Regeneration of Pair End MiSeq flowcell

[Saurabh_Raj]

Hi,
I am performing some high-throughput DNA-protein interactions (HiTS-FLIP) using a custom made TIRF microscope, where I image the sequenced MiSeq flowcell. My trouble is that I need to regenerate the dsDNA on the flow cell, which doesn't seem to happen. Here is a short protocol of what I am doing:
* Denature the flowcell with NaOH and wash with TE buffer
* Inject Read 2 sequencing primer. I have tried two primers for regeneration 1) 34bp long GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT that covers the whole Read2 region, and 2) GTGACTGGAGTTCAGACGTGT - 21bp long, that is outside the 13bp overlapping region of Read1 and Read2.
* The primers are then subjected to an annealing cycle where flowcell temperature is raised to 85C and then slowly brought to 40 C. At this point, all the excess primers are washed away.
* Then the primer is extended using Bst2.0 warmstart polymerase at 60C. I have also tested Klenow at 37C.
* After running the regeneration, I inject Read1 oligos labeled with Cy3 to check whether regeneration was successful or not.
The problem is, I do not see any difference in the intensity of Read1Cy3 oligo binding before and after dsDNA regeneration. This means that my regeneration is unsuccessful. I would appreciate if you have some inputs on the methodology.
Thanks!
Best regards,
Saurabh

[cmbetts]

Could you explain a little more about you're intending to accomplish? As I understand it, you're performing paired end sequencing on a MiSeq. You then take the spent flowcell, and perform the described molecular biology steps with the desired result being dsDNA anchored to the flowcell in the same orientation as at the end of read2. Is that correct?

If so, I don't think that the Read1Cy3 oligo is a valid readout. The structure of your desired product has the sequence identical to the read 1 primer and it's reverse compliment annealed to it from the regenerated strand. I wouldn't expect a Read 1 oligo to be informative here because you'd have to displace your new strand to hybridize to it. Have you considered labeling the Read2 extension primer? Or maybe use a small portion of labeled nucleotide in your extension mix? This may not be compatible with your final downstream use, but could be used until you are confident in your protocol where it can be discarded. This is all assuming that you have access to a relatively plentiful supply of discard flowcells from other users at your institution that were going to throw them away anyhow.

[Saurabh_Raj]

Hi,
Thank you for your message. We have access to discarded flowcells to test our protocol. Let me explain a bit further. The discarded PE sequence that we obtain from Illumina have sequence in the order:
3'-P7-Index-Read2-sequenced DNA-Read1-P5-5'
Once the flow cell is denatured to remove any residual fluorophore from sequencing, we should get the above mentioned ssDNA. I use Read2 complimentary as the regeneration primer. When the regeneration is done, complimentary of Read1 - Read1*Cy3 is injected to see whether regeneration was successful or not. When regeneration happens, Read1 region will be annealed with regenerated strand, as you mentioned. In that case, Read1*Cy3 should not bind to the regenerated strands, isn't it? But I still see binding as strong as that for ssDNA.

I have used the labeled Read2 extension primer. But it is hard to say whether extension happened or not as the primer will remain hybridized to the ssDNA even when extension fails. I am thinking to implement the usage of labeled dNTPs next.

Best regards,
Saurabh