REGULATION OF MYOSINS AND MYOSIN KINASES (PAKs)

曲霉肌球蛋白I的生化特性：构巢曲霉的单个肌球蛋白I在其重链（myoA）的TEDS位点含有一个Ser残基。myoA无效突变体的致死性通过表达野生型重链或其中该Ser被Ala或Glu替换的突变体来逆转。因此，要么这种肌球蛋白是肌球蛋白的肌动蛋白依赖性MgATP酶活性需要在TEDS位点处的Glu、Asp或磷酸化Ser或Thr的规则的例外，要么曲霉肌球蛋白I的基本功能不依赖于其催化活性。我们现在已经共表达了曲霉钙调蛋白（曲霉肌球蛋白I的轻链）与野生型重链或突变重链，其中Glu或Ala在TEDS位点取代Ser。野生型的肌动蛋白依赖性MgATPase活性为10/s，Glu突变体为3/s，Ala突变体没有或非常低的肌动蛋白依赖性MgATPase活性（所有三种酶具有相等的K/EDTA-ATPase活性）。因此，曲霉肌球蛋白I的基本功能似乎需要很少的肌动蛋白依赖性MgATPase活性。正在进行的研究表明，野生型肌球蛋白中的TEDS位点Ser在表达过程中被磷酸化，这可以解释其非常高的肌动蛋白依赖性MgATP酶活性。棘阿米巴肌球蛋白I重链激酶（MIHCK）（PAK家族成员）调节区的定位：在蛋白质水平上，只有少数PAK家族成员的特征得以确定。MIHCK是一种79 kDa的蛋白质，具有35 kDa的C-末端PAK样催化结构域，其在所有PAK中是保守的，并且具有p21结合结构域，其存在于大多数PAK中，在其调节N-末端区域内。MIHCK和其他PAK之间没有其他序列同源性区域。MIHCK活性通过多个位点（包括催化结构域中的一个）的自磷酸化而上调。就像哺乳动物的α-、β-和γ-PAK一样。MIHCK的自磷酸化由脂质和Rac或Cdc 42刺激。与哺乳动物PAK不同，MIHCK的自磷酸化需要酸性脂质，而p21仅在酸性脂质存在下激活。Ca-钙调蛋白结合MIHCK并消除脂质活化。我们以前提出，MIHCK是保持在其非活性状态的相互作用，其N-末端抑制区与其C-末端催化结构域。通过表达MIHCK的片段，我们现在已经将自抑制区域映射到p21结合位点，残基91-157，并将钙调蛋白结合位点映射到先前的碱性区域，残基51-79。含有残基51-157的片段抑制天然的、未磷酸化的MIHCK的活性，但对自磷酸化的、完全活性的MIHCK的活性没有影响，这表明这些片段抑制自磷酸化。棘阿米巴肌球蛋白I重链激酶（MIHCK）催化结构域转染HeLa细胞和NIH 3 T3细胞对粘着斑复合物的影响：我们用GFP偶联的MIHCK催化结构域瞬时转染HeLa细胞和NIH 3 T3细胞，MIHCK是一种活性远高于哺乳动物PAK的PAK。通过荧光显微镜观察，转染细胞中肌动蛋白或肌球蛋白II的分布没有明显变化。然而，通过黏着斑蛋白、talin和桩蛋白抗体染色确定，局灶性粘连的分布受到显著影响。用活性激酶转染的细胞的粘附复合物主要位于细胞外周，而粘附复合物均匀地分布在用两个催化失活的激酶突变体或单独的GFP转染的细胞中。与这些观察结果一致，用活性MIHCK催化结构域转染显著增加了失去与基质接触的细胞的分数。脱落细胞大小与对照细胞相同，但形态不同，既无坏死也无凋亡。用哺乳动物PAK的催化结构域转染HeLa细胞没有引起表型的实质性变化。这些观察结果表明，粘附复合物的重新分配的MIHCK催化结构域与其高催化活性。我们现在正试图确定MIHCK的底物，可能是负责这些影响。棘阿米巴肌球蛋白II棒替代Dictyostelium肌球蛋白II棒的影响：棘阿米巴肌球蛋白II的肌动蛋白依赖性MgATPase活性受棒末端三个丝氨酸残基磷酸化的负调节。为了研究N-末端头部和C-末端尾部之间的通信的结构基础，我们构建了三个嵌合肌球蛋白重链，其中网囊藻肌球蛋白II头部与野生型阿米巴肌球蛋白II棒或突变棒融合，其中三个Ser残基被替换为Asp或Ala。嵌合肌球蛋白在网囊藻肌球蛋白II无效细胞中表达，所述细胞在悬浮培养中既不分裂也不分化成固体基质上的子实体。如前所示，对于表达网囊藻肌球蛋白II头和骨骼肌球蛋白杆的嵌合体的细胞，用三种嵌合肌球蛋白转染的细胞以及用网囊藻野生型肌球蛋白II转化的细胞在悬浮培养物中生长。肌动蛋白依赖的MgATPase活性的三个嵌合体肌球蛋白约10倍高于野生型网囊藻肌球蛋白II和调节轻链磷酸化只有轻微激活。正在进行的实验表明，野生型嵌合体肌球蛋白II的阿米巴杆状体的磷酸化不会抑制其肌动蛋白依赖性MgATP酶活性。

BIOCHEMMICAL CHARACTERIZATION OF ASPERGILLUS MYOSIN I: The single myosin I of Aspergillus nidulans contains a Ser residue at the TEDS-site of its heavy chain (myoA). The lethality of the myoA null mutant is reversed by expression of wildtype heavy chain or mutants in which this Ser is replaced by Ala or Glu. Therefore, either this myosin is an exception to the rule that the actin-dependent MgATPase activity of myosins requires Glu, Asp or phosphorylated Ser or Thr at the TEDS-site or the essential function(s)of Aspergillus myosin I do not depend on its catalytic activity. We have now co-expressed Aspergillus calmodulin (the light chain of Aspergillus myosin I) with either wildtype heavy chain or mutant heavy chains in which Glu or Ala replace Ser at the TEDS-site. The actin-dependent MgATPase activity of the wildtype was 10/s, of the Glu-mutant 3/s, and the Ala mutant had no, or very low, actin-dependent MgATPase activity (all three enzymes had equal K/EDTA-ATPase activity). Thus, the essential function(s) of Aspergillus myosin I seem to require little, if any, actin-dependent MgATPase activity. Research in progress suggests that the TEDS-site Ser in the wildtype myosin is phosphorylated during expression which would explain its very high actin-dependent MgATPase activity. LOCALIZATION OF THE REGULATORY REGIONS OF ACANTHAMOEBA MYOSIN I HEAVY CHAIN KINASE (MIHCK), A MEMBER OF THE PAK FAMILY: Only a few members of the growing PAK family have been characterized at the protein level. MIHCK is a 79-kDa protein with a 35-kDa C-terminal PAK-like catalytic domain, that is conserved in all PAKs, and a p21-binding domain, that is present in most PAKs, within its regulatory N-terminal region. There are no other regions of sequence homology between MIHCK and other PAKs. MIHCK activity is up-regulated by autophosphorylation of multiple sites including one in the catalytic domain. Like