ergosome

Part III (you are reading this)

If you have ever watched a crowd surge toward an emergency, sirens blaring and lights flashing, you have seen the good side of a fast response. But the real test of any system is not only how quickly it mobilizes. It is whether it can stand down once the crisis is over.

The immune system faces this exact challenge: it can expand armies of lymphocytes in days, but when the threat passes, those armies must shrink cleanly and quietly. FAS (also known as CD95 or APO-1) is one of the signals that makes this possible. When it works, immune responses resolve and self-tolerance holds. When it fails, immune cells linger like responders who never go home, filling lymph nodes, spilling into tissues, and sometimes turning their weapons inward.

This post focuses on peripheral tolerance, where the immune system actively deletes activated or autoreactive cells that escaped earlier checkpoints. Few pathways illustrate the necessity of timely disposal better than Fas, an accidental antibody finding that became a cornerstone of immunology, mouse genetics, and human disease.

Two antibodies, two continents, one unsettling observation (1989)

In the late 1980s, monoclonal antibodies were already transforming biology. Most were used as labels or blockers: bind a protein, stop a ligand, map a marker. Then something unexpected happened in Tokyo.

Shin Yonehara and colleagues generated a monoclonal antibody, often referred to as CH-11, that did not merely bind a cell-surface protein. It killed the cell. Even more striking, the killing was enhanced when cells were sensitized with actinomycin D, a clue that what they were triggering was not nonspecific damage but a regulated program the cell could execute when survival transcripts were suppressed. In 1989, Yonehara reported this phenomenon and named the antigen “Fas” after the FS-7 cell line used during antibody generation, an oddly mundane origin for a receptor that would become synonymous with cellular suicide.

Almost simultaneously in Germany, Peter H. Krammer’s group, with Bernhard Trauth as first author, described an agonistic antibody against a surface molecule they called APO-1. Their antibody also triggered programmed cell death in tumor cells. The convergence was uncanny: two independent laboratories on different continents had found that binding a particular surface receptor could function as an off switch for cell survival.

At the time, discovering such antibodies was anything but straightforward. Researchers immunized mice with whole human cells, fused thousands of antibody-producing B cells into hybridomas, and then screened tens of thousands of clones by hand. Crucially, these screens were functional rather than descriptive: instead of asking whether an antibody bound a cell, they asked whether it did something dramatic, like halting growth or killing the cell outright. With no high-throughput assays, no genomics, and no rapid way to identify antibody targets, recognizing that a rare antibody was triggering a regulated death program rather than nonspecific toxicity required painstaking controls, intuition, and persistence. Confirming that the death was apoptosis relied on morphology, membrane blebbing, DNA fragmentation, and the absence of inflammation. Apoptosis is a tidy self-destruct mechanism. DNA is chopped into fragments, the membrane forms blebs, and the cell breaks into apoptotic bodies that can be cleared without triggering inflammatory chaos. In retrospect, the discovery of Fas and APO-1 looks inevitable, but at the time it was a low-probability find that depended as much on experimental courage as on technical skill.

By the end of 1989, the field was hooked. If Fas or APO-1 was a genuine death-inducing receptor, the questions were obvious: what was this receptor, why did immune cells carry it, and whatphysiological conditions triggered its expression?

From antigen to gene and the lpr mouse mystery (1991-1992)

Cloning Fas in the pre-genomic era (1991)

The next milestone required molecular cloning, and it reflected both the limits and ingenuity of early 1990s biology. In 1988, Yonehara approached Shigekazu Nagata, already known for work on cytokine receptors, to help isolate the gene encoding Fas. At the time, cloning meant building and screening cDNA libraries and validating candidates experimentally, not simply sequencing a sample and searching a database. The team had to build cDNA libraries from Fas-positive cells, express them in mammalian cells, and repeatedly enrich for the rare transfectants that gained anti-Fas antibody staining by flow cytometry. Early library screens even failed, forcing a switch to stronger mammalian expression systems.

