鼠脑驱动蛋白ATP水解机制的晶体学分析

鼠脑驱动蛋白是一类利用ATP水解释放的能量在微管系统上高连续性运动的常规驱动蛋白。了解ATP水解的化学能如何转化为机械动能是驱动蛋白研究中的重大课题。为此,鼠脑驱动蛋白单体(rK354)的晶体通过浸泡的方式引入ATP的结构类似物AMPPNP。rK354-AMPPNP复合物和rK354-ADP复合物结构的比较,揭示了开关区域Ⅱ的Glu237起连接ATP的γ-磷酸和驱动蛋白微管结合区的枢纽作用。     

鼠脑驱动蛋白的表达、纯化和结晶

驱动蛋白是一类利用水解ATP为ADP和磷酸的过程中释放的能量沿微管系统运动的蛋白。为了研究ATP中储存的化学能是如何转化为驱动蛋白的机械动能,鼠脑驱动蛋白的相关N-端区域在BL21-Codon Plus(DE3)-RP感受态大肠杆菌细胞中大量地表达。通过SP-强阳离子交换色谱和分子筛色谱的两步骤纯化,蛋白最终产量高达10 mg/L细胞培养液,蛋白纯度可以达到95%以上。纯化的蛋白具有水解ATP酶的活力,并与驱动蛋白抗体有特异性的反应。驱动蛋白可以在如下条件结晶:1.7 mol/L(NH4)2SO4,500 mmol/L NaCl,20%glycerol。晶体衍射的分辨率可以达到2.0。     

热休克蛋白90的N端与ATP类似物的晶体结构揭示其功能调控

分子伴侣热休克蛋白90(Hsp90)对于许多涉及细胞周期调控、信号转导以及细胞生长调控蛋白质的折叠、成熟及稳定是必需的．Hsp90的N端结构高度保守,包含一个ATP结合口袋并具有ATP酶活性,Hsp90的功能依赖于ATP与Hsp90结合后诱导的构象重排及之后的ATP水解．为了深入研究ATP与Hsp90结合后N端的结构及其功能状态,使用悬滴法共结晶了Hsp90的N端与ATP类似物AMPPNP及ATPγS的复合物,并利用分子置换法对其结构进行了解析．两个复合物晶体结构都捕获到了核苷酸的电子密度,尤其是γ-磷酸的电子密度,从而观察到γ-磷酸与蛋白质之间的相互作用．ATPγS中γ-磷酸的捕获证实了之前报道的结构中没有捕获到γ-磷酸是其处于无序状态而非被水解．单体状态下的人源Hsp90^N- AMPPNP与处于二聚体化的酵母Hsp90-AMPPNP结构对比可见S1和ATP lid的位置有明显区别,结构分析表明,E18-K100和N40-D127之间形成的氢键相互作用,在一定程度上阻碍了S1和ATP lid的摆动,很可能阻止了二聚体的形成．     

驱动蛋白的结构与功能

驱动蛋白是一类蛋白质超家族的总称,其中驱动蛋白-1(以下简称驱动蛋白)是目前已知的有机体内最小的马达蛋白.驱动蛋白能够催化三磷酸腺苷(adenosine triphosphate,ATP)分子的水解反应,将贮藏在ATP中的化学能转变为自身机械运动所需的机械能.驱动蛋白能够沿着微管连续定向运动,在细胞的有丝分裂和胞内物质运输中发挥重要作用.在真核细胞中,驱动蛋白主要以二聚体的形式存在,其结构主要包括4个部分,即马达头部、茎部、连接头部与茎部的颈链以及与"货物"相结合的尾部.驱动蛋白二聚体独特的结构特征以及各个组成部分协调的构象变化,保证了其沿微管的连续行走.目前,驱动蛋白的结构与功能之间的关系的研究取得了重要的进展.随着实验和计算水平的不断提高,彻底了解驱动蛋白的运动机理已经为期不远了.     

分子马达KIF1A的研究进展

驱动蛋白家族成员1A(kinesin family member 1A,KIF1A)属于向微管正向端移动的驱动蛋白第三家族,它能够利用三磷酸腺苷(adenosine triphosphate,ATP)水解释放的能量实现沿微管定向运动。KIF1A是轴突末端的突触囊泡体沿微管输运的重要载体,其马达结构域的突变将导致多种与神经有关的疾病和缺陷。文中主要综述了近年来KIF1A有关的生命过程和疾病的研究进展,介绍了KIF1A催化ATP水解反应的各中间态结构,同时基于这些结构信息,阐述了KIF1A的运动形式、核苷酸轮换机制和运动机理,并对今后的研究前景进行了展望,旨为KIF1A相关研究提供思路。     

