The Aquatic Species Program (ASP) was a relatively small research effort intended to look at the use of aquatic plants as sources of energy. While its history dates back to 1978, much of the research from 1978 to 1982 was focused on using algae to produce hydrogen. The program switched emphasis to other transportation fuels, in particular biodiesel, beginning in the early 1980s. This report provides a summary of the research activities carried out from 1980 to 1996, with an emphasis on algae for biodiesel production.

In 1995, DOE made the difficult decision to eliminate funding for algae research within the Biofuels Program. Under pressure to reduce budgets, the Department chose a strategy of more narrowly focusing its limited resources in one or two key areas, the largest of these being the development of bioethanol. The purpose of this report is to bring closure to the Biofuels Program’s algae research. This report is a summary and compilation of all the work done over the last 16 years of the program. It includes work carried out by NREL researchers at our labs in Golden, as well as subcontracted research and development activities conducted by private companies and universities around the country. More importantly, this report should be seen not as an ending, but as a beginning. When the time is right, we fully expect to see renewed interest in algae as a source of fuels and other chemicals. The highlights presented here should serve as a foundation for these future efforts.

What is the technology?

Biological Concepts

Photosynthetic organisms include plants, algae and some photosynthetic bacteria. Photosynthesis is the key to making solar energy available in useable forms for all organic life in our environment. These organisms use energy from the sun to combine water with carbon dioxide (CO2) to create biomass. While other elements of the Biofuels Program have focused on terrestrial plants as sources of fuels, ASP was concerned with photosynthetic organisms that grew in aquatic environments. These include macroalgae, microalgae and emergents. Macroalgae, more commonly known as “seaweed,” are fast growing marine and freshwater plants that can grow to considerable size (up to 60m in length). Emergents are plants that grow partially submerged in bogs and marshes. Microalgae are, as the name suggests, microscopic photosynthetic organisms. Like macroalgae, these organisms are found in both marine and freshwater environments. In the early days of the program, research was done on all three types of aquatic species. As emphasis switched to production of natural oils for biodiesel, microalgae became the exclusive focus of the research. This is because microalgae generally produce more of the right kinds of natural oils needed for biodiesel (see the discussion of fuel concepts presented later in this overview).

In many ways, the study of microalgae is a relatively limited field of study. Algae are not nearly as well understood as other organisms that have found a role in today’s biotechnology industry. This is part of what makes our program so valuable. Much of the work done over the past two decades represents genuine additions to the scientific literature. The limited size of the scientific community involved in this work also makes it more difficult, and sometimes slower, compared to the progress seen with more conventional organisms. The study of microalgae represents an area of high risk and high gains.

These photosynthetic organisms are far from monolithic. Biologists have categorized microalgae in a variety of classes, mainly distinguished by their pigmentation, life cycle and basic cellular structure. The four most important (at least in terms of abundance) are:

• The diatoms (Bacillariophyceae). These algae dominate the phytoplankton of the oceans, but are also found in fresh and brackish water. Approximately 100,000 species are known to exist. Diatoms contain polymerized silica (Si) in their cell walls.

All cells store carbon in a variety of forms. Diatoms store carbon in the form of natural oils or as a polymer of carbohydrates known as chyrsolaminarin.

• The green algae (Chlorophyceae). These are also quite abundant, especially in freshwater. (Anyone who owns a swimming pool is more than familiar with this class of algae).

They can occur as single cells or as colonies. Green algae are the evolutionary progenitors of modern plants. The main storage compound for green algae is starch, though oils can be produced under certain conditions.

• The blue-green algae (Cyanophyceae). Much closer to bacteria in structure and organization, these algae play an important role in fixing nitrogen from the atmosphere. There are approximately 2,000 known species found in a variety of habitats.

• The golden algae (Chrysophyceae). This group of algae is similar to the diatoms. They have more complex pigment systems, and can appear yellow, brown or orange in color. Approximately 1,000 species are known to exist, primarily in freshwater systems. They are similar to diatoms in pigmentation and biochemical composition. The golden algae produce natural oils and carbohydrates as storage compounds.

The bulk of the organisms collected and studied in this program fall in the first two classes—the diatoms and the green algae.

Microalgae are the most primitive form of plants. While the mechanism of photosynthesis in microalgae is similar to that of higher plants, they are generally more efficient converters of solar energy because of their simple cellular structure. In addition, because the cells grow in aqueous suspension, they have more efficient access to water, CO2, and other nutrients. For these reasons, microalgae are capable of producing 30 times the amount oil per unit area of land, compared to terrestrial oilseed crops.

Put quite simply, microalgae are remarkable and efficient biological factories capable of taking a waste (zero-energy) form of carbon (CO2) and converting it into a high density liquid form of energy (natural oil). This ability has been the foundation of the research program funded by the Office Fuels Development.

Hawaii Institute of Marine Biology Richard H. York, Jr.

1985

N/A

Microalgae were collected from a variety of sites in the Hawaiian islands, including ocean sites and inland saline habitats. The conductivity, dissolved oxygen content, pH, and temperature of each site was recorded. Individual cells were isolated via micropipetting and placed into glass tubes or fluorohalocarbon plastic bags containing either the original sample water, offshore seawater, SERI Type I medium, or SERI Type II medium. The plastic bags, which transmit the full visible solar spectrum, were placed in full sunlight without temperature control. This treatment was therefore believed to provide a good selection for strains that would be able to thrive under outdoor mass culture conditions. The glass tubes were incubated at 25°-26°C at 40 pE^m-2^s-1 under a 16h:8h light:dark regime; these conditions were less stressful than the outdoor
conditions, and therefore led to the recovery of less hardy strains. A large-scale outdoor enrichment culture was also prepared by pumping 1700 L of enriched seawater into a 5.5 m diameter open tank, then strains arising in this culture were isolated.

As a result of these procedures, 100 of the most rapidly growing strains were selected and maintained for further analysis. This group included members of the Chlorophyceae, Cyanophyceae, Bacillariophyceae, and Pyrrophyceae. Two strains, Chaetoceros SH 9-1 (CHAET38) and Cyclotella 14-89 (THALA6), were grown in outdoor cultures consisting of cells in 1-L fluorohalocarbon plastic bags. The highest growth rates measured for these strains were 2.12 doublings^-1 for CHAET38 and 1.43 doublings^-1 for THALA6. These growth rates were reported to correspond to 31 and 33 g dry weight^m-2^d-1, respectively, although how these values were derived is not clear.

