Collection Mémoires et thèses électroniques
Accueil À propos Nous joindre

2 Caractérisation moléculaire de deux homologues végétaux de BI-1 ayant la capacité de supprimer l’apoptose induite par Bax dans des cellules 293 humaines.

Table des matières

Ce chapitre fait l’objet d’une publication dans la revue Planta [216 (3), pp. 377-386, 2003] et constitue la version finale telle que publiée dans ce journal. Ce travail est reproduit ici avec la permission du journal. Cet article a été écrit suite à mes travaux. J’ai effectué l’ensemble des figures, à l’exception de la Figure 2-1 qui a été préparée par le second auteur. Ce dernier a également participé aux expériences ayant mené à l’élaboration de la Figure 2-3. Le troisième auteur m’a secondée pour certaines manipulations, entre autres le clonage de NtBI-1 et le marquage in situ présenté à la Figure 2-5.J’ai structuré et écrit en grande partie la première version de cet article, en étroite collaboration avec les deuxième et dernière auteurs. J’ai également préparé la version finale de l’article.

À ce jour, très peu d’homologues des régulateurs animaux de la mort cellulaire programmée (PCD) ont été identifiés chez les végétaux. Parmis ceux-ci figure Bax Inhibitor-1 (BI-1), qui possède, tout comme son équivalent humain, la capacité de supprimer la mort induite par Bax dans la levure. Puisque le rôle de BI-1 dans la régulation de la PCD végétale est encore inconnu, nous avons cloné BnBI-1 et NtBI-1 à partir de banques d’ADNc de canola ( Brassica napus L.) et de tabac ( Nicotiana tabacum L.). L’analyse des séquences déduites en acides aminés de BnBI-1 and NtBI-1 a révélé que ces protéines partagent un haut niveau d’identité avec les autres BI-1 d’origine végétale (73-95%) ou animale (26-42%). Des analyses comparatives avec les autres séquences de BI-1 disponibles a permis d’établir un modèle structural présentant sept domaines transmembranaires putatifs. De plus, la co-transfection transitoire de Bax avec BnBI-1 ou NtBI-1 dans des cellules humaines HEK ( human embryonic kidney 293) a révélé que les deux protéines pouvaient substantiellement inhiber l’apoptose induite par la surexpression de Bax. Des études de localisation ont aussi été conduites en utilisant des transformations stables dans des cellules BY-2 de tabac et de Saccharomyces cerevisiae , ou des transformations transitoires dans des feuilles de tabac, avec la protéine de fusion BnBI-1GFP sous le contrôle d’un promoteur constitutif. Tous les transformants montraient une distribution de la fluorescence typique d’une protéine du réticulum endoplasmique (RE). Des expériences de perméabilisation différentielle dans les cellules BY-2 exprimant la fusion BnBI-1GFP ont aussi montré que le C-terminus est localisé du côté cytosolique du RE. L’ensemble de nos résultats suggèrent que BI-1 serait évolutivement conservé et pourrait agir en tant que régulateur clé d’un sentier de mort commun entre les animaux et les végétaux.

Molecular characterization of two plant BI-1 homologues which suppress Bax-induced apoptosis in human 293 cells

Nathalie Bolduc 1 , Mario Ouellet 1 , Frédéric Pitre 1 and Louise F. Brisson 1

(1) Department of Biochemistry and Microbiology, Life and Health Sciences Research Building, Laval University, Québec, G1K 7P4, Canada

E-mail: louise.brisson@rsvs.ulaval.ca

Fax:1-418-6567176

To date, few homologues of animal programmed cell death (PCD) regulators have been identified in plants. Among these is the plant Bax Inhibitor-1 (BI-1) protein, which possesses, like its human counterpart, the ability to suppress Bax-induced lethality in yeast cells. As the role of BI-1 in the regulation of plant PCD remains to be elucidated, we cloned BnBI-1 and NtBI-1 from cDNA libraries of oilseed rape ( Brassica napus L.) and tobacco ( Nicotiana tabacum L.). The analysis of the deduced amino acid sequences of BnBI-1 and NtBI-1 indicated that these proteins share a relatively high level of identity with other plant BI-1 proteins (73-95%) as well as with animal BI-1 proteins (26-42%). Comparative analysis with other available plant BI-1 proteins allowed the establishment of a structural model presenting seven transmembrane domains. Moreover, transient co-transfection of Bax with BnBI-1 or NtBI-1 in human embryonic kidney 293 cells revealed that both proteins can substantially inhibit apoptosis induced by Bax overexpression. Localization studies were also conducted using stable transformation of tobacco BY-2 cells and Saccharomyces cerevisiae , or transient expression in tobacco leaves, with the fusion protein BnBI-1GFP under control of the cauliflower mosaic virus 35S promoter. All transformants showed a fluorescence pattern of distribution typical of an endoplasmic reticulum (ER) protein. Results from differential permeabilization experiments in BY-2 cells expressing BnBI-1GFP also showed that the C-terminus is located on the cytosolic side of the ER. Taken altogether, our results suggest that BI-1 is evolutionarily conserved and could act as a key regulator of a death pathway common to plants and animals.

Programmed cell death (PCD) is considered to be a vital phenomenon for multicellular organisms because of its involvement in removing unnecessary or harmful cells during normal development or under pathological conditions. In animals, cells committed to die exhibit characteristic morphological features, such as cell shrinkage, chromatin condensation, DNA and nuclear fragmentation, and formation of apoptotic bodies. Many proteins are implicated in the regulation of PCD, including the Bcl-2 family of proteins, composed of members activating or suppressing death. Their proper orchestration leads to controlled cell autolysis via the activation of effector proteins, mainly endonucleases and cysteine proteases (caspases; for a recent review, see Gallaher et al. , 2001).

