Microbes and Evolution: The World That Darwin Never Saw

Bacteria had to wait in the wings for many years before they could star in evolution experiments. The new molecular biologists pursued their reductionist methods, while evolutionary biologists, grounded in natural history, did not want to study things they couldn’t even see. It was also difficult to tell bacterial strains and species apart, and many evolutionary biologists were focused on using patterns of similarities and differences to unravel the relationships among organisms. But eventually the field of microbial evolution awakened, and for several reasons. In the 1970s, Carl Woese used differences in DNA sequences to study the evolutionary relationships among bacteria and other microbes, revealing extraordinary diversity beneath their outwardly simple appearances. At the same time, other evolution experiments were pursued with Escherichia coli. Avidians are computer programs that copy their own genomes, and they live in a virtual world that exists inside a computer. But their replication is imperfect, so Avidians sometimes mutate. While most mutations are deleterious, some provide an advantage that allows a mutant to obtain resources and replicate faster than its competitors in that virtual world.

Charles Darwin’s research on the spatial distribution of macroorganisms formed an integral part of his arguments for natural selection and played a foundational role in the burgeoning field of biogeography. Darwin recognized three biogeographic patterns that he described in terms of “great facts” in On the Origin of Species. In summary, these are: (i) environmental conditions alone cannot account for the dissimilarity of flora and fauna among geographically distinct regions; (ii) barriers to dispersal significantly contribute to these differences; and (iii) although the spatial variability in community composition within regions is substantial, these communities remain evolutionarily related. The prevailing view of Darwin and his contemporaries was that microorganisms (microbes) are dispersed globally and able to proliferate in any habitat with suitable environmental conditions. Microbial diversity shifts across the surface of the Earth as a consequence of the three forces highlighted by Darwin when considering plants and animals: the environment, dispersal, and diversification. The author’s laboratory is striving to overcome limitations of time, resources, and sequencing efficiency, by developing a spatial theory of community assembly that quantifies the evolutionary relatedness among individuals (or genomes or genes) within and among sample locations across the landscape. This new framework combines four crucial ingredients: spatial dispersal, by which individuals disperse seeds or move across a spatial landscape; evolutionary diversification, through which genetic novelty is introduced to a community; individual-based processes, so that the discrete nature of individuals is accounted for; and demographic stochasticity, the random nature of birth, death, and dispersal processes.

The complex networks of life are driven by the action of microbes, simultaneously synthesizing and consuming, moving metabolites and energy with relentless, tireless efficiency. Communities of microbes are deeply interdependent, and every bacterial species depends on one or more other species for survival. The consequences of this interdependence are profound, making cooperation and collaboration at least as important as competition in the shaping of bacterial communities. The ability to synthesize a toxic protein raises some vexing challenges not only for the target cells but for the producers themselves. How, after all, can a producer strain keep from becoming the victim of its own lethal weapon? It is because that the producing cell also synthesizes an immunity protein that binds tightly to the killing domain of the bacteriocin, rendering it temporarily inactive. This safety catch will remain on until the bacteriocin is safely inside its intended target cell. The delicate balance between producer cells and target cells plays out in a surprising minuet that has been likened to the childhood game of rock-paper-scissors. A bacteriocin-producing strain emerges and acts as "paper" in our game, defeating the sensitive strain as "paper" covers "rock." Eventually, no sensitive cells survive, and only bacteriocin producers rule. But the emergence of a resistant strain from within the producer population plays the role of “scissors.” Little by little, the population shifts from bacteriocin producers to resistant cells: scissors beats paper. Yet in the absence of producers, the cost of resistance is high and the benefits nonexistent.

This chapter focuses on the genetic and evolutionary aspects of antibiotic resistance, and how an unintended global experiment in selection has taught us about genetic variation and evolution in the bacteria. An antibiotic may fail to reach its intended target, for example, because it cannot cross the cell wall and get inside the bacterium. Genetic alterations can cause previously susceptible bacteria to become antibiotic resistant. This is a major reason why many antibiotics are ineffective against gram-negative bacteria, and why pharmaceutical companies have devoted great research efforts to developing variants of successful antibiotics with different activity spectra. The genetic pathways by which bacteria can evolve antibiotic resistance can be broadly divided into two classes: (i) those where genetic alterations reduce the effective interaction between the antibiotic and its target in the bacterial cell and (ii) those where genetic alterations reduce the chemical effectiveness of the antibiotic molecule itself. An important aspect to consider in the evolution of antibiotic resistance is the source of the genetic alteration responsible for the resistance. The global significance of horizontal gene transfer in providing the raw material for bacterial evolution in general, and antibiotic resistance specifically, is one of the major lessons we are still learning from our use of antibiotics. One way to evolve antibiotic resistance is by mutating the gene coding for the target of the antibiotic. A resistance phenotype can also be generated by any mechanism that inactivates the antibiotic.

