Abstract: Background and Objective: The phytophagous insects select their host plants according to plant tissue nutritional quality. Thus, the objective of this study was to correlate the direct and indirect effects of phosphate fertilization on the nutritional status of the soybean crop and its relationship with the occurrence of insect pests. Materials and Methods: The effect of phosphate fertilization on soybean was evaluated using two phosphate sources, Single Superphosphate (SSP) and Monoammonium Phosphate (MAP), applied at five rates of P2O5. A sampling of insects initiated from stage V5 and was done weekly. Leaves were collected for nutrient analysis stage R1, in stage R9 was harvest was carried out. SSP or MAP phosphate fertilization in soybean affected the incidence of Chrysodeixis includens, Helicoverpa armigera, Elasmopalpus lignosellus and Euschistus heros. Results: The plants treated with MAP had the infestation reduced compared with plants treated with SSP. Higher contents of Cu and Fe in the leaf reduces the incidence of insect-pests, whereas the opposite occurred with Mn. The occurrence of E. lignosellus reduced soybean yield. Conclusion: Therefore, the source and rates of phosphorus in soybean fertilization change the concentration of macro and micronutrients in the leaves and affect the behavior and incidence of pest species.
INTRODUCTION
In the insect-plant interaction, it is of note that phytophagous insects select their host plants according to plant tissue nutritional quality1. Plants obtain nutrients such as carbon (C), hydrogen (H) and oxygen (O) from carbon dioxide (CO2) and water (H2O), apart from which, 14 nutrients are recognized as essential for growth and development. Plant nutrients are divided into macronutrients and micronutrients. Macronutrients include: Nitrogen (N), Phosphorus (P), Potassium (K), Calcium (Ca), Magnesium (Mg) and Sulfur (S); and micronutrients include: Iron (Fe), Manganese (Mn), Zinc (Zn), Copper (Cu), Boron (B), Molybdenum (Mo), Chlorine (Cl) and Nickel (Ni)2,3.
Studies about the effect of some nutrients on insect-plant interaction are abundant in the literature, while the effect of others, such as P on plant fertilization correlated with the occurrence of insect pests are scarce4. Phosphorus is the main structural component of nucleic acids and lipids in the cell membrane. In addition to participating in plant regulation, which involves signaling molecules, it is essential for carbohydrate transfer in leaf cells5. P deficiency in the plant, besides affecting the amount of energy (ATP), also affects the synthesis of DNA and RNA6. Therefore, plants with P deficiency have reduced amino acid synthesis and hence less preference by the insects3.
Insects require levels of N and P which are much higher than those present in host plants, thus, from the nutritional point of view, plants are considered suboptimal foods for growth, development and reproduction of insects4,7.
Plants treated with higher-than-recommended doses of P have the accumulation of phenols and terpenes changed. These compounds act in the protection against biotic stresses, protecting plants from microorganisms and repelling insects, as well as in the interactions between plants and herbivores8,9. Terpenes, including monoterpenes, sesquiterpenes and terpene polymers interfere with the neural transmission and block phosphorylation in insects. The effect of 1% P in the diet drastically reduced the growth and survival of the grasshopper Schistocerca americana (Drury, 1773) (Orthoptera: Acrididae)3.
Path analysis is a technique widely used to more consistently delimit the relationship between cause and effect since the unfolding of correlations is dependent on the set of variables studied10. From the variables, one can estimate the direct and indirect effects of the parameters on the response variable and, thus, identify the correlations between the leaf nutrient contents and the occurrence of pests10.
Moreover, the need to increase crop yields and to understand the interaction between P-treated plants and phytophagous insects are critical in high yield crops. Therefore, the present study sought to correlate the direct and indirect effects of phosphate fertilization on the nutritional status of soybean plants and its relationship with the occurrence of insect pests. The soybean fertilization alters the incidence and behavior of pest species in the soybean crop.
MATERIALS AND METHODS
Experimental location: The study was conducted from Nov. 2014-April 2015, in the experimental field of São João Farm (9°3'25.69" S; 44°33'12.89" W, 570 m altitude), Piauí-Brazil. The climate of the region is classified as Awa warm (tropical rainy with the dry season in winter and an average temperature of the warmest month above 22°C), according to the Koppen classification. Average rainfall ranges from 900 to 1,300 mm year–1 distributed between November and April, with an average annual temperature of 26.6°C, although temperatures around 40°C are common during the year.
The experiment was conducted in an area cultivated for the first time after removal of native vegetation (Cerrado), of flat, smooth undulating relief. The soil is classified as dystrophic Yellow Latosol, with sandy loam texture11. Table 1 shows the chemical and physical characteristics of the soil before installing the experiment.
