7.4The Mechanism of DNA Replication


Outline of DNA Replication

DNA replication is a reaction involving polymerization of deoxyribonucleotide units to form macromolecular DNA. It is generally expressed as follows:

[dNMP]n + dNTP ↔ [dNMP]n+1 + PPi
(PPi refers to pyrophosphate*2)

With the loss of PPi from dNTP, it is added as dNMP to 3′-OH of [dNMP]n. This is why the direction of synthesis is said to be from 5′ to 3′. The direction of synthesis is not limited to DNA; it is also applied to RNA synthesis. The direction of synthesis of nucleic acids is 5′ to 3′.

*2 Phosphate (Pi) Pyrophosphate (PPi) [diagrams included in original]

Top of Page


DNA Polymerase

Table 7-3 DNA polymerases in E. coli

DNA polymerase links together deoxyribonucleotides. While its types in E. coli include I, II, and III, the main replication enzyme is DNA polymerase III. There are only 10 molecules of DNA polymerase III per E. coli cell, despite it being the primary enzyme involved in DNA replication (Table 7-3).

Among the many enzymes (α, β, γ, δ, and ε) in mammals, α, δ, and ε function in replication of nuclear DNA, γ in replication of mitochondrial DNA, and β and others in repair of DNA damage (Table 7-4). α is involved in the start of DNA strand synthesis, while δ and ε are involved in the following extension reactions of the continuous lagging and leading strands, respectively.

Table 7-4 DNA polymerases in eukaryotes

Top of Page


Replication Requires a Template

Fig. 7-5 Semiconservative replication using a template

In the replication process, while the original strands are unwound, new nucleotides are added to form pairs with the bases of the original strand (Fig. 7-5). This figure shows the DNA double helix model published by Watson and Crick in 1953, which suggested the possibility that DNA was replicated using a template. In fact, in the replication process, each original strand is used as a template to synthesize the new chain by pairing C with G and A with T. Thus, when replication is complete, there are two DNA double strands with exactly the same base sequence as the original double-stranded DNA. This replication method is termed semiconservative replication, with one strand of the final double strand being the original strand that served as a template (parent strand) and the other being a newly created strand (daughter strand). While many macromolecules are present in the body, including proteins and sugar chains, semiconservative synthesis using a template is a feature unique to DNA.


The Meselson–Stahl Experiment Indicates Semiconservative Replication

When E. coli are cultured in 15NH4+ (a heavy isotope of nitrogen), the nitrogen in all their nucleotide bases is replaced by 15N. Next, the E. coli cells are cultured in 14NH4+ (a normal isotope of nitrogen), DNA synthesis is promoted once and then a second time with DNA purified at each step, and equilibrium density gradient centrifugation*3 is performed with cesium chloride. DNA containing 15N has a higher buoyant density than that containing 14N, and thus, the two DNA are separated into different bands in the centrifuge tube. As shown in Column Fig. 7-1, the results indicate the occurrence of semiconservative replication.

Column Fig. 7-1 Meselson–Stahl experiment proving Semiconservative Replication

*3 Equilibrium density gradient centrifugation and buoyant density: When centrifugation is performed at high velocity in a concentration tube with a cesium chloride solution, a concentration gradient (density gradient) of cesium chloride is formed from the top to the bottom of the tube. Although a sharp gradient can be achieved with higher centrifugal force, under fixed centrifugal force, equilibrium is reached between the centrifugal force-based density gradient formation and diffusion-based homogenization, with that state maintained after the fixed gradient is achieved. DNA in solution accumulates to the point of a certain density of cesium chloride. This point is termed the buoyant density of DNA. As opposed to the density (specific gravity) of metal, the buoyant density of DNA varies depending on the composition of the solution. For example, human DNA in a solution of cesium chloride has a buoyant density of 1.7, while it has a buoyant density of 1.4 in cesium sulfate.

Top of Page


Replication Is Discontinuous

Double strands of DNA always run in opposite directions. This is true for completed DNA or DNA during synthesis. When considering DNA synthesis based on the Watson–Crick model (Fig. 7-6), synthesis of the daughter strand must progress from the 3′ to 5′ direction. However, the reaction performed by DNA polymerase always occurs in the 5′ to 3′ direction. Let us look at this point more closely.

Fig. 7-6 Discontinuous replication of lagging strand

When the daughter strand is synthesized with the two parent strands unwound, the double-stranded DNA appears to be triple stranded, with this region referred to as a replication fork (Fig. 7-6). When we examine the region of the replication fork at the point of DNA synthesis (replication point), one of the daughter strands (leading strand) is synthesized in the same direction of movement as the replication fork, while the synthesis of the other daughter strand (lagging strand) proceeds from 5′ to 3′, which is in the opposite direction of the replication fork (Fig. 7-6). Short DNA fragments of approximately 100 nucleotides are continuously synthesized on the lagging strand and then subsequently linked with each other. These short strands are called Okazaki fragments (named after their discoverer Reiji Okazaki), with this type of synthesis termed discontinuous replication.

