Douglas Black, Ph.D
Department of Microbiology, Immunology, and Molecular Genetics
MRL 6-780 | MRL 6-567 (lab)
The Regulation of Neuronal Gene Expression through Alternative Pre-mRNA SplicingOur lab is interested in the regulation of pre-mRNA splicing and the biochemical
mechanisms that control changes in splice sites. The sequences of metazoan
Alternative splicing is particularly common in genes expressed in the mammalian nervous system, where many proteins important for neuronal differentiation and function are made in diverse isoforms through controlled changes in splicing. Our lab works on a range of projects related to the control of pre-mRNA splicing in neurons. We aim to determine the mechanisms of action of splicing regulators and to understand their roles in neural development. We are focused on four regulatory factors: Polypyrimidine Tract Binding Protein (PTB), neuronal PTB, Fox-1, and Fox-2. PTB and nPTB are primarily splicing repressors, while Fox-1 and Fox-2 act to enhance splicing. Each of these proteins alters the splicing of a specific set of exons within the genome. In mechanistic studies, we examine the RNA binding properties of these proteins and analyze how they can alter spliceosome assembly. A second effort uses cell culture models and conditional knockout mice to understand how these proteins affect neuronal development. A final area of interest is in understanding how cell signaling pathways impact the splicing reaction. This project focuses on the effect of cell excitation on the splicing of ion channel transcripts and the role of this splicing in neuronal plasticity.
Mechanisms of Splicing Repression and Derepression
Our early work used the neural-specific N1 exon of the c-src gene as a model for a tissue specific splicing event. Using a variety of approaches, we dissected the molecular components that determine c-src N1 exon splicing and its repression in nonneuronal cells. Analyses of the cis-acting RNA elements controlling the exon demonstrated that the tissue specificity of N1 derives from a combination of enhancer and repressor sequences in the surrounding RNA. In neurons, an enhancer sequence in the downstream intron activates splicing of the exon. Splicing is specifically repressed in non-neuronal cells by sequences in the upstream 3' splice site and within the enhancer region. This combination of positive and negative regulation has proven to be a common feature in systems of tissue-specific splicing.
We use an in vitro splicing system developed in the lab to study the molecules that regulate N1 splicing. Splicing of N1 is repressed in nuclear extracts of non-neuronal HeLa cells. In contrast, splicing proceeds in extracts of WERI-1 retinoblastoma cells. Using this system, we have made particular progress in understanding the proteins that mediate repression of N1 exon splicing. This repression requires CUCUCU repressor elements that are bound by the polypyrimidine tract–binding protein (PTB). Depleting PTB from HeLa extracts using a specific antibody, or from cells by RNA interference, shows that PTB is required to prevent N1 exon splicing. PTB-binding sites both upstream and downstream from N1 are needed for splicing repression, and the presence of the downstream sites stabilizes PTB binding to the upstream sites. This implies that PTB assembles a cooperative repressor complex that bridges the exon. We are currently analyzing the structure of this complex and its interactions with the RNA. Fred Allain’s lab (ETH, Zurich) has shown by NMR spectroscopy that all four domains of PTB each bind a pyrimidine triplet. However, the orientation of two of these domains indicates that at least one of these triplets cannot be contiguous with the other three. This ability to form an RNA loop, which fits well with the placement of the repressor elements surrounding N1 and other exons regulated by PTB, is being incorporated into models of the PTB/RNA interaction needed for splicing repression.
We have also taken a proteomic approach to understanding the mechanism of PTB mediated splicing repression. We find that PTB blocks splicing at an early stage of spliceosome assembly, prior to the formation of what is called the E complex. The large pre-mRNP complexes that lead to the E complex were purified and their components were identified by mass spectrometry. Under conditions of PTB dependent splicing repression, we find that the essential splicing factor U2AF is blocked from assembling into the spliceosome. Under the conditions of our experiments, U2AF requires an interaction with components at the 5’ splice site to assemble into the E complex. Thus, PTB may be specifically blocking this interaction, a possibility we are currently investigating. We are very interested in these and other results in the field, which indicate that PTB may interact with a specific spliceosomal target in repressing splicing.
To repress splicing, CUCUCU elements that bind PTB are needed both upstream and downstream of the N1 exon. These elements are apparently sufficient to determine the differential splicing between HeLa and WERI-1 extracts. However in other cells, additional sequences are required for proper regulation of the exon. The downstream element UGCAUG binds the Fox proteins (see below), and is essential for enhanced splicing in neurons. The downstream region also contains additional regulatory elements that assemble a multi-protein complex comprised of the hnRNP H and KSRP proteins, as well as PTB. We are currently working to understand the roles of these other factors and their interplay with the Fox and PTB proteins. The structures of pre-mRNP complexes and the nature of their assembly are an understudied problem. In addition to elucidating how these RNA/Protein complexes affect splicing, we hope our biochemical analyses will shed light on the general properties of these structures that affect many different RNA-dependent processes.
