JBC Origene Your Gene Company

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M109459200 on December 10, 2001

J. Biol. Chem., Vol. 277, Issue 9, 7483-7492, March 1, 2002
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
277/9/7483    most recent
M109459200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhou, Y. T.
Right arrow Articles by Low, B. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhou, Y. T.
Right arrow Articles by Low, B. C.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

The BNIP-2 and Cdc42GAP Homology/Sec14p-like Domain of BNIP-Salpha Is a Novel Apoptosis-inducing Sequence*

Yi Ting ZhouDagger §, Unice J. K. SohDagger , Xun ShangDagger , Graeme R. Guy, and Boon Chuan LowDagger ||

From the Dagger  Cell Signaling and Developmental Biology Laboratory, Department of Biological Sciences, The National University of Singapore, Blk S2, 14 Science Drive 4, Singapore 117543, Republic of Singapore and the  Signal Transduction Laboratory, Institute of Molecular and Cell Biology, 30 Medical Drive, Singapore 117609, Republic of Singapore

Received for publication, October 1, 2001, and in revised form, December 12, 2001

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

We have cloned the cDNAs for two novel human proteins, designated BNIP-Salpha and beta  (for BNIP-2 Similar) that are homologous to BNIP-2, a previously known Bcl-2 and E1B-associated protein. The BNIP-S gene encodes two protein isoforms; the longer protein (BNIP-Salpha ) contains a complete BNIP-2 and Cdc42GAP Homology (BCH) domain, a novel protein domain that we recently identified, whereas its shorter variant (BNIP-Sbeta ) lacks the full BCH domain as a result of an alternative RNA splicing that introduces a nonsense intron. Primer-specific reverse-transcription PCR revealed that both BNIP-Salpha and BNIP-Sbeta mRNA are differentially expressed in various cells and tissues. The expression of BNIP-Salpha or the complete BCH domain, but not BNIP-Sbeta , causes extensive apoptosis in cells. Furthermore, BNIP-Salpha can form a homophilic complex via a unique sequence motif within its BCH domain, and deletion of this interacting motif prevents its pro-apoptotic effect. These results indicate the presence of two BNIP-S splicing variants as cellular regulators and that the BCH domain of BNIP-Salpha confers a novel apoptotic function. The significance of this is discussed.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Major cellular functions such as growth, differentiation, and death are governed by intracellular signaling that depends on molecular recognition between subsets of interacting proteins. The interaction between proteins depends on endogenous interaction sequences (domains) that orchestrate individual contacts. Although interaction domains within proteins were relatively recently discovered many are still being revealed and characterized, and few are understood in depth. These domains can either serve as protein docking/interaction sites, where multiple proteins can associate to form a functional complex or they may take part in binding to other molecules such as DNA or lipids, or be involved in enzymatic reactions (1-5). In some cases, domains act as endogenous regulatory switches, where they can interact with binding partners within the host protein and modify the tertiary conformation and function of the protein.

The recent refinement of various bioinformatic tools has allowed the rapid and reliable identification of many conserved domains. A survey of the protein domain data bases deposited in the NCBI (www.ncbi.nlm.nih.gov/) and InterPro data bases (www.ebi.ac.uk/interpro/) (6) revealed that many "domains" currently remain uncharacterized at the biochemical and molecular levels. In the post-genomic era the research emphasis is focusing on identifying the functions for the protein products of the ~30,000 genes in the human genome (7). As protein function depends on domain interactions it is essential to understand the interaction parameters, including the possible regulation, of each domain.

Apoptosis is an exquisitely controlled suicidal mechanism in cells. This process is fundamental to the growth of tissue and the development of an organism (8-11). Tissue modeling in the embryo relies on the balance between the proliferation and death of certain subsets of cells. In the adult animal if too many cells fail to die cancer can ensure, whereas excess apoptosis can lead to "degradation" disease conditions such as neurodegenerative diseases, liver failure, and myocardial infarction. Various functional domains have been deemed to be important for regulating apoptosis, as exemplified by the Bcl-2 domain (8, 12), baculoviral IAP repeats domain (13), caspase large and small subunits (14), caspase recruitment domain (15), "death" domain (16), death effector domain (17), NB-ARC domain (18), cell death-inducing DFF45-like effector N-terminal domain (19), TRAF domain (20), and BAG domain (21). Clearly additional domains that are important for apoptosis remain to be identified, and the full extent of their various interactions and regulation need to be fully characterized in order to obtain an integrated perspective of apoptosis.

One of the protein domains that our group first identified and characterized was the BCH1 domain, for BNIP-2 and Cdc42GAP Homology (22-24). This novel sequence contains about 145 amino acids and was initially shown to be common to two proteins: BNIP-2 and Cdc42GAP. BNIP-2 was originally shown to be an interacting protein for both the viral E1B p19KDa protein and the Bcl-2 anti-apoptotic protein (25), whereas Cdc42GAP (also known as p50-RhoGAP) is a GTPase-activating protein that down-regulates the small GTPase Cdc42 (26, 27). The BCH domain of BNIP-2 acts as a non-canonical GTPase-activating protein domain for Cdc42, which involves a novel arginine finger (23). The BCH domain of BNIP-2 also forms both homophilic associations with itself and a heterophilic complex with Cdc42GAP. In each case the association negatively regulates the associated GTPase-activating protein activity (24).

To further characterize the BCH domain and its potential role in cell signaling we set out to characterize the domain within the context of novel proteins that contain this sequence. Here we report the identification and characterization of two novel alternatively spliced isoforms of proteins that are similar to BNIP-2, termed BNIP-S (BNIP-2-Similar) and showed that the BCH domain of the BNIP-Salpha isoform confers its apoptotic activity via a motif that is also required for its homophilic interaction.

    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Bioinformatics-- To search for novel proteins that contain BCH domains, the peptide sequence of either full-length BNIP-2 or the BCH domain (residues 167-303) was used to carry out BLAST searches using the "tblastn" and "position-specific iterative BLAST" method against the current non-redundant sequence data base as well as human and mouse EST data bases (www.ncbi.nlm.nih.gov). All BCH domains were identified without iteration using unfiltered query sequences. One putative member was initially identified by the accession number AF193056 with several human ESTs (AI620433, W80385) covering sequences of its putative open reading frame. Subsequently, a mouse homolog was identified in the data base (BC005659). This gene was termed BNIP-S, for BNIP-2-Similar. The multiple sequence alignment was generated using ClustalW (28) (www2.ebi.ac.uk/clustalw/).

