European Journal of Pharmacology 734 (2014) 105–113

Contents lists available at ScienceDirect

European Journal of Pharmacology journal homepage: www.elsevier.com/locate/ejphar

Behavioural pharmacology

Septal serotonin depletion in rats facilitates working memory in the radial arm maze and increases hippocampal high-frequency theta activity Miguel Ángel López-Vázquez a,b, Elisa López-Loeza b, Naima Lajud Á vila c, Blanca Erika Gutiérrez-Guzmán d, J. Jesús Hernández-Pérez d, Yoana Estrada Reyes a, María Esther Olvera-Cortés d,n a

Laboratorio de Neuroplasticidad de los Procesos Cognitivos, Centro de Investigación Biomédica de Michoacán, Instituto Mexicano del Seguro Social, Mexico Laboratorio de Biofísica, Instituto de Física y Matemáticas, Universidad Michoacana de San Nicolás de Hidalgo, Mexico Laboratorio de Neuroendocrinología, Centro de Investigación Biomédica de Michoacán, Instituto Mexicano del Seguro Social, Mexico d Laboratorio de Neurofisiología Experimental, Centro de Investigación Biomédica de Michoacán, Instituto Mexicano del Seguro Social, Camino de la arboleda 300, Exhacienda de san José de la Huerta, Morelia, Mich C.P. 58341, Mexico b c

art ic l e i nf o

a b s t r a c t

Article history: Received 11 November 2013 Received in revised form 1 April 2014 Accepted 3 April 2014 Available online 15 April 2014

Hippocampal theta activity, which is strongly modulated by the septal medial/Broca's diagonal band neurons, has been linked to information processing of the hippocampus. Serotonin from the medial raphe nuclei desynchronises hippocampal theta activity, whereas inactivation or a lesion of this nucleus induces continuous and persistent theta activity in the hippocampus. Hippocampal serotonin depletion produces an increased expression of high-frequency theta activity concurrent with the facilitation of place learning in the Morris maze. The medial septum–diagonal band of Broca complex (MS/DBB) has been proposed as a key structure in the serotonin modulation of theta activity. We addressed whether serotonin depletion of the MS/DBB induces changes in the characteristics of hippocampal theta activity and whether the depletion is associated with learning in a working memory spatial task in the radial arm maze. Sprague Dawley rats were depleted of 5HT with the infusion of 5,7-dihydroxytriptamine (5, 7-DHT) in MS/DBB and were subsequently trained in the standard test (win-shift) in the radial arm, while the CA1 EEG activity was simultaneously recorded through telemetry. The MS/DBB serotonin depletion induced a low level of expression of low-frequency (4.5–6.5 Hz) and a higher expression of high-frequency (6.5–9.5 Hz) theta activity concomitant to a minor number of errors committed by rats on the working memory test. Thus, the depletion of serotonin in the MS/DBB caused a facilitator effect on working memory and a predominance of high-frequency theta activity. & 2014 Elsevier B.V. All rights reserved.

Keywords: Hippocampus Theta activity Serotonin Septum Working memory

1. Introduction Hippocampal theta activity is a sinusoidal-like slow oscillation that ranges from 4 to 12 Hz and is prominent in the rat hippocampus during voluntary behaviour (Vanderwolf, 1969; Whishaw and Vanderwolf, 1973) and during the display of behaviours related to the acquisition of environmental information (Eichenbaum et al., 1992; Vinogradova, 1995). The expression and characteristics of the hippocampal theta activity have been linked to place learning ability in humans and rats (Cornwell et al., 2008; Watrous et al., 2011; Olvera-Cortes et al., 2002, 2004). Conversely, a relationship has been observed between decreases in theta frequency and place

n

Corresponding author. Tel./fax: þ 52 44 33 22 26 00.

http://dx.doi.org/10.1016/j.ejphar.2014.04.005 0014-2999/& 2014 Elsevier B.V. All rights reserved.

learning impairments (Maho et al., 1988; Pan and McNaughton, 1997). The medial septum–diagonal band of Broca complex (MS/DBB) acts as a pacemaker for the hippocampal theta activity (Ford et al., 1989; Hangya et al., 2009; Petsche et al., 1962) as it sends projections to the hippocampus from cholinergic, GABAergic, and glutamatergic neurons (Amaral and Kurz, 1985; Colom et al., 2005; Freund, 1989), all of which appear to be important in the regulation of hippocampal theta activity (Hangya et al., 2009; Manns et al., 2002; Monmaur et al., 1993; Sotty et al., 2003). Lesions or inactivation of the medial septum disrupts hippocampal theta activity and causes learning deficits in both reference and working memory tasks (Givens and Olton, 1994; McNaughton et al., 2006; Mitchell et al., 1982; Mizumori et al., 1990; Rawlins et al., 1979). The septal area is an important structure in the control of hippocampal theta activity (Rawlins et al., 1979; Vertes and

