Marsdenia tenacissima extract alters crucial metabolites in cancer , determined by 1 H NMR based metabolomics approach

Altered metabolites level in the biosystems, is the potential cause of cancer, the primary reason of alteration of metabolism is change in nutrient consumption and waste excretion, as a result genetic mutation leads to cancer initiation and progression. Aberration of specific metabolites such as fumarate, succinate, 2-hydroxyglutarate may alter cell signaling. We collected liver and kidney samples and prepared for 1H NMR analysis, then executed NMR spectroscopy. We used a set of domestic R scripts to perform an unsupervised principal component analysis (PCA) and a supervised orthogonal signal correction partial least-squares discriminant analysis (OSC-PLS-DA). It signifies class discrimination for getting a clear separation, whereas PCA scores plot signifies the model group kept further away from the control group than drug group on the horizontal axis. In another PCA scores plots, most parts of the control group was overlapping with the drug group but was distant from the model group. Marsdenia tenacissima extract (MTE) (Chines name: Xiao-Ai-Ping, XAP) modulates level of crucial metabolites such as fumarate, lactate, succinate, determined by 1H NMR spectroscopy and their altered level contributes major role in cancer.


INTRODUCTION
MTE is an anti-cancer, traditional Chinese herbal medicine.It has been widely used in various diseases and also in cancer, for many decades.But previously, no metabolomics study is carried out on MTE.In this research study, we explored the anticancer action of MTE in order to metabolomics approach, that how the level of some metabolites directly or indirectly involved with progression and suppression of cancer, modulated by MTE.
Metabolomics is a kind of newly emerging subject in recent decades which focuses on the changes of all the metabolites produced by external stimulus in the biosystem.This process is used to track the biomarkers according to the changes of the metabolic fingerprint and illuminate the overall effect (Roy et al., 2017), now metabolomics methods have been widely employed in recent years (Griffin, 2004;Lou et al., 2015).We unveiled in this research study, that some crucial metabolites play a major role in cancer.Mainly, the overexpression of succinate (Selak et al., 2005) and lactate (Hirschhaeuser, Sattler, Mueller-Klieser, 2011) links with cancer formation.The aim of this research was to investigate the curative effect of MTE on MDA-MB-231 in nude mice at the dosage of 2.5g/kg, using an NMR based metabolomics approach complemented with the histological inspection.

Sample collection and preparation for 1 H NMR analysis
After three weeks of the treatment, all mice fasted overnight.The next day after the final administration of MTE and after anesthetization by chloral hydrate (350 mg kg -1 , i.p.) all of them were sacrificed.Their livers and kidneys were harvested and frozen in the liquid nitrogen and stored at -80 ℃ immediately.Before 1 H NMR analysis, frozen mice liver and kidney tissues were weighted (500 mg) and homogenized in 50% acetonitrile/H 2 O (3 mL) and centrifuged at 12000 rpm for 10 min at 4 °C.We collected the supernatant.And then, the supernatant was lyophilized and reconstituted in 550 μL D 2 O in phosphatebuffered solution (0.2 M Na 2 HPO 4 -NaH 2 PO 4 , pH 7.4).We used TSP (0.05 wt%) as an internal standard for D 2 O.The supernatant was transferred into a 5 mm NMR tube for 1 H NMR analysis after the solution was vortexed and centrifuged at 12000 rpm for 10 min to remove any debris.D 2 O was used for field frequency locking and TSP was used as a chemical shift reference ( 1 H, 0.00 ppm).

H NMR spectroscopy
All the 1 H NMR spectra of liver and kidney samples were acquired at 298 K on a Bruker Avance 500 MHz flow-injection spectrometer (Bruker GmbH, Karlsruhe, Germany) with a Bruker 5 mm probe, using a modified transverse relaxationedited Call-Purcell-Meiboom-Gill (CPMG) sequence (90(τ-180-τ) n-acquisition) with a total spin-echo delay (2nτ) of 10 ms to suppress the signals of proteins, with 128 transients collected into 32 768 (32 K) data points, a spectral width of 10 000 Hz, an acquisition time of 3.27 s.The spectra were Fourier transformed after multiplying the FIDs by an exponential weighting function corresponding to a line-broadening of 0.5 Hz.

