Effect of genotype, Cr(III) and Cr(VI) on plant growth and micronutrient status in Silene vulgaris (Moench)

Chromium released into the environment from industrial activities has become an important environmental concern. Silene vulgaris has been proven to be tolerant to many heavy metals, so it is considered an interesting species in the revegetation and restoration of polluted soils, but no information is available about its response to Cr. The objective of this work was to study uptake and influence on plant growth of Cr(III) and Cr(VI) in six genotypes (four hermaphrodites and two females) of S. vulgaris from different sites of Madrid (Spain). Plants were treated for 12 days with 60 μM of Cr(III) or Cr(VI) in semihydroponics. Dry weights, soil-plant analysis development values (SPAD) reading with chlorophylls and micronutrient and total Cr concentrations were determined. Metal uptake was higher in presence of Cr(VI) than of Cr(III) and poorly translocated to the shoots. In both cases S. vulgaris did not show visual toxicity symptoms, biomass reduction, or differences among SPAD values as consequence of Cr additions. However genotypes SV36 and SV38 showed Fe and Mn imbalance. This is the first report on the relatively good performance of hermaphrodite and female S. vulgaris genotypes in Cr uptake and physiological traits, but further studies will be necessary to elucidate the mechanisms by which the gender may influence these variables. S. vulgaris presented high diversity at genotypic level; the treatment with hexavalent Cr increased the differences among genotypes so the use of cuttings from an homogeneous genotype seems to be an adequate method for the study of this species. Additional key words: bladder campion; metal pollution; metal speciation; tolerance; nutrient balance. * Corresponding author: ana.elena.pradas@madrid.org Received: 20-09-12. Accepted: 26-06-12. Abbreviations used: ANA (1-naftil acetic acid); ATP (adenosine triphosphate); BCF (bioconcentration factor); EPA (United States Environmental Protection Agency); SPAD (soil-plant analysis development values); TF (translocation factor). Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA) Spanish Journal of Agricultural Research 2013 11(3): 685-694 Available online at www.inia.es/sjar ISSN: 1695-971-X http://dx.doi.org/10.5424/sjar/2013113-3536 eISSN: 2171-9292


