BNIP-2 induces cell elongation and membrane protrusions by interacting with Cdc42 via a unique Cdc42-binding motif within its BNIP-2 and Cdc42GAP homology domain

Structural basis for the distinct roles of non-conserved Pro116 and conserved Tyr124 of BCH domain of yeast p50RhoGAP

p50RhoGAP is a key protein that interacts with and downregulates the small GTPase RhoA. p50RhoGAP is a multifunctional protein containing the BNIP-2 and Cdc42GAP Homology (BCH) domain that facilitates protein-protein interactions and lipid binding and the GAP domain that regulates active RhoA population. We recently solved the structure of the BCH domain from yeast p50RhoGAP (YBCH) and showed that it maintains the adjacent GAP domain in an auto-inhibited state through the β5 strand. Our previous WT YBCH structure shows that a unique kink at position 116 thought to be made by a proline residue between alpha helices α6 and α7 is essential for the formation of intertwined dimer from asymmetric monomers. Here we sought to establish the role and impact of this Pro116. However, the kink persists in the structure of P116A mutant YBCH domain, suggesting that the scaffold is not dictated by the proline residue at this position. We further identified Tyr124 (or Tyr188 in HBCH) as a conserved residue in the crucial β5 strand. Extending to the human ortholog, when substituted to acidic residues, Tyr188D or Tyr188E, we observed an increase in RhoA binding and self-dimerization, indicative of a loss of inhibition of the GAP domain by the BCH domain. These results point to distinct roles and impact of the non-conserved and conserved amino acid positions in regulating the structural and functional complexity of the BCH domain.

BNIP-2 Activation of Cellular Contractility Inactivates YAP for H9c2 Cardiomyoblast Differentiation

Rho GTPases and Hippo kinases are key regulators of cardiomyoblast differentiation. However, how these signaling axes are coordinated spatiotemporally remains unclear. Here, the central and multifaceted roles of the BCH domain containing protein, BNIP-2, in orchestrating the expression of two key cardiac genes (cardiac troponin T [cTnT] and cardiac myosin light chain [Myl2]) in H9c2 and human embryonic stem cell-derived cardiomyocytes are delineated. This study shows that BNIP-2 mRNA and protein expression increase with the onset of cTnT and Myl2 and promote the alignment of H9c2 cardiomyocytes. Mechanistically, BNIP-2 is required for the inactivation of YAP through YAP phosphorylation and its cytosolic retention. Turbo-ID proximity labeling corroborated by super-resolution analyses and biochemical pulldown data reveals a scaffolding role of BNIP-2 for LATS1 to phosphorylate and inactivate YAP in a process that requires BNIP-2 activation of cellular contractility. The findings identify BNIP-2 as a pivotal signaling scaffold that spatiotemporally integrates RhoA/Myosin II and LATS1/YAP mechanotransduction signaling to drive cardiomyoblast differentiation, by switching the genetic programming from YAP-dependent growth to YAP-silenced differentiation. These findings offer insights into the importance of scaffolding proteins in bridging the gap between mechanical and biochemical signals in cell growth and differentiation and the prospects in translational applications.

Structural basis for p50RhoGAP BCH domain-mediated regulation of Rho inactivation

Spatiotemporal regulation of signaling cascades is crucial for various biological pathways, under the control of a range of scaffolding proteins. The BNIP-2 and Cdc42GAP Homology (BCH) domain is a highly conserved module that targets small GTPases and their regulators. Proteins bearing BCH domains are key for driving cell elongation, retraction, membrane protrusion, and other aspects of active morphogenesis during cell migration, myoblast differentiation, and neuritogenesis. We previously showed that the BCH domain of p50RhoGAP (ARHGAP1) sequesters RhoA from inactivation by its adjacent GAP domain; however, the underlying molecular mechanism for RhoA inactivation by p50RhoGAP remains unknown. Here, we report the crystal structure of the BCH domain of p50RhoGAP Schizosaccharomyces pombe and model the human p50RhoGAP BCH domain to understand its regulatory function using in vitro and cell line studies. We show that the BCH domain adopts an intertwined dimeric structure with asymmetric monomers and harbors a unique RhoA-binding loop and a lipid-binding pocket that anchors prenylated RhoA. Interestingly, the β5-strand of the BCH domain is involved in an intermolecular β-sheet, which is crucial for inhibition of the adjacent GAP domain. A destabilizing mutation in the β5-strand triggers the release of the GAP domain from autoinhibition. This renders p50RhoGAP active, thereby leading to RhoA inactivation and increased self-association of p50RhoGAP molecules via their BCH domains. Our results offer key insight into the concerted spatiotemporal regulation of Rho activity by BCH domain-containing proteins.

