- Email: [email protected]
Prolonged deficits of associative motor learning in cynomolgus monkeys after long-term administration of phencyclidine
Behavioural Brain Research 331 (2017) 169–176
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Prolonged deficits of associative motor learning in cynomolgus monkeys after long-term administration of phencyclidine
MARK
Bing Wua,b, Xu-dong Zhaoc, Hui-min Zhanga,b, Xuan Lia,b, Guang-yan Wua,b, Ying-shan Yangd, ⁎ Chao-yang Tiand, Jian-feng Suia,b, a
Department of Physiology, College of Basic Medical Sciences, Third Military Medical University, Chongqing 400038, China Experimental Center of Basic Medicine, College of Basic Medical Sciences, Third Military Medical University, Chongqing 400038, China c State Key Laboratory of Brain and Cognitive Sciences, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China d Hainan Jingang Biological Technology Co., Ltd., Haikou, Hainan 571100, China b
A R T I C L E I N F O
A B S T R A C T
Keywords: Phencyclidine (PCP) Eye-blink conditioning (EBC) Schizophrenia Cerebellum Learning
Phencyclidine (PCP) is a potent drug of abuse that induces sustained schizophrenia-like symptoms in humans by blocking neurotransmission at N-methyl-D-aspartate (NMDA)-type glutamate receptors. Alterations in NMDA receptor function have been linked to numerous behavioral deficits and cognitive dysfunction. Classical eyeblink conditioning (EBC), including delay (dEBC) and trace (tEBC) paradigms, provides an effective means to study the neurobiology of associative motor learning in rodents, mammals and primates. To assess whether administration of low-dosage PCP for extended periods has prolonged effect to alter associative motor learning, in this study 19 adult cynomolgus monkeys were administered PCP (0.3 mg/kg, intramuscularly) or saline twice a day for 14 days. Twelve–fifteen months after PCP or saline injection, monkeys received dEBC, tEBC, or pseudopaired training for 6 or 12 successive daily sessions, respectively. The results of this study show that percentage of conditioned response (CR) in dEBC increased as a function of training sessions in both PCP-treated and control monkeys and there was no significant CR% difference between the two groups. However, the CR timing in dEBC of PCP-treated monkeys was significantly impaired, as manifested by shorter CR peak latencies than those of the control group. PCP-treated animals showed significantly lower percentage of CR in tEBC compared to controls. PCP-treated animals were also more sensitive to outside stimuli in tEBC because the UR peak latency of PCPtreated group was significantly lower than the control group. These results indicated that cynomolgus monkeys manifested prolonged deficits in associative motor learning after long-term administration of phencyclidine.
1. Introduction Phencyclidine (PCP) is a psychotomimetic drug and a noncompetitive antagonist of the N-methyl-D-aspartate (NMDA) glutamate receptor. PCP-induced psychosis is associated with both positive and negative schizophrenia-like symptoms in humans [1]. Repeated administration can result in a long-lasting syndrome marked by neuropsychological deficits, social withdrawal, and affective blunting as well as hallucinations, paranoia, and delusions [2,3]. PCP also induces cognitive and behavioral dysfunctions that partially correspond to the positive and negative symptoms of schizophrenia [4–6]. Abnormalities in working memory, behavioral inhibition, and social interactions have been observed in schizophrenic individuals and in PCP-abusing humans [7–9]. Because the long-term abuse of PCP in humans may represent a pharmacological model of learning and memory deficits that are associated with schizophrenia, we explored the prolonged associative ⁎
learning effects of long-term exposure to PCP in the cynomolgus monkey (Macaca fascicularis). Classical eye-blink conditioning (EBC) has been extensively used to study the neurobiology of associative motor learning in rodents, mammals and primates [10,11]. EBC involves paired presentations of a tone or light as a conditioned stimulus (CS) and a periorbital shock or air puff as an unconditioned stimulus (US). Initially, the CS produces no obvious eye-blink responses while the US elicits reliable eye-blink responses before learning, named the unconditioned response (UR). After repeated pairings of the CS and US, the CS comes to elicit an eyeblink called the conditioned response (CR). The two widely employed paradigms of EBC are delay EBC (dEBC) and trace EBC (tEBC). In the dEBC, CS is presented before the US, and the two stimuli are coterminated. In tEBC, CS and US are presented separately in time so that a complete stimulus-free period exists between the two stimuli. EBC has been studied for more than half a century in humans [12–14] as well as
Corresponding author at: Department of Physiology, College of Basic Medical Sciences, Third Military Medical University, Chongqing 400038, China. E-mail address: [email protected] (J.-f. Sui).
http://dx.doi.org/10.1016/j.bbr.2017.05.035 Received 2 March 2017; Received in revised form 5 May 2017; Accepted 10 May 2017 Available online 23 May 2017 0166-4328/ © 2017 Elsevier B.V. All rights reserved.
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
in animals [11,15,16]. These studies led to a consensus that dEBC is dependent upon the cerebellum-brainstem circuits and that tEBC involves the hippocampus, prefrontal cortex and other brain areas in addition to cerebellum-brainstem circuits. While dEBC provides a wellvalidated method to investigate cerebellar timing function, tEBC seems to be a sensitive task related to functions of the prefrontal cortex and hippocampus [17,18]. As an ideal model for associative motor learning studies, changes of EBC in schizophrenia has received increasing attention in recent years [19–21]. Repeated exposure to PCP reportedly produces a sustained decrease in dopamine turnover within the prefrontal cortex, which is accompanied by a deficit in working memory and prefrontal cortex-dependent tasks in both rats and non-human primates [22–24]. In this study, we treated cynomolgus monkeys with PCP (0.3 mg/kg twice a day, intramuscularly) for 14 days, and 12–15 months later, we estimated the effect of PCP on establishment of both dEBC and tEBC. The goals of this study were (1) to assess whether administration of low-dosage PCP for extended periods has prolonged effect to alter associative learning ability in cynomolgus monkeys, and (2) to explore whether, in addition to damage to the frontal lobe and prefrontal cortex, extended PCP administration leads to impaired cerebellar function, which is involved in behavioral timing.
Fig. 1. Monkey conditioning goggles and primate chair. The goggles include the eye-blink detector and air-puff delivery pipe. A monkey wearing the conditioning goggles was sat in the primate chair with its legs and arms fixed. The right eye is monitored for eye-blinks, and the left eye is unobstructed.
(Fig. 1).
2. Materials and methods
2.4. Training and recording system
2.1. Subjects
During behavioral training, the CS was a 1-kHz, 80-dB pure tone. All sound stimuli were delivered by two speakers placed 60 cm to the left and right ears of the animals, respectively. A sound-level meter (type 2240, Brüel & Kjær) was used to calibrate the intensity of the CS tone. The US was a 100-ms, 5-psi (measured at the tip of a plastic pipe attached to goggles) source pressure air-puff delivered to the right cornea through a plastic pipe. Presentations of the CS and US were controlled by a self-made computer system. All signals including markers of the applied stimuli and signals from the infrared sensor were digitized by a data-acquisition system (RM6240BDJ, Cheng Yi, Chengdu, China) and were acquired and stored by the system software (v. 4.7). Data analysis was carried out on a dedicated Windows PC.
Nineteen healthy adult cynomolgus monkeys (9 females, 10 males) from the breeding colonies at the Hainan Jingang biological technology Co., Ltd. were served as subjects. The monkeys weighed between 3.5 and 6.5 kg at the beginning of behavioral tests, had no obvious eye diseases, and were sensitive to sounds stimuli. The monkeys were individually housed under standard conditions (a 12-h light/dark cycle with light on from 07:00 to 19:00, humidity 60%, 21 ± 2 °C, 3 times/ day deliveries of food and 1 time/day delivery of fruit, water available ad libitum). The experimental procedures were approved by the Animal Care Committee of the Third Military Medical University and were performed in accordance with the principles outlined in the NIH Guide for the Care and Use of Laboratory Animals.
2.5. Experimental procedures 2.2. Primate chair 2.5.1. PCP or Saline treatment A total of 19 adult cynomolgus monkeys were included in the experiment, including dEBC paired learning (n = 6); tEBC paired learning (n = 7); and pseudo-paired training for both dEBC and tEBC (n = 6). Monkeys were treated with daily intramuscularly injection of PCP (Chemsky [Shanghai] international Co., LTD.) at the dosage of 0.3 mg/kg (for PCP-treated group), or equal volume of saline (for control group), twice a day for 14 days [22]. Twelve–Fifteen months after injection of the PCP or Saline, monkeys in different groups received dEBC, tEBC, or pseudo-paired behavioral training, respectively.
