A CLK1-KKT2 Signaling Pathway Regulating Kinetochore Assembly in Trypanosoma brucei

Given the importance of kinetochore movement during metaphase in eukaryotes (23), we assessed the impact of T. brucei CLK1 (TbCLK1) activity on kinetochore dynamics using AB1 as a chemical tool. The expression and localization of kinetochore proteins, labeled with mNeonGreen (mNG), were assessed by confocal microscopy in the bloodstream form of the parasite (Fig. 1A). Similar to previous observations in procyclic form cells (15, 17), we observed different kinetochore timings and patterns of expression throughout the cell cycle. By using the kinetoplast (K) and nucleus (N) configuration to define each cell cycle stage (24), we observed that KKT2 and KKT3 are constitutively expressed until anaphase. KKT1 and KKIP1 gradually load from S phase onwards until the end of mitosis, while KKT4 and KKT5 expression is restricted to metaphase. Furthermore, KKT9 and KKIP7 expression diminish during anaphase, suggesting both proteins are acting as scaffolds for the recruitment of multiple other components. Treatment with 5× the 50% effective concentration (EC₅₀) of AB1 for 6 h caused dispersal, to various degrees, for KKT1, KKT2, KKT5, KKT9, KKT13, KKT14, and KKT20 from the defined foci of the kinetochore within the nucleus, while KKT3, KKT7, KKT11, KKIP1, and KKIP7 remained in distinct foci (Fig. 1B and C; see also Fig. S1A in the supplemental material). Automated focus detection using subpixel precise single-particle localization combined with image segmentation (25) and intensity quantification (26) determined that there was a significant reduction in focus intensity for KKT1, KKT2, KKT4, KKT5, and KKT9 but not KKT3 (Fig. 1D and Fig. S1B and C). No degradation of these proteins was observed after treatment (Fig. S1D). These results suggest that although KKT2 and KKT3 are centromere-anchored proteins (15), they respond differently to CLK1 inhibition, and that TbCLK1 is a critical regulator of inner kinetochore component dynamics.

KKT2 and KKT3 protein kinases are likely components of the trypanosome inner kinetochore with functional equivalence to the constitutive centromere-associated network (CCAN), a canonical component of the eukaryotic inner kinetochore (27). Defective KKT2 clustering was also observed after CLK1 RNA interference (RNAi) (22). It has been reported that phosphorylation of kinetochore proteins has critical roles in kinetochore organization and interaction during mitosis in mammals and yeast (28). Indeed, cell cycle-regulated changes in the phosphorylation of T. brucei kinetochore components have been reported recently, where the regulation is coordinated with phosphorylation of essential protein kinases, including CLK1 (19).

We speculated that KKT2 provides a platform on which the kinetochore multiprotein complex assembles and that phosphorylation orchestrates this process. To address whether KKT2 is a CLK1 substrate, we first analyzed mobility shifts of phosphorylated forms of KKT2 and KKT3 using Phos-tag gel electrophoresis (29). A low-mobility, nonphosphorylated form of KKT2 was detected after treatment with AB1 or after CLK1 depletion by RNAi, while KKT3 remained unaffected (Fig. S2). Six phosphorylation sites have been identified in KKT2 (S⁵, S⁸, S²⁵, S⁵⁰⁷, S⁵⁰⁸, and S⁸²⁸) (30), and we tested if these are important for KKT2 function by generating a KKT2 RNAi line (Fig. S3) with a recoded hemagglutinin (HA)-tagged version of KKT2 integrated into the tubulin locus. This constitutively expressed KKT2 (KKT2^R) is not susceptible to RNAi-mediated degradation, and KKT2^R complements the loss of function of KKT2 48 h after RNAi induction (Fig. 2A). Replacement of Ser for Ala in KKT2 at positions S⁵, S⁸, S²⁵, and S⁸²⁸ resulted in complementation of KKT2 function when expressed in the RNAi line. In contrast, dual replacement of the KKT2 phosphorylation sites S⁵⁰⁷ and S⁵⁰⁸ with Ala (KKT2^S507A-S508A) failed to complement depletion of the wild-type KKT2 with respect to parasite growth (Fig. 2B) or cell cycle progression after 48 h of induction (Fig. 2C). The efficacy of RNAi knockdown of the endogenous KKT2 alleles was retained in these derivative cell lines (Fig. S3A, B, and C), demonstrating that the complementation effects were imparted by the recoded alleles. KKT2^S507A-S508A had good expression levels in the cell (Fig. 2B, lower) but was mislocalized (Fig. S3D), providing a possible explanation for the phenotype observed. These defects phenocopy the effect of AB1 and show the importance of the two phosphorylation sites for the function of KKT2. To assess whether protein kinase activity is essential for KKT2 function, an active-site mutant was generated in KKT2^R (KKT2^K113A). A significant loss of function was observed after 48 h of induction, indicating protein kinase activity is essential for KKT2 function but not for regulating cell cycle progression (Fig. 2B and C).

