Site-specific bioconjugation of proteins for in vivo applications

Abstract

Bioconjugation of proteins has been used as the breakthrough approach to biological applications. In particular, recent advanced site-specific bioconjugation has led to the development of methods for various applications, including drug delivery and diagnosis. In order to achieve these approaches, biorthogonal chemistry and site-specific unnatural amino acids (UAAs) containing biorthogonal reactive handles were developed. So far, over 200 UAAs containing click chemistry reactive handles were successfully incorporated into protein and applied to extend the function of proteins without perturbation of their function and structure. Therefore, site-specific bioconjugation of protein has a great potential for biological applications. Here, I investigated site-specific bioconjugations of proteins using optimization of reactivity and stability for efficient in vivo applications. In Chapter Ⅱ, I investigated genetically encoded a phenylalanine analog containing a hydrogen-substituted tetrazine (frTet) and determined its ability to increase the IEDDA reaction rate and allow successful bioconjugation in vivo. The IEDDA reaction rate of superfolder green fluorescent protein (sfGFP) containing frTet (sfGFP-frTet) was 12-fold greater than that of sfGFP containing methyl-substituted tetrazine (sfGFP-Tet_v2.0) in vitro. Moreover, sfGFP-frTet encapsulated with pluronic-based nanocarriers, and transferred into nude mice or tumor-bearing mice for in vivo imaging revealed fluorescence recovery upon addition of trans- cyclooctene via the IEDDA reaction. These results demonstrated that the genetically encoded frTet allows an almost complete IEDDA reaction in vivo upon addition of trans-cyclooctene, enabling temporal control of in vivo bioconjugation in high yield. In Chapter Ⅲ, I investigated species-dependent albumin-FcRn interactions using site-specific conjugation of species-matched albumin (mouse serum albumin, MSA) and species-mismatched albumin (human serum albumin, HSA) into sfGFP (superfolder green fluorescence protein). The clickable unnatural amino acid containing p-azido phenylalanine was site-specifically introduced into a sfGFP followed by conjugation to an albumin species via a hetero-bifunctional linker and SPAAC (strain promoted azide-alkyne cycloaddition). Then, sfGFP-MSA and HSA were directly compared of serum half-life in non-transgenic mice. Conjugation of HSA led to very limited extension of the serum half-life of sfGFP in mice (16.3 h), compared to that of HSA in transgenic mice harboring an allele of mouse FcRn knock-out and expressing human FcRn (67 h). However, conjugation of mouse serum albumin (MSA) resulted in a serum half-life of sfGFP (27.7 h) comparable to that of MSA in mice (28.8 h). Therefore, these results supported that albumin–FcRn interactions are species dependent in vivo. In Chapter Ⅳ, I investigated that conjugation of multiple HSA molecules to a therapeutic protein significantly further extends the serum half-life via multivalent HSA-FcRn interactions. I chose urate oxidase (Uox), a tetrameric therapeutic protein used for the treatment of gout, as a model protein. And frTet was site-specifically introduced into position 174 of each subunit of Uox. The four HSA molecules (Uox-HSA4), one or two HSA molecules (Uox-HSA1 or Uox-HSA2) were prepared via IEDDA and SPAAC, respectively. The enzyme activity of all three Uox-HSA conjugates was comparable to that of unmodified Uox. And I found out that an increase in HSA molecules conjugated to Uox (multiple albumin-conjugated therapeutic protein) enhanced FcRn binding and consequently prolonged the serum half-life in vivo. In particular, the conjugation of four HSA molecules to Uox led to a prominent extension of serum half-life (over 21 h), which is about 16-fold longer than that of Uox-WT. In Chapter Ⅴ, in order to develop a thermostable and long-circulating urate oxidase, I chose urate oxidase originating from Arthrobacter globiformis (AgUox), which has been reported to be thermostable and less immunogenic than the Aspergillus form. Then I investigated the site-specific conjugation of HSA to AgUox based on the site-specific incorporation of a clickable unnatural amino acid (frTet) and inverse electron demand Diels-Alder reaction (IEDDA). AgUox containing frTet at position 196 (AgUox_196frTet) exhibited enzymatic activity and thermostability comparable to those of wild-type AgUox. Furthermore, AgUox_196frTet exhibited a high HSA conjugation yield without compromising its enzymatic activity, generating well-defined HSA-conjugated AgUox (AgUox_196HSA). In mice, the serum half-life of AgUox_196HSA was approximately 29 h, or roughly 16 times longer than that of wild-type AgUox. In Chapter Ⅵ, in order to improve the in vivo stability of albumin-conjugated therapeutic protein, I introduced the APN chemistry into the site-specific albumin conjugation method. To demonstrate the effectiveness of APN chemistry, I performed site-specific human serum albumin (HSA) conjugation into Urate oxidase originating from Arthrobacter globiformis using thiol-maleimide or -APN chemistry and then directly compared the serum half-life of conjugates. The substantial cleavage of thioester of AgUox-MAL-HSA was observed in vitro, whereas no cleavage of thiol-APN product of AgUox-APN-HSA was observed. Furthermore, the in vivo serum half-life of AgUox-APN-HSA in the late phase was significantly longer than that of AgUox-MAL-HSA indicating the enhanced in vivo stability of AgUox-APN-HSA. Overall, these results demonstrate that the thiol-APN chemistry enhanced the in vivo stability of HSA-conjugated therapeutic protein. This dissertation research lies in the investigation of site-specific bioconjugation of proteins for its applications in vivo. To achieve this purpose, I performed with site-specific incorporation of UAAs containing biorthogonal reactive handles into proteins and generating well-defined protein conjugates by bioconjugation chemistries. And the effectiveness of site-specific bioconjugation of proteins in vivo was evaluated by in vivo imaging and pharmacokinetic studies.