Introduction

If you were to take a human body and bust it up into elements, you’d find that it is predominantly made of carbon, hydrogen, oxygen, nitrogen, calcium and phosphorus. These elements account for more than 99% of the mass of a human body. With the exception of calcium, which makes up much of our bone mass, the same could be said of all living organisms (even those without bones). What’s more, these elements are more common in living bodies than outside of them. Carbon, in particular, is dramatically¹ more concentrated in the bodies of organisms than it is in our atmosphere or in our oceans.

The universe tends towards equilibrium, towards uniformity of concentrations. But life is emphatically not at equilibrium; equilibrium is death. Energy is required to break equilibrium, to take elements like carbon from the environment and concentrate them in our bodies. Ultimately, all of the energy available to the biosphere (i.e. all extant life) comes from the sun. Even the energy we put into the biosphere artificially in the form of fertilizers, tractors and refrigerated trucks derives from fossil fuels like coal and gasoline, which themselves ultimately derive from plants and other photosynthetic bodies. If our present-day existence has an ultimate cause it is photosynthesis.

Beyond simply being more concentrated in our bodies, organic carbon (i.e. carbon compounds found in living systems) is more “reduced” than it is in the atmosphere, meaning that the carbon atoms literally have more electrons. Oxygenic photosynthesis uses light energy to withdraw electrons from water, forming molecular oxygen (O₂). These electrons are then “donated” to carbon dioxide in order to make covalent bonds between carbon atoms, bonds that build complex molecules like cellulose, starches, proteins and nucleic acids (DNA and RNA), which in turn form the bodies of plants. Animals eat the plants and are then eaten by other animals, fungi and bacteria. Everyone dies, decomposes and oxidizes (loses electrons) only to be reduced again by photosynthetic organisms at some later date. It’s the circle of life, son.

The great genius of oxygenic photosynthesis is that it couples the reduction of carbon to an abundant source of energy (sunlight) and an abundant source of electrons (water). There was photosynthesis prior to the invention of oxygenic photosynthesis by the Cyanobacteria, ² but it did not draw electrons from water and it did not make oxygen. It relied on electron-donating elements that are much less abundant than water, elements like iron, sulfur and manganese. The evolution of oxygenic photosynthesis opened up enormous possibilities for life – suddenly the whole surface of the Earth’s oceans becomes habitable. Anywhere there is water and sunlight, photosynthesis can proliferate.

Oxygen, being both highly reactive and highly energetic,³ is perhaps the most significant molecule on the planet. And carbon is its fraternal twin – we can make our complex, carbon-based bodies only because our photosynthetic friends made us reduced carbon to eat and oxygen to breath: carbon and oxygen in one act. We humans, mammals, animals, fungi and many bacteria make a living by inverting the chemistry that oxygenic photosynthesis performs. Photosynthetic organisms build up complex carbon-based molecules and produce O₂ as a byproduct. We tear them down piece-by-piece, donating their electrons to hungry oxygen one at a time and harnessing the energy of that piecemeal combustion to go on living, producing our own local order in all it’s Golgi apparatuses, intestinal crypts, corneae and venae cavae.

I spend a lot of my time researching the mechanics of photosynthesis in present-day organisms. In studying contemporary photosynthesis, it’s often useful to understand where plants and their relatives came from, how they differ from each other and how photosynthesis evolved (at least as far as we understand). Once you delve into the topic, it is impossible to ignore how intimately the history of life and indeed the history of the Earth is intertwined with the evolution of oxygenic photosynthesis. The presence of oxygen is inscribed on the surface of our planet, deep in our sedimentary rocks, in the radiation of biological forms present on Earth today, and also the extinction of forms incapable of handling the scavenger of electrons that is O₂. Woodward Fischer (he goes by Woody) is a professor of Geobiology in Geological and Planetary Sciences at CalTech. I contacted him over e-mail and asked if we could have a conversation about his main research focus: the evolution of oxygenic photosynthesis. Woody graciously agreed

 

---

 

Interview

Thanks to Akiva Steinmetz-Silber for transcription.

AVI FLAMHOLZ: I took an introductory planetary science class a few years ago. The professor started the course with a provocative statement: that the geosphere and even the biosphere don’t need us — where by “us” I think he meant “humans.” I'm wondering what you think about that as somebody who studies the history of photosynthesis. Do you think the geosphere needs life? Not just humans, but what would happen if life stopped? How would that affect the Earth?

WOODY FISCHER: Yeah, so that's a great question and it's a big question that people are now starting to wrestle with in a very different way. If you were to go back 30 years ago, the idea was that the sets of processes that occur in the solid Earth, they set the environmental conditions at the surface of the Earth, and that's the arena that biology gets to play in. The idea was that the flow of information went in that direction—the solid Earth sets the arena and when catastrophic things occur to solid Earth, like volcanism, or meteor impacts from extraterrestrial sources—the biosphere noticed, and in general, in a bad way. But we've started to rethink how that mapping works in terms of the flow of information from the geosphere to the biosphere, and maybe more broadly, between the geosphere and the fluid Earth and the biosphere. That is made most clear by the evolution of photosynthesis and the rise of oxygen, because it is demonstrably the most important event in terms of environmental change on the surface of our planet, and it's driven entirely by a biological innovation. It's very clear that our planet did not have any oxygen until oxygenic photosynthesis, and thereafter every corner of our planet is touched by oxygenic photosynthesis or the oxygen produced by that process. So, we're starting to have a much more subtle discussion of the interaction between the solid Earth, the fluid Earth and the biosphere that inhabits the surface of the Earth.

AVI FLAMHOLZ: What do we know about the nature of the early Earth and what's so distinctive about that as compared to how the atmosphere and the geosphere are structured today?