In plants, it is believed that genetically controlled cell death is an integral part of development and defense response. For example, a number of biochemical and physiological studies have shown the occurrence of PCD during leaf and flower senescence, and xylogenesis, as well as during the establishment of the hypersensitive response (Greenberg, 1996; Pennell and Lamb, 1997). The existence of lesion-mimic mutants (recently reviewed in Shirasu and Schulze-Lefert, 2000), where a localized cell death is triggered due to a genetic defect, also shows the consequence of uncontrolled PCD, and resembles the inappropriate PCD occurring in several human diseases such as Alzheimer's, Parkinson's, AIDS and cancers (reviewed by Vaux and Korsmeyer, 1999). Although considerable differences in the mechanisms of PCD exist between kingdoms, it appears that many basic hallmarks of apoptosis are conserved among plants and animals (Danon et al. , 2000).

Cloning of plant homologues of the Bcl-2 family members has not been reported yet, and no such putative genes have been identified from Arabidopsis thaliana genome sequences (Aravind et al. , 2001) or from the Oryza sativa genome. However, immunological detection of Bcl-2 in tobacco (Dion et al. , 1997) and in wheat germ (Kuo et al. , 1997) suggests that some regulatory features between animal and plant death processes are similar at the molecular level. Actually, it has been reported that plants expressing animal or viral regulators of cell death (Bax, Bcl-XL, Bcl-2, CED-9 or Op-IAP) display a death or a survival phenotype depending on the nature of the transgene (Lacomme and Santa Cruz, 1999; Mitsuhara et al. , 1999; Dickman et al. , 2001; Kawai-Yamada et al. , 2001). Moreover, a plant homologue of the mammalian defender against apoptotic death 1 ( Dad1 ) has been cloned from A. thaliana and found to be as efficient as human Dad1 in rescuing mutant hamster cells from apoptosis (Gallois et al. , 1997), suggesting the conservation of an apoptosis suppression process between plants and animals.

Another mechanism of suppression conserved throughout evolution involves a regulator referred to as Bax Inhibitor (BI-1). The BI-1 gene was first isolated from a human cDNA library because of its ability to block cell death induced by the overexpression of the pro-apoptotic Bax in yeast (Xu and Reed, 1998). Plant homologues have been subsequently cloned from A. thaliana and O. sativa , and it has been shown that both plant BI-1 (pBI-1) proteins could suppress Bax-induced lethality in yeast (Kawai et al. , 1999; Sanchez et al. , 2000). The cloning of a homologue from Hordeum vulgare has also been reported (Hückelhoven et al. , 2001). Recently, it has also been demonstrated that AtBI-1 can suppress Bax-induced lethality in A. thaliana (Kawai-Yamada et al. , 2001), reinforcing the hypothesis of evolutionarily conserved cell death mechanisms between plants and animals. However, the precise role of BI-1 protein in plant cells and its mechanism of action remain to be determined. In this article, we present the cloning of two pBI-1 homologues from Brassica napus and Nicotiana tabacum , and show the high level of conservation of BI-1 proteins across the plant and animal kingdoms through computer analysis, localization studies and functional assays in animal cells.

BnBI-1 and NtBI-1 cDNA clones were obtained by cDNA library screening using a PCR approach as described by Amaravadi and King (1994). For this purpose, primers (Bn1: 5’-GATTCTCTTAAAAACTTCCGTCAGATTCT-3’ and Bn2: 5’-ATAGTCCATGTCACCGAGGTGTGC-3’) were designed to amplify a 567-bp fragment in the coding region of AtBI-1 . The rapeseed library was constructed inλZapExpress (Stratagene) from leaves of Brassica napus L. cv. Westar, and the tobacco library was constructed inλZapII from young leaves of Nicotiana tabacum L. cv. SR1 (pre-made library from Stratagene, cat. # 936002). Identity of full-length positive clones was confirmed by automatic sequencing and plasmids were named pBK-CMV-BnBI-1 and pBSK-NtBI-1. Predicted amino acid sequences were analyzed by using several programs to predict the localization (PSORT, ChloroP, TargetP, iPSORT) or the transmembrane regions and topology (SOSUI, TMpred, TMHMM, TopPred) of the cognate protein. To consolidate the different predicted transmembrane domains (TMs) into a single model, all predictions (each program for each sequence) were compared in aligned sequences, with a particular attention to plant proteins. To be included in a highly probable transmembrane domain (HPTM), a particular position had to be predicted as part of a TM for at least three of the five plant sequences by at least three algorithms out of four. An amino acid is included in a probable transmembrane region (PTM) when its position was predicted to be part of a TM in at least 20% of plant predictions.

All constructions for plant cell transformation were made in pBI121. The fusion protein BnBI-1GFP was obtained following the amplification of the open reading frame (ORF) of BnBI-1 from the vector pBK-CMV-BnBI-1 using primers identified as Bn3 and Bn4 (Bn3: 5’-AAGCTCTAGAAAACAAAAGGCATGGA-3’; Bn4: 5’-ACTTTTTGGATCCAGTTCCTCCTCCTCTT-3’) containing, respectively, an Xba I and a Bam HI site. The PCR product was then subcloned in the Xba I and Bam HI sites of pBI121 to give the pBI121-BnBI-1 plasmid. The green fluorescent protein (GFP) gene (S65T mutant) was provided as a cassette (35S-GFP-nos) introduced in a pUC vector. Primers (GFP1: 5’-ATGGATCCAAGGAGGTTCTATGGTGAGCAAGGGC-3’ and nosE: 5’-ACGGCCAGTGAATTCCCGATCTAGTAA-3’), containing, respectively, a Bam HI and an Eco RI site, were used to amplify both the entire gene encoding for the GFP and the nopaline synthase (nos) terminator. The PCR product was then subcloned in the Bam HI and Eco RI sites of pBI121-BnBI-1 or pBI121, replacing theβ-glucuronidase ( GUS ) gene and the nos terminator to produce the final constructs pBI-BnBI-1GFP and pBI-GFP, respectively. For the construction of pVT102U-BnBI-1GFP, a BnBI-1GFP fragment, obtained from pBI-BnBI-1GFP by digestion with Xba I and Pst I, was subcloned in the same sites in pVT102U (Vernet et al. , 1987), downstream of the alcohol dehydrogenase (ADH) promoter. All constructions in the pcDNA3 plasmid were done by insertion of the gene of interest into the multiple cloning site, using standard molecular biology techniques (Sambrook et al. , 1989). Bax and Bcl-2 genes were from mouse, while the GUS gene was excised from pBI121. The proper construction of all plasmids was confirmed by automatic sequencing.

