The hexameric helicases are ubiquitous proteins (related enzymes are found on both sides of bacterial-archaean divide) involved in unwinding double-stranded DNA and RNA. They bind to the single stranded form, and "walk" along it, forcing an unwinding at the bifurcation point of the double strand. Molecular weight studies of the active enzyme, and examination of the subunit content shows that the active forms are hexameric. Electron microscopic imaging of helicases from sources ranging from viruses to eukaryotes shows a symmetrical hexameric structure with a central hole. The single strand of DNA or RNA is threaded through the hole. Not all enzymes with helicase activity are hexameric; - the hexameric helicases are one class. In the hexameric class, the association formed is dependent on cofactors, - nucleotides and Mg2+, - and trimeric complexes can also be formed. However, the active enzyme is hexameric. Helicases are often found associated in multienzyme complexes, but the helicase isolated from these complexes retains activity and hexameric association under active conditions. The hexamers can be readily dissociated and reassociated. Functional studies using mixed hexamers, with one or more copies being inactive mutant forms, suggest that all six are required for function. If one of the 6 monomers is specifically inhibited, for example by covalent binding of 8-azido-ATP, function is lost.
The hexameric helicases are NTP hydrolases. The nucleotide triphosphate favored varies between types, and also with the substrate polynucleotide. We will assume that ATP is used in the following discussion. In the helicase reaction, ATP hydrolysis accompanies movement along the DNA. It had earlier been suggested from rapid kinetic studies that ATP hydrolysis at three of the nucleotide binding sites was orders of magnitude slower that the overall rate of catalysis (kcat). On this basis, it appeared that only three of the monomers were involved in binding to the DNA, and in catalytic turnover in ATP hydrolysis, and that the binding of alternate subunits in the hexameric ring might distorts the interfaces between adjacent monomers so that alternating sites are in a non-catalytic configuration. In this interpretation, at one of the three rapid sites, 1 ATP/hexamer is hydrolyzed faster than kcat, suggesting that the three are not equivalent. My colleague Maria Spies has pointed out to me that more recent work has called for a reinterpretation of this mechanism. To paraphrase her points, the presteady-state kinetic data that show different rates of hydrolysis reflect the first (or an isolated) ATP hydrolysis event. An ATP molecule bound in the ATPase site of one monomer does indeed result in the conformational change that would preclude ATP hydrolysis in the next site. This has two consequences: (1) ATP hydrolysis and the power-stroke propagates in a circle; and (2) ATP titration results in a peculiar dependence: high concentrations of ATP typically inhibit hexameric helicases. However, this does not mean there are intrinsic difference between adjacent subunits. If one looks at multiple ATP hydrolysis cycles, as soon as ATP is hydrolyzed in subunit 1, subunit 2 becomes capable of hydrolyzing ATP (2). In addition, not all hexameric helicases (SF3, SF4 and SF5), but only "Family 5" or Rho-like helicase, show an appropriate parallel with the F1-ATPase. In contrast to other families, the Rho-like enzymes possess an ATP binding/hydrolysis fold, which is the most similar to F1-ATPases. F1-ATPase, however, "evolved" to be "smarter" by having 3 alpha and 3 beta subunits.
Since ssDNA is itself asymmetric in cross-section, if a sequential mechanism occurs, each subunit involved in binding will see a different face of the threaded strand to the next. The picture that emerges from kinetic studies is of a mechanism in which the catalytic subunits are involved in a sequential mechanism in which they change configuration to match different binding faces. Local changes in affinity for a periodic pattern along the DNA, with transient unbinding and rebinding of each subunit, would “force” a progress along the DNA. These reactions are mechanistically linked to ATP hydrolysis, which provides the driving force.
As Maria also notes, the evolutionary perspective shows a chain of related ATPases: RecA as a simple and typical ATPase containing a P-loop (very conserved, very ubiquitous, as good of a marker for evolution distances as ribosomal RNAs), and might have evolved to a hexameric helicases (also ubiquitous but diverged into different families based on the slight differences in ATPase fold and other signature motifs), then to a Rho-type helicase, which alternates between closed ring and open (almost like RecA) conformations, has a P-loop like all other ATPases but looks very similar to a subunits of F1 (Q and R-loops). The parallel between the hexameric helicases and the F1 ATPase is quite striking. Both consist of hexameric rings, but the F1 ring is composed of 3 each of alternating a and b subunits. However, the sequence similarity and homologous tertiary structure of these two subunits strongly suggest that both evolved from a common ancestor, indicating that the ancestral structure was a homohexameric structure. The a and b subunits also show sequence and structural similarity to the Rho-type hexameric helicase monomers, suggesting that they all share a common ancestry.
The a and b subunits form a ring around a centrally located g subunit, which interacts asymmetrically with the three b subunits. In the Walker structure, this asymmetry of interaction is associated with an asymmetry of occupation of the three catalytic sites on the b subunits. One contains the ATP analogue AMP-PNP, one ADP, and one is empty. The a subunits all contain nucleotide, - the ATP analogue AMP-PNP. The binding sites are more or less symmetrically located at the interfaces between a and b subunits, with “regulatory” a subunit and catalytic b subunit sites alternating around the hexameric ring. The complete structures show distinct binding sites along the g subunit, which interact differentially with the interfacial contact spans of each b subunit. Since the contact sites are on three faces reflecting the pseudo-C3 symmetry, the sequential changes in configuration allow the hexameric ring to rotate the g subunit by “walking” along it!
From the above description, it will be apparent that the F1 ATPase mechanism is, in essentials, the same as the helicase mechanism. In view of the sequence and structural similarity, it seems quite likely therefore that the a, b-ring evolved from an ancestral homohexameric structure shared in common with the helicases. If the a, b-ring evolved, then the enzyme evolved.
It would be going beyond the evidence to claim that we understand the evolution of the ATP synthase. It must have happened before the archaeal and bacterial lines separated, in a process for which we have only the modern forms, and the tools of genomics to provide a guide. We can however claim that the complexity of the modern forms is not an indication that they were designed, or created in their present forms. They evolved from simpler precursors as would be expected from an evolutionary origin.
Dale Wigley has a nice site comparing the mechanisms of the two enzymes. This includes movies based on the structure of the T7 helicase (3).
1. Patel, S. S. and Picha, K. M. (2000) Structure and function of hexameric helicases. Annu. Rev. Biochem. 69, 651-697
2. Liao, J.C., Jeong, Y.J., Kim, D.E., Patel, S.S. and Oster, G.(2005) Mechanochemistry of t7 DNA helicase. J. Mol. Biol. 350, 452-75
3. Singleton, M.R., Sawaya, M.R., Ellenberger, T. and Wigley, D.B. (2000). Crystal structure of T7 gene 4 ring helicase indicates a mechanism for sequential hydrolysis of nucleotides. Cell 101, 589-600.