
==== Front
Genome Biol Evol
Genome Biol Evol
gbe
Genome Biology and Evolution
1759-6653
Oxford University Press UK

39240751
10.1093/gbe/evae192
evae192
Letter
AcademicSubjects/SCI01130
AcademicSubjects/SCI01140
A New Model and Dating for the Evolution of Complex Plastids of Red Alga Origin
https://orcid.org/0000-0001-6218-9804
Pietluch Filip Department of Bioinformatics and Genomics, Faculty of Biotechnology, University of Wroclaw, 50-383 Wroclaw, Poland

https://orcid.org/0000-0003-4855-497X
Mackiewicz Paweł Department of Bioinformatics and Genomics, Faculty of Biotechnology, University of Wroclaw, 50-383 Wroclaw, Poland

https://orcid.org/0009-0009-9937-4857
Ludwig Kacper Department of Bioinformatics and Genomics, Faculty of Biotechnology, University of Wroclaw, 50-383 Wroclaw, Poland

https://orcid.org/0000-0001-9077-439X
Gagat Przemysław Department of Bioinformatics and Genomics, Faculty of Biotechnology, University of Wroclaw, 50-383 Wroclaw, Poland

Eme Laura Associate Editor
Corresponding author: E-mail: przemyslaw.gagat@uwr.edu.pl.
9 2024
06 9 2024
06 9 2024
16 9 evae19231 8 2024
20 9 2024
© The Author(s) 2024. Published by Oxford University Press on behalf of Society for Molecular Biology and Evolution.
2024
https://creativecommons.org/licenses/by/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Complex plastids, characterized by more than two bounding membranes, still present an evolutionary puzzle for the traditional endosymbiotic theory. Unlike primary plastids that directly evolved from cyanobacteria, complex plastids originated from green or red algae. The Chromalveolata hypothesis proposes a single red alga endosymbiosis that involved the ancestor of all the Chromalveolata lineages: cryptophytes, haptophytes, stramenopiles, and alveolates. As extensive phylogenetic analyses contradict the monophyly of Chromalveolata, serial plastid endosymbiosis models were proposed, suggesting a single secondary red alga endosymbiosis within Cryptophyta, followed by subsequent plastid transfers to other chromalveolates. Our findings based on 97 plastid-encoded markers, 112 species, and robust phylogenetic methods challenge all the existing models. They reveal two independent secondary endosymbioses, one within Cryptophyta and one within stramenopiles, precisely the phylum Ochrophyta, with two different groups of red algae. Consequently, we propose a new model for the emergence of red alga plastid–containing lineages and, through molecular clock analyses, estimate their ages.

Chromalveolata
Cryptophyta
endosymbiosis
phylogeny
plastid
stramenopiles
National Science Centre 10.13039/501100004281 2017/26/D/NZ8/00444 Wrocław Centre for Networking and Supercomputing 10.13039/501100011009 307
==== Body
pmcSignificance

Cyanobacteria and eukaryote endosymbioses created a multitude of photosynthetic organelles called “plastids” that feed most life on our planet. For decades, scientists have been trying to untangle the puzzle of plastid evolution, i.e. when and how plastids were acquired and spread throughout the eukaryotic tree of life. Our new model, based on robust phylogenetic methods, challenges current hypotheses and offers fresh insights into the evolution of complex plastids of red alga origin in cryptophytes, haptophytes, stramenopiles, and alveolates.

Introduction

The ability to convert solar energy into adenosine triphosphate is the distinctive feature of cyanobacteria and plastids. The first plastids were direct cyanobacteria descendants and are accordingly called “primary plastids.” They occur in glaucophytes, red algae, and green algae/land plants, which form the superassemblage Archaeplastida (Löffelhardt 2014; Burki et al. 2020). Primary plastids are surrounded by two membranes that correspond to those of their cyanobacterial ancestors. Interestingly, many protist groups, including euglenids, chlorarachniophytes, cryptophytes, haptophytes, stramenopiles, and alveolates, carry plastids enveloped by three or more membranes. They are called “complex plastids,” and their origin has been a topical issue in evolutionary biology for decades (Archibald 2015; Adl et al. 2019).

To explain the evolution of complex plastids, Cavalier-Smith (1999) proposed the Cabozoa and Chromalveolata hypotheses. The first suggested that plastids of euglenids and chlorarachniophytes were derived from a single green alga endosymbiosis, whereas the second postulated a single red alga endosymbiosis for the origin of plastids in cryptophytes, haptophytes, stramenopiles, and alveolates; the latter included dinoflagellates, perkinsids, colpodellids, and apicomplexans. The Cabozoa hypothesis was quickly refuted by phylogenetic analyses, which revealed that neither the host cells nor green alga endosymbionts of euglenids and chlorarachniophytes are closely related (Archibald and Keeling 2002; Rogers et al. 2007). In contrast, the Chromalveolata hypothesis influenced our understanding of endosymbiosis for over a decade (Keeling 2009; Gould et al. 2015). However, phylogenetic analyses also disproved this model as nuclear markers did not support the monophyly of the Chromalveolata clade (Burki et al. 2012, 2016; Strassert et al. 2021).