In 1991, Naoto Itoh, Yonehara, Nagata, and colleagues finally cloned the human Fas cDNA and showed that Fas belongs to the TNF receptor superfamily. The protein carried a cysteine-rich extracellular domain and a cytoplasmic death domain. When expressed in cultured cells, Fas could mediate apoptosis: cells transfected with Fas would self-destruct when exposed to agonistic anti-Fas antibody. This turned a strange antibody effect into a new category of signaling molecule, the death receptor.

lpr mice were not hyperactive, they were failing to die (1992)

At the same time, immunologists had long been puzzled by a spontaneous mutant mouse strain called lpr, short for lymphoproliferation. Especially on autoimmune-prone backgrounds such as MRL, lpr mice developed massive lymph node and spleen enlargement, produced autoantibodies reminiscent of lupus, and suffered immune-complex kidney disease. Their lymph nodes filled with abnormal CD4-negative, CD8-negative double-negative T cells, an unusual population that became one of the most recognizable clues in immune tolerance genetics.

Mapping a mouse mutation in that era was slow and laborious. There was no whole-genome sequencing and no easy way to recreate candidate mutations to test causality. Geneticists had narrowed lpr to mouse chromosome 19, but the responsible gene remained unknown. Nagata’s newly cloned Fas gene offered a compelling candidate. In 1992, Rie Watanabe-Fukunaga in Nagata’s lab, working with NIH mouse geneticists Nancy Jenkins and Neal Copeland, examined Fas expression in lpr mice. The result was decisive: lpr mice showed almost no functional Fas expression, and a genetic lesion in the Fas gene explained the phenotype. Later work revealed different lpr alleles, including a retrotransposon insertion that disrupted transcription and a point mutation in the death domain, known as the lpr^cg allele, that rendered Fas nonfunctional.

This finding forged a direct link between defective apoptosis and autoimmunity. If activated T cells cannot receive the Fas time-to-die signal, they persist longer than they should, accumulate in lymphoid organs, and increase the likelihood of autoreactivity and tissue damage. The 1992 Nature paper went further, noting that Fas is normally expressed in the thymus and likely plays an important role in deleting autoreactive T cells, bridging central and peripheral tolerance.

How Fas kills: death by design

Mechanistically, Fas signaling turned out to be both simple and ruthless. When Fas is engaged by ligand or agonistic antibody, it trimerizes and assembles the death-inducing signaling complex, or DISC. The cytoplasmic death domain recruits the adaptor protein FADD, which in turn recruits procaspase-8, historically also called FLICE. Procaspase-8 molecules activate one another by proximity, then trigger downstream executioner caspases such as caspase-3. DNA fragments, cellular scaffolding collapses, and the cell is dismantled from within, usually without inflammation.

By the early 1990s, the picture was coming together: a TNF-family receptor on lymphocytes that could trigger apoptosis, a mutant mouse lacking this receptor that developed autoimmune lymphoproliferation, and a mechanistic pathway linking receptor engagement to caspase activation. One crucial piece was still missing: the physiological ligand.

Fas ligand and the gld mirror image (1993-1994)

The ligand appears (1993)

Every receptor needs a ligand. For Fas, the search concluded in late 1993 when Nagata’s group identified and cloned Fas Ligand, another TNF-family molecule expressed on activated T cells and natural killer cells. This discovery made Fas biologically intuitive. Cytotoxic lymphocytes could eliminate targets by engaging Fas, and immune responses could terminate through activation-induced cell death, a contraction phase that prevents chronic immune activation and preserves self-tolerance.

gld mice complete the genetic picture (1994)

At nearly the same time, attention returned to another autoimmune mouse strain: gld, for generalized lymphoproliferative disease. gld mice developed a phenotype almost indistinguishable from lpr mice, yet the mutation mapped elsewhere. In 1994, researchers showed that gld mice carry a point mutation in the Fas ligand gene, producing a nonfunctional ligand unable to trigger Fas-mediated death.

The symmetry was striking. lpr mice lacked a functional Fas receptor. gld mice lacked a functional Fas ligand. Different mutations, same outcome: immune cells that should die persisted, lymphoid organs enlarged, and autoimmunity followed. Together, these mutants established the Fas and FasL pair as indispensable for immune regulation and tolerance.

