Molecular Biology of the Cell - Volume 1 - Alberts B., Bray D., Lewis J., Raff M., Roberts K., Watson J. 1994

Molecular Organization of Cells
Basic Genetic Mechanisms
DNA Replication Mechanisms

Living organisms must not only maintain the integrity of DNA nucleotide sequences through DNA Repair, but also replicate their DNA with high fidelity before each Cell Division. During METABOLISM/36.html">DNA Replication, The rate of polymerization ranges from approximately 500 NUCLEOTIDES per second in Bacteria to about 50 nucleotides in mammals. Clearly, the Enzymes catalyzing replication must work both accurately and rapidly. This speed and accuracy are achieved by a specialized multienzyme complex that drives the replication process. Composed of several different Proteins, this complex constitutes a highly sophisticated and efficient 'replication apparatus'.

5.3.1. DNA replication, like DNA repair, is based on complementary base pairing [25]

The template activity of DNA is reflected in the fact that its nucleotide sequence is copied (either entirely or partially) via complementary base pairing (A with T or G with C) into a complementary nucleotide sequence of DNA or RNA. This process requires the recognition of each nucleotide in DNA by a free (unpolymerized) complementary nucleotide, and the mandatory Separation (at least temporarily) of the two DNA strands, so that the hydrogen-bonding donor and acceptor groups on each base become accessible for complementary pairing. In this way, incoming single nucleotides align in a specific order along the template DNA strand for enzymatic polymerization, yielding a new polynucleotide chain. In 1957, the first enzyme catalyzing nucleotide polymerization was discovered and named DNA polymerase. It was shown that the substrates for DNA polymerase are deoxyribonucleoside triphosphates, which polymerize on a single-stranded DNA template (the two-step mechanism of this polymerization is shown in Fig. 5-34 in connection with the Structure/133.html">Discussion of DNA repair). Later, RNA polymerase was also isolated, which utilizes ribonucleoside triphosphates as substrates.

During DNA replication, each of the two parental strands serves as a template for the synthesis of a new strand. Consequently, the extremely long nucleotide sequence of cellular DNA is replicated in what is known as a 'semiconservative' manner, and each of the two daughter Cells receives a new DNA double helix consisting of one parental strand and one newly synthesized strand (see Fig. 3-11).

5-23

5.3.2. The Replication fork is asymmetrical [26]

Studies conducted in the early 1960s on replicating Chromosomes, pulse-labeled with the radioactive DNA precursor ³H-thymidine, revealed a distinct, well-defined region of replication that moves along the parental DNA helix. Because of its Y-shape, this active region was named the replication fork. It is here that daughter DNA molecules are synthesized by a multienzyme complex containing DNA polymerase.

At that time, it seemed highly likely that the simplest mechanism of DNA replication would involve the continuous growth of both new strands, nucleotide by nucleotide, as the replication fork moves from one end of the DNA molecule to the other. However, because the two strands in the DNA double helix are antiparallel, one of the daughter strands must grow in the 5' → 3' direction, while the other must grow in the 3' → 5' direction. In that case, the replication fork would require two different DNA polymerases. One would extend the strand in the 5' → 3' direction (Fig. 5-34); here, each incoming monomer (deoxyribonucleoside triphosphate) provides the energy required for its addition to the chain (carried by its triphosphate group). The other DNA polymerase, moving in the 3' → 5' direction, would have to catalyze 'HEAD growth'; in this scenario, the energy needed to add each subsequent nucleotide would have to be carried by the end of the growing DNA strand. In reality, no such (3' → 5') DNA polymerase exists (Fig. 5-38), although biochemists are familiar with some other polymerization processes that proceed via a 'head growth' mechanism (see Fig. 2-34). How, then, does strand growth in the 3' → 5' direction occur? A possible answer to this question was suggested in the late 1960s by experiments with radioactively labeled DNA precursors. If growing cells are exposed to highly radioactive ³H-thymidine for just a few seconds, the label is incorporated only into the DNA synthesized at the very last moment—that is, the part immediately trailing the replication fork. This pulse-labeling method revealed that during bacterial DNA replication, fragments of 1000 to 2000 nucleotides are formed and transiently exist in the replication fork region (subsequently named 'Okazaki fragments'; in eukaryotes, they are much shorter, ranging from 100 to 200 nucleotides). Shortly thereafter, it was shown that the synthesis of these DNA fragments occurs exclusively in the 5' → 3' direction; the synthesized fragments are then joined into long DNA strands by the same enzyme that seals nicks in the DNA helix during repair, namely DNA ligase (see Fig. 5-35).

