Inhibition of BUB3 shunts glucose to glycolytic pathway by inducing PFKFB3 accumulation

Purpose: Metabolic reprogramming as a hallmark of cancer has countless connections with other biological behavior of tumor such as rapid mitosis. Mitotic checkpoint protein BUB3 as a key protein involved in the regulation of mitosis is modulated by PKM2, an important glycolytic enzyme. However the role of BUB3 in glucose metabolism remains unknown. Methods: We analyzed the TCGA data to evaluate BUB3 expression in certain tumors. The uptake of glucose and CO2 incorporation was tested by isotopic tracer methods. The lactate, NADPH, NADP and metabolic enzyme activities were tested by assay kits accordingly. Results: We show here that BUB3 is over expressed in cervical cancer and hepatocellular carcinoma. Interference of BUB3 increase the uptake of glucose and shunts the metabolic ux from pentose phosphate pathway to glycolytic pathway. The glycolysis metabolites lactate is increased by BUB3 interference whereas NADPH/NADP ratio is reduced. With regard to metabolic enzymes, interference of BUB3 increase PFKFB3 on protein level and enzyme activity, but not mRNA level. Moreover, the increasing of protein level is diminished when proteasome degradation pathway is blocked by MG132. Conclusions: BUB3 is a potential tumor promoter and plays certain roles in cancer cellular metabolic reprogramming. Taken together, this study demonstrated elevated BUB3 expression accompanied by tumor generation and poor prognosis. Inhibition of BUB3 shunts glucose from the pentose phosphate pathway to glycolytic pathway by inducing PFKFB3 accumulation. However, there is still be limitation in this study. First, the effect of BUB3 on glucose metabolism was studied only in Hela cells in vitro. More types of cancer cell lines and in vivo experiments are still been needed to verify this conclusion. Moreover, the exact mechanisms of how BUB3 regulates the PFKFB3 and HK2 protein level still need to be further interpreted.


Background
Reprogramming of cellular metabolism has been reckoned as one of the hallmarks of cancer [1]. Cancer cells often use glucose as a carbon source for glycolysis, even in oxygenated environments, which is known as the Warburg effect. By this type of glucose metabolism, cancer cells produced sufficient nucleotides, proteins, and lipids, which are necessary for the rapid growth and division [2]. And one of the key enzymes in this highly glycolytic metabolic type is 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 (PFKFB3).
PFKFB3 has been found to be up-regulated in numerous cancers [3]. It converts fructose-6-phosphate to fructose-2,6-bisP (F2,6BP). F2,6BP is an allosteric activator of 6-phosphofructokinase-1, stimulating glycolysis. It was more recently discovered that PFKFB3 degradation through the proteasome pathway, thus shunting glucose usage from glycolysis to the pentose phosphate pathway for NADPH generation [4]. These findings established a significant role of PFKFB3 in cancer glucose metabolism as a switcher between glycolysis and pentose phosphate pathway.
BUB3 mitotic checkpoint protein, also known as BUB3 or hBUB3, is involved with the regulation of the Spindle Assembly Checkpoint (SAC). BUB3 delays the start of anaphase by directing the localization of kinetochore during prometaphase [5]. SAC is activated during the process of cell division, preventing separation of duplicated chromosomes until each chromosome is properly attached to the spindle apparatus. When SAC is not active, BUB3 facilitates binding of APC/C and Cdc20 for substrate ubiquitination [6]. Tumor-specific pyruvate kinase M2 (PKM2), a key enzyme in aerobic glycolysis, binds to Bub3 during mitosis and phosphorylates Bub3 at Y207.
Thus facilitates BUB3 recruitment to kinetochores [7]. However, whether BUB3 plays a role in glucose metabolism remains unknown.
In the present study, we show that interference of BUB3 by siRNA shunts glucose from (pentose phosphate pathway) PPP to glycolytic pathway in cancer cells. It is possibly because inhibition of BUB3 stabilize PFKFB3 in protein level. Thus the relationship between BUB3 and PFKFB3, as well as the influence of BUB3 on glucose metabolism in cancer cells were revealed for the first time.

