Male and Female Plants of Salix viminalis Perform Similarly to Flooding in Morphology, Anatomy, and Physiology

Salix viminalis L., a dioecious species, is widely distributed in riparian zones, and flooding is one of the most common abiotic stresses that this species suffers. In this study, we investigated the morphological, anatomical, and physiological responses of male vs. female plants of S. viminalis to flooding. The results showed that the plant height and root collar diameter were stimulated by flooding treatment, which corresponded with higher dry weight of the stem and leaf. However, the dry weight of the underground part decreased, which might be due to the primary root having stopped growing. The little-influenced net photosynthesis rate (Pn) under flooding treatment could guarantee rapid growth of the aboveground part, while the unaffected leaf anatomical structure and photosynthetic pigment contents could ensure the normal operation of photosynthetic apparatus. Under a flooding environment, the production ratio of superoxide free radical (O2·-) and malondialdehyde (MDA) contents increased, indicating that the cell membrane was damaged and oxidative stress was induced. At the same time, the antioxidant enzyme system, including superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), and ascorbate peroxidase (APX), and osmotic adjustment substances, involving proline (Pro) and solute protein (SP), began to play a positive role in resisting flooding stress. Different from our expectation, the male and female plants of S. viminalis performed similarly under flooding, and no significant differences were discovered. The results indicate that both male and female plants of S. viminalis are tolerant to flooding. Thus, both male and female plants of S. viminalis could be planted in frequent flooding zones.


Introduction
Flooding is one of the primary abiotic stresses encountered by many plants, and it develops when the water content in the soil surpasses the field moisture capacity [1]. Soil flooding is usually caused by James) are more tolerant to flooding than males, which could be related to the greater frequency of females in flood-prone sites [39,40]. Salix viminalis L., a species of Salix, Salicaceae, is predominant in riparian zones, and the female plants are more prevalent in wetter zones that are prone to suffer from flooding. However, whether the male and female plants of S. viminalis perform differently (i.e., the females are more tolerant to flooding than males) under flooding has not been documented. Although there has been research to examine the responses of S. viminalis to soil flooding [41], no studies have been conducted to discuss possible gender differentiation of S. viminalis under flooding. In this study, we investigated the gender-specific responses of S. viminalis in morphology, anatomy, growth, gas exchange, chlorophyll pigment, oxidative stress degree, osmotic adjustment, and enzymatic antioxidants to flooding stress. The objectives of this study were to: (1) explore how flooding affects the morphological, anatomical, and physiological processes of dioecious S. viminalis, and (2) assess whether differences exist between male and female plants of S. viminalis under flooding stress.

Plant Materials and Experiment Design
A male parent of S. viminalis from Heilihe nature reserve and a female parent from Saihanwula nature reserve were used for controlled intraspecific hybridization. In the F1 progeny containing 570 seedlings, 10 male and 10 female individuals were selected randomly. One-year-old branches were harvested to propagate cuttings from each individual on March 29th, 2018. The cuttings with a length of about 10 cm were grown in nutritive bowls filled with peat soil. Then, five male and five female clones were replanted in 16 cm diameter × 15 cm deep plastic pots on May 12th and the substrate was a 2:1 mixture of garden soil to peat soil. The plants were placed in a greenhouse of the Chinese Academy of Forestry Sciences in Beijing and were watered regularly until flooding application.
On June 10th, plants of similar height were selected for flooding treatment. The experimental layout was a completely randomized block design with two main factors (gender and watering regime). Two watering regimes were applied: well-watered treatment as control and flooding treatment. In the control, the plants were watered every day to maintain soil moisture at field capacity. In the flooding treatment, the pots were placed into a larger plastic bucket filled with tap water to 4 cm above the soil surface and watered every day to maintain the water level. During the flooding treatment, the morphological, photosynthetic, and physiological parameters were investigated at 7, 14, 21, 30, and 45 days. For the physiological parameters, the leaves were cut into pieces and mixed, put into the liquid nitrogen, and then stored at -80 • C. There were five blocks in this research, and the experiment was conducted during the gowning season from June 10th to July 25th, 2018. During this period, the precipitation is large and flooding events are more prone to occur.

