Kinetic Studies On The Mechanisms Of RECQ Helicase

RecQ helicases are critical for the maintenance of genomic stability, which belongs toSF2. Defects in BLM, WRN and RecQ4will lead to serious human hereditary diseases:Bloom syndrome, Werner syndrome and Rothmund–Thomson syndrome, characterized bygenomic instability and a high predisposition to cancer. Over the past decade areas such ashuman RecQ helicases play roles in nucleic acid metabolism have been widely studied. Andthis will provide novel idea and probability to cancer treatment.RecQ5is one of the five human RecQ-family helicases. In human RecQ5protein existsin at least three different isoforms, namely RecQ5α, RecQ5β and RecQ5γ. RecQ5β has991amino acids, which has one helicase domain, one RQC domain, and a long non-homologousC-terminal region from other RecQ helicase. Interestingly, RecQ5β can not only unwindduplex DNA, but possesses DNA annealing activity. Although helicase are widely studied, thebiochemical properties and biological functions of RecQ5in human have yet to be determined.BLM contains1417amino acids, and exist as oligomeric structures in solution, the unwindingmechanism of which needs further study.The study may be useful for better understanding theaction mechanism catalyzed by RecQ helicases, as well as for understanding its biologicalfunctions.In this study, using stopped-flow method based on fluorescence resonance energytransfer, we have investigated systematically the DNA annealing and unwinding kineticproperties of RecQ5β, and the unwinding activity of BLM. The main results are as follows:1. We have systematically studied the annealing kinetics properties of RecQ5-catalyzed2nM two complementary45nt ssDNA substrates, at different enzyme concentrations, in theabsence or presence of0.1mM different nucleotide cofactors (ADP,ATPγS and AMPPNP).And by double-or triple-exponential fits of the data curves, we obtained the initial annealingrate at each enzyme concentration. It showed that:(a) There is an optimum enzymeconcentration, between10and20nM, for DNA annealing in each case;(b) All the maximuminitial annealing rates are quite similar for the four different cases, which seems to be incontradiction with the previous observation that ATP S and ADP may completely or partiallyinhibited the annealing activity of RecQ5.(c) RecQ5catalyzed DNA annealing efficiently in wide concentration ranges for the cases of apo and ATP S states, and in a much narrowerrange for the case of ADP. We defined an efficient enzyme concentration range when efficientannealing occurs. They are5.6-27.7nM,9.8-24.1nM,4.9-33.8nM and11.5-19.3nM,respectively, for apo,AMPPNP,ATP S and ADP.2. We have analyzed the equilibrium ssDNA binding kinetics of RecQ5by fluorescencepolarization anisotropy assay, with varying amounts of RecQ5adding to8nM3'F20-ntssDNA substrate, in the absence or presence of0.1mM different nucleotide cofactors (ADP,ATPγS and AMPPNP). It showed that, the anisotropy first increases with increasing enzymeconcentration, and then, at high enzyme concentrations, the anisotropy saturates. As theanisotropy reflects the extent to which the ssDNA is covered with bound enzyme molecules,the data curves indicate that the ssDNA is partially covered with the enzyme at lowconcentrations and then completely covered with the enzyme at high concentrations. From theefficient enzyme concentration range given previously and the equilibrium DNA binding datacurves, we found that efficient annealing occurs when the coverage of ssDNA by the enzymeis in the ranges of39.4%22.8%(apo),37.3%14.0%(AMPPNP),37.8%26.3%(ATP S)and49.1%9.9%(ADP), indicating that RecQ5-catalyzed DNA annealing proceedsoptimally at levels of enzyme sufficient to cover⁴0%-50%of the DNA strand, regardless ofthe nucleotide states.3. We have measured the dissociation kinetics of bound100nM RecQ5β helicase from10nM3'F20-nt ssDNA substrate by fluorescence polarization anisotropy assay, in theabsence or presence of0.1mM different nucleotide cofactors (ADP,ATPγS and AMPPNP). Itshowed that, the dissociation from this substrate of RecQ5exhibited similar behavioursunder conditions of apo,AMPPNP and ATP S states: occurring in two phases, a slow one anda fast one. In the case of ADP state, there is only one phase. By single-or double-exponentialfits of the dissociation kinetics data curves, we obtained the dissociation rates in the four cases(in the cases of apo,AMPPNP and ATP S, only the slow-phase rates are given). It showedthat, the helicase binds to the ssDNA substrate most tightly in ATP S state and most weakly inADP state. The helicase has similar affinities for ssDNA in the apo and AMPPNP states.4. We performed annealing kinetic assays using non-complementary F-and H-labeledssDNA substrates. It showed that in the cases of both apo and ATP S, the fluorescence signalchange for non-complementary substrates is much smaller than that for complementarysubstrates. This demonstrates that the fluorescence signal change in the assay not simplyresulted from the gathering of ssDNA molecules with the aid of RecQ5, but the real duplexformation. That is, what we have observed reflected the real ssDNA annealing properties ofRecQ5. 