As discussed earlier, environmental variables (particularly nutrient status) can have great effects on growth and the quantity and quality of lipids in microalgae. To determine the effects of several environmental variables alone and in combination on the growth and lipid contents of microalgae, a multivariate, fractional factorial design experiment was carried out with two promising diatoms, Navicula saprophila (NAVIC1) and C. muelleri (CHAET9). For these experiments, cells were grown in modified SERI Type II/25 medium in which the alkalinity was adjusted by adding sodium carbonate and sodium bicarbonate and the conductivity was adjusted by adding sodium chloride. Cultures were grown on the temperature gradient table described previously at 200 qE^m-2^s-1. The following variables were tested in the multivariate analysis: conductivity (20 to 80 mmho^cm-1), temperature (17° to 32°C), N (urea) concentration (0 to 144 mg^L-1), sodium silicate concentration (0 to 500 mg-L-1), and alkalinity (8.8 to 88 meq^L-1). In these experiments, growth was measured by changes in AFDW and lipid was measured by the use of the Nile Red fluorometric assay.

The results indicated that the N content and conductivity of the medium were the most important variables affecting lipid content (Nile Red fluorescence) of both NAVIC1 and CHAET9. As N levels and conductivity increased, the amount of neutral lipid per mg of AFDW decreased. The interaction of N and conductivity was an important determinant of lipid content as well. Silicon level and alkalinity were more important factors in determining the lipid content for CHAET9 than for NAVIC 1. N concentration was by far the most important factor in determining final cell mass for NAVIC 1, and was a major factor for cell mass yield in CHAET9 (along with the interaction of conductivity and alkalinity, which had a large negative impact on growth). Alkalinity was a major factor for growth of both NAVIC 1 and CHAET9. However, these experiments did not determine actual growth rates, but only the final cell yields; thus, how actual cell division rates compared with each other is not known.

These experiments indicate the importance of examining the interactions of environmental variables in determining the effects on growth and lipid production. However, the models generated by these kinds of experiments are specific for the strains being studied, and the results cannot necessarily be used to predict the effects of these variables on other strains. Furthermore, for such models to be truly predictive of growth and lipid production in an actual mass culture, much more sophiticated (and realistic) experimental setups would be required.

I Publications:

Chelf, P. (1990) “Environmental control of lipid and biomass production in two diatom species.” J. Appl. Phycol. 2:121-129.