Expt: I have isolated genomic DNA from LNCaP cells and would like to crosslink a small molecule (Chem-seq) to the DNA. The small molecule is an alkylator so it will crosslink DNA on its own. The buffer I am using for the crosslinking is TKMC (10 mM Tris-HCl [pH 7], 10 mM KCl, 10 mM MgCl2, 5 mM CaCl2). After crosslinking I want to fragment the genome to 100 - 600 bp (300-400 bp average). I am using a Bioruptor 300.

Question: Should the TKMC buffer be sufficient for sonication? Most references I have found use TE buffer (pH 7.5 - 8), and I can't find a reference using TKMC.

More info: I tried a direct comparison between the two buffers (TKMC v TE pH8) and the TE buffer seems to fragment to smaller sizes compared to the TKMC buffer.

For example: Biorupter 300, Low setting, 1 cycle = 30/30 ON/OFF @ 3 C, 5 ng/uL DNA in 300 uL

average frag size @ 30 cycles: TE = 350bp, TKMC = 500bp

Any suggestions? Using TE buffer for my alkylation crosslinking would not work due to presence of EDTA. Should I play around with the cycle ON/OFF times?

Any help would be awesome! Thanks.


After running a comparison between TKMC, TKMC + 1mM EDTA, TKMC + 10 mM EDTA there was a slight tightening up of the fragment range as I added EDTA. In addition, the fragment range (sizes) was smaller when EDTA was added (compared to TKMC w/o EDTA, after the same number of cycles). The results were not drastic but there was indeed a noticeable effect on fragment size and range. I will proceed with adding EDTA before sonication. Thank you for your suggestions!

  • $\begingroup$ As long as kept cold I would increase the on/off until you achieved the average size you are looking for. I'd start the next one with 40 cycles and examine. This manual page 17 seems to indicate this would be very appropriate diagenode.com/files/products/shearing_technology/bioruptor/… $\endgroup$
    – akaDrHouse
    May 18, 2017 at 19:12
  • 1
    $\begingroup$ I know nothing about this, but if divalent cations somehow affect fragmentation, you could try adding EDTA in excess after cross-linking. $\endgroup$
    – canadianer
    May 18, 2017 at 19:17
  • $\begingroup$ @canadianer Yes, of course, better than my suggestion of buffer exchange! $\endgroup$
    – Alan Boyd
    May 19, 2017 at 10:08
  • $\begingroup$ @AlanBoyd I'm kind of curious if it works. It's surprising to me that Mg would have such a drastic effect, but I'll have to read the paper you found. $\endgroup$
    – canadianer
    May 19, 2017 at 15:31
  • $\begingroup$ Note that there is 10 mM Mg and 5 mM Ca, so you may need to add in excess of 15 mM EDTA. $\endgroup$
    – canadianer
    May 25, 2017 at 16:22

1 Answer 1


Could you use a spun column to exchange the buffer? Some sort of gel filtration column; equilibrate column in TE; add your DNA sample, spin through excluded DNA. TKMC components will be included in the column and thus retained.

Update Have a look at this paper, and references therein. Apparently magnesium ions influence the flexibility of DNA. Clearly if the shape of your DNA in the two buffers is different this would impact upon the effects of shearing forces.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .