Non-photosynthetic predators are sister to red algae · Non-photosynthetic predators are sister to...

Letterhttps://doi.org/10.1038/s41586-019-1398-6

Non-photosynthetic predators are sister to red algaeryan M. r. Gawryluk1,3,5*, Denis V. tikhonenkov1,2,5*, elisabeth Hehenberger1,4, Filip Husnik1, Alexander P. Mylnikov2 & Patrick J. Keeling1*

Rhodophyta (red algae) is one of three lineages of Archaeplastida1, a supergroup that is united by the primary endosymbiotic origin of plastids in eukaryotes2,3. Red algae are a diverse and species-rich group, members of which are typically photoautotrophic, but are united by a number of highly derived characteristics: they have relatively small intron-poor genomes, reduced metabolism and lack cytoskeletal structures that are associated with motility, flagella and centrioles. This suggests that marked gene loss occurred around their origin4; however, this is difficult to reconstruct because they differ so much from the other archaeplastid lineages, and the relationships between these lineages are unclear. Here we describe the novel eukaryotic phylum Rhodelphidia and, using phylogenomics, demonstrate that it is a closely related sister to red algae. However, the characteristics of the two Rhodelphis species described here are nearly opposite to those that define red algae: they are non-photosynthetic, flagellate predators with gene-rich genomes, along with a relic genome-lacking primary plastid that probably participates in haem synthesis. Overall, these findings alter our views of the origins of Rhodophyta, and Archaeplastida evolution as a whole, as they indicate that mixotrophic feeding—that is, a combination of predation and phototrophy—persisted well into the evolution of the group.

Two previously undescribed eukaryovorous protists, Rhodelphis limneticus and Rhodelphis marinus (see Supplementary Information for taxonomic diagnosis), were isolated from a freshwater lake and marine coral sand, respectively. Rhodelphis are 10–13 μm, oval or tapered, slightly flattened cells with two subapical heterodynamic flagella (Fig. 1a–f and Extended Data Figs. 1a, 2a, b). Characteristic morphological features include umbrella-shaped glycostyles on the cell surface and flagellum (Fig. 1g, h, s and Extended Data Figs. 1d, 2c–e), perpendicularly oriented flagellar basal bodies (Fig. 1i, j and Extended Data Figs. 1e, k, 1, 2g, h) with outgoing striated structures and at least two fibrils (Fig. 1j–l and Extended Data Fig. 2e), one narrow and two wide microtubular bands (Fig. 1l–q and Extended Data Fig. 1h–l), a flagellar transition zone with a transverse plate at the cell surface and a proximal diaphragm through which a central pair of flagellar micro-tubules surrounded by a cylinder passes (Fig. 1q, r and Extended Data Fig. 1e–h), a sac-shaped double-layered smooth endoplasmic reticulum (Fig. 1 s–u and Extended Data Fig. 1o) and mitochondria with tubular cristae (Fig. 1 s–u and Extended Data Fig. 2g, h, j). Plastids were not observed.

To establish the evolutionary position of Rhodelphis, we sequenced transcriptomes from cultures and manually isolated cells, and generated a concatenated 153-taxon/253-protein supermatrix (153/253 dataset; 56,312 sites). Maximum-likelihood and Bayesian analyses recovered Rhodelphis as a well-supported sister to the red algae, with ultrafast bootstrap support of 97%, Shimodaira–Hasegawa-like approximate likelihood ratio test (SH-aLRT) support of 0.99 and a Bayesian poste-rior probability of 0.98 (Extended Data Fig. 3a, b). Analyses of a second supermatrix without the picozoan MS584-115 and Telonema (both of which had poor datasets; 151-taxon/253-protein supermatrix (151/253 dataset), 56,530 sites), recovered Rhodelphis and red algae with

complete statistical support (Fig. 2a, b and Extended Data Fig. 4a, b), although approximately unbiased tests were unable to distinguish between Archaeplastida monophyly or paraphyly (P = 0.6693 and P = 0.3397, respectively). To examine the possibility that long-branch attraction affected the position of Rhodelphis, we carried out fast-site removal analyses, which showed that support for the sisterhood of Rhodelphis and red algae remained high for both 153/253 (Extended Data Fig. 3c) and 151/253 datasets (Fig. 2c). To test whether mixed gene ancestry (for example, from horizontal gene transfer) affected the phylogenetic placement of Rhodelphis, we calculated internode certainty values6 from 253 bootstrapped single-gene trees and the 50 trees with the highest relative tree certainty. These analyses showed a degree of conflict that is similar to other ancient but well-established rela-tionships (such as opisthokont monophyly) and that the internode certainty scores for the Rhodelphis and red algae bipartition increase in the 50 best-supported trees (Extended Data Fig. 5a, b), which also recover Rhodelphis as sisters to the red algae in a concatenated phylog-eny (Extended Data Fig. 6). Coalescent species trees7 estimated from the 253 and 50 single-gene tree datasets also recover Rhodelphis and red algae as sisters, with full support (Extended Data Fig. 7a, b).

Most Archaeplastida are photoautotrophic and phagotrophy is very rare. However, phagotrophy must have existed for the archaeplastid ancestor to take up the plastid, and must have persisted at least until the protoplastid became a reliable source of both energy and nutrients. This suggests a key mixotrophic intermediate stage; however, because the few known phagotrophic archaeplastids have been interpreted as secondarily derived, it has been widely assumed that phagotrophy was lost early in the evolution of archaeplastids8. We therefore characterized the Rhodelphis genome and transcriptomes further to investigate the ancestral states of phagotrophy and other characteristics that are seem-ingly absent from red algal and archaeplastid ancestors. Altogether, these analyses suggest that Rhodelphis possess larger and more gene-rich nuclear genomes than most red algae (Extended Data Table 1a–c) and genes with many more introns (nearly 40,000 spliceosomal introns were identified in genome–transcriptome comparisons in R. limneti-cus). But the most notable differences between Rhodelphis and red algae are in gene content: Rhodelphis possess genes that are associated with centrioles, autophagy and synthesis of glycosylphosphatidylinositol, all of which are absent from red algae9. Notably, genes that encode flagellar proteins are absent from red algae, whereas Rhodelphis encode homologues of 209 out of 361 high-confidence Chlamydomonas flagel-lar proteins10, consistent with our microscopic observations.

Rhodelphis engulf whole bacteria and eukaryotic prey at the posterior end (Extended Data Fig. 1r–t and Supplementary Video 1), but a dis-tinct feeding apparatus is not evident. Phagotrophy in Rhodelphis there-fore differs from feeding in some prasinophytes (the only other known class in which phagotrophy occurs within Archaeplastida), which use a mouth-like opening, a tubular channel and a large permanent vac-uole to engulf, transport and digest bacterial cells11. Specific genetic markers of phagotrophy are difficult to define; however, genome-level predictive models12 indicate that the genetic repertoire of Rhodelphis is consistent with phagocytotic feeding (Extended Data Fig. 8).

1Department of Botany, University of British Columbia, Vancouver, British Columbia, Canada. 2Papanin Institute for Biology of Inland Waters, Russian Academy of Sciences, Borok, Russia. 3Present address: Department of Biology, University of Victoria, Victoria, British Columbia, Canada. 4Present address: GEOMAR – Helmholtz Centre for Ocean Research Kiel, Division of Experimental Ecology, Kiel, Germany. 5These authors contributed equally: Ryan M. R. Gawryluk, Denis V. Tikhonenkov. *e-mail: [email protected]; [email protected]; [email protected]

N A t U r e | www.nature.com/nature

https://doi.org/10.1038/s41586-019-1398-6

mailto:[email protected]



LetterreSeArCH

Taken together, the cellular motility and phagotrophic feeding by Rhodelphis and the ancestral photosynthetic capacity of red algae indi-cates that their common ancestor was mixotrophic.

Although an ancestor of Rhodelphis must have been photosynthetic, no plastids were observed using microscopy, so we searched for genetic evidence of a relic plastid, as complete loss of the plastid is exceedingly rare13,14. In archaeplastids, nucleus-encoded proteins are targeted to plastids through N-terminal transit-peptide leaders that are recognized by TIC/TOC import complexes. We identified homologues of several plastid import proteins in Rhodelphis, including TIC20, TIC22, TIC32

and TOC75 (Extended Data Fig. 9b–e), as well as many proteins with a putative plastid function (Fig. 3a and Supplementary Table 1). Notably, plastid proteins of Rhodelphis encode leader sequences that are similar to archaeplastid transit peptides (Extended Data Fig. 9a); they do not have bipartite leader sequences or specific homologues of SELMA complex subunits15 that are indicative of protein targeting to complex plastids by the endoplasmic reticulum. Taken together, the evidence from the analysis of the phylogenomics, plastid-targeting leaders and plastid import machinery is consistent with the conclusion that Rhodelphis have a primary plastid as with other members of Archaeplastida.

a b c d

e

f

g h i

j k l

m n o

p q

r s t u

gl

cvcv

gl bb2

tpbb1

gl

tp

tp

cmt

cmt cl

dp

dp

bb2 bb1

bb2

wmb1

fb

fb

ss

bb1st

sm

sm

bb2

cv

nn

wmb1

wmb2

wmb1

bb2

wmb2

wmb2nmb

ss

mrt

ser ser

ser

mt

mtcr

dc

ob

obbb1

af

pf

fv

rgl

rs

n

Fig. 1 | Cell morphology of R. limneticus. a–c, Living cells, visualized by light microscopy. d, Cyst, visualized by light microscopy. e, Scanning electron microscopy image highlighting the flagella. f–u, Cells, visualized by transmission electron microscopy. f, Longitudinal section. g, Glycostyles can be seen on the cell surface. h, Transverse section of the posterior flagellum covered with glycostyles. i–l, Arrangement of basal bodies, connecting structures and satellites. m–p, Arrangement of microtubular bands, microtubules and satellites. q, r, Structure of the flagellar transition zone. s, t, Transverse sections of the cell showing the sac-shaped smooth endoplasmic reticulum, nucleus and mitochondria. u, Area near the endoplasmic reticulum showing a single mitochondrion with dark condensations and vesicles with rudiments of glycostyles.

af, anterior flagellum; cl, cylinder; cv, contractile vacuole; cr, cristae; cmt, central microtubules; dc, dark condensation; dp, diaphragm; fb, fibril; fv, food vacuole; gl, glycostyles; bb1, basal body of posterior flagellum; bb2, basal body of anterior flagellum; mt, mitochondrion; mrt, microtubule; n, nucleus; nmb, narrow microtubular band; ob, osmiophilic body; pf, posterior flagellum; rgl, rudiments of glycostyles; rs, reserve substance; ser, sac of smooth endoplasmic reticulum; sm, single microtubules; ss, striated structure; st, satellite of basal body; tp, transverse plate; wmb1, wide microtubular band 1; wmb2, wide microtubular band 2. Scale bars, 10 μm (a–c), 5 μm (d, e), 1 μm (f), 0.2 μm (g, h, n–p, r), 0.5 μm (i–m, q, u) and 2 μm (s, t). These experiments were repeated 50 (a–d), 3 (e) and 7 (f–u) times, with similar results.


Letter reSeArCH

Consistent with their lack of pigmentation, Rhodelphis encode almost no proteins that are involved in photosynthesis. The exceptions are ferredoxin and ferredoxin NADP+ reductase, which probably partic-ipate in FeS cluster assembly through the plastidial SUF-type (sulfur assimilation) biosynthesis pathway (Fig. 3a). A mitochondrial-type iron sulfur cluster (ISC) pathway is also present in Rhodelphis, which sug-gests that the SUF system supports only plastid FeS proteins.

We found no evidence for the type II fatty acid biosynthesis pathway that is typical of plastids. Instead, Rhodelphis synthesize fatty acids in the cytosol, through a multidomain type I fatty acid synthase, which

is found in numerous animals, fungi and protists, but not in archae-plastids (Fig. 3a). Furthermore, Rhodelphis also lack the plastid methy-lerythritol phosphate (MEP) isoprenoid biosynthetic pathway, and use the cytosolic mevalonate (MVA) pathway instead (Fig. 3a). The MVA pathway is patchily distributed among red algae, and has been inde-pendently lost numerous times9, whereas the MEP pathway is found in most archaeplastids that have been examined to date. The distribu-tion of both pathways highlights the potential for long-term functional redundancy to shape gene loss and metabolic reorganization.

By contrast, most Rhodelphis haem biosynthetic enzymes are prob-ably targeted to plastids, with the exception of the first step that is cat-alysed by δ-aminolevulinic acid synthase (ALAS), which is probably a mitochondrial protein (Fig. 3a); this is probably the major reason that Rhodelphis plastids have not been lost. All of the putatively plastid- targeted haem proteins are of the typical plastid-type (phylogenetic trees are available through the Dryad Digital Repository), except fer-rochelatase (HemH), which branches with homologues from some red algae and dinoflagellates that are in turn nested within a group

b

Rhodelphis limneticus

Rhodelphis marinus

–/81

Chloroplastida

Glaucophyta

Cryptista

Rhodophyta

a

Ancoracysta twista

Thecamonas trahens

MalawimonasCollodictyon sp.

