Potential Pitfalls of the Humanized Mice in Modeling Sepsis.
Animal models are frequently employed as a precursor to clinical trials and even more broadly in research investigations. Despite their enormous contribution to the development of scientific knowledge, there is a well-funded and accepted appreciation of their limitations [1-3]. Animal and human physiology may be similar, but this is not a universal rule . For example, the toll-like receptor pathway is relatively conserved in both humans and Drosophila, but the clinical response to ligands varies between animals and humans . These differences are sufficiently profound to prevent the successful, direct translation of discoveries in mice into human clinical trials. However, the high prevalence and mortality resulting from sepsis necessitate further research in the field of critical care inflammation using animal models [2, 6].
Some of the issues hampering clinical implementation of discoveries in animal models are related to the shortcomings of the general methodology of animal experimentation . Animals are often inbred and kept in sterile conditions, leading to high homogeneity. While reducing interindividual variability is helpful in pure laboratory research, the applicability of the research may be compromised by the lack of diversity typical of human patients. It is also questionable how well the environment of the animal facility resembles natural conditions. Stable temperature, rigorously controlled diet, multiple barriers to prevent infection, and limitations on animal mobility due to housing constraints are far from typical clinical environments [2, 7]. Furthermore, disease models are only an approximation of illness and may not correspond well to clinical scenarios . For example, cecal ligation and puncture are a widely employed model of sepsis in rodents, but they have been criticized for not adequately reflecting typical clinical conditions and treatment [3, 9]. In particular, antibiotic and fluid management are dramatically different between lab animals and clinical settings [2, 10]. Lastly, rodent models offer only limited efficacy in testing several higher-order neurological functions such as cognition, memory, and emotional regulation which makes modeling of inflammation-induced dysfunction of central nervous system somewhat difficult.
Numerous papers have been published debating the utility of animal models [3, 4, 8, 9]. There is little doubt that animal models are essential research tools, but increased concerns have triggered a search for alternative methods to investigate important clinical questions [3, 4, 8, 11]. With the National Institutes of Health (NIH) emphasizing "translational research," there is added impetus to look for these alternatives.
One promising approach to bridging this translational gap is the application of "humanized mice." Here, we will review the basic understanding of their immunology and their applicability to human disease in the context of acute critical care illness. Our review discusses certain characteristics of the humanized mice model that are potentially limiting their capacity for direct translation of research findings into clinical therapy [9,12].
2. Development of "the Ideal" Humanized Mice
Humanized mice can be defined simply as mice carrying human genes or tissues such as leukocytes, stem cells, organs, and tumors . In this review, we will focus only on animals with a reconstituted immune system and not mice with transplanted human neoplastic tissues, human pancreatic islets, cardiomyocytes, or other biological materials. A viable humanized mouse model is contingent on the successful development of an immunodeficient host enabling the native immune system to combat the transplanted one. Consequently, the primary goal of humanized mice development is to engineer mice with increasingly deficient native immune systems to curb the rejection of the transplanted immune system [13-15]. The secondary goal is aimed at providing the grafted cells with a specific microenvironment in which they can successfully settle and thrive .
2.1. Development of Immunocompromised Host. The earliest model consisted of CB17 mice strain carrying the [Prkdc.sup.scid] mutation leading to severe combined immunodeficiency (SCID) [13, 16]. The first attempt was a simple transplant of human peripheral leukocytes into mice . HIV-1 was the primary impetus for the development of this model, notwithstanding issues with efficient grafting and graft stability with overrepresentation of NK cell activity. The development of nonobese diabetic- (NOD-) scid mice further improved the engraftment in comparison to previous immunodeficient host types . The success of this subsequent model was attributed to its decreased NK cell activity and the emergence of additional defects of innate immunity, which allowed for higher levels of engraftment [10, 18]. Null mutations in the IL-2 receptor [gamma] chain ([IL2rg.sup.null]) were particularly successful in attaining this goal. Consequently, mouse strains including IL-2 receptor y chain mutations such as NOD/Shi-scid/IL2r[[gamma].sup.-/-] (NOG), NOD-[Rag.sup.1-/-]IL2r[[gamma].sup.-/-] (NRG), NOD-scid/IL2r[[gamma].sup.-/-] (NSG), or BALB/c/Rag[2.sup.-/-]IL2r[[gamma].sup.-/-] (BRG) were widely adapted for their best hosting capability [15, 18-20]. The deletion of [beta]2-microglobulin (NOD-said B2[m.sup.-/-]), a critical component of MHC I, further improved engraftment compared to NOD-SCID mice . The most unique feature of the NOD-Rag[1.sup.-/-] mice was generation of the first radiation-resistant strain, though engraftment was less efficient as compared to previous strains [19, 21]. Progress continued in developing mice strains with increasingly deficient native immune systems improving graft success and stability. Subsequently, the newest types of humanized mice have satisfactory recovery of T cells, B cells, and NK cells despite some variability [22-24]. At the same time, they suffer from diminished monocyte and dendritic cell counts coupled with frequently underperforming function of their humanized leukocytes [25-27]. High variability in emergence of the human mononuclear component results from the variable expression of signal regulatory protein-[alpha] (SIRP-[alpha]) .
The immunocompromised host retains a significant part of the native immune system despite all aforementioned interventions. Consequently, all models are at risk of developing graft versus host disease (GVHD) since transplanted CD[8.sup.+] cells will interact with MHC class I of the host [15, 28, 29]. Incomplete eradication of the native immune system results in the development of lymphomas whereby more advanced humanized models have had much longer lag time and a lower propensity allowing for prolonged longitudinal studies .
2.2. Evolution of Grafting. First, researchers observed that human peripheral blood mononuclear cells (PBMCs) could engraft successfully [14, 31]. While injecting mice with PBMCs resulted in the reconstitution of T cells, the high risk of graft versus host disease limited such models to short-term experiments . In next step human fetal thymus and liver were implanted into the kidney capsule and CD[34.sup.+] cells from bone marrow (BLT model) . This system preserved the complex interaction between antigen presenting cells (APCs) and T cells, which is critical in the resolution of acute inflammation [8, 10, 32]. MHC restricted selection of T cells took place in this model. Furthermore, a secondary lymphatic system developed along the human mucosal immune system. In the next step, CD[34.sup.+] stem cells were used for grafting allowing for even more complete restoration of the immune system (HIS model) [33, 34]. This model featured high graft stability and diversity, proportional to the number of cells used for grafting [33-36].
