Effects of ciprofloxacin on turion germination and seedling development in two submerged aquatic plants


 Germination and seedling development are crucial processes for plant growth and survival, and asexual propagules are predominant reproduction organs for aquatic plants. Ciprofloxacin is widely detected in both terrestrial and aquatic ecosystems, while studies on its effects on germination and seedling development mainly focused on terrestrial plants. We evaluated effects of ciprofloxacin (0.1, 1, 5 and 10 mg/L) on turion germination and early establishment of two submerged plants species (Potamogeton crispus L and Hydrilla verticillata Royle). Results showed that ciprofloxacin did not impact germination rate and rooting rate of both species. However, 0.1 mg/L ciprofloxacin significantly accelerated germination of H. verticillata while 5-10 mg/L ciprofloxacin significantly delayed rooting process of both species. With rising ciprofloxacin concentrations and prolonging exposure time, seedling tolerance index and root number of both species decreased significantly, and shoot number decreased slightly in P. crispus but kept increasing in H. verticillata, suggesting better tolerance of the later under ciprofloxacin exposure. Root and shoot biomass accumulation of both species was significantly inhibited, which was partially due to ciprofloxacin toxicity on photosynthetic pigments. By integrating the biomarkers including plant antioxidants, lipid peroxidation degree and hydrogen peroxide contents, we found that the holistically toxicological effects of ciprofloxacin on seedlings of both species were enhanced with increasing ciprofloxacin concentrations. Overall, ciprofloxacin impacted turion germination process and harmed early establishment of these two submerged plants, which might suggest adverse effects of ciprofloxacin on the survival and expansion of submerged aquatic plant populations and their restoration effectiveness for degraded aquatic vegetations.


