Animal models continue to be a mainstay of basic research and preclinical studies. But since animals such as mice only approximate humans, they may generate findings of limited clinical relevance. Fortunately, animal models have become more humanized and much better at generating human-relevant findings, thanks to gene editing technology. Indeed, humanized animal models are readily available from suppliers. Such models include mice that have been engrafted with human cells or engineered to express human gene products.

Alongside the evolution of humanized animal models is the ongoing development of organ-on-a-chip (OOC) models. OOC models range in complexity. Some are single-organ chips. Others are multiorgan microphysiological systems that can capture the subtler mechanisms of systems biology. Accordingly, OOC models are beginning to play a role in drug development and preclinical safety testing, and their future holds even greater promise.

Focusing on immunology

According to Michael Seiler, PhD, vice president, commercial products, Taconic Biosciences, “A heightened focus on immunology as a contributor to systems biology influences how investigators use both immunocompetent and immunodeficient animal models.”

An example of an immunocompetent model is Taconic’s Diet Induced NASH B6, which replicates the progressive and chronic liver inflammation that develops in nonalcoholic steatohepatitis. The NASH B6, which has a fully functioning mouse immune system, can generate predictive results in therapeutic studies. Such models can support studies in various therapeutic areas. Still, an equally strong need exists for immunodeficient animal models, particularly those that gain facets of the human immune system through primary cell engraftments. Such animal models include humanized mice.

“Given the complexity of modeling the human immune system in vivo, we see an increased requirement for optionality in immunodeficient models,” Seiler remarks. “[The most notable options include] the ability to choose the host mouse as well as the primary human cells that will best contribute meaningfully to the experimental design and objective.”

Taconic Biosciences’ chart
Taconic Biosciences’ portfolio of human immune system mouse models includes the huNOG-EXL SA (Standard Access) model and the huNOG-EXL EA (Early Access) model. Both support high levels of human cell engraftment and the differentiation of human myeloid and lymphoid cells. The SA model is shipped to customers 10 weeks after engraftment, whereas the EA model is shipped within 2 weeks of engraftment, prior to the expansion of significant numbers of human immune cells in peripheral blood, enabling the study of slow-growing tumors on longer experimental timelines.

Taconic has expanded its portfolio of humanized models with the addition of the huNOG-EXL EA (early access) mouse model. The huNOG-EXL EA offers a study window that is longer than the one offered by the original huNOG-EXL model, which was launched in 2016. Like many standard humanized models, the huNOG-EXL is typically delivered weeks after engraftment because cell differentiation must be confirmed.

The huNOG-EXL EA retains the advantages of the huNOG-EXL. For example, it supports the differentiation of multiple human cell types known to affect adaptive immunity, including myeloid and lymphoid cells. Also, it is exceptionally reliable and consistent, reflecting Taconic’s experience injecting tens of thousands of mice with hematopoietic stem cells to generate the huNOG-EXL.

In addition, the huNOG-EXL EA can be provided soon after engraftment, prior to the model demonstrating full immune system humanization. This early-access model allows a tumor to be introduced sooner while ensuring that the model’s humanized immune system will mature as expected.

Another recent immunology model, a B-cell-deficient Jh syngeneic model on a B6 background, can help investigators study biologics more efficaciously. For example, the model allows investigators to avoid the complications and risks of anti-drug antibody development.

Taconic sees “exciting opportunities enabled by genetically engineered strains with well-characterized loss of function phenotypes,” Seiler declares. In addition, Taconic will soon announce a microbiome initiative to remove obstacles that have limited investigators to date.

Rethinking animal models

Recent animal model supply chain issues have made it more imperative than ever for researchers to use the most prudent models. Possibilities for moving to more accessible models—that is, models that can be more easily bred, maintained in-house, or even replaced—are now being considered with a greater sense of urgency.

The Jackson Laboratory (JAX) provides researchers worldwide with innovative animal models and preclinical in vivo services. These tools also align with a recently released FDA directive to seek alternatives to nonhuman primates. Several platforms, including immune-humanized mice and genetically diverse models, can address these needs and produce translationally relevant data.

