Viability of using organic substrates according to toxicity tests and the antioxidant activities of tomato seeds and seedlings

T.R. Marcon, A. Rafagnin-da-Silva, R.O. Meira, L.P.C. Guedes, J.M. Corsato, and A.M.T. Fortes. 2020. Viability of using organic substrates according to toxicity tests and the antioxidant activities of tomato seeds and seedlings. Int. J. Agric. Nat. Resour. Plant growth can be directly influenced by the physical, chemical, and biological characteristics of their substrates. We tested the viability of using alternative substrates derived from agroindustrial residues by evaluating their effects on tomato (Solanum lycopersicum L.) germination and growth as well as the antioxidant activities in seeds and seedlings. The extracts were classified based on their principal carbon source: cotton waste (SA); sugarcane waste (SB); napier grass (SN); tree prunings (SP); and sawdust (SS). The chemical attributes of the substrates were analyzed, the physiological characteristics of tomato seed germination were determined, and the activities of the antioxidant enzymes superoxide dismutase (SOD), catalase (CAT) and peroxidase (POD) were measured in tomato seeds and seedlings. The extracts from tree prunings and sawdust showed the highest germination indices and root lengths, while the cotton and sugarcane waste extracts showed the lowest values for the same variables, with high pH and electric conductivity values indicating possible toxicity; increased activities of antioxidant enzymes needed to correct physiological imbalances were detected. The substrates deemed most suitable for seedling production were those derived from tree prunings and sawdust.


Introduction
The oxidative protection systems of plant cells are responsible for ensuring metabolic balance and avoiding damage during plant development caused by excess production of reactive oxygen species (ROS) (Silveira et al., 2010).
Among the mechanisms responsible for oxidative protection are antioxidant enzymes that function to eliminate ROS through the degradation of free radicals (Curvêlo et al., 2013). The main antioxidant enzymes are superoxide dismutase (SOD), catalase (CAT), and peroxidase (POD) (Silveira et al., 2010).
SOD is a key enzyme involved in the catalysis of singlet oxygen molecules (O 2 •-) into atmospheric oxygen (O 2 ) and hydrogen peroxide (H 2 O 2 ) (Alscher et al., 2002). CAT and POD enzymes, in turn, protect the cells from damage caused by excessive H 2 O 2 by converting it into H 2 O and O 2 (Silveira et al., 2010;Bhatt & Tripathi, 2011).
Many factors can contribute to increases in cell ROS levels, such as extreme temperatures, water stress, and high substrate concentrations of heavy metals and salts (Chagas et al., 2013). Excess substrate salts reduce the osmotic potential of plants and cause water stress; excessive accumulation of salts in plant tissues eventually result in toxicity, causing cell membrane damage and delays in seed germination -thus reducing commercial yields (Silveira et al., 2010).
Metabolic imbalances in plant cells caused by substrate toxicity will result in increased ROS production, resulting in severe damage to cell membrane permeability as well as DNA and protein synthesis (Silveira et al., 2010) and leading to saline stress (Chagas et al., 2013).
Plant growth is directly influenced by the physical, chemical, and biological characteristics of the substrate (Silva & Queiroz, 2014). Tomatoes (Solanum lycopersicum L.) are one of the most widely consumed vegetables globally and are widely produced in Brazil (tomatoes are often employed as bioindicators in substrate toxicity tests (Santos et al., 2011)).
Given the influence that substrates can have on seedling quality and plant enzymatic activities (Watthier et al., 2016), it is of the utmost importance to test different substrates using bioindicator species (such as tomatoes) to judge their suitability for commercial use. We sought to evaluate the qualities of alternative substrates derived from agroindustrial residues and to evaluate any toxicity in relation to seed germination and the antioxidant activities of tomato seeds and seedlings.

Materials and Methods
Our analyses were conducted between August and November/2015 at the Plant Physiology Laboratory and the Laboratory of the Analysis of Agroindustrial Residues at the Universidade Estadual do Oeste do Paraná, Campus Cascavel, Paraná State, Brazil.
