Tetrodotoxin inactivation of pontine regions: Influence on sleep–wake states

Brain Research 1044 (2005) 42 – 50 www.elsevier.com/locate/brainres

Research report

Tetrodotoxin inactivation of pontine regions: Influence on sleep–wake states Larry D. Sanforda,T, Linghui Yanga, Xiangdong Tanga, Richard J. Rossb,c, Adrian R. Morrisonb a

Sleep Research Laboratory, Department of Pathology and Anatomy, Eastern Virginia Medical School, PO Box 1980, Norfolk, VA 23501, USA b Laboratory for Study of the Brain in Sleep, Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, PA 19104, USA c VA Medical Center, Philadelphia, PA 19104, USA Accepted 24 February 2005 Available online 8 April 2005

Abstract Studies using various methodologies have implicated n. reticularis pontis oralis (RPO) and n. subcoeruleus (SubC) in the generation of rapid eye movement sleep (REM). In rats, electrolytic lesions in these regions may give rise to the phenomenon of REM without atonia (REM-A), in which the electrophysiological features of REM are normal except that atonia is absent and elaborate behaviors may be exhibited. However, electrolytic lesions damage both cell bodies and fibers of passage, and the neural reorganization and adaptation that can occur post-lesion can complicate interpretation. Tetrodotoxin (TTX) is a sodium channel blocker that temporarily inactivates both neurons and fibers of passage and thus may be functionally equivalent to an electrolytic lesion, but without allowing time for neural adaptation. In this study, we examined the influence of microinjections of TTX into RPO and SubC on sleep in freely behaving rats. Rats (90 day old male Sprague–Dawley) were implanted with electrodes for recording EEG and EMG. Guide cannulae were implanted aimed into RPO or SubC. Each animal received one unilateral microinjection (TTXUH: 5.0 ng/0.2 Al) and two bilateral microinjections (TTXBL: 2.5 ng/0.1 Al; TTXBH: 5.0 ng/0.2 Al) of TTX, and control microinjections of saline alone (SAL). The injections were made 2 h following lights on, and sleep was recorded for the subsequent 22 h. Sleep was scored from computerized records in 10 s epochs. Recordings from the 10-h light period and the 12-h dark period were examined separately. TTX inactivation of RPO could decrease REM and non-REM (NREM), whereas inactivation of SubC produced relatively more specific decreases in REM with smaller effects on NREM. The results complement studies that have implicated RPO and SubC in REM generation. REM-A was not observed, suggesting that REM-A is a complex phenomenon that requires time for reorganization of the nervous system after insult. D 2005 Elsevier B.V. All rights reserved. Theme: Neural basis of behavior Topic: Biological rhythms and sleep Keywords: Tetrodotoxin; Nucleus reticularis pontis oralis; Nucleus subcoeruleus; Rapid eye movement sleep; Sleep

1. Introduction The nucleus reticularis pontis oralis (RPO), which forms the bulk of the medial pontine reticular formation, is important for the generation of rapid eye movement sleep

T Corresponding author. Fax: +1 757 446 5719. E-mail address: [email protected] (L.D. Sanford). 0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2005.02.079

(REM) and REM-related phenomena (reviewed in [44]). Electrolytic (e.g., [10,15]) and chemical lesions [47] that produce extensive damage to RPO in cats have been reported to reduce REM significantly or even eliminate it, leading to the suggestion that this region is essential for REM generation (reviewed in [44]). In addition, RPO in cats contains sites in which microinjections of the cholinergic agonist, carbachol, induce REM (e.g., [2–5,14,45]). Microinjections of hypocretin 1 and 2 [51], vasoactive intestinal

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50

peptide [8], adenosinergic agonists [27,28] and neurotrophin3 [52] have also been reported to enhance or induce REM when administered into RPO. A significant amount of recent work in RPO has focused on GABAergic regulation of arousal and REM. In cats, unilateral microinjections of both g-aminobutyric acid (GABA) and the GABAA agonist, muscimol (MUS), into RPO increased wakefulness at sites where carbachol induced REM, whereas microinjections of the GABAA antagonist, bicuculline (BIC), induced long-duration REM episodes that could occur without intervening non-REM (NREM) [48–50]. Increases in REM occurred in a dosedependent manner at concentrations of BIC V 10 mM although in some cats microinjections of higher concentrations induced wakefulness and bhyperexcitationQ [50]. In rats, we found that bilateral microinjections of lower concentrations of MUS and BIC decreased and increased REM, respectively, but the effect was much subtler than that reported in cats [40]. GABA in RPO may regulate REM by inhibiting acetylcholine release [46]. Another pontine region that has received considerable recent interest due to its apparent role in generating REM and REM-related phenomena is the nucleus subcoeruleus (SubC) [6,35]. Early work reported that bilateral lesions of the locus coeruleus and subcoeruleus in rats eliminated REM without significantly affecting NREM [36]. In contrast, lesions of the locus coeruleus alone resulted in disrupted sleep for up to 2 days followed by a return to normal [36]. More recently, the unilateral application of relatively high concentrations of BIC (8–10 mM) into the subcoeruleus/sublaterodorsal region by microiontophoresis in head-fixed [6] and by microinjection in freely moving [35] rats produced significant reductions in REM latency and increases in REM amounts. Triggering neurons for the generation of ponto-geniculo-occipital (PGO) waves in rats, one of the signature characteristics of REM, have been localized in the SubC [11]. In recent studies in rats, the application of carbachol into SubC has been found to enhance PGO wave density [30], whereas local microinjections of serotonin into SubC suppressed PGO wave activity without altering REM [12]. Additional evidence linking RPO and SubC to the generation of REM has come from studies on REM without atonia (REM-A), a phenomenon in which the electrophysiological features of REM are normal but skeletal muscle atonia is absent allowing overt movements and elaborate behaviors including orienting and even quadrupedal locomotion. In cats, behavior resembling predatory attack resulted from lesions extending into the ventral midbrain [16]. REM-A is defined by the presence of behavior and not just the presence of nuchal muscle tone [32]. It has been observed after electrolytic lesions in RPO and SubC in cats [16,17,19,38,43] and rats [31,33,39]. However, a critical limitation of electrolytic lesions is that both cell bodies and fibers of passage are damaged, and the neural reorganization and adaptation that occurs post-lesion