Breviate0.77

0.72

0.82

–

Rhodelphis marinus

Rhodelphis limneticus

Stramenopila

Alveolata

Rhizaria

Haptophyta

Centrohelida

Chloroplastida

Glaucophyta

Cryptista

Discoba

Opisthokonta

Amoebozoa

Rhodophyta

c

Archaeplastida

Rhodelphis + red algae

Opisthokonta

Cryptista + green plantsand glaucophytes

0

25

50

75

100

0 10 20 30 40 50

Thousand sites removed

Boo

tstr

ap s

upp

ort

(%)

Arc

haep

last

ida

Fig. 2 | The evolutionary position of Rhodelphis. a, b, Bayesian (a; CAT + GTR)20 and maximum-likelihood (b; 151/253 dataset; LG + C60 + F + G4)21 analyses place Rhodelphis as a sister to red algae (Rhodophyta). Bayesian analyses also recover the monophyly of Archaeplastida, whereas maximum-likelihood analyses do not. Black dots denote full statistical support for the Bayesian tree (a; Bayesian posterior probability = 1.0) or maximum-likelihood tree (b; maximum-likelihood ultrafast bootstrap = 100%/SH-aLRT = 1.0); support values below 1.0 or 100% are shown in cases in which either value is below this value; support values <0.7 or 70% are not shown (indicated by a ‘−’). Complete Bayesian and maximum-likelihood trees for 151/253 and 153/253 datasets are shown in Extended Data Figs. 3 and 4, respectively. c, Bootstrap support for maximum-likelihood trees (PROT + CAT + LG + F) after progressive removal of the fastest-evolving amino acid sites (151/253 supermatrix) shows that the Rhodelphis and red algae relationship is robust, but that Archaeplastida paraphyly (Cryptista, and green plants and glaucophytes) may be the result of long branch attraction. Support for Opisthokonta monophyly serves as a control for the presence of sufficient information for phylogenomic inference. A parallel analysis for 153/253 is shown in Extended Data Fig. 3c.

a b

Isoprenoidbiosynthesisacetyl-coA

mevalonate

IPP

Type Ifatty acid

biosynthesis(FASI)

Nucleus

Basalbodies

Gene-richIntron-rich

Flagella(swimming)

mtDNA

SCoA

Gly

MitochondrionTubularcristae

TOCTIC

FeS

Haem

e–

PlastidNon-

photosynthetic

Cytosol

Phagocytosis(eukaryotic prey)

Digestion(lysosomes)

ExocytosisMastigonemes

PBGDUROS

URODCPOX

PPOXFeCH

SufBSufE

FNR FDX

NifU

SufS

ALAD

SufDSufC

Galdieria sulphuraria

Cyanidioschyzon merolae

Porphyridium aerugineum

Madagascaria erythrocladiodes

Erythrolobus australicus

Erythrolophus madagascariensis

Rhodella maculata

Porphyra umbilicalisPorphyra purpurea

Porphyridium cruentum


Rhodella maculata

Chondrus crispusPyropia yezoensis

Compsopogon coeruleus

Erythrolobus australicusRhodosorus marinus

88

91

73

75

78

94

88

94

88

86

87

90

87

R. limneticus

LG + R9

Cyanobacteria

CPN600.3

R. marinus

ALAS

Fig. 3 | Genomic insights into Rhodelphis cell biology. a, Rhodelphis genome and transcriptome analyses reveal a relic plastid that generates FeS clusters and probably cooperates with mitochondria in haem biosynthesis, cytosolic isoprenoid and fatty acid synthesis, extensive nuclear mRNA splicing (coloured lines on chromosomes), phagocytosis (Extended Data Fig. 8 and Supplementary Video 1), basal bodies and flagella (a long flagellum was shortened to fit). Note that subcellular structures are not drawn to scale and the mitochondrial genome has not been verified to be circular. b, Maximum-likelihood phylogeny of Cpn60 with a predicted N-terminal transit peptide (Extended Data Fig. 9), indicating the presence of a primary plastid in Rhodelphis, although none were observed by microscopy. Rhodelphis Cpn60 groups basal to homologues from red algae and other eukaryotes with plastids derived from red algae (coloured wedges). Other putative plastid proteins are presented in Supplementary Table 1. IPP, isopentenyl pyrophosphate. Eukaryotic groups are indicated by coloured taxon names/collapsed clades: green, Viridiplantae; red, red algae; grey, glaucophytes; turquoise, cryptophytes; pink, stramenopiles; dark blue, haptophytes; yellow, dinoflagellates; orange, chrompodellids and apicomplexans; brown, opisthokonts, amoebozoans, apusozoans, ancyromonads and ciliates. Taxa or clades outlined in black are prokaryotic.


LetterreSeArCH

of bacterial homologues. Other red algae and lineages with red algal plastids—such as diatoms, cryptophytes and ochrophytes—have a typical cyanobacterial HemH16. This again suggests that there is redun-dancy of HemH types in the common ancestor of Rhodelphis and red algae.

Despite finding many nucleus-encoded putatively plastid-targeted proteins, we found no evidence that supports the existence of a plastid genome. No plastid DNA is present in the genomic datasets of R. limneticus, which were generated from over 300 million paired reads from both cultures and single cells. By contrast, mitochondrial DNA was readily identified; the genome could not be unambiguously assem-bled, but is >100 kilobases and individual contigs were connected in a single contig map. Similarly, no nucleus-encoded components of plastid genome replication, gene expression or translation systems were identi-fied in any Rhodelphis dataset. Notably, many of the nucleus-encoded, putatively plastid-targeted Rhodelphis proteins that we did identify are encoded in plastid genomes of red algae. Taken together, we interpret these observations as strong evidence for the complete loss of plastid DNA in Rhodelphis. This has only been reported in a few non- photosynthetic plastids17,18 and is in contrast to the gene-rich nature of the plastid genomes of red algae19.

In conclusion, phylogenomic analyses strongly support the place-ment of Rhodelphis as a sister lineage to red algae. Rhodelphis are flagellate predators with primary, non-photosynthetic plastids that are involved in haem biosynthesis; all of which indicates that the ancestor of Rhodelphis and red algae was very different from previ-ous models of the ancestors of red algae. This ancestor was probably a mixotrophic flagellate that obtained energy and nutrients from both photosynthetic plastids and phagotrophy, which suggests that phag-otrophy persisted within Archaeplastida until well after the diver-gence of red algae from green plants and glaucophytes. The gene- and intron-rich genomes and complex pattern of biosynthetic pathway retention also provide insights into the potential for functional redun-dancy to persist over considerable periods of evolutionary time, followed by differential loss. Indeed, Rhodelphis reveals that the absence of phagotrophy and many other characteristics of the Archaeplastida as a whole are due to multiple convergent losses rather than an already established ancestral state.

Online contentAny methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; andstatements of data and code availability are available at at https://doi.org/10.1038/s41586-019-1398-6.

Received: 27 February 2019; Accepted: 13 June 2019; Published online xx xx xxxx.

1. Burki, F. The eukaryotic tree of life from a global phylogenomic perspective. Cold Spring Harb. Perspect. Biol. 6, a016147 (2014).

2. Archibald, J. M. The puzzle of plastid evolution. Curr. Biol. 19, R81–R88 (2009). 3. Keeling, P. J. The number, speed, and impact of plastid endosymbioses in

eukaryotic evolution. Annu. Rev. Plant Biol. 64, 583–607 (2013). 4. Qiu, H., Price, D. C., Yang, E. C., Yoon, H. S. & Bhattacharya, D. Evidence of

ancient genome reduction in red algae (Rhodophyta). J. Phycol. 51, 624–636 (2015).

5. Yoon, H. S. et al. Single-cell genomics reveals organismal interactions in uncultivated marine protists. Science 332, 714–717 (2011).

6. Salichos, L. & Rokas, A. Inferring ancient divergences requires genes with strong phylogenetic signals. Nature 497, 327–331 (2013).

7. Zhang, C., Rabiee, M., Sayyari, E. & Mirarab, S. ASTRAL-III: polynomial time species tree reconstruction from partially resolved gene trees. BMC Bioinformatics 19, 153 (2018).

8. Spiegel, F. W. Contemplating the first Plantae. Science 335, 809–810 (2012). 9. Qiu, H., Yoon, H. S. & Bhattacharya, D. Red algal phylogenomics provides a

robust framework for inferring evolution of key metabolic pathways. PLoS Curr. 8, https://doi.org/10.1371/currents.tol.7b037376e6d84a1be34af756a4d 90846 (2016).

10. Pazour, G. J., Agrin, N., Leszyk, J. & Witman, G. B. Proteomic analysis of a eukaryotic cilium. J. Cell Biol. 170, 103–113 (2005).

11. Maruyama, S. & Kim, E. A modern descendant of early green algal phagotrophs. Curr. Biol. 23, 1081–1084 (2013).

12. Burns, J. A., Pittis, A. A. & Kim, E. Gene-based predictive models of trophic modes suggest Asgard archaea are not phagocytotic. Nat. Ecol. Evol. 2, 697–704 (2018).

13. Gornik, S. G. et al. Endosymbiosis undone by stepwise elimination of the plastid in a parasitic dinoflagellate. Proc. Natl Acad. Sci. USA 112, 5767–5772 (2015).

14. Xu, P. et al. The genome of Cryptosporidium hominis. Nature 431, 1107–1112 (2004).

15. Gould, S. B., Maier, U.-G. & Martin, W. F. Protein import and the origin of red complex plastids. Curr. Biol. 25, R515–R521 (2015).

16. Oborník, M. & Green, B. R. Mosaic origin of the heme biosynthesis pathway in photosynthetic eukaryotes. Mol. Biol. Evol. 22, 2343–2353 (2005).

17. Smith, D. R. & Lee, R. W. A plastid without a genome: evidence from the nonphotosynthetic green algal genus Polytomella. Plant Physiol. 164, 1812–1819 (2014).

18. Fernández Robledo, J. A. et al. The search for the missing link: a relic plastid in Perkinsus? Int. J. Parasitol. 41, 1217–1229 (2011).

19. Muñoz-Gómez, S. A. et al. The new red algal subphylum Proteorhodophytina comprises the largest and most divergent plastid genomes known. Curr. Biol. 27, 1677–1684 (2017).

20. Lartillot, N., Lepage, T. & Blanquart, S. PhyloBayes 3: a Bayesian software package for phylogenetic reconstruction and molecular dating. Bioinformatics 25, 2286–2288 (2009).

21. Nguyen, L.-T., Schmidt, H. A., von Haeseler, A. & Minh, B. Q. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol. 32, 268–274 (2015).

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© The Author(s), under exclusive licence to Springer Nature Limited 2019


https://doi.org/10.1038/s41586-019-1398-6

https://doi.org/10.1038/s41586-019-1398-6

https://doi.org/10.1371/currents.tol.7b037376e6d84a1be34af756a4d90846

https://doi.org/10.1371/currents.tol.7b037376e6d84a1be34af756a4d90846

Letter reSeArCH

MethodsData reporting. No statistical methods were used to predetermine sample size. The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment.Cell isolation and culture establishment. R. marinus (clone Colp-29) was obtained from near-shore marine coral sand off the coast of a small island, Bay Canh, near Con Dao Island, South Vietnam (8° 39′ 58.6362″ N, 106° 40′ 49.7562″ E). The sample was collected from a depth of 40 cm on 3 May 2015.

R. limneticus (clone Colp-38) was obtained from a near-shore freshwater sample including organic debris, taken by Y. V. Dubrovsky (IEE NAS Ukraine) on 9 August 2016 from Lake Trubin (floodlands of Desna River, 51° 23′ 49.9986″ N, 32° 22′ 8.0004″ E), near Yaduty village, Chernigovskaya oblast, Ukraine.

The samples were examined on the third, sixth and ninth day of incubation in accordance with previously described methods22. Following isolation by glass micropipette, R. marinus and R. limneticus were propagated on the bodonids Procryptobia sorokini strain B-69 and Parabodo caudatus strain BAS-1, respec-tively, which were grown in marine Schmalz–Pratt’s medium and spring water (Aqua Minerale, PepsiCo or PC Natural Spring Water, President’s Choice) using the bacterium Pseudomonas fluorescens as food23. R. limneticus is currently being stored in a collection of live protozoan cultures at the Papanin Institute for Biology of Inland Waters, Russian Academy of Sciences; however, R. marinus perished after several months of cultivation.Light and electron microscopy. Light microscopy observations of R. limneticus were made using a Zeiss AxioScope A.1 equipped with a differential interference contrast (DIC) water-immersion objective (63×) and an AVT HORN MC-1009/S analogue video camera. Observations of R. marinus were made using a Zeiss Axioplan 2 Imaging microscope equipped with a DIC objective (40×) and a Canon XL H1S video camera.

