This protocol provides a reliable method to establish humanized mice with both human immune system and liver cells. Dual reconstituted immunodeficient mice achieved via intrasplenic injection of human hepatocytes and CD34+ hematopoietic stem cells are susceptible to human immunodeficiency virus-1 infection and recapitulate liver damage as observed in HIV-infected patients.
Despite the increased life expectancy of patients infected with human immunodeficiency virus-1 (HIV-1), liver disease has emerged as a common cause of their morbidity. The liver immunopathology caused by HIV-1 remains elusive. Small xenograft animal models with human hepatocytes and human immune system can recapitulate the human biology of the disease's pathogenesis. Herein, a protocol is described to establish a dual humanized mouse model through human hepatocytes and CD34+ hematopoietic stem/progenitor cells (HSPCs) transplantation, to study liver immunopathology as observed in HIV-infected patients. To achieve dual reconstitution, male TK-NOG (NOD.Cg-Prkdcscid Il2rgtm1Sug Tg(Alb-TK)7-2/ShiJic) mice are intraperitoneally injected with ganciclovir (GCV) doses to eliminate mouse transgenic liver cells, and with treosulfan for nonmyeloablative conditioning, both of which facilitate human hepatocyte (HEP) engraftment and human immune system (HIS) development. Human albumin (ALB) levels are evaluated for liver engraftment, and the presence of human immune cells in blood detected by flow cytometry confirms the establishment of human immune system. The model developed using the protocol described here resembles multiple components of liver damage from HIV-1 infection. Its establishment could prove to be essential for studies of hepatitis virus co-infection and for the evaluation of antiviral and antiretroviral drugs.
Since the advent of antiretroviral therapy, there has been a substantial decrease in deaths related to HIV-1 monoinfection. However, liver disease has emerged as a common cause of morbidity in HIV-infected patients1,2. Coinfections of hepatitis viruses with HIV-1 infection are more common, accounting for 10% – 30% of HIV-infected persons in the United States3,4,5.
The host-specificity of HIV-1 and hepatitis viruses limits the utility of small animal models to study human-specific infectious diseases or to investigate multiple aspects of HIV-1-associated liver pathogenesis. Immunodeficient mice that permit the engraftment of human cells and/or tissues (termed humanized mouse models) are acceptable animal models for preclinical studies6,7,8. Since the introduction of humanized mice in the early 2000s, multiple preclinical studies of cholestatic human liver toxicity, human-specific pathogens, including HIV-1 and HIV-associated neurocognitive disorders, Epstein Barr virus, hepatitis, and other infectious diseases, have been investigated in these mice6,9,10,11. Multiple mouse models for CD34+ HSPCs and/or human hepatocyte transplantation have long been developed and have improved over time to study the disease pathogenesis of Hepatitis B virus (HBV)-associated liver disease12,13,14. Several models for HSPC and human hepatocyte (HEP) transplantation are based on strains, known as NOG (NOD.Cg-Prkdcscid Il2rgtm1Sug/JicTac)8,13, NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ)15, Balb/C-Rag2-/- γc-/- (Rag2tm1.1Flv Il2rgtm1.1Flv/J)12, and fah-/- NOD rag1-/-il2rγnull mouse16. However, each model has its own advantages and limitations; for example, AFC8 dual humanized mice for HEPs and human stem cells (HSCs) on a Balb/C-Rag2-/- γc-/- background enables the successful engraftment of immune cells and HSCs, but there is an absence of an antigen-specific T- and B-cell response in this model12. The major concerns in reconstituting double humanized mice include suboptimal engraftment, a lack of suitable models to support different tissues, mismatched conditions, immune rejection, or graft-versus-host disease (GVHD), and technical difficulties, such as risky manipulations with newborns and high mortality rates due to metabolic abnormalities13.
Although humanized mice have been used for HIV research for many years17,18,19, the use of humanized mice to study liver damage caused by HIV-1 has been limited20. We previously reported the establishment of a dual humanized TK-NOG mouse model and its application in HIV-associated liver disease8. This model shows the robust engraftment of liver and immune cells and recapitulates HIV infection pathogenesis. This discussion presents a detailed protocol, including the most critical steps in the transplantation of human hepatocytes. A description of the HSPCs required for a successful engraftment of HEPs and the establishment of a functional immune system in TK-NOG mice is also presented. The use of these mice to study HIV-associated liver immunopathogenesis is detailed. TK-NOG male mice carrying a liver-specific herpes simplex virus type 1 thymidine kinase (HSV-TK) transgene are used. Mouse liver cells expressing this transgene can easily be ablated after a brief exposure to a nontoxic dose of GCV. Transplanted human liver cells are stably maintained within the mouse liver without exogenous drugs21. The mice are also preconditioned with nonmyeloablative doses of treosulfan to create a niche in the mouse bone marrow for human cells8. Immunodeficient TK-NOG mice are intrasplenically injected with HEPs and multipotent HSPCs. The mice are then regularly monitored for blood and liver reconstitution by blood immunophenotyping and measurements of serum human-albumin levels, respectively. Mice with a successful reconstitution of more than 15% for both human immune cells and HEPs are intraperitoneally injected with HIV-1. The effect of HIV on the liver can be assessed as early as 4 – 5 weeks postinfection. It is critical to note that, because HIV-1 is used, all necessary precautions must be taken while handling the virus and injecting it into mice.
