Immune: On May 14, 1796, a country doctor in Berkeley, Gloucestershire named Edward Jenner took material from a cowpox sore on the hand of a milkmaid named Sarah Nelmes and inoculated it into the arm of James Phipps, an eight-year-old boy from the village.[9]
Cowpox was a mild, self-limiting disease. Six weeks later, Jenner inoculated Phipps with fresh material from a smallpox pustule - the standard variolation method then used to produce mild smallpox immunity, which carried a real risk of causing the disease.
Phipps showed no sign of smallpox. Jenner repeated the challenge again, and again the boy remained well. Jenner had demonstrated the principle of vaccination - that exposure to a related, harmless pathogen could protect against a deadly one - without having the slightest understanding of why it worked.
The mechanism Jenner had stumbled onto would take another century and a half to understand in molecular detail. Louis Pasteur confirmed germ theory in the 1860s and developed attenuated vaccines for anthrax and rabies. Robert Koch identified specific bacterial pathogens in the 1870s-1880s.
But the immunological machinery underlying vaccination - the specific cellular actors, the receptor diversity, the selection and memory formation processes - was not described in mechanistic terms until Macfarlane Burnet proposed clonal selection theory in 1957, and not explained at the molecular genetic level until Susumu Tonegawa discovered V(D)J recombination in 1976.
The Nobel Prize for Tonegawa's discovery came in 1987. The 2023 Nobel Prize in Physiology or Medicine went to Katalin Karik and Drew Weissman for the modified mRNA technology that enabled COVID-19 vaccines - a discovery that depended on understanding immunological mechanisms that were entirely unknown when Jenner watched James Phipps remain healthy.
The story of immunological learning is one of the most elegant in all of biology: a system of almost unimaginable diversity, generated by deliberate genetic randomization, refined by Darwinian selection inside the body, and capable of improvement with every encounter - the immune system as a learning machine.
"The power of vaccination is the power of memory - not ours, but the immune system's, encoded in cells that persist for decades, waiting." - Peter Medawar, The Uniqueness of the Individual (1957)
Key Definitions
Innate immunity - The immediate, non-specific first line of immune defense, present from birth and not requiring prior exposure to function.
Components include physical barriers (skin, mucous membranes), pattern recognition receptors (Toll-like receptors) that detect conserved microbial signatures, macrophages, neutrophils, natural killer (NK) cells, and the complement system.
Adaptive immunity - The specific, learned arm of the immune system, mediated by B cells (which produce antibodies) and T cells (which kill infected cells and coordinate responses). Requires days to weeks to deploy on first exposure but generates immunological memory for faster, stronger responses on re-exposure.
Antigen - Any molecular fragment (typically a protein or carbohydrate) that can be specifically recognized by the adaptive immune system and trigger an immune response. From "antibody generator."
B cells - Lymphocytes that, when activated, differentiate into plasma cells that secrete antibodies. Named for the bursa of Fabricius in birds where they were first identified; in mammals they mature in the bone marrow. Each B cell carries a unique membrane-bound antibody (B cell receptor) that defines its antigen specificity.
T cells - Lymphocytes that mature in the thymus. Helper T cells (CD4+) coordinate immune responses by signaling to B cells and cytotoxic T cells. Cytotoxic T cells (CD8+) kill cells displaying foreign antigens (infected cells, tumor cells). Each T cell carries a unique T cell receptor.
Clonal selection - The process by which an antigen selects the lymphocytes (B cells or T cells) whose receptors specifically bind it, triggering massive proliferation and differentiation of those cells. Proposed by Macfarlane Burnet in 1957; the foundational mechanism of adaptive immunity.[1]
V(D)J recombination - The genetic mechanism generating receptor diversity in B cells and T cells, assembling receptor genes from variable (V), diversity (D), and joining (J) gene segments through RAG1/RAG2 enzyme-mediated recombination. Nobel Prize to Susumu Tonegawa, 1987.
Affinity maturation - The process in germinal centers by which B cells undergo somatic hypermutation in their antibody genes and are selected for those mutations that improve antigen binding, generating progressively higher-affinity antibodies over the course of an immune response.
