In rice, the stage of development most sensitive to high temperature stress is flowering, and exposure at this stage can result in spikelet sterility, thereby leading to significant yield losses. In this study, protein expression patterns of rice anthers from Dianxi4, a high temperature tolerant Japonica rice variety, were compared between samples exposed to high temperature and those grown in natural field conditions in Korea. Shotgun proteomics analysis of three replicate control and high-temperature-treated samples identified 3,266 nonredundant rice anther proteins (false discovery rate < 0.01). We found that high levels of ATP synthase, cupin domain-containing proteins, and pollen allergen proteins were present in rice anthers. Comparative analyses of 1,944 reproducibly expressed proteins identified 139 differentially expressed proteins, with 95 increased and 44 decreased in response to high temperature conditions. Heat shock, DnaK family, and chaperone proteins showed highly increased expression, suggesting that the high temperature tolerance of Dianxi4 is achieved by stabilization of proteins in pollen cells. Trehalose synthase was also highly increased after heat treatment, suggesting a possible role for trehalose in preventing protein denaturation through desiccation.
Rice is one of the most important staple cereal crops in the world, along with wheat and corn. It is cultivated and consumed mainly in tropical and temperate regions, especially Asia. Cereal crops, such as rice, wheat, and corn, are particularly susceptible to changes in environmental conditions. The International Panel on Climate Change (IPCC) predicts that average global temperatures will increase by 1–4°C by 2100, and this global warming will cause the yields of major crops to decrease by more than 25% by 2050 [
Previous studies have described rice yield reductions of 7-8% for each 1°C increase in temperature [
In rice, high temperature stress leading to significant yield loss due to spikelet sterility has been studied in various ways. Since resistance to high temperature is a quantitative trait, mapping quantitative trait locus (QTL) using SNP markers is fundamental to the study of this characteristic. QTLs that conferred protection from heat stress have been detected in the tolerant N 22 variety on chromosome 4 (qHTSF4.1) and chromosome 1 (qHTSF1.1), explaining 12.6% and 17.6% of the total phenotypic variation, respectively [
Multidimensional protein identification technology (MudPIT), a representative shotgun proteomic method, has been introduced as a complimentary technique to 2D-PAGE [
The tiny size of rice pollen grains makes it almost impossible to collect pure samples to monitor proteome responses. Thus, rice anthers are usually collected for protein extraction. Here, we describe quantitative shotgun proteomics analysis, which allowed us to screen a large number of proteins using only a small amount of material from rice anthers. We used this method to compare protein expression patterns in Dianxi4.
The method of Ye et al. (2012) was modified for heat treatment in flowering stage. Dianxi4, a Chinese rice variety, and Ilpum, a Korean commercial variety, were grown at Konkuk University in an experimental rice paddy field in 15 × 30 cm rows. One day before anthesis, each individual plant including soil was scooped from paddy field with 15 cm diameter and put into a pot. Rice plants were put in a growth chamber at 06:30 AM. Temperature was gradually increased from 27°C to 38°C until 08:00 AM. At 18:30 PM, the temperature was gradually decreased to 24°C until 20:00 PM. The relative humidity was 70% with a 12 h day/12 h night cycle. After 5 days of high temperature treatment, plants were moved to a greenhouse and cultivated until the grain fully matured. As a control, plants with tillers at the same stage of development (bearing panicles) were scooped the same as in the method of the high-temperature-treated rice. The rice in a pot was placed in the rice filed by the end of the day of high temperature treatment; then the control plants were moved to greenhouse with the high-temperature-treated rice plants. Three individual plants for each variety were used and the spikelet fertility was measured from 9 panicles (three panicles per an individual plant). For the proteomic analysis, plants grown in paddy field were put into pots as descried previously. Rice plants were put in a growth chamber at 06:30 AM. Temperature was gradually increased from 27°C to 38°C until 08:00; then the temperature was maintained through day and night. At the next day at 05:00 AM, the temperature was gradually decreased to 24°C until 06:30 AM. The relative humidity was 70% with a 12 h day/12 h night cycle. For the control plants, the rice in a pot was placed in the rice filed. Three individual plants for each variety were used and rice anther for proteome analysis was harvested from one panicle for each plant.