Phage represents the ultimate mash-up of selfish genes producing hopeful monsters—that is, groups of genes that promote their own propagation but which can often result in dramatic changes for the host that acquires them. The author and his group accumulated enough metagenomic data to determine the scale at which the global virome was contributing to the metabolic potential of their hosts. Compared to the microbes, the phage were enriched in genes with functions such as nucleic acid metabolism. More striking was the fact that the viromes also carry many genes associated with other unexpected aspects of metabolism (vitamin, cofactor, cell wall, and capsule synthesis), as well as genes for virulence factors, stress response genes, and chemotaxis genes. The phage-driven evolution of microbial hosts extends beyond the phage themselves. It was shown that many of the changes in microbial metagenomes were actually responses to predation by phage and protists. A most exciting recent development has been the discovery that host cells have rapidly evolving sequences, known as clustered, regularly interspaced, short palindromic repeats (CRISPRs) that serve as a bacterial immune system to prevent phage infections in bacteria. The phage-microbial evolution is much more rapid and dynamic than Charles Darwin could have imagined. Phage transport genes all over the world and change the evolutionary trajectory of the microbes they infect by introducing genes that completely change the phenotypes of their hosts.

Everything that a cell does within a biofilm can have immediate consequences for its neighbors. This includes the most fundamental action of any microorganisms (microbes): cell division. It is argued strongly-and controversially-that we need to find ways to regulate human population growth in order to avert environmental tragedy, an argument that may be more likely to be heeded now than it was originally. Many pathogens for example bacteria, rely on both quorum sensing and secreted products to inflict harm upon us. This includes the secretion of enzymes that kill host tissue by Salmonella enterica (typhoid fever) and Pseudomonas aeruginosa; toxin production by Bacillus anthracis (anthrax) and Vibrio cholerae; and the widespread production of compounds that break down antibiotics, such as the β-lactamases that destroy penicillin. The chapter focuses on microbial societies and the potential for their own microtragedies. Traditional antibiotics act by killing or stopping cell division, and resistant that can grow in the presence of the antibiotic rapidly replace the original susceptible strains. Resistant mutants that re-evolve secretion can promote the growth of susceptible cells around them. And, more than this, the susceptible cells do not pay the cost of secretion, which can put the resistant strain at a competitive disadvantage. At least in principle, this can slow the rise of antibiotic resistance. By recognizing that microbes rely on both sociality and altruism to cause infection, a novel strategy for treatment is revealed.

The pseudogenes must be extremely rare in bacteria compared to plants and animals because bacteria grow rapidly and have small genomes. Work was carried to develop some clever genetic tricks for identifying such "host specificity" genes in Typhi. After years of hard work, it seems that the reason for restriction of Typhi to humans was not because of the acquisition of new genes but because Typhi had mutations in many different genes scattered around the chromosome. In the mid-1990s it became possible to determine the entire sequence of bacterial genomes. The genome sequences of different types of bacteria provided surprising insights into genes not known earlier. But the most striking insights came from comparisons of the genome sequences of closely related bacteria. In contrast to Salmonella Typhimurium, the human-specific pathogen Salmonella Typhi was one of these organisms with a large number of pseudogenes. In addition, many of the pseudogenes were in parts of the genome that had been identified as containing mutations that restrict host specificity. It had to be known whether the pseudogenes are responsible for restricting Typhi to the human host or whether the accumulation of pseudogenes is simply an accident. The genome sequencing studies clearly indicated that pseudogenes occur in bacteria, and the results showed that pseudogenes can influence the host range of bacterial pathogens.

The family Enterobacteriaceae, whose members are often referred to as the enterics, comprises several bacterial species that are known because of the diseases they cause in humans and/or in animals or plants of economic importance. This chapter presents four possible genetic scenarios to account for the differences that exist between the closely related bacterial species, Salmonella enterica and Escherichia coli. It also illustrates how variation in gene regulatory strategies can result in distinct bacterial behaviors. The comparative studies discussed in this chapter have revealed two distinct aspects that distinguish individual enteric species. First, organisms often vary in the environments in which the polymyxin B resistance proteins are produced, and this is often due to changes in the amino acid sequence of a regulatory protein. Second, two related organisms can synthesize the resistance proteins in response to the same signals but achieve dissimilar resistance levels as a consequence of the particular architecture that individual species utilize to promote expression of the resistance genes. The types of genetic differences that distinguish closely related species can be quite different depending on whether one is considering bacterial versus eukaryotic species.

