Serum Colorectal Cancer Biomarkers Unraveled by NMR Metabolomics: Past, Present, and Future

Systematic Review and Meta-Analysis: Taurine and Its Association With Colorectal Carcinoma

BACKGROUND

Colorectal cancer (CRC) is one of the most common cancers. Various options are available for treatment, but prognosis is still poor in the more advanced stages. Current screening methods are not as accurate for distinguishing between benign and malignant growths, resulting in unnecessary invasive procedures. Recently a focus has been placed on identifying metabolites. Of these, taurine has frequently been detected, and this particular compound has a multifactorial role in human physiology.

METHODS

We conducted a systematic review of studies up till November 2023. Searches were done in three databases- MEDLINE, CINAHL-Ebsco, and PubMed. Three independent reviewers filter titles, abstracts, and full-texts according to selection criteria. Ten studies (samples = 1714) were identified showing a differential level of taurine in CRC patient samples. Quality assessment accounted for the risk of bias of each study using the 'robvis' tool. Where meaningful comparisons could be made, meta-analyses were carried out using the 'R' program for precalculated effect sizes with 'metagen' in R. The 'meta' package was utilised for creation of forest plots.

FINDINGS

Taurine was shown to significantly increase odds of CRC. It was also significantly associated with being a discriminator for CRC as a diagnostic metabolite. This was maintained at various stages of CRC. Taurine had increased expression in CRC patients, especially when the matrix utilised was blood. Nevertheless, there was significant heterogeneity for some outcomes.

INTERPRETATION

In conclusion, these findings highlight the potential of using taurine as well as other bile acid metabolites (lithocholic and ursodeoxycholic acid) to diagnose CRC and illustrate the link with microbiome interactions. Overall increased taurine concentration are associated with significantly increased odds for CRC. There was mostly an increase in relative expression of taurine in CRC samples, excluding results from Wang et al.

LC-MS analysis of chiral amino acids in human urine reveals D-amino acids as potential biomarkers for colorectal cancer

Colorectal cancer (CRC) is a common malignant tumor in the gastrointestinal tract. Changes in amino acid metabolites have been implicated in tumorigenesis and disease progression. Biomarkers on the basis of chiral amino acids, especially D-amino acids, have not been established for early diagnosis of CRC. Quantification of chiral amino acids, especially very low concentrations of endogenous D-amino acids, is technically challenging. We report here the quantification of L- and D-amino acids in urine samples collected from 115 CRC patients and 155 healthy volunteers, using an improved method. The method of chiral labeling, liquid chromatography, and tandem mass spectrometry enabled separation and detection of 28 amino acids (14 L-amino acids, 13 D-amino acids and Gly). Orthogonal partial least squares discriminant analysis identified 14 targeted variables among these chiral amino acids that distinguished the CRC from the healthy controls. Binary logistic regression analysis revealed that D-α-aminobutyric acid (D-AABA), L-alanine (L-Ala), D-alanine (D-Ala), D-glutamine (D-Gln) and D-serine (D-Ser) could be potential biomarkers for CRC. A receiver operating characteristic curve analysis of combined multi-variables contributed to an area under the curve (AUC) of 0.995 with 98.3 % sensitivity and 96.8 % specificity. A model constructed with D-AABA, D-Ala, D-Gln, and D-Ser achieved an AUC of 0.988, indicating important contributions of D-amino acids to the association with CRC. Further analysis also demonstrated that the metabolic aberration was associated with age and the development of CRC, D-methionine (D-Met) was decreased in CRC patients with age over 50, and D/L-Gln in patients at stage IV was higher than patients at stage I. This study provides the signature of D-amino acids in urine samples and offers a promising strategy for developing non-invasive diagnosis of CRC.

Exploring microbiota-gut-brain axis biomarkers linked to autism spectrum disorder in prenatally chlorpyrifos-exposed Fmr1 knock-out and wild-type male rats

Fmr1 (fragile X messenger ribonucleoprotein 1)-knockout (KO) rats, modeling the human Fragile X Syndrome (FXS), are of particular interest for exploring the ASD-like phenotype in preclinical studies. Gestational exposure to chlorpyrifos (CPF) has been associated with ASD diagnosis in humans and ASD-like behaviors in rodents and linked to the microbiota-gut-brain axis. In this study, we have used both Fmr1-KO and wild-type male rats (F2 generation) at postnatal days (PND) 7 and 40 obtained after F1 pregnant females were randomly exposed to 1 mg/kg/mL/day of CPF or vehicle. A nuclear magnetic resonance (NMR) metabolomics approach together with gene expression profiles of these F2 generation rats were employed to analyze different brain regions (such as prefrontal cortex, hippocampus, and cerebellum), whole large intestine (at PND7) and gut content (PND40). The statistical comparison of each matrix spectral profile unveiled tissue-specific metabolic fingerprints. Significant variations in some biomarker levels were detected among brain tissues of different genotypes, including taurine, myo-inositol, and 3-hydroxybutyric acid, and exposure to CPF induced distinct metabolic alterations, particularly in serine and myo-inositol. Additionally, this study provides a set of metabolites associated with gastrointestinal dysfunction in ASD, encompassing several amino acids, choline-derived compounds, bile acids, and sterol molecules. In terms of gene expression, genotype and gestational exposure to CPF had only minimal effects on decarboxylase 2 (gad2) and cholinergic receptor muscarinic 2 (chrm2) genes.

