Dose-dependent anticonvulsant effects of linoleic and α-linolenic polyunsaturated fatty acids on pentylenetetrazol induced seizures in rats

Authors

  • Ameer Y. Taha,

    1. Department of Pharmacology, Faculty of Medicine, University of Toronto, Toronto, Canada
    2. University of Toronto Epilepsy Research Program, Faculty of Medicine, University of Toronto, Toronto, Canada
    Search for more papers by this author
  • Elvis Filo,

    1. Department of Pharmacology, Faculty of Medicine, University of Toronto, Toronto, Canada
    Search for more papers by this author
  • David W. L. Ma,

    1. Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, Canada
    2. Department of Human Health and Nutritional Sciences, College of Biological Science, University of Guelph, Guelph, Canada
    Search for more papers by this author
  • W. McIntyre Burnham

    1. Department of Pharmacology, Faculty of Medicine, University of Toronto, Toronto, Canada
    2. University of Toronto Epilepsy Research Program, Faculty of Medicine, University of Toronto, Toronto, Canada
    Search for more papers by this author

Address correspondence to Ameer Y. Taha, Department of Pharmaco-logy, University of Toronto, Medical Sciences Building, 1 King's College Circle, Toronto, ON, Canada. E-mail: a.taha@utoronto.ca

Summary

Purpose: Linoleic and α-linolenic polyunsaturated fatty acids, derived from plant oils, have been reported to reduce neuronal excitability ex vivo and in cell culture. The evidence derived from animal seizure models, however, has been contradictory. The goal of the present study was to assess the dose-dependent anticonvulsant effects of a fatty acid mixture containing linoleic and α-linolenic acids in a 4 to 1 ratio (the “SR-3” compound).

Methods: The maximal pentylenetetrazol seizure model and Long-Evans hooded rats were used.

Results: Daily intraperitoneal injection of SR-3 for 21 consecutive days raised omega-3 polyunsaturated fatty acid (n-3 PUFA) composition in the unesterified fatty acid fraction of brain lipids (p  < 0.05), and increased latency to seizure onset when administered at 200  mg/kg (p < 0.05), but not at 40  mg/kg (p > 0.05). There were no significant effects of SR-3 on seizure occurrence or on seizure severity (p > 0.05). A toxic effect of the SR-3 compound on peristalsis was observed at a dose of 400  mg/kg and above.

Conclusion: Linoleic and α-linolenic polyunsaturated fatty acids in a 4 to 1  ratio raises n-3 PUFA composition of unesterified fatty acids in the brain and increases resistance to pentylenetetrazol-induced seizures.

Epilepsy is a neurological disorder characterized by spontaneous, recurrent seizures (Burnham, 2006), which can be controlled in 60%–70% of patients by the use of anticonvulsant medications (Shorvon, 1996; Vining, 1999). Patients using anticonvulsants often experience drug-related side effects, such as fatigue, sedation, and nausea (Vining, 1999). Thus, new and less toxic therapies are required for people with epilepsy.

n-3 polyunsaturated fatty acids (n-3 PUFA), derived from seafood and plants such as flax, have been considered as a complementary to drug treatment for patients with epilepsy (Schlanger et al., 2002; Yuen et al., 2005; Bromfield et al., 2008). n-3 PUFA are diet-derived lipids, which are essential for normal brain function and development (Clandinin et al., 1980; Crawford et al., 2003). They are important structural components of neuronal membranes and are involved in modulating neurotransmission, cell signaling, and gene regulation (Rapoport, 2003; Kitajka et al., 2004).

The most abundant n-3 PUFA in the brain is docosahexaenoic acid (DHA; 22:6n-3). DHA, destined for the brain, can be synthesized in the liver from α-linolenic acid. It has been suggested that DHA synthesis is optimal when the n-6 PUFA linoleic acid is also present at a specific ratio. In particular, it has been proposed that a 4 to 1 ratio of linoleic acid and α-linolenic acid is best for raising brain DHA levels in rats. Linoleic and α-linolenic acid in a 4 to 1 ratio has been termed the “SR-3 compound” (Yehuda et al., 1996).

The best evidence suggesting possible anticonvulsant properties for n-3 PUFA, including α-linolenic, eicosapentaenoic, and docosahexaenoic acids, has come from studies involving cell cultures and ex vivo preparations (Vreugdenhil et al., 1996; Xiao & Li, 1999; Lauritzen et al., 2000; Young et al., 2000). These studies have shown that n-3 PUFA confer protection against seizures by increasing the threshold for action potentials and by extending the refractory period in neurons. This action appears to result from a partial inhibition of sodium and calcium voltage-gated channels.

The evidence derived from in vivo studies in animal seizure models, however, has been contradictory. It has been reported that coadministration of linoleic acid with α-linolenic acid in a 4 to1 ratio (i.e., the “SR-3” compound) raises brain DHA levels in rats (Yehuda et al., 1996) and increases resistance to pentylenetetrazol (PTZ)-induced seizures when it is injected intraperitoneally (i.p.) at 40  mg/kg for 21 consecutive days (Yehuda et al., 1994; Rabinovitz et al., 2004). These findings, however, have failed to be replicated in a recent study that used the same dose of the SR-3 mixture and the same seizure model (Taha et al., 2006).

