Discovery and characterization of regulatory mechanisms affecting the heteromeric acetyl-coenzyme a carboxylase in Arabidopsis

Abstract

Fatty acid biosynthesis (FAS) is an essential metabolic pathway used by all organisms to generate fatty acids. A staple component of this pathway is the enzyme acetyl-CoA carboxylase (ACCase), which catalyzes the committed step by converting acetyl-CoA to malonyl-CoA. The heteromeric form of this enzyme requires four different subunits for activity: biotin carboxylase, biotin carboxyl carrier protein (BCCP), and alpha- and beta-carboxyltransferase (CT). Heteromeric ACCase is present in prokaryotes and the plastids of most plants, and has been a focus of biotechnology research due to its prominent role in FAS. Many different regulatory mechanisms have been identified in both plants and E. coli. However, it is still unknown how most of these regulatory mechanisms are mediated. For example, ACCase is known to be feedback inhibited by 18:1-acyl carrier protein in plants, yet it is unknown how this inhibition is exerted on the enzyme. Therefore it was posited that other unknown factors, such as proteins or post-translational modifications, might play a role in ACCase regulation. To identify suspected regulatory factors associated with ACCase, we performed in vivo co-immunoprecipitation (co-IP) using subunit-specific antibodies to isolate the ACCase complex from Arabidopsis thaliana leaves. Quantitative mass spectrometry of these co-IPs revealed all four known subunits to ACCase and two unknown proteins annotated as 'biotin/lipoyl attachment domain containing' (BADC) proteins. The BADC proteins are a family of three proteins in A. thaliana and resemble the BCCP subunit to ACCase, but lack the conserved biotinylation motif. All three BADC proteins interacted with the two A. thaliana BCCP isoforms and the biotin carboxylase subunit of ACCase based on yeast two-hybrid and heterologous co-expression analyses. None of the BADC proteins were biotinylated in planta or when expressed in Escherichia coli, unlike BCCP controls. Gene orthologs to BADC were found only in plant and green algae species that contain a heteromeric ACCase suggesting BADC genes co-evolved with this form of ACCase. Expression of BADC proteins in a temperature-sensitive E. coli BCCP mutant in minimal media strongly inhibited cell growth through interaction with the homologous, bacterial ACCase. Also, addition of recombinant BADC protein to in vitro ACCase activity assays significantly reduced enzyme activity. Finally, partial silencing of one of the BADC genes in A. thaliana seed led to a slight, yet significant, increase in seed oil content. We conclude the BADC proteins are ancient BCCPs that acquired a new function through mutation of the biotinylation motif. We propose a poisoned complex model whereby BADCs function as negative regulators of ACCase by competing with BCCP for access to the holo-ACCase complex. In addition, a study was performed to identify the role of phosphorylation of the alpha-CT subunit. Multiple studies had identified two phosphorylation sites on the C-terminal domain of alpha-CT in A. thaliana. This C-terminal domain is not found in all plant species and has an unknown function. To determine the potential regulatory effect of phosphorylation on this domain, phosphomimic and phospho-deficient alpha-CT mutants were made and expressed in wild type A. thaliana. Multiple independent transgenic lines containing at least two-fold alpha-CT protein compared to empty vector controls were screened for seed oil content. The resulting data showed no clear phenotype that could be attributed to expression of the mutants. This result could be explained by a number of factors such as the presence of endogenous alpha-CT, the complexity of the seed oil phenotype, or a large margin of technical error in some lines. However, in vitro ACCase activity assays showed that a transgenic line overexpressing native alpha-CT contained increased specific activity of the enzyme compared to controls. Furthermore, analysis of transgenic lines expressing phosphomimic or phospho-deficient alpha-CT mutants also showed increased ACCase specific activity which was indistinguishable from the native alpha-CT overexpression line, regardless of the mutation. Therefore it appears that increased alpha-CT expression can increase ACCase activity by allowing for the formation of more active complexes. This observation suggests that alpha-CT is the limiting subunit of the ACCase complex in the stroma.