Subsequent models explaining the evolution of red alga–derived plastids and reconciling nuclear and plastid marker–based phylogenies involved serial endosymbioses. For example, Bodył et al. (2009a, 2009b, 2018) and Stiller et al. (2014) proposed that only the cryptophyte plastid directly descended from red algae, while the other red alga–derived plastids evolved through multiple transfers among the Chromalveolata lineages, with cryptophytes as their initial donors (supplementary fig. S1, Supplementary Material online).

As the evolution of Chromalveolata plastids remains unclear and the existing models are inconsistent, we decided to reinvestigate the history of red alga–derived plastids. Our phylogenies indicate that there are two independent red alga endosymbioses within chromalveolates, challenging the previous models. Consequently, we advance a new scenario for the evolution of red alga–derived plastids and employ molecular dating to estimate the age of the lineages containing them.

Materials and Methods

We employed 97 conserved plastid-encoded proteins from 112 organisms, to include all major red alga plastid lineages, from the NCBI reference sequence database (O’Leary et al. 2016) and GenBank (Sayers et al. 2021; supplementary table S12, Supplementary Material online). Alignments were created using the L-INS-i algorithm in MAFFT (Katoh and Standley 2013) and assessed in AliView (Larsson 2014), and phylogenetically informative sites were selected with trimAl (Capella-Gutiérrez et al. 2009) and ClipKIT (Steenwyk et al. 2020) based on the benchmark provided by Steenwyk et al. (2020). To balance the phylogenetic signal and reduce noise from poorly aligned regions, we constructed phylogenetic trees using the ML method in IQ-TREE (Minh et al. 2020) and RAxML (Bouckaert et al. 2014), employing seven trimming strategies. Alignments were concatenated into supermatrices using SequenceMatrix (Vaidya et al. 2011) and compared with PhyKIT (Steenwyk et al. 2021; supplementary table S1, Supplementary Material online). An additional ML tree was inferred in IQ-TREE under the CAT model, and three phylogenies along with chronograms were calculated using the Bayesian approach, two in Beast (Bouckaert et al. 2014) and one in MrBayes (Ronquist et al. 2012). These analyses were based on the smart-gap trimming alignment, offering a supermatrix with the highest number of variables (26,449) and parsimony-informative sites (22,039) and preserving almost 97% of all parsimony-informative sites from the original alignment.

To address data heterogeneity, we considered 97 potential partitions within all supermatrices. Model selection was performed using ModelFinder (Kalyaanamoorthy et al. 2017) in IQ-TREE and PartitionFinder (Lanfear et al. 2017) for RAxML, MrBayes, and Beast (supplementary tables S13 to S29, Supplementary Material online). We also employed the CAT model in IQ-TREE (Lartillot and Philippe 2004). We tested variants with 10, 20, and 30 profiles using substitution matrices: Poisson, LG, and Q.yeast. Based on Bayesian Information Criterion, we selected the best-fit model: LG + C20 + F + R10. To further account for data heterogeneity, we employed only relaxed molecular clocks. In MrBayes, we used an independent gamma rate (IGR) model (Ronquist et al. 2012), while in Beast, we applied an uncorrelated lognormal model and an uncorrelated exponential model (Bouckaert et al. 2014). To test the assumptions of stationarity and homogeneity in our data, we performed the maximum test of symmetry in IQ-TREE for each partition. Using Benjamini–Hochberg correction for multiple testing, we found that all 17 partitions passed the test.

To assess the confidence of the inferred phylogenies, nonparametric bootstrap tests (100 replicates) were used for ML analyses in IQ-TREE (1,000 for the CAT model) and RAxML, while posterior probabilities were employed for Bayesian approaches. In Beast, the Yule tree prior and exponential priors were applied to define calibration constraints, while in MrBayes, offset exponential priors, in combination with the birth–death tree prior, were applied. During the Markov chain Monte Carlo process, tree samples were saved every 1,000 iterations. To ensure robust parameter estimation, the first 10% of samples were discarded as burnin. The median heights were calculated for the final chronograms.

Results

Phylogenetic Analyses

All our phylogenies consistently rejected the monophyly of Chromalveolata plastids and showed that the clade of ochrophytes (photosynthetic stramenopiles) and cryptophytes + haptophytes (a mean bootstrap rate of 100% and posterior probability: 1) was polyphyletic. The trees provided strong to moderate support (bootstrap range: 61% to 88% and mean 78% and posterior probability range: 0.85 to 1 and mean 0.95) for the clustering of Cryptophyta and Haptophyta plastids with two red alga groups: Proteorhodophytina and Eurhodophytina (Figs. 1 and 2, supplementary figs. S2 to S20, Supplementary Material online). Ochrophyte plastids were either placed with thermoacidophilic red algae Cyanidiales (14 out of 20 trees) or branched between Cyanidiales and Proteorhodophytina/Eurhodophytina alongside with Cryptophyta/Haptophyta (6 out of 20 trees) (Figs. 1 and 2, supplementary figs. S2 to S20, Supplementary Material online, Table 1). The placement of ochrophytes with Cyanidiales (Fig. 1, topology A) achieved a maximum bootstrap rate of 68% in the IQ-TREE phylogeny based on the CAT model (supplementary fig. S20, Supplementary Material online), and two Bayesian trees provided a posterior probability of 1 for this clade (Fig. 2, supplementary fig. S3, Supplementary Material online). The alternative positioning of Ochrophyta (Fig. 1, topology B) reached a maximum bootstrap rate of 71% in the RAxML tree (supplementary fig. S18, Supplementary Material online). Considering the prevalence of the former topology in a larger number of trees, including the Bayesian ones, it appears to be more plausible.

