Secondary Metabolites and the Risks of Isaria fumosorosea and Isaria farinosa

Isaria fumosorosea and Isaria farinosa are important entomopathogenic fungi with a worldwide distribution and multiple host insects. However, the concerns about the safety risks of myco-pesticides have been attracting the attention of researchers and consumers. Secondary metabolites (SMs), especially the mycotoxins, closely affect the biosafety of Isaria myco-insecticides. In the last forty years, more than seventy SMs were identified and isolated from I. fumosorosea and I. farinose. The SMs of I. fumosorosea include the mycotoxins of non-ribosomal peptides (NRPs) (beauvericin and beauverolides), terpenes (trichocaranes and fumosorinone), lactone compounds (cepharosporolides), acids (dipicolinic acid and oxalic acid), etc. Meanwhile, the NRP mycotoxins (cycloaspeptides) and the terpene compounds (farinosones and militarinones) are the main SMs in I. farinosa. Although several researches reported the two Isaria have promised biosafety, the bioactivities and the safety risks of their SMs have not been studied in detail so far. However, based on existing knowledge, most SMs (i.e., mycotoxins) do not come from Isaria myco-insecticide itself, but are from the host insects infected by Isaria fungi, because only the hosts can provide the conditions for fungal proliferation. Furthermore, the SMs from Isaria fungi have a very limited possibility of entering into environments because many SMs are decomposed in insect cadavers. The biosafety of Isaria myco-insecticides and their SMs/mycotoxins are being monitored. Of course, SMs safety risks of Isaria myco-insecticides need further research.


Introduction
Isaria fumosorosea and Isaria farinosa-formerly known as Paecilomyces fumosoroseus and Paecilomyces farinosus, respectively-are important entomopathogenic fungi with a worldwide distribution and multiple host insects [1,2]. Although differing from the popular Beauveria bassiana and Metarhizium anisopliae species thoroughly researched in various areas, both I. fumosorosea and I. farinosa attract more attention. They have multiple hosts, do not show harmful effects linked to the use of chemical pesticides, and are considered to be environmentally friendly [3]. Besides their application as pest biocontrol agents, there were some experiments indicating the both fungi have potential uses in the biotransformation of flavonoids glycosides, steroids, etc. [4][5][6].
The terpene compound, fumosorinone (10) (Table 1, Figure 1) was isolated from the ACCC37775 strain of I. fumosorosea (Hebei University, Baoding, China). Fumosorinone (10) is structurally similar to tenellin and desmethylbassianin, but has different chain length and degree of methylation. Fumosorinone (10) acts as a classic non-competitive inhibitor of protein tyrosine phosphatase 1B (PTP1B) with an IC 50 of 14.04 µM, which suggests that it is a potential medicine for the treatment of type II diabetes and other associated metabolic syndromes. The gene cluster of fumosorinone biosynthesis includes a hybrid polyketide synthase-nonribosomal peptide synthetase gene, two cytochrome P450 enzyme genes, a trans-enoyl reductase gene, and other two transcription regulatory genes [42]. Fumosorinone (10) also showed cytotoxic against human cancer lines, including HeLa, MDA-MB-231, and MDA-MB-453 cell lines [43]. A compound similar to fumosorinone, fumosorinone A (11) (Table 1, Figure 1), was identified as well. It is also a PTP1B inhibitor [31].
The other acids, dipicolinic acid (DPA) (16) and oxalic acid (OXA) (17) ( Table 1, Figure 1), were found in the I. fumosorosea Pfrd strain (Centro Nacional de Referencia de Control Biológico, Tecomán, Colima, Mexico). DPA (16) was the most abundant metabolite with insecticidal activity against the third-instar nymphs of the whitefly in bioassays involving topical applications. DPA (16) was detected after 24 h when the fungus started growing in submerged cultures. The production of DPA (16) was directly correlated with fungal growth, but the maximal yield was only 0.041 g/L [45]. In submerged fermentation, carbon was significantly directed towards the synthesis of DPA (16) and OXA (17), especially under zinc limitation [46]. OXA (17) has antimicrobial and antioxidant activities [47,48] and can delay the sclerotial formation of Polyporus umbellatus [49], which is called as "Zhuling", a traditional Chinese medicine used for a wide range of ailments related to the edema, scanty urine, vaginal discharge, urinary dysfunction, jaundice, and diarrhea [50].
