Dendritic alterations after dynamic axonal stretch injury in vitro

Abstract

Traumatic axonal injury (TAI) is the most common and important pathology of traumatic brain injury (TBI). However, little is known about potential indirect effects of TAI on dendrites. In this study, we used a well-established in vitro model of axonal stretch injury to investigate TAI-induced changes in dendrite morphology. Axons bridging two separated rat cortical neuron populations plated on a deformable substrate were used to create a zone of isolated stretch injury to axons. Following injury, we observed the formation of dendritic alterations or beading along the dendrite shaft. Dendritic beading formed within minutes after stretch then subsided over time. Pharmacological experiments revealed a sodium-dependent mechanism, while removing extracellular calcium exacerbated TAI's effect on dendrites. In addition, blocking ionotropic glutamate receptors with the N-methyl-d-aspartate (NMDA) receptor antagonist MK-801 prevented dendritic beading. These results demonstrate that axon mechanical injury directly affects dendrite morphology, highlighting an important bystander effect of TAI. The data also imply that TAI may alter dendrite structure and plasticity in vivo. An understanding of TAI's effect on dendrites is important since proper dendrite function is crucial for normal brain function and recovery after injury.

Fig. 1

Axon stretch injury apparatus. Schematic representation of the steel well (A) in which two silicon barriers are placed over the elastic membrane to generate two cell-free gaps. Each barrier contains laddering micro-channels (B) that allow axon growth on 2 mm longitudinal tracks.

Fig. 2

Composite fluorescent photomicrographs of neuronal processes double-stained for (A) NF200 and (B) MAP2, showing that axon tracts are devoid of cell bodies and dendritic processes. Instead, cell bodies and dendrites are restricted to each side of the gap region. Both panels represent the same microscopic field of double-labeled rat cortical neurons after 11 days in culture. Scale bar=50 μm.

Fig. 3

Representative fluorescent photomicrographs of NF200-labeled neuronal processes illustrating the temporal evolution of axons' delayed elastic response to dynamic stretch injury. Before stretch, axons are straight. Immediately after injury, they develop large undulations before gradually recovering their initial shape and length. Scale bar=10 μm.

Fig. 4

Fluorescent photomicrographs of control (uninjured) neurons (A and B) immunostained for the dendritic marker MAP2, and neurons displaying beaded dendrites 5 min after axonal stretch (C, D, and E). The dashed line in (A) indicates the boundary between the axon-only gap area (left) and the neuronal cell body region (right). Scale bar=50 μm (A) and 25 μm (B, C, D, and E).

Fig. 5

Histogram illustrating the time-course of dendritic bead formation. After dynamic stretch injury of axons, cultures were fixed at various time-points, and double-stained for MAP2 and NF200. The number of beads per dendrite Length Unit (100 μm) was determined at ×400 magnification. Values represent mean±S.E.M. from 3 independent experiments. *p<0.0005 vs. 5 min.

Fig. 6

Histogram (A) and graph (B) showing inter-bead distances expressed as a percentage of bead pairs. After dynamic stretch injury of axons, cultures were fixed at various time-points and double-stained for MAP2 and NF200. Inter-bead distances were measured along dendritic processes at ×800 magnification, and were graphed as bead pairs. Values represent mean±S.E.M. from 3 independent experiments. *p<0.005 and **p<0.0001 vs. 5 min.

Fig. 7

Histogram (A) and graph (B) of bead perimeter distribution. Cells were axon-stretched, then fixed at various time-points and double-labeled with MAP2 and NF200. Bead perimeter was determined at ×800 magnification. Values represent mean±S.E.M. from 3 independent experiments.

Fig. 8

Pharmacological modulation of axonal stretch injury-induced dendritic beading. Fluorescence photomicrographs of dendritic processes immunostained with the dendritic marker MAP2. Cultures were fixed 5 min after dynamic stretch injury of axons. Compared with untreated cultures (A), pre-treatment with the VGSC inhibitor tetrodotoxin (TTX, 1 μM) (B), removal of extracellular sodium (C), or pre-incubation with the NMDA receptor antagonist MK-801 (20 μM) (D), prevented dendritic beading. Scale bar=25 μm.

Fig. 9

Effect of extracellular calcium on axon stretch injury-induced dendritic beading. Fluorescent photomicrographs of dendritic processes immunostained for MAP2 demonstrate smooth processes of uninjured cultures pre-incubated in calcium-free medium (A). However, 5 min after dynamic stretch injury of axons, dendrites displayed extensive beading (B). Histograms (C) and (D) illustrate quantitative analysis of dendritic beading and show a decreased number of beads (C) and an increased bead perimeter (D) in calcium-free medium after stretch, compared tocalcium-containing medium. Bead number and perimeter were determined at ×400 and ×800 magnification, respectively. Values represent mean±S.E.M. from 3 independent experiments. *p<0.0001 vs. calcium-containing medium. Scale bar=25 μm in (A and B).