Replicating And Repairing The Genome: From Basic Mechanisms To Modern Genetic Technologies
by one to two orders of magnitude (Figure 2.2). This activity is therefore referred to as “proofreading exonuclease.” As the “exo” in the name implies, exonucleases attack DNA from one of the two ends (from outside). The proofreading exonuclease in T7 and other replicative polymerases is a 3′ to 5′ exonuclease, because it attacks DNA from the 3′ end, which of course is the end of the growing strand found at the polymerase active site.
Figure 2.2.Excision of misincorporated residue by DNA polymerase. Replicative DNA polymerases generally have a 3′ to 5′ exonuclease activity that removes the vast majority of misincorporated residues. In the case of T7 DNA polymerase, the exonuclease activity resides in the same protein chain as the DNA polymerase activity, but the two active sites are some distance from each other. Upon misincorporation, the growing 3′ end is displaced from the polymerase active site to the exonuclease active site, where the 3′ terminal residue is excised (as a monophosphate). Synthesis resumes when the corrected 3′ end returns to the polymerase active site.
Notice in the polymerase structure (Figure 2.1A) that the exonuclease active site is far removed from the polymerase active site, and indeed the DNA primer-template within the complex must dramatically rearrange in order that a 3′ end relocates to the exonuclease active site and gets clipped. How does the enzyme favor exonuclease action when an incorrect base is incorporated? Incorporation of an incorrect base will invariably cause mispairing and thereby destabilize the double helix in its vicinity. For example, the end of the primer would become destabilized and fray when the rare tautomeric form of the incorporated base described above reverts to its normal form. Two forces then favor removal of the incorrect base. The exonuclease active site prefers a single-stranded end rather than one in a duplex primer-template, and the polymerase active site is inhibited from adding an additional base by a mispaired 3′ terminus in the primer-template (Figure 2.2). In this way, a misincorporated base has a high likelihood of ending up in the exonuclease active site, and once the terminal mismatched base is removed, the (correctly paired) duplex primer-template reforms and is favored to migrate back to the polymerase active site.
An important property of DNA polymerases, called processivity, is the ability to extend a primer (add a nucleotide residue) repeatedly without dissociating from the primer-template. A processive polymerase extends repeatedly, while a distributive polymerase incorporates a small number of nucleotide residues before dissociating and switching to a different primer-template. Processivity increases the rate of DNA replication because it eliminates the time-consuming process of recruiting a new polymerase whenever the replicating polymerase dissociates. It turns out that the processivity of all replicative polymerases is modulated extensively in order to carry out an efficient and concerted replication reaction. The most obvious need for modulating processivity is to accommodate the very different length requirements for replication on the leading versus the lagging strand.
The T7 DNA polymerase devoid of thioredoxin has quite a low processivity, incorporating only about 10–15 nucleotides before dissociating. However, the addition of thioredoxin dramatically increases processivity to an average of about 800 nucleotides per binding event. Concomitant with the increased processivity, the structure of the region of polymerase that binds thioredoxin is reorganized, and the polymerase complex interacts more extensively with the primer-template. Thioredoxin itself is situated over the duplex portion of the primer-template in the structure of the complex (Figure 2.1A). One model for the increased processivity conferred by thioredoxin is that the thumb domain with thioredoxin folds down over the duplex portion of the primer-template after nucleotide addition, essentially encircling the duplex region transiently. This mechanism would be somewhat analogous to that used in cellular DNA replication, where specialized proteins called sliding clamps tether the DNA polymerase to its template (see Chapters 3 and 4).
Although the increase in processivity conferred by thioredoxin is impressive, it is not sufficient to account for the rapid copying of the 40,000-base pair viral genome. Indeed, the processivity of T7 DNA polymerase is further enhanced by the helicase/primase hexamer when the proteins are appropriately arranged for leading-strand synthesis (Figure 2.3, also see below). In this case, DNA polymerase extending the leading strand continues for approximately 5000 nucleotides, on average, before dissociating from the nascent 3′ end. Remarkably, however, this dissociation does not significantly impede replication. There is another DNA polymerase binding site on the helicase/primase,2 and the dissociating polymerase can transiently bind to this site and then quickly return to the leading-strand 3′ end. In all, the helicase/primase increases DNA polymerase processivity to greater than 17,000 nucleotides, likely sufficient to copy the entire T7 genome.
Figure 2.3.T7 replication fork, the “traditional” view. The T7 replication fork is depicted with two DNA polymerase molecules (blue), the combined helicase/primase (orange) and the ssDNA-binding protein (green). The direction of helicase movement is indicated by the dotted arrow.
2.4Mechanisms of unwinding by the T7 helicase
The two major domains of the T7 helicase/primase protein have distinct functions — one catalyzes the unwinding of the parental DNA duplex (helicase) and the other provides RNA primers for lagging-strand synthesis of Okazaki fragments (primase). As we see in Chapters 3 and 4, cellular replication systems utilize two distinct proteins for these two functions, although even in those cases the two proteins interact functionally and communicate with each other.
Focusing first on DNA unwinding activity, T7 helicase translocates in the 5′ to 3′ direction along a single strand of DNA, unwinding any duplex region that is encountered during this translocation. Recent structural studies support an attractive “hand-over-hand” model for this translocation. Helicase complexes with ssDNA were resolved into multiple hexameric forms that resembled lock-washers rather than symmetrical rings. The image at the top left of Figure 2.4A shows one such hexamer, with the discontinuity of the lock-washer in between the subunits labeled HelA and HelF.3 The key structural transition in the model is that the subunit on the 5′ side of the discontinuity moves up to the 3′ end of the lock-washer, as the discontinuity shifts to the next interface (between HelE and HelF in the first step; Figures 2.4A and B). Given the 5′ to 3′ direction of movement along the DNA, the HelF subunit is essentially moving from the back to the front of the lock-washer. This movement comprises approximately 24 angstroms (10−10 meters) in the 5′ to 3′ direction. Long-range movement of the helicase along the DNA is accomplished by the sequential movement of individual subunits from the back to the front of the lock-washer, as the discontinuity essentially rotates around the ring (Figure 2.4B). Note that the structure at the end of six cycles of subunit movement is identical to the structure at the start, except that the hexamer has advanced along the ssDNA.
Figure 2.4.Movement and loading of the T7 replicative helicase. Recent structural studies strongly support a “hand-over-hand” mechanism for translocation of the T7 helicase along ssDNA (Gao et al., 2019). In panel A, closely related helicase-DNA structures illustrate key features of the mechanism (see text for detailed discussion). Gao et al. (2019) refer to the six subunits