To obtain the extracts, 5 g of the dry substrate was crushed and homogenized in 50 mL of distilled water (at 60 ºC) on a vibratory table for 20 minutes and then filtered (Zucconi et al., 1981). The tomato seeds were then sown to germinate in Petri dishes lined with filter paper moistened with 3 mL of the extract, with 25 tomato seeds per plate. The Petri dishes were incubated in a BOD germination chamber at 25 ºC under a 12hour photoperiod; germination was evaluated daily. At the end of seven days, we calculated the germination percentage (%) (Brasil, 2009), synchronization index, mean germination time (days) (Labouriau, 1983), and germination speed index (Edmond & Drapala, 1958). The germination index (%) (Zucconi et al., 1981) was calculated using the following formula: photoperiod; germination was evaluated daily. At the end the germination percentage (%) (Brasil, 2009), synchroniz time (days) (Labouriau, 1983), and germination speed inde The germination index (%) (Zucconi et al., 1981)  where where S t represents the percentage of germinated seeds in the treatment, Co t represents the root length in the treatment (cm), S c represents the percentage of control seeds germinated, and Co c represents the mean root length of the control seedlings (cm).
The physiological variables of tomato seedlings were evaluated in a one-factorial experiment employing a completely randomized design, considering the six treatments mentioned above (extracts from the organic substrates -SA, SB, SN, SP, SS, and the control -distilled water), each with four replicates.
The activities of the antioxidant enzymes of the tomato seeds and seedlings were examined in a 6 ˟ 3 two-factor arrangement, with the substrates designated as the first factor, while the times (2, 12, and 24 hours) when seeds were collected for analysis were designated as the second factor.

Statistical methods
The data from the one-factor and two-factor experiments were tested for normality (Shapiro-Wilk) and homogeneity of variance (Bartlett), and whenever necessary, transformations were performed. Analysis of variance (ANOVA) was then performed, and the means of the treatments were compared using the Tukey test. The treatments were also compared to the control using Dunnett's test. The statistical analyses were performed using R statistical software (R Development Core Team, 2017) at a 5% level of significance.

Results and Discussion
Analyses of the specific activities of the antioxidant enzymes (SOD, CAT and POD) in the first 24 hours of germination (Tab. 1) showed increased SOD and CAT activities after 12 hours of imbibition in the Control (T -distilled water) treatment group (Fig. 1A).
Thus, activation of the antioxidant defense systems occurs soon after seed hydration and reactivation of their quiescent metabolism, generating free radicals. Enzymes then act to minimize cell damage and maintain a balance between the production and elimination of ROS, thus ensuring seed germination (Barbosa et al., 2014). The seeds exposed to the cotton waste extract (SA) (Fig. 1B) demonstrated the same SOD enzyme pattern observed in the Control, with increased specific activity 12 hours after imbibition. CAT, however, showed greater activity within 2 hours of imbibition (similar to the activity shown by the control at 12 hours) and reduced activity after 12 and 24 hours of imbibition.
It could therefore be observed that CAT enzyme activity was required at the beginning of the germination process of seeds exposed to the cotton waste extract, presumably to eliminate hydrogen peroxide radicals.
The CAT enzyme acts as an effective catalyst for hydrogen peroxide (H 2 O 2 ) decomposition into H 2 O and O 2 (Dubey, 2011), and the increasing CAT activity observed within the first two hours of evaluation appeared relevant to repairing damage resulting from lipid peroxidation that mainly occurred during Phase I of seed germination (the Figure 1. Graphs demonstrating enzyme behaviors: superoxide dismutase -SOD (U1 mg 1 protein), catalase -CAT (H2O2 min -1 mg -1 of protein), and peroxidase -POD (µmol min -1 mg -1 protein) in tomato seeds in the first hours after germination (2, 12 and 24 hours) when exposed to different organic substrate extracts. water absorption stage); these repairs are facilitated by the increased catalase activity that ensures the elimination of free radicals (Umair et al., 2012). Tomato seeds exposed to cotton waste extract (Fig. 1B) showed reduced CAT activity in the first two hours after imbibition, but increased SOD activity, indicating compensatory effects of their activities in repairing oxidative damage (Barbosa et al., 2014). Tomato seeds exposed to sugarcane waste (Fig.  1C) and Napier grass (Fig. 1D) extracts showed almost constant SOD activity, with slight increases in CAT activity at 12 hours.
Seeds exposed to the tree prunings (Fig. 1E) and sawdust (Fig. 1F) extracts showed reduced SOD activity with increasing imbibition time. SOD is the first enzyme in the antioxidation defense system, a key enzyme that catalyzes the transformation of singlet oxygen (O 2 •-) into atmospheric oxygen (O 2 ) and hydrogen peroxide (H 2 O 2 ) (Alscher, Erturk & Heath, 2002).
Independent of the use (or not) of any type of organic substrate extract, POD activity remained practically constant during tomato seed metabolic reactivation.
POD enzyme activity can vary greatly, as it depends on numerous factors, such as the type of inducer, the timing of plant exposure, and the concentration of the inducer, so that standard behaviors are not generally observed (Bhatt & Tripathi, 2011).
