Management of cotton bollworms Helicoverpa armigera and Earias vittella by entomopathogenic nematodes

The bollworm complex consisting of Helicoverpa armigera and Earias vittella is a major threat in cotton production globally. The habit of developing resistance to many insecticides including Bt transgenic cotton necessitates the exploration of an alternate strategy to manage bollworms. The entomopathogenic nematodes (EPN) Steinernema carpocapsae strain APKS2 and Heterorhabditis bacteriophora strains KKMH1 and TRYH1 at different concentrations of 1 × 109 infective juveniles (IJs)·hm−2, 2 × 109 IJs·hm−2, and 3 × 109 IJs·hm−2 in 500 L of water were evaluated as a foliar spray in fields naturally infested with H. armigera and E. vittella located at Eastern Block and and Cotton Research Farm of Tamil Nadu Agricultural University, Coimbaotre, India during October 2010–February 2011 and October 2011–February 2012, respectively. In general, all three tested EPN strains reduced the larval population of H. armigera and E. vittella; reduced square and boll damage; and subsequently increased cotton yield compared with the untreated control. The S. carpocapsae APKS2 is most effective against H. armigera whereas both S. carpocapsae APKS2 and H. bacteriophora KKMH1 were equally effective against E. vittella. The higher dose of 3 × 109 IJs·hm−2 was highly significant in the reduction of H. armigera larvae. However, the doses 2 × 109 IJs·hm−2 and 3 × 109 IJs·hm−2 were equally effective for E. vittella control. The S. carpocapsae APKS2 at 3 × 109 IJs·hm−2 caused a 62.2% reduction of H. armigera larvae, 34% reduction of square damage, 58.5% reduction of boll damage, and yielded 45.5% more seed cotton than the untreated control plots. In E. vittella infested field, S. carpocapsae strain APKS2 and H. bacteriophora strain KKMH1 at 2 × 109 IJs·hm−2 resulted in 60.6%~62.4% larva reduction, 68.4%~70.7% square damage reduction, 66.6%~69.9% boll damage reduction and 45.9% yield increase over the untreated control. The effective EPN treatments were comparable to the chemical insecticide chlorpyriphos 20% emulsifiable concentrate spraying at 2 mL·L−1. This study has shown that EPN have great potential in the management of the bollworm complex in cotton. Foliar spraying EPN strain S. carpocapsae (APKS2) at 3 × 109 IJs·hm−2 and S. carpocapsae (APKS2) or H. bacteriophora (KKMH1) at 2 × 109 IJs·hm−2 five times at10 days intervals are the best for the management of H. armigera and E. vittella, respectively.

2019). Chemical control was also a concern due to the health risk of the applicator, devastation of natural enemies, the resurgence of minor pests and environmental pollution (Seenivasan and Murugan 2011;Akhtar and Farooq 2019). In recent years, the focus has been on the use of genetically modified Bt (Bacillus thuringiensis) cotton. Bt cotton has been proven to be effective against bollworms. However, resistance to the insecticidal crystal proteins of B. thuringiensis in cotton bollworms has been documented (Akhurst et al. 2003;Wu et al. 2004) and the sustainable use of Bt cotton for bollworm management is becoming highly questionable. In this situation, the introduction of an alternate bio-control agent for cotton bollworm management will be a promising approach.
In recent years, entomopathogenic nematodes (EPN) Steinernema spp. (Stainerneamtidae) and Heterorhabditis spp. (Heterorhabditidae) are emerging as a potential biological control agent for the management of insect pests. They have symbiotic bacteria Xenorhabdus (Steinernematidae) and Photorhabdus (Heterorhabditidae) in their gut and release the bacteria into the insect haemocoel after invasion. Most biocontrol agents require days or weeks to kill the pest, but entomopathogenic nematodes with their symbiotic bacteria can kill insects in 12~24 h. EPN also have many desirable attributes such as high reproduction potential, easy to mass produce, safety to humans and other invertebrates, ability to reach insects in cryptic habitats, compatible with many insecticides, easy to deliver through the irrigation system to manage soil insects or as a spray for foliar pests (Koppenhofer et al. 2000). EPN are used to control a wide variety of economically important insect pests (Shapiro-Ilan et al. 2006). Many cotton pests like bollworm Helicoverpa zea (Boddie), fall armyworm Spodoptera frugiperda (Smith), beet armyworm Spodoptera exigua (Hübner), cabbage looper Trichoplusia ni (Hübner), tobacco budworm Heliothis virescens (Fabricius) and the pink bollworm Pectinophora gossypiella (Saunders) are susceptible to entomopathogenic nematodes (Gaugler 2001). The infectivity of EPN species Steinernema carpocapsae (Weiser), Steinernema riobrave (Cabanillas and Poinar) and Steinernema feltiae (Weiser) on H. armigera and Earias insulana was established earlier (Glazer 1997). Gassmann et al. (2006) demonstrated the successful field control of pink bollworm using an EPN species S. riobrave on cotton. However, the information on field control of the American bollworm H. armigera and spotted bollworm E. vittella by EPN is scanty. Preliminarily, 27 stains of EPN comprising of 16 S. carpocapsae, 3 Steinernema siamkayai (Stock, Somsook and Reid), 1 Steinernema monticolum (Stock, Choo and Kaya) and 7 Heterorhabditis bacteriophora (Poinar) strains were isolated from the cotton ecosystem .
Among them, the strains of KKMH1 (H. bacteriophora), APKS2 (S. carpocapsae), and TRYH1 (H. bacteriophora) showed the advantages such as more virulence against H. armigera and E. vittella, high reproduction potential, tolerance of heat and desiccation under laboratory conditions (Seenivasan and Sivakumar 2014;Nagachandrabose 2021). The objective of this study is to test the bio-efficacy of EPN strains stored at the Department of Nematology, Tamil Nadu Agricultural University, India such as KKMH1 (H. bacteriophora), APKS2 (S. carpocapsae), and TRYH1 (H. bacteriophora) against H. armigera and E. vittella under field conditions.

