Search bioRxiv⌕ Search

Biology subjects

Gaenko, A.

Publications and source records attributed to Gaenko, A..

2 recordsLinked to original sources

A Multivalent Peptide-Polymer Conjugate Material Mimics STING to Therapeutically Activate Innate Immune Signaling

Stimulator of interferon genes (STING) is a promising therapeutic target for cancer immunotherapy, but agonists are often rendered ineffective by the loss of STING expression in cancer cells. Here we engineer a multivalent peptide-polymer conjugate material that can easily be delivered to the cytosol, where it mimics key protein interactions from the missing STING protein to directly activate downstream innate immune signaling. While previously developed STING mimicking therapeutics use nearly the full STING protein, this material contains only a 39 amino acid peptide from the STING C-terminal tail that includes interaction motifs for downstream kinase TBK1 and transcription factor IRF3. Conjugation of multiple peptide copies to a negatively charged polymer backbone mimics the multivalent protein-protein interactions of the oligomerized STING signaling complex, activating TBK1 and IRF3 as well as the transcription of downstream genes in both STING-proficient and STING-silenced cancer cell lines. We optimize a lipid nanoparticle formulation to deliver this conjugate material intracellularly, allowing for its application as an immunotherapy for ovarian cancer. Treatment with the STING mimicking conjugate material promoted the production of type I interferons, repolarization of myeloid cells to an anti-tumor phenotype, and recruitment of T cells to tumors in mice. This treatment ultimately led to tumor regression and extended survival in multiple mouse models of metastatic ovarian cancer. Overall, this work highlights the potential of peptide-polymer conjugate mimics of STING to therapeutically activate innate immune signaling.

bioengineering↗

CoFrEE: An Application to Estimate DNA Copy Number from Genome-wide RNA Expression Data

We introduce CoFrEE, a simple python-based approach to extracting copy number data from expression values that works with either RNAseq or array-based expression data. CoFrEE works best in tumor cohorts that include a subset of non-tumor tissues and is applied to processed (RSEM, RPKM or TPM) expression, rather than raw data. Experiments with real public data suggest CoFrEE can provide copy number estimations comparable to existing RNAseq-based approaches, with the advantage of also being applicable to the multitude of older expression-array cohorts.

bioinformatics↗