One of the latest platforms from JAX evaluates cytokine release syndrome (CRS). The platform, a peripheral blood mononuclear cell (PBMC)-engrafted humanized mouse, can be used to assess immune-stimulating therapeutics such as chimeric antigen receptor (CAR) T cells and bispecific antibodies. These therapeutics have demonstrated great promise and efficacy, but they can cause PBMCs to overexpress cytokines, giving rise to CRS.

Until recently, the only options for CRS preclinical screening were in vitro platforms and nonhuman primate models. Humanized mouse models, which offer greater speed, accessibility, and versatility, may be more useful.

“We have taken this platform a step further by performing the engraftment in mice lacking MHC class I and II, which dramatically delays the onset of acute graft-versus-host disease,” says Brian Soper, PhD, senior scientific engagement manager, JAX. “This allows performance of longer-term efficacy studies in a model that more closely reflects the human immune system.”

Another JAX initiative is to offer more genetically diverse models such as the J:ARC and HET3 mouse models. “These strains differ drastically,” Soper notes, “but they share value for research and preclinical testing in that their genetic heterogeneity allows us to model likely biological interactions in a diverse population.” Applications range from safety and toxicology screening to quantitative trait locus (QTL) mapping. HET3 also has a unique position in the aging field as it mimics some of the sex differences in aging in the human population.

Genetically diverse platforms facilitate the implementation of precision medicine treatments. For example, these platforms not only inform go/no-go decisions, but they also improve patient stratification, identifying patients for whom treatments might have greater safety and efficacy.

Studying SARS-CoV-2

A challenge of the pandemic has been the lack of animal models that could contribute to the study of human immune cells in SARS-CoV-2 viral infections. Such animal models, however, are starting to become available. For example, Charles River Laboratories has developed humanized mouse models that are susceptible to COVID-19. According to Steve Festin, the company’s senior director of scientific and commercial development, the new models could help investigators understand how different cells in the human immune system respond to early SARS-CoV-2 infection.

In collaboration and under license with GemPharmatech, Charles River recently introduced the hACE2-NCG mouse model. It is based on the triple-immunodeficient NCG mouse, and it reflects the use of a genetic modification strategy that knocks in human ACE2 at the mouse ACE2 locus. (The human ACE2 gene expresses hACE2, the receptor used by SARS-CoV-2 to enter cells.)

“In this model,” Charles River elaborates, “the intracellular domain of the mouse contains ACE2 under the transcriptional regulation of endogenous sequences, designed to mimic the physiological expression pattern of ACE2 in various tissue types, including the kidney, lungs, and intestines.” The company also indicates that its “genetically humanized” model is capable of being “immuno-humanized” with PBMCs or CD34-positive cells.

“As research regains momentum with the majority of laboratories back online, existing trends in oncology discovery and safety are expected to continue, along with greater emphasis on research models supporting cell and gene therapy initiatives,” Festin relates. “We will continue to focus our efforts on providing innovative and meaningful tools for biomedical research and our drug development partners.”

Contextualizing disease correctly

Compounds and targets need to be studied in the correct disease context to ensure a successful development path. Using the wrong models at the start of drug development may very well lead to the wrong drug, as demonstrated by the high attrition of new drug candidates.

Models of healthy and diseased human organs from Mimetas can enable assay development and the screening of thousands of compounds. The company is positioning the models for use in early phases of drug development, including target and compound discovery. Kidney, gut, lung, liver, brain, vasculature, bone marrow, tumor, placenta, and immune system models are available.

The core product is an organ-on-a-chip platform (OrganoPlate). Other products include ready-to-assay 3D tissue models (OrganoReady models), an instrument for taking transepithelial electrical resistance measurements (OrganoTEER), and instruments for driving perfusion flow (OrganoFlow S and OrganoFlow L). Mimetas services include compound profiling and screening services, as well as custom model and assay development services. To further support customers, Mimetas participates in drug development partnerships. Also, the company’s Phenotypic Screening Center is available to facilitate large-scale screening campaigns.

Mimetas OrganoPlate
This image shows a high-throughput compound screen on a Mimetas OrganoPlate that contains 64 tissue culture chips. Each chip consists of three channels—angiogenic factors; extracellular matrix gel; and human umbilical vein endothelial cells (HUVECs)—patterned with PhaseGuide ridges. Inset: A perfused endothelial tubule with angiogenic sprouts growing into the extracellular matrix on the left (not visible). Green: HUVECs; blue: nuclei.

