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
Mechanisms of Genetic Recombination

In the two previous sections of this chapter, we examined the mechanisms by which DNA nucleotide sequences are transmitted almost unchanged through Cell generations. Genetic stability is crucial for survival in the relatively short term, but the long-term survival of a species requires genetic variation to allow ADAPTATION TO A changing environment. Therefore, an important property of DNA is its capacity for rearrangements that can alter both the combination of genes in a given genome and their expression (its timing and level). DNA rearrangements are the result of genetic recombination. The events that constitute genetic recombination can be divided into two broad classes: general recombination and Site-Specific Recombination. In general recombination, genetic exchange occurs between homologous DNA nucleotide sequences, mostly between two copies of the same chromosome. One of the best-known Examples of this is the exchange of segments between homologous Chromosomes (homologs) during Meiosis. This exchange (Crossing-over), which occurs between closely paired chromosomes during the Cytology/cytology/16.html">Early stages of egg or sperm development (see Section 15.2.3), provides an opportunity to test different variants (alleles) of the same Gene in new combinations with other genes, thereby increasing the chances of survival in a changing environment (at least for some members of an interbreeding population (see Section 15.2.2). Meiosis is unique to eukaryotes, but the advantages of such gene combination are so great that prokaryotic organisms have also evolved processes such as mating and gene rearrangement through general genetic recombination. Site-specific recombination differs from general recombination in that it does not require DNA Homology. The exchange involves short, specific nucleotide sequences on one or both DNA helices involved in the process, which are recognized by a specific site-specific recombination enzyme. Thus, site-specific recombination alters the distribution of nucleotide sequences in The Genome. Sometimes these changes are timed and organized in a specific way, such as during the excision of an integrated bacteriophage from a bacterial chromosome. However, they can also be completely random, such as during the insertion of mobile elements into the genome.

As for the biochemistry of genetic recombination, as with METABOLISM/36.html">DNA Replication, most of what we know about these processes has been elucidated through studies on simple organisms, particularly E. coli and its Viruses.

5-31

5.4.1. General recombination processes are guided by base-pairing interactions between complementary strands of homologous DNA helices [36]

General recombination involves a series of intermediate steps that require some effort to understand. Furthermore, The Mechanism of strand exchange seems to differ slightly among different organisms. However, detailed genetic analysis of crosses in Bacteria, viruses, and Fungi suggests that, overall, the outcomes of general recombination are always the same:

1. Two homologous DNA Double helices are broken, and the broken ends of one homolog are joined to the corresponding ends of the other, resulting in two intact DNA helices, but now each consists of PARTS OF THE two original DNA molecules (Fig. 5-55).

2. The point of exchange (i.e., the site where the red helix joins the black one in Fig. 5-55) can occur at any position along the homologous nucleotide sequences of the chromosomes.

3. At the point of exchange, each polynucleotide strand of one helix is joined by base-pairing to a strand of the other helix, creating a staggered (heteroduplex) joint between the two different DNA helices (Fig. 5-56). Such joints can span several thousand Base Pairs. Exactly how they arise will be explained later.

4. No change in nucleotide sequences occurs at the point of exchange. The precision of breaking and rejoining is so high that not a single nucleotide is lost, added, or changed into another.

The mechanism of general recombination is such that exchange can occur only between two segments of DNA helices whose nucleotide sequences share a high degree of homology. This is ensured by the presence of a heteroduplex joint at the point of exchange, as such a joint can form only if complementary interactions between strands from the two original helices occur over a sufficiently long region. But how exactly does this staggered joint arise, and how do two homologous DNA helices destined to pair recognize the homology of their nucleotide sequences? As we will see, homologous regions first recognize each other directly through complementary base-pairing. Subsequently, base-pairing between complementary strands from the two DNA helices guides general recombination so that it occurs only within a sufficiently long region of homology between the two DNA nucleotide sequences. Even so, general recombination often leads to a redistribution of DNA nucleotide sequences; a heteroduplex joint may contain a small number of mismatched base pairs, and, more importantly, the two DNA helices undergoing crossing-over are usually not completely identical on either side of the joint.

