Intellectual disability-causing mutations in KIF11 impair microtubule dynamics and dendritic arborization

Wingfield, Jenna L.; Niese, Lukas; Avchalumov, Yosef; Liu, Xiaodan; Grover, Rahul; Nakahata, Yoshihisa; Raveendra, Bindu L.; Gonzalez-Santiago, Adrian E.; Carter, Jackson P.; Yasuda, Ryohei; Yuan, Jason X.-J.; Diez, Stefan; Puthanveettil, Sathyanarayanan V.

doi:10.1038/s41467-026-70522-z

Download PDF

Article
Open access
Published: 18 March 2026

Intellectual disability-causing mutations in KIF11 impair microtubule dynamics and dendritic arborization

Nature Communications volume 17, Article number: 4125 (2026) Cite this article

4662 Accesses
1 Citations
6 Altmetric
Metrics details

Subjects

Abstract

Microcephaly with or without chorioretinopathy, lymphedema, or intellectual disabilities (MCLID) is a rare disease caused by mutations in the mitotic motor KIF11. However, the specific neuronal functions of KIF11, its mechanisms of microtubule (MT) regulation, and the impact of MCLID mutations on KIF11 function remain underexplored. Here, using live-imaging, we find that KIF11 depletion in postmitotic neurons increases minus-end-out MT dynamics in both axons and dendrites. Introducing MCLID-associated KIF11 mutations, KIF11^Y82F and KIF11^ΔCterm, significantly reduces MT dynamics, impairs dendritic arborization, and decreases mEPSC frequency. Biochemical analyses reveal that the KIF11^ΔCterm mutant disrupts tetramer formation and MT crosslinking, while the KIF11^Y82F mutant reduces MT sliding velocity and ATP affinity. Temporal inhibition of KIF11 using a photo-controllable KIF11 increases MT dynamics and dendritic growth. Together, these data reveal that KIF11 is a MT dynamics rheostat and regulator of dendritic arborization in mature neurons, providing essential insights into the molecular mechanisms driving MCLID.

Cryo-EM unveils kinesin KIF1A’s processivity mechanism and the impact of its pathogenic variant P305L

Article Open access 02 July 2024

Pathogenic morphological signatures of perturbations in mitochondrial-related genes revealed by pooled imaging assay

Article Open access 01 August 2025

Microglia mitochondrial complex I deficiency during development induces glial dysfunction and early lethality

Article 24 July 2024

Introduction

Neurons, among the most asymmetric and largest cell types, rely on precise cytoskeletal control and directed long-distance transport for intercellular communication and proper brain wiring. Several studies, including our own, have demonstrated that kinesin superfamily (KIFs) molecular motor proteins are key modulators of both intra- and intercellular communication^{1,2,3,4,5,6,7,8,9,10}. These proteins play crucial roles in maintaining neuronal health, as well as in brain development and function¹¹. Most of the KIFs studied in neurons, particularly those involved in learning and memory, are kinesins directly responsible for the long-distance microtubule (MT) dependent transport of molecular and organelle cargos^1,12,13. However, emerging research indicates that certain KIFs, traditionally studied in the context of cell division, are also expressed and functional in mature neurons^14,15. However, the mechanisms underlying their roles in these post-mitotic cells remain poorly understood.

Elucidating the functions and mechanisms of mitotic KIFs in post-mitotic neurons is expected to uncover novel pathways that govern KIFs and neural circuit function. This line of research holds promise for advancing our understanding of disease mechanisms and could open new avenues for therapeutic development.

To address this significant knowledge gap, we investigated the mechanisms and functions of the mitotic KIF, KIF11, in hippocampal neurons. Notably, mutations in KIF11 (Kinesin-5/EG5) have been linked to microcephaly with or without chorioretinopathy, lymphedema, or intellectual disabilities (MCLID)^16,17,18. MCLID is a rare autosomal dominant disorder characterized by the phenotypic features that make up its name^16,19,20. Patients with MCLID are heterozygous, and disease-relevant Kif11 mutations, predicted to be loss-of-function, occur throughout the Kif11 gene. However, these predictions have not yet been experimentally confirmed. In rodent models, homozygous KIF11 knockout (KO) mice die before the blastocyst stage, whereas heterozygous mice develop normally, suggesting that Kif11 haploinsufficiency alone is unlikely to account for the observed disease phenotypes^21,22. Consequently, the mechanism by which Kif11 mutations lead to intellectual disabilities remains elusive.

The pathomechanism of MCLID has been largely attributed to KIF11’s role in mitosis, although this has not been thoroughly tested. KIF11 is a homotetrameric kinesin that generates force on MTs by binding them on both sides, rather than transporting cargo in dividing cells^23,24. During mitosis, KIF11 binds interpolar MTs and acts as a brake on other mitotic kinesins, ensuring proper spindle formation^25,26,27. Disruption in KIF11 function results in monopolar spindles and polyploidy^28,29. KIF11 binds both antiparallel MTs in the spindle and parallel astral MTs²⁹. In vitro studies reveal that KIF11 slides antiparallel MTs but only crosslinks parallel MTs^30,31. Although KIF11 is downregulated at the end of mitosis, it has also been found to be expressed in postmitotic neurons^32,33. Several studies have demonstrated roles for KIF11 in axon migration, branching, and dendritic arborization during maturation of sympathetic and hippocampal neurons^34,35,36. Despite the growing evidence of KIF11’s importance in developing neurons, the mechanism of its action in mature neurons and its potential contribution to MCLID remain unclear.

In a screen aimed at identifying MT motors involved in excitatory synaptic transmission in mouse primary hippocampal neurons, we unexpectedly found that Kinesin-5 (KIF11) acts as a repressor of excitatory synaptic transmission². Specifically, knockdown of KIF11 in mature neurons led to an increase in the frequency of excitatory post-synaptic currents, enhanced dendritic arborization, and a higher number of dendritic spines. Additionally, we observed a decrease in KIF11 expression accompanied by elevated levels of synaptic vesicle proteins, such as Synaptophysin and Piccolo. Despite these findings, the mechanism by which KIF11 restrains dendritic growth and presynaptic vesicle release in mature neurons remains unclear. Given the structural and electrophysiological alterations resulting from KIF11 perturbations in primary hippocampal cultures, and the limited understanding of how KIF11 mutations lead to intellectual impairment in MCLID, we aimed to gain mechanistic insight into KIF11’s role in mature neurons.

Using live imaging of MT dynamics and neuronal cargoes in KIF11 knockdown neurons, we identified KIF11 as a crucial rheostat of MTs and intracellular transport in mature neurons. Pharmacological inhibition of KIF11 suggests that it stabilizes MTs by binding and crosslinking parallel, rather than antiparallel, MTs in mature neurons. Furthermore, the expression of two distinct MCLID-associated mutations linked to intellectual disabilities suggests that while these mutant forms of KIF11 retain their MT-binding capability, though with impaired tetramer formation in the case of one mutant and reduced MT sliding capacity in the other, they negatively impact MT dynamics and dendritic complexity, mEPSC frequency, and arborization in vivo. Finally, the development of an optically controllable KIF11 revealed that temporal inhibition of KIF11 increases MT dynamics and dendritic growth, whereas temporal activation exerts force on MTs, diminishing both MT dynamics and dendritic growth. Collectively, these data establish KIF11 as a critical regulator of MT stability and dendritic arbor integrity in mature neurons. Additionally, they provide critical insights into the pathomechanism of MCLID, in which mutant KIF11 proteins exert unrestrained forces on MTs, destabilizing them and leading to a consequent reduction in dendritic arbor complexity.

Results

KIF11 crosslinks, rather than slides microtubules, in mature neurons

To explore how KIF11 regulates dendritic arborization, we considered its known role in controlling cytoskeletal components, particularly microtubules (MTs) in neurons, similar to KIF11-mediated stabilization of spindle MTs during cell division. In neurons, MTs are differentially organized: axons contain predominantly plus-end-out MTs, whereas dendrites exhibit a mixed polarity of MTs (Fig. 1A).

We first confirmed the presence of KIF11 in the dendrites of mature neurons by detecting KIF11 mRNA and protein in hippocampal neurons (Supplementary Fig. S1A-C). KIF11 has been shown to bind both parallel and antiparallel MTs, with its activity depending on the orientation of the MTs: KIF11 crosslinks and slides MTs when bound to antiparallel MTs and crosslinks them statically when bound to parallel MTs. To gain functional insights into whether KIF11 slides or crosslinks MTs in mature hippocampal neurons, we used two pharmacological inhibitors of KIF11 in primary hippocampal neurons expressing Dendra-EB1, a reporter for MT dynamics (Fig. 1B). The inhibitors included Ispinesib, which prevents KIF11 from binding to MTs³⁷, and BRD9876, which locks KIF11 in a rigor state, bound to MTs³⁸ (Fig. 1C). We hypothesized that if Ispinesib alone altered MT dynamics, KIF11 must be crosslinking MTs, whereas if BRD9876 had an effect, KIF11 was likely also involved in sliding antiparallel MTs in mature dendrites.

We first evaluated MT flux and found that neither compound affected the plus-end-out MTs. However, Ispinesib treatment significantly increased the dynamics of minus-end-out MTs (Fig. 1D-E, Supplementary Movie S1). Next, we assessed MT growth rates and observed that, while neither compound affected plus-end-out MTs, Ispinesib treatment significantly increased the growth rate of minus-end-out MTs (Fig. 1F).

We further examined how MT dynamics were affected along the dendrites. Ispinesib treatment resulted in a decrease in plus-end-out EB1 comets in distal dendrites (60-70 µm from the soma) and an increase in minus-end-out EB1 comets in proximal dendrites (20-30 µm and 90-100 µm from the soma). In contrast, BRD9876 treatment only decreased plus-end-out EB1 comets at 60-70 µm from the soma (Fig. 1G-H). These data suggest that KIF11 limits the presence of growing minus-end-out MTs by crosslinking and limits the growth of parallel MT bundles in mature neurons (Fig. 1I).

KIF11 mediates EB3 dynamics in mature neurons

To confirm the role of KIF11 in regulating MT dynamics within mature neurons, we investigated MT behavior in KIF11 knockdown neurons. We transfected primary hippocampal neurons with a far-red-tagged end-binding protein 3 (miRFP703-EB3) to track growing MTs, alongside either a negative control construct (NC-GFP) or a previously validated shRNA construct targeting KIF11 (shKIF11)². Imaging was conducted approximately 24 h post-transfection (Fig. 2A). Using this approach, we visualized EB3 comets, representing both plus- and minus-end-out MTs, in the axons and dendrites of NC-GFP and shKIF11 neurons (Fig. 2B-E, Supplementary Movie S2).

**Fig. 2: KIF11 regulates EB3 dynamics in mature neurons.**

Our results revealed a significant increase in the percentage of minus-end-out MTs in shKIF11 axons compared to control neurons (Fig. 2F). A similar increase was observed in shKIF11 dendrites (Fig. 2G). Additionally, KIF11 knockdown led to an increase in minus-end-out MT flux in both axons and dendrites (Fig. 2H, I).

We further assessed MT growth rates, discovering a decrease in the minus-end-out EB3 growth rate in axons, whereas no significant change was observed in dendrites (Fig. 2J, K).

When examining the distance traveled by EB3, which reflects MT growth, no significant change was observed in axons (Fig. 2L). However, a significant increase in the growth of minus-end-out MTs was detected in dendrites (Fig. 2M). Collectively, these data suggest that KIF11 normally constrains minus-end-out MT dynamics in both axons and dendrites, thereby limiting arborization.

To determine whether these alterations in MT dynamics were specific to certain types of dendrites or distributed throughout the entire arbor, we conducted a segmented analysis of EB3 dynamics using the dataset from Fig. 2. We observed an increase in the percentage of dynamic minus-end-out MTs in secondary and tertiary dendrites, but not in primary dendrites (Fig. 2N). Further analysis revealed a significant increase in minus-end-out EB3 flux specifically in secondary shKIF11 dendrites (Supplementary Fig. S2A-H). This indicates that KIF11 specifically restricts the number of growing minus-end-out MTs within dendrites, particularly in secondary and tertiary branches, thereby limiting the formation of new branches.

