Distribution of glial cells in the auditory brainstem: normal development and effects of unilateral lesion

Abstract

Auditory brainstem networks facilitate sound source localization through binaural integration. A key component of this circuitry is the projection from the ventral cochlear nucleus (VCN) to the medial nucleus of the trapezoid body (MNTB), a relay nucleus that provides inhibition to the superior olivary complex. This strictly contralateral projection terminates in the large calyx of Held synapse. The formation of this pathway requires spatiotemporal coordination of cues that promote cell maturation, axon growth, and synaptogenesis. Here we have examined the emergence of distinct classes of glial cells, which are known to function in development and in response to injury. Immunofluorescence for several astrocyte markers revealed unique expression patterns. Aldehyde dehydrogenase 1 family member L1 (ALDH1L1) was expressed earliest in both nuclei, followed by S100ß, during the first postnatal week. Glial fibrillary acidic protein (GFAP) expression was seen in the second postnatal week. GFAP-positive cell bodies remained outside the boundaries of VCN and MNTB, with a limited number of labeled fibers penetrating into the margins of the nuclei. Oligodendrocyte transcription factor 2 (OLIG2) expression revealed the presence of oligodendrocytes in VCN and MNTB from birth until after hearing onset. In addition, ionized calcium binding adaptor molecule 1 (IBA1)-positive microglia were observed after the first postnatal week. Following hearing onset, all glial populations were found in MNTB. We then determined the distribution of glial cells following early (P2) unilateral cochlear removal, which results in formation of ectopic projections from the intact VCN to ipsilateral MNTB. We found that following perturbation, astrocytic markers showed expression near the ectopic ipsilateral calyx. Taken together, the developmental expression patterns are consistent with a role for glial cells in the maturation of the calyx of Held and suggest that these cells may have a similar role in maturation of lesion-induced connections.

Keywords: astrocyte; auditory system; brainstem; deafferentation; microglia; oligodendrocyte.

Figure 1. Expression of glial markers in VCN at P0

(A) At P0, GFAP showed minimal expression in VCN but was evident in the adjacent central projection from the auditory nerve. (B) ALDH1L1 is expressed throughout VCN with less expression in DCN. (C) Similar to GFAP, S100β was expressed in the auditory nerve but excluded from VCN. (D) OLIG2 expression revealed limited cells along the ventral aspect of VCN and in the auditory nerve. (E and E′) IBA1 was expressed in a small number of cells in VCN that were mostly amoeboid in shape with few processes (panel E′ another representative section). Scale bar in E: 100 μm, applies to Panels A–E. Scale bar in E′: 50 μm.

Figure 2. Expression of glial markers in MNTB at P0

(A) GFAP expression was limited to the ventral portion of the brainstem outside MNTB. (B) S100β expression was minimal and limited to the dorsal region of MNTB. (C) ALDH1L1 was expressed by fibers surrounding MNTB principal neurons, and more generally throughout the brainstem. ALDH1L1 cells had several processes (C′). (D) OLIG2-positive cells were seen along the ventral brainstem. (E) IBA1- positive cells were found in brainstem, but not found in MNTB. Scale bar in E: 100 μm, applies to all panels except panel C′ (scale bar: 50 μm).

Figure 3. Expression of glial markers in VCN at P6

(A) Similar to P0, GFAP expression is limited to the auditory nerve. (B–C) ALDH1L1 and S100β have similar expression patterns in VCN, including the presence of immunopositive fibrous processes. S100β-positive cells are seen in auditory nerve and in VCN encapsulating the globular bushy cells (C′). (D) OLIG2-positive cells were seen throughout VCN. (E) Few IBA1-positive cells could be observed throughout the VCN. Scale bar in F: 100μm, applies to all panels except C′ (scale bar: 50 μm).

Figure 4. Expression of glial markers in MNTB at P6

(A) While all other astrocytic markers were present throughout brainstem, GFAP expression was limited to fibers in the ventral region of brainstem with more processes in MNTB than at younger age (GFAP at P6 was 595.54 ± 70.14 was significantly higher than GFAP at P0 (143.12 ± 68.79, p = 0.03). (B–C) Expression of ALDH1L1 and S100β revealed processes throughout MNTB. (D) OLIG2 immunofluorescence was seen in spherical cells within MNTB. (E) IBA1-positive cells were found throughout the brainstem, particularly at the ventral margin. IBA1-positive cells in MNTB mostly had small cell bodies with elongated processes (E′). Scale bar in F: 100 μm, applies to panels A–E. Scale bar in E′ is 50 μm.

Figure 5. Expression of glial markers in VCN at P14

(A) At P14, an increase in GFAP-positive fibers was seen in VCN in comparison to earlier ages (p=0.05; the GFAP O.D. at P0 was 294.74 ± 3.41 while the GFAP O.D. at P14 was 1158.50 ± 165.08). (B–E) Astrocytic markers GLAST, S100β and ALDH1L1 were all expressed in VCN. (E) Cellular OLIG3 expression was seen throughout VCN. (F) Several IBA1 expressing cells could be identified within VCN. Scale bar in F: 100 μm, applies to all panels.