Peroxy-ergosterol (22) (Table 1, Figure 1) was isolated from the RCEF1253 strain of I. fumosorosea (Anhui Agricultural University, Hefei, China) by high-speed-counter-current chromatography [52]. It has various bioactivities, such as cytotoxicity to cancer cells P-388, KB, A549, and HT-29, with ED50s of 0.4, 2.1, 2.7, and 1.4 µg/mL [53]. It also induces the apoptosis of the human leukemia cell HL-60 [54]. Furthermore, this fungus has a relatively high vitamin A content, which shows that it is a potential producer of vitamin A [55].

Secondary Metabolites (SMs) from Isaria farinosa
Cycloaspeptides F (25) and G (26) (Table 2, Figure 3), two new cyclic pentapeptides, and the known cycloaspeptides A (27), C (28), and bisdethiodi (methylthio) hyalodendrin (29) (Table 2, Figure  3) were isolated from the fermented rice substrate with the I. farinosa strain XJC04-CT-303 (Institute of Microbiology, Chinese Academy of Sciences, Beijing, China) that colonizes Cordyceps sinensis. Cycloaspeptides F (25) and G (26) inhibited the growth of MCF7 cells, which was comparable to the positive control 5-fluorouracil. They also had modest cytotoxic effects on HeLa cells [59]. Cycloaspeptide A (27) has a low cytotoxicity in human lung fibroblasts [60]. Cycloaspeptide C (28) is closely related to cycloaspeptide G (26), but its bioactivity is not reported. The gene cluster responsible for the biosynthesis of the cycloaspeptides were identified in Penicillium soppii and Penicillium jamesonlandense. Heterologous expression in Aspergillus oryzae has demonstrated that the minimal gene set required to produce both cycloaspeptide A and cycloaspeptide E is a 5-module NRPS and a new type of pathway-specific N-methyltransferase (N-MeT). Gene knock-outs and feeding studies have demonstrated that two modules of the NRPS preferentially accept and incorporate N-methylated amino acids, which are provided by the pathway-specific N-MeT. This is

Secondary Metabolites (SMs) from Isaria farinosa
Cycloaspeptides F (25) and G (26) (Table 2, Figure 3), two new cyclic pentapeptides, and the known cycloaspeptides A (27), C (28), and bisdethiodi (methylthio) hyalodendrin (29) (Table 2, Figure 3) were isolated from the fermented rice substrate with the I. farinosa strain XJC04-CT-303 (Institute of Microbiology, Chinese Academy of Sciences, Beijing, China) that colonizes Cordyceps sinensis. Cycloaspeptides F (25) and G (26) inhibited the growth of MCF7 cells, which was comparable to the positive control 5-fluorouracil. They also had modest cytotoxic effects on HeLa cells [59]. Cycloaspeptide A (27) has a low cytotoxicity in human lung fibroblasts [60]. Cycloaspeptide C (28) is closely related to cycloaspeptide G (26), but its bioactivity is not reported. The gene cluster responsible for the biosynthesis of the cycloaspeptides were identified in Penicillium soppii and Penicillium jamesonlandense. Heterologous expression in Aspergillus oryzae has demonstrated that the minimal gene set required to produce both cycloaspeptide A and cycloaspeptide E is a 5-module NRPS and a new type of pathway-specific N-methyltransferase (N-MeT). Gene knock-outs and feeding studies have demonstrated that two modules of the NRPS preferentially accept and incorporate N-methylated amino acids, which are provided by the pathway-specific N-MeT. This is a system not previously seen in secondary metabolism [61]. The diketopiperazine derivative, bisdethiodi (methylthio) hyalodendrin (29) (gliovictin), was isolated in 1973. It exhibited weak cytotoxic activity on KB (human epidermoid carcinoma of the mouth) with IC 50 of 42 µg/mL, and had IC 50 values of >50 µg/mL on HepG2 (human hepatocellular liver carcinoma cell line), A549 (human lung carcinoma cell line), HCC-S102 (hepatocellular carcinoma cell line), HuCCA-1 (human cholangiocarcinoma cancer cells), HeLa (cervical adenocarcinoma cell line), MDA-MB231 (human breast cell line), T47 D (human mammary adenocarcinoma cell line), HL-60 (human promyelocytic leukemia cell line), and P388 (murine leukemia cell line) [62].
