A protein isoform, or "protein variant" is a member of a set of highly similar proteins that perform the same or similar biological roles. A set of protein isoforms may be formed from alternative splicings or other post-translational modifications of a single gene. Through RNA splicing mechanisms, mRNA has the ability to select different protein-coding segments (exons) of a gene, or even different parts of exons from RNA to form different mRNA sequences. Each unique sequence produces a specific form of a protein. A set of protein isoforms may also result from a set of closely related genes that evolved from a single gene in the past.
The discovery of isoforms could explain the discrepancy between the small number of protein coding regions genes revealed by the human genome project and the large diversity of proteins seen in an organism: different proteins encoded by the same gene could increase the diversity of the proteome. Isoforms at the DNA level are readily characterized by cDNA transcript studies. Many human genes possess confirmed alternative splicing isoforms. It has been estimated that ~100,000 ESTs can be identified in humans. Isoforms at the protein level can manifest in deletion of whole domains or shorter loops, usually located on the surface of the protein.
One single gene has the ability to produce multiple proteins. All these proteins are different both in structure and composition and this process is regulated by alternative splicing of mRNA and have a large impact in proteome diversity. The specificity of produced proteins is derived by protein structure/function, development stage and even the cell type. It becomes more complicated when a protein has multiple subunits and each subunit has multiple isoforms.
For example, the 5' AMP-activated protein kinase (AMPK), an enzyme, which performs different roles in human cells, has 3 subunits:
In human skeletal muscle, the preferred form is α2β2γ1. But in the human liver, the most abundant form is α1β2γ1.