驱动蛋白结构与运动机制

驱动蛋白是一类能够利用ATP水解释放的化学能驱动其所携带的“货物”分子沿着微管（microtubule,MT）定向运动的分子马达,在细胞器运输、有丝分裂、轴突运输等方面有着重要的生理作用。随着驱动蛋白结合ADP、ATP和未结合核苷酸（APO）三种特征状态的晶体结构的解析,驱动蛋白构象变化的研究得到了进一步发展,而在力产生机制和运动模型方面仍然存在较大争议。本文以kinesin-1家族为例,分析了驱动蛋白三种特征状态结构的特点、状态结构间的构象转变,论述了驱动蛋白的力产生机制和整个迈步过程。并探讨了驱动蛋白的运动模型,同时采用分子动力学模拟比较了驱动蛋白的两种迈步方式,为深入研究驱动蛋白提供了一定的理论计算。最后,基于本课题组对复杂体系的研究,对驱动蛋白体系的控制机制提出了新的假设,并对未来的研究方向进行了展望。     

驱动蛋白的能量转换过程——从ATP分子结合到颈链对接

驱动蛋白是一种分子马达,同时也是一种核苷酸酶,它能够将ATP分子所携带的化学能转化为其沿微管蛋白行走的机械能,每消耗一个ATP分子行走一步。对于驱动蛋白如何将ATP的化学能转化为构象变化的机械能的研究一直是生物物理学研究的热点问题。本文从3个方面对此问题的研究进展进行了综述:ATP分子与驱动蛋白结合; ATP结合引起驱动蛋白头部产生转动;驱动蛋白头部转动引起驱动蛋白颈链向头部的对接。将这三个方面的内容合并起来就构成了驱动蛋白的能量传递路径。     

Hsp70蛋白自身磷酸化对其分子伴侣功能的影响

近年对分子伴侣蛋白Hsp70作用机制的研究发现,其ATP功能区域X光晶体结构有一个新的钙离子结合区域,这个新的功能区域与Hsp70分子的ADP结合、ATP水解及合成有关.有报道认为Hsp70蛋白的NDP激酶样作用,通过形成酸不稳定性自身磷酸化中间体催化γ 磷酸基团在ATP和ADP间传递,组氨酸H89与这个新的区域有密切关系,有可能与Hsp70蛋白形成自身磷酸化中间体有关.本研究运用基因定位诱导突变技术,将89位组氨酸以丝氨酸替代(H89S),通过比较Hsp70野生型及突变型蛋白的自身磷酸化过程的改变,及其对Hsp70蛋白体外荧光素酶活性影响的不同,初步探讨Hsp70作用机制.结果发现,突变的H89S蛋白自身磷酸化过程及体外变性荧光素酶重折叠受到抑制.野生型蛋白未受到影响,野生型Hsp70可以形成酸不稳定的自身磷酸化中间体,产生CDP依赖性解磷酸反应,而H89S突变型蛋白不能形成这种反应.89位组氨酸点突变能显著降低ATP酶交换反应及体外变性荧光素酶重折叠水平,但它的自身磷酸化可能并非唯一必需的介导位点或只是一个选择性的功能侧链.     

利用荧光素酶生物发光体系测定Mg~(2+)-ATP酶结合和水解活性

本文应用LKB公司的ATP测液建立了Mg~(2 )-ATP酶的ATP结合及水解活性的测定方法;利用国产荧光素酶粗品在连串反应体系中建立测定Mg~(2 )-ATP酶结合活性的方法,并与水解活性相比较.对Mg~(2 )-ATP酶的去脂样品,Mg~(2 )-ATP酶与卵磷脂复合物以及微粒体样所做的测定表明,上述两种方法是可靠、简便的,尤其是利用国产荧光素酶粗品建立的ATP结合活性的测定方法,能避免水解对结合活性测定的干扰,刘其它的酶-底物的结合研究有参考价值.     