Cell culture
The human cervical cancer cell line Hela and hepatocellular carcinoma HepG2 (Cell Bank of the Chinese Academy of Sciences, Shanghai, China) were maintained in DMEM medium. The media were supplemented with 10% FBS and 100 units/mL penicillin/streptomycin. Cell cultures were maintained in 5% CO 2 and air in a humidified 37°C incubator.

TCGA data analysis
The Cancer Genome Atlas (TCGA) online gene profile data for cervical cancer and hepatocellular carcinoma (http://cancergenome.nih.gov/) was downloaded. The relative mRNA expression levels of BUB3 were analyzed from a dataset of 304 cases of cervical cancer and 3 cases of normal cervical tissue, as well as a dataset of 371 cases of hepatocellular carcinoma and 50 cases of normal liver tissue.

Transfection of siRNA
Cells were transfected with oligo siRNAs using Lipofectamine 2000. The sequences of siRNA oligos used in this study are as follows: Non-specific, UUCUCCGAACGUGUCACGUTT, ACGUGACACGUUCGGAGAATT; BUB3, CAAGCAGGGUUAUGUAUUATT, UAAUACAUAACCCUGCUUGTT.

Measurement of cell viability
Cells were seeded in 96-well plates at 5x10 3 cells/well. After the indicated time, cell viability was determined using the CCK-8 assay (Dojindo) according to the manufacturer's instructions.

mRNA expression analysis
Total RNA was isolated using a Trizol kit (Omega, Norcross, GA, USA) and transcribed to cDNA with a cDNA synthesis kit (Takara, Otsu, Japan). Quantitative real-time PCR was performed using SYBR Green PCR Master Mix (Takara) and the transcript levels of genes were detected by using the StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Primers used for detection of specific genes are listed below.

Immunoblotting
Cells were lysed into the RIPA buffer containing protease inhibitors by incubating on ice for 30 min. Followed by centrifugation at 10 000 g for 15 min. The extracted proteins were subjected to electrophoresis on SDSpolyacrylamide gels and transferred to PVDF membranes (GE Healthcare, Buckinghamshire, UK), which were blocked and probed with specific primary antibodies with appropriate dilution at 4°C overnight. The membranes were then incubated with the horseradish peroxidase-conjugated secondary antibodies for 1h at room temperature, followed by three washes with 1 × TBST. The immunoreactive bands were visualized by ECL Plus system (Tanon, Shanghai, China).

[18F]-Fluorodeoxyglucose ([18F]-FDG) incorporation
FDG uptake assays were performed essentially as described The incorporation of [1-14C] or [6-14C] glucose into 14 CO 2 was determined as previously reported [9]. Briefly, cells were cultured in 10cm dishes, and the cells were exposed to DMEM supplemented with [1-14 C] glucose (0.1 µCi/ml) or [6-14 C] glucose(0.1 µCi/ml). The dish was placed in a container to collect CO 2 produced. Rates of glucose consumption were measured by incubating cells for 120 min at 37°C. Fresh air was pumped into the container by a ventilator. The 14 CO 2 was driven into a vial and trapped by Hyamine hydroxide. No-cell controls were included to correct for unspecific CO 2 trapping.

Mass spectrometry
The metabolites extraction was carried out as previously reported [4]. Glucose metabolites in the cell lysate were tested by liquid chromatography-mass spectrometer (LC-MS) (Agilent, CA, USA). The quantity of metabolites was normalized to the total protein levels.

Metabolic enzymes assays
Enzymes of glucose metabolism were assayed using an assay kit (Comin Biotechnology Co. Ltd, Suzhou, China) according to the manufacturer's recommended protocol.

Statistical analysis
Statistical analysis was performed using GraphPad Prism software. Student's t-test were used to analyze the statistical significance, in which p< 0.05 or less were considered as significant differences.

