Multi-input/multi-output Control of Cellular Processes by PROCISiR

Cells react to multiple environmental or autologous inputs by various behaviors. These biological processes are difficult to investigate because of their complexity. Small molecules are widely used as extrinsic inputs to investigate biological process in vitro, ex vivo and in vivo. While most systems allow single molecule input and single output that limits their applications. Recently, a research paper “Multi-input chemical control of protein dimerization for programming graded cellular responses” was published in Nature Biotechnology in October 2019. The paper developed a post-translational control system that can be used to detect diverse cellular responses. It utilizes the NS3a protease as a central receiver protein that is the target of multiple clinically approved drugs. Computationally designed reader proteins are used to translate different drug-NS3a reactions into diverse outputs.

This trispecific antibody was engineered to bind three different targets: 1) the cancer antigen CD38, which is an antigen highly expressed by malignant myeloma cells and the current target of monoclonal immunotherapy 2) CD3, which is a T cell receptor activation signal that leads to cytokine secretion, and 3) CD28, which is a survival signal inhibiting programmed cell death. The multi-functionality of this trispecific antibody allows the targeting of the tumor while providing co-stimulatory signals that promote T cell expansion and survival in order to increase elimination of the tumor. To create this antibody, researchers used the trispecific antibody platform that has been previously reported to create an antibody that bound three different sites on the HIV viral envelope.

Using the cross-over dual variable (CODV) bispecific antibody format, Wu et al. evaluated combinations of CD3 and CD28 variable fragments for sustained activation of T cells. Mutation of the binding sites revealed that each component was necessary for optimal T cell activation. In vitro testing of the optimal CD38/CD3xCD28 trispecific antibody preferentially stimulated proliferation of both CD4⁺ T helper 1 cells and memory and effector CD8⁺ T cells, possibly contributing to improved immunity against non-CD38 expressing tumor cells. Additionally, engagement of CD28 promoted T cell proliferation and expression of pro-survival protein Bcl-xL as expected. Interestingly, the engineered trispecific antibody showed a 3- to 4-log higher cytolytic activity in vitro against CD38^high and CD38^low compared to the clinically approved CD38-targeting monoclonal antibody daratumumab. In order to test the effects of the multi-functional trispecific antibody on tumor regression, a humanized mouse model of multiple myeloma was used. In vivo administration of this trispecific antibody exhibited a dose-dependent reduction in tumor burden even at low doses. Some cancer immunotherapies have been associated with cytokine-related toxicities in human. To assess the safety of the antibody, non-human primates were used. Administration of the trispecific antibody in non-human primates revealed potent stimulation of memory and effector T cell proliferation and targeting with low toxicity at tolerable doses.

This study shows the potential for novel multi-functional antibodies that can deliver two signals to T cells while simultaneously directing them to cancer cells, resulting in enhanced T cell activation and tumor targeting. These antibodies have opened the door to the ability to engineer antibodies that can target any cancer antigen or combination of antigens while precisely controlling the stimulation and activation of the immune system. The next step will be to evaluate antibody safety and efficacy against cancer in humans. Although anti-therapeutic response could still occur in clinical trials, current bispecific antibodies using a related format have been well tolerated in humans. This multi-specific targeted approach increases the immunotherapeutic options available and increases the ability to harness the power of the body’s immune system to defeat cancer.

Computational Design of NS3a Readers

With the Rosetta interface design method, they developed protein readers that selectively recognize NS3a-drug complex. They designed some candidate scaffolds recognizing danoprevir/NS3a, including LRRs, DHRs, ferredoxins and helical bundles. One of DHR designs that showed modest drug-dependent binding to NS3a was used for further optimization. The final construct of danoprevir/NS3 complex reader, DNCR2, showed apparent affinity for the NS3a/danoprevir complex and no binding to apo NS3a or free danoprevir. Further biochemical analysis confirmed that DNCR2/danoprevir/NS3a form a 1:1:1 complex. These results proved that DNCR2 is a perfect danoprevir/NS3a reader protein. Other two readers, GNCR1 for grazoprevir/NS3 complex and ANR for apo NS3a, were developed with similar method.

Validation of PROCISiR in Mammalian Cells

They next evaluated the function of PROCISIR in detecting multi-input/multi-output behaviors in mammalian cells. They observed that grazoprevir exclusively colocalized NS3a-mCherry with membrane-targeted GNCR1, while only danoprevir led to colocalization with mitochondrially targeted DNCR2. Likewise, membrane-targeted ANR prelocalized NS3a-mCherry to the plasma membrane, and danoprevir treatment recruited NS3a to the nucleus with nuclear targeted DNCR2. These colocalization experiments validated that the DNCR2, GNCR1 and ANR readers are selective for their targeted state of NS3a and can be used in concert.

PROCISiR Enables Programmable Transcriptional Control

Two fusion proteins, DNCR2-VPR (a transcriptional activator) and NS3a-dCas9, functioned together to promote endogenous gene expressions in HEK293 cells. Danoprevir acted as agonist and grazoprevir as antagonist to temporally and proportionally control the transcription. Addition of danoprevir activated CXCR4 expression, while addition of grazoprevir rapidly reversed this expression. Moreover, treatment with danoprevir and increasing grazoprevir proportions produced graded transcription outputs. Besides, they achieved proportional and graded control of two transcriptional outputs with two inputs. Thus the PROCISiR architecture can be used for temporal, graded and multiple transcriptional control.

Proportional Control of Cell Signaling with PROCISiR

Confocal microscopy of EGFP-DNCR2 and BFP-GNCR1 validated that the concentrations of danoprevir and grazoprevir provided graded and proportional regimes of DNCR2 and GNCR1 colocalization with NS3a. They next used danoprevir/grazoprevir combinations with DNCR2-TIAM (Rac1 guanine nucleotide exchange factor) and GNCR1-LARG (RhoA/B/C guanine nucleotide exchange factor) to control the activations of Rac and Rho GTPases, which work reversely in actin cytoskeleton and cell morphology. In cells coexpressing DNCR2-TIAM and GNCR1-LARG with membrane-localized NS3a, danoprevir treatment resulted in cell expansion due to the activity of Rac. Grazoprevir treatment caused cell contraction and stress fibers because of the Rho activation. With PROCISiR, they found that concurrent variation in Rho and Rac activity does not result in an average phenotype, but one closer to the Rho-only.

Conclusion

PROCISiR can control switching behavior and provide graded, proportional control over two cellular outputs at once. The control modalities in PROCISiR can be used to manipulate mammalian cellular processes and potentially engineer drug regulated cell therapies.

In this paper, Grazoprevir/NS3a reader designs were synthesized and cloned into pETCON plasmids by Genscript. As the leader in gene synthesis, Genscript have completed over 600,000 gene synthesis projects for scientists around the world. GenScript's gene synthesis service saves both time and money that makes your research easy.

• Foight, G.W., Wang, Z., Wei, C.T., Greisen, P.J., Warner, K.M., Cunningham-Bryant, D., Park, K., Brunette, T.J., Sheffler, W., Baker, D., & Maly, D.J. (2019). Multi-input chemical control of protein dimerization for programming graded cellular responses. Nature Biotechnology, 1-9.