Introduction
Chromium is a heavy metal used on a large scale in industry. In the last years its release into the enviroment from anthropogenic activities has become a major health and ecological hazard. On a worldwide basis, about 80% of the chromium mined goes to metallurgical applications, but it is also used in the manufacture of stainless steel, wood treatment, leather tanning and chrome plating due to its corrosion-resistant properties (Barnhart, 1997). Although Cr contamination does not pose a global risk, it could be a serious problem for the local environment (Bielicka et al., 2005).
Once it enters the environment, its toxicity is determined to a large extent by its chemical form, which is also responsible for its mobility and bioavailability (Kotas & Stasicka, 2000). The two most common and stable chemical species of Cr in the environment are Cr(III) and Cr(VI). The active-redox Cr(VI) is more water soluble and more bioavailable than Cr(III), and it has been classified as carcinogen of Group A by the EPA (United States Environmental Protection Agency; USEPA, 1998). The role of Cr(III) as essential anion in mammals and plants is under question because it also seems to be toxic at higher doses than Cr(VI) (Levina & Lay, 2008).
In plants grown in non-contaminated soils, Cr concentrations are usually less than 1 mg kg -1 dry weight, rarely exceed 5 mg kg -1 and are typically in the order of 0.02-0.2 mg kg -1 dry weight (Pezennec, 2007). To date, high concentrations of Cr have been detected in different species as Brassica oleracea (Zayed et al.,
Biomass reduction and leaf chlorosis are the first toxicity symptoms shown by plants when they grow in soil with high doses of metal and hence in Cr. Among other reasons, this is caused by an imbalance in micronutrient status (Shanker et al., 2005). The Fe accumulation was reduced in leaves in presence of Cr(III) or Cr(VI) in B. oleracea (Pandey & Sharma, 2003), Spinacia oleracea (Chatterjee & Chatterjee, 2000) or Zea mays (Mallick et al., 2010). Concentrations of other micronutrients, such as Cu, Zn and specially Mn, were greatly affected. Micronutrients concentration decreased with regard to the reference levels.
Silene vulgaris (Moench) Garcke, the bladder campion, is a perennial weed widely distributed in Europe, North America, Asia and North Africa. It occupies a great variety of habitats, including metalliferous soils. Flower phenology and insect pollination have lead to a predominantly outcrossing habit; therefore it exhibits a high level of genetic variability (Prentice & Giles, 1993). Spanish natural populations of S. vulgaris have been characterized by morphological traits and molecular markers (Alarcón et al., 2008) that show a high genetic diversity. This species has a gynodioecious mating system characterized by the co-ocurrence of female and hermaphrodites individuals within populations (Taylor et al., 1999). Individuals of these populations are scattered along road sides and agricultural fields; thus their ecology is likely to be affected by anthro-pogenic factors associated with roadside maintenance as well as natural processes (Olson et al., 2005).
Theoretical studies suggest that because females cannot gain fitness through pollen, they must increase their investment in seed quantity and/or quality in order to coexist with hermaphrodites (Charlesworth & Laporte, 1998). This increase in female fitness has been observed in most gynodioecious species, but the few studies that investigated physiological differences between genders in gynodioecious species have produced inconsistent results (Poot et al., 1996;Caruso et al., 2003;Schultz, 2003).
The tolerance of many S. vulgaris ecotypes has been proven for most of heavy metal and their combinations, especially Cd, Cu, Fe, Mn, Pb, Zn and Hg (Paliouris & Hutchinson, 1991;Ernst & Nelissen, 2000;Ciarkowska & Hanus-Fajerska, 2008). Data from these studies show S. vulgaris to be an interesting species in the early stages of revegetation and soil restoration. Regarding genotypes, Cu has been the only metal studied to evaluate differences to metal uptake among genotypes of S. vulgaris (Price & Abrahams, 1994). There is no information about physiological traits and heavy metal tolerance between genders of S. vulgaris and about the ability of this species to tolerate and accumulate Cr.
The objectives of this study were: a) to assess whether the presence of low concentrations of Cr(III) or Cr(VI) in nutrient solution led to differences in Cr uptake by S. vulgaris at genotype level; b) to evaluate the effectiveness of the genotypes using the classical approach of the growth efficiency of wild plants based on micronutrient status and state of chlorophylls.

Plant material and growth conditions
Six genotypes of S. vulgaris (Moench) were chosen from different populations of Madrid, Spain (Table 1). A permanent 10 × 10 m plot (divided into 1 m 2 quadrats) was established to vegetative propagate in Alcalá de Henares, Madrid (Spain). Hermaphrodites and female individuals were easily discriminated by watching for the presence or absence of mature anthers. Cuttings of each juvenile growth of mature genotype were collected in March 2009. The base of the cuttings was dipped in 0.1% hormone Inavarplanté 1(indol-3butyric acid (A,B), 0.1% 1-naftil acetic acid (ANA), 4% ziram for 5 min, and transferred into floating polystyrene trays. Twenty five cuttings per tray were induced to root in tap water, and set in a growth chamber under controlled environmental conditions (temperature 20/16°C, 164.527 µmol E m -2 s -1 , 16/8 hour photoperiod). They were allowed to root for 3 weeks until their roots reached a length of 2.0 ± 0.5 cm (Wierzbicka & Panufnik, 1998).
Four trays, with four cuttings of each genotype in each, were used as independent replicates of each treatment. The pH of the solutions was buffered with 2 mM of MES and adjusted to 5.5 with KOH. Nutrient solution was replenished daily and completely changed every 3 days. Aliquots (20 mL) of nutrient solution were collected before and after each change to check pH and the oxidation state of Cr. Plants were treated for 12 days.
The concentration of Cr(VI) in nutrient solution was determined just after changing the nutrient solution by UV-Vis light spectrophotometer (Thermo Spectronic Helios Alpha) using the EPA method 7196A (USEPA, 1992). The total Cr concentration was measured in the samples of nutrient solution previously acidified and then, by atomic absorption spectrophotometry (VARIAN fast sequential, model AA240FS). The total concentration of Cr(III) was calculated by subtracting the concentration of Cr(VI) from the total Cr concentration. Results from Cr speciation analysis indicated that the transformation among Cr species was not significant during the experiment.