A human syndrome finds its molecular cause (1967-1995)

Long before Fas was known, clinicians had described a rare childhood disorder characterized by chronic, non-malignant lymphadenopathy, splenomegaly, and autoimmune cytopenias, without infection or cancer. Frank Canale and Robert Smith first reported it in 1967. For decades, it remained a medical curiosity.

By the mid-1990s, immunologists had the tools and conceptual framework to connect this syndrome to the Fas pathway. In 1995, Frédéric Rieux-Laucat and Alain Fischer in Paris identified mutations in the FAS gene in children with Canale-Smith syndrome. That same year, an NIH team led by Jennifer Puck, Steven Straus, and Michael Lenardo independently identified FAS mutations in American patients with similar clinical features. The disorder was renamed Autoimmune Lymphoproliferative Syndrome, or ALPS. A key genetic insight was that many cases involve dominant-negative FAS mutations. A single mutated allele can sabotage Fas signaling even when the other allele is normal, explaining autosomal dominant inheritance in many families.

Clinically, ALPS often presents early in childhood with persistent enlarged lymph nodes and spleen, along with autoimmune destruction of blood cells such as hemolytic anemia and thrombocytopenia. A striking laboratory hallmark is the expansion of CD3-positive double-negative T cells, sometimes comprising 15 to 30 percent of circulating T cells. These cells mirror the abnormal population seen in lpr mice and have become a diagnostic signature of the disease.

ALPS also highlighted a deep connection between apoptosis and cancer risk. When lymphocytes that should die persist, they not only drive autoimmunity but also increase the chance of malignant transformation. Patients with ALPS have an elevated risk of lymphoma. The connection works in the opposite direction as well: many tumors evade immune attack by downregulating Fas or exploiting FasL to eliminate infiltrating T cells.

Epilogue: From accidental discovery to modern immunology (1996-Present)

The Fas story is a textbook example of biology progressing from accident to mechanism to medicine. In 1989, antibody-induced cell death looked like a laboratory oddity. In the early 1990s, Fas and FasL became molecular entities with defined domains, defined signaling partners, and defined genetics, validated by two classic mouse mutants. By 1995, the same pathway explained a rare human disease, transforming a clinical curiosity into a diagnosable genetic syndrome.

Since then, the picture has only grown more nuanced: we now know that Fas signaling is not always a one-way route to apoptosis. Depending on cellular context and downstream pathway integrity, CD95 can also engage non-apoptotic signaling, including inflammatory or survival pathways. This complexity helps explain why directly activating Fas has proven risky as a therapeutic strategy, since some tissues, particularly the liver, are exquisitely sensitive to death-receptor signaling.

Immune tolerance is not a single mechanism but a layered system. When one layer fails, the result can be immunological chaos. Imagine a self-reactive clonotype that somehow slips through thymic negative selection, which happens from time to time because AIRE dramatically expands the self-antigen “training set,” but the thymus is not an omniscient filter, and some potentially problematic T cells still graduate from thymus. That is where Fas starts to matter most: once a T cell is repeatedly stimulated, especially in a chronic setting like persistent self-antigen exposure (leading to autoimmunity), the immune system can invoke a built-in off switch called activation-induced cell death. In effect, Fas helps ensure that a self-reactive T cell that “passed the exam” but misbehaves later does not get to stick around indefinitely.

So, in the case of self-reactive T cells and an immune system that cannot stop responding, Fas has taught immunology one of its most counterintuitive truths: controlled cell death is essential for a proper immune system.

References

Part III (you are reading this)

AIRE: The Orchestrator of Self-Tolerance

2026-01-01T00:00:01+00:00

Part II (you are reading this)

Your immune system is a powerful guard dog, but it needs training. From the moment you are born, it has to learn the difference between intruders and you. Most of us never notice that training happening. But in rare cases, the lesson plan goes unexpectedly wrong, and the immune system grows up confused, lashing out at the body while letting some infections slip through.

This post follows the trail from baffling patient cases to an unexpected culprit called AIRE, a tiny genetic teacher that helps the immune system recognize itself.

