Chlorophyll Fluorescence Imaging Analysis for Elucidating the Mechanism of Photosystem II Acclimation to Cadmium Exposure in the Hyperaccumulating Plant Noccaea caerulescens

We provide new data on the mechanism of Noccaea caerulescens acclimation to Cd exposure by elucidating the process of photosystem II (PSII) acclimation by chlorophyll fluorescence imaging analysis. Seeds from the metallophyte N. caerulescens were grown in hydroponic culture for 12 weeks before exposure to 40 and 120 μM Cd for 3 and 4 days. At the beginning of exposure to 40 μM Cd, we observed a spatial leaf heterogeneity of decreased PSII photochemistry, that later recovered completely. This acclimation was achieved possibly through the reduced plastoquinone (PQ) pool signaling. Exposure to 120 μM Cd under the growth light did not affect PSII photochemistry, while under high light due to a photoprotective mechanism (regulated heat dissipation for protection) that down-regulated PSII quantum yield, the quantum yield of non-regulated energy loss in PSII (ΦNO) decreased even more than control values. Thus, N. caerulescens plants exposed to 120 μM Cd for 4 days exhibited lower reactive oxygen species (ROS) production as singlet oxygen (1O2). The response of N. caerulescens to Cd exposure fits the ‘Threshold for Tolerance Model’, with a lag time of 4 d and a threshold concentration of 40 μM Cd required for the induction of the acclimation mechanism.


Introduction
Cadmium is a non-essential heavy metal that can occur in the environment in high concentrations as a consequence of numerous human activities, thus becoming toxic to all organisms [1][2][3][4][5]. Plants have developed several exclusive and effective mechanisms for Cd detoxification and tolerance, including control of Cd influx and acceleration of Cd efflux, Cd chelation and sequestration, Cd remobilization, and scavenging of Cd-induced reactive oxygen species [5][6][7][8][9][10].
Hyperaccumulators are plant species that vigorously take up heavy metals, translocate them into the above-ground parts and isolate them into a risk-free state [4,11]. These plants can accumulate several percent of heavy metals in their dry mass [4]. Hyperaccumulators also have to stock the absorbed heavy metal in a manner that is not detrimental to vital enzymes and especially photosynthesis [12,13]. Hyperaccumulators can be used for phytoremediation and also for   The quantum yield of regulated nonphotochemical energy loss in PSII, that is the quantum yield for dissipation by down regulation in PSII The quantum yield of non-regulated energy loss in PSII Calculated as F s /F m 1 − q P The fraction of closed PSII reaction centers Calculated as 1 − q P

Changes in the Maximum Quantum Efficiency of PSII Photochemistry after Cd Exposure
At the beginning of exposure to 40 µM Cd, the maximum quantum efficiency of PSII photochemistry (F v /F m ) in N. caerulescens decreased significantly but increased to control values at 120 µM Cd ( Figure 1).

Changes in the Allocation of Absorbed Light Energy in PSII after Cd Exposure
The quantum yield of photochemical energy conversion in PSII (ΦPSΙΙ), at both growth light (GL) and high light (HL) intensity decreased significantly compared to the control, after 3 d at 40 μM Cd, while it improved during the 4 d ( Figure 2). However, ΦPSΙΙ increased to control values after 3 d at 120 μM Cd at GL and stabilized to control values after 4 days of exposure (Figure 2a). High light (HL) exposure to 120 μM Cd resulted in decreased ΦPSIΙ compared to controls (Figure 2b).

Changes in the Allocation of Absorbed Light Energy in PSII after Cd Exposure
The quantum yield of photochemical energy conversion in PSII (Φ PSII ), at both growth light (GL) and high light (HL) intensity decreased significantly compared to the control, after 3 d at 40 µM Cd, while it improved during the 4 d ( Figure 2). However, Φ PSII increased to control values after 3 d at 120 µM Cd at GL and stabilized to control values after 4 days of exposure (Figure 2a). High light (HL) exposure to 120 µM Cd resulted in decreased Φ PSII compared to controls (Figure 2b).

