Viruses are among the simplest biological entities on the planet, yet they pull off feats of molecular engineering that rival anything cells can do. A virus particle is little more than a protein shell wrapped around genetic material, but that minimal toolkit is enough to hijack complex cellular machinery, dodge sophisticated immune defenses, and generate staggering genetic diversity in a matter of hours. Understanding how viruses are built, how their genomes operate, and how they interact with host cells is not just academic curiosity. It underpins everything from vaccine design to cancer biology to gene therapy.
How a Virus Particle Is Put Together
Strip a virus down to basics and you get two things: a genome (DNA or RNA) and a protein coat called a capsid. The capsid’s job is protection and delivery. It shields the fragile genome from the environment and carries molecular tools for latching onto and entering a host cell. Capsid protein subunits are individually asymmetric, so the most stable arrangement packs them into highly symmetric shapes that minimize energy and maximize the contacts holding the structure together.1PubMed Central. Geometric architecture of viruses This is why so many viruses look like near-perfect geometric objects under an electron microscope.
The two most common capsid plans are helical and icosahedral. Helical capsids are rod-shaped, with protein subunits spiraling around the genome like a coiled staircase; tobacco mosaic virus is the classic example. Icosahedral capsids are roughly spherical, built from 20 triangular faces that assemble into a shape resembling a soccer ball. Many familiar human viruses, from common cold rhinoviruses to poliovirus, use this design. Some complex viruses combine both plans in one particle.
Beyond the capsid, many viruses carry an extra outer layer: a lipid envelope stolen from the host cell’s own membranes during an earlier round of infection. This envelope matters enormously for how the virus gets into new cells. Enveloped viruses fuse their lipid layer directly with a host cell membrane, while non-enveloped viruses rely on being swallowed into the cell through endocytosis and then punching through an internal membrane to reach the cytoplasm.2PubMed Central. Non-enveloped virus membrane penetration: New advances leading to new insights The membrane-penetrating agents non-enveloped viruses use are typically small hydrophobic peptides hidden inside the capsid until needed, and the way they interact with membranes varies widely from one virus family to another.3PubMed. Breach: Host Membrane Penetration and Entry by Nonenveloped Viruses
Viral Genomes and the Many Ways They Replicate
Cells all store their genetic information in double-stranded DNA. Viruses have no such constraint. Some carry double-stranded DNA, some single-stranded DNA, some double-stranded RNA, some single-stranded RNA in either positive or negative sense, and some use a reverse-transcription step to shuttle between RNA and DNA. This diversity is captured by the Baltimore classification system, which groups viruses into seven classes based on how they convert their genome into messenger RNA that the host’s ribosomes can read.4PubMed Central. The Baltimore Classification of Viruses 50 Years Later: How Does It Stand in the Light of Virus Evolution? The system has held up remarkably well over five decades, with only minor additions needed to accommodate newly discovered replication strategies.
Viral genomes are also impressively compact. Many viruses squeeze multiple genes into the same stretch of nucleotides by using overlapping reading frames, where shifting the reading window by one or two positions produces an entirely different protein from the same DNA or RNA sequence. Analysis across virus families shows that the number of these overlapping regions tends to increase with genome length, though each individual overlap gets shorter in larger genomes, consistent with the idea that gene overlap is a strategy for compressing information into small genomes.5PLoS Pathogens. Using networks to analyze and visualize the distribution of overlapping genes in virus genomes
RNA viruses in particular are genetic shape-shifters. Their mutation rates are orders of magnitude higher than those of DNA-based organisms, which fuels rapid adaptation but also creates something unusual: rather than existing as a single defined genome sequence, an RNA virus population is really a swarm of closely related but distinct variants, called a quasispecies.6PubMed Central. Quasispecies Nature of RNA Viruses: Lessons from the Past Evolution in these populations happens through constant disequilibria within the mutant cloud, not through single mutations marching through a uniform population.7PubMed Central. Viral quasispecies evolution This is one reason flu vaccines need updating every year and why HIV has been so difficult to pin down with a single vaccine: the target is not one virus but a shifting constellation of variants.
Getting Inside a Cell
A virus cannot infect just any cell. It needs to find one that displays the right molecular “lock” on its surface for the viral “key.” This receptor specificity is the main reason different viruses infect different tissues and different species, a property known as tropism.
