BIM and other BCL-2 family proteins exhibit cross-species conservation of function between zebrafish and mammals

Abstract

Here we investigate the function of zebrafish Bcl-2 family proteins and demonstrate important conservation of function across zebrafish and mammalian systems. We have isolated a zebrafish ortholog of mammalian BIM and show that it is the most toxic of the zebrafish BH3-only genes examined, sharing this characteristic with the mammalian BIM gene. The zebrafish bad gene shows a complete lack of embryonic lethality, but like mammalian BAD, its pro-apoptotic activity is regulated through phosphorylation of critical serines. We also found that the pattern of mitochondrial dysfunction observed by zebrafish BH3 domain peptides in a mammalian cytochrome c release assay recapitulates the pattern of embryonic lethality induced by the respective mRNA injections in vivo. In contrast to zebrafish Bim, Bid exhibited only weak binding to zebrafish Bcl-2 and moderate-to-weak overall lethality in zebrafish embryos and isolated mitochondria. Given that zebrafish Bcl-2 binds strongly to mammalian BID and BIM peptides and proteins, the protein identified as the zebrafish Bid ortholog has different properties than mammalian BID. Overall, our results demonstrate the high degree of functional conservation between zebrafish and mammalian Bcl-2 family proteins, thus validating the zebrafish as a model system to further dissect the molecular mechanisms that regulate apoptosis in future forward genetic and chemical modifier screens.

Figure 1

Alignment of zebrafish Bim and mouse and human BIM (EL) proteins. The zebrafish bim gene was identified and its protein product aligned with mouse and human BIM sequences (GenBank nos. AF032459 and AF032457, respectively) using ClustalW and Boxshade from the EMBnet server. The domain specific to the EL isoform lies in the area between the arrows and the nine-amino-acid core BH3 domain is underlined. *Conservation of human serine 69, shown previously to be phosphorylated by ERK1/2.^{, #}Possible conservation of human serine 87 (depending on where the zebrafish serine 68 (zS68) is aligned), shown previously to be phosphorylated by AKT. Also shown, the mammalian dynein light-chain (DLC) binding site does not appear to be conserved in zebrafish Bim. (b) Individual pairwise alignments were used to determine percent identity and overall similarity between protein sequences

Figure 2

Overexpression of the zebrafish BH3-only genes bim, noxa, and puma, but not bad or bid, causes widespread embryonic death via maternally provided apoptotic machinery. (a) One-cell stage embryos were injected with 100 ng/μl of mRNA encoding the indicated genes, and development of the embryos was evaluated by brightfield microscopy. One representative embryo from each group is shown. At the two- to four-cell stage, half of the embryos were fixed and analyzed for apoptosis by Casp3 assay. Fluorescence indicates the presence of activated caspase-3 in a representative embryo from each group. (b) Groups treated as in (a) are shown at the 8- to 16-cell stage to demonstrate that bim, noxa, and puma mRNA injections give rise to overall similar phenotypes. (c) One-cell stage embryos were injected with a range of doses of mRNA encoding the indicated genes, and embryos were scored for survival at 2 h post-injection. The concentration of wild-type mRNA in ng/μl is indicated for each gene. The total mRNA concentration of each injection was held constant at 100 ng/μl using egfp mRNA to make up the difference. BH3 mutant forms of each gene (bh3mut) described in (d) were injected at 100 ng/μl and included to demonstrate that the lethality is caused by the respective BH3-domain. Standard deviation is representative of three to six independent experiments. (d) BH3 domain mutations were created for each zebrafish BH3-only gene by changing the residue(s) denoted by an asterisk to alanine

