page 257 that surprisingly weak protein interactions can induce their formation. The results undermine a previous model of OSER formation, suggest a general mechanism that could drive membrane stacking in organelle biogenesis, and raise a warning flag for users of green fluorescent protein (GFP) tags.
Previous work suggested that OSER biogenesis entailed the tight, zipper-like dimerization of the cytoplasmic domains of certain ER-resident proteins. However, the authors found that naturally occurring OSER-inducing proteins can diffuse freely between OSER and ordinary reticular ER, indicating that they are not tightly bound in zipper structures. Weakly dimerizing GFP, and chimeric ER proteins with GFP on their cytoplasmic tails, could also induce OSER formation, but similar proteins with nondimerizing GFP tags could not.
When any of the weakly dimerizing proteins are expressed above a threshold level in a cell, several different types of OSER structures can form, suggesting that a single mechanism produces the spiraling, stacked, and crystalloid ER shapes seen in earlier studies. Mutations that cause the genetic diseases Charcot-Marie-Tooth syndrome and early onset torsion dystonia have been linked to mutant protein accumulation in the ER leading to OSER formation. The function of OSER in healthy cells remains unknown, but Snapp et al. suggest that the structures may help sequester lipid-soluble toxins.
Mass action by weakly binding proteins might be a general way to organize stacked organelle structures, including chloroplast thylakoids and the Golgi apparatus. The striking morphological changes induced by dimerizing GFP domains also suggest that the popular fluorescent tag could have unintended consequences. Although OSER formation does not appear overtly harmful to cultured cells, nondimerizing forms of the tag are probably a more prudent choice for most studies. ▪