Class="center">

Fig. 5-55. Breakage and rejoining of two homologous DNA double helices during general recombination. This process results in two crossover chromosomes.

Fig. 5-56. A staggered joint connecting two chromosomes at the site of crossing-over. The length of such joints often reaches several thousand NUCLEOTIDES.

5-32

5.4.2. General recombination is initiated at a break in one of the two strands of a DNA double helix [36]

Each of the two strands of a DNA molecule is wound around the other. Consequently, any complementary interactions between two homologous DNA double helices are possible only if a break first occurs in one of the two strands, freeing it for the necessary unwinding and rewinding. For the same reason, any reciprocal exchange of strands between two DNA double helices requires at least two breaks—that is, one single-strand break in each of the two double helices. Finally, it is clear that to form the staggered (heteroduplex) joint shown in Fig. 5-56, all four strands must be broken, because only then can each strand rejoin with a different partner. In general recombination, all these breaks and rejoinings are executed and coordinated in such a way that they can occur only when the two DNA helices contain sufficiently long regions of homologous nucleotide sequences.

Fig. 5-57. The reaction catalyzed by the recBCD protein, an enzyme involved in general recombination in E. coli. The protein binds to one end of a DNA double helix and travels toward the other end at a rate of about 300 nucleotides per second, using energy derived from the Hydrolysis of bound ATP. A DNA loop generated by the protein moves along with it.

When it reaches a specific eight-nucleotide sequence on the helix called a recognition site (such sequences occur at various locations on the E. coli chromosome), one of the strands is cleaved, releasing a short single-stranded "whisker." This "whisker" can initiate genetic recombination by pairing with a homologous helix (see Fig. 5-58).

Fig. 5-58. Schematic illustrating the initial single-strand exchange between two homologous DNA double helices during general recombination. A break in one of the DNA strands releases this strand, which then invades the second helix, forming a short paired region. Only two DNA molecules with complementary nucleotide sequences can pair in this manner and thereby initiate general recombination. Enzymes catalyzing all the steps shown here are known (see Figs. 5-57 and 5-60).

Experiments with many different organisms have revealed that a single break in just one of the two DNA strands is sufficient to initiate the events of general recombination. Factors that cause such single-strand breaks, such as chemical agents or certain types of radiation, have been found to initiate genetic recombination. Furthermore, it has been shown that one of the specific Proteins required for recombination in E. coli, namely the recBCD protein, introduces single-strand breaks into DNA molecules. The recBCD protein is a DNA-dependent ATPase that Functions as a DNA helicase—moving along the DNA helix and unwinding it to make its strands accessible. Under the action of the recBCD protein, which combines nuclease and helicase activities, a single-stranded region—a "whisker"—is generated on the DNA double helix (Fig. 5-57). Fig. 5-58 illustrates how the presence of such a single-stranded region can induce the initial interaction between two complementary regions of a DNA double helix.

5-34

5.4.3. DNA Hybridization can serve as a model for the complementary base-pairing step of general recombination [29, 37]

In its simplest form, the complementary interactions that play a central role in general recombination can be reproduced in in vitro experiments by the renaturation of DNA that has been separated into single strands. This renaturation (or hybridization) occurs when, As a result of random collisions of single DNA strands in solution, complementary nucleotide sequences align opposite one another and form a short stretch of double helix. This relatively slow helix nucleation step is followed by a very rapid "zippering" step: The Double Helix grows until the maximum possible number of Hydrogen Bonds is formed (Fig. 5-59). To form a new double helix in this way, the separated strands must be kept extended during annealing so that their bases are exposed. For this reason, in vitro DNA hybridization experiments are carried out at high temperatures or in the presence of organic Solvents such as formamide; under these conditions, even the short helices ("hairpins") that arise in a single DNA strand due to complementary interactions when it folds back on itself are "melted." Bacterial Cells, of course, cannot tolerate such harsh treatments. In them, the straightening of helices is achieved through the action of a special helix-destabilizing protein, or SSB protein. In E. coli, the SSB protein is required for both DNA Replication and general recombination; by binding cooperatively to the sugar-phosphate backbone of all single-stranded DNA regions, it maintains them in an extended conformation and makes the bases accessible. In this conformation, single-stranded DNA can pair either with nucleoside triphosphate molecules (during DNA replication) or with complementary regions of another single-stranded DNA (During genetic recombination). If DNA hybridization is carried out in vitro under conditions resembling those inside The Cell, the SSB protein accelerates DNA helix nucleation, and thus the entire annealing process, by more than 1000-fold.