Anthers of Dianxi4 were collected immediately after one day of high temperature treatment. To ensure that all anthers collected were at the same developmental stage, they were taken from upper part of only one panicle; anthers emerging from the spikelet were not used in this study. Only anthers inside in the spikelet and located in the middle of panicles approximately 3 cm in width were collected. Control anthers were collected on the same day using the same method. Harvested anthers were ground in liquid nitrogen using metal balls and protein was extracted from the resultant powder using extraction buffer (100 mM Tris-HCl, pH 8.5; 8 M Urea; 5 mM DTT; 1% LDS). The suspension was incubated at room temperature for 30 min, followed by centrifugation at 14,000 g for 15 min. The supernatant was retained and filtered through 0.45
Protein samples (50
A nanoflow HPLC instrument (Easy nLC, Thermo Fisher Scientific, San Jose, CA, USA) coupled on-line to a Q Exactive mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) was used. Analytical columns (12 cm; 75
Each peptide from the LC-MS/MS spectra was searched against the TIGR Rice Pseudomolecule protein database Release V7.0 (
The output of the Proteome Discoverer search was exported to Microsoft Excel to calculate normalized spectral counts (NSpC) [
Rice protein molecular weights (MW) and
GO annotations of rice proteins were retrieved from the TIGR Rice Pseudomolecule protein database; Release V7.0. GO singular enrichment analyses were performed in agriGO (
Ilpum, a commercial rice variety, was included in our initial experiments to compare the tolerance of the Dianxi4. One of the most heat-sensitive stages of rice pollen development is flowering; thus, high temperature was applied to the plants at the flowering stage. After 5-day heat-treatment both rice varieties were cultivated in a greenhouse and spikelet fertility was measured after the grain reached full maturity (Figure
Comparison of spikelet fertility of Ilpum and Dianxi4 at (orange) normal condition and (brown) heat stress.
Proteins from anthers of Dianxi4 rice plants exposed to high temperature treatment and grown under normal conditions were identified by shotgun proteomic analysis; anthers were used rather than pollen due to the small size of rice pollen grains. Individual anthers are also too small to provide sufficient proteins for proteomic analysis; therefore, groups of anthers were collected from single panicles and the pooled proteins extracted. For biological replication, anthers were collected from different individual plants. Shotgun proteomics analysis of three replicates each of the control and high-temperature-treated samples identified 3,266 nonredundant proteins (Supplementary Table
Distribution
Relative protein quantities can be estimated from the normalized Spectral Count with a high degree of accuracy. In general, proteome studies of rice leaves have found RuBisCO protein and photosynthesis proteins to be present in large quantities. Interestingly, our analysis of total proteins identified from rice anthers revealed that pollen allergen proteins (LOC_OS06g451810, LOC_OS09g23899, LOC_OS09g23999, and LOC_OS04g25190), ATP synthase, and cupin domain-containing proteins were present in large quantities. Pollen allergen proteins are allergenic to humans but are present in high amounts in plants, suggesting that they have physiological roles required for pollen germination and growth [
For comparative analysis, the relative quantities of the identified proteins were calculated using the label-free spectral count (SC) method. Not all 3,266 identified proteins were reproducibly identified in all control or heat-treated samples; therefore, we included only proteins identified in all three replicates of either category, with at least two SCs per replicate. Using these criteria, SCs of 1,944 reproducible proteins were normalized (NSpC) and the logarithmically transformed NSpCs (the natural log (Ln) of NSpC) were compared using a
Enriched GO terms of the differentially expressed proteins in Dianxi4 anther at high temperature.
