- Email: [email protected]
Repetitive acute pain in infancy increases anxiety but does not alter spatial learning ability in juvenile mice
Behavioural Brain Research 142 (2003) 157–165
Repetitive acute pain in infancy increases anxiety but does not alter spatial learning ability in juvenile mice Heather M. Schellinck∗ , Lianne Stanford, Matthew Darrah Department of Psychology, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4J1 Received 8 July 2002; received in revised form 2 December 2002; accepted 2 December 2002
Abstract We assessed the long-term behavioural effects of a single acute or repetitive inflammatory pain experienced during infancy. Groups of male and female CD1 mice were subjected to either an acute single pain, i.e. a tail clip or sham pain at P8, or acute repetitive pain in the form of needle pricks or sham pain from P8 to P14. All of the subjects were tested in the elevated plus maze at P30 and in the Morris Water Maze from P31 to P38. Mice in the acute single pain and sham groups did not differ on measures of anxiety in the plus maze. Mice in the repetitive pain group demonstrated significantly more anxious behaviours than controls in the elevated plus maze as they spent less time in the open arms, made fewer open arm entries, displayed fewer head dips and showed more stretch attend postures. There were no effects of single or repetitive pain treatments in the latency to find the hidden platform in the Morris Water Maze. Overall, these data suggest that acute repetitive pain experienced during infancy may increase anxiety later in life but it does not influence spatial learning as measured in the Morris Water Maze. The origin of the anxiogenic profile shown in the acute repetitive pain mice may be a result of changes in neural circuitry or context dependent learning and is currently under further investigation with this paradigm. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Repetitive pain; Anxiety; CD1 mice; Development; Learning and memory; Elevated plus maze; Morris Water Maze
1. Introduction The technical advances that have lead to enhanced survival for acutely ill newborns have been accompanied by an increase in potentially painful diagnostic and treatment procedures. Repeated heel pricks, intravenous drug delivery, endotracheal intubation and neonatal surgeries have become relatively common and often result in acute inflammatory pain in the neonate [5,18]. The mechanisms which contribute to inflammatory pain have been thoroughly investigated, primarily in preparations from adult animals, and involve alterations in intracellular signalling cascades as well as transcription-related increases in both receptors and neurotransmitters [13,17,19,29]. These responses may lead to short-term functional and structural changes in adult animals [1,2,15,22]. In newborns, these processes are experienced at a time when the brain is extraordinarily plastic with the nervous, endocrine and immune systems all undergoing functional and structural development [9]. Consequently, it has been hypothesised that exposure to painful stimuli during such a critical period could disrupt ∗
Corresponding author. Tel.: +1-902-494-6809; fax: +1-902-494-6585. E-mail address: [email protected] (H.M. Schellinck).
these neurodevelopmental processes and lead to permanent changes in the brain as well as in behaviour [4,25]. Several investigators have found long-term changes in response to inflammatory pain in neonatal rats in peripheral nerve terminals and within the spinal cord itself. Reynolds and Fitzgerald [22] assessed peripheral changes following skin wounds in P0, P7, P14 and adult rats. An examination of the axonal sprouting response revealed a long lasting, i.e. 21–42 days, cutaneous hyperinnervation. The innervation density was of a greater magnitude and duration when the wounds were performed in P0 and P7 rats compared with P14 animals. In the latter animals, changes were of a more transient nature similar to the effect found in adults. These results suggest the existence of a critical window for changes to create long-term effects. In the latter study, sensory thresholds were also measured in rats that received skin wounds on P0. They were found to be significantly lower than thresholds in same-age controls for at least 24 days. The authors suggested this provides evidence of pain or hypersensitivity long after the wound has healed. Recently, a persistent inflammatory pain in neonatal rats resulting from a single injection of Freund’s adjuvant (FA) in P0 and P3 neonatal rats has been shown to be associated with greater density and distribution in spinal cord laminae
0166-4328/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0166-4328(02)00406-0
158
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
I and II dorsal afferent axons in adult animals [24]. In contrast, such permanent changes were not demonstrated in adult animals that had previously been injected at P14. Behavioural responses to the initial injection of CFA were not assessed in adult animals. Nonetheless, when injected with formalin, adults that had been treated as neonates at an unspecified age exhibited pain responses earlier than controls during the late phase of the test [24]. To test the hypothesis that such structural changes could be paralleled by long-term behavioural effects, Anand et al. [3] assessed a number of behavioural responses of rat pups exposed to needle pricks daily from P0 to P7. Compared with controls, these rats showed decreased latency to respond to a thermal stimulus prior to weaning. There were no differences in hotplate latencies between adult experimental and control rats. As adults, rats that had received four paw pricks daily had an increased latency and decreased frequency to enter an open field and an increase in alcohol preference. The authors suggested that these behavioural effects in the adult rats subjected to repetitive neonatal pain were a result of increased anxiety. In a follow up study, Bhutta et al. [7] assessed the effects of formalin-induced inflammatory pain on P0 to P7 rats. When tested as adults, both male and female experimental animals had increased hotplate latencies compared with controls and males showed a decreased preference for alcohol. Males but not females showed a decrease in locomotor activity in a behavioural chamber. The different methods used to induce the early inflammatory pain in these experiments may have resulted in the mixed support for the hypothesis that early repetitive pain leads to increased anxiety in adulthood. Anand et al. [3] also determined that following repeated paw pricks as neonates adult rats recognised familiar conspecifics in a social discrimination task for a longer period of time than controls. This latter result is in agreement with other work that suggests that memory and/or performance is enhanced by a previous painful experience. Neonatal rats exposed to thermal pain had better performance on a two-way active avoidance task than control animals [6]. Similarly, young rats that had experienced foot shocks mastered a lever pressing task faster than controls [16]. The experiments thus far have reported physical and behavioural effects in rats (P0–P7) that are developmentally equivalent to preterm infants [10]. In this study, we have used a different procedure to determine if an early pain experience in older neonatal mice also results in long-term behavioural changes. We exposed P8 mice to a single event of acute pain in the form of a tail clip. A second group of mice were given daily paw pricks (P8–P14) on the dorsal surface of a front and contralateral rear paw. We choose to use animals of these ages as the maturity of an 8-day-old rodent has been reported to be comparable to that of a full-term infant [10]. We examined the effects of these procedures in P30 to P38 animals to determine if the hypothesised effects would manifest themselves in prepubertal animals. These procedures enabled us to investigate the long-term effects of a fairly typical experience in a full-term human infant.
Given the somewhat conflicting findings as to whether anxiety is affected by early pain [3,7], we used the elevated plus maze to assess anxiety. This test is known for its high ecological validity and reliable measurement of behaviours associated with anxiety and risk assessment in rodents [23,27]. To further investigate reports that pain in neonates enhances learning [6,16], we also assessed spatial learning in the Morris Water Maze.
