TY - JOUR
AU - Maberly, Stephen C.
AB - Abstract Darwin performed innovative observational and experimental work on the apparently paradoxical occurrence of carnivory in photosynthetic flowering plants. The nutritional use of particulate organic material which also supplies other elements is now known to be widespread in free-living algae as well as in organisms with endosymbiotic algae and with kleptoplastids. In addition to this direct nutritional role, phagotrophy, in the broad sense of internalization of photosynthetic organisms by a eukaryote, is essential for the occurrence of present-day endosymbiotic algae and kleptoplastid-containing protists, and was essential for the origin of plastids themselves. The endosymbiotic phenomena involving photosynthetic organisms clearly played a major role in combining genomes providing different metabolic functions, but, in our opinion, this does not demand a re-appraisal of evolution by natural selection. That the balance of physiological optimization for competition for resources and minimization of losses (e.g. through predation) is a fine one, and thus subject to a complex selective process, is illustrated by the diversity of mixotrophic strategies in extant phytoplankton. Carnivorous plants, Darwin, endosymbiosis, gene transfer to nucleus, kleptoplasty, mixotrophy, oxygenic photosynthesis, phytoplankton, plastids Introduction Darwin's 1875 book ‘Insectivorous plants’ was the inspiration for the excellent review by Ellison and Gotelli (2009) in this series of Darwin Reviews. In the present article we do not revisit the core of Darwin's work on carnivory by flowering plants, but rather consider the role of carnivory (or, more generally, phagotrophy) in the origin(s) of photosynthesis in eukaryotes, and the continued occurrence of phagotrophy in photosynthetic organisms other than flowering plants today. The discussion is in the context of evolution by natural selection (Darwin, 1859), and also touches, where relevant, on Darwin's work on geology (Darwin, 1842) and on zoological systematics (Darwin, 1851a, b, 1854a, b). Our neo-Darwinian approach to topics such as the origin of plastids, and to symbiosis more generally, regards symbiosis in the broad sense (de Bary, 1879) as a source of structural and metabolic innovations that are acted on by natural selection (Maynard-Smith and Szathmary, 1996) rather than as an essentially independent, and dominant, evolutionary mechanism (Margulis, 1998). The innovations come about by combining the genetic resources of organisms that had, of course, a common origin in the ‘Last Universal Common Ancestor’ but which had diverged structurally and nutritionally (Maynard-Smith and Szathmary, 1996). Phagotrophy and eukaryotic photosynthesis: the origin of (almost) all extant plastids Without phagotrophy at the cell level there would be no photosynthesis in eukaryotes. This bold statement may need some qualification, but is essentially correct in that the endosymbiotic event preceding genetic integration to produce plastids involves the internalization of a cyanobacterium using the endomembrane system. This is the same mechanism used at the cell level by extant phagotrophic photosynthetic mixotrophs and presumably also led to the development of what we recognize as mitochondria from ancestral proteobacteria in the past. Oxygenic photosynthesis evolved in cyanobacteria, or their immediate, now extinct, ancestors. All evidence indicates that the spread of photosynthesis into eukaryotes to produce the first chloroplasts was a result of endosymbiosis of a cyanobacterium by a non-photosynthetic eukaryote, a notion mentioned in a footnote by Schimper (1883) and developed by Mereschowsky (1905; see Martin and Kowallik, 1999) (Table 1). This primary endosymbiosis occurred at least 1.2 billion years ago, as indicated by the first known red alga Bangiomorpha (Butterfield, 2000; Deusch et al., 2008; cf. Cavalier-Smith, 2006). This critical (Baldauf, 2008; Deusch et al., 2008; Archibald, 2009; cf. Larkum et al., 2007) primary event gave rise to all extant chloroplasts, with the exception of the cyanelles of the euglyphid rhizarian Paulinella which is discussed below (Table 1). The primary event involved a (mutualistic) symbiosis between a cyanobacterium and a eukaryote which, like all eukaryotes at the time, was non-photosynthetic. For the photosynthetic symbiosis between two potentially free-living organisms to become a genetically integrated organism, with the cyanobacterium ultimately becoming the plastid, requires that the cyanobacterium is vertically transmitted when the host reproduces, and that the cyanobacterium is endosymbiotic; the order in which these events occurred is not known. If the symbiosis was initiated by phagocytosis of the cyanobacterium as a potential food item (phagotrophy), then internalization occurred with the origin of the association. If the original association was as an ectosymbiosis, the ultimate endosymbiosis would still exhibit involved phagocytosis, although it is not clear if it should be termed phagotrophy as the metabolites originating from the cyanobacterium, and not the ingested biomass, constitute the acquired food. Although only eukaryotes have phagocytosis, the β-proteobacterial aphid endosymbiont Buchnera itself has a γ-proteobacterial endosymbiont, with the possible wide evolutionary implications of endosymbisis by bacteria suggested by Lake (2009). Table 1. Number of endosymbioses involved in oxygenic photosynthetic organisms Photosynthetic organism/organelle in extant organism  Number of endosymbioses   One  Two  Three  Four  Plastid derived from a β-cyanobacterial (probably hetero-cystous) endosymbiont  Plastids of Plantae (Glaucocysto-phyta, Rhodophyta, Chlorophyta+Embryophyta)  Plastids of Chlorarachniophyta (rhizarian), Euglenophyta (excavate), Apicomplexas and peridinin-containing Dinophyta (alveolates), Crytophyta, Haptophyta, and Ochrista (chromistans)  Plastids of non-peridinin-containing Dinophyta    Plastid derived from an α-cyanobacterial endosymbiont  Cyanelle of the euglyphid rhizarian Paulinella        Kleptoplastids    Ulvophycean kleptoplastids of the ciliate Symbiodinium and ulvophycean and rhodophycean kleptoplastids of saccoglossan gastropods  Cryptophycean kleptoplastids of the ciliate Myrionecta (?), tribophycean kleptoplastids of sacoglossan gastropods  Kleptoplastids of Dinophysis caudata derived from the prey ciliate Myrionecta  Endosymbiotic cyanobacteria and algae  Cyanobacterial endosymbionts  Chlorophycean, prasinophycean, trebouxiophycean endosymbionts  Ochristan and cryptophycean (?) endosymbionts. Dinophycean (secondary plastids) endosymbionts  Dinophycean (tertiary plastids) endosymbionts (if such exist)  Photosynthetic organism/organelle in extant organism  Number of endosymbioses   One  Two  Three  Four  Plastid derived from a β-cyanobacterial (probably hetero-cystous) endosymbiont  Plastids of Plantae (Glaucocysto-phyta, Rhodophyta, Chlorophyta+Embryophyta)  Plastids of Chlorarachniophyta (rhizarian), Euglenophyta (excavate), Apicomplexas and peridinin-containing Dinophyta (alveolates), Crytophyta, Haptophyta, and Ochrista (chromistans)  Plastids of non-peridinin-containing Dinophyta    Plastid derived from an α-cyanobacterial endosymbiont  Cyanelle of the euglyphid rhizarian Paulinella        Kleptoplastids    Ulvophycean kleptoplastids of the ciliate Symbiodinium and ulvophycean and rhodophycean kleptoplastids of saccoglossan gastropods  Cryptophycean kleptoplastids of the ciliate Myrionecta (?), tribophycean kleptoplastids of sacoglossan gastropods  Kleptoplastids of Dinophysis caudata derived from the prey ciliate Myrionecta  Endosymbiotic cyanobacteria and algae  Cyanobacterial endosymbionts  Chlorophycean, prasinophycean, trebouxiophycean endosymbionts  Ochristan and cryptophycean (?) endosymbionts. Dinophycean (secondary plastids) endosymbionts  Dinophycean (tertiary plastids) endosymbionts (if such exist)  Based on Table 1 of Raven (1997), plus Williams and Walker (1999), Gustafson et al. (2000), Rumpho et al. (2000), McManus et al. (2004), Cavalier-Smith (2006), Curtis et al. (2006), Hansen and Fenchel (2006), Bhattacharya et al. (2007), Larkum et al. (2007), Usher et al. (2007), Baldauf (2008), Burkholder et al. (2008), Casalduero and Muniain (2008), Deusch et al. (2008), Nagai et al. (2008), Nowack et al. (2008); Park et al. (2008), Rumpho et al. (2008), Teugels et al. (2008), Archibald (2009), Evertson and Johnsen (2009), Garcia-Cuetos et al. (2009); Nakayama et al. (2009), and Vieira et al. (2009). Sanchez-Puerta and Delwiche (2008) propose that ochristan and alveolate plastids involve tertiary endosymbiosis of haptophyte cells, with the non-peridinin-containing dinoflagellates having plastids derived from quaternary endosymbiosis. View Large The loss from the cyanobacterial genome of a single gene required for independent life would have made the symbiosis obligate (Usher et al., 2007). This loss could be a loss from the endosymbiosis as a whole, or a loss from the endosymbiont to the host genome; evidence exists for both kinds of event. During the establishment of the symbiosis some cyanobacterial genes were lost, and a majority of the rest transferred to the host nucleus, leaving a minority, all involved in