≡ Menu

Alternatives to DNA-Based Life

The question of whether or not we would recognize extraterrestrial life if we encountered it used to occupy mathematician and historian Jacob Bronowski (1908-1974), who commented on the matter in a memorable episode of his 1973 BBC documentary The Ascent of Man.

“Were the chemicals here on Earth at the time when life began unique to us? We used to think so. But the most recent evidence is different. Within the last few years there have been found in the interstellar spaces the spectral traces of molecules which we never thought could be formed out in those frigid regions: hydrogen cyanide, cyano acetylene, formaldehyde. These are molecules which we had not supposed to exist elsewhere than on Earth. It may turn out that life had more varied beginnings and has more varied forms. And it does not at all follow that the evolutionary path which life (if we discover it) took elsewhere must resemble ours. It does not even follow that we shall recognise it as life — or that it will recognise us.”

Bronowski wanted to show how human society had evolved as its conception of science changed — the title is a nod to Darwin’s The Descent of Man (1871), and the sheer elegance of the production reflected the fact that the series was the work of David Attenborough, whose efforts had likewise led to the production of Kenneth Clarke’s Civilisation (1969), among many other projects. If the interplay of art and science interests you, a look back at both these series will repay your time.

As to Bronowski, who died the year after The Ascent of Man was first aired, I can only imagine how fascinating he would have found new work out of the Foundation for Applied Molecular Evolution in Alachua, Florida. Led by Steven Benner, a team of scientists has addressed the question of alien life so unlike our own that we might not recognize it. Along the way, it has managed to craft a new informational system that, like DNA, can store and transmit genetic information. The difference is that Benner and team use eight, not four, key ingredients.

Image: This illustration shows the structure of a new synthetic DNA molecule, dubbed hachimoji DNA, which uses the four informational ingredients of regular DNA (green, red, blue, yellow) in addition to four new ones (cyan, pink, purple, orange). Credit: Indiana University School of Medicine.

DNA, a double-helix structure like the new “hachimoji DNA” (the Japanese term ‘hachi’ stands for ‘eight,’ while ‘moji’ means ‘letter’), is based upon four nucleotides that appear to be standard for life as we know it on Earth. ‘Hachimoji’ DNA likewise contains adenine, cytosine, guanine, and thymine, but puts four other nucleotides into play to store and transmit information.

We begin to see alternatives to the ways life can structure itself, pointing to environments where a different kind of structure could survive whereas DNA-based life might not. That could be useful as we’re beginning to put spacecraft into highly interesting environments like Europa and Enceladus, but to get the most out of our designs, we need to have a sense of what we’re looking for. What kinds of molecules could store information in the worlds we’ll be exploring?

Thus Mary Voytek, senior scientist for astrobiology at NASA headquarters:

“Incorporating a broader understanding of what is possible in our instrument design and mission concepts will result in a more inclusive and, therefore, more effective search for life beyond Earth.”

Creating something unusual right here on Earth is one way to approach the problem, but of course there are others, and I am reminded of Paul Davies work and his own notions of what he calls ‘weird life.’ The Arizona State scientist, a prolific author in his own right, has examined the concept of a ‘second genesis,’ a fundamentally different kind of life that might already be here, having evolved on our planet and remaining on it in what we might call a ‘shadow biosphere.’

Finding alternate life on our own planet would relieve us of the burden of creating new mechanisms to make life work in our labs, so perhaps the thorough investigation of deep sea hydrothermal vents, salt lakes and high radiation environments may cut straight to the chase, if such life is there. In any case, finding a second genesis would make it far more likely that we’re going to find life on other worlds, and such life, as Davies reminds us, might be right under our noses. Like ‘hachimoji DNA,’ such life would challenge and stimulate all our assumptions.

The paper is Hoskika et al., “Hachimoji DNA and RNA: A genetic system with eight building blocks,” Science Vol. 363, Issue 6429 (22 Feb 2019), pp. 884-887 (abstract). Thanks to Byron Rogers for an early tip on this work.


Comments on this entry are closed.

