Polyploidy Improves Photosynthesis Regulation within the Ranunculus auricomus Complex (Ranunculaceae)

Simple Summary Genome duplication or multiplication, polyploidy, has contributed substantially to the evolutionary success of plants. Polyploidy is often connected to a higher resilience to environmental stress. We have chosen the goldilocks, the Ranunculus auricomus complex, to study effects of light stress. In this species complex, diploid (2x), tetraploid (4x), and hexaploid (6x) cytotypes occur in Central Europe in both shaded and sun-exposed habitats. In this study, we exposed them to different photoperiods in climate growth chambers to explore how the efficiency of photosynthesis varied between the various ploidies (2x, 4x, and 6x). We used fluorescence experiments exploring the proportion of light that is captured for photosynthesis and the resulting energy fluxes. In addition, quenching coefficients can be calculated that inform about the capability of a plant to deal with excess light. We found that the polyploids can quench excess light better, which concurs with their adaptation to open habitats and their predominantly asexual mode of reproduction that is probably favored by low stress levels in the reproductive tissues. Abstract Polyploidy has substantially contributed to successful plant evolution, and is often connected to a higher resilience to environmental stress. We test the hypothesis that polyploids tolerate light stress better than diploids. The Ranunculus auricomus complex comprises diploid (2x), tetraploid (4x), and hexaploid (6x) cytotypes, the former of which occur in shaded habitats and the latter more in open, sun-exposed habitats in Central Europe. In this study, we experimentally explored the effects of ploidy and photoperiod extension on the efficiency of photosystem II in the three cytotypes in climate growth chambers. Quantum yields and various coefficients that can be calculated from light curve, Kautsky curve, and fluorescent transient OJIP experiments provided support for the hypothesis that, in comparison to diploids, the improved regulation of excess light by more efficient photochemical and non-chemical quenching in polyploids might have facilitated the adaptation to unshaded habitats. We suggest how lower stress levels in reproductive tissues of polyploids might have favored asexual reproduction.


Introduction
Polyploidy, whole-genome multiplication, denotes the presence of double or multiple chromosome sets by either genome doubling in a single species (autopolyploidy), or by hybridization of two species with associated genome doubling (allopolyploidy) [1]. Polyploidy enhances stress tolerance in response to drastic environmental changes by enabling more extensive adaptations as consequences of gene and genome duplication [2] and acts Three cytotypes of the R. auricomus complex (2x, 4x, and 6x [32]), with different ecological amplitudes, were exposed to the more or less adapted (10 h) and a moderately extended photoperiod (16.5 h) [25]. We examined the photosynthetic efficiency by performing various chlorophyll fluorescence experiments, quantum yield in light-adapted plants and dark-adapted plants, relative electron transport rates (rETR) during increasing actinic light (light curves), fluorescence induction in actinic light and darkness (Kautsky curves), and fast fluorescence transients (OJIP) [41][42][43]. Fluorescence coefficients that were obtained from the Kautsky curves allowed the calculation of a number of coefficients that quantify photosystem II quenching processes, PQ and NPQ [44]. The OJIP analysis provides coefficients that inform about PSII energy fluxes and general performance [45,46]. In terms of exploring photosynthesis efficiency of different ploidies and the effects of an extended photoperiod on them, to the knowledge of the authors, a comparable comprehensive range of fluorescence experiments has not been performed previously. The expected insights are aimed to contribute to a better understanding of to what extent photosynthetic performance and stress response can contribute to the formation of apomictic seeds in plants, a trait to which plant breeders pay still a lot of attention as one mechanism to fix vigorous hybrid genotypes over generations [47,48]. Since many crop plants are polyploids, we further contribute to the understanding of photosynthesis performance of different cytotypes.

Plant Material
Ranunculus auricomus plants comprised the same individuals as in a previously published study [26]. Table 1 provides an overview and Table S1 shows which experiments have been performed on which individuals. The sampling covers the whole morphological diversity of leaf shape within the complex, as the diploid cytotype forms both divided and undivided basal leaves [49], the tetraploids have mostly divided leaves [36], while both hexaploid clones have mostly undivided leaves [38]. We did not observe any effects of light treatments on morphological traits. All seedlings developed under equal conditions outdoors in the Old Botanical Garden of the University of Goettingen. At all ploidy levels, plants represented closely related hybrid genotypes [26]. Table 1. Origin of 2x, 4x, and 6x Ranunculus auricomus plants.

