COMMENTARY Horizontal gene transfer in eukaryotic algal evolution Jason Raymond and Robert E. Blankenship* Department of Chemistry and Biochemistry, Arizona State University, Tempe, AZ 85287-1604 lthough seemingly innocuous, ther complexities in transcription and multiple times is also made clear in that the power of something as sim- translation, such as the need for correct it has occurred in various lineages after ple as a drop of water, given a splicing of RNA transcripts replete with the radiation of the three flavors of pri- few millions of years over introns. It can also be argued that sexual mary algae (red, green, and glaucocysto- Awhich to act continually, is immediately reproduction affords many eukaryotes phyte), leading to secondary algae with and awe-inspiringly obvious to someone the same advantage gained through quite varying plastid phenotypes. In pri- peering over the edge of the Grand HGT in bacteria. In this issue of PNAS, mary and secondary endosymbiosis Canyon. Acting over a similarly im- Archibald et al. (5) leap beyond the there has been a massive loss of genes mense time period but on vastly differ- case-by-case examples that typify eu- from the endosymbiont genome, many ent scales, this massive effect from a karyotic HGT and demonstrate that of which have been transferred into the weak, but constant, force embodies how HGT has played a significant role in the host genome, with each host-encoded natural selection has been able to direct evolution of a eukaryotic alga. In a col- plasmid-targeted gene now carrying a the evolution of once primitive, mal- lective analysis of 78 plastid-targeted transit peptide sequence that directs it adapted biological structures to the re- proteins from this alga, they show that, back to the plastid (secondary endosym- markable and almost inconceivably di- even by conservative measures, Ϸ21% biont-directed proteins also carry an verse molecular machines found within of these genes have likely been acquired additional signal sequence to get them extant organisms. This idea of natural by HGT. Their result stands to signifi- through the vestigial membrane system selection as a slow-and-steady workhorse cantly expand the number of established of the original plastid host) (14). Two was central to Charles Darwin’s evolu- cases of so-called transdomain HGT oc- groups of algae that evolved through tionary synthesis, as epitomized in his curring between prokaryotes and eu- secondary endosymbiosis, the crypto- oft-repeated pre´cis ‘‘Natura non facit karyotes and bolsters some novel ideas mondads and chlorarachniophytes (in- saltum,’’ Nature does not make leaps. on evolutionary mechanisms in phago- cluding B. natans) are particularly inter- Darwin would not live to see the discov- cytic eukaryotes (6). esting because they still contain a relict ery of genes as the vessel of inheritance The subject, Bigelowiella natans,isa nucleus called a nucleomorph, dramati- and random mutation as the propagator member of a class of algae known as cally reduced in size, from the originally of change, although these breakthroughs chlorarachniophytes that, in and of it- engulfed algae (10). would serve to reinforce his prescient self, is quite an evolutionary enigma. All Additional transfer of genes has un- ideas. plastid-containing eukaryotes acquired doubtedly occurred from the relict nu- To the contrary, the discovery of hor- the ability to do photosynthesis when, cleus into the host genome (15), al- izontal gene transfer (HGT) as a signifi- perhaps Ϸ2 billion years ago, a primitive though despite these complex transfer cant evolutionary driver may require an eukaryote engulfed a photosynthetic events these genes in the host genome addendum to the Darwinian synthesis. cyanobacterium. This so-called primary should have a phylogenetic signal consis- A growing body of evidence indicates endosymbiotic event gave early eu- tent with the engulfed algae and thus that many organisms, particularly pro- karyotes an extremely powerful meta- should be grouped more broadly with karyotes, can and do make evolutionary bolic ability that previously was manifest cyanobacteria. Genes present in the host leaps by sharing genes with one another, only among photosynthetic bacteria, and before endosymbiosis should cluster thereby opening a back door to an ad- also constituted, along with the enslave- with other eukaryotic genes and thereby aptation or ability that was already fine- ment of a proteobacterium that would can be used to classify the original host. tuned within another organism. Once become the mitochondrion, a massive For most nuclear-encoded algal genes thought to be an explanation of last re- horizontal transfer of genes into a prim- these stratifications are indeed observed, sort when the data were not robust itive eukaryote. The modern progeny of and in B. natans a variety of evidence enough to give unambiguous results, this primitive photosynthetic eukaryote clearly indicates a green algal endosym- with the