Fig. 5-59. During DNA hybridization in vitro, DNA double helices are formed anew from previously separated strands. The restoration of helices depends on the random collision of two complementary strands. Most such collisions are unproductive (as seen on the left side of the figure), but some lead to the pairing of complementary bases over a short region (i.e., helix nucleation). This is followed by rapid "zippering," and the double helix is complete. Through this trial-and-error process, each DNA strand can find its complementary partner among millions of "mismatched" strands. Apparently, general recombination is always initiated in this way: complementary partners recognize each other by trial and error.

5-34

5.4.4. The recA protein in E. coli allows single DNA strands to pair with a homologous region of a DNA double helix [38]

General genetic recombination is a more complex process than the simple DNA hybridization described above. In general recombination, a single DNA strand released from another double helix must invade a DNA double helix (see Fig. 5-58). In E. coli, this requires the recA protein. This protein is the product of the recA gene, which, as discovered in 1965, plays a major role in chromosome conjugation. Biochemists searched for a long time for the important but elusive product of this gene, and finally, in 1976, it was purified. It turned out to be a protein with a Molecular Weight of 38,000 daltons. Like the helix-destabilizing protein, it binds tightly to single DNA strands in large, cooperatively formed clusters; however, the recA protein also has some unique properties. In particular, it has two DNA-binding sites, allowing it to hold a single strand and a double helix together. These two DNA-binding sites enable the recA protein to catalyze synapsis between a DNA double helix and a homologous region of single-stranded DNA, as shown in Fig. 5-60. The key step in this reaction is the search for homology through initial base-pairing between complementary nucleotide sequences (step 2 in Fig. 5-60). This interaction initiates the pairing process (see Fig. 5-58) and thereby triggers the exchange of single-stranded regions between the two recombining DNA double helices. In vitro experiments have shown that the E. coli helix-destabilizing protein (SSB protein) and the recA protein act cooperatively to facilitate pairing reactions. This may be why genetic recombination in E. coli cells is drastically reduced if either of these proteins is defective.

Fig. 5-60. In vitro experiments show that several different complexes can form between single-stranded DNA coated with recA protein and a double helix. First (step 1), an unpaired-base complex is formed. As soon as a region with a homologous nucleotide sequence is found (step 2), this is converted into a paired-base complex in which, however, the strands are not intertwined. Such a complex is unstable because the DNA in it is in an unusual form: its two strands are either not wound into a helix at all, or are wound in such a way that regions of right-handed (i.e., normal) and left-handed helices alternate. In step 3, the strand exchange is stabilized. To achieve this, a nick (not shown here) must occur in one of the two strands forming the helix, and then one strand must be wound around the other.

Once synapsis has occurred, the short heteroduplex region, where strands belonging to two different DNA molecules have begun to pair, is extended by protein-directed branch migration, which is also catalyzed by the recA protein. Branch migration can occur at any point where two single DNA strands of identical sequence attempt to pair with the same complementary strand; an unpaired region of one strand displaces a paired region of the other, thereby shifting the branch point, although the total number of paired bases remains unchanged. Spontaneous branch migration occurs with equal probability in both directions, and therefore it is unlikely to lead to the efficient completion of the recombination process (Fig. 5-61, A). In the presence of the recA protein, this migration becomes directional, so that the heteroduplex region rapidly expands, reaching several thousand paired nucleotides (Fig. 6-61, B).

Fig. 5-61. Two Types of branch migration observed in in vitro experiments. Spontaneous migration occurs in both directions, governed by chance, so the net displacement is very small. In contrast, migration mediated by the recA protein proceeds at a constant rate in only one direction; the energy for this is apparently supplied by the polarized assembly of the recA protein on a single DNA strand, proceeding in the indicated direction.

