The human protein disulfide isomerase gene family

Abstract

Enzyme-mediated disulfide bond formation is a highly conserved process affecting over one-third of all eukaryotic proteins. The enzymes primarily responsible for facilitating thiol-disulfide exchange are members of an expanding family of proteins known as protein disulfide isomerases (PDIs). These proteins are part of a larger superfamily of proteins known as the thioredoxin protein family (TRX). As members of the PDI family of proteins, all proteins contain a TRX-like structural domain and are predominantly expressed in the endoplasmic reticulum. Subcellular localization and the presence of a TRX domain, however, comprise the short list of distinguishing features required for gene family classification. To date, the PDI gene family contains 21 members, varying in domain composition, molecular weight, tissue expression, and cellular processing. Given their vital role in protein-folding, loss of PDI activity has been associated with the pathogenesis of numerous disease states, most commonly related to the unfolded protein response (UPR). Over the past decade, UPR has become a very attractive therapeutic target for multiple pathologies including Alzheimer disease, Parkinson disease, alcoholic and non-alcoholic liver disease, and type-2 diabetes. Understanding the mechanisms of protein-folding, specifically thiol-disulfide exchange, may lead to development of a novel class of therapeutics that would help alleviate a wide range of diseases by targeting the UPR.

Figure 1

Schematic representation of the domain compositions of the 21 proteins in the PDI gene family. All proteins contain a short NH₂-terminal signal sequence designated in black. Catalytically active TRX-like domains (a or a’) are represented in red with active sites noted; inactive TRX domains in blue (b) and purple (b’); green represents the Asp/Glu rich Ca²⁺-binding domains; linker regions (gray); transmembrane domains (yellow); COOH-terminal ER-retention sequences in white with a.a. composition denoted. Although ERP29 does contain an a-like domain, this is based on homology and not catalytic efficiency. Sequences and domain compositions were acquired and verified from the National Center for Biotechnology Information (NCBI) database. Figure was adapted and modified from [4].

Figure 2

Clustering dendrogram of the human PDI gene family. Utilizing ClustalW alignment software of known protein sequences of the human PDI genes reveals divergence into AGR and CASQ subfamilies (red and blue, respectively).