recent availability of a wealth point to a single primary endosymbiotic biont origin. However, as Archibald of whole-genome data, HGT has not event, although some evidence argues et al. (5) show, many of the plastid- only become respectable but has otherwise (e.g., refs. 7 and 8). It is also targeted genes from B. natans clearly emerged as a central force in the evolu- certainly feasible that endosymbiosis diverge from this expectation. These tion of many different prokaryotes occurred multiple times, but many or- horizontally transferred genes span a (1–3). Of course, this idea came as ganisms were wiped out in the bottle- varied swath of functions, including no major surprise to many bacterial neck of subsequent global catastrophes chlorophyll biosynthesis, carbon fixation, geneticists, who for decades have been (e.g., global glaciations, refs. 9–11). and ribosome structure, and cluster with selecting prokaryotes for their ability to Not to be left behind, some eu- a similarly broad range of taxa other take up and express exogenous genes (as karyotes acquired photosynthesis than green algae. Several of their trees, Oswald Avery did some 60 years ago, through the same mechanism, although which, importantly, encompass much of demonstrating that DNA was the carrier not by engulfing a cyanobacterium but the available taxonomic sampling for of genetic information) (4). rather a eukaryotic alga (10, 12). each sequence, are particularly robust The impact of HGT on eukaryote ge- Termed secondary endosymbiosis, this based on bootstrap values and con- nomes has not been so clear-cut (3). process is believed to have given rise to served sequence motifs, providing strong The species concept of genetically segre- multiple independent groups of photo- gated germ lines has been tied to eu- synthetic organisms, all of which bear karyote taxonomy since its inception, the hallmark of plastids with three or See companion paper on page 7678. and the barriers against HGT in bacte- more bounding membranes (13). That *To whom correspondence should be addressed. E-mail: ria are magnified in eukaryotes by fur- secondary endosymbiosis has occurred [email protected]. www.pnas.org͞cgi͞doi͞10.1073͞pnas.1533212100 PNAS ͉ June 24, 2003 ͉ vol. 100 ͉ no. 13 ͉ 7419–7420 Downloaded by guest on September 24, 2021 mid analog exists in eukaryotes, Archibald et al. suggest that HGT in B. natans may occur in the same way it has for the many endosymbiotic events that have happened over the past 2 billion years, by engulfing other organisms (Fig. 1). Compared with the green alga Chlamydomonas reinhardtii, which is photoautotrophic and in which no paral- lel evidence of HGT is found, B. natans is mixotrophic, meaning that it can live phagocytotically and photosynthetically. Models have been proposed whereby small snippets of DNA from an en- gulfed microbe are able to escape diges- tion, e.g., from protist lysozomes, and migrate to and subsequently be incorpo- rated into the host genome (6, 16). One can imagine the series of fleetingly small probability events proceeding from en- gulfment to incorporation of a strand of foreign DNA into the genome to a new gene overcoming genetic drift to be- come fixed in the population. In a cer- Fig. 1. Stepwise conceptual image of one mechanism of HGT that, as proposed by Archibald et al. (5), tain sense, this is the same cumulative might operate in the chlorarachniophyte alga B. natans. Red arrows show phagocytosis and subsequent digestion of a bacterium or protist, from which foreign DNA has survived digestion and become effect as random mutations in single incorporated into the algal nucleus (flow of HGT-acquired genetic information indicated with blue genes or dripping water, but it now op- arrows). Although the genes studied herein by Archibald et al. are directed for function to the plastid, the erates on a different level, an entire significant number of horizontally transferred genes they found may only be the tip of the iceberg in gene. However, the essential point is phagocytic protists such as B. natans. that those probabilities are nonzero and over time have made a significant con- tribution to the genome of this organism support for HGT having played a signif- Most intriguingly, two of the genes from (6). The real boon of Archibald et al.’s icant role in the evolution of this their analysis indicate HGT from differ- hypothesis, perhaps best synopsized by organism. ent bacteria, significant not only as an Ford Doolittle’s epigram ‘‘you are what Because all of the genes studied by example of prokaryote-to-eukaryote you eat’’ (16), is that it is eminently test- Archibald et al. are encoded within the gene transfer but also because these ac- able as more eukaryotic genome data nucleus of B. natans but operate within quired genes initially would have not become available. It is already apparent the plastid, the signal and transit pep- had the proper leader sequence for that the magnitude of HGT varies dra- tides are absolutely necessary.