“Recent launches include OrganoStart and OrganoStart Pro packages, OrganoReady blood vessel and vascular bed products, and the high-throughput-screening-compatible OrganoPlate 3-lane 64,” says Paul Vulto, PhD, CEO, Mimetas. “Using 3,500 chips of the OrganoPlate 3-lane 64, we screened 1,546 compounds in duplicate in a 3D angiogenesis assay.1 We also published an assay2 mimicking the onset of inflammation in collaboration with Merck.”

The OrganoPlate comprises proprietary PhaseGuide technology that is designed to allow the horizontal layering of cells and gel matrices without artificial membranes. According to Mimetas, this technology can “enable precise, barrier-free definition of culture matrices and cells in 3D, supporting cell-cell interactions and unprecedented imaging and quantification.” In a recent paper,3 Mimetas scientists asserted that “the co-culture capabilities of the platform can be explored to create complex tissue configurations, for example, by incorporating mesenchymal and immune cells in the ECM adjacent to the epithelial tubes.”

Based on a microtiter plate footprint, the 40- to 96-chip OrganoPlate, comprised of inert materials such as glass and polystyrene, is intended to be fully compatible with microscopes, plate readers, and robotics systems.

“Our flexible suite of offerings, the large array of assays, and the rich ecosystem of instruments around the OrganoPlate make us the go-to company for physiologically relevant modeling,” Vulto asserts. “Throughput and automation of our platform render it well suited for early-stage drug development. Disease context is a crucial piece of the puzzle. Compounds are being progressed or halted based on data from our models.”

Organ-on-a-chip models

“To forward human predictive biology, we build biological models that emulate human physiology more closely,” says Lorna Ewart, PhD, chief scientific officer, Emulate. “Each organ is highly complex. Our philosophy is to start at the simplest level and then layer on complexity.”

Emulate’s Organ-Chips
Emulate’s Organ-Chips employ a transparent, flexible silicone polymer to support two parallel and independent microfluidic channels separated by a flexible, porous membrane. The upper channel is lined by epithelial cells; the lower channel, by organ-specific endothelial cells. Alongside these channels are vacuum channels that provide tissue-relevant mechanical forces. This arrangement recreates tissue-tissue interfaces.

The Emulate Organ-Chips employ a polydimethylsiloxane (PDMS) material to create two parallel independent microfluidic channels separated by a porous membrane. Tissue-tissue interfaces are enabled by taking epithelial cells and endothelial cells from the organ of choice, and then adding them to the upper channel and the lower channel, respectively.

Microfluidic technology applies a programmable stream of culture media across the cells simulating sheer stress, an important signal for the cells. “The PDMS also allows us to apply mechanical stress, another important cue for organs such as the lung and intestine,” Ewart notes. “This enhances the physiological relevance.”

Five product lines representing the lung, intestine, liver, kidney, and brain are supplied as Bio-Kits that include the Organ-Chips, Pod Portable Modules, qualified human cells, and chip activation reagents. Bio-Kit resources include validated protocols.

The chips are also offered on a standalone basis for researchers desiring to build their own models. Recent applications for the Colon-Intestine Chip (which incorporates prequalified biopsy-derived primary organoids and colonic endothelial cells) and the Brain-Chip (which incorporates five different cell types) show the introduction and pharmacological modulation of inflammation.

In a recent study, Emulate qualified its Liver-Chip for predictive toxicology. The performance of 780 Liver-Chips was assessed across a blinded set of 27 known hepatotoxic and nontoxic drugs. “We wanted to know how well the human Liver-Chip could predict drug-induced liver injury,” Ewart relates. “Results showed 87% sensitivity and 100% specificity when differentiating hepatotoxic from nonhepatotoxic small molecules.”1

Notably, all the included 22 hepatotoxic drugs had previously been classified as safe due to a lack of toxicity in animal models. Collectively, these compounds resulted in 208 patient fatalities and 10 liver transplants. If the human Liver-Chip had been used during preclinical screening, it is likely that many of these fatalities could have been avoided, Ewart suggests.

