Natural Science 301 Guidelines

This is an open forum area for all members for discussions on all issues of science and origins. This area will and does get volatile at times, but we ask that it be kept to a dull roar, and moderators will intervene to keep the peace if necessary. This means obvious trolling and flaming that becomes a problem will be dealt with, and you might find yourself in the doghouse.

As usual, Tweb rules apply. If you haven't read them now would be a good time.

Forum Rules: Here
See more
See less

Experimental evolution. Testing Darwin regarding the origin of multicellularity.

  • Filter
  • Time
  • Show
Clear All
new posts

  • Experimental evolution. Testing Darwin regarding the origin of multicellularity.

    Hi All,

    While finishing off my last essay, an interesting bit of research dealing with experimental evolution was published in Nature. 

    As sometimes happens, it caught my eye, and so I began reading with a hope to understanding it tolerably well, enough so that I could write up a series of posts explaining the experiment.


I think I do understand it and before I begin to describe the gory guts, let me introduce it and mention what I found so interesting about the research.

    The article is:-

    Life cycles, fitness decoupling and the evolution of multicellularity

    Its abstract and conclusion read as follows:-


    Originally posted by link above
    Cooperation is central to the emergence of multicellular life; however, the means by which the earliest collectives (groups of cells) maintained integrity in the face of destructive cheating types is unclear. One idea posits cheats as a primitive germ line in a life cycle that facilitates collective reproduction. Here we describe an experiment in which simple cooperating lineages of bacteria were propagated under a selective regime that rewarded collective-level persistence. Collectives reproduced via life cycles that either embraced, or purged, cheating types. When embraced, the life cycle alternated between phenotypic states. Selection fostered inception of a developmental switch that underpinned the emergence of collectives whose fitness, during the course of evolution, became decoupled from the fitness of constituent cells. Such development and decoupling did not occur when groups reproduced via a cheat-purging regime. Our findings capture key events in the evolution of Darwinian individuality during the transition from single cells to multicellularity.

    [big snip]

    Multicellular organisms are descendants of once free-living cells1, 3, 4. By virtue of their capacity for differential reproduction, ancestral free-living cells were units of selection33. During the transition to multicellularity, collectives of cells emerged that came to participate in Darwinian processes in their own right2, 5, 11, 19. The essential ingredient was a means of collective reproduction5, 11. This most seminal of Darwinian properties emerges afresh at each transition and requires explanation44. Here we have shown that cheating cells—those types seemingly most detrimental to the persistence of newly formed cooperative entities—can function as a germ line within a life cycle that facilitates the reproduction of collectives. Moreover, the two-phase life cycle presents selection with an altogether new kind of biological entity: each state becomes a different attribute of a single organism whose evolution is unified through a developmental programme45. When reproduction of collectives is via fragmentation (a single-phase life cycle), the traits that yield success at the higher level are largely those that determine success of single cells. This offers limited opportunity for the emergence of new kinds of biological individuality because properties of higher and lower levels remain aligned19, 38.

    Direct observation of early stages in an evolutionary transition requires that issues surrounding levels of selection be considered2, 5, 19. This necessarily leads to territory in which a range of perspectives is possible (see Supplementary Discussion). Our experimental design incorporates an ecology that is explicitly multi-level: both individual cells (that reproduce once every hour), and individual lineages (that reproduce once every 9*days) can be units of selection; however, selection operates on cells and lineages over different timescales. While selection on individual cells favours short-term success, short-term success is unlikely to facilitate persistence of lineages. Indeed, persistence requires more than simply switching between phenotypes: it involves a developmental programme that underpins expression of a collective phase in which a soma-like body is constructed from germ-like cells. Cells of the body must simultaneously play an ecological role (maintaining the body near oxygen via a robust mat phenotype) while producing the seeds of the next generation of bodies (the germ-like cells). Given sufficient variation among lineages, then selection over the longer timescale stands to conquer the short-term interests of individual cells. This appears to have happened in our CE regime with decoupling of fitness between levels supporting the view that selection has begun the process of transitioning to the higher (collective) level—with the lower level beginning to function for the good of the collective.

    The research has to do with using a bacterium named Pseudomonas fluorescens as a means for exploring the origins of multicellularity. This organism is well established as a test bed for these kinds of experiments because what it can do is go from single celled organisms to clusters of cooperating organisms (called “mats”) fairly readily. But to go from single cells to mats of cells, the individual cells have to give up their individualistic identity and cooperate. By doing this, they reap benefits not open to the cells if they remain as individuals. As individuals, the cells remain immersed in a broth that can be toxic, thanks to the accumulation of waste products which deprive the cells of oxygen. As cooperators forming mats though, things change because the mats float, inhabiting the fluid/air interface and thus allowing the cells in the mat to gain access to oxygen.


