Moving through the crowd. Where are we at understanding physiological axon growth?

Stretch triggers microtubule stabilization and MARCKS-dependent membrane incorporation in the shaft of embryonic axons

Neurons have a unique polarized nature that must adapt to environmental changes throughout their lifespan. During embryonic development, axon elongation is led by the growth cone,1 culminating in the formation of a presynaptic terminal. After synapses are formed, axons elongate in a growth cone-independent manner to accompany body growth while maintaining their ultrastructure and function.2,3,4,5,6 To further understand mechanical strains on the axon shaft, we developed a computer-controlled stretchable microfluidic platform compatible with multi-omics and live imaging. Our data show that sensory embryonic dorsal root ganglia (DRGs) neurons have high plasticity, with axon shaft microtubules decreasing polymerization rates, aligning with the direction of tension, and undergoing stabilization. Moreover, in embryonic DRGs, stretch triggers yes-associated protein (YAP) nuclear translocation, supporting its participation in the regulatory network that enables tension-driven axon growth. Other than cytoskeleton remodeling, stretch prompted MARCKS-dependent formation of plasmalemmal precursor vesicles (PPVs), resulting in new membrane incorporation throughout the axon shaft. In contrast, adolescent DRGs showed a less robust adaptation, with axonal microtubules being less responsive to stretch. Also, while adolescent DRGs were still amenable to strain-induced PPV formation at higher stretch rates, new membrane incorporation in the axon shaft failed to occur. In summary, we developed a new resource to study the biology of axon stretch growth. By unraveling cytoskeleton adaptation and membrane remodeling in the axon shaft of stretched neurons, we are moving forward in understanding axon growth.

Concept of Normativity in Multi-Omics Analysis of Axon Regeneration

Transcriptomes and proteomes can be normalized with a handful of RNAs or proteins (or their peptides), such as GAPDH, β-actin, RPBMS, and/or GAP43. Even with hundreds of standards, normalization cannot be achieved across different molecular mass ranges for small molecules, such as lipids and metabolites, due to the non-linearity of mass by charge ratio for even the smallest part of the spectrum. We define the amount (or range of amounts) of metabolites and/or lipids per a defined amount of a protein, consistently identified in all samples of a multiple-model organism comparison, as the normative level of that metabolite or lipid. The defined protein amount (or range) is a normalized value for one cohort of complete samples for which intrasample relative protein quantification is available. For example, the amount of citrate (a metabolite) per µg of aconitate hydratase (normalized protein amount) identified in the proteome is the normative level of citrate with aconitase. We define normativity as the amount of metabolites (or amount range) detected when compared to normalized protein levels. We use axon regeneration as an example to illustrate the need for advanced approaches to the normalization of proteins. Comparison across different pharmacologically induced axon regeneration mouse models entails the comparison of axon regeneration, studied at different time points in several models designed using different agents. For the normalization of the proteins across different pharmacologically induced models, we perform peptide doping (fixed amounts of known peptides) in each sample to normalize the proteome across the samples. We develop Regen V peptides, divided into Regen III (SEB, LLO, CFP) and II (HH4B, A1315), for pre- and post-extraction comparisons, performed with the addition of defined, digested peptides (bovine serum albumin tryptic digest) for protein abundance normalization beyond commercial labeled relative quantification (for example, 18-plex tandem mass tags). We also illustrate the concept of normativity by using this normalization technique on regenerative metabolome/lipidome profiles. As normalized protein amounts are different in different biological states (control versus axon regeneration), normative metabolite or lipid amounts are expected to be different for specific biological states. These concepts and standardization approaches are important for the integration of different datasets across different models of axon regeneration.

