Plants expressing murine pro-apoptotic protein Bid do not have enhanced PCD

Objectives The purpose of this study was to explore whether plant programmed cell death (PCD) cascade can sense the presence of the animal-only BH3 protein Bid, a BCL-2 family protein known to play a regulatory role in the signaling cascade of animal apoptosis. Results We have expressed the mouse pro-apoptotic protein Bid in Arabidopsis thaliana and in Nicotiana tabacum. We did not obtain any transformed plant constitutively expressing the truncated protein (tBid—i.e. the caspase-activated form) whereas ectopic expression of the full-length protein (flBid) does not interfere with growth and development of the transformed plants. To verify whether the presence of this animal pro-apoptotic protein modified stress responses and PCD execution, both N. tabacum and A. thaliana plants constitutively expressing flBid have been studied under different stress conditions triggering cell death activation. The results show that the presence of flBid in transgenic plants did not significantly change the responses to abiotic stress (H2O2 or NO) and biotic stress treatments. Moreover, the finding that no Bid active form was present in treated tobacco plants suggests an absence of a proper activation of Bid.


Introduction
Programmed cell death (PCD) is present both in animal and plant systems, where different modes of action have been recognized and studied. Among these, apoptosis and apoptotic-like PCD differentiate from other forms of cell death and share common morphological characteristics.
In animals, apoptotic cell death involves the Bcl-2 protein family as the initiator of the signaling pathway and a cascade of caspases for the execution of the death program [1]. So far, no Bcl-2 proteins have been identified in plants. Nevertheless, there is evidence that Bcl-2-like proteins exist. In 2001, a gene encoding for a homolog of the human gene BI-1 was isolated from Arabidopsis (AtBI) [2]. The existence of a functional Bax Inhibitor in plant indicated the possibility of a Bax-like mechanism in plants or the presence of a regulatory system similar to that of animals (i.e. pro-and anti-apoptotic mediators). Moreover, when the anti-apoptotic gene Bcl-XL was expressed in tobacco, death triggered by inducers of cellular ROS like UV, paraquat or TMV, was suppressed [3]. In addition, the expression of anti-apoptotic genes (Bcl-Xl and Ced-9) in tomato plants improved tolerance to viral infection and low temperature exposure [4]. These results indicated that a conserved apoptotic pathway might be present in plants with divergent proteins.
In order to study the presence of a common signaling pathway in the sequence of cellular processes occurring during PCD, we have expressed mouse Bid, a pro-apoptotic BH3-only protein, in plant systems. This protein is located at the crossroad of the extrinsic and intrinsic signaling pathways of apoptosis, driving death signal to mitochondria through interactions with pro-and antiapoptotic proteins [5]. Bid exists as a precursor (fulllength protein, flBid) and it becomes activated (truncated Bid, tBid) upon cleavage by Caspase 8 [6]. Other mechanisms, triggered by different stimuli, influence Bid activity such as the reversible binding with lysolipids or cardiolipin.
In the present study, we report the generation of transgenic Arabidopsis thaliana and Nicotiana tabacum plants expressing the mouse Bid protein and the response of transgenic plants to abiotic and biotic stresses.

Plant material and growth conditions
Arabidopsis thaliana wild type (Col-0), Nicotiana tabacum cv. Petit Havana SR1 and transgenic seeds were sown in Petri dishes on water-soaked Whatman filter paper and incubated for 3 days at 4°. Plants were grown under controlled greenhouse conditions (100-120 μmol m −2 s −1 , 16/8 light/dark).

Generation of flBid and tBid over-expressing Arabidopsis thaliana and Nicotiana tabacum lines
To create a Cauliflower Mosaic Virus (CaMV) 35S plant expression construct, the 588 bp long flBid and the 414 tBid coding sequences (GeneBank accession number MMU75506, https ://www.ncbi.nlm.nih.gov/nucco re/ U7550 6.1) were amplified from mouse cDNA by PCR using sequence specific primers (Additional file 1). The amplified sequences were inserted into the expression vector pH2GW7 by LR reaction (Gateway ® LR Clonase ™ , Invitrogen). The resulting constructs were introduced by electroporation into competent Agrobacterium tumefaciens strain GV3101 pMP90RK [8]. A. thaliana Col-0 wild type plants were transformed by floral dip [9]. N. tabacum cv. Petit Havana SR1 wild type plants were transformed by A. tumefaciens cocultivation. Transformed plants were selected on MS [7] medium containing hygromicin. Transgene insertion was verified by PCR and protein over-expression/truncation was confirmed by western blotting with the specific human/mouse Bid polyclonal anti-Bid antibody (R&D Systems). Seeds were harvested and homozygous plants were screened.

DNA laddering analysis
Five micrograms of extracted genomic DNA (DNeasy Plant Mini Kit Qiagen) were run for 60 min at 50 V on 1.0% agarose gel.

Abiotic stress analysis
Oxidative stress experiments For tobacco, discs of about 1 cm diameter were excised from healthy and fully expanded leaves and floated in a water containing different concentrations of H 2 O 2 for 24 h. For Arabidopsis leaf discs (6 mm diameter) from 6-week old plants were floated on water for 30 min. H 2 O 2 was added to 30 mM and conductance measured after 16 and 24 h by using the B-173 compact conductivity meter (HORIBA) as described [10]. For analysis of roots, five-day old seedlings were incubated in 25 mM H 2 O 2 for 5 min and observed after 24 h to detect apoptotic-like PCD symptoms [11].
Nitric oxide fumigation Four weeks old Arabidopsis plants were fumigated for 8 h under dark conditions with NO diluted in air at the given concentrations by using a fumigation system as described [12] to trigger cell death activation. NO concentration was validated by a chemiluminescence-based NO detector machine CLD70E (Eco physics ® ). Surviving plants were counted 7 days after treatment.
Biotic stress analysis HR electrolyte leakage assay to quantify avirulent pathogen induced cell death was performed as described in [13].