Figure 1.
Changes in the maximum quantum efficiency of PSII (Fv/Fm) in N. caerulescens plants grown at 0 (control), 40 or 120 μM Cd 2+ for 3 and 4 days.

Changes in the Allocation of Absorbed Light Energy in PSII after Cd Exposure
The quantum yield of photochemical energy conversion in PSII (ΦPSΙΙ), at both growth light (GL) and high light (HL) intensity decreased significantly compared to the control, after 3 d at 40 μM Cd, while it improved during the 4 d ( Figure 2). However, ΦPSΙΙ increased to control values after 3 d at 120 μM Cd at GL and stabilized to control values after 4 days of exposure (Figure 2a). High light (HL) exposure to 120 μM Cd resulted in decreased ΦPSIΙ compared to controls (Figure 2b). The quantum yield of regulated non-photochemical energy loss in PSII (ΦNPQ) decreased significantly compared to the control after 3 d at 40 μM Cd at GL and increased to control values during the 4 d ( Figure 3a). Exposure to 120 μM Cd resulted in decreased ΦNPQ at GL compared to controls during the 4 d (Figure 3a). At HL, ΦNPQ remained unchanged at 40 μM Cd, but increased significantly at 120 μM Cd (Figure 3b). The quantum yield of regulated non-photochemical energy loss in PSII (Φ NPQ ) decreased significantly compared to the control after 3 d at 40 µM Cd at GL and increased to control values during the 4 d ( Figure 3a). Exposure to 120 µM Cd resulted in decreased Φ NPQ at GL compared to controls during the 4 d (Figure 3a). At HL, Φ NPQ remained unchanged at 40 µM Cd, but increased significantly at 120 µM Cd (Figure 3b).  The quantum yield of non-regulated energy loss in PSII (ΦNO), a loss process due to PSII inactivity, at both GL and HL intensity, increased significantly compared to the control after 3 d exposure to 40 μM Cd, while during the 4 d it decreased compared to 3 d ( Figure 4). After exposure to 120 μM Cd for 3 d at GL, ΦNO retained the same values compared to the controls, but increased during the 4 d ( Figure 4a). However, ΦNO decreased more than the control values at 120 μM Cd at HL (Figure 4b). The quantum yield of non-regulated energy loss in PSII (Φ NO ), a loss process due to PSII inactivity, at both GL and HL intensity, increased significantly compared to the control after 3 d exposure to 40 µM Cd, while during the 4 d it decreased compared to 3 d ( Figure 4). After exposure to 120 µM Cd for 3 d at GL, Φ NO retained the same values compared to the controls, but increased during the 4 d ( Figure 4a). However, Φ NO decreased more than the control values at 120 µM Cd at HL (Figure 4b).
The quantum yield of non-regulated energy loss in PSII (ΦNO), a loss process due to PSII inactivity, at both GL and HL intensity, increased significantly compared to the control after 3 d exposure to 40 μM Cd, while during the 4 d it decreased compared to 3 d ( Figure 4). After exposure to 120 μM Cd for 3 d at GL, ΦNO retained the same values compared to the controls, but increased during the 4 d ( Figure 4a). However, ΦNO decreased more than the control values at 120 μM Cd at HL (Figure 4b).

Non-Photochemical Quenching and Electron Transport Rate in Response to Cd
Non-photochemical quenching (NPQ) that reflects heat dissipation of excitation energy, decreased significantly compared to the control after 3 d at 40 μM Cd at GL, while it improved during the 4 d ( Figure 5a). Exposure to 120 μM Cd resulted in decreased NPQ at GL compared to controls during the 4 d (Figure 5a). At HL, NPQ decreased significantly compared to the control after 3 d exposure to 40 μM Cd, and increased to control values during the 4 d, while after exposure to 120 μM Cd increased significantly compared to the controls (Figure 5b).