Figure 3

Zebrafish Bad enhances radiation-induced apoptosis and its activity is likely regulated by serine phosphorylation. (a) Embryos were either uninjected or injected with 100 ng/μl of mcherry-bad and analyzed for mcherry expression by fluorescence microscopy 2 h after injection. One representative embryo from each group is shown. Asterisks denote the yolk region. Arrows point to the blastomeres of the animal pole. (b) Embryos were injected at the one-cell stage with mRNA encoding mcherry (as a negative control) or mcherry-bad. At 24 hpf, embryos were either left unexposed or exposed to 15 Gy γ-radiation and harvested 3 h later. The Casp3 assay was performed, and apoptotic cells were visualized by fluorescence microscopy. (c) Embryos were injected with mRNA encoding mcherry, mcherry-bad, mcherry-bad bh3mut (mcherry-bad L99A, see Figure 2c), or mcherry-bad-2SD and either left unexposed or exposed to 4 Gy γ-radiation and harvested and analyzed as in (b). In the whole embryo (above), the white box denotes the region of the spinal cord documented below. (d) Five to ten embryos from each group of embryos analyzed in (c) were blindly selected and apoptotic cells were documented by fluorescence microscopy. Fluorescence intensity from a representative region of the spinal cord from each embryo was quantified by Volocity software and displayed as arbitrary units of fluorescence (pixel) intensity. Student’s t-tests were performed as indicated to measure statistical significance. **P<0.05. (e) Alignment of part of zebrafish Bad and mouse and human BAD proteins. Asterisks denote serines 84 and 103 in zebrafish Bad that show conservation with mouse serines 136 and 155 (and human serines 99 and 118), respectively. (f) Messenger RNA encoding mcherry (control), bad, and bad 2SA was injected into zebrafish embryos at the one-cell stage, and survival was measured at the indicated time points. (g) Embryos surviving at 5-h post-injection in the experiment described in (f) were analyzed by Casp3 assay. Asterisks denote the yolk region. Arrows point to the blastomeres of the animal pole

Figure 4

Zebrafish BH3-only peptides display a toxicity in isolated mouse mitochondria that is similar to lethality of the respective full-length proteins in vivo. (a) To design zebrafish BH3 domain peptides, zebrafish BH3-only protein sequences were aligned with human or mouse peptides as described previously.^, (b) Mitochondria (0.5 mg/ml) isolated from wild-type MEFs were treated with zebrafish, mouse, or human BH3 domain peptides (10 μM). Percent cytochrome c release was measured by ELISA. (c) BH3 domain peptides were analyzed as in (b) but with mitochondria isolated from BAX^–/–, BAK^–/– MEFs (DKO). Standard deviations of at least three independent measurements are in parentheses

Figure 5

Zebrafish Bcl-2 has evolved to bind a distinct set of BH3-only proteins. Binding affinities between GST-zBcl-2 and BH3 domain peptides were analyzed by fluorescence polarization. (a) Representative binding curves show selective binding between antiapoptotic zebrafish Bcl-2 and zebrafish and mouse BH3-only family members. (b) Dissociation constants (in nM) were calculated for each interaction and then corrected for impurities in the protein (62.8% pure protein, 37.2% impurities). Standard deviations of at least three independent measurements are in parentheses. Yellow blocks signify high affinity binding, blue blocks and minus sign signify no observed binding (K_d>700 nM). Activators are in purple, sensitizers in green. n.a., not assayed. We previously published the GST-BCL-2 binding data, and permission was kindly granted by Cancer Cell to reprint the data here for the sake of comparison

Figure 6

Zebrafish Bcl-2 protects FL5.12 cells from IL-3 withdrawal-induced apoptosis, likely through sequestration of mouse Bim. (a) Zebrafish Bcl-2 interacts with mouse BIM and BID. A GST pull-down assay was performed to analyze zebrafish Bcl-2 binding interactions. GST-zBcl-2 was immobilized on glutathione-agarose beads and combined with LY-1 cell lysate (100 μg). Proteins interacting with GST-zBcl-2 were analyzed by western blot with antibodies recognizing BIM and BID. (b) Overexpression of zebrafish Bcl-2 protects FL5.12 cells from IL-3 withdrawal-induced apoptosis. FL5.12 cells were stably transfected with vector alone or pCMV-3XFLAG-zBcl-2 in the presence of IL-3. Cells were then cultured in the presence or absence of IL-3 for 30 h and analyzed by Annexin V assay. The average and standard deviation of three independent experiments are shown. (c) Zebrafish Bcl-2 binds more BIM in FL5.12 cells upon withdrawal of IL-3. Lysates were prepared from cells treated as described in (b). Lysates were combined with Protein A beads and incubated with FLAG antibody (7 μg). Proteins interacting with FLAG (vector) and FLAG-zBcl-2 were analyzed and compared by western blot with antibodies to the indicated BCL-2 family proteins