2. Materials and methods 2.1. Animals CD1 mice (N = 56), offspring of adult males and females purchased from Charles River Canada (St. Constant, Quebec) and bred at Dalhousie University, were used in this experiment. The day of birth was designated as Postnatal Day 0 (P0). Litters were culled to a group of four males and four females on P3 and weaned on P21. Prior to weaning, the mice were housed with their dams in polycarbonate cages (30 cm × 15 cm × 10 cm) with wood-chip bedding. Following weaning, the mice were housed in same sex groups of four in polycarbonate cages (30 cm × 15 cm × 10 cm). They were maintained on a 12 h/12 h light/dark cycle with lights off at 13:30 h and provided with Purina Agribrand diet 5001 and water ad libitum. On P8, males and females from each litter were quasi-randomly assigned to each of the following four treatment groups: acute single pain, sham single pain, acute repetitive pain and sham repetitive pain. All animals were treated in accordance with the Canadian Council on Animal Care guidelines and the experimental protocol was approved by the Dalhousie University Committee on Laboratory Animals. 2.2. Pain treatments 2.2.1. Single acute and sham pain On Day 8, during the dark phase of the L/D cycle, two male and two female mice from each litter were temporarily placed in a holding cage. One male and one female mouse received a 2.5 mm tail clip with surgical scissors. The remaining male and female in the sham single pain group were tail marked with a permanent marker at the 2.5 mm mark but no cut was made. For identification purposes, mice were marked with nontoxic acrylic artists ink (Daler-Rowney, Jamesburg, NJ) following the procedure. The four mice were then returned to their home cage. On Days 9–14, these mice were transferred daily to a clean holding cage to minimise the handling differences between these groups and the repetitive pain groups. 2.2.2. Repetitive acute and sham pain On Day 8, during the dark phase of the L/D cycle, the remaining four mice were removed from the litter. One male and one female were given needle pricks on the dorsal
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
surface of a front and contralateral rear paw with a 26-gauge needle piercing the paw. The remaining two animals received sham pricks by placing the flat edge of the needle on the dorsal surface of both paws. Both treatments were continued daily up to and including Day 14. The pain treatment was counterbalanced such that two litters received the pain in the right front and rear left paw and two litters received the pain in the left front and right rear paw. 2.3. Elevated plus maze The elevated plus maze consisted of four arms (30 cm × 5 cm) in the shape of a cross, extending from a central square (5 cm × 5 cm) and elevated 45 cm above the ground. Two of the opposing arms had a 4 mm lip around the edges (open arms) and the remaining two arms were enclosed with 15 cm transparent Plexiglas walls (closed arms). The floor of the maze was composed of black Plexiglas. The test room was illuminated by a 60 W red light bulb. A video camera linked to a TV and a VCR were used to record each trial. On Day 30, during the dark phase of the L/D cycle, the mice were tested individually in the maze. Each mouse was lowered onto the central square and their behaviour in the maze videotaped for 5 min for subsequent analysis with Hindsight Ver 1.5 [28]. Video analysis was completed with the observer blind to the experimental conditions for the following behaviours: (1) amount of time spent in the open arms (four paws in arm); (2) number of entries into the open arms; (3) locomotor activity; (4) frequency of head dipping, i.e. looking over the edge of an open arm; (5) number of stretch attend postures, i.e. pausing while stretched out with head lowered and rear tilted upward; (6) time spent rearing, i.e. standing up on hind legs with or without pressing the front paws on the Plexiglas wall; (7) time spent grooming. After each mouse completed the maze, the presence or absence of fecal boli was recorded and the entire maze was cleaned with 70% ethanol. Behaviour in the elevated plus maze was assessed prior to Morris Water Maze testing so that the handling associated with the latter paradigm would not confound the measurement of anxiety. 2.4. Morris Water Maze A Morris Water Maze (MWM) was created from a plastic swimming pool (diameter: 110 cm; height: 20 cm) filled with 20–22 ◦ C water to a depth of 14 cm. The water was made opaque by the addition of 500 ml of nontoxic white liquid Tempura Paint (Schola Brand, Marieville, Quebec, Canada). The circular concealed platform was constructed from white plastic (diameter: 10 cm; height: ∼14 cm) such that it was just below the surface of the water. During Sessions 1–3 (P31–P33), the concealed platform was placed in the south-east quadrant; during Sessions 4–6 (P34–P36), the platform was moved to the north-west
159
quadrant. During Session 7, the platform was removed and a probe test completed. During Session 8 (P38), the platform was placed in the NE quadrant and made visible with the addition of a 3 cm high red platform so that it was 2.5 cm above the water. Fixed spatial cues including several posters with geometric designs and office furniture were located around the maze in the test room. A video camera linked to a TV and a VCR were used to record each session. On P31 to P36 and P38, the mice were tested in this maze for four trials a day for 6 days during the dark phase of the L/D cycle. The mouse was lowered from a plastic cup into the maze at one of the four marked principal points of a compass such that for each four trial sessions, it entered the water once from each north, south, east and west point. The order of entry was quasi-randomised over sessions. The mice were given 60 s to find the platform after which time they were directed to the platform with a plastic container and forced to remain on the platform for 20 s. Latency to find the platform was recorded with a stopwatch. On P37, the mice were tested for a single 60 s probe trial during which the platform was absent. For the probe trial, the amount of time spent in each quadrant and in the outer sixth annulus of the pool (thigmotaxis) was recorded from video following completion of the session. 2.5. Statistical analysis Two of the fifty-six mice were incorrectly sexed prior to the beginning of the experiment such that there were only five mice in the male paw prick control group. Thus, sex differences were not assessed. 2.5.1. Elevated plus maze The effects of the single acute pain on time spent in the open arms, number of open arm entries, frequency of line crosses, stretch attend postures, head dips, duration of grooming and rearing and were assessed with two-tailed t-tests. The amount of time spent in the open arms was calculated as the percentage of time spent in the open arms divided by the amount of time that the mice spent in the open and closed arms combined (excluding any centre square time). Similarly, the effect of repetitive pain on the same measures were also were examined by t-tests. Differences were considered to be significant if P < 0.05. 2.5.2. Morris Water Maze The effects of single acute pain on latency to find the hidden platform during 3 days of acquisition training was assessed in a two-way mixed ANOVA. To complete this analysis, the individual scores of each animal on the four trials each day were averaged to give a mean latency per day. Similarly, the latency to find the hidden platform was also assessed during 3 subsequent days of reversal training. Latency to find the hidden platform during the last acquisition and first reversal days was assessed in a two-tailed
160
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
t-test. The total amount of time spent in each quadrant during the probe trial in which the platform was removed was examined in a 4 × 2 ANOVA. The degree of thigmotaxia expressed during the probe trial was assessed in a two-tailed t-test as was the mean time to find the visible platform on the day following the probe trial. Newman–Keuls post hoc tests were used to find differences across groups where required.
3. Results 3.1. Elevated plus maze There were no significant differences between mice that received the single tail cut and sham cut controls in any of the eight behaviours scored in the elevated plus maze (all ts(26) < 1.31), Fig. 1 illustrates these results.
Fig. 1. Mean + S.E.M. for the (A) number of line crosses, (B) percent of time spent in the open arms, (C) number of head dips, (D) number of open arm entries, (E) number of stretch attend postures, (F) number of rearing bouts, (G) probability of the presence of fecal boli and (H) duration of grooming for the sham tail cut mice (white bars) and the tail cut mice (black bars).
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
Mice that received repetitive paw pricks made fewer open arm entries (t (26) = −3.80, P < 0.0008), spent less time in the open arms of the elevated plus maze (t (26) = −3.32, P < 0.003), displayed fewer head dips (t (26) = −2.55, P < 0.02) and showed more stretch attend postures than the sham pricked mice (t (26) = 2.47, P < 0.02). There were
161
no differences between the paw pricked and control mice for the number of lines the mice crossed (t (26) = −1.26, P > 0.22), time spent rearing (t (26) = 0.75, P > 0.46) or grooming (t (26) = −0.24, P > 0.81) or probability of defecation (t (26) = 1.59, P > 0.12). Fig. 2 illustrates these results.
Fig. 2. Mean (+S.E.M.) for the (A) number of line crosses, (B) percent of time spent in the open arms, (C) number of head dips, (D) number of open arm entries, (E) number of stretch attend postures, (F) number of rearing bouts, (G) probability of the presence of fecal boli and (H) duration of grooming, for the sham paw pricked mice (white bars) and the paw pricked mice (black bars). The asterisk (∗) illustrates significant differences between sham and paw pricked mice (P < 0.05).
