Endophytic establishment of Beauveria bassiana and Metarhizium anisopliae in maize plants and its effect against Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae) larvae

The aim of the present study was to evaluate the artificial establishment of Beauveria bassiana and Metarhizium anisopliae as endophytes in maize plants, and its effect in controlling the fall armyworm, Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae) larvae. The commercial strains B. bassiana Bb-18 and M. anisopliae Ma-30 were concentrated at 1 × 108 conidia ml−1, and the soil drench method was used to establish them as endophytes in maize plant parts. The biological control assays were conducted under laboratory conditions on second and fourth larval instars of S. frugiperda. B. bassiana colonized roots, stems, and leaves of maize tissues. However, a high occurrence of B. bassiana was obtained in roots than leaves and stems with 25, 10, and 5 isolations, respectively, whereas M. anisopliae was only acquired on roots. Both entomopathogenic fungi caused (100%) mortality on the second instar larvae. In addition, B. bassiana and M. anisopliae killed (87 and 75%) of the fourth larval instars, respectively. The fungus M. anisopliae caused the highest sporulation rates during the study. These results suggest that the entomopathogenic fungi might contribute to a sustainable S. frugiperda management in maize production in Cuba.


Background
The fall armyworm, Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae), is distributed in tropical and subtropical regions (Pogue 2002). This insect pest causes damage to several host plants (maize, peanuts, cotton, soybean, and forage grasses); however, the maize (Zea mays L.) constitutes the main agricultural crop preferred by S. frugiperda (Virla et al. 2008). The most serious damage produced by this pest is continuous consumption of the young shoots reducing the photosynthetic area of the plant.
Currently, the most employed method to reduce populations of S. frugiperda is the spraying of chemical insecticides, but despite its fast mode of action, the larvae have developed resistance as effects of this method of control, and it causes environmental pollution (Berón and Salerno 2006). Furthermore, S. frugiperda larvae remain feeding inside the plant shoots reducing the contact with insecticides applied for their control (Braga Maia et al. 2013). For these reasons, the use of biological control as an ecofriendly alternative could be effective to control S. frugiperda. Among the most biological control measures used stand out the entomopathogenic fungi (EPF); Beauveria bassiana (Balsamo-Crivelli) Vuillemin, and Metarhizium anisopliae (Metschnikoff) Sorokin, because they can cause infection at all life stages (Hajek and St Leger 1994). These EPF can also act as saprophytes on the organic matter and live as endophytes of several plants (Vega 2008). However, the endophytic colonization by EPF might be more widespread than currently realized and may provide a source of indirect interaction between fungi and insects. Furthermore, studies related to the endophytic colonization of B. bassiana and M. anisopliae in maize in Cuba and its use as biological control on S. frugiperda is already scarce.
The aims of this study were therefore to evaluate the artificial establishment of B. bassiana and M. anisopliae as endophytes in maize plants, and their effect in controlling S. frugiperda larvae.

