Glycogen storage disease type Ib is caused by mutations in the glucose 6-phosphate transporter (G6PT) in the endoplasmic reticulum membrane in liver and kidney. Twenty-eight missense and two deletion mutations that cause the disease were previously shown to reduce or abolish the transporter's activity. However, the mechanisms by which these mutations impair transport remain unknown. On the basis of the recently determined crystal structure of its Escherichia coli homologue, the glycerol 3-phosphate transporter, we built a three-dimensional structural model of human G6PT by homology modeling. G6PT is proposed to consist of 12 transmembrane alpha-helices that are divided into N- and C-terminal domains, with the substrate-translocation pore located between the two domains and the substrate-binding site formed by R28 and K240 at the domain interface. The disease-causing mutations were found to occur at four types of positions: (I) in the substrate-translocation pore, (II) at the N-/C-terminal domain interface, (III) in the interior of the N- and C-terminal domains, and (IV) on the protein surface. Whereas class I mutations affect substrate binding directly, class II mutations, mostly involving changes in side chain size, charge, or both, hinder the conformational change required for substrate translocation. On the other hand, class III and class IV mutations, often introducing a charged residue into a helix bundle or at the protein-lipid interface, probably destabilize the protein. These results also suggest that G6PT operates by a similar antiport mechanism as its E. coli homologue, namely, the substrate binds at the N- and C-terminal domain interface and is then transported across the membrane via a rocker-switch type of movement of the two domains.