- Email: [email protected]
Behavioural effects of dimethyl sulfoxide (DMSO): Changes in sleep architecture in rats
Toxicology Letters 157 (2005) 221–232
Behavioural effects of dimethyl sulfoxide (DMSO): Changes in sleep architecture in rats Mar´ıa Cavas ∗ , David Beltr´an, Jos´e F. Navarro Departamento de Psicobiolog´ıa y Metodolog´ıa de las Ciencias del Comportamiento, Facultad de Psicolog´ıa, Universidad de M´alaga, 29071 M´alaga, Spain Received 15 October 2004; received in revised form 16 February 2005; accepted 17 February 2005 Available online 15 April 2005
Abstract Dimethyl sulfoxide (DMSO) is an efficient solvent for water-insoluble compounds, widely used in biological studies and as a vehicle for drug therapy, but few data on its neurotoxic or behavioural effects is available. The aim of this work is to explore DMSO’s effects upon sleep/wake states. Twenty male rats were sterotaxically prepared for polysomnography. Four concentrations of DMSO (5%, 10%, 15%, and 20%, in saline) were examined. DMSO or saline were administered intraperitoneally at the beginning of the light period. Three hours of polygraphic recording were evaluated for stages of vigilance after treatment. Sleep/wake parameters and EEG power spectral analyses during sleep were investigated. Results show no significant effect after 5% or 10% DMSO treatment. DMSO 15% increased mean episode duration of light slow wave sleep (SWS), decreasing mean episode duration of deep SWS and of quiet wake (QW). DMSO 20% increased light SWS enhancing number of episodes, while decreased deep SWS mean episode duration. EEG power spectra of sigma and delta activity were also affected by DMSO. Therefore, DMSO at 15% and 20% affects sleep architecture in rats, increasing light SWS and reducing deep SWS. Being aware of DMSO behavioural effects seems important since experimental artefacts caused by DMSO can lead to the erroneous interpretation of results. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Sleep; Wakefulness; Power spectra; DMSO; Solvent; Vehicle; Rat
1. Introduction Dimethyl sulfoxide (DMSO) [(CH3 )2 SO] is a molecule with a highly polar domain and two apo∗ Corresponding author. Tel.: +34 952 13 25 09; fax: +34 952 13 26 21. E-mail address: [email protected] (M. Cavas).
lar groups, making it soluble in both aqueous and organic media. Due to these physico-chemical properties, DMSO is a very efficient solvent for water-insoluble compounds and is a hydrogen-bound disrupter (Santos et al., 2003). DMSO has been available for medical study since 1963. Since then, it has been frequently used as a solvent in biological studies and also as a vehicle for drug therapy. Although DMSO has been
0378-4274/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2005.02.003
222
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
extensively used for many years, few data on its behavioural or neurotoxic effects are available. The aim of this work is to explore any possible effect over sleep and wake states of DMSO. For this purpose we examine the effects of four widely used concentrations of DMSO (5%, 10%, 15% and 20%) dissolved in saline on sleep/wake states in rats.
2. Methods 2.1. Animals Twenty male Wistar rats, weighting 275–300 g, from Service of Laboratory Animals, University of M´alaga, Spain, were used. 2.2. Substances Dimethyl sulfoxide (Sigma, Germany) was prepared at 5%, 10%, 15%, and 20% concentrations, dissolved in saline. Concentrations were chosen on the basis of those widely used on behavioural studies. The injection volume was 5 ml/kg. 2.3. Manipulations Animals were maintained for one week before surgery on a 12 h light/12 h dark schedule (lights on at 9:00 a.m.) in a temperature controlled room (22 ± 1 ◦ C) with free access to food and water. Under general anaesthesia (Thiopental, 60 mg/kg i.p., Sigma, Germany) the rats were implanted with stainless steel electrodes screwed into the skull over the frontoparietal (lateral, 2 and 2 mm posterior to bregma) and occipital (lateral, 1 and 2 mm anterior to lambda) cortex for recording electroencephalographic activity. Two stainless-steel curved needles were inserted and fastened to the dorsal neck muscles for recording electromyographic activity. A ground screw electrode was placed over the parietal cortex. The free ends of each electrode were placed in a six-pin plastic connector that was cemented in place with acrylic cement. After surgery, the rats were housed individually and allowed to recover for 10 days. Ten days after surgery the animals were habituated for 5 days to experimental conditions in order to minimize the stress involved in experimental procedures.
2.4. Apparatus and recording Animals were habituated to recording conditions which consisted on a sound-attenuated chamber (Letica) fitted with a cable connector for electroencephalography (EEG) and electromyography (EMG) recording via a flexible cable system including a swivel device which allowed the animals to move freely. Signals were recorded with Power Lab (ADInstruments) hardware, amplified and digitised with a sample frequency of 400 samples per second. Treatment of the signal was performed using Chart 4.0 software. EEG signals were high-pass filtered with a 0.3 Hz filter and a low-pass filtered with 60 Hz filter. A Notch filter of 50 Hz was used to eliminate influence from the alternate current frequency. EMG signal was high-pass filtered with a 10 Hz filter and low-pass filtered with a 1 kHz filter. Polygraphic recordings were evaluated for stages of vigilance. Five categories of sleep/wake states based on the wave form were considered (Robert et al., 1999; Datta and Hobson, 2000): (a) Active waking, cortical EEG waves are predominantly low voltage (40–60 V) waves and high frequency (35–50 Hz) with mixed theta rhythm (4–10 Hz), the nuchal EMG exhibits high tonus and occasional bursts of activity consistent with head and neck movements made by the animal. (b) Quiet waking, cortical EEG shows no marked difference from observations during active waking, but there is less prominent EEG theta-activity and the EMG tone is markedly reduced compare with its levels in active waking. During slow wave sleep (SWS), cortical EEG progressively slows and increases its amplitude. SWS is characterised by the presence of low frequency (0.1–10 Hz) and high amplitude (100–400 V) waves. On the basis of cortical EEG, SWS may be subdivided into (c) Light SWS and (d) Deep SWS. Light SWS is characterised by low frequency and medium amplitude (100–200 V) waves, higher amplitude waves occupy less than 50% of the epoch. EMG is noticeably reduced compared to quiet waking. During deep SWS waves are slower and higher in amplitude compared to light SWS, and delta waves occupy more than 50% of the epoch. EMG tone is minimal. (e) REM sleep is assigned when cortical EEG consists of high frequency (20–40 Hz) and low amplitude waves (50–80 V), with presence of EEG theta-activity,
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
mostly a 7 Hz rhythm. Nuchal EMG shows muscle atonia. This visual scoring was completed with the analysis of the spectral power of the EEG bands during each epoch to ensure the objective evaluation of the polygraphic recording. Spectral analysis was performed using Chart 4.0 software. The EEG signal was subjected to a Fast Fourier Transform (FFT) using 2 s intervals for each 10 s epoch. The FFT was computed on 1024 points corresponding to 2 s epochs with a resolution of 0.39 Hz. A seven-point smoothing window was applied allowing a minimum frequency of 0.39 Hz and a maximum frequency of 50.00 Hz in the spectrum computation. For each 2 s interval, the EEG power densities in the delta (0.39–4.30 Hz), theta (4.69–8.98 Hz), sigma (9.38–14.06 Hz), beta (14.45–30.08 Hz) and gamma band (30.47–50.00 Hz) were computed (Maloney et al., 1997). The epoch was said to belong to a specific vigilance state when criteria previously described for each state were fulfilled for more than 50% of scoring epoch, this is at least three out of five FFT 2 s intervals (Bjorvatn et al., 1995, 1998). 2.4.1. Power spectral analysis of the sleep EEG To complete the sleep polygraphic recordings, the analysis of the EEG power spectra during sleep was performed using the FFT routine described above. For the power spectral analysis, ten 2 s samples of each sleep state from each one hour period were computed. Only epochs that were preceded by 15 s and followed by 15 s of the corresponding state on the digital recordings were selected for analysis of the spectral power (Corsi-Cabrera et al., 2001).
223
2.6. Evaluation and analysis of data Polygraphic recordings were evaluated as follows: the electrographic activity of 10 s epochs were analysed and assigned to one state of vigilance. One episode was considered when three consecutive 10 s epochs of the same behavioural state were observed. For each sleep/wake state five parameters were evaluated: duration, number of episodes, mean episode duration, latency. Total time of sleep, sleep efficiency, number of sleep cycles and mean cycle duration were also measured. For the EEG power spectra analysis, power densities values were log-transformed prior to statistical analysis to reduce effects of non-Gaussian distribution (Bjorvatn et al., 1995; Ursin and Bjorvatn, 1998). The spectra and band activity were displayed and reported using power density units (V2 /Hz). Experimental data are expressed as mean ± S.E.M. Data is presented and analysed in periods of time: three time blocks of one hour time each block, and total recording time, to study drugs effects across time. For statistical analysis the differences between values obtained after DMSO administration and those obtained after saline were detected by Wilcoxon ranges test within the indicated time blocks. This test is the best indicated for repeated measures studies when data do not fulfil the requirements for parametric tests. A p-value <.05 was used as the cut-off for statistical significance. The experiment followed UE guidelines (86/609/ EEC) and all efforts were made to minimize the number of animals used and their suffering in every step of the experiment.
2.5. Experimental protocols 3. Results Four groups of rats were used, each group receiving a concentration of DMSO (5%, 10%, 15%, and 20%). After habituation, on experimental day one, animals received a control solution (saline) administered intraperitoneally at the beginning of the light period (9:00 a.m.). Polygraphic recordings started 30 min after injection and were performed for 3 h to study drugs effects across time. The day after, experimental day 2, animals received the assigned concentration of DMSO at the beginning of the light period. Recording started 30 min after and lasted 3 h. Animals were tested twice only (control and drug).
3.1. Sleep and waking effects DMSO at 5% or 10% dissolved in saline compared with saline administration alone, showed no statistically significant effect over sleep/wake states at any of the time blocks studied. At 15%, DMSO showed no statistically significant effect over sleep/wake states during first hour of recording. During second hour, mean episode duration of deep slow wave sleep (p = .043) and of quiet wake (p = .043) decreased. During third hour of recording, an increase
224
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
on mean episode duration of light slow wave sleep was observed (p = .043). At 20%, DMSO showed no statistically significant effect during first hour of recording. During second hour, time spent on light slow wave sleep increased (p = .043), enhancing the number of episodes (p = .042)
of this type of sleep. During third hour no statistically significant effect was observed. Total time spent in light slow wave sleep increased (p = .043), enhancing number of episodes (p = .042) of this type of sleep. Mean episode duration of deep slow wave sleep was decreased (p = .043).
