GeneticsDNA III: The Replication of DNA

by Nathan Lents, Ph.D.

Did you know?

Did you know that DNA in a human body must make exact copies of itself not merely thousands of times, not millions of times, and not even billions of times, but staggeringly, more than a trillion times? Whether a yeast, a bacterium, or a human cell, every living cell must be able to copy all of its DNA with amazing accuracy.


In the field of molecular biology, scientists examine how DNA encodes all the complexities of living things. This third module in the DNA series focuses in the process by which DNA is replicated. The module describes the DNA synthesis assay, where scientists were able to replicate DNA in a test tube. Advancements in understanding the features and properties of DNA replication are discussed.

Terms you should know
  • discontinuous = having interruptions, gaps, or breaks; not proceeding in a smooth unbroken flow; disconnected, fragmentary
  • crude = unrefined or unprocessed; occurring in a natural state
  • extract = a substance that is pulled out or separated from its source; a preparation that contains the active principle or material to be studied
  • in vitro = in a test tube; in an artificial environment outside of a living organism
Table of contents

The discovery that DNA is the material that forms our genes (see our DNA I: The Genetic Material module) opened the door to the modern field of molecular biology, sometimes called molecular genetics, in which scientists examine how DNA encodes all of the great complexities of living things. One of the first major advances of the new field of molecular biology was the deciphering of the DNA molecule's structure - the double helix (see our DNA II: The Structure of DNA module).

Part of the motivation behind scientists' extensive efforts to discover the structure of DNA was the long-held scientific principle that "structure begets function." In other words, what a cell or molecule does, and how it does it, is determined by its shape and structure. This makes sense even in our everyday experience. Consider a hammer or a screwdriver. These important tools can do what they do because of their unique shape. If we changed their shape, they wouldn't work very well. Shape drives function. The same is true for DNA.

to top

The synthesis of DNA

As mentioned in our DNA II module, the moment James Watson and Francis Crick first gazed upon their newly built model of DNA, they could see clues about one of the major properties that they knew DNA must somehow exhibit: self-replication. The mystery of self-replication had confused scientists for many years. But one thing was certain: Every cell, whether a yeast, a bacterium, or a human cell, must be able to copy all of its genes, all of its DNA. This is because when a cell divides in two, both resulting cells are genetically identical to each other and to the original parent cell. The sheer number of times that the DNA in your body has been replicated (and accurately) is astounding.

You began life as a single cell, a zygote, the result of the fusion of a sperm and an egg. Since then, you have developed into an organism with somewhere between 10 and 100 trillion cells (>10,000,000,000,000). And, with certain rare exceptions, every single one of your trillions of cells has the same DNA sequence as the one cell did when you were just a zygote. How does all of this copying of DNA take place?

As mentioned, the structure of the double-stranded DNA molecule gave powerful hints as to how DNA might be accurately copied. Specifically, the complementary base-pairing of DNA follows a strict pattern that allows us to accurately predict what one strand of DNA looks like just by looking at the other, complementary strand. Put another way, if someone took a regular DNA molecule, pulled the two strands apart, and showed us only one strand, we could accurately list the series of nucleotides of the missing strand.

Watson and Crick saw this possibility when they ended their paper saying, "It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material." This possible copying mechanism is called semi-conservative DNA replication, because if a cell would duplicate its DNA in this manner, the DNA helix would split and half of both of the new double helices would retain DNA from the original strand (Figure 1). While this scheme makes good sense, it was just a logical guess at first. It wasn't until the late 1950s that Matthew Meselson and Franklin Stahl performed the scientific experiment that showed that the replication of DNA was indeed semi-conservative (see our Meselson and Stahl: Models of DNA Replication.).


Figure 1: Schematic of DNA replication according to the rules of Watson-Crick base-pairing. In this model, the two strands of the original DNA molecule are first pried apart. Then, complementary nucleotides (A with T, G with C, etc.) are added opposite the nucleotides in both of the original strands. The result is two DNA molecules, both identical to the original strand (and thus to each other), and both with one old strand and one new strand.

