marine drugs

Article Fragilides U–W: New 11,20-Epoxybriaranes from the Sea Whip Gorgonian Coral Junceella fragilis

1,2, 3, 3 2 4 Tung-Pin Su †, Chien-Han Yuan †, Yi-Ming Jhu , Bo-Rong Peng , Zhi-Hong Wen , Yu-Jen Wu 5, Tung-Ying Wu 6,7,*, Hong-Wen Liu 8,* and Ping-Jyun Sung 1,2,4,9,10,*

1 Graduate Institute of Marine Biology, National Dong Hwa University, Pingtung 944, Taiwan; [email protected] 2 National Museum of Marine Biology and Aquarium, Pingtung 944, Taiwan; [email protected] 3 Department of Otolaryngology, Kaohsiung Armed General Hospital, Kaohsiung 802, Taiwan; [email protected] (C.-H.Y.); [email protected] (Y.-M.J.) 4 Department of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung 804, Taiwan; [email protected] 5 Department of Nursing, Meiho University, Pingtung 912, Taiwan; [email protected] 6 Department of Biological Science & Technology, Meiho University, Pingtung 912, Taiwan 7 Department of Food Science and Nutrition, Meiho University, Pingtung 912, Taiwan 8 Antai Medical Care Corporation Antai Tian-Sheng Memorial Hospital, Pingtung 928, Taiwan 9 Chinese Medicine Research and Development Center, China Medical University Hospital, Taichung 404, Taiwan 10 Graduate Institute of Natural Products, Kaohsiung Medical University, Kaohsiung 807, Taiwan * Correspondence: [email protected] (T.-Y.W.); [email protected] (H.-W.L.); [email protected] (P.-J.S.); Tel.: +886-8-779-9821 (ext. 8754) (T.-Y.W.); +886-8-882-5037 (P.-J.S.); Fax: +886-8-779-3281 (T.-Y.W.); +886-8-882-5087 (P.-J.S.) These authors contributed equally to this work. †


Received: 16 November 2019; Accepted: 13 December 2019; Published: 15 December 2019


Abstract: Three new 11,20-epoxybriaranes—fragilides U–W (1–3), as well as two known metabolites, junceellonoid D (4) and junceellin (5), were obtained from the octocoral Junceella fragilis. The structures of briaranes 1–3 were elucidated by spectroscopic methods and briaranes 3 and 5 displayed inhibition effects on inducible nitric oxide synthase (iNOS) release from RAW264.7.

Keywords: Junceella fragilis; fragilide; briarane; gorgonian; junceellonoid; junceellin; iNOS

1. Introduction Gorgonian corals belonging to the genus Junceella (family Ellisellidae) [1–3], distributed abundantly in the coral reefs of tropical Indo-Pacific Ocean have been found to produce briarane diterpenoids, natural products of a marine origin, in abundance [4]. In our further research into the natural products of Junceella fragilis (Ridley 1884) (Figure1), which was distributed extensively in the waters of Southern Taiwan, have resulted in the isolation of three new 11,20-epoxybriaranes– fragilides U–W (1–3) along with two know compounds junceellonoid D (4)[5] and junceellin (5)[6–15] (Figure1). An anti-inflammatory assay was employed to evaluate the activity of these isolates against the release of inducible nitric oxide synthase (iNOS) from macrophage cells RAW264.7.

Mar. Drugs 2019, 17, 706; doi:10.3390/md17120706 www.mdpi.com/journal/marinedrugs Mar. Drugs 2019, 17, 706 2 of 9 Mar. Drugs 2019, 17, x 2 of 9

1 2: R = OH, 6: R = H 3

4: R1 = R2 = OH, 5: R1 = R2 = OAc 7 Junceella fragilis

FigureFigure 1. StructuresStructures of of fragilides fragilides U–W U–W ( (11––33),), junceellonoid junceellonoid D D ( (44),), junceellin junceellin ( (55),), robustolide robustolide F F ( (66),), juncenolidejuncenolide M M (= (= frajunolidefrajunolide S) S) ( (77),), and and a a picture picture of of JunceellaJunceella fragilis. fragilis.

