Comparative Toxicity of Helicoverpa armigera and Helicoverpa zea (Lepidoptera: Noctuidae) to Selected Insecticides

Until recently, the Old World bollworm (OWB) Helicoverpa armigera (Hübner) and the corn earworm Helicoverpa zea (Boddie) (Lepidoptera: Noctuidae) were geographically isolated. Both species are major pests of agricultural commodities that are known to develop insecticide resistance, and they now coexist in areas where H. armigera invaded the Americas. This is the first study to compare the susceptibility of the two species to conventional insecticides. The susceptibility of third instar H. armigera and H. zea larvae to indoxacarb, methomyl, spinetoram, and spinosad was determined using a diet-overlay bioassay in a quarantine laboratory in Puerto Rico. Mortality was assessed at 48 h after exposure for up to eight concentrations per insecticide. Spinetoram exhibited the highest acute toxicity against H. armigera, with a median lethal concentration (LC50) of 0.11 µg a.i./cm2, followed by indoxacarb and spinosad (0.17 µg a.i./cm2 for both) and methomyl (0.32 µg a.i./cm2). Spinetoram was also the most toxic to H. zea (LC50 of 0.08 µg a.i./cm2), followed by spinosad (0.17 µg a.i./cm2) and methomyl (0.18 µg a.i./cm2). Indoxacarb was the least toxic to H. zea, with an LC50 of 0.21 µg a.i./cm2. These findings could serve as a comparative reference for monitoring the susceptibility of H. armigera and H. zea to indoxacarb, methomyl, spinetoram, and spinosad in Puerto Rico, and may facilitate the detection of field-selected resistance for these two species and their potential hybrids in areas recently invaded by H. armigera.


Introduction
The noctuid moths Old World bollworm (OWB), Helicoverpa armigera (Hübner, 1809) and corn earworm, Helicoverpa zea (Boddie, 1850), are major lepidopteran pests attacking crops worldwide. The latter is restricted to the New World and attacks more than 120 host species in 29 plant families [1][2][3][4]. Helicoverpa armigera feeds on more than 180 hosts in 70 plant families, and it is widely distributed Service permit number P526P-15-04600). No information is available on the previous exposure of this population to insecticides.
The H. zea colony was started with larvae collected in Isabela, Puerto Rico, on unsprayed pigeon pea on the 11 November 2015 (18 • 0 34 S; 66 • 53 33 W), and it was replenished multiple times with additional specimens collected from corn in the same area to maintain colony vigor. There are no large row crop operations in this area of Puerto Rico, and the H. zea individuals were collected from unsprayed experimental plots at the University of Puerto Rico-Isabela Agricultural Experimental Station.

Species Confirmation
Morphological and molecular tools (real-time PCR analysis) were used to determine species. Male genitalia were extracted and analyzed following the methods described by Brambila [44]. Males and females of both Helicoverpa species were identified by real-time PCR with specific primers for the internal transcribed spacer 1 (ITS1) region, as well as the sequencing of cytochrome c oxidase subunit I (COI) and Cytb regions [14]. Both populations were maintained at the CEQIS and were shared with other laboratories to be used as a reference for future studies and screening of other populations.

Rearing Procedure
Larvae were reared individually in 30 mL transparent plastic cups containing an artificial moth diet (Frontier Agricultural Sciences, Product # F9630B, Newark, DE, USA) until pupation. Pupae were transferred to Petri dishes with autoclaved vermiculite (Vigoro ® , Lake Forest, IL, USA). One day before adult emergence, pupae were placed in white 5-gallon (19 L) plastic buckets (15.6" × 11.8") with lids lined with cheesecloth (DeRoyal, BIDF2012380-BX, Powell, TN, USA) that served as an oviposition substrate. Adult moths were provided with a 10% sucrose solution. The oviposition substrate was replaced daily and stored in 3.8 L Ziploc ® (Racine, WI, USA) bags with thin strips of diet. Third instar larvae were transferred to cups with an artificial diet (described above). Colonies were maintained at 25 ± 2 • C, 65 ± 9% relative humidity (RH), and a 14:10 light:dark (L:D) photoperiod, with the exception of female pupae. They were placed in incubators (Sanyo ® , MLR-351H, New York, NY, USA) set at a lower temperature (22 ± 1.5 • C, 75 ± 4% RH, and a 14:10 L:D photoperiod) to synchronize their emergence with that of the adult males [45,46]. Prior to this study, H. armigera and H. zea were reared for 11 and 24 generations, respectively.