Charles Darwin really could have used bacteria, organisms for which you can really measure mutation rates. Astronomical populations of bacteria can be grown with and without stress and their mutation types can be characterized. In fact bacteria helped Lederberg, Luria, and Delbrück show that new mutations can form without any required "stress." Mutants arise continuously in perfectly happy bacterial populations. With great excitement Darwin showed the data (obtained using bacteria) suggesting that Lederberg, Luria, and Delbrück had missed something-Darwin might have been closer to the mark than people realized. The key to evaluating Cairns claims lies in the details of his selection conditions. Small-effect mutations are extremely common, and they can contribute serially to very fast strain improvement. It seems that the mechanisms for DNA replication and repair have evolved to be most effective at preventing mutation types with large phenotypic effects. All of laboratory bacterial genetics use the same principle strong selection. The trick is to block all parent cell growth (no new mutations) and prevent growth of those annoying frequent small-effect mutations. Darwin’s idea of stress-induced mutation may be wrong, but natural selection can take on breathtaking power when common small-effect mutations are allowed to contribute. These effects are revealed by bacterial populations. Under selection, the high speed of genetic adaptation is easily mistaken for an increase in mutation rate—maybe even Darwin underestimated the power of selection.

Bacteria do not reproduce sexually; they have one set of genes rather than two, and their cells simply divide-they never fuse with each other like eggs and sperm. Short pieces of the DNA that genes are made of sometimes move from one bacterial cell to another, and this can randomly mix their different versions of genes if it happens often enough, just like real sexual reproduction does. For a long time microbiologists assumed that these parasites were a bacterial way of having sex-that is, that the evolutionary forces responsible for our sexual reproduction had indeed acted on bacteria too, producing ways to move DNA between cells. Nutrients and temperature conditions are usually very different from those the bacteria experience in their natural environments, partly because they are more convenient and partly because one really does not know what the natural conditions are. But studying bacteria under such unnatural conditions makes it very easy to misinterpret what they do. Most bacteria use nutritional signals, just as the DNA-equals-food explanation predicts. Just as people who read only the newspaper headlines think every lottery ticket is a big winner, a scientist who reads only the research papers about bacterial genome sequences could easily conclude that every new gene is an improvement. In either case, the discovery that bacteria do not have sex means that a big key to the puzzle of the evolution of sex will come from finding out why bacteria need so much less mixing than plants and animals.

The author’s methods for direct visualization of individual mutation and recombination events in living Escherichia coli bacteria are based only on the shrewd use of a natural DNA modification (methylation of A in the GATC sequences) and of two fluorescent proteins. A special protein involved in the process of DNA error correction (called mismatch repair) binds exclusively, extensively, and stably to uncorrected DNA copy errors, forming a fluorescent focus in the living cell. In recombination, a fluorescent version of the E. coli SeqA protein, which binds exclusively the hemimethylated DNA (a DNA duplex with only one strand methylated), allows us to monitor the integrity of permanently hemimethylated DNA over an unlimited number of generations. The molecular phylogenies of preexisting mutations from genomes of many sequenced natural isolates of E. coli suggest that the acquisition of a mutation at any locus in the genome is about 100 times more likely to occur by recombination than by de novo mutation. The results of a study conducted by the author, appeared as expected from the E. coli– Salmonella enterica serovar Typhimurium crosses: when both partners had mutator histories, genetic recombination dropped 10-fold in mut+ cells but not in cells with mut defects; when only one partner had mutator history, recombination dropped only 3- to 4-fold; but when neither had a mutator history, there was no effect on recombination.

This chapter talks about the life and times of the ocean’s smallest photosynthetic cell. The author says that it is difficult to describe the thrill of studying Prochlorococcus. Like most scientific advances, the unveiling of Prochlorococcus involved new technologies, diverse approaches, teamwork, and luck. It became clear immediately that although they shared the "signature" characteristics of Prochlorococcus, MED4 and SS120 were not the same: MED4 could grow at high light intensities that killed SS120, while SS120 could grow under extremely low-light conditions that could not sustain MED4; i.e., the cells were adapted to the light intensities found where they were captured. For every cell that is produced, there is another that is eaten by small predatory cells that must rely on others for their food. This keeps the Prochlorococcus population in check and begins the flow of energy through the marine food web. Since they reproduce by making identical copies of themselves, we know that at any moment in time there must be lineages of identical Prochlorococcus cells in the oceans. Prochlorococcus can photosynthesize, and thrive through diversity, and their federation can adapt.