Plant analysis
SPAD (soil-plant analysis development) index was measured to estimate the chlorophyll state. This is a green colour index related to chlorophyll content. The average of six determinations per leaf was recorded in the four plants of each genotype per tray using a Minolta Chlorophyll Meter SPAD-502. All plants were harvested. The roots and aerial parts were separated and washed thoroughly with MilliQ water. A full sample was made with four plants of the same genotype in each tray, thus there were four trays and a total of four independent replicates per treatment and genotype were analysed. The vegetal material was dried in a forced air oven for 48 h at 70°C. Subsequently, the dry weights were recorded and aerial parts and roots were ground separately. Dried samples (0.25 mg) were digested in an Anton Paar Microwave Reaction System 3000 by adding 6 mL of 65% HNO 3 and 2 mL of 33% H 2 O 2. After cooling, the digests were filtered (Whatman f ilter paper nº 541) and brought up to a volume of 25 mL. Total concentrations of Cr, Fe, Mn and Zn were measured by flame atomic absorption spectrophotometer (Varian fast sequential model AA240FS) and Cu concentration by Sequential ICP-AES Liberty AX.
Tobacco leaves were used as certified reference materials (CTA-VL2, tobacco leaves). The recovery percentages were 88% in Fe, 99% in Zn, 93% in Cu, 88% in Mn and 109% in Cr.
Effect of Cr(III) and Cr(VI) on Silene vulgaris genotypes 687

Parameters of chromium accumulation in the plant
The following indexes were calculated to compare Cr uptake and translocation to the shoot as result of the different Chromium treatments. The translocation factor (TF) is defined as the ratio between the total metal concentration of shoots and roots. It shows the plant's ability to translocate heavy metals from the roots to the harvestable aerial part (Mattina et al., 2003): where C aerial and C root are the total Cr concentration in aerial part and roots respectively.
The bioconcentration factor (BCF) or phytoextraction rate was described as the heavy metal concentration in total plant divided by heavy metal concentration in nutrient solution (Kumar et al., 1995).
where C solution is the Cr concentration in the nutrient solution.

Statistical treatment
Statistical analyses were performed used the statistical package SPSS version 16.0. Data from all varia-bles considered were analysed by general lineal model (GLM) with doses of Chromium (no Cr, 60 µM Cr(III) or 60 µM Cr(VI) ) and S. vulgaris genotypes (SV19, SV21, SV27, SV30, SV36 and SV38) as experimental factors at α = 0.05 using the F-test. GLM was followed by a post hoc Duncan test to assess the significance of differences among treatments and among genotypes for each parameter.

Results
The statistical values (F) and empirical p-values of both factors and their interaction for all variables considered are presented in Table 2. Fig. 1 gives the total Cr concentration in the studied genotypes of S. vulgaris. Chromium uptake was strongly affected by the source of Cr in nutrient solution (p = 0.000), but not by the genotype. Chromium mainly accumulated in the roots of the genotypes exposed to Cr(VI) and Cr(III). The highest Cr concentrations were achieved in roots of plants growth with Cr(VI) with values between 374 and 481 mg Cr kg -1 dw.

