Homologous Recombination Protein Machines

Organisms from all branches of life encode enzymes that catalyze the biochemical steps of recombination. In some cases, members of homologous protein families provide the same function in all organisms. In contrast, other recombination stops are catalyzed by different classes of proteins in different organisms but with the same genera] outcome. Our most detailed understanding of the mechanism of recombination comes from studies of E. coli and its phage. Thus, in the following sections, we first focus on the proteins that promote recombination in E. coli via a major DSB-repair pathway, known as the RecBCD pathway. Homologous recombination in eukaryotic cells, and the proteins involved in these events, are considered in later sections.

Table 10-1 lists the proteins that catalyze critical recombination steps in bacteria as well as those that serve these same functions in eukaryotes (the budding yeast S. cerevisiae is the best-understood example). These proteins provide activities needed to complete important steps in the DSB-repair pathway. In addition to these dedicated recombination proteins, DNA polymerases, single-stranded DNA-binding proteins, topo-isomerases, and ligases also have critical roles in the process of genetic exchange.

Notice that absent from the list in Table 10-1 is an E. coii protein that introduces DSBs in DNA. despite the fact that recombination via the RecBCD pathway requires a DSB on one of the recombining two DNA molecules. As discussed above, in bacteria, no specific protein has been found that carries out this task. Rather, breaks generated as a result of DNA damage or failure of a replication fork are the major source of these initiating events in chromosomal DNA.

The following sections describe the E. coii recombination proteins and how they perform their functions during recombination by the

TABLE 10-1 Pro kary otic and Eukaryotic Factors that Catalyze Recombination Steps

E. coli Protein Recombination Step Catalyst

Eukaryotic Protein Catalyst

Pairing homologous DNAs RecA protein and strand invasion

Rad 51

Ocm1 (in meiosis)

Introduction of DSB


Spott (in meiosis) HO (for mating-!ypc switching)

Processing DNA breaks to RecBCD generate single strands nelicase/nuctease

For invasion

MHX protein (also called Rad 50^58/80


Assembly of strand exchange proteins

RecBCD and Ree FOR Rad52 and Rad59

Holliday junction recognition RtivAB complex and branch migration


Resolution oT Holliday RuvC


Perhaps MusB 1 and others

DSB-repair pathway. These proteins are discusser) in the order in which they appear during the reaction pathway. First, we will see how the RecBCD enzyme processes DNA at the site of the DSB to generate single-stranded regions. Next, the structure and mechanism of RecA, the strand-exchange protein, is described. RecA, after assembling on the single-stranded DNA, finds regions of sequence homology in the DNA molecules and generates new base-pairing partners between these regions. The RuvA and RuvE proteins that drive DNA branch migration are then described. Finally, the Holliday junction-resolving enzyme, RuvC, will be considered.

The RecBCD Helicase/Nuclease Processes Broken DNA Molecules for Recombination

DNA molecules with single-stranded DNA extensions or tails are the preferred substrate for initiating strand exchange between regions of homologous sequence. The RecBCD enzyme processes broken DNA molecules to generate these regions of ssDNA. RecBCD also helps load the RecA strand-exchange protein onto these ssDNA ends. In addition, as we will see, the multiple enzymatic activities of RecBCD provide a means for cells to "choose" whether to recombine with, or destroy, DNA molecules that enter a cell.

RecBCD is composed of three subimits (the products of thn recB, rerC, and recD genes) and has both DNA helicase and nuclease activities, it binds to DNA molecules at the site of a double-stranded break and tracks along DNA using the energy of ATP-hydrolysis. As a result of its action, the DNA is unwound, with or without the accompanying nucleolytic destruction of one or both of the DNA strands. The activities of RecBCD are controlled by specific DNA sequence elements known as chi sites (for cross-over hotspot instigator). Chi sites were discovered because they stimulate the Frequency of homologous recombination.

Figure 10-5 shows a schematic of RecBCD processing a DNA molecule containing a single chi site to activate this DNA for recombination. RecBCD enters the DNA at the site of thn double-strand break

FIGURE 10-5 steps of DNA processing by RecBCD. Note- thai RecBCD protetn could have entered this DMA mclecUe from either or both broken ends However, chi sites Function only in one onentation. On the DNA molecule shown, the chi site is oriented such that it will only modify a RecBCD enzyme that is moving from right to left. The RecBCD enzyme has two DMA helicases: RecD, moves rapidly on the 5'-ending strand (bottom strand) and RecB, which moves slowly on the 3'-ending strand {top strand). Because these two subunits travel at different speeds, the DMA molecules accumulate a single strand DNA loop on the top strand during unwinding. A red X is shown on the RecD subunit, after the enzyme has encountered the chi site, to denote the inactivation or loss of this subunit

t rr and moves along the DNA, unwinding the strands. The Recti and RecD subunits are both DNA helicases, that is, enzymes that use ATP hydrolysis to melt DNA base pairs (see Chapter 8). The nuclease activities of RecBCD frequently cleave each strand during unwinding and thereby destroy the DNA.

