Homologous Recombination

Gene targeting technology for humanized, Knockin & conditional Knockout models

Homologous recombination is a genetic recombination in which nucleotide sequences are exchanged between two similar or identical DNA molecules.

It is commonly applied in mouse genetics. Also called gene targeting, this method can be used to specifically replace a copy of a gene by integrating a gene different from that of origin.

Of therapeutic interest are the creation of Knockout mice using embryonic stem cells to deliver artificial genetic material in order to repress the target mouse gene.

Furthermore, this technology enables a robust and efficient Knockin of a particular mutation, reporter, or human gene sequence into endogenous loci, leading to accurate and physiologically more relevant human disease models.

Such models show significant impact on research experiments and drug development programs spanning activities from target validation to ADMET studies.

‍

Typical mouse model creations using homologous recombination

Humanized cell membrane receptors for compound testing

Cell membrane receptors such as GPCR, cytokine receptors, Ig Receptors (FcR) ligand-gated anion channels, voltage-gated sodium channels, etc. are important drug targets. Substituting the mouse protein by a human homolog enables efficacy and allows affinity studies to be run with increased predictability and accuracy.
genOway has an extensive experience in humanizing receptors (more than 100 models successfully created) to insure that the transcription, translation and expression is efficient and mimics the human situation.
Humanized mouse models for cell membrane receptors have demonstrated the value they provide to drug development programs and the "Humanization & Knockin Technology" is a key technology to achieve that goal.

Case study: Humanization of the cell membrane receptor CCR2 (also called CD192) for in vivo testing of small molecule specificity and efficacy.

Adapted from Prosser et al, DDR 2002, Targeted Replacement of Rodent CCR2 With the Human Orthologue CCR2B: A Mouse Model for In Vivo Analysis of Human Target-Selective Small Molecule MCP-1 Receptor Antagonists.

Human CCR2 is Functional

The human CCR2 gene is expressed and functional in humanized animals, substituting the inactivated murine CCR2.

Expression studies are done on mRNA (data not shown).
Functionality: Normal inflammatory response is obtained to intraperritoneal thioglycollate injection (data not shown).

Humanized CCR2 Model is a Highly Valuable Screening Tool

Activity of a human selective CCR2 antagonist is better predicted and mimicked in humanized CCR2 models than in wild type models.

The potency of SB-399721 at inhibiting JE-stimulated chemotaxis of peritoneal cells from CCR2 Knockin mice (IC50 = 820 ± 83 nM) was approximately 7-fold greater than that at peritoneal cells from wild type littermates (IC50 5,500 ± 900 nM).

‍

Humanized cell membrane molecules for in vivo evaluation of antibodies and biologics

The exponential development of monoclonal antibody and biologics has paved the way for innovative therapies. In vivo evaluation of monoclonal antibodies requires new types of animal models. Because of their high specificity, monoclonal antibodies cannot be properly evaluated on mouse proteins. Evaluation can only occur in a humanized animal model where the antibody will bind to the human antigen. Most of the time, it is required that after humanization the transduction pathway stays functional, which means that the human antigen be able to functionally substitute for the mouse homolog. Availability of a humanized model has become a compulsory step for the evaluation of monoclonal antibodies and biologics. Relying on its experience in humanized models (more than 250 models successfully created), genOway is recognized for its excellence in the development of humanized models for biologics testing.

Case study: Humanization of CTLA-4 (Cytotoxic T-lymphocyte antigen) for in vivo testing of monoclonal antibodies specificity and efficacy.

CTLA-4, a high affinity receptor for CD80 and CD86, can inhibit T-cell activation. Blocking of CTLA-4 has been shown to promote anti-tumor immunity.

Adapted from Lute et al. Blood 2005. Human CTLA-4 knock-in mice unravel the quantitative link between tumor immunity and autoimmunity induced by anti–CTLA-4 antibodies.

Human CTLA-4 Gene is Expressed

Human CTLA-4 gene is expressed in humanized animals, substituting the inactivated murine CTLA-4 gene.

Expression studies were performed by FACS on unstimulated CD4 splenocytes.