6x
Hexaploid plants (6x) were grown from seeds of natural hybrids of R. carpaticola × R. cassubicifolius from Slovakia (original clone 29 from a forest margin and clone 35 from a meadow [25,38]). Originally, both clones were combined to obtain sufficient replicate numbers.

Photoperiod Experiments in Climate Chambers
Growth conditions for control and stress treatments were optimized in a previous study [25]. The climatic chambers were set at a temperature of 18 • C, 60% humidity, and an average light intensity of about 250 µmol photons m −2 s −1 . The photoperiod spanned 10 h for control and 16.5 h for light stress treatment. Consequently, the plants received 9.0 mol m −2 d −1 and 14.8 mol m −2 d −1 , respectively.

Photosynthesis
We analyzed the effect of extended photoperiod on photosynthesis efficiency as a proxy of stress conditions. All photosynthesis analyses were performed on 3-11 plants per ploidy level and treatment. A fully developed basal leaf (upper side) that supported the inflorescence was chosen for the measurement. Fluorescence analyses started in the first weeks after sprouting in March 2019 when plants produced flower buds. Flower buds are covered by green sepals as photosynthetic tissue, of which no measurements were possible with the available equipment. Basal leaves had to suffice under assumptions that effects on photosynthesis were comparable in both leaf types.
The photosynthesis performances were observed by measurement of chlorophyll fluorescent intensity with a PAM fluorometer, PAR-FluorPen FP 110 LM/S (Photon Systems Instruments, Drásov, Czech Republic). Table S2 shows the used coefficients and their calculations on the basis of the obtained fluorescence data. By using the preprogrammed device protocol, first, we measured the leaves without pre-dark adaptation to record the PSII potential quantum yield y (φ PSII ). Then, plants were dark-adapted for at least 30 min for further parameter measurements, i.e., a light curve (LC) to determine the relative electron transport rate (rETR), Kautsky curve fluorescence induction decay (KC) to explore photochemical (PQ) and non-photochemical quenching (NPQ), and fast fluorescence transient curve (OJIP). Maximum quantum efficiency of PSII (φ max ) was determined in all dark-adapted experiments (LC, KC, and OJIP).

Light Curves (LC)
After dark adaptation, the experiment was started with an initial saturating super pulse in the dark, and subsequently, similar pulses were applied during an actinic light phase with increasing intensity, 10, 20, 50, 100, 300, and 500 µmol photons m −2 s −1 . Higher light intensities were not used due to a strong intensity decrease at 500 µmol photons m −2 s −1 actinic light. The measuring pulse was set to 0.09 µmol photons m −2 s −1 , the saturating and super pulses to 2400 µmol photons m −2 s −1 . The coefficient rETR was calculated as shown in Table S2.

Kautsky Curve Fluorescence Induction Decay (KC)
The measuring and super pulses were the same as in the light curve analysis (LC). The actinic light was set to 300 µmol photons m −2 s −1 . After the initial saturating super pulse, actinic light was switched on and lasted for 1 min. Within this time period, five additional super pulses were applied, the first one after 7 s and the following in 12 s intervals. The subsequent dark period lasted 88 s, during which three super pulses were applied, the first one after 11 s and the following in 26 s intervals. The fluorescence parameters F 0 , F M , F M , F(t), F 0 , and F M ( Figure S1) allowed calculating a number of coefficients (Table S2), among which non-photochemical quenching (NPQ), energy-dependent non-photochemical quenching coefficient (q E ), and photoinhibitory photochemical quenching coefficient (q I ) informed about non-photochemical quenching (NPQ) and photochemical quenching (PQ) together with the photochemical quenching coefficients q P and q L about photochemical quenching (for detailed explanations, see Table S2).

