Regions of the genome that can safely accommodate new therapeutic genes, without causing other, unintended changes in a cell’s genome that could pose a risk to patients—known as genomic safe harbors (GSHs)—could enhance the efficacy, persistence, and predictability of gene and cell therapies.
Now, a collaborative effort has developed a computational approach to identify GSH sites. For two out of 2,000 predicted GSH sites, the team provided an in-depth validation with adoptive T cell therapies and in vivo gene therapies for skin diseases in mind. By engineering the identified GSH sites to carry a reporter gene in T cells, and a therapeutic gene in skin cells, respectively, they demonstrated safe and long-lasting expression of the newly introduced genes.
The study is published in Cell Reports Methods in the paper, “Discovery and validation of human genomic safe harbor sites for gene and cell therapies.”
Finding GSHs with potential for clinical translation has been a challenge. A GSH needs to be accessible by genome editing technologies, free of physical obstacles like genes and other functional sequences, and allow high, stable, and safe expression of a “landed” therapeutic gene. Thus far, only a few candidate GSHs have been explored—all with certain caveats. In addition, candidate GSHs have not been analyzed for the presence of regulatory elements nor whether inserted genes change global gene expression patterns in cells across the entire genome.
“While GSHs could be utilized as universal landing platforms for gene targeting, and thus expedite the clinical development of gene and cell therapies, so far no site of the human genome has been fully validated and all of them are only acceptable for research applications,” said George Church, PhD, Wyss Institute core faculty member. “This makes the collaborative approach that we took toward highly-validated GSHs an important step forward. Together with more effective targeted gene integration tools that we develop in the lab, these GSHs could empower a variety of future clinical translation efforts.”
A computational pipeline allowed the team to predict regions in the genome with potential for use as GSHs by harnessing the wealth of available sequencing data from human cell lines and tissues. “In this step-by-step whole-genome scan we computationally excluded regions encoding proteins, including proteins that have been involved in the formation of tumors, and regions encoding certain types of RNAs with functions in gene expression and other cellular processes,” said Erik Aznauryan, PhD, research fellow in the Church lab at the Wyss Institute. “We also eliminated regions that contain so-called enhancer elements, which activate the expression of genes, often from afar, and regions that comprise the centers and ends of chromosomes to avoid mistakes in the replication and segregation of chromosomes during cell division. This left us with around 2,000 candidate loci all to be further investigated for clinical and biotechnological purposes.”
Out of the 2,000 identified GSH sites, the team randomly selected five and investigated them in common human cell lines by inserting reporter genes into each of them using a rapid and efficient CRISPR-Cas9-based genome editing strategy. “Two of the GSH sites allowed particularly high expression of the inserted reporter gene—in fact, significantly higher than expression levels achieved by the team with the same reporter gene engineered into two earlier-generation GSHs. Importantly, the reporter genes harbored by the two GSH sites did not upregulate any cancer-related genes,” said Aznauryan.
To evaluate the two most compelling GSH sites in human cell types (Rogi1 and Rogi2) with interest for cell and gene therapies, the team investigated them in immune T cells and skin cells, respectively. T cells are used in a number of adoptive cell therapies for the treatment of cancer and autoimmune diseases that could be safer if the receptor-encoding gene was stably inserted into a GSH. Also, skin diseases caused by harmful mutations in genes controlling the function of cells in different skin layers could potentially be cured by insertion and long-term expression of a healthy copy of the mutated gene into a GSH of dividing skin cells that replenish those layers.
“We introduced a fluorescent reporter gene into two new GSHs in primary human T cells obtained from blood, and a fully functional LAMB3 gene, an extracellular protein in the skin, into the same GSHs in primary human dermal fibroblasts, and observed long-lasting activity,” said Denitsa Milanova, PhD, Wyss Institute technology development fellow.
“An extensive sequencing analysis that we undertook in GSH-engineered primary human T cells clearly demonstrated that the insertion has minimal potential for causing tumor-promoting effects, which always is a main concern when genetically modifying cells for therapeutic use,” said Sai Reddy, PhD, associate professor of systems and synthetic immunology at ETH Zurich in the department of biosystems science & engineering, Basel, Switzerland. “The identification of multiple GSH sites, as we have done here, also supports the potential to build more advanced cellular therapies that use multiple transgenes to program sophisticated cellular responses, this is especially relevant in T cell engineering for cancer immunotherapy.”
“This collaborative interdisciplinary effort demonstrates the power of integrating computational approaches with genome engineering while maintaining a focus on clinical translation,” said Donald Ingber, MD, PhD, Wyss Institute founding director. “The identification of GSHs in the human genome will greatly augment future developmental therapeutics efforts focused on the engineering of more effective and safer gene and cellular therapies.”