Chromium uptake
Concentration of Cr in the shoots of S. vulgaris genotypes only increased in treatments with Cr(VI), which ranged between 6 and 20 mg Cr kg -1 dw. The    Figure 1. Total metal concentrations (mg kg -1 dw) in shoots and roots of Silene vulgaris genotypes: a) chromium, b) iron, c) zinc, d) copper and e) manganese. Control, white bars; Cr(III) treatment, grey bars and Cr(VI) treatment, black bars. Different lowercase and capital letters mean significant differences among treatments and genotypes, respectively (Duncan's test, p < 0.05, mean ± SE, n = 4).  (Table 3).
The bioconcentration factor (BCF), which evaluates the plant ability to take up the metals from the nutrient solution, is given in Table 3 for each genotype at the end of the experiment. The highest values were achieved when Cr(VI) was added to the nutrient solution. This factor decreased by 80% for all the genotypes treated with Cr(III) compared to Cr(VI). In relation to genotypes, there were no significant differences in Cr concentrations or BCF values. Though not statistically significant, there is a tendency of female individuals to accumulate less Cr in their tissues than hermaphrodites. The genotype SV21 seemed to be the most efficient in Cr uptake, as it showed the highest bioconcentration factor. Table 4 presents the dry weight for roots and shoots of the S. vulgaris genotypes after 12 days of exposure to the different treatments. Significant differences were mainly due to genotypes (p = 0.002 for shoots dry weights and p = 0.000 for roots) rather than to Cr treatments. As regards treatment with Cr(VI), the Duncan test showed that the hermaphrodite genotype SV36 presented the highest dry weight in shoots and roots while the female genotypes (SV38 and SV19) showed growth inhibition. To compare the effect of Cr treatments on chlorophyll content, SPAD values were measured for each genotype of S. vulgaris after 12 days of treatment started. The results are given in Table 5. 690 A. E. Pradas-del-Real et al. / Span J Agric Res (2013) 11 (3): 685-694 Table 3. Translocation factor (TF, ratio between the total metal concentration of shoots and roots) and bioconcentration factor (BCF, heavy metal concentration in total plant divided by heavy metal concentration in nutrient solution) of Silene vulgaris genotypes after 12 days of treatment with 60 µM of Cr(III) or Cr(VI)

Plant growth and SPAD index
TF Cr(III) 0.015 ± 0.005 b 0.021 ± 0.003 b 0.031 ± 0.005 b 0.027 ± 0.005 0.040 ± 0.008 b 0.022 ± 0.001 b Cr(VI) 0.026 ± 0.013 a 0.031 ± 0.004 a 0.021 ± 0.004 a 0.026 ± 0.006 0.018 ± 0.002 a 0.030 ± 0.005 a BCF Cr(III) 3 ± 2 4 ± 2 b 10 ± 5 b 2.6 ± 0.3 b 2.6 ± 0.7 b 3.6 ± 1.2 b Cr(VI) 13 ± 1 31 ± 14 a 23 ± 2 a 13 ± 2 a 20 ± 4 a 21 ± 3 a Different lowercase letters mean significant differences between Cr treatments (Duncan's test, p < 0.05, mean ± SE, n = 4).  In this setting, there were no significant differences in treatments (p = 0.066) but once again, there were differences among genotypes (p = 0.000). The leaves of female genotypes (SV19 and SV38) displayed the lowest SPAD values. Note that this trend is not the same as biomass yield, where the maximum and minimum values were kept for the same genotypes regardless of treatment. The SPAD index values showed differences among genotypes due to the effect of Cr(VI). The females (SV19 and SV38) presented the lowest SPAD when plants were treated with Cr(VI). Fig. 1 gives the concentrations of Fe, Zn, Cu and Mn in shoots and roots of S. vulgaris genotypes. Iron concentrations in shoots of genotypes SV36 and SV38 have decreased by 55% and 68% respectively, due to Cr(VI) in the nutrient solution. This fact was accompanied by increasing of Fe concentration in roots and significant to SV36 genotype, in which Fe increased by 160% over the control. Genotypes SV38, SV30, and SV36 treated with Cr(VI) have reduced by 40% Mn concentration in aerial parts compared to controls. These genotypes showed increments in Mn concentrations of roots, but not statistically significant except for genotype SV30 (up to 30% more than the control). Zinc and Cu concentrations in roots and aerial parts are not consistent enough to show if there is any influence of Cr uptake.