Fast Fluorescence Transient Analysis (OJIP)
OJIP (alternatively O-J-I-P) experiments followed Strasser et al. [43]. The OJIP curve was induced by a pulse of red light of 3000 µmol photons m −2 s −1 . The relative fluorescence intensity of the OJIP curve at time points O, J, I, and P, 0.5, 2, 30, and 1000 ms, respectively (F 0 , F J , F I , and F P ), was determined after dark adaptation. Several OJIP-specific coefficients can be calculated (Table S2). Additional individuals had to be included as the number of available remaining basal leaves from the hitherto investigated individuals was insufficient (Table S1).

Statistical Analysis
All statistical analyses were conducted with R (version 4.0.2 for Windows, R Foundation for Statistical Computing, [50]). Data handling and visualization were performed using the packages dplyr [51], tidyr [52], and ggplot2 [53]. Boxplots were created with ggpubr [54]. Differences in the photoperiods' effect on the photosynthetic performance of the different ploidies in the two photoperiods were explored with ANOVA and a 95% Tukey or 95% Duncan multiple range test (glmmTMB [55]).

Cytotype 6x_35 Showed Lower PSII Potential Quantum Yield (φ PSII ) in Both Photoperiods
The extended photoperiod did not affect the light-adapted φ PSII in the tested cytotypes, with clone 6x_35 being the exception that showed significantly lower φ PSII values ( Figure 1a). Initially, both hexaploid cytotypes were regarded as sufficiently uniform to contribute to the investigated 6x cytotypes. The high variability in the experimental data and the calculated coefficients caused us to re-evaluate to what extent each of the two clones, 6x_29 and 6x_35, contributed to the observed variation. One clone, 6x_29, was found to yield rather homogenous data compared to the other cytotypes. The other clone, 6x_35, by contrast, deviated by predominantly yielding lower φ PSII values. Consequently, and in contrast to the original experimental plan, the two clones were treated separately in attempts not to hamper statistical data evaluation by inhomogeneous sample groups. the different ploidies in the two photoperiods were explored with ANOVA and Tukey or 95% Duncan multiple range test (glmmTMB [55]).

Cytotype 6x_35 Showed Lower PSII Potential Quantum yield (ϕPSII) in Both Photope
The extended photoperiod did not affect the light-adapted ϕPSII in the teste types, with clone 6x_35 being the exception that showed significantly lower ϕPSII ( Figure 1a). Initially, both hexaploid cytotypes were regarded as sufficiently unif contribute to the investigated 6x cytotypes. The high variability in the experiment and the calculated coefficients caused us to re-evaluate to what extent each of t clones, 6x_29 and 6x_35, contributed to the observed variation. One clone, 6x_2 found to yield rather homogenous data compared to the other cytotypes. The othe 6x_35, by contrast, deviated by predominantly yielding lower ϕPSII values. Conseq and in contrast to the original experimental plan, the two clones were treated sep in attempts not to hamper statistical data evaluation by inhomogeneous sample gr

Cytotype 6x_35 Showed Lower PSII Maximum Quantum Efficiency (ϕmax) in Both Photoperiods
Likewise, the extended photoperiod did not affect the dark-adapted maximum tum yield of PSII (ϕmax) with the only exception again being clone 6x_35 that differ nificantly from the others (Figure 1b). Interestingly, after dark adaptation, 6x_35 that had been exposed to the extended photoperiod showed a lower median than case of ϕPSII, in which the median of the 10 h-exposed plants was lower.
(a) (b) Figure 1. Quantum yields of R. auricomus cytotypes exposed to two different photoperiods tential quantum yield, ϕPSII, from light-adapted plants; (b) maximum quantum yield, ϕm dark-adapted plants. For summary statistics, see Table S3, letters represent 95% Tukey. B show the 25th, median, and 75th percentile range, and jitter plots represent the exact data d tion. Figure 1. Quantum yields of R. auricomus cytotypes exposed to two different photoperiods: (a) potential quantum yield, φ PSII , from light-adapted plants; (b) maximum quantum yield, φ max , from dark-adapted plants. For summary statistics, see Table S3, letters represent 95% Tukey. Boxplots show the 25th, median, and 75th percentile range, and jitter plots represent the exact data distribution.

