Electrocorticographic patterns dominated by low-frequency waves in camphor-induced seizures

Camphor is an aromatic terpene compound found in the essential oils of many plants, which has been used for centuries as a herbal medicine, especially in children. However, many studies have shown that camphor may have major side effects, including neurological manifestation, such as seizures. In the present study, we investigated the electrocorticographic patterns of seizures induced by camphor in male adult Wistar rats. Each rat received 400 mg/kg (i.p.) of camphor prior to monitoring by electrocorticography. The application of camphor resulted a rapid evolution to seizure and marked changes in the electrocorticographic readings, which presented characteristics of epileptiform activity, with an increase in the total power wave. The decomposition of the cerebral waves revealed an increase in the delta and theta waves. The analysis of the camphor traces revealed severe ictal activity marked by an increase in the polyspike wave. Our data thus indicate that camphor may cause seizures, leading to tonic–clonic seizures. Clearly, further studies are necessary to better elucidate the mechanisms through which camphor acts on the brain, and to propose potential treatments with anticonvulsant drugs that are effective for the control of the seizures.


Results
Behavior and brain wave patterns in camphor-induced seizure. A rapid evolution to seizure was observed in the animals injected with camphor (CPR). The first stage of the seizure (stage 1) included the raising the vibrissae and immobility, and occurred 202.4 ± 13.8 s after the injection of camphor, followed by head jerking (stage 2), at 248.4 ± 13.3 s. Stage 3, which involved spasms of the forelimbs, began at 281.9 ± 26.5 s, while the final stage was characterized by clonic seizure without any transient loss of the posturing reflex, followed by focal seizure that evoked secondary generalization, starting at 566.0 ± 122.7 s. This final stage lasted 24.56 ± 9.4 min, on average.
The electrocorticographic (ECoG) control presented a regular trace with amplitude of 0.04 mV (Fig. 1A, left) with the majority of the spectral power below 10 Hz, which is normal (Fig. 1B, left). The ECoG of the pentylenetetrazol (PTZ) group (Fig. 1A, center) was characterized by oscillations with peaks of amplitude, typical of a seizure. A cyclical pattern of generalized polyspike-wave discharges (GPWD) can also be noted during wakefulness, with a trace amplitude of 1 mV and the spectral amplitude density increasing to higher frequencies (Fig. 1B, center).
In the case of the camphor, there was a latency of 566.0 ± 122.7 s to the onset of the seizure induced by the substance. The ECoG traces are compatible with a seizure, characterized by repeated and irregular spike-waves and polyspike-wave discharges (Fig. 1A, right), with mean wave amplitude of 1 mV. The ECoG trace of the camphor-induced seizure also presented a prevalence of low-frequency waves, which was also observed in the frequency strength distribution of up to 50 Hz (Fig. 1B, right).
The decomposition of the spectral power distribution revealed greater amplitude of low frequency waves in the camphor group than in the control group. However, the PTZ group presented the greatest oscillations The spectral power distribution of the delta, theta, alpha, beta, and gamma cerebral waves in the three study groups. (D) Distribution of the linear frequency of up to 40 Hz in the three groups. The data are expressed as the means ± SD (n = 9 animals per group; *** p < 0.0001). (PTZ: pentylenetetrazole; CPR: camphor).