Figs. 1–6. Figs. 1–5: Time spent in each vigilance state during each time block in control and treatment sessions. Fig. 6: Time spent asleep during each time block in control and treatment sessions. Asterisk ( ) denotes a statistically significant difference between values obtained after i.p. administration of saline during control sessions and those obtained after DMSO during treatment sessions. Saline–DMSO 5% (), saline–DMSO 10% (), saline–DMSO 15% (♦), and saline–DMSO 20% (). Symbols above the x-axis indicate DMSO treatment that produces significant differences.
Table 1 Effects of DMSO (5%, 10%, 15% and 20%) administered intraperitoneally on sleep architecture of rats Treatment
Period (h)
Light slow wave sleep No. of episodes
Mean episode duration
5.0 3.25 8.75 17.00
± ± ± ±
2.30 2.06 6.18 10.51
.81 ± .26 1.08 ± .94 1.06 ± .18 .91 ± .19
0–1 1–2 2–3 Total
5.75 9.50 10.25 25.50
± ± ± ±
5.50 5.80 3.30 13.22
.98 ± .35 .88 ± .23 .86 ± .16 .88 ± .19
Saline
0–1 1–2 2–3 Total
4.60 8.80 9.40 22.80
± ± ± ±
1.51 2.94 4.72 4.08
.91 ± .25 1.06 ± .33 .93 ± .17 .99 ± .06
DMSO 10%
0–1 1–2 2–3 Total
3.04 7.40 7.20 18.00
± ± ± ±
2.40 3.20 3.11 6.32
.93 ± .31 .90 ± .09 .90 ± .29 .91 ± .19
0–1 1–2 2–3 Total
4.40 6.40 9.20 20.00
± ± ± ±
3.13 2.40 5.06 5.52
.87 ± .23 1.16 ± .47 .81 ± .21 .99 ± .06
0–1 1–2 2–3 Total
5.40 7.40 12.60 25.40
± ± ± ±
1.94 4.15 4.97 8.47
.98 ± .38 1.00 ± .23 1.17 ± .25* 1.06 ± .16
Saline
0–1 1–2 2–3 Total
6.20 3.80 5.80 15.80
± ± ± ±
3.03 1.30 4.91 7.22
1.15 ± .44 .82 ± .25 .87 ± .21 .99 ± .23
DMSO 20%
0–1 1–2 2–3 Total
6.60 10.20 9.40 26.20
± ± ± ±
4.66 5.01* 3.91 11.7*
1.02 ± .28 .93 ± .22 .96 ± .15 .95 ± .15
DMSO 5%
Saline
DMSO 15%
Latency
No. of episodes
Mean episode duration
27.25 ± 15.56
3.50 4.75 7.00 15.25
± ± ± ±
3.00 3.30 2.82 7.88
1.62 ± .89 2.42 ± .50 2.24 ± .91 2.38 ± .86
19.75 ± 13.97
1.25 5.25 5.75 12.25
± ± ± ±
1.89 4.71 4.57 7.93
.81 ± .02 1.21 ± .55 .80 ± .26 1.04 ± .35
26.23 ± 3.48
3.40 8.00 6.00 17.40
± ± ± ±
1.67 2.12 1.22 2.88
2.48 ± 1.70 2.29 ± 1.10 1.85 ± .59 2.17 ± .79
44.13 ± 28.02
3.20 7.60 7.20 18.00
± ± ± ±
3.96 1.94 1.78 4.63
.65 ± .14 1.75 ± .73 2.18 ± .63 1.84 ± .67
21.46 ± 11.55
1.80 5.00 6.40 13.20
± ± ± ±
1.30 2.64 1.14 3.27
2.54 ± 2.06 2.43 ± 1.30 2.21 ± .34 2.46 ± .64
21.63 ± 15.90
1.20 4.00 9.20 14.40
± ± ± ±
1.64 4.12 4.54 6.91
.72 ± .25 1.04 ± .39* 1.63 ± 1.26 1.54 ± 1.03
10.90 ± 8.25
4.00 4.00 8.40 16.40
± ± ± ±
3.39 2.82 4.33 8.01
1.28 ± .81 2.30 ± 1.97 2.18 ± .55 2.05 ± .55
25.93 ± 27.44
4.80 5.60 7.00 17.40
± ± ± ±
2.77 3.20 5.24 6.58
1.61 ± .75 1.17 ± .43 1.16 ± .23 1.34 ± .31*
Paradoxical sleep Latency
No. of episodes
Mean episode duration
Latency
57.41 ± 54.91
.50 2.25 5.00 7.75
± ± ± ±
.57 1.50 2.00 2.62
1.16 ± .00 1.18 ± .70 1.83 ± .23 1.62 ± .30
83.54 ± 42.15
56.12 ± 34.09
.00 3.00 3.00 6.00
± ± ± ±
.00 2.44 .81 3.16
1.23 ± .60 1.61 ± .95 1.60 ± .88
92.16 ± 38.60
34.20 ± 8.76
.60 3.00 3.40 7.00
± ± ± ±
.89 1.87 1.81 2.44
.75 ± .11 2.53 ± 1.48 2.64 ± 2.15 2.02 ± .84
79.93 ± 24.96
47.70 ± 27.46
.60 1.80 4.00 6.40
± ± ± ±
.89 2.16 2.34 4.03
1.00 ± .70 1.46 ± .77 2.18 ± .34 1.97 ± .44
78.43 ± 54.92
39.50 ± 18.00
.20 2.00 3.80 6.00
± ± ± ±
.44 1.41 1.64 1.87
3.83 ± . 2.34 ± 1.76 1.98 ± 1.45 1.76 ± .74
94.66 ± 44.46
48.30 ± 28.46
.40 1.20 2.40 4.00
± ± ± ±
.89 1.64 1.14 3.16
.50 ± . 1.69 ± 1.05 1.74 ± .82 1.77 ± .82
119.30 ± 45.37
27.36 ± 18.08
1.40 2.20 4.00 7.60
± ± ± ±
1.51 1.30 2.91 4.27
1.81 ± 1.35 1.54 ± .58 2.47 ± 1.15 1.90 ± .41
49.63 ± 24.22
34.73±28.77
2.20 4.40 4.40 11.00
± ± ± ±
2.16 3.50 2.88 7.77
1.21 ± .43 1.92 ± .63 1.68 ± .53 1.63 ± .34
47.00 ± 36.05
225
Values are provided as mean (in minutes for duration, mean episode duration, and latency) ± S.E.M. Differences were detected by Wilcoxon ranges test within the indicated time periods. ∗ Statistically significant difference between values obtained after i.p. administration of DMSO and those obtained after saline. A p value <.05 is considered for statistical significance.
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
0–1 1–2 2–3 Total
Saline
Deep slow wave sleep
226
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
Tables 1 and 2 show the effects of saline and DMSO (5%, 10%, 15% and 20%) upon sleep and wakefulness of rats. Figs. 1–5 show group data (one group per line) for time spent in each vigilance state during each time block in control and treatment sessions. Fig. 6 illus-
trates time spent asleep. Figs. 7–11 show individual data (one animal per line) for those slow wave sleep parameters where statistical differences were found between values obtained after control sessions and after treatment with DMSO sessions.
Table 2 Effects of DMSO (5%, 10%, 15% and 20%) administered intraperitoneally on wakefulness of rats Treatment
Period (h)
Active wake No. of episodes
Quiet wake Mean episode duration
Saline
0–1 1–2 2–3 Total
11.25 7.25 5.00 23.50
± ± ± ±
2.50 2.62 4.24 8.96
2.15 ± .43 3.23 ± 1.12 1.89 ± .12 2.43 ± .49
DMSO 5%
0–1 1–2 2–3 Total
9.25 7.00 7.00 23.25
± ± ± ±
2.98 3.91 5.09 10.78
3.42 ± 1.34 2.14 ± .608 5.19 ± 6.61 3.16 ± 1.60
0–1 1–2 2–3 Total
9.20 4.60 5.60 19.40
± ± ± ±
2.04 2.07 1.51 1.94
2.70 ± 1.03 1.92 ± .68 3.42 ± .65 2.64 ± .48
0–1 1–2 2–3 Total
11.80 6.40 4.40 22.60
± ± ± ±
3.03 3.04 2.96 5.59
2.21 ± .60 2.20 ± 1.03 3.36 ± 2.14 2.44 ± .87
0–1 1–2 2–3 Total
8.20 6.80 6.00 21.00
± ± ± ±
2.58 2.16 2.91 3.46
3.89 ± 1.97 2.14 ± .84 2.76 ± 1.04 2.87 ± .90
0–1 1–2 2–3 Total
10.60 7.80 4.40 22.80
± ± ± ±
2.88 2.94 2.70 6.01
2.28 ± .97 2.83 ± 1.28 2.01 ± 1.22 2.52 ± .87
Saline
0–1 1–2 2–3 Total
9.40 8.00 5.40 22.80
± ± ± ±
4.03 3.24 5.12 9.88
2.19 ± .78 3.73 ± 3.11 1.31 ± .63 2.58 ± 1.01
DMSO 20%
0–1 1–2 2–3 Total
8.60 5.60 4.20 18.40
± ± ± ±
2.40 3.97 4.60 8.26
2.90 ± 1.13 4.05 ± 4.69 4.21 ± 3.30 3.02 ± .72
Saline
DMSO 10%
Saline
DMSO 15%
Latency
No. of episodes
Mean episode duration
Latency
.91 ± 1.33
13.50 8.00 7.00 28.50
± ± ± ±
1.73 2.44 4.00 7.72
1.16 1.45 1.09 1.23
± ± ± ±
.46 .42 .25 .34
1.25 ± .86
.58 ± .78
13.00 11.00 10.75 34.75
± ± ± ±
4.24 4.16 6.50 13.50
1.04 1.11 .99 1.06
± ± ± ±
.16 .43 .42 .21
2.12 ± 1.92
1.86 ± 1.84
8.60 4.20 5.00 17.80
± ± ± ±
3.78 2.58 2.91 7.66
1.23 1.49 .99 1.23
± ± ± ±
.30 .69 .50 .26
4.90 ± 8.08
3.50 ± 4.76
10.60 5.80 3.80 20.20
± ± ± ±
5.89 3.56 2.16 9.52
1.56 1.45 1.07 1.33
± ± ± ±
.45 .63 .31 .34
4.73 ± 9.21
2.03 ± 2.05
9.40 7.60 6.40 23.40
± ± ± ±
1.94 3.28 3.97 8.61
1.20 1.53 1.09 1.29
± ± ± ±
.30 .64 .36 .19
4.53 ± 7.27
1.80 ± 2.14
12.40 9.40 5.40 27.20
± ± ± ±
3.71 3.64 2.88 8.87
1.33 1.14 1.97 1.22
± ± ± ±
.34 .33* 2.43 .22
.43 ± .61
1.70 ± 1.87
11.20 9.40 6.40 27.00
± ± ± ±
5.89 4.15 5.81 13.41
1.07 1.46 1.12 1.24
± ± ± ±
.32 .40 .41 .28
10.73 ± 21.77
2.33 ± 3.90
8.80 8.20 6.20 23.20
± ± ± ±
2.94 4.65 5.26 10.35
1.02 1.25 1.08 1.16
± ± ± ±
.26 .56 .53 .43
1.96 ± 2.87
Values are provided as mean (in minutes for duration, mean episode duration, and latency) ± .S.E.M. Differences were detected by Wilcoxon ranges test within the indicated time periods. ∗ Statistically significant difference between values obtained after i.p. administration of DMSO and those obtained after saline. A p value <.05 is considered for statistical significance.
