Autoimmune disease beyond the thyroid: copper, iron, vitamin A and endothelin-1

13 November 2026

By Libbie Wallace

Back to the thyroid post: Thyroid

Across autoimmune diseases, the four factors are not equally supported. Iron and endothelin-1 have the strongest evidence: misplaced iron and ferroptosis drive tissue damage in lupus, rheumatoid arthritis and MS, and blocking endothelin-1 is an approved treatment in systemic sclerosis. Vitamin A has a clear tolerance mechanism and some trials, with modest results. Copper is essential for immune regulation, but no study shows deficiency causing autoimmune disease in people, and inflammation raises serum copper enough to hide low stores.

The common thread is inflammation moving these markers around: hepcidin traps iron in cells, ceruloplasmin and copper rise, retinol falls and ET-1 climbs. That makes blood tests hard to read and causation hard to prove. The same thread links this tab to the thyroid model on the first tab.

Copper: needed for immune control, but masked by inflammation

Copper deficiency weakens the immune system rather than directly triggering autoimmunity. In most autoimmune diseases studied, serum copper is higher than normal. The catch is that this rise largely reflects inflammation, which can hide low stores.

What copper does for immunity

In animal models, copper deficiency reduces antibody production by B cells, IL-2 secretion by helper T cells, cytotoxic T-cell killing, and neutrophil numbers (Front Oncol review, 2022).
An American Journal of Clinical Nutrition review reported that even marginal deficiency — where standard copper markers were unaffected — reduced T-cell proliferation and interleukin levels, and impaired neutrophils’ ability to kill microbes (Percival, 1998).
A British Journal of Nutrition review noted that accurate diagnosis of marginal deficiency remains elusive, and proposed immune markers such as IL-2 production and neutrophil function as more sensitive indicators (Bonham et al., 2002).

IL-2 matters for autoimmunity because regulatory T cells depend on it to survive. Low IL-2 is a recognised feature of lupus. Whether marginal copper deficiency contributes to that in people has not been tested.

Rheumatoid arthritis: copper runs high. A 1978 Annals of the Rheumatic Diseases study of 189 patients found both serum copper and ceruloplasmin raised, with an inverse relationship between serum iron and copper (Scudder et al.). A 2022 study with meta-analysis confirmed higher copper and lower zinc in RA. Copper tracked disease activity, CRP and ESR, while blood iron fell as activity rose (Biol Trace Elem Res, 2022). That pattern — copper up, iron down, both tied to inflammation — is the acute-phase response at work.

Lupus: no consistent change. One study found no difference in copper or ceruloplasmin between lupus patients and controls, and a weak negative correlation between copper and disease activity (SLEDAI) (Ann Rheum Dis abstract).

Multiple sclerosis: a mimic and a model

Copper deficiency myelopathy causes neurological symptoms with demyelination and white matter lesions in humans (Metallomics, 2024), and can be mistaken for MS or B12 deficiency.
The standard laboratory model of MS-like demyelination uses cuprizone, a copper chelator (Mol Neurodegener, 2022). A 2024 mouse study linked demyelination to loss of soluble brain copper, and found a brain-penetrating copper compound was protective (Metallomics, 2024). A 2022 study disagreed, concluding cuprizone damage comes from a toxic cuprizone-copper complex rather than copper depletion (ASN Neuro, 2022). The question is unresolved, and cuprizone demyelination is not autoimmune.

Bottom line for copper. Copper is necessary for normal immune regulation, and deficiency is under-recognised (see the thyroid tab). But no study shows copper deficiency causing an autoimmune disease in people. In active autoimmune disease, raised serum copper and ceruloplasmin mostly reflect inflammation — a reason to treat a "normal" or high copper result as uninformative about tissue copper, not as evidence of adequate or excess copper (see next section).

Raised serum copper is not adequate copper

A high serum copper in autoimmune disease tells you the liver is pouring ceruloplasmin into the blood as part of the inflammatory response. It does not tell you that the tissues have enough copper. Animal and human studies show that blood copper can be normal or high while the organs, and the copper enzymes inside cells, are running short.