106

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

Kocsis, 1997), and it is also crucial in the processing of information that is related to hippocampal function (Mitchell et al., 1982). While serotonin has been proposed as the only neurotransmitter that desynchronises hippocampal theta activity, the cerebral depletion of serotonin induces persistent and continuous theta activity in the hippocampus (Maru et al., 1979; Vertes, 1986; Vertes and Kocsis, 1997; Vinogradova et al., 1999), an effect that is mediated principally in the SM (Crooks et al., 2012). In view of the relevant role of theta oscillations in the processing of information by the hippocampus, it is reasonable to postulate that the regulation of theta activity by the 5-HT could assume a modulator role in the processing of information by the hippocampus, a premise that has received scant attention to date. Stimulation of the median raphe nucleus (MR), the main input of serotonin to the hippocampus, desynchronises the hippocampal EEG signal (Kitchigina et al., 1999; Maru et al., 1979; Vertes and Martin, 1988), and lesions of the MR result in the presence of persistent and continuous trains of theta activity (Maru et al., 1979; Vertes, 1986; Vinogradova et al., 1999). Septal neuronal activity is modulated by serotonin acting on cholinergic, GABAergic (Acsady et al., 1996; Jeltsch-David et al., 2008), and possibly glutamatergic neurons through different receptors, such as 5-HT1A and 5-HT2A (expressed on cholinergic and GABAergic septo-hippocampal neurons) (Luttgen et al., 2004, 2005). As serotonergic fibres target parvalbumin-expressing GABAergic neurons in MD/DBB, these cells synapse on hippocampal GABAergic basket and chandelier cells. Accordingly, it has been proposed that this arrangement can generate rhythmic disinhibition and produce a powerful inhibition of hippocampal interneurons (Freund, 1989; Freund and Antal, 1988; Leranth and Vertes, 1999). Serotonin also inhibits cholinergic and non-cholinergic neurons on MS (Alreja, 1996). Additionally, it has been proposed that the serotonergic fibres that innervate the medial septum could exert an inhibitory influence on the rhythmic firing of septal cells (Assaf and Miller, 1978). Serotonin also modulates learning and memory processes, although its role is unclear. In rats, enhanced learning in spatial discrimination has been reported after cerebral and hippocampal serotonin depletion (Altman et al., 1990; Normile et al., 1990), and improvement in working memory performance has been detected after prefrontal serotonin depletion (Perez-Vega et al., 2000). Recently, improvement in place learning concurrent with increased high-frequency theta activity during the search for the platform in the Morris maze was reported after hippocampal serotonin depletion (Gutiérrez-Guzmán et al., 2011), while deficient learning associated with a reduced expression of hippocampal high-frequency theta activity after serotonin depletion in the supramammillary nucleus and posterior hypothalamus (SUM/PH) was observed in rats (Gutiérrez-Guzmán et al., 2012). Thus, the effects of serotonin on theta activity may be strongly mediated through MS/DBB, which is the pacemaker of hippocampal theta activity and is related to learning and memory processes. The aim of this study was to evaluate the effect of serotonin MS/DBB reduction in the modulation of working memory and theta-related activity.

2. Materials and methods 2.1. Animals Forty Sprague Dawley male rats weighing between 350 and 450 g were used. The experiments were performed in accordance with the National Institute of Health guide regarding the care and use of laboratory animals (NIH Publications no. 80-23) and with the “Norma Oficial Mexicana para el uso de animales de

laboratorio” (NOM-062-ZOO-1999). Furthermore, the experiments were approved by the research ethics committee of the Instituto Mexicano del Seguro Social. All rats were maintained under standard housing conditions in the animal facility. The rats were assigned to one of two groups, the 5,7-DHT group (n ¼15) or the vehicle group (n ¼15). 2.2. Surgery Under pentobarbital anaesthesia (30 mg/kg ip), the 5,7-DHT rats received microinjections of 5,7-dihydroxytriptamine (5,7-DHT) into the MS (0.6 mm anterior from the bregma, 1.5 mm right of the midline, 151 from the vertical and 6.8 mm dorsoventral from the cranial surface) and into the DBB (0.6 mm anterior from the bregma, 0.5 mm bilateral of the midline and 7.8 mm dorsoventral from the cranial surface ). The rats subsequently received 30 mg/kg of desipramine (ip) to protect noradrenergic terminals 30 min before the 5,7-DHT injections. The dose administered was 1 mg of 5,7-DHT freebase in 0.1 ml of saline solution with ascorbic acid (0.1%). The volume injected into the MS was 0.5 μl, whereas 0.4 ml were bilaterally infused into the DBB. The rate of infusion was 0.1 ml/min using a Hamilton syringe of 25 gauge infusion stainless steel. During the same operation, a sub-group of eight rats was implanted with a chronic monopolar electrode in the hippocampal CA1 area, specifically 4.0 mm posterior from the bregma, 2.2 mm right of the midline, and 3.7 mm dorsoventral from the cranial surface. All coordinates reference the Watson and Paxinos atlas (Paxinos and Watson, 1996). The recording electrode was constructed of 25-gauge stainless steel insulated with epoxy resin and containing a small recording surface in the tip. The electrode was soldered to a telemetry transmitter that was placed under the skin of the neck of the rat. The vehicle group rats received the infusion of the vehicle solution in the coordinates and volumes as previously described for the 5,7-DHT group. Eight rats from the vehicle group were also implanted with the recording electrode as described for the 5,7-DHT group rats. After the surgery, the rats were subjected to food restriction (80% of their normal consumption ad libitum) for two weeks. They then underwent the radial arm maze training, which adhered to standard procedures. 2.3. Radial arm training The radial arm consisted of one central arena (30 cm in diameter) from which eight radial arms were extended (60 cm  10 cm). The rats were exposed to one habituation period of 10 min in the maze. On the following day, a second habituation period was conducted, and the rats were exposed to the fruit cereal (Fruit Loops), which was used as the reward. One day after the rats consumed the reward in the maze, the training began. All eight arms were baited for the test. Each rat was placed in the central area of the maze for a 30 s period, during which time access to the arms was blocked by a white cylinder. At the end of the 30 s period, the cylinder was moved and the trial was begun. The experimenter recorded the visited arms until 10 min had elapsed or until the rat had consumed all of the rewards placed in the containers at the ends of the arms. The rat was later removed from the maze and returned to its home cage for 20 min, after which it was again placed in the maze for a second trial. The rats were trained over seven consecutive days. The total errors – omission errors (the rat entered the arm did not proceed to the end of the arm or did not consume the cereal) and re-entry errors (entries previously visited arms) – were compared. The number of re-entry errors and the time to complete the task (mean of the two trials per day) were also compared. Inter-group comparisons were made using an ANOVA for repeated measures (days of training as the

