This one is a treat. An opportunity to blog about my ideas on science! It seems that most of my efforts these days are focused on BiOS, patent transparency and innovation strategies. Science is still an important part of my life, but my dismay at the way it has been co-opted and made less relevant to society has left a bitter taste.
Still, there are new fields that are breathtaking in their implications (to me at least) and which do not lend themselves to being ‘owned’, but – at least at this stage in their development – rather shared.
The single most exciting development in the biological sciences to occur in my lifetime is the idea that microbes are not only ubiquitous but that they may be the most important component that drives the evolution of macro-organisms.
In fact, I’d venture to say that multicellular eukaryotes only exist in nature as complexes of organisms in which microbial genomes are critical, essential contributors to the fitness of the overall ‘individual’ (which itself needs redefining).
Back in September of 1994 I gave an invited presentation at a Symposium at Cold Spring Harbor sponsored by Perkin Elmer Corporation: “A Decade of PCR“. The symposium was only a couple of days, was a celebration of the impact and future of PCR on life sciences, and featured Jim Watson, Kary Mullis, and a number of other prominent speakers. I was given the task of talking about Agriculture, Environment and the Third World. Rather dauntingly broad marching orders. But I decided that I’d try something fun out on the audience, which was a pretty substantial group of scientists.
While setting the scene with our observations about the complexity of agricultural systems, I wanted mostly to talk about ‘Hologenomics’, a term I coined and defined for that seminar. As with most of my ideas, I’ve never written it down, but I have spoken about it alot, and thought about it even more. (This comment is a response to a quote from my PhD thesis advisor, who wrote in my letters of recommendation that “Jefferson thinks incredibly quickly, but speaks even faster”….sigh).
For years I’ve been studying the metabolism of glucuronide conjugates by Escherichia coli, and many other microbes associated with vertebrates. Vertebrates, including many humans (!) have a remarkable mode of turning over small molecules from their systems. Most small molecules, including steroid hormones, vitamins, xenobiotics (including most pharmaceuticals and dietary secondary metabolites) are excreted from the body as conjugates of glucuronic acid known as ‘glucuronides’ .
Glucuronic acid is simply glucose in which the 6-carbon is in the form of a carboxylic acid or carboxylate. The conjugates are typically (but not always) beta-1-O-linked glycosides of the aglycone. This conjugation is done in many tissues of the body, including of course the liver, but also epidermis and almost everywhere we look. The conjugates so generated tend to be much more water soluble than the aglycone (for instance, testosterone), and thus can more readily be transported in the aqueous circulatory system, where they can be transfered (recognizing the highly stereospecific glucuronic acid moiety) to an excretion pathway. The conjugates find their way to the kidneys for excretion in urine; the pancreas/bile ducts where they are dumped into the upper intestine; and the axils where they are excreted in apocrine and merocrine excretions (eg. sweat).
Turns out that the majority of steroid hormones are excreted this way, into the bile, where they end up feeding the enormous diversity of bugs in the upper and lower intestine. E. coli is just one of the myriad microbes there, and not even the most prominent one.
But it seems that the excretion of the hormone conjugates (and you can substitute almost any word for ‘hormone’ there, including ‘drug’, ‘toxin’, ‘metabolite’ etc) does not see the story end. Rather, the microbial populations have evolved the ability to transport these compounds selectively (in our own work, through a glucuronide transporter we call glucuronide permease, encoded by gusB locus of E. coli), where the aglycone (the steroid for instance) is cleaved off (by glucuronidase, encoded by gusA) and the sugar (glucuronic acid) metabolized by the microbe. The freed aglyone is then available for re-absorption in the gut.
Thus, thousands of compounds actually cycle through the intestine of vertebrates. Conjugated in the body, transported by circulatory system to a distal site of excretion, imported into microbial systems, cleaved to free the aglycone, which is then reabsorbed into the body. This has been called enterohepatic circulation, back when it was thought that the liver was the only site of glucuronic acid conjugation. I’ve read that as much as 65% of the circulating testosterone, for instance, has already passed through an enterohepatic circulation route.
Glucuronidation, and sulfation to a lesser extent (though this depends on species of molecule and species of macrobiont), are used on most steroids and most drugs to excrete. Hence the cleavage and re-presentation of the aglycone by the microbial populations are critical to controlling the level of these compounds.
Well these compounds include the entire range of steroids that have seminal impacts (excuse the pun) on reproduction, and reproductive fitness, including fertility, fecundity and mate choice. Estrogens, progesterone, testosterone, estradiols, all of them are excreted as glucuronides, and reabsorbed after microbial cleavage. Even the pheromones excreted from the armpits (or legpits…axils…) are generally non-aromatic. Some would say VERY non-aromatic. Perhaps better to say, they are excreted as water soluble, non-volatile compounds. It is the skin microbes that cleave the compounds to release the volatile steroids, such as androstenols, which impact on mate selection and population behaviour of the macrobiont (the big animal we see, for instance the human, mouse, fish, etc)..
