WHEN EVOLUTIONARY EXPLANATIONS are invoked to explain the existence or persistence of specific traits, the normative causal reasoning invokes a benefit to the individuals carrying the genes for those traits, a concept widely understood as individual fitness. Natural selection then results as an outcome of competition for the most limiting resources; nature providing a grand stage upon which all of the theatrical plays of life are worked out, with some characters thriving, and others denied contribution to the next generation: for them, their genetic fate a dismal tragedy.
Tomes of very clever ideas, and deep understanding on how complex evolution could become were worked out in for 90 years in the pages of journals that published theoretic works by such great minds as Sewall Wright, whose ideas on balancing selection explain how two traits, both slightly deleterious, could offset a certain march toward genetic doom, leading to the persistence of apparently deleterious traits in evolving populations over time. He and others worked out ways of thinking about complex genetic interactions, such as epistasis and pleiotropy. All shifts in gene frequencies in most populations, it was thought, could be well understood as a balance between the relative contribution of alleles to benefits in terms of survival and reproduction, and inconsistencies and unlikely or seemingly impossible outcomes could result from shifting balances across adaptive landscapes.
Utterly brilliant, this body of work was confronted with new thinking upon the the arrival of even the earliest modicum of data. Motoo Kimura showed, for example, that most of the changes in gene frequencies over time where not likely to involve a specific genetic contribution to differential survival and contribution, but were rather simply a mathematical necessity, driven by mutation pressure, with most nucleotide substitutions occur simply because there was not enough room in the population to hold all of the alleles at a particular locus. This is most easily understood in very small populations, in which the smallness of the population makes random factors that influence the survival of specific genetic variants much more important than they would be were those alleles found in a larger population. Alleles could come into existence by chance, and then, driven by chance, drift to fixation to the exclusion of other alleles, regardless of their relative contribution to survival and reproduction. Under Kimura’s neutral theory of evolution, and its descendant theories, mutation pressure places the limit on the rate of evolution, and most (the largest percentage) of genetic changes over time do not particularly influence survival or reproduction, or if they do, those changes are carried along by chance.
We should understand that evolution is a combination of Darwinian natural selection and neutral drift, both rates limited by the amount of genetic variation in a population, which itself is limited by factors such as mutation rates, and, for a given subpopulation, access to mates from other subpopulations that might contain different alleles and genotypes.
We understood then, and now, that mutations are not restricted to nucleotide substitutions, and that large, deleterious genomic differences can have large impacts on the relative survival and reproduction of individuals, and that many traits observable in a population are impacted by the effects of the environment on specific methylation patterns in parents’ (mostly mothers’) chromosomes, and that changes that occur during gametogenesis both of the nucleotide, genomic and epigenetic levels can be found in offspring. Most of the important changes in phenotypes over time likely involve traits that are related to differential survival and reproduction, both of which can be measured objectively. In other words, what is important in terms of providing an evolutionary explanation can be consider both in terms of counting shifts in gene frequencies, and in their impact on survival and reproduction.
When I began studying genomic and genetic changes related to disease, I performed most of my studies in the service of others. I worked hard to try to insure that the data analysis techniques and models we employed were at least empirically reproducible, if not founded upon a solid theoretical basis. When technologies brought the ability to assay entire genomes, transcriptomes, and proteomes, I worked overtime, often deep into the early morning hours, comparing alternative methods for data representation, normalization, finding differences, and interpreting in search of the most empirically justified frameworks to tell, from a single data set, which gene, proteins, or methylation patterns were likely to be truly important to a particular disease. I started in cancer (nearly all types), and from there moved into nearly every other domain in medicine, including immunology, diabetes, pulmonology, neurology, and so on. In any given week I’d be involved in 4-5 different studies, and when finally tallied, during my 16 years working exclusively on these problems, I contributed results to over 100 studies. Many of the studies were, appropriately, at the pilot level.