A new maleimide-bearing compound, farinomalein (39) (Table 2, Figure 3), was isolated from the strain HF599 (National Institute of Fruit Tree Science, Tsukuba, Japan). It showed potent activity against the plant pathogen Phytophthora sojae [68]. It can be synthesized in two steps from a readily available γ-hydroxybutenolide [69]. The activities against Phytophthora sojae and Aphanomyces cochlioides were confirmed [70].
I. farinosa can produce a water soluble anthraquinone-related red pigment with good stability after being exposed to salt solution (96.1% stability after treatment with sodium chloride), acid (72.1% stability with citric acid), heat (86.2% stability at 60 • C), and sunlight (99.4% stability). It shows a potential for pigment production [74,75].

Risks of the Secondary Metabolites (SMs) from both Isaria Myco-Insecticides
In recent years, the concerns about the safety risks of myco-pesticides and their SMs have been attracting the attention of researchers and consumers. In fact, the popular myco-insecticides, Beauveria bassiana. and Metarhizium anisopliae, have been proposed as low-risk environmental

Risks of the Secondary Metabolites (SMs) from both Isaria Myco-Insecticides
In recent years, the concerns about the safety risks of myco-pesticides and their SMs have been attracting the attention of researchers and consumers. In fact, the popular myco-insecticides, Beauveria bassiana. and Metarhizium anisopliae, have been proposed as low-risk environmental alternatives to chemical insecticides for controlling agricultural pests and disease vectors [27,28]. This is because their safety for humans and the environment were well evaluated [76][77][78][79], while the mycotoxins they produced were considered unlikely to enter food chains [80].
However, there were few experiments on the safety of Isaria myco-insecticides. I. fumosorosea was the subject of two reports involving safety analysis, while I. farinosa has not been paid attention yet. The I. fumosorosea monospore culture EH-506/3 (BMFM-UNAM 834, Universidad Nacional Autonoma de Mexico, Mexico City, Mexico) was subjected to a biosafety test by applying a 2 g/kg of animal body weight dose on the shaved skin of 16 New Zealand rabbits, with an exposure time of 24 h. The results indicated that none of the rabbits showed clinical signs of any disease, and their body weight corresponded to the expected weight for a healthy rabbit. The test data supports the safety of I. fumosorosea EH-506/3 when applied to the skin [29]. Another toxicity test on I. fumosorosea was completed in China. The toxicities of acute oral, dermal, and inhalation to rats were recorded as LD 50 > 5000 mg/kg, LD 50 (4 h) > 2000 mg/kg, and LC 50 (2 h) > 2000 mg/m 3 , respectively. No irritation action was observed in rabbit eyes, and no dermal sensitization reaction was found on the treated rabbit skin. These results suggested that I. fumosorosea has low toxicities of acute oral, dermal toxicity, and inhalation, and it can be graded as a weak sensitizer [81].
Overall, there are six destinations (i.e., target organisms, non-target organisms, soil, water, atmosphere, and humans) involved in the production and application of Isaria myco-insecticide formulations ( Figure 4). The most important destination is target organisms, including the pests and crops, when Isaria myco-insecticide is released in fields. In practice, a few fungal spores of myco-insecticide probably land on insect surfaces, but the fungus will proliferate on the infected insect hosts, which suffer a pathogenic process from spore germination and the formation of the next generation of spores. Many mycotoxins of entomopathogenic fungi are probably biosynthesized in hemocoel to conquer the host's immunity [82]. The target crops are probably the main destination, especially when the Isaria myco-insecticide is applied by stem-leaf treatment in fields. This is because most of the fungal spores are dropped on the plants with canopy covering the ground. In addition, the endophytic characteristics of entomopathogenic fungi might produce some SMs because Isaria fungi, similar to B. bassiana and M. anisopliae, can colonize plants [83][84][85]. Non-target organisms are an important destination of myco-insecticides as well. They represent a big category, including animals, plants, and microbes, which are not the targets of myco-insecticides but have chances to contact myco-insecticides. Among them, non-target insects might be the most important destination, because some of these insects probably are the hosts of myco-insecticidal fungi. There have been many reports published giving evidence that entomopathogenic fungi infect silkworms [86], bees [87,88], and natural enemy insects [89]. Of course, more studies found that the myco-insecticides are safe to non-target insects if they are used correctly [90,91].
