Do bio-insecticides affect only insect species? Behavior, regeneration, and sexual reproduction of a non-target freshwater planarian

Bio-insecticides have been increasingly used worldwide as ecofriendly alternatives to pesticides, but data on their effects in non-target freshwater organisms is still scarce and limited to insects. The aim of this study was to determine the lethal and sub-lethal effects of the bio-insecticides Bac Control (based on Bacillus thuringiensis kurstaki—Btk) and Boveril (based on Beauveria bassiana—Bb) on regeneration, behavioral, and reproductive endpoints of the freshwater planarian Girardia tigrina. The estimated LC50–48h were > 800 mg a.i./L for Btk and 60.74 mg a.i./L for Bb. In addition, exposure to Btk significantly decreased locomotion and feeding activities of planarians (lowest observed effect concentration (LOEC) of 12.5 mg a.i./L Btk) and fecundity rate (LOEC = 3.12 mg a.i./L Btk), whereas exposure to Bb significantly delayed regeneration (LOEC = 0.75 mg a.i./L Bb) and decreased fecundity rate (1.5 mg a.i./L Bb) of planarians. Thus, both bio-insecticides induced deleterious sub-lethal effects on a non-insect freshwater invertebrate species. However, only Bb-based formulation affected the survival, fecundity rate, and regeneration at concentrations below the maximum predicted environmental concentration (PEC = 247 mg/L). Thus, care should be taken when using such formulations as alternatives to chemical insecticides near aquatic ecosystems.


Introduction
Bio-insecticides have been increasingly used since the 1980s as alternatives to reduce the impacts on the environment and public health posed by application of chemical insecticides for the control of pests (Gupta and Dikshit 2010;Singh et al. 2018;Vivekanandhan et al. 2018). These biological compounds are based on microorganisms (bacteria, fungi, virus, or protozoa) and have been considered ecofriendly due to their specificity, low toxicity, fast decomposition, and efficacy when used in low concentrations (Gupta and Dikshit 2010;Kandpal 2014;EPA 2016). Therefore, they are considered safe to non-target organisms and humans (Mazid et al. 2011;Subbanna et al. 2019). Due to all their advantages and safety perception by consumers, the demand of microbiological products increased over the last 5 years showing a global market value of US $ 5.2 billion in 2020 in opposition to US $ 2.3 billion in 2015 (Kumar 2015). According to the Environmental Protection Agency (EPA), 299 active Responsible Editor: Philippe Garrigues * Renato Almeida Sarmento rsarmento@uft.edu.br ingredients have been registered and more than 1400 biobased formulations were available in the market in 2016 (EPA 2016). Over the years, strains of Bacillus thuringiensis occupied prime position in bio-pesticide' market followed by entomopathogenic fungi, such as Beauveria bassiana (Thakore 2006;Olson 2015;Subbanna et al. 2019).
Bacillus thuringiensis kurstaki (Btk) is a facultative anaerobic gram-positive bacterium naturally occurring in soil, water, air, and plants (Machado et al. 2017). It is known as an entomopathogenic bacterium that produces parasporal crystal proteins deathly toxic to insects after ingestion (Sanahuja et al. 2011;Castagnola and Stock 2014). The strains of B. thuringiensis (Btk) can produce different types of crystal proteins containing δ-endotoxins that specifically affect certain orders of insects after previous solubilization and activation in their midgut (OECD 2012). The predicted environmental concentration in surface water (PEC sw ) for different strains of B. thuringiensis kurstaki is lower than 100 μg a.i. Btk/L (EFSA 2012).
Beauveria bassiana (Bb) is a filamentous fungus that belongs to the class of deuteromycetes (Sandhu et al. 2012;Berlitz et al. 2014). Briefly, it affects hosts through penetration of their cuticle by its conidia, followed by internal colonization of their tissues combining both mechanical pressure and enzymatic activities (Mascarin and Jaronski 2016). The grown hyphae invade the hemocoel of hosts and use nutrients of hemolymph producing toxins that ultimately lead to death (Sayed and Behle 2017;Rustiguel et al. 2018). This species of fungus is commonly found worldwide as a saprophyte in soils, an endophyte in plants, and acting as entomopathogen for arthropods (Rehner et al. 2011;Berlitz et al. 2014). The predicted environmental concentration in surface water (PEC sw) for B. bassiana (different strain) ranges from 35.3 to 247.3 mg/L of commercial compound (EFSA 2015).
Despite all the advantages pinpointed to these bioinsecticides and the studies concerning their effects on target and non-target insects (Lajmanovich et al. 2015;Allgeier et al. 2019;Challa et al. 2019;Dornelas et al. 2020a), there is a lack of knowledge about their potential ecological effects to freshwater ecosystems (EFSA 2012;EFSA 2015). Moreover, some of these biological formulations of insecticides are also being used directly in aquatic ecosystems to control mosquito populations that are vectors of human pathogens responsible for important diseases (Pelizza et al. 2010;Singh et al. 2018), which may increase their concentration in freshwater systems. Microbial insecticides-based Btk have been considered environmentally friendly, since they were considered harmless to non-target species due to its target-oriented mode of action (Álvarez and Biosca, 2017). In fact, 48h and 96h LC 50 previously estimated (Table 1) for vertebrate (fish and amphibians) species were high and above 100 mg a.i. Bt/L (Becker and Margalit 1993;Karmrin 1997;WHO 1999), except for the frog Leptodactylus latrans with a LC 50 of 22.45 mg/L (Lajmanovich et al. 2015). Moreover, studies performed with non-target insects showed that microbial insecticides were more acutely toxic for insects than other invertebrates, such as the genus Hydra (Becker and Margalit 1993) (Table 1).
Thus, this study aims to determine the effects of the bioinsecticides Bac Control (based on B. thuringiensis kurstaki) and Boveril (based on B. bassiana) on the freshwater planarian Girardia tigrina Girard (Paludicola: Dugesiidae), not only to evaluate and compare their acute effects posed to planarians (mortality) with other studied species, but also to assess their sub-lethal effects using more sensitive endpoints, such as locomotion, feeding, regeneration, and sexual reproduction. In fact, freshwater planarians have been successfully used in environmental toxicology studies to assess the sub-lethal effects caused by different contaminants on such endpoints (Ofoegbu et al. 2016;Rodrigues et al. 2016;Saraiva et al. 2018;Wu and Li 2018;López et al. 2019;Ofoegbu et al. 2019a, b;Saraiva et al. 2020;Dornelas et al. 2020b;Simão et al. 2020;Simão et al. 2021).
Freshwater planarian's characteristics like their broadly geographic distribution, easy experimental manipulation and maintenance in laboratory (Guecheva et al. 2003;Knakievicz 2014;Knakievicz and Ferreira 2008), and regeneration capacity (Reddien and Alvarado 2004) make them ideal to evaluate sub-lethal effects that might be more relevant in terms for environmental risk assessment. Moreover, planarians are aquatic invertebrates found in a range of water systems (Knakievicz et al. 2006;McConnell 1965;Tyler 2000) and they prey on insect larvae that are common targets of microbial insecticides (Allgeier et al. 2019;Benzina et al. 2018;Bordalo et al. 2020;Hart and Merz 1998;Reddien and Alvarado 2004;Vila-Farré and Rink 2018).