Fig. 1. Two tree topologies obtained based on various trimming strategies (supplementary table S1, Supplementary Material online). In both trees, cryptophyte and haptophyte plastids are clustered with those from Proteorhodophytina/Eurhodophytina red algae, whereas stramenopile plastids are grouped either with Cyanidiales red algae a) or branch between Cyanidiales and the clade containing Proteorhodophytina/Eurhodophytina alongside Cryptophyta/Haptophyta b).

Fig. 2. Time-calibrated phylogeny of red alga and red alga–derived plastids. The tree was inferred with Beast using an exponential model, based on ClipKIT smart-gap alignment (supplementary table S1, Supplementary Material online), along with substitution models provided in supplementary table S2, Supplementary Material online, and calibrated with constraints (numbers in white circles) listed in supplementary table S3, Supplementary Material online. Node numbers to help reading time estimates are given in supplementary fig. S21, Supplementary Material online, and the estimates in Table 2 and supplementary table S4, Supplementary Material online, and the bar rate at key nodes is 95% HPD (Highest Posterior Density). Nodes with posterior probability lower than 1 are marked with green circles. Abbreviations on the geological timescale for Eras: Cz—Cenozoic, Mz—Mesozoic, Pz—Paleozoic, Nptz—Neoproterozoic, Mptz—Mesoproterozoic, and Pptz—Paleoproterozoic.

Table 1 Results of phylogenetic analyses categorized by software, trimming strategy, and obtained topologies

Software	Trimming strategya	Topology A	Topology B	Figure	
Beast exponential	ClipKIT smart-gap	X		2	
Beast lognormal	ClipKIT smart-gap	X		S2	
MrBayes IGR	ClipKIT smart-gap	X		S3	
IQ-TREE	ClipKIT gappy	X		S4	
IQ-TREE	ClipKIT kpic	X		S5	
IQ-TREE	ClipKIT kpic-smart-gap		X	S6	
IQ-TREE	ClipKIT smart-gap	X		S7	
IQ-TREEb	ClipKIT smart-gap	X		S20	
IQ-TREE	trimAl automated1		X	S8	
IQ-TREE	trimAl gappyout	X		S9	
IQ-TREE	trimAl strict		X	S10	
IQ-TREE	original alignment		X	S11	
RAxML	ClipKIT gappy	X		S12	
RAxML	ClipKIT kpic	X		S13	
RAxML	ClipKIT kpic-smart-gap	X		S14	
RAxML	ClipKIT smart-gap	X		S15	
RAxML	trimAl automated1	X		S16	
RAxML	trimAl gappyout		X	S17	
RAxML	trimAl strict		X	S18	
RAxML	original alignment	X		S19	
Topology A: Ochrophyta plastids are placed with Cyanidiales; topology B: Ochrophyta plastids branch between Cyanidiales and Proteorhodophytina/Eurhodophytina alongside Cryptophyta/Haptophyta clade (Fig. 1).

aStatistics for alignment trimming strategies are provided in supplementary table S1, Supplementary Material online.

bCAT model.

Notably, we observed a link between the tree topologies and supermatrix length. Longer supermatrices, with a higher number of variable and parsimony-informative sites, predominantly yielded trees supporting ochrophytes’ grouping with Cyanidiales (9 out of 14 trimming strategies). In contrast, shorter supermatrices favored the alternative topology (5 out of 14 trimming strategies). The untrimmed supermatrix produced both topologies depending on the software: in the RAxML tree, ochrophytes clustered with Cyanidiales (topology A), while in the IQ-TREE phylogeny, they branched between Cyanidiales and Proteorhodophytina/Eurhodophytina alongside Cryptophyta/Haptophyta (topology B) (Table 1, supplementary figs. S11 and S19, Supplementary Material online). Irrespective of ochrophytes’ placement (topology A vs. topology B), our results suggest two separate secondary red alga endosymbioses in Chromalveolata lineages: one driving photosynthesis in the Cryptophyta/Haptophyta clade and the other in Ochrophytes. Importantly, tests of topologies assuming the monophyly of cryptophytes, haptophytes, and ochrophytes, the number of amino acid sites, and testing of individual partitions also supported the model for two red alga plastid transfers to chromalveolates as well as topology A (supplementary fig. S22 and tables S5 to S8, Supplementary Material online).

In our phylogenies, alveolates, specifically colpodellids (Chromerida sp. RM11 and Chromera velia) and dinoflagellates (Durinskia baltica and Kryptoperidinium foliaceum), are placed separately within ochrophytes. Colpodellids form a sister clade with Eustigmatophyceae (bootstrap range: 83% to 99% and mean 95% and posterior probabilities 1), while dinoflagellates branch among Nitzschia, Cylindrotheca, and Pseudo-nitzschia species (bootstraps 100% and posterior probabilities 1). Importantly, dinoflagellates are not monophyletic; D. baltica is grouped with Nitzschia palea, and K. foliaceum is their sister, suggesting independent ochrophyte (diatom) endosymbioses in these lineages. Additionally, a dinoflagellate, Karlodinium veneficum, clusters with haptophytes (bootstraps 100% and posterior probabilities 1) (Fig. 2, supplementary figs. S2 to S20, Supplementary Material online).

