Antiprotozoal activity of different Xenorhabdus and Photorhabdus bacterial secondary metabolites and identification of bioactive compounds using the easyPACId approach.

Natural products have been proven to be important starting points for the development of new drugs. Bacteria in the genera Photorhabdus and Xenorhabdus produce antimicrobial compounds as secondary metabolites to compete with other organisms. Our study is the first comprehensive study screening the anti-protozoal activity of supernatants containing secondary metabolites produced by 5 Photorhabdus and 22 Xenorhabdus species against human parasitic protozoa, Acanthamoeba castellanii, Entamoeba histolytica, Trichomonas vaginalis, Leishmania tropica and Trypanosoma cruzi, and the identification of novel bioactive antiprotozoal compounds using the easyPACId approach (easy Promoter Activated Compound Identification) method. Though not in all species, both bacterial genera produce antiprotozoal compounds effective on human pathogenic protozoa. The promoter exchange mutants revealed that antiprotozoal bioactive compounds produced by Xenorhabdus bacteria were fabclavines, xenocoumacins, xenorhabdins and PAX peptides. Among the bacteria assessed, only P. namnaoensis appears to have acquired amoebicidal property which is effective on E. histolytica trophozoites. These discovered antiprotozoal compounds might serve as starting points for the development of alternative and novel pharmaceutical agents against human parasitic protozoa in the future.

Introduction

Infectious diseases are caused by the invasion and continued presence of pathogenic microorganisms such as viruses, bacteria, fungi, protozoa, nematodes, etc. in a host`s body organ, tissue, or cells. Protozoa in particular, such as Acanthamoeba castellanii (A. castellanii), Entamoeba histolytica (E. histolytica), Trichomonas vaginalis (T. vaginalis), Leishmania tropica (L. tropica) and Trypanosoma spp., are eukaryotic single-celled organisms that are the leading cause of numerous untold deaths and devastating chronic diseases worldwide, especially in underdeveloped and developing countries of sub-Saharan Africa, Asia and South America[1-3]. They are transmitted directly or indirectly through contact, air contaminated food or water, or by vectors from infected humans and animals to healthy others[1]. Poverty, inadequate sanitation and unhygienic living conditions, malnutrition, suitable climatic factors, ineffective anti-parasitic drugs, inept vector control interventions, insecticide resistance are some of the factors that contribute to the persistence and incidence of such parasitic infectious diseases in various parts of the world[4-6].

Since protozoan parasites are eukaryotic organisms that share functional homology with mammalian cells, currently available drugs for the treatments of parasitic diseases are generally toxic to human cells and have adverse side effects[7,8]. Owing to these undesired effects and considering the development of resistant strains of parasites against pharmaceutical products, new drugs with different modes of action on target parasites and minimal toxicity to host cells are urgently required[9,10].

Natural products (or secondary metabolites) have been proven to be an important starting point for the development of new drugs. Screening natural products provide the chance of discovering new molecules with unique structure, high activity, and selectivity[11]. The most important natural product sources in nature are fungi[12,13], plants[14] and bacteria[11,15,16]. Various fungi and bacteria produce antimicrobial compounds as secondary metabolites to compete with other organisms. One of the sources of novel bioactive therapeutics against parasites are insect pathogenic Photorhabdus and Xenorhabdus bacteria. These bacteria encode several putative biosynthetic pathways for natural product biosynthesis[17-19] of which several of them are conserved since they fulfill important ecological functions in their ecological niche[20]. Photorhabdus and Xenorhabdus bacteria are associated with entomopathogenic nematodes which are obligate and lethal insect parasitic organisms[21,22]. When these nematodes penetrate an insect host, they release their mutualistic bacteria into the insect hemolymph and within 48 h the insect host is killed because of bacterial toxins and enzymes[23,24]. Furthermore, to protect the nematode-infected cadaver from opportunistic microorganisms (e.g., bacteria, fungi, protozoa, and viruses) both Xenorhabdus and Photorhabdus bacteria produce a variety of natural products that have antimicrobial activities[19,25,26]. Although several studies have reported the antibacterial[27-30], antifungal[29-35], and insecticidal[36-38] activities, only very few studies have investigated the antiprotozoal effect of the secondary metabolites produced by these bacteria[39,40]. Currently, more than 40 different species of Photorhabdus and Xenorhabdus bacteria have been identified[23,41] that produce different sets of natural products[42]. The aim of our study was to investigate natural products produced by five Photorhabdus and 22 Xenorhabdus species against human parasitic protozoa, A. castellanii, E. histolytica, T. vaginalis, L. tropica, and Trypanosoma cruzi (T. cruzi), and the identification of novel bioactive antiprotozoal compounds by using the easyPACId (easy Promoter Activated Compound Identification) approach[43].