转运蛋白ABCG1/4 和ABCA1对脑胆固醇的调节作用

脑是富含胆固醇的器官,机体大约有25%的胆固醇集中在脑组织中.ATP结合盒超家族转运蛋白对脑组织中胆固醇的膜外转运和动态平衡起着重要的调节作用.研究发现,ATP结合盒超家族转运蛋白亚体ABCG1、ABCG4和ABCA1在成体脑组织中存在不同程度的表达,一种或多种亚体的缺失可以导致神经退行性病变.然而,ATP结合盒超家族转运蛋白亚体对脑发育过程中脑胆固醇动态变化的调节缺乏相关性的报道.在本研究中,从低胆固醇饮食喂养的C57BL/6J小鼠中获取出生后不同发育时期的脑组织,对ABCG1、ABCG4和ABCA1的mRNA与蛋白质表达水平进行测定,并对脑组织和血清中ATP结合盒超家族转运蛋白的表达水平与胆固醇水平的相关性进行研究.同时,使用ABCG1、ABCG4单一基因敲除鼠和ABCG1、ABCG4双基因敲除鼠,研究ATP结合盒超家族转运蛋白对与胆固醇合成的相关基因表达的影响以及对脑组织胆固醇代谢的调节作用.结果发现,ABCG1、ABCG4和ABCA1在机体多个器官中均有表达,但ABCG1和ABCG4在小鼠脑组织中表达量最高.在脑组织发育过程中,ABCG1和ABCG4mRNA水平呈现明显的表达时效性,小鼠于出生后42天达到峰值,而ABCA1 mRNA的表达水平无明显变化.血清和脑组织中中酯化型胆固醇水平呈双高峰分布,也于出生后42天达到最高.基因敲除鼠模型显示,单一敲除ABCG1或者ABCG4基因对脑组织胆固醇水平无明显影响,而ABCG1和ABCG4基因的同时缺失导致脑胆固醇水平显著升高,并明显降低胆固醇合成相关基因的表达水平.本研究表明,在脑发育成熟过程中,ATP结合盒超家族转运蛋白亚体ABCG1和ABCG4,而非ABCA1,以调节脑胆固醇的膜外转运;ABCG1和ABCG4互补调控脑胆固醇的动态平衡.     

Kinesin and myosin ATPases: variations on a theme.

The enzymes kinesin and myosin are examples of molecular motors which couple ATP hydrolysis to directed movement of biological structures. Myosin has been extensively studied and its structure and mechanism of coupling are known in detail. Much less is known about kinesin, but many of its major properties are similar to those of myosin. Both enzymes have two catalytic head groups at the end of a long alpha-helical rod. The head groups contain the sites for ATP hydrolysis and interaction with their respective partners for movement (microtubules or F-actin). In each case the binding and hydrolysis of ATP is rapid and the steady state ATPase rate is limited by a slow step in the region of product release. This slow release of product is accelerated by interaction with actin or microtubules coupled to changes in binding affinity. As there is no evidence for a close evolutionary link between kinesin and myosin, these and other similarities may represent convergence to set of common functional properties which are constrained by the requirements of protein structure and the use of ATP hydrolysis as a source of energy. It will be of particular interest to determine if these common properties are also shared by the large number of divergent proteins which have recently been discovered to possess a domain which is homologous to the head group of kinesin.     

Conformational change of the loop L5 in rice kinesin motor domain induced by nucleotide binding

Loop L5 of kinesin is located near the ATPase site, in common with kinesins of various animal species. The rice plant-specific kinesin K16 also has a corresponding loop that is slightly shorter than that of mouse brain kinesin. The present study was designed to monitor conformational changes in loop L5 during ATP hydrolysis. For this purpose, we introduced one reactive cysteine into the L5 of rice kinesin and modified it with fluorescent probes. The cysteine in L5 was labeled with a fluorescent probe 2-(4'(iodoacetamide) anilino-naphthalene-6-sulfonic acid sodium salt) [IAANS]. IAANS was incorporated into L5 at an almost equimolar ratio in the absence of nucleotides. In contrast, the incorporated amount was reduced to 0.62 and 0.32 mol IAANS/mol motor domain in the presence of ATP and ADP, respectively. Upon nucleotide addition, the fluorescent intensity of IAANS incorporated into L5 was significantly reduced to 63% and 51% for ATP and ADP, respectively. These results suggest that L5 of rice kinesin significantly changes its conformation during ATP hydrolysis.     

Kinesin as an Electrostatic Machine

Kinesin and related motor proteins utilize ATP fuel to propel themselves along the external surface of microtubules in a processive and directional fashion. We show that the observed step-like motion is possible through time-varying charge distributions furnished by the ATP hydrolysis cycle while the static charge configuration on the microtubule provides the guide for motion. Thus, while the chemical hydrolysis energy induces appropriate local conformational changes, the motor translational energy is fundamentally electrostatic. Numerical simulations of the mechanical equations of motion show that processivity and directionality are direct consequences of the ATP-dependent electrostatic interaction between the different charge distributions of kinesin and the microtubule.     

Alternating site ATPase pathway of rat conventional kinesin

The pathway of ATP hydrolysis by rat kinesin was established by pre-steady-state kinetic methods. A 406-residue long N-terminal fragment was shown by sedimentation equilibrium analysis to form a dimer with a K(d) of 46 nm. The pathway of ATP hydrolysis follows the Gilbert-Johnson pathway determined previously for a similarsized N-terminal fragment of Drosophila conventional kinesin. However, the rates of ADP release were at least 3-fold faster, and ATP hydrolysis was approximately 5-fold faster. Paralleling our previous mechanistic data, these results support an alternating site ATPase pathway, including a captive head state as an intermediate in the kinesin ATPase cycle. The kinetic data presented in this report once again point to the importance of the captive head state and argue against a pathway that short-circuits this key intermediate. In addition, several unique aspects of the rat kinesin kinetics reveal new aspects of the ATPase-coupling mechanism. These studies provide a baseline set of kinetic parameters against which future studies of rat kinesin mutants may be evaluated and directly correlated with the structure of the dimeric kinesin.     