Memory B cells and T cells - Long-lived lymphocytes that persist after an immune response resolves, encoding specificity for the antigen that triggered their formation. They respond faster and more robustly than naive lymphocytes on re-exposure to the same antigen.
Herd immunity - The indirect protection of unimmunized individuals in a population arising when a sufficient proportion of others are immune, breaking transmission chains before they reach susceptible people.
mRNA vaccines - A vaccine platform that delivers messenger RNA encoding a pathogen antigen (rather than the antigen directly), causing the recipient's own cells to produce the antigen and mount an immune response. Enabled by Karik and Weissman's discovery of modified mRNA that avoids triggering excessive innate immune reactions.[3]
Innate vs. Adaptive Immunity Compared
| Feature | Innate Immunity | Adaptive Immunity |
|---|---|---|
| Speed of response | Minutes to hours | Days to weeks (first exposure) |
| Specificity | Generic - detects broad pathogen patterns (PAMPs) | Highly specific - recognizes particular antigens |
| Memory | None - same response every time | Yes - faster and stronger on re-exposure |
| Key cells | Macrophages, neutrophils, NK cells, dendritic cells | B cells (antibodies), helper T cells (CD4+), cytotoxic T cells (CD8+) |
| Diversity mechanism | Pre-programmed receptor patterns | V(D)J recombination generates ~10^18 possible receptor combinations |
| Activated by | PAMPs (conserved microbial signatures) | Specific antigen presented by MHC molecules |
| Vaccination target | Provides danger signals; required for adaptive activation | Primary target - generates antibodies and memory cells |
| Improves with experience | No | Yes - affinity maturation improves antibody quality over days |
The Two Arms of the Immune System
Innate Immunity: Speed Without Specificity
The innate immune system is the body's immediate response to infection - present from birth, operating within minutes to hours, and requiring no prior exposure to the specific pathogen to function.
Its pattern recognition receptors (Toll-like receptors, NOD-like receptors, RIG-I-like receptors) detect conserved molecular patterns shared by classes of microorganisms - bacterial lipopolysaccharide, flagellin, double-stranded RNA - that are not present in healthy human cells.
These "pathogen-associated molecular patterns" (PAMPs) trigger innate responses: macrophages engulf and destroy pathogens; natural killer cells kill infected host cells that have downregulated their MHC class I markers; neutrophils flood the site of infection; the complement cascade punches holes in bacterial membranes.
The innate system is fast but not specific: it responds the same way to every bacterial infection, every viral infection, regardless of the specific pathogen. It cannot improve with experience - a second infection with the same bacterium triggers exactly the same innate response as the first.
But it does something critically important: it provides early containment while the slower adaptive response is being organized, and its inflammatory signaling provides the "danger signals" that activate the adaptive system and shape the character of the adaptive response that follows.
Adaptive Immunity: Specific and Learned
The adaptive immune system operates on a completely different logic: specificity over speed. Instead of pre-programmed responses to generic patterns, it generates individual, highly specific responses to particular antigens - and retains a cellular memory of those responses.
A B cell whose antibody binds the specific protein on a flu virus will be selected, expanded, and remembered. A year later, the same flu protein will activate those memory cells within hours, not the days or weeks required for a naive primary response.
The adaptive system requires days to two weeks to deploy on first exposure - the primary immune response - during which the innate system holds the infection at bay while adaptive cells are being selected and expanded. This is why you feel sick for a week with a new pathogen: the adaptive system needs time to gear up.
But once the response is established and memory cells are formed, subsequent exposures are handled rapidly, often before symptoms develop.
Clonal Selection: The Central Mechanism
Macfarlane Burnet's clonal selection theory (1957) solved a puzzle that had defeated immunologists for decades: how can the immune system specifically recognize and respond to essentially any antigen, including synthetic molecules that have never existed in nature?
The alternative theories - instructive theories, which supposed that antibodies somehow adopted the shape of the antigen that instructed them - had failed on biochemical grounds.
Burnet's insight: the diversity comes first. Before any antigen arrives, the adaptive immune system has already generated a vast library of lymphocytes - each with a unique, randomly generated receptor - that collectively cover essentially the entire space of possible antigens.