BUB3 is overexpressed in cervical cancer.
It has been proposed that BUB3 increase colorectal cancer risk by high-throughput sequencing analysis [10].
However the clinical significance of BUB3 on other cancer types remains unclear. We extracted TCGA (The Cancer Genome Atlas) online gene profile data and performed a comprehensive evaluation. Our analysis revealed that BUB3 mRNA was overexpressed significantly in primary solid tumors (cervical cancer and hepatocellular carcinoma) compared with normal solid tissue ( figure 1A and 1B). Interference of BUB3 with siRNA in Hela cells showed relatively slower proliferation rate compared with non-specific control (figure1c). To test if BUB3 have regulating effect on glucose metabolism in cancer cells. We interference the expression of BUB3 with siRNA. Interference efficiency was evaluated with western blot. Firstly, we tested the effect of interference of BUB3 on glucose uptake with 18F-FDG ( Figure 3A). 18F-FDG is a glucose analog, with the positron-emitting radionuclide fluorine-18 substituted for the normal hydroxyl group at the C-2 position in the glucose molecule. The uptake of 18F-FDG by tissues is a marker for the cell uptake of glucose, which in turn is closely correlated with certain types of cell metabolism. As is shown in figure 2A, the uptake of 18F-FDG of Hela cells was elevated by the interference of BUB3, which indicate inhibition of BUB3 could promote glucose uptake in certain cells.
In order to determine the actual conversion of glucose to the downstream pathways, [1-4C] (Figure 2A), but the [1-14C] glucose incorporated into 14 CO 2 was decreased ( Figure 2B). This result suggested that interference of BUB3 shifts glucose metabolic flux form PPP to the TCA cycle.
To further confirm that the pentose phosphate pathway was inhibited by siBUB3, cellular NADPH and NADP were determined. As is shown in figure 2C, the NADPH/NADP ratio was decreased when BUB3 was inhibited.
Since the 14 CO 2 decomposed from [6-14C] glucose in the TCA cycle has to through glycolytic pathway first. We asked if the glycolysis was up-regulated by BUB3 interference. As expected, lactate production was increased when BUB3 was inhibited ( Figure 2D). These results suggest that BUB3 inhibition could promote both glycolysis and TCA cycle, but inhibit PPP. Or, in other words, interference of BUB3 shunts glucose to glycolytic pathway from the pentose phosphate pathway.
Differences in glucose biotransformation between Hela cells interference with BUB3 siRNA and non-specific control were examined by metabolome analysis based on liquid chromatography mass spectrometry (LC-MS). It was showing that the sum of metabolites belonging to glycolysis was increased, while those belonging to PPP were decreased significantly by BUB3 interference (Figure 2B-2G).

