Recently congress recommended that NASA create an Ocean Worlds Exploration Program whose primary goal is “to discover extant life on another world using a mix of Discovery, New Frontiers, and flagship class missions”. Pretty awesome. In February I was invited to participate on the science definition team for the Roadmap to Ocean Worlds (ROW) initiative. The ROW is step one in NASA’s response to the congressional recommendation. Over the last few months the science definition team has been meeting remotely to hash through some challenging questions around the central themes of identifying ocean worlds, defining their habitability, and understanding the life that might live in or on them. This week we will have our first and only opportunity to meet in person as a team. As one of very few biologists in the group, and possible the only ecologist, I’ll be giving a short talk on polar microbial ecology as it pertains to discovering and understanding life on other worlds. As a way to organize my thoughts on the subject and prep for the talk I’m experimenting with writing out the main points of the presentation as an article.
I decided to title the talk (and this post) Polar microbial ecology and the search for totally normal life because it reflects an idea that I and others have been advocating for some time: there isn’t anything that special about life in extreme environments (gasp!). Life functions there in pretty much the same way that it functions everywhere else. One way to think of this is that polar microbes (and other extremophiles) are uniquely adapted but not unique; they follow the standard rules of ecology and evolution, including how they acquire energy and deal with stress. The nuances of how they interact with their environment reflect the unique characteristics of that environment, but the microbes are generally kind of conventional.
Act I: Challenges and opportunities for polar microbes
As with microbes in any environment, polar microbes are presented with a range of challenges and opportunities that define how they live. From a microbial perspective challenges are stressors, or things that cause damage. Opportunities energy sources. Very often these things are the same. Take sunlight, for example. This is the single most important source of energy on the planet, but get a little too much of it (or any of it, to be more accurate), and you start doing damage to yourself. As a result you have to invest some of the acquired energy into offsetting this stress. We can illustrate this with a very simple conceptual model.
The above figure shows a hypothetical (but plausible) relationship between energy and stress. Energy is shown both on the x-axis and as the blue 1:1 line. Stress is the orange line. In this scenario as energy increases, stress increases logarithmically. The energy that is available for growth (that which is not dedicated to offsetting the damage caused by stress) is energy – stress, or the difference between the blue and orange lines. That looks like this:
The units for biomass are, of course, arbitrary, what matters is the shape of the curve. Past a critical energy value the amount of biomass an ecosystem can sustain actually decreases as energy increases. At the point at which stress = energy, no biomass is being accumulated. It isn’t a coincidence that the shape of the biomass curve is the same as experimentally determined temperature curves for bacterial isolated belonging to different temperature classes (below). In those curves the energy in the system is increasing as temperature rises. This enhances enzymatic reactions (producing more biomass) until there is so much energy that the system becomes disordered (enzymes denature). This process is directly analogous to the one presented here, but takes place at the population level rather than the ecosystem level.
One thing that I would like to make clear up front is that low temperature itself is not a stressor. Low temperature alone doesn’t kill single-celled organisms, it only impedes their ability to deal with other stress. Because of this energy and stress have a very complicated relationship with temperature in cold ecosystems. For example:
- As temperature decreases, energy decreases.
- As energy decreases, stress decreases (remember that in our conceptual model stress is a function of energy).
- As temperature decreases, access to substrate (and by proxy energy) increases.
- As temperature decreases, inhibitory compounds (which are stress) increases.
So many of the effects of low temperature are in conflict. Highly adapted psychrophiles play off these conflicts to establishes niches in cold environments. We can illustrate this idea by focusing on the last two bullet points, which are specific to environments that contain ice. Impure ice (which virtually all environmental ice is) does not form a solid structure when it freezes. It forms a porous matrix, with the porosity determined by the concentration of solutes and the temperature. You can actually view this in 3D using sophisticated X-ray tomography techniques, as shown in this figure from Pringle et al., 2009.
