What Is a Fully Human Antibody Mouse Model, and How Does It Work?


What Is a Fully Human Antibody Mouse Model, and How Does It Work?
A fully human antibody mouse model carries human immunoglobulin (Ig) gene segments that enable its B cells to generate antibodies with human variable regions after immunization. Depending on the platform design, these antibodies may also contain human constant regions. In models that retain mouse constant regions, the selected human variable regions can be recombinantly expressed with human constant regions to generate fully human antibody candidates.
When therapeutic antibody discovery starts with a conventional mouse antibody, the selected antibody often requires humanization. This step can alter binding affinity and require further engineering. Fully human antibody mouse models bypass the conventional CDR-grafting humanization step and, alongside phage and yeast display, provide established routes for discovering human antibody candidates. The resulting candidates still require evaluation and, where necessary, further optimization for therapeutic development.
What Is a Fully Human Antibody Mouse Model?
In a fully human antibody mouse model, also called a human Ig mouse in this article, the mouse's own B cells rearrange human immunoglobulin (Ig) gene segments as they develop. This process generates a diverse set of antibody variable regions. In a conventional antibody, the heavy- and light-chain variable regions together form the antigen-binding site. Some fully human antibody mice retain mouse constant regions. To obtain a fully human antibody from these mice, researchers express the selected human variable regions with human constant regions. A fully human antibody can still elicit an anti-drug antibody response in patients.
The phrase "humanized antibody mouse model" can be ambiguous. In mouse antibody humanization, researchers first select an antibody from a conventional mouse. They then graft its complementarity-determining regions (CDRs) onto human variable-region frameworks and may retain selected mouse residues to preserve binding. The resulting molecule is a humanized antibody.

How Fully Human Antibody Mouse Models Are Made and How They Produce Antibodies
The fully human antibody mouse model is engineered once. Its offspring inherit the human gene segments and use them during B-cell development and the response to immunization.
- Human Ig genes are introduced into the germline. Depending on the platform, human gene segments replace their mouse counterparts at the native loci, be introduced as transgenes at other genomic locations, or be carried on artificial chromosomes. Endogenous mouse Ig genes may also be modified or inactivated to limit the production of murine antibody chains.
- The mouse's B cells generate and refine the repertoire. During B-cell development, V, D and J gene segments rearrange to form heavy-chain variable regions, while V and J segments rearrange to form light-chain variable regions. After immunization, somatic hypermutation introduces additional changes into the antibody variable regions. Selection within germinal centers can enrich B cells with higher-affinity receptors, a process known as affinity maturation. Class-switch recombination changes the heavy-chain constant region without changing the antigen-binding variable regions, thereby altering the antibody isotope and potentially its effector functions. In models retaining mouse constant regions, this switching occurs among murine isotypes.
- Antigen-specific antibodies are recovered and sequenced. Researchers monitor serum antibody titers and recover B cells from tissues such as the spleen or lymph nodes. They can generate hybridomas and screen the antibodies they produce, or screen individual B cells directly. Paired heavy- and light-chain variable-region sequences from selected clones can then be used for recombinant expression.

Types of Human Immunoglobulin Mice: How Platform Designs Differ
Two design choices shape the antibody repertoire and the format recovered from the mouse.
The first is how much of the antibody is human. Some platforms encode both human variable and constant regions. Others retain mouse constant regions, which can help preserve compatibility with the mouse B-cell development and signaling environment. The resulting antibodies contain human variable regions but murine constant regions in vivo, and can be reformatted with human constant regions for therapeutic development. To produce a fully human IgG from a model that retains mouse constant regions, researchers express the selected human variable regions with human constant regions.
The second is how the human genes are introduced. Some early models used randomly integrated minilocus transgenes containing a subset of human gene segments [2]. Other platforms replace variable-region genes at the native mouse loci or carry large human Ig loci on an artificial chromosome. In 2014 studies, researchers replaced mouse variable-region genes with their human counterparts at the native loci while retaining mouse constant regions. In the assays tested, these mice mounted antibody responses comparable to those of wild-type controls [3-4].
HUGO-Mab™, the fully human monoclonal antibody mouse model in Cyagen's HUGO-Ab™ family, illustrates how these choices work together.
| Design axis | Option | What you gain | What to watch |
|---|---|---|---|
| How much of the antibody is human | Human variable regions, mouse constant regions | Can support normal B-cell development | A constant-region swap during expression |
| Human variable and constant regions | Fully human antibodies produced in vivo | Response kinetics can vary by strain | |
| How the human genes get in | Randomly integrated minilocus transgene | Simpler to build | Only the included gene segments are available |
| In situ replacement or artificial chromosome transfer | Can accommodate large repertoires | Regulatory context differs between designs |
How the HUGO-Mab™ Mouse Is Built
Cyagen uses its TurboKnockout™ embryonic stem cell targeting technology to replace the heavy-chain, κ light-chain and λ light-chain variable-region loci in situ with their human counterparts while retaining mouse constant regions. Cyagen lists C57BL/6NCya, BALB/cAnCya and SJL backgrounds and recommends starting immunization at 6 to 8 weeks of age.
Cyagen reports serum IgG subclass profiles comparable to wild-type controls. In Cyagen's anti-PD-L1 case study, single-B-cell screening yielded 1,466 paired heavy- and light-chain variable-region sequences from one immunized HUGO-Mab™ mouse and 1,248 from one immunized wild-type C57BL/6N mouse. Of these, 413 and 142 pairs, respectively, had distinct CDR amino acid sequences. In a PD-L1 protein-binding ELISA, three HUGO-Mab™ clones and two wild-type clones had lower EC₅₀ values than atezolizumab. In a separate epitope-binning assay using biolayer interferometry, clone SAI-427-4 showed no detectable competition with atezolizumab for PD-L1 binding.

