Kinesin-5 motor proteins perform essential mitotic functions by providing the force that separates the spindle poles apart during spindle assembly, maintenance, and elongation reviewed in Goulet and Moores, 2013; Kapoor, 2017; Mann and Wadsworth, 2019; Scholey et al., 2016; Singh et al., 2018; Waitzman and Rice, 2014. These motors are homotetramers, with two pairs of catalytic domains located on opposite sides of a central minifilament (Acar et al., 2013; Gordon and Roof, 1999; Kashina et al., 1996; Scholey et al., 2014). This unique structure enables kinesin-5 motors to crosslink the antiparallel microtubules (MTs) of the spindle and to slide them apart by moving in the plus-end direction of the two MTs that they crosslink (Kapitein et al., 2005; Shapira et al., 2017; Shimamoto et al., 2015). Because of the MT architecture within the mitotic spindles, plus-end directed motility of kinesin-5 motors is essential to separate the spindle poles (Goulet and Moores, 2013; Mann and Wadsworth, 2019; Singh et al., 2018). Moreover, since the catalytic domains of kinesin-5 motors are located at their N-termini, it was previously believed that such N-terminal motors are exclusively plus-end directed. Recently, however, several studies have demonstrated independently in single-molecule motility assays that three kinesin-5 motors – the Saccharomyces cerevisiae Cin8 and Kip1 and the Schizosaccharomyces pombe Cut7 – move processively toward the minus-end of the MTs and switch directionality under different experimental conditions (Britto et al., 2016; Düselder et al., 2015; Edamatsu, 2014; Fallesen et al., 2017; Fridman et al., 2013; Gerson-Gurwitz et al., 2011; Roostalu et al., 2011; Shapira and Gheber, 2016). The above-described body of work notwithstanding, the molecular mechanism and physiological implications of such bi-directional motility remain elusive.

The majority of kinesin motors share a mechanical structural element of 10–18 amino acids, termed the neck linker (NL), which connects the two catalytic domains of the kinesin dimers, enabling them to step on the same MT. For N-terminal plus-end directed kinesins, the NL is located at the C-terminal end of the catalytic domain, between the last helix of the catalytic domain (α6) and the neck helix (α7) that is required for dimerization (Kozielski et al., 1997; Vale and Fletterick, 1997). It has been proposed that during ‘hand-over-hand’ stepping in the plus-end direction, the NL isomerizes between a disordered ‘undocked’ conformation in the presence of ADP and an ordered, motor-domain-bound ‘docked’ conformation in the presence of ATP, with the docked conformation pointing toward the plus-end of the MT (Gigant et al., 2013; Goulet et al., 2014; Rice et al., 1999; Rosenfeld et al., 2001; Tomishige et al., 2006). It is believed that the docked conformation is stabilized by N-terminal sequences upstream of the motor domain, termed the cover strand (CS), which form a β-sheet with the first half of the docked NL, termed β9 (Geng et al., 2014; Goulet et al., 2012; Hwang et al., 2008; Khalil et al., 2008; von Loeffelholz et al., 2019). Additional stabilization of the docked NL conformation is provided by an asparagine residue in the middle of the NL that is predicted to serve as a latch (N-latch), holding the docked NL along the core motor domain, which is conserved in processive plus-end directed kinesins (Hwang et al., 2008). The docked NL is also stabilized by an interaction between amino acids in the second half of the NL, termed β10, and amino acids in β7 and the loop 10 of the motor domain (Hwang et al., 2008). The importance of the isomerization of the NL between the docked and undocked conformations – as a force-generating transition in kinesin motors (Cao et al., 2014; Clancy et al., 2011; Gigant et al., 2013; Goulet et al., 2014; Shang et al., 2014) – is believed to lie in its role in coordinating nucleotide binding and hydrolysis (Clancy et al., 2011) and regulating motor velocity, processivity, and force production (Budaitis et al., 2019; Düselder et al., 2012; Higuchi and Endow, 2002; Hughes et al., 2012; Muretta et al., 2015; Muretta et al., 2018; Schief and Howard, 2001; Shastry and Hancock, 2011).

During plus-end directed stepping of pairs of motor domains connected via the NL, the transition to, and stabilization of, the docked NL conformation bring the trailing head forward in the plus-end direction and are thus critical for plus-end directed motility. Since bi-directional kinesin motors can move in both plus- and minus-end directions on the MTs, the dynamics of the NL isomerization in these kinesins has to allow stepping in both directions. Thus far, however, the role of the NL in regulating the motor functions of bi-directional kinesins has not been reported. To address this issue, we designed a series of NL variants of the bi-directional kinesin-5 motor, Cin8, and examined their functions in vivo and in vitro. We also examined the H-bond arrangement of the docked NL conformations of these NL variants. We showed that, among these variants, those that were partially active exhibited a smaller number of stabilizing H-bonds between the docked NL and motor domain, vs. wild type (wt) Cin8. In addition, we showed that elimination of a conserved backbone H-bond between a glycine in the N-latch position of Cin8 resulted in a non-functional variant, indicating that stabilization of this H-bond is critical for the functionality of Cin8. Partial replacement of the NL of Cin8 with homologous sequences from the NLs of plus-end directed kinesins resulted in the generation of non-functional Cin8 variants that could neither move in the minus-end direction nor crosslink MTs in vitro. In one such variant, containing sequences from the NL of the plus-end directed vertebrate kinesin-5 Eg5, the number of stabilizing H-bonds between the N-latch asparagine and the motor domain was larger than that in wt Cin8, indicating that additional stabilization of the N-latch position is incompatible with the functionality of Cin8. In this variant, a single replacement of the conserved N-latch asparagine with glycine, as originally present in Cin8, decreased the number of N-latch stabilizing H-bonds and rescued the majority of the defects of the non-functional variant, both in vivo and in vitro. Thus, we propose that exact H-bond stabilization that allows a certain degree of flexibility in NL docking is critical for the functionality of the bi-directional kinesin-5 Cin8.

To examine the role of the NL in regulating the functionality of Cin8, we generated a series of variants in which some of the amino acids of the Cin8 NL sequence were replaced with homologous sequences from other kinesin motors (Figure 1A and B). Based on amino acid alignment, we created Cin8 variants containing NL sequences from two exclusively plus-end directed motors, human kinesin 1 KHC (designated here as Cin8_NLKHC) and Xenopus laevis kinesin-5 Eg5 (Cin8_NLEg5) or sequences from the bi-directional S. pombe kinesin-5 Cut7 (Cin8_NLCut7). In addition, we created Cin8 NL variants in which single amino acids were replaced with amino acids from the above-mentioned kinesin motors (Figure 1B). These variants were examined in a series of in vivo and in vitro assays with the aim to study the effect of the above-described replacements on the functionality of Cin8.

Figure 1 with 1 supplement see all

Download asset Open asset

Cin8 variants containing NL sequences from plus-end directed kinesin motors are not functional in cells

To assess the intracellular functionality of the NL variants, we first examined their ability to support yeast viability as the sole source of kinesin-5 function. Since at least one of the two S. cerevisiae kinesin-5 motors, Cin8 and Kip1, is essential for cell viability (Roof et al., 1992; Saunders and Hoyt, 1992), the NL variants were examined on a low-copy centromeric plasmid in cells carrying chromosomal deletions of CIN8 and KIP1, following the shuffling-out of the Cin8 plasmid that covered the double deletion (Figure 1C; Avunie-Masala et al., 2011; Goldstein et al., 2019; Goldstein et al., 2017). We found that NL variants carrying a single amino acid replacement at positions where the consensus of most bi-directional kinesin-5 motors differs from the consensus of the exclusively plus-end directed kinesin-5 motors (variants Cin8-K516M, Cin8-Q520E, Cin8-M526T, and Cin8-D528K) were mostly viable (Figure 1C), thereby indicating these replacements do not significantly affect the functionality of Cin8 in vivo.

In contrast to the single amino acid replacements, Cin8 variants containing partial NL sequences from exclusively plus-end directed kinesin motors (Cin8_NLKHC and Cin8_NLEg5) failed to grow at all the examined temperatures (Figure 1C), thereby indicating that these variants were not functional in cells. In addition, cells deleted for Cin8 (but with functional Kip1) and expressing the Cin8_NLEg5 variant exhibited significantly longer doubling times than cells expressing wt Cin8 (Table 1), as is consistent with the impaired functionality of this variant. Importantly, cells expressing the Cin8 variant with the NL sequence of the bi-directional Cut7 (Cin8_NLCut7) were viable at 23°C and 26°C, but not at temperatures above 30°C (Figure 1C), thereby indicating that at least some of the Cin8 functions are preserved in the Cin8 variant containing the NL sequence from a different bi-directional kinesin. The doubling time of cells expressing the Cin8_NLCut7 variant was significantly longer than that of cells expressing wt Cin8 in both the presence and absence of Kip1 (Table 1), which is consistent with the partial functionality of this variant. Moreover, in cells with functional Kip1, variants containing the NL sequence from the plus-end directed Eg5 and bi-directional Cut7 accumulated with monopolar spindles (Figure 2A, B), indicating that both variants were defective in the spindle assembly function in S. cerevisiae cells.

Table 1

	cin8Δkip1Δa†	cin8Δa
wt Cin8	152 ± 1 (3)	127 ± 4 (3)
Cin8-G522N	175 ± 2 (3)*	n.d.
Cin8NLCut7	210 ± 7 (3)**	143 ± 3 (4)*
Cin8NLEg5-NG	191 ± 6 (4)**	n.d.
Cin8NLEg5	n.d.	167 ± 4 (4)**

1. ^aAverage doubling times ± SEM (min). The number of experiments is shown in parentheses (see also the Source data 1). 3HA-tagged Cin8 variants were expressed in the kip1Δcin8Δ and cin8Δ strains.

   ^†In the cin8Δkip1Δ strain, NL variants and wt Cin8 were examined following shuffling-out of the parental pMA1208 plasmid (see Materials and methods).

2. *p < 0.05, **p < 0.01, compared to wt Cin8.

Table 1—source data 1

Source data for Table 1 .

https://cdn.elifesciences.org/articles/71036/elife-71036-table1-data1-v2.xlsx

Download elife-71036-table1-data1-v2.xlsx

Figure 2

Download asset Open asset

Figure 2—source data 1

Cell cycle analysis of cells expressing NL variants of Cin8.

https://cdn.elifesciences.org/articles/71036/elife-71036-fig2-data1-v2.xlsx

Download elife-71036-fig2-data1-v2.xlsx

Figure 2—source data 2

Area of localization of NL variants of Cin8 at the poles in cells with monopolar spindles.

https://cdn.elifesciences.org/articles/71036/elife-71036-fig2-data2-v2.xlsx

Download elife-71036-fig2-data2-v2.xlsx

To elucidate which function of Cin8 is maintained in the partially viable Cin8_NLCut7, but not in the non-viable Cin8_NLEg5 variant, we visualized the cellular localization of these variants, tagged with 3GFP, in cells containing functional Kip1 and bearing a tdTomato-tagged spindle-pole-body (SPB) protein Spc42 (Spc42-tdTomato) (Fridman et al., 2013; Goldstein et al., 2017). Prior to bipolar spindle formation, when the two SPBs had not yet separated, wt Cin8 and Cin8_NLCut7 concentrated near the SPBs, at the minus-ends of nuclear MTs (Figure 2C–E). This localization pattern is consistent with the minus-end directed motility of these variants on the nuclear MTs (Figure 2D; Shapira et al., 2016). In contrast, Cin8_NLEg5 exhibited diffusive localization in the nucleus (Figure 2C–E). Quantitative analysis in cells with pre-assembled spindles did indeed indicate that Cin8_NLEg5 occupied a significantly larger area than wt Cin8 and Cin8_NLCut7 (see Materials and methods) (Figure 2E), indicating that localization of this variant is significantly more diffusive in the nucleus, probably due to reduced affinity to nuclear MTs. In addition to the diffusive nuclear localization and in contrast to wt Cin8 and Cin8_NLCut7, Cin8_NLEg5 also exhibited residual attachment to the nuclear MTs (Figure 2C, white arrow). Such localization, similar to the previously reported pattern of the tailless Cin8 mutant that had lost its minus-end directionality preference (Düselder et al., 2015; Shapira et al., 2016) suggests that the minus-end directed motility is impaired in the Cin8_NLEg5 variant, although it is maintained in the partially functional Cin8_NLCut7. This is probably the reason for the ability of Cin8_NLCut7, but not Cin8_NLEg5, to support yeast viability.

In summary, the above experiments indicate that replacement of the Cin8 NL sequence with the sequence from plus-end directed kinesin motors produced variants that are not functional in cells, probably due to decreased affinity to MTs and abolished minus-end directed motility on nuclear MTs prior to spindle assembly. In contrast, replacement of the Cin8 NL sequence with a sequence from the bi-directional Cut7 probably maintained minus-end directed motility, resulting in a partially functional variant of Cin8 that can support cell viability as a sole source of kinesin-5.