43

can complicate interpretation of the effects that the actual damage produces. Methods of temporarily inactivating or reversibly blesioningQ brain regions permit examination of the immediate effects of blocking neural activity. Tetrodotoxin (TTX) is a sodium channel blocker that temporarily inactivates both neurons and tracts for up to 12 h and potentially longer [29]. Inactivation of a brain region by TTX is thought to be equivalent functionally to that produced by an electrolytic lesion but allows studies to be conducted without allowing time for neural adaptation [29]. In this study, we examined the influence of microinjections of TTX into RPO and SubC on sleep in freely behaving rats in order to determine the effect of inactivating neurons and tracts in these regions without the potential confound produced by neural adaptation.

2. Methods The subjects were 21 male Sprague–Dawley rats of approximately 90 days of age at the time of surgery. The rats were given ad libitum access to food and water. The recording room was kept on a 12:12 light–dark cycle with lights on from 07:00 to 19:00 h, and ambient temperature was maintained at 24.5 F 0.5 8C. The rats were implanted with skull screws for recording the electroencephalogram (EEG) and with stainless steel wire electrodes sutured to the dorsal neck musculature for recording the electromyogram (EMG). Leads from the recording electrodes were routed to a nine-pin miniature plug that mated to one attached to a recording cable. Guide cannulae (26 ga.) for microinjections were bilaterally implanted with their tips aimed 1.0 mm above RPO (AP 8.3, ML F 1.1, DV 6.5) or SubC (AP 9.5, ML, F 1.5, DV 6.8) according to coordinates in the atlas of Kruger et al. [20]. The recording plug and cannulae were affixed to the skull with dental acrylic and anchor screws. All surgical procedures were performed stereotaxically under aseptic conditions. The rats were anesthetized with isoflurane (5% induction; 2% maintenance). Buprenorphine (0.5 mg/kg) was administered for potential postoperative pain. The rats were allowed a minimum of 14 days to recover prior to beginning the experiment. All procedures were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Experimental Animals and were approved by Eastern Virginia Medical School’s Animal Care and Use Committee (Protocol # 02-029). The animals were habituated to the handling procedures and recording chamber over the course of four recording sessions prior to receiving any drug microinjections. During the first two sessions, the rats were placed in the recording chamber, connected to the cable and left undisturbed for 22 h, as they would be for the experimental recording sessions. For the next two sessions, the

44

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50

animals received pre-recording handling and restraint just as they would for the experimental sessions. Then, they were connected to the cable for 22 h as during the initial cable habituation days. Following habituation sessions, the rats received microinjections of the various concentrations of each drug and the same volume of saline control in a counterbalanced order. For microinjections, injection cannulae (33 ga.), which projected 1 mm beyond the tip of the guide cannulae, were secured in place within the guide cannulae. The injection cannulae were connected to lengths of polyethylene tubing that in turn were connected to 1.0 Al Hamilton syringes. The injection cannulae and tubing were prefilled with the solution to be injected. In an attempt to approximate different types and sizes of lesions, we examined the effects of unilateral and bilateral microinjections and of different volumes of the solution containing TTX. Thus, each animal received one unilateral microinjection containing TTX in a higher volume (TTXUH: 5.0 ng/0.2 Al) and two bilateral microinjections containing TTX in a lower and higher volume (TTXBL: 2.5 ng/0.1 Al; TTXBH: 5.0 ng/0.2 Al). Microinjections of appropriately matching volumes of saline alone (SAL) were given unilaterally or bilaterally as controls for each TTX microinjection. At least 7 days elapsed between injections. The injections were made 2 h following lights on. The solutions were slowly infused over 2 min (0.1 Al/min), and the injection cannulae were allowed to stay in place for 1.0 min after the microinjection was finished. After receiving the injections, the rats were placed in the sleep recording chambers and connected to the cables. Polygraphic studies were conducted for the subsequent 22 h. Each rat was studied in its home cage, which was placed into a sound-attenuating chamber. For electrophysiological recording, a lightweight, shielded recording cable was connected to the plug on the rat’s head. The cable was attached to a spring-mounted swivel that permitted free movement of the rat within its cage. Sleep was visually scored from computerized records in 10 s epochs. Trained observers visually determined wakefulness, NREM, and REM using standard electrographic criteria [40]. Wakefulness was scored based on the presence of low-voltage, fast EEG; high amplitude, tonic EMG level; and phasic EMG bursts that could be associated with gross body movements. NREM was scored based on the presence of spindles interspersed with slow waves, lower muscle tone and no gross body movements or EEG desynchronization. For scoring REM, onset was considered to have occurred immediately following the last sleep spindle of NREM that appeared in conjunction with decreasing or fully relaxed muscle tone. Afterward, REM was scored continuously during the presence of low voltage, fast EEG, theta rhythm and muscle atonia. The records were also visually examined for the presence of REM EEG features in association with muscle tone and movement artifacts, which would have indicated the possible occurrence of REM-A. However, this was never observed.