RT-PCR Cloning of BNIP-S Isoforms and Plasmid Constructions-- To obtain the full-length cDNA of BNIP-S, total RNA was isolated from 293T cells using the RNeasy (Qiagen) according to the manufacturer's instructions. 5 µg of this RNA was subjected to the first-strand cDNA synthesis with Expand Reverse Transcriptase (Roche Molecular Biochemicals) primed with oligo(dT) (Operon) for 60 min at 42 °C in a total volume of 20 µl. 0.5 µl of this cDNA was then amplified by the high fidelity, long-template Taq polymerase enzyme (Roche Molecular Biochemicals) using specific primers corresponding to either the full-length BNIP-S or various fragments of it. PCR conditions were: initial denaturation 94 °C, 2 min; subsequent cycling (30 cycles) at 94 °C, 10 s; annealing at 60 °C, 30 s; extension at 68 °C, 2 min; and final extension at 68 °C, 7 min. These PCR primers contained BamHI and XhoI restriction sites on the forward and reverse primers, respectively, to facilitate their subsequent cloning.

The full-length PCR products were gel-purified (Qiagen) and cloned into a FLAG epitope-tagged expression vector, pXJ40 (Dr. E. Manser, Institute of Molecular and Cell Biology, Singapore). cDNA fragments encoding the various domains of BNIP-S were generated from the full-length template using specific primers in a standard PCR and then gel purified for cloning. Deletion mutants of BNIP-Salpha were generated by PCR using specific primers facilitated by restriction sites. For each construct, several clones were chosen and sequenced entirely in both directions using the ABI PRISM BigDye Terminator Cycle Sequencing kit (Applied Biosystem). A GenomeWalker kit (CLONTECH) was used to amplify the corresponding gene fragment of BNIP-S and identify the region of insert as an intron. All plasmids were purified using Qiagen miniprep kit for subsequent use in transfection experiments. Escherichia coli strain DH5alpha was used as host for the propagation of the clones. Reagents used were of analytical grade, and standard protocols for molecular manipulations and media preparation were as described (29).

Semiquantitative RT-PCR-- To distinguish the mRNA expression level of BNIP-Salpha , BNIP-Sbeta , and BNIP-2 in various cells and tissues, RT-PCR using the full-length BNIP-S or BNIP-2 gene primers was employed. Total RNA was isolated using the RNeasy kit (Qiagen) from either various cultured cell lines or from various organs obtained from a 2-week-old male mouse and primed for the first-strand cDNA synthesis as described above. Equal amounts of the reverse transcription product was then subjected to PCR amplification for BNIP-Salpha and BNIP-Sbeta . The house-keeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used to normalize the level of expression. The results were verified in at least two independent experiments with varying numbers of PCR cycles to ensure near-linear amplification. To further validate the weak expression of BNIP-S in mouse tissues/organs, PCR products were hybridized with a gene-specific probe, and signals were detected using the CSPD® chemiluminescence-based Southern blot analyses system according to the instructions of manufacturer (Roche Molecular Biochemicals).

Cell Culture and Transfection-- Human MCF-7, 293T, MCN45, and KMN74 were grown in RPMI 1640 medium (Hyclone) supplemented with 10% (v/v) fetal bovine serum (Hyclone), 2 mM L-glutamine, 100 units/ml penicillin, and 100 µg/ml streptomycin (all from Sigma), and maintained at 37 °C in a 5% CO2 atmosphere. HeLa cells were grown in Dulbecco's modified Eagle's medium high glucose, whereas HT29 and HCT116 were grown in McCoy's medium (Sigma). Cells at 90% confluence in 100-mm plates or 6-well plates were transfected with 2 µg or 0.4 µg of the indicated plasmids using Effectene cationic lipids, according to the manufacturer's instructions (Qiagen).

Precipitation/"Pull-down" Studies and Western Blot Analyses-- Control cells or cells transfected with expression plasmids were lysed in 1 ml of lysis buffer (50 mM HEPES, pH 7.4, 150 mM sodium chloride, 1.5 mM magnesium chloride, 5 mM EGTA, 10% (v/v) glycerol, 1% (v/v) Triton X-100, a mixture of protease inhibitors (Roche Molecular Biochemicals), and 5 mM sodium orthovanadate). The lysates that were directly analyzed, either as whole-cell lysates (25 µg) or aliquots (500 µg), were used in affinity precipitation/pulldown experiments with various GST fusion proteins (5 µg), as previously described (23, 24). Samples were run in SDS/PAGE gels and analyzed by Western blotting with FLAG monoclonal antibody (Sigma).

Immunofluorescence-- Cells were seeded on coverslips in 6-well plates, transfected with various expression constructs for 16 h, and then stained for immunofluorescence detection using confocal fluorescence microscopy as previously described (30). FLAG-tagged proteins were detected with monoclonal anti-FLAG followed by Texas Red® dye-conjugated goat anti-mouse IgG (Jackson ImmunoResearch). For cells expressing GFP-tagged recombinants, the morphology of cells was examined directly under a fluorescent microscope at the times indicated after the transfection.

beta -Galactosidase Assays-- Cells grown in 35-mm plates to 70% confluence were transiently transfected with 0.4 µg of either vector control or expression plasmids for full-length BNIP-Salpha , BNIP-Sbeta , or their various deletion domains, together with 0.1 µg of pCMV-beta -galactosidase as the marker. These cells were transfected with the marker at one-fourth the level of test cDNAs such that every cell that expressed beta -galactosidase and subsequently stained blue should also express the query protein. After 16 h, cells were fixed, incubated in 5-bromo-4-chloro-3-indolyl beta -D-galactopyranoside buffer to mark the beta -galactosidase expressing cells, and visualized by phase contrast light microscopy (31). The percentages of apoptotic cells were assessed from 6-9 random fields containing at least 300 cells and were scored positive if rounded and stained dark blue. At least three separate experiments were performed with each analysis done in duplicate.

Annexin V Binding and Nuclear Staining Assays-- Exposure of phosphatidylserine on outer membrane is a key sign of apoptosis, which can be readily recognized by annexin V. Cells grown in 60-mm plates were transfected with 2 µg of vector or expression constructs of BNIP-Salpha for 6 h. They were immediately assayed for annexin V binding and propidium iodide staining using the ApoAlert annexin V-EGFP apoptosis kit (CLONTECH) according to the manufacturer's instruction and examined by microscopy. Cells that bound annexin V but did not take up propidium iodide underwent early apoptosis, whereas cells that were positive for both markers indicated late apoptosis. Live cells were negative for both staining, and necrotic cells were only stained by propidium iodide.