When an antigen arrives, it is recognized by the rare lymphocyte whose randomly generated receptor happens to bind it.
That lymphocyte receives signals to proliferate massively - producing a clone of thousands of identical cells, all recognizing the same antigen - and to differentiate into effector cells (plasma cells secreting antibodies, or cytotoxic T cells killing infected cells) and memory cells.
The elegance of the theory is that it explains both specificity (each clone has a unique receptor) and memory (clones persist after the antigen is cleared) with a single mechanism. It also explained tolerance: lymphocytes that react strongly to self-antigens must be eliminated during development, or the result is autoimmunity.
Burnet and Peter Medawar shared the 1960 Nobel Prize in Physiology or Medicine for this framework and for Medawar's experimental demonstration of acquired immunological tolerance.
V(D)J Recombination: Generating the Diversity
The clonal selection theory described what must happen - enormous pre-existing receptor diversity - but did not explain how the genome, with a finite number of genes, could encode the receptor diversity needed.
Susumu Tonegawa's experiments in 1976-1977 provided the answer: the receptor genes are not inherited as complete sequences but are assembled from gene segments by a unique form of somatic recombination that occurs only in lymphocytes during their development.[2]
The Mechanism
For antibody heavy chains, the immune system uses three families of gene segments: V (variable), D (diversity), and J (joining) segments.
In each developing B cell, the RAG1 and RAG2 enzymes randomly select one V segment, one D segment, and one J segment from the available repertoire (approximately 40-50 functional V segments, 25 D segments, and 6 J segments for heavy chains) and splice them together into a single coding sequence.
The resulting V-D-J junction encodes the hypervariable region of the antibody - the part that directly contacts antigen and determines specificity.
Additional diversity comes from junctional diversity: RAG1/RAG2 do not cut the DNA at precisely defined positions, and exonucleases trim the ends before ligation, so the junctions between segments gain or lose nucleotides randomly.
This imprecision at the junction - the most critical part of the antigen-binding region - exponentially increases diversity beyond what simple combinatorial assembly would produce.
For light chains (and T cell receptor alpha chains), V and J segments are joined without a D segment, but the same principle applies.
The estimated diversity from V(D)J recombination alone - combining combinatorial and junctional diversity for both heavy and light chains - exceeds 10^11 unique antibody sequences.
Paired with the somatic hypermutation that occurs after antigen stimulation, the total potential diversity of the human antibody repertoire is essentially unlimited for practical purposes.
Tonegawa's discovery revealed that the immune system uses deliberate genetic instability - somatic recombination that would be pathological in any other context - as a strategy for generating diversity. The genome does not encode every possible receptor; it encodes a molecular toolkit for generating receptors on the fly.
Selection in the Thymus: Learning Self from Non-Self
T cells are generated in the bone marrow but complete their maturation in the thymus, where they undergo a rigorous selection process that distinguishes the useful from the dangerous.
Positive selection occurs in the thymic cortex: T cells whose randomly generated receptors can recognize self-MHC (major histocompatibility complex) molecules - the proteins that present antigen fragments to T cells - receive survival signals.
T cells whose receptors cannot recognize self-MHC die by neglect. This step ensures that only T cells capable of recognizing antigen in the context of the body's own presenting molecules survive.
Negative selection occurs in the thymic medulla: T cells whose receptors bind self-MHC plus self-peptides (fragments of the body's own proteins) too strongly receive death signals. This step - central tolerance - eliminates T cells that would attack the body's own tissues.
Approximately 95% of T cells die during thymic selection; only the minority with receptors that recognize self-MHC but do not react too strongly to self-peptides exit the thymus as functional, tolerant T cells.
The AIRE gene (autoimmune regulator), expressed in medullary thymic epithelial cells, drives the production of tissue-specific antigens in the thymus - allowing negative selection to eliminate T cells that would be reactive to liver, pancreas, or thyroid proteins, for instance.
AIRE mutations cause autoimmune polyendocrinopathy - a syndrome in which multiple endocrine organs are attacked - because T cells reactive to endocrine tissue escape thymic deletion.