Micronutrient concentration
It should be noted that differences in micronutrient concentrations were found among genotypes. But the trend showed that hermaphrodite individuals had micronutrient concentrations lower than females. This is the case of hermaphrodite SV36 which presented the lowest micronutrient concentrations, significant only for Zn and Mn concentration when plants were treated with Cr(VI). Maximum micronutrient concentrations were reached by females and both hermaphrodites SV30 and SV27, especially for SV30, that achieved the highest concentrations of Zn, Cu and Fe when plants were treated with Cr(VI).

Discussion
The genotypes of S. vulgaris grown in nutrient solution did not show any visual or physiological toxicity symptoms after being treated with 60 µM of Cr(III) or Cr(VI) during 12 days. With the exception of Cr uptake, which was higher in the plants treated with Cr(VI) than with Cr(III), the main differences in the plant development that were found related more to the genotypes than to the Cr forms in the nutrient solution.
Total Cr concentrations in dry tissues of S. vulgaris exposed to Cr(III) and Cr(VI) were in the same range as other species treated with similar doses and time of exposure as Helianthus annuus and Glycine max (Mei et al., 2002), Salsola cali (Gardea-Torresdey et al., 2005) or Vigna radiata .
As shown in Cr tissue concentrations and bioconcentration factor, all genotypes of S. vulgaris studied here presented a significantly greater uptake of Cr(VI) than Cr(III). These differences have already been explained based on a different Cr uptake mechanism by the plant. Skeffington et al. (1976) carried out inhibitor studies with barley seedlings which demonstrated that Cr(III) and Cr(VI) do not share a common uptake mechanism. Cr(VI) is actively taken up in metabolically driven processes, in contrast to Cr(III) which is passively taken up and retained by cation exchange sites of the cell wall. This fact explained why plants take up more Cr(VI) than Cr(III). This was later confirmed by other authors (Zayed et al., 1998;Gardea-Torresdey et al., 2005).
The six genotypes of S. vulgaris presented TFs < 1 of Cr in all treatments. Chromium was mainly accumulated in roots and poorly translocated to aerial parts, as was previously reported by McGrath (1982) and Sharma et al. (1995). The authors proposed that poor translocation of Cr to the shoots could be due to sequestration of Cr in the vacuoles of the root cells to render it non-toxic. Furthermore, Han et al. (2004) and Arduini et al. (2006) showed that Cr is only translocated at toxic levels well above the dose applied in this study.
Among the effects shown to be caused by Cr toxicity to the plant are the detriment of dry weight in both root and shoots and leaf chlorosis. None of these symptoms were shown in S. vulgaris genotypes at the applied dose. It is remarkable that previous studies found biomass decreases in Z. mays, Solanum lycopersicum and B. oleracea (di Toppi et al., 2002) treated with Cr(VI) at 5 mg L -1 and in Sorghum bicolour at 50 µM . Similar doses also affected the chlorophyll contents of other species like Vigna radiata (Samantary, 2002), Salvinia maritima (Nichols et al., 2000) and Z. mays .

