1
The Discovery of Archaea

Patrick FORTERRE1,2

1 Institut Pasteur, Université Paris Cité, France

2 Université Paris-Saclay, Gif-sur-Yvette, France

1.1. Introduction

The discovery of archaea by Carl Woese and his collaborators was one of the greatest breakthroughs in 20th-century biology. This finding went largely unnoticed by many biologists, particularly those focused on eukaryotes, and remains unknown to nearly the entire general public. By unveiling a previously unknown segment of the living world, the discovery of archaea has, for the first time, allowed us to consider the history of life on Earth in its entirety. This discovery has spectacularly revived debates on the nature of the last common ancestor of current living beings and the origin of eukaryotes (our origin). Challenging the prokaryote-eukaryote dichotomy was one of Carl Woese's major endeavors. This struggle appears far from over, especially as the emphasis is once again on phenotypic similarities between archaea and bacteria at the genomic level. The scenario favored by most biologists continues to reflect the classical view that eukaryotes emerged from the association of different prokaryotes.

These reflections recently led me to write a review titled "Carl Woese: Still ahead of our time" on the 10th anniversary of his death (Forterre 2022a). In this context, it seems particularly important to revisit the history of the discovery of archaea to understand the intricacies of the major debates which continue to captivate 21st-century evolutionists.

1.2. Prokaryotic-eukaryotic dichotomy

In the late 1960s, molecular biologists imposed the division of the living world into two major groups: eukaryotes and prokaryotes (Sapp 2005). This classification was long considered a reflection of the history of life. The simpler prokaryotes were thought to have appeared first, followed much later by eukaryotes. The term "prokaryote" ("before the nucleus") reflects this idea. Quickly, the terms "prokaryotes" and "bacteria" became synonymous for most biologists. The term "prokaryote", which originally referred to a type of cellular structure (absence of a true nucleus), eventually took on an evolutionary connotation: all bacteria were believed to be close relatives, descending from the same ancestral "prokaryotic" bacterium. Prokaryotes were considered "primitive" compared to eukaryotes (Stanier and Van Niel 1962; Whittaker 1969). Most biologists accepted that eukaryotes descended from prokaryotes similar to current bacteria, possibly through a symbiotic merger of several different bacteria (Margulis 1975) (Figure 1.1(a)).

For many biologists, the molecular mechanisms of prokaryotes were assumed to be similar across the group, and studying one bacterium at the molecular level was thought sufficient to understand them all! Consequently, discoveries made on the model bacterium Escherichia coli were believed to be applicable to all bacteria. This view, still tacitly accepted by many biologists and taught in many universities, proved to be false. The work initiated in the late 1970s by the American molecular biologist Carl Woese revealed this (for reviews and books summarizing the history of this work, see Woese (1981), Forterre (2007, 2016, 2022b), Sapp and Fox (2013) and Quammen (2018)).

In the early 1970s, Carl Woese set himself the goal of constructing an evolutionary tree of life that would include microorganisms, bacteria and protists. Until then, microorganisms had never been integrated into a comprehensive universal phylogeny, as these were primarily based on the fossil record and the comparison of phenotypic characteristics. However, this approach was only feasible for "macro-organisms", mainly animals and plants. Consequently, the evolutionary models of microorganisms could only be based on theoretical considerations, primarily focused on the presumed evolution of metabolic types (from anaerobiosis to aerobiosis) and certain morphological characteristics considered more or less "primitive". For instance, mycoplasmas were sometimes regarded as the most primitive living beings because they lacked a cell wall (Figure 1.1(a)).

Figure 1.1. Two visions of evolution in the late 1970s

COMMENT ON FIGURE 1.1.- (a) According to Lynn Margulis, eukaryotes result from the fusion of various types of bacteria, assimilated to prokaryotes (Margulis 1975). (b) According to Carl Woese, three kingdoms of prokaryotes emerge from an ancestor more "primitive" than current prokaryotes, the progenote. Eukaryotes originate from an extinct prokaryote lineage, the urcaryotes (Woese and Fox 1977).