The Mystery of the Missing Immune Tolerance (1926 to 1946)

In the early 20th century, doctors began noticing a strange pattern in a small number of children. These patients did not just have one problem. They had a cluster of them: thrush infections that kept coming back no matter what clinicians tried; dangerously low calcium from failing parathyroid glands, sometimes severe enough to trigger seizures; and later, signs of Addison’s disease, when the adrenal glands stopped producing essential hormones, often accompanied by a deep bronzed pigmentation that looked almost like an unnatural tan.

Individually, each diagnosis was recognizable. Together, they were baffling. In 1926, Dr. Ernest Schmidt in Germany noted the curious association between persistent Candida infections and underactive parathyroid glands. By 1946, clinicians recognized that these issues often appeared alongside autoimmune adrenal failure, defining a syndrome that seemed to combine immune weakness and immune misdirection in the same person.

Much later, this condition would be given two names that referred to the same disease: APS-1 (Autoimmune Polyglandular Syndrome Type 1) and APECED (Autoimmune Polyendocrinopathy-Candidiasis-Ectodermal Dystrophy). The names captured the breadth of what patients experienced, but they did not solve the underlying question: why would the immune system attack multiple organs while still failing to control a common fungus?

How the Syndrome Came Into Focus (1980s to 1990)

As more cases accumulated, the outline of the disorder became clearer, especially through the work of Finnish pediatric endocrinologist Jaakko Perheentupa. In an era when rare diseases were often described through isolated case reports, Perheentupa did something unusually patient: he followed families over years, sometimes decades, watching the syndrome unfold rather than treating it as a static diagnosis.

In a landmark 1990 study, he and colleagues documented 68 cases, many in Finland, helping establish what APS-1 looked like over time rather than at a single moment. One insight stood out immediately: the syndrome rarely arrived all at once. Many patients developed two major features in childhood, with the third appearing later. This staging explained why earlier reports had seemed inconsistent or incomplete.

The second insight was genetic. APS-1 behaved like an autosomal recessive condition, clustering in families and appearing more often in communities with shared ancestry. Clinicians converged on a practical diagnostic rule: at least two of the classic triad, candidiasis, hypoparathyroidism, and Addison’s disease.

By the early 1980s, researchers also recognized that this syndrome was distinct from a more common adult-onset disorder that also included Addison’s disease. In 1981, immunologist Markus Neufeld formalized the naming: APS Type 1 for the childhood-onset triad, and APS Type 2 for adult-onset Addison’s plus other autoimmune endocrinopathies. Around the same period, the term APECED gained popularity because it highlighted features beyond the endocrine glands, especially ectodermal changes affecting teeth, nails, skin, and hair. Other labels appeared briefly, but APS-1 and APECED became the enduring names.

APS-1 vs APECED: What’s in a Name?

APS-1 and APECED describe the same disease.

Term	Emphasis
APS-1	Autoimmune, polyglandular nature
APECED	Full clinical spectrum including ectodermal defects

The acronym may be awkward, but it reflects an important truth: this disorder is more than just endocrine autoimmunity.

A Global Clue: Founder Populations (1990s)

One of the strongest hints that APS-1 was caused by a single gene came from geography. Worldwide, the disease was extremely rare, but it appeared much more often in a few genetically isolated populations: Finland, Sardinia, and Iranian Jewish communities. This pattern was classic for a founder effect, a mutation arising in a small ancestral group and becoming common over generations. Before whole-genome sequencing existed, geography itself functioned as a genetic tool. Population history narrowed the search space long before molecular methods could.

Israeli geneticist John (Yehuda) Zlotogora reported that many Iranian Jewish cases shared the same founder mutation, suggesting a single ancestral origin. In parallel, clinicians like Piero Betterle in Italy and Eystein Husebye in Norway broadened the clinical map across Europe, showing that APS-1 looked remarkably consistent in its core features while varying in which organs were affected and when.

By the early 1990s, the consensus was clear. APS-1 was not random, and it was not multifactorial in the usual sense. It looked like a single recessive gene disorder. The hunt for that gene was on.

