PDA

View Full Version : Would an autocatalyst dominate its environment? Why? Why not?



Colin Robinson
2012-Dec-18, 11:12 PM
from the thread "Metabolism First Strikes Back"


Scientists like Kauffman might alter their views upon pondering some more on what it was that 'brought him to the dance' .. and yet, he still ends up coming up with the shakier, more philosophically oriented view of 'inevitability'.

Autocatalysis might still play a role … but why should we think that such a factor necessarily dominates the outcome?

Why an organic autocatalyst would (or wouldn't) dominate its environment...

By an organic autocatalyst, I mean any agent (whether a single organic molecule, or a simple set of molecules, or a living organism) which facilitates a chemical reaction (or set of reactions) which then produces more of the agent. The autocatalytic agent is not necessarily the sole product of the reaction(s), but it has to be one of the products (otherwise it doesn't qualify as an autocatalyst).

I'm not talking about an environment at chemical equibrium, I'm talking about a non-equilibrium environment, like Titan today, or Earth in the Hadean epoch. An environment where there are potential reactants (on Titan, acetylene with hydrogen; on Hadean Earth carbon dioxide with hydrogen) but their reaction will not proceed without a catalyst to facilitate it.

One big reason why an autocatalyst would dominate its environment: diffusion. (Major theme of Darryl Strobel's study about hydrogen on Titan.) Diffusion means that randomly moving reactants tend to migrate from area of high concentration to areas of low concentration. Even if the autocatalyst only operates in one little region, it will lower the concentration of reactants in its own region, and therefore reactants from other regions (of higher concentration) will tend to diffuse towards the catalyst.

It is like being a predator in a jungle where herds of prey animals have a suicidal urge to move towards you. Which would make it rather easy to dominate the jungle...

So where's the catch? Why wouldn't an autocatalyst dominate its environment?

Well, it might eventually run out of reactants... but on Titan right now, that wouldn't be an immediate worry, since photochemistry in the upper atmosphere is continually producing acetylene and hydrogen out of methane.

Or... the autocatalyst we are considering might become a reactant itself... To be operating as an autocatalyst at all, it would need to be comparatively stable (metastable), but no organic molecule is immune from chemical change...

What if other autocatalysts emerge which can facilitate reactions which consume the first autocatalyst?

TooMany
2012-Dec-18, 11:54 PM
from the thread "Metabolism First Strikes Back"



Why an organic autocatalyst would (or wouldn't) dominate its environment...

By an organic autocatalyst, I mean any agent (whether a single organic molecule, or a simple set of molecules, or a living organism) which facilitates a chemical reaction (or set of reactions) which then produces more of the agent. The autocatalytic agent is not necessarily the sole product of the reaction(s), but it has to be one of the products (otherwise it doesn't qualify as an autocatalyst).

I'm not talking about an environment at chemical equibrium, I'm talking about a non-equilibrium environment, like Titan today, or Earth in the Hadean epoch. An environment where there are potential reactants (on Titan, acetylene with hydrogen; on Hadean Earth carbon dioxide with hydrogen) but their reaction will not proceed without a catalyst to facilitate it.

One big reason why an autocatalyst would dominate its environment: diffusion. (Major theme of Darryl Strobel's study about hydrogen on Titan.) Diffusion means that randomly moving reactants tend to migrate from area of high concentration to areas of low concentration. Even if the autocatalyst only operates in one little region, it will lower the concentration of reactants in its own region, and therefore reactants from other regions (of higher concentration) will tend to diffuse towards the catalyst.

It is like being a predator in a jungle where herds of prey animals have a suicidal urge to move towards you. Which would make it rather easy to dominate the jungle...

So where's the catch? Why wouldn't an autocatalyst dominate its environment?

Well, it might eventually run out of reactants... but on Titan right now, that wouldn't be an immediate worry, since photochemistry in the upper atmosphere is continually producing acetylene and hydrogen out of methane.

Or... the autocatalyst we are considering might become a reactant itself... To be operating as an autocatalyst at all, it would need to be comparatively stable (metastable), but no organic molecule is immune from chemical change...

What if other autocatalysts emerge which can facilitate reactions which consume the first autocatalyst?

That chemical cycles are sustainable is demonstrated by the oxygen/CO2 cycle on Earth persisting for 100's of millions of years and powered by the Sun, but the actors involved are already very complex chemical beasts.

I like the idea of an "autocatalyst" (reminds me of the nanons) and trying to relate such processes to the eventually formation as life as we know it. I don't think there will be a clear line where you can say, life emerged exactly when X happened. I.e. it seems likely that life emerged gradually though a progressive evolution, just like the part of the history of life where substantial direct evidence is available.