For scanning electron microscopy, cells from a culture in exponential growth phase were fixed with 2.5% glutaraldehyde (final concentration). The cells were mounted on a glass coverslip coated with poly-l-lysine for 30 min and subsequently rinsed three times with 0.1 M sodium cacodylate buffer (pH 7.34), which was diluted twice with spring water (PC Natural Spring Water, President’s Choice). Next, cells were fixed in 1% osmium tetroxide for 1 h. The fixed cells were rinsed three times with distilled water, 10 min each time, and dehydrated with a graded ethanol series from 30% to absolute ethanol (10 min per step), followed by 100% hexamethyldisilazane (three times, 15 min each) and dried at 65 °C. Dry glass coverslips were mounted on aluminium stubs, coated with gold–palladium, and observed with a Hitachi S4700 scanning electron microscope (Hitachi High-Technologies Corporation).

For transmission electron microscopy, cells were centrifuged, fixed in a cocktail of 0.6% glutaraldehyde and 2% osmium tetroxide (final concentration) prepared using a 0.1 M cacodylate buffer (pH 7.2) for freshwater cells, or Schmaltz–Pratt medium for marine cells at 1 °C for 30–60 min and dehydrated in an alcohol and acetone series (30, 50, 70, 96 and 100%; 20 min per step). Finally, cells were embed-ded in a mixture of Araldite and Epon24. Ultrathin sections were obtained with an LKB ultramicrotome. Transmission electron microscopy observations were obtained using a JEM-1011 (JEOL) electron microscope.Preparation of libraries and sequencing. RNA and genomic DNA isolation. Сells grown in clonal laboratory cultures were collected when the cultures had reached peak abundance and after the prey had been eaten (based on daily light microscopy observations). Cells were collected by centrifugation (1,000g, room temperature) onto a 0.8-μm membrane of a Vivaclear mini column (Sartorius Stedim Biotech, VK01P042); this was done separately for RNA and DNA extractions. Total RNA was then extracted using an RNAqueous-Micro Kit (Invitrogen, AM1931) and converted into cDNA using the Smart-Seq2 protocol25. Additionally, cDNA of R. limneticus was obtained from 20 single cells using the Smart-Seq2 protocol: cells were manually picked from the culture using a glass micropipette and transferred to a 0.2-ml thin-walled PCR tube containing 2 μl cell lysis buffer (0.2% Triton X-100 and RNase inhibitor (Invitrogen)). Total DNA was extracted from the filters using the MasterPure Complete DNA and RNA Purification Kit (Epicentre, MC85200).

The small subunit (SSU) rRNA genes of R. marinus and R. limneticus were amplified by polymerase chain reaction (PCR) using the general eukar-yotic primers GGF (CTTCGGTCATAGATTAAGCCATGC) and GGR (CCTTGTTACGACTTCTCCTTCCTC) and 18SFU and 18SRU26, respectively. PCR products were subsequently cloned (R. marinus) or sequenced directly (R. limneticus) using Sanger dideoxy sequencing.

R. limneticus transcriptome sequencing was performed on the Illumina MiSeq platform with read lengths of 300 bp using the NexteraXT protocol (Illumina, FC-131-1024) to construct paired-end libraries. R. marinus transcriptome sequenc-ing was performed on the Illumina HiSeq platform (UCLA Clinical Microarray Core) with read lengths of 100 bp using the KAPA stranded RNA-seq kit (Roche) to construct paired-end libraries.

R. limneticus DNA was extracted from cultures (containing R. limneticus, prey and bacteria) using the MasterPure DNA Purification Kit (Epicentre) or obtained from three individually picked cells through whole-genome amplification using the TruePrime Single Cell WGA kit v.2.0 (Expedeon) according to the manufacturer’s instructions. Libraries were generated at The Centre of Applied Genomics and 151-bp paired-end reads were sequenced on an Illumina HiSeq X. Whole-genome amplified DNA (TruePrime Single Cell WGA kit v.2.0 (Expedeon)) was sequenced on MinION, Oxford Nanopore Technologies using the Ligation Sequencing Kit 1D (SQK-LSK108, Oxford Nanopore Technologies). Covaris shearing was omit-ted to preserve long fragments. DNA was initially treated with T7 endonuclease to remove extremely branched DNA structures resulting from whole-genome amplification.Sequencing dataset assembly and decontamination. Sequence quality and adaptor contamination of reads from transcriptomic datasets were assessed with FastQC27. Reads were trimmed with Trimmomatic-0.32 ILLUMINACLIP28, with a maximum of two mismatches, a palindromeClipThreshold of 30 and a simpleClipThreshold of 10. Low-quality sequences were discarded, using a sliding window of 4 bp, a minimum quality score of 25 and a minimum trimmed length of 35 bp.

A strand-specific R. marinus HiSeq transcriptome was assembled with Trinity v.2.0.629, with the --SS_lib_type flag set to RF. MiSeq transcriptomes of R. limne-ticus from culture or 20-cell preparations were assembled in essentially the same manner, but without strand specificity. Transdecoder was used to infer the most likely open reading frame sequences, informed by blastp30 and hmmscan queries of the Swissprot and Pfam databases, respectively (E-value cut-off = 1 × 10−5). CD-HIT31 was used to reduce the redundancy of the inferred protein dataset by clustering proteins with ≥95% identity. Extensive in silico decontamination was performed to remove sequences derived from the eukaryotic prey P. sorokini (R. marinus), P. caudatus (R. limneticus) and co-cultured bacteria. We used mega-blast to identify transcripts that were ≥95% identical to sequences of previously generated P. sorokini and P. caudatus MiSeq transcriptome datasets, along with previously generated HiSeq transcriptome datasets from protists that feed on P. sorokini or P. caudatus (that is, sequences that these datasets had in common were probably derived from P. sorokini or P. caudatus). Contaminating bacterial sequences were identified using megablast and blastp queries of the NCBI nt and nr databases. Nucleotide sequences that were ≥80% identical to bacterial entries were removed. For a protein sequence from the Rhodelphis datasets to be classified as bacterial, each of the top-15 blastp hits had to be most similar to a bacterial homologue and ≥70% identical to a bacterial protein. Assessment of transcriptome completeness was performed by searching ‘eukaryote’ protein datasets with BUSCO v.3.0.132, using default parameters.

The R. limneticus genome was assembled from HiSeq X reads (either culture or whole-genome-amplified DNA from single cells (WGA)) and MinION reads using SPAdes v.3.11.133 with kmer lengths of 21, 33, 55, 77, 99 and 121, and with the ‘–sc’ flag activated for the single-cell assembly. For each of the WGA and culture genome assemblies, contigs from co-cultured contaminants (for example, kine-toplastid prey and bacteria) were identified using Autometa34 and removed. As expected, the culture dataset was heavily contaminated, whereas WGA contigs were predominantly from R. limneticus. Assessments of genome assembly were performed using QUAST v.5.0.235.

A search for putative spliceosomal introns in R. limneticus was performed by aligning transcripts to the culture assembly, requiring a minimum 95% identity threshold, using GMAP v.2016-08-1636. Spliceosomal introns with GT/AG splice boundaries were extracted from the GMAP output with a custom Python script and visualized with WebLogo37. The total proportion of transcripts from R. limne-ticus mapping to the nuclear genome sequence was determined with isoblat v.338.

Searches for a plastid genome were carried out by querying Rhodelphis tran-scriptome and genome datasets with red algal plastid-encoded RNAs and proteins, and by searching for rRNAs of plastidial/cyanobacterial affinity with phyloFlash v.3.039. All plastid-type proteins that were found are probably encoded by DNA in the nucleus and no plastidial/cyanobacterial rRNAs were found.Phylogenomic dataset preparation and analysis. Construction of a phylogenomic supermatrix was performed essentially as previously described40, using nearly the same dataset. In brief, blastp was used to identify Rhodelphis homologues, with an expect value threshold of ≤1 × 10−30. Alignments were generated with MAFFT L-INS-I v.7.21241 and trimmed automatically with BMGE v.1.1242 based on the BLOSUM75 substitution matrix. Single-protein maximum-likelihood phylogenies—derived from 20 independent heuristic searches with RAxML v.8.1.643 using the PROTGAMMALGF model—were used to screen for paralogues and sequences that were probably derived from prey contamination. Individual trimmed alignments were concatenated with SCaFOs v.1.2.544, requiring a minimum of 15% coverage for inclusion. The final concatenated alignment included 153 taxa, 253 proteins and 56,312 amino acid sites (153/253 dataset); R. marinus and R. limneticus were well-represented (R. marinus, 98% of genes and 99% of sites; R. limneticus, 94% of genes and 93% of sites). Another alignment

LetterreSeArCH

was generated without the picozoan MS584-11 and Telonema, leaving 151 taxa, 253 proteins, and 56,530 sites (151/253 dataset); R. marinus and R. limneticus were again well represented (R. marinus, 98% of genes and 99% of sites; R. limneticus, 94% of genes and 93% of sites).

Maximum-likelihood phylogenomic tree reconstruction was performed using IQ-TREE v.1.5.521 with the LG matrix combined with the C60 protein mixture model and four gamma categories (that is, LG + C60 + F + G4). The results of 1,000 ultrafast bootstrap and SH-aLRT replicates are reported as a measure of statistical support for bipartitions. Bayesian analyses were carried out by run-ning three independent Markov chain Monte Carlo chains with PhyloBayes MPI v.1.720, using an infinite mixture model and four discrete gamma cate-gories (CAT + GTR + G4). Chains were run for more than 7,200 generations for both the 153/253 and 151/253 datasets, and the first 1,500 generations were discarded as burn-in. As is frequently seen in large-scale phylogenomic analyses, the chains failed to converge in each case; for the 153/253 dataset, maxdiff = 1 and meandiff = 0.0128946, and for the 151/253 dataset, maxdiff = 1 and mean-diff = 0.0204329. Bayesian posterior probabilities are reported as a measure of statistical support for bipartitions.

The effect of fast-evolving sites on phylogenomic inference was examined by progressively removing the fastest-evolving sites in intervals of 3,000 and using RAxML (PROT + CAT + LG + F model) to generate 100 rapid-bootstrap replicates for each alignment. The fastest-evolving sites were identified using AgentSmith, based on the tree topologies generated in the IQ-TREE LG + C60 + F + G4 153- and 151-taxon trees. The support for the sister relationship of Rhodelphis and red algae was assessed for both trees; support for Archaeplastida monophyly versus Cryptista, glaucophytes and green plants (Cryptista and GG) was similarly tested for both datasets. Support for Picozoa, Rhodelphis and red algae was assessed only for the 153-taxon tree. In each case, the monophyly of Opisthokonta was tested as a positive control. Alternative tree topologies were generated from the maximum- likelihood trees using Mesquite v.3.545 and tested using the approximately unbiased test, as implemented in IQ-TREE v.1.5.521.

We carried out a number of analyses to test whether mixed gene ancestry could be affecting the phylogenetic analysis. RAxML43 was used to compute internode certainty scores6 by mapping bootstrapped single-gene trees (generated as above, except with the PROTCATLGF model) to the 151/253 maximum-likelihood phylogenomic tree (Extended Data Fig. 4b). We subsequently identified the 50 single- gene trees with the highest relative tree certainty (RTC) values with RAxML43 (average RTC score for all 253 single-gene trees is 0.175; average for best 50 is 0.362), and computed internode certainty scores for this subset of the data, as above. To test whether the 50 highest RTC single-gene datasets recover the sister relationship of Rhodelphis and red algae, we concatenated the alignments with SCaFOs v.1.2.544 and performed phylogenetic analysis in IQ-TREE using the LG + PMSF + G model along with 1,000 ultrafast bootstrap and SH-aLRT rep-licates. ASTRAL-III7 was used to calculate overall species trees from individual bootstrapped trees under a ‘coalescence’ framework, using default parameters and 100 multilocus bootstrap replicates.Identification of putative plastid-targeted proteins. A screen for Rhodelphis proteins bearing putative N-terminal plastid transit peptides was performed with TargetP46, with the ‘Plant’ flag activated. Proteins with a score corresponding to a 70% probability of plastid localization were retained for further manual inspection. Additionally, we queried the NCBI nr database with Rhodelphis protein sequences in search of proteins that were most similar to characterized plastidial or cyanobac-terial homologues, and queried the Rhodelphis genome and transcriptome assem-blies with proteins encoded by red algal plastid genomes. We considered proteins to potentially be plastidial if they fulfilled both of the above criteria, or if they were constituents of a metabolic unit of members which predominantly met the criteria.Plastid protein phylogenies. Proteins of R. marinus and R. limneticus identified as being putatively plastid-targeted were either added to existing alignments (FeS clus-ter and haem biosynthesis pathway proteins) or used as queries in a blastp search (E-value threshold of 1 × 10−5) against a comprehensive custom database, which contained representatives of plastid-bearing and non-plastidial groups (archaeplas-tids, cryptophytes, haptophytes, stramenopiles, dinoflagellates, chrompodellids, apicomplexans as well as opisthokonts, amoebozoans/apusozoans/ancyromonads and ciliates). For the TIC22 and TOC75 alignments, additional known homologues were used as queries, to extend the number of taxa obtained with R. marinus and R. limneticus. Results from blast searches were parsed for hits with a minimum query coverage of 50% and E < 1 × 10−25 or E < 1 × 10−5 (TIC/TOC and YCF proteins, respectively). The number of bacterial hits was constrained to 20 hits per phylum (for the Fibrobacteres–Chlorobi–Bacteroidetes group, most classes of Proteobacteria, the Planctomycetes–Verrucomicrobia–Chlamydiae group, Spirochaetes, Actinobacteria, Cyanobacteria (unranked) and Firmicutes) or 10 per phylum (remaining bacterial phyla) as defined by NCBI taxonomy. For initial tree reconstruction, corresponding sequences were aligned with MAFFT using the ‘--auto’ option, ambiguously aligned positions were trimmed off with trimAL