Due to a lack of species-specific cytokines and inefficient ability of host mesenchymal cells to support function of the leukocytes, an emergence of a more complete human immune system was impaired [23, 37, 38]. Lack of interlocking cytokine networks contributed to the poor regulation of leukocyte populations . Nanoparticles, plasmids, and lentiviruses were tried to boost the supportive environment of mice bone marrow since optimal cytokine environment is pivotal for the emergence of a complete immune system [21, 38, 39]. Despite these efforts and several other modifications, the development of most specialized cells was hampered [40, 41]. Even more alarming was the emergence of unintended consequences of genetic manipulation aimed at boosting the production of human cytokines. For example, Billerbeck et al. found that increased expression of stem cells factor, IL-3, and GM-CSF led not only to the increased expression of monocytes but also to the significant skewness of the circulating T cells toward Fox3(+) T cells . Aberration of composition of leukocyte population is one of the critical phenomena of sepsis. Consequently, an introduction of any bias may affect the validity of results and their translation into clinical practice. Currently, engineering of MIRTG mice, which can produce four human cytokines endogenously, is the most sophisticated model even among similar models [23, 40, 41]. Assessment of MIRTG mice performance in sepsis follows.
2.3. Additional Techniques. Ancillary techniques surrounding grafting procedures have evolved as well. Initially, all animals were subjected to irradiation to remove the native immune system [15-17, 19, 33, 34, 36, 40]. However, collateral damage to supportive structures of bone marrow and organs was often present. Humanized mice with Rag2 mutations appeared to be less sensitive to the effects of radiation [19,41]. Some mice strains such as those carrying the c-kit mutation did not need irradiation, but they have not been evaluated in sepsis studies . Finally, chemical ablation can be used, but it appears to have a detrimental effect on animal survival .
Since murine monocytes and other indigenous phagocytic cells are responsible for poor engraftment, several targeted techniques aimed at their eradication were developed such as the use of anti-murine SIRP[alpha], CD47, or Ncf1 knockout, and clodronate liposomes [42-44].
Additionally, reconstitution of the immune system was found to be more efficient and better tolerated in newborn versus adult mice especially if augmented by the injection of human grafting factors .
3. The Performance of Immune System Components in Humanized Mice
Reconstitution of the human immune system is a relatively slow process that may take up to 3 months . B cells are the earliest leukocytes to reconstitute, followed by T cells . B cells in humanized mice produce all classes of immunoglobulins [22, 53, 58-60]. These data should be viewed with caution since the maturation of B cells and antibody class switching from IgM to IgG are particularly inefficient as compared to the human system [61, 62]. Effective immunoglobulin class switching is critical for recovery from an acute infectious process. The lack of human IL6 is considered one of the most important reasons for this inefficiency. Only recently, a novel model of humanized mice with active human IL-6 gene was described. Interestingly, while the immunoglobulin switch was more efficient in this model, the maturation of B cells remained impaired . Evaluation of these IL-6 boosted mice should be noteworthy as IL-6 is one of the critical cytokines in the development of sepsis. Additionally, the introduction of human stem cell factor, granulocyte macrophage stimulating factor, and IL-3 into the BLT model resulted in more efficient maturation and optimized immunoglobulin production at baseline and after viral infection .
T cells obtained from humanized mice after reconstitution are proficient as measured by the delayed hypersensitivity reaction, but their ability to respond to antigen was suboptimal . Introduction of the BLT model partially resolved this problem as human T cells rely on grafted fetal thymus for clonal selection . The performance of T cells depended on several factors but was mostly mediated by stimulation via MHC class II or IL-2R . Since humanized mice are xenotransplanted animals, it is unclear how a state of tolerance to different MHC antigens affects their performance. Lack of MHC-matched APCs would impair that process [25, 33, 64]. Conversely, an additional transplant of sensitized dendritic cells alleviated the problem and presented an interesting opportunity for future research . Altering the cytokine environment was another way to improve the presence of APCs like dendritic cells (DCs) [33, 54, 65-67]. In critical care illnesses, like sepsis, dendritic cells emerge from circulating monocytes that stimulate optimal T cell responses [10, 32]. Recently, only one study has investigated monocyte function in sepsis using humanized mice . Other T cell populations may encounter similar difficulties as well. Reconstitution of the T cells in mucosal membranes depends on the presence of [gamma][delta] chain, but its expression was variable in humanized mice . Since this subtype of T cells plays a critical role in the emergence of the tolerance and the modulation of complex T cell responses, it is unclear how the deficit will affect the evolution of the immune response in the setting of critical care illness. T cells with regulatory properties ([T.sub.reg]) are present in humanized mice, but their role seems to be conflicting in terms of function and number, and it is greatly influenced by the cytokine environment [37, 69]. It is also worth mentioning that balance between different T cell populations may be abnormal in humanized mice and was linked to a deficiency of the human cytokine network [22, 29, 37, 45, 48].
Myeloid cells are the last to reconstitute after grafting. Stem cell factor, M-CSF, GM-CSF, and IL-4 are supportive in speed and efficiency of the recovery as well as in functional maturation [23, 70]. Slow and incomplete reconstitution of the myeloid line results in the inability of T cells to mount a proficient response to antigen challenges. Enhanced recovery of the myeloid compartment was seen in MISTRG and MITRG mice . MO obtained from humanized mice resemble neonatal cells in their ability to upregulate CD80/CD86, two critical factors in modulating T cell function . MO from humanized animals were shown to generate a robust T cell response and cytokine production after sepsis [35, 55]. Supplementation of the humanized mice with in vitro generated allogeneic DCs can restore T cell responsiveness [46, 64, 65]. DCs emerge in some humanized models on their own or after Flt3 supplementation [46, 49]. Such in vivo-generated and antigen-sensitive DCs can trigger T cell response to a specific antigen .