Data pre-processing and peak assignments
The 1 H NMR spectra were phased and baselinecorrected manually using Bruker Topspin 3.0 software (Bruker GmbH, Karlsruhe, Germany).The spectra were converted to ASCII-format files by using MestReNova (Version 8.0.1, Mestrelab Research SL) and then imported into "R" (http://cran.r-project.org/) for multivariate data analysis using internally developed R scripts.The spectra were split into 0.015 ppm average integrated spectral regions (buckets) between 0.2 and 10 ppm.Then, the data were probability quotient-normalized to account for variation in sample dilutions.We assigned the metabolites in the NMR spectra according to the previous literature and queried performed in open access databases, such as the Madison metabolomics consortium database (http:// mmcd.nmrfam.wisc.edu/)and the human metabolome database (http://www.hmdb.ca/).We also used Chenomx NMR suite 7.7 software and the statistical total correlation spectroscopy (STOCSY) technique to assign the metabolites.

Data analysis
We used a set of domestic R scripts to perform an unsupervised PCA and a supervised OSC-PLS-DA.PCA was performed to show a general overview of the metabolic pattern through using the NMR data and then, OSC-PLS-DA was performed to show differences in metabolic alterations across the liver and kidney tissues.Each OSC-PLS-DA model was confirmed by repeated two-fold cross-validation, as follows: the validity of the model against overfitting was assessed using the parameter R 2 Y, and the predictive ability was described by Q 2 .
Color-coded loading and S-plots were constructed to illustrate the variables that contributed to the grouping.The fold changes and the associated p-values adjusted by BH (Benjamini-Hochberg) methods of different metabolites in liver and kidney samples were calculated and listed in Table I (for liver samples) and Table II (for kidney samples).

Multivariate analysis of 1 H NMR spectra
We used the supervised OSC-PLS-DA to remove variations that were unrelated to class discrimination for getting a clear separation.In the PCA scores plots of the liver (Figure 1A), the model group kept further away from the control group than drug group on the horizontal axis.In another PCA scores plots (Figure 1B for kidney), most parts of the control group were overlapping with the drug group but were distant from the model group.
The loading plot (Figure 1B, C, E, F) was constructed by a covariance-based pseudo-spectrum and it was used to show color-coded according to the absolute values of the correlation coefficients (r 2 ).The weight of a variable in the discrimination model was determined by calculating the square of its r 2 and ranged from zero (blue areas) to high values (red areas).Changes in metabolites were directly visualized as fold-changes in these plots (Figure 2).They are color-coded according to the p-values of differences among the groups.
In this study, we also analyzed the antitumor activity of MTE with 1 H NMR-based metabolomics approach.OPLS-DA analysis of NMR data from livers and kidneys revealed that MTE induced severe metabolic perturbations (Figure 3).Both of the livers and kidneys' model groups were found with the significant higher level of succinate.It can be catalyzed into fumarate by succinate dehydrogenase (SDH).SDH is also a tumor suppressor.The succinate accumulated in the mitochondria due to SDH inhibition is transported to the cytosol.Elevated cytosolic succinate inhibits HIF-prolyl hydroxylase (PHD) and thereby Hypoxia-inducible factor-α (HIF-α) hydroxylation.HIF-α makes up HIF combined with HIF-β.In many tumors, oxygen availability becomes limited (hypoxia) very quickly during cancer development.Intermittent hypoxia may paradoxically be the only reason for increased tumor glycolysis (Zu, Guppy, 2004).The major regulator of the response to hypoxia is the HIF transcription factor, forcefully, emphasizing the importance of HIF in tumor development or sustention.In addition, it was demonstrated that HIF activation resulting from von Hippel-Lindau (VHL) mutations promotes metastasis by inducing the expression of met and CXR4 (Selak et al., 2005).The VHL gene product (pVHL) is part of an E3 ubiquitin ligase complex that binds to the oxygen-dependent degradation (ODD) domain of HIF-α in an oxygen-dependent manner and targets it for degradation (Pugh, Ratcliffe, 2003;Semenza, 2002).Consequently, pVHL binding to HIF-α is decreased, and elevated HIF activity induces expression of genes that facilitate angiogenesis, metastasis, and metabolism, leading to  (Selak et al., 2005).To sum up, succinate may function as an intracellular messenger between mitochondria and the cytosol and has a profound effect on cytosolic enzymes (PHD) and consequently on nuclear events (gene expression by HIF) (Figure 4).Significant alteration level of metabolites with their related expression and relevant action are summarized in the Table III.All the above indicates that the float of      -Halfwerk et al., 2000;Tessem et al., 2008), succinate (Yang, Pollard, 2013), glutamate (Fazzari et al., 2015;Willard, Koochekpour, 2013) are related to the cancer progression.Conversely, lactose (Goodman et al., 2002)  Here we used xenograft mice model for this study because all mammalian cells have similar biological pathways to regulate growth, differentiation, replication, and death.It has been also observed that Mice and Human have similar metabolic homogeneity and physiology disease pathogenesis.Additionally, mice acquire mutations by the equivalent spectrum of tumor suppressor genes and proto-oncogenes as Human (Balmain, Harris, 2000) (Demetrius, 2005).So, this research outcome also will be effective on Human as well.