Rac1 and EGFR cooperate to activate Pak in response to nutrient stress

The interplay between nutrient scarcity and signal transduction circuits is an important aspect of tumorigenesis that regulates many aspects of cancer progression. Glutamine is a critical nutrient for cancer cells, as it contributes to biosynthetic reactions that sustain cancer proliferation and growth. In tumors, because nutrient utilization can often outpace supply, glutamine levels can become limiting and oncogene-mediated metabolic rewiring triggers signaling cascades that support nutrient stress survival. Recently, we identified that in pancreatic ductal adenocarcinoma (PDAC) cells, glutamine depletion can trigger p21-activated kinase (Pak) activation through EGFR signaling as a means to circumvent metabolic stress. Here, we elucidate that glutamine starvation, as well EGF stimulation, can enhance the presence of many different Pak phosphoforms, and that this activation only occurs in a subset of PDAC cells. Pak is a well-established effector of Rac1, and while Rac1 mutant variants can modulate the metabolic induction of Pak phosphorylation, Rac1 inhibition only partially attenuates Pak activation upon glutamine depletion. We decipher that in order to efficiently suppress metabolic activation of Pak, both EGFR and Rac1 signaling must be inhibited. These results provide a mechanistic understanding of how glutamine-regulated signal transduction can control Pak activation in PDAC cells.

BNIP-2 retards breast cancer cell migration by coupling microtubule-mediated GEF-H1 and RhoA activation

Microtubules display dynamic turnover during cell migration, leading to cell contractility and focal adhesion maturation regulated by Rho guanosine triphosphatase activity. This interplay between microtubules and actomyosin is mediated by guanine nucleotide exchange factor (GEF)-H1 released after microtubule depolymerization or microtubule disconnection from focal adhesions. However, how GEF-H1 activates Rho upon microtubule disassembly remains elusive. Here, we found that BNIP-2, a BCH domain-containing protein that binds both RhoA and GEF-H1 and traffics with kinesin-1 on microtubules, is important for GEF-H1-driven RhoA activation upon microtubule disassembly. Depletion of BNIP-2 in MDA-MB-231 breast cancer cells decreases RhoA activity and promotes cell migration. Upon nocodazole-induced microtubule disassembly, the interaction between BNIP-2 and GEF-H1 increases, while knockdown of BNIP-2 reduces RhoA activation and cell rounding via uncoupling RhoA-GEF-H1 interaction. Together, these findings revealed that BNIP-2 couples microtubules and focal adhesions via scaffolding GEF-H1 and RhoA, fine-tuning RhoA activity and cell migration.

BNIP-2 binds phosphatidylserine, localizes to vesicles, and is transported by kinesin-1

BNIP-2 shows high homology with the Cayman ataxia protein, caytaxin, which functions as a kinesin-1 adapter bridging cargos and kinesin light chains (KLCs). BNIP-2 is known to induce cell shape changes when over-expressed in culture cells, but its physiological functions are mostly unknown. BNIP-2 interacts with KLC through the conserved WED motif in the N-terminal region of BNIP-2. Interaction with KLC and transportation by kinesin-1 are essential for over-expressed BNIP-2 to elongate cells and induce cellular processes. Endogenous BNIP-2 localizes to the Golgi apparatus, early and recycling endosomes and mitochondria, aligned with microtubules, and moves at a speed compatible with kinesin-1 transportation. The CRAL-TRIO domain of BNIP-2 specifically interacts with phosphatidylserine, and the vesicular localization of BNIP-2 requires interaction with this phospholipid. BNIP-2 mutants which do not bind phosphatidylserine do not induce morphological changes in cells. These data show that similar to caytaxin, BNIP-2 is a kinesin-1 adapter involved in vesicular transportation in the cytoplasm and that association with cargos depends on interaction of the CRAL-TRIO domain with membrane phosphatidylserine.

KIF5B transports BNIP-2 to regulate p38 mitogen-activated protein kinase activation and myoblast differentiation

Cdo bridges scaffold proteins BNIP-2 and JLP to activate p38MAPK during myoblast differentiation. KIF5B is a novel interacting partner of BNIP-2 and promotes myogenic differentiation. KIF5B-dependent transport of BNIP-2 is essential for its promyogenic effects.

The Cdo-p38MAPK (p38 mitogen-activated protein kinase) signaling pathway plays important roles in regulating skeletal myogenesis. During myogenic differentiation, the cell surface receptor Cdo bridges scaffold proteins BNIP-2 and JLP and activates p38MAPK, but the spatial-temporal regulation of this process is largely unknown. We here report that KIF5B, the heavy chain of kinesin-1 motor, is a novel interacting partner of BNIP-2. Coimmunoprecipitation and far-Western study revealed that BNIP-2 directly interacted with the motor and tail domains of KIF5B via its BCH domain. By using a range of organelle markers and live microscopy, we determined the endosomal localization of BNIP-2 and revealed the microtubule-dependent anterograde transport of BNIP-2 in C2C12 cells. The anterograde transport of BNIP-2 was disrupted by a dominant-negative mutant of KIF5B. In addition, knockdown of KIF5B causes aberrant aggregation of BNIP-2, confirming that KIF5B is critical for the anterograde transport of BNIP-2 in cells. Gain- and loss-of-function experiments further showed that KIF5B modulates p38MAPK activity and in turn promotes myogenic differentiation. Of importance, the KIF5B-dependent anterograde transport of BNIP-2 is critical for its promyogenic effects. Our data reveal a novel role of KIF5B in the spatial regulation of Cdo–BNIP-2–p38MAPK signaling and disclose a previously unappreciated linkage between the intracellular transporting system and myogenesis regulation.

Importance of extended protease substrate recognition motifs in steering BNIP-2 cleavage by human and mouse granzymes B

Background

Previous screening of the substrate repertoires and substrate specificity profiles of granzymes resulted in long substrate lists highly likely containing bystander substrates. Here, a recently developed degradomics technology that allows distinguishing efficiently from less efficiently cleaved substrates was applied to study the degradome of mouse granzyme B (mGrB).

