Carbamazepine (CBZ), a drug commonly used for the treatment of epilepsy, can cause hypersensitivity reactions in 1 to 1000 to 1 to 10,000 new exposures. Lymphocyte transformation tests (LTT), in conjunction with medical history and clinical findings, have shown that CBZ hypersensitivity is immune-mediated [1]. Heterotrimeric G-proteins are important mediators of intracellular signal transduction. Recently, a C825T polymorphism was detected in exon 10 of the gene GNB3, which encodes the G-protein β3-subunit. The 825T allele results in the expression of a splice variant termed G β3-s, which despite a deletion of 41 amino acids is functionally active [2]. Expression of G β3-s leads to increased cellular immune responses both in vivo and in vitro [3–5]. Therefore, in this study we have investigated whether C825T polymorphism in the GNB3 gene acts as a predisposing factor for CBZ hypersensitivity. In addition, we performed LTTs in eight hypersensitive patients to determine whether the proliferative response to CBZ is affected by the GNB3 genotype.

DNA samples from 61 CBZ hypersensitive patients (22 with severe and 39 with mild hypersensitivity), 44 patients on CBZ therapy for at least 6 months with no adverse reactions to the drug, and 231 healthy individuals were genotyped for the C825T polymorphism in the GNB3 gene. Ethics committee approval was obtained. Genotyping was performed using an ABI PRISM 7000 Sequence detection System. Statistical analysis was done using the chi² and unpaired t-tests. Lymphocytes (1.5x10⁵) from hypersensitive patients were incubated with CBZ (10–100  µg ml^-1) in vitro for 6 h and proliferation measured by the incorporation of [³H] thymidine (0.5  µCi). Maximum stimulation indices (SI, cpm in drug treated cells/cpm in cells with vehicle alone) were then calculated.

We found no difference in the frequency of the T allele between our hypersensitive patients (0.32), CBZ tolerant patients (0.30) and healthy controls (0.32). However, there was a higher CBZ-specific lymphocyte proliferation in patients with the CT genotype (median SI 26.4, 95% CI 9.6, 45.1) than with the CC genotype (median SI 8.4, 95% CI 4.1, 13.2), although this just failed to reach statistical significance
( p = 0.12).

There was no association between the C825T polymorphism in the GNB3 gene and CBZ hypersensitivity. Interestingly the maximum lymphocyte proliferation obtained on CBZ exposure in vitro may be dependent on the GNB3 genotype, although our data are limited due to the small numbers studied. Nevertheless, further investigation into the role of G-protein subunits in the mechanism of allergic reactions is warranted.

1. Naisbitt DJ, et al. Mol Pharmacol 2003; 63: 732.
2. Siffert W, et al. Nat Genet 1998; 18: 45.
3. Virchow S, et al. FEBS Letters 1998; 436: 155.
4. Lindemann M, et al. FEBS Letters 2001; 495: 82.
5. Lindemann M, et al. Virology 2002; 297: 245.