Cytotype 6x_35 Showed Lower PSII Maximum Quantum Efficiency (φ max ) in Both Photoperiods
Likewise, the extended photoperiod did not affect the dark-adapted maximum quantum yield of PSII (φ max ) with the only exception again being clone 6x_35 that differed significantly from the others (Figure 1b). Interestingly, after dark adaptation, 6x_35 plants that had been exposed to the extended photoperiod showed a lower median than in the case of φ PSII , in which the median of the 10 h-exposed plants was lower.

OJIP Experiment
This OJIP experiment focuses just on the effect of one super pulse on chlorophyll fluorescence after dark adaptation, albeit in a very meticulous way. Figure 3 presents a number of coefficients that quantify specific energy fluxes, ABS/RC, DI 0 /RC, ET0/RC, and TR 0 /RC, as well as a general performance index, PI ABS . Only cytotype 6x_35 differed from the others investigated. This became especially evident in the case of the photoperiod extension that caused significant increases in the energy flux coefficients, such as absorption flux per PS II reaction center (ABS/RC), dissipated energy (DI 0 /RC), electron transport flux (ET 0 /RC), and trapped energy per reaction center (TR 0 /RC). The performance index PI ABS decreased.
Albeit revealing no specific insights, Figure 2 points to the subtle differences between ploidies and photoperiods that merited further exploration. For example, (1) OJIP curves of the 6x_35 varied much more than the other cytotypes ( Figure 2a); (2) the light curves of the tetraploids differed clearly from the other cytotypes (Figure 2b), and (3) different photoperiods tended to cause some differentiation in the polyploid Kautsky curves but not those of the diploid cytotypes (Figure 2c).

OJIP Experiment
This OJIP experiment focuses just on the effect of one super pulse on chlorophyll fluorescence after dark adaptation, albeit in a very meticulous way. Figure 3 presents a number of coefficients that quantify specific energy fluxes, ABS/RC, DI0/RC, ET0/RC, and TR0/RC, as well as a general performance index, PIABS. Only cytotype 6x_35 differed from the others investigated. This became especially evident in the case of the photoperiod extension that caused significant increases in the energy flux coefficients, such as absorption flux per PS II reaction center (ABS/RC), dissipated energy (DI0/RC), electron transport flux (ET0/RC), and trapped energy per reaction center (TR0/RC). The performance index PIABS decreased.  Table S4, letters represent 95% Duncan. Boxplots show the 25th, median, and 75th percentile range, and jitter plots represent the exact data distribution.  Table S4, letters represent 95% Duncan. Boxplots show the 25th, median, and 75th percentile range, and jitter plots represent the exact data distribution.

Light Curve (LC) Experiment
At lower PPFDs, 10-100 µmol m −1 s −1 , the 6x_35 clone showed the lowest rETR whilst the other ploidies did not differ (Figure 4). At values ≥ 300 µmol m −2 s −1 , a remarkable change occurred, in which the 4x plants took over in showing the lowest values.
At lower PPFDs, 10-100 µ mol m −1 s −1 , the 6x_35 clone showed the lowest rETR whilst the other ploidies did not differ (Figure 4). At values ≥ 300 µmol m −2 s −1 , a remarkable change occurred, in which the 4x plants took over in showing the lowest values.
In terms of photoperiod extension, significant differences only appeared in lower PPFDs < 50 µ mol m −1 s −1 . Plants that were exposed to the longer photoperiod, 16.5 h, showed higher rETRs. In 4x plants, this effect was visible in PPFDs ≤ 50 µmol m −2 s −1 , in 2x plants only at 10 µ mol m −1 s −2 . Figure 4. Relative electron transport rate (rETR) from light curve (LC) experiments of R. auricomus cytotypes exposed to two different photoperiods in increasing actinic light PPFD; for summary statistics, see Table S5, letters represent 95% Duncan. Boxplots show the 25th, median, and 75th percentile range, and jitter plots represent the exact data distribution.

