Humanized Mice as Unique Tools for Human-Specific Studies

With an increasing human population, medical research is pushed to progress into an era of precision therapy. Humanized mice are at the very heart of this new forefront where it is acutely required to decipher human-specific disease pathogenesis and test an array of novel therapeutics. In this review, “humanized” mice are defined as immunodeficient mouse engrafted with functional human biological systems. Over the past decade, researchers have been conscientiously making improvements on the development of humanized mice as a model to closely recapitulate disease pathogenesis and drug mechanisms in humans. Currently, literature is rife with descriptions of novel and innovative humanized mouse models that hold a significant promise to become a panacea for drug innovations to treat and control conditions such as infectious disease and cancer. This review will focus on the background of humanized mice, diseases, and human-specific therapeutics tested on this platform as well as solutions to improve humanized mice for future clinical use.


Introduction
Fundamental understandings of many biological processes that occur in humans have evolved from experimental studies on animal models, particularly non-human rodents and non-human primates (NHPs) (Hatziioannou and Evans 2012;Phillips et al. 2014). A major technical barrier in translating these discoveries to treatments is caused by differences in the biological systems between animals and humans (Greek and Rice 2012;Mestas and Hughes 2004;Shanks et al. 2009; Van der Worp et al. 2010). For example, functional Toll-like receptor 10 (TLR10) is absent in mice (Oosting et al. 2014) and cell expression marker CD28 is expressed on 100% of CD4 + and CD8 + T cells in mice but only on 80% of CD4 + and 50% CD8 + T cells in humans (Beyersdorf et al. 2015). Due to these differences, it is common that animal models are refractory to many infectious (Bäumler and Fang 2013;Carlton et al. 2008;Fauci 1988;Pain et al. 2008;Ploss et al. 2009), therapeutic (McKenzie et al. 1995Rehman et al. 2011), or immunomodulatory agents (Attarwala 2010;Tsoneva et al. 2017) that are human-specific.
To address the limitations of translating discoveries on non-human animal models to clinical applications, a platform known as "humanized mice" was engineered to simulate humans at a cellular and molecular level (Bosma et al. 1983;Pearson et al. 2008). Humanized mice generated in recent years encompass functional human immune systems with expansive capabilities (Rongvaux et al. 2014) and are unprecedented platforms used for understanding disease pathogenesis and evaluation of compounds to treat a variety of human diseases which include but are not limited to, cancer Ito et al. 2009;Miyakawa et al. 2004;Pan et al. 2017), infectious disease (Amaladoss et al. 2015;Frias-Staheli et al. 2014;Keng et al. 2015;Yajima et al. 2008), autoimmune disease (Gunawan et al. 2017;Viehmann Milam et al. 2014;Young et al. 2015;Zayoud et al. 2013), and graft-versus-host disease (GvHD) (King et al. 2008;Kirkiles-Smith et al. 2009;Tobin et al. 2013;Zhao et al. 2015).
This review covers the background of humanized mice, diseases modelled on these platforms, human-specific therapeutics tested, and suggestions for overcoming remaining challenges to improve humanized mouse models for clinical applications.

Evolving History of Humanized Mice
There has been a constant pursuit to engineer novel immunodeficient mouse models via gene deletion or backcrossing strains with mutations in essential molecular compartments such as, T cells, B cells, macrophages, natural killer (NK) cells, cytokines, TLRs, and transcription factors . The aim of introducing these mutations is to reduce murine cells and increase the engraftment of human cells and tissues to better recapitulate human immune responses (Aryee et al. 2014;Billerbeck et al. 2011;Chen et al. 2009;Rongvaux et al. 2014;Yao et al. 2016).
