Genetic tolerance to low temperatures in irrigated rice1

The aim of this study was to characterise genetic variability to low-temperature tolerance in the emergence stage of irrigated-rice genotypes under controlled and field conditions. Thirty-seven genotypes were evaluated, interspersed with controls of different levels of low-temperature tolerance. The design used was of randomised blocks with three replications. The genotypes were submitted to germination and emergence in controlled-temperature trials at 13 oC (Trial I) and 17 oC (Trial II), and in field environments with no cover (Trial III) and with a cover of polyethylene (Trial IV), where the soil temperature was monitored. The evaluations were carried out daily by counting the emerged seedlings and later inferring the speed of emergence index (SEI). The results showed (experiment I to IV) that the SEI varied from 0.61 to 4.23, increasing by 0.4259 for each 1 oC increase in soil temperature. The genotypes Ostiglia, Diamante, Baldo, Carnaroli, Selenio, Loto e Amarelo, of subspecies japonica, and Ringo Miara-AC and Arelate, of subspecies indica, are promising sources of genetic tolerance to low temperature. Temperatures of less than 17 oC reduce the number of emerged seedlings and delay development in each of the rice genotypes under test, with a greater or lesser negative effect on the initial plant stand per unit area.


INTRODUCTION
Rice (Oryza sativa L.) is the staple food of half the world's population (HAO; LIN, 2010). Due to the constant growth in population, there is a growing demand for the cereal, demonstrating the need for increases in production. In recent decades, there have been significant increases in the production potential of rice, which were achieved by advances in genetics (BRESEGHELLO et al., 2011;STRECK et al., 2018), as well as improvements in crop management. Among the successful changes in crop management, anticipating the sowing season has been suggested, so that the reproductive stage coincides with months with the smallest probability of low temperatures and the highest intensity of solar radiation (MERTZ et al., 2009;STEINMETZ et al., 2009;DEIBLER;SILVA, 2013).
Low temperatures are a common problem in rice cultivation, where sensitivity affects production of the cereal (ZHOU et al., 2012). Temperatures of less than 20 ºC can be harmful to the rice crop, and are common in temperate and subtropical areas (NANDA; SESHU, 1979), reducing the percentage and speed of seedling emergence (XU et al., 2008). The occurrence of such low temperatures during the early stages can impair the establishment and stand of the crop , affecting plant growth and development, which may lead to uneven grain maturation (BOSETTI et al., 2012). In addition, if the low temperatures occur later, they can cause a loss in grain yield due to spikelet sterility.
Low temperature is a factor which is unpredictable and abiotic in nature, as such, its negative effects on rice are difficult to control at the management level, which makes the genetic tolerance of cultivars the most viable alternative in the search for a solution to minimise damage and stabilise grain yield in areas subject to cold. Research aimed at obtaining genotypes with a greater tolerance to cold using the existing variability of available collections, is therefore a strategy of fundamental importance.
To develop rice cultivars with adequate tolerance to low temperature during the germination and seedling emergence stage, it is necessary to investigate the genetic resources available from various regions of the world that might serve as a genetic source for the subsequent transfer of genes involved in the capacity for low-temperature germination to elite strains or commercial cultivars.
The characterisation and subsequent selection of genotypes that are tolerant to low temperatures during the initial period of development is difficult due to the complex genetic basis of the trait and the lack of control of low-temperature stress under field conditions (CRUZ; MILACH, 2000). On account of the difficulties of field evaluations for cold tolerance, it is necessary to use strategies for conducting experiments under controlled conditions in order to characterise this trait.
In this context, the aim of this study was to characterise genetic variability to low-temperature tolerance in the emergence stage of irrigated rice genotypes under controlled and field conditions.

MATERIAL AND METHODS
The experiment was conducted at the Terras Baixas Station of Embrapa Clima Temperado, in the district of Capão do Leão, Rio Grande do Sul, Brazil (31°48'16" S, 52°24'46" W, at an altitude of 15 m). The Köppen-Geiger climate classification defines the climate as predominantly of type Cfa, simplified as humid subtropical.
The seeds were tested for germination, employing three replications of 100 seeds for each cultivar, and left to germinate at a temperature of 25 ºC in rolls of paper towel (Germitest ® ) moistened with an amount of water equal to 2.5 times the weight of the substrate. Counts were taken of normal seedlings 7 and 14 days after sowing at the Seed Analysis Laboratory of Embrapa Clima Temperado, Pelotas, Rio Grande do Sul.