2.2. Results Results and and Discussion Discussion FragilideFragilide UU ( 1)( was1) was isolated isolated as an amorphousas an amorphous powder and powder displayed and a sodiated displayed pseudomolecular a sodiated pseudomolecularion at m/z 589.22583 ion inat them/z ( +589.22583)-HRESIMS, in the which (+)‐HRESIMS, suggested thatwhich the suggested molecular that formula the molecular of 1 was formulaC28H38O of12 (calcd.1 was forC28H C2838OH1238 (calcd.O12 + Na, for 589.22555) C28H38O12 (+Ω Na,= 10). 589.22555) The IR spectrum ( = 10). of The1 showed IR spectrum the presence of 1 1 1 1 showedof hydorxy the ( νpresencemax 3445 of cm hydorxy− ), γ-lactone (νmax ( ν3445max 1780cm–1), cm γ‐−lactone), and ester (νmax ( ν1780max 1733cm–1 cm), and− ) groups.ester (ν Analysismax 1733 cmof the–1) groups.1H, 13C Analysis NMR, and of distortionless the 1H, 13C NMR, enhancement and distortionless by polarization enhancement transfer by (DEPT) polarization spectra togethertransfer with(DEPT) the molecularspectra together formula, with suggested the molecular that there mustformula, be an suggested exchangeable that proton. there Themust13 Cbe NMR an exchangeablespectrum (Table proton.1), in combination The 13C NMR with DEPTspectrum and heteronuclear(Table 1), in single combination quantum coherencewith DEPT (HSQC) and heteronuclearspectra, revealed single the presencequantum of coherence four acetoxy (HSQC) groups spectra, (δC 21.7, revealed 21.2, 20.9, the 20.8, presence 4 CH 3of; δ Cfour170.1, acetoxy 169.8, × groups169.4, 169.0, (C 21.7, 4 C),21.2, a γ20.9,-lactone 20.8, moiety 4  CH (3δ; CC175.9), 170.1, and 169.8, a trisubstituted 169.4, 169.0, 4 olefin  C), ( δaC γ‐142.0,lactone C; moiety 120.6, CH). (C 13 × 175.9),Base on and the a trisubstitutedC NMR data andolefin numbers (C 142.0, of C; unsaturation, 120.6, CH).1 Basewas on established the 13C NMR as a diterpenoiddata and numbers featuring of unsaturation,with four rings. 1 Anwas exocyclic established epoxy as group a diterpenoid was deduced featuring from thewith signals four ofrings. an oxygenated An exocyclic quaternary epoxy groupcarbon was and deduced an oxymethylene from the atsignalsδC 62.5 of and an oxygenated 59.1, respectively, quaternary and further carbon supported and an oxymethylene by the chemical at shiftsC 62.5 of and oxymethylene 59.1, respectively, protons and at δ Hfurther2.85 (1H, supported d, J = 4.4 by Hz) the and chemical 2.98 (1H, shifts dd, ofJ = oxymethylene4.4, 1.6 Hz). Moreover, protons ata methylH 2.85 singlet (1H, d, ( δJH = 1.09,4.4 Hz) 3H, and s), a 2.98 methyl (1H, doublet dd, J = ( δ4.4,H 1.15, 1.6 Hz). 3H, d,Moreover,J = 7.6 Hz), a methyl a vinyl singlet methyl ( (δHH 1.09,2.00, 3H,3H, s), d, Ja =methyl1.6 Hz), doublet three pairs(H 1.15, of aliphatic 3H, d, J methylene= 7.6 Hz), a protons vinyl methyl (δH 2.04, (H 1H, 2.00, m; 3H, 2.62, d, 1H,J = 1.6 ddd, Hz),J = three18.4, pairs4.0, 2.4 of Hz; aliphatic 1.13, 1H, methylene m; 2.30, protons 1H, m; 2.11, (H 2.04, 1H, m;1H, 1.81, m; 2.62, 1H, m), 1H, two ddd, aliphatic J = 18.4, methine 4.0, 2.4 Hz; protons 1.13, ( δ1H,H 2.36, m; 2.30,1H, dd, 1H,J m;= 5.6, 2.11, 1.6 1H, Hz; m; 2.33, 1.81, 1H, 1H, q, m),J = 7.6two Hz), aliphatic five oxymethine methine protons protons (H ( δ2.36,H 5.91, 1H, 1H, dd, br J = s; 5.6, 5.65, 1.6 1H, Hz; d, 2.33,J = 5.6 1H, Hz; q, 5.04,J = 7.6 1H, Hz), d, Jfive= 8.8 oxymethine Hz; 5.03, 1H, protons br s; 4.71,(H 5.91, 1H, 1H, d, J =br4.8 s; 5.65, Hz), 1H, an olefin d, J = proton5.6 Hz; ( δ5.04,H 5.67, 1H, 1H, d, Jdq, = 8.8J = Hz;8.8, 5.03, 1.6 Hz), 1H, four br s; acetate 4.71, 1H, methyls d, J = ( δ4.8H 2.24,Hz), 2.01,an olefin 1.99, 1.98,proton each (H 3H 5.67, s),1H, and dq, a hydroxyJ = 8.8, 1.6 proton Hz), 1 × four(δH 4.85, acetate 1H, methyls br s) were (H observed 2.24, 2.01, in 1.99, the 1.98,H NMR each spectrum 3H  s), (Tableand a 2hydroxy). proton (H 4.85, 1H, br s) were observed in the 1H NMR spectrum (Table 2).

