Structural Enzymology: A Quantitative Approach

Project Leader
Prof. Rik K. Wierenga, Ph.D.

Biocenter Oulu and Faculty of Biochemistry and Molecular Medicine, University of Oulu

Background and Significance

CoA dependent enzymes form important families of enzymes, catalysing reactions that are thioester dependent. It has intrigued enzymologists what the specific role of CoA is in these thioester dependent reactions. The CoA moiety is important not only for binding of the substrate but also for additional CoA-enzyme interactions at the transition state, which lower the free energy barrier of the catalysed reaction (Jencks, 1987). The mechanism of this transition state stabilisation has not yet been elucidated. Other interesting questions concern the importance of hydrogen bond networks that influence the electrostatic stabilisation achieved by oxyanion holes, when they interact with the negatively charged oxyanion of the transition state of the thioester intermediates (Guo and Salahub, 2008). These questions are addressed by studying enzymes of the β-oxidation pathway. Some of these enzymes are monofunctional and some are multifunctional. We also study the substrate channelling properties of these multifunctional enzymes (Yao and Schulz, 1996).

Collagen prolyl 4-hydroxylase (C-P4H) (Myllyharju, 2008) hydroxylates the Y-prolines of  the X-Y-Gly repeats of procollagen. Collagen is a triple helical protein, built up from three collagen molecules that adopt the poly(proline)-type-II conformation. Many Y-residues are prolines and very often these prolines are hydroxyprolines, being of critical importance for the stability of the collagen triple helix. C-P4H is an α2β2 tetramer. The α-chain has three domains of which the N-terminal domain is important for forming the α2-dimer, the middle domain is the peptide-substrate-binding (PSB) domain and the C-terminal domain is the catalytic domain (CAT). The b-chain is also known as proteindisulfide isomerase (PDI). Our C-P4H research efforts are aimed at resolving the cryoEM or crystal structure of C-P4H, which will reveal the precise spatial location of the PSB domain with respect to the CAT domain and which will be critically important for understanding all aspects of the C-P4H reaction mechanism, including the function of the β/PDI subunit.

Recent Progress

MFE1, multifunctional enzyme, type-1, has two active sites, being an enoyl-CoA hydratase active site and a 3S-hydroxyacyl-CoA dehydrogenase active site. The structure of MFE1 can be subdivided in five domains (Figure 1). Domain A and domain B form the hydratase active site, whereas domain C, D and E form the dehydrogenase active site. Structurally domain B is part of the dehydrogenase part. For certain MFE1 substrates it has been shown that substrate channelling can be observed, for example the substrate 2E,4E-decadienoyl-CoA is converted into 3-keto-decanoyl-CoA, for which enzyme kinetic data suggest that the intermediate is not released into bulk solvent. Several MFE1 structures are available. In each of these crystal structures (of rat peroxisomal MFE1) the crystal packing is the same with two molecules in the crystallographic asymmetric unit. Interestingly, these two molecules adopt different conformations. The systematic comparison of the structures of these two molecules, molecule A and molecule B, has provided some interesting ideas on the functional importance of the conformational flexibility of MFE1 (Kasaragod et al., 2017). Two hinge regions can be identified, effecting (i) the position of the A-domain with respect to the HAD part and (ii) effecting the position of the C-domain with respect to the D/E-domains. The detailed structural analysis shows that in molecule B the dehydrogenase part is fully open, whereas the A-domain is positioned near the dehydrogenase part, and therefore also near the linker helix (domain B). In molecule A the dehydrogenase part is more closed and domain A has moved away from the dehydrogenase part, and therefore is further away from the linker helix (domain B). Domain A and domain B together form the crotonase fold and it is known that the properties of domain B (also known as helix-10 of the crotonase fold) are important for the catalytic efficiency of the crotonase fold. Therefore these comparisons suggest a mechanism in which a liganded, closed conformation of the dehydrogenase part favors an open conformation of the hydratase part, facilitating release of the product formed in this active site. Using various structural enzymological approaches we currently try to test this hypothesis further using various approaches, in particular using the substrate 2E,4E-decadienoyl-CoA (synthesised by our collaborator Dr. Werner Schmitz, University of Wurzburg, Germany).