To address whether CLK1 phosphorylates KKT2 directly at S^507-508 residues, we expressed a recombinant peptide (amino acids [aa] 486 to 536) of KKT2, including mutations of S507 and S508 residues. We demonstrated that recombinant CLK1 could phosphorylate recombinant KKT2 in vitro at positions S507 and S508 (Fig. 2D). Given the conservation of KKT2 S⁵⁰⁸ in kinetoplastids, we then raised a phosphospecific antibody against KKT2^S508 to monitor KKT2 phosphorylation through the T. brucei cell cycle and after treatment with AB1. The antibody specifically recognizes phosphorylation of KKT2^S508, as phosphorylated KKT2^S508 was depleted following KKT2 or CLK1 RNAi (Fig. 2E, upper) or after treatment with AB1 in a cell line where KKT2 was endogenously tagged with Ty and mNG (Fig. 2E, lower; both endogenous KKT2 and Ty-mNG KKT2 are detected). In addition, the KKT2 phosphoantibody detects phosphorylated KKT2 in all the recoded mutants except the KKT2^S507A-S508A double mutant (Fig. S3E). KKT2^S508 phosphorylation was found to increase in S-phase after hydroxyurea synchronization and progressively decreased toward G₁ phase (Fig. 2F), in correlation with the recent demonstration of dynamic KKT2 S508 phosphorylation during the cell cycle (31). Together, these data show that KKT2 phosphorylation is downstream of CLK1 in a kinetochore-specific signaling cascade and occurs during early metaphase.

We next assessed whether KKT2 phosphorylation is required for recruitment of proteins to the trypanosome kinetochore. KKT1 and KKT9 recruitment were impaired in the KKT2^RS^507A-S508A::KKT2-induced cell line (Fig. 3A to C) but not the KKT2^R K^113A-induced line (Fig. S4A), underlining the importance of KKT2 phosphorylation by CLK1 for kinetochore assembly. Individual expression of phosphomimetics S^507E and S^508E impaired KKT1 and KKT9 recruitment but also affected the timing of events during mitosis, with a notable defect in nuclear abscission (Fig. S4B).

Faithful chromosome segregation relies on the interaction between chromosomes and dynamic spindle microtubules (32). Furthermore, spindle elongation is important for correct segregation of chromosomes during anaphase (33). To further examine if CLK1 inhibition impairs microtubule spindle dynamics, we observed the expression of the mitotic spindle by staining the parasites with KMX-1 antibody and analyzing the microtubule-associated protein 103 kDa (MAP103) (Fig. S5) (34). This showed that treatment with AB1 does not affect microtubule spindle formation (Fig. 4A). Considering that CLK1 inhibition during metaphase results in an arrest in late anaphase (19, 22), it is likely that the function of CLK1 during cytokinesis is related to either the control of kinetochore-spindle microtubule attachment errors or its interactions with the chromosomal passenger complex (CPC). Of note, it has been reported that T. brucei aurora kinase B has an important role during metaphase-anaphase transition and the initiation of cytokinesis via regulation of the CPC (35–37) and nucleolar and other spindle-associated proteins (NuSAPs) (38).