Table 1. 13C NMR data for briaranes 1–3.

Position 1 a 2 a 3 b 1 47.0, C c 46.1, C 49.1, C 2 72.2, CH 72.1, CH 75.4, CH 3 37.5, CH2 37.6, CH2 136.4, CH

Mar. Drugs 2019, 17, 706 3 of 9

Table 1. 13C NMR data for briaranes 1–3.

Position 1 a 2 a 3 b 1 47.0, C c 46.1, C 49.1, C 2 72.2, CH 72.1, CH 75.4, CH 3 37.5, CH2 37.6, CH2 136.4, CH 4 70.6, CH 73.4, CH 128.2, CH 5 142.0, C 144.1, C 135.9, C 6 120.6, CH 58.0, CH 128.8, CH 7 76.5, CH 79.0, CH 80.0, CH 8 80.4, C 81.9, C 80.9, C 9 67.4, CH 68.9, CH 67.1, CH 10 39.9, CH 39.8, CH 39.5, CH 11 62.5, C 62.6, C 60.8, C 12 23.6, CH2 23.5, CH2 29.4, CH2 13 24.2, CH2 24.0, CH2 25.5, CH2 14 73.2, CH 73.5, CH 78.9, CH 15 14.6, CH3 15.4, CH3 16.0, CH3 16 21.4, CH3 121.0, CH2 48.0, CH2 17 42.4, CH 43.8, CH 45.2, CH 18 6.6, CH3 7.1, CH3 6.9, CH3 19 175.9, C 175.2, C 174.8, C 20 59.1, CH2 58.2, CH2 50.4, CH2 Acetate methyls 21.7, CH3 21.9, CH3 21.7, CH3 21.2, CH3 21.0, CH3 21.4, CH3 20.9, CH3 21.0, CH3 21.3, CH3 20.8, CH3 Acetate carbonyls 170.1, C 171.0, C 170.4, C 169.8, C 170.1, C 169.7, C 169.4, C 169.7, C 169.3, C 169.0, C a b c Spectra measured at 100 MHz in CDCl3. Spectra measured at 150 MHz in CDCl3. Multiplicity deduced by distortionless enhancement by polarization transfer (DEPT) and heteronuclear single quantum coherence (HSQC) spectra.

Table 2. 1H NMR data (J in Hz) for briaranes 1–3.

