Antimicrobial Diterpene Alkaloids from an Agelas citrina Sponge Collected in the Yucatán Peninsula

Three new diterpene alkaloids, (+)-8-epiagelasine T (1), (+)-10-epiagelasine B (2), and (+)-12-hydroxyagelasidine C (3), along with three known compounds, (+)-ent-agelasine F (4), (+)-agelasine B (5), and (+)-agelasidine C (6), were isolated from the sponge Agelas citrina, collected on the coasts of the Yucatán Peninsula (Mexico). Their chemical structures were elucidated by 1D and 2D NMR spectroscopy, HRESIMS techniques, and a comparison with literature data. Although the synthesis of (+)-ent-agelasine F (4) has been previously reported, this is the first time that it was isolated as a natural product. The evaluation of the antimicrobial activity against the Gram-positive pathogens Staphylococcus aureus, Streptococcus pneumoniae, Enterococcus faecalis showed that all of them were active, with (+)-10-epiagelasine B (2) being the most active compound with an MIC in the range of 1–8 µg/mL. On the other hand, the Gram-negative pathogenes Acinetobacter baumannii, Pseudomonas aeruginosa, and Klebsiella pneumoniae were also evaluated, and only (+)-agelasine B (5) showed a moderate antibacterial activity with a MIC value of 16 μg/mL.


Introduction
One of the most common sponges in tropical and subtropical areas around the world are marine sponges of the genus Agelas (class Demospongiae, order Agelasida, family Agelasidae) [1]. Covering the period from 1971 to November 2021, 355 compounds have been isolated from Agelas sponges with interesting biological activities [2]. Many of these metabolites are of mixed biogenetic origins, as illustrated by alkaloids (especially pyrrole alkaloids and terpenoid alkaloids) [3,4], glycosphingolipids, sterols, and carotenoids that display diverse biological activities and unique structural features [5][6][7].
In particular, the nitrogenated diterpenes that include hypotaurocyamines (agelasidines) and N 9 -adeninium alkaloids (agelasines) represent an important class of compounds isolated from this genus and they show interesting biological activities, including antibacterial, cytotoxic, antifouling, antifungal, and inhibitory effects on Na + /K + -ATPase [8].
compounds isolated from this genus and they show interesting biological activities, including antibacterial, cytotoxic, antifouling, antifungal, and inhibitory effects on Na + /K + -ATPase [8].
As part of our ongoing efforts to find new bioactive natural compounds from marine organisms collected off the coast of the Yucatán Peninsula (Mexico) [9][10][11][12], specimens of the sponge A. citrina were investigated due to the antibacterial activity displayed by their organic extracts. Former studies of the sponge A. citrina reported the presence of diterpene alkaloids [13] and pyrrole-imidazole alkaloids [14]. Our preliminary chemical investigation of this sponge led to the isolation of the major known component (−)-agelasine B, which showed significant antibacterial activity against two Gram-positive bacteria S. aureus strains [10]. Further bioactivity-guided fractionation of the CH2Cl2 and aqueous methanolic fractions led to the isolation of three new compounds (1, 2, and 3), as well as three known compounds (4, 5, and 6). All isolated compounds (1-6) were evaluated for their antibacterial activity. Herein, we describe the isolation, structure elucidation, and antibacterial activity of 1-6.