Metabolomic landscape of overall and common cancers in the UK Biobank: A prospective cohort study

Information about the NMR metabolomics landscape of overall, and common cancers is still limited. Based on a cohort of 83,290 participants from the UK Biobank, we used multivariate Cox regression to assess the associations between each of the 168 metabolites with the risks of overall cancer and 20 specific types of cancer. Then, we applied LASSO to identify important metabolites for overall cancer risk and obtained their associations using multivariate cox regression. We further conducted mediation analysis to evaluate the mediated role of metabolites in the effects of traditional factors on overall cancer risk. Finally, we included the 13 identified metabolites as predictors in prediction models, and compared the accuracies of our traditional models. We found that there were commonalities among the metabolic profiles of overall and specific types of cancer: the top 20 frequently identified metabolites for 20 specific types of cancer were all associated with overall cancer; most of the specific types of cancer had common identified metabolites. Meanwhile, the associations between the same metabolite with different types of cancer can vary based on the site of origin. We identified 13 metabolic biomarkers associated with overall cancer, and found that they mediated the effects of traditional factors. The accuracies of prediction models improved when we added 13 identified metabolites in models. This study is helpful to understand the metabolic mechanisms of overall and a wide range of cancers, and our results also indicate that NMR metabolites are potential biomarkers in cancer diagnosis and prevention.

Precise visualization and ROS-dependent photodynamic therapy of colorectal cancer with a novel mitochondrial viscosity photosensitive fluorescent probe

BACKGROUND

Colorectal cancer (CRC) is a prominent global cancer with high mortality rates among human beings. Efficient diagnosis and treatment have always been a challenge for CRC management. Fluorescence guided cancer therapy, which combines diagnosis with therapy into one platform, has brought a new chance for achieving precise cancer theranostics. Among this, photosensitizers, applied in photodynamic therapy (PDT), given the integration of real-time imaging capacity and efficacious treatment feasibility, show great potential to serve as remarkable tools. Although much effort has been put into constructing photosensitizers for locating and destroying CRC cells, it is still in high need to develop novel photosensitizers to attain specific detection and fulfil effective therapy.

METHODS

Probe HTI was rational synthesized for the diagnosis and treatment of CRC. Spectrometric determination was carried out first, followed by the 1O2 generation ability test. Then, HTI was displayed in distinguishing CRC cells from normal cells Further, the PDT effect of the photosensitizer was studied in vitro. Additionally, HTI was used in CRC BALB/c nude mice model to validate its viscosity labelling and tumor suppression characteristics.

RESULTS

We successfully fabricated a mitochondrial targeting probe, HTI, together with remarkable viscosity sensitivity, ultralow background interference, and excellent 1O2 generation capacity. HTI was favorably applied to the viscosity detection, displaying a 11-fold fluorescent intensity enhancement in solvents from 1.57 cp to 2043 cp. Then, it was demonstrated that HTI could distinguish CRC cells from normal cells upon the difference in mitochondrial viscosity. Moreover, HTI was qualified for producing 1O2 with high efficiency in cells, supported by the sparkling signals of DCFH after incubation with HTI under light irradiation. More importantly, the viscosity labelling and tumor suppression performance in CRC CDX model was determined, enriching the multifunctional validation of HTI in vivo.

CONCLUSIONS

In this study, HTI was demonstrated to show a sensitive response to mitochondrial viscosity and possess a high 1O2 generation capacity. Both in vitro cell imaging and in vivo tumor treatment trials proved that HTI was effectively served as a robust scaffold for tumor labeling and CRC cells clearance. This breakthrough discovery held immense potential for advancing the early diagnosis and management of CRC through PDT. By leveraging HTI's properties, medical professionals could benefit from improved diagnostic accuracy and targeted treatment in CRC management, ultimately leading to enhanced patient outcomes.

