Phytotoxicity of atrazine combined with cadmium on photosynthetic apparatus of the emergent plant species Iris pseudacorus

The combined pollution, instead of single pollution, has become a widespread contamination phenomenon in aquatic environment. However, little information is now available about the joint effects of the combined pollution, especially co-existed pesticides and heavy metals, on aquatic plants. In the present study, using continuous excitation chlorophyll fluorescence parameters and the OJIP transient, comparisons of herbicide atrazine (ATZ) phytotoxicity on Iris pseudacorus between in the presence and absence of cadmium (Cd) were evaluated over an exposure period of three weeks under laboratory conditions. Results showed that both ATZ and Cd were toxic to I. pseudacorus. The ratio Fv/Fo, specific electron transport energy (ET0/RC), and photochemistry efficiency (PIabs and PItotal) of this emergent plant species at individual ATZ and Cd concentrations were significantly lower than those of the control. ATZ mainly inhibited electron transport beyond QA at PSII acceptor side as indicated by the sharp rise of the J-step level of fluorescence rise kinetics. A pronounced K-step and the loss of I-step due to the damage on the OEC and PSI also occurred when ATZ was at or above 1.0 mg·L−1. In comparison to ATZ alone, ATZ combined with Cd resulted in a lower amplitude rise in J-step with apparent J-I and I-P phases; and significantly lower Fo with higher Fv/Fo, as well as greater ET0/RC with higher values of PIabs and PItotal. However, the adverse influences of ATZ combined with Cd on the above indicators were still significant as compared with the control. Therefore, the coexistence of Cd alleviated the individual phytotoxicities of ATZ, whereas combined pollution of ATZ and Cd still induced the decline in photosynthetic performance of I. pseudacorus, and its potential ecological impacts on the aquatic vegetation cannot be ignored. Our findings offer a better understanding of the joint effects of the pesticide and heavy metal on non-target aquatic plants, and provided valuable insights into the interaction of these pollutants in aquatic environment.


Introduction
Atrazine (ATZ), a triazinic herbicide, is the second most widely consumed pesticide worldwide (Singh et al. 2018). Due to its extensive and long-term application, ATZ has become a widespread pollutant in water environments with an expected environmental concentration of 2.7 mg L −1 (Peterson et al. 1994;Albright et al. 2013). The lowest observed effect concentration of ATZ for the growth of emergent plant is 2 mg L −1 (Wang et al. 2014). Exposure to various herbicides in surface waters is one of the main factors responsible for the loss of native plant vegetation (Knauert et al. 2010). Additionally, ATZ is still a subject of continuous concern due to its potential endocrine and carcinogenic activity (Jablonowski et al. 2011). Meanwhile, Cd is a kind of typical heavy metal pollutant in aquatic systems (Xin et al. 2020). The amount of Cd occurring in water is generally less than 1 mg L −1 (Das and Jana 2004;Bhardwaj et al. 2020), whereas extraordinarily high concentrations of total Cd up to 5 mg L −1 in waters have been reported (Thornton and Walsh 2001). Such concentrations potentially pose a threat to human health because of its harmful effects on the food chain (Das and Jana 2004). Furthermore, considering that the co-presence of pesticides and heavy metals in aquatic systems has become a widespread contamination phenomenon, interactions between them may influence the fate of these pollutants; thus, the ecotoxicological effects of one contaminant may be enhanced (synergism) or eliminated (antagonism) due to the presence of others (Starling et al. 2019).
To date, only a small number of studies reported the joint biological toxicity of ATZ and Cd. These studies indicated that ATZ combined with Cd showed less than additive toxicity to the earthworm (Eisenia fetida) (Wang et al. 2012;Yu et al. 2019) and rice (Oryza sativa) (Su et al. 2005) owing to the formation of ATZ/Cd complexes, yet none of these investigations addressed the joint toxicity of these contaminant mixtures to aquatic plants which are of great importance in stabilizing function and structure of freshwater ecosystems. In addition, ATZ in coexistence with Cd displayed a prolonged half-life in water (Xie et al. 2021). The prolonged persistence of ATZ implies an increase in potential ecological risks. Therefore, there is a clear need to focus on the combined effects of ATZ and Cd on aquatic plants. It is well established that ATZ is an inhibitor of photosynthesis (Gronwald et al. 1989). Also, Cd inhibits electron transfer reactions in PSII (Sigfridsson et al. 2004). Moreover, photosynthesis is the basic physico-chemical process for plant growth and survival, and is usually considered a top priority in probing toxicological action of a compound toward plants (Gao et al. 2018). Thus, photosynthetic characteristics of plants can provide more direct and fundamental evidences for joint action induced by these two photosynthesis inhibitors. Iris pseudacorus, a completely emergent plant with high ornamental value, is commonly distributed in nature and artificial water bodies, and has been frequently used for phytoremediation of wastewaters. The purpose of this study was to evaluate the combined effects of ATZ and Cd on photosynthesis performance of I. pseudacorus through chlorophyll a fluorescence induction OJIP transients. To accomplish the objective, the present study investigated the changes of chlorophyll fluorescence parameters and differential kinetics in this plant species under ATZ stress in the presence and absence of Cd over an exposure period of three weeks.

