The spike protein of SARS-CoV-2, whether produced during infection or generated briefly by vaccination, can linger in the body far longer than originally expected and trigger a cascade of harmful effects across multiple organ systems. Researchers have detected spike protein in human tissues months after the initial exposure, and laboratory studies have connected its presence to blood vessel inflammation, abnormal clotting, heart cell damage, brain inflammation, and premature cellular aging. These findings form much of the biological basis for long COVID and, in rarer cases, post-vaccination symptoms, and the picture is still evolving as new mechanisms come to light.
Spike Protein Sticks Around Longer Than Anyone Predicted
Early assumptions held that the spike protein would be cleared from the body within days to weeks. That turns out to be wrong for a meaningful fraction of people. A study examining skull bone marrow from patients who died of non-COVID causes in 2021 and 2022 found spike protein in about a third of the samples, even though these individuals had recovered from their infections well before death.1Cell Press. SARS-CoV-2 spike protein persists in the skull-meninges-brain axis and induces neuroinflammation and neurodegeneration in mice Separately, researchers measuring blood serum in people with long COVID found both infection-derived and vaccine-derived spike protein persisting for months, challenging the idea that the body rapidly eliminates it.2European Review for Medical and Pharmacological Sciences. Presence of viral spike protein and vaccinal spike protein in the blood serum of patients with long-COVID syndrome
The spike protein’s persistence appears to involve immune cells themselves. In people reporting prolonged symptoms after vaccination, flow cytometry detected the S1 subunit of spike protein inside a specific class of immune cells called nonclassical monocytes in 11 out of 12 symptomatic patients, with detection as late as 245 days after vaccination.3Taylor & Francis. Detection of S1 spike protein in CD16+ monocytes up to 245 days in SARS-CoV-2-negative post-COVID-19 vaccine syndrome (PCVS) individuals The protein was essentially hitching a ride inside circulating immune cells, which may explain how it reaches distant organs and continues to provoke inflammation long after the original exposure.
Damage to Blood Vessels Starts at the Lining
One of the most consequential long-term effects involves the endothelium, the thin layer of cells that lines every blood vessel in the body. When the spike protein’s S1 subunit contacts endothelial cells, it ramps up production of inflammatory signaling molecules like IL-6 and MCP-1 and increases markers of cell injury, including microparticle formation.4Nature. SARS-CoV-2 spike protein induces endothelial inflammation via ACE2 independently of viral replication This inflammatory cascade does not require live virus to be present. The spike protein alone, binding to the ACE2 receptor on these cells, is enough to set it off.
That endothelial inflammation has downstream consequences for the complement system, part of the body’s innate immune defense. In lab experiments, spike protein activation of endothelial cells triggered heavy deposition of complement components (C3 and the membrane attack complex C5b-9) on cell surfaces, which in turn promoted platelet clumping by increasing von Willebrand factor and P-selectin release. Blocking ACE2 or inhibiting complement halted this platelet aggregation.5Frontiers in Immunology. SARS-CoV-2 Spike Protein 1 Activates Microvascular Endothelial Cells and Complement System Leading to Platelet Aggregation Roughly 30% of severe COVID-19 patients in that study had detectable spike protein circulating in their blood, suggesting a direct connection between persistent spike and vascular complications.
Fibrinolysis-Resistant Microclots
Beyond the acute platelet clumping, the spike protein contributes to a more insidious clotting problem. Researchers have identified dense, fibrin-based microclots in the plasma of long COVID patients that resist the body’s normal clot-dissolving machinery.6Europe PMC. A central role for amyloid fibrin microclots in long COVID/PASC: origins and therapeutic implications These are not ordinary blood clots. The fibrin in them takes on an amyloid-like structure, meaning the protein fibers fold into a tough, stacked arrangement that enzymes struggle to break apart.
A more recent biochemistry study showed exactly how this happens at the molecular level. A specific fragment of the spike protein (a sequence spanning amino acids 685 to 701) can itself form amyloid fibrils, and when those fibrils are mixed with fibrinogen during clot formation, the resulting fibrin networks become markedly resistant to plasmin, the enzyme your body normally uses to dissolve clots.7ACS Publications. SARS-CoV-2 Spike Protein Amyloid Fibrils Impair Fibrin Formation and Fibrinolysis The clinical implication is that even small amounts of lingering spike protein could seed microclots that the body cannot clean up on its own, potentially contributing to fatigue, exercise intolerance, and poor oxygen delivery to tissues.
