In Vitro and In Vivo Assessment of the Efficacy of Bromoageliferin, an Alkaloid Isolated from the Sponge Agelas dilatata, against Pseudomonas aeruginosa

The pyrrole-imidazoles, a group of alkaloids commonly found in marine sponges belonging to the genus Agelas, display a wide range of biological activities. Herein, we report the first chemical study of the secondary metabolites of the sponge A. dilatata from the coastal area of the Yucatan Peninsula (Mexico). In this study, we isolated eight known alkaloids from an organic extract of the sponge. We used NMR and MS analysis and comparison with existing databases to characterize the alkaloids: ageliferin (1), bromoageliferin (2), dibromoageliferin (3), sceptrin (4), nakamuric acid (5), 4-bromo-1H-pyrrole-2-carboxylic acid (6), 4,5-dibromopyrrole-2-carboxylic acid (7) and 3,7-dimethylisoguanine (8). We also evaluated, for the first time, the activity of these alkaloids against the most problematic multidrug-resistant (MDR) pathogens, i.e., the Gram-negative bacteria Pseudomonas aeruginosa, Klebsiella pneumoniae and Acinetobacter baumannii. Bromoageliferin (2) displayed significant activity against P. aeruginosa. Comparison of the antibacterial activity of ageliferins 1–3 (of similar structure) against P. aeruginosa revealed some relationship between structure and activity. Furthermore, in in vitro assays, 2 inhibited growth and biofilm production in clinical strains of P. aeruginosa. Moreover, 2 increased the survival time in an in vivo Galleria mellonella model of infection. The findings confirm bromoageliferin (2) as a potential lead for designing new antibacterial drugs.


Introduction
Multidrug-resistant bacterial infections represent a serious global health problem [1], causing an estimated 700,000 deaths a year worldwide. If this rising trend in antibiotic resistance is not reversed in the coming years, it could lead to 10 million people dying every year and the economic impact of approximately 1% reduction of the world's gross domestic product (GDP) and there would be a 5-7% methanolic fraction was subjected to reversed-phase HPLC, yielding compounds 1-3 and 5-7 ( Figure  1).