The gold spaces in this figure are pore spaces within the ice. As the ice gets colder the pore spaces become smaller and the solutes contained in them become more concentrated. That’s because whatever was dissolved in the starting material (e.g. salt, sugars, small particles) is excluded from the ice as it forms crystals, the excluded material ends up in the pore spaces. The colder the ice, the smaller the spaces, the more concentrated the solutes. Bacteria and ice algae that are also excluded into these spaces will experience a much higher concentration of nutrients, potential sources of carbon, etc. This equates to more energy, which helps them offset the stress of being in a very salty environment. In very cold, relatively pure ice, solute concentrations can be 1000x those of the starting material. Neat, right?
Here’s what decreasing temperature actually means for a bacterial cell trying to make a go of it in a cold environment:
Our earlier plot of biomass depicted cells in the “growth” phase shown here. Only very early in the plot, when energy was very low, and at the very end, when stress was very high, did biomass approach zero. That zero point is called the maintenance phase. Maintenance phase happens when decreasing temperature suppresses the ability to deal with stress to the point that a cell cannot invest enough resources into creating biomass to reproduce. The cell is investing all its energy in offsetting the damage caused by stress.
At increasingly low temperatures the synthesis of new biomass becomes increasingly less efficient, requiring a proportionally greater expenditure of energy (we say that bacterial growth efficiency decreases). At the end of the maintenance phase the bacterial cell is respiring, that is it is creating energy, but this is resulting in virtually no synthesis or repair of cellular components. This leads inevitably to cell death when enough cellular damage has accumulated. It’s sad.
If we lift our heads up out of the thermodynamic weeds for a moment we can consider the implications of all this for finding life on another cold world (all the good ones are cold):
On the assumption that everyone’s asleep by this point in the presentation I’m trying to be funny, but the point is serious. You hear a lot of silly (my bias) talk in the astrobiology community about life at the extremes; how low can it go temp wise etc. Who cares? I’m not particularly interested in finding a maintenance-state microbial community, nor do we stand a very good chance of detecting one even if it was right under our (lander’s) nose. The important contribution to make at this points is to determine where life is actively growing on the relevant ocean worlds. Then we can try to figure out how to reach those places (no mean task!). The point of this whole exercise it to find extant life, after all…
Act II: Where are polar microbes distributed and why?
I can think of 8 polar environments that are particularly relevant to ocean worlds:
- Supraglacial environment (glacier surface)
- Interstitial glacier environment (glacier interior)
- Subglacial environment (where the glacier meets rock)
- Sea ice surface (top’o the sea ice)
- Interstitial sea ice (sea ice interior)
- Sea ice-seawater interface (where the sea ice meets the seawater)
- Water column below ice (the ocean! or a lake)
- The sediment-water interface (where said lake/ocean meets mud)
Each of these environments has a range of stress, energy, and temperatures, and this to a large extent defines their ability to support biomass. For the purpose of this discussion I’ll report biomass in units of bacteria ml-1. To give some sense of what this means consider that what the waste-treatment profession politely calls “activated sludge” might have around 108 bacteria ml-1. Bottled water might have 103 bacteria ml-1. Standard ocean water ranges between 104 bacteria ml-1 and 106 bacteria ml-1 at the very highest end. One further note on biomass… our conceptual model doesn’t account for grazing, or any other trophic transfer of biomass, because that biomass is still in the biological system. Bacteria ml-1 doesn’t reflect this, because bacteria are consumed by other things. So the values I give for the ecosystems below are at best a loose proxy for the capacity of each ecosystem to support biomass.
Let’s consider the sea ice-seawater interface and the sea ice surface in a little more detail. The sea ice-seawater interface is interesting because it has, among all polar microbial environments, probably the greatest capacity to support biomass (we’re talking about summertime sea ice, of course). It is located at the optimum balance point between energy and stress, as shown in the figure below. Despite the fact that the sea ice surface has relatively high biomass, it is located at the extreme right side of the plot. The trick is that biomass has accumulated at the sea ice surface as a result of abiotic transport, not in situ growth. I explored this pretty extensively back in the very early days of my PhD (see here and here).
