Blood disorders such as sickle cell anemia, immunodeficiencies, and bone marrow failure syndromes reduce life expectancy and quality of life for millions of people around the world. Bone marrow transplants can be a patient’s only treatment option, but these involve a number of painful and costly procedures. To make room for the transplant, clinicians must kill the diseased cells in a process called immunoablation, which subjects the patient to highly toxic drugs that have many side effects. Immunoablation can leave a patient sterile, immunodeficient, and predisposed to tumors and genetic mutations. 

In a recent study, researchers at the University of Pennsylvania developed a new system for re-engineering hematopoietic stem cells (HSC) in vivo, which may one day lead to approaches that don’t require immunoablation or even eliminate the need for bone marrow transplants altogether (1). Led by Hamideh Parhiz, a pharmaceutical engineer at the Perelman School of Medicine, and Stefano Rivella, a pediatric hematologist at the Children’s Hospital of Philadelphia, the team designed a range of mRNA that coded for gene-editing proteins and proteins that facilitated cell death. They delivered the mRNA directly to the bone marrow by making targeted modifications to lipid nanoparticles (LNP), which were similar to the ones used to package COVID-19 mRNA vaccines. “The goal and the hope [is] that this technology will not be as toxic [as immunoablation],” said Rivella.

The researchers first created HSC-directed LNP by targeting CD117, a receptor found on the surfaces of all HSC. When CD117 binds its target ligand, the HSC absorbs them both (2). The researchers suspected that HSC would bind and absorb the LNP and the mRNA packaged inside if the LNP were decorated with antibodies against CD117. They first tested this in vitro with an mRNA that codes for the light-producing enzyme luciferase. Stem cells treated with CD117-LNP lit up hundreds of times stronger than those treated with LNP decorated with antibodies against CD45 receptors, which are found in fully-differentiated bone marrow cells. They then injected their CD117-LNP into mouse veins, and within 24 hours, saw the mouse femurs light up. By comparison, the femurs did not light up when they replaced the CD117 antibodies with antibodies that had no specific target.

“Historically, people always had this idea that you could just attach some sort of ligand or antibody to the outside of nanoparticles and they would just go to the cells or tissues where you wanted them to go, but that was just not true.” said Anna Blakney, an RNA delivery expert at the University of British Columbia who was not involved in the study. “Showing that you can conjugate an antibody and that it targets the specific cells, but also that you can get it to the bone marrow, I think that's quite groundbreaking.”

Papers like this don't just happen. It was a great collaboration and fantastic team behind this. 
- Hamideh Parhiz, Perelman School of Medicine 

Parhiz and Rubella’s team’s next used their CD117-LNP as a delivery system for DNA-modifying agents like Cre-recombinase, which scans the genome for a pair of identical LoxP sequences and removes whatever is between them (3). They packaged Cre-recombinase mRNA inside the CD117-LNP and injected them into mice with HSC in which LoxP sites flanked sequences that stop the expression of various fluorescent proteins. Successful LNP uptake led to removal of the stop sequences and caused the cell to light up. Four months after injecting Cre-LNP, the researchers observed fluorescence in the majority of the mouse red blood cells as well as in their B and T cells, which all descend from HSC. “This is kind of proof that we actually target the true stem cells,” Parhiz said. 

The researchers next wanted to see if their system could edit single bases in the human genome. This type of editing would benefit patients with sickle cell anemia, where converting just one adenine to guanine changes the pathogenic gene into a nonpathogenic variant (4). They used two LNP: one that contained mRNA that codes for a CRISPR-Cas9 adenine base editor and one that contained the guide RNA to bring the editor to the sickle cell mutation. When they injected both into a human blood sample, the percentage of sickle cells dramatically decreased.

“Obviously, we hope that soon every single mutation can be corrected, but this may not be possible in the short term,” Rivella said. Many mutations remain untreatable or uncharacterized. “In those cases, already providing a less toxic way to do immunoablation can be very powerful and very important for these patients.” 

To find out whether their LNP could be a viable alternative to immunoablation, the team packaged mRNA that coded for p53 up-regulated modulator of apoptosis (PUMA), a protein that promotes cell death, inside the CD117-LNP. When they delivered the LNP to mice with diseased bone marrow cells, PUMA killed the cells and allowed treated mice to subsequently take in healthy cells from an external donor, whereas untreated mice could not accept donor cells. 

Rivella and Parhiz both attributed this accomplishment to the partnership between their respective labs. “Papers like this don’t just happen,” Parhiz said. “It was a great collaboration and fantastic team behind this.” Rivella added, “We've been working together very well, and we hope to do many more, more things together.”

The researchers plan to explore the versatility of antibodies and the diversity of cell type specific receptors in the body to determine what other cell types can serve as potential targets for their system. “Thankfully, whatever cell type that so far we wanted to target, we were able to,” Parhiz said. “I don’t want to claim that all of them will be successful at the same level… but we are trying basically everything.” 

References

1. Breda L. et al. In vivo hematopoietic stem cell modification by mRNA delivery. Science  381, 436-443 (2023). 
2. Yee N.S. et al. Mechanism of down-regulation of c-CD117 receptor. Roles of receptor tyrosine kinase, phosphatidylinositol 3'-kinase, and protein kinase C.J Biol Chem  50, 31991-31998 (1994). 
3. Madisen L. et al. A robust and high-throughput Cre reporting and characterization system for the whole mouse brain.Nat Neurosci  13, 133-40 (2010). 
4. Newby GA. et al. Base editing of haematopoietic stem cells rescues sickle cell disease in mice.Nature  595, 295-302 (2010).