The in vitro oxidative folding of proteins has been studied for over sixty years, providing critical insight into protein folding mechanisms. Hirudin, the most potent natural inhibitor of thrombin, is a 65-residue protein with three disulfide bonds, and is viewed as a folding model for a wide range of disulfide-rich proteins. Hirudin’s folding pathway is notorious for its highly heterogeneous intermediates and scrambled isomers, limiting its folding rate and yield in vitro. Aiming to overcome these limitations, we undertake systematic investigation of diselenide bridges at native and non-native positions and investigate their effect on hirudin’s folding, structure and activity. Our studies demonstrate that, regardless of the specific positions of these substitutions, the diselenide crosslinks enhanced the folding rate and yield of the corresponding hirudin analogues, while reducing the complexity and heterogeneity of the process. Moreover, crystal structure analysis confirms that the diselenide substitutions maintained the overall three-dimensional structure of the protein and left its function virtually unchanged. The choice of hirudin as a study model has implications beyond its specific folding mechanism, demonstrating the high potential of diselenide substitutions in the design, preparation and characterization of disulfide-rich proteins.

The oxidative folding of proteins has been studied for over sixty years, providing critical insight into protein folding mechanisms. Hirudin, the most potent natural inhibitor of thrombin, is a 65-residue protein with three disulfide bonds, and is viewed as a folding model for a wide range of disulfide-rich proteins. Hirudin’s folding pathway is notorious for its highly heterogeneous intermediates and scrambled isomers, which plague its folding rate and yield in vitro. Aiming to overcome these limitations, we undertook a systematic investigation of diselenide bridges at native and non-native positions and investigated their effect on hirudin’s folding, structure and activity. Our studies demonstrated that, regardless of the specific positions of these substitutions, the diselenide crosslinks enhanced the folding rate and yield of the corresponding hirudin analogs, while reducing the complexity and heterogeneity of the process. Moreover, crystal structure analysis confirmed that the diselenide substitutions maintained the overall structure of the protein and left the function virtually unchanged. The choice of hirudin as a study model has implications beyond its specific folding mechanism, demonstrating the high potential of diselenide substitutions in the design, preparation and characterization of disulfide-rich proteins.

No abstract

No abstract

Aristolochic acid (AA) is a potent human carcinogen and nephrotoxin found in preparations of Aristolochia plants used in Chinese Traditional Medicine. Following biotransformation to form N-sulfonyloxyaristolactam (AL-I-NOSO3), this intermediate undergoes heterolytic cleavage of the sulfate group to generate a reactive cyclic nitrenium ion, the ultimate DNA binding species. Recently, we showed that primary human hepatocytes significantly increase renal toxicity of AA in the integrated human liver-kidney “organs on-chips” model. Therefore, we propose that AA is activated in the liver by forming AL-I-NOSO3, which is transported to the kidney protected from decomposition by binding to serum albumin. We employed mass spectrometric, fluorimetric and X-crystallography based approaches to dissect mechanisms of interactions between human serum albumin (HSA), AA, N-hydroxyaristolactam and AL-I-NOSO3. First, we demonstrate that HSA stabilizes otherwise labile N-sulfonyloxyaristolactam. Quenching of the native fluorescence of HSA due to the presence of a sole molecule of tryptophane-214, allowed us to conclude that all three compounds have similar affinities to IIA drug binding pocket of HSA. Subsequently, we obtained a high-resolution X-ray structure of AA bound to HSA in domain IB (1.9°A, pdb: 6HSC). Since prior to crystallization HSA was enriched with sodium myristate and site IB in circulation is occupied by fatty acids, our results imply that the IB pocket is the primary high affinity binding site for AA and its active forms. To assess whether AL-I-NOSO3 covalently binds protein, we incubated human plasma and purified HSA with this active AA. Immunoblotting of reacted HSA using antibodies that recognize aristolactam(AL)-adducted DNA suggests irreversible covalent adduction of AL to HSA. A combined approach using mass spectrometry instruments and enzymatic digestion revealed that AL is adducted to HSA at the following sites: Trp-214, Tyr-138 and Tyr-141. The former amino acid is located in the IIA drug binding site of HSA, while the latter two can be found in our HSA/AA structure in the site IB in the vicinity to AA molecule, corroborating our X-crystallography and fluoremetric data. Based on these studies we propose that AL-I-NOSO3 has a dual mode of interactions with HSA. If AL-I-NOSO3 decomposes prior to HSA binding, aristolactam will become irreversibly trapped with HSA. This binding to HSA would serve as mechanism of detoxication of AA species. However, if AL-I-NOSO3 binds to HSA prior to decomposition, it should be protected by HSA and transported to target tissues in its intact form. Citation Format: Sergei Pomyalov, Radha Bonala, Robert Rieger, Irina Zaitseva, Charles Iden, John Haley, Robert Turesky, Francis Johnson, Thomas Rosenquist, Arthur P. Grollman, Gil Shoham, Viktoriya S. Sidorenko. Molecular mechanisms by which a bioactivated human carcinogen is transported to target tissues [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4660.