Worker attine ants carefully tend the fungus and forage for substrates to use as nutrients to support its growth. In exchange, the fungus serves as the primary food source for the colony. Functionally, this is similar to human agriculture, but instead of growing plants, the ants grow fungus. The foraging activity of these ants is so prodigious that they are one of the most dominant herbivores of Neotropical ecosystems. Pseudonocardia are members of the Actinobacteria, a group well known for their ability to produce potent antibiotics. Indeed, the majority of antibiotics used in human medicine are derived from Actinobacteria. So, it is perhaps not surprising that through natural selection fungus-growing ants “discovered” the benefit of obtaining antibiotics from these bacteria. This relationship, however, not one-sided, as Pseudonocardia benefits from its symbiosis with the ants. Studies indicate that over their long evolutionary history, attine ants have been continuously threatened with famine induced by Escovopsis infections of their fungus garden; to help mitigate this threat, the ants have been employing antibiotic-producing bacteria for millions of years. The fungus garden serves as the external digestive system for the ants; the garden converts leaves into energy for the ants. It has long been assumed that nutrient provisioning to the ants from the garden is only mediated through the ants’ fungal mutualist. Through the formation of symbiotic associations with their fungus and bacteria, the ants obtain access to the metabolic capacity of these microbes, including the capacity to degrade plant biomass and produce antibiotics, respectively.

By the early 1990s, it was becoming clear that carriage of H. pylori increased risk for peptic ulcer disease and for gastric cancer, where both are consequences of the Helicobacter pylori induced gastric inflammation, although in separate ways. In recent years, using more sophisticated analytical techniques, there has come greater support for the notion that H. pylori has colonized humans since before the out-of-Africa migrations of about 58,000 years ago, and that as humans migrated to all parts of the world, they brought their H. pylori strains with them. It has been hypothesized that for H. pylori and humans, there are actually a series of equilibria, nested in one another, that create the governing boundaries at each level. Postulated benefits for humans for early-life carriage of H. pylori include resistance to colonization by exogenous pathogens (through manipulation of gastric pH and gastric immunity, as well as direct competition), as well as close regulation of metabolism, through gastric leptin (5 to 10% of body total) and ghrelin (60 to 80% of body total). In summary, H. pylori has evolved over a long period of time as a highly interactive member of the human gastrointestinal microbiota. There is extensive evidence that H. pylori coevolved with its human hosts, enabling its nearly universal gastric persistence. The interaction had little or no cost (and possible some benefit) to its early-in-life hosts, but also conferred certain late-in-life disease costs, including gastric cancer.

As Charles Darwin has taught us, evolution happens through natural selection. Small things, which in any given instance might not appear significant, can grow to have great importance when the environment changes. Bacteria could be used to clean up the messes humans made of their environment if the conditions were right for them to grow. Microorganisms are amazing. They invented the metabolic machinery that sustains all life today. Indeed, the cellular compartments where energy is made in the human body and plants are nothing more than the remnants of ancient bacteria that were engulfed by and entered into symbioses with other cells long ago. But the vast majority of bacteria play important roles in sustaining both human health and the health of our planet. Conventionally, when structurally complex organic molecules are identified from any given sample, they are interpreted as biomarkers of cyanobacteria—the microorganisms that invented oxygenic photosynthesis, the conversion of water to oxygen in sunlight—and hence, they are assumed to be biomarkers of oxygenic photosynthesis itself. The remaining challenge will be to identify new biomarkers to help constrain the timing of major metabolic breakthroughs, such as oxygenic photosynthesis, and shed light on how these breakthroughs came to be. In anticipation of the continued appearance of new clues, the development of increasingly powerful technologies, and the aid of talented colleagues, the author has no doubt that the pursuit of the molecular equivalents of billion-year-old Rosetta Stones will provide enough excitement to last my lifetime.

It is a known fact that bacteria (and archaea) are genetically, phylogenetically, and physiologically very diverse. Microbial microevolution has been and is still being analyzed in very elegant laboratory experiments in which microbes are allowed to mutate, adapt, and evolve in test tubes under very stringent and therefore reproducible and adjustable conditions. The "Evolution Canyon" (EC) system in Israel turned out to be a suitable sampling place for such microevolutionary studies. For more than 20 years, adaptation and speciation of macroorganisms have been explored in EC, with Drosophila flies and wild barley being two of the most prominent model organisms. Compared to proteins and genomes, the cell membrane is hardly in the focus of mainstream research. The potential research topics appear to be endless. The author focuses on addressing some of these issues, and expects that comparative genome sequence comparisons will soon aid one in understanding how these evolutionary changes are implemented on the molecular level of genes and their expression. It is perhaps too naïve to expect that the observed traits indeed evolved precisely in those sites from where the Bacillus simplex bacteria have been isolated, taking into account that there are probably thousands of such east-west-directed canyons on the globe and that sporulating bacteria like bacilli easily migrate with the wind from continent to continent. But most probably, such sharp microclimatic contrasts in immediate proximity reinforce within-species evolutionary splits, and therefore EC represents a beautiful site to study microevolution in natural habitats.

