Osirisynes G-I, New Long-Chain Highly Oxygenated Polyacetylenes from the Mayotte Marine Sponge Haliclona sp.

Chemical study of the CH2Cl2−MeOH (1:1) extract from the sponge Haliclona sp. collected in Mayotte highlighted three new long-chain highly oxygenated polyacetylenes, osirisynes G-I (1–3) together with the known osirisynes A (4), B (5), and E (6). Their structures were elucidated by 1D and 2D NMR spectra and HRESIMS and MS/MS data. All compounds were evaluated on catalase and sirtuin 1 activation and on CDK7, proteasome, Fyn kinase, tyrosinase, and elastase inhibition. Five compounds (1; 3–6) inhibited proteasome kinase and two compounds (5–6) inhibited CDK7 and Fyn kinase. Osirisyne B (5) was the most active compound with IC50 on FYNB kinase, CDK7 kinase, and proteasome inhibition of 18.44 µM, 9.13 µM, and 0.26 µM, respectively.


Introduction
Natural products play a highly significant role as a source of new and approved drugs for the treatment of human diseases. Approximately 70% of small molecule drugs were produced between 1981 and 2006, possessing an important link to a natural product source [1,2]. Over the last few decades, the oceans (covering 70% of the earth) have represented a widely promising source of new biologically active natural compounds [3] with considerably different characteristics in comparison with those of the terrestrial ones [4].
In health, enzymes play key roles in different cellular processes and their deregulation has been considered as one of the first causes of aged-related diseases, including cancer [16,17] and Alzheimer's disease [18,19]. As good drug candidates, natural enzyme activators or inhibitors have received an increasing amount of attention for their potential pharmacological applications, especially those from marine origin [20].
In our continuing search for bioactive metabolites from marine invertebrates, the undescribed sponge Haliclona sp. collected in Mayotte (Indian Ocean), was investigated. The organic crude extract of this animal exhibited a potent inhibitory activity against proteasome as well as a significant inhibitory activity against tyrosinase. Bioassay-guided partitioning and separation by chromatographic methods led to the isolation of the three known long-chain highly oxygenated polyacetylenes osirisynes A (4), B (5), and E (6) together with three new long-chain highly oxygenated polyacetylenes osirisynes G-I (1-3) (Figure 1). The purification and structure elucidation by spectral data including HRESIMS, MS/MS, and 2D NMR and comparison with published data [12], are reported herein. The biological evaluations of the latter new compounds against seven different targets involved in aging or age-related diseases are described as well.

Chemistry
The CH2Cl2-MeOH extract of the lyophilized sponge Haliclona sp. was first subjected to a normalphase silica gel column chromatography to yield 12 fractions. Fraction 9 was subjected to repetitive reverse-phase semi-preparative and analytical HPLC to yield six pure compounds (1-6) ( Figure 2). Among them, three are known and were identified as osirisynes A (4), B (5), and E (6) by comparison with published spectroscopic data; the other three are new and were named osirisynes G-I (1-3). Their elucidation is described below.