Human embryonic kidney 293 cells were maintained in Iscove's modified Dulbelcco's medium (IMDM) supplemented with 10% fetal bovine serum (FBS). Cell death assays were performed essentially as described by Xu and Reed (Xu and Reed, 1998), with minor modifications. Briefly, cells were seeded (8 x 105) in 60-mm dishes in 4 ml of culture medium and transfected 24 h later by a calcium phosphate precipitation method (Sambrook et al. , 1989), with pcDNA3 plasmids encoding Bax, Bcl-2, BnBI-1, NtBI-1 or nothing (control). Transfections were performed with a total of 9 µg DNA (adjusted with control vector). Precipitates were removed 7 h later and replaced with fresh medium. Both floating and adherent cells (after trypsinization) were collected 24 hr after transfection and the percentage of dead cells was determined by trypan blue dye-exclusion assay (0.1% trypan blue in PBS), counting a minimum of 300 cells per assay. Fragmentation of nuclei associated with apoptosis was also verified by 4’,6-diamidino-2-phenylindole (DAPI) staining as described previously by Zha and co-workers (Zha et al. , 1996). Transfection efficiency was estimated to be at least 65% by co-transfection with a GUS reporter plasmid (pcDNA3-GUS). For this purpose, 1 µg of pcDNA3-GUS was included in transfection experiments and GUS activity was detected by incubating cells overnight at 37°C in the presence of 5-bromo-4-chloro-3-indolyl-β-glucuronide (X-Gluc) as described elsewhere (Gallagher, 1992). The percentage of cells presenting a blue precipitate was then determined under light microscopy.

Cell suspension cultures of N. tabacum cv. BY-2 were maintained in modified MS medium [MS salts; Murashige and Skoog, 1962), 100 mg/l myo -inositol, 255 mg/l KH2PO4, 30 g/l sucrose, 0.2 mg/l 2,4-dichlorophenoxyacetic acid, 1 mg/l thiamine HCl] in the dark, with agitation at 100 rpm. Cells were subcultured every 7 days by transferring 1.1 ml to 50 ml of fresh medium in a 250-ml Erlenmeyer flask. Stable transformation was achieved by a standard co-culture method (for 3 days in the presence of 20 µM acetosyringone) using Agrobacterium tumefaciens strain LBA4404 bearing the appropriate binary vector (pBI-GFP or pBI-BnBI-1GFP). Transformants were selected on solid growth medium supplemented with 100 µg/l kanamycin (Sigma).

Plant materials for transient expression or mRNA analysis ( N. tabacum cv. Xanthi NC and B. napus cv. Westar) were grown in a greenhouse under a 16 hr photoperiod. Transient expression in tobacco leaves was performed essentially as described by Rubino et al . (2001) with some modifications. Briefly, Agrobacterium cells were grown to late logarithmic phase, washed once in sterile infiltration medium [MS salts; (Murashige and Skoog, 1962); 10 mM Mes, pH 5.6; 20 g/l sucrose; 200 µM acetosyringone], and resuspended in the same medium to an OD600of 3.0. The bacterial preparation was pressure-infiltrated into the abaxial surface of fully expanded tobacco leaves (6-week-old plants) with a 10-ml syringe without needle. Transient expression of GFP was observed 3-6 days later by confocal microscopy.

In order to isolate BI-1 homologues from rapeseed and tobacco, a set of primers was designed to amplify a 567-bp fragment in the coding region of AtBI-1 . These primers amplified a fragment of the expected length when using whole cDNA libraries derived from leaves of rapeseed or tobacco as templates (data not shown). Sequencing of these fragments revealed high identity with AtBI-1 , suggesting that these primers could serve for the screening of the two libraries. These screenings resulted in the isolation of two full-length cDNA clones named BnBI-1 and NtBI-1 (GenBank accession numbers AF390555 and AF390556). BnBI-1 consists of 1,010 bp and predicts a 741-bp open reading frame encoding a 27.5-kDa protein of 247 amino acids. NtBI-1 consists of 1,293 bp and predicts a 747-bp open reading frame encoding a 27.6-kDa protein of 249 amino acids.

The analysis of the deduced amino acid sequences of BnBI-1 and NtBI-1 indicates that these proteins share a relatively high level of identity with other pBI-1 proteins, as well as withanimal BI-1 (aBI-1) proteins (Figure 2-1A). For example, BnBI-1 and NtBI-1 share, respectively, 93% and 75% identity with AtBI-1, while they are 66-68% identical to OsBI-1 and HvBI-1. All predicted pBI-1 proteins share 38-42% identity with the human or rat BI-1, while the identity drops to 26-28% when compared with Drosophila BI-1. Similarity between the predicted proteins spans over the entire length of the sequence, except for the N-termini, where many amino acids are lacking in the alignment. Divergence between predicted sequences is represented as an inferred tree in Figure 2-1B. It is noteworthy that plant sequences are grouped together apart from animal sequences, with a separation between monocotyledonous and dicotyledonous species, the latter being divided according to their cognate family (Solonaceae and Brassicaceae). This inferred tree is thus closely related to accepted evolutionary classification.