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

repeated measure) and Tukey's test. Intra-group comparisons (between training days per group) were conducted using the ANOVA for blocks and Tukey's test. 2.4. EEG records Three telemetry recording plaques (Physiotel receiver, Mod RPC-1, DSI) that covered the area of the maze were placed underneath the maze such that when the rat was in the central arena, all three plaques recorded the EEG signal emitted by the transmitter, when the rat moved toward an arm, only two plaques transmitted the EEG signal, and when the rat went into an arm, only one plaque transmitted the signal until the rat reached the halfway point of the arm, after which the plaque no longer transmitted the EEG signal. In this way, we were able to select from the recordings only the EEG samples that corresponded to the journey of the rat from the centre area toward one arm (arm choice). The plaques were connected to one amplifier (Grass Neuro-Data Acquisition System, Model 15) through a consolidation matrix (BCM-100), and the EEG data, with a sampling frequency of 1024 Hz and with filters set between 1 and 100 Hz, were stored in a PC for later offline analysis. One daily basal EEG recording was taken from the rat in the central area of the maze during the 30 s period when the arms were blocked with a white cylinder. Following that, the EEG underlying the behavioural test was continuously recorded. It was only possible to obtain one second EEG samples from the first five days of training (exploration and 4 training days) because in the last three days, the rats travelled from one arm to another sufficiently fast such that the plaques were unable to transmit the signal. The EEG signal was submitted for the Fourier analysis (FFT) to obtain the power spectra. The theta band power (4–12 Hz) was divided into three sub-bands: low-frequency theta activity (4–6.5 Hz), high-frequency theta activity (6.5–9.5 Hz) and maximum-frequency theta activity (9.5–12 Hz). The relative power (RP) was obtained as the percentage of absolute power (AP) of each sub-band of theta in relation to the absolute power of the entire theta band (for example, 4–6.5 Hz RP¼4–6.5 Hz AP  100/4–12 Hz AP), the RP at each test stage was calculated after the data were transformed in natural logarithms (nl) according to the equation y¼nl(x/(100 x)), and the mean values for each stage and trial were compared using an ANOVA for repeated measures and Tukey's test post hoc. As a standard requirement, a Po0.05 was considered to be significant for all comparisons. Finally, five additional rats were lesioned in the MS/BDB, as previously described, and the serotonin and 5 hydroxyl indole acetic acid (5HIAA) concentrations in the MS/DBB were measured using HPLC, and the data were compared with results from five intact rats (Student's t-test). Samples including MS and DBB were obtained cutting a slice containing the region of interest and then punching a sample of tissue with a pipette tip, the size of which was adjusted to cleave the samples without including adjacent tissue. The tissue samples were homogenated (HCl 1 N) and centrifuged, and the content of serotonin and 5HIAA (pg/mg of fresh tissue) of the supernatant was determined using a Lichrocart purospher star column (150  4.6, RP-18 endcaped, 5 mm, MERK KGaA, Darmstadt; Germany) with a mobile phase (pH 3.1) composed of citric acid (50 mM), H3PO4 (50 mM), EDTA (20 mg), octanesulfonic acid (120 mg/L) and methanol (10%). The flow rate was 1.5 mL/min. An electrochemical detector (Atec Lyden VT-03) with a work potential of 680 mV was adjusted to the pH of the mobile phase and used. The data were compared with Student's t-test.

107

removed, and 40-mm sections were sliced with a cryostat and stained to localise the positions in which the tips of the cannula had been placed during the 5,7-DHT infusion. Only rats with correct electrode placement and those where the tip of the cannula had been correctly placed were included in the EEG results. Accordingly, the vehicle group had a final number of ten rats in the behavioural test. From these ten, hippocampal EEG recordings were obtained from five of the rats. The 5,7-DHT group consisted of nine rats for the behavioural performance test, seven of which were monitored for EEG recording.

3. Results 3.1. Histology The position of the cannule was histologically verified. The 5,7DHT and vehicle solution infusions produced mechanical damage that was evident in the cortical damage caused by the insertion of the gauge, the breaking of fibres in the corpus callosum caused by the gauge trajectory, an accumulation of glial cells in a line following the tract of the gauge and a small area of glial accumulation in the place where the tip of the cannule remained during the infusion. The lesion procedure, assessed in five separated rats, resulted in a 60% reduction of serotonin and an approximate 30% reduction of 5HIAA compared with the amounts of the two in the MS/DBB of intact rats (t¼2.555, DF¼ 8, P ¼0.034 and t¼4.257, DF¼ 8, P ¼ 0.003, for 5-HIAA and serotonin, respectively) (Fig. 1). 3.2. Behaviour Inter-group comparisons were conducted using an ANOVA for repeated measures and Tukey's post hoc test. In this comparison, the 5,7-DHT group had a minor number of total errors (omissions and re-entries to arm) compared with the vehicle group [main effect; F (1,17) ¼15.114, P ¼0.001], and a significant effect of the training day was also observed [F (6,102)¼ 2.585, P¼0.023], whereas the interaction of group and day was not significant (Fig. 2A). The total number of errors committed by the two groups was also compared considering only the four days when the EEG recordings were taken. Significant differences were observed for the group [F (1,17)¼ 15.892, P¼0.001] but not for the day or for the interaction of group and day. The intra-group comparisons of the total number of errors committed by the rats on each day of training (mean from the two daily trials) were conducted for each group to assess their learning curve using a block design and Tukey's post hoc test. Training for

2.5. Histology The rats were intracardially perfused with a buffer solution followed by a paraformaldehyde solution. The brains were then

Fig. 1. Concentration of SM/BDB 5-HT and 5-HIIA (percentages from the intact rats). Mean 7 S.E.M. þ , intact vs. 5,7-DHT group, P o 0.05.

108

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

Fig. 2. (A) Number of total errors committed by the two groups of rats in the radial arm task. Both groups reduced the number of errors on days 6 and 7 compared with the first day of training. The 5,7-DHT group had a minor number of errors compared to the vehicle group (main effect). (B) Number of re-entry errors committed by the two groups of rats. The 5,7-DHT group had a minor number of errors compared to the vehicle group (main effect). (C) Time required by the two groups of rats to complete the task. The 5,7-DHT group had shorter times than the vehicle group (main effect). Mean 7S.E.M. *Day 1 vs. subsequent days, Po 0.05; þvehicle group vs. 5,7-DHT group.

the vehicle group exhibited a significant effect [F (6, 54) ¼8.342, P o0.001] as paired comparisons showed that this group significantly reduced their number of total errors on days six (P ¼0.027) and seven (P ¼0.024) when compared with their first day of training (Fig. 2A). In the intra-group comparison of total errors, training have a significant effect on the 5,7-DHT group

[F (6, 48) ¼5.376, P o0.001]. Similar to the vehicle group, this group significantly reduced the number of total errors on days six and seven compared with the first training day (P ¼0.006 and P¼ 0.004, respectively) (Fig. 2A). In comparing the number of re-entries to arms, a significant effect of group was observed when considering the seven days of