To stain for chromatin integrity, cells transfected with 2 µg of vector or expression constructs of BNIP-Salpha were incubated for 10 min with 10 µg/ml of Hoechst dye (Sigma), washed with phosphate-buffered saline, and examined by microscopy. Cells undergoing apoptosis exhibited an irregular pattern of staining with Hoechst dye compared with homogenous staining in live cells (32).

    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Identifying and Cloning Novel Proteins Containing BCH Domains-- To isolate putative clones encoding proteins that bear a similar BCH domain to BNIP-2, we performed PSI-BLAST by using either the full-length sequence of BNIP-2 or its BCH domain against the non-redundant, as well as human and mouse EST data bases of GenBankTM. Our searches revealed several novel protein/DNA sequences that harbor or encode BCH domains with varying degrees of homology. One clone was initially identified (accession number AF193056) with several human ESTs (AI620433, W80385) spanning parts of its putative open reading frame. Subsequently, a mouse homolog was also identified in the data base (BC005659). This cDNA was termed BNIP-S, for BNIP-2-Similar.

The open reading frame and parts of the 5'- and 3'-untranslated regions of the sequence AF193056 was amplified by RT-PCR, as shown in Fig. 1A. Several pairs of primers, indicated by the arrows, were designed against the open reading frame in order to isolate the full-length or partial cDNA from human 293T cells. To maximize the chance of obtaining the cDNA, total RNA was isolated from 293T cells grown in media with or without 10% serum, pooled and subjected to RT-PCR as described under "Materials and Methods." The PCR products were resolved in 1% agarose gel, stained, and visualized (Fig. 1B). The results show that all primer pairs generated various DNA fragments of the predicted sizes. However, when the full-length primer pairs (lane 1) or primer pairs that covered specifically the 3'-end of the open reading frame (including the new BCH domain) were used (lanes 4 and 5), there was an unexpected, but distinct additional fragment. No bands were detectable in samples without reverse transcriptase, confirming that these products were not derived from genomic DNA (data not shown). The result implies that there were two transcripts of BNIP-S in 293T cells with the larger fragment carrying a longer 3'-end at its open reading frame. To verify this possibility, the predicted 825-bp (full-length fragment-alpha ) and the additional longer fragment (full-length fragment-beta ) cDNAs were separated and gel purified, and multiple clones were then isolated and sequenced for direct identification.


View larger version (50K):
[in this window]
[in a new window]
  		Fig. 1.   Cloning of BNIP-S variants. A, depicted is the BNIP-S cDNA sequence obtained by primer-specific RT-PCR cloning from 293T cells based on the previously deposited entry AF193056. The first methionine residue was identified downstream of an in-frame stop codon (*) and fits well with the Kozak consensus for translational initiation site as predicted (43) (hri.co.jp/atgr/). Triangular arrow indicates the site where an insertion was identified for the novel BNIP-Sbeta splicing variant as shown in D. B, RT-PCR was performed using sets of primers as indicated by forward (F) and reverse (R) arrows in A and revealed two isoforms of cDNA that differ in size at their 3'-ends. Lane 1: primer set 1F/3R; lane 2: 1F/1R; lane 3: 1F/2R; lane 4: 3F/3R; lane 5: 2F/3R. C, genomic DNA amplification with these primers revealed only one type of DNA product. Lane 1: primer set 1F/3R; lane 2: 1F/2R; lane 3: 2F/3R; lane 4: 3F/3R. D, sequence of the insert in beta -fragment of BNIP-S cDNA and genomic PCR products are underlined, while the consensus site for splicing (XGT ... AG) is indicated by upper case. Introduction of an in-frame stop codon in the insert (intron) truncates the protein into BNIP-Sbeta isoform.

Two Novel Isoforms of BNIP-S Are Generated by Alternative RNA Splicing-- Sequencing of both cDNA fragments revealed that there was an insertion of 154 bp in the larger transcript within the sequence encoding the BCH domain, but the rest of the sequences were essentially identical. This raised the possibility that such an insertion may have occurred by means of an alternative RNA splicing. To test this hypothesis, the same sets of full-length PCR primers were used to amplify human genomic DNA. In all cases, only one size of fragment was obtained (Fig. 1C). The full-length fragment (lane 1) was subsequently cloned, sequenced and shown to be identical to the fragment-beta obtained by RT-PCR. This result confirms that fragment-beta (BNIP-Sbeta ) was derived by the retention of the 154-bp intron during the process of mRNA maturation, whereas this intron was removed in fragment-alpha (BNIP-Salpha ). Consistent with the notion of intron retention is the finding that there is the sequence (XGT ... AG) at the junction of the 154-bp insertion (Fig. 1D). This sequence conforms to the consensus signature for exon/intron splice site in the eukaryotic system (33).

Overall, BNIP-Salpha is 72% homologous (46% identity) to BNIP-2 with extensive homology (85%) in the region encompassing the BCH domain but with lower homology at the N terminus (Fig. 2A). While the longer protein (275 amino acids; termed BNIP-Salpha ) contains the complete BCH domain, the 154-bp insertion of the intron in BNIP-Sbeta cDNA introduced an in-frame stop codon in its unspliced mRNA resulting in a truncation within the BCH domain, hence a shorter peptide (204 amino acids). When compared with the mouse sequence, the human BNIP-Salpha is shorter by 53 amino acids at the N terminus. Interestingly, this additional region in mouse BNIP-Salpha is homologous to the N terminus of BNIP-2. The BCH domain of BNIP-Salpha is predicted to comprise several alpha -helices, and like those of BNIP-2 and Cdc42GAP, it also shares about 25% homology to the phospholipid-binding domain (Sec-14p) of the yeast phosphatidylinositol transfer protein, Sec-14 (34) (Fig. 2B). This raises some interesting issues on the potential structural and functional relationship between these two protein domains, as will be discussed later.


