Re: How does Taq Polymerase know when to stop DNA synthesis

First, please accept my apologies for the lengthy delay in my response. I would have responded immediately had the answer been immediately apparent. It was not, and then I forgot about the message, but now that I've been reminded of it, the answer seems to have emerged with 'incubation time'.

How does one end up getting discrete PCR products of a given length, despite the fact that Taq polymerase does not know when to stop? For it does not - it's a remarkably robust and efficient enzyme, that keeps on chugging along the single stranded DNA template, generating double stranded product - as long as there is template and available deoxynucleotide 'building blocks', from which to synthesize the complementary strand of DNA.

Why doesn't one end up with a population of DNA fragments of very different lengths? Why, if you use primers [that work] to amplify, say, a 123 bp region, you invariably end up with mainly products of 123 bp?

The reason is because of the nature of the reaction itself.

Presumably, this is what happens in the first few PCR cycles of a reaction in which a product of 123 bp is expected:

1) Cycle no. 1. One primer binds to the cDNA or DNA, upon which the Taq enzyme tethers to the 3' end of the primer and starts synthesizing DNA in a 3' direction. It doesn't know when to stop, so it keeps synthesizing DNA until the end of the 'synthesis' phase of the reaction, when the mixture is heated up to about 95 degrees C, the strands separate, and the Taq enzyme detatches. The mixture now contains long strands of DNA, which are likely to vastly exceed 123 bp in length.

2) Cycle no. 2. The other primer binds to the strands generated by cycle 1. Because of its sequence design, this primer hybridizes with these strands so that its 5' end (or the 'outside' end of the primer is 123 bp from the 'starting end' of products generated in the first cycle. Once it's hybridized, synthesis proceeds, but it can only proceed as long as there is template - after reaching the end of the template (which corresponds with the beginning of the strands generated in cycle 1), the reaction ends. Whether the Taq actually falls off the end of the DNA, or whether it sits there until the denaturing step (heat at 95 degrees C), I'm not sure, but either way, the synthesis has to stop because there is no more template. At the end of cycle 2, you can expect to have mixture of very long DNA strands, generated from the first reaction, and strands of 123 bp, generated from the second one.

3) Cycle no. 3. Ditto cycle 1, but this time, there will be two template DNAs, the very long strands generated from the first reaction, and the 123 bp templates generated from the second. Only 123 bp products will be obtained from 123 bp templates, and some more long strands will be generated by the long strands.

4) Cycle no. 4. Ditto cycle 2 - only 123-bp products will be generated from templates - whehter the template is long or short.

5) ditto cycle 3.

As 123-bp products are generated in each reaction - and longer ones are generated only in every other one, the 123-bp products come to dominate the mixture very quickly, and after 30 cycles (which is typical for a PCR amplification), the 123-bp products will vastly outnumber the longer strands - so much so that the latter will not be visible when the products are run out on a gel and stained with ethidium bromide - which 'lights up' the products when the gel is illuminated with a UV lamp.

I hope that this is clear. If not, please let me know, and I'll draw a diagram, which might help.

With best wishes,

Bette Phimister
b.phimister@natureny.com
fax: 212 545 8341