Early stages of axonal regeneration in the goldfish optic tract: an electron microscopic study.

Two hours after the goldfish optic tract was cut, the severed axons in the retinal stump of the tract showed ballooning of the axoplasm and myelin sheath in the region of the cut, with accumulation in the swollen axon of various organelles, including dense cored vesicles. By day 1 the myelin sheath had degenerated back to a node of Ranvier and the tip of the severed axon had formed a myelin-free terminal bulb with a well-organized core of 9-10 nm filaments. By 2 days, such terminal bulbs were often seen to be extended on a neck of cytoplasm a few micrometers in length, presumably indicating axonal outgrowth. In addition, occasional small bundles of axon sprouts were first seen at this time. The sprouts had a diameter of about 2 micrometers and contained a central core of 9-10 nm filaments surrounded by a mantle of cell organelles (smooth endoplasmic reticulum, mitochondria and diverse vesicles), with few if any microtubules. Sprouts within a bundle were separated by fairly uniform 10-15 nm spaces. Beginning at 3 days, significant numbers of microtubules appeared in the sprouts, and there was an increasing proportion of small diameter (greater than or equal to 0.3 micrometer) sprouts. Thus it was not until 3 days that the sprouts took on the appearance usually considered to be typical of regenerating axons. By 6 days a dense layer of glial cells or macrophages formed a cap over the cut surface of the tract. Penetrating this layer were bundles containing up to 20-30 axon sprouts and also single axons which may have been serving as 'pioneering' fibres to which later-emerging axons would attach. There was no evidence that the regenerating axons were guided by the glial cells. At 6 days astroglia began to separate individual axons within the bundles but oligodendrocytes were still inactive at this time.