Lately I have been working on generating a 200 bp fragment that will contain a 69 bp randomized region. I will do this by using two large oligos with a 30bp overlapping region. As shown in the figure below, the right oligo has a illumina sequencing primer site, extra bases and the overlapping region. On the other hand, the left oligo has a illumina sequencing primer site, 3 fixed bases, 69 randomized base, and the overlapping region with the left oligo.
The illumina Nextera sequencing priming sites on the flanks (see figure below) will allow the fragment to be sequenced in both directions,if needed. Additionally this region will act as template for the primers, used by Nextera, to attach the barcodes and the adapters to the fragments to be sequenced (see figure below, barcodes are shown in grey and light blue, adapters in yellow and dark blue). Generating this 200 bp fragment is an essential part of my thesis, since I will use them as input DNA for uptake experiments using competent Acinetobacter baylyi and Thermus thermophilus. By comparing the sequences from the fragments that were taken up and from the input DNA fragments, I will try to determine if this bacteria have an uptake biases for a particular sequence motif.
In a first experiment, I annealed the two oligos using 1.1 ug of the right (38pmol) and left (26.5 pmol) oligos, as well as Tris and water in a 20 ul reaction. The oligos were then denatured at 95 degrees, then the temperature was decrease at 0.1 degree/second until it reached 65 degrees for 5 minutes.
Then, I used 2 ul of this annealed oligos reaction as part of a 6 PCR reactions using primers that will amplify the 200 bp fragment.
Besides the 2 ul of anneal oligos, 4 of the 6 reactions used: 200uM dNTPs, 0.2uM primers, 1X buffer, 2 units of Onetaq (NEB).
One reaction (called extension only) used: 2 ul of anneal oligos, 200uM dNTPs, 1X buffer, 2 units of Onetaq (NEB), but no PCR primers.
One reaction (No taq) used: 2 ul of anneal oligos, 200uM dNTPs, 1X buffer, 0.2uM primers, but no taq polymerase.
Amplification program includes an extension step:
65 degrees 10 min
Followed by a standard PCR protocol:
95 degrees 2 min
95 degrees 30 sec ]
55 degrees 30 sec ] 1, 5, 10, 20 cycles
68 degrees 30 sec ]
95 degrees 5 min
The “taq only” and “extension only”samples were taken out of the thermocycler after the extension step. The rest of the 4 reactions were amplified for 1, 5, 10, 20 cycles.
Results showed that:
- Comparing notaq and extension only samples we can see that the extension works, producing a 200 bp fragment
- PCR cycles seem to be amplifying an unspecific band at 250 bp.
In a next experiment, I was looking to determine why I am getting this 250 bp band. My, hypothesis was that not-fully synthesized “intermediate” sequences, which are common in long oligo synthesis, were responsible of the unspecific band. With that in mind, I tested if denaturing, re-annealing and re-extending oligos several times (without any PCR amplification, only extension) would remove, at least partially this “intermediate” oligos that could be responsible of the 250 bp band observed.
In this experiment, I used ~220 ug of the right (38pmol) and left (26.5 pmol) oligos, as well as Tris and water in a 20 ul reaction.
The oligos were then anneal and extended using three distinct protocols:
- Samples were denatured at 94 degrees, then the temperature was decrease at 0.1 degree/second (ramp) until it reached 58 degrees for 2 minutes and then temperature was increased by 10 minutes. Next, I used 5 cycles of: a 94 degrees denaturation (45 seconds), 58 degrees (1 minute) and 65 degrees (2 minutes).
- Samples were denatured at 94 degrees, then the temperature was decrease at 0.1 degree/second (ramp) until it reached 58 degrees for 2 minutes and then temperature was increased by 10 minutes
- Samples were denatured at 94 degrees and then anneal and extended at 65 degrees 10 minutes.
Results showed that:
- Denaturing and re-annealing/extending the oligos (1 protocol) did not help, and as a matter of fact it produced a 250bp unspecific fuzzy band that looks very similar to what I saw in the first experiment.
- This unspecific band seems to be produced by mis-priming of the long oligos and not mis-priming of PCR primers.
In this point I still believed that oligo mis-priming originated because of not-fully synthesized oligo intermediate, so I gel purified the anneal/extended product of the second reaction. Next, I did a PCR using a 1/100 dilution of the anneal extended gel purified and non-purified product with 4 different annealing temperatures each (62, 60, 58, 55.7 degrees).
Results showed a 250 bp band with a very faint 200bp band. It is surprising that even after gel purification I still get a 250bp band, that it is even more intense that the expected 200bp band.
At this point, it seems that there is any problem with the oligo sequences themselves, so I looked at the oligo design carefully in order to discover if there is any region of the oligos prone to miss-priming.
By looking at the sequences of the oligos, I realized that a 16 bases region of the overlaping region from the left primer (third row of the figure below) anneal perfectly (region in black rectangule) with extra bases from the extended forward strand (second row of the figure below)
This mis-priming would leave a 5 ‘ overhang that could easily get extended from the 3′ end of the forward strand and the 3’ end of the reverse strand, generating a 245bp sequence.
Additionally the sequence located in the extra bases of the left primer is partially complementary to each other (ccgcatcAGGTGGCACGAGgatgcgg).
This mis-priming is consistent with the results from the second experiment in which I tested three annealing/extension protocols. The first and second protocols had different results, despite that the first part of them is identical (94 degree denaturation followed by a rap decrease of 0.1 degree/sec to 58 degrees annealing). However, when I denaturate and re-anneal/extend the oligos in the first protocol, then I see the ~250bp band. Maybe the slow ramp gave enough time at higher temperatures (~65 degrees) to anneal the oligos correctly.
Eventually I will have to re-design and re-order at least one of both oligos, since even if I am successful amplifying the 200 bp fragment, I do not like any strange mis-priming effects when barcodes and adapters are introduced later on during the library prep step previous to sequencing the fragments.