What does the origin of chlorophylls have to do with the origin of cell membranes?
If one reads the published works and discussions on the evolution of chlorophylls and cell membranes one might conclude that the two have nothing to do with each other. When chlorophylls appeared for the first time, cell membranes had already evolved… because it is assumed that chlorophylls originated after the most recent common ancestor of bacteria. In other words, according to current views, the earliest bacteria and the earliest forms of life, were not photosynthetic.
Grab a cup of coffee or tea, sit down, get comfortable and let’s challenge this view a little.
OK, let me provide some context first.
The possibility that bacteria were ancestrally photosynthetic was emphasized by Woese et al. (1985). I quote:
Oparin long ago argued that the primitive oceans were a "soup" of energy-rich biochemicals, and consequently, the first organisms arising therein were nonphotosynthetic, fermentative heterotrophs—there being no need initially for biological systems to generate their own energy-rich compounds.
Photosynthetic species arose only later, as the oceanic supply of energy-rich biochemicals became exhausted. Oparin's view was taken to mean that the first bacteria were heterotrophs and that photosynthetic bacteria arose only later, as one particular subline in the otherwise nonphotosynthetic bacterial world. Although we still have no idea whether such an idea is correct, it tends to be presented to each new generation of microbiologists as unassailable truth.
I’m sorry Carl, but no one heard you.
Basically, Carl and colleagues realised that the different groups of bacteria were too distant from each other, in such a way that, assuming vertical descent, the earliest photosynthetic ancestor would be placed at the root.
I arrived to the same conclusion 30 years later when I re-assessed the evolution of the photosynthetic reaction centre proteins (Cardona 2015).
However, those ideas were later challenged, and really never considered again properly, when the notion was introduced that horizontal gene transfer could hypothetically explain the scattered distribution of photosynthesis.
However, I have argued that the molecular evolution of photosynthesis does support Carl and colleagues’ suspicions, based on the observation that the early evolutionary steps in the emergence of reaction centres, the photosystems, must antedate the diversification of the major clades of bacteria. Combine that with the fact that not only there is a massive phylogenetic distance between the lineages that have photosystems, but also that great distance is matched by that of their photosynthetic machinery… add to that the evidence for photosynthesis more than 3.4 billion years ago: and then it starts to look very likely that the first photosynthetic reaction centres appeared deep within the tree of bacteria, if not before.
More recently, I have started to compare the distances and rates of evolution of the different proteins of the photosystems to those of other very ancient enzymes. They show patterns of evolution that are identical to other enzymes that are thought to have arisen before the last universal common ancestor (LUCA)… this is unpublished stuff that I am actually writing right now. But see this previous blog post.
It opens the possibility that the origin of photosynthesis—and I am not thinking of a complex biochemical process, but of simple enzymes containing photoactive cofactors—may trace back to before the LUCA. I’m thinking of the possibility that photochemistry was an ingredient in the origin of life… not just any photochemistry though, a photochemistry that led to the oxidation of water (Cardona and Rutherford 2019).
I’m going to quote Carl again: Therefore, although our conclusion is not a compelling one, it does demand that the archaic and now suspect notion that all bacteria have arisen from a common nonphotosynthetic ancestor no longer be accepted and perpetuated as dogma.
Thank you. My mission now is to make this conclusion, a compelling one.
Ok, what about the origin of chlorophylls and membranes?
I came across an interesting paper about the evolution of membrane lipids:
Investigating the Origins of Membrane Phospholipid Biosynthesis Genes Using Outgroup-Free Rooting by Gareth A. Coleman, Richard D. Pancost, Tom A. Williams (Coleman et al. 2019)—Thank you for publishing open access, I will be using some of your figures in this blog post!
One of the main distinguishing features of bacteria and archaea are their membranes: “the lipid divide”.
Quoting: Canonically, Archaea have isoprenoid chains attached to a glycerol-1-phosphate (G1P) backbone via ether bonds and can have either membrane spanning or bilayer-forming phospholipids (Lombard et al. 2012a). Most Bacteria, as well as eukaryotes, classically have acyl (fatty-acid) chains attached to a glycerol-3-phosphate (G3P) backbone via ester bonds and form bilayers (Lombard et al. 2012a), although a number of exceptions have been documented (Sinninghe Damsté et al. 2002, 2007; Weijers et al. 2006; Goldfine 2010).
See the figure below, Fig. 1 from Coleman et al. (2019).
Membrane lipids in bacteria and archaea, see the reference for all the details.