A special primate chair was designed according to the monkey's physical characteristics to ensure that the monkey can sit in the chair comfortably with its limbs easy to be fixed. The design of the chair prevented monkeys from touching their heads with their hands or feet. Once a monkey was secured in the chair, the circular bayonet on the top section of the chair can limit the up and down movement of its head and body, the head could only rotate around. 2.3. Eye-blink Detection Goggles To detect monkeys’ eye-blink behavior, a specially designed pair of goggles was constructed and worn by the monkey during behavior training. Eye-blinks were measured with an infrared sensor consisting of an infrared emitter (FBCB30, HengSheng, Shenzhen, China) and an infrared detector (TBBB30, HengSheng, Shenzhen, China) encased side by side and aligned with converging optical axes. The converging optical axes were aligned for maximum sensitivity to detect eye-blinks at a distance of 0.5-cm from the infrared sensor surface to the pupil surface. The infrared sensor and a plastic pipe with 0.2-cm diameter were established and secured together with dental acrylic on the goggle’s right lens. The goggles were held in place by a rubber strap that attached to the lateral edges of the goggles and around the back of the subject’s head (under the ears). The rubber strap looseness was adjusted to ensure that animals worn goggles comfortably and firmly
2.5.2. Habituation All subjects were first habituated to sitting in the primate chair and wearing eye-blink detector goggles for 2 h each day for 6 days. No stimulation was applied to the monkeys during the habituation period. 2.5.3. Behavioral training Six monkeys (PCP-treated: n = 3; control: n = 3) included in dEBC paired learning underwent delay eyeb-link conditioned training for 6 successive daily sessions after habituation; 7 monkeys (PCP-treated: n = 3; control: n = 4) included in tEBC paired learning underwent trace eye-blink conditioned training for 12 successive daily sessions. In addition, in order to control the possible non-associative learning, 6 monkeys (PCP-treated: n = 3; control: n = 3) were included in pseudo170
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 2. Experimental design and temporal relationship between the conditioned stimulus (CS), unconditioned stimulus (US), and analysis periods for the eye-blink conditioning (EBC). (A) Schematic diagram showing the delay eye-blink conditioning paradigm used in the present study. The CS was presented 400 ms before the US onset and remained on while the 100-ms air-puff US was delivered and they terminated together. In each trial, we analyzed the parameters of the conditioned eye-blink response (CR; 100–400 ms after CS onset) and the unconditioned eye-blink response (UR; 400–700 ms after CS onset); (B) Schematic diagram showing the trace eye-blink conditioning paradigm used in the present study. In this paradigm, the CS (100 ms, tone) was paired with the US (100 ms, air-puff) and the CS was always presented before the onset of the US, a temporal gap (500 ms) occurs between the offset of the CS and the onset of the US. In each trial, we analyzed the parameters of the conditioned eye-blink response (CR; 200–600 ms after CS onset) and the unconditioned eye-blink response (UR; 600–1000 ms after CS onset).
the CS onset; (2) a “CR” period, 100–400 ms after the CS onset; and (3) a “UR” period, 400–700 ms after the CS onset (Fig. 2A). For tEBC training, each CS-US paired trial was also subdivided into three discontinuous analysis periods: (1) the “baseline” period, −1400–0 ms before the CS onset; (2) a “CR” period, 200–600 ms after the CS onset; and (3) a “UR” period, 600–1000 ms after the CS onset (Fig. 2B). A significant eye-blink movement was defined as an increase in amplitude that was greater than the mean baseline amplitude, plus four times the standard deviation of the baseline activity. In addition, the significant eye-blink movement required a minimal duration of 30 ms. The significant eye-blink movement during the “SB”, “CR” and “UR” period was counted as a SB, a CR or a UR, respectively (Fig. 3). The percentage of CR (CR%) was defined as the ratio of the number of trials that contained the CR to the total number of trials. The CR peak latency was defined as the time interval from the CS onset to the peak of the CR. Similarly, the UR peak latency was defined as the time interval from the US onset to the peak of the UR.
paired training to serve as control of paired learning of both dEBC and tEBC. They received the delay pseudo-conditioned training when served as pseudo-paired control group of dEBC, and the trace pseudo-paired training when served as pseudo-paired control group of tEBC. In order to obtain more information about spontaneous eye-blink (SB) characteristics of 13 monkeys during paired training, we continuously recorded the SB responses for 10-min before each delay or trace conditioned training session. Each daily training session consisted of 100 CS-US paired trials. The trials were separated by a variable intertrial interval of 15 s–35 s. To avoid the influence of SB on CR recording and analysis, every trial was started artificially during this period by typing a key of the computer keyboard when the recording system demonstrated a smooth and steady signal baseline. For the delay conditioning paradigm, the US terminated simultaneously with the offset of the CS (500 ms duration), and the inter-stimulus interval was 400 ms (Fig. 2A); for the trace conditioning paradigm, a stimulus trace interval of 500 ms was interposed between the CS (100 ms duration) termination and the US onset (Fig. 2B). For pseudo-paired training of both dEBC and tEBC paradigms, the CS was 500 ms or 100 ms, for dEBC and tEBC respectively, and the US was presented at a random interval between 1 s and 10 s after the CS onset. All experiments were performed during the light phase of the light/dark cycle. After the delay or trace conditioning training session, the monkeys received 2 days of extinction training. Extinction testing parameters were identical to acquisition training except the US was always omitted.
2.7. Statistical analysis All the experimental data were expressed as the mean ± standard error of the mean (SEM). The statistical significance was determined by a least significant difference (LSD) post hoc tests, following a two-way repeated measures analyses of variance (ANOVA) and a separate oneway repeated measures ANOVA using the SPSS for Windows package (v. 13.0). A value of P < 0.05 was considered to be statistically significant.
2.6. Behavioral data analysis
3. Result
For delay and trace conditioning trials, 2000-ms or 2400-ms time periods were analyzed respectively. All data presented in this paper are measurement of the right upper eyelid movements. The parameters of eye-blink responses were analyzed using custom software. For dEBC training, each CS-US paired trial was subdivided into three discontinuous analysis periods: (1) the “baseline” period, −1000–0 ms before
3.1. Spontaneous eye-blinks There was no difference between the PCP-treated and control group in the average frequency of spontaneous eye-blink across the 6 sessions. 171
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
3.2.2. CR peak latency Normally as learning occurs over the conditioning sessions, timing of the CR should shift to better anticipate the onset of the air puff; CRs that occur shortly after CS onset are less adaptive than those that occur just prior to US onset. The present results demonstrate that repeated exposure to PCP produces a sustained impairment on CR timing. This was confirmed by a two-way repeated measures ANOVA on the CR peak latency; there was a significant group effect [F (1, 4) = 10.867, P = 0.030] and session effect [F (5, 20) = 10.872, P < 0.001]. Furthermore, a separate one-way ANOVA revealed that the CR peak latency of control group was significantly higher than PCP-treated group on session 2, 4 and 6 [F (1, 4) = 8.235, P = 0.045; F (1, 4) = 15.376, P = 0.017; F (1, 4) = 9.175, P = 0.039] (Fig. 5B). 3.2.3. UR peak latency To investigate the prolonged effects of PCP on the passive eye-blink responses elicited by external stimuli, UR peak latency was analyzed for all paired-learning monkeys including PCP-treated and control animals in both dEBC and tEBC groups. A two-way repeated measures ANOVA on the UR peak latency of dEBC confirmed that there was no significant difference between the two groups; there was no significant group effect [F (1, 4) = 0.099, P = ns], no significant group-by-session interaction [F (5, 20) = 1.321, P = ns] (Fig. 5C). 3.3. Acquisition of trace eye-blink conditioning by the cynomolgus monkeys 3.3.1. Conditioned responses Fig. 6A shows that although both PCP-treated and control groups for tEBC paired-training showed tEBC learning as evidenced by increased CRs℅ across successive training sessions, subjects of PCP-treated group demonstrated significantly poorer conditioning performance relative to control group. This was confirmed by a two-way repeated measures ANOVA on the CR%; there were significant effects of group [F (1, 5) = 20.063, P = 0.007], significant effects of session [F (11, 55) = 23.547, P < 0.001], and significant effects of group-by-session interaction [F (11, 55) = 2.033, P = 0.046]. Furthermore, a separate one-way ANOVA revealed that the CR% of control group was significantly higher than that of PCP-treated group on session 5, 6, 10, 11 and 12 [F (1, 5) = 30.817, P = 0.003; F (1, 5) = 16.949, P = 0.009; F (1, 5) = 11.145, P = 0.021; F (1, 5) = 7.252, P = 0.043; F (1, 5) = 9.847, P = 0.026]. Fig. 6A also shows that no learning occurs during the trace pseudo-conditioning training, there were no significant difference between the PCP-treated and control group. The two-way repeated measures ANOVA revealed no significant group effect [F (1, 4) = 5.908, P = ns], no significant session effect [F (11, 44) = 0.636, P = ns] and no significant group-by-session interaction [F (11, 44) = 0.704, P = ns]. Furthermore, LSD post-hoc tests revealed that the CR % of trace conditioned groups were significantly higher than the CR% of the pseudo-conditioned groups (P < 0.001; Fig. 6A). Additionally, it has been noticed that after 2 days of extinction sessions, CR% values in tEBC paired learning groups decreased to 39.85% or 35.24%, in PCPtreated or control monkeys respectively (data not shown).