Non-Photochemical Quenching and Electron Transport Rate in Response to Cd
Non-photochemical quenching (NPQ) that reflects heat dissipation of excitation energy, decreased significantly compared to the control after 3 d at 40 µM Cd at GL, while it improved during the 4 d ( Figure 5a). Exposure to 120 µM Cd resulted in decreased NPQ at GL compared to controls during the 4 d (Figure 5a). At HL, NPQ decreased significantly compared to the control after 3 d exposure to 40 µM Cd, and increased to control values during the 4 d, while after exposure to 120 µM Cd increased significantly compared to the controls (Figure 5b). The electron transport rate (ETR), at both GL and HL intensity, decreased significantly compared to the control after 3 d at 40 μM Cd, while it improved during the 4 d ( Figure 6). However, ETR increased to control values after 3 d exposure to 120 μM Cd at GL and stabilized to control values after 4 d exposure (Figure 6a). High light exposure to 120 μM Cd resulted in decreased ETR compared to the controls (Figure 6b).  The electron transport rate (ETR), at both GL and HL intensity, decreased significantly compared to the control after 3 d at 40 µM Cd, while it improved during the 4 d ( Figure 6). However, ETR increased to control values after 3 d exposure to 120 µM Cd at GL and stabilized to control values after 4 d exposure (Figure 6a). High light exposure to 120 µM Cd resulted in decreased ETR compared to the controls (Figure 6b).
The electron transport rate (ETR), at both GL and HL intensity, decreased significantly compared to the control after 3 d at 40 μM Cd, while it improved during the 4 d ( Figure 6). However, ETR increased to control values after 3 d exposure to 120 μM Cd at GL and stabilized to control values after 4 d exposure (Figure 6a). High light exposure to 120 μM Cd resulted in decreased ETR compared to the controls (Figure 6b).

Changes in the Redox State of PSII after Cd Exposure
The redox state of QA (qP) that is a measure of the fraction of open PSII reaction centers, at both GL and HL intensity, decreased significantly compared to the control after 3 d at 40 μM Cd, while it improved during the 4 d ( Figure 7). However, qP increased to control values after 3 d exposure to 120 μM Cd at GL and stabilized to control values after 4 d exposure (Figure 7a). High light exposure to 120 μM Cd resulted in a more reduced redox state of QA compared to controls, i.e., a lower fraction of open PSII reaction centers (Figure 7b).

Changes in the Redox State of PSII after Cd Exposure
The redox state of QA (q P ) that is a measure of the fraction of open PSII reaction centers, at both GL and HL intensity, decreased significantly compared to the control after 3 d at 40 µM Cd, while it improved during the 4 d ( Figure 7). However, q P increased to control values after 3 d exposure to 120 µM Cd at GL and stabilized to control values after 4 d exposure (Figure 7a). High light exposure to 120 µM Cd resulted in a more reduced redox state of QA compared to controls, i.e., a lower fraction of open PSII reaction centers (Figure 7b).

