Host-Pathogen Interactions

A host-pathogen interaction is best understood as a set of matched moves and countermoves across an evolutionary timescale, layered on top of the moment-to-moment biochemistry of a single infection. At the biochemical level, pathogens present molecular patterns — surface proteins, nucleic acid structures, metabolic products — that host receptors have evolved to recognize, from germline-encoded pattern recognition receptors (PRRs) of innate immunity to the somatically diversified receptors of B and T lymphocytes. Evasion, in this framework, is any pathogen trait that reduces the efficiency of recognition, signaling, effector killing, or memory formation at one or more of these steps. Because evasion traits that improve pathogen fitness are inherited and selected exactly as any other trait, and because microbial populations are enormous and fast-replicating, evasion mechanisms are not rare accidents; they are near-inevitable outcomes of sustained immune pressure. The key conceptual distinction is between evasion and subversion. Evasion mechanisms reduce the probability that a pathogen is detected or bound at all — antigenic variation, capsule formation that masks underlying antigens, and intracellular residence in compartments innate sensors do not survey are all evasion strategies. Subversion mechanisms actively redirect or cripple immune machinery that has already engaged the pathogen — blocking phagolysosome fusion after phagocytosis, secreting proteases that cleave antibodies or complement components, or inducing regulatory T cells that dampen the response to the pathogen itself. A third category, exploitation, uses host machinery for pathogen benefit, as when HIV uses CD4 and CCR5 as an entry receptor pair, turning the very molecule that identifies helper T cells into a doorway. Coevolution gives this taxonomy its dynamism. Because host defenses (antibody specificity, MHC allele diversity, T cell receptor repertoires) are themselves genetically variable and heritable, pathogens that evade a common host allele have a fitness advantage, and hosts carrying rare alleles that still recognize the pathogen have a corresponding advantage — a process formalized as negative frequency-dependent selection, closely related to the Red Queen hypothesis (organisms must constantly adapt merely to maintain fitness relative to co-evolving antagonists). This is a principal explanation for the extraordinary polymorphism of the MHC/HLA locus in vertebrate populations: no single allele combination is universally optimal, because pathogens continually adapt to whichever allele is locally common.

Step 1 — Antigenic variation defeats acquired memory by changing the target. The clearest case is African trypanosomiasis. Trypanosoma brucei coats its entire cell surface with roughly 10⁷ copies of a single glycoprotein, the variant surface glycoprotein (VSG), encoded by a repertoire of well over 1,000 VSG genes and pseudogenes in the genome, only one of which is transcribed at a time from a specialized subtelomeric expression site. When host antibodies rise against the dominant VSG and clear most of that population, a minority subpopulation that has already switched its active VSG gene (via gene conversion or in situ promoter switching) survives and expands, producing a new wave of parasitemia with a serologically distinct coat. This produces the classic relapsing-remitting parasitemia curve of sleeping sickness: each peak is cleared by antibody, each subsequent peak is a new antigenic variant. Influenza performs a milder, population-scale version: antigenic drift accumulates point mutations in hemagglutinin (HA) and neuraminidase (NA) epitopes under continuous antibody selection pressure in the human population, gradually eroding vaccine and natural-immunity protection over a few years; antigenic shift occurs when two influenza strains co-infect a cell (classically in pigs, which express both avian- and human-type sialic acid receptors) and reassort whole gene segments, producing an abruptly novel HA subtype against which almost no one has memory — the mechanism behind pandemics such as 1918 (H1N1), 1957 (H2N2), 1968 (H3N2), and 2009 (H1N1).

Step 2 — Intracellular residence hides pathogens from antibody and complement. Antibodies and complement act primarily in extracellular fluid; a pathogen that lives inside a host cell is largely invisible to them until the cell dies or presents pathogen peptides on MHC. Mycobacterium tuberculosis, phagocytosed by alveolar macrophages, secretes the protein ESAT-6 and other ESX-1 system effectors that arrest phagosome maturation, preventing phagosome-lysosome fusion and acidification, so the bacterium persists in a modified phagosome rather than being degraded. Listeria monocytogenes takes this further: its pore-forming toxin listeriolysin O (LLO) ruptures the phagosomal membrane, releasing the bacterium into the nutrient-rich cytosol, where it polymerizes host actin via the surface protein ActA to propel itself directly into adjacent cells — spreading cell-to-cell without ever re-entering the extracellular space where antibody could act.