Experimental sites
Two field experiments were conducted during 2010-2012 at Tamil Nadu Agricultural University, Coimbatore, India. Experiment I was conducted from October 2010 to February 2011 at Eastern Block which lies 11° 12′ N and 77° 03′ E at an altitude of 426.74 m (Location I). Soil texture was a clay loam (33% clay, 22% silt, 30% sand), pH 8.4, cation exchange capacity (CEC) 11.1 cmol (p + ) kg −1 , organic carbon 34 g·kg −1 , electrical conductivity 0.17 dS·m −1 , and available N, P, and K in the soil were 225, 19.9, and 570 kg·hm −2 , with Ca, Mg, and Zn at 38, 14.6, and 1.29 μg·g −1 , respectively. The site had a history of H. armigera infestation and the larval population was maintained by planting cotton cv. MCU5 in the field without insecticide spray before experimenting. Experiment II was conducted from October 2011 to February 2012 at the Cotton Research Farm which lies 11° 50′ N and 77° 50′ E at an altitude of 426.72 m above mean sea level (Location II). Soil texture was a sandy clay loam (24% clay, 32% silt, 48% sand), pH 7.8, CEC 10.6 cmol (p + ) kg −1 , organic carbon 3.4 g·kg −1 , electrical conductivity 0.24 dS·m −1 , and available N, P, and K in the soil were 180, 19, and 485 kg·hm −2 , with Ca, Mg, and Zn at 32, 16, and 6 μg·g −1 , respectively. The site of the experiment II had a history of E. vittella infestation, and the cotton cv. MCU5 was used to maintain the larval population without insecticide spray before the experimental year. Weather parameters like monthly mean rainfall, monthly rainy days, maximum and minimum temperature, monthly average relative humidity, and monthly mean evaporation prevailed in the fields during experimental period were retrieved from the Department of Agro Climate Research Centre, Coimbatore, Tamil Nadu, India and summarized in Table 1.

EPN culture
Three EPN strains KKMH1 (H. bacteriophora), APKS2 (S. carpocapsae), and TRYH1 (H. bacteriophora), earlier isolated by baiting soil samples from different cotton fields of Tamil Nadu, India, were acquired from the Department of Nematology, TNAU, Coimbatore, India. The EPN were cultured in the laboratory on the last instar larvae of the rice grain moth, Corcyra cephalonica (Lepidoptera: Pyralidae) at (27 ± 4) °C as described by Kaya and Stock (1997). The infective juveniles (IJs) released from C. cephalonica larval cadavers were collected in sterile distilled water using a modified White's trap (Kaya and Stock 1997), sterilized in 0.05% formalin (v/v) solution and maintained on aerated sterile water in plastic tissue-culture flasks at 15 °C. The IJs younger than two weeks old were acclimatized at room temperature for 6 h before field application.

