Appropriate Activity Assays Are Crucial for the Specific Determination of Proline Dehydrogenase and Pyrroline-5-Carboxylate Reductase Activities

Abstract

Accumulation of proline is a widespread plant response to a broad range of environmental stress conditions including salt and osmotic stress. Proline accumulation is achieved mainly by upregulation of proline biosynthesis in the cytosol and by inhibition of proline degradation in mitochondria. Changes in gene expression or activity levels of the two enzymes catalyzing the first reactions in these two pathways, namely pyrroline-5-carboxylate (P5C) synthetase and proline dehydrogenase (ProDH), are often used to assess the stress response of plants. The difficulty to isolate ProDH in active form has led several researchers to erroneously report proline-dependent NAD⁺ reduction at pH 10 as ProDH activity. We demonstrate that this activity is due to P5C reductase (P5CR), the second and last enzyme in proline biosynthesis, which works in the reverse direction at unphysiologically high pH. ProDH does not use NAD⁺ as electron acceptor but can be assayed with the artificial electron acceptor 2,6-dichlorophenolindophenol (DCPIP) after detergent-mediated solubilization or enrichment of mitochondria. Seemingly counter-intuitive results from previous publications can be explained in this way and our data highlight the importance of appropriate and specific assays for the detection of ProDH and P5CR activities in crude plant extracts.

Keywords: electron acceptor; enzyme activity assay; mitochondria; proline dehydrogenase; protein extraction; pyrroline-5-carboxylate reductase.

FIGURE 1

Reactions and enzymes known to produce or degrade proline in plants. GSA, glutamate-5-semialdehyde; P5C, pyrroline-5-carboxylate; P5CR: P5C reductase; ProDH, proline dehydrogenase, Q/QH₂, oxidized/reduced ubiquinone.

FIGURE 2

Enzyme activities in bacterial extracts. Soluble protein extracts from E. coli cells carrying an empty vector or overexpressing either GST:ProDH1ΔN12, GST:ProDH2ΔN13 or 6xHis:P5CR were used for different enzyme assays. (A) Coomassie-stained gel with the positions of molecular weight markers and those expected for the recombinant proteins. (B) Proline-dependent 2,6-dichlorophenolindophenol (DCPIP) reduction (red columns) and P5C-dependent NADPH oxidation (light blue columns) were measured at pH 7.5, while proline-dependent NAD⁺ reduction (blue columns) was measured at pH 10. Data are the average (±SD) of technical triplicates. Independent protein preparations gave very similar results; n.d., not detected.

FIGURE 3

Activities of purified recombinant GST:ProDH1ΔN12. Crude protein extract of bacteria overexpressing GST:ProDH1ΔN12 and affinity-purified GST:ProDH1ΔN12 (indicated by an arrowhead, calculated molecular weight is 80.3 kDa) were assayed for proline-dependent 2,6-dichlorophenolindophenol (DCPIP) reduction and P5C-dependent NADPH oxidation at pH 7.5, as well as for proline-dependent NAD⁺ reduction at pH 10. Note that the scales of the two y-axes are in an opposite ratio than in Figure 2. Data are the average (±SD) of technical triplicates. Two further, independent protein preparations gave very similar results; n.d., not detected. The inset shows a Coomassie-stained, denaturing protein gel of the assayed fractions. Lane 1: insoluble proteins, lane 2: soluble extract [in the presence of 0.1% (w/v) dodecyl maltoside], lane 3: flow-through of the glutathione agarose column, lane 4: purified GST:ProDH1ΔN12.

FIGURE 4

Enzymatic activities of purified P5CR. 6xHis:P5CR was purified and its activity was analyzed photometrically under various conditions. (A) P5C-dependent oxidation of NADPH at pH 7.5, the physiological forward-reaction of P5CR, was followed over time with different amounts of enzyme in the assay. (B) Proline-dependent reduction of NAD⁺, the reverse reaction of P5CR, was monitored at two different pH values and with different amounts of enzyme. (C) Reduction of 2,6-dichlorophenolindophenol (DCPIP) at pH 7.5 with various amounts of enzyme in the presence or absence of proline. (D) Reduction of DCPIP at pH 10 with various amounts of enzyme in the presence or absence of proline. All data are from single kinetic analyses. Very similar results were obtained with at least three independently purified enzyme preparations.

FIGURE 5

ProDH and P5CR activity levels in Arabidopsis seedlings in response to dark-induced senescence. Twelve-day-old Arabidopsis seedlings grown on 0.5x Murashige and Skoog solid medium were transferred to water and darkness for 3 days to trigger senescence and ProDH1 and ProDH2 expression. Enzyme activities were measured in soluble protein extracts prepared with or without 0.5% (v/v) triton-X-100 from wildtype (Col-0) seedlings and prodh1-4/prodh2-2 double mutants. ProDH activity (red bars) was calculated from the proline-dependent reduction of the artificial electron acceptor 2,6-dichlorophenolindophenol (DCPIP) at pH 7.5. P5CR activity (light blue bars) was measured through P5C-dependent oxidation of NADPH at pH 7.5. Reverse P5CR activity (blue bars) was detected based on proline-dependent NAD⁺ reduction at pH 10. Data are means (±SD) of three independent biological replicates. n.d.: ProDH activity not detected.

FIGURE 6

ProDH and P5CR activities in isolated mitochondria. Crude mitochondrial fractions were prepared from Arabidopsis wildtype (Col-0) and prodh1-4/prodh2-2 double mutant leaves after 5 days of dark treatment. Isolated mitochondria were assayed for proline-dependent 2,6-dichlorophenolindophenol (DCPIP) reduction at pH 7.5 (red columns) and proline-dependent NAD⁺ reduction at pH 10 (blue columns). P5C-dependent NADPH oxidation at pH 7.5 was not detected (n.d.). Data are means (±SD) of three independent preparations of mitochondria.