WOODY FISCHER: More or less all of our knowledge about the composition of the oceans and the atmosphere, or even that there were oceans, or the size and chemistry of the atmosphere, it all comes from the observations of very old sedimentary rocks. It's actually remarkable that we can take observations of ancient sedimentary rocks that are billions of years old and invert aspects of their chemistry. The thing that emerges most strikingly from that work is that the early Earth is, in many ways, similar to today’s Earth — it's characterized by ocean basins that are similar today, and we know that many of those basins had life present. Evidence of life is found in microfossils, particular bits of carbon, or more subtle interpretations of the major chemical cycles—the carbon cycle, the iron cycle and the sulfur cycle. But one of the things that we do know that emerges very quickly from looking at that record of early Earth is that there's no oxygen present. There's absolutely zero oxygen present in the waters, and in the atmosphere, and oxygen is not playing a role, for example, when the crust is weathering. All of the redox sensitive elements in minerals appear mostly in their reduced⁴ form. When those materials got weathered, for example into soil, and then ultimately into grains of sand that are transported by a river, they didn’t rust. So they did not interact with oxygen. And we can deduce, therefore, that oxygen wasn't present in those environments, so all of our ideas about what the early Earth looked like fundamentally come from observations of sedimentary rocks.

AVI FLAMHOLZ: How do we know how old these sedimentary rocks are? How do we that one is older than the other, or that there's some absolute time that has elapsed since they were formed?

WOODY FISCHER: For hundreds of years, geologists have used a couple of really simple geometric rules to ordinate rocks that are in contact with each other. That is to say, sedimentary rocks are laid down flat and, through their crosscutting relationships, you can develop a kind of relative chronology. But it wasn't until the advent of radiometric dating — that is, looking for, for example, uranium-bearing minerals that decay to lead – that you could actually, by measuring the uranium content and the lead content, the parent and the daughter products of that decay, develop an absolute chronology. ⁵ That’s critically important for some of these really old rocks in particular, because they don't have other things that tell us about their age, like fossils. So we rely critically on the uranium-lead system to obtain an absolute age.

AVI FLAMHOLZ: In an intro biology class, you’d normally learn this stuff in the reverse order, right? You learn that there's oxygen in the atmosphere and there's carbon in the biosphere — we eat the carbon, and we use the oxygen to combust it to generate energy. But actually oxygenic photosynthesis generates atmospheric oxygen and also reduced carbon – it’s responsible for both of them. What do we think are the outcomes of having both carbon and oxygen become abundant after the advent of photosynthesis?

WOODY FISCHER: There were a couple of revolutions in a number of different ways. The one that is probably the most profound is this: once photosynthesis figures out how to oxidize water, it's a really big deal because there's plenty of sunlight on our planet, there's plenty of water on our planet. It opens up a huge ecological area for organisms that can do this metabolism. It's also a very energy-rich metabolism, so they're actually able to harvest a substantial amount of energy just by producing organic matter and oxygen. The flipside of that is that oxygen provides a tremendous challenge. It's incredibly energy-rich, so if we can figure out how to use it to eat organic carbon, and organisms — we think it may have taken a little bit, you know, millions of years, but organisms figured out how to do that and conserve a substantial amount of energy, and that's a really big breakthrough. So, photosynthesis generates the possibility of breathing oxygen for the first time. Once you can breathe oxygen, because you can generate so much energy, it leads to a bunch of really interesting evolutionary phenomena, for example, complex multicellularity arises. Multi-cellularity occurs in organisms that harvest a substantial amount of energy using oxygen, and basically have energy to burn, to explore new ways of being. If you've read some Jared Diamond, I always say it's akin to agriculture. There's this argument that agriculture is this really big boost to human society, because, all of a sudden, with surplus food, you can start to specialize. So you end up with craftsmen, and you end up with scribes, and scholars, whereas when everyone's a hunter-gatherer, everyone's got to be invested in getting food all the time. Aerobic respiration⁶ is like agriculture, because it allows cells to harvest so much energy that they can start to explore new types of biology.

AVI FLAMHOLZ: It's also maybe like agriculture in a more subtle way. Jared Diamond says that feudalism allows some people to specialize because others are focusing on agriculture. There's a division of labor. But agricultural life is actually not that pleasant — life might even be harder for a farmer than a hunter-gatherer. In the same way, your work describes the double-edged sword of oxygen as being both a source of energy, and also an enormous source of toxicity for cells. We colloquially think of oxygen as being the source of life, so I’m hoping you can speak to why it is that oxygen is actually fundamentally very harmful.

WOODY FISCHER: What it basically boils down to is that oxygen is energy-rich and also super reactive. Oxygen is fundamentally incompatible with the most basic ways that cells are constructed. One of the things that's really amazing is the lengths to which aerobes⁷ must go to just to live the lifestyle that they do. It is obviously worth it, but almost certainly, when those first cells figured out how to make oxygen, it was an absolute catastrophe, because the cells were completely ill-equipped to deal with this toxin.

AVI FLAMHOLZ: For the sake of the drama, could you describe what happens when oxygen is produced inside a cell? What are some of the atrocious things it can do, and what are the properties of oxygen that make that possible?

WOODY FISCHER: There are three things I think are really nasty. It's not exactly oxygen that's the problem. It's when oxygen either gives up an electron to form reactive oxygen species, or in some ways becomes a little more energetic, for example by absorbing some light energy. Before it even gets into the cell, this creates a giant problem by the oxidation of lipids. It's going to interact with the membrane and oxidize the membrane, and that fundamentally changes the fluid properties of the membrane, which requires constant repair. Once O₂ starts to get into the compartments of the cell, it's going to interact with any proteins that have, for example, iron-sulfur clusters, which are the major kind of workhorses of redox chemistry present in all cells. O₂ irreversibly oxidizes those clusters, making them useless. This is most famously known for the protein nitrogenase, a really important protein that takes N₂ gas from the atmosphere and turns it into useful nitrogen. ⁸ Nitrogenase, for example, is completely incompatible with oxygen. The third thing is that oxygen is great at breaking strands of DNA and RNA, which degrades genomic information and can be an incredibly toxic thing. There's a fourth that is critically important for photosynthetic organisms. Because they make oxygen, photosynthetic organisms are going to have all of their chlorophyll-studded photosynthetic machinery in the presence of oxygen. When light energy is available, this proximity rapidly produces a remarkable oxidant called singlet oxygen. And that is inescapable. So, put another way, all of the photosynthetic machinery that is ultimately a part of how oxygen is made is completely incompatible with the presence of oxygen, and there's no getting around that fundamental chemistry, except to keep oxygen concentrations low near the machinery that's used for photosynthesis.