The catalysis of branch migration is linked to another feature of the recA protein. In addition to having two DNA-binding sites, this protein (like the recBCD protein) has an additional site for binding and hydrolyzing ATP; that is, it is a DNA-dependent ATPase. It binds to DNA much more tightly when ATP, rather than ADP, is bound to it. Moreover, incoming recA molecules preferentially bind to one end of the protein filament, and ATP is hydrolyzed to ADP in the process. Thus, it can be seen that the recA protein filaments assembling on DNA share much in common, in terms of assembly dynamics, with the tubulin or Actin filaments that form the Cytoskeleton; this is evidenced, in particular, by the fact that the directional movement of the recA protein along the DNA strand can drive the branch migration reaction, as shown in Fig. 5-61, B.

5-35

5.4.5. General genetic recombination usually involves a crossed-strand exchange [39]

The difficult and slow step in general genetic recombination is the single-strand exchange between two double helices (see Fig. 5-58). After this initial exchange, the homologous nucleotide sequences of the two interacting helices are aligned in register, and therefore the expansion of the paired region and the initiation of new exchanges between the two helices proceed rapidly. During these events, the excision of A number of nucleotides and local DNA resynthesis, similar to those occurring during DNA Repair, are frequently observed. However, There are many possible pathways, so different organisms often use mechanisms that differ in detail at this stage. Most mechanisms involve a crossed-strand exchange between two DNA helices as an intermediate step. One of the simplest pathways for The formation of such a Structure is shown in Fig. 5-62.

In a crossed-strand exchange structure (also called a Holliday junction), two homologous DNA double helices, after the initial pairing step, are held together by a reciprocal exchange of two of the four strands—one strand from each helix. No disruption of base-pairing is required to maintain this structure. The structure has several interesting and important properties. 1. The exchange point between the two homologous DNA helices, located where their two strands cross (Fig. 5-62), can move rapidly back and forth along the helix (branch migration). 2. The structure formed by crossed-strand exchange contains two crossed and two uncrossed strands. This structure can exist in various isomeric forms resulting from the rotation of its constituent elements relative to one another, as shown in Fig. 5-63. Isomerization changes THE POSITION OF the two pairs of strands: the two previously crossed strands become uncrossed, and vice versa.

In order to regenerate two separate DNA helices and thus terminate the pairing process, a break must occur in each of the two crossed strands. If this break occurs before the crossed-strand structure undergoes isomerization, the two original DNA helices will separate almost unaltered—each having only one of its strands modified over a short stretch (Fig. 5-63, top). If, however, the Cleavage of the two crossed strands occurs after isomerization, a portion of each original DNA helix will become joined (by a staggered joint) to a portion of the other helix; in other words, crossing-over will occur between the two helices (Fig. 5-63, bottom).

Fig. 5-62. Crossed-strand exchange. Many possible pathways lead from the structure shown in Fig. 5-58 (single-strand exchange) to the crossed-strand structure. One such pathway is shown here. Although the top diagram probably gives a more accurate representation of the crossed-strand structure, the bottom diagram makes it easier to understand the isomerization reaction illustrated in Fig. 5-63.

Fig. 5-63. Isomerization of a crossed-strand structure. In the absence of isomerization, cleavage of the two crossed strands causes the exchange to terminate without crossing-over (top). If isomerization occurs, cleavage of the crossed strands yields two crossover chromosomes (bottom). It is therefore believed that isomerization is required for the breakage and reunion of two homologous DNA double helices during general genetic recombination.

Fig. 5-64. General genetic recombination between two homologous chromosomes leading to crossing-over. Isomerization of the crossed-strand structure occurs as shown in Fig. 5-63.

Isomerization is thought to be required for crossing-over to occur between two chromosomes. Fig. 5-64 shows how this process might take place between two sister chromatids in mitotic cells or between non-sister chromatids during meiosis. Although isomerization must occur spontaneously at a certain frequency, in cells it may be accelerated or regulated in some other way. Some form of regulation likely occurs during meiosis, when two pairing DNA double helices are held closely together in the synaptonemal complex.