Antimicrobial Susceptibility Testing
Minimum inhibitory concentration (MIC) assays were performed with all of the alkaloids (1-8) against reference bacterial strains and clinical isolates (see Table 1). Four categories of antibacterial activity were established: high activity (MIC ≤ 8 mg/L); moderate activity (MIC = 16-32 mg/L), Mar. Drugs 2020, 18, 326 4 of 15 low activity (MIC = 64 mg/L) and no activity (MIC ≥ 128 mg/L). Compounds 2 and 3 displayed the highest activity against the Gram-negative bacterium P. aeruginosa. Higher MICs were obtained for the remaining purified compounds extracted from A. dilatata in tests with the pathogens. Bromoageliferin (2) proved to be the most active against P. aeruginosa of the alkaloids isolated from A. dilatata. The MIC values for 2 and P. aeruginosa strains indicated high activity against ATCC 27853 (8 mg/L) and moderate activity against PAO1 (32 mg/L). To further study the activity of 2 against P. aeruginosa, 4 clinical isolates were included in the study. Bromoageliferin (2) displayed moderate activity against P. aeruginosa clinical isolates 29-200 SV, 30-127 VI, 30-223 SV and 30-230 SV, with a MIC value of 32 mg/L, while the corresponding MIC value for imipenem was 2 mg/L. Low concentrations of bromoageliferin (2) did not inhibit the growth of the tested strains of A. baumannii or K. pneumoniae. Additionally, 2 has also been reported to display antibacterial against Gram-positive bacteria, including Micrococcus luteus and Bacillus subtilis [22,28] and the human pathogen methicillin-resistant Staphylococcus aureus (MRSA) [29], and also against the Gram-negative bacteria Escherichia coli [22,28] and Rhodospirillum salexigens (a marine bacterium known to form biofilms) [30]. This work represents the first assessment of the antibacterial activity of bromoageliferin (2) against the multidrug resistant pathogen P. aeruginosa.
Also noteworthy is the activity of dibromoageliferin (3) against ATCC 27853 and PAO1 reference strains of P. aeruginosa (MIC, 32 mg/L) and against the two strains of K. pneumoniae included and the A. baumannii strain RYC 52763/97 (MIC, 64 mg/L). Ageliferin (1) showed a similar ability to inhibit growth against K. pneumoniae and P. aeruginosa (MIC, 64 mg/L). A. baumannii displayed greater resistance to these ageliferin-derived compounds.
Although 1 and 3 have previously been reported to display antibacterial activity against M. luteus [28], B. subtilis [22,24,28], S. aureus [24] and E. coli [22,24,28], this is the first report of an assessment of the antibacterial activity of 1 and 3 against the human pathogenic bacteria A. baumannii, K. pneumoniae and P. aeruginosa.
Isolation of the three ageliferins (1-3), which show slight structural differences, and assessment of their antibacterial activity allowed us to identify a relationship between the structure of the compounds and their activity against the P. aeruginosa ATCC 27853 strain. The structural comparison of the three ageliferins (1)(2)(3) in relation to the MIC values for this strain indicated that the presence of a second bromine atom at C-2 of pyrrol A ring increases the antibacterial activity. Thus, 1, which does not contain the atom, was less active than 2 and 3, in which the atom does occur (1 was eight times less active than 2 and two times less active than 3). However, the presence of a second bromine atom at C-2 of pyrrol B ring decreased the antibacterial activity because 2, which bears a hydrogen at C-2 of the Mar. Drugs 2020, 18, 326 5 of 15 pyrrol B ring was more active than 3, which bears a bromine atom at that position (2 was four times more active than 3) (Figure 2).
Taking into account the remarkable MIC values observed for bromoageliferin (2) against P. aeruginosa, which is particularly problematic in serious infections such as cystic fibrosis, we wished to gain further insight into the antibacterial activity of this compound by performing biofilm biomass inhibition analysis and a survival assay with Galleria mellonella.
Taking into account the remarkable MIC values observed for bromoageliferin (2) against P. aeruginosa, which is particularly problematic in serious infections such as cystic fibrosis, we wished to gain further insight into the antibacterial activity of this compound by performing biofilm biomass inhibition analysis and a survival assay with Galleria mellonella.