I had on several occasions had the opportunity to work with experts in immunology, working on cancer. There was a good reason for that. Part of my attraction to come work at the University of Pittsburgh Cancer Institute was Dr. Ronald Herberman, who had assembled a cracker jack team of immunologists focused on various aspects of cancer. While working as a post-doc under Dr. Masatoshi Nei, I became immersed in his excellent studies on the evolution of the very complex MHC loci, and the amazing adaptive immune system, which uses combinatoric guessing at matching epitopes found in nature. While at Penn State and at UPCI, it took a while, but over time I became intimately familiar with the roles of the various types of cells, tissues, and signals. It was an amazing time, and I had never been happier in my position. From breast cancer to melanoma, pancreatic cancer and lung cancer, I had a chance to see some of the world’s foremost researchers working to understand both the causes of and processes of cancer, driven by mutations, and the optimal routes to treatment. Biomarkers were a huge part of the experience, as well as a search for their biological significance (function).
My interests in autoimmune disorders sank in while writing three books: Ebola: An Evolving Story; Cures vs. Profits:Successes in Translational Research, and The Environmental and Genetic Causes of Autism. The experience of writing those three books allowed me to go into deep study of the primary literature on disease causality specifically on factors that impact our immune systems. Factors that most profoundly influenced my understanding that leads to this Evolutionary Theory of Autoimmunity of course were threefold:
(1) intensive study on the role of cytokine storms in the pathogenesis caused by Ebola infection, specifically the devastating positive feedback loop caused by our immune systems’ responses to the effects of Ebola on our tissues, with releases of cytokines that activate escalating responses to cellular damage
(2) the theoretical and empirical basis for artificial immunization, and its attendant consequences (most often attempted using vaccines), and
(3) being inundated with information on the fact that dozens and dozens of conditions known as autoimmune disorders defy explanation.
Readers of “Causes” will come to understand that autism can often involve a form of ‘innate’ autoimmunity, in which specialized cells in the brain that play the role of scavenger of cellular debris and killers of intruders, like white blood cells, can become chronically activated in the presence of the persistent activating signal of the excitotoxic amino acid glutamate. Ironically, I cannot consume monosodium glutamate, because it brings an onset of migraine headaches. This same amino acid, in some kids’ brains, stays at high levels and rather than ceasing their destruction, say, of unhealthy brain tissue, they go on the attack and destroy both dendrites and neural precursor cells, which release cytokines signalling cellular damage, injury, and death, causing sustained microglial activation. The original cause of the high glutamate levels in the brain of autistics may vary from case to case; it may involve a mutation in a glutamate receptor found on astrocytes, or it could be environmental damage to astrocytes via metals, which preferentially bind to astrocytes, and which have been shown to localize both to the nuclear pore and the cell membrane The literature on traumatic brain injury and stroke is now rapidly filling with how helpful it is to keep microglial cells quiet before they set off a positive feedback of cellular destruction that keeps them chronically activated.
They will also come to understand that many of the adverse events seen in vaccines, and autoimmune disorders, which in most doctor’s minds have root causes that defy explanation, are in fact induced by the presentation of foreign antigens into the human in the presence of an adjuvant. Dr. Yehuda Schoenfeld presented this idea in 2010:
Shoenfeld Y, Agmon-Levin N. 2010. ‘ASIA’ – autoimmune/inflammatory syndrome induced by adjuvants. J Autoimmun. 2011 Feb;36(1):4-8. doi: 10.1016/j.jaut.2010.07.003.
Other studies by Schoenfeld and colleagues have found that ASIA has been reported from nearly every vaccine on the market, and that we may be able to predict who is at highest risk of developing autoimmunity from after vaccines:
Soriano A, Nesher G, Shoenfeld Y. 2015. Predicting post-vaccination autoimmunity: who might be at risk? Pharmacol Res. 2015 Feb;92:18-22. doi: 10.1016/j.phrs.2014.08.002. Epub 2014 Sep 30.
I suspect that in time most autoimmune disorders, which are known to involve adaptive immune systems attacking healthy human tissue, will be seen to ultimately have an evolutionary explanation. Part of it will seen an resulting from adaptive advantage to pathogens in their ability to effect their hosts in a manner in which transmission to new hosts is made more likely. While none of which could possibly be seen as resulting from adaptive advantage to humans, we will explore that in detail later. It must also be remembered that a part of the explanation could also be come to be seen to be the result of mere chance.