Here is where it really starts to get interesting however.

    The cells get access to oxygen by forming mats. In doing so, they have to expend food and energy producing a glue that binds them together. So, if “cheats” can learn to live with the cooperators, then the cheats also access the oxygen but don’t have to give anything to the group to allow it to continue.

    That is, cheats in the mat gain the benefits of being in the mat, without having to expend food and energy that allow the other cells to participate in mat formation.


As a result, the cheats outcompete the cooperators and soon overwhelm them and the mat dies. It disintegrates.

    So mats only last for a while before dying.

    As a result, experiments on the origins of multicellularity have focused on systems which purge the cheats from the habitat. It seemed an intuitive thing to do.

    However, the results were of limited success.

    Now here is where it begins to get even more interesting. The researchers wondered if there were ways in which the cheats may actually help the mats, not by preventing their deaths, but rather by acting as a kind of sex cell and on the death of the mat, the cheats being able to act as the nucleus for the next generation of mat, simply by virtue of the fact that a cheat cell also has the genome of the cooperating cell which it carries on into the future as the beginning of the next generation.

    The experiment was able to show that there is a condition in which this is what happens. The cooperating mat cells behave as our somites - the non sex cells (skin, liver, hair, brain, eye cells) which ultimately die. The cheats in the mat behave as our sex cells, propagating the genome of the mat into the next generation, forming the nucleus of the new mat.


Like the cheats, our sex cells really don’t contribute to the daily continuation of the body. They kind of bludge off it, using the food and energy produced by the rest of the body. But where they come into their own is on the death of the organism. When the organism dies, it’s the sex cells that get the new generation underway. Thus it was with the cheats.

    Their research was not so much about the origins of sex, but rather it was about the origins of multicellularity.

    The other fascinating thing was this. What is good for the individual cells in the mat, is not necessarily good for the mat itself. That is, if selection acts on the mat in a positive way, then it may at the same time, act on the individual cells in a negative manner, and vice versa. And certainly selection acts on cells over the life time of a cell, with is very much less than selection acting on a mat over the life time of a mat. Mats have a life time measured in days. Cells have a life time measured in hours.

    Yet in this experiment, the mat and the cells became decoupled, such that selection at one level did not overly affect selection at the other level.

    Hopefully I can describe the experiment to you, and give you an idea about what they found out, and what the genetics were that allowed this system of cooperating mat cells + cheat cells was all about, and how the cheats played a crucial role in bringing about the next generation of mat.


This set of essays may be slow in coming. I don’t know. But it is that hectic time of the year and besides, I’m busy digging up the back yard, remodeling it. Nevertheless, reading and learning some interesting idea is always a lot more fun than hot sweaty work in dust under a blazing sun or in mud and slush when a storm comes.  

    To be continued ....

    Last edited by rwatts; 11-23-2014, 01:14 AM.

  • #2
    Before I begin to describe the paper in earnest, I need to point out that was is on offer here is the testing of a hypothetical scenario concerning the origin of multicellularity. In their Supplementary Information, the authors offer an example of one such scenario and while the example is a good one, it left me puzzled because of one important aspect which I think was wrong. The error is certainly not fatal to the kind of scenario they offer. But it seemed an odd thing to write, (I think). I shall deal with this towards the end of the essays.

    The Introduction

    The fossil record shows that before multicellular complex life existed, the earth was inhabited by single cells, small clusters of cells, and cells living in symbiotic relationships making up, for example, stromatolites. Essentially the earth was a place for bacteria to live out their lives (and it still is). Hence for single cells to join up to become multicellular organisms, a degree of cooperation is required. That is, cooperation is a central feature in the evolution of multicellularity.

    Under laboratory conditions, it’s easy these days to get undifferentiated, individual cells to evolve into groups of cooperating cells at a basic level. However, thanks to the evolution of cheats from within the group, the groups don’t last long. The problem is that to be within a group, a cell has to spend resources making something that keeps the group together. There will be a benefit in being a member of the group, until a cheat comes along. The thing about cheats is that they exist within the group, and gain the benefits from being in the group, but they contribute nothing do the group, and so have energy to spare allowing cheats out reproduce the group members. And so they do, to the point that the group collapses.

    Because of this, experimenters have looked to scenarios in which groups evolved in such a way that cheats were always suppressed by one means or another. But this had very limited success.

    So the research group of Hammerschmidt, Rose, Kerr and Rainey (hereafter referred to as HRKR or H2RK :)) wondered if the cheats might facilitate the origins of multicellularity by playing a critical role in a multicellular life-cycle, providing the generation of cheats were to be controlled in some way.

