3.1. Changes in macrophytes’ morpho-physiological status
The morphological examination of the studied macrophytes (Supplementary figures 1-10), revealed evident symptoms of decolourisation, chlorosis, necrosis for some plants despite similar biomass and height to controls. This was the case for C. riparia, C. longus, C. rotundus, I. pseudacorus, J. effusus, and M. aquatica, suggesting that these plants did not acclimatize well within the 15 days incubation time. In contrast, the morphology of P. australis, S. holoschoenus, and T. angustifolia was indicative of healthy plants. The details related to the physiological profiles and metal(loid) uptake are discussed in the proceeding sections.
3.1.1. Plant height
The compiled images taken in successive intervals of 1, 5, 10 and 15 days are presented in the supplementary material (Supplementary Fig 1-10). These images were used for the estimation of plant height. Figure 1.a. summarises the total plant height for each of the plants subjected to clean (C) and polluted groundwater (T). After 15 days of exposure total plant heights (cm) were measured. Significant differences were only found for I. pseudacorus (5.49 ± 2.49), L. salicaria (8.13 ± 3.16), and P. australis (3.49 ± 1.27); for these species, plants exposed to contaminated water showed 48, 46, and 46% lower height than their respective controls. For all other species no significant difference were found between treatments (C. riparia, C. longus, C. rotundus, J. effusus, M. aquatica, S. holoschoenus, and T. angustifolia). Height and biomass are reliable indicators of a plant's final growth. In addition, plant performance can be monitored by the growth rate. For this purpose, percentage change in plant heights were also recorded at 5-day intervals (Figure 1.b). On day 10 reduced heights compared to the controls were noted for C. riparia, C. longus, C. rotundus, J. effusus, L. salicaria, M. aquatica and T. angustifolia. However, on day 15 similar heights to the controls were observed suggesting the initial stress due to metal(loid)s and extreme water conditions, while plant recovering if the exposure duration to metal(loid) contamination is increased.
3.1.2. Studied plant physical parameters upon metal(loid)s exposure
The studied wetland plants were exposed to polluted groundwater for 15 days and subsequently harvested. Fresh and dried biomass of aerial parts and roots, along with plant height, were quantified (Table 1). Except for C. riparia and M. aquatica, significant differences in the shoot fresh biomass were observed between the control and the polluted-water treatment. Differences in shoot biomass were positive or negative depending on the plant species. In the case of C. longus, J. effusus, and T. angustifolia (37178 ± 3284, 70992 ± 7788, and 7225 ± 481, respectively in mg) a decrease in fresh shoot biomass of 65, 31, and 35%, as compared to the controls, was recorded. However, for C. rotundus, I. pseudacorus, L. salicaria, P. australis, and S. holoschoenus, an increase in fresh shoot biomass was observed when plants were exposed to polluted water. The observed fresh biomass (in mg) for these plants was 25088 ± 3087, 106340 ± 2451, 55107 ± 5128, 9940 ± 285, 24195 ± 2591. For plants exposed to contaminated water the increase in fresh shoot biomass was highest for P. australis and S. holoschoenus (184 and 139%, respectively), while for C. rotundus, I. pseudacorus, and L. salicaria the increment in fresh shoot biomass was 69, 39, and 39%, accordingly. The dried shoot biomass for all studied plants was found to vary significantly. Interestingly, dried shoot biomass (mg) of P. australis (697± 72) and S. holoschoenus (2072 ± 21) was significantly lower (41 and 13%, respectively) for the water-contaminated treatment in comparison to the controls. The relative increment in dried shoot biomass for plants cultivated with polluted water was in decreasing order: C. longus (881%), I. pseudacorus (758%), M. aquatica (525%), T. angustifolia (431%), C. riparia (270%), J. effusus (87%), C. rotundus (37%), and L. salicaria (26%). For the water storage capacity (WSC) in the shoot (g g-1 DW) all plants showed significant variabilities except L. salicaria (5.1 ± 0.9). The plants exposed to contaminated water for the species C. longus (0.5 ± 0.1), T. angustifolia (1.6 ± 0.1), M. aquatica (4.9 ± 0.5), C. riparia (1.6 ± 0.3), I. pseudacorus (2.6 ± 0.2), and J. effusus (3.6 ± 0.7) showed a lower shoot WSC (98, 92, 85, 87, 88, and 68%, respectively) compared to the controls. The opposite trend was observed for the species C. rotundus (8.3 ± 1.3), P. australis (13.4 ± 2), and S. holoschoenus (10.7 ± 1.1). The largest increase in shoot WSC with respect to the controls was observed for P. australis and S. holoschoenus with an increment of 573 and 231%, respectively. Similarly, an increased shoot WSC of 27% was reported for C. rotundus plants exposed to contaminated water in comparison to the clean water controls.