Effect of Cr(III) and Cr(VI) on Silene vulgaris genotypes
Chromium, due to its structural similarity with some essential elements, can affect mineral nutrition of plants in a complex way. Competition mechanisms for transport bindings in the plant resulting in a decrease in micronutrient uptake and translocation have also been described (Shanker et al., 2005). Results from Cr(VI) treatment showed Mn decrease in the shoots of genotypes SV30, SV36 and SV38. They also showed a decrease of Fe in SV36 and SV38, which in case of SV36 is accompanied by a greater accumulation of Fe in roots. These results agree with those of Gardea-Torresdey et al. (2004) and Gopal et al. (2009) that reported decreases of Mn and Fe contents in stems and leaves of cauliflower and spinach respectively. Gardea-Torresdey et al. (2004) found that Fe was significantly concentrated in the root of Cr(VI)-treated plants of Convultus arvensis as in genotype SV36. It must be taken into account that treatment with Cr(VI) increased the differences in micronutrient concentrations among genotypes. No effect has been found related to Cr(III) treatment.
The imbalance in Fe and Mn found in Cr(VI) treatment does not happen in all genotypes and is not translated in biomass or SPAD values reductions, suggesting that S. vulgaris presents a relatively high tolerance to Cr compared with other species treated with similar doses. The lack of toxicity symptoms could be related to the low dosage applied (close to hormesis). It should be taken into account that this experiment was conducted using Cr concentrations far less than those that cause toxicity because the objective was to evaluate the influence of the Cr speciation in the genotype development at environment concentrations.
Based on results obtained, there is a high level of variability among S. vulgaris genotypes. Each genotype studied presented differences in all variables studied in this work, except in Cr uptake. It could be difficult to choose the most efficient genotype in a possible treatment of Cr contaminated sites. Considering results of biomass and micronutrients concentrations, the common def inition of "eff iciency" for wild plants could be used. Efficiency is defined as the quantity of dry matter produced per gram of nutrient and it is simply the inverse of tissue concentration (Small, 1972). Given this definition, hermaphrodite genotypes and especially SV21 and SV36 would be more efficient than females because they present higher biomass and lower nutrient concentrations, especially in Cr(VI) treatment. Tissue concentration may be affected by processes like luxury consumption or substantial sto-rage. Therefore a more useful measure of efficiency might be respiration, photosynthetic or net assimilation rate Arduini et al., 2006).
In relation to SPAD values, like the measure of the state of chlorophylls, the hermaphrodites, and especially SV21, could be considered more efficient also because they presented higher values than females. In general, hermaphrodite genotypes showed similar behaviour in Cr uptake and tolerance, but SV21 seems to be the best candidate to be used in future assays. SV21 presented the highest BCF and it was the genotype that showed the highest concentrations of Cr in the aerial part without any alteration in its micronutrient balance. On the other hand, female genotypes were less efficient in all variables studied here, especially SV38. Both SV38 and SV19 seem to have different behaviour. SV38 translocated more Cr than other genotypes in relation with its poor uptake. This leads to a decrease in the levels of Fe, Mn and Zn in the aerial part. On the other hand SV19 did not accumulate great amounts of Cr and it did not show any micronutrients imbalance probably due to an exclusory mechanism. Some plants simply avoid toxicity by preventing uptake of the metal as have been reported by De Vos et al. (1991) in Cu-treated Silene cucubalus plants.
Comparison of physiological traits between genders is scarce for gynodioecious species, and the results are generally inconsistent, with some suggesting that females have higher photosynthetic rates (Caruso et al., 2003) and others suggesting the opposite (Schultz, 2003). At least three physiological mechanisms might underlie such sex differences: i) higher carbon demand in developing fruit; ii) greater nitrogen demand in developing seeds and iii) greater rate of active translocation of resources from leaves to developing fruit. All require a limited pool of resources (carbon, nitrogen, ATP) shared between the developing reproductive structures and supporting leaf tissue. As metal uptake is controlled by many physiological traits, it was expected to find sex differences in Cr uptake, but further studies will be necessary to elucidate the mechanisms by which the gender may influence metal uptake in S. vulgaris.
Uptake and effects of Cr are related to the dose and time of exposure and more studies are necessary to better understand tolerance and Cr uptake of S. vulgaris and to select the most effective genotype in field.
To our knowledge, this is the first report on the relative good performance of hermaphrodite and female S. vulgaris genotypes in Cr uptake and physiological 692 A. E. Pradas-del-Real et al. / Span J Agric Res (2013)  traits. S. vulgaris uptakes mainly the hexavalent form of Cr and accumulates it mainly in roots. This species presents relative high tolerance to Cr at the applied dose. This work confirms the high variability in physiological traits of S. vulgaris, even in genotypes from neighbouring areas. It seems that the treatment with hexavalent Cr increases the differences among genotypes; hence the use of cuttings from a homogeneous genotype seems to be an adequate method for the study of metal behaviour in this species.