Fertilization management and plot size: Two sources of P2O5 were used as a source of variation: Single Superphosphate (SSP) (18% P2O5, 16% Ca and 8% S) and Monoammonium Phosphate (MAP) (58% P2O5 and 12% N). Five rates of P2O5 (0, 100, 200, 300, 400 kg) were applied to both P2O5 sources. The fertilizer was broadcast and incorporated by disc harrowing, going twice over the field. Soybean cultivar BRS Sambaíba RR was sown in a row spacing of 0.45 cm and 12 plants per linear meter, with a population around 250 thousand plants ha, in mid-January 2015. The experiment was arranged in a randomized blocks design with four replications, totaling 40 experimental plots of 11×5 m (55 m2).
| Table 1: | Soil chemical characteristics before the experiment installation, Currais-PI |
| P: Phosphorus, K: Potassium, Ca: Calcium, Mg: Magnesium, Al: Aluminum, V%: Base saturation, OM: Organic matter, CEC: Cation exchange capacity | |
Insect-pest monitoring: The occurrence of the insects Spodoptera spp., Helicoverpa armigera (Hübner, 1808), Anticarsia gemmatalis (Hübner,1818), Chrysodeixis includens (Walker, 1858) (Lepidoptera: Noctuidae) and Euschistus heros (Fabricius, 1974) (Hemiptera: Pentatomidae) was monitored by the beating cloth sampling method, weekly, from the vegetative stage V5 to the crop maturation stage (vegetative stage V5 to V7 and reproductive stage R1 to R8)12. Damage by Elasmopalpus lignosellus (Zeller, 1848) (Lepidoptera: Pyralidae) was monitored by counting plants attacked per linear meter, taking two samples from each plot. The insects were collected, stored in bottles containing 70% alcohol and taken to the Agricultural Laboratory of the Federal University of Piauí (UFPI/CPCE) for identification, according to the dichotomous key of Carrano-Moreira13. Only caterpillars and stink bugs were considered in the samples. Finally, the phytosanitary treatment was performed according to the monitoring indication and farming operation's schedule.
Soil nutrient analysis after experiment implementation: Soil sampling was performed at the R1 stage of the crop12. Samples were taken from the 0-20 cm layer, with 5 samples from each plot, four between rows and one within the row, forming a composite sample. In the laboratory, the soil samples were air-dried and sieved through a 2 mm sieve. Nutrient levels were determined according to the methodology described by Ribeiro et al.14.
Leaf nutrient analysis: Leaves sampling was carried out during the R2 stage of the crop12. The third trifoliate leaf from the apex on the main stem was collected from a total of 30 plants per plot14. Leaves were oven-dried at 65°C and sent for macro and micronutrient analysis.
Leaf area and yield: Leaf area was determined by collecting one plant from the central area of each plot and estimating the area with a LI-3100c leaf area meter (LI-COR, Inc. Lincoln, NE, USA). The yield was estimated by hand-harvesting 3 m2 (three lines of 1 m in length) in the central area of each plot. The seeds harvested in each plot were processed, weighed and moisture was determined and corrected to 13%.
Statistical analysis: Two statistical methods were used; the first was a 2×5 double factorial analysis for sources and rates of phosphorus. When interactions were detected, the unfolding analyses were performed for significant factors at 5% probability by the F test.
The second method was a triple factorial analysis of variance, considering two qualitative factors (sources and rates) and a quantitative factor (insect occurrence). Insect fluctuation data were transformed to √ (x+0.5) to meet normality. The software application used to analyze the data was SAS®15.
Pearson correlation coefficients were estimated between the characters evaluated (total incidence of insects and nutrient levels in the soil and plant) and, from these correlations, the path coefficient was estimated according to Cruz et al.16, using the program Genes17. Correlation coefficients were interpreted according to Berman et al.18: 0-0.19, none/negligible, 0.2-0.39 low, 0.4-0.59 moderate, 0.6-0.79 marked, 0.81-1 high.
RESULTS
The analysis of variance showed a significant effect of the factors source and rate on leaf nutrient levels (Table 2). Leaf area showed no significant difference, however, the variable yield was influenced by both the P source (P = 29.98 p<0.001; CV: 7.44%) and the P2O5 rate (P = 17.75, p<0.001, CV: 7.43%).
