Revisiting Classical Biological Control Through the New Associations Between the Invasive Pest Spodoptera frugiperda J.E. Smith (Lepidoptera: Noctuidae) and Local Parasitoids

1. Introduction

The appearance of invasive pests is a problem that requires attention because it may threaten agriculture and the variety of local species [1] and cause biotic homogenization [2]. Spodoptera frugiperda J.E. Smith (Lepidoptera: Noctuidae) is an invasive pest from America. It has become a new pest in Indonesia since early 2019 [3]. This pest has a wide range of distribution. It has now spread to 32 provinces in Indonesia, including Sumatra [4], Java [5], Kalimantan [6], and Sulawesi [7]. Spodoptera frugiperda infestations should be severely considered since a population of 0.2 to 0.8 S. frugiperda larvae/plant can decrease maize productivity by 20 to 50% [3]. Reports of damage due to S. frugiperda have been reported in several countries, such as Ethiopia and Kenya (32 - 47%) [8], Zimbabwe (32 - 48%) [9], Ghana (22 - 67%) [10], and Indonesia (60%) [11]. Spodoptera frugiperda has reportedly replaced the position of Asian corn borer Ostrinia furnacalis Guenée (Lepidoptera: Crambidae) as the primary pest of maize in China [12]. Rizali et al. [2] mentioned that the presence of S. frugiperda significantly decreases the intensity of attack of other lepidopteran pests and indirectly causes negative effects on the diversity of their natural enemies (particularly predators) in different maize fields in Indonesia.

Several studies have been reported since S. frugiperda was first reported in Indonesia. The studies were mostly focused on the presence/absence, diversity, infestation level, and ecology of S. frugiperda [5,13,14,15,16]. Research on the performance of local parasitoid species in Indonesia in parasitizing S. frugiperda on a lab scale has even been reported [17,18]. Surveys on the infestation level of S. frugiperda, the association between S. frugiperda with local parasitoids, and its associated parasitism rate have also been carried out but limited within a specific period [2,5,19,20,21,22,23,24,25,26,27,28]. Preliminary research in Yogyakarta, Indonesia, showed a low attack rate from local parasitoids toward S. frugiperda [29]. Thus, there is scattered information regarding the possibility of a new association between S. frugiperda and local parasitoids.

According to classical biological control theory, invasive pest control can be carried out by importing parasitoids from the original habitat of the pest, which have been proven as potential biological agents [30]. For example, the introduction of Rodolia cardinalis Mulsant (Coleoptera: Coccinellidae) to control the cotton aphid, Icerya purchasi Maskell (Hemiptera: Margarodidae) in California [31], and the introduction of Urophora affinis Frauenfeld and U. quadrifasciata Meigen (Diptera: Tephritidae) to control spotted knapweed Centaurea maculosa Lam. and C. diffusa Lamarck in western North America [32]. In Indonesia, the introduction of biological control agents was also carried out several times, such as the introduction of Curinus coeruleus Mulsant (Coleoptera: Coccinellidae) from Hawaii to control Heteropsylla cubana Crawford (Hemiptera: Psyllidae) on mimosoid tree Leucaena leucocephala [33], the introduction of Anagyrus lopezi De Santis (Hymenoptera: Encyrtidae) from Thailand to control cassava mealybugs Phenacoccus manihoti Matile-Ferrero (Hemiptera: Pseudococcidae) [34], and the introduction of Cecidochares connexa Macquart (Diptera: Tephritidae) from Columbia to control the invasive species - siam weed Chromolaena odorata L. [35].

The latest example provides important information that 28 years after introduction, C. connexa has been reported to be associated with several local parasitoid species [36,37,38,39]. Another example is the association between leaf miner Liriomyza spp. and local parasitoids in Indonesia. Liriomyza spp. is an invasive pest found in Indonesia from America in the 1990s [40]. Liriomyza spp. was reported to be associated with 11 parasitoid species in 2000 [41], and in 2022, Liriomyza spp. was associated with 18 local parasitoid species [42]. These findings support Hokkanen and Pimentel’s old hypothesis [43] on the possibility of new associations that can be formed between herbivores and their local natural enemies. Therefore, this research aimed to study the diversity of parasitoids associated with S. frugiperda for three years after it was first discovered in Indonesia. This is important to study as an effort to prepare local biological agents potentially to be used in controlling S. frugiperda in the future.