Experimental design
Experiments I and II consisted of three EPN strains APKS2, KKMH1, and TRYH1 applied at three rates of 1 × 10 9 , 2 × 10 9 , and 3 × 10 9 IJs·hm −2 . A positive control was a recommended chemical control practice for bollworm management as per Crop Production Guide such as chlorpyriphos 20% emulsifiable concentrate (EC) at 2 mL·L −1 (Anonymous 2021) and an untreated control were also maintained. The experiments were laid out in a randomized complete block design with four replicates for each treatment. Each plot was 20 m 2 (5 × 4 m) with 96 plants. Plots within blocks were separated by 1-m buffers and blocks were separated by 2-m buffers. Cotton (Gossypium hirsutum var. MCU 5, an H. armigera and E. vittella susceptible variety) was planted on 15 October 2010 in location I and 30 October 2011 in location II in rows at 30 × 60 cm spacing. Each plot comprised six rows with 16 plants in each row. The EPN were applied 5 times at 10 days intervals, at a rate of 1 × 10 9 , 2 × 10 9 , and 3 × 10 9 IJs·hm −2 in 500 L of water, using a knapsack sprayer (diaphragm pump, cone nozzle) with an operating pressure of 300 kPa. Glycerin 1 mL·L −1 was added as an anti-desiccant to EPN treatments 1~9 at 0.1% (v/v). Additionally, in order to lower the risk of desiccation, EPN was applied between 5:00 pm and 7:00 pm. The positive control chemical pesticide chlorpyriphos at 2 mL·L −1 was also applied five times like EPN spray. The first sprayings of EPN strains and chemical pesticides were started 40 days after planting (DAP). Separate sprayers were used for the EPN spraying and chemical spraying. In addition to the foliar application, the soil below the plant was also drenched with EPN suspension and chlorpyriphos at 100 mL·m −2 , so that final instar larvae and pre-pupal stages of the insect were also exposed to nematode as well as chlorpyriphos suspension (pupation is reported to take place in the soil). Uniform foliar applications were made to ensure that the whole plant was saturated with the EPN suspension. The topsoil (to a depth of 1 cm) of the root zone area of the plant (up to 10 cm diameter) was also uniformly wetted with the suspension from the sprayer. The field was irrigated 24 h before spraying, to increase the activity of the EPN.

Crop husbandry
Cultural practices were followed as per Crop Production Techniques of Agricultural Crops (Anonymous 2021). NPK fertilizers at 120, 80, and 50 kg·hm −2 were applied as recommended. Half amount of the nitrogen was applied at planting and the other half at 30 DAP, and all P and K were applied only at planting. The crop was irrigated at the rate of 4~5 L·m −2 ·day −1 of water by drip irrigation with 1 bar pressure. Since there was no severe incidence of any disease or other pests, no fungicides or other insecticides were applied as plant protection measures.

Assessment of larva, damage and yield
The H. armigera was the major bollworm in location I whereas in location II E. vittella was the major bollworm present. Data on the number of live larvae of each pest at seven days after each spray were recorded from 10 randomly tagged plants per experimental plot and means were reported. The damage to squares and bolls was assessed by counting the diseased and the total number of squares and bolls. Based on the numbers of damaged and undamaged squares/bolls, the percentages of damaged squares and bolls at each observation were calculated and the pooled mean was reported. Harvesting was done by handpicking. Seed cotton yield (kg·hm −2 ) per plot was recorded from each harvest and pooled.

Statistical analysis
The data were analyzed using SPSS 16.0 software (SPSS Inc., Chicago, IL, USA). Shapiro-Wilk test was performed to test the normality of data. To maintain homogeneity, larval population values were transformed to square root and cotton square damage/boll damages were converted to arcsine for analysis. The data were subjected to analysis of variance and means were separated by performing Tukey-Kramer test at P < 0.05. The transformed data were back-transformed for presentation.