Introduction
Submerged plants, the main primary producers in shallow fresh waters, could provide habitats, shelters and food sources for shes, aquatic macroinvertebrates and waterbirds, contributing to their important roles in the maintenance of structure and function of shallow-water ecosystems ( (Phillips et al. 2016). For instance, in eutrophic water bodies, toxic microcystins produced by massive cyanobacterial blooms would deteriorate water quality and seriously threaten the growth and survival of submerged plants (Ha and P ugmacher 2013). Additionally, as emerging pollutants, uoroquinolone antibiotics have become one of the most common categories of harmful pollutants in lakes (Chen et al. 2020), rivers (Pan et al. 2020) and even seawaters (Wu et al. 2022). Polluted waters containing uoroquinolone antibiotics could impact aquatic community structure and might act as possible causes of the decline and disappearance of aquatic plants (Robinson et al. 2005).
Cipro oxacin is a uoroquinolone antibiotic that has been widely used in aquaculture, treatments of human ailments and pharmaceutical industries. Consequently, agricultural wastewater, hospital e uent, domestic sewage and industrial emissions are the main sources of cipro oxacin pollution in natural water bodies (Adeleye et al. 2022). For example, cipro oxacin (mean concentration, detected frequency) was detected in drinking water (169.2 ng/L, 94.9%), surface water (28 ng/L, 100%) and ground water (77.2 ng/L, 45%) (Boy-Roura et al. 2018; Pan et al. 2020; Chen et al. 2018). Enormously high concentrations of cipro oxacin in e uents of wastewater treatment plants (14 mg/L) and lakes (6.5 mg/L) have also been reported by Fick et al (2009). In estuarine water and seawater of Laizhou Bay in northern China, cipro oxacin was one of the most dominant antibiotics which contributed to over 70% of the total antibiotic burden (Lu et al. 2022). Moreover, by inducing antibiotic resistance ), cipro oxacin could cause adverse effects on human health at even low concentrations (ng/L) (Koczura et al. 2012;Marchant 2018). Therefore, cipro oxacin has been proposed to be prioritized and strictly controlled to reduce its negative effects on food chains and ecological systems (Han et al. 2020).
Generally, environmental stresses could impact vegetative growth and physiological stability of aquatic plants (Bornette and Puijalon 2010; Yu et al. 2022). For example, cipro oxacin (0.01-1mg/L) signi cantly decreased the photosynthetic pigment contents, disrupted PSII integrity, affected leave growth and triggered ROS accumulations and oxidative stress in a oating plant Eichhornia crassipes (Yan et al. 2019). Root developments, chlorophyll contents, SOD and POD activities and root activity of the emergent plant species Phragmites australis were all evidently inhibited by a mixture of cipro oxacin, oxytetracycline and sulfamethazine (Liu et al. 2013). As for submerged plants, Ebert et al (2011) found the 7-day NOEC (the no-observed-effect concentration) of cipro oxacin based on root elongation of Myriophyllum spicatum L was higher than 0.98 mg/L. These studies mainly focused on vegetative growth stage. However, researches on other life history stages such as germination and reproduction are limited. As an example, it has been reported that cipro oxacin at 0.2-2 mg/L did not affect the germination rate of maize seeds but signi cantly reduced the germination time (Gomes et al. 2019). But no investigations on the effects of cipro oxacin on germination phase of aquatic species have been reported.
Vegetative propagation is one of the main modes of reproduction for submerged plants, and turions of submerged plants are modi ed vegetative organs produced via asexual processes (Adamec 2018). With functional similarity to seeds, turions also have the ability to generate progeny plants . What is more, due to their tolerance to extreme conditions such as coldness, darkness and hypoxiation of the underwater environments, turions play key roles in maintaining submerged plant community (Adamec 2008). Nowadays, in China, sowing turions of submerged plants have become one of the most common practices for restoration of damaged submerged vegetation in lakes and ponds (Jian et al. 2003;. Undoubtedly, germination of turions and early establishment are crucial stages that could directly in uence the biomass accumulations and population expansion of submerged plants. However, ecotoxicological effects of cipro oxacin on turion germination and seedling development of submerged plants have not been reported. Hydrilla verticillata is a dominant submerged species in many freshwater lakes, while Potamogetom crispus could be widely found in shallow lakes, ponds, rice paddies and rivers (Jian et al. 2003;Li et al. 2021). Additionally, due to their good performances in purifying water, these two submerged plants have been widely used as pioneer species in restoration of submerged plants for governing eutrophication (Li et al. 2021; Wang et al. 2017). In the present study, we investigate effects of cipro oxacin on turions germination, root development, seedling tolerance, and photosynthetic pigment contents, H 2 O 2 content, lipid peroxidation degree, seven antioxidants of P. crispus and H. verticillata. Our aims were to determine how the germination and early establishment of these two species were impacted by cipro oxacin, which might help to predict population dynamics and provide theoretical guidance for restoration of degraded shallow waters polluted by cipro oxacin.

Materials And Methods
2.1 Plant materials and experimental design P. crispus turions (green turions, weighing 0.68 ± 0.12 g, n=250) were hand collected in September 2020 from Lake Diaocha (30°69'N; 113°72'E) in Hubei Province, PR China. After cleaning thoroughly, the turions were transported into a climate chamber on 20 September. Turions of H. verticillata (weighing 0.15 ± 0.01 g, n=250) were hand collected in January 2021 from Lake Diaocha. After cleaning thoroughly, the turions were transported into a climate chamber on 5 January.
During the whole experiment, the turions/seedlings of P. crispus or H. verticillata were cultivated in 10% Hoagland's solution in a transparent container (Hoagland and Arnon 1950) at 15 ± 2°C or 25 ± 2°C with a 12/12 light/dark cycle and a photon ux density of 30-40 µmol photons m −2 s −1 . The turions/seedlings were exposed to various concentrations of cipro oxacin (0, 0.1, 1, 5 and 10 mg/L) and turions (seedlings) and nutrient solution without cipro oxacin was set as the control group. The whole observation lasted for 35 days (d), where turions were kept in a solution volume of 2 L during germination (0-7d), and seedlings (8-35d) were kept in 5 L solution to meet enough space for their growth. Germination experiment was performed in ve replicates with each replicate containing 10 uniform turions. At 8d, uniform seedlings from different treatments were respectively selected for further investigation, where each treatment had ve replicates and each replicate contained ve seedlings. To maintain approximately constant concentrations of cipro oxacin and nutrients, they were changed every 48 hours during the whole observation. Germination parameters were investigated in the rst 7d, and rooting parameters were investigated at 6-14d. Seedling growth and tolerance indicators of seedlings were investigated at 7, 14, 21, 28 and 35d respectively. Plant biomass and physiological responses of seedlings were analyzed at 35d.

