OCCURRENCE AND ABUNDANCE OF GREEN-FLUORESCING DINOFLAGELLATES IN SURFACE WATERS OF THE NORTHWEST ATLANTIC AND NORTHEAST PACIFIC OCEANS1

We have cultured green fluorescing heterotrophic dinoflagellates whose continuous green fluorescence is due to an unidentified compound, probably a flavin, that excites with blue (～460 nm) light and emits green (～535 nm) light. No evidence of bioluminescence was found, but we note that compounds with similar fluorescence characteristics have been associated with bioluminescence in other taxa. These cells, all naked gymnodinoids, are widespread and abundant in the Northwest Atlantic and Northeast Pacific Oceans (10³–10⁵ L^?1). They comprise 4–100% of the total heterotrophic dinoflagellate component which, in turn, is usually equivalent magnitude to the phototrophic naked dinoflagellate component of the phytoplankton community.     

Bioluminescence of colonial radiolaria in the western Sargasso Sea

Colonial radiolaria (Protozoa: Spumellarida) were a conspicuous feature in surface waters of the Sargasso Sea during the April (1985) Biowatt cruise. The abundance of colonies at the sea surface at one station was estimated to be 23 colonies · m⁻².Bioluminescence by colonial radiolaria, representing at least six taxa, was readily evoked by mechanical stimuli and measured by fast spectroscopy and photon-counting techniques. Light emission was deep blue in color (peak emissions between 443 and 456 nm) and spectral distributions were broad (average half bandwidth of 80 nm). Single flashes were 1–2 s in duration at ≈23 °C, with species-dependent kinetics which were not attributed to differences in colony morphology, since colonies similar in appearance could belong to different species (even families) and display different flash kinetics. Although the presence of dinoflagellate symbionts was confirmed by the presence of dinoflagellate marker pigments in the colonies, luminescence in the radiolaria examined most likely did not originate from symbiotic dinoflagellates because of (1) differences in the emission spectra, (2) unresponsiveness to low pH stimulation, (3) differences in flash kinetics and photon emission of light emission, and (4) lack of light inhibition.The quantal content of single flashes averaged 1 × 10⁹ photons flash⁻¹, and colonies were capable of prolonged light emission. The mean value of bioluminescence potential based on measurements of total mechanically stimulated bioluminescence was 1.2 × 10¹¹ photons · colony⁻¹. It is estimated that colonial radiolaria are capable of producing ≈2.8 × 10¹² photons · m⁻² of sea surface. However, this represented only 0.5% of in situ measured bioluminescence potential.     

Behavioral responses of the coastal copepod Acartia hudsonica (Pinhey) to simulated dinoflagellate bioluminescence

Light pulses were used to mimic dinoflagellate bioluminescence and test its effects on the swimming behavior of Acartia hudsonica (Pinhey). The horizontal swimming patterns of the copepod were tracked and described using a video-computer system. Single flashes of light of 60 ms duration, with a wavelength of peak emission of 475 nm and an intensity of 2 μE · m^?2 · s^?1 caused a “startle” response consisting of a short burst of high speed swimming. A series of these flashes repeated every 5 s resulted in higher average swimming speed, more swimming speed bursts, and straighter paths. These behavioral changes are similar to those previously found for A. hudsonica in the presence of bioluminescent dinoflagellates. The effects of altering the intensity, duration, and color of the simulated dinoflagellate flash were also tested. Our results support the hypothesis that dinoflagellate bioluminescence is a highly evolved adaptation for repelling nocturnal grazers.     

