How Vaccines Work: The Immunology Behind Immunity

In 1796, a country doctor in rural Gloucestershire made a wager on a principle no one had yet articulated. Edward Jenner inoculated eight-year-old James Phipps with material taken from a cowpox blister on the hand of a milkmaid named Sarah Nelmes. Six weeks later, he deliberately exposed the boy to smallpox — a disease killing roughly 400,000 Europeans every year. James did not fall ill. Jenner had discovered vaccination, though he had no idea why it worked. The word "virus" did not yet exist. The immune system was a black box. The germ theory of disease lay three-quarters of a century in the future.

Two centuries after Jenner's experiment, we understand the mechanism with extraordinary precision, and that understanding has produced one of the greatest technological achievements in human history. Vaccines have eradicated smallpox entirely — a disease that killed an estimated 300 million people in the twentieth century alone — and have reduced the global burden of polio, measles, diphtheria, tetanus, and dozens of other infectious diseases by orders of magnitude. The World Health Organization estimates that vaccination currently prevents between 3.5 and 5 million deaths annually.

Yet misconceptions about how vaccines work remain widespread, and those misconceptions have real public health consequences. The immunological science behind vaccination is not only practically important — it is genuinely fascinating, a story about one of the most sophisticated biological systems ever to evolve.

The Immune System: Two Lines of Defence

To understand vaccines, you must first understand what they are activating. The mammalian immune system consists of two interconnected but functionally distinct arms: the innate immune system and the adaptive immune system.

The innate immune system is ancient in evolutionary terms — shared with invertebrates and even plants in its most basic forms. It responds to infection within minutes to hours using a fixed repertoire of receptors that recognise broad patterns common to many pathogens: lipopolysaccharide (a component of bacterial cell walls), viral double-stranded RNA, fungal chitin. These pattern-recognition receptors, called toll-like receptors and first described by Jules Hoffmann and his colleagues at the University of Strasbourg in the 1990s, trigger rapid inflammatory responses that aim to contain infection before the adaptive system is ready. Hoffmann and Bruce Beutler shared the Nobel Prize in Physiology or Medicine in 2011 for this work.

The adaptive immune system is evolutionarily younger, found only in vertebrates, and operates on a completely different logic. Rather than recognising fixed pathogen patterns, it generates an almost unlimited repertoire of receptors — B cell receptors and T cell receptors — through random genetic recombination. The diversity produced is staggering: estimates suggest the human immune system can generate on the order of 10¹⁵ to 10¹⁸ distinct receptor configurations. When one of these randomly-generated receptors happens to match a pathogen antigen, the cell bearing it proliferates dramatically — clonal expansion — producing an army of identical cells all targeting the same molecular structure.

The crucial feature of the adaptive system — the one that makes vaccination possible — is immunological memory. After an infection resolves, most of the expanded clone of antigen-specific cells dies off. But a small subset survives as long-lived memory cells, maintaining elevated antibody levels and persisting in bone marrow and lymphoid tissues for years, sometimes lifetimes. When the same pathogen is encountered again, memory cells respond within hours rather than days, producing antibodies at concentrations orders of magnitude higher than the initial response and doing so fast enough to clear the infection before symptoms develop. This is why people generally do not catch the same cold twice in the same season, and why surviving certain infections confers lasting immunity.

The Vaccine Principle: Training Without the Disease

Vaccines exploit immunological memory without requiring the individual to survive the full danger of natural infection. They present the immune system with antigens — molecules that the immune system can recognise as foreign — without the pathogen's capacity to cause serious disease. The immune system mounts an adaptive response, generates memory cells, and the individual is now primed: if they later encounter the real pathogen, the memory response activates before the infection can cause harm.

The central challenge of vaccine design is presenting enough antigen to generate robust memory while avoiding the dangers of the disease itself. Different vaccine platforms solve this problem in different ways, reflecting a century of innovation in immunology and molecular biology.