KIF11 regulates synaptophysin flux and velocity in neurons

Our previous data demonstrate that functional depletion of KIF11 leads to significant destabilization of dendritic MTs. We next investigated whether this instability would affect the transport of cargoes that rely on these MTs for movement. Previous work from our lab revealed that KIF11 depletion resulted in the accumulation of synaptophysin (a presynaptic protein) without affecting PSD95 (a postsynaptic protein) levels, so we focused our analysis on these two MT motor cargos².

To examine synaptophysin dynamics, we co-transfected mature hippocampal neurons with mRuby-Synaptophysin³⁹ and either NC-GFP or shKIF11 constructs (Fig. 3A, Supplementary Movie S3) and focused on analyzing its dynamics within axons. KIF11 knockdown did not significantly impact the percentage of mobile mRuby-Synaptophysin puncta compared to the NC-GFP control (Fig. 3B-D). Nevertheless, mRuby-Synaptophysin anterograde flux and velocity did significantly increase with KIF11 knockdown, while retrograde flux, velocity, and distance traveled remained unaffected (Fig. 3B-C,E,F). These results suggest that depleting KIF11 causes synaptophysin accumulation in neurons by increasing the abundance and velocity of synaptophysin being transported into axons.

**Fig. 3: KIF11 regulates synaptophysin flux in neurons.**

Next, we assessed PSD95 dynamics by co-transfecting mature hippocampal neurons with mApple-PSD95 and either NC-GFP or shKIF11 constructs (Fig. 3H, Supplementary Movie S4) and focused our analysis within dendrites. We observed no change in the percentage of mobile mApple-PSD95 trafficking in KIF11 knockdown neurons nor its flux (Fig. 3I-L). Although KIF11 depletion reduced the anterograde and retrograde velocity of mApple-PSD95, this did not change the distance PSD95 trafficked (Fig. 3M,N). Consistent with our previous findings (in Swarnkar et al. 2018)², KIF11 depletion does not alter PSD95 abundance in dendrites.

Transient KIF11 inhibition disrupts the transport of both mitochondria and lysosomal vesicles

We next investigated whether functional depletion of KIF11 affects intracellular trafficking beyond synaptic proteins. To address this, we examined the transport of key organelles—mitochondria and lysosomal vesicles (LVs)—in mature neurons. Mitochondria and LVs are transported by various MT motors on both stabilized and dynamic MTs, making them ideal targets for assessing broader transport alterations.

To gain functional insights, we pharmacologically inhibited KIF11 using Ispinesib and BRD9876 (Supplementary Fig. S3A). We labeled mitochondria and LVs in DIV16 hippocampal neurons with MitoTracker Green and LysoTracker Red, respectively, to ensure complete labeling (Supplementary Fig. S3B, Supplementary Movie S5).

Our analysis of mitochondrial transport revealed that Ispinesib treatment decreased anterograde flux and increased retrograde flux, whereas BRD9876 did not produce significant changes compared to the DMSO control (Supplementary Fig. S3C-D). Given that BRD9876 did not significantly impact MT dynamics in our previous analyses, we next evaluated LVs using only DMSO and Ispinesib. Ispinesib treatment led to a trend of increased anterograde flux of LVs and significantly increased retrograde flux (Supplementary Fig. S3E-F, Supplementary Movie S6). Together, these findings indicate that KIF11 inhibition significantly disrupts the transport of both mitochondria and LVs in mature hippocampal neurons.

Intellectual disability causing mutations in KIF11 reduce the complexity of dendritic arborization and disrupt microtubule dynamics

We next sought to further tease apart the influence of KIF11 on dendritic arborization and MT dynamics by exploring the consequences of KIF11 overexpression in primary hippocampal neuronal cultures. Additionally, we wanted to assess the impact of KIF11 mutations associated with severe intellectual disabilities and minor microcephaly. We focused on two specific mutations: Y82F (Y81F in Mus musculus) and H768Qfs*7 (a frameshift mutation from a 2304_2305 deletion, which results in an early stop codon after 768, represented as KIF11^ΔCterm) (Fig. 4A)¹⁹. Both of these patients are heterozygous for their corresponding mutations.

**Fig. 4: KIF11 MCLID mutations impact dendritic arborization and microtubule dynamics.**

We began by examining dendritic arborization. Primary hippocampal neurons were transfected with either a control construct (NC-GFP), KIF11 overexpression construct (CAG-Dendra-KIF11), or constructs carrying the KIF11^Y81F or KIF11^ΔCterm murine versions of patient mutations. Neurons were imaged approximately 24 h later to assess dendritic morphology (Fig. 4B, C). Although these mutations are linked to microcephaly, no significant difference in soma area was observed among the conditions (Fig. 4D). Sholl analysis, which evaluates dendritic complexity based on the number of times dendrites intersect with concentric circles emanating from the neuronal soma, demonstrated that KIF11 overexpression, as well as the KIF11^Y81F or KIF11^ΔCterm mutations, led to a marked reduction in dendritic complexity, with the most severe effects observed in KIF11-OE and KIF11^Y81F (Fig. 4E). These findings, in combination with our previous study Swarnkar, et al.², further confirm the essential role of KIF11 in regulating dendritic arborization.

Given that KIF11 depletion increases MT dynamics, we hypothesized that KIF11 overexpression might decrease MT dynamics. To test this, we co-transfected neurons with miRFP703-EB3 and either NC-GFP, KIF11-OE, or the mutant constructs and imaged them ~24 h later (Supplementary Fig. S4A-E, Supplementary Movie S7). We observed a significant decrease in the flux of plus-end-out MTs in KIF11-OE compared to both NC-GFP and KIF11^ΔCterm neurons, and a slight but nonsignificant decrease in KIF11^Y81F (Fig. 4F). Additionally, the KIF11-OE and mutant neurons exhibited a reduced flux of minus-end-out MTs compared to NC-GFP (Fig. 4G). Although there were no statistically significant differences in the percentage of minus-end-out MTs between eGFP and the KIF11 constructs, KIF11^Y81F reduced the percentage of minus-end-out MTs compared to WT KIF11 (Fig. 4H). Furthermore, KIF11-OE reduced the plus-end-out MT growth compared to all other groups, while KIF11^Y81F reduced minus-end-out MT growth compared to eGFP and KIF11^ΔCterm (Fig. 4I,J). No differences were detected in the growth rate of plus-end-out MTs (Fig. 4K). However, KIF11-OE neurons exhibited an increased growth rate for minus-end-out MTs, suggesting that the few remaining growing MTs are destabilized (Fig. 4L). Overall, KIF11 overexpression and the KIF11^Y81F mutation significantly constrain MT growth dynamics.

During MT dynamics imaging in KIF11-OE and KIF11 mutant neurons, we noted numerous bent, twisted, and looped MTs with KIF11 expression but not with NC-GFP (Supplementary Fig. 4F-I). Previous in vitro studies suggested that KIF11 might act as an MT polymerizer, producing curved and looped MTs during polymerization⁴⁰. However, we observed that these loops often lacked EB3 comets and were not initiated by EB3 comets over more than 8 h of imaging. These observations imply that the loops are not due to new MT growth. Instead, we captured the process of MTs bending, twisting, and forming loops on themselves (Supplementary Fig. S4J, Supplementary Movie S8). This twisting likely reflects a combination of torque and tension applied to the MTs⁴¹. Additionally, we observed dead and dying neurons in KIF11-OE and KIF11^Y81F cultures with bent and fragmented dendrites, indicating that these MTs eventually succumb to the applied tension and break (Supplementary Fig. S3K-L). These findings align with the concept that KIF11 exerts rotational force when crosslinking antiparallel/parallel MTs, as seen in vitro with full-length KIF11 in a 3D MT sliding assay³⁰. In summary, the observation of twisted and looped MTs in KIF11 overexpression conditions supports the role of KIF11 as a MT crosslinker in neurons.

KIF11 overexpression and MCLID mutations reduce synaptic transmission

While we observed that KIF11 overexpression and MCLID mutation reduce dendritic complexity, it is unclear if these defects result in meaningful disruptions in neuronal communication. Therefore, to further understand the effects of Kif11 MCLID mutations on synaptic transmission, we measured miniature excitatory postsynaptic currents (mEPSCs) in primary hippocampal cultures expressing either control NC-GFP or KIF11 constructs (Fig. 5A). mEPSCs are spontaneous synaptic events driven by the neurotransmitter release from a single synaptic vesicle. It is measured by applying TTX, which blocks action potentials. Therefore, a change in the frequency of mEPSCs suggests alterations in pre-synaptic vesicular release, whereas a change in the amplitude corresponds to the availability/functionality of receptors on the post-synaptic terminal. mEPSCs analyses showed no changes in the amplitude between NC-GFP, wildtype KIF11, KIF11^Y81F, or KIF11^ΔCterm (Fig. 5B, C). However, analysis of mEPSC frequency revealed significant reductions in Kif11-OE, KIF11^Y81F, or KIF11^ΔCterm neurons compared to the NC-GFP control group (Fig. 5D). Cumulative probability analysis of the data shows these same differences in amplitude and frequency (Fig. 5E, F). Together, these results suggest that KIF11-OE and MCLID mutations suppress the frequency of presynaptic vesicle release.

MCLID mutations differentially impact KIF11 tetramerization and MT sliding

While we observed that both KIF11^Y81F and KIF11^ΔCterm reduced dendritic complexity, minus-end-out MT flux, and synaptic transmission, similar to WT KIF11, there were several notable differences (Fig. 4C-E,G, Fig. 5). KIF11^Y81F more significantly decreased minus-end-out growth, while KIF11^ΔCterm increased growth compared to WT KIF11 (Fig. 4J). Additionally, KIF11^ΔCterm allowed visualization of dendritic spine-like structures (Supplementary Fig. S4I), suggesting it is not as tightly bound to MTs and may diffuse into the cytoplasmic space. These observations indicate that while these mutants still retain some MT binding capabilities, KIF11 motility or degree of MT binding is likely differentially impaired.

To further understand the effect of MCLID mutations on KIF11 function, we sought to more closely look at where they occur and how they biochemically impact KIF11. Notably, the residue Y82 is positioned between the ATP-binding pocket and the tail-binding domain (Fig. 6A). Considering the importance of the coordination of tail association in the mechanochemical cycle of the motor domain, the conservative mutation Y82F may be enough to disrupt it, or ATP binding or hydrolysis⁴². Additionally, H768 residue (mutated in KIF11^ΔCterm), located within the bilateral assembly (BASS) domain, has been identified to participate in the packing of antiparallel helices⁴³. Thus, the KIF11^ΔCterm mutant is both mutated within the BASS domain and truncated before the tail domain, potentially impacting both the formation of the tetramer and the coordination of the mechanochemical cycling (Fig. 6B).

**Fig. 6: Intellectual disability-causing mutations differentially affect KIF11’s microtubule crosslinking and sliding capacity.**

To examine how MCLID mutations affect KIF11, we purified human KIF11 constructs (Supplementary Fig. S5A) and performed a series of biochemical assays. We first examined the ability of these monomers to form tetramers through Mass Photometry (Fig. 6C; Supplementary Fig. S5B-D). 99% of WT KIF11 formed tetramers (Fig. 6C; Supplementary Fig. S5B). Similarly, ~97% of KIF11^Y82F formed tetramers (Fig. 6C; Supplementary Fig. S5C). However, only ~7% of KIF11^ΔCterm were able to form tetramers, with the majority (~79%) forming dimers (Fig. 6C; Supplementary Fig. S5D). We next investigated if the MCLID mutations affected KIF11’s ability to bind MTs. To assess this, we immobilized fluorescently labeled MTs on a glass coverslip using antibodies, then added fluorescently tagged KIF11 constructs, and washed out excess materials (Supplementary Fig. S5E). Both WT and KIF11^Y82F were found to localize throughout fixed MTs at comparable concentrations (Supplementary Fig. S5F-H). In contrast, KIF11^ΔCterm showed reduced MT binding (Supplementary Fig. S5F-H). However, by increasing KIF11^ΔCterm concentration 10-fold, we observed increased MT binding (Supplementary Fig. S5F-H). This suggests that the motor domain’s MT binding capacity is not directly affected. Instead, the reduced binding likely results from impaired tetramer formation (Supplementary Fig. S5F-H). Next, we evaluated the ability of the different KIF11 mutants to crosslink MTs. To test this, we immobilized fluorescently labeled MTs on a glass coverslip as described above, then added untagged KIF11s and fluorescent cargo MTs. Unbound cargo MTs were then washed out by flushing ADP buffer (Fig. 6D). WT and KIF11^Y82F displayed similar abilities to overlap/crosslink the cargo MTs to the fixed MT (Fig. 6E,F). KIF11^ΔCterm (at 10-fold higher concentration) displayed reduced MT crosslinking compared to both WT and KIF11^Y82F (Fig. 6E,F).