The outer kinetochore components KNL-1 and Ndc80 complex regulate axon and neuronal cell body positioning in the C. elegans nervous system

The KMN (Knl1/Mis12/Ndc80) network at the kinetochore, primarily known for its role in chromosome segregation, has been shown to be repurposed during neurodevelopment. Here, we investigate the underlying neuronal mechanism and show that the KMN network promotes the proper axonal organization within the C. elegans head nervous system. Postmitotic degradation of KNL-1, which acts as a scaffold for signaling and has microtubule-binding activities at the kinetochore, led to disorganized ganglia and aberrant placement and organization of axons in the nerve ring - an interconnected axonal network. Through gene-replacement approaches, we demonstrate that the signaling motifs within KNL-1, responsible for recruiting protein phosphatase 1, and activating the spindle assembly checkpoint are required for neurodevelopment. Interestingly, while the microtubule-binding activity is crucial to KMN's neuronal function, microtubule dynamics and organization were unaffected in the absence of KNL-1. Instead, the NDC-80 microtubule-binding mutant displayed notable defects in axon bundling during nerve ring formation, indicating its role in facilitating axon-axon contacts. Overall, these findings provide evidence for a noncanonical role for the KMN network in shaping the structure and connectivity of the nervous system in C. elegans during brain development.

Doublecortin reinforces microtubules to promote growth cone advance in soft environments

Doublecortin (DCX) is a microtubule-associated protein critical for brain development. Although most highly expressed in the developing central nervous system, the molecular function of DCX in neuron morphogenesis remains unknown and controversial. We demonstrate that DCX function is intimately linked to its microtubule-binding activity. By using human induced pluripotent stem cell (hiPSC)- derived cortical i 3 Neurons genome engineered to express mEmerald-tagged DCX from the endogenous locus, we find that DCX-MT interactions become highly polarized very early during neuron morphogenesis. DCX becomes enriched only on straight microtubules in advancing growth cones with approximately 120 DCX molecules bound per micrometer of growth cone microtubule. At a similar saturation, microtubule-bound DCX molecules begin to impede lysosome transport, and thus can potentially control growth cone organelle entry. In addition, by comparing control, DCX-mEmerald and knockout DCX -/Y i 3 Neurons, we find that DCX stabilizes microtubules in the growth cone peripheral domain by reducing the microtubule catastrophe frequency and the depolymerization rate. DCX -/Y i 3 Neuron morphogenesis was inhibited in soft microenvironments that mimic the viscoelasticity of brain tissue and DCX -/Y neurites failed to grow toward brain-derived neurotrophic factor (BDNF) gradients. Together with high resolution traction force microscopy data, we propose a model in which DCX-decorated, rigid growth cone microtubules provide intracellular mechanical resistance to actomyosin generated contractile forces in soft physiological environments in which weak and transient adhesion-mediated forces in the growth cone periphery may be insufficient for productive growth cone advance. These data provide a new mechanistic understanding of how DCX mutations cause lissencephaly-spectrum brain malformations by impacting growth cone dynamics during neuron morphogenesis in physiological environments.

Tubulin isotype regulation maintains asymmetric requirement for α-tubulin over β-tubulin

How cells regulate α- and β-tubulin to meet the demand for αβ-heterodimers and avoid consequences of monomer imbalance is not understood. We investigate the role of gene copy number and how shifting expression of α- or β-tubulin genes impacts tubulin proteostasis and microtubule function in Saccharomyces cerevisiae. We find that α-tubulin gene copy number is important for maintaining excess α-tubulin protein compared to β-tubulin protein. Excess α-tubulin prevents accumulation of super-stoichiometric β-tubulin, which leads to loss of microtubules, formation of non-microtubule assemblies of tubulin, and disrupts cell proliferation. In contrast, sub-stoichiometric β-tubulin or overexpression of α-tubulin has minor effects. We provide evidence that yeast cells equilibrate α-tubulin protein concentration when α-tubulin isotype expression is increased. We propose an asymmetric relationship between α- and β-tubulins, in which α-tubulins are maintained in excess to supply αβ-heterodimers and limit the accumulation of β-tubulin monomers.