In summary, these studies show that the function of several leukocyte populations can be restored during reconstitution of the immune system. However, the complex nature of the process, dependence on numerous interventions, and unclear functional competency of leukocytes undermine the robustness of the humanized model.
4. Current Studies of Humanized Mice and Sepsis
Humanized mice were used successfully to study HIV [19, 30]. Most recently, several other viral infections were successfully modeled in humanized mice including Zika and West Nile virus [59, 71]. Introduction of Epstein-Barr virus reproduced several traits of the infection with high fidelity in humanized mice . The ability to replicate the trajectory of viral hepatitis, longevity, and mimicry of response to subsequent infections made humanized mice especially suitable for finding the optimal drug to cure hepatitis C . The endothelial inflammation of the highly lethal dengue virus and other pathological agents causing hemorrhagic fever were investigated in humanized mice, but concerns were raised regarding the accuracy of the model [60, 72, 73]. Of primary concern was the ability of the xenotransplant to mimic vasculitis and interactions between mice endothelium and human immune system [55, 74-76]. This illustrates a typical shortcoming of humanized mice when the interaction between two organs encounters an interspecies difference that can be overcome only through further modification of the models (Table 1). Whether these modifications make the model closer to reality or more artificial remains to be ascertained.
The success of humanized mice in mimicking viral infections established high expectations for using them to study sepsis [2, 32].
Sepsis is a highly prevalent and serious condition that has a profound and prolonged impact on morbidity and mortality [6, 40, 55, 74-77]. Unsinger et al. demonstrated that humanized mice replicate several key features of the septic process, including apoptosis and exaggerated cytokine production . Bone marrow suppression closely resembling the natural history of sepsis was seen as well . The critical role of HMGB1, TLR4, and Notch in sepsis and apoptosis was also demonstrated using humanized mice [67, 76-78]. However, the degree to which these processes replicate the complexity of the septic response is difficult to assess fully. For example, IL-15, a critical cytokine for the development of sepsis-related apoptosis, can be studied in humanized mice only after modification of the model still resulting in production of age-dependent IL-15 [32, 39, 78]. Introduction of several human cytokines improved cell recovery but introduced artificially skewed populations . It was demonstrated that extended depression in bone marrow function is mediated by methylation changes in the PU.1 gene, and retransplantation of the postsepsis surviving mice with allogeneic stem cells partially restored immune reactivity . However, M-CSF production and biological activities were limited to transplanted stem cells since the mice environment did not provide indigenously produced cross-reactive cytokines [66, 70]. It becomes evident that these observations may not reflect clinical reality with so many features absent from the model.
Other humans disease models successfully mimicked in humanized mice include Toxic Shock Syndrome Toxin-1(TSST-1-) mediated shock and staphylococcal infections [40, 76]. To date, only a few research investigations have focused on sepsis in humanized mice [10, 27, 55, 67, 76, 77]. The CLP model of sepsis is by far the most popular for studying sepsis despite several shortcomings of the CLP itself and some unique features of humanized mice undergoing sepsis.
Further studies with humanized mice tested clinical compounds for the treatment of sepsis. Wang et al. tested the potential of curcumin analogs to reverse lung injury secondary to sepsis in humanized mice . Autologous stem cell transplants showed a potential to reverse some of the postseptic immune system aberrations, but clinical relevance remains to be seen . In another example, antibodies to human-specific toxins were tested . Finally, a researcher utilized humanized mice to test indomethacin as the modulator of the immune response in neonatal sepsis under the assumption that the functional immaturity of the grafted immune system was a good model of neonatal immunity [79,80]. These results are guarded due to insufficient evidence as to whether immature humanized mice and neonatal immunology are in fact equivalent.
5. Limitations of Humanized Mice Models to Study Sepsis
Humanized mice seem to be an appealing choice to investigate the pathology and treatment of sepsis [2, 10, 32, 75]. However, interest remains relatively low. PubMed cites approximately 90 publications for which "sepsis" and "humanized mice" are keywords. Many of the authors assume that several limitations of humanized mice prevent broader implementation into mainstream septic research (Table 1). Humanized mice earned justifiable praise from several researchers. However, the inherited problems of this model have been acknowledged by a few [8, 10, 12, 29, 42, 62, 77, 81].
First, there exist fundamental differences between human and mice physiology . More specifically, humanized mice exhibit several differences in the natural history of sepsis. Weight loss and mortality are greater in humanized mice than wild-type mice when short-term and long-term data are analyzed [35, 55, 67, 81]. Only the introduction of extensive measures (antibiotics, fluid resuscitation, and diet modification) resulted in animal survival exceeding a couple days [35, 55]. Prolonged studies were complicated by the emergence of GVHD and lymphomas as well as the suppressive effect of preserved components of the indigenous mice immune system [15, 29, 82]. Additionally, the recovery of the granulocyte compartment required supplementation of human G-CSF . Considering that granulocytes are a pivotal defense against microbial infections, this need for supplementation is a shortcoming of humanized mice in mimicking their function and heterogeneity significantly limiting that model [2, 32]. The deficit in the granulocyte compartment may also underlie early mortality in sepsis and require a more aggressive therapeutic approach . Furthermore, all leukocyte types demonstrate a sign of functional immaturity unless remedial measures are implemented [25, 31, 36, 46, 66]. The corrective modification may create an artificial condition on its own or only be partially effective [37, 38]. In the example of [T.sub.reg] expansion in humanized cells, the imbalance may not reflect the natural history of sepsis or postseptic leukocyte population changes . Other shifts in T cell population composition were reported, but virtually no study investigated population heterogeneity of monocytes, NK, and B cells in the context of the response to infection [35, 45, 55]. Finally, another limitation exists in that some of the pivotal cells of the immune system are not replaced in grafting. Particularly, microglia are not part of the humanized grafted system with potentially profound negative effects on the modeling of the central nervous system's effects from sepsis.