View larger version (41K):
[in this window]
[in a new window]
  		Fig. 2.   Multiple sequence alignmentof BNIP-S isoforms with BNIP-2. A, protein sequences of BNIP-S isoforms were aligned with BNIP-2 using the ClustalW method with blossum matrix 60 (www2.ebi.ac.uk/clustalw/) that revealed regions of high homology and unique regions. Amino acids were color-coded to represent basic (magenta), acidic (blue), small or hydrophobic (red), and hydroxyl or amine (green). Regions delineated by arrows indicate conserved BCH domains. The following symbols are used: (*) denotes identical residues; (:) denotes highly conserved residues, and (.) represents lower but significant conservation among all members. The alpha -helical secondary structures were predicted based on a consensus secondary structure prediction by Jpred (jura.ebi.ac.uk:8888/) and are indicated as bars above the sequences, while regions R, S, and T of BNIP-Salpha used in subsequent deletional studies are indicated by double-underlines. B, alignment of regions in BNIP-Salpha , BNIP-2, and Cdc42GAP with the Sec-14p lipid-binding domain. Arrows indicate the boundary of the BCH domain, while the symbols under the residues denote degree of homology among these proteins, as used for the alignment in A. GenBankTM accession numbers used for the alignments are human BNIP-2 (U15173), mouse BNIP-Salpha (BC005659), human Cdc42GAP (Z23024), and yeast Sec-14p (NP_013796).

Differential Expression of BNIP-S Isoforms in Cell Lines and Mouse Tissues-- To understand the potential roles of the BNIP-Salpha and BNIP-Sbeta , we examined their relative pattern of expression in various established cell lines from different tissue origins as well as in mouse tissues/organs. Since BNIP-Salpha and BNIP-Sbeta mRNA vary by the relatively short 154-bp insertion, their relative abundance would not be easily distinguishable by conventional Northern analyses. Furthermore, their relatively high degree of homology with BNIP-2 could result in cross-hybridization of the BNIP-S probes to the mRNA of BNIP-2 or other potentially yet undiscovered homologs. To avoid such confusion, total RNA were isolated from target cells, reverse-transcribed to cDNA, and amplified by PCR using primers that were only specific to BNIP-Salpha , BNIP-Sbeta , or BNIP-2. Results in Fig. 3A show that BNIP-Salpha mRNA was predominantly detected in human breast cancer MCF-7 (lane 2) and monkey kidney fibroblast COS-1 cells (lane 4), whereas the expression of BNIP-Sbeta in these two cell lines was very low. In contrast, there were relatively high levels of BNIP-Sbeta mRNA in human HeLa cervical epithelial cells (lane 1) and human 293T embryonic transformed kidney epithelial cells (lane 3). Furthermore, two human stomach cancer lines, MCN45 and KMN74, and two human colon epithelial cell lines, HT29 and HCT116, expressed low amounts of BNIP-Salpha mRNA but no detectable BNIP-Sbeta (lanes 5-8). In contrast to the differential expression of the BNIP-S isoforms, BNIP-2 mRNA was ubiquitously expressed in all cell lines tested. Equal expression of the house-keeping GAPDH confirms the equal loading of samples for the RT-PCR.


View larger version (32K):
[in this window]
[in a new window]
  		Fig. 3.   Differential expression of BNIP-Salpha and BNIP-Sbeta mRNA. A, various cells were grown in appropriate medium containing 10% serum, and total RNA were isolated for semi-quantitative RT-PCR for the detection of expression levels of BNIP-S isoforms, BNIP-2, and GAPDH. Results were representative of at least two separate determinations. Lanes 1: HeLa; lane 2: MCF-7; lane 3: 293T; lane 4: COS1; lane 5: MCN45; lane 6: KMN74; lane 7: HT29, lane 8: HCT116. B, different organs of a 2-week-old male mouse were obtained, and their total RNA isolated for RT-PCR to detect the expression level of BNIP-S isoforms, BNIP-2, and GAPDH. The full-length PCR product of BNIP-Salpha was subjected to internal amplification using CBCH region-specific primers (second panel). C, PCR products (from B, top panel) was subjected to Southern blot analyses using BNIP-Salpha gene-specific cDNA probe. Lane 1: liver; lane 2: lung; lane 3: kidney, lane 4: brain; lane 5: intestine; lane 6: penis, lane 7: heart; lane 8: testis.

Next, the expression levels of BNIP-Salpha and BNIP-Sbeta were examined in different organs of a 2-week-old male mouse by utilizing similar RT-PCR protocols (Fig. 3B). Intriguingly, all tissues examined such as liver, lung, kidney, brain, intestine, heart, and testis (lanes 1-5, 7-8; top panel) exhibited weak signals corresponding to BNIP-Salpha but not BNIP-Sbeta , and the only organ showing elevated levels of BNIP-Salpha expression was the penis (lane 6). To further validate that these weak bands observed were truly BNIP-Salpha , the first PCR products were used as templates for amplification with internal PCR primers corresponding to region CBCH (amino acids 91-275; see Fig. 6A). The products would better distinguish the BNIP-Salpha (554 bp) and BNIP-Sbeta (706 bp). The results (second panel) confirmed the initial finding, which was further supported by Southern blot analyses using a gene-specific probe (Fig. 3C) and partial sequencing of the mouse cDNA from the liver. In comparison, BNIP-2 was relatively highly expressed across all tissues and organs, whereas GAPDH expression levels in those samples were identical.

BNIP-Salpha and BNIP-Sbeta Exert Different Effects on Cell Growth and Morphology-- There is no immediate explanation for the pattern of expression for the two isoforms of BNIP-S. Their apparent differential distribution in various cells and tissues may suggest a physiologically and developmentally regulated expression pattern and indicate that they may also play different roles in the cells. As a first stage in investigating their potential physiological function(s), BNIP-Salpha and BNIP-Sbeta were overexpressed in MCF-7 cells to investigate their intracellular distribution and any effects they may have on cell growth and morphology. MCF-7 cells were transfected for 16 h with FLAG-tagged full-length cDNAs for BNIP-Salpha and BNIP-Sbeta and processed for analysis by confocal microscopy (Fig. 4A, left panels). The result shows that 70-80% of the cells transfected with full-length BNIP-Salpha exhibited a dramatic rounding, where BNIP-Salpha itself was found mainly localized in the cytosol. In addition, some of these cells had elongated extensions that resembled "beads-on-a thread" structure (indicated by arrows) where BNIP-Salpha was also concentrated. In contrast, all cells transfected with BNIP-Sbeta exhibited similar morphology as the control cells (transfected with vector alone) but with most of the protein being localized in the nucleus (indicated by arrows). Similar observations were made using direct labeling of transfected cells utilizing a similar protocol except for GFP being conjugated to the target proteins (Fig. 4A, right panels). The cells over-expressing BNIP-Salpha again displayed drastic rounding, one of the common features of cells undergoing apoptosis, whereas BNIP-Sbeta (localized to the nucleus; indicated by arrow) had no effect on the cell morphology.