Material and methods

Bacterial sources and preparation of cell-free supernatants

The cell-free supernatants of 22 Xenorhabdus and 5 Photorhabdus species were tested against human parasitic protozoa (Table 1). All bacteria strains were obtained from the Bode lab and were kept at − 80 °C as stock culture until use.

Table 1

Bacterial species used in antiprotozoal activity tests.

	Bacteria species
1	Xenorhabdus nematophila ATCC 19061
2	X. bovienii SS-2004
3	X. vietnamensis DSM 22392
4	X. cabanillasii JM26-1
5	X. szentirmaii DSM 16338
6	X. stockiae DSM 17904
7	X. ehlersii DSM 16337
8	X. koppenhoferi DSM 18168
9	X. indica DSM 17382
10	X. maulenoii DSM 17908
11	X. poinarii G6
12	X. griffiniae DSM 17911
13	X. ishibashi DSM 22670
14	X. doucetiae DSM 17909
15	X. innexi DSM 16336
16	X. japonica DSM 16522
17	X. khoisanae
18	X. beddingii DSM 4764
19	X. budapestensis DSM 16342
20	X. miraniensis DSM 17902
21	X. hominickii DSM 179903
22	X. kozodoii DSM 17907
23	Photorhabdus kayaii DSM 15194
24	P. namnaoensis PB 45.5
25	P. laumondii TTO1
26	P. akhurstii DSM 15138
27	P. thracensis DSM 15199

A loopfull of bacteria taken from stock culture was inoculated to Luria Bertani (LB) (Merck) agar medium and incubated at 30 °C for 24 h. A single colony was picked and inoculated to 10 ml sterilized Tryptic Soy Broth (TSB) medium (Merck) and cultivated at 30 °C for 24 h to be used as overnight culture. Subsequently, 1 ml from overnight culture was transferred to 50 ml sterilized TSB medium and incubated at 30 °C and 150 rpm for 120 h (it is known that these bacteria produce the most secondary metabolite after 120 h)[30,44]. To obtain cell-free supernatant, the bacterial broth was centrifuged at 10,000 rpm for 10 min at 4 °C. The supernatant was collected carefully and filtered through a 0.22 μm Millipore filter (ISOLAB)[45]. An aliquot of the filtrated suspension was streaked onto NBTA agar to verify the absence of bacterial cells[46]. The supernatants were poured into the 50 ml sterile centrifuge tubes (Corning, NY) and kept at − 20 °C for up to 2 weeks prior to use[47,48].

In vitro cultures of parasitic protozoons

Axenic cultures of A. castellanii trophozoites (ATCC 30010) were maintained in liquid PYG (protease peptone—yeast extract—glucose) medium supplemented with penicillin G (500 U/ml) and streptomycin (50 μg/ml)[49] (Pérez-Serrano et al. 2000). The cultures were refreshed weekly in 25 ml cell culture flasks (Sigma) and incubated at 30 °C, until use[50,51]. Cells from the culture medium were harvested by centrifugation at 2000 rpm for five minutes and washed three times with Phosphate-Buffered Saline (PBS). Acanthamoeba castellanii trophozoites adhering to flasks were collected by placing the flasks on ice for 30 min with gentle agitation[52,53].

