|Home | About | Journals | Submit | Contact Us | Français|
Small animal models such as mice have been extensively used to study human disease and to develop new therapeutic interventions. Despite the wealth of information gained from these studies, the unique characteristics of mouse immunity as well as the species specificity of viral diseases such as human immunodeficiency virus (HIV) infection led to the development of humanized mouse models. The earlier models involved the use of C. B 17 scid/scid mice and the transplantation of human fetal thymus and fetal liver termed thy/liv (SCID-hu) 1, 2 or the adoptive transfer of human peripheral blood leukocytes (SCID-huPBL) 3. Both models were mainly utilized for the study of HIV infection.
One of the main limitations of both of these models was the lack of stable reconstitution of human immune cells in the periphery to make them a more physiologically relevant model to study HIV disease. To this end, the BLT humanized mouse model was developed. BLT stands for bone marrow/liver/thymus. In this model, 6 to 8 week old NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) immunocompromised mice receive the thy/liv implant as in the SCID-hu mouse model only to be followed by a second human hematopoietic stem cell transplant 4. The advantage of this system is the full reconstitution of the human immune system in the periphery. This model has been used to study HIV infection and latency 5–8.
We have generated a modified version of this model in which we use genetically modified human hematopoietic stem cells (hHSC) to construct the thy/liv implant followed by injection of transduced autologous hHSC 7, 9. This approach results in the generation of genetically modified lineages. More importantly, we adapted this system to examine the potential of generating functional cytotoxic T cells (CTL) expressing a melanoma specific T cell receptor. Using this model we were able to assess the functionality of our transgenic CTL utilizing live positron emission tomography (PET) imaging to determine tumor regression (9).
The goal of this protocol is to describe the process of generating these transgenic mice and assessing in vivo efficacy using live PET imaging. As a note, since we use human tissues and lentiviral vectors, our facilities conform to CDC NIH guidelines for Biosafety Level 2 (BSL2) with special precautions (BSL2+). In addition, the NSG mice are severely immunocompromised thus, their housing and maintenance must conform to the highest health standards (http://jaxmice.jax.org/research/immunology/005557-housing.html).
Fresh fetal tissue or tissue shipped overnight on ice from various organ procurement agencies can be used. Often fetal tissue is not sterile, as evidenced by the production of 1,000 bacterial colonies on blood agar per milliliter of surrounding medium. We routinely wash the tissue twice in 40 ml of sterile PBS. While this does not remove all bacteria, it can make a difference between a successful transplant series and an outcome in which most of the recipient mice succumb to bacterial infection. As an added step, to further disinfect the tissue, we culture the cell suspension in the presence of antibiotics.
The goal of this step is to transplant a human fetal thymus/liver organoid under the NSG mouse kidney capsule. This organoid better mimics the process of human T cell selection and maturation processes as the human hematopoietic stem cells will use the human thymus and not the mouse as the site for their differentiation to different T cell and other lymphoid lineages. The use of CD34− cells in the transplantation process is to re-generate the fetal liver stroma and allowing for better transplantation and growth of the implant. The age of NSG mice used is 6–8 weeks old.
The goal of this step in the generation of BLT mice is to populate the moue bone marrow with human hematopoietic stem cells. The transplanted implant from procedure (2) is not sufficient to support full reconstitution of the human immune system in these mice. During the secondary transplant, we sub-lethally irradiate the mice to deplete murine bone marrow cells thus generating “space” for the implantation of the human CD34+ cells.
For our studies we examine glucose uptake ([18F]-fluorodeoxyglucose ([18F]FDG) thus mice are fasted 4–6 hr prior to imaging. MicroPET/CT scans are done using the microPET Inveon scanner (Siemens Preclinical Solutions) and MicroCATII CT scanner (Siemens PreclinicalSolutions). Image analysis is done using OsiriX (Pixmeo, Switzerland) software. The goal is to measure the metabolic activity of the tumor and ultimately to use PET imaging as an alternative to physically measuring tumor regression. As seen in Figure 2, we have encountered tumors that based on physical appearance and size are not targeted but live PET imaging revealed extensive tissue necrosis. This methodology can serve as a more sensitive and accurate indicator of tumor regression.
A flow chart of the transplantation process is shown in Figure 1A. A picture of the thy/liv implant is shown in Figure 1B. The thymic tissue develops normally and has a physiological distribution of human CD4 and CD8 T cells. Following reconstitution, the animals carry a human immune system with normal distribution of CD4, CD8 T cells and other immune cell lineages.
The discrepancy between tumor size and live tissue is shown in Figure 2. While the CT scan (grey area) indicated a large tumor growth, in vivo PET imaging showed that it was mostly necrotic and scar tissue (Figure 2). This underscores the utility of PET imaging as a more sensitive and accurate way to assess tumor regression and clearance.
The modified BLT humanized mouse model coupled with in vivo PET imaging are powerful tools to study chronic human diseases. This system takes the BLT mouse model and advances it beyond the limited scope for HIV research. In addition, it is a great system in which we can examine various gene therapy protocols as well as diagnostic techniques before they can reach the clinical setting. The latter coupled with the low cost of using mice versus primates makes this a very useful model.
The PET imaging technology allowed us to assess the efficacy of our approach. If we relied exclusively on physical measurements of the tumor, we would have underestimated the potency of the antitumor response generated by our transgenic T cells. The extensive scarring and necrotic tissue gave the appearance of a large tumor, which in reality was dead tissue.
In conclusion, the utility of the modified BLT mouse model can be extended to other disease models. While some disadvantages still persist such as the shorter lifespan of mice, this can be a very strong tool to in vivo assess many aspects of human immunity, test and develop novel therapeutic interventions.
We would like to thank Alvin Welch and Larry Pang for their technical assistance. This work was funded in part by the National Institutes of Health (NIH) award P50 CA086306, the California Institute for Regenerative Medicine (CIRM) grants RC1-00149-1 and RS1-00203-1; CIRM New Faculty Award RN2-00902-1, the Caltech-UCLA Joint Center for Translational Medicine, UCLA Center for AIDS Research (CFAR) NIH/NIAID AI028697, the UCLA AIDS Institute, the CIRM Tools and Technology Award RT1-01126, and the UC Multicampus Research Program and Initiatives from the California Center for Antiviral Drug Discovery (number MRPI-143226).
A complete version of this article that includes the video component is available at http://dx.doi.org/10.3791/4181.
No conflicts of interest declared.