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

training [F (1, 17) ¼13.402, P ¼0.002] or only the first four days of training when the EEG records were obtained [F (1, 17) ¼15.642, P ¼0.001] (Fig. 2B). No other significant differences were observed for this comparison. In the intra-group comparison of the re-entry errors, the vehicle group rats showed a significant effect of the training [F (6, 54)¼ 4.902, Po0.001] as paired comparisons showed that the number of errors was significantly minor on days six and seven (P¼0.046) only when compared with the second day of training, but not when compared with the first day of training when most of the omission errors occurred (Fig. 2B). A similar result was observed for the 5,7-DHT group as training exhibited a significant effect [F (6, 48)¼ 4.678, P¼0.001] with a reduction of errors on the seventh (P¼0.048) day compared with the second day of training but not with the first day (Fig. 2B). This finding was attributable to the occurrence of the omission errors, particularly on the first day of training. The inter-group comparison of the time required to complete the trials showed significant effect for the group factor when comparing both the seven days of training [F (1, 17) ¼13.391, P ¼0.001] and the first four days of training [F (1,17) ¼5.061, P ¼0.038]. However, no differences were observed with respect to the day or the interaction group and the day. Training exhibited a significant effect in the intra-group comparisons with respect to the time required by the vehicle rats to accomplish the task [F (6, 54)¼ 2.835, P¼0.018], while the paired comparisons showed a bias toward the minimal time required to solve the task on the fifth day of training (P¼0.062) compared with that of the first day of training (Fig. 2C). The 5,7-DHT group also indicated that the training day had a significant effect on the time required to realise the task [F (6, 48)¼ 14.108, Po0.001], and paired comparisons showed that the time to realise the task was shorter on days three to seven compared to that of the first day (Pr0.01) and on days five to seven (Pr0.008) compared to that of the second day of training (Fig. 2C). Finally, to determine whether the behaviour of the implanted rats was different from that of the non-implanted rats, a comparison of the behavioural variables between implanted (n¼5) and non-implanted (n¼ 5) rats was conducted for the vehicle group. Considering the seven days of training, no significant differences for group were observed in total errors [F (1,8)¼ 0.258, P¼0.625], omission errors [F (1, 8)¼0.890, P¼0.373], or time required to

109

complete the trials [F (1, 8)¼1.088, P¼0.327]. Furthermore, no significant differences were observed with respect to the interaction of group and day for any of the variables. Of the 5,7-DHT group, as only two rats without implants were included in the behavioural results, a comparison between implanted and non-implanted rats was not conducted. 3.3. EEG We compared the RP of the EEG obtained on days one to four of training. Inter-group comparisons were conducted using an ANOVA for repeated measures and considering the factors group, sub-band of theta frequency and behavioural condition and the days as repeated measure (exploration day and four training days), along with Tukey's post hoc test. Intra-group comparisons were conducted using a block design that included three factors: day of training (exploration and days 1–4), behavioural condition (basal and test, the latter included exploration and four training sessions) and sub-band of frequency (low frequency between 4 and 6.5 Hz and high frequency between 6.5 and 9.5 Hz). The Tukey test was used post hoc. The intra-group comparisons of the vehicle group showed differences in theta RP between bands and behavioural condition [F (1, 80) ¼ 11.388, P¼ 0.001]. Paired comparisons showed that this group had more high-frequency RP compared to low-frequency RP both in the basal (P ¼0.006) and in the test (Po0.001) conditions. Regarding the interactions of the factors, day and sub-band of frequency, this group exhibited no significant effects [F (4, 80) ¼ 2.018, P ¼0.100] (Fig. 3). Furthermore, there were no significant effects in the interactions of sub-band of frequency, training days and behavioural conditions [F (4, 80) ¼0.573, P¼ 0.683]. The intra-group comparison of the 5,7-DHT group exhibited differences between behavioural stage and sub-band of frequency [F (1, 99) ¼35.264, P o0.001], similar to the differences exhibited by the vehicle group. The high-frequency RP was greater than the low-frequency RP both in the basal (P o0.001) and in the test conditions (P o0.001). Furthermore, the low-frequency RP was lower during the test condition than in the basal condition (P o0.001), and there was higher high-frequency RP during the test condition than in the basal condition (P o0.001). In addition,

Fig. 3. Relative power (ln) of the low-frequency (4–6.5 Hz) and high-frequency (6.5–9.5 Hz) EEG signals recorded from the two groups of rats. Comparisons were significant considering behavioural conditions and the sub-band of frequency for the two groups of animals (top); the effect considering the day of training and sub-band of frequency was significant for the 5,7-DHT group (bottom). Mean7 S.E.M. *Low frequency vs. high frequency; þbasal vs. test (exploration and training) conditions, Pr 0.006.

110

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

this group showed a significant effect of the interaction between day of training and sub-band of frequency [F (4, 99) ¼3.214, P ¼0.016]. The paired comparisons showed that for all days (exploration and days 1–4 of training), the high-frequency theta RP was higher than the low-frequency theta RP (Po 0.001). In other words, in these rats, a dominance of high-frequency theta activity was associated with better behavioural performance in comparison to the vehicle rats (Fig. 3). No effects of the interaction among day of training, behavioural condition and sub-band of frequency were observed [F (4, 99) ¼ 1.243, P ¼0.298]. In the inter-group comparisons, a significant difference between groups was observed for the sub-band of frequency [F (1, 38)¼ 4.353, P¼0.044] as the paired comparisons showed that the 5,7-DHT group had minor low-frequency RP (Po0.001) and major highfrequency RP (Po0.001) compared with the vehicle group. Additional inter-group differences were observed for behavioural condition and sub-band of frequency [F (1, 38)¼6.696, P¼0.014] as paired comparisons showed that the 5,7-DHT group had less low-frequency RP and more high-frequency RP than did the vehicle group in both the basal (P¼0.008 and P¼0.013, respectively) and the test conditions (Po0.001) (Fig. 4). Finally, a significant difference between groups when considering day and sub-band of frequency was observed [F (4, 152)¼4.250, P¼ 0.003]. The 5,7-DHT lesion group had lower low-frequency RP than did the vehicle group on days two (P¼0.017), three (P¼0.011) and four (P¼ 0.025), and it had higher high-frequency RP than did the vehicle group on the same three days (P¼0.030, P¼ 0.019 and P¼0.025, respectively) (Fig. 4). Thus, better performance was evidenced by a lower number of errors during the

Fig. 4. Relative power (ln) of the low-frequency (4–6.5 Hz) and high-frequency (6.5–9.5 Hz) EEG signals recorded from the two groups of rats. Inter-group comparisons were significant considering the sub-band of frequency (top), the interaction of the behavioural stage and the sub-band of frequency (middle), and the interaction of the day of training, the behavioural stage and the sub-band of frequency (bottom). Mean 7 S.E.M. þ Vehicle vs. 5,7-DHT group, Po 0.05.

learning of the working memory test was associated with the major expression of high-frequency theta activity in rats with lesions of SM/DBB serotonergic terminals.

