Ketamine can be produced by Pochonia chlamydosporia: an old molecule and a new role

Background: Infection by nematodes is a problem for human health, livestock, and agriculture, as it causes decits in host health, increases production costs, and incurs less food supply. The control of these parasites is usually done using anthelmintics, which, in most cases, have not been fully effective. Therefore, the search for new molecules with anthelmintic potential is necessary. Methods: In the present study, we isolated and characterized molecules from the nematophagous fungus Pochonia chlamydosporia and tested these compounds on three nematodes: Caenorhabditis elegans, Ancylostoma ceylanicum, and Ascaris suum. Results: The ethyl acetate extract showed nematicidal activity on the nematode model C. elegans. We identied the major substance present in two subfractions of this extract as ketamine. Then, we tested this compound on C. elegans and the parasites A. ceylanicum and A. suum using hamsters and mice as hosts, respectively. We did not nd a difference between the animal groups when considering the number of worms recovered from the intestines of animals treated with ketamine and albendazole (P (cid:0) 0.05). The parasite burden of larvae recovered from the lungs of mice treated with ketamine was similar to those treated with ivermectin. Conclusion: These results presented here demonstrate the nematicidal activity of ketamine in vitro and in vivo. Thus, conrming the nematicidal potential of the molecule present in the fungus P. chlamydosporia may consist a new method of controlling parasites.


Background
Nematode infection is a problem for human health, as it contributes negatively to growth and cognitive development. In animal production and agriculture, damages go beyond organic de cits because they incur in increased production costs and consequently less food supply [1][2][3]. The control of these parasites is usually carried out using anthelmintics that, in most cases, have not been totally effective. In addition, resistance to the main group of drugs used in the treatment and control is already a consolidated problem [4]. Thus, the search for new drugs and control alternatives is necessary.
Several studies in the literature have described the direct or indirect nematicidal action of some organisms, such as fungi, featuring one of the alternatives of biological control. The administration of nematophagous fungi is considered a promising alternative in the prophylaxis of parasitic gastrointestinal helminthiasis [5]. Willcox and Tribe [6] reported the association between the fungus Pochonia chlamydosporia and the nematode Heterodera schachtii, as the primary and secondary metabolites of nematophagous fungi have been known to be involved in this interaction [5,7,8]. In the review presented by Li et al. [9], 179 substances from fungi, including nematophagous, were reported as potential nematicidal drugs; therefore, these fungi can be considered an important source of new drugs.
In the history of drug production and discovery, often a molecule that is characterized with a well-de ned function can, with new studies, be employed for a new purpose. This practice is called drug repositioning [10]. In helminthology, some drugs that were not originally used against helminths have shown high e cacy against some worms, such as me oquine for Echinococcus multilocularis [11]. Sertraline, paroxetine, and chlorpromazine have long histories of clinical use as antidepressant or antipsychotic medicines. However, according to Weeks et al. (2018) [12], they may represent new classes of anthelmintic drugs. In this perspective, other drugs that act via ion channels or neurotransmitter receptors may have promising functions for helminth control. Here, we extracted, characterized, and evaluated P. chlamydosporia metabolites against nematodes. In these studies, it is important to have good models to test drug effectiveness. In this context, Caenorhabditis elegans is the main model for a drug-screening study, since it presents characteristics that facilitate the assays, such as the feasibility of laboratory maintenance [13] and the taxonomic and functional proximity to important parasitic nematodes [14]. However, validation of the experiments in an animal model is necessary. It is important to consider that we also carried out tests on two parasitic nematodes, Ancylostoma ceyalnicum and Ascaris suum, using the infection model, hamsters and mice, respectively. Therefore, in the present work, we have extracted and characterized a new metabolite of P. chlamydosporia: the molecule ketamine. We evaluated this molecule on nematodes, C. elegans, A. suum, and A. ceylanicum, and, here, we present unprecedented results of its nematicide activity.