Class="center">

Fig. 5-38. At first glance, the simplest mechanism of DNA replication appears to be the one shown in this (incorrect!) diagram. Both daughter strands would grow continuously by adding nucleotides in the 5' → 3' direction (bottom) and the 3' → 5' direction (top), respectively. However, no enzyme exists that can catalyze nucleotide addition in the 3' → 5' direction.

Fig. 5-39. STRUCTURE OF THE replication fork. Both daughter strands are synthesized in the 5' → 3' direction. To achieve this, the lagging DNA strand must be synthesized as a series of short fragments (Okazaki fragments).

The replication fork is asymmetrical (Fig. 5-39). Of the two daughter DNA strands being synthesized, one is made continuously and the other discontinuously. The former is called the leading strand, and the latter is the lagging strand. The Synthesis of the lagging strand lags behind because each Okazaki fragment can only be formed after the advance of the leading strand exposes the corresponding template region. Although this entire strand is synthesized overall in the 3' → 5' direction, each of its individual fragments is synthesized in the 5' → 3' direction. Because DNA on the lagging side of the fork is synthesized via a discontinuous 'backstitching' mechanism, no enzyme other than a (5' → 3') DNA polymerase is required at the replication fork.

5-24

5.3.3. The high fidelity of DNA replication implies the existence of a proofreading mechanism [27]

The fidelity of copying during DNA replication is so high that, on average, only about one error occurs for every 1-10⁹ complementary Base Pairs formed during the replication of the mammalian genome, which contains 3-10⁴ base pairs (see Section 9.1.3). This accuracy is far greater than would be expected, considering that replication involves more than just The formation of standard complementary base pairs. In normal DNA, rare tautomeric forms of all four bases transiently arise with a frequency of 10^-4-10^-5. These forms form mismatched pairs. For instance, a rare tautomeric form of C pairs with A instead of G, resulting in a mutation (Fig. 5-40). Thus, the high fidelity of DNA replication is determined by the presence of proofreading mechanisms that eliminate such errors.

Fig. 5-40. An example of a base mismatch occurring during DNA replication: in its thermodynamically unfavorable tautomeric form, cytosine readily forms Hydrogen Bonds with adenine.

Fig. 5-41. Diagram illustrating the proofreading process (error correction) during DNA Synthesis catalyzed by bacterial DNA polymerases. A similar proofreading mechanism is believed to operate in Eukaryotic cells.

One important proofreading mechanism depends on the unique properties of DNA polymerase. Unlike RNA polymerases, DNA polymerases cannot initiate the synthesis of a new polynucleotide chain by simply joining two nucleotides together; they can only add new nucleotides to the pre-existing 3'-OH end of a polynucleotide strand that is paired with the template DNA strand (see Fig. 5-34). This pre-existing strand, to which nucleotides are added, is called a primer. DNA molecules with a primer that has an unpaired 3'-OH end cannot serve as templates. Bacterial DNA polymerases, however, can work with them. Upon binding to such DNA molecules, they utilize their intrinsic (3' → 5') exonuclease activity to remove (by Hydrolysis) any mismatched nucleotides at the primer terminus. They excise exactly as many nucleotides as needed to create a paired primer end, thereby generating an active template. By acting in this manner, DNA polymerase Functions as a 'self-correcting' enzyme, correcting its own errors made during polymerization. Fig. 5-41 illustrates how this type of proofreading can be used to remove mismatched C—A pairs formed by the rare tautomeric form of cytosine.

The requirement for a correctly paired terminus is precisely what enables DNA polymerase to correct its own errors. Such an enzyme could presumably initiate DNA synthesis in the complete absence of a primer only if it lost The ability to distinguish between paired and unpaired ends. At the same time, RNA polymerases involved in Gene Transcription (see Section 5.1.1) apparently do not require self-correction, because transcription errors are not passed on to the next generation, and occasional defective molecules do not play a critical role. RNA polymerases can initiate the synthesis of new polynucleotide chains without a primer, with errors occurring at a frequency of 10^-4 during both RNA Synthesis and Translation—that is, the translation of mRNA nucleotide sequences into protein Amino acid sequences.