plastid function, in the plastid genome (Martin et al., 2002; Bhattacharya et al., 2007; Deusch et al., 2008). Most of the cyanobacterially derived genes involved in plastid function were transferred to the nucleus, but the majority of the genes transferred are involved in processes other than those intimately associated with photosynthesis and plastid maintenance. Deusch et al. (2008) estimated that 16–18% of nuclear genes in a red alga, a green alga, and two flowering plants originated from the plastid ancestor. The relatively constant percentage of cyanobacterial genes, despite the 6-fold range in total gene number (5 300–32 000) among the four eukaryotes, may seem remarkable in view of the postulated single primary endosymbiotic event, but is, in principle, explicable in terms of genome size reduction in the very small cells of the red alga and genome duplication(s) in the flowering plants. Moustafa and Bhattacharya (2008) come to rather different conclusions: they suggest in a specimen analysis that only 3.5–6% of the nuclear genes in the completely sequenced green alga Chlamydomonas reinhardtii are derived from the cyanobacterial plastid ancestor; most of the transferred genes are related to plastid function (Reyes-Prieto et al., 2006). There are also a number of genes in the Kingdom Plantae derived from chlamydial bacteria, again mainly involved with plastid functions (Moustafa et al., 2008); this horizontal gene transfer could also have involved phagotrophy. The suspected unique symbiotic event that produced all plastids, other than those of Paulinella, gave rise directly to the photosynthetic glaucocystophytes, rhodophytes, and the green algae and their descendants, the embryophytic plants, together constituting the Plantae (Table 1). Deusch et al. (2008) suggest that the plastid ancestor was a heterocystous diazotrophic cyanobacterium, based on a comparison of the completely sequenced genomes of nine cyanobacteria with those of a red alga, a green alga, and two flowering plants. However, the completed cyanobacterial genomes do not explore the complete range of cyanobacterial diversity. The further spread of plastids through the eukaryotes as genetically integrated organelles involved secondary and tertiary endosymbiosis through phagocytosis, and thus presumably reflect mixotrophic modes of nutrition. Secondary endosymbiosis involves a green or red alga as an endosymbiont in a non-photosynthetic eukaryotic host. When the photosynthetic symbiont is a green alga the outcome was the chlorarachniophyte (rhizarian) and euglenoid (excavate-discicristate) algae (Table 1). With a red algal photosynthetic symbiont, the product was the chromalveolate algae (Graham and Wilcox, 2000). The chromalveolates comprise, in the alveolates, the dinoflagellates (Dinophyceae, of which some extant species are secondarily, i.e. apomorphically, non-photosynthetic), and Chromera and the non-photosynthetic apicomplexan parasites that retain plastids as their apicoplasts in the alveolates (Moore et al., 2008), as well as the ciliates for which genomic evidence suggests loss of plastids (Table 1). The photosynthetic members of the chromist clade of the chromalveolates are the cryptophytes, haptophytes, and heterokontophytes (=ochristans); in all these three groups there are extant non-photosynthetic members. Tertiary endosymbiosis involves a range of photosynthetic symbionts and a non-photosynthetic host eukaryote which has lost an earlier capacity for photosynthesis. These tertiary endosymbioses appear to be restricted to the dinoflagellates (Table 1, although see below), a group which today shows great nutritional variety. The number of independent endosymbiotic events involved in the original acquisition(s) of photosynthesis in the chromalveolates is still a matter of debate, although it seems clear that two independent events were involved for the chlorarachniophytes and the euglenoids (Archibald, 2009). One view is that there were separate secondary endosymbioses for the alveolates and for the chromists, possibly with more than one event within the chromists (Falkowski and Raven, 2007; Archibald, 2009). A recent view (Sanchez-Puerta and Delwiche, 2008) is that a single secondary endosymbiosis occurred in the cryptophycean–haptophycean clade, with the plastid in the ochrists and alveolates arising from tertiary endosymbiosis. Accordingly, the plastids in the dinoflagellates, other than those containing the carotenoid peridinin and Form II Rubisco, arose from quaternary endosymbioses. A further twist was added by Moustafa et al. (2009), who found more genes of green algal (prasinophycean) than of red algal origin in the nuclei of the two completely sequenced diatom genomes, those of Phaeodactylum tricornutum and Thalassiosira pseudonana. This is consistent with the photosynthetic ancestors of diatoms and hence the ancestor of (chrom-)alveolates having a green alga-derived plastid before the red alga-derived plastids that they now have. As pointed out by Dagan and Martin (2009), the findings of Moustafa et al. (2009) are likely to be controversial. Phagotrophy and eukaryotic photosynthesis: Paulinella The occurrence of photosynthesis and plastids in the euglyphid amoeba was for a long time associated with the superficially similar organelles in the glaucocystophyte algae such as Cyanophora paradoxa. Similarities between the ‘cyanelles’ in these two phyologentically widely separated organelles included the retention of the peptidoglycan wall of the cyanobacterial ancestor which is absent from other plastids, and the presence of a carboxysome-like structure: carboxysomes are otherwise confined to bacterial autotrophs, including cyanobacteria. A possible functional role for the combination of these two ancestral (symplesiomorphic) features was suggested by Raven (2003): see also Fathinejad et al. (2008). Determination of the plastid genome size showed that while the Cyanophora cyanelle genome had similar gene numbers to the other plastids examined, Paulinella had many more genes in the cyanelle genome. While this could be rationalized as reflecting further progress along the path of gene loss from the glaucocystophyte cyanelles than from the cyanelle of Paulinella, sequencing showed that the Paulinella cyanelle genome was derived from a different group of cyanobacteria (Nowack et al., 2008) (Table 1). Instead of the probably heterocystous, multicellular β-cyanobacterial ancestor of the great bulk of plastids, including the cyanelles of the glaucocystophytes, the Paulinella cyanelle was derived from a unicellular α-cyanobacterium from the Prochlorococcus–Synechococcus clade (Nowack et al., 2008). Despite the larger number of genes in the Paulinella cyanelle genome than in other plastids, it is still by a significant margin smaller than the α-cyanobacterial genome. Among the missing genes are those for some essential biosyntheses, e.g. those of the tricarboxylic acid cycle and the synthesis of most amino acids, so that the cyanelle is clearly unable to perform independent photolithotrophic growth. The required genes could have been transferred to the host genome, or lost from the symbiosis; in either case genes in the host genome are needed for the synthesis and functioning of the cyanelles. While Nowack et al. (2008) provided no evidence to distinguish ‘loss’ from ‘transfer’ as the cause of the absence from the cyanelle genome of any or all of the missing genes, Nakayama and Ishida (2009) have shown that one photosynthetic gene (PsaE, of incompletely known function) has been transferred to the host genome. This work on Paulinella has thus provided clear evidence of a second independent origin of a photosynthetic organelle. Phagotrophy and eukaryotic photosynthesis: kleptoplastids Kleptoplastids are plastids derived from algal food by certain phagotrophs, and retained in a photosynthetically functional state for up to several months (Rumpho et al., 2000, 2008; Evertson et al., 2007; Casalduero and Muniain, 2008; Evertson and Johnsen, 2009; Vieira et al., 2009). Kleptoplastids are also suspected to function in inorganic nitrogen assimilation in organisms that would otherwise typically be unable to utilize such forms of nutrition (Teugels et al., 2008; cf. the possible nitrogen nutrition connection in the origin of primary plastids from diazotrophic β-cyanobacteria: Deusch et al., 2008). It is likely that all ingested photosynthetic prey remain capable of some degree of photosynthesis after ingestion into their protistan predator. However, changes in the thylakoid membranes of the wall-less chlorophycean Dunaliella primolecta occur rapidly after ingestion by the heterotrophic dinoflagellate Oxyrrhis marina (Öpik and Flynn, 1989), suggesting that the development of a symbiotic relationship would require that the ingested phototroph is at least partially resistant to digestion. Kleptoplasty seems to be confined to the marine environment, and has been demonstrated in several flagellates and ciliates as well as in some sacoglossan gastropod molluscs (Williams and Walker, 1999; McManus et al., 2004; Park et al., 2008) (Table 1). In these cases the phagotrophy by which the plastids are ingested is by necessity at the cell level. In the metazoan sacoglossans this involves phagotrophy at the gut lumen level followed by partial intracellular digestion in the gut epithelium cells; such intracellular digestion (supplementing extracellular digestion) is relatively widespread in molluscs (Table 2). Table 2. Phylogenetic and ecological diversity of phagotrophy in O2-evolving phototrophs with genetically integrated plastids, with endosymbiotic algae, or with kleptoplastids Taxon    Occurrence of phagotrophy in photosynthetic membersa  Ecological status of phagotrophic photosynthetic membersb  Plantae Glaucocystophyta    – (i)  NA  Plantae Rhodophyta    – (i)  NA  Plantae Chlorophyta  Prasinophyceae  + (i)  sw  Plantae Embryophyta  Magnoliophyta  + (i)  T, rhi, fw, ple  Plantae Rhodophyta    – (i)  N.A.  