Fig. 3. Representative examples of various forms of eye-blink responses encountered during the course of this experiment. The top line represents the unconditioned stimulus (US) of EBC training paradigm. Panel A is an example of an unconditioned response (UR) alone that occurred after the US onset. Panel B to D are examples of conditioned response (CR) and URs. Panel B is a robust CR with longer latency and large amplitude. Panel C shows a short-latency, small-amplitude response. Panel D shows a CR with long latency and high amplitude, and it also shows a UR and an alpha response (this data has not been analyzed in this study).
The average SB frequency of the two groups reduced with the gradual adaptation to the training environment. This was confirmed by a twoway repeated measures ANOVA on the frequency of SB; there was no significant effects of group [F (1, 11) = 0.131, P = ns], no significant effects of group-by-session interaction [F (5, 55) = 0.290, P = ns], but there were significant effects of session [F (5, 55) = 7.046, P < 0.001] (Fig. 4A). There was also no difference between the two groups on the amplitude of SB. The two-way repeated measures ANOVA revealed no significant group effect [F (1, 11) = 2.355, P = ns], no significant session effect [F (5, 55) = 0.329, P = ns] and no significant group-bysession interaction [F (5, 55) = 0.874, P = ns] (Fig. 4B). 3.2. Acquisition of delay eye-blink conditioning by the cynomolgus monkeys 3.2.1. Conditioned responses The CR% increased as a function of sessions in both PCP-treated and control groups and there was no significant difference between the two groups (Fig. 5A). This result was confirmed by a two-way repeated measures ANOVA, there were no significant effects of group [F (1, 4) = 2.962, P = ns], but significant effects of session [F (5, 20) = 65.084, P < 0.001]. During the delay pseudo-conditioning training, there were no difference between the PCP-treated and control group. The two-way repeated measures ANOVA revealed no significant group effect [F (1, 4) = 4.099, P = ns], no significant session effect [F (5, 20) = 0.413, P = ns] and no significant group-by-session interaction [F (5, 20) = 1.441, P = ns]. Furthermore, LSD post-hoc tests revealed that the CR % of delay conditioned groups were significantly higher than the CR% of the pseudo-conditioned groups (P < 0.001; Fig. 5A). Additionally, we noticed that after 2 days of extinction sessions, CR% values in dEBC paired learning groups decreased to 34.67% or 32.33%, in PCP-treated or control monkeys respectively (data not shown).
3.3.2. CR peak latency As shown in Fig. 4, there was no difference between the PCP-treated and control group on CR peak latency. It was confirmed by a two-way repeated measures ANOVA on CR peak latency; there was no significant session effect [F (11, 55) = 0.540, P = ns], no significant group effect [F (1, 5) = 1.145, P = ns], and no group-by-session interaction [F (11, 55) = 1.033, P = ns] (Fig. 6B). 3.3.3. UR peak latency We also examined the UR peak latency across 12 days of tEBC training sessions (Fig. 6B). The UR peak latency didn’t change with the increase of training, but the UR peak latency of control group was significantly higher than that of PCP-treated group. It was confirmed by 172
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 4. The frequency and amplitude of spontaneous eye-blink (SB) responses. (A) The frequency of SB in PCP-treated (n = 6) and control (n = 7) monkeys in EBC paired learning groups across the 6 training sessions; (B) The amplitude of SB in PCP-treated (n = 6) and control (n = 7) monkeys in EBC paired learning groups across the 6 training sessions. Both frequency and amplitude of the SB are given as mean ± standard error (SEM). Error bars indicate the SEM.
training (sessions 1–3), schizophrenic monkeys and monkeys from the control group both demonstrated similar lower CR rates. Furthermore, there were no differences in the frequency and amplitude of spontaneous eye-blinks between the two groups. All animals exhibited high SB rates in early training sessions, possibly due to the stress reaction, but these decreased gradually. This observation suggests that there are no differences in motor responses between the two groups. Finally, UR characteristics of decreased peak latency in tEBC, indicating a high sensitivity to aversive stimuli, were not responsible for lower CR rates in schizophrenic monkeys. In theory, these characteristics should have facilitated effects on CR performance. Therefore, it is reasonable to infer that the differences in improving the CR rates of tEBC between the two groups are caused by discrepancies in the ability to associate CS with US. The control cynomolgus monkeys can associate the two different stimuli (tones and a puff of air to the eye) better and faster, and, thus, they can establish motor learning at a faster rate. The molecular mechanism underlying prolonged EBC deficits in cynomolgus monkeys after long term PCP exposure requires further study. As a nonspecific NMDA receptor antagonist, PCP blocks activity of NMDA receptors at postsynaptic membranes, leading to the release of extra glutamate at presynaptic membranes. Recently, the relationship between the impairment of cerebellar function and the incidence of schizophrenia has attracted much attention. The famous “cognitive dysmetria” hypothesis states that the cerebellum plays an important role in schizophrenia through the cortico-cerebellar-thalamic-cortical (CCTC) circuit. According to this hypothesis, impaired cerebellar function makes the sensory and motor processes unable to coordinate and integrate with the cognitive information and thus results in a wide range of schizophrenia symptoms. Our study demonstrated that the CR
a two-way repeated measures ANOVA on UR peak latency; there was no significant session effect [F (11, 55) = 0.238, P = ns], no group-bysession interaction [F (11, 55) = 1.551, P = ns], but there was a significant group effect [F (1, 5) = 38.212, P = 0.002]. Furthermore, a separate one-way ANOVA revealed that the UR peak latency of control group was significantly higher than that of experimental group on session 7–12 [F (1, 5) = 17.067, P = 0.009; F (1, 5) = 38.401, P = 0.002; F (1, 5) = 6.722, P = 0.049; F (1, 5) = 46.361, P = 0.001; F (1, 5) = 55.635, P = 0.001; F (1, 5) = 48.259, P = 0.001] (Fig. 6C).
4. Discussion In the initial portion of this study, we showed that long-term exposure to PCP 12–15 months before at low doses did not change the frequency and amplitude of spontaneous eye-blink response. In the second portion of this study, compared with the control group, the PCPtreated group also successfully established dEBC. However, the CR timing in dEBC of PCP-treated monkeys was significantly impaired, which manifested as shorter CR peak latency than in the control group. In the third portion of this study, the PCP-treated group and control group showed a gradual increase in CR% during the tEBC training, but the CR% of the PCP-treated group was significantly lower than that in the control group after 12 sessions of training. Compared with the control group, the PCP-treated group animals were more sensitive to the outside stimuli in tEBC since the UR peak latency of PCP-treated group was significantly lower than that in the control group. Prolonged EBC deficiencies in cynomolgus monkeys with schizophrenia induced by long-term exposure to PCP should not be caused by the impairment of motor or sensory function. In the early stage of the 173
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 5. Acquisition of the delay eye-blink conditioned response (CR) for the PCP-treated (n = 3) and control (n = 3) groups across 6 acquisition training sessions. (A) CR percentage, (B) CR peak latency, and (C) Unconditioned response (UR) peak latency are given as mean ± standard error (SEM). Error bars indicate the SEM. *P < 0.05 vs control.
associative motor learning after cessation of PCP treatment in cynomolgus monkeys, even though this learning requires minimal motivation and attention. This finding adds to the growing literature documenting cognitive deficits in schizophrenia. The long-lasting behavioral effects after repeated PCP treatment might be related with neurochemical and neuroanatomical changes within the prefrontal cortex (PFC) or cerebellum [22,36]. Non-human primates have advantages over other kinds of mammals in terms of brain structure and functional evolution. EBC tasks have seldom been applied previously in studies related to schizophrenia, and, to our knowledge, this study is the first attempt at establishing an EBC model in cynomolgus monkeys and to measure its alteration in cynomolgus monkey model of schizophrenia. Although there are some limitations in this study, such as a small sample size and lack of assessment of illness severity, etc., the present work provides important data with regard to the dysfunction of associative motor learning,
timing in dEBC of PCP-treated monkeys was significantly impaired than that of control animals. Given that dEBC provides a well-validated model to investigate cerebellar timing function due to its dependence upon the cerebellum, our finding indicates a potential cerebellar impairment in PCP modeling of schizophrenia. However, in our study, the performance deficits of tEBC in cynomolgus monkeys with schizophrenia may be associated with the dysfunction of the prefrontal cortex, because the tEBC task is dependent upon prefrontal cortex function, in addition to the cerebellum [25]. It has been reported that enduring frontal lobe dysfunction occurs after the long-term intake of PCP [22]. It has been noted that some behavioral deficits remained after repeated PCP treatment ceased, including impaired working memory in the T-maze/radial arm maze [26–28], and Morris water maze tasks [29], attention/latent learning [30–32], fear conditioning learning [33], and recognition in social discrimination [34,35], among other deficits. The present results illustrate an enduring deficit of EBC 174
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 6. Acquisition of the trace eye-blink conditioned response (CR) for the PCP-treated (n = 3) and control (n = 4) groups across 12 acquisition training sessions. (A) CR percentage, (B) CR peak latency, and (C) Unconditioned response (UR) peak latency are given as mean ± standard error (SEM). Error bars indicate the SEM. *P < 0.05 vs control.
which requires minimal motivation and attention in schizophrenia patients.
