New Marine Antifouling Compounds from the Red Alga Laurencia sp.

Six new compounds, omaezol, intricatriol, hachijojimallenes A and B, debromoaplysinal, and 11,12-dihydro-3-hydroxyretinol have been isolated from four collections of Laurencia sp. These structures were determined by MS and NMR analyses. Their antifouling activities were evaluated together with eight previously known compounds isolated from the same samples. In particular, omaezol and hachijojimallene A showed potent activities (EC50 = 0.15–0.23 µg/mL) against larvae of the barnacle Amphibalanus amphitrite.


Introduction
Biofouling is a major cause of increased fuel consumption of ships and facilitates the spread of invasive marine organisms [1]. After many years of utilizing toxic antifoulants to control biofouling, the IMO (International Maritime Organization) banned the use of organotin compounds in 2008, paving the way for the development of environmentally friendly fouling-resistant coatings [2]. In addition, effective and new fouling treatment is needed to control and curb the introduction of non-indigenous species (NIS), that are introduced into local waters partly via ballast water uptake and discharge in vessels. The ballast water management convention will enter into force in September 2017. However, to effectively control NIS introduction, an effective antifouling treatment approach is still needed [3]. Search for natural antifouling compounds to solve biofouling has been "a work in progress" for almost four decades. Lately, a number of natural antifouling products have been reported and some of them have shown potent activities [4,5]. Furthermore, recent efforts to identify molecular mechanisms of antifouling compounds improve the possibility of their industrial importance [6]. As expected, industrial utilization of natural products as antifouling agents would warrant systematic synthetic efforts, and these have been implemented [5]. The total synthesis of 10-isocyano-4-cadinene [7] and dolastatin 16 [8] were achieved by our group [9,10]. The red alga Laurencia sp. is a rich source of biologically active secondary metabolites [11,12] and produces antifouling compounds such as elatol [13], omaezallene [14], and 2,10-dibromo-3-chloro-7-chamigrene [15]. As a continuous effort, our group has been searching for novel and potent antifouling compounds from Laurencia sp. by using barnacle larvae and epiphytic diatom assays. Hence, here we report the structures ( Figure 1) and activities of new antifouling compounds from Laurencia sp. importance [6]. As expected, industrial utilization of natural products as antifouling agents would warrant systematic synthetic efforts, and these have been implemented [5]. The total synthesis of 10isocyano-4-cadinene [7] and dolastatin 16 [8] were achieved by our group [9,10]. The red alga Laurencia sp. is a rich source of biologically active secondary metabolites [11,12] and produces antifouling compounds such as elatol [13], omaezallene [14], and 2,10-dibromo-3-chloro-7-chamigrene [15]. As a continuous effort, our group has been searching for novel and potent antifouling compounds from Laurencia sp. by using barnacle larvae and epiphytic diatom assays. Hence, here we report the structures ( Figure 1) and activities of new antifouling compounds from Laurencia sp.  (1) and Intricatriol (2) Laurencia sp., which was collected in Omaezaki, Japan, was extracted with MeOH. The extract yielded omaezol (1), intricatriol (2), in addition to the previously isolated omaezallenes and intricatetraol [14]. The molecular formula of 1 was determined to be C20H35BrO2 (m/z 368.1707, calcd. for C20H33BrO, 368.1709 [M − H2O] + ) by HR-EIMS (High Resolution-Electron Ionization Mass Spectrometry), suggesting three degrees of unsaturation. The existence of a hydroxy group was revealed by an IR absorption at 3386 cm −1 . 13 C NMR data (Table 1) showed the presence of one double bond, suggesting a bicyclic structure. COSY correlations connected C-1 to C-3. HMBC peaks from H-19 to C-1, 5, 9, 10 and from H-20 to C-3, 4, 5 concluded a cyclohexane ring ( Figure S1). Based on 13 C NMR chemical shifts, a bromine atom is attached to C-1 (δ 67.6) and a hydroxy group is connected to C-4 (δ71.4). A side chain was determined by COSY correlation (H-11/H-18 and H-12/H-13) and HMBC peaks (H-18/C-7, 12 and H-14/C- 13,16,17). Another hydroxy group is attached to C-7 (δ 74.6). Overlapping of NMR chemical shifts (C-6; δ 32.1, C-8; δ 31.8, C-9; δ 38.7, C-10; δ 39.4, H-6; δ 1.48, H-8; δ 1.49) hampered the positioning of the two remaining methylenes. However, HSQC-TOCSY peaks from H-8 to C-9 and H-6 to C-5 concluded the planar structure of 1. NOESY correlations determined relative configurations of the trans-decalin ( Figure 2).  (2) Laurencia sp., which was collected in Omaezaki, Japan, was extracted with MeOH. The extract yielded omaezol (1), intricatriol (2), in addition to the previously isolated omaezallenes and intricatetraol [14]. The molecular formula of 1 was determined to be C 20 (Table 1) showed the presence of one double bond, suggesting a bicyclic structure. COSY correlations connected C-1 to C-3. HMBC peaks from H-19 to C-1, 5, 9, 10 and from H-20 to C-3, 4, 5 concluded a cyclohexane ring ( Figure S1). Based on 13 C NMR chemical shifts, a bromine atom is attached to C-1 (δ 67.6) and a hydroxy group is connected to C-4 (δ71.4). A side chain was determined by COSY correlation (H-11/H-18 and H-12/H-13) and HMBC peaks (H-18/C-7, 12 and H-14/C- 13,16,17). Another hydroxy group is attached to C-7 (δ 74.6). Overlapping of NMR chemical shifts (C-6; δ 32.1, C-8; δ 31.8, C-9; δ 38.7, C-10; δ 39.4, H-6; δ 1.48, H-8; δ 1.49) hampered the positioning of the two remaining methylenes. However, HSQC-TOCSY peaks from H-8 to C-9 and H-6 to C-5 concluded the planar structure of 1. NOESY correlations determined relative configurations of the trans-decalin ( Figure 2).  (Table 1) were similar to intricatetraol [16,17], but were not symmetric. The 2D NMR interpretations ( Figure S2) showed that both compounds had the same carbon skeleton. However, 13 C chemical shift changes were observed (C-7: δ 84.1 in intricatetraol, δ 113.4 in 2, and C-11: δ 77.5 in intricatetraol, δ 84.7 in 2). C-7 in 2 was proposed to be an acetal carbon and C-11 was suggested to be a part of an ether instead of a secondary alcohol to make another fivemembered ring. In conclusion, the planar structure of 2 was determined and its configuration was assumed to be the same as intricatetraol [16,17], which was extracted from the same sample.