Actually, it would be surprising if a dose of 40  mg/kg of the SR-3 mixture could increase resistance to PTZ-induced seizures, because the 40 mg/kg dose represents an increase of only 1.2% in a rat's daily intake of these fatty acids. Higher SR-3 doses would be more likely to achieve a physiologically relevant rise in brain DHA, and this might be accompanied by a significant increase in seizure threshold.

The goal of the present study was to assess the possible anticonvulsant effects of the SR-3 mixture in a dose-response paradigm involving higher doses of the compound. n-3 PUFA composition in brain phospholipids and unesterified fatty acids was also determined.

Materials and Methods

SR-3 preparation

The SR-3 compound was prepared by mixing 0.90 mg/ml nonesterified linoelic acid and 0.92  mg/ml nonesterified α-linolenic acid at a 4 to 1 ratio, in a vehicle that consisted of 0.73  ml mineral oil and 0.02  ml α-tocopherol (all from Sigma-Aldrich, St. Louis, MO, U.S.A.). Four different doses of the SR-3 compound were prepared: 40, 200, 400, and 1000  mg/kg. Each dose was dissolved in the same fixed volume of 0.73  ml mineral oil and 0.02  ml α-tocopherol. After preparation, the SR-3 mixture was stored at −20°C, until further use, to minimize oxidation of the linoleic and α-linolenic polyunsaturated fatty acids.

Subjects and treatments

All experimental procedures were conducted in accordance to the guidelines of the Canadian Council of Animal Care and approved by the Animal Care Committee of the Faculty of Medicine of the University of Toronto.

One-month-old male Long Evans Hooded rats (Charles River, La Prairie, QC, Canada), weighing on average 151  g at the start of the experiment, served as subjects. Subjects were individually housed in plastic cages with corn-cob bedding in a vivarium maintained on a 12  h light, 12  h dark cycle (lights on at 7 a.m.), and at a temperature of 21°C. Water and rat chow (2018 18% Protein Rodent Diet; Teklad Global, Madison, WI, U.S.A.) were available ad libitum. The rat chow contained (in g/kg diet) 189 protein, 60 fat, 554 carbohydrates, 38 fiber, 59 ash, and 100 moisture. The fat component of the diet mainly contained (in percent of total fatty acids) palmitate (13.5%), stearate (2.7%), oleate (22.3%), linoleate (55.5%), and α-linolenate (4.9%).

After 7 days in the facility, subjects were randomly divided into five groups that initially received daily i.p. injections (starting at 11 a.m.) of: (1) 0.9% saline (0.035  ml, n  = 10); (2) mineral oil vehicle (0.035  ml, n  = 7); (3) SR-3 40  mg/kg (in 0.035  ml vehicle, n  = 8); (4) SR-3 400  mg/kg (in 0.035  ml vehicle, n  = 8); or (5) SR-3 1000  mg/kg (in 0.035  ml vehicle, n  = 8). The saline and mineral oil groups served as controls. All subjects were intended to be injected with their respective treatments for 21 consecutive days. By the tenth day of the experiment, however, it was clear that the injections were causing toxicity in the 1000  mg/kg group. The symptoms consisted of low weight gain, low food intake, and bloating, which appeared to be caused by impaired peristalsis. This group was therefore terminated, and the subjects were euthanized with CO2. These symptoms were also evident to a lesser extent in the group that received the 400  mg/kg daily dose. Therefore, from days 10 to 15, these rats were injected with saline and not the SR-3. The bloating was gone, and food intake had returned to normal by day  16. The injections were therefore resumed, but at a lower dose of 200  mg/kg. These animals then became the 200  mg/kg group. Overall, they received 16 days of SR-3 injections, the last 9 days at 200  mg/kg.

All subjects were weighed each day prior to receiving the injections. Food intake was also measured every day by measuring the difference in weight in the stainless steel dish that contained the food.

Seizure testing

On day 22, after 21 days of treatment, the subjects were weighed and subsequently seizure-tested using the maximal pentylenetetrazol (PTZ) procedure (Fisher, 1989), starting at 11 a.m. At high (maximal) doses, PTZ models tonic–clonic generalized seizure attacks in humans (Fisher, 1989). As previously described (Taha et al., 2006), subjects received 80  mg/kg PTZ via the i.p route. Subjects were then placed in an open field and videotaped for 30 min. Videotapes were subsequently scored by two independent blinded observers. Seizure latency, severity, and occurrence were scored. Seizure latency (seconds) was scored as the interval between injection and the onset of a myclonic jerk or forelimb clonus. Seizure severity was scored according to the following scale: Stage 1, myoclonic jerks; stage 2, forelimb or hindlimb clonus; stage 3, forelimb or hindlimb tonus; and stage 4, running fits. Scores were averaged based on the maximum seizure score displayed by the subject over the 30-min observation period in order to yield a measure of seizure severity (out of 4).  Subjects that displayed a running fit, however, were immediately euthanized using a lethal injection of sodium pentobarbital (MTC Pharmaceuticals, Cambridge, ON, Canada), and their seizure score was ranked as stage 4. Accordingly, 50%, 25%, 14%, and 13% of the saline, mineral oil, SR-3 40  mg/kg and SR-3 200  mg/kg groups were respectively euthanized due to a stage 4 running fit. Seizure occurrence was scored according to the severity scale, seizures being scored as present if any of the stages described above was present and absent if none of them was present.