4. Discussion While the serotonin modulation of learning has been largely sustained by various experimental approaches, the results to date have been complex and dependent on the behavioural tasks and the experimental approach employed to modify the serotonergic activity. With respect to spatial learning, no effects were observed after the 5,7-DHT cerebral serotonin reduction in rats trained on a working memory task (delay not matching to position) (Ruotsalainen et al., 1998). Furthermore, the intracerebroventricular infusion of 5,7-DHT did not alter performance of male rats on the Morris water maze and radial arms maze (Lehmann et al., 2002b). While Asin et al. (1985) evaluated the effect of cerebral serotonin reduction through MR infusion of 5,7-DHT on a working memory test using the radial arm maze in comparison with electrolytic lesions of MR and observed deficiencies only after an electrolytic lesion of RM, they did not examine the effects of a reduction in serotonin. Murtha and Pappas (1994) did not find effects on behavioural performance of rats tested using the Morris water maze (spatial reference memory) and radial arm (spatial working memory test) after the reduction of the hippocampal serotonin concentration through the infusion of 5,7-DHT on the fimbria fornix and cingulum bundle. Similarly, no effect on learning was reported after specific hippocampal serotonin reduction (Lehmann et al., 2002a). Among others, this work supports the premise that cerebral or hippocampal serotonin depletion alone has no effect on learning and memory in spatial tasks. However, other authors have reported deficiencies in working memory after cerebral serotonin depletion (Cassaday et al., 2003) or the potentiation of those deficits induced by cholinergic lesions with concomitant serotonin depletion (Lehmann et al., 2002b; Murtha and Pappas, 1994; Richter-Levin and Segal, 1991). Still others have reported evidence of spatial learning improvement after cerebral (Altman et al., 1989; Sarihi et al., 2000), forebrain (Ward et al., 1999), prefrontal (Perez-Vega et al., 2000) or hippocampal (Altman et al., 1990; Gutiérrez-Guzmán et al., 2011) serotonin depletion. To the best of our knowledge, there are no studies that evaluate the specific consequences of SM/DBB serotonin depletion on learning. However, while it has been demonstrated that the activation of 5-HT1A receptors in the medial septum or MS/DBB induces impairment in water maze learning (Koenig et al., 2011) and working memory (Jeltsch et al., 2004), the same pharmacological strategy was reported to have no effect on performance on the water maze test (Elvander-Tottie et al., 2009). The present results indicate that spatial working memory is enhanced specifically after MS/DBB serotonin depletion as evidenced by the minor number of errors committed by the serotonin-depleted rats throughout the training. This facilitation was accompanied by an increase in the RP corresponding with the high-frequency theta band and a concomitant reduction in the low-frequency theta RP. The present results support an inhibitory role of serotonin in spatial working memory acting on SM/DBB. In accordance with this issue, it has been observed that the activation of the 5HT1A receptors due to the intra-septal administration of 8-OH-DPAT produces deficiencies in spatial working memory as evidenced on the water maze test (Jeltsch-David et al., 2008; Jeltsch et al., 2004). Additionally, the agonist infusion induced place learning and retrieval impairment on a spatial memory task as demonstrated in the Morris maze, only when the infusion was administered before training (Koenig et al., 2011). Because the lesion of cholinergic neurons by the 192-IGg-saporin

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

administration did not produce spatial learning impairment (Koenig et al., 2011), it is possible that the effects of 8-OH-DPAT could be mediated by its action on the GABAergic (Alreja et al., 2000) and glutamatergic neurons. Recently, it has been demonstrated that septal GABAergic cells play a principal role in the generation of hippocampal theta activity (Hangya et al., 2009). In support of the principal role of MS in serotonergic theta activity modulation, it has been observed in urethaneanaesthetised rats that persistent hippocampal theta activity induced by procaine application to the MR was blocked by septal inactivation with procaine or by the administration of septal atropine (Crooks et al., 2012). These results suggest that the modulator effect of the raphe on the hippocampal theta activity depends on the activity of the SM neurons in anaesthetised rats. Additionally, a principal role of SM as a recipient of the raphe influences for the serotonergic modulation of theta activity was previously demonstrated in anaesthetised rats and in awake rabbits (Crooks et al., 2012; Kinney et al. 1996; Vinogradova et al., 1999). Our results are consistent with the role of MS/DBB as a principal target of the serotonergic influence in hippocampal theta activity modulation, given that the selective depletion of serotonin to the MS/DBB induced a higher expression of thighfrequency theta activity (6.5–9.5 Hz) in rest conditions, as well as during the processing of spatial working memory in freely moving rats. Furthermore, this change in theta activity expression was related to better performance on the radial arm task, a task that involved septal (Hepler et al., 1985) and hippocampal processing (Floresco et al., 1997). The possibility that intra-septal infusion of 8-OHDPAT could induce a disturbance in the rhythmic firing of these neurons has been considered by other researchers (Jeltsch-David et al., 2008); however, this possibility remains to be analysed. Although there are few studies that assess the consequences of serotonergic manipulations locally in the MS/DBB on electrical activity, it has been determined that the systemic administration of 5HT2C agonists induces a reduction in the rhythmic firing in the MS concurrent with a reduction in the hippocampal theta activity, whereas the administration of antagonists for these receptors has the opposite effect, increasing the rhythmic firing of MS neurons and increasing the theta rhythmicity in the hippocampus (Hajos et al., 2003). The increased high-frequency theta activity observed in serotonin-MS/DBB depleted rats possibly reflects the lack of influence of serotonin on all of the neuronal types and on all of the classes of 5-HT receptors, which is conducive to a global effect of increased rhythmic influences of MS/DBB on the hippocampus. The present results, when added to previous findings that show that hippocampal serotonin depletion induces facilitation in place learning ability is related to predominantly high-frequency theta activity in the underlying EEG (Gutiérrez-Guzmán et al., 2011), indirectly indicates that serotonin acts as an inhibitor of theta oscillations and a modulator of hippocampal information processing on both the septum and the hippocampus. However, in other studies, it was observed that the depletion of serotonin in the SUM/PH induces deficiencies on the same place learning test that are associated with a lack of learning-related changes in hippocampal theta activity and with behavioural impairments (GutiérrezGuzmán et al., 2012). The SUM/PH has been related to the codification of the frequency and/or the amplitude of the theta activity (Bland et al., 1990; Kirk and McNaughton, 1993). Accordingly, the role of serotonin in hippocampal theta activity modulation is complex and could vary in the different relays of the synchronising ascending system depending on the contribution of each relay to the hippocampal theta expression. However, the effects of serotonin depletion on learning are related with changes