Step 3 — Direct subversion of complement, antibody, and antigen presentation. Many bacteria produce surface proteins that bind the Fc region of antibodies in the wrong orientation (Protein A of Staphylococcus aureus, Protein G of Streptococcus), sterically blocking Fc-receptor-mediated phagocytosis while leaving the antibody "used up" on the surface. Others recruit host complement regulators: Neisseria gonorrhoeae and Borrelia burgdorferi both bind host Factor H, the negative regulator of the alternative complement pathway, decorating their own surface with a host "self" signal that shuts down C3b amplification. Herpesviruses encode proteins that downregulate MHC class I on the infected cell surface (human cytomegalovirus's US2/US3/US6/US11 gene products each target a different step of MHC I biosynthesis and degradation) to escape CD8⁺ cytotoxic T cell recognition — but this creates a new vulnerability, because cells lacking MHC I trigger natural killer cell attack via the "missing-self" mechanism, forcing CMV to also express a decoy MHC-I-like molecule (UL18) to blunt that NK response.

Step 4 — Active immunosuppression and the destruction of immune cells themselves. The most direct evasion is killing the effector cells of adaptive immunity. HIV enters CD4⁺ T helper cells via CD4 plus a coreceptor (CCR5 early in infection, sometimes CXCR4 later), integrates its reverse-transcribed genome into host DNA, and, upon activation, drives massive CD4⁺ T cell depletion both by direct viral cytopathic effects and by chronic immune activation leading to apoptosis of bystander cells — progressively dismantling the very coordination layer (helper T cells) that adaptive immunity depends on, a process detailed mechanistically in Immunodeficiency and HIV/AIDS. Many pathogens also actively recruit host regulatory mechanisms: Leishmania and helminths such as Schistosoma induce regulatory T cells (Tregs) and anti-inflammatory cytokines (IL-10, TGF-β) that dampen the very Th1/Th2 responses needed to clear them, permitting chronic, low-grade, tolerated infection rather than sterilizing clearance.

Step 5 — Molecular mimicry, sometimes at the cost of autoimmunity. Some pathogens display surface epitopes structurally similar to host self-antigens, reducing the odds that self-tolerant T and B cells will respond strongly. Streptococcus pyogenes M protein shares structural motifs with human cardiac myosin and other tissue proteins; in a minority of infected individuals, antibodies and T cells raised against M protein cross-react with heart valve and joint tissue, producing rheumatic fever — mimicry that reduces initial detection can, when the response finally breaks through, misdirect that response onto self tissue.

Establishing antigenic variation as a molecular mechanism, not just a clinical curiosity. In the 1970s, George Cross and colleagues purified the variant surface glycoprotein of Trypanosoma brucei and showed biochemically that antibodies raised against parasites from one relapse peak failed to recognize parasites isolated from a subsequent peak, even though both derived from the same clonal infection. Subsequent molecular cloning (led by groups including those of Piet Borst and George Cross through the 1980s) showed that the trypanosome genome carries a large family of VSG genes, most silent, with only one active at a time expressed from a specialized telomeric expression site, and that switching occurs by duplicative gene conversion (copying a silent VSG gene into the active expression site) or by in situ activation of a different expression site. This directly demonstrated the mechanism proposed to explain the clinical pattern of relapsing fever in African trypanosomiasis: each parasitemic wave is antigenically distinct at the molecular level, precisely matching the serological distinctness observed clinically, and antibody-mediated clearance itself is the selective force driving each successive switch.