v.1.247 using a gap threshold of 20% and trees were calculated using FastTree48 with default options. Resulting phylogenies and underlying alignments were manually inspected, and obvious contaminations and paralogues were removed. Cleaned sequence files as well as the pre-existing alignments with added R. marinus/ R. limneticus sequences were filtered using prequel v.1.0.249 to remove stretches of non-homologous characters and then aligned using the G-INS-I algorithm of MAFFT in combination with VSM50, using an αmax of 0.6, to reduce over-alignment. Alignments were trimmed as above and final trees were reconstructed using IQTREE21, using ModelFinder51 to identify the best model for each alignment based on the Bayesian information criterion. Node supports were calculated with 1,000 ultrafast bootstrap (UFBoot) replicates52.Comparison of Rhodelphis gene repertoire to red algae. We used the KEGG Automatic Annotation Server53, using the bi-directional best hit, to functionally annotate Rhodelphis genes, along with genes encoded by the red algae Cyanidioschyzon merolae, Galdieria sulphuraria, Chondrus crispus and Porphyra purpureum. The resulting KEGG Orthology assignments were used to infer the overlap or difference in gene repertoire between Rhodelphis and red algae. Similarly, we used OrthoFinder v.2.0.054 to assess the overlap in orthologous gene sets between the same species (Extended Data Table 1c). Genome-based assessments of Rhodelphis trophic mode were done with R. marinus and R. limneticus Transdecoder proteins using PredictTrophicMode_Tool.R12.Reporting summary. Further information on research design is available in the Nature Research Reporting Summary linked to this paper.

Data availabilityRaw transcriptome and genome reads from R. limneticus and R. marinus are deposited in GenBank (PRJNA544719), along with full SSU rRNA gene sequences for R. marinus (MK966712) and R. limneticus (MK966713). Assembled tran-scriptomes and genomes, along with raw light and electron-microscopy images, individual gene alignments, concatenated and trimmed alignments, single-gene trees, and maximum-likelihood and Bayesian tree files for the 151-taxon and 153-taxon datasets have been deposited in Dryad (https://doi.org/10.5061/dryad.tr6d8q2). The family Rhodelphidae (urn:lsid:zoobank.org:act:80B5C004-2954-4A57- A411-482BCD29E85D), genus Rhodelphis (urn:lsid:zoobank.org:act:6D09D4D9-D9FC-4D0C-8FB2-55FD9DDEAD53) and species Rhodelphis limneticus (urn:lsid: zoobank.org:act:695ACD0B-8151-4609-97FC-A044A312BE22) and Rhodelphis marinus (urn:lsid:zoobank.org:act:84233191-4710-43D1-A2DA-914B8E7B7E01) have been registered with the Zoobank database (http://zoobank.org/).

Code availabilityAll unpublished code is available upon reasonable request from the corresponding authors. 22. Tikhonenkov, D. V., Mazeĭ, IuA. & Embulaeva, E. A. [Degradation succession of

heterotrophic flagellate communities in microcosms]. Zh. Obshch. Biol. 69, 57–64 (2008).

23. Tikhonenkov, D. V. et al. Description of Colponema vietnamica sp.n. and Acavomonas peruviana n. gen. n. sp., two new alveolate phyla (Colponemidia nom. nov. and Acavomonidia nom. nov.) and their contributions to reconstructing the ancestral state of alveolates and eukaryotes. PLoS ONE 9, e95467 (2014).

24. Luft, J. H. Improvements in epoxy resin embedding methods. J. Biophys. Biochem. Cytol. 9, 409–414 (1961).

25. Picelli, S. et al. Full-length RNA-seq from single cells using Smart-seq2. Nat. Protocols 9, 171–181 (2014).

26. Tikhonenkov, D. V., Janouškovec, J., Keeling, P. J. & Mylnikov, A. P. The morphology, ultrastructure and SSU rRNA gene sequence of a new freshwater flagellate, Neobodo borokensis n. sp. (Kinetoplastea, Excavata). J. Eukaryot. Microbiol. 63, 220–232 (2016).

27. Andrews, S. FastQC: a quality control tool for high throughput sequence data. version 0.10.1 https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (2010).

28. Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).

29. Grabherr, M. G. et al. Full-length transcriptome assembly from RNA-seq data without a reference genome. Nat. Biotechnol. 29, 644–652 (2011).

30. Altschul, S. F. et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402 (1997).

31. Li, W. & Godzik, A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22, 1658–1659 (2006).

32. Simão, F. A., Waterhouse, R. M., Ioannidis, P., Kriventseva, E. V. & Zdobnov, E. M. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31, 3210–3212 (2015).

33. Bankevich, A. et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J. Comput. Biol. 19, 455–477 (2012).

34. Miller, I. J. et al. Autometa: automated extraction of microbial genomes from individual shotgun metagenomes. Nucleic Acids Res. 47, e57 (2019).

35. Gurevich, A., Saveliev, V., Vyahhi, N. & Tesler, G. QUAST: quality assessment tool for genome assemblies. Bioinformatics 29, 1072–1075 (2013).

https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA544719

https://www.ncbi.nlm.nih.gov/nuccore/MK966712

https://www.ncbi.nlm.nih.gov/nuccore/MK966713

https://doi.org/10.5061/dryad.tr6d8q2

http://zoobank.org/

https://www.bioinformatics.babraham.ac.uk/projects/fastqc/

Letter reSeArCH

36. Wu, T. D. & Watanabe, C. K. GMAP: a genomic mapping and alignment program for mRNA and EST sequences. Bioinformatics 21, 1859–1875 (2005).

37. Crooks, G. E., Hon, G., Chandonia, J.-M. & Brenner, S. E. WebLogo: a sequence logo generator. Genome Res. 14, 1188–1190 (2004).

38. Ryan, J. F. Baa.pl: a tool to evaluate de novo genome assemblies with RNA transcripts. Preprint at https://arxiv.org/abs/1309.2087 (2013).

39. Gruber-Vodicka, H. R., Seah, B. K. B. & Pruesse, E. phyloFlash – rapid SSU rRNA profiling and targeted assembly from metagenomes. Preprint at https://www.biorxiv.org/content/10.1101/521922v1 (2019).

40. Burki, F. et al. Untangling the early diversification of eukaryotes: a phylogenomic study of the evolutionary origins of Centrohelida, Haptophyta and Cryptista. Proc. R. Soc. B 283, 20152802 (2016).

41. Katoh, K. & Standley, D. M. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol. Biol. Evol. 30, 772–780 (2013).

42. Criscuolo, A. & Gribaldo, S. BMGE (Block Mapping and Gathering with Entropy): a new software for selection of phylogenetic informative regions from multiple sequence alignments. BMC Evol. Biol. 10, 210 (2010).

43. Stamatakis, A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30, 1312–1313 (2014).

44. Roure, B., Rodriguez-Ezpeleta, N. & Philippe, H. SCaFoS: a tool for selection, concatenation and fusion of sequences for phylogenomics. BMC Evol. Biol. 7 (Suppl. 1), S2 (2007).

45. Maddison, W. P. & Maddison, D. R. Mesquite: a modular system for evolutionary analysis. Version 3.5 https://www.mesquiteproject.org/ (2018).

46. Emanuelsson, O., Nielsen, H., Brunak, S. & von Heijne, G. Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J. Mol. Biol. 300, 1005–1016 (2000).

47. Capella-Gutiérrez, S., Silla-Martínez, J. M. & Gabaldón, T. trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics 25, 1972–1973 (2009).

48. Price, M. N., Dehal, P. S. & Arkin, A. P. FastTree: computing large minimum evolution trees with profiles instead of a distance matrix. Mol. Biol. Evol. 26, 1641–1650 (2009).

49. Whelan, S., Irisarri, I. & Burki, F. PREQUAL: detecting non-homologous characters in sets of unaligned homologous sequences. Bioinformatics 34, 3929–3930 (2018).

50. Katoh, K. & Standley, D. M. A simple method to control over-alignment in the MAFFT multiple sequence alignment program. Bioinformatics 32, 1933–1942 (2016).

51. Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K. F., von Haeseler, A. & Jermiin, L. S. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat. Methods 14, 587–589 (2017).

52. Hoang, D. T., Chernomor, O., von Haeseler, A., Minh, B. Q. & Vinh, L. S. UFBoot2: improving the ultrafast bootstrap approximation. Mol. Biol. Evol. 35, 518–522 (2018).

53. Moriya, Y., Itoh, M., Okuda, S., Yoshizawa, A. C. & Kanehisa, M. KAAS: an automatic genome annotation and pathway reconstruction server. Nucleic Acids Res. 35, W182–W185 (2007).

54. Emms, D. M. & Kelly, S. OrthoFinder: solving fundamental biases in whole genome comparisons dramatically improves orthogroup inference accuracy. Genome Biol. 16, 157 (2015).

Acknowledgements We thank L. Nguyen-Ngoc, H. Doan-Nhu, E. S. Gusev, Y. Dubrovsky and the staff of the Russian-Vietnam Tropical Centre, Coastal Branch for assistance with sample collection and trip management; S. A. Karpov for assistance with interpretation of transmission electron microscopy images; Compute/Calcul Canada for computational resources, especially the Orcinus (Westgrid) and Mammouth Parallèle II (Calcul Québec) clusters. This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada to P.J.K. (grant 227301). Field work in Vietnam is part of the project ‘Ecolan 3.2’ of the Russian-Vietnam Tropical Centre. R.M.R.G. was supported by a fellowship from the Canadian Institutes of Health Research and a grant from the Tula Foundation to the Centre for Microbial Diversity and Evolution. Cell isolation and culturing, generation of material for sequencing, light and electron microscopy and analysis were supported by the Russian Science Foundation to D.V.T. (grant 18-14-00239). F.H. is supported by an EMBO fellowship (ALTF 1260–2016).

Author contributions R.M.R.G., D.V.T. and P.J.K. designed the study. D.V.T. isolated and cultured cells. D.V.T. and F.H. generated material for sequencing and F.H. removed contaminant sequences from the genome assembly. D.V.T. and A.P.M. performed microscopy experiments. R.M.R.G. performed phylogenomic, genomic and transcriptomic analyses. E.H. performed phylogenetic analysis of plastid proteins. R.M.R.G., D.V.T. and P.J.K. wrote the manuscript with input from all authors.

Competing interests The authors declare no competing interests.

Additional informationsupplementary information is available for this paper at https://doi.org/ 10.1038/s41586-019-1398-6.Correspondence and requests for materials should be addressed to R.M.R.G., D.V.T. or P.J.K.Peer review information Nature thanks Geoffrey McFadden and the other anonymous reviewer(s) for their contribution to the peer review of this work.Reprints and permissions information is available at http://www.nature.com/reprints.

https://arxiv.org/abs/1309.2087

https://www.biorxiv.org/content/10.1101/521922v1

https://www.biorxiv.org/content/10.1101/521922v1

https://www.mesquiteproject.org/

https://doi.org/10.1038/s41586-019-1398-6

https://doi.org/10.1038/s41586-019-1398-6

http://www.nature.com/reprints

http://www.nature.com/reprints

LetterreSeArCH

Extended Data Fig. 1 | Cell structure of R. limneticus. Related to Fig. 1. a, b, Scanning electron microscopy images showing the flagella and mastigonemes on the posterior flagellum. c, Section of an anterior flagellum. d, Section of a posterior flagellum. e–g, Arrangement of the transitional zone of a flagellum with transverse plate and cylinder. h, Wide microtubular band 2 accompanies the posterior flagellum. i–l, Cell sections from anterior to posterior. m, Single microtubules inside the cytoplasm. n, A rhizoplast connects the basal body of the posterior flagellum to the nucleus. o, Area of cell with nucleus, rudiments of glycostyles inside the vesicles and smooth endoplasmic reticulum. p, Contractile vacuole. q, Osmiophilic body. r, s, Phagocytosis of eukaryotic prey and bacteria. t, Cell section showing food vacuole with

several engulfed bacterial cells. bc, bacterium; cl, cylinder; cv, contractile vacuole; fv, food vacuole; gl, glycostyles; bb1, basal body of posterior flagellum; bb2, basal body of anterior flagellum; mn, mastigonemes; mrt, microtubule; n, nucleus; nmb, narrow microtubular band; ob, osmiophilic body; pf, posterior flagellum; pr, eukaryotic prey; rgl, rudiments of glycostyles; rp, rhizoplast; ser, sac of smooth endoplasmic reticulum; sm, single microtubule; ss, striated structure; st, satellite of basal body; tp, transverse plate; wmb1, wide microtubular band 1; wmb2, wide microtubular band 2. Scale bars, 5 μm (a, b), 0.5 μm (c, d, i–o), 0.2 μm (e–h, q), 1 μm (p) and 2 μm (r–t). These experiments were repeated three (a, b) and seven (c–t) times, with similar results.