These findings prompted us to explore whether MCLID mutations affect KIF11’s MT sliding capacity. Therefore, we performed a MT sliding assay (assay set-up shown in Fig. 6D in the presence of ATP) for the various KIF11 constructs. In the case of KIF11^ΔCterm, MT sliding was observed only at 10-fold higher concentration compared to the other two constructs. Interestingly, KIF11^Y82F exhibited significantly reduced MT sliding velocity compared to both WT KIF11 and KIF11^ΔCterm. Additionally, KIF11^ΔCterm showed higher variability in sliding velocity (Fig. 6G-H; Supplementary Movie S9). Given that its velocity was sixfold slower compared to other constructs, we hypothesized that the point mutation of KIF11^Y82F might result in impaired ATP binding or ATP hydrolysis. To investigate this, we examined how increasing ATP concentrations impacted MT sliding velocity for both WT and KIF11^Y82F constructs (Supplementary Fig. S5I,J). While we observed increased sliding velocity with increased ATP concentration, KIF11^Y82F still exhibited a lower maximum MT sliding velocity (4.7 nm/s) compared to WT KIF11 (17.5 nm/s) (Supplementary Fig. S5I,J). Additionally, by fitting the data with the Michaelis-Menten equation, we obtained significantly higher K_m for KIF11^Y82F compared to WT KIF11 (0.13 mM vs. 0.07 mM), indicating that KIF11^Y82F has a lower affinity for ATP compared to WT KIF11 (Supplementary Fig. S5I,J). In summary, while MCLID mutations still retain MT binding capabilities, they distinctly alter KIF11’s functionality on MTs; with KIF11^Y82F having reduced ATP binding and impaired stepping along MT, while KIF11^ΔCterm impairs tetramer formation and thus its MT crosslinking ability (Fig. 6I).

Expression of MCLID mutations in vivo impairs dendritic arborization with different severities

The electrophysiology effects and differential effects of MCLID mutations on tetramer formation and MT sliding further intrigued us about how KIF11 might contribute to disease pathology in patients. To gain a broad view of how these mutations impact dendritic arbor when in an in vivo developmental context, we used In Utero Electroporation (IUE) to express our eGFP, WT KIF11, KIF11^Y81F, or KIF11^ΔCterm in developing hippocampi in E15.5 in WT mice (Fig. 7A). While this approach does not allow us to examine the role of KIF11 haploinsufficiency, it does allow us to examine what happens in vivo when both WT KIF11 and patient mutations are present in the same neuron, as is probable in these heterozygous patients. We sought to look at the consequences of MCLID mutations early in development (i.e., after dendritic primary branching but before rapid elongation), so we isolated brains from postnatal day 7 (P7)⁴⁴. Additionally, we wanted to investigate the effects of MCLID mutations later in development (i.e., after secondary and tertiary branches have formed), so we also isolated brains from postnatal day 21 (P21) mice⁴⁴. We then fixed, cryoprotected, sectioned, and imaged these brains (Fig. 7A). We focused on pyramidal neurons within the dorsal CA1 of the hippocampus, a critical region for learning and memory (Supplementary Fig. S6A, B). At P7, via sholl analysis, we see that WT KIF11-OE significantly reduces dendritic arborization from 50 µm from the soma onward compared to eGFP (Fig. 7B, C, Supplementary Movie S10). KIF11^Y81F has reduced arborization from 10 µm to 20 µm from the soma compared to WT KIF11-OE, and increased arborization compared to WT KIF11-OE at 50 µm and KIF11^ΔCterm from 30 µm to 70 µm (Fig. 7B, C). In addition to the previously stated differences, KIF11^ΔCterm has reduced arborization from 20 µm to 70 µm from the soma (Fig. 7B, C). When we look at total branch points per neuron, we see that WT KIF11-OE increases branch points compared to all other conditions, but no difference between other groups (Fig. 7D). The reduced arborization quantified via sholl analysis, yet increased branch points for WT KIF11-OE may indicate that the branches are overall retracted and do not make it through the concentric circles of sholl analysis to be counted. Indeed, when we look at the length of these individual branches, we see that WT KIF11-OE has decreased branch length compared to eGFP and KIF11^ΔCterm, and KIF11^Y81F has reduced branch length compared to eGFP (Fig. 7E). However, when total branch lengths are examined, we find that all KIF11 conditions have reduced branch length sums (Fig. 7F). Collectively, we observe that at P7, WT KIF11-OE overexpression reduced dendritic complexity, with KIF11^ΔCterm having a more significant reduction, and KIF11^Y81F having almost no effect on dendritic complexity but a significant reduction in dendritic branch length.

**Fig. 7: MCLID mutations reduce hippocampal dendritic complexity in vivo.**

When we look later in development, at P21, we see some of these findings hold, while others change. Similar to at P7, at P21, WT KIF11-OE decreases dendritic complexity from 50 µm to 110 µm from the soma compared to eGFP (Fig. 7G,H, Supplementary Movie S11). KIF11^Y81F shows reduced dendritic complexity at 20 µm from the soma compared to eGFP but has improved dendritic complexity compared to WT KIF11-OE and KIF11^ΔCterm from 50 µm to 90 µm from the soma (Fig. 7G,H). KIF11^ΔCterm has significantly reduced dendritic complexity from eGFP and KIF11^Y81F from 40 µm to 90 µm from the soma (Fig. 7G,H). When we look at the number of branches per neuron, we see that KIF11^ΔCterm has significantly fewer branch points than all other conditions, while eGFP, WT KIF11-OE, and KIF11^Y81F are not significantly different from each other (Fig. 7I). When we quantify the lengths of these branches, WT KIF11-OE reduces individual branch length compared to eGFP and KIF11^Y81F (Fig. 7J). KIF11^Y81F has shorter branches than eGFP but longer branches than WT KIF11-OE and KIF11^ΔCterm (Fig. 7J). KIF11^ΔCterm has shorter branches than eGFP and KIF11^Y81F (Fig. 7J). When we quantify the sum of all the branches, WT KIF11-OE decreased total branch length from eGFP, while KIF11^ΔCterm decreased total branch length compared to both eGFP and KIF11^Y81F (Fig. 7K). In summary, overexpressing WT KIF11-OE in P21 neurons reduces dendritic complexity and branch length, KIF11^ΔCterm enhances the severity of this reduction, and KIF11^Y81F also reduces dendritic branch length but overall rescues dendritic complexity defects seen with WT KIF11-OE overexpression.

Local inhibition of KIF11 stimulates new dendritic growth

Our data reveal that KIF11 depletion increases MT dynamics and dendritic complexity, whereas KIF11 overexpression reduces both MT dynamics and dendritic complexity. These effects suggest that KIF11 plays a significant role in regulating MTs and dendritic arborization on a global scale. However, it remains unclear whether these changes result from KIF11’s direct or indirect impact on MTs and dendritic structure.

To better elucidate KIF11’s role in mature neurons, we sought to locally inhibit KIF11 and assess its effects on MT dynamics and dendritic branching. We employed the pdDronpaV system for this purpose. The pdDronpaV system allows for reversible dimerization of pdDronpaV molecules upon exposure to violet (405 nm) light and dissociation upon exposure to cyan (488 nm) light⁴⁵. By tagging KIF11 at the N-terminus with pdDronpaV, we hypothesized that the monomeric form of pdDronpaV would not affect KIF11’s function, similar to Dendra-KIF11 (Fig. 8A). However, dimerization of pdDronpaV would distort the motor and tail domains of KIF11, impairing its ability to bind MTs effectively (Fig. 8A).

**Fig. 8: Local inhibition of KIF11 stimulates new dendritic growth.**

We first tested the dimerization capability of pdDronpaV within the KIF11 tetramer by transfecting HEK293T cells with either the previously validated pdDronpaV-CDK5⁴⁵ or our pdDronpaV-KIF11, along with miRFP703-EB3. We observed that exposure to cyan light reduced the fluorescence of pdDronpaV-CDK5, while violet light increased the fluorescence intensity of pdDronpaV-KIF11, indicating successful dimerization (Fig. 8B-C, Supplementary Movie S12). Importantly, neither cyan nor violet light affected miRFP703-EB3 fluorescence, suggesting that changes in pdDronpaV fluorescence were due to dimerization rather than alterations in EB3 expression or fluorescence.

Next, we investigated the effects of temporally inhibiting KIF11 in primary hippocampal neurons by co-expressing pdDronpaV-KIF11 and miRFP703-EB3. Inhibition of pdDronpaV-KIF11 with violet light for 5 min resulted in a significant increase in the flux of growing plus-end-out and minus-end-out MTs (Fig. 8D-F, Supplementary Movie S13). This increased MT growth was accompanied by enhanced dendritic growth (Fig. 8G,H). Collectively, these findings demonstrate that KIF11 functions as a repressor of MT dynamics and dendritic growth in mature neurons.

Local activation of KIF11 results in twisted microtubules and dendritic branches

Our observations revealed that prolonged overexpression (>24 h) of KIF11 resulted in twisted and compressed MTs, prompting us to investigate how KIF11 might exert torque on these structures. Specifically, we sought to understand whether this might provide insight into the orientation of KIF11 binding to MTs. Pharmacological inhibition studies (Fig. 1) indicated that KIF11 binds to parallel MT bundles. However, the large loops and contraction of dendritic arbors observed with KIF11 overexpression (Supplementary Fig. S4F-I) suggest that KIF11 may also interact with anti-parallel MTs, showing sliding and thus more bending. Smaller kinks and loops, on the other hand, are more characteristic of mainly parallel MT bundles, as there is less sliding activity occurring (Fig. 9A).

**Fig. 9: KIF11 enacts torque on parallel microtubule bundles in mature neurons.**

To determine whether the observed effects were due to constitutive overexpression of KIF11 or if our pharmacological inhibition interpretations were incorrect, we acutely activated pdDronpa-KIF11 with cyan light for 10 min and assessed the impact on dendrites and MT dynamics (Fig. 9B, Supplementary Movie S14). Acute activation of KIF11 resulted in a significant reduction in the flux of both plus-end-out and minus-end-out MTs in the targeted dendrites (Fig. 9B-D). Additionally, this acute activation led to MT kinking and looping, similar to constitutive KIF11 overexpression (Fig. 9B, E-G). Notably, this acute activation did not produce the extensive twisted MTs seen with sustained overexpression (Fig. 9H). These findings suggest that KIF11 may predominantly affect parallel MTs in mature neurons, exerting torque that limits MT dynamics and contributes to aberrant dendritic arborization (Fig. 9I).

Discussion

Early in brain development, the organization of microtubules (MTs) and actin dynamics determines whether neurites become axons or dendrites. Axons align MTs with their plus-ends oriented away from the soma, while dendrites maintain a mixed MT polarity^46,47,48,49. As dendrites mature, their MTs become more organized into parallel bundles of like-MTs, guided by cytoskeletal organization and adhesion cues, such as Wnt signaling^47,50. This organization and stability of MTs and actin are crucial for maintaining and modulating dendritic arborization through both activity-independent and activity-dependent mechanisms^51,52. In mature mammalian dendrites, plus-end-out MTs exhibit dynamic behaviors, whereas minus-end-out MTs are stabilized^53,54,55,56. This orientation and stability are critical for regulating cargo trafficking and neuronal morphology^{7,53,54,57,58}.

KIF11 regulates minus-end-out microtubule dynamics

Our observations reveal that KIF11 knockdown increases retrograde EB3 dynamics, as evidenced by increased flux but reduced speed (Fig. 2H-J). This reduced speed suggests that the retrograde EB3 comets may originate from the minus end of plus-end-out MTs. Patronin/CAMSAP2, a minus-end binding protein, affects minus-end MT growth in Drosophila dendrites⁵⁹. However, the interaction between KIF11 and CAMSAP2 remains unclear. The observed reduction in growth rate should be more substantial—at least fourfold slower—to align with in vitro polymerization rates of the minus end⁶⁰. Another possibility is that KIF11 may normally prevent minus-end-out MTs from entering axons from the cell body, though this proposed mechanism of MT sliding has not been observed in neurons⁶¹. Given that primary dendrites were unaffected by KIF11 inhibition and the increase in minus-end-out MTs occurred more than 50 µm away from the cell body, this explanation seems unlikely (Fig. 1G,H; Fig. 2N). Additionally, KIF11 might restrict gamma-tubulin or other nucleating factors from entering axons and initiating minus-end-out MT growth. KIF11 is also found densely decorating distal MAP2-negative projections in our cultures (Supplementary Fig. S1), complicating the understanding of its role in this process. Further studies are needed to explore these possibilities.