162
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
3.2. Morris Water Maze For the acute pain mice and their controls, there was no main effect of pain (single pain versus sham) in the latency to find the hidden platform during the 3 days of acquisition training (F (1, 26) = 1.12, P > 0.30) but there was a decrease in latency over the 3 test days for both groups (F (2, 52) = 24.13, P < 0.0001) with no interaction between test day and pain (F (2, 52) < 1). There was no disruption in the latency to find the platform when it was moved from the acquisition quadrant to the reversal quadrant between Days 3 and 4 (F (1, 26) = 1.20, P > 0.28) and there was no effect of pain and no interaction between pain group and day (F s < 1). During the reversal days, the mice showed a significant decrease in escape latency over the 3 days (F (2, 52) = 7.73, P < 0.002) with no main effect of pain and no interaction between test day and pain (F s < 1; Fig. 3A). During the 60 s probe trial, the total amount of time spent in each quadrant was significantly different (F (3, 78) = 16.56, P < 0.0001). Newman–Keuls post hoc tests revealed that more time was spent in the quadrant which most recently had the platform (reversal) compared to the other three quadrants. There were no differences found between
the pain and control groups (F (1, 26) = 1.08, P > 0.31) and no significant interaction between quadrant and group (F (3, 78) < 1) as shown in Fig. 3B. An analysis of thigmotaxic behaviour showed no experimental group differences in duration (t (26) = −0.174, P > 0.86; means ± S.E.M.: 11.01 ± 1.69 (pain) and 11.43 ± 1.77 (sham)) or frequency (t (26) = −0.76, P > 0.45; 9.71 ± 1.04 (pain) and 10.93 ± 1.21 (sham)). Finally, the latency required for the mice to reach the visible platform showed no differences between groups (t (26) = 1.38, P > 0.18; Fig. 3B). Paw pricked mice and their controls were compared over the first 3 days of acquisition in the Morris Water Maze and there was an reduction in the latency to find the platform across days (F (2, 52) = 25.84, P < 0.0001). Overall, there was no difference between the paw pricked mice and controls (F (1, 26) = 2.51, P > 0.13) and no interaction between pain group and day (F (2, 52) < 1). There was no increase in latency to find the platform when comparing the third day of acquisition to the first day of reversal (F (1, 26) < 1), no difference between the two groups (F (1, 26) < 1) and no significant interaction between day and experimental manipulation (F (1, 26) = 3.18, P > 0.08). During the reversal days, the mice showed a decrease in latency to find the hidden platform over the 3 days (F (2, 52) =
Fig. 3. (A) Mean latency ± S.E.M. for the tail cut (black circles) and sham tail cut mice (white circles) to find the hidden platform in the original location (acquisition), the opposite quadrant (reversal) and when it was made visible and moved to a third location in the Morris Water Maze (visible). (B) Mean duration + S.E.M. spent by the sham mice (white bars) and the tail cut mice (black bars) in each quadrant of the pool during the probe trial.
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
163
Fig. 4. (A) Mean latency ± S.E.M. for the paw pricked (black squares) and sham paw pricked mice (white squares) to find the hidden platform in the original location (acquisition), the opposite quadrant (reversal) and when it was made visible and moved to a third location (visible). (B) Mean duration + S.E.M. spent by the sham mice (white bars) and the tail cut mice (black bars) in each quadrant of the pool during the probe trial.
17.18, P < 0.0001, with means ± S.E.M. of 28.71 ± 2.54 (Day 4), 17.71 ± 2.11 (Day 5) and 15.72 ± 1.51 (Day 6)). The main effect of group (F (1, 26) < 1) and the interaction of group and day (F (2, 52) = 1.05, P > 0.36) were not significant (Fig. 4A). During the probe trial, the total amount of time spent in each quadrant was significantly different (F (3, 78) = 24.55, P < 0.0001). Newman–Keuls revealed that the mice spent more time in the quadrant from which the platform was removed (reversal) compared to the remaining three quadrants, and that the amount of time spent in the left quadrant was greater than that spent in either the opposite quadrant or the right quadrant. There were no differences found between the experimental groups (F (1, 26) < 1), and no significant interaction of the experimental group and quadrant (F (3, 78) = 1.97, P > 0.12). An analysis of thigmotaxic behaviour during the probe trial showed no differences between the experimental and control groups for duration, (t (26) = −1.31, P < 0.20; 11.04 ± 1.35 (pain), 14.00 ±1.81 (sham)) or for frequency of thigmotaxic bouts (t (26) = −1.58, P < 0.12; 9.93 ± 0.8 (pain) and 12.29 ± 1.26 (sham)). Finally, there were no differences observed between the experimental groups in the latency of the mice to find the visible platform (t (26) = −1.86, P < 0.07; Fig. 4B).
4. Discussion In these experiments, we have demonstrated that infant mice that experience acute repetitive pain from P8 to P14 show a pattern of behaviour in the elevated plus maze as juveniles that is characteristic of anxiety, i.e. a significant increase in stretch attends and a decrease in head dips and time spent in the open arms and frequency of open arm entries. In contrast, mice that experienced a single acute pain on P8 did not show anxious behaviours as juveniles. Mice exposed to either a single or repetitive pain experience performed as well as littermate controls on the Morris Water Maze. Overall, these results indicate that there are long-term behavioural effects of early repetitive pain on anxiety in mice that can be demonstrated in prepubertal animals. These effects manifest themselves as an increase in anxiety in the elevated plus maze but not as an alteration in performance in a spatial learning task. Two previous studies [3,7] have also found that early repetitive pain in either the form of a needle prick or a formalin injection produced long-term behavioural effects in adult rats. We have extended these findings by demonstrating that mice as well as rats are vulnerable to the effects of repetitive pain. Nonetheless, there are some important differences in these studies. The work of Anand and coworkers
164
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
[3,7] was designed to model the effects of pain on preterm infants and as such exposed rats to pain from P1 to P7. The mice in our study were older (P8–P14) when exposed to pain and more analogous to full-term infants. Thus, we have demonstrated that animals with more developed peripheral and central nervous systems are also vulnerable to the effects of repetitive pain. This suggests that the concept of a critical period in which pain must be experienced to have long-term effects should be expanded to include animals within the age range of P8 to P14. The mice in our study were also younger (P30–P38) when tested for long-term effects. Thus, it is clear that anxiety associated with early pain is manifest in mice in a prepubertal state. These findings point to the need for early assessment and treatment of anxiety-like symptoms in children exposed to pain, regardless of their status as term or preterm infants. Mice exposed to pain and mice in the sham conditions performed equally well in the Morris Water Maze with both groups showing a decrease in latency to find the hidden platform, no reduction in performance after the location of the hidden platform was reversed and no differences in search strategy during the probe trial. Our failure to find evidence of enhanced performance in the Morris Water Maze as a result of early acute or repetitive pain is not in agreement with the results of previous reports on the influence of pain on learning [6,16]. Anand et al. [3] also found that adult rats that had specifically experienced early repetitive pain recognised conspecifics for a longer time period than controls. The authors suggested that anxiety-related hypervigilance may have led to the latter result. This state of anxiety could have also accounted for the differences in learning noted in the other reports in that more vigilant animals may become aware of task demands more quickly, thus leading to enhanced performance. Anxiety has been suggested to lead to inefficient search strategies and poor performance on the Morris Water Maze [8,11]. In our experiment, as measured by frequency and duration of thigmotaxic behaviour in the probe trial test, anxiety in the Morris Water Maze did not differ between animals that experienced pain and those that had not. The need for further investigation of the effect of pain on learning in a number of paradigms is required before any substantive conclusions can be drawn. Our results did not support the hypothesis that a single acute pain would cause long-term behavioural changes in mice as had previously been found in humans [26]. It is possible that the type of pain administered in this experiment, i.e. tail cuts in mice, is not comparable to the single pain experience, i.e. circumcision, assessed in humans. Nonetheless, other factors that might have influenced the response to pain in humans should also be taken into consideration. In particular, there may be context dependent memory effects in humans such that the experience of pain in similar environments, i.e. a clinic and a hospital, is sufficient to evoke the emotional aspects of pain regardless of the lack of long-term nociceptive changes. In our experiment, the rooms in which the tail cuts were performed, i.e. the odour laden animal
room versus the sterile testing room, were very different. Thus, it seems much less likely that the context would evoke any psychological aspect of the previous painful event that could manifest itself as increased anxiety. There have been an increasing number of reports that human infants exposed to pain and other stressful experiences at birth show altered responses to pain during infancy and childhood. Infants circumsized at birth show increased behavioural response to vaccination at 4 months of age [26]. Stressful birth conditions are also associated with high levels of cortisol following vaccinations at 6 months of age [21]. Extremely low birth weight (ELBW) infants interviewed at 8–10 years rated the intensity of painful medical events as higher than their full-term peers; moreover, those who spent more time in intensive care gave higher affective ratings to common painful events [12]. ELBW infants also show a different pattern of cardiac response to finger lance at 4 months of age [20]. Hack et al. [14] have also argued that the attentional deficits, poorer social skills and adaptive behaviour problems in LBW infants may well be associated with the extensive painful procedures experienced by these children at birth. Given the affective nature of the responses in many of these studies, it seems possible that the effects may be related to structural and functional alterations in brain regions associated with processing of emotion rather than perturbations in noiciceptive circuitry. Because of the obvious ethical considerations, it is almost impossible to examine the causal relationship between early pain and long-term behavioural effects in humans. The animal paradigm described in this experiment has provided us with a means to investigate both the nocieptive and emotional mechanisms of pain that underlie these long-term changes.