Tracing the roots of humanized mice, the discovery of non-human animal models xenotransplanted with cells and tissues of human origin was credited to the invention of C.B-17-Prkdc scid (CB17-scid) mice (Bosma et al. 1983). Derived from backcrossing C57BL/Ka and BALB/c, this mouse features loss of function mutation in a gene known as protein kinase, DNA-activated, catalytic polypeptide (PRKDC). In normal physiological conditions, PRKDC is essential for resolving breaks in DNA strands during variable, diversity, and joining [V(D)J] recombination for the development of T and B cells Finnie et al. 1996;Lieber et al. 1988;Taccioli et al. 1998). Non-functional PRKDC gene leads to impaired development of T and B cells resulting in syndrome known as severe combined immunodeficiency (scid) (Bosma and Carroll 1991). Despite efforts in creating CB17-scid mice, this model was not used in many experiments due to the poor engraftment of human hematopoietic stem cells (HSCs) (Bosma et al. 1983).
Further research saw the transfer of scid mutation onto a mouse of non-obese diabetic (NOD) background, creating NOD-scid mice which lacked T cells, B cells, and NK cells. This mouse allowed a slightly higher level of human cell reconstitution (Van der Loo et al. 1998). However, the biggest breakthrough in humanized mice only occurred when mutant interleukin 2 receptor α (IL2rα) gene was introduced into NOD-scid mice, creating NOD-scid-γc null mice (NSG or NOG), which exhibited defective mouse cytokines IL-2, IL-4, IL-7, IL-9, and IL-15 (Ishikawa et al. 2005;Ito et al. 2002;Shultz et al. 2005). Knock-out of recombination activating gene (RAG ) 1 or 2 (RAG1 null and RAG2 null ) caused even greater immunodeficiencies including an absence of NK cells, T cells, B cells, and impaired macrophage and dendritic cell (DC) subsets (Harris and Badowski 2014;Watanabe et al. 2007). However, an absence of human leukocyte antigen (HLA) in these models resulted in engrafted human pre-T cells being "educated" and selected on mouse thymic epithelium and major histocompatibility complexes (MHCs) (Shultz et al. 2010). Due to this limitation, engrafted human T cells were unable to recognise human antigen-presenting cells, and hence, these mice had impaired immunoglobulin (Ig) class switching and disorganised secondary lymphoid structures (Shultz et al. 2010(Shultz et al. , 2012. To overcome this hurdle, HLA class I and II transgenes were added into NSG mice allowing the development of human T-cell repertoires and responses (Brehm et al. 2013;Shultz et al. 2010).
To overcome technical barriers, a few methods to improve the functional human biological systems in mice is to inject humanized mice with recombinant proteins (Huntington et al. 2009;Van Lent et al. 2009), hydrodynamically inject DNA plasmids (Chen et al. 2009), induce lentivirus expression of cytokines (Van Lent et al. 2009), or introduce knock-in gene replacement as so to increase the repertoire of cytokines to support human cells (Billerbeck et al. 2011;Lim et al. 2017;Nicolini et al. 2004;Rongvaux et al. 2011). An example of a technique that is effective does not require complex procedures and can be readily applied in any laboratory is the injection of plasmid DNA (IL-15 and Fms-like tyrosine kinase 3/fetal liver kinase-2 (FLT3/FLK2) ligand) via hydrodynamic tail-vein injection (Chen et al. 2009). Upon application of this method, the expression levels of human cytokines were present for 2-3 weeks, while the levels of functional NK cells remained high for more than a month (Chen et al. 2009). Unlike mice induced to constitutively express cytokines which may activate cells and skew them toward unideal lineages, hydrodynamic injection enables researchers to control the exact timing of cytokine induction, allowing flexible manipulation of the model. On top of this, cytokine-stimulated NK cells expressed activation and inhibitory receptors; attacked in vitro target cells, and responded well to viral infections within an in vivo setting (Chen et al. 2009).