111

in hippocampal theta activity in a consistent manner. When serotonin depletion results in lower hippocampal high-frequency theta activity expression, it is associated with marginal efficiency in learning. On the contrary, when serotonin depletion causes higher hippocampal high-frequency theta activity, it is associated with enhanced learning. Thus, serotonin, hippocampal theta activity and learning appear to be closely related. Our present results are consistent with the hypothesis of Jeltsch-David et al. (2008), which proposes that low serotonergic tonus in the medial septum may increase the cholinergic and GABAergic influence on the septum and hippocampus, thereby constituting an encoding/ consolidation state. However, studies that assess the effects of an increase in serotonergic activation on hippocampal theta activity and learning processes are worthwhile and have yet to be realised. Finally, the minimal time employed by the MS/DBB group in task solving could be related to the lower number of arms visited by this group of rats. However, the role of the septo-hippocampal system, on the one hand, and the role of serotonin, on the other hand, with respect to the modulation of anxiety raises the possibility that serotonin depletion could alter the anxiety level of MS/DBB-depleted rats. The reduction in time elapsed to complete the trials could be related to a reduced anxiety level, which, in turn, enables the serotonin-depleted rats to travel faster through the arms. However, it has been reported that anxiolityc drugs reduce the frequency of theta activity evoked by reticular stimulation (Zhu and McNaughton, 1994). Because we did not specifically evaluate anxiety levels, this possibility remains to be addressed. Finally, we cannot rule out the possibility that some marginal reduction of hippocampal serotonin could occur after the SM/DBB serotonergic lesions. However, the magnitude of the changes observed after the MS/DBB infusion of 5,7-DHT with respect to both behaviour and EEG variables was major compared to the magnitude of the changes observed after specific hippocampal depletion (Gutiérrez-Guzmán et al., 2011). Furthermore, while the changes were expressed in all behavioural stages (basal and learning), a difference was observed after the hippocampal serotonin depletion in which high-frequency theta activity increased only during the search for the platform in the Morris maze. In other words, the high-frequency theta activity did not increase in the rest condition. Thus, the serotonergic modulation of hippocampal theta activity seems to have much greater weight when acting on the MS/DBB than when acting directly on the hippocampus. In conclusion, septal serotonin denervation in rats induced the increased expression of high-frequency theta activity in hippocampal CA1 EEGs, as well as better performance in a spatial working memory task.

Acknowledgements This work was supported by the “Fondo de Investigación en Salud, No. FIS/IMSS/PROT/C2007/067”. References Acsady, L., Arabadzisz, D., Katona, I., Freund, T.F., 1996. Topographic distribution of dorsal and median raphe neurons with hippocampal, septal and dual projection. Acta Biol. Hung. 47, 9–19. Alreja, M., 1996. Excitatory actions of serotonin on GABAergic neurons of the medial septum and diagonal band of Broca. Synapse 22, 15–27. Alreja, M., Wu, M., Liu, W., Atkins, J.B., Leranth, C., Shanabrough, M., 2000. Muscarinic tone sustains impulse flow in the septohippocampal GABA but not cholinergic pathway: implications for learning and memory. J. Neurosci. 20, 8103–8110. Altman, H.J., Normile, H.J., Galloway, M.P., Ramirez, A., Azmitia, E.C., 1990. Enhanced spatial discrimination learning in rats following 5,7-DHT-induced serotonergic deafferentation of the hippocampus. Brain Res. 518, 61–66.