Cipro oxacin concentrations analysis
The actual concentrations of cipro oxacin were determined 1h after being added into the experimental containers, which were 0.1135 ± 0.0022, 1.0483 ± 0.0047, 4.5034 ± 0.0484 and 9.7506 ± 0.0868 mg/L (n=3) for the nominal concentrations of 0.1, 1, 5 and 10 mg/L, respectively. Cipro oxacin concentrations were analyzed by high performance liquid chromatography (HPLC, Agilent1100, USA). The mobile phase was 0.025 mol/L phosphoric acid (pH=2.5, regulated by triethylamine) and acetonitrile, where the volume ratio of phosphoric acid and acetonitrile was 83:17, using C18 column (Agilent, 20RBAX SB-C18, 4.6 mm×250 mm, 5 µm particle size), with a ow rate of 1.0 ml/min and excitation/emission wavelength at 278/453 nm. The limit of detection was 0.1 µg/L. External standards were used for quantitative determination. The calibration curves showed good linearity for the analyte, with correlation coe cients of 0.9994.

Estimation of germination and rooting
P. crispus turions were considered germinated when 5 mm of shoot had emerged, and the turions of H. verticillata were considered germinated once evident elongation (≥ 1cm) of the buds (internodes) had emerged. Germination rate (percentage) and germination time were documented in the rst 7d. For both species, rooting rate (percentage) and rooting time were recorded when 5 mm of radicle (root) had emerged. Both germination rate and rooting rate were displayed as x percentage (%) of the control. Germination time and rooting time were calculated according to Ellis and Roberts (1981): Where D was the number of days since the beginning of the experiment, n was the number of germinated/rooted turions on day D.

Seedling growth, tolerance indicators and IC50 assessment
Shoot number and root number were used to assess growth status of both P. crispus and H. verticillata seedlings under cipro oxacin treatment. Due to the structural differences between these two species, the shoot of H. verticillata referred to the stolon branch in this study. Root tolerance index (RTI) and shoot tolerance index (STI) were used to assess seedling tolerance to cipro oxacin exposure. RTI and STI were Inhibition rate (IR, %) based on RTI and STI was calculated according to the following equation: IR = (µ 0 -µ t ) / µ 0 × 100, Where µ 0 was the value of the control and µ t was the corresponding value of treatment groups at the same sampling date. To acquire cipro oxacin effect on seedling tolerance as time elapsed, regression models between the IR values and cipro oxacin concentrations were built at 7, 14, 21, 28 and 35d using liner regression techniques performed in R software (version 4. 0. 4), and the inhibitory concentrations (IC50, inhibitory concentration causing 50% of an endpoint) with a 95% con dence interval (95% CI) were calculated. Regression models and details were shown in supplementary information.

Plant biomass
At 35d, after blotting water with lter paper, the shoot biomass and root biomass of both P. crispus and H. verticillata seedlings were estimated by an electronic analytical balance (accuracy 0.1 mg), where the biomasses were fresh weights (FW).