State Basic Research Development Program of China (973 program, No. 2014CB541600) and the National Natural Science Foundation of China (No. 81171249).
Author contributions References Jian-feng Sui designed the study. Jian-feng Sui and Bing Wu interpreted the results and wrote the paper. Hui-min Liu, Xuan Li and Guang-yan Wu performed and helped with data analysis. All authors discussed and commented on the manuscript.
[1] D.C. Javitt, S.R. Zukin, Recent advances in the phencyclidine model of schizophrenia, Am. J. Psychiatry 148 (10) (1991) 1301–1308. [2] A.S. Carlin, I. Grant, K.M. Adams, et al., Is phencyclidine (PCP) abuse associated with organic mental impairment? Am. J. Drug Alcohol Abuse 6 (3) (1979) 273–281. [3] J. Cosgrove, T.G. Newell, Recovery of neuropsychological functions during reduction in use of phencyclidine, J. Clin. Psychol. 47 (1) (1991) 159–169. [4] S. Castellani, P.M. Adams, Effects of dopaminergic drugs on phencyclidine-induced behavior in the rat, Neuropharmacology 20 (4) (1981) 371–374. [5] S. Levin, Frontal lobe dysfunctions in schizophrenia–I. Eye movement impairments, J. Psychiatr. Res. 18 (1) (1984) 27–55. [6] J.D. Jentsch, R.H. Roth, The neuropsychopharmacology of phencyclidine: from NMDA receptor hypofunction to the dopamine hypothesis of schizophrenia, Neuropsychopharmacology 20 (3) (1999) 201–225. [7] C. Carter, L. Robertson, T. Nordahl, et al., Spatial working memory deficits and their relationship to negative symptoms in unmedicated schizophrenia patients, Biol.
Conflict of interests The authors declare no competing financial interests. Acknowledgements We are grateful to Yuan-ye Ma for technical assistance and results analysis. This work was mainly supported by grants from the Major 175
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
rats, Neuroscience 240 (2013) 336–348. [24] J.D. Jentsch, A. Tran, D. Le, et al., Subchronic phencyclidine administration reduces mesoprefrontal dopamine utilization and impairs prefrontal cortical-dependent cognition in the rat, Neuropsychopharmacology 17 (2) (1997) 92–99. [25] G. Wu, G. Liu, H. Zhang, et al., Optogenetic stimulation of mPFC pyramidal neurons as a conditioned stimulus supports associative learning in rats, Sci. Rep. 5 (2015) 10065. [26] M.R. Stefani, B. Moghaddam, Effects of repeated treatment with amphetamine or phencyclidine on working memory in the rat, Behav. Brain Res. 134 (1–2) (2002) 267–274. [27] Z. Li, C.H. Kim, J. Ichikawa, et al., Effect of repeated administration of phencyclidine on spatial performance in an eight-arm radial maze with delay in rats and mice, Pharmacol. Biochem. Behav. 75 (2) (2003) 335–340. [28] J.P. Marquis, S. Goulet, F.Y. Dore, Schizophrenia-like syndrome inducing agent phencyclidine failed to impair memory for temporal order in rats, Neurobiol. Learn Mem. 80 (2) (2003) 158–167. [29] J.D. Andersen, B. Pouzet, Spatial memory deficits induced by perinatal treatment of rats with PCP and reversal effect of D-serine, Neuropsychopharmacology 29 (6) (2004) 1080–1090. [30] J.S. Rodefer, E.R. Murphy, M.G. Baxter, PDE10A inhibition reverses subchronic PCP-induced deficits in attentional set-shifting in rats, Eur. J. Neurosci. 21 (4) (2005) 1070–1076. [31] A. Deschenes, S. Goulet, F.Y. Dore, Rule shift under long-term PCP challenge in rats, Behav. Brain Res. 167 (1) (2006) 134–140. [32] P.J. Fletcher, C.C. Tenn, Z. Rizos, et al., Sensitization to amphetamine, but not PCP, impairs attentional set shifting: reversal by a D1 receptor agonist injected into the medial prefrontal cortex, Psychopharmacology (Berl.) 183 (2) (2005) 190–200. [33] Z. Abdul-Monim, G.P. Reynolds, J.C. Neill, The effect of atypical and classical antipsychotics on sub-chronic PCP-induced cognitive deficits in a reversal-learning paradigm, Behav. Brain Res. 169 (2) (2006) 263–273. [34] S. Harich, G. Gross, A. Bespalov, Stimulation of the metabotropic glutamate 2/3 receptor attenuates social novelty discrimination deficits induced by neonatal phencyclidine treatment, Psychopharmacology (Berl.) 192 (4) (2007) 511–519. [35] R. Depoortere, G. Dargazanli, G. Estenne-Bouhtou, et al., Neurochemical, electrophysiological and pharmacological profiles of the selective inhibitor of the glycine transporter-1 SSR504734, a potential new type of antipsychotic, Neuropsychopharmacology 30 (11) (2005) 1963–1985. [36] T. Enomoto, Y. Noda, A. Mouri, et al., Long-lasting impairment of associative learning is correlated with a dysfunction of N-methyl-D-aspartate-extracellular signaling-regulated kinase signaling in mice after withdrawal from repeated administration of phencyclidine, Mol. Pharmacol. 68 (6) (2005) 1765–1774.
Psychiatry 10 (9) (1996) 930–932. [8] M. Hertzmann, R.C. Reba, E.V. Kotlyarov, Single photon emission computed tomography in phencyclidine and related drug abuse, Am. J. Psychiatry 147 (2) (1990) 255–256. [9] J.C. Wu, M.S. Buchsbaum, W.E. Bunney, Positron emission tomography study of phencyclidine users as a possible drug model of schizophrenia, Yakubutsu. Seishin. Kodo. 11 (1) (1991) 47–48. [10] T.J. Carew, C.L. Sahley, Invertebrate learning and memory: from behavior to molecules, Annu. Rev. Neurosci. 9 (1986) 435–487. [11] I. Gormezano, N. Schneiderman, E. Deaux, et al., Nictitating membrane: classical conditioning and extinction in the albino rabbit, Science 138 (3536) (1962) 33–34. [12] G. Weidemann, E. Best, J.C. Lee, et al., The role of contingency awareness in singlecue human eyeblink conditioning, Learn. Mem. 20 (7) (2013) 363–366. [13] R.E. Clark, Classical conditioning and brain systems: the role of awareness, Science 280 (5360) (1998) 77–81. [14] H. Huang, B. Wu, Q. Li, et al., Awareness is essential for differential delay eyeblink conditioning with soft-tone but not loud-tone conditioned stimuli, Neurosci. Bull. 30 (3) (2014) 433–440. [15] A.B. Steinmetz, T.C. Harmon, J.H. Freeman, Visual cortical contributions to associative cerebellar learning, Neurobiol. Learn. Mem. 104 (2013) 103–109. [16] D.H. Lindquist, Hippocampal-dependent Pavlovian conditioning in adult rats exposed to binge-like doses of ethanol as neonates, Behav. Brain Res. 242 (2013) 191–199. [17] G.Y. Wu, J. Yao, B. Hu, et al., Reevaluating the role of the hippocampus in delay eyeblink conditioning, PLoS One 8 (8) (2013) e71249. [18] G.Y. Wu, J. Yao, L.Q. Zhang, et al., Reevaluating the role of the medial prefrontal cortex in delay eyeblink conditioning, Neurobiol. Learn. Mem. 97 (3) (2012) 277–288. [19] B.C. Reeb-Sutherland, N.A. Fox Eyeblink Conditioning, A non-invasive biomarker for neurodevelopmental disorders, J. Autism Dev. Disord. 45 (2) (2015) 376–394. [20] A.R. Bolbecker, J.S. Kent, I.T. Petersen, et al., Impaired cerebellar-dependent eyeblink conditioning in first-degree relatives of individuals with schizophrenia, Schizophr. Bull. 40 (5) (2014) 1001–1010. [21] H. Picard, I. Amado, S. Mouchet-Mages, et al., The role of the cerebellum in schizophrenia: an update of clinical, cognitive, and functional evidences, Schizophr. Bull. 34 (1) (2007) 155–172. [22] J.D. Jentsch, D.J. Redmond, J.D. Elsworth, et al., Enduring cognitive deficits and cortical dopamine dysfunction in monkeys after long-term administration of phencyclidine, Science 277 (5328) (1997) 953–955. [23] T. Katayama, M. Okamoto, Y. Suzuki, et al., Phencyclidine affects firing activity of ventral tegmental area neurons that are related to reward and social behaviors in
176
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Prolonged deficits of associative motor learning in cynomolgus monkeys after long-term administration of phencyclidine
MARK
Bing Wua,b, Xu-dong Zhaoc, Hui-min Zhanga,b, Xuan Lia,b, Guang-yan Wua,b, Ying-shan Yangd, ⁎ Chao-yang Tiand, Jian-feng Suia,b, a
Department of Physiology, College of Basic Medical Sciences, Third Military Medical University, Chongqing 400038, China Experimental Center of Basic Medicine, College of Basic Medical Sciences, Third Military Medical University, Chongqing 400038, China c State Key Laboratory of Brain and Cognitive Sciences, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China d Hainan Jingang Biological Technology Co., Ltd., Haikou, Hainan 571100, China b
A R T I C L E I N F O
A B S T R A C T
Keywords: Phencyclidine (PCP) Eye-blink conditioning (EBC) Schizophrenia Cerebellum Learning
Phencyclidine (PCP) is a potent drug of abuse that induces sustained schizophrenia-like symptoms in humans by blocking neurotransmission at N-methyl-D-aspartate (NMDA)-type glutamate receptors. Alterations in NMDA receptor function have been linked to numerous behavioral deficits and cognitive dysfunction. Classical eyeblink conditioning (EBC), including delay (dEBC) and trace (tEBC) paradigms, provides an effective means to study the neurobiology of associative motor learning in rodents, mammals and primates. To assess whether administration of low-dosage PCP for extended periods has prolonged effect to alter associative motor learning, in this study 19 adult cynomolgus monkeys were administered PCP (0.3 mg/kg, intramuscularly) or saline twice a day for 14 days. Twelve–fifteen months after PCP or saline injection, monkeys received dEBC, tEBC, or pseudopaired training for 6 or 12 successive daily sessions, respectively. The results of this study show that percentage of conditioned response (CR) in dEBC increased as a function of training sessions in both PCP-treated and control monkeys and there was no significant CR% difference between the two groups. However, the CR timing in dEBC of PCP-treated monkeys was significantly impaired, as manifested by shorter CR peak latencies than those of the control group. PCP-treated animals showed significantly lower percentage of CR in tEBC compared to controls. PCP-treated animals were also more sensitive to outside stimuli in tEBC because the UR peak latency of PCPtreated group was significantly lower than the control group. These results indicated that cynomolgus monkeys manifested prolonged deficits in associative motor learning after long-term administration of phencyclidine.
