Glucosidase II β-subunit, a novel substrate for caspase-3-like activity in rice, plays as a molecular switch between autophagy and programmed cell death

Endoplasmic reticulum (ER) stress activates unfolded protein response (UPR) and autophagy. However, prolonged, severe stresses activate programmed cell death (PCD) in both animal and plant cells. Compared to the well-studied UPR pathway, the molecular mechanisms of ER-stress-induced PCD are less understood. Here, we report the identification of Gas2, the glucosidase II β subunit in the ER, as a potential switch between PCD and autophagy in rice. MS analysis identified Gas2, GRP94, and HSP40 protein in a purified caspase-3-like activity from heat stressed rice cell suspensions. The three corresponding genes were down-regulated under DTT-induced ER stress. Gas2 and GRP94 were localized to the ER, while HSP40 localized to the cytoplasm. Compared to wild-type, a Gas2 RNAi cell line was much sensitive to DTT treatment and had high levels of autophagy. Both caspase-3 and heat-stressed cell suspension lysate could cleave Gas2, producing a 14 kDa N-terminal fragment. Conditional expression of corresponding C-terminal fragment resulted in enhanced caspase-3-like activity in the protoplasts under heat stress. We proposed that mild ER stress causes down-regulation of Gas2 and induces autophagy, while severe stress results in Gas2 cleavage by caspase-3-like activity and the cleavage product amplifies this activity, possibly participating in the initiation of PCD.


Results
Gas2/HSP90/HSP40 formed caspase-3 related protein complex in rice suspension cells under heat stress. Heat stress is a known inducer of caspase-like activities and PCD in Arabidopsis 29 , tobacco 30 , and other plant species. Previously, heat treatment has been demonstrated to induce caspase-3-like activity 31 and PCD in rice suspension cells 32 . In order to identify the putative caspase-3 in rice, caspase-3-like activity was monitored in whole cell lysate and attempts were made to purify this activity by gel filtration, mono-Q hydrophobic chromatography, and biotinyl-DEVD-CHO affinity chromatography. Two elution peaks (I and II) were noted using a NaCl linear gradient (Fig. 1a) that corresponded to significant caspase-3-like activity (Fig. 1b), and these two peaks were identified by LC-MS/MS (Table 1). Three proteins were identified from peak I, which included Gas2, glucose-regulated protein 94 (GRP94), and TPR-containing protein. Additionally, three proteins were identified from peak II, which included methylmalonate-semialdehyde dehydrogenase, peptidyl prolyl cis trans isomerase, and the beta subunit of ATP synthase. Sequence homologue analysis indicated that the TPR-containing protein has a HSP binding motif, STI1, at the C terminus (data not shown), which indicted that this protein likely belongs to the HSP40 family 33 . No caspase-like proteases were identified from peak I and peak II purification, possibly due to the low abundant of this caspase activity. As such, caspase-3-related proteins from peak I were the main focus of the rest of the study.

Expression profile of Gas2 under different stresses.
Under physiological conditions, Gas2 is a non-catalytic subunit of Glucosidase II, which participates in the glycosylation of the proteins in the ER. In order to identify the biological roles of Gas2 under abiotic stresses, Gas2 expression was measured by quantitative real-time polymerase chain reaction under several conditions, including under DTT-induced ER stress. Treatment with 6.4 mM DTT induced expression of OsbZIP50 34 , the marker gene of UPR, which indicated that DTT was able to induce ER stress in rice suspension cells (Fig. 2a) and, under these conditions, Gas2 was down-regulated (Fig. 2b). In addition to ER stress, Gas2 was also down-regulated under different abiotic stresses, such as cold (Fig. 2c), osmotic (Fig. 2d), and salt (Fig. 2e) stresses. Sub-cellular localization. As Gas2 forms a caspase-3 related protein complex with GRP94 and TPR-containing HSP40, the sub-cellular localization of these three proteins was examined. As expected, Gas2 and GRP94 were localized to the ER (Fig. 3a,b), while TPR-containing HSP40 was localized to the cytoplasm (Fig. 3c). Since caspase-3-like activity was present in the cytoplasm, this suggested that under specific stress conditions, TPR-containing HSP40 may form a complex with Gas2 and GRP94 at the interface between ER and cytoplasm, as the ER is known to provide attachment sites for many cytosolic proteins.
