A few weeks ago I posted some questions regarding protocols and reagents
to study subcellular localization of proteins in yeast. As promised, here
is a quick summary of some of the comments I received.
Johannes Hegemann's group has prepared yeast expression vectors which
allow for either N- or C-terminal GFP fusions with your protein of
interest. The fusion is driven from the MET25 promoter and therefore
allows for regulated expression. These have been successfully used to
localize proteins to the nucleus, mitochondria and vacuole (see Niedenthal
et al. (1996) YEAST 12:773-786).
The host strain seems to be important for good success in the
immunofluorescence work. The red pigment which accumulates in ade1 or ade2
mutants autofluoresces and gives high backgound so if possible, it is
probably a good idea to steer away from these strains. Also the use of
formamide (eg. for in situ hybridization) generates a yellow fluorescence
which also interferes with the GFP signal.
Approach the use of overexpression vectors carefully as reviewers can have
a problem with respect to mislocalization of the fusion protein,
especially at high levels. It may be smartest to try both strong promoters
(ie. ADH or GAL) as well as the native promoter of your gene. Of course,
the problem with this is generation of a signal below the threshold of
detection. If the knock-out of your gene yields a readily measurable
phenotype, reversion of the phenotype by overexpression of the fusion
protein could be used as an argument that at least some of it localizes
An excellent review on this topic is provided by Zinser and Daum (1995)
YEAST 11: 493-536.
Many thanks to those who responded to my original questions.
Department of Biochemistry
Montreal, Quebec, Canada