Behavioural Brain Research 93 (1998) 143 – 155
Mild traumatic lesion of the right parietal cortex of the rat: Selective behavioural deficits in the absence of neurological impairment Sandy Hogg 1,a, Paul C. Moser a,*, David J. Sanger b b
a Synthe´labo Recherche, 10 rue des Carrie`res, 92500 Rueil-Malmaison, France Synthe´labo Recherche, 31 a6enue Paul Vaillant-Couturier, 92220 Bagneux, France
Received 26 May 1997; accepted 8 October 1997
Abstract Fluid impact models are widely used to study the histological and neurochemical consequences of traumatic brain injury and although behavioural consequences have also been studied, behavioural changes are often confounded by non-specific neurological deficits. In the present study we investigated behavioural effects of a unilateral mild traumatic lesion of the right lateral parietal cortex. This region is implicated in a number of basic and complex behaviors, and we therefore analyzed the performance of rats in a diverse range of behavioural procedures. The lesion had no effects on general neurological function, motor activity (activity boxes, rota-rod and paw reaching tests), habituation to a novel environment (holeboard), spatial learning ability (Morris water maze) or anxiety (elevated plus-maze). However, the lesioned animals demonstrated lower levels of exploration than the control group when novel objects were placed beneath some of the holes in the holeboard. Lesioned animals also differed from controls in their performance in passive and active avoidance procedures. In a step-through passive avoidance test the lesioned rats performed worse than the sham-operated controls, i.e. they had significantly lower entry latencies on the 2nd day. In contrast, in the active avoidance task the lesioned animals performed better than sham-operated rats, demonstrating a better ability to learn to avoid and escape from the shock. These diverse results in different tests of learning and memory, in particular the impairment in passive avoidance and the improvement in active avoidance behavior, are difficult to reconcile with a simple effect of the lesion on cognitive performance per se. The complete absence of general neurological deficits following the mild traumatic injury rules out the possibility that the observed behavioural changes reflect a non-specific impairment. These results demonstrate that mild traumatic lesion of the right parietal cortex can induce relatively selective behavioural changes that may serve to study functional recovery after trauma. However further work is required to establish the underlying deficit(s) that has led to the behavioural effects described here. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Cerebral trauma; Neurologic function; Motor activity; Paw reaching; Water maze; Passive avoidance; Active avoidance; Plus-maze; Holeboard
1. Introduction Each year in the USA approximately 2 million people —an estimated 460000 of whom are hospitalised [14] —suffer from head injuries of sufficient severity to result in brain trauma. A certain proportion of these * Corresponding author. Tel.: +33 141391396; fax: + 33 141391306. 1 Present address: H. Lundbeck A/S, Ottiliavej 9, DK-2500 Copenhagen-Valby, Denmark. 0166-4328/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0166-4328(97)00146-0
people die, whilst those that survive are likely to suffer from any of a large number of physical or behavioural disabilities, such as motor inco-ordination or paralysis, as well as speech and cognitive impairments [30]. There is, therefore, a great need for effective therapeutic intervention, either neuroprotective or neuroregenerative. Traumatic injury of the central nervous system results in neuronal damage by at least two mechanisms. There is a direct, mechanical lesion which is surrounded by an area in which indirect damage occurs. In this
144
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
penumbral region a cascade of events occur, which are not dissimilar to those observed in classical ischaemic lesions and include haematoma, increased release of neurotransmitters (including excitatory amino acids) and alterations in cell membrane permeability (allowing increased calcium uptake). The biochemical sequalae to cerebral trauma, which are extensive, are reviewed in detail elsewhere [17,18]. By virtue of its physical location, the cortex is most susceptible to traumatic injury and the extent of damage beyond this structure (both necrotic and diffuse damage) is dependent on the force of the physical impact. One of the more widely used methods of inducing traumatic brain injury which provides a high degree of consistency, with respect to both the size of the lesion produced and the survival rate of the animals, is fluid percussion. Induction of traumatic lesions using fluid percussion involves the exertion of a brief, controllable and measurable liquid impact against the brain. The lesions produced are reproducible and have enabled the study of histological and neurochemical consequences of cerebral trauma [17,18,32]. Using histological endpoints, pharmacological manipulations have indicated the neuroprotective efficacy of a number of compounds [34]. However, it is important to know if these are indicative of an attenuation of a functional deficit as this is the desired outcome of therapeutic intervention. Analyses of behavioural consequences of traumatic head injury could provide these functional correlates and a number of studies have already been carried out to investigate this issue [8,21,26]. For example, lesions produced by exerting a brief fluid impact (2.1 – 2.9 atm) vertically on to the brain midway between lambda and bregma either centrally over the sagittal sinus or over the dorsolateral parietal cortex produced spatial learning or retention deficits in the Morris water maze [8,27]. Traumatic lesion of a more lateral part of the parietal cortex using a lower percussion pressure (1.5 – 1.6 atm) directed horizontally onto the cortex has also been carried out to investigate the neurochemical correlates of brain injury [32] and the effects of potentially neuroprotective compounds [33,34]. The aim of the investigation reported here was to establish behavioural consequences of this traumatic lesion of the parietal cortex. The parietal or association cortex was originally considered to be involved in the processing and integration of complex sensory information. This role is not questioned although it appears that ‘control centres’ for many diverse functions are located, in part, in the parietal cortex [2]. Clinical evidence has implicated the posterior parietal cortex in disorders of gross and fine motor control spatial perception, attention (normally a neglect which is directed to the side of the body contralateral to the site of the lesion), sensation and affect [16]. As numerous behavioural outcomes of lesions to
this area are therefore possible, a battery of tests was carried out in rats having undergone traumatic lesion of the lateral parietal cortex to analyse general neurological function, forced and spontaneous motor activity, fine motor control and coordination, acquisition of passive and active avoidance tasks, spatial learning ability and anxiety.
