The soil moisture content, superoxide anion contents and proline contents of HN44 and HN65 under different treatments were shown in Fig 1, 2 and 3 respectively.
Relationship between soil moisture content and superoxide anion content
According to the changes observed in the superoxide anion content under different soil moisture contents, with the help of Origin software, an equation was fitted by using the exponential (E index) curve fitting method (Fig 4). The fitting equation of HN44 is y=0.50+1.89×10
8e
-1.21x, with R
2=0.962; the fitting equation of HN65 is y=0.58+2.19×10
9e
-1.36x, with R
2=0.992. In these equations, y is the superoxide anion content and x is the soil moisture content. The absolute value of the change rate (denote as |r|) can reflect the variation trend of the superoxide anion content with soil moisture content. According to the above formula, the change rate of the superoxide anion content was obtained by taking the derivative of the fitting curves of the two varieties.
When |r| was greater than 0 and less than 0.1, the soil moisture content of HN44 was 31.5% - 19.66%, that of HN65 was 31.5%-19.41% and the superoxide anion content of soybean leaves had no significant change When |r| was greater than 0.1 and less than 1, the soil moisture content of HN44 was 19.66%-17.77%, that of HN65 was 19.40%-17.71% and the superoxide anion content in soybean leaves increased slowly. When |r| was greater than 1, the soil moisture contents of HN44 and HN65 were 17.77%-14.5% and 17.70%-14.5%, respectively and the superoxide anions contents of soybean leaves increased rapidly. The variation trends of the superoxide anion contents were similar between the two varieties. This is consistent with the research of
Gao (2007),
Mo (2007) and
Ren (2017),
etc. with the decrease of soil moisture content, the content of superoxide anion in leaves is increasing.
Relationship between soil moisture content and proline content
According to the changes observed in the proline contents under different soil moisture contents, an equation was fitted (Fig 5). The fitting equation of HN44 is y= 27.12 + 1.20×10
7e
-0.71x, with R
2= 0.996; the fitting equation of HN65 is y= 23.54+4.28×10
14e
-1.91x, with R
2= 0.984. In these equations, y is the content of proline and x is the soil moisture content.
According to the above formula, the change rate of the proline content was obtained by taking the derivative of the fitting curves of the two varieties. When the |r| of the proline content was greater than 0 and less than 0.1, the soil moisture content of HN44 was 31.5% - 25.73%, that of HN65 was 31.5%-19.18% and the proline content of soybean leaves did not increase significantly. When |r| was greater than 0.1 and less than 1, the soil moisture content of HN44 was 25.73%-22.48%, that of HN65 was 19.18%-17.99% and the proline content of the two soybean varieties increased slowly. When |r| was greater than 1, the soil moisture content of HN44 was 22.48%-14.5%, that of HN65 was 17.96%-14.5% and the proline content in soybean leaves increased rapidly. These results showed that drought-tolerant varieties could obviously respond to mild and moderate drought and synthesize proline earlier than drought-sensitive varieties to alleviate the effect of drought, while sensitive varieties could synthesize proline in large quantities under severe drought and their regulatory abilities were obviously weaker than those of drought-tolerant varieties.
The results showed that drought-tolerant varieties had a strong ability to accumulate osmotic adjustment substances; these substances can respond to drought earlier and reduce the damage caused by drought stress. In this experiment, the response times of the proline contents of the studied varieties with different drought tolerances to soil water content differed from the results of
Sun et al., (2005) and
Cai et al., (2008).
Correlation analysis of the superoxide anion and prolines in soybean leaves
A correlation analysis was conducted between the superoxide anion and proline contents in the two studied soybean varieties (Fig 6) and it can be seen that the superoxide anion and proline contents were significantly positively correlated.
The fitting equation of HN44 is y=99.36x-7.38 and HN65 is y= 65.34x -19.64. The fitting correlation is preferably. The abscissa is the superoxide anion, content, the ordinate is the proline content. The data showed that there was a positive correlation between proline content and superoxide anion content in both cultivars, but the slope of HN44 was much higher than HN65, indicating that with the increase of superoxide anion content, the increase of proline content in drought-tolerant cultivars was significantly higher than that in sensitive cultivars. The results showed that proline played an important role in regulating superoxide anion content and alleviated the damage of superoxide anion on soybean under drought conditions.