A ) Alignment generated with Clustal_X (Thompson et al. , 1997) and displayed using BOXSHADE (GCG package). Amino acids identical and similar to a computer-generated consensus sequence are shaded black and gray , respectively. Bars with Roman numbers show predicted plant HPTMs ( black ) and their variable ends (PTMs; halftone ). Charged residues of plant TMs are identified within the bars. B) Unrooted tree generated with Clustal_X and displayed by TreeView (Page, 1996). Branch lengths are drawn to scale and numbers indicate how many times this tree was produced out of 1,000 attempts using the bootstrap option of Clustal_X. BI-1 sequences used: Bn , Brassica napus (accession number AAK73101); Nt , Nicotiana tabacum (AAK73102); At , Arabidopsis thaliana (BAA89541); Hv , Hordeum vulgare (CAC37797); Os , Oryza sativa (BAA89540); hu , human (P55061); Rn , Rattus norvegicus (P55062); Dm , Drosophila melanogaster (Q9VSH3).

Hydropathy analysis and targeting-motif detection for both BnBI-1 and NtBI-1 predicted proteins strongly suggest their association with plant membranes. According to PSORT, iPSORT and ChloroP programs, both proteins could be located within the thylakoid membranes. On the other hand, basic residues at their C-terminal end look like some nuclear targeting sequences (Dingwall and Laskey, 1991), which were also detected by PSORT. Furthermore, it seems that BI-1 proteins display five to seven transmembrane helices (TMs), depending on the software used for the analysis. As illustrated in Figure 2-1A, we propose a model for pBI-1 composed of seven TMs, determined following the comparison of predicted TMs for the five aligned pBI-1 sequences (see Materials and methods ). These TMs are divided into highly probable (HPTM) and probable (PTM) transmembrane domains, in order to illustrate the variation in TM ends between different sequences. Almost all HPTM domains (I-VI) of this model also apply to aBI-1, while the PTM ones might be slightly different. Although predicted structural features seem to be highly conserved among different species, most programs failed to attribute TM VII to the animal sequences used in this study. It is of interest that a few charged residues are present in some TMs, particularly in the plant TM VII, which display features of an amphipathic helix (charged residues mostly aligned on the same side of theα-helix), suggesting interactions with other membrane proteins. It is worth noting that the most favored topology placed the C-termini on the cytosolic side for both plant and animal BI-1 proteins, thus placing the N-terminal end on the inner side of the membrane for pBI-1 and in the cytosol for aBI-1.

Southern blot analysis at high stringency revealed that BnBI-1 is present as a two-copy gene, while NtBI-1 is present as a single-copy gene (data not shown). Northern blot analysis (Figure 2-2) showed that mRNA levels remained relatively constant regardless of the organ, both in B. napus and N. tabacum . However, an increased accumulation was observed in senescing flowers, a situation known to involve PCD (Rubinstein, 2000). The length of messengers detected on Northern blots was estimated to 1,200 bp for BnBI-1 and 1,350 bp for NtBI-1 , values that are consistent with the length of the cloned cDNAs.

A ) BnBI-1 accumulation in leaves (1), flowers (2), senescing flowers (3) and shoots (4). B) NtBI-1 accumulation in leaves (1), flowers (2), senescing flowers (3) and cultured suspension cells (4). Ethidium bromide-stained gels with approximately equal amounts of total RNA in each lane are also shown ( lower panels ).

The functional assay developed by Xu and Reed (1998) to evaluate the ability of human BI-1 to suppress Bax-induced apoptosis in mammalian cells was used to estimate the suppressor capacity of its plant homologues. Human embryonic kidney 293 cells were transfected with an expression plasmid encoding Bax (pcDNA3-Bax) in combination with equal amounts of either pcDNA3 parental vector (used as a negative control), pcDNA3-BnBI-1, pcDNA3-NtBI-1 or pcDNA3-Bcl-2 (used as a positive control). Using this model system, expression of Bax resulted in the death of more than 30% of the cells, compared with less than 5% following transfection with parental vector (Figure 2-3A). However, when cells expressed both Bax and BnBI-1, NtBI-1 or Bcl-2, the percentage of dead cells was substantially lower, with 12, 11 and 10%, respectively, indicating that both pBI-1 homologues and Bcl-2 can suppress Bax-induced apoptosis at a similar level. Production of Bax and Bcl-2 in transfected cells was attested by immunoblot analysis (Figure 2-3A).

A ) Cells were either transfected with control vector (9 µg), or co-transfected with pcDNA3-Bax (3 µg) together with control vector (6 µg) or pcDNA3 plasmids encoding Bcl-2, BnBI-1 or NtBI-1 (6 µg). The percentages of dead cells were determined using trypan blue dye-exclusion assay ( top ). Protein extracts from transfected cells were subjected to immunodetection using antibodies specific for Bcl-2 and murine Bax proteins ( bottom ). B) Dose-dependent suppression of Bax-induced cell death by Bcl-2 and NtBI-1. Cells were either transfected with 0.1-6 µg pcDNA3-NtBI-1, or with 3 µg pcDNA3-Bax and 0.1-6 µg pcDNA3 encoding Bcl-2 or NtBI-1 (total DNA adjusted to 9 µg with control vector). C) Nuclear morphology (DAPI staining) in human 293 cells transfected with 9 µg control vector ( top ), 3 µg pcDNA3-Bax ( middle ) or 3 µg pcDNA3-Bax and 6 µg pcDNA3-NtBI-1 ( bottom ). Bar = 10 µm. Quantitative data shown ( A, B ) are means ± mean deviation of triplicate experiments.

