Search

Abstract: The leader protease (Lbpro) of foot-and-mouth disease virus frees itself during translation from the viral polyprotein by cleavage between its own C terminus and the N terminus of the subsequent protein, VP4. Lbpro also specifically cleaves the host proteins eukaryotic initiation factor (eIF) 4GI and 4GII, thus disabling host cell protein synthesis. We used NMR to study full-length Lbpro as well as a shortened species lacking six C-terminal amino acid residues (sLbpro) to examine the mechanism of self-processing, the quaternary structure and the substrate specificity. Both Lbpro forms have the same structure in solution as in the crystal. In the solution structure of sLbpro, the 12 residue C-terminal extension was flexible and disordered. In contrast, the 18 residue C-terminal extension of full-length Lbpro was bound by the substrate-binding site of a neighbouring molecule, resulting in the formation of a stable dimer in solution. The Lbpro dimer could not be dissociated by increasing the ionic strength or by dilution. Furthermore, titration with model peptides mimicking the substrates destabilised the dimer interface without dissociating the dimer. The peptides were, however, bound by sLbpro in the canonical substrate binding site. Peptide binding gave rise to chemical shifts of residues around the sLbpro substrate binding site. Shifts of Asn146 and Glu147 indicated that these residues might form the enzyme''s S1′ site and interact with the P1′ arginine residue of the eIF4GI cleavage site. Furthermore, differences in substrate specificity between sLbpro and Lbpro observed with an in vitro translated protein indicate some involvement of the C terminus in substrate recognition. [Copyright &y& Elsevier]

Abstract: The gain of neurotoxic function in amyotrophic lateral sclerosis (ALS) has been linked to misfolding of the homodimeric enzyme Cu/Zn superoxide dismutase (SOD). Here, we present the crystal structure of fully cysteine-depleted human SOD (SODCallA), representing a reduced, marginally stable intermediate on the folding pathway in vivo that has also been implicated as neurotoxic precursor state. A hallmark of this species is that it fails to dimerise and becomes trapped as a monomer in the absence of the active-site metals. The crystallographic data show that removal of the C57–C146 disulphide bond sets free the interface loop IV in the apo protein, whereas the same loop remains unaffected in the holo protein. Thus, the low dimerisation propensity of disulphide-reduced apoSOD seems to be of entropic origin due to increased loop flexibility in the monomeric state: in the disulphide-reduced holo protein this gain in configurational entropy upon splitting of the dimer interface is reduced by the metal coordination. [Copyright &y& Elsevier]

Abstract: TPST1 is a human tyrosylprotein sulfotransferase that uses 3′phosphoadenosine-5′phosphosulfate (PAPS) to transfer the sulfate moiety to proteins predominantly designated for secretion. To achieve a general understanding of the cellular role of human tyrosine-directed sulfotransferases, we investigated targeting, structure and posttranslational modification of TPST1. Golgi localisation of the enzyme in COS-7 and HeLa cells was visualised by fluorescence imaging techniques. PNGase treatment and mutational studies determined that TPST1 bears N-linked glycosyl residues exclusively at position Asn60 and Asn262. By alanine mutation of these asparagine residues, we could determine that the N-linked oligosaccharides do not have an influence on Golgi retention of TPST1. In concert with N and C-terminal flanking residues, the transmembrane domain of TPST1 was determined to act in targeting and retention of the enzyme to the trans-Golgi compartment. This domain exhibits a pronounced secondary structure in a lipid environment. Further in vivo FRET studies using the transmembrane domain suggest that the human tyrosylprotein sulfotransferase may be functional as homodimer/oligomer in the trans-Golgi compartment. [Copyright &y& Elsevier]

The Rel/NF-κB family of eukaryotic transcription factors bind DNA with high specificity and affinity as homo- or heterodimers to mediate a diverse range of biological processes. By comparison, the nuclear factor of activated T-cells (NFAT) family has been recognised as Rel homologues due to structural similarities between the DNA-binding domains, yet they bind DNA as lower-affinity monomers. The structural and functional overlap between the NF-κB and NFAT families suggests that they may be evolutionarily divergent from a common, monomeric ancestor but have evolved different mechanisms to achieve high-affinity binding to their target DNA sequences. In order to understand the origin of these mechanistic differences, we constructed two chimeric proteins, based on molecular modelling, comprising the DNA-binding domain of NFAT and the dimerisation domain of NF-κB p50, differing only in the position of the splice site. Biophysical characterisation of the wild-type and chimeric proteins revealed that one of the chimeras bound DNA as a high-affinity, NF-κB-like cooperative dimer, whilst the other bound as a lower-affinity, NFAT-like monomer, demonstrating the importance of the interdomain linker in controlling the intrinsic ability of NFATc to form dimers. In addition, we have studied the rate of exchange of monomers between preformed NF-κB dimers and have determined, for the first time, the intrinsic homodimerisation constant for NF-κB p50. These data support a model in which NF-κB proteins bind DNA both in vitro and in vivo as high-affinity preformed homo- or heterodimers, which in an unbound form can still exchange monomer units on a physiologically relevant timescale in vivo. [Copyright &y& Elsevier]