KIF11 differentially regulates dendritic arborization throughout neuronal development

Our study found that KIF11 depletion in mature neurons (DIV14 or later) leads to increased dendritic arborization² and KIF11 overexpression decreases dendritic arborization (Fig. 2D-E; Fig. 4C-E). This contrasts with earlier studies showing decreased arborization following KIF11 loss-of-function in immature neurons (DIV7 or earlier)^34,36. However, others have also found KIF11 overexpression reduces dendritic complexity in mature neurons⁶². We attribute this discrepancy to differences in MT organization between maturing and established arbors. In both cases, dendrites exhibit mixed polarity, but in mature neurons, MTs are organized into parallel bundles⁵³. Early in development, KIF11 likely binds to antiparallel MTs, promoting neurite sliding and outgrowth, similar to its role at the mitotic spindle²⁵. In mature neurons, KIF11 interacts with parallel MTs, where it cannot promote sliding but may induce rotational forces³⁰. These differences may be why we see that expression of the Y82F mutant, which retains its crosslinking ability but not sliding ability, has less severe impacts on the dendritic arbor than WT KIF11 when expressed earlier in dendritic development (In vivo- Fig. 7) but demonstrates a more severe effect when added after tertiary branches have been established (In vitro-Fig. 4). Additionally, given that stable, acetylated MTs are positioned centrally in mature dendrites, the growth from minus-end-out MTs could be central to creating new branches in the absence of KIF11⁶³. It remains unclear whether KIF11 directly organizes parallel MT bundles or if its effects are secondary to other molecular organizers.

KIF11 is not the only mitotic kinesin to restrict dendritic arborization. For example, conditional knockout (after P14) of the MT depolymerizer KIF2A (kinesin-13) results in overextended dendrites in dentate granule cells, presumably due to reduced depolymerization⁶⁴. Additionally, our previous work found that in cortical neuronal cultures, KIF3B (kinesin-2) knockdown increased dendritic arborization; in the prelimbic cortex, KIF3B knockdown increases dendritic spine density and impairs the extinction of fear memory^4,65. Another group conducted a more mechanistic investigation into how KIF3B regulates arborization by utilizing KIF3B^+/- mice, which exhibit schizophrenic traits and neuron hyperbranching⁶⁶. In Yoshihara et al., researchers find that KIF3B normally carries collapsin response mediator protein 2 (CRMP2), a protein needed in dendrites for F-actin bundling and proper lamellipodial dynamics. They show MTs being excluded from lamellipodia in developing growth cones, highlighting the interplay of actin and MTs in branching. These findings align with earlier studies, where developing rat neurons exhibit filopodia-like projections that form dendrites, initiated by actin dynamics and subsequently stabilized by MTs⁶⁷. However, Yoshihara et al. reveal this is due to secondary effects of impaired cargo delivery, rather than direct impacts on the cytoskeleton, like KIF11⁶⁶.

KIF11 is not a microtubule polymerizer in mature mammalian neurons

Overexpression of KIF11 results in curled MTs, similar to those seen in vitro with a KIF11-KHC chimera lacking the KIF11’s tail domain⁴⁰. KIF11 was previously proposed to act as an MT polymerase. However, we find it unlikely that KIF11 functions as a polymerase in mature hippocampal neurons. Expression of WT, KIF11^Y81F, and KIF11^ΔCterm all resulted in decreased dynamics of plus- and minus-end-out MTs, rather than an increase. While some observed loops might be due to polymerase activity, and we cannot resolve protofilaments, the compression and twisting behaviors suggest KIF11 also crosslinks and exerts rotational forces on MTs. Both pharmacological and genetic depletion of KIF11 increases MT dynamics and growth, indicating that KIF11 limits rather than enhances polymerization.

KIF11 overexpression and MCLID mutations reduce microtubule dynamics and dendritic arborization

Our study also explored how KIF11 mutations, predicted to cause loss-of-function, relate to intellectual disabilities. We focused on mutations from patients with minor microcephaly (~3 SD below the mean) yet diagnosed with severe intellectual disabilities, to isolate the effects of KIF11 mutations from those due to microcephaly alone¹⁶. We examined mutations in both the N- (KIF11^Y81F) and C-terminus (KIF11^ΔCterm) within mature neurons (Fig. 4 and Supplementary Fig. S4). Expression of MCLID patient mutations KIF11^ΔCterm and KIF11^Y81F produced phenotypes consistent with force exertion on MTs, and these mutant KIF11 proteins colocalized with MTs, suggesting they retain MT-binding ability. Analysis of these mutants with an MT gliding assay also supports this, as both KIF11^ΔCterm and KIF11^Y82F retained the ability to bind and slide MTs (Fig. 6E-H). Although it is unclear if they form functional spindles, these findings imply they may still function in mature or developing neurons, contributing to disease phenotypes.

KIF11 overexpression and MCLID mutations reduce neuronal communication

Remarkably, expression of WT KIF11, KIF11^ΔCterm, and KIF11^Y82F impacted synaptic communication, all by reducing mEPSC frequency in primary hippocampal neuronal cultures. As previously mentioned, this reveals that KIF11 overexpression and MCLID mutations negatively impact presynaptic vesicle release probability. This reciprocates our previous findings that KIF11-KD increases mEPSC frequency². Swarnkar and colleagues found that the increase in mEPSC frequency was regulated presynaptically, as knockdown of the presynaptic protein, piccolo, rescued the mEPSC frequency increase, as did treatment with APV/D-AP5 (an inhibitor of NMDAR, which blocked presynaptic NMDARs)². Results from this study further illuminated the presynaptic role of KIF11, as KIF11-KD increased dynamic minus-end-out MTs in axons (Fig. 2B,C,F,H) and increased anterograde mRuby-Synaptophysin flux and velocity (Fig. 3A-G). Nevertheless, these results do not exclude the possibility that KIF11’s role in regulating the postsynaptic dendritic arbor may also contribute to this change in synaptic communication. Indeed, the lack of differences between the mutants and WT closely matches the in vitro arborization analysis, with KIF11^ΔCterm exhibiting a less dramatic phenotype than KIF11^Y81F. Therefore, by regulating the amount of input the neuron receives, based on dendritic branch complexity, the output (presynaptic vesicle release) may also be regulated. Further studies are needed to better elucidate the presynaptic and postsynaptic contributions of KIF11, especially in the context of MCLID.

MCLID mutations differentially impair KIF11 function

KIF11^Y81F expression had nearly indistinguishable effects from WT KIF11 on dendritic arborization in vitro, suggesting the mutant retains near-complete functional capacity. However, in the MT sliding assays, KIF11^Y82F showed reduced ATP affinity and thus significantly slower MT sliding velocity (Fig. 6G,H; Supplementary Fig. S5I-J). While Y82F is a conservative mutation, its position near the nucleotide-binding site and tail-binding site may imply that the missing hydroxyl group is enough to disrupt ATP binding. In contrast, KIF11^ΔCterm expression had less impact on dendritic arborization in vitro and on plus-end-out MT dynamics. KIF11^ΔCterm exhibited more diffuse localization, allowing the visualization of dendritic spine-like protrusions (Supplementary Fig. S4I). However, in vivo, KIF11^ΔCterm resulted in a more severe phenotype than WT KIF11-OE as neurons matured within the hippocampus from P7 to P21 (Fig. 7). We also note that at P21, the KIF11^ΔCterm induced highly variable neuronal morphology phenotypes, as shown in Fig. 7G. Neurons expressing the other constructs produced more uniform neuronal structure within each condition, compared to KIF11^ΔCterm, where the upper representative neuron had many short, retracted branches, while the lower representative image had longer dendrites with essentially no secondary branching (Fig. 7G). The biochemical assays reveal KIF11^ΔCterm primarily forms dimers rather than tetramers, significantly impairing its ability to crosslink MTs (Fig. 6C-F). Additionally, the KIF11^ΔCterm displayed more variable velocity in the MT sliding assay, similar to the Eg5-Δtail-GFP construct previously studied, as the tail domain is essential for coordinating with the motor domain in the homotetramer⁴². Therefore, we hypothesize that in vivo, KIF11^ΔCterm displays these severe but irregular phenotypes due to variability in its multimerization state and truncated tail domain. KIF11^ΔCterm monomers, dimers, and trimers may also have more opportunity to form a mixed oligomer with the endogenous KIF11, although the extent and frequency of such occurrence remain unmeasured. Nonetheless, both mutations appear to retain the ability to exert force on MTs, likely contributing to MCLID pathophysiology.

In summary, we demonstrate that the mitotic kinesin KIF11 binds to parallel MT bundles in mature neurons to regulate minus-end-out MT growth and dendritic branching. This regulation extends beyond dendrites to mature axons. Disruption of MT stability affects synaptophysin, mitochondrial, and lysosomal trafficking in dendrites. Overexpression studies reveal that KIF11 exerts rotational forces in mature neurons. Additionally, patient mutations associated with intellectual disabilities retain the ability to generate force on MTs. Finally, the creation of a photo-inhibitable KIF11 reveals that localized, temporal inhibition is sufficient to increase MT dynamics and dendritic growth. These findings establish KIF11 as a MT dynamics rheostat in mature neurons, supporting an innovative model for dendritic arborization where minus-end-out MTs initiate new branch points. Future studies should investigate the origin of these dynamic minus-end-out MTs, KIF11’s regulatory partners, and consequences to behavior in vivo.

Methods

Neuronal cultures and transfection

For this study, in vitro experiments were performed in primary hippocampal cell cultures obtained from CD1 pups. The dissections were performed during the light phase of the cycle. Housing (ambient temperature of 22 °C, 28–72% humidity, 12-h light-dark cycle) and experimental procedures were approved and supervised by the Institutional Animal Care and Use Committee of the Wertheim UF Scripps Institute following their Guide for the Care and Use of Laboratory Animals.

Primary hippocampal mouse neuronal cultures were prepared from the brains of embryonic day 16–17 CD1 mice of both sexes. For imaging, cells were plated at 1.7 × 10⁵ on poly-D-lysine-coated (500 μg/ml PDL in Borate Buffer) 35 mm MatTek dishes with a 1.5 coverslip. Cultures were plated in Neurobasal medium (Invitrogen) supplemented with 10% fetal bovine serum and penicillin/streptomycin mix and grown in Neurobasal medium supplemented with 2% B27 (Invitrogen), 0.5 mM glutamax, and penicillin/streptomycin mix at 37 °C in 5% CO₂.

Neurons were transfected on Day in Vitro (DIV) and with the amount of plasmid indicated for each experiment. Transfections were performed using Lipofectamine™ 2000 (Invitrogen) and Oz Biosciences COMBIMAG (Fischer Scientific Cat. No. NC2043583) according to the Oz Biosciences magnetofection protocol on a magnetic plate (Oz Biosciences # MF10000). Note that for neurons, after washing out the magnetofection reagents, the media should be replaced with a 1:1 ratio of conditioned: new media.

