Since the invention of gene targeting in mice (Capecchi, 1989), there has been an upsurge of interest in all aspects of the anatomy of the mouse central nervous system (Capecchi, 2005; Watson et al., 2012). The use of mutant strains in the study of spinal cord injury could be enhanced by a basic knowledge of the position of the major spinal cord tracts. Since there is limited information on the position of spinal tracts in the mouse, we have used data from other mammals to predict the most likely position of tracts that have not been traced experimentally in the mouse.
Information on the location of the major spinal cord tracts in the mouse is sparse. We have collected published data on the position of these tracts in the mouse and have used data from other mammals to identify the most likely position of tracts for which there is no mouse data. We have plotted the position of six descending tracts (corticospinal, rubrospinal, medial and lateral vestibulospinal, rostral and caudal reticulospinal) and eight ascending tracts (gracile; cuneate; postsynaptic dorsal columns; dorsolateral, lateral, and anterior spinothalamic; dorsal and ventral spinocerebellar) on diagrams of transverse sections of all mouse spinal cord segments from the first cervical to the third coccygeal segment. Anat Rec, 2012. © 2012 Wiley Periodicals, Inc.
Why is Information on Spinal Cord Tracts Hard to Find?
In the silver degeneration stain era, almost all studies reported on the complete course of tracts of interest. Following the introduction of modern retrograde tracing techniques in the early 1970, we entered an era when clear pictures of the course of a tract have became a rarity. Even with the development of new anterograde tracing methods, there are relatively few studies that adequately picture the course of a tract. There are exceptions, such as the anterograde tracing studies of Liang et al. (2012a, 2012b). Fortunately, the anatomy of spinal tracts is very conserved among mammals, the corticospinal tract being the only one that varies significantly (Watson and Harvey, 2009). Because of the consistency of position of the tracts in mammals, we have used data from rat, cat, monkey, and human studies to predict the position of some tracts in the mouse.
The descending tracts in the mouse arise from a very wide range of cell groups (Liang et al., 2011). As expected, the majority of the descending fibers were shown to arise in the cerebral cortex, the red nucleus, the hindbrain reticular nuclei, and the vestibular nuclei. This study shows the actual or likely position of the tracts arising from these major centers. The tracts arising from other groups will not be considered in this report.
Space limitations have restricted the number of spinal cord sections we could display in this article. However, we intend to include drawings of tract position in all spinal cord levels in a mouse spinal cord atlas intended for publication in 2013. We are happy to provide more information on other spinal cord segments on request. Note that a single image of the likely position of spinal cord tracts in the mouse cervical spinal cord appears in a book chapter (Sengul and Watson, 2012).
MATERIALS AND METHODS
We researched the literature to first identify reports of the anatomy of these tracts in the mouse. We next searched for papers on the anatomy of the spinal tracts in other mammals, principally rat, cat, monkey, and human. We plotted the data on a published set of sections of the mouse spinal cord from the first cervical to the last coccygeal segment (Watson et al., 2009). In each drawing, the descending tracts in the lateral and ventral funiculi are drawn and labeled on the left, and the ascending tracts are drawn and labeled on the right. The tracts of the dorsal funiculus are drawn on both left and right sides. The abbreviations we used are consistent with those of Paxinos and Watson (2007), in which tracts are represented by lower-case letters and cell groups are represented by upper case letters. We confirmed the position of some of the tracts with reference to BDA tracing experiments (Fig. 2A,B), gene expression (Fig. 2C,D), and histochemical and immunomarkers (Fig. 3).
Figure 1 shows the actual or putative position of each of the six major descending tracts and the eight major ascending tracts in drawings of transverse sections from C8, T7, L3, and S2.
The Spinal Cord Position of the Six Major Descending Tracts
Consistent with other rodents and marsupials, the corticospinal tract in the mouse crosses in the caudal hindbrain and occupies the most ventral part of the dorsal funiculus (Watson and Harvey, 2009) (Fig. 3 calbindin [Cb], SMI-32, and acetylcholinesterase [AChE]). The dorsal corticospinal tract (dsc) consists of small axons, which enables it to be easily distinguished from the large axons of the gracile and cuneate fasciculi. The position of dsc in the mouse is nicely demonstrated in a study of the projection of Thy1-eYFP-labeled pyramidal neurons (Porrero et al., 2010) (Fig. 2C).
The rubrospinal tract of the mouse crosses the midline in the caudal midbrain and forms a contralateral tract in the dorsolateral corner of the lateral funiculus (Liang et al., 2012b) (Figs. 2A, 3), in a position similar to that found in cat (Nyberg-Hansen and Brodal, 1965). The position of the rubrospinal tract in the mouse is marked by calretinin (Cr) staining (Watson et al., 2009; Liang et al., 2012b), which assisted us in plotting its position at different levels.
The course and location of the vestibulospinal tracts follows the findings of silver degeneration studies in the cat (Nyberg-Hansen and Mascitti, 1964) and an unpublished tracing study of Dr. Andy Liang (Fig. 2B). These studies found that the lateral vestibulospinal tract (lvs), which is uncrossed, lies in the ventral funiculus and undergoes a lateral shift during its descent to lumbosacral levels; the medial vestibulospinal tract (mvs), which is crossed, is located within the medial longitudinal fasciculus, next to the ventral median fissure, and does not reach lumbar levels.
