalij Enders, Dmitrij Lisak, Lars Schneider, and Axel

SC RI

Methner#

Focus Program Translational Neuroscience (FTN), Rhine Main Neuroscience Network (rmn2) and Department of Neurology, University Medical Center of the Johannes Gutenberg

NU

University Mainz, Mainz, Germany

*Current address: Cologne Excellence Cluster on Cellular Stress Responses in Aging-

MA

Associated Diseases and Department I of Internal Medicine, University Hospital of Cologne, Cologne, Germany #

ED

Address for correspondence: Axel Methner MD, Johannes Gutenberg University Medical

Center Mainz, Department of Neurology, Langenbeckstr. 1, D-55131 Mainz, Germany, Tel.:

CE

PT

+49-6131-17-2695, Fax: +49-6131-17-5967, Email: [email protected]

AC

Running title: AIG1

ACCEPTED MANUSCRIPT Abstract (199<250) Androgen-induced gene 1 (AIG1) is a transmembrane protein implicated with survival (its

PT

expression level was shown to correlate with the survival of patients suffering from hepatocellular carcinoma) and Ca2+ signaling (over-expression of AIG1 increased

SC RI

transcription mediated by the Ca2+-dependent nuclear factor of activated T cells). We aimed to shed light on this less-studied protein and investigated its tissue expression, genomic organization, intracellular localization and membrane topology as well as its effects on cell

NU

death susceptibility and the Ca2+ content of the endoplasmic reticulum. Immunoblotting of

MA

mouse tissues demonstrated highest expression of AIG1 in the liver, lung and heart. AIG1 has a complex genomic organization and expresses several splice variants in a tissuedependent manner. Analyzing the topology of AIG1 in the ER membrane using a protease-

ED

protection assay suggested that AIG has five transmembrane domains with a luminal N- and cytosolic C-terminus and a hydrophobic stretch between the third and fourth membrane

PT

domain that does not cross the membrane. AIG1 over-expression slightly increased

CE

susceptibility to oxidative stress, which correlated with an increased ER Ca2+ concentration in two different cell lines. Together, these results indicate that AIG1 plays a role in the control of

AC

the intracellular Ca2+ concentration and cell death susceptibility.

Highlights: 

Androgen-induced gene 1 (AIG1) is expressed mainly in the liver, lung and heart



AIG1 has three splice isoforms with a tissue-specific expression



AIG1 has five transmembrane domains with a luminal N- and cytosolic C-terminus



AIG1 over-expression slightly increases susceptibility to oxidative stress



AIG1 over-expression increases the ER Ca2+ concentration

ACCEPTED MANUSCRIPT Introduction Androgen-induced gene 1 (AIG1) is a protein of approximately 27 kDa with five or six

PT

proposed transmembrane domains of an unknown membrane topology, previously cloned from human dermal papilla cells and shown to be inducible by androgen (Seo et al., 2001).

SC RI

AIG1 bears 35% sequence homology to FAR-17a (Seo et al., 2001), an androgen-inducible gene which has been identified in the flank organ of male Syrian hamsters (Ideta et al., 1998). Later, Wu et al. demonstrated that AIG1 is a new interaction partner of the Ring

NU

Finger And CHY Zinc Finger Domain Containing 1 (RCHY1 also known as Pirh2) E3 ligase by yeast two-hybrid screening. They confirmed the interaction between AIG1 and RCHY1 on

MA

the molecular level as well as functionally by showing that AIG1 expression reduced p53 reporter gene activity, a read-out for RCHY1 activity.

They also showed that AIG1

ED

expression activates the nuclear factor of activated T cells (NFAT) signaling pathway (Wu et al., 2011) suggesting a mechanism of action involving alteration of the intracellular Ca 2+

PT

homeostasis.