At the end of the 30-min seizure test, subjects were euthanized via a lethal i.p. injection of sodium pentobarbital, following which the whole brain and liver were excised and snap-frozen in liquid nitrogen. The samples were stored at −80°C for later analysis.

Brain lipid analysis

The left hemisphere of the brains was used for phospholipid and unesterified free fatty acid analysis, with the right hemisphere being reserved for possible future analyses. Total lipids were extracted according to the extraction method of Folch et  al. (1957), following the addition of 1 mg diheptadecanoyl L-α-phosphatidylcholine and 1.5  mg nonesterified heptadecaenoic acid (Sigma, St. Louis, MO, U.S.A.) in chloroform as internal standards, to approximately 0.8 g brain tissue. The samples were then homogenized in chloroform/methanol (2:1, vol/vol) to extract total lipids. Saline (0.9%) was added an hour later in order to separate the polar phase. Following distinct separation of the phases, the lower chloroform phase was transferred to new 15-ml glass screw cap tubes with Teflon-lined caps, dried under a gentle stream of nitrogen, and reconstituted in 2  ml hexane.

Phospholipids and free fatty acids in the brain total lipid extracts were fractionated by lipid thin-layer chromatography (TLC) using 20 × 20 cm silica gel plates (LK6D plates, precoated with 250 μm Silica Gel 60A; Whatman, Florham Park, NJ, U.S.A.). Separate lanes were spotted with phospholipids or free fatty acid standards. The plates were developed  using  hexane, diethyl ether, and acetic acid (80:20:1 by volume) in covered glass tanks for 35 min. Bands corresponding to phospholipids and free fatty acids were viewed under ultraviolet light after lightly spraying with 8-anilino-1-naphthalenesulfonic acid. The bands were scraped off each plate into 15-ml glass screw cap tubes with Teflon-lined caps and directly methylated by incubating with 2   ml hexane and 2  ml 14% methanolic BF3 at 100°C for 1  h. Deionized water (2  ml) was then added to separate the phases. The upper hexane phase was extracted, dried under nitrogen, and reconstituted in hexane for analysis by gas  chromatography.

Fatty acid composition of the SR-3 compound

The fatty acid composition of each component of the SR-3 mixture (i.e., linoleic acid, α-linolenic acid, and mineral oil) was verified by gas chromatography. Approximately 0.1  ml each compound was dissolved in 2  ml hexane and directly derivatized in 14% boron trifluoride in methanol (2 ml; Sigma) for 1  h at 100°C. Deionized water (2  ml) was then added to separate the phases. The upper hexane phase was extracted, dried under nitrogen, and reconstituted in hexane for analysis by gas chromatography.

Fatty acid methyl ester analysis by gas-chromatography

Fatty acid methyl esters (FAME) in phospholipids and unesterified fatty acids of brain were analyzed using an Agilent 6890 gas-chromatography system (Agilent Technologies, Santa Clara, CA, U.S.A.) equipped with a flame ionization detector and a SP2560 fused silica capillary column [100 m, 0.25 μm film thickness, 0.25 mm inner diameter (ID); Supelco, Sigma-Aldrich, St. Louis, MO, U.S.A.). One microliter FAME from each sample was injected in splitless mode. The injector and detector ports were set at 250°C. Methyl esters were eluted using a temperature program set initially at 60°C for 5 min, increasing 10°C/min until 170°C, then 5°C/min until 175°C, 2°C/min until 185°C, 1°C/min until 190°C, and 10°C/min until 240°C. Helium was used as a carrier gas, at a constant flow rate of 1.3 ml/min. Fatty acid peaks were identified by comparing the retention time of each peak against the retention times of an authentic fatty acid standard of known composition (GLC463; NuCheck Prep., Elysion, MN, U.S.A.).

Fatty acid profiles of the SR-3 constituents were determined on a 30 m × 25 mm capillary column (DB-23; J & W Scientific, Folsom, CA, U.S.A.) in the Agilent 6890 gas-chromatography system equipped with a flame ionization detector. One microliter FAME from each sample was injected into the column in splitless mode, using helium gas as a carrier, at a constant flow rate of 0.7 ml/min. A three-stage temperature program was used to acquire the FAME profile. Initial temperature setting was at 50°C with a 2-min hold, followed by a ramp up at 20°C/min to 170°C and a 1-min hold, and a final 3°C/min ramp up to 212°C, followed by a 10-min hold. Fatty acid peaks were identified by comparing the retention time of each peak against the retention times of authentic fatty acid standards of known composition (GLC463).

