We have tried a new technology to generate temperature-sensitive (ts)
mutants [see Dohmen, R. J., Wu, P. and Varshavsky, A. (1994) Heat-inducible
degron, a method for constructing temperature-sensitive mutants. Science, 263,
1273-1276]. Basically one simply constructs a fusion consisting of a portable
ts protein-degradation cassette fused to the 5' end of the gene of interest.
This construct is then integrated to the chromosomal site of the gene of
interest, creating a truncated version and the recombinant
temperature-sensitive "degradable" (td) version flanking the plasmid DNA on the
The gene of interest in our case is DBP5 (DEAD-Box protein; an essential
putative RNA helicase gene of unknown function; cellular localization unknown).
At the first glance, this technology works extremely well, in that the dbp5-td
strain failed to grow on either YPD or SD-Ura solid media at 37oC (analyzed by
streak-out or spotting method). Although after 3-5 days' incubation at this
temperature, there are a small number (1 to 6 in each streak out) of revertants
(?) popping out on the plate. Introduction of the ubr1-deletion allows the td
strain to regain growth at 37oC, demostrating the temperature-sensitivity is
due to N-degron protein degradation pathway.
But the good news ends here, because we could not reproduce the plate
results in the liquid culture. We have tried several times for the temperature
shift growth curve studies. In all cases, the td strain grows equally well as
the wild type strain (in selective medium) for the first 9-10 hours after being
shifted to the nonpermissive temperature. We have tried shifting at various
OD600, such as 1 (mid-log phase), 0.8, or 0.3 (early log phase), and obtained
practically the same results. The td strain showed approximately 50% reduction
of growth (1.0 versus 1.6 of wild-type strain) for a period of at least 7-8
hours subsequent to the first 10 hr of shifting to 37oC. However, the td
strain still eventually reach near saturation (with comparable OD as the wt) 30
hr after the temperature shift. This result is completely at odds with the
tight growth arrest phenotype on the plate.
We have measured the spontaneous reversion frequency of the td strains in
the liquid culture. The numbers we got were 5.4x10e-7, 7.2x10e-7, and 3x10e-6
from 3 independent experiments using two td isolates. Thus, we believe that
the problem is unlikely due to exceptionally high spontaneous reversion
frequency (derived from the the N-degron pathway).
We have inspected the cell morphology 10 to 30 hours after the shift in
the liquid culture and found no obvious signs of strange shapes (such as
super-enlarged shapes) which may explain the high OD after shift.
At this moment, we are still scratching our heads. Would any one out
there like to venture some explanations of this seemingly strange observation?
The central question is how do the cell growth (physiology) differ from each
other on plate and in the liquid medium? Could it be due to the difference of
osmolarity, nutrient diffusion rate, or something else?
Department of Molecular Genetics
The Ohio State University
484 West 12th Ave.
Columbus, OH 43210
chang.108 at osu.edu