Since NFAT signaling is a proliferative and survival-promoting pathway (reviewed in (Shou et

CE

al., 2015)) and p53 is involved in apoptosis induction in response to cellular stresses such as

AC

DNA damage (reviewed in (Reinhardt and Schumacher, 2012)), these results suggest that AIG1 is involved in cell death decisions. In line with this, it was previously shown that AIG1 mRNA expression levels were reduced in the majority of human hepatocellular carcinomas when compared to matched, non-cancerous liver tissue, and the survival time of those patients with downregulated AIG1 was much shorter than for patients with upregulated AIG1 expression (Wu et al., 2011). Recently, AIG1 was identified as a novel fusion partner of Nuclear Factor 1/B (NFIB) in chromosomal rearranged salivary adenoid cystic carcinoma (Mitani et al., 2011) and of Golgi SNAP Receptor Complex Member 1 (GOSR1) in Sezary Syndrome, a rare form of cutaneous T-cell lymphoma (Iżykowska et al., 2013). Furthermore, in a screen for resistance against organophosphate pesticides, AIG1 knockout correlated with resistance to chlorpyrifos, an organophosphate pesticide, in the mammalian cell line KBM7 (Zhu et al., 2015).

ACCEPTED MANUSCRIPT We identified AIG1 as a potential interactor of Transmembrane Bax inhibitor motif containing 6 (TMBIM6), also known as Bax inhibitor-1, by screening an adult brain library for TMBIM6 interacting proteins in yeast using the split-ubiquitin technique – a technique used to identify

PT

interaction partners of membrane proteins not amenable to conventional yeast-2-hybrid

SC RI

screens (Wang et al., 2004). TMBIM6’s anti-apoptotic functions and its involvement in the maintenance of the cellular Ca2+ homeostasis (reviewed in (Bultynck et al., 2014; Henke et al., 2011)) perfectly matched the abovementioned potential functions of AIG1, but we were,

NU

unfortunately, not able to prove this interaction unequivocally. We therefore decided to report here only the tissue expression and membrane topology of AIG1 as well as its effects on ER

MA

and oxidative stress and the intracellular Ca2+ homeostasis in two unrelated cell lines. Our results suggest that AIG1 is involved in the control of the ER Ca2+ content and has some

PT

Experimental procedures

ED

effect on cell death susceptibility against oxidative stress.

Cell culture

CE

Mouse embryonic fibroblasts (MEFs) have been described previously (Bultynck et al., 2012; Kiviluoto et al., 2013; 2012). Neuro2a (N2a) cells were established from a spontaneous

AC

neuroblastoma of an albino strain A mouse in 1967 (Olmsted et al., 1970). HT22 cells are murine hippocampal cells subcloned from the HT4 cell line. Since they lack ionotropic glutamate receptors, HT22 cells serve as a model for endogenous oxidative stress induced by the addition of glutamate which blocks glutathione synthesis (Albrecht et al., 2010; Maher and Davis, 1996). MEF and N2a cells were cultured in high glucose containing Dulbecco’s Modified Eagle’s Medium (DMEM high glucose, Invitrogen) supplemented with 10% fetal bovine serum (FBS) and 100 µg/ml streptomycin and 100 units/ml penicillin. HT22 cells were grown in DMEM high glucose supplemented with 5% FBS, 100 µg/ml streptomycin and 100 units/ml penicillin. Transfections and generation of stable cell lines

ACCEPTED MANUSCRIPT High-purity plasmids were prepared using Nucleobond AX-500 columns (Machery and Nagel) and DNA was transfected with Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. For generation of stably AIG1-expressing cell lines, cDNAs of AIG1

PT

or empty vector were inserted into the pPB-CAG-EBNXN vector (obtained via the Sanger

SC RI

Institute) containing PiggyBac transposon sites, which facilitate stable insertion into the genome of a target cell when co-expressed with a transposase (Yusa et al., 2011). The cDNAs were N-terminally tagged with a hemagglutinin (HA) epitope and C-terminally

NU

connected to an internal ribosomal entry site (IRES) followed by the yellow fluorescent protein Venus. These constructs were co-transfected with a plasmid encoding the

MA

transposase in a ratio of 1:4 and 48 h after transfection, Venus positive cells were selected on a MoFlo XDP (Beckman-Coulter) cell sorter. After three to four rounds of cell sorting, approximately 99% of cells were positive for Venus fluorescence and considered as stably

ED

transfected. Successful AIG1 expression was confirmed by western blotting.