Selfsim
2012-Dec-19, 10:26 AM
Hmm ... autocatalytic reactions possess the key feature of being nonlinear. Certain terms in the equations describing the rates of these reactions can lead to fixed or multiple stable states of the system. Phase transitions occur as the system moves from one state to another. Simple autocatalytic reactions can even oscillate between these different phase states, depending on the concentrations of a reactant, relative to another. Spatial and temporal order are characteristics of these types of reactions. My favourite example of these types of reactions is the Belousov-Zhabotinski reaction (mentioned previously), others are; the Briggs-Rauscher and the iodine clock reactions.
(Reference here). (http://en.wikipedia.org/wiki/Autocatalytic_reaction#Autocatalytic_reactions)

Wherever there is feedback ... (like where the autocatalytic agent itself, is one of the products, and its concentration varies), there is non-linearity, which causes the mentioned dynamically varying 'non-equilibrium' state.

So when I asked: "why should we think that such a factor necessarily dominates the outcome?", perhaps more appropriately, the question might be better expressed as: "how sensitive might these reactions be to external (other) environmental influences, such that the spatial (or temporal) order which they normally produce, becomes unpredictable?" (For example, I see that in biology, glycolisis can oscillate (http://en.wikipedia.org/wiki/Autocatalytic_reaction#Biological_example) between an active and inactive form. In other words, 'the autocatalytic reactions can turn the process off and on'.

So, what is the envelope of applicability in the hunt for abiogenesis? How determinable, or more specifically, how predictable is the outcome?

Paul Wally
2012-Dec-19, 02:35 PM
One big reason why an autocatalyst would dominate its environment: diffusion. (Major theme of Darryl Strobel's study about hydrogen on Titan.) Diffusion means that randomly moving reactants tend to migrate from area of high concentration to areas of low concentration. Even if the autocatalyst only operates in one little region, it will lower the concentration of reactants in its own region, and therefore reactants from other regions (of higher concentration) will tend to diffuse towards the catalyst.

On the question of why an autocatalyst wouldn't dominate its environment, we will have to answer the question of what happens to the autocatalyst when it doesn't have access to reactants. Does it just hang around indefinitely until the reactants eventually arrive or does it disintegrate? One could say that the autocatalyst wouldn't dominate its environment if the the rate of disintegration is greater than or equal to the the rate of production. Rate of production will of course be limited by the diffusion rate of the reactants.



It is like being a predator in a jungle where herds of prey animals have a suicidal urge to move towards you. Which would make it rather easy to dominate the jungle...

I think the predator-prey analogy is a good one. It got me thinking of the Lotka-Volterra predator-prey equations (http://en.wikipedia.org/wiki/Lotka–Volterra_equation). Perhaps the same equations, or slight modifications thereof, could be used to model the dynamics between autocatalyst and reactants.



So when I asked: "why should we think that such a factor necessarily dominates the outcome?", perhaps more appropriately, the question might be better expressed as: "how sensitive might these reactions be to external (other) environmental influences, such that the spatial (or temporal) order which they normally produce, becomes unpredictable?" (For example, I see that in biology, glycolisis can oscillate (http://en.wikipedia.org/wiki/Autocatalytic_reaction#Biological_example) between an active and inactive form. In other words, 'the autocatalytic reactions can turn the process off and on'.

Regarding your question of why we should think that an autocatalytic reaction should necessarily dominate the outcome, I would say it doesn't necessarily.
However, we should look at this from a more global perspective rather than considering single reactions in isolation, like the ones you're mentioning. So, from a global perspective, there is a large variety of different reactions, some would dominate while others won't. Only the ones who dominate gets selected, so natural selection is of relevance here. The unpredictability or randomness that you're mentioning actually plays an important role in self-organizing complexity because it introduces variety into the global system. The environment then selects from this random variety the ones that work.
The following quote from Wikipedia captures the idea of self-organization (http://en.wikipedia.org/wiki/Self-organization) from randomness:


Wikipedia: Self-organization

The principle of "order from noise" was formulated by the cybernetician Heinz von Foerster in 1960.[5] It notes that self-organization is facilitated by random perturbations ("noise") that let the system explore a variety of states in its state space. This increases the chance that the system would arrive into the basin of a "strong" or "deep" attractor, from which it would then quickly enter the attractor itself. A similar principle was formulated by the thermodynamicist Ilya Prigogine as "order through fluctuations"[6] or "order out of chaos".[7]



So, what is the envelope of applicability in the hunt for abiogenesis? How determinable, or more specifically, how predictable is the outcome?

We may not be able to predict specific outcomes of specific reactions, but on a global level we could still make the general prediction that the environment will select the dominant reactions, whatever they might be.

TooMany
2012-Dec-19, 04:25 PM
We may not be able to predict specific outcomes of specific reactions, but on a global level we could still make the general prediction that the environment will select the dominant reactions, whatever they might be.

I.e. evolution does not begin after what we consider to be life forms, evolution is also responsible for the formation of life. What works, works.

Selfsim
2012-Dec-19, 09:04 PM
Regarding your question of why we should think that an autocatalytic reaction should necessarily dominate the outcome, I would say it doesn't necessarily. I agree .. it looks a bit like a 'moving', perhaps even, 'randomly moving', target (over certain scales)(??)


