KNTC1 introduces segmental heterogeneity to mitochondria

Abstract

Mitochondria contribute to cellular metabolism by providing a specialised milieu for energising cells by incorporating and processing the metabolites. However, heterogeneity between mitochondria has only partially been elucidated. Mitochondria dynamically alter their morphology and function during the life of an animal, when cells proliferate and grow. We here show that Kntc1, a highly evolutionarily conserved protein, translocates from the Golgi apparatus to linear mitochondrial segments (LMSs) upon glutamine deprivation and plays an essential role in maintaining LMSs. The LMSs to which Kntc1 localised exhibited an increase in the mitochondrial membrane potential, suggesting the role of Kntc1 in functioning as a reservoir for the energy-generating potential. Suppression of Kntc1 led to glutamine consumption and lactate production, thus impacting cellular metabolism, eventually leading to anchorage-independent growth of cells. Indeed, a KNTC1 variant was identified in a patient with ovarian cancer, suggesting that segmental regulation of the mitochondrial function is essential for maintaining tissue integrity.

Keywords: Bent mitochondrial segment; Glutamine metabolism; KNTC1; Linear mitochondrial segment; Mitochondrial structural heterogeneity.

Fig. 1.

Kntc1 senses glutamine loss. (A) Confocal images of dividing NIH3T3 cells immunostained for Kntc1 (green) showing kinetochore-associated localisation of Kntc1. α-tubulin (pink) was stained to visualise the mitotic spindle. Nuclei were stained with DAPI (blue). (B) Kntc1 forms a filamentous structure upon glutamine deprivation for 16 h. Arrowheads indicate Kntc1 signals in GM130-stained Golgi apparatus (normal condition) and the linear structures (glutamine deprivation). (C) Quantification of cells exhibiting the relocalisation of Kntc1 signals. (D) Immunostaining for Kntc1 in NIH3T3 cells transfected with control oligonucleotide or siRNA against Kntc1. Cells were stained with Kntc1 and GM130 antibodies 72 h after the transfection. (E) Real-time qPCR analysis for Kntc1 in NIH3T3 cells transfected with control oligonucleotide or siRNA against Kntc1. RNA was extracted 48 h after transfection. Data were normalised to Polr2a and show expression relative to that of siCtrl. (F) Western blot analysis of Kntc1 showing the knockdown efficiency utilising the siRNA against Kntc1. GAPDH was used to show equal loading. (G) The filamentous structure of Kntc1 was induced by treatment with acivicin, which blocks glutamine metabolism. Boxed areas are shown at 1.6× magnification on the right. ***P<0.001, unpaired two-tailed Student's t-test. All scale bars: 10 μm

Fig. 2.

Kntc1 localises to linear mitochondrial segments. (A) Kntc1 localises to mitochondria upon glutamine loss in NIH3T3 cells. After treatment with acivicin or glutamine deprivation Kntc1 forms linear segments. Mitochondria were stained with MitoTracker Red CMXRos. Scale bars: 10 μm. Boxed areas are shown at 2× magnification underneath. (B) Kntc1 forms linear segments in mitochondria. The length (in μm) and linearity (reciprocal of curvature) of mitochondria and Kntc1 macrostructures was measured using the Fiji software Kappa. (C) Statistical analysis of linearity (in μm) of the mitochondrial segments that were marked by Kntc1 localisation. (D) Schematic describing preferential localisation of Kntc1 to the linear mitochondrial segment (LMS) but not the bent mitochondrial segment (BMS). ***P<0.001, unpaired two-tailed Student's t-test.

Fig. 3.

Kntc1 depletion compromises mitochondrial linearity. (A) Mitochondrial morphology was assessed in NIH3T3 cells, which were depleted for Kntc1 and treated with acivicin. Mitochondria were stained with MitoTracker Red CMXRos. Nuclei were stained with DAPI. Scale bars: 10 μm. Boxed areas are shown at 2× magnification underneath. (B) Linearity analysis of mitochondria, indicating that Kntc1 is indispensable for the linear mitochondrial structure. (C) Statistical analysis for the segmental linearity in mitochondria. ***P<0.001, one-way ANOVA with post hoc Tukey's test.

Fig. 4.

Kntc1-LMS colocalises with Mfn1 and Mfn2. (A) NIH3T3 cells were treated with mdivi-1, a DRP1 inhibitor, for 16 h before immunofluorescence analysis against Kntc1. Mitochondria were stained with MitoTracker Red CMXRos. (B) Acivicin treatment of NIH3T3 cells facilitates LMS formation by Kntc1 accompanied by Mfn1 localisation to the LMS. Arrowheads indicate the segments to which Kntc1 and Mfn1 localise. (C) Acivicin treatment of NIH3T3 cells facilitates LMS formation by Kntc1 accompanied by Mfn2 localisation to the LMS. Arrowheads indicate the segments to which Kntc1 and Mfn2 localise. Boxed areas are shown magnified on the right. All scale bars: 10 μm.

