Structural and Ultrastructural Changes to Type I Spiral Ganglion Neurons and Schwann Cells in the Deafened Guinea Pig Cochlea

Abstract

Sensorineural hearing loss is commonly caused by damage to cochlear sensory hair cells. Coinciding with hair cell degeneration, the peripheral fibres of type I spiral ganglion neurons (SGNs) that normally form synaptic connections with the inner hair cell gradually degenerate. We examined the time course of these degenerative changes in type I SGNs and their satellite Schwann cells at the ultrastructural level in guinea pigs at 2, 6, and 12 weeks following aminoglycoside-induced hearing loss. Degeneration of the peripheral fibres occurred prior to the degeneration of the type I SGN soma and was characterised by shrinkage of the fibre followed by retraction of the axoplasm, often leaving a normal myelin lumen devoid of axoplasmic content. A statistically significant reduction in the cross-sectional area of peripheral fibres was evident as early as 2 weeks following deafening (p < 0.001, ANOVA). This was followed by a decrease in type I SGN density within Rosenthal's canal that was statistically significant 6 weeks following deafening (p < 0.001, ANOVA). At any time point examined, few type I SGN soma were observed undergoing degeneration, implying that once initiated, soma degeneration was rapid. While there was a significant reduction in soma area as well as changes to the morphology of the soma, the ultrastructure of surviving type I SGN soma appeared relatively normal over the 12-week period following deafening. Satellite Schwann cells exhibited greater survival traits than their type I SGN; however, on loss of neural contact, they reverted to a non-myelinating phenotype, exhibiting an astrocyte-like morphology with the formation of processes that appeared to be searching for new neural targets. In 6- and 12-week deafened cochlea, we observed cellular interaction between Schwann cell processes and residual SGNs that distorted the morphology of the SGN soma. Understanding the response of SGNs, Schwann cells, and the complex relationship between them following aminoglycoside deafening is important if we are to develop effective therapeutic techniques designed to rescue SGNs.

Keywords: Schwann cell; deafness; nerve degeneration, nerve regeneration, cochlear implant; spiral ganglion neuron.

FIG. 1

Schematic diagram of the histological sectioning used for SGN and peripheral fibre analysis. a The resin-embedded cochlea was bisected down the midline (dotted line) and the basal half-turn was removed (thick black line). All analysis was carried out on the basal half-turn only. b The basal half-turn was further divided into two regions (regions 1 and 2). A thick block (~1 mm) was cut in the ‘transverse’ plane in each of the two regions. A wedge-shaped ‘longitudinal’ block was then cut for the cross-sectional analysis of the peripheral fibres. c Nerve and residual cells in the organ of Corti (OC), at the habenula perforata (HP) within the osseous spiral lamina (OSL), and SGN cell bodies (located in Rosenthal’s canal) were examined in transverse sections. d Longitudinal blocks were cut at a distal or medial OSL position to obtain OSL cross sections that were used in analysis of the peripheral fibres as indicated in the schematic representation of a longitudinal OSL cross section. Peripheral fibre analysis involved measuring the area of the fibre axoplasm within the myelin sheath and categorisation of each fibre as anatomically normal or degenerated based on whether the area of the axoplasm occupied greater or less than 90 % of the myelin lumen (see schematic examples of these fibres).

FIG. 2

Representative light microscope images illustrating the deafness-induced changes in the basal turn from each cohort in the present study. a Normal hearing cochlea. b 2-week deafened. c 6-week deafened. d 12-week deafened. Scale bar = 100 μm. Arrowheads indicate the flattened sensory epithelium and arrows indicate the Rosenthal’s canal with reduced SGN density.

FIG. 3

The effect of duration of deafness on terminal nerve endings in the organ of Corti for region 1 (see Fig. 1). Low (top panel) and high magnification (bottom panel) images taken at the location of the inner hair cell (IHC) near the habenula perforata. a, e Images from a normal hearing cochlea showing terminal nerve endings, an IHC, and supporting cells (IP—inner pillar cell, TC—tunnel of Corti, arrow points to the tunnel spiral bundle). At high magnification (e), the basal portion of an IHC is contacted by an afferent fibre (a), which is in turn contacted by two efferent (e) vesiculated endings. b, f Images from a 2-week deafened cochlea. The IHC has degenerated but an inner pillar cell (IP) is still clearly visible. f Numerous nerve endings, both afferent (a) and efferent (e), were present within the residual structures of the organ of Corti. c, g Images from a 6-week deafened cochlea. The inner pillar cells (IP) were still clearly visible and the organ of Corti had some structural integrity as indicated by the presence of the tunnel of Corti (TC), which had begun to collapse. g The higher magnification image shows few remaining small afferent and efferent nerve endings (arrow a and e). d, h Images from a 12-week deafened cochlea. A monolayer of cells appeared at the place of the degenerated organ of Corti. At the habenula perforata (arrowhead), a small number of nerve profiles were still visible extending in between the basilar membrane and the monolayer epithelium (arrow). Scale bars (a–d = 5 μm, e–h = 1 μm).