Data presentation and statistical analysis

All data are presented as means ± SEM. The fatty acid profile data for brain phospholipids and unesterified fatty acids are expressed as a percentage of total fatty acids and not absolute concentrations (mg/g wet tissue). This is because (1) no significant differences between the groups were observed in the total pool of phospholipids and unesterified fatty acids, and therefore the fatty acid percent composition data in general reflected the absolute concentration data, and (2) the variability in the percent composition data is lower as compared to the absolute concentration data, and therefore the chances of a type II error are minimized. Data analysis was performed on Statistical Analysis Software (version 8.02; SAS Institute, Cary, NC, U.S.A.). The group that received the SR-3 at 1000  mg/kg was excluded from the statistical analyses, because it was discontinued. A two-way repeated measures analysis of variance (ANOVA) was used to determine the effects of treatment and time on body weight gain and food intake. A one-way ANOVA was used to determine the effect of treatment on seizure latency, severity, and fatty acid concentrations of phospholipids and free fatty acids. Outliers falling more than 2 standard deviations from the mean were excluded from the statistical analyses. Post hoc Tukey t-tests were applied when appropriate. The chi-square (χ2) test was used to assess differences in seizure occurrence. Statistical significance was accepted at p  < 0.05.

Results

Fatty acid profile of the SR-3 constituents

The fatty acid composition of the components used to mix the SR-3 compound was determined in each stock bottle by gas chromatography. The composition of linoleic acid and α-linolenic acid, expressed as a percentage of total fatty acids, was 95.8 ± 1.9 (n  = 3) and 99.7 ± 0.3 (n  = 3), respectively. As expected, no fatty acids were detected in the mineral oil (n  = 2).

Body weights

Body weights of control and experimental subjects are presented in  Fig.  1. All subjects gained weight over time (p  < 0.05). There was a significant difference in body weights, however, between controls and the experimental subjects that received the SR-3 at 200  mg/kg (p  < 0.05). Weight gain in the SR-3 200  mg/kg rats was significantly lower than weight gain in the control saline and mineral oil subjects or the subjects that received the SR-3 at 40  mg/kg (p  < 0.05). In the SR-3 200  mg/kg group, the lower body weight was evident on day  4 and persisted until the end of the experiment.

Figure 1.

 Effect of treatment on body weight gain. The effects of daily saline (blue circle, solid line), mineral oil (orange square, solid line), SR-3 40  mg/kg (green diamonds, dotted line), SR-3 200  mg/kg (red triangle, dotted line), and SR-3 1000  mg/kg (green-blue circle, dotted line) injections on body weight gain. Subjects that received the SR-3 at 1000  mg/kg were excluded because they were terminated before the study ended. Body weight gain of the SR-3 200  mg/kg group was significantly lower over time, as compared to the saline, mineral oil, and SR-3 40  mg/kg groups: p  < 0.05 for significant main effect of treatment and time on body weight gain by two-way repeated measures ANOVA.

Food intake

The data related to food intake are presented in  Fig.  2. All rats consumed more food over time (p  < 0.05). Food intake, however, in subjects that received the SR-3 200  mg/kg was significantly lower, as compared to food intake in subjects that were injected with saline, mineral oil, or SR-3 40  mg/kg (p  < 0.05). The differences in food intake between the SR-3 200  mg/kg and the other groups were no longer statistically different during the last 10 days of the experiment.

Figure 2.

 Effect of treatment on food intake. The effects of saline (blue circle, solid line), mineral oil (orange square, solid line), SR-3 40  mg/kg (green diamonds, dotted line), SR-3 200  mg/kg (red triangle, dotted line), and SR-3 1000  mg/kg (green-blue circle, dotted line) injections on food intake. Subjects that received the SR-3 at 1000  mg/kg were excluded because they were terminated before the study ended. Food intake of the SR-3 200  mg/kg group was significantly lower during the first 12 days, as compared to the saline, mineral oil, and SR-3 40  mg/kg groups: p  < 0.05 for significant main effect of treatment and time on food intake and an interaction between treatment and time by two-way repeated measures ANOVA.

Possible physiological signs of toxicity—liver weight and percent liver of body weight

Liver weight and liver weight as a percentage of total body weight are indirect markers of treatment-induced toxicity, with greater liver mass being indicative of potential treatment-induced stress. The data for liver weight and liver percentage of body weight are presented in  Fig.  3, A and B, respectively. As shown in  Fig.  3A, liver weight was significantly lower in rats that received the SR-3 200  mg/kg treatment, as compared to those that received the saline, mineral oil, or SR-3 40  mg/kg treatments (p  < 0.05). As indicated by  Fig.  3B, however, no significant differences were observed between the groups when the liver weight was expressed as a percentage of total body weight (p > 0.05).

Figure 3.

 Liver weight and percent liver weight of total body weight. Data are mean ± SEM of n  = 7–9 for each group. Bars marked with different letters differed significantly from each other, as determined by one-way ANOVA and Tukey's post hoc test. (A) Liver weight. The difference between the SR-3 200  mg/kg group and the other groups was statistically significant. (B) Percent liver weight of total body weight. There were no significant differences among the different groups.

Seizure occurrence

All animals in the control and experimental groups exhibited seizure activity after PTZ administration, except for one rat in the saline group and one rat in the SR-3 200 mg/kg group. There were no statistically significant differences between the groups (p > 0.05).

Seizure latency

The data related to seizure latency are presented in  Fig.  4. Outliers that were excluded from the statistical analysis included one rat in the saline group, one rat in the mineral oil group, and one rat in the SR-200  mg/kg group, with latencies of 5.8 min, 3.4 min, and 12.3 min, respectively. It is not surprising to have a few rats that do not seize within the first 2 min following PTZ injection (Krall et al., 1978; Taha et al., 2006). A dose titration was not performed in a subset of rats, and it is therefore likely that the PTZ dose used in this study was below 100% of the effective dose (ED100).