Antibodies produced by HUGO-Mab™ mice contain human heavy- and light-chain variable regions paired with mouse constant regions. To generate fully human antibody candidates, researchers can recombinantly express the naturally paired human variable regions with selected human heavy- and light-chain constant regions. The choice of human IgG subclass helps determine Fc function. The resulting candidates will go through testing for binding, biological activity and developability.
When to Use a Fully Human Antibody Mouse Model for Discovery
These models are useful when a program calls for naturally paired human heavy- and light-chain variable regions that have undergone affinity maturation in vivo. For membrane protein targets, the antigen format also matters. Immunization with cells displaying the target can help preserve extracellular epitopes in their membrane context.
Common light-chain models can also supply binding arms for bispecific antibodies. For example, Cyagen's HUGO-Light™ mouse pairs diverse human heavy-chain variable regions with a fixed human κ light-chain variable region. This allows researchers to select binding arms for two different targets that share the same light chain, simplifying light-chain pairing during bispecific antibody assembly.A common light chain can reduce light-chain mispairing during bispecific antibody assembly, but correct heavy-chain heterodimerization remains a separate engineering consideration. Strategies such as Knobs-into-Holes (KiH) may be used alongside common light-chain designs.

Immune tolerance can limit antibody responses against human antigens that share highly conserved epitopes with their mouse counterparts. In such cases, knockout of the endogenous mouse ortholog may help reduce this tolerance barrier. Cyagen's HUGO-Ab-eKO™ platform combines target-knockout mouse engineering with human antibody discovery models to support antibody generation against challenging, highly conserved targets. The suitability of this approach depends on the target's biological role and the specific immunization strategy.. Antigen design can focus the response, but the epitopes recognized by the recovered antibodies still depend on which B cells respond.
A fully human antibody mouse model allows researchers to recover human variable regions that have undergone selection in vivo. The model's design determines whether the antibodies produced in the mouse already contain human constant regions or require recombinant expression in the intended human IgG format. For each candidate, binding alone is not sufficient to establish therapeutic potential. Functional activity, specificity, and developability must be evaluated in the final antibody format intended for development.
👉 [Explore Cyagen's Fully Human Antibody Discovery Platform]
👉 [Discuss Your Antibody Discovery Project with Our Scientists]
Reference
[1] Mallbris L, Davies J, Glasebrook A, Tang Y, Glaesner W, Nickoloff BJ. Molecular Insights into Fully Human and Humanized Monoclonal Antibodies: What are the Differences and Should Dermatologists Care? J Clin Aesthet Dermatol. 2016 Jul;9(7):13-5. Epub 2016 Jul 1. PMID: 27672407; PMCID: PMC5022998.
[2] Taylor LD, Carmack CE, Schramm SR, Mashayekh R, Higgins KM, Kuo CC, Woodhouse C, Kay RM, Lonberg N. A transgenic mouse that expresses a diversity of human sequence heavy and light chain immunoglobulins. Nucleic Acids Res. 1992 Dec 11;20(23):6287-95. doi: 10.1093/nar/20.23.6287. PMID: 1475190; PMCID: PMC334518.
[3] Murphy AJ, Macdonald LE, Stevens S, Karow M, Dore AT, Pobursky K, Huang TT, Poueymirou WT, Esau L, Meola M, Mikulka W, Krueger P, Fairhurst J, Valenzuela DM, Papadopoulos N, Yancopoulos GD. Mice with megabase humanization of their immunoglobulin genes generate antibodies as efficiently as normal mice. Proc Natl Acad Sci U S A. 2014 Apr 8;111(14):5153-8. doi: 10.1073/pnas.1324022111. Epub 2014 Mar 25. PMID: 24706856; PMCID: PMC3986188.
[4] Macdonald LE, Karow M, Stevens S, Auerbach W, Poueymirou WT, Yasenchak J, Frendewey D, Valenzuela DM, Giallourakis CC, Alt FW, Yancopoulos GD, Murphy AJ. Precise and in situ genetic humanization of 6 Mb of mouse immunoglobulin genes. Proc Natl Acad Sci U S A. 2014 Apr 8;111(14):5147-52. doi: 10.1073/pnas.1323896111. Epub 2014 Mar 25. PMID: 24706858; PMCID: PMC3986150.