NL variants of Cin8 exhibit a different H-bond arrangement in the docked orientation

To seek an explanation for the differences in activity of the NL variants of Cin8, and in particular, for the finding that replacing the N-latch asparagine with glycine rescued the defects of the Cin8_NLEg5 variant, but not the Cin8_NLCut7 variant, we examined the structural configuration of the NL docking by using homology modeling. To this end, we generated structural models of the variants in a nucleotide-bound state in which the NL is docked by exploiting four known structures of kinesin motors published previously, all in the presence of adenylyl-imidodiphosphate (AMP-PNP) (Figure 3 and Figure 3—figure supplement 1, and Table 2) (see Materials and methods). In the modeling, we focused on the H-bond array between the NL and the motor domain (Figure 3 and Figure 3—figure supplement 1, dashed lines). In this context, we note that it has previously been suggested that the non-motor N-terminus of the plus-end directed kinesin-5 Eg5 forms a β-sheet cover strand bundle (CSB) with β9 of the docked NL, which stabilizes the docked NL configuration (Goulet et al., 2012; Goulet et al., 2014), similarly to reports for kinesin-1 motors (Hwang et al., 2008; Khalil et al., 2008). A recent cryo-EM study of MT-bound S. pombe Cut7 indicates that a similar CSB formation may also take place during NL docking in bi-directional kinesin-5 motors (von Loeffelholz et al., 2019), despite the fact that the non-motor N-terminal region is considerably longer in bi-directional kinesin-5 motors compared to exclusively plus-end directed kinesin-5 and kinesin-1 motors (Singh et al., 2018). Thus, it is likely that CSB formation stabilizes NL docking in Cin8 and may affect NL dynamics. Indeed, our modeling revealed that β9 of the NLs of all our Cin8 variants forms two H-bonds with the non-motor N-terminal residues, L73 and I75 (Figure 3—figure supplement 1, red dashed lines), supporting the notion that the N-terminal CS of Cin8 stabilizes the docked conformation of the NL. However, since a large part of the N-terminal non-motor sequence is absent in our models (due to a lack of structural data for this region), at this stage we cannot estimate accurately to what extent these extended non-motor N-terminal sequences contribute to the stabilization of NL docking. Therefore, the other H-bonds between β9 and the motor domain present in our models (Figure 3—figure supplement 1, black dashed lines) are likely to be less informative. In addition, the formation of H-bonds in this region does not correlate with the function of Cin8. For example, wt Cin8 and Cin8_NLCut7 exhibit drastically different viability in cells (Figure 1C), but they form the same number of H-bonds between β9 and the motor domain (Figure 3—figure supplement 1, black dashed lines). Thus, we conclude that the differences in H-bond formation between β9 and the motor domain, as depicted by our modeling, cannot explain the differences in functionality between the NL variants of Cin8.

Figure 3 with 1 supplement see all

Download asset Open asset

Table 2

Protein	Residue i	Atom	Residue J	Atom
wt Cin8	F412	O	F524	N
	F412	O	S523	Oγ
	F524	N	S523	Oγ
	G157	O	G522	N
Cin8NLEg5	F412	O	K524	N
	G157	O	N522	Nδ
	G157	O	N522	N
	I414	O	N522	Nδ
Cin8-G522N	F412	O	F524	N
	G157	O	N522	N
Cin8NLEg5-NG	F412	O	K524	N
	G157	O	G522	N
Cin8NLCut7	G157	O	N522	N
Cin8NLCut7-NG	F412	O	L524	N

1. H-bonds are based on geometric criteria. Here, we used stringent criteria: the distance between the donor and acceptor heavy atoms must be <0.36 nm, and the acceptor-donor hydrogen angle must be <30°.

We next examined the H-bond formation between the N-latch position and β10 of the NL and the motor domain. According to our models, wt Cin8 forms a conserved backbone H-bond between G157 in a loop between α1 and β3 and the N-latch G522 in the NL (Hwang et al., 2008). In addition, wt Cin8 forms two H-bonds between F412 in β7 and S523 and F524 in the NL and an additional H-bond between S523 and F524 of the NL (Figure 3 and Table 2). This array of H-bonds was not recapitulated in any of the variants we examined, which might explain why none of these variants is able to exhibit the full function of Cin8 in cells (Figures 1 and 2 and Table 1). The two partially functional variants, Cin8-G522N and Cin8_NLEg5-NG, also form the conserved backbone H-bond between G522/N522 of the NL and G157 in a loop between α1 and β3. However, both these variants form only one H-bond between F245/K245 in the NL and F412 in β7. Finally, the Cin8_NLCut variant, which exhibits severely impaired activity in cells (Figures 1 and 2 and Table 1), forms only one backbone H-bond between N522 of the NL and G157 and completely lacks a H-bond between the NL and β7 (Figure 3 and Table 2). These results suggest that H-bond stabilization between NL and β7 is important, but not critical, for the function of Cin8.

Interestingly, the non-functional Cin8_NLEg5 variant also forms only one H-bond between F412 in β7 and K524 in the NL. However, according to our models, it forms a highly stabilized H-bond array involving the N-latch asparagine (N522) of the NL, which replaces the G522 of Cin8. In Cin8_NLEg5, similarly to wt Cin8, a conserved backbone H-bond is formed between G157 and N522. In addition, two H-bonds are formed between the amide nitrogen of N255, one with G157 and the other with I414 in β7 (Figure 3 and Table 2). Similarly to Cin8_NLEg5, the partially functional Cin8_NLEg5-NG variant also forms only one H-bond between F412 in β7 and K524 in the NL. However, in contrast to Cin8_NLEg5, it does not form the two stabilizing H-bonds between the amide nitrogen of N522, and G157 and I414. Since the Cin8_NLEg5-NG variant is partially functional in cells, while Cin8_NLEg5 is not (Figures 1 and 2 and Table 1), we propose that the extra stabilization of the NL during docking, provided by H-bonds of the amide group of N522, is one of the reasons for lack of functionality of Cin8_NLEg5 in cells. Finally, the non-functional Cin8_NLCut7-NG forms only one H-bond between F412 in β7 and L524 of the NL and is unable to form the highly conserved H-bond between G157 and G522 (Figure 3 and Table 2). This might lead to significant decrease of stabilization of the NL docking and to Cin8_NLCut7-NG to failing to perform essential functions in cells. Taken together, our data indicate that NL docking to the Cin8 core motor domain is governed by H-bonds formed in two regions, one between G522 in the N-latch position and G157 and the other between F412 in β7 and residues in positions 523 and 254 of the NL. Of these two regions, the exact stabilization of the H-bond formation at the N-latch position has a more significant impact on the function of Cin8 in cells.

Configuration of the H-bond during NL docking regulates the motile properties of Cin8 in vitro

To correlate between the intracellular phenotypes of the NL variants and their motor functions, we examined their activity in vitro. In these assays, we characterized the activity of GFP-tagged full-length NL variants, overexpressed and purified from S. cerevisiae cells, on fluorescently labeled GMPCPP-stabilized MTs (Pandey et al., 2021; Shapira et al., 2017). We first compared the MT affinities of the NL variants by determining the average number of MT-attached motors per MT length at equal motor protein concentrations. We found that all the NL variants exhibited significantly lower levels of MT-bound motors than wt Cin8 (Figure 4A), thereby indicating that the NL affects the MT-affinity of Cin8. Moreover, we also found that the majority of Cin8_NLEg5 and Cin8_NLCut7-NG variants, which were unable to support yeast cell viability as a sole source for kinesin-5, exhibited no motor motility at 140 mM KCl (not shown). We also examined the motility of Cin8_NLEg5 at a lower salt concentration, at which the affinity of the motors to MTs is increased (Shapira and Gheber, 2016). We found that under these conditions, Cin8_NLEg5 exhibited only bi-directional diffusive motility, with no minus-end directed bias, in contrast to wt Cin8, which exhibited processive minus-end directed motility (Figure 4B). These results indicate that, consistent with its intracellular localization (Figure 2C–E), the Cin8_NLEg5 variant exhibited very low MT affinity and was unable to move towards the minus-end of the MTs.

Figure 4 with 2 supplements see all

Download asset Open asset

Figure 4—source data 1

Affinity of NL variants of Cin8 to the MTs.

https://cdn.elifesciences.org/articles/71036/elife-71036-fig4-data1-v2.xlsx

Download elife-71036-fig4-data1-v2.xlsx

Figure 4—source data 2

MD analysis of NL variants of Cin8.

https://cdn.elifesciences.org/articles/71036/elife-71036-fig4-data2-v2.xlsx

Download elife-71036-fig4-data2-v2.xlsx

We then examined the motility of the functional Cin8 NL variants at a saturating ATP concentration and at a high ionic strength (140 mM KCl). We found that under these conditions, consistent with previous reports (Gerson-Gurwitz et al., 2011; Shapira et al., 2016; Shapira et al., 2017), all Cin8 NL variants moved processively in the minus-end direction of the MTs (Figure 4C). We have previously demonstrated that one of the factors that affects the directionality and velocity of Cin8 is its accumulation in clusters on MTs (Shapira et al., 2017). Thus, to examine the effects of mutations in the NL sequence on motility but not on motor clustering, we sorted out single molecules of Cin8 from a total population of moving Cin8 particles, based on their fluorescence intensities (Pandey et al., 2021) (see Materials and methods). By following the fluorescence intensity of Cin8 particles as a function of time, we observed single events of intensity decrease, of ~45 arbitrary intensity units (a.u.), most probably originating from the photobleaching of single GFP molecules (Figure 4—figure supplement 1A). Since single Cin8 motors are tetramers comprised of four identical subunits (Hildebrandt et al., 2006), the maximal fluorescence intensity of a single Cin8 molecule containing four GFP molecules is expected to be ∼180 a.u. The intensity distribution of the total Cin8-GFP population was consistent with this notion, in that it exhibited a major intensity peak containing ~65% of Cin8 particles with intensity <180 a.u. (Figure 4—figure supplement 1B). The maximal intensity of the peak was ~120 a.u., consistent with the average intensity of single Cin8 molecules containing one, two, three, or four fluorescent GFP molecules (Figure 4—figure supplement 1B). Based on this analysis (see Materials and methods), we defined ‘single Cin8 molecules’ as particles of intensity lower than 180 a.u. and analyzed the motile properties only of those Cin8 molecules, thereby ensuring analysis of mainly single molecules of tetrameric Cin8 NL-variants (Pandey et al., 2021).

We quantified the motility of single molecules of NL variants by tracing their positions on the MTs at each time point, followed by mean displacement (MD) analysis (see Materials and methods). Consistent with previous reports (Gerson-Gurwitz et al., 2011; Roostalu et al., 2011; Shapira and Gheber, 2016; Shapira et al., 2017), we found that under high ionic strength conditions, single molecules of wt Cin8 exhibited fast processive movements towards the minus ends of MTs, with a high average velocity of −318±4 nm/s (Figure 4C and D). Remarkably, in contrast to the diffusive motility of the Cin8 variant containing the NL sequence of the plus-end directed Eg5 (Cin8_NLEg5) (Figure 4B, orange arrows), mutating asparagine at position 522 to glycine (Cin8_NLEg5-NG) dramatically changed the motile behavior, inducing processive minus-end directed motility (Figure 4B–D). Although the average velocity of the Cin8_NLEg5-NG variant was lower than that of wt Cin8 (Figure 4D), the processive minus-end directed motility of this variant under the high ionic strength conditions indicates that glycine at position 522 in the NL modulates the minus-end directed motility of Cin8. Consistently, in wt Cin8, mutation of glycine in this position to asparagine resulted in slower minus-end directed motility, compared with that of the original wt Cin8 (Figure 4D). Furthermore, Cin8_NLCut7, the Cin8 variant with the NL sequence of the bi-directional Cut7 containing asparagine at position 522 (Figure 1C), also exhibited processive minus-end directed motility, but with reduced velocity compared with that of wt Cin8 (Figure 4D).

Finally, we also examined the run length of motile trajectories of single NL variants (Figure 4—figure supplement 2). We found that the average run length of the motile Cin8_NLCut7 was significantly shorter than that of wt Cin8. Among the viable NL variants, Cin8_NLCut7 exhibited the lowest ability to support cell viability as a sole source of kinesin-5 (Figure 1C). Thus, our results suggest that that the ability to produce processive minus-end directed motility is one of the motility traits that are important for the intracellular functioning of Cin8.

The NL of Cin8 regulates its MT-crosslinking

The intracellular functions of kinesin-5 motors are attributed to their ability to crosslink the interpolar MTs of the spindle (Goulet and Moores, 2013; Mann and Wadsworth, 2019; Singh et al., 2018). Thus, to establish additional correlations between intracellular phenotypes and the motor activity of the NL variants, we examined their MT-crosslinking activity in vitro. In this assay, purified Cin8 variants were mixed with fluorescently labeled GMPCPP-stabilized MTs in solution, followed by imaging and quantitation of the fluorescence intensity of the MTs (see Materials and methods). Consistent with previous reports (Gheber et al., 1999), we found that Cin8 induced the accumulation of MT bundles, thereby increasing the apparent fluorescence intensity of these bundles compared with that of single MTs (Figure 5). The MT-bundling activity of wt Cin8 was concentration dependent, inducing MT bundling at a minimal molar ratio of ~1:40 Cin8 motors to tubulin dimer (Figure 5D).

