6
$\begingroup$

Was it a 1 and done thing? Plants seem to have developed photosynthesis by the endosymbiosis of cyanobacteria. Is the latter the one time in Earth's history that the process independently came about?

Any answers would be greatly appreciated

$\endgroup$
4
$\begingroup$

The emergence of photosynthesis has occurred rather independently in multiple organisms. One way of looking at this is to look at the taxonomic distribution of photosynthesis reaction centers, which points to 5-10 origins according to current data (mostly in bacteria). That number has some big error bars as we are talking about very deep, old branches of the tree of life.

There are other criteria that we could use- for example the evolution of carbon fixation. However, that includes also chemoautotrophy, and is thus a bit more expansive.

Note that by some definitions the acquisition of cyanobacteria in plants may not even represent an independent emergence of photosynthesis, as the cyanobacteria already did photosynthesis!

If your question is more narrowly "how many times did photosynthesis evolve in eukaryotes" (not bacteria or archaea), then you would also need to include animals that do photosynthesis in a way similar to (though I believe less efficiently than) plants. They seem to similarly use endocytosis to acquire photosynthetic machinery from photosynthetic cells that they eat.

Update In response to comment from @shigeta, I'll add that the review that I link focuses on the reaction centers (RC) genes. They note that the RC genes probably had a single emergence, but that horizontal gene transfer is probably responsible for the current distribution of the organismal trait of photosynthesis:

Significant evidence indicates that the current distribution of photosynthesis in bacteria is the result of substantial amounts of horizontal gene transfer, which has shuffled the genetic information that codes for various parts of the photosynthetic apparatus, so that no one simple branching diagram can accurately represent the evolution of photosynthesis (Raymond et al., 2002).

$\endgroup$
  • $\begingroup$ Hey @maximilianPress a good citation but the conclusion at the bottom of the review is: "Current evidence suggests that the earliest photosynthetic organisms were anoxygenic, that all photosynthetic RCs have been derived from a single source, and that antenna systems and carbon fixation pathways have been invented multiple times." you can see on their phylogenetic tree that the RC1 and RC2 appear on all these branches, which implies they came from the same ancestor, lost in time. $\endgroup$ – shigeta Oct 28 '20 at 17:49
  • $\begingroup$ @shigeta a good point, which highlights the importance of whether you mean the emergence of photosynthesis (the organismal trait) or genes for the reaction centers. Indeed the RC genes are probably a single emergence, which were simply passed around by horizontal gene transfer. (Note that there is no ambiguity about eukaryotes independently evolving photosynthesis several times, through endosymbiosis!) The exact count of events comes down to a matter of how you like to parameterize your models. This exact issue is why I give a range of values. I'll edit answer. $\endgroup$ – Maximilian Press Oct 28 '20 at 17:59
  • $\begingroup$ ... well in this case the structures of RC1 and RC2 and their ubiquity in photosynthesis suggest a single source, lost in time. usually parallel evolution creates completely different protein structures for same function. a classic example are 'crystallins' - several genes have filled in for crystallins - the protein families are quite divergent. $\endgroup$ – shigeta Oct 29 '20 at 16:19
  • 1
    $\begingroup$ @shigeta the genes or proteins have a single origin. This is in the same way that e.g. Type III secretion system genes all have a single origin. However, this does not mean that Type III secretion-mediated pathogenicity has a single origin. Type III secretion systems get passed around on plasmids constantly, such that the parallel origins of this pathogenicity are certainly due to HGT. See e.g. the parallel evolution of Shigella from E. coli- Shigellas are polyphyletic in origin! Photosynthesis is like pathogenicity, only it evolves much less frequently. This is what the authors mean. $\endgroup$ – Maximilian Press Oct 29 '20 at 17:51
  • 1
    $\begingroup$ Agree - you can't expect any more than this when you are looking at a ~ billion year old gene! $\endgroup$ – shigeta Nov 5 '20 at 20:56
2
$\begingroup$

Im going to expand my comment into an answer because the discussion was just too interesting.

the conclusion at the bottom of the review cited is: "Current evidence suggests that the earliest photosynthetic organisms were anoxygenic, that all photosynthetic RCs have been derived from a single source, and that antenna systems and carbon fixation pathways have been invented multiple times." you can see on their phylogenetic tree that the RC1 and RC2 appear on all these branches, which implies they came from the same ancestor, lost in time.

@maximillianPress points out in his comment that lateral gene transfer is very common and this creates the exact pattern Many gene functions can arise more than once, but when you look at the gene, if a protein is achieves a function, like RC1 and RC2 their shape and detailed sequence will not resemble each other.

A classic example is the evolution of the eye. Eyespots and photosensors with lenses have evolved many times in microbial evolution and in may have evolved separately in molluscs vs vertebrates depending on how you define it. Light detection is a microbial innovation and the opsin light sensors in animals bear a strong resemblance to those in archaebacteria.

Enzymes form a clearer example. Studies of some of the carbohydrate metabolizing enzymes show that some specific roles are filled by multiple very divergent protein families. The role of a protein may change over time but its very unusual for the way the protein folds to change.

An third example is the crystallins whose function is to form the lens of the eye which has to remain stable for a lifetime. In different animals these genes are radically different - distinctly different proteins can fulfill this role. that is a gene role which may have evolved twice, but the role is fairly broad so its been filled by different proteins in different parts of the evolutionary tree more than once.

My thought is that the lateral transfer of microbial genes may confuse the evolutionary tree but the difficulty of creating a photoreaction center is much higher. There is only one bacterial reaction center protein family, so the predominance of evidence is that this phenomenon only evolved once and may have been transmitted by lateral gene transfer many times since.

$\endgroup$
  • 1
    $\begingroup$ I think that I agree with all this, happy that we've had a good discussion on this. With regard to the reconstruction of trees of life and the influence of HGT, a big distinction is the tree of genes vs. the tree of genomes vs. the tree of cells. Gene trees are noisy, genome trees have a central tendency but fuzzy, and the tree of cells has to exist but our only way of knowing it is the tree of genomes. For some of the (now older) work on this see science.sciencemag.org/content/311/5765/1283 and genomebiology.biomedcentral.com/articles/10.1186/… $\endgroup$ – Maximilian Press Nov 5 '20 at 19:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.