    The bacterium Pseudomonas fluorescens is a major test vehicle for multicellularity research because it readily forms groups of cells when populations of individual cells are grown in a spatially structured microcosm. This kind of a medium is one which has structures which are spaced apart, at the surface of the growing medium, and the importance of this will be seen very shortly. An early paper regarding the use of P. fluorescens is here:-

    Evolution of cooperation and conflict in experimental bacterial populations

    As the authors of that paper note:-


    Originally posted by link just above (bolding mine)
    Defecting genotypes evolved in populations founded by the cooperating type and were fitter in the presence of this type than in its absence. In the short term, defectors sabotaged the viability of the group; but these findings nevertheless show that transitions to higher orders of complexity are readily achievable, provide insights into the selective conditions, and facilitate experimental analysis of the evolution of individuality.

    Anyway, what happens is that normally, P. fluorescens cells are smooth (SM) in outline, but thanks to several different types of spontaneous mutation, cells can group together and form what are called “wrinkly spreader” (WS) mats. These various spontaneous mutations cause the cells to over-produce a cell-cell binding glue and so when a cell divides the glue keeps mother and daughter joined. The daughter cell does not break off to become an individual cell. Hence, in the experimental broth, small growing mats appear, and they float and become anchored to the structures at the surface. What is important here is that because they float, the growing mats find themselves at the air/broth interface and thus accessing the oxygen in the air, something the individual cells in the broth cannot do so readily. The SM cells increasingly find themselves in a steadily growing anaerobic environment as the broth begins to run short of oxygen.

    But there is a catch for the cooperating cells in the mat. The production of the glue is expensive for each cell. Cells have to spend resources producing it. Sooner or later, daughter cells evolve thanks to spontaneous mutations, daughter cells which lack the ability to make the glue. They revert back to SM but, being members of the mat, accrue the benefits of being there (accessing the oxygen), while providing nothing in return (the production of the glue). Hence they have spare energy to out reproduce the mat cells and so they do, building up to the point that they overwhelm the mat and the mat disintegrates. It effectively dies. As the authors explain:-


    Originally posted by link at OP
    ... The net effect is a cellular mat that colonizes the air–broth interface. Although glue production is costly to individual cells, the trait spreads29 because the group of mat-forming cells reaps an advantage (access to oxygen) that is denied to individual cells9.

    The life span of WS mats is brief: selection acting on individual cells favours mutant types that cheat. Cheating cells are phenotypically SM and no longer produce adhesive glues9; nonetheless they take advantage of the benefit that accrues from being part of the mat. In the absence of any mechanism of cheater repression, cheats prosper—ultimately weakening the fabric of the mat to the point where it collapses9.
    So cheats mess things up. However, as the authors explain, the “the tension between cooperating and cheating cells could fuel evolution”.

    How so?

    They ask the reader to consider a newly formed mat. The mat has no means of reproduction. Like the soma (non sex cells) of our bodies, the mat is at an evolutionary dead end. Our bodies die, the soma die. The mat dies, the WS cells die. However, the emergence of cheats in mats is always guaranteed and while they ultimately destroy the mat, because they derive from mat cells, they carry the genotype of those cells. If they can regenerate the mat then cycling between the WS groups of cells and SM cells brings about a primitive life cycle. The authors refer the reader to Figure 1a, left panel (The figures will be explained a bit later).

    With the establishment of a primitive life cycle then there is the possibility that the mats themselves may begin to participate “as units of selection in the process of Darwinian evolution”.


To be continued ....


    • #3
      Still here soaking it up!
      Glendower: I can call spirits from the vasty deep.
      Hotspur: Why, so can I, or so can any man;
      But will they come when you do call for them? Shakespeare’s Henry IV, Part 1, Act III:

      go with the flow the river knows . . .


      I do not know, therefore everything is in pencil.


      • #4
        Embracing cheats

        The initial experimental set up to test the idea that SM (smooth) cells may act as primitive germ cells for new generations of mats of WS (wrinkly spreader) cells, was as shown in Figure 1a. This is called the “cheat embracing (CE)” version of the experiment. On the right is the experimental setup where cheats are purged (CP) from the system and this will be discussed later.

        The researchers ran 120 different lines of this experiment where each line was split into two phases. In the first phase, a single WS cell was incubated in a broth for six days, during which time it had to form a mat in which some cheating SM cells arose via spontaneous mutation. The mat had to remain viable until the end of the six days.

        In the second phase, some SM cells were removed and incubated in their own broth. At the end of this second phase, some of the SM cells had to have mutated back to WS type.

The cycle was repeated when a single WS cell was taken to begin a new phase 1.