In archaea, the second isoprenoid chain is bound by digeranylgeranylglyceryl phosphate synthase (DGGGPS). DGGGPS is widely distributed in both archaea and bacteria. In addition, DGGGPS belongs to a larger family of enzymes, UbiA-like, which includes among others, chlorophyll synthase: ChlG, as it is known in cyanobacteria; or BchG, as it is known in the anoxygenic phototrophs. I’ll refer to both as ChlG.
Coleman et al. (2019) concluded based on the phylogeny of DGGGPS alone, the following:
The wide distribution of this enzyme across both Archaea and Bacteria, and the occurrence of both domains on either side of the root, for both rooting methods, suggest either multiple transfers into Bacteria from Archaea, or that DGGGPS was present in LUCA and inherited in various archaeal and bacterial lineages, followed by many later losses in and transfers between various lineages.
In both cases, DGGGPS is tremendously ancient: either as old as archaea or as old as life itself.
Have in mind that one of the key points of the paper is that the authors implemented an “outgroup-free rooting” approach.
The authors did provide a phylogenetic tree showing the entire UbiA family, which they used to help distinguish DGGGPS from other sequences, Supplementary Figure S25.
Supplementary Figure S25. UbiA full tree, 227 sequences, 69 positions, inferred under LG+C60 model. DGGGPS sequences in blue, chlorophyll a synthase in green, protoheme IX farnesyltransferase in red, and 4-hydroxybenzoate octaprenyltransferase in black.
Have a second good look at Figure S25 again.
ChlG connects the chlorophyll ring to its tail. This occurs by a reaction very similar to that of DGGGPS (blue square below).
Comparison of enzymatic reactions.
Protoheme IX farnesyltransferase connects the heme b ring to a tail turning it into heme o/a. This enzyme was named CyoE in E. coli. I’m not very familiar with the proper nomenclature of these enzymes across life, so I will just call it CyoE.
4-hydroxybenzoate octaprenyltransferase is well characterized in the synthesis of ubiquinone in proteobacteria and eukaryotes, and it is probably involved in the synthesis of other quinones in other prokaryotes.
What we can see in Figure S25 is that ChlG made a monophyletic clade and clustered together with DGGGPS.
One might say: “maybe large branch attraction? I mean, the trees are really not well resolved.” But then we can also see that ChlG and DGGGPS actually catalyse a very similar reaction, distinc to those catalysed by UbiA and CyoE, so it makes sense that they would cluster together… do these enzymes share a more recent common ancestor?
Remember what was concluded about the root and evolution of DGGGPS. The outgroup-free-calculated root was placed were the green arrow is located. According to the authors the ancestral DGGGPS node could be pre-LUCA or deep within archaea.
Now I ask, what is it going to be rooted to? I would answer that it would be rooted to its nearest sister clade: ChlG.
Remember that based on the evolution of reaction centres I concluded, independently, that the origin of photosynthesis occurred, at the very least, deep within bacteria: so, are we seen a bacteria/archaea split?
And where we have seen something like this before? Yes, in the evolution of protochlorophyllide and chlorophyllide reductases and their relationships with their archaeal counterparts involved in methanogenesis. I have argue that the early evolution of nitrogenase-like enzymes may have been driven by a bacterial/photosynthesis archaeal/methanogenesis split. See this: On the evolution of chlorophyll synthesis, methanogenesis, and nitrogen fixation. Was the ancestor of Bacteria photosynthetic?
So we have several possible rooting options, I suppose, see the image below. Panel A and B are my favourites, with a slight preference for A based on biochemistry.
I have annotated Fig. S25 with purple B and A to illustrate how the enzymes are distributed in bacteria and archaea. What I note is possible bacteria/archaea splits also within UbiA and within CyoE, with the region containing strains of archaea always including some strains of bacteria. So, what we are actually seeing is perhaps what you see in panel C. That means that we could have a triplication of the ancestral enzyme before the LUCA, leading to one lineage ancestral to ChlG and DGGGPS, a second one ancestral to CyoE, and the other one to UbiA.
There is something that I have not mentioned yet. It has to do with CyoE… well heme o/a, as far as I know, are mainly found as the cofactors of the quinol oxidase, dioxygen reductase, heme-copper respiratory complexes. And it just happens that the main core protein of this complex has been suggested in several independent studies, to trace to the LUCA. Not only this has been concluded from the phylogeny of the proteins themselves (Brochier-Armanet et al. 2009), but also they have turned up in reconstructions of the LUCA proteome (Ouzounis et al. 2006), even against the author’s will sometime (Weiss et al. 2016).