  • Alex Tolley February 27, 2019, 17:15

    While the 8 base DNA (and RNA) implies that a vastly larger amino acid suite could be used in protein building (assuming 3 base codons), I am skeptical this is necessary, or even a good system for evolution. I see it more as a good technique for biotechnology. The paper itself shows that the 4 new bases are not optimal for DNA coding and transcription, and we have little idea how these possible new codons (512 vs 64) could be translated and with what extra amino acids.

    I see this experiment as proof of concept that there are more possible combinations of bases for encoding than teh 4 terrestrial life uses, and therefore there is the possibility that other geneses have selected different bases (as well as genetic codes).

    Assuming more experiments show that translation can be made to work, then what would be interesting is to determine the pros and cons of these alternative DNA and RNA structures for life in different environments – temperature, pH, pressure, etc. This would lead to research on different amino acids and hence proteins to be similarly tested. [There are obvious ideas to explore for biotech applications, which is where I suspect commercial pressures will pursue.]

    • hiro February 28, 2019, 19:49

      Is it possible to change from 3 base codons to 5 or 7 base codon now that we have PBS-GZ-ACT to construct a bigger system?

  • Jer February 27, 2019, 18:26

    Similar to the AI discussions, I would suggest that ‘complexity’ would be the metric. Is there any inorganic matter more complex than the simplest organic matter? This is out of my wheelhouse but I point people to TED talks by: Craig Venter on creating functioning cells from scratch, Lee Cronin on the definitions of life (is it exclusive to biology?), etc… Further to spirit of this article, I suggest searching for and defining complexity within the universe, then examining the context of its creation and current situation – some may be Life, so might be ‘Other’?

  • Dave Moore February 27, 2019, 18:26

    It’s an interesting idea. An 8 base system would allow the same 64 coding possibilities with twin base pairs as opposed to DNA’s current triplet system to reach the same number. This would be an advantage (particularly if phosphorous was in short supply) in reducing the amount of DNA needed in each cell. However, the laws of probability mitigate against it.

    I’ve had the thought that life started off with a 2 base system before evolving the 4 base system. You will note that Adenine and Guanine like Thymine and Cytosine are variations on each other.

    If you start life off with a 50 base pair RNA molecule that can replicate itself, (this has been done in the lab) and have a mechanism that randomly assembles base pairs until the self-replicating arrangement is arrived at then there is one chance in 10 to the 34 of getting the right combination, assuming there is an equal chance each time of getting the correct base pair with the same chance of the chain terminating. In other words 5 to the power of 50. With a 2 base pair system, it’s 3 to the 50 or one chance in 10 to the 23, which is about a mole of the material as opposed to 10 to the 11 mole with the 4 base pair system.

    I maintain that the most likely way life arose is the most likely way i.e. the highest probability system of all the systems available. This would point to a 2 base pair system at the origin of life.

    An 8 base pair system would be far more improbable, and while it offers some advantage to complex life forms, it doesn’t offer much of an advantage to simple forms as they don’t have a lot of DNA.

    • Alex Tolley February 27, 2019, 22:18

      Using 8 bases and a 2-base codon would make for a greater impact of point mutations, wouldn’t it? This might make the hill climbing of a fitness landscape more difficult, as changes would be larger. Unless life sticks to this code, we then have to explain the transition to a 4 base, 3-base codon. It would be a lot easier to start with a 4 base system with a 2-base codon and then transition to a 3-base codon.

    • Andrei February 28, 2019, 6:56

      I do not think they did mean to say that life would have started evolving from that 8 base system.
      Life probably did start on the 2 base system, progress were probably slow as it could be as it were more of natural experimentation – without any pressure of metabolism or need to avoid danger – it were entirely prebiotic. The question have been asked if RNA life came first, but serious? With the hydroxyl group in RNA it is less stable than DNA.
      I find PNA more likely as the starting molecule.

