Article Highlight | 9-Nov-2026

Cyclic and catenane IL-2 enable long-acting treatment of autoimmune diseases

Protein topology engineering preserves IL-2's Treg-favoring activity while stretching its serum half-life from 1.25 hours to up to 6.23 hours, cutting dosing frequency in lupus, rheumatoid arthritis and Sjögren's syndrome mouse models

Innovation Press

Low-dose interleukin-2 (IL-2) therapy works by expanding regulatory T cells, the immune system's brakes, yet the cytokine's short serum half-life — about 75 minutes in mice — forces frequent injections and caps how much can safely be given. A study published in The Innovation Life reports a fix that adds almost nothing to the protein: tie its own ends together. A team led by chemical biologist Wen-Bin Zhang at Peking University, together with rheumatologists Jing He and Zhanguo Li at Peking University People's Hospital, engineered two topological variants of IL-2. In one (c-IL-2), the protein's termini are ligated into a seamless ring. In the other (cat-IL-2), two ring-shaped copies are mechanically interlocked into a catenane. Both are genetically encoded and produced directly inside E. coli cells by split-intein protein splicing with a dimeric entangling motif called p53dim, so no refolding is needed after purification. Circular dichroism spectra and binding assays showed that neither topology disturbed IL-2's native fold or receptor binding. The payoff came in pharmacokinetics: wild-type IL-2 cleared from mouse blood with a half-life of 1.25 hours, while c-IL-2 and cat-IL-2 lasted 6.23 and 5.67 hours — about 5.0 and 4.5 times longer. Slower clearance meant less frequent dosing. In an MRL/lpr lupus mouse model, once-weekly low-dose c-IL-2 or cat-IL-2 expanded regulatory T cells and reduced proteinuria, skin lesions and kidney injury, matching weekly or thrice-weekly wild-type IL-2. The weekly regimen also eased joint swelling and bone erosion in a collagen-induced arthritis model and, in a Sjögren's syndrome model, c-IL-2 reduced salivary gland inflammation and improved secretion. No obvious systemic toxicity was observed at the doses and time frames tested. The variants kept their selectivity in human material: in peripheral blood mononuclear cells from systemic lupus erythematosus patients, both preferentially expanded regulatory T cells and activated downstream STAT5 signaling without raising NK-cell proportions. RNA sequencing of patient-derived CD4+ T cells showed lowered expression of genes involved in immune activation, antigen presentation and inflammation, suggesting a shift toward immune tolerance. The authors caution that dosing, long-term safety and immunogenicity still require systematic evaluation before clinical translation; the therapeutic evidence so far rests on mouse models and ex vivo human cells. The team has filed a Chinese patent on the variants as autoimmune therapeutics. Because the strategy rewires backbone connectivity rather than attaching bulky modifications, the authors argue it should extend to other cytokines, offering a minimalist route to long-acting protein biologics. 

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.