In contrast with the pattern recorded for the delta and theta waves, in the case of the waves of above 8 Hz, such as the alpha, beta and gamma waves, while significant variation was found among treatments overall, there was no significant difference between the control and camphor groups. Surprisingly, in fact, these waves were responsible for the greatest differences observed between the PTZ and CPR groups. In the case of the alpha waves, despite the significant overall variation (F (2, 24) = 248.9, p < 0.0001), the camphor group was not different from the control group (control: 0.01 ± 0.002 mV 2 /Hz × 10 -3 vs CPR: 0.12 ± 0.05 mV 2 /Hz × 10 -3 , p = 0.7441), although the alpha power of the PTZ was more than 23-fold greater than that recorded in the CPR group (PTZ: 2.875 ± 0.53 mV 2 /Hz × 10 -3 ; p < 0.0001 for both groups; Fig. 2F). A similar pattern was observed in the case of the beta waves, with significant variation among groups overall (F (2, 24) = 68.45, p < 0.0001), but not between the control and camphor groups (control: 0.003 ± 0.0009 mV 2 /Hz × 10 -3 and CPR: 0.40 ± 0.08 mV 2 /Hz × 10 -3 ; p = 0.8063). By contrast, the PTZ group presented a 16.4-fold increase in beta power in comparison with the CPR group (PTZ: 6.56 ± 2.31 mV 2 /Hz × 10 -3 ; p < 0.0001 for both comparisons; Fig. 2G). In the case of the gamma waves (F (2,24) = 79.49, p < 0.0001), the PTZ presented a 25-fold increase in the wave power in comparison with the camphor group (PTZ: 0.40 ± 0.13 mV 2 /Hz × 10 -3 vs CPR: 0.016 ± 0.004 mV 2 /Hz × 10 -3 ; p < 0.0001; Fig. 2H). Similarly, a significant difference was found only between the control and the PTZ group (control: 0.002 ± 0.0003 mV 2 /Hz × 10 -3 vs PTZ group, p < 0.0001) and not in comparison with the camphor group (p = 0.9155). Clearly, then, our findings indicate that camphor-induced seizures alter only low-frequencies waves.

Discussion
Herbal medicines and essential oils are popular worldwide, especially in oriental medicine, in China, and other parts of Asia. These substances are used for a diverse range of complaints, including coughs, colds, microbial infections, and pain, and are used by approximately 75% of the population in Asia 5 . However, the potential toxicity of herbal medicines has received little attention, especially when taken in unknown quantities in composite substances 10 . There are, however, many case reports of herbal medicines and essential oils causing severe toxicity or even death, and these cases include camphor, whether applied topically, inhaled or orally 2 .
Camphor is used in a range of pharmaceutical formulations, including ointments, oils, vaporizer solutions, patches, mothballs or in its solid form 2 . The mechanisms of camphor toxicity are poorly known, although Vatanparast and Andalib-Lari 11 suggested that camphor causes a direct blockade of the K+ channels and an increase in the influx of Ca 2+ in the neurons, which has serious effects on the brain, such as hallucinations and seizures 2,12 . Camphor may also cause changes in hormones and the reproductive organs 13,14 .
There are many recorded cases of seizures induced by camphor, especially in children 9,12,[15][16][17][18][19] . Despite these records, little is known about this phenomenon. The results of the present study show clearly that camphor is capable of inducing tonic-clonic seizures in rats, with effects similar to those described in humans 8,9,12,15,18 . There are many types of seizure, which present different electroencephalographic patterns and behavioral correlates 20,21 . Here, we described the patterns of camphor-induced seizure, and found moderate neuronal hyperexcitability, with a predominance of low-frequency waves (delta and theta), which are related to tonic-clonic seizures 22 . These findings corroborate those of Jalilifar et al. 23 , which showed that motor alterations and behavioral seizure progression are associated with an increase in delta oscillations.
Interestingly, the wave pattern observed in the camphor-induced seizure presented the characteristics of a generalized spike-wave discharge, concentrated in low frequency waves, below 8 Hz (delta and theta). These frequencies may be faster and more irregular at the onset of the seizure and decelerate gradually towards its termination, while intra-discharge irregularities may also occur, as observed in the ECoG traces obtained from the camphor group. The amplitude of these spike-wave complexes typically peaks over the frontal areas and displays an anterior to posterior gradient, frequently fading over the occipital areas 24 . This pattern is very indicative of absence seizures 24 . Given these characteristics and the available data, we propose that camphor causes absence seizures, which evolve to generalized tonic-clonic seizures. Although only a single pair of electrodes was used in the present study, which may result in lower spatial accuracy than other systems, Johnstone et al 25 and Hemington and Reynolds 26 confirmed the validity of this approach for EEG records and diagnostics.
