The structure of a virion

Stability outside the host is the single most useful thing a virion’s structure tells you, and it follows directly from the outermost layer. An enveloped virion presents a lipid bilayer to the world. That bilayer is held together by the same weak hydrophobic forces as any cell membrane, so it is disrupted by detergents and soaps, by alcohols such as ethanol and isopropanol at 60–80 per cent, by drying, by heat and by extremes of pH. Once the bilayer is breached the particle cannot fuse with the next cell, even if its genome and capsid are intact. A naked virion presents a protein capsid instead. Proteins are more robust than membranes, so non-enveloped viruses survive acid, bile salts, drying, moderate heat and many disinfectants, and some remain infectious on hard surfaces for days to weeks. Norovirus is the classic example: fewer than a hundred particles can cause illness, it survives stomach acid at pH 2, it resists freezing and heating to 60 °C, and it is not reliably killed by alcohol hand rub, so chlorine at 1 000–5 000 ppm for about a minute is required on contaminated surfaces. This is beyond-notes material, marked in the sources below, because the supplied notes define the envelope at lines 1321–1326 but never draw out its fragility. The consequence organises prevention. A virus that cannot survive drying must move directly from a wet source to a new host, which pushes transmission towards respiratory droplets and aerosols, sexual and blood contact, and close contact: influenza, measles, HIV and hepatitis B. A virus that survives drying and acid can take the long route — faecal–oral spread, contaminated water and food, and fomites such as door handles and toys, with outbreaks in closed wards and schools: norovirus, rotavirus, hepatitis A and poliovirus. That is why alcohol hand rub is sufficient for influenza and HIV but not for a norovirus outbreak, where soap and water are used because physically removing the particles matters more than chemically killing them, and why hands soiled with faeces or vomit must always be washed rather than rubbed. Two figures sharpen the rule: 60–70 per cent ethanol inactivates influenza virus on a hand within about 15–30 seconds, whereas norovirus needs chlorine; and HIV loses most of its infectivity within hours of drying, which is why casual contact does not transmit it.

Bacteriophage T4 is the standard illustration that virion complexity is set by the delivery problem, not by the sophistication of the host. T4 has a prolate icosahedral head, about 111 nanometres long and 78 wide, holding a single double-stranded DNA molecule of roughly 169 000 base pairs packed under great pressure. Below the head sits a collar, then a contractile sheath surrounding an inner tube, then a baseplate, then six long tail fibres and six short ones. Infection begins when the long fibres recognise lipopolysaccharide or outer-membrane proteins on Escherichia coli; the baseplate changes shape and anchors, and the sheath contracts like a syringe, driving the tube through the outer membrane and the thin peptidoglycan layer so that the DNA is injected into the cytoplasm. A tail-associated lysozyme digests the peptidoglycan locally to let the tube through. The particle is built from about 50 genes and assembles in separate, independently constructed sub-assemblies — head, tail and fibres — which then join, a pathway worked out in the 1960s by Edgar and Wood using conditional-lethal mutants; some sub-assemblies even form spontaneously in a test tube, the classic demonstration that capsids are self-assembled structures. Compare that with HIV, which infects a far more complex host but carries only nine genes and about 15 proteins in a particle roughly 120 nanometres across, or with adenovirus, a non-enveloped icosahedral virus with 12 fibre spikes at its vertices and a genome of about 36 000 base pairs. Complexity is not a ladder from simple hosts to complex ones. A bacterial host is wrapped in a wall that must be breached mechanically, so the phage invests genes in a molecular drill; a human host attacks the virus with antibodies, so HIV invests in a glycan shield and in changing its peplomers rather than in a drill. Selection acts on the obstacle in front of the particle.

The viral enzymes are the part of virion structure with the most direct clinical consequence, because they supply what the host cell cannot and therefore become the targets of antiviral drugs. Host cells contain DNA-dependent DNA polymerase, which copies DNA, and DNA-dependent RNA polymerase, which copies DNA into RNA. They contain no enzyme that copies RNA into RNA and none that copies an RNA genome back into DNA. A virus with an RNA genome must therefore package an RNA-dependent RNA polymerase, or the first round of replication cannot begin; influenza carries its own RNA polymerase for exactly this reason. Retroviruses go further. HIV packages reverse transcriptase, which copies its single-stranded RNA genome into double-stranded DNA while travelling towards the nucleus; integrase, which inserts that DNA into a host chromosome to form a provirus (the viral DNA copy integrated into the host chromosome); and protease, which cuts the long viral polyprotein into working pieces as new particles are assembled. Because reverse transcriptase has no proofreading, it makes roughly one error in every ten thousand bases, so the HIV population in one patient becomes a cloud of variants within months; this is the mechanism behind drug resistance and the reason treatment combines at least three drugs aimed at two or more targets. Antiviral drugs are named after these enzymes: tenofovir and zidovudine inhibit reverse transcriptase, dolutegravir and raltegravir inhibit integrase, and protease inhibitors block the cutting step. Herpesviruses show a different pattern, because acyclovir is a harmless prodrug until the viral thymidine kinase phosphorylates it, after which it inhibits the viral DNA polymerase. Influenza neuraminidase is also an enzyme, but it is not carried inside the particle to start infection; it sits on the envelope as a peplomer and releases new virions from the cell surface, and oseltamivir blocks it. For the nurse this explains why adherence is a structural issue and not a moral one: sub-therapeutic drug levels select the variants that already carry a resistant enzyme, so a missed dose is a step in the evolution of resistance.