Parasitism and Development of Tetrastichus howardi (Hymenoptera: Eulophidae) in Immature Spodoptera frugiperda (Lepidoptera: Noctuidae)

Tetrastichus howardi (Olliff, 1893) (Hymenoptera: Eulophidae) is an endoparasitoid mainly of Lepidoptera. Its potential as a biological control agent has been investigated for several agricultural and forestry species in different countries. The objective of our study was to evaluate the biological characteristics of T. howardi in larvae, prepupae and pupae of Spodoptera frugiperda (Smith, 1797) (Lepidoptera: Noctuidae), under laboratory conditions. The first bioassay was performed with S. frugiperda larvae in the fifth and sixth instar and prepupal phases, which were exposed to the parasitism of 7 female T. howardi for 24 h. The second bioassay was conducted with S. frugiperda pupae, with parasitism allowed for 24 h, at six parasitoid of T. howardi densities, for one host pupa of S. frugiperda (1, 7, 14, 21, 28 or 35 parasitoids: 1 host). The parasitism of T. howardi in S. frugiperda larvae was confirmed by larval mortality and the emergence of parasitoid descendants. Female T. howardi parasitized and emerged from S. frugiperda larvae of fifth and sixth instars, prepupae, and pupae under laboratory conditions, which is the first record of such in the literature.

pupa. Parasitism was allowed for 24 h and the pupae were kept in an acclimatized room at 25±2 °C, 70±10% RH, and a 14 h photophase until the emergence of adult parasitoids. The natural mortality of the host was calculated using the Abbott's (1925) formula.

Experimental Evaluation
The biological characteristics evaluated for the two experiments were: percentage of parasitism and emergence of parasitoids; life cycle duration (egg-adult); progeny (number of individuals per host); progeny per female (number of individuals per parasitoid); longevity of adults (with food), and sex ratio (number of females/total number of individuals).

Statistical Analysis
The experimental design used was completely randomized (DIC). In the first experiment, three treatments (larval phases) and five repetitions were performed, with each repetition containing ten individuals. For the second experiment, six treatments were prepared (T. howardi density). Each treatment consisted of five replicates, each containing a group of 10 pupae, totaling 50 experimental units per treatment. Data on life cycle duration, percentage of parasitism and emergence of parasitoids, number of individuals per host and per female parasitoid, sex ratio, and longevity were submitted to analysis of variance and, when significant, to regression with 5% of probability (Sigmaplot 12.0 Software). The choice of the most suitable equation was based on the Determination Coefficient (R²), the significance of the Regression Coefficients (ᵝi), and the Regression by the F Test (at 5% probability).

Parasitism of S. frugiperda Fifth and Sixth Instar Larva, Prepupae and Pupae by T. howardi
Adult female T. howardi were able to parasitize, develop, and emerge from S. frugiperda fifth and sixth instar larvae as well as prepupae.
The highest parasitism and emergence of T. howardi were observed in sixth instar larvae, at 28.00±6.23% and 76.39±10.91%, respectively (Table 1). The parasitized S. frugiperda larvae had a swollen, dry, and rigid appearance, with brown and black coloration. The parasitism and emergence from prepupal S. frugiperda were greater than 90% (Table 1).
T. howardi had the shortest life cycle (egg-adult) in the prepupal phase of S. frugiperda, with 20.10±0.22 days ( Table 1). The progeny of T. howardi increased as the instars of S. frugiperda increased.
The greatest number of T. howardi descendants were obtained in the prepupal phase of S. frugiperda, with 363.27±44.61 individuals. The most progeny of T. howardi was obtained in the prepupal phase of S. frugiperda, with 363.27±44.61 offspring. The progeny per female T. howardi was also higher in prepupal S. frugiperda, with 47.61±6.09 individuals per female (Table 1).
The sex ratio of T. howardi that emerged from the fifth and sixth instar and prepupal larvae was similar, with an overall mean of 0.90±0.02 (Table 1).
There was no difference in the longevity of adult female T. howardi emerged from the fifth and sixth instars and prepupal S. frugiperda, with an overall mean of 20.40±0.33 days. The longevity of males was lower when adults emerged from prepupal S. frugiperda, with 18.70±0.40 days (Table 1). Note. Means followed by the same letter in the column do not differ from each other by Tukey's Test at 5% probability. N = number of repetitions.