Analysis of Biofilm Biomass Inhibition
Bromoageliferin (2) has been reported to possess anti-biofilm activity against the marine R-proteobacterium R. salexigens [30]. Identification of a 2-aminoimidazole (2-AI) subunit in these bioactive brominated pyrrol alkaloids led to the suggestion that this structural motif, in tandem with the bicyclic core of bromoageliferin, may be the key pharmacophore that imparts biological activity [39]. For this reason, bromoageliferin (2) has been used as template for designing a library of simplified bromoageliferin scaffolds, such as trans-bromoageliferin analogue (TAGE, see Figure 1), which has proven to be very effective in inhibiting biofilm formation in strains of P. aeruginosa [39], A. baumannii, Bordetella bronchiseptica and S. aureus [40]. Furthermore, some of these simplified analogues suppress resistance of multiple antibiotic classes across a broad-spectrum of clinically important bacteria [40][41][42]. The parent natural product, bromoageliferin (2), was later reported to inhibit biofilm formation in two Mar. Drugs 2020, 18, 326 6 of 15 representative human pathogens, A. baumannii and S. aureus [29]. In the present study, we wished to evaluate, for the first time, the anti-biofilm activity of bromoageliferin (2) in P. aeruginosa strains.
The concentration-dependence of the effect of 2 on biofilm reduction formation was detected with both the PAO1 and ATCC 27853 reference strains ( Figure 3). A significant decrease in the ability of P. aeruginosa PAO1 strain to generate biofilm was observed in the presence of bromoageliferin (2) at concentrations of 8 mg/L (p = 0.0119) and 16 mg/L (p ≤ 0.0001), relative to the control without compound. Regarding strain ATCC 27853, significant differences were found after addition of respectively 4 mg/L of 2 (p = 0.0140) or 8 mg/L (p = 0.0135) to the culture, relative to the control.
Mar. Drugs 2020, 18, x 6 of 16 the present study, we wished to evaluate, for the first time, the anti-biofilm activity of bromoageliferin (2) in P. aeruginosa strains.
The concentration-dependence of the effect of 2 on biofilm reduction formation was detected with both the PAO1 and ATCC 27853 reference strains (Figure 3). A significant decrease in the ability of P. aeruginosa PAO1 strain to generate biofilm was observed in the presence of bromoageliferin (2) at concentrations of 8 mg/L (p = 0.0119) and 16 mg/L (p ≤ 0.0001), relative to the control without compound. Regarding strain ATCC 27853, significant differences were found after addition of respectively 4 mg/L of 2 (p = 0.0140) or 8 mg/L (p = 0.0135) to the culture, relative to the control. The data obtained regarding the inhibition of biofilm production by bromoageliferin (2) are consistent with those previously obtained with two simplified synthetic analogues of 2 against P. aeruginosa [39]. Indeed, a concentration of 100-200 µM of these analogues was required to inhibit 50% of biofilm production, while for 2 we observed 30-40% biofilm inhibition at concentrations of 8 or 16 mg/L (11.45 or 22.9 µM), depending on the P. aeruginosa strain used. Furthermore, inhibition of bacterial growth in the presence of the simplified synthetic analogues of bromoageliferin was evaluated by means of growth curves and was found to occur at 400-500 µM. Thus, the original compound 2 appears to have a greater capacity to inhibit bacterial growth, with MICs of 8-32 mg/L (11.45-45.83 µM) obtained in the present study.
Little is known about the antimicrobial mechanism of action of bromoageliferin (2). Although 2 displays high activity against P. aeruginosa, the growth inhibition is not as significant with the other pathogens tested, i.e., A. baumannii and K. pneumoniae. However, the inhibitory effect on biofilm production in different bacteria seems to be demonstrated. The genes encoding structural subunits of fimbriae fimA and mfa1 of the oral pathogen Porphyromonas gingivalis show altered expression when the bacteria is grown in the presence of small molecules of bromoageliferin-derivates [43]. These molecules prevented P. gingivalis from binding to Streptococcus gordoni to form a mixed species biofilm community. The possible targets (fimA and mfa1) are involved in attachment and biofilm formation, which may partly explain the anti-virulence effect of bromoageliferin (2) observed in this study.

In vivo Efficacy of Bromoageliferin Against P. aeruginosa
Mammalian animal models are considered the gold standard for screening new drugs. However, they have important ethical and administrative restrictions and are costly. The in vivo Galleria mellonella (wax moth) model is suitable for studying P. aeruginosa infections and the results obtained correlate well with those obtained in mammals. Although the wax moth does not have an adaptive immune system, it does possess an immune system analogous to the innate immune system in humans, and the model is therefore suitable for studying acute infections [44,45]. We therefore The data obtained regarding the inhibition of biofilm production by bromoageliferin (2) are consistent with those previously obtained with two simplified synthetic analogues of 2 against P. aeruginosa [39]. Indeed, a concentration of 100-200 µM of these analogues was required to inhibit 50% of biofilm production, while for 2 we observed 30-40% biofilm inhibition at concentrations of 8 or 16 mg/L (11.45 or 22.9 µM), depending on the P. aeruginosa strain used. Furthermore, inhibition of bacterial growth in the presence of the simplified synthetic analogues of bromoageliferin was evaluated by means of growth curves and was found to occur at 400-500 µM. Thus, the original compound 2 appears to have a greater capacity to inhibit bacterial growth, with MICs of 8-32 mg/L (11.45-45.83 µM) obtained in the present study.
Little is known about the antimicrobial mechanism of action of bromoageliferin (2). Although 2 displays high activity against P. aeruginosa, the growth inhibition is not as significant with the other pathogens tested, i.e., A. baumannii and K. pneumoniae. However, the inhibitory effect on biofilm production in different bacteria seems to be demonstrated. The genes encoding structural subunits of fimbriae fimA and mfa1 of the oral pathogen Porphyromonas gingivalis show altered expression when the bacteria is grown in the presence of small molecules of bromoageliferin-derivates [43]. These molecules prevented P. gingivalis from binding to Streptococcus gordoni to form a mixed species biofilm community. The possible targets (fimA and mfa1) are involved in attachment and biofilm formation, which may partly explain the anti-virulence effect of bromoageliferin (2) observed in this study.

