Nucleo-cytoplasmic actin relationships in Stockholm

Abstract

The Wenner-Gren Foundations symposium “Actin and Actin-associated Proteins from Genes to Polysomes” took place at the Wenner-Gren Center in Stockholm, Sweden, on September 7–10, 2011. As the name of the symposium implied, the organizing committee, consisting of local organizers Piergiorgio Percipalle, Neus Visa and Ann Kristin Östlund Farrants from Stockholm and Thoru Pederson from Worcester, MA USA, had boldly decided to embrace the unconventional roles of actin, namely its connections to the gene expression apparatus all the way from the nuclear genes to the cytoplasmic protein synthesis machineries. The organizers assembled a respectable crowd of 25 speakers with very diverse backgrounds, but a common interest in understanding how actin and its associated factors may function outside their conventional roles in the cytoskeleton. As many people presented unpublished work, I will not go into detail of these talks but will rather aim to highlight the discussed topics and their possible implications to this exciting research field.

Keywords: :

• actin
• nucleoskeleton
• nucleus
• polysomes
• transcription

Our journey following actin from the gene to the polysomes began rightfully with a talk from the grand lady in the nuclear actin field, Brigitte Jockusch from Braunschweig, Germany. She gave a concise summary of the history of this research topic, which has been full of controversy. Actin was first identified in the nucleus almost 50 years ago, but for a long time serious doubts about the whole concept of nuclear actin hampered the development of the field. Dr Jockusch highlighted the main sources of skepticism, namely the concerns about cytoplasmic actin contaminating nuclear fractions and the inability to detect canonical actin filaments in the nucleus by phalloidin staining—a hallmark of cytoplasmic filamentous actin. Many of the early concerns have mostly evaporated due to technical advances, and it is nowadays widely accepted that actin is indeed present in the nucleus, but the absence of “normal” actin filaments in the nucleus still plagues us. Most of the traditional roles of actin in the cytoplasm, such as the interplay between myosin motors during muscle contraction or promotion of membrane protrusion during cell motility, rely on either actin filaments or their polymerization process, respectively. Therefore the apparent absence of actin filaments in the nucleus has been rather puzzling. An active research area has therefore been the development of different probes and techniques to visualize the functional form of actin in the nucleus. In this regard, Cora-Ann Schoenenberger from Basel presented her latest efforts in developing actin antibodies that recognize specific forms of actin, such as the lower-dimer, which is an antiparallel dimer of actin monomers they found associated with the endocytic pathway, but also in the nucleus.¹ Dyche Mullins from San Francisco told us about their elegant approach of using actin-binding domains as probes for different actin structures within the nucleus. His studies strongly imply that there is some type of polymeric actin also in the cell nucleus, and his probes will be a valuable asset for the field in the future. As nuclear actin clearly has some characteristics that are different from the cytoplasmic actin networks, it is easy to draw parallels to other non-conventional actins, such as those found in bacteria. Robert Robinson from Singapore reminded us of the different ways of building an actin filament by using both bacterial and eukaryotic actins as models. He suggested that the differences in prokaryotic and eukaryotic actin designs are due to the evolution of actin-binding proteins, which have preserved the filament structure in eukaryotes but not in prokaryotes. Evolution of the actin cytoskeleton was also discussed by Richard Meagher from Athens, GA USA, who described the general competence of the different actins to fulfil each other’s roles, and thereby the functional significance of the different actins in a multicellular organism.