Entamoeba histolytica (ATCC 30459) strain was kindly provided by Dr. Charles Graham Clark from the London School of Hygiene and Tropical Medicine. Entamoeba histolytica trophozoites were cultured axenically in LYI medium (880.0 ml LYI Broth, 20.0 ml Vitamin Mixture, 100.0 ml Heat Inactivated Adult Bovine Serum) supplemented with penicillin G (500 U/ml) and streptomycin (50 μg/ml)[54]. The cultures were routinely maintained by subculturing into screw capped test tube containing 7 mL of LYI medium[55,56].

Trichomonas vaginalis (ATCC 30001) trophozoites were grown in Diamond’s trypticase yeast-extract maltose (TYM) medium (0.5 mg of L-cysteine HCl, 0.1 g of ascorbic acid, 0.4 g of K2HPO4, 0.4 g of KH2PO4, 10 g of trypticase, 2.5 g of maltose and 10 g of yeast extract in one ml of distilled water, pH:6) supplemented with 100 IU/ml streptomycin, 100 IU/ml penicillin and 10% heat-inactivated Fetal Bovine Serum (FBS). T. vaginalis subcultures were cultured regulaly to maintain viability and for use in the assays[57].

Leishmania tropica (ATCC 50129) promastigotes were routinely cultured at 27 °C in RPMI-1640 medium (Sigma) supplemented with 10% heat-inactivated FBS (Cegrogen, Stadtallendorf-Germany). The culture was sustained in 25 ml flasks and stationary phase of promastigotes were obtained[58].

Trypanosoma cruzi (CBU-TC01) trypomastigotes were obtained from the parasite biobank of Manisa Celal Bayar University School of Medicine Department of Parasitology Manisa, Turkey. The trypomastigotes were incubated at 27 °C in RPMI 1640 medium supplemented with 10% heat-inactivated FBS, 200 U penicillin/ml, and 0.2 mg streptomycin/ml. Subcultures were maintained in 25 ml flasks until use in the experiments[59,60].

In vitro antiprotozoal activity of bacterial secondary metabolites

Except for E. histolytica, the microdilution method was used to assess the antiprotozoal activity of the bacterial supernatants against A. castellanii, T. vaginalis, L. tropica and T. cruzi. The four parasites were seeded in 96-well Microtiter plates (Greiner, Germany) and the supernatants were applied at serial concentrations ranging from 10% to 1.25%. Briefly, Trophozoites of A. castellanii and T. vaginalis were adjusted to 5 × 104 and 2 × 106 cells/mL, respectively. The density of L. tropica promastigotes and trypomastigotes of T. cruzi were adjusted to 1 × 106 cells/mL[61,62]. Plates with the isolates were incubated at 30 °C for 24 h, 37 °C for 48 h and 27 °C for 72 h for A. castellanii, T. vaginalis, L. tropica, and T. cruzi respectively. Screw capped test tubes were used for E. histolytica instead of the plates used for other parasites. Entamoeba histolytica trophozoites (200 µl of 3 × 105 cells/mL) were inoculated into the tubes containing 1.8 ml of fresh axenic LYI medium with the bacterial supernatants at final concentrations of 10%, 5%, 2.5%, and 1.25%. The tubes were incubated at 37 °C for 48 h[56].

Two methods were used to determine the antiprotozoal effects of the bacterial supernatants in vitro. To assess the anti-leishmanial activity was performed by using the XTT (sodium 3,39- [1- (phenylaminocarbonyl)-3,4-tetrazolium]-bis (4-methoxy-6-nitro) benzene sulfonic acid hydrate) cell proliferation kit (Roche Molecular Biochemicals, Mannheim, Germany) as previously described[63].

Cell viability assay test was used for A. castellanii, T. vaginalis, E. histolytica and T. cruzi. The assay was evaluated by adding 0.1% trypan blue stain (TB) [the number of live (unstained) and dead (stained)] using a hemocytometer[64,65]. The parasite mortality (in %) for each bacterial supernatants sample was caluclated according to the formula: % Mortality of parasites = (Control Negative-Test sample) × 100%/Control negative. Only 100% inhibition of the parasite was considered when no motile parasite was observed.