I Publications:

York, Jr., R. H. (1987) “Collection of high energy yielding strains of saline microalgae from the Hawaiian Islands.” FY1986 Aquatic Species Program Annual Report, Solar Energy Research Institute, Golden, Colorado, SERI/SP-231-3071, pp. 90-104.

Put less emphasis on outdoor field demonstrations and more on basic biology

Much work remains to be done on a fundamental level to maximize the overall productivity of algae mass culture systems. The bulk of this work is probably best done in the laboratory. The results of this program’s demonstration activities have proven the concept of outdoor open pond production of algae. While it is important to continue a certain amount of field work, small scale studies and research on the

basic biological issues are clearly more cost effective than large scale demonstration studies.

і Take Advantage of Plant Biotechnology

We have only scratched the surface in the area of genetic engineering for algae. With the advances occurring in this field today, any future effort on modifying algae to increase natural oil production and overall productivity are likely to proceed rapidly. The genetic engineering tools established in the program serve as a strong foundation for further genetic enhancements of algae.

і Start with what works in the field

Select strains that work well at the specific site where the technology is to be used. These native strains are the most likely to be successful. Then, focus on optimizing the production of these native strains and use them as starting points for genetic engineering work.

і Maximize photosynthetic efficiency.

Not enough is understood about what the theoretical limits of solar energy conversion are. Recent advances in our understanding of photosynthetic mechanisms at a molecular level, in conjunction with the advances being made in genetic engineering tools for plant systems, offer exciting opportunities for constructing algae which do not suffer the limitations of light saturation photoinhibition.

і Set realistic expectations for the technology

Projections for future costs of petroleum are a moving target. DOE expects petroleum costs to remain relatively flat over the next 20 years. Expecting algal biodiesel to compete with such cheap petroleum prices is unrealistic. Without some mechanism for monetizing its environmental benefits (such as carbon taxes), algal biodiesel is not going to get off the ground.

Look for near term, intermediate technology deployment opportunities such as wastewater treatment.

Excessive focus on long term energy displacement goals will slow down development of the technology. A more balanced approach is needed in which more near term opportunities can be used to launch the technology in the commercial arena. Several such opportunities exist. Wastewater treatment is a prime example. The economics of algae technology are much more favorable when it is used as a waste treatment process and as a source of fuel. This harks back to the early days of DOE’s research.

Photosynthetic organisms, including plants, algae, and some photosynthetic bacteria, efficiently utilize the energy from the sun to convert water and CO2 from the air into biomass. The Aquatic Species Program (ASP) at SERI[1] was initiated as a long term, basic research effort to produce renewable fuels and chemicals from biomass. It emphasized the use of photosynthetic organisms from aquatic environments, expecially species that grow in environments unsuitable for crop production. Early in the program, macroalgae, microalgae, and emergents were investigated for their ability to make lipids (as a feedstock for liquid fuel or chemical production) or carbohydrates (for fermentation into ethanol or anaerobic digestion for methane production). Macroalgae (seaweeds) are fast-growing marine or freshwater plants that can reach considerable size; for example, the giant brown kelp can grow a meter in 1 day and as long as 60 m. Emergents are plants such as cattails or rushes that grow partially submerged in bogs or marshes. Macroalgae and emergents were found to produce small amounts of lipid, which function mainly as structural components of the cell membranes, and produce carbohydrate for use as their primary energy storage compound. In contrast, many microalgae, (microscopic, photosynthetic organisms that live in saline or freshwater environments), produce lipids as the primary storage molecule. By the early 1980s, the decision was made to focus ASP research efforts on the use of microalgal lipids for the production of fuels and other energy products. The studies on the growth and chemical composition of macroalgae and emergents will not be discussed in this report. However, interested readers are referred to reports by subcontractors J. D. Ryther, Harbor Branch Foundation, Florida (seaweeds), and D. Pratt, from the University of Minnesota, St. Paul (emergents) listed in the Bibliography.

Microalgae, like higher plants, produce storage lipids in the form of triacyglycerols (TAGs). Although TAGs could be used to produce of a wide variety of chemicals, work at SERI focused on the production of fatty acid methyl esters (FAMEs), which can be used as a substitute for fossil-derived diesel fuel. This fuel, known as biodiesel, can be synthesized from TAGs via a simple transesterification reaction in the presence of acid or base and methanol. Biodiesel can be used in unmodified diesel engines, and has advantages over conventional diesel fuel in that it is renewable, biodegradable, and produces less SOX and particulate emissions when burned. The technology is available to produce biodiesel from TAGs, and there are growing biodiesel industries both in the United States and Europe that use soybean or rapeseed oil as the biodiesel feedstock. However, the potential market for biodiesel far surpasses the availability of plant oils not designated for other markets. Thus, there was significant interest in the development of microalgal lipids for biodiesel production.

Microalgae exhibit properties that make them well suited for use in a commercial-scale biodiesel production facility. Many species exhibit rapid growth and high productivity, and many microalgal species can be induced to accumulate substantial quantities of lipids, often greater than 60% of

their biomass. Microalgae can also grow in saline waters that are not suitable for agricultural irrigation or consumption by humans or animals. The growth requirements are very simple, primarily carbon dioxide (CO2) and water, although the growth rates can be accelerated by sufficient aeration and the addition of nutrients. A brief overview of the characteristics of the major microalgal classes can be found in Section II. A.2.

A major undertaking by ASP researchers in the early stages of the program was to identify candidate microalgal species that exhibited characteristics desirable for a commercial production strain. Resource analyses carried out by SERI (discussed in Section III. C.) indicated that the desert regions of the southwestern United States were attractive areas in which to locate microalgal-based biofuel production facilities. This, in part, dictated the required strain characteristics. These characteristics included the ability of the strains to grow rapidly and have high lipid productivity when growing under high light intensity, high temperature, and in saline waters indigenous to the area in which the commercial production facility is located. In addition, because it is not possible to control the weather in the area of the ponds, the best strains should have good productivity under fluctuating light intensity, temperature, and salinity.

