The Holobiont Paradigm: Reframing the Human Organism

1.1 From Animalcules to Algorithms: A Historical Trajectory of Human-Microbe Conceptualization

The conceptual repositioning of microbes from incidental inhabitants to constitutive partners of the human organism did not occur as a single epistemic rupture; it unfolded across four centuries, each era contributing a specific instrument of observation and a corresponding theoretical prejudice that had to be dismantled before the next could be erected. To grasp why the holobiont paradigm feels simultaneously revolutionary and overdue, one must reconstruct this trajectory, because the assumptions of each prior stage persist as silent constraints in contemporary experimental design.

The first rupture occurred in 1674–1683, when Antonie van Leeuwenhoek, using single-lens microscopes of his own fabrication reaching magnifications of roughly 275×, reported "animalcules" inhabiting dental plaque, fecal matter, and seminal fluid. His letters to the Royal Society constitute the founding observation of human-associated microbial ecology, yet their conceptual yield was constrained by the absence of any theoretical scaffold in which to place them. Animalcules were curiosities, neither pathological nor physiological, because the dominant Galenic and later iatromechanical frameworks of disease had no category for invisible, ubiquitous, non-pathogenic life. Leeuwenhoek's observations were therefore archived rather than integrated — a fate that, as we shall see, would repeat with each new technical advance until a sufficiently integrative theory emerged.

The second rupture arrived in the second half of the nineteenth century through the convergence of Louis Pasteur's fermentation studies (1857–1861) and Robert Koch's postulates (formulated 1884). This pair of advances produced the germ theory of disease, which — for all its indisputable therapeutic value — installed a profoundly antagonistic framing of host-microbe relations: the microbe as invader, the host as defended territory, the immune system as a military apparatus. This conceptual architecture was so productive that it actively suppressed interest in commensal or mutualistic associations for nearly a century. Investigators who observed abundant microbes in apparently healthy tissues either dismissed them as contaminants or, when pressed, hypothesized that they must be latent pathogens awaiting opportunity. The cultural and methodological infrastructure of microbiology — agar plates, selective media, colony counting — was built around the detection and elimination of suspect organisms. The vast majority of human-associated microbes, which resist cultivation under standard laboratory conditions, were rendered epistemically invisible precisely by the instruments that had revealed their pathogenic cousins.

A third, quieter rupture began in the 1960s and accelerated through the 1990s with the recognition that anaerobic cultivation, specialized gas mixtures, and longer incubation times could recover a substantially broader fraction of the human microbiota. Named organisms such as *Bacteroides thetaiotaomicron* (1971, Salyers and colleagues) and the methanogenic archaeon *Methanobrevibacter smithii* (1982, Miller and colleagues) demonstrated that strict anaerobes and archaea — not just facultative bacteria — populated the healthy gut. Yet the conceptual weight of this expansion remained limited because the underlying logic was still one of inventory: cataloging individual species rather than understanding community-level function. Microbial ecology in this period retained a character reminiscent of nineteenth-century natural history — comprehensive, descriptive, but lacking the mechanistic claims that would compel a re-evaluation of host biology.

The fourth rupture, and the one that opened the conceptual space for the holobiont paradigm, was methodological rather than theoretical: the development of culture-independent, sequence-based surveys of microbial communities. The pivotal advance was the design of universal PCR primers targeting the bacterial and archaeal 16S ribosomal RNA gene, first systematically deployed by Pace, Olsen, and colleagues in the mid-1980s and culminating in Norman Pace's influential 1997 review advocating "molecular phylogenetic surveys" as a substitute for cultivation. The 16S gene contains both highly conserved regions (useful as primer-binding sites) and nine hypervariable regions (useful as taxonomic identifiers), making it an ideal molecular chronometer. By sequencing this gene from environmental DNA, investigators could enumerate microbial taxa without ever growing them — dissolving the methodological bias that had rendered most of the microbiota invisible for over a century.

Two parallel technical currents accelerated this shift. First, the advent of high-throughput short-read sequencing platforms (454 pyrosequencing in 2005, Illumina sequencing shortly thereafter) drove the cost of megabase-scale sequencing from tens of thousands to a few dollars, making population-level surveys feasible. Second, the development of shotgun metagenomics — sequencing all DNA in a sample rather than only amplified 16S amplicons — extended the analytic reach from taxonomic identification to functional gene content, enabling inferences about community metabolism directly from sequence data. By 2007–2010, the launch of the Human Microbiome Project (HMP, NIH) and the MetaHIT consortium (European Commission) had transformed what had been a niche specialty into a central preoccupation of biomedical science.

The cumulative effect of these four ruptures was the production of an empirical observation so stark that it demanded theoretical revision: the human body harbors microbial cells in numbers roughly comparable to (and by recent revised estimates within a 1:1 to 1.3:1 ratio of) human cells, and the collective microbial genome — the microbiome — encodes between roughly 2 million and 20 million non-redundant genes, dwarfing the approximately 20,000 protein-coding genes of the human genome by two to three orders of magnitude. The conceptual problem this observation posed was not merely quantitative. If the microbiome represents a metabolic and immunological organ whose genomic complexity exceeds that of the host itself, then any definition of the "human" that excludes it is operationally incomplete. This is the conceptual pressure to which the holobiont and hologenome concepts are responses.

The term *holobiont* itself was introduced by Lynn Margulis and Dorion Sagan in 1991 to denote the host plus its persistent symbiotic microbiota as a single ecological unit. The *hologenome*, a term introduced by Eugene Rosenberg and Ilana Zilber-Rosenberg in 2007–2008, extends this concept to the genetic level: the combined genetic repertoire of host and symbionts, considered as a target of selection. These terms are not merely lexical novelties; they encode substantive theoretical commitments about the unit of selection, the nature of individuality, and the appropriate frame for biomedical experimentation. Whether those commitments are warranted is the subject of Section 1.4. Before that critical evaluation, however, we must examine in detail the empirical foundations on which the holobiont framework rests.

1.2 The Quantitative and Functional Inventory: How Much Biology Is Not "Human"?

The case for reframing the human organism as a holobiont rests on three empirical pillars: quantitative dominance, functional indispensability, and evolutionary entanglement. We address each in turn, treating them as independent lines of evidence that converge on the same conceptual conclusion.

1.2.1 Quantitative dominance revisited

The often-cited claim that bacterial cells outnumber human cells ten-to-one derives from a 1972 estimate by Thomas Luckey and was propagated through textbooks for decades without re-examination. A rigorous revision by Sender, Fuchs, and Milo (2016), which accounted for the volumes of the rel