Figure 4.

 Seizure latency following PTZ administration. Data are mean ± SEM of n  = 6–8 for each group. Bars marked with different letters differed significantly from each other, as determined by one-way ANOVA and Tukey's post hoc test. The difference between the SR-3 200 mg/kg group and the other groups was statistically significant.

As indicated by  Fig.  4, subjects that received the SR-3 at 200  mg/kg exhibited an increase in latency to seizure onset. Their latency was approximately three-fold higher, as compared to latencies in the other three groups, which had similar, shorter latencies. Statistical analysis showed that seizure latency in the SR-3 200  mg/kg group was significantly greater than latencies in the saline, mineral oil, and SR-3 40  mg/kg groups (p  < 0.05).

Seizure severity

Seizure scores were obtained for all the subjects that exhibited seizure activity and were averaged within each experimental group. Averaged seizure scores were similar in all of the four experimental groups, with group averages being 3.2 ± 0.3, 3.1 ± 0.2, 3.3 ± 0.2, and 2.3 ± 0.4 for the saline, mineral oil, SR-3 40, and SR-3 200 groups, respectively. Statistical analysis revealed no significant differences among the groups (p = 0.2), although a strong tendency towards a reduction in seizure severity was noticeable in the SR-3 200  mg/kg group.

Brain phospholipid fatty acid composition

In order to define differences in the localization of n-3 PUFA within brain lipids, total lipid extracts were subjected to TLC to separate membrane phospholipids and unesterified free fatty acids.

The data for brain phospholipid fatty acid profile are shown in  Table  1. Total brain phospholipid concentrations, expressed as  mg/g tissue, as well as phospholipid fatty acid percent composition, expressed as a percentage of total phospholipids, did not differ significantly between the groups (p > 0.05).

Table 1.   Brain phospholipid fatty acid composition, expressed as a percentage of total fatty acids within the phospholipid lipid pool
 SalineMineral oilSR-3 40  mg/kgSR-3 200  mg/kg
  1. Data are mean ± SEM of n  = 8–10/group; p  > 0.05 by one-way ANOVA. ND, not detected.

14:00.2 ± 0.030.1 ± 0.010.1 ± 0.010.1 ± 0.01
16:022.9 ± 0.822.3 ± 1.021.8 ± 1.422.0 ± 0.8
18:018.7 ± 0.517.1 ± 0.616.5 ± 0.517.1 ± 0.5
19:00.2 ± 0.10.5 ± 0.20.8 ± 0.10.8 ± 0.1
20:01.0 ± 0.22.0 ± 0.32.4 ± 0.42.6 ± 0.4
22:01.0 ± 0.30.9 ± 0.21.0 ± 0.30.9 ± 0.3
24:00.9 ± 0.30.6 ± 0.10.4 ± 0.050.5 ± 0.1
16:1 n-90.02 ± 0.0050.02 ± 0.010.02 ± 0.010.01 ± 0.01
16:1 t-90.3 ± 0.030.3 ± 0.010.3 ± 0.020.2 ± 0.01
18:1 t-9/t110.1 ± 0.10.3 ± 0.10.1 ± 0.10.2 ± 0.1
18:1 n-919.6 ± 0.418.1 ± 0.518.0 ± 0.417.8 ± 0.5
18:1 n-114.2 ± 0.14.1 ± 0.14.2 ± 0.14.0 ± 0.1
20:1 n-50.1 ± 0.10.1 ± 0.10.3 ± 0.20.1 ± 0.1
20:1 n-80.2 ± 0.10.9 ± 0.40.6 ± 0.20.5 ± 0.2
20:1 n-112.2 ± 0.12.1 ± 0.22.2 ± 0.22.2 ± 0.2
22:1 n-90.3 ± 0.10.3 ± 0.10.3 ± 0.10.3 ± 0.1
24:1 n-90.4 ± 0.040.5 ± 0.10.3 ± 0.040.3 ± 0.04
18:2 n-62.4 ± 0.54.4 ± 0.65.2 ± 0.84.7 ± 0.6
20:2 n-60.5 ± 0.10.6 ± 0.21.1 ± 0.30.7 ± 0.3
20:3 n-60.5 ± 0.10.7 ± 0.20.8 ± 0.20.6 ± 0.1
20:4 n-68.9 ± 0.18.7 ± 0.28.3 ± 0.38.7 ± 0.3
22:2 n-60.2 ± 0.10.3 ± 0.10.2 ± 0.10.2 ± 0.1
22:4 n-62.9 ± 0.12.7 ± 0.12.6 ± 0.12.7 ± 0.1
22:5 n-60.4 ± 0.020.4 ± 0.030.4 ± 0.020.4 ± 0.02
18:3 n-30.8 ± 0.11.1 ± 0.21.2 ± 0.31.2 ± 0.2
20:3 n-30.2 ± 0.10.2 ± 0.10.3 ± 0.10.2 ± 0.1
20:5 n-30.03 ± 0.020.005 ± 0.0030.02 ± 0.01ND
22:3 n-3ND0.1 ± 0.050.1 ± 0.030.1 ± 0.02
22:5 n-30.1 ± 0.010.1 ± 0.010.1 ± 0.010.1 ± 0.01
22:6 n-310.7 ± 0.310.5 ± 0.510.3 ± 0.410.7 ± 0.5
Total saturates44.9 ± 0.743.4 ± 0.943.1 ± 1.244.1 ± 0.5
Total monounsaturates27.4 ± 0.626.7 ± 0.426.3 ± 0.425.6 ± 0.5
Total polyunsaturates27.7 ± 0.529.9 ± 0.730.6 ± 0.930.3 ± 0.7
Total n-6 polyunsaturates15.9 ± 0.517.8 ± 0.818.5 ± 1.018.0 ± 0.7
Total n-3 polyunsaturates11.9 ± 0.212.1 ± 0.212.1 ± 0.212.3 ± 0.4
n-6/n-31.3 ± 0.11.5 ± 0.11.5 ± 0.11.5 ± 0.1
Total phospholipids (mg/g)19.7 ± 1.320.5 ± 1.223.4 ± 1.920.4 ± 2.1