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 4.   BNIP-Salpha , but not BNIP-Sbeta , induces extensive morphological changes and cell death. A, MCF-7 cells were transfected with FLAG-tagged or GFP-tagged expression plasmids for BNIP-Salpha or BNIP-Sbeta or their corresponding vectors alone. Cells were processed, and the proteins detected by indirect confocal immunofluorescence (IF) or directly detected for GFP proteins. B, MCF cells were transfected with FLAG-tagged BNIP-Salpha or vector alone and analyzed for binding of annexin V to the external cell membrane. C, MCF-7 cells transfected with GFP expression vector alone or GFP expression plasmids for BNIP-Salpha were visualized for GFP expression, and cells were stained with Hoechst dye for nuclei integrity. D, MCF-7 cells transfected with FLAG-tagged expression vector or FLAG expression plasmids for BNIP-Salpha together with beta -galactosidase as marker. Cells were processed for beta -galactosidase assay and visualized to assess morphological changes. Arrows indicate features that are described in the text.

To further investigate the notion that BNIP-Salpha could indeed cause apoptosis, several other biochemical and morphological assays indicative of this process were also performed as described under "Materials and Methods." First, annexin V binding for phosphatidylserine on the outer membrane and propidium iodide staining for nuclei showed that cells transfected with FLAG-tagged BNIP-Salpha exhibited extensive annexin V binding without significant propidium iodide staining. This indicates an early apoptosis event where phosphatidylserine translocated to the outer membrane while cells were relatively still intact (Fig. 4B). Second, when cells were transfected with GFP-BNIP-Salpha , there were irregular chromatin staining patterns with Hoechst DNA dye when compared with the homogenous nuclei staining in GFP vector controls (Fig. 4C; arrows indicate transfected cells). Third, co-transfection of BNIP-Salpha with a beta -galactosidase reporter gene also revealed features of cell-rounding, shrinkage and apparent detachment (Fig. 4D). To ensure that these effects were not cell type-specific the experiments were repeated in COS-1 cells (monkey fibroblast), HeLa (human cervical cancer epithelial) and neuro2A (mouse neuronal) cells. In all cases similar results were obtained (data not shown).

To further explore the dynamics of the apoptotic activity induced by BNIP-Salpha , MCF-7 cells were transfected with GFP-BNIP-Salpha and were microscopically observed for changes in cell shape at 2-h intervals over a period of 16 h (Fig. 5). At the first time point (6 h post-transfection), cells expressing BNIP-Salpha exhibited extensive elongation and thinning in regions where the protein was predominantly localized. Subsequent observations revealed the retraction and collapse of this extension into a smaller beads-on-a thread structure. Subsequent to these extensions, the main mass of the cell was progressively becoming rounder and more compact. These morphological changes were not apparent in cells expressing GFP control (data not shown).


View larger version (35K):
[in this window]
[in a new window]
  		Fig. 5.   Dynamic changes of cell morphology induced by BNIP-Salpha . MCF-7 cells were transfected with a GFP-tagged expression plasmid of BNIP-Salpha and visualized for morphological changes over the time course indicated. Arrows indicate loci for dynamic changes in cells expressing BNIP-Salpha . No such changes were noted in control cells (data not shown).

Taken together, these results are consistent with the notion that the onset of BNIP-Salpha expression in the cell causes apoptosis. Since the BNIP-Sbeta protein lacks the complete BCH domain and that it does not cause apoptosis, it appeared that the BCH domain itself may be instrumental in the induction of the apoptotic process.

The BCH Domain Mediates BNIP-Salpha -induced Apoptosis-- To further investigate whether the BCH domain mediates BNIP-Salpha -induced apoptosis, full-length FLAG-tagged BNIP-Salpha , BNIP-Sbeta , or various deletion mutants that bear or lack the BCH domain were constructed (Fig. 6A) and expressed in MCF-7 for a series of experiments. The levels of expression were assessed by Western analysis using anti-FLAG. Fig. 6B shows that all constructs expressed intact proteins of relatively similar levels except the BCH domain, the level of which was always consistently much lower than the others (about 10%). Its detection by Western analysis required overexposure of the film (data not shown).


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 6.   The BCH domain of BNIP-Salpha is a pro-apoptotic protein domain. A, various constructs were made from BNIP-S based on predicted structural parameters derived from their known homology to BNIP-2 protein (see Fig. 2A). B, MCF-7 cells were transfected with FLAG-tagged versions of the domains depicted in Fig. 6A analyzed by Western analyses using FLAG antibodies. The signal for the BCH domain needs prolonged exposure, thus is not included here. Lanes 1, 2 and 8: unrelated constructs as controls; lane 3: full-length BNIP-Salpha , lane 4; full-length BNIP-Sbeta , lane 5: N domain, lane 6: NBCH domain, lane 7: CBCH domain. C, MCF-7 cells were transfected with the constructs depicted in A and analyzed for their ability to cause cell death by measurement of cell rounding utilizing the standard beta -galactosidase marker cotransfection methodology. The magnitude of the apoptotic activity was measured against the vector control and represented as a bar graph. Results are averages ± half the ranges for two determinations that are representative of at least three separate experiments.

Next, MCF-7 cells were transfected for 16 h with the described expression plasmids together with beta -galactosidase cDNA as a marker for apoptosis-induced cell rounding and shrinkage. This assay system was chosen for their relative simplicity and efficacy (32). Cells were subsequently examined for viability and changes in cell morphology by phase contrast microscopy. The total number of rounded, detaching, dense/dark-blue cells was scored as positive for apoptosis as opposed to "healthy" cells (flat, cuboidal, and attached) and was expressed as a percentage of total cells that were stained blue. Fig. 6C indicates that nearly 70% of the MCF-7 cells that expressed BNIP-Salpha underwent substantial cell rounding with the occasional thinning and elongation of some cells (similar to those described above). In contrast, cells transfected with control vector alone or BNIP-Sbeta remained cuboidal and attached. Consistent with the hypothesis that the BCH domain of BNIP-Salpha was able to induce apoptosis, expression of the fragments designated CBCH (that includes BCH domain) and BCH domain itself also induced potent apoptotic response, whereas expression of NBCH (that lacks the BCH domain) failed to exert any apoptotic effects.