THE ROLE OF CALCIUM IN FLOW‐STIMULATED BIOLUMINESCENCE OF THE RED TIDE DINOFLAGELLATE LINGULODINIUM POLYEDRUM

Many marine planktonic dinoflagellates emit flashes of light in response to either laminar or turbulent flows as well as direct mechanical stimulation. The production of a flash of light is known to be mediated by a proton‐mediated action potential across the vacuolar membrane; the mechanotransduction process initiating this action potential is unknown. Here we report on an investigation into the role of Ca⁺² in the mechanotransduction process regulating bioluminescence in the red tide dinoflagellate Lingulodinium polyedrum. Calcium ionophores and low concentrations of the membrane‐disrupting agent digitonin stimulated bioluminescence only when calcium was present in the media or added with the agent, indicating that the flash‐triggering vacuolar action potential is specifically stimulated by a calcium influx. A variety of known calcium channel blockers or antagonists inhibited mechanically stimulated bioluminescence but did not affect cellular bioluminescent capacity. In many cases the inhibitory affect occurred after only a brief exposure. In addition, gadolinium (Gd⁺³), a blocker of many stretch‐activated ion channels, caused potent inhibition of mechanically stimulated bioluminescence. The order of potency of the transition metals tested was La⁺³ > Gd⁺³ > Co⁺² > Mn⁺² > Ni⁺², similar to their potency as blockers of known calcium channels. Experiments with a quantified shear flow demonstrated that flow‐stimulated bioluminescence depended on the level of extracellular calcium. Future work will elucidate the signaling pathway involving calcium‐mediated flow‐stimulated mechanotransduction. Our goal is to use bioluminescence as a proxy for the initial cellular mechanotransduction events triggered by fluid flow.     

Pharmacological investigation of the bioluminescence signaling pathway of the dinoflagellate Lingulodinium polyedrum: evidence for the role of stretch‐activated ion channels

Dinoflagellate bioluminescence serves as a whole‐cell reporter of mechanical stress, which activates a signaling pathway that appears to involve the opening of voltage‐sensitive ion channels and release of calcium from intracellular stores. However, little else is known about the initial signaling events that facilitate the transduction of mechanical stimuli. In the present study using the red tide dinoflagellate Lingulodinium polyedrum (Stein) Dodge, two forms of dinoflagellate bioluminescence, mechanically stimulated and spontaneous flashes, were used as reporter systems to pharmacological treatments that targeted various predicted signaling events at the plasma membrane level of the signaling pathway. Pretreatment with 200 μM Gadolinium III (Gd³⁺), a nonspecific blocker of stretch‐activated and some voltage‐gated ion channels, resulted in strong inhibition of both forms of bioluminescence. Pretreatment with 50 μM nifedipine, an inhibitor of L‐type voltage‐gated Ca²⁺ channels that inhibits mechanically stimulated bioluminescence, did not inhibit spontaneous bioluminescence. Treatment with 1 mM benzyl alcohol, a membrane fluidizer, was very effective in stimulating bioluminescence. Benzyl alcohol‐stimulated bioluminescence was inhibited by Gd³⁺ but not by nifedipine, suggesting that its role is through stretch activation via a change in plasma membrane fluidity. These results are consistent with the presence of stretch‐activated and voltage‐gated ion channels in the bioluminescence mechanotransduction signaling pathway, with spontaneous flashing associated with a stretch‐activated component at the plasma membrane.     

PHOTOINHIBITION OF MECHANICALLY STIMULABLE BIOLUMINESCENCE IN THE HETEROTROPHIC DINOFLAGELLATE PROTOPERIDINIUM DEPRESSUM (PYRROPHYTA)1

Photoinhibition of mechanically stimulable bioluminescence (MSL) in the heterotrophic dinoflagellate Protoperidinium depressum Bailey was investigated using samples collected from the Massachusetts and southern Texas coasts. The times for both photoinhibition of MSL (ca. 10 min) and dark recovery from photoinhibition of MSL (ca. 45 min) in this species were similar to those reported for autotrophic dinoflagellates. The degree of photoinhibition of MSL was a linear function of the logarithm of photon flux density (PFD). The threshold PFDs for the photoinhibition of MSL were 0.02, 0.6, and 21 μmol photons · m^?2· s^?1 for broad-band blue, green, and red light, respectively. These PFDs are lower than those required for photoinhibition of MSL by the autotrophic dinoflagellates Pyrocystis lunula and Ceratium fusus. We speculate that photosynthetic pigments in autotrophic dinoflagellates shield the photoreceptor that causes photoinhibition of MSL, thus lowering the sensitivity of these dinoflagellates to light. When field-collected P. depressum were kept in the laboratory without growth for a week, photoinhibition of MSL's sensitivity to light increased progressively along with 1) a decrease in its bioluminescence capacity (BCAP), 2) a decrease in the ratio of MSL to BCAP (MSL/BCAP), and 3) a decrease in the orange pigmentation (probably carotenoid) of the dinoflagellate. The action spectrum for photoinhibition of MSL in P. depressum was characterized primarily with a broad peak in the blue extending into the green. We suggest that carotenoid was not a photoreceptor for the photoinhibition of MSL in P. depressum because the peak of the action spectrum was too broad and extended too far into the green part of the spectrum, and because the orange pigment present decreased as photoinhibition of MSL became more sensitive to light.     

