Lupus Patients’ Immune Cells Target Proteins Made by Splicing Errors

Lupus neoantigens, protein fragments that healthy cells never make, are produced by faulty mRNA splicing in patients’ neutrophils and are drawing responses from their T and B cells, according to a study published online by Nature on 28 October. The finding offers a different explanation for why the immune system turns on the body in lupus.

The standard view is that lupus begins when the immune system loses tolerance for the body’s own normal proteins. The new work, co-led by Tomas Mustelin of the University of Washington School of Medicine and Stephen Schoenberger of the La Jolla Institute for Immunology, suggests the immune system may instead be reacting to proteins that really are new. “The importance of our finding is that they suggest the possibility of a completely new paradigm of autoimmune disease,” Mustelin said in UW Medicine’s announcement on 29 October.

Where the lupus neoantigens come from

When a cell copies a gene into messenger RNA, it cuts out sections that are not meant to be read and joins the rest together. This process, mRNA splicing, normally follows a fixed pattern. When it goes wrong, the joined-up message can code for a stretch of amino acids that exists nowhere else in the body.

The team sequenced RNA from neutrophils, the white blood cells that act as a first line of defence against infection. Samples came from 34 patients with systemic lupus erythematosus, the most common form of lupus, along with 20 healthy donors and 15 people with rheumatoid arthritis, which served as a comparison autoimmune disease. The researchers found 271 abnormal, or “cryptic”, splicing events in 130 genes in the lupus neutrophils, all absent from both control groups. The most common were skipped exons, and many of the affected genes are switched on by type I interferons, signalling molecules already linked to lupus.

According to UW Medicine, the faulty messages also escape the cell’s system for destroying nonsense RNA, so they can be translated into protein. Using mass spectrometry, the team found 21 of the new peptides on the surface of patients’ cells, displayed by class I MHC molecules, the cell’s way of showing fragments to passing T cells.

T and B cells treated the fragments as foreign

The immune system appears to have noticed the lupus neoantigens. When T cells from patients were grown alongside peptides covering the new sequences, both CD4 and CD8 T cells multiplied and quickly released interferon-gamma, a sign of activation. “We found that the T cells are easily reactivated by the presence of the neoantigens in laboratory cultures,” Mustelin said.

The team also sequenced one T cell receptor and put it into a reporter T cell, which then responded to the neoantigen too. On the antibody side, 17 of the 34 patients had IgG antibodies against one peptide from a mis-spliced gene called IFI44L. The equivalent stretch of the normal protein did not draw the same response.

In the La Jolla Institute’s release, Schoenberger described the conditions this creates. “When a cell type whose function is pro-inflammatory, such as neutrophils, also expresses potentially immunogenic mutations, the conditions for immune attack are met,” he said.

A cancer immunology idea turned on autoimmune disease

Neoantigens are familiar from cancer, which is Schoenberger’s field. Tumours make altered proteins that T cells can learn to attack, and that is the basis of personalised treatments such as the mRNA therapy from Merck and Moderna that cut melanoma recurrence or death risk by 42%, and of off-the-shelf approaches such as the KRAS pancreatic cancer vaccine that primed T cells in nearly every patient even though it missed its main goal. In cancer the aim is to strengthen that response. In lupus, the same mechanism may be doing harm.

That reversal is the most interesting part of the paper. If the immune system is responding properly to proteins that really are foreign, lupus looks less like a broken immune system and more like a normal one aimed at the wrong target. Schoenberger put it bluntly: autoimmunity “can be caused by the immune system doing exactly what it’s supposed to do”. The authors argue this would also explain why lupus flares and settles: production of the fragments could rise and fall over time.

The model also changes where researchers might look for treatments. Active lupus currently calls for strong immunosuppression, as the paper notes. Work on more targeted immune drugs, such as the T cell drug invika tested in an NIH-led cancer trial, shows how much interest there is in steering immune cells more precisely. If lupus neoantigens from faulty mRNA splicing are the source, the authors’ suggested test is clear: block their production and see whether patients improve. The paper points to lower expression of several splicing factors in lupus neutrophils as possible culprits, but does not identify a single cause.

The evidence has limits that matter. The study is small, at 34 patients, and each was sampled once. The faulty splicing did not track disease activity scores or steroid use, which is not what a simple “more fragments, worse disease” story would predict. The authors estimate their sequencing caught only about 60% of splicing events, so lupus neoantigens may be more common than the study shows. And Schoenberger himself frames the central question as open: “whether neoantigens are a cause or consequence of this process”. Inflammation might itself disturb splicing in neutrophils. Like another Nature paper this week on DNA errors in developing neurons, it describes something cells get wrong, not yet a treatment.

Clinical trials will decide whether it matters

UW Medicine’s release is explicit that the authors caution that whether the new sequences “are truly important in the pathology of their disease awaits additional study”, and that this “can be assessed only in clinical trials with agents that block their production”. Neither release mentions such a trial. Mustelin discloses consulting fees from several drug companies in the paper.

The research was funded by the US National Institutes of Health, with support from foundations and other funders. The next signs to watch for are whether other groups find the same lupus neoantigens in larger and longer-term patient groups, and whether anyone moves a splicing-targeted drug towards a trial.

Sources

• Nature: Najjar et al., T and B cells recognize splicing-derived neutrophil neoantigens in lupus (open access, published online 28 Oct 2026), DOI 10.1038/s41586-026-11219-7
• UW Medicine Newsroom: Lupus findings open new paradigm for autoimmune disease (29 Oct 2026)
• La Jolla Institute for Immunology: Mutated immune cells raise red flags in lupus (28 Oct 2026)

Related stories

• Personalised mRNA Therapy Cuts Melanoma Recurrence or Death Risk 42%
• KRAS Pancreatic Cancer Vaccine Misses Main Goal but Primes T Cells