Thomas A Phillips: Imaging actin organisation and dynamics in 3D

The dynamics and organisation of F-actin have been widely studied in a breadth of cell types on classical two-dimensional (2D) surfaces. Recent advances in optical microscopy have enabled interrogation of these cytoskeletal networks in cells within three-dimensional (3D) scaffolds, tissues and in vivo. Emerging studies indicate that the dimensionality experienced by cells has a profound impact on the structure and function of the cytoskeleton, with cells in 3D environments exhibiting cytoskeletal arrangements that differ to cells in 2D environments. However, the addition of a third (and fourth, with time) dimension leads to challenges in sample preparation, imaging and analysis, necessitating additional considerations to achieve the required signal-to-noise ratio and spatial and temporal resolution. Here, we summarise the current tools for imaging actin in a 3D context and highlight examples of the importance of this in understanding cytoskeletal biology and the challenges and opportunities in this domain.

Introduction

Actin is a highly conserved 42 kDa molecule found in all eukaryotic organisms; it plays an essential role in the organisation of intracellular architecture of cells to enable division, shape changes, protrusion and migration. There are six actin isoforms in mammals (encoded by separate genes) that are almost identical in structure but are expressed at different levels, depending on the cellular context (Kashina, 2020). These are broadly classified into non-muscle and muscle isoforms, with diverse structures that support cell-type-specific functions. Non-muscle β-actin has been most intensively studied biochemically and is the only isoform that is critical for viability, with some redundancy shown between other isoforms (Bunnell et al., 2011). In cells, actin exists in monomeric globular (G-actin) and filamentous (F-actin) forms, with F-actin formed from multiple G-actin monomers. ATP-bound G-actin binds the barbed (+) end of F-actin and, following ATP hydrolysis, disassociates from the pointed (−) end, in a process known as treadmilling, which is critical for cellular adaptation and function (Carlier and Shekhar, 2017). This process is regulated and directed by numerous actin-binding proteins that control processes such as capping, branching and sequestration of G-actin (reviewed in Pollard, 2016). Formin proteins promote unbranched actin polymerisation via conserved formin homology (FH) domains by preventing access to capping or severing proteins, such as cofilin proteins, thus halting formation of structures such as stress fibres or filopodia (Courtemanche, 2018). The Arp2/3 complex promotes nucleation of 70° branched F-actin; this enables assembly of more-complex ‘dendritic’ architecture to support cell shape changes and new protrusions, such as those found in lamellipodia (Papalazarou and Machesky, 2021).

In addition to actin nucleators, capping proteins can promote both polymerisation of branched actin filaments and prevent overextension of filament growth at the plasma membrane (Edwards et al., 2014). Actin is also found in the nucleus, in both G- and F-forms, where it contributes to chromatin architecture, transcription and survival (Hurst et al., 2019). Thus, actin dynamics and organisation crucially depend on associated proteins and are context dependent.

Much of our current understanding of F-actin organisation comes from in vitro studies of cells on 2D substrates, which do not recapitulate the complex environment in vivo. Most adherent cells, such as fibroblasts and smooth muscle cells, adopt a flat, well-spread morphology on 2D surfaces, and they assemble large F-actin stress fibres and lamellipodia. In 3D environments, these cells tend to exhibit more elongated, spindle-like organisation with fewer F-actin stress fibres compared to those in 2D environments (Caswell and Zech, 2018; Hakkinen et al., 2011). Fibroblasts can also assemble lobopodia and pseudopodia, which are blunt cortical actin or multiple extended protrusions, respectively, and are unique to 3D settings (Caswell and Zech, 2018).

Variations in 3D extracellular matrix (ECM) density, stiffness or alignment can all further influence the organisation of actin (Tran and Kumar, 2021). For some cell types, such as simple epithelia or endothelial cells, growth on a 2D surface is not dramatically unlike an in vivo setting in terms of dimensionality. However, the mechanical properties and curvature of the 2D substrate, as well as lack of apical flow, mucus or other physiologically important cues, can result in altered morphology and F-actin organisation compared to what is seen in vivo (Fedi et al., 2021; Rougerie et al., 2020). This highlights the importance of studying actin assembly and the factors controlling this in more physiologically relevant 3D environments if we are to understand the biology of these processes in development, homeostasis and disease. In this Review, we will discuss tools for visualising actin organisation and the current state-of-the-art of imaging platforms in the context of 3D actin imaging. Examples of 3D actin dynamics in model organisms and in vitro mammalian cell-based models, such as spheroids and organoids, immune cells, fibroblasts and neurons are discussed, as are current approaches and limitations for analysing 3D actin image datasets.

Read full publication here