PT

Immunoblotting

CE

To investigate sex-specific differences in tissue expression of AIG1, we isolated the indicated organs from one male and one female C57BL/6 mouse at the age of 12-16 weeks. Cells and

AC

tissues were lysed in ice-cold RIPA buffer (Sigma) supplemented with protease inhibitor cocktail mini complete (Roche), loaded onto 8–16% SDS gels (Thermo scientific) and transferred to nitrocellulose membranes with the iBlot system (Invitrogen). Membranes were blocked in 3% non-fat dry milk dissolved in phosphate buffered saline with 0.05% Tween (PBS-T) for at least 1 h and incubated overnight at 4°C with primary antibodies against actin (Millipore, #MAB1501, 1:4000), AIG1 (Acris, #AP20067PU-N, 1:500) or Ha-Tag (Abcam, #ab9110, 1:4000) diluted in PBS-T containing 3% non-fat dry milk. Afterwards, membranes were incubated with anti-mouse (rabbit) IgG (Fc) infrared fluorescence-conjugated (Licor, 1:30000) secondary antibody diluted in PBS-T with 3% non-fat dry milk. The membranes were scanned for infrared fluorescence at 680 or 800 nm using the Odyssey scanner system (Licor).

ACCEPTED MANUSCRIPT Cell-viability assay Cell viability was measured with the cell titer blue reagent exactly as has been described

PT

previously (Henke et al., 2012).

SC RI

Calcium-live cell imaging

Single-cell Ca2+ imaging was performed on a BD Pathway 855 as has been described previously (Noack et al., 2012).

NU

Confocal microscopy

Cells were transfected, treated as indicated, fixed for 20 min in 4% paraformaldehyde and

MA

visualized using a Zeiss LSM510 microscope.

ED

Protease protection assay

N2a cells were transiently transfected with the indicated constructs and 48 h later treated

PT

with 20 µM digitonin for 90 sec. Then, consecutive images were taken at the indicated time points after addition of 4 mM trypsin, which extinguishes fluorescence from EGFP and/or

AC

Results

CE

mCherry moieties present in the cytosol.

Expression of AIG1 in male and female tissues The expression of AIG1 was previously investigated on mRNA levels using Northern blots, identifying a major 1.5 kb and a minor 1.2 kb band at a relatively high level in the heart, ovary, testis, liver, kidney, small intestine and colon and at a low level in the spleen, prostate, brain, skeletal muscle and pancreas. It was undetectable in peripheral blood leukocytes, thymus, lung, and placenta (Seo et al., 2001). In this investigation, the blot was unfortunately not labeled and it is therefore not possible to clarify which tissue expresses which isoform. Conventional RT-PCR of three male human samples (dermal papilla cell, dermal sheath cell, skin fibroblast) and one female sample (dermal sheath cell) suggested that the level of expression was higher in males than females (Seo et al., 2001). We aimed to reproduce

ACCEPTED MANUSCRIPT these results at the protein level by immunoblotting using an antibody directed against an Nterminal sequence of AIG1 that clearly identified over-expressed AIG1 (see below). We analyzed tissues from male and female mice and used Actin as a loading control. In line with

PT

the results of Seo et. al., we observed strong expression of AIG1 in the liver, heart and

SC RI

kidney and lower expression in the spleen and brain. In contrast to the undetectable mRNA in the lung, we observed a strong protein signal of AIG1 in lung lysates of both male and female and contrarily to the mRNA signal we observed no AIG1 protein signal in the testis

NU

(Fig. 1). The suspected increased expression in male tissues was very weak if at all present at the protein level (Fig. 1).