Isolation and Identification of Agelasines
Specimens of the sponge A. citrina collected on Cozumel Island of the Yucatán Penisula (Mexico) were extracted several times with CH3OH/CH2Cl2 to produce an organic extract, which was partitioned following our standard partitioning procedure [15] into several fractions of differing polarities. The CH2Cl2 and aqueous methanolic fractions were submitted to Solid Phase Extraction (SPE) with RP-18 cartridges and finally to RP-HPLC to produce pure compounds 1-6 ( Figure 1). The 1 H and 13 C NMR spectral data, along with an HSQC spectrum of 1 in DMSO-d6 (Table 1), were indicative of an agelasine-type diterpene alkaloid compound. The adenine moiety was identified by the characteristic signals of two methines at δH/δC 8.46 (1H, s, H-  The 1 H and 13 C NMR spectral data, along with an HSQC spectrum of 1 in DMSO-d 6 (Table 1), were indicative of an agelasine-type diterpene alkaloid compound. The adenine moiety was identified by the characteristic signals of two methines at δ H /δ C 8.46 (1H, s, H-2 )/155.4 (C-2 ) and 9.51 (1H, s, H-8 )/140.8 (C-8 ); a N-methyl group at δ H /δ C 3.88 (3H, s)/31.1; and three non-protonated sp 2 carbons at δ C 152.4 (C-6 ), 149.0 (C-4 ), and 109.2 (C-5 ). The diterpene portion of 1 was assigned from the 1 H and 13 -20); two quaternary carbons at δ C 32.9 (C-4) and 38.5 (C-10); and one non-protonated carbon linked to oxygen at δ C 72.1 (C-8). Moreover, the 1 H and 13 C NMR signals corresponding to three methylenes at δ H /δ C 1.55 (1H, m, H-11a)-1.27 (1H, m, H-11b)/22.7 (C-11), 2.22 (1H, td, J = 12.2, 6.1 Hz, H-12a)-2.09 (1H, td, J = 12.2, 4.9 Hz, H-12b)/42.2 (C-12), and 5.13 (2H, d, J = 7.3 Hz, H-15)/46.6 (C-15); a vinyl methyl at δ H /δ C 1.79 (3H, s, H 3 -16)/16.5 (C-16); an olefinic methine at δ H /δ C 5.44 (1H, t, J = 7.0 Hz, H-14)/114.6 (CH); and a quaternary sp 2 carbon at δ C 147.0 (C-13) confirmed the presence of the 3-methylpentenyl chain. The analysis of COSY and HMBC spectra was able to confirm the planar structure of 1 ( Figure 2). The 1D NMR spectral data of 1 were similar to those for agelasine T, a previously reported agelasine isolated from an unidentified Agelas collected in Okinawa [16]. However, the difference of the carbon chemical shift of the C-17 methyl group at δ C 23.6 in 1 instead of the reported value of that group at δ C 30.7 in agelasine T, suggested that they differed in the spatial disposition of the C-17 methyl group, and consequently in the stereochemistry at C-8 position. The analysis of COSY and HMBC spectra was able to confirm the planar structure of 1 ( Figure 2). The 1D NMR spectral data of 1 were similar to those for agelasine T, a previously reported agelasine isolated from an unidentified Agelas collected in Okinawa [16]. However, the difference of the carbon chemical shift of the C-17 methyl group at δC 23.6 in 1 instead of the reported value of that group at δC 30.7 in agelasine T, suggested that they differed in the spatial disposition of the C-17 methyl group, and consequently in the stereochemistry at C-8 position. The relative configuration was assigned by 2D and selective 1D-NOESY experiments of 1. The E-geometry of the Δ 13 double bond was established by the observed NOESY correlation between the signals assigned to the methylene CH2-15 at δH 5.13 and vinyl methyl CH3-16 at δH 1.79. On the other hand, the irradiation of methyl protons ( Figure S6) at the position CH3-17 (δH 1.00) showed a 2.0% NOE response to the methyl protons at the position CH3-20 (δH 0.74); the irradiation of methyl protons ( Figure S7) at the position CH3-18 (δH 0.84) also displayed NOE to the methyl protons at the position CH3-20 (δH 0.74) (3.0%) (see Figure 3A). Thus, these NOE correlations suggest that these three methyl groups are located on the same face of the molecule, assigned as the β face. Furthermore, the carbon chemical shift of CH3-17 (δC 23.6) resonates similarly to that of a synthetic terpene having a bicyclic ring bearing the same spatial disposition of the hydroxy group (δC 24.2) [17].  