Gabapentin's minimal action on markers of rat brain arachidonic acid metabolism agrees with its inefficacy against bipolar disorder

Abstract

In rats, FDA-approved mood stabilizers used for treating bipolar disorder (BD) selectively downregulate brain markers of the arachidonic acid (AA) cascade, which are upregulated in postmortem BD brain. Phase III clinical trials show that the anticonvulsant gabapentin (GBP) is ineffective in treating BD. We hypothesized that GBP would not alter the rat brain AA cascade. Chronic GBP (10 mg/kg body weight, injected i.p. for 30 days) compared to saline vehicle did not significantly alter brain expression or activity of AA-selective cytosolic phospholipase A(2) (cPLA(2)) IVA or secretory (s)PLA(2) IIA, activity of cyclooxygenase-2, or prostaglandin E(2) or thromboxane B(2) concentrations. Plasma esterified and unesterified AA concentration was unaffected. These results, taken with evidence of an upregulated AA cascade in the BD brain and that approved mood stabilizers downregulate the rat brain AA cascade, support the hypothesis that effective anti-BD drugs act by targeting the brain AA cascade whereas ineffective drugs (such as GBP) do not target this pathway, and suggest that the rat model might be used for screening new anti-BD drugs.

Published by Elsevier Ltd.

Figures

Fig. 1

PLA2 mRNA, protein and activity levels in brains of gabapentin treated and control rats. Mean mRNA for cPLA2 (p = 0.02) (a); sPLA2 (p = 0.98) (d); and iPLA2 (p = 0.33) (g) in control (open bars) and gabapentin (filled bars) treated rat brain (n = 6 in control and gabapentin groups), measured using RT-PCR. mRNA levels in brain are normalized to the endogenous control (β2 microglobulin) and relative to control level (calibrator), using the ΔΔCT method. Mean brain protein levels with representative immunoblots for cPLA2 IVA (n = 8 per group; p = 0.88) (b); sPLA2 IIA (n = 8 per group; p = 0.60) (e); and iPLA2 VIA (h) (n = 7 per group, gabapentin due to one outlier in each group that was removed; p = 0.15) from control and gabapentin treated rats. Data are ratios of optical densities of cPLA2, sPLA2, or iPLA2 expressed as percentage of control. Mean activity of cPLA2 (n = 8, control and n = 7, gabapentin, one sample omitted due to activity level below blank; p = 0.38) (c); sPLA2 (n = 8 per group; p = 0.49) (f); and iPLA2 (n = 8 per group; p = 0.14) (i) in control and gabapentin treated rat brain. Values are mean ± SEM, *p< 0.05

Fig. 2

Mean COX mRNA, protein and activity in brain of gabapentin treated and control rats. Mean mRNA levels for COX-1 (p = 0.90) (a) and COX-2 (p = 0.46) (d) in control (open bars) and gabapentin (filled bars) treated rat brain (n = 6 per group), measured using RT-PCR. Data are mRNA level normalized to the endogenous control (β2 microglobulin) and relative to control level (calibrator), using the ΔΔCT method. Mean protein levels with representative immunoblots for COX-1(p = 0.83) (b) and COX-2 (p = 0.049) (e) in control and gabapentin treated rat brain (n = 8 per group). Data are ratios of optical densities of COX-1 or COX-2 expressed as percentage of control. Mean activity of COX-1 (n=6, control (one sample omitted due to activity below level of blank and one outlier removed) and n=8, gabapentin; p = 0.41) (C) and COX-2 (n=7, control (one sample omitted due to activity below level of blank) and n=8, gabapentin; p = 0.91) (f) in control and gabapentin treated rat brain. Values are mean ± SEM, *p< 0.05

Fig. 3

PGE2 and TXB2 concentrations in brain of gabapentin treated and control rats. Mean concentrations (ng/ml) of PGE2 (n = 8, control and n = 9 gabapentin (one control sample omitted because its concentration was 6-fold greater than the mean, suggesting inadequate microwave fixation; p = 0.85)(a) and TXB2 (b) in control (open bars) and gabapentin (filled bars) treated rat brain (n = 9 per group; p = 0.85), determined by an ELISA assay. Values are mean ± SEM.