In this post, we explore the wide variety of inflammatory bowel disease (IBD) mouse models available to researchers, including their unique characteristics. Understanding important differences between these models is key to selecting the optimal one for your specific research questions.
Selecting a Mouse Model for IBD Studies
Choosing a mouse model for IBD preclinical studies is not always straightforward. This is because you have many different model types and subtypes to choose from (see table below for key models). Selecting the most appropriate model is largely influenced by the biology and pathology of the target and of an agent and the disease stage required to test the agent. Different models will have advantages and disadvantages that should be weighed in advance.
Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4435576/pdf/main.pdf
Commonly Used IBD Mouse Models
| Model Examples | Location of Response | Prevalent Type of Response | Immunopathology | |
|---|---|---|---|---|
| Chemically Induced | DSS | Colon | Epithelial damage | Th1 for acute and Th1/2 for chronic DSS. NLRP6-deficient mice, an inflammasome protein produced exclusively in epithelial cells, results in more severe DSS colitis |
| TNBS | Colon | Epithelial damage Immune-mediated | TH1-mediated and infiltration of the lamina propria with CD4+ T cells, neutrophils, and macrophages. Blocking TNFSF15 completely blocked TNBS colitis in mouse. |
|
| Oxazolone | Colon | Epithelial damage Immune-mediated | IL-13 and NKT mediated Easier induced in SJL/J or C57BL/10 mice and C57BL/6 mice are resistant to this colitis. Chronic colitis can also be induced in BALB/c |
|
| Spontaneous Mutation | SAMP1/Yit | Ileum | Immune-mediated | sever inflammation of the terminal ileum and caecum driven by a TH1 |
| C3H/HeJBir (C3Bir) | Colon | Immune-mediated | Th1 | |
| Adoptive T Cell Transfer | CD4+CD45RBhi | Colon | Immune-mediated | Th1 and / or Th17 |
| Genetically Engineered | IL-10-/- | Colon | Immune-mediated | Established Th1 |
| Microbiome Induced | Germ-free IL-10-/- +/- microbial transfer | Colon | Immune-mediated | Th1 / Th17 |
The varied models reproduce different IBD features (to varying extents) for histology, therapeutic response, disease location, etc. This post reviews the main model types, looking at their unique characteristics and potential uses.
Chemically Induced IBD Mouse Models
The most widely used models have chemically induced disease (e.g., by dextran sulfate sodium (DSS), trinitrobenzene sulfonic acid (TNBS), or oxazolone). These models’ main advantages are that they are relatively cheap and also quick and easy to develop. Different models in this class have specific uses based upon their different disease induction methods.
DSS-Induced Colitis Model
DSS is a negatively charged sulfated polysaccharide. It damages epithelial cells in mice. Innate immune cells then release cytokines, causing inflammation in the colon, characterized by ulcers and granulocyte infiltration.
Common uses for the DSS-induced colitis model include studying factors that maintain or reestablish epithelium integrity during or after injury and how the innate immune system is involved in intestinal inflammation.
These models also respond to cyclosporine A, providing a relevant model to assess new agents that target the same immune mechanisms (e.g., new immunosuppressants).
TNBS-Induced Colitis Model
TNBS is a haptenizing agent. It is a small molecule that is not antigenic by itself but causes an immune response when it binds to host proteins. TNBS results in a preclinical mouse model replicating clinical Crohn’s disease (CD). The immune response is Th1 mediated, characterized by infiltration of CD4+ T cells, neutrophils, and macrophages. Transversely spreading inflammation develops, resulting in transmural colitis.
TNBS-induced colitis models are ideal to study the immunologic aspects of CD and test the efficacy of potential immunotherapies.
Oxazolone-Induced Colitis Model
Oxazolone is also a haptenizing agent, but it induces a different kind of inflammation than TNBS does. This model is more like clinical ulcerative colitis, including similarities in immunopathogenesis. The immune response is Th2 mediated, resulting in diffuse colonic inflammation.
This model was used to study delayed-type hypersensitivity reactions in the skin. It is also used to assess agents targeting Th2-mediated mechanisms.
Chemically Induced Model Limitations
Any chemically induced IBD model has several variables to consider. The same protocols should always be used. To make sure studies are reproducible, you should closely monitor chemical batch, strain, gender, animal source, chemical supplier, dosing level, frequency, and duration. This type of model can also be severe, with TNBS showing increased severity over DSS models.
Spontaneous Mutation IBD Mouse Models
This group of models develop IBD due to spontaneously occurring mutations. This is not due to transgenic overexpression or gene suppression. Commonly used models include SAMP1/Yit and C3H/HeJBir (C3Bir), with both representing chronic intestinal inflammation.
Samp1/YitFc Colitis
Mice of the SAMP1/Yit substrain are used to model clinical CD based on disease location, histological features, and treatment response. Spontaneous inflammation of the terminal ileum (the primary location of clinical CD lesions) is seen, with immense infiltration of activated CD4+ and CD8α+TCRαβ+ T cells into the lamina propria.
IBD development and disease progression follow a well-defined time course in this substrain, providing a good model to assess new therapies for different disease stages. This substrain also responds to anti-TNF treatment in a similar way to clinical patients, showing usefulness and translatability for assessing similar new agents.
