Causes for sequencing overlap traces in one direction but not another - (Nov/04/2013 )

I am doing some standard ligation cloning here and running into a systematic issue.

I am attempting to ligate a 1 kb insert (obtained by overlap extension PCR) into NcoI and KpnI sites of a particular plasmid. I get plenty of colonies on my transformations, and very few colonies when the insert is left out of the ligation. 

I sequenced 4 colonies so far, using two different forward primers and one reverse primer for each colony. All sequencing primers lay down on the plasmid backbone just before (and after in the case of the reverse primer) where I would expect the insert to be, and should read into the insert.

The reverse sequencing comes back exactly as expected. The insert is there and all the mutations I have made that can be seen from the reverse direction are present. 

However, every single forward sequencing reaction comes back with overlap traces, irrespective of the sequencing primer used. The overlaps in the sequencing traces start about 10 bases upstream of the NcoI site, i.e. just before the insert.

The only thing that I can think of is that somehow the plasmid is ligating to another plasmid molecule through their NcoI sites? This would create two sites for the forward sequencing (but also two sites for reverse sequencing...which doesn't make sense).

Cloning is not a new thing for me but this particular issue is, and any suggestions as to the source of the sequencing overlaps would be most appreciated. 

I am redoing the ligation using the same plasmid backbone from a different source in case the plasmid itself that I used just happened to be bad.

have you tried pcr to determine the size of the insert?

that should tell you if you have ligated more than your intended insert.

you may be seeing slippage in the sequencing with your forward primers. do you have any repeats in your insert near the forward primers?

mdfenko on Tue Nov 5 12:17:25 2013 said:

have you tried pcr to determine the size of the insert?

 

that should tell you if you have ligated more than your intended insert.

 

you may be seeing slippage in the sequencing with your forward primers. do you have any repeats in your insert near the forward primers?

I appreciate your reply.

I did PCR the 4 clones using flanking primers that sit just outside of the insert, and all clones had  the insert of the expected size. So thankfully there weren't any unintentional ligation products.

I think you may be right about the forward primers being the problem. There is a 5bp sequence that repeats three times not far from the start of the insert. One of my two forward primers sits down on one of the repeats, so it's possible there is some kind of mis-priming.

I have sent more samples of those same four clones to sequencing using primers that sit down in the insert reading in opposite directions. This should give me some good clues as to whether primer or plasmid is the issue with the overlapping reads.

you can also try pcr using the sequencing primers. if you get multiple bands then you'll know that you are seeing mispriming or slippage (based on the size difference).

I had tried using one of the sequencing primers for PCR and the PCR came out clean, even though the sequencing had overlaps.

I got the new sequencing results back where I used primers that read from the middle of the insert toward the 5' and 3' ends. The results came back clean, showing the correct insert with all the right mutations ligated into the proper place in the plasmid. So I have had all along exactly what I wanted. I think you were right that there was some kind of slippage or mis-priming in my original forward sequencing primers. Not sure why this happened, but I am happy to have the correct construct.

It is kind of fun to know it worked because I made 7 mutations in a single round of overlap extension PCR. To do this, I had to amplify my insert into 5 pieces and then stitch them all back together in a single PCR. I was surprised I could using stitch 5 fragments together at once...usually I only do two fragments. Nice to see this method is robust.