Soil is another important destination, especially when Isaria myco-insecticide is released through soil treatments in fields (Figure 4). Through the drifting pathway from application and dropping pathway from target pest cadavers, fungal phages and mycotoxins can enter the soil system. In fact, entomopathogenic fungi can persist and survive in soil for a long time [92][93][94][95]. Beauveria spp., Metarhizium spp., Paecilomyces spp., and Isaria spp. can be often isolated from the soil [1]. The entomopathogens in soil can be detected after myco-insecticides are used [96], but there are no reports about the growth and proliferation of Isaria fungi and the presence of their SMs in soil. There are also no cases of soil fungi affecting human health.
Water and atmosphere are the destinations of the drifting myco-insecticides ( Figure 4). In fact, many entomopathogenic fungi can persist and survive in water. I. farinosa and I. fumosorosea, similarly to M. anisopliae, can infect aquatic insects like mosquitoes [97,98]. However, there are few studies about SMs of entomopathogenic fungi in water. Milner et al. reported that the Metarhizium biopesticide is very unlikely to pose any hazard to aquatic organisms [99]. The atmosphere is another destination of Isaria myco-insecticides, where fungal entomopathogens are obtained from drifting myco-insecticides and the spore dispersal of natural fungi. However, fungi cannot either persist for a long time or proliferate in the air [100]. Also, fungi might be exchanged between soil, water, and atmosphere systems, although there are few studies on this aspect. To the date, there are also no cases of entomopathogenic fungi from water and atmosphere influence on human health.
Humans contact Isaria myco-insecticides through direct and indirect ways (Figure 4). Only people who are involved in the production process of myco-insecticides or use them in farms directly contact the fungi. Several studies have reported that fungal spores lead to allergies in workers who worked in factories of B. bassiana and M. anisopliae for long periods [27,28]. Perhaps, most people indirectly contact the fungi through foods, soil, the atmosphere, and water contaminated by the fungi. However, there have not been any case reports about people's health affected by indirectly contacting entomopathogenic fungi.
Undoubtedly, the biosafety risks of Isaria myco-insecticides are closely related to the sources and fates of the SMs (especially the mycotoxins) produced by entomopathogenic fungi.
In fact, the SMs of myco-insecticide itself are very limited. Because the active ingredients of myco-insecticide formulations are the spores of the fungal entomopathogen, the fungi cannot proliferate in the formulation and cannot produce new SMs. Most SMs possibly exist in the spore cells rather than outside the spores [80]. Therefore, the main sources of SMs basically include the target pests or host insects infected by fungal entomopathogen of myco-insecticide, because the host insects support the fungi with conditions for proliferation. The source of SMs and mycotoxins is the growing entomopathogenic filamentous fungus. In addition, endophytic entomopathogenic fungi might be an SM source, because some Isaria strains can colonize plants and become endophytic fungi [101,102].
Currently, we do not know the detailed fates of the fungal SMs of myco-insecticides. Obviously, there are certain possibilities that the SMs of entomopathogenic fungi enter environments, however, there have not been any reports that show evidence of the entry of SMs from myco-insecticides into environments. In fact, a few research cases indicate that mycotoxins are scarcely released into environments from insects. For example, destruxin analogues were shortly decomposed by M. anisopliae after the host insects died, which was presumably due to the activity of hydrolytic enzymes in the insects' cadavers. This appeared to be independent of host or soil type and biota. The study supported that destruxins are essentially restricted to the host and pathogen and are unlikely to contaminate the environment or enter the food chain [103,104].
To date, it has been found that most mycotoxins that contaminate environments and food chains come from the crops and products infected by fungal phytopathogens, such as Fusarium spp., Aspergillus spp., etc., rather than fungal entomopathogens [38,105], despite the fact that both phytopathogenic and entomopathogenic fungi often produce the same mycotoxins [106]. For example, Schenzel et al. reported that beauvericins were detected in drainage water where wheat was inoculated with Fusarium spp., which is a producer of beauvericins [107].
atmosphere systems, although there are few studies on this aspect. To the date, there are also no cases of entomopathogenic fungi from water and atmosphere influence on human health.
Humans contact Isaria myco-insecticides through direct and indirect ways (Figure 4). Only people who are involved in the production process of myco-insecticides or use them in farms directly contact the fungi. Several studies have reported that fungal spores lead to allergies in workers who worked in factories of B. bassiana and M. anisopliae for long periods [27,28]. Perhaps, most people indirectly contact the fungi through foods, soil, the atmosphere, and water contaminated by the fungi. However, there have not been any case reports about people's health affected by indirectly contacting entomopathogenic fungi.