Growth Characteristics (Height, Diameter, and Biomass)
The height and root collar diameter of each individual plant were measured at 0, 7, 14, 21, 30, and 45 days. The plant height was measured by a measuring tape with a precision of 0.1 cm, and the root collar diameter was evaluated by Vernier caliper with an accuracy of 0.01 cm. At the end of the experiment, the root, stem, and leaf were harvested and each component was dried to a constant weight and weighed, respectively.

Leaf Gas Exchange Parameters
Gas exchange parameters were measured in the 3rd~5th fully expanded and mature leaf from the top of the stem between 9:00 and 12:00 in the morning using the Li-Cor 6400XT photosynthesis measuring system (LI-COR, Lincoln, NE, USA) equipped with a red-blue light-emitting diode light source. The parameters included net photosynthesis (P n ), stomatal conductance (G s ), internal CO 2 concentrations (C i ), and transpiration rate (T r ). The optimal parameters were set as follows:

Chlorophyll Pigment
About 0.20 g of leaf tissue, 0.2 × 0.2 cm in size, was immersed into 80% acetone (v:v) for 24 h, approximately under room temperature until all of the tissue turned white. Meanwhile, the tube was shaken at intervals. Then, the absorbance of the extracting solution was measured at 470, 646, and 663 nm. The content of chlorophyll a (Chl a), chlorophyll b (Chl b), and carotenoid (Caro) were calculated by the equation described by Lichtenthaler [42].

Proline Content
Pro was extracted and measured according to the procedure by Bates et al. [43]. About 0.30 g of leaf tissue was immersed into 5 ml of 3% aqueous sulfosalicylic acid solution. After being extracted in a boiling water bath and filtered, 2.0 ml of glacial acetic acid and 2 ml of 2.5% acid ninhydrin were added into 2 ml of the supernatant. Then, the mixed solution was incubated in a boiling water bath. After the mixture was cooled, 5 ml of toluene was added and the absorbance was measured at 520 nm.

Production Ratio of Superoxide Free Radical, Malonaldehyde Content, Solute Proteins Content, and Autioxidant Enzyme Activities Determination
The frozen leaves were used to test the physiological indicators, including the production ratio of O 2 ·-, the MDA content, the solute proteins (SP) content, the SOD activity, the POD activity, the CAT activity, and the APX activity. All of these indicators were analyzed by applying the respective assay kit (Comin Biotechnology Co. Ltd., Suzhou, China).

Anatomical Structure Observation
The 5th fully expanded and mature leaf from the top of the stem was used for anatomical analysis. The same parts of the leaf were cut into 2 mm strips perpendicular to the main vein and immediately fixed in formalin, acetic acid, and alcohol solution (FAA) solution (70% alcohol/glacial acetic acid/40% formalin = 90:5:5, v/v) and stored at 4 • C. Then, the materials were, respectively, dehydrated with gradient ethanol of 75% (4 h), 85% (2 h), 90% (2 h), 95% (1 h), and 100% concentrations (two changes, 30 min). Next, they were immersed in mixed liquor with ethanol and xylene (10 min) and xylene (two changes, 10 min). Thereafter, the samples were embedded in paraffin by the Embedding Center (JB-P5, Wuhan Junjie Electronic Co., Ltd.), and subsequently crosscut into 4-µm-thick slices by a microtome (RM2016, Shanghai Leica Instrument Co. Ltd.). After dyeing in saffron for 2 h and solid green for 40 s, the cross-sections were observed and photographed using a NIKON ECLIPSE E100 microscope (Nikon Corporation, Japan).

Statistical Analysis
All data were analyzed by SPSS 18.0 software (SPSS Inc., Chicago, IL, USA). Before the ANOVAs, the data were checked for normality, as well as the homogeneity of variances using Shapiro-Wilk test and Levene's test, respectively. Individual differences among the means for each investigated time and each part of plant were identified by Duncan tests of one-way ANOVAs at a significance level of p < 0.05. Two-way ANOVAs were used to test the effects of gender, flooding, and their interaction, which was carried out using the multivariate General Linear Model procedure with Type III sum of squares.