In the 1950s, Meselson and Stahl, Watson and Crick, and many other scientists explored the properties of DNA using the intestinal bacterium Escherichia coli. Because a few rare strains of E. coli have been found to cause gastrointestinal illness, E. coli is frequently associated with outbreaks of food poisoning. But actually, most strains of E. coli are harmless and our large intestines are filled with this bacterium. E. coli was among the first routinely used "model organisms," a species that is chosen for extensive study in the laboratory because it offers certain practical advantages that make research easier. E. coli, in particular, is among the fastest growing organisms on earth, with a generation time of under 20 minutes in ideal conditions. Since long before they knew what DNA was, scientists had noticed that the amount of DNA in an E. coli cell (and any other cell for that matter) doubles prior to cell division. The pool of DNA in the cell is then split equally between the two "daughter cells" that result, so that both have the same amount of DNA that the original bacterium had before replication. Because all of this happens in E. coli in about 20 minutes, it was the logical organism for early molecular biologists to select.

Comprehension Checkpoint

Why did molecular biologists choose E. coli for laboratory studies?

to top

In vitro DNA replication

While Meselson and Stahl and others were testing the possible hypothetical models of DNA replication, other scientists set out to understand its molecular mechanism by re-creating it in a test tube. This process is called in vitro reconstitution and is often used in the field of biochemistry as a way of simplifying a complex cellular event so that it happens in isolation and can thus be observed and manipulated at will. The scientists who were first able to reconstitute DNA replication in a test tube were Arthur Kornberg and his wife Sylvy and the research team that they led. They achieved this incredible feat through a painstaking process of successive chemical purification of different proteins and other components from large batches of E. coli bacteria. By separating and purifying individual components, the Kornberg research team made several important discoveries about how DNA replication occurs.

These discoveries all began with the development of a critically important technique – the DNA synthesis assay. An assay is a quantitative laboratory measurement of a certain biological or chemical process, usually in a test tube (in vitro). The DNA synthesis assay is a technique for measuring the synthesis of new DNA molecules. The Kornberg laboratory was the first to develop this assay, and the assay itself is quite simple. First, DNA polymers are easily separated from free nucleotides because DNA is not soluble in solutions that contain trichloroacetic acid (TCA), while free nucleotides are. If a scientist adds TCA to a liquid mixture of DNA and free nucleotides, and the DNA will precipitate out, while the nucleotides will remain dissolved in the liquid. The DNA precipitate can then be easily separated from the liquid by centrifugation.

The second important feature of the DNA synthesis assay is its use of radioactively labeled nucleotides. A scientist can add radioactive nucleotides when preparing a DNA synthesis assay, and then later, if DNA synthesis has occurred, some of the radioactive label will be incorporated into the TCA-insoluble DNA. This provides evidence that some of the labeled nucleotides were polymerized into a new DNA molecule. This DNA synthesis assay is very simple to execute and also very quantitative, which means that it gives very reliable and reproducible numerical values that can be used to calculate how much DNA was made and how fast the synthesis took place.

Armed with this assay, the Kornberg laboratory was the first to report the synthesis of DNA outside of a living cell. The popular press of the time announced that Arthur Kornberg had "created life in a test tube."

Of course, this was hardly the case, but the new ability to synthesize DNA in vitro captured the attention of the general population and is recognized as one of the crucial successes paving the way for the emergence of genetic engineering in the 1970s and 80s. Initially, the laboratory synthesis of DNA was extremely slow (much slower than it occurs in a cell), and it occurred only when crude extracts of E. coli were added to the test tubes. Crude extracts contain all the contents of the cells – all proteins, nucleotides, DNA, RNA, lipids, carbohydrates, etc. Nevertheless, the DNA synthesis assay was a good starting point in which Kornberg and others could begin to dissect the process of DNA replication in detail.

The first discovery and arguably the most important occurred in 1955: Kornberg's research team purified the enzyme from the crude extract that is chiefly responsible for the synthesis of DNADNA polymerase. When purified DNA polymerase is added to the DNA synthesis assay, the synthesis of DNA occurs hundreds of times more rapidly than when it is not added. However, the in vitro synthesis of DNA still required the addition of small amounts of crude cell extract. This is because DNA polymerase does not make DNA all by itself – there are many other factors required and not all of these were known at the time. The Kornberg lab and others around the world worked to purify other important components from the crude extract, in the hopes that one day they could make DNA using only the necessary factors and no crude extract.

Some of these required components were obvious, while others were unexpected. For example, it was very quickly discovered that nucleotides were required for the synthesis of DNA, which isn't very surprising because it was well known, even in the 1950s, that nucleotides are the building blocks of DNA. However, only nucleotides in the tri-phosphate form could be used as DNA building blocks (Figure 2). Later studies demonstrated why this is so - the breaking of the high-energy terminal phosphate bond of each new nucleotide added to a growing DNA molecule provides the energy for the polymerization reaction.

