Thursday 22 June 2023
On irreducible complexity
Answering Farina on Behe’s Work: Irreducible Complexity
Editor’s note: We are delighted to welcome Dr. McLatchie as a new colleague. He is Resident Biologist and Fellow at the Center for Science and Culture.
On the Internet, some people are worth responding to and others are not. We have written enough lately in response to YouTuber Dave Farina following his recent debate with Rice University synthetic chemist Dr. James Tour. However, Mr. Farina has also posted a popular video criticizing Dr. Michael Behe’s work, and I will turn to that now.
I am a specialist in molecular and cell biology, and was thus interested in Farina’s video reviewing Michael Behe’s three books — Darwin’s Black Box, The Edge of Evolution, and Darwin Devolves. In this and several subsequent articles, I will offer a rebuttal to Mr. Farina’s analysis of Dr. Behe’s work. Here, I will address Farina’s commentary on Darwin’s Black Box – in particular, his alleged counterexamples of irreducibly complex systems having evolved by natural processes.
Evolution of the Cit+ Mutant
The video contends that Behe’s concept of irreducible complexity is empirically falsified by documented examples of systems that meet Behe’s definition of being “irreducibly complex” yet evolving by unguided mechanisms. The first exhibit is Richard Lenski’s long-term evolution experiment with Escherichia coli, in which it was observed that, after some 33,000 generations (15 years), bacterial cells evolved the ability to grow on citrate under aerobic conditions.1 The genetic basis for this ability was subsequently identified.2 E. coli already possesses the ability to grow on citrate under anaerobic conditions, facilitated by a citrate transporter protein encoded by the gene citT. A 2933 base pair stretch of DNA, containing the citT gene, underwent duplication. The result was that a copy of the hitherto unexpressed citT gene was placed under the control of the promoter of the adjacent gene, rnk, which as a consequence drove expression under aerobic conditions. This is a relatively simple change that does not require multiple co-dependent mutations to bring it about. Indeed, this instance of adaptation does not even require the origin of any novel genes and proteins, or even the modification of existing ones. The citT gene already codes for a citrate transporter which imports citrate into the cell in the absence of oxygen. The duplication event led to a loss of regulation of the citT protein such that it was expressed under both oxygen-rich and oxygen-deficient conditions, rather than in only oxygen-deficient ones.
Furthermore, the ability of the cells to grow on citrate under aerobic conditions was optimized by several other mutations.3 As Behe himself summarizes:
Even before the critical mutation occurred, a different mutation in a gene for a protein that makes citrate in E. coli degraded the protein’s ability to bind another metabolite, abbreviated NADH, which normally helps regulate its activity. Another, later, mutation to the same gene decreased its activity by about 90 percent. Why were those mutations helpful? As the authors write, ‘When citrate is the sole carbon source, [computer analysis] predicts optimal growth when there is no flux through [the enzyme]. In fact, any [of that enzyme] activity is detrimental.’ And if something is detrimental, random mutation will quickly get rid of it. Further computer analysis by the authors suggested that the citrate mutant would be even more efficient if two other metabolic pathways that were normally turned off were both switched on. They searched and discovered that two regulatory proteins that usually suppress those pathways had been degraded by point mutations; the traffic lights were now stuck on green.4
In other words, these mutations that optimized the ability of the cells to grow on citrate were in fact damaging. Since they supported the present needs of the organism, they were preserved by natural selection. As Behe argues at length in Darwin Devolves, the majority of mutations that are subject to positive selection are damaging rather than constructive, since there are far more ways to acquire an advantage by breaking something than there are ways to do so by building something. Readers may find interesting Behe’s response to Richard Lenski’s review of Darwin Devolves, which you can find here.
