A plant calmodulin-binding motor is part kinesin and part myosin

Kinesins and myosins are molecular motors that move on microtubules and actin filaments, respectively. These motor proteins are involved in a variety of cellular processes such as intracellular transport, spindle formation and chromosome segregation in eukaryotes (Langford, 1995; Hirokawa, 1998). Although kinesins and myosins hydrolyze ATP to generate force for their movement on cytoskeletal filaments and have similar structural organization (head/motor, stalk and tail) there is virtually no amino acid (aa) sequence identity between the microtubuleand actin-based motors (Kull et al., 1996). We report here that a novel calmodulin-binding protein from plants has domains that are present in microtubuleand actin-based motors, suggesting that this class of plant motors is unique in having kinesin and myosin features. In a protein–protein interaction based screening with calmodulin we recently isolated a cDNA from Arabidopsis encoding a novel calmodulin (CaM)-binding protein, KCBP, with a kinesin-like motor domain at the C-terminus of the protein (Reddy et al., 1996). Motility studies with the C-terminal kinesin-like motor domain alone have shown that it is a minus-end directed microtubule motor (Song et al., 1997). KCBP is unique among all known kinesins and kinesin-like proteins in having a CaMbinding domain (CBD). The homolog of KCBP has been cloned from three other plant systems, suggesting that it is ubiquitous in flowering plants (Narasimhulu and Reddy, 1998). However, a KCBP homolog has not been found in yeast and Caenorhabditis elegans, whose genomes have been completely sequenced, or in any other non-plant systems. In our ongoing analysis of KCBP function, we analyzed KCBP sequence using BLASTP and SMART programs (http://coot.embl-heidelberg.de/SMART) and identified, in addition to kinesin-like motor domain at the C-terminus, a region that is conserved in the tails of some members of myosins (Figure 1). The C-terminal region (amino acids 860–1217) of KCBP showed similarities with the motor region of a large number of microtubulebased motors of which KIFC3, a C-terminal motor, showed the highest sequence similarity (43% identity and 56% similarity, Figure 1A). The tail of myosin VIIa has two long repeats (about ∼460 aa per repeat), each containing a MyTH4 (myosin tail homology 4) domain (∼110 aa) and a talin-like (∼350 aa) domain. Although the MyTH4 domain and the talin-like regions exist singly in some myosins (IV and XII) and in the band 4.1 superfamily (examples include talin and ERM proteins), respectively, the presence of these two regions together (MyTH4+talin-like) occurs only in myosin VIIa (as two repeats) and myosin X (one repeat) (Chen et al., 1996). The N-terminal tail (aa 121–612) of KCBP showed significant similarities to the MyTH4 and talin-like (band 4.1) regions present in myosins VIIa and X. The MyTH4 and the talin-like domains of KCBP showed the highest similarity with the myosin VIIa tail (29% identity and 45% similarity in the MyTH4 domain and 23% identity and 37% similarity in the talin-like region) (Figure 1B). The MyTH4 domain and talin-like region have not been found in any other known members of the kinesin superfamily. Therefore, we conclude that the KCBP is a molecular hybrid consisting of a motor domain from microtubule-based motors and a tail region of actin-based motors (Figure 1C). The significance of MyTH4 and talin-like domains in KCBP and myosins is not known at this time, but the existence of talin-like region together with MyTH4 domain in KCBP, myosin VIIa and X is interesting and points to some functional significance. Based on what is known about KCBP and myosin VIIa, we propose that these domains may be involved in one or more of the following functions. One possibility is that the KCBP and myosin VIIa-like proteins could interact with some unknown common protein(s) through their tail homology regions to either cross-bridge microtubule and actin filaments or facilitate cargo exchange between these two types of molecular motors. An elegant study by Lille and Brown (1992) showed that a lethal mutation resulting from a defect in a myosin motor could be complemented by a kinesin motor protein which also suggests that actinand microtubule-motors are likely to interact with some common set of proteins. Furthermore,