Background

An organism’s nervous system is highly dynamic and enables cognition, growth, breathing, physical and emotional sensation, and other daily activities. The nervous system is comprised of a vast network of specialized cells, called neurons, that work together to transmit signals between different parts of the body (Abbasi et al., 2018). Each neuron has a cell body and extensions of different lengths. Shorter extensions, called dendrites, receive and transmit signals to the neuronal cell body (Lovinger, 2008; Reference 20). Longer extensions, called axons, carry electrical impulses away from the cell body to muscles, glands, and distant neurons (Lovinger, 2008). At the cellular level, nervous system function relies on precise wiring of these neurons, breaches to which lead to neurological disorders (Goodman and Shatz, 1993; Chédotal and Richards, 2010; Engle, 2010; Kolodkin and Tessier-Lavigne, 2011; Ren et al., 2019). Specialized extensions located at the tips of dendrites and axons, called neuronal growth cones, play a key role in ensuring proper axonal growth and guidance to establish functional connectivity (Lowery and Vactor, 2009; Vitriol and Zheng, 2012). Neuronal growth cones are highly motile sensory units that can detect extracellular cues and convert them into intracellular signals (Suter and Forscher, 2000; Dent et al., 2011). Growth cones use these signals to remodel cytoskeletal proteins, including F-actin, which constitute structures like lamellipodia and filopodia that are needed for motility. Thus, growth cones control axonal growth and guidance and the formation of neuronal circuitry. Therefore, elucidating the signaling mechanisms that dictate actin organization and dynamics during growth cone motility is imperative to understand the process of nervous system development and regeneration.

Cortactin is an actin-binding protein known to regulate cytoskeletal dynamics and cell migration in non-neuronal cells. Cortactin consists of an amino terminal acidic domain (NTA), followed by an F-actin-binding domain, an alpha-helical domain, a proline-rich region (PRR), and an SH3 domain at the C-terminus (Figure 1) (Ammer and Weed, 2008; Decourt et al., 2009). It was first characterized as a substrate for Src kinase and regulates several actin-related processes, including cell migration, tumor metastasis, and response to pathogens (Schnoor et al., 2018). Cortactin is important in the assembly, branching, and stabilization of cytoskeletal structures, such as lamellipodia and filopodia (Helgeson and Nolen, 2013; He et al., 2015). While cortactin is known to undergo multiple post-translational modifications, including phosphorylation (Schnoor et al., 2018; Ren et al., 2019) and acetylation, not much is known about the exact mechanisms underlying its fundamental roles in neuronal growth cones and filopodia formation in neurons.

Using cultured Aplysia bag cell neurons as a model, recent findings demonstrated that phosphorylation of a single tyrosine 499 residue of cortactin mediated by the Src2 kinase is important for the formation of filopodia and the regulation of actin organization and dynamics in growth cones (He et al., 2015; Ren et al., 2019). We have also demonstrated with purified proteins that Aplysia Src2 phosphorylates Aplysia cortactin in vitro (Ren et al., 2019). Here, we describe in detail the robust and cost-effective assay to assess the direct phosphorylation of cortactin by Src2 kinase used in our previous study (Ren et al., 2019). Using E. coli (BL21-DE3) cells as a host, we developed a reproducible two-step protocol to produce pure cortactin and Src kinase proteins at high yields.

Src kinases are phosphoryl transferases that transfer the γ-phosphate of ATP to tyrosine residues on specific substrate proteins. A common method used to detect kinase activity is a radioactive ATP kinase assay (Karra et al., 2017). This kinase assay protocol tracks the transfer of the radio-isotope ³²P from ATP [γ-³²P] to the substrate. Incorporation of this radiolabeled phosphate into the kinase substrate is then measured using autoradiography. Despite the sensitivity of this assay using optimized protocols with human Src and cortactin (Tehrani et al., 2007), we were unable to detect phosphorylation of Aplysia cortactin by Src kinase in vitro. Because bacteria also possess kinases, we reasoned and found to be true that our bacterially expressed and purified Src and cortactin proteins were non-specifically phosphorylated by bacterial kinases (Shrestha et al., 2012) during the purification process, therefore preventing Src2-mediated phosphorylation of cortactin. We circumvented these issues by co-purifying Aplysia Src2 and cortactin in the presence of a phosphatase called YopH (encoded by Yersinia enterocolytica) (Zhang et al., 1992), followed by sequential affinity and size exclusion chromatography, which yielded pure, unphosphosphorylated proteins that could be used in an in vitro kinase assay. By incorporating YopH into the purification scheme, we removed residual phosphate groups from the proteins that may have been introduced during the heterologous expression step. Here, we describe the co-purification process and present data obtained using this method, revealing the first direct biochemical evidence that Aplysia Src2 phosphorylates Aplysia cortactin and that Y505 (Cort-FYF) is preferentially modified in vitro, whereas Y499 is the essential tyrosine that is phosphorylated by Src2 in neuronal growth cones (see Figure 6). This procedure could be more generally adopted to remove non-specific phosphorylation of other proteins expressed in a prokaryotic host.

Figure 1. Schematic representation of the functional domain structure of Cortactin (human). N = N terminus, C = C-terminus, NTA = amino-terminal acidic region, SH3 = Src homology 3 domain. Schematics were created with Biorender.com.