Brain unesterified fatty acid composition

The data related to brain fatty acid profile of the unesterified fatty acid fraction are presented in Table 2. Total concentrations of unesterified fatty acids did not significantly differ between the groups (p > 0.05). There were no significant differences in the percent composition of total saturated, monounsaturated, and polyunsaturated fatty acids within the unesterified fatty acid fraction (p > 0.05). Total n-3 PUFA percent composition, however, was highest in the SR-3 200 mg/kg group, as compared to the saline, mineral oil, or SR-3 40 mg/kg groups. Statistical analysis showed that n-3 PUFA percent composition in the SR-3 200 mg/kg group was significantly different from n-3 PUFA percent composition in the saline group (p < 0.05), but not the mineral oil and SR-3 40 mg/kg groups. Total n-6 PUFA percent composition was significantly lower in the SR-3 40 mg/kg group as compared to the saline and mineral oil groups (p < 0.05), but not the SR-3 200 mg/kg group. The n-6 to n-3 PUFA ratio was significantly lower in the SR-3 40  mg/kg and SR-3 200 mg/kg groups as compared to the saline and mineral oil groups (p < 0.05). This was due to the increase in the percent composition of n-3 PUFA in the SR-3 200 mg/kg group and the slight decrease in n-6 PUFA percent composition in the SR-3 40 mg/kg group.

Table 2.   Brain unesterified free fatty acid composition, expressed as a percentage of total fatty acids within the free fatty acid lipid pool
 SalineMineral oilSR-3 40  mg/kgSR-3 200  mg/kg
  1. Data are mean ±sem of n  = 8–10/group. Values with different superscripts significantly different at p  < 0.05 by one-way ANOVA and Tukey's post hoc test. ND, not detected.

14:01.2 ± 0.40.6 ± 0.20.8 ± 0.32.1 ± 1.0
16:015.7 ± 0.914.3 ± 0.518.2 ± 1.213.5 ± 2.1
18:024.7 ± 1.123.6 ± 0.724.3 ± 1.221.2 ± 2.9
19:01.3 ± 0.40.8 ± 0.20.6 ± 0.11.4 ± 0.5
20:03.0 ± 1.3a9.0 ± 1.4b3.8 ± 0.9a3.8 ± 0.6a
22:03.9 ± 0.81.8 ± 0.71.6 ± 0.62.3 ± 0.5
24:03.9 ± 0.91.8 ± 0.71.8 ± 0.62.9 ± 0.6
16:1 n-9ND ND0.01 ± 0.01ND
16:1 t-90.3 ± 0.050.4 ± 0.020.3 ± 0.010.3 ± 0.03
18:1 t9/t110.1 ± 0.040.1 ± 0.10.1 ± 0.020.4 ± 0.2
18:1 n-911.2 ± 0.812.6 ± 0.514.6 ± 1.411.3 ± 1.5
18:1 n-72.9 ± 0.2a3.1 ± 0.1a,b3.6 ± 0.2b3.2 ± 0.1a,b
20:1 n-50.03 ± 0.03ND0.01 ± 0.01ND
20:1 n-80.1 ± 0.1ND0.03 ± 0.020.01 ± 0.01
20:1 n-111.5 ± 0.41.8 ± 0.81.3 ± 0.22.5 ± 0.8
22:1 n-90.3 ± 0.20.2 ± 0.010.2 ± 0.010.5 ± 0.2
24:1n-90.2 ± 0.10.2 ± 0.030.2 ± 0.031.1 ± 0.6
18:2 n-69.1 ± 1.7 a6.5 ± 0.9a,b2.5 ± 0.4b3.8 ± 0.5b
20:2 n-60.1 ± 0.040.1 ± 0.030.2 ± 0.021.1 ± 0.6
20:3 n-60.4 ± 0.10.5 ± 0.10.5 ± 0.10.4 ± 0.1
20:4 n-613.0 ± 1.614.9 ± 0.714.5 ± 1.114.2 ± 1.1
22:2 n-60.1 ± 0.030.1 ± 0.030.1 ± 0.020.03 ± 0.02
22:4 n-61.1 ± 0.11.3 ± 0.21.7 ± 0.32.1 ± 0.4
22:5 n-60.2 ± 0.10.2 ± 0.10.3 ± 0.041.3 ± 0.7
18:3 n-30.4 ± 0.10.6 ± 0.10.6 ± 0.10.8 ± 0.2
20:3 n-30.1 ± 0.040.1 ± 0.020.1 ± 0.020.9 ± 0.5
20:5 n-30.1 ± 0.030.1 ± 0.030.1 ± 0.032.1 ± 1.3
22:3 n-31.1 ± 0.30.7 ± 0.11.4 ± 0.41.2 ± 0.2
22:5 n-30.4 ± 0.10.3 ± 0.10.4 ± 0.10.9 ± 0.5
22:6 n-33.8 ± 0.64.4 ± 0.66.1 ± 1.14.8 ± 1.0
Total saturates53.6 ± 1.951.8 ± 1.551.1 ± 1.747.2 ± 3.1
Total monounsaturates16.6 ± 1.218.3 ± 1.120.5 ± 1.819.3 ± 1.0
Total polyunsaturates29.8 ± 1.329.8 ± 1.128.5 ± 0.733.5 ± 2.8
Total n-6 polyunsaturates24.0 ± 0.9a23.6 ± 0.6a19.8 ± 1.1b22.9 ± 0.9a,b
Total n-3 polyunsaturates5.8 ± 0.6a6.2 ± 0.6a,b8.6 ± 0.8a,b10.6 ± 2.2b
n-6/n-34.4 ± 0.3a 4.1 ±0.4a,b2.5 ± 0.3b2.8 ± 0.5b
Total fatty acids (mg/g)1.8 ± 0.51.0 ± 0.12.8 ± 1.11.3 ± 0.3

