Báo cáo khoa học: Enhanced sensitivity to hydrogen peroxide-induced apoptosis in Evi1 transformed Rat1 fibroblasts due to repression of carbonic anhydrase III docx

Enhanced sensitivity to hydrogen peroxide-induced apoptosis in Evi1 transformed Rat1 fibroblasts due to repression of carbonic anhydrase III P. Roy 1 , E. Reavey 1 , M. Rayne 1 , S. Roy 1 , M. Abed El Baky 1 , Y. Ishii 2 and C. Bartholomew 1 1 Department of Biological & Biomedical Sciences, Glasgow Caledonian University, City Campus, Glasgow, UK 2 Graduate School of Pharmaceutical Sciences, Kyushu University, Fukuoka, Japan Introduction Multiple mechanisms have been proposed for Evi1’s contribution to cancer progression, including enhanced cell proliferation, impaired differentiation and evasion of apoptosis [1]. Evasion of apoptosis has been observed in both haemopoietic and epithelial cells with a variety of agents and suggests that Evi1 is a survival factor. For example, either deregulated or enforced expression of Evi1 has been shown to protect HEK293, HEC-1B and Jurkat cells from UV-induced apoptosis [2], U937 cells from tumour necrosis factor- a-induced apoptosis [2], SiHa cells from interferon- a-induced apoptosis [3] and both rat intestinal epithe- lial cells and HT-29 cells from transforming growth factor-b (TGFb)- and paclitaxel-induced apoptosis [4]. Keywords apoptosis; carbonic anhydrase III; Evi1; H 2 O 2 Correspondence C. Bartholomew, Department of Biological & Biomedical Sciences, Glasgow Caledonian University, City Campus, Cowcaddens Road, Glasgow G4 OBA, UK Fax: +44 141 331 3208 Tel: +44 141 331 3213 E-mail: c.bartholomew@gcal.ac.uk (Received 18 August 2009, revised 9 November 2009, accepted 16 November 2009) doi:10.1111/j.1742-4658.2009.07496.x EVI1 is a nuclear zinc finger protein essential to normal development, which participates in acute myeloid leukaemia progression and transforms Rat1 fibroblasts. In this study we show that enforced expression of Evi1 in Rat1 fibroblasts protects from paclitaxel-induced apoptosis, consistent with previously published studies. Surprisingly, however, these cells show increased sensitivity to hydrogen peroxide (H 2 O 2 )-induced apoptosis, dem- onstrated by elevated caspase 3 catalytic activity. This effect is caused by a reduction in carbonic anhydrase III (caIII) production. caIII transcripts are repressed by 92–97% by Evi1 expression, accompanied by a similar reduc- tion in caIII protein. Reporter assays with the rat caIII gene promoter show repressed activity, demonstrating that Evi1 either directly or indi- rectly modulates transcription of this gene in Rat1 cells. Targeted knock- down of caIII alone, with Dicer-substrate short inhibitory RNAs, also increases the sensitivity of Rat1 fibroblasts to H 2 O 2 , which occurs in the absence of any other changes mediated by Evi1 expression. Enforced expression of caIII in Evi1-expressing Rat1 cells reverts the phenotype, restoring H 2 O 2 resistance. Together these data show that Evi1 represses transcription of caIII gene expression, leading to increased sensitivity to H 2 O 2 -induced apoptosis in Rat1 cells and might suggest the basis for the development of a novel therapeutic strategy for the treatment of leukae- mias and solid tumours where EVI1 is overexpressed. Abbreviations AKT, protein kinase B; caIII, carbonic anhydrase III; DCF-DA, 2¢-7¢-dichloroflourescene diacetate; DMEM, Dulbecco’s modified Eagle’s medium; DsiRNA, Dicer-substrate short inhibitory RNA; FACS, fluorescent activated cell sorter; H 2 O 2 , hydrogen peroxide; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide; PI3K, phosphoinositide-3-kinase; ROS, reactive oxygen species; RTQ, real-time quantitative; siRNA, short inhibitory RNA; TGFb, transforming growth factor-b. FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS 441 Variant forms of Evi1 are also antiapoptotic, protect- ing murine acute myeloid leukaemia cells that have been treated with arsenic trioxide [5]: in this case the agent actually targets degradation of AML1 ⁄ EVI1 in order to induce programmed cell death. Recent evi- dence also supports a survival role in nonpathological conditions, as mice require Evi1 to maintain adequate numbers of haemopoietic stem cells [6] and this is also consistent with a more general requirement for Evi1 for cell survival during murine development [7]. Evi1 is a 145 kDa nuclear protein member of the cys2his2 zinc finger family [8]. It possesses multiple domains that have been identified by both sequence homology and functional activity, including: two dis- tinct zinc finger motifs of seven and three repeating units at the N-terminus and towards the C-terminus, respectively [8]; a central repressor domain [9] and a C-terminal acidic domain [8]. These domains have been shown to interact with other molecules, including DNA [10,11] and proteins [1] and are responsible for mediat- ing Evi1 inhibition of apoptosis. Interactions of various molecules with these motifs enable Evi1 to impair or activate particular signalling pathways, including TGFb [12,13], c-Jun N-terminal kinase (JNK) [2] and phos- phoinositide-3-kinase ⁄ protein kinase B (PI3K ⁄ AKT) [4]. Intervention of critical signalling molecules are the basis for Evi1-mediated enhanced cell survival. A number of agents have been used to study the impact of Evi1 on apoptosis in cells, including UV light, tumour necrosis factor-a, TGFb, interferon-a, arsenic trioxide and taxol (paclitaxel). Hydrogen per- oxide (H 2 O 2 ) also induces apoptosis, but the impact of Evi1 expression on its apoptotic-inducing capability has not been investigated previously. Either exoge- nously supplied or endogenous H 2 O 2 generate reactive oxygen species (ROS), which if unchecked cause oxida- tive stress, resulting in damaged cellular DNA, lipids and proteins that interfere with cell function. To com- bat oxidative stress, complex antioxidant defence mechanisms