Second, the time after grafting is critical for modulating immune responses. A minimum of 8 weeks is required for NSG grafting but longer periods are related to the emergence of lymphoma and GVHD [28, 50, 60, 82]. The question of how to measure the age of humanized mice remains. Possible measurements include host's age, grafting time, or the age of the grafted human immune system. This important question has some biological underpinnings since the production of cytokines in response to pathogens has been reported as a variable dependent on the amount of time after grafting . Since sepsis disproportionally affects the lower and upper extremes of age, younger mice are still reconstituting their immune system while older ones are at higher risk of immune system disorders; humanized mice may not be the most suitable model in age-related studies of sepsis [2, 6, 32, 83].
To improve grafting, animals are frequently irradiated. Despite wide acceptance of this process, the effect of irradiation on humanized mouse physiology and sepsis remains complex and extends beyond just the immune system [56, 84]. Certain humanized models do not require irradiation, but none of them have been evaluated in septic conditions [19, 24, 85].
Third, gut flora has a significant effect on the performance of the immune system and is increasingly perceived as one of the modifiers for sepsis trajectory [2, 32]. Humanized mice have mouse gut flora interacting with the human immune system, but significant alterations are also introduced by irradiation [41, 84]. Some attempts to establish human-specific flora in humanized animals have been tried to increase the fidelity of the models since gut immune interaction is gaining increased recognition in the pathology of sepsis .
Fourth, sepsis is a multiorgan disease with poorly defined etiologies. The multiple interactions between pathogen and host organs as well as between the host organs themselves are crucial for understanding the clinical trajectory [2, 32, 85]. Mesenchymal cells, the central nervous system, and endothelium are frequently quoted culprits of unfavorable outcomes in sepsis [57, 85, 86]. Several deficits in the crosstalk between the grafted human immune system and mice organs may create a translation barrier for implementation of humanized mice into sepsis research . Their xenotransplant nature inherently limits their usefulness as it was already suggested in studies of hemorrhagic fever or meningitis [57, 86]. Additionally, the endothelium is critical for immune system performance, and humanized mice are deemed a less favorable model for them unless significant modification is introduced . Moreover, on the level of immune system interaction between APC and effector cells is notably ineffective. Vagal nerve or subcortical structures have also been suggested as playing a role in sepsis outcomes [2, 32, 86]. Meanwhile, it is unclear how the murine nervous system interacts with the human immune system during critical care illness. Lastly, mesenchymal cells are not able to have crosstalk efficiently with the grafted human immune system unless modified .
These are significant limitations of humanized mice as a model of sepsis due to the xenotransplant nature of these animals. In contrast, humanized mice are robust models to study pathogen cycles or relatively simple, most likely conservative, immunological processes since these illnesses trigger very specific immune system response or have a life cycle that is independent of the immune system [40, 59, 71, 72, 82].
Fifth, one often-overlooked limitation of the humanized mouse model is the statistical approach to the grafted animals. For example, several mice can be injected with the same stem cell. It is then debatable whether these mice represent one organism/ecosystem repeated several times or if they are independent experiments. Using the same cells in highly modified NSG mice would intuitively support the former conclusion. On the other hand, the interaction of the immune system with the mouse body results in unavoidable (and perhaps desirable) stochastic variation in the characteristics of the immune system.
Finally, one cannot ignore the costly nature of the humanized mice model. The high initial cost derives from the cost of stem cells used for grafting since the efficiency of grafting is proportional to the cells used for grafting. Furthermore, the animals must be housed in exceptionally protective conditions and may require a special maintenance regimen. Often, mice production is lengthy and, in some cases, restricted by patents.
In conclusion, humanized mice promise a better approximation of human physiology and are often recommended as a tool to bridge the gap between rodent models and clinical scenarios. However, such a view is potentially oversimplified. Our review shows that humanized mice have several unclear biases, which may profoundly affect their ability to mimic clinically relevant scenarios of sepsis. Furthermore, the significant cost and limitations associated with this model may not justify their use. There is little doubt that humanized mice are a useful tool to study the partial mechanism of sepsis, but the complexity of this disease demands more sophisticated models whereby several complex systems may interact with each other.
Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this paper.
The authors would like to express appreciation to Lucia DiNapoli for proofreading this manuscript. This work is supported by the National Institute of Health NIGMS 1K23GM120630-01A1.
 A. Gonzalez-Fernandez, J. Faro, and C. Fernandez, "Immune responses to polysaccharides: Lessons from humans and mice," Vaccine, vol. 26, no. 3, pp. 292-300, 2008.
 R. P. Dellinger, M. M. Levy, A. Rhodes et al., "Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Surviving Sepsis Campaign Guidelines Committee including the Pediatric Subgroup," Intensive Care Medicine, vol. 41, no. 2, pp. 580-637, 2013.
 M. F. Osuchowski, D. G. Remick, J. A. Lederer et al., "Abandon the mouse research ship? Not just yet!," Shock, vol. 41, no. 6, pp. 463-475, 2014.
 J. Mestas and C. C. W. Hughes, "Of mice and not men: differences between mouse and human immunology," The Journal of Immunology, vol. 172, no. 5, pp. 2731-2738, 2004.
 E. P. McCarron, D. P. Williams, D. J. Antoine et al., "Exploring the translational disconnect between the murine and human inflammatory response: Analysis of LPS dose-response relationship in murine versus human cell lines and implications for translation into murine models of sepsis," Journal of Inflammation Research, vol. 8, pp. 201-209, 2015.
 H. E. Wang, R. S. Devereaux, D. M. Yealy, M. M. Safford, and G. Howard, "National variation in United States sepsis mortality: A descriptive study," International Journal of Health Geographics, vol. 9, article no. 9, 2010.
 D. Masopust, C. P. Sivula, and S. C. Jameson, "Of mice, dirty mice, and men: Using mice to understand human immunology," The Journal of Immunology, vol. 199, no. 2, pp. 383-388, 2017.
 P. McGonigle and B. Ruggeri, "Animal models of human disease: Challenges in enabling translation," Biochemical Pharmacology, vol. 87, no. 1, pp. 162-171, 2014.