The highest infestation of E. lignosellus was recorded up to 57 Days After Emergence (DAE), with about 0.5 plants per linear meter with symptoms of infestation observed at the highest P2O5 rate, 400 kg ha–1 (Fig. 1a). The population of leaf-cutting lepidopterans peaked at about 57 DAE.
| Table 2: | Nutrient content in soybean leaves for a two Source of P2O5, Currais-PI |
| Means followed by the same letters on the line do not differ by Tukey's test at 5% probability | |
| Fig. 1(a-e): | Fluctuation of the caterpillar population in the soybean crop as a function of the treatments assessed in the harvest year 2014/2015, MAP: Monoammonium Phosphate, SSP: Single superphosphate |
The population of A. gemmatalis peaked, with one insect per linear meter, when the P source used for the plants was SSP, in comparison with MAP (Fig. 1b).
Chrysodeixis includens showed a similar response to P2O5 rates for caterpillar incidence, with the population peak ranging from 8 to 16 insects per linear meter and the highest incidence was observed in plants that received 400 kg ha–1 of P2O5 (Fig. 1c).
| Fig. 2(a-b): | Fluctuation of Euschistus heros population in the soybean crop as a function of the treatments assessed in the harvest year 2014/2015 (a) MAP and (b) SSP |
MAP: Monoammonium Phosphate, SSP: Single superphosphate. *Significant interaction for P source, P2O5 rate and insect evaluation (P = 1.81, DF: 32, 360; p <0.05; CV: 17.85%) |
|
| Table 3: | Estimates of direct (DE) and indirect (IE) effects of nutrients on the correlations with insect occurrence |
| S: Solo, P: Plant | |
Considering the P sources, the incidence recorded in the treatments with SSP was about 6 insects per linear meter, greater than in treatments with MAP (Fig. 1d). H. armigera had similar behavior in relation to the two P sources, with about 2.5 insects per linear meter for plants fertilized with SSP and MAP (Fig. 1e).
The highest incidence of E. heros was observed at 89 DAE. At 9 DAE, the plants fertilized with MAP presented the lowest pest pressure at the rate 300 kg ha–1 P2O5, while those treated with SSP at the same P2O5 rate had the highest population peak, with about 0.8 insects per linear meter. However, the opposite effect was observed for the controls, considering the P sources (Fig. 2).
| Table 4: | Estimates of direct (DE) and indirect (IE) effects on the correlations between insects occurrence on the soybean yield |
The analysis of the Pearson's correlation between insects and all parameters evaluated showed a positive correlation between H. armigera and leaf Magnesium (0.43**) and A. gemmatalis and leaf sulfur (0.31**). Results of the path analysis showed the estimates of correlations between insect incidence and soil (S) and plant (P) nutrient content. Data of the path analysis showed means of predominantly large magnitude, with values between (0.421 and 1.450) positive and negative (Table 3).
From the data in Table 2, we can infer the effects between nutrients, those acting directly and those which associate with another nutrient, on the incidence of the insects A. gemmatalis, C. includens, H. armigera, E. lignosellus and E. heros. Thus, it is verified that Fe (both in the soil and in the plants) had direct and indirect negative effects of medium and large magnitude of -0.468 and -1.450, respectively. That is, Fe negatively affects the incidence of C. includens, H. armigera, E. lignosellus and E. heros.
In a similar manner, sulfur (S) presented direct and indirect negative effects in most of the results, with medium and large magnitude of -0.436 and -1.052, respectively, which may affect the incidence of C. includens, H. armigera., E. lignosellus and E. heros in plants (Table 3).
The direct and indirect effects of Mn were positive, predominant, with medium and large magnitude of 0.421 and 0.755, respectively, which favored the higher incidence of A. gemmatalis, C. includens, H. armigera, E. lignosellus and E. heros.
Cobalt (Co) presented positive indirect effects, with correlations of large magnitude and values between 0.597 and 1.249. Copper (Cu) had negative direct and indirect effects of medium magnitude, with values between -0,431 and -0,857, with a lower incidence of C. includens, H. armigera, E. lignosellus and E. heros. Zinc (Zn) showed indirect negative effects of high magnitude -0.877, with lower caterpillar incidence of A. gemmatalis, C. includens and E. lignosellus. However, Zn favored the incidence of E. heros stink bugs, with positive direct effects of 0.692 (Table 3).
The other nutrients or chemical elements in the soil or plants that had some direct or indirect correlation with the insects showed a small magnitude (between 0.2 to 0.3) and had small relevance to insect occurrence.
The path analysis in Table 4 shows that the occurrence of C. includens, A. gemmatalis, H. armigera and the stink bug E. heros had no significant effect on yield, not compromising the soybean crop. However, the E. lignosellus caterpillar had direct negative effects of medium magnitude -0.433. The other insects presented low direct and indirect correlations, almost nil, for yield, including C. includens with low positive direct correlation of 0.285 and A. gemmatalis also with low correlation of 0.051. H. armigera and E. heros also presented low correlations of 0.048 and 0.002, respectively.