AVI FLAMHOLZ: It's kind of incredible that this stuff works.

WOODY FISCHER: Exactly, it shouldn't work. It absolutely shouldn't work. You know, you would think it's just full of cross-purposes everywhere, and yet it's absolutely worth it.

AVI FLAMHOLZ: So the capacity to do oxygenic photosynthesis, that is, photosynthesis that takes light energy and uses it to withdraw electrons from water and create oxygen, allows for species to have both a pretty strong competitive advantage relative to their contemporaries. It also allows for the generation of a much more complex biosphere through the production of both atmospheric oxygen that organisms can respire, and also reduced carbon that can be used by other organisms to build their bodies, to create biomass, and conserve energy through respiration. I know that you interrogate the history of this whole system — the timing of it’s evolution — through the geological record, and also through comparative genomics. If I look at two sequences and they're more similar to each other than they each are to a third sequence, then it makes sense that they're probably more closely related to each other, and therefore more recently related to each other than to the third guy. But the part that I've never fully understood is this: there seem to be methods for figuring out the absolute times from the differences between protein sequences. How does that work?

WOODY FISCHER: Yeah, so there are seemingly disparate records, a sedimentary record and a comparative genomics record. But of course, the thing that unites them is the same history. The Earth only has one history, so they're very different different windows into that history. The idea of my nascent field “geobiology” is that we can leverage some of their strengths and avoid some of their weaknesses. Ultimately, it's with some understanding and appreciation of extant biology that we're going to interpret sedimentary rocks. ⁹ Just like you mentioned, when we compare genes and proteins, or any type of phylogeny, we can get some idea of relative time, but ultimately that phylogeny is going to have to get calibrated by the sedimentary record, by the fossil record. ¹⁰ Figuring out how to do that well has been a really big challenge. We're interested in the evolution of photosynthesis. Photosynthesis is something that originates within the cyanobacteria. There is a fossil record of cyanobacteria, but it's pretty bad. You might imagine that microbial fossils are in general hard to interpret – there’s not a lot of morphology — for the most part, microbes express their diversity through metabolism, and not through morphology, which is the opposite of animals and plants.

AVI FLAMHOLZ: Also, they're quite small.

WOODY FISCHER: Yeah, they're really small. Consequently, you'll hear people argue about some of the earliest fossils. What people are arguing about is not which species these might be, but whether they're microfossils at all. At the scale of kind of micron-sized bubbles in rock, there's a bunch of features that you might mistake as former cells. One thing that we’ve tried to do is to build as much as possible on aspects of the fossil record that we feel really good about. For cyanobacteria that comes down to interrogating the fossil record of algae and plants. As cyanobacteria become chloroplasts, we can use the algal fossil record, which is reasonable, and the plant fossil record, which is excellent, to estimate evolutionary rates¹¹ for certain lineages of cyanobacteria. We can then ask: are these genes relatively recent or are they very old?

An age-old debate, actually, is where do the Cyanobacteria come from? Who are their closest living relatives? We didn't even have an answer until about three or four years ago. Now, we're starting to get a picture of who some of their close living relatives are, and it's totally changing our understanding of the timing of oxygenic photosynthesis.¹² Now we recognize that the cyanobacteria that do oxygenic photosynthesis are just a tiny little branch of what you would now call the cyanobacterial phylum, which includes a number of diverse members, none of whom do photosynthesis. So photosynthesis in cyanobacteria isn't even as old as the cyanobacterial phylum, which isn't even as old as the radiation of the major bacterial phyla from one another, etc. That is all relative time,¹³ but that’s already telling us that we shouldn't expect oxygenic photosynthesis to be nearly as old as our planet, for example.

AVI FLAMHOLZ: The thing that confuses me here is that I've never understood what assumptions you have to make to get an absolute time out of these relatives comparisons of protein sequences. If I can just paraphrase what you said, that you have fossils whose dates you know based on these radiometric tracing. Those fossils, especially for plants, are really clear, right—you see a leaf in the fossil, and it's obviously a leaf, and you know how old it is, and you can pretty much tell what species it is by the shape, and so on. But then you need to make some assumptions about how fast mutations occur, in order to go from that leaf fossil to a number that you can apply to the rest of the sequences, right? Because, as you said, you don’t have reliable cyanobacterial fossils. So how do you go about taking a number that you estimate for a leaf and applying it to the cyanobacterial genomic record?

WOODY FISCHER: We’re always interested in extrapolating to some point in time that is older than one that we can calibrate with fossils. At some point in time your rate is well calibrated, you can demonstrate that it's behaving well through the fossil record, and you're always interested in stepping a little bit deeper. The quality of this “molecular clock” is only as good as you're willing to extrapolate. The farther that you extrapolate from your calibrated data set, the worse and worse the estimate is. So, with Patrick¹⁴ we've tried to estimate how old crown-group¹⁵ cyanobacteria are. We now have to step deeper than anything that we can calibrate, and so consequently, we carry a large degree of uncertainty with that. Our best estimates for crown-group cyanobacteria are something to the tune of 2 billion years ago,¹⁶ but they carry uncertainties that are measured in hundreds of millions of years, which is a huge uncertainty. That said, it is good that we can say, “it's not three and a half billion years ago, it's not four billion years ago.” That has real value. That's the kind of logic by which molecular clocks are applied. It's important to keep in mind that those molecular clocks are not strict chronometers like radiometric dates. They're much more of an extrapolation from a distribution that hopefully you have calibrated well.