Opisthokonta Porifera    + (c)  fw, sw  Opisthokonta Cnidaria    + (c)  fw, sw  Opisthokonta Turbellaria    + (c)  sw  Opisthokonta Mollusca  gastropod  + (kp) (c)  sw    bivalve  + (c)  fw, sw  Chromista Cryptophyta    + (i)  fw, sw, pla  Chromista Haptophyta    + (i)  fw, sw, pla  Chromista Ochrista  Chrysophyceae  + (i)  fw, sw, pla    Synurophyceae  –  NA    Bacillariophyceae  –  NA    Phaeophyceae  –  NA    Silicoflagellata  –  NA  Alveolata Apicomplexa    –? (i)  NA  Alevolata Dinophyta    + (i). (c), (kp)  fw, sw  Alveolata Ciliata    + (c), (kp)    Rhizaria Acantharia    + (c)  sw, pla  Rhizaria Chlorarachniophyta    + (i)  sw, pla  Rhizaria Euglyphia Paulinella    + (i)  fw, sed  Rhizaria Foraminifera    + (c)  sw, pla, sed  Rhizaria Radiolarians    + (c)  sw, pla  Excavata Euglenophyta    – (i)  NA  Taxon    Occurrence of phagotrophy in photosynthetic membersa  Ecological status of phagotrophic photosynthetic membersb  Plantae Glaucocystophyta    – (i)  NA  Plantae Rhodophyta    – (i)  NA  Plantae Chlorophyta  Prasinophyceae  + (i)  sw  Plantae Embryophyta  Magnoliophyta  + (i)  T, rhi, fw, ple  Plantae Rhodophyta    – (i)  N.A.  Opisthokonta Porifera    + (c)  fw, sw  Opisthokonta Cnidaria    + (c)  fw, sw  Opisthokonta Turbellaria    + (c)  sw  Opisthokonta Mollusca  gastropod  + (kp) (c)  sw    bivalve  + (c)  fw, sw  Chromista Cryptophyta    + (i)  fw, sw, pla  Chromista Haptophyta    + (i)  fw, sw, pla  Chromista Ochrista  Chrysophyceae  + (i)  fw, sw, pla    Synurophyceae  –  NA    Bacillariophyceae  –  NA    Phaeophyceae  –  NA    Silicoflagellata  –  NA  Alveolata Apicomplexa    –? (i)  NA  Alevolata Dinophyta    + (i). (c), (kp)  fw, sw  Alveolata Ciliata    + (c), (kp)    Rhizaria Acantharia    + (c)  sw, pla  Rhizaria Chlorarachniophyta    + (i)  sw, pla  Rhizaria Euglyphia Paulinella    + (i)  fw, sed  Rhizaria Foraminifera    + (c)  sw, pla, sed  Rhizaria Radiolarians    + (c)  sw, pla  Excavata Euglenophyta    – (i)  NA  Based on Raven (1997), plus Kempf (1984), Fitt et al. (1986), Calderon-Saenz and Schnetter (1989), Maranger et al. (1998), Granéli and Carlson (1999), Williams and Walker (1999), Gustafson et al. (2000), Jones (2000), Farmer et al. (2001), Bell and Laybourn-Parry (2003), Pierce et al. (2003), McManus et al. (2004), Cavalier-Smith (2006); Curtis et al. (2006), Hansen and Fenchel (2006), Evertson et al. (2007), Baldauf (2008), Burkholder et al. (2008), Deusch et al. (2008), Moore et al. (2008); Nowack et al. (2008), Teugels et al. (2008), Archibald (2009), Evertson and Johnsen (2009), Garcia- Cuetos et al. (2008), Moustafa et al. (2009), Nakayama and Ishida (2009), and Raven and Giordano (2009). a Key for occurrence of phagotrophy in photosynthetic members: (i) genetically integrated plastids; (c) symbiotic algal cells; (kp) kleptoplastids. b Key for ecological status of phagotrophic photosynthetic members: fw, freshwater, sw, seawater; hap, haptophyte (attached to a large-grained solid substrate); NA, not applicable; pla, planktonic (small, in water body); ple, pleustophyte (large, in water body, or resting on sediment); rhi, rhizophyte (large, roots or rhizoids in a fine-grained sediment); sed, small, in or on sediment. View Large Kleptoplasty can involve some fascinating aspects of natural history. One is the case of Strombidium spp., a ciliate of high intertidal rock-pools. Molecular phylogenetic analysis shows that the kleptoplastids originate from the ulvophycean green alga Ulva sp. of the tubular Enteromorpha morphology that also occurs in the rock-pools (McManus et al., 2004). The oral equipment of ciliates is not capable of ingesting the vegetative thallus of the macroalga, and McManus et al. (2004) found that the ciliates swarmed round the sites on the thallus that produce motile reproductive cells and ingested these meroplanktonic flagellate cells. Another example is the sequence of phagotrophic events that account for the kleptoplastids of the dinoflagellate Dinophysis caudata (Nagai et al., 2008; Park et al., 2008). Here molecular phylogenetic analysis shows that the plastids in this dinoflagellate arise from the prey item, the ciliate Myrionecta rubra (formerly Mesodinium rubrum) (Park et al., 2008). While there is evidence consistent with the view that plastids of Myrionecta are kleptoplastids derived from cryptophyte prey items (Gustafson et al., 2000), a subsequent study (Hansen and Fenchel, 2006) supports the older view that the chloroplasts of Myrionecta are still within ingested cryptophyte cells. Johnson et al. (2006, 2007) argue for the use of kleptochloroplasts (i.e. removed from the prey cell) but that the cryptophyte nuclei are also retained to aid control of the captured chloroplasts. A further possibility for the genus Dinophysis was raised by Garcia-Cuetos et al. (2009), who provide evidence that Dinophysis acuminata has permanent chloroplasts, rather than kleptoplastids, of cryptophyte origin. Since the cryptophyte plastids arose by secondary endosymbiosis from red algae, the sequence leading to plastids in Dinophysis spp. involved one or more of the following alternatives: (i) cyanobacterium → green algal primary plastid subsequently lost (Moustafa et al., 2009) - red algal primary plastid – cryptophyte secondary plastid → Myrionecta rubra primary kleptoplastid (Gustafson et al., 2000) → Dinophysis caudata secondary kleptoplastid, or (ii) cyanobacterium → green algal primary plastid subsequently lost (Moustafa et al., 2009) - red algal primary plastid → cryptophyte secondary plastid → symbiotic cryptophyte symbiosis in Myrionecta rubra (Hansen and Fenchel 2006) → Dinophysis caudata primary kleptoplastid, or (iii) cyanobacterium → green algal primary plastid subsequerntly lost (Moustafa et al., 2009) → red algal primary plastid → cryptophyte secondary plastid → cryptophyte endosymbiosis in Myrionecta rubra (Hansen and Fenchel, 2006) → cryptophyte endosymbisis in Dinophysis acuminata (Garcia-Cuetos et al., 2009). The longevity and degree of photosynthetic function of kleptoplastids has been categorized for sacoglossans by Williams and Walker (1999) and Evertson et al. (2007). Long-term photosynthetic function of kleptoplastids poses the problem of repair of ‘wear and tear’ to enzymes and pigments since many of the genes coding for the photosynthetic proteins and the enzymes of pigment synthesis are in the algal nuclear rather than the plastid genome. While some of the hosts, for example, ciliates and dinoflagellates, were at one time photosynthetic, others, for example, marine sacoglossan gastropods, were never photosynthetic. In both cases, the host genome would, at least before kleptoplastidy began, lack the extraplastidic genes needed to maintain most of the plastid proteins active in the face of damage and proteolysis. Flynn and Mitra (2009) suggest that the control of the longevity of the interaction may hinge on the flow of metabolites from the kleptochloroplasts repressing the production (or activity) of digestive enzymes by the host. If the signal for this regulation is the cellular C-status, then the failure of the kleptochloroplasts will automatically signal the release of enzymes responsible for their destruction, and thence for the predatory activity required for the acquisition of their replacement. A suggestion that has proved difficult to test is that the host nuclear genome has acquired photosynthetic genes from host algae that can function in maintaining the kleptoplastids in a functional state. However, it is now known that such gene transfer has occurred (Pierce et al., 2003; Rumpho et al., 2008) (Table 3). The precise origin of the algal genes in the host genome is not clear; it is known that, despite some earlier assumptions (Raven et al., 2001), a sacoglossan species can harbour plastids from four algal species from two genera, with plastids from more than one species in a single digestive cell (Curtis et al., 2006). In almost all cases the plastids are from coenocytic green algae of the class Ulvophyceae (e.g. species of Acetabularia, Caulerpa, Codium, Halimeda, and Penicillus) although in a few sacoglossans they are from large-celled floridiophycean red algae such as Griffithsia sp., or the coenocytic tribophycean ochristan alga Vaucheria. It is possible, or likely, that the genes in the nuclear genome are interchangeable in terms of the species of green or red or tribophyte algae. Table 3. Probable extent of acquisition of materials, energy and information from items subject to phagotrophic ingestion into cells as purely food items (phagotrophy) or as intracellular symbionts or as