One researcher who has studied the evolution of citrate metabolism in E. coli is microbiologist Dr. Scott Minnich, a Fellow with Discovery Institute’s Center for Science and Culture. Farina is aware of this fact, and states that “Ironically, one of Behe’s colleagues at Discovery Institute, Dr. Scott Minnich — a professor in the department of agricultural and life sciences at the University of Idaho — co-authored a paper describing this pathway in detail, inadvertently highlighting its irreducible nature. I’m sure he got a slap on the wrist from Meyer and pals for that.” However, the Minnich paper5 in fact supports Behe’s interpretation of these results rather than Lenski’s — that is, that “no new functional coded element was gained or lost, just copied.”6 I am therefore doubtful that Farina has read the paper. Lenski’s original paper had suggested that the adaptation process involved three steps — potentiation (involving initial neutral mutations), actualization (the promoter fusion event described above), and refinement (increasing expression of the dctA gene, encoding a transporter that facilitates recovery of lost succinate during citrate import). In the view of Lenski and his colleagues, the reason the trait took some 15 years to evolve is because of the need for the initial neutral mutations (“historical contingency”) that by themselves do not promote growth but are necessary for the later actualization event. Minnich’s lab demonstrated that the mutants observed by the Lenski lab could be isolated much more rapidly — within 14 days rather than 15 years (and in as few as a hundred generations rather than 33,000) — and that the length of time it took in Lenski’s experiment more probably reflects an artifact of the experimental conditions than a requirement of evolution.
Vpu Protein
The video offers a few other supposed examples of irreducibly complex traits evolving, including the Vpu protein in HIV-1.7 Farina comments,
This now has a novel function — inactivating a protein of the human immune system called tetherin. Human tetherin is different enough from the same protein in chimpanzees such that simian immunodeficiency virus, from which HIV evolved, can’t counter it. This new trait requires three to seven specific mutations, leading to several completely new protein binding sites. HIV only jumped from chimps into humans around 1930, so this is another recent example of an irreducibly complex trait evolving.
Vigan and Neil (2010) “carried out an extensive mutagenesis of the HIV-1 NL4.3 Vpu TM domain to identify three amino acid positions…that are required for tetherin antagonism.”8 Mutating two of these positions (namely, A14L and W22A) showed a marked defect, whereas the A18L mutant showed only a minor defect. Thus, it is not even the case that all three amino acid positions are crucial for tetherin antagonism. Furthermore, viruses have enormous population sizes and extremely high mutation rates. They are therefore able to evolve complex traits requiring multiple co-dependent mutations far more readily than more complex organisms, including bacteria. Thus, for those two reasons, this is really not a good example to cite in support of Farina’s contention about irreducible complexity arising in more complex organisms.
Horizontal Gene Transfer to the Rescue?
As a further example, the video offers “the many animals that eat photosynthetic algae which appear to be on their way to becoming photosynthetic themselves via endosymbiosis.” Farina cites a paper on “Horizontal gene transfer of the algal nuclear gene psbO to the photosynthetic sea slug Elysia chlorotica.”9 The paper discusses the acquisition of plastids by the sea slug Elysia chlorotica by ingestion of the photosynthetic algae Vaucheria litorea. Though more than 90 percent of the proteins required for plastid metabolism are encoded in the nuclear genome of the algae, the plastids are nonetheless still able to photosynthesize within the sea slug. The paper determines that the essential plastid proteins are supplied by the sea slug itself, and that the genes which support photosynthesis have been acquired through horizontal gene transfer. But this does not involve the evolution of any new complex traits. The genes and proteins already existed but were simply transferred from one organism to another.
Rapid Lizard Evolution?