Although the evolution of the dioxygen reductases is tremendously controversial, I think the phylgenetic trees that I have seen on these clearly show that they are extremely ancient, even within the scenario in which they emerged from NO reductases. Their early origin would be consistent with what I have concluded on the evolution of photosynthetic reaction centres: that reacton centres likely originated in an oxygenic context, to split water and before the diversification of the major groups of bacteria, including cyanobacteria themselves. Therefore, aerobic respiration also emerges early before the known groups of bacteria (and archaea?) had time to diversify into the groups that we know today.
And… UbiA is used to make quinones, right? Which are used in photosynthesis and respiration.
Are we not seeing in Fig. S25 the evolution of the world’s core bioenergetics?
The synthesis of membranes where photosynthetic (chlorophyll/quinone) and respiratory (heme/quinone) complexes are located. Their diversification linked to the photsynthetic oxidation of water to oxygen and the utilisation of that produced oxygen… since the very beginning.
I learnt at the 2019 Geobiology conference in Banff that before the evolution of oxygenic photosynthesis, primary productivity would be less than 1% of today’s value. Imagine that! Ward et al. (2019) also calculated that in the absence of oxygenic photosynthesis primary productivity would be something like 0.1% of today’s value.
This is the thing: it seems that life has always been pretty productive, as even the extremely rare and oldest rocks (3.8-3.9 billion years old) show traces and signatures of life. What are the odds of that if primary productivity then was only 0.1 to 1% of the current level? My point is… what if there was never a “before the evolution of oxygenic photosynthesis”? What if there is not a discrete point in time when we can say, “on this day, the 25th of March of the year 3.0 billion B.C., oxygenic photosynthesis originated.” What if all blurs into a primordial cocktail of photochemical reactions, some of which resulted in the oxidation of water, that trace back to the same chemistry that led to life?
Here I have mentioned two family of the enzymes required for the synthesis of chlorophyll with roots that can reach down to the earliest events in the evolution of life: BchG and the protochlorophyllide/chlorophyllide reductases. These are some of the last enzymes in the pathway. I will write a bit about the evolution of Mg-chelatase in a different post at a later stage, the “first” enzyme in the pathway, because its evolution is consistent with the oxygenic origin of photosynthesis I propose. I believe within its evolution there is also evidence of the early losses of photosynthesis across the tree of life of bacteria… stay tuned.
References
Brochier-Armanet, C., E. Talla and S. Gribaldo (2009). "The multiple evolutionary histories of dioxygen reductases: Implications for the origin and evolution of aerobic respiration." Mol Biol Evol 26(2): 285-297. DOI: 10.1093/molbev/msn246.
Cardona, T. (2015). "A fresh look at the evolution and diversification of photochemical reaction centers." Photosynth Res 126(1): 111-134. DOI: 10.1007/s11120-014-0065-x.
Cardona, T. and A. W. Rutherford (2019). "Evolution of Photochemical Reaction Centres: More Twists?" Trends in plant science. DOI: 10.1016/j.tplants.2019.06.016.
Coleman, G. A., R. D. Pancost and T. A. Williams (2019). "Investigating the Origins of Membrane Phospholipid Biosynthesis Genes Using Outgroup-Free Rooting." Genome Biol Evol 11(3): 883-+. DOI: 10.1093/gbe/evz034.
Ouzounis, C. A., V. Kunin, N. Darzentas and L. Goldovsky (2006). "A minimal estimate for the gene content of the last universal common ancestor: Exobiology from a terrestrial perspective." Res Microbiol 157(1): 57-68. DOI: 10.1016/j.resmic.2005.06.015.
Ward, L. M., B. Rasmussen and W. W. Fischer (2019). "Primary Productivity Was Limited by Electron Donors Prior to the Advent of Oxygenic Photosynthesis." J Geophys Res-Biogeo 124(2): 211-226. DOI: 10.1029/2018jg004679.
Weiss, M. C., F. L. Sousa, N. Mrnjavac, S. Neukirchen, M. Roettger, S. Nelson-Sathi and W. F. Martin (2016). "The physiology and habitat of the last universal common ancestor." Nat Microbiol 1(9): 16116. DOI: 10.1038/nmicrobiol.2016.116.
Woese, C. R., B. A. Debrunnervossbrinck, H. Oyaizu, E. Stackebrandt and W. Ludwig (1985). "Gram-Positive Bacteria - Possible Photosynthetic Ancestry." Science 229(4715): 762-765.