2. Materials and methods
2.1. Animals Adult male Sprague-Dawley rats (6–7 weeks old and weighing 160–200 g at the time of surgery, Iffa Credo, L’Arbresle, France) were used for all experimental procedures. They were housed in groups of five from the date of arrival (7 days before the start of experimentation) and allowed free access to food and water (except where detailed in Section 2.3). Different groups of rats were employed for each of the different behavioural procedures described, except as specified below. A total of 175 animals were used in the studies reported here.
2.2. Surgery Cerebral trauma or sham operation were carried out according to the method of Toulmond et al. [32,34]. Briefly, animals were anaesthetised with sodium pentobarbitone (60 mg/kg, i.p.) and placed in a stereotaxic frame. The top and the right side of the skull were exposed. A hole (approximately 4 mm in diameter) was trephined at the level of the right parietal cortex (its centre being located 3.5 mm posterior to bregma, 7 mm lateral to the midline and 3.5 mm below the upper surface of the skull). A teflon tube (1.6 mm internal diameter, 3.2 mm external diameter) was placed in contact with the dura and fixed into the craniotomy with dental acrylic cement such that a fluid percussion would be directed horizontally onto the cortex (parallel with the intra-aural line). The opposite end of the tube was connected to a solenoid valve which was in turn connected to a Beckman HPLC pump using the same internal diameter tubing. The pump/tubing system was filled with sterile, injectable water and once the pump had reached the predetermined pressure (5 bar), fluid percussion injury of moderate severity was induced by a brief (100 ms) opening of the solenoid valve (model 017650D, Bu¨rkert, Germany). The total length of tubing after the valve was 96 cm. The actual pressure applied to each animal was monitored using an electronic transducer linked to an oscilloscope and was between 1 and 1.5 bar. Following brain trauma the tubing was removed, the scalp sutured and the animals placed under a heating-lamp for at least 2 hours before being returned to the animal holding room. Sham-oper-
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
ated animals were prepared in the same way with the exception that the liquid impact was not carried out. The lesions produced using this technique in rats of this size and strain have an average volume of approximately 10 mm3. In Fig. 1 sample data (from animals tested in the passive avoidance and rota-rod experiments reported here) are presented showing largest and smallest lesion sizes.
2.3. Beha6ioural procedures The same groups of control and lesioned animals were assessed for their neurological scores and rota-rod performance. Independent groups of animals were used for all other tests. In all experiments (except the water maze and the neurological scoring) an ‘odour’ rat was tested (and its data ignored) before the test proper; this served to scent the apparatus such that all the animals were tested under the same conditions. Test apparatus were wiped with a damp cloth between animals to remove urine and faeces.
2.4. Neurological and motor beha6iours 2.4.1. Neurological scoring Rats underwent a variety of tests which assessed their level of consciousness, rate and depth of respiration, cranial nerve function, motor and sensory function and
Fig. 1. Representative brain slices taken from animals killed 7 days after traumatic lesion of the right parietal cortex. Light and dark shaded areas represent, respectively, the largest (14.25 mm3) and smallest (5.05 mm3) lesions observed. The mean ( 9SEM) lesion size for 41 rats was 9.65 (9 1.14) mm3. The numbers indicate the distance in mm from the bregma [22].
145
behaviour. A neurologic score (0–267) was assigned on each of days 1, 2, 3, 4, 7, 8, 9, 10 and 14 following surgery according to a revised schedule of Cahn et al. [3] (Table 1).
2.4.2. Rota-rod All sessions consisted of a 5-min trial using a procedure in which the rotation speed increased from 3 to 30 rpm. The duration for which the rats could stay on the rota-rod was measured. Animals that fell off during the first 10 s were put back on. In addition the first session was immediately preceded by a 2-min habituation period where the rats were placed on the apparatus at its slowest speed (3 rpm) and replaced each time they fell. Animals were tested on days 1–5 following surgery. 2.4.3. Locomotor acti6ity Locomotor activity was measured in clear perspex boxes measuring 37×37×15 cm high. Infrared cells were located at the centre of each side such that there were two beams which crossed at the centre of the cage. Activity was measured as the total number of beam breaks made during a 30-min test. The same group of animals was tested on days 1, 7 and 8 following surgery. 2.4.4. Paw reaching Paw reaching was measured using a modified version of the ‘staircase test’ [19] which consisted of a grey perspex box (400× 70× 130 mm high) covered with a clear perspex lid. The inside of this box could be considered as two, equally sized chambers, the first (200× 70×130 mm) was empty and provided a start box into which an animal was placed at the beginning of each test session. In the second chamber there was a central, raised platform (72 mm above the floor of the box) which ran the length of the chamber such that there was a narrow trough (15 mm) on each of its sides. The narrowness and reduced height of the second chamber prevented the animals from turning round, such that they could only reach into the trough to the left of the central platform with their left paw and likewise into the right trough with their right paw. Removable staircases (seven steps each, the height difference between each step being 8 mm) were placed in the troughs and food pellets (45 mg, 20M precision food pellets, Campden Instruments, UK) were placed on the steps (four per step) such that the animals were required to lie on the central platform and reach down into the troughs to retrieve the pellets. Rats were trained in the paw reaching task before surgery. For 3 days before the start of the training procedure the animals were placed on a limited food regimen and were maintained at 85–90% of their free feeding weight by allowing them free access to food for only 30 min at the end of each day. During this time
146
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
Table 1 Neurological scoring schedule Level of consciousness Normal Convulsions Stupor Coma Respiration Normal Dyspnea Gasping Cranial nerve function Corneal reflex (×2) Normal Sluggish Absent Pupil size (×2) Normal Miosis Mydriasis Light reflex (×2) Normal Sluggish Absent Ptosis (times 2) Absent Present Auditory response Normal No response Motor and sensory function Traction reflex Four paws Three paws Two paws No grasping Motor response to painful stimulus (foot pinch) Normal Sluggish No reaction Behaviour Bristling of fur Absent Present Tremor Absent Present Movements Frequent Rare Gait Normal Abnormal Grooming Present Absent Sniffing and rearing Frequent Rare Absent Curiosity Present Absent
0 25 50 100 0 20 40
0 2 4 0 2 4 0 2 4 0 4
the animals were handled and weighed daily and habituated to the pellets to be used in the paw reaching task. Training started with 5 days of habituation to the test box where food pellets were placed throughout the box for the first 2 days, on the central platform and the staircases for the next 2 days and on the 5th day pellets were located only on the steps of the staircase (four per step). On each of the days animals were allowed 10 min free exploration of the box. During the second, and all subsequent, weeks of training the pellets were placed on the steps only and, once again, the rats were permitted free exploration for 10 min. At the end of this time the location of the remaining pellets was noted and the number of pellets which had been eaten was calculated individually for the left and right sides of the platform. Animals were trained for 2 weeks and put back on a normal free feeding schedule for 5 days before the trauma was carried out. Following surgery animals were allowed to feed freely for a further 4 days before the restricted diet recommenced. Testing in the paw reaching task was recommenced on the 6th day after trauma or sham operation.