5. We performed annealing kinetic assays with another reaction buffer:20mMTris-acetate,pH7.9,50mM KOAc,10mM Mg(OAc)2,1mM DTTå’Œ50mg/ml BSA. Itshowed that ATP S indeed inhibited the annealing activity of RecQ5in the different reactionbuffer. This also demonstrates that our observations reflected the real DNA annealing.6. We have compared and analyzed the DNA annealing kinetics,equilibrium DNAbinding kinetics and dissociation kinetics of RecQ5β^1-662(the C-terminal deleted fragment ofRecQ5β), in apo state. The DNA annealing kinetics results showed that the maximum initialannealing rate of RecQ5β^1-662was significantly (about one order of magnitude) lower thanthat of RecQ5β full length. For RecQ5β^1-662, it was³%s^-1and RecQ5β full length was³0%s^-1. The efficient enzyme concentration range was obtained as4.6-16.3nM forRecQ51-662.The equilibrium DNA binding kinetics results showed that efficient annealingoccurs when the coverage of ssDNA by RecQ51-662is in the range of52.9%23.3%,indicating the RecQ51-662-catalyzed DNA annealing also proceeds optimally when thehelicase covers half of the DNA strand, just like the full-length enzyme. It would beinteresting for further studies to see if this observed feature is equally applicable to otherRecQ family helicases. The dissociation kinetics results showed that as in the case of RecQ5full length, the dissociation of RecQ51-662from ssDNA occurred in two phases, a slow oneand a fast one. The slow phase rate is0.450.04s^-1, almost the same as that of RecQ5fulllength in apo state (0.510.12s^-1). It is quite interesting to note that although RecQ5andRecQ51-662have almost the same affinity for ssDNA, the two enzymes have significantlydifferent DNA annealing efficiencies. This indicates that the annealing ability of certainhelicases such as the present RecQ5is not simply resulted from their ssDNA bindingbehaviour, but rather, is an intrinsic property of these helicases that should be determined bytheir specific structures. In the case RecQ5, the C-terminal region is indeed mainlyresponsible for its DNA annealing activity, as previously noted.7. We have chosen RecQ5β^1-467and RecQ5β^1-662(two N-terminal fragments of RecQ5β)to determine the single-turnover kinetic conditions of RecQ5β unwinding:2nM16-bp duplexwith different3'-ssDNA tails (10-50nt) is as the optimum substrate,37°C is as the optimumtemperature, the concentration of dT56is2μM, the concentration of ATP is1.5mM, theenzyme concentration is^-100nM, and for RecQ5β^1-662,2M DNA trap is needed for betterunwinding.8. Under the single-turnover kinetic conditions obtained above, we have systematicallystudied and compared the DNA unwinding kinetic properties of RecQ5β^1-467,RecQ5β^1-567andRecQ5β^1-662(three N-terminal fragments of RecQ5β) with DNA substrates of different3'-ssDNA tail lengths. It showed that, for each fragment, its unwinding amplitude and rate increased with the increase of the3'-tail length in the DNA substrate. The maximumamplitude was73.5%,57.6%and35.5%for RecQ5β^1-467, RecQ5β^1-567and RecQ5β^1-662,respectively. It indicated that for a substrate with certain length of3'-tail, the unwindingamplitude of RecQ5β was associated with the length of the fragment, namely, it decreasedwith the increase of the length. In addition, for each of the three RecQ5β fragments, when the3'-tail length of the DNA substrates was shorten, one slow unwinding process occurred andwith its increase, the unwinding amplitude of the fast unwinding process increased obviously.RecQ5β-catalyzed DNA unwinding depended on the3'-tail length, that the competence ofRecQ5β-catalyzed DNA unwinding depended on the3'-tail length of the DNA substrate. Itindicates that RecQ5β molecules are cooperative in DNA unwinding. This is a surprisingreport in the helicase SF2, which is different from that of E.coli RecQ helicase.9. The N-terminal467amino acids is enough for enabling RecQ5β to unwind DNA, andthe remaining amino acid sequence can inhibit the unwinding activity of RecQ5β. RecQ5βitself has a certain regulatory mechanism to control the activity. The N-terminal467aminoacids contains a complete functional structure of the helicase domain, which is an importantinspiration for RecQ5domain partition.10. We have studied the dissociation kinetic behaviors of BLM from DNA substrate indifferent nucleotide states. It found that the low processivity of BLM is mainly resulted fromits high dissociation rate from the substrate when it is in ADP·Pi and ADP states,3.9and4.8s1respectively. That BLM tends to detach from the DNA substrate in ADP·Pi and ADP states,thus it can not unwind efficiently.