Sweet Corn Sentinel Monitoring Network
Galen P. Dively, Emeritus Professor, Department of Entomology
Bt corn and Bt cotton producing insecticidal toxins derived from Bacillus thuringi
ensis (Bt) are widely adopted in the U.S. to control lepidopteran pests, which have resulted in major benefits to growers and the general public. However, resistance evolution in corn earworm/bollworm (Helicoverpa zea) populations has become a major threat to the sustainability of these crops. To detect resistance and implement mitigation measures before control failures occur, industry registrants of Bt crops are required to annually monitor resistance in target pest populations. For H. zea, the monitoring approach consists of discriminating dose bioassays of larvae collected from non-Bt host plants in major production areas, and investigations of unexpected pest damage in Bt crop fields. So far, industry monitoring has not reported any significant changes in the baseline level of susceptibility to Bt toxins in H. zea populations. Yet, recent studies conducted in Maryland and several southeastern states report widespread field-evolved resistance in H. zea to all Cry toxins in Bt corn and Bt cotton.
More effective monitoring approaches are clearly needed to identify resistance early enough to enable proactive mitigation measures. Previous work in Maryland demonstrated that sentinel Bt sweet corn planted side-by-side with its non-Bt isoline can function as an in-field diagnostic screen to monitor changes in control efficacy and the phenotypic frequency of resistance to Cry and Vip3A toxins expressed in Bt field corn and cotton. Starting in 2017, a sentinel monitoring network has been tracking changes in H. zea susceptibility each year. Syngenta and Seminis companies provided sweet corn seed that is repackaged in Maryland and distributed to volunteer collaborators to establish field trials. Each trial involved Bt sweet corn hybrids (expressing Cry1Ab, Cry1A.105+Cry2Ab2, and Cry1Ab and Vip3A) planted side by side with their non-Bt isolines. All trials used the same ear sampling/data collection protocol to generate metrics showing differences in control efficacy between Bt and non-Bt, changes in the density and age of surviving larvae, and resultant kernel damage. Additionally, the network simultaneously monitored susceptibility changes and regional differences in European corn borer (Ostrinia nubilalis), fall armyworm (Spodoptera frugiperda), and western bean cutworm (Striacosta albicosta) populations.
During the past three years, collaborators established 41 trials in 2020, 52 trials in 2021, and 53 trials in 2022, located in 26 states (TX, LA, AL, MS, AZ, FL, GA, SC, NC, VA, MD, DE, PA, NJ, NY, NH, VT, OH, IN, IA, IL, NE, SD, WI, MN, MI) and 5 Canadian provinces (ON, QC, PEI, NS, NB). Trials in 11 states, ON and NS included multiple plantings at different times and/or locations. In MD, multiple plantings were established on research farms at Salisbury, Queenstown, Beltsville and Keedysville. Altogether, a total of 47,905 ears were examined to record the location and amount of kernel damage (recorded as cm2), larval density by instar stage, and signs of exit holes. Overall, 109 of the 146 trials reported high H. zea infestations and infestations and kernel consumption in more than 50% of the non-Bt ears. Highest infestations occurred at the southeastern and mid-Atlantic locations where
successful H. zea overwintering occurs, whereas the lowest infestations were mainly recorded in the North Central and Northeast states and Canadian provinces, where populations are sourced by migrant moths. Overall levels of H. zea infestations and larval densities in Cry expressing ears were slightly lower relative to the non-Bt isolines. O. nubilalis feeding injury (<1.5%) was recorded at only 30 of the 146 trials and associated with either missing or very few live larvae. Trials with consistent year-to-year O. nubilalis infestations were located where the surrounding landscape likely contained relatively less Bt field corn acreage. The absence of O. nubilalis infestations concurs with reports ofareawide suppression of populations due to the high adoption of Bt field corn. More importantly, no O. nubilalis survival or feeding injury was found in a total of 32,786 ears examined from the Bt sweet corn plots. S. frugiperda infested only 2.1% of all non-Bt ears sampled and at only 41 of the 146 trials. Ear infestations varied widely across trial locations and monitoring year, depending on the seasonal recruitment of S. frugiperda populations in the south and the frequency and direction of storm fronts that enabled migrant moths to reach northern locations. Highest ear infestations (16 to 29%) were consistently recorded in TX. Although data on this pest are limited, Cry1A.105+Cry2Ab2 appeared to be more effective against S. frugiperda than Cry1Ab. Ear infestations of S. albicosta larvae were uncommon, as only recorded in 1.4% of all non-Bt ears sampled and at 14 of the 146 trials, all located in NE, MI and the Canadian Provinces.
Phenotypic frequencies of resistance (PFR) for Cry1Ab, Cry1A.105+Cry2Ab2, and Cry1Ab+Vip3A were estimated as the ratio of mean number of surviving H. zea larvae per Bt ear to the mean number per non-Bt isoline ear. We assumed that any live larvae associated with kernel damage in a Bt ear indicates some level of resistance to the expressed toxins, which could result in mature larvae surviving to contribute resistance alleles in the next generation. The following summarizes the phenotypic frequencies for each single or pyramided Bt toxin compared to previous sentinel monitoring results.