Down-regulation of Gas2 induced autophagy in rice suspension cell. In order to characterize the function of Gas2 in rice, a Gas2 RNAi transgenic cell line was constructed, and the expression of Gas2 was reduced significantly compared to wild-type (Fig. 4a). Gas2 RNAi cell line demonstrated normal growth using Scientific RepoRts | 6:31764 | DOI: 10.1038/srep31764 standard medium (data not shown), however, after DTT treatment, the expression of OsbZIP50 increased in the GAS2 RNAi line compared to the wild-type (Fig. 2a), which indicated that DTT-induced UPR was amplified in the GAS2 RNAi line. At the same time, cytosolic Ca 2+ levels were increased > 2-fold in the transgenic lines (Fig. 4b,c). Consistent with noted increased Ca 2+ levels in the cytoplasm, the autophagosome was observed by LTR staining in transgenic line, but not in wild-type after DTT treatment (Fig. 4d), and this effect could be inhibited by 3-MA, a specific inhibitor of autophagosome formation. Electric microscopy also indicated the formation of double membrane autophagosomes in the RNAi line under DTT treatment (Fig. 4e). Taken together, these results indicated that down-regulation of Gas2 in rice led to higher sensitivity to DTT and accelerated DTT-induced autophagy in rice suspension cells.
Gas2 is a novel substrate of caspase-3-like activity in rice. Gas2/GRP94/HSP40 forms a caspase-3 related complex, which suggests a relationship between Gas2 and caspase-3-like activity. In order to examine this relationship, recombinant OsGas2 was expressed in and purified from E. coli. The purified Gas2-His 6 was incubated with caspase-3 (Sigma-Aldrich) for 5 h at 37 °C and the products were analyzed by immunoblotting using an anti-His-tag antibody. A cleavage product with a MW of 14 kDa could be detected (Fig. 5a), indicating that caspase-3 digested the recombinant OsGas2 and produced a His 6 -tagged 14 kDa N-terminal fragment. In order to verify this result, cytosolic fractions which exhibited caspase-3-like activity were prepared from heat-treated suspension cells. Purified Gas2-His 6 was incubated with the cytosolic fractions for 5 h at 37 °C and the cleavage products were detected by immunoblotting, as was the 14 kDa cleavage product (Fig. 5a). Consistent with this result, a putative caspase-3 cleavage site (DEYD 113 S) could be found in the OsGas2 sequence, and caspase-3 and cytosolic fractions mentioned above could not cleave the D 110 A/D 113 A mutant (Fig. 5a). These data demonstrate for the first time that OsGas2 is a novel substrate of caspase-3-like activity in rice and that cleavage at D 113 resulted in a 14 kDa N-terminal fragment and a corresponding C-terminal fragment (Fig. 5b).
Cleavage product of Gas2 amplified caspase-3-like activity in rice protoplasts. In order to identify the possible function of the cleavage products of Gas2, the N-and C-fragments of Gas2 were expressed in rice mesophyll protoplasts using a Dex-induced conditional expression vector and Dex effectively induced the fragment expression (Fig. 6a). Compared to the expression of the N-fragment and vector control, C-fragment expression resulted in a significant increase in caspase-3-like activity after heat treatment (Fig. 6b). Interestingly, no autophagosomes were detected in protoplasts expressing the two fragments after DTT treatment (data not shown).  Discussion ER stress is a known inducer of autophagy in animal cells. Elicitors of ER stress, such as TM and DTT, can also induce autophagosome formation in Arabidopsis 3 . In this study, DTT treatment was demonstrated to induce OsbZIP50 expression, the marker gene of ER stress in rice, and a Gas2 RNAi cell line was much more sensitive to DTT than wild-type, which suggested that Gas2 might be involved in DTT-induced ER stress in rice suspension cell (Fig. 2). Further, for the first time Gas2 has been implicated in serving a dual function in the crosstalk between autophagy and PCD. As shown in Fig. 4, DTT-induced Ca 2+ release (Fig. 4b,c) and autophagy (Fig. 4d,e) was more pronounced in RNAi line than in the wild-type. Similar to these results, Yang et al. 27 reported that knockdown of Gas2 in HeLa cell induces autophagy through the mammalian target of rapamycin, although the UPR pathway was not disturbed and no changes in the steady-state concentration of cytosolic Ca 2+ were observed. These opposing results indicate that, although Gas2 has conserved functions between the animal and plant cells, the molecular mechanisms might be different. On the other hand, it has been reported that Gas2 could interact with IP 3 R, the calcium channel in ER, and participate in the release of Ca 2+ from ER in animal cells 35 . Although this study did not investigate the relationship between Gas2 and IP 3 R in rice, the results demonstrated that Gas2 participates in the regulation of cytosolic Ca 2+ concentration in plants.