AVI FLAMHOLZ: I've struggled with this as well, I think: finding the balance between wanting to give and answer and being responsible with the data you have. It's a tough one. Given that you're interrogating the evolutionary history of the early Earth, how do you deal with the fact that we may never know answers to some of the questions you're posing?

WOODY FISCHER: This is, I think, a really great question. Let me give you an example of the place where we've ended up, that is a little frustrating, but very real. We're very interested in oxygenic photosynthesis. We know that oxygenic photosynthesis emerged from anoxygenic¹⁷ photosynthesis, so a big part of this problem involves understanding the evolutionary history of anoxygenic photosynthesis. This is another place where metagenomic sequencing—the ability to go into the environment and sequence DNA more or less agnostically — has been a really big breakthrough. Now we have genomic representation from all kinds of different bacteria that we didn't even know existed, and it's allowed us to do a much more complete survey of anoxygenic photosynthesis. One of the things that we recognized in this data is anoxygenic photosynthesis is widespread in the bacteria, more widespread than we thought previously. It's present in seven phyla and most of those phyla have a tremendous amount of metabolic diversity — they have some photosynthetic members and some members who are not photosynthetic but use other metabolisms. So it looks like none of those groups were ancestrally photosynthetic. That is, for all of the groups that we know do photosynthesis today, you can't make the case that any of them was ancestrally phototrophic. So lateral gene transfer¹⁸ must play an incredibly important role in the evolution of photosynthesis. And yet some group out there invented photosynthesis, but it doesn't look like it's any of the groups that are present today. If we want to trace photosynthesis deeper than the groups that are present today, we have to leave the organisms behind and start talking about the proteins and genes, because we can't properly place them in any particular organisms. It's been popular for a long time to say “I study photosynthesis in cyanobacteria,” or “I study photosynthesis in Chloroflexi.” There's always this implicit interpretation that “of course I study the group that invented photosynthesis.” But with our current data, it appears to me that none of them are ancestrally phototrophic, and we actually have no idea of which group that invented photosynthesis. That’s sad because we would love to know. It's possible, you know, that group went extinct 3 billion years ago, and we're just not going to be able to resolve the earliest history of photosynthesis.

AVI FLAMHOLZ: In one of your reviews, you had this phrase I really liked, that evolutionary history is written by the winners. That really emphasizes the point that all we see when we look at biological diversity is the biological diversity that's exists today. Anything that went extinct in history, we've lost, unless we have a fossil of it.

WOODY FISCHER: I think that point is incredibly important. But we shouldn't feel like it's crippling, because all it requires is that we become innovative, that we develop an imagination that allows us to extract principles of what types of things should have come before on the basis of relatively simple chemical reasoning, and some ideas about how evolution works. We might be able to make some hay with those kinds of principles…

You know there's this really long-lived idea that emerged from some Precambrian paleontologists. The idea is that there's no microbial extinction. For example, you have to brush your teeth twice a day. Like once wasn't good enough. The bacteria came back. But teeth aren’t really a good example, I think, because we know there must be tremendous microbial extinction. One of the reasons that we know that is when we do phylogenetics, we populate, for example, a gene or a protein tree with as many genes and proteins as we can possibly find, and we there are always long branches.¹⁹ It is impossible to explain those long branches without extinction of all the groups that would have once existed on that branch.

AVI FLAMHOLZ: Or there are things that we haven't sequenced yet, right?

WOODY FISCHER: Yeah, so the argument is “we're just waiting to find it.” I think there's plenty of opportunity for metabolisms, physiologies, ways of being, to just disappear. They just don't make sense anymore. We already know there's a huge richness of anaerobic biology on the planet today. I'm sure it pales in comparison to all the different modes of anaerobic life before the rise of oxygen, all the different types of fermentation that were possible then. Maybe they're just not important anymore, because it's important to do something else. One of the things that we've looked at quite a bit is the idea that manganese oxidation²⁰ played a role in the evolution of water oxidation. But manganese phototrophy is not known to exist today.²¹ Maybe it exists in groups that are waiting to be discovered, but it's also possible that it went extinct because if you figure out how to split water, everything changes. Manganese isn't as abundant as water and it's not going to be very abundant once oxygen's present in the environment. ²² You know what, maybe it's just not worth doing anymore? My spidey sense strongly suggests that extinction is common in microbial groups — there are physiologies and ways of being that just don't make sense anymore.

AVI FLAMHOLZ: We kind of skipped the intro because I led in with this question about the geosphere. I could write an intro, but I think you're better posed than me to give one. So: for me, the metaphor has been thinking about the atmosphere and the biosphere as being a battery, where you have spatial separation of reductants and oxidants. Photosynthesis basically generates the battery by producing O₂ gas and also reduced organic carbon molecules, which separate from each other and can be brought back together by organisms like us that eat reduced carbon and breathe air. Is that a good metaphor for somebody coming outside the field?

WOODY FISCHER: I usually explain a couple of points. First, I like to think about energy. Nearly all the available energy is in visible light that hits our planet. Oxygenic photosynthesis is all about how to turn visible light into something that's useful. It's so profound that it is the core engine behind the biosphere. Even though much of the photosynthesis that happens today today is due to algae in the oceans and plants on land, it all owes itself to this singular bacterial invention. When I'm trying to describe why this is important, I think it's easy to just say, “this is the most important bioenergenetic innovation in the history of life on our planet.” It ramps up the carbon cycle to eleven, and that's going to impact all the other biogeochemical chemical cycles, by both the production of carbon and the production of oxygen. The iron cycle notices oxygen, the sulfur cycle notices, and the nitrogen cycle notices oxygen. All of the major biogeochemical cycles now are entangled with oxygenic photosynthesis.