A multi-faceted effort was carried out to:

• isolate microalgae from a variety of saline habitats (including oceans, lakes, ponds, and various ephemeral water bodies),

• screen those isolates for the ability to grow under a variety of conditions,

• analyze the biochemical components of the strains (especially with respect to lipids), and

• determine the effects of environmental variables on the growth and lipid composition of selected strains.

This effort involved in-house researchers and subcontractors from academia, industry, and other government laboratories. Section II. A.1. documents the efforts of SERI in-house researchers in the area of microalgal strain isolation and screening. It also describes the methodologies developed and employed during the isolation, screening, and characterization phases of the work. Section II. A.2. describes parallel efforts conducted by SERI subcontractors. An account of the history and current status of the NREL Microalgae Culture Collection is presented in Section II. A.3.

Although the collection and screening efforts produced a number of viable candidate strains, no one algal strain was identified that exhibited the optimal properties of rapid growth and high constitutive lipid production. Many microalgae can be induced to accumulate lipids under conditions of nutrient deprivation. If this process could be understood, it might be possible to manipulate either the culture conditions, or to manipulate the organisms themselves, to increase lipid accumulation in a particular strain. Therefore, studies were initiated both at SERI and by

ASP subcontractors to study the biochemistry and physiology of lipid production in oleaginous (oil-producing) microalgae. Work performed by several ASP subcontractors was designed to understand the mechanism of lipid accumulation. In particular, these researchers tried to determine whether there is a specific “lipid trigger” that is induced by factors such as nitrogen (N) starvation. Subcontractors also studied ultrastructural changes induced in microalgae during lipid accumulation. They also initiated efforts to produce improved algae strains by looking for genetic variability between algal isolates, attempting to use flow cytometry to screen for naturally-occurring high lipid individuals, and exploring algal viruses as potential genetic vectors. The work performed by ASP subcontractors is described in Section II. B.1.

Although some of the efforts of the in-house SERI researchers were also directed toward understanding the lipid trigger induced by N starvation, they showed that silica (Si) depletion in diatoms also induced lipid accumulation. Unlike N, Si is not a major component of cellular molecules, therefore it was thought that the Si effect on lipid production might be less complex than the N effect, and thus easier to understand. This initiated a major research effort at SERI to understand the biochemistry and molecular biology of lipid accumulation in Si-depleted diatoms. This work led to the isolation and characterization of several enzymes involved in lipid and carbohydrate synthesis pathways, as well as the cloning of the genes that code for these enzymes. One goal was to genetically manipulate these genes in order to optimize lipid accumulation in the algae. Therefore, reseach was performed simultaneously to develop a genetic transformation system for oleaginous microalgal strains. The successful development of a method to genetically engineer diatoms was used in attempts to manipulate microalgal lipid levels by overexpressing or down-regulating key genes in the lipid or carbohydrate synthetic pathways. Unfortunately, program funding was discontinued before these experiments could be carried out beyond the prelimilary stages.

Cost-effective production of biodiesel requires not only the development of microalgal strains with optimal properties of growth and lipid production, but also an optimized pond design and a clear understanding of the available resources (land, water, power, etc.) required. Section III reviews the R&D on outdoor microalgae mass culture for production of biodiesel, as well as supporting engineering, economic and resource analyses, carried out and supported by ASP during the 1980s and early 1990s. It also covers work supported by DOE and its predecessor agency, the Energy Research and Development Administration (ERDA), during the 1970s and some recent work on utilization of CO2 from power plant flue gases.

From 1976 to 1979, researchers at the University of California-Berkeley used shallow, paddle wheel mixed, raceway-type (high-rate) ponds to demonstrate a process for the simultaneous treatment of wastewater and production of energy (specifically methane). Starting in 1980, the ASP supported outdoor microalgal cultivation projects in Hawaii and California, using fresh and seawater supplies, respectively, in conjunction with agricultural fertilizers and CO2. The two projects differed in the types of algae cultivated and the design of the mass culture system, with the project in California continuing to develop the high-rate pond design, and the Hawaii project studying an (initially) enclosed and intensively mixed system. From 1987 to 1990, an “Outdoor Test Facility” was designed, constructed and operated in Roswell, New Mexico, including two

1,000 m2 high-rate ponds. This last project represented the culmination of ASP R&D in large-scale algal mass culture R&D. These studies are described in Section III. A. Some supporting laboratory studies and development of an “Algal Pond Model” (APM) are also reviewed at the end of that section. The conclusion from these extensive outdoor mass culture studies was that the use of microalgae for the low-cost production of biodiesel is technically feasible, but still requires considerable long-term R&D to achieve the high productivities required.

Section III. B. reviews the resource assessments, for water, land, CO2, etc., carried out by the ASP, primarily for the southwestern United States. These studies demonstrated the potential availability of large brackish and saline water resources suitable for microalgae mass cultures, large land and CO2 resources. They suggest that the potential production of microalgae-derived biodiesel may represent more than 10% of U. S. transportation fuels, although full resource exploitation would be significantly constrained in practice. Several engineering and economic cost analyses were also supported by DOE and the ASP, and these are reviewed in Section III. C., including recent work by the ASP and DOE on power plant flue gas utilization for greenhouse gas (CO2) mitigation.

The overall conclusion of these studies was that in principle and practice large-scale microalgae production is not limited by design, engineering, or net energy considerations and could be economically competitive with other renewable energy sources. However, long-term R&D would be required to actually achieve the very high productivities and other assumptions made in such cost analyses. Section III. D. provides recommendations for future research that could make this technology commercially feasible.

City College of the City University of New York Jane C. Gallagher 3/86 — 12/87

ZK-4-04136-5; ZK 4-04-136-04

The purpose of these studies was to investigate the intragenetic variability (e. g., between various isolates of a single species) in microalgae with potential for high lipid production. The rationale for this work with respect to the ASP is that variability within and between species of microalgae

has implications for algal collection strategies, for strain selection for high lipid producers, and for genetic manipulation of microalgae by classical breeding or genetic engineering.