5-22

5.3.4. DNA replication in the 5' → 3' direction enables efficient proofreading

It is highly likely that the unidirectional Nature of DNA replication (5' → 3') is dictated by the stringent accuracy requirements of the process. If a DNA polymerase existed that added deoxyribonucleoside triphosphates to a growing polynucleotide chain such that the chain grew in the 3' → 5' direction, the activating triphosphate group would be carried by the growing 5' end of the chain rather than by the incoming mononucleotide. In this case, polymerization errors could not be removed by simple hydrolysis, because the resulting free 5' end would immediately terminate DNA synthesis. Clearly, a newly mismatched base at the 3' end is much easier to remove than a similar base added to the 5' end of a DNA strand. Therefore, although the actual mechanism of DNA replication shown in Fig. 5-39 appears at first glance to be significantly more complex and cumbersome than the incorrect hypothetical mechanism shown in Fig. 5-38, this real-world mechanism is capable of providing much higher fidelity precisely because DNA synthesis proceeds exclusively in the 5' → 3' direction.

5-25

5.3.5. Synthesis of short primer molecules on the lagging-strand template requires a special enzyme [28]

Once a replication fork has opened, the DNA polymerase synthesizing the leading strand always has a paired 3' end available to initiate the synthesis of a new strand. The situation is quite different for the DNA polymerase responsible for lagging-strand synthesis. It takes only about 4 seconds to synthesize a short DNA fragment, after which it must switch to synthesize a completely different fragment on a new region of the template strand located some distance from the first (see Fig. 5-39). To do this, it requires a primer with a paired 3' end each time, and therefore a mechanism capable of producing such primers. This mechanism involves an enzyme called DNA primase. It uses ribonucleoside triphosphates to synthesize short RNA primers, which in eukaryotes are about 10 nucleotides long (Fig. 5-42). These primers are synthesized at intervals on the lagging-strand template, where they are elongated by DNA polymerase, thereby initiating a new Okazaki fragment each time. The DNA polymerase molecule continues this elongation until it runs into the RNA primer attached to the 5' end of the previous DNA fragment. To ensure the formation of a continuous DNA strand from these many fragments, a special DNA repair system comes into play, rapidly removing the RNA primer and replacing it with DNA. The process is completed by DNA ligase, which joins the 3' end of the new DNA fragment to the 5' end of the previous fragment (Fig. 5-43).

Why is preference given to an erasable RNA primer rather than a DNA primer that would not need to be removed? As noted above, a self-correcting polymerase cannot initiate the synthesis of polynucleotide chains de novo; this implies the converse: an enzyme that starts chains de novo cannot be capable of efficient self-correction. Therefore, any enzyme that catalyzes the initiation of Okazaki fragment synthesis would inevitably produce a relatively inaccurate copy (with at least 1 error in 10⁵). This would mean an enormous increase in the mutation rate, even though The amount of such copies retained in the final product would constitute no more than 5% of the entire genome (for example, 10 nucleotides in a fragment of 200 nucleotides). It is natural to think, therefore, that using RNA rather than DNA AS A primer offered a major advantage, because ribonucleotides automatically flag these sequences as "suspect copies" that must be removed.

Fig. 5-42. Schematic of the reaction catalyzed by primase, an enzyme that synthesizes short RNA primers on the lagging DNA strand. Unlike DNA polymerase, this enzyme can initiate the synthesis of a new polynucleotide chain by joining two nucleoside triphosphates. After forming a short polynucleotide, primase stops. DNA polymerase can then add nucleotides to the free 3' end.

5-26

5-33

5.3.6. Special proteins facilitate the unwinding of the DNA double helix ahead of the replication fork [29]

The DNA double helix must unwind as the replication fork advances so that incoming deoxyribonucleoside triphosphates can pair with the parental template strand. Under normal conditions, however, the DNA double helix is highly stable; the paired bases are held together so tightly that separating the two DNA strands in a test tube requires temperatures approaching the boiling point of Water (90°C). For this reason, most DNA polymerases can copy only a DNA molecule whose template strand has already separated from the other strand. Special proteins are required to open the DNA double helix and make the appropriate template strand accessible to DNA polymerase. These proteins are of two types.