Influenza offers a textbook example. Avian flu strains preferentially bind to sugar molecules on cell surfaces that end with one particular type of linkage, while human-adapted strains prefer a different linkage type. In the human airway, the human-preferred receptor appears on both ciliated and mucus-secreting goblet cells, giving the virus broad cellular tropism. In mice, that receptor is absent, which helps explain why certain human flu strains do not efficiently infect mouse lungs.8PubMed Central. Influenza virus receptor specificity and cell tropism in mouse and human airway epithelial cells The principle extends across virus families: small molecular variations in receptor proteins between species can determine whether a virus can jump from one host to another. Research on murine leukemia virus, for instance, shows that the viral surface protein locks onto a specific extracellular loop of its receptor that varies substantially between species, governing cross-species tropism.9PubMed Central. Structural insights into cationic amino acid transport and viral receptor engagement by CAT1
Once a virus finds its receptor, it still has to get its genome across a membrane. Enveloped viruses accomplish this through specialized fusion proteins that undergo dramatic shape changes. Despite enormous structural diversity across virus families, all characterized fusion proteins converge on a common sequence of events: the protein transforms from its resting state into an extended form that inserts a hydrophobic “fusion peptide” into the target membrane, then folds back on itself to pull the viral and cell membranes together until they merge.10PubMed Central. Structures and mechanisms of viral membrane fusion proteins: multiple variations on a common theme The fusion process moves through stages of membrane contact, partial merging called hemifusion, and finally pore formation. Different virus families have evolved at least four distinct trigger mechanisms for kicking off this conformational change, ranging from low pH in endosomes to receptor binding at the cell surface.
Taking Over the Factory Floor
Once inside, a virus faces a fundamental problem: it carries only a handful of genes but needs to produce thousands of copies of itself using machinery designed to serve the cell’s own needs. The solution is to repurpose the host’s equipment.
One of the most visible signs of this takeover is the formation of “viral factories,” specialized compartments that viruses build by reshaping the host cell’s own membranes and internal skeleton. These structures go by various names depending on the virus: some create bubble-like spherules, others build double-membrane vesicles, and still others form large cytoplasmic masses called viroplasms or set up shop inside the nucleus.11PubMed Central. Host cytoskeleton and membrane network remodeling in the regulation of viral replication These factories concentrate the raw materials for genome copying and particle assembly in one place while physically shielding viral RNA or DNA from the cell’s antiviral sensors.
Viruses also commandeer the cell’s protein-making machinery. Normally, a cell’s ribosomes read messenger RNA that carries a specific chemical cap at one end, the signal that says “translate me.” Several virus families have evolved ways to bypass this requirement entirely. Some viral RNAs contain internal ribosome entry sites that let ribosomes latch on in the middle of the message, making translation independent of the cap. Others use cap-independent translational enhancers located even in the far end of the RNA molecule to redirect ribosomes to the uncapped starting point.12Pleiades Publishing / Biochemistry (Moscow). Non-Canonical Translation Initiation Mechanisms Employed by Eukaryotic Viral mRNAs Some viruses go further, actively degrading or inactivating the host’s own capped messages so that the ribosomes have no choice but to translate viral RNA instead.
Dodging the Immune Response
The host is not passive during all of this. Vertebrate cells have a layered defense system anchored by interferons, signaling proteins that sound an alarm and activate antiviral genes throughout the body. Viruses, in turn, have developed a remarkable arsenal of countermeasures. A detailed examination of these strategies has catalogued at least ten common approaches that viruses use to block or subvert the interferon response, from sequestering the signaling molecules to degrading the sensors that detect viral RNA in the first place.13PubMed Central. Ten Strategies of Interferon Evasion by Viruses
Another critical arm of immune defense is the display of viral protein fragments on the cell surface via molecules called MHC class I. This display is essentially the cell waving a flag that says “I’m infected,” which alerts killer T cells to come destroy it. Many viruses have evolved ways to sabotage this pathway, blocking MHC molecules at multiple points: their initial production, their assembly, their transport to the surface, or their stability once they arrive.14PubMed Central. MHC-I pathway disruption by viruses: insights into immune evasion and vaccine design for animals SARS-CoV-2 provided a striking recent example. While other viruses and synthetic immune activators readily boosted MHC class I levels on the cell surface, cells infected with SARS-CoV-2 showed little or no increase in surface MHC expression, pointing to a targeted suppression of the pathway.15Nature Communications. SARS-CoV-2 inhibits induction of the MHC class I pathway by targeting the STAT1-IRF1-NLRC5 axis Understanding exactly which step each virus disrupts has direct implications for designing vaccines that can still trigger robust T cell responses.