Historically, microalgae have been classified based on morphological similarities. Previous studies by Dr. Gallagher and others (see Gallagher 1986) suggested significant physiological variability between isolates of a single species. In these studies, various isolates of a species grown under identical conditions (to control for environmentally induced changes in gene expression) often showed significant differences in characteristics such as nutrient uptake, growth rates, and pigment content. These results indicated that there may be inherent genetic differences between the individual strains. The studies by Dr. Gallagher compared electrophoretic banding patterns of specific proteins to obtain quantitative estimates of the genetic differences between isolates of two genera of oil-producing microalgae.

The organisms studied were A. coffeiformis (class Bacilliarophyceae) and Nannochloropsis spp. (class Eustigmatophyceae). The basic approach was to streak the isolated algae onto agar plates, then to pick single colonies and restreak the cells to ensure that each strain was unialgal. The isolates were propagated under identical growth conditions to minimize differences caused by environmentally induced changes in gene expression. Each strain was examined using light microscopy (LM) and scanning electron microscopy (SEM) to look for morphological differences and to confirm species identity. Crude protein extracts from each strain were separated by polyacrylamide gel electrophoresis. The gel was then stained to detect several specific enzymes, including phosphoglucose isomerase, hypoxanthine dehydrogenase, a — ketoglutarate dehydrogenase, malate dehydrogenase, a-hydroxybutarate dehydrogenase, and tetrazolium oxidase. (Dr. Gallagher also tried unsuccessfully to stain for several other enzymes. Poor staining may be a consequence of the location of these enzymes within cellular membranes in microalgae.) An extract from the diatom Skeletonema costatum (clone SKEL) was run on each gel to serve as an internal standard, and the migration pattern for each enzyme was reported as the ratio of the migration distance for the unknown Amphora or Nannochloropsis enzyme to the migration of the known enzyme from Skeletonema. An example of this type of experiment is shown in Figure III. B. 1. This method allowed for detection of very small differences in the migration patterns of the various forms of the enzymes. These differences could result from subtle variations in protein charge or conformation due to one or several amino acid changes. Isolates that showed two bands for a specific enzyme were assumed to be heterozygous at that allele.

For the studies of Amphora, Dr. Gallagher obtained 47 isolates, 32 of which were isolated from a salt marsh in Woods Hole, Massachusetts, on the same day in August 1985. Another six strains had been isolated from the same site during the summers of 1979 or 1980 and maintained in culture, and five strains were obtained from laboratory cultures maintained by other investigators. It is unclear from Dr. Gallagher’s reports how many of the 47 Amphora isolates were tested as described earlier.

All strains that were subsequently analyzed for enzyme banding patterns were first examined by LM and SEM. Microscopy confirmed that all the strains were A. coffeiformis, although some

variation was observed in the morphology of the frustule between strains, for example, in the presence or absence of costae (rib-like protrusions) or in the shape or number of punctae (holes). These changes were assumed to be due to genetic differences between the strains, as unialgal clones maintained in culture for 6-7 years did not show variations in frustule morphology between individuals. Genetic similarity was calculated based on the electrophoretic banding patterns using the statistical methods of Nei (1972). The zymograms indicated significant variation between isolates of the same species, even between strains isolated from the same site on the same day. These differences were not correlated with the extent of morphological variation, and some morphologically identical strains showed differences in the protein banding patterns.

For Nannochloropsis, 115 strains were obtained, all from culture collections. The electrophoretic banding patterns also indicated significant genetic diversity between strains, even between samples isolated from the same location. However, the zymogram data for Nannochloropsis was limited due to the high percentage of “null” alleles (no staining of some enzymes) in some isolates. It was unclear whether this was caused by undetermined genetic differences between the isolates (and between Amphora and Nannochloropsis), or due to difficulties in extraction of the proteins from Nannochloropsis. More data would be needed to fully analyze the genetic differences between isolates of this genus.

What are the implications of this research for the Aquatic Species Program? The significant amount of genetic diversity between individuals of a species, even when isolated from very similar locations, suggests that researchers involved in collecting microalgal strains as potential lipid producers should obtain more than one isolate from each site. In fact, these results suggest that it may be adequate to sample fewer sites to obtain a sufficiently varied collection of microalgal strains.

In a previous study (discussed in Gallagher 1985), Dr. Gallagher described experiments performed on isolates of the diatom S. costatum similar to those described here. The data suggested significantly less genetic variation between isolates of Skeletonema, even between strains isolated from different locations, than was seen between Amphora strains isolated from the same environment. This difference was attributed to the fact that Amphora is an attached, benthic organism that produces amoeboid gametes, whereas Skeletonema is planktonic, and produces swimming sperm. These “lifestyle” differences would result is lower potential for gene flow between Amphora populations, although the presence of heterozygotes indicates interbreeding among Amphora at localized sites. These observations suggest that benthic organisms may have greater genetic diversity than planktonic forms.

Based on the data in this study, Dr. Gallagher also concluded that breeding or genetic engineering of microalgae may be more successful using morphologically similar phenotypes, as her results suggest less diversity at the protein level between identical morphotypes. However, genetic engineering research during the past 15 years in other organisms indicates that cells can often express genes from very different species, so these differences between strains probably will not affect the expression of genes transferred between these similar organisms.

While working under the SERI subcontract, Dr. Gallagher also participated in a study that provided evidence that the carotenoid violaxanthin functions as a major light harvesting pigment in Nannochloropsis (Owens et al. 1987). Carotenoids generally are considered accessory pigments in photosynthetic organisms, involved primarily in photoprotection, fluorescence quenching, and light harvesting. Nannochloropsis is a member of the class Eustigmatophyceae, which are unusual in that they can contain violaxanthin as up to 60% of their total pigments. These authors used room temperature fluorescence excitation and emission data to provide the first evidence that violaxanthin can function in photosynthetic light harvesting.

Understanding the fundamental processes involved in microalgal photosynthesis is important to the ASP since light-driven photosynthesis results in the production of chemical reductants that drive the synthetic dark reactions; lipids are storage products that can be produced from excess photosynthate. One possible implication is that carotenoids absorb at different wavelengths than chlorophyll, absorbing green light that penetrates into the water column. This feature could be beneficial for mass culture of organisms, allowing denser cultures to grow in a deep raceway.

