Early Cold-Induced Peroxidases and Aquaporins Are Associated With High Cold Tolerance in Dajiao (Musa spp. 'Dajiao')

Abstract

Banana is an important tropical fruit with high economic value. One of the main cultivars ('Cavendish') is susceptible to low temperatures, while another closely related specie ('Dajiao') has considerably higher cold tolerance. We previously reported that some membrane proteins appear to be involved in the cold tolerance of Dajiao bananas via an antioxidation mechanism. To investigate the early cold stress response of Dajiao, here we applied comparative membrane proteomics analysis for both cold-sensitive Cavendish and cold-tolerant Dajiao bananas subjected to cold stress at 10°C for 0, 3, and 6 h. A total of 2,333 and 1,834 proteins were identified in Cavendish and Dajiao, respectively. Subsequent bioinformatics analyses showed that 692 Cavendish proteins and 524 Dajiao proteins were predicted to be membrane proteins, of which 82 and 137 differentially abundant membrane proteins (DAMPs) were found in Cavendish and Dajiao, respectively. Interestingly, the number of DAMPs with increased abundance following 3 h of cold treatment in Dajiao (80) was seven times more than that in Cavendish (11). Gene ontology molecular function analysis of DAMPs for Cavendish and Dajiao indicated that they belong to eight categories including hydrolase activity, binding, transporter activity, antioxidant activity, etc., but the number in Dajiao is twice that in Cavendish. Strikingly, we found peroxidases (PODs) and aquaporins among the protein groups whose abundance was significantly increased after 3 h of cold treatment in Dajiao. Some of the PODs and aquaporins were verified by reverse-transcription PCR, multiple reaction monitoring, and green fluorescent protein-based subcellular localization analysis, demonstrating that the global membrane proteomics data are reliable. By combining the physiological and biochemical data, we found that membrane-bound Peroxidase 52 and Peroxidase P7, and aquaporins (MaPIP1;1, MaPIP1;2, MaPIP2;4, MaPIP2;6, MaTIP1;3) are mainly involved in decreased lipid peroxidation and maintaining leaf cell water potential, which appear to be the key cellular adaptations contributing to the cold tolerance of Dajiao. This membrane proteomics study provides new insights into cold stress tolerance mechanisms of banana, toward potential applications for ultimate genetic improvement of cold tolerance in banana.

Keywords: Musa spp. ‘Cavendish’; Musa spp. ‘Dajiao’; aquaporin; cold tolerance; peroxidase; quantitative proteomics.

FIGURE 1

Experimental design and schematic diagram of the workflow used in this study. A two-step method was used for extraction of membrane proteins in Cavendish and Dajiao seedlings in response to cold stress (10°C for 0, 3, and 6 h), and three sets of biological replicate samples were analyzed by an iTRAQ-based 2D-LC/MS/MS workflow for examining proteome changes. Phobius, Scampi-single, and TMHMM programs were used for filtering and predicting membrane proteins. Some important candidate proteins identified were evaluated and verified by RT-PCR, MRM, recombinant green fluorescent protein (GFP)-based subcellular localization, and enzyme activity analyses.

FIGURE 2

Phenotype, reactive oxygen species (ROS) accumulation, malondialdehyde (MDA) content, and relative conductivity in Cavendish and Dajiao under cold stress. Six-leaf-stage seedlings of Cavendish and Dajiao were treated at 10°C for 0, 3, and 6 h (B,E,H, respectively), and one representative leaf is shown, respectively (A,C,D,F,G,I). The level of superoxide radicals and H₂O₂ in the leaves of Cavendish and Dajiao was assessed by NBT (J,L) and DAB (K,M) staining, respectively. MDA content (N) and relative conductivity (O) in the leaves of Cavendish and Dajiao were measured under treatment at 10°C for 0, 3, 6, 24, and 48 h. Data are the mean ± SE (n = 5). Different letters indicate significant differences at p ≤ 0.05 between control and treatment.

FIGURE 3

Comparison of log₂ iTRAQ ratio (115/114 and 116/114) for 1,834 proteins identified in all three Dajiao biological replicates: set1, set2, and set3. Plots of 115/114 or 116/114 ratios for each of the quantified proteins between two of the three sets generated comparable quantification results, as determined by a linear regression analysis that revealed a slope of around 0.96, 0.97, or 0.99 in Dajiao for 115/114, among the pairs of set1/set2 (A), set1/set3 (B), and set2/set3 (C), respectively. Similar plots were also conducted for the 116/114 ratio, yielding a slope of around 0.96, 0.96, or 0.97 in Dajiao, among the pairs of set1/set2 (D), set1/set3 (E), and set2/set3 (F), respectively.

FIGURE 4

Comparison of GO molecular functions for Cavendish and Dajiao MPs (A) and DAMPs (B) under cold stress.

FIGURE 5

Subcellular localization analysis of Peroxidase P7. The Peroxidase P7-GFP fusion protein and GFP (used as a control), driven by the CaMV 35S promoter, were separately transformed into Arabidopsis thaliana protoplasts (A–H) and Cavendish protoplasts (I–P) and visualized by fluorescence microscopy. Images were taken of the representative cells expressing GFP or Peroxidase P7-GFP fusion protein under UV light (A,E,I,M), green light (B,F,J,N), or bright field (C,G,K,O). The merged images are shown in (D,H,L,P) respectively.

FIGURE 6

Time-course expression and MRM analysis of representative differentially abundant proteins under cold stress. Relative expression of five aquaporins (A–E) and three peroxidases (F–H), and their expression patterns in proteomics data (I). The horizontal arrow, upwardly tilted arrow, and downwardly tilted arrow represent no change, and increased and decreased expression, respectively. Relative quantitation of representative DAMPs between Cavendish and Dajiao by MRM analysis (J). The MRM data were obtained by summing the six or nine peak areas, generated from three biological replicates, for each protein containing two or three typical peptides, along with three transition ion pairs for each targeted peptide. Asterisks indicate significant differences between the transgenic lines and wild type under the same conditions (^∗p ≤ 0.05).

FIGURE 7

Treatment with NaN₃ compromised cold tolerance of Dajiao. Phenotype of the Cavendish and Dajiao plants pretreated with NaN₃ (B) or water (C), followed by cold stress at 10°C for 24 h (F,G); representative leaves of Dajiao are shown (A,D,E,H). H₂O₂ accumulation (I,J) in leaves of Dajiao analyzed at the end of cold treatment. (K,L) Show the comparison of POD and CAT activities in Cavendish and Dajiao plants treated with water or 5 mM NaN₃ for 12 h. Different lowercase letters above columns indicate a significant difference at p ≤ 0.05 between the columns by Duncan’s test using SPSS statistical software (version 16.0, SPSS Inc. Chicago, IL, United States).

FIGURE 8

Peroxidase activity assays for Cavendish and Dajiao. Peroxidase activity assays reveal the temporal changes in soluble peroxidase activity (A) and ionically bound CWP activity (B) for both Cavendish and Dajiao under 10°C treatment for the indicated time. Values of relative activity for each column indicate means ± SD of three biological replicates. Different lowercase letters above columns indicate a significant difference at p ≤ 0.05 between the columns by Duncan’s test using SPSS statistical software (version 16.0, SPSS Inc. Chicago, IL, United States).