Field experiment I
The population density of H. armigera in the cotton crop before treatments ranged 9.7~10.3 larvae per 10 plants (Table 2). Spraying three EPN strains at three doses effectively reduced the population of H. armigera. There was a significant difference in the number of bollworms per 10 plants among the three strains (APKS2, KKMH1, and TRYH1) and three doses (1 × 10 9 IJs·hm −2 , 2 × 10 9 IJs·hm −2 , and 3 × 10 9 IJs·hm −2 ) tested (P < 0.05). The larval population of H. armigera was significantly (P < 0.05) lower in plots sprayed with APKS2 whereas the untreated control plots supported more larval populations. Among three EPN dosages tested, EPN sprayed at 3 × 10 9 IJs·hm −2 in 500 L of water recorded significantly fewer H. armigera larval populations. When considering both strain and dosages, S. carpocapsae APKS2 at 3 × 10 9 IJs·hm −2 was significantly effective in reducing the H. armigera population by 62.2% compared with the control (P < 0.05). In the chlorpyriphos treated plot, the number of larvae decreased by 72.0% compared with the control. Another method to determine the bio-efficacy of EPN on the bollworm was by counting the damaged squares and bolls in the trial plots. Table 2 showed that both chlorpyriphos and EPN decreased the injury of H. armigera on the plants compared with the untreated control but there was a significant difference among EPN strains and dosages (P < 0.05). Among EPN strains, APKS2 reduced square and boll damage compared with KKMH1 and TRYH1. The higher rate of 3 × 10 9 IJs·hm −2 had significantly less square and boll damage than the lower rates. Square and boll injury were significantly low in Chlorpyriphos followed by APKS2 at 3 × 10 9 IJs·hm −2 (P < 0. 05).

Field experiment II
Results of the field trial showed that the EPN strains reduced the E. vittella larval populations, but there were differences in the efficacy among the EPN and rates tested ( Table 3). The population of E. vittella was significantly reduced in the plots sprayed with APKS2 and KKMH1 compared with TRYH1. Among three EPN dosages test, EPN spray at 3 × 10 9 IJs·hm −2 and 2 × 10 9 IJs·hm −2 recorded significantly fewer E. vittella larval populations whereas in low dosage (1 × 10 9 IJs·hm −2 ) larval populations were higher. When considering both strain and dosage, the APKS2 and KKMH1 strains applied at the two highest rates were similarly effective   at reducing E. vittella numbers. Square damage was significantly reduced in APKS2 and KKMH1 (Table 2). Considering the different rates of application of EPN, 3 × 10 9 IJs·hm −2 had significantly less square and boll damage than the 1 × 10 9 IJs·hm −2 and 2 × 10 9 IJs·hm −2 rates. However, the standard chemical spray (chlorpyriphos) recorded the lower larval population and square damage percentage compared with all the EPN strains at all three dosages sprayed. A similar trend was observed for boll damage percent. The yield was significantly high in APKS2 and KKMH1 over TRYH1 and the untreated control. EPN strains applied at 2 × 10 9 IJs·hm −2 and 3 × 10 9 IJs·hm −2 recorded significantly higher yield than the low rate applied at 1 × 10 9 IJs·hm −2 and the untreated control.
Yield followed a similar pattern as seen with the square and boll damage. Results showed that foliar spraying of EPN strains APKS2 or KKMH1 at 2 × 10 9 IJs·hm −2 and 3 × 10 9 IJs·hm −2 were considered prominent eco-friendly treatments for the control of E. vittella in cotton.