2. Materials and Methods

2.1. Sampling Location Determination

Sampling locations were determined using a stratified random sampling method in four central districts in Yogyakarta, including Sleman, Bantul, Kulon Progo, and Gunung Kidul. Parasitoid sampling activities were carried out in all sub-districts in each district. A total of 2-3 villages were selected from each sub-district as sampling points. From each village, a maize field was chosen as a sampling point using GPS Essentials (mictale.com), resulting in 143 sampling points (Figure 1, Supplementary file). Sampling of parasitoids was carried out on maize fields during the vegetative phase (2-3 weeks old) because the highest S. frugiperda infestation occurs at this stage [13].

2.2. Sampling

The survey was carried out from September 2019 to June 2022. Sampling was carried out once on each field. Parasitoids were collected by collecting hosts (eggs, larvae, and pupae) of S. frugiperda, found on maize plants in every field. Sampling was carried out purposively by taking eggs, larvae, and pupae found. The samples obtained were brought from the field to the laboratory using an insect-rearing plastic container (21 x 21 cm) and then kept individually in plastic cups (400 ml) until moths or parasitoids appeared under laboratory conditions (26±1 °C, 60-80% r.h.). Parasitoids that appeared were counted, recorded, and grouped based on similar morphological characteristics, then preserved in a 1.5 ml microtube filled with 70% ethanol for further identification.

2.3. Parasitoid Identification

The parasitoids that appeared were identified at the Plant Protection Laboratory, Department of Agrotechnology, Universitas Muhammadiyah Yogyakarta. Identification of parasitoids was carried out by observing and matching the morphological characteristics of the parasitoids with some relevant literature [23,44,45,46]. The identified parasitoids were then documented and measured using TrueChromeII, TCapture 5.1 software (Fuzhou Tucsen Photonics Co., Ltd., China), and a Leica S6E Stereo Microscope (Leica Microsystems, Germany) at the Biological Control Laboratory, Department of Plant Protection, Faculty of Agriculture, IPB University. All identified parasitoids were confirmed at the Ecology and Systematics Laboratory, Faculty of Applied Science and Technology, Ahmad Dahlan University.

2.4. Data Analysis

The diversity (species richness and abundance) and parasitism rate of S. frugiperda parasitoids were tabulated using a pivot table on Microsoft Office Excel 365. Parasitoid distribution was mapped based on sampling points (regional administrative data) using ArcGis 10 (ESRI, Environmental Systems Research Institute, California, USA).

3. Results

The results showed that 15 species of parasitoids were associated with S. frugiperda. 80% (12/15) of the parasitoids discovered were new associations (Table 1). Four species were egg parasitoids, eight were larval parasitoids, and three were pupal parasitoids. The egg parasitoids found were Hymenoptera sp.1 (Figure 2a), Hymenoptera sp.2 (Figure 2b), Trichogramma sp. (Figure 2c), and Telenomus remus (Figure 2d). The larval parasitoids found were Hymenoptera, such as Apanteles sp. (Figure 3a), Campoletis sp. (Figure 3b), Coccygidium sp. (Figure 3c), Eupelmus sp. (Figure 3d), Microplitis sp. (Figure 3e), Stenobracon sp. (Figure 3f), and Diptera such as Archytas marmoratus (Figure 4a) and Megaselia sp. (Figures 4b). Based on its morphological characteristics, we confirmed that the last Dipteran species is M. scalaris (Figure 5). Meanwhile, the pupal parasitoids found were Brachymeria femorata (Figure 6a), Brachymeria lasus (Figure 6b), and Charops sp.

However, not all parasitoid species were found in every location. Eight species of parasitoids were found in Bantul, seven in Sleman, six in Gunung Kidul, and only three in Kulonprogo. The richest and most abundant species of parasitoids were found in Bantul (8748 parasitoids). This amount is far higher than in other places. Two thousand seven hundred forty-eight parasitoids were obtained in Sleman, 924 in Gunung Kidul, and 494 in Kulon Progo (Table 1).

Based on a mapping analysis of the parasitoid species distribution found in the field, S. frugiperda has spread almost throughout the Yogyakarta region (Figure 7A). Telenomus remus was the most dominant parasitoid because of its abundance. However, T. remus was only distributed in a few areas (Figure 8), with the highest abundance found in Bantul.

