Independent Science News
Mon, 06 Feb 2017 14:01 UTC
The mythologising of DNA
Highly respected scientists make very strong claims for the powers of DNA. In his autobiography, Nobel Laureate Kary Mullis called it "The King of molecules" and "The big one". Maybe he read DNA: The Secret of Life, a popular science book that calls DNA the molecule that "holds the key to the very nature of living things". Its author should know. He is Nobel Laureate, James Watson, co-discoverer of the structure of DNA. Even institutions have strong opinions when it comes to DNA; the website of the US National Institutes of Health claims "Genes are at the center of everything that makes us human".
My edition of The Secret of Life features on its back cover Eric Lander. Lander is the celebrated brains behind modern human genetics. He is also the head of the Broad Institute at MIT. In his blurb, Lander endorses "The secret of life" trope. Just below him on the jacket is Professor of genetics Mary-Claire King. She writes: "This is the story of DNA and therefore the story of life, history, sex, money, drugs, and still-to-be-revealed secrets." According to Prof. King, DNA is life.
The Watson view of genetics dominates education too. The standard US high school biology textbook "Life", of which we own the 1997 edition, frames the entirety of biology around DNA, thereby giving it the biochemical status of life's centrepiece.
Some biologists will say that these views are extreme and unrepresentative. They are, and part of this article is to explain why extreme views about DNA dominate the public discourse. But its main purpose is to contrast the portrayal of DNA by virtually all biologists with the narrow scientific treatment they apply to other biological molecules. Our existence also depends on proteins, fats, carbohydrates and RNA (Ribonucleic Acid); but no one says "it's in my protein". But here is a question: is it any less scientifically preposterous to say something "is in my DNA"?
To take a ruthless look at that question is thus the purpose of this article. Does DNA have any claim to being in control? Or at the centre of biological organisation?
The answer is that DNA is none of the things Watson, Lander, and Collins claim, and that even the standard nuanced biologist's view of life is wrong. This is provable in many ways but mainly by a new science of life that is emerging from almost complete obscurity. This new science explains the features of living beings in productive new ways that DNA-centric, genetic determinist, biology has not, and cannot. DNA is not the language of God. It is not even the language of biology.
Organisms are systems
The evidence that DNA is not a biological controller begins with the fact that biological organisms are complex systems. Outside of biology, when we consider any complex system, such as the climate, or computers, or the economy, we would not normally ask whether one component has primacy over all the others. We consider it obvious that complex systems are composed of subsystems, each being necessary for the larger whole. Each subsystem has its specific niche but no one subsystem exerts a privileged level of causation.
The same applies to living organisms. At the level of the physiology of an individual organism we do not apply an exclusive or special causative role to the heart, the liver, the skin, or the brain, because a body is a system. All parts are necessary.
At the smaller biological scales of organs too, distinct cell types maintain, operate, and repair themselves and each other. Similarly, at the cellular level, no one disagrees that organelles and other molecular structures are interacting but independent subparts of the whole.
At the level of macromolecules, however, a curious thing happens. Biologists abandon systems thinking entirely. Instead, we apply the famous central dogma of biology, which is that DNA makes RNA makes Protein (Crick, 1970). This formulation creates an origin story that begins with DNA.
The first mistake of the dogma, however, is to call it "central". If an organism is a system, then there is no centre. The second error is that the pathway described is factually incorrect. The pathway should be a loop since the DNA does not come from nowhere: to make every DNA molecule requires proteins and RNA and DNA. More broadly, the synthesis of DNA cannot be done without a whole cell, just as the making of any RNA or any protein also takes a whole cell.
If we wanted to be more accurate still, we would say it takes a whole organism to make each of these components. Even this description would be incomplete, since, undeniably, it takes an ecosystem, including, in the case of humans, a gut microbiota and a food supply. The full formulation of the central dogma is therefore a loop embedded in a web. But the central dogma taught to millions of students every year takes an entirely different intellectual path. It arbitrarily confers on DNA a special place: firstly, by not closing the loop, and second, by placing DNA at its beginning. The central dogma is thus merely a representation formed from arbitrarily constructed boundaries. It is not biological reality.