Acknowledgements This research was supported by a NSERC operating grant to H.M. Schellinck and a K.M. Hunter Charitable Foundation/CIHR Doctoral research award and the Spatz Doctoral Training Award 2002 (Alzheimer Society of Canada and IA-CIHR) to L.E. Stanford.
References [1] Aldskogius H, Hermanson A, Jonsson CE. Reinnervation of experimental superficial wounds in rats. Plast Reconstr Surg 1987;79: 595–9. [2] Alvares D, Torsney C, Beland B, Reynolds M, Fitzgerald M. Modelling the prolonged effects of neonatal pain. In: Sandkuhler J, Bromm B, Gebhart GF, editors. Nervous system plasticity and chronic pain. Progress in brain research, vol. 129. Amsterdam: Elsevier; 2000. p. 365–73. [3] Anand KJ, Coskun V, Thrivikraman KV, Nemeroff CB, Plotsky PM. Long-term behavioral effects of repetitive pain in neonatal rat pups. Physiol Behav 1999;66:627–37.
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165 [4] Anand KJ, Scalzo FM. Can adverse neonatal experiences alter brain development and subsequent behavior? Biol Neonate 2000;77: 69–82. [5] Barker DP, Simpson J, Pawula M, Barrett DA, Shaw PN, Rutter N. Randomised, double blind trial of two loading dose regimens of diamorphine in ventilated newborn infants. Arch Dis Child Fetal Neonatal Ed 1995;73:F22–6. [6] Bernardi M, Genedani S, Bertolini A. Behavioral activity and active avoidance learning and retention in rats neonatally exposed to painful stimuli. Physiol Behav 1986;36:553–5. [7] Bhutta AT, Rovnaghi C, Simpson PM, Gossett JM, Scalzo FM, Anand KJ. Interactions of inflammatory pain and morphine in infant rats: long-term behavioral effects. Physiol Behav 2001;73:51–8. [8] Brooks AI, Cory-Slechta DA, Murg SL, Federoff HJ. Effects of behaviourally induced stress on spatial memory acquisition and retention in mice: a biochemical evaluation. Abstr Soc Neurosci 2000;26:2239. [9] Constantine-Paton M. The plastic brain. Neurobiol Dis 2000;7: 515–9. [10] Fitzgerald M, Anand KJ. The developmental neuroanatomy and neurophysiology of pain. In: Schechter NL, Berde CB, Yaster M, editors. Pain in infants, children, and adolescents. Baltimore, MD: Williams & Wilkins; 1993. p. 11–32. [11] Grootendorst J, de Kloet ER, Dalm S, Oitzl MS. Reversal of cognitive deficit of apolipoprotein E knockout mice after repeated exposure to a common environmental experience. Neuroscience 2001;108: 237–47. [12] Grunau RE, Whitfield MF, Petrie J. Children’s judgements about pain at age 8–10 years: do extremely low birthweight (≤1000 g) children differ from full birthweight peers? J Child Psychol Psychiatry 1998;39:587–94. [13] Guenther S, Reeh PW, Kress M. Rises in [Ca2+ ]i mediate capsaicinand proton-induced heat sensitization of rat primary nociceptive neurons. Eur J Neurosci 1999;11:3143–50. [14] Hack M, Taylor HG, Klein N, Eiben R, Schatschneider C, Mercuri-Minich N. School-age outcomes in children with birth weights under 750 g. N Engl J Med 1994;331:753–9. [15] Hermanson A, Dalsgaard CJ, Bjorklund H, Lindblom U. Sensory reinnervation and sensibility after superficial skin wounds in human patients. Neurosci Lett 1987;74:377–82.
165
[16] Kim DG, Lee S, Lim JS. Neonatal footshock stress alters adult behavior and hippocampal corticosteroid receptors. Neuroreport 1999;10:2551–6. [17] Li P, Wilding TJ, Kim SJ, Calejesan AA, Huettner JE, Zhuo M. Kainate-receptor-mediated sensory synaptic transmission in mammalian spinal cord. Nature 1999;397:161–4. [18] McIntosh N. Pain in the newborn, a possible new starting point. Eur J Pediatr 1997;156:173–7. [19] Miki K, Zhou QQ, Guo W, Guan Y, Terayama R, Dubner R, et al. Changes in gene expression and neuronal phenotype in brain stem pain modulatory circuitry after inflammation. J Neurophysiol 2002;87:750–60. [20] Oberlander TF, Grunau RE, Whitfield MF, Fitzgerald C, Pitfield S, Saul JP. Biobehavioral pain responses in former extremely low birth weight infants at four months’ corrected age. Pediatrics 2000;105:e6. [21] Ramsay DS, Lewis M. The effects of birth condition on infants’ cortisol response to stress. Pediatrics 1995;95:546–9. [22] Reynolds ML, Fitzgerald M. Long-term sensory hyperinnervation following neonatal skin wounds. J Comp Neurol 1995;358:487–98. [23] Rodgers RJ, Dalvi A. Anxiety. Neurosci Biobehav Rev 1997;21: 801–10. [24] Ruda MA, Ling QD, Hohmann AG, Peng YB, Tachibana T. Altered nociceptive neuronal circuits after neonatal peripheral inflammation. Science 2000;289:628–31. [25] Schellinck HM, Anand KJ. Consequences of early experience: lessons for rodent models of newborn pain. In: McGrath PJ, Finley GA, editors. Chronic and recurrent pain in children and adolescents. Progress in pain research and management, vol. 13. Seattle, WA: ISAP Press; 1999. p. 39–55. [26] Taddio A, Katz J, Ilersich AL, Koren G. Effect of neonatal circumcision on pain response during subsequent routine vaccination. Lancet 1997;349:599–603. [27] Wall PM, Messier C. Methodological and conceptual issues in the use of the elevated plus maze as a psychological measurement instrument of animal anxiety-like behavior. Neurosci Biobehav Rev 2001;25:275–86. [28] Weiss S. Hindsight Ver. 1.5. Computer software and manual. Winnersh, UK: Vernalis; 1995. [29] Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–9.