Ok, so what is it about hologenomics and evolution? Well any sensible student of evolution will note that the traits that I mentioned, which are hugely impacted by steroid balance, which itself is determined (!) by the turnover of the glucuronide conjugates, are the very core traits that are key to natural selection, namely fertility, fecundity and mate choice.
Holy Cow! The evolutionarily most important traits of a mouse (or human) are encoded by bacteria!
And more importantly, by the balance of extraordinarily complex populations of bacteria in diverse settings. As anyone who has read this far (ie no one) would know, as we get better at looking for microorganisms we get much better at finding them – and it seems that the vast majority of the genomes associated with a human are non-human. Or are they? They are microbial (including eubacteria, but also protists, fungi, archae perhaps) but why are they not ‘human’?
They are causally associated with the behaviour and performance characteristics that have allowed us (or indeed any macrobiont) to persist and flourish under natural selection. So really, that means they are integral to ‘human’. In fact, I would argue they’re as ‘human’ as ‘we’ are. You see, we’re used to seeing our big sloppy nucleus and chromosomes and thinking ‘Yup, that’s the human genome; the book of life’…well, it seems that that’s only the scaffold! The real genome is what I call a hologenome. The apobiont (that would be what looks like me, if I had NO microbes associated with me) is not a holobiont (a complete organism) until the population structure makes it so. So the human genome hasn’t been sequenced, just the parts in a bag that replicates slowly. The little bags that replicate, migrate and reassort rapidly, and which clearly have mission critical functions, are only now being discovered.
And their importance is not yet appreciated. And its not just animal! All macrobionts (big multicellular organisms), including plants, are also a pastiche, an amalgam of many genomes. And so a rice plant is not really a rice plant in the absence of the microbial populations that allow it to perform in the environment.
This concept has big implications for our world view. For science. For medicine and agriculture. It is the tip of a very big iceberg. Its has been said in the last ten years by many scientists that most of the microorganisms in the world are unculturable (some would say that about humans too). Its true that as we begin our exploration of the microbiome we find vast numbers and diversity of microbes that have never been cultivated on media in the laboratory. The likelihood is that they never will. I guess that’s part of why I think they are a part of a hologenome that contributes to the overall natural selective fitness of the holobiont (and perhaps a holosystem?), and can only be thought of in that light.
So if we drop a mouse in a blender, we get (besides a visit from the site bioethics committee) more ‘bacterial’ than ‘mouse’ genomes. By a factor of nearly a hundred! This is the new wisdom. But its wrong. We definitely get more ‘microbiont’ than ‘macrobiont’ genomes. But the mouse is not the macrobiome. The mouse is the sum of these genomes – the holobiont expressing the hologenome – and it is that sum that contributes to the success of the mouse (not of course the blender mouse itself, who would not look at its role as a big success) under evolutionary selection.
And what gets very exciting about this is the following: If I were a macro-genetic-engineer (eg. teleologically speaking) I’d design most of the dull, workaday, scaffold functions to be encoded by slowly evolving, slowly replicating packets – ie nuclear genes. But I’d put the really critical, environmentally responsive functions in packets that could amplify quickly and specifically, could be traded (or not) through well crafted horizontal transfer mechanisms, which could replicate, and hence evolve, very rapidly. In short, I’d put the really cool important stuff in the microbial genomes, and leave the big macrobiont nucleus as a pretty pedestrian framework.
So are we really doing a great job yet at studying human (or mouse, or rouse – isn’t that the singular of rice?) genomics? Nope. Not until we think of the suite of genomes – the apogenomes and their aggregation into a hologenome – as the performance unit of natural selection. A nd therefore of the repository of the logic of life.
It would mean that the logic of ‘eukaryotic’ genomes will be incomplete, maybe even dead wrong, in the absence of the full appreciation of the impact of populations of microbes, which are lynchpins.
So I’m wondering, how do we set up experimental systems to prove this beyond a shadow of a doubt? How do we show that evolution works at that hologenome? Or doesn’t.
I’m rather optimistic that the glucuronide system may afford a model. And the new tools of massive high-throughput environmental sequencing will be a platform. I will be musing on this rather alot.
So let’s keep talking about hologenomics, and even its practical implications.
In a talk in South Africa at the 14th Conference of the South African Society of Biochemistry and Molecular Biology in 1997, in Grahamstown South Africa, I called this Ecotherapeutics.
The idea that adjusting the balance and constitution of the microbial populations would have huge repercussions on disease and health. Both plant and animal. And that this could constitute the next revolution in life sciences, in which the interventions could be intrinsically ‘un-ownable’, context dependent and robust.