        Because the cycling between WS and SM relied on spontaneous mutation the researchers were a bit surprised that the system manage to pass through six cycles before all 120 lines had died out. The main cause of death was insufficient production of SM cells. These really were vital to the regeneration of new mats. Yet the fact that up to six cycles did occur before the final line had died off suggested to the researchers some “capacity for innovation.” As they wrote:-

        Originally posted by the paper
        Such a capacity might, under different circumstances, provide opportunity for evolutionary refinement to the point where cycling through phases could come under developmental control.
        Because most non neutral mutations are harmful, sooner or later a life cycle would be disrupted. So life cycle persistence would depend on the ability of successful lines to split such that a viable mutation would go to one line and a less viable or harmful mutation would go to another line, with the eventual elimination of the lines carrying the harmful mutation. In this way “life-cycle-enhancing mutations which are beneficial over the longer time scale of the life-cycle” might be able to “outrun life-cycle disrupting mutations”.

        Essentially, what was missing from their experiment was a competitive environment with selection acting on the mats.

        So, they rearranged their experiment to allow for mat selection to operate. Things changed dramatically.

        To be continued ....


        • #5
          In my last post, I described how the authors, HRKR, ran their experiment using 120 lines, but with no selection, and after six life cycles, every line had died out. Yet the fact that they had managed to go to six life cycles on some lines, suggested that the system had the capacity for innovation. So the researchers redesigned the experiment to introduce selection.

          In the new setup, the 120 flasks, or microcosms were grouped into 15 “replicate populations” of 8 lines each. The same life cycling was done as described in the previous post.

However, selection was introduced as follows. When a line died off, that is, it failed to complete a phase of the life cycle, the authors reasoned that this provided an opportunity for a viable lines to “export their success” to the new microcosm, the one just vacated. So in the next phase of the life cycle, they replaced the dead line with one from the remaining and still viable lines. But they did not replace it with the best of the remaining lines. Rather they selected the replacement randomly. This can be seen in Figure 2a.

          The main causes of extinction were failure to produce either SM (germ cells) or WS (new mat cells). Extinction occurred at a very high rate, approximately 5 replacements per generation per replicate. That is, in Figure 2, for example, to go through both phases, approximately five extinctions occurred.

          But, after ten life-cycle generations, each replicate population housed viable lines. (Previously, every line had died off after 6 life-cycle generations). As the authors note:-


          Originally posted by paper
          Selection on lineage viability - and concomitantly fecundity - was thus central to persistence.

          The reader might protest at this stage, that it was the experimenters doing the selection. However, it needs to be pointed out that they were testing a hypothetical scenario and this will be discussed further towards the end. And the whole point of random selection was to remove experimenter bias. The thing is, the death of a particular line does in fact open up that space for a viable line to occupy.

          To analyse the experiment and determine the course of evolution in the context of variation+selection, the scientists needed to measure the fitness of the evolved lines relative to the fitness of the ancestral types. Rather than competing evolved lines and cells against ancestral lines or cells, they tested evolved and ancestral lines against a standard, a “lacZ marked reference strain” of the bacterium. So the evolved was competed against this reference. The ancestral was competed against this reference. Then the relative fitness of evolved versus ancestral was compared.

          How this was actually done is described in Extended data, Figure 1a, and the methods section (which is behind the paywall). Note that Figure 1, when describing how representative types of SM or WS cells were obtained, puzzlingly mentions that this gave “14 such types, one representing each replicate”. The reason for the 14 types as opposed to 15 looks, on reading the methods section, to have been because one sample failed to thaw when taken from the freezer.

          Fitness of mats was defined as the ability of a mat to leave mat offspring relative to the ability of the marked competitor. This is seen in Figure 1b in the link above. Again be careful. Looking at the figure, line fitness (mat fitness, I think) in the example is printed as a ratio, less than one, whereas in the graph to be discussed soon, it is shown as a figure greater than 1. It depends on how one looks at it. If one ratios the numbers relative to one, the fraction will result. If one ratios the numbers relative to each other, then values greater than one can result. These are just different ways of saying the same thing.


Fitness of cells was established by estimating the total number of cells (WS and SM) at the end of phase 1. Other measures were also determined and these will be described later.

          Now it gets interesting. Figure 2b shows a small graph. At the top left is a plot of mat fitness or fitness of the evolved lines. To the bottom right is cell fitness.

          What they noted was that mat fitness had improved significantly and this was in the situation in which cheats were integral to mat life cycle. However, individual cell fitness had decreased significantly. In both cases, this was relative to the ancestral types. The ancestral plot is the “ANC” line across the middle, expressed as 1.0. Mat fitness plots at between 1.5 and 2.0 whereas cell fitness plots at less than 1.

          Because of this the authors argue that they have detected a fitness decoupling between the mats and the individual cells. Mat fitness did not depend on cell fitness. What was good for the mat was not necessarily good for the individual cell. They write that:-

          Originally posted by paper
          This is consistent with theoretical predictions that during major evolutionary transitions selection shifts from the lower (cell) to the higher (collective) level2, 38. With such a shift arises a new kind of biological individual whose emergence is likely to curtail the independent evolution of lower-level entities5.