One further source of skepticism toward nuclear actin has been the lack of clear functions. Despite the fact that Egly et al. suggested already 25 years ago that actin may play a role in transcription,² the functional significance of nuclear actin has only emerged during the last ten years. Even if we do not still completely understand the structural basis of nuclear actin, a growing number of functions especially at the level of the gene have been linked to it, and many findings point toward the requirement for polymeric actin in these processes. In my own talk, I summarized our recent findings on how actin shuttles in and out of the nucleus. Identification of the proteins mediating these processes has allowed us to show that the maintenance of the nuclear actin levels by the active transport process is absolutely required for the general transcriptional activity of the cell. The role of actin in a more specialized case of transcription was also covered by Kei Miyamoto from Cambridge, who described their recent study on transcriptional reprogramming in Xenopus oocytes, and highlighted the need for regulated actin polymerization in this process.³ This feature seems to be shared also with other nuclear processes, such as DNA double strand break repair, as presented by Michael Hendzel from Edmonton, and transcriptional activation of estrogen receptors highlighted by Concetta Ambrosino from Italy. She dissected their recent data on the estrogen receptor interactome, which includes many known actin polymerization regulating proteins.^4,5 The need for regulating actin polymerization in the nucleus was also emphasized by Roger Karlsson from Stockholm, who highlighted the role of the actin monomer-binding protein profilin in this process. The relative levels of actin monomers and filaments can also play a role in regulating specific sets of genes, as exemplified by MRTF-A, which is a transcriptional coactivator of SRF. Richard Treisman from London presented their latest studies on the structural basis of how MRTF-A senses actin monomer levels in the cell.⁶ Besides, and likely in conjunction with the transcription process, actin is also a component of many chromatin remodelling complexes, and Ann-Kristin Östlund Farrants discussed the presence (or absence) of actin in the SWI/SNF chromatin remodelling complex. On the other hand, Piergiorgio Percipalle from Stockholm emphasized the importance of specific interactions between actin and heterogeneous nuclear ribonucleoproteins and their role in the regulation of transcription elongation through establishment of permissive chromatin and in the co-transcriptional assembly of ribonucleoprotein complexes. These mechanisms have been suggested to play a role in connecting multiple molecular machineries.⁷ Also Neus Visa from Stockholm discussed the importance of actin in the crosstalk between chromatin remodelling, RNA polymerase function and pre-mRNA processing. A very exciting area of research is currently the functional organization of the genome and the issue of whether the location of a gene within the nucleus matters is one of the fundamental open questions in cell biology. This problem was touched upon also at the Wenner-Gren meeting when Rolf Ohlsson from Stockholm described his efforts to map in detail the genome-wide pattern of chromosome interactions and their functional meaning. Andrew Belmont from Urbana described his recent advancements in engineering gene loci, which can be used to study the dynamics of specific loci within the nucleus, and hopefully in the future also permit detailed dissection of pathways that mediate these events. It is likely that actin and factors associated with it play a role in this process.⁸ One way by which actin could contribute to genome organization is through the “nucleoskeleton,” which is a network of filaments underlying the nuclear envelope and lamina that plays a role in genome organization and integrity.⁹ Kathy Wilson from Baltimore gave an overview of the proteins associated with the nucleoskeleton, concentrating on the LEM domain family proteins and their interactions with actin. Besides acting as an organizer in the nuclear periphery, the nucleoskeleton is linked to the cytoplasmic cytoskeleton via the LINC complexes and Greg Gundersen from New York presented their data on how this connection is used by the cell to position the nucleus during cell movement. This implies that the cytoplasmic and nuclear actin networks are intimately connected and that communication between them can be used to transmit information between the cytoplasmic and nuclear compartments. This topic was also taken up by Jun-Lin Guan from Ann Arbor, who discussed focal adhesion kinase (FAK)-mediated signaling in cancer and how this signaling pathway may also control nuclear actin through N-WASp.

Actin never operates alone and is perhaps one of the proteins with the most interaction partners in the cell. Therefore, it is not surprising that also in the nucleus actin acts in concert with its binding partners to mediate the gene regulatory events. The founder of the nuclear motor protein field, Primal de Lanerolle from Chicago, made an inventory of the myosins localized in the nucleus to date and presented his recent data on the function of Myosin V in splicing. Pavel Hozak from Prague concentrated on different forms of Myosin IC in the nucleus and suggested that nuclear phosphoinositides may play an important role in controlling the activities of these motor proteins. Besides myosins, many actin-related proteins (Arps), which share the same fold as actin, are found in the nucleus and are linked especially to different chromatin remodelling complexes, as discussed by Masahiko Harata from Japan. These talks highlighted the fact that in order to understand how actin functions in the nucleus we also need to consider its nuclear interactions and develop methods to study these interactions in the context of different nuclear processes.

The role of actin and its associated factors in gene expression does not end when the pre-mRNA is transcribed, but it seems that actin, in fact, accompanies the mRNA to the cytoplasm and may even play a role in the translation process. Sm proteins play an essential role in the nuclear splicing process, but Greg Matera from Chapel Hill proposed that these proteins also operate in the cytoplasm, especially in the context of localizing mRNAs to specific sites within the cell. Thus, it appears that actin is not the only protein “moonlighting” in different cellular compartments. Significant crosstalk may exist between the actin cytoskeleton and localized translation. This was highlighted by Stefan Huttelmeier from Halle, who discussed the regulatory circuits of IGF2BP1 (also known as ZBP1), which controls, for example, localized actin translation and by Clive Bramham from Bergen, who presented their findings on how the fate of Arc (activity-regulated cytoskeleton-associated protein) mRNA is regulated especially in the context of long-term potentiation in the brain. These talks illustrated that actin may influence the mRNA targeting and translation processes and vice versa, the proteins participating in these processes may influence actin dynamics.

Taken together, this Wenner-Gren symposium was very timely, and clearly demonstrated how this field has taken significant steps forward during the last few years. Today, there is no doubt that actin is present in the nucleus, and functional studies have shown that it has a very important role also in this cellular compartment especially in the context of gene expression, and this fact was evident in many talks at the symposium. One of the clear strengths in this field is the very distinct background of scientists battling with this research theme and this was apparent from the speaker list for the symposium. This fact also underscores the need for collaboration, as this field merges two complicated areas, gene expression and the complex nature of actin. Those studying gene expression may not completely grasp the finely-tuned regulation of different aspects of actin dynamics, while the actin biochemists, like me, still feel bewildered by the complex interplay between the huge macromolecular transcription complexes. Nevertheless, by combining these areas of expertise, we have an opportunity to make some sense into how actin contributes to functional organization of the nucleus, and this Wenner-Gren symposium with stimulating social events certainly helped to foster these types of collaborations. The most important outcome of the meeting was anyhow the feeling that we are in the process of making significant progress in understanding how actin contributes to gene expression, both in and outside the nucleus. This will serve as a motivation to us tackling these research questions daily and in the words of one of the participants, “The meeting was hugely enjoyable and has inspired me to more actively pursue structural aspects of actin in the nucleus.”