Human saliva offers many advantages over other biofluids regarding its use and value as a bioanalytical medium for the identification and prognostic monitoring of human diseases, mainly because its collection is largely non-invasive, is relatively cheap, and does not require any major clinical supervision, nor supervisory input. Indeed, participants donating this biofluid for such purposes, including the identification, validation and quantification of surrogate biomarkers, may easily self-collect such samples in their homes following the provision of full collection details to them by researchers. In this report, the authors have focused on the applications of metabolomics technologies to the diagnosis and progressive severity monitoring of human cancer conditions, firstly oral cancers (e.g., oral cavity squamous cell carcinoma), and secondly extra-oral (systemic) cancers such as lung, breast and prostate cancers. For each publication reviewed, the authors provide a detailed evaluation and critical appraisal of the experimental design, sample size, ease of sample collection (usually but not exclusively as whole mouth saliva (WMS)), their transport, length of storage and preparation for analysis. Moreover, recommended protocols for the optimisation of NMR pulse sequences for analysis, along with the application of methods and techniques for verifying and resonance assignments and validating the quantification of biomolecules responsible, are critically considered. In view of the authors’ specialisms and research interests, the majority of these investigations were conducted using NMR-based metabolomics techniques. The extension of these studies to determinations of metabolic pathways which have been pathologically disturbed in these diseases is also assessed here and reviewed. Where available, data for the monitoring of patients’ responses to chemotherapeutic treatments, and in one case, radiotherapy, are also evaluated herein. Additionally, a novel case study featured evaluates the molecular nature, levels and diagnostic potential of 1H NMR-detectable salivary ‘acute-phase’ glycoprotein carbohydrate side chains, and/or their monomeric saccharide derivatives, as biomarkers for cancer and inflammatory conditions.

This communication represents Part III of our series of reports based on the applications of human saliva as a useful and conveniently collectable medium for the discovery, identification and monitoring of biomarkers, which are of some merit for the diagnosis of human diseases. Such biomarkers, or others reflecting the dysfunction of specific disease-associated metabolic pathways, may also be employed for the prognostic pathological tracking of these diseases. Part I of this series set the experimental and logistical groundwork for this report, and the preceding paper, Part II, featured the applications of newly developed metabolomics technologies to the diagnosis and severity grading of human cancer conditions, both oral and systemic. Clearly, there are many benefits, both scientific and economic, associated with the donation of human saliva samples (usually as whole mouth saliva) from humans consenting to and participating in investigations focused on the discovery of biomolecular markers of diseases. These include usually non-invasive collection protocols, relatively low cost when compared against blood sample collection, and no requirement for clinical supervision during collection episodes. This paper is centred on the employment and value of ‘state-of-the-art’ metabolomics technologies to the diagnosis and prognosis of a wide range of non-cancerous human diseases. Firstly, these include common oral diseases such as periodontal diseases (from type 1 (gingivitis) to type 4 (advanced periodontitis)), and dental caries. Secondly, a wide range of extra-oral (systemic) conditions are covered, most notably diabetes types 1 and 2, cardiovascular and neurological diseases, and Sjögren’s syndrome, along with a series of viral infections, e.g., pharyngitis, influenza, HIV and COVID-19. Since the authors’ major research interests lie in the area of the principles and applications of NMR-linked metabolomics techniques, many, but not all, of the studies reviewed were conducted using these technologies, with special attention being given to recommended protocols for their operation and management, for example, satisfactory experimental model designs; sample collection and laboratory processing techniques; the selection of sample-specific NMR pulse sequences for saliva analysis; and strategies available for the confirmation of resonance assignments for both endogenous and exogenous molecules in this biofluid. This article also features an original case study, which is focussed on the use of NMR-based salivary metabolomics techniques to provide some key biomarkers for the diagnosis of pharyngitis, and an example of how to ‘police’ such studies and to recognise participants who perceive that they actually have this disorder but do not from their metabolic profiles and multivariate analysis pattern-based clusterings. The biochemical and clinical significance of these multidimensional metabolomics investigations are discussed in detail.

Ammonia (NH3) has been shown to be a key biomarker for a wide variety of diseases, such as hepatic and chronic kidney diseases (CKD), and cancers. It also has relevance to the oral health research area, and, hence, its determination in appropriate biofluids and tissues is of much importance. However, since it contains exchangeable >N-H protons, its analysis via 1H NMR spectroscopy, which is a widely employed technique in untargeted metabolomic studies, is rendered complicated. In this study, we focused on the 1H NMR analysis of this biomarker in less invasively collected human saliva samples, and we successfully identified and quantified it as ammonium cation (NH4+) in post-collection acidulated forms of this biofluid using both the standard calibration curve and standard addition method (SAM) approaches. For this purpose, n = 27 whole mouth saliva (WMS) samples were provided by healthy human participants, and all donors were required to follow a fasting/oral environment abstention period of 8 h prior to collection. Following acidification (pH 2.00), diluted WMS supernatant samples treated with 10% (v/v) D2O underwent 1H NMR analysis (600 MHz). The acquired results demonstrated that NH4+ can be reliably determined in these supernatants via integration of the central line of its characteristic 1:1:1 intensity triplet resonance (complete spectral range δ = 6.97–7.21 ppm). Experiments performed also demonstrated that any urease-catalysed NH3 generation occurring post-sampling in WMS samples did not affect the results acquired during the usual timespan of laboratory processing required prior to analysis. Further experiments demonstrated that oral mouth-rinsing episodes conducted prior to sample collection, as reported in previous studies, gave rise to major decreases in salivary NH4+ levels thereafter, which renormalised to only 50–60% of their basal control concentrations at the 180-minute post-rinsing time point. Therefore, the WMS sample collection method employed significantly affected the absolute levels of this analyte. The LLOD was 60 μmol/L with 128 scans. The mean ± SD salivary NH4+ concentration of WMS supernatants was 11.4 ± 4.5 mmol/L. The potential extension of these analytical strategies to the screening of other metabolites with exchangeable 1H nuclei is discussed, as is their relevance to the monitoring of human disorders involving the excessive generation and/or uptake of cellular/tissue material, or altered homeostasis, in NH3.