Gender-Specific Responses of Morphology to Flooding
All of the plants grew vigorously and no withering, defoliation, or death occurred during the study. At the end of experiment, about 3~5 leaves at the bottom of the plant under the flooding Forests 2020, 11, 321 5 of 16 treatment turned yellow. White hypertrophied lenticels and adventitious roots developed at the base of the submerged shoot after 3 and 5 days of flooding treatment, respectively. However, there were no obvious differences between male and female plants.

Gender-Specific Responses of Plant Growth to Flooding
S. viminalis plants continued to grow for both the control and flooding treatment groups during the experiment (Figure 1). Flooding did not significantly affect the plant height or root collar diameter before the first 21 days, but the two parameters for flooding treatment were significantly higher than the control at 30 and 45 days. However, the plant height and root collar diameter between male and female plants showed no obvious differences throughout the experiment ( Figure 1). Besides, the growth parameters were not significantly affected by gender × flooding interaction. of the submerged shoot after 3 and 5 days of flooding treatment, respectively. However, there were no obvious differences between male and female plants. S. viminalis plants continued to grow for both the control and flooding treatment groups during the experiment (Figure 1). Flooding did not significantly affect the plant height or root collar diameter before the first 21 days, but the two parameters for flooding treatment were significantly higher than the control at 30 and 45 days. However, the plant height and root collar diameter between male and female plants showed no obvious differences throughout the experiment ( Figure 1). Besides, the growth parameters were not significantly affected by gender × flooding interaction. At the end of the flooding treatment, we found that the underground biomass of male and female plants was significantly decreased, including primary root and total root, in spite of the emergence of adventitious root, while the aboveground biomass (stem and leaf) was significantly increased ( Figure 2). However, the total dry weight presented no significant differences between the control and the flooding treatment. Besides, the dry weight of each constituent and sapling had no significant differences between male and female plants. In addition, the dry weight was not At the end of the flooding treatment, we found that the underground biomass of male and female plants was significantly decreased, including primary root and total root, in spite of the emergence of adventitious root, while the aboveground biomass (stem and leaf) was significantly increased ( Figure 2). However, the total dry weight presented no significant differences between the control and the flooding treatment. Besides, the dry weight of each constituent and sapling had no significant differences between male and female plants. In addition, the dry weight was not significantly affected by gender × flooding interaction. of the submerged shoot after 3 and 5 days of flooding treatment, respectively. However, there were no obvious differences between male and female plants. S. viminalis plants continued to grow for both the control and flooding treatment groups during the experiment (Figure 1). Flooding did not significantly affect the plant height or root collar diameter before the first 21 days, but the two parameters for flooding treatment were significantly higher than the control at 30 and 45 days. However, the plant height and root collar diameter between male and female plants showed no obvious differences throughout the experiment ( Figure 1). Besides, the growth parameters were not significantly affected by gender × flooding interaction. At the end of the flooding treatment, we found that the underground biomass of male and female plants was significantly decreased, including primary root and total root, in spite of the emergence of adventitious root, while the aboveground biomass (stem and leaf) was significantly increased ( Figure 2). However, the total dry weight presented no significant differences between the control and the flooding treatment. Besides, the dry weight of each constituent and sapling had no

Gender-Specific Responses of Leaf Anatomical Structure to Flooding
The male and female plants displayed similar cross-section structures, including one-layer upper and lower epidermis, bi-layered compact palisade parenchyma, 2-3-cell-thick loose spongy parenchyma and transversal vascular bundles, etc. (Figure 3). After flooding treatment for 45 days, the anatomical structure of males and females were not affected significantly and palisade cells were still arranged regularly ( Figure 3). The male and female plants displayed similar cross-section structures, including one-layer upper and lower epidermis, bi-layered compact palisade parenchyma, 2-3-cell-thick loose spongy parenchyma and transversal vascular bundles, etc. (Figure 3). After flooding treatment for 45 days, the anatomical structure of males and females were not affected significantly and palisade cells were still arranged regularly (Figure 3).