DNA helicases were first isolated as proteins that, by binding to single-stranded DNA, catalyze ATP hydrolysis. As already noted in Chapter 3, ATP hydrolysis can cyclically alter the shape of a protein molecule, causing the protein to perform mechanical work (see Section 3.4.11). This very principle underlies the rapid movement of DNA helicases along single-stranded DNA. When encountering a region of double helix, these enzymes continue to move along their strand, thereby unwinding The Double Helix (Fig. 5-44). The unwinding of the DNA helix at the replication fork is probably carried out by two cooperating DNA helicases, one moving along the leading strand and the other along the lagging strand. Clearly, these two helicases must move in opposite directions along single-stranded DNA, meaning they must be different enzymes. Indeed, both types of DNA helicases have been identified. Studies in bacteria have shown that the lagging-strand DNA helicase plays the major role. We will discuss the reasons for this below.

Helix-destabilizing proteins (also called single-strand DNA-binding proteins, or SSB proteins) bind to single-stranded DNA without covering the bases, leaving them accessible for base-pairing. They cannot unwind long DNA molecules on their own, but by binding to single-stranded DNA, they facilitate any unwinding process; for example, they help DNA helicase unwind the double helix at the replication fork. On the lagging-strand template, SSB proteins bind cooperatively to single-stranded DNA regions, preventing the formation of "hairpins"—small double-Helical structures that could interfere with DNA synthesis by DNA polymerase (Fig. 5-45).

Fig. 5-43. Individual steps in the synthesis of each lagging-strand DNA fragment. In eukaryotes, RNA primers are synthesized on the lagging strand at intervals of approximately 200 nucleotides, and each primer is 10 nucleotides long.

5.3.7. Proteins at the replication fork act cooperatively to form a "replication machine" [30]

So far, we have discussed DNA replication as if it were carried out by a mixture of replication proteins acting independently of one another. In reality, however, most of these proteins are associated in a large multienzyme complex that moves rapidly along DNA. This complex is like a tiny "sewing machine": its "parts" are individual proteins, and its energy source is nucleoside triphosphate hydrolysis. The complex has been well studied only in E. coli bacteria and some Viruses, but there is every reason to believe that a very similar mechanism operates in eukaryotes as well (see Section 9.3.3).

The diagram in Fig. 5-46, which shows the replication fork in detail, illustrates how the individual parts of this "replication machine" function. Two identical DNA polymerases operate at the fork—one on the leading strand and one on the lagging strand. The DNA helix is unwound by the combined action of the DNA polymerase working on the leading strand and the DNA helicase moving along the lagging strand; this process is facilitated by cooperatively binding molecules of helix-destabilizing protein. While the DNA polymerase on the leading strand works continuously, the enzyme on the lagging strand pauses and restarts at regular intervals, using short RNA primers synthesized by DNA primase for polymerization.

Replication efficiency is greatly enhanced by the tight association of all these protein components. The primase molecule is directly linked to the DNA helicase, forming a structure on the lagging strand called a primosome, which moves with the replication fork and synthesizes RNA primers as it goes. The DNA polymerase molecule working on the lagging strand also moves in concert with the other proteins, synthesizing a series of new Okazaki fragments; to accommodate this, the DNA template strand is thought to fold back on itself, as shown in Fig. 5-47. The replication forks are thus integrated into a single large structure (with a total mass of > 10⁶ daltons) that moves rapidly along the DNA, enabling coordinated and efficient DNA synthesis on both strands of the fork.

Fig. 5-44. Action of DNA helicases. A small DNA fragment is annealed to a long single-stranded DNA, forming a short region of double helix. This helix is unwound as the helicase moves along the single-stranded DNA, catalyzing a reaction that requires ATP In addition to the enzyme. The energy source for the helicase movement is ATP hydrolysis (see Fig. 3-63).

Fig. 5-45. Effect of helix-destabilizing proteins on The structure of single-stranded DNA. Because each protein molecule prefers to bind to another molecule that has already bound (known as cooperative binding), these proteins form long clusters that straighten the DNA template strands and facilitate polymerization. Hairpin structures in free single-stranded DNA are formed by random base-pairing in short regions containing mutually complementary nucleotide sequences; they resemble the short helices that occur in all RNA molecules.

Fig. 5-46. Major Types of proteins acting at the replication fork (the diagram shows their localization on the DNA). The complex of DNA primase and DNA helicase on the lagging DNA strand is known as the primosome.

Behind the moving "replication machine," a series of unjoined Okazaki fragments remains on the lagging strand, still containing at their 5' ends the RNA primers required to initiate synthesis. These RNA primers must be removed and the fragments joined by repair enzymes operating behind the replication fork (see Fig. 5-43).