Figure 5

Download asset Open asset

Figure 5—source data 1

MT bundling by NL variants of Cin8.

https://cdn.elifesciences.org/articles/71036/elife-71036-fig5-data1-v2.xlsx

Download elife-71036-fig5-data1-v2.xlsx

With the exception of Cin8_NLEg5, all NL variants (Figure 1C) induced MT bundling in a concentration-dependent manner, with the fluorescence intensity of bundles induced by high concentrations of motors being significantly higher than that of MTs without the addition of motors (Figure 5D). Moreover, at high motor protein concentrations, the functional NL variants, Cin8-G522N, Cin8_NLCut7 and Cin8_NLEg5-NG, induced MT bundling to a significantly greater extent compared with the non-functional Cin8_NLEg5. This finding indicates that the MT-bundling activity of the NL variants is strongly correlated with the ability of these variants to support cell viability as a sole source of kinesin-5 (Figure 1C). However, all the functional NL variants exhibited reduced ability to bundle MTs, compared with wt Cin8, a finding that is consistent with the reduced affinity for MTs of these variants (Figure 4A). The Cin8-G255N variant exhibited reduced bundling ability at low and intermediate protein concentrations but reached the MT-bundling levels of wt Cin8 at high protein concentrations. In contrast, both the Cin8_NLCut7 and Cin8_NLEg5-NG variants exhibited reduced MT bundling compared with wt Cin8 at low and high motor concentrations, with Cin8_NLCut7 exhibiting better MT-bundling activity than Cin8_NLEg5-NG (Figure 5D). Finally, in contrast to the Cin8_NLEg5 variant, which failed to induce MT bundling, a single change of the asparagine at position 522 of Cin8_NLEg5 to glycine, produced a variant that induced MT bundling at high protein concentrations (Figure 5D), thereby indicating that glycine at position 522 is important for the MT-crosslinking and bundling activity of Cin8.

Mutations in the NL reduce the efficiency of plus-end directed antiparallel MT sliding

We have previously suggested that for fungal kinesin-5 motors to perform their spindle assembly function in closed mitosis, they need first to localize near the SPBs by their minus-end directed motility and then to reverse directionality to plus-end directed motility between antiparallel MTs in order to produce antiparallel MT sliding and separate the SPBs apart (Shapira et al., 2017). We therefore examined the ability of the NL variants to reverse directionality in an in vitro MT sliding assay, in which one set of MTs was immobilized to a glass surface, followed by the addition of Cin8 with an additional set of MTs (Figure 6) (see Materials and methods). By using polarity-marked MTs, both velocity and directionality of MT sliding could be determined for each NL variant. Consistently with previous results (Shapira et al., 2017), we found that in the presence of 125 mM KCl and ~3 ng/ml wt Cin8, MT sliding was observed. In more than 50% of cases, this sliding was plus-end directed, with the minus-end of the moving MT leading (Figure 6 and Figure 6—video 1). Under the same conditions, MT capturing or sliding was not observed in the presence of the NL variants, consistent with the lower affinity of the NL variants for the MTs (Figure 4A) and their reduced ability to bundle MTs (Figure 5). To increase the occurrence of MT sliding, we lowered the KCl concentration to 100 mM and increased the motor concentration until MT sliding was observed (Figure 6B and Figure 6—videos 2–4). Similarly to wt Cin8, in more than 50% of the cases, sliding produced by the NL variants was plus-end directed (Figure 6B), indicating that all the viable NL variants can reverse directionality. The velocity of plus-end MT sliding induced by Cin8-G522N was faster than that of wt Cin8 (Figure 6B and Figure 6—video 2). The significance of this finding requires further investigation (which we are currently undertaking). Finally, the velocity of plus-end directed MT sliding of the Cin8_NLCut7 and Cin8_NLEg5-NG variants was significantly slower than that of wt Cin8 (Figure 6B and Figure 6—videos 3 and 4). Taken together, our data indicate that although the NL variants can reverse to plus-end directionality during antiparallel MT sliding, their MT sliding is less efficient and requires higher motor concentrations compared to wt Cin8.

Figure 6 with 7 supplements see all

Download asset Open asset

Figure 6—source data 1

MT sliding by NL variants of Cin8.

https://cdn.elifesciences.org/articles/71036/elife-71036-fig6-data1-v2.xlsx

Download elife-71036-fig6-data1-v2.xlsx

For all the examined variants, including wt Cin8, we observed bi-directional and minus-end directed MT sliding in addition to the plus-end directed MT sliding discussed above (Figure 6, Figure 6—figure supplement 1 and Figure 6—videos 5 and 6). The occurrence of such events was similar for the different variants (Figure 6B). We have recently suggested that Cin8 can reverse directionality from minus- to plus-end directed motility due to forces opposing the directional motility, which we referred to as drag (Pandey et al., 2021). Such forces can result, for example, from numerous motors interacting with the same crosslinked MTs during antiparallel MT sliding. In accordance with this notion, in the absence of a considerable applied drag force, MT sliding can also be minus-end directed, similarly to the single molecule motility under the high salt conditions (Figure 4B,C). Based on this model, we propose that the bi-directional and minus-end directed MT sliding we observed here results from a low local concentration of Cin8 motors in the overlapping MT region and/or reduced MT affinity of the NL variants (Figure 4A). However, the fact that all the examined Cin8 motors can produce plus-end directed MT sliding suggests that if sufficient motors are present locally between overlapping MT of the spindle, antiparallel plus-end directed MT sliding can take place to produce an outwardly directed force that separates the SPBs apart.

The role of the NL dynamics in regulating the motor functions of plus-end directed kinesins has been addressed in a number of studies. However, the current study is the first to demonstrate that mutations in the NL modulate the motile properties and intracellular functions of a bi-directional N-terminal kinesin motor. The data presented here indicates that mutations in the NL regulate MT affinity, minus-end directed motility, and MT crosslinking and antiparallel sliding functions of the bi-directional kinesin-5, Cin8. In turn, these motor functions affect the ability of Cin8 to localize at the spindle poles prior to spindle assembly, to form mitotic spindles, and to support cell viability as a sole source of kinesin-5.

All data generated or analysed during this study are included in the manuscript and supporting files. All Source data files have been provided.

1. 1. Acar S
   2. Carlson DB
   3. Budamagunta MS
   4. Yarov-Yarovoy V
   5. Correia JJ
   6. Niñonuevo MR
   7. Jia W
   8. Tao L
   9. Leary JA
   10. Voss JC
   11. Evans JE
   12. Scholey JM

   (2013) The bipolar assembly domain of the mitotic motor kinesin-5

   Nature Communications 4:1343.

   https://doi.org/10.1038/ncomms2348

   • PubMed
   • Google Scholar
2. 1. Avunie-Masala R
   2. Movshovich N
   3. Nissenkorn Y
   4. Gerson-Gurwitz A
   5. Fridman V
   6. Kõivomägi M
   7. Loog M
   8. Hoyt MA
   9. Zaritsky A
   10. Gheber L

   (2011) Phospho-regulation of kinesin-5 during anaphase spindle elongation

   Journal of cell science 124:873–878.

   https://doi.org/10.1242/jcs.077396

   • PubMed
   • Google Scholar
3. 1. Bell KM
   2. Cha HK
   3. Sindelar CV
   4. Cochran JC

   (2017) The yeast kinesin-5 Cin8 interacts with the microtubule in a noncanonical manner

   Journal of Biological Chemistry 292:14680–14694.

   https://doi.org/10.1074/jbc.M117.797662

   • Google Scholar
4. 1. Blackwell R
   2. Edelmaier C
   3. Sweezy-Schindler O
   4. Lamson A
   5. Gergely ZR
   6. O'Toole E
   7. Crapo A
   8. Hough LE
   9. McIntosh JR
   10. Glaser MA
   11. Betterton MD

   (2017) Physical determinants of bipolar mitotic spindle assembly and stability in fission yeast

   Science Advances 3:e1601603.

   https://doi.org/10.1126/sciadv.1601603

   • PubMed
   • Google Scholar
5. 1. Block SM

   (2007) Kinesin motor mechanics: binding, stepping, tracking, gating, and limping

   Biophysical Journal 92:2986–2995.

   https://doi.org/10.1529/biophysj.106.100677

   • PubMed
   • Google Scholar
6. 1. Britto M
   2. Goulet A
   3. Rizvi S
   4. von Loeffelholz O
   5. Moores CA
   6. Cross RA

   (2016) Schizosaccharomyces pombe kinesin-5 switches direction using a steric blocking mechanism

   PNAS 113:E7483–E7489.

   https://doi.org/10.1073/pnas.1611581113

   • PubMed
   • Google Scholar
7. 1. Budaitis BG
   2. Jariwala S
   3. Reinemann DN
   4. Schimert KI
   5. Scarabelli G
   6. Grant BJ
   7. Sept D
   8. Lang MJ
   9. Verhey KJ

   (2019) Neck Linker docking is critical for Kinesin-1 force generation in cells but at a cost to motor speed and processivity

   eLife 8:e44146.

   https://doi.org/10.7554/eLife.44146

   • PubMed
   • Google Scholar
8. 1. Cao L
   2. Wang W
   3. Jiang Q
   4. Wang C
   5. Knossow M
   6. Gigant B

   (2014) The structure of apo-kinesin bound to tubulin links the nucleotide cycle to movement

   Nature Communications 5:5364.

   https://doi.org/10.1038/ncomms6364

   • PubMed
   • Google Scholar
9. 1. Chang Q
   2. Nitta R
   3. Inoue S
   4. Hirokawa N

   (2013) Structural basis for the ATP-induced isomerization of kinesin

   Journal of Molecular Biology 425:1869–1880.

   https://doi.org/10.1016/j.jmb.2013.03.004

   • PubMed
   • Google Scholar
10. 1. Chen GY
    2. Mickolajczyk KJ
    3. Hancock WO

    (2016) The Kinesin-5 Chemomechanical Cycle Is Dominated by a Two-heads-bound State

    The Journal of Biological Chemistry 291:20283–20294.

    https://doi.org/10.1074/jbc.M116.730697

    • PubMed
    • Google Scholar
11. 1. Clancy BE
    2. Behnke-Parks WM
    3. Andreasson JO
    4. Rosenfeld SS
    5. Block SM

    (2011) A universal pathway for kinesin stepping

    Nature Structural & Molecular Biology 18:1020–1027.

    https://doi.org/10.1038/nsmb.2104

    • PubMed
    • Google Scholar
12. 1. Dogan MY
    2. Can S
    3. Cleary FB
    4. Purde V
    5. Yildiz A

    (2015) Kinesin's front head is gated by the backward orientation of its neck linker

    Cell Reports 10:1967–1973.

    https://doi.org/10.1016/j.celrep.2015.02.061

    • PubMed
    • Google Scholar
13. 1. Dunnett CW

    (1955) A multiple comparison procedure for comparing several treatments with a control

    Journal of the American Statistical Association 50:1096–1121.

    https://doi.org/10.1080/01621459.1955.10501294

    • Google Scholar
14. 1. Düselder A
    2. Thiede C
    3. Schmidt CF
    4. Lakämper S

    (2012) Neck-linker length dependence of processive Kinesin-5 motility

    Journal of Molecular Biology 423:159–168.

    https://doi.org/10.1016/j.jmb.2012.06.043

    • PubMed
    • Google Scholar
15. 1. Düselder A
    2. Fridman V
    3. Thiede C
    4. Wiesbaum A
    5. Goldstein A
    6. Klopfenstein DR
    7. Zaitseva O
    8. Janson ME
    9. Gheber L
    10. Schmidt CF

    (2015) Deletion of the tail domain of the Kinesin-5 Cin8 affects its directionality

    Journal of Biological Chemistry 290:16841–16850.

    https://doi.org/10.1074/jbc.M114.620799

    • PubMed
    • Google Scholar
16. 1. Edamatsu M

    (2014) Bidirectional motility of the fission yeast kinesin-5, Cut7

    Biochemical and Biophysical Research Communications 446:231–234.

    https://doi.org/10.1016/j.bbrc.2014.02.106

    • PubMed
    • Google Scholar
17. 1. Edgar RC

    (2004) MUSCLE: a multiple sequence alignment method with reduced time and space complexity

    BMC Bioinformatics 5:113.

    https://doi.org/10.1186/1471-2105-5-113

    • PubMed
    • Google Scholar
18. 1. Fallesen T
    2. Roostalu J
    3. Duellberg C
    4. Pruessner G
    5. Surrey T

    (2017) Ensembles of Bidirectional Kinesin Cin8 Produce Additive Forces in Both Directions of Movement

    Biophysical Journal 113:2055–2067.

    https://doi.org/10.1016/j.bpj.2017.09.006

    • PubMed
    • Google Scholar
19. 1. Fridman V
    2. Gerson-Gurwitz A
    3. Shapira O
    4. Movshovich N
    5. Lakämper S
    6. Schmidt CF
    7. Gheber L

    (2013) Kinesin-5 Kip1 is a bi-directional motor that stabilizes microtubules and tracks their plus-ends in vivo

    Journal of Cell Science 126:4147–4159.

    https://doi.org/10.1242/jcs.125153

    • PubMed
    • Google Scholar
20. 1. Geng Y-Z
    2. Li T
    3. Ji Q
    4. Yan S

    (2014) Simulation study of interactions between kinesin’s Neck Linker and Motor Domain

    Cellular and Molecular Bioengineering 7:99–105.