2.5. Cogniti6e function and anxiety 0 5
0 2 5 10 0 5 10
0 10 0 10 0 10 0 10 0 10 0 5 10 0 10
2.5.1. Passi6e a6oidance The passive avoidance apparatus consisted of a perspex box (36× 18×20 cm high) divided into equally sized white and black compartments by a central wall. In this central wall was an opening (7× 7 cm) through which the animals could pass from one compartment to the other, which could be closed with a guillotine door. The floor of the white compartment was made of white perspex whilst that in the black compartment was a shockable grid constructed of 4 mm diameter metal bars, their centres spaced 11 mm apart. Animals were tested on days 5 and 6 following surgery. Food and water were removed from the animals’ cages 60 min before each test session. Testing consisted of two test sessions carried out 24 h apart. During the first the rats were placed in the light half of the apparatus facing away from the central wall The latency for each animal to enter the dark compartment was measured. Once in this section the door in the central wall was closed such that the animal could not pass back into the white section and a 0.6 mA, 2 s scrambled electric shock (grid floor shocker, Coulbourn Instruments, USA) was administered. The rat was immediately removed from the box by the tail and replaced in its home cage. During the second test session each rat was again placed in the light compartment facing the wall opposite the central partition and was observed for 3 min. The latency to enter, number of entries made into, and time spent in, the dark compartment were measured.
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
2.5.2. Acti6e a6oidance Active avoidance was measured using a shuttle box (Ugo Basile, model 7530) measuring 21× 48 × 22 cm high divided into two equally sized sections across its width by a metal partition with a circular opening in the centre large enough for the rat to pass through. The exterior walls were made from grey perspex, the cover of the box from clear perspex and the floor of the box from metal rods, 3 mm in diameter placed 12 mm apart through which a scrambled electric shock could be delivered. The floor was designed such that it would tilt when the rats passed through the opening in the central partition from one end of the box to the other. Behavioural testing was on days 4 – 9 following surgery. Each session was 20 min long and consisted of 40 trials. A trial consisted of 10 s presentation of both light (200 lux) and sound (70 db, 670 Hz) stimuli, during the last 5 s of which a scrambled 0.6 mA electric shock was delivered through the floor. The time interval between the start of two consecutive trials was 30 s. The stimulus was terminated either when the rat passed from one half of the box to the other (thus changing the tilt of the floor) or after 10 s, whichever was the sooner. Thus the rat could avoid a shock by moving through the opening in the partition during the first 5 s of the light and sound stimuli and escape it by moving during the second 5 s. The behaviours recorded were the avoidances of the shock (number of times the rat changed ends during the first 5 s of the light and sound stimuli), the number of escapes from the shock (number of times the rat changed ends during the second 5 s of stimuli, i.e. when the shock was being delivered), the number of failures (when the rat failed to cross from one side of the box to the other during the trial period), the latency to cross, measured from stimulus onset, and the number of inter-trial crossings (number of times the rats crossed from one half of the box to the other between trials, i.e. when neither conditioning stimuli nor shock was being delivered). 2.5.3. Holeboard The holeboard used was a perspex box 60 × 60× 35 cm with nine holes, each 4 cm in diameter, arranged in a 3× 3 configuration and equally spaced on the floor. Locomotor activity and rearing were measured by the interruption of infrared beams from cells located in the walls of the box 4.5 and 12.5 cm from the floor. Head dipping was measured by the interruption of infrared beams from cells located immediately beneath the edges of the holes. These allowed the recording of total time spent exploring the holes and number of head-dips. Number and duration of beam breaks were monitored and scores were entered directly into a computer such that behaviour could, if required, be regarded as a function of time and the head-dipping could be analysed for each hole independently.
147
Testing was started on day 5 following surgery and consisted of four 5-min trials which were carried out on consecutive days. On the first 3 days the holes were empty and on the 4th day novel objects (a bottle top, some staples and a magnetic stirrer) were placed under three of the holes (object holes) while the other six holes were left empty (non-object holes). The position of the objects was counter-balanced across the groups in a semi-random fashion using four possible configurations. The time spent head-dipping and number of head-dips in the object and non-object holes were analysed together to enable comparison of exploratory behaviour with that exhibited during the first 3 days of the test, and separately to enable separation of exploration directed specifically towards the objects from general investigatory behaviours.