Molecular Clock Analyses

The estimated ages are presented as means of mean dates from the three molecular clocks calculated (supplementary figs. S21, S23 and S24 and tables S4, S9 and S10 Supplementary Material online, Table 2). Based on our chronograms, the red alga lineage emerged during the Paleoproterozoic Era ∼1.92 billion years ago (Bya). The estimated age for the crown group of red algae is ∼1.75 Bya, corresponding to the divergence of Cyanidiales from the clade comprising Proteorhodophytina and Eurhodophytina (Fig. 2, supplementary figs. S2 and S3, Supplementary Material online).

Table 2 Molecular clock estimates for key nodes in million years

Node number and node name	Beast exponential	Beast log normal	Mrbayes IGR	Age meana	Age rangesb	
Mean	95% HPD	Mean	95% HPD	Mean	95% HPD	
114 Divergence of Red algae	1,874	1,800 to 2,018	1,882	1,800 to 2,045	1,998	1,807 to 2,329	1,918	1,800 to 2,329	
115 Split of Cyanidiales and Proteo./Eurho.	1,707	1,560 to 1,869	1,741	1,560 to 1,930	1,807	1,560 to 2,076	1,752	1,560 to 2,076	
116 Split of Cyanidiales and Ochrophytes	1,598	1,360 to 1,819	1,670	1,463 to 1,888	1,740	1,494 to 2,016	1,669	1,360 to 2,016	
117 Crown group Ochrophytes	1,451	1,181 to 1,695	1,539	1,337 to 1,744	1,568	1,320 to 1,849	1,519	1,181 to 1,849	
148 Divergence of K. foliaceum	70	21 to 135	93	44 to 142	88	38 to 143	84	21 to 143	
149 Split of Durinskia and Nitzschia	18	3 to 47	18	5 to 37	16	5 to 30	17	3 to 47	
167 Divergence of Colpodellids	1,179	937 to 1,428	1,233	1,023 to 1,454	1,246	1,008 to 1,503	1,220	937 to 1,503	
168 Crown group Colpodelids	896	719 to 1,080	997	788 to 1,205	886	671 to 1,111	926	671 to 1,205	
175 Crown group Cyanidiales	928	398 to 1,495	1,239	680 to 1,648	1,407	861 to 1,855	1,191	398 to 1,855	
178 Split of Proteo./Eurho and Haptophyta/Cryptophyta	1,498	1,182 to 1,758	1,539	1,273 to 1,803	1,543	1,210 to 1,882	1,527	1,182 to 1,882	
179 Crown group Proteo. and Eurho.	1,234	1,048 to 1,482	1,210	1,059 to 1,387	1,232	1,077 to 1,435	1,225	1,048 to 1,482	
209 Split of Haptophyta and Cryptophyta	1,101	761 to 1,456	1,157	798 to 1,487	1,290	849 to 1,722	1,183	761 to 1,722	
210 Crown group Cryptophyta	625	315 to 939	513	315 to 725	506	294 to 771	548	294 to 939	
217 Crown group Haptophyta	932	631 to 1,268	1,020	702 to 1,323	1,171	741 to 1,555	1,041	631 to 1,555	
219 Divergence of K. veneficum	598	399 to 853	656	442 to 881	754	472 to 1,058	669	399 to 1,058	
aCalculated as means of mean dates from the three chronograms.

bThe minimum and maximum values of 95% HPD intervals from the three chronograms.

Our analyses strongly indicate that Ochrophytes acquired their plastids directly from red algae, related to Cyanidiales. This event occurred after plastids of Cyanidiales and Ochrophytes diverged ∼1.67 Bya, but before the estimated age of the crown group Ochrophyta ∼1.52 Bya. Furthermore, we determined that ochrophytes transferred their plastids to colpodellids possibly as early as ∼1.22 Bya, when the colpodellid plastid diverged from Ochrophyta, but no later than ∼926 million years ago (Mya), corresponding to the separation age of studied colpodellids (Fig. 2, supplementary figs. S2 and S3, Supplementary Material online).

The provided chronograms suggest the possibility that Cryptophyta acquired plastids after Ochrophytes; however, the opposite scenario is also plausible due to overlapping estimated time frames. According to our clocks, Cryptophyta and Haptophyta plastids diverged from Proteorhodophytina and Eurhodophytina red algae ∼1.53 Bya, but prior to ∼1.18 Bya, i.e. the separation of cryptophytes and haptophytes (Fig. 2, supplementary figs. S2 and S3, Supplementary Material online).

In our studies, we also investigated plastid transfers to dinoflagellates: K. veneficum, K. foliaceum, and D. baltica. They acquired their photosynthetic organelles through three separate endosymbiotic events: ∼670, ∼84, and 17.5 Mya, respectively (Fig. 2, supplementary figs. S2 and S3, Supplementary Material online).