    https://doi.org/10.1007/s12195-014-0320-4

    • Google Scholar
21. 1. Gerson-Gurwitz A
    2. Movshovich N
    3. Avunie R
    4. Fridman V
    5. Moyal K
    6. Katz B
    7. Hoyt MA
    8. Gheber L

    (2009) Mid-anaphase arrest in S. cerevisiae cells eliminated for the function of Cin8 and dynein

    Cellular and molecular life sciences : CMLS 66:301–313.

    https://doi.org/10.1007/s00018-008-8479-2

    • PubMed
    • Google Scholar
22. 1. Gerson-Gurwitz A
    2. Thiede C
    3. Movshovich N
    4. Fridman V
    5. Podolskaya M
    6. Danieli T
    7. Lakämper S
    8. Klopfenstein DR
    9. Schmidt CF
    10. Gheber L

    (2011) Directionality of individual kinesin-5 Cin8 motors is modulated by loop 8, ionic strength and microtubule geometry

    The EMBO journal 30:4942–4954.

    https://doi.org/10.1038/emboj.2011.403

    • PubMed
    • Google Scholar
23. 1. Gheber L
    2. Kuo SC
    3. Hoyt MA

    (1999) Motile properties of the kinesin-related Cin8p spindle motor extracted from Saccharomyces cerevisiae cells

    Journal of Biological Chemistry 274:9564–9572.

    https://doi.org/10.1074/jbc.274.14.9564

    • PubMed
    • Google Scholar
24. 1. Gigant B
    2. Wang W
    3. Dreier B
    4. Jiang Q
    5. Pecqueur L
    6. Plückthun A
    7. Wang C
    8. Knossow M

    (2013) Structure of a kinesin-tubulin complex and implications for kinesin motility

    Nature Structural & Molecular Biology 20:1001–1007.

    https://doi.org/10.1038/nsmb.2624

    • PubMed
    • Google Scholar
25. 1. Goldstein A
    2. Siegler N
    3. Goldman D
    4. Judah H
    5. Valk E
    6. Kõivomägi M
    7. Loog M
    8. Gheber L

    (2017) Three Cdk1 sites in the kinesin-5 Cin8 catalytic domain coordinate motor localization and activity during anaphase

    Cellular and Molecular Life Sciences 74:3395–3412.

    https://doi.org/10.1007/s00018-017-2523-z

    • PubMed
    • Google Scholar
26. 1. Goldstein A
    2. Goldman D
    3. Valk E
    4. Loog M
    5. Holt LJ
    6. Gheber L

    (2019) Synthetic-Evolution reveals narrow paths to regulation of the Saccharomyces cerevisiae Mitotic Kinesin-5 Cin8

    International Journal of Biological Sciences 15:1125–1138.

    https://doi.org/10.7150/ijbs.30543

    • PubMed
    • Google Scholar
27. 1. Gordon DM
    2. Roof DM

    (1999) The kinesin-related protein Kip1p of Saccharomyces cerevisiae is bipolar

    The Journal of biological chemistry 274:28779–28786.

    https://doi.org/10.1074/jbc.274.40.28779

    • PubMed
    • Google Scholar
28. 1. Goulet A
    2. Behnke-Parks WM
    3. Sindelar CV
    4. Major J
    5. Rosenfeld SS
    6. Moores CA

    (2012) The structural basis of force generation by the mitotic motor kinesin-5

    The Journal of biological chemistry 287:44654–44666.

    https://doi.org/10.1074/jbc.M112.404228

    • PubMed
    • Google Scholar
29. 1. Goulet A
    2. Major J
    3. Jun Y
    4. Gross SP
    5. Rosenfeld SS
    6. Moores CA

    (2014) Comprehensive structural model of the mechanochemical cycle of a mitotic motor highlights molecular adaptations in the kinesin family

    PNAS 111:1837–1842.

    https://doi.org/10.1073/pnas.1319848111

    • PubMed
    • Google Scholar
30. 1. Goulet A
    2. Moores C

    (2013) New insights into the mechanism of force generation by kinesin-5 molecular motors

    International review of cell and molecular biology 304:419–466.

    https://doi.org/10.1016/B978-0-12-407696-9.00008-7

    • PubMed
    • Google Scholar
31. 1. Guydosh NR
    2. Block SM

    (2006) Backsteps induced by nucleotide analogs suggest the front head of kinesin is gated by strain

    PNAS 103:8054–8059.

    https://doi.org/10.1073/pnas.0600931103

    • PubMed
    • Google Scholar
32. 1. Higuchi H
    2. Endow SA

    (2002) Directionality and processivity of molecular motors

    Current opinion in cell biology 14:50–57.

    https://doi.org/10.1016/s0955-0674(01)00293-9

    • PubMed
    • Google Scholar
33. 1. Hildebrandt ER
    2. Gheber L
    3. Kingsbury T
    4. Hoyt MA

    (2006) Homotetrameric form of Cin8p, a Saccharomyces cerevisiae kinesin-5 motor, is essential for its in vivo function

    The Journal of Biological Chemistry 281:26004–26013.

    https://doi.org/10.1074/jbc.M604817200

    • PubMed
    • Google Scholar
34. 1. Hughes J
    2. Hancock WO
    3. Fricks J

    (2012) Kinesins with extended neck linkers: a chemomechanical model for variable-length stepping

    Bulletin of mathematical biology 74:1066–1097.

    https://doi.org/10.1007/s11538-011-9697-6

    • PubMed
    • Google Scholar
35. 1. Hwang W
    2. Lang MJ
    3. Karplus M

    (2008) Force generation in kinesin hinges on cover-neck bundle formation

    Structure 16:62–71.

    https://doi.org/10.1016/j.str.2007.11.008

    • PubMed
    • Google Scholar
36. 1. Jones DT
    2. Taylor WR
    3. Thornton JM

    (1992) The rapid generation of mutation data matrices from protein sequences

    Computer applications in the biosciences : CABIOS 8:275–282.

    https://doi.org/10.1093/bioinformatics/8.3.275

    • PubMed
    • Google Scholar
37. 1. Kapitein LC
    2. Peterman EJ
    3. Kwok BH
    4. Kim JH
    5. Kapoor TM
    6. Schmidt CF

    (2005) The bipolar mitotic kinesin Eg5 moves on both microtubules that it crosslinks

    Nature 435:114–118.

    https://doi.org/10.1038/nature03503

    • PubMed
    • Google Scholar
38. 1. Kapitein LC
    2. Kwok BH
    3. Weinger JS
    4. Schmidt CF
    5. Kapoor TM
    6. Peterman EJ

    (2008) Microtubule cross-linking triggers the directional motility of kinesin-5

    The Journal of cell biology 182:421–428.

    https://doi.org/10.1083/jcb.200801145

    • PubMed
    • Google Scholar
39. 1. Kapoor TM

    (2017) Metaphase spindle assembly

    Biology 6:8.

    https://doi.org/10.3390/biology6010008

    • PubMed
    • Google Scholar
40. 1. Kashina AS
    2. Baskin RJ
    3. Cole DG
    4. Wedaman KP
    5. Saxton WM
    6. Scholey JM

    (1996) A bipolar kinesin

    Nature 379:270–272.

    https://doi.org/10.1038/379270a0

    • PubMed
    • Google Scholar
41. 1. Khalil AS
    2. Appleyard DC
    3. Labno AK
    4. Georges A
    5. Karplus M
    6. Belcher AM
    7. Hwang W
    8. Lang MJ

    (2008) Kinesin's cover-neck bundle folds forward to generate force

    PNAS 105:19247–19252.

    https://doi.org/10.1073/pnas.0805147105

    • PubMed
    • Google Scholar
42. 1. Khmelinskii A
    2. Roostalu J
    3. Roque H
    4. Antony C
    5. Schiebel E

    (2009) Phosphorylation-dependent protein interactions at the spindle midzone mediate cell cycle regulation of spindle elongation

    Developmental cell 17:244–256.

    https://doi.org/10.1016/j.devcel.2009.06.011

    • PubMed
    • Google Scholar
43. 1. Khmelinskii A
    2. Schiebel E

    (2008) Assembling the spindle midzone in the right place at the right time

    Cell Cycle 7:283–286.

    https://doi.org/10.4161/cc.7.3.5349

    • PubMed
    • Google Scholar
44. 1. Klumpp LM
    2. Hoenger A
    3. Gilbert SP

    (2004) Kinesin's second step

    PNAS 101:3444–3449.

    https://doi.org/10.1073/pnas.0307691101

    • PubMed
    • Google Scholar
45. 1. Kozielski F
    2. Sack S
    3. Marx A
    4. Thormählen M
    5. Schönbrunn E
    6. Biou V
    7. Thompson A
    8. Mandelkow EM
    9. Mandelkow E

    (1997) The crystal structure of dimeric kinesin and implications for microtubule-dependent motility

    Cell 91:985–994.

    https://doi.org/10.1016/s0092-8674(00)80489-4

    • PubMed
    • Google Scholar
46. 1. Krzysiak TC
    2. Grabe M
    3. Gilbert SP

    (2008) Getting in sync with dimeric Eg5. Initiation and regulation of the processive run

    The Journal of biological chemistry 283:2078–2087.

    https://doi.org/10.1074/jbc.M708354200

    • PubMed
    • Google Scholar
47. 1. Leary A
    2. Sim S
    3. Nazarova E
    4. Shulist K
    5. Genthial R
    6. Yang SK
    7. Bui KH
    8. Francois P
    9. Vogel J

    (2019) Successive Kinesin-5 microtubule crosslinking and sliding promote fast, irreversible formation of a stereotyped bipolar spindle

    Current Biology 29:3825–3837.

    https://doi.org/10.1016/j.cub.2019.09.030

    • PubMed
    • Google Scholar
48. 1. Mann BJ
    2. Wadsworth P

    (2019) Kinesin-5 Regulation and Function in Mitosis

    Trends in cell biology 29:66–79.

    https://doi.org/10.1016/j.tcb.2018.08.004

    • PubMed
    • Google Scholar
49. 1. Massey FJ

    (1951) The Kolmogorov-Smirnov test for goodness of fit

    Journal of the American Statistical Association 46:68–78.

    https://doi.org/10.1080/01621459.1951.10500769

    • Google Scholar
50. 1. Movshovich N
    2. Fridman V
    3. Gerson-Gurwitz A
    4. Shumacher I
    5. Gertsberg I
    6. Fich A
    7. Hoyt MA
    8. Katz B
    9. Gheber L

    (2008) Slk19-dependent mid-anaphase pause in kinesin-5-mutated cells

    Journal of cell science 121:2529–2539.

    https://doi.org/10.1242/jcs.022996

    • PubMed
    • Google Scholar
51. 1. Muretta JM
    2. Jun Y
    3. Gross SP
    4. Major J
    5. Thomas DD
    6. Rosenfeld SS

    (2015) The structural kinetics of switch-1 and the neck linker explain the functions of kinesin-1 and Eg5

    PNAS 112:E6606–E6613.

    https://doi.org/10.1073/pnas.1512305112

    • PubMed
    • Google Scholar
52. 1. Muretta JM
    2. Reddy BJN
    3. Scarabelli G
    4. Thompson AF
    5. Jariwala S
    6. Major J
    7. Venere M
    8. Rich JN
    9. Willard B
    10. Thomas DD
    11. Stumpff J
    12. Grant BJ
    13. Gross SP
    14. Rosenfeld SS

    (2018) A posttranslational modification of the mitotic kinesin Eg5 that enhances its mechanochemical coupling and alters its mitotic function

    PNAS 115:E1779–E1788.

    https://doi.org/10.1073/pnas.1718290115

    • PubMed
    • Google Scholar
53. 1. Nitta R
    2. Kikkawa M
    3. Okada Y
    4. Hirokawa N

    (2004) KIF1A alternately uses two loops to bind microtubules

    Science 305:678–683.

    https://doi.org/10.1126/science.1096621

    • PubMed
    • Google Scholar
54. 1. Pandey H
    2. Reithmann E
    3. Goldstein-Levitin A
    4. Al-Bassam J
    5. Frey E
    6. Gheber L

    (2021) Drag-induced directionality switching of kinesin-5 Cin8 revealed by cluster-motility analysis

    Science Advances 7:eabc1687.

    https://doi.org/10.1126/sciadv.abc1687

    • PubMed
    • Google Scholar
55. 1. Parke CL
    2. Wojcik EJ
    3. Kim S
    4. Worthylake DK

    (2010) ATP hydrolysis in Eg5 kinesin involves a catalytic two-water mechanism

    Journal of Biological Chemistry 285:5859–5867.

    https://doi.org/10.1074/jbc.M109.071233

    • PubMed
    • Google Scholar
56. 1. Pettersen EF
    2. Goddard TD
    3. Huang CC
    4. Couch GS
    5. Greenblatt DM
    6. Meng EC
    7. Ferrin TE

    (2004) UCSF chimera--a visualization system for exploratory research and analysis

    Journal of Computational Chemistry 25:1605–1612.

    https://doi.org/10.1002/jcc.20084

    • PubMed
    • Google Scholar
57. 1. Rice S
    2. Lin AW
    3. Safer D
    4. Hart CL
    5. Naber N
    6. Carragher BO
    7. Cain SM
    8. Pechatnikova E
    9. Wilson-Kubalek EM
    10. Whittaker M
    11. Pate E
    12. Cooke R
    13. Taylor EW
    14. Milligan RA
    15. Vale RD