Sleep in the total recording period and from the 10-h light and the 12-h dark periods was examined. The measures were: total NREM, total REM and total sleep times (min); number of NREM and REM episodes; and REM percentage (total REM / total sleep  100). Analyses of the hourly distributions of NREM and REM during the total 22-h recording period and during separate 10-h light or 12-h dark periods were made with the twosample Kolmogorov–Smirnov (K–S) test, which determines whether there are significant differences between cumulative frequency distributions. Analyses of the totals for the 22-h total recording period and light and dark period totals were made with within-subjects ANOVAs. When appropriate, comparisons were made between each concentration and its appropriate saline control using Tukey’s tests. The criterion for significance was P b 0.05. Upon completion of the experiment, the rats were overdosed with sodium pentobarbital (150 mg/kg intraperitoneally) and perfused intracardially with 0.9% saline and 10% formalin. The brains were processed to determine cannula placements. For this purpose, 40 Am slices were made through the areas of interest with a cryostat, and the sections were stained with cresyl violet. Injection sites in RPO, SubC or outside either region were verified by comparing sections to those in the stereotaxic atlas [20]. Injection sites verified to be in RPO (n = 9) and in SubC (n = 8) were used in the analyses for each region. Injections verified as located outside either RPO or SubC (n = 4) were grouped together and analyzed separately.

3. Results 3.1. Control microinjections of saline into RPO and SubC Because we performed multiple microinjections into the same sites and were concerned that sleep could be altered across time, we conducted a separate analysis of the three saline microinjections that were performed as controls for each microinjection of TTX. There were no significant differences in NREM or REM, when considered as totals for the light and dark periods, after microinjections of saline into RPO or SubC. We also conducted K–S tests for the hourly data for each period (not shown) and found no significant differences in the distributions in any analysis period. This indicated that sleep returned to control levels after each microinjection of TTX. Therefore, we averaged across the 3 saline days in each region and used the mean for comparison with the different microinjections of TTX. 3.2. Hourly sleep patterns after microinjections of TTX into RPO Fig. 1 demonstrates the effects of unilateral (TTXUH) and bilateral (TTXBH) microinjections of the high volume of

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50

45

Fig. 1. Total sleep, NREM and REM plotted hourly across 22 h of recording after microinjection of saline (SAL) and TTXUH and TTXBH into RPO. The distributions of NREM and REM for total, light and dark recording periods for each TTX condition were compared to SAL with the K–S test (see Table 1). Error bars indicate SEM.

TTX into RPO on the hourly distributions of NREM and REM. The pattern of hourly sleep after TTXBL was similar to that after TTXBH and, to simplify the graph, the former is not plotted. However, the results for TTXBL are included in all analyses and in the graphs indicating other sleep parameters. Both unilateral and bilateral microinjections of TTX into RPO produced reductions in light period sleep followed by increases in dark period sleep. Analyses of these data with the K–S test (Table 1) indicated that the distribution of NREM in the light periods had been altered by microinjection of TTX into RPO for each variable. By comparison, the distribution of dark period NREM was altered only in TTXBH. The distribution of REM differed from control in each light, dark and total 22-h recording period except for the 22-h total after TTXUH. 3.3. Sleep parameters after microinjections of TTX into RPO 3.3.1. Total sleep and NREM The ANOVAs for comparisons of total sleep were significant for the light period, F(3,35) = 3.68, P b .026, and the dark period, F(3,35) = 3.88, P b .022. Post hoc comparisons found decreased light period total sleep and increased dark period total sleep after TTXBH (Table 2). No other analyses for total sleep were significant.

Table 1 Results of K–S test comparisons of hourly distributions after each microinjection of TTX into RPO NREM

REM

Total Light Dark Total Light Dark

TTXUH

TTXBL

TTXBH

ns 0.031 ns ns 0.007 0.005

ns 0.031 ns 0.014 0.0001 0.0001

ns 0.007 0.019 0.014 0.001 0.0001

All comparisons are to hourly distributions after microinjections of SAL. TTXUH and TTXBH are presented in Fig. 1. Total: 22-h recording period; Light: 10 h; Dark: 12 h. ns, not significant ( P N 0.05).

There were corresponding significant ANOVAs for NREM during the light period, F(3,35) = 3.68, P b .026, and the dark period, F(3,35) = 3.85, P b .022. These resulted from decreased light period total sleep and increased dark period total sleep after TTXBH (Table 2). The ANOVA for latency to NREM was significant, F(3, 35) = 3.41, P b .034, with significantly increased latency after TTXUH and TTXBH compared to saline (Table 2). No other analyses for total NREM were significant, and there were no significant differences found in the analysis of NREM episodes.

Table 2 Representative sleep parameters in total 22-h recording period, 10-h light period and 12-h dark period after microinjection of saline (SAL) and TTXUH, TTXBL and TTXBH into RPO SAL Mean (SEM)

TTXUH Mean (SEM)

Total sleep (min) T 691.5 (17.4) 629.2 (33.0) L 389.6 (6.2) 308.7 (43.0) D 301.8 (13.3) 320.5 (20.2) Total NREM (min) T 588.5 (16.8) 530.5 (34.3) L 343.7 (7.0) 279.9 (37.4) D 244.8 (12.6) 250.6 (14.5) NREM episodes (number) T 182.4 (6.8) 169.0 (10.4) L 93.6 (4.9) 77.7 (9.9) D 88.8 (3.6) 91.3 (5.4) Total REM (min) T 103.0 (5.7) 98.7 (8.8) L 46.0 (2.5) 28.8 (7.3) D 57.0 (4.7) 69.9 (9.3) REM episodes (number) T 53.1 (2.1) 55.2 (4.4) L 22.9 (1.1) 15.6 (3.6) D 30.2 (1.7) 39.7 (4.7) REM percentage (%) T 14.9 (0.8) 16.1 (1.6) L 11.8 (0.7) 7.9 (1.8) D 19.0 (1.5) 21.5 (2.2) Latency (min) NREM 42.2 (5.6) 110.8 (27.7)* REM 84.6 (10.1) 235.3 (48.9)**

TTXBL Mean (SEM)

TTXBH Mean (SEM)

648.5 (32.6) 277.5 (30.7) 371.0 (23.9)

621.4 (23.7) 228.5 (35.6)** 392.9 (27.5)*

547.7 (28.9) 262.3 (27.9) 285.5 (16.1)

520.1 (20.8) 211.8 (32.0)* 308.3 (22.0)*

158.0 (5.6) 73.8 (5.1) 84.2 (2.3)

156.8 (10.1) 67.1 (9.5) 89.7 (5.8)

100.7 (7.6) 15.2 (5.5)** 85.5 (9.4)**

101.3 (11.6) 16.7 (5.4)** 84.6 (10.4)**

49.9 (3.3) 9.3 (3.1)** 40.6 (3.2)*

48.9 (3.8) 9.7 (2.6)** 39.2 (3.9)

15.6 (1.0) 4.8 (1.6)** 22.6 (1.6)

16.2 (1.6) 6.0 (1.7)* 21.4 (2.4)

90.7 (18.7) 110.9 (22.8)* 373.3 (70.9)** 416.6 (66.1)***

Comparisons between conditions were made with Tukey’s tests (significant differences relative to SAL: *P b 0.05; **P b 0.01).