Top of Page


Replication Primers

DNA polymerase governs the previously shown reaction:

[dNMP]n + dNTP ↔ [dNMP]n+1 + PPi

However, no reaction occurs when n = 1. Without a so-called “primer” and with some number of nucleotides attached, new nucleotides cannot be added to a given location and DNA polymerase cannot initiate replication. However, RNA polymerase can synthesize RNA using DNA as a template even when n = 1. Prior to DNA synthesis, RNA primer is synthesized by primase, which has RNA polymerase activity, contained in the DNA polymerase α compound, and DNA synthesis proceeds through DNA polymerase α ahead of the RNA primer (Fig. 7-7). This was another discovery of Okazaki et al.

Fig. 7-7 RNA primers in DNA synthesis

On the lagging strand, DNA synthesis continues until it runs into the next RNA primer. To produce a continuous new DNA strand from the many separated pieces of DNA fragments made on the lagging strand, the RNA primer is degraded and DNA ligase*4 finally seals the gap between these short DNA strands.

*4 An enzyme that connects the break between 3′-OH and 5′-PO4 on one strand of a DNA double strand. It cannot connect the break even if one base is lost.


Accuracy of DNA Replication

DNA replication must be extremely accurate. If DNA is synthesized using nucleotides that do not follow the pairing rule, the base sequence will be altered in the nascent (daughter) strand, resulting in a mutation. If the effect of a mutation that occurs in the region of an important gene is significant, it may result in cell death. Alternatively, in some cases, the cell may become cancerous. In humans, one diploid somatic cell contains six billion base pairs, with 100 billion to one trillion cell divisions each day.

In the elongation of a nascent strand by DNA polymerase, there is a 10−6 to 10−4 possibility of incorrect nucleotides being added. Errors are corrected by the proofreading activity of DNA polymerase, which can remove incorrectly added nucleotides and replace them with the correct ones. Errors missed during this process are detected by a mechanism that excises the incorrect nucleotides and replaces them with the correct nucleotide, and this mechanism is known as the base pair mismatch repair mechanism. Cells have several of these repair systems that help maintain the final error rate in the range of 10−11 and 10−10. A reaction system that maintains errors this low is difficult to construct artificially, even in the field of precision engineering.

Top of Page


Origins of Replication and Termination Points

Prokaryotes have a single origin of replication (ori) on the long DNA strand, and replication forks proceed bidirectionally from this point. Because of the circular nature of prokaryotic DNA, the replication forks meet on the opposite side of the circle at a location termed the termination point (ter). Both the origin of replication and the termination points have characteristic base sequences, with particular proteins governing initiation and termination. A DNA unit from initiation to termination of replication is called a replicon. Prokaryotic DNA consists of one replicon, and replication of the E. coli replicon takes 40 min. With much larger DNA content than prokaryotes, eukaryotes have multiple origins of replication on each DNA molecule. Their DNA molecules consist of multireplicon. Replication forks proceed on both sides from these multiple origins of replication, and replication of each replicon takes approximately 1 h.


Many Enzymes Are Involved in Replication

Replication reactions are actually more complicated than the reactions we have shown thus far (Column Fig. 7-2). Helicase, which unwinds the parent double-stranded DNA, functions at the tip of the replication fork. Single-strand binding proteins stably maintain the single strand exposed by helicase. Primase synthesizes RNA primer. DNA polymerase elongates the DNA strand. The clamp loader functions to maintain the sliding clamp, which forms a ring around the template DNA strand, so that DNA polymerase binds firmly to the strand and sustains replication without dissociating from it. As stated in the text, DNA synthesis proceeds on the lagging strand while the RNA primer is being degraded. Specialized degradation enzymes play a role here because eukaryotic DNA polymerase lacks RNA degradation activity. Finally, DNA segments on the lagging strand are joined together by DNA ligase. Ahead of the proceeding replication fork, topoisomerase (DNA gyrase) releases the tension (supercoils) held by the parent strands by cleaving the DNA strands and rejoining them. Various enzymes and proteins with such functions form a large replication complex, and similar essential mechanisms function in organisms from bacteria to humans.