112

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

Altman, H.J., Ogren, S.O., Berman, R.F., Normile, H.J., 1989. The effects of p-chloroamphetamine, a depletor of brain serotonin, on the performance of rats in two types of positively reinforced complex spatial discrimination tasks. Behav. Neural Biol. 52, 131–144. Amaral, D.G., Kurz, J., 1985. An analysis of the origins of the cholinergic and noncholinergic septal projections to the hippocampal formation of the rat. J. Comp. Neurol. 240, 37–59. Asin, K.E., Wirtshafter, D., Fibiger, H.C., 1985. Electrolytic, but not 5,7-dihydroxytryptamine, lesions of the nucleus medianus raphe impair acquisition of a radial maze task. Behav. Neural Biol. 44, 415–424. Assaf, S.Y., Miller, J.J., 1978. The role of a raphe serotonin system in the control of septal unit activity and hippocampal desynchronization. Neuroscience 3, 539–550. Bland, B.H., Colom, L.V., Ford, R.D., 1990. Responses of septal theta-on and theta-off cells to activation of the dorsomedial-posterior hypothalamic region. Brain Res. Bull. 24, 71–79. Cassaday, H.J., Norman, C., Shilliam, C.S., Vincent, C., Marsden, C.A., 2003. Intraventricular 5,7-dihydroxytryptamine lesions disrupt acquisition of working memory task rules but not performance once learned. Prog. Neuropsychopharmacol. Biol. Psychiatry 27, 147–156. Colom, L.V., Castaneda, M.T., Reyna, T., Hernandez, S., Garrido-Sanabria, E., 2005. Characterization of medial septal glutamatergic neurons and their projection to the hippocampus. Synapse 58, 151–164. Cornwell, B.R., Johnson, L.L., Holroyd, T., Carver, F.W., Grillon, C., 2008. Human hippocampal and parahippocampal theta during goal-directed spatial navigation predicts performance on a virtual Morris water maze. J. Neurosci. 28, 5983–5990. Crooks, R., Jackson, J., Bland, B.H., 2012. Dissociable pathways facilitate theta and non-theta states in the median raphe-septohippocampal circuit. Hippocampus 22, 1567–1576. Eichenbaum, H., Otto, T., Cohen, N.J., 1992. The hippocampus – what does it do? Behav. Neural Biol. 57, 2–36. Elvander-Tottie, E., Eriksson, T.M., Sandin, J., Ogren, S.O., 2009. 5-HT(1A) and NMDA receptors interact in the rat medial septum and modulate hippocampaldependent spatial learning. Hippocampus 19, 1187–1198. Floresco, S.B., Seamans, J.K., Phillips, A.G., 1997. Selective roles for hippocampal, prefrontal cortical, and ventral striatal circuits in radial-arm maze tasks with or without a delay. J. Neurosci. 17, 1880–1890. Ford, R.D., Colom, L.V., Bland, B.H., 1989. The classification of medial septumdiagonal band cells as theta-on or theta-off in relation to hippocampal EEG states. Brain Res. 493, 269–282. Freund, T.F., 1989. GABAergic septohippocampal neurons contain parvalbumin. Brain Res. 478, 375–381. Freund, T.F., Antal, M., 1988. GABA-containing neurons in the septum control inhibitory interneurons in the hippocampus. Nature 336, 170–173. Givens, B., Olton, D.S., 1994. Local modulation of basal forebrain: effects on working and reference memory. J. Neurosci. 14, 3578–3587. Gutiérrez-Guzmán, B.E., Hernández-Pérez, J.J., González-Burgos, I., Feria-Velásco, A., Medina, R., Guevara, M.A., et al., 2011. Hippocampal serotonin depletion facilitates place learning concurrent with an increase in CA1 high frequency theta activity expression in the rat. Eur. J. Pharmacol. 652, 73–81. Gutiérrez-Guzmán, B.E., Hernández-Pérez, J.J., López-Vázquez, M.A., Fregozo, C.S., Guevara, M.A., Olvera-Cortés, M.E., 2012. Serotonin depletion of supramammillary/posterior hypothalamus nuclei produces place learning deficiencies and alters the concomitant hippocampal theta activity in rats. Eur. J. Pharmacol. 682, 99–109. Hajos, M., Hoffmann, W.E., Weaver, R.J., 2003. Regulation of septo-hippocampal activity by 5-hydroxytryptamine(2C) receptors. J. Pharmacol. Exp. Ther. 306, 605–615. Hangya, B., Borhegyi, Z., Szilagyi, N., Freund, T.F., Varga, V., 2009. GABAergic neurons of the medial septum lead the hippocampal network during theta activity. J. Neurosci. 29, 8094–8102. Hepler, D.J., Wenk, G.L., Cribbs, B.L., Olton, D.S., Coyle, J.T., 1985. Memory impairments following basal forebrain lesions. Brain Res. 346, 8–14. Jeltsch-David, H., Koenig, J., Cassel, J.C., 2008. Modulation of cholinergic functions by serotonin and possible implications in memory: general data and focus on 5-HT(1A) receptors of the medial septum. Behav. Brain Res. 195, 86–97. Jeltsch, H., Bertrand, F., Galani, R., Lazarus, C., Schimchowitsch, S., Cassel, J.C., 2004. Intraseptal injection of the 5-HT1A/5-HT7 agonist 8-OH-DPAT and working memory in rats. Psychopharmacology (Berl.) 175, 37–46. Kinney, G.G., Kocsis, B., Vertes, R.P., 1996. Medial septal unit firing characteristics following injections of 8-OH-DPAT into the median raphe nucleus. Brain Res. 708, 116–122. Kirk, I.J., McNaughton, N., 1993. Mapping the differential effects of procaine on frequency and amplitude of reticularly elicited hippocampal rhythmical slow activity. Hippocampus 3, 517–525. Kitchigina, V.F., Kudina, T.A., Kutyreva, E.V., Vinogradova, O.S., 1999. Neuronal activity of the septal pacemaker of theta rhythm under the influence of stimulation and blockade of the median raphe nucleus in the awake rabbit. Neuroscience 94, 453–463. Koenig, J., Lecourtier, L., Cosquer, B., Pereira, P.M., Cassel, J.C., 2011. Spatial memory alterations by activation of septal 5HT 1A receptors: no implication of cholinergic septohippocampal neurons. Psychopharmacology (Berl.) 214, 437–454. Lehmann, O., Bertrand, F., Jeltsch, H., Morer, M., Lazarus, C., Will, B., et al., 2002a. 5,7-DHT-induced hippocampal 5-HT depletion attenuates behavioural deficits produced by 192 IgG-saporin lesions of septal cholinergic neurons in the rat. Eur. J. Neurosci. 15, 1991–2006.

