Collection of organism and maintenance of cultures
Seawater samples were collected from a tide pool at Pachena Beach, Bamfield, British Columbia, Canada (48° 47.551' N, 125° 06.974' W) on June 18, 2010. The samples were inoculated in Provasoli's Enriched Seawater (PES) medium  and maintained at room temperature for one week. Tetraselmis sp. was isolated from the enrichment culture by micropipetting into sterile PES medium. Rapaza viridis n. gen. et sp. was also isolated from the enrichment culture into PES medium containing Tetralselmis sp. as a food source. Both cultures were incubated at 20°C under an illumination of 55-59 μmol photons/m2/s with 12:12 light:dark (L:D) light regime. The cultures were transferred every 10 to 14 days by bringing 1 ml of the culture into 25 ml of PES medium and prey cells. The cultures of R. viridis n. gen. et sp. and Tetraselmis sp. have been deposited into the American Type Culture Collection (ATCC), Manasas, VA, USA as PRA-360 and PRA-361, respectively.
Differential interference contrast (DIC) light micrographs were generated using a Zeiss Axioplan 2 imaging microscope equipped with Leica DC500 digital camera. Digital videos of live cells were generated using a Zeiss Axioplan 2 equipped with a Q imaging Microimager II digital camera and Q Capture v 2.8.1 software.
Cells of R. viridis were exposed to several different food sources in addition to Tetraselmis sp. ATCC PRA-361. Cultures of Tetraselmis sp. NEPCC365, Tetraselmis sp. NEPCC498, T. striata NEPCC487 and T. tetrathelle NEPCC 483 were obtained from the Canadian Center for the Culture of Microorganisms (CCCM) at the University of British Columbia, Vancouver, BC, Canada. A culture of Navicula sp. ATCC PRA-314 was obtained from the American Type Culture Collection (Manasas VA, USA). Cultures of the prasinophyte Pycnococcus sp., the chlorophyte Dunaliella sp., and the dinoflagellate Scrippsiella trochoidea were established in the lab from different marine environments independently.
Cultures of R. viridis were starved of prey cells and observed every 2-3 days until the death of all R. viridis cells was confirmed. We confirmed the existence of at least one chloroplast in the starved R. viridis cells using a Zeiss Axioplan 2 imaging microscope after one week, three weeks and five weeks from the start of starvation.
Cultures of R. viridis were also grown in the dark at 20°C. Prey cells (Tetraselmis sp., ATCC PRA-361) were added every 1-2 days until the death of all R. viridis cells was confirmed. As a control experiment, cultures grown under normal light conditions were also treated exactly like the cultures grown in the dark (e.g., the same volume of prey cells was added at the same intervals of time).
Scanning electron microscopy
A culture of R. viridis was mixed in 1% (v/v) OsO4 in seawater at room temperature. The fixed cells were mounted on polycarbonate Millipore filters (13 mm diam., 5 μm pore size) or glass plates coated with poly-L-lysine at room temperature for 15 min. The cells were rinsed with distilled water and dehydrated with a graded ethanol series from 30% to absolute ethanol before being critical point dried with CO2 using a Tousimis Critical Point Dryer. The dried cells were then coated with gold using a Cressington 208HR High Resolution Sputter Coater and observed with a Hitachi S-4700 field emission SEM.
Transmission electron microscopy
Two different cultures were examined with transmission electron microscopy (TEM): (1) a culture starved of Tetraselmis sp. cells for three weeks and (2) a culture that was fed cells of Tetraselmis sp. one hour prior to fixation. The cultures were pre-fixed in 2.5% (v/v) glutaraldehyde with 0.2 M sucrose in 0.1 M sodium cacodylate buffer (SCB) (pH 7.2) at room temperature for 2 hours. The pre-fixed cells were washed in 0.2 M SCB (pH 7.2) twice and post-fixed in 1% (w/v) osmium tetroxide in 0.2 M SCB (pH 7.2) at room temperature for 1 hour. The fixed cells were dehydrated through a graded series of ethanol and 100% acetone. The dehydrated cells were then infiltrated with acetone-Epon 812 resin mixtures and 100% Epon 812 resin. Ultra-thin serial sections were collected on copper Formvar-coated slot grids, stained with 2% (w/v) uranyl acetate and lead citrate, and observed using a Hitachi H7600 TEM.
DNA extraction, PCR amplification, alignment and phylogenetic analysis
Genomic DNA was extracted using the MasterPure Complete DNA and RNA purification Kit (Epicentre, WI, USA) from a culture of R. viridis that was starved of prey cells for three weeks. Polymerase chain reactions (PCR) were performed using PuRe Taq Ready-To-Go PCR beads kit (GE Healthcare, Buckinghamshire, UK). The nearly complete eukaryotic SSU rDNA gene was amplified using the eukaryotic universal primers 5'-TGATCCTTCTGCAGGTTCACCTAC-3' and 5'-GCGCTACCTGGTTGATCCTGCCAGT-3' with the same PCR protocol described by Breglia et al. (2010) . The amplified DNA fragments were purified from agarose gels using UltraClean 15 DNA Purification Kit (MO Bio, CA, USA), and subsequently cloned into the TOPO TA Cloning Kit (Invitrogen, CA, USA). One clone was sequenced using ABI Big-Dye reaction mix (BigDye 3.1) using the vector forward and reverse primers and also internal primers (nomet1134R: 5'-TTTAAGTTTCAGCCTTGCG-3' and SR4Eug: 5'-ACTGGAGGGCAAGYCTGGT-3') oriented in both directions. The new sequence was initially identified by BLAST analysis, confirmed with molecular phylogenetic analysis, and deposited into GenBank: AB679269.