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
227
Figs. 7–11. These figures show individual data (one animal per line) for those slow wave sleep parameters where statistical differences were found between values obtained after control sessions and after treatment with DMSO sessions. Asterisk ( ) denotes a statistically significant difference.
3.2. EEG power spectra effects during sleep To complete the study of the effects DMSO upon sleep architecture, EEG power spectra within sleep stages were analysed off-line. DMSO at 15% modified power densities during light and deep SWS. During light SWS, compared to saline, 15% DMSO reduced EEG power densities in the sigma band during third hour of recording (p = .026). The analysis
of the total three-hour period of recording showed a decrease in the power densities of the sigma waves (p = .009) after DMSO administration. During deep SWS, a decrease of the power densities in the sigma band was observed (p = .000) during the third hour of recording. This reduction in the power spectra of the sigma waves was significant also for the total three-hour recording period (p = .000).
228
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
Figs. 12–18. These figures show significant changes in power densities (V2 /Hz) within delta and sigma bands during light and deep SWS after DMSO treatment. Symbols: saline () – DMSO 15% (), and saline () – DMSO 20% (䊉).
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
DMSO at 20% showed no effect upon power spectra in the delta or sigma band during light SWS. During deep SWS, a reduction of power densities within the sigma band was observed during the third hour period of recording (p = .000) and for the total 3 h recording period (p = .000). Also, the power spectra within the delta band increased after DMSO 20% administration for the total 3 h recording period (p = .000). Figs. 12–18 show significant changes in power densities (V2 /Hz) within delta and sigma bands during light and deep SWS after DMSO treatment.
4. Discussion DMSO is a widely used solvent in biological studies. Despite its frequently use, few data is available on its behavioural effects. To our knowledge, the effects of DMSO upon sleep/wake states have not been studied before. In this work we report dose-dependent effects of DMSO on sleep/wake states in rats. The results from this experiment show that commonly used DMSO concentrations, 15% and 20%, increase light slow wave sleep at expense of deep slow wave sleep, which is reduced. Paradoxical sleep was not affected. Besides its properties as a solvent, and as a vehicle for drug therapy, DMSO has clinical properties by itself. It is only clinically approved for the treatment of interstitial cystitis (Parkin et al., 1997), but it has been used successfully as an adjuvant, for its analgesic properties (Demos et al., 1967), and in the treatment of musculoskeletal disorders (Halsted and Youngberg, 1981), pulmonary adenocarcinoma (Goto et al., 1996), rheumatologic (Abdullaeva and Shakimova, 1989) and dermatologic diseases (Swanson, 1985), chronic prostatitis (Shirley et al., 1978), and as a topical analgesic (Kingery, 1997). DMSO readily crosses the bloodbrain barrier (Broadwell et al., 1982) and has antiinflamatory (Kelly et al., 1994) and reactive oxygen species scavenger actions (Salim, 1992). At the central nervous system, it has been used in the treatment of traumatic brain edema (Ikeda and Long, 1990), schizophrenia (Smith, 1992) and it has been suggested for the treatment of Alzheimer’s disease (Regelson and Harkins, 1997). Besides its pharmacological applications, several systemic side-effects from the use of DMSO have been reported, namely nausea, vomiting (Davis et al., 1990),
229
diarrhea (O’Donnell et al., 1981), dermatologic reactions (Swanson, 1985), and cardiovascular effects (Hameroff et al., 1983). Some studies reveal also toxic effects, particularly on the peripheral nervous system (Cavaletti et al., 2000). DMSO penetrates the cell membrane and causes an increase in osmolality both inside and outside the cell, preventing any significant hemolysis due to the formation of an osmotic gradient (Franco et al., 1983). Cellular and molecular effects of DMSO have also been reported: DMSO interferes with intracellular calcium concentrations, blunting calcium increases, in a nonspecific action of DMSO. Data on other cellular and intracellular mechanisms as sodium and potassium currents and ATPase activities are controversial (Santos et al., 2003). Behavioural assessment of DMSO is spare, in despite of its widely use as a solvent in basic and clinical studies. At high concentrations, 32% and 64%, DMSO significantly decreased locomotor activity in male mice (Castro et al., 1995). Cavaletti et al. (2000) demonstrated that repeated intraperitoneal administration of DMSO produced a marked and dose-dependent reduction in nerve conduction velocity in rats, being 7.2% the highest concentration used. Nevertheless, Authier et al. (2002), using the same experimental protocol and the same range of concentrations (1.8%, 3.6% and 7.2%) could not replicate these results. At these DMSO concentrations, clinical status of the animals was good and no motor deficits were observed. To our knowledge, no other study has assessed behavioural effects of DMSO. In our experiment, used at 5% or 10% concentrations, DMSO has no effect on sleep/wake states in rats, but at 15% and 20% concentrations, DMSO modifies significantly sleep architecture. At 15% concentration, DMSO during second hour of recording, decreased mean episode duration of deep slow wave sleep, and it also affected waking, by decreasing mean episode duration of quiet wake. During third hour, light slow wave sleep was affected, increasing mean episode duration. A higher concentration, 20% of DMSO significantly increased, during second and third hour of polysomnographic recording, time spent in and number of episodes of light slow wave sleep. Total mean episode duration of deep slow wave sleep was reduced. Therefore, at 15% and 20% concentrations, DMSO interferes with sleep, increasing light slow wave sleep and decreasing deep slow wave sleep.
230
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
The results observed here clearly show a time window where the behavioural effects upon sleep/wake are consistent; this time window comprises second and third hour, since no significant effect was observed during first hour of recording. Therefore, DMSO behavioural effects upon sleep and wakefulness are observed 90 min after injection. The schedule of administration of DMSO in the works conducted by Cavaletti et al. (2000) and Authier et al. (2002) consisted on one intraperitoneal injection of DMSO a day during 10 days, and these injections were given after the behavioural tests, therefore there is no previous data about the latency for behavioural effects of DMSO. Using a higher range of concentrations of DMSO in mice, Castro et al. (1995) observed decreased locomotor activity after DMSO treatment. Locomotor activity in those animals receiving a 32% concentration of DMSO was significantly decreased at 10–75 min after administration and at 5–120 min in those receiving a 64% concentration of DMSO. DMSO is readily absorbed following administration by all routes and distributed throughout the body. It is metabolized in part by oxidation to methylsulfonylmethane and by reduction to dimethyl sulfide. These metabolites are excreted in the urine and feces (Kolb et al., 1967). In a study by Nishimura et al. (1989) looking specifically at distribution of DMSO in brain and vascular tissue in the rat, calculated tissue to plasma ratios were observed to be 1:1 two hours after infusion was initiated. This could give support to the results reported here, initiated 90 min after injection of DMSO. The analysis of EEG power spectra within delta and sigma bands during sleep stages reveals significant effects of DMSO at 15% and 20%, later on, during third hour of recording, and these effects are also significant for the total 3 h recording period. Polysomnographic recordings conducted here show a latency of 90 min for DMSO effects upon behavioural manifestations of sleep/wake stages. Within sleep stages, DMSO effects upon spectral power of sigma band appear later on, during third hour of recording. However, this is not unusual since drugs often differ in their latency for their different effects. The latency of DMSO effects upon sleep and wake also raises the question of the duration of the effects observed. In the present study, recordings were evaluated during three hours at the beginning of the light period. This 3 h period of recording is frequently used in stud-
ies upon sleep/wake. However, the present study has the limitation that it explores only one part of the light period. Since neurobiological and behavioural studies upon sleep/wake states are often conducted throughout a 24 h period, including light and dark phases, the magnitude of the changes reported here should be analysed in further studies to examine DMSO effects upon sleep and wake in an extended period of time. Exploring differences between light and dark periods would be also of great interest. Despite this limitation, the results reported in this study are original since they show for the first time that DMSO changes sleep architecture in rats. The experimental work conducted here does not permit us to know which mechanisms may underlie the effects of DMSO administration upon sleep states, or whether these effects are central or peripheric, or specific or non-specific actions of DMSO, but some explanations could be suggested. Due to its cardiovascular effects, reported diastolic and systolic hypertension, bradycardia and cardiac arrest, DMSO could affect sleep strongly. Also, as an anti-inflammatory agent, DMSO could interfere with sleep intensity. Additionally, through its reactive oxygen species scavenger actions, DMSO could interfere with sleep. Moreover, DMSO can reduce noradrenalin-mediated vasoconstriction, enhance cerebral blood flow (De la Torre et al., 1975), and it is an effective cholinesterase inhibitor (Watts and Hoogmed, 1984). These actions may contribute to its effects on sleep states. Furthermore, DMSO actions on intracellular systems, disrupting normal neurotransmision, may also be responsible for the results observed here. Thus, one possibility is that DMSO is interfering with intracellular calcium concentration and indirectly affects the massive calcium entry during spindles occurring within slow wave sleep. This argument is relevant for the results observed here, given the fact that, in the rat, an increased density of sleep spindles correlates well with the amount of deep SWS, pointing out the role of sleep spindles as the hallmark of consolidated (deep) SWS. Thus, in Wistar rats, spindles increase in number, duration, and amplitude as sleep deepens (Gottesmann, 1992; Terrier and Gottesmann, 1978). Moreover, to complete the findings of the polysomnographic recordings, the analysis of the power spectra of delta and sigma bands during light and deep SWS was conducted in the present study. These analyses show that power spectra within sigma band, mainly reflecting sleep spindles which contribute
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
to the bulk of power found in this 9.38–14.06 Hz range (Bjorvatn et al., 1998), during deep SWS are reduced after the administration of DMSO at 15% and 20% concentrations. This result is consistently found during third hour of recording and for the total 3 h recording period at both concentrations. This finding is accompanied by an increase of the power of delta waves within deep SWS, observed in the total 3 h period of recording after DMSO 20%. Therefore, the results here show a decrease of the power of the sigma band and an increase of the power of delta band after DSMO administration, which differs from the normal progress of sleep described in the rat (Gottesmann, 1992; Terrier and Gottesmann, 1978). There is to say that in the present study, DMSO at 15% also affected sigma band power spectra during light SWS, decreasing the power of sigma waves. However, at 20% concentration, DMSO did not affect significantly power spectra during light SWS. At the moment, it is difficult to assess this finding. Further studies specifically focused on EEG spectral power are needed to clarify these matters. It seems important to notice also that, at present, there is no data about whether DMSO may have a noxious effect on the animal, but if it does, an indirect impact on sleep-wake state by the treatment with DMSO should be expected. However, even though mechanisms through which DMSO affects sleep and wake are unknown so far, the results in this work show that this widely used solvent affects sleep architecture in rats. Being aware of DMSO behavioural effects seems important since experimental artefacts caused by DMSO can lead to the erroneous interpretation of results. Especially since sleep represents a physiological need, which alterations may interfere with several cognitive and behavioural processes. An increase awareness of the multidisciplinary utilisation of DMSO and other commonly used solvents in several research fields can be a valid contribution to avoid or minimise misunderstandings.