Establishing MHC restriction and its consequence for pathogen evasion. Rolf Zinkernagel and Peter Doherty, working with lymphocytic choriomeningitis virus (LCMV) in mice in 1973–74, found that cytotoxic T lymphocytes from an infected mouse could kill virus-infected target cells only if the targets shared the mouse's own MHC class I haplotype — T cells from an H-2ᵏ mouse killed LCMV-infected H-2ᵏ targets but not infected H-2ᵇ targets, even though both targets carried the identical viral antigen. This showed that T cell receptors recognize a composite ligand — foreign peptide bound in the groove of self MHC — rather than free viral antigen alone, work recognized by the 1996 Nobel Prize in Physiology or Medicine. This finding is the mechanistic foundation for understanding why MHC downregulation (as practiced by cytomegalovirus and many tumors) is such an effective evasion strategy: without self-MHC to present the peptide, CD8⁺ T cells simply have no ligand to recognize, regardless of how much viral antigen is present inside the cell — and it explains why NK cells evolved "missing-self" surveillance as the corresponding host countermeasure.

Establishing antigenic drift and shift with molecular and epidemiological data. Comparative sequencing of influenza hemagglutinin genes isolated from patients across successive decades (extensive work compiled by groups including Peter Palese, Edwin Kilbourne, and later global influenza surveillance networks coordinated by the WHO) showed a steady accumulation of amino acid substitutions clustering in five defined antigenic sites on the HA head, correlating tightly with the decline in efficacy of antibodies raised against earlier strains — direct molecular confirmation of drift. Phylogenetic and serological reconstruction of the 1957, 1968, and 2009 pandemics showed abrupt, discontinuous jumps in HA subtype (H1→H2, H2→H3, and a novel H1N1 reassortant respectively) rather than gradual accumulation, with genomic segment analysis showing mixed avian/swine/human gene segment ancestry — direct confirmation that reassortment (shift), not point mutation, generated pandemic strains. Together these two datasets separated drift and shift as mechanistically distinct evasion processes operating on different genetic substrates (point mutation versus segment reassortment) and different timescales (years versus decades).

Example 1 — Quantifying an influenza antigenic drift escape. Suppose a population has pre-existing neutralizing antibody against an epitope on hemagglutinin present at estimated protective titer in 70% of adults (herd immunity ≈ 0.70). A new drift variant arises carrying a single amino acid substitution in that epitope that reduces antibody binding affinity 20-fold, dropping effective neutralization in previously immune individuals to only 15% of that group (titer falls below the protective threshold in the rest). What is the new effective population susceptibility, and what does this predict about epidemic potential?

• Fraction previously immune and still protected: 0.70 × 0.15 = 0.105 (10.5% of the population remains protected via that specific response).
• Fraction previously immune but now susceptible: 0.70 × 0.85 = 0.595 (59.5%).
• Fraction never immune (always susceptible): 0.30.
• New total susceptible fraction = 0.595 + 0.30 = 0.895 (89.5%), compared with 0.30 before the mutation.

Using the simple threshold condition for epidemic growth, R₀ × (susceptible fraction) > 1 is required for spread; if R₀ for this influenza strain is 1.3, then before the mutation, 1.3 × 0.30 = 0.39 (no epidemic, below threshold), but after the mutation, 1.3 × 0.895 ≈ 1.16 (above threshold — an epidemic is now possible). This captures, quantitatively, why a single well-placed antigenic-site substitution can convert a population from largely immune to epidemiologically vulnerable — the essence of why influenza vaccines require near-annual reformulation, and why WHO strain-selection committees track exactly these substitution patterns.

Example 2 — Reasoning through a clinical scenario: reactivation tuberculosis in an immunosuppressed patient. A 54-year-old patient with a remote (30-year-old, untreated but asymptomatic) tuberculosis exposure begins infliximab (an anti-TNF-α monoclonal antibody) for rheumatoid arthritis. Eight months later they develop fever, night sweats, and a cavitary upper-lobe lesion on chest imaging; sputum is positive for acid-fast bacilli. Reasoning through the mechanism: M. tuberculosis survives for decades inside granulomas, organized structures of macrophages, epithelioid cells, and lymphocytes that wall off (but do not sterilize) the bacterium, a containment state maintained actively rather than passively. TNF-α is required for granuloma maintenance — it sustains macrophage activation, promotes apoptosis of infected cells, and helps recruit and organize the cellular architecture of the granuloma. Blocking TNF-α with infliximab does not "cause" a new infection; it removes the ongoing active suppression that had held a latent population of intracellularly-hiding bacteria in check. Bacteria that survived precisely because they resided inside macrophages, invisible to circulating antibody the entire time, are released from granuloma constraint and resume active replication — a direct clinical illustration of Step 2 mechanism above (intracellular hiding defeats humoral immunity, but requires ongoing active cellular containment, which anti-TNF therapy specifically disables). This is why anti-TNF biologics carry a labeled requirement for latent TB screening (interferon-gamma release assay or tuberculin skin test) before initiation.