46

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50

3.3.2. REM parameters The ANOVAs for comparisons of total REM were significant for the light period, F(3,35) = 7.85, P b .0008, and the dark period, F(3,35) = 6.47, P b .002. Post hoc comparisons found decreased light period total REM and increased dark period total REM after TTXBL and TTXBH (Table 2). The number of REM episodes was also changed in the light period F(3,35) = 6.10, P b .003, and the dark period, F(3,35) = 3.91, P b .021. A reduction in the number of REM episodes was found in the light period after TTXBL and TTXBH. This was followed by a significantly increased number of REM episodes in the dark period after TTXBL, but the increase after TTXBH did not reach significance. TTX also produced significant changes in light period REM percentage, F(3,35) = 5.27, P b .006, with significant reductions after both TTXBL and TTXBH. The ANOVA for latency to REM was significant, F(3, 35) = 9.11, P b .0003, with each dosage of TTX significantly increasing REM latency compared to saline (Table 2). 3.4. Hourly sleep patterns after microinjections of TTX into SubC Fig. 2 presents hourly plots of total NREM and REM after microinjections of saline, TTXUH and TTXBH into SubC. TTXBL is not shown. Analyses of the hourly data with the K–S test (Table 3) indicated that the distribution of NREM was significantly altered only during the dark period after TTXBL and TTXBH. By comparison, the distribution of REM was significantly altered during the light period after each microinjection of TTX into SubC. Significant alterations were also seen in the 22-h distribution of total REM after TTXBL and in the dark period distribution of total REM after TTXBH. 3.5. Sleep parameters after microinjections of TTX into SubC 3.5.1. Total sleep and NREM The ANOVAs for comparisons of total sleep were significant only for the dark period, F(3,31) = 5.95, P b

Table 3 Results of K–S test comparisons of hourly distributions after each microinjection of TTX into SubC NREM

REM

Total Light Dark Total Light Dark

TTXUH

TTXBL

TTXBH

ns ns ns ns 0.031 ns

ns ns 0.019 0.014 0.0001 ns

ns ns 0.001 ns 0.007 0.005

All comparisons are to hourly distributions after microinjections of SAL. TTXUH and TTXBH are presented in Fig. 3. Total: 22-h recording period; Light: 10 h; Dark: 12 h. ns, not significant ( P N 0.05).

.004. Post hoc comparisons found increased dark period total sleep after TTXBL and TTXBH (Table 4). No other analyses for total sleep were significant. There were no significant alterations in the analyses for total NREM, number of NREM episodes or latency to NREM. 3.5.2. REM parameters The ANOVAs for comparisons of total REM were significant for the light period, F(3,31) = 9.19, P b .004, and the dark period, F(3,31) = 7.14, P b .017. Light period total REM was decreased after TTXBL and TTXBH, and dark period total REM was increased after TTXBH (Table 4). The number of REM episodes was also changed in the light period F(3,31) = 8.48, P b .007, and the dark period, F(3,31) = 3.45, P b .034. There was a reduction in the number of REM episodes during the light period after TTXBL and TTXBH and an increase in the dark period after TTXBH. TTX produced significant changes in light period REM percentage, F(3,31) = 10.84, P b .0002, with significant reductions after both TTXBL and TTXBH. For each REM parameter, TTXBL values were also significantly reduced compared to those of TTXUH (which did not differ compared to saline), but not compared to those of TTXBH. The ANOVA for latency to REM was significant, F(3, 31) = 10.36, P b .0002, with each dosage of TTX significantly increasing REM latency compared to saline (Table 4). The latency to REM was also significantly greater after TTXBL compared to TXXUH.

Fig. 2. Total sleep, NREM and REM plotted hourly across 22 h of recording after microinjection of saline (SAL) and TTXUH and TTXBH into SubC. The distributions of NREM and REM for total, light and dark recording periods for each TTX condition were compared to SAL with the K–S test (see Table 2). Error bars indicate SEM.

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50 Table 4 Representative sleep parameters in total 22-h recording period, 10-h light period and 12-h dark period after microinjection of saline (SAL) and TTXUH, TTXBL and TTXBH into SubC SAL Mean (SEM)

TTXUH Mean TTXBL Mean (SEM) (SEM)

Total sleep (min) T 647.0 (36.1) 655.5 L 386.5 (17.3) 359.6 D 260.6 (23.8) 295.9 Total NREM (min) T 540.9 (31.6) 548.8 L 333.2 (16.5) 323.7 D 207.7 (19.1) 225.0 NREM episodes (number) T 177.5 (4.0) 178.8 L 90.0 (3.6) 89.3 D 87.5 (4.5) 89.5 Total REM (min) T 106.2 (6.8) 106.7 L 53.3 (3.9) 35.9 D 52.9 (6.1) 70.9 REM episodes (number) T 51.2 (4.5) 57.8 L 23.1 (2.8) 18.9 D 28.1 (2.5) 39.1 REM percentage (%) T 16.5 (0.7) 16.5 L 13.9 (1.1) 9.8 D 20.4 (1.4) 22.8 Latency (min) NREM 34.7 (3.4) 44.7 REM 80.3 (10.5) 214.7

TTXBH Mean (SEM)

(27.7) (20.6) (29.6)