MA

Genomic organization and tissue specific expression of splice variants of AIG1 We noted that the band recognized by the AIG1 antiserum migrated slightly faster in the liver

ED

of male and female mice, suggesting the presence of a different splice isoform of AIG1 or a tissue-specific post-translational

modification. We therefore analyzed

the genomic

PT

organization of AIG1 using the ensembl database (http://www.ensembl.org/). The murine

CE

Aig1 gene consists of six exons spread over a distance of 222.13 kb; three different splice isoforms are supported by expressed sequence tags and differ in the inclusion of exon 5 and

AC

the usage of two alternative sixth exons (Fig. 2A). We designed three primers capable of distinguishing these isoforms (Fig. 2B) and investigated the tissue expression of these variants using non-quantitative RT-PCR. This confirmed the suspicion raised by the immunoblots and confirmed a liver-specific expression of variant 2 (ENSMUST00000105534) with a predicted open reading frame of 543 bp and a molecular weight of 20.7 kDa (Fig. 2C). The isoform variant 2 differs from variants 1 and 3 in the number of transmembrane domains. The C-terminal transmembrane domain of variant 1 and 3 is completely missing in variant 2, and the last C-terminal transmembrane domain of variant 2 is truncated compared to variants 1 and 3. The hydrophilic C-terminal part of variant 2 consists only of 8 amino acids compared to 42 or 18 in variants 1 and 3 (Fig 2A). AIG1 is localized in the endoplasmic reticulum

ACCEPTED MANUSCRIPT AIG1 is a very hydrophobic protein with an ER localization signal at its cytosolic C-terminus, which has been previously identified as an interaction partner of the Pirh2 E3 ligase that negatively regulates p53 (Wu et al., 2011). To verify the presumed localization, we cloned

PT

AIG1 in frame with a C-terminal monomeric red fluorescent protein and co-expressed it in

SC RI

N2A cells with marker proteins localizing to the ER or mitochondria. We found a colocalization with the ER marker but not with mitochondria (Fig. 3). These results are in line with the data of Wu et al., which showed a perinuclear localization of AIG1 in HEK293 and

NU

Hela cells (Wu et al., 2011) .

AIG1 is an integral membrane protein with five potential transmembrane domains

MA

The exact topology of AIG1 in the ER membrane is still unknown. Bioinformatical analysis using TMpred (www.ch.embnet.org/software/TMPRED_form.html) suggests five or six

ED

transmembrane domains, with the presumed fourth transmembrane domain being clearly less hydrophobic than the other six (Fig. 4A), resulting in four different models of how AIG1

PT

might be inserted into the ER membrane (Fig. 4B). We therefore tried to clarify the topology

CE

of AIG1 in the ER membrane using a protease-protection assay. In this assay, the plasma membrane but not the ER membrane of cells over-expressing proteins tagged with

AC

fluorescent proteins is permeabilized with digitonin. Fluorescent proteins in the cytosol but not in the ER lumen can be digested by trypsin that has been added to the cells and as a result, luminal fluorescence remains whereas cytosolic fluorescence is extinguished (Lorenz et al., 2006). We over-expressed EGFP-AIG1 or AIG1-mRFP in Neuro2a cells; cytosolic EGFP served as control. Fluorescence of AIG1-mRFP and EGFP were extinguished whereas EGFP-AIG1 remained fluorescent, suggesting that the N-terminus of AIG1 resides in the ER lumen and the C-terminus in the cytosol which corresponds to model B (Fig. 4B and C). Effects of AIG1 on cell death and the ER Ca2+ concentration We next evaluated the effect of AIG1 on cell survival in different cell lines, the hippocampal cell line HT22 and mouse embryonic fibroblasts (MEFs). We generated cells stably over-

ACCEPTED MANUSCRIPT expressing AIG1 tagged with the small HA tag at the N-terminus (Fig. 5A) and subjected these cells to ER stress either elicited with the SERCA inhibitor thapsigargin (Fig.5B) or with tunicamycin, an inhibitor of glycosylation (Fig 5C) and to oxidative stress elicited with