Fungus cultivation
The P. chlamydosporia used in this work was kept in the fungal collection of the Veterinary Parasitology Laboratory of the Federal University of Viçosa. Culture discs with an approximate diameter of 4 mm were extracted from the tube containing the fungal isolate and then transferred to a 250-mL Erlenmeyer ask containing the potato dextrose liquid medium (Difco). These were incubated in a horizontal shaker (Tecnal) at 28 °C under agitation at 120 rpm, and, after 20 days, this material was ltered to separate the extract from the mycelial mass. The separated mass was used in the extraction process by maceration.
To obtain the methanolic extract, 400 g of the mycelium (a product of approximately 30 L of fungal culture) had been submitted to cold triple maceration for 7 days in methanol. The methanolic extract (MeOH) was concentrated under reduced pressure and lyophilized, obtaining 4.1 g of methanolic extract. 3 g of this extract were solubilized in MeOH, and then this solution was partitioned with dichloromethane, followed by ethyl acetate, affording the fractions CH 2 Cl 2 (0.61 g) and EtOAc (0.50 g). 0.4 g of the ethyl acetate fraction (EtOAc) was further fractionated by column chromatography in subfractions using silica gel as the stationary phase and then eluted with hexane: ethyl acetate gradients of increasing polarity. These subfractions were concentrated under reduced pressure in a rotary evaporator and lyophilized (Table 1).

GC-MS analysis
The ethyl acetate fraction and its seven subfractions obtained from the puri cation by column chromatography (F1-F9) were analysed by gas chromatography coupled with mass spectrometry (GC/MS). The analyses were performed using a Shimadzu GCMSQP2010 Plus, equipped with an AOC-10 automatic injection system. In the experiments, a 30-m-long capillary column Rxi-1 (100% polydimethylsiloxane), with an internal diameter of 0.25 mm and a lm thickness of 0.25 µm, was used. The injector temperature was 250 °C, and the temperature programme ranged from 150 °C to 300 °C at 3 °C/min. The injected volume was 1 µL in split-mode at a ratio of 10:1. MS analysis was carried out in a quadrupole MS system (QP-2010plus) operating at 70 eV under the same conditions described above.
Compounds were identi ed through comparison with mass spectral data in NIST 08 libraries.

UPLC-ESI-MS/MS
4.0 mg of the ethyl acetate fraction and the standard molecule (ketamine) were weighed directly in a microtube and 1 mL of methanol/water (1:1). The HPLC grade was added until reaching complete solubilization. After that, the ethyl acetate sample was diluted 4 x and ketamine was diluted 20 x, and then they were automatically injected onto the system.
The chemical composition of ethyl acetate fractions was investigated by ultra-performance liquid chromatography coupled with mass spectrometry in a Waters ACQUITY UPLC system (Waters, Milford, MA, USA) composed of a binary pump, auto sampler, in-line degasser, and photodiode array detector (190-500 nm) (Waters, USA). The analyses were performed on an Acquity UPLC BEH C18 column (2.1 × 50 mm i.d., 1.7 µm, Waters) at 40 °C, eluting with a linear gradient of water (A) and acetonitrile (B), both containing 0.1% v/v formic acid, at a ow rate of 0.3 mL/min (e.g., 5 to 95% of B in 10 min) before returning to the initial conditions in 1 min. A re-equilibration time of 2 min was maintained between runs. As previously described, our research group employed the spectroscopic conditions [15].
A mass spectrometer Xeco TM Triple Quadrupole MS (Waters Corp., Milford, USA) equipped with an electrospray ionization (ESI) source, operating in a negative and positive ionization mode, was used in the analysis. The cone gas ow was set to 90 L/h, and desolvation gas ow was set to 900 L/h at 350 °C.
The capillary voltage was set to 3.54 kV, cone gas voltage was set to 27 V, and source temperature was set to 120 °C. The data were accomplished in the scan mode, with mass range adjusted to m/z 100-700 Da. Ketamine (Holliday -Scott ®) was employed as the reference compound. The UPLC-ESI-MS/MS analysis was performed twice with the fungus grown at different times for con rmation of ketamine's presence.
Culturing Caenorhabditis elegans larvae and motility test The strain of C. elegans used in the experiment was kindly provided by Professor Carlos Winter of Universidade de São Paulo (USP). L3 larvae of C. elegans were grown on 8P NGM plates according to the methodology previously described [16]. After 7 days of culturing in a BOD incubator at 20 °C, the plates were washed with an M9 medium and ltered through three sieves with 40-, 30-, and 20-µm pores. L3 larvae retained in the 20-µm strainer were collected by backwashing. The obtained larvae were washed by centrifugation at 700 g for 4 min, followed by two washes with an M9 medium. For the nematicidal assay against C. elegans, the L3 was resuspended in M9, and approximately 1,000 larvae in 100 µL of suspension were added to each well in a 96-well microplate. The ethyl acetate extract was added in concentrations of 1,000, 100, 10, 1, 0.1, and 0.01 µg/mL. For methanol and dichloromethane, concentrations of 5,000 and 3,000 µg/mL were added, and ketamine was used up until the concentration of 20,000 µg/mL. In the negative control, the M9 medium was used with 0.05% dimethyl sulfoxide, and, for the positive control, ivermectin was applied in the same concentrations used for the tested extracts (1,000-0.01 µg/mL). Plates containing compounds and larvae were stored in a BOD incubator at 20 °C. After 48 h, 10 µL of a solution containing approximately 100 larvae were removed from each well for analysis, and quanti cation of the total number of paralyzed larvae was carried out using an optical microscope at 100 × magni cation. Larvae were considered paralyzed when presenting a straight body and absence of any mobility.

