Moderate injury in motor-sensory cortex causes behavioral deficits accompanied by electrophysiological changes in mice adulthood

Abstract

Moderate traumatic brain injury (TBI) in children often happen when there's a sudden blow to the frontal bone, end with long unconscious which can last for hours and progressive cognitive deficits. However, with regard to the influences of moderate TBI during children adulthood, injury-induced alterations of locomotive ability, long-term memory performance, and hippocampal electrophysiological firing changes have not yet been fully identified. In this study, lateral fluid percussion (LFP) method was used to fabricate moderate TBI in motor and somatosensory cortex of the 6-weeks-old mice. The motor function, learning and memory function, extracellular CA1 neural spikes were assessed during acute and subacute phase. Moreover, histopathology was performed on day post injury (DPI) 16 to evaluate the effect of TBI on tissue and cell morphological changes in cortical and hippocampal CA1 subregions. After moderate LFP injury, the 6-weeks-old mice showed severe motor deficits at the early stage in acute phase but gradually recovered later during adulthood. At the time points in acute and subacute phase after TBI, novel object recognition (NOR) ability and spatial memory functions were consistently impaired in TBI mice; hippocampal firing frequency and burst probability were hampered. Analysis of the altered burst firing shows a clear hippocampal theta rhythm drop. These electrophysiological impacts were associated with substantially lowered NOR preference as compared to the sham group during adulthood. These results suggest that moderate TBI introduced at motorsenory cortex in 6-weeks-old mice causes obvious motor and cognitive deficits during their adulthood. While the locomotive ability progressively recovers, the cognitive deficits persisted while the mice mature as adult mice. The cognitive deficits may be attributed to the general suppressing of whole neural network, which could be labeled by marked reduction of excitability in hippocampal CA1 subregion.

Fig 1. LFP, electrode implantation, injury, percussion trace and electrode location.

(A), Positions of LFP and recording electrodes. The large filled circle indicates the craniectomy site for LFP. The small black square shows where the recording microwires implanted. The two stars represent the screws where grounding silver leads connected with. The screw in the frontal skull, as well as the grounding screws, help to secure the whole electrode assembly. (B), Macroscopic findings were seen in LFP injured mouse on DPI 16 showing the lesion confined to the cortex. (C), The single LFP waveform demonstrates a rapid upstroke followed by a slower downstroke within 10 ms pulse duration. (D), An example of cresyl violet staining shows the actual location of the recording electrode in mouse CA1 subregion.

Fig 2. A typical burst firing recording and its inter-spike interval histogram.

(A), typically isolated burst firing waveform recorded from CA1 pyramidal cell. (B), inter-spike interval histogram of the isolated burst firing in CA1 cell.

Fig 3. The Motor function changes after TBI.

Mice motor performance in term of beam-walk latency (A) and foot-fault (B) were significantly impaired due to moderate TBI. Multiple Holm-Sidak t test were used to compare the daily based beam-walk or foot-fault between TBI and sham groups. Data are expressed as mean±SEM (n = 7). * p<0.05, ** p<0.01 and ***, P < 0.001 vs. control.

Fig 4. Object preference (mean percent exploration time) evaluation on DPI 9 and 16.

The sham group performs significantly better than TBI group on the novel object (object-C) testing on both DPI 9 (A) and 16 (B), indicating TBI consistently deteriorated cognitive memory after injury. Object-A, B, identical objects; Separate ANOVAs were used to compare the preference for all the familiar and novel objects during the acquisition phase (object-A, B test) and following novel object test phase (object-A/B and object-C test). Data are expressed as mean±SEM (n = 7–8). * p<0.05, ** p<0.01 vs. control.

Fig 5. TBI impaired learning and spatial memory retention in mice.

TBI mice demonstrated an increased latency to find the platform, compared to the sham mice in acquisition phase (A, DPI 1 to 5 and B, DPI 8 to 12). During trials in probe phase (DPI 6, p<0.05 and DPI 13, p<0.05), TBI mice spent a significantly less percentage of time in the target quadrant, compare to the sham mice. Data are expressed as mean±SEM (n = 7). * p<0.05, vs. control by control by Multiple Holm-Sidak t test or ANOVA with Newman-Keuls post hoc test.

Fig 6. Several aspects of burst firing in CA1 were altered following moderate TBI.

Significant decrease of overall mean frequency on DPI 14 (A left inset), and burst per minute on both DPI 7 and 14 (A right inset), increase of burst duration on DPI 14 (B), reduction of mean spikes in burst (C), significant increase of mean frequency in burst on 7 DPI (D), significant increases of mean ISI in burst (E) and mean interburst interval (F) on DPI 14. Data are expressed as mean±SEM (n = 23–24). * p<0.05, ** p<0.01, *** p<0.001 vs. control by unpaired t-test or ANOVA with Newman-Keuls post hoc test.

Fig 7. Results of burst analysis in the CA1 subregion after moderate TBI.

Representative instant firing, autocorrelogram and power spectral density in CA1 pyramidal cell of sham mice on DPI 7 (panel A) and 14 (panel C). Decreases of instant burst firing (left column) in CA1 pyramidal cell of TBI mice on DPI 7 (panel B) and on DPI 14 (panel D) are confirmed by the reductions of frequency in autocorrelogram (1ms bin, middle column) and drops of power spectral density within theta range (right column). Data in three graphs of each panel (A-D) are from the same cell.