Amyotrophic lateral sclerosis does not have a single cause. It arises from a collision of genetic vulnerability, environmental exposures, and a cascade of cellular failures that together overwhelm the motor neurons controlling voluntary movement. Research increasingly treats ALS as a multistep process, with population-based models estimating that roughly six distinct events or risk factors must accumulate before symptoms appear. That framing helps explain why the disease strikes some people with a clear family history and others with no obvious genetic link at all, and why decades of searching for “the cause” have instead uncovered dozens of contributing threads.
Genetic Roots of ALS
About one in ten ALS cases runs in families, traditionally labeled “familial ALS.” The remaining roughly ninety percent are called “sporadic,” meaning no obvious inherited pattern is present. But the line between the two is blurrier than those labels suggest, because many sporadic cases turn out to carry genetic risk variants that simply haven’t been traced through a family tree.
The first ALS gene identified was SOD1, which encodes an enzyme involved in clearing harmful oxygen molecules from cells. For years it was the only known genetic contributor. Then the field accelerated. Mutations in TARDBP (which makes the TDP-43 protein) and FUS were linked to ALS, each producing distinct patterns of damage inside neurons despite similar clinical symptoms. In a large U.S. clinical testing cohort, SOD1 mutations appeared in about 7.5% of tested cases, FUS in roughly 2%, and TARDBP in under 1%.1PubMed. SOD1, ANG, TARDBP and FUS mutations in amyotrophic lateral sclerosis: a United States clinical testing lab experience Those three genes account for pathologically distinct subtypes, even though the clinical picture of progressive weakness often looks similar from the outside.2PubMed. The genetics and neuropathology of amyotrophic lateral sclerosis
The biggest genetic discovery in ALS came later: a repeat expansion in the C9orf72 gene, now recognized as the most common genetic cause of both ALS and frontotemporal dementia. In this mutation, a short stretch of DNA is copied hundreds or even thousands of times, and the resulting RNA and abnormal proteins appear to damage cells through multiple mechanisms at once, including loss of the normal C9orf72 protein’s function and toxic buildup of repeat-derived products.3PubMed Central. C9orf72-mediated ALS and FTD: multiple pathways to disease More than thirty other genes have since been linked to ALS risk at varying levels of certainty, and large-scale genetic studies continue to add to the list.
TDP-43 and the Common Thread
If you want to understand what goes wrong at the molecular level in most ALS cases, TDP-43 is the place to look. This protein normally lives inside the nucleus of cells, where it helps manage RNA. In about 97% of ALS cases, regardless of whether a patient carries a known mutation, TDP-43 misbehaves: it moves out of the nucleus, clumps together in the cell’s cytoplasm, and forms dense aggregates.4PubMed Central. The role of TDP-43 mislocalization in amyotrophic lateral sclerosis That process damages cells in two ways at the same time. The nucleus loses TDP-43’s normal RNA-managing functions, and the clumps in the cytoplasm actively interfere with cellular machinery, blocking normal processes and hastening cell death.5PubMed Central. TDP-43-The key to understanding amyotrophic lateral sclerosis
As TDP-43 aggregates, the protein picks up chemical modifications like extra phosphate groups and ubiquitin tags. These changes make the clumps harder for the cell to clear, so the problem snowballs. Researchers have argued that preventing TDP-43 from misfolding in the first place, or boosting the cell’s ability to clean up the aggregates, could be among the most promising treatment strategies for ALS and frontotemporal dementia alike.6PubMed Central. TDP-43 Proteinopathy and ALS: Insights into Disease Mechanisms and Therapeutic Targets The exceptions are SOD1-linked ALS, which typically shows SOD1 protein aggregates rather than TDP-43, and FUS-linked cases, which have their own distinct pathology.
A Six-Step Disease
One influential model treats ALS not as the result of a single trigger but as a multistep chain of events. By plotting how ALS incidence rises with age in large populations, researchers found a mathematical relationship consistent with a process requiring about six sequential steps before the disease emerges.7PubMed Central. Analysis of amyotrophic lateral sclerosis as a multistep process: a population-based modelling study People carrying strong genetic mutations like C9orf72 or SOD1 appear to start with some of those steps already checked off, which is why they tend to develop symptoms earlier and why the model shows a reduced number of remaining steps needed in genetically positive patients.8PubMed Central. The multistep hypothesis of ALS revisited: The role of genetic mutations
What the remaining steps consist of is still debated, but the model accommodates the reality that ALS does not reduce to either genes or environment alone. Environmental exposures, occupational hazards, and lifestyle factors can plausibly fill in the remaining “steps” for someone who has some genetic susceptibility but not enough on its own to cause disease. This framework also helps explain why identical twins don’t always share an ALS diagnosis and why clusters of ALS cases pop up in populations with unusual exposures.