Figure 2: Only nucleotide tri-phosphates can be used for DNA synthesis. Although nucleotides can exist with one, two, or three phosphates attached to the 5' carbon of the pentose sugar, Kornberg found that only triphosphate nucleotide can be used as building blocks for DNA synthesis. Later work demonstrated that the reason for this requirement is that the breaking of the high-energy covalent bond between the phosphates provides the energy for forming the covalent bonds between adjacent nucleotides of DNA.

Another important point that the Kornberg laboratory noted was that the test tube DNA synthesis reactions required the presence of an intact copy-template DNA in order for DNA polymerase to make more DNA. In other words, even in a test tube, DNA polymerase cannot build "random" DNA molecules through the willy-nilly polymerization of nucleotides. It can only make copies of DNA molecules that already exist. Think of it this way - DNA polymerase is like a copy machine, NOT like a computer with new sentences can be created. A copy machine cannot print anything unless it has a template to work with. So when Kornberg added purified intact DNA molecules to the DNA synthesis assay, once again the speed of DNA polymerase increased dramatically. (Prior to this discovery, DNA synthesis was occurring only because tiny amounts of DNA template were present in the crude extract that is added to the assay mixture.)

Comprehension Checkpoint

DNA polymerase makes it possible to synthesize DNA molecules in a test tube, a key aspect of genetic engineering.

to top

How DNA is pulled apart

In addition to the hunt for more of the individual factors involved in DNA replication, the DNA synthesis assay allowed researchers to study the properties of DNA synthesis. As scientists around the globe began to study DNA polymerase and DNA replication, they knew that the semi-conservative model of DNA replication, as proven by Meselson and Stahl, requires that the two original template strands of DNA are pulled apart in order to be copied separately. However, it was not known how this happens. Scientists had observed that the two strands of DNA are held very tightly together by the hydrogen bonds between complementary nucleotide base-pairs of the two strands. In the laboratory, the only way the two strands could be separated was by heating the DNA to near-boiling temperatures. Obviously, it is not likely that living cells generate high heat in order to pry apart the two strands of DNA, so the question remained, "Inside a living cell, what pulls apart the two original strands of DNA so that they may be copied?"

Figure 3: DNA synthesis begins at many locations. DNA replication begins at specific chromosomal locations called origins of replication. Linear chromosomes have many origins, allowing DNA synthesis to occur rapidly.

Because double-stranded DNA is very stable, scientists suspected that there must be an elaborate mechanism for pulling the two strands apart. Two research groups, including Arthur Kornberg's, discovered the answer in the late 1970s: an enzyme they named DNA helicase. This enzyme is capable of prying the two strands of DNA apart so that the two individual strands can then serve as templates for DNA polymerase, according to the semi-conservative model.

It turns out, however, that when helicase first pries apart a section of DNA, it does not start at the end of the molecule in the case of linear DNA, nor does it select a place at random. The initial "melting" of DNA occurs at specific locations, called origins of DNA replication. Each of these creates a bulge in the DNA double helix that is visible by electron microscopy. These bulges are called replication bubbles and represent sites of DNA synthesis (Figure 3).

When a replication bubble opens up and DNA synthesis begins, replication proceeds in both directions, away from the origin. A DNA helicase enzyme leads the way, unzipping the parental DNA as replication proceeds in its wake. Both of these mobile regions of DNA synthesis are referred to as replication forks, which are the sites at which the replication of DNA is executed (Figure 4).

Figure 4: The replication fork. Following formation of a replication bubble, DNA synthesis proceeds in both directions, away from the original origin. A replication fork is the site in which the two parental DNA strands are being pried apart and DNA replication is taking place.

Comprehension Checkpoint

Replication bubbles are bulges in the DNA helix that indicate

to top

DNA primers and unidirectional synthesis

Once scientists began to focus on the events that occur at replication forks, they made several interesting observations that helped them to realize that DNA synthesis was much more complicated than they first imagined. The first such intriguing discovery was made by a young Japanese scientist named Tsuneko Okazaki, while working as a postdoctoral fellow with Kornberg at Stanford. Okazaki noticed that DNA polymerase cannot simply begin copying a template once it is pried apart from its complementary strand. Something more is needed to "kick-start" the copying of DNA before DNA polymerase can jump into action. Okazaki then discovered that she could coax DNA polymerase into performing DNA replication if she added a short piece of DNA that was complementary to part of the DNA template (Figure 5). Because this short DNA molecule served to get DNA synthesis started, Kornberg named them primers.