Chemistry
The CH 2 Cl 2 -MeOH extract of the lyophilized sponge Haliclona sp. was first subjected to a normal-phase silica gel column chromatography to yield 12 fractions. Fraction 9 was subjected to repetitive reverse-phase semi-preparative and analytical HPLC to yield six pure compounds (1-6) ( Figure 2). Among them, three are known and were identified as osirisynes A (4), B (5), and E (6) by comparison with published spectroscopic data; the other three are new and were named osirisynes G-I (1-3). Their elucidation is described below. Osirisyne G (1) was obtained as a white amorphous solid. The molecular formula, C47H72O12, was established from a HRESIMS molecular ion peak at m/z 827.4950 [M -H] -, indicating 12 degrees of unsaturation ( Figure S1). Analysis of the 1D and 2D 1 H and 13 C NMR data for 1 (CD3OD, Table 1, Figure S2-S5) revealed resonances and correlations consistent with those of a long-chain highly oxygenated polyacetylene, like osirisynes A-F [12] or fulvynes A-I [11]. The 1 H NMR spectrum of 1 recorded in CD3OD showed the presence of four olefinic protons (δH 5.88 (1H, ddd, J = 15.4, 6.2, 1. ) and a series of methylene groups in the range δH 2.50-1.30. The 13 C NMR spectrum of 1 showed the presence of a ketone C-19 (δC 214.5), a carboxylic acid C-1 (δC 161.3), eight sp carbons due to four triple bonds C-2, C-3, C-32, C-33, C-35, C-36, C-46, and C-47 (δC 79.9, 83.4, 85.0, 83.8, 80.9, 81.9, 84.7, 74.9), four sp 2 carbons due to two double bonds C-25, C-26, C-43, and C-44 (δC 132.3, 134.3, 136.0, 130.3), nine oxymethines C-4, C-5, C-6, C-27, C-31, C-34, C-38, C-42, and C-45 (δC 65.0, 78.5, 72.7, 73.5, 62.5, 52.3, 70.7, 72.2, 62.5), and several methylene groups were also present.  Figure S1). Analysis of the 1D and 2D 1 H and 13 C NMR data for 1 (CD 3 OD, Table 1, Figures S2-S5) revealed resonances and correlations consistent with those of a long-chain highly oxygenated polyacetylene, like osirisynes A-F [12] or fulvynes A-I [11]. The 1 H NMR spectrum of 1 recorded in CD 3 30. The 13 C NMR spectrum of 1 showed the presence of a ketone C-19 (δ C 214.5), a carboxylic acid C-1 (δ C 161.3), eight sp carbons due to four triple bonds C-2, C-3, C-32, C-33, C-35, C-36, C-46, and C-47 (δ C 79.9, 83.4, 85.0, 83.8, 80.9, 81.9, 84.7, 74.9), four sp 2 carbons due to two double bonds C-25, C-26, C-43, and C-44 (δ C 132.3, 134.3, 136.0, 130.3), nine oxymethines C-4, C-5, C-6, C-27, C-31, C-34, C-38, C-42, and C-45 (δ C 65.0, 78.5, 72.7, 73.5, 62.5, 52.3, 70.7, 72.2, 62.5), and several methylene groups were also present. All proton-bearing carbons were assigned by HSQC experiment. Analysis of the COSY and HMBC correlations aided in recognizing the partial structures a-d ( Figure 3) of the long alkyl chain of compound 1. The COSY correlations revealed the presence of the spin system C-4-C-5-C-6-C-7 for the partial structure a and the HMBC correlations between H-4, C-2, and C-3 and between H-5 and C-2 indicated the carbon resonances of the triple bond in fragment a. Hz, respectively). The relative configuration of compound 1 remained unassigned. Any attempts to obtain suitable derivatives (Mosher's esters) for a stereochemical analysis were unsuccessful due to the small amount of product and the high number of chiral carbons. The connectivities between partial structures a-d for 1 as well as the number of the linking methylene groups were established on the basis of the molecular formula and ESI-MS/MS data. The combinations of partial structures a + b + c + d represented 674 m.u. whereas the molecular structure weight was 828 m.u. The difference corresponding to 11 methylene groups determined the length of the alkyl chains between the different partial structures. For the ESI-MS/MS analysis, Cu II was used for compound ionization so the species sought were detected as [(M − H) + Cu II ] + . Indeed, a reduction of Cu II copper to Cu I can occur during the electrospray ionization process [21]. This reduction is accompanied by the formation of radical species, which can be at the origin of specific fragmentations during the experiments of collision-induced dissociation (CID) and cannot be obtained from the dissociation of protonated species [M + H] + for this type of compound [22]. Indeed, the CID of these protonated species is generally accompanied by the loss of non-specific