Spatiotemporal Variation of PSII Responses to Cd Exposure
The major veins (mid-vein, first-and second-order veins) in N. caerulescens leaves grown under control growth conditions at both GL and HL defined areas with a lower fraction of open PSII reaction centers or a more reduced redox state of QA, while mesophyll cells expressed larger spatial heterogeneity with a larger fraction of open PSII reaction centers or a more oxidized redox state (Figures 8e and 9d).
The maximum quantum efficiency of PSII photochemistry (Fv/Fm) show the smallest spatial heterogeneity even though it decreased significantly at 40 μM Cd and increased to control values at 120 μM Cd (Figure 8a). The quantum yield of photochemical energy conversion in PSII (ΦPSΙΙ) decreased significantly after 3 d at 40 μM Cd at GL, while it improved during the 4 d, showing a high spatiotemporal leaf heterogeneity (Figure 8b). Among the chlorophyll fluorescence parameters with high spatiotemporal heterogeneity observed at GL, were the images of the quantum yield of nonregulated energy dissipated in PSII (non-regulated heat dissipation, a loss process due to PSII inactivity) (ΦNO) (Figure 8d) and the images of the redox state of the PQ pool (qP) (Figure 8e). The most severely affected leaf area after 3 d at 40 μM Cd, was the left and right leaf side, while the central area was less affected (Figures 8d,e). At the left and right leaf side after 3 d exposure to 40 μM Cd, the quantum yield of non-regulated energy loss in PSII (ΦNO) increased; thus, these areas exhibited increased singlet oxygen ( 1 O2) production (Figure 8d), and also presented the lower qP values ( Figure   Figure 7. Changes in the photochemical fluorescence quenching, that is the relative reduction state of Q A , reflecting the fraction of open PSII reaction centers (q P ) measured at (a) 300 µmol photons m −2 s −1 or (b) 1000 µmol photons m −2 s −1 . N. caerulescens plants were grown at 0 (control), 40, or 120 µM Cd 2+ for 3 and 4 days.

Spatiotemporal Variation of PSII Responses to Cd Exposure
The major veins (mid-vein, first-and second-order veins) in N. caerulescens leaves grown under control growth conditions at both GL and HL defined areas with a lower fraction of open PSII reaction centers or a more reduced redox state of QA, while mesophyll cells expressed larger spatial heterogeneity with a larger fraction of open PSII reaction centers or a more oxidized redox state (Figures 8e and 9d).
The maximum quantum efficiency of PSII photochemistry (F v /F m ) show the smallest spatial heterogeneity even though it decreased significantly at 40 µM Cd and increased to control values at 120 µM Cd (Figure 8a). The quantum yield of photochemical energy conversion in PSII (Φ PSII ) decreased significantly after 3 d at 40 µM Cd at GL, while it improved during the 4 d, showing a high spatiotemporal leaf heterogeneity (Figure 8b). Among the chlorophyll fluorescence parameters with high spatiotemporal heterogeneity observed at GL, were the images of the quantum yield of non-regulated energy dissipated in PSII (non-regulated heat dissipation, a loss process due to PSII inactivity) (Φ NO ) (Figure 8d) and the images of the redox state of the PQ pool (q P ) (Figure 8e). The most severely affected leaf area after 3 d at 40 µM Cd, was the left and right leaf side, while the central area was less affected (Figure 8d,e). At the left and right leaf side after 3 d exposure to 40 µM Cd, the quantum yield of non-regulated energy loss in PSII (Φ NO ) increased; thus, these areas exhibited increased singlet oxygen ( 1 O 2 ) production (Figure 8d), and also presented the lower q P values (Figure 8e). However, in the left and right leaf side after 4 d exposure to Cd, Φ NO decreased (Figure 8d) and the redox state of the PQ pool increased (q P ) (Figure 8e). At exposure to 120 µM Cd at GL, leaf spatial heterogeneity decreased, and both Φ NO (Figure 8d) and q P (Figure 8e) stabilized to control values. Exposure of N. caerulescens to HL increased the spatiotemporal leaf heterogeneity (Figure 9) and the plants suffered more from Cd toxicity during the 3 d of exposure to 40 μM Cd, but they recovered during the 4 d. However, exposure to 120 μM Cd at HL revealed mild effects. This was realized by an increase in ΦNPQ (Figure 9b) that down-regulated PSII quantum yield (ΦPSΙΙ) (Figure 9a) and decreased the quantum yield of non-regulated energy loss in PSII (ΦNO) (Figure 9c). Exposure of N. caerulescens to HL increased the spatiotemporal leaf heterogeneity (Figure 9) and the plants suffered more from Cd toxicity during the 3 d of exposure to 40 µM Cd, but they recovered during the 4 d. However, exposure to 120 µM Cd at HL revealed mild effects. This was realized by an