Adaptive radiations are in fact commonplace; indeed, the vast majority of life’s diversity appears to have arisen through successive adaptive radiations, the most famous perhaps being the radiation that took place approximately 530 million years ago during the Cambrian period (the Cambrian radiation) and from which arose the major groups of complex animals. The author’s own interest in the emergence of diversity began with the common plant- and soil-colonizing bacterium Pseudomonas fluorescens and its propagation in unshaken test tubes (microcosms). Necessary to bring the study of evolution into line with advances in genetics and molecular biology is knowledge of the mutational origins of new phenotypes, and, critically, understanding of the connection between genotype and phenotype in sufficient detail so as to explain precisely why natural selection favors one type over another. Recent discovery of repeated molecular evolution—not just in laboratory populations of microbes but also in natural populations of insects and plants—has brought renewed interest to this possibility, but mechanistic insight of the hard-proof variety has been lacking. Indeed, the author’s recent work on the genetics of wrinkly spreader diversity provides evidence that genetic architecture does indeed contribute to the repeatability of evolution by biasing the molecular variation presented to selection. Among the most exciting of contemporary issues is genetic architecture—the genotype-to-phenotype map—and its influence on the evolution of populations. With the emergence of improved understanding of this map-its definition, its formulation, and the constraints it imposes-rules by which the outcome of phenotypic evolution might be predicted may just arise.

Charles Darwin and other naturalists of his time traversed the globe to survey the diversity of plants, animals, and insects and their distributions around the world. They combined these surveys with their contemporary understanding about the history of Earth to develop an evolutionary model for the origins of species through natural selection. One hundred fifty years later, microbial naturalists are uncovering a whole new world of diversity that will lead to a sea change in our understanding of evolution. Initial surveys of microbial diversity suggested that microbes did not fit this conventional model of evolution. Microbial naturalists were unable to identify mechanisms that would disconnect microbial populations from one another. Our initial analyses suggest that primary differences come not from natural selection by abiotic environmental variables between locations, but from interactions between microbes in each location, specifically between Sulfolobus and its viruses and other microbial parasites. The study of microbial evolution lags behind that of macroorganisms, and it seems that there are relatively few microbiologists choosing microbial naturalism. It is clear that the new generation of microbial naturalists who set out to explore the distribution of natural diversity now wields not only the strong lens of genomics but also the power of molecular biology and genetics, which will enable them to discover the fundamental laws of evolution that apply to all of life on Earth.

In this chapter on postphylogenetics, the author narrates how his lab detoured into the physiology and molecular biology of halobacteria (properly haloarchaea). Resistance to antibiotics, on the rise in hospitals around the world, was shown by Japanese microbiologists to be due to the transfer between them of small circular DNAs, called plasmids, bearing genes for such resistance. There is continuing debate about the significance of gene exchange, which is called either lateral gene transfer (LGT) or horizontal gene transfer-terms of identical meaning, but like personal computers and Macs, each with its own vigorous defenders. Phylotyping had been until now pursued as if its primary goal were the enumeration and identification of microbial species, as their distributions and abundance are influenced by biotic and abiotic forces.

Nature walks aside, studying the natural history of microorganisms is uniquely challenging for a host of reasons, one of which is that there are many different microbial species—or at least groupings that look and act like species. The author saw metagenomic data in the form of genes taken from seawater to identify marine bacteria and was fascinated to discover that although similar genes clustered to form groups, no two were identical. Bacterial and archaeal cells do not pair, recombine gametes, and undergo meiosis. Among others, two observations are critical to understanding the arguments about microbial speciation. The first is the importance of the vast size of microbial populations in nature, and the second is the subtler concept of neutral variation and sequence space. Microbes are the frontier beneath our feet, yet it has taken a long time for Charles Darwin’s ideas to approach their full potential for interpreting microbial diversity. The natural selection is probably very efficient at improving microorganisms to make them nearly ideal machines for reproducing in natural environments, but at the same time the age and size of these populations, the constantly changing environments they must adapt to, and their varied repertoires of DNA acquisition mechanisms make them extraordinarily diverse.