Where the body’s copper actually is. An adult holds about 50–120 mg of copper, almost two-thirds of it in the skeleton and muscle (NIH ODS). The liver is the key site that sets the plasma level (Institute of Medicine). With serum copper of roughly 0.7–1.4 mg/L and about 3 L of plasma, the circulation carries only a few milligrams — a small fraction of the total. A blood test samples the export stream from the liver, not the stores.

Why inflammation pushes the blood level up

Ceruloplasmin is an acute-phase protein. Its output from the liver is driven by IFN-γ, IL-1β and bacterial lipopolysaccharide, and is thought to supply extra copper to immune cells fighting infection (PLOS One review, 2022). In liver cells, IL-6 raises ceruloplasmin while cutting the iron exporter ferroportin (Di Bella et al., 2017).
This "hypercupraemia" is a recognised part of nutritional immunity — the body moving metals around to fight microbes. Mice lacking ceruloplasmin are highly susceptible to Salmonella and Staphylococcus aureus infection (USDA project summary).
Your reading is right: a raised serum copper is a sign of infection, inflammation or dysregulation — or of oestrogen, pregnancy or some cancers (NIH ODS) — not of good copper status.

Blood and tissue can point in opposite directions

An Advances in Nutrition review notes that numerous animal experiments show plasma copper normal or increased even though copper in the liver and other organs is low (PMC3226389).
In rats, inflammation lowered liver Cu/Zn-SOD activity at every copper intake, and ceruloplasmin rose with inflammation only in rats that had enough copper (DiSilvestro & Marten, J Nutr 1990). The authors concluded that inflammation raises the dietary copper needed to keep liver SOD working normally (same study).
In rheumatoid arthritis, a 1982 Annals of the Rheumatic Diseases study found serum ceruloplasmin activity up but red-cell SOD activity down, suggesting copper-dependent activity shifts from the cells into the plasma (Banford et al.). Red-cell SOD is also lower in lupus (PMC4024413).

The human depletion study that makes the point. Twelve post-menopausal women ate 0.57 mg copper a day for 105 days. Plasma copper and ceruloplasmin did not change. Over the same period, red-cell SOD fell from 3450 to 2600 U/g haemoglobin and platelet cytochrome c oxidase from 1740 to 810 U/g protein. The authors called for a paradigm shift in assessing copper status (Milne & Nielsen, Am J Clin Nutr 1996). The cellular enzymes ran down while the blood test stayed normal — even without inflammation.

A "normal" ceruloplasmin can hide inactive protein. Standard immunoassays count ceruloplasmin protein whether or not it carries copper. Copper-deficient rats had no ceruloplasmin enzyme activity yet still had about a quarter of their ceruloplasmin protein, as copper-free apoceruloplasmin; even in healthy plasma, apoceruloplasmin may be about half the total (Linder, review). In one clinic series, 8% of patients with neurological symptoms had normal ceruloplasmin concentration but low oxidase activity (peeref summary).

Better markers of tissue copper

Marker	What it shows	Availability
Ceruloplasmin oxidase activity ÷ ceruloplasmin protein	Share of ceruloplasmin actually loaded with copper; proposed as more sensitive than either alone (Gibson, Principles of Nutritional Assessment)	Oxidase assay in some specialist labs
Platelet cytochrome c oxidase	Mitochondrial copper enzyme; halved in the depletion study	Research only
Red-cell Cu/Zn-SOD activity	Cellular antioxidant enzyme; reflects intake	Research and some specialist labs
Red-cell CCS (copper chaperone for SOD)	Rose early in copper-deficient rodents; candidate marker of marginal deficiency (Nordic Nutrition Recommendations)	Research only

None of these is yet validated for marginal deficiency in routine practice (NNR 2023). In UK practice, the realistic step is to always read serum copper and ceruloplasmin alongside CRP. The Scottish trace element laboratory notes that copper deficiency can be masked by raised ceruloplasmin in infection, injury or chronic inflammatory disease (SAS Trace Elements).