Discussion

Our results challenge the Chromalveolata and serial endosymbioses models. The former claims a single red alga endosymbiosis with the ancestor of all Chromalveolata lineages, while the latter claims a single secondary red alga endosymbiosis within cryptophytes, followed by subsequent plastid transfers to other Chromalveolata lineages. Based on host relationships by Strassert et al. (2021), our phylogenies suggest two secondary red alga endosymbioses, one within Cryptophyta and one within Ochrophyta (Fig. 3). Such a possibility has already been mentioned by Dorrell and Bowler (2017) and obtained in Kim et al. (2015) trees. From our phylogenies, we can also infer two independent tertiary endosymbioses: (i) cryptophyte-to-haptophyte and (ii) ochrophyte-to-colpodellid.

Fig. 3. The multiple and serial endosymbiosis model proposed in this study. The arrows indicate plastid transfers, and the numbers in circles correspond to the level of endosymbiosis: 2—secondary and 3—tertiary. The phylogenetic relationships between main eukaryotic lineages (hosts) were based on Strassert et al. (2021). The presented model assumes that cryptophytes donated their plastids to haptophytes rather than haptophytes to cryptophytes since this scenario is more parsimonious considering the presence of a nucleomorph in cryptophyte plastids (a vestigial nucleus of red alga origin between the fourth and the third plastid membranes counting from outside). The alternative scenario would have required haptophytes to carry an endosymbiont with a nucleus and a plastid, which was next donated to cryptophytes, and then, the endosymbiont nucleus was completely reduced in haptophytes but not in cryptophytes.

Assuming that the colpodellid plastid represents the ancestral alveolate plastid, our results align with the serial endosymbioses models for this particular plastid transfer (Fig. 3, supplementary fig. S1, Supplementary Material online). Notably, our phylogenies and those of Ševčíková et al. (2015) group colpodellids with Eustigmatophyceae. Both lineages lack chlorophyll c, a defining characteristic of Chromalveolata, and instead rely on chlorophyll a, violaxanthin, and β-carotene for capturing light energy (Füssy and Oborník 2017; Amaral et al. 2021). This photosynthetic pigment composition further strengthens the hypothesis that ochrophytes, precisely Eustigmatophyceae, donated their plastid to the ancestor of alveolate plastid–containing lineages.

According to the serial endosymbioses hypotheses, the ochrophyte plastid was replaced in some dinoflagellates by other plastids, e.g. from cryptophytes, other ochrophytes, and haptophytes, and sometimes more than once (Gagat et al. 2014; Bodył 2018). This scenario is also reflected in our phylogenies as D. baltica and K. foliaceum are separated in our trees by N. palea, which indicates two independent ochrophyte endosymbioses in these species. D. baltica and K. foliaceum represent “dinotoms,” i.e. endosymbiotic consortia of a dinoflagellate and diatom. The former carries a pennate diatom endosymbiont, whereas the latter a centric one (Gagat et al. 2014; Tillmann et al. 2023). Moreover, K. veneficum carries a fucoxanthin plastid derived from haptophytes (Gagat et al. 2014).

Our model (Fig. 3) is more parsimonious than previous ones considering the presence of the eubacterial rpl36 gene copy in the plastomes of cryptophytes and haptophytes (Rice and Palmer 2006). It avoids extra assumptions, e.g. (i) cryptophyte-to-ochrophyte plastid transfer followed by the rpl36 replacement in the cryptophyte plastome and subsequent cryptophyte-to-haptophyte plastid donation or (ii) independent eubacterial rpl36 acquisitions by both lineages (Bodył, et al. 2009b).

Our model, however, might seem initially less parsimonious than serial endosymbioses models regarding the evolution of protein import into complex plastids. In short, protein import into four-membrane plastids of cryptophytes, haptophytes, and ochrophytes relies on a bipartite presequence: a signal peptide and a transit peptide, and four protein complexes: Sec61, SELMA, Toc, and Tic, located in the fourth (continuous with the host ER), third, second, and first plastid membrane (counting from outside), respectively (Stork et al. 2013). The signal peptide facilitates the crossing of the outermost membrane, whereas the transit peptide enables the crossing of the three remaining ones. A similar system operates in colpodellid/apicomplexan plastids; however, they use vesicular trafficking to connect the host ER with the plastid outermost membrane (Stork et al. 2012). Importantly, the Toc-Tic translocons represent cyanobacteria-host-derived complexes that operate in primary plastids along with transit peptides (Bodył, et al. 2009a; Richardson and Schnell 2020), and Sec61 is an ancient system responsible for protein movement into the ER (Osborne et al. 2005). The only major innovation in complex plastids regarding protein import is SELMA, which evolved from the red alga–derived ERAD complex previously used to export misfolded/aberrant proteins from the ER (Bolte et al. 2011; Felsner et al. 2011; Lau et al. 2016). Under the serial endosymbiosis hypotheses, the ERAD system was adapted only once, while our model requires two independent adaptations. However, unlike the Toc-Tic supercomplex, which represents a chimeric cyanobacterial host innovation, SELMA represents a simple adaptation of a red alga system already used for protein transport. Therefore, we do not consider two independent ERAD adaptations and two subsequent SELMA transfers in two independent tertiary endosymbioses significantly more challenging that one ERAD adaptation and three subsequent SELMA transfers in three independent higher-order endosymbioses. In both scenarios, the same number of the same set of genes is required to be transferred to the host genome.