The parasitism of T. howardi on S. frugiperda occurred in all larval instars of this host. After parasitism, S. frugiperda larvae had a swollen, dry, and rigid appearance with brown and black coloration. Seven fifth or sixth instar S. frugiperda larvae that tried to transition to the prepupal phase died, remaining in an intermediate phase between the two.
The percentage of parasitism, emergence, life cycle duration (egg-adult), sex ratio, and longevity of T. howardi females in S. frugiperda pupae were similar in the different densities evaluated ( Table 2).
The mean percentage of T. howardi parasitism and emergence in S. frugiperda pupae was 99% in the evaluated densities of female parasitoids. The duration of the life cycle (egg-adult) of T. howardi in pupae of S. frugiperda was similar, with an overall mean of 16.30±0.10 days.
The sex ratio of T. howardi, with a general mean of 0.85±0.02, did not differ when the number of females increased. The longevity of the parasitoid also exhibited no statistical difference between the densities tested, with means of 22.10±0.64 to 23.60±0.80. The density of the progeny produced in the S. frugiperda pupae was positively influenced by the density of T. howardi, with means ranging from 99.1±6.09 to 513.82±18.66 of offspring per pupa at densities 1:1 and 28:1, respectively ( Figure 1).

Discussion
Tetrastichus howardi parasitized, developed, and emerged in all tested phases of S. frugiperda, allowing its use for the biological control of this insect. Similar results were observed in the parasitism capacity of T. howardi in different stages of Helicoverpa armigera, (Hübner, 1809) (Lepidoptera: Noctuidae), in which 2 fifth instar larvae were parasitized with emergence of parasitoids in the pupal phase, with a progeny of 81 individuals; the egg-adult cycle was 19.5 days; the sex ratio was 0.88; the longevity of males was 19.6 days and females 20.3 days, demonstrating significant parasitism in the larval and pupal stages of H. armigera (Simonato, Oliveira, J Our study is the first to record, under laboratory conditions, the parasitism of T. howardi in the larvae and prepupae of S. frugiperda. This information is very relevant because this insect is easy to raise on an artificial diet and can be used as an alternative host for large-scale production (Parra, 2001).
In relation to parasitism by T. howardi in the fifth and sixth instar and prepupal larvae of S. frugiperda, we can consider, parasitism generally increased with the more advanced development of the host S. frugiperda, with values of 10.00±4.47; 28.00±6.23; and 90.00±5.24, respectively. The emergence of adult parasitoids was 37.50±15.14 for the fifth instar, 76.39±10.91 for the sixth instar, and 92.50±5.24 for the prepupal phase. This increase may be associated with the nutritional quality of the host larvae, which increases as it grows and develops (Cônsoli & Vinson, 2009). This increase may be associated with the nutritional quality of the host larvae, which increases as it grows and develops (Cônsoli & Vinson, 2009). These results demonstrate that caterpillars, pre-pupae and pupae of S. frugiperda fit perfectly as hosts for the multiplication of T. howardi and, as the host develops and grows, the more the parasitoids multiply better.
Another important factor is that larvae in more advanced phases can be more aggressive, demonstrating that larger hosts can defend themselves better than smaller ones, which are in early phases of development (Kouamé & Mackauer, 1991). The fifth instar larvae presented a defensive behavior that reduced parasitism by T. howardi, through body movements and regurgitation. A similar situation was observed for third, fourth, and fifth instar larvae of H. armigera that exhibited aggressive movements towards Eriborus argenteopilosus (Cameron) (Hymenoptera: Ichneumonidae), which interrupted the oviposition of the parasitoid female or moved her away (M. E. Pascua & L. T. Pascua, 2004).
In the sixth instar, the larva prepares to change to the prepupal phase; thus, it tends to remain immobile. After the last larval instar, the larva transforms into the prepupa, at which point it completely stops feeding and prepares to pupate. Initially, the prepupa remains shrunken, which facilitates parasitism, as occurred for Anticarsia gemmatalis Hübner, 1818 (Lepidoptera: Noctuidae), a soybean caterpillar (Fernandes, 2018).