The ongoing structural studies on the human CP4Hs have resulted in a model of the full length α2β2 tetramer. This model (Figure 2) is based on the crystal structure of the dimeric DD-construct of C-P4H-I (Ananatharajan et al., 2013), complemented with SAXS studies of the full length C-P4H-I tetramer, that were done by Petri Kursula (University of Bergen, Norway) (Koski et al., 2017). The DD-domain construct includes the N-terminal domain and the middle domain of the a-chain. The DD crystal structure shows a coiled-coil dimerization motif, suggesting that the α2β2 tetramer is built from an a2-dimer. The SAXS shape information and subsequent model fitting calculations suggest that the β/PDI subunits cap the α2-dimer, as visualised in Figure 2. In the current hypothesis of the reaction mechanism the PSB domain binds to a certain region of the procollagen substrate, after which the catalytic domain catalyses several hydroxylation cycles. This hypothesis is consistent with a processive reaction mechanism, as suggested by enzyme kinetic data (De Jong et al., 1991). The SAXS model suggests that the β/PDI subunit interacts mainly with the CAT domain of the α-chain. The latter prediction is in good agreement with recent studies on the properties of truncated variants of the α-chain, where it is found that a truncated variant of the α-chain, without the N-terminal dimerization domain and without the PSB domain, still forms a tight complex with β/PDI.

Figure 1. The structure of MFE1 (Kasaragod et al., 2017). A) Schematic drawing of the domains of rat peroxisomal MFE1. Domain A is at the N-terminus. It forms, together with domain B, the crotonase fold.  Domain B is also the linker helix between domain A and the dehydrogenase part. This helix is tightly interacting with the dehydrogenase part. The active site residues of the crotonase and the dehydrogenase active sites are also listed. B) The MFE1 fold. The same color scheme is used as in A), but the active site helix of domain A (helix-3) is in light blue. The upper part (domain A), together with the helix-10 (domain B), forms the crotonase fold. The lower part (domains C and D/E) forms the dehydrogenase part. In this structure (5MGB) acetoacetyl-CoA (labeled as CAA) is bound in the crotonase active site and NAD+ (labeled as NAD) is bound in the dehydrogenase active site. The two arrows visualize the hinge motions of (i) domain A with respect to the BCDE-domains and (ii) domain C with respect to the D/E-domains.

Figure 2. Schematic model of the human C-P4H-I α2β2 heterotetramer (Koski et al., 2017). The β/PDI subunits (consisting of four domains, a, b, b’, a’) are shown in light and dark grey colors. The two a-chains are in purple and orange. CC refers to the coiled-coil N-terminal dimerization domain, PSB refers to the peptide-substrate-binding, middle domain, L is the linker region and CAT refers to the C-terminal catalytic domain.  The predicted mode of binding of a partially hydroxylated procollagen peptide is also shown. This topology is such that the unhydroxylated peptide forms an initial enzyme-substrate complex with the PSB domain. The x region of the peptide refers to the unknown number of residues between the peptide regions that bind to the binding sites of the PSB and CAT domains, respectively. The region at its N terminal end extends to the catalytic site (black asterisk) and becomes hydroxylated. The model visualizes also the proposed disulfide bonds Cys276–Cys293 and Cys486–Cys511 of the α subunit. The active sites of the PDI/β subunits are indicated by red asterisks. The location of the known glycosylation sites of the α subunits are marked by grey polygons. KDEL of the β/PDI subunit concerns the ER-retention signal).

Future Goals

The research concerning the chemistry and transition state stabilisation of the catalysis by the CoA-dependent enzymes remains an important topic of our studies. We are also implementing stopped-flow and quenched-flow experiments to study the substrate channelling properties of MFE1 and TFE by pre-steady state fast kinetics. Also mass spec approaches to measure the presence/absence in the bulk solvent of intermediates of the MFE1 and TFE reactions have been initiated. At the molecular level, substrate channelling properties are similar to processivity properties, as observed for example in C-P4H. In the latter example the enzyme C-P4H remains bound to the substrate (procollagen) during several subsequent hydroxylation reactions. The crystallisation studies of C-P4Hs and their truncated variants will be continued, together with studying their enzymological properties. It will be important to “lock” the molecules within the same protein sample in an identical conformation to increase the chances of getting suitable crystals.  The possible heterogeneity of C-P4H can be related to partially wrongly formed SS-bridges, but also to the complicated reaction mechanism, which is predicted to have an “open” conformations (without bound peptide substrate, to exchange the cosubstrate) and a “closed” conformation (with peptide bound, to allow for the hydroxylation chemistry).

Other references

Anantharajan J, Koski MK, Kursula P, Hieta R, Bergmann U, Myllyharju J, Wierenga RK. The structural motifs for substrate binding and dimerization of the α subunit of collagen prolyl 4-hydroxylase. Structure. 21:2107-2118. 2013.

