I doubt it is ferritin sucking up metals; although the problem of
crystallizing ferritin is an issue in secreted preps. See this paper,
where they made lemonade (a PNAS paper) from this lemon (High Five
ferritin crystals): " Secreted ferritin was isolated from the
supernatants of baculovirus-infected T. ni cells as a contaminant during
purification of an unrelated recombinant 6xHis-tagged protein on a
Ni-agarose column..."
http://www.ncbi.nlm.nih.gov/pubmed/15896348
The reason why I don't think it is ferritin is if you buffer exchange
your entire secreted prep, you still end up with ferritin contaminating
your prep, but your resin is not ruined. Overall, methods that don't
remove any protein, but only small-molecule components perform perfectly
well preserving resin in secreted insect/mammalian preps. So, your
second guess is probably correct.
By the way, I have once quizzed the makers of one proprietary medium,
and their response was "Yes, IMAC won't work, and no, we cannot tell you
what it is that makes it not work, it is proprietary."
Best,
Engin
On 5/19/14, 9:38 AM, Keller, Jacob wrote:
Well, this is of only possible relevance, but in a previous lab, we
used sf9 cells/media quite a bit, and there was always an issue
similar to this, due to [we thought] ferritin being secreted into the
medium, and sucking up the metals. Many, in fact, crystallized
ferritin this way by mistake! Is the concentrated flow-through
colored, indicating protein-bound metals, or do the metals go through
a concentrator, indicating free or small-molecule-complexed metals?
What about pH?
What about Gibco's formulation---is it available? The mysterious
"dispersant" could easily be either some chelator like EDTA. You could
try ITC with Gibco in the cuvette, titrate with metals if you have a
machine handy---it's just an hour or so.
JPK
*From:*CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] *On Behalf
Of *Bernhard Rupp
*Sent:* Monday, May 19, 2014 10:14 AM
*To:* CCP4BB@JISCMAIL.AC.UK
*Subject:* [ccp4bb] HisTrap Trap
Hi Fellows,
my lab mates successfully expressed a glycoprotein in CHO cells in
serum free medium, and
the protein captures nicely on HisTrap Excel 1ml columns (obviously,
high yield is not my problem...).
We load ca 1L supernatant at 0.5 ml/min, and eluate with a steep
imidazole gradient. 20mM Imidazole buffer for regeneration.
Works fine (and often...see yield remark).
Overcome by common crystallographers' greed (nor creed), we switched
to stable xfected HEK293, and cell free medium Gibco CD 293.
The first run gave high final yields & cheers.
The second run less of either, because the small HisTrap column
essentially dissolved -- the medium collapsed,
Ni leaches out, kaput as kaput goes.
A 3rd run on a similar previously working column lead to the same result.
Only thing changed was the cells and medium. Same buffers, same
gradients, same Akta equipment, same lab techs.
Before I improve the statistics by ruining further columns, has
anybody experienced such a calamity that might
be blamable on secret media components or similar? There is a
mysterious 'proprietary dispersant' preventing
cell adhesion quoted....
Best wishes, BR
----------------------------------------------------------------------------------------
Bernhard Rupp
b...@ruppweb.org <mailto:b...@ruppweb.org>
b...@hofkristallamt.org <mailto:b...@hofkristallamt.org>
http://www.ruppweb.org/
-----------------------------------------------------------------------