2.5.4. Morris water maze The maze used was based on that employed by Morris [20] to measure the spatial learning ability of rats. It consisted of a circular pool measuring 1.5 m in diameter and 50 cm high which was filled with water to a level 20 cm from its upper rim. The water, which was made opaque by addition of 5 l of sterilised milk was maintained at 28–30°C and changed daily. For the purpose of definition of positions within the pool it was considered to consist of four quadrants; north-east, south-east, south-west and north-west. A square platform (10 × 10 cm) was placed at the centre of one of these quadrants such that its top was located 2.5 cm below the surface of the water. The platform position was fixed for the duration of the training. The entire maze was located in the centre of a well lit room measuring approximately 3.5 m2. Training started on day 5 following surgery and consisted of three trials a day for 4 days with the platform being positioned in the north-west quadrant. For the start of each trial the rat was released from one of three possible start positions (north, east or south) and allowed to swim until either it found the platform or 120 s had elapsed, whichever was the sooner. Behaviour was monitiored by video camera mounted vertically overhead which was linked, via a VP118 tracking system (HVS Image), to a video recorder, monitor and computer. Analyses of swim paths were carried out using the Watermaze for Windows program (version 1.02; Watermaze Software, Edinburgh, UK). The parameters measured were the latency of the rat to find the platform (training latency), the speed of the rat to move about the pool (swim speed) and the percentage of time that the rat spent 10 cm or less from the wall of the pool (wall edge analysis). On the 5th day a probe trial was carried out in the absence of the platform. Rats were allowed to swim freely in the pool for 120 s before being removed. The behaviours monitored were the swim speed, the per-
148
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
centage of time spent within 10 cm of the wall, the percentage of time spent in each of the four quadrants and the number of annulus crossings (an annulus being the imaginary location of the platform had it been placed at the centre of a quadrant).
2.5.5. Ele6ated plus-maze The plus-maze was contructed of cream coloured perspex and consisted of two opposite open arms (50× 10 cm) and two opposite closed arms (50× 10 cm) enclosed by 40 cm high walls. The open arms were covered with nylon mesh to help prevent the animals from slipping. The arms were connected by a central square, thus forming a ‘plus’ shape. The maze was elevated 50 cm from the floor and lit by dim light. A video camera was mounted vertically over the maze and the behaviour was scored from a monitor in an adjacent room. Each rat was placed in the central square of the plus-maze facing one of the open arms and behaviour was observed for 5 min by an observer blind to the surgical preparation of the animals. The number of the entries into the open and closed arms and the times spent in the open and closed arms were scored. The percentage of entries made onto the open arms and the percent time spent on the open arms and the time spent in the central square were subsequently calculated. Testing was carried out on day 3 following surgery.
3.1.2. Rota-rod The data obtained from testing the animals on the rota-rod are shown in Table 2. Rats were placed on the rota-rod for the first time on the 1st day after the lesion had been carried out and retested each day for 5 days. Two way analyses of variance for all the data points did not reveal any significant differences (group: F(1,19)= 0.4, ns; day: F(4,80)= 1.5, ns; group×day: F(4,72)= 0.7, ns). 3.1.3. Locomotor acti6ity Rats were placed individually into locomotor activity boxes for 30-min test periods. Each animal was tested three times, on the 1st, 7th and 8th day after surgery. Data for the overall activity scores during the 30-min test period are presented in Table 2. There were no significant effects of the trauma on the spontaneous motor activity of the rats at any of the time points following surgery.
Table 2 Mean ( 9SEM) behavioural scores for sham-operated and trauma rats tested for coordination on a rota-rod, locomotor activity and on the elevated plus-maze Time on rota-rod (s, n = 10)
2.6. Statistics Data were analysed using a range of statistical tests chosen independently for each behavioural test according to the type of data obtained. Details are given with the descriptions of the results obtained. Statistical procedures were carried out using SAS version 6.08.
Days following surgery 1 2 3 4 5
3.1.1. Neurological scoring There was no obvious mortality due to trauma (sham-operation, two out of 86 died; trauma, eight out of 89; x 2 =3.6, ns) nor was there a significant weight loss following the lesion (the average loss was approximately 5 g in the sham and 10 g in the lesioned rats). Even on day 1 following surgery the two groups of animals did not differ in their neurological scores. Neither apnea nor convulsions were observed in any animal subjected to trauma. The range of behaviours shown in Table 1 were repeatedly assessed during the first 2 weeks following surgery when both groups of rats appeared completely normal, yielding neurological scores of zero.
Trauma
108.2 9 14.3 169.49 14.3 161.0926.2 130.7 9 32.7 145.1 9 29.7
93.09 19.6 130.6 926.1 116.7 9 23.1 155.3 9 30.1 132.8 932.9
Locomotor activity (beam breaks, n = 10)
3. Results
3.1. Neurological and motor beha6iours
Sham
Days following surgery 1 7 8
307.3 932.9 353.9 934.0 255.9 9 29.8
326.4 927.4 335.3940.1 276.89 51.6
Elevated plus maze (sham: n= 15; trauma: n =18) Number of (%) Time spent (%) Number of Time spent (s)
open arm entries
32.8 91.8
28.5 9 1.9
on the open arm
19.0 92.2
15.2 9 2.0
closed arm entries in central square
12.3 9 1.0 40.9 9 5.0
10.3 9 0.7 34.5 93.6
Time spent on the rota-rod was assessed each day for the first 5 days following surgery. Spontaneous motor activity was assessed on either the 1st, 7th or 8th days following surgery (each time point represents a different group of rats) and exploration on the elevated plus-maze on the 3rd day following surgery.
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
149
into this section of the apparatus (PB 0.05, Mann– Whitney U test) than the trauma rats. Informal observation of the animals which failed to enter the dark compartment on the 2nd day of testing indicated that the lesioned animals continued to explore normally, whilst the sham-operated controls remained completely immobile, i.e. displayed freezing, with a higher rate of defecation.
Fig. 2. Mean ( 9 SEM) number of pellets eaten during the 10-min paw reaching task. Values for the number of pellets removed using the left (L) and right (R) paws for trauma (n= 10) and sham-operated (n = 10) rats are recorded separately. Behaviour was recorded for four training sessions before lesion of the right parietal cortex (or sham operation) and for four sessions after (commencing 6 days after surgery).