Plant materials and chemicals
Seeds of Iris pseudacorus were sterilized with sodium hypochlorite solution for 10 min, followed by thorough washing with deionized water. The seeds were germinated in moist peat soil for five weeks; uniform seedlings were selected and transferred to flasks (one plant per flask) containing 250 ml 10% Hoagland nutrient solution. The seedlings were acclimatized for one week with the culture solution prior to the start of the experiments.
ATZ, used as a formulated product (90% water dispersible granules, BUGAO®), was purchased from Zhongshan Chemical Industry Co., Ltd., China. Cd was used as Cd(NO 3 ) 2 ·4H 2 O which was of analytical grade with 99% purity.
Each treatment had three replications, and consisted of one plant per replicate in a flask holding 250 mL treatment solution. The experiments were carried out in a controlled environment growth chamber with a 14-h light period (350 µmol m −2 s −1 ) and a temperature cycle of 25/20 °C day/ night. The relative humidity was 70%. The culture nutrient solution was supplemented every 24 h to a volume of 250 mL (initial solution volume) to compensate for the nutrition consumption and water vaporization, and was replaced once a week in order to maintain a relatively constant ATZ level across the entire experiment. The chlorophyll fluorescence parameters were measured after three weeks of continuous cultivation.

Chlorophyll fluorescence measurement
Chlorophyll fluorescence measurements were performed in each plant with a Handy-PEA fluorometer (Plant Effi-ciencyAnalyzer, Hansatech Instruments Ltd., King's Lynn, Norfolk, UK). The light intensity was 3000 μmol (photons)·m −2 ·s −1 , and the leaves were dark-adapted for 30 min before measurements were started. For each treatment, 9 measurements (three replicates, three leaves in each replicate) were made. The chlorophyll fluorescence emission induced between 10 μs and 2 s by the light pulses was digitized by the instrument. The selected JIP-test parameters used to characterize the photosynthetic performance were listed in Table 1 (Strasser et al. 2004;Chen et al. 2016;Guo et al. 2020;Sun et al. 2020). These parameters derived from the original fast fluorescence transient OJIP measurements were calculated according to Chen et al. (2016) and Guo et al. (2020).

Data analysis
Analysis of variance (ANOVA) was performed using SAS9.1. The significance of differences between treated groups and the control was analyzed using the unequal variance two-tailed Student's t test, and significant differences among ATZ concentrations were determined by Duncan's test, and P value < 0.05 was considered significant.