Despite the fact that the sea ice surface doesn’t function as a microbial habitat, the concentration of biomass there might be relevant to our goal of finding extant life on another world. The accumulation of bacteria at the ice surface is largely the result of bacteria being passively transported with salt. Recall that saline ice is a porous matrix containing a liquid brine. All of the bacteria in the ice are also contained in the liquid brine; as the ice cools the pore spaces get smaller, forcing some of the brine and bacteria to the ice surface. Porous ice in say, the Europan ice shell would act similarly. Salt is easier to identify than life (in fact we know that there are large salt deposits on the surface of Europa), so we can target salty areas for deeper search. In astrobiology we like to “follow things” (because we get lost easy?); “follow the water” and “follow the energy” are often cited and somewhat useful axioms. So here we have a “follow the salt” situation.
Act III: Can we explore polar microbial ecology in life detection?
Act II ended on a life detection note, so let’s follow this through and consider other details of polar microbial ecology that might give us clues of what to look for if we want to identify life. Getting ecology to be part of this discussion (life detection) is non-trivial. Despite the fact that the whole field of astrobiology is really an elaborate re-visioning of classic ecology (who lives where and why), there isn’t a whole lot of interest in studying how organisms interact with their environment. This simply reflects the field’s disciplinary bias; the most active communities within astrobiology may well be astronomers and planetary geologists. If the most active community was ecologists we’d probably be ignoring all kinds of important stuff about how planets are formed, etc. To illustrate the utility of ecology for life detection here are two examples of ecological principles that might lead to life detection techniques.
Case 1 – Biological alteration of the microstructure of ice
Earlier I put forward that decreasing temperatures can lead to heightened stress because life in ice is exposed to higher concentrations of damaging substances at lower temperatures. The most obvious example is salt. In very cold sea ice the pore space salinity approaches 200 ppt, that’s roughly 6 times the salinity of seawater! Maintaining adequate internal water is a huge challenge for a cell under these conditions. One of the mechanisms for dealing with this is to produce copious amounts of exopolysaccharides (EPS), a hydrated gel (essentially mucous) containing polysaccharides, proteins, and short peptides. EPS buffers the environment around a cell, raising the activity of water and, in some cases, interacting with ice. This produces a highly modified internal structure, as shown in the images below. This alteration could be a useful way of identifying ice on an ocean world that has been modified by a biological community.
Case 2: Motility
Motility has been put forward before as an unambiguous signature of life, but the idea hasn’t really gained a lot of traction. Clearly one needs to be cautious with this – Brownian motion can look a lot like motility – but I can’t think of anything else that life does that is as easily distinguishable from abiotic processes. One additional challenge however, is that not all life is motile. Plants aren’t motile, at least most of them over the timescales that we care to stare at them for. Not all microbes are motile either, but I would argue that those that aren’t aren’t only because others are.
Consider the figure at right, which is a cartoon of two modes of bacterial life in the ocean. One of those modes is motile, and can be seen using flagella to follow chemical gradients (we call this chemotaxis) and optimize their location with respect to phytoplankton, their source of carbon. The second mode is much smaller and in the background; small-bodied non-motile cells that live on the diffuse carbon that they opportunistically encounter. This works because that would be an inefficient niche for motile bacteria to exploit. In the absence of motility however, chemical gradients constitute a very strong selective pressure to evolve motility. We can see evidence of this in the convergent evolution of flagellar motility (and other forms of motility) in all three domains of life. Although they may share a common chemotaxis sensory mechanism, the Bacteria, Archaea, and Eukarya all seem to have evolved flagellar motility independently. This means that it’s probably a pretty good feature to have, and is likely to be shared by microbes on other ocean worlds.
That was quite a lot, so to summarize:
- Microbial communities are oriented along gradients of energy and stress. At some optimal point along that gradient a maximum amount of biomass can be supported by surplus energy that is not being used to deal with stress (How much energy do you spend on stress? Think of how much more biomass you could have!).
- The relationships between energy, stress, and temperature are complicated, but Earth life generally works at T > -12 °C. This estimate is probably a little lower than the reality, because laboratory observations of growth at that temperature don’t accurately reflect environmental stress.
- Life is strongly biased towards surfaces and interfaces, these may provide enhanced opportunities for life detection (follow the salt!).
- The specific ecology of cold organisms can provide some further insights into life detection strategies. For example, motility might be an under-appreciated signature of life.