Kautsky Curce (KC) Experiment
The Kautsky curve experiment offered insights into non-photochemical and photochemical quenching, NPQ and PQ, of the various ploidies and how elongation of photoperiod affected these processes. Figure 5 shows the most important coefficients.
In terms of NPQ efficiency, the 4x and the 6x_35 plants showed the highest NPQ values that differed significantly from the 2x and 6x_29 plants when exposed to the shorter photoperiod. Its extension increased the levels to those of the 4x and 6x_35 plants. Notably, photoperiod extension did not affect the latter. The qE coefficient quantifies the amount of heat dissipation. The boxplots of qE reflected those of NPQ. The qI coefficient, which quantifies the photoinhibitory NPQ, indicated that especially the 6x_29 plants and, to a lesser extent, the 2x plants, developed rather weak activities compared to the other cytotypes. Relative electron transport rate (rETR) from light curve (LC) experiments of R. auricomus cytotypes exposed to two different photoperiods in increasing actinic light PPFD; for summary statistics, see Table S5, letters represent 95% Duncan. Boxplots show the 25th, median, and 75th percentile range, and jitter plots represent the exact data distribution.
In terms of photoperiod extension, significant differences only appeared in lower PPFDs < 50 µmol m −1 s −1 . Plants that were exposed to the longer photoperiod, 16.5 h, showed higher rETRs. In 4x plants, this effect was visible in PPFDs ≤ 50 µmol m −2 s −1 , in 2x plants only at 10 µmol m −1 s −2 .

Kautsky Curce (KC) Experiment
The Kautsky curve experiment offered insights into non-photochemical and photochemical quenching, NPQ and PQ, of the various ploidies and how elongation of photoperiod affected these processes. Figure 5 shows the most important coefficients.
In terms of NPQ efficiency, the 4x and the 6x_35 plants showed the highest NPQ values that differed significantly from the 2x and 6x_29 plants when exposed to the shorter photoperiod. Its extension increased the levels to those of the 4x and 6x_35 plants. Notably, photoperiod extension did not affect the latter. The q E coefficient quantifies the amount of heat dissipation. The boxplots of q E reflected those of NPQ. The q I coefficient, which quantifies the photoinhibitory NPQ, indicated that especially the 6x_29 plants and, to a lesser extent, the 2x plants, developed rather weak activities compared to the other cytotypes.
In terms of PQ efficiency, the PQ coefficient suggested a correlation with ploidy (lower values indicate higher efficiency). The q P and the q L coefficients quantify the number of reduceable (open) PSII reaction centers. They concur with PQ, with the exception that 6x_35 obviously quenched less efficiently than the other cytotypes, whereas the 6x_29 plants showed the best PQ values (lowest number of open reaction centers). Photoperiod extension reduced PQ in 2x, 4x, and 6x_29 plants, but not in 6x_35 plants.  Table S6, letters represent 95% Duncan. Boxplots show the 25th, median, and 75th percentile range, and jitter plots represent the exact data distribution.
In terms of PQ efficiency, the PQ coefficient suggested a correlation with ploidy (lower values indicate higher efficiency). The qP and the qL coefficients quantify the number of reduceable (open) PSII reaction centers. They concur with PQ, with the exception that 6x_35 obviously quenched less efficiently than the other cytotypes, whereas the 6x_29 plants showed the best PQ values (lowest number of open reaction centers). Photoperiod extension reduced PQ in 2x, 4x, and 6x_29 plants, but not in 6x_35 plants.

Ploidy and Photoperiod did Not Affect Quantum Yield in General
Quantum yields can be measured in darkness-adapted plants, ϕmax, and lightadapted plants, ϕPSII. The latter depends on the quality of the ambient light and thus provides a less comparable result than ϕmax. In this study, the different cytotypes gave similar results for both quantum yield types, the only exception being clone 6x_35 that showed remarkable reductions in both quantum yield types (Figure 1a,b). Quantum yield reduction is regarded as a general stress indicator [56]. Another study that compared Allium oleraceum L. cytotypes ≥ 4x [57] also found high similarities in their ϕmax values, as others did in a study on Phragmites australis (Cav.) Trin. ex Steud. cytotypes [58]. Likewise, ϕmax did not differ between 2x and 4x Lilium hybrids [59].
The photoperiod extension from 10.0 to 16.5 h caused no effect on ϕPSII and ϕmax. There are fewer studies that focus on the effects of photoperiod on quantum yield compared to those that focus on spectral composition. Leonardos and coworkers explored the effect of different light sources with different wavelength compositions in short day and long day photoperiods on the development of Chrysanthemum plants [60]. In this study, the authors  Table S6, letters represent 95% Duncan. Boxplots show the 25th, median, and 75th percentile range, and jitter plots represent the exact data distribution.