Geneticists, and sometimes other biologists, make this linear interpretation seem plausible, not with experiments—since their results contradict it—but by using highly active verbs in their references to DNA. DNA, according to them, "controls", "governs", and "regulates" cellular processes, while nouns like "expression" are also commonly used to ascribe functions to DNA. Biologists thus confer activist and willful superpowers on DNA. Ultimately, this can create circular arguments. DNA controls embryonic development or organism health because genes express themselves. QED.
However, there is no specific science that demonstrates that DNA plays the dominant role these words imply. Quite the opposite. For example, a recent publication in Nature magazine posited "An emerging consensus that much of the protein constituent of the cell is buffered against transcriptional variation. " i.e. is insulated from direct genetic quantitative influence (Chick et al., 2016). This buffering is nicely demonstrated by many experiments. One is the demonstration that the circadian rhythm of a bacterium can be reproduced, in the absence of any DNA, by just three proteins mixed together in a test tube. The rhythm was maintained for three days, even in the face of temperature changes (Nakajima et al., 2005).
Inevitably, any language used to describe DNA will necessarily be metaphorical and be of limited accuracy, but words like "govern" and "control" literally invent attributes for DNA (Noble, 2003). A much more precise metaphor for DNA would compare it to the library of Congress, since cells use DNA primarily as a storehouse of information. Consider that biologists could apply more neutral verbs such as "use", as in "cells use DNA to create proteins". If so, they would have created a very different status for DNA. Only librarians would have T-shirts saying "its in my DNA".
If we shed the wild metaphors and the central dogma, a more accurate way to think about biology emerges. If every molecule and every subsystem, regardless of scale, constrains and potentiates the other parts, then there is no need to infer a central controller. We can replace the DNA-centric model of biology with a relational model of complex interplay of feedback systems and emergent properties, of which the library of DNA is just one component. In this model, RNA is simply one of the inputs needed to make proteins and DNA is just one of the inputs needed to make RNA, and so on. Unlike the central dogma, such a proposition is consistent with the known facts of biology.
The formulation encapsulated by the central dogma and by biology textbooks is therefore an illusion. They are a classic case of what microbiologist Carl Woese has called the "reductionist fundamentalism". Reductionist fundamentalism differs from simple reductionism in that whereas simple reductionism is a valid scientific method, the former is an ideological preference for a simplistic explanation when a more holistic one is better supported by the evidence. In this case, the assigning of superpowers to DNA to explain observed biological activities when a better explanation would accept that many biochemical events have multiple causes and contributors. Oxford physiologist Denis Noble describes this fallacy as conferring on DNA "a privileged level of causation".
If not DNA, is there a "molecule of life"?
Many plant-infecting viruses lack DNA. They base their lifecycles on protein and they use RNA as their heritable material.
There are also plant pathogens, called viroids, that lack both DNA and protein. Viroids are thus composed solely of non-coding RNA. Lifeforms can therefore exist without either DNA or proteins—but there are none that that lack RNA.
Therefore, the answer to the opening question: "what kind of biomolecule is possessed by all living organisms?" is RNA. RNA stands for Ribonucleic Acid and for many reasons it is a better candidate for being a universal biomolecule than DNA.
RNA and DNA are chemically very similar. Even scientists confuse them, but their modest chemical distinctions confer very different properties. RNA is structurally very flexible (bendy), whereas DNA is highly inflexible; RNA is unstable and chemically reactive, whereas DNA is highly inert. A key difference is the number of chemical modifications that cells are able make to their four bases. In the case of DNA (whose bases are the nucleotides A,C,G and T), just two modifications are possible in most cells. These modifications are called methylation and acetylation. These two modifications alter the properties of DNA bases and they are the primary basis of the fashionable science of epigenetics.