have evolved to protect cells from oxidative injury. Established antioxidants include the enzymatic systems catalase, superoxide dismutase, glu- tathione peroxidases and peroxiredoxin III and nonen- zymatic systems including vitamins C, E and B 2 , coenzyme Q 10 , glutathione and carotene [14]. If the amount of ROS exceeds the capacity of the antioxi- dant machinery, then oxidative stress occurs [15]. The enzyme carbonic anhydrase III (caIII) (EC 4.2.1.1) is also thought to protect cellular proteins from oxidation [16]. Carbonic anhydrases are a family of 15 distinct isozymes that catalyse the reversible con- version of H 2 O+CO 2 and H + + HCO 3 ) [17]. caIII is unique, very abundant in liver, skeletal muscle and adipocytes and unlike other members of this family has low hydratase catalytic activity [18]. The function of caIII is unknown, but it is suggested that it has an antioxidant function and it has been shown to protect cells from H 2 O 2 -induced apoptosis [19,20]. In this study, we investigated H 2 O 2 -induced apoptosis in Rat1 cells expressing an Evi1 transgene. Surprisingly, we found that Evi1 expression increases sensitivity to H 2 O 2 -mediated apoptosis in complete contrast to the protective effect of other apoptosis- inducing agents. Increased sensitivity is primarily due to the transcriptional downregulation of caIII gene expression mediated by Evi1. Results Evi1-expressing Rat1 fibroblasts are resistant to taxol-induced apoptosis Independent, stable populations of Rat1 cells express- ing murine Evi1 were generated by infection with the p50M5.6neo retrovirus (Fig. 1A), produced by tran- sient transfection of EcoPak2Ô cells, and designated 5.61 and 5.62. Empty vector Neo1 and Neo2 cells were similarly created with the p50MX-neo retrovirus. Evi1 expression was confirmed by western blot analysis with a-Evi1 (1806), detecting a 145 kDa protein in 5.61 and oenASDS Evi1 RTLRTL p50M5.6neo Rat1 Neo1 Neo2 5.61 cells 5.62 cells 35 kDa 145 kDa α-Evi1 α-gapdh A B Fig. 1. Schematic representation of the murine Evi1-expressing recombinant retroviral vector p50M5.6neo and production of Evi1 in virus-infected Rat1 fibroblasts. (A) Viral long terminal repeats (LTR), the murine Evi1 gene, including the N-terminal and C-terminal zinc finger domains (black boxes), repressor domain (grey box) and acidic domain (striped box), the Neo gene (neo) and splice donor (SD) and splice acceptor (SA) sites for the production of subgenom- ic transcripts for the expression of neo. (B) Western blot analysis of whole cell protein extracts derived from the indicated cell lines and populations using a-Evi1 (1806) and a-gapdh (6C5) antibodies. The positions of 145 and 35 kDa Evi1 and gapdh proteins are shown. Evi1 enhanced oxidant-induced apoptosis P. Roy et al. 442 FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS 5.62 cells that is absent from Neo1, Neo2 and parental Rat1 cells (Fig. 1B). Previous studies have shown that Evi1 is a survival factor, protecting cells from apoptosis induced by a variety of agents. To determine if Evi1 also protects Rat1 cells from apoptosis, we treated our panel of cell populations with paclitaxel (taxol). Apoptosis was monitored by measuring caspase 3 catalytic activity. The results showed that taxol (1 lm, 16 h) induced sig- nificantly higher caspase 3 catalytic activity in Rat1, Neo1 and Neo2 cells than in Evi1-expressing 5.61 and 5.62 cells (Fig. 2). Taxol induced caspase 3 activity in all cell populations, but to a much lesser extent in 5.61 and 5.62 cells. These data show that Evi1 protects Rat1 cells from taxol-induced apoptosis, consistent with previous studies in other cell types. Rat1 fibroblasts expressing Evi1 have increased sensitivity to H 2 O 2 -induced apoptosis Although previous studies have shown that Evi1 pro- tects cells from a variety of inducers of apoptosis, the effects of H 2 O 2 have not yet been examined. Rat1 cells and 5.61 cells were exposed to various concen- trations of H 2 O 2 and cell viability monitored by 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay. The results showed that cell viability was reduced in a H 2 O 2 dose-dependent manner in both Rat1 and 5.61 cells (Fig. 3A). Surprisingly, the viabil- ity of 5.61 cells was significantly less than Rat1 cells at each concentration of H 2 O 2 (Fig. 3A). The viability of the entire panel of cell populations was examined by MTT assay following H 2 O 2 (750 lm) treatment and this confirmed that ectopic Evi1 expression decreased survival in Rat1 cells (Fig. 3B). This was supported by the dramatic morphological change observed in H 2 O 2 - treated 5.61 cells compared with parental Rat1 or empty vector control Neo cells treated with the same concentration of reagent (Fig. 3C). The morphological changes observed in 5.61 cells treated with H 2 O 2 resembled apoptosis. Therefore, we examined caspase 3 activation in cultures of the cell populations. The results showed that H 2 O 2 induced caspase 3 catalytic activity in all cell populations exam- ined, but the level of activation was significantly greater in 5.6 cell populations compared with parental Rat1 and vector control cells (Fig. 3D). Evi1 represses expression of the potential antioxidant caIII in Rat1 cells Previous studies have shown that arsenic trioxide induces apoptosis in leukaemia cells by degrading the Evi1 fusion protein, AML1 ⁄ EVI1 [5]. H 2 O 2 -induced Evi1 degradation could reduce cell survival in the pres- ence of this agent. Therefore, the stability of Evi1 transgene expression in 5.61 cells was examined by western blot analysis (a-Evi1, 1806) following H 2 O 2 treatment for 4, 10 and 16 h. However, the abundance of Evi1 protein remained unchanged during this time period, confirming that H 2 O 2 had no effect on protein stability (Fig. 4), eliminating this mechanism. Recently, we used microarray technology to identify Evi1-mediated induction and repression of