Test organisms
Girardia tigrina was obtained from the USP (University o f S ã o P a u l o ) a n d k e p t i n t h e L a b o r a t o r y o f Ecotoxicology at the UFT (Universidade Federal do Tocantins, Functional and Applied Ecology Research Group) in ASTM (American Society for Testing and Materials) hard water medium (ASTM 1980) with constant aeration, controlled temperature (22 ± 1°C), and constant dark conditions. Planarians were fed with bovine liver (ad libitum) for periods of 2 h once per week followed by renewal of ASTM medium. Seven days before bioassays, planarians were not fed and active organisms with no signs of injuries were used for experiments.   The length of test organisms used for the acute bioassays ranged from 8 to 10 mm and 25 organisms were used per experimental treatment (five planarians per replicate and 5 replicates). Exposure was carried out in Petri dishes (90×15 mm), containing 20 mL of each experimental solution and control treatment for each microbial insecticide. Planarians were exposed during 96 h in statically system, with constant temperature (22 ± 1°C), in dark and without being fed. Mortality was checked at the end of 48-h (to Bb and Btk) exposure. After the exposure period, the number of dead organisms was registered for each replicate in order to allow estimation of lethal concentrations (LCs 50 ).

Locomotion
The post-exposure effects of both bio-insecticides on planarian locomotor velocity (pLMV) were evaluated individually (using 12 organisms per condition) by placing the organism into a recipient (Ø = 35 cm) containing ASTM hard water medium and covered with gridlines (spaced 0.5 cm per gridline). Locomotor velocity was measured by the number of crossed and re-crossed gridlines over a period of observation of 3 min. Results were expressed as the number of gridlines crossed per minute.