  • Harold Daughety February 27, 2019, 22:16

    Geometry governs synthesis of proteins and then geometry governs how proteins are assembled into complex structures. An analogy is the assembly of complex mechanical devices via robots. The parts, whether made of rigid metal or made of interconnected atoms, fit in only one way to make a true copy. Chirality is another fixed requirement: the three dimensional molecule is possible only with the correct geometric isomer building blocks, else the growing protein molecule would branch off in a wrong direction and terminate the construction. DNA does far more than encode information, It is the encoded master template and assembly instructions. What is, works. What is contemplated, most likely would not.

    • Bruce D. Mayfield March 1, 2019, 11:01

      I concur, and you make points that I’ve tried making, only with better expression. Living cells and organisms are built from molecules, but otherwise, they are very machine-like. Only very advanced machines, complete with self-replication and adaptive reprogramming.

  • Thomas W. Hair February 27, 2019, 22:19

    I am sorry, but organic chemistry is a straightforward subject and all the machinations to come up with another way for self-replicating molecules (i.e., life) to do our thing is a bridge too too far. The interaction of carbon, hydrogen, oxygen , nitrogen, phosphorus, and sulfur cannot be replicated by any other elements on the periodic table. It is an absolute fact.

    • Alex Tolley February 28, 2019, 17:20

      Felisa Wolfe-Simon thought that her team had found organisms substituting arsenic for phosphorus. She was wrong, but it clearly wasn’t unthinkable.

  • Adam February 28, 2019, 5:20

    Hi Paul,

    Alex & Dave’s discussion is quite interesting – and a hail to Bronowski’s “Ascent of Man” is always welcome – but there’s another point worth making. The expansion of viable base-pair combinations means that the current ‘natural’ DNA triplet code with its specific four bases is now a *smaller* selection from all the possible DNA codes. This heightens the question as to *why* this particular triplet code – it’s one selection out of millions of possibilities. Even more possibilities now 8 viable base pairs have been demonstrated.

    Maybe we’re just one insignificant variation here on Earth, or maybe, as the late Vlad shCherbak and Max Makukov suggested a few years ago, our specific Code was a deliberate choice by a species that launched forth its Code via Directed Panspermia to seed the Galaxy? Nothing else is likely to endure for aeons like a Genetic Code and be spread far and wide so easily. Of course, as shCherbak & Makukov admit, the specific Code doesn’t hold a lot of data – a few hundred bits, basically a signature of the species that launched our specific DNA forth into the Galaxy. But that it’s *just* one possibility out of a huge range of *equally probable* options is why they suggested it might’ve been a deliberate choice.

    Intriguing speculation, but hard to imagine how it can be demonstrated without some independent corroborating evidence. I do wonder what form that might take – but after 4 or more aeons, it’s hard to imagine anything else that’s sufficiently enduring.


    • Alex Tolley February 28, 2019, 17:15

      The issue with abiogenesis occurring elsewhere is that the can is just kicked down teh road. How was the biology we have originally created?

      I’ve seen work that suggests the code we have, assuming the 4 bases we have, is fairly optimal. That might be an answer to why we see this particular biology on Earth today. If the same biology was found in different star systems, and that there was no clear reason for the superiority of one biology over another, the source of the [a]biogenesis would be in question. At some point, perhaps well before we have samples of life from elsewhere, we may be constructing life using different genetic codes, DNA-RNA bases, and different amino acids. This will tell us a lot about carbon-based life.

      • Andrei March 1, 2019, 6:30

        It is also lazy ‘The origin of life is somebody else problem’
        Life started elsewhere and came here trough panspermia – so we need not research it as we cannot find the solution here but far away at another star.
        Also dangerous, creationist would applaud this and say we should not waste money researching the origin of life then.