666.5 (29.2) 327.7 (28.9) 338.8 (25.5)*

689.9 (36.0) 321.1 (31.0) 368.8 (20.3)**

(27.7) (18.0) (28.1)

584.0 (26.1) 314.6 (24.0) 269.5 (21.4)

569.9 (27.9) 294.6 (24.8) 275.3 (15.1)

(9.9) (8.2) (6.1)

176.1 (10.2) 84.5 (6.6) 91.6 (5.1)

165.5 (8.1) 81.8 (4.3) 83.7 (7.2)

(3.9) (4.4)# (3.3)

82.5 (12.3) 13.2 (6.7)*** 69.3 (11.1)

120.0 (11.0) 26.6 (7.0)* 93.5 (8.9)***

(2.7) (2.3)# (3.2)

42.3 (6.2) 6.3 (2.8)*** 36.0 (5.7)

54.8 (2.0) 12.9 (2.5)* 41.9 (1.8)*

(0.9) (0.9)# (1.7)

12.3 (1.7) 3.2 (1.6)*** 20.3 (2.9)

17.2 (1.1) 7.5 (1.5)* 25.1 (1.8)

47

tion of RPO decreased sleep with significant reductions in both NREM and REM. In contrast, bilateral TTX inactivation of SubC significantly reduced REM in the light period but decreased NREM non-significantly. In the dark period, however, both low and high volumes of TTX resulted in increases in both REM and NREM. Unilateral application of TTX in RPO and SubC did not significantly alter sleep when light period totals were considered; however, unilateral microinjection into RPO, but not SubC, significantly altered the distribution of light period NREM. Our findings are consistent with suggestions that RPO is important for the maintenance of overall arousal state, not just REM [48–50], and are generally consistent with a single early report that lesions including SubC can reduce/ eliminate REM without significantly impacting NREM [36]. Thus, RPO may have a broader role in the regulation of arousal state, whereas SubC may function more specifically in the regulation of REM and related phenomena, such as the PGO wave [11,12]. Interestingly, total sleep after TTX application in both RPO and SubC did not significantly differ from baseline when the entire recording period (22 h) was considered,

(8.9) 40.1 (3.6) 67.9 (19.3) (59.5)* 393.5 348.6 (52.4)** 10.11(56.3)***

Comparisons between conditions were made with Tukey’s tests (significant differences relative to SAL: *P b 0.05; **P b 0.01; ***P b .001; significant differences between TTXUH and TTXBL: #P b 0.05).

3.6. Microinjection sites Fig. 3 illustrates the histological location of the microinjection sites in each animal. Sites indicated by a dark square were located inside the target area (RPO, n = 9; SubC, n = 8) and were included in the analysis. Sites indicated by a dark triangle were located outside the target areas (n = 4). These animals were grouped together and were analyzed separately using within-subjects ANOVAs. There were no significant differences in total NREM or total REM when considered as 22-h totals or as separate 10-h and 12-h dark periods.

4. Discussion 4.1. Sleep after inactivation of RPO and SubC This study complements those that have implicated RPO and SubC in the regulation of arousal state. Inactivation of both regions with TTX induced significant decreases in REM. However, inactivating each region had differing effects on NREM and wakefulness. Bilateral TTX inactiva-

Fig. 3. Line drawings illustrating microinjection sites in animals receiving TTX. Squares indicate sites in RPO or SubC in animals used in the main analyses. Sites indicated by triangles were outside RPO and SubC and were considered separately.

48

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50

even though there could be significant alterations in the light–dark distributions of most sleep parameters. The recordings were characterized by reductions in light period sleep and greater sleep during the dark period. This suggests that the TTX blockage of REM generation and other reductions in sleep during the light period were homeostatically compensated for with recovery sleep occurring during the dark period after the effects of TTX had dissipated. 4.2. Methods of inactivation Studies using both permanent and reversible lesion techniques have provided considerable insight into the neural mechanisms underlying sleep and wakefulness. Permanent lesions have been created with electrolytic and radio frequency lesions, which destroy both cell bodies and fibers of passage, and with chemical techniques, including the application of neurotoxic agents, such as kainic acid and ibotenic acid, both more sparing of fibers of passage (reviewed in [22]). In addition to the issues of neural reorganization and adaptation described earlier, another issue with permanent lesions is the potential for debilities produced by the damage. For example, in addition to changes in sleep, bilateral destruction of the locus coeruleus produced a urogenital syndrome consisting of hematuria, bladder distension and penile erection, and lesions in both locus coeruleus and SubC produced an aphagia–adipsia syndrome in addition to the urogenital syndrome [36]. Physical debilities also can be produced by reversible lesions, but the symptoms should subside as neural function recovers. In this study, we used TTX to inactivate reversibly two regions linked to the generation of REM. This technique has widespread use in examining neural function [53]. Local anesthetics (e.g., lidocaine, tetracaine) have been used as well [1,13,22,26,29,37]. Both TTX and lidocaine block voltage-dependent sodium channels, so that they temporarily mimic the condition existing after electrolytic lesions. However, lidocaine (1 ml of 4% lidocaine) produces a short-lasting blockade of about 20 min [22,26,29,37]. The maximum effective time for lidocaine administered via microinjection may be 3–4 h [9] although the effective time is usually on the order of 5–45 min, depending on the volume of the injection [22,26,29]. In contrast, the effect of TTX blockade (10 ng in 1 Al) is maximal 30–120 min after administration, decays exponentially and completely disappears within 24 h [53]. The extent of lesions produced by microinjections of TTX with a given volume has also been described. Zhuravin and Bures [53] examined pupillary diameter in anesthetized rats before and up to 24 h after injection of TTX at various distances from the Edinger–Westphal nucleus and found that 10 ng of TTX in a volume of 1 Al produced a functional lesion with a diameter of 3.0 mm. The diffusion radius constancy of microinjected TTX has been used to estimate the size of lesions produced by different volumes [23].