SPONTANEOUS AND STIMULATED BIOLUMINESCENCE OF THE DINOFLAGELLATE CERATOCORYS HORRZDA (PERIDINIALES)1

This is the first report of spontaneous bioluminescence in the autotrophic dinoflagellate Ceratocorys horrida von Stein. Bioluminescence was measured, using an automated data acquisition system, in a strain of cultured cells isolated from the Sargasso Sea. Ceratocorys horrida is only the second dinoflagellate species to exhibit rhythmicity in the rate of spontaneous flashing, flash quantum flux (intensity), and level of spontaneous glowing. The rate of spontaneous flashing was maximal during hours 2–4 of the dark phase [i.e. circadian time (CT)16–18 for a 14:10 h LD cycle (LD14:10)], with approximately 2% of the population flashing-min^?1, a rate approximately one order of magnitude greater than that of the dinoflagellate Gonyaulax polyedra. Flash quantum flux was also maximal during this period. Spontaneous flashes were 134 ms in duration with a maximum flux (intensity) of 3.1×10⁹ quanta-s^?1. Light emission presumably originated from blue fluorescent microsources distributed in the cell periphery and not from the spines. Values of both spontaneous flash rate and maximum flux were independent of cell concentration. Isolated cells also produced spontaneous flashes. Spontaneous glowing was dim except for a peak of 6.4× 10⁴quanta-s^?1 cell^?1, which occurred at CT22.9 for LD14:10 and at CT22.8 for LD12:12. The total integrated emission of spontaneous flashing and glowing during the dark phase was 4×10⁹ quantacell^?1, equivalent to the total stimulable luminescence. The rhythms for C. horrida flash and glow behavior were similar to those of Gonyaulax polyedra, although flash rate and quantum flux were greater. Spontaneous bioluminescence in C. horrida may be a circadian rhythm because it persisted for at least three cycles in constant dark conditions. This is also the first detailed study of the stimulated bioluminescence of C. horrida, which also displayed a diurnal rhythm. Cultures exhibited >200 times more mechanically stimulated bioluminescence during the dark phase than during the light phase. Mechanical stimulation during the dark phase resulted in 6.7 flashes. cell^?1; flashes were brighter and longer in duration than spontaneous flashes. Cruise-collected cells exhibited variability in quantum flux with few differences in flash kinetics. The role of dinoflagellate spontaneous bioluminescence in the dynamics of near-surface oceanic communities is unknown, but it may be an important source of natural in situ bioluminescence.     

Bubble stimulation efficiency of dinoflagellate bioluminescence

Dinoflagellate bioluminescence , a common source of bioluminescence in coastal waters , is stimulated by flow agitation . Although bubbles are anecdotally known to be stimulatory , the process has never been experimentally investigated . This study quantified the flash response of the bioluminescent dinoflagellate Lingulodinium polyedrum to stimulation by bubbles rising through still seawater . Cells were stimulated by isolated bubbles of 0 . 3–3 mm radii rising at their terminal velocity , and also by bubble clouds containing bubbles of 0 . 06–10 mm radii for different air flow rates . Stimulation efficiency , the proportion of cells producing a flash within the volume of water swept out by a rising bubble , decreased with decreasing bubble radius for radii less than approximately 1 mm . Bubbles smaller than a critical radius in the range 0 . 275–0 . 325 mm did not stimulate a flash response . The fraction of cells stimulated by bubble clouds was proportional to the volume of air in the bubble cloud , with lower stimulation levels observed for clouds with smaller bubbles . An empirical model for bubble cloud stimulation based on the isolated bubble observations successfully reproduced the observed stimulation by bubble clouds for low air flow rates . High air flow rates stimulated more light emission than expected , presumably because of additional fluid shear stress associated with collective buoyancy effects generated by the high air fraction bubble cloud . These results are relevant to bioluminescence stimulation by bubbles in two‐phase flows , such as in ship wakes , breaking waves , and sparged bioreactors . Copyright © 2015 John Wiley & Sons, Ltd.     