Environmental and Occupational Risks
Because most ALS cases are sporadic, researchers have spent decades hunting for environmental triggers. The evidence is often circumstantial, but several exposures keep appearing across studies. Heavy metals like lead and mercury, pesticides (especially organophosphates), industrial solvents, and agricultural chemicals have all been linked to elevated ALS risk.9PubMed Central. The role of environmental mercury, lead and pesticide exposure in development of amyotrophic lateral sclerosis The earliest environmental clue came from the Chamorro people of Guam, who during the 1950s and 1960s had extraordinarily high rates of ALS, thought to be related to dietary exposure to toxins from cycad seeds and flying foxes that fed on them.
Military service has emerged as another consistent risk factor. Veterans, particularly those who served in the Gulf War or World War II, appear to face elevated ALS risk. Studies point to several possible culprits: exposure to pesticides including Agent Orange, certain chemicals like exhaust and burning agents, heavy metals, and head trauma sustained during service.10PubMed Central. Military service and related risk factors for amyotrophic lateral sclerosis Male sex, smoking, strenuous exercise, and electrical injury have also been explored as possible risk factors, though the strength of evidence varies.11PubMed Central. Potential Environmental Factors in Amyotrophic Lateral Sclerosis Cyanobacteria, which produce a neurotoxic amino acid called BMAA, are under investigation as well, particularly in lakes and waterways where blooms have been tied to geographic clusters of ALS.12PubMed Central. Exposure to environmental toxicants and pathogenesis of amyotrophic lateral sclerosis: state of the art and research perspectives
None of these exposures, on their own, reliably cause ALS. That’s consistent with the multistep model: an environmental exposure might represent one or two of the required steps, but without the right genetic background and other accumulated insults, disease doesn’t develop.
What Breaks Down Inside Cells
Beyond the protein-aggregation problems described above, several cellular systems fail in ALS. One of the most studied is glutamate signaling. Glutamate is the brain’s main excitatory chemical messenger, and motor neurons are particularly sensitive to it. In ALS, several pathways conspire to flood motor neurons with too much glutamate stimulation: the astrocyte transporter that normally clears glutamate from the space between neurons stops working properly, certain receptors on motor neurons become more permeable to calcium, and the inhibitory circuits that normally dampen excitation weaken.13PubMed Central. Revisiting Glutamate Excitotoxicity in Amyotrophic Lateral Sclerosis and Age-Related Neurodegeneration The result is a toxic level of stimulation called excitotoxicity. Riluzole, one of the few approved ALS drugs, works by reducing glutamate’s effects.
Mitochondria, the cell’s energy factories, also deteriorate. They become structurally damaged, produce excessive amounts of harmful reactive oxygen molecules, and their normal transport along the long nerve fibers of motor neurons gets disrupted.14PubMed Central. Mitochondrial dysfunction in amyotrophic lateral sclerosis Motor neurons are among the longest cells in the body, so they’re especially dependent on efficient delivery of mitochondria to distant parts of the nerve fiber. When that delivery system fails, the far ends of the neuron starve of energy.
Another layer of dysfunction involves the traffic system that moves molecules between a cell’s nucleus and its cytoplasm. In healthy cells, this nucleocytoplasmic transport is tightly controlled. In ALS, it becomes impaired, meaning that critical proteins and RNA molecules end up in the wrong part of the cell, compounding the damage from TDP-43 mislocalization.15PubMed Central. Lost in Transportation: Nucleocytoplasmic Transport Defects in ALS and Other Neurodegenerative Diseases
The Neighbors Turn Hostile
Motor neurons do not die in isolation. The cells surrounding them, particularly astrocytes and microglia, play a double-edged role. Early in disease, microglia (the brain’s resident immune cells) attempt to protect neurons by clearing debris and releasing supportive signals. But as the disease progresses, both microglia and astrocytes shift toward a toxic state, releasing inflammatory molecules and reactive oxygen species that accelerate motor neuron death.16PubMed Central. Microglia and Astrocytes in Amyotrophic Lateral Sclerosis: Disease-Associated States, Pathological Roles, and Therapeutic Potential Once this shift passes a critical threshold, the inflammation becomes self-reinforcing and the process of motor neuron killing becomes a runaway reaction rather than a controlled immune response.17Experimental Neurobiology. Astrocytes and Microglia as Non-cell Autonomous Players in the Pathogenesis of ALS
Oligodendrocytes, which produce the insulating myelin sheath around nerve fibers in the brain and spinal cord, add yet another dimension. In ALS, oligodendrocytes fail to adequately support motor neurons by supplying them with lactate, a critical energy source. A transporter called MCT1, which normally shuttles lactate from oligodendrocytes to axons, is reduced in both ALS mouse models and patient tissue.18Trends in Cell Biology. Revisiting oligodendrocytes in amyotrophic lateral sclerosis using human multicellular stem cell models Oxidative stress also damages the precursor cells that would normally replace worn-out oligodendrocytes, leaving motor neurons further starved of metabolic support.19PubMed Central. Dysfunction of the oligodendrocytes in amyotrophic lateral sclerosis