Experimental setup and soil treatment
The experiment was carried out at the laboratory of Microbiology belonged to Universidad Central "Marta Abreu" de Las Villas, Santa Clara (22°24′ 49″ N, 79°57′ 58″ W), Cuba. Twenty kilograms of an Inceptisol soil (USDA Soil Taxonomy) were collected from an agroecological maize field located in "Encrucijada" municipality (22°37′ 01″ N, 79°51′ 58″ W). The soil was autoclaved 3 times at 121°C for 1 h. The efficacy of the sterilization was evaluated by diluting 1 g of soil in 20 ml sterile distilled water and then by inoculating 100 μl (soil + water solution) on Sabouraud Dextrose Agar (SDA) culture medium without antibiotics. Petri dishes (9 cm diameter) with the soil inoculations were incubated at 25°C and 75% relative humidity (RH) in the dark. Effectiveness of the soil sterilization was reached when no growth of bacterial or fungal microorganisms were detected.
Endophytic colonization of B. bassiana and M. anisopliae on maize plants The EPF used in the study were the commercial strains B. bassiana Bb-18 and M. anisopliae Ma-30. These fungi were previously isolated from Hypothenemus hampei Ferrari and stored at the EPF collection of the Facultad de Ciencias Agropecuarias, Universidad Central "Marta Abreu" de Las Villas. Sterile soil was deposited into 40 polyethylene bags (500 g capacity), and then 20 bags were inoculated with 10 ml of the commercial strain of B. bassiana Bb-18, and the other 20 bags were inoculated with 10 ml of M. anisopliae Ma-30 at a concentration of 1 × 10 8 conidia ml −1 . The soil contained into the bags was incubated during 7 days in a climatic chamber (Memmert, Germany) at 25°C and 75% RH in the dark to propitiate the establishment of the entomopathogenic fungi. Afterwards, maize grains cv 'P 78-45' were sown into the soil contained in the bags at a ratio of 1 grain per bag, and then the bags were placed into a germination chamber (TP, China) at 25°C, 75% RH, and 16 h of light and 8 h of dark (L16:D8). Once the maize plants reached the growth stage BBCH 12 (2 leaves unfolded) (Meier 2001), roots, stems, and leaves were collected and surface sterilized by dipping for 3 min in 1.5% sodium hypochlorite, 2 min in 70% ethanol, and then rinsed 3 times in sterile distilled water. The efficacy of sterilization was evaluated by inoculating 100 μl of the last rinse water on SDA culture medium.
Damaged plants resulting from the sterilization procedure were discarded to avoid the death of the endophyte tissue (Douglas et al. 2012). Afterwards, plants were dried by sterile paper towels and aseptically cut into small pieces (1 cm 2 ) in a laminar flow hood. The plant pieces were then inoculated on SDA culture medium contained in Petri dishes (9 cm diameter) with addition of chloramphenicol (250 mg/l w/v). The Petri dishes with the plant pieces were incubated before. A single spore of each EPF was harvested and re-inoculated on SDA culture medium to obtain colonies without contamination. For each EPF studied, 4 replications were done.

Insects assay
Both EPF strains, B. bassiana Bb-18 and M. anisopliae Ma-30, were supplied by the Plant Health Center from Villa Clara province. Conidia of each fungal strain were adjusted to a final concentration of 1 × 10 8 conidia ml −1 through a Neubauer hemocytometer chamber (Brand, Germany). The EPF were diluted in distilled water. Additionally, distilled water was used as the control treatment. To evaluate the effect of each fungal formulation against S. frugiperda, a pedigree of this insect was obtained under laboratory conditions (Chacón-Castro et al. 2009). Twenty larvae from the second instar of S. frugiperda and 20 from the fourth instar were dipped for 1 min into 15 ml of a conidial suspension of B. bassiana Bb-18 and M. anisopliae Ma-30, as well as 15 ml of distilled water (control treatment). After 10 min of treatments exposure, each larva was placed individually in a sterile Petri dish (15 cm diameter) with a portion of leaves and stem of maize plants (growth stage BBCH 12) for feeding. Petri dishes with S. frugiperda were incubated at 25°C, 90% RH, and L16:D8 photoperiod. Larvae were checked every day for mortality and production of mycosis. The experiment was repeated 4 times, and for each treatment 4 replications were used.

Data analysis
Data on frequencies of occurrence of endophytic B. bassiana and M. anisopliae in maize plant parts were analyzed by chi-square test. Chi-square statistics were calculated through a frequency table, and p values were corrected using Cramer's V. Analysis of variance (ANOVA) was applied to compare the mortality and sporulation induced by B. bassiana and M. anisopliae on S. frugiperda larvae. Means were separated using the Tukey DHS test. Chi-square, ANOVA, and LSD tests were run using STATGRAPHICS Plus 5.1 (Manugistics Inc.) with significance level of 0.05. However, a high occurrence (χ 2 = 5.06, df = 2, p = 0.0014) of this fungus was obtained in roots than leaves and stems with 25 ± 0.25, 10 ± 0.29, and 5 ± 0.65 isolations, respectively (Table 1). These results revealed that the B. bassiana colonization (62.5%) occurred on the maize roots showing differences (χ 2 = 2.25, df = 1, p = 0.0001) with the 12.5% of stems colonized by this fungus (Table 1). M. anisopliae was greater (χ 2 = 42.67, df = 2, p = 0.0001) acquired from roots, 32 ± 0.29 isolations, than from leaves and stems, where no fungal colonization was detected ( Table 1).

