Arabidopsis KNL1 recruits type one protein phosphatase to kinetochores to silence the spindle assembly checkpoint

Abstract

Proper chromosome segregation during cell division is essential for genomic integrity and organismal development. This process is monitored by the spindle assembly checkpoint (SAC), which delays anaphase onset until all chromosomes are properly attached to the mitotic spindle. The kinetochore protein KNL1 plays a critical role in recruiting SAC proteins. Here, we reveal that Arabidopsis KNL1 regulates SAC silencing through the direct recruitment of type one protein phosphatase (TOPP) to kinetochores. We show that KNL1 interacts with all nine TOPPs via a conserved RVSF motif in its N terminus, and this interaction is required for the proper localization of TOPPs to kinetochores during mitosis. Disrupting KNL1-TOPP interaction leads to persistent SAC activation, resulting in a severe metaphase arrest and defects in plant growth and development. Our findings highlight the evolutionary conservation of KNL1 in coordinating kinetochore-localized phosphatase to ensure timely SAC silencing and faithful chromosome segregation in Arabidopsis.

Arabidopsis KNL1 not only assembles the SAC at kinetochores but also recruits phosphatases to silence the SAC.

INTRODUCTION

The faithful segregation of chromosomes during cell division is essential for maintaining genomic integrity and ensuring proper growth and development of multicellular organisms. This critical process is monitored by the spindle assembly checkpoint (SAC), a surveillance mechanism that delays the onset of anaphase until each chromosome is properly attached and aligned within the spindle (1, 2). Failure to activate and satisfy the SAC can lead to chromosome missegregation and the formation of aneuploid daughter cells, which is a hallmark of many developmental disorders and diseases.

At the heart of the SAC signaling pathway is the kinetochore, a large multi-protein complex assembled at the centromeric region of each chromosome (3). The kinetochore scaffold protein KNL1 (kinetochore null protein 1) serves as a critical platform for the recruitment and regulation of SAC components at the kinetochore (4, 5). When kinetochores are not properly attached to spindle microtubules, KNL1 recruits key checkpoint proteins, including the BUB (budding uninhibited by benzimidazole) and MAD (mitotic arrest deficient) families (6–8). This triggers the activation of the SAC signaling cascade, which inhibits the anaphase-promoting complex/cyclosome (APC/C) and delays the onset of anaphase, thereby allowing time for proper chromosome attachment and alignment (2, 9). However, KNL1 also plays a crucial role in timely silencing of the SAC signal once all chromosomes have achieved proper biorientation within the spindle. This silencing process involves the inactivation and removal of the mitotic checkpoint complex (MCC) from the kinetochore, a critical step that is mediated by the dephosphorylation of key SAC components by protein phosphatases, such as PP1 (protein phosphatase 1) (10–12). This dual functionality of KNL1 in orchestrating both the activation and inactivation of the SAC is essential for ensuring the fidelity of chromosome segregation during cell division.

While the fundamental mechanisms coordinating the SAC are evolutionarily conserved across eukaryotes, the specific molecular architectures and regulatory strategies used by different organisms can vary substantially (13). This diversity is particularly evident in the case of plant KNL1 proteins, which exhibit substantial sequence divergence from their animal and fungus counterparts (14, 15). For example, plant KNL1 homologs lack the characteristic Met-Glu-Leu-Thr (MELT) motif repeats found in the middle region of fungal and animal KNL1 proteins (16, 17). These MELT repeats serve as hubs for recruiting BUB and MAD proteins to the kinetochore (7). Instead, plant KNL1 proteins have evolved distinct interaction domains to engage with the SAC signal. Several recent studies reveal that rather than relying on the classical MELT-based interaction modules seen in other eukaryotic systems, plant KNL1 proteins have developed specialized mechanisms to independently load plant-specific SAC components onto kinetochores (17, 18). The notable divergence in structural features and interaction motifs of plant KNL1 proteins compared to their nonplant counterparts highlights the need to further explore the plant-specific mechanisms governing the spatial and temporal control of the SAC.

Despite the poor sequence similarity of KNL1 across species and the variable predicted protein sizes, recent studies have identified and characterized several functional regions within KNL1 that serve as docking hubs for the recruitment of various SAC regulatory factors. These include the SILK motif with the consensus sequence (S/G)ILK and the RVSF motif with the consensus RVxF, both of which mediate the direct binding of protein phosphatase PP1 (11); various numbers of the MELT motif, as described above, that recruit BUB3 and BUB1 (8, 19, 20); as well as two KI motifs with the consensus sequence KI(D/N)FxxF(L/I)xRL that interact with BUB1 and BUBR1 (21, 22). The plant KNL1 proteins appear to have retained only the RVSF motif in their extreme N terminus, which is known for the recruitment of the phosphatase PP1 in other eukaryotic systems (23). This raises intriguing questions about the function mechanisms of plant KNL1 proteins. For example, does plant KNL1 use this conserved RVSF motif to recruit PP1 homologs as seen in other organisms? Or has plant KNL1 evolved unique strategies to ensure timely silencing of the SAC signal upon achievement of proper chromosome biorientation? The Arabidopsis thaliana genome encodes nine members of type one protein phosphatases (TOPPs), which are orthologous to PP1 in animals and fungi (24, 25). While TOPPs have been implicated in modulation of various hormone signaling pathways and stress response mechanisms in Arabidopsis (26–29), it is currently unknown whether any of these TOPP paralogs play a role in regulating the SAC signaling pathway.