Activity of ketamine on Ancylostoma ceylanicum
For this experiment, 50 female hamsters (Mesocricetus auratus), 4 to 6 weeks old, were divided into ve experimental groups containing 10 animals each. All animals were maintained in a controlled environment and infected with 75 third-stage larvae (L3) of A. ceylanicum at day 0 of the experiment. After 20 days of infection, when all animals were excreting eggs in the faeces, they were treated for 5 days, orally, according to the following experimental design: Group 1, PBS (phosphate buffer saline); Group 2, albendazole (Nova Química®) 0.60 mg; Group 3, ketamine 0.06 mg; Group 4, ketamine 0.60 mg; and Group 5, ketamine 6 mg (Holliday -Scott ®). The EPG (eggs per gram of faeces) was performed to monitor the quanti cation of eggs throughout the experiment. This quanti cation was performed every 2 days using the McMaster chamber [17]. After the treatment period, the animals were euthanized using an overdose of anaesthetic (45 mg/kg of xylazine associated with 240 mg/kg of ketamine, intraperitoneally). Then, the small intestine was removed and placed in a Petri dish containing PBS, and it was opened longitudinally for the recovery of adult worms.

Activity of ketamine on infection of Ascaris suum larvae
In this experiment, 60 mice of the C57Bl6 lineage were used, being divided into two groups of 30 animals: a group of 30 animals with treatment administered subcutaneously and the other group by oral route (gavage). The animals had been infected with 2,000 eggs of A. suum larvae and divided into groups, with six animals each, that received the following treatments: Group 1 PBS (phosphate buffered saline); Group 2 ivermectin 0.05 mg (Ouro Fino®); Group 3 ketamine 0.05 mg; Group 4 ketamine 0.25 mg; and Group 5 ketamine 0.50 mg (Holliday -Scott ®). The treatment was carried out for 7 days, except for the subcutaneous ivermectin group that received a single dose.
On the eighth day after infection, the animals were euthanized using an anaesthetic overdose (45 mg/kg of xylazine associated with 240 mg/kg of ketamine, intraperitoneally). To recover and quantify the number of larvae present in the lungs, the lung tissue collected was triturated in the Petri dish and placed in the Baermann apparatus for 4 h at 37 °C with PBS. After incubation, the sediment was collected and centrifuged at 600 xg for 10 min at 20 ºC, and the larvae were xed in 10% formalin and counted in the optical microscope (Leica Microsystems®) with a 100x magni cation.

Statistical analyses
The data of the recovered nematodes were submitted to the normality test, followed by the analysis of variance and Tukey's test; nonlinear regression analysis was used to calculate the ED 50 value utilized of a four parameter sigmoid curve using the statistical programme GraphPad Prism 7.0.