gene tran- scripts in Rat1 cells (E. R. Reavey & C. Bartholomew, unpublished results). Inspection of these data revealed transcriptional repression of caIII, which encodes a protein that has previously been shown to protect cells from H 2 O 2 -induced apoptosis [19,20]. The microarray data were confirmed by real-time quantitative RT-PCR (RTQ-RT-PCR) using total cel- lular RNA from Rat1 and derivative Neo and 5.6 cells. These data showed that caIII mRNA transcripts were repressed by 92–97% in 5.61 and 5.62 cells rela- tive to Rat1 and empty vector control cells Neo1 and Neo 2 (Fig. 5A). Western blot analysis with a-caIII (E-19) confirmed that caIII protein levels were also dramatically reduced in 5.61 and 5.62 cells (Fig. 5B), consistent with the RTQ-RT-PCR data. caIII gene promoter activity is repressed by Evi1 in Rat1 cells To determine if Evi1-mediated caIII repression occurs at the level of gene transcription, reporter assays were 160 000 180 000 200 000 *** 80 000 100 000 120 000 140 000 Relative caspase 3 activity (luminescence,RLU) 0 20 000 40 000 60 000 Fig. 2. Histogram showing relative caspase 3 catalytic activity of the indicated cell lines and populations in the absence (grey col- umns) or presence of 1 l M paclitaxel for 16 h (black columns). The columns represent the mean of an experiment performed in qua- druplicate and error bars the standard deviation. ***P £ 0.0002. P. Roy et al. Evi1 enhanced oxidant-induced apoptosis FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS 443 performed. A pGL3-basic vector containing )1485 to +55 of rat caIII gene promoter sequence [21], desig- nated p-1485 ⁄ +55 caIII luc, was created (Y. Ishii, unpublished data) and transfected into Rat1, Neo and 5.6 cells, together with the control vector pRLCMV. The activity of p-1485 ⁄ +55 caIII luc in the various cell types, normalized for pRLCMV activity, is shown in Fig. 6A. The results showed that the caIII gene promoter had at least 10-fold greater transcriptional activity in Rat1, Neo1 and Neo2 cells compared with the Evi1-expressing 5.61 and 5.62 cells. In contrast, the activity of a minimal thymidine kinase gene promoter construct (pGL2tkluc), normalized for pRLCMV, was similar in all cell types examined (Fig. 6B). These results show that Evi1 specifically repressed the transcriptional activity of the caIII gene promoter in Rat1 cells. caIII knockdown alone enhances H 2 O 2 -induced apoptosis in Rat1 cells and enhanced caIII expression is protective Short inhibitory RNAs (siRNA) were used to deter- mine if repression of caIII gene expression alone sensi- Cell viability (%) 0 20 40 60 80 100 120 Rat1 Neo 1 Neo 2 5.61 cells 5.62 cells Rat1 Neo 1 Neo 2 5.61 cells 5.62 cells 0 20 40 60 80 100 120 UT 100 150 200 250 500 750 Cell viability (%) H 2 O 2 (µM) Rat1 Neo2 5.61 cells 750 M H 2 O 2 Untreated 0 20 000 40 000 60 000 80 000 100 000 120 000 140 000 Rat1 Neo1 Neo2 5.61 cells 5.62 cells Rat1 Neo1 Neo2 5.61 cells 5.62 cells Relative caspase 3 activity (luminescence, RLU) ** *** C AB D Fig. 3. Cell viability and caspase 3 catalytic activity of cell lines and populations following 16 h treatment with H 2 O 2 . (A) Histogram of the percentage viability (MTT assay) of untreated cells (UT) and cells treated with the indicated concentration of H 2 O 2 for 16 h. Grey columns are Rat1 cells and black columns are 5.61 cells. The columns represent the mean of an experiment performed in quadruplicate and error bars the standard deviation. (B) Histogram of the percentage viability (MTT assay) of the indicated untreated (black columns) and 16 h of 750 l M H 2 O 2 treated (grey columns) cell lines and populations. The columns represent the mean of an experiment performed in quadrupli- cate and error bars the standard deviation. **P £ 0.002. (C) Photographs showing the morphology of either untreated or treated (16 h 750 l M H 2 O 2 ) indicated cell lines and populations. (D) Histogram showing relative caspase 3 catalytic activity of the indicated cell lines and populations in the absence (grey columns) or presence of 750 l M H 2 O 2 for 16 h (black columns). The columns represent the mean of an experiment performed in quadruplicate and error bars the standard deviation. ***P < 0.0001. UT 4 h 10 h 16 h UT 4 h 10 h 16 h 5.61 cells Neo1 α-Evi1 α-gapdh 145 kDa 35 kDa Fig. 4. Western blot analysis of whole cell protein extracts derived from Neo1 and 5.61 cell populations following treatment for 0 (UT), 4, 10 and 16 h with 750 l M H 2 O 2 using a-Evi1 (1806) and a-gapdh (6C5) antibodies. The positions of 145 kDa Evi1 and 35 kDa gapdh proteins are shown. Evi1 enhanced oxidant-induced apoptosis P. Roy et al. 444 FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS tizes Rat1 cells to H 2 O 2 . Both RTQ-RT-PCR and wes- tern blot analysis (a-caIII) were used to identify an effective Dicer-substrate siRNA (DsiRNA) that inhib- ited both caIII mRNA gene transcripts (98% reduc- tion, Fig. 7A, 10 nm siRNA1) and caIII protein (Fig. 7B, 10 nm siRNA1) when transfected into Rat1 cells. A control, nonspecific DsiRNA had no effect on either caIII mRNA or caIII protein when transfected into Rat1 cells at the same concentration (Fig. 7A,B, Non sp siRNA). The effect of caIII siRNA1 on H 2 O 2 sensitivity in transfected Rat1 cells was then investigated by moni- toring caspase 3 catalytic activity. siRNA1 transfected Rat1 cells treated with 750 lm H 2 O 2 for 16 h had at least double the caspase 3 activity observed with 750 lm H 2 O 2 -treated untransfected and nonspecific DsiRNA transfected Rat1 control cells (Fig. 7C). caIII knockdown with a second distinct siRNA (siRNA3) produced the same phenotype (Fig. S1). The results show that H 2 O 2 induced caspase 3 catalytic activity in all cells, but the level of activation was significantly greater in Rat1 cells transfected with a caIII-specific siRNA (Fig. 7C). Furthermore, caIII knockdown only sensitized Rat1 cells to