A new study reports a strategy for delivering gene-editing technology directly to human haematopoietic stem cells inside the body, using targeted lipid nanoparticles rather than viral vectors. Published in Nature Biomedical Engineering, the work by Zhiwei Luo, A.T. Zhu and Michael J. Mitchell describes an approach designed to overcome one of the most difficult problems in regenerative medicine: reaching rare, fragile stem cells in their natural biological environment while preserving their ability to generate blood and immune cells.
Haematopoietic stem cells, or HSCs, reside primarily in the bone marrow and continuously replenish the body’s blood and immune systems. Their ability to self-renew and produce multiple blood-cell lineages makes them attractive targets for treating inherited blood disorders, immune deficiencies and some cancers. Yet those same properties make them challenging to manipulate. Conventional gene-editing therapies often require stem cells to be removed from a patient, edited in a laboratory and then returned after conditioning treatment has cleared space in the bone marrow. An effective method for editing HSCs directly in vivo could simplify this process and potentially broaden access to cell and gene therapies.
The delivery system at the centre of the study is the lipid nanoparticle, a microscopic assembly of fats that can encapsulate and transport nucleic acids. Lipid nanoparticles became widely known through messenger RNA vaccines, but their potential extends far beyond vaccination. They can protect fragile genetic payloads from degradation, enter cells after administration and release their contents into the cytoplasm. For gene editing, such particles can be used to carry components that temporarily instruct a cell to cut, replace or regulate a selected DNA sequence. Unlike integrating viral vectors, many lipid-nanoparticle systems deliver their payload without permanently inserting a carrier genome into the recipient cell.
The major obstacle is specificity. When administered in the body, nanoparticles encounter proteins, membranes and immune cells throughout the circulation. Particles that accumulate in the liver, for example, may be highly effective for hepatic therapies but poorly suited to reaching bone-marrow stem cells. The researchers therefore developed targeted lipid nanoparticles intended to recognise and enter human HSCs more efficiently. Targeting can be achieved by displaying molecular ligands on the particle surface that bind receptors enriched on the desired cell type. Once attached, the particle may be internalised through receptor-mediated uptake, creating a route for the editing cargo to reach the cell’s interior.
This distinction between delivery and editing is crucial. A gene-editing system can be extraordinarily precise at the molecular level and still fail as a therapy if too little of it reaches the correct cells. HSCs represent only a small fraction of the cells in bone marrow, and they are surrounded by stromal cells, mature blood cells and other progenitors. A targeted particle must navigate this complex tissue, avoid premature clearance and release enough editing material inside the stem cell to generate a useful level of modification. At the same time, excessive exposure could damage cells or increase unintended editing, making the balance between potency and safety central to the design.
The study’s significance lies in its focus on human HSC biology rather than on delivery to a more accessible tissue. Editing these cells could have effects that persist for years because a successfully modified stem cell can divide and produce descendants across the blood system. That creates the possibility of correcting mutations at their source rather than repeatedly treating the symptoms produced by defective blood cells. It also raises the bar for safety: an alteration introduced into a long-lived stem-cell population could be inherited by many daughter cells, so researchers must evaluate both the intended genetic change and the possibility of unwanted genomic alterations.
Targeted lipid nanoparticles could offer several practical advantages over viral delivery platforms. Viruses have been engineered into powerful gene-transfer vehicles, but their manufacture, immune recognition, cargo limits and potential for persistent genetic activity can complicate treatment. Lipid nanoparticles are generally assembled from synthetic or semisynthetic components and can be designed to release transient editing instructions. Their chemistry can also be adjusted, allowing researchers to modify particle size, surface charge, stability and tissue distribution. These features make them a flexible platform, although they do not eliminate the challenges of immune responses, manufacturing consistency or delivery outside the liver.
The work also points toward a broader shift in gene therapy: from editing cells outside the body toward programming therapeutic changes in their native niches. In an ex vivo procedure, scientists can select cells, measure editing efficiency and remove poorly performing or damaged cells before infusion. In vivo treatment offers no such easy screening step. The nanoparticles must therefore perform their targeting, uptake and payload release within the patient, and the resulting cell population must be assessed through molecular and functional tests. Demonstrating meaningful editing in human HSCs is consequently an important milestone, but it is only one stage on the path toward clinical application.
Before such a method can be used routinely, researchers will need to establish how consistently the particles reach stem cells across individuals, how long the edited cells persist and whether blood production remains normal. Studies must also examine off-target editing, inflammatory reactions, dose limits and the behaviour of edited cells over extended periods. Questions about delivery to different bone-marrow compartments, the effects of preconditioning and the ability to adapt the system to different disease-associated mutations will be equally important. The therapeutic promise of the platform will ultimately depend not only on editing efficiency, but on whether it can deliver durable benefit with a risk profile acceptable for patients who may otherwise require lifelong treatment.
The report by Luo, Zhu and Mitchell marks a notable advance in the effort to make in vivo HSC gene editing technically achievable. By combining cell-selective targeting with the adaptable chemistry of lipid nanoparticles, the researchers address the delivery problem that has constrained many gene-editing concepts. The approach does not yet remove the biological and regulatory hurdles facing in vivo stem-cell therapy, but it provides a framework for pursuing treatments that act within the bone marrow rather than relying entirely on laboratory manipulation. If future studies confirm precise, durable and safe editing, targeted nanoparticles could help transform inherited blood disorders from conditions managed over a lifetime into diseases corrected at the level of the stem cells that sustain the blood system.
Subject of Research: Targeted lipid nanoparticle delivery for in vivo gene editing of human haematopoietic stem cells
Article Title: Targeted lipid nanoparticles unlock in vivo human haematopoietic stem cell gene editing
Article References: Luo, Z., Zhu, A.T. & Mitchell, M.J. Targeted lipid nanoparticles unlock in vivo human haematopoietic stem cell gene editing. Nature Biomedical Engineering (2026). https://doi.org/10.1038/s41551-026-01770-z
Image Credits: AI Generated
DOI: 10.1038/s41551-026-01770-z
Keywords: lipid nanoparticles, haematopoietic stem cells, in vivo gene editing, gene therapy, targeted delivery, bone marrow, regenerative medicine, nanomedicine