A financial framework assessed the economic impact of the technology and showed that the use of the Liver-Chip in pharmaceutical development programs for new candidate drugs could provide the equivalent of $3 billion in improved R&D productivity.4

Multiorgan microphysiological systems

Potential alternatives to animal models include multiorgan microphysiological systems (MPSs). Besides providing a way to implement the 3Rs (Reducing, Replacing, and Refining animal models to satisfy ethical considerations), MPSs may generate in vitro findings that are more clinically relevant than the findings from animal models. The main applications fall within disease modeling; absorption, distribution, metabolism, and excretion (ADME) testing; and toxicology/safety testing.

MPSs from CN Bio Innovations may incorporate several organ-on-a-chip systems, such as systems representing lung, liver, and gut. Systems representing single organs can be linked together into multiorgan systems to simulate processes such as drug absorption and metabolism, or to advance the study of interorgan interactions such as inflammation. CN Bio offers MPSs, validated 3D cell cultures, compatible consumables, and research services. The company maintains that its offerings can address drug development bottlenecks across many therapeutic areas, including metabolic and infectious diseases, oncology, and inflammation.

CN Bio Innovations Organ-on-a-chip (OOC) platforms diagram
Organ-on-a-chip (OOC) platforms developed by CN Bio Innovations include the PhysioMimix OOC Single-Organ System and the PhysioMimix OOC Multi-Organ System. These systems, which rely on organ-specific microphysiological system (MPS) consumable plates, can mimic complex human processes in the laboratory and help fast-track drug development. For example, the FDA demonstrated that the PhysioMimix liver-on-a-chip system can generate useful data for drug safety and metabolism applications.

The PhysioMimix range of single- and multiorgan MPSs harnesses microfluidic technology to mimic blood flow and precisely control the cellular microenvironment. In a collaborative study, CN Bio and FDA scientists demonstrated that data derived using the PhysioMimix liver-on-a-chip system are appropriate for use in drug safety and metabolism applications, evidencing its enhanced performance versus standard in vitro models.5

Many organ-on-a-chip systems currently available complement each other. “There is an inverse relationship between physiological relevance and throughput,” says David Hughes, DPhil, CEO, CN Bio. “Higher-throughput organ-on-a-chip solutions at the start of the screening cascade offer more binary, yes/no decisions. These are perfectly complemented by systems like ours, which can be used for smaller numbers of compounds in preclinical studies where the highest biological and translational relevance is essential.”

“Cultures can be maintained up to four weeks,” Hughes continues, “supporting, for example, studies of how chronic drug dosing may relate to the potential for drug-induced liver injury.” In addition, new, interconnected multiorgan (gut and liver) models recreate human processes in the laboratory such as first-pass metabolism to estimate drug bioavailability.

The open architecture system provides flexibility. New PhysioMimix models are fine-tuned to match their human counterparts as closely as possible through the application of inter- and intraorgan-specific flow rates that deliver human-relevant sheer forces. Purpose-built hardware and consumable plates support tissue-specific requirements such as oxygen gradients essential to the gut microbiome.


1. Soragni C, Ng CP, Heijmans J, et al. A robotised 1546 compound screen in a perfused 3D microfluidic angiogenesis assay. Poster presented at SLAS Europe 2021; June 22–25, 2021; Vienna, Austria.
2. De Haan L, Suijker J, van Roey R, et al. A Microfluidic 3D Endothelium-on-a-Chip Model to Study Transendothelial Migration of T Cells in Health and Disease. Int. J. Mol. Sci. 2021; 22(15):8234. DOI: 10.3390/ijms22158234.
3. Trietsch SJ, Naumovska E, Kurek D, et al. Membrane-free culture and real-time barrier integrity assessment of perfused intestinal epithelium tubes. Nat. Commun. 2017; 8(1): 262. DOI: 10.1038/s41467-017-00259-3.
4. Ewart L, Apostolou A, Briggs SA, et al. Qualifying a human Liver-Chip for predictive toxicology: Performance assessment and economic implications. bioRxiv. Preprint posted on December 29, 2021. DOI: 10.1101/2021.12.14.472674.
5. Rubiano A, Indapurkar A, Yokosawa R, et al. Characterizing the reproducibility in using a liver microphysiological system for assaying drug toxicity, metabolism, and accumulation. Clin. Transl. Sci. 2021; 14: 1049–1061. DOE: 10.1111/cts/12969.