H 2 O 2 treatment and had no effect upon apoptosis induced by treatment with taxol (Fig. S2). caIII expression was restored in 5.61 cells to deter- mine if the sensitivity to H 2 O 2 treatment could be reverted. Rat1 cells were transiently transfected with a caIII expression vector (pRC-sport 6caIII), which sig- nificantly increased cellular levels of the caIII protein (Fig. 8A). The increased levels of caIII protein in 5.61 cells protected them from H 2 O 2 treatment, compared with untreated or empty vector control transfected cells, as determined by measuring caspase 3 catalytic activity (Fig. 8B). Finally, we measured intracellular levels of ROS to determine if the basal oxidized state varied between Rat1 and 5.61 cells. Cells labelled with 2¢-7¢-dichloro- flourescene diacetate (DCF-DA) were examined by fluorescent automatic cell sorter (FACS). The results 2.00 *** 1.20 1.60 0.40 0.80 0.00 Neo1 Neo2 5.61 cells 5.62 cells caIII gene expression relative to Rat1 cells Rat1 Neo1 Neo2 5.61 cells α-caIII α-gapdh35 kDa 27 kDa A B Fig. 5. caIII gene expression in Rat1 cells and derivative cell popu- lations. (A) Histogram of caIII mRNA levels normalized for gapdh mRNA relative to normalized caIII mRNA in Rat1 cells, determined by RTQ-RT-PCR. The columns are the mean of an experiment per- formed in quadruplicate and the error bars the standard deviation. (B) Western blot analysis of whole cell protein extracts using a-caIII (E-19) and a-gapdh (6C5) antibodies. The positions of 27 kDa caIII and 35 kDa gapdh proteins are shown. ***P < 0.0001. B A Fig. 6. Reporter assays showing the activity of the caIII and mini- mal herpes simplex virus thymidine kinase (HSV tk) gene promot- ers in Rat1 and derivative cells. (A) Histogram of caIII gene promoter firefly luciferase reporter activity (p-1485 ⁄ +55 caIII luc) normalized for cytomegalovirus (CMV) immediate early enhancer ⁄ promoter renilla luciferase reporter activity (pRLCMV). The columns are the mean of an experiment performed in quadruplicate and error bars the standard deviation. ***P < 0.0001. (B) Histogram of HSV tk gene promoter firefly luciferase reporter activity (pGL2tkluc) normalized for pRLCMV. The columns are the mean of an experi- ment performed in quadruplicate and error bars the standard deviation. P. Roy et al. Evi1 enhanced oxidant-induced apoptosis FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS 445 showed that the mean fluorescence of 5.61 cells was significantly higher than that observed in Rat1 and Neo1 cells (Fig. 9). This shows that ROS were elevated in 5.61 cells, consistent with the observed reduction in caIII expression. Discussion We show here for the first time that ectopic expression of Evi1 sensitizes Rat1 cells to H 2 O 2 -induced apopto- sis. This represents the first description that Evi1 can actually stimulate cell death. Previous studies with a variety of agents have shown that Evi1 protects cells from apoptosis and functions as a survival factor, pro- viding one of multiple suggested roles that contribute to the development and progression of leukaemia. Consistent with this view, we also show that Evi1 pro- tects Rat1 cells from apoptosis induced by at least one of these agents (taxol) and therefore probably also acts as a survival factor in these cells. Evi1-mediated pro- tection from taxol-induced apoptosis in rat intestinal epithelial cells and colon cancer cells (HT-29) is due to UT Non sp siRNA siRNA 1 α- caIII α- gapdh 35 kDa 27 kDa 0 0.2 0.4 0.6 0.8 1 1.2 1.4 caIII gene expression relative to Rat1 cells ** Non sp siRNA 1 siRNA Relative caspase 3 activity (luminescence, RLU) 0 100 000 200 000 300 000 400 000 500 000 600 000 700 000 800 000 900 000 UT Non sp siRNA siRNA 1 UT 750 µM *** A B C Fig. 7. DsiRNA-mediated knockdown of caIII mRNA and protein and enhanced caspase 3 catalytic activity in Rat1 cells. (A) Histo- gram of caIII mRNA levels normalized for gapdh mRNA relative to normalized caIII mRNA in untreated Rat1 cells, determined by QRT- PCR in Rat1 cells transfected for 48 h with 10 n M of either a non- specific siRNA (Non sp siRNA) or a caIII-specific siRNA (siRNA 10 n M). The columns are the mean of an experiment performed in quadruplicate and the error bars the standard deviation. **P £ 0.0041. (B) Western blot analysis of whole cell extracts derived from Rat1 cells, transfected as described in (A), with a -caIII (E-19) and a-gapdh (6C5) antibodies. The positions of 27 kDa caIII and 35 kDa gapdh proteins are shown. (C) Relative caspase 3 cata- lytic activity in Rat1 cells transfected as described in (A), either treated (black columns) or untreated (grey columns) with 750 l M H 2 O 2 for 16 h. The columns are the mean of an experiment per- formed in quadruplicate and error bars the standard deviation. ***P < 0.0001. UT pRC-sport6caIII α-caIII α-gapdh35 kDa 27 kDa 1400 000 1600 000 1800 000 800 000 1000 000 1200 000 *** 200 000 400 000 600 000 Relative caspase 3 activity (luminescence, RLU) *** 0 UT 750 µ M B A Fig. 8. 5.61 cell protection from H 2 O 2 -induced caspase 3 catalytic activity by ectopic expression of caIII. (A) Western blot analysis of whole cell extracts derived from untransfected 5.61 cells (UT), empty vector transfected 5.61 cells (pRC CMV) and caIII expres- sion vector transfected cells (pRC-sport6caIII) with a-caIII (E-19) and a-gapdh (6C5) antibodies. The positions of 27 kDa caIII and 35 kDa gapdh proteins are shown. (B) Relative caspase 3 catalytic activity in untransfected 5.61 cells (grey columns) and 5.61 cells transfected with pRC CMV (white columns) or pRC-sport6caIII (black columns) with (H 2 O 2 ) or without (UT) 750 lM H 2 O 2 treat- ment. ***P £ 0.0007. Evi1 enhanced oxidant-induced apoptosis P. Roy et al. 446 FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS stimulation of PI3K and its downstream effector AKT [4]. The same mechanism probably also operates in Rat1 cells, but was not examined