Four trials were carried out simultaneously: I) Trial in a controlled environment at a constant temperature of 13 ºC; II) Trial in a controlled environment at a constant temperature of 17 ºC; III) Field trial during a period with a high probability of low air and soil temperatures; and IV) Field trial in a protected environment employing a transparent polyethylene tarpaulin in a low tunnel. Each trial was set up in a solodic Haplic Eutrophic Planasol, with the aim of removing any effect from this factor in the trials.
Trials I and II were carried out in a temperaturecontrolled environment, so as to maintain the treatment factors constant at 13 ºC and 17 ºC. The seeds of each genotype were sown in polypropylene trays filled with soil, where one row of 50 seeds, 0.5 metre in length, represented one experimental unit. Soil moisture was constantly controlled by thermocouple sensors. Genetic tolerance to low temperatures in irrigated rice Jasmine 85 Thailand Indica ----Trials III and IV were set up to simulate field conditions; however, Trial IV was covered with a transparent polyethylene tarpaulin in a low tunnel, thereby giving greater control over the environmental conditions. The trials were sown in August 2016 (period with a high probability of low air and soil temperatures), in experimental units comprising one row, 0.5 metres in length, containing 50 seeds of each genotype. The soil temperature under field conditions was monitored by means of thermocouple sensors (thermometers) located at a depth of three centimetres, which read the soil temperature and moisture every 60 minutes, collecting 24 temperatures daily. The data were recorded by data logger and transferred to a computer.
Each trial was monitored and evaluated daily, always at the same time (08:00), counting all the seedlings that emerged during that 24-hour period. The seedlings were counted when the coleoptile, developed from the embryo, broke the soil surface, initiating the process of emergence. Each emerged seedling was marked with a round wooden toothpick to avoid repeated counting. After evaluating each repetition, all the identified and marked seedlings were added to the daily-count spreadsheet. The counts ended once the emergence of each genotype had stabilised. From the daily count of emerged seedlings, the value of the desired response variable, the Speed of Emergence Index, was later inferred for each replication of the genotype under evaluation.
The analysis of cold tolerance was determined using the speed of emergence index (SEI) proposed by Maguire (1962), calculated with the formula SEI = (E1/ N1) + (E2/N2) + (E3/N3) +... + (En/Nn), where: E1, E2, E3, ... En represent the number of seedlings included in the first, second, third and last count; and N1, N2, N3, ... Nn are the number of days between sowing and the first, second, third and last count respectively. The speed of emergence index was corrected using the germination power (PG%) of each genotype to configure the weighted speed of emergence. The speed of emergence index data were submitted to descriptive analysis, analysis of variance, and the Scott-Knott mean-value grouping test at 5% probability. The relationship of temperature (quantitative treatment factor) to the independent variable, speed of emergence index, was inferred by linear regression analysis. All analyses were carried out using the GENES genetics and statistics computer software (CRUZ, 2013).

RESULT AND DISCUSSION
In the experiments conducted in the field and under the low tunnel, a total of 480 soil temperatures were recorded up to the end of the experimental period; their intensity and fluctuation can be seen in Figure 1. It was found that between the seventh and eighth day after sowing, higher temperatures occurred due to variations in climate. In the afternoon the temperature reached 35.9 ºC in the environment covered with a polyethylene tarpaulin, and 32.6 ºC in the outdoor environment. The minimum temperature seen for both environments was 8.5 ºC. The mean temperature for the covered environment was 21.5 ºC, while for the outdoor environment it was 18 ºC. This demonstrates the great difficulty in inferring the tolerance of genotypes, especially under field conditions. Mittler (2006) states that crop improvement programs have achieved limited success in improving cold tolerance, mainly due to the polygenic control of the trait, unpredictable weather, and the interaction between the response to cold and to environmental factors (soil, nutrients and moisture, among others).
It was found from Figure 2 that first emergence occurred in Trial IV (in the field in a low tunnel) from the seventh day after sowing, and continued until the fifteenth day. In the outdoor environment (Trial III), first emergence occurred only on the ninth day after sowing, continuing until 16 days after sowing.
For tests I and II under constant conditions of extreme low temperature for the crop, first emergence  occurred later, from the twelfth and ninth day respectively, continuing for a much longer period, until day 35 (Trial I) and day 28 (Trial II). The daily frequency of emergence of each genotype was low, approximately 0.55 and 1.38 seedlings day -1 ; in addition, at 13 ºC, the number of emerged seedlings was far lower than in the other experiments. According to Mertz et al. (2009), in an isoenzymatic analysis, a temperature of 13 °C causes a decrease in the activity of the enzyme esterase and an increase in the enzyme alcohol dehydrogenase. Furthermore, according to the authors, the start of esterase activity occurred only on the seventh day of the germination test, whereas, at the ideal temperature, it was already possible to see greater band intensity by the third day of germination.