Feeding activity
Nine planarians per condition were individually placed into Petri dish containing 20 mL of ASTM hard water medium and 25 larvae of Chironomus xanthus (total length 0.6 ± 0.1cm, 2°i nstars). The post-exposure feeding rate of planarians was determined by counting the number of chironomids larvae totally digested and consumed over 3 h. Results of feeding rate were expressed as the number of consumed larvae per hour.

Regeneration
After exposure, twelve organisms per treatment were decapitated with a precise single cut behind the auricles. After decapitation, each organism was individually placed in a Petri dish containing 20 mL of ASTM hard water only. Blastema regeneration (i.e., length in mm) was observed after 48-h decapitation and photoreceptors regeneration was monitored eve r y 1 2 h u n t i l c o m p l e t e r e g e n e r a t i o n , u s i n g a stereomicroscopy (MIKROS®) with an eyepiece micrometer. Blastema length was expressed as millimeters after 48-h regeneration, whereas photoreceptor formation was expressed as time in hours needed for formation of photoreceptors.

Reproduction
Sexual reproduction endpoints were determined on same conditions of Btk and Bb concentrations as described above, but the size of test organisms was 1.5 ± 0.1 cm, and the time of exposure 28 days. Four replicates per condition using ten organisms per replicate in a bottle glass containing 100 mL were used for each experimental condition including controls. Planarians were fed once a week followed by renewal of test solutions. The number of cocoons and newborn planarians were registered daily in order to determine fecundity and fertility rates (Knakievicz et al. 2006). Fecundity was expressed as the number of cocoons produced divided by the number of planarians per treatment. Fertility was expressed as the number of offspring produced divided by total cocoons produced.