Homophilic Binding via the BCH Domain of BNIP-S Correlates with Its Apoptotic Activity-- The pro-apoptotic activity of the BNIP-Salpha BCH domain directly indicates that this module can act autonomously. We have previously shown that the homologous BCH domains of BNIP-2 and Cdc42GAP can interact in homophilic and heterophilic contexts (24). We asked whether the BCH domain of BNIP-Salpha could likewise function as a protein-protein interaction domain and if such an interaction might play a key role in regulating the observed pro-apoptotic activity. GST recombinants of various domains of BNIP-Salpha were constructed (same to those depicted in Fig. 6A) and tested for their ability to bind the FLAG-tagged proteins of either itself (BNIP-Salpha ), BNIP-Sbeta , BNIP-2, or Cdc42GAP that were expressed in mammalian cells. An expression assessment blot is shown in Fig. 7A. Fig. 7B and C show that the full-length BNIP-Salpha , the CBCH fragment (which contains the BCH domain), and the BCH domain alone interacted strongly with BNIP-Salpha , BNIP-2, and Cdc42GAP but not with BNIP-Sbeta . On the other hand, fragments N and NBCH of BNIP-Salpha , which were devoid of the BCH domain, failed to bind to the test targets. These results confirm that the BCH domain of BNIP-Salpha also mediates homophilic and heterophilic interactions. Furthermore, based on the lack of binding and apoptotic activity seen with BNIP-Sbeta it may be assumed that the homophilic interaction of BNIP-Salpha may be necessary for its pro-apoptotic effect. To test this possibility we required a mutant form of BNIP-Salpha that lacked homophilic binding capacity. To this end, we took advantage of the failure of BNIP-Sbeta to mediate both binding and apoptotic activity and deduced that the interaction/apoptotic sequence lies in the "missing" sequence of the BCH domain (i.e. amino acids 204-275). We carried out a limited deletion of several amino acids in this sequence (region R) plus two other regions that are unique to BNIP-Salpha ; regions S and T (as indicated in Fig. 2A). These full-length mutants were expressed as FLAG-tagged constructs, and the resulting cell lysates were either analyzed by Western blotting or subject to pulldown experiments that utilized GST-BNIP-Salpha . The mutant proteins were expressed equally well and at similar levels to the wild type protein (Fig. 8A). In the pulldown experiments, the wild type, deletion-R and deletion-T mutants retained an ability to bind to the full-length BNIP-Salpha , whereas the deletion-S mutant demonstrated poor binding activity.


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 7.   The BCH domain of BNIP-S mediates homophilic and heterophilic interaction. GST-recombinants of various BNIP-S domains depicted in Fig. 6A were prepared as agarose beads and used for pulldown binding assays using cell lysates expressing FLAG-tagged proteins of BNIP-Salpha , BNIP-Sbeta , BNIP-2, or Cdc42GAP as the targets. A, cells expressing the FLAG-tagged proteins were analyzed for their relative expression levels by Western blot. Lane 1: BNIP-2; lane 2: Cdc42GAP; lane 3: BNIP-Salpha ; lane 4: BNIP-Sbeta . B and C, beads from the pulldown experiments were washed and processed for Western analyses using FLAG antibodies. Blots were stripped and stained with amido black to reveal equal loading of GST-recombinants. GST constructs used for B were: lane 1: BNIP-Salpha ; lane 2: BNIP-Sbeta ; lane 3: BNIP-2; lane 4: Cdc42GAP; lane 5: BNIP-2-BCH; lane 6: Cdc42GAP-BCH; lane 7: BNIP-Salpha -BCH, and lane 8: BNIP-Salpha -N. GST constructs used for C were: lane 1: BNIP-Salpha ; lane 2: BNIP-Sbeta ; lane 3: BNIP-2-BCH; lane 4: Cdc42GAP-BCH; lane 5: BNIP-Salpha -BCH; lane 6: BNIP-Salpha -CBCH; lane 7: BNIP-Salpha -N; lane 8: BNIP-Salpha -NBCH.


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 8.   A, identification of the binding motif in the BNIP-Salpha -BCH domain. Various deletion mutants of FLAG-tagged BNIP-Salpha were generated at the regions indicated in Fig. 2A and tested for their ability to affect protein interaction with GST-recombinants of BNIP-Salpha in the pull-down assays. GST beads were washed, and the pulldown products analyzed on SDS/PAGE and detected with FLAG antibodies. The blot was stripped and probed with GST antibodies to check for equal loading of beads. B, the binding motif within the BNIP-Salpha BCH is necessary for its pro-apoptotic activity. MCF-7 cells were transfected with FLAG-tagged expression constructs of either the wild type or deletion mutants at regions R, S, or T together with the beta -galactosidase marker and analyzed for their apoptotic status, as described above. Results are averages ± half the ranges for two determinations and are representative of at least three separate experiments.

The identification of an important sequence in the binding motif led us to examine whether this mutation had any impact on the apoptotic activity of the BNIP-Salpha BCH domain. MCF-7 cells were transfected with the mutants together with beta -galactosidase cDNA, and cell death assays were subsequently performed as before. It was observed that the deletion-S mutant (which did not form homophilic complex) failed to exert pro-apoptotic activity, whereas deletion mutants of region R and T were just as potent as the wild type (Fig. 8B). These findings suggest that the homophilic interaction of BNIP-Salpha via the BCH domains coincides with its apoptotic effects.

Taken together, our results provide evidence that the BCH domain in BNIP-Salpha constitutes a novel functional unit that potently induces apoptosis in eukaryotic cells. The controlled expression of the apoptotic BNIP-Salpha and its non-apoptotic splicing variant, BNIP-Sbeta , in various tissues could therefore play important roles in regulating cell growth and development.

    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

From a bioinformatics and "data-mining" starting point, we have identified and cloned the cDNAs for various proteins that contain the highly conserved BCH domain. In this work we report the cloning and characterization of two novel isoforms that are homologous to BNIP-2. The BNIP-Salpha isoform contains a complete BCH domain that mediates its pro-apoptotic effect, whereas the alternatively spliced isoform BNIP-Sbeta that lacks such domain does not. These two proteins may differentially regulate cell death in specific cell types or tissues.

The presence of modular protein domain(s) is a common feature in cell signaling pathways, including those that are involved in regulating apoptosis. Since apoptosis plays an important role in the development and maintenance of an organism it is necessary to identify domains that orchestrate this complex process. A number of these domains have been described. Our current data indicate that the BCH domain of BNIP-Salpha represents an addition to the list.

The identification of the BCH sequence as an apoptosis-inducing domain highlights some important implications. Firstly, the pro-apoptotic activity of the BCH domain coincides with its ability to form a homophilic complex. It is currently not clear whether such a complex can directly lead to apoptosis or it requires the binding of other downstream effectors. Another possibility is that region S, identified here as the binding motif for such a homophilic complex, may itself be the target for such effector binding. Similarly, we cannot rule out the possibility that the BNIP-Salpha BCH domain can mediate the apoptotic effect by interacting with other BCH-containing proteins, for example Cdc42GAP, and its target Cdc42. Work is currently underway to test out this hypothesis.