Photosynthetic pigment measurement
The contents (mg/g FW) of photosynthetic pigment were determined based on the method of Jampeetong and Brix (2009) with some modi cations. Fresh plant leaves (0.1g, FW) were cut into pieces and placed in 5-ml asks. Then, 5 ml 95% ethanol was added into the asks to extract the photosynthetic pigment in the dark for 24 h. Plant pigment contents were determined with a spectrophotometer at 470, 649 and 645 nm for chlorophyll a, chlorophyll b and carotenoids, respectively. Values were calculated according to Lichtenthaler and Wellburn (1983). All spectrophotometric analyses involved in this experiment were accomplished with a MAPADA UV-1200 spectrophotometer (Shanghai Meipuda Instrument Co. Ltd., Shanghai, China).

Lipid perioxidation measurement
Malondialdehyde (MDA) content (µmol/g FW) was used as a proxy of lipid peroxidation level. Plant leaves (0.1g, FW) were homogenized using 3 ml 10% (w/v) trichloroacetic acid (TCA), and then the homogenate was centrifugated at 4000 rpm for 10 min at 4°C. For every 1mL of supernatant, 1 mL 0.6% (w/v) thiobarbituric acid (TBA) was added. The reaction mixture was heated at 95°C for 25 min and centrifugated at 4000 rpm for 10 min at 4°C after being cooled quickly. The resulting supernatant was analyzed at 450, 532 and 600 nm, respectively (Cang and Zhao 2013).

Measurements of enzyme activity, soluble protein and H 2 O 2 content
Plant leaves (0.1g, FW) were homogenized using 50 mM phosphate buffer solution (pH 7.8) containing 1% (w/v) polyvinyl pyrrolidone (PVPP) and 0.1 mM ethylenediaminetetraacetic acid (EDTA) at 4°C. Then, the homogenate was centrifuged at 8000×g for 15 min at 4°C. The resulting supernatant was used for SOD, CAT, POD, PPO and APX activity, soluble protein and H 2 O 2 content assays.
SOD activity was measured at 560 nm according to Cang and Zhao (2013). The reaction mixture was 50 mM phosphate buffer solution (pH 7.8) containing 13 mM methionine, 0,075 mM NBT (nitro blue tetrazolium), 0.1 mM EDTA (ethylene diamine tetraacetic acid), 0.002 mM ribo avin and enzyme extract. The unit (U) of SOD activity (U/mg protein) was de ned as the amount that resulted in a 50% inhibition of the initial rate of the reaction in the absence of the enzyme. H 2 O 2 content (µmol/g FW) was determined according to Shi (2016). Brie y, for every 1 mL enzyme extract, 1 mL 5% (w/v) titanium sulfate was added. After 10 min, the reaction solution was centrifuged at 12000 rpm for 10 min at 4°C and the supernatant was then analyzed at 410 nm.
Soluble protein content was determined according to Bradford (1976) and the bovine serum albumin was used as standard. Brie y, the rection mixture contained 5 mL Coomassie brilliant blue G-250 solution and 0.1 mL crude extract. After 2 min of reaction, the soluble protein content (mg/g FW) was calculated by recording the absorbance at 595 nm.

Measurement of non-enzymatic antioxidants
Reduced ascorbic acid (AsA) was determined according to Cang and Zhao (2013) with some modi cations. Plant leaves (0.1g, FW) were ground by 4 mL 5% (w/v) TCA and then centrifuged at 4000 rpm for 10 min at 4°C. Then 1 mL supernatant was mixed with 1 mL 5% (w/v) TCA, 1 mL absolute ethanol, 0.5 mL 0.4% (w/v) H 3 PO 4 in absolute ethanol, 1 mL 0.5% (w/v) red phenanthroline (BP) in absolute ethanol and 0.5 mL 0.03% (w/v) FeCl 3 . The reaction mixture was then incubated at 30°C for 40 min, and the absorbance at 534 nm was recorded. And a standard curve with AsA was used.

Integrated biomarker response
To visualize and compare the stress degree among different treatment groups, the integrated biomarker response (IBR) including MDA, H 2 O 2 , SOD, CAT, POD, PPO, APX, AsA and GSH were computed according to Kim et al. (2010). The detailed analysis process was shown in supplementary information. Spearman correlation coe cients were calculated between the IBR levels in both species and cipro oxacin concentrations.