          But what were the phenotypic traits that underpinned lineage improvement?

          To be continued ....

          Last edited by rwatts; 11-29-2014, 01:28 PM.


          • #6
            Traits inherent in the individual cells themselves, must be able to explain the success or otherwise of each line to be able to go through a complete life cycle. In this case success was the ability to be able to produce the required cell types by the end of each phase, thereby allowing a line to continue. This might seem at odds with the idea that the mats themselves (clusters of cells) are the ultimate focal point. Nevertheless it’s the same with organisms such as ourselves. What happens in our individual cells defines so much of how well we succeed or otherwise at keeping alive, keeping well, keeping fit and thereby helping to bring about the next generation.

            The researchers suspected that selection for some kind of primitive developmental program may have occurred in those lines that were successful.

            This was in part because of how the cells reacted to the length of the two phases. They seemed to be tuned to these phases. To show this, the time for each phase was doubled and it became clear that the ability of the cells in each phase, to produced the required cell type (SM in phase 1, WS in phase 2) reached its maximum at six days in phase 1 and three days in phase 2. This ability was consistent for all lines. Following the reaching those maxima, performance more or less leveled off, and this can be seen in Figure 3a, most notably with respect to phase 1, where the black dots represent the derived or evolved lines and the grey dots represent the ancestral lines.

            They noted that while some of the ancestral lines had the ability to rapidly produce the required cell type, in the derived lines this ability had spread markedly.

            As opposed to selection for some kind of internal development program, it could be that increased cell density, rate of cell division or improved cell growth allowed for this switching. After all, the more cells there are, the greater the chance of some random mutation occurring that causes some cells to swap to the required type. 

Yet this was found not to be the case. When it came to issues of cell growth rates etc., there was not a big difference between ancestral and derived. In fact, in some cases, ancestral cells out performed the evolved cells. Figure 3b at the link above shows this kind of thing. Those black and grey dots are hardly separated.

            Other statistical analyses on the traits of lines, and on the single cells also ruled out competitive performance of single cells as providing the improved survival ability of the evolved lines. Regression and correlation analyses showed that the capacity to switch from WS to SM predicts the fitness in the derived or evolved lineages. Line fitness was also found to be unrelated to the competitive performance of single cells.


So the required cells arising via chance mutations in each phase did not appear to be explaining what was happening. Rather, how the cells responded to the phase they were in, suggested that they were responding to the phase as if a kind of internal developmental program was operating, a switch that could flick to produce the right cell type for the phase the cell was in.

The question then was - what were the genetics of this developmental program that allowed the lines to produce the required cell types at roughly the right numbers, depending on the phase the cells were in. That is, if the cells were in phase 1, and growing lots of mat producing cells, what were the genetics that allowed a certain number of cheats to arise with near certainty, such that these cheats became the basis of primitive germ cells, taking the mat genome on into the future when the mat itself died (thanks the existence of the cheats)? And what were the genetics that allowed the cheat cells, when alone and reproducing, to bring on the existence of a number of mat cells that would form the basis of a new generation of mats?

            The hunt was on for the underpinning genetics.

            To be continued ....

            Last edited by rwatts; 12-03-2014, 01:53 PM.


            • #7
              With data suggestive of some kind of genetic switch tuned to the particular phase the cells found themselves in, the researchers went looking. They examined the genome of the most successful lineage to see what happened there.

              The fittest line had been given the number 17. It was sequenced at specific generations and additional life cycles were run and the associated genomes sequenced.

              If spontaneous mutation (as opposed to genetic switching) was driving the cell type changes then seven mutations should separate WS8 from the ancestral SM. This is because the cell type sequence would be:-

              SM1 -> WS2 -> SM3 -> WS4 -> SM5 -> WS6 -> SM7 -> WS8

              - with each switch between a type being caused by a mutation to a gene locus known to cause that particular cell type. The data backed this up. Mutations to 5 loci known to cause SM/WS switching were found.

              However, at generation 11, things were very different.

              Perhaps around 21 mutations could be expected to be seen but in fact 53 were found when compared to the ancestral genome. 

This elevated rate was determined to be due to a single amino acid mutation in mutS, a protein that fixes errors in DNA following duplication. It increases the fidelity of DNA replication by a factor of 100 to 1000.

              That is, the mutation had compromised DNA replication fidelity, allowing the mutation rate to dramatically rise.

              The next question was whether or not mutS played a direct role in switching, or whether or not it allowed a different mutation to be the switch and it had simply hitch-hiked with the real switch causing mutation.