Figure 6. Expression of glial markers in MNTB at P14

(A and B) GFAP-positive fibers were observed penetrating MNTB from the ventral boundary and was significantly increased compared to P0 (p = 0.001; GFAP at P14 was 821.08 ± 23.11 compared to GFAP at P0 (143.12 ± 68.79). (C–D) Similar to younger ages, S100β and ALDH1L1 were expressed throughout the brainstem and inside MNTB, surrounding principal neurons. (E) OLIG2 expression showed few cells in MNTB. (F) IBA1 expression had cell bodies penetrating MNTB with many cells possessing intricate processes (inset). Scale bar in F: 100 μm, applies to all panels.

Figure 7. Expression of glial markers in VCN at P23

(A) At P23, an increase in GFAP-positive fibers observed inside VCN with several processes (highlighted in A′ from a representative section). (B–C) Relatively low levels of expression were seen for S100β and ALDH1L1 surrounding VCN neurons. (D) VCN was populated with OLIG2-positive cells, (E) IBA1-positive cells were considerably less numerous than at earlier ages. Scale bar in E: 100 μm, applies to all panels except for A′, which has a scale bar of 50 μm.

Figure 8. Expression of glial markers in MNTB at P23

(A) At P23 there was significantly less GFAP-positive processes found in MNTB compared to at P14 (p = 0.02; GFAP at P23 was 368.97 ± 37.02 and GFAP at P14 was 821.08 ± 23.11). (B) S100β had low level of expression in brainstem. In contrast, many ALDH1L1- positive cells were surrounding individual MNTB principal neurons (C–C′). (D) OLIG2 expression was located ventral to MNTB. (E) IBA1-positive microglia were observed in a sparse distribution throughout MNTB. Scale bar in E: 100μm, applies to all panels except panel C′ (scale bar: 50 μm).

Figure 9. Expression of glial markers in relation to the calyx at P6 and P14

(A–E) Glial cell types near P6 calyx. (A–B) Cells positive for ALDH1L1 and S100β were in close apposition to labeled calyces (calyces indicated with arrows). (C) Few GFAP-positive populations found within MNTB. (D) IBA1-positive cells were found scattered across MNTB, often close to the labeled calyx. (E) Oligodendrocytes fully surround postsynaptic neurons. (F–J) Glial cell types near P14 calyx. (F–G) All astrocytic markers found expressed close to developing calyx. ALDH1L1 and S100β-positive astrocytes were in close proximity to calyx and appeared to align closely to the MNTB neuron. (H) Several GFAP processes extended into MNTB and in some cases were seen in proximity to labeled calyces. (I) IBA1-positive cells were located near calyx and had larger cell bodies compared to younger age. (J) OLIG2-positive staining looked similar to younger age, often near the MNTB neuron. Scale bar in J, 25 μm; applies to all panels.

Figure 10. Expression of glial markers in relation to ectopic calyx seven days following CR

(A) ALDH1L1-positive astrocytes located along postsynaptic space unoccupied by newly formed calyx, similar to contralateral calyx (calyces indicated with arrows). (B) S100β cells and processes situated around MNTB neurons in both contralateral and ectopic ipsilateral calyx. (C) Several GFAP-positive processes extended into MNTB region, with a majority of cells contacting both contralateral and ectopic calyces (example of close apposition in C″). (D) Unlike astrocytic populations, few microglia found near calyx. (E) Very few OLIG2-positive expression near calyces. Scale bar in E; 25 μm; applies to all panels.

Figure 11. Expression of glial markers two and seven days following unilateral CR

(A) No significant difference in expression pattern of glial markers two days after CR (ANOVA, p = 0.23). ALDH1L1 staining in CR animals had a mean D/I ratio of 1.06 ± 0.03 compared to sham animals (1.05 ± 0.05). (B) Expression of S100β cells for CR groups had a D/I ratio of 1.0 ± 0.05 compared to sham group (0.96 ± 0.03). (C) OLIG2 expression had a similar result when comparing CR groups to sham (1.14 ± 0.07 and 0.77 ± 0.13, respectively). (D) Number of GFAP-positive processes was not different - the CR group had a mean D/I ratio of 0.85 ± 0.15 processes compared to sham operated animals, which had a D/I ratio of 1.38 ± 0.41. (E) Number of microglial cells comparing CR animals to sham was not significant (0.86 ± 0.09 compared to 0.88 ± 0.05). (F) One week after CR, no differences in glial marker distribution seen when comparing sham to CR groups (ANOVA, p = 0.13). Staining of ALDH1L1 showed that CR groups had a mean D/I ratio of 0.82 ± 0.06 and were not significantly different compared to the sham group (0.94 ± 0.07). (G) S100β expression between CR and sham was not different (0.9 ± 0.03 and 0.92 ± 0.03, respectively). (H) OLIG2 expression between CR and sham groups was not different with the CR group D/I ratio of 1.03 ± 0.07 and sham group D/I ratio of 1.04 ± 0.12. (I) When comparing GFAP processes, animals subjected to unilateral CR had a D/I ratio of 1.12 ± 0.17, a value greater than sham-operated animals but not significantly different (0.71 ± 0.15). (J) No change in between was seen in IBA1-positive cell density D/I ratio between CR and sham groups (1.0 ± 0.01 and 0.99 ± 0.01). Values are reported as mean ± SEM.