A Leaky Barrier
The spinal cord is supposed to be sealed off from the bloodstream by a tight barrier of specialized blood vessel cells. In ALS, this blood-spinal cord barrier breaks down, and it appears to do so early, even before motor symptoms begin.20PubMed Central. Blood-CNS barrier dysfunction in amyotrophic lateral sclerosis: Proposed mechanisms and clinical implications Post-mortem studies of ALS patients have found blood-derived proteins like hemoglobin, immunoglobulin, fibrin, and thrombin deposited in the spinal cord tissue where they don’t belong. Pericytes, the cells that help maintain the barrier’s structural integrity, are reduced in number.21PubMed Central. Blood-spinal cord barrier breakdown and pericyte reductions in amyotrophic lateral sclerosis
In mouse models of SOD1-linked ALS, leakage of blood components including hemoglobin and iron into the spinal cord triggers oxidative damage to motor neurons. Repairing the barrier early in the disease course slowed the disease process in those mice, suggesting that barrier breakdown is not just a consequence of disease but an active contributor to it.22PubMed Central. Blood-spinal cord barrier disruption contributes to early motor-neuron degeneration in ALS-model mice
Epigenetics as a Bridge Between Genes and Environment
If genes load the gun and environment pulls the trigger, epigenetics is arguably the mechanism that positions the finger on the trigger. Epigenetic modifications are chemical changes to DNA or the proteins that package it. They don’t alter the genetic code itself but affect which genes are turned on or off. In ALS, changes in DNA methylation patterns and histone modifications have been found in both brain tissue and blood.23PubMed Central. Epigenetics in amyotrophic lateral sclerosis: a role for histone post-translational modifications in neurodegenerative disease
A large blood-based study of nearly 10,000 samples identified 45 specific DNA locations where methylation differed between ALS patients and healthy controls. These sites clustered in genes involved in metabolism, cholesterol production, and immune function.24PubMed. Genome-wide study of DNA methylation shows alterations in metabolic, inflammatory, and cholesterol pathways in ALS Because methylation patterns can reflect past environmental exposures as well as disease activity, these findings offer a potential window into how toxic exposures or lifestyle factors physically change gene activity in ways that push cells toward the disease.
Dormant Viruses Waking Up
One of the more surprising lines of ALS research involves ancient virus sequences embedded in human DNA. About 8% of the human genome is made up of endogenous retroviruses, remnants of viral infections our ancestors survived millions of years ago. These sequences are normally silenced. But in a subset of ALS patients, one particular family of these viral remnants, called HERV-K (subtype HML-2), appears to reactivate. Researchers found HERV-K expressed in cortical and spinal motor neurons of ALS patients but not in healthy controls.25PubMed Central. Human endogenous retrovirus-K contributes to motor neuron disease
When the HERV-K envelope protein was expressed in human neurons in the lab, it caused nerve fibers to retract and break apart. Mice engineered to produce this protein developed progressive motor problems, lost brain volume in the motor cortex, and showed damage to synapses and DNA in their neurons. More recent analyses of nearly 2,000 ALS and control tissue samples confirmed that certain HERV-K loci were significantly upregulated in the spinal cord of ALS patients, though the extent varied from person to person.26iScience. Endogenous retroviruses are dysregulated in ALS This doesn’t mean a virus “causes” ALS in the traditional sense. Rather, something about the disease process, or the underlying susceptibility, appears to unlock viral sequences that then amplify neuronal damage. Clinical trials targeting HERV-K with antiretroviral drugs are underway.27Neurotherapeutics. Neurotherapeutics
The Gut Microbiome Connection
Research into gut bacteria and ALS is young, but findings so far suggest the gut is not irrelevant. ALS patients show signs of dysbiosis, with reduced microbial diversity and shifts in which species dominate. Intestinal inflammation and changes in gut-barrier integrity have been correlated with the disease.28PubMed Central. Microbiome and micronutrient in ALS: From novel mechanisms to new treatments One study found that ALS patients had increased abundances of certain bacterial groups and reduced levels of beneficial species. These patterns also differed between ALS subtypes, with spinal-onset and bulbar-onset patients showing distinct microbial profiles.29Scientific Reports. Examining the complex Interplay between gut microbiota abundance and short-chain fatty acid production in amyotrophic lateral sclerosis patients shortly after onset of disease
Whether these changes are a cause, a consequence, or an accelerant of ALS remains unclear. The concept of a “gut-microbiome-neuron axis” suggests that microbial metabolites could influence inflammation and nerve health systemically, but no one has yet established that altering the microbiome changes ALS outcomes in humans. It’s an area to watch, not an area to act on.