Undoubtedly, the biosafety risks of Isaria myco-insecticides are closely related to the sources and fates of the SMs (especially the mycotoxins) produced by entomopathogenic fungi.
In fact, the SMs of myco-insecticide itself are very limited. Because the active ingredients of myco-insecticide formulations are the spores of the fungal entomopathogen, the fungi cannot proliferate in the formulation and cannot produce new SMs. Most SMs possibly exist in the spore cells rather than outside the spores [80]. Therefore, the main sources of SMs basically include the target pests or host insects infected by fungal entomopathogen of myco-insecticide, because the host insects support the fungi with conditions for proliferation. The source of SMs and mycotoxins is the growing entomopathogenic filamentous fungus. In addition, endophytic entomopathogenic fungi might be an SM source, because some Isaria strains can colonize plants and become endophytic fungi [101,102].
Currently, we do not know the detailed fates of the fungal SMs of myco-insecticides. Obviously, there are certain possibilities that the SMs of entomopathogenic fungi enter environments, however, there have not been any reports that show evidence of the entry of SMs from myco-insecticides into environments. In fact, a few research cases indicate that mycotoxins are scarcely released into environments from insects. For example, destruxin analogues were shortly decomposed by M. anisopliae after the host insects died, which was presumably due to the activity of hydrolytic enzymes in the insects' cadavers. This appeared to be independent of host or soil type and biota. The study supported that destruxins are essentially restricted to the host and pathogen and are unlikely to contaminate the environment or enter the food chain [103,104].
To date, it has been found that most mycotoxins that contaminate environments and food chains come from the crops and products infected by fungal phytopathogens, such as Fusarium spp., Aspergillus spp., etc., rather than fungal entomopathogens [38,105], despite the fact that both phytopathogenic and entomopathogenic fungi often produce the same mycotoxins [106]. For example, Schenzel et al. reported that beauvericins were detected in drainage water where wheat was inoculated with Fusarium spp., which is a producer of beauvericins [107].  In conclusion, there are more than seventy SMs identified and isolated from I. fumosorosea and I. farinosa. Many of these are mycotoxins attracting people's concerns about the biosafety. The SMs of I. fumosorosea include the NRP mycotoxins (beauvericin and several beauverolides), terpenes (several trichocaranes and fumosorinone), lactone compounds (several cepharosporolides), and acids (dipicolinic acid and oxalic acid). In I. farinosa, the NRPs (several cycloaspeptides) and terpenes  In conclusion, there are more than seventy SMs identified and isolated from I. fumosorosea and I. farinosa. Many of these are mycotoxins attracting people's concerns about the biosafety. The SMs of I. fumosorosea include the NRP mycotoxins (beauvericin and several beauverolides), terpenes (several trichocaranes and fumosorinone), lactone compounds (several cepharosporolides), and acids (dipicolinic acid and oxalic acid). In I. farinosa, the NRPs (several cycloaspeptides) and terpenes indicates a pathway that has not been found to date. (SMs = secondary metabolites) (modified based on Hu et al. (2016) [81]).
In conclusion, there are more than seventy SMs identified and isolated from I. fumosorosea and I. farinosa. Many of these are mycotoxins attracting people's concerns about the biosafety. The SMs of I. fumosorosea include the NRP mycotoxins (beauvericin and several beauverolides), terpenes (several trichocaranes and fumosorinone), lactone compounds (several cepharosporolides), and acids (dipicolinic acid and oxalic acid). In I. farinosa, the NRPs (several cycloaspeptides) and terpenes (several farinosones and militarinones) were the main SMs. Currently, the bioactivities and mechanisms of action of the SMs in both Isaria have not been well studied, and neither have the risks of these compounds been carefully assessed. However, it is indicated that most SMs (mycotoxins) come from the host insects infected by Isaria fungi rather than the Isaria myco-insecticide itself, because the hosts provide all the conditions for fungal proliferation. Furthermore, the possibility of SMs from Isaria fungi entering into environments is very limited, because many SMs are decomposed in insect cadavers. Although more careful research in the future is essential, the biosafety of Isaria myco-insecticides and their SMs/mycotoxins in current is under control.