5. Conclusion DMSO is a very efficient solvent, widely used in biological studies and also as a vehicle for drug therapy. Little is known about its behavioural or neurotoxical effects. In the present study, DMSO effects upon sleep and wake states in rats are tested. DMSO at 5% and
231
10% has no effect. However, at 15% and 20% concentrations, DMSO interferes with sleep, increasing light slow wave sleep and decreasing deep slow wave sleep. Additionally, power spectral analysis of the sleep EEG shows a decrease in the power of sigma activity after treatment with DMSO at 15% and 20% concentrations. At 20%, DMSO also increased the power density of delta activity during deep SWS. Therefore, DMSO at these frequently used concentrations modifies sleep architecture in rats.
Acknowledgements Experimental procedures are in compliance with the European Communities Council Directive of 24 November 1986 (86/609/EEC). Work financed by Consejer´ıa de Educaci´on y Ciencia, Junta de Andaluc´ıa, Spain.
References Abdullaeva, G., Shakimova, B., 1989. An evaluation of the efficacy of treating rheumatoid arthritis with preparations for local use. Revmatologiia (Mosk) 4, 35–39. Authier, N., Dupuis, E., Kwasiborski, A., Eschalier, A., Coudor´e, F., 2002. Behavioural assessment of dimethylsulfoxide neurotoxicity in rats. Toxicol. Lett. 132, 117–121. Bjorvatn, B., Bjorkum, A., Neckelmann, D., Ursin, R., 1995. Sleep/waking and EEG power spectrum effects of a nonselective serotonin (5-HT) antagonist and a selective 5-HT reuptake inhibitor given alone and in combination. Sleep 18, 451–462. Bjorvatn, B., Fagerland, S., Ursin, R., 1998. EEG power densities (0.5–20 Hz) in different sleep-wake stages in rats. Physiol. Behav. 63, 413–417. Broadwell, R., Salcman, M., Kaplan, R., 1982. Morphologic effect of dimethyl sulfoxide on the blood–brain barrier. Science 217, 164–166. Castro, C., Hogan, J., Benson, K., Shehata, C., Landauer, M., 1995. Behavioural effects of vehicles: DMSO, etanol, Tween-20, and emulphor-620. Pharmacol. Biochem. Behav. 50, 521–526. Cavaletti, G., Oggioni, N., Sala, F., Pezzoni, G., Cavalletti, E., Marmiroli, P., Petruccioli, M.G., Frattola, L., Tredici, G., 2000. Effect on the peripheral nervous system of systemically administered dimethylsulfoxide in the rat: a neurophysiological and pathological study. Toxicol. Lett. 118, 103–107. Corsi-Cabrera, M., Perez-Garci, E., Del R´ıo-Portilla, Y., Ugalde, E., Guevara, M., 2001. EEG bands during wakefulness, slow-wave, and paradoxical sleep as a result of principal component analysis in the rat. Sleep 24, 374–380. Datta, S., Hobson, J.A., 2000. The rat as an experimental model for sleep neurophysiology. Behav. Neurosci. 114, 1239–1244.
232
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
Davis, J., Rowley, S., Braine, H., Piantadosi, S., Santos, G., 1990. Clinical toxicity of cryopreserved bone marrow graft infusion. Blood 75, 781–786. De la Torre, J., Kawanaga, H., Rowed, D., Johnson, C., Goode, D.J., Mullan, S., 1975. Dimethylsulfoxide in central nervous system trauma. Ann. NY Acad. Sci. 243, 362–389. Demos, C., Beckloff, G., Donin, M., Oliver, P., 1967. Dimethyl sulfoxide in musculoskeletal disorders. Ann. NY Acad. Sci. 147, 516–523. Franco, R., Weiner, M., Wagner, K., Martelo, O., 1983. Incorporation of inositol hexaphosphate into red blood cells mediated by dimethyl sulfoxide. Life Sci. 32, 2763–2768. Goto, I., Yamamoto-Yamagushi, Y., Honma, Y., 1996. Enhancement of sensitivity of human lung adenocarcinoma cells to growthinhibitory activity of interferon ␣ by differentiation inducing agents. Br. J. Cancer 74, 546–554. Gottesmann, C., 1992. Detection of seven sleep-waking stages in the rat. Neurosci. Biobehav. Rev. 16, 31–38. Halsted, B., Youngberg, S., 1981. The DMSO Handbook. Golden Quill Publishers, California. Hameroff, S., Otto, C., Kanel, J., Weinstein, P., Blitt, C., 1983. Acute cardiovascular effects of dimethyl sulfoxide. Ann. NY Acad. Sci. 411, 94–99. Ikeda, Y., Long, D., 1990. Comparative effects of direct and indirect hydroxyl radical scavengers on traumatic brain oedema. Acta Neurochir. Suppl. (Wien) 51, 74–76. Kelly, K., Hill, M., Youkhana, K., Wanker, F., Gimble, J., 1994. Dimethyl sulfoxide modulates NF-B and cytokine activation in lipopolysaccharide-treated murine macrophages. Infect. Immun. 62, 3122–3128. Kingery, W., 1997. A critical review of controlled clinical trials for peripheral neuropathic pain and complex regional pain syndromes. Pain 73, 123–139. Kolb, K., Jaenicke, G., Kramer, M., Schulze, P., 1967. Absorption, distribution and elimination of labeled dimethyl sulfoxide in man and animals. Ann. NY Acad. Sci. 141, 85–95. Maloney, K., Cape, E., Gotman, J., Jones, B., 1997. High-frequency ␥ electroencephalogram activity in association with sleep-wake states and spontaneous behaviors in the rat. Neuroscience 76, 541–555.
Nishimura, M., Jacob, P., Cassel, M., Pitts, L., 1989. Tissue distribution of intravenously administered dimethyl sulfoxide in the normal rat. Drug Metab. Dispos. 17, 224–226. O’Donnell, J., Burnett, A., Sheehan, T., Tansey, P., McDonald, G., 1981. Safety of dimethyl sulphoxide. Lancet 1, 498. Parkin, J., Shea, C., Sant, G., 1997. Intravesical dimethyl sulfoxide (DMSO) for interstitial cystitis—a practical approach. Urology 49, 105–107. Regelson, W., Harkins, S., 1997. Amyloid is not a tombstone—a summation. The primary role for cerebrovascular and CSF dynamics as factors in Alzheimer’s disease (AD): DMSO, fluorocarbon oxygen carriers, thyroid hormonal, and other suggested therapeutic measures. Ann. NY Acad. Sci. 826, 348–374. Robert, C., Guilpin, G., Limoge, A., 1999. Automated sleep staging systems in rats. J. Neurosci. Meth. 88, 111–122. Salim, A., 1992. Role of oxygen-derived free radical scavengers in the management of recurrent attacks of ulcerative colitis: a new approach. J. Lab. Clin. Med. 119, 710–717. Santos, N., Figueira-Coelho, F., Martins-Silva, J., Salda˜na, C., 2003. Multidisciplinary utilization of dimethyl sulfoxide: pharmacological, cellular, and molecular aspects. Biochem. Pharmacol. 65, 1035–1041. Shirley, S., Steward, B., Mirelman, S., 1978. Dimethyl sulfoxide in treatment of inflammatory genitourinary disorders. Urology 11, 215–220. Smith, R., 1992. A comprehensive macrophage-T-lymphocyte theory of schizophrenia. Med. Hypotheses 39, 248–257. Swanson, B., 1985. Medical use of dimethyl sulfoxide (DMSO). Rev. Clin. Basic Pharm. 5, 1–33. Terrier, G., Gottesmann, C., 1978. Study of cortical spindles during sleep in the rat. Brain Res. Bull. 3, 701–706. Ursin, R., Bjorvatn, B., 1998. Sleep-wake and EEG effects following adenosine A1 agonism and antagonism: similarities and interactions with sleep-wake and EEG effects following a serotonin reuptake inhibitor in rats. Sleep Res. On. 1, 119– 127. Watts, J., Hoogmed, R., 1984. Dimethyl sulfoxide inhibition of acetylcholinesterase in the mammalian heart. Biochem. Pharmacol. 33, 365–369.
