Scientists at Yale University and Carnegie Mellon University have used gene editing to effectively cure the genetic blood disorder β-thalassemia, in mouse fetuses, in utero. The approach, which harnesses peptide nucleic acid (PNA) technology pioneered at Carnegie Mellon’s Center for Nucleic Acids Science and Technology (CNAST), corrected the disease-causing β-globin gene mutation, without any adverse effects on the fetus or pregnancy, and with no evidence of off-target effects that can occur with other gene editing techniques, such as CRISPR. The in utero-treated mice showed dramatic improvements in β-thalassemia symptoms, higher haemoglobin levels and increased long-term survival, when compared with untreated controls.
Headed by Adele S. Ricciardi, an M.D./Ph.D. student at Yale University’s Department of Biomedical Engineering, and Danith H. Ly, Ph.D., professor of chemistry, Carnegie Mellon's Mellon College of Science, the researchers say their work could represent a springboard for developing equivalent approaches for human gene correction in utero. “This work may provide the basis for a safe and versatile method of fetal gene editing for human monogenic disorders,” they write in their published paper in Nature Communications, which is entitled, “In utero nanoparticle delivery for site-specific genome editing.”
Every year an estimated eight million children are born worldwide with a severe genetic disorder or birth defect. Of these, hemoglobinopathies, or disorders that affect the structure of hemoglobin, are the most commonly inherited, single-gene disorders, the authors explain.
Genetic disorders such as β-thalassemia can be diagnosed during early pregnancy, offering the potential for genetic correction in utero, but no such treatments exist, and children with β-thalassemia may need to undergo blood transfusions for life, or require bone marrow transplants. “Early in embryonic development, there are a lot of stem cells dividing at a rapid pace,” comments Dr. Ly. “If we can go in and correct a genetic mutation early on, we could dramatically reduce the impact the mutation has on fetal development or even cure the condition.”
While existing approaches to gene therapy, such as stem cell transplants and the delivery of genes using viral vectors, are challenging, the development of site-specific gene editing tools such as CRISPR-Cas9, offer a promising, alternative approach to genetic correction in vivo. “By delivering gene editing therapies in utero, it is possible to gain access to dividing stem and progenitor cell populations, which can result in propagation of the corrected gene in all progeny cells,” the authors write.
However, CRISPR-Cas-based techniques deliver enzymes that cut the host DNA at specific sequences, and the process isn't foolproof and can result in off-target gene editing. In contrast, the PNA-based technology developed by the CNAST researchers works by triggering the cell to engage its own DNA repair machinery to correct the faulty gene sequences, without using exogenous enzymes to snip the DNA. “CRISPR is much easier to use, which makes it ideal for laboratory research,” Dr. Ly adds. “But the off-site errors make it less useful for therapeutics. The PNA technique is more ideal for therapeutics. It doesn't cut the DNA, it just binds to it and repairs things that seem unusual.”
PNAs are synthetic molecules that combine a synthetic protein backbone with nucleobases present in DNA and RNA. The PNA molecules are paired with a donor DNA encoding the corrected hemoglobin gene sequence, and loaded into an FDA-approved poly(lactic-co-glycolic acid) (PLGA) nanoparticle. When delivered to cells the constructs effectively bind to the mutant gene sequence and trigger the cell to correct the sequence to that of the donor DNA template. “The PNAs contain nucleobases supported by a modified polyamide backbone and bind to their specific genomic target site via both Watson−Crick and Hoogsteen base-pairing, yielding PNA/DNA/PNA triplex structures that induce endogenous DNA repair to mediate the recombination of the donor DNA molecule containing the correct sequence and produce specific, in situ gene correction.”
In 2016 the team reported on use of the PNA technology in adult mice to correct the β-thalassemia gene defect and effectively cure animals of disease. For their latest work, the team went one step further, and used a technique similar to amniocentesis to deliver the NP-encapsulated PNA components to the mouse fetuses either intravenously – as a proxy for human umbilical vein transfusion – or directly into the amniotic fluid, at varying stages of gestation. Because mouse pregnancies are relatively short, each fetus received just a single treatment.
Initial tests in healthy mice suggested that the NP-delivered PNA/DNA constructs had no adverse effects on pup health, growth or survival either to weaning, or more long term, and there was no evidence of developmental abnormalities or tumor formation associated with the treatment. Mice that had received the NPs in utero were also themselves able to have successful pregnancies and healthy litters.
When the treatment was then tested in the mouse model of β-thalassemia, the team found that fetuses that had received the NP-encapsulated PNA and donor DNA components in utero grew into adult mice with significantly higher levels of hemoglobin than untreated, β-thalassemic mice. Encouragingly, levels of hemoglobin in the treated animals were in the normal range. Postnatally, the PNA/DNA recipients also exhibited much better red blood cell morphology, and reduced levels of spleen enlargement – “a 73% reduction in splenic weight in treated mice, compared to controls,” the authors write. The in utero-treated β-thalassemic mice also survived much longer than untreated control animals. “At 500 days after birth, in utero treated mice had 100% survival, in contrast to just 69% survival in the untreated group,” the authors point out. Encouragingly, deep sequencing analyses found no evidence of any off-target effects of PNA-based gene editing. “We looked at 50 million samples and couldn't find one offsite error when we used our PNA gene editing technique,” Dr. Ly comments.
An analysis of the treated animals’ bone marrow indicated that single treatment resulted in a gene editing frequency of about 6% in total bone marrow cells, but this was enough to result in improvements that were effectively curative. Interestingly, this level of gene correction was higher than was achieved when the technology was used in adult mice, which received four doses of nanoparticles. “These improvements suggest there may be an advantage to gene editing in the fetus because it is possible to access of rapidly cycling population of HSCs [hematopoietic stem cells] within the fetal liver,” the authors write.
The team also suggests that while a 6% gene editing frequency was achieved in the mouse fetus after a single treatment, the potential to carry out multiple rounds of treatment in animals and humans that have much longer pregnancies than mice, could result in even higher levels of gene editing. “Due to the low toxicity of one dose, we speculate that multiple treatments should be possible in humans or mammals with longer gestational periods, which may result in higher gene editing frequencies.
“Here we demonstrate that in utero delivery of PLGA nanoparticles loaded with PNA/DNA is a safe and effective means of achieving clinically relevant frequencies of site-specific, nonenzymatic gene editing in a mammalian fetus that results in sustained postnatal alleviation of disease,” the authors conclude. “These findings suggest that in utero gene editing has the potential to be safe and produce a clinical response substantial enough to reduce β-thalassemia-associated morbidity and mortality.”