I am trying to get my head around amplicon sequencing library prep but can't understand how the adapters are attached

Question

[ScottDansk]

In my lab we use a Fluidigm access array library prep strategy and then sequence on Illumina MiSeq

I have a list of the primers used and can see that we use CS1 and CS2 primers attached to target region primers

I will go through what I understand is happening and where I get confused using schematics:

1:Target region

5' --------------- 3'  sense strand
3' --------------- 5'  antisense strand

2:DNA is denatured and primers bind

(==== is either CS1 (sense) or CS2 (antisense) overhang & ~~~~ is target DNA primer)

sense strand:

           3' ~~~~==== 5'
5' --------------- 3'

antisense strand:

    3' --------------- 5'
5' ====~~~~ 3'

3:Synthesis of complement

sense strand:

3' <<<<<<<<<<<~~~~==== 5'
5' --------------- 3'

antisense strand:

    3' --------------- 5'
5' ====~~~~>>>>>>>>>>> 3'

4: Now you have ssDNA fragments with CS1 / CS2 primers

sense strand complement:

3' <<<<<<<<<<<~~~~==== 5'

antisense strand complement:

5' ====~~~~>>>>>>>>>>> 3'

This is where I am confused. How do you end up with libraries that have both the CS1 AND CS2 primers at either end? I understand this is what you end up with prior to sequencing primers and P7/P5 adaptors:

5' CS1 -------target DNA------- CS2 3'
3' CS1 -------target DNA------- CS2 5'

However looking at step 4, if reverse primers to CS1 and CS2 bind, synthesis cannot occur because it would be the incorrect direction:

++++ being CS1/CS2 reverse primers

sense strand complement:

3' <<<<<<<<<<<~~~~==== 5'
5'                ++++ 3'
#                 ^synthesis cannot occur here because it would be in the 3' > 5' direction

antisense strand complement:

5' ====~~~~>>>>>>>>>>> 3'
3' ++++                5'
#      ^synthesis cannot occur here because it would be in the 3' > 5' direction

SO, after step 3, are the original target DNA fragments washed away somehow (enrichment for fragments with CS1/CS2 primers:

3' <<<<<<<<<<<~~~~==== 5' CS1 fragment
5' --------------- 3' initial sense strand


    3' --------------- 5' antisense strand:
5' ====~~~~>>>>>>>>>>> 3' CS2 fragment

enrichment for CS1/CS2:

3' <<<<<<<<<<<~~~~==== 5' CS1 fragment

5' ====~~~~>>>>>>>>>>> 3' CS2 fragment

the target regions are complimentary so they can bind:

3'     <<<<<<<<<<<~~~~==== 5' CS1 fragment
5' ====~~~~>>>>>>>>>>>     3' CS2 fragment

so you can fill in the gaps with reverse CS1 / CS2 DNA:

3' ++++[<<<<<<<<<<<~~~~]==== 5' 
5' ====[~~~~>>>>>>>>>>>]++++ 3'

(everything in [] is the target DNA or amplicon)

Answer 1

[ATpoint]

This is where I am confused. How do you end up with libraries that have both the CS1 AND CS2 primers at either end? I understand this is what you end up with prior to sequencing primers and P7/P5 adaptors:

It's a multistep process.

In the very first PCR cycle you have these ssDNA fragments, with the primers being only present at one site each:

5'         -------target DNA------- CS2 3'    
3'     CS1 -------target DNA-------         5'

From there on the primers will not only target the genomic sites, but also created ssDNA fragments.

The left part of -------target DNA------- CS2 will be amplified by the forward, and the other one with only CS1 by the reverse primer, creating double-stranded DNA with primers on both sites. These products will enrich more and more with every cycle, creating the final PCR product.

Does this help?

[ScottDansk]

Hello, thanks so much for replying!

I am a bit confused because we have it the other way around:

your schematic puts CS1 and CS2 at the 3' end of the fragment is it contained in, but yours is 5' end:

Yours:

5'         -------target DNA------- CS2 3'

3'     CS1 -------target DNA-------         5'

Whereas my schematic:

sense strand complement:

3' -----target DNA------ CS1 5'

antisense strand complement:

5' CS2 ------target DNA------ 3'

It has to be this way because CS1 / CS2 primes synthesis, which happens 5' > 3' (step 2) and if the original CS1/CS2 primers are 3' they cannot prime the initial synthesis?

[GenoMax]

Figure 1 in this paper explains Fluidigm library prep: https://journals.asm.org/doi/10.1128/aem.03182-15

[ScottDansk]

hi thanks I have been looking at that! It still doesn't make sense to me though

In figure 1 you have original sample dna, then CS1/CS2 primers bind and prime the synthesis (but they have it from the diagram synthesis 3' > 5' ??????).

Figure 1 looks like this:

3' ==CS1==-----> 5'
5' ------------------------------------ 3'
3' ------------------------------------ 5'
                          5' <-----==CS2== 3'


5' ===CS1===------------------------------------===CS2=== 3'
3' ===CS1===------------------------------------===CS2=== 5'

I am questioning how it is possible that this directionality primes synthesis from 3' > 5' ?? that is impossible?

3' ==CS1==-----> 5'
5' ------------------------------------ 3'
3' ------------------------------------ 5'
                          5' <-----==CS2== 3'

[ScottDansk]

If you look at figure one (TAS) in here:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4440812/

It is the other way around & correct.

Still don't understand how you get from 2 ssDNA fragments with CS1/2 at the 5' end to bind preferentially over the genomic DNA and create dsDNA with CS1/CS2 at either end

[ScottDansk]

If you look at figure one (TAS) in here:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4440812/

It is the other way around & correct.

Still don't understand how you get from 2 ssDNA fragments with CS1/2 at the 5' end to bind preferentially over the genomic DNA and create dsDNA with CS1/CS2 at either end