Behavioural effects of dimethyl sulfoxide (DMSO): Changes in sleep architecture in rats Mar´ıa Cavas ∗ , David Beltr´an, Jos´e F. Navarro Departamento de Psicobiolog´ıa y Metodolog´ıa de las Ciencias del Comportamiento, Facultad de Psicolog´ıa, Universidad de M´alaga, 29071 M´alaga, Spain Received 15 October 2004; received in revised form 16 February 2005; accepted 17 February 2005 Available online 15 April 2005
Abstract Dimethyl sulfoxide (DMSO) is an efficient solvent for water-insoluble compounds, widely used in biological studies and as a vehicle for drug therapy, but few data on its neurotoxic or behavioural effects is available. The aim of this work is to explore DMSO’s effects upon sleep/wake states. Twenty male rats were sterotaxically prepared for polysomnography. Four concentrations of DMSO (5%, 10%, 15%, and 20%, in saline) were examined. DMSO or saline were administered intraperitoneally at the beginning of the light period. Three hours of polygraphic recording were evaluated for stages of vigilance after treatment. Sleep/wake parameters and EEG power spectral analyses during sleep were investigated. Results show no significant effect after 5% or 10% DMSO treatment. DMSO 15% increased mean episode duration of light slow wave sleep (SWS), decreasing mean episode duration of deep SWS and of quiet wake (QW). DMSO 20% increased light SWS enhancing number of episodes, while decreased deep SWS mean episode duration. EEG power spectra of sigma and delta activity were also affected by DMSO. Therefore, DMSO at 15% and 20% affects sleep architecture in rats, increasing light SWS and reducing deep SWS. Being aware of DMSO behavioural effects seems important since experimental artefacts caused by DMSO can lead to the erroneous interpretation of results. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Sleep; Wakefulness; Power spectra; DMSO; Solvent; Vehicle; Rat
1. Introduction Dimethyl sulfoxide (DMSO) [(CH3 )2 SO] is a molecule with a highly polar domain and two apo∗ Corresponding author. Tel.: +34 952 13 25 09; fax: +34 952 13 26 21. E-mail address: [email protected] (M. Cavas).
lar groups, making it soluble in both aqueous and organic media. Due to these physico-chemical properties, DMSO is a very efficient solvent for water-insoluble compounds and is a hydrogen-bound disrupter (Santos et al., 2003). DMSO has been available for medical study since 1963. Since then, it has been frequently used as a solvent in biological studies and also as a vehicle for drug therapy. Although DMSO has been
0378-4274/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2005.02.003
222
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
extensively used for many years, few data on its behavioural or neurotoxic effects are available. The aim of this work is to explore any possible effect over sleep and wake states of DMSO. For this purpose we examine the effects of four widely used concentrations of DMSO (5%, 10%, 15% and 20%) dissolved in saline on sleep/wake states in rats.
2. Methods 2.1. Animals Twenty male Wistar rats, weighting 275–300 g, from Service of Laboratory Animals, University of M´alaga, Spain, were used. 2.2. Substances Dimethyl sulfoxide (Sigma, Germany) was prepared at 5%, 10%, 15%, and 20% concentrations, dissolved in saline. Concentrations were chosen on the basis of those widely used on behavioural studies. The injection volume was 5 ml/kg. 2.3. Manipulations Animals were maintained for one week before surgery on a 12 h light/12 h dark schedule (lights on at 9:00 a.m.) in a temperature controlled room (22 ± 1 ◦ C) with free access to food and water. Under general anaesthesia (Thiopental, 60 mg/kg i.p., Sigma, Germany) the rats were implanted with stainless steel electrodes screwed into the skull over the frontoparietal (lateral, 2 and 2 mm posterior to bregma) and occipital (lateral, 1 and 2 mm anterior to lambda) cortex for recording electroencephalographic activity. Two stainless-steel curved needles were inserted and fastened to the dorsal neck muscles for recording electromyographic activity. A ground screw electrode was placed over the parietal cortex. The free ends of each electrode were placed in a six-pin plastic connector that was cemented in place with acrylic cement. After surgery, the rats were housed individually and allowed to recover for 10 days. Ten days after surgery the animals were habituated for 5 days to experimental conditions in order to minimize the stress involved in experimental procedures.
2.4. Apparatus and recording Animals were habituated to recording conditions which consisted on a sound-attenuated chamber (Letica) fitted with a cable connector for electroencephalography (EEG) and electromyography (EMG) recording via a flexible cable system including a swivel device which allowed the animals to move freely. Signals were recorded with Power Lab (ADInstruments) hardware, amplified and digitised with a sample frequency of 400 samples per second. Treatment of the signal was performed using Chart 4.0 software. EEG signals were high-pass filtered with a 0.3 Hz filter and a low-pass filtered with 60 Hz filter. A Notch filter of 50 Hz was used to eliminate influence from the alternate current frequency. EMG signal was high-pass filtered with a 10 Hz filter and low-pass filtered with a 1 kHz filter. Polygraphic recordings were evaluated for stages of vigilance. Five categories of sleep/wake states based on the wave form were considered (Robert et al., 1999; Datta and Hobson, 2000): (a) Active waking, cortical EEG waves are predominantly low voltage (40–60 V) waves and high frequency (35–50 Hz) with mixed theta rhythm (4–10 Hz), the nuchal EMG exhibits high tonus and occasional bursts of activity consistent with head and neck movements made by the animal. (b) Quiet waking, cortical EEG shows no marked difference from observations during active waking, but there is less prominent EEG theta-activity and the EMG tone is markedly reduced compare with its levels in active waking. During slow wave sleep (SWS), cortical EEG progressively slows and increases its amplitude. SWS is characterised by the presence of low frequency (0.1–10 Hz) and high amplitude (100–400 V) waves. On the basis of cortical EEG, SWS may be subdivided into (c) Light SWS and (d) Deep SWS. Light SWS is characterised by low frequency and medium amplitude (100–200 V) waves, higher amplitude waves occupy less than 50% of the epoch. EMG is noticeably reduced compared to quiet waking. During deep SWS waves are slower and higher in amplitude compared to light SWS, and delta waves occupy more than 50% of the epoch. EMG tone is minimal. (e) REM sleep is assigned when cortical EEG consists of high frequency (20–40 Hz) and low amplitude waves (50–80 V), with presence of EEG theta-activity,
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
mostly a 7 Hz rhythm. Nuchal EMG shows muscle atonia. This visual scoring was completed with the analysis of the spectral power of the EEG bands during each epoch to ensure the objective evaluation of the polygraphic recording. Spectral analysis was performed using Chart 4.0 software. The EEG signal was subjected to a Fast Fourier Transform (FFT) using 2 s intervals for each 10 s epoch. The FFT was computed on 1024 points corresponding to 2 s epochs with a resolution of 0.39 Hz. A seven-point smoothing window was applied allowing a minimum frequency of 0.39 Hz and a maximum frequency of 50.00 Hz in the spectrum computation. For each 2 s interval, the EEG power densities in the delta (0.39–4.30 Hz), theta (4.69–8.98 Hz), sigma (9.38–14.06 Hz), beta (14.45–30.08 Hz) and gamma band (30.47–50.00 Hz) were computed (Maloney et al., 1997). The epoch was said to belong to a specific vigilance state when criteria previously described for each state were fulfilled for more than 50% of scoring epoch, this is at least three out of five FFT 2 s intervals (Bjorvatn et al., 1995, 1998). 2.4.1. Power spectral analysis of the sleep EEG To complete the sleep polygraphic recordings, the analysis of the EEG power spectra during sleep was performed using the FFT routine described above. For the power spectral analysis, ten 2 s samples of each sleep state from each one hour period were computed. Only epochs that were preceded by 15 s and followed by 15 s of the corresponding state on the digital recordings were selected for analysis of the spectral power (Corsi-Cabrera et al., 2001).
223
2.6. Evaluation and analysis of data Polygraphic recordings were evaluated as follows: the electrographic activity of 10 s epochs were analysed and assigned to one state of vigilance. One episode was considered when three consecutive 10 s epochs of the same behavioural state were observed. For each sleep/wake state five parameters were evaluated: duration, number of episodes, mean episode duration, latency. Total time of sleep, sleep efficiency, number of sleep cycles and mean cycle duration were also measured. For the EEG power spectra analysis, power densities values were log-transformed prior to statistical analysis to reduce effects of non-Gaussian distribution (Bjorvatn et al., 1995; Ursin and Bjorvatn, 1998). The spectra and band activity were displayed and reported using power density units (V2 /Hz). Experimental data are expressed as mean ± S.E.M. Data is presented and analysed in periods of time: three time blocks of one hour time each block, and total recording time, to study drugs effects across time. For statistical analysis the differences between values obtained after DMSO administration and those obtained after saline were detected by Wilcoxon ranges test within the indicated time blocks. This test is the best indicated for repeated measures studies when data do not fulfil the requirements for parametric tests. A p-value <.05 was used as the cut-off for statistical significance. The experiment followed UE guidelines (86/609/ EEC) and all efforts were made to minimize the number of animals used and their suffering in every step of the experiment.
2.5. Experimental protocols 3. Results Four groups of rats were used, each group receiving a concentration of DMSO (5%, 10%, 15%, and 20%). After habituation, on experimental day one, animals received a control solution (saline) administered intraperitoneally at the beginning of the light period (9:00 a.m.). Polygraphic recordings started 30 min after injection and were performed for 3 h to study drugs effects across time. The day after, experimental day 2, animals received the assigned concentration of DMSO at the beginning of the light period. Recording started 30 min after and lasted 3 h. Animals were tested twice only (control and drug).
3.1. Sleep and waking effects DMSO at 5% or 10% dissolved in saline compared with saline administration alone, showed no statistically significant effect over sleep/wake states at any of the time blocks studied. At 15%, DMSO showed no statistically significant effect over sleep/wake states during first hour of recording. During second hour, mean episode duration of deep slow wave sleep (p = .043) and of quiet wake (p = .043) decreased. During third hour of recording, an increase
224
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
on mean episode duration of light slow wave sleep was observed (p = .043). At 20%, DMSO showed no statistically significant effect during first hour of recording. During second hour, time spent on light slow wave sleep increased (p = .043), enhancing the number of episodes (p = .042)
of this type of sleep. During third hour no statistically significant effect was observed. Total time spent in light slow wave sleep increased (p = .043), enhancing number of episodes (p = .042) of this type of sleep. Mean episode duration of deep slow wave sleep was decreased (p = .043).