Human CTLA-4 Gene is Functional

Human CTLA-4 genes functionally substitute murine CTLA-4 genes.

Normal development of lymphoid organs in humanized CTLA-4 mice.

Humanized CTLA-4 Model is a Highly Valuable Screening Tool

Reliable discrimination of therapeutic activity between anti-CTLA-4 antobodies with identical affinity and isotope in humanized CTLA-4 mice.

A complete rejection of tumors was observed in 2 out of 9 mice as well as a delayed tumor growth (data not shown) in the antibody 3-treated group.

‍

Humanized soluble molecules (Ig, transport proteins, interleukins, and hormones)

Secreted proteins are important targets and biomarkers for biomedical research. Humanization of these soluble molecules provide very valuable research models, such as the humanization of:

• Immunoglobulin genes for the in vivo development of chimeric and fully human antibodies.
• Transport proteins like albumin to study the impact of blood transport on the compound half-life or activity.
• Interleukins and hormones for in vivo studies of NCE or antibodies targeting these soluble molecules.

Humanized models can also be used to generate models tolerant to a given human antibody isotype and avoid a mouse immune response against therapeutics antibody when tested in mice.

Case study: Generation of humanized IgG1 model for in vivo assessment of therapeutic humanized IgG1 efficacy and the toxicity.

MOUSE LINE AVAILABLE AT GENOWAY. FOR MORE INFORMATION, PLEASE CONTACT US.

The constant regions of both light and heavy chain of IgG1 immunoglobulins were humanized (red boxes).

Variable regions are of mouse origin.

Humanized IgG1 Model is Fully Functional for Antibody Development and Testing

Upon stimulation, humanized IgG1 mice are producing human IgG1 and no mouse IgG1. The expression of other isotopes is unchanged. These mice are tolerant to human IgG1.

Humoral response was assessed by ELISA after stimulation with KLH. Mouse IgG1 immunoglobulins are not expressed. Human IgG1 is detected at standard level.

The model is producing all other mouse isotopes including IgM, IgD, IgG3, Ig2a, Ig2b, IgE, IgA (data not shown).

‍

Monitoring of Biomarker Expression

Biomarkers have become key tools for physiological studies. They enable:

• Detection of an early event in the response cascade.
• Quantification of a physiological response.

Knockin models could be designed in order to provide valuable tools for the monitoring of biomarker expression. Following this approach a reporter marker will be expressed concomitantly to the biomarker. To avoid interference with the biomarker biological activities, genOway combines its IRES technology with the Humanization Knockin technology in order to achieve bicistronic expression so that the biomarker expression is not altered by the expression of the reporter.

It is then possible to efficiently and easily monitor the biomarker expression without interfering with its functional network. This approach provides more reliable and valuable information under normal physiological conditions, but more interestingly, under pathological conditions, under special diets, under stress conditions, etc.

Such models have been developed for the assessment of compound polarizing the immune response, detection of protective immune response, detection of effector functions (cytotoxic, antitumor, helper cells, repressor cells).

Case study: Monitoring of IL4 expression.

MOUSE LINE AVAILABLE AT GENOWAY. FOR MORE INFORMATION, PLEASE CONTACT US.

Adapted from Mohrs et al. Immunity 2001. Analysis of type 2 immunity in vivo with a bicistronic IL4 reporter.

IL4 induces differentiation of naive Th cells to Th2 cells polarising the immune response toward a humoral effector response. IL4 remains the canonical marker of Th2 cells.

Expression of the reporter gene does not interfere with IL4 expression.

Monitoring of IL4 gene expression is done by 3'UTR insertion of the reporter gene (bicistronic expression via IRES technology). Reporter expression is proportional to IL4 gene expression.

Reliable and Sensitive Monitoring of IL4 Expression

The reporter is only detected in IL4 expressing cells, enabling a reliable detection and monitoring of the IL4 expression.

GFP expression correlates with IL4 expression and allows monitoring of early events of an effector immunity type 2. Established by FACS analysis on activated splenocytes.
Although all culture conditions induce cell division (activation), GFP+ cells are detected only in Th2 polarizing culture conditions. In addition, GFP can be detected as early as 36h after the onset of the culture as well as prior to cell division (40-45h).