When parasitizing fifth and sixth instar larvae, T. howardi emerged both from the larvae and pupae of S. frugiperda. The stage change of these larvae may have been influenced by parasitism or the cessation of feeding. The cessation of feeding promotes reduced metabolism; with this, the secretions of brain cells start to stimulate the prothoracic gland at levels sufficient to promote molt (Costa, Ide, & Simonka, 2006).
After parasitism, S. frugiperda larvae have a swollen, dry, and rigid appearance with brown and black coloration, because parasitoid females of the Eulophidae family inject venom during oviposition and reduce the number of circulating hemocytes in the host. This provides a favorable environment for the development of its immature phase, probably due to immunosuppression of the defense system, to prevent the parasitoid larvae from dying due to encapsulation, asphyxia, or activation of substances antagonistic to their development (Uckan, Sinan, Savasci, & Ergin, 2004;Nappi & Christensen, 2005;Carton, Poirié, & Nappi, 2008;Andrade et al., 2010). The same aspects were observed in the parasitism of A. gemmatalis larvae by T. howardi in laboratory and semi-field conditions (Fernandes, 2018).
The parasitism and emergence of T. howardi in prepupal S. frugiperda above 90%, demonstrates a great advantage for this parasitoid and its excellent potential to be used as an alternative host for rearing in biofactories, to increase the speed of the entire process. Similar results were found for T. howardi parasitism of in prepupal A. gemmatalis (Fernandes, 2018). This corroborates with several studies, where the use of younger or newly formed hosts facilitates excellent development of T. howardi, as females of this parasitoid prefer newly formed pupae and prepupal phases as described in the results of this and other studies (Moore & Kfir, 1995;Prasad, Aruna, Kumar, & Kariappa, 2007;Cruz et al., 2011;Barbosa et al., 2015).
Seven S. frugiperda larvae in the fifth or sixth instar that tried to make the transition to the prepupal phase died, remaining in an intermediate phase between the two, in which the host does not have a sufficient immune response, such as to the encapsulation of eggs, to complete its development cycle. Parasitism can cause changes in juvenile hormones, ecdysteroids, and neuropeptides in the host, preventing some larvae from evolving into the next stage (Strand & Peck, 1995).
The duration of the life cycle (egg-adult) of T. howardi emerged from S. frugiperda exposed to seven female parasitoids was statistically different depending on the phase. The life cycle was longer in the fifth instar (31.00±0.45) and shorter in the sixth instar (22.44±0.97) and prepupa (20.10±0.22). This may be because late instar larvae provide more nutrients to the parasitoids than younger larvae, resulting in rapid growth and development (M. E. Pascua & L. T. Pascua, 2004). The shorter life cycle of T. howardi is satisfactory for commercialization because it facilitates faster rearing of adults, meeting the requirements of biofactories.
The highest total progeny and progeny per female of T. howardi occurred in the prepupal phase of S. frugiperda, confirming its potential as an alternative host for mass rearing. T. howardi is a parasitoid of pupae and, thus, preferred the prepupal phase, which is the closest to their natural host stage. A similar preference was observed for the phases of D. saccharalis parasitized by T. howardi (Rodrigues et al., 2021). In the larval stage of S. frugiperda, changes occur in the structure of various tissues and their metabolism, which interfere with the availability of nutrients in the larvae's hemolymph (Cônsoli & Vinson, 2009), directly interfering with parasitism.
The sex ratio of T. howardi was similar in the fifth and sixth instars and prepupae of S. frugiperda larva, demonstrating a high capacity of this parasitoid to develop at any stage, if necessary, with an excellent result with a mean greater than 90%. This parameter is very relevant for biological control, since parasitoid females are the agents responsible for parasitism and successful pest control, providing an increase in mass rearing and its performance in the field (Rodrigues et al., 2021;Costa et al., 2014;Kumar, Baitha, & Bareliya, 2016).
The longevity of T. howardi females was similar in the three phases tested, with a general average of 20 days. The behavior of this species guarantees its reproductive success because the female is responsible for the parasitism and continuation of the species. This survival will guarantee the location of the host in the habitat, evaluating and choosing those that present the best quality for development of offspring, directly interfering with the performance of the progeny (Royer, Fournet, Brunel, & Boivin, 1999;Pereira, Zanuncio, Pastori, Pedrosa, & Oliveira, 2010a;Pereira et al., 2010b). Even if descendant adults do not emergence, the parasitism alone is enough to control and prevent the continuation of the arthropod-pest cycle.