Chlorophyll fluorescence rise kinetics OJIP curves
For the treatments with ATZ only, the level of J-step of fluorescence rise kinetics increased for all ATZ concentrations (Fig. 1A); the I-and P-step levels also increased at ATZ concentrations of 0.1, 0.5, and 1.0 mg·L −1 , but declined at ATZ concentrations of 2.0 mg·L −1 . ATZ at concentrations of 1.0 and 2.0 mg·L −1 led to the disappearance of J-I and I-P phases. In addition, the uptrend in F o and downtrend in F v /F o with the increase of ATZ concentration were observed, and the change trend showed a significant dose effect relationship (Fig. 1B). Fm decreased significantly in treatment with 2.0 mg·L −1 ATZ.
For treatments with Cd alone, Cd at concentrations of 5.0 mg·L −1 caused slight increases in J-, I-, and P-step with obvious J-I and I-P phases (Fig. 1C); F o , F m , and F v /F o did not change significantly compared with the control (Fig. 1D); when Cd concentration increased to 10.0 mg·L −1 , however, the J-step level increased while the P-step level Fluorescence intensity at the J-step (2 ms) of OJIP F I = F 30ms Fluorescence intensity at the I-step (30 ms) of OJIP F m = F P Maximal recorded fluorescence intensity, at the peak P of OJIP Relative variable fluorescence at the J-step Relative variable fluorescence at the I-step Approximated initial slope (in ms −1 ) of the fluorescence transient normalized on the maximal variable fluorescence F v Specific energy fluxes (per Q A -reducing PSII reaction center-RC) ABS/RC Absorption flux per RC TR 0 /RC Captured energy flux per RC (at t = 0) DI 0 /RC Dissipated energy flux per RC (at t = 0) ET 0 /RC Electron transport flux per RC (at t = 0) Performance indexes PI abs Performance index (potential) for energy conservation from photons absorbed by PSII to the reduction of intersystem electron acceptors PI total Performance index (potential) for energy conservation from the photons absorbed by PSII to the reduction of photosystem I electron end-acceptors   1E); F m and F v /F o were significantly lower than those of the control (Fig. 1F). For treatments with ATZ plus Cd, OJIP curves showed visible J-, I-, and P-step with lower F o and higher F v /F o in comparison with ATZ alone (Fig. 1C, E); the levels of J-, I-, and P-step were higher and the magnitude of J-I phase was still lower than those of the control except for 0.1ATZ + 5Cd; when ATZ concentrations were 1.0 and 2.0 mg·L −1 , F o and F m increased significantly, and F v /F o did not show significant compared with those of the control.
The double normalized fluorescence curves (presented as relative variable fluorescence V t on a logarithmic time scale) show invisible features of OJIP curve (Guo et al. 2020). Compared with the control, J-step increased drastically under ATZ stress whether in presence or absence of Cd, and ATZ alone induced a greater increase than ATZ combined Cd did ( Fig. 2A, C, E). Especially in the case of ATZ alone at concentrations above 0.1 mg·L −1 , J-step was closed to P-step with a disappearance of the IP phase of the OJIP curve; at the same time, a remarkable lift in I-step and a flattened I-step was observed ( Fig. 2A). The initial slope of the relative variable fluorescence kinetics, denoted as M 0 , expresses the net rate of the RCs' closure (Strasser et al. 2004). The M 0 values of single and combined treatments of ATZ and Cd were significantly larger than that of the control ( Fig. 2A, C, E; Table 2). ATZ alone at concentrations above 0.1 mg·L −1 had stronger effects on the M 0 value than ATZ combined with Cd did, and the difference was significant when Cd concentration was 10 mg·L −1 . The ANOVA result showed that ATZ and Cd were significant factors for the M 0 value and had statistically significant interaction effects (Table 2). V J , V I , and F J /F I further reflected the initial action site of ATZ. V J and V I reflect the accumulation of Q A − ). Both single ATZ and ATZ combined Cd increased significantly the values of V J , V I , and F J /F I compared with the control. ATZ only at levels above 0.1 mg·L −1 showed significant differences in values of V I , V I , and F J /F I from ATZ combined with 10.0 mg·L −1 Cd, whereas this difference was not observed when Cd concentration was 5.0 mg·L −1 (Fig. 2B, D, F). Under single ATZ and combined stresses of ATZ and Cd, V J showed higher increase magnitude than V I .