In this study, we reveal a mechanism by which Arabidopsis KNL1 regulates the SAC through the direct recruitment of TOPPs to kinetochores. We demonstrate that the interaction between KNL1 and TOPP, mediated by the conserved RVSF motif in the N terminus of KNL1, plays a crucial role in timely silencing of the SAC upon chromosome congression at the metaphase plate. Disruption of the KNL1-TOPP interaction leads to persistent SAC activation, characterized by the sustained localization of key SAC components at kinetochores, resulting in severe defects in mitotic progression and plant growth. Overall, our findings highlight a conserved feature by which KNL1 coordinates kinetochore-localized phosphatase to orchestrate SAC regulation and mitotic fidelity in Arabidopsis.

RESULTS

KNL1 interacts with TOPP through the conserved RVSF motif

To investigate the potential interaction between Arabidopsis KNL1 and the nine TOPP paralogs, we conducted a comprehensive screen using yeast two-hybrid (Y2H) and bimolecular fluorescence complementation (BiFC) assays. Our results revealed that the full-length KNL1 protein was able to robustly interact with all nine TOPP proteins present in the Arabidopsis genome (TOPPs 1 to 9) (Fig. 1, A and B). Sequence alignment analysis showed that the RVSF motif located at the N terminus of KNL1 exhibits a high degree of conservation across different plant species (fig. S1). Furthermore, AlphaFold structural predictions suggested a high-likelihood of interaction between the TOPP paralogs and the RVSF motif of Arabidopsis KNL1 (fig. S2). To test the role of the RVSF motif in mediating the KNL1-TOPP interaction, we generated a mutant version of KNL1, denoted KNL1^4A, in which the RVSF sequence was replaced with four alanine residues (AAAA). We found that this RVSF-to-AAAA mutation completely abolished the interactions between KNL1 and all nine TOPP paralogs in both our Y2H and BIFC assays (Fig. 1, A and B).

Fig. 1. KNL1 interacts with TOPP through the RVSF motif.

(A) Y2H assay to examine interactions between the nine TOPP paralogs (TOPP1-TOPP9) and KNL1 or the KNL1^4A mutant. The empty vector was used as a negative control (NA). Yeast cultures were spotted on vector-selective (-L/-W, left) and interaction-selective (-L/-W/-H/-A, right) media and imaged after 2 days of incubation at 30°C. (B) BIFC assay to visualize the interactions between KNL1 variants (fused with the N-terminal YFP fragment) and the TOPP proteins (fused with the C-terminal YFP fragment) in Nicotiana benthamiana leaves. Scale bars, 10 μm. (C) In vitro pull-down assays using recombinant GST-tagged KNL1 variants and MBP-tagged TOPP4 and TOPP8 proteins. The MBP-TOPP4/TOPP8 fusion proteins were immobilized on MBP-trap beads and incubated with the GST-KNL1 variants. (D) Coimmunoprecipitation analysis of the interaction between KNL1 and TOPP4/8. Total proteins were extracted from 3-day-old transgenic seedlings expressing pTOPP4/8::TOPP4/8-GFP and pKNL1::KNL1-FLAG and were incubated with anti-GFP beads. Both the input and immunoprecipitated samples were subjected to immunoblotting using anti-FLAG or anti-GFP antibodies. These experiments were repeated three times with similar results.

To further validate the KNL1-TOPP interaction and the critical role of the RVSF motif, we performed in vitro pull-down experiments using recombinant proteins. Our pull-down assays confirmed the direct binding of the KNL1–N-terminal fragment (KNL1^N) to two representative TOPP paralogs, TOPP4 and TOPP8. The glutathione S-transferase (GST)–KNL1^4A fusion was unable to be copurified by either MBP-TOPP4 or MBP-TOPP8 (Fig. 1C). Last, we carried out anti–green fluorescent protein (GFP) immunoprecipitation experiments using transgenic plants coexpressing TOPP4/8-GFP and KNL1-FLAG. These experiments demonstrated that when TOPP4 or TOPP8 was captured by anti-GFP affinity purification, the KNL1-FLAG protein was also detected in the complex (Fig. 1D). Collectively, these results establish that the Arabidopsis KNL1 protein directly interacts with all nine TOPP paralogs through a conserved RVSF motif in its N terminus.

KNL1 recruits TOPP to kinetochores

To investigate the subcellular localization of the nine TOPPs during mitosis, we transformed each TOPP protein fused to GFP under the control of their respective endogenous promoters into wild-type plants. Immunofluorescence microscopy revealed that most TOPP members exhibited similar localization patterns throughout the different stages of mitosis (figs. S3 to S11). When chromosome congression was initiated after nuclear envelope breakdown (NEB), eight TOPP (TOPP1 through TOPP8) proteins appeared as weak paired dots associated with chromosome regions, surrounded by a bipolar spindle structure. At metaphase, when the chromosomes were fully aligned at the metaphase plate flanked by the kinetochore fibers, the TOPP signals were observed at the two edges of the aligned chromosomes, colocalizing toward the ends of the kinetochore fibers. Upon anaphase onset, the most prominent TOPP signals tracked the shortening kinetochore fibers until they eventually reached the two spindle poles at telophase (Fig. 2, A to C, and figs. S3 to S10). In contrast, TOPP9, different from the other eight TOPP proteins, remained diffusely distributed in the cytoplasm throughout mitosis (fig. S11).

Fig. 2. KNL1 recruits TOPP to kinetochores.