Position 1 a 2 a 3 b 2 5.03 br s 4.96 dd (2.8, 2.8) 5.63 d (6.0) 3α/β 2.04 m; 2.62 ddd (18.4, 4.0, 2.4) 1.75 m; 2.38 m 6.04 dd (17.4, 6.0) 4 5.91 br s 4.85 m 6.42 d (17.4) 6 5.67 dq (8.8, 1.6) 5.49 br s 5.66 d (3.6) 7 5.04 d (8.8) 5.15 d (2.4) 5.62 d (3.6) 9 5.65 d (5.6) 5.52 d (6.4) 4.77 d (6.6) 10 2.36 dd (5.6, 1.6) 2.31 d (6.4) 3.32 d (6.6) 12α/β 1.13 m; 2.30 m 1.15 m; 2.30 m 1.29 m; 2.18 m 13α/β 2.11 m; 1.81 m 2.14 m; 1.78 m 2.08 m; 1.85 dddd (12.0, 12.0, 3.6, 1.8) 14 4.71 d (4.8) 4.86 d (4.4) 4.94 dd (3.6, 3.0) 15 1.09 s 1.13 s 0.87 s 16a/b 2.00 d (1.6) 5.74 s; 5.97 s 4.13 d (12.0); 4.19 d (12.0) 17 2.33 q (7.6) 2.29 q (7.6) 2.36 q (7.2) 18 1.15 d (7.6) 1.22 d (7.6) 1.18 d (7.2) 20a/b 2.85 d (4.4); 2.98 dd (4.4, 1.6) 2.83 d (4.0); 2.85 br d (4.0) 2.76 d (2.4); 3.40 dd (2.4, 2.4) OH-8 4.85 br s 4.63 br s 3.13 d (1.2) Acetate methyls 2.24 s 2.23 s 2.27 s 2.01 s 2.03 s 2.25 s 1.99 s 1.99 s 2.11 s 1.98 s a b Spectra measured at 400 MHz in CDCl3. Spectra measured at 600 MHz in CDCl3. Mar. Drugs 2019, 17, 706 4 of 9