H 2 O 2 treatment of wild-type and Bid expressing tobacco plants
To explore if the presence of flBid can interact with the plant cell death signalling pathway either accelerating the process or inducing resistance to stress, both A. thaliana and N. tabacum transformed plants constitutively expressing flBid were analyzed under stress conditions known to be able to induce programmed cell death activation.
Nicotiana tabacum leaf discs were treated with hydrogen peroxide (H 2 O 2 ) at different concentration (5, 10 and 20 mM) and for 24 h. H 2 O 2 was chosen because it can act as a secondary messenger both in cell signalling and during the PCD in response to pathogen attack (HR) [14].
Tobacco leaf discs were observed to evaluate chlorosis and cell death. Leaf discs prepared from 35S::flBid plants showed a similar pattern of chlorosis and cell death to untransformed. The only weak difference observed was in the 10 mM treatment where a higher level of chlorosis seems to be present in flBid leaf discs when compared with WT plants (Fig. 1).
The genomic DNA extracted from leaf discs treated with H 2 O 2 was analyzed to evaluate changes in the time course of DNA fragmentation (DNA ladder, a distinctive trait of apoptosis and programmed cell death) at 0, 1, 3 and 6 h of treatment. No differences were observed between the three lines transformed and the wild type tobacco plants (Additional file 3) and no DNA ladder formation was found.
In addition, we tried to evaluate the formation of tBid after oxidative stress induced by H 2 O 2 in leaf discs. The analysis did not reveal the appearance of the truncated protein following H 2 O 2 treatment in transgenic lines (Additional file 4), suggesting that there is no animal-like Caspase cleavage activity to process the Bid protein.

Nitric oxide fumigation of wild-type and Bid expressing Arabidopsis plants
A fine balance of reactive oxygen intermediates and NO is involved in the activation of PCD during the hypersensitive response to avirulent pathogens [15,16]. In the absence of pathogens, this can be mimicked by exogenous NO treatment [17]. To further enquire whether flBid expression in plants could interact with the cell death activation triggered by external PCD inducing stimuli we treated WT and 35S:flBid Arabidopsis homozygous T3 plants with nitric oxide gas at 160 ppm and 200 ppm in air. As shown in Table 1, no significant difference was found in the response to NO treatment at both NO concentrations between transgenic and wt plants. Therefore, the presence of flBid protein did not interact with the cell death signaling induced by NO. These data indicate that likely no activation of a caspaselike cascade and no cleavage activity similar to that present in animals for processing the Bid protein is present in this system as seen in tobacco, although the absence of a tBid form was not directly enquired.

Avirulent pathogen treatment of wild-type and Bid expressing Arabidopsis plants
Treatment with the avirulent pathogen Pseudomonas syringae pv. tomato carrying AvrB (PstAvrB) induce the activation of an hypersensitive response in plants which is accompanied by the induction of a programmed cell death at the infection site which aims to restrict pathogen growth and spread [18]. To further enquire whether flBid expression could affect program cell death establishment we treated WT plants and plants expressing flBid protein with PstAvrB. Cell death establishment was measured for 48 h by an optimized ion leakage assay [13]. However, no significant difference was found in 35S:flBid compared to wild-type plants (Fig. 2a) demonstrating that flBid protein does not contribute to the plant PCD occurring during the HR.

H 2 O 2 treatment of wild-type and Bid expressing Arabidopsis plants
Finally, A. thaliana plants constitutively expressing flBid were also analyzed under H 2 O 2 treatment conditions known to be able to induce PCD activation, similarly as done for tobacco plants, to determine if there is any flBid cross-talk.
PCD triggered by H 2 O 2 in leaves was measured following treatment with 30 mM H 2 O 2 in leaf discs [10]. No differences were found in the 35S:flBid answer compared to wild type (Fig. 2b).
Moreover, the induction of programmed cell death was further analyzed in roots in A. thaliana plants constitutively expressing the flBid protein, after treatment with H 2 O 2 with 1, 2, 5 and 10 mM H 2 O 2 for 24 h which can activate apoptosis [11]. The presence of changes in the cells of root hairs was observed under the microscope to detect root hair protoplast contraction and vitality (by FDA loading). It was found that A. thaliana flBid lines, after H 2 O 2 treatment, exhibit a reduced number of alive root hairs compared to wild type demonstrating that flBid leads to higher sensitivity to abiotic stresses in transformed plants, but the effect of the protein in these plants is different from PCD activation, as observed in tobacco leaf discs treated with H 2 O 2 (Fig. 1). This is likely due to an increase H 2 O 2 stress sensitivity (i.e. necrosis) since we found apoptosis was not affected (Additional files 5 and 6) thus confirming what we observed in tobacco leaves.

Conclusions
We have expressed mouse Bid pro-apoptotic BH3 protein in plants to enquire for common signaling pathway during PCD in animal and plants. Our data showed that the active Bid protein (tBid) didn't allow us to recover any transformed plant, suggesting that the expression of this protein is not compatible with normal plant growth. Furthermore, we also enquired the effect of the ectopic expression of full length Bid protein in plants applying different stress conditions triggering PCD. The results showed that biotic and abiotic stress treatment didn't lead to any significant change in PCD activation. All together these data suggested that a proper Bid processing activity induced by applied stress is missing in plants.

Limitations
Expression in two different plants systems of the active truncated Bid protein interestingly did not allow us to recover any transgenic line on the contrary of flBid. However, despite several biotic and abiotic stress treatments applied we couldn't successfully activate the flBid protein in the plant system to finally demonstrate the presence of a plant target for the active Bid.