This protocol has been approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Nebraska Medical Center.
NOTE: Obtain approval from the local IACUC before performing experiments on animals.
1. Processing of Umbilical Cord Blood and the Isolation of Human HSPCs
2. Preparation of Human Hepatocytes for Transplantation
3. Animal Handling, Screening, Genotyping, and Treatment for Human HSPC and Hepatocyte Transplantation
4. Engraftment Validation of the Human Liver by ELISA and the Human Immune System by Flow Cytometry
NOTE: Evaluate the reconstitution of the human liver and immune system monthly, starting 1 month posttransplantation by enzyme-linked immunosorbent assay (ELISA) and flow cytometry, respectively.
5. HIV Infection of TK-NOG Mice and Its Effect on the Human Liver and Immune System
The establishment of a dual humanized mouse model with human liver and immune cells can be easily monitored at each step with very simple ELISA and flow cytometry, respectively. Flow cytometry is regularly performed to evaluate the development of a functional immune system and to see the effect of HIV infection on immune cells. In dual humanized mice, the development of functional immune cells can range from 15% to 90% of the lymphocyte gate. Representative subsets of immune cells are shown in dot plots (Figure 3). For the evaluation of the engraftment of human hepatocytes, ELISA for human-specific albumin levels is performed monthly on mouse serum. Mice engrafted with both HSPCs and HEPs show human-specific albumin levels ranging from ~7 µg/mL to 377 µg/mL at one month, continuing to grow over the time of observation (6 months) (Figure 4). The effect of HIV infection on human immune cells in the blood of dual humanized mice is monitored by flow cytometry and on HEPs in the liver by human-specific albumin ELISA. By 5 weeks, HIV-1 causes a decrease in human albumin levels in the serum, as assessed by ELISA, and there is a depletion of human CK18+ hepatocytes in the liver sections of dual humanized mice, as evaluated by immunohistochemistry (Figure 5). A lower ratio of CD4:CD8 is typically observed, by flow cytometry, in the blood and liver of HIV-infected mice, compared to levels noted in the same mouse before infection (Figure 6). All reagents and materials important for the protocol are discussed in the Table of Materials.
Figure 1: Schematic of the enrichment of CD34+ cells from cord blood. (A) Cord blood is layered on lymphocytes separation medium (LSM) and centrifuged to isolate buffy coat. (B) LS columns are placed on a magnetic stand and rinsed with BSA buffer, followed by adding buffy coat. Cells positive for CD34 are trapped in the columns, and CD34– cells are eluted in separate tubes. Trapped CD34+ cells in column resins are plunged with a plunger, and the cells are collected in a new tube. Please click here to view a larger version of this figure.
Figure 2: Schematic view of the experimental design for the dual reconstitution of humanized liver and immune system mice, followed by HIV-1 infection. TK-NOG mice are injected with ganciclovir (GCV) at a dose of 6 mg/kg, 2x a day, on day -7 and day -5, followed by a treosulfan injection on days -3, -2 and -1. To screen the mice for the transplantation (Tx), an alanine aminotransferase (ALT) assay is performed one day before the surgery, and mice with ALT levels of >200 and <600 U/L are selected. After transplantation, the mice are checked for a reconstitution of the human immune system by flow cytometry (FACS) and for liver reconstitution by assessing their albumin level using ELISA. The mice are infected with HIV-1 5 weeks before they are sacrificed. Please click here to view a larger version of this figure.
Figure 3: Flow cytometry analysis gating strategy for the human cell distribution of blood. (A) First, lymphocytes are gated on whole blood based on FSC-A and SSC-A. (B) Single cells are gated on lymphocytes. (C) Human CD45+ leukocytes are gated on single cells using mouse CD45 and human CD45. (D) CD3+ T cells and CD19+ B cells are identified on gated CD45+ human leukocytes. (E) CD4+ T helper cells and CD8+ cytotoxic T cells are identified in gated CD3+ T cells. (F) CD14+ monocytes are gated from human CD45+ leukocytes. The results represented here are from one mouse transplanted with dual human hepatocytes and HSPCs. Please click here to view a larger version of this figure.