How the Heart Takes a Hit
Heart cells are particularly vulnerable because they are rich in ACE2 receptors. When human cardiomyocytes are exposed to the S1 subunit, the cellular response unfolds in two phases. In the first 24 hours, mitochondria actually rev up, increasing ATP production and fatty acid transport. But by 72 hours, the picture flips: mitochondrial respiration drops, reactive oxygen species (ROS) climb, and calcium floods into both the mitochondria and the cytoplasm. Electron microscopy at that point shows the mitochondria fragmenting and splitting apart.8PubMed Central. Spike Protein Impairs Mitochondrial Function in Human Cardiomyocytes: Mechanisms Underlying Cardiac Injury in COVID-19
Blocking ACE2 with neutralizing antibodies completely prevented these effects on ATP production, confirming that the spike protein enters heart cells through that specific receptor. The mitochondrial damage and calcium overload can explain, at least partly, the myocarditis, arrhythmias, and exercise intolerance that some people experience for months after COVID-19.
Crossing the Blood-Brain Barrier
The spike protein’s S1 subunit can cross the blood-brain barrier without the virus attached to it.9SpringerLink. SARS-CoV-2 Spike Protein S1 Exposure Increases Susceptibility to Angiotensin II-Induced Hypertension in Rats by Promoting Central Neuroinflammation and Oxidative Stress Once inside the brain, it triggers a series of inflammatory events that researchers have linked to many of the neurological complaints in long COVID. Spike-mediated damage to the vessels forming the blood-brain barrier, along with mast cell activation that destroys the tight junctions holding barrier cells together, opens the door for further immune cell infiltration. Microglial cells, the brain’s resident immune responders, become activated and proliferate. In the hippocampus, this neuroinflammation and microgliosis appear to impair memory through complement-dependent destruction of synapses.10Frontiers in Cellular Neuroscience. Unraveling the SARS-CoV-2 spike protein long-term effect on neuro-PASC
The skull-bone study mentioned earlier connected these dots further: in mice injected with spike protein, the protein traveled from skull bone marrow into the meninges and brain parenchyma, where it induced neuroinflammation and neurodegeneration. The fact that spike protein was found in human skull samples from people who died of unrelated causes raises uncomfortable questions about how many recovered COVID patients are carrying low-grade brain inflammation without obvious symptoms.1Cell Press. SARS-CoV-2 spike protein persists in the skull-meninges-brain axis and induces neuroinflammation and neurodegeneration in mice
Microglia exposed to the spike protein in culture also show significant mitochondrial disruption, with a roughly 1.4-fold increase in reactive oxygen species and dramatically increased oxygen consumption rates, suggesting these immune cells are being pushed into a state of metabolic overdrive that could sustain chronic neuroinflammation.11PubMed Central. Mitochondrial Dynamics in SARS-COV2 Spike Protein Treated Human Microglia: Implications for Neuro-COVID
Connections to Amyloid-Prone Proteins
One of the more alarming lines of research involves the spike protein’s ability to interact with proteins already linked to neurodegenerative diseases. Computational and biochemical experiments have shown that the S1 receptor-binding domain binds to amyloid-beta (associated with Alzheimer’s disease), alpha-synuclein (associated with Parkinson’s disease), tau, prion protein, and TDP-43. The binding occurs at heparin-binding sites and could theoretically seed or accelerate the aggregation of these proteins in the brain.12Elsevier. SARS-CoV-2 spike protein interactions with amyloidogenic proteins: Potential clues to neurodegeneration
This does not mean that COVID-19 causes Alzheimer’s or Parkinson’s. What it does suggest is a plausible mechanism by which persistent spike protein in the brain could nudge protein aggregation in a harmful direction, especially in people already predisposed. The binding affinity between S1 and alpha-synuclein was the strongest among the proteins tested, which is worth watching given early epidemiological signals of increased parkinsonism risk after COVID-19. These are still early-stage findings, and the jump from protein-protein binding data to clinical neurodegeneration is a large one.