Another method which requires more time and resources to create but eliminates the need for cytokine plasmid injection is the use of transgenic mice with knock-in genes, encoding for cytokines. Four examples of these enhanced immunodeficient mice are, first, NOD.Cg-Prkdcscid Il2rg tm1Sug Tg (SV40/HTLV-IL3, CSF2) 10-7Jic/JicTac (huNOG-EXL mouse), this strain of super immunodeficient mouse has a high rate of human cell engraftment and expresses both granulocyte/macrophage colony-stimulating factor (GM-CSF) and human IL-3 cytokines, controlled by SV40 promoter, which induces myeloid reconstitution and differentiation.
Second, NOD.Cg-Prkdc scid Il2rg tm1Wjl Tg (CMV-IL3, CSF2, KITLG) 1Eav/MloySzJ (NSG-SGM3 mouse) are knock-in mice expressing IL-3, GM-CSF and stem cell factor (SCF) under the control of human-specific cytomegalovirus (CMV) (Billerbeck et al. 2011;Yao et al. 2016). Even though this combination of genes supports human HSC engraftment, formation of myeloid leukocytes, and reduces B-lymphopoiesis post-BM transplantation this model lacks an improved red blood cell (RBC) reconstitution and the presence of SCF may destructively affect human stem cell compartments by supporting the growth and competitive repopulation of mouse cells (Billerbeck et al. 2011;Yao et al. 2016).
Fourth, NOD.Cg-Kit W−41J Prkdc scid Il2rg tm1Wjl /WaskJ (NSGW41) was created to overcome a lack of erythro-megakaryopoiesis in humanized mouse models. Without the need for irradiation, this KIT-deficient mouse demonstrated improved erythropoiesis and platelet formation as compared to other models of mice (Cosgun et al. 2014;Rahmig et al. 2016). After reconstitution, significant numbers of mature thrombocytes were present in the peripheral blood while human erythroblasts were seen in the BM. In addition, the morphology, composition, and enucleation ability of de novo generated human erythroblasts were similar with those in the human BM (Rahmig et al. 2016). However, as this model is relatively new, more studies are needed to further characterise the advances and limitations of this platform. Details of immunodeficient mouse models are listed in Tables 1 and 2. As existing models are far from perfect, it is important to work on components that enhance cell-cell interactions, support differentiation, and induce maturation of human cells, particularly that of myeloid and B cell compartments to create a model that faithfully recapitulates the human immune system.

Models of Human Diseases Established on Humanized Mice
The introduction of humanized mice provides immeasurable opportunities to advance medical research. These increasingly important pre-clinical models are not only easy to handle due to their small sizes, but they also have short reproductive cycles, an exceptional ability to produce a large number of young and are relatively affordable to maintain in animal facilities as they do not require highly specialised infrastructures that are used by NHPs (Fischer and Austad 2011). In addition, humanized mice allow human-specific pathogens to infect and replicate within them and are able to develop functional human-specific immune responses to an array of diseases.
Many mechanisms underlying diseases are not completely dissected; therefore, utilization of humanized mice allows researchers to understand important factors that facilitate the development of medical issues including infectious disease, cancer, autoimmunity, and GvHD. Currently, a mouse model that completely mimics every single human disease does not exist; therefore, research aims such as the consideration of specific parameters to be analyzed including genotype, phenotype of the model, and scientific budget must be thought through carefully to select a suitable platform.

Infectious Disease
Since the invention of humanized mice, multitudinous attempts have been made to recapitulate infectious diseases within these mice. A particular human-specific infectious pathogen that has been successfully studied on humanized mice is a retrovirus known as human immunodeficiency virus (HIV) (Araínga et al. 2016;Berges and Rowan 2011;Choudhary et al. 2009;Duyne et al. 2011;Li et al. 2014). Before humanized mice were introduced, the only nonhuman animal model available for dissecting HIV pathogenesis was the chimpanzee (Vanden Haesevelde et al. 1996). Because of cellular and molecular differences between HIV pathogenesis in humans and chimpanzees, restricted tropism of HIV and high-expense of using NHPs, the small, costeffective, and widely available humanized mice were used in place of the NHPs (Denton and Garcia 2011;Hatziioannou and Evans 2012;Miller et al. 2000).