    (1999) A structural change in the kinesin motor protein that drives motility

    Nature 402:778–784.

    https://doi.org/10.1038/45483

    • PubMed
    • Google Scholar
58. 1. Roof DM
    2. Meluh PB
    3. Rose MD

    (1992) Kinesin-related proteins required for assembly of the mitotic spindle

    The Journal of cell biology 118:95–108.

    https://doi.org/10.1083/jcb.118.1.95

    • PubMed
    • Google Scholar
59. 1. Roostalu J
    2. Hentrich C
    3. Bieling P
    4. Telley IA
    5. Schiebel E
    6. Surrey T

    (2011) Directional switching of the kinesin Cin8 through motor coupling

    Science 332:94–99.

    https://doi.org/10.1126/science.1199945

    • PubMed
    • Google Scholar
60. 1. Rosenfeld SS
    2. Jefferson GM
    3. King PH

    (2001) ATP reorients the neck linker of kinesin in two sequential steps

    The Journal of biological chemistry 276:40167–40174.

    https://doi.org/10.1074/jbc.M103899200

    • PubMed
    • Google Scholar
61. 1. Sakia RM

    (1992) The Box-Cox transformation technique: a review

    The Statistician 41:169–178.

    https://doi.org/10.2307/2348250

    • Google Scholar
62. 1. Sali A
    2. Blundell TL

    (1993) Comparative protein modelling by satisfaction of spatial restraints

    Journal of molecular biology 234:779–815.

    https://doi.org/10.1006/jmbi.1993.1626

    • PubMed
    • Google Scholar
63. 1. Saunders WS
    2. Hoyt MA

    (1992) Kinesin-related proteins required for structural integrity of the mitotic spindle

    Cell 70:451–458.

    https://doi.org/10.1016/0092-8674(92)90169-d

    • PubMed
    • Google Scholar
64. 1. Schief WR
    2. Clark RH
    3. Crevenna AH
    4. Howard J

    (2004) Inhibition of kinesin motility by ADP and phosphate supports a hand-over-hand mechanism

    PNAS 101:1183–1188.

    https://doi.org/10.1073/pnas.0304369101

    • PubMed
    • Google Scholar
65. 1. Schief WR
    2. Howard J

    (2001) Conformational changes during kinesin motility

    Current opinion in cell biology 13:19–28.

    https://doi.org/10.1016/s0955-0674(00)00169-1

    • PubMed
    • Google Scholar
66. 1. Scholey JE
    2. Nithianantham S
    3. Scholey JM
    4. Al-Bassam J

    (2014) Structural basis for the assembly of the mitotic motor Kinesin-5 into bipolar tetramers

    eLife 3:e02217.

    https://doi.org/10.7554/eLife.02217

    • PubMed
    • Google Scholar
67. 1. Scholey JM
    2. Civelekoglu-Scholey G
    3. Brust-Mascher I

    (2016) Anaphase B

    Biology 5:51.

    https://doi.org/10.3390/biology5040051

    • PubMed
    • Google Scholar
68. 1. Shang Z
    2. Zhou K
    3. Xu C
    4. Csencsits R
    5. Cochran JC
    6. Sindelar CV

    (2014) High-resolution structures of kinesin on microtubules provide a basis for nucleotide-gated force-generation

    eLife 3:e04686.

    https://doi.org/10.7554/eLife.04686

    • PubMed
    • Google Scholar
69. 1. Shapira O
    2. Goldstein A
    3. Al-Bassam J
    4. Gheber L

    (2016) Regulation and possible physiological role of BI-Directional motility of the Kinesin-5 Cin8

    Biophysical Journal 110:460a.

    https://doi.org/10.1016/j.bpj.2015.11.2462

    • Google Scholar
70. 1. Shapira O
    2. Goldstein A
    3. Al-Bassam J
    4. Gheber L

    (2017) A potential physiological role for bi-directional motility and motor clustering of mitotic kinesin-5 Cin8 in yeast mitosis

    Journal of cell science 130:725–734.

    https://doi.org/10.1242/jcs.195040

    • PubMed
    • Google Scholar
71. 1. Shapira O
    2. Gheber L

    (2016) Motile properties of the bi-directional kinesin-5 Cin8 are affected by phosphorylation in its motor domain

    Scientific reports 6:25597.

    https://doi.org/10.1038/srep25597

    • PubMed
    • Google Scholar
72. 1. Sharp DJ
    2. McDonald KL
    3. Brown HM
    4. Matthies HJ
    5. Walczak C
    6. Vale RD
    7. Mitchison TJ
    8. Scholey JM

    (1999) The bipolar kinesin, KLP61F, cross-links microtubules within interpolar microtubule bundles of Drosophila embryonic mitotic spindles

    The Journal of cell biology 144:125–138.

    https://doi.org/10.1083/jcb.144.1.125

    • PubMed
    • Google Scholar
73. 1. Shastry S
    2. Hancock WO

    (2010) Neck linker length determines the degree of processivity in kinesin-1 and kinesin-2 motors

    Current biology : CB 20:939–943.

    https://doi.org/10.1016/j.cub.2010.03.065

    • PubMed
    • Google Scholar
74. 1. Shastry S
    2. Hancock WO

    (2011) Interhead tension determines processivity across diverse N-terminal kinesins

    PNAS 108:16253–16258.

    https://doi.org/10.1073/pnas.1102628108

    • PubMed
    • Google Scholar
75. 1. Shimamoto Y
    2. Forth S
    3. Kapoor TM

    (2015) Measuring Pushing and Braking Forces Generated by Ensembles of Kinesin-5 Crosslinking Two Microtubules

    Developmental cell 34:669–681.

    https://doi.org/10.1016/j.devcel.2015.08.017

    • PubMed
    • Google Scholar
76. 1. Sindelar CV
    2. Budny MJ
    3. Rice S
    4. Naber N
    5. Fletterick R
    6. Cooke R

    (2002) Two conformations in the human kinesin power stroke defined by X-ray crystallography and EPR spectroscopy

    Nature structural biology 9:844–848.

    https://doi.org/10.1038/nsb852

    • PubMed
    • Google Scholar
77. 1. Sindelar CV
    2. Downing KH

    (2010) An atomic-level mechanism for activation of the kinesin molecular motors

    PNAS 107:4111–4116.

    https://doi.org/10.1073/pnas.0911208107

    • PubMed
    • Google Scholar
78. 1. Singh SK
    2. Pandey H
    3. Al-Bassam J
    4. Gheber L

    (2018) Bidirectional motility of kinesin-5 motor proteins: structural determinants, cumulative functions and physiological roles

    Cellular and Molecular Life Sciences 75:1757–1771.

    https://doi.org/10.1007/s00018-018-2754-7

    • PubMed
    • Google Scholar
79. 1. Tinevez JY
    2. Perry N
    3. Schindelin J
    4. Hoopes GM
    5. Reynolds GD
    6. Laplantine E
    7. Bednarek SY
    8. Shorte SL
    9. Eliceiri KW

    (2017) TrackMate: An open and extensible platform for single-particle tracking

    Methods 115:80–90.

    https://doi.org/10.1016/j.ymeth.2016.09.016

    • PubMed
    • Google Scholar
80. 1. Tomishige M
    2. Stuurman N
    3. Vale RD

    (2006) Single-molecule observations of neck linker conformational changes in the kinesin motor protein

    Nature structural & molecular biology 13:887–894.

    https://doi.org/10.1038/nsmb1151

    • PubMed
    • Google Scholar
81. 1. Tukey JW

    (1949) Comparing individual means in the analysis of variance

    Biometrics 5:99–114.

    https://doi.org/10.2307/3001913

    • PubMed
    • Google Scholar
82. 1. UGENE team
    2. Okonechnikov K
    3. Golosova O
    4. Fursov M

    (2012) Unipro UGENE: a unified bioinformatics toolkit

    Bioinformatics 28:1166–1167.

    https://doi.org/10.1093/bioinformatics/bts091

    • PubMed
    • Google Scholar
83. 1. Vale RD
    2. Fletterick RJ

    (1997) The design plan of kinesin motors

    Annual review of cell and developmental biology 13:745–777.

    https://doi.org/10.1146/annurev.cellbio.13.1.745

    • PubMed
    • Google Scholar
84. 1. von Loeffelholz O
    2. Peña A
    3. Drummond DR
    4. Cross R
    5. Moores CA

    (2019) Cryo-EM Structure (4.5-Å) of Yeast Kinesin-5-Microtubule Complex Reveals a Distinct Binding Footprint and Mechanism of Drug Resistance

    Journal of molecular biology 431:864–872.

    https://doi.org/10.1016/j.jmb.2019.01.011

    • PubMed
    • Google Scholar
85. 1. Waitzman JS
    2. Rice SE

    (2014) Mechanism and regulation of kinesin-5, an essential motor for the mitotic spindle

    Biology of the cell 106:1–12.

    https://doi.org/10.1111/boc.201300054

    • PubMed
    • Google Scholar
86. 1. Yildiz A
    2. Tomishige M
    3. Gennerich A
    4. Vale RD

    (2008) Intramolecular strain coordinates kinesin stepping behavior along microtubules

    Cell 134:1030–1041.

    https://doi.org/10.1016/j.cell.2008.07.018

    • PubMed
    • Google Scholar

Acceptance summary:

This paper dissects in great detail the properties of the neck linker of the bidirectional kinesin Cin8. Neck linker is the main mobile element for generating a step by the motor protein kinesin. The described observations on neck linker mutations provide valuable insights into the determinants of bidirectionality of Cin8 and motivate further investigation into the underlying mechanism.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "Flexibility of the neck-linker during docking is pivotal for function of bi-directional kinesin" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by a Senior Editor. The reviewers have opted to remain anonymous.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

All three reviewers felt that the experiments are appropriate and well-done and that the manuscript is well-written. However, as you will see from the specific comments below, all three reviewers felt the results and interpretations are limited to our understanding of Cin8 motility and do not provide sufficient knowledge to advance our understanding of how some kinesin-5 motors achieve bidirectional motion. All three felt this is a very nice story and should be published with the toning down of some interpretations and we hope the reviewers' comments will be helpful to you in this regard.

Reviewer #1:

The manuscript by Goldstein-Levitin et al. examines the role of a kinesin structural element called the neck linker (NL) in the function of an unusual kinesin-5 motor, the S cerevisiae kinesin-5 motor Cin8, which previous work has demonstrated to switch directionality from fast processive minus‐end directed to slow, processive plus‐end directed motility depending on ionic conditions or motor number.

Using both in vitro and in cell assays, the authors demonstrate that replacement of the entire NL by the analogous sequences from either a plus-end motor or bidirectional motor is detrimental to single motor motility and cellular function. In contrast, single point mutations either have no effect or decrease motility properties with minor effects on cellular function. Overall the results provide support for the hypothesis that the NL plays a role in kinesin-5 motility. However, I find the conclusion that "flexibility of the NL during docking is pivotal for function of bi-directional kinesin" unsupported for the reasons outlined below. The authors have carried out the proper experiments and these are well-done but I find the overall conclusions need additional consideration.

1. The authors are equating the presence of a G residue in the N-latch position of Cin8 with flexibility of the NL. It is not clear to me what they mean by flexibility (in the undocked position? in the ionic or backbone interactions in the docked position?). What is the evidence that the flexibility is changed?

2. For Cin8, the authors demonstrate that mutation of G in the N-latch position increases the doubling time but affects cell viability only at high temperatures. The authors conclude that "increased flexibility of the NL during docking of Cin8, compared to Eg5 (Figure 5) provides the necessary adaptation to allow bi-directional motility." (p.16, Discussion). However, the other bi-directional kinesin-5s (Kip1 and Cut7) have an N at the N-latch position and their NLs are thus presumably not flexible although the motors are functional and cells are viable.

3. On p. 14 in the Discussion, the authors state "The profound difference between the activity of Cin8NLEg5 and Cin8NLEg5-NG variants indicates that the glycine in position 522 is critical for the functions of Cin8." The problem is that this conclusion is based on a chimeric motor that does not occur naturally. In fact, if one considers Cin8 with its own NL sequence, both the WT and the G522N mutant Cin8 motors are functional and cells expressing these motors are viable under normal growth conditions. In addition, the other bi-directional kinesin-5s also have an N at this position and are functional.

4. The relationship between G vs N in the N-latch position and motor motility and function is overall unclear. The authors focused on 3 mutant versions in motility assays: Cin8-G522N (has N in N-latch position), Cin8NLCut7 (has N in N-latch position), and Cin8NLEg5-NG (has G in N-latch position). In single-molecule motility assays, all show decreased affinity to the MT and processive minus end-directly motility albeit with decreased velocity as compared to the WT Cin8 motor (has G in N-latch position). In MT bundling assays, all showed a reduced ability to bundle MTs. But their functions are different as are their effects on cell viability.

For example, the NLCut7 and G522N mutants both have an N in the N-latch position. Both show decreased MT affinity, slower minus end-directed motility, and decreased MT bundling activity. Both localize to the SPB in early mitosis. However the NLCut7 mutant generates more monopolar spindles and is not viable at normal yeast growth temperature whereas the G522N mutant generates normal spindle distributions and is viable. It thus appears that having a N in the N-latch position does not correlate with viability.