Autoimmunity Through Molecular Mimicry
The spike protein shares short amino acid sequences with a surprising number of human proteins, and this resemblance can confuse the immune system. Computational analysis has identified mimicry “hotspots” in the spike protein. One motif, TQLPP, is shared with thrombopoietin, a hormone involved in platelet production. Antibodies that cross-react with thrombopoietin could explain the thrombocytopenia (low platelet counts) seen in some COVID-19 patients. Another motif, ELDKY, appears in tropomyosin (a cardiac muscle protein) and PRKG1 (involved in platelet signaling and calcium regulation), providing a plausible route to both cardiac complications and clotting abnormalities.13Europe PMC. Potential Autoimmunity Resulting from Molecular Mimicry between SARS-CoV-2 Spike and Human Proteins
The scale of this mimicry appears to be larger than you would expect by chance. A 2024 review of accumulated evidence found that both computational approaches and actual antibody cross-reactivity experiments have turned up far more mimicry events between SARS-CoV-2 and human proteins than random chance would predict, and that many of the targeted human proteins are already known targets in autoimmune diseases.14Hindawi / PubMed Central. Evidence for Molecular Mimicry between SARS-CoV-2 and Human Antigens: Implications for Autoimmunity in COVID-19
Laboratory work has begun to validate these computational predictions. When researchers raised antibodies against spike peptides containing common short sequences shared with human proteins, most did not strongly cross-react with human tissues. But at least one, targeting the motif EPLDVL, showed high affinity for both the spike protein and a human protein (unc-80 homolog), and reacted broadly with proteins expressed in the small intestine, ovary, and stomach on Western blot.15MDPI. In Silico and In Vitro Evaluation of the Molecular Mimicry of the SARS-CoV-2 Spike Protein by Common Short Constituent Sequences (cSCSs) in the Human Proteome The authors noted this raises a serious possibility of both infection-induced and vaccine-induced autoantibodies in humans, though the clinical significance for most people remains unclear.
Premature Cellular Aging
Cells exposed to the spike protein show signs of accelerated senescence, the state in which a cell stops dividing and begins secreting inflammatory molecules. Human epithelial cells transfected with spike protein exhibited increased expression of senescence markers p16 and p21, elevated senescence-associated beta-galactosidase activity, and a burst of inflammatory signaling known as the senescence-associated secretory phenotype, or SASP.16PubMed Central. SARS-CoV-2 Spike Protein Induces Paracrine Senescence and Leukocyte Adhesion in Endothelial Cells The senescence was not confined to the directly exposed cells; neighboring cells also became senescent through paracrine signaling, meaning the effect spreads outward.
Further research has traced the molecular pathway: the spike protein upregulates a protein called Cdc42, which activates Wnt/beta-catenin signaling and drives the cell into a senescent state.17Frontiers. Cdc42 improve SARS-CoV-2 spike protein-induced cellular senescence through activating of Wnt/β-Catenin signaling pathway Senescent cells are not just inert; they actively inflame surrounding tissue and impair regeneration. A buildup of senescent cells in the lungs, blood vessels, or brain could contribute to the lingering fatigue and reduced organ function that characterize long COVID.
Lung Damage and Fibrotic Remodeling
In lung tissue specifically, the spike protein induces reactive oxygen species, DNA double-strand breaks, activation of TGF-beta signaling pathways, and cellular senescence. In mouse models, these effects produced changes resembling the early stages of pulmonary fibrosis, and the damage was worsened by prior or subsequent exposure to ionizing radiation.18Europe PMC. SARS-CoV-2 Spike Protein Induces Oxidative Stress and Senescence in Mouse and Human Lung The overlap with radiation injury is striking and suggests the two insults may compound each other in, for example, cancer patients who receive chest radiation after a COVID-19 infection.
A related line of investigation has found that the spike protein drives collagen deposition in lung tissue through a pathway involving STAT3 and an enzyme called thymidine phosphorylase. When the gene for thymidine phosphorylase was knocked out in mice engineered to express human ACE2, the collagen buildup and STAT3 activation triggered by spike protein were significantly reduced.19bioRxiv. Thymidine Phosphorylase Drives SARS-CoV-2 Spike Protein-Induced Lung Tumorigenesis This points to a druggable target that could potentially limit fibrotic remodeling in the lungs of long COVID patients.