Immunocytochemistry(ICC) analysis

DIV16 primary hippocampal neurons culture (in 35 mm Mattek dishes) were transfected with 0.5 µg of NC-GFP (Origene TR30013) or KIF11-OE (Genecoepia CS-Mm03465-Lv236-01) for Supplementary Fig. S1; or 0.5 µg NC-GFP (Origene TR30013), KIF11-OE (Genecoepia CS-Mm03465-Lv236-01), or Kif11^Y81F (Genecoepia CS-Mm03465-Lv236-02) for Supplementary Fig. S4L. At DIV18, neuronal cultures were washed with PBS and then fixed in a freshly prepared solution of 4% paraformaldehyde for 15 min. After three or more rinses in PBS, the cells were then incubated in 10% normal horse serum(NHS) (Gibco) in PBS-T with 0.1% Triton X-100 for 1 h at room temperature to reduce the non-specific binding of the primary antibody. Next, the cells were incubated overnight at 4 °C with primary antibodies in PBS-T 0.1% Triton X-100 with anti-Kif11 (1:1000, Abcam, #ab51976, mouse), anti-Map2 (1:1000, Abcam, #Ab32454, rabbit), and anti-beta3 Tubulin (1:1000, Synaptic Systems, 302304, Guinea pig) for Supplementary Fig. S1; or anti-alpha Tubulin (1:1000, ThermoFisher Scientific, PA1-38814, rabbit) for Supplementary Fig. S4L. Cells were then washed thoroughly with three rinses in PBS, followed by a 1 h room temperature incubation in PBS-T with secondary antibodies: 1:1000 Anti-mouse Alexa Fluor546 (Invitrogen, A10036, donkey), 1:1000 Anti-rabbit Alexa Fluor647 (Invitrogen, A31573, donkey), Anti-Guinea Pig Alexa Fluor 405 (Abcam, #Ab175678, goat). After washing three times with PBS, images were acquired using Zeiss ELYRA PS.1 Structured Illumination Super-Resolution Microscopy.

Structured Illumination Super-Resolution Microscopy (SIM)

Following immunocytochemistry, neurons were imaged using a Zeiss ELYRA PS.1 instrument (Carl Zeiss, Jena, Germany) at a resolution of 1028 × 1028 pixels with a Zeiss 63×/1.4 NA Plan Apochromatic objective. Each fluorescent channel, 405, 488 and 561 was acquired using three pattern rotations with three translational shifts. The final SIM projection images were made using ImageJ.

EB1 imaging and drug treatment

Primary hippocampal neurons, plated at low density (100 K cells in a 35 mm Mattek glass bottom dish), are transfected with 0.5 µg Dendra2-EB1 (Addgene Plasmid #57714) via lipofectamine DIV15-17. 6–12 h after transfection, feeding media is replaced with pre-warmed Hibernate-E. Immediately after replacing the media, the dish was placed in the prewarmed Tokai Hit stage top incubator at 37 °C. The cells are then imaged with the Olympus Fluoview FV3000 confocal microscope with the cellSens V4.1 software (64-bit) with the UPlanApo 60x/1.5 Oil N.A. objective, and with a 488 nm (20 mW) laser. Healthy cells that expressed Dendra-EB1 and have a 50 µm region of branched projection (likely dendrites) in the same focal plane, are selected. These neurons are then recorded in the 488 channel for 2 min at 1.25 frames per second. These movies are then opened in FIJI ImageJ and the ROI is manually drawn along the in-focus branched projections and processed with the KymoResliceWide plugin. The slope of the EB1 tracks seen on the kymograph, along with the directionality (moving towards(retrograde) or away(anterograde) from the soma), the length of the tracks, and the number of tracks are manually measured with the measure tool in FIJI ImageJ. These values were then used to calculate the MT growth rate (µm/s), the orientation of the MT (anterograde track = plus-end-out, retrograde tracks=minus-end-out), the duration of the growth, and the frequency of the growth, respectively.

EB3, mRuby-Synaptophysin, and mApple-PSD95 imaging and analysis

DIV14-16 Primary hippocampal mouse neurons, plated in 35 mm Mattek No1.5 dishes were simultaneously transfected via combiMag and Lipofectamine with 0.5 µg EB3-miRFP703 (Addgene #79994) or mRuby-Synaptophysin [(pEF Synaptophysin-mRuby was a gift from Edwin Chapman (Addgene plasmid # 188980; http://n2t.net/addgene:188980; RRID:Addgene_188980)] or mApple-PSD95 [mApple-PSD95-N-14 was a gift from Michael Davidson (Addgene plasmid # 54941; http://n2t.net/addgene:54941; RRID:Addgene_54941)] and either 0.5 µg NC-GFP (Origene TR30013) or KIF11-shRNA-A (Origene TG501174) or tagged KIF11 constructs. After 24–48 h of transfection, EB3-miRFP703 imaging was performed at 37 °C with 5% CO₂ in TOKAI HIT STX Stage Top Incubation Chamber, using a confocal microscope (FV3000; Olympus; UPLAPO60XOHR Objective in immersion oil). Time-lapse imaging was acquired with OlyVIA 16-bit Software (Olympus). Pyramidal hippocampal neurons expressing both the tagged constructs and either NC-GFP of shKIF11 or KIF11 overexpression constructs were selected for imaging. Time-lapse movies were recorded with a zero-second delay, with 3-frame averaging, first in the far-red channel (EB3) or red channel (Synaptophysin, PSD95), then in the green channel.

To distinguish between axons and dendrites, the researcher first trained herself to accurately recognize axons vs. dendrites in live neurons. To do this, she recorded movies from over 15 NC-GFP and shKIF11 neurons each and labeled processes either axons (these projections lacked spines and continued over 150 µm away from the soma with minimal branching and a consistent diameter, terminating in a growth cone) or dendrites (heavily decorated with spines and consistent branches starting <10 µm from the soma). Then these cells were fixed and stained with MAP2 (dendrite) and acetylated tubulin (MAP2- and acTub + = axon). The researcher then reimaged these dishes, found the previously identified neurons, and confirmed the accuracy of the labeling from the live imaging session. For NC-GFP/shKIF11 imaging of EB3 axons, primary, secondary, and tertiary branches at least 50 µm with clear EB3 puncta were the targeted regions of acquisition. For NC-GFP/shKIF11 imaging of synaptophysin, axons were targeted for acquisition, while dendrites were targeted for PSD95 acquisition. For NC-GFP/KIF11-OE/KIF11^Y81F/KIF11^ΔCterm and EB3 imaging, primary and secondary branches at least 50 µm in length with clear EB3 puncta were targeted for acquisition.

These movies are then opened in FIJI ImageJ and the ROI was manually drawn along the in-focus axon/dendrite and processed with the KymoResliceWide plugin. The slope of the tagged protein tracks seen on the kymograph, along with the directionality (moving towards(retrograde) or away(anterograde) from the soma), the length of the tracks, and the number of tracks are manually measured with the measure tool in FIJI ImageJ. The slopes were then used to calculate the MT growth rate (µm/s) or velocity (µm/s), the orientation of the MT/track (anterograde track= plus-end-out, retrograde tracks=minus-end-out), the duration of the growth, the distance of the MT growth/travel, and the frequency of the growth/traffic.

Sholl analysis—in neuronal cultures

After 24–48 h of transfection of hippocampal neurons using plasmid expressing eGFP or KIF11 constructs (custom-made by Genecoepia: KIF11-OE(CS-Mm03465-Lv236-01), Kif11^Y81F (CS-Mm03465-Lv236-02), and Kif11^ΔCTerm (CS-Mm03465-Lv236-03)), images of dendrites were collected at 37 °C with 5% CO₂ using a confocal microscope (FV3000; Olympus; UPLAPO60XOHR Objective in immersion oil). Z-stack images were acquired using OlyVIA 16-bit Software (Olympus) and maximum intensity projection images were used for Sholl analysis Plugin FIJI (ImageJ, NIH). The center of the soma is considered the midpoint, and the origin of the concentric radii was set from that point to the longest axis of the soma. The parameters set for analysis were: starting radius 20 μm, ending radius 100 μm, radius step size 10 μm. The maximum value of sampled intersections reflecting the highest number of processes/branches in the arbor was calculated and the number of intersections was plotted against the distance from the soma center in μm. Data were analyzed using two-way ANOVA with Tukey post hoc test.

Mitochondrial and lysosomal vesicle imaging and drug treatment

Primary hippocampal neurons, plated at low density (100 K cells in a 35 mm Mattek glass bottom dish) and imaged at DIV16. Prior to imaging, the feeding media were replaced with pre-warmed Hibernate-E containing 50 nM MitoTracker Green and 50 nM LysoTracker Red. After a 15-min incubation, the media is replaced with pre-warmed Hibernate-E and either DMSO, 1 µM Ispinesib, or 1 µM BRD9876. After a 15-min incubation, the dish is placed in a prewarmed Tokai Hit stage-top incubator at 37 °C. The cells are then imaged with the Olympus Fluoview FV3000 confocal microscope with the cellSens V4.1 software (64-bit) with the UPlanApo 60×/1.5 Oil N.A. objective and with a 488 nm (20 mW) and 561 nm:20 mW) laser. Healthy cells that contained clearly stained mitochondria and lysosomal vesicles, and a 100 µm region of axon or dendrite in the same focal plane are selected. These neurons are then recorded first in the 568 channel and then in the 488 channel for 2 min at 1.25 frames per second. These movies are then opened in FIJI ImageJ, and the ROI is manually drawn along the in-focus axon/dendrite and processed with the KymoResliceWide plugin. The number and directionality (moving towards(retrograde) or away(anterograde) from the soma of tracks per 100 µm to determine flux.

Electrophysiology

Coverslips with cultured primary hippocampal neurons were transfected at DIV14 with 0.5 µg of either NC-GFP, Kif11-OE, Kif11^Y81F, or Kif11^ΔCTerm plasmids via CombiMag (OZ bioscience) with Lipofectamine 2000 (Invitrogen). At DIV15–17, these coverslips were transferred to the perfusion chamber of an upright microscope and perfused with Extracellular bath solution (EBS) containing (in mM): 135 NaCl, 10 glucose, 3 CaCl₂, 2 KCl, 2 MgCl₂, and 5 HEPES, pH adjusted to 7.3–7.4 with NaOH, and 300–315 mOsm with sucrose. Patch-pipettes were pulled from borosilicate glass with a micropipette puller (Sutter Instrument Co.) and filled with an intracellular solution containing 100 mM K-gluconate, 1.7 mM KCl, 0.6 mM EGTA, 5 mM MgCl2, 10 mM Hepes, 4 mM ATP, and 0.1 mM GTP, adjusted to pH 7.2 with NaOH and 300–315 mOsm with sucrose. Recordings were performed to measure mEPSCs under voltage clamp. Whole-cell patch-clamp recordings were performed using an Axon Multiclamp 700b amplifier, 1440 A DigiData digitizer and pClamp software (Axon Instruments, Foster City, CA). Recordings were made at 50 kHz and subsequently filtered at 5 kHz. The recordings were conducted blindly, and four groups were used during these experiments; Control (NC-GFP), Kif11 overexpressed, Kif11^Y81F, and Kif11^ΔCTerm groups. Hippocampal neurons were cultured for at least 14 to 16 days to allow an extensive synaptic network to develop before recordings were made. The mEPSCs were recorded in the presence of TTX at 1 µM. The membrane potential was held at −60 mV during the recording of mEPSCs. The current clamp was recorded only to identify the health of the neurons. Only neurons with a resting membrane potential of less than −40 mV were used for our analysis. Action potentials were recorded in current clamp mode, I = 0. Membrane potential was maintained at least −40 mV in current clamp mode. All experiments were performed at room temperature (22–24 °C). The frequency and amplitude of EPSCs were analyzed using the template match search (pClamp) and basic statistical analysis performed to extract average amplitude and frequency. The amplitude and frequency of EPSCs expressed as a percentage of baseline level, calculated from an average of 3 min of the baseline-recording period. The amplitude and the frequency of EPSCs for each experiment were measured as the average of 5 min during the recording period.

Microtubule sliding assay with KIF11 mutants: Gene expression and protein purification

KIF11 expression and purification were conducted based on previous work³⁰. Viruses were generated using the FlexiBac system⁶⁸. SF9 cells (IPLB Sf21-AE, Merck 71104) at 1 million cells per mL were infected with virus (1:100, v/v) and genes were expressed for 96 h at 27 °C and 120 rpm in a New Brunswick Innova® 44 Incubator (Model#: M1282-0010). Cells were centrifuged at 300 × g for 10 min at 4 °C. Pellets were resuspended in PBS (1% of expression volume) with protease inhibitor, flash-frozen in liquid nitrogen, and stored at −80 °C. For purification, cell pellets were thawed on ice and resuspended in purification buffer (50 mM NaH₂PO₄, 300 mM KCl, 2 mM MgCl₂, 0.5 mM TCEP, 0.5 mM ATP, pH 7.5) with protease inhibitor. The lysate was cleared with an ultracentrifuge spin at 186,000 × g for 1 h at 4 °C. The supernatant was filtered through a 0.45 µm filter and loaded on a 1 mL HiTrap column using the system pump. The column was washed with immobilized metal affinity chromatography (IMAC) wash buffer (purification buffer with 20 mM imidazole), and the protein was eluted with IMAC elution buffer (purification buffer with 300 mM imidazole) with an elution gradient. Protein-containing fractions were pooled and concentrated with Amicon filters (cutoff 100 kDa). 3C protease[10 µg/ml] was then added to cleave the 6X-His tag overnight at 4 °C. The protein solution was diluted 6-fold to reduce the imidazole concentration and passed over the HiTrap column again. The flow through was concentrated to 0.5 mL, cleared at 17,000 × g for 10 min, and gel filtered over a Superose6 column with purification buffer. 5% glycerol was added, and the protein was flash-frozen in liquid nitrogen and stored at −80 °C.