Lehmann, O., Jeltsch, H., Lazarus, C., Tritschler, L., Bertrand, F., Cassel, J.C., 2002b. Combined 192 IgG-saporin and 5,7-dihydroxytryptamine lesions in the male rat brain: a neurochemical and behavioral study. Pharmacol. Biochem. Behav. 72, 899–912. Leranth, C., Vertes, R.P., 1999. Median raphe serotonergic innervation of medial septum/diagonal band of broca (MSDB) parvalbumin-containing neurons: possible involvement of the MSDB in the desynchronization of the hippocampal EEG. J. Comp. Neurol. 410, 586–598. Luttgen, M., Ogren, S.O., Meister, B., 2005. 5-HT1A receptor mRNA and immunoreactivity in the rat medial septum/diagonal band of Broca-relationships to GABAergic and cholinergic neurons. J. Chem. Neuroanat. 29, 93–111. Luttgen, M., Ove Ogren, S., Meister, B., 2004. Chemical identity of 5-HT2A receptor immunoreactive neurons of the rat septal complex and dorsal hippocampus. Brain Res. 1010, 156–165. Maho, C., Dutrieux, G., Ammassari-Teule, M., 1988. Parallel modifications of spatial memory performances, exploration patterns, and hippocampal theta rhythms in fornix-damaged rats: reversal by oxotremorine. Behav. Neurosci. 102, 601–604. Manns, J.R., Clark, R.E., Squire, L.R., 2002. Standard delay eyeblink classical conditioning is independent of awareness. J. Exp. Psychol.: Anim. Behav. Process. 28, 32–37. Maru, E., Takahashi, L.K., Iwahara, S., 1979. Effects of median raphe nucleus lesions on hippocampal EEG in the freely moving rat. Brain Res. 163, 223–234. McNaughton, N., Ruan, M., Woodnorth, M.A., 2006. Restoring theta-like rhythmicity in rats restores initial learning in the Morris water maze. Hippocampus 16, 1102–1110. Mitchell, S.J., Rawlins, J.N., Steward, O., Olton, D.S., 1982. Medial septal area lesions disrupt theta rhythm and cholinergic staining in medial entorhinal cortex and produce impaired radial arm maze behavior in rats. J. Neurosci. 2, 292–302. Mizumori, S.J., Perez, G.M., Alvarado, M.C., Barnes, C.A., McNaughton, B.L., 1990. Reversible inactivation of the medial septum differentially affects two forms of learning in rats. Brain Res. 528, 12–20. Monmaur, P., Ayadi, K., Breton, P., 1993. Hippocampal EEG responses induced by carbachol and atropine infusions into the septum and the hippocampus in the urethane-anaesthetized rat. Brain Res. 631, 317–324. Murtha, S.J., Pappas, B.A., 1994. Neurochemical, histopathological and mnemonic effects of combined lesions of the medial septal and serotonin afferents to the hippocampus. Brain Res. 651, 16–26. Normile, H.J., Jenden, D.J., Kuhn, D.M., Wolf, W.A., Altman, H.J., 1990. Effects of combined serotonin depletion and lesions of the nucleus basalis magnocellularis on acquisition of a complex spatial discrimination task in the rat. Brain Res. 536, 245–250. Olvera-Cortés, E., Cervantes, M., González-Burgos, I., 2002. Place-learning, but not cue-learning training, modifies the hippocampal theta rhythm in rats. Brain Res. Bull. 58, 261–270. Olvera-Coréts, E., Guevara, M.A., González-Burgos, I., 2004. Increase of the hippocampal theta activity in the Morris water maze reflects learning rather than motor activity. Brain Res. Bull. 62, 379–384. Pan, W.X., McNaughton, N., 1997. The medial supramammillary nucleus, spatial learning and the frequency of hippocampal theta activity. Brain Res. 764, 101–108. Paxinos, G., Watson, C., 1996. The Rat Brain in Esterotaxic Corrdinates. Pérez-Vega, M.I., Feria-Velásco, A., Gonázlez-Burgos, I., 2000. Prefrontocortical serotonin depletion results in plastic changes of prefrontocortical pyramidal neurons, underlying a greater efficiency of short-term memory. Brain Res. Bull. 53, 291–300. Petsche, H., Stumpf, C., Gogolak, G., 1962. The significance of the rabbit's septum as a relay station between the midbrain and the hippocampus. I. The control of hippocampus arousal activity by the septum cells. Electroencephalogr. Clin. Neurophysiol. 14, 202–211. Rawlins, J.N., Feldon, J., Gray, J.A., 1979. Septo-hippocampal connections and the hippocampal theta rhythm. Exp. Brain Res. 37, 49–63. Richter-Levin, G., Segal, M., 1991. The effects of serotonin depletion and raphe grafts on hippocampal electrophysiology and behavior. J. Neurosci. 11, 1585–1596. Ruotsalainen, S., MacDonald, E., Koivisto, E., Stefanski, R., Haapalinna, A., Riekkinen Jr., P., et al., 1998. 5-HT1A receptor agonist (8-OH-DPAT) and 5-HT2 receptor agonist (DOI) disrupt the non-cognitive performance of rats in a working memory task. J. Psychopharmacol. 12, 177–185. Sarihi, A., Motamedi, F., Naghdi, N., Rashidy-Pour, A., 2000. Lidocaine reversible inactivation of the median raphe nucleus has no effect on reference memory but enhances working memory versions of the Morris water maze task. Behav. Brain Res. 114, 1–9. Sotty, F., Danik, M., Manseau, F., Laplante, F., Quirion, R., Williams, S., 2003. Distinct electrophysiological properties of glutamatergic, cholinergic and GABAergic rat septohippocampal neurons: novel implications for hippocampal rhythmicity. J. Physiol. 551, 927–943. Vanderwolf, C.H., 1969. Hippocampal electrical activity and voluntary movement in the rat. Electroencephalogr. Clin. Neurophysiol. 26, 407–418. Vertes, R.P., 1986. Brainstem modulation of the hippocampus. Anatomy, physiology and significance. In: Isaacson, R.L., Pribram, K.H. (Eds.), The Hippocampus, vol. 4. Plenum, New York, pp. 41–75. Vertes, R.P., Kocsis, B., 1997. Brainstem–diencephalo–septohippocampal systems controlling the theta rhythm of the hippocampus. Neuroscience 81, 893–926.

M.Á. López-Vázquez et al. / European Journal of Pharmacology 734 (2014) 105–113

Vertes, R.P., Martin, G.F., 1988. Autoradiographic analysis of ascending projections from the pontine and mesencephalic reticular formation and the median raphe nucleus in the rat. J. Comp. Neurol. 275, 511–541. Vinogradova, O.S., 1995. Expression, control, and probable functional significance of the neuronal theta-rhythm. Prog. Neurobiol. 45, 523–583. Vinogradova, O.S., Kitchigina, V.F., Kudina, T.A., Zenchenko, K.I., 1999. Spontaneous activity and sensory responses of hippocampal neurons during persistent thetarhythm evoked by median raphe nucleus blockade in rabbit. Neuroscience 94, 745–753. Ward, B.O., Wilkinson, L.S., Robbins, T.W., Everitt, B.J., 1999. Forebrain serotonin depletion facilitates the acquisition and performance of a conditional visual discrimination task in rats. Behav. Brain Res. 100, 51–65.

113

Watrous, A.J., Fried, I., Ekstrom, A.D., 2011. Behavioral correlates of human hippocampal delta and theta oscillations during navigation. J. Neurophysiol. 105, 1747–1755. Whishaw, I.Q., Vanderwolf, C.H., 1973. Hippocampal EEG and behavior: changes in amplitude and frequency of RSA (theta rhythm) associated with spontaneous and learned movement patterns in rats and cats. Behav. Biol. 8, 461–484. Zhu, X.O., McNaughton, N., 1994. The interaction of serotonin depletion with anxiolytics and antidepressants on reticular-elicited hippocampal RSA. Neuropharmacology 33, 1597–1605.