Arabidopsis NAD-malic enzyme functions as a homodimer and heterodimer and has a major impact on nocturnal metabolism

Abstract

Although the nonphotosynthetic NAD-malic enzyme (NAD-ME) was assumed to play a central role in the metabolite flux through the tricarboxylic acid cycle, the knowledge on this enzyme is still limited. Here, we report on the identification and characterization of two genes encoding mitochondrial NAD-MEs from Arabidopsis (Arabidopsis thaliana), AtNAD-ME1 and AtNAD-ME2. The encoded proteins can be grouped into the two clades found in the plant NAD-ME phylogenetic tree. AtNAD-ME1 belongs to the clade that includes known alpha-subunits with molecular masses of approximately 65 kD, while AtNAD-ME2 clusters with the known beta-subunits with molecular masses of approximately 58 kD. The separated recombinant proteins showed NAD-ME activity, presented comparable kinetic properties, and are dimers in their active conformation. Native electrophoresis coupled to denaturing electrophoresis revealed that in vivo AtNAD-ME forms a dimer of nonidentical subunits in Arabidopsis. Further support for this conclusion was obtained by reconstitution of the active heterodimer in vitro. The characterization of loss-of-function mutants for both AtNAD-MEs indicated that both proteins also exhibit enzymatic activity in vivo. Neither the single nor the double mutants showed a growth or developmental phenotype, suggesting that NAD-ME activity is not essential for normal autotrophic development. Nevertheless, metabolic profiling of plants completely lacking NAD-ME activity revealed differential patterns of modifications in light and dark periods and indicates a major role for NAD-MEs during nocturnal metabolism.

Figure 1.

Phylogenetic tree of plant NAD-MEs. Mature proteins were aligned using the ClustalW (1.81) multiple alignment program (Thompson et al., 1994), and the alignment obtained was modified by visual inspection to exclude the sites containing gaps. The phylogenetic tree was constructed by the neighbor-joining method using the Phylip software package (Felsenstein, 1989). Statistical significance of each branch of the tree was evaluated by bootstrap analysis by 100 iterations of bootstrap samplings and reconstruction of trees by the neighbor-joining method. The topology obtained by this method is shown along with statistical significance of each branch. The following sequences are included: α-subunits from A. hypochondriacus (U01162; photosynthetic NAD-ME), Arabidopsis (At2g13560), potato (Z23023), and Oryza sativa (NM_001066235), and β-subunits from Arabidopsis (At4g00570), O. sativa (NM_001071533), and potato (Z23002).

Figure 2.

Expression analysis of AtNAD-ME in different tissues. A, Expression of AtNAD-ME1 transcript relative to AtNAD-ME2 mRNA levels for each organ analyzed by qRT-PCR. B, Relative expression of AtNAD-ME1 and AtNAD-ME2 in different organs with respect to leaf analyzed by qRT-PCR. The y axis refers to the fold-difference of a particular AtNAD-ME transcript level relative to its amount found in leaf. The asterisks indicate that the expression values obtained were statistically significantly different from the ones obtained for leaf as determined by the Student's t test (P < 0.05). C, Analysis of AtNAD-ME∷GUS expression. a to j, AtNAD-ME1∷GUS; k to s, AtNAD-ME2∷GUS. a to d and k to m, Seedlings at 3, 5, and 12 DAI. e and n, Three-week-old rosette. f and o, Longitudinal section through the stem. g and p, Root system of a 3-week-old plant. h and q, Close-ups of the roots. i and r, Inflorescence with flowers at different stages. j and s, Siliques.

Figure 3.

Recombinant Arabidopsis NAD-ME isoforms analyzed by gel electrophoresis. A, Coomassie-stained SDS-PAGE of recombinant NAD-ME isoforms. Five micrograms of purified recombinant AtNAD-ME1 and AtNAD-ME2 before (1) and after (2) enterokinase digestion was loaded in each case. The estimated molecular mass of the purified proteins is indicated on the right. B, Native-PAGE stained for NAD-ME activity. Approximately 20 milliunits of AtNAD-ME1 and AtNAD-ME2 were loaded, as well as a mixture of equal amounts of both proteins. A mitochondrial leaf crude extract (L, 20 milliunits) was also loaded in the gel. C, Western blot of native-PAGE using antibodies against A. hypochondriacus α-NAD-ME. Approximately 5 μg of NAD-ME1 and AtNAD-ME2 were loaded, as well as a mixture of equal amounts of both proteins. A mitochondrial leaf crude extract (L, 30 μg) was also loaded in the gel. Molecular mass markers (M) were run in parallel and stained with Coomassie Blue.