The two reticulospinal tracts are distinguished by their rostral or caudal origin in the hindbrain. Our illustrations of the course of the reticulospinal fibers follow the studies of Nyberg-Hansen (1965) and Nathan et al. (1996). They showed that the rostral reticulospinal tract (rrts) lies next to the ventral and ventromedial periphery of the ventral funiculus and maintains this position during its course (Fig. 2C). The caudal reticulospinal tract (crts) travels in the lateral funiculus close to the lateral border of the ventral horn. Like the rrts, the crts maintains the same relative position as it descends.
The Spinal Cord Position of the Eight Major Ascending Tracts
The largest ascending tracts are the gracile and cuneate fasciculi, the spinothalamic tracts, and the spinocerebellar tracts. However, there are many other small ascending fiber groups, most of which have been identified on the basis of retrograde tracing (Kayalioglu, 2009). We do not have enough data to predict the spinal cord position of these minor tracts.
Most of the dorsal funiculus in mammals is occupied by the gracile (gr) and cuneate (cu) tracts. These tracts are formed by the ascending axons of ipsilateral primary afferent neurons, and their anatomy is essentially the same in all mammals that have been studied (Kayalioglu, 2009). However, it has been shown that some fibers in the dorsal funiculus arise from spinal dorsal horn neurons. This bundle has been called the postsynaptic dorsal column (psdc) (Giesler et al., 1984). We have represented this pathway between the gracile and cuneate fasciculi, on the basis of its position in the rat (Giesler et al., 1984). The position of the gracile tract is nicely shown by Pitx2 expression in an image taken from the GENSAT site (Fig. 2D).
The position of the ventral and dorsal spinocerebellar tracts (vsc and dsc) is based on Terman et al. (1998) and Xu and Grant (1994). We have taken account of the finding in these studies that the dsc shifted dorsally and that the vsc shifted laterodorsally as they ascended the spinal cord from lumbar to cervical levels. The spinocerebellar tracts lie in contact with the surface of the lateral and ventrolateral funiculi. The dsc fibers are ipsilateral relative to their origin in dorsal root ganglia, but the vsc fibers cross in the ventral white commissure of the spinal cord close to their level of entry (Kayalioglu, 2009).
The spinothalamic fibers are traditionally divided into lateral (lst) and ventral (Ivst) tracts in human neurology textbooks, but an experimental study in rats showed an additional dorsolateral spinothalamic tract (dlst) (Apkarian and Hodge, 1989). We plotted the location of the three spinothalamic tracts according to Apkarian and Hodge (1989), Stevens et al. (1991), Zhang et al. (2000a, 2000b), and Friehs et al. (1995). The dlst and lst are deep to the spinocerebellar tracts (Fig. 1). We did not find a report of the course of the spinothalamic tracts in mice, but a retrograde tracing study showed that there are about 3,500 spinothalamic neurons that are distributed in the spinal cord in a manner similar to that found in rat, cat, and monkey (Davidson et al., 2010).
We believe that this contribution to spinal cord anatomy will be of value to researchers interested in experimental spinal cord injury in the mouse. It would of course be ideal to have experimental data on all of the major spinal cord tracts in the mouse, but in the absence of such information, we are forced to rely on studies on other mammals. However, those who use these diagrams should be constantly aware that many of the tract positions we have designated are based on extrapolation of data from other mammals. This means that that there is always a possibility that the situation in the mouse might be different in some way.
We are encouraged by evidence from other vertebrates that indicates that spinal cord tract anatomy is very consistent, with the exception of the corticospinal tract. For example, the positions of the avian rubrospinal tract (Wild et al., 1979), spinocerebellar tracts (Whitlock, 1952), and gracile and cuneate fasciculi (Wild, 1985) are very similar to those in mammals, despite the fact that these two vertebrate classes are separated in evolutionary terms by more than 300 million years (Kemp, 1980).
Among mammals, the only tract that is reported to vary in spinal cord position from group to group is the corticospinal tract; the major corticospinal tract may lie in the dorsal funiculus, the lateral funiculus, or in the ventral funiculus (Verhaart, 1962; Watson and Harvey, 2009). In our literature search, we have not found evidence of significant variation in the spinal cord position of any of the other major tracts in mammals.
As noted above, there are many minor descending and ascending tracts in addition to the major tracts outlined in our figures. Liang et al. (2010) identified over 30 centers in the brain that give rise to descending pathways to the spinal cord. Kayalioglu (2009) summarizes the wide variety of pathways that ascend from the spinal cord to different areas of the brain. While these minor tracts are generally small, specific tracts may be important to researchers interested in functional connections between the brain and spinal cord. An example is the study of the serotoninergic raphespinal projection in the analysis of recovery after spinal cord injury (Holmes et al., 2005).
We believe that the drawings we have made on the basis of extrapolation are likely to be correct in most respects, and should provide a useful interim guide, until experimental data on these tracts becomes available.