1 Variation of CFPs
The variations of 53 CFPs are determined for the leaves of 189 upland cotton cultivars (strains) at 12:00 and 17:00, respectively. 5 CFPs are not significantly different but 48 CFPs are significantly different between 12:00 and 17:00 (Fig.1; Table 1). This may be due to environmental difference such as temperature, strong light intensity and other stress at 12:00 and 17:00. DS reflects the degree of differences between the CFPs at 12:00 and 17:00. 23 DSs are negative and 25 DSs are positive, indicating significant differences in the DSs of different fluorescence CFPs (Table 1). In order to accurately identify the difference of CFPs of upland cotton cultivars (strains), it is necessary to detect the difference at 12:00 and 17:00.
2 Distribution of CFPs of upland cotton cultivars (strains)
Of 48 CFPs observed in 189 upland cotton cultivars (strains), 23 CFPs at 12:00 tend to lower values distribution than those at 17:00 in the distribution map (Fig. 2). The reduction of these CFPs at 12:00 reflects reduction of photosynthetic function of the leaves. Among the 23 CFPs, 8 CFPs such as Fv/Fm, Fv/Fo, φPo, PI(abs), PI(CSo), PI(CSm), SFI(abs) and D.F. tend to lower value distribution indicating that the photochemical activity of leaves decreased at 12:00 (Fig. 2A-H). 10RC/ABS tends to lower value distribution at 12:00 (Fig. 2I), indicating that coefficient of light energy reaction center decreases. ETo/CSo, ETo/CSm, Kp and SumK tend to lower value distribution (Fig. 2J-M), indicating that the photochemical rate decreased at 12:00. φPo/(1-φPo) and ψo/(1-ψo) tend to lower value distribution(Fig. 2N, O), indicating an increase in the ratio of thermal dissipation to the absorption,capture and electron transfer of light energy at 12: 00; Sm and Sm/T tend to lower value distribution (Fig. 2P, Q), indicating that the plastoquinone library decreases and the reduction rate of plastoquinone library decreases. Both ψo and φEo have a lower value distribution tendency (Fig. 2R, S), indicating that the efficiency of electron transfer was reduced. Fp, Vi, TRo/CSm and RC/CSm tend to lower value distribution (Fig. 2T-W).
Compared with 17:00, a total of 24 CFPs tend to higher value distribution at 12:00 (Fig. 2). The increasing of these CFPs shows reduction of photosynthetic function of the leaves. Fo tends to higher value distribution (Fig. 2X), indicating that energy dissipation increases at 12:00. ABS/RC, ABS/CSo, TRo/CSo, TRo/RC tend to higher value distribution (Fig. 2Y-AC), indicating an increase in the energy absorbed and captured by the unit area reaction center. DIo/RC, DIo/CSo, DIo/CSm tend to higher value distribution at 12:00 (Fig. 2AD-AF), indicating an increase in energy dissipation per unit area and unit reaction center. Fo/Fm and φDo tend to higher
value distribution (Fig. 2AG-AH),indicating an increase in the ratio of thermal dissipation to the absorption, capture and electron transfer of light energy. dVG/dTo and dV/dTo tend to higher value distribution (Fig. 2AI,AJ), indicating that the rate of optical reaction center closure increases. REo/CSo and REo/RC tend to higher value distribution (Fig. 2AK, AL), indicating an increase in the electron transfer per unit area and unit reaction center for PSI. φRo, ψRo tends to higher value distribution (Fig. 2AM, AN), indicating that the quantum production used for PSI electron transfer increases. The 8 CFPs such as F100μs, F300μs, F2ms, F30ms, Vj, δRo, δRo/(1- δRo) and PI(abs, total) tend to higher value distribution (Fig. 1AO-AV). RC/CSo tends to be distributed at both ends (Fig. 2AW), indicating that some cultivars (strains) have an upward trend and some cultivars (strains) have a downward trend.
Of the 48 CFPs, there are same trends such as Fo, F100μs, F300μs, F2ms, F30ms, and ABS/CSo, Kp and SumK, Fo/Fm and φDo at 12:00 and 17:00 ( Fig. 2).
