Scientists report on the development of a mouse model that recapitulates the immune system of a single adult human. In contrast with existing humanized mouse models with immune systems derived from transplanted fetal hematopoietic stem cells (HSCs), the immune system of the model described by the Massachusetts General Hospital and Columbia University Medical Center team is derived from a relatively few adult human HSCs. These could effectively be taken from any human volunteer or patient.
Megan Sykes, M.D., and colleagues, say the the resulting "personalized immune" (PI) mice generated a robust and diverse repertoire of fully functional T cells that were self-tolerant, and exhibited immune responses that mimicked those of the adult CD34+ cell donor. The team reports on its approach and initial analyses of a diabetic PI mice in Science Translational Medicine, in a paper titled “A Model for Personalized in Vivo Analysis of Human Immune Responsiveness.”
They claim the ability to generate PI mice will provide valuable new insights into the basis of immune disorders such as diabetes, and also allow the generation of models for evaluating an individual’s likely response to drug or immunotherapies, analyse the immunological and genetic basis of disease progression, and aid in the development of new treatments for immune disorders.
Research into the genetic basis of human immune diseases has to date largely been limited to studies in peripheral blood mononuclear cells (PBMCs). More recently, mouse models with human immune systems have been generated using grafts of human fetal thymus tissue seeded with fetal HSCs. These animals produce human T cells, B cells and myeloid cells, and mount antigen-specific immune responses. However, in order to be relevant to the study of specific human immune-mediated disorders, the immune systems of these engineered mice need to be reconstituted from adult HSCs, not just fetal cells.
Unfortunately, achieving human immune reconstitution and function in mice using adult HSCs hasn’t proven easy. Adult HSCs aren’t available in large quantities from volunteers and don’t engraft as efficiently as fetal CD34+ cells in immunodeficient mice, the researchers note. Moreover, even when large quantities of adult HSCs are available, they can be rejected by allogeneic thymocytes already present in the fetal thymus grafts.
The researchers report overcoming such issues and generating a truly personalized immune mouse that sustains peripheral reconstitution of T cells and antigen-presenting cells (APCs) from small numbers of adult allogeneic bone marrow CD34+ cells seeded into fetal human thymus grafts transplanted in the animal’s kidney capsule. Their successful approach hinged on depletion of pre-existing thymocytes in the fetal thymus grafts. This was necessary to prevent both rejection of the administered allogeneic CD34+ cells and the development of xenogeneic graft-versus host disease (GVHD) mediated by mature T cells derived from the fetal graft.
The investigators achieved this by transplanting human fetal thymus tissue that had been frozen and then thawed prior to grafting. To test whether this technique would support human immune reconstitution, sublethally irradiated NSG (NOD/SCID Gamma) mice received grafts of cryopreserved-thawed thymus tissue and an intravenous infusion of allogenic adult CD34+ cells. The animals were in addition treated with an anti-human CD2 monoclonal antibody to further ensure depletion of mature T cells originating from the graft.
Encouragingly, all the mice achieved human B cell and monocyte chimerism by six weeks, and generated peripheral T cells, which appeared by six weeks and peaked at 10–30% of PBMCs at 16 weeks after the CD34+ cell infusion. Animals that were subsequently sacrificed demonstrated markedly enlarged thymus grafts, and evidence of robust CD4/CD8 T cell differentiation. Importantly, none of the animals receiving the cryopreserved-fetal thymus tissue and CD2 antibodies developed evidence of GVHD, even after 20 weeks.
Having demonstrated that their approach supports the reconstitution of a human immune system, the team generated PI mice using adult CD34+ cells isolated by bedside bone marrow aspiration from a healthy human volunteer and a type 1 diabetes patient. Sublethally irradiated NSG mice received 1.8 × 105 adult CD34+ cells each plus a cryopreserved and thawed human fetal thymus graft and anti-human CD2 mAb. Control irradiated mice received CD34+ cells without thymus tissue.
Human chimerism was detected by week six in the thymus-grafted animals that received CD34+ cells from either the healthy volunteer or the diabetic patient, and this peaked at up to 80%. Moreover, recipients of thymus grafts plus intravenous CD34+ cells developed substantial CD3+ cell levels by eight weeks, and also generated CD19+ and CD14+ cells. Encouragingly, equivalent results were obtained when the experiments were repeated three times in multiple animals. In contrast, the control mice had significantly delayed T cell reconstitution.
T cell function in thymus-grafted mice reconstituted with the volunteer donor CD34+ cells was tested by subsequently grafting the animals with allogeneic human and pig skin. The grafts were rapidly rejected in all animals. In contrast, naïve untreated NSG mice were able to accept both allogeneic human and xenogeneic skin grafts, with no evidence of rejection. Importantly, mixed lymphocyte reaction assays using purified T cells from the spleens and lymph nodes of the PI mice showed self-tolerance and strong responses to allogeneic human stimulators.
Notably, while immune recognition of the mice mimicked that of the adult CD34+ cell donor, the T cell phenotypes were more predominantly “naïve” than those of the adult donors. “Thus, a rejuvenated version of the adult donors immune system is generated in PI mice,” the investigators note.
Further evaluation of PI animals with immune systems derived from either the diabetic or healthy volunteer confirmed the presence of regulatory T cells (Tregs) in the thymus grafts and in the peripheral blood, and a diverse repertoire of T cells equivalent to that seen in healthy humans. Interestingly, the blood of diabetic- and volunteer-derived mice demonstrated similar proportions of Tregs, which contrasts with some previous research indicating that Treg numbers are reduced in the blood of type 1 diabetics, the authors note. However, while both naive and memory T cells were present in the peripheral tissues of mice reconstituted from the CD34+ bone marrow cells of both the diabetic patient and the healthy control, the diabetic-derived PI mice exhibited much lower proportions of naive-type T cells.
“Immune reconstitution from adult bone marrow CD34+ cells of patients in NSG mice provides an immune system unaltered by disease, allowing comparison of individuals in a controlled and prospective manner,” the team concludes. And, they claim, while current peripheral blood-based analysis of immune disorders can’t necessarily distinguish immune dysregulation from the knock-on inflammatory cascades that lead to disease, “our model will allow assessment of genetically programmed, HSC-intrinsic immunoregulatory abnormalities in type 1 diabetes in relation to predisposing gene alleles.”
The authors believe PI mice will allow researchers to evaluate how individual patients might respond to specific drugs or immunotherapies. They could also aid in the development of individualized immunotherapies for transplant patients, or those with cancer or infections.