Crystal structure of the collagen prolyl 4-hydroxylase (C-P4H) catalytic domain complexed with PDI: Toward a model of the C-P4H α2ß2 tetramer.

Collagen prolyl 4-hydroxylases (C-P4H) are α2ß2 tetramers, which catalyze the prolyl 4-hydroxylation of procollagen, allowing for the formation of the stable triple-helical collagen structure in the endoplasmic reticulum. The C-P4H α-subunit provides the N-terminal dimerization domain, the middle peptide-substrate-binding (PSB) domain, and the C-terminal catalytic (CAT) domain, whereas the ß-subunit is identical to the enzyme protein disulfide isomerase (PDI). The structure of the N-terminal part of the α-subunit (N-terminal region and PSB domain) is known, but the structures of the PSB-CAT linker region and the CAT domain as well as its mode of assembly with the ß/PDI subunit, are unknown. Here, we report the crystal structure of the CAT domain of human C-P4H-II complexed with the intact ß/PDI subunit, at 3.8 Å resolution. The CAT domain interacts with the a, b', and a' domains of the ß/PDI subunit, such that the CAT active site is facing bulk solvent. The structure also shows that the C-P4H-II CAT domain has a unique N-terminal extension, consisting of α-helices and a ß-strand, which is the edge strand of its major antiparallel ß-sheet. This extra region of the CAT domain interacts tightly with the ß/PDI subunit, showing that the CAT-PDI interface includes an intersubunit disulfide bridge with the a' domain and tight hydrophobic interactions with the b' domain. Using this new information, the structure of the mature C-P4H-II α2ß2 tetramer is predicted. The model suggests that the CAT active-site properties are modulated by α-helices of the N-terminal dimerization domains of both subunits of the α2-dimer.

Assembly of the elongated collagen prolyl 4-hydroxylase α2ß2 heterotetramer around a central α2 dimer.

Collagen prolyl 4-hydroxylase (C-P4H), an α2ß2 heterotetramer, is a crucial enzyme for collagen synthesis. The α-subunit consists of an N-terminal dimerization domain, a central peptide substrate-binding (PSB) domain, and a C-terminal catalytic (CAT) domain. The ß-subunit [also known as protein disulfide isomerase (PDI)] acts as a chaperone, stabilizing the functional conformation of C-P4H. C-P4H has been studied for decades, but its structure has remained elusive. Here, we present a three-dimensional small-angle X-ray scattering model of the entire human C-P4H-I heterotetramer. C-P4H is an elongated, bilobal, symmetric molecule with a length of 290 Å. The dimerization domains from the two α-subunits form a protein-protein dimer interface, assembled around the central antiparallel coiled-coil interface of their N-terminal α-helices. This region forms a thin waist in the bilobal tetramer. The two PSB/CAT units, each complexed with a PDI/ß-subunit, form two bulky lobes pointing outward from this waist region, such that the PDI/ß-subunits locate at the far ends of the ßααß complex. The PDI/ß-subunit interacts extensively with the CAT domain. The asymmetric shape of two truncated C-P4H-I variants, also characterized in the present study, agrees with this assembly. Furthermore, data from these truncated variants show that dimerization between the α-subunits has an important role in achieving the correct PSB-CAT assembly competent for catalytic activity. Kinetic assays with various proline-rich peptide substrates and inhibitors suggest that, in the competent assembly, the PSB domain binds to the procollagen substrate downstream from the CAT domain.

Structural insights into the substrate-binding proteins Mce1A and Mce4A from Mycobacterium tuberculosis.

Mycobacterium tuberculosis (Mtb), which is responsible for more than a million deaths annually, uses lipids as the source of carbon and energy for its survival in the latent phase of infection. Mtb cannot synthesize all of the lipid molecules required for its growth and pathogenicity. Therefore, it relies on transporters such as the mammalian cell entry (Mce) complexes to import lipids from the host across the cell wall. Despite their importance for the survival and pathogenicity of Mtb, information on the structural properties of these proteins is not yet available. Each of the four Mce complexes in Mtb (Mce1-4) comprises six substrate-binding proteins (SBPs; MceA-F), each of which contains four conserved domains (N-terminal transmembrane, MCE, helical and C-terminal unstructured tail domains). Here, the properties of the various domains of Mtb Mce1A and Mce4A, which are involved in the import of mycolic/fatty acids and cholesterol, respectively, are reported. In the crystal structure of the MCE domain of Mce4A (MtMce4A39-140) a domain-swapped conformation is observed, whereas solution studies, including small-angle X-ray scattering (SAXS), indicate that all Mce1A and Mce4A domains are predominantly monomeric. Further, structural comparisons show interesting differences from the bacterial homologs MlaD, PqiB and LetB, which form homohexamers when assembled as functional transporter complexes. These data, and the fact that there are six SBPs in each Mtb mce operon, suggest that the MceA-F SBPs from Mce1-4 may form heterohexamers. Also, interestingly, the purification and SAXS analysis showed that the helical domains interact with the detergent micelle, suggesting that when assembled the helical domains of MceA-F may form a hydrophobic pore for lipid transport, as observed in EcPqiB. Overall, these data highlight the unique structural properties of the Mtb Mce SBPs.

Structural enzymology binding studies of the peptide-substrate-binding domain of human collagen prolyl 4-hydroxylase (type-II): High affinity peptides have a PxGP sequence motif.

The peptide-substrate-binding (PSB) domain of collagen prolyl 4-hydroxylase (C-P4H, an α2 ß2 tetramer) binds proline-rich procollagen peptides. This helical domain (the middle domain of the α subunit) has an important role concerning the substrate binding properties of C-P4H, although it is not known how the PSB domain influences the hydroxylation properties of the catalytic domain (the C-terminal domain of the α subunit). The crystal structures of the PSB domain of the human C-P4H isoform II (PSB-II) complexed with and without various short proline-rich peptides are described. The comparison with the previously determined PSB-I peptide complex structures shows that the C-P4H-I substrate peptide (PPG)3 , has at most very weak affinity for PSB-II, although it binds with high affinity to PSB-I. The replacement of the middle PPG triplet of (PPG)3 to the nonhydroxylatable PAG, PRG, or PEG triplet, increases greatly the affinity of PSB-II for these peptides, leading to a deeper mode of binding, as compared to the previously determined PSB-I peptide complexes. In these PSB-II complexes, the two peptidyl prolines of its central P(A/R/E)GP region bind in the Pro5 and Pro8 binding pockets of the PSB peptide-binding groove, and direct hydrogen bonds are formed between the peptide and the side chains of the highly conserved residues Tyr158, Arg223, and Asn227, replacing water mediated interactions in the corresponding PSB-I complex. These results suggest that PxGP (where x is not a proline) is the common motif of proline-rich peptide sequences that bind with high affinity to PSB-II.