The mRNA Revolution: A New Vaccine Architecture

The approval of mRNA vaccines for COVID-19 in late 2020 represented the culmination of roughly three decades of foundational research. The platform's origins trace directly to work by Katalin Karikó, then at the University of Pennsylvania, who spent years in the 1990s and 2000s investigating why synthetic mRNA triggered inflammatory reactions when introduced into human cells — an obstacle that had made mRNA therapeutics impractical.

Karikó and her collaborator Drew Weissman discovered in 2005 that substituting pseudouridine for uridine in synthetic mRNA — a small chemical modification mimicking a natural modification found in certain human RNAs — dramatically reduced the innate immune response to the synthetic molecule while preserving its ability to be translated into protein. This work, published in Immunity, was the key that unlocked practical mRNA therapeutics and would eventually earn Karikó and Weissman the Nobel Prize in Physiology or Medicine in 2023.

"The mRNA platform is not just a new vaccine technology — it is a new way of instructing the immune system. We can encode virtually any antigen and have a vaccine candidate ready for clinical testing in weeks rather than years."

— Professor Drew Weissman, Roberts Family Professor of Vaccine Research, Perelman School of Medicine, University of Pennsylvania

In an mRNA vaccine, the mRNA is encapsulated in lipid nanoparticles — tiny fat bubbles that protect the fragile RNA from degradation and facilitate its uptake by cells at the injection site. Once inside the cell, the mRNA is translated by ribosomes into the encoded protein — in the case of the COVID-19 vaccines, the spike protein of SARS-CoV-2. The spike protein is displayed on the cell surface and processed by the immune system as a foreign antigen, triggering both B cell and T cell responses. Crucially, the mRNA never enters the cell nucleus, cannot be integrated into chromosomal DNA, and degrades within days. The cell continues producing antigen only as long as the mRNA persists.

The speed advantage of mRNA platforms proved decisive during the COVID-19 pandemic. Once the SARS-CoV-2 genome was sequenced in January 2020, the mRNA sequence encoding the spike protein could be designed computationally within days. The BioNTech team had a vaccine candidate ready for toxicology testing within weeks — a timeline impossible for any traditional platform. Clinical trials confirmed efficacy in the high 90s percent against the original variant, and the platform's adaptability has since been applied to influenza, respiratory syncytial virus (RSV), and experimental cancer vaccines targeting patient-specific tumour antigens.

Adjuvants: Amplifying the Signal

Many vaccines, particularly subunit and protein vaccines, require adjuvants — substances added to the vaccine formulation to enhance the immune response. The reason relates to the two-arm structure of immunity described earlier. The adaptive immune system requires not just antigen (the signal) but also co-stimulatory signals from the innate system (sometimes called "danger signals") to mount a robust memory response. Without these co-stimulatory signals, antigen presentation alone can lead to tolerance — the immune system learning to ignore the antigen rather than responding to it.

The oldest adjuvant, alum (aluminium hydroxide or aluminium phosphate), has been used since the 1920s. It works partly by forming a depot at the injection site that slowly releases antigen, and partly by activating innate immune cells through mechanisms not fully understood until recently. More modern adjuvants include AS04 (alum plus MPL, a modified bacterial lipopolysaccharide, used in the hepatitis B and HPV vaccines), AS01B (used in the RTS,S malaria vaccine and the shingles vaccine), and squalene-based oil-in-water emulsions like MF59 and AS03, which activate the NLRP3 inflammasome and stimulate strong antibody responses to influenza antigens.

Herd Immunity: When Individual Vaccination Protects Others

One of vaccination's most important population-level effects is herd immunity — the indirect protection of unvaccinated or immunocompromised individuals when a sufficient proportion of the surrounding population is immune. The mechanism is straightforward: a pathogen spreads through social contact, and if most contacts of an infectious individual are immune, transmission chains are broken before they can generate epidemic spread.

