7.1.1 An Evolving History of Mouse Strains

The biggest limitation of human immune cell engraftment has been the lack of mouse strains that could support and maintain human immune cell growth. The nude mouse was first discovered and characterized in the 1960s and found absent of thymus and T lymphocytes (Flanagan 1966; Pantelouris 1968). The nude mice, as the first immunocompromised strain defective in adaptive immune responses, have been widely used in human tumor xenograft transplantation. However, the intact innate immunity in these nude mice rejects engraftment of nontumor tissues such as human bone marrow (BM) suspensions (Ganick et al. 1980).

The first breakthrough came with the discovery of immunocompromised mice harboring the Prkdc^scid mutation. In 1983, Bosma et al. descried a strain that carries a spontaneous, autosomal‐recessive mutation in the protein kinase, DNA‐activated, catalytic polypeptide (Prkdc) gene. This strain lacks both functional T and B lymphocytes, mimicking the clinical phenotypes of human severe combined immunodeficiency (SCID) and therefore subsequently named as SCID mice (Bosma et al. 1983). PRKDC is essential for resolving DNA double‐strand breaks during variable, diversity, and joining [V(D)J] recombination during the development of T and B cells (Lieber et al. 1988; Blunt et al. 1996; Taccioli et al. 1998). Successful engraftment of human peripheral blood mononuclear cells (PBMCs) (Mosier et al. 1988), BM‐derived HSPCs (Lapidot et al. 1992), and fetal tissues (McCune et al. 1988) have been reported. However, SCID mice still maintain a functional innate immune system; NK cells are present and could result in rejection of implanted cells.

At around the same time, the nonobese diabetic (NOD) mice was discovered by Makino et al. (Makino et al. 1980). The NOD strain was characterized to have impaired innate immunity associated with NK, macrophage, and dendritic cell malfunctions (Kataoka et al. 1983; Kikutani and Makino 1992; Anderson and Bluestone 2005). Eventually, the SCID mutation was backcrossed onto the NOD strain background. This resulted in the creation of the NOD SCID mice by Shultz et al. (Shultz et al. 1995). The NOD SCID strain demonstrated deficiencies in both the innate and adaptive immunity (Shultz et al. 1995), and has served since then as a preferred model by many investigators for human immune cell engraftment.

Nevertheless, the NOD SCID strain still has limitations. The mice were prone to developing thymomas and demonstrated “leakiness” – the eventual production of B and/or T cells over time due to incomplete V(D)J recombination defects (Shultz et al. 1995; Prochazka et al. 1992; Bosma 1992). There are residual NK activities in this NOD SCID strain, thus compromising engraftment efficiency (Yoshino et al. 2000). Interleukin‐2 receptor γ chain (IL‐2Rγ) gene encodes a cytokine receptor subunit constituting the receptor complexes for IL‐2, IL‐4, IL‐7, IL‐9, IL‐15, and IL‐21, and is critical for innate and adaptive lymphoid development (Sugamura et al. 1996). IL‐2Rγ‐deficient mice demonstrate markedly reduced NK‐, T‐, and B‐cell numbers and functions (Cao et al. 1995; DiSanto et al. 1995). Crossing IL‐2Rγnull mice with the NOD SCID mice results in the NOD SCID IL‐2Rγnull strain that are severely immunodeficient due to the lack of mature T, B, and NK cells, impaired cellular functions from the myeloid lineage, and prevention of cytokine signaling through multiple receptors (Ito et al. 2002; Shultz et al. 2005). These NOD SCID IL‐2Rγnull mice (e.g. NSG with a complete null mutation of IL‐2Rγ, NOG with a truncated intracellular signaling domain of IL‐2Rγ (Nagatani et al. 2019)) demonstrate improved engraftments of both tumoral and nontumoral human tissues.

The recombination activating gene 1 (Rag1) and gene 2 (Rag2) play important functions in the initiation of V(D)J rearrangement (Schatz and Ji 2011). Knock‐out of Rag1 or Rag 2 resulted in defective primary lymphoid organs that do not contain mature T and B lymphocytes (Mombaerts et al. 1992; Shinkai et al. 1992). By crossing IL‐2Rγnull mice with strains carrying the Rag1 or Rag2 deficiency, the NRG and BRG mice have been established (Pearson et al. 2008a; Baenziger et al. 2006). The NRG and BRG mice have also been used for human immune cell engraftment and are more resistant to irradiation and genotoxic drugs than mice with a SCID mutation in the DNA repair enzyme Prkdc.

Commonly used immunocompromised mouse strains for I‐O research are summarized in Table 7.1.

7.1.2 Emerging Strain Variants

As the demand for humanized mouse models increase, newer immune‐deficient strains have been established for specific experimental needs. Strain variants engineered based on the Prkdc^scid IL‐2Rγnull or Rag1/2null IL‐2Rγnull mutations are summarized in Table 7.2.