Ploidy and Photoperiod Did Not Affect Quantum Yield in General
Quantum yields can be measured in darkness-adapted plants, φ max , and light-adapted plants, φ PSII . The latter depends on the quality of the ambient light and thus provides a less comparable result than φ max . In this study, the different cytotypes gave similar results for both quantum yield types, the only exception being clone 6x_35 that showed remarkable reductions in both quantum yield types (Figure 1a,b). Quantum yield reduction is regarded as a general stress indicator [56]. Another study that compared Allium oleraceum L. cytotypes ≥ 4x [57] also found high similarities in their φ max values, as others did in a study on Phragmites australis (Cav.) Trin. ex Steud. cytotypes [58]. Likewise, φ max did not differ between 2x and 4x Lilium hybrids [59].
The photoperiod extension from 10.0 to 16.5 h caused no effect on φ PSII and φ max . There are fewer studies that focus on the effects of photoperiod on quantum yield compared to those that focus on spectral composition. Leonardos and coworkers explored the effect of different light sources with different wavelength compositions in short day and long day photoperiods on the development of Chrysanthemum plants [60]. In this study, the authors also compared high-pressure sodium lamps, similar to those that were used in this study, and found that photoperiod did not affect quantum yield that, in contrast to this study, was determined not by fluorescence but by CO 2 gas exchange measurements. Accordingly, the lower quantum yield of 6x_35 ( Figure 1) must have been caused by other factors.

Fast Fluorescence Transient Analyis (OJIP) Identified Stress in Cytotype 6x_35
The fast fluorescent transient analysis (OJIP), a highly detailed analysis of φ max , provides insights into energy fluxes and performance of PSII. In congruence with the quantum yield values, 2x, 4x, and 6x_29 did not differ in the calculable coefficients even when the photoperiod was extended. Only cytotype 6x_35 differed significantly (Figure 3), especially when the photoperiod was extended to 16.5 h. The increased absorption and trapped energy fluxes in the reaction centers (ABS/RC, TR 0 /RC) and the increased electron transport (ET 0 /RC) were reflected in higher dissipation energy fluxes (DI 0 /RC). These observations resembled those that are seen when plants are submitted to heat stress, temperatures ≥ 40 • C [43], or UV-C irradiation [61]. These effects were especially prominent in the 16.5 h photoperiod that aggravated the stress of the 6x_35 plants. Accordingly, the overall performance index PI ABS of 6x_35 plants decreased significantly.

Light Curve (LC) Experiments Showed That Increasing Actinic Light Caused 4x Cytotypes to Show Lower rETRs and 6x_35 to Resemble 2x and 6x_29 Cytotypes
The exploration of how quantum yield changes in the presence of increased actinic light is widely used to explore electron transport through PSII [41]. In this study, concerns about bad comparability do not apply because the investigated plants belong to a closely related species complex, in which similar pigment compositions and leaf structures can be assumed to prevail. A comparison of the rETR values ( Figure 4) showed that the 6x_35 plants showed the lowest rETR values at lower PPFDs (≤100 µmol m −2 s −1 ). These observations can be explained by the stressed state of plants from this clone.
However, when PPFD increased further, the 4x plants yielded the lowest rETR values and the 6x_35 plants started to resemble the 2x and 6x_29 cytotypes more closely. Differences of the two photoperiods, which were clearly visible at PPFDs ≤ 20 µmol m −2 s −1 , vanished at higher PPFDs. The stress of the 6x_35 cytotype that was observed in the OJIP analysis did not explain the effects that appeared in the LC experiment. Rather idiosyncratically, the rETR of the 6x_35 cytotype improved relatively with increasing PPFD. The OJIP experiment was carried out with dark-adapted plants and the LC experiment indicated that, rather idiosyncratically, increasing PPFD seemed to alleviate the stress of the 6x_35 plants.
Light period extension could have acted as a priming effect in 2x and 4x plants. At 10 µmol m −2 s −1 , the 16.5 h-exposed plants showed higher rETRs than the 10.0 h plants. It disappeared, however, at PPFDs ≥ 20 µmol m −2 s −1 , and was never visible in either of the 6x cytotypes, despite their pronounced differences in quantum yield. Optimization studies of light regimes for greenhouse-reared lettuce revealed that moderate extension of the light period may improve growth but daily received light amounts should not exceed certain dosages [62]. A certain priming potential was attributed to a combination of high light with high temperature in tomatoes [63]. These observations probably do not apply generally to all plant species. At least, albeit interpretation of the observed effects is difficult, subtle differences between the investigated cytotypes became evident.
Thus far, the discussed experiments did not include quenching mechanisms that have evolved to protect the photosynthesis in high-light conditions, NPQ and PQ [17,20]. The terms non-photochemical and photochemical are somewhat misleading because both of them involve chemical reactions. NPQ covers all quenching chemical reactions with the exception of the exciton trapping act, while PQ covers those affecting the exciton trapping act [64]. The Kautsky curve experiments (KC) informed about the quenching chemical reactions and represent the focus of the next section.