Several caspase-like enzymatic activities have been reported to participate in plant ER stress-induced PCD 36 . In this study, Gas2/GRP94/HSP40 was identified as a caspase-3-related complex, further demonstrating that Gas2 is a novel substrate for caspase-3-like activity. These results suggested that, under severe stress (e.g. heat treatment), heat shock proteins GRP94 and HSP40 provide a processing platform for Gas2, which is then cleaved by caspase-3-like activity at the DEYD 109 S site (Fig. 5). As a result, the C-terminal cleavage product of Gas2 further promotes caspase-3-like activity (Fig. 6b) and forms a caspase-3-amplifying feedback loop. Recently, several caspases were identified as regulators between autophagy and apoptosis in animal cells 37 -caspases can inhibit autophagy by cleaving autophagy-related proteins. After cleavage, the proteins can be converted into pro-apoptotic molecules to induce/promote apoptosis; for example, autophagy-related Atg4D can be cleaved by caspase-3 at the DEVD 63 K site 38 . Overexpression of the C-terminal cleavage products of Atg4D translocated to mitochondria and induced cell death in human cells 38 . Similar with these reports, the C-terminal cleavage product of Gas2 could also amplify the caspase-3-like activity. These results suggest that Gas2 might be a sensor of ER stress in plant cells. As shown in Fig. 7, under mild ER stress, Gas2 was down-regulated, which induce the release of Ca 2+ from ER to cytoplasm, probably through IP 3 R pathway, and then led to autophagy. Under severe ER stress, HSP40 recruit caspase-3-like activity in cytoplasm, and formed a HSP40/Caspase3/GRP94/Gas2 complex at the interface between ER and cytoplasm. In this complex, caspase-3-like activity cleaved Gas2 at DEYD site. As a product of the cleavage, C-fragment of Gas2 amplified caspase-3 like activity. C-fragment of Gas2 might play partial role in initiating PCD, as DNA fragmentation and cell death was not observed after C-terminal fragment expression, indicating that this amplifying feedback loop is not the only initiator of PCD. As previously mentioned, two caspase-3-related complexes were identified in this study. In addition to Gas2/GRP94/HSP40, FKBP/ ATP synthase beta subunit/MMDH was found to form another caspase-3-related complex. In animal cells, the FKBP/ATP synthase subunit has been shown to form PTP transition pores in the inner membrane of mitochondria, which participate in the release of cytochrome c and initiate apoptosis 39 . Although the exact role of this complex in plant PCD is unknown, the FKBP/ATP synthase subunit could participate in activation of caspase-3-like activity through a mitochondrial pathway rather than an ER pathway. This hypothesis needs further investigation to clarify the role of this complex in plant PCD. We are also aware of that in the model in Fig. 7, the mechanism by which GRP94 and Gas2 in the ER interacted with cytosolic HSP40 and caspase activity is still lacking. One possible explanation is that under severe ER stress, GRP94 participated in the retro-translocation of Gas2 from the ER to the cytosol, through ERAD pathway 40 . As a result, Gas2 might be cleaved by caspase-3-like activity at the cytoplasm side of ER membrane. However, the precise events need to be elucidated in the near future.  Purification and identification of caspase-3-like activity-related proteins. In order to purify the caspase-3-like activity-related proteins, 3-4 d rice cell suspensions 32 were treated at 48 °C for 15 min, and then were allowed to recover at 28 °C. 300 g (fresh weight) of treated cells were collected and then ground in  Protoplast preparation. Rice suspension cells from wild-type and GAS2 RNAi lines were routinely propagated and cultured at 28 °C. Protoplasts were isolated from 3-4 d suspension cells according to Maas et al. 42 . For the protoplast isolation from green tissue, rice seedlings were grown in the nutrient solution 28 °C for 7-10 days. 10 cm high seedlings were used for protoplast isolation according to Wang et al. 43 .