Humanized mice infected with HIV recapitulated the disease's progression, latency and virology, permitted longterm immunological studies and helped identify crucial factors such as viral infectivity factor, viral protein u, and negative factor which are essential for in vivo HIV replication (Yamada et al. 2015).
Humanized mouse model with a fully functional human immune system has also been infected with Dengue virus (DENV) (Frias-Staheli et al. 2014;Kuruvilla et al. 2007;Sridharan et al. 2013;Subramanya et al. 2010). These mice demonstrated fever, rash, viremia, erythema, thrombocytopenia, and production of anti-DENV IgM, IgG and a range of cytokines as observed in patients Rico-Hesse 2009, 2011). Another human-specific infectious pathogen studied on humanized mice, Plasmodium falciparum, is a causative agent of malaria (Amaladoss et al. 2015;Carlton et al. 2008;Chen et al. 2014;Good et al. 2015;Jiménez-Díaz et al. 2009;Soulard et al. 2015;Vaughan et al. 2012). For years, our understanding of malaria had been impeded by the lack of human-specific small animal models which can be infected by highly host-specific human Plasmodium species (Amaladoss et al. 2015;Chen et al. 2014;Pain et al. 2008) to recapitulate both erythrocytic and immunological disease pathogenesis observed in patients. Due to this, most in vivo experimental studies of malaria were conducted in rodents with mouse or rat-specific Plasmodium strains (Goodman et al. 2013). Differences in invasion and disease pathology between human and rodent parasite species hindered the translation of findings and evaluation of new therapeutics from rodents to humans (Amaladoss et al. 2015;Chen et al. 2014). This challenge has been tackled by incorporating RBC supplemented, immune cell-optimised (enhanced by hydrodynamic expression of human cytokines, IL-15, and FLT3/FLK2 ligand) humanized mice that supports multiple cycles of P. falciparum infection (Amaladoss et al. 2015;Chen et al. 2014).
Utilizing this model, research teams were able to identify the importance of human NK cells, DCs, and B cells in the control of parasitemia. Notably, how NK cells preferentially interacts with infected RBCs (iRBCs), resulting in the activation of NK cells, release of interferon (IFN)-γ, perforin, and granzyme to lyse and eliminate iRBCs in a contactdependent manner and the importance of adhesion molecule lymphocyte-associated antigen-1 and DNAX accessory molecule-1 which are required for NK cell interaction and clearance of iRBCs (Amaladoss et al. 2015;Chen et al. 2014). Besides facilitating the understanding of human immune responses to Malaria infection, the use of humanized mice also assists in evaluation of new therapeutics and vaccines (Good et al. 2015;Tsuji et al. 1995).
In addition to the human immune system, recent progress has been made to introduce humanization of the liver in humanized mice to support the study of hepatotropic pathogens such as hepatitis B virus and hepatitis C virus (HCV) (Bility et al. 2012;Keng et al. 2015;Strick-Marchand et al. 2015;Tan-Garcia et al. 2017;Washburn et al. 2011). It has been shown that these new humanized mice could be infected with human strains of hepatitis viruses and exhibit leukocyte infiltrations, liver inflammation, fibrosis, cirrhosis, and elevated cytokines similar to HCV-infected patients (Bility et al. 2014;Keng et al. 2015;Tan-Garcia et al. 2017;Washburn et al. 2011). Mouse models with human liver cells and matched human immune system provides an important platform for understanding disease pathogenesis of hepatitis viruses through human-specific cytokines, chemokines and immune cell regulations involved, potentially translating this knowledge into creation of anti-fibrotic and immunemodulatory therapeutics (Bae et al. 2015;Keng et al. 2015).