Gender-Specific Responses of Gas Exchange Parameters to Flooding
The gas exchange parameters, including P n , g s , C i , and E, were not significantly affected by flooding throughout the experiment, although they fluctuated over time ( Figure 4). The male and female plants of S. viminalis also presented no obvious differences. In addition, the gas exchange parameters were not significantly affected by gender × flooding interaction.
upper and lower epidermis, bi-layered compact palisade parenchyma, 2-3-cell-thick loose spongy parenchyma and transversal vascular bundles, etc. (Figure 3). After flooding treatment for 45 days, the anatomical structure of males and females were not affected significantly and palisade cells were still arranged regularly (Figure 3).   The gas exchange parameters, including Pn, gs, Ci, and E, were not significantly affected by flooding throughout the experiment, although they fluctuated over time ( Figure 4). The male and female plants of S. viminalis also presented no obvious differences. In addition, the gas exchange parameters were not significantly affected by gender × flooding interaction. The contents of Chl a, Chl b, and Car of S. viminalis plants were not significantly influenced by flooding treatment, and the three pigments also presented no significant difference between male and female plants ( Figure 5). In addition, chlorophyll pigments were not significantly affected by

Gender-Specific Responses of Chlorophyll Pigments to Flooding
The contents of Chl a, Chl b, and Car of S. viminalis plants were not significantly influenced by flooding treatment, and the three pigments also presented no significant difference between male and female plants ( Figure 5). In addition, chlorophyll pigments were not significantly affected by gender × flooding interaction. The gas exchange parameters, including Pn, gs, Ci, and E, were not significantly affected by flooding throughout the experiment, although they fluctuated over time (Figure 4). The male and female plants of S. viminalis also presented no obvious differences. In addition, the gas exchange parameters were not significantly affected by gender × flooding interaction. The contents of Chl a, Chl b, and Car of S. viminalis plants were not significantly influenced by flooding treatment, and the three pigments also presented no significant difference between male and female plants ( Figure 5). In addition, chlorophyll pigments were not significantly affected by gender × flooding interaction.

Gender-Specific Responses of Superoxide Free Radical and Lipid Peroxidation to Flooding
Flooding treatment led to increases in the production ratio of O 2 ·and MDA contents of S. viminalis ( Figure 6). The production ratio of O 2 ·under flooding treatment was significantly higher than that of the control after 14 days, while MDA contents increased significantly after 30 days of flooding treatment. However, both the production ratios of O 2 ·and MDA contents between male and female plants showed no significant differences. In addition, the two parameters were not significantly affected by gender × flooding interaction.
Forests 2020, 11, x FOR PEER REVIEW 8 of 16 Flooding treatment led to increases in the production ratio of O2 •-and MDA contents of S. viminalis ( Figure 6). The production ratio of O2 •-under flooding treatment was significantly higher than that of the control after 14 days, while MDA contents increased significantly after 30 days of flooding treatment. However, both the production ratios of O2 •-and MDA contents between male and female plants showed no significant differences. In addition, the two parameters were not significantly affected by gender × flooding interaction. Flooding treatment caused significant increases of SOD, POD, CAT, and APX activities in both male and female plants after 14 days, and the four enzyme activities were on the rise alongside the

Gender-Specific Responses of Enzyme Activities to Flooding
Flooding treatment caused significant increases of SOD, POD, CAT, and APX activities in both male and female plants after 14 days, and the four enzyme activities were on the rise alongside the flooding treatment time (Figure 7). However, there were no significant differences between males and females of S. viminalis, neither in the control nor the flooding treatment. Besides, the enzyme activities were not significantly affected by gender × flooding interaction. Flooding treatment led to increases in the production ratio of O2 •-and MDA contents of S. viminalis ( Figure 6). The production ratio of O2 •-under flooding treatment was significantly higher than that of the control after 14 days, while MDA contents increased significantly after 30 days of flooding treatment. However, both the production ratios of O2 •-and MDA contents between male and female plants showed no significant differences. In addition, the two parameters were not significantly affected by gender × flooding interaction. Flooding treatment caused significant increases of SOD, POD, CAT, and APX activities in both male and female plants after 14 days, and the four enzyme activities were on the rise alongside the flooding treatment time (Figure 7). However, there were no significant differences between males