Diversity of the Luciferin Binding Protein Gene in Bioluminescent Dinoflagellates – Insights from a New Gene in Noctiluca scintillans and Sequences from Gonyaulacoid Genera

Dinoflagellate bioluminescence systems operate with or without a luciferin binding protein, representing two distinct modes of light production. However, the distribution, diversity, and evolution of the luciferin binding protein gene within bioluminescent dinoflagellates are not well known. We used PCR to detect and partially sequence this gene from the heterotrophic dinoflagellate Noctiluca scintillans and a group of ecologically important gonyaulacoid species. We report an additional luciferin binding protein gene in N. scintillans which is not attached to luciferase, further to its typical combined bioluminescence gene. This supports the hypothesis that a profound re‐organization of the bioluminescence system has taken place in this organism. We also show that the luciferin binding protein gene is present in the genera Ceratocorys, Gonyaulax, and Protoceratium, and is prevalent in bioluminescent species of Alexandrium. Therefore, this gene is an integral component of the standard molecular bioluminescence machinery in dinoflagellates. Nucleotide sequences showed high within‐strain variation among gene copies, revealing a highly diverse gene family comprising multiple gene types in some organisms. Phylogenetic analyses showed that, in some species, the evolution of the luciferin binding protein gene was different from the organism's general phylogenies, highlighting the complex evolutionary history of dinoflagellate bioluminescence systems.     

Use of Bioluminescence Markers To Detect Pseudomonas spp. in the Rhizosphere

The use of bioluminescence as a sensitive marker for detection of Pseudomonas spp. in the rhizosphere was investigated. Continuous expression of the luxCDABE genes, required for bioluminescence, was not detectable in the rhizosphere. However, when either a naphthalene-inducible luxCDABE construct or a constitutive luxAB construct (coding only for the luciferase) was introduced into the Pseudomonas cells, light emission could be initiated just prior to measurement by the addition of naphthalene or the substrate for luciferase, n-decyl aldehyde, respectively. These Pseudomonas cells could successfully be detected in the rhizosphere by using autophotography or optical fiber light measurement techniques. Detection required the presence of 10³ to 10⁴ CFU/cm of root, showing that the bioluminescence technique is at least 1,000-fold more sensitive than β-galactosidase-based systems.     

MICROSOURCES OF BIOLUMINESCENCE IN PYROCYSTIS FUSIFORMIS (PYRROPHYTA)1

The large dinoflagellate, Pyrocystis fusiformis Murray, emits biolumtnescence on stimulation with dilute acid. The bioluminescence can be seen in the light microscope to originate in a spherical region just distal to the nucleus during the day and appears as a persistent glow which can be localized in an orange-brown sphere. At night, the bioluminescence, in response to stimulation, is a bright flash from microsources scattered throughout the cytoplasm. The orange sphere can no longer be seen nor does a bioluminescent glow originate from this central region on stimulation. This difference in the position of intracellular bioluminescence between day and night has allowed the identification in electron micrographs of structures which correspond to the source of bioluminescence during the day. Light is emitted from a spherical mass of vesicles which contain electron-dense short rods with rounded ends, sometimes crossed by electron-transparent narrow bands. At night, these vesicles can be recognized in the peripheral cytoplasm. It is proposed that these vesicles are the structural counterparts of the microsources of bioluminescence in P. fusiformis.     

Separation of the apoprotein and reconstitution of the holoprotein from the long-lived intermediate in bacterial bioluminescence

A long-lived intermediate in bacterial bioluminescence, which has been suggested to be an FMN flavoprotein, has been separated as an apoprotein plus free FMN and the holoprotein reconstituted by addition of FMN (K_a = 7 × 10⁵ M^?1). The apoprotein preparation reacts with long-chain aldehyde to give the full quantum yield observed for the complete system. Only after removal of all remaining FMN in the apoprotein preparation by prior dialysis of luciferase against KBr and inclusion of apoflavodoxin in the reaction mixture, can a dependence of the light output on FMN be observed. Bacterial bioluminescence therefore appears to be in the class of sensitized chemiluminescence with FMN acting as the specific sensitizing agent.     

COMPARATIVE STUDY OF LUMINESCENT AND NONLUMINESCENT STRAINS OF GONYAULAX EXCAVATA (PYRRHOPHYTA)1,2

Three of ten cultures of Gonyaulax excavata (Braarud) Balech isolated from the 1972 New England red tide are nonluminescent, Biochemical components of dinoflagellate bioluminescence were not detected in the extracts from these three isolates. Cells of the nonluminescent cultures were identical to those of luminescent cultures as compared by light microscopy, major body plate tabulation, cell size and growth. Both luminescent and nonluminescent cells were toxic as determined by using the mouse bioassay for paralytic shellfish poisoning. All the 122 clones made from one of the luminescent isolates were luminescent suggesting this feature is a stable trait. We conclude that these isolates represent luminescent and luminescent strains of G. excavata. This is the first intraspecific investigation of in vitro bioluminescent components between nonluminescent and luminescent strains of a dinoflagellate.     