  • ljk February 28, 2019, 10:42

    A chemistry course from the University of Colorado at Colorado Springs based on The Ascent of Man:


    I found his book online here:


    Episode 1 of the BBC Television series is here:


    An essay by Dr. Bronowski titled “The Creative Mind”:


    “Fifty years from now if an understanding of man’s origins, his evolution, his history, his progress is not in the common place of the school books we shall not exist.” ― Jacob Bronowski, The Ascent of Man

    • Alex Tolley February 28, 2019, 17:02

      “Fifty years from now if an understanding of man’s origins, his evolution, his history, his progress is not in the common place of the school books we shall not exist.” ― Jacob Bronowski, The Ascent of Man

      I remember the episode where he visited a concentration camp and was very emotional about what had happened and what it meant. In the US there seems to be a concerted effort by various groups to undermine the content in textbooks concerning the subjects he mentions in the quote. If he was alive today, he would be distressed at what is happening. Even in the relatively secular UK, a poll taken a decade ago suggested that Darwinian evolution was either not accepted or misunderstood.
      Contrary to initial expectations, social media seems to be facilitating the spread of anti-science, rather than science.

      • ljk March 4, 2019, 10:53

        Yes, with little drama he simply bent over a pool of water and pulled up a handful of what one initially thinks is silt or mud. Instead it is human ashes.

        Dr. Bronowski gives not the usual arguments one hears in regards to what the Nazis did to millions of Jews and other people who didn’t fit into their master plans for the human race. Instead he explains that this is what happens when people think they have absolute knowledge about anything. A lesson that many have yet to learn, if ever.



  • Michael C. Fidler February 28, 2019, 11:15

    What type of life could develop on Titan? This is a fascinating idea that is finally being assessed and many of our cherished ideals for how life developed are being tested. Laura pointed out an interesting aspect in a comment at the end of “Planet Formation: How Ocean Worlds Happen” that, “A recent work in Astrobiology states that ethane seas could be nine times more frequent than water seas in a recent paper “.
    The article ” Diving into Exoplanets: Are Water Seas the Most Common?” makes the astounding point that there are 16 time more exoplanets that have ethane, nitrogen and methane oceans then there are water worlds! Most of these planets are around M dwarfs but even G type dwarfs like our sun have more ethane and sulfuric acid ocean worlds then water worlds! Our prime examples is Titan for ethane and methane, so there are a lot more solvents out there for completely different types of life to develop. See the articles below for info on life on Titan, and how thinking out of box in an environment at -290 degrees fahrenheit makes for some mind stretching creative thinking!

    Titan’s oddly thick atmosphere may come from cooked organic compounds.

    Life on Titan.

    UIC’s mission: To model life on Saturn’s moon in the lab.

    • Michael C. Fidler March 2, 2019, 10:52

      Here is an earlier article from 2004 by WILLIAM BAINS that gives a very good and understandable introduction to the different liquids that form at distances from the sun.

      Many Chemistries Could Be Used to Build Living System.

      It has been widely suggested that life based around carbon, hydrogen, oxygen, and nitrogen is the only plausible biochemistry, and specifically that terrestrial biochemistry of nucleic acids, proteins, and sugars is likely to be “universal.” This is not an inevitable conclusion
      from our knowledge of chemistry. I argue that it is the nature of the liquid in which life evolves that defines the most appropriate chemistry. Fluids other than water could be abundant on a cosmic scale and could therefore be an environment in which non-terrestrial biochemistry could evolve. The chemical nature of these liquids could lead to quite different biochemistries, a hypothesis discussed in the context of the proposed “ammonochemistry” of the internal oceans of the Galilean satellites and a more speculative “silicon biochemistry” in liquid nitrogen. These different chemistries satisfy the thermodynamic drive for life through different mechanisms, and so will have different chemical signatures than terrestrial biochemistry. Key Words: Carbon, hydrogen, oxygen, and nitrogen life—Planetary liquid—

      An interesting point is a section on supercritical fluids at different temperatures and densities:

      Throughout this discussion, “fluid” has been equated with “liquid.” Very dense gases near their critical point [near-critical solvents or supercritical fluids (SCFs)] could also provide the dense, mobile environment in which macromolecular “solution” chemistry could occur. Supercritical carbon dioxide is well known for its industrial application as a solvent that can dissolve many materials insoluble in other, liquid solvent systems (Cygnarowicz and Seider, 1991), including a variety of complex biochemicals such as steroids, nitrogenous bases, and lipids.