Significant differences were also observed for the roots between controls and plants exposed to contaminated water. For fresh root biomass (mg) significant differences were observed between treatments for all species except for C. rotundus (8172 ± 430) and P. australis (2890 ± 347). A reduction in the fresh root biomass of C. riparia (2846 ± 566), L. salicaria (10123 ± 981), and M. aquatica (4557 ± 479) was observed for the contaminant-exposed treatments compared to the controls. Here, plants exposed to contaminated water produced 47, 34, and 25% less biomass, respectively, compared to plants grown with clean water. Fresh root biomass was significantly larger for I. pseudacorus (26602 ± 2438), S. holoschoenus (9437 ± 937), C. longus (21995 ± 2826), J. effusus (28180 ± 1260), and T. angustifolia (2264 ± 380). For these plants a relative increase of 88, 82, 77, 51, and 14% was recorded compared to the controls. The root-dried weights of C. riparia, C. longus, and C. rotundus did not differ between treatments. In contrast, for I. pseudacorus (7658 ± 463), J. effusus (11888 ± 1430), S. holoschoenus (3178 ± 136), and T. angustifolia (634 ± 41) a relative increase of 78, 42, 32, and 22% higher dried root biomass (mg) was recorded for plants exposed to metal(loid)s with respect to the controls. For C. longus, C. rotundus, I. pseudacorus, J. effusus, L. salicaria, and M. aquatica. no significant changes were noted for the root WSC between plants exposed to metal(loid)s and the controls. For C. riparia, I. pseudacorus, and T. angustifolia the root WSC decreased by 53, 40, and 9% in plants exposed to contaminated water compared to the controls. The opposite trend was observed for S. holoschoenus and P. australis with an increase in root WSC of 71 and 45%, respectively, in plants exposed to contaminated water.
Higher cumulative fresh root biomass was observed on day 15 for S. holoschoenus (120%), P. australis (94%), C. rotundus (47%), I. pseudacorus (47%), L. salicaria (19%), and M. aquatica (5.4%) plants exposed to contaminated water than for the controls. The opposite was reported for C. longus (-18%), C. riparia (-28%), T. angustifolia (-31%), and J. effusus (-50%). Higher cumulative dried root biomass was observed on day 15 for C. longus (509%), I. pseudacorus (499%), M. aquatica (260%), T. angustifolia (227%), C. riparia (127%), J. effusus (65%), C. rotundus (27%), S. holoschoenus (9.4%), and L. salicaria (5.1%) plants exposed to contaminated water than for the controls. P. australis was the only species with lower cumulative root dried biomass in plants exposed to contaminated groundwater (-34%) compared to the control. The water storage capability (%) on day 15 was higher for P. australis (335), S. holoschoenus (189), C. rotundus (16), and L salicaria (2.258) and lower for C. longus (90), T. angustifolia (-82), M. aquatica (-81), C. riparia (-81), I. pseudacorus (-81), and J. effusus (-60.82) in plants exposed to contaminated water than in the controls.
3.2. Meta(loid)s concentration in macrophytes, rhizospheric soil and groundwater.
The results of metal(loid)s uptake and compartmentalization are presented in Figure 2. Additional details related to the statistical analysis are provided in Supplementary Tables 2 (for the shoot) and 3 (for the root). Among the studied plants, one of the most prominent features observed was that the largest quantity of metal(loid)s generally was stabilized in the roots rather than in the shoots of the plants. However, arsenic (As), in μg kg-1 of the plant’s dried weight (DW), was found to be significantly highest in the aerial parts of P. australis (4.0 ±0.1, 4.1 ±0.2, and 5.1 ±0.5, on the 5th, 10th, and 15th day of exposure, respectively). Similarly, P. australis also showed the highest As uptake in the root on the 5th and 10th day with 42.3 ±2.9 and 43.7 ±1.7 μg kg-1 DW, respectively. Alongside with P. australis, C. riparia also showed higher As root uptake on the 10th day (47.1 ±0.4), while only C. longus showed higher As uptake in root on the 15th day (65.9 ±0.3) (Figure 2.a). The Cd uptake in the shoots and the roots showed a variable pattern depending on the species with highest levels for J. effusus on the 15th day (3.8) compared to other plants. For the roots, the highest level of Cd (μg kg-1 DW) was reported for C. longus, I. pseudacorus and P. australis on the 15th day (ranging between 4 to 5.5) (Figure 2.b). For Cu uptake (mg kg-1 DW) in the