DISCUSSION
In theory, herbivorous insects prefer host plants that will maximize their growth rate while avoiding host plants that lead to fitness decrease19. This may result from a number of factors, including a mistake on the female’s part and the differential fitness among the insect’s life stages. The results show that the source and rates of phosphorus affect the occurrence of different insect species throughout the soybean crop cycle. Occurrence of E. lignosellus was altered as a function of the phosphorus rates and the occurrence of C. includens and E. heros was affected by both the source and the increase in the rates of phosphorus.
The highest incidence of rice leaf folder, Cnaphalocrocis medinalis (Guenée, 1854) and stem borer Scirpophaga incertulas (Walker, 1863) (Lepidoptera: Pyralidae) was recorded from the basmati variety Punjab Bas-2. The incidence of leaf folder and stem borer increased with an increase in nitrogen level20. Plutella xylostella (Linnaeus, 1758) (Lepidoptera: Plutellidae) was reported to have a feeding preference for nitrogen-rich cabbage plants21. Plants with high nitrogen content produce bright green leaves that may attract insects, in addition to a higher concentration of proteins and free amino acids.
Phosphorus had a positive effect on various aphid performance parameters. High phosphorus concentrations decrease the incidence of whitefly in cotton3. In this study, is showed the direct and indirect effects of different nutrients in soybean leaf on the occurrence of different insect-pests of this crop. The plants grown in the treatments fertilized with MAP presented higher nitrogen levels, while those fertilized with SSP presented higher sulfur levels. These results are related to the composition of the P sources used, with SSP containing 8% of S and MAP containing 18% of N. Phosphorus fertilization improves micronutrient absorption since P can bind to micronutrients (P-Zn, P-Fe, P-Cu, P-Mn, P-Mo and P-B)22. Macronutrients and micronutrients such as Ca, Zn and S also reduced pest populations, which was also observed by Santos et al.23. Plants require secondary nutrients calcium, sulfur and magnesium in large quantities in the same way as nitrogen, phosphorus and potassium. Micronutrients, on the other hand, are necessary in small quantities and necessary for crop development. Deficiency of micronutrients can impair crop yields, lead to low absorption of other nutrients and problems in plant structure.
Excessive and/or inappropriate use of fertilizers lead to nutritional imbalances and lower insect resistance24. Nutritional balance is required to give the plant a certain level of insect resistance, altering its physical and biochemical properties. Chatterjee et al.25 associated the reduction of white fly incidence in tomato with the use of inorganic fertilizers combined with vermicompost and biofertilizers.
In this study, the managed to observe some patterns such as: the presence of H. armigera had a positive correlation with leaf area and leaf number, while the presence of C. includens had a positive correlation with leaf number. This may be linked with overcompensation that occurs when herbivory has a beneficial effect. In these cases, herbivory can stimulate the plant to increase efficiency in photoassimilate conversion and thus the less competition between leaves that acted as sinks and reproductive structures can increase the yield of flowers and pods in soybean26,27.
At lower or zero P rates, the incidence C. includens was lower, which may be related to the amount of nutrient required by this species6. The reduction in amino acid synthesis results in less preference by insects for P-deficient plants, reflecting the negative effect on feeding such as the case with A. gemmatalis, C. includens and E. lignosellus6.
It is necessary to study the physiology of the plant subjected to these treatments, to verify which changes occur in the plant that interferes with the behavior of insects
CONCLUSION
The source and rates of phosphorus in soybean fertilization alter the concentration of macro and micronutrients in the leaves and affect the incidence and behavior of pest species throughout the soybean crop cycle. In this study, the managed to observe some patterns such as: the presence of H. armigera had a positive correlation with leaf area and leaf number, while the presence of C. includens had a positive correlation with leaf number. This may be linked with overcompensation that occurs when herbivory has a beneficial effect. The source and rates of P, affect the incidence and behavior of pest species throughout the soybean crop cycle. Thus collecting information on soybean fertilization, may explain the insect's behavior in the plant.
SIGNIFICANCE STATEMENT
This study discovers that the effect of some nutrients on insect-plant interaction are abundant in the literature, while the effect of others, such as P on plant fertilization correlated with the occurrence of insect pests are scarce. This study will help the researcher to uncover the critical areas of plant-insect interaction and plant nutritional deficiency. Studies that still require data collection to understand some interactions.
ACKNOWLEDGMENTS
The authors would like to thank the Piauí Research Support Foundation (FAPEPI), the Higher Education Personnel Improvement Coordination - Brazil (CAPES) and CNPq for the financial support provided for this research.