Gender-Specific Responses of Osmotic Regulation to Flooding
Compared with the control, the flooding treatment increased Pro and SP contents significantly in S. viminalis, but there were no obvious differences between male and female plants (Figure 8). The Pro contents increased continually as time went on, while the SP contents presented a fluctuating variation throughout the experiment. Furthermore, the Pro and SP contents were not significantly affected by gender × flooding interaction.
Forests 2020, 11, x FOR PEER REVIEW 9 of 16 and females of S. viminalis, neither in the control nor the flooding treatment. Besides, the enzyme activities were not significantly affected by gender × flooding interaction. Compared with the control, the flooding treatment increased Pro and SP contents significantly in S. viminalis, but there were no obvious differences between male and female plants (Figure 8). The Pro contents increased continually as time went on, while the SP contents presented a fluctuating variation throughout the experiment. Furthermore, the Pro and SP contents were not significantly affected by gender × flooding interaction.

Morphology, Growth, and Anatomical Structure
S. viminalis is a species of obligated riparian trees that are often distributed along streams and rivers and other wet areas [44]. In this research, we studied the response of males vs. females of S. viminalis in an artificial flooding treatment with water up to 4 cm above the soil surface. Hypertrophied lenticels and adventitious roots were induced after 3 days of flooding treatment, which are specific changes of flood-tolerant species [1,3]. The hypertrophied lenticels can contribute to air exchange between the stem and root of a plant and the atmosphere [1,45], while the adventitious roots can replace some functions of older damaged roots and help plants to uptake water and nutrients normally [24,46]. Symptoms of chlorosis and defoliation of leaves can also be induced by flooding treatment [1,47]. However, only yellow leaves of the flooded plants were observed in our study, which might indicate that S. viminalis is slightly damaged by long-time flooding treatment.
Flooding can bring about a reduction of plant height and root collar diameter, and the biomass of the leaf, stem, and root can also decrease [13,39,48,49]. In our study, flooding irritated the growth of the aboveground part, accompanied by taller plant height and thicker root collar diameter, which was consistent with the larger dry weight of the leaf and stem. However, the growth of the underground part was suppressed, along with a lighter dry weight of the root. Our results concerning morphology and biomass are consistent with previous findings of purpul osier (Salix integra Thunb.), Salix variegate Franch, and marsh bluegrass (Poa leptocoma Trin.) [38,50,51], which indicate that flooding has a negative effect on the root system, while it promotes the growth of the aerial part. The elongation of the stem might be an effective strategy for S. viminalis under our experimental conditions with a shallow water level, which might contribute to escaping from severe flooding [52][53][54]. In this study, morphological changes, including the emergence of hypertrophied lenticels and adventitious roots and accelerated aboveground growth, were conducive to adapting riparian zones for S. viminalis.

Morphology, Growth, and Anatomical Structure
S. viminalis is a species of obligated riparian trees that are often distributed along streams and rivers and other wet areas [44]. In this research, we studied the response of males vs. females of S. viminalis in an artificial flooding treatment with water up to 4 cm above the soil surface. Hypertrophied lenticels and adventitious roots were induced after 3 days of flooding treatment, which are specific changes of flood-tolerant species [1,3]. The hypertrophied lenticels can contribute to air exchange between the stem and root of a plant and the atmosphere [1,45], while the adventitious roots can replace some functions of older damaged roots and help plants to uptake water and nutrients normally [24,46]. Symptoms of chlorosis and defoliation of leaves can also be induced by flooding treatment [1,47]. However, only yellow leaves of the flooded plants were observed in our study, which might indicate that S. viminalis is slightly damaged by long-time flooding treatment.
Flooding can bring about a reduction of plant height and root collar diameter, and the biomass of the leaf, stem, and root can also decrease [13,39,48,49]. In our study, flooding irritated the growth of the aboveground part, accompanied by taller plant height and thicker root collar diameter, which was consistent with the larger dry weight of the leaf and stem. However, the growth of the underground part was suppressed, along with a lighter dry weight of the root. Our results concerning morphology and biomass are consistent with previous findings of purpul osier (Salix integra Thunb.), Salix variegate Franch, and marsh bluegrass (Poa leptocoma Trin.) [38,50,51], which indicate that flooding has a negative effect on the root system, while it promotes the growth of the aerial part. The elongation of the stem might be an effective strategy for S. viminalis under our experimental conditions with a shallow water level, which might contribute to escaping from severe flooding [52][53][54]. In this study, morphological changes, including the emergence of hypertrophied lenticels and adventitious roots and accelerated aboveground growth, were conducive to adapting riparian zones for S. viminalis.
Anatomical characteristics are also important in improving the flooding tolerance of plants [55]. A previous study found that flooding-susceptible poplar displays an unstable anatomical structure in which the shape of palisade cells turns from long columns into circles [47]. However, the flooding-tolerant species still own regular palisade cells and are characterized by the presence of aerenchymatous tissue [47,56]. In our study, the anatomical structure of both male and female S. viminalis was not influenced by flooding, which could ensure the stability of the photosynthetic structure. However, aerenchymatous tissue was not observed in the leaves of S. viminalis, which might be due to the fact that CO 2 could diffuse normally and so the formation aerenchymatous tissue was not initiated.

