Poster CC16 - Engineering Indoleamine 2,3 Dioxygenase Cysteine Mutations for Enzymatic Stability

Introduction: Enzyme-based protein therapeutics face many challenges concerning maintaining catalytic function during handling and in vivo delivery. Engineered variants of these therapeutic enzymes allow for increased stability while preserving their wild-type function. One approach for this is targeted mutation of unstable amino acid complexes on the protein of interest. In particular, free surface cysteines can contribute to protein instability through intra- and inter-protein disulfide bond formation, which can cause protein misfolding and aggregation (Perry 1987, Fomenko 2008). Selected mutation of surface cysteines into less reactive amino acid residues has shown to increase protein stability for therapeutic use (Fettis 2018, Hayashi 2015). Our group has previously demonstrated the efficacy of the enzyme indoleamine 2,3 dioxygenase (IDO) as an immunomodulatory protein therapeutic (Bracho-Sanchez 2019, Bracho-Sanchez 2023). The research presented here demonstrates that selected mutation of the surface cysteines of IDO not only preserves its enzymatic and therapeutic function, but also increases its stability, advantageous for greater clinical translatability.

Materials and

Methods: Two IDO mutant variants were designed, with selected mutations of cysteines into serines at sites 112, 159, 206, and 308 for IDOC4S4 and at 112, 159, and 308 for IDOC5S3. Comparable IDO mutants were designed as a fusion protein with Galectin3, as previously described (Farhadi 2018, Bracho-Sanchez 2023). In silico analysis of the mutant forms versus wild type IDO was performed on AlphaFold v2 generated structures of all three proteins. Secondary structure predictions were generated using STRIDE (Frishman 1995) and Pairwise Structure Alignment (RCSB PDB). Proteins were recombinantly produced, and purification of the final product was confirmed via SDS-PAGE. To determine enzymatic stability, IDO mutants were incubated at 4°C, room temperature, or 37°C, and assessed regularly for catalytic function. Enzymatic activity was determined via incubation of IDO with tryptophan and the electron donors, methylene blue and ascorbic acid, and the presence of the tryptophan metabolite kynurenine was measured via kinetic absorbance at 321 nm.

Results, Conclusions, and Discussions: In silico analysis showed a high degree of similarity between the IDO mutants as compared to wild type IDO; predicted secondary structure composition for both variants were comparable to native IDO, and the root mean square deviation and template modeling scores for the mutants were between 1.26-1.27 Å and 0.97-0.98, respectively. Previous work has demonstrated that the cysteines in IDO do not form disulfide bridges or coordinate the heme unit within the active enzymatic site, and while we did not anticipate any substantial changes in predicted structure, different structural compositions were reported when IDO cysteines were mutated to alanines (Austin 2013). Our findings indicate that cysteine to serine mutations do not alter the physical structure of IDO. Recombinant production of the IDO mutants resulted in purified, active product and had comparable yields as recombinant production of wild type IDO. Additionally, there was no significant difference in average activity between IDOC4S4 and IDOC5S3; while the cysteine located at site 206 has previously been shown to play a key role in maintaining catalytic function (Austin 2013), our mutation had no impact on enzymatic activity. After prolonged storage at 4°C, room temperature, and 37°C, the IDO mutants showed better enzymatic stability across all temperatures as compared to wild type IDO, with IDOC4S4 having a significantly greater activity half-life for every storage condition. These data suggest that even though the mutated cysteines are not required for enzymatic function, they may contribute to IDO instability. Additionally, this improved thermostability will also serve as an advantage for clinical translatability, limiting the likelihood of enzymatic activity loss during handling. Conjugation of our IDO mutants to Galectin3, which serves as an in vivo anchoring moiety, also showed comparable in silico predicted structural similarity as our wild-type IDO-Galectin3 fusion protein, and demonstrated purified, active expression. Future work will be assessing the therapeutic efficacy of our mutant IDO-Gal3 fusions in a mouse model of psoriasis, to confirm that these mutations maintain IDO’s immunomodulatory capabilities.

Acknowledgements (Optional):