Affinity Maturation: Improving the Response
The primary immune response generates antibodies, but not necessarily the best possible antibodies. The body has a mechanism for improving them: affinity maturation in germinal centers.
After clonal selection and initial antibody production, some activated B cells migrate to germinal centers - specialized microenvironments that form transiently in lymph nodes and spleen after antigen stimulation.[4]
Inside germinal centers, B cells undergo somatic hypermutation: the variable regions of their antibody genes are mutated at a rate approximately one million times higher than the normal cellular mutation rate. Most mutations worsen or abolish antibody binding. Some improve it.
B cells with improved antibodies compete more effectively for the limited antigen displayed on follicular dendritic cells within the germinal center. Better binders receive survival signals (via T follicular helper cells); worse binders die.[8]
Over successive rounds of mutation and selection - a Darwinian process entirely within the individual immune response - the average binding affinity of the antibody population increases dramatically, often by several orders of magnitude compared to the initial antibodies.
Memory B cells emerging from germinal centers carry the high-affinity, affinity-matured antibody sequences. When they encounter their antigen again in a subsequent infection, they produce high-affinity antibodies immediately - not only faster but better than the primary response.
The immune system does not just remember; it remembers having improved its response, and deploys the improved version.
Memory Cells and the Secondary Response
The immunological memory that Jenner stumbled onto is encoded in long-lived memory B cells and memory T cells that persist for years or decades - sometimes for life - after the initial immune response.
Memory B cells retain the high-affinity, affinity-matured antigen receptors generated during the germinal center reaction. They circulate in blood and peripheral tissues, ready to respond rapidly to re-exposure.
When they encounter their antigen again, they differentiate quickly into antibody-secreting plasma cells, flooding the system with high-affinity antibodies within days - compared to the one to two weeks of the primary response.
Long-lived plasma cells - a distinct population that migrates to niches in the bone marrow - secrete antibodies continuously even in the absence of antigen, maintaining baseline antibody titers that provide immediate neutralization upon re-exposure before memory B cell activation is even required.[5]
Studies of people vaccinated against measles five to six decades ago detect both persisting antibody titers and circulating memory B cells, encoding immunity established in childhood.
Memory T cells similarly persist and respond more rapidly than naive T cells: they have lower activation thresholds, respond to smaller amounts of antigen, and deploy effector functions (cytokine secretion, cytotoxicity) more quickly.
Central memory T cells (in lymph nodes and spleen) and effector memory T cells (in peripheral tissues) represent distinct subpopulations with different properties, providing layered protection.
How Vaccines Exploit Immunological Memory
Vaccines present antigens to the immune system in a context that triggers clonal selection, affinity maturation, and memory cell formation - without causing the disease the pathogen would produce.
The result is that genuine first exposure to the pathogen is immunologically a second exposure: memory cells recognize it immediately, plasma cells produce pre-formed antibodies, and the pathogen is often cleared before symptoms develop.
Live-attenuated vaccines - using a weakened version of the pathogen that can replicate but cannot cause serious disease - produce the most robust and durable immunity because they most closely mimic natural infection, engaging both B cells and T cells in all phases of the response.
The measles vaccine, introduced in 1963, produces immunity that lasts essentially a lifetime in most recipients; the smallpox vaccine conferred protection for decades.
The mRNA vaccine platform, developed for COVID-19 from foundational work by Katalin Karik and Drew Weissman on modified nucleosides, is distinctive in its mechanism.
Rather than delivering an antigen directly, mRNA vaccines deliver instructions - the mRNA encoding the SARS-CoV-2 spike protein, with chemical modifications to the nucleosides that reduce innate immune reactions to the mRNA itself while preserving translation.
The recipient's own muscle cells at the injection site translate the mRNA into spike protein, which is then presented to the immune system.
The Nobel Prize awarded to Karik and Weissman in 2023 recognized that the nucleoside modification technology was the critical enabling step that made safe, effective mRNA vaccines possible.
Herd Immunity: Protecting the Unvaccinated
Herd immunity arises when the proportion of immune individuals in a population is high enough that transmission chains break even before reaching susceptible people.