kleptoplastids, or which occur as extracellular symbionts on or within multicellular host structures   Materials  Energy  Information  Endosymbiosis (truly intracellular)  X  XXX  XX  Kleptoplasty  X  XXX  X  Ectosymbiosis (not truly intracellular or clearly extracellular in multicellular organisms)  X  XXX  X  Phagotrophy  XXX  XX  –    Materials  Energy  Information  Endosymbiosis (truly intracellular)  X  XXX  XX  Kleptoplasty  X  XXX  X  Ectosymbiosis (not truly intracellular or clearly extracellular in multicellular organisms)  X  XXX  X  Phagotrophy  XXX  XX  –  See text for further discussion. View Large One intriguing aspect of the horizontal gene transfer from the nuclear genes of the algal food to the host nucleus is that the algal genes seem to be vertically transmitted, although they were transferred into somatic (i.e. gut epithelia) rather than germ-line cells of the metazoan (Pierce et al., 2003; Rumpho et al., 2008). Spatial separation from reproductive cells has been suggested as a reason for the lack of vertical transmission of, for example, diazotrophic symbiotic bacteria in root (occasionally stem) nodules of flowering plants (Douglas, 1994). Kleptoplasty is an intriguing phenomenon. The capacity for photosynthesis that it permits does not fit the definition of symbiosis of de Bary (1879), i.e. ‘the living together of differently named organisms’, since the kleptoplastids and some algal nuclear genes transferred to the host nucleus do not constitute an organism. The absence of the complete genetic apparatus for plastid function in the organisms with kleptoplastids also means that vertical transmission is impossible. The process of chloroplast acquisition must be repeated, typically within a generation of the ‘host’, with no replication of the plastid. Flynn and Mitra (2009) liken the process to asset-stripping, where the ingested structures are run for as long as possible, with minimal maintenance, and then digested when they fail to perform. Carnivory and eukaryotic photosynthesis: intracellular symbiosis conferring photosynthetic capacity on phagotrophs The processes through which phototrophs become integrated into non-phototrophic organisms represent the first stage envisaged for the origin of plastids outlined earlier in this paper. There are extant examples where the non-phototroph partner is of metazoan animal, rather than protistan, origin; the following list includes metazoan, as well as unicellular and colonial protistan and algal, hosts. The benthic metazoans that retain phagotrophy and have photosynthetic endosymbionts are close functional analogues of the insectivorous flowering plants studied by Darwin (1875). The endosymbioses involve cyanobacterial photobionts in some marine sponges and cyanobacterial diazotrophs in some marine and freshwater diatoms, green (trebouxiophycean) algae in some freshwater sponges, cnidarians, and bivalve molluscs, green (prasinophycean) algae in some marine trematodes, and dinoflagellates (Symbiodinium) in some marine rhizarians (foraminiferans, radiolarians, and acantharians), cnidarians and bivalve molluscs (Kempf, 1984; Fitt et al., 1986; Smith and Douglas, 1987; Farmer et al., 2001; Usher et al., 2007; Raven and Giordano, 2009) (Table 1). Similar endosymbioses also exist for many freshwater protistan ciliates such as Euplotes, Ophrydium, and Stentor where endosymbiotic Chlorella-like cells replace the photosynthetic capacity lost by ciliates during evolution. In most cases the host continues to feed phagotrophically (Tables 2, 3). For the metazoan hosts there are cases in which the symbionts entered the organism by phagotrophy at the whole organism level, but are not intracellular. Among the molluscs the photosynthetic marine bivalves have Symbiodinium in gut diverticulae that form in response to the symbionts (Fitt et al., 1986; Farmer et al., 2001), while in photosynthetic nudibranch gastropods the Symbiodinium is apparently intracellular (Kempf, 1984), as are the kleptoplastids of sacoglossan gastropods considered above. Even for unicellular hosts there is the possibility that the symbionts are in invaginations of the plasmalemma rather than truly within the cells, for example, the marine diazotrophic cyanobacterium Richelia symbiotic with species of Hemiaulus and Rhizosolenia (Usher et al., 2007; Wouters et al., 2009). These extracellular symbionts can be vertically transmitted, as can the clearly extracellular (but within cavities in the host) diazotrophic cyanobacteria of the floating aquatic fern Azolla (Usher et al., 2007). There are clearly possibilities, in the origin of chloroplasts, that vertical transmission of symbionts preceded endosymbiosis by phagotrophy or an analogous process, and that gene transfer to the host nucleus could also have occurred before endosymbiosis. It is generally considered that the photobionts are capable of independent existence, and indeed many have been cultured. This situation contrasts with that involving Paulinella cyanelles. However, culture of the photobiont away from the host has not been achieved in some cases, such as the diazotrophic cyanobiont Richelia from some planktonic marine diatoms. It is not known if a gene or genes needed for independent growth is/are absent from the genome of Richelia, or whether any cyanobacterial genes have been transferred to the diatom genome (with or without deletion from the cyanobacterial genome) (Usher et al., 2007; Wouters et al., 2009). Raven (1993) suggested that such transfer to the host genome, and expression in the host, could rationalize the occurrence of high-affinity influx transport mechanisms, for example, for ammonium, inorganic carbon, and phosphate in the plasmalemma of the host while such transporters are not found in non-symbiotic relatives of the host. This possibility apparently remains untested, but could be involved for both intracellular symbionts and for extracellular symbionts of multicellular organisms where the symbionts are within the organism but not within cells. Examples where the host may be involved in inorganic carbon concentrating mechanisms for extracellular symbionts are the marine intertidal lichen Lichina and the bivalve molluscs (Raven et al., 1990; Raven, 1993). An aspect of photosynthetic symbioses that is well-established is the use of pre-existing host structures and functions in the symbiosis in relation to photosynthetic functions (Raven, 1993; Raven and Giordano, 2009). Examples are flagellar (sponges) and circulatory (bivalves) mass flow of solution in supplying inorganic carbon for photosynthesis, and muscle-based changes in exposure of photosynthetic structures to light with apparent roles in avoiding photodamage at high irradiances and maximizing absorption of low irradiances (Raven, 1993). A link to Darwin in this group of phagocytosed phototrophs is the crucial role of mineralized cnidarians, with contributions from sponges and giant clams on coral reefs (Darwin, 1842; Herbert, 2005). However, there is no evidence for photosynthetic symbionts in the group of sessile metazoans in whose systematics Darwin specialized, the barnacles (Darwin, 1851a, b, 1854a, b), although intracellular non-photosynthetic symbiotic bacteria do occur in motile arthropods (e.g. Buchnera in aphids and Walbachia in many insects; Douglas and Raven, 2003). Another large group of sessile metazoans without photosynthetic symbioses are the ectoprocts. Possibly related to the gene loss from the photobiont is the occurrence of vertical or horizontal transmission of the photobiont. Vertical transmission is known for photobionts of the trematode Convoluta and of some sponges and cnidarians, but does not necessitate gene loss from the photobiont. The horizontal transmission known for some sponges and cnidarians, and all molluscs, would not necessarily have precluded gene loss if the photobionts could survive outside the host for an appropriate period to transition between hosts. The discussion here has emphasized intracellular symbioses. Extracellular symbioses, such as those between fungi and green algae or cyanobacteria in lichens, are presumably less likely to involve gene transfer from photobiont to host even when there is vertical transmission of symbionts. The reasoning here is a lack of opportunity, especially if both host and photobiont have cell walls, and a lack of evolutionary requirement if the photobiont is in contact with the medium and can acquire external nutrients without mediation by the host, for example, in most lichens. Phagotrophic nutrition in photosynthetic organisms Occurrence of phagotrophic nutrition in eukaryotes with genetically integrated photosynthesis Several algae with their typical nutritional mode of genetically integrated photolithotrophy also express the ancestral feeding mode of phagotrophy with intracellular digestion (Jones, 1994, 2000; Raven, 1997). Examples are found in the green algae (Pyramimonas), chlorarachniophyte and euglyphid (Paulinella) rhizarians, dinoflagellate alveolates, chrysophycean, raphidophycean, and tribophycean ochristans, cryptophycean cryptophytes and prymnesiophycean haptophytes (Calderon-Saenz and Schnetterer, 1989; Jones, 1994, 2000; Raven, 1997, Granéli and Carlson, 1999; Graham and Wilcox, 2000; Bell and Laybourn-Parry, 2003; Burkholder et