Freezing GSH-agarose for "pull-down fever (PDF)"

Fred dynamin at
Sat Mar 4 23:45:44 EST 2000

Anton Tutter wrote:
> i wouldn't do that if i were you!!!  i think peter misunderstood the
> questions -- the beads are to be stored long-term *after* the GST-fusion
> proteins are bound to the beads.  20% ethanol may denature the proteins.

I completely agree. No ethanol for my beads (I, however, am a different

> what i would do is rapid-freeze the bead/protein complexes in liquid
> nitrogen, making sure there was 20% glycerol in the final buffer before
> storing. then store at -70 or colder.  the proteins should last a good long
> time that way without losing any activity.  

Why -70?  We often find that 20% glycerol frezzes at that temp, vs -20? 
And why rapid freeze?  Does it offer some advantage? 

> i regularly do pulldowns with
> GST-fusions but i always bind my proteins fresh -- that is, i keep purified,
> dialyzed stocks of my proteins at -80 prior to applying them to the beads,
> and thaw them when it is time to bind them to the beads.  one hour of
> binding shoud be more than adequate before you add whatever extracts you are
> trying to pull down from.

I have thought of that too.  I am trying to eliminate the time involved
in eluting the fusion proteins and dialysing them, and then rebinding
them - roll it all into one - just freeze the initially washed beads. 
Also, some of our constructs don't survive dialysis!!
> what's the nature of your GST "bait" proteins?  i routinely use
> transcription factor activation domains as bait and they generally are
> resistant to at least two or three freeze/thaw cycles, provided there is 10%
> glycerol or more (also slight detergent sometimes helps maintain solubility,
> like 0.1% NP-40), and i freeze/thaw them as rapidly as possible (nitrogen
> freeze, room temp water bath thaw).

We use lost of GST bait constructs.
- Those that don't survive dialysis include a couple of protein kinase
catalytic subunits (although PKA is one example that is fine).  In this
case, we need to freeze the bacterial pellet and bind it to beads
whenever needed - works ok, just variable binding.
- Those that don't survive sodium azide include a couple of small G
proteins (ras-like).  In this case, we want to incubate the proteins
attached to the beads with GDP or GTP to "load" them.  The time involved
in characterising the beads after loading makes it *much* more
economical in time to make a big batch and keep the frozen.
- The rest are SH3 domains and a splicing factor we have.  These all
survive the above and work fine the way you suggest.  However, even
these have problems.  We, for example, need to use a battery of 4-10 SH3
domains to look at a specific protein binding.  I  takes a lot of time
to titrate the beads on mini-gels to equalise the protein content of
each lot.  It would be easier to do this only for each new bacterial
expression (ie once a year?).

Hence my original question.  Would it be feasible to simply bind our GST
fusion protein to the GSH beads, wash them, run a minigel to check
loading level, then freeze the lot in aliquots.  I am trying to find the
fastest, simplest approach that will be generally applicable to all our

Freezing in glycerol seems to be the best bet so far.  I will post the
results when we have had a go at it.

Phil R,
CMRI, Sydney

More information about the Proteins mailing list