What the evidence does not yet show

That autoimmune patients are copper-deficient in their tissues. The red-cell SOD data in RA are mixed: one study found SOD activity increased (Tuncer et al., 1999), and SOD can also fall from oxidative damage rather than copper lack.
That inflammation drains copper stores. In rats with acute inflammation, liver copper was not depleted; the extra copper for ceruloplasmin came without emptying stores (Conforti et al.). A review concluded RA appears to develop without being linked to copper or zinc deficiency (Milanino et al.).
That more copper helps. Copper-deficient rats developed adjuvant arthritis just as often as controls, but recovered more slowly and were immunosuppressed (Kishore et al., 1984). Yet removing copper with tetrathiomolybdate suppressed arthritis in rats (Arthritis Res Ther, 2005). Copper is likely needed for the immune system to resolve inflammation, but also fuels some of its damaging steps.

The fair conclusion: serum copper in active autoimmune disease is uninterpretable on its own, and tissue copper status in these patients is genuinely unknown. That is a research gap, not proof of either deficiency or sufficiency.

Iron: low in the blood, high in the inflamed tissue

Iron has the strongest mechanistic evidence of the four across autoimmune disease. The recurring pattern is the same one seen in Hashimoto’s: iron is scarce in the circulation but builds up in inflamed tissue and immune cells, where it drives oxidative damage and pushes T cells towards inflammation.

Lupus (SLE): iron in T cells and neutrophils

Neutrophil ferroptosis. Neutrophils from lupus patients and lupus-prone mice undergo ferroptosis. Autoantibodies and interferon-α in the serum suppress GPX4, raising lipid peroxides. Mice with half the normal neutrophil GPX4 developed autoantibodies, neutropenia, skin lesions and proteinuria, and a ferroptosis inhibitor reduced disease severity (Li et al., Nature Immunology 2021). The authors conclude that neutrophil ferroptosis is an important driver of neutropenia in SLE (abstract).
Iron loading in T cells. Activated T cells raise the transferrin receptor (CD71) to take up iron, and this is exaggerated in lupus-prone T cells. Blocking CD71 inhibited Th1 and Th17 cells while enhancing regulatory T cells (Voss et al., Science Immunology 2023). Excess T-cell iron impaired mitochondria and promoted Th17 differentiation; anti-CD71 treatment reversed lupus features in mice, and disease severity in patients correlated with CD71 on Th17 cells (Science Immunology summary).
Kidney iron. In lupus nephritis, iron in the proximal tubules drives lipid-peroxide damage, and giving hepcidin to lupus-prone mice reduced kidney iron and inflammatory cytokines (same source, PMC9936798).

Rheumatoid arthritis: systemic deficiency, synovial overload. A 2025 systematic review describes a paradoxical distribution in RA — systemic iron deficiency but iron overload in the synovium, which promotes ferroptosis. It also reports impaired GPX4 defences, and that some disease-modifying drugs (leflunomide, sulfasalazine) may act partly through ferroptosis (Clin Immunol, 2025). Synovium and synovial fluid from RA patients show more lipid peroxidation and iron, with much higher synovial-fluid iron in high-activity disease (Wu et al., Nature Communications 2022). In the blood, iron falls as disease activity rises (Biol Trace Elem Res, 2022).

There is a twist. In arthritic mice, inducing ferroptosis cut the number of aggressive synovial fibroblasts, and combining a low-dose ferroptosis inducer with the TNF blocker etanercept markedly slowed arthritis (Wu et al., 2022). Ferroptosis can therefore harm or help depending on which cells it kills — a reason for caution about simple "iron is bad" messages.