Letter reSeArCH

Extended Data Fig. 2 | Cell structure of R. marinus. a, b, Living cells, obtained by light microscopy. c, Longitudinal section of the cell, d, Region of the cell surface with glycostyles. e, Basal body of posterior flagellum with outgoing fibrils and striated structure. f, Section of the flagellum covered with glycostyles and dark granules. g, Emergence of a posterior flagellum. h, Basal body of the posterior flagellum and mitochondrion with tubular cristae. i, Formation of rudiments of glycostyles in perinuclear space. j, Nucleus, mitochondrion and reserve substance.

k, Osmiophilic formation and microtubules. cr, cristae; dg, dark granules; fb, fibril; gl, glycostyles; bb1, basal body of posterior flagellum; bb2, basal body of anterior flagellum; mt, mitochondrion; mrt, microtubules; n, nucleus, of, osmiophilic (dark) formation; pf, posterior flagellum; rgl, rudiments of glycostyles; rs, reserve substance; ser, sac of smooth endoplasmic reticulum; ss, striated structure. Scale bars, 10 μm (a, b), 2 μm (c), 0.2 μm (d, e), 0.5 μm (f–i, k) and 1 μm (j). These experiments were repeated ten (a, b) and three (c–k) times, with similar results.

LetterreSeArCH

0.3

Euplotes

Cryptococcus neoformans

Bodo saltans

Schizosaccharomyces pombe

Chlorella vulgaris

Hartmannella vermiformis

Paramecium tetraurelia

Palpitomonas bilix

Emiliania huxleyi

Batrachochytrium dendrobatidis

Pelagomonas calceolata

Reticulomyxa filosa

Rhodomonas abbreviata


Monosiga brevicollis

Isochrysis galbana

Daphnia pulex

Chlorarachnion reptans

Voromonas pontica GG

Breviate

Pleurochrysis carterae

Guillardia theta

Polysphondylium pallidum


Chrysochromulina brevifilum

Hemiselmis rufescens

Raphidiophrys ambigua

Rhodomonas sp.

Ostreococcus lucimarinus

Percolomonas cosmopolitus MMETSP0758

Chattonella subsalsa

Phytophthora

Schizochytrium aggregatum

Chlamydomonas reinhardtii

Ectocarpus siliculosus

Tetrahymena thermophila

Sterkiella histriomuscorum

Elphidium sp.

Aureococcus anophageferrens

Madagascaria erythrocladoides

Thraustochytrium sp.

Erythrolobus

Hemiselmis andersenii

Pseudopedinella elasticaRhizochromulina marina

Sawyeria marylandensis

Rhodella maculata

Selaginella moellendorffii

Noctiluca scintillans

Mallomonas sp.

Platyophrya macrostoma

Ustilago maydis

Hematodinium sp

Tsukubamonas globosa

Thecamonas trahens

Naegleria gruberi

Acanthamoeba castellanii

Protocruzia adherens

Cryptophyceae sp.

Malawimonas

Arabidopsis


Pinguiococcus pyrenoidosus

Capsaspora owczarzaki

Rhodosorus marinus

Bigelowiella natans

Chrysochromulina polylepis

Lottia gigantea

Vaucheria litorea

Neurospora crassa

Saprolegnia parasitica

Phaeocystis antarctica

Brachypodium distachyon

Fibrocapsa japonica

Paraphysomonas imperforata

Physcomitrella patens

Cryptosporidium muris

Heterosigma akashiwo

Phycomyces blakesleeanus

Jakoba

Glaucocystis nostochinearum

Babesia bovis

Spumella elongata

Dictyostelium discoideum

Colponema vietnamica

Telonema subtilis

Branchiostoma floridae

Volvox carteri


Collodictyon sp.

Phaeomonas parva

Plasmodium falciparum

Eutreptiella gymnastica

Goniomonas pacifica

Astrolonche serrata

Cyanoptyche gloeocystis

Coccomyxa sp.

Karenia brevis

Scyphosphaera apsteinii

Neobodo designis

Aureoumbra lagunensis

Homo sapiens

Toxoplasma gondii

Prymnesium parvum

Dictyocha speculum

Phaeodactylum tricornutum

Acanthocystis sp.

Dictyostelium purpureum

Reclinomonas americana

Raphidiophrys heterophryoidea

Porphyra

Calcidiscus leptoporus

Bolidomonas pacifica

Litonotus pictus

Cyanophora paradoxa

Cafeteria roebergensis

Asterionellopsis glacialis

Amphidinium carterae

Micromonas sp.

Seculamonas ecuadoriensis

Aplanochytrium

Vitrella brassicaformis

Odontella aurita

Cryptomonas curvata

Raineriophrys erinaceoides

Cafeteria sp.

Perkinsus marinus

Gloeochaete witrockiana

Roombia truncata

Ochromonas sp. MMETSP1177

Gromia sphaerica

Mimulus guttatus

Danio rerio

Colpodella

Nematostella vectensis

Aurantiochytrium limacinum

Ancoracysta twista

Blastocystis hominis

Oryza sativa

Phaeocystis sp.

Goniomonas sp.

Choanocystis sp.

Nannochloropsis gaditana


Chondrus crispus


Chromera velia

Pavlovales sp.

Populus trichocarpa

Thalassiosira pseudonana

Picobiliphyte MS584-11

--

--

--

0.99

0.99

0.97

0.83

--

--

0.89

--

--

0.98

0.99

R. limneticusR. marinus

Stramenopiles

Alveolata

Rhizaria

Haptophyta

Centrohelida

Archaeplastida

Cryptista

Excavata

Obazoa

153/253 dataset (56,312 sites)PhyloBayes CAT + GTR + G4

a b

0.2

Emiliania huxleyi


Prymnesium parvum










Fibrocapsa japonica



Collodictyon sp.

Colpodella

Gromia sphaerica

Lottia gigantea

Roombia truncata



Oryza sativa

Goniomonas pacifica









Naegleria gruberi

Chondrus crispus


Coccomyxa sp.


Spumella elongata




Ustilago maydis

Chlorella vulgaris

Chromera velia


Picobiliphyte MS584-11

Neobodo designis


Perkinsus marinus

Dictyocha speculum

Isochrysis galbana


Pseudopedinella elastica


Toxoplasma gondii


Cyanophora paradoxa

Bodo saltans


Guillardia theta

Aplanochytrium

Choanocystis sp.

Thecamonas trahens

Homo sapiens




Palpitomonas bilix





Phytophthora

Mallomonas sp.







Cafeteria sp.


Micromonas sp.


Odontella aurita


Elphidium sp

Bigelowiella natans


Acanthocystis sp.


Hematodinium sp

Cryptophyceae sp.



Neurospora crassa


Porphyra


Malawimonas


Litonotus pictus

Astrolonche serrata

Daphnia pulex

Mimulus guttatus



Volvox carteri

Vaucheria litorea

Euplotes




Breviate


Jakoba

Telonema subtilis




Goniomonas sp.

Cryptomonas curvata


Karenia brevis




Populus trichocarpa

Rhizochromulina marina

Rhodomonas sp



Danio rerio


Erythrolobus

Rhodella maculataRhodosorus marinus

Pavlovales sp.

Arabidopsis





Babesia bovis


Ancoracysra twista


Reticulomyxa filosa


Phaeocystis sp.

Phaeomonas parva

99.8/100

--/74

81/98

99.7/95

97.8/100

--/72

--/--

90.5/100

99.9/82

100/75

--/83

--/--

--/98

99.8/100

73.9/96

73.4/94

99.8/97

99.8/100

92.5/99

98.4/99

100/97

99.8/100

100/81

--/82

82.8/98

--/77

70.1/78

--/94

99.4/100

97/97

95.6/100

100/99

--/95

93.1/98

--/91

--/--

97.3/99

87.9/96

87.6/80

87.3/--


Stramenopiles

Alveolata

Rhizaria

Archaeplastida

Cryptista

Archaeplastida

Haptophyta

Centrohelida

Excavata

Obazoa

153/253 dataset (56,312 sites)IQ-TREE LG + C60 + F + G4

0

25

50

75

100

0 10 20 30 40 50

% b

oots

trap

sup

port

Archaeplastida

Rhodelphis + reds

Cryptista + green and glauc.

Opisthokonta

Pico + Rhodelphis + reds

Thousand sites removed

c

Extended Data Fig. 3 | Phylogenomic analysis of the concatenated 153/253 dataset. a, Bayesian tree using the CAT + GTR evolutionary model as implemented in PhyloBayes. b, Maximum-likelihood tree using the LG + C60 + F + G4 model as implemented in IQ-TREE. Black dots denote full statistical support (Bayesian posterior probability = 1.0, maximum-likelihood ultrafast bootstrap and SH-aLRT = 100%); support values <0.7/70% are not shown (indicated by ‘–’). c, Bootstrap support for maximum-likelihood trees (PROT + CAT + LG + F) after progressive

removal of the fastest evolving amino acid sites shows that both the Rhodelphis and red algae and the picozoa, Rhodelphis and red algae relationships are relatively robust to data removal. Similar to the 151/253 dataset, support for Archaeplastida paraphyly (Cryptista, green plants and glaucophytes (green and glauc.)) decreases with data removal, whereas Archaeplastida monophyly support increases. Support for Opisthokonta monophyly serves as a control for the presence of sufficient information for phylogenomic inference.

Letter reSeArCH

0.3

Reticulomyxa filosa

Pavlovales sp.


Jakoba



Populus trichocarpa


Cafeteria sp.





Isochrysis galbana


Dictyostelium discoideumPolysphondylium pallidum


Danio rerio


Daphnia pulex

Neobodo designis


Goniomonas sp.


Ancoracysta twista

Elphidium sp.

Rhodomonas sp.

Aplanochytrium



Homo sapiens



Cyanophora paradoxa




Thecamonas trahens




Malawimonas

Choanocystis sp.

Karenia brevis

Vaucheria litorea

Roombia truncata

Mimulus guttatus



Rhodella maculata

Ustilago maydis




Odontella aurita


Chromera velia


Lottia gigantea




Palpitomonas bilix






Litonotus pictus



Bigelowiella natans


Fibrocapsa japonica




Phaeocystis sp.





Neurospora crassa


Phaeomonas parva



Euplotes





Acanthocystis sp.

Porphyra

Goniomonas pacifica


Volvox carteri

Chondrus crispus

Colpodella

Collodictyon sp.

Astrolonche serrata

Hematodinium sp.Noctiluca scintillans

Cryptomonas curvata



Toxoplasma gondii


Rhodosorus marinus

Batrachochytrium dendrobatidisPhycomyces blakesleeanus


Arabidopsis

Chlorella vulgaris

Dictyocha speculum

Phytophthora

Mallomonas sp

Spumella elongata


Perkinsus marinus



Cryptophyceae sp.



Babesia bovis




Gromia sphaerica

Prymnesium parvum

Breviate


Micromonas sp.


Oryza sativa

Bodo saltans


Emiliania huxleyi

Coccomyxa sp.

Erythrolobus






Naegleria gruberi

Guillardia theta

0.86

1

--

0.77

--

0.81

--

--

0.86

0.99

0.96

0.72

0.82

--

--

0.97

R. marinusR. limneticus

Stramenopiles

Alveolata

Rhizaria

Archaeplastida

Cryptista

Haptophyta

Centrohelida

Excavata

Obazoa

151/253 dataset (56,530 sites)PhyloBayes CAT + GTR + G4

a

0.2


Cryptophyceae sp.


Goniomonas pacifica

Breviate





Cyanoptyche gloeocystis Rhodomonas abbreviata


Gromia sphaerica

Colpodella

Volvox carteri

Rhodomonas sp.



Cyanophora paradoxa




Populus trichocarpa

Palpitomonas bilix

Prymnesium parvum



Reticulomyxa filosa


Oryza sativa






Bodo saltans

Goniomonas sp.

Malawimonas

Chromera velia

Euplotes

Rhodosorus marinus



Thecamonas trahens


Neurospora crassa


Phaeocystis sp.

Daphnia pulex

Astrolonche serrata


Phaeomonas parva


Glaucocystis nostochinearumGloeochaete witrockiana

Danio rerio

Micromonas sp


Roombia truncata

Coccomyxa sp


Bigelowiella natans

Dictyocha speculum

Choanocystis sp.Acanthocystis sp.





Perkinsus marinus


Homo sapiens



Hematodinium sp

Litonotus pictus


Arabidopsis




Karenia brevis

Erythrolobus



Spumella elongata


Cafeteria sp.


Elphidium sp.


Chlorella vulgaris



Phytophthora


Odontella aurita

Toxoplasma gondii


Thraustochytrium sp



Mallomonas sp.

Mimulus guttatus

Guillardia theta

Emiliania huxleyi

Pavlovales sp.







Neobodo designis






Rhodella maculata




Aplanochytrium

Collodictyon sp.