3 Screening of CFPs
3.1 Screening of CFPs at 12:00
CFPs with high variation among different cultivars (strains) are suitable for identification of cotton cultivars (strains) with different photosynthetic properties. 38 CFPs were measured at 12:00 such as PI(abs), DIo/CSo, PI(CSm), δRo/(1 − δRo), PI(abs,total), DIo/RC, Sm/T, Kp, PI(CSo), Fo, REo/CSo, SumK, F100μs, ABS/CSo, DIo/CSm, F300μs, dVG/dTo, N, SFI(abs), Fv/Fo, φPo/(1-φPo), F2ms, Eto/CSo, Tro/CSo, ψo/(1-ψo), Sm, φDo, Fo/Fm, dV/dTo, REo/RC, D.F. F30ms, ABS/RC, 10RC/ABS, ETo/CSm, Vj, RC/CSo and δRo have a large degree of variation, and the coefficient of variation exceeds 0.1. These CFPs are suitable to screen upland cotton cultivars (strains). 15 CFPs such as Eto/RC, φEo, ρRo , φRo, Fp, Kn, Fm, ABS/CSm, RC/CSm, Tro/RC, ψo, TRo/CSm, Vi, Fv/Fm, Fm and φPo have low variability and coefficient of variation are less than 0.1. These CFPs are not suitable for screening cotton cultivars (strains) (Table 1; Table S3).
Of the 38 CFPs, the correlation coefficient of some CFPs is greater than 0.8, which are determined to be highly correlated. Only one of them was selected in highly related CFPs. Moreover, PI(abs, total)=PI(abs)·δRo/(1−δRo) [32]. So PI(abs, total) and PI(abs) are saved but δRo/(1−δRo) is abandoned. It is finally determined that 9 CFPs such as PI(abs, total), PI(abs), DIo/CSo, Sm/T, N, REo/CSo, Eto/CSo, Vj, RC/CSo are suitable for screening and evaluating the differences of CFPs in upland cotton cultivars (strains) (Table 2).
3.2 Screening of CFPs at 17:00
CFPs with high variation are suitable for identification of cotton cultivars (strains) with different photosynthetic properties. There are large variation coefficients in 41 CFPs measured at 17:00 such as δRo/(1−δRo), DIo/CSo, DIo/RC, dVG/dTo, PI(CSm), PI(abs), PI(abs,total), dV/dTo, PI(CSo), Vj, φDo, Fo/Fm, Sm/T, ψo/(1-ψo), F300μs, Sm/T, D.F. , ET/CSm, ABS/RC, Fo, Fv/Fo, φPo /(1-φPo), F100μs, ABS/CSo, DIo/CSm, δRo, φEo, F2ms, Kp, φRo, ψo, RC/CSm, F30ms, ρRo, SumK, Eto/RC, Eto/CSo, REo/CSo, Sm, N, REo/RC. The variation coefficient of these CFPs is more than 0.1 (Table 1, S3). So these CFPs are suitable to select upland cotton cultivars (strains) with different photosynthetic potential. The variation coefficient in the 12 CFPs such as 10RC/ABS, TRo/CSm, TRo/CSo, TRo/RC, Kn, Fv/Fm, φPo, Vi, Fm, ABS/CSm, RC/CSo and Fp is less than 0.1 (Table 1, S3). These CFPs are not suitable for selection of upland cultivars (strains).
Of the 41 CFPs at 17:00, the correlation coefficient of some CFPs is greater than 0.8, which is determined to be highly related (Table 3). Because of high correlation, it is not necessary to list them one by one but select one from high-related CFPs. Finally, it is determined that 8 CFPs such as PI(abs, total), DIo/CSo, Vj, PI(abs), Sm/T, Eto/CSo, REo/CSo and N are suitable for screening upland cotton cultivars (strains).
3.3 Comparison of CFPs obtained by screening at 12:00 and 17:00
The common CFPs obtained at 12:00 and 17:00 include PI(abs,total), PI(abs), DIo/CSo, Eto/CSo, Vj, Sm/T, REo/CSo, N. Compared with 17:00, RC/CSo is unique at 12:00. This result shows that there is no significant difference in the number of response centers per unit area of each strain under the non-stress environment at 17:00. But the values change greatly under the stress environment at 12:00. The results showed that CFPs were greatly affected by environment at 17:00 and 12:00.
4 Upland cotton cultivars (strains) screening
4.1 Upland cotton cultivars (strains) classification and screening
The selected 9 CFPs PI(abs,total), DIo/CSo, Sm/T, N, PI(abs), Eto/CSo, Vj, REo/CSo and RC/CSo are used to classify and review photosynthetic capacity of 189 upland cotton cultivars (strains).