Comparing our chronograms with clocks calculated by other researchers is challenging because of differences in markers (plastid vs. nuclear), species, calibration points, molecular clock methods, and tree topologies obtained. Nevertheless, meaningful comparisons can still be made by examining the ages of the crown groups of Ochrophyta, Cryptophyta, and Haptophyta from various studies (supplementary table S11, Supplementary Material online). Our estimations for red algae evolution (∼1.92 Bya, crown group ∼1.75 Bya) align with recent molecular clock studies (Strassert et al. 2021; Pietluch et al. 2022) and are consistent with the oldest fossils of red alga discovered (Rafatazmia chitrakootensis ∼1.6 Bya) (Bengtson et al. 2017). In our clocks, the crown group of Ochrophyta emerged ∼1.52 Bya, which significantly differs from previous estimations ranging from ∼575 to ∼1.22 Bya. This discrepancy is, however, understandable given the new placement of Ochrophyta in our phylogenies. The estimated age for colpodelids, which is ∼926 Mya, a lineage representing plastid-containing alveolates, is consistent with more recent nuclear marker–based clocks calculated by Parfrey et al. (2011) and Strassert et al. (2021). We inferred the age of the crown group of Cryptophyta as ∼550 Mya and Haptophyta as ∼1,050 Mya. These estimations agree with the lower boundaries for these points calculated by Strassert et al. (2021).

Our results challenge the current view on plastid evolution and suggest that secondary endosymbioses may not be rarer than endosymbioses of a higher order. The two red alga endosymbioses proposed in our scenario provide another example of secondary plastid acquisitions, adding to the already observed two independent endosymbioses of green alga plastids within euglenids and chlorarachniophytes.

Supplementary Material

evae192_Supplementary_Data

Author Contributions

P.G. conceived the research, wrote the manuscript, and managed the project; P.G. and F.P. designed the research; F.P. performed the phylogenomic and molecular clock analyses under the supervision of P.G.; P.M. and K.L. conducted the additional analyses; P.G., F.P., and P.M. analyzed the data; and P.G. and P.M. answered the reviews.

Supplementary Material

Supplementary material is available at Genome Biology and Evolution online.

Funding

This work was supported by the National Science Centre grant no. 2017/26/D/NZ8/00444 to P.G. and the National Science Centre grant no. 2019/35/N/NZ8/03366 to F.P. Some computations were carried out at the Wrocław Centre for Networking and Supercomputing under the grant no. 307.

Data Availability

All data sets used to perform phylogenetic and molecular clock analyses are available at https://zenodo.org/records/11953485.
==== Refs
Literature Cited

