Alternative splicing of Bcl-2-related genes: functional consequences and potential therapeutic applications.

Apoptosis is a morphologically distinct form of cell death. It is executed and regulated by several groups of proteins. Bcl-2 family proteins are the main regulators of the apoptotic process acting either to inhibit or promote it. More than 20 members of the family have been identified so far and most have two or more isoforms. Alternative splicing is one of the major mechanisms providing proteomic complexity and functional diversification of the Bcl-2 family proteins. Pro- and anti-apoptotic Bcl-2 family members should function in harmony for the regulation of the apoptosis machinery, and their relative levels are critical for cell fate. Any mechanism breaking down this harmony by changing the relative levels of these antagonistic proteins could contribute to many diseases, including cancer and neurodegenerative disorders. Recent studies have shown that manipulation of the alternative splicing mechanisms could provide an opportunity to restore the proper balance of these regulator proteins. This review summarises current knowledge on the alternative splicing products of Bcl-2-related genes and modulation of splicing mechanisms as a potential therapeutic approach.

BCL-2 family expression in human neutrophils during delayed and accelerated apoptosis.

The human neutrophil spontaneously undergoes apoptosis, but this type of cell death can be delayed or accelerated by a wide variety of agents. There are wide discrepancies in the literature regarding the expression of the Bcl-2 family of proteins in human neutrophils. Here, we show that A1, Mcl-1, Bcl-X(L), and Bad are major transcripts in human neutrophils and that levels of these transcripts are cytokine regulated. However, no Bcl-X(L) protein was detected in Western blots. Protein levels for the proapoptotic proteins Bad, Bax, Bak, and Bik remained constant during culture, despite changes in the levels of mRNA for these gene products. These proapoptotic proteins were extremely stable, having very long half-lives. In contrast, A1 and Mcl-1 transcripts were extremely unstable (with approximately 3-h half-lives), and Mcl-1 protein was also subject to rapid turnover. These results indicate that neutrophil survival is regulated by the inducible expression of the short-lived Mcl-1 and possibly the A1 gene products. In the absence of their continued expression, these prosurvival gene products are rapidly turned over, and then the activity of the stable death proteins predominates and promotes apoptosis.

Molecular control of neutrophil apoptosis.

Human neutrophils constitutively undergo apoptosis and this process is critical for the resolution of inflammation. Whilst neutrophil apoptosis can be modulated by a wide variety of agents including GM-CSF, LPS and TNF-alpha, the molecular mechanisms underlying neutrophil death and survival remain largely undefined. Recent studies have shown the involvement of members of the Bcl-2 protein family (especially Mcl-1 and A1) and caspases in the regulation and execution of neutrophil apoptosis. Cell surface receptors and protein kinases, particularly mitogen-activated protein kinases, also play critical roles in transducing the signals that result in neutrophil apoptosis or extended survival. This review summarises current knowledge on the molecular mechanisms and components of neutrophil apoptosis.

Apoptosis is rapidly triggered by antisense depletion of MCL-1 in differentiating U937 cells.

Mcl-1 is a member of the Bcl-2 protein family, which has been shown to delay apoptosis in transfection and/or overexpression experiments. As yet no gene knockout mice have been engineered, and so there is little evidence to show that loss of Mcl-1 expression is sufficient to trigger apoptosis. U937 cells constitutively express the antiapoptotic protein Bcl-2; but during differentiation, in response to the phorbol ester PMA (phorbol 12 beta-myristate 13 alpha-acetate), Mcl-1 is transiently induced. The purpose of this investigation was to determine the functional role played by Mcl-1 in this differentiation program. Mcl-1 expression was specifically disrupted by chimeric methylphosphonate/phosphodiester antisense oligodeoxynucleotides to just 5% of control levels. The depletion of Mcl-1 messenger RNA (mRNA) and protein was both rapid and specific, as indicated by the use of control oligodeoxynucleotides and analysis of the expression of other BCL2 family members and PMA-induced tumor necrosis factor-alpha (TNF-alpha). Specific depletion of Mcl-1 mRNA and protein, in the absence of changes in cellular levels of Bcl-2, results in a rapid entry into apoptosis. Levels of the proapoptotic protein Bax remained unchanged during differentiation, while Bak expression doubled within 24 hours. Apoptosis was detected within 4 hours of Mcl-1 antisense treatment by a variety of parameters including a novel live cell imaging technique allowing correlation of antisense treatment and apoptosis in individual cells. The induction of Mcl-1 is required to prevent apoptosis during differentiation of U937 cells, and the constitutive expression of Bcl-2 is unable to compensate for the loss of Mcl-1. (Blood. 2000;96:1756-1763)

In vivo localisation and stability of human Mcl-1 using green fluorescent protein (GFP) fusion proteins.

Mcl-1 is an anti-apoptotic member of the Bcl-2 family of proteins. We have expressed full length and mutated GFP:Mcl-1 fusion proteins to define structural motifs that control protein localisation and stability. When expressed in U-937 cells, full length Mcl-1 locates primarily within mitochondria and its half-life was approximately 3 h, which was identical to the native, endogenously expressed protein. When the terminal 20 amino acids from the C-terminus of the protein were detected, the protein was diffused in the cytoplasm, but its stability was unaffected. This confirms that this region is responsible for efficient targeting to mitochondria. Surprisingly, deletion of 104 amino acids (residues 79-183) that contain putative PEST sequences and other stability regulating motifs, did not affect protein stability.

Regulation of neutrophil FcgammaRIIIb (CD16) surface expression following delayed apoptosis in response to GM-CSF and sodium butyrate.