PT

glutamate (Fig 5D), which in these cells inhibits cystine import through the glutamate/cystine

SC RI

antiporter xCT. Reduced cystine, cysteine, is a building block of glutathione synthesis and a lack of this important antioxidant kills cells by endogenous oxidative stress (reviewed in (Albrecht et al., 2010)). Cell viability was quantitated by the cell titer blue assay measured as

NU

fluorescence intensities. In both cell lines, AIG1 had no effect against ER stress (Fig. 5B/C), but increased susceptibility against oxidative stress (Fig. 5D). Fluorescence values were not

MA

normalized to prevent masking of effects induced by cell density which are especially important for glutamate induced toxicity; here a higher cell density is protective per se. In regard of that, it is important to recognize that the fluorescence values of AIG1 expressing

ED

MEF cells, although starting at a higher level due to a higher cell density compared to empty

PT

vector control, were more severely reduced by glutamate treatment. In addition, AIG1 increased the ER Ca2+ content, which we analyzed by treating the HT22 and MEF cells with

CE

the SERCA inhibitor thapsigargin leading to passive efflux of Ca2+ from the ER (Fig. 6A) respectively with the Ca2+ ionophore ionomycin (Fig. 6B), both in the presence of EGTA to

AC

remove any influence of extracellular Ca2+.

Discussion

In this study, we have attempted to shed light on the membrane orientation, tissue- and sexspecific protein expression, and alternative splicing of AIG1 in the mouse. We also studied its effects on cell death susceptibility and the intracellular Ca2+ homeostasis in different cell lines. AIG1 is a hydrophobic ER resident protein most highly expressed in the liver, lung and heart. It has a potential loop domain between transmembrane domain 3 and 4 and, according to this model, which is based on the results of our protease-protection assay, the N-terminus should reside within the lumen of the ER. However, it might be possible that the EGFP moiety at the N-terminus alters the membrane insertion of AIG1.

ACCEPTED MANUSCRIPT The expression analysis for the different splice variants revealed tissue specific expression of variant 2 in the liver. The corresponding protein differs from variants 1 and 3 in the amount of transmembrane domains and in the length of the hydrophilic C-terminus. Since our topology

PT

analysis revealed the N-terminus is located luminally and the C-terminus in the cytoplasm,

SC RI

this change in the C-terminal amino acid sequence might impact on protein-protein interactions with cytoplasmic proteins. Since Pirh2, a known AIG1-interaction partner (Wu et al., 2011) and regulator for p53, is also located in the cytoplasm (Duan et al., 2007), one

NU

could speculate that in the liver splice variant 2 might show different properties concerning Pirh2 interaction and cell death regulation compared to AIG1 variants 1 and 3 expressed in

MA

other tissues.

Our over-expression results were consistent in neuronal HT22 cells and wildtype mouse embryonic fibroblasts and showed an increase in the cellular Ca2+ content and a rather small

ED

detrimental effect on oxidative stress elicited with glutamate. Such a cell death-promoting

PT

function of AIG1 is in line with previously reported data in which AIG1 knockout correlated with cellular resistance against the organophosphate pesticide chlorpyrifos (Zhu et al., 2015)

CE

and with a study in which AIG1 expression positively correlated with survival of patients suffering from hepatocellular carcinoma, one of the most frequent malignancies worldwide

AC

(Wu et al., 2011). The fact that AIG1 increased NFAT-dependent gene transcription in promoter assays in HEK293 cells (Wu et al., 2011) suggested an involvement of the Ca2+ homeostasis in the mechanism of action of AIG1. NFAT is a substrate of the Ca2+-sensitive phosphatase calcineurin; it is therefore probable that the increased cellular Ca2+ content that we observed in AIG1-over-expressing cells causes this increased nuclear NFAT via calcineurin activation, resulting in increased gene transcription from NFAT-response elements, which is not only important in lymphocyte activation (Gwack et al., 2007) but also in cancerogenesis (reviewed in (Buchholz and Ellenrieder, 2007)).

