Human lung alveolus chip infection model enables investigation of viral replication, inflammatory responses, and genetic off-target effects of a novel pan-influenza CRISPR therapy
By Benjamin Boettner
(BOSTON) – The Influenza A virus (IAV) has been the cause of six major flu pandemics, responsible for 50 to 100 million deaths globally. In the U.S. alone, it is estimated that, despite seasonally updated vaccines, IAV infections still lead to 140,000 to 710,000 hospitalizations and 12,000 to 52,000 deaths annually.
The development of antiviral treatments against IAV – or more durable vaccination approaches for that matter – has been extremely challenging because IAV readily develops resistance against them by changing its genetic makeup. To date, its ability to “mutate,” rearrange its genetic information, or even recombine it with that of other IAV viruses infecting the same cell has been an insurmountable challenge for drug developers, and presents a constant risk for new pandemic strains to emerge.
The search for an effective weapon against IAV’s ever-changing genetic makeup has been hampered by the absence of a suitable human in vitro model for testing new treatments. This challenge is compounded by the fact that animal models of IAV infection fail to accurately replicate human immune responses, and drug delivery to human lung tissue operates under different conditions than in animals. New approaches based on CRISPR gene editing technology are being explored, but the sequences being targeted are so human-specific that studies can’t be carried out in animal models in a meaningful way.
Now, a new collaborative study from the Wyss Institute for Biologically Inspired Engineering at Harvard University addressed these challenges by simultaneously leveraging a microfluidic “breathing” human lung alveolus chip (Lung Chip) model of IAV infection developed in the group of Founding Director Donald Ingber, M.D., Ph.D., drug delivery platforms advanced by Associate Director Natalie Artzi, Ph.D. and her group, as well as state-of-the-art CRISPR technology. The team achieved this by designing CRISPR machinery targeting a strongly conserved sequence in IAV’s genome, packaging it up in tiny nanoparticles with affinity to lung epithelial cells, and delivering the loaded particles to lung epithelial cells lining a microfluidic channel in the Lung Chip that were infected with a pandemic IAV. As a result, the load of the virus in the engineered tissue was reduced by more than 50% after a single administration of the treatment, and the host inflammatory response caused by the virus was significantly blunted. Importantly, only minimal off-target effects, as revealed by transcriptomic analysis, occurred in the system. Thus, this Organ Chip model that better mimics human IAV infection than other preclinical models enables the efficacy and safety of CRISPR RNA therapies to be evaluated in a more clinically relevant way than earlier approaches. The findings are published in Lab on a Chip.
Our findings demonstrate that the human Lung Chip model of IAV infection is a highly valuable preclinical testbed for CRISPR RNA therapeutics that act broadly across virus strains because it not only reports on their efficacy in a human-relevant manner but, importantly, also allows assessment of their potential off-target effects.
“Our findings demonstrate that the human Lung Chip model of IAV infection is a highly valuable preclinical testbed for CRISPR RNA therapeutics that act broadly across virus strains because it not only reports on their efficacy in a human-relevant manner but, importantly, also allows assessment of their potential off-target effects, which we find so far are minimal,” said Ingber. “Given the high likelihood of future pandemics and natural seasonal variation of IAV, such pan-IAV antiviral treatments could help us get ahead of the virus and, potentially, save thousands of lives.” Ingber is also the Judah Folkman Professor of Vascular Biology at Harvard Medical School and Boston Children’s Hospital, and the Hansjörg Wyss Professor of Biologically Inspired Engineering at SEAS.
When CRISPR meets Organ Chips
Efforts to develop antiviral treatments against IAV thus far have heavily relied on animal models like mice, hamsters, and ferrets that can be experimentally infected with IAV strains, whether laboratory-adapted or clinically relevant human isolates, and develop flu-like symptoms. However, they have multiple shortcomings: they differ from humans in anatomy, physiology, and genetic design, which could also affect the efficiency with which the virus enters lung cells and replicates in them. Importantly, also their immune systems function differently than that of humans in important aspects; and drug delivery to their lung tissues has different requirements than drug delivery to human lung tissue.
To overcome those limitations, Ingber’s group, with support from the National Institutes of Health (NIH)’s National Center for Advancing Translational Sciences, in 2017, launched an NIH-funded Human Organ Chip project to more accurately model human IAV infection. The team built on the Wyss Institute’s Organ Chip platform, advanced with support from the Defense Advanced Research Projects Agency (DARPA). Developed as part of that platform, the human Lung Chip, which mimics the lung’s tiny air sacs (alveoli) that allow the exchange of oxygen and CO2 and are also the site of IAV infection enabled the team in 2022 to model human-like IAV infection, and to uncover novel therapeutics that modulate the first-line immune response (innate immunity) of infected hosts to prevent life-threatening lung inflammation. Some of these therapeutics have moved on to human clinical trials for evaluation of their effects in patients with COVID-19.