al., 2008; Liu et al., 2009) (Table 2). Despite the range of photolithotrophic (and photosaprotrophic), and of phagotrophic, euglenoids, there appear to be no examples of euglenoids that combine phototrophy with phagotrophy (Graham and Wilcox, 2000). The extent of phototrophy and phagotrophy in these potentially mixotrophic algae varies widely as a function of the inherent capacities for the two trophic modes in a particular alga, whether some level of obligatory photosynthesis is required, and on the environmental conditions. Although not invoked in empirical classifications of functional forms of planktonic mixotrophs (Stoecker, 1998; Zubkov, 2009; Raven, 2009), Flynn and Mitra (2009) found the extent of obligatory photosynthesis to be a core requirement when modelling these organisms. The other group of carnivorous photosynthetic organisms with genetically integrated plastids are the phagotrophic (insectivorous) flowering plants (Darwin, 1875; Adlassnig et al., 2005; Ellison and Gotelli, 2009) (Table 2). These organisms have extracellular digestion of prey, just as is the case for particulate nutrients in fungi, arthropods, and vertebrates. The secretion of digestive (in this case proteolytic) enzymes is now known for axenic non-carnivorous plants, although here there is evidence for endocytotic uptake of protein by the cells prior to proteolysis (Paungfoo-Lonhierre et al., 2008). In several cases the sensing and trapping of prey involves a propagated action potential; these electrical phenomena are widespread among organisms with genetically-integrated photosynthesis (Findlay and Hope, 1976; Taylor, 2009). As with phagotrophy in the mixotrophic algae, the extent of carnivory in carnivorous plants varies with genotype and circumstances, although unlike the common occurrence of the former (Jones, 1994, 2000; Stoecker, 1998; Jeong et al., 2005; Unrein et al., 2007; Zubkov and Tarren, 2008) mixotrophy in higher plants remains very much the exception. An interesting possible link between phagotrophy and carnivorous higher plants is the occurrence of commensal microalgae in traps of Genlisea, Sarracenia, and Utricularia (Dudley, 1984; Gebühr et al., 2006; Plachno and Wolowski, 2008; Sirova et al., 2009). For at least Genlisea and Utricularia, there is evidence consistent with the algal commensals providing organic carbon to the host (Plachno and Wolowski, 2008; Sirova et al., 2009). The organic carbon could be derived from photosynthesis using inorganic carbon in the trap fluid, and perhaps by mixotrophy. Phagotrophic nutrition in eukaryotes with photosynthesis resulting from symbiotic cyanobacteria and algae or with kleptoplastids The minimum requirement for phagotrophy in these eukaryotes is in providing the photobionts in each new cohort when there is horizontal photobiont transmission. In the case of kleptoplasty there is a continuing need for feeding even when the plastids can function in the host for periods similar to the life span of the host (Rumpho et al., 2000; Evertson et al., 2007), since the kleptoplastids cannot divide and they will become diluted by growth of the host. When transmission is vertical there is no such need for phagotrophy to provide new symbionts, unless the host is ‘shopping’ for different genotypes of symbiont in response to, for example, environmental change or the symbionts have been expelled during some environmental perturbation (Lewis and Coffroth, 2004). Flynn and Mitra (2009) suggest that a role may be the acquisition of an anti-grazer capacity through involvement of the phototrophic component; this would explain the results of Perez et al. (1997) in which mixotrophic ciliates were noted to grow more slowly than their heterotrophic counterparts but also to be grazed more slowly (and hence, ultimately, to be at an advantage). These considerations do not address the extent to which host nutrition depends on phagotrophy, which may involve nutritional items other than those provided directly or indirectly by the activity of the photobiont. As with phagotrophy in potentially mixotrophic organisms with a genetically integrated capacity for photosynthesis, the extent of phototrophy and of phagotrophy in hosts with photobionts or kleptoplastids depends on both genetics and environment. How is the occurrence of phagomixotrophy, and its extent in a given organism under varying conditions, rationalized in terms of Natural Selection? What do phagomixotrophs get from phagotrophy? Organisms depend on information (genetic and environmental), energy and material (chemical) resources (Raven, 1984). Phagotrophy can, as indicated above, yield genetic innovation, bringing together genomes of very different evolutionary histories and metabolic potentials (Table 3). Aside from this, phagotrophy has the more mundane capacity to supply organic and inorganic chemical resources and hence, from organic materials, energy (Jones, 1994, 2000; Raven, 1995, 1997; Maranger et al., 1998; Granéli and Carlson, 1999; Raven et al., 2001; Weisse, 2002; Kamjunke et al., 2007; Paffenhöfer et al., 2007; Burkholder et al., 2008; Casalduero and Munian, 2008; Teugels et al., 2008; Evertson and Johnsen, 2009; Jones et al., 2009) (Table 3). This resource acquisition is more extensive than the single nutrient stress that may have originally promoted the act of phagotrophy. This is because of the stoichiometric implications of ingesting a complete nutritional source, rather than engaging in the usual primary phototrophic nutritional activity of piece-wise acquisition of nutrients and light (Flynn and Mitra, 2009). Thus complete packages of C, N, P, Fe (etc) are ingested, and the organisms so consumed may enable access to nutrient sources otherwise inaccessible (or poorly accessible) to the phototroph. There is also the possibility of acquiring iron by phagotrophy of non-living (but not necessarily completely inorganic) colloidal iron particles: this element can limit phytoplankton growth in significant areas of the world's oceans, the particulate iron not being readily accessible to non-phagotrophs (Nodwell and Price, 2001). Several attempts have been made to describe, measure and/or model costs and benefits of phagotrophy to photosynthetic organisms (Raven, 1995; Dolan and Pérez, 2000; Tittel et al., 2003; Adlassnig et al., 2005; Troost et al., 2005; Flöder et al., 2006; Ellison and Gotelli, 2009; Flynn and Mitra, 2009; Fig. 1). Raven (1995), for example, points out from the limited data available for planktonic members of the Chrysophyceae sensu stricto (i.e. excluding the Synurophyceae with no phagomixotrophic members), that pure photolithotrophs have lower maximum specific growth rates than phagomixotrophs, which, in turn, have lower rates than pure phagotrophs. This was attributed to the fraction of cellular resources devoted to the photolithotrophic plus the phagotrophic apparatus in the three cases, leaving a larger fraction of resources such as carbon, nitrogen, and phosphorus for the core processes of metabolism downstream of the provision of reduced organic carbon and nitrogen within the cell in the order pure phagotroph>phagomixotroph>pure photolithotroph (Raven, 1995). Operation of the dynamic mechanistic model of Flynn and Mitra (2009; Fig. 1) provides a platform to explore the ecological payback to compensate for such apparently contradictory negative costs. Critical amongst these, however, is the importance of net growth (gross growth – losses through predation etc), and recognition that without an understanding of the operations of the ecosystem in which these organisms reside we will remain ignorant of the natural selection pressures acting on them. Tittel et al. (2003) argue from their mathematical modelling that phagomixotrophs can, via trophic level interactions, outcompete specialist photolithotrophs and specialist phagotrophs, and can help to explain the occurrence of deep chlorophyll maxima in oligotrophic waters. Fig. 1. View largeDownload slide Different configurations of mixotroph physiology described by the model of Flynn and Mitra (2009). Bottom left: mixotroph performing photosynthesis using kleptoplastids from captured prey; digestion of chloroplasts is restricted by products of carbon fixation, so that when they fail then they are replaced. Top left: phototrophic mixotrophy obtaining additional nutrition by prey consumption. Int1 indicates competition for space within cells between chloroplasts and food vacuoles (only for phagotrophic species), Int2 indicates interactions between biosynthesis based upon newly fixed carbon or from prey digestion. Top right: combination of interactions from mixotroph and kleptochloroplastidic carbon-fixation and prey digestion. (This figure is available in colour at JXB online.) Fig. 1. View largeDownload slide Different configurations of mixotroph physiology described by the model of Flynn and Mitra (2009). Bottom left: mixotroph performing photosynthesis using kleptoplastids from captured prey; digestion of chloroplasts is restricted by products of carbon fixation, so that when they fail then they are replaced. Top left: phototrophic mixotrophy obtaining additional nutrition by prey consumption. Int1 indicates competition for space within cells between chloroplasts and food vacuoles (only for phagotrophic species), Int2 