In addition to its high abundance, T. remus had the highest parasitism rate (26-87%). Other parasitoids such as Hymenoptera sp.1, Campoletis sp., Coccygidium sp., and Microplitis sp. had maximum parasitism rates of 12.81%, 12.69%, 17.54%, and 19.84% respectively. Apanteles sp. had a parasitism rate of 2-30%. The parasitism rate of other parasitoids occurred at lower rates, such as Stenobracon sp., Brachymeria femorata, and B. lasus, with 1 – 2% parasitism rate. Meanwhile, the other parasitoids, including Hymenoptera sp.2, Megaselia scalaris, Archytas marmoratus, and Charops sp., had a less than 1% parasitism rate (Figure 9).

Periodically, only one parasitoid was found associated with S. frugiperda in 2019. The number of parasitoid species associated with S. frugiperda increased to eight, thirteen, and fifteen species in 2020, 2021, and 2022, respectively (Table 2).

Figure 9. Parasitism rate of Spodoptera frugiperda’s parasitoid in Special Region Yogyakarta, Indonesia.

Figure 9. Parasitism rate of Spodoptera frugiperda’s parasitoid in Special Region Yogyakarta, Indonesia.

4. Discussion

A survey was conducted to study the diversity and the new associations between S. frugiperda and local parasitoids in the Special Region of Yogyakarta for three years after S. frugiperda was discovered in Indonesia to revise from the old to the new association biological control theoretical approach.

Classical Biological Control emphasizes old associations that invasive pests are controlled by the natural enemy from the country of origin because local natural enemies cannot control the invasive pest [30]. Meanwhile, a new association between invasive pests and local natural enemies will not result in suppression/regulation of the pest because adaptation will take too long. In fact, a physiological equilibrium that results from coevolution between a parasitoid and its hosts might prohibit such long-associated parasitoids from acting as efficient biological control agents [47]. Adaptation can happen relatively quickly, as we found during this study. Elton [48] said that when a parasite species is introduced into an ecosystem with a host or hosts it has never been associated with, the parasite population often rises quickly, often to outbreak levels, and its host population is suppressed.

Our research shows an increase in the number of parasitoids associated with S. frugiperda. These findings indicate a similar pattern in other areas of Indonesia. For example, one parasitoid (T. remus) was found to be associated with S. frugiperda at the beginning of survey activities in 2019. According to Jindal et al. [49], no parasitoids were seen in India in 2019 because S. frugiperda had just recently infected the crop in the late season. Maharani et al. [5] also reported no parasitoids directly associated with S. frugiperda in Bandung and Garut, West Java, Indonesia in 2019.

However, A. marmoratus and Hymenoptera larvae were found from Mythymna separata obtained from the same field where S. frugiperda was collected. Furthermore, Pu’u and Mutiara [50] reported no parasitoids associated with S. frugiperda in Ende, East Nusa Tenggara, in 2020. Suroto et al. [24] reported one parasitoid (Apanteles sp.) associated with S. frugiperda in Banyumas, Central Java, in 2021. Then, Minarni et al. [21] reported the association of S. frugiperda with one egg parasitoid (T. remus) and three larval parasitoids from the Braconidae, Ichneumonida, and Chalcididae families in the same location in 2022.

Based on the abundance of parasitoids, this study also showed differences in the abundance found in the four districts in Yogyakarta. The area with the highest abundance of parasitoids was Bantul. This is due to the characteristics of the sampling site, where Bantul Regency serves as Yogyakarta’s primary maize-producing hub [51], making hosts (S. frugiperda) more accessible than in other districts. Kishinevsky et al. [52] said that an individual parasitoid would be more prevalent in a site if its host population is more numerous, as are its hosts.

Some parasitoids are similar to those found in the Western Hemisphere, while others differ. Chelonus insularis (Hymenoptera: Ichneumonidae) is identified as the primary parasitoid of S. frugiperda in most investigations conducted in North, Central, and South America. However, Eiphosoma laphygmae (Hymenoptera: Icheumonidae) is more recommended as a prospective candidate for introduction because of its specificity and significance as a parasitoid of the pest across most of its natural habitat [53]. These two parasitoids were not discovered during our research. They were not associated with S. frugiperda in other investigations, including those conducted in Cameroon [54] and India [55].