When neutrophils undergo apoptosis, they lose expression of the surface receptor CD16 (FcgammaRIIIb). Thus levels of surface CD16 are good indicators of apoptotic or non-apoptotic neutrophils. Shedding of CD16 occurs via the activity of a metalloproteinase that cleaves the receptor from the plasma membrane. Granulocyte-macrophage colony-stimulating factor (GM-CSF) and sodium butyrate both stimulate neutrophil gene expression, protect these cells from apoptosis, and maintain expression of surface CD16. In this report we have investigated whether these agents maintain surface expression of CD16 via (1) decreased shedding (2) increased mobilization of the internal pool of pre-formed CD16, or (3) via de novo biosynthesis of new receptor molecules. Although GM-CSF and sodium butyrate both preserved surface expression of CD16, GM-CSF actually accelerated the rate of shedding of this receptor. Maintenance of surface levels was achieved by substantial mobilization of the internal pool of CD16. Sodium butyrate, on the other hand, maintained surface expression without extensive store depletion via a mechanism that appeared to involve a decreased rate of shedding. In these experiments we could find no evidence for de novo biosynthesis of CD16 stimulated by either GM-CSF or sodium butyrate. These experiments indicate that multiple mechanisms exist for the maintenance of surface CD16 during rescue of neutrophils from apoptosis by different agents.

In vitro effects of GM-CSF on mature peripheral blood neutrophils.

GM-CSF can play a crucial role in regulating the neutrophil-mediated inflammatory response. This growth factor is a proliferative stimulus for bone marrow neutrophil stem cell precursors and has at least 3 important roles in regulating neutrophil-mediated immunity: a) a direct effect on the proliferation and development of neutrophil progenitors; b) synergistic activity with other haemopoietic growth factors; c) stimulation of the functional activity of mature neutrophils. The production of GM-CSF may be triggered directly by exogenous factors such as antigens and endotoxins, or indirectly through the release of cytokines by a variety of cells including lymphocytes, activated macrophages and endothelial cells exposed to products of mononuclear phagocytes. Such production of GM-CSF may serve to quickly release mature neutrophils from the bone marrow in response to infections. Moreover, enhancement of the function of mature neutrophils may also augment their ability to migrate to infective sites and then phagocytose and kill pathogens. Increased expression of CD11b/CD18 may play a fundamental part in this mechanism because this receptor is essential for the adhesion of neutrophils to the endothelium. Both phagocytosis and oxidative burst activity increase as a result of the action of GM-CSF and the increased expression of complement- and Fc-receptors can augment opsono-phagocytosis. A further level of neutrophil up-regulation occurs by increasing the functional life span of neutrophils by GM-CSF. Thus, by delaying neutrophil apoptosis, GM-CSF greatly extends the time over which neutrophils may function at inflammatory sites. GM-CSF can thus exert a variety of important regulatory controls of neutrophil function during bacterial infections. Both the number and the functional status of neutrophils is highly regulated by GM-CSF. It is also possible that GM-CSF produced within localised sites of acute inflammation or infection may attract, trap and then activate neutrophils within this site.

Mcl-1 expression in human neutrophils: regulation by cytokines and correlation with cell survival.

Human neutrophils possess a very short half-life because they constitutively undergo apoptosis. Cytokines, such as granulocyte-macrophage colony-stimulating factor (GM-CSF), and other agents can rescue neutrophils from apoptosis but the molecular mechanisms involved in this rescue are undefined. Here, we show by Western blotting that human neutrophils do not express Bcl-2 or Bcl-X but constitutively express Bax. However, cellular levels of these proteins are unaffected by agents which either accelerate or delay neutrophil apoptosis. In contrast, neutrophils express the antiapoptotic protein Mcl-1 and levels of this protein correlate with neutrophil survival. Thus, cellular levels of Mcl-1 decline as neutrophils undergo apoptosis and are enhanced by agents (eg, GM-CSF, interleukin-1beta, sodium butyrate, and lipopolysaccharide) that promote neutrophil survival. Neutrophils only possess few, small mitochondria, and much of the Mcl-1 protein seems to be located in nuclear fractions. These observations provide the first evidence implicating a Bcl-2 family member in the regulation of neutrophil survival. Moreover, this work also provides a potential mechanism whereby cytokine-regulated gene expression regulates the functional lifespan of neutrophils and hence their ability to function for extended time periods during acute inflammation.

Activation of human neutrophils by soluble immune complexes: role of Fc gamma RII and Fc gamma RIIIb in stimulation of the respiratory burst and elevation of intracellular Ca2+.

Activation of control, unprimed neutrophils with soluble immune complexes fails to generate a respiratory burst. However, if the cells are primed with either tumor necrosis factor-alpha or granulocyte-macrophage colony-stimulating factor prior to addition of soluble immune complexes, then a rapid and transient burst of reactive oxidant secretion is observed. In unprimed neutrophils the soluble immune complexes stimulate an intracellular Ca2+ transient that arises from the mobilization of intracellular Ca2+. However, in primed cells, an "extra" intracellular Ca2+ signal is observed that arises from Ca2+ influx. After removal of Fc gamma RIIIb by treatment with pronase or PI-PLC, the soluble immune complexes fail to activate a respiratory burst in unprimed neutrophils and the "extra" Ca2+ signal is not observed. These results indicate that during priming Fc gamma RIIIb becomes functionally activated and thence its ligation leads to stimulated Ca2+ influx and the generation of intracellular signals that lead to NADPH oxidase activation. Experiments using Fab/F(ab')2 fragments to specifically crosslink either Fc gamma RII or Fc gamma RIIIb and experiments with neutrophils from an individual with Fc gamma RIIIb gene deficiency confirm this important function for Fc gamma RIIIb in neutrophil activation.