Interference of BUB3 up-regulate PFKFB3 enzyme activities.
To study how glycolytic pathway is regulated by BUB3, firstly we determined if the glycolytic enzymes are affected by interference of BUB3. As is well known, hexokinase, 6-phosphofructokinase, pyruvate kinase and lactate dehydrogenase have important roles in glycolysis. We determined the activity of these key enzymes with enzyme activity assay kits accordingly. As shown in figure 4A and 4B, when the expression of BUB3 is inhibited, the activity of phosphofructokinase and hexokinase were increased, whereas the activity of pyruvate kinase and lactate dehydrogenase remained unchanged ( Figure 4C and 4D). This result suggested that phosphofructokinase and hexokinase might be regulated by BUB3.
3.4 Interference of BUB3 induce PFKFB3 accumulation in protein level by inhibiting proteasome degeneration pathway.
To further confirm that HK2 and PFKFB3 are regulated by BUB3. We determined the protein levels of these enzymes in cells by western blot. In cancer cells, HK2, PFKFB3, PKM2 and LDHA are reckoned as pivotal enzymes in glycolysis. As is shown in figure 5A, the protein levels of HK2 and PFKFB3 were up-regulated by interference of BUB3, whereas PKM2 and LDHA did not have significant change in protein levels. To test whether the transcription of these enzyme genes were changed by interference of BUB3, quantitative PCR (qPCR) was applied to determine the mRNA levels of these enzymes. As is shown in figure 5B, all of the four enzymes did not have any change in mRNA levels when BUB3 was inhibited. These results suggest that HK2 and PFKFB3 were regulated by BUB3 in protein levels, but not mRNA levels.
It was reported that PFKFB3 protein is degenerated by APC/C regulated proteasome pathway [11].
Furthermore, BUB3 was reported that could promote activation of APC/C [12]. Thus, we asked that whether the proteasome pathway played any role in the accumulation of HK2 and PFKFB3 protein induced by interference of BUB3. As is shown in figure 5C, when the proteasome is blocked by MG132, a classic proteasome inhibitor, the accumulation effect induced by BUB3 inhibition was diminished. This result suggests that BUB3 inhibition induced PFKFB3 and HK2 protein accumulation are depending on proteasome pathway.
3.5 Shifts between glycolysis and PPP modulated by BUB3 are depended on PFKFB3.
PFKFB3 is a key switcher between glycolysis and PPP metabolic pathway [4]. In our previous study, the PFKFB3binding protein was purified by immunoprecipitation and analyzed by LC-MS/MS [13]. BUB3 was revealed as potential binding partner of PFKFB3 by mass spectrometry. Next, we applied in vivo experiments to validate the protein-protein interaction between PFKFB3 and BUB3. Anti-BUB3 antibody was capable of immunoprecipitating PFKFB3 protein from Hela cell extracts ( Figure 5A).
In our previous study, we knock-out PFKFB3 gene in Hela cells with CRISPR/Cas9 [13].To verify whether BUB3 interference induced glucose biotransformation is depending on PFKFB3, [1-14C] and [6-14C] glucose oxidation were estimated in Hela cells with PFKFB3 knockout. As showed in Figure 5B and 5C, interference of BUB3 was not capable of shifting the metabolic flux between glycolysis and PPP in PFKFB3 knock-out Hela cells. This result indicates that the regulating effect of BUB3 on glucose biotransformation is, at least partially, rely on PFKFB3 expression.

Discussion
BUB3 is a core component of the key mitotic surveillance mechanism spindle assembly checkpoint (SAC). It prevents anaphase onset until all chromosomes have attached to microtubules successfully and properly. It has been proposed that BUB3 increase colorectal cancer risk by high-throughput sequencing analysis [10]. However the clinical significance of BUB3 on other cancer types remains unclear. Here, we extracted TCGA (The Cancer Genome Atlas) online gene profile data and performed a deep analysis. We found that compared with normal tissue, BUB3 mRNA levels were significantly increased in cervical cancer and hepatocellular carcinoma. Thus the presence of BUB does not procrastinate cell from devision, but ensures it goes properly.
It was reported that BUB3 is regulated by PKM2, a glycolysis enzyme which is instrumental in both aerobic glycolysis and gene transcription [7]. However, whether BUB3 plays a role in glucose metabolism remains unknown. In this study, we shown that glucose uptake was increased by interference of BUB3. BUB3 interference also accompanied with a shifting of metabolic flux from PPP to glycolysis. And in the metabolite level, NADPH/NADP ratio was diminished whereas lactate was increased after BUB3 interference. These results indicated that inhibition of BUB3 shunts glucose from PPP to glycolytic pathway.
6-Phosphofructo-2-kinase/fructose-2,6-bisphosphatase-3 (PFKFB3) is a master regulator of glycolysis by its ability to synthesize fructose-2,6-bisphosphate, a potent allosteric activator of 6-phosphofructo-1-kinase. It was reported that PFKFB3 is a substrate of APC/C. APC/C plays a crucial role in brain metabolism by promoting PFKFB3 ubiquntination and thus protein degeneration [14]. In this study, we find that interference of BUB3 in Hela cells showed an increase in PFKFB3 and HK2 protein level, as well as their enzyme activities, but not mRNA level. When blocking the proteasome degradation pathway by MG132, the accumulation effect of BUB3