Results
To evaluate the nematicidal activity of P. chlamydosporia, we tested the extracts obtained from the fungus on C. elegans larvae. Table 2 shows the results of the activity of three extracts, and it can be seen that EtOAc was presented as the most promising of them. Although CH 2 Cl 2 and MeOH also showed a larval paralysis percentage above 50%, they had high IC 50 values when compared to the value observed for EtOAc. We observed that EtOAc was able to paralyze 84% of the larvae at a concentration of 1,000 µg/mL, whereas CH 2 Cl 2 and MeOH did not show relevant activity at this same concentration, as it was necessary to perform the test at 5,000 µg/mL for these two extracts.
As EtOAc was the most promising, we fractioned it further, obtaining seven subfractions (Table 1). In the motility test, we observed that only F4 and F5 were active, and it presented the following ED 50 values: 50.9 µg/mL and 210.5 µg/mL, respectively (Fig. 1). The EtOAc and F4 and F5 reached the following percentages of immobility: 84, 85.5, and 75.7%, respectively, using the concentration of 1,000 µg/mL. After the fractionation of EtOAc, the activity remained in F4 and F5. 1 H NMR analysis of these samples (Fig. 1S) revealed characteristic signals of fatty acids and aromatic compounds in EtOAc. Using the GC/MS technique, we could identify the main substances present in EtOAc (Table 3). In subfraction 4, we identi ed ketamine and the acids cis-7-hexadecenoic, palmitic, and linoleic acid; however, there were no identi ed substances other than ketamine in F5 (Fig. 6S). The subfractions F1, F2, F7, F8, and F9 presented the same fatty acids but not ketamine (dates not showing).
Ketamine was present in the EtOAc fraction and in subfractions F4 and F5 only, and these subfractions showed paralyzing activity on C. elegans larvae. As some CG/MS analyses revealed greater purity in F5, we chose to perform two-dimensional NMR analyses for this sample. We con rmed the presence of ketamine F5 by HSQC, HMBC, and COSY (Table 1S and (Table 4).
Once the ketamine was commercially available and the production of metabolites in biomass quantity produced by the fungus was low, we also employed the ketamine standard on C. elegans larvae. Figure 1 shows the activity of the drug on C. elegans. As the commercial ketamine was active on the nematode model, we tested it in vivo on the animal models infected with A. ceylanicum, and A. suum. Figure 2 shows that ketamine exhibited activity on A. ceylanicum at the concentration of 6 mg, and it reduced worm burden parasites to approximately zero; thus, there is no difference (P 0.05) between the groups treated with albendazole (0.6 mg) and ketamine (6 mg). The EPG of the groups treated with albendazole (0.6 mg) and ketamine (6 mg) was zero after 4 days of treatment (Fig. 2). Figure 3 shows the ability of ketamine to impair the infection of A. suum larvae in C57Bl/6 mice. The averages of larvae recovered in the lungs of these animals are shown in this gure. Ketamine showed nematicidal activity in the two routes of the drug's administration, oral and subcutaneous, with no difference (P 0.05) between the administration routes. In addition, when ketamine was administered orally or subcutaneously, there was no difference in the number of larvae recovered (P 0.05) between ketamine (0.5 mg) and ivermectin.