Two negative and one positive control group were included in each experiment. Bacterial culture medium (TSB) and parasite medium was used as a negative control. Metronidazole (Specia Rhone Poulenc Rorer, Paris, France) for T. vaginalis and E. histolytica, Chlorhexidine (Sigma, Spain) for A. castellanii, N-methyl meglumine (Glucantime™, Rhone Poulenc, France) for L. tropica and Benzimidazole (Sigma, Spain) for T. cruzi were used as positive controls. Each assay was performed at least three times in triplicate.

Identification of antiprotozoal compounds using the easyPACId method

Generating promoter exchange mutant strains

The easyPACId approach method recently developed by Bode et al.[43] was used to identify the antiprotozoal compounds in Xenorhabdus spp. bacteria. Briefly, ∆hfq mutants of each bacterial species (X. budapestensis, X. cabanillasii, X. doucetiae, X. hominickii, X. nematophila, X. stockiae and X. szentirmaii) were first generated and then the native promoter regions of selected natural product biosynthetic gene clusters of these bacteria were exchanged with the chemically inducible promoter PBAD (addition of L-arabinose) via integration of the pCEP-KM plasmid[43,66]. This allows the selective production of a desired single natural product compound class and enables direct bioactivity analysis of the corresponding supernatant instead of time-consuming isolation of single compound(s) from analytically complex wild type extracts. The generation of the described Xenorhabdus spp. ∆hfq as well as Xenorhabdus spp. ∆hfq pCEP-KM-xy mutants listed in Table 3 (xy describes the locus of the first biosynthetic gene cluster) is described in detail by Bode et al.[43].

Table 3

Activity of promoter exchange mutants and their respective natural products against the Acanthamoeba castellanii, Trichomonas vaginalis, Leishmania tropica and Trypanasoma cruzi.

Activity of all easyPACId strains (delta hfq mutants) was determined after induction of the PBAD Promoter with L-arabinose. Bioactivities are shown for none (white) to highest activity (dark red) in the different assays. In order to confirm the active compounds, these are also shown non-induced (−) and induced (+).

Obtaining cell-free supernatants from different Xenorhabdus spp. ∆hfq promoter exchange mutants

A single Xenorhabdus spp. ∆hfq pCEP-KM-xy mutant colony, cultivated on LB agar supplemented with a 50 μg/mL final concentration of kanamycin at 30 °C for 48 h, was transferred into LB medium (10 mL) also supplemented with a 50 μg/mL final concentration of kanamycin and incubated at 150 rpm and 30 °C. Then, this overnight culture was inoculated into a fresh 20 mL LB with the final optical density (OD600) adjusted to 0.1. After an hour incubation at 30 °C, these cultures were induced with 0.2% L-arabinose and incubated again for 72 h at 150 rpm and 30 °C[43,67]. Cultures of non-induced mutants contained no L-arabinose. The cell-free supernatants were obtained by centrifugation at 10.000 rpm for 20 min in 50 ml Falcon tubes at 4 °C and filteration through a 0.22-µm Millipore filter (Thermo scientific, NY) to ensure total removal of bacterial cells[34]. The cell free supernatants were stored at − 20 °C and used within 2 weeks[48].

Testing the antiprotozoal activity of cell-free supernatants of mutant strains

Antiprotozoal activity of 5-day-old cell-free supernatants of wild type strains, as well as induced (with arabinose) and non-induced (without arabinose) promoter exchange mutant strains were tested in microdilution bioassay as previously described in the in vitro antiprotozoal activity tests section.

Anti-protozoal activity of bioactive extracts obtained from hfq mutants

As a last step, extracts containing identified anti-protozoal compounds were tested again on the parasite species at different concentrations ranged from 10 to 0.078% (v/v). The same experimental method used in antiprotozoal activity tests was carried out here.

Anti-protozoal bioactive compound extraction was performed by culturing induced X. nematophila Δhfq_pCEP_ kan_XNC1_1711 for xenocoumacin production and X. doucetiae Δhfq_PBAD_ PAX_km for PAX peptide production in LB (6L) with 2% XAD® resin at 30 °C for 3 days. Afterwards the resin was exhaustively extracted with methanol (3 × 2 L) at 24 ± 1 °C and concentrated under reduced pressure to give a crude extract enriched by the desired natural compound class. The extracts were then dissolved in DMSO and prepared as a stock solution with distilled water. Fabclavine was obtained by concentrating the supernatant of the induced X. cabanillasii Δhfq_128-129 culture 10-fold using an evaporator.