As a second comparison, the Eg5-NG mutant (G in N-latch position) and the G522N (N in N-latch position) both display decreased MT affinity and slower minus end-directed motility. The Eg5-NG mutant shows severely reduced MT bundling activity and the G522N mutant shows reduced MT bundling activity. Both localize to the SPB in early mitosis and generate normal spindle distributions. And both are viable. It thus appears that the residue in the N-latch position does not correlate with cellular function.

5. In my opinion, it would be most interesting to know whether the NL replacements or mutations alter the bi-directional behavior of Cin8. This is not clear to me from Figure 3. Can the authors provide information about the frequency of each type of motion (fast minus end, diffusive, slow plus end) for each motor at each salt concentration? While all of the mutant motors examined in these assays are slower than the WT motor, their frequency of transitioning to plus end growth is unclear. What salt condition was utilized for the data in 3D?

6. Does mutation of the N>G in the NL of Cut7 in the NLCut7 mutant rescue viability?

Reviewer #2:

The authors investigate the role of neck linker docking in the bidirection motility of Cin8. The N-latch idea is well established for kinesin-1 from Lang and Hwang's work. Most kinesin-5 have an N in this position, including two of the three bidirectional kinesin-5, but Cin8 has a glycine in this position. This is notable, because differences in neck linker docking are a good candidate for why fungal kinesin-5 are bidirectional.

The authors investigate chimaeras containing Eg5 or Cut7 neck linker replacement into Cin8. A number of assays are used including viability, in vivo localization, single-molecule velocity, and in vitro microtubule bunding. Pleasingly, the assays all give pretty consistent results. Cin8 doesn't like N replacing G, it can handle the Eg5 neck linker as long as you mutate N to G, and Cut7 NL works better than Eg5.

The paper is well written and the diversity of assays to demonstrate functionality is a strong point of the paper. The question of bidirectionality is important and still puzzling from a mechanistic standpoint.

A weakness is that there is clearly a complex interaction between the neck linker, the cover bundle and the catalytic core of the head, and so it is difficult to interpret what these mutations might be doing to the complex mechanochemistry of these motors. It is notable that the three key residues that interact with N in kinesin-1 are conserved in Cin8, but that means we don't really understand the glycine is doing in wild-type. That mutating it to N screws the motor up is notable, but doesn't shed a lot of light either on NL docking or on bidirectionality.

A smaller but important point is the use of the term "flexibility". Generally glycines increase flexibility of polypeptides. But that isn't actually rigorously shown here. And the effect of the glycine may not have anything to do with flexibility, and rather may all be about hydrogen bonds. In the discussion, the authors bring up a kinesin-1 with longer and hence more flexible neck linker that can be made to step backwards, but that is adding sequence and hence flexibility at the distal part of the neck linker that doesn't dock, so I don't think this comparison is necessarily apt.

The changes in speed are notable, but the mechanism of stepping is too complicated to just attribute the changes to neck linker docking. Because neck linker docking is so fundamental to the chemomechanical cycle, changing these interactions could be changing a lot about the motor.

Overall, this is important work for the field and to add another detail to the question of how fungal kinesin-5 can walk bidirectionally. However, rather than making us think differently about the problem, it is contributing important new details. That limits the impact of the study in my eyes.

Reviewer #3:

While kinesin motors have been intensively studied for over thirty years, and while there is consensus on mechanisms underlying "+" end directed movement and processivity, exceptions to the current "rules" exist, most notably concerning the ability of some kinesin 5 family members to move processively in both the "+" and "-" directions. This remarkable feature, which has been known for over a decade, begs a structure-based explanation. In this manuscript, the authors use a mutational approach to examine the structural features needed for bi-directional processive movement in Cin8. Their studies, spanning cell biology to single molecule motillity assays lead them to conclude that bi-directional motility requires "flexibility" in the neck linker, which in Cin8 is provided by the presence of a glycine in a position that in plus end directed kinesins is occupied by an apsaragine. They hypothesize that this single residue difference destabilizes the neck linker from docking along the motor surface toward the "+" end and increases said flexibility.

While the experiments appear to be well done and the text is clearly written, I have several major editorial concerns that I believe need to be addressed in a revised manuscript:

1. NL flexibility: The authors are vague as to what they mean here. One presumes, from their citing of literature that proposes an ordered to disordered NL transition that this flexibility implies increased disorder in the NL which somehow translates to "-" directed motion. However, I note that cryoEM reconstructions of several "+" directed kinesins, including Eg5 and Kif20A, show that the NL is in fact ordered in both pre- and post-hydrolytic states, pointing towards the "+" end in the former and the "-" end in the latter. In view of this, it seems much more likely that there is rather a dynamic equilibrium between two orientations in the NL (minus and plus directed) and that while the NL asparagine drives this equilibrium toward a structured (non flexible) NL in a plus orientation, its loss drives this equilibrium toward a similarly structured, non flexible NL pointing in the minus direction. This is not a minor point, since the term flexibility that the authors use here could mean NL conformational flexibility (ordered versus disordered) or directional flexibility (minus versus plus directed). I believe that the authors need to be considerably clearer about what they mean by "flexibility" and how they envision the loss of the stabilizing hydrogen bonds leading mechanistically to "-" directed motion.

2. Importance of NL glycine: Given the importance the authors place in Cin8 glycine 522 in determining directional flexibility, it is striking that two other bi-directional kinesin 5 family members (Kip1 and Cut7) have an arginine at this position. The authors cite a a cryoEM model (which incidentally while included in the references as #29 is not cited in the text of the discussion) indicating that the NL of cut7 is not stabilized by NL-motor hydrogen bonds to the same degree as wild type Cin8. However, I am not sure that a 4.5 Å resolution map can definitively allow for this conclusion, and I believe that the authors need to expand their discussion of this point, including the potential limitations of their interpretation in this light. Finally, the authors note that mutating the N latch in kinesin 1 increases motor velocity. This is a striking finding that appears to be consistent with their own studies of Cin8. Certainly one possibility is that hydrogen bonding between the NL and the motor core slows NL undocking, which becomes rate limiting in the single molecule assay. Regardless, some mechanism needs to be proposed to explain this besides the relatively uninformative statement that "stabilization of the NL docking plays similar role in controlling the velocity of plus- and minus-end directed N terminal kinesin motors.

[Editors’ note: The authors appealed the original decision. What follows is the authors’ response to the first round of review.]

Reviewer #1:

The manuscript by Goldstein-Levitin et al. examines the role of a kinesin structural element called the neck linker (NL) in the function of an unusual kinesin-5 motor, the S cerevisiae kinesin-5 motor Cin8, which previous work has demonstrated to switch directionality from fast processive minus‐end directed to slow, processive plus‐end directed motility depending on ionic conditions or motor number.

Using both in vitro and in cell assays, the authors demonstrate that replacement of the entire NL by the analogous sequences from either a plus-end motor or bidirectional motor is detrimental to single motor motility and cellular function. In contrast, single point mutations either have no effect or decrease motility properties with minor effects on cellular function. Overall the results provide support for the hypothesis that the NL plays a role in kinesin-5 motility. However, I find the conclusion that "flexibility of the NL during docking is pivotal for function of bi-directional kinesin" unsupported for the reasons outlined below. The authors have carried out the proper experiments and these are well-done but I find the overall conclusions need additional consideration.

1. The authors are equating the presence of a G residue in the N-latch position of Cin8 with flexibility of the NL. It is not clear to me what they mean by flexibility (in the undocked position? in the ionic or backbone interactions in the docked position?). What is the evidence that the flexibility is changed?

We thank the reviewer for this comment, and we agree that the term “flexibility” that we used to describe NL docking and dynamics could be ambiguous. Similar concerns are also expressed by Reviewer # 2 (comments 1 and 2) and by Reviewer # 3 (comment 1). To address this point, in collaboration with Dr. Itamar Kass, we performed homology modeling of the Cin8 motor domain of wt Cin8 and its NL mutants (Figures 3 and S2 and Table 2). The modeling is based on the following four PDB structures of kinesin motors published previously, all in the presence of AMPPNP—conditions under which the NL is docked and points in the plus-end direction of the MT: (a) 3HQD, X-ray structure of the Eg5 motor domain (Parke et al., 2010); (b) 1VFV, X-ray structure of KIF1A kinesin heavy chain isoform 5C (Nitta et al., 2004); (c) 3ZFC, X-ray structure of KIF4A isoform 4A (Chang et al., 2013); and (d) 6S8M, CryoEM structure of S. pombe kinesin-5 Cut7 decorating MTs (von Loeffelholz et al., 2019).

We used the generated model of the Cin8 motor domain to study the H-bond network between residues in the NL (⁵¹³KNIKNKPQLGSF⁵²⁴) and in the β7 strand of the motor domain. Calculation of the H-bonds is based on the distance between donor and acceptor atoms (smaller than 0.35 nm) and the angle between the acceptor, donor and hydrogen atoms (smaller than 30 degrees).

Our analysis revealed that mutations in the NL affect the population of H-bonds formed between the NL and β7 of the motor domain. Whereas in wt Cin8 three H-bonds were formed in this region, in variants with partial activity (Cin8-G522N and Cin8_NLEg5-NG) only two H-bonds were formed, and in the variant that was severely defective in intracellular functions (Cin8_NLCut7) only one H-bond was formed in this region. In addition, a conserved H-bond between the N-latch position (G522) and G157 in a loop between α1 and β3 of the motor domain appears to be critical for Cin8 function. Elimination of this bond resulted in a non-functional variant (Cin8_NLCut7-NG). Additional H-bonds stabilizing the N-latch docking to the motor domain (in Cin8_NLEg5) also resulted in a non-functional variant. In this variant, replacement of the N-latch asparagine by glycine (as is the case in Cin8) eliminated the additional H-bonds of the N-latch position and rescued the functional defects. Based on this analysis, we conclude that exact H-bond stabilization of the docked NL is critical for the function of Cin8.

Most importantly, the new analysis presented in Figure 3 and Table 2 diverts the discussion from focusing on the N-latch glycine to a more general view of H-bond stabilization involving additional amino acids of the NL (comment # 3 of Reviewer 1; comment # 4 of Reviewer # 2 and comment # 1 of Reviewer # 3). This more general view of NL docking in which the exact H-bond stabilization (at least of certain positions) is critical for the function of a bi-directional kinesin motor may be a general trait of bi-directional motors. As is mentioned above, this notion should be examined experimentally for additional bi-directional kinesin-motors.

2. For Cin8, the authors demonstrate that mutation of G in the N-latch position increases the doubling time but affects cell viability only at high temperatures. The authors conclude that "increased flexibility of the NL during docking of Cin8, compared to Eg5 (Figure 5) provides the necessary adaptation to allow bi-directional motility." (p.16, Discussion). However, the other bi-directional kinesin-5s (Kip1 and Cut7) have an N at the N-latch position and their NLs are thus presumably not flexible although the motors are functional and cells are viable.

As we emphasized in response to the previous point, based on the new structural analysis, in the revised version of the manuscript we do not focus on the N-latch glycine found in Cin8, but generally look at stabilizing H-bond formation between the docked NL and the motor domain. In fact, we observed that in the partially functional NL variants that contain an asparagine in the Nlatch position the number of stabilizing H-bonds (between the N-latch position and part of β10) is smaller than that in wt Cin8, which contains glycine in this position. This is true for the Cin8-G522N, Cin8_NLCut7 and Cin8_NLEg5-NG variants. Thus, it is not the glycine per se, but the arrangement of the H-bonds, that is important for Cin8 functionality. Based on our new structural analysis, we found a correlation between partial functionality of the variants and the smaller number of stabilizing Hbonds between the docked NL and the motor domain. We discuss this point in the Results section (pages 10-13) and in the discussion (pages 22-23).

3. On p. 14 in the Discussion, the authors state "The profound difference between the activity of Cin8NLEg5 and Cin8NLEg5-NG variants indicates that the glycine in position 522 is critical for the functions of Cin8." The problem is that this conclusion is based on a chimeric motor that does not occur naturally. In fact, if one considers Cin8 with its own NL sequence, both the WT and the G522N mutant Cin8 motors are functional and cells expressing these motors are viable under normal growth conditions. In addition, the other bi-directional kinesin-5s also have an N at this position and are functional.

We thank the reviewer for this comment and agree with the reviewer (and the other reviewers) that not all the intracellular phenotypes are completely consistent with the in vitro data and that in the previous version of the manuscript too much emphasis was put on the N-latch glycine. Based on the new analysis performed on the 3D models (Figure 3 and Table 2), additional amino acids from the NL are involved in the stabilization of the docked NL of some Cin8 mutants. Therefore, we discussed these results from a more general view point (rather than emphasizing only the N-latch glycine in Cin8) (pages 10-13).

That having been said, we also observed that over stabilization of the N-latch position, which is generated by replacing part of the NL sequence of Cin8 with sequences from the NL of the plus-end directed kinesin-5 Eg5, results in a non-functional variant. In this variant, a single replacement of the N-latch asparagine with glycine (as found in Cin8) decreases the number of stabilizing H-bonds at the N-latch position and rescues the functional defects. These data suggest that over stabilization of the N-latch position has a marked effect on the functionality of Cin8. We also found that complete elimination of H-bond stabilization of the N-latch position resulted in a non-functional variant. Thus, we conclude that exact stabilization of the N-latch position is critical for the function of Cin8 (page 11, last paragraph).