Bioassays
The same artificial diet used to maintain the colonies was used in the bioassays. Bioassay cups placed on 30-well trays were filled with 1 mL of diet per well (4.3 cm top diameter, 3.3 cm bottom diameter, and 3 cm height). A 100 ppm A.I. stock solution of each insecticide was serially diluted to obtain the test concentrations. Triton X-100 (0.1%, Sigma Aldrich, MO, USA) was used as a surfactant to obtain a uniform distribution over the diet surface. The control treatment was composed of distilled water and a surfactant. Up to eight concentrations of each insecticide were tested for each species. The insecticides were applied to the diet surface with a replicating pipette, ultimately delivering 140 µL per cup (equivalent to 20 µL per cm 2 ). The diet surface area in each cup was 7.0 cm 2 . After a 30 min drying period, one H. armigera or H. zea third instar larva was transferred to each cup using a fine paintbrush (AIT Art ® , 10/0, Danbury, CT, USA). The cups were closed with a perforated lid that allowed for gas exchange and stored in a climate chamber (25 ± 2 • C, 65 ± 9% RH, and a 14:10 L:D photoperiod). The bioassays were repeated four times for each species, and each replication consisted of 30 larvae per concentration. Larvae were inspected after 48 h and recorded as dead if there was no movement when gently touched with a fine paintbrush.

Statistical Analysis
Mortality data were subjected to Probit analysis (PROC PROBIT, SAS Institute 2000) [47] to estimate the lethal concentrations (LC 50 and LC 90 -insecticide concentrations (µg a.i./cm 2 ) required to kill 50% and 90% of larvae, respectively, in 48 h) and their confidence intervals (CIs). A likelihood test was conducted to determine whether the response of the two species differed significantly in either slope or intercept [48]. Pairwise comparisons were performed, and significance was declared when CIs did not overlap [48,49]. Significant differences among slopes were determined through a likelihood ratio test for parallelism and equality [48]. For each insecticide, the tolerance ratio (TR) was determined by dividing the LC 50 and LC 90 of the more susceptible species by the corresponding parameter of the other species.

Results
The Indoxacarb-induced mortality of third instar larvae for both H. armigera and H. zea was concentration-dependent (Table 2). Concentrations ranging from 0.0051 to 1.60 µg a.i./cm 2 caused 4-100% mortality. The LC 50 of indoxacarb on H. armigera was 0.17 µg a.i./cm 2 , and the LC 90 was 1.70 µg a.i./cm 2 ; they were slightly higher for H. zea at 0.21 and 2.64 µg a.i./cm 2 , respectively. The tolerance ratios for the LC 50 and LC 90 values were similar at 1.24 and 1.55-fold, with H. zea exhibiting a slightly lower susceptibility. The response for both species were also statistically similar, as indicated by the 95% fiducial limits overlap.
Methomyl produced the greatest variation in response between the species (Table 2). Concentrations from 0.0051 −3 to 2.88 µg a.i./cm 2 caused mortality ranging from 5% to 100% in both species; however, the LC 50 and LC 90 for H. armigera were 0.32 and 3.20 µg a.i./cm 2 , respectively, which were much higher than those for H. zea (0.18 and 1.88 µg a.i./cm 2 , respectively). The tolerance ratios were lower than 1.8-fold, indicating a similar response of these species to methomyl. Spinosad and spinetoram also induced high mortality for both species (Table 2). Concentrations ranging from 0.0051 to 1.60 µg a.i./cm 2 caused 3-100% mortality. The spinosad LC 50 value for both species was 0.17 for µg a.i./cm 2 . The spinetoram LC 50 values were 0.11 and 0.08 µg a.i./cm 2 for H. armigera and H. zea, respectively. In contrast, a lower LC 90 of spinosad was detected for H. armigera