Repetitive acute pain in infancy increases anxiety but does not alter spatial learning ability in juvenile mice Heather M. Schellinck∗ , Lianne Stanford, Matthew Darrah Department of Psychology, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4J1 Received 8 July 2002; received in revised form 2 December 2002; accepted 2 December 2002
Abstract We assessed the long-term behavioural effects of a single acute or repetitive inflammatory pain experienced during infancy. Groups of male and female CD1 mice were subjected to either an acute single pain, i.e. a tail clip or sham pain at P8, or acute repetitive pain in the form of needle pricks or sham pain from P8 to P14. All of the subjects were tested in the elevated plus maze at P30 and in the Morris Water Maze from P31 to P38. Mice in the acute single pain and sham groups did not differ on measures of anxiety in the plus maze. Mice in the repetitive pain group demonstrated significantly more anxious behaviours than controls in the elevated plus maze as they spent less time in the open arms, made fewer open arm entries, displayed fewer head dips and showed more stretch attend postures. There were no effects of single or repetitive pain treatments in the latency to find the hidden platform in the Morris Water Maze. Overall, these data suggest that acute repetitive pain experienced during infancy may increase anxiety later in life but it does not influence spatial learning as measured in the Morris Water Maze. The origin of the anxiogenic profile shown in the acute repetitive pain mice may be a result of changes in neural circuitry or context dependent learning and is currently under further investigation with this paradigm. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Repetitive pain; Anxiety; CD1 mice; Development; Learning and memory; Elevated plus maze; Morris Water Maze
1. Introduction The technical advances that have lead to enhanced survival for acutely ill newborns have been accompanied by an increase in potentially painful diagnostic and treatment procedures. Repeated heel pricks, intravenous drug delivery, endotracheal intubation and neonatal surgeries have become relatively common and often result in acute inflammatory pain in the neonate [5,18]. The mechanisms which contribute to inflammatory pain have been thoroughly investigated, primarily in preparations from adult animals, and involve alterations in intracellular signalling cascades as well as transcription-related increases in both receptors and neurotransmitters [13,17,19,29]. These responses may lead to short-term functional and structural changes in adult animals [1,2,15,22]. In newborns, these processes are experienced at a time when the brain is extraordinarily plastic with the nervous, endocrine and immune systems all undergoing functional and structural development [9]. Consequently, it has been hypothesised that exposure to painful stimuli during such a critical period could disrupt ∗
Corresponding author. Tel.: +1-902-494-6809; fax: +1-902-494-6585. E-mail address: [email protected] (H.M. Schellinck).
these neurodevelopmental processes and lead to permanent changes in the brain as well as in behaviour [4,25]. Several investigators have found long-term changes in response to inflammatory pain in neonatal rats in peripheral nerve terminals and within the spinal cord itself. Reynolds and Fitzgerald [22] assessed peripheral changes following skin wounds in P0, P7, P14 and adult rats. An examination of the axonal sprouting response revealed a long lasting, i.e. 21–42 days, cutaneous hyperinnervation. The innervation density was of a greater magnitude and duration when the wounds were performed in P0 and P7 rats compared with P14 animals. In the latter animals, changes were of a more transient nature similar to the effect found in adults. These results suggest the existence of a critical window for changes to create long-term effects. In the latter study, sensory thresholds were also measured in rats that received skin wounds on P0. They were found to be significantly lower than thresholds in same-age controls for at least 24 days. The authors suggested this provides evidence of pain or hypersensitivity long after the wound has healed. Recently, a persistent inflammatory pain in neonatal rats resulting from a single injection of Freund’s adjuvant (FA) in P0 and P3 neonatal rats has been shown to be associated with greater density and distribution in spinal cord laminae
0166-4328/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0166-4328(02)00406-0
158
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
I and II dorsal afferent axons in adult animals [24]. In contrast, such permanent changes were not demonstrated in adult animals that had previously been injected at P14. Behavioural responses to the initial injection of CFA were not assessed in adult animals. Nonetheless, when injected with formalin, adults that had been treated as neonates at an unspecified age exhibited pain responses earlier than controls during the late phase of the test [24]. To test the hypothesis that such structural changes could be paralleled by long-term behavioural effects, Anand et al. [3] assessed a number of behavioural responses of rat pups exposed to needle pricks daily from P0 to P7. Compared with controls, these rats showed decreased latency to respond to a thermal stimulus prior to weaning. There were no differences in hotplate latencies between adult experimental and control rats. As adults, rats that had received four paw pricks daily had an increased latency and decreased frequency to enter an open field and an increase in alcohol preference. The authors suggested that these behavioural effects in the adult rats subjected to repetitive neonatal pain were a result of increased anxiety. In a follow up study, Bhutta et al. [7] assessed the effects of formalin-induced inflammatory pain on P0 to P7 rats. When tested as adults, both male and female experimental animals had increased hotplate latencies compared with controls and males showed a decreased preference for alcohol. Males but not females showed a decrease in locomotor activity in a behavioural chamber. The different methods used to induce the early inflammatory pain in these experiments may have resulted in the mixed support for the hypothesis that early repetitive pain leads to increased anxiety in adulthood. Anand et al. [3] also determined that following repeated paw pricks as neonates adult rats recognised familiar conspecifics in a social discrimination task for a longer period of time than controls. This latter result is in agreement with other work that suggests that memory and/or performance is enhanced by a previous painful experience. Neonatal rats exposed to thermal pain had better performance on a two-way active avoidance task than control animals [6]. Similarly, young rats that had experienced foot shocks mastered a lever pressing task faster than controls [16]. The experiments thus far have reported physical and behavioural effects in rats (P0–P7) that are developmentally equivalent to preterm infants [10]. In this study, we have used a different procedure to determine if an early pain experience in older neonatal mice also results in long-term behavioural changes. We exposed P8 mice to a single event of acute pain in the form of a tail clip. A second group of mice were given daily paw pricks (P8–P14) on the dorsal surface of a front and contralateral rear paw. We choose to use animals of these ages as the maturity of an 8-day-old rodent has been reported to be comparable to that of a full-term infant [10]. We examined the effects of these procedures in P30 to P38 animals to determine if the hypothesised effects would manifest themselves in prepubertal animals. These procedures enabled us to investigate the long-term effects of a fairly typical experience in a full-term human infant.
Given the somewhat conflicting findings as to whether anxiety is affected by early pain [3,7], we used the elevated plus maze to assess anxiety. This test is known for its high ecological validity and reliable measurement of behaviours associated with anxiety and risk assessment in rodents [23,27]. To further investigate reports that pain in neonates enhances learning [6,16], we also assessed spatial learning in the Morris Water Maze.