Discussion
There are few investments in the discovery of new drugs for the treatment of parasites. Considering that there are works in the literature relating nemathophagous fungi extracts with nematicidal activity and commercial products using nemathophagous fungi [7,18,19], we performed extraction of the P. chlamydosporia metabolites with an organic solvent; the extract was fractionated with solvents of different polarity scale and evaluated the compounds for nematicidal capacity. We observed that EtOAc was able to paralyze 84% of the C. elegans larvae at a concentration of 1,000 µg/mL, whereas CH 2 Cl 2 and MeOH did not show any relevant activity at the same concentration, which was probably because of the pro le and quantity of substances present in EtOAc. The analyses by 1 H NMR and GC/MS showed the presence of fatty acids and aromatic compounds in EtOAc and CH 2 Cl 2 ; however, the GC/MS analysis showed that, in addition to these compounds, ketamine was present in EtOAc. The other substances present in this fraction were the following fatty acids: hexadecenoic, palmitic, linolelaidic, oleic, linoleic, and vaccenic, The palmitic, oleic, and linoleic acids have the previously described nematicidal activity [19,20]. The other fatty acids are most likely the constituents of the fungal structures, such as the cell membrane. Moreover, it has been hexadecenoic acid has biological activity reported for fungi and bacterial [21]. In F4, we identi ed ketamine and palmitic and linoleic acids, as these fatty acids have nematicidal activity that has already been described in the literature [19,20], and this can explain the fact that subfraction 4 had the lowest ED 50 compared to subfraction 5 and commercial ketamine; thus, the substances could be acting in synergy. On the other hand, F5 indicated only the presence of ketamine. Even though the role of fatty acids in nematode fungus interaction is not totally understood, some of these compounds can be present in the formation of adhesive traps in predatory fungi, with the example of linoleic acid [22]. However, the role of such metabolites may go beyond the formation of adhesive traps, since P. chlamydosporia does not form such structures.
The GC/MS analysis of F5 suggested the presence of ketamine with a high degree of purity, since it presented a single peak with a retention time of 25.5 min (Fig. 5S). Due to the fact that nematicidal activity was observed in this fraction and that the subfractions possessing ketamine, and this molecule were commercially available, we con rm the anthelmintic activity of ketamine on C. elegans; thus, we demonstrated that besides the fatty acids with anthelmintic activity known there, ketamine is responsible for the nematode death/paralysis process. Second [23], fungal metabolites capable of paralyzing and/or killing nematodes are involved in the rst contact between the fungus and the nematode, so it can be the rst method of immobilization, which explains the production of ketamine by the fungus.
The anthelmintic activity of ketamine was con rmed in vivo using hamsters and BALB/c mice. Hamsters are permissive hosts for A. ceylanicum infection, and the pathogeny caused by these worms in hamsters is similarly caused in humans [24]. The BALB/c mice have been a good alternative model for studying early Ascaris spp., and initial events of infection and immunology are similar for the hosts de A. suum and A. lumbricoides [25]. Ketamine was effective in reducing the burden parasites recovered from the hamsters' intestines. At a concentration of 6 mg, the worm burden was close to zero, showing results similar to albendazole. In the experiment on A. suum, we observed reduction in the number of larvae that reached the mice lungs, and the drug activity was not dependent on the route of administration, since there was no difference in results when the drug was administered subcutaneously or orally. This information is important because the oral route of administration is cheaper, easy to administer, and painless. Possibly, the activity of this drug is related to its ability to bind to GABA receptors and may consequently lead the nematode to paralysis [26,27]. There is no report in the literature on the anthelmintic activity of ketamine. It is worth noting that this drug is already used as a dissociative anaesthetic. Therefore, for its use as an anthelmintic, more studies are needed to avoid or minimize adverse reactions. Other studies are necessary to verify a better form of administration and/or association with another drug or vehicle once this drug is rapidly metabolized, having a half-life of approximately 15 minutes [28].

Conclusion
The unprecedented results presented here con rm the anthelmintic activity of ketamine in the tested models, thus demonstrating the possibility of repositioning ketamine and corroborating the potential of P. chlamydosporia in the discovery of new drug candidates that can be used in the treatment of several neglected diseases caused by nematode infections.
RTF and SRF were responsible for the conception of the study. LLB and DCB assisted in the obtaining nancial resources. LMC and JVA provided and cultivating the fungus. RMA, TAOM, FSB, LFF and EMLR participated the experimental biological assays. ATM, JHSG, LPSP, VNM and RMP participated the chemistry extraction and analyses experiments. All co-authors contributed to the interpretation of the results and revision of the manuscript, and approved the nal version of the manuscript for publication Figure 2 Activity of ketamine on the reduction of A. ceylanicum worm burden in hamsters (Mesocricetus auratus). Figure A shows the recovered worm burden from hamster intestines. Bars followed by the same letter signify that there was no statistical difference, according to the Tukey's test (P 0.05). Results were plotted as mean, ± mean standard deviation. Figure B shows the mean EPG values of the animals in their treatment groups, before (18th day) and after (24th day) the treatment had begun.