The Endocrine System Is Not Spared
The ACE2 receptor is not limited to the lungs and blood vessels. It is expressed in the pancreas, thyroid, ovaries, and testes, and each of those tissues also expresses TMPRSS2, the protease the virus needs to complete cell entry.20Oxford Academic. Impact of COVID-19 on the Endocrine System: A Mini-review This dual expression means the endocrine system has all the molecular hardware for spike-mediated damage. Clinically, thyroid inflammation, new-onset diabetes, and menstrual irregularities have all been reported after COVID-19, and the widespread distribution of ACE2 across hormone-producing glands provides a plausible explanation for why the virus’s effects can feel so systemic and unpredictable.
Genetic Variation Shapes Individual Risk
Not everyone who encounters the spike protein experiences the same degree of harm, and part of the explanation is genetic. The ACE2 gene varies among individuals, and some variants change how tightly the spike protein binds. Researchers identified one variant, D355N, that substantially restricts spike protein binding and limits infection both in cell culture and in animal models.21PubMed Central. Susceptibilities of Human ACE2 Genetic Variants in Coronavirus Infection
Broader analysis across ethnic populations found that 11 out of 20 ACE2 variants studied, predominantly found in Asian populations, were classified as deleterious to the spike-ACE2 interaction, meaning they reduced binding affinity and potentially lowered infection susceptibility. Another six variants, found across Asian and African populations, were categorized as potentially deleterious with a similar tendency.22Taylor & Francis Online. Comparative analysis of ACE2 mutations and their impact on SARS-CoV-2 susceptibility within diverse ethnic groups These differences in receptor architecture may partly explain why some people clear the spike protein quickly and feel fine, while others develop months of symptoms.
Measuring Spike Protein as a Diagnostic Tool
Researchers are beginning to treat circulating spike protein levels as a measurable biomarker for long COVID severity. One modeling study found a statistically significant relationship between spike protein concentration in blood and the number of long COVID symptoms a patient reported. The model estimated that symptom counts plateau around 20, with a half-maximal effect at a spike concentration of roughly 3.77 log copies per microliter. Among the inflammatory mediators associated with those symptoms, the chemokine CXCL8 had the lowest threshold for driving effects, followed by IL-6, IL-1beta, and TNF-alpha.23MDPI. Biomarker-Based Risk Assessment Strategy for Long COVID: Leveraging Spike Protein and Proinflammatory Mediators to Inform Broader Postinfection Sequelae If validated in larger studies, a simple blood test for spike protein levels could help clinicians identify who is at highest risk for prolonged illness and track whether treatments are actually reducing the protein burden.
Emerging Approaches to Clear Persistent Spike Protein
The therapeutic landscape is still early and mostly preclinical, but several strategies are being explored. A scoping review of 420 studies identified 40 that specifically assessed compounds or technologies capable of neutralizing, eliminating, or removing spike protein. The approaches range from monoclonal antibodies (the largest category at 25% of included studies) and neutralizing antibodies to nanomaterial-based systems, polyphenols derived from green tea, engineered proteins, and even device-based methods like lectin-affinity plasmapheresis, which filters the protein directly from blood.24Mathews Journal of Nutrition & Dietetics. Elimination/Neutralization of COVID-19 Vaccine-Produced Spike Protein: Scoping Review
Among the natural compounds, nattokinase, an enzyme from fermented soybeans, has attracted attention because it has both fibrinolytic properties (it helps dissolve clots) and has been shown in experiments to degrade spike protein directly.25PubMed Central. Strategies for the Management of Spike Protein-Related Pathology That dual action makes it a candidate for addressing both the microclot problem and the spike persistence problem simultaneously. However, most of this evidence comes from in vitro work and animal models. Rigorous clinical trials in humans are still needed, and anyone considering these approaches should discuss them with a clinician rather than self-treating based on preliminary data.
The bigger challenge is that clearing spike protein may not be enough on its own. If the protein has already triggered autoimmune processes, induced cellular senescence, or seeded amyloid-like microclots, removing the initial insult does not automatically reverse the downstream damage. Effective treatment for long-term spike protein effects will likely require a combination strategy: reducing the protein itself, calming the inflammatory response it provoked, and repairing the specific tissue damage in whichever organs were affected.