FIG. 4

The effect of deafening on peripheral fibres. Representative images of peripheral fibres within the OSL (medial OSL sections), at low (top panel a–d) and higher (bottom panel e–h) magnification. a, e In the normal hearing cochlea, regularly shaped and densely packed fibres, together with some Schwann cells, were evident. b, f After 2 weeks of deafening, the overall density of myelin profiles had not changed; however, at high magnification (f), more irregular shaped myelin profiles (arrow), reduction in fibre area, and evidence of early stage of axoplasm retraction were observed. c, g Following 6 weeks of deafness, the density of fibres had decreased dramatically. Many remaining myelin profiles exhibit retraction of the axoplasm, but the remaining myelin sheaths were well preserved. d, h After 12 weeks of deafness, the density of fibres was further reduced. An example of a Schwann cell (d: arrow) that appeared to have lost connection with its peripheral fibres appeared to have dedifferentiated into a non-myelinating phenotype. h A Schwann cell (large arrow) not only provided myelin to a fibre with normal morphology (arrowhead) but also ensheathed remnants of two other degenerated fibres (double arrows). Scale bars (a–d = 5 μm, e–h = 2 μm).

FIG. 5

Morphological changes in peripheral fibres with duration of deafness. a Median fibre area for normal (Norm) cochleae and following 2 (2w), 6 (6w), and 12 (12w) weeks of deafness measured at the medial and distal position within the OSL. There was a significant reduction in fibre area following the onset of deafness (main effect, ANOVA, p < 0.001) that stabilised after 6 weeks. The median, 25th, and 75th percentiles are illustrated. b Analysis of the distribution of fibres undergoing degeneration from anatomically normal fibres. The mean density of anatomically normal fibres (solid lines) decreased with duration of deafness (as indicated by the significance bars). The density of degenerating fibres (dashed lines) increased significantly 2 weeks following deafness and then plateaued. Total fibre density is indicated by open squares. Note that the total 2-week fibre density count is greater than the normal controls. This is likely to be a result of a slight bias as a result of the significant reduction in fibre area at 2 weeks. Error bars = SEM.

FIG. 6

Peripheral fibres within the OSL. a OSL cross sections from a normal hearing cochlea. Schwann cells (arrow and arrowhead) appear to be providing myelin for a number of peripheral fibres (white stars) that were wrapped by small cytoplasmic processes. A Schwann cell (arrowhead) ensheaths both a myelinated fibre and an unmyelinated fibre (double arrow). b, c Higher magnification images from a normal cochlea showing examples of two peripheral fibres (asterisk) that were ensheathed by one Schwann cell. d–f Examples of two peripheral fibres (asterisk) that were ensheathed by a single Schwann cell in deafened cochleae. g–i Examples of Schwann cells (arrow) that appeared to have dedifferentiated in response to the loss of their peripheral fibres. Two fibres (g: double arrows) have undergone significant degeneration and were almost completely devoid of axoplasm. Scale bars (a = 2 μm, b–f = 1 μm, g = 5 μm, h and i = 2 μm).

FIG. 7

Myelin sheath of a normal and degenerating peripheral fibre. a, c A peripheral fibre in a normal cochlea with an intact myelin sheath. b, d A peripheral fibre in a 6-week deafened cochlea that had begun to degenerate as evidenced by the collapse and retraction of the fibre’s axoplasm. The myelin sheath surrounding the degenerating fibre appeared normal. e, f Examples of large unmyelinated fibres (arrow) from a 6-week deafened cochlea in a cross section (e) and a longitudinal section (f) of the OSL. Schwann cells ensheathed but did not myelinate these fibres. The morphological features of these fibres (relatively large size, neurofilament, and microtubule content) are consistent with neighbouring type I SGN peripheral fibres. Scale bars (a = 1 μm, b = 0.5 μm, c and d = 0.1 μm, e and f = 1 μm).