RNA also has four bases (A, C, G, and U). But cells make more than one hundred comparable chemical modifications to them. The roles of these modifications are essentially a mystery, but presumably they help RNA perform its many cellular tasks.
RNA is also misunderstood. In a typical human cell, less than 1% of it makes proteins. The remaining 99% has a huge variety of structural, regulatory, and enzymatic functions. Most biologists though might as well be slaves to the central dogma in thinking that RNA is just the intermediate between DNA and protein. Only recently has RNA begun emerging from the shadow of DNA as a far more interesting molecule.
The deep explanation of these molecular differences is that RNA existed long before DNA. RNA probably predated even the invention of cells. It is enormously old. In consequence, it is so deeply and structurally embedded in living systems that it is very hard to study. Thus the paradoxical reason why we don't know much about RNA is not because it is unimportant, but because, unlike DNA, RNA is too important to cell function to selectively remove at will.
Consequently, to conform with current evolutionary understanding, we should really invert standard teaching and insist that the proper way to think about DNA is that it is a specialised form of RNA. DNA evolved structural rigidity and chemical inertness to make itself a more staid librarian for the safe storing of heritable information.
So, over evolutionary time DNA was chosen as a better librarian (this library metaphor originates with Colin Tudge and his excellent book Why DNA isn't selfish and people are nice); proteins turned out to be superior catalysts of chemical reactions; but RNA is more likely to have been the biomolecule around which life was really built. But RNA is no more a controller than is DNA.
Nor is DNA the centre of evolution
A common explanation for organising biology around DNA, and the one given by the authors of "Life", the textbook, is DNA's supposed role in the theory of evolution. For two reasons this explanation is highly questionable, however. Both reasons exemplify pervasive misunderstandings of the theory of evolution. One of these misunderstandings exaggerates the significance of Darwin's theory and the second, once again, gives to DNA credit it doesn't deserve.
The first misunderstanding is to assume that evolutionary theory is an explanation of life. Life, however, began long before Darwinian evolution and some of its fundamental patterns (cells, proteins, energy metabolism) emerged—so far as we can tell—long before DNA became the molecule of heredity (Carter, 2016). This distinction is important. In a textbook about "Life", for example, it is important to separate the origin of life from its maintenance so as not to unhelpfully exaggerate (i.e. confuse) what Darwin's theory explains; but in conflating the two, "Life" is only reflecting the misunderstanding of most biologists.
Second, the pre-Darwinian life of cells and metabolism arose thanks to the fact that complex systems have emergent and self-organising properties (e.g. Kauffman, 1993; Carter, 2016). The advent of DNA into these systems allowed Darwinian evolution to accelerate, but it did not eradicate emergent and self-organising properties. Rather, it colluded with them and helped create new ones. This means such properties are the likeliest explanation of large areas of biology. "Self-organization proposes what natural selection disposes" is how Batten and colleagues quaintly summarise alternatives to standard evolutionary theory which is pretty much rigidly genetic determinist (Batten et al., 2008).
A classic emergent property is the folding of proteins. DNA encodes the linear sequence of amino acids that constitute proteins, but every protein adopts one (or usually more) highly complex three dimensional shape (Munson et al., 1996). These shapes, along with charge and solubility, are largely responsible for a protein's properties. It is habitually, but lazily, presumed that DNA specifies all the information necessary for the formation of a protein, but that is not true. All protein shapes depend also on the integration of multiple sources of information. These sources include temperature, other cellular molecules like water and mineral ions, pH, energy molecules like ATP, protein folding aids called chaperones, and so forth. Beyond this, many proteins have functions, such as to be molecular channels and pumps, that emerge only at higher levels of structure, such as in the presence of other proteins.