Cancer
Immunodeficient mice that lack innate and adaptive immune cell compartments enable successful engraftment of many human tumors including tumor cell lines and primary solid and hematological tumors. Currently, there are three ways to study tumor growth and cancer immunology in humanized mice. First, tumor cell lines can be engrafted into humanized mice reconstituted with HSCs or PBMCs (Ito et al. 2009;Tsoneva et al. 2017;Wege et al. 2014). Breast cancer was modelled in mice by concurrently transplanting CD34 + HSCs and tumor cells into newborn mice or engrafting both PBMCs and tumor cells into BRG mice (Wege et al. 2014). In these models, human immune cells were able to traffic and infiltrate the microenvironment, enabling human tumor-immune system interactions to be studied (Wege et al. 2014). To more closely recapitulate human immune responses to tumor cell lines, MISTRG mice engrafted with CD34 + human FL cells were subcutaneously transplanted with a melanoma cell line, Me290 (Rongvaux et al. 2014). Similar to clinical scenarios, it was observed that myeloid cells infiltrated the tumor, numerous cells within the tumor expressed CD14 and CD163 which are commonly associated as macrophage markers, and CD163 + cells were most likely M2-like macrophages as they were HLA-DR low and CD206 high . It was hypothesised that tumor growth may have been mediated by M2-like macrophages that can induce cytokine production or release enzymes to promote vascularisation and immune suppression. Therefore, these mice  Moffat et al. (1995) 1 3 were treated with human-vascular endothelial growth factor (VEGF) inhibitor, Avastin ® . Humanized mice engrafted with Me290 responded to treatment by inhibiting tumor growth, suggesting that myeloid cells may support tumor growth via VEGF activity (Rongvaux et al. 2014). Second, immunodeficient mice can be engrafted with patient-derived xenografts (PDX) (Bankert et al. 2011;Her et al. 2017;Simpson-Abelson et al. 2008). Engraftment of patient-derived acute myeloid leukemia (AML) cells into newborn NSG resulted in high levels of human cell engraftment in the peripheral blood, spleen and BM of recipient mice ). Similar to observations in the clinics, these mice also had enlarged spleens and infiltration of AML cells into multiple organs. Even though AML remained unaltered during serial transplantation, many studies with engrafted PDXs into immunodeficient mice have demonstrated that heterogeneity of parental tumor was often only maintained in primary engraftment (Cassidy et al. 2015). Over time and tumor passage, human stromal was frequently compromised by infiltration and replacement with mousederived cells (Cassidy et al. 2015;Maykel et al. 2014). This model is ideal for understanding stroma-tumor interactions, which is integral for tumor growth and an important target for cancer therapy.
Third, for a comprehensive study of interactions between human immune cells and tumor in vivo, immunodeficient mice should be engrafted with PDX and human immune cells (Pan et al. 2017;Roth and Harui 2015). This humanized PDX model would not only have a complete tumor microenvironment but also an ability to display heterogeneity lost in tumors (Pan et al. 2017). However, a drawback of this model is the scarcity of autologous HSCs which affects the capacity to generate cohorts for research. To overcome this challenge, HSCs isolated from UBC, FL or G-CSF mobilised PBMCs can be expanded either by transduction with tat-MYC and tat-Bcl2 fusion proteins or cultured with a validated cocktail of growth factors to induce in vitro proliferation of HSCs (Bird et al. 2014;Yong et al. 2016). An example of this model is XactMice which are engrafted with in vitro expanded HSCs and autologous PDX samples from head and neck squamous cell carcinoma patients (Morton et al. 2016). Even though these mice had low levels of humanization in their peripheral blood, they demonstrated an increase in lymphatic vessels and the presence of CD45 + CD151 + cells, suggesting that these mice were able to recapitulate immune and stromal cell compartments of the tumor microenvironment (Morton et al. 2016).
While the current immunodeficient mouse strains are able to support the engraftment of most tumor cell lines, not all primary tumors for example prostate cancer can be easily engrafted (Roth and Harui 2015). Novel humanized oncological models are being innovated to address important questions on tumor-immune system interactions, mechanisms of tumor escape, therapeutic potential of immune modulation, as well as refining therapeutic solutions such as chemotherapy, NK cell therapy, checkpoint inhibitors and cytokine therapy. Tumor cell lines, and solid and hematological cancers tested on humanized mice are listed in Table 5.