In the Lung Chip, a human primary alveolar epithelium is created in one of two parallel running channels of a microfluidic chip the size of a memory stick, while human pulmonary vascular endothelial cells are cultured in the other channel to mimic a supporting blood vessel. Both channels are separated by a thin porous membrane to allow the free exchange of molecules and gases, as happens in alveoli of the human lung. By flowing culture medium through the “vascular channel,” the researchers can simulate the natural blood flow in alveoli; and by circulating air through the “epithelial channel” and cyclically stretching and relaxing the engineered tissue by pressurizing and depressurizing two hollow side chambers flanking the two channels, the researchers can mimic the lung’s breathing “air-liquid interphase.” To create a baseline for their new study, they used this system and added IAV virus particles of different strains, including the pandemic H3N2 strain, which caused a global flu pandemic with 1 million deaths in 1968, through the epithelial channel. This faithfully replicated many of the intricacies of IAV infection, including virus entry and replication in lung cells, the release of new viruses, and inflammatory and immune responses triggered by the infection.
However, the new study took the platform much further by demonstrating that it has unprecedented value as an effective preclinical tool for developing a new class of antiviral drugs that utilize CRISPR technology, which targets gene sequences that are highly species-specific. The gene-editing system had already been explored as an antiviral platform against IAV, but mostly in dish cultures and animal models. Important translational challenges like human tissue-specific CRISPR delivery and potential off-target effects of the virus-targeting CRISPR machinery on the gene expression of human lung cells, so far, could not be investigated.
Members of Ingber’s team, including Yuncheng Man, Ph.D., the first author and a postdoctoral fellow in Ingber’s lab at Boston Children’s Hospital, designed a version of the CRISPR machinery that uses two CRISPR RNAs (crRNAs) targeting two invariable regions in the so-called polymerase basic 1 (PB1) gene of IAV, which they found are conserved across the vast majority of IAV viruses that infect humans. When the researchers introduced the crRNAs along with an mRNA molecule encoding the RNA-destroying Cas13 enzyme, the crRNAs guided Cas13 to the PB1 target sequence of different IAVs that they used to infect cultured lung cells with. This resulted in a more than 80% reduction of PB1 RNA levels.
CRISPR: signed, sealed, and delivered
To study the therapeutic effects of their pan-IAV CRISPR therapy, the team had to package it up in particles to protect it from unspecific degradation and enable it to be delivered more efficiently to lung epithelial cells in the Lung Chip. “Finding the right type of particle that allows efficient packaging of the therapeutic RNA molecules and their internalization by human lung alveolar tissue is a challenge of its own,” said Man.
In their quest, the team greatly benefited from the drug delivery expertise of co-author Natalie Artzi and the platform that her group is advancing for enabling tissue-specific drug delivery. By comparing different nanoparticle chemistries and organizations, they identified a subtype of polymer-based nanoparticle to deliver greater amounts of molecular cargo to human primary lung alveolar epithelial cells than others.
“Delivery remains the key bottleneck for nucleic acid therapeutics. Our work on delivering CRISPR-based antivirals shows that nanoparticle formulation critically determines efficacy, particularly in primary human alveolar cells. Testing these systems in a human lung organ chip model provides physiologically relevant insight that cannot be captured in standard cell lines, helping to advance our understanding of how to translate CRISPR antivirals to patients,” said Artzi, who is also the Hansjörg Wyss Associate Professor of Biologically Inspired Engineering at Harvard Medical School and Brigham and Women’s Hospital.
When introduced into the air channel of the “breathing” human Lung Chips infected with IAV, the delivered CRISPR machinery strongly inhibited viral replication and lowered viral titers by more than 50%, compared to a CRISPR control machinery not targeting any specific RNA sequence. The researchers also demonstrated the therapeutic impact of their treatment by measuring the levels of inflammatory molecules produced by the lung tissue assembled in the Lung Chip, as well as the attachment of so-called peripheral blood mononuclear cells, a group of immune cells with inflammatory potential, to vascular endothelial cells after adding them to the media flown through the blood vessel-mimicking channel. Remarkably, a single treatment could significantly reduce the levels of inflammatory markers and immune cell recruitment in the infected Lung Chips, when compared to infected Lung Chips treated with the non-targeting CRISPR control.
Delivery remains the key bottleneck for nucleic acid therapeutics. Our work on delivering CRISPR-based antivirals shows that nanoparticle formulation critically determines efficacy, particularly in primary human alveolar cells.
“For the first time, we were able to assess potential off-target effects of an IAV-specific CRISPR therapy in living human lung tissues,” said Man. “Since the pan-IAV CRISPR system that we designed targets viral RNA in lung cells, we wanted to assess whether it would also non-specifically target RNAs that belong to the cells’ natural repertoire and thus cause changes in gene expression that could be a concern for safety.” The team first analyzed how the gene expression of human lung alveolar tissue on-chip responded to IAV and found that it faithfully replicated the immune response normally occurring in lung alveoli of patients with the flu – many of the same inflammatory genes increased their expression. “Importantly, we found that the non-targeting control CRISPR system did not affect gene expression in the Lung Chips, and that the genes changed in their expression by the targeted pan-IAV version did not share any sequence identities with the targeted IAV sequence, but were associated with downregulated host inflammatory pathways,” said Man, adding that “this strongly argues for the general safety of our approach.”
Other authors on the study included Ryan Posey, Haiquing Bai, Amanda Jiang, Pere Dosta, Diana Ocampo-Alvarado, Robert Plebani, Jie Ji, and Chaitra Belgur. The study was supported by Defense Advanced Research Projects Agency (DARPA) under Cooperative Agreement HR0011-22-2-0017, and the Wyss Institute at Harvard University.