molecules such as H 2 O molecules. The ESI-MS/MS spectra ( Figure 4) showed different fragment ions that indicated the presence of nine methylenes between a and b, one methylene between b and c, and one methylene between c and d. The fragmentation of the molecule was explained by different dissociation mechanisms ( Figure 5). In ESI + -MS/MS, the peak at m/z 846.4 corresponded to the loss of a CO 2 molecule and confirmed the presence of the carboxylic acid C-1. The presence of this carboxylic acid was also confirmed in ESI − -MS/MS by the presence of the ion [M − H] − at m/z 827.5 and the peak at m/z 783.5 corresponded to the loss of a CO 2 molecule. In ESI + -MS/MS, the peak at m/z 761.4 can be explained by a rearrangement between the three hydroxyls in C-4, C-5, and C-6, yielding to a loss of a H 2 O molecule and a 129.0 Da fragment corresponding to the C 5 H 3 O 3 formula. This mechanism confirmed the sequence C-1 to C-6. The peak at m/z 542.2 can be explained by a fragmentation between C-27 and C-28. This fragment confirmed the link between the partial structure b and c by one methylene. The peak at m/z 708.3 can be explained by a fragmentation between C-37 and C-38. The loss of a 182.1 Da fragment corresponding to a C 10 H 14 O 3 confirmed the link between the partial structure c and d by one methylene. In ESI − -MS/MS, the peak at m/z 601.4 was also due to a fragmentation between C-37 and C-38 and confirmed this link. These fragments (m/z 542. corresponding to 11 methylene groups determined the length of the alkyl chains between the different partial structures. For the ESI-MS/MS analysis, Cu II was used for compound ionization so the species sought were detected as [(M -H) + Cu II ] + . Indeed, a reduction of Cu II copper to Cu I can occur during the electrospray ionization process [21]. This reduction is accompanied by the formation of radical species, which can be at the origin of specific fragmentations during the experiments of collision-induced dissociation (CID) and cannot be obtained from the dissociation of protonated species [M + H] + for this type of compound [22]. Indeed, the CID of these protonated species is generally accompanied by the loss of non-specific molecules such as H2O molecules. The ESI-MS/MS spectra ( Figure 4) showed different fragment ions that indicated the presence of nine methylenes between a and b, one methylene between b and c, and one methylene between c and d. The fragmentation of the molecule was explained by different dissociation mechanisms ( Figure 5). In

Biological Activity
While the CH 2 Cl 2 -MeOH extract from Haliclona sp. sponge presented significant anti-tyrosinase activity (31.1%) (Figure 6), one of these molecules, osirisyne E (6), was already described as an enzyme inhibitor. This molecule, as well as osyrisines C and F, had shown Na + /K + ATPase and reverse transcriptase (RT) inhibitory activities [13]. The six isolated osirisynes (1-6) were submitted to a biological evaluation against seven different targets involved in aging or age-related diseases.
These targets include biological assays on catalase and sirtuin 1 activation and on CDK7, Fyn kinase, tyrosinase, elastase, and proteasome inhibition. targets include biological assays on catalase and sirtuin 1 activation and on CDK7, Fyn kinase, tyrosinase, elastase, and proteasome inhibition. Catalase is a common enzyme located at the peroxisome that prevents cell oxidative damage (oxidized proteins, lipids, and DNA) by converting hydrogen peroxide (H2O2) into water (H2O) and dioxygen (O2). This antioxidant enzyme prevents the accumulation of hydrogen peroxide, which is continuously produced by metabolic reactions and belongs to the reactive oxygen species (ROS), in cellular organelles and tissues. Indeed, ROS are associated to the pathogenesis of numerous diseases including age-related diseases [23][24][25]. Finding natural products that are catalase activators can increased the intracellular antioxidant defense system capacity and can be useful in preventing these diseases.
Sirtuin 1 is a member of the sirtuin family of proteins, a group of very promising targets for antiaging approaches [26] with activities linked to crucial biological processes like regulating ribosomal DNA recombination, gene silencing, DNA repair, chromosomal stability, and longevity [27].