Autoimmunity
Disparities in the immune system between mice and men restrict the use of mouse models which develops spontaneous autoimmunity (Covassin et al. 2013). To overcome this challenge, Gunawan et al. (2017) engrafted PBMCs from systemic lupus erythematosus (SLE) patients to create a human-specific disease-based immune system which demonstrated that human T and B cells were present in the peripheral blood and spleen of humanized mice and were important to lupus development. Similar to patients, when these mice were treated with dexamethasone, spleen weight, and proteinuria decreased. Mice with a human immune system xenografted with patient samples allow a spectrum of disorders such as SLE (Andrade et al. 2011;Gunawan et al. 2017) and type I diabetes (Shultz et al. 2007;Unger et al. 2012;Viehmann Milam et al. 2014) to be evaluated for the identification of screening markers, retrieval of antigen-specific autoantibodies, and drug tests. Autoimmune diseases that have been studied using humanized mice as a platform are listed in Table 6.

Graft-versus-host Disease
The occurrence of GvHD is a life-threatening complication that may develop following transplantations (Hu et al. 2011;Hu and Yang 2012). Even though GvHD has been intensively analyzed in non-humanized animal models, many humanspecific mechanisms and treatments cannot be tested due to incongruence between humans and mice. Humanized mice are excellent substitutes to investigate exact human immune responses of GvHD and its related therapeutics (Ali et al. 2012;King et al. 2008;Kirkiles-Smith et al. 2009;Tobin et al. 2013;Wang et al. 2011;Zhao et al. 2015). An example of a humanized mouse model applied in GvHD studies is the engraftment of human PBMCs into immunodeficient mice (Ali et al. 2012). Post-transplantation, these mice demonstrated human lymphocytes infiltration into peripheral blood, spleen, lymph nodes, and BM of the mice, had enhanced tissue homing cells with a T-effector memory (T EM ) phenotype and high levels of cutaneous lymphocyte antigen, recapitulating the exact pathogenesis of GvHD as observed in patients (Ali et al. 2012;Wang et al. 2011). Utilizing humanized mice to understand human-specific mechanisms of rejection provides a strong pre-clinical platform for the design of novel immunotherapies (Fogal et al. 2011;Onoe et al. 2011;Tobin et al.   Mice were successfully engrafted with a functional human immune system. Oncolytic vaccinia virus therapy, particularly CTLA4 scAb increased CD56 + NK cells and decreased virus titers Tsoneva et al. (2017) 1 3 2013), especially those targeting T EM cell driven GvHD (Ali et al. 2012). Transplant rejection studies that have been conducted on humanized mice are listed in Table 7.

Human-Specific Drug Tests on Humanized Mouse Models
Non-human animal models are commonly used to test an array of human-specific therapeutics during pre-clinical trials. Due to a lack of human specificity, it is common for pre-clinical trials to inadequately identify exact pharmacokinetics, pharmacodynamics, and side-effects of therapeutics, which may result in debilitating and life-threatening situations when tested on humans (Horvath et al. 2012;Rehman et al. 2011;Xu et al. 2014). To improve from unsuccessful clinical trials, it is important to use validated and costeffective animal models with high human specificity such as humanized mouse models to expand the traditional armamentarium of therapeutics for treatment of patients with complicated and progressive conditions. Therapeutics successfully tested in mice with a functional human immune system includes an antiviral drug, peginterferon alpha-2a (Peg-IFNα2a) which demonstrated signs of HCV inhibition such as decreased human IFN-γ production, level of serum alanine aminotransferase, copies of HCV ribonucleic acid (RNA), and absence of leukocyte infiltration or fibrosis in the liver (Keng et al. 2015). Similar to clinical scenarios, humanized mice administered with Ipilimumab developed autoimmune disease with signs of weight loss, anti-nuclear antibodies, and adrenalitis. In addition, a biologic highly specific for human CD28, theralizumab, was tested in humanized mice engrafted with PBMCs (Weißmüller et al. 2016). These mice demonstrated severe reduction in CD45 + human cells, rapid drop of body temperature, elevated levels of cytokines, and succumbed to treatment within 6 h after antibody administration, recapitulating adverse effects observed in clinical scenarios (Weißmüller et al. 2016).