In mammals, kinetochore assembly is enhanced by mitotic phosphorylation of the Dsn1 kinetochore protein by aurora kinase B, generating kinetochores capable of binding microtubules and promoting the interaction between outer and inner kinetochore proteins (39). In T. brucei, aurora kinase B (TbAUK1) plays a crucial role in spindle assembly, chromosome segregation, and cytokinesis initiation (37). Therefore, we asked if CLK1 and AUK1 are part of the same signaling pathway. We showed that treatment with AB1 does not affect spindle formation (Fig. 4A), in contrast to the inhibition of AUK1 (40). AUK1 is a key component of the trypanosome CPC (41). To understand if CPC dynamics are impaired by CLK1 inhibition, we monitored the localization of CPC1 throughout the cell cycle before, after AB1 treatment, and following AUK1 inhibition by Hesperadin (42). After treatment with AB1, CPC1 showed a dispersed nuclear pattern that progressively disappeared after abscission of the nucleus (Fig. 4B, middle). This was different from AUK1 inhibition by Hesperadin, which prevented translocalization of the CPC from the spindle midzone, impairing initiation of cytokinesis (Fig. 4B, right). Finally, we confirmed that AUK1 is not involved in kinetochore assembly, since neither KKT2^S508 phosphorylation nor KKT2 localization was affected by AUK1 inhibition by Hesperadin (Fig. 4C). In addition, a cohort of divergent spindle-associated proteins has been described that is required for correct chromosome segregation in T. brucei (18). Therefore, we analyzed the subcellular localizations of NuSAP1 and NuSAP2 during the cell cycle after CLK1 inhibition. NuSAP2 expression in the central portion of the spindle after metaphase release was compromised by CLK1 inhibition, while NuSAP1 remained unaffected (Fig. 4D). NuSAP2 is a divergent ASE1/PRC1/MAP65 homolog, a family of proteins that localizes to kinetochore fibers during mitosis, playing an essential role in promoting the G₂/M transition (43). Considering that NuSAP2 and KKT2 colocalize during interphase and metaphase (18), it is likely that KKT2 regulation by CLK1 influences posterior spindle stability and cytokinesis.

All transgenic T. brucei brucei parasites used in this study were derived from monomorphic T. brucei brucei 2T1 bloodstream forms (63) and were cultured in HMI-11 (HMI-9 [GIBCO] containing 10%, vol/vol, fetal bovine serum [GIBCO], Pen-Strep solution [penicillin at 20 U ml⁻¹, streptomycin at 20 mg ml⁻¹]) at 37°C and 5% CO₂ in vented flasks. Selective antibiotics were used: 5 μg ml⁻¹ blasticidin or hygromycin and 2.5 μg ml⁻¹ phleomycin or G418. RNAi was induced in vitro with tetracycline (Sigma-Aldrich) in 70% ethanol at 1 μg ml⁻¹. The endogenous Ty, mNeonGreen experiment was performed using the pPOTv6 vector (64). The generation of inducible TbCLK1 and KKT2 RNAi was done as previously described (20).

Recoded KKT2 was synthesized by Dundee Cell Products. The recoded KKT2 sequence (KKT2^R) codes for the same amino acid sequence as KKT2 but only shares 94.23% nucleotide identity. All segments of identity between KKT2 and KKT2^R are less than 20 bp long. KKT2^R was inserted into the plasmid pGL2243 using XbaI and BamHI restriction sites, generating pGL2492. This plasmid is designed to constitutively express KKT2 from the tubulin locus, with the addition of a C-terminal 6× HA tag. To express catalytically inactive KKT2 and phosphomutants, the active-site lysine (K¹¹³) and serine (S⁵, S⁸, S²⁵ S^507-S508, and S⁸²⁸) were changed to alanine by mutating pGL2492, carrying the coding sequence for KKT2, using site-directed mutagenic PCR. A list of primers is provided in Text S1 in the supplemental material. To generate individual KKT2 recoded mutants, correspondent KKT2^R plasmids (described above) were transfected into the KKT2 RNAi cell line. Localization of endogenous KKT1 and KKT9 in KKT2^R mutants was assessed by microscopy after transfection of the correspondent mNG-KKT1 or mNG-KKT9 pPOTv6 vector into each recoded cell line.