Next, the video notes that “There are lizards which are transitioning from laying eggs to live birth, in one case doing both in a single litter,” citing a paper from 2019.10 But it seems more likely that the Saiphos equalis was designed with the capability of switching between laying eggs (“oviparous”) and carrying embryos internally until completely developed (“viviparous”). Indeed, as Farina noted, there was even one case of a female giving birth to a live infant and laying eggs in the same litter. Moreover, it has been shown that very similar genes are expressed between the two modes, suggesting “that reproductive mode is relatively labile in this species,” and that “there may be fewer physiological barriers to transitions between parity modes than previously thought, an assertion that is supported by our gene expression data, in which oviparous and viviparous individuals expressed genes with many of the same functions during the reproductive cycle.”11 Furthermore, it has been previously shown that differences in parity mode between two species of Phrynocephalus are due to differences in gene expression rather than the result of many mutations.12
Multicellular Algae
As a final example, the video argues that “There are the unicellular algae that evolved in the lab to become permanently multicellular,” citing a paper by Herron et al. (2018) — “De novo origins of multicellularity in response to predation.”13 In the study, populations of the unicellular green alga Chlamydomonas reinhardtii were subjected to selective pressure by the introduction of the filter-feeding predator Paramecium tetraurelia. They found that two of the five populations developed multicellular structures. However, the multicellular populations lacked motility and the multicellular structures did not evolve multiple cell types. Moreover, as the authors of the paper note, “The ability of wild-type C. reinhardtii to form palmelloids [i.e. multicellular structures] suggests that the founding population in our experiment already possessed a toolkit for producing multicellular structures.” While the strains that evolved in the experiment are obligately multicellular (meaning that being composed of multiple cells is an essential and permanent part of their life cycle), the authors suggest that the genetic basis of the evolved multicellularity phenotype “involves the co-option of a previously existing plastic response.” If this is the case, the authors note, “the shift from a primarily unicellular (but facultatively multicellular) to an obligately multicellular life cycle may have required only a change from facultative to obligate expression of the genes involved in palmelloid formation.” In other words, the transition from being able to exist as single-celled organisms, while forming multicellular structures under certain conditions, to being permanently multicellular may have involved a shift from being able to turn the relevant genes on or off to the genes being permanently locked on. See also this short article by Michael Behe where he addresses a similar paper.
Straw Man Arguments
A common tactic in debate is setting up and then knocking down a straw man argument — that is, a version of your interlocuter’s argument that has been modified to render it easier to refute. The Farina video complains that ID proponents turn scientific research into a game of whack-a-mole by moving on to some other system and claiming that it is irreducibly complex when prior similar claims have turned out to be false. He comments, “No matter how many times creationists point to a supposedly unevolvable system and then biologists figure out its evolutionary history, there’s always another system creationists will point to and say, ‘But this this one — surely, this one — couldn’t have evolved.” However, I am not persuaded that any of the systems that have been proposed by Behe and other proponents of ID to be irreducibly complex have been shown to be evolvable by naturalistic processes. It would be one thing if there were detailed, plausible evolutionary scenarios that could account for thousands of complex biological systems save for a relative handful. Instead, it is the case that essentially none of the complex systems found in living organisms can be plausibly accounted for in this way.
The video also grossly misrepresents the structure of the inference from irreducible complexity to design. As Farina summarizes, “Humans build things; therefore, everything must have been built.” In fact, the inference is much more nuanced. The argument says that well-matched arrangements of parts that work together to achieve some higher-level objective are the sorts of systems that are habitually associated with conscious agents — since intelligent beings, unlike naturalistic processes, are capable of goal-directedness. Therefore, on the hypothesis that a mind was involved in the creation of biological systems, the existence of irreducibly complex systems is not particularly surprising. However, if the design hypothesis is false, such systems become wildly surprising. Given the top-heaviness of this likelihood ratio (or “Bayes factor”), these systems favor a design hypothesis. Moreover, since each requires a significant appeal to chance, each one is epistemically independent, meaning that the Bayes factors for the many thousands of irreducibly complex systems multiply together, forming a massive cumulative case for design.
Notes
Blount ZD, Borland CZ, Lenski RE. Historical contingency and the evolution of a key innovation in an experimental population of Escherichia coli. Proc Natl Acad Sci U S A. 2008; 105(23):7899-906.
Blount ZD, Barrick JE, Davidson CJ, Lenski RE. Genomic analysis of a key innovation in an experimental Escherichia coli population. Nature. 2012;489(7417):513-8.