Thus DNA specifies proteins and their functions only up to a very limited point. It is possible to disregard all such non-genetic contributions and ascribe to DNA all the properties of a protein or a process (or a whole organism). Most scientists do, but doing so is an ultra-determinist position. It writes emergent properties, such as protein folding, entirely out of the functioning of life. It again confers onto DNA superpowers it does not have.
Emergent properties are only one example of why the relationship between DNA and evolution is much more tenuous than is normally portrayed. Patrick Bateson of Cambridge University, whose perspective is not emergent properties but animal behaviour, explained evolution much more accurately than most when he wrote: "Whole organisms survive and reproduce differentially and the winners drag their genotypes with them. This is the engine of Darwinian evolution".
Thus we can explain why Charles Darwin invented his theory of evolution without knowing DNA even existed, because, even for evolution, DNA still is not "The big one", but it is standard for biologists to teach that DNA is more important to evolution than any other component of living organisms.
Explaining genocentric biology
When Dorothy journeyed to the Emerald City she discovered that The Wizard of Oz was only "a common man". He was devoid of magic powers and so could not help her friends. But there was at least something behind the facade. The same is true for DNA.
Most cellular molecules are highly reactive and transient chemical substances. That means they are difficult to extract, and hard to study. So it is with RNA and proteins.
DNA, however, is a much more practical point of intervention in biology. It is stable and robust and simple enough to be isolated on a reproducible basis and copied precisely. With an hour of training, high school students can do it. With a bit more training, DNA can be altered and, in some species, replaced. Hence the alarm over garage hacking of DNA.
This explains, in a nutshell, why our understanding of gene regulatory networks runs far ahead of our understanding of other disciplines of biology. It is because DNA is the low hanging fruit of biology.
Scientific dissent around DNA
"The human body completely changes the matter it is made of roughly every 8 weeks, through metabolism, replication and repair. Yet, you're still you - with all your memories, your personality... If science insists on chasing particles, they will follow them right through an organism and miss the organism entirely." mathematical biologist Robert Rosen is supposed to have said. And indeed, examine any multicellular organism and concealed under its relatively calm surface are circulatory systems, churning stomachs, lymphatic drainage systems, electrical impulses, biomolecular machines and so forth.
These systems cause every part of an organism to continuously move, contract, twist, vibrate, strain and grow. What defines living organisms, in the final analysis, is their dynamic and animate nature. This is why, when we want to know if an organism has legally died we don't examine its DNA, we measure its heartbeat or brain function. Animate properties require animate components, like RNA and proteins.
Yet by organizing our understanding of life largely around DNA (recall Mary-Claire King's "DNA is life"), biologists have curiously chosen the cellular constituent that is probably the least representative of life's dynamic nature.
For this reason there are dissenters in biology. Some are prominent. Some are not. They all have questioned whether biology is not much more complex and interesting than our present DNA-based framing can make room for (e.g. Kaufman, 1993; Strohman, 1997; Rose, 1999; Woese 2004; Annila and Baverstock 2014; Friston et al., 2015).
These dissenters like to note, for example, the general absence of medico-scientific breakthroughs following the sequencing of the human genome and the ever-more-detailed-analysis-of-tiny-scraps-of-human-DNA (Ioannidis, 2007; Dermitzakis and Clark, 2009; Manolio et al., 2009).
Some go much further in their critiques than others. Carl Woese, perhaps the best known bacteriologist since Pasteur, argued before his death that genetic determinism is a dead end, its vision of biology is "spent" (Woese, 2004).
There perhaps is no finer example of this than the field of tissue engineering. Tissue engineers claim to have made "incredible" progress making whole human organs in vitro for transplanting and other medical uses, yet these organs are all non-functional (Badylak, 2016). They don't have blood vessels or immune systems or nerve networks, they are just human cells on an ear-shaped scaffold or a hand-shaped scaffold and so, among their many deficiencies, they are short-lived because they have no regenerative properties.