Discussion
The type of damage on PSII that has frequently been identified as the main target of Cd toxicity on photosynthesis strongly depends on light conditions [4,[43][44][45][46]. At GL, the damage of the PSII function is mainly due to the impairment that results from the replacement by Cd 2+ of the Mg 2+ ion in the chlorophyll molecules of the light-harvesting complex II, while in HL it is mainly from direct damage to the PSII reaction center [4,[44][45][46].
N. caerulescens leaves grown under control growth conditions at both GL and HL show a spatial heterogeneity in PSII functionality (Figures 8,9). This spatial heterogeneity may be attributed to 'patchy stomatal behavior', in which stomata in adjacent regions exhibit significantly different mean apertures from each other, resulting in significantly different stomatal conductance (gs) [47,48]. Stomatal conductance decreases when the stomata close; this is used as an indicator of the extent of stomatal opening [49,50]. It is assumed that spatial variation in the quantum efficiency of PSII photochemistry (ΦPSΙΙ) arises from local differences in internal CO2 concentrations, which in turn result from changes in stomatal conductance due to patchy stomatal behavior [51]. A body of evidence suggests that patterns of ΦPSΙΙ can be used to calculate stomatal conductance [51][52][53][54][55].

Discussion
The type of damage on PSII that has frequently been identified as the main target of Cd toxicity on photosynthesis strongly depends on light conditions [4,[43][44][45][46]. At GL, the damage of the PSII function is mainly due to the impairment that results from the replacement by Cd 2+ of the Mg 2+ ion in the chlorophyll molecules of the light-harvesting complex II, while in HL it is mainly from direct damage to the PSII reaction center [4,[44][45][46].
N. caerulescens leaves grown under control growth conditions at both GL and HL show a spatial heterogeneity in PSII functionality (Figures 8 and 9). This spatial heterogeneity may be attributed to 'patchy stomatal behavior', in which stomata in adjacent regions exhibit significantly different mean apertures from each other, resulting in significantly different stomatal conductance (g s ) [47,48]. Stomatal conductance decreases when the stomata close; this is used as an indicator of the extent of stomatal opening [49,50]. It is assumed that spatial variation in the quantum efficiency of PSII photochemistry (Φ PSII ) arises from local differences in internal CO 2 concentrations, which in turn result from changes in stomatal conductance due to patchy stomatal behavior [51]. A body of evidence suggests that patterns of Φ PSII can be used to calculate stomatal conductance [51][52][53][54][55].
At the beginning of exposure to 40 µM, Cd Φ PSII decreased significantly at the left and right leaf sides (Figures 8b and 9a), with a simultaneous decrease in Φ NPQ (Figures 8c and 9b) resulting in an increase of the quantum yield of non-regulated non-photochemical energy loss (Φ NO ) (Figures 8d and 9c). The increase in Φ NO indicates that photochemical energy conversion and photoprotective regulatory mechanism were insufficient, pointing to serious problems of the plant to cope with the absorbed light energy [56,57]. Φ NO consists of chlorophyll fluorescence internal conversions and intersystem crossing, which indicate the formation of singlet oxygen ( 1 O 2 ) via the triplet state of chlorophyll ( 3 chl *) [13,58,59]. After 3 d exposure to 40 µM Cd, N. caerulescens leaves exhibited increased 1 O 2 production at the left and right leaf sides, since Φ NO increased significantly at those areas. Thus, although Cd 2+ is a redox-inert element, it produces reactive oxygen species [28]. The simultaneous reduced PQ pool that was observed mainly at the left and right leaf sides mediated stomatal closure probably through the generation of mesophyll chloroplastic hydrogen peroxide (H 2 O 2 ) [60]. The stomatal closure at these areas implies decreased transpiration rates that slow down Cd supply.