OLIGONUCLEOTIDE PRIMERS FOR THE DETECTION OF BIOLUMINESCENT DINOFLAGELLATES REVEAL NOVEL LUCIFERASE SEQUENCES AND INFORMATION ON THE MOLECULAR EVOLUTION OF THIS GENE1

Bioluminescence is reported in members of 18 dinoflagellate genera. Species of dinoflagellates are known to have different bioluminescent signatures, making it difficult to assess the presence of particular species in the water column using optical tools, particularly when bioluminescent populations are in nonbloom conditions. A “universal” oligonucleotide primer set, along with species and genus‐specific primers specific to the luciferase gene were developed for the detection of bioluminescent dinoflagellates. These primers amplified luciferase sequences from bioluminescent dinoflagellate cultures and from environmental samples containing bioluminescent dinoflagellate populations. Novel luciferase sequences were obtained for strains of Alexandrium cf. catenella (Whedon et Kof.) Balech and Alexandrium fundyense Balech, and also from a strain of Gonyaulax spinifera (Clap. et Whitting) Diesing, which produces bioluminescence undetectable to the naked eye. The phylogeny of partial luciferase sequences revealed five significant clades of the dinoflagellate luciferase gene, suggesting divergence among some species and providing clues on their molecular evolution. We propose that the primers developed in this study will allow further detection of low‐light‐emitting bioluminescent dinoflagellate species and will have applications as robust indicators of dinoflagellate bioluminescence in natural water samples.     

Variability in the bioluminescence response of the dinoflagellate Pyrocystis lunula

Light emission in dinoflagellates is induced by water motions. But although it is known that mechanical stimulations of these organisms trigger the bioluminescent response, the exact mechanism that involves some cell membrane excitations by fluid motions is not yet fully understood and is still controversial. We show in this experimental study that the accelerated shear flow, created by abrupt rotations of one or both co-axial cylinders of a Couette shearing chamber excites the light emission from cultured dinoflagellates Pyrocystis lunula. Following our first results published earlier that state that pure laminar shear does not excite the main bioluminescent response in dinoflagellates, our present experiments show that both shear and acceleration in the flow are needed to trigger the bioluminescent response. Besides, the probability to stimulate this bioluminescent response under acceleration and shear is deduced from the response curves. This response follows a Gaussian distribution that traduces a heterogeneity in individual cell thresholds for the stimulation of bioluminescence in a dinoflagellate population. All these results will have a repercussion in the possible applications of dinoflagellate bioluminescence in flow visualizations and measurements. Moreover, this study opens a new way in studying mechanically-induced stimulus thresholds at the cell level.     

Evidence for the role of G-proteins in flow stimulation of dinoflagellate bioluminescence

Luminescent dinoflagellates respond to flow by the production of light. The primary mechanotransduction event is unknown, although downstream events include a calcium flux in the cytoplasm, a self-propagating action potential across the vacuole membrane, and a proton flux into the cytoplasm that activates the luminescent chemistry. Given the role of GTP-binding (G) proteins in the mechanotransduction of flow by nonmarine cells and the presence of G-proteins in dinoflagellates, it was hypothesized that flow-stimulated dinoflagellate bioluminescence involves mechanotransduction by G-proteins. In the present study, osmotic swelling of cells of the dinoflagellate Lingulodinium polyedrum was used as a drug delivery system to introduce GDPbetaS, an inhibitor of G-protein activation. Osmotically swollen cells produced higher levels of flow-stimulated bioluminescence at a lower threshold of shear stress, indicating they were more flow sensitive. GDPbetaS inhibited flow-stimulated bioluminescence in osmotically swollen cells and in cells that were restored to the isosmotic condition following hypoosmotic treatment with GDPbetaS. These results provide evidence that G-proteins are involved in the mechanotransduction of flow in dinoflagellates and suggest that G-protein involvement in mechanotransduction may be a fundamental evolutionary adaptation.     