Rolf Zinkernagel and Peter Doherty's discovery of MHC restriction is one of immunology's cleanest examples of an unplanned observation reshaping an entire field. Working at the John Curtin School of Medical Research in Canberra in the early 1970s, they were studying how mice cleared lymphocytic choriomeningitis virus (LCMV) infection of the meninges, expecting a relatively simple story of cytotoxic T cells killing infected cells. Instead, their killing assays produced a puzzling pattern: T cells from an infected mouse would efficiently lyse virus-infected cells from a genetically identical (syngeneic) mouse, but not virus-infected cells from a different mouse strain, even when both carried the same virus. Rather than dismiss this as assay noise, they systematically mapped the requirement to the major histocompatibility complex, showing T cells demand a match at MHC class I loci in addition to viral antigen. Doherty, a veterinarian by training, brought a strong instinct for careful comparative pathology; Zinkernagel, trained in both medicine and immunology, pushed the genetic dissection. Their 1974 papers reframed cellular immunology entirely around the concept of "self plus antigen" recognition, a principle with direct consequences for understanding why pathogens that manipulate MHC expression are so evolutionarily successful.

history

• 1973 — Zinkernagel and Doherty begin cytotoxic T cell assays with LCMV-infected mice in Canberra.
• 1974 — Publication establishing that T cell killing requires matching self-MHC plus viral antigen (MHC restriction).
• 1975–1980s — Molecular characterization of trypanosome VSG genes and expression-site switching (Cross, Borst and colleagues) formalizes antigenic variation as a genetic mechanism.
• 1996 — Zinkernagel and Doherty awarded the Nobel Prize in Physiology or Medicine for MHC restriction.
• 1996–present — Structural biology (peptide-MHC crystal structures) and genome sequencing of pathogens (M. tuberculosis 1998, HIV reservoirs, influenza global surveillance) extend the mechanistic evasion catalog described in this article.

Understanding immune evasion mechanistically, rather than descriptively, has directly shaped modern vaccine design, diagnostics, and therapeutics. Because influenza escapes antibody responses through mutations concentrated in variable HA head epitopes, the current generation of "universal" influenza vaccine candidates targets the conserved HA stalk region, a domain under strong structural constraint (mutations there tend to destroy membrane-fusion function, so the virus cannot easily escape without a fitness cost) — a design directly derived from understanding antigenic drift's molecular geometry. Because HIV persists in a latent proviral reservoir inside resting memory CD4⁺ T cells invisible to both antibody and cytotoxic T cells, "shock and kill" and "block and lock" cure strategies attempt either to reactivate latent virus for clearance or to lock it more permanently into silence — both approaches are direct responses to the evasion mechanism of latency itself. Because M. tuberculosis survives by arresting phagosome maturation, host-directed therapies now under trial aim to pharmacologically force phagosome-lysosome fusion or enhance autophagy in infected macrophages, rather than targeting the bacterium directly. Understanding Factor H hijacking by Neisseria and Borrelia has informed engineered vaccine antigens (e.g., factor H binding protein constructs in meningococcal vaccines) designed to block the mimicry interaction itself. And recognition that MHC polymorphism is maintained by pathogen-driven balancing selection has informed both transplant medicine (why perfect HLA matching is hard to find in outbred populations) and vaccine equity considerations (why a vaccine epitope effective in one HLA-diverse population may need supplementation for full population coverage) — all downstream, practical consequences of treating host-pathogen interaction as an ongoing evolutionary negotiation rather than a solved contest.

Ask the tutor: How does negative frequency-dependent selection specifically explain why no single MHC/HLA allele ever reaches fixation in a human population?