The most prevalent parasitoid identified in this investigation is T. remus. This study reported similar results from other investigations [56,57], where T. remus was the dominant parasitoid for S. frugiperda because this parasitoid has a high abundance and parasitism rate (26-87%). Kumela et al. [8] also reported that T. remus was a parasitoid of S. frugiperda eggs, with the highest parasitism rate (69.3%) in Kenya and Southern China (30-50%) [57]. Sari et al. [17] reported the potential of T. remus as a biological agent of S. frugiperda in Indonesia, with a parasitism rate of 69.40%. This value is quite comparative compared to other egg parasitoids such as T. chilotraeae [58]. The parasitism rate of T. remus may be higher in a situation with many potential hosts. Junaedi et al. [59] said that the availability of hosts for parasitoid survival could increase the parasitism rate. During sampling in the field, the population of S. frugiperda eggs was abundant. The high parasitism rate is also due to T. remus’ ability to find and recognize its host [60]. Goulart et al. [61] said that the ability of T. remus to search and recognize its hosts is better than other S. frugiperda egg parasitoids such as Trichogramma pretiosum.

Telenomus remus is a native egg parasitoid from Malaysia and Papua New Guinea [62]. Telenomus remus is a common egg parasitoid used to control pests in the Noctuidae group, especially the Spodoptera genus [63]. Telenomus remus has been introduced to many countries, including America [63], despite having little success in controlling S. frugiperda [64]. In contrast to studies in Indonesia and other countries like Africa (Benin, Cameroon, Côte d’Ivoire, Ghana, Kenya, Niger, Nigeria, Uganda, South Africa, Tanzania, and Zambia) and Asia (China, India, and Nepal) [65] which claimed that T. remus is a potential biological control agent for controlling S. frugiperda. This could occur because S. frugiperda is not native to these regions. When a parasitoid can attack a new host species, the host typically suffers greatly [66]. In contrast to America, the existence of other parasitoids cannot successfully control S. frugiperda because they have evolved to its original natural enemies, making it resistant to effective control by other parasitoids [47].

Other egg parasitoids found were Hymenoptera sp.1 and Hymenoptera sp.2. Morphologically, these two parasitoid species are different from T. remus, even though they both come from the Platygasteridae family, and these parasitoids are characterized by eight flagellum segments and a wider second metasoma segment, different from T. remus, which has dilation in the third metasoma [67]. Unfortunately, due to minimal sample conditions, identification could not be carried out to the genus level. Platygasteridae was found, with a lower abundance than T. remus. This might happen because the two are not the primary parasitoids of S. frugiperda eggs. Hymenoptera sp.1 has a similar morphological character to Platygaster oryzae, the main parasitoid of Asian rice gall midge Orseolia oryzae [68]. However, Hymenoptera sp.1 species could not be confirmed, even though this parasitoid emerged from the rearing of S. frugiperda egg clusters collected from maize fields adjacent to rice fields. Meanwhile, Hymenoptera sp.2 has a different body color from P. oryzae. Platygaster oryzae has a metallic black body color [69,70], while the parasitoids found were a bright yellow. The discovery of two different egg parasitoid species from T. remus and Trichogramma indicates the existence of two new associations between S. frugiperda and local egg parasitoids.

Another new association has been found between S. frugiperda and larval parasitoids from Hymenoptera order, including Apanteles sp., Microplitis sp., Campoletis sp., Coccygidium sp., Eupelmus sp., and Stenobracon sp. All of these parasitoids are present in the Western hemisphere. However, Apanteles like A. ruficrus imported from the Australia [71], M. manilae from the Thailand [72], and C. chloridae from the India [73] to the US. Apanteles sp. is a larval parasitoid with the highest parasitism rate (2-30%) compared to other S. frugiperda larval parasitoids. Supeno et al. [23] and Suroto et al. [24] also reported the incidence of parasitism of Apanteles on S. frugiperda larvae with a 17 - 22% parasitism rate, higher than other S. frugiperda larval parasitoids. Moreover, Microplitis also has a significant parasitism rate (0.5 – 19.84%). The genus Microplitis, reported as a larval parasitoid of S. frugiperda, includes M. manilae [72]. Microplitis is a genera widely distributed throughout all biogeographic zones, with macrolepidopterans as their primary hosts [74]. The results of this study also indicate that Apanteles and Microplitis are parasitoids of S. frugiperda larvae potentially developed as biological agents. Association of S. frugiperda larvae with Apanteles and Microplitis was also reported from Bogor, West Java, with 0.39 and 12.3% parasitism rates, respectively [75]. Surprisingly, Stenobracon sp. and Eupelmus sp. have never been found in the western hemisphere or elsewhere. Meanwhile, Coccygidium sp. is reported from other regions. Campoletis and Coccygidium had 0.2-12.69% and 0.2-17.54% parasitism rate, respectively. Campoletis has also been reported in India, with a 2-4% parasitism rate [55]. Meanwhile, Coccygidium was also reported in India, with a 0.001% parasitism rate [55], and in Ghana, with a 3.9-19.3% parasitism rate [44].