Photosynthesis
The photosynthetic rate generally decreases for non-flooding-tolerant species during the flooding period, and the earliest response is stomatal closure [15,49]. However, flooding-tolerant species can maintain a high level or even an unaffected photosynthetic rate [14,57]. Although the P n of S. viminalis fluctuated over time, it was not significantly affected by flooding. The photosynthetic response of S. viminalis was in accordance with species of the same genus, such as S. integra cv. qingpi [1]. Usually, g s has a positive correlation with P n [58], which also applied to this study. The high g s may increase gas exchange rates, as well as the total volumes, and hence plants assimilate more photosynthates for growth [59]. Also, the appearance of hypertrophied lenticels and adventitious roots in a short time is conducive to keeping the stoma open and maintaining an unaltered photosynthetic rate [60]. The pigment concentrations in the leaves of plants are closely associated with photosynthesis. Many previous studies have found that the Chl a, Chl b, and Car contents usually decline under flooding conditions, along with a reduction of photosynthesis [15,49,61]. However, the Chl a, Chl b, and Car contents in this study were unchanged, which could ensure the normal operation of photosynthesis. The observation of photosynthesis and chlorophyll indicate that S. viminalis were tolerant to flooding.

Oxidative Stress
Plants exposed to flooding conditions usually accumulate ROS, such as O 2 ·- [62]. The ROS react with unsaturated fatty acids, which results in destruction of the membranes, while MDA is the final production of membrane lipid peroxidation [62,63]. We observed an enhanced production ratio of O 2

·-
and MDA contents for both male and female S. viminalis plants under flooding treatment; the former and the later showed significant increases after 14 days and 30 days of treatment, respectively. Similar results have been shown in previous studies [19,64], and usually, the production ratios of O 2 ·and MDA contents are positively correlated with flooding treatment time [64,65]. The increased production ratios of O 2 ·and MDA contents can damage the membrane lipids and plants cells, break the cytomembrane structure, and cause oxidative stress [62,64,66]. Our results indicate that the flooding could induce oxidative stress and cause damage to the leaf cells of S. viminalis.

Antioxidant System
Plants have developed an antioxidant system, including a series of enzymes (SOD, POD, CAT, etc.) and substances to defend or alleviate the detrimental effect of ROS by metabolism under abiotic stresses [64]. Usually, the antioxidant enzyme activities are positively correlated with the self-protective ability of plants [67]. SOD is the first line of defense and it catalyzes the dismutation of O 2 ·to O 2 and H 2 O 2 timely and effectively in the cytosol, chloroplasts, and mitochondria [34]. SOD activity significantly increased after 14 days of flooding treatment in this study, suggesting that the detoxification of O 2 ·in S. viminalis is effective. POD, CAT, and APX can catalyze the decomposition of H 2 O 2 to water and oxygen, eventually clearing up H 2 O 2 in cells [23,34,64]. The three enzymes increased significantly in our study, which contributed to neutralizing the product H 2 O 2 from the previous SOD. Similar result has been obtained in açai berry (Euterpe oleracea Mart.) [62] treated with flooding. The enhanced activities of antioxidant enzymes could increase the antioxidant ability of S. viminalis, and thus were helpful to reduce oxidative stress.