Adl  SM, Bass  D, Lane  CE, Lukeš  J, Schoch  CL, Smirnov  A, Agatha  S, Berney  C, Brown  MW, Burki  F, et al  Revisions to the classification, nomenclature, and diversity of eukaryotes. J Eukaryot Microbiol.  2019:66 (1 ):4–119. 10.1111/jeu.12691.30257078
Amaral  R, de Melo  JSS, dos Santos  LMA. Pigments from Eustigmatophyceae: an interesting class of microalgae for carotenoid production. J Appl Phycol.  2021:33 (1 ):371–384. 10.1007/s10811-020-02312-z.
Archibald  JM . Genomic perspectives on the birth and spread of plastids. Proc Natl Acad Sci U S A. 2015:112 (33 ):10147–10153. 10.1073/pnas.1421374112.25902528
Archibald  JM, Keeling  PJ. Recycled plastids: a ‘green movement’ in eukaryotic evolution. Trends Genet.  2002:18 (11 ):577–584. 10.1016/S0168-9525(02)02777-4.12414188
Bengtson  S, Sallstedt  T, Belivanova  V, Whitehouse  M. Three-dimensional preservation of cellular and subcellular structures suggests 1.6 billion-year-old crown-group red algae. PLoS Biol.  2017:15 (3 ):e2000735. 10.1371/journal.pbio.2000735.28291791
Bodył  A . Did some red alga-derived plastids evolve via kleptoplastidy? A hypothesis. Biol Rev Camb Philos Soc.  2018:93 (1 ):201–222. 10.1111/brv.12340.28544184
Bodył  A, Mackiewicz  P, Stiller  JW. Early steps in plastid evolution: current ideas and controversies. Bioessays. 2009a:31 (11 ):1219–1232. 10.1002/bies.200900073.19847819
Bodył  A, Stiller  JW, Mackiewicz  P. Chromalveolate plastids: direct descent or multiple endosymbioses?  Trends Ecol Evol.  2009b:24 (3 ):119–121. 10.1016/j.tree.2008.11.003.19200617
Bolte  K, Gruenheit  N, Felsner  G, Sommer  MS, Maier  U-G, Hempel  F. Making new out of old: recycling and modification of an ancient protein translocation system during eukaryotic evolution. BioEssays. 2011:33 (5 ):368–376. 10.1002/bies.201100007.21425305
Bouckaert  R, Heled  J, Kühnert  D, Vaughan  T, Wu  C-H, Xie  D, Suchard  MA, Rambaut  A, Drummond  AJ. BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput Biol.  2014:10 (4 ):e1003537. 10.1371/journal.pcbi.1003537.24722319
Burki  F, Kaplan  M, Tikhonenkov  DV, Zlatogursky  V, Minh  BQ, Radaykina  LV, Smirnov  A, Mylnikov  AP, Keeling  PJ. Untangling the early diversification of eukaryotes: a phylogenomic study of the evolutionary origins of Centrohelida, Haptophyta and Cryptista. Proc R Soc Lond B Biol Sci.  2016:283 (1823 ):20152802. 10.1098/rspb.2015.2802.
Burki  F, Okamoto  N, Pombert  J-F, Keeling  PJ. The evolutionary history of haptophytes and cryptophytes: phylogenomic evidence for separate origins. Proc R Soc Lond B Biol Sci.  2012:279 (1736 ):2246–2254. 10.1098/rspb.2011.2301.
Burki  F, Roger  AJ, Brown  MW, Simpson  AGB. The new tree of eukaryotes. Trends Ecol Evol.  2020:35 (1 ):43–55. 10.1016/j.tree.2019.08.008.31606140
Capella-Gutiérrez  S, Silla-Martínez  JM, Gabaldón  T. Trimal: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics. 2009:25 (15 ):1972–1973. 10.1093/bioinformatics/btp348.19505945
Cavalier-Smith  T . Principles of protein and lipid targeting in secondary symbiogenesis: euglenoid, dinoflagellate, and sporozoan plastid origins and the eukaryote family tree1,2. J Eukaryot Microbiol.  1999:46 (4 ):347–366. 10.1111/j.1550-7408.1999.tb04614.x.18092388
Dorrell  RG, Bowler  C. Chapter three—secondary plastids of stramenopiles. In: Hirakawa  Y, editor. Advances in botanical research. Vol. 84 . Academic Press; 2017. p. 57–103. 10.1016/bs.abr.2017.06.003.
Felsner  G, Sommer  MS, Gruenheit  N, Hempel  F, Moog  D, Zauner  S, Martin  W, Maier  UG. ERAD components in organisms with Complex red plastids suggest recruitment of a preexisting protein transport pathway for the periplastid membrane. Genome Biol Evol.  2011:3 :140–150. 10.1093/gbe/evq074.21081314
Füssy  Z, Oborník  M. Chapter six—chromerids and their plastids. In: Hirakawa  Y, editor. Advances in botanical Research. Academic Press; 2017. p. 187–218. 10.1016/bs.abr.2017.07.001.
Gagat  P, Bodył  A, Mackiewicz  P, Stiller  JW. Tertiary plastid endosymbioses in dinoflagellates. In: Löffelhardt  W, editor. Endosymbiosis. Springer; 2014. p. 233–290. 10.1007/978-3-7091-1303-5_13.
Gould  SB, Maier  U-G, Martin  WF. Protein import and the origin of red Complex plastids. Curr Biol.  2015:25 (12 ):R515–R521. 10.1016/j.cub.2015.04.033.26079086
Kalyaanamoorthy  S, Minh  BQ, Wong  TKF, von Haeseler  A, Jermiin  LS. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods.  2017:14 (6 ):587–589. 10.1038/nmeth.4285.28481363
Katoh  K, Standley  DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol.  2013:30 (4 ):772–780. 10.1093/molbev/mst010.23329690
Keeling  PJ . Chromalveolates and the evolution of plastids by secondary endosymbiosis1. J Eukaryot Microbiol.  2009:56 (1 ):1–8. 10.1111/j.1550-7408.2008.00371.x.19335769
Kim  JI, Yoon  HS, Yi  G, Kim  HS, Yih  W, Shin  W. The plastid genome of the cryptomonad Teleaulax amphioxeia. PLoS One. 2015:10 (6 ):e0129284. 10.1371/journal.pone.0129284.26047475
Lanfear  R, Frandsen  PB, Wright  AM, Senfeld  T, Calcott  B. PartitionFinder 2: new methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Mol Biol Evol.  2017:34 (3 ):772–773. 10.1093/molbev/msw260.28013191
