Of All That Science Can See, What it Can’t Blinds Us to the Rest

We have hit a roadblock when it comes to finding the full range of microbes that surround us and inhabit us. From here on out, progress can be made only by innovation.

We live in a microbial world, teeming with organisms invisible to the naked eye. Ever since the ‘eureka’ moment of the Dutchman Antoine van Leeuwenhoek, who with his homemade microscope saw microbes (or ‘animalcules’ as he called them) for the first time, our knowledge of the microbial word has only grown. The development of the germ theory of disease in Europe reached a cusp in the 19th century when systematic investigation by several microbe hunters, including Louis Pasteur and Robert Koch, definitely identified certain microbes as the causative agents of some diseases. Thus, these invisibly tiny creatures were not mere curiosities but of central importance to our health and well-being. So where do we stand in terms of knowing our microbial neighbours?

Today, we are witnessing something unusual. While physicists are using large telescopes to probe farther reaches in the universe, and are discovering minuscule particles using humongous machines, biologists are discovering more and more tiny organisms every day, occasionally figuring out what these organisms might be doing on our planet (and even beyond!).

Discovering and classifying organisms is an old field. Even before Charles Darwin came up with the idea of evolution, which has since served as the foundation on which the classification of organisms is interpreted, a man called Carolus Linnaeus collected tens of thousands of plants, animals and sea shells, and based on their shape developed a system of classifying them. This system is still being used today. When someone tells us that humans are technically Homo sapiens, they are in fact referring to the Linnaean classification system, where Homo represents our genus and sapiens our species. Among our predecessors are Homo erectus, who are not-quite sapien Homos. A mouse would not even be a Homo and instead be a Mus. All Homo sapiens are very closely related to one another, and are evolutionarily related to the Homo erectus, but less so to the Mus, and so on. One could classify macroscopic organisms into rather sensible divisions by just looking at them. But microbes were a problem.

What shapes us?

As more microorganisms, especially bacteria, were observed under microscopes over a century ago, it became clear that shapes were not going to take us very far in classifying them. These creatures looked like rods or spheres or springs and little more. It was also not clear whether the same type of microorganism would change shapes depending on the stage of its life, like a werewolf. Other chemical observables, for example the ability of an organism to ferment cereals into beer, helped clear the scene. But questions persisted. Would the equivalent of the Linnaean classification even work for microbes? In other words, were they a form of the sort of life we are used to seeing around us?

Fast forward to the present and we know that it does, but with some important caveats that we’d rather not get into.

A status quo that had emerged was the division of life into two broad classes: eukaryotes and prokaryotes. Eukaryotes included organisms whose cells contained a nucleus, a closed body within the cell that housed the DNA. This class of organisms included yeasts, plants, insects, animals, birds and essentially all macroscopic organisms. The bacterial cell does not have a nucleus and was deemed to belong to the second class of organisms, the prokaryotes. This classification, which was enabled by microscopic examination of cellular structures, had overturned the earlier belief that bacteria were some kind of plant, and different from us and animals.

Whereas the shape and other observables are often useful in classifying things, they are really one step removed from evolution’s real playing field: the genetic material, or the DNA (for most organisms). But for more than two centuries since Linnaeus’s work, robust classification based on DNA content was beyond us. In the 1970s, our newfound ability to sequence DNA changed all that.

Carl Woese and George Fox, at the University of Illinois at Urbana, compared sequences of a gene called the 16S ribosomal RNA, found in bacteria, and its relative, the 18S ribosomal RNA, found in eukaryotes, to show that there were in fact three major kingdoms of life. The sequences of most genes change fairly rapidly over evolutionary time, thus making sequence comparisons among distant organisms very difficult. But there is light at the end of the tunnel, and the 16S/18S ribosomal RNA sequence was just that. It is a part of the protein synthesis machinery, and happens to change very slowly over evolutionary time. As a result, it is recognisable from its sequence with a high degree of confidence across most of the diverse life forms on Earth today, and therefore an ideal reference for comparison.

The microbial census

Woese and Fox found that there were in fact two types of prokaryotes: the bacteria and the archaea. Many archaea were later found to be members of many extreme environments, and to have properties that have many similarities to the cells of the complex eukaryotes. Let us not get into the whorl of microbial classification here, except to say that the use of 16S ribosomal RNA sequencing for identifying and classifying bacteria and archaea is still a method of choice today. It is also the basis of what is called the microbial census, which in many ways is not very different from a human population census.

Laboratories from around the world have sequenced the 16S ribosomal RNA gene of bacteria from many different environments. These experiments had for long been performed on bacteria isolated from the environment and then grown in the laboratory. This is problematic primarily because most bacteria in our natural world cannot be grown in the laboratory and therefore cannot be catalogued by these approaches. More recently, however, thanks to the ideas of a man called Norman Pace in the 1990s, we have started isolating bacterial DNA directly from the environment for sequencing. These efforts have clearly expanded our catalogues of microbial diversity.

In all these studies, 16S ribosomal RNA genes are sequenced and classified, based on similarity, into groups called operational taxonomic units (OTUs). The number and diversity of OTUs found in an environment is often considered a measure of the richness of the microbial community residing there.

What does the census look like today? Using a reference dataset of nearly 1.5 million ribosomal RNA sequences available in public databases, Patrick Schloss and his colleagues from the University of Michigan, Ann Arbor, showed that we have sampled up to 200,000 bacterial OTUs, with archaeal sequences being a poor cousin by numbers. This is small compared to our estimates that there might be up to a billion bacterial species on the planet. Nevertheless, with credit to the acceptance of Pace’s ideas, over 80% of these sequences were obtained directly from the environment without resorting to growth in a laboratory.

To find less of the same

The microbial census highlights two inadequacies in our knowledge. One is that the current methods of discovering new bacterial species, which may be biased in favour of the more abundant members of a community, may be nearing saturation. In other words, the more we sequence 16S ribosomal RNA genes the way we do today, the less likely that we will discover new OTUs. This calls for new approaches to detect rare members of a community. These approaches might involve combinations of new computational methods to analyse experiments in which the entire genetic material of a bacterial community, and not just the sequence of the 16S ribosomal RNA, is sequenced. Newer techniques in which the DNA of a single cell, and not a pot of millions and billions of cells, is sequenced might also help. In short, we might have hit a roadblock in terms of knowing our neighbours and are at an inflection point from which progress can be made by innovation.

Second: we seem to discover more bacteria from certain types of communities – for example, those residing in the human body more than those that reside in very rich environments such as the soil. We had always thought that the nutritionally sparse but diverse soil will host bacteria with very large genomes, enabling the bacterium to utilise these diverse nutrients to make ends meet. A few days ago, a paper deposited in the preprint archive BioRxiv (pronounced “bio-archive”), showed that a highly abundant soil bacterium, poorly covered in ribosomal RNA gene databases, carried a small and streamlined genome, something that had been seen previously in abundant oceanic microbes that deal with a narrow range of limiting nutrients. This highlights our inadequate understanding of microbial niche management in the soil environments that we see everywhere.

The microbial census indicates that we have made significant progress in knowing our microbial neighbours but might have hit a roadblock. Such severely biased knowledge also limits our understanding of the myriad contributions that microbes make to the functioning of our ecosystems and so on our lives.

Aswin Sai Narain Seshasayee runs a laboratory researching bacterial biology at the National Centre for Biological Sciences, Bengaluru. Beyond science, his interests are in classical art music and history.