The proportion of the population that must be immune to achieve herd protection depends on the pathogen's basic reproduction number (R₀) — the average number of secondary infections generated by a single case in a fully susceptible population. The higher the R₀, the greater the proportion that must be immune to interrupt transmission.

Disease	R₀ (approx.)	Herd Immunity Threshold	Vaccine Efficacy
Measles	12–18	~95%	97% (2 doses MMR)
Pertussis (whooping cough)	12–17	~92–94%	~85% (acellular)
Polio	5–7	~80–86%	~99% (IPV, 3 doses)
Seasonal influenza	1.2–1.4	~33–44%	40–60% (varies by season)
COVID-19 (original)	2.5–3.5	~60–72%	~95% against original (mRNA)

The measles figures illustrate both the power and the fragility of herd immunity. Measles is one of the most contagious pathogens known, with an R₀ of 12 to 18 in unvaccinated populations. This means that achieving herd protection requires roughly 95% of the population to be immune — a threshold requiring sustained vaccination coverage that leaves essentially no room for hesitancy or exemption. Measles outbreaks in Western countries over the past decade have occurred almost exclusively in communities where vaccination rates have fallen below that threshold, demonstrating that herd immunity is not a permanent achievement but a continual one, requiring ongoing vigilance.

Vaccine Safety: What the Evidence Shows

The safety record of licensed vaccines, established through clinical trials enrolling tens to hundreds of thousands of participants and post-marketing surveillance systems monitoring billions of doses, is remarkable. Serious adverse events are rare and extensively documented. The most common side effects — injection-site soreness, mild fever, fatigue — reflect the innate immune response to the vaccine components and typically resolve within days. They are, in a sense, evidence that the vaccine is working: the innate immune signals that cause temporary discomfort are the same signals that prime the adaptive response to generate lasting memory.

Serious adverse events do occur with some vaccines at very low rates. Myocarditis has been observed following mRNA COVID-19 vaccination, particularly in young males after the second dose — at a rate of approximately 1–4 per 100,000 doses in the highest-risk group (males aged 16–29), according to a 2022 analysis in the New England Journal of Medicine. Most cases were mild and resolved spontaneously. COVID-19 myocarditis rates in unvaccinated individuals were substantially higher. The pattern illustrates a principle fundamental to vaccine evaluation: risks must always be compared against the baseline risk of the disease, not against zero — which exists nowhere in medicine.

The Future of Vaccination: Cancer, HIV, and Personalised Immunity

The mRNA platform has opened research horizons that would have seemed fantastical two decades ago. Cancer vaccines — using mRNA encoding the unique somatic mutations (neoantigens) present in an individual patient's tumour cells — are in advanced clinical trials. Because tumours evolve to evade the immune system, therapeutic cancer vaccines aim to reinvigorate immune responses that the tumour has suppressed, rather than preventing infection. Early results from trials of personalised mRNA vaccines for melanoma and non-small-cell lung cancer have been encouraging, with some trials showing significant reductions in recurrence when combined with checkpoint inhibitor immunotherapy.

HIV has resisted decades of vaccine development efforts because of its extraordinary capacity to mutate, evade antibody recognition, and establish latent reservoirs in immune cells themselves. New approaches based on broadly neutralising antibodies — antibodies that target conserved structural features of the virus shared across many variants — combined with germline-targeting immunogens designed to educate the immune system to produce those antibodies, represent the most promising current strategy. Phase I trials are underway.

From Jenner's empirical wager in 1796 to personalised cancer vaccines encoded in synthetic RNA, the trajectory of vaccination science has been one of progressively deeper understanding transforming progressively more ambitious intervention. The immune system, as centuries of investigation have revealed, is not a fixed wall but a learning system — one that can, with the right instruction, be taught to recognise threats that have not yet arrived. That is the extraordinary promise Jenner glimpsed without understanding, and that modern immunology has only begun to fulfil.