NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Lodish H, Berk A, Zipursky SL, et al. Molecular Cell Biology. 4th edition. New York: W. H. Freeman; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Molecular Cell Biology

Molecular Cell Biology. 4th edition.

Show details

Section 12.1General Features of Chromosomal Replication

We first consider three general features of DNA replication: the semiconservative and bidirectional growth of new strands from a common site. Understanding these properties of the replication process provides the foundation for our discussion in the next section of the complex protein machinery that carries out replication.

DNA Replication Is Semiconservative

The base-pairing principle inherent in the Watson-Crick model suggested that the two new DNA strands are copied from the two old strands. Although this mechanism provides for exact copying of genetic information, it raised a new question: Is replication a conservative or semiconservative process? In the first mechanism, the two new strands form a new duplex and the old duplex remains intact, whereas in the second mechanism, each old strand becomes paired with a new strand copied from it (Figure 12-1).

Figure 12-1. Conservative and semiconservative mechanisms of DNA replication differ in whether the newly synthesized strands pair with each other (conservative) or with an old strand (semiconservative).

Figure 12-1

Conservative and semiconservative mechanisms of DNA replication differ in whether the newly synthesized strands pair with each other (conservative) or with an old strand (semiconservative). All available evidence supports semiconservative replication (more...)

The first definitive evidence supporting a semiconservative mechanism came from a classic experiment by M. Meselson and W. F. Stahl. E. coli cells initially were grown in a medium containing ammonium salts prepared with “heavy” nitrogen (15N) until all the cellular DNA contained the isotope. The cells were then transferred to a medium containing the normal “light” isotope (14N), and samples were removed periodically from the cultures. The DNA in each sample was analyzed by density-gradient equilibrium centrifugation, which can separate heavy-heavy (H-H), light-light (L-L), and heavy-light (H-L) duplexes into distinct bands. The actual banding patterns observed were consistent with semiconservative replication and inconsistent with conservative replication. Subsequent experiments of a different design with cultured plant cells demonstrated for the first time semiconservative DNA replication in eukaryotic chromosomes. Apparently all cellular DNA in both prokaryotic and eukaryotic cells is replicated by a semiconservative mechanism. For further information on these early experiments on semiconservative replication, see Classic Experiment 12.1 on the accompanying CD-ROM.

Most DNA Replication Is Bidirectional

Several possible molecular mechanisms of DNA strand growth would result in semiconservative DNA replication. In one of the simplest possibilities, one new strand derives from one origin and the other new strand derives from another origin (Figure 12-2a). Only one strand of the duplex grows at each growing point. In this mechanism, which operates in linear DNA viruses such as adenovirus, the ends of the DNA molecules serve as fixed sites for the initiation and termination of replication. A second possibility entails one origin and one growing fork (the point where DNA replication occurs), which moves along the DNA in one direction with both strands of DNA being copied (Figure 12-2b). Certain bacterial plasmids replicate in this manner. A third possibility is that synthesis might start at a single origin and proceed in both directions, so that both strands are copied at each of two growing forks (Figure 12-2c). The available evidence suggests that the third alternative is generally employed by prokaryotic and eukaryotic cells: that is, DNA replication proceeds bidirectionally from a given starting site, with both strands being copied at each fork. Thus two growing forks emerge from a single origin site.

Figure 12-2. Three mechanisms of DNA strand growth that are consistent with semiconservative replication.

Figure 12-2

Three mechanisms of DNA strand growth that are consistent with semiconservative replication. The third mechanism — bidirectional growth of both strands from a single origin — appears to be the most common (more...)

In the circular DNA molecules present in bacteria, plasmids, and some viruses, one origin often suffices, and the two resulting growing forks merge on the opposite side of the circle to complete replication (see Figure 7-2). However, the long linear chromosomes of eukaryotes contain multiple origins; the two growing forks from a particular origin continue to advance until they meet the advancing growing forks from neighboring origins. Each region served by one DNA origin is called a replicon.

Evidence for Bidirectional Replication

The first experimental support for bidirectional replication in eukaryotic cells was obtained by fiber autoradiography of labeled DNA molecules from cultured mammalian cells (Figure 12-3). Such studies have revealed clusters of active replicons, each of which contains two growing forks moving away from a central origin, thus providing unambiguous evidence of bi-directional growth. Most cellular DNA and many viral DNA molecules replicate bidirectionally. Such viruses serve as excellent models for the study of cellular DNA replication.

Figure 12-3. Demonstration of bidirectional growth of cellular DNA chains by fiber autoradiography.

Figure 12-3

Demonstration of bidirectional growth of cellular DNA chains by fiber autoradiography. If cultured replicating mammalian cells are exposed first to high and then to low concentrations of [3H]thymidine, the resulting DNA will be heavily labeled (“hot”) (more...)