Apart from the Hymenoptera order, there were also S. frugiperda larvae parasitoids from the Diptera order, such as Archytas marmoratus and Megaselia scalaris. Archytas marmoratus is a potential parasitoid for S. frugiperda in the American field [76]. However, the parasitism rate found in this study was very low. The parasitism level of M. scalaris was also very low. Megalia scalaris was first reported in Asian regions, including India [77] and China [78]. Megalia scalaris was also found in the Mexican region [79]. This finding is also the first report of an association between S. frugiperda and M. scalaris in Indonesia. Megaselia scalaris is an insect found in various regions, usually in decaying organic matter [80]. Besides being reported as a larval parasitoid of S. frugiperda, M scalaris was also reported to be associated with peach fruit fly, Bactrocera zonata (Saunders), Mediterranean fruit fly, Ceratitis capitata (Wiedemann) in Egypt [81], and fruit-piercing moths Thyas coronota (Fabricius) (Lepidoptera: Erebidea) in India [82]. However, according to a recent study, M. scalaris is not recommended as a potential biological control agent for S. frugiperda. Megaselia scalaris prefers to consume deceased larvae instead of acting as an endoparasitoid with parasitism rates of 2.2 and 0.7% in third- and fifth-instar larvae of S. frugiperda, respectively [83].

Besides the new association between S. frugiperda and several egg and larval parasitoids, there was another new association with three pupal parasitoids, such as Brachymeria lasus and B. femorata from the Chalcididae family, and Charops sp. from Ichneumonid family. Because these parasitoids also have never been documented in the western hemisphere. Several chalcidid families reported to be associated with S. frugiperda include B. flavipes (= robusta) (Fabricius), B. ovata (Say), Conura femorata (Fabricius), C. hirtifemora (Ashmead), C. igneoides (Kirby), C. immaculata (Cresson), and C. meteori (Burks) [84]. These two S. frugiperda pupal parasitoids were only found in Bantul. Brachymeria lasus and B. femorata usually attack hidden insects, such as the banana leafroller caterpillar Erinota thrax (Lepidoptera: Hesperiidae) [85], fire caterpillars, and other Noctuidae families. The proximity of banana plants to the sampling site in Bantul Regency suggests that there may be other hosts for these parasitoids. However, these two parasitoids have also been found parasitizing the pupae of S. frugiperda in Egypt [86]. Lastly, Charops sp. was identified through its pupal characteristics. Charops sp. was also reported as a pupal parasitoid of S. frugiperda in Cameroon [87], Ghana, and Benin [54] but has never been reported from the western hemisphere as well.

Our findings support Hokkanen and Pimentel’s theory on using the New Association biological control approach for controlling S. frugiperda with local parasitoids such as T. remus and Microplitis sp. Telenomus remus has been used as a biological agent for S. exigua [88], and Microplitis manilae has also been around for a long time and is associated with other Spodoptera species, such as S. litura in Indonesia [89]. This evidence led us to conclude that these parasitoids existed earlier than S. frugiperda in Indonesia. Hokkanen and Pimental [43] also said that the original host of the most effective new association biocontrol agents is closely connected to the new host of the agent was introduced against. The original and subsequent hosts have typically belonged to the same genus. Longer taxonomic ''jumps'' to a new host from distinct families of hosts are also feasible, as exemplified by a pupal parasitoid B. lasus. Therefore, T. remus and Microplitis sp. are thus possible biological control agent options for use in control initiatives within the new association of biological control concepts equipped to regulate S. frugiperda.

5. Conclusions

This research supports Hokkanen and Pimentel’s theory of using the new association biological control approach for controlling S. frugiperda. In contrast to Kenis [53], who recommends a classical biological control approach since introducing natural enemies may lead to an environmental risk. Telenomus remus is the most dominant parasitoid with a high abundance and parasitism rate compared to other parasitoids. Telenomus remus is a promising opportunity because T. remus is an egg parasitoid. Parasitoids can control pests early in life, thereby avoiding further attacks.