Evolution Icon Evolution
Intelligent Design Icon Intelligent Design

Vindicated But Not Cited: Paper in Nature Heredity Supports Michael Behe’s Devolution Hypothesis

Casey Luskin
Photo: Michael Behe discusses “Biological Truth & Myth: Insights from the Foundation of Life" (screenshot).

Last year I wrote an article here at Evolution News titled “Harvard Molecular Geneticist Vindicates Michael Behe’s Main Argument in Darwin Devolves.” I discussed a 2020 paper by Andrew Murray in the journal Current Biology. That paper stated:

In laboratory-based experimental evolution of novel phenotypes and the human domestication of crops, the majority of the mutations that lead to adaptation are loss-of-function mutations that impair or eliminate the function of genes rather than gain-of-function mutations that increase or qualitatively alter the function of proteins.

Murray’s paper vindicated Behe’s thesis which also argued that “random mutation and natural selection are in fact fiercely devolutionary.” That is since “mutation easily breaks or degrades genes, which, counterintuitively, can sometimes help an organism to survive, so the damaged genes are hastily spread by natural selection.” (Darwin Devolves, p. 10) He continues:

Darwinian evolution proceeds mainly by damaging or breaking genes, which, counterintuitively, sometimes helps survival. In other words, the mechanism is powerfully devolutionary. It promotes the rapid loss of genetic information. [p. 37, emphasis in original.]

Vindicated Again

Now Behe has been vindicated again by a 2021 paper in Nature Heredity, which agrees that “loss-of-function” mutations are prevalent in the evolutionary process:

Views on loss-of-function mutations — those abolishing a gene’s biomolecular activity — have changed considerably over the last half century. Early theories of molecular evolution that emerged during the 1960’s and 1970’s saw little potential for loss-of-function mutations to contribute to adaptation (Maynard Smith 1970). Except in the case of inactivated gene duplicates, nonfunctional alleles were often assumed to be lethal, with adaptation being generally regarded as a process explained only by the fixation of single, mutationally rare alleles that improved or altered a gene’s function (Orr 2005). Only relatively recently, through discoveries enabled by the availability of molecular sequence data, were alternative views of adaptive loss-of-function alleles formalized, most notably with the “less is more” ideas proposed by Olson (1999). Classical paradigms of molecular evolution had by that time been challenged, for example, by evidence that natural loss-of-function variants of CCR5 lead to reduced HIV susceptibility in humans (Libert et al. 1998). Discoveries during the subsequent two decades have continued to support the idea that loss of function contributes to adaptation (Murray 2020), with cases of adaptive or beneficial loss of function being discovered across diverse organisms, genes, traits, and environments.

(Monroe et al., 2021, “The population genomics of adaptive loss of function,” Nature Heredity)

You might notice that the final citation in the quote above is to Murray (2020) — that’s the same Andrew Murray mentioned above, the Harvard geneticist I wrote about last year who similarly proposed that evolutionary adaptations frequently proceed by breaking functionality at the molecular level. Professor Murray’s paper was appropriately cited, but unfortunately this 2021 paper conspicuously avoids citing Behe’s 2010 paper in the respected Quarterly Review of Biology, “Experimental evolution, loss-of-function mutations, and ‘the first rule of adaptive evolution’.” There, Behe makes similar arguments. Nor does it cite Behe’s 2019 book Darwin Devolves. But it’s enough to accept the vindication even if we don’t get the citation. 