Larsson  A . AliView: a fast and lightweight alignment viewer and editor for large datasets. Bioinformatics. 2014:30 (22 ):3276–3278. 10.1093/bioinformatics/btu531.25095880
Lartillot  N, Philippe  H. A Bayesian mixture model for across-site heterogeneities in the amino-acid replacement process. Mol Biol Evol.  2004:21 (6 ):1095–1109. 10.1093/molbev/msh112.15014145
Lau  JB, Stork  S, Moog  D, Schulz  J, Maier  UG. Protein–protein interactions indicate composition of a 480 kDa SELMA complex in the second outermost membrane of diatom complex plastids. Mol Microbiol.  2016:100 (1 ):76–89. 10.1111/mmi.13302.26712034
Löffelhardt  W . The single primary endosymbiotic event. In: Endosymbiosis. Springer; 2014. p. 39–52. 10.1007/978-3-7091-1303-5_3.
Minh  BQ, Schmidt  HA, Chernomor  O, Schrempf  D, Woodhams  MD, von Haeseler  A, Lanfear  R. IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Mol Biol Evol.  2020:37 (5 ):1530–1534. 10.1093/molbev/msaa015.32011700
O’Leary  NA, Wright  MW, Brister  JR, Ciufo  S, Haddad  D, McVeigh  R, Rajput  B, Robbertse  B, Smith-White  B, Ako-Adjei  D, et al  Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. Nucleic Acids Res.  2016:44 (D1 ):D733–D745. 10.1093/nar/gkv1189.26553804
Osborne  AR, Rapoport  TA, van den Berg  B. Protein Translocation by the Sec61/Secy Channel.  Annu Rev Cell Dev Biol. 2005:21 :529–550. 10.1146/annurev.cellbio.21.012704.133214.16212506
Parfrey  LW, Lahr  DJG, Knoll  AH, Katz  LA. Estimating the timing of early eukaryotic diversification with multigene molecular clocks. Proc Natl Acad Sci U S A. 2011:108 (33 ):13624–13629. 10.1073/pnas.1110633108.21810989
Pietluch  F, Mackiewicz  P, Sidorczuk  K, Gagat  P. Dating the photosynthetic organelle evolution in Archaeplastida, Paulinella and secondary-plastid bearing lineages. bioRxiv 2022.07.01.497312. 10.1101/2022.07.01.497312., 3 July 2022, preprint: not peer reviewed.
Rice  DW, Palmer  JD. An exceptional horizontal gene transfer in plastids: gene replacement by a distant bacterial paralog and evidence that haptophyte and cryptophyte plastids are sisters. BMC Biol.  2006:4 (1 ):31. 10.1186/1741-7007-4-31.16956407
Richardson  LGL, Schnell  DJ. Origins, function, and regulation of the TOC–TIC general protein import machinery of plastids. J Exp Bot.  2020:71 (4 ):1226–1238. 10.1093/jxb/erz517.31730153
Rogers  MB, Gilson  PR, Su  V, McFadden  GI, Keeling  PJ. The complete chloroplast genome of the chlorarachniophyte Bigelowiella natans: evidence for independent origins of chlorarachniophyte and euglenid secondary endosymbionts. Mol Biol Evol.  2007:24 (1 ):54–62. 10.1093/molbev/msl129.16990439
Ronquist  F, Teslenko  M, van der Mark  P, Ayres  DL, Darling  A, Höhna  S, Larget  B, Liu  L, Suchard  MA, Huelsenbeck  JP. Mrbayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst Biol.  2012:61 (3 ):539–542. 10.1093/sysbio/sys029.22357727
Sayers  EW, Cavanaugh  M, Clark  K, Pruitt  KD, Schoch  CL, Sherry  ST, Karsch-Mizrachi  I. GenBank. Nucleic Acids Res.  2021:49 (D1 ):D92–D96. 10.1093/nar/gkaa1023.33196830
Ševčíková  T, Horák  A, Klimeš  V, Zbránková  V, Demir-Hilton  E, Sudek  S, Jenkins  J, Schmutz  J, Přibyl  P, Fousek  J. Updating algal evolutionary relationships through plastid genome sequencing: did alveolate plastids emerge through endosymbiosis of an ochrophyte?  Sci Rep.  2015:5 (1 ):10134. 10.1038/srep10134.26017773
Steenwyk  JL, Buida  TJ  III, Labella  AL, Li  Y, Shen  X-X, Rokas  A. PhyKIT: a broadly applicable UNIX shell toolkit for processing and analyzing phylogenomic data. Bioinformatics. 2021:37 (16 ):2325–2331. 10.1093/bioinformatics/btab096.33560364
Steenwyk  JL, Buida  TJ  III, Li  Y, Shen  X-X, Rokas  A. ClipKIT: a multiple sequence alignment trimming software for accurate phylogenomic inference. PLoS Biol.  2020:18 (12 ):e3001007. 10.1371/journal.pbio.3001007.33264284
Stiller  JW, Schreiber  J, Yue  J, Guo  H, Ding  Q, Huang  J. The evolution of photosynthesis in chromist algae through serial endosymbioses. Nat Commun.  2014:5 (1 ):5764. 10.1038/ncomms6764.25493338
Stork  S, Lau  J, Moog  D, Maier  U-G. Three old and one new: protein import into red algal-derived plastids surrounded by four membranes. Protoplasma. 2013:250 (5 ):1013–1023. 10.1007/s00709-013-0498-7.23612938
Stork  S, Moog  D, Przyborski  JM, Wilhelmi  I, Zauner  S, Maier  UG. Distribution of the SELMA translocon in secondary plastids of red algal origin and predicted uncoupling of ubiquitin-dependent translocation from degradation. Eukaryot Cell.  2012:11 (12 ):1472–1481. 10.1128/EC.00183-12.23042132
Strassert  JFH, Irisarri  I, Williams  TA, Burki  F. A molecular timescale for eukaryote evolution with implications for the origin of red algal-derived plastids. Nat Commun.  2021:12 (1 ):1879. 10.1038/s41467-021-22044-z.33767194
Tillmann  U, Wietkamp  S, Kretschmann  J, Chacón  J, Gottschling  M. Spatial fragmentation in the distribution of diatom endosymbionts from the taxonomically clarified dinophyte Kryptoperidinium triquetrum (= Kryptoperidinium foliaceum, Peridiniales). Sci Rep.  2023:13 (1 ):8593. 10.1038/s41598-023-32949-y.37237053
Vaidya  G, Lohman  DJ, Meier  R. SequenceMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics. 2011:27 (2 ):171–180. doi:10.1111/j.1096-0031.2010.00329.x.34875773