Statistical analysis

Differences in antiprotozoal activity of the supernatants were compared with one-way ANOVA and the means separated using Tukey’s test. P values < 0.05 were considered as significant[68]. The results are reported as mean ± SD for all values.

Ethics approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Results

In vitro antiprotozoal activity tests

Acanthamoeba castellanii

The in vitro activity assays against the trophozoite of A. castellanii showed that 12 of 22 Xenorhabdus species exhibited effective antiprotozoal activity whereas, none Photorhabdus strain showed any activity (Table 2). At 10% supernatant concentration, A. castellanii cell mortality ranged between 78 and 100% depending on Xenorhabdus species. Relative to the negative control, 12 of 22 tested Xenorhabdus supernatants caused significant A. castellanii mortality (F = 1828.80; df = 28, 232; P < 0.0001). Chlorhexidine used as positive control showed 100% mortality and no statistical difference was observed between chlorhexidine and X. budapestensis, X. cabanillasii, X. doucetiae and X. innexii supernatants. At 5% concentration of bacterial supernatants, the highest level of mortality (> 95%) was exhibited by X. budapestensis, X. cabanillasii, X. innexii and chlorhexidine (which were not statistically different from each other). All Xenorhabdus species with antiprotozoal activity presented statistically significant mortality compared to the negative controls (F = 1357.38; df = 28, 232; P < 0.0001). In the following concentration (2.5%), the supernatants of X. budapestensis, X. cabanillasii and X. innexii exhibited more than 90% mortality on A. castellanii trophozoites and no significant difference was observed between this group and chlorhexidine (Table 2). Xenorhabdus miraniensis and X. nematophila supernatants caused the lowest mortality (55%) on the trophozoites. Despite this, there was a significant difference between all effective 12 Xenorhabdus supernatant treatments and negative controls (F = 653.63; df = 28, 232; P < 0.0001). At the lowest concentration of tested bacterial supernatants (1.25%), X. budapestensis and X. innexii species showed equal mortality with chlorhexidine. Following these species, supernatants of X. cabanillasii, X. doucetiae, X. hominickii, X. stockiae, and X. szentirmaii were more effective compared with the other treatments. Even at highly diluted concentrations of the supernatants compared to negative controls, significant mortalities were obtained (F = 550.64; df = 28, 232; P < 0.0001) (Table 2).

Table 2

Antiprotozoal activity of cell-free culture supernatants of Xenorhabdus and Photorhabdus spp. against different human parasitic protozoa species.

Bioactivities are shown for none (white) to highest activity (dark red) in the different assays. Negative control-1: Bacterial culture medium (TSB), Negative control-2: Parasite growth medium. Positive controls: Metronidazole for T. vaginalis and E. histolytica, Chlorhexidine for A. castellanii, N-methyl meglumine for L. tropica and Benzimidazole for T. cruzi. Columns with a common superscript letter do not differ significantly at P = 0.05).

Trichomonas vaginalis

It was noted that 10 of the 22 tested Xenorhabdus supernatants were significantly lethal against T. vaginalis when compared with negative controls. Of these, X. cabanillasii, X. doucetiae, X. hominickii, X. kozodoii, X. mauleonii, X. miraniensis and X. vietnamensis species displayed more than 90% mortality against T. vaginalis at 10% supernatant concentration (F = 334.60; df = 29, 240; P < 0.0001) (Table 2). At 5% concentration, though the positive control metronidazole caused statistically significant mortality compared to the supernatant treatments, all the other effective bacterial supernatants exhibited mortality ranging between 57 and 93% (F = 288.74; df = 29, 240; P < 0.0001). Among the effective bacterial supernatants, X. miraniensis, X. mauleonii, X. kozodoii and caused 92, 85 and 83% T. vaginalis mortality, respectively, at 2.5% concentration (F = 288.74; df = 29, 240; P < 0.0001). There was a significant difference between the effective treatments and controls (F = 237.78; df = 29, 240; P < 0.0001). Xenorhabdus miraniensis and X. vietnamensis still exhibited more than 80% mortality, when the supernatants were diluted to the lowest concentration (1.25%) (F = 255.82; df = 29, 240; P < 0.0001). However, none of the Photorhabdus species caused T. vaginalis mortality at any concentration (Table 2).