CDK7 is one of the cyclin-dependent kinases (CDKs), known for their critical roles in cell cycle regulation but also involved in other physiological process like DNA repair and transcription [28]. This kinase has been reported in recent studies to be crucial for the pathogenesis of certain cancer types driven by transcription of a key set of genes and has been validated as a therapeutic target for cancers [29][30][31].
Fyn is a member of the Src family of protein tyrosine kinases (PTKs), an important class of molecules in human biology. Fyn's biological functions are diverse, and include signaling via the T cell receptor, regulation of brain function, as well as adhesion mediated signaling [32]. Recent studies highlight the involvement of this kinase in different age-related diseases such as cancers [33] or Alzheimer's disease [34]. Fyn interacts with both protein Tau and amyloid β-peptide, two key players responsible for the major pathologic hallmarks of Alzheimer's disease [35,36] and inhibitors of this kinase seem to be a promising novel approach therapy of this disease [18].
Tyrosinase, a copper-containing metalloenzyme, is a key enzyme involved in melanogenic processes [37]. In humans, melanin helps defend skin from the damage caused by UV light, however, excess levels of melanin can cause various dermatological disorders including hyperpigmentations, melisma, freckles, and age spots. Many tyrosinase inhibitors have been used in cosmetics and pharmaceutical products for the prevention of overproduction of melanin in the epidermis, however, Catalase is a common enzyme located at the peroxisome that prevents cell oxidative damage (oxidized proteins, lipids, and DNA) by converting hydrogen peroxide (H 2 O 2 ) into water (H 2 O) and dioxygen (O 2 ). This antioxidant enzyme prevents the accumulation of hydrogen peroxide, which is continuously produced by metabolic reactions and belongs to the reactive oxygen species (ROS), in cellular organelles and tissues. Indeed, ROS are associated to the pathogenesis of numerous diseases including age-related diseases [23][24][25]. Finding natural products that are catalase activators can increased the intracellular antioxidant defense system capacity and can be useful in preventing these diseases.
Sirtuin 1 is a member of the sirtuin family of proteins, a group of very promising targets for anti-aging approaches [26] with activities linked to crucial biological processes like regulating ribosomal DNA recombination, gene silencing, DNA repair, chromosomal stability, and longevity [27].
CDK7 is one of the cyclin-dependent kinases (CDKs), known for their critical roles in cell cycle regulation but also involved in other physiological process like DNA repair and transcription [28]. This kinase has been reported in recent studies to be crucial for the pathogenesis of certain cancer types driven by transcription of a key set of genes and has been validated as a therapeutic target for cancers [29][30][31].
Fyn is a member of the Src family of protein tyrosine kinases (PTKs), an important class of molecules in human biology. Fyn's biological functions are diverse, and include signaling via the T cell receptor, regulation of brain function, as well as adhesion mediated signaling [32]. Recent studies highlight the involvement of this kinase in different age-related diseases such as cancers [33] or Alzheimer's disease [34]. Fyn interacts with both protein Tau and amyloid β-peptide, two key players responsible for the major pathologic hallmarks of Alzheimer's disease [35,36] and inhibitors of this kinase seem to be a promising novel approach therapy of this disease [18].
Tyrosinase, a copper-containing metalloenzyme, is a key enzyme involved in melanogenic processes [37]. In humans, melanin helps defend skin from the damage caused by UV light, however, excess levels of melanin can cause various dermatological disorders including hyperpigmentations, melisma, freckles, and age spots. Many tyrosinase inhibitors have been used in cosmetics and pharmaceutical products for the prevention of overproduction of melanin in the epidermis, however, side effects may occur following the chronic exposure to these compounds, so research of new tyrosinase inhibitors is still important for the development of new cosmeceuticals agents [38].