3.2.2. Acti6e a6oidance Training in the active avoidance test started on the 5th day following surgery. Data were analysed using two-way, repeated measures analyses of variance and are presented graphically in Fig. 4. Analysis of the data showed overall evidence for avoidance learning, demonstrated by a significant effect of repeated testing on the number of avoidances of the shock (F(5,95)=5.35, PB0.001), failures to escape (F(5,95)= 4.21, PB0.01) and latency to escape (F(5,95)= 7.27, PB0.0001). Visual inspection of the data in Fig. 4 suggests that the trauma rats performed
3.1.4. Paw reaching test Animals were trained to perform the paw reaching test before trauma and testing was continued for 3 weeks from the 6th day after surgery. Comparisons were made for the data obtained before and after surgery between lesioned and control animals for each side and additionally for each group of rats the number of pellets eaten from the sides ipsi- and contra-lateral to the lesion were compared (Fig. 2). Analyses of variance indicated that there were no significant differences between the trauma and sham-operated rats on either the side ipsi- or contra-lateral to the lesion. Performance was comparable for the sides ipsi- and contra-lateral to the site of traumatic lesion for both the lesioned and sham-operated rats. 3.2. Cogniti6e function and anxiety 3.2.1. Passi6e a6oidance Passive avoidance testing was carried out on the 5th and 6th days following surgery. There were no significant baseline differences between the sham-operated and trauma rats, i.e. their latencies to enter the dark compartment on the 1st day of testing were not significantly different (Fig. 3). During the 3-min test period on the 2nd day a number of significant differences between the groups of rats were observed (Fig. 3); 90% of the sham-operated but only 30% of the trauma rats avoided the dark compartment (P =0.02, x 2-test). In addition the mean entry latency for the sham rats was significantly higher than for the lesioned rats (P= 0.001, Mann–Whitney U test). The sham group spent significantly less time in the dark compartment (PB 0.01, Mann–Whitney U test) and made fewer entries
Fig. 3. Data obtained from step-through passive avoidance testing are presented as mean ( 9 SEM for n =10) values for latencies to enter the dark compartment on days 1 (open bars) and 2 (closed bars) of testing, number of entries made into, and time spent in, the dark compartment on day 2 of testing and the percentage of number of rats avoiding the dark compartment on the 2nd day. * PB 0.05, ** PB 0.01 vs. sham-operated rats: Mann – Whitney U test for entry latencies, time spent in dark compartment and number of entries made into the dark compartment; x 2-test for percent of avoidance of dark compartment.
150
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
Fig. 4. Performance of sham-operated (closed circles, n= 10) or trauma lesioned rats (open circles, n = 11) in an active avoidance learning task. The graphs show the mean ( 9 SEM) number of avoidances and number of escapes from shock, number of failures to escape from or avoid the shock, and mean latency to cross, measured from the time of stimulus onset.
better than the sham-operated controls. This is supported by the statistical analysis of both failures to escape (group effect: F(1,19) = 1.19, ns; group× day interaction: F(5,95)= 2.66, P B 0.05) and latency to escape (group effect: F(1,19) =1.45, ns; group× day interaction: F(5,95)= 2.54, P B 0.05). Although the other measures shown in Fig. 4 suggest a trend for better performance in lesioned rats, neither the group, nor the interaction terms were significant for numbers of avoidances (group effect: F(1,19) = 1.66, ns; group× day interaction: F(5,95) =1.72, ns) and no significant effects were obtained for escapes or for intertrial crossings (data not shown).
3.2.3. Holeboard The behaviour of lesioned and control rats on the holeboard are shown in Fig. 5. The data were analysed using separate two-way repeated measures analyses of variance for the first 3 days and for days 3 and 4. This enabled the habituation to the apparatus observed during the first 3 days to be analysed separately from the response to stimulus-change achieved by putting objects in three of the holes. During the frst 3 days there was a significant effect of repeated testing for each of the parameters (locomotor activity: F(2,40)=22.2, P B 0.001; rearing: F(2,40)= 6.9, P B0.01; number of head-dips: F(2,40) = 46.0, PB 0.001; duration of head-dipping: F(2,44) =53.3,
PB 0.001) but no effect of group and no group by day interaction, suggesting that both sham-operated and trauma rats showed the same degree and rate of habituation to the apparatus (Fig. 5). Placing novel objects beneath three of the holes had no overall effect on locomotor activity (F(1,20=0.81, ns) but it did produce a difference between the groups which had not been evident on day 3 (this is shown by a significant group by day interaction (days 3 and 4 only): F(1,20)=5.99, PB 0.05). A similar effect was observed on rearing although in this case there was also an overall difference between the two groups (group: F(1,20)=4.7 PB 0.05; group× day interaction: F(1,20)= 4.9 PB 0.05). Significant group, day and interaction effects were also observed in the number and duration of head-dipping indicating that the sham-operated rats responded much more to the objects (number of head-dips: group F(1,20)= 5.4, PB 0.05; day F(1,20)= 12.8, PB 0.01; group× day interaction: F(1,20)=16.8, PB 0.001. Duration of head-dipping: group F(1,20)= 5.6, PB 0.05; day: F(1,20)= 23.1, PB 0.001; group × day interaction: F(1,20)= 8.0, P=0.01). When the results from day 4 were analysed separately for the holes with and without objects it was clear that trauma rats spent less time exploring the object and non-object holes than the sham rats. However both groups of rats spent more time head-dipping at the holes containing objects than those which were empty. In addition the relative amount of exploration of the object and non-object holes (in terms of both head-dips and time spent head-dipping) was similar for each of the groups (Fig. 5).
3.2.4. Water maze Training in the water maze started on the 5th day following surgery. Two-way repeated measures analyses of variance were applied to data obtained from the 4 days testing in the water maze. For the training latencies there was a highly significant effect of repeated testing indicating that there was an improvement in the animals ability to find the platform (F(11,187)=12.57, PB 0.0001). There was however no difference between the sham-operated and traumatised animals (Fig. 6; effect of trauma: F(1,17)= 0.08, ns; trauma× test interaction: F(11,187)=0.9, ns). Analyses of data for swim speed and time spent within 10 cm of the walls was hampered due to equipment failure and a loss of values for the 2nd day of testing. Statistical tests were however carried out on the data from the 1st, 3rd and 4th days of testing. There was a clear effect of repeated testing on swim speed (F(8,36)=4.72, PB 0.001; Fig. 6) but no difference between the sham-operated and traumatised rats (F(1,17)= 0.01, ns) and no trauma by testing interaction (F(8,136)= 0.79, ns). For the time spent within 10 cm of the wall of the maze there was again a significant
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
151
Fig. 5. Mean values ( 9 SEM, n =11) for horizontal (locomotor activity) and vertical (rearing) motor activity (number of beam breaks), number and duration(s) of head-dipping for sham-operated (closed squares) and trauma-lesioned rats (open circles) tested in a holeboard apparatus for 5 min/day for 4 consecutive days. On the 4th day novel objects were placed under three of the holes. Data for the number of head-dips/hole and time spent head-dipping/hole(s) for holes with objects (shaded bars) or without objects (clear bars) on the 4th day of testing is shown in the two right-hand panels. * P B 0.05, ** P B0.01, Newman–Keuls test comparing sham and trauma rats on day 4; $ PB 0.05, $$ PB 0.01 Newman– Keuls test comparing performance for days 3 and 4; c PB 0.05, c c PB 0.01 for comparisons between holes with and without objects, Student’s t-test for paired values.
