Protein Function Refuted

A mouse knockout calls into question the presumed function of a protein long considered important for steroid hormone biosynthesis.

Feb 1, 2014
Abby Olena

TYPICAL TESTES: Control mouse testis (left) and TSPO-knockout testis (right) develop similarly, refuting long-held assumptions about TSPO’s function.KANAKO MOROHAKU, CORNELL UNIVERSITY


The paper
K. Morohaku et al., “Translocator protein/peripheral benzodiazepine receptor is not required for steroid hormone biosynthesis,” Endocrinology, 155:15-20, 2013.

The context
The first step in steroid hormone production is the transportation of cholesterol into the mitochondria. For years, scientists have considered two candidates vital to this process: steroidogenic acute regulatory protein (StAR) and channel-like translocator protein (TSPO). Experiments in knockout mice showed that StAR, which binds cholesterol, is essential for hormone production. But the bulk of evidence regarding TSPO, found in the outer mitochondrial membrane, has been generated using cell lines because global TSPO knockout mice die as embryos.

The surprise
A team led by Vimal Selvaraj of Cornell University deleted TSPO only in the testicular Leydig cells of mice. Much to his surprise, the mice displayed normal testicular development, testosterone production, and StAR expression. “It completely refuted the existing theory or dogma that was established for TSPO,” says Selvaraj.

The future
Selvaraj is convinced that TSPO is not essential for testosterone production and he hopes his results will encourage researchers to go in new directions toward understanding the mechanism of cholesterol transport.

The debate
Walter Miller of the University of California, San Francisco, who was not involved in the work, cautions against tossing out the earlier in vitro evidence. “The role of TSPO in steroidogenesis remains controversial,” Miller says, adding that TSPO is just the latest on a list of proteins with biochemical evidence for a role in steroid synthesis, but for which mouse knockouts have no abnormal phenotype. “Maybe mice are just lousy models for what is happening in human steroidogenesis.”