  • wdk February 28, 2019, 17:40

    If a species is an example of a genetic book with a certain structure, then it is possible that the content has certain rules which include limits to the content such as paragraphs and chapters.

    This is definitely not my field. But I have noticed such oddities as the fact that turtles having shells do not have teeth. And that bats have wings because their forelegs are adapted to be such. In other words they did not sprout additional appendages like angels.

    Then one day a decade or so back I noticed a short Science article about discovery of a fossil turtle with only half a shell and – guess what – a half a set of teeth.

    Looking from outside at the field of genetic coding and considering how much of DNA content must be about protein construction, I have to wonder why there should be such arbitrary limitations on physiology. Perhaps it is somehow inherent in the current prevailing code of 4 bases in 3 unit codons arranged in long sequences with other delimiters. No doubt the message with 8 bases, etc. would have larger variations just like license plates with added alpha numeric settings ( numbers and alphabet letters, additional spaces), but it’s not clear (to me) what the current limits are. Human genetic code content, I am given to understand, is not necessarily more voluminous than that for various vegetables (say a turnip) or other flora or fauna. But as commenters indicate above, a more elaborate code would not necessarily make adaptive evolution any easier or likely to get started.

    However, it just might make it possible to pack more gear in your carrying case. So perhaps somewhere where the locals feel they had already reached a cul-de-sac, they might just have explored that route – so turtles could get both teeth and shells and mammals wouldn’t have to give up their hands to fly.

    • Alex Tolley March 1, 2019, 0:34

      It wouldn’t matter what bases, coding, or proteins are available, evolution doesn’t work the way you seem to imply. Typically existing structures are adapted by incremental changes. Thus wings in both birds and bats started at as forelimbs and then slowly changed to wings. The arguments have always been about what fitness advantage partial changes had. Evolutionists assume that there must have been fitness advantages, or Darwinian selection doesn’t work. Creationists assume no fitness advantage exists, reinforcing the need for design.

      There are constraints on phenotypes, but not of the “half shell, half teeth” variety of your example. Protein space is vast and barely explored by organisms, let alone the space of forms. However, novel properties of proteins using different amino acids might offer unexpected advantages in unexplored environmental niches, or advantages in existing ones. At this point we can only speculate.

      What I do expect is that commercial applications with engineered microorganisms will appear this century. These applications may be novel proteins using different amino acids as industrial or medical products. Or possibly novel compounds as a result of new metabolic pathways, or even simply faster production of existing products. Commerce will drive the applications.

      The 21st century was always predicted to be the “biological century”. The possibilities are expanding faster and wider than we could have anticipated just a few decades ago. Exciting times.

      • wdk March 1, 2019, 19:18

        Alex Tolley,
        Thanks for the examination of the question. Still, as you said:

        “Typically existing structures are adapted by incremental changes. Thus wings in both birds and bats started at as forelimbs and then slowly changed to wings. The arguments have always been about what fitness advantage partial changes had. Evolutionists assume that there must have been fitness advantages, or Darwinian selection doesn’t work…”

        Well, in posing the question, I conceded at the start that from arguments posted above, complex coding might not be well suited to evolutionary or adaptive processes. Rather it might be useful for
        complex function or processes in a present day rather than a
        distant future to which a species might adapt.

        Of course, we could consider the alternative route to wings without
        using the fore arms and hands. I would suppose that in theory a mammal could be given an additional set of limbs with feathers, but the process we have observed here on earth is quite different. In fact, it must have been even quite vexing to watch creatures continuously attempting to adapt to flight with their dinosaur or mammal fore arms and failing quite miserably… And in today’s society we would indeed describe human attempts at such as “natural selection at work”.

        But saying this, it is not a critique of the theory, but rather a wonder that it inpractice works in such instances at all. If we should actually be able to reach other worlds, adaptations required are likely to be enormous, not likely to be simply a trait selected and passed on from those that successfully took off their pressure helmets. How adaptation could be engendered poses both moral and technical difficulties to weigh against the frontiers possibly afforded.