effect of repeated testing (F(8,136) = 24.76, P B 0.001) but there was neither a difference between the two groups of animals nor a significant group by testing interaction (group: F(1,17) =0.22, ns; interaction: F(8,136)=1.24, ns; Fig. 6). During the probe trial both groups of rats showed a clear preference for the north-west (training) quadrant than the other quadrants during the 1st and 2nd minutes of the probe trial and the amount of time spent by each group in the different quadrants did not differ (Fig. 6). This was supported by repeated measures ANOVA with quadrant as the repeated measure where only the effect of quadrant was significant (1st minute: treatment F(1,17)= 0.89, ns; quadrant F(3,51) = 20.35, P B 0.001; interaction F(3 51)=0.49, ns; 2nd minute: treatment F(1,17)=0.49, ns; quadrant F(3,51) = 9.5, P B 0.001; interaction F(3,51) =1.88, ns). During the 1st minute of the probe trial both shamoperated and trauma rats made a significantly greater number of annulus crossings in the north-west quad-
rant than in any of the other quadrants (Fig. 6) but this preference was extinguished by the 2nd minute of the trial.
3.2.5. Ele6ated plus-maze Four behavioural parameters were analysed for animals tested on the elevated plus-maze: the percentage of entries made onto the open arms the percentage of time spent on the open arms, the number of closed-arm entries and the time spent in the central square. There were no significant differences between sham-operated and lesioned rats on any of the measures (Table 2).
4. Discussion The majority of studies concerning traumatic brain injury in rodents have attempted to evaluate deficits in learning and memory. However, some of these have involved the induction of non-specific neurological im-
152
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
Fig. 6. Left-hand panels: mean ( 9 SEM) latency to find the platform, swim speeds and percentage of time spent within 10 cm of the wall for sham-operated (closed circles, n=10) and lesioned rats (open circles, n =9) tested in the Morris water maze (three trials per day for 4 days). Day 2 data were lost for the swim speed and percentage of time measures due to equipment failure. Right-hand panels: Data obtained from a 2-min probe trial carried out on the 5th day in the Morris water maze were broken down into two time bins (0 – 60 s and 60 – 120 s). The percentage of time spent in the quadrants and the number of annulus crossings made in each of the four quadrants are presented (mean 9 SEM) for both sham-operated rats (upper panels) and trauma-lesioned rats (lower panels). The platform was placed in the north-west quadrant during training. * P B0.05 for comparisons with north-west quadrant, Dunnetts t-test (percentage of time) or Mann – Whitney U test (annulus crossings).
pairments which might have compromised the interpretation of cognitive deficits. One of the aims of the present series of experiments, therefore, was to determine if it was possible to detect specific and limited behavioural changes following the mild traumatic brain injury used in our laboratory in the absence of general neurological impairment. The fluid percussion-induced lesion of the right parietal cortex used in the present study produced a consistent and reproducible lesion volume and, in contrast with previously published reports on the use of traumatic lesions in rats, did not lead to mortality (e.g. Lyeth et al. [15] who observed 77% mortality in their lesioned group and Sun and Faden [29], who reported 28% mortality). Although clinical data [30] and experiments carried out in rats [11,12] have implicated lesions of the parietal cortex in motor incoordination and hemispheric neglect, a traumatic lesion of the right hemisphere under the conditions reported here did not affect the animals’ behaviour in a range of general
behavioural tests nor in those designed to study either gross (rota-rod, locomotor activity) or fine, lateralised (paw reaching) motor performance. Additionally, the lesion parameters used in these experiments did not affect performance in the Morris water maze or in the habituation phase in the holeboard despite evidence that aspiration and traumatic lesions of this region can perturb spatial memory [4,7,13,25,28]. These apparent differences might be due to both the small lesion size used in the present experiments and the precise location of the fluid-percussion impact. The traumatic lesions carried out here were centred on the lateral aspect of the parietal cortex, whereas Kolb and Walkey [12], Crowne et al. [4] and Smith et al. [27] carried out lesions of the dorsal parietal cortex in order to produce their behavioural deficits. Additionally, the percussion pressures used in other studies were high enough to produce significant damage to the underlying hippocampal structures [9,25,27]. Hicks et al. [9] carried out correlative analyses and
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
demonstrated that the extent of neuronal loss in both the hilar region of the dentate gyrus and the CA3 region of the hippocampus were significantly correlated with the performance deficit in the water maze task. They commented that the reduced ability of their lesioned rats to locate the platform is likely, therefore, to be due to lesions of the hippocampus rather than the cortex. Thus, our lack of effect in the Morris water maze would suggest that the functional consequences of our lesions are not due to hippocampal damage. The holeboard results demonstrate that whilst both the control and lesioned animals similarly responded and habituated to a novel environment, they behaved differently when details within that environment were altered. In the sham-operated controls the presence of novel objects in the experimental context strongly stimulated exploratory behaviours. Lesioned animals continued to habituate on motor parameters and number of head-dips and there was a reduced (cf sham) objectinduced augmentation of the time spent head-dipping. These differences were not due to the inability of the trauma animals to see the objects as they spent significantly longer head-dipping at the holes containing objects than those without. The phenomenon observed on the 4th day is similar to that observed by File and Fluck [6] who showed that a stimulus change (achieved by moving objects from previously baited to unbaited holes) produced a more profound stimulation of exploratory activity in animals which had been handled for the 18 days prior to the test than in those which had only been handled for 4 days, a difference they suggested was due to relative stress levels and habituation to manipulation and novelty. In the present experiments however, both sham and lesioned rats were handled in a similar manner and there was no difference between the groups of rats in other behavioural tests which have been extensively