Figs. 1–6. Figs. 1–5: Time spent in each vigilance state during each time block in control and treatment sessions. Fig. 6: Time spent asleep during each time block in control and treatment sessions. Asterisk ( ) denotes a statistically significant difference between values obtained after i.p. administration of saline during control sessions and those obtained after DMSO during treatment sessions. Saline–DMSO 5% (), saline–DMSO 10% (), saline–DMSO 15% (♦), and saline–DMSO 20% (). Symbols above the x-axis indicate DMSO treatment that produces significant differences.
Table 1 Effects of DMSO (5%, 10%, 15% and 20%) administered intraperitoneally on sleep architecture of rats Treatment
Period (h)
Light slow wave sleep No. of episodes
Mean episode duration
5.0 3.25 8.75 17.00
± ± ± ±
2.30 2.06 6.18 10.51
.81 ± .26 1.08 ± .94 1.06 ± .18 .91 ± .19
0–1 1–2 2–3 Total
5.75 9.50 10.25 25.50
± ± ± ±
5.50 5.80 3.30 13.22
.98 ± .35 .88 ± .23 .86 ± .16 .88 ± .19
Saline
0–1 1–2 2–3 Total
4.60 8.80 9.40 22.80
± ± ± ±
1.51 2.94 4.72 4.08
.91 ± .25 1.06 ± .33 .93 ± .17 .99 ± .06
DMSO 10%
0–1 1–2 2–3 Total
3.04 7.40 7.20 18.00
± ± ± ±
2.40 3.20 3.11 6.32
.93 ± .31 .90 ± .09 .90 ± .29 .91 ± .19
0–1 1–2 2–3 Total
4.40 6.40 9.20 20.00
± ± ± ±
3.13 2.40 5.06 5.52
.87 ± .23 1.16 ± .47 .81 ± .21 .99 ± .06
0–1 1–2 2–3 Total
5.40 7.40 12.60 25.40
± ± ± ±
1.94 4.15 4.97 8.47
.98 ± .38 1.00 ± .23 1.17 ± .25* 1.06 ± .16
Saline
0–1 1–2 2–3 Total
6.20 3.80 5.80 15.80
± ± ± ±
3.03 1.30 4.91 7.22
1.15 ± .44 .82 ± .25 .87 ± .21 .99 ± .23
DMSO 20%
0–1 1–2 2–3 Total
6.60 10.20 9.40 26.20
± ± ± ±
4.66 5.01* 3.91 11.7*
1.02 ± .28 .93 ± .22 .96 ± .15 .95 ± .15
DMSO 5%
Saline
DMSO 15%
Latency
No. of episodes
Mean episode duration
27.25 ± 15.56
3.50 4.75 7.00 15.25
± ± ± ±
3.00 3.30 2.82 7.88
1.62 ± .89 2.42 ± .50 2.24 ± .91 2.38 ± .86
19.75 ± 13.97
1.25 5.25 5.75 12.25
± ± ± ±
1.89 4.71 4.57 7.93
.81 ± .02 1.21 ± .55 .80 ± .26 1.04 ± .35
26.23 ± 3.48
3.40 8.00 6.00 17.40
± ± ± ±
1.67 2.12 1.22 2.88
2.48 ± 1.70 2.29 ± 1.10 1.85 ± .59 2.17 ± .79
44.13 ± 28.02
3.20 7.60 7.20 18.00
± ± ± ±
3.96 1.94 1.78 4.63
.65 ± .14 1.75 ± .73 2.18 ± .63 1.84 ± .67
21.46 ± 11.55
1.80 5.00 6.40 13.20
± ± ± ±
1.30 2.64 1.14 3.27
2.54 ± 2.06 2.43 ± 1.30 2.21 ± .34 2.46 ± .64
21.63 ± 15.90
1.20 4.00 9.20 14.40
± ± ± ±
1.64 4.12 4.54 6.91
.72 ± .25 1.04 ± .39* 1.63 ± 1.26 1.54 ± 1.03
10.90 ± 8.25
4.00 4.00 8.40 16.40
± ± ± ±
3.39 2.82 4.33 8.01
1.28 ± .81 2.30 ± 1.97 2.18 ± .55 2.05 ± .55
25.93 ± 27.44
4.80 5.60 7.00 17.40
± ± ± ±
2.77 3.20 5.24 6.58
1.61 ± .75 1.17 ± .43 1.16 ± .23 1.34 ± .31*
Paradoxical sleep Latency
No. of episodes
Mean episode duration
Latency
57.41 ± 54.91
.50 2.25 5.00 7.75
± ± ± ±
.57 1.50 2.00 2.62
1.16 ± .00 1.18 ± .70 1.83 ± .23 1.62 ± .30
83.54 ± 42.15
56.12 ± 34.09
.00 3.00 3.00 6.00
± ± ± ±
.00 2.44 .81 3.16
1.23 ± .60 1.61 ± .95 1.60 ± .88
92.16 ± 38.60
34.20 ± 8.76
.60 3.00 3.40 7.00
± ± ± ±
.89 1.87 1.81 2.44
.75 ± .11 2.53 ± 1.48 2.64 ± 2.15 2.02 ± .84
79.93 ± 24.96
47.70 ± 27.46
.60 1.80 4.00 6.40
± ± ± ±
.89 2.16 2.34 4.03
1.00 ± .70 1.46 ± .77 2.18 ± .34 1.97 ± .44
78.43 ± 54.92
39.50 ± 18.00
.20 2.00 3.80 6.00
± ± ± ±
.44 1.41 1.64 1.87
3.83 ± . 2.34 ± 1.76 1.98 ± 1.45 1.76 ± .74
94.66 ± 44.46
48.30 ± 28.46
.40 1.20 2.40 4.00
± ± ± ±
.89 1.64 1.14 3.16
.50 ± . 1.69 ± 1.05 1.74 ± .82 1.77 ± .82
119.30 ± 45.37
27.36 ± 18.08
1.40 2.20 4.00 7.60
± ± ± ±
1.51 1.30 2.91 4.27
1.81 ± 1.35 1.54 ± .58 2.47 ± 1.15 1.90 ± .41
49.63 ± 24.22
34.73±28.77
2.20 4.40 4.40 11.00
± ± ± ±
2.16 3.50 2.88 7.77
1.21 ± .43 1.92 ± .63 1.68 ± .53 1.63 ± .34
47.00 ± 36.05
225
Values are provided as mean (in minutes for duration, mean episode duration, and latency) ± S.E.M. Differences were detected by Wilcoxon ranges test within the indicated time periods. ∗ Statistically significant difference between values obtained after i.p. administration of DMSO and those obtained after saline. A p value <.05 is considered for statistical significance.
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
0–1 1–2 2–3 Total
Saline
Deep slow wave sleep
226
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
Tables 1 and 2 show the effects of saline and DMSO (5%, 10%, 15% and 20%) upon sleep and wakefulness of rats. Figs. 1–5 show group data (one group per line) for time spent in each vigilance state during each time block in control and treatment sessions. Fig. 6 illus-
trates time spent asleep. Figs. 7–11 show individual data (one animal per line) for those slow wave sleep parameters where statistical differences were found between values obtained after control sessions and after treatment with DMSO sessions.
Table 2 Effects of DMSO (5%, 10%, 15% and 20%) administered intraperitoneally on wakefulness of rats Treatment
Period (h)
Active wake No. of episodes
Quiet wake Mean episode duration
Saline
0–1 1–2 2–3 Total
11.25 7.25 5.00 23.50
± ± ± ±
2.50 2.62 4.24 8.96
2.15 ± .43 3.23 ± 1.12 1.89 ± .12 2.43 ± .49
DMSO 5%
0–1 1–2 2–3 Total
9.25 7.00 7.00 23.25
± ± ± ±
2.98 3.91 5.09 10.78
3.42 ± 1.34 2.14 ± .608 5.19 ± 6.61 3.16 ± 1.60
0–1 1–2 2–3 Total
9.20 4.60 5.60 19.40
± ± ± ±
2.04 2.07 1.51 1.94
2.70 ± 1.03 1.92 ± .68 3.42 ± .65 2.64 ± .48
0–1 1–2 2–3 Total
11.80 6.40 4.40 22.60
± ± ± ±
3.03 3.04 2.96 5.59
2.21 ± .60 2.20 ± 1.03 3.36 ± 2.14 2.44 ± .87
0–1 1–2 2–3 Total
8.20 6.80 6.00 21.00
± ± ± ±
2.58 2.16 2.91 3.46
3.89 ± 1.97 2.14 ± .84 2.76 ± 1.04 2.87 ± .90
0–1 1–2 2–3 Total
10.60 7.80 4.40 22.80
± ± ± ±
2.88 2.94 2.70 6.01
2.28 ± .97 2.83 ± 1.28 2.01 ± 1.22 2.52 ± .87
Saline
0–1 1–2 2–3 Total
9.40 8.00 5.40 22.80
± ± ± ±
4.03 3.24 5.12 9.88
2.19 ± .78 3.73 ± 3.11 1.31 ± .63 2.58 ± 1.01
DMSO 20%
0–1 1–2 2–3 Total
8.60 5.60 4.20 18.40
± ± ± ±
2.40 3.97 4.60 8.26
2.90 ± 1.13 4.05 ± 4.69 4.21 ± 3.30 3.02 ± .72
Saline
DMSO 10%
Saline
DMSO 15%
Latency
No. of episodes
Mean episode duration
Latency
.91 ± 1.33
13.50 8.00 7.00 28.50
± ± ± ±
1.73 2.44 4.00 7.72
1.16 1.45 1.09 1.23
± ± ± ±
.46 .42 .25 .34
1.25 ± .86
.58 ± .78
13.00 11.00 10.75 34.75
± ± ± ±
4.24 4.16 6.50 13.50
1.04 1.11 .99 1.06
± ± ± ±
.16 .43 .42 .21
2.12 ± 1.92
1.86 ± 1.84
8.60 4.20 5.00 17.80
± ± ± ±
3.78 2.58 2.91 7.66
1.23 1.49 .99 1.23
± ± ± ±
.30 .69 .50 .26
4.90 ± 8.08
3.50 ± 4.76
10.60 5.80 3.80 20.20
± ± ± ±
5.89 3.56 2.16 9.52
1.56 1.45 1.07 1.33
± ± ± ±
.45 .63 .31 .34
4.73 ± 9.21
2.03 ± 2.05
9.40 7.60 6.40 23.40
± ± ± ±
1.94 3.28 3.97 8.61
1.20 1.53 1.09 1.29
± ± ± ±
.30 .64 .36 .19
4.53 ± 7.27
1.80 ± 2.14
12.40 9.40 5.40 27.20
± ± ± ±
3.71 3.64 2.88 8.87
1.33 1.14 1.97 1.22
± ± ± ±
.34 .33* 2.43 .22
.43 ± .61
1.70 ± 1.87
11.20 9.40 6.40 27.00
± ± ± ±
5.89 4.15 5.81 13.41
1.07 1.46 1.12 1.24
± ± ± ±
.32 .40 .41 .28
10.73 ± 21.77
2.33 ± 3.90
8.80 8.20 6.20 23.20
± ± ± ±
2.94 4.65 5.26 10.35
1.02 1.25 1.08 1.16
± ± ± ±
.26 .56 .53 .43
1.96 ± 2.87
Values are provided as mean (in minutes for duration, mean episode duration, and latency) ± .S.E.M. Differences were detected by Wilcoxon ranges test within the indicated time periods. ∗ Statistically significant difference between values obtained after i.p. administration of DMSO and those obtained after saline. A p value <.05 is considered for statistical significance.
