Development of a CAR T-cell based agranulocytosis mouse model

Agranulocytosis, the most severe form of granulocytopenia, is characterized by a complete loss of granulocytes in the blood. Mouse models of chronic agranulocytosis are lacking, yet such a model could provide useful insights into the regulation of neutrophil and stem cell homeostasis as well as support infectious disease and antibiotic testing. We postulate that ablation of neutrophils with anti-granulocyte CAR T-cells might be able to generate a long-lasting agranulocytosis mouse model.

To prevent immunization, a murinized second generation CAR‑backbone was complemented with a scFv derived from the Gr1 (granulocyte differentiating antigen 1) hybridoma and expressed on murine T‑cells by retroviral transduction. In addition to Ly6G (lymphocyte antigen 6 complex locus) Gr1 also binds Ly6C, albeit with lower affinity. The human cell line MDA-MB‑468 was engineered to express either Ly6G or Ly6C to function as target cells. Functionality was assessed by a FACS‑based killing assay. CAR T-cells were further analyzed for their phenotype and exhaustion markers during culture. C57BL/6 mice were lymphodepleted with cyclophosphamide one day prior to CAR T-cell injection and peripheral blood granulocytes were monitored for 4 weeks.

CAR T-cells showed specific in vitro killing of Ly6G expressing target cells compared to control T-cells and parental MDA‑MB‑468 target cells. As murine T-cells lowly express Ly6C, some fratricide was observed during CAR T culture. Independent of CAR expression, T‑cells change from naïve to central memory/ effector phenotype in culture. Directly after T-cell separation a transient upregulation of several exhaustion markers was observed, which regresses by day 5. Moreover, TIM-3, CTLA-4 and LAG3 indicate exhaustion. In vivo data revealed delayed regeneration of WBCs, especially monocytes, after application of CAR T-cells compared to the control group, but, likely due to loss of CAR-T, at the current conditions chronic ablation was not yet achieved.

We have generated a highly functional murine CAR-construct that specifically kills Ly6G expressing cells. With the ongoing mouse experiments, we aim to establish a model of long-term immune neutropenia via Gr1 targeting CAR T‑cells. The model will address the role of neutrophils in immature hematopoiesis in vivo and will investigate CAR T longevity, anergy and exhaustion, as well as drugs in neutropenic infection models.

The authors declare no conflict of interest