The immune system is arguably one of the most impressive systems in the human body, able to identify foreign invaders and destroy them while leaving the “self” alone. Its success depends on its ability to determine what belongs to the self in the first place: the cells and tissues that belong to the human it operates within, and the larger picture of organs, muscles, and nerves. Early immunologists considered this distinction so fundamental that an immune attack against the self appeared biologically impossible.
But the discovery of autoimmune diseases bulldozed that thinking and revealed an even more complex system of protection, one that is not inherent but constantly learning, tolerating, and fighting against self-destruction.
Impossible!
The immune system is built to protect the body from foreign invaders, pathogens, and cells that shouldn’t be there by seeking them out and destroying them, like the body’s greatest fleet of military missiles. Autoimmune disease begins when these missiles are turned inward to the self, and medicine has no simple switch to turn them away.
Patients today with rheumatoid arthritis (RA), systemic lupus erythematosus, and multiple sclerosis (MS), among others, frequently describe their diseases to the layperson as a result of their immune system mistakenly attacking their disease’s bodily structure of choice. In RA, the damage primarily affects the joints, although the disease can involve other organs. Lupus can affect nearly any part of the body, while MS damages structures within the central nervous system.
If I were to travel back in time with this same explanation and go to immunologists of the early 20th century, telling them my immune system is attacking myself, there’s a good chance I wouldn’t be believed. Why? In the early 20th century, scientists could not fathom that the immune system was capable of anything other than protecting the body in which it lived.
A 20th Century House of Horrors
At the time, the immunology dogma was that the immune system was so innately fine-tuned to protect the body that it was impossible to think it could cause any harm to the self. This understanding was called ”horror autotoxicus,” a term coined by Paul Ehrlich. Ehrlich was a German-born scientist and physician who worked in the fields of hematology, antimicrobial chemotherapy, and immunology.
He came up with the term “horror autotoxicus,” the horror of self-toxicity, in 1906 in response to some scientists postulating the existence of autoimmune diseases. Ehrlich believed it went against nature for the body to cause self-injury. To prove this, Ehrlich and his assistant conducted experiments with animals, such as goats, injecting blood from a different animal or from the same animal into one animal and observing how the immune system would react. The result was that if an animal had been injected with blood from a different species, antibodies would swarm and attack those cells. Conversely, if they had been injected with the blood of the same animal, Ehrlich noted no change.
This experiment reinforced Ehrlich’s belief in the impossibility of the immune system attacking healthy cells, because why would it? It had no reason to. In Ehrlich’s book “Collected Studies on Immunity,” he wrote, “the organism possesses certain contrivances by means of which the immunity reaction, so easily produced by all kinds of cells, is prevented from acting against the organism’s own elements and so give rise to autotoxins.”
Just two years after he published that statement, Ehrlich shared the 1908 Nobel Prize in Physiology and Medicine with Élie Metchnikoff for their work on immunity.
Although Ehrlich had become one of immunology’s leading scientists and proposed the mechanics that govern antibody responses against foreign cells, Ehrlich had overestimated the reliability of these mechanics. But because of Ehrlich’s authority on the matter, it would take decades for the mechanics of autoimmune disease to be accepted, even with mounting evidence.
Autoimmunity without Autoimmune Disease
The mounting evidence came from some scientists who were discovering the very thing Ehrlich said was against nature. Just a few years before Ehrlich received praise for the biological safeguards that prevent antibodies from attacking one’s own cells, two scientists had discovered the opposite.
Julius Donath and Karl Landsteiner, two Austrian-born physicians, began studying a rare blood condition, later named paroxysmal cold hemoglobinuria (PCH). Patients presented with extreme red blood cell destruction and dark urine after being exposed to cold temperatures. Upon further investigation, Donath and Landsteiner found antibodies that bound to the red blood cells of these patients and destroyed them when the body was warmed back up.
PCH today is understood as a very rare form of autoimmune anemia, and one of the first known autoimmune diseases thanks to the Austrian physicians.
Across Asia, a similar discovery was unfolding in Japan. Dr. Hakaru Hashimoto had just graduated from medical school in Imperial Japan and, while studying thyroid tissue extracted from four thyroid glands of middle-aged women, noticed an unusual pattern. Each gland showed dense lymphocytic infiltration (white blood cells) and cellular degeneration of normal thyroid cells. In 1912, Hashimoto published his findings in a German medical journal, marking a landmark in the recognition of thyroid disease that would later be called Hashimoto’s disease. But the connection to the immune system was still missing.