CAT-mediated repair was apparently efficient in protecting the cells from oxidative damage due to excessive accumulation of H 2 O 2 (Silveira et al., 2010), as there were significant reductions in its activity within 24 hours of germination.
It is therefore possible to conclude that the chemical, physical, and biological characteristics of the extracts will affect the seedlings in terms of their antioxidant pathways, the enzyme activities of plant cell metabolism (Watthier, 2016) (Fig. 1), and the physiological processes related to tomato seed germination (Fig. 2). *Averages marked with asterisks represent statistically significant differences compared with the Control (T) by the Dunnett test (at a 5% level of significance). Similar letters indicate that the results do not differ from each other by the Tukey test (at a 5% level of significance). The seeds exposed to the sawdust extract (SS) demonstrated mean germination rates of 94% (G%), the best result for any of the extracts tested (Fig. 2F), although they were not significantly different from seeds exposed to sugarcane waste (SB), Napier grass (SN), and tree pruning (SP) extracts (Figs. 2C to 2E). They were, however, significantly different from the control (T) ( Fig.  2A) and the cotton extract (SA) (Fig. 2B), both of which showed 87% germination (considering a 5% level of significance).
The use of sawdust together with other agroindustrial wastes in composting processes has been shown to produce substrates that are amenable to lettuce and wild cabbage seedling production, with similar or better results compared to those obtained from commercial substrates, indicating its potential use in vegetable cultivation and corroborating our results for the sawdust-based substrate (which demonstrated a higher germination percentage than the control).
In addition to the sawdust extract substrate, the extracts of Napier grass, sugarcane waste, and tree prunings showed high average tomato seed germination percentage values, corroborating the results obtained for the germination speed index (GSI), with the highest mean values observed in seeds sown on the sawdust (Fig. 2F), tree pruning ( Fig. 2E) and sugarcane waste extracts (Fig. 2C) (with germination speed indices of 6.71, 6.38 and 6.27, respectively).
Satisfactory results were similarly reported when using sugarcane waste extracts mixed with sand and peanut husks in the cultivation of cherry tomatoes (Sindy cultivar) (Gonçalves et al., 2014).
The lowest mean germination speed index was observed with seeds sown onto the cotton waste extract (SA) (4.87) (Fig. 2B); this treatment also showed the highest mean germination time (4.71), differing statistically from all the other treatments (Fig. 2).
Tomato seeds sown onto the tree pruning (SP) (Fig. 2E) and sawdust (SS) extracts (Fig. 2F) also had low mean germination times (GMT) (approximately three and a half days), with no significant difference between them, the sugarcane waste extract (SB) (Fig. 2C) or the control ( Fig.  2A); the tomato seeds sown onto the SP extract demonstrated the greatest ability to germinate in the shortest time interval.
These results are apparently related to extract characteristics such as pH, electrical conductivity (Tab. 1), as their chemical properties are indicative of the concentrations of salt ions, which are linked to toxicity (Silveira et al., 2010).

The sawdust (SS) and tree pruning (SP) extracts demonstrated electrical conductivity values
The cotton (SA) and sugarcane (SB) wastes and Napier grass (SN) demonstrated electrical conductivity values above 3.5 dS m -1 (Tab. 1) and are therefore not recommended for cultivating vegetables due to osmotic-saline stress caused by excess salts (Gruszynski, 2002).
As the tree pruning (SP) and sawdust (SS) extracts showed the lowest electrical conductivity values, pH, and nutrient concentrations (N, P, K, Ca and Mg) when compared to the other extracts evaluated, our results corroborate another study that likewise observed lower values of those factors in a 100% sawdust extract (Silveira et al., 2010).
The germination index (GI%) is one of the most sensitive parameters for indicating substrate toxicity, and it is widely used to classify substrate quality, as it takes into account both root growth (responsive to even low substrate toxicity) and seed germination (responsive to high degrees of toxicity) (Zucconi et al., 1981).
On average, the lowest germination index values were observed in seeds sown onto substrate extracts based on sugarcane (62%) (Fig. 2C) and cotton (69%) wastes (Fig. 2B); their values differed statistically from the control (100%) and from the other treatments (Fig. 2) -indicating the probable presence of organic compounds with moderate toxicity, as substrates having germination index values between 60 and 80% will generally show moderate inhibition of seedling growth (Silva & Villas-Bôas, 2007).
Seeds exposed to the tree pruning (Fig. 2E) and sawdust extracts (Fig. 2F) demonstrated average germination index values of 116% and 110%, respectively, which were significantly greater than the control (100%) and those obtained from the other treatments. The germination index values above 100% observed in the pruning and sawdust treatments indicated the stimulation of seed germination, with beneficial effects on seedling development (Delgado et al., 2010).