Xenogeneic Graft‐Versus‐Host Disease (xGvHD) usually develops a few weeks after human PBMC engraftment. This phenotype is attributed to the mismatch of major histocompatibility complex (MHC) between human T cells and mouse cells (Sanmamed et al. 2016). As a result, engrafted mature human T cells recognize murine xeno‐antigens presented by murine MHC, target mouse organs and tissues, and ultimately kill the animals (King et al. 2009). To address this issue, mouse MHC class I and/or II knockout strains have been established. The resulting strains demonstrated an attenuated xGVHD development post human PBMC transplantation (Covassin et al. 2013; Brehm et al. 2019).

In addition, mice engineered to express human HLA genes have also been established. An absence of human leukocyte antigen (HLA) molecules resulted in human pre‐T cells being educated and selected on mouse thymus and MHC. Human HLA class I or class II transgenic mice demonstrated high engraftment of HLA class‐restricted immune cells after CD34+ humanization, supporting studies on human T‐cell repertoire and responses (Shultz et al. 2010; Covassin et al. 2011).

Besides T cells, multiple immune cell populations, such as NK cells and monocytes, also play crucial functions during tumor progression and are now being actively explored for immunotherapies. The poor survival of NK cells and the myeloid compartment has long been a limitation of humanized mouse models due to the lack of key human hematopoietic growth factors and cytokines. To solve this problem, newer immunocompromised strains harboring knockin of human cytokine genes have been developed.

For example, the NSG‐IL15 and hIL‐15 NOG strains with transgenic human IL‐15 expression, demonstrate enhanced NK‐cell development and survival in mouse blood, spleen, and BM (Brehm et al. 2018; Katano et al. 2017). Mice with knock‐in of human IL‐15 and SIRPα (SRG‐15) have also been generated based on the BALB/c Rag2^−/− IL2rgc^−/− strain (Herndler‐Brandstetter et al. 2017). Signal‐regulatory protein alpha (SIRPα) binds to human CD47 and produces anti‐phagocytic signals which inhibit murine macrophages from engulfing engrafted human cells (Takenaka et al. 2007). These SRG‐15 mice demonstrated improved development and functional maturation of circulating and tissue resident human NK and CD8+ T cells and promoted the development of tissue‐resident innate lymphoid cell (ILC) subsets (Herndler‐Brandstetter et al. 2017).

The NOG‐EXL strain is engineered to express both human granulocyte/macrophage colony‐stimulating factor (GM‐CSF) and IL‐3, and it supports increased reconstitution and differentiation of the human myeloid lineage compared to its parental NOG strain (Fukuchi et al. 1998; Ito et al. 2013). Similarly, the triple transgenic NSG‐SGM3 mice expressing human GM‐CSF, IL‐3 and Kit ligand (KITLG; also known as stem cell factor, SCF) support improved reconstitution of myeloid lineages and regulatory T‐cell populations compared to its parental NSG strain (Billerbeck et al. 2011). These NSG‐SGM3 mice were demonstrated with improved engraftment of diverse hematopoietic lineages and primary AML samples (Billerbeck et al. 2011; Wunderlich et al. 2010). Another model, the MISTRG mice, which express human GM‐CSF, IL‐3, SIRPα, macrophage colony‐stimulating factor (M‐CSF) and thrombopoietin (TPO), was designed to support a greater level of human immune cell reconstitution, particularly in the myeloid compartment (Rongvaux et al. 2014). However, the MISTRG strain develops severe anemia due to poor erythropoiesis and results in short lifespan and was eventually discontinued commercially (Rongvaux et al. 2014). Recently, a c‐kit mutation, which suppresses mouse stem cell self‐renewal and differentiation, has been introduced into NSG mice and the resulting strains, NBSGW and NSGW41, demonstrate multilineage human hematopoietic reconstitution without preconditioning irradiation (Cosgun et al. 2014; McIntosh et al. 2015). Hanazawa’s group generated human IL‐6‐expressing NOG mice (NOG‐hIL‐6 transgenic mice) allowing the spontaneous development and maintenance of monocytes/macrophages by the engraftment of HSPCs. The immunosuppressive myeloid cells detected in NOG‐hIL‐6 transgenic mice bearing tumors were CD163+ TAM like cells, not HLA‐DR^lo/− MDSC like cells, thus only partly enabled reconstitution of the tumor microenvironment () (Hanazawa et al. 2018).

7.2.1 Hu‐PBMC Model

The Hu‐PBMC model, also known as human peripheral blood leukocyte (Hu‐PBL) model, provides a simple and cost‐effective approach of humanization. Reconstitution of immunocompromised mice with human PBMCs allows the study of HIS, particularly the CD8+ T‐cell population, and its responses to cancer in an in vivo model.

PBMCs originate from hematopoietic stem cells (HSCs) and are constituted by cells from both the lymphoid and myeloid lineages (Autissier et al. 2010; Kleiveland 2015). About 70–90% of PBMCs are lymphocytes, with the most significant portion being CD3+ T cells (45–70%), followed by CD19+ B cells (5–15%), and NK cells (5–10%). T and B cells in PBMCs isolated from healthy donors largely exist in a naïve state and once activated, they will induce cell‐mediated and humoral immune responses, respectively. Monocytes account for 10–30% of cells in PBMC. They circulate in the blood and continually migrate into peripheral tissues. Upon stimulation, monocytes differentiate into dendritic cells and macrophages, which could mediate both the innate and adaptive immune responses. Dendritic cells (1–2%) and HSCs (0.1–0.2%) are also found at low frequencies in PBMCs.