However, we should look at this from a more global perspective rather than considering single reactions in isolation, like the ones you're mentioning. So, from a global perspective, there is a large variety of different reactions, some would dominate while others won't.It would depend on the temporal and spatial scales of the observation, I think. It seems the potential exists to find an observational scale which might suit one's own chosen perspective, what's more(??) :(
What is needed, is data which confines the observation scales … (and we don't have that data).


Only the ones who dominate gets selected, so natural selection is of relevance here. The unpredictability or randomness that you're mentioning actually plays an important role in self-organizing complexity because it introduces variety into the global system. The environment then selects from this random variety the ones that work. The randomness aspects could easily be compounded by other external (to the system) influences … (which is where I was coming from, originally). If these impact the initial (startup) conditions of the reaction, then the outcomes between two separate trials could potentially be completely different, say at the macroscopic level (or even at the level of the complex molecules so formed). There are a myriad of ways proteins can wrap themselves (for eg) .. all resulting in different outcomes.
So, your suggestion is that 'natural selection' might sort all these out, eh? (Which is reasonable, I think. :) ).
I think current evidence shows the 'modern' molecule topology itself, is a major player in this as well. If we're dealing with a three dimensional molecule, then the numbers of variables would be enormous.

If environmental natural selection is the key player in sorting all this out also, then potentially, even minor differences between two separate, but similar, environments, could make all the difference (in terms of getting the same long term result from successive trials). (This is what I mean when I say 'what brought Kauffman to the dance').

How does one eliminate the scenario of the 'non-viable' outcome (ie: in terms of progressing towards viable life)? Is this 'equiprobable', when compared with the 'viable' outcome? (Surely this would be testable?)

What if the reaction stops, and fails to reach the equivalent of some sort of 'critical mass'?


We may not be able to predict specific outcomes of specific reactions, but on a global level we could still make the general prediction that the environment will select the dominant reactions, whatever they might be.But the 'specific outcome' at the molecular level, is what we're trying to predict isn't it?

Colin Robinson
2012-Dec-20, 01:36 AM
We may not be able to predict specific outcomes of specific reactions, but on a global level we could still make the general prediction that the environment will select the dominant reactions, whatever they might be.
But the 'specific outcome' at the molecular level, is what we're trying to predict isn't it?

Some aspects of a chemical system may more more predictable than others.

The more predictable side has to do with the basic reactants and waste products. I mean simple molecules such as carbon dioxide, water, oxygen, hydrogen, methane.

A system of chemical reactions is like a machine, in the sense that it needs an energy source in order to keep moving. (I'm not talking about states of dormancy here. Of course a car does not have to burn petrol when it is sitting in a garage...)

In a chemical environment which is oxidizing rather than reducing (i.e. rich in oxygen rather than hydrogen), methane can be used a source of energy by combining it with oxygen, eventually (thru a series of chemical steps) producing water and carbon dioxide.

In a chemical environment which is reducing rather than oxidizing, methane cannot be used a source of energy, but carbon dioxide can be used as an energy source by combining it with hydrogen.

The less predictable side has to do with comparatively complex catalysts (e.g. polypeptide enzymes). According to Stuart Kauffman, for each reactant (ligand), there are very many possible catalysts that can bind to it and facilitate its reaction with another ligand.

Kauffman's conclusion is that while it is highly likely that autocatalytic sets will emerge from a chemical system of sufficient diversity and activity, it is not possible to predict in advance which catalytic molecules these sets will contain.

publiusr
2012-Dec-21, 10:31 PM
This link may be of use
http://www.spacedaily.com/reports/Origin_of_life_emerged_from_cell_membrane_bioenerg etics_999.html

Cougar
2012-Dec-22, 01:40 AM
Kauffman's conclusion is that while it is highly likely that autocatalytic sets will emerge from a chemical system of sufficient diversity and activity....

Correct, it's largely the sufficient diversity of molecules and reactants. And it's not a matter of the autocatalysts 'dominating' the planet, rather them beginning to utilize a portion of their surroundings....


What if the reaction stops, and fails to reach the equivalent of some sort of 'critical mass'?

Must not have had enough diversity. :) Or, yes, any number of additional things. But Kauffman doesn't just handwave. He does the math. And his book At Home in the Universe does a fair job with the analogies and descriptions. It blew me away when it came out. :rolleyes-default:

Colin Robinson
2012-Dec-23, 08:04 PM
This link may be of use
http://www.spacedaily.com/reports/Origin_of_life_emerged_from_cell_membrane_bioenerg etics_999.html

Many thanks for this helpful summary of a very recent paper about the role of geo-chemistry in the origin of life.

Here is a link to an online pdf of the full paper (which you don't have to pay to read)

Nick Lane and William F. Martin, The Origin of Membrane Bioenergetics, published in the journal Cell (http://download.cell.com/pdf/PIIS0092867412014389.pdf)