alternative splicing
/ all-TER-nuh-tiv SPLY-sing /
Imagine a set of LEGO sections labelled 1, 2, 3, 4, 5. From the same box of pieces you could build 1-2-3-4-5, or skip part 3 to build 1-2-4-5, or use a different version of part 2. Same parts, different finished models. Alternative splicing is the cell doing this with a gene's exons: from one pre-mRNA it can make several different mRNAs by choosing which exons to keep and which to leave out.
Recall that splicing removes introns and joins exons. In alternative splicing, the spliceosome treats certain exons as optional or chooses among competing splice sites. An exon can be included or skipped; two exons can be mutually exclusive (use one or the other, never both); a splice site can be shifted to lengthen or shorten an exon; even some introns can be retained. Each choice yields a different mature mRNA, called an isoform, and usually a different protein. The choices are not random — they are steered by regulatory proteins that bind the pre-mRNA and tip the spliceosome one way or another, and they differ between tissues, developmental stages, and conditions.
This is the discovery that overturned the old slogan one gene, one protein. The human genome has only about 20,000 protein-coding genes — fewer than expected — yet the great majority of them are alternatively spliced, so the actual protein repertoire is far larger. One famous insect gene can in principle be spliced into tens of thousands of forms. Alternative splicing is a central reason a modest gene count can build an organism as complex as a human, and errors in it underlie many diseases.
The same calcitonin gene is spliced one way in thyroid cells to make the hormone calcitonin, and another way in nerve cells to make a different protein, CGRP — one gene, two tissues, two proteins, decided entirely by which exons are kept.
One gene, spliced differently in different cells, yields different proteins — overturning one gene, one protein.
Not every splice isoform a cell makes is functional — some are dead ends destined for nonsense-mediated decay, and a cell can use this deliberately, splicing a message into a doomed form to dial its protein down. So an abundance of isoforms does not by itself prove a matching abundance of useful proteins.