The initial slope of the standardized fluorescence transient F t /F o
To ascertain the differences in the rate of the electron transfer from P 680 to Q A between ATZ-Cd treated and control plants, the initial slopes of the standardized fluorescence transient (the fluorescence values were expressed as F t /F o (Strasser et al. 2004)) at a linear time scale from 0.02 to 0.15 ms were presented in Fig. 3. The initial slope of single ATZ at the level of 0.1 and 0.5 mg·L −1 was significantly greater than that of the control, while it markedly decreased when ATZ concentration was up to 2.0 mg·L −1 (Fig. 3A, E). In contrast to the control, the plant treated with single Cd showed a significant increase in the initial slope, and there was significant difference among two Cd concentrations ( Fig. 3B, C, E). The initial slope of ATZ combined with Cd was significantly larger than that of the control, and was significantly larger than that of ATZ alone when ATZ concentrations were 1.0 and 2.0 mg·L −1 .

L-band
L-band is known as an indicator of the grouping of the PSII units or energetic connectivity between antenna and PSII RCs (Guo et al. 2020). For single ATZ treatment, the L-band increased with increasing ATZ concentrations. ATZ at the concentration of 1.0 and 2.0 mg·L −1 increased L-band most obviously ( Fig. 4A) with significantly higher W L and F L /F J (Fig. 4B); ATZ at other concentrations yielded negative L-band values, and the levels of W L and F L /F J for 0.1 mg·L −1 ATZ were significantly different from those of the control. For the treatments with Cd alone, L-band slightly increased (Fig. 4C, E), and there was no significant difference in W L and F L /F J compared with the control (Fig. 4D, F). ATZ combined with Cd, except for 0.1ATZ + 5Cd, decreased L-band (Fig. 4C, E) with significantly smaller W L and F L /F J compared with the control (Fig. 4D, F). In addition, when ATZ concentrations were 1.0 and 2.0 mg·L −1 , the W L and F L /F J values for ATZ alone were significantly larger than those of ATZ combined with Cd.

K-band
The inhibition of the donor side of PSII results in the appearance of K-step in the chlorophyll a fluorescence rise (Lazár 2006). The double normalized chlorophyll a fluorescence curve in the 0.02-2 ms time range revealed the changes of K-band (Fig. 5). I. pseudacorus exhibited a prominent rise of fluorescence intensity at K points of the transient curve in response to ATZ alone at concentrations of 1.0 and 2.0 mg·L −1 (Fig. 5A), and the values of W K and F K /F J were significantly higher than those of the control (Fig. 5B). The exception was the treatments with ATZ alone at 0.1 mg·L −1 which showed a negative K-band with a significantly decreased F K /F J (Fig. 5A, B), but this significant difference was not observed in W K (Fig. 5B). Compared with the control, Cd alone induced the occurrence of K-band, and only Cd at 10 mg·L −1 increased significantly the values of W K and F K /F J . It was observed that the K-band was negative for 2.0 mg·L −1 ATZ combined with Cd (Fig. 5C, E), and F K /F J decreased significantly compared to that of the control, while W K did not increase pronouncedly (Fig. 5D, F). On the other hand, The relative variable fluorescence The relative variable fluorescence