Figure 4.

SDS- and native-PAGE of mitochondrial extracts analyzed for activity or by western blot. A, SDS-PAGE of leaf mitochondrial extracts (L, 50 μg of total protein) analyzed by western blot using antibodies against AtNAD-ME1 (a-AtME1) or AtNAD-ME2 (a-AtME2), or, alternatively, against A. hypochondriacus α-NAD-ME (a-AhαME). As control, recombinant AtNAD-ME1 (3 μg, ME1) and AtNAD-ME2 (3 μg, ME2) were loaded. B, Native-PAGE of Arabidopsis mitochondrial extracts stained for NAD-ME activity. Approximately 20 milliunits of NAD-ME from mitochondrial crude extracts from leaf (L), stem (S), root (R), and flower (F) was loaded in each lane. Molecular mass markers were run in parallel and stained with Coomassie Blue. C, The active band from leaf mitochondrial crude extracts (excised band from B–L) was excised and analyzed by SDS-PAGE followed by western-blot analysis using antibodies against AtNAD-ME1 (a-AtME1) or AtNAD-ME2 (a-AtME2). As controls, recombinant purified AtNAD-ME1 and AtNAD-ME2 (3 μg, ME1 and ME2) were loaded. D, Western-blot analysis of native-PAGE of mitochondrial crude extracts (30 μg) from leaf (L), stem (S), root (R), and flower (F) using antibodies against AtNAD-ME1 (a-AtME1) or AtNAD-ME2 (a-AtME2). As control, recombinant purified AtNAD-ME1 and AtNAD-ME2 (3 μg, ME1 and ME2) were loaded. The molecular masses of the marker proteins run in parallel are indicated.

Figure 5.

Identification of nad-me insertion lines. A, AtNAD-ME gene structure showing the locations of the T-DNA insertions in the knockout mutants. The orientation of the T-DNA insertion is indicated as left border (LB). B, Semiquantitative RT-PCR showing the absence of the corresponding AtNAD-ME transcript in the single (nad-me1 and nad-me2) and double (1 × 2) T-DNA knockout lines. PCR products of 750 (NAD-ME1) and 928 (NAD-ME2) bp were amplified using 35 cycles. As loading control, a 521-bp Actin2 cDNA fragment was amplified by 29 cycles. C, NAD-ME activity (units/mg) in different organs of Arabidopsis wild type and the T-DNA insertion lines. The bars indicate the sd of the measurements from three different crude extract preparations. The activity measurement was performed three independent times with each crude extract preparation, with <5% sd within each preparation.

Figure 6.

Diurnal changes in NAD-ME activity and expression level in wild-type leaves. A, Total NAD-ME activity at the end of the day and night periods. The bars indicate the sd of the measurements from three different crude extracts preparations. The activity measurement was performed three independent times with each crude extract preparation. The asterisk indicates that the NAD-ME activity determined at the end of the night period was statistically significantly higher than the one at the end of the light period as determined by the Student's t test (P < 0.05). B, SDS-PAGE of leaf crude extracts (50 μg of total protein) prepared at the end of the light and night periods and analyzed by western blot using antibodies against AtNAD-ME1 (a-AtME1) or AtNAD-ME2 (a-AtME2). As control of the amount of protein loaded, antibodies against A. viridis PEPc (a-AvPEPc) were also used for detection. The estimated molecular mass of the immunoreactive bands is indicated on the left. C, Expression of AtNAD-ME1 and AtNAD-ME2 transcripts by the end of the night period relative to the expression by the end of the day period analyzed by qRT-PCR. The bars indicate the sd of measurements from three different biological replicates. For both genes, the relative expression values obtained at the end of the night period were statistically significantly higher than the ones at the end of the light period as determined by the Student's t test (P < 0.05).

Figure 7.

Scheme representing the flux of metabolites in Arabidopsis lacking NAD-ME activities. The direction of the arrows indicates accumulation or depletion of the respective metabolite. During the night period, mitochondrial pyruvate derives from glycolysis and from vacuolar malate reserves through the action of NAD-ME in leaves.