Many biologists suspect at least part of this paradigm problem, but they rarely act on it. The sole noticeable official response to the obvious fact that organisms are highly complex systems has been to shovel modest funding in the direction of 'systems biology'.
One is bound to note that even this systems biology is rarely the study of systems. Instead, biologists have overwhelmingly used systems biology funds not to further the understanding of complex systems but to scale up and mechanise their reductionism.
Thus no scientific specialism or institution has articulated the profound inadequacy of viewing organisms as collections of gene regulatory networks or moved towards assembling an alternative paradigm (or paradigms) to replace it (Strohman, 1997).
This intellectual near-vacuum is nevertheless being steadily filled by individual scientists, mostly on the margins, with promising, even revolutionary, theoretical developments and experimental findings that explain biological phenomena in ways that transcend genetics.
A short guide to alternative paradigms of life
A Helmholtz machine is a sensory device that makes a prediction about reality and crosschecks it against that reality. It then estimates the difference between the two. Bayesian statistics is a mathematical method of doing the same: estimating differences between expectation and reality.
A new theory of neurobiology, called the Bayesian brain theory, proposes that the brain is the biological equivalent of these (reviewed in Clark, 2013). Brains make predictions, measure the mismatches with their expectations and pass those mismatches up to higher neural circuits. These higher circuits repeat the process and if mismatches persist then these are passed on to yet 'higher' mental levels.
The Bayesian brain hypothesis is quite new and predictive neurons might seem superficially improbable, yet the hypothesis appears to explain numerous aspects of brain structure and brain function; for example, how the brain can treat widely different stimuli (visual, sensual, oral, aural, etc.) essentially with the same neural mechanisms and structures. It also appears to show how the brain can integrate action and perception. The theory also provides a substantive explanation of learning: learning is the updating of the predictive model. The Bayesian brain hypothesis may even explain how brains evolved higher levels of consciousness over evolutionary time periods: by adding new layers of prediction.
A particular strength of the Bayesian brain hypothesis is that it corresponds to the actual spatial organisation of neurons in the primate cortex in which ranks of "predictive" neurons and "sensory" neurons send signals in opposing directions which lets them cancel each other out (except for the mismatches).
The structure-based predictive learning system proposed by the Bayesian brain hypothesis is of interest here because it relegates detailed genetic explanations of many phenomena, including arguably all consciousness, to the margins (Friston, 2010). Genes and proteins may fill in the details but many of the key elements of brain function: learning, action, and perception, derive primarily from structure alone. I.e., like protein folding, they are emergent properties of organisation.
Emergent properties are equally important in other areas of biology. An example is the vascular system of plants. Trees can transport water from unsaturated sources hundreds of feet into the air. Transpiration, as it is called, requires no energy input. Rather, it takes advantage purely physical properties of hydrophilic xylem tissues (tubes) and the properties of water itself. Without transpiration, which already operates, but only very weakly, in soils, plants could not exceed a couple of inches in height, nor tolerate dry conditions (Wheeler and Stroock, 2008). Thus, the defining characteristic of plants (apart from photosynthesis) is their clever exploitation of a simple physical property of water.
A further example is the arches of the human foot. These are longitudinal and transverse diaphragms composed of bone and connective tissue whose emergent property is both to dissipate forces at impact and operate as springs to transfer energy from impact into forward motion. Arches reduce the energy needed to walk or run.
In the discipline of biochemistry, a recent development is the proposed existence of metabolons. Metabolons are three-dimensional spatial arrangements of enzymes. Metabolons explain how the product of an ostensibly minor metabolic pathway can nevertheless constitute 30% of the weight of a seedling and so drive away pests (Laursen et al., 2017).
A more conventional class of self-organising properties found in biology are homeostatic feedback loops. They too are phenomena largely independent of gene functions with key roles in explaining the activities and properties of living organisms. The three proteins noted earlier that can recreate a bacterial circadian rhythm are just one example (Nakajima et al., 2005).
