Classic Papers: Earliest Axons in the Early Bird Embryo (Windle and Austin, 1935)

Neurons are often born in regions of the nervous system distant from the targets with which they must connect during development. As it extends to the target, each neuronal process, or axon, is led by the growth cone, the expanded “tip” that responds to both attractive and repulsive directional cues. Currently, much developmental neurobiology research focuses on the molecular nature of these cues and on the ways in which growth cones respond to such cues. However, these fascinating studies were preceded by a great amount of research on the initiation, timing, and progress of axon growth and connections in various animal species, which provided an essential temporal context for the expression patterns of guidance cues and receptors. In a classic paper published in 1935, Windle and Austin describe the routes of the earliest axons in the central nervous system of the chick embryo, assessed at several stages between the second and sixth days of incubation. First, I’ll summarize the methods and results in this classic paper, and then describe some recent research on the reticulospinal system, one of the earliest pathways to form in the vertebrate embryo.

I first read the Windle and Austin paper as a postdoctoral fellow, when I was characterizing the growth of the earliest central axons of cranial sensory neurons into the hindbrain in chick embryos, using the fluorescent lipophilic dye, DiI. Compared to Windle and Austin, I had a lot of sophisticated equipment and labeling techniques at my disposal; in the end, however, my results were almost identical to theirs, with respect to the timing of arrival of the earliest central axons, and I had a nice historical confirmation for a section of the Neuron paper my mentor and I published. Windle and Austin used a refinement of the procedure developed by Golgi – used extensively by Cajal – a pyridine silver-reduction method, to reveal “neurofibrillar differentiation” by deposition of silver in growing axons in chick embryos of different stages. Windle already had several publications, using this method to trace axons in rat and cat embryos, and on the development of motor behavior in rat and chick embryos as well. The researchers analyzed their silver-stained embryos using a compound binocular microscope with oil immersion objectives, and projected the locations of principal axon pathways onto stage-appropriate pasteboard models of chick brains, to produce the detailed figures for the paper. Since this paper preceded the standardized Hamburger and Hamilton (1951) staging criteria for chick embryos (incubator temperatures vary considerably between labs), I annotated the results section with H & H stages, using the somite numbers and brain outlines provided by the authors.

Hamburger and Hamilton (1951) Stage 16 chick embryo.

The earliest “neurofibrillary differentiation” was noted at the 18-somite stage, on the second day of incubation, in the midbrain; by the 20-somite stage, axons have developed from a few hindbrain (rhombencephalic) neurons, as well as from neurons that will pioneer the pathway called the medial longitudinal fasciculus. By the 27-somite stage (H&H stage 16), there are numerous reticular neurons in the hindbrain, particularly at the level of the developing inner ear. Many of these reticular neurons send their axons across the midline, to form the ventral commissure, before turning caudally towards the spinal cord. Windle and Austin also note at this stage that the neurons of the oculomotor nucleus have begun to send axons to their targets, as have the motor neurons of the trigeminal and facial nuclei. The authors summarized their results in a detailed table, which indicates the order of appearance of the axon tracts in the chick embryo, and whether these axons cross the midline. Comparison with rat and cat embryos is also included. By the end of the second day of incubation in the chick embryo, the following tracts could be detected with the Golgi silver-staining method: reticulospinal, medial longitudinal fasciculus, hypothalamo-subthalamic, tectobulbar, tectospinal, spinobulbar, and thalamo-tegmental.

The reticular formation is one of many neuroanatomy topics that drive medical students nuts: it receives inputs from everywhere and sends outputs everywhere, it has multiple nebulous locations and nuclei in the midbrain, pons, and medulla, and its neurons use just about every neurotransmitter under the sun. Cajal recognized that the reticular formation was organized according to connections and neuron size, and currently there are three main groups of nuclei recognized in humans (unpaired median raphe, and paired medial and lateral). The reticular formation nuclei receive sensory information from a variety of brain and spinal cord regions, and through their widespread output, influence critical functions such as respiration, cardiovascular activity, attention, arousal, and sleep. The Windle and Austin paper focused on the descending, or reticulospinal, axons arising from hindbrain nuclei, and thus I’ll add a bit on the connections with spinal cord interneurons, related to the following functions: to influence movements of proximal muscles, to control body posture, and to affect coordination of head and eye movements.

As you might expect from its functions, the reticular formation is a phylogenetically ancient neural system. Recent work has focused on the comparative anatomy of reticular projections, and modern researchers have a variety of axon tracing techniques at their disposal. Cruce and colleagues (1999) used retrogradely transported axonal tracers (horseradish peroxidase and Fluoro-Gold) to identify groups of brainstem neurons that projected to the spinal cord, in two cartilaginous fishes, the Thornback Guitarfish, and the Horn Shark. They identified numerous distinct reticular nuclei in these elasmobranchs, consistent with a complex organization similar to the reticular formation in other vertebrates. Similarly, New and colleagues (1998 ) traced axons descending to the spinal cord in the Channel Catfish, and found that the majority of neurons projecting to the spinal cord are located in the reticular formation of the hindbrain. Both ascending and descending reticular formation projections are of great clinical importance in humans, as they can be damaged or destroyed by strokes, spinal cord injuries, and astrocytomas.

References:

Cruce, W.L.R., Stuesse, S.L., and Northcutt, R.G. (1999) Brainstem neurons with descending projections to the spinal cord in two elasmobranch fishes: Thornback Guitarfish, Platyrhinoidis triseriata, and Horn Shark,Heterodontus francisci. J. Comparative Neurology 403, 534-560.

Hamburger, V. and Hamilton, H.L. (1992 republish) A series of normal stages in the development of the chick embryo. 1951. Developmental Dynamics 195, 231-272.

New, J.G., Snyder, B.D., and Woodson, K.L. (1998 ) Descending neural projections to the spinal cord in the Channel Catfish, Ictalurus punctatus. Anatomical Record 252, 235-253.

Windle, W.F., and Austin, M.F. (1935) Neurofibrillar development in the central nervous system of chick embryos up to 5 days’ incubation. J. Comparative Neurology 63, 431-463.