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
227
Figs. 7–11. These figures show individual data (one animal per line) for those slow wave sleep parameters where statistical differences were found between values obtained after control sessions and after treatment with DMSO sessions. Asterisk ( ) denotes a statistically significant difference.
3.2. EEG power spectra effects during sleep To complete the study of the effects DMSO upon sleep architecture, EEG power spectra within sleep stages were analysed off-line. DMSO at 15% modified power densities during light and deep SWS. During light SWS, compared to saline, 15% DMSO reduced EEG power densities in the sigma band during third hour of recording (p = .026). The analysis
of the total three-hour period of recording showed a decrease in the power densities of the sigma waves (p = .009) after DMSO administration. During deep SWS, a decrease of the power densities in the sigma band was observed (p = .000) during the third hour of recording. This reduction in the power spectra of the sigma waves was significant also for the total three-hour recording period (p = .000).
228
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
Figs. 12–18. These figures show significant changes in power densities (V2 /Hz) within delta and sigma bands during light and deep SWS after DMSO treatment. Symbols: saline () – DMSO 15% (), and saline () – DMSO 20% (䊉).
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
DMSO at 20% showed no effect upon power spectra in the delta or sigma band during light SWS. During deep SWS, a reduction of power densities within the sigma band was observed during the third hour period of recording (p = .000) and for the total 3 h recording period (p = .000). Also, the power spectra within the delta band increased after DMSO 20% administration for the total 3 h recording period (p = .000). Figs. 12–18 show significant changes in power densities (V2 /Hz) within delta and sigma bands during light and deep SWS after DMSO treatment.
4. Discussion DMSO is a widely used solvent in biological studies. Despite its frequently use, few data is available on its behavioural effects. To our knowledge, the effects of DMSO upon sleep/wake states have not been studied before. In this work we report dose-dependent effects of DMSO on sleep/wake states in rats. The results from this experiment show that commonly used DMSO concentrations, 15% and 20%, increase light slow wave sleep at expense of deep slow wave sleep, which is reduced. Paradoxical sleep was not affected. Besides its properties as a solvent, and as a vehicle for drug therapy, DMSO has clinical properties by itself. It is only clinically approved for the treatment of interstitial cystitis (Parkin et al., 1997), but it has been used successfully as an adjuvant, for its analgesic properties (Demos et al., 1967), and in the treatment of musculoskeletal disorders (Halsted and Youngberg, 1981), pulmonary adenocarcinoma (Goto et al., 1996), rheumatologic (Abdullaeva and Shakimova, 1989) and dermatologic diseases (Swanson, 1985), chronic prostatitis (Shirley et al., 1978), and as a topical analgesic (Kingery, 1997). DMSO readily crosses the bloodbrain barrier (Broadwell et al., 1982) and has antiinflamatory (Kelly et al., 1994) and reactive oxygen species scavenger actions (Salim, 1992). At the central nervous system, it has been used in the treatment of traumatic brain edema (Ikeda and Long, 1990), schizophrenia (Smith, 1992) and it has been suggested for the treatment of Alzheimer’s disease (Regelson and Harkins, 1997). Besides its pharmacological applications, several systemic side-effects from the use of DMSO have been reported, namely nausea, vomiting (Davis et al., 1990),
229
diarrhea (O’Donnell et al., 1981), dermatologic reactions (Swanson, 1985), and cardiovascular effects (Hameroff et al., 1983). Some studies reveal also toxic effects, particularly on the peripheral nervous system (Cavaletti et al., 2000). DMSO penetrates the cell membrane and causes an increase in osmolality both inside and outside the cell, preventing any significant hemolysis due to the formation of an osmotic gradient (Franco et al., 1983). Cellular and molecular effects of DMSO have also been reported: DMSO interferes with intracellular calcium concentrations, blunting calcium increases, in a nonspecific action of DMSO. Data on other cellular and intracellular mechanisms as sodium and potassium currents and ATPase activities are controversial (Santos et al., 2003). Behavioural assessment of DMSO is spare, in despite of its widely use as a solvent in basic and clinical studies. At high concentrations, 32% and 64%, DMSO significantly decreased locomotor activity in male mice (Castro et al., 1995). Cavaletti et al. (2000) demonstrated that repeated intraperitoneal administration of DMSO produced a marked and dose-dependent reduction in nerve conduction velocity in rats, being 7.2% the highest concentration used. Nevertheless, Authier et al. (2002), using the same experimental protocol and the same range of concentrations (1.8%, 3.6% and 7.2%) could not replicate these results. At these DMSO concentrations, clinical status of the animals was good and no motor deficits were observed. To our knowledge, no other study has assessed behavioural effects of DMSO. In our experiment, used at 5% or 10% concentrations, DMSO has no effect on sleep/wake states in rats, but at 15% and 20% concentrations, DMSO modifies significantly sleep architecture. At 15% concentration, DMSO during second hour of recording, decreased mean episode duration of deep slow wave sleep, and it also affected waking, by decreasing mean episode duration of quiet wake. During third hour, light slow wave sleep was affected, increasing mean episode duration. A higher concentration, 20% of DMSO significantly increased, during second and third hour of polysomnographic recording, time spent in and number of episodes of light slow wave sleep. Total mean episode duration of deep slow wave sleep was reduced. Therefore, at 15% and 20% concentrations, DMSO interferes with sleep, increasing light slow wave sleep and decreasing deep slow wave sleep.
230
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
The results observed here clearly show a time window where the behavioural effects upon sleep/wake are consistent; this time window comprises second and third hour, since no significant effect was observed during first hour of recording. Therefore, DMSO behavioural effects upon sleep and wakefulness are observed 90 min after injection. The schedule of administration of DMSO in the works conducted by Cavaletti et al. (2000) and Authier et al. (2002) consisted on one intraperitoneal injection of DMSO a day during 10 days, and these injections were given after the behavioural tests, therefore there is no previous data about the latency for behavioural effects of DMSO. Using a higher range of concentrations of DMSO in mice, Castro et al. (1995) observed decreased locomotor activity after DMSO treatment. Locomotor activity in those animals receiving a 32% concentration of DMSO was significantly decreased at 10–75 min after administration and at 5–120 min in those receiving a 64% concentration of DMSO. DMSO is readily absorbed following administration by all routes and distributed throughout the body. It is metabolized in part by oxidation to methylsulfonylmethane and by reduction to dimethyl sulfide. These metabolites are excreted in the urine and feces (Kolb et al., 1967). In a study by Nishimura et al. (1989) looking specifically at distribution of DMSO in brain and vascular tissue in the rat, calculated tissue to plasma ratios were observed to be 1:1 two hours after infusion was initiated. This could give support to the results reported here, initiated 90 min after injection of DMSO. The analysis of EEG power spectra within delta and sigma bands during sleep stages reveals significant effects of DMSO at 15% and 20%, later on, during third hour of recording, and these effects are also significant for the total 3 h recording period. Polysomnographic recordings conducted here show a latency of 90 min for DMSO effects upon behavioural manifestations of sleep/wake stages. Within sleep stages, DMSO effects upon spectral power of sigma band appear later on, during third hour of recording. However, this is not unusual since drugs often differ in their latency for their different effects. The latency of DMSO effects upon sleep and wake also raises the question of the duration of the effects observed. In the present study, recordings were evaluated during three hours at the beginning of the light period. This 3 h period of recording is frequently used in stud-
ies upon sleep/wake. However, the present study has the limitation that it explores only one part of the light period. Since neurobiological and behavioural studies upon sleep/wake states are often conducted throughout a 24 h period, including light and dark phases, the magnitude of the changes reported here should be analysed in further studies to examine DMSO effects upon sleep and wake in an extended period of time. Exploring differences between light and dark periods would be also of great interest. Despite this limitation, the results reported in this study are original since they show for the first time that DMSO changes sleep architecture in rats. The experimental work conducted here does not permit us to know which mechanisms may underlie the effects of DMSO administration upon sleep states, or whether these effects are central or peripheric, or specific or non-specific actions of DMSO, but some explanations could be suggested. Due to its cardiovascular effects, reported diastolic and systolic hypertension, bradycardia and cardiac arrest, DMSO could affect sleep strongly. Also, as an anti-inflammatory agent, DMSO could interfere with sleep intensity. Additionally, through its reactive oxygen species scavenger actions, DMSO could interfere with sleep. Moreover, DMSO can reduce noradrenalin-mediated vasoconstriction, enhance cerebral blood flow (De la Torre et al., 1975), and it is an effective cholinesterase inhibitor (Watts and Hoogmed, 1984). These actions may contribute to its effects on sleep states. Furthermore, DMSO actions on intracellular systems, disrupting normal neurotransmision, may also be responsible for the results observed here. Thus, one possibility is that DMSO is interfering with intracellular calcium concentration and indirectly affects the massive calcium entry during spindles occurring within slow wave sleep. This argument is relevant for the results observed here, given the fact that, in the rat, an increased density of sleep spindles correlates well with the amount of deep SWS, pointing out the role of sleep spindles as the hallmark of consolidated (deep) SWS. Thus, in Wistar rats, spindles increase in number, duration, and amplitude as sleep deepens (Gottesmann, 1992; Terrier and Gottesmann, 1978). Moreover, to complete the findings of the polysomnographic recordings, the analysis of the power spectra of delta and sigma bands during light and deep SWS was conducted in the present study. These analyses show that power spectra within sigma band, mainly reflecting sleep spindles which contribute
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
to the bulk of power found in this 9.38–14.06 Hz range (Bjorvatn et al., 1998), during deep SWS are reduced after the administration of DMSO at 15% and 20% concentrations. This result is consistently found during third hour of recording and for the total 3 h recording period at both concentrations. This finding is accompanied by an increase of the power of delta waves within deep SWS, observed in the total 3 h period of recording after DMSO 20%. Therefore, the results here show a decrease of the power of the sigma band and an increase of the power of delta band after DSMO administration, which differs from the normal progress of sleep described in the rat (Gottesmann, 1992; Terrier and Gottesmann, 1978). There is to say that in the present study, DMSO at 15% also affected sigma band power spectra during light SWS, decreasing the power of sigma waves. However, at 20% concentration, DMSO did not affect significantly power spectra during light SWS. At the moment, it is difficult to assess this finding. Further studies specifically focused on EEG spectral power are needed to clarify these matters. It seems important to notice also that, at present, there is no data about whether DMSO may have a noxious effect on the animal, but if it does, an indirect impact on sleep-wake state by the treatment with DMSO should be expected. However, even though mechanisms through which DMSO affects sleep and wake are unknown so far, the results in this work show that this widely used solvent affects sleep architecture in rats. Being aware of DMSO behavioural effects seems important since experimental artefacts caused by DMSO can lead to the erroneous interpretation of results. Especially since sleep represents a physiological need, which alterations may interfere with several cognitive and behavioural processes. An increase awareness of the multidisciplinary utilisation of DMSO and other commonly used solvents in several research fields can be a valid contribution to avoid or minimise misunderstandings.