Results and discussion
The obtained results showed that B. bassiana was endophyticaly established on roots, stems, and leaves of maize plants. These findings are in accordance with the results obtained by Mahmood et al. (2019), who revealed that all inoculated maize plants in their experiment contained endophytic B. bassiana. In addition, the highest colonization levels were (61%) in the oldest inoculated leaves and (19%) in the youngest non-inoculated leaves indicating the movement of the endophyte inside plants. It has been shown that after a foliar spray of B. bassiana on maize plants the hyphae can penetrate inside the xylem and act as entophyte in leaves tissues (Wagner and Lewis 2000). However, the present study showed that M. anisopliae was only detected on roots. Besides, Pilz et al. (2011) demonstrated that M. anisopliae spores applied on maize leaves were able to survive for no longer than 3 days after application, whereas on the soil surface a high increase of fungus densities were found after treatments. Other studies have been demonstrated that M. anisopliae was commonly associated with plant roots (Keyser et al. 2015). Other experiments revealed that B. bassiana reached a higher colonization on leaves than roots of Sorghum bicolor (L.) Moench when the grains were treated by a formulation of this fungus (Tefera and Vidal 2009). This result suggested that the endophytic colonization of B. bassiana can vary with the  host plant part and the method used. Renuka et al. (2016) observed that stems of Z. mays were also colonized by B. bassiana; however, the availability and persistence on the plant tissues varied according to their age (high persistence in young tissues). Despite several authors have demonstrated that B. bassiana as endophyte of maize, others have been revealed that after the inoculation of this EPF on maize seeds, no colonization of plant tissues was obtained (Tall and Meyling 2017).

Effect of B. bassiana and M. anisopliae on S. frugiperda
The survivorship experiments showed that B. bassiana started to kill larvae from the second instar of S. frugiperda on the fourth day after treatment, reaching 11% mortality rate. After that, mortality increased by this fungus reaching 100% mortality rate on ninth day after treatment. Meanwhile, M. anisopliae started its mortality on the third day after treatment with 19% larval mortality (Fig. 1).
At the fourth instar, the EPF M. anisopliae and B. bassiana started their mortality at third and fourth days after treatments, respectively. The mortality caused by B. bassiana was increased, reaching up to 87% on 11th day after the application. However, the fungus could not kill 100% of the larvae. On the other hand, M. anisopliae caused a maximum of 75% mortality rate (Fig. 2). The control treatment did not have any effect on second and fourth instars of S. frugiperda larvae.
The results are in contrast with the findings obtained by Akutse et al. (2019), who demonstrated that B. bassiana ICIPE 676 caused moderate mortality of 30% to the second instar larvae of S. frugiperda, whereas M. anisopliae ICIPE 78, ICIPE 40, and ICIPE 20 caused egg mortality of 87, 83, and 79.5%, respectively. This denotes that the virulence of these EPF may vary according to the strain origin.
Studies conducted to compare the sporulation level of B. bassiana and M. anisopliae demonstrated that the effectiveness of 3 B. bassiana strains was greater than the effect caused by M. anisopliae, when these EPF were applied on Spodoptera litura F. Furthermore, B. bassiana caused the best candidate related to virulence and germination rates, whereas the highest enzymatic activity  was resulted by the use of M. anisopliae (Petlamul and Prasertsan 2015). Generally, several studies have been conducted to evaluate the effect of B. bassiana against lepidopteran species, but the effect of M. anisopliae on S. frugiperda larvae has been less reported and the present research constitute an advance in this regard.
Although there were no differences between the mortality produced by B. bassiana and M. anisopliae on S. frugiperda larvae, the sporulation rate was higher in the case of M. anisopliae than in B. bassiana. The EPF M. anisopliae had the ability to reach a 52.8% of sporulation on cadavers of S. frugiperda at 3 days after the mortality (Ibarra-Aparicio et al. 2005).
This result is important because the sporulation of the EPF from its host means the formation of a new source of inoculum for the infection of new populations of insect pests. Therefore, the fungus that shows the highest percentage of sporulation is more likely to be disseminated by biotic and abiotic agents.

Conclusion
The present study indicated that B. bassiana and M. anisopliae occurred as endophytes in maize plant parts, mainly in root tissues. Further, it highlights the importance of these EPF as potential biological control agents of S. frugiperda in maize fields in Cuba. These results suggest that the EPF studies could contribute to sustainable pest management of maize production.