Catching the Disease Earlier
One practical consequence of understanding what causes ALS is the ability to detect it sooner. Currently, diagnosis relies on clinical examination and ruling out other conditions, a process that often takes a year or longer. Biomarker research has focused on neurofilament light chain (NfL), a protein released into spinal fluid and blood when nerve fibers are damaged. In people already showing early ALS symptoms, NfL levels in both spinal fluid and blood are sharply elevated and can distinguish ALS from other neurological conditions with high sensitivity and specificity.30PubMed. Multicenter evaluation of neurofilaments in early symptom onset amyotrophic lateral sclerosis
Even more striking, NfL appears to rise before symptoms begin. In people who carry ALS-causing mutations but haven’t yet developed symptoms, NfL levels were elevated roughly twelve months before the earliest clinical signs appeared.31PubMed Central. Neurofilament light: A candidate biomarker of presymptomatic amyotrophic lateral sclerosis and phenoconversion Prediagnostic blood samples from people who went on to develop sporadic ALS also showed elevated NfL, suggesting this marker could eventually enable earlier diagnosis in the general population, not just in genetically at-risk families.32PubMed Central. Prediagnostic Neurofilament Light Chain Levels in Amyotrophic Lateral Sclerosis Earlier detection matters because future therapies will likely be most effective when started before extensive motor neuron loss has occurred.
Gene-Targeted Treatments
The clearest proof that understanding causes can change treatment is tofersen, an antisense oligonucleotide drug designed specifically for people with SOD1 mutations. Tofersen works by intercepting the messenger RNA that cells use to build the harmful SOD1 protein, reducing how much of it gets made. Delivered by injection into the spinal fluid, it lowered SOD1 protein concentrations in the spinal fluid of treated patients.33PubMed. Phase 1-2 Trial of Antisense Oligonucleotide Tofersen for SOD1 ALS A larger phase 3 trial confirmed that tofersen reduced SOD1 protein levels and showed signals of slowed decline, though the primary clinical endpoint was not met on the original timeline.34PubMed. Trial of Antisense Oligonucleotide Tofersen for SOD1 ALS The drug was approved by the FDA in 2023 under an accelerated pathway.
Tofersen only helps the small fraction of patients with SOD1 mutations. But it established a template: identify the genetic cause, design a molecule that dials down the harmful protein, and deliver it to the nervous system. Similar antisense strategies are being developed for C9orf72 repeat expansions and other genetic targets.35PubMed Central. Antisense Oligonucleotides (ASOs) in Motor Neuron Diseases: A Road to Cure in Light and Shade For the vast majority of patients with sporadic ALS, gene-targeted drugs are not yet an option, but therapies aimed at TDP-43 aggregation, neuroinflammation, and glutamate excitotoxicity are all in various stages of development.
Where Does the Disease Actually Start in the Body
Even among ALS researchers, there’s an unresolved debate about where the disease process physically begins. Three competing hypotheses have persisted for decades. The “dying-forward” hypothesis proposes that the disease starts in the brain’s motor cortex, where overactive upper motor neurons send toxic levels of excitatory signaling down to lower motor neurons in the spinal cord, killing them through a kind of excitotoxic cascade. The “dying-back” hypothesis flips this, arguing that damage begins at the nerve terminals in muscles and travels backward into the spinal cord and eventually the brain. A third model suggests upper and lower motor neurons degenerate independently of each other, affected by the same disease process but not necessarily in a cause-and-effect relationship.36PubMed Central. History of ALS and the competing theories on pathogenesis: IFCN handbook chapter
The answer likely varies by patient and by genetic subtype. Evidence from neuroimaging and clinical neurophysiology has increasingly supported the dying-forward model in many sporadic cases, but SOD1 mutations, for instance, sometimes present with primarily lower motor neuron signs, which fits the dying-back model better. This isn’t just an academic question. If the disease starts in the cortex, early interventions aimed at cortical excitability could theoretically slow progression. If it starts at the periphery, protecting nerve-muscle junctions might be more valuable. The reality is that effective treatment will probably need to address multiple sites simultaneously, which circles back to why ALS has been so resistant to single-target therapies.