indicates interactions between biosynthesis based upon newly fixed carbon or from prey digestion. Top right: combination of interactions from mixotroph and kleptochloroplastidic carbon-fixation and prey digestion. (This figure is available in colour at JXB online.) The use of an alternative carbon supply from phagotrophy might be related to the absence of an inorganic carbon concentrating mechanism (CCM) in the Chrysophyceae and in the aquatic carnivorous flowering plants Aldovandra and Utricularia (Adamec, 1995, 2009; Maberly and Madsen, 2002; Raven et al., 2005; Maberly et al., 2009). It must be noted that not all of the photosynthetic Chrysophyceae have a phagotrophic capacity, and the sister-class Synurophyceae have no phagomixotrophic members yet also lack CCMs (Raven et al., 2005; Bhatti and Colman, 2008; Maberly et al., 2009). However, Clegg et al. (2007) found that a model based on observed tactic responses to gradients of light, CO2, and O2 in the laboratory accounted for the natural distribution of the phagomixotrophic chrysophycean flagellate Dinobryon sertuloides at least as well as it did with four non-phagotrophic flagellate phytoplankton species. Clegg et al. (2007) suggested that high CO2 levels could account for much of the distribution of Dinobryon without invoking the distribution of bacterial prey. More generally, there is a need for more work on possible mechanisms of prey detection in phagomixotrophs. Troost et al. (2005) consider the evolutionary specialization of phagomixotrophs into pure photolithotrophs or pure phagotrophs. They conclude that intrinsic properties (e.g. the costs of maintaining the capacity to express phagotrophy and mixotrophy, and the production and operation of these two trophic mechanisms) are the important determinants of evolutionary outcomes; extrinsic factors, such as light and nutrient availability, have a negligible influence on the outcome. Such an observation could be considered consistent with the fact that the great majority of photosynthetic eukaryotes that acquired photosynthesis by primary, secondary or tertiary endosymbiosis involving phagotrophy have not retained phagotrophic nutrition. Thus, while some seaweeds can gain nutritional advantage from non-phagotrophic use, via microbial activity, of particulate organic matter deposited on the thallus (Schaffelke, 1999), phagomixotrophy by marine macroalgae is only known from fiction (Martel, 2002). However, while the bulk of the biogeochemical fluxes of the oceans may be driven by what are essentially pure phototrophs, these organisms also reside in a strongly physical-directed environment dominated by the availability of inorganic nutrients. In more mature systems, which are physically quieter (less turbulence and upwelling), where most elements are tied to dissolved or particulate organics, mixotrophs become increasingly important. In terrestrial systems also, phagotrophy is associated with low inorganic-nutrient availability. The study of mixotrophy does indeed provide a rich ground for exploring natural selection. Conclusions Phagotrophy in the broad sense was vital for the spread of photosynthesis as organelles into eukaryotes, and is also important in the origin of intracellular symbioses involving photosynthetic micro-organisms. Phagomixotrophy remains widespread phylogenetically among algae, and is ecologically widespread, especially in oligotrophic water bodies. That this is so, despite the apparent clear metabolic benefit in heterotrophy for the individual versus the dominance of pure phototrophic eukaryotes in nutrient-rich ecosystems, points to the continued importance of natural selection in aquatic systems. We still await resolution of Darwin's ‘abominable mystery’ (Friedman, 2009), the origin and, especially, the rapid diversification of flowering plants. Comments on an earlier draft of the manuscript by Aaron Ellison and an anonymous reviewer have been very helpful. The University of Dundee is a Scottish Registered Charity, No. SC015096. References Adamec L.  Ecological requirements and recent European distribution of the aquatic carnivorous plant Aldrovandra vesiculosa L. : a review,  Folia Geobotanica et Phytotaxonomica ,  1995, vol.  30 (pg.  53- 61) Google Scholar CrossRef Search ADS   Adamec L.  Photosynthetic CO2 affinity of the aquatic carnivorous plant Utricularia australis (Lentibulariaceae) and its investment in carnivory,  Ecological Research ,  2009, vol.  24 (pg.  327- 333) Google Scholar CrossRef Search ADS   Adlassnig W,  Peroutka M,  Lambers H,  Lichtsheidl IK.  The roots of carnivorous plants,  Plant and Soil ,  2005, vol.  274 (pg.  127- 140) Google Scholar CrossRef Search ADS   Archibald JM.  The puzzle of plastid evolution,  Current Biology ,  2009, vol.  19 (pg.  R81- R88) Google Scholar CrossRef Search ADS PubMed  Baldauf SL.  An overview of the phylogeny and diversity of eukaryotes,  Journal of Systematics and Evolution ,  2008, vol.  46 (pg.  263- 273) Bell EM,  Laybourn-Parry J.  Mixotrophy in the Antarctic phytoflagellate, Pyramimonas gelidicola (Chlorophyta: Prasinophyceae),  Journal of Phycology ,  2003, vol.  39 (pg.  644- 649) Google Scholar CrossRef Search ADS   Bhattacharya D,  Archibald JM,  Weber APM,  Reyes-Prieto A.  How do endosymbionts become organelles? Understanding early events in plastid evolution,  BioEssays ,  2007, vol.  29 (pg.  1239- 1246) Google Scholar CrossRef Search ADS PubMed  Bhatti S,  Colman B.  Inorganic carbon acquisition in some synurophyte algae,  Physiologia Plantarum ,  2008, vol.  133 (pg.  33- 40) Google Scholar CrossRef Search ADS PubMed  Burkholder JM,  Gilbert PM,  Skelton HM.  Mixotrophy, a major mode of nutrition for harmful algal species in eutrophic waters,  Harmful Algae ,  2008, vol.  8 (pg.  77- 83) Google Scholar CrossRef Search ADS   Butterfield NJ.  Bangiomorpha pubescens n. gen., n. sp.: implications for the evolution of sex, multicellularity, and the Mesoproterozoic–Neoproteozoic radiation of eukaryotes,  Palaeobiology ,  2000, vol.  263 (pg.  386- 404) Google Scholar CrossRef Search ADS   Calderon-Saenz E,  Schnetterer R.  Morphology, biology and systematics of Cryptochroma perforata (Chlorarachniophyta), a phagotrophic marine alga,  Plant Systematics and Evolution ,  1989, vol.  163 (pg.  165- 176) Google Scholar CrossRef Search ADS   Casalduero FG,  Muniain C.  The role of kleptoplastids in the survival rates of Elysia timida (Risso, 1818): (Sacoglossa: Opisthobranchia) during periods of food storage,  Journal of Experimental Marine Biology and Ecology ,  2008, vol.  357 (pg.  181- 187) Google Scholar CrossRef Search ADS   Cavalier-Smith T.  Cell evolution and Earth history: stasis and revolution,  Philosophical Transactions of the Royal Society of London B ,  2006, vol.  361 (pg.  969- 1006) Google Scholar CrossRef Search ADS   Clegg MR,  Maberly SC,  Jones RI.  Behavioural response as a predictor of seasonal depth distribution and vertical niche separation in freshwater phytoplanktonic flagellates,  Limnology and Oceanography ,  2007, vol.  52 (pg.  441- 455) Google Scholar CrossRef Search ADS   Curtis NE,  Massey SE,  Pierce SK.  The symbiotic chloroplasts in the sacoglossan Elysia clarki are from several algal species,  Invertebrate Biology ,  2006, vol.  125 (pg.  336- 345) Google Scholar CrossRef Search ADS   Dagan T,  Martin W.  Seeing green and red in diatom genomes,  Science ,  2009, vol.  324 (pg.  1651- 1652) Google Scholar CrossRef Search ADS PubMed  Darwin C. ,  Geological observations of South America. The structure and distribution of coral reefs ,  1842 London Smith, Elder Darwin C. ,  Monograph of the living Cirripedia ,  1851 London Ray Society Darwin C. ,  Monograph of the fossil Lepadidae ,  1851 London Palaeontological Society Darwin C. ,  Monograph of the fossil Cirripedia ,  1854 London Ray Society Darwin C. ,  Monograph of the fossil Balanidae ,  1854 London Palaeontological Society Darwin C. ,  On the origin of species by means of natural selection: or, the preservation of favoured races in the struggle for life ,  1859 London John Murray Darwin C. ,  Insectivorous plants ,  1875 London John Murray Deusch O,  Landan G,  Roettger M,  Guenheit N,  Kowallik KV,  Allen JF,  Martin W,  Dagan T.  Genes of cyanobacterial origin in plant nuclear genomes point to a heterocyst-forming plastid ancestor,  Molecular Biology and Evolution ,  2008, vol.  25 (pg.  748- 761) Google Scholar CrossRef Search ADS PubMed  De Bary HA. ,  Die Erscheinung der Symbiose ,  1879 Stuttgart published privately Dolan JR,  Pérez MT.  Costs, benefits and characteristics of mixotrophy in marine oligotrichs,  Freshwater Biology ,  2000, vol.  45 (pg.  227- 238) Google Scholar CrossRef Search ADS   Douglas AE. ,  Symbiotic interactions ,  1994 Oxford Oxford University Press Douglas AE,  Raven JA.  Genomes at the interface between bacteria and organelles,  Philosophical Transactions of the Royal Society B ,  2003, vol.  358 (pg.  5- 17) Google Scholar CrossRef Search ADS   Dudley KL.  Algal communities in the leaf fluid of the pitcher plant, Sarracenia purpurea L,  Proceedings of the Pennsylvania Academy of Science ,  1984, vol.  58 (pg.  213- 215) Ellison AM,  Gotelli NJ.  