Based on published studies [23,53], we estimate that our low and high dosages produced lesions with diameters of approximately 0.3 mm and 0.6 mm, respectively. By comparison, at the injection sites, the width of RPO is approximately 1.1 to 1.2 mm and the width of SubC is approximately 0.5 to 0.6 mm [20]. However, because the region of inactivation also is influenced by TTX binding to sodium channels across the diffusion gradient [53], the higher concentration of TTX we used could have produced larger effective lesions. In either case, a relatively large area of RPO and virtually all of SubC would have been inactivated with the larger dosage. Microinjection studies in the two regions we examined have generally used only unilateral application in both cats (e.g., [2–5,14,45,53]) and rats (e.g., [7,8,27]). Thus, it would be important to know if significant differences between bilateral and unilateral inactivation exist. Our attempt was moderately successful in that we did observe differences in the effects of different bilateral volumes microinjected into RPO on the amount of total sleep and total NREM. We also saw differences in the effects of unilateral vs. bilateral microinjections in both SubC and RPO. However, the effects of both volumes of the TTX solution microinjected bilaterally into RPO were similar with respect to the magnitude of effects on REM. In SubC, both bilateral volumes produced a significant reduction in REM compared to control; however, only the low volume produced a significant reduction compared to the unilateral microinjection of TTX. This result may have been due to effects on adjacent regions with the larger volume, and to greater specificity to the REM-altering effect of shutting down SubC with the smaller volume. Various other methods have been used to inactivate regions temporarily in sleep research and other disciplines. GABA receptors are widespread throughout the brain, and temporary pharmacological inactivation of a neural region is exemplified by the use of the inhibitory neurotransmitter GABA and its agonists (reviewed in [18,22,26,29]. These can produce inactivation of cell bodies for extended periods while sparing conduction in fibers of passage [18,26]. Studies [48–50] microinjecting GABA and GABA agonists such as muscimol into RPO in cats found alterations in arousal similar to those we found with TTX microinjected into RPO in rats. Most GABAergic neurons are involved in local circuitry [34], and GABAergic interneurons have been suggested as a potential source of inhibitory regulation for RPO [25,48,49]. If true, the effects of TTX and muscimol could involve functional inactivation of the same local neurons and local fibers involved in regulating arousal in RPO. 4.3. Potential relevance for REM-A REM-A in animals was instrumental in the identification of REM behavior disorder (RBD) in humans. Both involve a release of behavior in REM [42]. RBD is characterized by

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50

bvigorous and frequently violent dream-enacting behaviorsQ and is most often observed in older men [24,41]. RBD often occurs in conjunction with the onset of narcolepsy, and neurodegenerative, cerebrovascular or other neurologic disorders [24,41], though interestingly, up to 55% of RBD may be idiopathic with no identifiable neuropathology. This observation prompted Mahowald et al. [24,41] to suggest that idiopathic RBD may result from subtle changes in the brain that compromise descending inhibition to the brainstem and that structural damage to the pons, such as produced by lesions producing REM-A in animals, may rarely be the cause of RBD in humans [24,41]. However, over time, roughly 50% of those patients diagnosed with idiopathic RBD have later developed various neurodegenerative diseases [24]. In rats, electrolytic lesions of RPO can produce REM-A [31,33,37], and SubC may be an important region in producing REM-A as well, as demonstrated in cats [43]. Relatively small unilateral lesions were sufficient to eliminate the atonia of REM in rats [37], whereas larger bilateral lesions in either RPO or the adjacent RPC could lead to behavioral release during REM-A [31,33,37]. The present results indicate that functional inactivation in these regions is not sufficient to produce REM-A. One potential caveat is that the TTX lesion in RPO would not be complete. However, complete lesions of RPO are not necessary to produce REM-A in rats [39], and it is difficult to imagine that inactivating a larger area of a region that is important for REM generation would have resulted in REMA. A recent study in humans also found reduced REM associated with lesions in brainstem regions implicated in REM generation [21]. Although limited in scope, the results suggest that physical damage alone does not necessarily lead to REM-A or RBD in humans. The fact that RBD is often associated with diseases that are characterized by widespread neural atrophy that occurs over time and is rarely associated with specific lesions may provide a clue. In animals, also, post-lesion REM-A often fully manifests over time [16] (though its expression is more rapid in rats than in cats [39]). This suggests that neural reorganization and adaptation that occur in response to damage and consequent altered functionality in brainstem circuits controlling atonia may play a role in REM-A induced by electrolytic lesions as well, although other lesion sites that induce REM-A need to be examined.

5. Conclusion The time course of reversible lesions produced by TTX is long enough to allow the examination of the effects of inactivating specific regions on sleep and wakefulness. In this study, the results suggest that inactivation of RPO induces significant decreases in both NREM and REM. In contrast, it is possible to inactivate SubC and produce a relatively specific reduction in REM without significant alterations in

49

NREM sleep. The present data support findings from studies using permanent lesion techniques indicating that these regions have roles in the generation of REM. A significant difference between the effects of TTX inactivation and electrolytic lesions is that TTX lesions failed to produce behavioral release in REM. This suggests that the process of reorganization and adaptation that can occur post-lesion may play a role in the development of REM-A. This suggestion is consistent with the notion that RBD in humans develops over time in conjunction with neurodegenerative diseases.

Acknowledgments This work was supported by NIH research grants MH64827 and NS36694. We thank Jihua Xiao, Brian Parris and Isabel Liu for their assistance in conducting this experiment.