1. Introduction Phencyclidine (PCP) is a psychotomimetic drug and a noncompetitive antagonist of the N-methyl-D-aspartate (NMDA) glutamate receptor. PCP-induced psychosis is associated with both positive and negative schizophrenia-like symptoms in humans [1]. Repeated administration can result in a long-lasting syndrome marked by neuropsychological deficits, social withdrawal, and affective blunting as well as hallucinations, paranoia, and delusions [2,3]. PCP also induces cognitive and behavioral dysfunctions that partially correspond to the positive and negative symptoms of schizophrenia [4–6]. Abnormalities in working memory, behavioral inhibition, and social interactions have been observed in schizophrenic individuals and in PCP-abusing humans [7–9]. Because the long-term abuse of PCP in humans may represent a pharmacological model of learning and memory deficits that are associated with schizophrenia, we explored the prolonged associative ⁎
learning effects of long-term exposure to PCP in the cynomolgus monkey (Macaca fascicularis). Classical eye-blink conditioning (EBC) has been extensively used to study the neurobiology of associative motor learning in rodents, mammals and primates [10,11]. EBC involves paired presentations of a tone or light as a conditioned stimulus (CS) and a periorbital shock or air puff as an unconditioned stimulus (US). Initially, the CS produces no obvious eye-blink responses while the US elicits reliable eye-blink responses before learning, named the unconditioned response (UR). After repeated pairings of the CS and US, the CS comes to elicit an eyeblink called the conditioned response (CR). The two widely employed paradigms of EBC are delay EBC (dEBC) and trace EBC (tEBC). In the dEBC, CS is presented before the US, and the two stimuli are coterminated. In tEBC, CS and US are presented separately in time so that a complete stimulus-free period exists between the two stimuli. EBC has been studied for more than half a century in humans [12–14] as well as
Corresponding author at: Department of Physiology, College of Basic Medical Sciences, Third Military Medical University, Chongqing 400038, China. E-mail address: [email protected] (J.-f. Sui).
http://dx.doi.org/10.1016/j.bbr.2017.05.035 Received 2 March 2017; Received in revised form 5 May 2017; Accepted 10 May 2017 Available online 23 May 2017 0166-4328/ © 2017 Elsevier B.V. All rights reserved.
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
in animals [11,15,16]. These studies led to a consensus that dEBC is dependent upon the cerebellum-brainstem circuits and that tEBC involves the hippocampus, prefrontal cortex and other brain areas in addition to cerebellum-brainstem circuits. While dEBC provides a wellvalidated method to investigate cerebellar timing function, tEBC seems to be a sensitive task related to functions of the prefrontal cortex and hippocampus [17,18]. As an ideal model for associative motor learning studies, changes of EBC in schizophrenia has received increasing attention in recent years [19–21]. Repeated exposure to PCP reportedly produces a sustained decrease in dopamine turnover within the prefrontal cortex, which is accompanied by a deficit in working memory and prefrontal cortex-dependent tasks in both rats and non-human primates [22–24]. In this study, we treated cynomolgus monkeys with PCP (0.3 mg/kg twice a day, intramuscularly) for 14 days, and 12–15 months later, we estimated the effect of PCP on establishment of both dEBC and tEBC. The goals of this study were (1) to assess whether administration of low-dosage PCP for extended periods has prolonged effect to alter associative learning ability in cynomolgus monkeys, and (2) to explore whether, in addition to damage to the frontal lobe and prefrontal cortex, extended PCP administration leads to impaired cerebellar function, which is involved in behavioral timing.
Fig. 1. Monkey conditioning goggles and primate chair. The goggles include the eye-blink detector and air-puff delivery pipe. A monkey wearing the conditioning goggles was sat in the primate chair with its legs and arms fixed. The right eye is monitored for eye-blinks, and the left eye is unobstructed.
(Fig. 1).
2. Materials and methods
2.4. Training and recording system
2.1. Subjects
During behavioral training, the CS was a 1-kHz, 80-dB pure tone. All sound stimuli were delivered by two speakers placed 60 cm to the left and right ears of the animals, respectively. A sound-level meter (type 2240, Brüel & Kjær) was used to calibrate the intensity of the CS tone. The US was a 100-ms, 5-psi (measured at the tip of a plastic pipe attached to goggles) source pressure air-puff delivered to the right cornea through a plastic pipe. Presentations of the CS and US were controlled by a self-made computer system. All signals including markers of the applied stimuli and signals from the infrared sensor were digitized by a data-acquisition system (RM6240BDJ, Cheng Yi, Chengdu, China) and were acquired and stored by the system software (v. 4.7). Data analysis was carried out on a dedicated Windows PC.
Nineteen healthy adult cynomolgus monkeys (9 females, 10 males) from the breeding colonies at the Hainan Jingang biological technology Co., Ltd. were served as subjects. The monkeys weighed between 3.5 and 6.5 kg at the beginning of behavioral tests, had no obvious eye diseases, and were sensitive to sounds stimuli. The monkeys were individually housed under standard conditions (a 12-h light/dark cycle with light on from 07:00 to 19:00, humidity 60%, 21 ± 2 °C, 3 times/ day deliveries of food and 1 time/day delivery of fruit, water available ad libitum). The experimental procedures were approved by the Animal Care Committee of the Third Military Medical University and were performed in accordance with the principles outlined in the NIH Guide for the Care and Use of Laboratory Animals.
2.5. Experimental procedures 2.2. Primate chair 2.5.1. PCP or Saline treatment A total of 19 adult cynomolgus monkeys were included in the experiment, including dEBC paired learning (n = 6); tEBC paired learning (n = 7); and pseudo-paired training for both dEBC and tEBC (n = 6). Monkeys were treated with daily intramuscularly injection of PCP (Chemsky [Shanghai] international Co., LTD.) at the dosage of 0.3 mg/kg (for PCP-treated group), or equal volume of saline (for control group), twice a day for 14 days [22]. Twelve–Fifteen months after injection of the PCP or Saline, monkeys in different groups received dEBC, tEBC, or pseudo-paired behavioral training, respectively.