Figure 5-47. Schematic diagram illustrating current views on the arrangement of replication proteins in a moving replication fork. Instead of the two-dimensional structure shown in Fig. 5-46, this diagram shows how the DNA on the lagging strand folds back, creating a complex of two DNA polymerases—one for the leading strand and one for the lagging strand. Furthermore, this folding brings the 3' end of each completed Okazaki fragment close to the start site of the next fragment (cf. Fig. 5-46). By remaining in close contact with the other replication proteins, the lagging-strand DNA polymerase molecule can work continuously at the same replication fork; upon releasing a completed DNA fragment, it moves to the nearest new RNA primer to initiate synthesis of the next fragment. Note that in this diagram, one of the daughter DNA helices points down and to the right, while the other points up and to the left.

5.3.8. DNA replication errors in bacterial cells are corrected by a special proofreading system that recognizes base mismatches

In bacteria such as E. coli, cell division occurs every 30 minutes, making it relatively easy to detect rare individuals with altered traits in a large cell population. For example, a class of mutants has been isolated that exhibit a dramatically increased rate of spontaneous Mutations, which is associated with the presence of specific Mutator Genes in their cells. A well-known mutator gene encodes a defective form of the 3' → 5' proofreading exonuclease, which is a subunit of DNA polymerase (see Section 5.3.3). If a defect affects this protein, DNA polymerase loses its ability to proofread effectively, and numerous errors accumulate in the DNA that would otherwise be corrected during normal replication.

Studies of E. coli mutants carrying mutator genes have revealed another system that normally corrects replication errors missed by the proofreading exonuclease. This mismatch proofreading system, also known as the Mismatch Repair system, differs from the previously discussed DNA repair systems because it does not depend on the presence of abnormal nucleotides in the DNA that must be recognized and excised. Instead, it detects distortions on the outer surface of the helix caused by the poor fit of normal but noncomplementary bases. If this proofreading system simply recognized mismatches in replicated DNA and randomly excised one of the two mismatched nucleotides, it would make a mistake half the time by 'correcting' the template strand rather than the newly synthesized one, leaving the average error rate unchanged. For effective correction, the system must be able to distinguish between mismatched nucleotides and selectively remove them only from the newly synthesized strand (i.e., correcting replication errors specifically).

In E. coli cells, this recognition process is linked to the methylation of specific adenine residues in DNA. Methyl groups are added to all A residues in the GATC sequence, but only after some time has passed since the A was incorporated into the newly synthesized DNA strand. The new strands differ from the old ones because, immediately behind the replication fork, only the new strands contain unmethylated GATC sequences. Mismatch correction is carried out by a large multienzyme complex that scans both strands of the DNA double helix. This complex removes only the mismatched nucleotides, but it does so only after detecting an unmethylated GATC sequence on the same strand. Consequently, nucleotides are excised exclusively from the new strand, thereby correcting replication errors (Fig. 5-48).

Figure 5-48. Schematic of an experiment illustrating the operation of the mismatch repair system that corrects DNA replication errors in bacteria. A special protein complex removes mismatched nucleotides from the newly synthesized DNA strand behind the replication fork; this repair complex recognizes the new DNA strand by the unmethylated GATC sequences found within it. The diagram shows three DNA molecules with the same 'incorrect' nucleotide pair; however, in one molecule (A), methylated GATC sequences occur in both strands, in another molecule (B), there are no such methylated sequences at all, and in the third (C), they are present in only one of the strands. If these DNA molecules are treated with a cell extract containing the repair complex, the result shown here is obtained. The DNA molecule on the right side of the figure reproduces the situation found immediately behind the replication fork: the bottom strand corresponds to the new strand, which has not yet been methylated.

In eukaryotic cells, neither of these two proofreading mechanisms found in bacteria has yet been identified. However, the accuracy of replication in mammals and E. coli is approximately the same, suggesting that both described types of proofreading also exist in eukaryotes. It should be noted, however, that mammalian DNA lacks methylated A residues; therefore, the mechanism used by the mismatch repair system to recognize the newly synthesized strand must be different in this case.