Various pathways for induction of apoptosis have already been established. These include the involvement of certain caspases, sequestration, gene expression, and inhibition of certain pro-survival factors such as Bcl-2 or releasing the cofactors that are necessary for the execution of the death signal (8-11). Our data showed that BNIP-Salpha -induced apoptosis could not be blocked by pre-treatment with general caspase inhibitors, and that under physiological pull-down assays, it did not interact with two well known pro-survival factors, Bcl-2 and Bcl-xL. Furthermore, coexpression of increasing amount of Bcl-2 or Bcl-xL in the presence of BNIP-Salpha also failed to rescue cells from undergoing apoptosis (data not shown). These data imply that BNIP-Salpha -induced apoptosis did not directly involve at least some of the known caspases and the Bcl-2/Bcl-xL inhibition. Further structural and functional analysis of the sequence motif and the BCH domain is therefore necessary to help identify the detailed mechanism of its pro-apoptotic function. To this end, we are planning to use the proteomics approach to identify other targets of this sequence.

Secondly, it is worth noting that various conserved BCH domains are also present in many uncharacterized proteins in Caenorhabditis elegans, Drosophila, Arabidopsis, and humans.2 It remains to be seen how many of these proteins have the similar roles in their respective host organisms and whether they possess varying degrees of specificity in their functions. We are currently examining what effects closely related BCH domain-containing proteins, such as BNIP-2 and Cdc42GAP, have on mammalian cell physiology.

Thirdly, it is intriguing that the BCH domain shares a considerable degree of homology to the Sec-14p domain of the phosphatidylinositol transfer protein (34), which was previously documented as a lipid-binding entity (35). This raises several interesting questions: Do BCH and Sec-14p domains have common properties and functions? Apparently they evolved from the same ancestral sequence but the question remains; have they evolved in binding capability and function or do they still recognize a similar cellular target? In essence, does the BCH domain of the BNIP-2 family possess lipid-binding activity and does the Sec-14p-containing proteins confer apoptotic activity via the Sec-14p domain? We are currently investigating these questions by the structural and genetics approach. In this regard, we propose that the BCH domain be referred to as the BCH/Sec-14p-like domain unless their unique properties and functions are shown to be clearly distinguishable.

Fourthly, one of the major surprises from the announcement of the human genome sequence is the relatively low number of genes, estimated to be around 30,000-45,000 in the human genome (7). This number does not directly translate into the number of possible proteins that can occur because each gene has the capacity to give rise to many isoforms of its derivative proteins (36, 37). The capability of alternative RNA splicing to generate multiple types of protein isoforms from a single gene contig allows proteome complexity and confers a greater repertoire for physiological regulation.

The identification of BNIP-Salpha and BNIP-Sbeta isoforms provides an example of alternative RNA splicing that confers distinctly different properties on the resultant proteins. Interestingly, the mechanism involving differential RNA splicing has also been widely observed recently in the family of proteins regulating apoptosis. They include the Bcl-2 family (38, 39), inhibitors of apoptosis family (40), caspases (41), and mammalian Ced-4 homologues such as DEFCAP (42). Variant proteins generated via this mechanism could carry out completely different functions or even opposing cellular effects. This raises an interesting question as to how such selective RNA splicing is regulated in response to changes in the cell's homeostasis.

In summary, using the bioinformatics approach coupled with molecular cloning and functional studies, we have identified a novel family of cellular regulators that are homologous to BNIP-2, designated BNIP-S. The gene encodes two isoforms, BNIP-Salpha and BNIP-Sbeta , via an alternative RNA splicing that generates species that differ only by the complete BCH/Sec-14p-like domain. Such a complete domain is instrumental for BNIP-Salpha -induced apoptosis and thus represents a novel apoptosis-inducing sequence.

    FOOTNOTES

* This work was supported in part by a grant from the Academic Research Fund, National University of Singapore.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§ Recipient of Research Scholarship from National University of Singapore.

|| To whom correspondence should be addressed. Tel.: 65-874-7834; Fax: 65-779-2486; E-mail: dbslowbc@nus.edu.sg.

