Xiaoying Zhou’s research while affiliated with Tsinghua University and other places

Human Vγ9Vδ2 T cells respond to microbial infections and malignancy by sensing diphosphate-containing metabolites called phosphoantigens, which bind to the intracellular domain of butyrophilin 3A1, triggering extracellular interactions with the Vγ9Vδ2 T cell receptor (TCR). Here, we examined the molecular basis of this “inside-out” triggering mechanism. Crystal structures of intracellular butyrophilin 3A proteins alone or in complex with the potent microbial phosphoantigen HMBPP or a synthetic analog revealed key features of phosphoantigens and butyrophilins required for γδ T cell activation. Analyses with chemical probes and molecular dynamic simulations demonstrated that dimerized intracellular proteins cooperate in sensing HMBPP to enhance the efficiency of γδ T cell activation. HMBPP binding to butyrophilin doubled the binding force between a γδ T cell and a target cell during “outside” signaling, as measured by single-cell force microscopy. Our findings provide insight into the “inside-out” triggering of Vγ9Vδ2 T cell activation by phosphoantigen-bound butyrophilin, facilitating immunotherapeutic drug design.

Motivated by the clinical observation that interruption of the mevalonate pathway stimulates immune responses, we hypothesized that this pathway may function as a druggable target for vaccine adjuvant discovery. We found that lipophilic statin drugs and rationally designed bisphosphonates that target three distinct enzymes in the mevalonate pathway have potent adjuvant activities in mice and cynomolgus monkeys. These inhibitors function independently of conventional "danger sensing." Instead, they inhibit the geranylgeranylation of small GTPases, including Rab5 in antigen-presenting cells, resulting in arrested endosomal maturation, prolonged antigen retention, enhanced antigen presentation, and T cell activation. Additionally, inhibiting the mevalonate pathway enhances antigen-specific anti-tumor immunity, inducing both Th1 and cytolytic T cell responses. As demonstrated in multiple mouse cancer models, the mevalonate pathway inhibitors are robust for cancer vaccinations and synergize with anti-PD-1 antibodies. Our research thus defines the mevalonate pathway as a druggable target for vaccine adjuvants and cancer immunotherapies.

Atomic force microscopy (AFM) force curve assay. (A) Two representative force curves showing maximum adhesion forces (blue circles) between individual Vγ9Vδ2 T cell and LX-2 cell untreated (left) or treated (right) with BPH-1236 and the corresponding work (shaded area) required for full separation between the two cells. (B) Histograms of work between the two cells with or without BPH-1236 treatment. Each group contains at least 50 data points from five pairs of cells with 10 cycles. ***P < 0.001.

Activated hepatic stellate cells (aHSCs) are now established as a central driver of fibrosis in human liver injury. In the presence of chronic or repeated injury, fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) can occur, so there is interest in down-regulating aHSCs activity in order to treat these diseases. Here, we report that Vγ9Vδ2 T cells are reduced in patients with liver cirrhosis, stimulating us to investigate possible interactions between Vγ9Vδ2 T cells and aHSCs. We find that Vγ9Vδ2 T cells kill aHSCs and killing is enhanced when aHSCs are pretreated with BPH-1236, a lipophilic analog of the bone resorption drug zoledronate. Cytotoxicity is mediated by direct cell-to-cell contact as shown by Transwell experiments and atomic force microscopy, with BPH-1236 increasing the adhesion between aHSCs and Vγ9Vδ2 T cells. Mechanistically, BPH-1236 functions by inhibiting farnesyl diphosphate synthase, leading to accumulation of the phosphoantigen isopentenyl diphosphate and recognition by Vγ9Vδ2 T cells. The cytolytic process is largely dependent on the perforin/granzyme B pathway. In a Rag2−/−γc−/− immune-deficient mouse model, we find that Vγ9Vδ2 T cells home-in to the liver, and when accompanied by BPH-1236, kill not only orthotopic aHSCs but also orthotopic HCC tumors. Collectively, our results provide the first proof-of-concept of a novel immunotherapeutic strategy for the treatment of fibrosis–cirrhosis–HCC diseases using adoptively transferred Vγ9Vδ2 T cells, combined with a lipophilic bisphosphonate.

Effector memory phenotype of Vγ9Vδ2 T cells stimulated by zoledronate or BPH-1236. Vγ9Vδ2 T cells were assessed by flow cytometric staining for TCR Vδ2, CD45RA and CD27. Flow plots in all panels are representative of at least three independent experiments.

Activated HSCs promote the growth, migration, and metastasis of human liver cancer cells. (A) The growth of Huh 7/Luc cells under the control medium or LX-2 cells conditioned medium (CM) for 72 h, as determined by IVIS imaging. Data are presented as mean ± SEM of four replicates from a representative experiment of two independent experiments. *P < 0.05. (B) Representative micrograph of the areas between scratch fronts after 48 h. The scratched Huh 7/Luc cells were treated with the control medium or LX-2 cells condition medium for 48 h. Data are presented of six replicates from a representative experiment of two independent experiments. (C) Representative images of liver metastasis in Huh 7/LX-2 cells or Huh 7 cells spleen xenografts (n = 5 per group). 1 × 106 Huh 7 cells and 5 × 105 LX-2 cells or 1 × 106 Huh 7 were injected into spleen of Rag2−/−γc−/− mice at day 0, and livers/tumors were harvested at day 43. Black arrows and blue circles highlight the tumors in liver. (D) Survival rate in Huh 7/LX-2 cells or Huh 7 cells spleen xenograft mice; n = 5 per group.

Homing behavior of Vγ9Vδ2 T cells in Rag2−/−γc−/− mice. Vγ9Vδ2 T cells were labeled with XenoLight DiR, and monitored with the IVIS imaging system on day 1, 3, 5 after i.v. adoptive transfer (n = 3). One of the two independent experiments is shown.

Time lapse movie of the interaction between CFSE (green) labeled LX-2 cells and LysoTracker Red (red) labeled Vγ9Vδ2 T cells. LX-2 cells were pretreated with 5 µM BPH-1236 for 4 h, then co-incubated with Vγ9Vδ2 T cells. Two-color and brightfield images were acquired every 2 min. Acquisition time is displayed in h: mm: ss. Scale bar represents 50 µm.

Phenotypic screening regains new momentum in pharmaceutical industry owing to its exceeded success over target-based screenings. Most phenotypic screenings rely on nonspecific biochemical readout regarding cellular viability which hampers the discovery of novel druggable mechanism of actions (MOAs). Here we present Contractility-based bi-Content micro-Collagen Chip (3CChip), which innovates cellular contractility as a biomechanics-related phenotype for drug screening. Bi-content analysis on cell contractility (imaged by iPhone) and viability suggests that the label-free contractility-based analysis exhibits superior sensitivity to compounds targeting contractile elements (e.g. focal adhesion, cytoskeleton), resulting in enlarged target pool for drug assessment. Six typical readout patterns of drug response are summarized according to relative positions of contraction/viability curves and drug targets are profiled into three categories (biomechanical, biochemical and house-keeping) by 3CChip, which would benefit subsequent target identification. The simple-to-use and effective 3CChip offers a robust platform for micro-tissue-based functional screening and may lead to a new era of mechanism-informed phenotypic drug discovery.