5. Conclusion DMSO is a very efficient solvent, widely used in biological studies and also as a vehicle for drug therapy. Little is known about its behavioural or neurotoxical effects. In the present study, DMSO effects upon sleep and wake states in rats are tested. DMSO at 5% and
231
10% has no effect. However, at 15% and 20% concentrations, DMSO interferes with sleep, increasing light slow wave sleep and decreasing deep slow wave sleep. Additionally, power spectral analysis of the sleep EEG shows a decrease in the power of sigma activity after treatment with DMSO at 15% and 20% concentrations. At 20%, DMSO also increased the power density of delta activity during deep SWS. Therefore, DMSO at these frequently used concentrations modifies sleep architecture in rats.
Acknowledgements Experimental procedures are in compliance with the European Communities Council Directive of 24 November 1986 (86/609/EEC). Work financed by Consejer´ıa de Educaci´on y Ciencia, Junta de Andaluc´ıa, Spain.
References Abdullaeva, G., Shakimova, B., 1989. An evaluation of the efficacy of treating rheumatoid arthritis with preparations for local use. Revmatologiia (Mosk) 4, 35–39. Authier, N., Dupuis, E., Kwasiborski, A., Eschalier, A., Coudor´e, F., 2002. Behavioural assessment of dimethylsulfoxide neurotoxicity in rats. Toxicol. Lett. 132, 117–121. Bjorvatn, B., Bjorkum, A., Neckelmann, D., Ursin, R., 1995. Sleep/waking and EEG power spectrum effects of a nonselective serotonin (5-HT) antagonist and a selective 5-HT reuptake inhibitor given alone and in combination. Sleep 18, 451–462. Bjorvatn, B., Fagerland, S., Ursin, R., 1998. EEG power densities (0.5–20 Hz) in different sleep-wake stages in rats. Physiol. Behav. 63, 413–417. Broadwell, R., Salcman, M., Kaplan, R., 1982. Morphologic effect of dimethyl sulfoxide on the blood–brain barrier. Science 217, 164–166. Castro, C., Hogan, J., Benson, K., Shehata, C., Landauer, M., 1995. Behavioural effects of vehicles: DMSO, etanol, Tween-20, and emulphor-620. Pharmacol. Biochem. Behav. 50, 521–526. Cavaletti, G., Oggioni, N., Sala, F., Pezzoni, G., Cavalletti, E., Marmiroli, P., Petruccioli, M.G., Frattola, L., Tredici, G., 2000. Effect on the peripheral nervous system of systemically administered dimethylsulfoxide in the rat: a neurophysiological and pathological study. Toxicol. Lett. 118, 103–107. Corsi-Cabrera, M., Perez-Garci, E., Del R´ıo-Portilla, Y., Ugalde, E., Guevara, M., 2001. EEG bands during wakefulness, slow-wave, and paradoxical sleep as a result of principal component analysis in the rat. Sleep 24, 374–380. Datta, S., Hobson, J.A., 2000. The rat as an experimental model for sleep neurophysiology. Behav. Neurosci. 114, 1239–1244.
232
M. Cavas et al. / Toxicology Letters 157 (2005) 221–232
Davis, J., Rowley, S., Braine, H., Piantadosi, S., Santos, G., 1990. Clinical toxicity of cryopreserved bone marrow graft infusion. Blood 75, 781–786. De la Torre, J., Kawanaga, H., Rowed, D., Johnson, C., Goode, D.J., Mullan, S., 1975. Dimethylsulfoxide in central nervous system trauma. Ann. NY Acad. Sci. 243, 362–389. Demos, C., Beckloff, G., Donin, M., Oliver, P., 1967. Dimethyl sulfoxide in musculoskeletal disorders. Ann. NY Acad. Sci. 147, 516–523. Franco, R., Weiner, M., Wagner, K., Martelo, O., 1983. Incorporation of inositol hexaphosphate into red blood cells mediated by dimethyl sulfoxide. Life Sci. 32, 2763–2768. Goto, I., Yamamoto-Yamagushi, Y., Honma, Y., 1996. Enhancement of sensitivity of human lung adenocarcinoma cells to growthinhibitory activity of interferon ␣ by differentiation inducing agents. Br. J. Cancer 74, 546–554. Gottesmann, C., 1992. Detection of seven sleep-waking stages in the rat. Neurosci. Biobehav. Rev. 16, 31–38. Halsted, B., Youngberg, S., 1981. The DMSO Handbook. Golden Quill Publishers, California. Hameroff, S., Otto, C., Kanel, J., Weinstein, P., Blitt, C., 1983. Acute cardiovascular effects of dimethyl sulfoxide. Ann. NY Acad. Sci. 411, 94–99. Ikeda, Y., Long, D., 1990. Comparative effects of direct and indirect hydroxyl radical scavengers on traumatic brain oedema. Acta Neurochir. Suppl. (Wien) 51, 74–76. Kelly, K., Hill, M., Youkhana, K., Wanker, F., Gimble, J., 1994. Dimethyl sulfoxide modulates NF-B and cytokine activation in lipopolysaccharide-treated murine macrophages. Infect. Immun. 62, 3122–3128. Kingery, W., 1997. A critical review of controlled clinical trials for peripheral neuropathic pain and complex regional pain syndromes. Pain 73, 123–139. Kolb, K., Jaenicke, G., Kramer, M., Schulze, P., 1967. Absorption, distribution and elimination of labeled dimethyl sulfoxide in man and animals. Ann. NY Acad. Sci. 141, 85–95. Maloney, K., Cape, E., Gotman, J., Jones, B., 1997. High-frequency ␥ electroencephalogram activity in association with sleep-wake states and spontaneous behaviors in the rat. Neuroscience 76, 541–555.
Nishimura, M., Jacob, P., Cassel, M., Pitts, L., 1989. Tissue distribution of intravenously administered dimethyl sulfoxide in the normal rat. Drug Metab. Dispos. 17, 224–226. O’Donnell, J., Burnett, A., Sheehan, T., Tansey, P., McDonald, G., 1981. Safety of dimethyl sulphoxide. Lancet 1, 498. Parkin, J., Shea, C., Sant, G., 1997. Intravesical dimethyl sulfoxide (DMSO) for interstitial cystitis—a practical approach. Urology 49, 105–107. Regelson, W., Harkins, S., 1997. Amyloid is not a tombstone—a summation. The primary role for cerebrovascular and CSF dynamics as factors in Alzheimer’s disease (AD): DMSO, fluorocarbon oxygen carriers, thyroid hormonal, and other suggested therapeutic measures. Ann. NY Acad. Sci. 826, 348–374. Robert, C., Guilpin, G., Limoge, A., 1999. Automated sleep staging systems in rats. J. Neurosci. Meth. 88, 111–122. Salim, A., 1992. Role of oxygen-derived free radical scavengers in the management of recurrent attacks of ulcerative colitis: a new approach. J. Lab. Clin. Med. 119, 710–717. Santos, N., Figueira-Coelho, F., Martins-Silva, J., Salda˜na, C., 2003. Multidisciplinary utilization of dimethyl sulfoxide: pharmacological, cellular, and molecular aspects. Biochem. Pharmacol. 65, 1035–1041. Shirley, S., Steward, B., Mirelman, S., 1978. Dimethyl sulfoxide in treatment of inflammatory genitourinary disorders. Urology 11, 215–220. Smith, R., 1992. A comprehensive macrophage-T-lymphocyte theory of schizophrenia. Med. Hypotheses 39, 248–257. Swanson, B., 1985. Medical use of dimethyl sulfoxide (DMSO). Rev. Clin. Basic Pharm. 5, 1–33. Terrier, G., Gottesmann, C., 1978. Study of cortical spindles during sleep in the rat. Brain Res. Bull. 3, 701–706. Ursin, R., Bjorvatn, B., 1998. Sleep-wake and EEG effects following adenosine A1 agonism and antagonism: similarities and interactions with sleep-wake and EEG effects following a serotonin reuptake inhibitor in rats. Sleep Res. On. 1, 119– 127. Watts, J., Hoogmed, R., 1984. Dimethyl sulfoxide inhibition of acetylcholinesterase in the mammalian heart. Biochem. Pharmacol. 33, 365–369.