Aspects of Microalgae Biology Related to Cultivation for Biofuel Production: Microbial Phycosphere Interactions and Adaptive Physiology

Abstract

The interest in microalgal cultivation for the production of renewable biofuel and other applications has grown significantly over the last several decades in close correspondence with increasing concerns over global climate change and energy security. These unicellular organisms fix atmospheric carbon photosynthetically while offering a wide range of desirable characteristics that are not found in other categories of feedstock crops. Owing to their ubiquitous distribution in nature, microalgae have evolved into species capable of growing at extremely rapid rates under many varied conditions and to synthesize very high concentrations of lipids upon induction. Screening, identification, and characterization efforts have yielded promising results, projecting certain algal strains to be capable of producing significantly more biodiesel per unit of land than any other studied crop. There remain challenges to economical large-scale production at each aspect of the process including strain selection/optimization, improving cultivation, harvesting biomass, and product extraction. Life-cycle analyses and modeling predictions have concluded that cultivation in outdoor, open style reactor systems are critical to meet national and international goals for increased biofuel usage. This type of system not only exposes the target algal strains to unpredictable environmental conditions, but also exposes them to other microorganisms, whose total interactions result in a phycosphere, or zone of microbial influence. The relationships between algae and their associated phycosphere members range from mutualistic to parasitic, depending on the relative abundances of species and on environmental factors. Understanding how these relationships affect algal growth in their industrial context provides guidance for successful manipulation of microbial consortia for maximum productivity. Minimization of inputs is an important goal, regardless of the growth system chosen. Several algal strains have been manipulated by adaptive laboratory evolution to produce greater amounts of biomass, carotenoids, lipids, and other products without additional nutrient inputs. This process is not only capable of producing novel strains with valuable characteristics, but also offers an opportunity to dissect the adaptive physiological and metabolic mechanisms employed by the algae. Here I focus on aspects of algal biology related to improving and stabilizing growth characteristics for optimizing biomass cultivation. The experiments described herein follow two major themes of algal biology. First, the effects of bacterial phycosphere members are considered from the perspectives of overall community ecology and individual species’ relationships to the host algae. Second, reflections on the adaptability of microalgae to oligotrophic conditions to determine factors allowing for more efficient macronutrient metabolism and their impacts on lipid content intended for conversion to biodiesel. To determine the influence of bacterial members of the phycosphere on algae grown in open, outdoor, reactors at an industrial pilot scale, my coauthors and I used high throughput sequence data from 16S rRNA V4 region amplicons to characterize complete prokaryotic assemblages and their relationships to a variety of important environmental and engineered factors. These experiments were conducted with collaborators on the Regional Algal Feedstock Testbed (RAFT) project, which operated a variety of open raceway style reactors to test aspects of microalgae cultivation in the Southwest United States. The primary algal production species used during the project was a strain of the chlorophyte, Chlorella sorokiniana (Shihira and Krauss 1965), designated DOE1412. Regular sampling across 41 growth cycles spanning two separate seasons yielded one of the most comprehensive surveys of prokaryotic phycospheres in mass algal cultures to date. Longitudinal analyses of taxonomic assignments of 16S rRNA sequence data revealed that phycosphere diversity increased significantly with the amount of time in outdoor culture, and that there were consistent patterns associated with the stationary, growth, and death phases of algal cultures. Furthermore, computation of a Health Index metric that compared observed algal growth to model-predicted growth identified strong positive correlations with bacteria belonging to the Burholderiaceae and Allorhizobium-Neorhizobium-Pararhizobium-Rhizobium clade, determined by rank-based multinomial regression. The results also confirmed the predominance of the algicidal bacteria, Vampirovibrio chlorellavorus, during the summer months, and that a benzalkonium chloride treatment regimen suppressed its growth along with several other bacterial taxa. To understand the molecular mechanisms by which an algal strain adapted to more efficient growth under phosphorus limited conditions, my coauthors and I collected periodic transcriptomic and lipidomic data from algal cell populations subjected adaptive laboratory evolution in continuous cultures. The green microalgae, Auxenochlorella protothecoides, was grown in continuous culture with 100 times less inorganic phosphate (Pi) than the amount in the typical growth media for more than 41 generations. The resulting algal populations grew more rapidly in low Pi media than unadapted progenitor cells at an average maximal rate of 0.72 d-1 and 0.54 d-1 respectively. The fatty acid (FA) profiles of the adapted cells shifted minorly from the wild type A. protothecoides, with relative increases in one monounsaturated species being compensated by decreases in another, indicating very little effect on the biodiesel product. The mechanisms by which the algae adapted to low Pi growth were typified by an early and late stage wherein transcriptomic and lipid profiles differed significantly in samples collected prior to ~ 11 generations compared to results after ~ 34 generations. The short-term changes in gene expression were associated with shifts in major metabolic pathways including carbon metabolism, oxidative phosphorylation, glycolysis, and gluconeogenesis. By comparison, certain transcripts showing decreased expression, reflected increased fatty acid turnover, and a stable decrease in photosynthesis-related gene expression. The most significant changes in lipids occurred in the glycerolipid class with adapted populations containing 306% monogalactosyldiacylglycerol (MGDG) and 189% sulfoquinovosyldiacylglycerol (SQDG) of the amounts in unadapted cells by the final sampling time point at 41 generations after fluctuating throughout the experiments. These results provided insights into some basic variations of low Pi adaptation displayed by A. protothecoides compared to other algal species and to plants that could be utilized for optimization of other production strains. Lastly, my collaborators and I investigated the viral-like element, prophage/plasmid content, of available genomes of a pernicious bacterial pathogen of a biofuel production microalgae species using comparative genomic approaches. By combining three prophage identification algorithms, we were able to identify 14 putative integrated prophage regions among the three currently published Vampirovibrio chlorellavorus bacterial genome assemblies. In addition to the previously described circular plasmids in the genome assembly of an isolate collected from a Ukrainian lake, Vc_UKR, at least one prophage was found in all genomes. Because the other two assemblies were derived from isolates collected in an algal biofuel production reactor in the Southwest United States, these findings implicate a heterogenous distribution of viral-like elements likely contributing to fitness and pathogenicity of V. chlorellavorus populations. The Vc_AZ2 and Vc_UKR both contained prophage regions homologous to genomic regions of Melainabacteria sp. nbed3b74 (GenBank; GCA_902168245.1), a closely related soil borne bacteria found to contribute to disease suppression in plant rhizospheres. Additionally, functional predictions indicated that prophage regions contained genes involved in pathogenicity, including 10 toxin-antitoxin system genes, 19 transcriptional regulators, and other bacterial virulence factors. One of these was the identification of TonB-dependent receptor (TBDR) genes that are known to require several accessory genes to complete their function of collecting extracellular iron across the outer bacterial membrane. Intriguingly, homologs to TBDR-related genes were found in chromosomal locations. These findings represent the first reports of predicted MGE involvement in the pathogenicity of an algicidal bacteria.