Metabolomics aims to provide a global snapshot of all small-molecule metabolites in cells and biological fluids, free of observational biases inherent to more focused studies of metabolism. However, the staggeringly high information content of such global analyses introduces a challenge of its own; efficiently forming biologically relevant conclusions from any given metabolomics dataset indeed requires specialized forms of data analysis. One approach to finding meaning in metabolomics datasets involves multivariate analysis (MVA) methods such as principal component analysis (PCA) and partial least squares projection to latent structures (PLS), where spectral features contributing most to variation or separation are identified for further analysis. However, as with any mathematical treatment, these methods are not a panacea; this review discusses the use of multivariate analysis for metabolomics, as well as common pitfalls and misconceptions.

Background: Principal Component Analysis (PCA) and Orthogonal Projections to Latent Structures Discriminant Analysis (OPLS-DA) are powerful statistical modeling tools that provide insights into separations between experimental groups based on high-dimensional spectral measurements from NMR, MS or other analytical instrumentation. However, when used without validation, these tools may lead investigators to statistically unreliable conclusions. This danger is especially real for Partial Least Squares (PLS) and OPLS, which aggressively force separations between experimental groups. As a result, OPLS-DA is often used as an alternative method when PCA fails to expose group separation, but this practice is highly dangerous. Without rigorous validation, OPLS-DA can easily yield statistically unreliable group separation. Methods: A Monte Carlo analysis of PCA group separations and OPLS-DA cross-validation metrics was performed on NMR datasets with statistically significant separations in scores-space. A linearly increasing amount of Gaussian noise was added to each data matrix followed by the construction and validation of PCA and OPLS-DA models. Results: With increasing added noise, the PCA scores-space distance between groups rapidly decreased and the OPLS-DA cross-validation statistics simultaneously deteriorated. A decrease in correlation between the estimated loadings (added noise) and the true (original) loadings was also observed. While the validity of the OPLS-DA model diminished with increasing added noise, the group separation in scoresspace remained basically unaffected. Conclusion: Supported by the results of Monte Carlo analyses of PCA group separations and OPLS-DA cross-validation metrics, we provide practical guidelines and cross-validatory recommendations for reliable inference from PCA and OPLS-DA models.

Aberrant energy metabolism is a hallmark of cancer. To fulfill the increased energy requirements, tumor cells secrete cytokines/factors inducing muscle and fat degradation in cancer patients, a condition known as cancer cachexia. It accounts for nearly 20% of all cancer-related deaths. However, the mechanistic basis of cancer cachexia and therapies targeting cancer cachexia thus far remain elusive. A ketogenic diet, a high-fat and low-carbohydrate diet that elevates circulating levels of ketone bodies (i.e., acetoacetate, β-hydroxybutyrate, and acetone), serves as an alternative energy source. It has also been proposed that a ketogenic diet leads to systemic metabolic changes. Keeping in view the significant role of metabolic alterations in cancer, we hypothesized that a ketogenic diet may diminish glycolytic flux in tumor cells to alleviate cachexia syndrome and, hence, may provide an efficient therapeutic strategy. We observed reduced glycolytic flux in tumor cells upon treatment with ketone bodies. Ketone bodies also diminished glutamine uptake, overall ATP content, and survival in multiple pancreatic cancer cell lines, while inducing apoptosis. A decrease in levels of c-Myc, a metabolic master regulator, and its recruitment on glycolytic gene promoters, was in part responsible for the metabolic phenotype in tumor cells. Ketone body-induced intracellular metabolomic reprogramming in pancreatic cancer cells also leads to a significantly diminished cachexia in cell line models. Our mouse orthotopic xenograft models further confirmed the effect of a ketogenic diet in diminishing tumor growth and cachexia. Thus, our studies demonstrate that the cachectic phenotype is in part due to metabolic alterations in tumor cells, which can be reverted by a ketogenic diet, causing reduced tumor growth and inhibition of muscle and body weight loss.

Staphylococcus aureus must rapidly adapt to a variety of carbon and nitrogen sources during invasion of a host. Within a staphylococcal abscess, preferred carbon sources such as glucose are limiting, suggesting that S. aureus survives through the catabolism of secondary carbon sources. S. aureus encodes pathways to catabolize multiple amino acids, including those that generate pyruvate, 2-oxoglutarate, and oxaloacetate. To assess amino acid catabolism, S. aureus JE2 and mutants were grown in complete defined medium containing 18 amino acids but lacking glucose (CDM). A mutation in the gudB gene, coding for glutamate dehydrogenase, which generates 2-oxoglutarate from glutamate, significantly reduced growth in CDM, suggesting that glutamate and those amino acids generating glutamate, particularly proline, serve as the major carbon source in this medium. Nuclear magnetic resonance (NMR) studies confirmed this supposition. Furthermore, a mutation in the ackA gene, coding for acetate kinase, also abrogated growth of JE2 in CDM, suggesting that ATP production from pyruvate-producing amino acids is also critical for growth. In addition, although a functional respiratory chain was absolutely required for growth, the oxygen consumption rate and intracellular ATP concentration were significantly lower during growth in CDM than during growth in glucose-containing media. Finally, transcriptional analyses demonstrated that expression levels of genes coding for the enzymes that synthesize glutamate from proline, arginine, and histidine are repressed by CcpA and carbon catabolite repression. These data show that pathways important for glutamate catabolism or ATP generation via Pta/AckA are important for growth in niches where glucose is not abundant, such as abscesses within skin and soft tissue infections. IMPORTANCE S. aureus is a significant cause of both morbidity and mortality worldwide. This bacterium causes infections in a wide variety of organ systems, the most common being skin and soft tissue. Within a staphylococcal abscess, levels of glucose, a preferred carbon source, are limited due to the host immune response. Therefore, S. aureus must utilize other available carbon sources such as amino acids or peptides to proliferate. Our results show that glutamate and amino acids that serve as substrates for glutamate synthesis, particularly proline, function as major carbon sources during growth, whereas other amino acids that generate pyruvate are important for ATP synthesis via substrate-level phosphorylation in the Pta-AckA pathway. Our data support a model whereby certain amino acid catabolic pathways, and acquisition of those particular amino acids, are crucial for growth in niches where glucose is not abundant.

Metabolomics investigates global metabolic alterations associated with chemical, biological, physiological, or pathological processes. These metabolic changes are measured with various analytical platforms including liquid chromatography-mass spectrometry (LC-MS), gas chromatography-mass spectrometry (GC-MS) and nuclear magnetic resonance spectroscopy (NMR). While LC-MS methods are becoming increasingly popular in the field of metabolomics (accounting for more than 70% of published metabolomics studies to date), there are considerable benefits and advantages to NMR-based methods for metabolomic studies. In fact, according to PubMed, more than 926 papers on NMR-based metabolomics were published in 2021—the most ever published in a given year. This suggests that NMR-based metabolomics continues to grow and has plenty to offer to the scientific community. This perspective outlines the growing applications of NMR in metabolomics, highlights several recent advances in NMR technologies for metabolomics, and provides a roadmap for future advancements.