CONCLUSION
It has been revealed through 1 H NMR-based metabolomics approach that MTE modulates the alteration of crucial metabolites, mainly succinate and lactate, which contribute major role in cancer formation.Consequently, pVHL binding to HIF-α is decreased, and elevated HIF activity induces expression of genes that facilitate angiogenesis, metastasis, and metabolism, leading to more aggressive tumors (Table I).

FIGURE 1 -
FIGURE 1 -OPLS-DA analysis of liver and kidney extract 1 H NMR data of control, model, and MTE groups.Score plots (A for liver, D for kidney) and the corresponding loading plots of OPLS-DA (B and C for liver, E, and F for kidney) color-coded with the absolute value of correlation coefficients.The color bar corresponds to the weight of the corresponding variable in the discrimination of statistically significant (red) or non-significant (blue).Positive and negative peaks indicate a relatively decreased and increased metabolite level.

FIGURE 3 -
FIGURE 3 -Fold change plots color-coded with p-values adjusted by Benjamini Hochberg method indicating the significance of altered metabolites in the liver of (A), XAP vs. control (B), model vs. control (C), and lung of XAP vs. control (D), model vs. control.The blue and red dashed lines represented variations of 20% and 100%, respectively.

FIGURE 4
FIGURE 4 -A schematic model that summarizes the role of succinate in the mitochondrion-to-cytosol signaling pathway.Succinate accumulated in the mitochondria due to SDH inhibition is transported to the cytosol.Elevated cytosolic succinate inhibits PHD and thereby HIF-α hydroxylation.Consequently, pVHL binding to HIF-α is decreased, and elevated HIF activity induces expression of genes that facilitate angiogenesis, metastasis, and metabolism, leading to more aggressive tumors (TableI).

TABLE I -
Identified metabolites in the livers from different groups with log 2 (FC) and p-values a Multiplicity: singlet (s), doublet (d), triplet (t), quartets (q), multiplets (m).b Color coded according to the log2(FC), red represents increased and blue represents decreased concentrations in glyphosate exposed groups.Color bar c p-values corrected by BH (Benjamini Hochberg) methods were calculated based on a parametric Student's t-test or a nonparametric Mann-Whitney test (dependent on the conformity to normal distribution).*p < 0.05, **p < 0.01, ***p < 0.001

TABLE II -
Identified metabolites in the kidney from different groups with log 2 (FC) and p-values a Multiplicity: singlet (s), doublet (d), triplet (t), quartets (q), multiplets (m).b Color coded according to the log2(FC), red represents increased and blue represents decreased concentrations in glyphosate exposed groups.Color bar c p-values corrected by BH (Benjamini Hochberg) methods were calculated based on a parametric Student's t-test or a nonparametric Mann-Whitney test (dependent on the conformity to normal distribution).*p < 0.05, **p < 0.01, ***p < 0.001

TABLE III -
Different level of metabolites: In liver the level of the metabolites, alanine, succinate, glutamate, lactose and fumarate were aberrated significantly.Alanine (Leij (Bonuccelli et al.,2010))al role in the inhibition of cancer progression.However, in kidney, the level of succinate and lactate(Bonuccelli et al., 2010)were notably reduced, besides this the level of 3-Hydroxybutyrate(Bonuccelli et al.,2010)was augmented remarkably, which takes part a major role in tumor progression After a consecutive 3 weeks oral MTE administration, the level of succinate was significantly lowered and back to normal.In another way, MTE was proved to exhibit antitumor action in vivo.
Depletion of the level of fumarate links to accumulation of succinate, leads to activation of HIF-1α.…..succinate can reflect the growth of cancer cells in vivo.