Results

In vitro kinetic degradome analysis resulted in the identification of 37 mGrB cleavage events, 9 of which could be assigned as efficiently targeted ones. Previously, cleavage at the IEAD₇₅ tetrapeptide motif of Bid was shown to be efficiently and exclusively targeted by human granzyme B (hGrB) and thus not by mGrB. Strikingly, and despite holding an identical P4-P1 human Bid (hBid) cleavage motif, mGrB was shown to efficiently cleave the BCL2/adenovirus E1B 19 kDa protein-interacting protein 2 or BNIP-2 at IEAD₂₈. Like Bid, BNIP-2 represents a pro-apoptotic Bcl-2 protein family member and a potential regulator of GrB induced cell death. Next, in vitro analyses demonstrated the increased efficiency of human and mouse BNIP-2 cleavage by mGrB as compared to hGrB indicative for differing Bid/BNIP-2 substrate traits beyond the P4-P1 IEAD cleavage motif influencing cleavage efficiency. Murinisation of differential primed site residues in hBNIP-2 revealed that, although all contributing, a single mutation at the P3′ position was found to significantly increase the mGrB/hGrB cleavage ratio, whereas mutating the P1′ position from I₂₉ > T yielded a 4-fold increase in mGrB cleavage efficiency. Finally, mutagenesis analyses revealed the composite BNIP-2 precursor patterns to be the result of alternative translation initiation at near-cognate start sites within the 5′ leader sequence (5′UTR) of BNIP-2.

Conclusions

Despite their high sequence similarity, and previously explained by their distinct tetrapeptide specificities observed, the substrate repertoires of mouse and human granzymes B only partially overlap. Here, we show that the substrate sequence context beyond the P4-P1 positions can influence orthologous granzyme B cleavage efficiencies to an unmatched extent. More specifically, in BNIP-2, the identical and hGrB optimal IEAD tetrapeptide substrate motif is targeted highly efficiently by mGrB, while this tetrapeptide motif is refractory towards mGrB cleavage in Bid.

Bmcc1s interacts with the phosphate-activated glutaminase in the brain

Bmcc1s, a brain-enriched short isoform of the BCH-domain containing molecule Bmcc1, has recently been shown to interact with the microtubule-associated protein MAP6 and to regulate cell morphology. Here we identified kidney-type glutaminase (KGA), the mitochondrial enzyme responsible for the conversion of glutamine to glutamate in neurons, as a novel partner of Bmcc1s. Co-immunoprecipitation experiments confirmed that Bmcc1s and KGA form a physiological complex in the brain, whereas binding and modeling studies showed that they interact with each other. Overexpression of Bmcc1s in mouse primary cortical neurons impaired proper mitochondrial targeting of KGA leading to its accumulation within the cytoplasm. Thus, Bmcc1s may control the trafficking of KGA to the mitochondria.

Expression of Caytaxin protein in Cayman Ataxia mouse models correlates with phenotype severity

Caytaxin is a highly-conserved protein, which is encoded by the Atcay/ATCAY gene. Mutations in Atcay/ATCAY have been identified as causative of cerebellar disorders such as the rare hereditary disease Cayman ataxia in humans, generalized dystonia in the dystonic (dt) rat, and marked motor defects in three ataxic mouse lines. While several lines of evidence suggest that Caytaxin plays a critical role in maintaining nervous system processes, the physiological function of Caytaxin has not been fully characterized. In the study presented here, we generated novel specific monoclonal antibodies against full-length Caytaxin to examine endogenous Caytaxin expression in wild type and Atcay mutant mouse lines. Caytaxin protein is absent from brain tissues in the two severely ataxic Atcay^jit (jittery) and Atcay^swd (sidewinder) mutant lines, and markedly decreased in the mildly ataxic/dystonic Atcay^ji-hes (hesitant) line, indicating a correlation between Caytaxin expression and disease severity. As the expression of wild type human Caytaxin in mutant sidewinder and jittery mice rescues the ataxic phenotype, Caytaxin’s physiological function appears to be conserved between the human and mouse orthologs. Across multiple species and in several neuronal cell lines Caytaxin is expressed as several protein isoforms, the two largest of which are caused by the usage of conserved methionine translation start sites. The work described in this manuscript presents an initial characterization of the Caytaxin protein and its expression in wild type and several mutant mouse models. Utilizing these animal models of human Cayman Ataxia will now allow an in-depth analysis to elucidate Caytaxin’s role in maintaining normal neuronal function.

SmgGDS antagonizes BPGAP1-induced Ras/ERK activation and neuritogenesis in PC12 cell differentiation

BPGAP1 controls morphogenesis, migration, and ERK signaling by the concerted action of its multiple domains. Its BCH domain targets K-Ras and induces robust ERK activation and neuronal differentiation in a process antagonized by SmgGDS. The results highlight unique cross-talk of two regulators of GTPases in Ras/ERK signaling and differentiation.