mammalian alpha-, beta- and gamma-PAK. autophosphorylation of MIHCK is stimulated by lipids and Rac or Cdc42. Unlike mammalian PAKs, autophosphorylation of MIHCK requires acidic lipids, and p21s activate only in the presence of acidic lipids. Ca-calmodulin binds to MIHCK and eliminates lipid activation. We had previously proposed that MIHCK is maintained in its inactive state by interaction of its N-terminal inhibitory region with its C-terminal catalytic domain. By expressing fragments of MIHCK, we have now mapped the autoinhibitory region to the p21-binding site, residues 91-157, and mapped the calmodulin-binding site to the preceding basic region, residues 51-79. Fragments that contain residues 51-157 inhibit the activity of native, unphosphorylated MIHCK but have no effect on the activity of autophosphorylated, fully active MIHCK, suggesting that these fragments inhibit autophosphorylation. TRANSFECTION OF HELA CELLS AND NIH 3T3 CELLS BY THE CATALYTIC DOMAIN OF ACANTHAMOEBA MYOSIN I HEAVY CHAIN KINASE (MIHCK) AFFECTS FOCAL ADHESIION COMPLEXES: We transiently transfected HeLa and NIH 3T3 cells with GFP-coupled catalytic domain of MIHCK, a PAK whose activity is much higher than that reported for mammalian PAKs. By fluorescence microscopy, there were no significant changes in the distribution of actin or myosin II in the transfected cells. However, as determined by staining with antibodies against vinculin, talin and paxillin, the distribution of focal adhesions was significantly affected. Adhesion complexes of cells transfected with active kinase were localized mostly in the cell periphery whereas adhesion complexes were uniformly distributed in cells transfected with two catalytically inactive kinase mutants or GFP alone. Consistent with these observations, transfection with active MIHCK catalytic domain substantially increased the fraction of cells that lost contact with the substratum. The detached cells were neither necrotic nor apoptotic and were the same size as control cells, but with different shape. Transfection of HeLa cells with the catalytic domain of mammalian PAK caused no substantial change in phenotype. These observations suggest that the redistribution of adhesion complexes by the MIHCK catalytic domain are associated with its high catalytic activity. We are now attempting to identify the substrates of MIHCK that may be responsible for these effects. EFFECT OF SUBSTITUTION OF THE ACANTHAMOEBA MYOSIN II ROD FOR THE DICTYOSTELIUM MYOSIN II ROD: The actin-dependent MgATPase activity of Acanthamoeba myosin II is negatively regulated by phosphorylation of three serine residues at the end of rod. To investigate the structural basis of the communication between the N-terminal head and C-terminal tail, we constructed three chimeric myosin heavy chains in which the Dictyostelium myosin II head was fused with either wild type Acanthamoeba myosin II rod or mutant rods with the three Ser residues replaced by either Asp or Ala. The chimeric myosins were expressed in Dictyostelium myosin II null cells that neither divide in suspension culture nor differentiate to fruiting bodies on solid substratum. As shown previously for cells expressing a chimera of Dictyostelium myosin II head and skeletal myosin rod, cells transfected with the three chimeric myosins grew in suspension culture as well as cells transformed with Dictyostelium wildtype myosin II. The actin-dependant MgATPase activities of the three chimeric myosins were about 10-fold higher than wildtype Dictyostelium myosin II and only slightly activated by regulatory light chain phosphorylation. Experiments in progress indicate that phosphorylation of the Acanthamoeba rod of the wildtype chimeric myosin II does not inhibit its actin-dependent MgATPase activity.