Statistical analyses
All values were expressed as the mean ± standard deviation (SD). The Levene and Kolmogorov-Smirnov tests were used to verify homoscedasticity and normality criteria, respectively. One-way analysis of variance (ANOVA) was used for the statistical analysis. And a least signi cant difference (LSD) test was used to separate differences between pairs of treatments, where all the differences were considered signi cant at p < 0.05 (SPSS 23.0, IBM Inc., Chicago, IL, USA). Graphs were produced using Sigma-Plot 12.5 (Systat Software, Inc., USA). No shoots and roots of H. verticillata formed at 7d and thus no relevant analysis was performed then.

Germination and rooting
Germination rates of both P. crispus and H. verticillata were not affected by cipro oxacin exposure (p > 0.05; Fig. 1A). Rooting rate of P. crispus decreased slightly but not signi cantly with rising cipro oxacin concentrations (p > 0.05; Fig. 1B Cipro oxacin did not affect the germination time of P. crispus (p > 0.05; Fig. 1C), while 0.1 mg/L treatment signi cantly decreased the germination time of H. verticillata (p < 0.05; Fig. 1C). 10 mg/L cipro oxacin signi cantly increased the rooting time of P. crispus (by 42.5%) when compared with the control (p < 0.05; Fig. 1D). Rooting time of H. verticillata also showed slight increases with increasing cipro oxacin concentrations, reaching levels signi cantly higher than the control in 5 mg/L treatment (p < 0.05; Fig. 1D).

Tolerance indicators and seedling growth
When compared with the control, cipro oxacin at 1-10 mg/L and 0.1-10 mg/L signi cantly decreased RTI of P. crispus at 7-21d and 28-35d, respectively (p < 0.05; Fig. 2A) 3.3 Time-dependent IC50 based on STI and RTI IC50 values of cipro oxacin based on RTI and STI of both species decreased over time (Tab. 1), indicating tolerance of these two plant species was attenuated with prolonged exposure of cipro oxacin. What's more, regardless of time and species, IC50 values based on RTI were respectively lower than those based on STI, suggesting RTI was more sensitive to cipro oxacin exposure. NA a referred to "not available" due to lack of signi cance of the models (no calculation of the IC50 values), or because no radicle or shoot emerged then.

Plant biomass
As shown in Fig. 3A, B, cipro oxacin at 1-10 mg/L signi cantly decreased both the shoot biomass and root biomass of P. crispus (p < 0.05). Signi cant decreases of the biomasses in H. verticillata were observed at cipro oxacin concentrations of 0.1-10 mg/L (p < 0.05). As shown in Fig. 5C, cipro oxacin sharply elevated SOD activity of P. crispus (p < 0.05), reaching levels 1.6-5.0 times higher than the control. However, SOD activity of H. verticillata showed no signi cant differences when compared with the control (Fig. 6C; p > 0.05). CAT activity of P. crispus was signi cantly inhibited, which declined by up to 22.9%, 40.4% and 59.1% at 1, 5 and 10 mg/L cipro oxacin compared with the control, respectively ( Fig. 5D;

Integration of biomarker responses
As shown in Fig. 5-6, physiological biomarkers in both species responded differently to varying cipro oxacin concentrations. Therefore, for comparison, nine physiological biomarkers were integrated and displayed as star plots (Fig. 7). The IBR values of both P. crispus and H. verticillata showed good positive correlations with cipro oxacin concentrations (r = 0.991, p < 0.05; r = 0.943, p < 0.05, respectively), indicating that higher concentrations of cipro oxacin caused higher degree of stress on these two species.