BPGAP1 is a Rho GTPase-activating protein (RhoGAP) that regulates cell morphogenesis, cell migration, and ERK signaling by the concerted action of its proline-rich region (PRR), RhoGAP domain, and the BNIP-2 and Cdc42GAP homology (BCH) domain. Although multiple cellular targets for the PRR and RhoGAP have been identified, and their functions delineated, the mechanism by which the BCH domain regulates functions of BPGAP1 remains unclear. Here we show that its BCH domain induced robust ERK activation leading to PC12 cell differentiation by targeting specifically to K-Ras. Such stimulatory effect was inhibited, however, by both dominant-negative mutants of Mek2 (Mek2-K101A) and K-Ras (K-Ras-S17N) and also by the small G-protein GDP dissociation stimulator (SmgGDS). Consequently SmgGDS knockdown released this inhibition and resulted in a superinduction of K-Ras activation and PC12 differentiation mediated by BCH domain. These results demonstrate the versatility of the BCH domain of BPGAP1 in regulating ERK signaling by involving K-Ras and SmgGDS and support the unique role of BPGAP1 as a dual regulator for Ras and Rho signaling in cell morphogenesis and differentiation.

Modularity and functional plasticity of scaffold proteins as p(l)acemakers in cell signaling

Cells coordinate and integrate various functional modules that control their dynamics, intracellular trafficking, metabolism and gene expression. Such capacity is mediated by specific scaffold proteins that tether multiple components of signaling pathways at plasma membrane, Golgi apparatus, mitochondria, endoplasmic reticulum, nucleus and in more specialized subcellular structures such as focal adhesions, cell-cell junctions, endosomes, vesicles and synapses. Scaffold proteins act as "pacemakers" as well as "placemakers" that regulate the temporal, spatial and kinetic aspects of protein complex assembly by modulating the local concentrations, proximity, subcellular dispositions and biochemical properties of the target proteins through the intricate use of their modular protein domains. These regulatory mechanisms allow them to gate the specificity, integration and crosstalk of different signaling modules. In addition to acting as physical platforms for protein assembly, many professional scaffold proteins can also directly modify the properties of their targets while they themselves can be regulated by post-translational modifications and/or mechanical forces. Furthermore, multiple scaffold proteins can form alliances of higher-order regulatory networks. Here, we highlight the emerging themes of scaffold proteins by analyzing their common and distinctive mechanisms of action and regulation, which underlie their functional plasticity in cell signaling. Understanding these mechanisms in the context of space, time and force should have ramifications for human physiology and for developing new therapeutic approaches to control pathological states and diseases.

Bmcc1s, a novel brain-isoform of Bmcc1, affects cell morphology by regulating MAP6/STOP functions

The BCH (BNIP2 and Cdc42GAP Homology) domain-containing protein Bmcc1/Prune2 is highly enriched in the brain and is involved in the regulation of cytoskeleton dynamics and cell survival. However, the molecular mechanisms accounting for these functions are poorly defined. Here, we have identified Bmcc1s, a novel isoform of Bmcc1 predominantly expressed in the mouse brain. In primary cultures of astrocytes and neurons, Bmcc1s localized on intermediate filaments and microtubules and interacted directly with MAP6/STOP, a microtubule-binding protein responsible for microtubule cold stability. Bmcc1s overexpression inhibited MAP6-induced microtubule cold stability by displacing MAP6 away from microtubules. It also resulted in the formation of membrane protrusions for which MAP6 was a necessary cofactor of Bmcc1s. This study identifies Bmcc1s as a new MAP6 interacting protein able to modulate MAP6-induced microtubule cold stability. Moreover, it illustrates a novel mechanism by which Bmcc1 regulates cell morphology.

Cross-species analyses identify the BNIP-2 and Cdc42GAP homology (BCH) domain as a distinct functional subclass of the CRAL_TRIO/Sec14 superfamily

The CRAL_TRIO protein domain, which is unique to the Sec14 protein superfamily, binds to a diverse set of small lipophilic ligands. Similar domains are found in a range of different proteins including neurofibromatosis type-1, a Ras GTPase-activating Protein (RasGAP) and Rho guanine nucleotide exchange factors (RhoGEFs). Proteins containing this structural protein domain exhibit a low sequence similarity and ligand specificity while maintaining an overall characteristic three-dimensional structure. We have previously demonstrated that the BNIP-2 and Cdc42GAP Homology (BCH) protein domain, which shares a low sequence homology with the CRAL_TRIO domain, can serve as a regulatory scaffold that binds to Rho, RhoGEFs and RhoGAPs to control various cell signalling processes. In this work, we investigate 175 BCH domain-containing proteins from a wide range of different organisms. A phylogenetic analysis with ∼100 CRAL_TRIO and similar domains from eight representative species indicates a clear distinction of BCH-containing proteins as a novel subclass within the CRAL_TRIO/Sec14 superfamily. BCH-containing proteins contain a hallmark sequence motif R(R/K)h(R/K)(R/K)NL(R/K)xhhhhHPs (‘h’ is large and hydrophobic residue and ‘s’ is small and weekly polar residue) and can be further subdivided into three unique subtypes associated with BNIP-2-N, macro- and RhoGAP-type protein domains. A previously unknown group of genes encoding ‘BCH-only’ domains is also identified in plants and arthropod species. Based on an analysis of their gene-structure and their protein domain context we hypothesize that BCH domain-containing genes evolved through gene duplication, intron insertions and domain swapping events. Furthermore, we explore the point of divergence between BCH and CRAL-TRIO proteins in relation to their ability to bind small GTPases, GAPs and GEFs and lipid ligands. Our study suggests a need for a more extensive analysis of previously uncharacterized BCH, ‘BCH-like’ and CRAL_TRIO-containing proteins and their significance in regulating signaling events involving small GTPases.