Methods
RNA extraction from rice seedlings. Total RNA was isolated from rice protoplasts or roots of seedlings using the Trizol reagent (Takara, Japan). Total RNA (5 μ g) was used for cDNA synthesis with the PrimeScript ™ RT Reagent Kit with gDNA Eraser (Takara, Japan) according to the manufacturer's instructions.

Real-time quantitative reverse transcription polymerase chain reaction analysis. Quantitative
real-time polymerase chain reaction was conducted using a CFX 96 fluorescent quantitative PCR apparatus (Bio-Rad, Hercules, CA, USA). Gene-specific primers were used to detect the OsGAS2 transcripts and 17S rRNA was used as an internal control. Each sample was performed in three independent experiments and the primer sequences are detailed below.
OsGAS2-F: TTGGTAAGGAGAAGGAGTTC; OsGAS2-R: AGGCTGGTGGTACTATGT.  Subcellular localization of the OsGAS2, OsGrp94 and HSP40 protein in rice mesophyll protoplasts. For the subcellular localization study, the full length OsGAS2 (Os01g0276800), OSGrp94 (Os06g0716700) and putative HSP40 (Os02g0100300) cDNA were amplified using gene-specific primers (BamH I and Kpn I site are shown in italics): OsGAS2: F-CGGGATCCATGGGGCTCCACGCGATCC; R-GGGGTACCGCGAGTTCATCATGGTCACGCT. OsGrp94: F-CGGGATCCATGCGCAAGTGGGCGCTCTCC; R-GGGGTACCCTACAGCTCGTCCTTATCATA. HSP40: F-CGGGATCCATGGATGCTTCTCGCGTTGGC; R-GGGGTACCTTACTGGGACCCATTGAATTT. The PCR products were confirmed by sequencing and then gel-purified with the AxyPrep DNA Gel Extraction Kit (Axygen, China). The products were then double digested with BamH I and Kpn I and cloned into the subcellular localization vector pXZP008 44 . A reporter gene encoding green fluorescent protein (GFP) was fused to OsGAS2, OsGrp94 and OsHSP40, which is driven by the cauliflower mosaic virus 35S promoter (35S: OsGAS2/OsGrp94/OsHSP40-GFP), and then transformed and transiently expressed in rice mesophyll protoplasts. After incubation for 16 h at 26 °C in the dark, transformed protoplasts were stained with 100 nM ER-Tracker ™ Blue-White DPX (E12353, Molecular Probes, USA) for 30 min at 28 °C in darkness and washed twice with W5 solution. Fluorescence was visualized using a PerkinElmer UltraVIEW VoX confocal microscope. The ER was identified using E12353 as a control. The sequencing-confirmed PCR products were gel-purified and double digested with Nde I/EcoR I, cloned into pET-28a with His 6 -tag, and then transformed into E. coli DE3. Expression was induced using 0.3 mM isopropyl-b-D-1-thiogalactopyranoside (IPTG) at 37 °C for 4 h. Cells were harvested and resuspended in lysis buffer (50 mM NaH 2 PO 4 (pH 8.0), 300 mM NaCl). After ultrasonic disruption, the mixture was then centrifuged at 10,000 × g for 1 h at 4 °C. The resulting supernatant was loaded onto a Ni-NTA resin column (GenScript, China). After washing the column, His-tagged OsGAS2 was eluted using elution buffer (50 mM NaH 2 PO 4 (pH 8.0), 300 mM NaCl, and 250 mM imidazole).

