Guanidine isothiocyanate in PCR - (Jan/30/2014 )

I need to do PCR on human cell genomic DNA, very small amount (around 5000 cells) that were frozen in 500ul QIAGEN RLT buffer (not that I could do anything about that).

Now RLT is proprietary, but supposedly contains a high concentration of guanidine isothiocyanate and some other nasty things. 

My original idea was just lyse the cells by heating and just put into PCR, howevrer I got most of them in Trizol (isolation required) or RLT. I can isolate from Trizol thought that is a pain in the ass with DNA and the small number of cells is problematic (I've done that before, but repeatedly thought of better way to do that), but I can't even use any standard isolation protocol from RLT and some Q colums I have, because it's generaly designed for RNA isolation. In addition, elution volumes from columns are to high. I'm kind of affraid I wouldn't get any DNA, from whole volume used (which I idealy should not use whole).

So I'm thinking of some quick way how to get PCR-suffucient DNA out of the RLT solution, or at least minimize the guanidine content. Pelet the lysate on high g (if possible) and take the GITC out? Compensate the PCR reaction for GITC and use only a very small amount?

Any other ideas?

Just for the record.

I took 200 ul of the RLT, precipitated with the same volume of isopropanol for 10 minutes, centrifuged 15 min/16 000g/4 deg (probably could be less), washed pelet (completely invisible) with 200 ul of 100% EtOH and 70% EtOH (centrifuged inbetween as aforementioned), dissolved in 10 ul of TE, used 1 ul of this to PCR (with BSA <10 ug> only or BSA+0,8M betaine) and got pretty nice bands (betaine slightly better), definitelly enough for my purpose :)

Using 1 ul of RLT as is in the reaction, gave no product.