The expression and localization of Prune2 mRNA in the central nervous system

A family of Bcl-2/adenovirus E1B 19kDa-interacting proteins (BNIPs) plays critical roles in several cellular processes such as cellular transformation, apoptosis, neuronal differentiation, and synaptic function, which are mediated by the BNIP2 and Cdc42GAP homology (BCH) domain. Prune homolog 2 (Drosophila) (PRUNE2) and its isoforms -C9orf65, BCH motif-containing molecule at the carboxyl terminal region 1 (BMCC1), and BNIP2 Extra Long (BNIPXL) - have been shown to be a susceptibility gene for Alzheimer's disease, a biomarker for leiomyosarcomas, a proapoptotic protein in neuronal cells, and an antagonist of cellular transformation, respectively. However, precise localization of PRUNE2 in the brain remains unclear. Here, we identified the distribution of Prune2 mRNA in the adult mouse brain. Prune2 mRNA is predominantly expressed in the neurons of the cranial nerve motor nuclei and the motor neurons of the spinal cord. The expression in the dorsal root ganglia (DRG) is consistent with the previously described reports. In addition, we observed the expression in another sensory neuron in the mesencephalic trigeminal nucleus. These results suggest that Prune2 may be functional in these restricted brain regions.

Mammalian diseases of phosphatidylinositol transfer proteins and their homologs

Inositol and phosphoinositide signaling pathways represent major regulatory systems in eukaryotes. The physiological importance of these pathways is amply demonstrated by the variety of diseases that involve derangements in individual steps in inositide and phosphoinositide production and degradation. These diseases include numerous cancers, lipodystrophies and neurological syndromes. Phosphatidylinositol transfer proteins (PITPs) are emerging as fascinating regulators of phosphoinositide metabolism. Recent advances identify PITPs (and PITP-like proteins) to be coincidence detectors, which spatially and temporally coordinate the activities of diverse aspects of the cellular lipid metabolome with phosphoinositide signaling. These insights are providing new ideas regarding mechanisms of inherited mammalian diseases associated with derangements in the activities of PITPs and PITP-like proteins.

Animal models of dystonia: Lessons from a mutant rat

Dystonia is a motor sign characterized by involuntary muscle contractions which produce abnormal postures. Genetic factors contribute significantly to primary dystonia. In comparison, secondary dystonia can be caused by a wide variety of metabolic, structural, infectious, toxic and inflammatory insults to the nervous system. Although classically ascribed to dysfunction of the basal ganglia, studies of diverse animal models have pointed out that dystonia is a network disorder with important contributions from abnormal olivocerebellar signaling. In particular, work with the dystonic (dt) rat has engendered dramatic paradigm shifts in dystonia research. The dt rat manifests generalized dystonia caused by deficiency of the neuronally restricted protein caytaxin. Electrophysiological and biochemical studies have shown that defects at the climbing fiber-Purkinje cell synapse in the dt rat lead to abnormal bursting firing patterns in the cerebellar nuclei, which increases linearly with postnatal age. In a general sense, the dt rat has shown the scientific and clinical communities that dystonia can arise from dysfunctional cerebellar cortex. Furthermore, work with the dt rat has provided evidence that dystonia (1) is a neurodevelopmental network disorder and (2) can be driven by abnormal cerebellar output. In large part, work with other animal models has expanded upon studies in the dt rat and shown that primary dystonia is a multi-nodal network disorder associated with defective sensorimotor integration. In addition, experiments in genetically engineered models have been used to examine the underlying cellular pathologies that drive primary dystonia. This article is part of a Special Issue entitled "Advances in dystonia".

Identification of the BCL2/adenovirus E1B-19K protein-interacting protein 2 (BNIP-2) as a granzyme B target during human natural killer cell-mediated killing

Cytotoxic lymphocytes eliminate infected cells and tumours via the perforin-mediated delivery of pro-apoptotic serine proteases known as granzymes. Granzyme B triggers apoptosis via the cleavage of a repertoire of cellular proteins, leading to caspase activation and mitochondrial depolarization. A simple bioinformatics strategy identified a candidate granzyme B cleavage site in the widely expressed BNIP-2 (BCL2/adenovirus E1B-19K protein-interacting protein 2). Granzyme B cleaved recombinant BNIP-2 in vitro and endogenous BNIP-2 was cleaved during the NK (natural killer) cell-mediated killing of tumour cells. Cleavage required the site identified in the bioinformatics screen and was caspase-independent. Expression of either full-length BNIP-2 or a truncated molecule mimicking the granzyme B cleaved form was pro-apoptotic and led to the caspase-dependent cleavage of BNIP-2 at a site distinct from granzyme B cleavage. Inhibition of BNIP-2 expression did not affect the susceptibility to NK cell-mediated killing. Furthermore, target cells in which BID (BH3-interacting domain death agonist) expression was inhibited also remained highly susceptible to NK cell-mediated killing, revealing redundancy in the pro-apoptotic response to human cytotoxic lymphocytes. Such redundancy reduces the opportunity for escape from apoptosis induction and maximizes the chances of immune-mediated clearance of infected cells or tumour cells.