Conditional expression of OsGAS2 fragments in rice mesophyll protoplasts. OsGAS2 fragments:
N-and C-framents were amplified using gene-specific primers (XhoI and SpeI sites are shown in italics): OsGAS2-N: F-CGGGATCCATGGCCTCCAGGCCGCCGCTC OsGAS2-N: R-GGGGTACCAAATCATACTCATCACTCCCGTCGCAG OsGAS2-C: F-CGGGATCCATGAGCAATGTCACTTGCAAGAATAC OsGAS2-C: R-GGGGTACCAAGAGTTCATCATGGTCACGCTGG The products were double digested with Xho I/Spe I, and then directionally cloned into the dexamethasone (Dex)-inducible binary vector pTA7002. The resulting constructs and the vector control (20-25 μ g) were transformed into rice mesophyll protoplasts.
Caspase-3 like activity assay. Protoplasts transformed with the pTA7002 vector control or pTA7002-N/C were resuspended in W5 medium (supplemented with 10 mM Dex) and incubated in a six-well culture plate at 28 °C in darkness for specific amounts of time. After incubation, the protoplasts were treated at 48 °C for 15 min and allowed to recover at 28 °C for 6 h. The protoplasts were suspended in lysis buffer (50 mM HEPES (pH 7.4), 100 mM NaCl, 250 mM sucrose, 0.1% CHAPS, 1 mM DTT, 0.1 mM EDTA, 1 mg/mL pepstatinin, 8 mg/mL aprotinin, 10 mg/mL leupeptin). The suspensions were homogenized on ice and centrifuged at 12,000 × g for 5 min at 4 °C, the soluble cytosolic fractions were collected and incubated in assay buffer (50 mM HEPES-KOH (pH 7.5), 10% glycerol, 50 mM KCl, 2.5 mM MgCl 2 , and 1 mM DTT) supplemented with 70 mM N-acetyl-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin (Ac-DEVD-AMC, EnzoLife Sciences, USA). The reaction system was incubated without light at 30 °C for 1 h. The fluorescence was measured in a microplate reader (TECAN, USA) at 380/460 nm. The difference in fluorescence before and after incubation was normalized against the corresponding protein concentration. All assays were performed using three independent samples. Electron microscopy. DTT-treated protoplasts were fixed with 2.5% (w/v) glutaraldehyde in CPW-9M (pH 5.8) at 4 °C overnight and post-fixed with 1% (w/v) osmium tetroxide for 3 h at room temperature. After staining with 1% (w/v) uranyl acetate at room temperature for 2 h, the protoplasts were dehydrated using several different mixtures of ethanol, water, and propylene oxide, and then embedded with Spurr's resin. Sections Scientific RepoRts | 6:31764 | DOI: 10.1038/srep31764 measuring ~70 nm thick were cut and stained with uranyl acetate and lead nitrate. The section samples were observed with an electron microscope (H-7650, Hitachi).
Measurement of cytosolic Ca 2+ content. The Ca 2+ -sensitive fluorescent dye, Fluo-3-AM was used to detect the relative change of the Ca 2+ -dependent fluorescence in protoplasts. Protoplasts were incubated with 5 μ M preheated Fluo-3-AM (Dojindo Laboratories, Japan) for 30 min at 37 °C, as specified by the manufacturer's instructions. Then the cells were washed twice and re-suspended in HBSS solution containing 0.4 M mannitol. The Fluo-3 fluorescence was measured with a PerkinElmer UltraVIEW VoX confocal microscope using excitation and emission wavelengths of 488 and 525 nm, respectively. The fluorescence intensity of the protoplasts was measured by delimiting the individual protoplast, and the mean fluorescence of the protoplasts was measured and analyzed with Image Pro Plus. 8-10 protoplasts were included for each treatment.
LSDs labeling. Protoplasts treated with DTT and/or 3-MA were collected and incubated with 50 nM LysoTracker Red DND-99 (Molecular Probes, USA) for 20 min at 37 °C, washed with CPW-9M buffer three times, and then examined by a PerkinElmer Ultra VIEW VoX confocal microscope. The wavelength of excitation was 561 nm and the emission signals were measured at 590 nm.