Organ-On-A-Chip Platform For Studying Immune Response To Vaccines

By Deborah Borfitz 

June 21, 2022 | A first in the world of organs-on-a-chip, researchers at the Wyss Institute for Biologically Inspired Engineering at Harvard University have come up with a lymphoid follicle (LF) model that can be used to probe the function of the immune system and predict its response to vaccines—including the three (Pfizer, Moderna, and J&J) currently in use for the prevention of COVID-19. “We hit upon the recipe that works and, in retrospect, it makes sense given what happens naturally,” says senior staff scientist Girija Goyal, Ph.D. 

When human B and T cells were being cultured inside a microfluidic device, they started to spontaneously self-organize into 3D structures akin to germinal centers within LFs that form in response to vaccination, she explains. There the B cells make protective antibodies and differentiate into a plasma cell. 

That entire loop can now be studied with the new LF chip in lieu of using cell culture slides or non-human primates. The engineering feat was recently reported in Advanced Science (DOI: 10.1002/advs.202103241). 

Critical factors in coaxing the cultured B and T cells to form functional lymphoid follicles were their placement in a 3D extracellular matrix gel, limiting their movement, and the organ-on-a-chip microfluidic device that allowed perfusion as would normally occur in the lymph nodes, Goyal says. 

The culture media alone is not enough, she emphasizes. The dynamic perfusion is what prompts the cells to self-assemble into much larger 3D multicellular aggregates after two to four days—and this was consistent and reproducible across multiple donors. 

This was a surprise, says Goyal, given that when this biology was studied in animals it was proposed that other cells of the lymph nodes such as stromal cells and antigen-presenting myeloid cells were expected to be the drivers of follicle formation. But, as has been also shown in an animal model, “lymphocytes by themselves autonomously assemble without cues provided by these structural cells.” 

Molecular Reprogramming 

Pivoting from the original experiment to figure out what the mysterious 3D structures were, the researchers found the cells were secreting a chemical called CXCL13—a hallmark of LF formation within lymph nodes and in other parts of the body that occurs in response to chronic inflammation, she continues. They also found that B cells within the self-assembled LFs expressed an enzyme called activation-induced cytidine deaminase (AID) that is critical for activating B cells against specific antigens and is not present in B cells that are circulating in the blood. 

This was molecular confirmation that the B cells were being reprogrammed into a lymph node-like phenotype, even though they were taken from circulating blood, Goyal explains. Notably, neither CXCL13 nor AID were present in cells cultured in a standard 2D dish. 

“What is functionally important is that AID is the enzyme that allows B and T cells to class switch, which means all these cells start by producing one type of antibody known as IgM, but then switch to produce high affinity IgG antibodies when activated,” she says. AID is required for their conversion to produce this IgG form of antibody that provides better protection from pathogens. 

These findings align with what is known clinically, adds Goyal. It is also known that the function of AID and CXCL13 correlates with successful vaccination in people in clinical trials. 

Additionally, study results indicate the LF chips might be useful in determining how the immune system responds to different types of pathogens, she says. After treating the chips with the cytokine IL-4 and an anti-CD40 antibody, researchers detected the presence of long-lived antibody-producing plasma cells in the chips that clustered within the LFs as they would in vivo. “We don’t know how long the plasma cells will live, but at least we know that the right molecular markers are being formed.” 

From Flu to Polio 

For their study, Harvard researchers focused on influenza by inoculating the chips with an inactivated research formulation of the H5N1 flu vaccine, with and without the adjuvant SWE (squalene-in-water emulsion) to boost immune response, and the commercially available Fluzone influenza vaccine. LF chips treated with the H5N1 vaccine and booster, as well as those receiving Fluzone, produced significantly more plasma cells and anti-influenza antibodies than B and T cells grown in 2D cultures or LF chips that received the vaccine without SWE, says Goyal.  

Now, the research team is partnering with a pharmaceutical company to study vaccines against COVID and rabies, she reports. They are also writing a grant in hopes of conducting a side-by-side comparison of the three in-use COVID vaccination platforms in terms of their ability to induce antibodies and antigen-specific T cells. 

Earlier in the pandemic, scientists from the Wyss Institute used a novel lung airway chip (famously developed by Donald Ingber) to screen drugs for use against influenza and SARS-CoV-2. Results of that investigation were published in Nature Biomedical Engineering (DOI: 10.1038/s41551-021-00718-9). 

LF chips could be used to test multiple types of vaccines, since their first purpose is to induce antibodies and destroy pathogens, Goyal says. They can’t directly test vaccine efficacy—a COVID vaccine, for example, would require the airway chip to see if protective immune cells reached the lung—but the chips can indicate if the immune system is reacting in a favorable way by creating antibodies to block the virus. 

In experiments now underway, she adds, the research team is attempting to use LF chips to learn if a vaccine is harmful. Specifically, they are recapitulating the body’s inflammatory response to different vaccines and hoping to correlate their cytokine profile with their known toxicity. 

Most organs-on-chips being used at the Wyss Institute for preclinical research are about the size of a USB memory stick, says Goyal. At present, a total of 48 chips lined by cells from two to eight human donors can be run in an incubator the size of a mini fridge.  

“We’re working now to miniaturize it to see if we can increase the throughput,” she continues. But even as-is, the LF chip is “very comparable” to a mouse study in terms of effort, and much more efficient than a nonhuman primate study involving no more than four to eight larger animals.  

In the case of polio vaccines, Goyal and her colleagues have already started trying to connect the LF chip to other model organ systems for testing intramuscular vaccines and they hope to expand that to intranasal vaccines soon, she reports. Ultimately, they want to connect their model to a gut-on-a-chip device to study oral polio vaccines. 

Connecting the LF and gut chips would also have value in investigations of other inflammatory insults to the intestines, including inflammatory bowel disease (IBD), she notes. The two-organ system could be used to look at how IBD-caused changes to the intestinal microbiome impact the immune system and even the brain if linked to chips lined by brain cells.