FIG. 8

Effect of deafness duration on type I SGN cell bodies. Representative images of SGN soma at low (top panel) and higher (bottom panel) magnification. a, e Examples from a normal hearing cochlea. a The soma of the type I SGNs were typically oval shaped, myelinated, and densely packed. The higher magnification image (e) shows the classical ultrastructure of a type I SGN cell body, covered by a few layers of myelin sheath, densely packed polyribosomes, numerous mitochondria, and lipidic granules (arrowhead). Nucleoli (arrow) are clearly visible in the cell nucleus. b, f Examples from a 2-week deafened cochlea. Very few, if any, changes in cellular ultrastructure were observed. A normal myelin layer (arrow) and polyribosomes (arrowhead) were evident. c, g Examples from a 6-week deaf cochlea. The cell density (quantified data in i) and soma area (quantified data in j) had decreased significantly. However, most of the remaining type I SGNs, although smaller, appeared to have a normal myelin sheath (arrow) and intracellular content (g; arrowhead). d, h Examples from a 12-week deafened cochlea. A progressive loss of type I SGNs was observed (65 % reduction by 12 weeks). However, again, the intracellular contents of most of the remaining type I SGNs appeared to be consistent with a normal cochlea including intact myelin (arrow). The graphs indicate type I SGN density (i) and soma area (j) plotted against duration of deafness. There was an ongoing decrease in type I SGN density (*ANOVA, p < 0.001) in addition to a decrease in soma area (*ANOVA, p < 0.003) that had stabilised by 12 weeks of deafness. Error bars represent standard error of the mean (SEM). Scale bars (a–d = 5 μm, e–h = 1 μm).

FIG. 9

Degeneration processes of type I SGNs following 6 weeks of deafness. a A type I SGN (arrow) in the process of degeneration via apoptosis. However, the Schwann cell (arrowhead) and its myelin sheath surrounding the neuron appeared normal. b Example of a type I SGN in a more advanced stage of apoptotic degeneration (arrow). The Schwann cell exhibited a normal soma (asterisk) and formed a ‘bed’ around the degenerating neuron. However, demyelination was also evident (double arrow) and some myelin remnants (arrowhead) can be seen. c Example of a type I SGN undergoing degeneration via a cytolytic process. The soma of the Schwann cell is indicated by asterisk. d Example of a more advanced cytolysis of a type I SGN still ensheathed by its satellite Schwann cell (arrowhead). Scale bars = 5 μm.

FIG. 10

Ectopic morphology of type I SGNs in 6-week deafened cochleae. Deafened cochleae contained type I SGNs that exhibited an altered morphology. In some cases, this was associated with neuronal sprouting. Although the cytoplasmic structure and myelin of these neurons appeared anatomically normal, the morphology varied substantially from the oval shape type I SGN observed in normal cochleae (e.g. Fig. 8a). Neural sprouting arising from the SGN soma (arrow; a and b) or the axon hillock (arrow; c) was observed in some SGNs of deafened cochleae. All examples in this figure were taken from 6-week deafened cochleae although similar features were observed in the 12-week deafened cohort. Scale bars (a = 5 μm, b and c = 2 μm).

FIG. 11

Schwann cell dedifferentiation. Images from a 6-week deafened cochlea. a An example of a Schwann cell (arrow) that had lost its SGN (asterisk) and had begun to dedifferentiate into a non-myelinating phenotype. Nearby is a type I SGN with normal morphology, myelinated by its Schwann cell (arrowhead). b An example of a Schwann cell (arrow) dedifferentiating after the loss of the SGN. This Schwann cell had an astrocyte-like profile, with long processes (double arrows) extending from the soma. c Another dedifferentiated Schwann cell extended processes from its soma and contacted the glial sheath of a surviving SGN (arrow). d A dedifferentiated Schwann cell (arrow) contacted new targets (arrowheads). An elongated process contacted two neurons already ensheathed (and myelinated) by their own Schwann cells. The region within the framed box is represented in higher magnification in e showing that the process appeared to alter the morphology of the contacted SGNs. f Higher magnification (framed in e) of this contact region showing more clear filaments between the dedifferentiated glial process membrane (arrow) and the glial sheath (asterisk) of one of the two neighbouring SGNs. Scale bars (a–c = 5 μm, d = 10 μm, e = 5 μm, f = 0.5 μm).

FIG. 12

Schematic representation of the key features of type I SGN degeneration and the dedifferentiation of myelinating Schwann cells in the transverse orientation. a In the normal cochlea, a number of peripheral fibres make synaptic connections with an inner hair cell. There is a full complement of SGNs with their soma in Rosenthal’s canal. Schwann cells are associated with both types of SGNs but only myelinate type I SGNs. b Two weeks following hair cell loss, the axoplasm within the peripheral fibres begins to shrink and retract towards the soma while the myelin sheath remains intact. There is no SGN loss at this stage. c After 6 weeks of deafness, retraction of the peripheral fibres is more extensive, and ~50 % of the type I SGNs have degenerated via apoptotic or necrotic mechanisms. The myelin sheath associated with early degenerated SGNs has also collapsed. Schwann cells lose connection with the degenerated SGNs and dedifferentiate temporally leaving behind a glial bed and sending out processes to sample the local environment. Residual type I SGNs exhibit altered morphology. d After 12 weeks of deafness, about two thirds of the SGNs have degenerated. Greater numbers of Schwann cells undergo dedifferentiation into a non-myelinating phenotype.