5-27

5.3.9. Replication forks originate at replication origins [32]

In both bacteria and mammals, the formation of replication forks begins with the appearance of a special structure called a replication bubble. This is a small region where the two strands of the parental DNA double helix have separated and been used as templates for DNA synthesis (Fig. 5-49). For bacteria and some viruses that replicate in eukaryotic cells, it has been shown that the replication bubble forms at specific nucleotide sequences in the DNA molecule, which are called replication origins. These sequences consist of approximately 300 nucleotides. It is assumed that similar replication origins exist in eukaryotic chromosomes, although there is no definitive proof of this yet (see Section 9.3.2).

The process of replication fork initiation has been successfully reproduced in vitro in some cases. These experiments showed that in bacteria and Bacteriophages, the initiation of replication forks begins as shown in Fig. 5-50. Multiple copies of an initiator protein bind to specific sites at the replication origin, forming a large protein complex. This complex then recruits DNA helicase and loads it onto a single-stranded region of DNA adjacent to the origin. DNA primase is also recruited, forming a primosome which, moving away from the replication origin, synthesizes An RNA primer, enabling the synthesis of the first DNA strand to begin. The remaining proteins then rapidly assemble into two replication Protein Complexes that move away from the replication origin in opposite directions (see Fig. 5-49); they continue to synthesize DNA until both forks have traveled all the way to the end of the template.

Some additional details regarding the initiation of replication forks in eukaryotic chromosomes will be discussed in Chapter 9, in the section dealing with the Cell Nucleus.

Figure 5-49. Hypothetical mechanism for the formation of replication forks at replication origins (see also Fig. 5-50).

Figure 5-50. A simplified diagram illustrating the initial steps of replication fork formation at replication origins in E. coli and bacteriophage λ. Uncovering this mechanism required in vitro experiments using a mixture of highly purified proteins.

Subsequent steps lead (by a pathway that is not yet clear) to the initiation of three more DNA strands (Fig. 5-49). In E. coli, the initiator protein is the dnaA protein, and the primosome consists of the dnaB (DNA helicase) and dnaG (DNA primase) proteins.

5.3.10. DNA topoisomerases prevent DNA tangling during replication [33]

By depicting the DNA double helix as we have done so far—incorrectly, as a flat 'ladder'—we have ignored the 'winding problem.' However, for every 10 base pairs replicated at the replication fork, the parental double helix must make one complete turn around its axis. Consequently, for the replication fork to move forward, the entire chromosome ahead of it would have to rotate rapidly (Fig. 5-51), which for long chromosomes would require a massive expenditure of energy. During DNA replication, this problem is solved differently: by creating a kind of 'swivel' in the helix, using a special class of proteins called DNA topoisomerases.

A DNA topoisomerase acts as a kind of 'reversible nuclease.' It first breaks a DNA strand and then covalently attaches to the broken end. The covalent protein–DNA bond is relatively high-energy because it preserves the energy of the cleaved phosphodiester bond. Consequently, the reaction that breaks the strand is reversible and requires no additional energy input. In this respect, this mechanism differs significantly from that of DNA ligase, which we discussed earlier (see Fig. 5-35).

There are Different types of DNA topoisomerases. Type I topoisomerase breaks only one of the two strands of the DNA double helix, allowing the two DNA sections on either side of the break to rotate freely relative to each other around the phosphodiester bond opposite the break, which acts as the aforementioned 'swivel' (Fig. 5-52). Any tension in the DNA helix causes it to rotate in a direction that relieves this tension. Therefore, during DNA replication, rotation occurs only over a short stretch of the helix—in the region immediately ahead of the replication fork. A similar problem arising during Introduction/24.html">DNA Transcription is solved in the same way.

Figure 5-51. The 'winding problem' that arises during DNA replication. For the replication fork (in bacteria) to move forward at a rate of 500 nucleotides per second, the parental DNA helix ahead of the fork must rotate at 50 revolutions per second.

Figure 5-52. A reversible reaction that creates a single-strand break in DNA. In eukaryotes, this reaction is catalyzed by type I DNA topoisomerase. Enzymes of this group form a transient covalent bond with DNA.

Figure 5-53. An example of a reaction separating two interlinked circular DNA molecules, catalyzed by type II DNA topoisomerase. The action of these enzymes (unlike Reactions Catalyzed by type I DNA topoisomerases) is coupled with ATP hydrolysis, and some of them can introduce additional tension into the DNA helix. Type II DNA topoisomerases are found in both PROKARYOTES AND EUKARYOTES and, in all likelihood, participate in many DNA-related reactions.