A special primate chair was designed according to the monkey's physical characteristics to ensure that the monkey can sit in the chair comfortably with its limbs easy to be fixed. The design of the chair prevented monkeys from touching their heads with their hands or feet. Once a monkey was secured in the chair, the circular bayonet on the top section of the chair can limit the up and down movement of its head and body, the head could only rotate around. 2.3. Eye-blink Detection Goggles To detect monkeys’ eye-blink behavior, a specially designed pair of goggles was constructed and worn by the monkey during behavior training. Eye-blinks were measured with an infrared sensor consisting of an infrared emitter (FBCB30, HengSheng, Shenzhen, China) and an infrared detector (TBBB30, HengSheng, Shenzhen, China) encased side by side and aligned with converging optical axes. The converging optical axes were aligned for maximum sensitivity to detect eye-blinks at a distance of 0.5-cm from the infrared sensor surface to the pupil surface. The infrared sensor and a plastic pipe with 0.2-cm diameter were established and secured together with dental acrylic on the goggle’s right lens. The goggles were held in place by a rubber strap that attached to the lateral edges of the goggles and around the back of the subject’s head (under the ears). The rubber strap looseness was adjusted to ensure that animals worn goggles comfortably and firmly
2.5.2. Habituation All subjects were first habituated to sitting in the primate chair and wearing eye-blink detector goggles for 2 h each day for 6 days. No stimulation was applied to the monkeys during the habituation period. 2.5.3. Behavioral training Six monkeys (PCP-treated: n = 3; control: n = 3) included in dEBC paired learning underwent delay eyeb-link conditioned training for 6 successive daily sessions after habituation; 7 monkeys (PCP-treated: n = 3; control: n = 4) included in tEBC paired learning underwent trace eye-blink conditioned training for 12 successive daily sessions. In addition, in order to control the possible non-associative learning, 6 monkeys (PCP-treated: n = 3; control: n = 3) were included in pseudo170
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 2. Experimental design and temporal relationship between the conditioned stimulus (CS), unconditioned stimulus (US), and analysis periods for the eye-blink conditioning (EBC). (A) Schematic diagram showing the delay eye-blink conditioning paradigm used in the present study. The CS was presented 400 ms before the US onset and remained on while the 100-ms air-puff US was delivered and they terminated together. In each trial, we analyzed the parameters of the conditioned eye-blink response (CR; 100–400 ms after CS onset) and the unconditioned eye-blink response (UR; 400–700 ms after CS onset); (B) Schematic diagram showing the trace eye-blink conditioning paradigm used in the present study. In this paradigm, the CS (100 ms, tone) was paired with the US (100 ms, air-puff) and the CS was always presented before the onset of the US, a temporal gap (500 ms) occurs between the offset of the CS and the onset of the US. In each trial, we analyzed the parameters of the conditioned eye-blink response (CR; 200–600 ms after CS onset) and the unconditioned eye-blink response (UR; 600–1000 ms after CS onset).
the CS onset; (2) a “CR” period, 100–400 ms after the CS onset; and (3) a “UR” period, 400–700 ms after the CS onset (Fig. 2A). For tEBC training, each CS-US paired trial was also subdivided into three discontinuous analysis periods: (1) the “baseline” period, −1400–0 ms before the CS onset; (2) a “CR” period, 200–600 ms after the CS onset; and (3) a “UR” period, 600–1000 ms after the CS onset (Fig. 2B). A significant eye-blink movement was defined as an increase in amplitude that was greater than the mean baseline amplitude, plus four times the standard deviation of the baseline activity. In addition, the significant eye-blink movement required a minimal duration of 30 ms. The significant eye-blink movement during the “SB”, “CR” and “UR” period was counted as a SB, a CR or a UR, respectively (Fig. 3). The percentage of CR (CR%) was defined as the ratio of the number of trials that contained the CR to the total number of trials. The CR peak latency was defined as the time interval from the CS onset to the peak of the CR. Similarly, the UR peak latency was defined as the time interval from the US onset to the peak of the UR.
paired training to serve as control of paired learning of both dEBC and tEBC. They received the delay pseudo-conditioned training when served as pseudo-paired control group of dEBC, and the trace pseudo-paired training when served as pseudo-paired control group of tEBC. In order to obtain more information about spontaneous eye-blink (SB) characteristics of 13 monkeys during paired training, we continuously recorded the SB responses for 10-min before each delay or trace conditioned training session. Each daily training session consisted of 100 CS-US paired trials. The trials were separated by a variable intertrial interval of 15 s–35 s. To avoid the influence of SB on CR recording and analysis, every trial was started artificially during this period by typing a key of the computer keyboard when the recording system demonstrated a smooth and steady signal baseline. For the delay conditioning paradigm, the US terminated simultaneously with the offset of the CS (500 ms duration), and the inter-stimulus interval was 400 ms (Fig. 2A); for the trace conditioning paradigm, a stimulus trace interval of 500 ms was interposed between the CS (100 ms duration) termination and the US onset (Fig. 2B). For pseudo-paired training of both dEBC and tEBC paradigms, the CS was 500 ms or 100 ms, for dEBC and tEBC respectively, and the US was presented at a random interval between 1 s and 10 s after the CS onset. All experiments were performed during the light phase of the light/dark cycle. After the delay or trace conditioning training session, the monkeys received 2 days of extinction training. Extinction testing parameters were identical to acquisition training except the US was always omitted.
2.7. Statistical analysis All the experimental data were expressed as the mean ± standard error of the mean (SEM). The statistical significance was determined by a least significant difference (LSD) post hoc tests, following a two-way repeated measures analyses of variance (ANOVA) and a separate oneway repeated measures ANOVA using the SPSS for Windows package (v. 13.0). A value of P < 0.05 was considered to be statistically significant.
2.6. Behavioral data analysis
3. Result
For delay and trace conditioning trials, 2000-ms or 2400-ms time periods were analyzed respectively. All data presented in this paper are measurement of the right upper eyelid movements. The parameters of eye-blink responses were analyzed using custom software. For dEBC training, each CS-US paired trial was subdivided into three discontinuous analysis periods: (1) the “baseline” period, −1000–0 ms before
3.1. Spontaneous eye-blinks There was no difference between the PCP-treated and control group in the average frequency of spontaneous eye-blink across the 6 sessions. 171
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
3.2.2. CR peak latency Normally as learning occurs over the conditioning sessions, timing of the CR should shift to better anticipate the onset of the air puff; CRs that occur shortly after CS onset are less adaptive than those that occur just prior to US onset. The present results demonstrate that repeated exposure to PCP produces a sustained impairment on CR timing. This was confirmed by a two-way repeated measures ANOVA on the CR peak latency; there was a significant group effect [F (1, 4) = 10.867, P = 0.030] and session effect [F (5, 20) = 10.872, P < 0.001]. Furthermore, a separate one-way ANOVA revealed that the CR peak latency of control group was significantly higher than PCP-treated group on session 2, 4 and 6 [F (1, 4) = 8.235, P = 0.045; F (1, 4) = 15.376, P = 0.017; F (1, 4) = 9.175, P = 0.039] (Fig. 5B). 3.2.3. UR peak latency To investigate the prolonged effects of PCP on the passive eye-blink responses elicited by external stimuli, UR peak latency was analyzed for all paired-learning monkeys including PCP-treated and control animals in both dEBC and tEBC groups. A two-way repeated measures ANOVA on the UR peak latency of dEBC confirmed that there was no significant difference between the two groups; there was no significant group effect [F (1, 4) = 0.099, P = ns], no significant group-by-session interaction [F (5, 20) = 1.321, P = ns] (Fig. 5C). 3.3. Acquisition of trace eye-blink conditioning by the cynomolgus monkeys 3.3.1. Conditioned responses Fig. 6A shows that although both PCP-treated and control groups for tEBC paired-training showed tEBC learning as evidenced by increased CRs℅ across successive training sessions, subjects of PCP-treated group demonstrated significantly poorer conditioning performance relative to control group. This was confirmed by a two-way repeated measures ANOVA on the CR%; there were significant effects of group [F (1, 5) = 20.063, P = 0.007], significant effects of session [F (11, 55) = 23.547, P < 0.001], and significant effects of group-by-session interaction [F (11, 55) = 2.033, P = 0.046]. Furthermore, a separate one-way ANOVA revealed that the CR% of control group was significantly higher than that of PCP-treated group on session 5, 6, 10, 11 and 12 [F (1, 5) = 30.817, P = 0.003; F (1, 5) = 16.949, P = 0.009; F (1, 5) = 11.145, P = 0.021; F (1, 5) = 7.252, P = 0.043; F (1, 5) = 9.847, P = 0.026]. Fig. 6A also shows that no learning occurs during the trace pseudo-conditioning training, there were no significant difference between the PCP-treated and control group. The two-way repeated measures ANOVA revealed no significant group effect [F (1, 4) = 5.908, P = ns], no significant session effect [F (11, 44) = 0.636, P = ns] and no significant group-by-session interaction [F (11, 44) = 0.704, P = ns]. Furthermore, LSD post-hoc tests revealed that the CR % of trace conditioned groups were significantly higher than the CR% of the pseudo-conditioned groups (P < 0.001; Fig. 6A). Additionally, it has been noticed that after 2 days of extinction sessions, CR% values in tEBC paired learning groups decreased to 39.85% or 35.24%, in PCPtreated or control monkeys respectively (data not shown).