If DNA from replicating eukaryotic cells is extracted and examined by electron microscopy, so-called replication “bubbles,” or “eyes,” extending from multiple replication origins are visible. Although such micrographs do not constitute conclusive evidence for unidirectional or bidirectional fork movement, electron-microscope studies of bubbles in viral DNA have provided evidence for bidirectional replication. If circular viral DNA molecules at different stages of replication are cut with a restriction endonuclease that recognizes a single site, the positions at the center of the replication bubble with respect to the restriction site can be determined (Figure 12-4). The most common result from such analyses is a series of ever larger bubbles whose centers map to the same site, indicating bidirectional replication of both DNA strands from that site. Thus both fiber autoradiography and electron microscopy have indicated that bidirectional DNA replication is the general rule.

Figure 12-4. Demonstration of bidirectional chain growth from a single origin in viral DNA.

Figure 12-4

Demonstration of bidirectional chain growth from a single origin in viral DNA. The replicating viral DNA from SV40-infected cells was cut by the restriction enzyme EcoRI, which recognizes a single site, and examined by electron microscopy. The electron micrographs (more...)

Number of Growing Forks and Their Rate of Movement

In E. coli cells, it takes about 42 minutes to replicate the single circular chromosome, which contains exactly 4,639,221 base pairs and is about 1.4 mm in length. Since the chromosome is duplicated from one origin by two growing forks, we can calculate that the rate of fork movement is about 1000 base pairs per second per fork. The rate of growing fork movement determined from fiber autoradiography of E. coli cells labeled for various times agrees with this calculated value, indicating that the fiber-labeling technique can provide a reasonable estimate of the rate of growing fork movement in vivo.

The rate of fork movement in human cells, based on fiber-labeling experiments, is only about 100 base pairs per second per fork. The entire human genome of 3 × 109 base pairs replicates in 8 hours; in this time, one fork theoretically could replicate ≈3 × 106 base pairs, suggesting that the human genome must contain a minimum of 1000 growing forks. However, fiber autoradiography and electron microscopy indicate that growing forks are spaced closer than 3 × 106 base pairs apart. A more likely estimate is that the human genome contains 10,000 – 100,000 replicons, each of which is actively replicating for only part of the 8 hours required for replication of the entire genome.

DNA Replication Begins at Specific Chromosomal Sites

Perhaps the most important decision every cell has to make is whether, and when, to replicate its DNA. DNA replication, like RNA synthesis and many other biological processes, is controlled at the initiation step. Such control would be most efficient if there are specific sites on chromosomes at which DNA replication always begins in vivo. As noted already, electron-microscope studies have shown that animal viruses have replication bubbles whose centers are always in the same approximate site (see Figure 12-4). Similar results have been obtained with circular bacterial and plant viruses and for bacterial, yeast, and mammalian plasmids. More detailed molecular studies indicate that replication of these DNAs actually begins at a defined sequence of base pairs near the center of these bubbles, called the replication origin.

Genetic and recombinant DNA experiments provide another way to define a replication origin experimentally as a stretch of DNA that is necessary and sufficient for replication of a circular DNA molecule, usually a plasmid or virus, in an appropriate host cell. In yeast, this definition has been refined to include sequences that direct replication once per S phase, the period of the cell cycle in which chromosomal duplication takes place. This important characteristic of DNA replication in eukaryotic cells will be considered in Chapter 13. We discuss three types of replication origins to illustrate some general conclusions about their nature: the E. coli oriC, yeast autonomously replicating sequences, and the simian virus 40 (SV40) origin. The detailed knowledge now available about the proteins required to start replication at the E. coli origin and the accumulating information about other origins and their use in vitro all suggest that most cellular DNA replication begins at specific sequences, possibly using similar mechanisms.

E. coli Replication Origin

The E. coli replication origin oriC is an ≈240-bp DNA segment present at the start site for replication of E. coli chromosomal DNA. Plasmids or any other circular DNA containing oriC are capable of independent and controlled replication in E. coli cells. Comparison of oriC with the origins of five other bacterial species including the distant species Vibrio harveyi, a marine bacterium, revealed that all contain repetitive 9-bp and AT-rich 13-bp sequences, referred to as 9-mers (dnaA boxes) and 13-mers, respectively (Figure 12-5). As we will see later, these are binding sites for the DnaA protein that initiates replication. In addition, the E. coli genome contains a segment of DNA with a relatively high A+T content adjacent to oriC; this sequence appears to be important in facilitating local melting of the duplex to reveal the two single-stranded DNA segments onto which the DNA replication machinery is loaded.