Upon encountering the chi sequence, the nuclease activity of the RecBCD enzyme is altered. As RecBCX) moves into the sequence distal to the chi site (with respect to the broken DNA site at which the enzyme entered), it no longer cleaves the DNA strand with 3' —» 5' polarity. Furthermore, after the encounter with the chi site, the other DNA strand (the one with the 5' ~*3' polarity) is cleaved even more frequently than it was prior to the chi site. As a result of this change in activity, a duplex DNA molecule is converted into one with a 3' single-stranded extension terminating with the chi sequence at the 3' end. This structure it, ideal for assembly of RecA and initiation of strand exchange (see below). The molecular basis of the change in RecBCD's enzyme activity after the encounter with a chi site is unclear, but appears to be associated with either the inactivation or loss of the RecD sub-unit. The ssDNA tail generated by RecBCD must be coated by the RecA protein for recombination to occur. However, cells also contain single-strandnd DNA-binding protein (SSB) that can bind to this DNA, To ensure that RecA, nit hoi than SSB binds these ssDNA tails, RecBCD interacts directly with RecA and promotes its assembly.

Chi sites increase the frequency of recombination about tenfold. This stimulation is most pronounced directly adjacent to the chi site. Although elevated recombination frequencies are observed for at tout 10 kb distal to the chi site, they drop off gradually over this distance (Figure 10-6). The observation that recombination is stimulated specifically only on one "side" of the chi site was initially puzzling. It is now clear, however, why this pattern is observed: the DNA between the D5B (where RecBCD enters) and the chi site is cut into small pieces by the enzyme and is therefore not available for recombination. In contrast, DNA sequences met by RecBCD after its encounter with chi are preserved in a recombinagenic, single-stranded form and are specifically loaded with RecA.

The ability of chi sites to control the nuclease activity of RecBCD also helps bacterial cells protect themselves from foreign DNA that may enter via phage infection or conjugation. The eight-nucleolide chi site [GCTGGTGG) is highly overre presented in the E. coli genome: whereas it is predicted to occur only once every 65 kb, or about BO times, the chromosomal sequence reveals the presence of 1,009 chi sites! Because of this overrepresentation. E. coli DNA that enters an E. coli cell is likely to be processed by RecBCD in a manner that generates the 3' ssDNA tails, and thus activated for recombination. In contrast, DNA from another species (in which E. coli chi sites are not overrepresented) will Sack frequent chi sites. RecBCD action on this DNA will lead to its extensive degradation, rather than activation for recombination.

In summary, the DNA-degradation activity of RecBCD has multiple consequences: this degradation is needed to process DNA at a break site for the subsequent steps of RecA assembly and strand invasion. In this manner, RecBCD promotes recombination. However, because RecBCD degrades DNA to activate it, the overall process of homologous recombination must also involve DNA synthesis to regenerate the degraded strands. In addition, RecBCD sometimes functions simply to destroy DNA—as it does when foreign DNA lacking frequent chi sites enters cells, In this way, KecBCD can protect cells from the potentially deleterious consequences of taking up foreign sequences, which, for example, may carry a bacteriophage or other harmful agent.

donor DNA {linear}

relative recombination frequency recipient chromosome


donor DNA {linear}

relative recombination frequency recipient chromosome


FIGURE 10-6 Polar action of chi. This schematic sltcrws that a chi site specifically elevates recombination frequencies directly at the site, as welt as tn the distal sequences. The recombination event shown represents exchange between a transferred linear DNA segment introduced into a ceil by transduction or conjugation arid the badenal chromosome. The actual DNA segments participating may be much longer, for example, phage transduc tion often delivers an approximately BO kb segment of DNA. The f. coli chromosome is approximately 5 Mb.

KccA Protein Assembles on Single-Stranded DNA and Promotes Strand Invasion

RecA is the central protein in homologous recombination. It is the founding memljer of a family of enzymes called strand-exchange proteins. These proteins catalyze the pairing of homologous DNA molecules. Fairing involves both the search for sequence matches between two molecules and (he generation of regions of base pairing between these molecules.

The DNA pairing and strand-exchange activities of RecA can be observed using simple DNA substrates in vitro; examples of DNA pairing and strand-exchange reactions useful for demonstrating the biochemical activities of RecA are shown in Figure 10-7. The important features of these DNA molecules are: (1) DNA sequence complementarity between the two partner molecules; (2) a region of single-stranded DNA on at least one molecule to allow RecA assembly; and (3) the presence of a DNA end within the region of complementarity, enabling the UNA strands in the newly-formed duplex to intertwine.

The active form of RecA is a protein-DNA filament (Figure 10-8). Linlike most proteins involved in molecular biology, that function in smaller discrete protein units, such as monomers, dimers, or hexamers, the RecA filament is huge and variable in size; filaments that contain approximately 100 subunits of RecA and 300 nucleotides of UNA are common. The filament can accommodate one, two, three, or even lour strands of DNA. As described below, filaments with either one or three bound strands are most common in recombination intermediates.

The structure of DNA within the filament is highly extended compared to either uncoateri ssDNA or a standard BTorm helix, On average, the distance between adjacent bases is 5 A rather than the 3.4 A spacing normally observed (Chapter 6). Thus, upon RecA binding, the length of a DNA molecule is extended approximately 1.5-fold (Figure lO.Ba). It is within this RecA-fi lament that the search for homologous DNA sequences is conducted and the exchange of DNA strands executed.

To form a filament, subunits of RecA bind cooperatively to DNA. Ret A binding and assembly are much more rapid on single-stranded than

FIGURE 10-7 Substrates for RecA strand exchange.

FIGURE 10-7 Substrates for RecA strand exchange.

0 0

Post a comment