Kautsky Curve (KC) Analyses Identify Polploids as More Efficient Photosynthesis Quenchers
In this experiment, the actinic light was set to 300 µmol m −2 s −1 , which represents the generally assumed saturation PPFD of photosynthesis in plants [65]. Its advantage is that it allows for exploring the quenching processes in the investigated plant leaves ( Figure 5). In terms of NPQ, the polyploid 4x and 6x_35 cytotypes proved to be the most efficient ones. The coefficient q E , which quantifies heat dissipation, provided a highly similar picture to that of coefficient NPQ. The coefficient q I , which quantifies photoinhibitory NPQ, the minor component of NPQ, yielded similar results. NPQ seemed to increase in polyploids. Rakić and co-workers, who compared the diploid Ramonda nathaliae Panč et Petrov with the hexaploid R. serbica Panč in terms of photosynthesis after rewetting following a desiccation period, to which these rock-dwelling Gesneriaceae are exposed during the summer months, reported similar findings [66]. The hexaploid R. serbica showed higher NPQ levels than the diploid R. nathaliae. A study on artificial Lilium hybrids confirmed these observations [59]. The big differences in the coefficient values between the 6x_29 and 6x_35 cytotype may, at first glance, contradict the hypothesis that polyploids are more efficient in NPQ. In congruence with Achenbach and co-workers [58], ploidy alone does not suffice to explain photosynthesis efficiency data. If, however, the adaption history to forest habitats with a tree canopy and to open meadows is taken into account, the observed patterns make more sense, in which the open habitat-adapted 4x and 6x_35 plants represent the more efficient NPQ types and the forest-dwelling 2x and 6x_29 cytotypes the less efficient NPQ types.
PQ describes the chemistry at the PSII reaction centers. The coefficients q P , and even more so q L , are regarded as the most informative [67]. They quantify the number of open (unreduced) PSII reaction centers. The lower the number, the more efficient PQ becomes. By the majority, and as suggested by q P and q L , the polyploids were able to use PQ better than the diploids. The only exception is cytotype 6x_35, in which PQ is on the same level as in the diploids. The coefficient PQ, corresponding to NPQ, deviated by assigning similar efficiencies to 6x_29 and 6x_35, but is not considered as meaningful as q P and q L .
Light period extension only slightly increased the NPQ capabilities of the 2x cytotype. Generally, in terms of NPQ and PQ, polyploidy contributed more to differences than the light period extension that was applied in this study.