Construction of a Tn5 derivative encoding bioluminescence and its introduction in Pseudomonas, Agrobacterium and Rhizobium

Summary A simple method based upon the use of a Tn5 derivative, Tn5-Lux, has been devised for the introduction and stable expression of the character of bioluminescence in a variety of gram-negative bacteria. In Tn5-Lux, the luxAB genes of Vibrio harveyi encoding luciferase are inserted on a SalI-BglII fragment between the kanamycin resistance (Km^r) gene and the right insertion sequence. The transposon derivative was placed on a transposition suicide vehicle by in situ recombination with the Tn5 suicide vector pGS9, to yield pDB30. Mating between Escherichia coli WA803 (pDB30) and a strain from our laboratory, Pseudomonas sp. RB100C, gave a Km^r transfer frequency of 10^-6 per recipient, a value 10 times lower than that obtained with the original suicide vehicle pGS9. Tn5-Lux was also introduced by insertion mutagenesis in other strains of gram-negative soil bacteria. The bioluminescence marker was expressed in the presence of n-decanal, and was monitored as chemiluminescence in a liquid scintillation counter. The recorded light intensities were fairly comparable among the strains, and ranged between 0.2 to 1.8x10⁶ cpm for a cell density of 10³ colony forming units/ml. Nodules initiated by bioluminescent strains of Rhizobium leguminosarum on two different hosts were compared for intensity of the bioluminescence they produced.     

PHOTOINHIBITION OF MECHANICALLY STIMULABLE BIOLUMINESCENCE IN THE AUTOTROPHIC DINOFLAGELLATE CERATIUM FUSUS (PYRROPHYTA)1

Ceratium fusus (Ehrenb.) Dujardin was exposed to light of different wavelengths and photon flux densities (PFDs) to examine their effects on mechanically stimulable bioluminescence (MSL). Photoinhibition of MSL was proportional to the logarithm of PFD. Exposure to I μmol photons·m^?2s^?1 of broadband blue light (ca. 400–500 nm) produced near-complete photoinhibition (≥90% reduction in MSL) with a threshold at ca. 0.01 μmol photons·m^?2·s^?1. The threshold of photoinhibition was ca. an order of magnitude greater for both broadband green (ca. 500–580 nm) and red light (ca. 660–700 nm). Exposure to narrow spectral bands (ca. 10 nm half bandwidth) from 400 and 700 nm at a PFD of 0.1 μmol photons·m^?2·s^?1 produced a maximal response of photoinhibition in the blue wavelengths (peak ca. 490 nm). A photoinhibition response (≥ 10%) in the green (ca. 500–540 nm) and red wavelengths (ca. 680 nm) occurred only at higher PFDs (1 and 10 μmol photons·m^?2·s^?1). The spectral response is similar to that reported for Gonyaulax polyedra Stein and Pyrocystis lunula Schütt and unlike that of Alexandrium tamarense (Lebour) Balech et Tangen. The dinoflagellate's own bioluminescence is two orders of magnitude too low to result in self-photoinhibition. The quantitative relationships developed in the laboratory predict photoinhibition of bioluminescence in populations of C. fusus in the North Atlantic Ocean.     

DISTRIBUTION AND GENETIC DIVERSITY OF THE LUCIFERASE GENE WITHIN MARINE DINOFLAGELLATES1

Dinoflagellates are the most abundant protists that produce bioluminescence. Currently, there is an incomplete knowledge of the identity of bioluminescent species arising from inter‐ and intraspecific variability in bioluminescence properties. In this study, PCR primers were designed to amplify the dinoflagellate luciferase gene (lcf) from genetically distant bioluminescent species. One of the primer pairs was “universal,” whereas others amplified longer gene sequences from subsets of taxa. The primers were used to study the distribution of lcf and assess bioluminescence potential in dinoflagellate strains representing a wide variety of taxa as well as multiple strains of selected species. Strains of normally bioluminescent species always contained lcf even when they were found not to produce light, thus demonstrating the utility of this methodology as a powerful tool for identifying bioluminescent species. Bioluminescence and lcf were confined to the Gonyaulacales, Noctilucales, and Peridiniales. Considerable variation was observed among genera, or even species within some genera, that contained this gene. Partial sequences of lcf were obtained for the genera Ceratocorys, Ceratium, Fragilidium, and Protoperidinium as well as from previously untested species or gene regions of Alexandrium and Gonyaulax. The sequences revealed high variation among gene copies that obscured the boundaries between species or even genera, some of which could be explained by the presence of two genetic variants within the same species of Alexandrium. Highly divergent sequences within Alexandrium and Ceratium show a more diverse composition of lcf than previously known.