4. The relationship between G vs N in the N-latch position and motor motility and function is overall unclear. The authors focused on 3 mutant versions in motility assays: Cin8-G522N (has N in N-latch position), Cin8NLCut7 (has N in N-latch position), and Cin8NLEg5-NG (has G in N-latch position). In single-molecule motility assays, all show decreased affinity to the MT and processive minus end-directly motility albeit with decreased velocity as compared to the WT Cin8 motor (has G in N-latch position). In MT bundling assays, all showed a reduced ability to bundle MTs. But their functions are different as are their effects on cell viability.

For example, the NLCut7 and G522N mutants both have an N in the N-latch position. Both show decreased MT affinity, slower minus end-directed motility, and decreased MT bundling activity. Both localize to the SPB in early mitosis. However the NLCut7 mutant generates more monopolar spindles and is not viable at normal yeast growth temperature whereas the G522N mutant generates normal spindle distributions and is viable. It thus appears that having a N in the N-latch position does not correlate with viability.

As a second comparison, the Eg5-NG mutant (G in N-latch position) and the G522N (N in N-latch position) both display decreased MT affinity and slower minus end-directed motility. The Eg5-NG mutant shows severely reduced MT bundling activity and the G522N mutant shows reduced MT bundling activity. Both localize to the SPB in early mitosis and generate normal spindle distributions. And both are viable. It thus appears that the residue in the N-latch position does not correlate with cellular function.

In view of the new analysis that we performed, we completely agree with the reviewer that the presence of asparagine in the N-latch position per se does not correlate with cell viability. In the revised version of the manuscript, we show that cell viability correlates with decreased H-bond stabilization of the docked NL. For example, both Cin8-G522N and Cin8_NLEg5-NG exhibit reduced Hbond stabilization between the docked NL and β7 of the motor domain (Figure 3 and Table 2), and viability of cells expressing these variants as the sole source of kinesin-5 is reduced (Figure 1C). In Cin8_NLCut7 the H-bonds between the docked NL and β7 are eliminated completely, with only one Hbond remaining between the N-latch position and D137, and cell viability is reduced dramatically (Figure 1C). Based on this result, we concluded that the stabilization between docked NL and β7 is important, but not critical for Cin8 function (page 10 and page 11, the first paragraph).

In addition, to address the important point regarding the correlation between motor motility and cell viability, we performed an additional analysis of the run length of single molecules of wt Cin8 and its NL variants (Figure S4). We found that only the Cin8_NLCut7 variant exhibits a significantly shorter run length compared to wt Cin8. In addition, the Cin8_NLCut7 variant requires a high protein concentration to produce efficient antiparallel MT sliding (Figure 6B), indicating less efficient sliding. Thus, although cell viability is a complex phenomenon, we propose that the significantly reduced viability of cells expressing Cin8_NLCut7 results, at least in part, from a combination of different motile defects, which correlate with the reduced H-bond stabilization of NL docking (page 16, last paragraph and page 22, lines 17-25).

Finally, we might have mistakenly implied, in the previous version of the manuscript, that N-latch glycine is the only important factor controlling NL dynamics and Cin8 function, but very likely, this is not the case. In the revised version of the manuscript, we have clarified this point and have emphasized the contribution of other amino acids in the NLs of the mutants that affect NL docking dynamics and function in vivo and in vitro (pages 10-13).

5. In my opinion, it would be most interesting to know whether the NL replacements or mutations alter the bi-directional behavior of Cin8. This is not clear to me from Figure 3. Can the authors provide information about the frequency of each type of motion (fast minus end, diffusive, slow plus end) for each motor at each salt concentration? While all of the mutant motors examined in these assays are slower than the WT motor, their frequency of transitioning to plus end growth is unclear. What salt condition was utilized for the data in 3D?

We thank the Reviewer for this comment. To address this important comment, we examined the directionality of antiparallel MT sliding induced by wt Cin8 and the different NL variants. It has previously been shown that one of the conditions under which Cin8 reverses directionality from minus- to plus-end directed motility occurs when crosslinking two antiparallel MTs and mediating their sliding apart by plus-end directed motility on the two MTs (Gerson-Gurwitz et al., 2011; Roostalu et al., 2011; Shapira et al., 2017). In addition, we note that – based on our recently published model (Shapira et al., 2017) – directionality reversal to plus-end directed motility during antiparallel MT sliding is an important motor function essential for mitotic spindle assembly.

In the current study, MT sliding experiments were performed using polarity-marked MTs, as previously described (Shapira et al., 2017). Surprisingly, we observed two different MT sliding modes (Figures 6 and S5): plus-end directed sliding, characterized by MT sliding with its unlabeled minus-end leading (Figure 6), and bi-directional/minus-end directed sliding, characterized by at least one episode of minus-end directed MT sliding, with its labeled plus-end leading (Figure S5). Our data indicates that all the functional variants that can support cell viability as single source for kinesin-5 function can produce plus-end directed antiparallel sliding. The occurrence of such sliding was similar for all the examined variants. These findings support the notion that plus-end directed antiparallel sliding is required for intracellular kinesin-5 function.

For the partially active variants, the motor protein concentration under which antiparallel MT sliding was observed was considerably higher than that for wt Cin8. This result is consistent with the lower ability of the NL variants to crosslink antiparallel MTs (Figure 5). Finally, the velocity of MT sliding was slowest when mediated by the Cin8_NLCut7 and Cin8_NLEg5-NG variants, which can account, in part, for the reduced functionality of these variants in cells (Figure 1C).

6. Does mutation of the N>G in the NL of Cut7 in the NLCut7 mutant rescue viability?

We thank the Reviewer for this interesting question. To address this point, we generated the Cin8_NLCut7-NG variant, which contains the NL of Cut7 but with the N-latch asparagine of Cin8replaced with glycine. Surprisingly, we found that, in contrast to the Cin8_NLEg5-NG variant, the Cin8_NLCut7-NG variant was completely non-functional in cells (Figure 1C); it accumulated with monopolar spindles (Figure 2A) and exhibited diffusive localization in the nucleus (Figure 2C-E). This variant also exhibited a significantly reduced MT-binding ability (Figure 4A). Examination of the H-bond configuration of this variant revealed that it failed to form a conserved backbone H-bond between the N-latch position and a glycine in a loop between α1 and β3 of the motor domain of Cin8 (G157) (Figure 3 and Table 2). Based on this result, we conclude that this conserved backbone H-bond is critical for Cin8 function (page 9 last paragraph, page 22, lines 26-31).

Reviewer #2:

The authors investigate the role of neck linker docking in the bidirection motility of Cin8. The N-latch idea is well established for kinesin-1 from Lang and Hwang's work. Most kinesin-5 have an N in this position, including two of the three bidirectional kinesin-5, but Cin8 has a glycine in this position. This is notable, because differences in neck linker docking are a good candidate for why fungal kinesin-5 are bidirectional.

The authors investigate chimaeras containing Eg5 or Cut7 neck linker replacement into Cin8. A number of assays are used including viability, in vivo localization, single-molecule velocity, and in vitro microtubule bunding. Pleasingly, the assays all give pretty consistent results. Cin8 doesn't like N replacing G, it can handle the Eg5 neck linker as long as you mutate N to G, and Cut7 NL works better than Eg5.

The paper is well written and the diversity of assays to demonstrate functionality is a strong point of the paper. The question of bidirectionality is important and still puzzling from a mechanistic standpoint.

We thank the Reviewer for these comments.

A weakness is that there is clearly a complex interaction between the neck linker, the cover bundle and the catalytic core of the head, and so it is difficult to interpret what these mutations might be doing to the complex mechanochemistry of these motors. That mutating it to N screws the motor up is notable, but doesn't shed a lot of light either on NL docking or on bidirectionality.

A smaller but important point is the use of the term "flexibility". Generally glycines increase flexibility of polypeptides. But that isn't actually rigorously shown here. And the effect of the glycine may not have anything to do with flexibility, and rather may all be about hydrogen bonds. In the discussion, the authors bring up a kinesin-1 with longer and hence more flexible neck linker that can be made to step backwards, but that is adding sequence and hence flexibility at the distal part of the neck linker that doesn't dock, so I don't think this comparison is necessarily apt.

We fully agree with the Reviewer that the interaction of the NL with the catalytic motor domain is complex and that focusing on the N-latch glycine/asparagine of Cin8 was an oversimplification. We also agree with comment #2 in that the term “flexibility” was poorly defined and characterized and that stabilization by H-bonds should have been considered in the previous version of the manuscript. To address this issue (and similar comments raised by Reviewers #1 and #3), we generated homology models of the Cin8 motor domain of wt Cin8 and its NL mutants (Figure 3, S2 and Table 21). Our modeling is based on four PDB structures published previously for kinesin motors; please see a detailed description in our response to comment #1 of Reviewer #1. We used the generated model of the Cin8 motor domain to calculate the H-bonds between the NL (⁵¹³KNIKNKPQLGSF⁵²⁴) and the motor domain (Figure 3, S2 and Table 2 of the revised manuscript).

Our analysis revealed that in wt Cin8 one conserved backbone H-bond is formed between the Nlatch G522 and G157. Based on the study of Hwang at al. (2008), two backbone H-bonds are formed between the N-latch asparagine and the motor domain, one with a glycine homologous to G157 of Cin8 and one with β7 of the motor domain. This suggests a reduced H-bond stabilization of the Nlatch position of Cin8, compared to kinesin-1. Importantly, our analysis revealed that the number of H-bonds formed between the NL and the motor domain is different from that between wt Cin8 and NL-mutants. In all the partially functional variants, the number of H-bonds formed between the Nlatch position and β10 of the NL and the motor domain is smaller than that in wt Cin8. In the Cin8_NLCut7 variant, which exhibits severely defective activity (Figure 1,2), no H-bonds are formed in this region, indicating that these H-bonds are important but not critical for Cin8 function. Finally, our analysis showed that complete elimination of H-bonds at the N-latch position and overstabilization of this position in the docked NL led to the production of the non-functional variants, Cin8_NLCut7-NG and Cin8_NLEg5, respectively. In the Cin8_NLEg5 variant, replacement of the N-latch asparagine with glycine, as in Cin8, eliminated additional H-bonds at this position and rescued the majority of functional defects. Thus, we conclude that exact stabilization of the N-latch position is critical for the function of the bi-directional Cin8.

Finally, regarding the flexibility of NL docking and bi-directional motility of Cin8: Based on our modeling and H-bond analysis, we observed reduced H-bond stabilization of Cin8 compared to kinesin-1 (Hwang et al., 2008). In addition, overstabilization of this position by additional H-bonds resulted in a non-functional variant. Therefore, a certain degree of flexibility is important in this position. Although this “flexibility” may be different for different bi-directional kinesins, in general terms, the reduced stabilization of NL docking that we observed in Cin8 can, in fact, constitute the molecular mechanism required for the bi-directional stepping (see the last section of the Discussion).

It is notable that the three key residues that interact with N in kinesin-1 are conserved in Cin8, but that means we don't really understand the glycine is doing in wild-type.

Although in amino acid sequence alignments, presented in the previous version of the manuscript, it appears that glycine occupies the N-latch position only in Cin8 (Figure 2), in fact, there are other kinesin-5 motors that contain glycine in this position (Figure S1). Therefore, the mechanism of NL docking and dynamics applicable to Cin8 is not unique to Cin8 and probably represents a more general mechanism. In addition, as discussed in response to the previous point, the N-latch glycine of Cin8 does form one of the backbone H-bonds reported for kinesin-1 (Hwang et al., 2008), which indicates that, to some degree, the molecular mechanism involving the N-latch asparagine in kinesin-1 is recapitulated in the N-latch glycine of Cin8.

The changes in speed are notable, but the mechanism of stepping is too complicated to just attribute the changes to neck linker docking. Because neck linker docking is so fundamental to the chemomechanical cycle, changing these interactions could be changing a lot about the motor.

We agree with the Reviewer that the H-bond stabilization mechanism discussed here (Figure 3 and Table 2) – or the “flexibility” discussed in the previous version of the manuscript – do not exclude other structural-functional mechanisms that contribute to NL docking and dynamics. For example, the formation of a β-sheet cover strand between the non-motor N-terminal sequences and the β9 of the NL is probably important in the stepping of Cin8, similarly to other kinesin motors (Goulet et al., 2012; Goulet et al., 2014; Hwang et al., 2008; Khalil et al., 2008; von Loeffelholz et al., 2019) (please see page 10, lines 12-28). Based on our new structural analysis (Figure 3 and Table 2), there are differences in the H-bond arrangement between wt Cin8 and the NL variants. In all the partially functional variants (Cin8-G522N, Cin8_NLEg5-NG and Cin8NLCut7), the number of H-bonds formed in this region is smaller, indicating a correlation between H-bond stabilization and Cin8 functionality.