Osmotic Substances
Osmotic adjustment has been regarded as one of the most crucial protective mechanisms for plants to adapt to stressful environments, while Pro and SP are major osmotic substances [16,34]. Pro can act as a cell protector and reducer of osmotic potential, which is involved in many cellular processes to keep a balance of ROS [19]. SP is also an important osmotic protective substance that plays a role in regulating osmotic potential [68]. Previous studies have found that flooding significantly increases Pro content in the leaves of pigeonpea (Cajanus cajan L. Millsp.) [69] and bambara groundnut (Vigna subterranea L. Verdc) [70], while accumulated SP has been discovered in Acorus tatarinowii Schott under flooding conditions [71]. Accumulated Pro and SP contents were observed in our study, indicating that osmotic adjustment was irritated and acted as an effective way for S. viminalis to respond to flooding. However, there is some dispute about the accumulation of Pro. An et al. [21] found that Pro accumulation does not contribute to maintaining water balance and it may be an indicator of injury for fig leaves under flooding condition.

Gender-Specific Responses
It is generally assumed that differences exist in males and females of dioecious plants due to different resource allocation, i.e., that females apply more resources in defense while males invest more resources in growth [72,73]. Gender differences usually appear or are greater under adverse conditions [74], and such differences exist in many plants, such as Populus cathayana [35], European aspen (Populus tremula L.) [75], and grey willow (Salix glauca L.) [76]. Gender-specific differences in flooding resistance vary in plant species, even when the plants are in the same genus. A previous study showed that males of eastern cottonwood (Populus deltoides Marsh) have better cellular defense mechanisms against damage caused by waterlogging stress, whereas females are more responsive to waterlogging stress [48]. In S. variegate, the males can adjust more flexibly their resource allocation, and so they are more tolerant to flooding [77]. However, the female plants of P. angustifolia are more tolerant to flooding, which is strengthened by the increased occurrence of female cottonwoods in streamside zones [39,40]. The favored occurrence of female plants in wetter sites is also apparent in some willows, and it is observed that female willows grow more vigorously than males [78,79].
In our study, flooding treatment caused morphological and physiological changes in S. viminalis, but male and female plants performed similarly and no significant gender-specific differences were discovered. This result is different from previous studies on gender differences of S. viminalis, which found that female plants may be more severely affected by fungal pathogens and high temperatures than males [80,81]. However, gender differences of S. myrsinifolia are not much affected by simulated climatic changes (enhanced CO 2 and temperature or their combination) at the pre-reproductive stage [73], and no specific difference was found in P. angustifolia in response to seasonal changes in water availability [82]. In our study, no obvious differences were found between the two genders, suggesting that similar patterns of morphology, anatomy, and physiology may be adopted by the two genders to co-exist in riparian zones, which is consistent with two dioecious riparian shrub species, namely, Salix myrsinifolia and Salix lapponum [83]. Besides, all plants survived 45 days of flooding with no leaf abscission, fast growth aboveground, unaffected photosynthesis, and strong antioxidant capacity and osmotic adjustment ability, which suggests that both male and female plants of S. viminalis are flood-tolerant. Thus, it is concluded that the greater frequency of females in riparian zones is not influenced by flooding, and other factors, such as demographic parameters, could also influence the spatial distribution of genders [84].To better understand the response of male and female S. viminalis to flooding, further studies at the level of molecular biology are needed.

Conclusions
The combination of morphological, anatomical, and physiological parameters can help us to better understand the gender-specific responses of S. viminalis to flooding. The present study suggests that the growth of plant height and root collar diameter is stimulated by flooding treatment, accompanied by increased dry weight of the aboveground part. The little-affected photosynthesis could guarantee the rapid growth of plants, while the unaffected leaf anatomical structure and photosynthetic pigment contents could ensure the normal operation of photosynthetic apparatus. However, the dry weight of the underground part decreased at the end of the experiment, which might be owing to the fact that the primary root stopped growing. The flooding treatment caused oxidative stress in S. viminalis, but they could resist the stress by enhancing antioxidant enzyme activities and osmotic adjustment ability. Different from our expectation, gender-specific differences in S. viminalis were not obviously detected, and the male and female plants performed similarly to flooding. Comprehensive indicators indicate that both male and female plants of S. viminalis are tolerant to flooding, which would guarantee a sustainable population and maintain a stable riparian ecosystem. Thus, both male and female plants of S. viminalis could be planted in frequent flooding zones.