A new lipid nanoparticle system has enabled researchers to genetically modify human haematopoietic stem and progenitor cells (HSPCs) inside humanized mice, addressing one of the most difficult challenges in gene therapy: reaching rare, quiescent stem cells without compromising their ability to produce blood cells over the long term.
The study, published in Nature Biomedical Engineering, describes an engineered lipid nanoparticle known as CD34/LNP^DP. The platform is designed to deliver messenger RNA and CRISPR/Cas gene-editing components specifically to cells carrying CD34, a surface marker commonly associated with human HSPCs. By coupling the nanoparticle to an anti-CD34 antibody, the researchers aimed to improve cellular targeting while avoiding the need to remove stem cells from the body for laboratory manipulation.
HSPCs reside primarily in the bone marrow and are responsible for replenishing the entire blood and immune system throughout life. Their biology makes them attractive targets for treating inherited blood disorders, but also makes them unusually difficult to engineer. Many long-term haematopoietic stem cells remain in a dormant, or quiescent, state. This limits the efficiency of approaches that depend on active cell division or high rates of cellular uptake. In addition, excessive manipulation can damage stem-cell function, reducing the ability of edited cells to repopulate the blood system.
To identify a suitable delivery vehicle, the investigators screened 15 different lipid nanoparticles for their capacity to transport reporter messenger RNA into human HSPCs. Lipid nanoparticles are microscopic assemblies of fats and related molecules that can protect nucleic acids in the bloodstream and facilitate their entry into cells. The selected formulation, LNP^DP, showed effective delivery in human HSPCs under ex vivo conditions, where cells are treated outside the body, and also demonstrated activity in vivo when directed toward CD34-positive cells.
The antibody-conjugated system was then tested with CRISPR/Cas editing cargos. In this approach, messenger RNA can encode the Cas nuclease, while a guide RNA directs the nuclease to a chosen DNA sequence. Once inside the target cell, the Cas protein creates a precise break in the genome. The cell’s own repair machinery then introduces a mutation or uses a supplied template to modify the target sequence. The researchers reported high editing efficiency in human HSPCs treated ex vivo with CD34/LNP^DP, suggesting that the formulation could support gene editing without relying on viral vectors.
The next experiments examined whether the nanoparticles could work inside living organisms. Humanized mice, which carry human blood-forming cells, received the CD34/LNP^DP system through intrafemoral administration, an approach that delivers the material directly into the marrow cavity of the thigh bone. This route places the nanoparticles close to the HSPC population and may reduce the barriers associated with systemic delivery. The treatment produced efficient editing in human HSPCs and did not cause detectable disruption of overall haematopoiesis, according to the study.
One of the first genetic targets was the erythroid-specific enhancer of BCL11A. This regulatory DNA element controls the activity of BCL11A in red blood cell precursors. BCL11A normally suppresses the production of fetal haemoglobin after birth. Disrupting its erythroid enhancer can therefore reactivate fetal haemoglobin, or HbF, a form of haemoglobin that is normally abundant before birth and can compensate for defective adult haemoglobin in disorders such as sickle cell disease and some forms of beta-thalassaemia.
In the humanized mice, editing the BCL11A enhancer within HSPCs was followed by sustained HbF reactivation in erythroid cells during long-term observation. The result is important because a durable therapeutic effect requires editing of stem cells that continue to generate blood-cell descendants over time, rather than only modifying short-lived progenitor cells. The findings indicate that CD34/LNP^DP can reach a biologically meaningful HSPC compartment and preserve the transmission of the edit through blood production.
The researchers also explored a second disease model involving a mutation in ELANE, a gene associated with severe congenital neutropaenia. Variants in ELANE can impair the development and maturation of neutrophils, leaving patients vulnerable to recurrent and potentially life-threatening infections. In humanized mice carrying an ELANE mutation, intrafemoral administration of CD34/LNP^DP directed CRISPR editing toward exon 2 of the gene. The treatment achieved robust editing in human HSPCs and partially improved the defect in neutrophil development during long-term follow-up.
Although the results do not yet establish a clinical treatment, they demonstrate a potential alternative to conventional ex vivo HSPC gene therapy. Current strategies often require patients’ stem cells to be collected, purified, edited in a specialized laboratory and reinfused after conditioning treatment. An in vivo approach could eventually simplify this process, but it must overcome major challenges, including precise tissue distribution, immune reactions, dose control, off-target editing and the need to reach enough long-term stem cells to produce a durable benefit. The study’s use of an anti-CD34 antibody provides a targeting mechanism, while the nanoparticle protects and transports the gene-editing cargo.
The work also highlights the importance of delivery technology in the development of genetic medicines. CRISPR systems are capable of rewriting disease-associated DNA, but their therapeutic value depends heavily on whether they can reach the correct cells at an effective dose while minimizing unintended exposure. By combining a selected lipid formulation with antibody-mediated recognition of CD34, the researchers created a platform that targets a clinically relevant human cell population in its native bone-marrow environment. Further studies will be needed to evaluate safety, editing precision, manufacturing requirements and performance in larger and more physiologically representative models. Even so, the findings position CD34-targeted lipid nanoparticles as a promising route toward durable in vivo engineering of the human blood-forming system.
Subject of Research: In vivo genetic engineering of human haematopoietic stem and progenitor cells using CD34-targeted lipid nanoparticles.
Article Title: Engineered lipid nanoparticles for in vivo and durable editing of haematopoietic stem cells within humanized mice
Article References: Du, J., Luo, Z., Xie, D. et al. “Engineered lipid nanoparticles for in vivo and durable editing of haematopoietic stem cells within humanized mice.” Nature Biomedical Engineering (2026). https://doi.org/10.1038/s41551-026-01765-w
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s41551-026-01765-w
Keywords: lipid nanoparticles, haematopoietic stem cells, HSPCs, CRISPR/Cas gene editing, CD34 targeting, in vivo gene therapy, BCL11A, fetal haemoglobin, ELANE, neutropaenia, humanized mice