Obtaining PBMC samples is relatively straightforward. With appropriate Institutional Review Board (IRB) approval, peripheral whole blood can be collected by venipuncture from volunteers following informed consent and local regulatory requirements. From healthy blood, PBMC yield ranges between 0.5 and 2 × 10⁶ cells per mL blood (Bittersohl and Steimer 2016). The most common method to isolate PBMCs from whole peripheral blood is density gradient centrifugation (Ulmer et al. 1984). This process separates different cell populations according to their varying densities throughout a density gradient medium. After centrifugation, the PBMC fraction appears as a thin white layer at the interface between the plasma and the medium. Maintaining proper temperatures (experimental procedures, reagent and sample storage, etc.) and reducing processing time could help improve PBMC viability and recovery and ensure reliability and consistency of downstream applications.

Isolated PBMCs can be engrafted via intravenous (IV), intraperitoneal (IP), or subcutaneous (SC) co‐injection with mixed human tumor cells. The intrasplenic route has also been documented (Steyaert et al. 2007) but is more time consuming and requires surgical proficiency. Detailed methods for human PBMC and HSPC engraftment have been previously reviewed (Pearson et al. 2008b). Despite heterogenous immune cell populations found in PBMCs prior to engraftment, T cells are the primary subpopulation that resides and remains functional in the recipient mouse. Human T cells can be maintained for up to four to six weeks, with an activated memory and effector phenotypes (Pearson et al. 2008b; King et al. 2008). B cells and most innate immune cells, however, tend to die off within the first few days of engraftment, likely due to dominant expansion of T cells and lack of appropriate human cytokines required for survival (Curran et al. 2020).

Commercially available immunocompromised strains such as NOD SCID and NOD/SCID/IL‐2Rγ knockout (e.g. NSG, NOG) have been proved to be efficient at engrafting mature human T cells from PBMCs. The availability of healthy PBMC donors and relatively simple and cost‐effective cell isolation procedures makes the Hu‐PBMC model a powerful tool for I‐O studies. Robust and reproducible T‐cell responses, combined with wide distribution of T cells in multiple organs, allow effective preclinical evaluation of immunomodulatory therapeutics in vivo. A single donor alone would provide enough PBMCs for an in vivo study with dozens of mice, minimizing inter‐ and intra‐group variations resulted from different donors (Li et al. 2019). Moreover, rapid humanization process prepares readily usable humanized mice within days of engraftment.

The major drawback of the Hu‐PBMC model is development of xGvHD, usually at a few weeks post human PBMC engraftment. The induction of xGvHD is due to MHC mismatch between human T cells and mouse cells (Sanmamed et al. 2016). The onset and severity of xGvHD correlate with the expansion of activated human T cells and attacking mouse tissues in the murine host (Ehx et al. 2018). As a result, the experimental window of the Hu‐PBMC model is usually limited to four to six weeks (De La Rochere et al. 2018). To solve this issue, researchers have established newer strains of mice deficient in MHC class I and/or II and demonstrated delayed xGvHD onset and increased survival (Covassin et al. 2013; Brehm et al. 2019). Preconditioning irradiation of NSG mice before injecting PBMCs is not required as mature human leukocytes are injected. However, it has been reported that a sublethal dose of irradiation facilitates engraftment and colonization of human PBMCs in the murine host; this leads to faster and enhanced T‐cell expansion but also accelerates xGvHD, resulting in a shorter survival time compared to nonirradiated mice receiving PBMCs (King et al. 2009; King et al. 2008).

Considering the genetic diversity among human population (e.g. HLA type and expression), donor‐to‐donor difference in engraftment kinetics, and variations in progression and severity of xenogeneic T‐cell activation, investigators are encouraged to prescreen a panel of PBMC donors and set up additional studies to understand how therapeutics work across multiple genetically diverse donors (Li et al. 2019; Verma et al. 2019).

7.2.2 Hu‐CD34+ HSPC Model

Human HSPCs have the potential to differentiate into all human hematopoietic lineages, thereby serving as a reliable source for HIS reconstitution in immunocompromised mice. CD34 is a transmembrane phosphoglycoprotein protein and a cell–cell adhesion factor (Krause et al. 1996; Lanza et al. 2001). It was first described on HSCs by Civin et al. (1984) and Tindle et al. (1985) independently, and is considered as a key surface marker for HSPCs. Although combined markers have been proposed to precisely define HSPCs, the single, less discriminating marker CD34 could still capture most cells with the long‐term repopulating activities (Hogan et al. 1997; Kawashima et al. 1996; Gao et al. 2001

Stay updated, free articles. Join our Telegram channel

Join