in this study. Sensitization to H 2 O 2 -induced apoptosis, determined by caspase 3 catalytic activity, was seen in both Evi1- expressing Rat1 cells (5.6 cells) and in caIII knockdown Rat1 cells. Furthermore, Evi1-expressing Rat1 cells had a 90% reduction in caIII gene transcripts and protein. Together, these results confirm that Evi1-mediated stim- ulation of H 2 O 2 -induced cell death is due to the reduc- tion in cellular levels of the caIII protein. The results presented here suggest that caIII protects Rat1 cells from H 2 O 2 -derived ROS and therefore acts as an antioxidant. However, the biological activity of caIII is an enigma. Unlike other products of this gene family, caIII has very low catalytic activity and so it is unlikely that it functions in hydrating carbon dioxide [18]. Furthermore, knockout mice, deficient in caIII, have normal growth, development and lifespan under laboratory conditions, suggesting that the protein is not essential [22]. However, several recent observations suggest that caIII is an important antioxidant, consis- tent with the observations here. caIII is highly abun- dant in skeletal muscle, a tissue of high oxygen consumption and antioxidant activity. Microarray analysis of skeletal muscle of wild-type and caIII-defi- cient knockout mice revealed that caIII has a possible role in the glutathione-mediated antioxidative system [23]. This is supported by biochemical evidence show- ing that caIII undergoes rapid reversible S-glutathiola- tion or irreversible oxidation in mildly and exhaustively stressed muscle, respectively [23]: in the presence of glutathione, glutathione peroxidases restore reversibly S-glutathiolated caIII. This mechanism would explain both the protective effect of ectopic caIII expression observed in NIH3T3 cells [20] and the increased sensitivity of caIII knockdown Rat1 or Evi1- expressing Rat1 cells exposed to H 2 O 2 . Several possibilities exist to explain the caIII repres- sion effect observed in the present study. Abundant caIII gene transcripts and protein occur in Rat1 cells, which, like all fibroblasts examined (C. Bartholomew, unpublished results), normally express low levels of endogenous evi1. Therefore, caIII gene expression and protein production normally occur efficiently in the presence of evi1 in Rat1 cells. The simplest explanation is that merely elevated cellular levels of Evi1 are suffi- cient to repress caIII transcription. Consistent with this, previous studies have shown that the abundance of Evi1 is crucial to 32Dcl3 granulocyte differentiation [24], suggesting that differential changes in gene expression must occur that are dependent on the quan- tity of cellular levels of the Evi1 protein. However, other possibilities exist. Multiple naturally occurring evi1 isoforms occur and it might be that it is the relative increase in the abundance of the Evi1 full-length form [25] in Rat1 cells (5.6 cells), observed here, that signifi- cantly represses caIII gene expression. It is possible that only some isoforms of Evi1 repress caIII expression, whereas perhaps other forms either have no effect or the opposite effect to the full-length form. Some studies have shown that the MDS1 ⁄ EVI1 isoform has Rat1 MFI –250 ROS levels Neo1 MFI –144 ROS levels 5.61 cells MFI –905 ROS levels 10 0 10 1 10 2 10 3 10 4 10 0 10 1 10 2 FL1-H FL1-H 10 3 10 4 10 0 10 0 10 1 10 1 10 2 10 2 FL1-H 10 3 10 3 10 4 10 4 10 5 10 0 10 1 10 2 10 3 10 4 10 5 10 0 10 1 10 2 10 3 10 4 10 5 M1 M1 M1 Fig. 9. Histogram of fluorescence intensity (x-axis) versus cell num- ber (y-axis) of indicated DCF-DA-labelled cells analysed by FACS. The mean fluorescence intensity (MFI) for the region designated M1 is shown for each cell type. P. Roy et al. Evi1 enhanced oxidant-induced apoptosis FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS 447 the opposite effect of the full-length form. For example, it is reported that the MDS1 ⁄ EVI1 isoform enhances the growth inhibitory effects of TGFb [26], whereas the full-length form blocks this response [12]. caIII is a very abundant protein, particularly in liver, muscle and adipocytes. However, very little is known about its transcriptional regulation. Transcription of the rat caIII gene is inhibited by the aryl hydrocarbon recep- tor ligand 3-methylchlanthrene in hepatocytes and in the livers of rats fed an ethanol-supplemented diet [21,27]. Human CAIII mRNA is induced in muscle of athletes training under hypoxic conditions [28]. One study has been conducted with the caIII gene promoter, with a preliminary analysis of an active 2.8 kb human CAIII gene promoter in myogenic cells and a significant loss of activity upon deletion to )715 bp [29]. Reporter assays with the caIII gene promoter ()1485 ⁄ +55) showed that this region contains strong promoter activity in Rat1 cells, consistent with a previ- ous analysis of the human promoter [29]. This region also has the cis-regulatory elements necessary for Evi1- mediated transcriptional repression. Evi1-mediated repression could be caused by binding directly to promoter sequences. Previous studies with artificial promoter reporter constructs have shown that Evi1 can function as a DNA-binding transcriptional repressor protein [9]. Evi1 binds several corepressor molecules, CtBP, the histone methyltransferase SUV39H1 and the histone deacetylase HDAC1 [30–32], each of which mediates transcriptional repres- sion. However, to date no genes that are direct targets for transcriptional repression have been identified. Furthermore, very few genes have been identified that are directly regulated and induced by Evi1; GATA2 being the best characterized [6,33]. Inspection of the )1485 ⁄ +55 rat nucleotide sequence for Evi1 protein- binding sites with matinspector software revealed mul- tiple potential sites. This suggests that caIII may be a direct target for Evi1-mediated transcriptional repres- sion, although binding and