(A and B) Triple-label immunolocalization of the TOPP paralogs, microtubules, and DNA in wild-type (WT) root meristematic cells undergoing mitosis. TOPP4-GFP (A) and TOPP8-GFP (B) are pseudocolored in green, microtubules are in red, and DNA is in blue. (C) Summary of the localization patterns of all nine TOPPs in Arabidopsis cells. (D) Colocalization of TOPP4-GFP or TOPP8-GFP with KNL1-FLAG in mitotic cells. The merged image has TOPP4 or TOPP8 pseudocolored in green, KNL1 in red, and DNA in blue. (E and F) Localization of TOPP4-GFP (E) and TOPP8-GFP (F) in the knl1 mutant background. The merged image has TOPP4 and TOPP8 pseudocolored in green, microtubules in red, and DNA in blue. (G) Live-cell imaging of wild-type seedlings expressing TOPP4-GFP, with representative snapshot images acquired from movie S1. (H) Assessment of the dynamic changes in TOPP4-GFP fluorescence intensity over time, as observed in the live-cell imaging experiment shown in (G). Scale bars, 5 μm.

To determine whether the dot-like TOPP signals specifically targeted the kinetochores, we expressed TOPP4-GFP and TOPP8-GFP fusions in a knl1 background line complemented with a functional KNL1 protein tagged with the FLAG peptide. Immunofluorescence microscopy revealed that TOPP4-GFP and TOPP8-GFP signals precisely colocalized with the KNL1-FLAG signal, which appeared as paired dots on bioriented kinetochores (Fig. 2D). These colocalization results suggest that the TOPP paralogs associate with kinetochores during mitosis.

Further analysis of the subcellular localization of the different TOPP paralogs revealed additional insights. During interphase and cytokinesis, when the two daughter nuclei were formed, TOPP5, TOPP6, TOPP8, and TOPP9 were found to be present in the cytoplasm, while the other TOPP proteins accumulated within the nucleus (Fig. 2C and figs. S3 to S11). This observation implies that, in addition to their roles at the kinetochores during mitosis, many of the TOPP proteins may also have important functions in the nucleus at other stages of the cell cycle potentially involved in dephosphorylating key nuclear substrates.

The interaction and colocalization between KNL1 and TOPP prompted us to examine whether the kinetochore localization of TOPP is dependent on the presence of KNL1. We transformed constructs expressing each of the TOPP paralogs into the knl1 mutant background. The immunofluorescence analysis revealed that the kinetochore-associated signals of all the TOPP members that had previously localized to the kinetochores (TOPP1 through TOPP8) were completely lost in the knl1 mutant cells (Fig. 2, E and F, and figs. S3 to S10). This finding demonstrates that KNL1 is required for the proper recruitment of TOPPs to the kinetochores during mitosis.

The kinetochore-localized TOPP signals exhibited a distinct temporal pattern compared to the typical SAC signals. While the SAC components, such as BMF3 or MAD1, showed strong kinetochore signals before chromosome alignment and were specifically enriched at the kinetochores of unaligned chromosomes, these signals then disappeared once the chromosomes were fully aligned at the metaphase plate (30, 31). In contrast, the TOPP signals at kinetochores were relatively weak or did not stand out prominently from the background before metaphase but became more pronounced when all chromosomes were aligned at the metaphase plate (Fig. 2, A and B). This suggested that the retention of TOPP at kinetochores may be influenced by the status of microtubule attachment. We then performed experiments using the microtubule-depolymerizing agent oryzalin. Our results showed that in the presence of oryzalin, where the proper kinetochore-microtubule attachment is disrupted, most metaphase cells exhibited either a loss or a notable decrease in the kinetochore localization of the TOPP proteins. The TOPP signals at kinetochores were not affected in anaphase cells (fig. S12). Live-cell imaging analysis further corroborated these observations, showing that the TOPP’s kinetochore localization started to increase in signal intensity as the chromosomes were aligning and segregating (Fig. 2, G and H; fig. S13; and movies S1 and S2). This inverse localization patterns between TOPP and SAC proteins at kinetochores suggests that TOPPs may play an antagonistic role to the SAC machinery in regulating mitotic progression, similar to observations in other eukaryotic systems (32).

Functional consequences of disrupting KNL1-TOPP interaction

To investigate the functional significance of the KNL1-TOPP interaction, we transformed KNL1^4A-FLAG, which was shown to abolish the binding with TOPPs, under the control of the native KNL1 promoter into the knl1 mutant background. While the expression of unmodified KNL1 was able to fully rescue the growth defects associated with the knl1 mutation, the KNL1^4A transgene failed to restore the normal plant phenotype. Instead, these plants expressing KNL1^4A exhibited even more severe growth inhibition compared to the original knl1 mutant (Fig. 3A and fig. S14). Specifically, the mutant plants expressing KNL1^4A displayed distinctly deformed, dark purple–colored rosette leaves and were unable to produce any flowers, ultimately leading to the death of the plants within approximately 4 weeks (Fig. 3A). This notable phenotype contrasted sharply with the wild-type–like appearance of the knl1 mutants complemented with the unmodified KNL1 protein.

Fig. 3. Expressing KNL1^4A cannot rescue knl1 plants.

(A) Growth phenotypes of 4-week-old Arabidopsis plants, including WT, knl1, and the knl1 mutant expressing either unmodified KNL1 or KNL1^4A. (B to D) Live-cell imaging of WT plant cells (B), knl1 mutant cells (C), and knl1 mutant cells expressing the KNL1^4A construct (D), along with GFP-TUB6 and histone-RFP. Representative snapshot images are acquired from movies S3 to S5. Scale bars, 5 μm.