2. Materials and methods 2.1. Animals CD1 mice (N = 56), offspring of adult males and females purchased from Charles River Canada (St. Constant, Quebec) and bred at Dalhousie University, were used in this experiment. The day of birth was designated as Postnatal Day 0 (P0). Litters were culled to a group of four males and four females on P3 and weaned on P21. Prior to weaning, the mice were housed with their dams in polycarbonate cages (30 cm × 15 cm × 10 cm) with wood-chip bedding. Following weaning, the mice were housed in same sex groups of four in polycarbonate cages (30 cm × 15 cm × 10 cm). They were maintained on a 12 h/12 h light/dark cycle with lights off at 13:30 h and provided with Purina Agribrand diet 5001 and water ad libitum. On P8, males and females from each litter were quasi-randomly assigned to each of the following four treatment groups: acute single pain, sham single pain, acute repetitive pain and sham repetitive pain. All animals were treated in accordance with the Canadian Council on Animal Care guidelines and the experimental protocol was approved by the Dalhousie University Committee on Laboratory Animals. 2.2. Pain treatments 2.2.1. Single acute and sham pain On Day 8, during the dark phase of the L/D cycle, two male and two female mice from each litter were temporarily placed in a holding cage. One male and one female mouse received a 2.5 mm tail clip with surgical scissors. The remaining male and female in the sham single pain group were tail marked with a permanent marker at the 2.5 mm mark but no cut was made. For identification purposes, mice were marked with nontoxic acrylic artists ink (Daler-Rowney, Jamesburg, NJ) following the procedure. The four mice were then returned to their home cage. On Days 9–14, these mice were transferred daily to a clean holding cage to minimise the handling differences between these groups and the repetitive pain groups. 2.2.2. Repetitive acute and sham pain On Day 8, during the dark phase of the L/D cycle, the remaining four mice were removed from the litter. One male and one female were given needle pricks on the dorsal
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
surface of a front and contralateral rear paw with a 26-gauge needle piercing the paw. The remaining two animals received sham pricks by placing the flat edge of the needle on the dorsal surface of both paws. Both treatments were continued daily up to and including Day 14. The pain treatment was counterbalanced such that two litters received the pain in the right front and rear left paw and two litters received the pain in the left front and right rear paw. 2.3. Elevated plus maze The elevated plus maze consisted of four arms (30 cm × 5 cm) in the shape of a cross, extending from a central square (5 cm × 5 cm) and elevated 45 cm above the ground. Two of the opposing arms had a 4 mm lip around the edges (open arms) and the remaining two arms were enclosed with 15 cm transparent Plexiglas walls (closed arms). The floor of the maze was composed of black Plexiglas. The test room was illuminated by a 60 W red light bulb. A video camera linked to a TV and a VCR were used to record each trial. On Day 30, during the dark phase of the L/D cycle, the mice were tested individually in the maze. Each mouse was lowered onto the central square and their behaviour in the maze videotaped for 5 min for subsequent analysis with Hindsight Ver 1.5 [28]. Video analysis was completed with the observer blind to the experimental conditions for the following behaviours: (1) amount of time spent in the open arms (four paws in arm); (2) number of entries into the open arms; (3) locomotor activity; (4) frequency of head dipping, i.e. looking over the edge of an open arm; (5) number of stretch attend postures, i.e. pausing while stretched out with head lowered and rear tilted upward; (6) time spent rearing, i.e. standing up on hind legs with or without pressing the front paws on the Plexiglas wall; (7) time spent grooming. After each mouse completed the maze, the presence or absence of fecal boli was recorded and the entire maze was cleaned with 70% ethanol. Behaviour in the elevated plus maze was assessed prior to Morris Water Maze testing so that the handling associated with the latter paradigm would not confound the measurement of anxiety. 2.4. Morris Water Maze A Morris Water Maze (MWM) was created from a plastic swimming pool (diameter: 110 cm; height: 20 cm) filled with 20–22 ◦ C water to a depth of 14 cm. The water was made opaque by the addition of 500 ml of nontoxic white liquid Tempura Paint (Schola Brand, Marieville, Quebec, Canada). The circular concealed platform was constructed from white plastic (diameter: 10 cm; height: ∼14 cm) such that it was just below the surface of the water. During Sessions 1–3 (P31–P33), the concealed platform was placed in the south-east quadrant; during Sessions 4–6 (P34–P36), the platform was moved to the north-west
159
quadrant. During Session 7, the platform was removed and a probe test completed. During Session 8 (P38), the platform was placed in the NE quadrant and made visible with the addition of a 3 cm high red platform so that it was 2.5 cm above the water. Fixed spatial cues including several posters with geometric designs and office furniture were located around the maze in the test room. A video camera linked to a TV and a VCR were used to record each session. On P31 to P36 and P38, the mice were tested in this maze for four trials a day for 6 days during the dark phase of the L/D cycle. The mouse was lowered from a plastic cup into the maze at one of the four marked principal points of a compass such that for each four trial sessions, it entered the water once from each north, south, east and west point. The order of entry was quasi-randomised over sessions. The mice were given 60 s to find the platform after which time they were directed to the platform with a plastic container and forced to remain on the platform for 20 s. Latency to find the platform was recorded with a stopwatch. On P37, the mice were tested for a single 60 s probe trial during which the platform was absent. For the probe trial, the amount of time spent in each quadrant and in the outer sixth annulus of the pool (thigmotaxis) was recorded from video following completion of the session. 2.5. Statistical analysis Two of the fifty-six mice were incorrectly sexed prior to the beginning of the experiment such that there were only five mice in the male paw prick control group. Thus, sex differences were not assessed. 2.5.1. Elevated plus maze The effects of the single acute pain on time spent in the open arms, number of open arm entries, frequency of line crosses, stretch attend postures, head dips, duration of grooming and rearing and were assessed with two-tailed t-tests. The amount of time spent in the open arms was calculated as the percentage of time spent in the open arms divided by the amount of time that the mice spent in the open and closed arms combined (excluding any centre square time). Similarly, the effect of repetitive pain on the same measures were also were examined by t-tests. Differences were considered to be significant if P < 0.05. 2.5.2. Morris Water Maze The effects of single acute pain on latency to find the hidden platform during 3 days of acquisition training was assessed in a two-way mixed ANOVA. To complete this analysis, the individual scores of each animal on the four trials each day were averaged to give a mean latency per day. Similarly, the latency to find the hidden platform was also assessed during 3 subsequent days of reversal training. Latency to find the hidden platform during the last acquisition and first reversal days was assessed in a two-tailed
160
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
t-test. The total amount of time spent in each quadrant during the probe trial in which the platform was removed was examined in a 4 × 2 ANOVA. The degree of thigmotaxia expressed during the probe trial was assessed in a two-tailed t-test as was the mean time to find the visible platform on the day following the probe trial. Newman–Keuls post hoc tests were used to find differences across groups where required.
3. Results 3.1. Elevated plus maze There were no significant differences between mice that received the single tail cut and sham cut controls in any of the eight behaviours scored in the elevated plus maze (all ts(26) < 1.31), Fig. 1 illustrates these results.
Fig. 1. Mean + S.E.M. for the (A) number of line crosses, (B) percent of time spent in the open arms, (C) number of head dips, (D) number of open arm entries, (E) number of stretch attend postures, (F) number of rearing bouts, (G) probability of the presence of fecal boli and (H) duration of grooming for the sham tail cut mice (white bars) and the tail cut mice (black bars).
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
Mice that received repetitive paw pricks made fewer open arm entries (t (26) = −3.80, P < 0.0008), spent less time in the open arms of the elevated plus maze (t (26) = −3.32, P < 0.003), displayed fewer head dips (t (26) = −2.55, P < 0.02) and showed more stretch attend postures than the sham pricked mice (t (26) = 2.47, P < 0.02). There were
161
no differences between the paw pricked and control mice for the number of lines the mice crossed (t (26) = −1.26, P > 0.22), time spent rearing (t (26) = 0.75, P > 0.46) or grooming (t (26) = −0.24, P > 0.81) or probability of defecation (t (26) = 1.59, P > 0.12). Fig. 2 illustrates these results.
Fig. 2. Mean (+S.E.M.) for the (A) number of line crosses, (B) percent of time spent in the open arms, (C) number of head dips, (D) number of open arm entries, (E) number of stretch attend postures, (F) number of rearing bouts, (G) probability of the presence of fecal boli and (H) duration of grooming, for the sham paw pricked mice (white bars) and the paw pricked mice (black bars). The asterisk (∗) illustrates significant differences between sham and paw pricked mice (P < 0.05).