References [1] D.J. Albert, F.J. Madryga, An examination of the functionally effective spread of 4 microliters of slowly infused lidocaine, Behav. Neural Biol. 29 (1980) 378 – 384. [2] H.A. Baghdoyan, A.P. Monaco, M.L. Rodrigo-Angulo, F. Assens, R.W. McCarley, J.A. Hobson, Microinjection of neostigmine into the pontine reticular formation of cats enhances desynchronized sleep signs, J. Pharmacol. Exp. Ther. 231 (1984) 173 – 180. [3] H.A. Baghdoyan, M.L. Rodrigo-Angulo, R.W. McCarley, J.A. Hobson, Site-specific enhancement and suppression of desynchronized sleep signs following cholinergic stimulation of three brainstem regions, Brain Res. 306 (1984) 39 – 52. [4] H.A. Baghdoyan, M.L. Rodrigo-Angulo, R.W. McCarley, J.A. Hobson, A neuroanatomical gradient in the pontine tegmentum for the cholinoceptive induction of desynchronized sleep signs, Brain Res. 414 (1987) 245 – 261. [5] H.A. Baghdoyan, R. Lydic, C.W. Callaway, J.A. Hobson, The carbachol-induced enhancement of desynchronized sleep signs is dose dependent and antagonized by centrally administered atropine, Neuropsychopharmacology 2 (1989) 67 – 79. [6] R. Boissard, D. Gervasoni, M.H. Schmidt, B. Barbagli, P. Fort, P.H. Luppi, The rat ponto-medullary network responsible for paradoxical sleep onset and maintenance: a combined microinjection and functional neuroanatomical study, Eur. J. Neurosci. 16 (2002) 1959 – 1973. [7] P. Bourgin, P. Escourrou, C. Gaultier, J. Adrien, Induction of rapid eye movement sleep by carbachol infusion into the pontine reticular formation in the rat, NeuroReport 6 (1995) 532 – 536. [8] P. Bourgin, A. Ahnaou, A.M. Laporte, M. Hamon, J. Adrien, Rapid eye movement sleep induction by vasoactive intestinal peptide infused into the oral pontine tegmentum of the rat may involve muscarinic receptors, Neuroscience 89 (1999) 291 – 302. [9] M.J. Cambiasso, E. Chiaraviglio, The involvement of the hypothalamic preoptic area on the regulation of thirst in the rat, Behav. Neural Biol. 58 (1992) 190 – 195. [10] G. Carli, A. Zanchetti, A study of pontine lesions suppressing deep sleep in the cat, Arch. Ital. Biol. 103 (1965) 751 – 788. [11] S. Datta, D.F. Siwek, E.H. Patterson, P.B. Cipolloni, Localization of pontine PGO wave generation sites and their anatomical projections in the rat, Synapse 30 (1998) 409 – 423. [12] S. Datta, V. Mavanji, E.H. Patterson, J. Ulloor, Regulation of rapid eye

50

[13] [14]

[15]

[16]

[17]

[18]

[19] [20] [21] [22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

[30]

[31]

[32] [33]

L.D. Sanford et al. / Brain Research 1044 (2005) 42–50 movement sleep in the freely moving rat: local microinjection of serotonin, norepinephrine, and adenosine into the brainstem, Sleep 26 (2003) 513 – 520. C. Flicker, M.A. Geyer, Behavior during hippocampal microinfusions: III. Lidocaine versus picrotoxin, Brain Res. 257 (1982) 129 – 136. M. Garzon, I. De Andres, F. Reinoso-Suarez, Sleep patterns after carbachol delivery in the ventral oral pontine tegmentum of the cat, Neuroscience 83 (1998) 1137 – 1144. E. Gutierrez-Rivas, I. de Andres, J. Gomez-Montoya, F. ReinosoSuarez, The influence of the rostropontine-ventrolateral region on the sleep–wakefulness cycle, Experientia 34 (1978) 61 – 62. J.C. Hendricks, A.R. Morrison, G.L. Mann, Different behaviors during paradoxical sleep without atonia depend on pontine lesion site, Brain Res. 239 (1982) 81 – 105. K. Henley, A.R. Morrison, A re-evaluation of the effects of lesions of the pontine tegmentum and locus coeruleus on phenomena of paradoxical sleep in the cat, Acta Neurobiol. Exp. (Wars) 34 (1974) 215 – 232. J.M. Hupe, G. Chouvet, J. Bullier, Spatial and temporal parameters of cortical inactivation by GABA, J. Neurosci. Methods 86 (1999) 129 – 143. M. Jouvet, F. Delorme, Locus coeruleus et sommeil paradoxal, C. R. Soc. Biol., Paris 159 (1965) 895 – 899. L. Kruger, S. Saporta, L. Swanson, Photographic Atlas of the Rat Brain, Cambridge Univ. Press, New York, 1995. M.E. Landau, J.Y. Maldonado, B. Jabbari, The effects of isolated brainstem lesions on human REM sleep, Sleep Med. 6 (2005) 37 – 40. S.G. Lomber, The advantages and limitations of permanent or reversible deactivation techniques in the assessment of neural function, J. Neurosci. Methods 86 (1999) 109 – 117. C.A. Lorenzini, E. Baldi, C. Bucherelli, G. Tassoni, Time-dependent deficits of rat’s memory consolidation induced by tetrodotoxin injections into the caudate–putamen, nucleus accumbens, and globus pallidus, Neurobiol. Learn. Mem. 63 (1995) 87 – 93. M. Mahowald, C. Schenck, REM sleep behavior disorder, in: M. Kryger, T. Roth, W. Dement (Eds.), Principles and Practice of Sleep Medicine, W.B. Saunders, Philadelphia, 1994, pp. 574 – 588. K.J. Maloney, L. Mainville, B.E. Jones, c-Fos expression in GABAergic, serotonergic, and other neurons of the pontomedullary reticular formation and raphe after paradoxical sleep deprivation and recovery, J. Neurosci. 20 (2000) 4669 – 4679. J.G. Malpeli, Reversible inactivation of subcortical sites by drug injection, J. Neurosci. Methods 86 (1999) 119 – 128. G.A. Marks, C.G. Birabil, Enhancement of rapid eye movement sleep in the rat by cholinergic and adenosinergic agonists infused into the pontine reticular formation, Neuroscience 86 (1998) 29 – 37. G.A. Marks, J.P. Shaffery, S.G. Speciale, C.G. Birabil, Enhancement of rapid eye movement sleep in the rat by actions at A1 and A2a adenosine receptor subtypes with a differential sensitivity to atropine, Neuroscience 116 (2003) 913 – 920. J.H. Martin, C. Ghez, Pharmacological inactivation in the analysis of the central control of movement, J. Neurosci. Methods 86 (1999) 145 – 159. V. Mavanji, S. Datta, Activation of the phasic pontine-wave generator enhances improvement of learning performance: a mechanism for sleep-dependent plasticity, Eur. J. Neurosci. 17 (2003) 359 – 370. M. Mirmiran, bOneiricQ behavior during active sleep induced by bilateral lesions of the pontine tegmentum in juvenile rats, in: W. Koella (Ed.), Sleep: Sixth European Congress of Sleep Research, Karger, Basel, 1983, pp. 236 – 239. A.R. Morrison, The pathophysiology of REM-sleep behavior disorder (letters to the editor), Sleep 21 (1998) 446. J. Mouret, F. Delorme, M. Jouvet, Lesions of the pontine tegmentum and sleep in rats, C. R. Seances Soc. Biol. Fil. 161 (1967) 1603 – 1606.