Figure 4: Albumin concentration is measured by ELISA in the serum of dual humanized mice. The mice are transplanted with both human hepatocytes (HEPs) and CD34+ hematopoietic stem/progenitor cells (HSPCs) (n = 11). Serum is collected at different times at 1, 4, and 6 months posttransplantation, and dilutions are made to adjust the unknown sample concentrations in the range of standards. Each symbol represents an individual mouse value. The results represent the median, as well as individual values. * P < 0.05, by one-way ANOVA. This figure has been modified from Dagur et al.8. Please click here to view a larger version of this figure.
Figure 5: Effect on HIV-1 on albumin levels in serum and the depletion of CK18+ human hepatocytes in the liver of dual humanized mice. (A) Albumin concentrations are monitored in uninfected mice (n = 9) transplanted with both human HEPs and HSPCs at 1 and 4 months. The mice are infected (n = 10) with HIV at 4 – 5 months posttransplantation and sacrificed 5 weeks postinfection. Each symbol represents an individual mouse value. The results represent the median, as well as individual values. * P < 0.05, by one-way ANOVA. This figure has been modified from Dagur et al.8. (B) Five-micron liver sections from uninfected (HEPs + HSPCs, left panel) and HIV-infected TK-NOG mice (HEPs + HSPCs + HIV, right panel) are fixed, paraffin embedded, and stained for anti-human cytokeratin-18 (CK18) antibody. HIV-1 causing a depletion of CK18+ hepatocytes is evidenced by a less occupied area by the CK18+ human cells. The results represented here are from one uninfected and one HIV-infected mouse transplanted with dual human hepatocytes and HSPCs. Scale bars = 100 µm. Please click here to view a larger version of this figure.
Figure 6: Ratio of CD4+ cells to CD8+ T cells in peripheral blood and in the liver of dual reconstituted uninfected and HIV-1-infected mice. For dual reconstituted uninfected mice: closed circle; HEPs + HSPCs; blood n = 7; liver n = 6. For HIV-1 infected mice: open circles; HEPs + HSPCs + HIV; blood n = 10; liver n = 11. The results represent the median, as well as individual values. * P < 0.05, by one-way ANOVA test between HIV-infected and uninfected mice. This figure has been modified from Dagur et al.8. Please click here to view a larger version of this figure.
The liver is compromised and damaged in HIV-infected patients24. Experimental small animal models for studying human liver diseases in the presence of HIV-1 is extremely limited, despite the availability of a few cotransplanted animal models with CD34+ HSPCs and hepatocytes7,12,25. In in vitro experiments, hepatocytes are shown to have low-level HIV-1 infection26. Humanized mice that carry both types of human cells are a desirable model. The liver of mice reconstituted with only human immune system has been shown to be affected by HIV infection under the experimental depletion of human regulatory T cells20,27. However, the difference in immune and functional properties of mouse and human hepatocytes may underline the differences in their responses to HIV-1 and immune cells. In this review, a protocol is described to reconstitute both human immune system and liver and to address HIV-1-associated liver immunopathology, as observed in HIV-1-infected patients. TK-NOG male mice were selected due to their liver-selective high mRNA expression of HSV-TK transgene and the susceptibility of GCV toxicity to mouse transgenic liver21. Moreover, they can be maintained for long periods after transplantation without the use of exogenous drugs and do not develop spontaneous systemic diseases28. To establish human immune system and liver reconstitution, ablation of the mouse immune system and damage to mouse-specific liver cells are required and achieved using nonmyeloablative doses of treosulfan and GCV, as shown previously in TK-NOG male mice13,23. Mice are injected with GCV and treosulfan at the age of 6 – 8 weeks, as the expression of transgene and GCV-induced hepatic injury as assessed by ALT levels are optimal, then, for providing niche-to-transplanted human cells21. Mice showing ALT levels of >200 U/L, but <600 U/L, are usually selected for transplantation. Mice showing ALT levels of >600 U/L are at a greater risk of death as human hepatocytes are not able to rescue the damaged mouse liver function.
