Regulation of Myosin by Phosphorylation of the Heavy Chain

AMII重链磷酸化的重要性的第一个证据是我们在20世纪80年代初的数据，表明其肌动蛋白激活的镁ATPase活性被两个重链的29个残基的非螺旋尾段上的1到3个丝氨酸残基的磷酸化所抑制。从20世纪80年代中期到90年代中期，我们花了大约10年的时间研究了这种抑制的可能机制。一个一致的结果是，非磷酸化(活性)肌球蛋白和磷酸化(非活性)肌球蛋白的杂聚体的肌动蛋白激活的ATPase活性明显低于同等的均聚物混合物，即当两者在同一细丝中时，AMII的调节似乎处于细丝水平，而磷酸化的AMII失活非磷酸化的AMII。有趣的是，尾巴C末端的磷酸化位点与球状结构中的ATPase位点相距80 nm，中间由螺旋螺旋隔开。当时，无论是我们还是其他人，都没有任何证据表明非螺旋尾段的磷酸化影响了细丝结构。 我们通过研究重组野生型和突变型AMII的微丝结构和肌动蛋白激活的ATPase活性来重新研究这个问题。我们现在通过定量质谱学发现，与肌球蛋白II重链激酶孵育后，全长AMII的非螺旋尾段(NHT)中的四个丝氨酸和全长AMII运动域的第二环中的Ser639、重肌球蛋白(HMM)和亚片段1(S1)被磷酸化。所有肌球蛋白结构的肌动蛋白激活的ATPase都被磷酸化抑制，由于HMM和S1没有肌球蛋白尾巴，因此肌动蛋白激活的ATPase活性的抑制必须源于Ser639的磷酸化。与这一结论一致的是，全长AMII的S639A突变体不受NHT-丝氨酸的磷酸化或其突变为谷氨酸的抑制。相反，S1和全长AMII的S639D突变体在酶活性上都是无效的，与全长AMII的NHT-丝氨酸的磷酸化状态无关。对重组野生型和S639D S1的肌球蛋白和肌动球蛋白ATPase循环中每一步的动力学分析表明，肌球蛋白ADP-PI复合体中PI的释放是S639D突变主要抑制的步骤。 虽然非螺旋尾段中丝氨酸的磷酸化不会影响酶的活性，但我们发现它确实影响了AMII微丝的结构。如旋转阴影细丝的电子显微镜所见，全长AMII及其NHT-丝氨酸被磷酸化或突变为谷氨酸的双极四聚体、六聚体和八聚体微丝与未磷酸化的、未突变的AMII相比，裸区更长，头部更紧密，与Ser639的磷酸化状态无关。因此，我们已经证明，AMII的肌动蛋白激活的ATPase活性是通过运动区的Ser639的磷酸化来调节的，AMII的细丝结构是通过NHT中四个丝氨酸中的一个或多个的磷酸化来改变的。这些结果对哺乳动物非肌肉肌球蛋白II的微丝结构的调控具有重要意义，其NHT-丝氨酸和苏氨酸在体外和体内都可以被磷酸化。

The first evidence for the importance of heavy chain phosphorylation of AMII was our data in the early 1980s showing that its actin-activated MgATPase activity was inhibited by phosphorylation of 1 to 3 serine residues in the 29-residue non-helical tailpiece of each of the two heavy chains. For a period of about 10 years from the mid-1980s to mid-1990s, we had investigated the possible mechanism of this inhibition. The one consistent result was that heteropolymers of unphosphorylated (active) myosin and phosphorylated (inactive) myosin had significantly less actin-activated ATPase activity than equivalent mixtures of homopolymers, i.e. regulation of AMII seemed to be at the level of filaments with phosphorylated AMII inactivating unphosphorylated AMII when both are in the same filaments. Interestingly, the phosphorylation sites at the C-terminal end of the tail are 80-nm distant from the ATPase site in the globular separated by a coiled-coil helix. At that time, neither we nor others had any evidence that phosphorylation of the non-helical tailpiece affected the filament structure. We have reinvestigated this problem by studying the filament structure and actin-activated ATPase activity of recombinant wild-type and mutant AMIIs. We now find, by quantitative mass spectrometry, that incubation with myosin II heavy chain kinase phosphorylates four serines in the non-helical tailpiece (NHT) of full-length AMII, and Ser639 in loop 2 of the motor domain of full-length AMII, heavy meromyosin (HMM) and sub-fragment 1 (S1). The actin-activated ATPase of all of the myosin constructs is inhibited by phosphorylaion and, since HMM and S1 lack the myosin tail, inhibition of actin-activaed ATPase activity must result from phosphorylation of Ser639. Consistent with this conclusion, the S639A mutant of full-length AMII is not inhibited by phosphorylation of the NHT-serines or their mutation to glutamate. Conversely, S639D mutants of both S1 and full-length AMII are enzymatically inactive irrespective of the phosphorylation state of the NHT-serines of full-length AMII. Kinetic analysis of each step in the myosin and actomyosin ATPase cycles of recombinant wild-type and S639D S1 showed that the release of Pi from the actomyosin ADP-Pi complex is the step predominantly inhibited by the S639D mutation. Although phosphorylation of the serines in the non-helical tailpiece does not affect enzymatic activity, we find it does affect the structure of the AMII minifilaments. As seen by electron microscopy of rotary shadowed filaments, bipolar tetrameric, hexameric and octameric minifilaments of full-length AMII with their NHT-serines either phosphorylated or mutated to glutamate have longer bare zones and more tightly clustered heads than non-phosphorylated, non-mutated AMII irrespective of the phosphorylation state of Ser639. Thus, we have shown tha the actin-activated ATPase activity of AMII is regulated by phosphorylation of Ser639 in the motor domain and the filament structure of AMII is modified by phosphorylation of one or more of the four serines in the NHT. These results have significant implications for the regulation of the filament structure of mammalian non-muscle myosins IIs whose NHT-serines and threonines can also be phosphorylated both in vitro and in vivo.