Fig. 5.

The LMS harbours a specific metabolic profile. (A,B) Colocalisation of Kntc1 (green) and mitochondria (magenta) labelled with MitoTracker Red CMXRos (CMXROS) in mouse NIH3T3 cells. Nuc, nucleus. Boxed rectangular areas in A show colocalisation of Kntc1 and MitoTracker Red CMXRox signals and are shown magnified in B. (C) Colocalisation of Kntc1 and MitoTracker Red CMXRos signals reflecting the membrane potential. The intensity of MitoTracker CMXRos is higher in mitochondrial segments to which Kntc1 localises (black arrowheads), which possess higher membrane potential, than in mitochondrial segments to which Kntc1 does not localise (white arrowheads). RFI, relative fluorescence intensity. (D) Colocalisation of Cox1 (green) and CMXRos (magenta). Boxed rectangular area in the middle image shows colocalisation of Cox1 and CMXROS and is shown magnified on the right. (E) Mutually exclusive localisation of Cox1 and mitochondria (CMXRos). (F) Colocalisation analysis of Kntc1 (green) and Cox1 (magenta). Boxed rectangular area is shown magnified in the right image, with arrowheads indicating the Kntc1-localising segment without Cox1 expression. (G) Mutually exclusive localisation of Kntc1 and Cox1. (H) Schematic for the association of LMS with the high membrane potential and not with Cox1. Scale bars: 5 μm. Nuclei were stained with DAPI (blue).

Fig. 6.

Kntc1 loss provokes glutamine overuse. (A) Kntc1 mRNA levels in Neuro2a cells transfected with siRNA targeting Kntc1 (siKntc1) or control siRNA (siCtrl). Data were normalised to those of Gapdh. (B) Growth analysis of cells that had been depleted of Kntc1 and were grown under normal nutrient conditions (top) or under conditions of glutamine depletion. Scale bars: 200 μm. (C) Quantification of cells described in B. (D) Image of a culture plate of Neuro2a cells transfected with siKntc1 or siCtrl. A change in colour from orange to yellow was observed for the culture medium of Kntc1-depleted cells (bottom). (E) Lactate production analysed in culture medium of siKntc1 or siCtrl cells. (F) Glutamine consumption measured in the culture medium of siKntc1 or siCtrl cells. (G) Glutamine consumption in the culture medium of KNTC1-transfected cells. (H) Glutamate levels measured in siKntc1 or siCtrl cells. (I) α-ketoglutarate levels measured in siKntc1 or siCtrl cells. (J) Assessment of glutamine dependency of Kntc1-depleted cells. The effect of acivicin treatment (+) or control (−) on cell growth was analysed in siKntc1 or siCtrl cells. *P<0.05, **P<0.01, ***P<0.001 unpaired two-tailed Student's t-test (A,C,E-I). ***P<0.001, one-way ANOVA with post hoc Tukey's test (J).

Fig. 7.

KNTC1 loss is associated with tumorigenesis. (A) KNTC1 knockdown efficiency in Kuramochi cells 48 h after transfection of siRNA targeting KNTC1 (siKNTC1) or control siRNA (siCtrl). Expression levels of KNTC1 were normalised to those of POLR2A. **P<0.01, unpaired two-tailed Student's t-test. (B) Soft agar anchorage-independency analysis performed with siKNTC1 or siCtrl cells as described in A. To rescue the KNTC1 depletion, the KNTC1 plasmid that was mutated so that the KNTC1 mRNA is not targeted by the siRNA was concomitantly transfected. The soft agar was stained with Crystal Violet to visualise the colonies. (C) Quantification of colony numbers as described in B. ***P<0.001, one-way ANOVA followed by Tukey's multiple comparison tests. (D) Western blot analysis of KNTC1 in Kuramochi cells transfected with siKNTC1 or a KNTC1-expressing plasmid. GAPDH is shown as the loading control. (E) Schematic of the genomic structure of KNTC1 showing the p.Ser40Phefs*2 variant identified in the patient. Usage of the ScanProsite tool identified a basic leucine zipper (bZIP) domain. a.a., amino acids. (F) Histological analysis of ovarian cancer. Boxed areas surround carcinoma cells that formed solid cell clusters within the ovary (f1,f3) or around the fallopian tube (f2), with magnified images on the right. (G) Schematic describing the role of Kntc1 in cell metabolism.