Cry1Ab: The level of H. zea phenotypic resistance has significantly increased, since Cry1Ab sweet corn was commercially introduced in 1996. Based on trials each year in Maryland, overall PFRs averaged 0.28 during 1996-2003 and 0.64 during 2004-2016. Now, PFRs averaged 1.07 in 2022, compared to 1.06 (2021), 0.95 (2020), 0.76 (2019), and 0.85 2018) and 0.99 (2017). The percentage of damaged ears and kernel consumption per Bt ear, along with larval development delays, have remained about the same during the last three years. However, the most disconcerting finding is that 51% of the 2022 trials reported higher H. zea densities per Bt ear compared to densities per non-Bt ear (PFR>1). This larval density difference is the result of cannibalistic behavioral changes in larvae receiving sublethal doses of Cry1Ab. Although many young larvae feed together initially in an ear, they become aggressively cannibalistic once they reach the 4 th instar stage, and thus often only one mature larva is found in a non-Bt ear. Sublethal Cry intoxication is known to inhibit the cannibalistic behavior of late instars, allowing more larvae to feed and survive together in Bt ears. If this behavioral inhibition continues as resistance increases, it is possible that a Bt plant could produce more H. zea moths emerging compared to recruitment from a non-Bt plant. Obviously, this would have serious resistance management implications; however, it is still unknown as to how many larvae develop to pupate and successfully emerge as normal reproductive adults; and, more importantly, contribute resistant alleles in the next generation. Given this high frequency of phenotypic resistance and widespread decline in Cry1Ab control efficacy against H. zea, most field corn hybrids expressing only Cry1Ab have been phased out of commercial use and replaced by pyramided Bt hybrids expressing multiple toxins. However, one remaining concern is that the cross resistance of Cry1Ab with other Cry toxins may continue to reduce the durability of the pyramided hybrids.
Cry1A.105+Cry2Ab2: These pyramided toxins expressed in corn were registered for use in 2010 and initially provided effective control of H. zea. However, phenotypic frequencies have steadily increased since 2010, averaging 0.19 during 2010-2013 and 0.41 during 2014-2016. Sentinel network results continue to show further resistance development, with PFRs averaging 0.67 (2017), 0.93 (2018), 0.70 (2019), 0.89 (2020), 0.95 (2021), and 0.92 (2022). Thirty-two % of the trials since 2020 reported H. zea densities per Bt ear greater than densities in non-Bt ears. Over the last three years, there has been a slight but consistent increase in phenotypic frequency, kernel consumption, and older instars surviving per ear, suggesting that H. zea populations continue to develop higher levels of resistance to these Cry toxins. These findings concur with recent studies reporting high resistance ratios and increased field failure of the Cry1A.105 and Cry2Ab2 toxins in controlling H. zea infestations in Bt corn and Bt cotton. Unfortunately, the widespread H. zea resistance to Cry toxins make it difficult for any regulatory mitigation action by EPA or industry registrants to reduce or prevent further H. zea resistance to these toxins.
Cry1Ab and Vip3A: Previous studies in MD and MN during 2013-2016 found virtually no H. zea survival or damage in Vip3A-expressing sweet corn ears. However, sentinel trials starting in 2017 began to report larval survival with the expansion of the monitoring network to more southern locations. During 2017-2019, 0.72% of the 9,369 Vip3A ears sampled had minor tip damage associated primarily with 2th-3rd instars. Furthermore, results by year show a small but noticeable increase in the number and age of surviving larvae. Of the 20,312 ears sampled during 2020-2022, 156 ears (0.77%) had minor damage (<0.5 cm2, primarily on the tip), but only 25 of these ears (0.12%) were infested with a total of 82 live larvae (78% early instars). Trials reporting most of the ear damage and older larvae were southern locations (TX, LA, MS, AL, NC). Assuming all ears with live larvae were expressing Vip3A, the overall PFR estimated from trials conducted during 2020-2022 is 0.0044, based on a total of 82 larvae found in 20,163 Vip3A ears compared to 10,682 larvae found in 11,622 non-Bt isoline ears sampled. This level of phenotypic resistance is consistent with laboratory studies reporting an estimated frequency of 0.0065 for Vip3Aa resistance alleles in Texas H. zea populations. These studies and sentinel monitoring results show evidence of an increase in phenotypic resistance since 2017, indicating early signs of H. zea resistance to Vip3A, particularly in the southern locations. However, the Vip3A expressed in sweet corn, field corn and cotton still provides excellent ear protection against H. zea. Nevertheless, given the high levels of H. zea resistance to Cry toxins and their ineffectiveness against this pest, the redundancy control advantage of the pyramided Bt crops is likely compromised, which may lead to faster evolution of Vip3A resistance, especially when considering multiple generations of selection per season and increased use of Vip3A field corn and cotton to improve control of H. zea in the South. In short, the time for proactive measures for the Vip3A toxin is passing quickly, so we urgently need best management practices implemented to delay further Vip3A resistance.