If you think about it in the context of the history of our planet, oxygen is a relatively recent innovation. It occurs at about the halfway mark. The idea that we might use oxygen as a marker for life on planets outside of our solar system is exactly because of its history on Earth. It's not until after oxygen that things pick that up. ²³ The last point that I think is really valuable is that this photosynthetic metabolism evolved only once. The are many things on our planet that evolve a bunch of different times, and even things related to photosynthesis that evolve a bunch times. C4 photosynthesis is a great example²⁴ where we can demonstrate that it evolved independently dozens of times, and that's because it’s a ready solution to an ever present problem. Oxygenic photosynthesis evolved only once. As we try to think about solar fuels, we might wonder how we are going to emulate this thing that plants do so effortlessly, and maybe even make it better? All of that thinking is wrapped up in contingency. Why is it that photosynthesis evolves once? Is it because it is such a hard process that there's really only one good solution? Or maybe there are things that plants and algae and cyanobacteria have to do that solar fuel technology doesn’t have to do at all? Maybe we should think very broadly about all possible solutions. We don't have answers yet, but that's all wrapped up in our understanding about where photosynthesis comes from.

AVI FLAMHOLZ: Oxygenic photosynthesis evolved one time in the cyanobacteria, and that's the progenitor of all the green photosynthesis on the planet. You’ve said that maybe it’s really hard to orchestrate and that’s why it only evolved once. Aside from the point that it produces molecular oxygen, what’s so hard about it?

WOODY FISCHER: We have a lot of interaction with chemists that are working on making of molecular oxygen, a process that's associated with the water splitting end of photosynthesis. Surely I can design a lot better catalyst than these living things have using cobalt, for example. Why not use cobalt? Some of the best water oxidation catalysts use iridium catalysts. These are incredibly exotic materials that are well suited for certain chemistry. Why didn’t photosynthesis pick those? They’re just rare. Life isn't going to pick something that's incredibly rare.

OK, that's one answer, but we actually have come to a very different answer, and that's just our observations that we've made of sedimentary rocks that were made just prior to the rise of oxygen. We’ve been asking questions about what manganese looks like in this environment. There's a lot of manganese in surface seawater, and there's a lot of light in surface seawater. Before you can split water, the currency for the biosphere are electron-rich substrates like iron, or organic compounds, or reduced sulfur compounds, or molecular hydrogen, and many of those are stripped out of the environment—iron in particular is one that we can measure most readily. And so the biosphere is starved for a source of electrons. Today, we think of the biosphere as being starved by nitrogen or phosphorous — if I can put fertilizer²⁵ on a plot of land, I can get a higher yield from that. You could go do that on the early Earth all you want — put as much phosphorous or as much fixed nitrogen into the ecosystem, and it won't be any more productive, because it actually needs a source of electrons. There's an opportunity for an organism living in late Archean seawater to pull electrons off manganese if it can figure out how. We think that that is a process that ultimately could put the progenitors of Oxyphotobacteria, this group of cyanobacteria that invented oxygenic photosynthesis, on the path to oxidizing water.²⁶

That wasn't a decision those cells made about how this is going to be elegant chemistry, or how it's going to be useful in acidic media, or whatever. There's absolutely nothing rational about that. Those cells know nothing about water splitting. All they know is that they can make a living by oxidizing manganese. Maybe the choice of manganese was doesn't necessarily have to do with whether it's useful, or valuable, for this chemistry, or it's even relatively abundant on the planet—maybe it just has to do with the fact that, you know what, surface seawater in parts of the world 2.5 billion years ago had a lot of manganese. That's one example of where a kind of geobiological approach to thinking about some of these things might free up your chemical imagination a little bit.

AVI FLAMHOLZ: There’s a debate, at least in my part of the biological sciences about whether there are really principles in biology.

WOODY FISCHER: I love the idea that one might be able to extract principles. Not because it necessarily leads to truths, but it can be really valuable from of pedagogical standpoint. Let me give you one example from geomorphology, since you're at Berkeley. There's a very famous geomorphologist there, a guy named Bill Dietrich, and he coined a set of “ geomorphic transport laws.” They are relations that were used to describe how, for example, water moves sediment around on the surface of the planet. They're not really laws, and it's hugely complex, and any physicist would say, “There's nothing in this that's at all like a law,” but it was really valuable for the community to state some of these things explicitly, and then think about ways to test them. If we can start doing things like that for biology, it would be wonderful. The problem is, it's been so hard to extract things that are sufficiently testable to have a lot of value. Now, in geobiology and microbial ecology, the thing that I can think of that is closest to a law is, “Everything is everywhere, and the environment selects.”²⁷

If I take a beaker, put some rich media in there and I heat it to 100 C and leave it open on the roof of the building here, ²⁸ where it’s outrageously hot, ultimately I will be able to culture a strain of microbes that is similar to what you might find in a hydrothermal system in Yellowstone, or at the bottom of the ocean. The idea is that cells are mixed all over the surface of the Earth. All I have to do is create an environment that works for that cell, and they’re going to grow. And so that is as close as anyone has come in to defining a law in the area of geobiology. It's very difficult to prove wrong, and appears to have some value in describing which groups of organisms are similar. That's a long-winded answer, but I think this idea that we can extract truths that can be really useful — maybe not as truths, but as ways to help organize activities.

AVI FLAMHOLZ: To strengthen the questions about truths and principles: if I heat that media to say 120 C, that's above the hydrolysis temperature of the peptide bond, surely nothing will live in it, right? So the principle that life uses proteins²⁹ to do stuff applies to tell us that we're not going to get anything growing at that temperature.

WOODY FISCHER: Absolutely, but in that case, you would find something. 121 C is the current limit!

AVI FLAMHOLZ: (Laughter) Really!?

WOODY FISCHER: That shouldn't happen, right? I think principles can be really valuable, because there are always exceptions that prove the rule. So, you would say, “Oh, well of course, 120 C, no one's living above that, because they literally can't hold themselves together,” except then you find an example and you wonder “how did these guys overcome that problem that I know they must have?” Sure enough, there is some trick about how they tick in as many charged amino acids in their proteins as they can — something like that. So, I think that principles can have really good value, especially when they're things that are rooted in some physical or chemical phenomena that you can just bank on. Even when they're broken, it's not because the physics or chemistry's wrong, it's because there's some elegant solution, there's some elegant way around it.