Citing away to the relevant literature (excluding Behe), this 2021 paper continues:

Today, reductive genome evolution is viewed as a powerful force of adaptation (Wolf and Koonin 2013) and gene loss is considered an important source of adaptive genetic variation (Albalat and Cañestro 2016; Murray 2020). … While the existence of adaptive loss of function is no longer seriously disputed, the assumed maladaptive nature of loss of function from early theories can persist in the language of population genetics such as in the continued use of deleterious as a synonym for loss-of-function (Moyers et al. 2018). … he existence of a category of alleles distinguished by a derived loss of biochemical function has been described by various names: “amorphic” (Muller 1932), “loss-of-function” (Jones 1972), “nonfunctional” (Nei and Roychoudhury 1973), “knockout” (Kulkarni et al. 1999),”null” (Engel et al. 1973), “pseudogene” (Jacq et al. 1977), or simply “gene loss” (Zimmer et al. 1980). Total gene loss is the most obvious case of loss of function. Comparisons of gene content between distantly related species have revealed considerable evidence for adaptation via complete deletion of genes or even entire sets of functionally related genes (Wang et al. 2006; Blomme et al. 2006; Will et al. 2010; McLean et al. 2011; Griesmann et al. 2018; van Velzen et al. 2018; Sharma et al. 2018; Huelsmann et al. 2019; McGowen et al. 2020; Baggs et al. 2020).

A Theoretical Basis for Behe

The article goes on to explore various mutational mechanisms which lead to loss of function in genes, and complete gene loss, providing a theoretical basis for Behe’s thesis:

First principles and empirical evidence indicate that many types of mutations can have effects that are equivalent to total gene loss, and for the purposes of this review, we employ this definition of complete gene losses being functionally equivalent to other loss-of-function mutations such as premature stop codons. However, there is the practical difficulty that these different types of mutations vary in how easily they can be detected and correctly annotated as loss-of-function alleles (Fig. 3). Insertions and deletions that interrupt the reading frame of a protein coding region (frameshift mutations), for example, might be readily classified as loss-of-function alleles because the downstream amino acid sequence will be severely disrupted. Yet a frameshift mutation at the extreme 3′ end of a coding region affecting only a few amino acids might be functionally distinct from a frameshift mutation at the extreme 5′ end disrupting the entire coding sequence. One simple heuristic to address this ambiguity is a threshold, measured by the portion of the gene affected by functionally disruptive mutations, at which an allele is classified as loss-of-function. This approach can be used to classify premature stop codons, frameshift, splice site disruptions, start loss, and inframe insertions and deletions. In humans (MacArthur et al. 2012; Karczewski et al. 2020) and Arabidopsis thaliana (Monroe et al. 2018; Baggs et al. 2020), loss-of-function mutations affecting only a small fraction (e.g., <10%) of total coding sequence in a gene were ignored when classifying loss-of-function variants.

The paper then explores methods for detecting when loss-of-function has been selected by natural selection:

Loss-of-function alleles were once often held up as a paragon of deleterious genetic variation. Today a more nuanced appreciation for their potential role in adaptation has emerged. This new paradigm inspires investigations into deeper questions about the causes and consequences of adaptation by genetic loss of function. For example: Do species or populations differ in their capacity to adapt via loss of function, and if so, why? Does the high effective mutation rate of loss-of-function alleles lead to bias in the probabilities of different evolutionary outcomes? What is the contribution of adaptive loss of function to the phenomena of antagonistic pleiotropy and reproductive isolation? How does adaptation by loss of function affect long term evolutionary trajectories of populations and future evolvability?

An Important Question

Focus on the last question: What does the prevalence of function-damaging mutations imply for the evolutionary process? This question is an important one, and it’s one to which Michael Behe, through the lens of intelligent design, has proposed an answer. The answer is essentially that a mechanism that proceeds primarily by breaking molecular features cannot easily account for the origin of new functional biological features at the molecular level. Here’s what Behe writes in Darwin Devolves:

From the dawn of life to the present, beneficial degradation has been a constant background — there’s no way to avoid it. From the beginning the Darwinian mechanism has been self-limiting, capable to an extent of eliminating or modifying preexisting molecular systems and in the process giving rise to new varieties of creatures below the biological classification level of family (described in Chapter 6), but incapable of building functionally complex molecular structures. To explain them, we must look elsewhere. (p. 251)

The literature is looking at the same data that intelligent design proponents are looking at, making similar observations, and asking similar questions. While ID proponents don’t always get recognized in the literature for their contributions, a careful analysis nonetheless shows that ID thinking is highly relevant to answering questions that everyone is asking.