Babesia bovis


Porphyra



Cryptomonas curvata

Ustilago maydis


Jakoba

Lottia gigantea


Fibrocapsa japonica


Isochrysis galbana





Ancoracysta twista

Nannochloropsis gaditanaVaucheria litorea

Naegleria gruberi

Chondrus crispus


--/91

89/97

73.3/89

99.9/100

--/81

94.4/98

85.2/96

99.7/100

93.7/99

92.6/98

--/98

99.8/100

--/--

96.5/96

99.9/73

--/--

71.5/95

70.7/--

94.9/99

77.4/96

--/84

90.9/100

99.8/100

98.5/--

87/98

99.2/100

--/85

99.4/100

--/93

100/72

90.6/99

72.3/79

--/8899.9/100


Stramenopiles

Alveolata

Rhizaria

Haptophyta

Centrohelida

Archaeplastida

Cryptista

Archaeplastida

Excavata

Obazoa

151/253 dataset (56,530 sites)IQ-TREE LG + C60 + F + G4

b

Extended Data Fig. 4 | Phylogenomic analysis of the concatenated 151/253 dataset. a, Bayesian tree using the CAT + GTR evolutionary model as implemented in PhyloBayes. b, Maximum-likelihood tree using the LG + C60 + F + G4 model as implemented in IQ-TREE. Black dots denote full statistical support (Bayesian posterior probability = 1.0, maximum-likelihood ultrafast bootstrap and SH-aLRT = 100%); support values <0.7/70% are not shown.

LetterreSeArCH

0.225,0.185

0.069,0.078

0.133,0.117

0.221,0.1620.704,0.704 Bodo saltans

Neobodo designisEutreptiella gymnastica

0.216,0.252

0.083,0.083 Naegleria gruberiSawyeria marylandensis

0.595,0.595

Percolomonas cosmopolitus 758Percolomonas cosmopolitus 759Tsukubamonas globosa

0.141,0.2720.000,0.122 Jakoba

Seculamonas ecuadoriensisReclinomonas americana

-0.994,-0.125

-1.000,-0.131

-0.993,-0.108

-1.000,-0.108

-0.959,-0.122

0.328,0.354

0.346,0.371

0.958,0.958

0.299,0.3270.034,0.050 Arabidopsis

Populus trichocarpaMimulus guttatus

0.998,0.998 Brachypodium distachyonOryza sativa

0.098,0.195 Physcomitrella patensSelaginella moellendorffii

0.312,0.340

0.573,0.573

0.999,0.999 Chlamydomonas reinhardtiiVolvox carteri

-0.002,-0.001Chlorella vulgarisCoccomyxa sp.

0.724,0.724 Micromonas sp.Ostreococcus lucimarinus

0.122,0.183

0.117,0.211 Cyanophora paradoxa

0.083,0.083

Glaucocystis nostochinearumGloeochaete witrockianaCyanoptyche gloeocystis

0.045,0.063

0.235,0.189

0.286,0.316

1.000,1.000

Cryptomonas curvata

0.237,0.272

0.263,0.295

0.036,0.052 Cryptophyceae sp.

0.041,0.091

Rhodomonas abbreviataRhodomonas sp.Guillardia theta

0.992,0.992 Hemiselmis anderseniiHemiselmis rufescens

0.981,0.981 Goniomonas pacificaGoniomonas sp.Roombia truncataPalpitomonas bilix

0.018,0.108

0.268,0.299

0.166,0.141

0.103,0.1300.090,0.132 Chondrus crispus

PorphyraRhodella maculata

0.001,0.019

0.006,0.039

0.191,0.137 Compsopogon coeruleusMadagascaria erythrocladoides

0.310,0.310

Erythrolobus

0.001,0.045

Porphyridium aerugineumPorphyridium cruentumRhodosorus marinus

0.075,0.128 Cyanidioschyzon merolaeGaldieria sulphuraria

0.568,0.568 Rhodelphis marinusRhodelphis limneticus

-0.039,-0.052

-0.031,-0.030

0.582,0.582

0.859,0.859

0.098,0.066

0.064,0.075

-0.021,-0.0260.119,0.081 Calcidiscus leptoporus

Pleurochrysis carteraeScyphosphaera apsteinii

0.939,0.939 Emiliania huxleyiIsochrysis galbana

0.420,0.342Chrysochromulina brevifilum

0.718,0.718

Chrysochromulina polylepisPrymnesium parvum

0.988,0.988 Phaeocystis antarcticaPhaeocystis sp.Pavlovales sp.

0.587,0.587Acanthocystis sp.

0.090,0.132

Choanocystis sp.

-0.272,-0.087


0.331,0.331

Raphidiophrys ambiguaRaphidiophrys heterophryoideaAncoracysta twista

-0.998,-0.128

0.077,0.148

0.280,0.212

0.280,0.212Astrolonche serrata

0.999,0.999Elphidium sp.Reticulomyxa filosaGromia sphaerica

0.945,0.945 Bigelowiella natansChlorarachnion reptans

-0.989,-0.108

0.172,0.145

0.006,0.077

0.977,0.977Aplanochytrium

0.598,0.598


0.160,0.201Schizochytrium aggregatumThraustochytrium sp.

0.129,0.116

0.227,0.186

-0.997,-0.129

0.366,0.389

0.363,0.3860.126,0.171

0.292,0.303 Asterionellopsis glacialisPhaeodactylum tricornutumOdontella auritaThalassiosira pseudonanaBolidomonas pacifica

0.156,0.134

0.335,0.3600.270,0.270 Aureococcus anophageferrens

Pelagomonas calceolataAureoumbra lagunensis

0.306,0.334

Dictyocha speculum

0.145,0.276Pseudopedinella elasticaRhizochromulina marina

-0.278,-0.069

-0.488,-0.101

0.032,0.053

0.268,0.2820.964,0.964 Chattonella subsalsa

Heterosigma akashiwoFibrocapsa japonica

0.166,0.295

Ectocarpus siliculosusVaucheria litoreaNannochloropsis gaditana

-0.127,-0.044

0.354,0.379Mallomonas sp.

0.129,0.174

0.128,0.128 Ochromonas sp. 1177Spumella elongataParaphysomonas imperforata

0.500,0.500 Phaeomonas parvaPinguiococcus pyrenoidosus

0.973,0.973 PhytophthoraSaprolegnia parasitica

0.048,0.119-0.054,-0.077 Blastocystis hominis

Cafeteria sp.Cafeteria roebergensis

0.110,0.100


0.096,0.091

0.237,0.234

0.098,0.144

0.092,0.2280.862,0.862 Euplotes

Sterkiella histriomuscorumLitonotus pictus

0.593,0.593

0.257,0.290 Paramecium tetraureliaTetrahymena thermophilaPlatyophrya macrostomaProtocruzia adherens

0.089,0.135

0.105,0.101

0.084,0.0880.050,0.067

0.065,0.112 Babesia bovisPlasmodium falciparumToxoplasma gondiiCryptosporidium muris

0.097,0.164

0.105,0.170Chromera velia

0.029,0.080ColpodellaVoromonas pontica GGVitrella brassicaformis

0.239,0.273

Perkinsus marinus

0.357,0.357

Hematodinium sp.

0.333,0.333


0.049,0.031Karenia brevisAmphidinium carterae

-0.082,-0.030

0.037,0.053

0.216,0.253


0.098,0.093

0.090,0.060

0.832,0.832 Cryptococcus neoformansUstilago maydis

0.090,0.059Neurospora crassaSchizosaccharomyces pompePhycomyces blakesleeanus

0.120,0.106

0.199,0.163

0.245,0.262

0.027,0.089

0.086,0.125Branchiostoma floridae

0.388,0.388Danio rerioHomo sapiens

-0.046,-0.056 Daphnia pulexLottia giganteaNematostella vectensisMonosiga brevicollisCapsaspora owczarzaki

0.000,0.069 BreviateThecamonas trahens

0.112,0.137

-0.994,-0.679 Collodictyon sp.Malawimonas

0.510,0.375

1.000,1.0000.736,0.736 Dictyostelium discoideum

Dictyostelium purpureumPolysphondylium pallidum

0.497,0.497 Hartmannella vermiformisAcanthamoeba castellanii

0.609,0.609

0.096,0.096

0.993,0.993 Eutreptiella gymnastica

1.000,1.000Bodo saltansNeobodo designis

0.100,0.235

0.128,0.128

0.996,0.996 Percolomonas cosmopolitus 759Percolomonas cosmopolitus 758

-0.997,-0.543Naegleria gruberiSawyeria marylandensis

-0.986,-0.986

0.999,0.999 Seculamonas ecuadoriensisJakobaReclinomonas americanaTsukubamonas globosa

-0.994,-0.379

-0.993,-0.418

0.612,0.612

0.773,0.773

0.774,0.774

1.000,1.000

0.993,0.993Guillardia theta

-0.039,-0.039

Cryptophyceae sp.

-0.857,-0.857

Rhodomonas abbreviataRhodomonas sp.

-0.995,-0.6801.000,1.000 Hemiselmis rufescens

Hemiselmis anderseniiCryptomonas curvata

1.000,1.000 Goniomonas sp.Goniomonas pacificaRoombia truncataPalpitomonas bilix

-0.993,-0.379

-0.993,-0.378

1.000,1.000

0.773,0.773

0.127,0.127

1.000,1.000 Volvox carteriChlamydomonas reinhardtii

-0.238,-0.289

Coccomyxa spChlorella vulgaris

0.992,0.992Ostreococcus lucimarinusMicromonas sp

1.000,1.000

0.110,0.110 Selaginella moellendorffiiPhyscomitrella patens

0.987,0.987

0.773,0.773-0.153,-0.153 Arabidopsis

Populus trichocarpaMimulus guttatus

1.000,1.000 Brachypodium distachyonOryza sativa

0.996,0.996

0.991,0.991 Cyanophora paradoxaGlaucocystis nostochinearum

0.678,0.678Cyanoptyche gloeocystisGloeochaete witrockiana

0.110,0.110

1.000,1.000

0.993,0.993

-0.666,-0.2870.096,0.096 Chondrus crispus

PorphyraRhodella maculata

-0.081,-0.081

Rhodosorus marinus

-0.082,-0.082

1.000,1.0000.082,0.082 Porphyridium cruentum

Porphyridium aerugineumErythrolobus

0.758,0.758

Compsopogon coeruleusMadagascaria erythrocladoides

-0.674,-0.674 Galdieria sulphurariaCyanidioschyzon merolae

0.128,0.128 Rhodelphis limneticusRhodelphis marinus

-0.994,-0.379

-0.039,-0.039

-0.039,-0.039

1.000,1.000

1.000,1.000

-0.496,-0.496

0.515,0.515

-0.365,-0.365-0.338,-0.338 Pleurochrysis carterae

Calcidiscus leptoporusScyphosphaera apsteinii

1.000,1.000 Isochrysis galbanaEmiliania huxleyi

0.520,0.5200.991,0.991 Chrysochromulina polylepis

Prymnesium parvumChrysochromulina brevifilum

1.000,1.000 Phaeocystis antarcticaPhaeocystis sp.Pavlovales sp.

1.000,1.000

0.082,0.082

-0.674,-0.6740.997,0.997 Raphidiophrys heterophryoidea

Raphidiophrys ambiguaRaineriophrys erinaceoidesChoanocystis sp.Acanthocystis sp.Ancoracysta twista

-0.998,-0.380

-0.371,-0.167

0.124,0.124

1.000,1.000 Bigelowiella natansChlorarachnion reptans

-0.947,-0.947

-0.945,-0.945 Astrolonche serrata

0.876,0.876Reticulomyxa filosaElphidium spGromia sphaerica

0.124,0.124

-0.497,-0.511

0.612,0.612

1.000,1.000 PhytophthoraSaprolegnia parasitica

0.622,0.622

-0.497,-0.218

-0.496,-0.218

0.987,0.987

0.127,0.1270.082,0.082 Pseudopedinella elastica

Rhizochromulina marinaDictyocha speculum

1.000,1.000

1.000,1.000 Pelagomonas calceolataAureococcus anophageferrensAureoumbra lagunensis

0.096,0.096

1.000,1.000Bolidomonas pacifica

1.000,1.000

0.703,0.703 Thalassiosira pseudonanaOdontella aurita

0.612,0.612

Asterionellopsis glacialisPhaeodactylum tricornutum

1.000,1.000 Phaeomonas parvaPinguiococcus pyrenoidosus

0.110,0.110

0.998,0.998 Ectocarpus siliculosusVaucheria litorea

0.110,0.110

1.000,1.000 Chattonella subsalsaHeterosigma akashiwoFibrocapsa japonica

-0.081,-0.0811.000,1.000

0.100,0.1000.773,0.773 Spumella elongata

Ochromonas sp. 1177Mallomonas sp.Paraphysomonas imperforataNannochloropsis gaditana

1.000,1.000

0.128,0.128 Aurantiochytrium limacinum

-0.869,-0.721

Schizochytrium aggregatumThraustochytrium sp.Aplanochytrium

0.772,0.772-0.082,-0.218 Cafeteria sp.