In the decomposition of the camphor wave profile (Fig. 3), the delta waves of the baseline and mild ictal activity periods were similar, with a significant increase only being observed in the period of severe ictal activity (Fig. 4). A similar pattern was observed in the case of the theta waves. This supports the hypothesis that the camphor-induced seizure was characterized by low frequency waves, with symptomatic manifestation, as described Jalilifar et al. 27 . The wave profile was also marked by the beta waves, which varied significantly between the three periods (Figs. 3E and 4). The available data do in fact highlight the role of the beta waves in generalized seizures, which indicates that this wave can be used to define the transition between the different morphological states, that is, the baseline (Fig. 3A1), mild ictal activity (Fig. 3A-2, spike-wave), and severe icital activity ( Fig. 3A-3, polyspike-wave).
Overall, then, camphor provoked convulsive symptoms, marked by altered behavior and electrocorticographic activity. Changes were also observed in total power and low frequency waves (delta and theta). Our findings show that camphor may cause absence seizures, which may lead to tonic-clonic seizures. This emphasizes the need for further research to better elucidate the mechanisms through which camphor acts on the brain, including whether there is glial cell activity, neuronal death or long-term functional impairment. The preset study nevertheless provides important insights for the development of clinical protocols for the administration of anticonvulsant drugs for seizure control.

Methods
Study animals. Twenty-seven adult males Wistar rats (250 ± 30 g) were obtained from the Central Animal Facility of the Federal University of Pará in Belém, Brazil. The animals were housed at a controlled temperature, of 23 ± 2 ºC and 12/12 h light-dark cycle, with food and water available ad libitum. All experimental procedures were conducted in accordance with the principles of laboratory animal care 28  Description of the seizure-related behavior. Seizure-related behavior was observed following the injection of the camphor, when the animals were maintained in a standard white cage (48 cm × 38 cm × 21 cm). Latency to the seizure was recorded and behavioral modifications were classified in four clinically-identifiable After the baseline, both epileptiform events present the greatest spike voltage in the low frequency waves (delta and theta) and in particular, in the high frequency waves (alpha, beta and gamma), as well as significant increases in β-spectral density from the baseline levels. Baseline period (BLP); Mild Ictal Activity (MIA) and Severe Ictal Activity (SIA). Spike-wave discharges (SWD). Spike-wave (SW). Polyspike and sharp wave discharges (PSWD Electrocorticographic recordings and data analyses. The ECoG recordings and offline data analyses followed the procedures described by Estumano et al. 31 . The animals were first anesthetized with ketamine hydrochloride (80 mg/kg, i.p.) and xylazine hydrochloride (10 mg/kg, i.p), and once the corneal reflex had been abolished, they were placed in a stereotaxic apparatus and stainless steel electrodes (1.0 mm diameter tip exposure) were placed on the duramater at the coordinates of the bregma-0.96 mm and ± 1.0 mm lateral. Five days after surgery, the ECoGs were obtained using a digital data acquisition system and the offline analyses were conducted. The analyses were run at a frequency of up to 50 Hz, and split into bands as in Estumano et al. 31 , that is, delta (1-4 Hz), theta (4-8 Hz), alpha (8-12 Hz), beta (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28), and gamma (28-40 Hz), for interpretation of the wave dynamics during the development of the seizure. The recordings followed a predefined protocol: the animals were immobilized carefully for 10 min for habituation, to avoid interference in the records. The basal ECoG activity was then recorded for 30 min, for use as the control in the ECoG analyses. The PTZ (positive control) or camphor was then administered, and the ECoG activity was recorded for a further 30 min. The animals were then euthanized to avoid further distress.
Statistical analyses. The normality of the data was verified using the Kolmogorov-Smirnov test, and the homogeneity of the variances was confirmed by the Levene statistic. As the residuals were distributed normally and the variances were homogeneous, comparisons of the mean amplitudes of the treatment and control values were based on a one-way ANOVA followed by Tukey's test for multiple comparisons. The data are presented as the mean ± standard deviation (SD), and the F and p values are included, when pertinent. A p < 0.05 significance level was considered for all analyses. All statistical procedures were un in the GraphPad Prism software, version 8 (Graph-Pad Software Inc., San Diego, CA, USA).