Elastase is a proteinase enzyme capable of degrading elastin, the main component of the elastic fibers responsible for the mechanical properties of connective tissue [39,40]. In the skin, the elastic fibers, together with the collagenous fibers, form a network under the epidermis and are responsible for skin elasticity. Therefore, elastase, by breaking down elastin, decreases the skin elasticity and increases skin aging, resulting in visible skin changes like wrinkles. So, elastase inhibitors are important cosmeceuticals agents by preventing loss of skin elasticity.
Proteasomes are protein complexes containing a common core, referred to as the 20S proteasome, that degrade unneeded or damaged proteins by proteolysis, this process is often called the ubiquitin-proteasome pathway [41]. This ubiquitin-proteasome pathway may be critical in cell cycle regulation, and due to these multiple functions, proteasome malfunctions are involved in a certain number of pathologies, in particular those linked to aging such as cancers and neurodegenerative diseases. Therefore, inhibitors of proteasome are highly sought for the treatment of these diseases [42].
Five compounds (1; 3-6) inhibited proteasome activity and two compounds (5-6) inhibited CDK7 and Fyn kinase (Table 3). Osirisyne B (5) was the most active compound with IC 50 on FynB kinase, CDK7 kinase, and proteasome inhibition of 18.44 µM, 9.13 µM, and 0.26 µM, respectively. The activity of these compounds seems to be related to the number of oxygen atoms. Osirisynes B (5) and E (6) are the two compounds with the lowest number of oxygens and presented inhibition activities on three enzymes, FYN kinase, proteasome, and CDK7. Furthermore, comparison between osirisyne B (5) and osirisyne E (6) highlighted the increase of the inhibition activity with the presence of a ketone instead of a hydroxyl function. In our continuing search for bioactive metabolites, studies on this family of compounds from sponges are still going on and a structure-activity relationship (SAR) study will be made when more similar compounds are isolated and tested.

General Experiment Procedures
Optical rotations were measured on an MCP 200 Anton Paar modular circular polarimeter at 25 • C (MeOH, c in g/100 mL) in a 10 × 5 mm i.d., 0.2 mL sample cell. 1 H and 13 C NMR data were acquired with a Bruker Avance III-600 MHz spectrometer equipped with a 5 mm TCI Cryoprobe in 1.7 mm o.d. capillary tube at 300 K. Chemical shifts were referenced using the corresponding solvent signals (δ H 3.31 and δ C 49.00 for CD 3 OD 99.96%-d). The spectra were processed using 1D and 2D NMR MNova software.
HRESIMS and MS/MS spectra were recorded using a Waters SYNAPT G2 HDMS mass spectrometer with an API source. For MS/MS, parameters of ionization were: Samples were solubilized in 300 µL of MeOH and then diluted to 1/10 in a solution of 1 mM CuSO 4 in MeOH.
The sponge was lyophilized with Cosmos −80 • C CRYOTEC and extracted with Dionex ASE 300. MPLC separations were carried out on Buchi Sepacore flash systems C-605/C-615/C-660 and a glass column (230 × 15 mm i.d.) packed with Macherey-Nagel MN Kieselgel silica gel (60−200 µm). Precoated TLC sheets of silica gel 60, Alugram SIL G/UV254, were used, and spots were visualized on the basis of the UV absorbance at 254 nm and by heating silica gel plates sprayed with formaldehyde−sulfuric acid or Dragendorff reagents. Analytical HPLC was carried out using a Phenomenex Gemini C 18 (150 × 4.6 mm i.d., 3 µm) column and was performed on a Thermo Scientific Dionex Ultimate 3000 system equipped with a photodiode array detector and a Corona detector with Chromeleon software. Semi-preparative HPLC was carried out using a Phenomenex Geminin C 18 (250 × 10 mm i.d., 5 µm) column and was performed on a Thermo Scientific Dionex Ultimate 3000 system equipped with a photodiode array detector. All solvents were analytical or HPLC grade and were used without further purification.