No statistically significant differences were detected in terms of the synchronization indices (U) of the different treatments or the control.
The activities of SOD, CAT and POD enzymes in tomato seedlings exposed to the aqueous extracts of the different organic substrates are illustrated in Fig. 3.
The tomato seedlings showed similar enzyme activities in response to all the treatments in terms of SOD (Fig. 3A) and POD (Fig. 3C) activities (with no significant difference between them), indicating no stress response to the extracts during that phase of seedling development.
Catalase activity (CAT) (Fig. 3B), however, showed higher than average activity in tomato seedlings exposed to sugarcane waste (SB) and sawdust (SS) extracts, and they differed significantly from the other extracts and the control.
Although treatment with the sawdust extract (SS) was not significantly different in terms of catalase activity, it was apparently effective in controlling the oxidative imbalance in the cellular medium, as the seedlings exposed to that treatment showed the longest root length values and highest shoot growth values (Fig. 4), indicating that their development was not impaired.
Catalase activity apparently was not sufficiently effective in overcoming the stress induced in tomato seedlings by the sugarcane waste extract (SB), with no significant differences seen in regards to the other treatments (Fig. 3B).
Plants exposed to stress conditions concentrate their energy on antioxidant defense to overcome these stress (Sunaina & Singh, 2014). Such redirection of metabolic energy can impair full plant development and have effects such as reducing the lengths of their roots. Similar results have been observed with the roots of seedlings germinated on sugarcane waste extracts (SB), indicating a possible toxic oxidative-stress effect of that substrate.
Upon examining the shoot and root lengths of the tomato seedlings, it was observed that all the treatments resulted in average shoot lengths near 5 cm, which statistically differed from the control (3.36 cm) (Fig. 4).
The lower shoot lengths of tomato seedlings grown under control (distilled water) conditions can be explained by the lack of nutrients in that substrate.
The mean root lengths of the tomato seedlings were lower in the cotton (SA), sugarcane (SB) waste and Napier grass (SN) treatments (4.39, 3.66 and 4.81 cm, respectively), and they differed *The means all demonstrated statistically significant differences from that of the Control (T) according to the Dunnett test (at a 5% level of significance). The same letters above the columns indicate no significant differences from each other by the Tukey test (at a 5% level of significance). *Averages demonstrating significant differences from that of the Control (T) according to the Dunnett test (considering a 5% level of significance). The same letters above or below the columns indicate that the substrates did not differ one from each other by the Tukey test (considering a 5% level of significance).
statistically from the control (6.32 cm) as well as from the other extracts tested (Fig. 4); the high electrical conductivities (Tab. 1) of those extracts could explain these results. Substrates must have adequate pH and EC values (Kratz et al., 2013)or their high concentrations of soluble salts can cause "burning", necrosis, or poor root development [which apparently occurred with the use of extracts of cotton (SA), sugarcane wastes (SB) and Napier grass (SN).
The substrate produced from composting 100% Napier grass was found to demonstrate the poorest result in terms of the heights of the tomato plants (2.3 cm) when compared to a commercial substrate (8.4 cm) (Leal et al., 2007), corroborating the results of the present study in which the Napier grass extract resulted in the smallest root lengths.
A combination of 33% Napier grass with 66% crotalaria was found, however, to yield high height values in tomatoes, lettuce, and beet plants (Leal et al., 2007), enabling the use of Napier grass in mixtures with other organic materials.
The greatest root lengths of tomato seedlings were observed in the extracts derived from tree prunings (SP) and sawdust (SS) (7.01 and 6.44 cm, respectively), although they did not statistically differ from those of the control (6.32 cm); these values indicate the absence of toxic components in those substrates.
Waste material resulting from the pruning of urban trees could therefore provide material to form high-quality organic composts that are useful for seedling production while simultaneously minimizing environmental impacts and production costs.
Our results show that the chemical, physical, and biological characteristics of organic substrates can affect seedling quality (Watthier et al., 2016), in terms of the physiological processes of seed germination (Fig. 2), seedling shoot and root lengths (Fig. 4), and antioxidant pathways of plant cell metabolism (through changes in superoxide dismutase, catalase, and peroxidase activities) ( Fig. 1 and 3).
The main conclusions are the following. Extracts derived from tree prunings and sawdust showed the greatest positive effects on seed germination, potentially enhancing the growth and development of tomato seedlings in substrates that are free of toxicity and oxidative stress and, therefore, appear to be the most suitable for tomato seedling production.