I Publications:

Gallagher, J. C. (1986). “Population genetics of microalgae.” In Algal Biomass Technologies: An Interdisciplinary Perspective (Barclay, W.; McIntosh, R., eds.), Beihefte zur Nova Hedwigia, Heft 83, Gebruder Borntraeger, Berlin-Stugart, pp. 6-14.

Gallagher, J. C. (1987a). “Patterns of genetic diversity in three genera of oil-producing microalgae” (abstr.), FY1987Aquatic Species Program Annual Report, (Johnson, D. A.; Sprague, S., eds.), Solar Energy Research Institute, Golden, Colorado, SERI/SP-231-3206, p. 209.

Gallagher, J. C. (1987b). “Genetic variation in oil-producing microalgae.” FY 1986 Aquatic Species Program Annual Report, (Johnson, D. A., ed.), Solar Energy Research Institute, Golden, Colorado, SERI/SP-231-3071, p.331-336.

Owens, T. G.; Gallagher, J. C.; Alberte, R. S. (1987) “Photosynthetic light-harvesting function of violaxanthin in Nannochloropsis spp. (Eustigmatophyceae).” J. Phycol. 23:79-85.

I Additional References:

Nei, M. (1972) Amer. Natur. 106:283.

By 1987, SERI researchers and subcontractors had collected approximately 3,000 algal strains. Most of these strains had not been well characterized, especially with respect to lipid production capabilities. As a consequence, work commenced on the development of a simple screening procedure to estimate the lipid contents of cells to determine which strains had the best potential as biofuel production organisms. Ideally, the procedure should be simple and reproducible so that it could be used as a standard method in numerous laboratories. The researchers hoped that such a screening tool would allow the size of the strain collection to be reduced to a manageable number (~200) representing the most promising strains.

Development of a rapid screen for lipid content.

In an attempt to develop a reproducible, easy-to-use screening procedure to identify algal strains with high lipid contents, Dr. Keith Cooksey (an ASP subcontractor at Montana State University) suggested that investigators explore the possibility of using the lipophilic dye Nile Red (9- diethylamino-5H-benzo{a}phenoxazine-5-one) to stain cells. Nile Red was first isolated from Nile Blue by Greenspan et al. (1985), who showed that Nile Red will fluoresce in a nonpolar environment and could serve as a probe to detect nonpolar lipids in cells. Nile Red permeates all structures within a cell, but the characteristic yellow fluorescence (approximately 575 nm) only occurs when the dye is in a nonpolar environment, primarily neutral storage lipid droplets. Earlier work within the ASP by Dr. Steve Lien had shown the utility of Nile Blue in microscopically assessing the lipid content of algal cells (Lien 1981). The active ingredient in these Nile Blue preparations may in fact have been Nile Red. Parallel efforts to develop a Nile Red staining procedure were carried out by SERI researchers and ASP subcontractors, notably Drs. Cooksey and Sommerfeld.

Cooksey et al. (1987) used the diatom Amphora coffeiformis to optimize the Nile Red staining procedure. The dye was dissolved in acetone and used at a concentration of 1 mg/mL of cell suspension. In this species, the fluorescence of the dye in live stained cells was stable for only 2­7 minutes; fluorescence measurements had to be completed rapidly to ensure consistent results. The kinetics of fluorescence in stained cells varied in different species, presumably due to differences in the permeability of cell walls to the stain, and differences in how the lipid is stored in the cells, i. e., as large or small droplets. Fixing the stained cells with formaldehyde or ethanol preserved the Nile Red fluorescence for 2 hours, but cells that were chemically fixed before Nile Red staining did not exhibit the characteristic yellow fluorescence. When Nile Red fluorescence was measured in algal cultures over time, the fluorescence increased as the culture became N deficient. The fluorescence level was linearly correlated with an increase in the total lipid content, determined gravimetrically, in a growing culture of algal cells. Fractionation of the lipids by silicic acid column chromatography demonstrated that the increase in lipid was due primarily to an increase in neutral lipids rather than in the polar lipids or glycolipids, which are found primarily in cell membranes.

Additional development of the Nile Red screening procedure occurred at SERI and at Milt Sommerfeld’s laboratory at Arizona State University. The resultant protocol involved taking a fixed volume of a diluted algal culture (typically 4 mL), adding 0.04 mL of a Nile Red solution (0.1 mg/mL in acetone), and determining the fluorescence after 5 min using a fluorometer equipped with the appropriate excitation and emission filters.

Although the use of Nile Red allowed various microalgae to be rapidly screened for neutral lipid accumulation, interspecies comparisons may be subject to misinterpretation because of the species-specific staining differences described earlier. Nonetheless, before Nile Red was used, quantitating lipids from cells was very time consuming. It required the extraction of lipid from a large number of cells using organic solvents, evaporation of the solvent, and determination of the

amount of lipid by weighing the dried extract. Consequently, the use of Nile Red as a rapid screening procedure can still have substantial value.

Screening for growth in high conductivity media.

The estimation of lipid content using a simple procedure such as the Nile Red assay is clearly an important component of a rapid screening procedure for identifying promising strains, but an equally important component is a means to identify strains that grow rapidly under the expected culture conditions. Reports detailing the amount and types of saline groundwater available in the southwestern United States, along with data concerning the high rates of evaporation in this region, indicated that tolerance of algal strains to high conductivity (higher than 50 mmho^cm-1) could be important. Therefore, an additional component of the secondary screening procedure developed to reduce the number of strains being maintained by ASP researchers. Algae were tested for the ability to grow at high conductivity (55 mmhccm"1, both Type I and Type II media), high temperature (30°C), and high light intensity (average of 200 pE^m-2^s-1, 12 h light: 12 h dark cycle) in cultures that were continually agitated via aeration. To prevent osmotic shock, strains were adapted to higher conductivities via a stepwise transfer into media with increasing conductivity at 2-day intervals. Tubes were used that could be placed directly in a spectrophotometer (i. e., 25 mm diameter, 50 mL volume), allowing the culture density to be measured without removing a sample. The tubes also held enough medium to allow samples to be taken for Nile Red lipid analysis (both for N-sufficient and N-deficient cells), and for ash-free dry mass determinations. The tubes were placed in a rack and illuminated by fluorescent lamps from below for the screening procedure. Optical density measurements were taken twice daily for 4 days during exponential growth to determine growth rates. Samples were removed for Nile Red fluorometric analysis during exponential growth and after 2 days (Arizona State University) or 4 days (SERI) of N deficiency.

