Hepatitis B virus core protein phosphorylation: Identification of the SRPK1 target sites and impact of their occupancy on RNA binding and capsid structure

Abstract

Hepatitis B virus (HBV) replicates its 3 kb DNA genome through capsid-internal reverse transcription, initiated by assembly of 120 core protein (HBc) dimers around a complex of viral pregenomic (pg) RNA and polymerase. Following synthesis of relaxed circular (RC) DNA capsids can be enveloped and secreted as stable virions. Upon infection of a new cell, however, the capsid disintegrates to release the RC-DNA into the nucleus for conversion into covalently closed circular (ccc) DNA. HBc´s interactions with nucleic acids are mediated by an arginine-rich C terminal domain (CTD) with intrinsically strong non-specific RNA binding activity. Adaptation to the changing demands for nucleic acid binding during the viral life cycle is thought to involve dynamic phosphorylation / dephosphorylation events. However, neither the relevant enzymes nor their target sites in HBc are firmly established. Here we developed a bacterial coexpression system enabling access to definably phosphorylated HBc. Combining Phos-tag gel electrophoresis, mass spectrometry and mutagenesis we identified seven of the eight hydroxy amino acids in the CTD as target sites for serine-arginine rich protein kinase 1 (SRPK1); fewer sites were phosphorylated by PKA and PKC. Phosphorylation of all seven sites reduced nonspecific RNA encapsidation as drastically as deletion of the entire CTD and altered CTD surface accessibility, without major structure changes in the capsid shell. The bulk of capsids from human hepatoma cells was similarly highly, yet non-identically, phosphorylated as by SRPK1. While not proving SRPK1 as the infection-relevant HBc kinase the data suggest a mechanism whereby high-level HBc phosphorylation principally suppresses RNA binding whereas one or few strategic dephosphorylation events enable selective packaging of the pgRNA/polymerase complex. The tools developed in this study should greatly facilitate the further deciphering of the role of HBc phosphorylation in HBV infection and its evaluation as a potential new therapeutic target.

Posttranslational modifications of HBV core protein

“…Using anti-HA immunoprecipitation coupled to mass spectrometry we identified phosphorylation on all eight hydroxy amino acids in CTD of HBc (Lubyova et al, 2017). Finally, the extensive phosphorylation of CTD of HBc was also corroborated by the work of Heger-Stevic and colleagues (Heger-Stevic et al, 2018). With combination of Phos-tag SDS-PAGE electrophoresis, mass spectrometry and mutagenesis, they identified seven out of eight possible phosphoacceptor sites -Ser155, Thr160, Ser162, Ser168, Ser170, Ser176, and Ser178 as targets for SRPK1.…”

“…With combination of Phos-tag SDS-PAGE electrophoresis, mass spectrometry and mutagenesis, they identified seven out of eight possible phosphoacceptor sites -Ser155, Thr160, Ser162, Ser168, Ser170, Ser176, and Ser178 as targets for SRPK1. The sole exception was Ser181 (Heger-Stevic et al, 2018).…”

“…As the small-sized HBV genome encodes neither a protein kinase nor phosphatase, HBc depends on cellular kinases and phosphatases for phosphorylation and dephosphorylation reaction. In this respect, several kinases were suggested, among others protein kinase C (PKC) (Kann and Gerlich, 1994;Wittkop et al, 2010), casein kinase 2 (CK2) (Enomoto et al, 2006), glyceraldehyde-3-phosphate dehydrogenase protein kinase (GAPD-PK) (Duclos-Vallee et al, 1998), a 46-kDa serine protein kinase (Kau and Ting, 1998), serine arginine protein kinase 1 and 2 (SRPK1/2) (Daub et al, 2002;Heger-Stevic et al, 2018), cyclin-dependent kinase 2 (CDK2) (Ludgate et al, 2012), and polo-like-kinase 1 (PLK1) (Diab et al, 2017). The kinases have a multitude of potential targets as HBc protein sequence contains 36 amino acids (almost 20%) as potential acceptors of phosphate group.…”

“…Next to these major phosphorylation sites, additional minor phosphor-acceptor sites at S/T positions in CTD were identified (Daub et al, 2002;Diab et al, 2017;Heger-Stevic et al, 2018;Jung et al, 2014;Lubyova et al, 2017). Further mutations of Ser176 and Ser178 (178 and 180 in a 185-aa HBc variant) to alanines in HBc-Ser155/162/170Ala triple mutant almost completely removed all residual SRPK-mediated phosphorylation (Daub et al, 2002).…”

“…ARDI and ARDIII were recently shown to associate with two co-dependent nuclear localization signals, while ARDII and ARDIV behave like two independent nuclear export/ cytoplasm retention signals (Li et al, 2010). The versatile roles of core protein in pgRNA and DNA encapsidation, subcellular transport of nucleocapsid including binding to nuclear pore and release of rcDNA to nucleoplasm were shown to be facilitated by dynamic phosphorylation and de-phosphorylation events occurring predominantly on serine residues within CTD (Heger-Stevic et al, 2018;Zhao et al, 2018). The linker region (aa 141-149) was previously assumed to merely serve as a spacer between NTD and CTD.…”