Figure 5: DNA polymerase requires a primer. DNA polymerase cannot begin to copy a template DNA unless a small part of the new copy DNA is already in place. Shown in light green in this figure, these short polymers are called "primers."

The discovery of primers was a major advance because now the scientists knew all the crucial components that were needed to perform an efficient DNA synthesis reaction in vitro. They no longer had a need for crude cell extract. Furthermore, the discovery of primers led to another curious observation by scientists, including Tsuneko Okazaki and her husband Reiji Okazaki, both former trainees of Kornberg, who had returned to Japan and formed their own research group. The Okazakis noticed that when a DNA synthesis reaction is set up and a primer is added, DNA synthesis begins at the primer and proceeds in only one direction. Curiously, they did not observe replication of the DNA region on the other side of the primer.

Returning to the structural model of DNA built by Watson and Crick, the Okazaki research team realized that DNA polymerization was only occurring at one end of the primer, the 3' end, and continuing in that direction. This was not simply a peculiar artifact of in vitro DNA synthesis. DNA replication inside all living cells also proceeds only in one direction - 5' to 3' (Figure 6). This property is called unidirectional DNA synthesis.

Figure 6: Unidirectional DNA synthesis. DNA synthesis can only proceed in one direction. This is because new nucleotides can only be added to a growing DNA polymer by addition onto the free hydroxyl group at the 3' end. The other end, the 5' end, has no free hydroxyl group.

to top

Okazaki fragments

Once it was realized that DNA synthesis proceeds in only one direction, Okazaki, Kornberg, and the entire community of DNA scientists realized that this posed a serious problem for their understanding of the DNA replication fork. There was extensive evidence that DNA synthesis proceeds on both strands of the DNA template after the two strands are pulled apart, and they had seen how DNA polymerase enzymes follow behind DNA helicase, synthesizing the new DNA strands alongside both original template strands (the semi-conservative model of DNA replication). But how could this be if DNA synthesis can proceed only in one direction (Figure 7)?

Figure 7: Unidirectional DNA synthesis poses a problem for the replication fork. It was discovered that DNA synthesis can proceed in only one direction, but scientists had already observed that DNA synthesis does indeed occur on both strands of a replication fork. These two observations appeared to contradict each other.

It was Reiji Okazaki who first postulated the solution to this conundrum. He imagined that the only possible way that DNA replication can occur on both strands of a replication fork, but still proceed only in the 5' to 3' direction was if DNA synthesis was continuous on one of the strands, trailing steadily behind the DNA helicase, but discontinuous on the other strand, proceeding in short stretches away from the replication fork. These short stretches are called Okazaki fragments in honor of Reiji Okazaki: however, it was the work of the whole Okazaki research team, the Kornberg research team, and several others that confirmed Okazaki's hypothesis regarding discontinuous replication of DNA (Figure 8). It was Kornberg who coined the terms leading strand for the strand in which DNA replication is continuous, and lagging strand for the strand in which DNA synthesis occurs in short discontinuous Okazaki fragments of ~300 nucleotides of DNA.

Figure 8: Okazaki fragments are stretches of discontinuous DNA replication. One side of the replication fork allows steady, continuous replication and is called the leading strand. The other side, the lagging strand, must employ discontinuous replication that occurs in short stretches called Okazaki fragments.

Tragically, Reiji Okazaki died seven years after his famous discovery of discontinuous DNA replication. A native of Hiroshima, he was 15 years old when the first atomic bomb was dropped and was heavily irradiated while searching for his parents amongst the rubble. He suffered the effects of radiation sickness, finally succumbing to Leukemia at the age of 44. The Okazakis and Kornbergs were both great examples of husband-wife teams of scientists (Figure 9).

Figure 9: The Okazakis and Kornbergs. From left, Reiji and Tsuneko Okazaki, Alfred and Sylvy Kornberg. c1975. image © A-IMBN

Following these major breakthroughs, scientists moved relatively quickly in mapping out the other major players of the replication fork (Figure 10). For example, it was discovered by the Kornberg lab that the primer that is necessary to initiate DNA synthesis inside cells is actually made of RNA, not DNA, and is put in place by an enzyme called DNA primase. This RNA is eventually replaced with DNA by a specialized version of DNA polymerase, called DNA polymerase I (DNA pol I), while the main workhorse of DNA polymerase is actually DNA polymerase III (DNA pol III).