Currently, dual humanization is shown by the transplantation of human CD34+ HSPCs and fetal liver cells; however, the manipulation of newborn animals creates technical problems13,14. HSPCs can be derived or isolated from multiple sources, such as fetal liver cells (FLCs), embryonic stem cells (ESCs), and CB. However, ethical issues constrain the use of ESCs and FLCs. CB has no such restriction and is a most useful alternative to obtain HSPCs, as well as being a precious source of primitive hematopoietic stem and progenitor cells that can reconstitute the functional immune system. Cord blood should not be older than one day when used to isolate HSPCs, as the yield of HSPCs is highly affected by age. The purity of the isolated HSPCs needs to be checked before cryopreserving the cells. The cross-contamination of CD3+ T cells is avoided, as it may lead to systemic mouse graft-versus-host disease and acute allorejection of HEPs while transplanting with mismatched cells.
Commercially available hepatocytes were used as a source for liver reconstitution8,13. Adult hepatocytes are preferred for establishing liver reconstitution due to their increased efficiency in engraftment and sustainability for a long period of time29.
The presence of human immune system in a mouse model increased ALB levels, as shown previously30,31. However, the efficiency of hepatocytes and immune system reconstitution may vary with different sources of donor cells and depend on the recipient mouse. So, each mouse needs to be assessed for engraftment, and the most critical part is to utilize the antibodies or reagent that are human-specific and do not cross-react with mouse cells. The human-specific reagents and antibodies used in the study presented here are detailed in the Table of Materials. If antibodies other than provided in the Table of Materials are used for the study, be sure to check for the human specificity.
The optimal condition would be the transplantation of syngeneic cells; however, that is technically difficult to achieve. Wherever possible, HSPCs and hepatocytes should be pooled from donors with partially matched human leukocyte class-1 antigens (like HLA-A2).
To screen mice for HIV studies, blood is drawn at multiple time points to determine the optimal immune and liver reconstitution; flow cytometry and ELISA are preferred as they can be performed with only a little amount of blood. Blood cells and serum from the same sample could be used for flow cytometry and ELISA, respectively. It is important to make proper dilutions of serum at each time point (1,000 – 40,000 range) to evaluate ALB levels so that the unknown concentrations can be brought within the range of standard concentrations (kit range: 6.25 – 400 ng/mL).
Proinflammatory cytokines in response to HIV-1 infection in the presence of human immune system can also be useful in addressing the interaction of hepatocytes and immune cells. The model is useful for showing the immunopathogenesis of HIV-1-induced liver disease, given that it recapitulates liver damage in the same manner as in humans, evidenced by a low ratio of CD4:CD8, a decrease in ALB levels, human hepatocyte death, and liver immune activation. The model also has some limitations, such as a low level of cytotoxic T cells activity and impaired immunoglobulin class switching. Due to the presence of both human immune system and liver, the model presented here is promising for studi coinfections of HIV-1 and hepatitis viruses, chronic hepatitis infection (to clarify the mechanisms of the anti-hepatitis immune response), and as a cirrhosis model.
The authors have nothing to disclose.
This work was supported by the National Institute of Health grant R24OD018546 (to L.Y.P. and S.G.). The authors would like to thank Weizhe Li, Ph.D., for the help in surgical procedures, Amanda Branch Woods, B.S., Yan Cheng for immunohistology, UNMC flow cytometry research facility members Director Phillip Hexley, Ph.D., Victoria B. Smith, B.S., and Samantha Wall, B.S., UNMC advanced microscopy core facility members Janice A. Taylor, B.S., and James R. Talaska, B.S., for the technical support. The authors acknowledge Drs. Mamoru Ito and Hiroshi Suemizu from CIEA for providing TK-NOG mice and Dr. Joachim Baumgart for providing treosulfan. The authors thank Dr. Adrian Koesters, UNMC, for her editorial contribution to the manuscript.