shoots, T. angustifolia showed the highest concentrations of all species at each successive exposure interval, 125.4 ±0.9, 244.4 ±0.4, and 303.1 ±5.4, respectively. J. effusus also showed a significantly higher shoot uptake on the 15th day (347.1 ±44.6 mg kg-1 DW). For the roots no successive Cu uptake trend was observed over time; the highest concentration (1199) was reported on the 15th day for M. aquatica (Figure 2.c). S. holoschoenus showed significantly higher Fe (mg kg-1 DW) uptake in the shoot at each interval compared to the other plants, with highest Fe concentration (2842 ±113) on the 15th day. For root Fe uptake the highest values (mg kg-1 DW) were noted for J. effusus with 9758 ±3224 on the 5th day and 12132 ±618 on the 10th day as well as for C. longus (21859 ±784) on the 15th day (Figure 2.d). The highest Ni uptake in the shoots was noted for J. effusus (314 ±2) on the 15th day. For the roots the highest Ni uptake was recorded generally for C. riparia and C. longus at each time interval with a maximum value of ~360 on the 15th day (Figure 2.e). The highest Pb uptake (μg kg-1 DW) in the shoot was noted for P. australis with a maximum concentration of 1.5 ±0.4 on day 15. For Pb uptake in roots, S. holoschoenus showed a significantly higher increase at all time intervals compared to other species, with a maximum value of 6.1 ±0.5 (μg kg-1 DW) on 15th day (Figure 2.f). The Zn uptake in the shoot (mg kg-1 DW) was variable among the different plants. The highest Zn uptake was recorded for J. effusus on the 15th day (212 ±8). For the roots the highest Zn uptake was observed for C. riparia with 218 ±11, on day 15. In addition, high Zn root uptake in mg kg-1 DW was reported for C. longus (233 ±12) on day 15, for S. holoschoenus (201 ±7, and 209 ±15, respectively) on days 10 and 15, as well as for T. angustifolia (215 ±6) on day 10.
The profiles for metal(loid) content in the rhizosphere of the macrophytes is presented in Table 2. Among the most notable trend was the successive increase in the metal(loid) concentration in the rhizospheric soil as time progressed, with the highest stabilization on day 15. An exception to this was As; here, the highest stabilization was achieved by I. pseudacorus on day 5. The highest As rhizospheric retention (77 ±4 μg kg-1 DW) was performed by J. effusus. For Cd stabilization in rhizospheric soil (μg kg-1 DW), the best performing plants were L. salicaria, C. longus, M. aquatica, and P. australis revealed significantly with 24 ±4, 26 ±1, 24 ±1, and 23 ±2, respectively. J. effusus and L. salicaria showed the highest rhizospheric stabilization for Cu (2529 ±215, and 2694 ±137 mg kg-1 DW, respectively). The highest stabilization for Fe in soil was observed for J. effusus with 9826 ±74 mg kg-1 DW. L. salicaria significantly showed the highest concentrations in rhizospheric soil for both Ni (1835 ±146 μg kg-1 DW) and Zn (798 ±85 mg kg-1 DW). For Pb, the highest soil stabilization was noted for I. pseudacorus (9 μg kg-1 DW).
The concentrations of metal(loid)s remaining in the polluted groundwater after the 15 days exposure time are presented in Figure 3. A significant reduction in metal(loid) concentrations in water was observed for most species studied because of either translocation into the plants, precipitation/adsorption in the rhizospheric soil or a combination of the two. The As was found lowest (~0.22 mg L-1) in ground water with J. effusus, however, this retention was mainly in the rhizospheric soil (Supplementary Table 4). For Cd, the lowest level (1.60 ±0.03 mg L-1) in treated groundwater were noted for the I. pseudacorus and seeing the rhizospheric level, the Cd is translated in the plant (Figure 2, Table 4). The Cu levels were lowest (~130 mg L-1) with I. pseudacorus, J. effusus, and C. riparia. The primary reason for this removal was due to plant uptake for C. riparia, while for I. pseudacorus, and J. effusus. Ni levels were lowest (~57 mg L-1) and were found for C. riparia, I. pseudacorus, and J. effusus which due to both soil adsorption and plant uptake. For Zn the lowest levels were noted (~33 mg L-1) with for C. riparia, I. pseudacorus, and J. effusus, that was mainly due to plant uptake of Zn in the roots. The ranges of percentage of metal(loid) removal from the contaminated groundwater after cultivation of wetland plants for 15 days was as follows: As (19-36%), Cd (12-28%), Cu (7-21%), Fe (1-31%), Ni (46-55%), Pb (97-98%), and Zn (44-52%).