To further dissect the underlying cellular defects caused by the disruption of the KNL1-TOPP interaction, we monitored mitotic progression in living cells using a histone H1.2–red fluorescent protein (RFP) marker labeling chromosomes and a GFP-TUB6 marker labeling microtubules. While the knl1 mutant cells exhibited characteristic chromosome congression and segregation defects, with persistent misaligned chromosomes and premature anaphase onset when compared with wild-type cells, the KNL1^4A-expressing cells were able to properly congress their chromosomes to the metaphase plate (Fig. 3, B to D, and movies S3 to S5). However, the KNL1^4A cells then exhibited a notable delay in the transition from metaphase to anaphase. These cells remained arrested at metaphase with fully aligned chromosomes for over 2 hours, in stark contrast to the rapid progression through mitosis (within 10 min) observed in the wild-type and knl1 mutant cells (Fig. 3, B to D, and movies S3 to S5). The persistent metaphase arrest in KNL1^4A-expressing cells might result in severe defects in plant growth and development.

KNL1-TOPP interaction is required for timely SAC silencing

To further investigate the functional consequences of disrupting the interaction between KNL1 and TOPP, we expressed the KNL1^4A-FLAG protein under the control of the KNL1 promoter in wild-type plants to better monitor and quantify mitotic progression. This was necessary because the sick growth of knl1/KNL1^4A plants made it difficult to find enough cells for live-cell imaging. The expression of the KNL1^4A mutant protein in wild-type plants also resulted in a growth inhibition phenotype, characterized by dwarf and bushy inflorescence development (Fig. 4A and fig. S15A). Immunofluorescence analysis revealed that KNL1^4A was still able to properly localize to the kinetochores throughout mitosis, similar to the unmodified KNL1 protein (fig. S16).

Fig. 4. KNL1-TOPP interaction is required for timely SAC silencing.

(A) Growth phenotypes of 7-week-old Arabidopsis plants, including WT plants, WT plants expressing KNL1^4A, the bmf3 mutant, bmf3 mutant plants expressing KNL1^4A, the mad1 mutant, and mad1 mutant plants expressing KNL1^4A. (B) Live-cell imaging of mitotic cells expressing GFP-TUB6 and histone-RFP in WT and WT/KNL1^4A backgrounds. Representative snapshot images are acquired from movies S6 and S7. (C) Live-cell imaging of mitotic cells expressing GFP-TUB6 and histone-RFP in bmf3 and bmf3/KNL1^4A backgrounds. Representative snapshot images are acquired from movies S8 and S9. (D) Live-cell imaging of mitotic cells expressing GFP-TUB6 and histone-RFP in mad1 and mad1/KNL1^4A backgrounds. Representative snapshot images are acquired from movies S10 and S11. (E) Triple-label immunolocalization of BMF3 (green), microtubules (red), and DNA (blue) in control and KNL1^4A-expressing cells. (F) Triple-label immunolocalization of MAD1 (green), microtubules (red), and DNA (blue) in control and KNL1^4A-expressing cells. (G) Quantification of the duration from NEB to anaphase onset in the different genotypes. Scale bars, 5 μm.

We then used live-cell imaging approaches, labeling chromosomes with a histone H1.2-RFP marker and microtubules with a GFP-TUB6 marker, to monitor mitotic progression in the KNL1^4A-expressing wild-type cells. While these cells exhibited normal chromosome congression and segregation dynamics, we observed a notable delay in the metaphase-to-anaphase transition compared to the control wild-type cells (Fig. 4B, fig. S15D, and movies S6 and S7). Quantitative analysis showed that the time from NEB to anaphase onset was approximately 1493 ± 189.8 s (n = 12) in the KNL1^4A-expressing cells, compared to only 578 ± 83.07 s (n = 22) in the wild-type control cells (Fig. 4G).

To investigate the potential mechanism underlying this metaphase delay, we examined the localization of key SAC components, including BUB1/MAD3 family (BMF) proteins BMF3 and MAD1. In complemented control cells, the BMF3-GFP and GFP-MAD1 signals were detected at the kinetochores before chromosome alignment but disappeared once all chromosomes were properly bioriented at the metaphase plate (Fig. 4, E and F). In contrast, in KNL1^4A-expressing cells, both BMF3 and MAD1 signals persistently decorated the bioriented kinetochores of congressed chromosomes that have established interkinetochore tension at metaphase (Fig. 4, E and F), suggesting a sustained activation of the SAC. When the chromosomes started to segregate at anaphase, the SAC signals disappeared in both the control and KNL1^4A-expressing cells (fig. S17).

To further confirm that the prolonged metaphase observed in the KNL1^4A-expressing cells was indeed due to persistent SAC signaling, we introduced the KNL1^4A-FLAG transgene into Arabidopsis lines harboring mutations in the SAC components BMF3 or MAD1. The growth inhibition phenotype caused by KNL1^4A expression in wild-type plants was completely abolished in either bmf3 or mad1 mutant plants (Fig. 4A and fig. S15, B and C). Moreover, live-cell imaging analysis revealed that the metaphase delay caused by the KNL1^4A expression in wild-type cells was also eliminated in the SAC-deficient bmf3 and mad1 mutant backgrounds (Fig. 4, C, D, and G; fig. S15, E and F; and movies S8 to S11). Collectively, these results demonstrate that the direct interaction between KNL1 and TOPP, mediated by the conserved RVSF motif, is crucial for the timely silencing of the SAC when all chromosomes have achieved proper bipolar attachment within the spindle.