Statistical analyses
The 48h

Discussion
Exposure to Btk-and Bb-based insecticides decreased the survival of planarians. However, only the estimated LC 50 value for Bb was below the PECsw and the concentrations recommended for its application in the field (EFSA 2015; Agboyi et al. 2020). This shows that adults of G. tigrina, a non-target and non-insect species, seem to be very sensitive to formulations based on Bb, but not so sensitive for Btk when compared to chironomids, Daphnia magna, the frog Leptodactylus latrans, and zebrafish (Table 1).
Survival tests performed with adults of G. tigrina showed that Bb-based formulations increased the mortality of planarians at estimated PECsw demonstrating that care should be taken when using this bio-insecticide near aquatic ecosystems. Data on other aquatic invertebrates, besides insects, is still very scarce in the literature and further toxicity studies should be performed, since such formulations have not been considered to represent a risk to aquatic environments (EFSA 2012;EFSA 2013). Furthermore, the soil fungus B. bassiana is not considered a potential risk to aquatic compartments, which can also be a valuable reason for the limited number of studies found (EFSA 2013).
Although LC 50 data are a simple approach to assess the toxicity of Btk and Bb, it represents the first line of evidence for environmental risk assessment (Lajmanovich et al. 2015). In fact, endpoints assessed on planarians at the organismal and population levels showed to be much more sensitive than its survival, as expected according to Sokolova and co-workers (2012) and other studies previously performed with planarians (Ofoegbu et al. 2016;Rodrigues et al. 2016;Saraiva et al. 2018;Wu and Li 2018;López et al. 2019;Ofoegbu et al. 2019a, b;Saraiva et al. 2020;Dornelas et al. 2020b).
The microbial formulation based on Btk decreased the locomotion, feeding activity, and fecundity of adult planarians for concentrations as low as 12.5, 12.5, and 3.12 mg a.i./L, respectively. Nevertheless, these concentrations affecting behavior and fecundity were still much higher than the PECsw value for Btk. In contrast, Bb-based formulation delayed the regeneration of blastema and decreased the fecundity of planarians at concentrations similar to the predicted environmental concentrations for surface water. Organisms exposed to moderate stress have high metabolic costs and high levels of energy consumption when trying to allocate energy to maintain defense mechanisms (Sokolova et al. 2012;Campos et al. 2017;Monteiro et al. 2019). No information of specific action mode of Btk or Bb regarding planarians is available. However, exposure to bio-insecticides can interfere with the immune response of freshwater invertebrates exacerbated by decreased feeding activity, and consequently the allocation of energy to other physiological processes, in an attempt to overcome an inflammatory response (Bordalo et al. 2020).
Planarians have exceptionally robust regenerative abilities and regulate neoblast proliferation (stem cells) in response to changes in metabolic status and wounding (Elliott and Sanchez Alvarado, 2013;Rink 2013). It is intriguing that the delay in blastema' regeneration of adults was not enough to affect the regeneration of photoreceptors. Some xenobiotics can target very specific pathways, thus not affecting the planarian regeneration system (Hagstrom et al. 2015). The dissimilar responses on behavioral (affected by Btk) and Fig. 1  regeneration (affected by Bb, but not photoreceptor formation) endpoints are not surprising as they have distinct modes of action in insects (Sanahuja et al. 2011;Castagnola and Stock 2014;Mascarin and Jaronski 2016;Sayed and Behle 2017;Rustiguel et al. 2018;Bordalo et al. 2020). In this context, Bordalo et al. (2020) observed different responses between bio-insecticides-based B. thuringiensis and B. bassiana to life history traits in Chironomus riparius larvae. However, fecundity rate of planarians was also affected by both bio-insecticides showing that exposure to both formulations can induce adverse population level effects in planarians.
There is still scarce information on sub-lethal effects of bioinsecticides on freshwater organisms and, to our knowledge, only a few studies were performed addressing development of larvae and emergence on Chironomus xanthus (Dornelas et al. 2020a, b) and Chironomus riparius (Charbonneau et al. 1994;Kästel et al. 2017;Bordalo et al. 2020). These previous studies with C. xanthus showed that Btk-and Bb-based formulations decreased growth rate of larvae and affected emergence, whereas Btk-based formulations did not affect time to emergence on study using C. riparius. Moreover, it was observed for insects that the toxicity of microbial insecticides may vary depending on the host species (Sanjayan et al. 1996;Beetz et al. 2008) and a few studies showed that this might be dependent on the mechanisms of response against bacterial and fungal and more specifically their immune response by recognition of microbial components and the triggering of a cascade of reactions leading to activation of phenoloxidase that is crucial for the melanization of pathogens and repair of damaged tissues (Cerenius et al. 2008;Grizanova et al. 2014). The host needs to cope also with reactive oxygen species and prevent oxidative damage as a consequence of immune response activation (González-Santoyo and Córdoba-Aguilar 2012; Saraiva et al. 2020).
Recent studies report the evaluation of the immune response of bio-insecticides based on B. thuringiensis and B. bassiana on C. xanthus (through total hemocyte count) and C. riparius (through phenoloxidase activity) (Dornelas et al. 2020a;Bordalo et al. 2020). Phenoloxidase activity and oxidative damage have been previously assessed on planarians (Pang et al. 2010;Saraiva et al. 2020). Therefore, further research studies concerning the use of those biochemical endpoints would be desirable to further unravel the mechanisms of response of planarians to bacterial-and fungalbased insecticides, as well as, implications for consumption and allocation of energy. Interestingly, both microbial insecticides decreased the number of cocoons deposited by each planarian, but the number of newborns originated on each cocoon was not affected, showing that their development was not compromised by both formulations.
Overall, toxicity and sub-lethal effects of B. bassiana have been demonstrated at PECsw which leads us to conclude that those formulations are not harmless to freshwater compartments and might not be safe to non-target freshwater invertebrate species. Concerning Btk-based formulation, further studies should be pursued, since its use is increasing considerably, and our current study showed deleterious effects on planarians' reproduction at much lower concentrations than the ones causing acute toxicity. Therefore, our study provides important information for the risk assessment of bioinsecticides in freshwater ecosystems. To our knowledge, this is the first study concerning the reproductive effects of Btkand Bb-based formulations in a non-target, non-insect freshwater invertebrate. Finally, the results obtained in this study are important to understand the possible direct effects of Fig. 3 Fecundity rate (mean ± SEM) of G. tigrina exposed during 4 weeks to sub-lethal concentrations of B. B. thuringiensis and B. bassiana on non-target aquatic organisms and validate the use of Girardia tigrina as a potential bioindicator species.

Conclusions
Our results showed evidence that microbial insecticides based on Bb might not be as ecologically friendly as previously suggested especially if used in the vicinity of aquatic ecosystems. This study also highlights the importance of using planarians as bioindicators of environmental contamination by bio-pesticides. Given the effects observed on fecundity rates of planarians exposed to environmentally relevant concentrations of Bb and Btk, it is critical to consider reproductive endpoints to better evaluate how microbial insecticides affect natural populations of non-target freshwater invertebrates.
Data and materials availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Author contribution LCRS conducted experiments, analyzed data, and wrote the manuscript (writing-original draft, review and editing). ASPD conducted experiments. ASS, ASPD, CG, JLTP, AMVMS, and RAS analyzed data and review and edit the manuscript; RAS and AMVMS conceived (funding acquisition) and designed research. All authors read, made corrections, and approved the manuscript.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication Not applicable.
Competing interests The authors declare no competing interests.