Energy conservation and photosynthetic performance
In the JIP-test parameters, the absorbed light energy (ABS/ RC), captured light energy (TR 0 /RC), thermally dissipated light energy (DI 0 /RC), and the energy used for electron transfer (ET 0 /RC) on the unit of active center were used to indicate the energy transformation in the RSII (Liu et al. 2020). For the treatment with single ATZ, there was no significant difference in ABS/RC (Fig. 6A) and DI 0 /RC (Fig. 6D) between ATZ at levels below 0.5 mg·L −1 and the control, while ATZ at levels of 1.0 and 2.0 mg·L −1 increased significantly ABS/RC and DI 0 /RC; ATZ at the level of 1.0 mg·L −1 also caused a significant increase in TR 0 / RC (Fig. 6B); ET 0 /RC decreased significantly at all ATZ concentrations (Fig. 6C). For the treatment with single Cd, ABS/RC, TR 0 /RC, and DI 0 /RC increased significantly, while ET 0 /RC did not differ from the control. For the combined treatment with ATZ and Cd, ABS/RC and DI 0 /RC at high level ATZ (1.0 and 2.0 mg·L −1 ) were significantly lower than those of corresponding single ATZ treatments; ET 0 /RC of 1.0 and 2.0 mg·L −1 ATZ was significantly higher than that of ATZ alone when Cd concentration was 10.0 mg·L −1 , but this significant difference was not observed when Cd concentration was 5.0 mg·L −1 . PI abs is the performance index for energy conservation from exciton to the reduction of intersystem electron acceptors, and PI total is performance index for energy conservation from exciton to the reduction of PSI end acceptors (de Souza et al. 2020). For the treatment of single ATZ or single Cd, Fig. 4 The fluorescence rise kinetics normalized by F O and F K as W OK on a linear time scale from 0.02 to 0.3 ms, and the values of W L and F L /F J in I. pseudacorus treated with ATZ alone (A, B), treated with ATZ plus 5 mg·L −1 Cd (C, D), and treated with ATZ plus 10 mg·L −1 Cd (E, F). Each parameter followed by different small case letters is significantly different (P < 0.05) among different treatments according to the Duncan's test    (Fig. 7A) and PI total (Fig. 7B) declined significantly compare with the control. For the combined treatment with ATZ and Cd, the values of PI abs and PI total were greater than those of the corresponding single ATZ treatment when ATZ concentrations were at levels 1.0 and 2.0 mg·L −1 .

Discussion
In the present study, the individual and combined effect of ATZ and Cd on photosystems behavior in I. pseudacorus leaves was evaluated using the fluorescence rise OJIP kinetics. The prompt fluorescence kinetics based on the energy cascade is useful for quantifying variations in the donor and acceptor sides of PS II, and is now widely employed to investigate electron transport in plant leaves because of its fast, non-invasive, precision and inexpensive characteristics (Guo et al. 2020;Sun et al. 2020). A typical chlorophyll a fluorescence curve has four basic steps: O (20 μs) refers to the initial fluorescence level (F o ), J (2 ms) and I (30 ms) are intermediate level, and P is the peak level (F m ) (Gao et al. 2018;Guo et al. 2020). In the OJIP curve, the O-J phase represents the reduction of the acceptor side of PSII, the J-I phase represents the reduction of the PQ pool, and the I-P phase reflects the reduction of the acceptor side of PSI (Ji et al. 2018).

