The number of primary auditory afferents in the rat

Abstract

Published estimates of the number of primary auditory afferents in the rat differ by as much as 30%. We undertook to determine if the widely varying estimates were related to methodological differences, especially the difference between counting cells in Rosenthal's canal and fibers in the cochlear nerve. Type I ganglion cells and myelinated cochlear nerve fibers in the same ears were counted in Long-Evans and Sprague-Dawley strains. Type II spiral ganglion cells were also counted. In each strain the numbers of myelinated fibers and type I ganglion cells were essentially the same. Means for the Long-Evans were 18,036 fibers and 17,749 cells. Means for Sprague-Dawleys were higher: 19,444 fibers and 19,229 cells. The mean number of type II ganglion cells was also greater in Sprague-Dawley than in Long-Evans rats: 1,388 and 1,170, respectively. Cell and fiber counts from the two ears of the same animal differed on average by only 1%. The number of auditory afferents did not change with age over the range (2–10 months) studied here. Several methodological differences have probably contributed to the varying estimates of type I primary auditory afferents, but the discrepancies are not inherent in counts of fibers and spiral ganglion cells.

Introduction

Counts of spiral ganglion cells in Rosenthal's canal are often used to assess the status of the auditory nerve in human temporal bones (e.g., Hinojosa and Marion, 1983; Nadol et al., 1989) and the cochleas of experimental animals (e.g., Webster and Webster, 1981; Keithley and Feldman, 1983). An advantage of counting spiral ganglion cells rather than cochlear nerve fibers is that the spatial distribution of the surviving cells is more easily ascertained, which is important when there is interest in relating auditory capabilities to the survival of cochlear nerve fibers innervating different frequency regions of the cochlea. The extent to which counts of spiral ganglion cells and cochlear nerve fibers provide similar estimates of first-order auditory afferents is not clear because the sources of error are not well understood and may be unique to the two methods. Some reports have suggested that under pathological conditions there may be appreciable disparities between the two counts. For example, Ylikoski et al. (1981)found surprisingly small or moderate losses of cochlear nerve fibers associated with nearly total loss of spiral ganglion cells in several patients who had previously undergone labyrinthectomy. No such within subject differences have been reported for animals to the best of our knowledge, but two reports suggest that counts of spiral ganglion cells and cochlear nerve fibers may not always yield similar estimates of first-order auditory afferents. Keithley and Feldman (1979)reported a median number of 15,800 spiral ganglion cells in 1–2-month-old male rats of the Sprague-Dawley strain. That number included 1059 type II cells whose axons are unmyelinated, leaving a total of 14,741 type I cells with myelinated axons. Similar numbers were found for animals 6 months old (13,500) and 12 months old (14,934). Hoeffding and Feldman (1988), on the other hand, reported a median number of 21,218 myelinated fibers in the cochlear nerves of 2–3-month-old male rats of the same strain. A median number of 19,928 was found in animals 6 months old, but the difference was not statistically reliable. Hoeffding and Feldman recognized the large discrepancy of approximately 25–30% between the cell and fiber counts and tentatively attributed it to a difference in the precision of counting cells and counting fibers, the implication being that it is possible to obtain more accurate estimates by counting myelinated axons rather than cell somas. However, the lower estimates of Keithley and Feldman were supported by the cell counts of Berglund and Ryugo (1991)who reported a mean of 14,305 type I spiral ganglion cells, and by an estimate of 14,000 fibers derived from fiber counts in electron micrographs (Dannhof and Bruns, 1993). On the other hand, counts of type I spiral ganglion cells in two samples of Long-Evans rats in this laboratory (unpublished observations), undertaken as part of a physiological study, yielded mean values of 17,318 and 19,609, which are intermediate to the cell counts of Keithley and Feldman and the fiber counts of Hoeffding and Feldman. Intermediate values were also reported by Rueda et al. (1987)in their study of spiral ganglion development and by Burda et al. (1988)who reported mean totals of type I plus type II spiral ganglion cells of 16,450–18,830 for three rat strains. The spiral ganglion cell counts made in our laboratory noted above were done as part of the development of an electrophysiological method to estimate the number of surviving cochlear nerve fibers in deaf rats, a method that we hoped could be extended to profoundly deaf humans who are candidates for cochlear implants or are already implanted with such devices. The apparent discrepancy in the cell and fiber counts suggested that our cell counts might have seriously underestimated the number of primary auditory afferents in our experimental animals. To resolve this issue spiral ganglion cells and cochlear nerve fibers from the same ears were counted in two series of animals. The first was comprised of pigmented rats of the Long-Evans strain, the strain used in our physiological studies. The second series, consisting of albino Sprague-Dawley rats, was undertaken when the fiber and cell counts from the Long-Evans animals, both lower than Hoeffding and Feldman's fiber counts in the Sprague-Dawley, suggested there might be a difference between the strains.