Mass photometry

Mass photometry measurements were conducted using a TwoMP system (Refeyn, Oxford, UK) operated at ambient temperature. Precision coverslips (24 × 50 mm, Thorlabs CG15KH) were cleaned by sequential rinsing with isopropanol and deionized water, then dried under a stream of clean nitrogen gas. A 3 mm multi-well gasket was positioned on the coverslip, and 18 μL of BRB80 buffer (80 mM PIPES, 2 mM MgCl₂, 0.5 mM EGTA, pH 6.8) was dispensed into one well. The Refeyn system automatically focused on the glass-buffer interface. For each measurement, 2 µl of sample solution were added and the signals were immediately recorded for 60 s each. Samples were diluted in BRB80 to 1–10 nM range to detect 500–1500 landing events. Mass calibration was performed using BSA, IgG, Thyroglobulin standards (Sigma) diluted in BRB80 Buffer to 10 nM concentration. Raw mass photometry data were processed in the DiscoverMP software (Refeyn, Oxford, UK). Molecular masses were determined by comparison to the calibration curve established with protein standards.

Microtubule sliding assay with KIF11 mutants: 2D sliding experiments

To assemble six flow chambers, seven strips of Nescofilm were placed on the coverslip with structures and covered with a regular coverslip with DDS coating. The Nescofilm was melted on a hot plate. Every second channel was sealed with two-component silicone (Twinsil picodent) to avoid cross-contamination of channels. Channels were flushed with: (I) 0.2 mg/mL TAMRA 5G5 antibody (Invitrogen, RRID AB_2536728) solution in PBS for 5 min, (II) 1% F127 (w/v in PBS) solution for at least 1 h, (III) 3 washes with BRB80, (IV) fixed microtubules in motility buffer (MB-ADP, BRB80 with 10 µM taxol, 200 µg/mL casein (from bovine milk, Sigma C7078), 10 mM DTT, 0.1% (v/v) Tween-20, 20 mM D-glucose, 1 mM ADP), (V) MB-ADP wash, (VI) KIF11 (10 nM KIF11-EGFP; 80 nM KIF11ΔCterm-mNG or 15 nM KIF11Y82F-eGFP) in MB-ADP for 5 min, (VII) MB-ADP wash, (VIII) cargo microtubules in MB-ADP, (IX) wash with MB-ADP++ (MB-ADP with 200 µg/mL glucose oxidase and 20 µg mL⁻¹ catalase), (xii) MB- ATP + + (MB-ADP + + with 1 mM ATP instead of ADP).

Microtubule polymerization

Tubulin was purified from pig brains (acquired from a local slaughterhouse) according to standard protocols (Castoldi and Popov, 2003) and labeled in-house. Specifically, tubulin was polymerized, and microtubules were isolated by centrifugation in a glycerol cushion. Microtubules were mixed with the labeling dye (TAMRA succinimidyl ester, ThermoFischer; Alexa Fluor 488 Carboxylic Acid, 2,3,5,6-Tetrafluorophenyl ester, ThermoFischer; ATTO 647N Amine-reactive NHS-ester, ATTO- TEC) in a 10 to 20-fold molar excess of the dye and incubated for 30 min. The reaction was quenched with 0.5 M potassium glutamate, and the tubulin was purified with depolymerization- polymerization cycles. Labeled tubulin was mixed in a ratio of 1:3 with unlabeled tubulin for experiments. Fixed and cargo microtubules were grown with tubulin with guanylyl-(α,β)-methylene- diphosphonate (GMP-CPP) and stabilized with taxol. For fixed microtubules, 40 µL elongation mix containing 1.25 mM GMP-CPP, 1.25 mM MgCl₂, and 4.5 µM TAMRA labeled tubulin in BRB80 (80 mM Pipes at pH 6.9, 1 mM MgCl₂, 1 mM EGTA) was incubated on ice for 5 min and then for 30 min at 37 °C. Microtubules were pelleted (17,000 × g, 15 min) and resuspended in elongation mix (1.25 mM GMP-CPP, 1.25 mM MgCl₂, and 0.5 µM TAMRA-labeled tubulin in BRB80) and grown for 2–3 days at 32 °C. Microtubules were pelleted (17,000 × g, 8 min) and gently resuspended in BRB80 with 10 µM taxol (BRB80X). They were kept at room temperature for several days for annealing. To grow cargo microtubules, a polymerization mix of 1.25 mM GMP-CPP, 1.25 mM MgCl₂, and 5 µM Atto647N labeled tubulin was incubated on ice for 5 min and then for 8 min at 37 °C. Microtubules were pelleted (17,000 × g, 15 min) and resuspended in BRB80X.

Optical image acquisition of microtubule sliding assays

Optical image acquisition: Optical imaging was performed using an inverted fluorescence microscope (Axio Observer Z1; Carl Zeiss Microscopy GmbH) with a 63× oil-immersion 1.46NA objective (Zeiss) in combination with an EMCDD camera (iXon Ultra; Andor Technology) controlled by Metamorph (Molecular Devices Corporation). A LED white light lamp (Sola Light Engine; Lumencor) in combination with a TRITC filterset (ex 520/35, em 585/40, dc 532: all Chroma Technology Corp.), an Atto647N filterset (ex 628/40, em 692/40, dc 635) and a GFP filter set (ex 475/35, em 525/45), corresponding to TAMRA labeled microtubules, Atto647N labeled microtubules and EGFP labeled motors/Atto488 labeled microtubule extensions, respectively, were used for epifluorescence imaging. The sliding of cargo microtubules was imaged in the Atto647N channel for 10 min with 10 frames/s with an exposure time of 100 ms. Fixed microtubules were imaged in the TRITC channel for 100 frames at 10 frames/s with an exposure time of 100 ms after imaging the cargo microtubules.

Microtubule sliding assay with altered ATP concentration

To assemble six flow chambers, seven strips of Nescofilm were placed on the coverslip with structures and covered with a regular coverslip with DDS coating. The Nescofilm was melted on a hot plate. Every second channel was sealed with two component silicone (Twinsil picodent) to avoid cross-contamination of channels. Channels were flushed with: (I) 1:60 TetraSpeck(v/v, 200 nm) bead solution in PBS(for 2 min, (II) 0.2 mg/mL TAMRA 5G5 antibody (Invitrogen, RRID AB_2536728) solution in PBS for 5 min, (III) 1% F127 (w/v in PBS) solution for at least 1 h, (IV) 3 washes with BRB80 (80 mM Pipes at pH 6.9, 1 mM MgCl 2, 1 mM EGTA), (V) fixed microtubules in motility buffer (MB-ADP, BRB80 with 10 µM taxol, 200 µg/mL casein (from bovine milk, Sigma C7078), 10 mM DTT, 0.1% (v/v) Tween-20, 20 mM D-glucose, 1 mM ADP), (VI) MB-ADP wash, (VII) KIF11 (10 nM KIF11; 80 nM KIF11ΔCterm or 15 nM KIF11Y82F) in MB-ADP for 5 min, (VIII) MB-ADP wash, (IX) cargo microtubules in MB-ADP, (X) wash with MB-ADP++ (MB-ADP with 200 µg/mL glucose oxidase and 20 µg mL−1 catalase), (XI) MB- ATP++ (BRB80 with 10 µM taxol, 200 µg/mL casein,10 mM DTT, 0.1% (v/v) Tween-20, 20 mM D-glucose, 200 µg/mL glucose oxidase and 20 µg mL−1 catalase, ATP with a concentration of 0.01 mM, 0.1 mM, 0.2 mM, 0.5 mM, 1 mM, 2 mM respectively).

In utero electroporation (IUE) and tissue processing

IUE was performed as described previously⁶⁹ using Swiss Webster mice purchased from Charles River. The surgeries and euthanasia were carried out during the light part of the cycle. Housing (ambient temperature of 22 °C, 35–60% humidity, 12-h light-dark cycle) and experimental procedures were approved and supervised by the Institutional Animal Care and Use Committee of Max Planck Florida, following their Guide for the Care and Use of Laboratory Animals. Embryonic day 15.5 timed-pregnant Swiss Webster mice were deeply anesthetized with isoflurane (2.5%), Buprenorphine-ER LAB (1.0 mg/kg, Wedgewood Pharmacy) was subcutaneously administered for analgesia, and the uterine horns were exposed by laparotomy. EGFP control or each KIF11 plasmid was mixed with TurboRFP plasmid (final concentration 0.75 µg/µl each) in phosphate-buffered saline (PBS) solution containing the dye Fast Green FCF (0.1 mg/ml, Sigma-Aldrich, F7258). Each plasmid mix was injected (1–2 μl) through the uterine wall into the lateral ventricles of embryos. Then, the embryo’s head was carefully held with a custom-made triple electrode (Nepa Gene)⁷⁰. Electrical pulses (50 V for 50 ms, 4 times with 950 ms intervals) were delivered with an electro-kinetic transfection system (NEPA21, Nepa Gene). After electroporation, the uterine horns were returned to the abdominal cavity, and embryos were allowed to continue normal development. Mouse sex was not evaluated, but they were presumably mixed sex.

At P7, the mouse brain was isolated after decapitation under anesthesia with isoflurane and kept in 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB) at 4 °C overnight. At P21, mice were transcardially perfused with 4% PFA in 0.1 M PB, and the brains were kept in a 4% PFA solution at 4 °C overnight. The following day, brains were transferred to 30% sucrose in PBS for overnight incubation at 4 °C. Next, brains were embedded in Tissue-Tek O.C.T Compound (Sakura), frozen at −80 °C, then sectioned at 50 µm via a cryostat. Brain sections were mounted on gelatin-coated glass slides, embedded with fluoro-gel (Electron Microscopy solutions) with DAPI, cured overnight, and imaged with a FV3000 Olympus Confocal Microscope. DAPI was used to identify the dorsal CA1. Images were collected from eGFP and tagged-KIF11-expressing neurons localized within the dorsal CA1. As transfection/expression with TurboRFP was rare, only the GFP signal was consistently acquired. Approximately 35 µm thick Z-stacks (0.35 µm step sizes) were collected to capture the entirety of the eGFP/tagged-KIF11 expressing neurons.

Sholl/branch point/branch length analysis—in vivo

Z-stacks of the eGFP/tagged-KIF11 expressing neurons were opened in the Imaris software (Oxford Instruments, 64-bit, version 10.2.0). Due to the variation in cell body shapes, we used the starting point and filament model for tracing and quantifying these neurons. The starting point was set at the center of the dimmest point within the cell body, which is presumably where the nucleus is located, as it would exclude GFP-tagged KIF11. The filament tracer was used for automated detection and tracing of dendrites. Manual tracing was performed if parts of branches were missed, background signal was included as part of the neuron, or neighboring neurons were conjoined. Sholl analysis (using concentric spheres every 10 µm, starting at 20 µm from the center of the soma), number of dendritic branch points, and dendritic segment length were calculated from the traced models.

Optical manipulation of KIF11

The pdDronpaV-KIF11 construct was custom synthesized with mouse Kif11 (NM_010615.1) with N-pdDronpaV tag⁴⁵ with a custom linker (GSSGGGGSGGGGSGGGGS), in the lentiviral vector pEZ-Lv236 with CAG promoter by Genecopeia (CS-Mm03465-Lv236-04). The pdDronpa-CDK5 (pLL3.7m-psCDK5, Addgene #89365) construct was used as a control for optimizing pdDronpaV association and dissociation.

