DEVELOPING A CHLOROPHYLL-BASED PREDICTION MODEL FOR EICOSAPENTAENOIC ACID (EPA) PRODUCTION IN PHYTOPLANKTON

Abstract

Phytoplankton are ideal candidates for sustainable high-scale production of polyunsaturated fatty acids such as eicosapentaenoic acid (EPA). However, a rapid and reliable technique to model EPA production has yet to be validated. Chlorophyll-a (Chl-a) and EPA are both constituents of chloroplast membranes and are thought to vary similarly with temperature and irradiance. In contrast to EPA, estimation of Chl-a is rapid, inexpensive, and precise. We predicted that Chl-a would be a reliable proxy for EPA production in microalgae. To test this hypothesis, three EPA-producers, Nannochloropsis oculata, Pavlova lutheri, and Thalassiosira pseudonana, were grown at three temperatures (15, 20, and 25 ºC) and three irradiances (22, 105, and 260 µmol photons m-2 s-1). Overall FA profiles of the three species were different and general patterns in the responses to light and temperature were shared among species. The % EPA (relative to total mass FA) rose as irradiance decreased at a given temperature. As expected, the same trend was observed for the ratio of EPA to cell carbon in P. lutheri and T. pseudonana but there was no change in N. oculata. The highest production rates were found at moderate and high irradiances where specific growth rates were highest, despite decreased biomass fraction. There was little influence of temperature in the range tested. EPA/Chl-a did not show a consistent behavior among the three species, with the maximum value ranging from 1-4 g g-1. However, in P. lutheri, the ratio was remarkably predictable, varying consistently with irradiance at all three temperatures. The proposed EPA/Chl-a model can aid in screening out inferior EPA producers using a threshold of 1 g g-1, and provides a database for future EPA/Chl-a analysis in phytoplankton.