 J. Seok, HS. Warren, AG. Cuenca, MN. Mindrinos, HV. Baker, W. Xu et al., "Genomic responses in mouse models poorly mimic human inflammatory diseases," Proceedings of the National Acadamy of Sciences of the United States of America, vol. 110, no. 9, pp. 3507-3512, 2013.
 J. Unsinger, J. S. McDonough, L. D. Shultz, T. A. Ferguson, and R. S. Hotchkiss, "Sepsis-induced human lymphocyte apoptosis and cytokine production in "humanized" mice," Journal of Leukocyte Biology, vol. 86, no. 2, pp. 219-227, 2009.
 J. P. Sundberg, A. Berndt, B. A. Sundberg et al., "The mouse as a model for understanding chronic diseases of aging: the histopathologic basis of aging in inbred mice," Pathobiology of Aging & Age-related Diseases, vol. 1, article 7179, 2011.
 M. A. Brehm, A. Cuthbert, C. Yang et al., "Parameters for establishing humanized mouse models to study human immunity: Analysis of human hematopoietic stem cell engraftment in three immunodeficient strains of mice bearing the IL2r[gamma]null mutation," Clinical Immunology, vol. 135, no. 1, pp. 84-98, 2010.
 G. C. Bosma, R. P. Custer, and M. J. Bosma, "A severe combined immunodeficiency mutation in the mouse," Nature, vol. 301, no. 5900, pp. 527-530, 1983.
 D. E. Mosier, R. J. Gulizia, S. M. Baird, and D. B. Wilson, "Transfer of a functional human immune system to mice with severe combined immunodeficiency," Nature, vol. 335, no. 6187, pp. 256-259, 1988.
 M. A. King, L. Covassin, M. A. Brehm et al., "Human peripheral blood leucocyte non-obese diabetic-severe combined immunodeficiency interleukin-2 receptor gamma chain gene mouse model of xenogeneic graft-versus-host-like disease and the role of host major histocompatibility complex," Clinical & Experimental Immunology, vol. 157, no. 1, pp. 104-118, 2009.
 S. W. Christianson, D. L. Greiner, R. A. Hesselton et al., "Enhanced human CD4+ T cell engraftment in beta2-microglobulin-deficient NOD-scid mice," Journal of immunology (Baltimore, Md), vol. 158, no. 8, pp. 3578-3586, 1950.
 M. Ito, H. Hiramatsu, K. Kobayashi et al., "NOD/SCID/[gamma]cnull mouse: An excellent recipient mouse model for engraftment of human cells," Blood, vol. 100, no. 9, pp. 3175-3182, 2002.
 P. Lan, N. Tonomura, A. Shimizu, S. Wang, and Y.-G. Yang, "Reconstitution of a functional human immune system in immunodeficient mice through combined human fetal thymus/liver and CD[34.sup.+] cell transplantation," Blood, vol. 108, no. 2, pp. 487-492, 2006.
 L. D. Shultz, P. A. Lang, S. W. Christianson et al., "NOD/LtSz-Rag1(null) mice: An immunodeficient and radioresistant model for engraftment of human hematolymphoid cells, HIV infection, and adoptive transfer of NOD mouse diabetogenic T cells," The Journal of Immunology, vol 164, no. 5, pp. 2496-2507, 2000.
 S. P. McDermott, K. Eppert, E. R. Lechman, M. Doedens, and J. E. Dick, "Comparison of human cord blood engraftment between immunocompromised mouse strains," Blood, vol. 116, no. 2, pp. 193-200, 2010.
 L. D. Shultz, P. A. Schweitzer, S. W. Christianson et al., "Multiple defects in innate and adaptive immunologic function in NOD/LtSz-scid mice," Journal of immunology (Baltimore, Md), vol. 154, no. 1, pp. 180-191, 1950.
 Y. Watanabe, T. Takahashi, A. Okajima et al., "The analysis of the functions of human B and T cells in humanized NOD/shiscid/[gamma]cnull (NOG) mice (hu-HSC NOG mice)," International Immunology, vol. 21, no. 7, pp. 843-858, 2009.
 A. Rongvaux, T. Willinger, J. Martinek et al., "Development and function of human innate immune cells in a humanized mouse model," Nature Biotechnology, vol. 32, no. 4, pp. 364-372, 2014.
 B. E. McIntosh, M. E. Brown, B. M. Duffin et al., "Nonirradiated NOD,B6.SCID Il2r[gamma]-/- kitW41/W41 (NBSGW) mice support multilineage engraftment of human hematopoietic cells," Stem Cell Reports, vol. 4, no. 2, pp. 171-180, 2015.
 C. Gille, T. W. Orlikowsky, B. Spring et al., "Monocytes derived from humanized neonatal NOD/SCID/IL2R[gamma] null mice are phenotypically immature and exhibit functional impairments," Human Immunology, vol. 73, no. 4, pp. 346-354, 2012.
 C. S. Netherby, M. N. Messmer, L. Burkard-Mandel et al., "The granulocyte progenitor stage is a key target of irf8-mediated regulation of myeloid-derived suppressor cell production," The Journal of Immunology, vol. 198, no. 10, pp. 4129-4139, 2017.
 Y. Wang, X. Shan, Y. Dai et al., "Curcumin Analog L48H37 Prevents Lipopolysaccharide-Induced TLR4 Signaling Pathway Activation and Sepsis via Targeting MD2," The Journal of Pharmacology and Experimental Therapeutics, vol. 353, no. 3, pp. 539-550, 2015.
 N. Hilger, J. Glaser, C. Muller et al., " Attenuation of graft-versus-host-disease in NOD scid IL-2R[gamma]," Cytometry Part A, vol. 89, no. 9, pp. 803-815, 2016.
 N. Ali, B. Flutter, R. Sanchez Rodriguez et al., "Xenogeneic Graft-versus-Host-Disease in NOD-scid IL-2Rynull Mice Display a T-Effector Memory Phenotype," PLoS ONE, vol. 7, no. 8, p. e44219, 2012.
 B. K. Berges and M. R. Rowan, "The utility of the new generation of humanized mice to study HIV-1 infection: Transmission, prevention, pathogenesis, and treatment," Retrovirology, vol. 8, article no. 65, 2011.