Effects of ATZ alone
The herbicide ATZ, as a PSII inhibitor, mainly lies in the fact that it can bind to the Q B -binding niche on the D1 protein of PSII, thus blocks the photosynthetic electron transport at PS II and excitation energy transfer from PS II to PS I (Majewska et al. 2018). In the present work, ATZ alone caused a significant increase in F o and a dramatic decline in F v /F o except for ATZ concentration of 0.1 mg·L −1 . F o fluorescence mainly depends on the initial exciton density in the PSII antenna pigment and, the structure state of the excitation energy transfer between the antenna pigments and the antenna pigment to the RCII (Krause and Weis 1984). Higher F o is symptomatic of a stronger overall emission which derived from functional changes within photosynthetic membranes (Kriedemann et al. 1985), corresponding to energy losses in the pigments of the PSII antenna and less efficient energy transfer among the antennae complexes toward the RCII (Kalaji et al. 2018), and subsequently causing a decrease in the rate constant of energy trapped by RCIIs (Hassannejad et al. 2020). Therefore, this increase in F o was mainly for two reasons: (1) the increase of initial exciton density resulted from the blocking of electron transport beyond Q A ; and (2) the low connectivity of PSII units caused low energy transfer from LHCII to RCII and thus higher fluorescence from LHCII (Tovuu et al. 2013;Ghassemi-Golezani and Lotfi 2015). The F v /F o , responding very sensitively to any changes in F v and/or F o , is also used to evaluate the photosynthetic quantum conversion efficiencies in dark-adapted (Babani and Lichtenthaler 1996;Govindachary et al. 2004). The progressive and significant decrease in F v /F o with increasing ATZ concentrations indicated that the water-splitting system of the donor side of PSII might be seriously impacted by high concentration ATZ (Mosadegh et al. 2019). F m could reflect the status of the electronic pass through the PSII. It primarily depends on the amount of Q A − (Singh-Rawal et al. 2010), and is high when the Q B binding site is occupied by a PSII herbicide (Prášil et al. 2018). In addition, increasing F m is associated with the rise of PSII activity due to conformational changes in the D1 protein (Ghassemi-Golezani and Lotfi 2015). In this study, an increasing trend in F m of treatments of single ATZ at concentrations of 0.1 and 0.5 mg·L −1 illustrated better physiological regulation ability and adaptability of I. pseudacorus to ATZ stress at this level. A gradually significant decline in the ratio F v /F o of treatments with ATZ alone above 0.1 mg·L −1 displayed that the potential photochemical efficiency of PSII decreased with increasing ATZ concentrations, which was associated with a disruption of watersplitting system (Borkowska 2002). Furthermore, significant changes in F v /F o between ATZ concentrations indicated that F v /F o was a sensitive indicator of ATZ induced limitation of photosynthesis in I. pseudacorus.
In this study, the relative variable fluorescence induction kinetic curves on a logarithmic time scale showed that J-I and I-P phases disappeared due to a sharp increase of J-and I-step closed to the same level as P-step in the treatments with ATZ alone at levels above 0.1 mg·L −1 . These results implied that, under single ATZ stress, Q A was reduced entirely and the electron flow from Q A − to Q B was blocked (Chen et al. 2007). Additionally, the loss of I-step from chlorophyll fluorescence induction curves can be linked to the inactivation of PSI and suppression of the cyclic phosphorylation (Cetner et al. 2020). The increased value of V J and V I parameters suggests the accumulation of reduced Q A and PQ, which cannot transfer electrons to the dark reactions (Kalaji et al. 2014). It was noticeable that the remarked increase of V J and F J /F I was mainly attributed to the increase in F J further confirming that PSII electron transport rate decreased in I. pseudacorus (Singh-Rawal et al. 2010). A sharp rise of J-step contributed to the large accumulation of Q A − , causing an increase in the net rate of the RCs' closure. As indicated by the larger values of M 0 in treatments of single ATZ relative to the control, it also proved that ATZ interrupted PSII electron transport beyond Q A (Guo et al. 2020) .