Kinesin and cytoplasmic dynein binding to brain microsomes.

Movement of cellular organelles in a directional manner along polar microtubules is driven by the motor proteins, kinesin and cytoplasmic dynein. The binding of these proteins to a microsomal fraction from embryonic chicken brain is investigated here. Both motors exhibit saturation binding to the vesicles, and proteolysis of vesicle membrane proteins abolishes binding. The maximal binding for kinesin is 12 +/- 1.7 and 43 +/- 2 pmol per mg of vesicle protein with or without 1 mM ATP, respectively. The maximal binding for cytoplasmic dynein is 55 +/- 3.8 and 73 +/- 3.7 pmol per mg of vesicle protein with or without ATP, respectively. These values correspond to 1-6 sites per vesicle of 100-nm diameter. The nonhydrolyzable ATP analog, adenyl-5'-yl imidodiphosphate (AMP-PNP), inhibited kinesin binding to vesicles but increased kinesin binding to microtubules. An antibody to the kinesin light chain also inhibited vesicle binding to kinesin. In the absence but not presence of ATP, competition between the two motors for binding was observed. We suggest that there are two distinguishable binding sites for kinesin and cytoplasmic dynein on these organelles in the presence of ATP and a shared site in the absence of ATP.     

Nucleotide-free kinesin hydrolyzes ATP with burst kinetics

Bovine brain kinesin binds ADP tightly and contains a stoichiometric amount of ADP at its active site when isolated in the presence of free Mg2+ (Hackney, D. D. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 6314-6318). EDTA in excess of Mg2+ weakens ADP binding and nucleotide-free kinesin can be prepared by gel filtration with excess EDTA. On addition of ATP, this nucleotide-free enzyme catalyzes the rapid hydrolysis of a stoichiometric amount of ATP in a burst phase followed by much slower continued ATP hydrolysis limited by the release of ADP from the active site. This burst reaction is evident both by formation of [32P]Pi from [gamma-32P]ATP and by formation of [alpha-32P]ADP from [alpha-32P]ATP. At 1.1 nM kinesin active sites, the observed rate of the burst phase increases linearly with ATP over the 1-20 nM range yielding a bimolecular rate of net ATP binding and hydrolysis of 2.5 microM-1 s-1. The intercept at zero ATP is 0.008 s-1 which equals the ADP release rate at 0.008-0.009 s-1. This predicts a Km for ATP of approximately 3.5 nM and measurements of the dependence on ATP concentration of the steady state rate and amount of bound ADP are consistent with a Km of this magnitude.     

Chemomechanical coupling of the forward and backward steps of single kinesin molecules

The molecular motor kinesin travels processively along a microtubule in a stepwise manner. Here we have studied the chemomechanical coupling of the hydrolysis of ATP to the mechanical work of kinesin by analysing the individual stepwise movements according to the directionality of the movements. Kinesin molecules move primarily in the forward direction and only occasionally in the backward direction. The hydrolysis of a single ATP molecule is coupled to either the forward or the backward movement. This bidirectional movement is well described by a model of Brownian motion assuming an asymmetric potential of activation energy. Thus, the stepwise movement along the microtubule is most probably due to Brownian motion that is biased towards the forward direction by chemical energy stored in ATP molecules.     

Kinesin's walk: Springy or gated head coordination?

Conventional kinesin (kinesin-1) is a motor protein that performs a vital function in the eukaryotic cell: it actively transports cargo to required destinations. Kinesin pulls cargo along microtubule tracks using twin linked motor domains (heads) that bind the microtubule, hydrolyse ATP, and alternately step forward. The detail of the kinesin walk has yet to be discovered but a prominent theory is that the mechanism is rectified Brownian motion (RBM) biased by linker zippering. There is evidence that an ATP binding gate coordinates the heads. The hypothesis proposed here is that the gate is unnecessary, that entropic linker strain is sufficient to enable procession. An agent-based computer simulation has been devised to explore head coordination in the RBM model. Walking was found to emerge in silico without a gate to synchronise the heads. Further investigation of the model by applying a range of hindering loads resulted in backstepping or detachment with similar characteristics to behaviour observed in vitro. It is unclear whether kinesin waits at an obstacle but adding an ATP hydrolysis gate to the model in order to force waiting resulted in the model behaving less realistically under load. It is argued here that an RBM model free of gating is a good candidate for explaining kinesin procession.