Human embryonic kidney (HEK)293T (American Type Culture Collection (ATCC; Manassas, VA, USA(CRL-11268)) were maintained in high-glucose Dulbecco’s Modified Eagle Medium (DMEM, Thermo Fisher) supplemented with 5% fetal bovine serum (FBS, Thermo Fisher) and 2 mM glutamine (Sigma, Thermo Fisher) at 37 °C in air with 5% carbon dioxide. HEK293T cells were transfected at 75-90% confluency with Lipofectamine 2000 (Invitrogen) in 35 mm MatTek dishes with 1.5 coverglass bottoms. Primary Hippocampal neurons were plated and transfected as previously described for EB3 imaging.

Cells were imaged within 16 h of transfection using an Olympus Fluoview FV3000 confocal microscope with the cellSens V4.1 software (64-bit) with the UPlanApo 60×/1.5 Oil N.A. objective. The imaging protocol for pdDronpa monomerization and dimerization was first optimized in HEK293T cells, starting with the protocol outlined in Zhou et al. 2017 and adjusted to prevent cell death, minimal photobleaching, and a clear increase in fluorescence intensity with 405 nm dimerization. pdDronpaV was imaged with a 10% neutral density filter, the 485/30-nm excitation filter, the 505-nm dichroic mirror, and the 525/40-nm emission filter. miRFP703-EB3 was imaged with a 10% neutral density filter, a 640-nm excitation filter, a 730-nm dichroic mirror, and a 640/730-nm emission filter. Dronpa variants were photo-switched off (enabling KIF11 activation) using the same excitation and dichroic filters as for imaging, but without a neutral density filter, for 5 min. DronpaV was reactivated (enabling KIF11 inactivation) with 405 nm (50 mW) at 0.1% power for 5 min.

Fluorescence was analyzed in ImageJ by measuring corrected total cell fluorescence (CTCF) = Integrated Density – (Area of selected cell X Mean fluorescence of background readings) and normalizing values to Frame 0 s. The intensities and durations of light used in each experiment were noted in the figures. EB3 Flux was calculated as described above. New branch points were characterized from average projection intensity images from the FarRed (EB3) channel, as new growth longer than 5 µm. Loop number and size were also calculated from the average projection intensity images from the FarRed (EB3) channel.

Statistical analysis

Levels of significance in this study are based on p-values calculated by GraphPad Prism 9 and 10 (GraphPad Software) and derived using Student’s t test, Mann-Whitney U test, One-way or Two-way ANOVA. Tukey, Dunnett, or Sidak tests were used for post hoc analyses. Significance was defined as <0.05. Outliers were defined with a Grubb’s test (Alpha = 0.05). Details of each experiment, including the number of replicates and statistical specifications, are provided in the figure legends, results, and Source Data file.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data generated in this study are provided in the Source Data file. The in vitro arborization images used in this study are available in the figshare database under 10.6084/m9.figshare.31092946. The in vivo arborization files are available in the Zenodo database under 10.5281/zenodo.18303030. Source data are provided with this paper.

References

Morikawa, M., Tanaka, Y., Cho, H. S., Yoshihara, M. & Hirokawa, N. The molecular motor KIF21B mediates synaptic plasticity and fear extinction by terminating Rac1 activation. Cell Rep. 23, 3864–3877 (2018).
Article CAS PubMed Google Scholar
Swarnkar, S., Avchalumov, Y., Raveendra, B. L., Grinman, E. & Puthanveettil, S. V. Kinesin family of proteins Kif11 and Kif21B act as inhibitory constraints of excitatory synaptic transmission through distinct mechanisms. Sci. Rep. 8, 17419 (2018).
Article PubMed Central ADS PubMed Google Scholar
Swarnkar, S. et al. Molecular motor protein KIF5C mediates structural plasticity and long-term memory by constraining local translation. Cell Rep. 36, 109369–109369 (2021).
Article PubMed Central CAS PubMed Google Scholar
Joseph, N. F. et al. Molecular motor KIF3B in the prelimbic cortex constrains the consolidation of contextual fear memory. Mol. Brain 14, 162 (2021).
Article PubMed Central CAS PubMed Google Scholar
Puthanveettil, S. V. et al. A new component in synaptic plasticity: upregulation of kinesin in the neurons of the Gill-Withdrawal reflex. Cell 135, 960–973 (2008).
Article PubMed Central CAS PubMed Google Scholar
Alsabban, A. H., Morikawa, M., Tanaka, Y., Takei, Y. & Hirokawa, N. Kinesin Kif3b mutation reduces NMDAR subunit NR2A trafficking and causes schizophrenia-like phenotypes in mice. EMBO J. 39, e101090 (2020).
Article CAS PubMed Google Scholar
Muhia, M. et al. The kinesin KIF21B regulates microtubule dynamics and is essential for neuronal morphology, synapse function, and learning and memory. Cell Rep. 15, 968–977 (2016).
Article PubMed Central CAS PubMed Google Scholar
Fan, R. & Lai, K. O. Understanding how kinesin motor proteins regulate postsynaptic function in neuron. FEBS J. 289, 2128–2144 (2022).
Article CAS PubMed Google Scholar
Zhou, R., Niwa, S., Guillaud, L., Tong, Y. & Hirokawa, N. A molecular motor, KIF13A, controls anxiety by transporting the serotonin type 1 A receptor. Cell Rep. 3, 509–519 (2013).
Article CAS PubMed Google Scholar
Aiken, J. & Holzbaur, E. L. F. Cytoskeletal regulation guides neuronal trafficking to effectively supply the synapse. Curr. Biol. 31, R633–r650 (2021).
Article PubMed Central CAS PubMed Google Scholar
Hirokawa, N., Niwa, S. & Tanaka, Y. Molecular motors in neurons: transport mechanisms and roles in brain function, development, and disease. Neuron 68, 610–638 (2010).
Article CAS PubMed Google Scholar
Guedes-Dias, P. & Holzbaur, E. L. F. Axonal transport: driving synaptic function. Science 366, eaaw9997 (2019).
Article PubMed Central CAS PubMed Google Scholar
Wang, N. & Xu, J. Functions of kinesin superfamily proteins in neuroreceptor trafficking. BioMed. Res. Int. 2015, 639301 (2015).
PubMed Central PubMed Google Scholar
Silverman, M. A. et al. Expression of kinesin superfamily genes in cultured hippocampal neurons. Cytoskeleton 67, 784–795 (2010).
Article PubMed Central CAS PubMed Google Scholar
Joseph, N.F., Swarnkar, S., Puthanveettil, S.V. Double duty: mitotic kinesins and their post-mitotic functions in neurons. Cells 10, https://doi.org/10.3390/cells10010136 (2021).
Jones, G. E. et al. Microcephaly with or without chorioretinopathy, lymphoedema, or mental retardation (MCLMR): review of phenotype associated with KIF11 mutations. Eur. J. Hum. Genet. 22, 881–887 (2014).
Article CAS PubMed Google Scholar
Malvezzi, J. V. M. et al. KIF11 microdeletion is associated with microcephaly, chorioretinopathy and intellectual disability. Hum. Genome Var. 5, 18010 (2018).
Article PubMed Central CAS PubMed Google Scholar
Balikova, I. et al. Ocular manifestations of microcephaly with or without chorioretinopathy, lymphedema or intellectual disability (MCLID) syndrome associated with mutations in KIF11. Acta Ophthalmol. 94, 92–98 (2016).
Article CAS PubMed Google Scholar
Schlögel, M. J. et al. No evidence of locus heterogeneity in familial microcephaly with or without chorioretinopathy, lymphedema, or mental retardation syndrome. Orphanet J. Rare Dis. 10, 52 (2015).
Article PubMed Central PubMed Google Scholar
Güneş, N. et al. A novel mutation of KIF11 in a child with 22q11.2 deletion syndrome associated with MCLMR. Mol. Syndromol. 9, 266–270 (2018).
Article PubMed Central PubMed Google Scholar
Chauvière, M., Kress, C. & Kress, M. Disruption of the mitotic kinesin Eg5 gene (Knsl1) results in early embryonic lethality. Biochem. Biophys. Res. Commun. 372, 513–519 (2008).
Article ADS PubMed Google Scholar
Wang, Y., Smallwood, P., Williams, J. & Nathans, J. A mouse model for Kinesin Family Member 11 (Kif11)-associated familial exudative vitreoretinopathy. Hum. Mol. Genet. 29, 1121–1131 (2020).
Pandey, H., Popov, M., Goldstein-Levitin, A. & Gheber, L. Mechanisms by which Kinesin-5 motors perform their multiple intracellular functions. Int. J. Mol. Sci. 22, 6420 (2021).
Article PubMed Central CAS PubMed Google Scholar
Walczak, C. E., Verma, S. & Mitchison, T. J. XCTK2: a kinesin-related protein that promotes mitotic spindle assembly in Xenopus laevis egg extracts. J. Cell Biol. 136, 859–870 (1997).
Article PubMed Central CAS PubMed Google Scholar
Ferenz, N. P., Gable, A. & Wadsworth, P. Mitotic functions of kinesin-5. Semin. Cell Dev. Biol. 21, 255–259 (2010).
Article PubMed Central CAS PubMed Google Scholar
Enos, A. P. & Morris, N. R. Mutation of a gene that encodes a kinesin-like protein blocks nuclear division in A. nidulans. Cell 60, 1019–1027 (1990).
Article CAS PubMed Google Scholar
Miki, H., Setou, M., Kaneshiro, K. & Hirokawa, N. All kinesin superfamily protein, KIF, genes in mouse and human. Proc. Natl. Acad. Sci. 98, 7004–7011 (2001).
Article PubMed Central ADS CAS PubMed Google Scholar
Kashina, A. S., Scholey, J. M., Leszyk, J. D. & Saxton, W. M. An essential bipolar mitotic motor. Nature 384, 225 (1996).
Article PubMed Central ADS CAS PubMed Google Scholar
Sharp, D. J. et al. The bipolar kinesin, KLP61F, cross-links microtubules within interpolar microtubule bundles of Drosophila embryonic mitotic spindles. J. Cell Biol. 144, 125–138 (1999).
Article PubMed Central CAS PubMed Google Scholar
Meißner, L., Niese, L., Schüring, I., Mitra, A. & Diez, S. Human kinesin-5 KIF11 drives the helical motion of anti-parallel and parallel microtubules around each other. EMBO J. 43, 1244–1256 (2024).
Article PubMed Central PubMed Google Scholar
Kapitein, L. C. et al. The bipolar mitotic kinesin Eg5 moves on both microtubules that it crosslinks. Nature 435, 114–118 (2005).
Article ADS CAS PubMed Google Scholar
Yount, A. L., Zong, H. & Walczak, C. E. Regulatory mechanisms that control mitotic kinesins. Exp. Cell Res. 334, 70–77 (2015).
Article PubMed Central CAS PubMed Google Scholar
Ferhat, L. et al. Expression of the mitotic motor protein Eg5 in postmitotic neurons: implications for neuronal development. J. Neurosci. 18, 7822–7835 (1998).
Article PubMed Central CAS PubMed Google Scholar
Kahn, O. I., Sharma, V., González-Billault, C. & Baas, P. W. Effects of kinesin-5 inhibition on dendritic architecture and microtubule organization. Mol. Biol. Cell 26, 66–77 (2015).
Article PubMed Google Scholar
Nadar, V. C., Ketschek, A., Myers, K. A., Gallo, G. & Baas, P. W. Kinesin-5 is essential for growth-cone turning. Curr. Biol. 18, 1972–1977 (2008).
Article PubMed Central CAS PubMed Google Scholar
Freixo, F. et al. NEK7 regulates dendrite morphogenesis in neurons via Eg5-dependent microtubule stabilization. Nat. Commun. 9, 2330 (2018).
Article PubMed Central ADS PubMed Google Scholar
Lad, L. et al. Mechanism of inhibition of human KSP by Ispinesib. Biochemistry 47, 3576–3585 (2008).
Article CAS PubMed Google Scholar
Chen, G. Y. et al. Eg5 inhibitors have contrasting effects on microtubule stability and metaphase spindle integrity. ACS Chem. Biol. 12, 1038–1046 (2017).
Article PubMed Central ADS CAS PubMed Google Scholar
Courtney, N. A., Briguglio, J. S., Bradberry, M. M., Greer, C. & Chapman, E. R. Excitatory and inhibitory neurons utilize different Ca(2+) sensors and sources to regulate spontaneous release. Neuron 98, 977–991.e975 (2018).
Article PubMed Central CAS PubMed Google Scholar
Chen, Y. & Hancock, W. O. Kinesin-5 is a microtubule polymerase. Nat. Commun. 6, 8160 (2015).
Article PubMed Central ADS CAS PubMed Google Scholar
Krieg, M. et al. Genetic defects in β-spectrin and tau sensitize C. elegans axons to movement-induced damage via torque-tension coupling. eLife 6, e20172 (2017).
Article PubMed Central PubMed Google Scholar
Bodrug, T. et al. The kinesin-5 tail domain directly modulates the mechanochemical cycle of the motor domain for anti-parallel microtubule sliding. eLife 9, e51131 (2020).
Article PubMed Central CAS PubMed Google Scholar
Scholey, J. E., Nithianantham, S., Scholey, J. M. & Al-Bassam, J. Structural basis for the assembly of the mitotic motor Kinesin-5 into bipolar tetramers. eLife 3, e02217 (2014).
Article PubMed Central PubMed Google Scholar
Goikolea-Vives, A. & Stolp, H. B. Connecting the neurobiology of developmental brain injury: neuronal arborisation as a regulator of dysfunction and potential therapeutic target. Int. J. Mol. Sci. 22, https://doi.org/10.3390/ijms22158220 (2021).
Zhou, X. X., Fan, L. Z., Li, P., Shen, K. & Lin, M. Z. Optical control of cell signaling by single-chain photoswitchable kinases. Science 355, 836–842 (2017).
Article PubMed Central ADS PubMed Google Scholar
Cheng, P. -l & Poo, M. -m Early events in axon/dendrite polarization. Annu. Rev. Neurosci. 35, 181–201 (2012).
Article CAS PubMed Google Scholar
Lanoue, V. & Cooper, H. M. Branching mechanisms shaping dendrite architecture. Dev. Biol. 451, 16–24 (2019).
Article CAS PubMed Google Scholar
Ricolo, D., Castro-Ribera, J. & Araújo, S. J. Cytoskeletal players in single-cell branching morphogenesis. Dev. Biol. 477, 22–34 (2021).
Article CAS PubMed Google Scholar
Sakakibara, A., Ando, R., Sapir, T. & Tanaka, T. Microtubule dynamics in neuronal morphogenesis. Open Biol. 3, 130061 (2013).
Article PubMed Central PubMed Google Scholar
Wayman, G. A. et al. Activity-dependent dendritic arborization mediated by CaM-Kinase I activation and enhanced CREB-dependent transcription of Wnt-2. Neuron 50, 897–909 (2006).
Article CAS PubMed Google Scholar
Urbanska, M., Blazejczyk, M. & Jaworski, J. Molecular basis of dendritic arborization. Acta Neurobiol. Exp. 68, 264–288 (2008).
Article Google Scholar
Dent, E. W. Dynamic microtubules at the synapse. Curr. Opin. Neurobiol. 63, 9–14 (2020).
Article PubMed Central CAS PubMed Google Scholar
Tas, R. P. et al. Differentiation between oppositely oriented microtubules controls polarized neuronal transport. Neuron 96, 1264–1271.e1265 (2017).
Article PubMed Central CAS PubMed Google Scholar
Baas, P. W. & Black, M. M. Individual microtubules in the axon consist of domains that differ in both composition and stability. J. Cell Biol. 111, 495–509 (1990).
Article PubMed Central CAS PubMed Google Scholar
Kleele, T. et al. An assay to image neuronal microtubule dynamics in mice. Nat. Commun. 5, 4827 (2014).
Article PubMed Central ADS CAS PubMed Google Scholar
Yau, K. W. et al. Dendrites in vitro and in vivo contain microtubules of opposite polarity and axon formation correlates with uniform plus-end-out microtubule orientation. J. Neurosci. 36, 1071–1085 (2016).
Article PubMed Central CAS PubMed Google Scholar
Thyagarajan, P., Feng, C., Lee, D., Shorey, M. & Rolls, M. M. Microtubule polarity is instructive for many aspects of neuronal polarity. Dev. Biol. 486, 56–70 (2022).
Article PubMed Central CAS PubMed Google Scholar
Feng, C. et al. Patronin-mediated minus end growth is required for dendritic microtubule polarity. J. Cell Biol. 218, 2309–2328 (2019).
Article PubMed Central CAS PubMed Google Scholar
Cao, Y. et al. Microtubule minus-end binding protein CAMSAP2 and kinesin-14 Motor KIFC3 control dendritic microtubule organization. Curr. Biol. 30, 899–908.e896 (2020).
Article PubMed Central CAS PubMed Google Scholar
Strothman, C. et al. Microtubule minus-end stability is dictated by the tubulin off-rate. J. Cell Biol. 218, 2841–2853 (2019).
Article PubMed Central CAS PubMed Google Scholar
Guha, S., Patil, A., Muralidharan, H. & Baas, P. W. Mini-review: microtubule sliding in neurons. Neurosci. Lett. 753, 135867 (2021).
Article PubMed Central CAS PubMed Google Scholar
Lucero, E. M. et al. Increased KIF11/kinesin-5 expression offsets Alzheimer Aβ-mediated toxicity and cognitive dysfunction. iScience 25, 105288 (2022).
Article PubMed Central ADS CAS PubMed Google Scholar
Katrukha, E. A., Jurriens, D., Salas Pastene, D. M. & Kapitein, L. C. Quantitative mapping of dense microtubule arrays in mammalian neurons. eLife 10, e67925 (2021).
Article PubMed Central CAS PubMed Google Scholar
Homma, N. et al. KIF2A regulates the development of dentate granule cells and postnatal hippocampal wiring. Elife 7, https://doi.org/10.7554/eLife.30935 (2018).
Joseph, N. F., Grinman, E., Swarnkar, S. & Puthanveettil, S. V. Molecular motor KIF3B acts as a key regulator of dendritic architecture in cortical neurons. Front. Cell. Neurosci. 14, https://doi.org/10.3389/fncel.2020.521199 (2020).
Yoshihara, S. et al. Betaine ameliorates schizophrenic traits by functionally compensating for KIF3-based CRMP2 transport. Cell Rep. 35, https://doi.org/10.1016/j.celrep.2021.108971 (2021).
Dailey, M. E. & Smith, S. J. The dynamics of dendritic structure in developing hippocampal slices. J. Neurosci. 16, 2983–2994 (1996).
Article PubMed Central CAS PubMed Google Scholar
Lemaitre, R. P., Bogdanova, A., Borgonovo, B., Woodruff, J. B. & Drechsel, D. N. FlexiBAC: a versatile, open-source baculovirus vector system for protein expression, secretion, and proteolytic processing. BMC Biotechnol. 19, 20 (2019).
Article PubMed Central PubMed Google Scholar
Mikuni, T., Nishiyama, J., Sun, Y., Kamasawa, N. & Yasuda, R. High-throughput, high-resolution mapping of protein localization in mammalian brain by in vivo genome editing. Cell 165, 1803–1817 (2016).
Article PubMed Central CAS PubMed Google Scholar
Dal Maschio, M. et al. High-performance and site-directed in utero electroporation by a triple-electrode probe. Nat. Commun. 3, 960 (2012).
Article ADS Google Scholar