Latency and Persistence
Not every virus burns through its host and moves on. Some viruses establish long-term residency by going dormant inside cells, a state known as latency. The mechanisms controlling when a virus enters latency, how it maintains its genome during dormancy, and what triggers reactivation are diverse and vary across virus families and even strains.16PubMed Central. Epigenetics and Genetics of Viral Latency
The key challenge for a latent virus is keeping its genome intact inside a dividing cell without being destroyed. Many herpesviruses and other DNA viruses solve this by forming circular DNA molecules called episomes that sit in the cell nucleus as free-floating mini-chromosomes, separate from the host’s own DNA. Hepatitis B virus and human papillomavirus use a similar strategy. Other viruses go a step further and stitch their DNA directly into the host chromosome, the way retroviruses like HIV do using a dedicated enzyme called integrase. Some herpesviruses can also integrate into host DNA, but they do so through a different mechanism that repurposes the cell’s own DNA-repair machinery.17Cell Host & Microbe. Mechanisms of Viral Latency
For viruses that persist as episomes, specialized viral proteins act as maintenance workers, tethering the episome to host chromosomes so it gets copied and distributed to daughter cells when the cell divides. Epstein-Barr virus uses a protein called EBNA1 for this purpose, the related herpesvirus KSHV uses a protein called LANA, and HPV relies on its E2 protein. These episome maintenance proteins also regulate how many viral genome copies accumulate and which viral genes stay active or silent.18PubMed Central. Control of Viral Latency by Episome Maintenance Proteins This silent persistence is not merely a curiosity: it is directly linked to the cancer-causing potential of these viruses, because the episomes can eventually destabilize and drive uncontrolled cell growth.
How New Virus Particles Leave the Cell
After a virus has replicated its genome and assembled new particles, those particles need to escape. For non-enveloped viruses, this is often brute force: the cell simply bursts open and releases its viral cargo. For enveloped viruses, exit is more elegant and involves budding, where the new virus particle pushes outward through a host membrane, pinching off and stealing a lipid coat in the process.
The budding process is not something viruses handle alone. Many enveloped viruses hijack the host cell’s own membrane-sculpting equipment, a set of protein complexes known collectively as ESCRT. Cells normally use ESCRT machinery to pinch off small membrane bubbles during routine housekeeping tasks like sorting proteins. HIV commandeered this system, and since that discovery, the ESCRT pathway has been recognized as a major escape route for enveloped viruses across many families.19PubMed Central. Virus budding and the ESCRT pathway HIV-1 in particular has become a model system for studying how a virus recruits ESCRT components to the site of budding to facilitate the final membrane cut that frees the new particle.20PubMed Central. Virus Hijacks Host Proteins and Machinery for Assembly and Budding, with HIV-1 as an Example
Spillover and Cross-Species Jumps
When a virus acquires the ability to infect a new host species, the consequences can be dramatic. The emergence of SARS, MERS, Ebola, and SARS-CoV-2 all involved viruses leaping from animal reservoirs into humans. These spillover events are not random accidents. They require a virus to clear a series of molecular hurdles: binding a receptor on the new host’s cells, evading that host’s innate immune defenses, and replicating efficiently enough to sustain transmission. Adaptive mutations, recombination between viral strains, and long histories of coevolution between viruses and their original hosts all shape which viruses have the molecular compatibility to make the jump.21Animal Diseases. Molecular mechanisms of viral host tropism and cross-species adaptation: a sequential molecular gatekeeping model of spillover The receptor-tropism principles described above for influenza and murine leukemia virus apply broadly: small changes in receptor-binding residues can open or close the door to a new species.
Ancient Viruses in Our Own DNA
Integration into the host genome is not always a dead end for a virus. Sometimes, millions of years later, the host turns the tables and starts using viral genes for its own purposes. Roughly eight percent of the human genome consists of sequences descended from ancient retroviruses that infected our ancestors and became permanently embedded. Most of these endogenous retroviral sequences have accumulated mutations and no longer produce functional viruses. But a handful have been “domesticated” by the host.