4.1.1 Screening of cultivars (strains) based on PI(abs, total)
PI(abs, total) is the most general parameter in the JIP-test. It allows extending the
research of photosynthetic electron transport activity beyond PS II, involving changes in intersystem electron transport and PS I related processes [33]. According to PI(abs, total) at 17:00, 189 cotton cultivars (strains) are divided into four categories: A, B, C, and D. According to the comparison of the values of PI(abs, total) at 12: 00 and 17: 00 and DSs, the four categories of upland cotton cultivars (strains) A, B, C and D are divided into 12 subcategories(1-12) (Fig. 3A; Table S4).
4.1.2 Screening of cultivars (strains) based on PI(abs)
PI(abs) is a sensitive indicator applied for physiological and environmental screenings, including three functional steps of the photosynthetic activity (light absorption, excitation energy trapping, and conversion of excitation energy to electron transport) [34]. According to the value of PI(abs) at 17:00, 189 cotton cultivars (strains) are divided into four categories: A, B, C and D. Further by the comparison of the values of PI(abs) at 12:00 and 17:00 and DSs, the four categories of upland cotton cultivars (strains) A, B, C and D are divided into 10 subcategories (1-10) (Fig. 3B; Table S5).
4.1.3 Screening of cultivars (strains) based on Eto/CSo
Eto/CSo represents electron transfer per unit area, reflecting the reduction of QA re-oxidation at the reaction center [35-37]. In this experiment, Eto/CSo, Eto/RC, and Eto/CSm have a relatively high correlation. Eto/CSo can be used as an important parameter for reflecting photosynthetic electron transfer. According to the value of Eto/CSo at 17:00, 189 upland cotton cultivars (strains) are divided into three categories: A, B, and C. Further, according to the numerical comparison of Eto/CSo and DSs , the three major cotton cultivars (strains) A, B, and C are divided into 9 subcategories (1-9) (Fig. 3C; Table S6).
4.1.4 Screening of cultivars (strains) based on REo/CSo
REo/CSo and REo/RC have a relatively high correlation in this experiment. REo/CSo can be used as an important parameter for reflecting electron transfer at the end of PSⅠ[38]. According to the value of REo/CSo at 17:00, 189 upland cotton cultivars (strains) are divided into three categories: A, B, and C. According to the numerical comparison and DSs of Eto/CSo at 12:00 and 17:00, the three major cotton cultivars (strains) A, B, and C are divided into 6 subcategories (1-6) (Fig. 3D; Table S7).
4.1.5 Screening of cultivars (strains) based on DIo/CSo
DIo/CSo represents thermal dissipation per unit area (at T = 0) [39]. DIo/CSo, DIo/CSm and DIo/CSm have a high correlation. So DIo/CSo can be used as representative CFPs of energy dissipation. According to fluorescence parameter value under non-stress condition at 17:00, 189 cotton cultivars (strains) are divided into three categories: A, B, and C. Further, according to the numerical comparison of DIo/CSo at 12:00 and 17:00 and DSs, the three major upland cotton cultivars (strains) A, B, and C are divided into 11 subcategories (1-11) (Fig. 3E; Table S8).
4.1.6 Screening of cultivars (strains) based on Sm/T
Sm/T expresses the average fraction of opening PSⅡ RCs in the time span from Fo to Fm [40,41]. According to the value of Sm/T under non-stress at 17:00, 189 cotton cultivars (strains) are divided into three categories: A, B, and C. Further, according to Sm/T at 12:00 and 17:00 and DSs, the three major cotton cultivars (strains) A, B, and C are divided into 7 subcategories (1-7) (Fig. 3F; Table S9).
4.1.7 Screening of cultivars (strains) based on N
N represents the number of times of QA restored from the time starting illuminating to maximal fluorescence Fm. It is related to the size of the receptor library [42]. According to the value of N non-stress at 17:00, 189 cotton cultivars (strains) are divided into three categories: A, B, and C. Further, according to the value of N at 12:00 and 17:00 and DSs, the three major cotton cultivars (strains) A, B, and C are divided into 11 subcategories (1-11) (Fig. 3G; Table S10).
4.1.8 Screening of cultivars (strains) based on Vj
Vj is the fraction of closed reaction centres (QA-/QA value) at 2 ms of illumination. Vj is established by the reduction rate of QA to QA- and the oxidation rate of QA- to QA [43]. According to the value of Vj non-stress at 17:00, 189 cotton cultivars (strains) are divided into three categories: A, B, and C. Further, according to the value of Vj at 12:00 and 17:00 and DSs, the three major cotton cultivars (strains) A, B, and C are divided into 7 subcategories (1-7) (Fig. 3H; Table S11).