References Albrecht, P., Lewerenz, J., Dittmer, S., Noack, R., Maher, P., Methner, A., 2010. Mechanisms of oxidative glutamate toxicity: the glutamate/cystine antiporter system xc- as a

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI

PT

neuroprotective drug target. CNS Neurol Disord Drug Targets 9, 373–382. Buchholz, M., Ellenrieder, V., 2007. An emerging role for Ca2+/calcineurin/NFAT signaling in cancerogenesis. Cell Cycle 6, 16–19. Bultynck, G., Kiviluoto, S., Henke, N., Ivanova, H., Schneider, L., Rybalchenko, V., Luyten, T., Nuyts, K., De Borggraeve, W., Bezprozvanny, I., Parys, J.B., De Smedt, H., Missiaen, L., Methner, A., 2012. The C terminus of Bax inhibitor-1 forms a Ca2+-permeable channel pore. Journal of Biological Chemistry 287, 2544–2557. doi:10.1074/jbc.M111.275354 Bultynck, G., Kiviluoto, S., Methner, A., 2014. Bax Inhibitor-1 Is Likely a pH-Sensitive Calcium Leak Channel, Not a H+/Ca2+ Exchanger. Science signaling 7, pe22–pe22. doi:10.1126/scisignal.2005764 Duan, S., Yao, Z., Hou, D., Wu, Z., Zhu, W.-G., Wu, M., 2007. Phosphorylation of Pirh2 by calmodulin-dependent kinase II impairs its ability to ubiquitinate p53. EMBO J. 26, 3062– 3074. doi:10.1038/sj.emboj.7601749 Gwack, Y., Feske, S., Srikanth, S., Hogan, P.G., Rao, A., 2007. Signalling to transcription: store-operated Ca2+ entry and NFAT activation in lymphocytes. Cell Calcium 42, 145– 156. doi:10.1016/j.ceca.2007.03.007 Henke, N., Albrecht, P., Pfeiffer, A., Toutzaris, D., Zanger, K., Methner, A., 2012. Stromal interaction molecule 1 (STIM1) is involved in the regulation of mitochondrial shape and bioenergetics and plays a role in oxidative stress. Journal of Biological Chemistry 287, 42042–42052. doi:10.1074/jbc.M112.417212 Henke, N., Lisak, D.A., Schneider, L., Habicht, J., Pergande, M., Methner, A., 2011. The ancient cell death suppressor BAX inhibitor-1. Cell Calcium 50, 251–260. doi:10.1016/j.ceca.2011.05.005 Ideta, R., Seki, T., Adachi, K., Nakayama, Y., 1998. The isolation and characterization of androgen-dependent genes in the flank organs of golden Syrian hamsters. Dermatology (Basel) 196, 47–50. Iżykowska, K., Zawada, M., Nowicka, K., Grabarczyk, P., Braun, F.C.M., Delin, M., Möbs, M., Beyer, M., Sterry, W., Schmidt, C.A., Przybylski, G.K., 2013. Identification of multiple complex rearrangements associated with deletions in the 6q23-27 region in Sézary syndrome. J Investig Dermatol 133, 2617–2625. doi:10.1038/jid.2013.188 Kiviluoto, S., Luyten, T., Schneider, L., Lisak, D., Rojas-Rivera, D., Welkenhuyzen, K., Missaen, L., De Smedt, H., Parys, J.B., Hetz, C., Methner, A., Bultynck, G., 2013. Bax Inhibitor-1-mediated Ca(2+) leak is decreased by cytosolic acidosis. Cell Calcium 54, 186–192. doi:10.1016/j.ceca.2013.06.002 Kiviluoto, S., Schneider, L., Luyten, T., Vervliet, T., Missiaen, L., De Smedt, H., Parys, J.B., Methner, A., Bultynck, G., 2012. Bax inhibitor-1 is a novel IP₃ receptor-interacting and sensitizing protein. Cell Death Dis 3, e367. doi:10.1038/cddis.2012.103 Lorenz, H., Hailey, D.W., Lippincott-Schwartz, J., 2006. Fluorescence protease protection of GFP chimeras to reveal protein topology and subcellular localization. Nat Methods 3, 205–210. doi:10.1038/nmeth857 Maher, P., Davis, J.B., 1996. The role of monoamine metabolism in oxidative glutamate toxicity. Journal of Neuroscience 16, 6394–6401. Mitani, Y., Rao, P.H., Futreal, P.A., Roberts, D.B., Stephens, P.J., Zhao, Y.-J., Zhang, L., Mitani, M., Weber, R.S., Lippman, S.M., Caulin, C., El-Naggar, A.K., 2011. Novel chromosomal rearrangements and break points at the t(6;9) in salivary adenoid cystic carcinoma: association with MYB-NFIB chimeric fusion, MYB expression, and clinical outcome. Clin. Cancer Res. 17, 7003–7014. doi:10.1158/1078-0432.CCR-11-1870 Noack, R., Frede, S., Albrecht, P., Henke, N., Pfeiffer, A., Knoll, K., Dehmel, T., Meyer Zu Hörste, G., Stettner, M., Kieseier, B.C., Summer, H., Golz, S., Kochanski, A., WiedauPazos, M., Arnold, S., Lewerenz, J., Methner, A., 2012. Charcot-Marie-Tooth disease CMT4A: GDAP1 increases cellular glutathione and the mitochondrial membrane potential. Hum Mol Genet 21, 150–162. doi:10.1093/hmg/ddr450 Olmsted, J.B., Carlson, K., Klebe, R., Ruddle, F., Rosenbaum, J., 1970. Isolation of microtubule protein from cultured mouse neuroblastoma cells. Proc Natl Acad Sci USA 65, 129–136. Reinhardt, H.C., Schumacher, B., 2012. The p53 network: cellular and systemic DNA