The BNIP-2 and Cdc42GAP homology (BCH) domain of p50RhoGAP/Cdc42GAP sequesters RhoA from inactivation by the adjacent GTPase-activating protein domain

The BNIP-2 and Cdc42GAP Homology (BCH) domain from p50RhoGAP sequesters RhoA from inactivation by the adjacent GAP domain and it confers unique Rho-binding profile from that of GAP domain. This suppression is further augmented by an intramolecular interaction, adding to a new paradigm for regulating p50RhoGAP signaling.

The BNIP-2 and Cdc42GAP homology (BCH) domain is a novel regulator for Rho GTPases, but its impact on p50-Rho GTPase-activating protein (p50RhoGAP or Cdc42GAP) in cells remains elusive. Here we show that deletion of the BCH domain from p50RhoGAP enhanced its GAP activity and caused drastic cell rounding. Introducing constitutively active RhoA or inactivating GAP domain blocked such effect, whereas replacing the BCH domain with endosome-targeting SNX3 excluded requirement of endosomal localization in regulating the GAP activity. Substitution with homologous BCH domain from Schizosaccharomyces pombe, which does not bind mammalian RhoA, also led to complete loss of suppression. Interestingly, the p50RhoGAP BCH domain only targeted RhoA, but not Cdc42 or Rac1, and it was unable to distinguish between GDP and the GTP-bound form of RhoA. Further mutagenesis revealed a RhoA-binding motif (residues 85-120), which when deleted, significantly reduced BCH inhibition on GAP-mediated cell rounding, whereas its full suppression also required an intramolecular interaction motif (residues 169-197). Therefore, BCH domain serves as a local modulator in cis to sequester RhoA from inactivation by the adjacent GAP domain, adding to a new paradigm for regulating p50RhoGAP signaling.

Pro-apoptotic activity of mBNIP-21 depends on its BNIP-2 and Cdc42GAP homology (BCH) domain and is enhanced by coxsackievirus B3 infection

Our previous study reported that mouse BNIP-21 (mBNIP-21) induces apoptosis through a mitochondria-dependent pathway. To map the functional domains of mBNIP-21, we performed mutational analyses and demonstrated that the BNIP-2 and Cdc42GAP homology (BCH) domain is required for apoptosis induction by mBNIP-21 targeting the mitochondria and inducing cytochrome c release. This pro-apoptotic activity was enhanced by coxsackievirus infection. However, deletion of the Bcl-2 homology 3 (BH3)-like domain, a well-known cell 'death domain' in proapoptotic Bcl-2 family proteins, did not affect the activity of mBNIP-21. These data were further supported by transfection of a mouse Bax (mBax) mutant, whose BH3 was replaced by the mBNIP-21 BH3-like domain. This replacement significantly reduced the pro-apoptotic activity of mBax. We also found that the predicted calcium binding domain has no contribution to the mBNIP-21-induced apoptosis. Further mapping of the motifs of BCH domain demonstrated that deletion of the hydrophobic motif proximal to the C-terminal of the BCH significantly reduced its proapoptotic activity. These findings suggest that mBNIP-21, as a member of the BNIP subgroup of the Bcl-2-related proteins, functions without need of BH3 but its BCH domain is critical for its activity in inducing cell elongation, membrane protrusions and apoptotic cell death.

Cayman ataxia protein caytaxin is transported by kinesin along neurites through binding to kinesin light chains

Deficiency of caytaxin results in hereditary ataxia or dystonia in humans, mice and rats. Our yeast two-hybrid screen identified kinesin light chains (KLCs) as caytaxin-binding proteins. The tetratricopeptide-repeat region of KLC1 recognizes the ELEWED sequence (amino acids 115-120) of caytaxin. This motif is conserved among BNIP-2 family members and other KLC-interacting kinesin cargo proteins such as calsyntenins. Caytaxin associates with kinesin heavy chains (KHCs) indirectly by binding to KLCs, suggesting that caytaxin binds to the tetrameric kinesin molecule. In cultured hippocampal neurons, we found that caytaxin is distributed in both axons and dendrites in punctate patterns, and it colocalizes with microtubules and KHC. GFP-caytaxin expressed in hippocampal neurons is transported at a speed (∼1 μm/second) compatible with kinesin movement. Inhibition of kinesin-1 by dominant-negative KHC decreases the accumulation of caytaxin in the growth cone. Caytaxin puncta do not coincide with vesicles containing known kinesin cargos such as APP or JIP-1. A part of caytaxin, however, colocalizes with mitochondria and suppression of caytaxin expression by RNAi redistributes mitochondria away from the distal ends of neurites. These data indicate that caytaxin binds to kinesin-1 and functions as an adaptor that mediates intracellular transport of specific cargos, one of which is the mitochondrion.