5.4.6. General genetic recombination combined with limited DNA Synthesis leads to Gene Conversion [40]

One of the fundamental laws of genetics states that both parents make an equal contribution to the genetic constitution of the offspring, since the offspring receives one complete set of genes from the mother and another from the father. Thus, when four haploid cells are formed from a single diploid cell by meiosis (Section 15.2.1), exactly half of all the genes in each of these cells should be maternal and the other half paternal. Verifying the validity of this statement for a complex Organism, such as a human, is of course impossible. Fortunately, there are organisms, such as fungi, in which all four daughter cells resulting from meiosis of a single cell can be isolated and analyzed. Such analysis has shown that there are exceptions to strict genetic rules. Sometimes meiosis yields three copies of the maternal variant (allele) of a given gene and only one copy of the paternal allele, indicating The conversion of one of the two copies of the paternal allele into a copy of the maternal allele. This phenomenon is known as gene conversion. Gene conversion is often associated with general genetic recombination, and this phenomenon may play an important role in the evolution of certain genes (see Section 10.5.2). Gene conversion is believed to be a direct consequence of two mechanisms: general genetic recombination and DNA repair.

During meiosis, heteroduplex joints are formed at crossover points between homologous maternal and paternal chromosomes. If The nucleotide sequences of maternal and paternal DNA differ slightly, a few mismatched base pairs are formed. The resulting disruption of the DNA double helix can be corrected by the repair machinery (see Section 5.2.7): it either removes some nucleotides from the paternal strand and replaces them with nucleotides complementary to the maternal strand, or performs the opposite operation, repairing the maternal strand. The result of this Mismatch Repair is gene conversion. Several other mechanisms can also bring about gene conversion, but in all cases, an event related to general genetic recombination is required to bring together two DNA copies with very similar nucleotide sequences. Since this creates an extra copy of one of the two sequences, some DNA synthesis must also occur. Genetic analysis shows that gene conversion usually occurs only over a short stretch of DNA, and in many cases, only a portion of a single gene is altered.

Fig. 5-65. A hypothetical mechanism of general recombination leading to gene conversion. In step 1, DNA polymerase begins synthesizing an extra copy of one of the strands of the red helix, displacing the original strand from the helix as a single strand. This single strand pairs with a homologous region of the black helix in the manner illustrated in Fig. 5-60. In step 2, a short unpaired region of the black strand is degraded, completing The transfer of The nucleotide sequence from one helix to the other. The overall result of these changes is usually revealed in the next Cell Cycle, after DNA replication separates the two mismatched strands (step 3).

Gene conversion can also occur during mitosis, although somewhat less frequently. As in meiosis, it probably arises here from the Repair of DNA heteroduplexes containing mismatched base pairs. Fig. 5-65 illustrates another hypothetical mechanism of gene conversion that is applicable to both meiosis and mitosis.

5-36

5.4.7. Site-specific recombination enzymes insert specific DNA nucleotide sequences into genomes and excise them [41]

Site-specific recombination differs from general recombination in that a specific recombinational enzyme recognizes specific nucleotide sequences in one or both of the recombining DNA molecules. Base pairing is not required (even in systems where it does occur, no more than a few base pairs are involved in heteroduplex formation). This form of recombination enables various types of mobile DNA sequences to move within chromosomes or to pass from one chromosome to another.

Fig. 5-66. Site-specific recombination by which bacteriophage λ DNA integrates into the host cell chromosome (E. coli). The specific sites recognized by integrase (gray circle) are specific DNA nucleotide sequences. Here they are symbolized by red rectangles (see also Figs. 5-74 and 9-19).