Hachijojimallenes A (3) and B (4)
Hachijojimallenes A (3) and B (4) were isolated together with N-methyl-2,3,6-tribromoindole [18] and pinnaterpene C [19] from a red alga, Laurencia sp., which was collected in Hachijojima Island, Japan. Based on HR-EIMS data, the molecular formula of 3 was determined to be C15H19Br3O3 (m/z 466.8851, calcd. for C15H18Br3O2, 466.8852 [M − OH] + ). The existence of a bromoallene moiety was suggested by an IR absorption at 1957 cm −1 , 13 C NMR chemical shifts (δ 74.1, 99.4, and 201.9), and 1 H NMR chemical shifts (δ 5.45 and 6.08) ( Table 2). COSY correlations established a carbon skeleton C-  (Table 1) were similar to intricatetraol [16,17], but were not symmetric. The 2D NMR interpretations ( Figure S2) showed that both compounds had the same carbon skeleton. However, 13 C chemical shift changes were observed (C-7: δ 84.1 in intricatetraol, δ 113.4 in 2, and C-11: δ 77.5 in intricatetraol, δ 84.7 in 2). C-7 in 2 was proposed to be an acetal carbon and C-11 was suggested to be a part of an ether instead of a secondary alcohol to make another five-membered ring. In conclusion, the planar structure of 2 was determined and its configuration was assumed to be the same as intricatetraol [16,17], which was extracted from the same sample.
The molecular formula of 4 was determined to be C 15 Table 2) was observed in C-9 (3; δ 102.1, 4; δ 119.7) and C-12 (3; δ 51.3, 4; δ 75.6). In addition, HMBC was observed from H-12 to C-9. In conclusion, the hemiacetal in 3 was replaced by acetal in 4 to form an ether between C-9 and C-12, which was debrominated at C-12. NOEs of 4 supported its relative configurations, similar to those of 3.

Debromoaplysinal (5)
Re-investigation of the red alga L. okamurae, collected at Oshoro Bay, Hokkaido, Japan, yielded debromoaplysinal (5). The molecular formula of 5 was determined to be C 15 Figure S4) clarified its planar structure, which is similar to aplysinal and debromoaplysinol. In fact, chemical shifts in the trisubstituted benzene of 5 are similar to those of debromoaplysinol, and those in the cyclopentane of 5 are similar to those of aplysinal. NOESY peaks between H-13 and H-14 suggested cis configuration of Me-13 and the aldehyde. Irradiation of H-3 by DPFGSE 1D NOE showed the enhancement of H-13. In conclusion, H-3, Me-13, and the aldehyde are located on the same face.

