Cellular energy allocation analysis of multiple marine bivalves using near infrared spectroscopy. (July 2018)

• Record Type: Journal Article
• Authors: Bartlett, Jill K.; Maher, William A.; Purss, Matthew B.J.
• Is Part Of: Ecological indicators. Volume 90(2018)
• Page Start: 247
• Page End: 256
• Publication Date: 2018-07
• Subjects: Near infrared spectroscopy -- Bivalve -- Cellular energy allocation -- Protein content -- Lipid content -- Glycogen content -- Electron transport system activity; Environmental monitoring -- Periodicals; Environmental management -- Periodicals; Environmental impact analysis -- Periodicals; Environmental risk assessment -- Periodicals; Sustainable development -- Periodicals; 333.71405
• DOI: 10.1016/j.ecolind.2018.03.007
• Languages: English
• ISSNs: 1470-160X

Abstract

Graphical abstract: Highlights: NIRS was used to build quantitative models of protein, lipid, carbohydrate and ETS. 5 marine bivalve species were included in each model. A data driven approach was used to build each model. Model outputs were used to calculate cellular energy allocation (CEA). Model application demonstrated CEA response to seasonal change and metal stress. Abstract: This paper proposes an alternative procedure for calculating Cellular Energy Allocation (CEA) in marine bivalve species. CEA is a measure of the difference in total energy available and energy consumption and is used to assess the health of organism fitness for growth and reproduction. In this study near infra-red spectral (NIRS) quantitative data-driven modelling techniques were applied to total energy available (total protein, lipid and glycogen content) and energy consumption (electron transport system activity (ETS)) of five marine bivalve species ( Saccostrea glomerata, Ostrea angasi, Crassostrea gigas, Mytilus galloprovincialis and Anadara trapezia ) to produce models for measuring each bioenergetic component and calculating CEA, a biomarker of energetic stress response. Comparison of predicted to measured data showed low prediction errors (RMSEP) and high ratio of predictive error to inter-quartile distance (RPIQ) values. RMSEP for protein, lipid, glycogen and ETS models were 8.7%, 6.8%, 7.6% and 12% of the dynamic range of the validation datasets respectively while RPIQ were 3.6, 5.2, 4 and 3.7 respectively. Conversion of model outputs to energetic equivalents and calculation of total energy stores as per the CEA method demonstrate a close relationship between measured and modelled results (105%±3%). These metrics indicate that the four energy models are accurate, robust and reliable for quantitative application to all five species within the sample variation of the energetic storage measures. Applying NIRS quantitative models using freeze dried tissue eliminates water interference in NIRS capture and provides improved sample homogeneity for both spectral capture and chemical analysis. This technique enables simultaneous analyses of protein, lipid, glycogen and ETS activity to be achieved with a much higher throughput of samples (100–200 samples daily) than conventional energetic storage analysis techniques. In addition, the technique is non-destructive to tissue samples and freeze dried samples can be stored for additional analyses. The value of CEA as a response measure in S. glomerata are included, with examples of changes in CEA over four seasons and the response of oysters to a known metal contamination gradient.