Team:Freiburg/Applied Design
Applied Design
CAR T cells are a promising new approach in cancer therapy. Recently, a therapy by Novartis against B cell precursor acute lymphoblastic leukemia has been approved by the FDA as the first CAR T cell therapy ever. Despite the successful usage of this approach in clinical trials, there are still concerns about the safety of CAR T cells, as severe off-target effects can occur (Morgan et al., 2010).
With our CARTELTM cells we were able to design an AND gate system that limits activation of therapeutic T cells to the tumor microenvironment, thus preventing off-target effects.
The logic gate we designed, is based on the modular character of the cellular hypoxia sensing system. Under normoxic conditions one of the hypoxia inducible factors (HIF1A) is constantly degraded, while hypoxia stabilizes HIF1A and leads to its dimerisation with HIF1B. The HIF1A-HIF1B dimer can enter the nucleus and induce expression of genes under control of hypoxia response elements. We knocked down the endogenous HIF1A and reintroduced it under control of a promoter activated either in response to VEGF or an acidic environment. CAR is expressed under control of HRE, thus being restricted to areas providing both inputs at once, hypoxia and VEGF/low pH. The three inputs were chosen as they are common features of the tumor microenvironment of most solid tumors. While hypoxia is the central factor of our AND gate, acidity and VEGF can be exchanged. Both of these inputs are integrated via a dedicated receptor.
The endogenous VEGF receptor belongs to the family of receptor tyrosine kinases and initiates intracellular calcium signaling, finally activating the transcription factor NFAT (nuclear factor of activated T cells). We used this pathway to regulate HIF1A expression by reintroducing the knocked out HIF1A gene under control of the NFAT responsive element of the CTLA4 promoter. Acidic conditions were integrated by using the proton sensing G-protein coupled receptor TDAG8 (T cell death associated gene 8), which increases intracellular cAMP levels upon activation. This pathway was used to regulate HIF1A expression under control of the cAMP response element (CRE). Using these pathways, CAR expression can be restricted to tissues that show two characteristic traits of tumor microenvironments at the same time, either acidity and hypoxia, or VEGF and hypoxia.
By introduction of a chemically inducible suicide gene we were able to apply a further safety level in our therapeutic cells. In the case that potentially fatal off-target effects occur despite restricted CAR expression, the suicide gene allows us to quickly shut off the T cell response by inducing apoptosis and thereby eliminating all CAR T cells. We used the herpes simplex virus I Thymidine Kinase (TK) as a suicide gene that can be introduced into therapeutic T cells to provide a further level of security. This feature can be used to induce apoptosis in TK expressing cells via treatment with ganciclovir, a common compound already used to treat herpes infections (Fillat et al., 2003).
Figure 1: Schematic representation of the functional mechanism of ganciclovir. a) Structures of ganciclovir, guanine nucleotide and ganciclovir triphosphate. b) Ganciclovir enters the cell (1) and is phosphorylated by the expressed HSV-TK (2). Cellular thymidine kinases lead to two additional phosphorylations of ganciclovir (3). The ganciclovir-triphosphate is a guanine analogue which inhibits the cellular DNA synthesis processes (4). This leads to apoptosis.
We were able to show that the receptor and response elements utilized in our system were triggered by the desired inputs. From our model we can infer that the activation of our CARTELTM cells is also restricted temporospatially to cancerous lesions as T cells do migrate slowly in tissue. The suicide gene was as well shown to efficiently induce cell death upon treatment with ganciclovir.
These results underline a principally functional and applicable concept that still has to be improved in order to provide a quick and sharp on-off response. The expression system we describe, bears great potential and applicability for cancer immunotherapy. Designing the AND gate we especially sought to enhance the treatment of solid tumors as these are still difficult to fight using CAR T cells due to severe off-target effects (Morgan et al., 2010, Hartmann et al., 2017). The proposed safety mechanisms could greatly reduce such side effects by restricting CAR expression to cancerous lesions and enabling physicians to eliminate CAR T cells that went out of control.
Proof of Concept
Modeling