biological activity of any of these motifs require experimental investigation. Alternatively, repression of caIII gene expression could be indirect. Evi1 has been shown to interact with a number of transcription factors, including PU.1, RUNX1, GATA1, E2F1 and SMAD3 [12,34–37] and signalling molecules such as JNK and PI3K ⁄ AKT [2,4] to inhibit their biological activities. Therefore, it remains possible that Evi1 might repress caIII gene expression by interacting with a transcription factor or by inhibit- ing a signalling pathway that is normally required for the high level of expression observed in Rat1 cells. Inspection of the caIII )1485 ⁄ +55 promoter region with matinspector software showed several potential binding sites for E2F family proteins, but not for any of the other transcription factors Evi1 has been shown to interact with. The precise mechanism by which Evi1 represses caIII gene expression awaits a more detailed analysis of the )1485 ⁄ +55 gene promoter region. These data show that Evi1 represses transcription of caIII in Rat1 cells (5.6 cells) and as a consequence these cells are vulnerable to oxidative stress. This raises the possibility that Evi1 regulates caIII in other cell types and if the caIII protein is an important antioxidant, then they too would be vulnerable to oxidative stress. EVI1 is overexpressed in some human neoplasias, including acute myeloid leukaemias [38] and hepatocel- lular carcinoma [4]. CAIII is also very abundant in nor- mal liver and presumably is an important antioxidant in this tissue. Interestingly, CAIII expression is reduced in human hepatocellular carcinoma [39], although it is not known which tumours overexpress EVI1. EVI1 might be responsible for CAIII repression in some cases and perhaps different mechanisms operate in others. CAIII expression in haemopoietic cells and leukaemia cells has not been described. There is some evidence that CAIII might operate as an antioxidant in erythrocytes in cer- tain anaemias [40], suggesting that it might be important in protecting haemopoietic cells from oxidative stress. It would be interesting to assess the expression level of CAIII in normal haemopoetic cells and leukaemia cells to determine if it is reduced in these neoplasias and if levels are inversely proportional to EVI1 expression. If this is the case, then tumours overexpressing EVI1 might be vulnerable to therapeutic agents that induce oxidative stress. Materials and methods Preparation of plasmid DNA Plasmids p50MX-neo, p50M5.6neo, pGL2tkluc, pBluescript KSII, pCMVcar3 (I.M.A.G.E. Id 4195712) and pRLCMV have all been described previously [9] and were obtained from Promega UK (Southampton, UK), Stratagene (La Jolla, CA, USA) and Source Bioscience, geneservice (Cam- bridge, UK). The construction of pGL3-caIII ()1485 ⁄ +55) has not been published (Y. Ishii, unpublished data). Plas- mid DNAs were prepared by affinity chromatography using Nucleobond Ò PC500EF gravity flow columns according to the manufacturer’s instructions (Macherey-Nagal, Du ¨ ren, Germany). Cell culture Rat1 and EcoPak2 cells were cultured in complete medium comprising Dulbecco’s modified Eagle’s medium (DMEM, Evi1 enhanced oxidant-induced apoptosis P. Roy et al. 448 FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS Lonza Group, Basel, Switzerland, BE12-604F) supple- mented with 5% newborn calf serum (Sigma-Aldrich, Poole, UK, N4637) or 10% fetal calf serum (Lonza Group, DE14-801F), respectively, and 2.5 mm glutamine, 50 lgÆmL )1 penicillin, 50 unitsÆmL )1 streptomycin (Lonza Group, BE17-605E and BE17-603E), 37 °C, 5% CO 2 . For retrovirus production, EcoPak2 cells (Clontech-Takara Bio Europe, Saint-Germain-en-Laye, France) were plated on collagen (Sigma-Aldrich, C38671) coated dishes and tran- siently transfected with either p50M5.6-neo or p50MX-neo using the calcium phosphate coprecipitate method described previously [41]. Virus was harvested and used to infect Rat1 fibroblasts, as described previously [9]; infected cells were selected in complete medium supplemented with 50 lgÆmL )1 G418 (Invitrogen, Paisley, UK). For paclitaxel and H 2 O 2 treatment, cells were incubated in complete med- ium supplemented with either 1 lm paclitaxel (Sigma- Aldrich, T7191) or 100–750 lm H 2 O 2 (Sigma-Aldrich, 21676) for 16 h. Western blotting Protein extracts, SDS ⁄ PAGE and western blotting were performed as described previously [9] with either a-caIII (Santa Cruz Biotechnology, Santa Cruz, CA, USA, E-19), a-Evi1 (1806) or a-gapdh (Fitzgerald Industries, North Acton, MA, USA, 6C5) and diluted 1 ⁄ 200 or 1 ⁄ 5000 (1806 and 6C5). Appropriate horseradish peroxidase-conjugated anti-goat (Sigma-Aldrich, A5420), anti-rabbit (Sigma- Aldrich, A9169) or anti-mouse (Sigma-Aldrich, A9044) IgG secondary antibodies were used at 1 ⁄ 5000 dilutions and detection was performed by enhanced chemiluminescence (Pierce, Rockford, IL, USA, 32209). DNA-mediated transfection and reporter assays Rat1 cells and derivatives were transfected using Fugene6 Ò (Roche Diagnostics, Mannheim, Germany, 11815091001). For reporter assays, cells were transfected with recombinant pGL3-caIII ()1485 ⁄ +55) and pRLCMV plasmid DNAs. Constant DNA concentrations were maintained with pBlue- script KSII. Cells (5 · 10 3 ) were incubated with a 1 : 6 ratio DNA : FuGENE6 Ò , prepared as described by the manufac- turer, in 96-well plates (Costar, New York, NY, USA, 3917) for 48 h. Cells were lysed, and luciferase activity determined using the dual-luciferase reporter assay system (Promega, TM046) in a Fluostar OPTIMA luminometer (BMG LABTECH, Offenburg, Germany). Oligonucleotides Gene-specific oligonucleotides were designed using primer express software version 3.0 (Applied Biosystems), synthe- sized and supplied by Eurogentec (Seraing, Belgium): 5¢ rat caIII: ccgggactattggacctacca 3¢ rat caIII: cagtagcagccacacaatgca 5¢ HEX, 3¢ TAMRA rat caIII probe: cttcaccacgccaccctgc