A Cdo-Bnip-2-Cdc42 signaling pathway regulates p38alpha/beta MAPK activity and myogenic differentiation

The p38α/β mitogen-activated protein kinase (MAPK) pathway promotes skeletal myogenesis, but the mechanisms by which it is activated during this process are unclear. During myoblast differentiation, the promyogenic cell surface receptor Cdo binds to the p38α/β pathway scaffold protein JLP and, via JLP, p38α/β itself. We report that Cdo also interacts with Bnip-2, a protein that binds the small guanosine triphosphatase (GTPase) Cdc42 and a negative regulator of Cdc42, Cdc42 GTPase-activating protein (GAP). Moreover, Bnip-2 and JLP are brought together through mutual interaction with Cdo. Gain- and loss-of-function experiments with myoblasts indicate that the Cdo–Bnip-2 interaction stimulates Cdc42 activity, which in turn promotes p38α/β activity and cell differentiation. These results reveal a previously unknown linkage between a cell surface receptor and downstream modulation of Cdc42 activity. Furthermore, interaction with multiple scaffold-type proteins is a distinctive mode of cell surface receptor signaling and provides one mechanism for specificity of p38α/β activation during cell differentiation.

Nerve growth factor stimulates interaction of Cayman ataxia protein BNIP-H/Caytaxin with peptidyl-prolyl isomerase Pin1 in differentiating neurons

Mutations in ATCAY that encodes the brain-specific protein BNIP-H (or Caytaxin) lead to Cayman cerebellar ataxia. BNIP-H binds to glutaminase, a neurotransmitter-producing enzyme, and affects its activity and intracellular localization. Here we describe the identification and characterization of the binding between BNIP-H and Pin1, a peptidyl-prolyl cis/trans isomerase. BNIP-H interacted with Pin1 after nerve growth factor-stimulation and they co-localized in the neurites and cytosol of differentiating pheochromocytoma PC12 cells and the embryonic carcinoma P19 cells. Deletional mutagenesis revealed two cryptic binding sites within the C-terminus of BNIP-H such that single point mutants affecting the WW domain of Pin1 completely abolished their binding. Although these two sites do not contain any of the canonical Pin1-binding motifs they showed differential binding profiles to Pin1 WW domain mutants S16E, S16A and W34A, and the catalytically inert C113A of its isomerase domain. Furthermore, their direct interaction would occur only upon disrupting the ability of BNIP-H to form an intramolecular interaction by two similar regions. Furthermore, expression of Pin1 disrupted the BNIP-H/glutaminase complex formation in PC12 cells under nerve growth factor-stimulation. These results indicate that nerve growth factor may stimulate the interaction of BNIP-H with Pin1 by releasing its intramolecular inhibition. Such a mechanism could provide a post-translational regulation on the cellular activity of BNIP-H during neuronal differentiation.

BNIP2 extra long inhibits RhoA and cellular transformation by Lbc RhoGEF via its BCH domain

Increased expression of BCH-motif-containing molecule at the C-terminal region 1 (BMCC1) correlates with a favourable prognosis in neuroblastoma, but the underlying mechanism remains unknown. We here isolated BNIPXL (BNIP2 Extra Long) as a single contig of the extended, in-vitro-assembled BMCC1. Here, we show that in addition to homophilic interactions, the BNIP2 and Cdc42GAP homology (BCH) domain of BNIPXL interacts with specific conformers of RhoA and also mediates association with the catalytic DH-PH domains of Lbc, a RhoA-specific guanine nucleotide exchange factor (RhoGEF). BNIPXL does not recognize the constitutive active G14V and Q63L mutants of RhoA but targets the fast-cycling F30L and the dominant-negative T19N mutants. A second region at the N-terminus of BNIPXL also targets the proline-rich region of Lbc. Whereas overexpression of BNIPXL reduces active RhoA levels, knockdown of BNIPXL expression has the reverse effect. Consequently, BNIPXL inhibits Lbc-induced oncogenic transformation. Interestingly, BNIPXL can also interact with RhoC, but not with RhoB. Given the importance of RhoA and RhoGEF signaling in tumorigenesis, BNIPXL could suppress cellular transformation by preventing sustained Rho activation in concert with restricting RhoA and Lbc binding via its BCH domain. This could provide a general mechanism for regulating RhoGEFs and their target GTPases.

Cleavage of BNIP-2 and BNIP-XL by caspases

BNIP-2 and BNIP-XL are BCH domain-containing proteins that are implicated in programmed cell death. It has been reported that overexpression of BNIP-2 in neuroblastoma cell lines resulted in massive cell death, whereas BNIP-XL was upregulated during NGF-depletion-induced apoptosis in neuroblastoma and was involved in the regulation of differentiation, survival, and aggressiveness of tumor cells. Despite their importance in apoptosis, our understanding of BNIP-2 containing proteins is limited. In this communication, we demonstrate that both BNIP-2 and BNIP-XL are cleaved by caspases during apoptosis. Significantly, the caspase cleavage sites on BNIP-2 are located on its N-terminal EF-hand motif, while that on BNIP-XL is located upstream of the C-terminal BCH domain. Our results suggest that the caspase-mediated cleavage of BNIP-2 and BNIP-XL could result in the release of the BCH domain or smaller fragments that are crucial for their proapoptotic activities.