Figure 12-5. Consensus sequence of the minimal bacterial replication origin based on analyses of genomes from six bacterial species.

Figure 12-5

Consensus sequence of the minimal bacterial replication origin based on analyses of genomes from six bacterial species. Similar sequences constitute each origin; the 13-bp repetitive sequences (orange) are rich in adenine and thymine residues. The 9-bp (more...)

After E. coli DNA replication has initiated, the replication origins in the two daughter DNA duplexes become linked to specific proteins at different points on the plasma membrane. As the cell wall that divides the cell in two forms, this linkage ensures that one of the new DNA duplexes is delivered to each daughter cell (Figure 12-6). There is no visible condensation and decondensation of the DNA in bacterial cells, as there is in eukaryotic cells during mitosis.

Figure 12-6. DNA replication and cell division in a prokaryote.

Figure 12-6

DNA replication and cell division in a prokaryote. (a) In a bacterial cell, the partially replicated circular chromosome (blue) is attached to the plasma membrane at the origins of the two daughter DNAs (step 1 ). The origins of the replicated chromosomes (more...)

Yeast Autonomously Replicating Sequences

Each yeast chromosome, like all eukaryotic chromosomes, has multiple origins of replication. Cloning experiments indicate that about 400 origins exist in the 17 chromosomes of S. cerevisiae; more than a dozen of these have been characterized in detail. Each yeast origin sequence, called an autonomously replicating sequence (ARS), confers on a plasmid the ability to replicate in yeast and is a required element in yeast artificial chromosomes (see Figure 9-40). Detailed mutational analysis of one ≈180-bp ARS called ARS1 revealed only one essential element, a 15-bp segment, designated element A, stretching from position 114 through 128. Three other short segments — the B1, B2, and B3 elements — increase the efficiency of ARS1 functioning. Comparison of the sequences required for functioning of many different DNA segments that act as ARSs led to recognition of an 11-bp consensus sequence:

Image ch12e1.jpg
Element A in ARS1 is identical at 10 of 11 positions of this consensus sequence, and element B2 at 9 of 11.

DNase footprinting (see Figure 10-6) has shown that a complex of six different proteins called the ORC (origin-recognition complex) binds specifically to element A in ARS1 in an ATP-dependent manner. This complex also binds specifically to other ARSs tested. The ORC remains bound to an ARS throughout the cell cycle and during replication becomes associated with other proteins, an event that apparently triggers initiation of DNA synthesis. Yeast cells with mutations in any of these six ORC proteins are defective in DNA replication. All eukaryotic cells express homologs of these proteins, attesting to their importance in initiation of DNA replication. Chapter 13 details how initiation of DNA replication is coupled to specific steps in the cell cycle.

SV40 Replication Origin

A 65-bp region in the SV40 chromosome is sufficient to promote DNA replication both in animal cells and in vitro. Three segments of the SV40 origins are required for activity, as demonstrated by testing of origins containing specific mutations. As discussed below, researchers have used mammalian proteins and plasmids carrying the SV40 replication origin to study the molecular mechanisms of mammalian DNA replication.

Three Common Features of Replication Origins

Although the specific nucleotide sequences of replication origins from E. coli, yeast, and SV40 are very different, they share several properties. First, replication origins are unique DNA segments that contain multiple short repeated sequences. Second, these short repeat units are recognized by multimeric origin-binding proteins. These proteins play a key role in assembling DNA polymerases and other replication enzymes at the sites where replication begins. And third, origin regions usually contain an AT-rich stretch. This property facilitates unwinding of duplex DNA because less energy is required to melt A·T base pairs than G·C base pairs (see Figure 4-9b). Origin-binding proteins control the initiation of DNA replication by directing assembly of the replication machinery to specific sites on the DNA chromosome.


  •  The general principles of DNA replication seem to apply, with little modification, to all cells.
  •  Viewed at the level of whole chromosomes, DNA synthesis is initiated at special regions called replication origins. A bacterial chromosome has one origin, whereas each eukaryotic chromosome has many.
  •  Copying of the DNA duplex at the growing fork is semiconservative: that is, each daughter duplex contains one old strand and its newly made complementary partner (see Figure 12-1).
  •  Most chromosomes have bidirectional origins. DNA synthesis usually proceeds bidirectionally away from an origin via two growing forks moving in opposite directions; this movement produces a replication bubble (see Figure 12-2c).
  •  Replication origins typically contain multiple short repeated sequences. These unique DNA segments are recognized by multimeric origin-binding proteins, which in turn assemble other replication enzymes at the origin.
Image ch7f2
Image permission
Image ch9f40
Image ch10f6
Image ch4f9

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2000, W. H. Freeman and Company.
Bookshelf ID: NBK21650


  • Cite this Page
  • Disable Glossary Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...