Leishmania tropica

Xenorhabdus budapestensis, X. cabanillasii, X. hominickii, X. indica, X. innexii and X. stockiae supernatants caused 100% mortality at the highest tested concentration (10%) against the promastigote form of L. tropica. No differences occurred between this treatment group and positive control (P > 0.05). Following those bacteria species X. vietnamensis, X. miraniensis, and X. szentirmaii supernatants exhibited 97, 96 and 90% mortality, respectively. The other effective bacterial supernatants presented mortalities that ranged between 53 and 82%. There was a significant difference between all effective treatment groups and negative control (F = 880,33; df = 29, 240; P < 0.0001) (Table 2).

Similarly, at 5% concentration, X. budapestensis, X. cabanillasii, X. hominickii, X. indica, X. innexii, X. miraniensis, X. stockiae and Stibogluconate caused 98–100% Leishmania mortalities. No significant difference was observed among these groups. Although X. doucetiae presented the least Leishmania mortality (40%), there were statistically significant differences between all effective treatments and negative controls (F = 232.16; df = 29, 240; P < 0.0001) (Table 2). At 2.5% supernatant concentration, X. budapestensis, X. cabanillasii, X. hominickii, X. indica, X. innexii, X. miraniensis, X. stockiae and X. vietnamensis showed the highest efficacy (93–100%), whereas X. griffinae and X. szentirmaii supernatants exhibited the lowest mortalities (33 and 31%, respectively). However, there was a significant difference between all effective treatments and negative control groups (F = 425.10; df = 29, 240; P < 0.0001) (Table 2).

Bacterial supernatant of X. budapestensis, X. cabanillasii, X. hominickii, X. indica, X. innexii, X. miraniensis and X. stockiae were still as effective as N-methyl meglumine even at 1.25% concentration. There was a significant difference between the effective bacterial supernatants and negative controls (F = 393.67; df = 29, 240; P < 0.0001) (Table 2).

Trypanosoma cruzi

At 10% concentration, 10 of 27 bacterial supernatants showed antiprotozoal activity against T. cruzi trypomastigotes. Among the bacterial species, X. cabanillasii and X. miraniensis exhibited 100% mortality followed by X. nematophila, X. indica, X. hominickii, X. budapestensis and X. innexi supernatants (88–97%). There was a statistically significant difference between effective bacterial supernatants and negative controls (F = 288.53; df = 29, 240; P < 0.0001) (Table 2).

When the bacterial supernatants were diluted to 5% concentration, X. cabanillasii and X. miraniensis still exhibited 100% mortality. Xenorhabdus indica, X. nematophila, X. budapestensis and X. innexi supernatans displayed between 94 and 81% mortality. The effect of X. hominickii dropped drastically from 91 to 43%. Significant differences were observed between negative controls and the effective supernatants (F = 178.60; df = 29, 240; P < 0.0001). At 2.5% concentration of the supernatants, X. miraniensis and X. cabanillasii had the highest mortality (98–100%) and there was no significant difference between this bacterial species and Benzimidazole (positive control). Among the other bacterial supernatans, X. budapestensis, X. indica, X. innexi and X. nematophila also caused relatively high mortality which ranged between 80 and 92%. However, X. doucetiae, X. hominicki, X. stockiae, and X. vietnamiensis lost their antiprotozoal effects at this concentration. A significant statistical difference was observed between the six effective bacterial supernatants and negative controls (F = 589.60; df = 29, 240; P < 0.0001) (Table 2).

At the lowest tested concentration (1.25%), X. cabanillasii and X. miraniensis maintained their high activity (97 and 98% mortality, respectively). Xenorhabdus indica followed this group with a mortality of 90%. There was a statistical difference between negative control groups and six bacterial supernatants (F = 929.48; df = 29, 240; P < 0.0001) (Table 2).