Type II topoisomerase covalently binds to both strands of the DNA double helix and transiently introduces a double-strand break. Enzymes of this type are activated by chromosomal regions where helices cross over. Upon binding to such a crossover, the topoisomerase: 1) reversibly breaks one of the two Double helices, thereby creating a sort of 'gate' for the other, 2) forces the second double helix to pass through this break, and 3) reseals both broken strands and then dissociates from the DNA. Acting in this manner, type II topoisomerases very rapidly separate two interlinked circular DNA molecules (Fig. 5-53). In the same way, they prevent DNA molecules from becoming tangled, which would otherwise inevitably pose a serious problem during replication. Temperature-sensitive Yeast mutants are known that produce topoisomerase II which is inactivated at 37°C. If these yeast cells are heated to this temperature, their chromosomes remain tangled during mitosis and cannot separate. Just how essential a 'tool' like topoisomerase II is for untangling chromosomes will be understood by anyone who has ever tried to untangle a hopelessly knotted fishing line without scissors.

5-28

5.3.11. DNA replication in eukaryotes and prokaryotes is fundamentally similar [24]

Almost everything we know about DNA replication has been learned from experiments with purified multienzyme systems of bacteria and bacteriophages capable of carrying out DNA replication in vitro. The preparation of such systems in the 1970s was greatly facilitated by the isolation of mutants in a variety of different replication genes, which could be used to identify and purify the corresponding proteins (Fig. 5-54).

In eukaryotes, the enzymology of DNA replication is not yet understood in detail, mainly because obtaining the necessary mutant forms is much more difficult. However, the basic scheme of Replication in Prokaryotes and eukaryotes, including the geometry of the replication fork and the requirement for an RNA primer, appears to be the same. The main difference is that eukaryotic DNA is replicated not as naked DNA, but in the form of Chromatin, in which it is tightly bound to proteins belonging to the histone class. In Chapter 8, we will learn that Histones form disk-shaped complexes around which eukaryotic DNA is wound, resulting in regularly repeating structures called nucleosomes. Nucleosomes are spaced along the DNA molecule at intervals of 200 base pairs. This may explain why new fragments of the lagging DNA strand are initiated in eukaryotes at intervals 10 times shorter (100 to 200 nucleotides) than in bacteria (1000 to 2000 nucleotides). Furthermore, if nucleosomes serve as barriers that temporarily halt the progress of DNA polymerase, the presence of chromatin (rather than naked DNA) may also explain why replication forks move about 10 times slower in eukaryotes than in bacteria.

Figure 5-54. Obtaining mutants in bacteria and bacteriophages with various DNA replication defects has opened up opportunities to identify and purify enzymes performing previously unknown functions required for prokaryotic DNA replication. The temperature-sensitive mutants used here belong to the so-called conditional mutants; typically, their enzyme functions normally at low temperatures and is inactive at high temperatures. In 'non-conditional' mutants with replication defects, DNA synthesis does not occur at either low or high temperatures, and therefore these mutants are doomed to die. In a modified form, such 'in vitro complementation assays' are also useful in the biochemical study of many other processes.

Summary

A self-correcting DNA polymerase catalyzes the polymerization of nucleotides on both strands of the DNA double helix in the 5' → 3' direction, copying the template with high fidelity. Because the two strands of the DNA double helix are antiparallel, only one of the two strands (called the leading strand) can be synthesized continuously in the 5' → 3' direction. The other, lagging strand is synthesized as short fragments in a 'backstitching' manner. Self-correcting DNA polymerase is unable to initiate the synthesis of a new strand. Therefore, to initiate fragments of the lagging DNA strand, short RNA primer molecules are used, which are later removed and replaced by DNA.

The process of DNA replication requires the concerted action of many proteins. It involves: 1) DNA polymerase and DNA primase, which catalyze the polymerization of nucleoside triphosphates; 2) DNA helicases and helix-destabilizing proteins, which help unwind the DNA helix to be copied; 3) DNA ligase and an enzyme that degrades RNA primers, which are needed to seal the discontinuously synthesized fragments of the lagging DNA strand; 4) DNA topoisomerases, which help resolve the problems of DNA helix winding and tangling; 5) initiator proteins, which bind to specific DNA sequences at THE ORIGIN OF replication and facilitate the formation of a new replication fork. At the origin of replication, a protein complex consisting of DNA helicase and DNA primase (called a primosome) first binds to the DNA template; other proteins are then added to this complex, forming a multienzyme complex—the 'replication machine'—which carries out DNA synthesis.