Discussion
This is the first study to compare the response of H. armigera and H. zea to broad spectrum and selective insecticides. Earlier studies with biological and chemical insecticides have evaluated the two species separately due to their former geographic isolation [1][2][3][4][5][6][7][8][9][10]. Among the insecticides tested in this study, high levels of resistance of H. armigera to methomyl were reported in Pakistan [50][51][52], India [53,54], and Greece [55]; in contrast, low levels of resistance were reported in populations from Spain and Turkey [56,57], and no resistance was reported in invasive populations of H. armigera in Brazil [58]. In the U.S.A., a low frequency of resistance alleles to methomyl in H. zea populations from Virginia was reported [59]. However, Vemula et al. [60] found variations in the tolerance of H. zea to methomyl between bean crop seasons in Texas and New Mexico.
Populations of H. armigera from Australia were highly susceptible to indoxacarb, with toxicity ratios between 1.2 and 3.5 among several populations. The most tolerant strain had an LC 50 value of 0.518 mg/mL [61]. However, follow-up studies identified field populations with up to a 198-fold resistance [62]. In addition, a population of H. armigera from China subjected to 11 generations of selection to indoxacarb resistance decreased its susceptibility by 4.43-fold (LC 50 increased from 5.93 to 26.25 mg L −1 ) [63]. Helicoverpa assulta Guenée, another related species, also demonstrated resistance to this pesticide in China [64]. In south-eastern U.S.A., first instar larvae of H. zea under high indoxacarb pressure were very susceptible, with LC 50 values ranging from 1.05 to 1.54 ppm using diet overlay bioassays [65], and no evidence of resistance was found in cotton fields in the U.S.A. [66].
The use of spinosyns, which include spinosad and spinetoram, to control Helicoverpa spp. has increased in recent years. Spinetoram has been reported to have high efficacy against Helicoverpa species under field conditions [67,68]. Interestingly, spinosad resistance is associated with a reduced fitness, as reflected in prolonged egg, larval, and pupal periods and decreased pupal survival and overall fecundity [66]. However, a remarkable variation in H. armigera population susceptibility, especially to spinosad, was reported in Pakistan [69], and populations in China developed more than 20-fold resistance after 15 generations [70]. In contrast, low levels of resistance to spinosad were reported in Pakistan [71] and populations of H. armigera from two intensive cotton growing areas in India [72]. The results in our study are similar to Pereira's [73], who found two-fold variations in the susceptibility to spinosad among different populations of H. armigera in Brazil, thus suggesting low levels of resistance; unfortunately, after a few years of exposure, resistance increased, resulting in a 22% survival (LC 99 ). Helicoverpa zea susceptibility to spinosad is also variable by population. In the U.S.A., high LC 50 values were obtained for H. zea third instar larvae [65]; the authors suggested that this was due to the reduced rates used in cotton systems. In contrast, López Jr. et al. [74] indicated that this pesticide is highly effective against H. zea adults in insecticide-baited traps in the southern U.S.A.
Our results indicated that spinetoram is highly toxic to both Helicoverpa species. This insecticide is considered an important alternative for controlling Helicoverpa pests, especially for Cry1Ac-resistant populations [68]. Xie et al. [67] found spinetoram to be effective against H. armigera, inducing high mortality rates and sublethal effects similar to spinosad in H. armigera populations from China [66]. Visnupriya and Muthukrishnan [75] also reported low LC 50 values for spinetoram on H. armigera, ranging from 1.94 to 5.20 ppm. There have been few reports of Helicoverpa species resistance to spinetoram; nevertheless, if usage patterns and exposure to sublethal concentrations of spinetoram increase, selection for resistance to it is also likely to rise [68].
The diet overlay bioassay is a valuable tool for monitoring changes in susceptibility to insecticides in Helicoverpa species [66,76]. This bioassay has been used to evaluate a range of insecticides (permethrin, thiodicarb, chlorfenapyr, cypermethrin, di-ßubenzuron, cyanamid, emamectin, benzoate, and spinosad) on larvae of Diatraea saccharalis (Fabricius, 1794), H. armigera, H. zea, and Spodoptera frugiperda (Smith, 1797), among other lepidopteran pests [77][78][79][80]. The overlay diet bioassay may better simulate the field application of insecticides than commonly used techniques such as a diet-incorporation bioassay. It allows for the even distribution of insecticides over the diet surface, thus simulating field deposition of insecticides over the surface of the larval feeding substrate. There is a caveat, as Roush et al. [81] pointed out that laboratory colonies are formed from a small number of individuals that lack the high frequency of alleles that confer field populations with resistance to insecticides, so results for laboratory populations could differ from field-selected resistance.

Conclusions
The recent establishment of H. armigera populations in the H. zea native range, as well as the potential for hybridization of these two species, may form a Helicoverpa complex in the Western Hemisphere. Monitoring the susceptibility of this complex to insecticides is essential for implementing IRM programs to prevent control failures. We present data on an invasive population of H. armigera from Brazil and a population of H. zea from Puerto Rico that showed similar responses to indoxacarb, methomyl, spinetoram, and spinosad. These populations can be used as a reference for future studies to develop baselines for monitoring field-selected resistance in Helicoverpa species. Funding: This research was funded by USDA APHIS-UF Cooperative Agreement No. 16-8130-0744-CA and APHIS-UPR AP17PPQS&T00C189. The findings and conclusions in this preliminary publication have not been formally disseminated by the U.S. Department of Agriculture and should not be construed to represent any Agency determination or policy. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the USDA; USDA is an equal opportunity provider and employer.