During the 4 d exposure to 40 µM Cd, Φ PSII increased at the left and right leaf sides (Figures 8b and 9a), with a simultaneous increase in Φ NPQ (Figures 8c and 9b) resulting in a decrease of Φ NO (Figures 8d and 9c) compared to 3 d exposure. This response is attributed to both the possible Cd detoxification mechanism achieved by vacuolar sequestration, that seems to be the main mechanism for Cd detoxification [61][62][63], and to the reduced plastoquinone (PQ) pool that mediated stomatal closure and decreased Cd supply at the affected leaf area, leading to the acclimation of N. caerulescens to Cd exposure. Under exposure to 120 µM Cd at HL, the quantum yield of non-regulated energy loss in PSII (Φ NO ) decreased even more than control values, and thus exhibited lower singlet oxygen ( 1 O 2 ) production. This was due to the photoprotective mechanism that can divert absorbed light to other processes such as thermal dissipation, preventing the photosynthetic apparatus from oxidative damage [64][65][66][67][68][69][70].
The observed spatial heterogeneity in the quantum yield of linear electron transport (Φ PSII ) in N. caerulescens leaves exposed to 40 µM Cd for 3 d (Figures 8b and 9a) is in accordance to elemental imaging using laser ablation inductively-coupled plasma mass spectrometry, performed on whole leaves of the hyperaccumulator N. caerulescens that revealed differences in the supply of Cd over the whole leaf area, suggesting a heterogeneous distribution across the leaf [71]. Useful information can be obtained by combining chlorophyll fluorescence images, followed by laser ablation inductively-coupled plasma mass spectrometry on whole leaves of the hyperaccumulator N. caerulescens exposed to Cd.
It seems that spatiotemporal variations in the redox state of the PQ pool related to stomatal conductance, an indicator of the extent of stomatal opening [50], are interconnected to the heterogeneous distribution of Cd over the entire leaf area [71]. Thus, the spatial heterogeneity in the redox state of the PQ pool throughout the whole leaf area (Figures 8e and 9d) reveals a spatial supply of Cd across the leaf. Recently, Cd 2+ root influx has been shown to exhibit spatiotemporal patterns [72]. A heterogeneous distribution of a reduced PQ pool gives rise to a spatial distribution of H 2 O 2 accumulation [73]. Still, reactive oxygen species (O 2 − , H 2 O 2 ) production corresponds to spatial accumulation metal patterns [74]. In our work, the response of N. caerulescens to Cd exposure fits the 'Threshold for Tolerance Model', with a lag time or/and a threshold concentration required for the induction of a tolerance mechanism [75][76][77][78]. Concurrent to this model, mild stress or short exposure times can produce significant effects on plants, while moderate stress or longer exposure times have less or no effect [79].
In accordance with this model, 40 µM Cd and 3d exposure time caused significant effects on PSII functioning, while 120 µM Cd or 4d exposure time have less or no effect. A lag-time of 4d exposure to 40 µM Cd was required for N. caerulescens to activate stress-coping mechanisms.

Conclusions
Acclimation to Cd exposure was achieved through the possible Cd detoxification mechanism done by vacuolar sequestration and the reduced plastoquinone (PQ) pool signaling that mediated stomatal closure and decreased Cd supply at the affected leaf area. The response of N. caerulescens to Cd exposure fits the 'Threshold for Tolerance Model', with a lag time of 4 d and a threshold concentration of 40 µM Cd required for the induction of the acclimation mechanism through the reduced PQ pool that mediated stomatal closure probably by the generation of mesophyll chloroplastic hydrogen peroxide (H 2 O 2 ) [60], which acts as a fast acclimation signaling molecule [73,80], as well as activates the Cd detoxification mechanism through vacuolar sequestration [61][62][63]. The mode of Cd damage on PSII strongly depends on the irradiance conditions [4,[43][44][45][46]. Chlorophyll fluorescence imaging analysis is a non-invasive tool to assess the physiological status of plants and detect the impacts of environmental stress [81][82][83], permitting also the visualization of the spatiotemporal variations in PSII efficiency [76]. As it was shown in our experiments, it is also capable of elucidating the mechanism of photosystem II acclimation to Cd exposure.