Tsuji Lab

Principal Investigator

The Tsuji Lab's current research is applied and translational, seeking to further our understanding of the host immune system for the development of effective vaccines against infectious diseases and cancer. The laboratory has recently established a novel humanized mouse model that mimics the human immune system (HIS) by utilizing recombinant adeno-associated virus (AAV)-based gene transfer technology. We have demonstrated that AAV vector-mediated gene delivery is a simple and efficient method to transfer multiple human cytokines and HLA genes to immunodeficient NSG mice before the engraftment of human hematopoietic stem cells. This distinct approach greatly improves not only the rate of HIS reconstitution, but also their functions, which include those of T cells, B cells, and dendritic cells. Thus, our HIS mice bridge the considerable gap between murine and human models, thereby reducing time and cost while accelerating the transition of candidate products from the bench to clinic.

Utilizing this humanized mouse model, the Tsuji Lab is currently investigating:

  1. The efficacy of new HIV and malaria vaccines.
  2. The adjuvant effect of our CD1d-binding, NKT-cell stimulatory glycolipid for vaccines.
  3. The therapeutic effect of our immunostimulatory glycolipid against various cancers.
  4. The adjuvant effect of our recently identified PD1-inhibitory peptides for vaccines.


Humanized Mouse Models for Vaccine/Adjuvant Research

The Tsuji Lab has recently established a humanized mouse model in which human genes encoding cytokines and HLA molecules are introduced by adeno-associated virus serotype 9 (AAV9) vector-mediated delivery. In transduced immunodeficient NSG mice, the human immune system (HIS) is reconstituted upon engraftment of human hematopoietic stem cells (HSCs), which includes functional human CD8+ T cell, CD4+ T cells, and B cells [ref. 1, 2, 3, 4, 5, 6, 7]. This approach contrasts other humanized mouse models by achieving a high level (>85%) of reconstitution of human CD45+ cells without significant mouse to mouse variation. Furthermore, the transduction of HLAs/hucytokines allows our HIS mice to induce a potent human T-cell response as well as a human IgG response against antigens derived from human pathogens.

We have also generated a humanized mouse model in which both human CD1d and HLA-A2 are co-expressed by way of AAV9 vectors [ref. 8]. These humanized mice can mount functional human invariant natural killer T (iNKT) cells as well as human CD8+ T cells, thus enabling us to study the function of CD1d-binding NKT-stimulating glycolipids in the context of our Project 2 (see below).

Most recently, we have succeeded in generating humanized mice that not only express human group 1 CD1 molecules, but also possess functional CD1-restricted human T lymphocytes that are developed and differentiated from HSCs. This was done by first transfecting ES cells with large fragments of human genomic DNA which contained human group 1 CD1 genes. After generating human group 1 CD1-transgenic (hCD1-TG) mice, they were backcrossed onto NSG (NOD/SCID-Rag2nullIL-2Rγnull) mice for more than 10 generations, and thus, hCD1-TG NSG mice were established. The neonates of the hCD1-TG NSG mice were then transduced by AAV9 vectors with genes that encode human hematopoietic cytokines to generate the HIS in the mice. After engrafting with human HSCs, group 1 CD1-restricted human T cells were successfully differentiated and developed from HSCs in hCD1-TG NSG mice. We named such hCD1-TG NSG mice having functional human T cells and other human immune competent cells (B and NK cells and DCs), hCD1-TG HIS mice. These mice are optimal for the study of host immunity against Mycobacterium tuberculosis (Mtb), as they are mainly composed of lipids that can bind CD1a, CD1b, or CD1c molecules.

A CD1d-binding NKT Stimulating Glycolipid as Vaccine Adjuvant and Cancer Immunotherapy

From a glycolipid library consisting of more than 100 α-galactosylceramide analogs, we have previously identified a clinical lead called 7DW8-5 which can exert potent stimulatory activity on iNKT cells and dendritic cells and display an adjuvant effect superior to that of α-GalCer [ref. 9, 10, 11, 12, 13, 14]. 7DW8-5 has demonstrated an excellent safety profile and potent immune enhancing activity in non-human primate studies [ref. 15] as well as highly localized biodistribution in the draining lymph nodes upon intramuscular injection in a mouse model [ref 16, 17]. We are currently conducting preclinical testing of 7DW8-5 as an immunotherapy agent against cancer, such as ovarian and breast cancers using humanized mice (see Project 1) [ref. 18]. Analysis of its antitumor properties are being conducted in collaboration with Dr. Arnold Han Lab in the Department of Medicine at Columbia University using single cell RNA-seq in order to determine the TCR repertoire, CD markers, and cytokine profile at a single-cell level. Lastly, we are partnering with Sanaria Inc., a world leading organization in developing a human malaria vaccine based on radiation-attenuated malaria sporozoites, in order to conduct a Phase I clinical trial to see whether our glycolipid adjuvant can not only increase the efficacy of their established malaria vaccine, but also help decrease the dose of their vaccine without affecting its efficacy or safety profile.

PD-1 Inhibitory Peptides to Improve Malaria Vaccine

In malaria-infected individuals, PD-1 is highly elevated on CD4+ T cells suggesting that PD-1-mediated immune dysfunction may limit protective immunity against the parasite. In a rodent malaria model, PD-1 was shown to severely dampen CD8+ T cell proliferation and prolong chronic infection. Leidos, Inc. a largest U.S. government contractor, has recently identified several PD-1 peptides that bind both human and mouse PD-1 from a peptide phage display library. In order to determine whether PD-1 inhibitory peptides can enhance the immunogenicity of a malaria vaccine, we immunized a group of mice with a recombinant adenovirus expressing P. yoelii circumsporozoite protein (AdPyCSP), followed by the injection of PD-1 inhibitory peptides. Another group of mice received AdPyCSP alone. When we compared the immunogenicity of the two groups of mice, we found that PD-1 inhibitory peptides could significantly enhance the level of PyCSP-specific CD8+ T-cell response, ultimately leading to a more potent protective immunity against rodent malaria infection. We have also found that PD-1 inhibitory peptides could enhance the immunogenicity of a human malaria vaccine, a recombinant adenovirus expressing P. falciparum circumsporozoite protein (AdPfCSP), in the HIS mice described in Project 1. We are currently determining the mechanisms by which PD-1 inhibitory peptides display their immunomodulatory effect.

Lab Members