Expression and localization of Cayman ataxia-related protein, Caytaxin, is regulated in a developmental- and spatial-dependent manner

Mutation of the gene encoding Caytaxin causes human Cayman ataxia by interfering with normal splicing and, in mutant rodents, by reducing normal transcription, which leads to ataxia, dystonia, and mental retardation: These observations suggest that Caytaxin may be crucial for higher brain functions such as motor learning. We generated antibodies against mouse Caytaxin. Interestingly, we found that the expression of Caytaxin is regulated during brain development while quantitative real time RT-PCR indicated that the mRNA level did not change between postnatal days 7 (P7) and P14 in the cerebellum and hippocampus, implying that the expression of Caytaxin may be controlled by a post-transcriptional mechanism. Immunostaining analyses demonstrated that Caytaxin was localized in many brain areas including the cerebellum and hippocampus. Furthermore, Caytaxin was localized to the presynaptic cytosol by the subcellular fractionation of mouse brain and an observation that was confirmed by co-localization studies with synapsin I and VGLUT1. The above data, disease phenotypes, and mutant animals suggest that Caytaxin may be essential for synaptic function. Thus, identifying the role of Caytaxin in synapse maturation may lead to the development of therapeutic interventions for cerebellar ataxia as well as mental disorders.

Brain-specific BNIP-2-homology protein Caytaxin relocalises glutaminase to neurite terminals and reduces glutamate levels

Human Cayman ataxia and mouse or rat dystonia are linked to mutations in the genes ATCAY (Atcay) that encode BNIP-H or Caytaxin, a brain-specific member of the BNIP-2 family. To explore its possible role(s) in neuronal function, we used protein precipitation and matrix-assisted laser desorption/ionisation mass spectrometry and identified kidney-type glutaminase (KGA) as a novel partner of BNIP-H. KGA converts glutamine to glutamate, which could serve as an important source of neurotransmitter. Co-immunoprecipitation with specific BNIP-H antibody confirmed that endogenous BNIP-H and KGA form a physiological complex in the brain, whereas binding studies showed that they interact with each other directly. Immunohistochemistry and in situ hybridisation revealed high BNIP-H expression in hippocampus and cerebellum, broadly overlapping with the expression pattern previously reported for KGA. Significantly, BNIP-H expression was activated in differentiating neurons of the embryonic carcinoma cell line P19 whereas its overexpression in rat pheochromocytoma PC12 cells relocalised KGA from the mitochondria to neurite terminals. It also reduced the steady-state levels of glutamate by inhibiting KGA enzyme activity. These results strongly suggest that through binding to KGA, BNIP-H could regulate glutamate synthesis at synapses during neurotransmission. Thus, loss of BNIP-H function could render glutamate excitotoxicity or/and deregulated glutamatergic activation, leading to ataxia, dystonia or other neurological disorders.

Functional network analysis reveals extended gliomagenesis pathway maps and three novel MYC-interacting genes in human gliomas

Gene expression profiling has proven useful in subclassification and outcome prognostication for human glial brain tumors. The analysis of biological significance of the hundreds or thousands of alterations in gene expression found in genomic profiling remains a major challenge. Moreover, it is increasingly evident that genes do not act as individual units but collaborate in overlapping networks, the deregulation of which is a hallmark of cancer. Thus, we have here applied refined network knowledge to the analysis of key functions and pathways associated with gliomagenesis in a set of 50 human gliomas of various histogenesis, using cDNA microarrays, inferential and descriptive statistics, and dynamic mapping of gene expression data into a functional annotation database. Highest-significance networks were assembled around the myc oncogene in gliomagenesis and around the integrin signaling pathway in the glioblastoma subtype, which is paradigmatic for its strong migratory and invasive behavior. Three novel MYC-interacting genes (UBE2C, EMP1, and FBXW7) with cancer-related functions were identified as network constituents differentially expressed in gliomas, as was CD151 as a new component of a network that mediates glioblastoma cell invasion. Complementary, unsupervised relevance network analysis showed a conserved self-organization of modules of interconnected genes with functions in cell cycle regulation in human gliomas. This approach has extended existing knowledge about the organizational pattern of gene expression in human gliomas and identified potential novel targets for future therapeutic development.

BNIP-Sα induces cell rounding and apoptosis by displacing p50RhoGAP and facilitating RhoA activation via its unique motifs in the BNIP-2 and Cdc42GAP homology domain

Changes in cell morphology are linked to many cellular events including cytokinesis, differentiation, migration and apoptosis. We recently showed that BNIP-Salpha induced cell rounding that leads to apoptosis via its BNIP-2 and Cdc42GAP Homology (BCH) domain, but the underlying mechanism has not been determined. Here, we have identified a unique region (amino acid 133-177) of the BNIP-Salpha BCH domain that targets RhoA, but not Cdc42 or Rac1 and only the dominant-negative form of RhoA could prevent the resultant cell rounding and apoptotic effect. The RhoA-binding region consists of two parts; one region (residues 133-147) that shows some homology to part of the RhoA switch I region and an adjacent sequence (residues 148-177) that resembles the REM class I RhoA-binding motif. The sequence 133-147 is also necessary for its heterophilic interaction with the BCH domain of the Rho GTPase-activating protein, p50RhoGAP/Cdc42GAP. These overlapping motifs allow tripartite competition such that overexpression of BNIP-Salpha could reduce p50RhoGAP binding to RhoA and restore RhoA activation. Furthermore, BNIP-Salpha mutants lacking the RhoA-binding motif completely failed to induce cell rounding and apoptosis. Therefore, via unique binding motifs within its BCH domain, BNIP-Salpha could interact and activate RhoA while preventing its inhibition by p50RhoGAP. This concerted mechanism could allow effective propagation of the RhoA pathway for cell rounding and apoptosis.