Advancing the Quality Diagnosis and Monitoring of Aquatic Pollution ^†

Abstract

Introduction: Aquatic chemical pollution is among the most worrying threats to ecosystem health. There is an ever-increasing variety of pollutant substances detected across the source-to-sea continuum, causing loss of biodiversity and ecological disequilibrium. Achieving cleaner and healthier systems relies on carrying out sustained, cost-effective, diagnosis and aquatic effects monitoring, within the adaptive management cycle. The available methods are, however, cumbersome, which creates a clear need for innovative expeditious approaches for low-cost surveillance monitoring. In the last decade, Raman Spectroscopy (RS) has gained wide recognition for application to biological questions, for its ability to uncover the complexity of molecules and their interactions. Various fields, from pharmacology to disease diagnosis and prognosis, have suffered an innovation revolution through the application of RS. In this technique inelastic light scattering of a small part of photons of an incident electromagnetic monochromatic light beam (ranging from near-infrared to visible or ultraviolet) is caused by the molecular vibration of chemical bonds. This results in shifts in energy, which indicate discrete vibrational modes of polarisable molecules, providing qualitative and quantitative assessments of the chemical composition and molecular structure of the sample. The technique shows high sensitivity, no need for sample preparation and the possibility of use in non-invasive and label-free analysis. Objective: The aim of this work is to present and discuss evidence about the application of Raman Spectroscopy (RS) to environmental diagnosis and aquatic effect monitoring of pollution. Methodology: The technique was applied to different biological models, i.e., diatoms, zebrafish embryos and larvae and freshwater snails. Quality assessments with diatoms were tested in environmental monitoring, while assessments with other models were done upon exposure to metals and organic contaminants. Results and conclusions: The Raman spectra obtained from the samples analysed comprised bands detected within the 800 to 2000 cm⁻¹ wavenumber range. These were related to bond vibrations of carbohydrates, DNA phosphate groups, proteins or CH, NH and OH stretching in lipids and proteins. Data analysis using chemometric methods clearly distinguished pollutant exposure from control sites or treatments, pointing out the potential for surveyance monitoring. The next steps include the comparison with other sensitive methods (e.g., locomotion and avoidance behaviours, omics methods) to assess efficiency and bring further mechanistic understanding.