To further investigate the dose-dependency relationship between pBI-1 concentration and suppression of Bax-induced death, cell death assays were performed with different quantities of Bcl-2- and NtBI-1-encoding plasmids (Figure 2-3B). In both cases, co-transfection of cells with 3 µg of pcDNA3-Bax and 0.1 or 1.0 µg of death suppressor-encoding plasmids did not substantially reduce the percentage of dead cells when compared with a total absence of death suppressor (evaluated to 25±1% in this experiment; data not shown). Nevertheless, appreciable reduction in the percentage of dead cells was observed when pcDNA3-Bcl-2 reached 3 µg, while 6 µg of pcDNA3-NtBI-1 was needed to inhibit cell death efficiently. It is of note that transfection of 293 cells with 0.1-6 µg of pcDNA3-NtBI-1 plasmid did not promote cell death (less than 5% cell death). Figure 2-3C shows nuclear fragmentation associated with Bax expression in 293 cells, which is only rarely observed when cells are transfected with control vector alone or co-transfected with pcDNA3-Bax (3 µg) and pcDNA3-NtBI-1 (6 µg).

The cellular localization of pBI-1 was observed by confocal microscopy using the construction pBI-BnBI-1GFP in which the GFP protein was fused to the C-terminal end of BnBI-1. GFP was observed either in stably transformed tobacco BY-2 cells cultivated in vitro or observed in planta following a transient-expression assay in tobacco leaves, and compared with cells transformed with untargeted GFP, or GFP targeted to the ER by way of a KDEL signal sequence at its C-terminus (ER-GFP; Collings et al. , 2000). The identity of organelles such as nuclei and vacuoles was verified by observing cells under Nomarski optics.

Plant cells were viewed under a confocal microscope at an excitation wavelength of 488 nm ( A, C, E, G, I-K ) and 543 nm ( J-K ), or using Nomarski optics ( B, D, F, H ). A-H) Tobacco BY-2 cells stably transformed with GFP ( A-B ), BnBI-1GFP ( C-F ) or ER-GFP ( G-H ). I-J) Transient expression of BnBI-1GFP in tobacco leaf epidermal ( I ) and palisade mesophyll ( J ) cells. In J , fluorescence emissions from chlorophyll and GFP are shown in red and green, respectively. K) Staining of mitochondria with MitoTracker Orange (red fluorescence) in BY-2 cells expressing BnBI-1GFP (green fluorescence). L-N) Yeast cells transformed with BnBI-1GFP were stained with DAPI, then examined for GFP fluorescence ( L ), DAPI fluorescence ( M ) or with Nomarski optics ( N ). Bars = 10 µm ( A-K ), 5 µm ( L-N ).

BY-2 cells expressing GFP showed diffuse fluorescence distributed throughout the cytoplasm and in the nucleoplasm (Figure 2-4A, B), as previously described by Köhler et al. (1997). In contrast, the fluorescence was mainly detected within and in the vicinity of the nuclear envelope as well as in the cytoplasm in BY-2 cells expressing BnBI-1GFP, while no fluorescence was observed in the nucleoplasm (Figure 2-4C-F). Examination of this cell line at higher magnification (Figure 2-4E, F) showed a fine fluorescent network extending into the cytoplasm, which strongly suggests an association with membranes. This subcellular distribution of the fluorescent signal fits perfectly with that observed in BY-2 cells expressing ER-GFP (Figure 2-4G, H), as well as with that reported in a number of recent studies for ER-targeted GFP (Boevink et al. , 1996;Collings et al. , 2000;Koizumi et al. , 2001;Dunoyer et al. , 2002). It is noteworthy that a strong perinuclear fluorescence was observed in nuclei isolated from BnBI-1GFP-expressing cells (data not shown), suggesting the association of BnBI-1 with the perinuclear region of the ER.

Visualization of BnBI-1GFP in tobacco leaves following transient expression showed an intracellular pattern of distribution closely related to that observed in BY-2 cells expressing BnBI-1GFP (Figure 2-4I, J). In palisade mesophyll cells, BnBI-1GFP did not co-localize with the red autofluorescence from chloroplasts (Figure 2-4J). Moreover, use of MitoTracker to visualize mitochondria in living BY-2 cells expressing BnBI-1GFP failed to show a co-localization with mitochondria (Figure 2-4K).

The fusion protein BnBI-1GFP was also expressed in yeast. Saccharomyces cerevisiae cells expressing GFP were stained with DAPI and pictures were taken under epifluorescence microscopy (Figure 2-4L-N). Comparison of the fluorescence from the GFP fusion (Figure 2-4L) with that from the DAPI-stained nuclear DNA (Figure 2-4M) showed that the protein is mainly localized in the perinuclear area, and to a lesser extent in the cytoplasm and cell periphery. This pattern of expression is typical of ER-localized proteins in yeast (Rose et al. , 1989;Preuss et al. , 1991;Hampton et al. , 1996).

To test the topological orientation of the C-terminal extremity of BnBI-1 protein in ER membranes, non-transformed and stably transformed tobacco cells were differentially permeabilized using either digitonin or Triton X-100, then processed for immunofluorescence microscopy using anti-GFP antibody. Digitonin selectively permeabilizes the plasma membrane but not internal cellular membranes, while Triton X-100 perforates both, thus allowing the selective immunodetection of cytosol-located antigens when using digitonin (Dyer and Mullen, 2001).

Control experiments with untransformed cells or transformed cells with omission of the primary antibody showed little unspecific background staining, regardless of the permeabilization procedure (data not shown). As shown in Figure 2-5A, permeabilization of cells expressing ER-GFP with Triton X-100 allowed the immunodetection of GFP in a pattern similar to the ER-GFP fluorescence. However, immunodetection was restricted to a few sites in digitonin-permeabilized cells (Figure 2-5B), possibly resulting from some leakage of ER-GFP during the cell fixation and labeling. These observations confirm that antigen sites located in the ER lumen are accessible in cells permeabilized with Triton X-100 but not with digitonin. As shown in Figure 2-5C and D, permeabilization of membranes with either Triton X-100 or digitonin in BnBI-1GFP-expressing cells resulted in similar patterns of immunodetection. In both cases, the fluorescent signals coming from the immunodetection were superposable on those coming from GFP fused to BnBI-1, indicating that labeling was highly specific. Taken together, these observations suggest that the C-termini of BnBI-1GFP would be located on the cytosolic side of the ER membrane.