27G1/2" needles | BD biosciences | 305109 | |
30G1/2" needles | BD biosciences | 305106 | |
5 mL polystyrene round-bottom tube 12 x 75 mm style | Corning | 352054 | |
BD 1 mL Tuberculin Syringe Without Needle | BD biosciences | 309659 | |
BD FACS array bioanalyzer | BD Biosciences | For purity check of eluted CD34+ cells | |
BD FACS array software | BD Biosciences | Software to analysis acquired CD34+ cell on FACS array | |
BD FACS lysing solution | BD Biosciences | 349202 | To lyse red blood cells |
BD LSR II | BD Biosciences | Instrument for acquisiton of flow cytometry samples | |
BD Vacutainer Plastic Blood Collection Tube | BD biosciences | BD 367874 | To collect Cord blood |
Bovine Serum Albumin | Sigma-aldrich | A9576 | |
Buprenorphine | Controlled substance and pain-killer | ||
CD14-PE | BD Biosciences | 555398 | Specific to human |
CD19-BV605 | BD Biosciences | 562653 | Specific to human |
CD34 MicroBead Kit, human | Miltenyi Biotec | 130-046-702 | For isoation of CD34+ HSPC |
CD34-PE, human | Miltenyi Biotec | 130-081-002 | Antibody used for purity check of eluted CD34+ cells |
CD3-AF700 | BD Biosciences | 557943 | Specific to human |
CD45-PerCPCy5.5 | BD Biosciences | 564105 | Specific to human |
CD4-APC | BD Biosciences | 555349 | Specific to human |
CD8-BV421 | BD Biosciences | 562428 | Specific to human |
Cell counting slides | Bio-rad | 1450015 | |
ChargeSwitch gDNA Mini Tissue Kit | Thermofisher scientific | CS11204 | for extraction of genomic DNA from ear piece |
Cobas Amplicor system v1.5 | Roche Molecular Diagnostics | bioanalyzer to measure viral load | |
Cotton-tipped applicators | McKesson | 24-106-2S | |
Cytokeratin-18 (CK18) | DAKO | M7010 | Specific to human |
DMSO (Dimethyl sulfoxide) | Sigma-aldrich | D2650-5X5ML | |
Extension set Microbore Slide Clamp(s) Fixed Male Luer Lock. L: 60 in L: 152 cm PV: 0.55 mL Fluid Path Sterile | BD biosciences | 30914 | Attached to dispensing pippet and to load with HSPC and HEP suspesion |
FACS Diva version 6 | BD Biosciences | flow cytometer software required for acqusition of sample | |
Fetal Bovine Serum (FBS) | Gibco | 10438026 | |
FLOWJO analysis software v10.2 |
FLOWJO, LLC | flow cytometry analysis software | |
Ganciclovir | APP Pharmaceuticals, Inc. | 315110 | Prescripition drug |
Greiner MiniCollect EDTA Tubes | Greiner bio-one | 450475 | |
Hepatocytes thawing medium | Triangle Research Labs | MCHT50 | |
Horizon Open Ligating Clip Appliers | Teleflex | 537061 | To hold the ligating clips |
Hospira Sterile Water for Injection | ACE surgical supply co. Inc. | 001-1187 | For dilution of Buprenorphine (pain-killer) |
Human Albumin ELISA Quantitation Set | Bethyl laboratories | E80-129 | For assesing human albumin levels in mouse serum |
Human hepatocyte | Triangle Research Labs | HUCP1 | Cryopreserved human hepatocytes, induction qualified |
Iris Scissors, Straight | Ted Pella, Inc. | 13295 | |
Lancet | MEDIpoint | Goldenrod 5 mm | |
LS columns | Miltenyi Biotec | 130-042-401 | Used to entrap CD34+ microbeads (positive selection) |
Lymphocyte Separation Medium (LSM) | MP Biomedicals | 50494 | For isoation of lymphocytes from peripheral blood |
MACS MultiStand | Miltenyi Biotec | 130-042-303 | holds Qudro MACS seperator and LS columns |
McPherson-Vannas Micro Dissecting Spring Scissors | Roboz Surgical Instrument Co. | RS-5605 | Used to make an incision on skin to expose spleen |
Micro Dissecting Forceps | Roboz Surgical Instrument Co. | RS-5157 | to hold and pull out spleen from peritoneal cavity |
mouse CD45-FITC | BD Biosciences | 553080 | mouse-specific |
PBS (Phosphate Buffered Saline) | Hyclone | SH30256.02 | |
Qudro MACS separator | Miltenyi Biotec | 130-090-976 | holds four LS columns |
RPMI 1640 medium | Gibco | 11875093 | |
StepOne Plus Real Time PCR | Applied Biosystems | Instrument used to genotype | |
Stepper Series Repetitive Dispensing Pipette 1ml | DYMAX CORP | T15469 | Used to dispense HSPC and HEP supension in controlled manner |
Suturevet PGA synthetic absorbale suture | Henry Schein Animal Health | 41178 | Suturing of skin and peritoneum |
TaqMan Gene Expression Master Mix | Thermofisher scientific | 4369016 | |
TC20 automated cell counter | Bio-rad | 1450102 | |
TK-NOG mice | Provided by the Central Institute for Experimental Animals (CIEA, Japan; Drs. Mamoru Ito and Hiroshi Suemizu) | ||
Treosulfan | Medac GmbH | Provided by Dr. Joachim Baumgart (medac GmbH) | |
Trypan Blue | Bio-rad | 1450022 | |
Vannas-type Micro Scissors, Straight, 80mm L | Ted Pella, Inc. | 1346 | Used to make an incision on skin to expose spleen |
Weck hemoclip traditional titanium ligating clips | Esutures | 523700 | To ligate the spleen post-injection |