Darwin review. Energetics and the evolution of carnivorous plants: Darwin's ‘most wonderful plants in the world’,  Journal of Experimental Botany ,  2009, vol.  60 (pg.  19- 42) Google Scholar CrossRef Search ADS PubMed  Evertson J,  Johnsen G.  In vivo and in vitro differences in chloroplast functionality in the two north Atlantic sacoglossans (Gastropoda, Opisthobranchia) Placida dendritica and Elysia viridis,  Marine Biology ,  2009, vol.  156 (pg.  847- 859) Google Scholar CrossRef Search ADS   Evertson J,  Burghardt I,  Johnsen G.  Retention of functional chloroplasts in some sacoglossans from the Indo-Pacific and Mediterranean,  Marine Biology ,  2007, vol.  151 (pg.  2159- 2166) Google Scholar CrossRef Search ADS   Falkowski PG,  Raven JA. ,  Aquatic photosynthesis ,  2007 2nd edn Princeton, NJ Princeton University Press Farmer MA,  Fitt WK,  Trench RH.  Morphology of the symbiosis between Corculum cardissa (Mollusca: Bivalvia) and Symbiodinium croculerum (Dinophyceae),  Biological Bulletin ,  2001, vol.  200 (pg.  336- 343) Google Scholar CrossRef Search ADS PubMed  Fathinejad S,  Steiner JM,  Siegfried R,  Martin M,  Allmaier G,  Burey SC,  Norikazu O,  Hideya F,  Loffelhardt W,  Bohnert H.  A carboxysomal carbon-concentrating mechanism in the cyanelles in the ‘coelocanth’ of the algal world, Cyanophora paradoxa?,  Physiologia Plantarum ,  2008, vol.  133 (pg.  27- 32) Google Scholar CrossRef Search ADS PubMed  Findlay GP,  Hope AB.  Lüttge U,  Pitman MG.  Electrical properties of plant cells: methods and findings,  Encyclopedia of plant physiology, New series, Vol. 2. Transport in plants II ,  1976 Berlin Springer-Verlag(pg.  53- 92) Fitt WK,  Fisher CR,  Trench RK.  Contribution of the symbiotic dinoflagellate Symbiodinium microadriaticum to the nutrition and survival of larval and juvenile tridacnid clams,  Aquaculture ,  1986, vol.  55 (pg.  5- 22) Google Scholar CrossRef Search ADS   Flőder S,  Hansen T,  Ptacnik R.  Energy-dependent bacterivory in Ochromonas minima: a strategy promoting the use of substitutable resources and survival at insufficient light supply,  Protist ,  2006, vol.  157 (pg.  291- 302) Google Scholar CrossRef Search ADS PubMed  Flynn KJ,  Mitra A.  Building the ‘perfect’ beast; modelling mixotrophic plankton,  Journal of Plankton Research ,  2009, vol.  9 (pg.  77- 93) Friedman WE.  The meaning of Darwin's ‘abominable mystery’,  American Journal of Botany ,  2009, vol.  96 (pg.  5- 21) Google Scholar CrossRef Search ADS PubMed  Garcia-Cuetos L,  Moestrup Ø,  Hansen PJ,  Daugbjerg N.  The toxic dinoflagellate Dinophysis acuminate harbors permanent chloroplasts of cryptomonad origin, not kleptoplastids,  Harmful Algae ,  2008  (in press) doi:10.1016/j.hal.2009.07.002 Gebühr C,  Pohlon E,  Schmidt AR,  Küsel K.  Development of microalgal communities in the phytoplankton of allochthonous populations of Sarracenia purpurea (Sarraceniaceae),  Plant Biology ,  2006, vol.  8 (pg.  849- 860) Google Scholar CrossRef Search ADS PubMed  Graham LE,  Wilcox LW. ,  Algae ,  2000 Upper Saddle River, NJ Prentice Hall Granéli E,  Carlson P.  Anderson DM,  Cymbella AD,  Hallegraeff GM.  The ecological significance of phagotrophy in photosynthetic flagellates,  Physiological ecology of harmful algal blooms. NATO ASI Series ,  1999, vol.  Vol. G41 (pg.  539- 557) Gustafson DEJr,  Stoecker DK,  Johnson MD,  Van Heukelem WF,  Sneider K.  Cryptophyte algae are robbed of their organelles by the marine ciliate Mesodinium rubrum,  Nature ,  2000, vol.  405 (pg.  1049- 1052) Google Scholar CrossRef Search ADS PubMed  Hansen PJ,  Fenchel T.  The bloom-forming ciliate Mesodinium rubrum harbors a single permanent endosymbiont,  Marine Biology Research ,  2006, vol.  2 (pg.  169- 177) Google Scholar CrossRef Search ADS   Herbert S. ,  Charles Darwin, geologist ,  2005 Ithaca Cornell University Press Jeong HJ,  Yoo YD,  Park JY,  Song JY,  Kim ST,  Lee SH,  Kim KY,  Yih WH.  Feeding by phototrophic red-tide dinoflagellates: five species newly revealed and six species previously known to be mixotrophic,  Aquatic Microbial Ecology ,  2005, vol.  40 (pg.  133- 150) Google Scholar CrossRef Search ADS   Johnson MD,  Oldach D,  Delwiche CF,  Stoecker DK.  Retention of transcriptionally active cryptophyte nuclei by the ciliate Myrionecta rubra,  Nature ,  2007, vol.  445 (pg.  426- 428) Google Scholar CrossRef Search ADS PubMed  Johnson MD,  Tengs T,  Oldach D,  Stoecker DK.  Sequestration, performance, and functional control of cryptophyte plastids in the ciliate Myrionecta rubra (Ciliophora),  Journal of Phycology ,  2006, vol.  42 (pg.  1235- 1246) Google Scholar CrossRef Search ADS   Jones H,  Cockell CS,  Goodson C,  Price N,  Simpson A,  Thomas B.  Experiments on mixotrophic protists and catastrophic darkness,  Astrobiology ,  2009, vol.  9 (pg.  563- 571) Google Scholar CrossRef Search ADS PubMed  Jones RI.  Mixotrophy in planktonic protists as a spectrum of nutritional strategies,  Marine Microbial Food Webs ,  1994, vol.  8 (pg.  87- 96) Jones RI.  Mixotrophy in planktonic protists: an overview,  Freshwater Biology ,  2000, vol.  45 (pg.  219- 226) Google Scholar CrossRef Search ADS   Kamjunke N,  Heinriches T,  Gaedke U.  Phosphorus gain by bacterivory promotes the mixotrophic flagellate Dinobryon spp. during re-oligotrophication,  Journal of Plankton Research ,  2007, vol.  29 (pg.  39- 46) Google Scholar CrossRef Search ADS   Kempf SC.  Symbiosis between the zooxanthella Symbiodinium (=Gymnodinium) microadriaticum and four species of nudibranch,  Biological Bulletin ,  1984, vol.  166 (pg.  116- 126) Google Scholar CrossRef Search ADS   Lake JA.  Evidence for an early prokaryotic endosymbiosis,  Nature ,  2009, vol.  460 (pg.  967- 971) Google Scholar CrossRef Search ADS PubMed  Larkum AWD,  Lockhart PJ,  Howe CJ.  Shopping for plastids,  Trends in Plant Science ,  2007, vol.  12 (pg.  189- 195) Google Scholar CrossRef Search ADS PubMed  Lewis CL,  Coffroth MA.  The acquisition of exogenous algal symbionts by an octacoral after bleaching,  Science ,  2004, vol.  304 (pg.  1490- 1492) Google Scholar CrossRef Search ADS PubMed  Liu H,  Probert I,  Uitz J,  Claustre H,  Aris-Brosou S,  Frada M,  Not F.  Extreme diversity in noncalcifying haptophytes explains a major pigment paradox in open oceans,  Proceedings of the National Academy of Sciences, USA ,  2009  (in press) doi: 10.1073/pnas0905841106 Maberly SC,  Madsen TV.  Freshwater angiosperm carbon concentrating mechanisms: processes and patterns,  Functional Plant Biology ,  2002, vol.  29 (pg.  393- 405) Google Scholar CrossRef Search ADS   Maberly SC,  Ball LA,  Raven JA,  Sültemeyer D.  Inorganic carbon acquisition by chrysophytes,  Journal of Phycology ,  2009  (in press) Martel Y. ,  Life of Pi ,  2002 Edinburgh Canongate Maranger R,  Bird D,  Price NM.  Iron acquisition by photosynthetic marine phytoplankton from ingested bacteria,  Nature ,  1998, vol.  396 (pg.  248- 251) Google Scholar CrossRef Search ADS   Margulis L. ,  Symbiotic planet: a new look at evolution ,  1998 New York Basic Books Martin W,  Kowallik KV.  Annotated English translation of Mereschowsky's 1905 paper ‘Über Natur und Urspring der Chromatophoren im Pflanzenreiche’,  European Journal of Phycology ,  1999, vol.  34 (pg.  287- 295) Martin W,  Rujan T,  Richly E,  Hansen A,  Cornelson S,  Lins T,  Leister D,  Stoebe B,  Hasegawa M,  Penny D.  Evolutionary analysis of Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genes in the nucleus,  Proceedings of the National Academy of Sciences, USA ,  2002, vol.  99 (pg.  12246- 12251) Google Scholar CrossRef Search ADS   Maynard-Smith J,  Szathmáry E. ,  The major transitions in evolution ,  1997 New York Oxford University Press McManus GB,  Zhang H,  Lin S.  Marine planktonic ciliates that prey on macraoalgae and enslave their chloroplasts,  Limnology and Oceanography ,  2004, vol.  49 (pg.  308- 313) Google Scholar CrossRef Search ADS   Mereschkowsky C.  Über Natur und Ursprung der Chromatophoren im Pflenzenreiche,  Biologisches Centralblatte ,  1905, vol.  25 (pg.  593- 604) Moore RB,  Obornik M,  Janoušovec J, et al.  A photosynthetic alveolate closely related to apicomplexan parasites,  Nature ,  2008, vol.  451 (pg.  959- 963) Google Scholar CrossRef Search ADS PubMed  Moustafa A,  Bhattacharya D.  Phylosort: a user-friendly sorting tool and its application to estimating the cyanobacterial contribution to the nuclear genome of Chlamydomonas,  BMC Evolutionary Biology ,  2008, vol.  8 pg.  6  Google Scholar CrossRef Search ADS PubMed  Moustafa A,  Reyes-Prieto A,  Bhattacharya D.  Chlamydiae has contributed at least 55 genes to Plantae with predominantly plastid functions,  PLoS ONE ,  2008, vol.  3 pg.  e2205  Google Scholar CrossRef Search ADS PubMed  Moustafa A,  Beszert B,  Maier UG,  Bowler C,  Valentin K,  Bhattachyra D.  Genomic footprints of a cryptic plastid endosymbiosis in diatoms,  Science ,  2009, vol.  324 (pg.  1724- 1726) Google Scholar CrossRef Search ADS PubMed  Nagai S,  Nishitani G,  Tomaru Y,  Sakiyama S,  Kamiyana T.  Predation by the toxic dinoflagellate Dinophysis fortii on the ciliate Myrionecta rubra and observations on the sequestration of ciliate chloroplasts,  Journal of Phycology ,  2008, vol.  44 (pg.  909- 922) Google Scholar CrossRef Search ADS PubMed  Nakayama T,  Ishida K-i.  Another primary acquisition of a primary photosynthetic organelle is underway in Paulinella chromatophora,  Current Biology ,  2009, vol.  19 (pg.  