              They tested this scenario with the number 17 line, by reverting some of the cells back to the wild type, (by repairing the mutS) to see what happened. The ability of those cells to switch was seriously impaired, indicating that mutS was directly involved in switching.


But how?

              They noted that various mutations had become fixed in line 17, but one seemed to be of more importance. This was a frameshift mutation in a tract of 7 guanine residues in a protein known as wspR. (A frameshift mutation is a mutation which causes the DNA reading machinery to begin reading in a different location, causing a change to the amino acids of the resulting protein). This frameshift mutation caused an additional guanine residue to join to the others in the SM cell type.

              wspR is a protein involved in a chemosensory pathway (the Wsp pathway) concerned with the production of the glue that binds cells together, depending on environmental cues. Thirty nine other genes have been identified as capable of causing WS cell types and thus the mats.

              This tract of guanine residues overlies the active site on the wspR protein and the researchers wondered whether or not this guanine residue addition was acting as the switch. That is, its presence or absence put the switch into the on or off state.

              The nature of this kind of mutation (lots of repeats) and its outcome looked very much as if the associated locus giving rise to the wspR protein was behaving as a “contingency locus”, a gene that can generate variation depending on environmental circumstances.

              So, following further life cycle tests and genome sequencing, they were able to demonstrate this switch in action. Take a look at Figure 7b

 On the left is the evolved version of line 17 and on the right is line 17 but reverted back to its wild/ancestral type. The wild type does not have the putative switch and as you can see every line goes extinct for one reason or another during the additional lifecycles. The evolved version with the mutS mutation however has only two extinctions, and the guanine expansion happens (except in a couple of cases) when the SM cell type exists and the contraction happens when the WS type exists.

              They write:-

              Originally posted by link in OP
              As shown in Extended Data Fig. 7b many of the transitions between WS and SM from generation 10 onward correlated with expansion (wspR OFF) or contraction (wspR ON) of the guanine tract. In a control experiment, performed with line-17 mutSWT, the slower and less reliable transition between phenotypic states did not, with a single exception, involve the tract of guanine residues.
              - and conclude that section with these observations:-


              Originally posted by link at OP
              The existence of a genetic switch in line-17 is highly advantageous to the collective: it strengthens heritability between recurrences; integrates both phases into essentially a single entity; and constitutes a critical first step in the emergence of differentiation4. Genome sequencing showed that the mutS-dependent switch arose in just a single lineage, but fixed in all eight populations of the replicate. Its evolution was reliant on earlier mutations in Wsp that preserved functionality of the pathway while ensuring constitutive activation of WspR25. It also depended on mutations elsewhere in the genome that exhausted alternate genetic routes to WS. Together, this set of prior mutations, in conjunction with mutS (A1489C), conferred special significance to the tract of guanine residues in wspR. While dependency on mutS might seem a dangerous liaison, the life cycle provides ample opportunity for purifying selection to maintain integrity of the SM type and refine the switch.

Note that last sentence. While messing the DNA repair mechanism might appear a dangerous thing to do, in this lineage it had the major impact on the formation of the genetic switch which allowed new generations of the mat to appear with near certainty, subsequent to the death of a mat. Given a primitive reproductive mechanism, the mats themselves now become open to variation and selection and the author argue that this would allow for subsequent purifying selection - removal of the potential harm caused by messing with mutS repair mechanism.

              Remember the beginning of this essay, how it was quickly established with this experimental system, that while it was easy to make cells cooperate, the emergence of cheats always brought about the death of the cooperating system because the cheats could always outcompeted the cooperating cells?

              Then it was thought that purging the cheats from the experiments was the best way to go. So what happens in comparison to the experiment just described, when cheats were purged from the system?

              To be continued ...


              • #8
                A reason for running the experiment in cheat purging (CP) mode was to better understand the results of the cheat embracing experiment.

                The only difference between the two kinds of experiment was that a WS cell began the new life cycle as opposed to an SM cell in the CE mode. This corresponded to cheats destroying the mat but it being an old mat cell (the WS type) that got the new mat underway.

                Predictions were that this system should have greater evolutionary potential using the previous definitions of fitness for both cells and mats. And indeed this was the case. After 10 life cycles, the evolved cells at least showed greater fitness than did the corresponding cells from the CE experiment. 

However, there was an interesting difference, and Figure 4b shows it. Cell fitness has increased along with mat fitness. Compare this with Figure 2b where while mat fitness increased, cell fitness decreased. This time mat fitness seemed to depend on cell fitness whereas before, both fitnesses were decoupled.


This observation confirmed for them that in the CE regime, the performance in mat fitness was driven by the alternating phases in the life cycles as opposed to being driven by the cells and their individual fitness.