Blastocystis hominisCafeteria roebergensis

0.110,0.110


0.082,0.218

0.100,0.100

0.622,0.622

0.999,0.999

1.000,1.000 Karenia brevis

-0.669,-0.669Amphidinium carteraeNoctiluca scintillansHematodinium sp.Perkinsus marinus

0.128,0.128

0.096,0.096

0.082,0.082 Toxoplasma gondii

0.760,0.760

Plasmodium falciparumBabesia bovisCryptosporidium muris

0.124,0.124

0.124,0.124 Chromera velia

0.980,0.980

ColpodellaVoromonas pontica GGVitrella brassicaformis

0.096,0.096


0.096,0.096

0.096,0.0960.290,0.290 Sterkiella histriomuscorum

EuplotesLitonotus pictus

0.993,0.993


0.773,0.773

Paramecium tetraureliaTetrahymena thermophila

-0.082,-0.218

-0.057,-0.057

0.124,0.124

0.987,0.987

0.996,0.996

0.096,0.232

0.082,0.218 Branchiostoma floridae

1.000,1.000

Homo sapiensDanio rerio

0.316,0.316

Daphnia pulexLottia giganteaNematostella vectensis


1.000,1.000Batrachochytrium dendrobatidis

0.099,0.099


0.609,0.609

1.000,1.000 Cryptococcus neoformansUstilago maydis

0.609,0.609

Neurospora crassaSchizosaccharomyces pompe

-0.999,-0.682 Thecamonas trahensBreviate

0.086,0.222

0.086,0.222

0.128,0.1281.000,1.000

0.920,0.920 Dictyostelium purpureumDictyostelium discoideumPolysphondylium pallidumHartmannella vermiformis

-1.000,-0.139 Acanthamoeba castellaniiCollodictyon sp.Malawimonas


a b

Extended Data Fig. 5 | Internode certainty analyses for the 151/253 and 151/50 datasets. Internode certainty and internode certainty-all scores were calculated with RAxML v.8.1.6 and mapped onto the 151/253 maximum-likelihood tree topology presented in Extended Data Fig. 4b. a, Internode certainty and internode certainty-all scores for 253 individual bootstrapped maximum-likelihood trees used to generate the concatenated alignment for Extended Data Fig. 4b. b, Internode certainty

and internode certainty-all scores for the 50 trees among the 253-tree dataset that have the highest RTC scores, which are expected to improve the robustness of phylogenomic analyses. Internode certainty scores for the sister relationship of Rhodelphis and red algae are higher for the 50 best-supported trees, indicating that they also support this relationship, and with fewer conflicting signals.

Letter reSeArCH

0.2


Schizosaccharomyces pompe

Litonotus pictus

Choanocystis sp.




Acanthocystis sp.

Prymnesium parvum






Hematodinium sp.

Neurospora crassa

Pavlovales sp.




Phaeomonas parva


Thecamonas trahens



Cryptophyceae sp.






Goniomonas pacifica

Jakoba

Emiliania huxleyi

Coccomyxa sp.

Cryptomonas curvata


Erythrolobus








Rhodosorus marinus

Naegleria gruberi

Perkinsus marinus

Roombia truncata


Arabidopsis



Reticulomyxa filosa

Populus trichocarpa




Babesia bovis

Aplanochytrium



Lottia gigantea

Goniomonas sp.




Cafeteria sp.

Breviate

Ancoracysta twista

Chromera velia



Cyanophora paradoxa


Astrolonche serrata


Elphidium sp



Phytophthora

Oryza sativa


Volvox carteri


Micromonas sp.

Phaeocystis sp.




Palpitomonas bilix

Fibrocapsa japonica




Mimulus guttatus

Chlorella vulgaris




Dictyocha speculum






Bodo saltans

Porphyra

Neobodo designis




Rhodomonas sp.


Homo sapiens

Mallomonas sp.

Rhodella maculata

Malawimonas


Collodictyon sp.


Chondrus crispus

Danio rerio


Isochrysis galbana






Colpodella


Guillardia theta

Vaucheria litorea

Gromia sphaerica

Odontella aurita

Toxoplasma gondii


Spumella elongata



Bigelowiella natans

Karenia brevis

Euplotes


Ustilago maydis


Daphnia pulex


99.6/100

99.5/100

100/95

89.3/100

99.6/100

82/96

92.5/98

100/96

--/76

98.7/--

76.2/87

100/95

95.2/80

88.4/98

99/99

92.2/87

--/8599.5/100

99.6/100

--/--

89.1/99

98.3/87

95.4/100

--/89

99.7/100

--/97

99.6/100

90.3/99

81.6/91

99.9/100

--/80

92.9/95

89.8/--

92.1/99

--/72

--/79

84.9/99

83.2/99

--/70

--/71

100/95

99.8/100

--/--

97.5/--

99.9/99

--/93

95.4/92

86.1/98

99.3/79

Rhodelphis marinusRhodelphis limneticus

Stramenopiles

Rhizaria

Alveolata

Haptophyta

Centrohelida

Archaeplastida

Cryptista

Excavata

Obazoa

151/50 dataset (21,886 sites)IQ-TREE LG + PMSF + G

Extended Data Fig. 6 | Phylogenomic analysis based on concatenation of 50 single-gene datasets with highest RTC scores. Maximum-likelihood tree using the LG + PMSF + G model as implemented in IQ-TREE (151 taxa, 50 proteins, 21,886 sites). Black dots denote full

statistical support (maximum-likelihood ultrafast bootstrap and SH-aLRT = 100%); support values <0.7/70% are not shown. The sister relationship of Rhodelphis and red algae still receives full statistical support with a highly reduced, phylogenetically well-supported dataset.

LetterreSeArCH

2.0

Reticulomyxa filosa


Toxoplasma gondii



Volvox carteri

Ancoracysta twista

Gromia sphaerica




Breviate




Cafeteria sp.


Goniomonas pacifica

Colpodella



Chromera velia

Mimulus guttatus



Homo sapiens


Dictyocha speculum



Porphyra





Naegleria gruberi

Euplotes

Bodo saltans



Perkinsus marinus

Micromonas sp.

Lottia gigantea

Aplanochytrium

Neurospora crassa





Thecamonas trahens



Arabidopsis

Rhodella maculata

Isochrysis galbana


Goniomonas sp.

Chrysochromulina brevifilumPrymnesium parvum



Malawimonas


Collodictyon sp.



Roombia truncata

Phaeomonas parva


Cryptophyceae sp.

Ustilago maydis


Chondrus crispus

Rhodomonas sp.

Cryptomonas curvata


Populus trichocarpa

Palpitomonas bilix




Spumella elongata

Heterosigma akashiwoFibrocapsa japonica

Phaeocystis sp.



Oryza sativa



Vaucheria litorea

Mallomonas sp.



Elphidium sp.

Acanthocystis sp.


Guillardia theta

Polysphondylium pallidumDictyostelium purpureum

Emiliania huxleyi




Chlorella vulgaris




Coccomyxa sp.

Bigelowiella natans


Neobodo designis




Astrolonche serrata

Jakoba






Choanocystis sp.



Daphnia pulex

Phytophthora


Pavlovales sp.


Karenia brevis

Erythrolobus

Rhodosorus marinus



Danio rerio

Babesia bovis






Hematodinium sp.


Litonotus pictus



Odontella aurita

Cyanophora paradoxa

0.93

0.89

--

--

--

0.98

--

0.88

--

0.85

0.96

--

--

--

--

--

0.95

0.85

0.98

0.9

--

--

--

0.76

0.95

0.97

0.8

0.88

--

--

0.99

--

--

0.89

0.86

0.76

0.99

--

--

--

--

--

0.7

0.76

0.76


2.0


Cafeteria sp.


Babesia bovis


Cryptomonas curvata



Neobodo designis


Chondrus crispus



Rhodomonas sp.



Goniomonas sp.


Cryptophyceae sp.




Lottia gigantea





Guillardia theta

Euplotes



Odontella aurita

Hematodinium sp.



Jakoba




Daphnia pulex

Ustilago maydis







Perkinsus marinus

Phaeocystis sp

Cyanophora paradoxa


Collodictyon sp.



Astrolonche serrata


Fibrocapsa japonica

Litonotus pictus



Homo sapiens





Porphyra

Emiliania huxleyi

Rhodella maculata





Gromia sphaerica

Vaucheria litorea


Mallomonas sp.

Acanthocystis sp.


Phytophthora


Naegleria gruberi

Populus trichocarpa


Chromera velia

Elphidium sp.

Thecamonas trahens

Coccomyxa sp.




Rhodosorus marinus





Colpodella

Choanocystis sp.

Phaeomonas parva

Palpitomonas bilix

Neurospora crassa


Isochrysis galbana


Malawimonas


Oryza sativa






Ancoracysta twista


Reticulomyxa filosa

Micromonas sp.




Toxoplasma gondii

ArabidopsisMimulus guttatus

Goniomonas pacifica


Pavlovales sp.


Volvox carteri

Bodo saltans

Spumella elongata

Chlorella vulgaris

Danio rerio




Erythrolobus

Prymnesium parvum





Aplanochytrium





Roombia truncata

Karenia brevis

Dictyocha speculum

Breviate

Bigelowiella natans

--

0.98

0.94

0.99

0.75

0.7

--

--

0.96

0.91

0.74

0.77

--

--

0.93

--

--

--

0.99

0.75

0.73

--

0.88

--

0.79

--

0.73

0.99

0.83

--

0.72

--

--

--

0.84

0.83

--

0.81

--

--

--

--

--

--

0.81

--

0.76

0.84

--

0.98

--

--

0.81

--

--


a b

ASTRAL-III species tree253 ML gene trees151 taxa

ASTRAL-III species tree50 ML gene trees151 taxa

Extended Data Fig. 7 | A coalescence phylogenomic framework recovers Rhodelphis as sister to red algae based on individual gene trees. Individual bootstrapped gene trees were generated with RAxML v.8.1.6 and used to generate a species tree with ASTRAL-III under default parameters and 100 bootstrap replicates. Support values <0.7/70% are

not shown. a, Species trees were made from all 253 single-gene trees from the 151/253 dataset. b, Species trees were made from the 50 trees with the highest relative tree certainty. The sister relationship of Rhodelphis and red algae is recovered with both datasets, and is consistent with concatenated phylogenomic analyses.

Letter reSeArCH

S. moe

llend

orffii

P. pat

ens

C. mer

olae

P. abie

s

O. sat

iva

B. dist

achy

on

A. tha

liana

M. g

utta

tus

P. sor

ghi

S. cer

evisi

ae

S. pom

be

C. cor

onat

us

Chlore

lla sp

.