Multiple sclerosis: iron rims mark smouldering disease. At post-mortem, 15-30% of MS lesions show chronic inflammatory activity, with rims of iron-laden microglia and macrophages (Mult Scler J, 2022). Iron-sensitive MRI can see these as paramagnetic rim lesions. In 159 people with relapsing-remitting MS, 36% had at least one rim lesion; having one was linked to worse physical and cognitive scores as total lesion burden rose (Marcille et al., Sci Rep 2022). Over a mean 10-year follow-up of 172 people, rim lesions marked more frequent relapses and greater long-term disability progression (Mult Scler J, 2024).

Bottom line for iron. Across lupus, RA and MS, the evidence points to iron misplaced into inflamed tissue and immune cells — not to high total body iron. That fits the ceruloplasmin and hepcidin mechanisms on the thyroid tab: inflammation raises hepcidin and locks iron inside cells, while circulating iron looks low.

Vitamin A: tolerance signal with modest clinical results

Vitamin A has the clearest immune mechanism of the four — retinoic acid tips T cells towards tolerance — and it has actually been trialled in autoimmune disease. Results so far are modest: better symptom scores in MS, encouraging small studies in lupus nephritis, and no proof that supplements change disease course.

The mechanism. Retinoic acid promotes regulatory T cells and helps keep mucosal immunity in check. In vitamin A deficiency, inflammatory T-cell reactions may go inadequately opposed and become dominant (Ross, review). In a mouse model of type 1 diabetes, retinoic acid reduced T-cell expression of the Th1 drivers T-bet and STAT4 (same source).

Multiple sclerosis: the most-studied disease

Blood levels predict MRI activity. In 88 Norwegian people with relapsing-remitting MS followed for two years, each 1 µmol/L rise in serum retinol cut the odds of new gadolinium-enhancing lesions by 49% and new T2 lesions by 42% (Løken-Amsrud et al., Mult Scler J 2013). Retinol also predicted new lesions six months ahead (same study). In a small Japanese study, lower retinol was linked to faster brain volume loss (Yokote et al., 2017).
Supplementation trial. In a one-year placebo-controlled trial in 101 Iranian patients, 25,000 IU/day retinyl palmitate for six months then 10,000 IU/day improved the MS Functional Composite score, but did not change EDSS disability, relapse rate or active brain lesions (Bitarafan et al., Arch Iran Med 2015). The same trial reported less fatigue and depression (Iran J Allergy Asthma Immunol, 2016). A companion analysis recorded a rise in CRP and advised close monitoring during high-dose supplementation (Jafarirad et al., 2013).

Lupus nephritis: retinoids as drugs. In lupus-prone mice, all-trans retinoic acid (ATRA) reduced kidney damage and prolonged survival (Kinoshita et al., J Immunol 2003), and worked better combined with prednisolone (Nozaki et al., Clin Exp Immunol 2005). Two patients whose nephrotic syndrome did not respond to steroids improved on ATRA 10 mg/day (Am J Kidney Dis, 2010). A small phase II study saw urine protein fall from 3.2 to 0.4 g/g creatinine at six months on ATRA. However, 6 of 14 on a synthetic retinoid dropped out with side effects, and the group comparison did not reach significance (Ann Rheum Dis 2016 abstract). These are prescription retinoids, not dietary vitamin A.

Inflammatory bowel disease and type 1 diabetes

Vitamin A deficiency is rare in well-nourished countries but does affect people with inflammatory bowel disease — more in Crohn’s than ulcerative colitis — and after bariatric surgery (StatPearls, NCBI).
Impaired vitamin A metabolism affecting autoimmunity and type 1 diabetes onset is a long-held hypothesis, but not established (Trasino & Gudas, review).

The caveat. The same retinoic acid that promotes tolerance at rest can activate pathogenic T cells during active inflammation (see the thyroid tab). Serum retinol is homeostatically controlled and does not drop until stores are significantly limited; it is also affected by infection, zinc and iron status (Medscape). Low retinol in active disease may partly be a marker of inflammation rather than true deficiency.