The appearance of the K-step is due to the accumulation of Y Z + (a tyrosine residue in the D1 subunit), which results from an imbalance within PSII between the electrons leaving the RC at the acceptor side and the electrons donated by the donor side, and is widely used to represent the state of the active OEC centers at the PSII donor side (Strasser 1997;Strasser et al. 2004). W K and the ratio F K /F J reflect the changes in the amplitude in the K-step (Strasser 1997;Lu and Zhang 2000). The considerable increase in W K and F K /F J was observed in the treatments with single ATZ at or above 1.0 mg·L −1 , indicating a noticeable imbalance between electron flow from the OEC to Y Z and electron flow from P 680 to Q A and beyond (Strasser 1997), and the inactivation of the PSII donor side (Mosadegh et al. 2019). The L-band can give information on the energy connectivity or grouping among PS II units. A positive L-band indicates the lower energetic connectivity while a negative L-band is a sign of greater grouping and more efficient energy exchange between the neighboring PSII units (Kalaji et al. 2018;Martins et al. 2018). ATZ alone at higher levels (≥ 1.0 mg·L −1 ) caused a prominent lift in L-band with the significant rise in the values of W L and F L /F J indicating a decrease in energetic connectivity among PSII units. However, I. pseudacorus grown in a medium with ATZ alone of 0.1 showed a negative L-band with significantly decreased W L and F L /F J , implicating higher connectivity of antennae to RCII units and more efficient utilization of excitation energy (Mlinarić et al. 2017). This corresponded well with the increase in the initial slope of F t /F o curve. These results suggested that this plant species exhibited relatively high tolerance to ATZ stress at this level through enhancing the electron supply from P 680 to Q A and energetic connectivity. To minimize damage to the photosynthetic apparatus, excess energy can be dissipated as nonphotochemical processes, including thermal dissipation (Silva et al. 2021). In this study, the gradual increase in DI 0 / RC with increasing ATZ concentration suggested that partial excitation energy was dissipated in nonphotochemical processes, especially achieving significant differences at higher levels of ATZ (≥ 1.0 mg·L −1 ). The significant increase in the nonphotochemical quenching was also observed by Silva et al. (2021) in determining the sensitivity level of Handroanthus heptaphyllus to ATZ. Though this increase diminished the fraction of energy delivered to PSII centers, it played an important role in the mitigation against photodamage, which reflected the tolerance of I. pseudacorus to ATZ by activating the photoprotective responses (Cazzaniga et al. 2013).
Though the target action site of ATZ was located on the electron acceptor side in PSII, some indirect effects, such as the inactivation of OEC and PSI, can be seen from the appearance of K-step and the disappearance of I-step, respectively. Therefore, ATZ eventually caused the overall suppression of plant photosynthesis owing to the less grouping of the PSII units and the imbalance of the electron flow . Besides, the overall impact of ATZ on the photosynthesis was likely linked to the oxidative stress in plants trigger by blocking of photosynthetic electron transport (Zhang et al. 2014;Guo et al. 2020).