Catalase Activation Assays
Catalase activity was measured using The Amplex ® Red Catalase Assay Kit according to the manufacturer's instructions (Thermo fisher Scientific) [43]. Briefly, catalase first reacts with H 2 O 2 to produce water and oxygen (O 2 ). Next, the Amplex Red reagent reacts with any unreacted H 2 O 2 in the presence of horseradish peroxidase (HRP) to produce the highly fluorescent oxidation product, resorufin. The fluorescence was measured on a Polar Star Omega (BMG Labtech) plate reader.

Sirtuin 1 Activation Assays
Sirt1 activity was measured using the SIRT1 Fluorometric Drug Discovery Kit according to the manufacturer's instructions (Enzo Life Sciences). Briefly, this assay uses a small lysine-acetylated peptide, corresponding to K382 of human p53, as a substrate. The lysine residue is deacetylated by SIRT1, and this process is dependent on the addition of exogenous NAD + . The assay was carried out at 37 • C using Greiner white, small volume 384-well plates. First, 4 µL of substrate Fluor de Lys (final concentration 25 µM) were mixed with 4 µL of extract previously diluted 1/100 in assay buffer and 2 µL of the enzyme were added. After an incubation of 15 min at 37 • C, 10 µL of Developer 1X solution (composed by buffer, developer 5X, and Nicotinamide 50 mM) was added and incubated for 45 min at 37 • C.
Afterwards, the fluorescence was measured on a Polar Star Omega (BMG Labtech) plate reader. The fluorescence generated is proportional to the quantity of deacetylated Lysine (i.e., corresponding to Lysine 382). All measurements were performed in triplicate and the final DMSO concentration is 0.1 %. SIRT1 inhibitors nicotinamide (2 mM), suramin (100 µM), and sirtinol (100 µM) were used to confirm the specificity of the reaction. Calculation of net fluorescence included the subtraction of a blank consisting of buffer containing no NAD+ and was expressed as a percentage of control.
Then, 5 µL substrate/ATP mix was added (final concentration substrate 30 µM and ATP 125 µM) and the assay was incubated for another 2 h. Afterwards, 5 µL of the 15 µL assay reaction were transferred to new wells and 5 µL ADP Glo Reagent was added to terminate the kinase reaction and deplete the remaining ATP. After 40 min, 10 µL of Kinase Detection Reagent was added to convert ADP to ATP and allow the newly synthesized ATP to be measured using a luciferase/luciferin reaction. After another 40 min, the assay plates were measured in Luminescence mode on a Tecan M1000 plate reader.
The light generated, i.e., luminescent signal, is proportional to the ADP concentration produced and is correlated with kinase activity. The IC 50 was calculated using XLfit.
In general, it can be stated that the ADP-Glo Kinase Assay worked properly, as can be seen for the control compound Staurosporine, which was run in parallel to the assay.

Proteasome Inhibition Assays
The assay was carried out at room temperature (22 • C) using Corning 4514 black low volume 384-well plates. All measurements were performed in duplicate and the final DMSO concentration was 3.3%. The assay buffer contained 100 mM Tris pH 7.5 and 1 mM MgCl 2 . (Table 5). For the positive control, 9.5 µL protein dilution (final concentration 9 nM) were mixed with 0.5 µL of compound dilution ranging from 30-0.015 µM (assay end conc. 1000-0.5 nM). For the selected hits, 9.5 µL protein dilution (final concentration 9 nM) were mixed with 0.5 µL of compound dilution ranging from 1-0.0005 mg/mL (assay end conc. 33-0.016 µg/mL) and preincubated for 90 min. Then, 5 µL substrate mix was added (final concentration 5 µM) and incubated for 60 min. The assay plates were measured in Fluorescence mode on a Tecan M1000 plate reader (ex 380 nm, em 460 nm). The IC 50 was calculated using XLfit.
For assay set-up and validation, the tool compound ONX-0914 was utilized. For this tool compound, a dose-dependent decrease in the Fluorescence signal was observed similar to previous results.