C. rein

hard

tii

V. car

teri

N. cali

forn

iae

R. irre

gular

is

B. den

drob

atidi

s

A. mac

rogy

nus

A. cas

tella

nii

A. sub

globo

sum

D. disc

oideu

m

D. mela

noga

ster

M. m

uscu

lus

T. tra

hens

B. nat

ans

C. tob

in

F. alba

B. salt

ans

R. filo

sa

P. tet

raur

elia

T. the

rmop

hila

C. tet

ram

itofo

rmis

R. lim

neticu

s

R. mar

inus

GO.0001503_ossification_8GO.0001501_skeletal.system.development_12GO.0018209_peptidyl.serine.modification_7GO.0019692_deoxyribose.phosphate.metabolic.process_5GO.0008015_blood.circulation_10GO.0007517_muscle.organ.development_8GO.0051298_centrosome.duplication_7GO.0019835_cytolysis_5GO.9999999_noGOannot_75GO.0045936_negative.regulation.of.phosphate.metabolic.process_16GO.0007601_visual.perception_8GO.0002790_peptide.secretion_18GO.0051896_regulation.of.protein.kinase.B.signaling_4GO.0051260_protein.homooligomerization_9GO.0022610_biological.adhesion_42GO.0000910_cytokinesis_11GO.0030097_hemopoiesis_17GO.0001822_kidney.development_8GO.0061448_connective.tissue.development_6GO.0006915_apoptotic.process_42GO.0060322_head.development_26GO.0051336_regulation.of.hydrolase.activity_34GO.0031152_aggregation.involved.in.sorocarp.development_9GO.0046835_carbohydrate.phosphorylation_3GO.1903829_positive.regulation.of.cellular.protein.localization_8GO.0042060_wound.healing_15GO.0060429_epithelium.development_38GO.0009880_embryonic.pattern.specification_6GO.1902904_negative.regulation.of.supramolecular.fiber.organization_8GO.1902905_positive.regulation.of.supramolecular.fiber.organization_8GO.0040011_locomotion_49GO.0001932_regulation.of.protein.phosphorylation_25GO.0035264_multicellular.organism.growth_7GO.0033299_secretion.of.lysosomal.enzymes_5GO.0007423_sensory.organ.development_15GO.0048666_neuron.development_41GO.0007040_lysosome.organization_6GO.0051495_positive.regulation.of.cytoskeleton.organization_11GO.0007399_nervous.system.development_77GO.0001894_tissue.homeostasis_13GO.0005975_carbohydrate.metabolic.process_42GO.0032388_positive.regulation.of.intracellular.transport_7GO.0033036_macromolecule.localization_125GO.0032743_positive.regulation.of.inter leukin.2.production_3GO.0010906_regulation.of.glucose.metabolic.process_4GO.0030218_erythrocyte.differentiation_5GO.0030587_sorocarp.development_16GO.0010212_response.to.ionizing.radiation_9GO.0045494_photoreceptor.cell.maintenance_8GO.0007409_axonogenesis_14GO.0030041_actin.filament.polymerization_13GO.0019915_lipid.storage_5GO.0051017_actin.filament.bundle.assembly_7GO.0030036_actin.cytoskeleton.organization_34GO.0009612_response.to.mechanical.stimulus_7GO.0009166_nucleotide.catabolic.process_12GO.0008361_regulation.of.cell.size_7GO.0031647_regulation.of.protein.stability_6GO.0045666_positive.regulation.of.neuron.differentiation_14GO.0010770_positive.regulation.of.cell.morphogenesis.involved.in.differentiation_8GO.0010506_regulation.of.autophagy_10GO.0050767_regulation.of.neurogenesis_24GO.0034097_response.to.cytokine_14GO.0071902_positive.regulation.of.protein.serine.threonine.kinase.activity_8GO.0007041_lysosomal.transport_7GO.0061462_protein.localization.to.lysosome_8GO.0016057_regulation.of.membrane.potential.in.photoreceptor.cell_1GO.0033132_negative.regulation.of.glucokinaseGO.0034394_protein.localization.to.cell.surface_3GO.2000058_regulation.of.ubiquitin.dependent.protein.catabolic.process_6GO.1901031_regulation.of.response.to.reactive.oxygen.speciesGO.0046683_response.to.organophosphorus_5GO.0010507_negative.regulation.of.autophagy_3GO.0055070_copper.ion.homeostasis_2GO.0098876_vesicle.mediated.transport.to.the.plasma.membrane_9GO.0014074_response.to.purine.containing.compound_6GO.0008045_motor.neuron.axon.guidance_3GO.0045806_negative.regulation.of.endocytosis_3GO.0046952_ketone.body.catabolic.process_1GO.0044655_phagosome.reneutralization_3

0 0.2 0.4 0.6 0.8 1

Value

Color Key

c

R. limneticus

−1

0

1

−4 −2 0 2

PC1 (73%)

PC

2 (4

%)

Groups

Rhodelphis

R. marinus Phagocytotic eukaryote

Phagomixotrophic eukaryote

a

R. limneticus

−1

0

1

−4 −2 0 2

PC1 (73%)

PC

2 (4

%)

Groups

Rhodelphis

R. marinusNon-photosynthetic eukaryote

Phagomixotrophic eukaryote

Photosynthetic eukaryote

b

Non-phagocytotic eukaryote

PhagocytesNon-phagocytes

Extended Data Fig. 8 | Genomic support for Rhodelphis as non-photosynthetic phagotrophs. a, b, Principal component (PC) plots of gene ontology (GO) category score from free-living phagocytes (a) and photosynthetic organisms (b). a, Rhodelphis associate with phagocytes, but not with photosynthetic eukaryotes. Dashed ellipses represent 95% confidence intervals based on training datasets using a model defined

by free-living phagocytes (a; n = 86 GO categories, 474 proteins) and photosynthesis model (b; n = 37 GO categories, 243 proteins). c, Heat map of phagocyte GO terms showing (as in a) that Rhodelphis gene repertoires are similar to phagocytes. Analyses were performed using PredictTrophicMode_Tool.R.

LetterreSeArCH

Extended Data Fig. 9 | Rhodelphis encode plastid-targeted proteins with N-terminal targeting sequences and homologues of the TIC/TOC plastid import system. a, An alignment of plastid-type chaperonin 60 (related to Fig. 3b) shows that Rhodelphis nuclear genomes encode plastid-targeted proteins that have clear N-terminal extensions (cTP) relative to plastid-encoded red algal homologues (names in red) and cyanobacterial

homologues (names in cyan), but lack a signal peptide (SP) characteristic of proteins targeted to complex secondary or tertiary plastids, as found in Plasmodium (orange). b–e, Rhodelphis nuclear genomes encode bona fide homologues of plastid protein import subunits TIC20, TIC32, TIC22 and TOC75, which are specific genetic markers for plastid presence.

Letter reSeArCH

extended data table 1 | summary of Rhodelphis genome and transcriptome data

a, Assembly statistics for the R. limneticus WGA dataset. HiSeq X 150-bp paired-end and MinION reads were assembled with SPAdes v.3.11.1, the assembly (scaffolds) was decontaminated using Autometa and assessed with QUAST v.5.0.2. N50 and N75 values refer to the shortest contig lengths that make up 50% and 75% of the genome assembly, respectively. L50 and L75 refer to the smallest number of contigs that account for 50% and 75% of the assembly length, respectively. b, Summary of BUSCO reports for both Rhodelphis species based on decontaminated transcriptome data. R. limneticus results are from a combination of single-cell and culture datasets. Percentage of transcripts mapping to the R. limneticus genome was determined with isoblat and introns were inferred from GMAP alignments. High-quality R. marinus genome data were not generated before the death of the culture. c, Orthologous group (OG) distribution across red algae and Rhodelphis. OrthoFinder v.2.0.0 was used to infer orthologous groups between proteins inferred from R. marinus and R. limneticus transcriptomes and red algal genomes. Rhodelphis genomes are gene-rich in comparison to published red algal genomes, and retain components of evolutionarily conserved structures (such as flagella and centrioles) that have been lost in red algae. Species is abbreviated as ‘spec’.

1

nature research | reporting summ

aryO

ctober 2018

Corresponding author(s): Ryan Gawryluk

Last updated by author(s): May 26, 2019

Reporting SummaryNature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.

StatisticsFor all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.

n/a Confirmed

The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement

A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly

The statistical test(s) used AND whether they are one- or two-sided Only common tests should be described solely by name; describe more complex techniques in the Methods section.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons

A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings

For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes

Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated

Our web collection on statistics for biologists contains articles on many of the points above.

Software and codePolicy information about availability of computer code

Data collection Sequence quality: FastQC v0.10.1 Transcriptome assembly: Trimmomatic v0.32;Trinity v2.0.6;Transdecoder v5.0.2;CD-HIT v4.6 Genome assembly: SPAdes v3.11.1

Data analysis Transcriptome and genome analysis: QUAST v5.0.2; GMAP v2016-08-16; Autometa, cloned Jan 29, 2019 ( https://bitbucket.org/jason_c_kwan/autometa); phyloFlash v3.0;R v3.5.0;BUSCO v3.0.1;BLAST+ v2.2.30;WebLogo v3 (http://weblogo.threeplusone.com/create.cgi);isoblat v0.3 Comparative genomics: OrthoFinder v2.0.0;KEGG Automatic Annotation Server (KAAS) (https://www.genome.jp/kegg/kaas/);TargetP v1.1 Phylogenomic and phylogenetic: mesquite v3.5;trimAL v1.2;MAFFT v7.212;SCaFOs v1.2.5;BMGE v1.12;RAxML v8.1.6;FastTree v2.1.7 SSE3;IQ-TREE v1.5.5;PhyloBayes MPI v1.7;AgentSmith;ASTRAL-III v5.6.2;Prequal v1.0.2

For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

DataPolicy information about availability of data

All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: - Accession codes, unique identifiers, or web links for publicly available datasets - A list of figures that have associated raw data - A description of any restrictions on data availability

Raw transcriptome and genome reads from R. limneticus and R. marinus have been deposited in the GenBank Sequence Read Archive (SRA). Full SSU rRNA gene sequences have been deposited in GenBank under the accessions MK966712 (R. marinus) and MK966713 (R. limneticus). Assembled transcriptomes and genomes

2


aryO

ctober 2018

have been deposited in Dryad (ACCESSION), along with raw light and electron microscopy images, individual gene alignments, concatenated and trimmed alignments, and ML and Bayesian tree-files for the 151-taxon and 153-taxon datasets . Zoobank accessions are also provided for family (urn:lsid:zoobank.org:act:80B5C004-2954-4A57-A411-482BCD29E85D), genus (urn:lsid:zoobank.org:act:6D09D4D9-D9FC-4D0C-8FB2-55FD9DDEAD53), and species (R. limneticus, urn:lsid:zoobank.org:act:695ACD0B-8151-4609-97FC-A044A312BE22; R. marinus, urn:lsid:zoobank.org:act:84233191-4710-43D1-A2DA-914B8E7B7E01).

Field-specific reportingPlease select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences

For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Ecological, evolutionary & environmental sciences study designAll studies must disclose on these points even when the disclosure is negative.

Study description In this study, we describe two species from a novel phylum of predatory eukaryotic microbes, Rhodelphidia. We performed detailed ultrastructural, transcriptomic/genomic, and phylogenomic analyses, showing that Rhodelphis are the sister lineage to red algae, and that they likely retain a non-photosynthetic plastid involved in heme biosynthesis.

Research sample This research describes two new species, Rhodelphis limneticus and R. marinus, from a new phylum of predatory eukaryotic microbes that is the sister lineage to red algae. The organisms were collected from freshwater lake sediments and seawater sediments, respectively.

Sampling strategy Sample size is not relevant to the present study.

Data collection Samples were collected from freshwater and marine sediments, and the new organisms were subsequently grown in the laboratory. Microscopic data were recorded by D Tikhonenkov. Sequencing data were generated by the UCLA Clinical Microarray Core (R. marinus), The Centre of Applied Genomics (Toronto, Canada; R. limneticus), and in-house using a minION (F Husnik). Transcriptome/genome data were assembled by R Gawryluk.

Timing and spatial scale Sampling relevant to the present study was carried out only two times: once from marine coral sand off of a small island, Bay Canh, near Con Dao Island, South Vietnam on May 3, 2015, and once from a near-shore freshwater sample on August 9, 2016. We had no reason to expect to find the organisms that we did, so there is no specific rationale to sampling sites.

Data exclusions Sequencing data from prey organisms were excluded from the analyses as best as possible. To do this for transcriptomes (R.marinus and R. limneticus), we subtracted transcripts derived from prey (kinetoplastids and co-cultured bacteria) from the total datasets. For genomic datasets, we used automated binning (autometa) along with limited manual curation in order to reduce prey contributions to the genome datasets. The raw data associated with this are still accessible in the raw read files deposited in the NCBI SRA database.

Reproducibility Microscopic analyses were conducted several times. Phylogenomic analyses were carried out with a number of different approaches (maximum likelihood, Bayesian etc.) and all associated datasets have been made available.

Randomization Randomization is not relevant to the present study because organisms were not allocated into groups.

Blinding Blinding was not relevant to the present study.

Did the study involve field work? Yes No

Field work, collection and transportField conditions Climatic conditions in the field were not recorded and are not relevant to the study.

Location 1) Bay Canh island near Con Dao Island, South Vietnam (8.666288 N, 106.680488 E). 2) Lake Trubin (flood-lands of Desna River, 51.397222 N, 32.368889 E), near Yaduty village, Chernigovskaya oblast, Ukraine

Access and import/export Habitats were accessed via a motor boat (location 1) and a car (location 2). No permissions were required for sampling in the selected sampling sites.

Disturbance No disturbances to the sites were caused; we sampled a small amount of water and sand/debris from a marine and lake habitat.

Reporting for specific materials, systems and methods

3


aryO

ctober 2018

We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material, system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.

Materials & experimental systemsn/a Involved in the study

Antibodies

Eukaryotic cell lines

Palaeontology

Animals and other organisms

Human research participants

Clinical data

Methodsn/a Involved in the study

ChIP-seq

Flow cytometry

MRI-based neuroimaging

Eukaryotic cell linesPolicy information about cell lines

Cell line source(s) Two clonal cultures of protists were isolated from marine coral sand and a freshwater sample containing organic debris.

Authentication Phase contrast light microscopy and 18S rRNA gene sequencing was used for authentication.

Mycoplasma contamination This is not relevant to protist cell culture.

Commonly misidentified lines(See ICLAC register)

This is not relevant to protist cell culture.

Animals and other organismsPolicy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research

Laboratory animals The study did not involve laboratory animals.

Wild animals The study did not involve wild animals (or any animals).

Field-collected samples Monoeukaryotic cultures of Rhodelphis marinus and R. limneticus were established by isolating cells with a glass micropipette. Cultures were maintained at room temperature. R. marinus and R. limneticus were propagateed using the kinetoplastid protists Procryptobia sorokini B-69 and Parabodo caudatus BAS-1 as prey, respectively. The kinetoplastids were grown in marine Schmalz-Pratt’s medium and spring water and preyed upon Pseudomonas fluorescens.

Ethics oversight No ethical approval was required. The organisms described here are novel eukaryotic microbes (protists) that feed on other protists and pose no risks.

Note that full information on the approval of the study protocol must also be provided in the manuscript.

Non-photosynthetic predators are sister to red algae · Non-photosynthetic predators are sister to...

Documents

Transcript of Non-photosynthetic predators are sister to red algae · Non-photosynthetic predators are sister to...