Inhibiting TOPPs led to delayed SAC silencing

To further investigate the functional relationship between KNL1 and TOPPs, we generated a dominant-negative variant of the TOPP4 by introducing a threonine-to-methionine substitution at amino acid 246 (TOPP4^T246M). This mutation had been previously shown to be an effectively dominant-negative inhibitor of TOPPs function (33). When we expressed the TOPP4^T246M mutant protein under the control of the native TOPP4 promoter in wild-type plants, we observed that the TOPP4^T246M-expressing plants exhibited notably similar growth phenotypes as the previously reported topp4-1 mutant plants (33). Specifically, these transgenic plants displayed a dwarf and bushy inflorescence architecture (Fig. 5A and fig. S18). Notably, this phenotype closely resembled the defects seen in plants expressing the KNL1^4A protein, which disrupts the interaction between KNL1 and the TOPP proteins. This observation suggests a functional link between the KNL1-TOPP axis and the regulation of plant growth and development.

Fig. 5. Inhibiting TOPP results in delayed SAC silencing.

(A) Growth phenotypes of 6-week-old WT plants and WT plants expressing TOPP4^T246M. (B) Growth phenotypes of 6-week-old bmf3 mutant and bmf3 mutant plants expressing TOPP4^T246M. (C and D) Live-cell imaging of mitotic cells expressing GFP-TUB6 and histone-RFP in WT (C) and WT/TOPP4^T246M (D) backgrounds. Representative snapshot images are acquired from movies S12 and S13. (E) Quantification of the duration from NEB to anaphase onset in different seedlings. (F and G) Triple-label immunolocalization of MAD1 (green), microtubules (red), and DNA (blue) in control (F) and TOPP4^T246M-expressing (G) cells. (H and I) Live-cell imaging of mitotic cells expressing GFP-TUB6 and histone-RFP in bmf3 (H) and bmf3/TOPP4^T246M (I) backgrounds. Representative snapshot images are acquired from movies S16 and S17. Scale bars, 5 μm.

To directly compare the cellular consequences of impairing TOPPs function versus disrupting the KNL1-TOPP interaction, we monitored mitotic progression in TOPP4^T246M-expressing cells using the same live-cell imaging approach with fluorescently labeled chromosomes and microtubules. Consistent with the observations in KNL1^4A-expressing wild-type cells, TOPP4^T246M-expressing wild-type cells displayed an obvious delay in the metaphase-to-anaphase transition (Fig. 5, C and D, and movies S12 and S13). Quantitative analysis revealed that the time from NEB to anaphase onset was approximately 1120 ± 118.5 s (n = 22) in the TOPP4^T246M-expressing cells, compared to only 578 ± 83.07 s (n = 12) in the wild-type control cells (Fig. 5E). To further corroborate the importance of TOPPs activity in the timely silencing of the SAC, we conducted pharmacological experiments in Arabidopsis using the TOPP inhibitor tautomycin (34). Our results showed that a 3-hour treatment with this inhibitor in wild-type plants could also result in a metaphase delay when compared to the mock treatment (fig. S19).

We also examined the localization of SAC component MAD1 in TOPP4^T246M-expressing cells. Similar to the observations in KNL1^4A-expressing cells, the MAD1 signals persistently decorated the kinetochores of the aligned chromosomes at metaphase in TOPP4^T246M-expressing cells, in contrast to control cells where MAD1 signals disappeared once chromosomes were properly bioriented (Fig. 5, F and G). To verify that the prolonged metaphase observed in the TOPP4^T246M-expressing cells was caused by unsilenced SAC signaling, we crossed the TOPP4^T246M transgene into the bmf3 mutant background, which is deficient in a key SAC component. While expressing the TOPP4^T246M protein still led to a compact inflorescence phenotype in the bmf3 mutant background, the growth inhibition effects were less severe compared to the wild-type background (Fig. 5B). Moreover, live-cell imaging showed that the metaphase delay caused by TOPP4^T246M expression in wild-type cells was also abolished in the SAC-deficient bmf3 mutant plants (Fig. 5, H and I, and movies S16 and S17).

Together, these results demonstrate that inhibition of TOPP function, whether through expression of the dominant-negative TOPP4^T246M mutant or pharmacological inhibition using tautomycin, phenocopies the effects observed upon disruption of KNL1-TOPP interaction by the KNL1^4A mutation. Both perturbations lead to a persistent metaphase arrest due to sustained SAC signaling, ultimately resulting in severe defects in plant growth and development.

DISCUSSION

As the central scaffold protein at kinetochores, KNL1 plays a pivotal and multifaceted role in orchestrating the spatiotemporal regulation of the SAC during mitosis. The results of this study offer valuable understanding of how Arabidopsis KNL1 coordinates the assembly of PP1 paralogs to govern the SAC. These coordinated actions are vital for maintaining the accuracy of chromosome segregation during cell division in flowering plants.

Distinct versus conserved strategies used by Arabidopsis KNL1 to coordinate SAC activation and silencing

In most eukaryotes, KNL1 recruits SAC components to kinetochores to activate the checkpoint signaling. Unattached kinetochores trigger MPS1 kinase–mediated phosphorylation of KNL1’s MELT repeats, which serve as a docking site for BUB3-BUB1 and BUB3-BUBR1 complexes (6, 8, 35). These complexes recruit the MAD1-MAD2 complex, leading to the production of the MCC and the delay of metaphase-to-anaphase transition (1, 2). However, our recent study reveals that Arabidopsis KNL1 follows a distinct activation mode due to the absence of MELT motifs. Instead, plant KNL1 has evolved lineage-specific characteristics to recognize and recruit SAC components, suggesting a specialized and potentially divergent mechanism for SAC activation in plants (18). In Arabidopsis, BMF3 and BUB3.3 proteins can independently recruit to the kinetochore based on the KNL1 protein (18, 31). Among the SAC components, only BMF3 and MAD1 exhibit typical unattached kinetochore localization, while others show persistent kinetochore or cytoplasmic distributions (30). These findings highlight the divergent mechanisms used by plant KNL1 for SAC recruitment, distinguishing them from the established pathways in fungi and animals.

