Subunit organisation and assembly of the 2-Oxoglutarate dehydrogenase multienzyme complex (OGDC)

Al-Alaway, Adel Ibrahim Ahmad (2013) Subunit organisation and assembly of the 2-Oxoglutarate dehydrogenase multienzyme complex (OGDC). PhD thesis, University of Glasgow.

Abstract

The family of mammalian 2-oxoacid dehydrogenase complexes (PDC, BCOADC and OGDC) are stable, high Mr assemblies composed of multiple copies of 3 separate enzymes (E1, E2 and E3) that catalyse key stages in carbohydrate and amino acid metabolism. Their respective E1 and E2 enzymes are complex-specific while E3 is the identical gene product in all 3 complexes.

In general terms, the oligomeric E2 ‘cores’ provide the structural and mechanistic framework to which their partner E1 and E3 enzymes are tethered tightly but non-covalently. However, the mode of E1 and E3 binding differs significantly from complex to complex. In the BCOADC, its cubic E2 core is composed of 24 identical subunits to which E1b and E3 bind stably in a mutually-exclusive fashion via multiple subunit binding domains (SBDs). In a variation of this theme, the icosahedral (60-meric) E2-PDC core comprises 2 types of subunit, E2 and an E2-related polypeptide, E3 binding protein (E3BP). In this case, E1p and E3 bind independently to specific SBDs located on E2 and E3BP, respectively. In contrast, OGDC differs significantly from its counterparts as its 24-meric E2 core does not contain any apparent SBDs. In addition, there is no equivalent to E3BP in this complex. Hence, how stable complex formation is achieved for the OGDC is still an area of active research, particularly in view of increasing evidence implicating OGDC deficiency as a major causative factor in a variety of neurodegenerative and oxidative stress disorders.

Previous subunit-specific proteolysis, enzymatic and immunological studies on native bovine OGDC in our laboratory have suggested that an intact E1o N-terminal region is vital for maintaining the structural integrity of the complex. In particular, a single cleavage of E1o at Arg77 results in complete loss of OGDC function stemming from dissociation of both E3 and a large, active E1o species (E1') from the native E2 core assembly.

The principal aim of this thesis was to establish the location and precise nature of the domains responsible for protein-protein interactions between the constituent E1o, E2o and E3 enzymes of OGDC and their roles in assembly, taking into account our previous data and the unique domain organisation of E2. It was also a goal to produce a recombinant version of the human OGDC for future biomedical studies including genetic analysis of naturally-occurring mutant forms.

Initially, the cloning, expression and purification of a series of E1o N-terminal constructs (His-tag, GST or MBP fusion proteins: 60, 90 and 153 a.a.s in length) is described extending from Ser1 to Phe153 of mature human E1o. Access to 10-30 mg of highly-purified E1o N-terminal peptides was required to enable testing of the ability of this region to interact with E3 (and also E2) employing a range of biochemical and biophysical techniques. High-level expression of full-length E1o was also achieved; however, attempts to produce active E1o in soluble form proved unsuccessful. Recombinant human E2o and E3 were both produced as soluble active enzymes in high yield.

A preliminary structural characterisation of the E1o N-terminal region was also undertaken employing synthetic peptides, circular dichroism and a basic bio-informatics approach. These studies demonstrated that the N-terminal region had the potential to form 2 short α-helical segments linked by regions of unstructured and flexible polypeptide chain. Moreover, a 3D-structural prediction for mature, full-length human E1o confirmed that its N-termini were highly accessible, extending above the enzyme surface and situated in close proximity at one end of the homodimer. Although there was no apparent sequence homology, our data also suggested that this region had several of the main structural features of the E3-SBD located on E3BP.

Direct evidence that the N-terminal region of E1 bound to E3 post-translationally was obtained using peptide array technology, alanine scanning, isothermal titration calorimetry (ITC), affinity chromatography and gel filtration (GFC). Interaction of E1o N-terminal peptides with E3 was salt-sensitive and reduced over the range 0-0.5M NaCl, i.e. similar to that required to promote E3 dissociation from intact OGDC. Only the longer E1o-90 and E1o-153 constructs complexed with E3 and evidence suggested that steric hindrance by the fusion partner blocked E3 binding to the short E1o-60 GST fragment.

As expected, in the absence of any obvious SBD on E2o, no direct interaction could be detected between E2o and E3 using ITC or GFC. In addition, no post-translation association occurred between our purified E1o constructs and fully-assembled, oligomeric E2o. In contrast, co-expression of E1o-90 and E1o-153 constructs with E2o (but not E1o-60) resulted in the formation of an E1o-90/E2o GST or E1o-153/E2o GST sub-complex. Moreover, these N-terminal E1o/E2o sub-complexes supported stable E3 binding as judged by affinity chromatography and GFC. Taken together, the data presented in this thesis have established that the E1o N-terminal region is pivotal for mediating formation of a stable multi-enzyme assembly by directing self-integration with the E2o core during or immediately after synthesis and subsequently promoting high-affinity E3 binding in a manner reminiscent of E3BP integration with the oligomeric E2-PDC core.

To test the functional importance of the putative E3-binding domain on E1o, the E1o constructs were employed as inhibitors of OGDC and PDC activity since it was anticipated that they should displace E3 from its normal binding site in the intact complexes. Incubation of OGDC or PDC with the E1o-90 and E1o-153 (but not E1o-60) constructs caused preferential inhibition of OGDC activity. Conversely, an E3BP-SBD construct was a more effective inhibitor of PDC suggesting that the mode of E3 interaction differs significantly in these 2 complexes.

In summary, our data have provided (a) new insights into the structure, organisation and mode of assembly of the mammalian OGDC; (b) suggested new approaches to producing a recombinant model OGDC as an important biological tool for future biomedical and genetic studies and (c) raised new questions concerning the subunit composition, architecture, and stoichiometry of its core assembly.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: 2-Oxoglutarate Dehydrogenase Multienzyme Complex, 2-Oxoglutarate Dehydrogenase, 2-oxoacid dehydrogenase complex, PDC, BCOADC, OGDC, E1, E2, E3, E1o, E2o, E2 core, E2o core,Pyruvate dehydrogenase, Dihydrolipoyl succinyltransferase, Dihydrolipoamide dehydrogenase, E3-binding protein, E3BP, N-terminal, Subunit binding domain, SBD, alpha ketoglutarate dehydrogenase, alpha keto acid.
• Subjects: Q Science > QH Natural history > QH301 Biology
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Molecular Biosciences
• Supervisor's Name: Lindsay, Prof. John Gordon
• Date of Award: 2013
• Depositing User: Dr. ADEL AL-ALAWY
• Unique ID: glathesis:2013-4600
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 02 Oct 2013 12:35
• Last Modified: 02 Oct 2013 12:37
• URI: https://theses.gla.ac.uk/id/eprint/4600