Page 47 - Read Online

Tu et al. J Cancer Metastasis Treat 2018;4:58 I http://dx.doi.org/10.20517/2394-4722.2018.67 Page 13 of 16

transduction pathways. Studies have shown that up-regulated glycolysis can be influenced through diverse
PTMs including phosphorylation, acetylation, glycosylation and oxidation of glycolytic enzymes as well as
[61]
other signaling mediators (reviewed ). It is not unlikely that the observed elevation of glycolytic activity in
the metastatic cells was the result of PTMs and gene expression were lowered as compensating mechanism.
Further studies would need to be carried out to investigate whether proteomic changes are correlated with
transcriptomic observations.

The XF mitochondrial stress test revealed that the elevated oxidative phosphorylation observed in the meta-
static cells is independent of leaky mitochondria and is mainly explained by the enhanced production of
ATP. The result further suggests a higher energy demand in the metastatic MDA-MB-231HM.LNm5 line
compared to the parental line. Additionally, we found increased expression of all five complexes of the mi-
tochondrial electron transport chain, which are the mediators of oxidative phosphorylation. Although this
elevation was modest in magnitude, it may be sufficient to shift the entire metabolic profile of the cells.

In addition to the XF analyzer, metabolic status could also be measured by a variety of assays such as direct
measurements of various metabolic enzymes, substrates, or ATP as surrogates of total energy metabolism.
Although these metabolic assays each have their limitations and are mostly single-point measurements, it
would have added valuable verification of our XF observation.

IDH2 expression was significantly reduced in MDA-MB-231HM.LNm5 cells while IDH1 levels remain un-
changed. Interest in this family of enzymes in relation to cancer biology arose from reports of recurring mu-
[62]
tations in IDH1 and IDH2 genes in several cancers including colorectal cancer and gliomas . The function-
ality of these mutants and their impact on cancer progression has been the focus of many studies. Currently,
inhibitors of mutant IDH1 and IDH2 are in Phase I/II clinical trials for both solid and myeloid tumors. In
[63]
breast cancer, IDH gene mutations are detected at a frequency of less than 5% . Compared to the substan-
tial focus on mutant forms of IDH, little is known about the role of wild-type IDH1, and even less of wild-
type IDH2, in cancer progression and metastasis. Hepatocellular carcinoma patients with reduced levels of
[64]
IDH2 in tumors were at increased risk of metastatic progression and showed worse prognosis . Similarly,
[65]
in osteosarcoma, IDH2 levels were inversely correlated with pathological grade and metastasis . The sug-
gestion from these correlative observations, that wild-type IDH2 suppresses metastatic processes, is further
supported by our data. In addition, our findings suggest that the mechanism by which IDH2 may inhibit
metastasis is independent of cellular energy pathways.

Our transcriptomic findings warrant further studies that directly investigate the role of the abovementioned
DEGs in metastatic behaviors of breast cancer cells. Knockdown and/or ectopic overexpression of genes of
interest found in our study, such as BCS1L or IDH2, in the metastatic MDA-MB-231HN.LNm5 and/or non-
metastatic MDA-MB-231 cells may reveal the relationship between these genes and metastatic phenotypes
including metabolic reprogramming. Moreover, related animal experiments involving the manipulation
of the expression of these genes of interest would further characterize their contribution in breast cancer
growth and progression.

We acknowledge the limitation of having carried out the metabolic and transcriptomic studies in cultured
cells. The clinical relevance of human cell line models has been questioned. Indeed, there is not always a
linear correlation between in vitro proliferation or motility and spontaneous metastatic capacity in vivo, as
other cellular phenotypes, influencing intravasation, extravasation and survival in the circulation (among
others) also play a role. However, to determine precise ECAR and OCR measurement in vivo would be tech-
nically challenging. Future studies involving metabolic and transcriptomics analysis of tumour cells isolated
in situ are required.