Column Fig. 7-2 Overview of DNA replication

Top of Page


Regulation of Replication Initiation

From initiation to termination, DNA synthesis takes approximately 40 min in E. coli. From termination of DNA synthesis, cell division takes approximately 20 min. The time of both processes cannot be reduced. However, when E. coli are cultured under the best conditions, cell division takes 20 min, which results in an eightfold increase in cell number within 1 h. This is because the next round of DNA synthesis begins from the origin of replication 20 min after the initiation of DNA synthesis. Because there are multiple replication forks, this is termed multifork replication. At the time of cell division, ongoing multifork DNA replication is distributed between two cells.

On the other hand, in mammalian cells, origins of replication do not function again until DNA synthesis and cell division have been terminated . This is termed S-phase re-replication prevention. In this phenomenon, the licensing factor, a protein that provides permission for replication, is deactivated at the origins of replication after they have been used once, and a new licensing factor in the cytoplasm must bind to the origins of replication to reinitiate replication. Cytoplasmic licensing factors cannot pass through the nuclear membrane but can bind to the origins of replication when the nuclear membrane disappears for cell division. However, this is merely the license (permission, authorization) for replication, and a preliminary process is required to actually initiate replication, as we will see in Chapter 13.

Although the basics of replication are similar between prokaryotes and eukaryotes, there are a number of quantitative differences (Table 7-5).


End Replication Problem

The 5′ terminal region of the nascent strand of eukaryotic linear DNA is not replicated during replication. With an interior region from the 3′ end of the parent strand serving as a template, RNA primer synthesis of the nascent strand is initiated, followed by DNA synthesis of the nascent strand. While the RNA primer is eventually digested, the nascent strand cannot fill the region from the area where the RNA primer binds to the 3′ end of the template strand. Thus, the 5′ end of the nascent strand is always shorter than the template strand. This is the end replication problem. Thus, a cell with shorter DNA is produced whenever replication occurs. The linear DNA terminal region in eukaryotes comprises repeating short base sequence units called telomeres, with a shared base sequence of 5′-(TTAGGG)n-3′ in mammals. In humans, this region is 10–20 kilo base pairs in length. Most eukaryotes possess an enzyme called telomerase with an RNA template for this base sequence, which extends telomere DNA using the RNA template. Thus, telomeres do not become shorter. However, in humans, there is a limit to the total number of cell divisions, resulting in a finite replicative life span because of the fact that telomerase is eventually not expressed in the course of the development of human somatic cells. This is considered to determine human aging and life span. The 2009 Nobel Prize in Physiology or Medicine was awarded for the discovery of telomeres and telomerase.

Top of Page


DNA Damage and Repair

Fig. 7-8 Excision repair of DNA damage

DNA is continuously subjected to damage. Many natural and artificial chemical substances bind to DNA bases, form base–base bonds, or cleave DNA strands. Radiation, including ultraviolet and cosmic rays, can modify bases and cleave strands. DNA has been facing these dangers since the emergence of life, and all organisms are equipped with a range of mechanisms to detect and repair DNA damage. For example, damage to bases caused by ultraviolet rays or certain types of chemical substances is repaired by the so-called excision repair mechanisms in which the surrounding nucleotides of the damaged base are cleaved and deleted and the generated gap is filled by DNA polymerase (Fig. 7-8). Defects in the genes coding for the enzymes involved in this mechanism result in a hereditary disease called xeroderma pigmentosum, which makes patients prone to cancer. Many genetic diseases involve abnormalities in the repair enzyme system, which may lead to high incidence of cancer or even accelerated senescence (progeria). In other words, thanks to the many routinely functioning repair enzyme systems, genetic abnormalities in DNA can be repaired and their abnormal accumulation is kept low. Unrepaired damage may cause mutations.


Base Modification and Replication of Genetic Information Following DNA Replication

Following replication in eukaryotes, the cytidine base at a specific position on a DNA sequence is methylated by methyltransferase (see Selection 4 of Chapter 10).

While DNA methylation is biologically significant at a number of levels, one important point is that gene function is suppressed in highly methylated regions (Chapter 10). The differentiated functions of cells are determined by expression of genes (i.e., functioning of genes) that are responsible for them. In a hepatocyte, for example, while genes are expressed for the cell to function as a hepatocyte, expression of the genes for it to differentiate into a neuron or epithelial cell are suppressed. Only hepatocytes and not neurons or epithelial cells are produced from hepatocyte proliferation because their characteristics as hepatocytes are transmitted to progeny cells through DNA methylation.

The aforementioned recent research developments indicate that while DNA replication refers to the accurate replication of the base sequence in DNA, replication of genetic information should include replication of DNA methylation information, which specifies the status of gene expression.

Another biological significance of methylation is the discrimination of bases that need to be repaired and correct (not in need of repair) bases when mismatch repair occurs immediately following replication. Unmethylated strands are nascent strands that are subjected to repair of mismatched bases.

Top of Page