1.3. Two domains for prokaryotes: archaebacteria and eubacteria

1.3.1. Ribosomal RNA as a molecular marker: a historical choice

To establish a classification of microorganisms on an objective basis that reflects their actual, rather than presumed, kinship relationships, Carl Woese and his collaborators at the University of Urbana in Illinois began in the early 1970s to determine the kinship relationships between different bacteria by comparing the RNAs present in the small subunit of their ribosomes, specifically the 16S ribosomal RNA (16S rRNA) (Figure 1.2(a)). Woese and his team chose ribosomal RNAs because these informational macromolecules are present in all organisms (with the exception of viruses, which will be discussed in Chapter 2). Additionally, the various functions they perform have been conserved throughout evolution, suggesting a "neutral" mode of evolution unaffected by changes in function. In other words, these macromolecules would have consistently evolved at the same rate across different cell lineages. This concept allows for the use of a molecular clock, meaning that the differences observed at the sequence level would directly reflect the evolutionary distances between organisms.

Why 16S rRNA rather than 5S or 23S rRNA? When Carl Woese began his research program in the early 1970s, studying all three ribosomal RNA molecules simultaneously was not feasible. He chose 16S rRNA because it contains more sequence information than the smaller 5S rRNA, while being easier to analyze than the (larger) 23S rRNA. Since the development of PCR and massive genome sequencing, the advantage of 16S rRNA over 23S rRNA has disappeared. Today, it is possible to quickly obtain the sequences of both molecules. However, this was not the case in the early 1970s. At that time, DNA sequencing was not yet developed, and RNA sequencing was a very lengthy and complex technique. Carl Woese and his collaborators overcame this difficulty by using a technique developed by the English biochemist Frederick Sanger (Sanger et al. 1965). The idea was to determine the evolutionary distance between two organisms by comparing the catalogs of oligonucleotides derived from the digestion of their RNA by a specific ribonuclease, ribonuclease T1. The genetic fingerprint of an organism corresponded to all the oligonucleotides identified in it. These oligonucleotides, previously made radioactive, were separated from each other by chromatography according to their size and composition, and then visualized by autoradiography (Figure 1.3(a)). It was then necessary to compare the catalog of the studied species with those of all the other species already analyzed.

This procedure required extensive development work, which explains Carl Woese's choice of an RNA molecule of intermediate size, neither too large nor too small (Figure 1.3(c)). First, the organisms had to be grown in large quantities to isolate their 16S (or 18S in eukaryotes) rRNAs. These molecules were then digested with ribonuclease T1, which systematically cuts the RNA sequence to the right of a G. For example, the portion of an RNA molecule corresponding to the sequence AUUUCGUUGAAU would have yielded the three oligonucleotides: AUUUCG, UUG and AAU. The different oligonucleotides present in the mixture obtained (a few dozen) were then separated by two-dimensional chromatography on very large sheets of paper. The cells from which the 16S RNA molecules were derived had been cultured in a medium containing radioactive phosphorus (P-32). The radioactive oligonucleotides then appeared as black spots on the photographic films placed for some time above the sheet of paper (Figure 1.3(a)). The arrangement of these spots was characteristic of the species studied. The oligonucleotides corresponding to each spot could then be eluted and their sequence identified by a combination of chemical and enzymatic methods. Fortunately, this laborious last step quickly became unnecessary, as the positions of the spots were reproducible from one analysis to the next, allowing for the visual identification of the corresponding oligonucleotide.

Figure 1.2. Carl Woese and the Triumph of Reductionism

COMMENT ON FIGURE 1.2.- (a) The secondary structure of 16S rRNA, highlighting variable regions useful as an evolutionary marker in blue. The inset depicts Carl Woese's finger pointing to a specific spot of the archaea, corresponding to an oligonucleotide containing a modified base. (b) Carl Woese receives the Crafoord Prize in 2003 from the hands of the King of...