The relative configuration was assigned by 2D and selective 1D-NOESY experiments of 1. The E-geometry of the ∆ 13 double bond was established by the observed NOESY correlation between the signals assigned to the methylene CH 2 -15 at δ H 5.13 and vinyl methyl CH 3 -16 at δ H 1.79. On the other hand, the irradiation of methyl protons ( Figure S6) at the position CH 3 -17 (δ H 1.00) showed a 2.0% NOE response to the methyl protons at the position CH 3 -20 (δ H 0.74); the irradiation of methyl protons ( Figure S7) at the position CH 3 -18 (δ H 0.84) also displayed NOE to the methyl protons at the position CH 3 -20 (δ H 0.74) (3.0%) (see Figure 3A). Thus, these NOE correlations suggest that these three methyl groups are located on the same face of the molecule, assigned as the β face. Furthermore, the carbon chemical shift of CH 3 -17 (δ C 23.6) resonates similarly to that of a synthetic terpene having a bicyclic ring bearing the same spatial disposition of the hydroxy group (δ C 24.2) [17]. The analysis of COSY and HMBC spectra was able to confirm the planar structure of 1 ( Figure 2). The 1D NMR spectral data of 1 were similar to those for agelasine T, a previously reported agelasine isolated from an unidentified Agelas collected in Okinawa [16]. However, the difference of the carbon chemical shift of the C-17 methyl group at δC 23.6 in 1 instead of the reported value of that group at δC 30.7 in agelasine T, suggested that they differed in the spatial disposition of the C-17 methyl group, and consequently in the stereochemistry at C-8 position. The relative configuration was assigned by 2D and selective 1D-NOESY experiments of 1. The E-geometry of the Δ 13 double bond was established by the observed NOESY correlation between the signals assigned to the methylene CH2-15 at δH 5.13 and vinyl methyl CH3-16 at δH 1.79. On the other hand, the irradiation of methyl protons ( Figure S6) at the position CH3-17 (δH 1.00) showed a 2.0% NOE response to the methyl protons at the position CH3-20 (δH 0.74); the irradiation of methyl protons ( Figure S7) at the position CH3-18 (δH 0.84) also displayed NOE to the methyl protons at the position CH3-20 (δH 0.74) (3.0%) (see Figure 3A). Thus, these NOE correlations suggest that these three methyl groups are located on the same face of the molecule, assigned as the β face. Furthermore, the carbon chemical shift of CH3-17 (δC 23.6) resonates similarly to that of a synthetic terpene having a bicyclic ring bearing the same spatial disposition of the hydroxy group (δC 24.2) [17].  The similarity of 1 H and 13 C NMR spectral in CDCl 3 of 2 to those of 1 suggested that 2 was also an agelasine-type alkaloid with an N 9 -methyladenine unit.
The relative configuration of 2 was deduced from NOESY data and confirmed by a IPAP-HSQMBC experiment. The E-configuration of the ∆ 13 double bond of 2, as in agelasine B (5), was deduced from NOE correlations observed in the NOESY experiment between the signals assigned to the methylene CH 2 3 -20) are oriented on the same face of the molecule, assigned as β-face.
The drastic change in the chemical shift of H-10 in 2 at δ H 2.28 (1H, brd, J = 12.9 Hz) in relation to that in agelasine B (5) at δ H 1.29 (1H, m) (Table 1), and considering that agelasine B (5) has the same spatial disposition for the three methyl groups H 3 -17, H 3 -19, and H 3 -20, suggested they must differ in the remaining stereogenic center at C-10. Similar chemical shifts and the same HMBC and NOESY correlations were observed in 2 when the NMR spectra were run in C 6 D 6 (Table S1). Therefore, these results suggest a cis-clerodane disposition for compound 2, instead of the trans-clerodane disposition for agelasine B (5).
Curiously, when the NMR spectra of 2 was repeated this time in CDCl 3 , an important displacement of the carbon and proton chemical shifts was observed in the NMR spectra after 24 h, probably due to the acidic character of the deuterated solvent. Thus, the carbon and proton chemical were assigned again by 2D-NMR experiments (see Table S1 and Figures S23-S31). The low value of the 3 J CH between H-10 and C-19 of 3.42 Hz measured in an IPAP-HSQMBC experiment of this sample ( Figure S31) agrees with the proposed cis-clerodane disposition for 2. Unfortunately, the 3 J CH between H-10 and C-19 in agelasine B (5) could not be determined due to the insufficient amount. In order to obtain a direct proof for the cisvs. trans-ring connections in decaline-like systems using the IPAP-HSQMBC experiment, we measured the corresponding values in the available synthetic model compounds 7 and 8 ( Figure 4) [18]. In this