This newly developed rapid screening protocol was subsequently used both in Milt Sommerfeld’s laboratory and at SERI to screen many microalgal isolates. Keith Cooksey’s laboratory also examined numerous strains using this procedure. Sommerfeld’s laboratory examined approximately 800 strains that had been collected over the previous 2 years of the subcontract. Only 102 of these strains survived transfer into media having a conductivity of 55 mmho^cm-1. Of these strains, 40 grew in both Type I/55 and Type II/55 media, 42 grew only in Type I/55 medium, and 19 grew only in Type II/55 medium. The 10 fastest-growing strains, along with their preferred media, are shown in Table II. A.2.

The lipid production potential of these strains was evaluated by the use of the fluorometric Nile Red assay. In Type I/55 medium, 49 strains had a higher apparent lipid content after 2 days of N-deficient growth, whereas 13 strains had the same or lower apparent lipid levels in response to N deficiency. In Type II/55 medium, 42 strains had higher lipid levels, and 7 strains had lower or unchanged lipid levels as a consequence of N deficiency. Of note was that the mean lipid level in cells grown in Type II/55 medium was nearly twice that of Type I/55-grown cells.

The strains exhibiting the highest Nile Red fluorescence levels are shown in Table II. A.3. All of these strains are diatoms, confirming the propensity of this group to accumulate lipids.

Table II. A.3. Strains from the Arizona State University collection having the highest Nile Red fluorescence.

These researchers also ranked strains according to the estimated lipid productivity of rapidly growing cells, based on the calculated growth rates and estimated lipid contents of exponential phase cells. The top strains resulting from this analysis are shown in T able II. A.4. However, the optimal strategy for maximizing lipid yield in actual mass culture facilities may require an “induction” step (i. e., manipulation of the culture environment, possibly involving nutrient deficiency). The ranking of strains would obviously be very different in that case.

Table II. A.4. Strains in the Arizona State University collection with the highest apparent lipid productivity during exponential growth, based on Nile Red staining.

SERI researchers also started to evaluate various strains by the rapid screening procedure. Initial work focused on 25 partially characterized strains. These strains were analyzed for growth and Nile Red fluorescence in exponentially growing cultures and in cultures grown under N-deficient conditions for 4 days. The results of the SERI and Sommerfeld laboratories cannot be compared directly, because Nile Red units are expressed differently and the time duration of N deficiency was not the same. The best strains of the 25 tested (based on the highest Nile Red fluorescence normalized to ash-free dry weight (AFDW) and rapid exponential growth) were determined to be CHAET9 (muelleri), NAVIC2 (Navicula saprophila), and NITZS12 (Nitzschia pusilla).

Twenty-eight strains of Chaetoceros were also examined using this screening protocol. The best strains indentified were CHAET21, CHAET22, CHAET23, and CHAET25 (all muelleri). All but the latter strain were isolated from various regions of the Great Salt Lake in Utah.

The departure of Dr. Bill Barclay and Dr. Jeff Johansen from the ASP, along with a greater emphasis on genetic improvement of strains, marked the end of the in-house collection and screening work. As a consequence, many of the 3,000 strains collected by ASP researchers during the course of this research effort were never analyzed via this rapid screening protocol. Nonetheless, enough strains had been analyzed at SERI and at the laboratories of various subcontractors to obtain a substantial number of promising strains. The next step was to determine their ability to grow in actual outdoor mass culture ponds. This work is described in Section III of this report.

The goal of this research was to characterize the organellar genomes of Chlorella and other microalgae. As organellar DNA is thought to be highly conserved evolutionarily, the idea was to use similarities or differences between chloroplast or mitochondrial DNA as a measure of the taxomonic relatedness of algal strains. This information could be useful for experiments involving somatic cell fusion or gene transfer, as these procedures would likely have a higher chance of success between more closely related strains. Studies of the organellar genomes could also lead to the identification of promoters or replication origins that could be used to develop vectors for algal transformation. Due to the lack of significant progress on the first three goals, Dr. Meints concentrated the efforts of his laboratory on this project for the last 2 years of the subcontract.

The first step was to develop methods for isolation of chloroplast and mitochondrial DNA from Chlorella N1a. Based on protocols used for higher plants, Dr. Meints exploited the differences in the C/G content between chloroplast DNA and nuclear DNA to separate the two genomes using density gradient centrifugation. Chloroplast DNA was identified by hybridization with heterologous chloroplast DNA markers. The chloroplast genome of Chlorella N1a was found to be circular, containing approximately 120 kbp of DNA. A restriction map of the chloroplast genome was produced and several genes were localized on the map by hybridization with chloroplast gene sequences from maize. Most chloroplast genomes contain two inverted repeats, each of which contains a copy of the 23S, 16S, and 5S ribosomal RNA genes. These repeats are flanked by a short and long single copy DNA region. Although Dr. Meints initially reported that Chlorella N1a chloroplast DNA contained this inverted repeat structure (Meints 1987), a subsequent article reported that the chloroplast genome of Chlorella N1a contains only a single copy of the ribosomal RNA gene region (Schuster et al. 1990b). This result was confirmed by Dr. Meints via a recent personal communication. Although most other green algae, including other chlorellans, contain chloroplast DNA similar to that commonly seen in most higher plants, i. e., containing two inverted repeats, this unusual chloroplast structure has been seen in two legumes (peas, broad beans), conifers, some red algae, and in at least one other green alga, Codium.

Restriction analysis of chloroplast DNA from several exsymbiont and free-living strains of Chlorella showed variations between the strains that indicate genetic divergence and that suggest gene transfer and cell fusion between these species may be problematic. The results suggest that chloroplast DNA structure may be a useful taxonomic parameter, but more study is needed before definite conclusions on algal taxonomy or cell-cell compatability based on chloroplast DNA structure can be made.