Cells treated for 6 h with compounds or dimethyl sulfoxide (DMSO) and were centrifuged at 1,400 × g for 10 min before washing twice with Trypanosoma dilution buffer (TDB)-glucose at room temperature. Suspensions were centrifuged at 1,000 × g for 5 min, pipetted into 6-well microscope slides, and dried at room temperature (RT). Cells were fixed with 25 μl of 2% paraformaldehyde diluted in phosphate-buffered saline (PBS) and incubated at room temperature for 5 min. Cells were washed in PBS to remove paraformaldehyde prior to washing twice more with PBS and permeabilized with 0.05% NP-40 for 10 min. Cells were washed twice in PBS and dried at RT. Mounting medium with 4′,6-diamidino-2-phenylindole (DAPI) was added to each well with a coverslip. Slides were kept at 4°C before viewing using a Zeiss LSM 880 with Airyscan on an Axio Observer.Z1 invert confocal microscope.

Ty-NuSAP1 and Ty-NuSAP2 were detected by indirect immunofluorescence by using a mouse Imprint monoclonal anti-Ty1 antibody (clone BB2). Briefly, cells were harvested by centrifugation at 1,400 × g for 10 min at room temperature, washed, and resuspended in TDB-glucose. A total of 2 × 10⁵ cells were dried on slides, fixed in 1% paraformaldehyde (PFA) for 1 h, washed with PBS, blocked with 50% (vol/vol) fetal bovine serum for 30 min, and then incubated with anti-TY (1:800) diluted in 0.5% blocking reagent for 1 h. Alexa-Fluor 488 (anti-mouse) was used as the secondary antibody (Invitrogen). Cells were DAPI stained and visualized using a Zeiss LSM 880 with Airyscan on an Axio Observer.Z1 inverted confocal microscope.

To study spindle formation, wild-type bloodstream forms were treated or not for 6 h with AB1 (5× EC₅₀) or CLK1 RNAi cells were treated or not with tetracycline for 24 h. Parasites were harvested by centrifugation at 1,400 × g for 10 min and then washed twice with TDB-glucose at room temperature. Samples were fixed for 10 min in 2%, wt/vol, formaldehyde in PBS, followed by 5 min incubation with 1 M Tris, pH 8.5, to quench the fixation. The fixed cells were washed with PBS, suspended in PBS, and adhered to SuperFrost Plus adhesion slides for 15 min. Attached parasites were then permeabilized with methanol at –20°C for 15 min and rehydrated with PBS, followed by incubation with blocking buffer (5% bovine serum albumin, 0.1% Triton X-100 in PBS) for 1 h at room temperature. Cells were immunostained at room temperature for 1 h with KMX-1 antibody to detect the mitotic spindle. After three washes (0.1% Triton X-100 in PBS), samples were incubated for 1 h with an Alexa Fluor 488-conjugated goat anti-mouse IgG (used at 1:300) secondary antibody. Finally, after three more washes, the slides were mounted in ProLong diamond antifade mountant with DAPI and examined by fluorescence microscopy. For analysis, 2K1N and 2K2N populations (n = 80) were considered, and statistical significance determined using the Holm-Sidak t test, with α = 0.05.

For cell cycle analysis, bloodstream-form T. brucei cell lines were incubated or not for 6 h with AB compounds at a final concentration of 5× the individual EC₅₀ for each compound (averaged from viability assays). Control cultures were treated with 0.5 μl DMSO. Cultures were pelleted and cells were collected and washed once in TDB supplemented with 5 mM EDTA and resuspended in 70% methanol. Cells were centrifuged at 1,400 × g for 10 min to remove methanol and washed once in TDB with 5 mM EDTA. Cells were resuspended in 1 ml 1× TDB with 5 mM EDTA, 10 μg ml⁻¹ propidium iodide, and 10 μl of RNase A. Cell suspensions in 1.5-ml tubes were wrapped in foil to avoid bleaching by light. Cells were incubated for 30 min at 37°C in the dark until fluorescence-activated cell sorting (FACS) analysis. Cells were analyzed for FACS using a Beckman Coulter CyAn ADP flow cytometer (excitation, 535 nm; emission, 617 nm). Cell cycle phase distribution was determined by fluorescence.