AVI FLAMHOLZ: Something incredibly persistent about the biology.

WOODY FISCHER: Yeah, absolutely.

AVI FLAMHOLZ: OK, a wind-down questions for you. I read your website, and it seems like you interested in Mars. You wrote “There's hope that Mars will teach us about the early Earth,” and that fascinates me. What is there to be learned from Mars, seemingly a very different planet in a lot of respects, about the early Earth?

WOODY FISCHER: I think this is a huge breakthough: in the past ten years, we’ve recognized that Mars has a sedimentary record. We didn't know that until those two rovers, MER-A and MER-B, and now, finally, MSL Because it has a sedimentary record, we can ask questions about its history. It gets even better than that, because we can now do comparative planetology. We can ask: what aspects of Mars sedimentary record look similar or different to that of the early Earth? But it gets way more interesting, because observations of Mars' sedimentary record tell us that it was warmer and wetter in the past. From observations of sedimentary rocks on Mars, we have discovered very ancient wet environments that were completely habitable. We could envision microbes growing. If I could inoculate this environment with some soil from the Earth, something would almost certainly grow there. That’s amazing! We didn't necessarily expect that.

From the broad perspective, early Earth was habitable and inhabited. Early Mars was habitable, but we don't know if it was inhabited. We can start to ask questions about what does Mars oxygen looks like. What does Mars water look like? How is it that Mars ends up going down a very different trajectory than Earth? If early Mars and the early Earth are fairly similar, how is it that one planet ends up sustaining life for 4 billion years, and the other one is apparently empty? It's about as close as you're ever going to be able to get to asking questions about how unique are we and how special are we. We do so on Mars using much the same tools that we use for the early Earth, in terms of interrogating the textures and chemistry of those sedimentary rocks. There's this incredible opportunity to contextualize the history of our planet, just by looking at our red neighbor.

AVI FLAMHOLZ: You mentioned that we didn't know that Mars had a sedimentary record. Why was that a question? Just because of the question of water?

WOODY FISCHER: Yeah, so we knew that Mars wasn't wet anymore, and we know that it's mostly made of this rock type basalt, which you might find on Hawaii. Basalt makes up the bottom of the oceanic crust, bottom of the oceans everywhere, and so the planet doesn't look particularly weathered. If it's not very weathered, maybe there hasn't been any water. Then there shouldn't be any sedimentary rocks, or maybe there would just be a thin veneer. It was really amazing to find that there were actually wet environments on early Mars, and a huge thickness of sedimentary rocks. There are kilometers of sedimentary rocks preserved in many places. In many places, we really do have Grand Canyon-style stratigraphy to interrogate. That's a record of Mars history that no one was expecting to find. Where MSL is right now, for example, was a lakebed. It's driving around on sedimentary rocks that were deposited in a lake bed, and that's an amazing environment to study the chemistry of that lake water, what redox processes were there, and how long was it there. Now we can ask questions like that in a way that's completely serious.

AVI FLAMHOLZ: If there was life on Mars when it was wetter, what are the odds we'd be able to find that out in a conclusive fashion?

WOODY FISCHER: Certainly if there were plants on Mars, we'd nail it. We'd have no problem—the fossils would be very clear. But because we're basically not going to have fossils, we're going to be talking about textural or chemical interpretations. These can be very subtle inferences, very indirect ways of knowing. We’re going to have the same exact issues that we wrestle with when we're trying to interpret the earliest evidence of life on Earth. When we say, “how old is life on Earth?” You will get a different answer for each geologist that you talk to. That's because it's not enough to say, “Oh, in some of these sedimentary rocks that are 3.7 billion years old, we have found graphitic carbon.” OK, but there's carbon in meteorites, there's carbon in the Solar System that has nothing to do with biology. You can't demonstrate conclusively that it’s biological in origin. All right, well, it has a carbon isotopic composition that's consistent with carboxylation via the Calvin cycle.³⁰ But there are plenty of abiotic syntheses, for example, Fischer–Tropsch syntheses, that make hydrocarbons with a similar characteristic, etc. So as we try to interrogate these things, we'll have exactly the same challenge on Mars as we did on Earth. Many things will be suggestive of the types of environments that we associate with the biosphere, but we're not going to be able to demonstrate it robustly, ultimately maybe until we find a fossil. It might be a microfossil, but it's got to be of sufficient quality that, whatever the number is, 99% of paleontologists would agree that this thing's a micro-fossil. That's how we do life detection.

AVI FLAMHOLZ: Like the toothpaste! “Three out of five dentists agree...” “99% of paleontologists agree that there's life on Mars.”

WOODY FISCHER: Exactly, but that's what it will take. That's really what it will take. So, we talk about stromatolites all the time. Are you familiar with stromatolites?

AVI FLAMHOLZ: Yeah, but I don't think everyone who will read this is, so…

WOODY FISCHER: Yeah, I won't get into it. But the point is that stromatolites, some of them are made in the presence of microbial mass, others are made in the complete absence of microbial mass. So just wait until we find stromatolites on Mars! It'll kick off great excitement, and then a bunch of letters to Nature and Science saying, “You guys can't say that!”

AVI FLAMHOLZ: OK, here's the big one, since we got all the way here, and you study Mars and photosynthesis. Practically speaking, how would you terraform Mars? If we needed to live on Mars, what would we need to do to make Mars modestly habitable for humans?