Download references

Acknowledgements

We gratefully acknowledge the funding that supported the work; NIH grants 5R01MH094607-05,1R01MH119541-01A1, and R01MH118444 and NSF grant 2231247 (for S.V.P.); NIH grant 1F32MH131420-01A1 and a Training Grant in Alzheimer’s Drug Discovery from the Lottie French Lewis Fund of the Community Foundation for Palm Beach and Martin Counties, Florida (for J.L.W.); NIH grant 5R35NS116804-06 (for R.Y.); the German Research Foundation DI 1226/8 and DI 1226/10 (for S.D.) and the Free state of Saxony (PhD stipend to L.N.). We sincerely thank Ms. Emelie Lewis for her help with ICC experiments, and Drs. HaJeung Park and Kevin Knight for modeling the KIF11 structure, and Dr. Laura Meißner for initial discussions of microtubule sliding experiments. The Protein Biochemistry (PEPC) facility at MPI-CBG Dresden is acknowledged for support with the Mass Photometry experiments.

Author information

Authors and Affiliations

Department of Neuroscience, The Herbert Wertheim UF Scripps Institute for Biomedical Innovation & Technology, Jupiter, FL, USA
Jenna L. Wingfield, Bindu L. Raveendra, Adrian E. Gonzalez-Santiago, Jackson P. Carter & Sathyanarayanan V. Puthanveettil
B CUBE—Center for Molecular Bioengineering, TUD Dresden University of Technology, Dresden, Germany
Lukas Niese, Rahul Grover & Stefan Diez
Department of Molecular Medicine, The Herbert Wertheim UF Scripps Institute for Biomedical Innovation & Technology, Jupiter, FL, USA
Yosef Avchalumov & Jason X.-J. Yuan
Max Planck Florida Institute for Neuroscience, Jupiter, FL, USA
Xiaodan Liu, Yoshihisa Nakahata & Ryohei Yasuda
Cluster of Excellence Physics of Life, TUD Dresden University of Technology, Dresden, Germany
Stefan Diez
Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany
Stefan Diez

Authors

Jenna L. Wingfield
View author publications
Search author on:PubMed Google Scholar
Lukas Niese
View author publications
Search author on:PubMed Google Scholar
Yosef Avchalumov
View author publications
Search author on:PubMed Google Scholar
Xiaodan Liu
View author publications
Search author on:PubMed Google Scholar
Rahul Grover
View author publications
Search author on:PubMed Google Scholar
Yoshihisa Nakahata
View author publications
Search author on:PubMed Google Scholar
Bindu L. Raveendra
View author publications
Search author on:PubMed Google Scholar
Adrian E. Gonzalez-Santiago
View author publications
Search author on:PubMed Google Scholar
Jackson P. Carter
View author publications
Search author on:PubMed Google Scholar
Ryohei Yasuda
View author publications
Search author on:PubMed Google Scholar
Jason X.-J. Yuan
View author publications
Search author on:PubMed Google Scholar
Stefan Diez
View author publications
Search author on:PubMed Google Scholar
Sathyanarayanan V. Puthanveettil
View author publications
Search author on:PubMed Google Scholar

Contributions

J.L.W., L.N., and S.V.P. conceived of and designed experiments. J.L.W. conducted all live-imaging experiments and arborization imaging and analyzed the data. L.N. and R.G. conducted microtubule sliding and KIF11 motility measurements under the supervision of S.D. Y.A. and J.P.C. conducted the mEPSC experiments with supervision from J.X.-J.Y. and S.V.P., respectively. X.L. and Y.N. performed IUE with supervision from R.Y. A.E.G.-S. performed ICC imaging and analysis. B.L.R. prepared reagents and helped with experimental planning. J.L.W. wrote and revised the manuscript with input from all the authors. S.V.P. supervised the study. All authors read and commented on the manuscript.

Corresponding author

Correspondence to Sathyanarayanan V. Puthanveettil.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information (download PDF )

Description of Additional Supplementary Files (download PDF )

Supplementary Movie 1 (download GIF )

Supplementary Movie 2 (download GIF )

Supplementary Movie 3 (download GIF )

Supplementary Movie 4 (download GIF )

Supplementary Movie 5 (download GIF )

Supplementary movie 6 (download GIF )

Supplementary Movie 7 (download GIF )

Supplementary Movie 8 (download GIF )

Supplementary Movie 9 (download GIF )

Supplementary Movie 10 (download AVI )

Supplementary Movie 11 (download AVI )

Supplementary Movie 12 (download GIF )

Supplementary Movie 13 (download GIF )

Supplementary Movie 14 (download GIF )

Reporting Summary (download PDF )

Transparent Peer Review file (download PDF )

Source data

Source Data (download XLSX )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Wingfield, J.L., Niese, L., Avchalumov, Y. et al. Intellectual disability-causing mutations in KIF11 impair microtubule dynamics and dendritic arborization. Nat Commun 17, 4125 (2026). https://doi.org/10.1038/s41467-026-70522-z

Download citation

Received: 26 August 2024
Accepted: 02 March 2026
Published: 18 March 2026
Version of record: 07 May 2026
DOI: https://doi.org/10.1038/s41467-026-70522-z