Gene Hunters and the AIRE of Discovery (mid-1990s)

That hunt accelerated in the mid-1990s, when Finnish geneticist Leena Peltonen-Palotie and collaborators used large family pedigrees to map APS-1 to chromosome 21q22.3. Peltonen-Palotie was known for treating Finland’s population history not as a limitation, but as an experimental advantage. At a time when many geneticists favored large, diverse cohorts, she argued that bottlenecks and founder effects could turn rare diseases into solvable problems.

APS-1 became one of her signature successes. In 1997, two independent teams arrived at the same answer almost simultaneously. A Finnish-German consortium associated with Peltonen’s work and an international NIH-backed group led by Kentaro Nagamine identified the same previously unknown gene whose disruption explained the syndrome.

In the competitive climate of 1990s human genetics, simultaneous discovery often led to disputes over priority. In this case, it did the opposite. Two independent paths converged on the same gene, removing doubt and quickly unifying the field around a single name: AIRE, short for Autoimmune Regulator.

From the start, AIRE’s protein sequence looked like it belonged in the nucleus rather than on the cell surface. It contained two PHD zinc finger domains, proline-rich regions, and LXXLL motifs associated with transcriptional regulation. PHD stands for Plant Homeodomain, a zinc-binding motif found in many chromatin-associated regulators. This architecture suggested that AIRE controlled other genes rather than acting as a structural or signaling protein. It looked like a master switch, exactly the kind of mechanism one might expect in a disorder where immune tolerance collapses across multiple organs.

APS-1 also made history. It became one of the first clear examples of a single-gene autoimmune disease discovered outside the HLA region.

Founder Mutations Identified (late 1990s)

The gene discovery immediately explained the founder populations. Different groups tended to carry different recurring mutations:

Population	Common Mutation
Finns	R257X (70-90% homozygous)
Sardinians	R139X (~82% of alleles)
Iranian Jews	Y85C (missense, unstable protein)

Each population carried its own genetic signature, but the conclusion was the same. When AIRE failed, self-tolerance failed with it.

AIRE’s Job: Teaching the Immune System ‘Self’ (2001 to 2002)

Finding AIRE solved the genetic mystery, but it raised a deeper question: what did AIRE actually do? The answer came from work focused on an organ most people rarely think about: the thymus, where developing T cells learn the difference between self and not-self. In 2001, immunologist Bruno Kyewski described something that initially sounded almost implausible. Certain thymic cells expressed an eclectic mix of genes normally restricted to entirely different organs.

At the time, tissue specificity was considered almost sacred. The idea that a thymic cell might express insulin, eye proteins, or liver enzymes sounded like experimental noise. Kyewski argued otherwise: he proposed that the thymus deliberately created an internal self portrait by turning on tissue-specific genes in the wrong place, for the right reason. These specialized cells, medullary thymic epithelial cells, used this so-called promiscuous gene expression to present a broad sampling of self-antigens to developing T cells during training.

The decisive tests came in 2002. Two groups independently created Aire-deficient mice. One of the most influential efforts came from the lab of Diane Mathis and Christophe Benoist, a scientific partnership known for turning abstract immunological ideas into clean experiments. Their team, with Mark Anderson playing a central role, removed Aire and watched a familiar pattern unfold: multi-organ autoimmunity, high autoantibody levels, and immune attacks on tissues that should have been protected.

What mattered most was not just that the mice became sick, but how. Without Aire, many tissue-specific antigens were no longer expressed in the thymus. The immune system graduated with gaps in its education. Self-reactive T cells that should have been eliminated survived, escaped into circulation, and later attacked organs that had never appeared in the thymic lesson plan. Parallel work linked to Peltonen’s collaborations confirmed that Aire’s critical action occurred within the thymic environment itself.

Mark Anderson’s later work added another twist. APS-1 patients often produce autoantibodies against type I interferons, a finding that helped explain why infection susceptibility and autoimmunity can coexist in the same syndrome. It also provided clinicians with a distinctive diagnostic signal, connecting bench discoveries back to patient care.

Epilogue: From Rare Disease to Foundational Immunology

We now know that as part of the central tolerance mechanism, developing T cells are constantly tested in the thymus. If a T cell reacts too strongly to self, it is removed, a process known as negative selection. AIRE’s key role is to help thymic cells display a wide range of self-antigens, including proteins normally found only in peripheral tissues. When AIRE is missing or broken, the immune system does not get the full picture of self, and dangerous self-reactive cells can slip through.