 T. Lapidot, F. Pflumio, M. Doedens, B. Murdoch, D. E. Williams, and J. E. Dick, "Cytokine stimulation of multilineage hematopoiesis from immature human cells engrafted in SCID Mice," Science, vol. 255, no. 5048, pp. 1137-1141, 1992.
 J. Cohen, J. L. Vincent, and N. K. Adhikari, "Sepsis: a roadmap for future research," The Lancet Infectious Diseases, vol. 15, no. 5, pp. 581-614, 2015.
 A. K. Palucka, J. Gatlin, J. P. Blanck et al., "Human dendritic cell subsets in NOD/SCID mice engrafted with CD34+ hematopoietic progenitors," Blood, vol. 102, no. 9, pp. 3302-3310, 2003.
 S. Watanabe, S. Ohta, M. Yajima et al., "Humanized NOD/SCID/IL2Rynull mice transplanted with hematopoietic stem cells under nonmyeloablative conditions show prolonged life spans and allow detailed analysis of human immunodeficiency virus type 1 pathogenesis," Journal of Virology, vol. 81, no. 23, pp. 13259-13264, 2007.
 K. Laudanski, N. Lapko, M. Zawadka, B. X. Zhou, G. Danet-Desnoyers, and G. S. Worthen, "The clinical and immunological performance of 28 days survival model of cecal ligation and puncture in humanized mice," PLoS ONE, vol. 12, no. 7, 2017.
 M. C. Andre, A. Erbacher, C. Gille et al., "Long-term human CD34+ stem cell-engrafted nonobese diabetic/SCID/IL2Rynull mice show impaired CD8+ T cell maintenance and a functional arrest of immature NK cells," The Journal of Immunology, vol. 185, no. 5, pp. 2710-2720, 2010.
 E. Billerbeck, W. T. Barry, K. Mu, M. Dorner, C. M. Rice, and A. Ploss, "Development of human CD4+FoxP3+ regulatory T cells in human stem cell factor-, granulocyte-macrophage colony-stimulating factor-, and interleukin-3-expressing NOD-SCID IL2R[gamma]null humanized mice," Blood, vol. 117, no. 11, pp. 3076-3086, 2011.
 S. Jangalwe, L. D. Shultz, A. Mathew, and M. A. Brehm, "Improved B cell development in humanized NOD-scid IL2R[gamma]null mice transgenically expressing human stem cell factor, granulocyte-macrophage colony-stimulating factor and interleukin-3," Immunity, Inflammation and Disease, vol. 4, no. 4, pp. 427-440, 2016.
 Q. Chen, M. Khoury, and J. Chen, "Expression of human cytokines dramatically improves reconstitution of specific human-blood lineage cells in humanized mice," Proceedings of the National Acadamy of Sciences of the United States of America, vol. 106, no. 51, pp. 21783-21788, 2009.
 M. W. Melkus, J. D. Estes, A. Padgett-Thomas et al., "Humanized mice mount specific adaptive and innate immune responses to EBV and TSST-1," Nature Medicine, vol. 12, no. 11, pp. 1316-1322, 2006.
 J. S. Knibbe-Hollinger, N. R. Fields, T. R. Chaudoin et al., "Influence of age, irradiation and humanization on NSG mouse phenotypes," Biology Open, vol. 4, no. 10, pp. 1243-1252, 2015.
 Z. Hu, N. Van Rooijen, and Y.-G. Yang, "Macrophages prevent human red blood cell reconstitution in immunodeficient mice," Blood, vol. 118, no. 22, pp. 5938-5946, 2011.
 T. Yamauchi, K. Takenaka, S. Urata et al., "Polymorphic Sirpa is the genetic determinant for NOD-based mouse lines to achieve efficient human cell engraftment," Blood, vol. 121, no. 8, pp. 1316-1325, 2013.
 K. J. Lavender, R. J. Messer, B. Race, and K. J. Hasenkrug, "Production of bone marrow, liver, thymus (BLT) humanized mice on the C57BL/6 Rag2-/-[gamma]c-/-CD47-/- background," Journal of Immunological Methods, vol. 407, pp. 127-134, 2014.
 D. Rajesh, Y. Zhou, E. Jankowska-Gan et al., "Th1 and Th17 immunocompetence in humanized NOD/SCID/IL2r[gamma]null mice," Human Immunology, vol. 71, no. 6, pp. 551-559, 2010.
 E. R. Jarman, K. Perschke, E. Montermann et al., "Deficient cytokine response of human allergen-specific T lymphocytes from humanized SCID mice and reconstitution by professional antigen--presenting cells," The Journal of Allergy and Clinical Immunology, vol. 105, no. 5, pp. 967-974, 2000.
 K. Takenaka, T. K. Prasolava, J. C. Y. Wang et al., "Polymorphism in Sirpa modulates engraftment of human hematopoietic stem cells," Nature Immunology, vol. 8, no. 12, pp. 1313-1323, 2007.
 A. Kimura and T. Kishimoto, "IL-6: Regulator of Treg/Th17 balance," European Journal of Immunology, vol. 40, no. 7, pp. 1830-1835, 2010.
 Y. Ding, A. Wilkinson, A. Idris et al., "FLT3-ligand treatment of humanized mice results in the generation of large numbers of CD141+ and CD1c+ dendritic cells in vivo," The Journal of Immunology, vol. 192, no. 4, pp. 1982-1989, 2014.
 A. M. Coughlan, C. Harmon, S. Whelan et al., "Myeloid engraftment in humanized mice: Impact of granulocyte-colony stimulating factor treatment and transgenic mouse strain," Stem Cells and Development, vol. 25, no. 7, pp. 530-541, 2016.
 M. Carretta, B. de Boer, J. Jaques et al., "Genetically engineered mesenchymal stromal cells produce IL-3 and TPO to further improve human scaffold-based xenograft models," Experimental Hematology, vol. 51, pp. 36-46, 2017.
 J. Bergounioux, M. Coureuil, E. Belli et al., "Experimental evidence of b acterial colonization of human coronary microvasculature and myocardial tissue during meningococcemia," Infection and Immunity, vol. 84, no. 10, pp. 3017-3023, 2016.