[34] R.G. Paredes, A. Agmo, GABA and behavior: the role of receptor subtypes, Neurosci. Biobehav. Rev. 16 (1992) 145 – 170. [35] M.S. Pollock, R.E. Mistlberger, Rapid eye movement sleep induction by microinjection of the GABA-A antagonist bicuculline into the dorsal subcoeruleus area of the rat, Brain Res. 962 (2003) 68 – 77. [36] B. Roussel, J.F. Pujol, M. Jouvet, Effects of lesions in the pontine tegmentum on the sleep stages in the rat, Arch. Ital. Biol. 114 (1976) 188 – 209. [37] J. Sandkuhler, B. Maisch, M. Zimmermann, The use of local anaesthetic microinjections to identify central pathways: a quantitative evaluation of the time course and extent of the neuronal block, Exp. Brain Res. 68 (1987) 168 – 178. [38] L.D. Sanford, A.R. Morrison, G.L. Mann, J.S. Harris, L. Yoo, R.J. Ross, Sleep patterning and behaviour in cats with pontine lesions creating REM without atonia, J. Sleep Res. 3 (1994) 233 – 240. [39] L. Sanford, C. Cheng, X. Tang, A. Silvestri, G. Mann, R. Ross, A.R. Morrison, Sleep and behavior in rats with pontine lesions producing REM without atonia [online], Sleep Res. Online 4 (2001) 1 – 5. [40] L.D. Sanford, X. Tang, J. Xiao, R.J. Ross, A.R. Morrison, GABAergic regulation of REM sleep in reticularis pontis oralis and caudalis in rats, J. Neurophysiol. 90 (2003) 938 – 945. [41] C.H. Schenck, S.R. Bundlie, M.G. Ettinger, M.W. Mahowald, Chronic behavioral disorders of human REM sleep: a new category of parasomnia, Sleep 9 (1986) 293 – 308. [42] C.H. Schenck, S.R. Bundlie, M.W. Mahowald, Delayed emergence of a parkinsonian disorder in 38% of 29 older men initially diagnosed with idiopathic rapid eye movement sleep behaviour disorder, Neurology 46 (1996) 388 – 393. [43] M.N. Shouse, J.M. Siegel, Pontine regulation of REM sleep components in cats: integrity of the pedunculopontine tegmentum (PPT) is important for phasic events but unnecessary for atonia during REM sleep, Brain Res. 571 (1992) 50 – 63. [44] J.M. Siegel, Brainstem mechanisms generating REM sleep, in: M. Kryger, T. Roth, W. Dement (Eds.), Principles and Practices of Sleep Medicine, Saunders, Philadelphia, PA, 2000, pp. 112–133. [45] G. Vanni-Mercier, K. Sakai, J.S. Lin, M. Jouvet, Mapping of cholinoceptive brainstem structures responsible for the generation of paradoxical sleep in the cat, Arch. Ital. Biol. 127 (1989) 133 – 164. [46] J. Vazquez, H.A. Baghdoyan, GABAA receptors inhibit acetylcholine release in cat pontine reticular formation: implications for REM sleep regulation, J. Neurophysiol. (2004) 2198 – 2206. [47] H.H. Webster, B.E. Jones, Neurotoxic lesions of the dorsolateral pontomesencephalic tegmentum–cholinergic cell area in the cat: II. Effects upon sleep–waking states, Brain Res. 458 (1988) 285 – 302. [48] M.C. Xi, F.R. Morales, M.H. Chase, Evidence that wakefulness and REM sleep are controlled by a GABAergic pontine mechanism, J. Neurophysiol. 82 (1999) 2015 – 2019. [49] M.C. Xi, F.R. Morales, M.H. Chase, A GABAergic pontine reticular system is involved in the control of wakefulness and sleep, Sleep Res. Online 2 (1999) 43 – 48. [50] M.C. Xi, F.R. Morales, M.H. Chase, Induction of wakefulness and inhibition of active (REM) sleep by GABAergic processes in the nucleus pontis oralis, Arch. Ital. Biol. 139 (2001) 125 – 145. [51] M.C. Xi, S.J. Fung, J. Yamuy, F.R. Morales, M.H. Chase, Induction of active (REM) sleep and motor inhibition by hypocretin in the nucleus pontis oralis of the cat, J. Neurophysiol. 87 (2002) 2880 – 2888. [52] J. Yamuy, M.J. Rojas, P. Torterolo, S. Sampogna, M.H. Chase, Induction of rapid eye movement sleep by neurotrophin-3 and its colocalization with choline acetyltransferase in mesopontine neurons, Neuroscience 115 (2002) 85 – 95. [53] I.A. Zhuravin, J. Bures, Extent of the tetrodotoxin induced blockade examined by pupillary paralysis elicited by intracerebral injection of the drug, Exp. Brain Res. 83 (1991) 687 – 690.