Unlike the distinct strategy for assembly the SAC, this study shows that Arabidopsis KNL1 uses a conserved mechanism for silencing the SAC. This occurs through the recruitment of TOPP, plant-specific homologs of PP1, to kinetochores via the RVSF motif in KNL1. This KNL1-TOPP interaction parallels the well-known partnership between KNL1 and PP1 catalytic subunits observed in other eukaryotic systems. In these contexts, direct binding of PP1 to the conserved RVSF motif in KNL1 is crucial for timely SAC silencing and proper progression through mitosis (36, 37). Our findings in Arabidopsis extend these mechanistic insights to the land plants, highlighting evolutionarily conserved strategies used by eukaryotic cells to coordinate kinetochore-localized phosphatase activity with mitotic checkpoint signaling regulation. While fungal and animal KNL1 proteins have both SILK and RVSF motifs for PP1 binding (11, 38), Arabidopsis KNL1 only contains the RVSF motif. Despite this structural difference, the interaction between Arabidopsis KNL1 and TOPP remains critical for SAC silencing. Disrupting this connection, either by mutating the RVSF motif in KNL1 or using dominant-negative inhibition of TOPPs, leads to a persistent metaphase arrest caused by continuous SAC signaling. This emphasizes the essential nature of this regulatory module for maintaining genome stability and ensuring faithful cell division across diverse organisms.

Potential targets and mechanisms of TOPP-mediated SAC silencing

The temporal dynamics of TOPP at kinetochores, which become more prominent during metaphase chromosome alignment, contrast with the localization patterns of SAC proteins. Plant SAC components like BMF3 and MAD1 exhibit stronger kinetochore signals before chromosome alignment (18, 30, 31), while TOPP signals are relatively weak until all chromosomes are properly bioriented. This inverse relationship suggests that TOPP may have an antagonistic role in timely inactivation of the SAC machinery, similar to the function of animal PP1 phosphatases in this process (32). As chromosomes achieve stable bipolar attachment and align at the metaphase plate, increased recruitment of TOPP at kinetochores could lead to dephosphorylation and removement of SAC components, thereby silencing the checkpoint and allowing the cell to proceed into anaphase.

While our study has revealed the pivotal role of the KNL1-TOPP axis in SAC silencing, the specific molecular targets and mechanisms by which TOPP switches off the checkpoint signaling cascade remain to be elucidated. In fungi and animals, a well-established mechanism involves the direct dephosphorylation of KNL1 itself by PP1 at the MELT repeats (12, 35). This dephosphorylation event removes SAC proteins from kinetochores, resulting in checkpoint inactivation (39). However, the recruitment of SAC components by plant KNL1 does not appear to follow the conventional MPS1-based MELT phosphorylation principle. Plant KNL1 proteins lack the conserved MELT motifs, and MPS1 deletion does not affect the kinetochore targeting of SAC proteins in plants (30, 31). This suggests that plant TOPP may use alternative mechanisms to silence the SAC. A likely possibility is that the key substrates of the TOPP could include the core kinetochore-associated SAC proteins that are known to interact with the N terminus of plant KNL1. Our previous findings have demonstrated that Arabidopsis KNL1 recruits the SAC components BUB3.3 and BMF3 independently through a eudicot-specific domain (ESD) located within its N-terminal region (31). This ESD motif is situated in close proximity to the RVSF motif that mediates the KNL1-TOPP interaction. This spatial arrangement suggests that BUB3.3 and BMF3 could be potential targets of the recruited TOPP during the silencing of the SAC.

However, the precise phosphorylation sites on SAC components that are targeted by TOPP, and how this dephosphorylation event is coordinated with the prior phosphorylation by mitotic kinases like MPS1 and Aurora (11, 12, 37, 40, 41), remain to be elucidated in plants. Identifying specific TOPP substrates and mapping intricate phosphorylation-dephosphorylation cycles will be crucial for understanding how the KNL1-TOPP axis antagonizes SAC kinases activities to achieve timely checkpoint silencing. In addition, it is important to investigate whether TOPP solely act as terminal effectors in dephosphorylating SAC components or whether they also regulate the upstream SAC kinases. Further investigations are needed to unravel the precise molecular mechanisms by which coordinated action of kinetochore-localized kinases and phosphatases regulates the phosphorylation state of key SAC components during mitosis.

SAC silencing pathways in eukaryotes are complex and involve additional regulatory mechanisms beyond just the recruitment of PP1 to kinetochores. Another important protein phosphatase family, PP2A-B56, is also known to play crucial roles in the inactivation of the SAC in various eukaryotic lineages. In animal cells, PP2A-B56 is recruited to kinetochores through its association with BUBR1 (42–44). The Arabidopsis genome encodes nine PP2A-B56 regulatory subunit paralogs (24, 25), but it remains unknown whether these plant PP2A-B56 paralogs are also recruited to kinetochores by BUB proteins and whether they function in SAC silencing in plants. Furthermore, in animal cells, the dynein-dependent stripping of the MAD2 protein from kinetochores has been shown to play a crucial role in the efficient silencing of the SAC (45–47). However, plants do not encode the dynein motor complex, instead having an enlarged kinesin family. This raises the intriguing possibility that plants may have evolved alternative kinesin-dependent mechanisms to facilitate the removal of SAC components from attached kinetochores, thereby contributing to the timely inactivation of the checkpoint signaling. In addition to the kinetochore-based SAC silencing mechanisms, the disassembly of the MCC at cytoplasm is equally important for turning off the SAC. This process is predominantly achieved by the APC15-dependent ubiquitylation of CDC20 (38, 39) and TRIP13-p31^comet-dependent conformational change of MAD2 (48–50). In Arabidopsis, homologs of APC15, TRIP13, and p31^comet have been identified (51–53), but their specific roles in SAC silencing have not yet been investigated. Further exploration of the roles of PP2A-B56, kinesin-dependent SAC component removal, and the MCC disassembly machinery in plants will be crucial to fully elucidate the evolutionary integration of SAC regulating pathways across eukaryotes.