demonstrated to be affected by the way in which animals are treated prior to testing (e.g. the initial exploration of a hole-board or of an elevated plus-maze are changed by the stresses associated with surgery and chronic handling [1]). A more likely explanation is to be found in previous studies linking parietal cortex function and spatial information processing. Several experiments have demonstrated that aspiration of the posterior parietal cortex can reduce the response to spatial novelty in the rat [23,24] and differences in response to spatial novelty between inbred mouse strains also support a role for the parietal cortex in this aspect of cognition [31]. Our results are consistent with these observations. However, it should be noted that we observed no differences in water maze performance, suggesting that not all aspects of spatial information processing are affected by fluid percussion of the lateral parietal cortex. Clear behavioural differences between the control and lesioned rats were observed in the active and
153
passive avoidance procedures. In the active avoidance task the trauma rats showed better and faster learning of the task compared with sham-operated controls. In contrast, trauma rats showed worse performance than sham-operated rats in the passive avoidance task, usually interpreted as an impairment in memory for the shock associated with the dark compartment. This latter finding is consistent with results recently reported by Yamaguchi et al. [35] who showed that traumatic lesion of the lower right parietal cortex resulted in a retention deficit in a step-through passive avoidance test. The interpretation of this impairment in passive avoidance behaviour as a mnemonic deficit is difficult to maintain in view of the improved active avoidance behaviour and lack of impairment in the Morris water maze. Other factors that may help to explain these results include differences between trauma and sham-operated rats in anxiety, general activity or sensitivity to footshock. The first two of these possibilities have already been discussed and the likelihood that changes in motor or emotional behaviour have played a part in any lesioninduced differences in behaviour dismissed. It is also unlikely that the lesioned rats were less sensitive to shock than the sham-operated controls as they responded appropriately to it in the active avoidance test. These results, indicating that traumatic lesions disrupted acquisition of a passive avoidance response but improved learning in the active avoidance task, may seem somewhat contradictory. However, it is well known that a two-way active avoidance response is very difficult for a rat to learn (as shown in the present study by the relatively poor performance of the sham-operated controls) because, to successfully avoid the shock, the animals must overcome their natural tendency to freeze when presented with a conditioned stimulus associated with an aversive shock. Anxiolytic drugs, such as benzodiazepines, have been shown to disrupt passive avoidance learning but improve acquisition of the active avoidance task [5]. The similar behavioural effects of the traumatic lesion in the present study may have resulted from a lesion-induced decrease in fear or anxiety although this hypothesis is not supported by the lack of difference between control and lesioned animals in the elevated plus-maze. Also differences in anxiety levels between the two groups would be expected to provoke differences in behaviour in other tests such as habituation to the hole-board and in general activity, and this was not observed. It remains a possibility, however, that the level of fear specifically invoked in the active avoidance test differs between the groups. Another possibility, suggested by the observation that trauma rats appeared to demonstrate less freezing behaviour in the passive avoidance task than controls, is that a deficit in conditioned responses to shock in trauma-lesioned rats can explain the results in active and passive avoidance tasks. Thus, a reduced freezing
154
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155
response will enhance exploration and thereby impair passive avoidance behaviour whereas acquisition of an active avoidance response requires movement and might therefore be improved by a reduced freezing response to shock. Further experiments designed to study in more depth differences in freezing behaviour between trauma and sham-operated rats are presented in a companion paper to the present manuscript [10]. In conclusion, the lesions carried out in the present studies, which were directed at a discrete section of the lateral parietal cortex, did not result in gross behavioural changes or neurological impairments. However, our results demonstrate that this fluid percussion injury induced behavioural changes in active and passive avoidance and in response to novelty in the holeboard. These effects do not appear to be attributable to changes in motor behaviour, level of anxiety or sensitivity to shock, and they may be useful as sensitive markers of a mild fluid percussion injury to study functional recovery after trauma. However further work is required to firmly establish the underlying deficit(s) that has led to the behavioural effects described here.
Acknowledgements The authors are grateful to Corinne Perron for her expert technical assistance.
References [1] Adamec RE, Sayin U, Brown A. The effects of corticotrophin releasing factor (CRF) and handling stress on behaviour in the elevated plus-maze test of anxiety. J Psychopharmacol 1991;5:175 – 86. [2] Anderson RA. Inferior parietal lobe function in spatial perception and visuomotor integration. In: Mountcastle VB, Plum F, Geiger SR, editors. Handbook of Physiology. Bethesda: American Physiological Society, 1987:483–518. [3] Cahn J, Borzeix MG, Cahn R. Effect of nimodipine on cerebral edema and cation shift, cognitive function, neurological recovery, CBF and EEG after transient incomplete forebrain ischemia in the rat. In: Bartko D, editor. New Trends in Clinical Neuropharmacology. John Libbey, 1988:27–38. [4] Crowne DP, Novotny MF, Maier SE, Vitols R. Effects of unilateral parietal lesions on spatial localization in the rat. Behav Neurosci 1992;106:808–19. [5] Fernandez-Teruel A, Escorihuela RM, Nunez JF, Zapata A, Boix F, Salazar W, Tobena A. The early acquisition of the two-way (shuttle box) avoidance as an anxiety-mediated behavior: Psychopharmacological validation. Brain Res Bull 1991;26:173 – 6. [6] File SE, Fluck E. Handling alters habituation and response to stimulus change in the holeboard. Pharmacol Biochem Behav 1994;49:449 – 53. [7] Foreman N, Save E, Thinus-Blanc C, Buhot M-C. Visually guided locomotion, distractibility, and the missing-stimulus effect in hooded rats with unilateral or bilateral lesions of parietal cortex. Behav Neurosci 1992;106:529–38.