  • hiro February 28, 2019, 19:44

    A speculated conjecture: we can’t obtain more than 2^2^2 letters DNA, anything more than that # will become highly unstable hence breaking down in finite time.

  • Coacervate Droplet March 1, 2019, 1:04

    Very interesting article, thanks!

    Oxidation of glucose releases ~17 kj/g according to my BOTEnvelop calculation.
    Compare that to the reduction of acetylene to ethane and it looks like ~11 kj/g.

    Enzymes bind to transition states here. I can see no reason why “cryozymes” would not do the same on Titan. Did Titan find a way of doing the acetylene reduction stepwise with useful metabolic tradeoffs along the way?

    Gosh it would be so nice to have a jar of mud from the bottom of Ontario Lacus in my dewar to play with.

  • Spaceman March 1, 2019, 11:31

    Interesting discussion, but when we talk about DNA/RNA as being the ‘optimal coding’ molecules, are we saying that they are the optimal coding molecules for life that developed on Earth or that they are the optimal coding molecules period? I am skeptical that DNA/RNA is the best possible coding combo in the Universe….perhaps the most we can say is that DNA/RNA may be optimal in some environments and not others and that this coding combo is just one of many in the biochemical parameter space. Further, coding combos will be selected from the parameter space depending on what material is locally available and what the physical conditions are locally? What about a nucleic acid with bases that are completely different from the A, T, C, and G we have in our DNA and/or a completely different chemical backbone? I can imagine an extraterrestrial civilization concluding that their coding molecule is the optimal one and ours as an aberration. Also, Alex brings up an interesting point: if we detect DNA/RNA based life elsewhere, then how could we be sure that it was not panspermia as opposed to a genuine second genesis?

  • Gary Wilson March 1, 2019, 14:02

    I think as with most theories, the idea of an 8 base genetic code will require thought and experimentation. We naturally think that the 4 base code is optimal because it is what has evolved here on earth, which remains our only sample space. If we do reach other planets with life (or moons) we can begin to learn what is actually possible. Let’s get going!

    • hiro March 1, 2019, 19:19

      We could replace the ordinary cells of bacteria or other simple organisms (maybe lab rats) with this new 8 base and then observe how these creatures behave. Constructing one from single-cell to multi-cell is much harder.

  • Michael C. Fidler March 2, 2019, 0:08

    This article was a real challenge for me and the basic concepts of organic chemistry and DNA/RNA is not something new to be, maybe some outreach is needed for the general populace so it is more palapale. Maybe this is one of the problems with biology and Darwinism, since most people understand the magic of the DNA crystal but seeing the diversity of life it is difficult to connect the two. I am seeing many articles coming up in just the last few days about hachimoji DNA but as in this CNN post:
    One unusual point that struck me is the 64 base comparing the Genetic Code of DNA to Binary Code and the very ancient Toaist I Ching use of 64.

    • Michael C. Fidler March 4, 2019, 15:59

      Born rule, wave probabilities – yin/yang and the orgin of life? What of Jennisons phase locked cavities and quantum entanglement and universal electrons. Cold Worlds where life is super conductive with planet waves. What of life’s code being transmitted to planets via quantum entanglements to increase the probabilities at the quantum level – The Life Rays!

  • MadRocketSci March 3, 2019, 2:06

    I’m surprised at some of the comments.

    1. Good work! This is very interesting.

    2. I don’t know why the default assumption is that DNA is the only way to store information in a cell. Don’t we have at least one other molecule in some primitive microorganisms (a kind of roll-up protein instead of the DNA double helix) that produces proteins? DNA is what we have because it was the first thing that happened to work well, and took off exponentially. We see this pattern a lot in nature – the first thing past the post gets copied ad infinitum and takes over the world. That certainly doesn’t mean it’s the *only* thing that can store and replicate information. One of the reasons why it would be exceedingly interesting to find alien life – even bacteria-like-things probably have tons of heretofore unimagined chemical innovations.