Entamoeba histolytica

Unlike A. castellanii, E. histolytica, a different amebic parasite, was resistant to the secondary metabolites of Xenorhabdus and Photorhabdus bacteria. Only P. namnaoensis species showed significant mortality compared to the negative controls (F = 1.02; df = 28,232; P < 0.0001) (Table 2). Positive control (metronidazole) and P. namnoensis supernatant caused 100% and 53% cell mortality on E. histolytica trophozoites, respectively. However, the other 26 of 27 tested species exhibited only between 0 and 6% mortality at 10% concentration.

Identification of antiprotozoal compounds

The promoter exchange mutants in Δhfq background revealed that antiprotozoal bioactive compounds produced by Xenorhabdus bacteria were fabclavines, xenocoumacins, xenorhabdins and PAX peptides (Table 3, Fig. 1). The supernatants obtained from induced mutants showed very high mortality against the parasite cells, non-induced mutants of the same compounds exhibited no activity (Table 3).

Figure 1

Chemical structure of antiprotozoal natural products from Xenorhabdus species.

Fabclavines produced by X. cabanillasii, X. hominickii and X. stockiae species had antiprotozal activity against A. castellanaii, T. vaginalis L. tropica and T. cruzi parasites. Fabclavines produced by X. budapestensis was not effective against T. vaginalis. Xenorhabdus szentirmaii also produces fabclavines being only effective against A. castellanii and L. tropica with no antiprotozoal activity against T. vaginalis and T. cruzi.

Xenocoumacins produced by X. nematophila species was the bioactive antiprotozoal compound against all tested pathogens. In contrast to other species, X. doucetiae species produce more than one antiprotozoal compound. Δhfq_PBAD_PAX_km of X. doucetiae producing PAX peptides exhibited antiprotozoal effect on A. castellanii and T. vaginalis, but L. tropica was killed by xenocoumacins and xenorhabdins. Xenorhabdus doucetiae Δhfq_PBAD_xcnA_km showed antiprotozoal activity only with xenocoumacins against T. cruzi (Table 3). The active compound in P. namnaoensis which was the only species that caused mortality on E. histolytica trophozoites was not identified due to the lack of promoter exchange mutants of this species.

Supernatants containing xenocoumacins, fabclavines and PAX peptides showed variable activity depending on parasite species and concentrations; no mortality was observed in the control (Fig. 2). Overall fabclavine molecules were highly effective on all tested parasite species even at very low concentrations.

Figure 2

Antiprotozoal activity of supernatants containing bioactive compounds. Xenocoumacins, fabclavines and PAX peptides were obtained from Xenorhabdus nematophila, Xenorhabdus cabanillasii and Xenorhabdus doucetiae, respectively.

Discussion

Our data revealed that Xenorhabdus and Photorhabdus produce antiprotozoal compounds effective on human pathogenic protozoa. However, not all Xenorhabdus or Photorhabdus species showed this activity. Except for E. histolytica, only some of Xenorhabdus species exhibited antiprotozoal activity. It was reported that Xenorhabdus bacteria produce broad-spectrum compounds with various activity against several organisms such as bacteria, fungi, insects, nematodes, mites, protozoa etc. to protect and bioconvert the host cadaver[69-72]. With the easyPACId approach we were able to assign the described activities on respective natural products from Xenorhabdus.