Cellular Functions of Splicing Regulators
In WERI-1 extract, where PTB activity is reduced, an additional protein is present called nPTB. Neuronal PTB is 74% identical to PTB in sequence, but is specifically expressed in neurons in the brain, whereas PTB is mostly excluded from neurons and expressed in glia. Neuronal PTB binds to the same CUCUCU regulatory elements as PTB, but is apparently neutral for N1 exon splicing. It does not repress splicing of neuronal exons in vitro and only mildly affects their splicing in vivo. In contrast, other exons seem to be equally repressed by PTB and nPTB. In addition to the cellular activities nPTB and its effects on splicing in vitro, we are studying how and when the choice is made during neuronal differentiation to express PTB or nPTB. Most interestingly, we find that nPTB expression is regulated post-transcriptionally. NPTB mRNA is present in most if not all cells, but this mRNA produces protein only in specific cell types, notably neurons and testis. In many cells, nPTB expression is repressed by PTB protein, presenting an interesting cross-regulation of one protein by the other. In other cells, specific microRNAs repress nPTB expression. We are currently characterizing this translational regulation of nPTB in more detail.
To characterize the roles of PTB and nPTB in neuronal cell biology, we are developing mouse strains carrying alleles of these genes that can be conditionally knocked out in particular cell lineages. PTB and nPTB apparently regulate overlapping but not identical sets of exons. To identify these exon sets and examine their cellular roles, we and others are developing microarrays of splice-junction-specific oligonucleotides to assay large numbers of alternative exons in parallel. We are using these arrays in conjunction with the gene knockouts in mice, and with RNA interference mediated PTB knockdown, to identify large sets of PTB regulated exons. The ability to do genome wide analyses of alternative splicing will help us in a number of projects.
The Fox proteins are cell-type specific splicing enhancers.
The enhancer region downstream of N1 is a complex array of RNA regulatory elements that are required both for splicing enhancement and for splicing repression. The most significant element for the enhancer activity is the hexanucleotide UGCAUG, which is also essential for other splicing enhancers and is found downstream of a whole set of neuronally-regulated exons. This key hexanucleotide element is recognized by several mammalian homologs of the C. elegans Fox-1 protein. We found that these RNA binding proteins are strong activators of splicing for certain neuronal exons. Through the use of multiple promoters and alternative splicing events, the mouse Fox-1 and Fox-2 genes each encode a large family of related proteins that are expressed in muscle, heart, and neurons. Interestingly, the human homolog of Fox-1 (called A2BP1 for Ataxin 2 binding protein) is implicated in cerebellar ataxias, as well as other forms of neurological disease. These disorders may arise from problems of splicing control. Thus, in addition to their mechanisms of splicing enhancement, we are extremely interested in the cellular roles of the Fox proteins and how Fox regulated exons might affect neuronal development and function. Using similar approaches to nPTB, we are studying the Fox proteins in cell culture and in conditional knockout mice.
The Regulation of Ion Channel Alternative Splicing by Cell Stimulation
A poorly understood aspect of splicing regulation is how extracellular stimuli and cell signaling pathways direct changes in splicing. This is especially interesting in the nervous system, where many proteins that determine neuronal excitation are modulated by alternative splicing. To study how specific signaling cascades alter splicing patterns, we examined the regulation of the STREX exon in the calcium-activated potassium channel transcript (BK channel). BK channels are important in shaping the action potential kinetics of a number of excitable cells and are highly regulated at the level of splicing. The STREX exon adds a peptide insert into the calcium-binding domain of the channel that both makes it more sensitive to intracellular calcium and voltage and allows its modulation by Protein Kinase A. We found that culturing excitable cells in depolarizing media containing high potassium led to the repression of STREX exon inclusion. This reduction in STREX splicing is blocked by the CaM Kinase inhibitor KN93 and by the L-type calcium channel blocker nifedipine, implicating calcium signaling in this splicing repression. As a model for this regulation, we expressed the STREX exon from a transfected minigene and found that STREX splicing is repressed by a cotransfected constitutive CaMK IV. This approach allowed us to map specific RNA regulatory elements in the BK channel transcript needed to respond to the CaMK signal. We identified a calcium-responsive RNA element (CaRRE-1) that can confer CaMK-dependent repression on a heterologous exon. We are very interested in the mechanism of this repression and how it responds to culture conditions.
The CaMK effect on ion channel splicing has implications for the control of neuronal excitability. Alternative splicing alters the activity of many other proteins that mediate the induction and propagation of action potentials. If splicing is controlled by cell excitation, then these splicing events are likely one component of the plastic changes in cell activity that underlie many aspects of neurophysiology. Additional CaRRE elements were also found in two regulated exons of the NMDA Receptor 1, which are also regulated by CaMK IV. NR1 exon 21 contains a CaRRE element within the exon itself, instead of the 3’ splice site, and encodes a peptide that controls the trafficking and assembly of the NR1 subunit. Exon 21 also contains an additional Calcium responsive element, different from the CaRRE-1, which we are currently characterizing. These elements in ion channel exons provide a unique entry point for the dissection of signaling pathways that impinge on the splicing apparatus, and we are working to identify the proteins that bind to them. In an alternative approach, we are studying whether known splicing regulators can be altered in expression or activity by specific signaling pathways. For example, we have found that the nucleo/cytoplasmic distribution of PTB is regulated by phosphorylation by Protein Kinase A. Our goal in these projects is to understand how the information of an extracellular stimulus can be transmitted to the cell nucleus to alter splicing.
For more information, please visit the Black Lab Website