For rice to achieve rapid and uniform emergence, the soil temperature should be equal to or higher than 20 ºC (STEINMETZ et al., 2008). Therefore, at lower soil temperatures, there is a high risk of delay in emergence, with fewer emerged seedlings as a result. This delay and reduction in emerged seedlings directly affect the plant stand in the field, which in turn, implies a reduction in crop productivity. This happens due to the reduced number of fertile panicles, a result of the number of plants per unit area, this variable being one of the components that most affects grain production (RANI et al., 2015).
The results also showed that as the temperature increased, the number of days to reach 50% of emerged seedlings decreased. In Trial I, half of the emerged seedlings was reached 11 days after emergence of the first seedling, this being the trial that took the most days to reach 50%. In trials II and III, this occurred five days after emergence of the first seedling, explained by the fact that for both trials the mean temperature was close, 17 o C and Genetic tolerance to low temperatures in irrigated rice 18 o C respectively. Trial IV took the least time for 50% of the seedlings to emerge, requiring only three days.
The end of the counts and of the experiment was reached when there was no further seedling emergence in each of the genotypes in each trial.
An analysis of the statistical parameters in relation to evaluating tolerance in irrigated-rice genotypes at low temperatures in the emergence stage, determined from the speed of emergence index (SEI), showed differences in performance between the genotypes under study. The variation coefficient (CV) ranged from 5.07% to 31.50%.
The grouping based on the Scott-Knott test at 5% probability (Table 2) separated the genotypes into groups in the four trials, making it possible to characterise these genotypes in terms of tolerance to low-temperature abiotic stress on germination. In Trial I, with the lowest temperature, the analysis showed good accuracy for the control cultivars, since the Diamante and Norin Mochi genotypes presented as tolerant to cold, the Tomoe Mochi genotype as medium-tolerant to cold and the BRS Pampa, BRS Querência and BRS Firmeza genotypes as susceptible to cold.
In the same trial, Ostiglia, Baldo, Carnaroli, Selenio, Loto and Amarelo belonging to subspecies japonica stood out. However, the Ringo Miara-AC and Arelate genotypes of subspecies indica, proved to be similar to the genotypes of subspecies japonica, also being promising sources of cold tolerance. The results corroborate those of Mertz et al. (2009), who concluded that the Lemont and Oro cultivars of subspecies japonica, showed greater tolerance compared to the BRS Agrisul and BRS Chuí cultivars (subspecies indica), when submitted to a temperature of 13 °C during the germination-emergence stage. Among the 10 genotypes that showed the highest speed of emergence index,  The trials conducted in the field, with and without a cover of polyethylene, showed great variation between the mean values for SEI, where genotypes of subspecies indica had a similar SEI with a greater number of emerged seedlings, compared to the trials at a lower temperature.
In Trial I, the minimum and maximum SEI found between genotypes were 0 and 1.39 respectively. For Trial II, the SEI varied from 0.46 to 2.99, a range of 2.53. In trials III and IV in the field, the minimum found was 0.35 and 0.93, and the maximum was 5.54 and 6.39, with a range of 5.19 and 5.46 respectively. This shows that the response of the genotypes under evaluation was affected by the trials.
In general, as the temperature decreased there was a significant reduction in the speed of emergence index. Comparing Trial I with the others trials, the mean speed of emergence index varied between 0.61 and 4.23. From the regression equation in Figure 3, the mean speed of emergence index increases linearly by 0.4259 for each 1 o C increase in temperature. The coefficient of determination (R 2 ) of the first-order regression equation was 0.9878, showing that a large part of the variation in the speed of emergence index can be explained by the variation in soil temperature.
Genetic tolerance to low temperatures in irrigated rice CONCLUSIONS 1. The genotypes Ostiglia, Diamante, Baldo, Carnaroli, Selenio, Loto and Amarelo, belonging to subspecies japonica, and Ringo Miara-AC and Arelate, of subspecies indica, are promising sources of genetic tolerance to low temperature; 2. Temperatures of less than 17 ºC reduce the number of emerged seedlings and delay development in each of the rice genotypes under test, having a greater or lesser negative effect on the initial plant stand per unit area.