Then, 5 µL substrate/ATP mix was added (final concentration substrate 10 µM and ATP 100 µM) and the assay was incubated for another 2 h. Afterwards, 5 µL of the 15 µL assay reaction were transferred to new wells and 5 µL ADP Glo Reagent was added to terminate the kinase reaction and deplete the remaining ATP. After 40 min, 10 µL of Kinase Detection Reagent was added to convert ADP to ATP and allow the newly synthesized ATP to be measured using a luciferase/luciferin reaction. After another 40 min, the assay plates were measured in Luminescence mode on a Tecan M1000 plate reader. The light generated, i.e., luminescent signal, is proportional to the ADP concentration produced and is correlated with kinase activity. The IC 50 was calculated using XLfit.
In general, it can be stated that the ADP-Glo Kinase Assay worked properly, as can be seen for the control compound Staurosporine which was run in parallel to the assay.

Tyrosinase Inhibition Assays
The anti-tyrosinase capacity of the extracts or compounds was assayed using an enzymatic method as previously described [45] with minor modifications. Briefly, in a 96-well microplate, 1/15 M PBS (pH 6.8), the tested samples, and 92 U/mL of mushroom tyrosinase (Sigma-Aldrich) were mixed and incubated for 10 min at room temperature, avoiding light exposure. Following the addition of 2.

Elastase Inhibition Assays
Elastase enzyme activity was evaluated using elastase from porcine pancreas (PPE) type IV and N-succinyl-Ala-Ala-Ala-p-nitroanilide as substrates, as previously described [46]. The amount of released p-nitroaniline, which was hydrolyzed by elastase, was measured spectrophotometrically at 405 nm. The reaction mix was constituted of 70 µL Trizma-base buffer (50 mM, pH 7.5), 10 µL of the extract tested (the final concentration of the extracts was 100 µg/mL), and 5 µL of PPE (0.4725 U/mL), in a 96-well microplate. The samples were incubated for 15 min at room temperature, avoiding light exposure. Subsequently, 15 µL from 0.903 mg/mL N-succinyl-Ala-Ala-Ala-p-nitroanilide were added and the samples were incubated at 37 • C for 30 min. Then, the absorbance of p-nitroaniline production was measured in the reader Infinite 200 PRO series (Tecan). Elastatinal was used as positive control. Experiments were performed in duplicate. The reagents of the assay were purchased from Sigma-Aldrich. The percentage of elastase inhibition was calculated as follows:  (1) where Abs control is the absorbance of the elastase in the Trizma base buffer, sample solvent, and substrate, and Abs sample is the absorbance of the elastase in the Trizma base buffer, extract or elastatinal, and substrate. Blank experiments were performed for each sample with all the reagents except the enzyme.

Conclusions
In conclusion, three new long-chain highly oxygenated polyacetylenes, osirisynes G-I (1-3), were isolated from Haliclona sp. together with three known long-chain highly oxygenated polyacetylenes osirisynes A (4), B (5), and E (6). The CH2Cl2-MeOH extract from Haliclona sp. sponge presented significant anti-tyrosinase activity (31.1%), however none of the Haliclona sp. extracts affected elastase activity significantly. Furthermore, five compounds (1; 3-6) showed interesting activities on three different biological assays: CDK7, proteasome, and Fyn kinase inhibition. Osirisyne B (5) was the most active compound with IC 50 on FYNB kinase, CDK7 kinase, and proteasome inhibition of 18.44 µM, 9.13 µM, and 0.26 µM, respectively. This molecule differed from osirisyne E (6) only by the presence of a ketone (C-19) instead of an allylic oxymethine and differed from osirisynes A (4), G (1), and H (2) by a lower number of oxymethines, so structure-activity relationship studies between osirisyne B (5) and these three enzymes could be undertaken in order to develop more selective and stable analogues for their therapeutic potential.