gag 5¢ rat gapdh: gggcagcccagaacatca 3¢ rat gapdh: ccgttcagctctgggatgac 5¢ 6-FAM, 3¢ TAMRA rat gapdh probe: ccctgcatccactgg tgctgcc Preparation of total cellular RNA, cDNA synthesis and RTQ-RT-PCR RNA was prepared from semiconfluent cultures of cells using the Trizol method (Invitrogen, 1559-026). Total cellular RNA (1 lg) was used to synthesize cDNA with Supermix III first-strand strand synthesis for QPCR according to the man- ufacturer’s instructions (Invitrogen, 11752). Five per cent of the cDNA reaction was used for RTQ-PCR using the ABso- lute QPCR mix (ABgene, Epsom, UK, AB-4136), gene-spe- cific oligonucleotide primers and dual-labelled probes, 95 °C, 15 min followed by 40 cycles 95 °C, 30 s, 60 °C, 30 s in an OPTICON 2 DNA engine (MJ Research, Watertown, MA, USA). The efficiency of the RTQ-PCR reactions was calculated using the formula efficiency = )1+10 ()1 ⁄ slope) against the standard curve of each assay over a gradient of template concentration with each gene. The efficiencies for caIII and gapdh primers ⁄ probes were 90 and 101%, respectively. Rela- tive expression levels between caIII and gapdh were deter- mined using the arithmetic comparative 2 )DDCt method [42] and were determined relative to caIII in Rat1 cells (cali brator). Caspase 3 assay Cells were incubated in 96-well dishes (Costar 3917), trea- ted with various agents and apoptosis determined using the Caspase 3 ⁄ 7-Glo Ò assay according to the manufacturer’s instructions (Promega, G8090), measuring luminescence with a Fluostar OPTIMA luminometer (BMG LABTECH). MTT assays MTT assays were performed on cells grown in 96-well tissue culture plates following treatment with H 2 O 2 . Cells were treated with 500 lgÆmL )1 MTT (Sigma-Aldrich, M5655) in complete medium, 37 °C, 5% CO 2 , 1 h. The medium was removed and replaced with 100 lL dimethylsulfoxide (Sigma-Aldrich, 472301) and absorbance measured at 570 nm using an MRX plate reader (Dynatech Laboratories, Guernsey Channel Island, UK). The final absorbance was determined by subtracting the absorbance of treated wells lacking cells. The formazan concentration was determined using the formula: c (formazan concentration; lm)=A (absorbance) ⁄ e (extinction coefficient) 1 (path length). P. Roy et al. Evi1 enhanced oxidant-induced apoptosis FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS 449 Knockdown of rat caIII Rat caIII knockdown was achieved in Rat1 cells using Tri- FECTa DsiRNA Kit RNC.RNAI.NO19292.9 (Integrated DNA Technologies, Leuven, Belgium). SiRNA1 (5¢-CCA UUGAACUGCAUACUAAAGACAT-3¢,5¢-AUGUCUU UAGUAUGCAGUUCAAUGGGU-3¢) was found to be the most effective and used for these studies. The control DsiRNA sequence used was (5¢-CUUCCUCUCUUUCUC UCCCUUGUGA-3¢,5¢ UCACAAGGGAGAGAAAGA GAGGAAGGA-3¢). In total, 1 · 10 5 Rat1 cells per well were seeded in a 12-well tissue culture plate in compete medium and incubated, 37 °C, 5% CO 2 . Twenty-four hours later the medium was removed and replaced with 600 lL compete medium. After 1 h, 1.5 lL Silentfect Ò (BioRad, Hercules, CA, USA, 170-3360) in 50 lL DMEM was mixed with 50 lL DMEM containing DsiRNA and added to the cells, giving a final DsiRNA concentration of 10 nm. The cells were incubated, 37 °C, 5% CO 2 , for 48 h prior to isolation of whole cell protein extracts or treatment with H 2 O 2 or 24 h for isolation of total cellular RNA. ROS assay The ROS assay was performed by labelling cells with DCF-DA [43]. Cells grown in complete medium were labelled for 30 min with 20 lm DCF-DA (Sigma-Aldrich, 35845), 37 °C, 5% CO 2 . The cells were trypsinized, pelleted and washed three times with ice-cold phosphate-buffered saline (Lonza Group, BE17-516F), then analysed for fluorescence by FACS (FACSCaliber, Becton Dickinson, Oxford, UK). Acknowledgements This work was funded by a Glasgow Caledonian University PhD studentship and Overseas Research Student Award Scheme (PR) and in part by the Leukaemia Research Fund (CB, 08018). References 1 Weiser R (2007) The oncogene and developmental regu- lator EVI1: expression, biochemical properties, and bio- logical functions. Gene 396, 346–357. 2 Kurokawa M, Mitani K, Yamagata T, Takahashi T, Izutsu K, Ogawa S, Moriguchi T, Nishida E, Yazaki Y & Hirai H (2000) The evi-1 oncoprotein inhibits c-Jun N-terminal kinase and prevents stress-induced cell death. EMBO J 19, 2958–2968. 3 Buonamici S, Li D, Mikhail FM, Sassano A, Platanias LC, Colamonici O, Anastasi J & Nucifora G (2005) EVI1 abrogates interferon-alpha response by selectively blocking PML induction. J Biol Chem 280, 428–436. 4 Liu Y, Chen L, Ko TC, Fields AP & Thompson EA (2006) Evi1 is a survival factor which conveys resistance to both TGFbeta- and taxol-mediated cell death via PI3K ⁄ AKT. Oncogene 25, 3565–3575. 5 Shackelford D, Kenific C, Blusztajn A, Waxman S & Ren R (2006) Targeted degradation of the AML1 ⁄ MD- S1 ⁄ EVI1 oncoprotein by arsenic trioxide. Cancer Res 66, 11360–11369. 6 Yuasa H, Oike Y, Iwama A, Nishikata I, Sugiyama D, Perkins A, Mucenski ML, Suda T & Morishita K (2005) Oncogenic transcription factor Evi1 regulates hematopoietic stem cell proliferation through GATA-2 expression. EMBO J 24, 1976–1987. 7 Hoyt PR, Bartholomew C, Davis AJ, Yutzey K, Gamer LW, Potter SS, Ihle JN & Mucenski ML (1997) The Evi1 proto-oncogene is required at midgestation for neural, heart, and paraxial mesenchyme development. Mech Dev 65, 55–70. 8 Morishita K, Parker DS, Mucenski ML, Jenkins NA, Copeland NG & Ihle JN (1988) Retroviral activation of a novel gene encoding a zinc finger protein in IL-3-dependent myeloid leukemia cell lines. Cell 54, 831–840. 9 Bartholomew C, Kilbey A, Clark AM & Walker M (1997) The Evi-1 proto-oncogene encodes a transcrip- tional repressor activity associated with transformation. Oncogene 14, 569–577. 10 Delwel R, Funabiki T, Kreider BL, Morishita K & Ihle JN (1993) Four of the seven zinc fingers of the Evi-1 myeloid-transforming gene are required for sequence- specific binding to GA(C ⁄ T)AAGA(T ⁄ C)AAGATAA. Mol Cell Biol 13, 4291–4300. 