Isolating mitochondrial DNA from Chlorella N1a was technically problematic, and the mitochondrial genome isolated was first presumed to be a plasmid. Unlike some species in which mitochondrial DNA has a G/C content similar to that of nuclear DNA, in Chlorella N1a, the mitochondrial DNA had a low G/C content similar to that of chloroplast DNA, and the two genomes banded very closely on the density gradients. As with the chloroplast DNA,

heterologous probes were used to identify the mitochondrial DNA and to localize specific mitochondrial genes on the restriction map. The gene organization in the Chlorella N1a mitochondrial DNA was similar to that in higher plants, and distinct from the organization of mitochondrial genes in animals and fungi. It has been proposed that mitochondria in plants and green algae originated from a separate endosymbiotic event as compared to animals and fungi. This is supported by Dr. Meints’ results.

Although not included in the original Statement of Work, Dr. Meints also reported under this task other related research efforts in his laboratory toward the development of a genetic transformation system for microalgae. Libraries were prepared from Chlorella N1a nuclear DNA and DNA from the algal virus. The goal of this project was to identify DNA sequences that could be used to develop transforming vectors, such as origins of replication, regulatory regions for gene expression, or algal genes to use in selectable marker systems. A library of the viral DNA was prepared in a lambda vector, which allowed for the sequencing of the viral genome and studies of viral gene stucture and expression. This work led to several significant discoveries that were published after SERI funding stopped, including the cloning of the major viral capsid protein (Graves and Meints, 1992), and the identification of a viral gene promoter that also functioned in higher plants (Mitra and Higgins, 1994).

Dr. Meints’ laboratory also made several attempts to produce a library of Chlorella nuclear DNA, with little success. This appeared to be due to modification (probably methylation) of the algal DNA that resulted in degradation of the DNA by the bacterial host used for library construction. Several ways around this problem were proposed, including the use of a yeast cloning system or the use of a bacterial host that did not contain the enzymes for degradation of methylated DNA. A cDNA library was produced successfully before the end of the SERI-funded research efforts.

Dr. Meints and his coworkers and collaborators produced a large quantity of data during the 4 years of SERI-funded research and during the following years. They made significant contributions to the study of the biology, biochemistry, and molecular biology of a eukaryotic algal virus, and to the biology and molecular biology of the algal hosts, particularly with respect to the algal organellar genomes. Unfortunately, because of the specificity of the virus/algal interactions, the results obtained were not directly applicable to the development of a transformation system for the oleaginous algal strains of interest to NREL. The research also generated some valuable technical information, regarding toxicity of microalgae to common osmotica, construction of genomic DNA libraries, and organellar genome isolation, which could be useful for further studies of algal molecular biology and the development of genetic engineering techniques. The studies of the algal virus also resulted in the identification of a new restriction endonuclease (Jin et al. 1994) and a new adenine methyltransferase (Stefan et al. 1991), as well as a viral promoter sequence that can function in plants (Mitra and Higgens 1994).

Like many good ideas (and certainly many of the concepts that are now the basis for renewable energy technology), the concept of using microalgae as a source of fuel is older than most people realize. The idea of producing methane gas from algae was proposed in the early 1950s1. These early researchers visualized a process in which wastewater could be used as a medium and source of nutrients for algae production. The concept found a new life with the energy crisis of the 1970s. DOE and its predecessors funded work on this combined process for wastewater treatment and energy production during the 1970s. This approach had the benefit of serving multiple needs—both environmental and energy-related. It was seen as a way of introducing this alternative energy source in a near-term timeframe.

In the 1980s, DOE’s program gradually shifted its focus to technologies that could have large-scale impacts on national consumption of fossil energy. Much of DOE’s publications from this period reflect a philosophy of energy research that might, somewhat pejoratively, be called “the quads mentality.” A quad is a short-hand name for the unit of energy often used by DOE to describe the amounts of energy that a given technology might be able to displace. Quad is short for “quadrillion Btus”—a unit of energy representing 1015 (1,000,000,000,000,000) Btus of energy. This perspective led DOE to focus on the concept of immense algae farms.

Such algae farms would be based on the use of open, shallow ponds in which some source of waste CO2 could be efficiently bubbled into the ponds and captured by the algae (see the figure below).

The ponds are “raceway” designs, in which the algae, water and nutrients circulate around a racetrack. Paddlewheels provide the flow. The algae are thus kept suspended in water. Algae are circulated back up to the surface on a regular frequency. The ponds are kept shallow because of the need to keep the algae exposed to sunlight and the limited depth to which sunlight can penetrate the pond water. The ponds are operated continuously; that is, water and nutrients are constantly fed to the pond, while algae-containing water is removed at the other end. Some kind of harvesting system is required to recover the algae, which contains substantial amounts of natural oil.

The concept of an “algae farm” is illustrated on the next page. The size of these ponds is measured in terms of surface area (as opposed to volume), since surface area is so critical to capturing sunlight. Their productivity is measured in terms of biomass produced per day per unit of available surface area. Even at levels of productivity that would stretch the limits of an aggressive research and development program, such systems will require acres of land. At such large sizes, it is more appropriate to think of these operations on the scale of a farm.

There are quite a number of sources of waste CO2. Every operation that involves combustion of fuel for energy is a potential source. The program targeted coal and other fossil fuel-fired power plants as the main sources of CO2. Typical coal-fired power plants emit flue gas from their stacks containing up to 13% CO2. This high concentration of CO2 enhances transfer and uptake of CO2 in the ponds. The concept of coupling a coal-fired power plant with an algae farm provides an elegant approach to recycle of the CO2 from coal combustion into a useable liquid fuel.

Other system designs are possible. The Japanese, French and German governments have invested significant R&D dollars on novel closed bioreactor designs for algae production. The main advantage of such closed systems is that they are not as subject to contamination with whatever organism happens to be carried in the wind. The Japanese have, for example, developed optical fiber-based reactor systems that could dramatically reduce the amount of surface area required for algae production. While breakthroughs in these types of systems may well occur, their costs are, for now, prohibitive—especially for production of fuels. DOE’s program focused primarily on open pond raceway systems because of their relative low cost.

The Aquatic Species Program envisioned vast arrays of algae ponds covering acres of land analogous to traditional farming. Such large farms would be located adjacent to power plants. The bubbling of flue gas from a power plant into these ponds provides a system for recycling of waste CO2 from the burning offossil fuels.