Figure 10: Other important factors that work in the replication fork. As research continued, a more complete picture of the events of the replication fork came into view. As shown here, DNA synthesis is a complicated process performed by the coordinated function of many factors.

Further, it was discovered that the individual Okazaki fragments of the lagging strand need to be covalently bonded together. The enzyme that seals the Okazaki fragments together is called DNA ligase (Figure 11). Because this enzyme "seals" two stretches of DNA together, DNA ligase would later prove to be an essential tool in genetic engineering, as DNA molecules from different sources were "cut and pasted" to make new combinations and new DNA sequences.

Figure 11: DNA ligases seals Okazaki fragments together. Because each Okazaki fragment is made separately, they need to be "sealed together" or else the lagging strand of DNA would have breaks in it. The enzyme DNA ligase seals these fragments together so that, when replication is complete, the lagging strand is just as smooth and unbroken as the leading strand.

Comprehension Checkpoint

During DNA synthesis, Okazaki fragments are seen on the _______ strand.

to top

The Kornbergs win the Nobel prize

For all of the important discoveries that led to our understanding of the molecular events that take place at the DNA replication fork, Alfred Kornberg was honored with the Nobel Prize in Chemistry in 1959. Throughout his life, Dr. Kornberg mentored many young scientists who went on to great accomplishments of their own, including the Okazakis, whose pioneering work with discontinuous DNA replication led to the discovery of Okazaki fragments. Other famous students of the so-called "Kornberg school" include research leaders around the world in both academia and the biotechnology industry. In fact, it is no surprise that the biotech industry itself started mainly in the San Francisco bay area, because Kornberg spent most of his career at Stanford University, just 30 miles south of San Francisco. Among the most successful students of Arthur Kornberg is his son Roger Kornberg, who claims to have "grown up in the lab" watching his father make crucial discoveries about DNA synthesis. With his own research team, Roger painstakingly studied the processes of RNA synthesis, also called gene transcription, which has many parallels to DNA synthesis.

Just as Arthur Kornberg earned the Nobel Prize in 1959 for deciphering the events of DNA synthesis, his son Roger was awarded the Nobel Prize in 2006 for a lifetime of research on RNA synthesis. The success of this father-son duo demonstrates how the mentoring of the next generation of scientists is among the most important work that scientists perform, a reality further emphasized by the fact that the large majority of scientific research takes place in the academic setting and involves young scientists-in-training as the foot soldiers of discovery.

Key Concepts

  • Once the structure of the DNA molecule was discovered, scientists could immediately envision a possible copying mechanism based on the rules of nucleotide-base pairing.
  • In order to study and observe DNA replication more directly, scientists in the 1950s devised techniques to perform DNA replication in a test tube, called the DNA synthesis assay.
  • By using the DNA synthesis assay, scientists were able to observe the features and properties of DNA replication and test various hypotheses about how the process works.
  • The process of DNA replication was identified by several teams of researchers all working together to break down the process into multiple steps that could more easily be studied individually.


Lehman, I. R., Bessman, M. J., Simms, E. S., and Kornberg, A.  Enzymatic Synthesis of Deoxyribonucleic Acid. I. Preparation of Substrates and Partial Purification of an Enzyme from Escherichia coli.  (1958)  J. Biol. Chem. v233, 163-170.

Bessman, M. J., Lehman, I. R., Simms, E. S., and Kornberg, A.  Enzymatic Synthesis of Deoxyribonucleic Acid. II. General Properties of the Reaction. (1958) J. Biol. Chem. v233, 171-177.

Meselson, M. and Stahl, F.W. The Replication of DNA in Escherichia coli. (1958)  PNAS v44: 671?82. doi:10.1073/pnas.44.7.671.PMID 16590258.

Kornberg, A.  (1991).  For the love of enzymes: The odyssey of a biochemist. Boston: Harvard University Press  

R. Okazaki, T. Okazaki, K. Sakabe, K. Sugimoto, and A. Sugino. Mechanism of DNA chain growth. I. Possible discontinuity and unusual secondary structure of newly synthesized chains.  (1968)  PNAS. v59:598-605.

Nathan Lents, Ph.D. “DNA III” Visionlearning Vol. BIO-3 (2), 2010.