De Jong L, van der Kraan I, de Waal A. The kinetics of the hydroxylation of procollagen by prolyl 4-hydroxylase. Proposal for a processive mechanism of binding of the dimeric hydroxylating enzyme in relation to the high kcat/Km ratio and a conformational requirement for hydroxylation of -X-Pro-Gly- sequences. Biochim. Biophys. Acta. 1079: 103-111, 1991.

Guo, H, Salahub DR. Cooperative hydrogen bonding and enzyme catalysis. Angew. Chem. Int. Ed. 37:2985-2990, 1998.

Jencks WP. Economics of enzyme catalysis. Cold Spring Harb. Symp. Quant. Biol. 52:65-73, 1987.

Myllyharju J. Prolyl 4-hydroxylases, key enzymes in the synthesis of collagens and regulation of the response to hypoxia, and their roles as treatment targets. Ann Med. 40:402-417, 2008.

Yao KW, Schulz H. Intermediate channeling on the trifunctional beta-oxidation complex from pig heart mitochondria. J Biol Chem. 271:17816-17820. 1996.

Publications 2017-

Harijan RK, Kiema TR, Syed SM, Qadir I, Mazet M, Bringaud F, Michels PAM, Wierenga RK. Crystallographic substrate binding studies of Leishmania mexicana SCP2-thiolase (type-2): unique features of oxyanion hole-1. PEDS, 30:225-233, 2017.

Kasaragod P, Midekessa GB, Sridhar S, Schmitz W, Kiema TR, Hiltunen J, Wierenga RK. Structural enzymology comparisons of multifunctional enzyme, type-1 (MFE1): the flexibility of its dehydrogenase part. FEBS Open Bio. 7:1830-1842, 2017.

Koski, MK., Anantharajan J, Kursula P, Dhavala P, Murthy AV, Bergmann U, Myllyharju J, Wierenga RK. Assembly of the elongated collagen prolyl 4-hydroxylase α2β2 heterotetramer around a central α2 dimer. Biochem J. 474:751-769, 2017.

Wierenga RK, Ringe D. The EMBO biocatalysis conference "The biochemistry and chemistry of biocatalysis: from understanding to design". PEDS, 30:141-141, 2017.

Research Group Members

Project Leader:
Rik Wierenga (University of Oulu)

Senior and Post-doctoral Investigators:
Rajaram Venkatesan (principal investigator, TFE, Academy of Finland)
Tiila-Riikka Kiema (senior postdoc, thiolases, IVDH, MFE1, maintenance of the diffraction unit, Biocenter Oulu)
Kristian Koski (senior postdoc, C-P4H, University of Oulu strategic funding targeted for BF operations)
Gabriele Cordara (postdoc, TIM, ECI2, Biocenter Oulu, Academy of Finland)

PhD Students:
Abhinandan Murthy (PhD student, human C-P4H, University of Oulu)
Shiv Kumar Sah-Teli (PhD student, TFE, Academy of Finland)
Shruthi Sridhar (PhD student, MFE1, Biocenter Oulu)
Subhadra Dalwani (PhD student, TFE, Academy of Finland)

ProGradu Students:
Ramita Sulu (P4H)
Mikko Hynonen (TFE)

Undergraduate Students:
Akke-Pekka Törmänen (TIM, ECI2)
Hossam ElDien Montaser (TFE)
Sarah Wazir  (TFE)  
Nguyen Vinh Hong Tran (thiolase)
Danilo Kimio Hirabae De Oliveira  (C-P4H)

Laboratory Technicians:
Ville Ratas (technician, crystallisation of proteins, Biocenter Oulu)
Ed Daniel (programmer, development of IceBear, Academy of Finland, Diamond Light Source, UK)

Foreign Scientists, 13

National and International Activities

Instruct-FI, BF, vice-chair of the Biocenter Finland Structural Biology network, Instruct-FI

University of Oulu, organiser International X-ray course: “X-ray and neutron diffraction studies of macromolecules: from data collection to structures”, May 15-19, 2017

International Activities:

BESSY, Helmholtz-Berlin, Germany, member of the macromolecular crystallography SSP evaluation panel

EU Projects (present and in progress):

iNEXT, EU, coordinator of the Oulu iNEXT TID-centre

Instruct-ULTRA, EU, coordinator of the call1-project: “Towards standard APIs for the exchange of metadata between homelab LIMS software and ISPyB”

Last updated: 15.10.2018