Expanding the functional repertoire of KNL1-TOPP axis

Several recent studies provide intriguing insights into the specific action modes of the KNL1 protein in plant cell division, which appears to differ from other eukaryotic systems (17, 18). While KNL1 deletion is typically lethal in various organisms, our recent findings demonstrate that in Arabidopsis, the absence of KNL1 leads to severe developmental defects but is not directly lethal (18). This suggests that plants have evolved a remarkable tolerance for a certain degree of chromosome missegregation during mitosis, in contrast to the stringent requirements observed in other eukaryotes.

In this study, we further expanded our observations and uncovered an unexpected twist. When KNL1 knockout plants were complemented with a mutant version of KNL1 (RVSF-to-AAAA), the resulting plants exhibited lethality. This is in stark contrast to the viability of KNL1-deficient plants and indicates that the temporal control of the SAC may be far more critical in the plant context than simply the recruitment of SAC components to kinetochores. In addition, in complemented seedlings carrying the KNL1 mutant, it was rare to observe cells undergoing mitosis. This again contrasts with KNL1-deficient plants where cells could still progress through mitosis, albeit with chromosome missegregation issues. These findings collectively suggest that the interaction between KNL1 and TOPP proteins plays a pivotal and multifaceted role not only in SAC regulation but also in broader coordination of the cell cycle machinery in plant cells.

In fungi and animals, PP1 at kinetochores is reported to regulate microtubule-kinetochore interactions (36, 54–56), a function that may also be disrupted by the perturbation of KNL1-PP1 binding in plants. This raises the intriguing possibility that the KNL1-TOPP axis in plants has evolved specific adaptations to integrate the microtubule-kinetochore attachments with the timely inactivation of the SAC. Further research is necessary to elucidate the precise molecular mechanisms by which the association of TOPP with KNL1 helps to fine-tune this critical balance in plant cell division. These insights may uncover specific regulatory strategies that plants have evolved to maintain genome stability and facilitate robust organismal development, even in the face of some degree of chromosome segregation issues during mitosis. Unraveling the expanded functional repertoire of the KNL1-TOPP axis in plants could yield important perspectives on the evolution of cell cycle control mechanisms across eukaryotes.

MATERIALS AND METHODS

Plant materials and growth conditions

A. thaliana ecotype Columbia was used as the wild-type control in this study. The knl1 (SALK_068970), bmf3 (SALK_032111), and mad1 (SALK_073889) mutant lines were described previously (18). All plants were grown in an environmentally controlled growth chamber maintained at 22°C under a 16-hour light and 8-hour dark photoperiod.

For genetic manipulations and transgene expression, the Agrobacterium tumefaciens strain GV3101 was used to transform the Arabidopsis plants via floral dip. Seedlings for live-cell imaging and immunolocalization experiments were grown on solid medium containing one-half strength Murashige and Skoog (MS) nutrients and 0.8% phytagel. For the pharmacological experiments, Arabidopsis seedlings were initially grown on one-half–MS solid medium. These seedlings were then transferred to one-half–MS plates supplied with either 100 nM oryzalin or 10 μM tautomycin dissolved in dimethyl sulfoxide (DMSO). Plates containing equal volumes of DMSO were used as the mock control treatment.

Plasmid construction

To generate the TOPP-GFP fusion constructs, the genomic regions encompassing the promoter and coding sequences of each TOPP gene (TOPP1 to TOPP9) were amplified and cloned into the pENTR4 entry vector using the Gibson Assembly method with BamHI and EcoRI sites. The entry vector containing the genomic KNL1 sequence had been described previously (18). Site-directed mutagenesis was used to create the KNL1^4A and TOPP^T246M mutant constructs using the wild-type KNL1 and TOPP4 entry vectors as templates, respectively. These entry vectors containing the native and mutant sequences were then recombined with the destination vectors pGWB650 (for GFP fusion) or pGWB510 (for FLAG fusion) using LR recombination reactions to generate the final expression constructs. The binary vectors harboring BMF3-GFP, GFP-MAD1, histone H1.2-TagRFP, and GFP-TUB6 had been reported in our previous studies (18, 31, 57, 58).

Y2H assay

For Y2H, full-length cDNAs for the proteins of interest were amplified from an Arabidopsis cDNA library and cloned into the pENTR4 entry vector. The subcloned cDNAs were then recombined into the pGADT7-GW (AD) (Addgene, plasmid #61702) and pGBKT7-GW (BD) (Addgene, plasmid #61703) destination vectors using LR recombination reactions. The resulting constructs were transformed into the yeast strain AH109. The transformed yeast cells were spotted onto two types of selective media, control media lacking leucine and tryptophan (-L/-W) and selection media lacking leucine, tryptophan, histidine, and adenine (-L/-W/-H/-A). The yeast colonies were allowed to grow at 30°C for 2 days before being photographed.

In vitro protein expression and pull-down assay

To produce recombinant proteins, entry clones containing coding sequences of KNL1^N, KNL1^N-4A, TOPP4, and TOPP8 were recombined with pDEST565 (Addgene, plasmid #11520) or pDEST566 (Addgene, plasmid #11517) by Gateway cloning through LR clonase. These vectors allowed the production of GST- or MBP-tagged fusion proteins in Escherichia coli BL21 host cells. The recombinant fusion proteins were purified using either Glutathione Agarose Resins (Yeasen Biotechnology, catalog no. 20507ES) or MBPSep Dextrin Agarose Resins (Yeasen Biotechnology, catalog no. 20515ES), according to the manufacturer’s instructions.