That understanding feels straightforward today, but much of it once sounded strange. The idea that the thymus would deliberately violate tissue specificity, or that a single gene could orchestrate such a broad tolerance program, ran against intuition. Autoimmunity was often framed as irreducibly complex. AIRE made a simpler and more unsettling point. Sometimes one broken part is enough to destabilize the whole system.

Even the gene hunt itself reflects a moment in time. In the 1990s, identifying a disease gene was closer to detective work than pipeline biology. Founder populations, family pedigrees, and careful mapping through anonymous stretches of DNA did the heavy lifting. When two independent teams converged on AIRE in 1997, it felt less like routine progress and more like resolution.

Today, APS-1/APECED remains rare, and treatment is still largely supportive, hormone replacement for endocrine failure and antifungals for chronic infections. But the discovery of AIRE transformed the condition from a puzzling clinical triad into a foundational lesson about the immune system. It showed that immune tolerance depends on a real biological process, in a real place, driven by specific genes. It also changed how the thymus is viewed, not as a passive organ you outgrow, but as an active classroom with a surprisingly broad syllabus.

In that sense, APS-1 did something remarkable. It took a rare disorder and used it to explain something universal: how the immune system learns when to fight, and when to leave home alone. And it reminds us that what feels obvious in hindsight is often the result of a few brave ideas that once sounded strange.

References

Part II (you are reading this)

RAG1 and RAG2: Discovery, Mechanism, and Evolution of the V(D)J Recombinase

2025-12-29T00:00:01+00:00

Part I (you are reading this)

The adaptive immune system depends on a controlled act of genomic violence. To recognize an almost infinite variety of pathogens, developing lymphocytes deliberately break and rejoin their own DNA, assembling antigen receptor genes from modular fragments. For decades, immunologists understood the outcome of this process but not its cause: what enzyme could cut the genome so precisely, and why did it act only in immune cells?

The answer emerged at the end of the 1980s with the discovery of RAG1 and RAG2, two genes that together form the molecular “scissors” of V(D)J recombination. Their identification not only solved a long-standing mystery in immunology but also revealed that the origins of adaptive immunity lie in an ancient, repurposed genetic element. This account follows the scientific hunt for the RAG genes and the far-reaching consequences of their discovery.

The Hunt for the Immunological “Scissors” (1970s–1980s)

In the 1970s, Susumu Tonegawa (a Japanese molecular biologist trained in Tokyo and California) stunned the scientific world by showing that antibody genes rearrange in B cells, which was a finding that earned him the 1987 Nobel Prize. This process, now called V(D)J recombination, explained how a finite genome could produce the vast diversity of antibodies and T-cell receptors. Yet a mystery remained: what molecular “scissors” cut and rejoin DNA to make these receptor genes? Through the 1980s, immunologists imagined a dedicated V(D)J recombinase that was selectively expressed in lymphocytes was behind this phenomenon. The race was on to find it. Remarkably, Tonegawa did not participate in this race for too long as after securing his place in immunology history, he switched fields to neuroscience, applying his scientific boldness to memory and learning research.

By the late 1980s, one approach to find the elusive recombinase was brute-force genetics. Among those intrigued was David Baltimore, already famous for co-discovering reverse transcriptase (Nobel Prize, 1975) and now pivoting to immunology. Baltimore, at MIT’s Whitehead Institute, suspected that transferring the right gene into non-immune cells might confer the ability to perform V(D)J recombination. It was a high-risk idea—like finding a needle in a genomic haystack—but worth a shot.

Cloning RAG1 and RAG2: Breathing Life into Fibroblasts (1988–1990)

In 1988, Baltimore’s team (most notably graduate students David Schatz and Marjorie Oettinger) developed an ingenious assay. They engineered a reporter plasmid containing artificial gene segments and RSSs (Recombination Signal Sequences) that would produce a detectable signal only if V(D)J recombination occurred. This reporter was transfected into NIH-3T3 fibroblasts, cells normally incapable of gene rearrangement, along with immune-cell DNA.