                They explored the adaptations in the CP regime that led to increased lineage performance and found that in this regime there was no improvement in the ability of WS to switch to SM cell types. In fact fewer SM types were produced when compared to the CE regime and furthermore, it took longer for the SM types to arise within a mat. They considered that cheater suppression may have begun to evolve within the CP regime, and certainly there was nothing like a genetic switch tuned to the phases the cells were in.

                So the evidence suggested that improved line fitness was simply due to traits within the individual cells themselves. Improvements in lines was happening because selection was operating on the individual cells themselves and not operating on the mats.

                The differences were clear. In the cheat embracing experiment, mat fitness was related to mats not cells, and in the cheat purging experiment, it was related to cell fitness. The ability to switch cell types was opposite for both regimes as well. There was a negative correlation in the cheat purging experiment as opposed to the embracing experiment, suggesting that the switch depended very much on mutational changes in individual cells when purging was done. In the purging experiment, selection seemed to be operating only at the level of the cells, where as in the cheat embracing experiment, it seemed to be operating at both levels with cell fitness decreasing while mat fitness increased.

                Therefore, what did the researchers conclude from all this?

                To be continued ...

                Last edited by rwatts; 12-13-2014, 10:52 PM.


                • #9
                  In their summary/perspective section to conclude the article, the researchers note that multicellular organisms are descendants of free living single celled organisms. The fossil record supports this contention. Because single cells have the capacity for differential reproduction, then they were the units of selection. But when multicellular organisms arose, they became the focus of Darwinian selection in their own right. What made this possible was a means of reproduction. When multicellular organisms could reproduce then differential success at reproduction happened and thereby these organisms now became open to selection.

                  So how did the emergence of reproduction occur with the origin of multicellularity?

                  This experiment showed a possible pathway:-


                  Originally posted by link at OP
                  ... cheating cells—those types seemingly most detrimental to the persistence of newly formed cooperative entities—can function as a germ line within a life cycle that facilitates the reproduction of collectives.

                  The experiment also produced a novel kind of organism for selection to operate on, a two state organism where each state represents a different aspect of the organism. The first state was the mat and its growth and subsequent generation of a few germ cells. The second state was the death of the mat with the germ cells surviving, and switching to cells capable of generating a new generation of mat. These two phases were unified by the evolution of an underlying developmental program which incorporated a switch to flick between states and to cause a new organism (the mat) to form following the death of the old organism.


In contrast to this kind of entity was the single state system of previous experiments whereby mats grew, were overwhelmed by the subsequent development of cheats, causing the organism to die. The cells that remained went on to form new mats.

                  In this earlier system, the success of the mats really depended on the fitness of the individual cells from which the mats grew.

                  The authors noted that during the early stages of an evolutionary transition, 
levels of selection need to be considered because conflicts can arise, as indicated by this experiment. The design of the experiment established a two level ecology. On one level was that of the individual cells, reproducing every hour. Selection acting on these, clearly favoured short term success, given this hourly reproductive cycle. On the other level however, that of the cooperating cells, the mat, the cycle was 9 days. Here selection also acted but on a much larger time scale.

                  So if selection was acting for short term success, then it was unlikely to facilitate the persistence of the lineages which had cycles lasting days.

                  They argue that not only was a switch required to evolve, but also an underlying developmental program was needed to underpin the development of the multicellular organism, the mat. The switch would ensure the right kind of cell, depending on the phase of the life cycle. The program would bring about the onset of the mat. Therefore, as with a sexually reproducing multicellular organism like us, these experimental organisms had primitive germ cells which ensured life beyond the death of the mat. The switch and developmental program ensured the emergence of somite like cells from the germ cells, bringing on the formation of new mats.

                  The cells forming the body or the mat, played an ecological role. They allowed the mat to exist at the broth/surface interface where there was access to abundant oxygen for growth. But the body cells also played a reproductive role by producing some seed or germ cells (the cheat cells) which ensured the beginning of the next generation.


And so they conclude:-


                  Originally posted by link at OP
                  Given sufficient variation among lineages, then selection over the longer timescale stands to conquer the short-term interests of individual cells. This appears to have happened in our CE regime with decoupling of fitness between levels supporting the view that selection has begun the process of transitioning to the higher (collective) level—with the lower level beginning to function for the good of the collective.

                  The authors describe the kind of situation their experiment was attempting to simulate in a Supplimentary Discussion paper.

                  On the second page, about two paragraphs down, they offer a thought experiment with the “glass boundaries” of their particular experiment removed. The paper, including the thought experiment is worth a read. However, given that this experiment is all about the origin of multicellularity, I thought it odd that they relied on reeds (multicellular organisms) to provide support columns for the mats. Surely a shallow flat pool with protruding rocks, or the vertical edge of a lake shore would have been a bit more realistic. 

                  But I quibble.


Anyway, in my next post, I will conclude this set of essays.