For the pull-down assays, the GST-tagged KNL1^N and KNL1^N-4A proteins, as well as GST alone, were incubated with equal amounts of the MBP-tagged TOPP4 or TOPP8 proteins immobilized on the MBP-trap resins. The binding reactions were performed in pull-down buffer [20 mM tris-HCl (pH 8.0), 150 mM NaCl, and 0.2% Triton X-100] at 4°C for 2 hours with constant rotation. After washing for five times with wash buffer [20 mM tris-HCl (pH 8.0), 150 mM NaCl, and 0.1% Triton X-100], the beads were collected by centrifugation. Bound proteins were eluted by boiling in 50 μl of 1× SDS sample buffer for 10 min at 100°C and examined by SDS–polyacrylamide gel electrophoresis and immunoblotting using anti-MBP and anti-GST antibodies (Yeasen Biotechnology, catalog no. 30401ES and no. 30901ES; 1:10,000).

Coimmunoprecipitation assay

Total proteins were extracted from the 3-day-old transgenic Arabidopsis plants expressing either pTOPP4::TOPP4-GFP/pKNL1::KNL1-FLAG or pTOPP8::TOPP8-GFP/pKNL1::KNL1-FLAG. These protein extracts were incubated with anti-GFP beads (Miltenyi Biotech, catalog no. 130-091-125) in immunoprecipitation buffer containing 10 mM tris (pH 7.5), 0.5% Nonidet P-40, 2 mM EDTA, 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, and 1% protease inhibitor cocktail (Thermo Fisher Scientific, catalog no. A32965). The beads were then collected, washed at least five times with the immunoprecipitation buffer, recovered, and mixed with 1× SDS loading buffer. Both the input (total proteins) and immunoprecipitated samples were subjected to immunoblotting using anti-FLAG (Cell Signaling Technology, catalog no. 8146; 1:2000) or anti-GFP (Thermo Fisher Scientific, catalog no. MA5-15256; 1:2000) antibodies.

Transient expression of fusion proteins in tobacco leaves

The wild-type Nicotiana benthamiana plants were grown in soil at 26°C under a 16-hour light and 8-hour dark cycle. Leaves of 5-week-old plants were infiltrated with A. tumefaciens strain GV3101 carrying the plasmids of interest. The A. tumefaciens cultures were grown overnight at 28°C, harvested by centrifugation, washed twice with one-half–MS liquid medium, and resuspended to an optical density at 600 nm of 0.8 before infiltration.

For the BiFC experiments, the entry clones containing the coding sequences of the tested proteins were recombined into the pGTQL1211YN [N-terminal yellow fluorescent protein (YFP) fusion] (Addgene, plasmid #61704) and pGTQL1221YC (C-terminal YFP fusion) (Addgene, plasmid #61705) vectors using LR clonase. The A. tumefaciens strains carrying the nYFP and cYFP constructs were mixed in a 1:1 ratio and coinfiltrated into young N. benthamiana leaves. After 48 hours, the YFP fluorescence resulting from the reconstitution of the split YFP fragments was examined using confocal microscopy with 488-nm excitation and 500- to 550-nm emission.

Immunolocalization and fluorescence microscopy

Immunofluorescence staining experiments were performed using Arabidopsis root meristematic cells. Root tips from 5-day-old Arabidopsis seedlings were excised and fixed for 45 min in PME buffer [50 mM Pipes (pH 6.9), 5 mM MgSO₄, 1 mM EGTA, and 4% formaldehyde] at room temperature. After fixation, the cells were treated with 1% Cellulase (Onozuka RS) for 10 min to facilitate permeabilization. The fixed cells were then immobilized on slides and permeabilized further with 0.5% Triton X-100 for 15 min and methanol for 10 min before incubation with the primary antibodies. Primary antibodies used in this study included GFP recombinant rabbit monoclonal antibody (Thermo Fisher Scientific, catalog no. G10362; 1:400), mouse anti-FLAG monoclonal antibody (Cell Signaling Technology, catalog no. 8146; 1:400), and DM1A mouse α-tubulin monoclonal antibody (Abcam, catalog no. ab7291; 1:800). Secondary antibodies were Alexa Fluor 488–conjugated goat anti-rabbit immunoglobulin G (IgG) and Alexa Fluor 555–conjugated goat anti-mouse IgG (Thermo Fisher Scientific, catalog no. A32731 and no. A32727; 1:1000). The stained cells were observed using an Eclipse 600 fluorescence microscope equipped with a 100× Plan-Apo objective [numerical aperture (NA) 1.45, Nikon]. Images were acquired by a panda sCMOS camera (PCO Imaging).

For live-cell imaging, root meristematic cells from 5-day-old Arabidopsis seedlings were observed using an LSM880 spinning-disk confocal microscope with a 100× oil-immersion objective (NA 1.4, Carl Zeiss). GFP was excited with a 488-nm laser and detected between 500 and 550 nm, while TagRFP were excited with a 561-nm laser and detected between 570 and 650 nm. Time-lapse imaging was performed using the ZEN software package (Carl Zeiss) and processed in ImageJ/Fiji (https://fiji.sc/).

Supplementary Materials

The PDF file includes:

Figs. S1 to S19

Table S1

Legends for movies S1 to S17

Other Supplementary Material for this manuscript includes the following:

Movies S1 to S17