The mathematics are driven by the pathogen's basic reproduction number (R0): the average number of secondary infections generated by one case in a fully susceptible population. The herd immunity threshold - the proportion that must be immune - is approximately 1 minus (1/R0).[6]
Measles, with an R0 of 12-18, requires 92-95% population immunity to achieve herd immunity. A 5-8% unvaccinated population is sufficient to sustain measles transmission if clustered geographically.
The resurgence of measles outbreaks in the 2010s following vaccine hesitancy reflected exactly this dynamic: aggregate coverage remained above 90%, but local geographic clusters fell below the threshold.
Smallpox, with a lower R0 (approximately 5-7), was amenable to eradication through vaccination campaigns that achieved herd immunity in each region. The WHO declared smallpox eradicated in 1980 - the only human pathogen yet eradicated - representing the most complete achievement of global immunological learning in history.
Edward Jenner's observation in his village in Gloucestershire in 1796 led, through 184 years of accumulated immunological science and public health effort, to the permanent elimination of a disease that had killed perhaps 300 million people in the twentieth century alone.
Trained Innate Immunity and Long COVID
A recent revision to the classical picture of immunological memory concerns the innate immune system. The dogma that innate immunity cannot improve with experience has been challenged by discoveries of "trained immunity" in macrophages and NK cells.
Exposure to certain pathogens or vaccines (the BCG tuberculosis vaccine in particular) appears to reprogram macrophages epigenetically, altering their gene expression in ways that make them respond more vigorously to subsequent, unrelated challenges.[7]
This non-specific memory of the innate system may partly explain why BCG vaccination in infancy reduces childhood mortality from unrelated infections.
The immune dysregulation of long COVID has also raised new questions about immunological memory: some long COVID symptoms appear to involve persistent viral antigen, chronic immune activation, or autoimmune processes triggered by SARS-CoV-2 molecular mimicry - demonstrating that immunological learning can go wrong, generating persistent inflammation and antibodies against self-tissues rather than purely protective memory.
For related topics, see how vaccines work, how the human immune system works, and how pandemics spread.
Sources & Further Reading
- Burnet, F. M. (1957). A Modification of Jerne's Theory of Antibody Production Using the Concept of Clonal Selection. Australian Journal of Science, 20, 67-69.
- Tonegawa, S. (1983). Somatic Generation of Antibody Diversity. Nature, 302(5909), 575-581. DOI: 10.1038/302575a0
- Karik, K., Buckstein, M., Ni, H., & Weissman, D. (2005). Suppression of RNA Recognition by Toll-like Receptors: The Impact of Nucleoside Modification and the Evolutionary Origin of RNA. Immunity, 23(2), 165-175. DOI: 10.1016/j.immuni.2005.06.008
- Victora, G. D., & Nussenzweig, M. C. (2012). Germinal Centers. Annual Review of Immunology, 30, 429-457. DOI: 10.1146/annurev-immunol-020711-075032
- Plotkin, S. A. (2010). Correlates of Protection Induced by Vaccination. Clinical and Vaccine Immunology, 17(7), 1055-1065. DOI: 10.1128/CVI.00131-10
- Anderson, R. M., & May, R. M. (1985). Vaccination and Herd Immunity to Infectious Diseases. Nature, 318(6044), 323-329. DOI: 10.1038/318323a0
- Netea, M. G., et al. (2020). Defining Trained Immunity and Its Role in Health and Disease. Nature Reviews Immunology, 20(6), 375-388. DOI: 10.1038/s41577-020-0285-6
- Crotty, S. (2019). T Follicular Helper Cell Biology: A Decade of Discovery and Diseases. Immunity, 50(5), 1132-1148. DOI: 10.1016/j.immuni.2019.04.011
- Jenner, E. (1798). An Inquiry into the Causes and Effects of the Variolae Vaccinae. Sampson Low.
Part of This Series
The Immune System: Your Body's Defense Mechanism
How the Human Immune System Works: Innate vs. Adaptive Immunity
How Vaccines Work: Immune Response Explained
What Is Pharmacology?
Exploring the Human Microbiome and Its Importance
How Antibiotics Work: Mechanisms and Resistance Explained
The Science of Pain: Nociception, the Brain, and Chronic Pain