The most striking examples are the syncytin genes, which encode proteins derived from the envelope proteins of ancient retroviruses. These proteins are essential for the formation of the placenta in mammals. They mediate the cell-to-cell fusion that creates the syncytial layers at the boundary between mother and fetus. What makes this story especially remarkable is that syncytin genes were captured independently, through convergent evolution, in multiple mammalian lineages. Mice, rabbits, primates, and other groups each acquired their own syncytin from different retroviral infections at different times.22PubMed Central. Paleovirology of ‘syncytins’, retroviral env genes exapted for a role in placentation Knockout experiments in mice confirmed that syncytin genes are not just helpful accessories. They are absolutely required for placenta development and embryo survival.22PubMed Central. Paleovirology of ‘syncytins’, retroviral env genes exapted for a role in placentation More broadly, endogenous retroviruses and other transposable elements have contributed regulatory sequences that shape gene expression patterns across mammalian development, including differences in placental architecture between species.23PubMed Central. Endogenous Retroviruses and Placental Evolution, Development, and Diversity
Giant Viruses and Where Viruses Fit in the Tree of Life
The traditional image of a virus as tiny and genetically minimal took a hit in the early 2000s with the discovery of giant viruses. These particles, first identified in amoebae, are physically large enough to be seen under a standard microscope and carry genomes rivaling those of small bacteria. More provocatively, their genomes encode proteins that were previously considered signatures of cellular life, including components of the translation system that cells use to build proteins from RNA.24PubMed. Giant viruses with an expanded complement of translation system components That discovery prompted early speculation that giant viruses might represent a fourth domain of life, alongside bacteria, archaea, and eukaryotes. Most researchers now think these translation genes were acquired piecemeal from hosts over evolutionary time rather than inherited from an independent cellular ancestor, but the debate has forced a genuine rethinking of what counts as a virus and where viruses sit in the broader picture of life’s evolution.
Turning Viral Biology Into Medicine
Understanding viral structure and host interactions at molecular resolution has direct payoffs for medicine. Structural studies of viral enzymes, for example, reveal the precise shapes of the pockets where drugs can bind. High-resolution imaging of HIV-1’s reverse transcriptase bound to antiviral drugs has illuminated exactly how drug-resistant mutations alter the binding site, guiding the design of next-generation inhibitors that can still work against mutant strains.25PubMed Central. Cryo-EM structures of wild-type and E138K/M184I mutant HIV-1 RT/DNA complexed with inhibitors doravirine and rilpivirine Similar structural work on the topoisomerase enzyme of African swine fever virus has revealed that its architecture closely resembles eukaryotic topoisomerases that are already successful drug targets in cancer and antibacterial therapy, opening a potential avenue for antiviral design against this devastating livestock pathogen.26PubMed Central. Cryo-EM structures of African swine fever virus topoisomerase
Viruses are not just targets for medicine; they are also tools. Adeno-associated virus, or AAV, has become the workhorse of gene therapy because it is small, non-pathogenic, and remarkably efficient at delivering DNA into human cells. Engineering efforts have refined AAV vectors to improve how efficiently they deliver their cargo, to redirect which tissues they target by modifying the capsid, and to reduce the chance of the immune system neutralizing the vector before it can do its job.27PubMed. Engineering adeno-associated virus vectors for gene therapy Several AAV-based therapies have already been approved for conditions ranging from inherited blindness to spinal muscular atrophy, with dozens more in clinical trials. The same principles of tropism, receptor binding, and immune evasion that make natural viruses effective pathogens are what make engineered viruses effective delivery vehicles.
Bacteriophages and Their Own Brand of Molecular Machinery
Viruses that infect bacteria, called bacteriophages or simply phages, operate by many of the same principles as viruses that infect animals, but their structural engineering includes some unique features. Many phages use a contractile tail to inject their DNA through the tough outer layers of a bacterial cell. The tail works somewhat like a molecular syringe: an outer sheath contracts, driving an inner tube through the cell wall. Structural studies of individual phages have revealed the mechanical details of this process. One phage that targets the bacterium responsible for a severe gut infection uses a contraction that shortens its tail by about a fifth, with accompanying changes in the sheath’s diameter and internal geometry that break the contacts holding the sheath to the inner tube, freeing the tube to punch through.28Life Science Alliance. Molecular mechanism of bacteriophage contraction structure of an S-layer–penetrating bacteriophage
Phage biology has practical implications well beyond basic science. Phage therapy, the idea of using bacteriophages to kill antibiotic-resistant bacteria, has experienced a resurgence in clinical interest as drug-resistant infections have become a growing crisis. The exquisite specificity of phage-host interactions, governed by the same receptor-binding logic seen in animal viruses, means a phage can be chosen to target a particular bacterial strain with minimal collateral damage to beneficial microbes. Meanwhile, phage-derived enzymes have become indispensable tools in molecular biology and biotechnology, from CRISPR-associated proteins that bacteria originally evolved as anti-phage defenses to the DNA-modifying enzymes that power modern genomics.