Scripps Institution of Oceanography Ralph A. Lewin 1985 — 1986 N/A

This subcontract focused on the collection and characterization of picopleustonic algae, which are defined as algae (including the prokaryotic cyanophytes) that are very small (1-5 pm) and that live on the surface of the water. In February 1985, water samples were taken from various sites in the Caribbean Sea, including sites near the U. S. Virgin Islands (St. John, St. Thomas, and St. Croix), Tortola, Puerto Rico, Curasao, Panama, and the Florida Keys. 130 samples (250 mL each) were collected and filtered through a 3-8-pm filter to remove larger cells. Smaller cells were collected on a 0.45 pm nitrocellulose filter, which was rolled up and placed in the original sampling water that had passed through the filter. These samples were placed under natural lighting at 20°C to 25°C until transported to the laboratory. The filters were then transferred into a tube of sterile enriched seawater (containing additional N and other nutrients) and incubated at 25°C under continuous illumination from a fluorescent lamp at 30 pE^m-2^s-1. In an attempt to stimulate lipid accumulation via nutrient deficiency, a portion of each culture was transferred after 4 weeks of growth to a fresh tube of unenriched seawater and then allowed to grow under the same conditions. After 4 more weeks, a film of cells was often observed floating on the surface of the cultures. Small samples of these cells were transferred to fresh enriched seawater. After incubation for an additional 2 weeks, the cells in these cultures were microscopically examined, and cultures that were dominated by diatoms, cyanophytes, and flagellates were

discarded, leaving approximately 60 cultures of small (1-5-pm) green cells. Unialgal cultures were established from these cells by isolating colonies on agar plates. Of these purified cultures, there were 14 isolates of Stichococcus, 21 isolates ofNannochloris, four strains of Chlorella, and several representatives of other genera. Stichococcus, Nannochloris, and Chlorella are all chlorophytes. Because cyanophytes typically do not accumulate lipids, they were eliminated from further study in this subcontract. The researchers anticipated that isolating strains in this manner would enrich for lipid-accumulating microalgae.

The isolated strains were tested for the ability to grow in freshwater; all the Stichococcus and Chlorella strains grew well in freshwater, suggesting perhaps a brackish water origin for these strains. Only six of the 21 Nannochloris strains could grow on non-marine media.

To quantitatively determine lipid content in the isolated strains, 1-L cultures were grown for 3 weeks in enriched seawater under continuous illumination at 30 pE^m-2^s-1. The cultures were bubbled with 0.5% CO2 in air. Cells were harvested, frozen and lyophilized, and then extracted three times with a chloroform/methanol mixture (2:1 v/v). After the solvents evaporated, the lipid mass was determined gravimetrically and normalized to the cellular AFDW. The 13 Stichococcus strains had lipid contents ranging from 9% to 59% of the AFDW, with an average of 33%. The lipid contents of the 21 Nannochloris strains ranged from 6% to 63%, with an average of 31%. Data were not presented for the lipid contents of the four Chlorella strains, although three strains of the eustigmatophyte Nannochloropsis that were isolated from Qingdao, China were examined; for this genus, the lipid content ranged from 31% to 68%, with an average of 46%. These reported lipid contents may be slight overestimates, in that there was apparently no attempt to remove somewhat polar materials that may have also been extracted via the use of an aqueous washing step.

Some preliminary experiments were also conducted during the course of this subcontract regarding the growth of the eustigmatophyte Nannochloropsis (strain Nanno-Q, one of the Qingdao strains). This strain is euryhaline, and is able to grow in seawater as well as brackish water with one-tenth the salinity of seawater. The cells grew to a higher final yield with nitrate as the N source than with ammonium, and the lipid content rose substantially when the N source was initially supplied at levels below 200 pM (as determined by the percentage of cells that were floating due to elevated lipid levels). A number of Nannochloropsis strains that had been obtained primarily from the Culture Collection of Marine Phytoplankton at Bigelow Laboratory (West Boothbay Harbor, Maine) were analyzed with respect to maximum cell yields after 4 weeks of growth at different temperatures. Most of the strains had temperature optima at or below 25°C, although one strain that had been collected near Long Island, New York had a temperature optimum of 33°C.

This subcontract examined a group of microalgae that had not received much attention in the ASP until that point. The small size of picopleustonic algae could hinder harvesting efficiency in a mass culture facility, which would have a negative impact on the economics of biodiesel production. However, if the cells could be made to consistently float due to high lipid levels, this

property might facilitate harvesting. Outdoor testing of the most promising strains would help to evaluate this group of microalgae.

.

The in-house collection effort was focused on collecting strains from inland saline habitats, particularly in Colorado, New Mexico, and Utah. The reasoning behind collecting strains from these habitats was that the strains would be adapted to at least some of the environmental conditions in mass culture facilities in the southwestern United States (i. e., high light intensity and high temperatures). They would also be well suited for growth in the saline waters available for use in such facilities. In addition, many of the aquatic habitats in this region are shallow, and therefore subject to large variations in temperature and salinity; thus, the strains collected in this region might be expected to better withstand the fluctuations that would occur in a commercial production pond. Cyanobacteria, chrysophytes, and diatoms often dominate inland saline habitats. The latter were of particular interest to the program because of their propensity to accumulate lipids. There had never been a large-scale effort to collect strains with this combination of characteristics; therefore, they were not available from culture collections.

The stated objectives of the SERI culture collection and screening effort were to:

• Assemble and maintain a set of viable mono-specific algal cultures stored under conditions best suited to the maintenance of their original physiological and biochemical characteristics.

• Develop storage techniques that will help maintain the genetic variability and physiological adaptability of the species.

• Collect single species cultures of microalgae from the arid regions of Colorado,

Utah, and New Mexico for product and performance screening.

• Develop media which are suitable for their growth.

• Evaluate each species for its temperature and salinity tolerances, and quantify growth rates and proximate chemical composition for each species over the range of tolerated conditions.

Each objective was met during the course of research within the ASP. The following pages describe in detail the major findings of the work conducted by SERI researchers.

All bands are graphed as a ratio of the distance travelled by protein bands in Amphora to the distance travelled by stand bands in Skeletonema costatum. PGI: phosphoglucose isomerase; XDH: hypoxanthine dehydrogenase; TO: tetrazolium oxidase; ADH: analine dehydrogenase. (Source: Gallagher 1987).