Fig. 3. Representative examples of various forms of eye-blink responses encountered during the course of this experiment. The top line represents the unconditioned stimulus (US) of EBC training paradigm. Panel A is an example of an unconditioned response (UR) alone that occurred after the US onset. Panel B to D are examples of conditioned response (CR) and URs. Panel B is a robust CR with longer latency and large amplitude. Panel C shows a short-latency, small-amplitude response. Panel D shows a CR with long latency and high amplitude, and it also shows a UR and an alpha response (this data has not been analyzed in this study).
The average SB frequency of the two groups reduced with the gradual adaptation to the training environment. This was confirmed by a twoway repeated measures ANOVA on the frequency of SB; there was no significant effects of group [F (1, 11) = 0.131, P = ns], no significant effects of group-by-session interaction [F (5, 55) = 0.290, P = ns], but there were significant effects of session [F (5, 55) = 7.046, P < 0.001] (Fig. 4A). There was also no difference between the two groups on the amplitude of SB. The two-way repeated measures ANOVA revealed no significant group effect [F (1, 11) = 2.355, P = ns], no significant session effect [F (5, 55) = 0.329, P = ns] and no significant group-bysession interaction [F (5, 55) = 0.874, P = ns] (Fig. 4B). 3.2. Acquisition of delay eye-blink conditioning by the cynomolgus monkeys 3.2.1. Conditioned responses The CR% increased as a function of sessions in both PCP-treated and control groups and there was no significant difference between the two groups (Fig. 5A). This result was confirmed by a two-way repeated measures ANOVA, there were no significant effects of group [F (1, 4) = 2.962, P = ns], but significant effects of session [F (5, 20) = 65.084, P < 0.001]. During the delay pseudo-conditioning training, there were no difference between the PCP-treated and control group. The two-way repeated measures ANOVA revealed no significant group effect [F (1, 4) = 4.099, P = ns], no significant session effect [F (5, 20) = 0.413, P = ns] and no significant group-by-session interaction [F (5, 20) = 1.441, P = ns]. Furthermore, LSD post-hoc tests revealed that the CR % of delay conditioned groups were significantly higher than the CR% of the pseudo-conditioned groups (P < 0.001; Fig. 5A). Additionally, we noticed that after 2 days of extinction sessions, CR% values in dEBC paired learning groups decreased to 34.67% or 32.33%, in PCP-treated or control monkeys respectively (data not shown).
3.3.2. CR peak latency As shown in Fig. 4, there was no difference between the PCP-treated and control group on CR peak latency. It was confirmed by a two-way repeated measures ANOVA on CR peak latency; there was no significant session effect [F (11, 55) = 0.540, P = ns], no significant group effect [F (1, 5) = 1.145, P = ns], and no group-by-session interaction [F (11, 55) = 1.033, P = ns] (Fig. 6B). 3.3.3. UR peak latency We also examined the UR peak latency across 12 days of tEBC training sessions (Fig. 6B). The UR peak latency didn’t change with the increase of training, but the UR peak latency of control group was significantly higher than that of PCP-treated group. It was confirmed by 172
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 4. The frequency and amplitude of spontaneous eye-blink (SB) responses. (A) The frequency of SB in PCP-treated (n = 6) and control (n = 7) monkeys in EBC paired learning groups across the 6 training sessions; (B) The amplitude of SB in PCP-treated (n = 6) and control (n = 7) monkeys in EBC paired learning groups across the 6 training sessions. Both frequency and amplitude of the SB are given as mean ± standard error (SEM). Error bars indicate the SEM.
training (sessions 1–3), schizophrenic monkeys and monkeys from the control group both demonstrated similar lower CR rates. Furthermore, there were no differences in the frequency and amplitude of spontaneous eye-blinks between the two groups. All animals exhibited high SB rates in early training sessions, possibly due to the stress reaction, but these decreased gradually. This observation suggests that there are no differences in motor responses between the two groups. Finally, UR characteristics of decreased peak latency in tEBC, indicating a high sensitivity to aversive stimuli, were not responsible for lower CR rates in schizophrenic monkeys. In theory, these characteristics should have facilitated effects on CR performance. Therefore, it is reasonable to infer that the differences in improving the CR rates of tEBC between the two groups are caused by discrepancies in the ability to associate CS with US. The control cynomolgus monkeys can associate the two different stimuli (tones and a puff of air to the eye) better and faster, and, thus, they can establish motor learning at a faster rate. The molecular mechanism underlying prolonged EBC deficits in cynomolgus monkeys after long term PCP exposure requires further study. As a nonspecific NMDA receptor antagonist, PCP blocks activity of NMDA receptors at postsynaptic membranes, leading to the release of extra glutamate at presynaptic membranes. Recently, the relationship between the impairment of cerebellar function and the incidence of schizophrenia has attracted much attention. The famous “cognitive dysmetria” hypothesis states that the cerebellum plays an important role in schizophrenia through the cortico-cerebellar-thalamic-cortical (CCTC) circuit. According to this hypothesis, impaired cerebellar function makes the sensory and motor processes unable to coordinate and integrate with the cognitive information and thus results in a wide range of schizophrenia symptoms. Our study demonstrated that the CR
a two-way repeated measures ANOVA on UR peak latency; there was no significant session effect [F (11, 55) = 0.238, P = ns], no group-bysession interaction [F (11, 55) = 1.551, P = ns], but there was a significant group effect [F (1, 5) = 38.212, P = 0.002]. Furthermore, a separate one-way ANOVA revealed that the UR peak latency of control group was significantly higher than that of experimental group on session 7–12 [F (1, 5) = 17.067, P = 0.009; F (1, 5) = 38.401, P = 0.002; F (1, 5) = 6.722, P = 0.049; F (1, 5) = 46.361, P = 0.001; F (1, 5) = 55.635, P = 0.001; F (1, 5) = 48.259, P = 0.001] (Fig. 6C).
4. Discussion In the initial portion of this study, we showed that long-term exposure to PCP 12–15 months before at low doses did not change the frequency and amplitude of spontaneous eye-blink response. In the second portion of this study, compared with the control group, the PCPtreated group also successfully established dEBC. However, the CR timing in dEBC of PCP-treated monkeys was significantly impaired, which manifested as shorter CR peak latency than in the control group. In the third portion of this study, the PCP-treated group and control group showed a gradual increase in CR% during the tEBC training, but the CR% of the PCP-treated group was significantly lower than that in the control group after 12 sessions of training. Compared with the control group, the PCP-treated group animals were more sensitive to the outside stimuli in tEBC since the UR peak latency of PCP-treated group was significantly lower than that in the control group. Prolonged EBC deficiencies in cynomolgus monkeys with schizophrenia induced by long-term exposure to PCP should not be caused by the impairment of motor or sensory function. In the early stage of the 173
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 5. Acquisition of the delay eye-blink conditioned response (CR) for the PCP-treated (n = 3) and control (n = 3) groups across 6 acquisition training sessions. (A) CR percentage, (B) CR peak latency, and (C) Unconditioned response (UR) peak latency are given as mean ± standard error (SEM). Error bars indicate the SEM. *P < 0.05 vs control.
associative motor learning after cessation of PCP treatment in cynomolgus monkeys, even though this learning requires minimal motivation and attention. This finding adds to the growing literature documenting cognitive deficits in schizophrenia. The long-lasting behavioral effects after repeated PCP treatment might be related with neurochemical and neuroanatomical changes within the prefrontal cortex (PFC) or cerebellum [22,36]. Non-human primates have advantages over other kinds of mammals in terms of brain structure and functional evolution. EBC tasks have seldom been applied previously in studies related to schizophrenia, and, to our knowledge, this study is the first attempt at establishing an EBC model in cynomolgus monkeys and to measure its alteration in cynomolgus monkey model of schizophrenia. Although there are some limitations in this study, such as a small sample size and lack of assessment of illness severity, etc., the present work provides important data with regard to the dysfunction of associative motor learning,
timing in dEBC of PCP-treated monkeys was significantly impaired than that of control animals. Given that dEBC provides a well-validated model to investigate cerebellar timing function due to its dependence upon the cerebellum, our finding indicates a potential cerebellar impairment in PCP modeling of schizophrenia. However, in our study, the performance deficits of tEBC in cynomolgus monkeys with schizophrenia may be associated with the dysfunction of the prefrontal cortex, because the tEBC task is dependent upon prefrontal cortex function, in addition to the cerebellum [25]. It has been reported that enduring frontal lobe dysfunction occurs after the long-term intake of PCP [22]. It has been noted that some behavioral deficits remained after repeated PCP treatment ceased, including impaired working memory in the T-maze/radial arm maze [26–28], and Morris water maze tasks [29], attention/latent learning [30–32], fear conditioning learning [33], and recognition in social discrimination [34,35], among other deficits. The present results illustrate an enduring deficit of EBC 174
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
Fig. 6. Acquisition of the trace eye-blink conditioned response (CR) for the PCP-treated (n = 3) and control (n = 4) groups across 12 acquisition training sessions. (A) CR percentage, (B) CR peak latency, and (C) Unconditioned response (UR) peak latency are given as mean ± standard error (SEM). Error bars indicate the SEM. *P < 0.05 vs control.
which requires minimal motivation and attention in schizophrenia patients.