11 Funabiki T, Kreider BL & Ihle JN (1994) The carboxyl domain of zinc fingers of the Evi-1 myeloid transform- ing gene binds a consensus sequence of GAAGAT- GAG. Oncogene 9, 1575–1581. 12 Kurokawa M, Mitani K, Irie K, Matsuyama T, Takahashi T, Chiba S, Yazaki Y, Matsumoto K & Hirai H (1998) The oncoprotein Evi-1 represses TGF-beta signalling by inhibiting Smad3. Nature 394, 92–96. 13 Izutsu K, Kurokawa M, Imai Y, Maki K, Mitani K & Hirai H (2001) The corepressor CtBP interacts with Evi-1 to repress transforming growth factor beta signal- ing. Blood 97, 2815–2822. 14 Ames BN, Shigenaga MK & Hagen TM (1993) Oxidants, antioxidants, and the degenerative diseases of aging. Proc Natl Acad Sci USA 90, 7915–7922. 15 Wu WS (2006) The signalling of ROS in tumor progres- sion. Cancer Metastasis Rev 25, 695–705. 16 Wistrand PJ (2002) Carbonic anhydrase III in liver and muscle of male rats. Purification and properties. Ups J Med Sci 107, 77–88. 17 Lehtonen J, Shen B, Vihinen M, Casini A, Scozzafava A, Supuran CT, Parkkila AK, Saarnio J, Kivela ¨ AJ, Evi1 enhanced oxidant-induced apoptosis P. Roy et al. 450 FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS [...]... concerted upregulation of methionine metabolic pathway enzymes, and downregulation of carbonic anhydrase- III, in betaine supplemented ethanol-fed rats Biochem Biophys Res Commun 381, 523–527 Zoll J, Ponsot E, Dufour S, Doutreleau S, VenturaClapier R, Vogt M, Hoppeler H, Richard R & Fluck M ¨ (2006) Exercise training in normobaric hypoxia in Evi1 enhanced oxidant-induced apoptosis 29 30 31 32 33 34... Bogdanovich S & Forster R (2004) Anti-oxidative response of carbonic anhydrase III in skeletal muscle IUBMB Life 56, 343–347 Khanna-Gupta A, Lopingco MC, Savinelli T, Zibello T, Berliner N & Perkins AS (1996) Retroviral insertional activation of the EVI1 oncogene does not prevent G-CSF-induced maturation of the murine pluripotent myeloid cell line 32Dcl3 Oncogene 12, 563–569 Alzuherri H, McGilvray R,... (2005) Differential expression of carbonic anhydrase isoenzymes in various types of anemia Clin Chim Acta 351, 79–86 Wigler M, Silverstein S, Lee LS, Pellicer A, Cheng Y & Axel R (1977) Transfer of purified herpes virus FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal compilation ª 2009 FEBS 451 Evi1 enhanced oxidant-induced apoptosis P Roy et al thymidine kinase gene to cultured mouse cells Cell... Yatsula B, Lin S, Read AJ, Poholek A, Yates K, Yue D, Hui P & Perkins AS (2005) Identification of binding sites of EVI1 in mammalian cells J Biol Chem 280, 30712–30722 Laricchia-Robbio L, Premanand K, Rinaldi CR & Nucifora G (2009) EVI1 impairs myelopoiesis by deregulation of PU.1 function Cancer Res 69, 1633–1642 Senyuk V, Sinha KK, Li D, Rinaldi CR, Yanamandra S & Nucifora G (2007) Repression of RUNX1... RUNX1 activity by EVI1: a new role of EVI1 in leukemogenesis Cancer Res 67, 5658–5666 Laricchia-Robbio L, Fazzina R, Li D, Rinaldi CR, Sinha KK, Chakraborty S & Nucifora G (2006) Point mutations in two EVI1 Zn fingers abolish EVI1GATA1 interaction and allow erythroid differentiation of murine bone marrow cells Mol Cell Biol 26, 7658– 7666 Chi Y, Senyuk V, Chakraborty S & Nucifora G (2003) EVI1 promotes... proliferation by interacting with BRG1 and blocking the repression of BRG1 on E2F1 activity J Biol Chem 278, 49806–49811 Nucifora G, Laricchia-Robbio L & Senyuk V (2006) EVI1 and hematopoietic disorders: history and perspectives Gene 368, 1–11 Kuo WH, Chiang WL, Yang SF, Yeh KT, Yeh CM, Hsieh YS & Chu SC (2003) The differential expression of cytosolic carbonic anhydrase in human hepatocellular carcinoma Life... of CA XIII, a novel member of the carbonic anhydrase isozyme family J Biol Chem 279, 2719–2727 Carter ND, Lonnerholm G, Meyerson BJ & Wistrand PJ ¨ (2001) Androgen-linked control of carbonic anhydrase III expression occurs in rat perivenous hepatocytes; an immunocytochemical study Ups J Med Sci 106, 67–76 Parkkila S, Halsted CH, Villanevva JA, Vaananen HK ¨ ¨ ¨ & Niemela O (1999) Expression of testosterone-depen¨... enzyme, carbonic anyhdrase III and oxidative stress in experimental alcoholic liver disease Dig Dis Sci 44, 2205–2213 Raisanen SR, Lehenkari P, Tasana M, Rahkila P, Harkonen PL & Vaananen HK (1999) Carbonic anhy¨ ¨ ¨ drase III protects cells from hydrogen peroxide- induced apoptosis FASEB J 13, 513–522 Ishii Y, Akazawa D, Aoki Y, Yamada H & Oguri K (2005) Suppression of carbonic anhydrase III mRNA... are not copy-edited or typeset Technical support issues arising from supporting information (other than missing files) should be addressed to the authors Supporting information The following supplementary material is available: Fig S1 DsiRNA-mediated knockdown of caIII mRNA and protein and enhanced caspase 3 catalytic activity in Rat1 cells 452 FEBS Journal 277 (2010) 441–452 ª 2009 The Authors Journal... (2006) Conservation and expression of a novel alternatively spliced Evi1 exon Gene 384, 154–162 Sood R, Talwar-Trikha A, Chakrabarti SR & Nucifora G (1999) MDS1 ⁄ EVI1 enhances TGF-beta1 signaling and strengthens its growth-inhibitory effect but the leukemia-associated fusion protein AML1 ⁄ MDS1 ⁄ EVI1, product of the t(3;21), abrogates growth-inhibition in response to TGF-beta1 Leukemia 13, 348–357 . Enhanced sensitivity to hydrogen peroxide-induced apoptosis in Evi1 transformed Rat1 fibroblasts due to repression of carbonic anhydrase III P. Roy 1 , E. Reavey 1 ,. H 2 O 2 -induced apoptosis in Rat1 cells expressing an Evi1 transgene. Surprisingly, we found that Evi1 expression increases sensitivity to H 2 O 2 -mediated apoptosis in complete contrast to the. effect of other apoptosis- inducing agents. Increased sensitivity is primarily due to the transcriptional downregulation of caIII gene expression mediated by Evi1. Results Evi1- expressing Rat1 fibroblasts