Correlation between seizure latency and n-3 PUFA levels within the unesterified fatty acid composition

Pearson's correlation analysis was performed to determine whether the observed changes in seizure latency were correlated with n-3 PUFA levels in the unesterified fatty acid fraction. As shown in  Fig. 5, seizure latency was positively correlated to n-3 PUFA percent composition within the unesterified fatty acid fraction (r = 0.65, p < 0.001).

Figure 5.

 Correlation between seizure latency and brain n-3 PUFA composition. Correlation between seizure latency and n-3 PUFA composition of brain unesterified fatty acids. Blue circles represent the saline group, orange squares represent the mineral oil group, green diamonds represent the SR-3 40  mg/kg group, and red triangles represent the SR-3 200 mg/kg group. Seizure latency was positively correlated to n-3 PUFA composition of brain unesterified fatty acids (r = 0.65, p < 0.001 by Pearson's correlation).

Discussion

The findings of the present study suggest that the chronic administration of the SR-3 compound, a mixture of linoleic and α-linolenic and acids, significantly increases seizure latency in the maximal PTZ model at a dose of 200  mg/kg. No effect was seen at the lower dose of 40  mg, whereas toxicity was seen at the higher doses of 400 and 1000  mg/kg.

The finding of increased seizure latency at 200  mg/kg suggests that chronic administration of n-3 PUFA has the ability to increase seizure threshold. These data are in general agreement with past studies that have shown anticonvulsant properties of n-3 PUFA in cell cultures and ex vivo preparations (Vreugdenhil et al., 1996; Xiao & Li, 1999; Lauritzen et al., 2000; Young et al., 2000).

The failure to find anticonvulsant effects at a dose of 40   mg/kg of the SR-3 preparation is in agreement with the past findings of our own research group (Taha et al., 2006) and in contrast to past reports by Yehuda and colleagues (Yehuda et al., 1994; Rabinovitz et al., 2004). The reasons for these differing findings are not clear, since similar experimental paradigms and the same strain of rats were used in the different experiments (Rabinovitz et al., 2004). The higher dose of 200  mg/kg achieved a larger rise in brain n-3 PUFA composition within the unesterified fatty acid fraction, and this was accompanied by a significant increase in seizure latency.

Brain n-3 PUFA percent composition in the unesterified fatty acid, but not the phospholipid fraction was highest in the animals that received the SR-3 200  mg/kg ( Table  2). This was associated with longer seizure latency ( Fig.  5), suggesting that n-3 PUFA, in their unesterified form, as opposed to their incorporated form, protect against seizures. These findings are consistent with a previous study that showed that tail vein infusion of unesterified n-3 PUFA protects against focal and generalized seizures induced by electrical stimulation in the cortex (Voskuyl et al., 1998).

Brain concentrations of the unesterified fatty acid fraction in this study exceeded values reported in the literature by at least 34-fold (Deutsch et al., 1997). These higher unesterified fatty acid concentrations probably reflect ischemia-induced release of free fatty acids from phospholipid membranes due to decapitation (Rapoport, 1995; Deutsch et al., 1997; Bazinet et al., 2005). Deutsch et  al. (1997) reported that the contribution of ischemia-induced release of free fatty acids overestimates the  actual unesterfied fatty acid pool in brain by at least seven-fold. The magnitude of increase in the unesterified fatty acids reported by Deutsch et  al. (1997) is still lower than what we observed (34-fold increase), but this is likely due to the additional effects of PTZ on free fatty acid release from membrane phospholipids (Bazan, 1971). Although microwaving the brains prior to decapitation would provide a more accurate estimate of the amount of unesterified fatty acid concentrations in brain (Rapoport, 1995; Deutsch et al., 1997; Bazinet et al., 2005), the contribution of ischemia and PTZ-induced release of free fatty acids does not explain the differing n-3 PUFA profiles between the four groups ( Table  2).