way, the corresponding value of the cis-decalin-like system in 7 ( 3 J C7H5 = 2.9 Hz) resulted to be similar to that of 2 ( Figure S32). As expected, the corresponding value of the trans-decalin-like system ( 3 J C7H5 = 8.42 Hz) was much larger ( Figure S33). and proton chemical were assigned again by 2D-NMR experiments (see Table S1 and Figures S23-S31). The low value of the 3 JCH between H-10 and C-19 of 3.42 Hz measured in an IPAP-HSQMBC experiment of this sample ( Figure S31) agrees with the proposed cisclerodane disposition for 2. Unfortunately, the 3 JCH between H-10 and C-19 in agelasine B (5) could not be determined due to the insufficient amount. In order to obtain a direct proof for the cis-vs. trans-ring connections in decaline-like systems using the IPAP-HSQMBC experiment, we measured the corresponding values in the available synthetic model compounds 7 and 8 ( Figure 4) [18]. In this way, the corresponding value of the cisdecalin-like system in 7 ( 3 JC7H5 = 2.9 Hz) resulted to be similar to that of 2 ( Figure S32). As expected, the corresponding value of the trans-decalin-like system ( 3 JC7H5 = 8.42 Hz) was much larger ( Figure S33). The 1 H-NMR and 13 C NMR spectral data of 3 in CDCl3 ( Table 2)   The 1 H-NMR and 13 C NMR spectral data of 3 in CDCl 3 ( Table 2)  The 1D NMR spectral data of 3 were similar to those of (+)-agelasidine C, isolated from Agelas nakamurai collected in Okinawa [19]. The presence of an additional hydroxyl group in 3 in relation to (+)-agelasidine C was suggested by the difference of an oxygen (16 Da) between their molecular formula. This was confirmed by the presence of an additional oxymethine at δ H /δ C 4.09 (1H, t, J = 6.5 Hz, H-12)/76.3 (C-12) in the 1 H and 13 C NMR spectra of 3, instead of the corresponding methylene group present in (+)-agelasidine C. The HMBC correlations between the vinyl methyl protons at δ H 1.71 (3H, s, H 3 -16) and the oxymethine carbon at δ C 76.3 (C-12), the quaternary sp 2 carbon at δ C 149.2 (C-13) and the olefinic methine carbon at δ C 110.0 (C-14), located the hydroxyl group at C-12 position (Figure 2). The analysis of COSY and HMBC spectra was able to confirm the constitution of 3 ( Figure 2). These data indicate that 3 is a new agelasidine, which was named 12-hydroxyagelasidine C. The relative configuration of the monocyclic ring system in 3 was confirmed by a comparison of its 13 C NMR spectral data with that of (+)-agelasidine C [19] and other agelasidines [20] having the same relative configuration. The E-configuration of the ∆ 9 and ∆ 13 double bonds was established from the NOESY correlation between the signals assigned to the methylene CH 2 -11 at δ H 2.27 and the vinyl methyl CH 3 -17 at δ H 1.61, and between the methylene CH 2 -15 at δ H 3.84-3.80 and the vinyl methyl CH 3 -16 at δ H 1.71, respectively. Absolute configuration at C-12 in 3 was determined by the Mosher's method using the MTPA esters [21]. The 12-hydroxyagelasidine C (3) was treated with R-(−)and S-(+)-α-methoxy-α-(trifluoromethyl) phenyl acetic acid (MTPA-OH) to afford the Sand R-MTPA esters. The analysis of the 1 H NMR and 1 H− 1 H COSY led to the assignment of both esters' chemical shifts in proximity of C-12. The ∆δ SR values between Rand S-MTPA esters of 3 at 12-OH were negative for H-11/10/17 and positive for H-14/15/18 ( Figure 5), which suggested the absolute configuration of C-12 as R.

General Experimental Chemical Procedures
Optical rotations were measured on a JASCO DIP-1000 polarimeter, with an Na (589 nm) lamp and filter. 1 H, 13 C, and 2D NMR spectra were recorded on a Bruker spectrometer of 950 MHz, equipped with a 5 mm Cryo Probe (NEO console); a Bruker spectrometer (800 MHz for 1 H and 200 MHz for 13 C) equipped with a 5 mm cryo probe and an NEO console; a Bruker (800 MHz for 1 H and 200 MHz for 13 C) spectrometer equipped with a 3 mm cryo probe and a NEO console; a Bruker spectrometer (700 MHz for 1 H and 175 MHz for 13 C) equipped with a 5 mm cryo probe and an Avance III console; and a Bruker Avance 500 spectrometer (500 MHz for 1 H and 125 MHz for 13 C) equipped with a 5 mm cryo probe, using DMSO-d 6 and CDCl 3 as solvents. Chemical shifts are reported in δ scale relative to DMSO-d 6 (δ 2.50 ppm for 1 H NMR, δ 39.51 ppm for 13 C NMR) and CDCl 3 (δ 7.26 ppm for 1 H NMR, δ 77.0 ppm for 13 C NMR).