ACCEPTED MANUSCRIPT

MA

NU

SC RI

PT

damage responses in aging and cancer. Trends Genet. 28, 128–136. doi:10.1016/j.tig.2011.12.002 Seo, J., Kim, J., Kim, M., 2001. Cloning of androgen-inducible gene 1 (AIG1) from human dermal papilla cells. Mol Cells 11, 35–40. Shou, J., Jing, J., Xie, J., You, L., Jing, Z., Yao, J., Han, W., Pan, H., 2015. Nuclear factor of activated T cells in cancer development and treatment. Cancer Letters 361, 174–184. doi:10.1016/j.canlet.2015.03.005 Wang, B., Pelletier, J., Massaad, M.J., Herscovics, A., Shore, G.C., 2004. The yeast splitubiquitin membrane protein two-hybrid screen identifies BAP31 as a regulator of the turnover of endoplasmic reticulum-associated protein tyrosine phosphatase-like B. Mol. Cell. Biol. 24, 2767–2778. Wu, G., Sun, M., Zhang, W., Huo, K., 2011. AIG1 is a novel Pirh2-interacting protein that activates the NFAT signaling pathway. Front Biosci (Elite Ed) 3, 834–842. Yusa, K., Zhou, L., Li, M.A., Bradley, A., Craig, N.L., 2011. A hyperactive piggyBac transposase for mammalian applications. Proc Natl Acad Sci USA 108, 1531–1536. doi:10.1073/pnas.1008322108 Zhu, J., Dubois, A., Ge, Y., Olson, J.A., Ren, X., 2015. Application of human haploid cell genetic screening model in identifying the genes required for resistance to environmental toxicants: Chlorpyrifos as a case study. J Pharmacol Toxicol Methods 76, 76–82. doi:10.1016/j.vascn.2015.08.154

ED

Acknowledgments

This work was funded by the DFG ME1922/9-1 to A.M. and the Forschungskommission der

PT

Heinrich-Heine Universität Düsseldorf. We thank Susanne Thomsen for technical assistance

AC

CE

and Darragh O’Neill for proofreading the manuscript.