Considering the strengths, limitations, and potential developments of humanized mice, the current data indicate that these models are beneficial tools for researchers to investigate short and long-term studies of in vivo therapeutic interactions and toxicities to mitigate risks and ensure the safety of healthy volunteers and patients exposed to candidate agents during clinical trials. Therapeutics that has been tested on humanized mice is listed in Table 8.

Future Directions and Conclusion
To address gaps in humanized mice, scientists working in different biomedical disciplines are attempting a myriad of approaches including boosting human cell reconstitution,     Weißmüller et al. (2016) reducing graft rejections, supporting critical immune cell subsets, and improving human-specific responses toward pathogens to maximise the potential of humanized mice as a pre-clinical platform. Despite an optimistic outlook of humanized mice, there are considerable obstacles associated with the model that has to be solved as soon as possible. This includes scarce sources of human cells and tissues, particularly obtained from fetal samples due to ethical restrictions. A solution for this limitation is underway as teams around the world perfect induced pluripotent stem cell (iPSC) technology, which enables the use of patient-specific iPSCs allowing a renewable source of autologous cells sans immune rejection .
In humanized mice, secondary lymphoid structures are either missing or disorganised; this curtails essential humoral responses, resulting in impairments for both class switching and affinity maturation post-immunisation. To overcome this, lymphoid tissue inducer cells should be introduced without affecting IL2rg receptors (Lim et al. 2017). Alternatively, immunodeficient mice can be engrafted with both FL and cells that support FL cell growth from the same clinical donor and supplemented with cytokines (e.g., IL-1β, IL-2, IL-7, and GM-CSF), so that differentiation and maturation of HSCs can take place to improve functional immune cells including macrophages, follicular DC, and T helper cell reconstitution (Chen et al. 2009;Lim et al. 2017;Yong et al. 2016).
An absence of essential human cytokines hinders optimal HSC engraftment, differentiation, and maturation of functional immune cells. To tackle this issue, mouse models can be hydrodynamically boosted with plasmids encoding cytokines (Chen et al. 2009). Despite this improvement, binding of human cytokines may be hindered by residual mouse cytokines or may induce mouse cells to proliferate and displace the engraftment of human cells due to the cross-reactivity between some human and mouse cytokines. Eliminating this problem entirely would require absolute depletion of murine cells or the introduction of high affinity human-specific cytokines and growth factors.
Human cell engraftment is being negatively affected by mouse cells (RBCs and innate immune cells) that were not completely depleted during the construction of immunodeficient mice. To improve this, additional gene knock-outs could be added to current strains of immunodeficient mice to further reduce mouse RBCs, granulocytes and macrophage functions (Hu et al. 2011;Hu and Yang 2012), however, because of the low human erythrocyte engraftment, excessive reduction of mouse RBCs might result in anemic mice which has short lifespans, are weak and not suitable for experiments (Rongvaux et al. 2014). A long-term solution would be to optimise and increase the engraftment rate of human RBCs in humanized mice, so that all traces of mouse RBCs can be removed (Hu and Yang 2012). ATG anti-thymocyte globulin, HSPCs hematopoetic stem and progenitor cells Long-termism, critical analysis, and adequate troubleshooting to solve existing problems in humanized mice would undoubtedly provide exciting opportunities for the establishment of new and improved humanized models with increased human immune cell engraftment and enhanced functionality that would greatly benefit the community.