                  Final post to come ...


                  • #10
                    In conclusion. 

                    The origin of multicellularity has long been a mystery - and naturally it remains one.

                    In experiments it is quite easy to get many kinds of cells that normally operate individually, to join up into mats of cells. However, genetic mutations will always ensure that some cells turn cheats. That is, they live in the mat, drawing benefits by living in the mat, but contribute nothing in return to the mat. And so these cheat cells have resources to burn and end up out competing and out reproducing the mat cells. Hence the mat dies.

                    Obviously, the way to conduct experiments on the origins of multicellularity, is to work out plausible routes by which cheat removal from the system may have evolved.

                    However the experiment described here tested the idea that the cheats, which caused mat death, may act as primitive germ cells and thereby allow the reformation of a new generation of mat, following the death of the old mat. After all, cheat cells do have, more or less, the same genome as the mat cells.

                    And this is what was found. 

In fact, in the most successful of the experimental lineages, it was found that the evolution of a genetic switch and an underlying primitive developmental program brought on a novel kind of multicellular organism in which the switch was sensitive to the phase the organism was in. When the mat was developing, the switch would be thrown and the development of cheat cells more or less ensured. This would cause the mat to die, with the cheat cells being the survivors. The switch would act again, to ensure that some of them switched back to mat growing cells, and thus a new mat would begin to develop.

                    And so it went.

                    This had the effect of allowing mats to form and die in cycles that went well beyond the six (maximum) cycles found in experiments in which cheats were always purged from the system.

                    One of the properties of the new organism was that mat success did not depend on individual cell success. Individual cell fitness dropped off while mat fitness improved. That is, cell fitness was decoupled from mat fitness, in contrast to the cell purging system where mat fitness depended very much on cell fitness.

                    With the new organism then, selection was operating independently on two distinct levels, that of the individual cells and that of the organism as a whole.

                    The organism itself could now participate in Darwinian evolution in its own right.

                    This brings to an end this set of essays. I suspect that I missed a lot and got some things wrong. Or that I misunderstood the importance of some concepts and did not give them the attention I should have. Nevertheless, I think I understood the gist of the experiment enough to have learnt something from it, and to have been able to pass those ideas on.

                    The end.

                    Last edited by rwatts; 12-17-2014, 03:13 PM.


                    • #11

                      Please stop annoying these anti-evolution folks with facts.

                      They're like garlic to a vampire.



                      • #12
                        Originally posted by klaus54 View Post

                        Please stop annoying these anti-evolution folks with facts.

                        They're like garlic to a vampire.


                        Problem is, I've come across another 5 or 6 papers that look equally as interesting, which have only just been published. The question is whether or not I can understand them well enough to do a write up.

                        Time will tell. :)

                        You will note that the only criticism I get from those representing creation science, are rants.

                        I'm having a look at one of these articles now regarding the origin of leaf mimic butterflies.


                        • #13
                          Judging from this layperson's writeup:-

                          A single, billion-year-old mutation helped multicellular animals evolve

                          - the paper:-

                          Evolution of an ancient protein function involved in organized multicellularity in animals

                          - could be interesting. If one has the time to slog through it that is. :(


                          • #14
                            Please stop discussing scientific evidence, lest Adrift disapprove.

                            Originally posted by Adrift View Post
                            So, basically you're just using this thread as a blog, is that right? I mean, outside of this back and forth I don't really see any discussion here, and this is a discussion forum.
                            Anyway, I'll need to dig into this as it relates to group selection. After all, a standard mantra amongst some biologists is that groups selection is equivalent to aggregated individual selection. That's wrong, however, if you're interpretation of the data is correct. Which confirms what I suspected: group selection is not equivalent to aggregated individual selection, and thus group selection might explain the presence of traits that are not adequately explained by individual selection. That would have implications for what I'm interested in: group-selection-based explanations of moral psychology.
                            "Instead, we argue, it is necessary to shift the debate from the subject under consideration, instead exposing to public scrutiny the tactics they [denialists] employ and identifying them publicly for what they are."


                            Related Threads


                            Topics Statistics Last Post
                            Started by TheLurch, 03-20-2023, 10:36 AM
                            3 responses
                            Last Post rogue06
                            by rogue06
                            Started by rogue06, 03-17-2023, 09:35 AM
                            6 responses
                            Last Post rogue06
                            by rogue06
                            Started by rogue06, 03-12-2023, 04:30 PM
                            1 response
                            Last Post Sparko
                            by Sparko
                            Started by rogue06, 03-07-2023, 11:07 AM
                            1 response
                            1 like
                            Last Post shunyadragon  
                            Started by Ronson, 01-02-2023, 02:12 PM
                            42 responses
                            Last Post Sparko
                            by Sparko