In Vivo Efficacy of Bromoageliferin against P. aeruginosa
Mammalian animal models are considered the gold standard for screening new drugs. However, they have important ethical and administrative restrictions and are costly. The in vivo Galleria mellonella (wax moth) model is suitable for studying P. aeruginosa infections and the results obtained correlate well with those obtained in mammals. Although the wax moth does not have an adaptive immune system, it does possess an immune system analogous to the innate immune system in humans, and the model is therefore suitable for studying acute infections [44,45]. We therefore decided to test the in vivo efficacy of bromoageliferin (2) in a G. mellonella survival assay, in larvae infected with the P. aeruginosa ATCC 27853 strain (Figure 4). decided to test the in vivo efficacy of bromoageliferin (2) in a G. mellonella survival assay, in larvae infected with the P. aeruginosa ATCC 27853 strain (Figure 4).
Although the survival rate to end point did not increase in larvae treated with bromoageliferin (2) relative to untreated larvae, a delay in death of the treated larvae was observed throughout the experiment. The mean survival time of larvae in the treated group was 18.3 h, compared with 13.5 h in the untreated larvae. Interestingly, at 20 h, once all untreated larvae were dead, a survival rate of 37.5% was observed in those treated with bromoageliferin (2). Therefore, significant differences were observed in mean survival time between treated and untreated larvae (p = 0.0035). Higher survival rates were not observed with higher concentrations of bromoageliferin (2) (5 and 20 mg/kg) (data not shown).

Additional Reported Activities for 1-8
In order to summarize the broad range of activities observed for 1-8, we list here other previously reported biological activities Ageliferin (1) has previously been reported to act as an antiviral agent (Herpes simplex virus-type 1 and Vesicular stomatitis) [22], antifouling agent (Balanus amphitrite amphitrite) [22] and potent actomyosin ATPase activator [21], and also to display activity against the somatostatin receptor and vasoactive intestinal peptide (VIP) receptor [46]. By contrast, compound 1 did not display antifungal (Penicillium atrovenetum and Saccharomyces cerevisiae) [22], cytotoxic (Artemia salina [47] and monkey kidney cells [22] or antifouling activity (Barnacle improvisus) [47] and yielded a negative response in a biochemical prophage induction assay (PIA) [22]. Bromoageliferin (2) has previously been reported to act as an antiviral agent (H. simplex-type 1 and V. stomatitis) [22], potent actomyosin ATPase activator [21], inhibitor of voltage-operated, but not store-operated calcium entry in PC12 cells [48] and as a potent feeding deterrent (Thalassoma bifasciatum) [49]. Other biological studies of compound 2 report no antifungal activity (P. atrovenetum and S. cerevisiae) [22], cytotoxic activity (monkey kidney cells) [22] or activity in the biochemical prophage induction (BIA) assay [22] and also no antitumoral activity against three human tumor cell lines (A549 lung cancer cells, HT29 colonic cancer cells and MDA-MB-231 breast cancer cells) [50]. Compound 3 has been reported to display antiviral activity (H. simplex virus-type 1 and V. stomatitis) [22], potent actomyosin ATPase activity [21], to inhibit voltage-operated, but not store-operated calcium entry in PC12 cells [48], and to display potent feeding deterrent activity (T. bifasciatum) [49]. Dibromoageliferin (3) did not display antifungal activity (P. atrovenetum and S. cerevisiae) [22], cytotoxic activity (monkey kidney cells) [22], antitumoral activity against three human tumor cell lines (A549 lung cancer cells, HT29 colonic cancer cells and MDA-MB-231 breast cancer cells) [50] or activity in a biochemical prophage induction assay [22]. Sceptrin (4) displayed antiviral activity (H. Although the survival rate to end point did not increase in larvae treated with bromoageliferin (2) relative to untreated larvae, a delay in death of the treated larvae was observed throughout the experiment. The mean survival time of larvae in the treated group was 18.3 h, compared with 13.5 h in the untreated larvae. Interestingly, at 20 h, once all untreated larvae were dead, a survival rate of 37.5% was observed in those treated with bromoageliferin (2). Therefore, significant differences were observed in mean survival time between treated and untreated larvae (p = 0.0035). Higher survival rates were not observed with higher concentrations of bromoageliferin (2) (5 and 20 mg/kg) (data not shown).

