(229z) Designed Ankyrin Repeat Proteins in Chimeric Antigen Receptor Engineered T Cells

 

Elizabeth Siegler¹, Yu Jeong Kim², Pin Wang³

¹Department of Biomedical Engineering, University of Southern California, Los Angeles, CA 90089

²Department of Pharmacology and Pharmaceutical Sciences, University of Southern California, Los Angeles, CA 90089

³Mork Family Department of Chemical Engineering and Materials Science, University of Southern California, Los Angeles, CA 90089

Cancer immunotherapy has the potential to achieve long term remissions by harnessing the patientâ??s own immune system to attack cancer cells. Chimeric antigen receptors (CARs) are engineered proteins that offer a method of reprogramming autologous immune cells to recognize and target cancer cells. CAR-engineered T cells are not MHC restricted and can be engineered to recognize specific tumor-associated surface antigens (TAA). Typically, this recognition is due to a single chain variable fragment (scFv) which is derived from the desired antibody and fused to intracellular signaling domains. Antigen binding to the scFv triggers a signaling cascade that activates the CAR-T cell against the TAA-expressing cancer cell.

CAR-T therapy has been largely successful in hematological cancers, but has faced many setbacks in treating solid tumors. Often this is because CAR-T cells become exhausted and lose their cytotoxic capabilities after long exposures to TAA. Many factors are thought to contribute to CAR-T exhaustion, but one important factor may be due to crosslinking of the scFvs on the CAR-T cell surface. scFvs tend to aggregate, which can stimulate the CAR-T cell even in absence of the appropriate antigen. Prolonged stimulation leads to eventual exhaustion and compromised function, including diminished cytokine release and decreased cytotoxicity.

Antibody mimetic proteins also are capable of binding to specific antigens. One such protein family is designed ankyrin repeat proteins (DARPins), which are based off naturally occurring ankyrin proteins. In general, these proteins are smaller than scFvs and consist of two or three repeat units of 33 amino acids each. Constructs with specific binding affinities, sizes, or other desired qualities can be generated from DARPin libraries. DARPins are more stable and less prone to aggregation compared to scFvs and are being explored as an alternative to scFv use in CARs. In particular, we are comparing two Her2-binding DARPin-CARs to a traditional anti-Her2 scFv CAR.

We have selected two DARPins, G3 and 929, to compare to the scFv 4D5. G3 is the smallest of the three Her2 targeting proteins, which may make it more easily expressed during lentiviral transduction. G3 also has the higest binding affinity (k_d = 90 pm), which may make the CAR more sensitive to Her2. We are also testing the DARPin 929, which binds to a different Her2 domain than either 4D5 or G3. Most anti-Her2 scFvs, including trastuzumab-derived 4D5, bind to Her2 domain IV, which is partially obstructed when Her2 dimerizes. 929 binds to domains I-III, which may be more easily accessible than domain IV.

To date, our group has performed side-by-side in vitro testing of G3, 929, and 4D5 CAR-T cells. Each of these antibody binding domains were cloned into 3^rd generation CD28-41BB-CD3ζ CARs. CAR surface expression, interferon-γ release, and cytotoxicity are comparable between the three CARs. However, T cell exhaustion is not readily observable in vitro.  Our next steps include in vivo testing, in which we predict the DARPin CARs will be superior to the scFv CAR due to decreased aggregation and exhaustion.

KEYWORDS: cancer therapy, cancer immunotherapy, chimeric antigen receptors (CARs), designed ankyrin repeat proteins (DARPins)

References

 

1. Chang, Z.; Silver, P.; Chen, Y., Identification and selective expansion of functionally superior T cells expressing chimeric antigen receptors. J Transl Med. 2015, 13 (161), 1-16.
2. Chmielewski, M.; Hombach, A.; Abken, H., Antigen-specific T-cell activation independently of the MHC: chimeric antigen receptor-redirected T cells. Front Immunol. 2013, 4 (371), 1-5.
3. Münch, R.; Mühlebach, M.; Schaser, T.; Kneissl, S.; Jost, C.; Plückthun, A.; Cichutek, K.; Buchholz, C., DARPins: An efficient targeting domain for lentiviral vectors. Mol Ther 2011, 19, 686-693.
4. Riddell, S.; Srivastava, S., Engineering CAR-T cells: design concepts. Trends Immunol 2015, 36 (8), 494-502.