162
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
3.2. Morris Water Maze For the acute pain mice and their controls, there was no main effect of pain (single pain versus sham) in the latency to find the hidden platform during the 3 days of acquisition training (F (1, 26) = 1.12, P > 0.30) but there was a decrease in latency over the 3 test days for both groups (F (2, 52) = 24.13, P < 0.0001) with no interaction between test day and pain (F (2, 52) < 1). There was no disruption in the latency to find the platform when it was moved from the acquisition quadrant to the reversal quadrant between Days 3 and 4 (F (1, 26) = 1.20, P > 0.28) and there was no effect of pain and no interaction between pain group and day (F s < 1). During the reversal days, the mice showed a significant decrease in escape latency over the 3 days (F (2, 52) = 7.73, P < 0.002) with no main effect of pain and no interaction between test day and pain (F s < 1; Fig. 3A). During the 60 s probe trial, the total amount of time spent in each quadrant was significantly different (F (3, 78) = 16.56, P < 0.0001). Newman–Keuls post hoc tests revealed that more time was spent in the quadrant which most recently had the platform (reversal) compared to the other three quadrants. There were no differences found between
the pain and control groups (F (1, 26) = 1.08, P > 0.31) and no significant interaction between quadrant and group (F (3, 78) < 1) as shown in Fig. 3B. An analysis of thigmotaxic behaviour showed no experimental group differences in duration (t (26) = −0.174, P > 0.86; means ± S.E.M.: 11.01 ± 1.69 (pain) and 11.43 ± 1.77 (sham)) or frequency (t (26) = −0.76, P > 0.45; 9.71 ± 1.04 (pain) and 10.93 ± 1.21 (sham)). Finally, the latency required for the mice to reach the visible platform showed no differences between groups (t (26) = 1.38, P > 0.18; Fig. 3B). Paw pricked mice and their controls were compared over the first 3 days of acquisition in the Morris Water Maze and there was an reduction in the latency to find the platform across days (F (2, 52) = 25.84, P < 0.0001). Overall, there was no difference between the paw pricked mice and controls (F (1, 26) = 2.51, P > 0.13) and no interaction between pain group and day (F (2, 52) < 1). There was no increase in latency to find the platform when comparing the third day of acquisition to the first day of reversal (F (1, 26) < 1), no difference between the two groups (F (1, 26) < 1) and no significant interaction between day and experimental manipulation (F (1, 26) = 3.18, P > 0.08). During the reversal days, the mice showed a decrease in latency to find the hidden platform over the 3 days (F (2, 52) =
Fig. 3. (A) Mean latency ± S.E.M. for the tail cut (black circles) and sham tail cut mice (white circles) to find the hidden platform in the original location (acquisition), the opposite quadrant (reversal) and when it was made visible and moved to a third location in the Morris Water Maze (visible). (B) Mean duration + S.E.M. spent by the sham mice (white bars) and the tail cut mice (black bars) in each quadrant of the pool during the probe trial.
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
163
Fig. 4. (A) Mean latency ± S.E.M. for the paw pricked (black squares) and sham paw pricked mice (white squares) to find the hidden platform in the original location (acquisition), the opposite quadrant (reversal) and when it was made visible and moved to a third location (visible). (B) Mean duration + S.E.M. spent by the sham mice (white bars) and the tail cut mice (black bars) in each quadrant of the pool during the probe trial.
17.18, P < 0.0001, with means ± S.E.M. of 28.71 ± 2.54 (Day 4), 17.71 ± 2.11 (Day 5) and 15.72 ± 1.51 (Day 6)). The main effect of group (F (1, 26) < 1) and the interaction of group and day (F (2, 52) = 1.05, P > 0.36) were not significant (Fig. 4A). During the probe trial, the total amount of time spent in each quadrant was significantly different (F (3, 78) = 24.55, P < 0.0001). Newman–Keuls revealed that the mice spent more time in the quadrant from which the platform was removed (reversal) compared to the remaining three quadrants, and that the amount of time spent in the left quadrant was greater than that spent in either the opposite quadrant or the right quadrant. There were no differences found between the experimental groups (F (1, 26) < 1), and no significant interaction of the experimental group and quadrant (F (3, 78) = 1.97, P > 0.12). An analysis of thigmotaxic behaviour during the probe trial showed no differences between the experimental and control groups for duration, (t (26) = −1.31, P < 0.20; 11.04 ± 1.35 (pain), 14.00 ±1.81 (sham)) or for frequency of thigmotaxic bouts (t (26) = −1.58, P < 0.12; 9.93 ± 0.8 (pain) and 12.29 ± 1.26 (sham)). Finally, there were no differences observed between the experimental groups in the latency of the mice to find the visible platform (t (26) = −1.86, P < 0.07; Fig. 4B).
4. Discussion In these experiments, we have demonstrated that infant mice that experience acute repetitive pain from P8 to P14 show a pattern of behaviour in the elevated plus maze as juveniles that is characteristic of anxiety, i.e. a significant increase in stretch attends and a decrease in head dips and time spent in the open arms and frequency of open arm entries. In contrast, mice that experienced a single acute pain on P8 did not show anxious behaviours as juveniles. Mice exposed to either a single or repetitive pain experience performed as well as littermate controls on the Morris Water Maze. Overall, these results indicate that there are long-term behavioural effects of early repetitive pain on anxiety in mice that can be demonstrated in prepubertal animals. These effects manifest themselves as an increase in anxiety in the elevated plus maze but not as an alteration in performance in a spatial learning task. Two previous studies [3,7] have also found that early repetitive pain in either the form of a needle prick or a formalin injection produced long-term behavioural effects in adult rats. We have extended these findings by demonstrating that mice as well as rats are vulnerable to the effects of repetitive pain. Nonetheless, there are some important differences in these studies. The work of Anand and coworkers
164
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165
[3,7] was designed to model the effects of pain on preterm infants and as such exposed rats to pain from P1 to P7. The mice in our study were older (P8–P14) when exposed to pain and more analogous to full-term infants. Thus, we have demonstrated that animals with more developed peripheral and central nervous systems are also vulnerable to the effects of repetitive pain. This suggests that the concept of a critical period in which pain must be experienced to have long-term effects should be expanded to include animals within the age range of P8 to P14. The mice in our study were also younger (P30–P38) when tested for long-term effects. Thus, it is clear that anxiety associated with early pain is manifest in mice in a prepubertal state. These findings point to the need for early assessment and treatment of anxiety-like symptoms in children exposed to pain, regardless of their status as term or preterm infants. Mice exposed to pain and mice in the sham conditions performed equally well in the Morris Water Maze with both groups showing a decrease in latency to find the hidden platform, no reduction in performance after the location of the hidden platform was reversed and no differences in search strategy during the probe trial. Our failure to find evidence of enhanced performance in the Morris Water Maze as a result of early acute or repetitive pain is not in agreement with the results of previous reports on the influence of pain on learning [6,16]. Anand et al. [3] also found that adult rats that had specifically experienced early repetitive pain recognised conspecifics for a longer time period than controls. The authors suggested that anxiety-related hypervigilance may have led to the latter result. This state of anxiety could have also accounted for the differences in learning noted in the other reports in that more vigilant animals may become aware of task demands more quickly, thus leading to enhanced performance. Anxiety has been suggested to lead to inefficient search strategies and poor performance on the Morris Water Maze [8,11]. In our experiment, as measured by frequency and duration of thigmotaxic behaviour in the probe trial test, anxiety in the Morris Water Maze did not differ between animals that experienced pain and those that had not. The need for further investigation of the effect of pain on learning in a number of paradigms is required before any substantive conclusions can be drawn. Our results did not support the hypothesis that a single acute pain would cause long-term behavioural changes in mice as had previously been found in humans [26]. It is possible that the type of pain administered in this experiment, i.e. tail cuts in mice, is not comparable to the single pain experience, i.e. circumcision, assessed in humans. Nonetheless, other factors that might have influenced the response to pain in humans should also be taken into consideration. In particular, there may be context dependent memory effects in humans such that the experience of pain in similar environments, i.e. a clinic and a hospital, is sufficient to evoke the emotional aspects of pain regardless of the lack of long-term nociceptive changes. In our experiment, the rooms in which the tail cuts were performed, i.e. the odour laden animal
room versus the sterile testing room, were very different. Thus, it seems much less likely that the context would evoke any psychological aspect of the previous painful event that could manifest itself as increased anxiety. There have been an increasing number of reports that human infants exposed to pain and other stressful experiences at birth show altered responses to pain during infancy and childhood. Infants circumsized at birth show increased behavioural response to vaccination at 4 months of age [26]. Stressful birth conditions are also associated with high levels of cortisol following vaccinations at 6 months of age [21]. Extremely low birth weight (ELBW) infants interviewed at 8–10 years rated the intensity of painful medical events as higher than their full-term peers; moreover, those who spent more time in intensive care gave higher affective ratings to common painful events [12]. ELBW infants also show a different pattern of cardiac response to finger lance at 4 months of age [20]. Hack et al. [14] have also argued that the attentional deficits, poorer social skills and adaptive behaviour problems in LBW infants may well be associated with the extensive painful procedures experienced by these children at birth. Given the affective nature of the responses in many of these studies, it seems possible that the effects may be related to structural and functional alterations in brain regions associated with processing of emotion rather than perturbations in noiciceptive circuitry. Because of the obvious ethical considerations, it is almost impossible to examine the causal relationship between early pain and long-term behavioural effects in humans. The animal paradigm described in this experiment has provided us with a means to investigate both the nocieptive and emotional mechanisms of pain that underlie these long-term changes.