At more elemental and universal levels of life are unifying theories of cells and metabolism, many of which relate life to the operation of fundamental physical forces. The father of all such theories was arguably Nicolas Rashevsky, who died in 1972. He is survived by his students Robert Rosen and AH Louie. Others include physicist Erwin Schrödinger, author of "What is life?"; Stuart Kauffman, author of "The Origins of Order" (1993); Steven Rose "Lifelines: Biology beyond determinism" (1997); Enrico Coen "The Art of Genes" (1999); Denis Noble, "The Music of Life" (2003) and Dance to the Tune of Life: Biological Relativity (2017); and Annila and Baverstock who argue life is the inevitable outcome of the second law of thermodynamics (Annila and Baverstock, 2014; see also Friston et al., 2015). These, and other omitted thinkers, have gone far in assembling the potential raw material for a scientific revolution. One that leaves the framework of gene regulatory networks far behind.
The closest that of any of these theories come to definitively falsifying genetic determinism as a life-concept, however, would be a theory of the origin of life itself that positions metabolism at the centre.
Readers may be familiar with the concept of the RNA world, which is theorised to have predated the supposed "modern DNA world". But more convincing than an RNA world, for which there is little evidence, is a new theory, the peptide-RNA world.
The central piece of evidence of the peptide-RNA origin thesis (Carter, 2016) is that the enzyme (called aminoacyl-tRNA synthetase) that nowadays links RNA to proteins—and which therefore connects the RNA world to the protein world—comes in two basic forms (in all organisms). The evolutionary origin of these two forms (called Class I and Class II enzymes), however, is strangely irreconcilable. Class I and II molecules perform almost identical functions (though with different amino acids) yet have nothing structurally in common. Except for one thing. Their most conserved aminoacids, those at their active catalytic centre, can be derived from opposite strands of the same small RNA molecule (Carter 2016). In other words, the two proteins that let RNA make all modern proteins are derived from opposite strands of a single very primitive small RNA molecule that encoded them both.
The implication of this compelling observation is to intimately link metabolism and replication at a very early stage of life's origins. RNA was the assembler of primitive proteins and the purpose of those proteins was catalysis, i.e. to guide and enhance metabolism. What the peptide-RNA origin thesis therefore does is to replaces the RNA world—which is a replication-first theory—with a metabolism-first theory in that RNA is enhancing a metabolism that already predated it.
DNA and politics
"Human biology is actually far more complicated than we imagine. Everybody talks about the genes that they received from their mother and father, for this trait or the other. But in reality, those genes have very little impact on life outcomes. Our biology is way too complicated for that and deals with hundreds of thousands of independent factors. Genes are absolutely not our fate. They can give us useful information about the increased risk of a disease, but in most cases they will not determine the actual cause of the disease, or the actual incidence of somebody getting it. Most biology will come from the complex interaction of all the proteins and cells working with environmental factors, not driven directly by the genetic code". (Anand et al., 2008)
This quotation, spoken (but not written), by Craig Venter, the legendary genome sequencer, suggests that even many geneticists secretly appreciate a clear need for alternative paradigms.
At the same time the Venter quote prompts a deep question: How is it that, if organisms are the principal objects of biological study, and the standard explanation of their origin and operation is so scientifically weak that it has to award DNA imaginary superpowers of "expression" and "control" to paper over the cracks, have scientists nevertheless clung to it?
Why is it that, rather than celebrating and investing in Rashevsky, Kauffman, Noble, et al., as pioneers of necessary and potentially fruitful and unifying paradigms, have these researchers been ignored by mainstream biology?
What is the big attraction of genetic determinism?
A compelling and non-intuitive explanation for the monomania of biology does exist. It is set out in a second and forthcoming article: The Meaning of Life. It is an explanation that requires going behind the window dressing of science and examining its active and symbiotic relation to power in modern political systems.