Discussion
Results of this study demonstrated the biocontrol potential of EPN to attack cotton bollworms H. armigera and E. vittella when applied as foliar spray under field conditions. Field application of EPN for insect pest control has already been attempted in several crops on multiple insect genera (Odendaal et al. 2016a, b;Goettig and Herz 2018;Helmberger et al. 2018;Platt et al. 2019;Steyn et al. 2019a, b;Jaffuel et al. 2020;Kim et al. 2021  In our field experiments, foliar application of all EPN reduced larval populations of H. armigera and E. vittella and provided control of square or boll damage caused by these insects. Our research findings add to the earlier reports on cotton against pink bollworm, P. gossypiella (Gassmann et al. 2006). The control of H. armigera and E. vittella larval due to EPN application was also reported by several investigators in crops such as beans (Glazer and Novan 1990), chickpea (Ali et al. 2008), corn (Ali et al. 2007, and pigeon pea (Vyas et al. 2002).
This study showed a 62.2% and 64.7% reduction of H. armigera and E. vittella, respectively, due to EPN application. In general, EPN efficacy was reported to be highly variable with low (˂10%) to high (˃60%) in field situations (Arthurs et al. 2004). Field application of EPN was not reported as promising in many experiments due to several factors such as lack of tolerance of EPN IJs to extreme temperature, UV radiation, desiccation, and relative humidity (Nyasani et al. 2008). Seenivasan and Sivakumar (2014) have established that the EPN strains used in this study namely KKMH1 (H. bacteriophora), APKS2 (S. carpocapsae), and TRYH1 (H. bacteriophora) have the potential to tolerate heat up to 40 °C for 2 h and to tolerate rapid and slow desiccation pressure to some extent. The adjuvant used in this study glycerol has been reported to reduce the negative effect of desiccation (Prabhuraj et al. 2005). Spraying EPN in the evening was demonstrated to minimize the effects of UV radiation (Ali et al. 2008). Hence, the improvement of field efficacy reported in this study was attributed to the use of thermo and desiccation-tolerant strains, the use of anti-desiccation adjuvant in spray liquid and judicious time of field application.
In this study S. carpocapsae strain APKS2 caused a significantly higher reduction of H. armigera suggesting that H. armigera is more susceptible to S. carpocapsae than H. bacteriophora. Similar results were also reported in other lab studies (Glazer 1992;Ali et al. 2007). Both S. carpocapsae strain APKS2 and H. bacteriophora strain KKMH1 controlled E. vittella equally effectively. The finding is in accordance with the earlier laboratory assay in which both strains caused 92.7% mortality of E. vittella (Seenivasan and Sivakumar 2014). EPN, even strains of the same EPN species, differ in their pathogenicity to different insect species (Nyasani et al. 2008). In this study, H. bacteriophora strain TRYH1 caused significantly more bollworm numbers with more boll and square damages than the H. bacteriophora strain KKMH1, which might be due to the different geographical origins and environmental adaptations of the two strains.
Biological control of H. armigera and E. vittella by EPN was not only influenced by the nematode species/ strains, but was also affected by the dose applied. For H. armigera control, an increase of dose from 1 to 3 × 10 9 IJs·hm −2 caused a decrease of H. armigera larva, its damage and increased yield in cotton plants. An increase in EPN concentration led to an increased H. armigera control, validating the results of previous studies (Ebssa et al. 2001(Ebssa et al. , 2004. This study also demonstrated that a high concentration of 3 × 10 9 IJs·hm −2 is needed to achieve better control of H. armigera. This finding is supported by earlier workers who have also shown the better performance of Steinernema mashoodi (Ahmad et al. 2015) and S. carpocapsae (Hussain et al. 2014) against H. armigera on chickpea; S. carpocapsae and S. feltiae against codling moth, Cydia pomonella on apple (Lacey et al. 2006); Steinernema wesieri against Plutella xylostella on cabbage (Nyasani et al. 2008); and H. indica against flush worm, Laspeyresia bipunctata on tea (Devrajan et al. 2010) at the dose of 3 × 10 9 IJs·hm −2 . Higher doses of EPN generally yield better insect control (Chen et al. 2003). Field doses of more than 2.5 × 10 9 IJs·hm −2 are commonly applied to ensure that a sufficient number of IJs is exposed to the target host insect for providing better control (Chen et al. 2003). For E. vittella control the dose of 2 × 10 9 IJs·hm −2 was equally effective as that of 3 × 10 9 IJs·hm −2 , which indicates that the dose of 2 × 10 9 IJs·hm −2 is optimum to get the required E. vittella control. Most probably, the smaller body size of E. vittella than H. armigera larvae is the reason that such low concentrations of IJs are required to obtain high control. Ebssa et al. (2001) reported similar results of differences in application dose with different sizes of target insects.

Conclusion
A foliar spray of S. carpocapsae strain APKS2 at 3 × 10 9 IJs·hm −2 is efficacious in controlling H. armigera up to 62.2% and reducing square (34.1%) and boll damage (58.5%) in cotton under field conditions. However, in E. vittella infested fields sprayed with S. carpocapsae strain APKS2 or H. bacteriophora strain KKMH1 at 2 × 10 9 IJs·hm −2 is recommended to achieve 60.6%~62.4% larva reduction, 68.4%~70.7% and 66.6%~69.9% reduction in square and boll damage, respectively. The spray application of EPN was proved to be useful in controlling H. armigera or E. vittella and increasing cotton yield 45.5%~45.9% higher than the untreated control. Despite the limitations associated with using EPN in the foliar arena, it was very encouraging that the effect of H. armigera or E. vittella control we obtained was comparable to that obtained using chemical insecticide. S. carpocapsae strain APKS2 and H. bacteriophora strain KKMH1 appear from our result to have the most potential, but future studies should focus on the performance of these EPN isolates under different climatic zones. In addition, the impact of these EPN strains on natural enemies