Life without complex I: proteome analyses of an Arabidopsis mutant lacking the mitochondrial NADH dehydrogenase complex

Abstract

The mitochondrial NADH dehydrogenase complex (complex I) is of particular importance for the respiratory chain in mitochondria. It is the major electron entry site for the mitochondrial electron transport chain (mETC) and therefore of great significance for mitochondrial ATP generation. We recently described an Arabidopsis thaliana double-mutant lacking the genes encoding the carbonic anhydrases CA1 and CA2, which both form part of a plant-specific 'carbonic anhydrase domain' of mitochondrial complex I. The mutant lacks complex I completely. Here we report extended analyses for systematically characterizing the proteome of the ca1ca2 mutant. Using various proteomic tools, we show that lack of complex I causes reorganization of the cellular respiration system. Reduced electron entry into the respiratory chain at the first segment of the mETC leads to induction of complexes II and IV as well as alternative oxidase. Increased electron entry at later segments of the mETC requires an increase in oxidation of organic substrates. This is reflected by higher abundance of proteins involved in glycolysis, the tricarboxylic acid cycle and branched-chain amino acid catabolism. Proteins involved in the light reaction of photosynthesis, the Calvin cycle, tetrapyrrole biosynthesis, and photorespiration are clearly reduced, contributing to the significant delay in growth and development of the double-mutant. Finally, enzymes involved in defense against reactive oxygen species and stress symptoms are much induced. These together with previously reported insights into the function of plant complex I, which were obtained by analysing other complex I mutants, are integrated in order to comprehensively describe 'life without complex I'.

Keywords: Arabidopsis thaliana; carbonic anhydrase; complex I; mitochondrial metabolism; photosynthesis; proteomics; respiratory chain..

Fig. 1.

Comparative analysis of the mitochondrial proteomes of Arabidopsis wt and ca1ca2 lines. Mitochondria were isolated as described in the Materials and Methods. Total mitochondrial protein was separated by 2D IEF/SDS PAGE and proteins were stained by Coomassie blue. Three replicates were produced per line and used for the calculation of master gels (Delta 2D software package, Decodon, Germany). The molecular masses of standard proteins are given to the left of the 2-D gel (in kDa). Isoelectric focusing range is from pH 3 (left) to pH 11 (right). Proteins indicated in pink are more abundant in the mutant (>1.5-fold increase); proteins indicated in green are less abundant in the mutant (>1.5-fold decrease). Spots indicated by numbers were identified by mass spectrometry (for results see Supplementary Table S1).

Fig. 2.

Comparative analysis of the mitochondrial membrane proteomes of Arabidopsis wt and ca1ca2 lines. Mitochondria were isolated as described in the Materials and methods. Mitochondrial membrane proteins were separated by 2D BN/SDS PAGE and proteins were stained by colloidal Coomassie (A). Three replicates were produced per fraction and used for the calculation of a master gel (Delta 2D software package, Decodon, Germany) (B). The molecular masses of standard proteins are given to the left of the 2-D gels (in kDa). OXPHOS complexes are boxed in (B); their identities are given above the gels (I, complex I; V, complex V; III₂, dimeric complex III; I+III₂, supercomplex formed of complex I and dimeric complex III; F₁, F₁ part of complex V; IV, complex IV; II, complex II). Proteins indicated in pink are less abundant in the mutant (>1.5-fold decrease); proteins indicated in green are more abundant in the mutant (>1.5-fold increase). Spots indicated by numbers were identified by mass spectrometry (for results see Supplementary Table S2).

Fig. 3.

Comparative analysis of the mitochondrial membrane proteomes of Arabidopsis wt and ca1ca2 lines by differential gel electrophoresis (DIGE). Mitochondria were isolated as described in the Materials and Methods. Mitochondrial membrane proteins of wt and ca1ca2 were labeled with different CyDyes and separated by 2D BN/SDS PAGE. Proteins were stained by colloidal Coomassie (A). The same gel was used for fluorescence detection of the two CyDyes (B). The molecular masses of standard proteins are given to the left of the 2-D gel (in kDa). The identities of selected mitochondrial protein complexes are given above the gels (I, complex I; V, complex V; III₂, dimeric complex III; I+III₂, supercomplex formed of complex I and dimeric complex III; F₁, F₁ part of complex V; IV, complex IV; II, complex II). Proteins indicated in red are less abundant in the ca1ca2 mutant (>1.5-fold decrease) and proteins indicated in green are more abundant in the ca1ca2 mutant (>1.5-fold increase). Proteins given in yellow are not changed in abundance. Spots indicated by numbers were identified by mass spectrometry (for results see Supplementary Table S3). Note: if compared to the comparative experiment shown in Fig. 2, several subunits of OXPHOS complexes appear to be absent in the DIGE approach. This is due to the fact that CyDye labeling takes place at native conditions. As a consequence, only proteins exposed to the surface of protein complexes are labeled. In contrast, image evaluation based on the Delta 2D approach (Fig. 2) allows visualization of all subunits of a protein complex.

Fig. 4.

Subcellular localization of proteins of altered abundances in the ca1ca2 line as obtained by label free quantitative shotgun proteomics. Total protein was extracted from wt and ca1ca2 mutant plants at a similar developmental stage. Proteins were identified and quantified by shotgun MS (for details see the Material and Methods). Predicted localizations of proteins of changed abundances between the two lines were obtained from the SUBA3 database (http://suba3.plantenergy.uwa.edu.au/). (A) Predicted localization of all proteins changed (P-value <0.05). (B) Predicted localization of proteins more abundant in ca1ca2 compared to wt. (C) Predicted localization of proteins less abundant in ca1ca2 mutant compared to wt. Numbers indicate amounts relative to the sum of altered protein species (%).

Fig. 5.

Functional annotation of proteins identified by quantitative label-free shotgun MS. The 318 proteins of differential abundances between wt and ca1ca2 mutant plants were grouped into functional BINs using MapMan. BINs are given to the right. Proteins more abundant in ca1ca2 are indicated in green and proteins that are less abundant in red (P-value <0.05). Each square represents one protein. BINs without any identified protein are with a grey dot (for results see Supplementary Table S4).

Fig. 6.

Life without complex I. The figure is based on altered protein levels in ca1ca2 plants relative to wildtype plants as obtained by label-free quantitative shotgun proteomics. Green arrows within the grey boxes indicate increased protein levels in the double-mutant, and red arrows indicate decreased levels. Absence of complex I causes reorganization of the cellular respiration system. Since electron insertion into the first segment of the mETC is not possible, increased electron insertion at later segments takes place (induction of complexes II, IV). This requires increased oxidation of organic substrates (induction of enzymes of glycolysis, the TCA cycle, and amino acid catabolism). Mitochondrial ATP formation most likely is still reduced, which requires increased fermentation. The growth rate of the double-mutant is drastically reduced. This is reflected by reduced amounts of the two photosystems, Calvin cycle enzymes, and enzymes of the tetrapyrrole biosynthesis pathway. Furthermore, altered metabolism and electron transport pathways in the mitochondria and chloroplasts cause increased ROS formation and stress symptoms. Several components of the ROS and stress defense system are induced in the double-mutant, as is the alternative oxidase, a well-known stress indicator in plants. Note: causal events indicated by black arrows do not necessarily indicate primary effects, but may well represent indirect consequences. For further details see the discussion section.