State Basic Research Development Program of China (973 program, No. 2014CB541600) and the National Natural Science Foundation of China (No. 81171249).
Author contributions References Jian-feng Sui designed the study. Jian-feng Sui and Bing Wu interpreted the results and wrote the paper. Hui-min Liu, Xuan Li and Guang-yan Wu performed and helped with data analysis. All authors discussed and commented on the manuscript.
[1] D.C. Javitt, S.R. Zukin, Recent advances in the phencyclidine model of schizophrenia, Am. J. Psychiatry 148 (10) (1991) 1301–1308. [2] A.S. Carlin, I. Grant, K.M. Adams, et al., Is phencyclidine (PCP) abuse associated with organic mental impairment? Am. J. Drug Alcohol Abuse 6 (3) (1979) 273–281. [3] J. Cosgrove, T.G. Newell, Recovery of neuropsychological functions during reduction in use of phencyclidine, J. Clin. Psychol. 47 (1) (1991) 159–169. [4] S. Castellani, P.M. Adams, Effects of dopaminergic drugs on phencyclidine-induced behavior in the rat, Neuropharmacology 20 (4) (1981) 371–374. [5] S. Levin, Frontal lobe dysfunctions in schizophrenia–I. Eye movement impairments, J. Psychiatr. Res. 18 (1) (1984) 27–55. [6] J.D. Jentsch, R.H. Roth, The neuropsychopharmacology of phencyclidine: from NMDA receptor hypofunction to the dopamine hypothesis of schizophrenia, Neuropsychopharmacology 20 (3) (1999) 201–225. [7] C. Carter, L. Robertson, T. Nordahl, et al., Spatial working memory deficits and their relationship to negative symptoms in unmedicated schizophrenia patients, Biol.
Conflict of interests The authors declare no competing financial interests. Acknowledgements We are grateful to Yuan-ye Ma for technical assistance and results analysis. This work was mainly supported by grants from the Major 175
Behavioural Brain Research 331 (2017) 169–176
B. Wu et al.
rats, Neuroscience 240 (2013) 336–348. [24] J.D. Jentsch, A. Tran, D. Le, et al., Subchronic phencyclidine administration reduces mesoprefrontal dopamine utilization and impairs prefrontal cortical-dependent cognition in the rat, Neuropsychopharmacology 17 (2) (1997) 92–99. [25] G. Wu, G. Liu, H. Zhang, et al., Optogenetic stimulation of mPFC pyramidal neurons as a conditioned stimulus supports associative learning in rats, Sci. Rep. 5 (2015) 10065. [26] M.R. Stefani, B. Moghaddam, Effects of repeated treatment with amphetamine or phencyclidine on working memory in the rat, Behav. Brain Res. 134 (1–2) (2002) 267–274. [27] Z. Li, C.H. Kim, J. Ichikawa, et al., Effect of repeated administration of phencyclidine on spatial performance in an eight-arm radial maze with delay in rats and mice, Pharmacol. Biochem. Behav. 75 (2) (2003) 335–340. [28] J.P. Marquis, S. Goulet, F.Y. Dore, Schizophrenia-like syndrome inducing agent phencyclidine failed to impair memory for temporal order in rats, Neurobiol. Learn Mem. 80 (2) (2003) 158–167. [29] J.D. Andersen, B. Pouzet, Spatial memory deficits induced by perinatal treatment of rats with PCP and reversal effect of D-serine, Neuropsychopharmacology 29 (6) (2004) 1080–1090. [30] J.S. Rodefer, E.R. Murphy, M.G. Baxter, PDE10A inhibition reverses subchronic PCP-induced deficits in attentional set-shifting in rats, Eur. J. Neurosci. 21 (4) (2005) 1070–1076. [31] A. Deschenes, S. Goulet, F.Y. Dore, Rule shift under long-term PCP challenge in rats, Behav. Brain Res. 167 (1) (2006) 134–140. [32] P.J. Fletcher, C.C. Tenn, Z. Rizos, et al., Sensitization to amphetamine, but not PCP, impairs attentional set shifting: reversal by a D1 receptor agonist injected into the medial prefrontal cortex, Psychopharmacology (Berl.) 183 (2) (2005) 190–200. [33] Z. Abdul-Monim, G.P. Reynolds, J.C. Neill, The effect of atypical and classical antipsychotics on sub-chronic PCP-induced cognitive deficits in a reversal-learning paradigm, Behav. Brain Res. 169 (2) (2006) 263–273. [34] S. Harich, G. Gross, A. Bespalov, Stimulation of the metabotropic glutamate 2/3 receptor attenuates social novelty discrimination deficits induced by neonatal phencyclidine treatment, Psychopharmacology (Berl.) 192 (4) (2007) 511–519. [35] R. Depoortere, G. Dargazanli, G. Estenne-Bouhtou, et al., Neurochemical, electrophysiological and pharmacological profiles of the selective inhibitor of the glycine transporter-1 SSR504734, a potential new type of antipsychotic, Neuropsychopharmacology 30 (11) (2005) 1963–1985. [36] T. Enomoto, Y. Noda, A. Mouri, et al., Long-lasting impairment of associative learning is correlated with a dysfunction of N-methyl-D-aspartate-extracellular signaling-regulated kinase signaling in mice after withdrawal from repeated administration of phencyclidine, Mol. Pharmacol. 68 (6) (2005) 1765–1774.
Psychiatry 10 (9) (1996) 930–932. [8] M. Hertzmann, R.C. Reba, E.V. Kotlyarov, Single photon emission computed tomography in phencyclidine and related drug abuse, Am. J. Psychiatry 147 (2) (1990) 255–256. [9] J.C. Wu, M.S. Buchsbaum, W.E. Bunney, Positron emission tomography study of phencyclidine users as a possible drug model of schizophrenia, Yakubutsu. Seishin. Kodo. 11 (1) (1991) 47–48. [10] T.J. Carew, C.L. Sahley, Invertebrate learning and memory: from behavior to molecules, Annu. Rev. Neurosci. 9 (1986) 435–487. [11] I. Gormezano, N. Schneiderman, E. Deaux, et al., Nictitating membrane: classical conditioning and extinction in the albino rabbit, Science 138 (3536) (1962) 33–34. [12] G. Weidemann, E. Best, J.C. Lee, et al., The role of contingency awareness in singlecue human eyeblink conditioning, Learn. Mem. 20 (7) (2013) 363–366. [13] R.E. Clark, Classical conditioning and brain systems: the role of awareness, Science 280 (5360) (1998) 77–81. [14] H. Huang, B. Wu, Q. Li, et al., Awareness is essential for differential delay eyeblink conditioning with soft-tone but not loud-tone conditioned stimuli, Neurosci. Bull. 30 (3) (2014) 433–440. [15] A.B. Steinmetz, T.C. Harmon, J.H. Freeman, Visual cortical contributions to associative cerebellar learning, Neurobiol. Learn. Mem. 104 (2013) 103–109. [16] D.H. Lindquist, Hippocampal-dependent Pavlovian conditioning in adult rats exposed to binge-like doses of ethanol as neonates, Behav. Brain Res. 242 (2013) 191–199. [17] G.Y. Wu, J. Yao, B. Hu, et al., Reevaluating the role of the hippocampus in delay eyeblink conditioning, PLoS One 8 (8) (2013) e71249. [18] G.Y. Wu, J. Yao, L.Q. Zhang, et al., Reevaluating the role of the medial prefrontal cortex in delay eyeblink conditioning, Neurobiol. Learn. Mem. 97 (3) (2012) 277–288. [19] B.C. Reeb-Sutherland, N.A. Fox Eyeblink Conditioning, A non-invasive biomarker for neurodevelopmental disorders, J. Autism Dev. Disord. 45 (2) (2015) 376–394. [20] A.R. Bolbecker, J.S. Kent, I.T. Petersen, et al., Impaired cerebellar-dependent eyeblink conditioning in first-degree relatives of individuals with schizophrenia, Schizophr. Bull. 40 (5) (2014) 1001–1010. [21] H. Picard, I. Amado, S. Mouchet-Mages, et al., The role of the cerebellum in schizophrenia: an update of clinical, cognitive, and functional evidences, Schizophr. Bull. 34 (1) (2007) 155–172. [22] J.D. Jentsch, D.J. Redmond, J.D. Elsworth, et al., Enduring cognitive deficits and cortical dopamine dysfunction in monkeys after long-term administration of phencyclidine, Science 277 (5328) (1997) 953–955. [23] T. Katayama, M. Okamoto, Y. Suzuki, et al., Phencyclidine affects firing activity of ventral tegmental area neurons that are related to reward and social behaviors in
176