Effects of Cd alone
Cd was also responsible for these adverse changes in photosynthetic characteristics in I. pseudacorus leaves, as can be seen from the positive K-band and significant decreases in PI abs and PI total caused by Cd alone. Other studies also indicated that Cd inhibited both the donor and the acceptor site of PSII and, thus, reduced the photosynthetic capacity of plants (Sigfridsson et al. 2004;Kalaji et al. 2016;Huihui et al. 2020).

Effects of ATZ combined with Cd
Compared with ATZ alone, ATZ combined with Cd presented lower F o and higher F v /F o , which meant less light energy dissipation in antenna and smaller structural alterations on the donor side of PSII (Mathur et al. 2011). This was consistent with what was indicated clearly by the higher ET 0 /RC and lower DI 0 /RC. The fluorescence rise transient obtained from the treatments of ATZ combined with Cd, especially at high ATZ concentration (≥ 1.0 mg·L −1 ), showed an apparent I-P phase, and a low fluorescence intensity at K point relative to ATZ alone, as well as negative L-band values, suggesting that the coexistence of Cd conduced to maintain the balance of electron transfer from the donor side to the acceptor side of PSII, and was helpful for the maintenance of the energetic connectivity and efficient consumption of the excitation energy and stability of the system upon ATZ treatments. It was noteworthy that no significant effect of single ATZ at 2 mg·L −1 on W K meant that the significant decrease in F K /F J was attributed mainly to the decrease of the J-step level, suggesting that the presence of Cd mainly reduced the damage to the acceptor side in this situation. This coincided exactly with the action mechanism of ATZ by which ATZ displaced the secondary quinone acceptor, Q B , from its binding site at the D1 protein of PSII. ATZ stressinduced damage to the structure and function of PSII and PSI in leaves of I. pseudacorus was alleviated by the presence of Cd, as indicated by the significantly higher values of PI abs and PI total in combined treatments relative to single ATZ. The inference was also supported by the initial slope of the standardized fluorescence transient, an indication of the electron transfer rate from P 680 to Q A , which was higher in combined treatments relative to the corresponding single ATZ treatments, suggesting that the electron transfer rate in PSII was enhanced in the presence of Cd. These results were quite matched with a former report, which showed that ATZ in binary mixtures with Cd was less toxic than ATZ alone to rice seedlings (Su et al. 2005). The enhanced photosynthetic performance in the presence of Cd may be related to the formation of the complex between ATZ and Cd 2+ . ATZ contains one or more binding sites (most probably NH) and they can potentially participate in the formation of the metal ion-ATZ dimer complexes (Su et al. 2005;Singh et al. 2020). A previous study confirmed that Cd can potentially complex with either monomeric or dimeric ATZ to form both anhydrous and hydrated complex forms (Meng and Carper 2000). Moreover, the N atoms of ATZ have stronger complexation ability with Cd 2+ than with other divalent metal ions, such as Zn 2+ and Cu 2+ (Martin et al. 1998). Since Cd 2+ occupied the binding site of ATZ with D1 protein, the competitive binding to the Q B -niche of the D1 protein was limited, contributing to slighter inhibitory effects of ATZ on I. pseudacorus. For most of the selected JIP-test parameters, there were no statistically significant differences between the two Cd concentrations selected in this study. This was probably associated with the concentration ratio of ATZ to Cd. A former research indicated that at the 1:1 molar concentration ratio, the formation of the complex between ATZ and Cd notably reduced the individual toxicities to rice seedlings (Su et al. 2005). In addition, the lower magnitude of the J-P phase in comparison with the control indicated that the PSII donor side was still partially inhibited by ATZ combined with Cd. Also, the electron transport flux and energy conservation capacity of the photosynthetic apparatus decreased, seen as the lower ET 0 /RC, as well as the significant decrease in PI abs and PI total relative to the control. Therefore, ATZ combined with Cd still had inhibitory effects on photosynthetic performance of I. pseudacorus.

Limitations and perspectives
In the present study, molar concentration ratios of ATZ to Cd, ranging from 0.005 to 0.2, were far less than 1. For better understanding of the toxicity of mixture of ATZ and Cd to plants, the ratio ranges both above and below 1 should be applied in the future study. In addition, experimental data of ATZ accumulation in both root and shoot of I. pseudacorus should be provided for clearly understanding ATZ uptake and transfer by plants in the presence of Cd, furtherly to get deeper insights in the action mode of ATZ under combined pollution.

Conclusion
ATZ showed the similar inhibitory activity on electrons transfer from Q A to Q B , mainly increasing the J-step on prompt fluorescence curves of non-target aquatic plant species, I. pseudacorus. It also had indirect impacts on the donor side because of the imbalance of electron in and out of RCs and subsequently the less efficient energy exchange between the neighboring PSII units. I. pseudacorus exhibited tolerance to ATZ at concentrations not above 0.5 mg·L −1 by increasing the rate of the electron transfer from P 680 to Q A and energetic connectivity between PSII units, but was sensitive to ATZ at higher levels (≥ 1 mg·L −1 ). Due to the formation of ATZ-Cd complex, ATZ mixed with Cd displayed weaker influence on the photosynthesis ability relative to ATZ alone, whereas still had harmful effects on the photosynthetic apparatus of I. pseudacorus. These results imply that the potential ecological risk of combined pollution of ATZ and Cd in waters still exist and cannot be ignored, even though lower than that of ATZ alone.
Author contribution All authors contributed significantly to the preparation of this manuscript. Qinghai Wang and Xiaoe Que designed the research, Dongyu Xie and Lei Peng performed experiments, Qinghai Wang and Dongyu Xie analyzed data and drafted the manuscript, Cui Li performed plant cultivation and management, Chuansheng Chen supported methodology. All authors read the manuscript and approved the final manuscript.