Monitoring with Unmodified IL4 Activity

IL4 biological and effector function is preserved in GFP reporter mice. GFP expression enables a reliable monitoring of IL4 responses.

FACS analysis of freshly isolated cells from different organs.

Protection against Nocardia brasiliensis is Th2 mediated. An efficient Th2 priming results in worm expulsion within 10 days. Priming of CD4 cells was analyzed in lung, mesenteric lymph node (MLN), peripheric lymph node (PLN) and splenocytes. Apparition of IL4 producing cells is consistent with the worm trafficking and demonstrates that the presence of GFP does not alter the protective Th2 immunity.

‍

Monitoring of Off-Target Effects

Xenobiotics can directly or indirectly modulate target genes' expression or functionality. Determining and quantifying these non-specific activities is an important step in most drug development programs, as in ADMET studies (i.e., SXR modulating CYPP450 3A family).

Genetically modified animal models can help in predicting and measuring non-specific effects:

• Humanization of these target genes in animal models enables a more predictable detection of non-specific activities.
• Reporters can be inserted in biomarker gene loci to measure induction or repression of biomarker expression.

Case study: Humanized CAR (constitutive androstane receptor) mice allow more predictable DMPK studies than WT mice.

Adapted from Scheer et al. J Clin Invest 2008. A novel panel of mouse models to evaluate the role of human pregnane X receptor and constitutive androstane receptor in drug response.

CAR plays a role in transcriptional regulation of drug metabolism.

Humanized CAR Model is a More Relevant Tool for DMPK Studies

CITCO is a species-specific CAR inducer activating huCAR in humanized mice.

This mimicks a human-only response in a murine environment.

Midazolam and bupropion pharmacokinetics in CITCO-treated WT and huCAR mice.

Increased clearance of the drugs in humanized mice.

‍

Labeling for Cell Trafficking

This Knockin technology targets the insertion of a reporter in any cell-specific gene enabling a tight regulation of the reporter expression and consequently, a reliable cell labeling.

Genetically modified models have been developed to:

• Label one-cell lineage using one biomarker specific to the cell lineage of interest.
• Label one subpopulation of one-cell lineage, i.e. activated cells, as defined by the expression of a specific biomarker.

Case study 1: A Sox9-eGFP Knockin model enables to follow stage-specific changes of the Sertoli-germ cell mass in developing testis.

Adapted from Nel-Themaat et al. Dev Dyn 2009. Morphometric analysis of testis cord formation in Sox9-EGFP mice.

Labelled Cells Enable Trafficking Studies

Between E11.5 and E12.5, the testis increase in thickness and the first signs of differentiation appeare.

Formation of small protrusions of the Sertoli-germ cell mass.

Distinct, non-fluorescing areas in the tissue mass.

Case study 2: Knockin mice expressing a hemagglutinin (HA)-epitope-tagged dopamine transporter (DAT) allow to study endogenous transporter trafficking.

Adapted from Rao et al. FASEB J 2012. Epitope-tagged dopamine transporter knock-in mice reveal rapid endocytic trafficking and filopodia targeting of the transporter in dopaminergic axons.

Labelled Vesicules Enable Transport Studies

Localization of DAT in axons of dopamin (DA) neurons. Region of an axonal process.

Intracellular HA-DAT vesicles displayed rapid bidirectional movement. Red arrows indicate vesicles moving in the retrograde direction; green arrows indicate vesicles moving anterograde. Time (s) is indicated.

‍

Models for translational medicine: Mimic human diseases - cell deletion using suicide genes

Use of cell ablation techniques enables the ability to delete upon request one cell type of interest from an organism. Cell depletion is a powerful research tool with two main applications:

• The mimicry of human diseases or drug side effects. Depleting cell populations in vivo can reliably mimic such physio-pathologies (hepatitis, neurodegenerative diseases, etc.).
• The improvement of xenograft efficiency. It may also be useful to partially or totally remove endogenous cells from the recipient organ / organism.