[8] Hamm RJ, Lyeth BG, Jenkins LW, O’Dell DM, Pike BR. Selective cognitive impairement following traumatic brain injury in rats. Behav Brain Res 1993;59:169 – 73. [9] Hicks RR, Smith DH, Lowenstein DH, Saint Marie R, McIntosh TK. Mild experimental brain injury in the rat induces cognitive deficits associated with regional neuronal loss in the hippocampus. J Neurotrauma 1993;10:405 – 14. [10] Hogg S, Sanger DJ, Moser PC. Mild traumatic lesion of the right parietal cortex in the rat: (2) Characterisation of a conditioned freezing deficit and its reversal by dizocilpine Behav Brain Res 1997; (In press). [11] Holm S, Mogensen J. Contralateral somatosensory neglect in unrestrained rats after lesion of the parietal cortex of the left hemisphere. Acta Neurobiol Exp 1993;53:569 – 76. [12] Kolb B, Walkey J. Behavioural and anatomical studies of the posterior parietal cortex in the rat. Behav Brain Res 1987;23:127 – 45. [13] Kraemer PJ, Brown RW, Baldwin SA, Scheff SW. Validation of a single-day Morris water maze procedure used to assess cognitive deficits associated with brain damage. Brain Res Bull 1996;39:17 – 22. [14] Levi L, Guilburd JN, Lemberger A, Soustiel JF, Feinsod M. Diffuse axonal injury: analysis of 100 patients with radiological signs. Neurosurgery 1990;27:429 – 32. [15] Lyeth BG, Dixon CE, Jenkins LW, Hamm RJ, Alberico A, Young HF, Stonnington HH, Hayes RL. Effects of scopolamine treatment on long-term behavioural deficits following concussive brain injury to the rat. Brain Res 1988;452:39 – 48. [16] Lynch JC. The functional organization of posterior parietal association cortex. Behav Brain Sci 1980;3:485 – 534. [17] McIntosh TK. Novel pharmacologic therapies in the treatment of experimental traumatic brain injury: a review. J Neurotrauma 1993;10:215 – 61. [18] McIntosh TK. Neurochemical sequalae of traumatic brain injury: therapeutic implications. Cerebrovasc Brain Metab Rev 1994;6:109 – 62. [19] Montoya CP, Campbell-Hope LJ, Dunnett SB. The ‘staircase test’: a measure of independent forelimb reaching and grasping abilities in rats. J Neurosci Methods 1991;36:219 – 28. [20] Morris R. Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Methods 1984;11:47–60. [21] Muir JK, Lyeth BG, Hamm RJ, Ellis EF. The effect of acute cocaine or lidocaine on behavioural function following fluid percussion brain injury in rats. J Neurotrauma 1995;12:87–97. [22] Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. Sydney: Academic Press, 1982. [23] Save E, Buhot M-C, Foreman N, Thinus-Blanc C. Exploratory activity and response to spatial change in rats with hippocampal or posterior parietal cortical lesions. Behav Brain Res 1992;47:113 – 27. [24] Save E, Poucet B, Foreman N, Buhot M-C. Object exploration and reactions to spatial and nonspatial changes in hooded rats following damage to parietal cortex or hippocampal formation. Behav Neurosci 1992;106:447 – 56. [25] Sinson G, Voddi M, McIntosh TK. Nerve growth factor administration attenuates cognitive but not neurobehavioural motor dysfunction or hippocampal cell loss following fluid-percussion brain injury in rats. J Neurochem 1995;65:2209 – 16. [26] Smith DH, Lowenstein DH, Gennarelli TA, McIntosh TK. Persistent memory dysfunction is associated with bilateral hippocampal damage following experimental brain injury. Neurosci Lett 1994;168:151 – 4. [27] Smith DH, Okiyama K, Thomas MJ, Claussen B, McIntosh TK. Evaluation of memory dysfunction following experimental brain injury using the morris water maze. J Neurotrauma 1991;8:259– 69.
S. Hogg et al. / Beha6ioural Brain Research 93 (1998) 143–155 [28] Smith DH, Okiyama K, Thomas MJ, McIntosh TK. Effects of excitatory amino acid receptor antagonists kynurenate and indole-2-carboxylic acid on behavioural and neurochemical outcome following experimental brain injury. J Neuroscience 1993;13:5383 – 92. [29] Sun FY, Faden AI. Neuroprotective effects of 619C89, a use-dependent sodium channel blocker, in rat traumatic brain injury. Brain Res 1995;673:133–40. [30] Temkin NR, Holubkov R, Machamer JE, Winn HR, Dikmen SS. Classification and regression trees (CART) for prediction of function at 1 year following head trauma. J Neurosurg 1995;82:764 – 71. [31] Thinus-Blanc C, Save E, Rossi Arnaud C, Tozzi A, Ammassari Teule M. The differences shown by C57BL/6 and DBA/2 inbred mice in detecting spatial novelty are subserved by a different
.
[32]
[33]
[34]
[35]
155
hippocampal and parietal cortex interplay. Behav Brain Res 1996;80:33 – 40. Toulmond S, Duval D, Serrano A, Scatton B, Benavides J. Biochemical and histological alterations induced by fluid percussion brain injury in the rat. Brain Res 1993;620:24 – 31. Toulmond S, Rothwell NJ. Interleukin-1 receptor antagonist inhibits neuronal damage caused by fluid percussion injury in the rat. Brain Res 1995;671:261 – 6. Toulmond S, Serrano A, Benavides J, Scatton B. Prevention by eliprodil (SL 82.0715) of traumatic brain damage in the rat. Existence of a large (18 h) therapeutic window. Brain Res 1993;620:32 – 41. Yamaguchi T, Suzuki Y, Yamamoto M, Nakamura T, Yamaura A. Indeloxazine hydrochloride improves impairment of passive avoidance performance after fluid percussion brain injury in rats. Neuropharmacology 1996;35:329 – 36.
.