Serum nuclear magnetic resonance metabolomics analysis of human metastatic colorectal cancer: Biomarkers and pathway analysis

We describe the use of nuclear magnetic resonance metabolomics to analyze blood serum samples from healthy individuals (n = 26) and those with metastatic colorectal cancer (CRC; n = 57). The assessment, employing both linear and nonlinear multivariate data analysis techniques, revealed specific metabolite changes associated with metastatic CRC, including increased levels of lactate, glutamate, and pyruvate, and decreased levels of certain amino acids and total fatty acids. Biomarker ratios such as glutamate-to-glutamine and pyruvate-to-alanine were also found to be related to CRC. The study also found that glutamate was linked to progression-free survival and that both glutamate and 3-hydroxybutyrate were risk factors for metastatic CRC. Additionally, gas chromatography coupled to flame-ionization detection was utilized to analyze the fatty acid profile and pathway analysis was performed on the profiled metabolites to understand the metabolic processes involved in CRC. A correlation was also found between the presence of certain metabolites in the blood of CRC patients and certain clinical features.

Butyrate producers, "The Sentinel of Gut": Their intestinal significance with and beyond butyrate, and prospective use as microbial therapeutics

Gut-microbial butyrate is a short-chain fatty acid (SCFA) of significant physiological importance than the other major SCFAs (acetate and propionate). Most butyrate producers belong to the Clostridium cluster of the phylum Firmicutes, such as Faecalibacterium, Roseburia, Eubacterium, Anaerostipes, Coprococcus, Subdoligranulum, and Anaerobutyricum. They metabolize carbohydrates via the butyryl-CoA: acetate CoA-transferase pathway and butyrate kinase terminal enzymes to produce most of butyrate. Although, in minor fractions, amino acids can also be utilized to generate butyrate via glutamate and lysine pathways. Butyrogenic microbes play a vital role in various gut-associated metabolisms. Butyrate is used by colonocytes to generate energy, stabilizes hypoxia-inducible factor to maintain the anaerobic environment in the gut, maintains gut barrier integrity by regulating Claudin-1 and synaptopodin expression, limits pro-inflammatory cytokines (IL-6, IL-12), and inhibits oncogenic pathways (Akt/ERK, Wnt, and TGF-β signaling). Colonic butyrate producers shape the gut microbial community by secreting various anti-microbial substances, such as cathelicidins, reuterin, and β-defensin-1, and maintain gut homeostasis by releasing anti-inflammatory molecules, such as IgA, vitamin B, and microbial anti-inflammatory molecules. Additionally, butyrate producers, such as Roseburia, produce anti-carcinogenic metabolites, such as shikimic acid and a precursor of conjugated linoleic acid. In this review, we summarized the significance of butyrate, critically examined the role and relevance of butyrate producers, and contextualized their importance as microbial therapeutics.

NMR-Based Metabolomics to Evaluate Individual Response to Treatments

The aim of this chapter is to highlight the various aspects of metabolomics in relation to health and diseases, starting from the definition of metabolic space and of how individuals tend to maintain their own position in this space. Physio-pathological stimuli may cause individuals to lose their position and then regain it, or move irreversibly to other positions. By way of examples, mostly selected from our own work using ¹H NMR on biological fluids, we describe the effects on the individual metabolomic fingerprint of mild external interventions, such as diet or probiotic administration. Then we move to pathologies (such as celiac disease, various types of cancer, viral infections, and other diseases), each characterized by a well-defined metabolomic fingerprint. We describe the effects of drugs on the disease fingerprint and on its reversal to a healthy metabolomic status. Drug toxicity can be also monitored by metabolomics. We also show how the individual metabolomic fingerprint at the onset of a disease may discriminate responders from non-responders to a given drug, or how it may be prognostic of e.g., cancer recurrence after many years. In parallel with fingerprinting, profiling (i.e., the identification and quantification of many metabolites and, in the case of selected biofluids, of the lipoprotein components that contribute to the ¹H NMR spectral features) can provide hints on the metabolic pathways that are altered by a disease and assess their restoration after treatment.

NMR in Metabolomics: From Conventional Statistics to Machine Learning and Neural Network Approaches

NMR measurements combined with chemometrics allow achieving a great amount of information for the identification of potential biomarkers responsible for a precise metabolic pathway. These kinds of data are useful in different fields, ranging from food to biomedical fields, including health science. The investigation of the whole set of metabolites in a sample, representing its fingerprint in the considered condition, is known as metabolomics and may take advantage of different statistical tools. The new frontier is to adopt self-learning techniques to enhance clustering or classification actions that can improve the predictive power over large amounts of data. Although machine learning is already employed in metabolomics, deep learning and artificial neural networks approaches were only recently successfully applied. In this work, we give an overview of the statistical approaches underlying the wide range of opportunities that machine learning and neural networks allow to perform with accurate metabolites assignment and quantification.Various actual challenges are discussed, such as proper metabolomics, deep learning architectures and model accuracy.