Aromatic amino acids play a harmonizing role in prostate cancer: A metabolomics-based cross-sectional study

Abstract Background Prostate cancer (PCa) is a common health problem worldwide. The rate of this disease is likely to grow by 2021. PCa is a heterogeneous disorder, and various biochemical factors contribute to the development of this disease. The metabolome is the complete set of metabolites in a cell or biological sample and represents the downstream end product of the omics. Hence, to model PCa by computational systems biology, a preliminary metabolomics-based study was used to compare the metabolome profile pattern between healthy and PCa men. Objective This study was carried out to highlight energy metabolism modification and assist the prognosis and treatment of disease with unique biomarkers. Materials and Methods In this cross-sectional research, 26 men diagnosed with stage-III PCa and 26 healthy men with normal PSA levels were enrolled. Urine was analyzed with proton nuclear magnetic resonance (1H-NMR) spectroscopy, accompanied by the MetaboAnalyst web-based platform tool for metabolomics data analysis. Partial least squares regression discriminant analysis was applied to clarify the separation between the two groups. Outliers were documented and metabolites determined, followed by identifying biochemical pathways. Results Our findings reveal that modifications in aromatic amino acid metabolism and some of their metabolites have a high potential for use as urinary PCa biomarkers. Tryptophan metabolism (p < 0.001), tyrosine metabolism (p < 0.001), phenylalanine, tyrosine and tryptophan biosynthesis (p < 0.001), phenylalanine metabolism (p = 0.01), ubiquinone and other terpenoid-quinone biosynthesis (p = 0.19), nitrogen metabolism (p = 0.21), and thiamine metabolism (p = 0.41) with Q2 (0.198) and R2 (0.583) were significantly altered. Conclusion The discriminated metabolites and their pathways play an essential role in PCa causes and harmony.


Introduction
Prostate cancer (PCa) is a significant health problem, reported as the second most frequently  (4). However, none of these methods are reliable (5), and the success rate of treatment significantly increases with the precise and early diagnosis of the disease (6).
Metabolomics is a fast-growing field in clinical research and provides new insight into various conditions and endpoints of omics cascade.
Nowadays, metabolomics-based studies with proton nuclear magnetic resonance ( 1 H-NMR) spectroscopy as analytical tools are the most applied methodologies in the metabolomics study of cancer (7,8).
It has been reported that PCa cells display significantly altered metabolism compared to healthy ones in aggressive and recurring conditions and are associated with gene regulation of cancer-associated pathways. In tumor cells, energy requirements are met by reprogramming metabolic pathways to survive and proliferate. It has been stated that high levels of glucose are essential in the metabolism of many cancer types.
Glucose is mainly consumed through the glycolysis pathway; however, fatty acid biosynthesis and amino acids such as glutamine and serine also generally increase in the cancer cell (9). Therefore, in the present case-control retrospective study, we compared the metabolome profile pattern between healthy and PCa men by 1 H-NMR spectroscopy. We determined the discriminated metabolites of both groups, highlighting the energy metabolism modification and facilitating data for more computational systems biology modeling approaches in the prognosis and treatment of this disease.

Sample collection and preparation
In this cross-sectional study, 26  Serums and urine samples were collected in a preservative-free and sterile container. Serum was separated and kept at -80°C until assayed for total and free PSA. Urines samples were also centrifuged at 4000 rpm for 15 min at 4°C, and the supernatants were stored at -80°C until analyzed.

PSA assay
Total and free PSA were assayed in duplicate by the Elisa colorimetric method. The plate was read at 450 nm with the reference wavelength at 630 nm in an Elisa reader.

Spectral preprocessing
The spectrum of all samples was preprocessed using a custom-written ProMetab (v.3.3) function file, a metabolomics data processing device that changes raw Bruker NMR spectra into a format for multivariate study in the MATLAB environment. The region at 0-10 ppm of the spectra is segmented into bins of 0.005. The water peak at 4.7 ppm chemical shift was removed, and the resulting spectrum was normalized, followed by scaling with the Pareto scaling approach, employing the square root of the standard deviation as the scaling factor.