Hydroxyurea-induced synchronization of cell lines was obtained by incubating parasites in exponential growth phase with 10 μM hydroxyurea (HU) (Sigma-Aldrich) for 6 h. Removal of HU from the culture medium was achieved by centrifuging cells at 1,400 × g for 10 min, washing twice with fresh (drug-free) medium, and resuspending cells in medium lacking HU. Subsequently, samples were collected each hour for posterior cell cycle analysis by propidium iodide staining.

KKT2 and KKT3 phosphorylation profile were analyzed by using a SuperSep Phos-tag precast gel (29) according to the manufacturing protocol. Briefly, Ty-mNG KKT2 and Ty-mNG KKT3 were incubated with 5× AB1 EC₅₀ for 18 h and collected for analysis by Western blotting in an EDTA-free radioimmunoprecipitation assay (RIPA) lysis buffer. In parallel, the expression of both proteins was analyzed after 24 h for TbCLK1 RNAi. After electrophoresis, the gel was washed 5 times with 10 mM EDTA transfer buffer to improve transference. The membrane then was transferred to a polyvinylidene difluoride (PVDF) membrane using a 0.1% SDS Tris-glycine transfer buffer at 90 mA overnight at 4°C. The membrane was blocked for 1 h with 10% bovine serum albumin (BSA) and KKT2 and KKT3 phosphorylation pattern was analyzed by using an anti-Ty1 antibody (see Text S1 for details).

Anti-phospho KKT2 S⁵⁰⁸ was raised against a synthetic phosphopeptide antigen C-GTRVGS(pS*)LRPQRE-amide, where pS* represents phosphoserine. The peptide was conjugated to keyhole limpet hemocyanin (KLH) and used to immunize rabbits. Phosphopeptide-reactive rabbit antiserum was first purified by protein A chromatography. Further purification was carried out using immunodepletion by nonphosphopeptide resin chromatography, after which the resulting eluate was chromatographed on a phosphopeptide resin. Anti-antigen antibodies were detected by indirect enzyme-linked immunosorbent assay with unconjugated antigens passively coated on plates, probed with anti-IgG-horseradish peroxidase conjugate, and detected with 2,2′-azinobis(3-ethylbenzthiazolinesulfonic acid) substrate. Posterior antigen specificity was confirmed by Western blotting using KKT2 RNAi and endogenous tagged KKT2 cell lines. Custom antibody was produced by Thermo Fisher Scientific.

For Western blotting, parasites were washed with TDB supplemented with 20 mM glucose. After centrifugation, the samples were resuspended in the RIPA buffer (number 9806S; New England Biolabs) supplemented with protease and phosphatase inhibitors obtained from Promega and Roche Life Science, respectively. All samples were quantified by Bradford protein assay (Bio-Rad), 25 μg of protein was loaded, resolved in a 4 to 20% NuPAGE Bis‐Tris gel (Invitrogen) in NuPAGE morpholinepropanesulfonic acid running buffer, and transferred onto Hybond‐C nitrocellulose membranes (GE Healthcare) at 350 mA for 2 h or, for high-molecular-weight proteins, overnight at 4°C.

After transfer, membranes were washed once in 1× TBST (Tris-buffered saline [TBS], 0.01% Tween 20 [Sigma-Aldrich]) for 10 min and then incubated for 1 h in blocking solution (1× TBST, 5% BSA) or, if required, overnight at 4°C. Next, the membrane was rinsed for 10 min in 1× TBST and placed in blocking buffer containing the required primary antisera for 1 h at room temperature or overnight at 4°C. The membrane was then washed 3 times with TBST and placed in blocking solution containing the appropriate fluorescent secondary antisera for 1 h. A list of antibodies is provided in Text S1.

All statistical analyses were performed using GraphPad Prism 8 (http://www.graphpad.com/scientific-software/prism/). The appropriate tests were conducted and are detailed in the corresponding figure legends.