In addition, the new results presented in the revised manuscript indicate that the exact stabilization of the N-latch position is critical for Cin8 function, with elimination and over-stabilization in this position resulting in non-functional Cin8 variants (Cin8_NLEg5 and Cin8_NLCut7-NG). In the nonfunctional CIn8_NLEg5, a single replacement of the N-latch asparagine with glycine, as is present in Cin8, reduces the number of stabilizing H-bonds of the N-latch position and rescues the majority of functional defects in vivo and in vitro. Thus, although other factors undoubtedly affect motor stepping, the strong correlation between H-bond formation and functionality of Cin8 indicates that this is an important factor controlling NL dynamics, motor activity and intracellular functions. As demonstrated here for Cin8, these principles may apply to other bi-directional kinesin motors.

Overall, this is important work for the field and to add another detail to the question of how fungal kinesin-5 can walk bidirectionally. However, rather than making us think differently about the problem, it is contributing important new details. That limits the impact of the study in my eyes.

The new analysis that we performed (Figure 3 and Table 2) demonstrates how H-bond formation regulates the functionality of a bi-directional kinesin. The current paper is thus the first report of such a structural analysis, coupled with experimental data, for a bi-directional kinesin. Our work establishes two new rules for NL-docking of Cin8: (1) the H-bonds formed between the NL and β7 of the motor domain are important but not critical for the function of Cin8; and (2) the exact Hbond stabilization of the N-latch position is critical for the function of Cin8. These factors, especially the second one, are likely to be important for bi-directional stepping. Cin8 serves as an example of a bi-directional kinesin, and more work on additional bi-directional motors is required to establish the generality of these principles (please see the last section of the Discussion). In this respect, the current study is not just another interesting detail, but adds to our understanding of what is required for bi-directional stepping.

Reviewer #3:

While kinesin motors have been intensively studied for over thirty years, and while there is consensus on mechanisms underlying "+" end directed movement and processivity, exceptions to the current "rules" exist, most notably concerning the ability of some kinesin 5 family members to move processively in both the "+" and "-" directions. This remarkable feature, which has been known for over a decade, begs a structure-based explanation. In this manuscript, the authors use a mutational approach to examine the structural features needed for bi-directional processive movement in Cin8. Their studies, spanning cell biology to single molecule motillity assays lead them to conclude that bi-directional motility requires "flexibility" in the neck linker, which in Cin8 is provided by the presence of a glycine in a position that in plus end directed kinesins is occupied by an apsaragine. They hypothesize that this single residue difference destabilizes the neck linker from docking along the motor surface toward the "+" end and increases said flexibility.

While the experiments appear to be well done and the text is clearly written, I have several major editorial concerns that I believe need to be addressed in a revised manuscript:

We thank the Reviewer for this comment and for stating that we should submit a revised version of this manuscript.

1. NL flexibility: The authors are vague as to what they mean here. One presumes, from their citing of literature that proposes an ordered to disordered NL transition that this flexibility implies increased disorder in the NL which somehow translates to "-" directed motion. However, I note that cryoEM reconstructions of several "+" directed kinesins, including Eg5 and Kif20A, show that the NL is in fact ordered in both pre- and post-hydrolytic states, pointing towards the "+" end in the former and the "-" end in the latter. In view of this, it seems much more likely that there is rather a dynamic equilibrium between two orientations in the NL (minus and plus directed) and that while the NL asparagine drives this equilibrium toward a structured (non flexible) NL in a plus orientation, its loss drives this equilibrium toward a similarly structured, non flexible NL pointing in the minus direction. This is not a minor point, since the term flexibility that the authors use here could mean NL conformational flexibility (ordered versus disordered) or directional flexibility (minus versus plus directed). I believe that the authors need to be considerably clearer about what they mean by "flexibility" and how they envision the loss of the stabilizing hydrogen bonds leading mechanistically to "-" directed motion.

We fully agree with the Reviewer that the term “flexibility” was poorly defined and characterized by us. To address this comment (and similar comments of the other Reviewers), we have now performed additional structural analysis, based on homology modeling of the Cin8 motor domain, to examine H-bond formation between the NL (⁵¹³KNIKNKPQLGSF⁵²⁴) and the motor domain. This analysis reveals that, as the reviewer suggested, addition of H-bonds stabilizing the docked N-latch position (such as in the Cin8_NLEg5 variant) correlates with the abolished functionality (Figure 1, 3 and Table 1 and 2). (Please also see our response to comment # 1 of Reviewer # 1 and comments # 1 and 2 of Reviewer # 2). The new analysis also reveals that there is a correlation between decrease in stabilizing H-bonds between β10 of the NL and β7 of the motor domain and decreased functionality (please see pages 10-13). These results shift the emphasis to a broader view of H-bond formation. We did find that in the non-functional Cin8_NLEg5 variant, which forms additional H-bonds to stabilize the N-latch position, replacing the N-latch asparagine with glycine, as is present in Cin8, reduces the number of stabilizing H-bonds and rescues the majority of functional defects. These results, in fact, support the view outlined here by the Reviewer that there might be a dynamic equilibrium between two conformations of the NL and that additional H-bonds in the non-functional variant shift this equilibrium towards one of the conformations, which interferes with the function of Cin8. In the revised manuscript, we discuss this mechanism suggested by the Reviewer (Page 24, lines 12-20).

2. Importance of NL glycine: Given the importance the authors place in Cin8 glycine 522 in determining directional flexibility, it is striking that two other bi-directional kinesin 5 family members (Kip1 and Cut7) have an arginine at this position. The authors cite a a cryoEM model (which incidentally while included in the references as #29 is not cited in the text of the discussion) indicating that the NL of cut7 is not stabilized by NL-motor hydrogen bonds to the same degree as wild type Cin8. However, I am not sure that a 4.5 Å resolution map can definitively allow for this conclusion, and I believe that the authors need to expand their discussion of this point, including the potential limitations of their interpretation in this light.

We thank the Reviewer for this question. To address this point and to obviate reliance on one structure of 4.5Å resolution, we generated 3D models of the Cin8 motor domain, including the NL (Figure 3 of the revised manuscript). The models were based on four PDB structures published previously for kinesin motors, all in the presence of AMP-PNP, namely, conditions under which the NL is docked and points in the plus-end direction of the MT: (a) 3HQD, X-ray structure of Eg5 motor domain (Parke et al., 2010); (b) 1VFV, X-ray structure of KIF1A kinesin heavy chain isoform 5C (Nitta et al., 2004); (c) 3ZFC, X-ray structure of KIF4A isoform 4A (Chang et al., 2013); and (d) 6S8M, cryo-EM structure of S. pombe kinesin-5 Cut7 decorating MTs (von Loeffelholz et al., 2019). Such modeling increases the reliability of our conclusions. (Please also see our response to point # 1of reviewer # 1).

It should be noted that there are other kinesin-5 motors that contain glycine at the N-latch position (Figure S1), which suggests that the NL functionality involving this glycine may be more general and not apply only to one protein. However, in the revised version of the paper, we focus on the stabilization of the NL via H-bond formation and not on a specific sequence. In fact, we show that based on our models, some variants that contain N-latch asparagine, instead of the glycine that is present in Cin8, form a smaller number of H-bonds between NL and β7, compared to wt Cin8 (such as Cin8-G522N, Cin8_NLCut7 and Cin8_NLEg5-NG; Figure 3 and Table 2 of the revised version). Although we believe that the high “sensitivity” to precise H-bond stabilization of certain positions in the NL that we have demonstrated here for Cin8 is a common trait of bi-directional kinesin motors, it is likely that the precise details and the amino acids that participate in such stabilization are different for the different motor proteins. Additional experiments are needed to examine this mechanism for other bi-directional kinesin motors.

Finally, the authors note that mutating the N latch in kinesin 1 increases motor velocity. This is a striking finding that appears to be consistent with their own studies of Cin8. Certainly one possibility is that hydrogen bonding between the NL and the motor core slows NL undocking, which becomes rate limiting in the single molecule assay. Regardless, some mechanism needs to be proposed to explain this besides the relatively uninformative statement that "stabilization of the NL docking plays similar role in controlling the velocity of plus- and minus-end directed N terminal kinesin motors.

We fully agree with the Reviewer on this point. We have performed structural analysis in order to be more precise in our statements and claims regarding stabilization (vs. flexibility) of certain configurations of the NL. We also referred to the mechanism suggested by this Reviewer in the first comment to explain the reduction in minus-end directed velocity when the docked NL is stabilized by H-bonds (Page 24, lines 12-20).

Barber-Zucker, S., R. Uebe, G. Davidov, Y. Navon, D. Sherf, J.H. Chill, I. Kass, R. Bitton, D. Schüler, and R. Zarivach. 2016. Disease-Homologous Mutation in the Cation Diffusion Facilitator Protein MamM Causes Single-Domain Structural Loss and Signifies Its Importance. Sci Rep. 6.

Chang, Q., R. Nitta, S. Inoue, and N. Hirokawa. 2013. Structural basis for the ATP-induced isomerization of kinesin. Journal of molecular biology. 425:1869-1880.

Fodor, J., B.T. Riley, I. Kass, A.M. Buckle, and N.A. Borg. 2019. The Role of Conformational Dynamics in Abacavir-Induced Hypersensitivity Syndrome. Sci Rep. 9:019-47001.

Gerson-Gurwitz, A., C. Thiede, N. Movshovich, V. Fridman, M. Podolskaya, T. Danieli, S. Lakamper, D.R. Klopfenstein, C.F. Schmidt, and L. Gheber. 2011. Directionality of individual kinesin-5 Cin8 motors is modulated by loop 8, ionic strength and microtubule geometry. The EMBO journal. 30:4942-4954.

Goulet, A., W.M. Behnke-Parks, C.V. Sindelar, J. Major, S.S. Rosenfeld, and C.A. Moores. 2012. The structural basis of force generation by the mitotic motor kinesin-5. J Biol Chem. 287:4465444666.

Goulet, A., J. Major, Y. Jun, S.P. Gross, S.S. Rosenfeld, and C.A. Moores. 2014. Comprehensive structural model of the mechanochemical cycle of a mitotic motor highlights molecular adaptations in the kinesin family. Proc Natl Acad Sci U S A. 111:1837-1842.

Hwang, W., M.J. Lang, and M. Karplus. 2008. Force generation in kinesin hinges on cover-neck bundle formation. Structure. 16:62-71.

Kass, I., D.E. Hoke, M.G.S. Costa, C.F. Reboul, B.T. Porebski, N.P. Cowieson, H. Leh, E. Pennacchietti, J. McCoey, O. Kleifeld, C. Borri Voltattorni, D. Langley, B. Roome, I.R. Mackay, D. Christ, D. Perahia, M. Buckle, A. Paiardini, D. De Biase, and A.M. Buckle. 2014. Cofactor-dependent conformational heterogeneity of GAD65 and its role in autoimmunity and neurotransmitter homeostasis. Proc Natl Acad Sci U S A. 111:E2524-2529.

Khalil, A.S., D.C. Appleyard, A.K. Labno, A. Georges, M. Karplus, A.M. Belcher, W. Hwang, and M.J. Lang. 2008. Kinesin's cover-neck bundle folds forward to generate force. Proceedings of the National Academy of Sciences. 105:19247-19252.

Nitta, R., M. Kikkawa, Y. Okada, and N. Hirokawa. 2004. KIF1A alternately uses two loops to bind microtubules. Science. 305:678-683.

Parke, C.L., E.J. Wojcik, S. Kim, and D.K. Worthylake. 2010. ATP Hydrolysis in Eg5 Kinesin Involves a Catalytic Two-water Mechanism. J Biol Chem. 285:5859-5867.

Roostalu, J., C. Hentrich, P. Bieling, I.A. Telley, E. Schiebel, and T. Surrey. 2011. Directional switching of the Kinesin cin8 through motor coupling. Science. 332:94-99.

Shapira, O., A. Goldstein, J. Al-Bassam, and L. Gheber. 2017. A potential physiological role for bidirectional motility and motor clustering of mitotic kinesin-5 Cin8 in yeast mitosis. J Cell Sci. 130:725-734.

Singh, S.K., H. Pandey, J. Al-Bassam, and L. Gheber. 2018. Bidirectional motility of kinesin-5 motor proteins: structural determinants, cumulative functions and physiological roles. Cell Mol Life Sci:1757-1771.

von Loeffelholz, O., A. Pena, D.R. Drummond, R. Cross, and C.A. Moores. 2019. Cryo-EM Structure (4.5-A) of Yeast Kinesin-5-Microtubule Complex Reveals a Distinct Binding Footprint and Mechanism of Drug Resistance. Journal of molecular biology. 431:864-872.

Author details

1. Alina Goldstein-Levitin

   Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva, Israel

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Methodology, Writing - original draft, Writing - review and editing

   Competing interests

   No competing interests declared

2. Himanshu Pandey

   Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva, Israel

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Writing - original draft, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0629-7525

3. Kanary Allhuzaeel

   Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva, Israel

   Contribution

   Formal analysis, Investigation

   Competing interests

   No competing interests declared

4. Itamar Kass

   1. Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva, Israel
   2. InterX LTD, Ramat-Gan, Israel

   Contribution

   Investigation, Writing - original draft

   Competing interests

   is affiliated with InterX LTD. The author has no other competing interests to declare.

5. Larisa Gheber

   Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva, Israel

   Contribution

   Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Validation, Methodology, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   lgheber@bgu.ac.il

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-3759-4001

• 637

  Page views

• 110

  Downloads

• 1

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.