C3H/HeJBir (C3Bir) Colitis
Mice of the C3H/HeJBir substrain develop a predominantly right-sided colitis characterized by acute and chronic inflammation. The substrain provides a valuable model for studies on IBD immuno-pathogenesis, revealing increased B and T cell reactivity to antigens in the enteric bacterial flora and high levels of serum IgG antibodies against select bacterial antigens.
Spontaneous Mutation Model Limitations
One major limitation is the time needed for full disease penetrance. For example, for the SAMP1/Yit substrain, 100% penetrance takes approximately 30 weeks. This can result in long and costly study timelines.
Adoptive T Cell Transfer IBD Mouse Model
The adoptive T cell transfer model induces chronic small bowel and colonic inflammation, which resembles some key aspects of human IBD. CD4+CD45RBhi T cells (which are CD25-) are sorted and isolated from donor BALB/c splenocytes. Cell transfer to a syngeneic immunodeficient SCID or RAG2-/- recipient generates a model with primary inflammation in the colon.
This inflammation is attributed to a lack of Treg cells in the naïve T cell population. The adoptive T cell transfer model is therefore used to study the role of pathogenic T cells in mucosal inflammation. A great deal of Tregs and other T cell population research has relied on this model.
Adoptive T Cell Transfer Model Limitations
As this model uses immunodeficient mice, a full overview of colitis development is not possible.
Genetically Engineered Mouse Models of IBD
Genetically engineered mouse models of IBD spontaneously develop colitis and/or ileitis. Many of the models harbor susceptibility genes identified in human IBD. The most well-known model is the constitutive knockout IL-10-/- mouse, still used in studies 25 years after it was developed.
IL-10-/- Knockout Mouse
Spontaneous colitis develops in the IL-10 KO mouse, as its Treg cells cannot produce IL-10. Specifically, colonic inflammation is seen, characterized by an inflammatory infiltrate of lymphocytes, macrophages, and neutrophils. Disease severity can be modulated by the background strain used, with more severe disease observed in BALB/c versus C57BL/6 mice.
Colitis develops over 24 weeks and results in models useful for studying different immune mechanisms of IBD. Tregs can also be studied because they do keep some function (with the obvious caveat that IL-10 is removed).
IL-10-/- Knockout Mouse Limitations
One limitation is that substantial variability in colitis development can occur between facilities because the model is highly dependent on microbiome differences. It also has a long disease development time. Colitis onset can be accelerated and synchronized via piroxicam, but this needs to be carefully validated for dose, formulation, and mouse age and microbial status.
No weight loss is observed in this model. You can monitor other clinical indicators, but some of them are only seen with severe disease. A more efficient, sensitive, and validated method to monitor inflammation levels is to measure lipocalin 2 in the feces.
Microbiome-Induced IBD Mouse Models
Intestinal bacteria are implicated in immune-mediated intestinal inflammation in IBD models. For example, in IL-10 knockout mice (and other IBD models), stimulation of the mucosal immune system by the microbiome is critical for colitis development.
We still do not know the precise pathogenic role of the microbiome in human IBD, partly because the microbiome is analyzed after IBD develops. However, the importance of intestinal symbiotic bacteria and flora disorders in IBD pathogenesis has been widely investigated.
Evidence indicates that antibiotics result in improvements and attenuated mucosal inflammation in IBD patients. In addition, a novel fecal microbiota transplant (FMT) strategy has recently been under preclinical and clinical investigation in mice. FMT aims to reverse disease by adjusting intestinal flora disturbances and restoring the homeostasis between the host and intestinal microorganisms.
Introducing select bacteria (e.g., Helicobacter hepaticus, alone or with other bacteria) or microbiota cocktails to germ-free IL-10-/- animals results in accelerated colitis development, in 2–4 weeks. Microbiome-induced mouse models therefore provide a useful tool for evaluating antimicrobial therapeutics and studying microbes or microbial communities that can be beneficial or pathogenic.
Microbiome-Induced Model Limitations
Limitations include variable colonization and that the germ-free conditions needed before induction are hard to control. Germ-free mice can also have issues with food digestion, which needs to be taken into account during model induction.
Emerging pathways and models
A protective role for IBD has been reported for the T cell dysfunction / exhaustion, in particular CD39+ CD8+ T cells. As a result, co-inhibitory markers associated with T cell exhaustion may present both markers of disease status as well as for therapeutic candidates. In clinic, CD39-based gene signatures are associated with flare-free survival.
Another recent advance in the field is the identification of β-hexosaminidase (β-hex), which is highly conserved, an enzyme that is produced by commensals from the Bacteroidetes phylum. This antigen drives CD4+CD8αα+ intraepithelial lymphocytes (CD4IELs) differentiation in the gut. Both CD4IELs and regulatory T cells recognize β-hex peptides. When transferred into mice, β-hex–specific CD4 T cells committed into CD4IELs that localized to the small intestine, and suppressed inflammation in a regulatory T cell–independent manner.
Conclusion
IBD mouse models are an invaluable tool for preclinical research and drug development. While no single model is fully clinically relevant, different models are used to study different and specific disease mechanisms. By understanding the differences, pros, and cons across the available range of models, researchers can select the most appropriate one per study.