Ethical considerations
Written consent was obtained from all participants, and the design of this study was approved by Payeme-Noor University Ethics Committee (Code: IR.PNU.REC.1398.150). Table I  Moreover, the topology map of altered biochemical pathways in the test and control groups is shown in Figure 3. Degree centrality was defined as the number of links that occurred upon a node. In our investigation, three major biochemical pathways with a p-value < 0.05 were used in our discussion, while other pathways were kept apart (Table II).

Results
Cross-validation (10fold-CV) was performed to validate our results. The Q2 value, which indicates the validity of the PLS-DA discrimination, was 0.198, whereas R2 was 0.58322 with an accuracy of 0.7954 in component 3. Experimental and healthy samples were obtained from men aged 40-60 according to the above data. PSA: Prostate-specific antigen, compression of total and free PSA ratio in both groups also described

Discussion
The prostate displays an exceptional metabolism that changes during primary neoplasia to destructive PCa (11). Although several biochemical studies have been carried out on PCa, very few systematic reviews have been performed to discover the metabolite pattern alterations in the urine of these men (12). So, in need of raw systematic metabolome data for PCa modeling, we compared the urinary metabolome profile of PCa men with that of healthy men for further modeling of PCa by systems biology methods.
Urine is considered the primary source of released PCa biomarkers (13). Change in sarcosine is reported as one of the essential factors for identifying PCa (14). It has been reported that the urinary excretion of isoleucine in PCa men is considered one of the most altered parameters in PCa (15 It has been stated that serotonin, which is produced from tryptophan in prostate cells, had a critical role in the growth, inhibition, and induction of apoptosis in tumor prostate cells (17). Serotonin is also linked to some metabolites such as N-Acetyl serotonin, melatonin, and 5hydroxyindoleacetate as a precursor. Therefore, any alteration in serotonin can affect these metabolites. In our study, melatonin was one of the altered metabolites in tryptophan metabolism.
According to the results of Hevia and colleagues, melatonin can be detected easily in PCa cells. They also showed a significant relationship between cellular melatonin uptake and anti−proliferative melatonin activity in some prostatic cancer cells (18).
Similarly, in other studies, it has been demonstrated that that melatonin can inhibit the growth of PCa cells by affecting the androgen-sensitive and insensitive epithelium of PCa cells (19). Also, glucose uptake in glycolysis, the Krebs cycle, and the pentose phosphate pathway is limited by melatonin in PCa cells (18). It has been exhibited that some kynurenine pathway metabolites such as L-tryptophan, kynurenine, anthranilate, and indolelactate were altered in the serum of the PCa men they studied (16). Our findings are in line with a recent study, which indicated that the urinary concentration of indoleacetamide and indoleacetate in tryptophan metabolism was altered in PCa males (Figure 4) reported that it is a potent factor for treating human tumors (27). Moreover, it has been shown that an increase in plasma concentrations of phenylacetic acid in end-stage renal failure leading to the inhibition of iNOS expression (28).
Phenylethylamine is a neurotransmitter, and it is biosynthesized from tryptophan by enzymatic decarboxylation (29). There is an increase in the urinary concentration of phenylethylamine in people with autism and minimal brain dysfunction (30).

2-hydroxyphenylacetate
is present in all biofluids and even in the prostate, kidney, and bladder tissues and is associated with colorectal and lung cancer (31).
The concentration of 2-hydroxyphenylacetic acid was altered in our study.
2-hydroxyphenylacetic acid is a substrate of the enzyme oxidoreductases. Recently, the potential impact of the combined inhibition of 3a-oxidoreductases and 5a-reductases on PCa it has been reported (32). Our findings also reveal that the urinary level of phenylalanine, L-tyrosine, phenylacetic acid, and 2-hydroxyphenylacetate metabolites were altered in the phenylalanine metabolic pathway.

Conclusion
We conclude that the discriminated metabolites in aromatic amino acid metabolism play an essential role in causing PCa and its metabolic harmony. Further systematic approaches are required to complete and validate this notion in computational systems biological modeling of PCa, to help work towards offering unique biomarkers for clinical practice.