The differing fatty acid profiles observed in the unesterified fatty acid pool most likely reflect differences in the release of free fatty acids from phospholipid membranes due to PTZ administration. PTZ, in addition to being excitatory, has been shown to increase neuroinflammation by increasing the production of proinflammatory prostaglandins (Seregi et al., 1990). n-3 PUFAs, such as eicosapentaenoic (20:5n-3) and docosahexaenoic acids (22:6n-3), have been reported to protect against neuroinflammation through their autacoid metabolites, which are eicosanoids and docosanoids, respectively (Hong et al., 2003; Marcheselli et al., 2003; Lukiw et al., 2005). The increased composition of n-3 PUFA in the SR-3 200  mg/kg group is most likely indicative of increased utilization of n-3 PUFA for eicosanoid and docosanoid production, to counteract the proinflammatory effects of PTZ. There is some evidence indicating that preventing or reducing neuroinflammation in brain by using antiinflammatory agents such as aspirin, can protect against PTZ-induced seizures (Tandon et al., 2003; Tu & Bazan, 2003; Dhir et al., 2006; Oliveira et al., 2008). Thus, the release of anti-inflammatory lipid mediators derived from n-3 PUFA is a possible explanation for the observed anticonvulsant effects of the SR-3 mixture.

We cannot exclude the possibility that ketone bodies, such as acetone, acetoacetate, and β-hydroxybutyrate, which also have anticonvulsant properties (Likhodii & Burnham, 2002; Rho et al., 2002; Likhodii et al., 2003; Ma et al., 2007), may have contributed to the observed anticonvulsant effects of the SR-3 compound. This is because the lower body weight gain in the SR-3 200  mg/kg group ( Fig.  1), which occurred despite similar food intake from days 12 to 21 relative to the other three groups ( Fig.  2), is suggestive of enhanced β-oxidation and possibly ketosis (Cunnane, 2004). Polyunsaturated fatty acids such as α-linolenic acid have been reported to decrease weight gain in mice (Cunnane et al., 1986) by activating the transcription of genes involved in β-oxidation (Ide et al., 1996). The increase in β-oxidation may have resulted in an elevation of ketone bodies. Levels of ketone bodies, such as acetone, were not measured in this study because all subjects received PTZ, which has previously been shown to increase basal levels of acetone in plasma (Nylen, 2005). It is worth noting, however, that in contrast to humans, rats are incapable of chronically sustaining or achieving clinical levels of ketosis (>2 mM) after being placed on a high-fat ketogenic diet, which is reported in several studies to chronically raise plasma ketone bodies in humans, but only transiently in rats (Likhodii et al., 2000; Taha et al., 2005; Musa-Veloso et al., 2006). Thus, the anticonvulsant effects of the SR-3 are unlikely due to elevated ketone bodies.

It is not surprising that the observed increase in total unesterified fatty acid concentrations was not reflected by a decrease in the concentration of total phospholipids. This is because, on a relative basis, phospholipids make up the majority of total brain lipids. On a quantitative basis, they exceeded the concentration of unesterified fatty acids by approximately 8- to 20-fold (Tables 1 and 2). Thus, it would be difficult to account for the release of free fatty acids from phospholipids by measuring changes in phospholipid concentrations. Alternatively, the use of radiolabeled fatty acid tracers in future studies may provide more insight into the release of free fatty acids from membrane phospholipids (Rapoport, 2003).

Toxicity was found at the doses of 400 and 1000  mg/kg. The symptoms consisted of low weight gain, low food intake, and bloating. These symptoms appeared to be caused by impaired peristalsis, which may have been caused by a direct effect of the injected fatty acids on the gastrointestinal tract following their diffusion through the peritoneal cavity. Future experiments will involve injection of the SR-3 compound via the subcutaneous (s.c.) route to avoid localized exposure of the gastrointestinal tract to high doses of the SR-3. It is possible that higher SR-3 doses will be tolerated when administered via s.c. or oral routes and that these may have larger effects on seizure severity and occurrence as well as on seizure latency. Administering the SR-3 through the oral route is of considerable importance, as this will have practical implications for future clinical studies.

While the present findings need to be confirmed by further studies, they do provide support for the idea that n-3 PUFA might provide a treatment for patients with epilepsy (Schlanger et al., 2002; Yuen et al., 2005; Bromfield et al., 2008). A diet rich in n-3 PUFA might supplement the anticonvulsant effects of antiepileptic drugs or the ketogenic diet (Fuehrelein et al., 2005; Dahlin et al., 2007; Taha & Burnham, 2007) or even possibly offer an alternative to anticonvulsant drug therapy.

Acknowledgments

We would like to thank Mr. Jerome Cheng for assisting in scoring the seizures. Funding for this study was provided by the Bahen Chair in Epilepsy Research grant to W.M.B., the Natural Sciences and Engineering Research Council grant to D.W.L.M., and the Canadian Institutes of Health Research doctoral research award (Fredrick Banting and Charles Best Canada Graduate Scholarships) to A.Y.T.

Conflict of interest: The work described within is consistent with the journal's guidelines for ethical publication. The authors declare that there are no competing personal or financial interests.

Ancillary