Site-specific recombination was first described for bacteriophage lambda. It is through this process that this phage integrates into and is excised from the E. coli chromosome. In its integrated state, bacteriophage lambda replicates as an integral part of the host cell DNA (see Section 5.5.6). When a phage particle enters a cell, the enzyme lambda integrase, encoded by one of the phage genes, is synthesized in that cell. This enzyme catalyzes the recombination process, which begins with many copies of the integrase protein binding tightly to specific nucleotide sequences on the circular bacteriophage chromosome. The resulting DNA-protein complex then binds to another specific DNA sequence, this time on the bacterial chromosome, thereby bringing the bacterial and bacteriophage chromosomes into close proximity (Fig. 5-66). Having brought them together, the integrase catalyzes the necessary DNA cleavage and ligation reactions; a short region where the nucleotide sequences are homologous is used to form a small staggered joint at the point of contact (Fig. 5-67, A). Integrase possesses DNA topoisomerase activity, but the individual steps of recombination follow one another so rapidly that it is impossible to detect any intermediate DNA forms, which presumably do occur.

The mechanism of site-specific recombination also ensures the excision of the phage from the bacterial chromosome, after which its rapid multiplication in the bacterial cell begins. The excision reaction is catalyzed by a complex that includes, in addition to integrase, another bacteriophage protein that the virus begins to produce only if the host cell is subjected to stress (see Fig. 9-20).

Many Other Enzymes that catalyze site-specific recombination are similar to lambda integrase in that they also require a short region of homology in the two segments of DNA to be joined. This requirement implies a fairly high selectivity of each of these Enzymes for the DNA sequences to be recombined.

In another class of site-specific recombination enzymes, this selectivity is less pronounced. Like lambda integrase, each of these enzymes recognizes a specific DNA sequence in the mobile genetic element whose recombination it catalyzes. What distinguishes these enzymes from lambda integrase is that they do not require a specific target sequence, nor do they form a staggered (heteroduplex) joint. Instead, they produce a staggered cut in the target DNA, generating free ends of the DNA strands that are then covalently joined to the specific DNA sequence of the mobile genetic element (Fig. 5-67, B). As a result, the entire mobile element becomes integrated into the target DNA molecule. Short single-stranded gaps remain on both sides of the integrated mobile element in the recombinant DNA molecule, which are filled in by DNA polymerase, completing the recombination process. As is clear from Fig. 5-61, B, this creates two short identical nucleotide sequences on both sides of the integrated element; in all likelihood, the site-specific recombination enzymes recognize these identical sequences flanking the mobile element.

Fig. 5-67. Two mechanisms used by different classes of site-specific recombination enzymes. In both cases, a specific enzyme (shown in gray) binds to a specific nucleotide sequence in the chromosome of the mobile genetic element (indicated by hatching) and holds this sequence in close contact with a specific region of the target chromosome. A. The enzyme makes a staggered cut on both sides of a very short homologous sequence on both chromosomes (12 nucleotides in the case of lambda integrase) and then joins the partner strands with a short staggered joint. B. The enzyme makes a staggered cut in the target chromosome and joins its protruding ends to the bluntly cut ends of the mobile element. In this case, two short identical nucleotide sequences appear on both sides of the integrated element—duplications of the corresponding region of the target DNA (from 3 to 12 nucleotides, depending on the enzyme).

One such non-selective enzyme has been isolated in active form from bacteriophage Mu. Like DNA topoisomerase, it is capable of catalyzing all cleavage and rejoining reactions without any energy source (such as ATP). This enzyme belongs to a bacteriophage, but there is every reason to believe that similar enzymes exist in other organisms, such as bacteria, fruit flies, or humans. This assumption stems from the fact that Mobile Genetic Elements are similar in all of these organisms.

Conclusion

Genetic recombination mechanisms enable the movement of large DNA fragments from chromosome to chromosome. The reaction pathways evolved for this purpose ensure that as two DNA helices break and rejoin, they undergo minimal damage, allowing the easy restoration of two intact chromosomes. There are two classes of recombination events. In general recombination, the initial reactions depend on complementary interactions occurring over extensive regions between the strands of the two DNA double helices involved in recombination. General recombination can therefore occur only between two homologous DNA molecules; and, although chromosomes exchange genes in the process, the overall sequence of gene arrangement in the chromosome is not disrupted. In site-specific recombination, the pairing reactions depend on the recognition—mediated by a specific protein—of the two nucleotide sequences that are to recombine; no significant homology is required. Site-specific recombination typically alters the relative arrangement of nucleotide sequences in chromosomes.