Additional Reported Activities for 1-8
In order to summarize the broad range of activities observed for 1-8, we list here other previously reported biological activities.

General Experimental Chemical Procedures
Optical rotations were measured in a JASCO DIP-1000 polarimeter (JASCO, Tokyo, Japan), with a Na (589 nm) lamp and filter. 1 H, 13 C and 2D NMR spectra were recorded in a Bruker Avance 500 spectrometer, at 500 and 125 MHz, respectively, with CD 3 OD and D 2 O as solvents. HRESIMS experiments were performed in an Applied Biosystems QSTAR Elite system or a Thermo MAT95XP spectrometer. HPLC separations were performed in the Agilent 1100 liquid chromatography system equipped with a solvent degasser, quaternary pump, and diode array detector (Agilent Technologies, Waldbronn, Germany) with a semipreparative reversed phase column (Luna C18: 5 µ, 100 Å, 250 × 10 mm, Phenomenex, Lane Cove, Australia). Precoated silica gel plates (Merck, Kieselgel 60 F254, 0.25 mm, Merck Millipore, Merck KGaA, Darmstadt, Germany) were used for TLC analysis and the spots were visualized under a UV light (254 nm) or by heating the plate pretreated with H 2 SO 4 /H 2 O/AcOH (1:4:20).

Sponge Collection
The sponge A. dilatata was collected by SCUBA from the waters surrounding Cozumel Island, Quintana Roo (20 • 43 55.03" N/87 • 00 24.70" W), at depths ranging from 10 to 15 m, in October 2016. The sponges were frozen immediately after collection. A voucher specimen E25-1 was deposited in the Phylum Porifera Gerardo Green National Collection of the Institute of Marine Sciences and Limnology (ICMyL) at the National Autonomous University of Mexico (UNAM), Mexico City.

Extraction and Isolation
Sliced bodies of A. dilatata (wet weight, 431.2 g; dry weight, 113.0 g) were exhaustively extracted with CH 3 OH-CH 2 Cl 2 (1:1, 3 × 1.5 L) at room temperature. The combined extracts were concentrated under reduced pressure to yield 20.0 g of a crude residue that was first partitioned between CH 2 Cl 2 and H 2 O (1:1 v/v). The resulting aqueous portion was extracted with n-butanol (200 mL) to yield the n-butanol fraction (3.25 g). The organic phase was concentrated under reduced pressure and partitioned between 10% aqueous CH 3 OH (400 mL) and hexane (2 × 400 mL) to produce 227.4 mg of the hexane fraction, after removal of the solvent under reduced pressure. The H 2 O content (% v/v) of the methanolic fraction was adjusted to 50% aqueous CH 3 OH, and this mixture was extracted with CH 2 Cl 2 (100 mL) to yield 109.4 mg of the CH 2 Cl 2 fraction and 150.4 mg of the remaining aqueous methanolic fraction, after removal of the solvent under reduced pressure.

Structural Characterization
Ageliferin (1) Table 2. Bacterial strains were frozen in Luria-Bertani (LB) with 10% glycerol and stored at −80 • C until analysis, when they were grown at 37 • C in LB medium.