Published, JBC Papers in Press, December 10, 2001, DOI 10.1074/jbc.M109459200

2 Y. T. Zhou, U. J. K. Soh, X. Shang, G. R. Guy, and B. C. Low, unpublished data.

    ABBREVIATIONS

The abbreviations used are: BCH, BNIP-2 and Cdc42GAP Homology; RT, reverse transcription; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GST, glutathione S-transferase; GFP, green fluorescence protein; EST, expressed sequence tag.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1. Hunter, T. (2000) Cell 100, 113-127[CrossRef][Medline] [Order article via Infotrieve]
2. Sudol, M., and Hunter, T. (2000) Cell 103, 1001-1004[CrossRef][Medline] [Order article via Infotrieve]
3. Collins, T., Stone, J. R., and Williams, A. J. (2001) Mol. Cell. Biol. 21, 3609-3615[Free Full Text]
4. Pawson, T., and Nash, P. (2000) Genes Dev. 14, 1027-1047[Free Full Text]
5. Pawson, T., and Scott, J. D. (1997) Science 278, 2075-2080[Abstract/Free Full Text]
6. Apweiler, R., Attwood, T. K., Bairoch, A., Bateman, A., Birney, E., Biswa, M., Buche, P., Cerutti, L., Corpet, F., Croning, M. D., Durbin, R., Falquet, L., Fleischmann, W., Gouzy, J., Hermjakob, H., Hulo, N., Jonassen, I., Kahn, D., Kanapin, A., Karavidopoulou, Y., Lopez, R., Marx, B., Mulder, N. J., Oinn, T. M., Pagni, M., Servant, F., Sigrist, C. J., and Zdobnov, E. M. (2001) Nucleic Acids Res. 29, 37-40[Abstract/Free Full Text]
7. Baltimore, D. (2001) Nature 409, 814-815[CrossRef][Medline] [Order article via Infotrieve]
8. Adams, J. M., and Cory, S. (2001) Trends Biochem. Sci. 26, 61-66[CrossRef][Medline] [Order article via Infotrieve]
9. Evan, G. I., and Vousden, K. H. (2001) Nature 411, 342-348[CrossRef][Medline] [Order article via Infotrieve]
10. Meier, P., Finch, A., and Evan, G. (2000) Nature 407, 796-801[CrossRef][Medline] [Order article via Infotrieve]
11. Porter, A. (1999) Trends Cell Biol. 9, 394-401[CrossRef][Medline] [Order article via Infotrieve]
12. Reed, J. C. (1998) Oncogene 17, 3225-3236[CrossRef][Medline] [Order article via Infotrieve]
13. Muchmore, S. W., Chen, J., Jakob, C., Zakula, D., Matayoshi, E. D., Wu, W., Zhang, H., Li, F., Ng, S. C., and Altieri, D. C. (2000) Mol. Cell 6, 173-182[CrossRef][Medline] [Order article via Infotrieve]
14. Watt, W., Koeplinger, K. A., Mildner, A. M., Heinrikson, R. L., Tomasselli, A. G., and Watenpaugh, K. D. (1999) Struct. Fold. Des. 7, 1135-1143[Medline] [Order article via Infotrieve]
15. Zou, H., Henzel, W. J., Liu, X., Lutschg, A., and Wang, X. (1997) Cell 90, 405-413[CrossRef][Medline] [Order article via Infotrieve]
16. Loetscher, H., Pan, Y. C., Lahm, H. W., Gentz, R., Brockhaus, M., Tabuchi, H., and Lesslauer, W. (1990) Cell 61, 351-359[CrossRef][Medline] [Order article via Infotrieve]
17. Chinnaiyan, A. M., O'Rourke, K., Tewari, M., and Dixit, V. M. (1995) Cell 81, 505-512[CrossRef][Medline] [Order article via Infotrieve]
18. Van der Biezen, E. A., and Jones, J. D. (1998) Curr. Biol. 8, R226-R227[CrossRef][Medline] [Order article via Infotrieve]
19. Otomo, T., Sakahira, H., Uegaki, K., Nagata, S., and Yamazaki, T. (2000) Nat. Struct. Biol. 7, 658-662[CrossRef][Medline] [Order article via Infotrieve]
20. Mosialos, G., Birkenbach, M., Yalamanchili, R., VanArsdale, T., Ware, C., and Kieff, E. (1995) Cell 80, 389-399[CrossRef][Medline] [Order article via Infotrieve]
21. Takayama, S., Kochel, K., Irie, S., Inazawa, J., Abe, T., Sato, T., Druck, T., Huebner, K., and Reed, J. C. (1996) Genomics 35, 494-498[CrossRef][Medline] [Order article via Infotrieve]
22. Low, B. C., Lim, Y. P., Lim, J., Wong, E. S., and Guy, G. R. (1999) J. Biol. Chem. 274, 33123-33130[Abstract/Free Full Text]
23. Low, B. C., Seow, K. T., and Guy, G. R. (2000) J. Biol. Chem. 275, 14415-14422[Abstract/Free Full Text]
24. Low, B. C., Seow, K. T., and Guy, G. R. (2000) J. Biol. Chem. 275, 37742-37751[Abstract/Free Full Text]
25. Boyd, J. M., Malstrom, S., Subramanian, T., Venkatesh, L. K., Schaeper, U., Elangovan, B., D'Sa-Eipper, C., and Chinnadurai, G. (1994) Cell 79, 341-351[CrossRef][Medline] [Order article via Infotrieve]
26. Barfod, E. T., Zheng, Y., Kuang, W. J., Hart, M. J., Evans, T., Cerione, R. A., and Ashkenazi, A. (1993) J. Biol. Chem. 268, 26059-26062[Abstract/Free Full Text]
27. Lancaster, C. A., Taylor-Harris, P. M., Self, A. J., Brill, S., van Erp, H. E., and Hall, A. (1994) J. Biol. Chem. 269, 1137-1142[Abstract/Free Full Text]
28. Thompson, J. D., Higgins, D. G., and Gibson, T. J. (1994) Nucleic Acids Res. 22, 4673-4680[Abstract/Free Full Text]
29. Sambrook, J., and Russell, D. W. (2001) Molecular Cloning: A Laboratory Manual , 3rd Ed. , Cold Spring Harbor Press, Cold Spring Harbor, NY
30. Lim, J., Wong, E. S., Ong, S. H., Yusoff, P., Low, B. C., and Guy, G. R. (2000) J. Biol. Chem. 275, 32837-32845[Abstract/Free Full Text]
31. Chan, S. L., Tan, K. O., Zhang, L., Yee, K. S., Ronca, F., Chan, M. Y., and Yu, V. C. (1999) J. Biol. Chem. 274, 2461-2468
32. Spector, D. L., Goldman, R. D., and Leinwand, L. A. (1998) Cells: A Laboratory Manual , Cold Spring Harbor Press, Cold Spring Harbor, NY
33. Iida, Y. (1990) J. Theor. Biol. 145, 23-33
34. Aravind, L., Neuwald, A. F., and Ponting, C. P. (1999) Curr. Biol. 9, 195-197
35. Phillips, S. E., Sha, B., Topalof, L., Xie, Z., Alb, J. G., Klenchin, V. A, Swigart, P., Cockcroft, S., Martin, T. F, Luo, M., and Bankaitis, V. A. (1999) Mol. Cell 4, 187-197[CrossRef][Medline] [Order article via Infotrieve]
36. International Human Genome Sequencing Consortium. (2001) Nature 409, 860-921[CrossRef][Medline] [Order article via Infotrieve]
37. Croft, L., Schandorff, S., Clark, F., Burrage, K., Arctander, P., and Mattick, J. S. (2000) Nat. Genet. 24, 340-341[CrossRef][Medline] [Order article via Infotrieve]
38. Bingle, C. D., Craig, R. W., Swales, B. M., Singleton, V., Zhou, P., and Whyte, M. K. (2000) J. Biol. Chem. 275, 22136-22146[Abstract/Free Full Text]
39. Sun, Y. F., Yu, L. Y., Saarma, M., Timmusk, T., and Arumae, U. (2001) J. Biol. Chem. 276, 16240-16247[Abstract/Free Full Text]
40. Ashhab, Y., Alian, A., Polliack, A., Panet, A., and Yehuda, D. B. (2001) FEBS Lett. 495, 56-60[CrossRef][Medline] [Order article via Infotrieve]
41. Cote, J., Dupuis, S., Jiang, Z., and Wu, J. Y. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 938-943[Abstract/Free Full Text]
42. Hlaing, T., Guo, R. F., Dilley, K. A., Loussia, J. M., Morrish, T. A., Shi, M. M., Vincenz, C., and Ward, P. A. (2001) J. Biol. Chem. 276, 9230-9238[Abstract/Free Full Text]
43. Kozak, M. (1992) Annu. Rev. Cell Biol. 8, 197-225<