DNA replication: polymerases

By: John Brunstein   

In this month’s installment of “The Primer,” we’ll build on previous material from the first two installments (MLO, December 2012, pp. 18-20, and MLO, January 2013, pp. 26-28) to cover core aspects of how DNA replication occurs. In the context of a living cell, DNA replication is tightly integrated with cell cycle control and growth inhibition through complex mechanisms involving many cofactors and signalling molecules. We shall instead focus more on the very mechanistic steps of the actual DNA replication chemistry, as they can be reconstituted and controlled in vitro. Our interest in this is not only academic but also practical, as later “Primer” chapters will demonstrate how harnessing DNA replication is an important part of many MDx procedures.

DNA replication is performed by a class of enzymes known as DNA polymerases, frequently in MDx context just referred to as “polymerases.” While enzymes of this class from different biological sources differ in terms of physical structure, capacity to catalyze side reactions, accuracy (fidelity) with which they work, and specific kinetic parameters, all DNA polymerases share the same core functional behavior in many key aspects. These requirements and behaviors include the following:

  • DNA polymerases require a pre-existing template—a single strand of DNA which is used to direct formation of its complement.
  • At least some small portion of the template must have a double-stranded section, providing an available 3′ —OH group annealed to template.
  • DNA polymerases “grow” a single nascent strand in a 5′ to 3′ direction with respect to the growing strand, starting from this 3’—OH.
  • The reaction environment must contain a supply of deoxyribonucleoside triphosphates (dNTPs), which provide the building blocks for the nascent strand.
  • The environment needs to be maintained at physiological pH (usually achieved by buffering agents) and needs to include a supply of divalent cations—most frequently Mg++—which act in making the dNTPs chemically reactive.

Assuming suitable pH, dNTP availability, and cations, a minimal structure for a DNA polymerase to be able to work with could then be represented in text as Figure 1.

Figure 1



Figure 1

The bottom strand is the template which the polymerase will copy, and the short upper strand provides the required annealed 3′ —OH (on the final T, here) from which the polymerase will grow the nascent strand 5′ towards 3′.

An immediate question most people have at this point is, “Where does the small initiating section of nascent strand come from?” It’s a good question, and a complete answer covering multiple biological contexts is beyond the scope of this series. A short answer is that it arises in different ways in different biological systems, but in vitro (and the way we’ll primarily be concerned with) it can most easily be created by addition of short, defined sequence oligonucleotides made synthetically. Short single-stranded defined DNA sequences of this type—referred to as primers—are the specialty product of numerous life science companies, and for a researcher or clinical lab obtaining a custom primer is as simple as going online or faxing a request to one of these companies. Synthesis and delivery take mere days and are quite inexpensive. Methods for the design of such primers include use of special software for predicting sequence Tm (as covered in last month’s installment) and other parameters such as searching for unwanted complementarity to non-target sequences.  For now we’ll not get into that, and just accept that it’s there.

When a DNA polymerase is presented with an annealed template—primer pair such as that shown in Figure 1, and provided with the other environmental requirements previously listed, it catalyzes the following reaction:

  • The DNA polymerase binds to the primer/template pair, with the enzyme active site placed over the 3′ —OH of the primer.
  • dNTPs in the environment diffuse into the polymerase active site. If the dNTP is the complement of the next downstream template nucleotide (i.e., if it is a deoxycitidylate, in Figure 1), then its capacity to form H-bonds with the template strand “G” located here causes the dNTP to pause long enough in position for the enzyme to act. Any other dNTP which diffuses into the active site doesn’t align properly and rapidly diffuses back out.
  • When a dNTP pauses in the active site, the polymerase catalyzes formation of a chemical bond between the 3′ —OH of the primer (nascent strand) and the innermost (alpha) phosphate of the incoming dNTP. The outer two (beta and gamma) phosphate groups of the dNTP are displaced by this reaction.
  • The nascent chain has thus just grown longer, on its 3′ end, by one nucleotide. The DNA polymerase now moves down one nucleotide length, to position its active site over the new 3′ —OH
  • This cycle repeats over and over, growing the nascent strand in a template-directed fashion, one dNTP at a time, to create the template’s complement strand.

Thus, if we were to consider the example shown in Figure 1 after a few cycles, it would look like Figure 2:

Figure 2



Figure 2

(The new nucleotides are shown in lowercase for clarity.) The reader may want to mentally continue this process, and should proceed with c,t,a,g,g,c, and so forth.

While the above description holds true in general for all DNA polymerases, individual polymerases from different sources differ in terms of their speed, accuracy (rarely, the wrong or non-complementary dNTP can be incorporated in a growing chain; high fidelity polymerases have additional proofreading functions to detect and correct these mutations), ability to resist various potentially inhibitory substances, and thermostability (capacity to withstand high temperatures without losing functionality). Additionally in many MDx applications, polymerases may be modified chemically or by specific antibodies to have a hot start capacity, which means the enzyme is inactive when initially mixed into a reaction and only becomes reactive after an initial heating step. This is used to reduce possible side reactions the enzyme might catalyze during test setup. Selection of the specific DNA polymerase to be used in an MDx application requires a consideration of all of these factors in context of the test to be performed, and helps to explain why so many different DNA polymerase sources and formulations are available to researchers and clinical labs. The polymerase which works well in one application may not be optimal in another, and often experimental data in comparative studies must be obtained to determine what formulation works best in a particular assay.

Let’s focus a bit more on two aspects of polymerase activity which are ubiquitous. The first of these relates to where the polymerase starts working; as described earlier, this is at the 3′ end of a primer which must be annealed to a template strand. This fact, combined with the fact that such primers of a user-defined sequence can be chemically synthesized and last month’s discussion of complementarity and annealing of two complementary DNA single strands, leads to the observation that it must be possible in theory to direct a DNA polymerase in vitro to start work at any arbitrary but known DNA sequence. All that is required is synthesis and addition to the reaction of a synthetic primer uniquely complementary to the desired polymerase starting point. Recall in this context our earlier definition of “unique,” and that here it can mean anything from “unique to species” to “unique to particular organism,” depending on our knowledge of an organism’s DNA sequence and its degree of conservation to other isolates and closely or distantly related strains.

The second aspect is where the polymerase stops working. The answer here is very simple: a DNA polymerase will stop synthesizing a nascent strand if it runs out of template, much like a train having to stop if there’s no more track in front of it; or a DNA polymerase will stop synthesis if it is physically or chemically “knocked off” the template, or if the nascent strand is dissociated from the template strand. One obvious way in which this can happen would be through thermal manipulation of the reaction; essentially any two DNA strands will denature at temperatures slightly under 100°C. Thus, raising the temperature to this range on a reaction in which a polymerase is actively synthesizing a nascent strand can also stop the process.

This ability to localize the “starting” and “stopping” of a DNA polymerase in vitro to a user-mutable definition of a “unique” region will be the basis for one of the most revolutionary techniques in molecular biology and the core of many diagnostic techniques, the polymerase chain reaction (PCR) whose mechanism will be covered in more detail in the April 2013 installment of this series. Before then, however, we will briefly detour next month to the topic of sample extraction methods, or where and how we obtain the DNA and/or RNA templates from samples which we may then examine with our MDx methods.

John Brunstein, PhD John Brunstein, PhD, a member of the MLO Editorial Advisory Board, is President and CSO of British Columbia-based PathoID, Inc.

DNA replication: polymerases
John Brunstein
John Brunstein, PhD, is a member of the MLO Editorial Advisory Board. He serves as President and Chief Science Officer for British Columbia-based PathoID, Inc., which provides consulting for development and validation of molecular assays