Through serial genomic transfections, they isolated a genomic locus that conferred recombination activity. In late 1989, they identified a gene encoding a 1040–amino acid protein that activated V(D)J recombination in fibroblasts. They named it RAG1 (Recombination Activating Gene 1). RAG1 expression was restricted to lymphoid tissues—exactly where the scissors should be active.

This was a great accomplishment but a piece of the puzzle was still missing: RAG1 alone was not sufficient for maximal activity, potentially hinting at a missing partner that was yet to be discovered. 2 years after their discovery of RAG1, in June 1990, Oettinger and Schatz reported RAG2, a gene adjacent to RAG1 that synergizes with it to produce full recombination activity. Interestingly, RAG2 was lacking obvious enzymatic motifs but without RAG2, RAG1 was a dull knife. Only together, RAG1 and RAG2 reconstituted the V(D)J recombinase at expected levels.

The molecular scissors were a two-part machine. After these breakthough findings, then-graduate-students David Schatz and Marjorie Oettinger eventually became leaders thmselves and continued working on V(DJ) recombination and the RAG proteins. David Baltimore later reflected that identifying RAG proteins was among his lab’s proudest achievements.

Proof by Knockout: No RAG, No Immune System (1992)

Discovery in vitro was only the beginning. In 1992, two groups independently generated RAG knockout mice. Published back-to-back in Cell, the results were unequivocal.

Peter Mombaerts (of Tonewaga lab) showed that in RAG1-KO mice, there were no mature B or T lymphocytes; thymus and lymph nodes were severely underdeveloped; immunoglobulin and TCR genes remained unrearranged; and interestingly the mice phenotype resembled a known human disease (more on this later). In addition to these, with this knockout mouse, Mombaerts showed that despite low-level brain expression reports of RAG1, RAG1-KO mice were neurologically normal. Like his mentor, Mombaerts also switched to neurogenetics and eventually become a leader in olfactory system research.

In parallel, Yuko Shinkai (of Frederick Alt lab) showed that in RAG2-KO mice, early lymphoid progenitors were present; V(D)J recombination initiation was completely failing; there was no detectable rearranged antigen receptor genes; and reintroduction of RAG2 rescued recombination in cultured cells. Shinkai, later, followed RAG2’s link to histone modifications and built his careers around epigenetics by studying chromatin regulation.

Together, these two amazing knockouts and their characterization proved that RAG1 and RAG2 were absolutely required for adaptive immunity. The details of how this pair of proteins accomplished recombination followed later on.

RAG Mechanism and Evolution: Jumping Genes to Immune Genes

Biochemical studies in the mid-1990s revealed that: RAG1 contained the catalytic DDE motif, typical of transposases; RAG2 acted as an essential cofactor; and core RAG1/2 proteins alone could catalyze recombination in vitro. Then came the stunning discovery of RAG proteins’ ability to mediate DNA transposition under artificial conditions, which led to the transposon hypothesis that RAG genes originated from an ancient mobile genetic element. It wasn’t until 2016 that the discovery of ProtoRAG in amphioxus provided compelling evolutionary evidence to this claim.

Epilogue: Legacy of the RAG Story

The discovery of RAG1 and RAG2 resolved a decades-old mystery and reshaped immunology, genetics, evolution, and medicine. What began as a search for DNA scissors revealed the molecular basis of immune diversity, the evolutionary domestication of transposons, and the genetic roots of immunodeficiency and autoimmunity. We now know that severe Combined Immunodeficiency (SCID) is a human disease that is caused by null RAG mutations. When these RAG mutations are hypomorphic, they then lead to Omenn syndrome where patients show very limited T cell development and autoimmune-like inflammation. In some cases, bone marrow transplantation from healthy donors can cure these conditions by restoring functional RAG genes.

From clever reporter assays to definitive knockout models, the RAG story exemplifies how bold ideas, technical rigor, and human curiosity converge to answer fundamental biological questions. It is very interesting that two unassuming genes changed how we understand the immune system and life’s capacity for innovation.

References

Part I (you are reading this)