To this purpose a suicide gene will be targeted using the Humanization & Knockin Technology so that it will only be expressed in the cell type to be deleted. At this stage the suicide gene has no effect on the cell. When a drug specific to the suicide gene used is delivered to the animals, the suicide gene will metabolize the drug creating toxic metabolites resulting in the death of the cells expressing the suicide gene (but not the other cells which do not respond to the drug).

Case study: Diphteria toxin-induced cell knockout.

Adapted from Wu et al. Development 2006. Motoneurons and oligodendrocytes are sequentially generated from neural stem cells but do not appear to share common lineage-restricted progenitors in vivo.

Target Cell Population is Killed

Mouse embryo expressing the Diphteria toxin (DT) in Olig1+ cells (Olig1-DTA) show absence of motoneurons.

I, J, M, N) Brain section from a wild-type mouse, showing normal presence of motoneurons.

K, L, O, P) Motoneurons from transgenic mouse expressing DT (Olig1-DTA) are largely missing.

‍

Models for translational medicine: Mimic human diseases - tissue- and/or time-specific expression of disease causal genes

Some disease models require the over-expression of disease causal genes in specific cell populations (inflammation, neurodegenerative diseases) and / or within well-defined time window. These mouse models enable scientists to study the onset of the pathology and to evaluate compound efficacy.

To this purpose mouse models could be generated using the Humanization & Knockin Technology using either permissive loci ("safe harbour") or target gene loci.

Case study: Inducible disease causal gene overexpression results in leukemia and mouse death.

Adapted from Carofino et al. Dis Model Mech 2013. A mouse model for inducible overexpression of Prdm14 results in rapid-onset and highly penetrant T-cell acute lymphoblastic leukemia (T-ALL).

Model Mimics Human Disease

Mice overexpressing the Prdm14 gene mimic efficiently the human disease (acute lymphoblastic leukemia) reproducing the human phenotype.

After induction, mice had enlarged thymi and spleens (arrows) as well as kidneys, livers and lymph nodes (data not shown).

After induction, mice developed and succumbed to acute lymphoblastic leukemia very rapidly.

‍

Models for translational medicine: Mimic human diseases - humanization with human mutant genes

Genomics studies provide statistical correlation between the presence of mutations and the pathology (onset, development, lethality, etc.) which does not present evidence of causality. The generation of mouse models displaying a substitution of the wild-type mouse gene by a mutant version (mouse or human) enables the study of mutation roles in the disease and, consequently, the mechanisms of disease etiology and progression.

Case study: Inducing a human disease in mice - Paget's disease-like disorder.

Adapted from Daroszewska et al. Hum Mol Genet 2011. A point mutation in the ubiquitin-associated domain of SQSMT1 is sufficient to cause a Paget's disease-like disorder in mice.

Model Recapitulates Human Disease

Mice presenting the human point mutation P394L develop a Paget's disease-like disorder.

MicroCT analysis of long bones from P394L mutant and wild-type (WT) mice. Mutant mice are showing multiple focal osteolytic lesions penetrating the cortex (arrows).

Axial microCT image of lumbar vertebra 5.

(N) WT mouse showing normal morphology and trabecular structure.

(O) P394L+/+ mouse showing an osteosclerotic lesion replacing most of the vertebral body.

‍

Tissue-specific transgene expression

Specific pattern and level of expression of a transgene (Cre or Flp recombinase, reporter genes, safe harbour, target gene, etc.) are key parameters of relevant transgenic models.

Knocking in the transgene, which means targeting its insertion into one specific and predetermined location of the genome, enables the optimal regulation of its expression. The complete endogenous mouse promoter (all regulatory elements being present, included enhancers and repressors located kilo bases away) will control the transgene regulation.

This Knockin technology has become the gold standard technology for tissue-specific expression since it enables to determine the transgene expression pattern and its level of expression. Many types of models have been created including:

• Tissue-specific expression of Cre and Flp recombinase (⤏ see also site-specific recombination)
• Monitoring gene expression using a reporter gene like fluorescent proteins (⤏ see also reporter mice, or cell lines)
• Inserting transgene in a safe harbour like Rosa26, Hprt locus (⤏ see also Quick Knockin mice, or cell lines)
• Humanizing the gene of interest (⤏ see also humanized Knockin mouse models)

Case study 1: Repopulation kinetics of intestinal stem cells in tamoxifen-inducible Cre mice.

