QUANTITATIVE TRAIT LOCI (QTL) MAPPING OF VALPROIC ACID (VPA) RESPONSIVENESS IN MICE
Abstract number :
2.128
Submission category :
Year :
2003
Submission ID :
640
Source :
www.aesnet.org
Presentation date :
12/6/2003 12:00:00 AM
Published date :
Dec 1, 2003, 06:00 AM
Authors :
Thomas N. Ferraro, Gregory T. Golden, George G. Smith, James F. Martin, Wade H. Berrettini, Russell J. Buono Psychiatry, University of Pennsylvania, Philadelphia, PA; Research Service, Department of Veterans Affairs Medical Center, Coatesville, PA
Responsiveness to anticonvulsant drugs is a complex trait that manifests clinically as interindividual differences between patients with respect to both therapeutic and toxic effects. We are using inbred strains of mice to map QTL which mediate responsiveness to VPA. We plan to identify causative genes in this model and ultimately translate the findings to humans with epilepsy.
We tested the ability of VPA (100-300 mg/kg, i.p.) to elevate maximal electroshock seizure threshold (MEST) in 10 inbred mouse strains. MEST was determined using a single daily shock schedule involving incremental current increases. The current which elicited a maximal seizure (tonic hindlimb extension) was taken as the threshold. VPA was administered 30 min prior to each shock trial. Control mice were injected with saline. The most responsive (129X1/Sv) and least responsive (BALB/c and 129S1/SvIM) strains were intercrossed to yield F1 and then F2 generation progeny which were also tested for responsiveness to VPA (200 mg/kg). Contingency tables and multipoint linkage analysis will be used to examine relationships between genome-wide microsatellite marker data and responsiveness of F2 mice.
VPA increased MEST 2-fold in 129X1 mice and 1.3- to 1.5-fold in BALB/c and 129S1 mice. Phenotype data on F2 progeny from both a BALB/c x 129X1 intercross (n = 250) and a 129S1 x 129X1 intercross (n = 200) reveal a continuous range of values, the distribution of which approximate a normal curve. Tails of the distribution overlap with parental strain phenotypes. Heritability estimates indicate that about two-thirds of the phenotypic variance in both crosses is explained by genetic factors. In order to have over 90% power to detect genes of relatively small effect (i.e. explaining 10-20% of total phenotypic variance), genome scan experiments will begin when F2 populations reach n = 400.
We conclude that the ability of VPA to increase MEST in mice is a highly heritable complex trait. The magnitude of the differences that we documented between the relatively responsive 129X1 strain and the relatively non-responsive BALB/c and 129S1 strains suggest that QTL mapping methods may be used successfully to identify the linkage groups which contain genes that can influence the anticonvulsant effects of this widely prescribed drug.
[Supported by: NIH Grant NS39516]