Fixed cells were permeabilized using either Triton X-100 ( A, C ) or digitonin ( B, D ), labeled with anti-GFP antibody and Cy3-conjugated anti-rabbit antiserum, then viewed using confocal microscopy at an excitation wavelength of 543 nm ( left panels ). Cells were also viewed at 488 nm to see GFP fluorescence in membranes ( middle panels ). Superposition of the two images is also shown ( right panels ). A-B) ER-GFP-expressing cells. C-D) BnBI-1GFP-expressing cells.Bar = 10 µm.

From the wealth of genomic information and with the completion of the Arabidopsis and rice genome sequences, we have to notice that few animal cell-death regulators appear to have sequence homologues in plants (Aravind et al. , 2001). However, the cloning of BI-1 from Arabidopsis and rice and the discovery of their capability to successfully inhibit Bax-induced lethality in yeast (Kawai et al. , 1999; Sanchez et al. , 2000) suggested an important conservation of this protein in evolution. In this work, BI-1 clones have been isolated from rapeseed and tobacco. Comparative study of various sequences was then facilitated with these two new clones (see Figure 2-1). Actually, our sequence analysis of a number of BI-1 proteins from plant and animal kingdoms showed a high level of similarity. This similarity was particularly relevant among plant members, with the expected separation between monocotyledonous and dicotyledonous species.

Analysis of plant BI-1 sequences (Figure 2-1) revealed the presence of seven potential transmembrane domains referred to as TM I to TM VII, indicative of an integral membrane protein. The overall position of these predicted TMs in pBI-1 proteins matches those found in aBI-1, with the exception of TM VII, predicted exclusively in plants, suggesting a conservation of the protein structure in evolution. It is interesting that some TMs carry charged or polar residues, a situation giving unfavorable thermodynamic characteristics to transmembrane helices. These residues could be implicated in the consolidation of protein tertiary structure, or in the stabilization of interactions with other proteins. Moreover, the presence of charged residues aligned on the same side of the predictedα-helix VII could result in the formation of an amphipathic helix, which could be stabilized in protein-protein interactions. The presence of numerous transmembrane helices with charged and polar residues might also imply that BI-1 would function as an ion channel, as previously suggested by Xu and Reed (Xu and Reed, 1998).

It has been previously shown that both plant and animal BI-1 can suppress Bax-induced lethality in yeast (Xu and Reed, 1998;Kawai et al. , 1999; Sanchez et al. , 2000), and now we show that pBI-1, like its human counterpart (Xu and Reed, 1998), can suppress Bax-induced apoptosis in human 293 cells (Figure 2-3). Interestingly, it has been recently reported that AtBI-1 suppresses Bax-induced lethality in Arabidopsis (Kawai-Yamada et al. , 2001). Taken together with our computer analysis, these observations raise the possibility of an evolutionarily conserved cell-death pathway, where BI-1 would act as a negative regulator of death. However, Yu and co-workers (Yu et al. , 2002) recently demonstrated that AtBI-1 can induce death in human fibrosarcoma HT1080 cells, whereas our results show the opposite in human embryonic kidney 293 cells. We have also found that BnBI-1 and NtBI-1 mRNAs accumulated to a higher level in flowers undergoing senescence (Figure 2-2), and Sanchez and his co-workers (Sanchez et al. , 2000) have shown that AtBI-1 mRNA is up-regulated following wounding or pathogen attack. Moreover, a pBI-1 homologue found in a Pinus taeda xylem cDNA library (Center for Computational Genomics and Bioinformatics, University of Minnesota) was found to be a relatively abundant messenger (about 600th out of 15,000 cDNAs; John MacKay, personal communication). Up-regulation of pBI-1 in PCD situations could suggest that in some cellular contexts it would function as a pro-apoptotic protein, maybe depending on the presence or absence of specific binding partners. Accordingly, Xu and Reed (Xu and Reed, 1998) have demonstrated by in vivo cross-linking and co-immunoprecipitation studies that human BI-1 can interact with Bcl-2 and Bcl-XLbut not Bax or Bak, suggesting that BI-1 can interfere with a number of death regulators. Therefore, it is possible that potential endogenous binding partners would be different in HT1080 and 293 cells, leading to different responses when pBI-1 is expressed.

There is much evidence arguing for the integration of life and death signals in the plant mitochondrion, as in mammalian mitochondria (Jones, 2000). pBI-1, as a non-mitochondrial protein (this study; Kawai-Yamada et al. , 2001), would potentially act downstream of mitochondria. Our localization studies in stably transformed tobacco BY-2 cells indicated that, like its human counterpart (Xu and Reed, 1998), BnBI-1GFP is mostly localized to the ER and the perinuclear region (Figure 2-4). This result is in agreement with those of Kawai-Yamada and co-workers (Kawai-Yamada et al. , 2001)while expressing AtBI-1GFP. The same conclusions came from our transitory expression in tobacco leaves. Moreover, we have shown that expression of BnBI-1GFP in yeast gives a similar pattern of expression, suggesting that inhibition of Bax lethality in humans, yeast and plants would act through similar pathways.