Quandt EM, Gollihar J, Blount ZD, Ellington AD, Georgiou G, Barrick JE. Fine-tuning citrate synthase flux potentiates and refines metabolic innovation in the Lenski evolution experiment. Elife. 2015; 4:e09696.
Behe MJ, Darwin Devolves: The New Science About DNA That Challenges Evolution (HarperOne, 2020), 139.
Van Hofwegen DJ, Hovde CJ, Minnich SA. Rapid Evolution of Citrate Utilization by Escherichia coli by Direct Selection Requires citT and dctA. J Bacteriol.2016;198(7):1022-34.
Behe MJ, Darwin Devolves: The New Science About DNA That Challenges Evolution (HarperOne, 2020), 188.
Lim ES, Malik HS, Emerman M. Ancient adaptive evolution of tetherin shaped the functions of Vpu and Nef in human immunodeficiency virus and primate lentiviruses. J Virol. 2010; 84(14):7124-34.
Vigan R, Neil SJ. Determinants of tetherin antagonism in the transmembrane domain of the human immunodeficiency virus type 1 Vpu protein. J Virol. 2010; 84(24):12958-70.
Rumpho ME, Worful JM, Lee J, Kannan K, Tyler MS, Bhattacharya D, Moustafa A, Manhart JR. Horizontal gene transfer of the algal nuclear gene psbO to the photosynthetic sea slug Elysia chlorotica. Proc Natl Acad Sci U S A. 2008; 105(46):17867-71.
Laird MK, Thompson MB, Whittington CM. Facultative oviparity in a viviparous skink ( Saiphos equalis). Biol Lett. 2019; 15(4):20180827.
Griffith OW, Brandley MC, Belov K, Thompson MB. Reptile Pregnancy Is Underpinned by Complex Changes in Uterine Gene Expression: A Comparative Analysis of the Uterine Transcriptome in Viviparous and Oviparous Lizards. Genome Biol Evol. 2016; 8(10):3226-3239.
Gao W, Sun YB, Zhou WW, Xiong ZJ, Chen L, Li H, Fu TT, Xu K, Xu W, Ma L, Chen YJ, Xiang XY, Zhou L, Zeng T, Zhang S, Jin JQ, Chen HM, Zhang G, Hillis DM, Ji X, Zhang YP, Che J. Genomic and transcriptomic investigations of the evolutionary transition from oviparity to viviparity. Proc Natl Acad Sci USA.2019;116(9): 3646-3655.
M.D. Herron et al. (2019), “De novo origins of multicellularity in response to predation,” Scientific Reports 9: 2328.
Primeval nanomachines troll Darwinism again?
Local Fitness Landscape Mapped Out For Green Fluorescent Protein
As we have discussed many times, proteins are a show-stopper for evolution. Proteins consist of dozens, hundreds and even thousands of amino acids and, like most machines, they don’t work very well until most of the parts (amino acids in this case) are in place. Half of the amino acids don’t give you half the function of a protein. Now, a new paper reinforces the problem of protein evolution.
One approach to studying how evolution could create new protein designs is to start with some sort of random sequence of amino acids, see how well it works, and try to evolve it to obtain a protein. This is difficult because the protein design space is astronomically huge and proteins are sparse within that space. Any random sequence of amino acids will merely give you junk. Furthermore, the fitness landscape is flat and doesn’t provide the guidance evolution needs to move toward functional proteins.
Another approach is to start at the end and work backwards. In other words, start with the finished product—a functional protein—and see what the fitness landscape looks like as you swap in different amino acids. This is difficult because, unfortunately for evolution, the fitness landscape drops off precipitously as you move away from the native protein design. Modifying only a few percent of the amino acids leads to a rapid loss of function.
The new paper takes this second approach. It uses a bioluminescent protein known as the green fluorescent protein, taken from the jellyfish, Aequorea victoria. It is a wonderful study that systematically mapped out the protein’s function (as measured by the protein’s fluorescence) for a total of 51,715 different protein sequences that are nearby the native sequence.
