The 3C (chromosome conformation capture) technology for studying chromatin 3D organization starts by a cross-linking step using formaldehyde to find segments of DNA interacting. In my understanding the cross-linking happens only between the DNA and proteins forming a DNA-protein-DNA complex. The "only" is from a DNA stand-point as of course protein-protein cross-links will also occur.

Can formaldehyde cross-link two double stranded DNA molecules? Is the 3C approach also capturing direct DNA-DNA interactions?

I looked at the literature, especially Dekker et al. description of the method, but couldn't find this information. Thanks.

  • 1
    $\begingroup$ Yes, it can: pubs.acs.org/doi/abs/10.1021/ja00074a005 $\endgroup$
    – canadianer
    Commented Mar 21, 2015 at 4:34
  • 1
    $\begingroup$ @canadianer Thanks. Had a look at it and some papers they reference but apparently this is only true for single stranded DNA which is not the kind of DNA-DNA cross-link that might happen during 3C. $\endgroup$ Commented Mar 21, 2015 at 4:52
  • $\begingroup$ I've only skimmed the paper, but it constantly refers to interstrand cross links and cross linking of DNA duplexes.… $\endgroup$
    – canadianer
    Commented Mar 21, 2015 at 4:59
  • 1
    $\begingroup$ @canadianer Interstrand cross-linking refers to a cross-link between opposite DNA strands. Not really what I am looking for as in the case of the 3C it is two distinct dsDNA interacting. $\endgroup$ Commented Mar 21, 2015 at 5:01
  • $\begingroup$ years ago my lab mixed plasmid DNA and short polylysine peptides with terminal cystines so the peptides would polymerize on the DNA. Then they mixed in glutaraldehyde to crosslink the peptides. The resulting complex was stable in mouse liver weeks, but of course the DNA itself was useless. $\endgroup$
    – user137
    Commented Mar 21, 2015 at 7:06

1 Answer 1


The fact that there is no inter-strand cross-linking between different double strands might be just because the cross-linker cannot bridge the distance between amines of different bases on different double strands.

The formaldehyde based linking of nucleobases has been described by Chaw et al. (1980) where formaldehyde bridged a gap of app. 3 angstroms (2x 1.3 A for N-C bond).

Even if you take the most condensed form of DNA as in nucleosomes (see X-ray model of nucleosome with bound DNA) you still have at least 20 angstroms between the bases of different double strands. At the same time there are only 3 angstroms between the nucleobases of the same double strand, agreeing well with the length of a methyl-bridge.


You must log in to answer this question.

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