Both our computer analysis and experimental data showed that the short extra-membranous C-termini of plant BI-1 would be located on the cytosolic side of membranes (Figure 2-5). This extremity is well conserved among plant members, suggesting a potential domain for interaction with cytosolic proteins. Actually, Yu and co-workers (Yu et al. , 2002) reported an apparent increase in cell death following transfection of HT1080 cells with a carboxyl-terminal (14 amino acids) truncated AtBI-1. On the other hand, they also mentioned unpublished results indicating that this mutant was unable to suppress Bax-induced lethality in yeast. Taken altogether, these lines of evidence let us hypothesize that the C-terminus is an important domain for pBI-1 activities. However, the N-terminus is more divergent, suggesting that no particular function is associated with this domain.

The mode of action of Bax in plants and yeast remains to be determined, but we can hypothesize that Bax fortuitously activates a death pathway that is evolutionarily conserved, involving direct or indirect interactions with conserved regulatory proteins, that would be otherwise activated under physiological or pathological necessity. Bax and some other Bcl-2 family members exhibit a pore-forming activity (reviewed in Schendel et al. , 1998), and thus it can be speculated that Bax would act in plants through pore formation in the mitochondrial outer membrane via its C-terminal transmembrane domain, which is essential for Bax-induced cell death in plants (Lacomme and Santa Cruz, 1999; Kawai-Yamada et al. , 2001). Accordingly, pBI-1, located in the ER, could not act directly on mitochondria but rather on signaling molecules or pathways activated by mitochondrial perturbations. The ER is a calcium reservoir (Sanders et al. , 1999), and considering the hypothetical ion channel activity of BI-1 it could be implicated in the up-regulation or down-regulation of cytosolic calcium. The animal mitochondrion is thought to integrate various stresses, and under pathological conditions associated with increase in cytosolic free Ca2+, cellular ATP depletion and oxidative stresses, mitochondrial Ca2+triggers opening of the mitochondrial permeability transition (MPT) pore, which lead to cytochrome  c escape (Cai et al. , 1998;Crompton, 1999;Jones, 2000). In plants, there are several cases where calcium induces cell death when ATP pools are reduced (Jones, 2000), and elevation of cytosolic free calcium is a common element of plant PCD signaling (Beers and McDowell, 2001). Further investigations regarding the role of calcium in the action of BI-1 in plants might reveal new insight into the regulation and execution of plant PCD.

Amaravadi, L., & King, M.W. (1994) A rapid and efficient, nonradioactive method for screening recombinant DNA libraries. Biotechniques 16: 98-103.

Aravind, L., Dixit, V.A., & Koonin, E.V. (2001) Apoptotic molecular machinery: vastly increased complexity in vertebrates revealed by genome comparisons. Science 291: 1279-1284.

Beers, E.P., & McDowell, J.M. (2001) Regulation and execution of programmed cell death in response to pathogens, stress and developmental cues. Curr Opin Plant Biol 4: 561-567.

Boevink, P., Santa Cruz, S., Hawes, C., Harris, N., & Oparka, K.J. (1996) Virus-mediated delivery of the green fluorescent protein to the endoplasmic reticulum of plant cells. Plant J 10: 935-941.

Cai, J., Yang, J., & Jones, D.P. (1998) Mitochondrial control of apoptosis: the role of cytochrome c . Biochim Biophys Acta 1366: 139-149.

Collings, D.A., Carter, C.N., Rink, J.C., Scott, A.C., Wyatt, S.E., & Allen, N.S. (2000) Plant nuclei can contain extensive grooves and invaginations. Plant Cell 12: 2425-2440.

Crompton, M. (1999) The mitochondrial permeability transition pore and its role in cell death. Biochem J 341: 233-249.

Danon, A., Delorme, V., Mailhac, N., & Gallois, P. (2000) Plant programmed cell death: a common way to die. Plant Physiol Biochem 38: 647-655.

Dickman, M.B., Park, Y.K., Oltersdorf, T., Li, W., Clemente, T., & French, R. (2001) Abrogation of disease development in plants expressing animal antiapoptotic genes. Proc Natl Acad Sci U S A 98: 6957-6962.

Dingwall, C., & Laskey, R.A. (1991) Nuclear targeting sequences - a consensus? Trends Biochem Sci 16: 478-481.

Dion, M., Chamberland, H., St-Michel, C., Plante, M., Darveau, A., Lafontaine, J.G., & Brisson, L.F. (1997) Detection of an homologue of bcl-2 in plant cells. Biochem Cell Biol 75: 457-461.

Dunoyer, P., Ritzenthaler, C., Hemmer, O., Michler, P., & Fritsch, C. (2002) Intracellular localization of the peanut clump virus replication complex in tobacco BY-2 protoplasts containing green fluorescent protein-labeled endoplasmic reticulum or Golgi apparatus. J Virol 76: 865-874.

Dyer, J.M., & Mullen, R.T. (2001) Immunocytological localization of two plant fatty acid desaturases in the endoplasmic reticulum. FEBS Lett 494: 44-47.

Gallagher, S.R. 1992. Gus protocols: using the gus gene as a reporter of gene expression. Academic Press, New York.

Gallaher, B.W., Hille, R., Raile, K., & Kiess, W. (2001) Apoptosis: live or die - hard work either way! Horm Metab Res 33: 511-519.

Gallois, P., Makishima, T., Hechtt, V., Despres, B., Laudie, M., Nishimoto, T., & Cooke, R. (1997) An Arabidopsis thaliana cDNA complementing a hamster apoptosis suppressor mutant. Plant J 11: 1325-1331.

Gehrig, H.H., Winter, K., Cushman, J., Borland, A., & Taybi, T. (2000) An improved RNA isolation method for succulent plant species rich in polyphenols and polysaccharides. Plant Mol Biol Rep 18: 360-376.

Greenberg, J.T. (1996) Programmed cell death: a way of life for plants. Proc Natl Acad Sci U S A 93: 12094-12097.

Hampton, R.Y., Koning, A., Wright, R., & Rine, J. (1996) In vivo examination of membrane protein localization and degradation with green fluorescent protein. Proc Natl Acad Sci U S A 93: 828-833.

Hückelhoven, R., Dechert, C., Trujillo, M., & Kogel, K.H. (2001) Differential expression of putative cell death regulator genes in near-isogenic, resistant and susceptible barley lines during interaction with the powdery mildew fungus. Plant Mol Biol 47: 739-748.

Jones, A. (2000) Does the plant mitochondrion integrate cellular stress and regulate programmed cell death? Trends Plant Sci 5: 225-230.

Kawai, M., Pan, L., Reed, J.C., & Uchimiya, H. (1999) Evolutionally conserved plant homologue of the Bax inhibitor-1 (BI-1) gene capable of suppressing Bax-induced cell death in yeast1. FEBS Lett 464: 143-147.

Kawai-Yamada, M., Jin, L., Yoshinaga, K., Hirata, A., & Uchimiya, H. (2001) Mammalian Bax-induced plant cell death can be down-regulated by overexpression of Arabidopsis Bax Inhibitor-1 (AtBI-1). Proc Natl Acad Sci U S A 98: 12295-12300.

Köhler, R., Zipfel, W., Webb, W., & Hanson, M. (1997) The green fluorescent protein as a marker to visualize plant mitochondria in vivo . Plant J 11: 613-621.

Koizumi, N., Martinez, I.M., Kimata, Y., Kohno, K., Sano, H., & Chrispeels, M.J. (2001) Molecular characterization of two Arabidopsis Ire1 homologs, endoplasmic reticulum-located transmembrane protein kinases. Plant Physiol 127: 949-962.

Kuo, W.N., McNabb, M., Kanadia, R.N., Dopson, N., & Morgan, R. (1997) The presence/absence of Bcl-2, Ca2+/calmodulin-dependant protein kinase IV, calretinin and p53 in baker's yeast and wheat germ. Cytobios 91: 7-13.

Lacomme, C., & Santa Cruz, S. (1999) Bax-induced cell death in tobacco is similar to the hypersensitive response. Proc Natl Acad Sci U S A 96: 7956-7961.

Mitsuhara, I., Malik, K.A., Miura, M., & Ohashi, Y. (1999) Animal cell-death suppressors Bcl-XLand Ced-9 inhibit cell death in tobacco plants. Curr Biol 9: 775-778.

Murashige, T., & Skoog, F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 15: 473-497.

Page, R.D. (1996) TreeView: an application to display phylogenetic trees on personal computers. Comput Appl Biosci 12: 357-358.

Pennell, R.I., & Lamb, C. (1997) Programmed cell death in plants. Plant Cell 9: 1157-1168.

Preuss, D., Mulholland, J., Kaiser, C.A., Orlean, P., Albright, C., Rose, M.D., Robbins, P.W., & Botstein, D. (1991) Structure of the yeast endoplasmic reticulum: localization of ER proteins using immunofluorescence and immunoelectron microscopy. Yeast 7: 891-911.

Rose, M.D., Misra, L.M., & Vogel, J.P. (1989) KAR2, a karyogamy gene, is the yeast homolog of the mammalian BiP/GRP78 gene. Cell 57: 1211-1221.

Rubino, L., Weber-Lotfi, F., Dietrich, A., Stussi-Garaud, C., & Russo, M. (2001) The open reading frame 1-encoded ('36K') protein of Carnation Italian ringspot virus localizes to mitochondria. J Gen Virol 82: 29-34.

Rubinstein, B. (2000) Regulation of cell death in flower petals. Plant Mol Biol 44: 303-318.

Sambrook, J., Fritsch, E.F., & Maniatis, T. 1989. Molecular cloning: a laboratory manual. 2nd. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Sanchez, P., de Torres Zabala, M., & Grant, M. (2000) AtBI-1, a plant homologue of Bax inhibitor-1, suppresses Bax-induced cell death in yeast and is rapidly upregulated during wounding and pathogen challenge. Plant J 21: 393-399.

Sanders, D., Brownlee, C., & Harper, J.F. (1999) Communicating with calcium. Plant Cell 11: 691-706.

Schendel, S.L., Montal, M., & Reed, J.C. (1998) Bcl-2 family proteins as ion-channels. Cell Death Differ 5: 372-380.

Shirasu, K., & Schulze-Lefert, P. (2000) Regulators of cell death in disease resistance. Plant Mol Biol 44: 371-385.

Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., & Higgins, D.G. (1997) The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25: 4876-4882.

Trelease, R.N., Lee, M.S., Banjoko, A., & Bunkelmann, J. (1996) C-terminal polypeptides are necessary and sufficient for in vivo targeting of transiently-expressed proteins to peroxisomes in suspension-cultured plant cells. Protoplasma 195: 156-167.

Vaux, D.L., & Korsmeyer, S.J. (1999) Cell death in development. Cell 96: 245-254.

Vernet, T., Dignard, D., & Thomas, D.Y. (1987) A family of yeast expression vectors containing the phage F1 intergenic region. Gene 52: 225-233.

Xu, Q., Jürgensmeier, J.M., & Reed, J.C. (1999) Methods of assaying Bcl-2 and Bax family proteins in yeast. Methods 17: 292-304.

Xu, Q., & Reed, J.C. (1998) Bax inhibitor-1, a mammalian apoptosis suppressor identified by functional screening in yeast. Mol Cell 1: 337-346.

Yu, L.-H., Kawai-Yamada, M., Naito, M., Watanabe, K., Reed, J.C., & Uchimiya, H. (2002) Induction of mammalian cell death by a plant Bax inhibitor. FEBS Lett 512: 308-312.

Zha, H., Fisk, H.A., Yaffe, M.P., Mahajan, N., Herman, B., & Reed, J.C. (1996) Structure-function comparisons of the proapoptotic protein Bax in yeast and mammalian cells. Mol Cell Biol 16: 6494-6508.

© Nathalie Bolduc, 2005