The results confirmed what earlier studies had indicated: the protein function drops off dramatically with only a relatively small number of substitutions. But the study also explored the effect of multiple substitutions. It is well known that the effect of two substitutions, for example, are not always simply the sum of their individual effects. They can interact with each other in either positive or negative ways. This is referred to as epistasis.
The new study found that negative epistasis was strong and prevalent. As one of the researchers explained:
We were really surprised when we finally had a chance to look at exactly how the interactions between mutations occur. We also did not expect that almost all the mutations that are only slightly damaging on their own can destroy fluorescence completely when combined together.
It was well understood that evolving a protein is an astronomically unlikely event, and these results indicate it is even more difficult. Those negative results, however, were not reported in the paper. Instead, the paper discussed possible ways that one green fluorescent protein, found in one particular species, may have evolved into other green fluorescent proteins, found in other species. The implications for the initial evolution of a protein were ignored.
The Origin of Life=the origin of information?
Information and Life’s Origin — A Retrospective View
In the late 1980s, I first encountered the science of information theory and its application to theories of the origin of life. My introduction to the subject was a book, Origins and Destiny1, by research scientist Robert Gange. He described the idea of a generalized form of the traditional Second Law of Thermodynamics, based on quantum statistical mechanics. In conjunction with the understanding that the formation of life from non-life constitutes a rise in specified complexity rather than increasing order, this generalized Second Law raised a theoretical barrier against any natural origin of life scenario.
The traditional Second Law of Thermodynamics is viewed as an inviolable arbiter of possible outcomes for all physical processes. In particular, any conceivable proposal for a perpetual motion machine can, without analysis, be rejected based on the Second Law. With regars to the origin and development of life, the generalized Second Law states that any “alleged natural explanation…will be untrue in the same way a patent examiner in Washington, DC, knows an alleged invention for a perpetual motion machine is untrue.”2
A Postdoc Wonders
As a young postdoctoral researcher in physics, I wondered why the quantum statistical version of the Second Law wasn’t more widely recognized as an aspect of physical reality that limited any proposed natural mechanism for the origin of life. However, since the background physics of the generalized Second Law is typically buried in graduate-level texts on quantum statistical mechanics, perhaps its low visibility isn’t surprising.
Since my own physics research trajectory took me from experimental plasma physics to integrated optics to computational nano-electronics, my professional work did not directly intersect with research on the physical aspects of information theory. However, my personal interest in seeing how evidence for intelligent design held up to scientific advances kept my attention focused on relevant developments in cosmology and molecular biology.
In the early 1990s, the graduate library at the University of Washington provided access to research articles by Hubert Yockey on the application of information theory to the origin of life. Noting that “The information content of amino acid sequences cannot increase until a genetic code with an adaptor function has appeared,” Yockey states,
Nothing which even vaguely resembles a code exists in the physico-chemical world. One must conclude that no valid scientific explanation of the origin of life exists at present.3
Unlocking Life’s Enigma
By the mid 1990s, I had read The Mystery of Life’s Origin4, in which the authors distinguish between thermal entropy (related to the distribution of energy in the system), and configurational entropy (related to the distribution of mass — for example, the specific sequence of amino acids comprising a protein). Their analysis of multiple proposals for natural mechanisms to overcome the thermodynamic barrier represented by the high degree of configurational entropy in living systems led them to conclude that all such mechanisms are “clearly inadequate to account for the configurational entropy work of coding.”5
In the documentary video “Unlocking the Mystery of Life,” the premise of intelligent design was cogently defended as an alternative to the shortcomings of naturalism to explain such biochemical enigmas as irreducible complexity and the specific sequences of amino acids comprising functional proteins. The latter example of nature’s inability to increase the information content of a closed system over time led biophysicist Dean Kenyon to disavow his own research, published earlier in his textbook, Biochemical Predestination.