MULTIPLE SCLEROSIS is a devastating progressively degenerative disease that, at its heart, involves stripping of axonal sheaths from victims, leaving their nerve cells in the brain and the spinal cord unable to transmit signal due to a lack of insulation and improper ionic balancing associated with nerve impulses. Clearly, whatever has caused demyelination, and demyelination itself, cannot be seen in Darwinian terms to be adaptive to humans. This is about as far as evolutionary thinking has applied. The dismissal of an evolutionary explanation based on the ideas that genes that encode ‘for autoimmunity’ would be quickly removed from a population is only halfway thought through. The more rapidly evolving organism on the scene, of course, is the pathogen itself.
From a pathogen’s perspective, for example, making its host immobile, especially if that host is a member of a social species, will increase the duration and frequency of contact between the affected and the unaffected, leading to increased transmission of the pathogen. Thus, from a pathogen’s fitness perspective, injuring its host, without killing it, would be adaptive. The pathogen itself could do the damage, such as infecting specific types of brain cells and impairing movement. Or, it could have the same effect by turning the hosts’ immune system against itself. We could expect that surface epitopes seen on viral proteins would be on adaptive landscape that could lead them, as a result of chance mutation after mutation, to become increasingly similar to key host proteins. In fact, many pathogens that infect humans have been infecting other primates for millenia, and the common, successful ones causing hardly any or no symptoms at all in their native hosts. But many pathogens do cause disease in their native hosts, including paralysis and death.
Within the human population, then, we could see that whichever proteins that are attacked by the immune system as a result of the development of cross-reactive antibodies would be on an adaptive landscape upon which there would be advantage to moving, random mutation by random mutation, further away from the offending pathogen’s epitope sequence.
We can make predictions on the basis of this theory: genes that encode spefic surface epitopes on pathogens that cause autoimmune disorders should show high rates of non-synonymous substitutions that cause increased conformational similarity to that of their hosts’ matching protein sequence, and hosts should be observed to show high rates of non-synonymous substitutions, as the run away from the offending pathogens’ epitope sequence and structure. A third prediction would be that the hosts’ proteins are much more constrained than the pathogens; they are, after all, already serving an essential function in the host, whereas the surface antigens on a pathogen may be more evolutionarily labile, freer to explore a wider area on its evolutionary adaptive landscape. (For more ideas on the importance of individual mutations vs. rates, see this article).
The first step toward testing this evolutionary arms’ race theory of autoimmunity is to determine which pathogens hold the best-matching epitopes. The second step is to confirm that cross-reactive antibodies exist in autoimmune patients to specific epitopes of interest. This step also happens to be potentially very useful to patients both in terms of helping their doctors understand their disease, and could lead to individualized treatment via immunomodulation therapies. The third step is to measure, whenever possible, the rates of synonymous and non-synonymous substitutions over long enough periods of time to catch the pathogens’ protein racing toward the human sequence, and the matching human protein racing away from the pathogen sequence.
There are attendant logical predictions about variation in the rates of specific autoimmune disorders attributable to specific pathogens across human populations if genetic variation exists within the self-targeted proteins involved in autoimmune disorders.
We have begun these steps, pathogen by pathogen, autoimmune disease by autoimmune disease, at the Institute for Pure and Applied Knowledge, importantly beginning with the proteins that are already known to be targeted by our immune system during autoimmune disease. Our findings thus far are quite promising, and may lead truly useful information, such as the knowledge of which pathogen’s epitopes should excluded from vaccine, for fear of inducing autoimmunity in a preventable manner.
One realization that we have had is that it is possible that the mechanism of pathogenesis, that is, the ways in which communicable pathogens cause disease in humans (and animals), may be precisely they same way they cause autoimmune disease. From the mildest sniffle, or the most raging fever, it seems likely that pathogen/host protein similarities, driven mostly by adaptive evolution in the pathogen, may explain most – if not all – of the symptoms by which we diagnose communicable diseases. That would be very good to know.
My next stop in this journey will be the literature – the massive scientific literature on host/pathogen interactions – to see what is already known on this fascinating topic. You can support our efforts at ipaknowledge.org.
-James Lyons-Weiler, PhD
October 31, 2016
I thank Celeste McGovern at Greenmedinfo for her well-written and informative post,