Proof Against the Self
The 1950s were among the most decisive decades in the recognition of autoimmune disease. When it was still loosely recognized, scientists around the world were working to prove that the immune system was, in fact, attacking the body’s own cells in various experiments. Doctors Ernest Witebsky and Noel Rose started with rabbits. In 1956, Witebsky and Rose injected rabbits with thyroglobulin, a protein produced by the thyroid, to determine whether the rabbits would mount an immune response. They had already seen foreign thyroglobulin induce an immune response, so what would happen if the rabbit’s own thyroglobulin was injected?
The answer: an autoimmune response. The rabbits’ thyroids exhibited significant inflammation and tissue damage upon examination, and Witebsky and Rose continued even further. They tested human blood samples from patients with Hashimoto’s, looking for thyroglobulin-specific antibodies. They were found. Rose later recalled Whitebsky telling him, “You have fulfilled the postulates and proven that a human disease can be caused by autoimmunity.”
Meanwhile, Doctors Deborah Doniach and Ivan Roitt would produce complementary evidence with Hashimoto’s antibodies in patients with the condition. Alongside their own published research and the animal studies, the existence of autoimmunity was becoming clear. It was no longer an impossibility that the body could turn on itself, but it was scientifically provable. What was missing was the mechanisms.
Learning “Self”
What followed these scientific landmarks was decades of research into how the immune system knows not to attack the self, and what happens when that knowledge goes astray. The answer is not that the immune system automatically knows not to unleash havoc on the self (much like Ehrlich thought), but that it tolerates the self, a learned tolerance that requires multiple layers of security.
The immune system has layers of tolerance that prevent B and T cells from going AWOL and attacking self tissues. Central tolerance acts as a screening or exam for these cells. T cells undergo this screening in the thymus, while B cells are screened primarily in the bone marrow. Their receptors are partly generated at random, causing some cells to have a stronger reactivity to self than others. Cells that react too strongly undergo cell death or inactivity, and in the case of B cells, receptor editing to lessen self-reactivity. It is the first, and toughest, act of security for these cells before they’re released into the bloodstream.
What happens next is a long line of regulations constantly keeping the cells in check. Peripheral tolerance partly relies on regulatory immune cells (T-Regs) performing safety checks on other immune cells, ensuring they remain inactive or are suppressed in their responses. If those immune cells are found to be self-reactive, they are sent calming signals, or told to initiate cell death.
The prevention of autoimmune disease depends on overlapping systems of immune tolerance that monitor self-reactive cells throughout their development and after they enter circulation. This system, like a set of locks with specific keys, is meant to work down a chain of commands, signals, and ordered tasks. But what happens when something is unlocked, and you lose the key, leaving you unable to lock the mechanism back into safety?
A Perfect Immunological Storm
The safety mechanisms that are supposed to keep the immune system a servant to the self have been uncovered and scrutinized with a fine-tooth comb, winning both the 1960 and 2025 Nobel Prizes and piecing together groundbreaking insights into how these systems contribute to autoimmune diseases.
All the science has led to the understanding that autoimmune diseases, no matter how they work mechanically because of these haywire systems, do not stem from a single cause. What has sent scientists down a rabbit hole for decades is the search for all the possible contributions that cause immune cells to turn on the self.
Combinations of triggers range from genetic susceptibility and infections to environmental exposures and disruptions of the gut microbiome, like different picks that can test different locks in the immune tolerance system.
The House Still Standing
Understanding, after over a century, the mechanisms by which the immune system turns on the self only goes so far. It still raises questions about why cells decide to attack an RA patient’s joints, versus an MS patient’s nerves, or a lupus patient’s kidneys, and why in that moment they chose to do so.
Knowing as much as we do, it feels as if we’re on the precipice of understanding, or solving, the immune system’s horror of itself, just barely able to reach out and tangibly grasp the underlying explanation for why a disease affects one person, organ, or stage of life.
There is much to be discovered. Ehrlich’s “horror autotoxicus” was not wrong. It still stands as the template for proper safeguards: what the immune system is supposed to do to avoid self-destruction. But those safeguards are not incontrovertible law. The immune system was never incapable of attacking us; it’s only learning every day not to.