11,12-Dihydro-3-hydroxyretinol (6)
Re-investigation of the red alga L. nipponica collected at Muroran, Hokkaido, Japan, yielded 11,12-dihydro-3-hydroxyretinol (6), along with three known halogenated chamigrene-type sesquiterpenoids [20,21]. The molecular formula of 6 was deduced to be C 20 (Table 3) as well as the HSQC experiment of 6 showed the presence of eight sp 2 carbons, three vinyl methyls, a pair of geminal methyls, an oxymethylene, an oxymethine, four methylenes, and a quaternary carbon. These signals explained four degrees of unsaturation, implying that one ring was present in 6. COSY and HMBC data ( Figure S5) clearly indicated the gross structure of 6. The E configurations of double bonds were deduced from the 1 H-1 H coupling constants ( 3 J 7-8 15.9 Hz) and 13 C-NMR chemical shifts of vinyl methyls (C-19; δ 13.2, and C-20; δ 17.2). Unfortunately, the configuration at C-3 was not determined due to its limited amount.

Discussion
From a chemical point of view, 1 is a rare halo-diterpenoid, prenylated selinene-type compound like anhydroaplysiadiol [23]. Oxasqualenoids such as 2 have been proposed to be synthesized by the epoxide-opening cascades. These reactions could be catalyzed by vanadium-dependent bromoperoxidases [24]. However, we do not know how different cyclization patterns are controlled to produce 2 and the similar compound intricatetraol [16,17]. Although most of C15 acetogenins from Laurencia contain only ether rings, 3 and 4 contain a carbocyclic ring, similar to lembynes [25]. Compound 5 is one of common laurane-type sesquiterpenes. Retinols occur in nature only in animals and in the limited green algae Caulerpa sp. [26] This is the first record of a retinane-type diterpene (6)

Discussion
From a chemical point of view, 1 is a rare halo-diterpenoid, prenylated selinene-type compound like anhydroaplysiadiol [23]. Oxasqualenoids such as 2 have been proposed to be synthesized by the epoxide-opening cascades. These reactions could be catalyzed by vanadium-dependent bromoperoxidases [24]. However, we do not know how different cyclization patterns are controlled to produce 2 and the similar compound intricatetraol [16,17]. Although most of C 15 acetogenins from Laurencia contain only ether rings, 3 and 4 contain a carbocyclic ring, similar to lembynes [25]. Compound 5 is one of common laurane-type sesquiterpenes. Retinols occur in nature only in animals and in the limited green algae Caulerpa sp. [26] This is the first record of a retinane-type diterpene (6) isolated from Laurencia. Although the genus of Laurencia is one of the most studied organisms in marine natural product chemistry [11,12], recent isolation techniques yielded six new compounds in this study. Twelve compounds showed antifouling activities against the barnacle larvae. Notably, the activities of 1 and 3 were equivalent to CuSO 4 (EC 50 = 0.18 µg/mL). Three compounds showed antifouling activities against diatoms. However, most of compounds isolated from Laurencia sp. so far have not been tested for antifouling activity. Our results suggest that Laurencia is a potential source of antifouling compounds.

Plant Material
Algal samples of Laurencia sp. were collected at Omaezaki, Shizuoka Prefecture, and Hachijojima Island, Tokyo, Japan. L. okamurae was collected at Oshoro Bay and L. nipponica was collected at Muroran, Hokkaido, Japan. The voucher specimens are deposited in the Herbarium of Graduate School of Science, Hokkaido University.

Debromoaplysinal
The air-dried algae (100 g) were soaked in MeOH (3 L) for seven days. The MeOH solution was concentrated in vacuo, and the residue was partitioned between EtOAc and H 2 O. The EtOAc layer was then washed with water, dried over anhydrous Na 2 SO 4 , and evaporated in vacuo to obtain the crude extract (4.3 g). The extract (2.1 g) was fractionated by Si gel column chromatography with a step gradient (hexane and EtOAc). The fraction (1.4 g) eluted with hexane-EtOAc (3:1) was further subjected to two sets of Si gel column chromatography and PTLC with toluene to give debromoaplysinal (5) (0.9 mg). The other known compounds (laurinterol 8, 30.6 mg; isolaurinterol 9, 12.4 mg; debromolaurinterol 10, 11.0 mg; α-bromocuparene 11, 1.1 mg) were isolated from the same fraction (hexane-EtOAc 3:1) in a similar way.

Antifouling Assay
An antifouling assay against larvae of the barnacle Amphibalanus amphitrite was conducted according to the previous literature [9]. The antifouling assay against diatoms Nitzschia sp. and Cylindrotheca closterium was conducted as follows. The diatoms were maintained in test tubes under 16 h light and 8 h dark cycle conditions at 25 • C in modified Jorgensen's medium. Pure compounds were applied to a sample zone (16 mm in diameter) of cellulose TLC aluminum sheet (58 × 58 mm), and then each treated sheet was placed in Petri dish (15 × 90 mm in diameter). After that, 25-mL aliquots of modified Jorgensen's medium were introduced into Petri dishes with treated sheets, and inoculated with 1 mL of cultivated diatom suspension in equivalent cell densities. Petri dishes were sealed with Parafilm to ensure a closed system, and were incubated under the same conditions. Cell growth was estimated daily, and the adhesive condition was evaluated and compared to those of the control after seven days of incubation.