Discussion
Many studies reported that seed germination of crops was insensitive to various antibiotics such as uoroquinolones, lincosamides, macrolides, sulfonamides, tetracyclines (Hillis et al. 2011;Bellino et al. 2018;Pan and Chu 2016). In our study, cipro oxacin employed did not in uence both the germination rate and rooting rate of P. crispus and H. verticillata, which suggested that the effects of antibiotics on turion germination were nearly not available. Some cellular processes might be negatively affected by antibiotics (Gambonnet et al. 2001), but germination was considered as a highly conserved process during which su cient nutrients, carbohydrates, and proteins were stored and available for the appearance of germ or radicle (Hillis et al. 2011). Turions functionally resemble seeds, serving multiple functions including carbohydrate storage, propagation, and dispersal (Jian et al. 2003). In this light, germination and rooting of turions of P. crispus and H. verticillata could be considered tolerant to cipro oxacin exposure. Notably, without impacts on germination and rooting rates, cipro oxacin could accelerate or delay the germination and rooting process, which were assessed by germination time and Additionally, nitric oxide was a main factor that mediated the auxin response during rooting process (Pagnussat et al. 2004). Cipro oxacin has been proved to disturb cellular nitric oxide production (Aiassa Our results showed that cipro oxacin especially at higher concentrations (5-10 mg/L) had retarding effects on the rooting process of H. verticillata and P. crispus, and these might because cipro oxacin disturbed the cellular nitric oxide level, which could further impact auxin response as well as rooting performance of turions.
Regardless of exposure time, RTI of both species decreased signi cantly with increasing cipro oxacin concentrations, indicating toxicity of cipro oxacin on RTI was concentration-dependent. Unlike RTI, STI of P. crispus at 7d was not in uenced by cipro oxacin treatment. What's more, cipro oxacin promoted STI of H. verticillata at 14-21d. This might because cipro oxacin addition provided more carbon and nitrogen sources for plant growth (Mao et al. 2021), resulting in but better growth performance and higher STI over these short exposures. However, after longer exposures, STI of both species also decreased with increasing cipro oxacin concentrations, indicating hysteresis effects of cipro oxacin toxicity on STI. RTI of P. crispus treated with 1-5 mg/L cipro oxacin and H. verticillata treated with 0.1-5 mg/L cipro oxacin decreased signi cantly over exposure time, with STI of both species in 1-5 mg/L treatments decreasing signi cantly with time prolonging. These results suggested seedling tolerance of H. verticillata and P. crispus to cipro oxacin decreased over time, and both RTI and STI were reliable and sensitive indicators for cipro oxacin toxicity on plant early establishment. Moreover, considering that IC50 values for RTI were lower than those for STI, we concluded that RTI was more sensitive than STI, regardless of species and exposure time. Previous studies also demonstrated high sensitivity of root elongation to antibiotics including tetracycline, sulfamethazine, nor oxacin, erythromycin and chloramphenicol compared to shoot elongation (Pan and Chu 2016). Root number and shoot number could re ect seedling growth status directly. Regardless of exposure time, root number of both species decreased signi cantly with increasing cipro oxacin concentrations, while shoot number was less affected. Roots are the main organ for absorbing antibiotics and growth of roots was more severely inhibited than that of shoots (Guo et al. And worse growth performance of P. crispus and H. verticillata might be related to plant defense against oxidative stress induced by cipro oxacin, during which cellular energy was highly consumed and thus energy available for growth process might be limited.  (Mittler et al. 2004). In P. crispus, activated SOD activity produced large quantities of H 2 O 2 . Meanwhile, POD and APX activities, AsA and GSH contents were signi cantly elevated. These antioxidants worked well in removing ROS induced by cipro oxacin, thereby controlling H 2 O 2 levels tightly and maintaining stability of cell membranes in P. crispus. However, CAT and PPO activities responded negatively under cipro oxacin exposure. As a photosensitizer, CAT would be likely inactive once photosynthetic pigment contents were reduced (Smirnoff 1995), which was in line with the observed reductions in pigments contents. Decreased activity of PPO has been considered as one of the outcomes of plant enhanced antioxidant capacity (Boeckx et al. 2015), indicating the local oxygen was maintained at levels that could not cause photoinhibition or oxidative damages. For H. verticillata, SOD activity increased slightly but not signi cantly in 5-10 mg/L treatments. At the same time, activated POD and APX could remove H 2 O 2 catalyzed by SOD activity. Compared with P. crispus, the effects of SOD in detoxifying ROS in H. verticillata was relatively weaker. This might because the quantities of superoxide anions were beyond the capacity of SOD, which would inhibit the outputs of this enzyme. Likely, as a compensative response, PPO activity was signi cantly activated, indicating high amounts of oxygen were removed by the oxidation of phenolic compounds. Additionally, AsA and GSH contents increased markedly especially at high cipro oxacin concentrations, indicating their vital roles in quenching ROS and protecting cells from oxidative damages. To conclude, rstly, though P. crispus and H. verticillata seemingly employed different antioxidants to prevent oxidative damages, AsA and GSH played key roles in both species. Both AsA and GSH are water soluble and easily accessible with potent antioxidant properties, and our results might provide some research ideas for alleviation of phytotoxicity of cipro oxacin. Secondly, the POD-H 2 O 2 decomposition system was involved in the degradation of chlorophylls (Kar and Choudhuri 1987), and the observed elevations of POD activity could be invoked to account for the reductions of chlorophylls contents in both species. Thirdly, under exposure to antibiotics like cipro oxacin, defense against oxidative stress could elevate cellular energy consumption, causing lower amounts of energy available for growth (Aderemi et al. 2018). To speculate, under cipro oxacin exposure, P. crispus and H. verticillata might allocate much more cellular energy to detoxify ROS and maintain physiological stability, resulting in growth inhibitions related to insu cient energy supply.
By integrating different biomarkers, IBR levels were employed to compare stress degree related to various environmental conditions, where a higher IBR level indicated more serious impact in an organism (Kim et al. 2010). We found IBR values increased with rising cipro oxacin concentrations. That was, the higher the cipro oxacin concentrations were, the more serious the damage to both submerged plants. These results were consistent with those of plant tolerance index, photosynthetic pigment contents and plant biomass accumulations, suggesting toxicity effects of cipro oxacin on these two species were concentration-dependent. On the other hand, the selected biomarkers responded e ciently to cipro oxacin pollution, and the IBR could serve as a reliable parameter for quantitative evaluation of the toxicological effects of cipro oxacin toward aquatic plants.