Endothelin-1: the one factor already treated in the clinic

Endothelin-1 is the only one of the four with an approved treatment in autoimmune disease. Endothelin receptor blockers prevent new finger ulcers in systemic sclerosis and treat pulmonary hypertension linked to connective tissue disease. Elsewhere, ET-1 is raised in active disease and tracks inflammation.

Systemic sclerosis (scleroderma): ET-1 at the centre. ET-1 and its receptors are overexpressed in scleroderma, and its vessel-narrowing and growth-promoting effects drive the disease’s vasculopathy (Rheumatology News, 2006). Ischaemic digital ulcers affect 35–60% of patients (RAPIDS-2, Ann Rheum Dis).

RAPIDS-1: bosentan, a dual endothelin receptor blocker, cut new digital ulcers by 48% over 16 weeks in 122 patients (same source).
RAPIDS-2: in 188 patients, bosentan reduced new ulcers by 30% over 24 weeks (1.9 vs 2.7) (Matucci-Cerinic et al., 2011). It did not speed healing of existing ulcers, and was linked to swelling and raised liver enzymes (PubMed abstract).

Lupus: ET-1 tracks disease activity. A 2023 study of 314 lupus patients found serum ET-1 higher than in healthy controls and in other rheumatic diseases. It correlated with disease activity (SLEDAI, rs = 0.46) and ESR, distinguished lupus from healthy people well (AUC 0.91), and was higher in patients with proteinuria or anti-Sm antibodies (Li et al., Sci Rep 2023). Raised plasma ET-1 in lupus was first reported in 1991 (Ann Rheum Dis).

Rheumatoid arthritis. Serum ET-1 is higher in RA than in controls and rises with CRP (cited in Li et al., 2023). ET-1 is also raised in psoriasis (same source).

Multiple sclerosis: blood flow to the brain. Plasma ET-1 is raised in MS. In a 2013 PNAS study, ET-1 appeared to be released into the cerebral circulation from reactive astrocytes in plaques, and a single 62.5 mg dose of bosentan restored reduced cerebral blood flow (D’haeseleer et al., PNAS 2013). Resting astrocytes show no ET-1, but those in demyelinated plaques show high levels, driven by inflammatory cytokines (Hostenbach et al.). However, a 4-week randomised trial of bosentan in 27 relapsing-remitting patients was negative. The authors suggested future trials select patients with high ET-1 in blood or spinal fluid (ROCHIMS trial, Front Neurol 2020).

Bottom line for ET-1. Across scleroderma, lupus, RA, MS and thyroid autoimmunity, ET-1 rises with immune-driven inflammation and narrows small vessels in the target organ. In scleroderma, blocking it has proven clinical benefit; elsewhere it remains a marker and a research target.

Where the evidence sits, disease by disease

Disease	Copper	Iron	Vitamin A	Endothelin-1
Hashimoto’s / Graves’	Mixed; shifts with thyroid status	Deficiency common; ferroptosis in thyroid tissue (early)	Low in Hashimoto’s; lowers TSH in trials	Raised in both
Lupus (SLE)	No consistent change	Neutrophil ferroptosis; T-cell iron via CD71; kidney iron (strong lab evidence)	Retinoic acid helped lupus nephritis in mice and small human series	Raised; tracks activity and proteinuria
Rheumatoid arthritis	Serum raised by inflammation; tissue status unknown	Low in blood, high in synovium; ferroptosis can help or harm	Mechanistic only	Raised; tracks CRP
Multiple sclerosis	Deficiency mimics MS; cuprizone model debated	Iron-rim lesions predict disability	Low retinol predicts new lesions; supplement trial improved function score only	Lowers brain blood flow; small bosentan trial negative
Systemic sclerosis	—	—	—	Central; bosentan cuts new digital ulcers (RCTs)
Type 1 diabetes, IBD	—	—	Hypothesis in type 1 diabetes; deficiency occurs in Crohn's	—

A dash means no relevant evidence was found in this search, not that none exists.