The SSU rRNA sequence from R. viridis was added to a 39-taxon alignment focused on the diversity of euglenids using representative kinetoplastids and diplonemids as an outgroup. Ambiguously aligned positions and gaps were excluded from alignment, leaving 805 unambiguously aligned positions; the alignment is available from the authors upon request.
Maximum likelihood (ML) analysis was performed on the 39-taxon alignment using PAUP* version 4.0b10 . Prior to starting the ML analysis, we used Akaike information criterion (AIC) test as implemented in the software jModelTest 0.1.1.  to find the model of evolution that best fits the data set for the Maximum likelihood (ML) analysis. The result indicated that the TIM1ef, which allows for equal base frequencies and four substitution rates (AC = GT; AT = CG; AG, CT) [27, 28], plus gamma model should be used for this data set. The parameters were as follows: assumed nucleotide frequencies are equal; substitution rate matrix with A-C substitutions = 1.0000, A-G = 2.7002, A-T = 0.7141, C-G = 0.7141, C-T = 3.8573, G-T = 1.0000; proportion of sites assumed to be invariable = 0 and rates for variable sites assumed to follow a gamma distribution with shape parameter = 0.3930. The ML tree was implemented using the heuristic search option with TBR branch swapping. Bootstrap analyses  were carried out for ML with 500 replicates to evaluate statistical reliability.
The alignment was also analyzed with Bayesian methods using the MrBayes 3.1.2 . The program was set to operate the GTR model with a gamma distribution and four Monte-Carlo-Markov chains (MCMC) starting from a random tree. A total of 500,000 generations were calculated with trees sampled every 100 generations. The first 1,250 trees in each run were discarded as burn-in using the sumt command. Posterior probabilities correspond to the frequency at which a given node was found in the post burn-in trees.
The SSU rDNA nucleotide sequences included in 39-taxon analyses for this paper are available from the GenBank database under the following accession numbers: Anisonema acinus [GenBank:AF403160], Bihospites bacati [GenBank:HM004354], Bodo saltans [GenBank:AY998648], Calkinsia aureus [GenBank:EU753419], Colacium sp. [GenBank:DQ140154], Dimastigella mimosa [GenBank:DQ207576], Dinema sulcatum [GenBank:AY061998], Diplonema ambulator [GenBank:AF380996], Diplonema papillatum [GenBank:AF119811], Discoplastis spathirhyncha [GenBank:AJ532454], Distigma proteus [GenBank:AF106036], Entosiphon sp. [GenBank:AY425008], Entosiphon sulcatum [GenBank:AF220826], Euglena gracilis [GenBank:AF283308], Euglena longa (as Astasia longa) [GenBank:AF112871], Euglena quartana (as Khawkinea quartana) [GenBank:U84732], Euglena stellata [GenBank:AF081590], Euglena viridis [GenBank:AF445460], Eutreptia viridis [GenBank:AF157312], Eutreptiella gymnastica [GenBank:AF081590], Eutreptiella pomquetensis [GenBank:AJ532398], Ichthyobodo necator [GenBank:AY224691], Lepocinclis buetschlii [GenBank:AF096993], Menoidium cultellus [GenBank:AF295019], Monomorphina sp. [GenBank:DQ140130], Neobodo designis [GenBank:AF209856], Notosoleus ostium [GenBank:AF403159], Peranema sp. [GenBank:AY048919], Peranema trichophorum [GenBank:AF386636], Petalomonas cantuscygni [GenBank:AF386635], Phacus aenigmaticus [GenBank:AF283313], Ploteotia costata [GenBank:AF525486], Rapaza viridis [GenBank: AB679269], Rhabdomonas costata [GenBank:AF295021], Rhynchomonas nasuta [GenBank:AY998642], Rhynchopus sp. [GenBank:AF380997], Strombomonas triquetra [GenBank:DQ140153], Trypanosoma evansi [GenBank:AY904050], Trypanosoma sp. [GenBank:EF375883].
A digital archive of this paper is available from PubMed Central and print copies are available from libraries in the following five museums: Natural History Museum Library (Cromwell Road, London, SW7 5BD, UK), American Museum of Natural History (Department of Library Services, Central Park West at 79th St., New York, NY, 10024, USA), Muséum national d'Histoire naturelle (Direction des bibliothèques et de la documentation, 38 rue Geoffroy Saint-Hilaire, 75005 Paris, France), Russian Academy of Sciences (Library for Natural Sciences of the RAS Znamenka str., 11, Moscow, Russia) and Academia Sinica (Life Science Library, 128 Sec. 2 Academia Rd, Nankang Taipei 115, Taiwan R.O.C.).