Acknowledgements This research was supported by a NSERC operating grant to H.M. Schellinck and a K.M. Hunter Charitable Foundation/CIHR Doctoral research award and the Spatz Doctoral Training Award 2002 (Alzheimer Society of Canada and IA-CIHR) to L.E. Stanford.
References [1] Aldskogius H, Hermanson A, Jonsson CE. Reinnervation of experimental superficial wounds in rats. Plast Reconstr Surg 1987;79: 595–9. [2] Alvares D, Torsney C, Beland B, Reynolds M, Fitzgerald M. Modelling the prolonged effects of neonatal pain. In: Sandkuhler J, Bromm B, Gebhart GF, editors. Nervous system plasticity and chronic pain. Progress in brain research, vol. 129. Amsterdam: Elsevier; 2000. p. 365–73. [3] Anand KJ, Coskun V, Thrivikraman KV, Nemeroff CB, Plotsky PM. Long-term behavioral effects of repetitive pain in neonatal rat pups. Physiol Behav 1999;66:627–37.
H.M. Schellinck et al. / Behavioural Brain Research 142 (2003) 157–165 [4] Anand KJ, Scalzo FM. Can adverse neonatal experiences alter brain development and subsequent behavior? Biol Neonate 2000;77: 69–82. [5] Barker DP, Simpson J, Pawula M, Barrett DA, Shaw PN, Rutter N. Randomised, double blind trial of two loading dose regimens of diamorphine in ventilated newborn infants. Arch Dis Child Fetal Neonatal Ed 1995;73:F22–6. [6] Bernardi M, Genedani S, Bertolini A. Behavioral activity and active avoidance learning and retention in rats neonatally exposed to painful stimuli. Physiol Behav 1986;36:553–5. [7] Bhutta AT, Rovnaghi C, Simpson PM, Gossett JM, Scalzo FM, Anand KJ. Interactions of inflammatory pain and morphine in infant rats: long-term behavioral effects. Physiol Behav 2001;73:51–8. [8] Brooks AI, Cory-Slechta DA, Murg SL, Federoff HJ. Effects of behaviourally induced stress on spatial memory acquisition and retention in mice: a biochemical evaluation. Abstr Soc Neurosci 2000;26:2239. [9] Constantine-Paton M. The plastic brain. Neurobiol Dis 2000;7: 515–9. [10] Fitzgerald M, Anand KJ. The developmental neuroanatomy and neurophysiology of pain. In: Schechter NL, Berde CB, Yaster M, editors. Pain in infants, children, and adolescents. Baltimore, MD: Williams & Wilkins; 1993. p. 11–32. [11] Grootendorst J, de Kloet ER, Dalm S, Oitzl MS. Reversal of cognitive deficit of apolipoprotein E knockout mice after repeated exposure to a common environmental experience. Neuroscience 2001;108: 237–47. [12] Grunau RE, Whitfield MF, Petrie J. Children’s judgements about pain at age 8–10 years: do extremely low birthweight (≤1000 g) children differ from full birthweight peers? J Child Psychol Psychiatry 1998;39:587–94. [13] Guenther S, Reeh PW, Kress M. Rises in [Ca2+ ]i mediate capsaicinand proton-induced heat sensitization of rat primary nociceptive neurons. Eur J Neurosci 1999;11:3143–50. [14] Hack M, Taylor HG, Klein N, Eiben R, Schatschneider C, Mercuri-Minich N. School-age outcomes in children with birth weights under 750 g. N Engl J Med 1994;331:753–9. [15] Hermanson A, Dalsgaard CJ, Bjorklund H, Lindblom U. Sensory reinnervation and sensibility after superficial skin wounds in human patients. Neurosci Lett 1987;74:377–82.
165
[16] Kim DG, Lee S, Lim JS. Neonatal footshock stress alters adult behavior and hippocampal corticosteroid receptors. Neuroreport 1999;10:2551–6. [17] Li P, Wilding TJ, Kim SJ, Calejesan AA, Huettner JE, Zhuo M. Kainate-receptor-mediated sensory synaptic transmission in mammalian spinal cord. Nature 1999;397:161–4. [18] McIntosh N. Pain in the newborn, a possible new starting point. Eur J Pediatr 1997;156:173–7. [19] Miki K, Zhou QQ, Guo W, Guan Y, Terayama R, Dubner R, et al. Changes in gene expression and neuronal phenotype in brain stem pain modulatory circuitry after inflammation. J Neurophysiol 2002;87:750–60. [20] Oberlander TF, Grunau RE, Whitfield MF, Fitzgerald C, Pitfield S, Saul JP. Biobehavioral pain responses in former extremely low birth weight infants at four months’ corrected age. Pediatrics 2000;105:e6. [21] Ramsay DS, Lewis M. The effects of birth condition on infants’ cortisol response to stress. Pediatrics 1995;95:546–9. [22] Reynolds ML, Fitzgerald M. Long-term sensory hyperinnervation following neonatal skin wounds. J Comp Neurol 1995;358:487–98. [23] Rodgers RJ, Dalvi A. Anxiety. Neurosci Biobehav Rev 1997;21: 801–10. [24] Ruda MA, Ling QD, Hohmann AG, Peng YB, Tachibana T. Altered nociceptive neuronal circuits after neonatal peripheral inflammation. Science 2000;289:628–31. [25] Schellinck HM, Anand KJ. Consequences of early experience: lessons for rodent models of newborn pain. In: McGrath PJ, Finley GA, editors. Chronic and recurrent pain in children and adolescents. Progress in pain research and management, vol. 13. Seattle, WA: ISAP Press; 1999. p. 39–55. [26] Taddio A, Katz J, Ilersich AL, Koren G. Effect of neonatal circumcision on pain response during subsequent routine vaccination. Lancet 1997;349:599–603. [27] Wall PM, Messier C. Methodological and conceptual issues in the use of the elevated plus maze as a psychological measurement instrument of animal anxiety-like behavior. Neurosci Biobehav Rev 2001;25:275–86. [28] Weiss S. Hindsight Ver. 1.5. Computer software and manual. Winnersh, UK: Vernalis; 1995. [29] Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–9.