I would suggest doing some rigorous quality-control on your protein
stock, e.g., mass-spec'ing. Also, it may help to aliquot small
aliquots into pcr tubes and freeze at -80, and use only those stocks
for crystallizations. I had exactly this problem, and finally
concluded that the problem was heterogeneity of the protein stock.
JPK
On Tue, Nov 23, 2010 at 11:24 AM, Joe <[log in to unmask]> wrote:
> Hi all,
> I recently have problems reproducing some conditions identified from high
> throughput screenings.
> The initial screening (10 mg/ml protein, 0.5 ul well+ 0.6 ul protein, 23 C)
> gave rise to at least three hits from different screen kits.
> The follow-up grid optimization (1.5 ul well + 1.5 ul protein) varying
> precipitant concentrations and pHs did not produce any crystals for all
> three conditions. Instead, the drops have heavier precipitation
> background. The following experiments have been done in order to get
> crystals back.
> 1. Seeding, with various precipitant concentrations
> 2. Varying volume ratios between well solution and protein (from 2: 1 to 1:
> 2 v/v).
> 3. Using original solutions from screen kits.
> 4. Hanging drops and sitting drops,
> 5. Dispensing protein first or well solutions first.
> 6. Using robot to set up drops on 96-well plate to see if I can reproduce
> the original hits.
> But none of them has worked so far. This is the first time I encountered
> such a scale-up issue. I am running out of ideas, so hope you could give me
> some suggestions. Thank you in advance.
> --
> Best regards,
>
> Joe
|