In the late 1990s, my academic schedule allowed me to pursue other source material on the physics of the generalized Second Law, and I obtained copies of two main textbooks on statistical mechanics referenced by Gange. The author of one of these texts gives a fundamental definition of information that emphasizes the relationship between information and the mind:
Information is the entity which makes the difference between knowing and not knowing, between being faced with a number of possibilities and between knowing the one that actually prevails.6
Only in the context of the mind is the difference between knowing and not knowing a meaningful distinction. The author of the other text, Arthur Hobson, affirms that given an initial measurement of a system, predictions of the system at a later time “cannot contain more information (but may contain less information)” than the initial data describing the system.7 This limitation on natural processes, based upon the laws of quantum mechanics, prohibits a system (even the entire universe bounded by the cosmic horizon) from progressing from a state of lower information (pre-life) to a state of higher information (post-life) by any combination of natural forces.
Chance and Necessity
William Dembski affirms the proscriptions of the generalized Second Law by showing that chance and necessity are insufficient to ratchet up the complex specified information (CSI) content of a closed system over time:
What natural causes cannot do, however, is originate CSI. This strong proscriptive claim, that natural causes can only transmit CSI but never originate it, I call the Law of Conservation of Information. It is this law that gives definite scientific content to the claim that CSI is intelligently caused.8
In 2005, while mentoring a senior honors student on a study of the boundaries of science, I read Stephen Meyer’s groundbreaking article outlining intelligent design as a source of biological information.9 Complementing the conclusions of other lines of research, Meyer’s analysis and conclusions added to the scientific literature affirming the validity of the generalized Second Law as a fundamental boundary of nature that disallows an unguided origin of life.
Have the last twenty years produced any experimental research demonstrating how unguided natural processes can successfully produce the complex, specified arrangements of components inherent in functional biomolecules? Any dearth of success in this field is not for lack of trying, nor is it surprising that scientists have not been able to coax even a single biochemically relevant protein into existence from a mix of ingredients likely to be available on the early Earth.
The Limits of Unguided Processes
A Rice University professor of chemistry, James Tour, who is renowned for his research in synthetic organic chemistry, draws this conclusion regarding any naturalistic origin-of-life scenario:
the requisite molecules (lipids, proteins, nucleic acids, and carbohydrates) are so unlikely to have occurred in the states and quantities needed, that we could never have gotten to the point of figuring out the genesis of the requisite code or information.10
Unguided natural processes, according to the generalized Second Law, cannot systematically increase the information content of a closed system over time. Since the origin of life, represented by the formation of a single-cell organism, constitutes just such an increase in information, natural processes are not expected to be the cause. Intelligence is the only recognized source of information, as inherent in living organisms. What we know of the laws of physics supports this conclusion.
Notes
Robert Gange, Origins and Destiny (Waco, Texas: Word Books, 1986).
Gange, Origins, (1986), 91.
Hubert P. Yockey, “Self Organization Origin of Life Scenarios and Information Theory,” J. Theor. Biol. 91 (1981), 13-31.
Charles B. Thaxton, Walter L. Bradley, and Roger L. Olsen, The Mystery of Life’s Origin: Reassessing Current Theories (New York: Philosophical Library, 1984).
Thaxton, et al, Mystery (1984), ch. 8 (available online: https://www.ldolphin.org/mystery/chapt8.html ).
Amnon Katz, Principles of Statistical Mechanics: The Information Theory Approach (San Francisco: W. H. Freeman & Co., 1967), 14.
Arthur Hobson, Concepts in Statistical Mechanics (New York: Gordon and Breach Science Publishers, 1971), 142-145.
William A. Dembski, “Intelligent Design as a Theory of Information,” (1997); https://www.discovery.org/a/118/ .
Stephen C. Meyer, “The Origin of Biological Information and the Higher Taxonomic Categories,” Proc. of the Biological Society of Washington, 117(2), (2004), 213-239.
https://www.jmtour.com/personal-topics/evolution-creation/.
Subscribe to:
Posts (Atom)