ACCEPTED MANUSCRIPT Figure Legends Figure 1 Sex-specific expression of AIG1. Tissue expression of AIG1 was analyzed by

PT

immunoblotting. Protein lysates from different tissues of one female and one male C57BL/6 mouse were analyzed on an SDS-gel, transferred to a nitrocellulose membrane and probed

SC RI

with anti-AIG1 antibody. Actin served as loading control. Size is indicated.

Figure 2 Genomic organization and tissue specific expression of splice variants of

NU

AIG1. A) Genomic organization of three splice isoforms of AIG1 (upper panel) and protein sequences of the corresponding proteins (lower panel). Transmembrane domains are

MA

underlined, the sequence encoded by exon V is in bold, and the different C-termini shaded. B) Localization of PCR primers capable of distinguishing the three isoforms. C) Non-

ED

quantitative RT-PCR showing the expression of the indicated splice isoforms in different

PT

mouse tissues. The arrows indicate the expected bands.

Figure 3 AIG1 localizes in the endoplasmic reticulum. Subcellular localization of AIG1

CE

was analyzed by confocal microscopy. AIG1-mRFP was expressed in N2A cells together

AC

with ER-targeted EGFP (upper panels) or mitochondrially-targeted EGFP (lower panels). A clear co-localized signal was obtained from ER-GFP together with AIG1-mRFP. No colocalization was detectable for AIG1-mRFP with mitochondrial (mt) EGFP.

Figure 4 AIG1 is an integral membrane protein with a C-terminus facing the cytosol. A) Hydrophobicity plot of AIG1 generated by TMpred. The localization of potential transmembrane domains is indicated by roman numerals. B) Models depicting the orientation of AIG1 in the membrane according to the indicated bioinformatics analyses. C) Proteaseprotection assay shows localization of the AIG1 C-terminus in the cytosol and the N-terminus in the ER lumen. N2a cells were transiently transfected with the indicated constructs and 48 h later treated with 20 µM digitonin for 90 sec. Consecutive images were then taken at the indicated time points after addition of 4 mM trypsin, which extinguishes fluorescence from

ACCEPTED MANUSCRIPT fluorescent proteins present in the cytosol.

Figure 5 Effects of AIG1 on cell death susceptibility. A) AIG1 was stably expressed in

PT

HT22 and MEF cells and expression was verified by immunoblotting against the indicated

SC RI

antigens, actin served as loading control. B-D) Survival of AIG1 expressing HT22 (left) or MEF (right) cells was analyzed under different stress conditions. 5000 cells, seeded in 96well plates, were subjected to ER stress induced by different concentrations of thapsigargin

NU

(B) or tunicamycin (C) or endogenous oxidative stress induced by the indicated concentrations of glutamate (D). 16 h after stress onset, cell survival was quantitated with the

performed in triplicate at least.

MA

cell-titer blue assay. Graphs represent mean of three independent experiments each

ED

Figure 6 AIG1 increases the ER calcium concentration. The ER Ca2+ content of AIG1-

PT

expressing HT22 or MEF cells was released with thapsigargin in EGTA (A) or with ionomycin in EGTA (B) and measured by Ca2+ live cell imaging with Fura2-AM. Graphs represent the

AC

CE

mean of three independent experiments.

AC

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Abbreviation List Androgen-induced gene 1 (AIG1)

fetal bovine serum (FBS)

SC RI

Golgi SNAP Receptor Complex Member 1 (GOSR1) hemagglutinin (HA) internal ribosomal entry site (IRES)

Neuro2a (N2a)

MA

nuclear factor of activated T cells (NFAT)

NU

Mouse embryonic fibroblasts (MEFs)

Nuclear Factor 1/B (NFIB)

phosphate buffered saline with 0.05% Tween (PBS-T)

ED

reverse-transcription polymerase chain reaction (RT-PCR)

CE

PT

Transmembrane Bax inhibitor motif containing 6 (TMBIM6)

AC

PT

Dulbecco’s Modified Eagle’s Medium (DMEM)