 A. K. Wege, M. W. Melkus, P. W. Denton, J. D. Estes, and J. V. Garcia, "Functional and phenotypic characterization of the humanized BLT mouse model," Current Topics in Microbiology and Immunology, vol. 324, pp. 149-165, 2008.
 G. Salguero, A. Daenthanasanmak, C. Munz et al., "Dendritic cell-mediated immune humanization of mice: Implications for allogeneic and xenogeneic stem cell transplantation," The Journal of Immunology, vol. 192, no. 10, pp. 4636-4647, 2014.
 N. Lapko, M. Zawadka, J. Polosak et al., "Long-term Monocyte Dysfunction after sepsis in humanized mice is related to persisted activation of macrophage-colony stimulation factor (M-CSF) and demethylation of PU.1, and it can be reversed by blocking m-csf in vitro or by transplanting naive autologous stem cells in vivo," Frontiers in Immunology, vol. 8, no. 401, 2017.
 I. Brook, R. I. Walker, and T. J. Macvittie, "Effect of antimicrobial therapy on bowel flora and bacterial infection in irradiated mice," International Journal of Radiation Biology, vol. 53, no. 5, pp. 709-716, 1988.
 K. Melican, F. Aubey, and G. Dumenil, "Humanized mouse model to study bacterial infections targeting the microvasculature," Journal of Visualized Experiments, no. 86, 2014.
 H. D. Brightbill, S. Jeet, Z. Lin et al., "Antibodies specific for a segment of human membrane IgE deplete IgE-producing B cells in humanized mice," The Journal of Clinical Investigation, vol. 120, no. 6, pp. 2218-2229, 2010.
 S. Biswas, H. Chang, P. T. N. Sarkis, E. Fikrig, Q. Zhu, and W. A. Marasco, "Humoral immune responses in humanized BLT mice immunized with West Nile virus and HIV-1 envelope proteins are largely mediated via human CD5 + B cells," The Journal of Immunology, vol. 134, no. 4, pp. 419-433, 2011.
 S. Jaiswal, K. Smith, A. Ramirez et al., "Dengue virus infection induces broadly cross-reactive human IgM antibodies that recognize intact virions in humanized BLT-NSG mice," Experimental Biology and Medicine, vol. 240, no. 1, pp. 67-78, 2015.
 D. M. Brainard, E. Seung, N. Frahm et al., "Induction of robust cellular and humoral virus-specific adaptive immune responses in human immunodeficiency virus-infected humanized BLT mice," Journal of Virology, vol. 83, no. 14, pp. 7305-7321, 2009.
 J. Villaudy, R. Schotte, N. Legrand, and H. Spits, "Critical assessment of human antibody generation in humanized mouse models," Journal of Immunological Methods, vol. 410, pp. 18-27, 2014.
 H. Yu, C. Borsotti, J.-N. Schickel et al., "A novel humanized mouse model with significant improvement of class-switched, antigen-specific antibody production," Blood, vol. 129, no. 8, pp. 959-969, 2017.
 A. Harui, S. M. Kiertscher, and M. D. Roth, "Reconstitution of huPBL-NSG mice with donor-matched dendritic cells enables antigen-specific T-cell activation," Journal of Neuroimmune Pharmacology, vol. 6, no. 1, pp. 148-157, 2011.
 M. A. M. Willart, K. Deswarte, P. Pouliot et al., "Interleukin-1[alpha] controls allergic sensitization to inhaled house dust mite via the epithelial release of GM-CSF and IL-33," The Journal of Experimental Medicine, vol. 209, no. 8, pp. 1505-1517, 2012.
 Y. Peretz, Z. F. Zhou, F. Halwani, and G. J. Prud'homme, "In vivo generation of dendritic cells by intramuscular codelivery of FLT3 ligand and GM-CSF plasmids," Molecular Therapy, vol. 6, no. 3, pp. 407-414, 2002.
 C. Ye, J.-G. Choi, S. Abraham et al., "Human macrophage and dendritic cell-specific silencing of high-mobility group protein B1 ameliorates sepsis in a humanized mouse model," Proceedings of the National Acadamy of Sciences of the United States of America, vol. 109, no. 51, pp. 21052-21057, 2012.
 P. W. Denton, T. Nochi, A. Lim et al., "IL-2 receptor [gamma]-chain molecule is critical for intestinal T-cell reconstitution in humanized mice," Mucosal Immunology, vol. 5, no. 5, pp. 555-566, 2012.
 T. Onoe, H. Kalscheuer, N. Danzl et al., "Human Natural Regulatory T Cell Development, Suppressive Function, and Postthymic Maturation in a Humanized Mouse Model," The Journal of Immunology, vol. 187, no. 7, pp. 3895-3903, 2011.
 Y. Li, Q. Chen, D. Zheng et al., "Induction of functional human macrophages from bone marrow promonocytes by M-CSF in humanized mice," The Journal of Immunology, vol. 191, no. 6, pp. 3192-3199, 2013.
 M. Dorner, J. A. Horwitz, B. M. Donovan et al., "Completion of the entire hepatitis C virus life cycle in genetically humanized mice," Nature, vol. 501, no. 7466, pp. 237-241, 2013.
 J. G. Kuruvilla, R. M. Troyer, S. Devi, and R. Akkina, "Dengue virus infection and immune response in humanized RAG2-/-[gamma]c-/(RAG-hu) mice," Virology, vol. 369, no. 1, pp. 143-152, 2007.
 J. Mota and R. Rico-Hesse, "Humanized mice show clinical signs of dengue fever according to infecting virus genotype," Journal of Virology, vol. 83, no. 17, pp. 8638-8645, 2009.
 S. J. Libby, M. A. Brehm, D. L. Greiner et al., "Humanized nonobese diabetic-scid IL2r null mice are susceptible to lethal Salmonella Typhi infection," Proceedings of the National Acadamy of Sciences of the United States of America, vol. 107, no. 35, pp. 15589-15594, 2010.