HRESIMS experiments were performed on the Applied Biosystems QSTAR Elite system or a Thermo MAT95XP spectrometer. HPLC separations were performed on the Agilent 1100 liquid chromatography system equipped with a solvent degasser, quaternary pump, and diode array detector (Agilent Technologies, Waldbronn, Germany) using a semipreparative reversed phase column Luna C18 (5 µ, 100 Å, 250 × 10 mm).

Animal Material
The sponge A. citrina was collected by hand and traditional SCUBA-diving off Cozumel Island, Quintana Roo (20 • 41 00.00" N/87 • 01 32.66" W) at depths ranging from 10 to 15 m in March 2017, and frozen immediately after collection. A voucher specimen CZE56 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. citrina (wet weight, 729.6 g; dry weight, 375.3 g) were exhaustively extracted with CH 3 OH-CH 2 Cl 2 (1:1, 3 × 1.5 L) at 25 • C for 24 h each maceration. The combined extracts were concentrated under reduced pressure to produce 6.1 g of a crude residue. A total of 6.0 g was first partitioned between CH 2 Cl 2 /H 2 O (1:1 v/v) to produce aqueous and organic phases. The organic phase was concentrated under reduced pressure and partitioned between 10% aqueous CH 3 OH (400 mL) and hexane (2 × 400 mL). The hexane portion produced, after removing the solvent under reduced pressure, 672.2 mg of the hexane fraction (FH). The H 2 O content (% v/v) of the methanolic fraction was adjusted to 50% aqueous CH 3 OH, and the mixture was extracted with CH 2 Cl 2 (100 mL) to afford, after removing the solvent under reduced pressure, 3.6 g of the CH 2 Cl 2 fraction (FD) and 755.8 mg of the remaining aqueous methanolic fraction (FM) (Scheme S1).
The CH 2 Cl 2 fraction (3.6 g) was subjected to a Solid Phase Extraction (SPE) with RP-18 (Merck KGaA) using a discontinuous gradient from H 2 O to CH 3 OH and then CH 2 Cl 2 .
The aqueous methanolic fraction (755.8 mg) was submitted to a Solid Phase Extraction (SPE) with RP-18 (Merck KGaA) using a discontinuous gradient from H 2 O to CH 3 OH and then CH 2 Cl 2 . The separation of the fraction eluted with H 2 O/CH 3 OH (1:1, 181.5 mg) by RP-HPLC using a mobile phase consisting of 30 min of a gradient from 35% to 100% of CH 3 OH in H 2 O (v/v, each containing 0.04% trifluoroacetic acid), followed by a 10 min isocratic at 100% of CH 3 OH at a flow rate of 2.0 mL/min, afforded (+)-12-hidroxy agelasidine C (3) (7.0 mg; R t = 29.8 min) and (+)-agelasidine C (6) (27.0 mg; R t = 34.0 min).

Structural Characterization
(+)-8-epiagelasine T (1  Funding: This work was supported by grants RTI2018-093634-B-C22 from the State Agency for Research (AEI) of Spain, cofunded by the FEDER Programme from the European Union (MCIN/AEI/ 10.13039/501100011033/FEDER) and BLUEBIOLAB (0474_BLUEBIOLAB_1_E), Programme INTER-REG V A of Spain-Portugal (POCTEP). This work was supported by Projects PI17/01482 and PI20/01212 awarded to AB, all within in the National Plan for Scientific Research, Development and Technological Innovation 2017-2020 and funded by the ISCIII-General Subdirection of Assessment and Promotion of the Research-European Regional Development Fund (FEDER) "A way of making Europe". The work was also supported by CIBERINFEC (CIBER de Enfermedades Infecciosas). The study was also funded by project IN607D 2021/12 (GAIN-Agencia Gallega de Innovación-Consellería de Economía, Emprego e Industria) awarded to AB. The study was also funded by projects GRC2018/039 from Xunta de Galicia. Dawrin Pech-Puch received his postdoctoral fellowship from the National Council of Science and Technology (CONACYT) of Mexico. This work was supported by the Max Planck Society and the DFG (Gr1211/19-1)/CAPES 418729698 project.

Data Availability Statement:
All datasets related to this article can be obtained from the authors.