4.1.9 Screening of cultivars (strains) based on RC/CSo
RC/CSo represents the number of unit area reaction centers (at T = 0) [44]. Since the variation coefficient of RC/CSo at 17:00 is small, RC/CSo does not suit to screen cotton varieties, cotton cultivars (strains) are classified using RC/CSo value at 12:00. According to RC/CSo value under 12:00, 189 cotton cultivars (strains) are divided into three categories: A, B, and C. Further, according to RC/CSo numerical comparison and DSs at 12:00 and 17:00, the three major cotton cultivars (strains) A, B, and C are divided into 5 subcategories (1-5) (Fig. 3I; Table S12).
4.2. Review of Typical Cotton cultivars (strains)
According to PI(abs) at 12:00 and 17:00, Ekangmian9, Emian18 and SuQ1 show strong photosynthetic properties. The values of PI(abs) are both high at 12:00 and 17:00, indicating that they have high PS II photosynthetic activity. But PI(abs) values are slightly lower at 12:00 than 17:00, indicating that they are resistant to stress environment at 12:00. The values of RC/CSo, Eto/CSo and N are at a moderate level in 189 cultivars and are slightly lower at 12:00 than 17:00, indicating the number of active reaction centers, the quantum yield of PS II electron transfer and the size of the PSII receptor library are at a moderate level and may be less subject to environmental change. The values of DIo/CSo are low and is slightly lower at 12:00 than 17:00, indicating that the three cultivars (strains) have low energy dissipation. The values of Sm/T are at a moderate level at 17:00 and increased significantly at 12:00, indicating that the ratio of active reaction center to inactivated reaction center is at a moderate level and may increase significantly under stress at 12:00. The value of PI(abs,total) is high, is significantly higher at 12:00 than 17:00, indicating the three cultivars (strains) have PS I photosynthetic performance, which may increase in stress. The values of Vj is low but is higher at 12:00 than 17:00, indicating the ratio of QA-/QA in the electronic delivery chain at 2ms is low, and may increase under adverse conditions. The REo/CSo value is at a moderate level, with little change from 12:00 to 17:00, indicating the PS I electron transfer efficiency is at a moderate level and may be less affected by the environment. These results show that reduced QA can be oxidized in time and the energy dissipation is low, which are the main reason for the strong photosynthetic performance of Ekangmian9, Emian18 and SuQ1.
According to PI(abs) at 12:00 and 17:00, the photosynthetic performance of Jifeng 908 is at a moderate level in 189 cultivars. PI(abs) are at middle level and are slightly lower at 12:00 than 17:00, indicating PS II photosynthetic performance is at a medium level and reduced under stress. DIo/CSo, PI(abs, total), N, RC/CSo, Eto/CSo, REo/CSo and Vj are in middle level and are higher at 12:00 than 17:00, which show that the dissipated energy, the photosynthetic performance of PS I, the size of receptor bank of PS II, the number of active reaction centers, the electron transfer quantum yield of PS I and PS II, and the ratio of QA- to QA in the electron transfer chain at 2ms are in moderate and elevated under stress conditions. The high Sm/T values decrease at 12:00 compared to 17:00 indicating that the ratio of the active reaction center to the inactivation reaction center is high and may decrease under stress. These results show that Sm/T may not be the main parameter affecting photosynthetic performance. The electron transfer, the number of active reaction centers, the size of PS II receptor bank and dissipation energy are all at a moderate level. Therefore, Jifeng 908 has moderate photosynthetic properties of PS I and PS II.
According to PI(abs) at 12:00 and 17:00, the photosynthetic performance of Zhongzi 2858 is low. PI(abs) values are low and lower at 12:00 than 17:00, indicating PSII photosynthetic activity is low and reduced due to stress. The values of Eto/CSo, PI(abs, total) and Sm/T are low and is higher at 12:00 than 17:00, indicating that the quantum yield of PS II electron transfer, PS I photosynthetic performance and the ratio of active reaction centres to inactivation reaction centres are low and rise under stress. The high DIo/CSo and RC/CSo values, which are higher at 12:00 than 17:00, indicating high energy dissipation, reaction centres and increase under stress. The N and REo/CSo are at a medium level and are higher at 12:00 than 17:00, indicating that the size of PS II receptor bank and the quantum output of PS I electron transfer are at a medium level and rise under stress. The high Vj values, which are lower at 12:00 than 17:00, show that the ratio of QA- to QA in electronic delivery chain at 2ms is high and decrease under stress. These results show the low quantum yield of PS II electron transfer, the low proportion of active reaction centres and the high energy dissipation are the main reasons for the low photosynthetic performance.