GO term |
|
Description | Number in input list | Number in rice genome |
|
FDR |
---|---|---|---|---|---|---|
GO:0003824 | F | Catalytic activity | 71 | 13508 |
|
|
GO:0005737 | C | Cytoplasm | 87 | 11866 |
|
|
GO:0044424 | C | Intracellular part | 94 | 14514 |
|
|
GO:0044444 | C | Cytoplasmic part | 79 | 10930 |
|
|
GO:0005622 | C | Intracellular | 94 | 15144 |
|
|
GO:0005739 | C | Mitochondrion | 22 | 1611 |
|
|
GO:0044464 | C | Cell part | 103 | 19532 |
|
|
GO:0043229 | C | Intracellular organelle | 74 | 12251 |
|
|
GO:0043226 | C | Organelle | 74 | 12251 |
|
|
GO:0005623 | C | Cell | 109 | 22048 |
|
|
GO:0043227 | C | Membrane-bounded organelle | 71 | 11878 |
|
|
GO:0043231 | C | Intracellular membrane-bounded organelle | 71 | 11878 |
|
|
GO:0009536 | C | Plastid | 35 | 4703 |
|
|
GO:0005829 | C | Cytosol | 27 | 3345 |
|
|
GO:0005618 | C | Cell wall | 14 | 1179 |
|
|
GO:0030312 | C | External encapsulating structure | 14 | 1189 |
|
|
GO:0005773 | C | Vacuole | 17 | 1723 |
|
|
GO:0043232 | C | Intracellular non-membrane-bounded organelle | 13 | 1237 |
|
|
GO:0043228 | C | Non-membrane-bounded organelle | 13 | 1237 |
|
|
GO:0005730 | C | Nucleolus | 7 | 495 |
|
|
Mapman analysis for expression pattern of the differentially expressed proteins responding to heat stress.
Among the 96 proteins showing increased expression in heat-treated anthers relative to control anthers, nine were HSPs, and all were expressed at high levels (Table
Upregulated rice anther protein against heat stress.
Accession | Description | Expression ratio |
---|---|---|
LOC_Os02g52150.2 protein | Heat shock 22 kDa protein, mitochondrial precursor, putative, expressed | 57.21 |
LOC_Os05g44340.1 protein | Heat shock protein 101, putative, expressed | 471.87 |
LOC_Os03g18200.2 protein | Heat shock protein DnaJ, putative, expressed | 27.42 |
LOC_Os03g15960.1 protein | HSP20/alpha crystallin family protein, putative, expressed | 104.49 |
LOC_Os03g16030.1 protein | HSP20/alpha crystallin family protein, putative, expressed | 96.61 |
LOC_Os03g16040.1 protein | HSP20/alpha crystallin family protein, putative, expressed | 96.13 |
LOC_Os01g08860.1 protein | HSP20/alpha crystallin family protein, putative, expressed | 54.86 |
LOC_Os04g36750.1 protein | HSP20/alpha crystallin family protein, putative, expressed | 53.11 |
LOC_Os03g14180.1 protein | HSP20/alpha crystallin family protein, putative, expressed | 51.85 |
LOC_Os02g08490.1 protein | Chaperone protein clpB 1, putative, expressed | 2.60 |
LOC_Os03g16920.1 protein | DnaK family protein, putative, expressed | 774.59 |
LOC_Os05g35400.1 protein | DnaK family protein, putative, expressed | 262.28 |
LOC_Os07g17120.1 protein | Late embryogenesis abundant protein, putative, expressed | 32.79 |
LOC_Os05g02530.1 protein | Glutathione S-transferase, N-terminal domain-containing protein, expressed | 1.47 |
LOC_Os07g28480.1 protein | Glutathione S-transferase, putative, expressed | 19.70 |
LOC_Os01g55830.1 protein | Glutathione S-transferase, putative, expressed | 1.43 |
LOC_Os05g25850.1 protein | Superoxide dismutase, mitochondrial precursor, putative, expressed | 1.30 |
LOC_Os01g53000.1 protein | Trehalose synthase, putative, expressed | 78.82 |
Of the proteins showing decreased expression after heat treatment, the majority showed minimal decreases in expression (Supplementary Table
In summary, through quantitative shotgun proteomic analysis of rice anthers, we identified 3,266 nonredundant rice proteins, one of the largest rice anther protein identification experiments to date. ATP synthase, cupin domain-containing proteins, and pollen allergens were present in the rice anther in large amounts, suggesting a potentially important physiological role for rice pollen allergens. A comparison of protein expression patterns in rice anthers cultivated under normal conditions or exposed to high temperature treatment at the flowering stage suggested that the tolerance of Dianxi4 may be due to highly increased expression of heat shock, DnaK family, chaperone, and trehalose synthase proteins, which maintain protein homeostasis in pollen cells in response to high temperatures.
The authors declare that there is no conflict of interests regarding the publication of this paper.
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2012R1A1A1004138) and “Cooperative Research Program for Agriculture Science & Technology Development (Mapping/Detection of genes conferring resistance to rice brown planthopper and cold stress and development of breeding materials PJ01104202)” Rural Development Administration Republic of Korea.