 W. Ernst, N. Zimara, F. Hanses, D. N. Mannel, B. Seelbach-Gobel, and A. K. Wege, "Humanized mice, a new model to study the influence of drug treatment on neonatal sepsis," Infection and Immunity, vol. 81, no. 5, pp. 1520-1531, 2013.
 J. Knop, F. Hanses, T. Leist et al., "Staphylococcus aureus Infection in Humanized Mice: A New Model to Study Pathogenicity Associated with Human Immune Response," The Journal of Infectious Diseases, vol. 212, no. 3, pp. 435-444, 2015.
 T. Skirecki, J. Kawiak, E. Machaj et al., "Early severe impairment of hematopoietic stem and progenitor cells from the bone marrow caused by CLP sepsis and endotoxemia in a humanized mice model," Stem Cell Research & Therapy, vol. 6, no. 1, article no. 142, 2015.
 A. Rodewohl, J. Scholbach, A. Leichsenring, M. Koberle, and F. Lange, "Age-dependent cellular reactions of the human immune system of humanized NOD scid gamma mice on LPS stimulus," Journal of Innate Immunity, vol. 23, no. 3, pp. 258-275, 2017.
 W. Ernst, E. Kusi, S. Fill Malfertheiner, E. Reuschel, L. Deml, and B. Seelbach-Gobel, "The effect of Indomethacin and Betamethasone on the cytokine response of human neonatal mononuclear cells to gram-positive bacteria," Cytokine, vol. 73, no. 1, pp. 91-100, 2015.
 F. Schlieckau, D. Schulz, S. Fill Malfertheiner, K. Entleutner, B. Seelbach-Goebel, and W. Ernst, " A novel model to study neonatal," American Journal of Reproductive Immunology, vol. 80, no. 1, p. e12859, 2018.
 A. Prince, H. Wang, K. Kitur, and D. Parker, "Humanized mice exhibit increased susceptibility to staphylococcus aureus pneumonia," The Journal of Infectious Diseases, vol. 215, no. 9, pp. 1386-1395, 2017.
 K. Schmitt and R. Akkina, "Ultra-Sensitive HIV-1 Latency Viral Outgrowth Assays Using Humanized Mice," Frontiers in Immunology, vol. 9, 2018.
 I. R. Turnbull, J. J. Wlzorek, D. Osborne, R. S. Hotchkiss, C. M. Coopersmith, and T. G. Buchman, "Effects of age on mortality and antibiotic efficacy in cecal ligation and puncture.," Shock (Augusta, Ga.), vol. 19, no. 4, pp. 310-313, 2003.
 K.-H. Song, S.-Y. Jung, S.-H. Kho et al., "Effects of low-dose irradiation on mice with Escherichia coli-induced sepsis," Toxicology and Applied Pharmacology, vol. 333, pp. 17-25, 2017.
 M. Schouten, W. J. Wiersinga, M. Levi, and T. van der Poll, "Inflammation, endothelium, and coagulation in sepsis," Journal of Leukocyte Biology, vol. 83, no. 3, pp. 536-545, 2008.
 G. Pena, B. Cai, L. Ramos, G. Vida, E. A. Deitch, and L. Ulloa, "Cholinergic regulatory lymphocytes re-establish neuromodulation of innate immune responses in sepsis," The Journal of Immunology, vol. 187, no. 2, pp. 718-725, 2011.
Krzysztof Laudanski [iD], (1) Michael Stentz, (2) Matthew DiMeglio [iD], (3) William Furey, (3) Toby Steinberg, (1) and Arpit Patel (1)
(1) Department of Anesthesiology and Critical Care, University of Pennsylvania, Philadelphia, PA (19104), USA
(2) Department of Anesthesiology and Intensive Care, Emory University, Atlanta, GA (30322), USA
(3) Philadelphia College of Osteopathic Medicine, Philadelphia, PA (19131), USA
Correspondence should be addressed to Krzysztof Laudanski; email@example.com
Received 22 May 2018; Revised 17 July 2018; Accepted 13 August 2018; Published 2 September 2018
Academic Editor: Jean-Marc Cavaillon
Table 1: Pitfalls of humanized models and the means to compensate them. Problem definition Impact on progress in Ways to overcome research limitations The artificial Increased Creating a more condition of housing susceptibility of the realistic environment animals to infection, for housing decreased immunity, Clinical relevance of Separation of the Developing more septic model in research outcome from clinically relevant general clinical reality. model by introducing fluid resuscitation and antibiotics, comparisons between humanized and non- humanized animals. Homogeneity of animals Decrease robustness Increasing diversity, of the findings. developing models with different strains, engaging in cross-species research, grafting with different stem cells. Preservation of mice Incomplete or Development of more native immune system inefficient grafting, profoundly GVHD, the emergence immunosuppressed of lymphomas hosts, eradication ofthe residual immune system, knock-out of SIRP-[alpha]. Lack of supportive Inefficient grafting, Supplementation of human cytokine inefficient cytokine human cytokines via environment network and immune various means system regulation Poor recovery of Incomplete Introduction of HIS, certain leukocyte restoration of the BLT, MITGR models, population immune system, supplementation of ineffective and human cytokines via clinically irrelevant genetic engineering, responses Immunoglobulin Inability to mimic Development of human switching humoral responses IL-6 producing mice, introduction of additional cytokine modification Functional immaturity The inappropriate Supplementation of of human leukocytes response, adequate cytokine difficulties in environment, ex vivo translating cell maturation, and supplementation Poor inter-organ Difficulty mimicking Additional communication complex interaction transplantation to between organs in better mimic inter/ sepsis organ interaction in the autologous/ allogeneic system, the introduction of human intestinal flora Problem definition Ref. The artificial [2, 4, 7,12] condition of housing Clinical relevance of [2, 8,10] septic model in general Homogeneity of animals Preservation of mice [13,16,17,19, 24, 28, 30, 34, 4547] native immune system Lack of supportive [31, 37, 48-52] human cytokine environment Poor recovery of [10,16,19,22, certain leukocyte 24, 25, 33, 34, population 36, 44, 53] Immunoglobulin [38, 54] switching Functional immaturity [35, 51, 55] of human leukocytes Poor inter-organ [35, 54-57] communication