Adapted from Sangiorgi et al. Nat Genet 2008. Bmi1 is expressed in vivo in intestinal stem cells.

Expression Obtained in the Targeted Subpopulation

Bmi1+ expressing Cre+ cells repopulate crypts of the small intestines.

Cre function is detected (lacZ staining) in the Bmi1+ cell lineage.

Repopulation starts day 2 (a) after tomoxifen injection. First fully labelled crypts are visible at day 17 (i).

Case study 2: Neurons and oligodendrocytes Cre recombinase-specific expression.

Adapted from Battiste et al. Development 2007. Ascl1 defines sequentially generated lineage-restricted neuronal and oligodendrocyte precursor cells in the spinal cord.

Expression Obtained in the Targeted Cell Lineage

Ascl1 expressing Cre+ cells give rise to neurons and oligodendrocytes.

Cre function is detected (lacZ staining) in developing sympathetic neurons.

Expression is just beginning at E10.5 but is clearly evident by E11.5.

Case study 3: Salivary gland Cre recombinase-specific expression.

Adapted from Bullard et al. Dev Biol 2008. Ascl3 expression marks a progenitor population of both acinar and ductal cells in mouse salivary glands.

Expression Obtained in Biomarker Expressing Cells

Cre recombinase expression recapitulates endogenous Ascl3 expression.

Arrowheads indicate duct cells of the submandibular gland in which Ascl3 expressing Cre+ cells are labeled red using antibodies to Cre recombinase.

Ac, acinar cells; Du, duct cells.

Target validation & Knockout models: Constitutive Knockout models with monitoring of the gene expression pattern

Using this Knockin technology, an inserted reporter gene provides the expression pattern of the gene of interest. Expression under normal physiological conditions is valuable information, but more interestingly, the model enables a reliable and simple monitoring of the expression under pathological conditions, under special diets or stresses, etc.

Case study: Generation of a MCT1 Knockout / LacZ Knockin model for validating MCT1 as a target for obesity.

Adapted from Lengacher et al, PLoS One 2013, Resistance to diet-induced obesity and associated metabolic perturbations in haploinsufficient monocarboxylate transporter 1 mice.

MCT1^+/- mice exhibit resistance to diet-induced obesity.

MCT1 ^+/- mice are almost undistinguishable from MCT1^+/+ mice under normal chow diet.

When fed a high fat diet, MCT1^+/+ mice gained considerably weight and become obese.

In contrast MCT1^+/- fed the same diet remained lean.

Efficient Monitoring of Target Gene Expression

LacZ Knockin allows monitoring of gene expression.

LacZ is expressed under the control of the endogenous MCT1 promoter and its expression provides insights on the expression pattern of MCT1.

Target validation & Knockout models: Conditional Knockout model with monitoring of the gene ablation efficacy

Conditional Knockout models, either in a tissue-specific manner or in an inducible manner, are now routinely used models to study gene functions and validate targets. The conditional deletion effectiveness (in how many cells of the tissue of interest, both alleles are deleted) is a key parameter for the quality and reliability of the model. To monitor this efficacy and validate the reliability of the model, scientists are now using this technology to target a reporter gene in the locus of interest which is switched on when the deletion occurs. Measuring the reporter gene expression quantifies the level of deletion obtained in a tissue or at the cell level indicating which cell is knocked out and which one is not. By providing a means to validate the conditional Knockout efficacy, this technology enables the creation of reliable and valuable conditional Knockout models for target validation.

Case study: Tracking of Knockout cells in vivo when no good antibody is available.

Adapted from Schnütgen et al, Nat Biotechnol 2003, A directional strategy for monitoring Cre-mediated recombination at the cellular level in the mouse.

Efficient Monitoring of the Inactivated Gene

Monitoring a Cre-mediated deletion of a gene of interest (Retinoic acid Receptor gamma).

Before induction, no staining was detected (top).

After induction, reporter activity was detected in Rar gamma-expressing cells (bottom).