The emergence of T. howardi males is essential for copulation with females to occur in the first hours after emergence (González, Oca, & Ravelo, 2003). The longevity of T. howardi males was lower in the prepupal phase of S. frugiperda, confirming that this phase is a good alternative for breeding this species. In the absence of the pupa, the prepupa will have as promising results as the pupal stage.
In the field all, stages of S. frugiperda are found; thus, parasitism occurring in all larval phases is extremely important to control the fall armyworm in corn.
In relation to parasitism of S. frugiperda pupae by T. howardi with different densities, we can consider that, the percentages of parasitism and emergence of T. howardi from S. frugiperda pupae were not influenced by the density of female parasitoids. This demonstrates the host's suitability for the development of the parasitoids, which must satisfy their nutritional requirements and prevent the host's immune system from eliminating them (Strand & Pech, 1995;Cônsoli & Vinson, 2009).
The increase in the density of T. howardi per S. frugiperda pupa did not affect the duration of the parasitoid cycle. Similar results were obtained for the parasitism of T. howardi, at various densities, on D. saccharalis (Vargas, 2013). Generally, the development period is shortened at high densities of parasitoids. When T. howardi were reared in pupae of Tenebrio molitor Linnaeus, 1758 (Coleoptera: Tenebrionidae), the increased number of females per pupa of the coleopteran reduced the life cycle of this parasitoid, which was 20.19±0.36 days at 1:1 density and 18.88±0.55 days at 32:1 density (Oliveira, 2013). This was also evidenced in the reproduction of Trichospilus diatraeae Cherian & Margabandhu, 1942 (Hymenoptera: Eulophidae) in T. molitor pupae (Favero et al., 2013). Palmistichus elaeisis Delvare & La Salle, 1993 (Hymenoptera: Eulophidae) in pupae of Bombyx mori Linnaeus, 1758 (Lepidoptera: Bombycidae) obtained between 49 and 589 descendants at densities of 1:1, 9:1, 18:1, 27:1, 36: 1, 45:1, and 54:1 (Pereira et al., 2010). The differences between these results may be related to the capacity to support the host (Pereira et al., 2010), competition, as well as the age and size of the parasitoid (Godfray, 1994).  Vol. 14, No. 6; The sex ratio of T. howardi per pupa of S. frugiperda resulted in 0.85±0.02 females. This result was similar to the sex ratio of T. howardi when reared in the pupa of T. molitor, with a proportion of 94% females in the progeny (Oliveira, 2013). The low number of males in the descendant population is also characteristic of other parasitoids in the Eulophidae family, including T. diatraeae and P. elaeisis (Chichera et al., 2012;Pastori et al., 2012a;Favero et al., 2013).
The best densities for total progeny per pupa were 21 and 28, with approximately 500 individuals in each density. The progeny produced by the S. frugiperda pupa was positively influenced by the density of T. howardi ( Figure  1). Many parasitoids in the Eulophidae family can parasitize hosts of different sizes, developmental stages, families, and orders. Because the host species affects the development of parasitoids, different results for their biological parameters have been observed (Baitha, Jalali, Rabindra, Venkatesan, & Rao, 2004, La Salle & Polaszek, 2007Prasad et al., 2007;Silva-Torres, Pontes, Torres, & Barros, 2010;Pastori et al., 2012aPastori et al., , 2012b. The density of parasitoid females is reported to influence the production of total P. elaeisis progeny per T. arnobia pupa, with a minimum of 30 and a maximum of 724 offspring of the parasitoid produced at densities of 1 and 15 per host pupa, respectively (Barbosa, Zanuncio, Pereira, Kassab, & Rossoni, 2016). Similarly, 10 T. diatraeae females per pupa of A. gemmatalis are indicated for mass rearing of this parasitoid, with the best density producing an average of 300 individuals per pupa (Oliveira et al., 2018).
The total progeny of T. howardi was also positively influenced by the density of females per prepupa of A. gemmatalis, where the total progeny ranged from 47.27±4.43 (1:1) to 389.40±4.03 descendants (25:1). For parasitism in the larval phase, the total progeny ranged from 73.11±8.89 to 271.88±38.44 of T. howardi descendants produced at densities of 1 and 10 parasitoids per host, respectively (Fernandes, 2018).