Microdilution Method: Minimum Inhibitory Concentration
The minimum inhibitory concentrations (MIC) of 1-8 were evaluated against bacterial strains by the microdilution method, as described by Clinical and Laboratory Standards Institute (CLSI), with some modifications [67]. Dimethylsulfoxide (DMSO) was used to dissolve the crude extracts, at a maximum concentration of 1.2% v/v in the well with the highest concentration of the plate (128 mg/L). Briefly, the strains were cultured overnight in Mueller Hinton II (MH) agar plates (Becton Dickinson) at 37 • C, and the turbidity of the bacterial suspensions was standardized at 0.5 on the McFarland scale to prepare the inocula. Wells were inoculated with approximately 1 × 10 6 colony forming units/mL. Two-fold serial dilutions of compounds were performed in 96-wells microplates, in Mueller Hinton II broth medium (Sigma, St. Louis, MO, USA). The range of extract concentrations used for MIC analysis was 0.5-128 mg/L. One well was used in each row as positive growth control, composed of growth media and bacterial suspension, and another well, used as a negative control, consisted of medium without bacterial inoculum. Solvent controls of DMSO were included to determine whether the used concentration interfered with bacterial growth. The β-lactam antibiotic imipenem, which displays a broad spectrum of activity against Gram-negative bacteria, was used as a control for the microdilution methodology. The minimum inhibitory concentration was determined after incubation for 20-24 h at 37 • C and was established as the lowest concentration of the compound in which the bacterial strains did not grow. All extracts were tested in triplicate.

Biofilm Inhibition Assay
P. aeruginosa strains ATCC 27853 and PAO1 were cultivated on MH agar for 18 h at 37 • C and used to inoculate 5 mL of MH broth. These cultures were, in turn, grown overnight at 37 • C with shaking. A 1:100 dilution of each strain (initial inoculum of approx. 1 × 10 7 CFU/mL) was then incubated for 24 h in 24-well plates. Assays were performed in the presence of sub-MICs of bromoageliferin.
Bacterial growth was then measured at OD 600nm , in an Epoch 2 Microplate Spectrophotometer (BioTek Instruments, VT, USA), to determine the total cell biomass. Afterward, medium with bacteria was removed from the wells and then, they were washed with phosphate-buffered saline (PBS). Biofilm formation was determined by staining with a final concentration of 10% crystal violet per well, washing vigorously with PBS and solubilizing in 30% (v/v) acetic acid. The OD 580nm /OD 600nm ratio was calculated to normalize the amount of produced biofilm to the total cell biomass, thus avoiding variations due to different culture conditions. A minimum of 5 replicates were analyzed per condition. A Student's t-test was carried out with GraphPad Prism (GraphPad Software, San Diego, CA, USA), in order to evaluate the statistical significance of observed differences (p ≤ 0.05).

Galleria Mellonella Treatment
The efficacy of bromoageliferin (2) treatment was evaluated in a G. mellonella survival assay, as previously described [68]. The larvae were obtained from BioSystems Technology. Briefly, P. aeruginosa ATCC 27853 was grown to an OD 600nm of 0.7 at 37 • C, centrifuged, washed and resuspended in sterile PBS. Two groups of 15 larvae were injected with 10 µL of bacterial suspension containing 5 × 10 2 CFU/mL. The treatment evaluated was bromoageliferin (2 mg/kg), and the same volume of sterile PBS was used as control. Higher bromoageliferin concentrations (5 and 20 mg/kg) were also tested. The groups were incubated at 37 • C in darkness, and survival was monitored during a period of 30 h. The resulting survival curves were plotted using the Kaplan-Meier method and analysed using the log-rank (Mantel-Cox) test.

Conclusions
We describe the isolation and structural characterization of eight known alkaloids 1-8, most of which are bromopyrrols, isolated from specimens of the sponge A. dilatata collected on Cozumel Island (Yucatan Peninsula, Mexico), in the first chemical assessment of the secondary metabolites from this sponge. Although the antibacterial activity of these compounds has previously been reported, the activity of most of the compounds was evaluated for the first time in relation to problematic multidrug resistant pathogens, i.e., the Gram-negative bacteria K. pneumoniae, P. aeruginosa and A. baumannii. Bromoageliferin (2) and dibromoageliferin (3) display both moderate antibacterial activity against the P. aeruginosa PAO1 strain while they show high and moderate antibacterial activity, respectively, against the P. aeruginosa ATCC27853 strain. Compound 2 inhibited growth and biofilm production in P. aeruginosa, and increased the survival time of larvae in the in vivo G. mellonella assay. Moreover, a relationship between structure and activity was deduced from the antibacterial analysis of the three isolated ageliferins with similar structures (1-3). In summary, in vitro and in vivo findings for the multidrug resistant pathogen P. aeruginosa indicate bromoageliferin (2) as a promising lead for optimization in the design of new antimicrobial therapies.