WOODY FISCHER: So the biggest thing is that you need to be able to—you need to start to stabilize water on the surface, which basically means you need to give Mars back an atmosphere. We think it once had a much more substantial atmosphere than it does today, but it's lost much of it. So you have of replenish it. Ideally you would also replenish it with CO₂ because it's a greenhouse gas, so it would allow you to warm up the surface of the planet with the available solar energy. Once you got that done, we think that there's enough water in the subsurface that you might be able to get plenty of water for the surface. Then you would go about the task of growing some plants.³¹ People don't talk about it very much, but there is oxygen on Mars. It's actually the third most abundant gas in the atmosphere. It's not super-abundant, but it's probably abundant enough at least for microbes to breathe. We're not going to get there yet, we're still going to be breathing out of tanks until plants turn over enough of the CO₂ in the atmosphere to produce enough oxygen that we could breathe.

AVI FLAMHOLZ: But we need to net add CO₂ to the atmosphere. Would we have to bring that from Earth, or is there some other place to get it from?

WOODY FISCHER: Well, that's the thing that's so amazing. We don't actually know. So, we think Mars lost a lot of CO₂—we think its atmosphere had a lot of CO₂, and it lost it. Whether it was lost to space through photochemical reactions, or lost to the crust, we don’t know. The way that CO₂ gets lost to the crust is through weathering reactions that ultimately converts it into calcium carbonate salts. There's some suggestion that Mars may have a lot of calcium carbonate salts in the crust. If that was the case, then you could imagine harvesting CO₂ from those carbonate salts. Now, you've got to get acid from somewhere in order to make that happen. I don't know if we're going to bring a spaceship of HCL that we can dump on the crust, or launch a spaceship with a big acid beaker that can just break on the surface of the planet.

AVI FLAMHOLZ: There must be some bacteria that makes a living with a low amount of oxygen, itself doing the weathering with self-generated acid and minute amounts of water.

WOODY FISCHER: Yep, yep, yep. That's an idea.

AVI FLAMHOLZ: Here we are. (laughter) I think that's as good as anyplace to stop.

---

¹ Between 100 and 1000 times more concentrated. Read more here: http://book.bionumbers.org/what-is-the-elemental-composition-of-a-cell/

² The Cyanobacteria are a clade of bacteria in which oxygenic photosynthesis – photosynthesis with two photosystems and chlorophyll, producing molecular oxygen – first arose. As far as we know oxygenic photosynthesis arose only once in the cyanobacteria. All other green-lineage photosynthesis, including algae and grasses, persimmon trees, douglas firs and hipster succulent arrangements are all derived from the Cyanobacteria through the endosymbiotic (internally symbiotic) engulfment of a photosynthetic cyanobacterium by an ancient eukaryote. https://en.wikipedia.org/wiki/Symbiogenesis

³ When we say that oxygen is “energetic,” what we really mean is that it has a strong tendency to accept electrons. Not only can molecular oxygen (O₂) accept electrons from nearly any organic donor molecule, but it can do so spontaneously and quickly from particular classes of organic molecules (i.e. without any catalyst). This reactivity is a big problem and also an enormous opportunity for biology. On the one hand O₂ will spontaneously grab electrons that were “intended” for other biochemical purposes, sometimes destroying important bits of the cell in the process. On the other hand, the tendency of O₂ to accept electrons can be harnessed (through the process of respiration) to store chemical energy for organisms to use later. Many bacterial, fungi and animals, ourselves included, can generate energy in the absence of O₂. This is what happens in when our muscles run out of oxygen and make lactic acid. However, respiratory metabolism – i.e. breaking down food and giving the electrons in it to oxygen – produces up to 15 times more chemical energy from every morsel of food.

⁴ Oxygen is a powerful oxidant, meaning recipient of electrons. Elements that react with oxygen rust, or become oxidized, meaning that they lose electrons in the interaction. The fact that they appear in their reduced for in sediment implies that oxygen was very scarce when that sediment was formed: they would have oxidized had oxygen been present.

⁵ Certain minerals admit uranium into their structure but disallow lead. Since isotopes of uranium decay into lead, we know that all the lead in these minerals was once uranium. Based on independent experimental characterization, we also know how long it takes for various uranium isotopes to decay into lead – the so-called half-life. Using these two pieces of information, we can back-calculate how old the mineral is, i.e. when all of the lead inside of it was uranium.

⁶ i.e. the capacity to utilize oxygen to break down organic (reduced) carbon compounds and generate energy for the purpose of organizing life (breaking equilibrium in myriad ways, big and small).

⁷ Aerobes are organisms that utilize oxygen. Many organisms, especially bacteria and Archaea are anaerobes, meaning that they do not utilize oxygen at all and are even poisoned by it. Woody is saying here that this poisoning is inherent – that it is due to the chemical nature of life and that even obligate aerobes, which require oxygen to live, must cope with the toxicity of this molecule. The toxicity is due to the fact that oxygen is a particularly good oxidant – recipient of electrons – really the best one that participates in biological reactions. The fundamental problem is that many of the most central chemical reactions in life involve the targeted donation of electrons. Oxygen, being such a glutton for electrons, can steal electrons from these reactions, electrons intended to for other destinations. The intervention of oxygen has many effects beyond simply mucking with the progress of these reactions, including inactivation of proteins and modification of nucleic acids like DNA and RNA. All cells, even obligate anaerobes, have molecular tools for dealing with the presence of oxygen. They simply differ in terms of how much oxygen they can tolerate.

⁸ N2 gas in the atmosphere is nearly inert. It takes a tremendous amount of energy to break the triple bond of N2 and form ammonia, which can be utilized directly organisms. Indeed, synthetic production of ammonia through the Haber-Bosch process requires enormous energy input in the form of high pressures and temperatures. Haber-Bosch is one of the great achievements of modern chemistry, responsible for both the production of synthetic fertilizers and WWI era explosives. https://en.wikipedia.org/wiki/Haber_process

⁹ And vice-versa.

¹⁰ Because we can use the decay radioisotopes to determine the absolute age, in millions or billions of years, of geological samples (as described above). So we can use comparative genomics to figure out whether some photosynthesis gene from trees is more closely related to the version in kelp or cyanobacteria, but we then look at the geological record to figure out roughly when the common ancestor of trees and kelp was present on the Earth. This is what Woody means by the phylogeny getting “calibrated by the sedimentary record.” Similarly, we can determine an approximate date for the emergence of O₂ in Earth’s atmosphere – the so-called “Great Oxidation Event” occurred about 2.5 billion years ago – by examining the “rusting” or “oxidation” of metals in the sedimentary record. Oxygenic photosynthesis must have evolved before this date because oxygenic photosynthesis produced the oxygen responsible for the rusting.