3.3. MLR analysis on the impact of metal(loid) uptake and polluted water exposure duration in the shoot and root compartments
The impact of specific metals on plant biomass was assessed by means of multiple linear regression (MLR) equations considering metal(loid) uptake and duration of exposure as independent variables and dried plant biomass as dependent variable. Results are displayed in Supplementary Figure 11 for shoots and Supplementary Figure 12 for roots. In this model, a p < 0.05 was considered statistically significant, and a coefficient of regression (R2) with a value close to 1 was considered a good/strong model fit. The data used for this analysis were taken from plant dried biomass (compartment-wise) at the interval of 5, 10, and 15 days of exposure. The number of observations (n) for dependent and independent variables was 9.Supplementary Figure 11.a-j displays the MLR models for shoots. For shoot biomass the least favourable model fit R2 were 0.659 for I. pseudacorus, 0.752 for S. holoschoerus and 0,811 for L. salicaria. For all other species the model fit R2 ranged between 0.892 for C.rotundus and 0.990 for J.effesus. The contribution of metal concentrations and exposure time to the total model variability differed significantly between species. On occasion, one single variable accounted for more than 50-75% of the total data variability. This was the case for C. riparia (Cu), C. longus (Fe), I.pseudacorus (Fe) and L. salicaria (exposure time). For the other species at least two but generally three or more variables accounted for >75% of the model variability. Supplementary Figure 12.a-j displays the MLR models for roots. Here the model fit R2 was >0,96 for all species. As for the shoots the contribution of metal concentrations and exposure time to the total model variability differed significantly between species. In addition, no common pattern was observed for the roots and shoots of the same species, that is, the variables accounting for most of the variability observed in the shoots were not necessarily the same ones those reported for the roots. For C. riparia and C. rotundus Fe accounted for approximately 50% of the total variability observed, whereas for I.pseudacorus the same was observed for As. For the other species at least two but generally three or more variables accounted for >50% of the model variability.
3.4. Bioaccumulation factor, bioconcentration factor and translocation factor of the studied macrophytes
The bioaccumulation factor (BAF) for each metal(loid) and plant on days 5, 10, and 15 are shown in Table 3. For As, the highest BAF was noted generally on the 5th or 10th day of exposure, except in the case of I. pseudacorus. The highest BAF for As (1.3-1.5) on the 5th and on the 10th day (1.1) was recorded for C. riparia and P. australis; on the 15th day for M. aquatica (0.2). The highest BAF for Cd on the 10th (1.2) and 15th day (0.3) was observed for T. angustifolia. Similarly, T. angustifolia also showed the highest BAF for Cu on days 10 (0.9) and 15 days (0.5) as well as for Ni with 1,5 on day 10 and 0.3 on day 15. Likewise, J. effusus exhibited a BAF of 0.3 for Ni on day 15. C. rotundus (0.5) and S. holoschoenus (0.7) exhibited the highest BAF for Fe on day 15. For Pb, the highest BAF on day 5 was noted for T. angustifolia (1.2), while P. australis showed the highest BAF values on the 10th and 15th days (0.9 and 0.7, respectively).
The bioconcentration factors (BCF) for each metal(loid) and plant on days 5, 10, and 15 are shown in Table 4. For As, the BCF increased as time progressed, except in the case of I. pseudacorus and L. salicaria. The highest BCF for As on day 15 was found for C. longus, P. australis, and S. holoschoenus (~1.5). For Cd, the highest BCFs (~0.4) on day 15 was recorded for I. pseudacorus. For Cu the BCF increase in time was observed only for C. longus, J. effusus, and L. salicaria. The greatest BCF for Cu on day 15 was noted for C. riparia (1.4) and C. rotundus (1.2). For Fe a BCF increase in time was observed for all plants except for J. effusus. M. aquatica showed the highest BCF for Fe (4.9) on day 15. The highest BCF for Pb was found on days 10 and 15 for all plants. The highest BCF for Fe was recorded for S. holoschoenus (2.2-3.2). Interestingly, the highest BCF for Ni (137) and Zn (9.75) were observed for C. riparia on day 5, whereas the BCF significantly decrease to ≤1 on days 10 and 15.
The translocation factors (TF) for each metal(loid) and plant on days 5, 10, and 15 are presented in Table 5. L. salicaria exhibited the highest TF for As on the 5th (2.5) and 10th (0.2) day, whilst M. aquatica showed the highest TF (0.2) on day 15. For Cd, J. effusus, showed an increasing TF trend as time progressed with 1.4 on day 10 and 2.2 on day 15. The highest TF for Cu were also observed for J. effusus with 0.8 on day 10 and 0.7 on day 15, whereas C. rotundus showed the highest TFs (~0.7) for Fe on days 10 and 15. J. effusus also showed the highest TF for Ni (2.5) on day 15. For Pb P. australis showed the highest TF at all exposure intervals with 3.9 on day 5, 1.5 on day 10 and 1.8 on day 15. For Zn I. pseudacorus (1.2) and L. Salicaria showed the greatest TF on day 5, whereas T. angustifolia (1.1) showed the largest TFs on day 10 and 15.