In mammals, cell cycle progression is dependent upon highly ordered events controlled by cyclins and cyclin-dependent kinases (Cdks)1. Through its cytoplasmic, nuclear, and centrosomal localization, CyclinB1/Cdk1 is able to synchronize different events in mitosis such as nuclear envelope breakdown and centrosome separation2. CyclinB1/Cdk1 protects mitotic cells against apoptosis3 and promotes mitochondrial fission, a critical step for an equal distribution of mitochondria to the newly formed daughter cells4.
In proliferating mammalian cells, mitochondrial ATP is generated via oxidative phosphorylation (OXPHOS) machinery (electron transport chain), which is composed of 5 multi-subunit complexes; complex I - complex V (CI-CV). Nicotinamide adenine dinucleotide (NADH):ubiquinone oxidoreductase or complex I (CI) is the largest and least understood of the five complexes5. The complex consists of 45 subunits, 14 of which form the catalytic core. Once assembled, the complex assumes an L-shaped structure with one arm protruding into the matrix and the other arm embedded in the inner membrane6,7. Mutations in CI subunits are the cause of a variety of mitochondrial disorders8. A functionally efficient CI in OXPHOS is required not only for overall mitochondrial respiration9, but also for successful cell cycle progression10. Unravelling the mechanisms underlying the functioning of this membrane-bound enzyme complex in health and disease could enable the development of novel diagnostic procedures and advanced therapeutic strategies. In a recent study, we have found that the CyclinB1/Cdk1 complex translocates into mitochondria in the (Gap 2) G2/(Mitosis) M phase and phosphorylates CI subunits to enhance mitochondrial energy production, potentially to offset increased energy needs of cells during cell cycle11. Here we showcase experimental procedures and strategies that can be used to study mitochondrial translocation of otherwise nuclear/cytoplasmic kinases, their mitochondrial substrates as well as functional consequences of their mitochondrial localization using CyclinB1/Cdk1 as an example.
The finding that the CyclinB1/Cdk1 complex translocates into mitochondria when needed prompted the studies of mitochondria-specific overexpression and knockdown of this complex. To achieve mitochondria-specific expression of proteins, one can add a mitochondria targeting sequence (MTS) in the N-terminus of the protein of interest. Mitochondria targeting sequences allow the sorting of the mitochondrial proteins into the mitochondria where they normally reside12. We have used an 87 base mitochondria targeting sequence derived from the precursor of human cytochrome c oxidase subunit 8A (COX8) and cloned it into Green Fluorescent Protein (GFP)-tagged CyclinB1 or Red Fluorescent Protein (RFP)-tagged Cdk1 containing plasmids in frame. This method allowed us to target CyclinB1 and Cdk1 into the mitochondria, specifically changing the mitochondrial expression of these proteins without affecting their nuclear pool. By fluorescently tagging these proteins, we were able to monitor their localization in real time. Similarly, we have introduced MTS into a plasmid containing RFP-tagged dominant negative Cdk1, which allowed us to specifically knock down the mitochondrial expression and functions of Cdk1. It is essential to distinguish between mitochondrial and nuclear functions of the kinases that have dual localizations like Cdk1. Engineering MTS into the N-terminal of these dual functional kinases offers a great strategy that is easy to be employed and effective.
Since Cdk1 is a cell cycle kinase, it is fundamental to determine the cell cycle progression when Cdk1 is localized into mitochondria. To achieve this, we have utilized a new method to monitor DNA content in cells. Traditional methods include using BrdU (bromodeoxyuridine), a synthetic thymidine analogue, which incorporates into the newly synthesized DNA during the S phase of the cell cycle to substitute thymidine. Then the cells that are actively replicating their DNA can be detected using anti-BrdU antibodies. One disadvantage of this method is that it requires denaturation of DNA to provide access for the BrdU antibody by harsh methods like acid or heat treatment, which may result in inconsistency among results13,14. Alternatively, we utilized a similar approach to monitor the actively dividing cells with a different thymidine analog, EdU. EdU detection does not require harsh DNA denaturation as mild detergent treatment enables the detection reagent to access the EdU in newly synthesized DNA. The EdU method has proven to be more reliable, consistent and with potential for high-throughput analysis15.
Finally, to determine the mitochondrial substrates of Cdk1, we used a proteomics tool called 2D-DIGE, which is an advanced version of classical two-dimensional gel electrophoresis. Two dimensional electrophoresis separates proteins according to their isoelectric point in the first dimension and molecular weight in the second. Since post-translational modifications such as phosphorylation affect the isoelectric point and molecular weight of the proteins, 2D gels can detect the differences between phosphorylation statuses of proteins within different samples. The size (area and intensity) of protein spots changes with the expression level of proteins, allowing quantitative comparison between multiple samples. Using this method, we were able to differentiate the phosphorylated proteins in wild type versus mutant mitochondria-targeted Cdk1 expressing cells. The specific protein spots that showed in the wild type but were missing in the mitochondria-targeted mutant Cdk1 preparation were isolated and identified via mass spectrometry.
In traditional 2D gels, triphenylmethane dyes are used to visualize the proteins on the gel. 2D-DIGE uses fluorescent protein labels with minimal effect on protein electrophoretic mobility. Different protein samples can be labeled with different fluorescent dyes, mixed together and separated by the identical gels, allowing the co-electrophoresis of multiple samples on a single gel16. This minimizes the gel-to-gel variations, which is a critical problem in gel-based proteomics studies.