¹¹ The oldest algal fossil we have is of the red algae Bangiomorpha. It is about 1 billion years old. But red algae are eukaryotes, meaning that they have compartments in their cells and have chloroplasts that derive genetically from free-living cyanobacteria. So photosynthesis clearly evolved well before Bangiomorpha did. In order to figure out when that happened, we need to project back in time from Bangiomorpha, the earliest event we can “calibrate” to the fossil record of algae. To do this we need to estimate an evolutionary rate – the amount of time that it takes for genes to accumulate a single mutation. Then we can count up the number of mutations between the cyanobacterial photosynthetic genes and the Bangiomorpha genes, say, in order to estimate how long ago they diverged from each other. This approach comes along with a big assumption – that the observed rate of mutations after the emergence of Bangiomorpha (which we have a real date for) is similar to the rate of mutations before Bangiomorpha. Unless we find definitive fossils of cyanobacteria, this assumption is sadly unavoidable.

¹² Until a few years ago, it appeared that the closest relatives of cyanobacteria were quite distantly related to them. This was evidenced by rather sizable differences between cyanobacteria and their closest known relatives in terms of the sequences of important genes (like the ribosome). However, recently two groups of much closer cyanobacteria relatives were discovered in recent metagenomic studies – attempts to sequence all the microbes in a particular environment. Because these newly discovered relatives are not photosynthetic, they give us a better handle on the cyanobacteria – a reference point for guessing when cyanobacteria evolved the capacity to perform oxygenic photosynthesis.

¹³ i.e. without attaching a “number of years ago” to it, but just saying “this happened before that.”

¹⁴ Woody is referring to this paper http://onlinelibrary.wiley.com/doi/10.1111/gbi.12200/full

¹⁵ i.e. photosynthetic cyanobacteria, as opposed to their close relatives that, as mentioned above, are not photosynthetic.

¹⁶ i.e. about 500 million years AFTER the great oxidation event.

¹⁷ i.e. not producing oxygen.

¹⁸ Lateral transfer involves movement of genes between potentially unrelated organisms. Lateral transfer stands in contrast to vertical transfer, or transfer by descent, which is what we are used to in thinking about the reproduction of animals and plants. Sexual reproduction of animals mixes the genes of the parents into their offspring, generating a child that is different than either parent. Lateral transfer involves the transfer of a few genes, typically, from one bacterium to another. Like sexual reproduction, lateral transfer produces genetic diversity, but a very different kind of genetic diversity where entirely new genes may be introduced in the transfer but only a very small number of genes are affected.

¹⁹ Long branches on a phylogenetic tree represent relationships between groups that are not very closely related. Since we know that evolution takes place by mutations to DNA sequences and we know that most mutations are small, e.g. one letter being removed or being swapped with another, a long branch implies a lot of individual mutations and therefore a lot of intermediate forms of the gene in question. It’s hard to imagine that in all cases where we see long branches none of those intermediate forms branched off, did it’s own thing for a while, and then went extinct.

²⁰ This is a fascinating and subtle issue. The protein that does the chemistry of oxidizing water for oxygenic photosynthesis is called the water oxidizing complex (WOC) and it uses manganese as part of its mechanism. The manganese is very central and very important chemically. But this protein is not similar to any other proteins that we know of – as far as we know it is an evolutionary singularity. So the question is: how did this whole manganese based oxidation of water get going? We have nothing to compare it to, so we are at a bit of a loss. Based on some compelling geochemical evidence, Woody has proposed that manganese was an important electron donor for photosynthesis before oxygenic photosynthesis arose. The problem with this explanation is that no known bacteria do manganese based photosynthesis today – they all use other metals. There are a couple reasons this could be, but Woody is here lobbying for the explanation that the WOC arose in a cyanobacterium that previously used manganese. Once there was oxygenic photosynthesis, it was so much better than using manganese that all the cyanobacteria switched over or died out.

²¹ Except in a very limited sense in cyanobacteria, where electrons are withdrawn from manganese as part of “setting up” the water oxidizing complex.

²² Because it becomes oxidized very rapidly in the presence of O₂ and then precipitates. If O₂ was present, manganese would be depleted from the oceans of early Earth and, therefore, much less useful for aquatic organisms (which was everyone 2 billion years ago).

²³ e.g. the Cambrian explosion in vertebrate diversity happened about 500 million years ago and is thought to be related to an increased availability of oxygen, perhaps due to the emergence and success of land plants.

²⁴ C4 photosynthesis is a modified form of photosynthesis found in some land plants like sugar cane and corn. It requires some metabolic changes and a particular anatomical arrangement where some cells do the work of harvesting light energy and other cells do the work of converting that energy into reduced carbon. Comparative genomics research has shown that this morphology and metabolism evolved independently many times in the plant lineage. C4 photosynthesis is thought to improve plant growth at elevated temperatures. http://onlinelibrary.wiley.com/doi/10.1111/j.1469-8137.2004.00974.x/full

²⁵ Plant fertilizers typically contain fixed nitrogen (in the form of ammonia) and phosphorus.

²⁶ And thereby making oxygen.

²⁷ A principle first stated by Baas Becking, a very early geobiologist. https://en.wikipedia.org/wiki/Lourens_Baas_Becking

²⁸ In Southern California.

²⁹ Proteins being made of peptide bonds.

³⁰ The cycle that oxygenic photosynthetic organisms use to harness light energy and use it to reduce carbon dioxide, thereby producing biomass. https://en.wikipedia.org/wiki/Calvin_cycle

³¹ To make oxygen for humans to breath and food for humans to eat.

---