The progeny produced by each female T. howardi, at density 1, had a satisfactory result, producing almost 100 offspring. The progeny of T. howardi produced by females in S. frugiperda pupae was inversely proportional to the density of female parasitoids ( Figure 2). Hence, one female T. howardi is considered sufficient for the parasitism, development, and emergence of this parasitoid. We propose the possibility of optimizing breeding in the laboratory, to obtain a greater number of parasitoids per host using the proportion of only one female for each host pupa, adding results and reducing maintenance costs. Similar data were found in the progeny of T. howardi per female in D. saccharalis, which presented an overall mean of 106.92±6.04, when a female T. howardi was exposed to different densities of D. saccharalis pupae (Vargas, 2013).
Up to a density of 7, T. howardi presented good progeny, with quality for its rearing in the laboratory, aimed at a possible use in applied biological control. At densities 14, 21, 28, and 35, each female of T. howardi produced less progeny. With gregarious parasitoids, when the number of females per host is increased, the average number of female offspring decreases simultaneously (Rabinovich, Jorda, Bernstein, & 2000), because female progeny exhibits greater conspecific competition within a single host (Dorn & Beckage, 2007). This was clearly evidenced in this research with T. howardi reared in S. frugiperda pupae, as the number of females tended to reduce as a function of density.
Due to the high rate of parasitism and emergence in pupae (94%) by a female T. howardi, the high biological potential of these parasitoid females is evidenced, proving that immature T. howardi efficiently exploited the nutritional resources of the host. Parasitoid females have developed strategies throughout their co-evolution process with arthropod-pests to overcome immune defenses and make the most of their nutritional resources, using, for example, substances from the ovary to block host's defenses, kill or paralyze the host, or protect the eggs against the host's immune system (Kaeslin, Pfister-Wilhelm, Molina, & Lanzerein, 2005). These strategies indicate the direct relationship of the parasitoid in the suppression of the host's cellular defense, reflected in the effective development of immature forms of the progeny (Andrade et al., 2010).
The longevity of T. howardi females ranging from 22.10±0.64 to 23.60±0.80 was similar at different densities. In mass rearing of parasitoids, survivability is one of the requirements for quality control (Carneiro, Fernandes, & Cruz, 2009). In addition, greater longevity is a favorable characteristic for the species, as it gives parasitoid females longer time to search for hosts in the field under conditions of host scarcity (Foerster & Avanci, 1999).
Spodoptera frugiperda in its pupal stage has great potential as an alternative host for T. howardi rearing, because in addition to being easy to handle, it allows parasitism and emergence rates of 100%, with 250 parasitoids per pupa, which is important for biological control laboratories aiming at field releases.
Consideration should be given to the possibility of using T. howardi in combination with other parasitoid species of lepidopteran pest eggs, such as Trichogramma spp. or Ooencyrtus submetallicus Howard, 1897 (Hymenoptera: Encyrtidae) (Faca et al., 2021), to control S. frugiperda, which is a proposal for future research. Additionally, the jas.ccsenet.org Journal of Agricultural Science Vol. 14, No. 6; population dynamics of the parasitoid together with the behavior of the Lepidoptera pest in the field should also be taken into account in future studies, given that the pupae of this Lepidoptera are usually located in the first centimeters of the soil and are not located by the parasitoid in natural conditions. The number of female parasitoids that emerged per host, as well as their duration in days, are important factors to be verified in the quality control of laboratory rearings and mass releases, as they interfere with biological aspects of the parasitoid, such as the rate of parasitism, viability, progeny, and sex ratio (Carneiro et al., 2009).
From the results obtained and the discussion raised, this work provides a basis and opens the possibility for future studies with other species of lepidopterans, with possible implementations of new protocols for rearing and field releases of this parasitoid species, in a synchronized way, to balance pest populations in the field.

Conclusions
Tetrastichus howardi parasitized, developed, and emerged in all immature stages of S. frugiperda, which is the first record in the literature.
Overall, the total progeny produced by the S. frugiperda pupa increased as the density of T. howardi increased.
The progeny produced by each female T. howardi in the S. frugiperda pupae decreased as the parasitoid density increased. One female was sufficient for parasitism.