Photosynthesis, Habitat Adaptation, and Apomixis
This study utilized 2x, 4x, and 6x cytotypes from the Ranunculus auricomus species complex. Especially in the more sophisticated fluorescence experiments, which in terms of information quality exceeded simple quantum yield measurements of either light-or dark-adapted plants, the photosynthesis efficiency of polyploids was found to be better adapted in open habitat-conditioned plants, especially in the 4x cytotype.
The second open habitat cytotype, 6x_35, yielded rather controversial results. High value variations in photosynthesis efficiency of polyploids can occur as a result of higher intracytotype variation and niche breadth [57]. In terms of NPQ, 6x_35 resembled the 4x cytotype, in terms of PQ the 2x cytotype. The 6x_35 plants appeared to be more stressed when dark-adapted. With increasing actinic light, the quantum yield was again more similar for the 2x and 6x_29 cytotype and the 4x cytotype showed the lowest values, which may be due to the efficient quenching capabilities, as would be expected for open habitatadapted species. The canopy-adapted 2x and 6x_29 cytotypes were less efficient in terms of NPQ. In terms of PQ, the 6x_29 followed the polyploid trend to be more efficient. The results of this study point more to NPQ than to PQ as an adaptive mechanism to open habitats. Probably, the success of a cytotype depends on the co-ordination extent of both quenching processes. Suboptimal co-ordination, as perhaps in 6x_35, may be responsible for the observed idiosyncratic results.
For the experiments in this study, the plants were grown under equal garden conditions, but different pre-adaptations from original habitats of the accessions may still have influenced their photosynthesis performance. Diploids originated from crosses of forest plants [38] and pre-adaptations to low-light conditions can be recognized in their photosynthesis. They responded to the extended photoperiod more intensively than the polyploids. Tetraploids were raised from light-adapted meadow plants, and the hexaploid clone 35 (=VRU2 in [57]) also originated from a typical open habitat-adapted mother plant from a meadow population with more sunlight exposure [38,68], whereas hexaploid clone 29 (=TRE in [57]) originated from a plant growing in a shaded habitat on a forest margin. Altogether, tetraploids appeared to be better adapted to high-light conditions by their efficient NPQ and PQ. Hexaploids, however, strongly differentiated according to their provenances, although 6x_35 behaved idiosyncratically, but not in terms of NPQ. All cytotypes originate from sites in Central Europe (at 47-48 • latitude and from the same altitudinal zone), and hence we can rule out that differential light intensities due to strong latitude (or altitude) gradients, as observed in a study over the whole of Europe [30], would have influenced the evolution of their photosynthetic performance. The variability of polyploid ecotypes in R. auricomus may rely on a greater variance in gene expression patterns as they have highly heterozygous genomes [30]. In a comparison of diploid and tetraploid Glycine max, overexpression of oxidative stress-regulating genes correlated to differential photosynthetic performance and adaptation to higher light intensities [8]. Additionally, epigenetic control mechanisms may play a role. A study on cytosine methylation of diploid and tetraploid R. kuepferi revealed not only different methylation profiles between cytotypes, but also indicated two different epigenetic groups within tetraploids, correlating with different temperature exposures [69].
The photosynthesis performance of R. auricomus cytotypes, however, does relate to mode of ovule formation, as predominantly sexual diploids showed the highest sensitivity to an extended photoperiod and the lowest quenching capacities concomitant to the highest proportions of sexual ovules. In tetraploids, apomictic ovules prevailed [26]. Hexaploids, however, exhibited a very large variance in the proportions of sexual ovules, in congruence with observations on photosynthesis efficiency [26]. Variation was mostly attributed to two different ecotypes, one adapted to the canopy, the other to open habitats. We assume that differential levels of oxidative stress affect the mode of reproduction, similarly as has been suggested for different ploidy levels in the brassicacean genus Boechera [24].
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/biology10080811/s1, Figure S1, Exemplary Kautsky curve of KC experiments, indication of fluorescence parameters and coefficients; Table S1, Sample and clone codes of plant accessions that were used in the various photosynthesis fluorescence experiments; Table S2, Calculation and definition of photosynthesis coefficients; Table S3, Summary statistics and 95% Tukey multiple range test of φ PSII and φ MAX among cytotypes exposed to different photoperiods; Table S4, Summary statistics and 95% Duncan multiple range test of coefficients from transient fluorescence analyses (OJIP); Table S5; Summary statistics and 95% Duncan multiple range test of relative electron transport rates (rETR) among cytotypes; Table S6; Summary statistics and 95% Duncan multiple range test of coefficients calculated from Kautsky curve experiments.