R284- R285) Google Scholar CrossRef Search ADS PubMed  Nodwell LM,  Price NM.  Direct use of inorganic colloidal iron by marine mixotrophic phytoplankton,  Limnology and Oceanography ,  2001, vol.  46 (pg.  765- 777) Google Scholar CrossRef Search ADS   Nowack ECM,  Melkonian M,  Glöckner G.  Chromatophore genome sequence of Paulinella sheds light on acquisition of photosynthesis by eukaryotes,  Current Biology ,  2008, vol.  18 (pg.  410- 418) Google Scholar CrossRef Search ADS PubMed  Öpik H,  Flynn KJ.  The digestive process of the dinoflagellate Oxyrrhis marina Dujardin, feeding on the chlorophyte, Dunaliella primolecta Butcher: a combined study of ultrastructure and free amino acids,  New Phytologist ,  1989, vol.  113 (pg.  143- 151) Google Scholar CrossRef Search ADS   Paffenhöfer G-A,  Sherr BF,  Sherr EB.  From small scales to the big picture: persistence mechanisms of planktonic grazers in the oligotrophic ocean,  Marine Biology ,  2007, vol.  28 (pg.  243- 253) Park MY,  Park JS,  Kim M,  Yih W.  Plastid dynamics during survival of Dinophysis caudata without its ciliate prey,  Journal of Phycology ,  2008, vol.  44 (pg.  1154- 1163) Google Scholar CrossRef Search ADS PubMed  Perez MT,  Dolan JR,  Fukai E.  Planktonic oligotrich ciliates in the NW Mediterranean: growth rates and consumption by copepods,  Marine Ecology Progress Series ,  1997, vol.  155 (pg.  89- 101) Google Scholar CrossRef Search ADS   Paungfoo-Lonhierre C,  Lonhierre TGA,  Rentsch D,  Robinson N,  Christie M,  Webb RI,  Gamage HK,  Carroll BJ,  Schenk PM,  Schmidt S.  Plants can use protein as a nitrogen source without assistance from other organisms,  Proceedings of the National Academy of Sciences, USA ,  2008, vol.  105 (pg.  4524- 4529) Google Scholar CrossRef Search ADS   Pierce SK,  Massey SE,  Hanten JJ,  Curtis NE.  Horizontal transfer of functional genes between multicellular organisms,  Biological Bulletin ,  2003, vol.  204 (pg.  237- 240) Google Scholar CrossRef Search ADS PubMed  Plachno BJ,  Wolowski K.  Algae commensal community in Genlisea traps,  Acta Societas Botanicorum Poloniae ,  2008, vol.  77 (pg.  77- 86) Raven JA. ,  Energetics and transport in aquatic plants ,  1984 New York, NY AR Liss Raven JA.  Energy and nutrient acquisition by autotrophic symbioses and their asymbiotic ancestors,  Symbiosis ,  1993, vol.  14 (pg.  33- 60) Raven JA.  Sandgren CD,  Smol JP,  Kristinasen J.  Comparative aspects of chrysophyte nutrition with emphasis on carbon, phosphorus and nitrogen,  Chrysophyte algae. Ecology, phylogeny and development ,  1995 Cambridge Cambridge University Press(pg.  95- 118) Raven JA.  Phagotrophy in phototrophs,  Limnology and Oceanography ,  1997, vol.  42 (pg.  198- 205) Google Scholar CrossRef Search ADS   Raven JA.  Carboxysomes and peptidoglycan walls of cyanelles: possible physiological functions,  European Journal of Phycology ,  2003, vol.  38 (pg.  47- 53) Google Scholar CrossRef Search ADS   Raven JA.  Contribution of anoxygenic and oxygenic phototrophy and chemolithotrophy to carbon and oxygen fluxes in aquatic environments,  Aquatic Microbial Ecology ,  2009  56, 177–192 Raven JA,  Ball LA,  Bardall J,  Giordano M,  Maberly SC.  Algae lacking carbon concentrating mechanisms,  Canadian Journal of Botany ,  2005, vol.  83 (pg.  879- 890) Google Scholar CrossRef Search ADS   Raven JA,  Giordano M.  Biomineralization by photosynthetic organisms: evidence of co-evolution of the organisms and their environment?,  Geobiology ,  2009, vol.  7 (pg.  140- 154) Google Scholar CrossRef Search ADS PubMed  Raven JA,  Johnston AM,  Handley LL,  McInroy SG.  Transport and assimilation of inorganic carbon by Lichina pygmaea under emersed and submersed conditions,  New Phytologist ,  1990, vol.  114 (pg.  407- 417) Google Scholar CrossRef Search ADS   Raven JA,  Walker DI,  Jensen KR,  Handley LL,  Scrimgeour CM,  McInroy SG.  What fraction of the organic carbon in sacoglossans is obtained from photosynthesis by kleptoplastids? An investigation using the natural abundance of stable carbon isotopes,  Marine Biology ,  2001, vol.  138 (pg.  537- 545) Google Scholar CrossRef Search ADS   Reyes-Prieto A,  Hackett JD,  Soares MB,  Bonaldo MF,  Bhattacharya D.  Cyanobacterial contribution to algal nuclear genomes is primarily limited to plastid functions,  Current Biology ,  2006, vol.  16 (pg.  2320- 2325) Google Scholar CrossRef Search ADS PubMed  Rumpho ME,  Summer EJ,  Manhart JR.  Solar-powered sea slugs. Mollusc/algal chloroplast symbiosis,  Plant Physiology ,  2000, vol.  123 (pg.  29- 38) Google Scholar CrossRef Search ADS PubMed  Rumpho ME,  Worful JM,  Lee J,  Kannan K,  Tyler MS,  Bhattacharya D,  Moustafa A,  Manhart JR.  Horizontal gene transfer of the algal nuclear gene psbO to the photosynthetic sea slug Elysia chlorotica,  Proceedings of the National Academy of Sciences, USA ,  2008, vol.  105 (pg.  17867- 17871) Google Scholar CrossRef Search ADS   Sanchez-Puerta M,  Delwiche CF.  A hypothesis for plastid evolution in chromalveolates,  Journal of Phycology ,  2008, vol.  44 (pg.  1097- 1107) Google Scholar CrossRef Search ADS PubMed  Schaffelke B.  Particulate organic matter as an alternative nutrient source for tropical Sargassum species (Fucales, Phaeophyceae),  Journal of Phycology ,  1999, vol.  35 (pg.  1150- 1157) Google Scholar CrossRef Search ADS   Schimper AFW.  Über die Entwicklung der Chlorophyllkőrner und Farbkőrper,  Botanisches Zeitunung ,  1883, vol.  41 (pg.  105- 114)  121–131, 137–146, 153–162 Sirova D,  Borovec J,  Cerna B,  Rejmankova E,  Adamec L,  Vrba J.  Microbial community development in the traps of aquatic Utricularia species,  Aquatic Botany ,  2009, vol.  90 (pg.  129- 136) Google Scholar CrossRef Search ADS   Smith DC,  Douglas AE. ,  The biology of symbiosis ,  1987 London Edward Arnold Stoecker DK.  Conceptual models of mixotrophy in planktonic protists and some ecological and evolutionary implications,  European Journal of Protistology ,  1998, vol.  34 (pg.  281- 290) Google Scholar CrossRef Search ADS   Taylor AR.  A fast Na+/Ca2+-based action potential in a marine diatom,  PloS ONE ,  2009, vol.  4 pg.  e4916  Google Scholar CrossRef Search ADS PubMed  Teugels B,  Bouillon S,  Veuger B,  Middelburg JJ,  Koedam N.  Kleptoplastids mediate nitrogen acquisition in the sea slug Elysia viridis,  Aquatic Biology ,  2008, vol.  4 (pg.  15- 21) Google Scholar CrossRef Search ADS   Tittel J,  Bissinger V,  Zippel B,  Gaedke U,  Bell EM,  Lorke AM,  Kanjunke N.  Mixotrophy: combined resource use to outcompete specialists,  Proceedings of the National Academy of Sciences, USA ,  2003, vol.  100 (pg.  12776- 12781) Google Scholar CrossRef Search ADS   Troost TA,  Kooi BW,  Kooijman SALM.  When do mixotrophs specialize? Adaptive dynamics theory applied to a dynamic energy budget model,  Mathematical Biosciences ,  2005, vol.  193 (pg.  159- 182) Google Scholar CrossRef Search ADS PubMed  Unrein F,  Massana R,  Alonso-Saéz L, et al.  Significant year-round effect of small mixotrophic flagellates on bacterioplankton in an oligotrophic coastal system,  Limnology and Oceanography ,  2007, vol.  52 (pg.  456- 469) Google Scholar CrossRef Search ADS   Usher KM,  Bergman B,  Raven JA.  Exploring cyanobacterial mutualisms,  Annual Review of Ecology, Evolution, and Systematics ,  2007, vol.  38 (pg.  255- 273) Google Scholar CrossRef Search ADS   Vieira S,  Calado R,  Coelho H,  Serôdio J.  Effects of light exposure on the retention of kleptoplastic photosynthetic activity in the sacoglossan mollusc Elysia viridis,  Marine Biology ,  2009, vol.  156 (pg.  1007- 1020) Google Scholar CrossRef Search ADS   Weisse T.  The significance of inter- and intraspecific variations in bacterivorous and herbivorous protists,  Antonie von Leeuwenhoek ,  2002, vol.  81 (pg.  327- 341) Google Scholar CrossRef Search ADS   Williams SI,  Walker DI.  Mesoherbivore–macroalgal interactions: feeding ecology of sacoglossan sea slugs (Mollusca, Opisthobranchia) and their effects on their food algae,  Oceanography and Marine Biology Annual Review ,  1999, vol.  37 (pg.  87- 138) Wouters J,  Raven JA,  Minnhagen S,  Graneli E,  Janson S.  The luggage hypothesis: comparison of two phototrophic hosts with nitrogen-fixing cyanobacteria and implications for analagous life strategies for kleptoplastids/secondary endosymbioses in dinoflagellates,  Symbiosis ,  2009  (in press) Zubkov MV.  Photoheterotrophy in marine prokaryotes,  Journal of Plankton Research ,  2009, vol.  31 (pg.  933- 938) Google Scholar CrossRef Search ADS   Zubkov MV,  Tarran GA.  High bacterivory by the smallest phytoplankton in the North Atlantic Ocean,  Nature ,  2008, vol.  455 (pg.  224- 226) Google Scholar CrossRef Search ADS PubMed  © The Author [2009]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
TI - Phagotrophy in the origins of photosynthesis in eukaryotes and as a complementary mode of nutrition in phototrophs: relation to Darwin's insectivorous plants
JO - Journal of Experimental Botany
DO - 10.1093/jxb/erp282
DA - 2009-09-18
UR - https://www.deepdyve.com/lp/oxford-university-press/phagotrophy-in-the-origins-of-photosynthesis-in-eukaryotes-and-as-a-xfHmJ0XFDR
SP - 3975
EP - 3987
VL - 60
IS - 14
DP - DeepDyve
ER -