Axonal regeneration proceeds through specific axonal fusion in transected C. elegans neurons

Abstract

Functional neuronal recovery following injury arises when severed axons reconnect with their targets. In Caenorhabditis elegans following laser-induced axotomy, the axon still attached to the cell body is able to regrow and reconnect with its separated distal fragment. Here we show that reconnection of separated axon fragments during regeneration of C. elegans mechanosensory neurons occurs through a mechanism of axonal fusion, which prevents Wallerian degeneration of the distal fragment. Through electron microscopy analysis and imaging with the photoconvertible fluorescent protein Kaede, we show that the fusion process re-establishes membrane continuity and repristinates anterograde and retrograde cytoplasmic diffusion. We also provide evidence that axonal fusion occurs with a remarkable level of accuracy, with the proximal re-growing axon recognizing its own separated distal fragment. Thus, efficient axonal regeneration can occur by selective reconnection and fusion of separated axonal fragments beyond an injury site, with restoration of the damaged neuronal tract.

Figure 1

Axonal regeneration in the C. elegans mechanosensory neurons. (A) A wild-type zdIs5(Pmec-4::GFP) animal, illustrating the ALM and PLM mechanosensory neurons. Anterior is left and ventral is down in these and all other images. Reconnection between proximal and distal segments in ALM (B) and PLM (C), twenty-four hours post-axotomy. (D,E) Degeneration of the distal axon occurs in the absence of reconnection. Filled arrowheads point to site of axotomy; open arrowheads show fusion site; asterisks highlight intestinal auto-fluorescence; scale bars: 25 μm. (F) Quantification of regenerative phenotypes, with regrowth defined as sprouting from proximal end, distal contact representing the proportion of regrowing axons that made proximal-distal contact, and distal maintenance showing the percentage of reconnecting neurons that preserved the distal axon segment. Error bars: standard error of proportion; n values: 77 and 95 for ALM and PLM, respectively.

Figure 2

Analysis of axonal reconnection with transmission electron microscopy. Serial thin sections of a wild-type animal 24 hrs post-axotomy are shown, along with a scheme to demonstrate how the PLM axon regrew around the laser damage zone. High magnification micrographs show the original axon (blue) and the new process (yellow) as it branched off from the original axon (section #1342), travelled inwards from the body wall (section #1376), traversed the damage site (section #1412 – #1480) and then fused as a very thin process (yellow) to the distal axon segment (blue) at section #1509. The fusion site is shown at higher power in section #1508, where the white arrow indicates the fusion zone. Local membrane whorls and voids (red asterisks) caused by collateral laser damage in the hypodermis can be seen in sections #1412, #1450 and #1480. Extracellular space below the PLM axon in section #1342 is swollen by mantle protein (black arrow), a characteristic feature of the mechanosensory neurons. Scale bars: 0.5 μm and 0.2 μm in sections #1509 and #1508, respectively.

Figure 3

Axonal fusion re-establishes cytoplasmic continuity. Kaede was converted in PLM in the bracketed regions 24 hrs post-axotomy to analyze anterograde (A) and retrograde (B) cytoplasmic diffusion. Representative images show green and red fluorescence before conversion in panels (i) and (ii), and after conversion in panels (iii) and (iv). Proximal-distal reconnection permitted anterograde (A) and retrograde (B) diffusion of red fluorescent Kaede across the site of axotomy (A,iv; B,iv). Filled arrowheads point to site of axotomy; open arrowheads show fusion site; scale bars: 25 μm. (C) Quantification of animals displaying either anterograde or retrograde diffusion of Kaede in the absence (no contact) or presence of reconnection for both ALM and PLM. Error bars: standard error of proportion; n values within each graph. Note the presence of a weak background signal before conversion through the red filter (A; B).

Figure 4

Specificity in axonal fusion. (A) Schematic diagrams of the zdIs5(Pmec-4::GFP); vdEx166(Plad-2::mCherry) strain where ALM and PLM were visualized with green fluorescence and ALN and PLN with red fluorescence. (B) Quantification of the number of axons displaying specific or aspecific fusion following axotomy of either ALM or PLM (left bars, one axon cut), or after simultaneous axotomy of both ALM and ALN, or PLM and PLN (right bars, two axons cut). Error bars: standard error of proportion; n values within each graph.