Conclusion
Germination and rooting of turions of P. crispus and H. verticillata were tolerant to cipro oxacin employed in this study. However, cipro oxacin could delay rooting process by increasing rooting time. During plant early establishment, both RTI and STI were reliable and sensitive endpoints to assess cipro oxacin toxicity on seedling development. Additionally, roots came to be more sensitive than shoots under cipro oxacin treatment. With better growth performance of shoots, H. verticillata appeared to be more tolerant to cipro oxacin exposure than P. crispus. Plant biomass accumulations of both species were inhibited by cipro oxacin, which could be partially contributed from phytotoxicity on photosynthetic pigment. Cipro oxacin induced oxidative stress and activated the antioxidant system of both species.
Key antioxidant enzymes for ROS detoxication differed between P. crispus and H. verticillata, while nonenzymatic antioxidants (AsA and GSH) functioned well in both species, which might provide valuable research ideas for alleviation of phytotoxicity of cipro oxacin. As the rst study that investigated toxicity cipro oxacin on germination and seedling development of submerged turions, this research might provide a new perspective to reveal the aquatic environments risks of antibiotics pollution. Toxicity

Competing interests
All authors declared that there were no competing interests.     (H) and GSH content (I) of P. crispus exposed to cipro oxacin. Bars represent the mean of ve replicates ± standard deviation. Different letters represent statistically signi cant differences among different treatments (p < 0.05, LSD test). (H) and GSH content (I) of H. verticillata exposed to cipro oxacin. Bars represent the mean of ve replicates ± standard deviation. Different letters represent statistically signi cant differences among different treatments (p < 0.05, LSD test).