The bioactivity of fabclavines could be confirmed for X. budapestensis, X. cabanillasii, X. hominickii, X. stockiae and X. szentirmaii mutants. Biochemically fabclavines are peptide/polyketide hybrids connected to a polyamine moiety generated by a fatty acid/polyketide synthase with similarity to enzymes producing polyunsaturated fatty acids (PUFAs)[67,73,74]. Fabclavines 1a and 1b exhibit various bioactivities against different bacterial, fungal and protozoal organisms[73] and due to such broad-sprectrum activity, fabclavines might serve as protective agents against saprophytic food competitors/microorganisms that attack insect cadavers; this enables Xenorhabdus/Steinernema to maintain a monoculture in the infected insect[73]. Fabclavines are structurally very similar to zeamines identified in Serratia plymuthica so might similarly permeabilize artificial bacterial and eucaryotic model membranes[75,76]. A structurally yet-undentified fabclavine derivative from X. innexi (Xlt) induces membrane degradation at low concentrations in selected mosquito cell lines which led to apoptosis[77]. Production of fabclavine is widespread in Xenorhabdus strains whereas, except for Photorhabdus asymbiotica, other Photorhabdus species do not produce fabclavines[42]. This can explain partially why none of our tested Photorhabdus species showed antiprotozoal activity. However, X. bovienii is a producer of only the polyamine part of fabclavine[74] and it did not exhibit any activity. There are 32 different types of fabclavine with important variations among their activity[67].

Xenorhabdus nematophila and X. doucetiae species do not produce fabclavine[42] but they are effective species on tested parasites except for E. histolytica. According to promoter exchange data, it became obvious that X. nematophila and X. doucetiae perform this task with different compounds. Xenocoumacins are produced using a hybrid nonribosomal peptide synthetase (NRPS) multienzyme (XcnA-N) and polyketide synthase (PKS)[78-80]. When tested for biological activity against T. b. rhodesiense, T. cruzi, L. donovani and Plasmodium falciparum, good activities were observed against T. b. rhodesiense and P. falciparum[80]. Possibly, xenocoumacins inhibit protein biosynthesis in these organisms[80,81]. However, xenocoumacins are not widely distributed in Xenorhabdus spp. as one would expect. Among 25 Xenorhabdus and Photorhabdus strains, xenocoumacins or the corresponding biosynthetic gene cluster were only be identified from seven Xenorhabdus subspecies (X. nematophila, X. indica, X. miraniensis, X. stockiae, X. kozodoii, X. mauleonii and X. doucetiae)[42]

According to our data, we have determined that xenorhabdins and PAX peptides produced by X. doucetiae are other effective secondary metabolites. Xenorhabdins are dithiolopyrrolone compounds[82] and it is reported that they have antibacterial, antifungal, and insecticidal activities[83-85]. Their suggested mode of action is the inhibition of RNA synthesis affecting translation as similar to xenocoumacins[86,87]. PAX peptides are lysine-rich cyclolipopeptides. Gualtieri et al.[88] first described five PAX peptides from X. nematophila and then additional eight PAX peptides were identified, and their structures elucidated by Fuchs et al.[73]. Three NRPS genes (paxABC) are responsible for the biosynthesis of the PAX compounds. These peptides have antifungal and antibacterial activity. They exhibited strong anti-fungal activity against the opportunistic human pathogen Fusarium oxysporum as well as several plant pathogenic fungi[88].

Interestingly, among the tested 27 Xenorhabdus and Photorhabdus strains only P. namnoensis appears to have acquired amoebicidal property which is effective on E. histolytica trophozoites. The bioactive compound responsible for this activity and its mode of action needs to be identified in the future.

The determined bioactive compounds may offer new opportunities for treating important parasitic diseases or be useful as lead compounds in the development of new antiprotozoal agents. For this purpose, new bioactive compounds should have no or very low cytotoxicity on human cells. Bode et al.[43] tested the efficacy of bioactive compounds isolated from Xenorhabdus and Photorhabdus bacteria on the human microvascular endothelial cell (EC) line (CDC.EU.HMEC-1). Fabclavine, PAX peptide, xenocoumacin and xenorhabdin had no or low impact on the metabolic activity, apoptosis and cell cycle G2-block. However, xenocoumacin and xenorhabdin exhibited toxic effects on cell proliferation.

In conclusion, this is the first extensive study screening the anti-protozoal activity of Xenorhabdus and Photorhabdus secondary metabolites against important human parasites A. castellanii, E. histolytica, T. vaginalis, L. tropica and T. cruzi and using the easyPACId technique to identify new potential antiprotozoal compounds. Future studies should investigate in detail the mode of action of these promising antiprotozoal compounds. Also, after a close structural investigation of these NPs, novel and safer pharmaceutical drugs can be potentially designed and synthesized.