Coupling information in the correlation spectroscopy (COSY) analysis enabled the proton sequences from H-2/H2-3/H-4, H-6/H-7, H-9/H-10, H2-12/H2-13/H-14, and H-17/H3-18 (Figure2), Mar. DrugsMar. Drugs Mar.2019Mar. , DrugsDrugs201917, x, 17 20192019, x ,, 1717,, xx 4 of 94 of 94 4 ofof 99 which was assembled with a heteronuclear multiple bond correlation (HMBC) experiment (Figure2). The HMBC between protons and quaternary carbons, such as H-2, H-10, H-13β, H-14, H3-15/C-1; H‐7, HH‐37,‐16/C HH‐‐37,‐7,16/C‐ 5; HH 3H3‐‐‐16/C16/C‐5;9, HH‐‐5;‐5;9,10, HH ‐H‐9,9,10,3 ‐H18/CH H‐‐10,10,3‐18/C‐ 8; HH 3H3‐‐‐18/C18/C‐8;9, HH‐‐8;‐8;9,10, HH ‐H‐9,9,10,2 ‐H12,H H‐‐10,10, 2H‐12, ‐H13H 2H2‐α‐12,12,,‐ 13H Hα2H‐,20/C‐ ‐13H13α2α‐20/C,‐, 11; HH2 2‐‐and‐20/C20/C11; andH‐‐11;11;‐17, Handand H‐17,3 ‐H18/CH H‐‐17,17,3‐18/C‐ 19, HH3 3‐‐‐18/C18/C19, ‐‐19,19, H-7, H -16/C-5; H-9, H-10, H -18/C-8; H-9, H-10, H -12, H-13α,H -20/C-11; and H-17, H -18/C-19, permittedpermittedpermittedpermitted elucidation elucidation 3 elucidationelucidation of the of carbonthe ofof carbon thethe skeleton carboncarbon 3skeleton skeletonofskeleton 1 .of A 1 vinyl. ofofA 11vinyl. . methylAA 2vinylvinyl methyl at methylmethyl C ‐at5 wasC ‐atat25 wasCconfirmedC‐‐55 wasconfirmedwas confirmedconfirmed by the by the by3by thethe permitted elucidation of the carbon skeleton of 1. A vinyl methyl at C-5 was confirmed by the HMBC HMBCHMBC betweenHMBCHMBC between betweenHbetween3‐16/C H3‐16/C‐ 4, HH 3C3‐‐‐16/C‐16/C4,5, CC‐‐‐‐4,5,4,6 CandC‐‐‐5,65, andHCC‐‐66/C6 Handand‐‐16;6/C HH and‐‐16;‐6/C6/C andfurther‐‐16;16; furtherandand supporting furtherfurther supporting supportingsupporting by an by allylic an byby allylic an ancoupling allylicallylic coupling couplingcoupling between H3-16/C-4, C-5, C-6 and H-6/C-16; and further supporting by an allylic coupling between betweenbetween betweenHbetween‐6 Hand‐6 andHH‐3‐6‐616 Handand 3(‐J16 =HH 1.6(33J‐‐ 1616= Hz). 1.6((JJ = = Hz). The1.61.6 Hz).Hz).Themethyl methylTheThe group methylmethyl group on group groupC on‐1 (MeC on‐on1 ‐ (Me15)CC‐‐11 ‐was (Me15)(Me was‐substantiated‐15)15) wassubstantiatedwas substantiatedsubstantiated by the by the byby thethe H-6 and H3-16 (J = 1.6 Hz). The methyl group on C-1 (Me-15) was substantiated by the HMBC from HMBCHMBC fromHMBCHMBC from H3‐ 15/C from fromH3‐15/C‐ 1, HH C33‐‐‐15/C15/C1,2, C‐‐2,1,10,1, CC C‐‐10,2,2,‐14; CC C‐ ‐10,and‐10,14; C CHand‐‐14;14;‐2, H andHand‐‐2,10, HH H‐‐‐2,10,2,‐ 14/C HH H‐‐10,10,‐14/C‐15. HH ‐The‐‐14/C14/C15. epoxyThe‐‐15.15. epoxy TheThe group epoxyepoxy group at groupCgroup ‐at11/20 C ‐ at11/20at was CC‐‐11/2011/20 was waswas H3-15/C-1, C-2, C-10, C-14; and H-2, H-10, H-14/C-15. The epoxy group at C-11/20 was confirmed by the confirmedconfirmedconfirmed by the by HMBCthe by HMBCthe between HMBC between Hbetween2‐20/C H2‐20/C‐ 10,H2 2‐C‐20/C10,‐11; C‐ 10,and‐11; C andH‐11;‐20a/C Hand‐20a/C‐ 12;H‐20a/C and‐12; andfurther‐12; furtherand supporting further supporting supporting by a by a by a confirmed by the HMBC between H ‐20/C‐10, C‐11; and H‐20a/C‐12; and further supporting4 by1 a1 4HMBC1 4 11 between4 1 1 1 H2-20/C-10, C-11; and H-20a/C-12; and further supporting by a long range J- H– H long longrange longlongrange J‐ H–rangerange J‐HH– correlation4JJ‐‐H1H–H– correlation1HH correlationcorrelation between between Hbetweenbetween‐10 H (‐10H H2.36) H(‐‐10H10 2.36) and((HH 2.36) 2.36)Hand‐20b Handand ‐(20b H HH 2.98) ‐(‐20b20bH 2.98) (((J H=H 2.98) 1.62.98)(J = Hz). 1.6 ((JJ == Hz).A 1.61.6 hydroxy Hz).AHz). hydroxy AA hydroxyhydroxy correlation between H-10 (δH 2.36) and H-20b (δH 2.98) (J = 1.6 Hz). A hydroxy group attaching at C-8 groupgroup attachinggroupgroup attaching attaching attachingat C ‐at8 wasC ‐ at8at was toCC‐ ‐8infer8 wastowas infer that toto infer inferthatan HMBC anthatthat HMBC anan of HMBCHMBC a ofhydroxy a hydroxyofof aa protonhydroxyhydroxy proton at proton protonH at 4.85 H at at4.85to  HCH ‐to4.854.857, CC ‐ ‐to7,to8, CandC‐‐7,8,7, CandC‐‐8,8, andand was to infer that an HMBC of a hydroxy proton at δH 4.85 to C-7, C-8, and C-9. Moreover, HMBC from C‐9. CMoreover,‐9. CMoreover,C‐‐9.9. Moreover,Moreover, HMBC HMBC from HMBCHMBC from the fromoxymethine fromthe oxymethine thethe oxymethineoxymethine protons protons at protonsprotons H at 5.03 H at at5.03(H ‐HH2), (H5.035.03 5.65‐2), (H(H 5.65(H‐‐2),2),‐9), (H5.655.65 and‐9), (H(H 4.71and‐‐9),9), 4.71 (Handand‐14) (H4.714.71 to‐14) (H(H to‐‐14)14) toto the oxymethine protons at δH 5.03 (H-2), 5.65 (H-9), and 4.71 (H-14) to the acetate carbonyls at δC 169.0, the acetatethe acetatethethe carbonyls acetateacetate carbonyls carbonylscarbonyls at C at 169.0, C atat169.0, 169.4,CC 169.0,169.0, 169.4, and 169.4,169.4, 170.1,and 170.1,andand placed 170.1,170.1, placed the placedplaced acetoxy the acetoxy thethe groups acetoxyacetoxy groups on groupsgroups C on‐2, C on‐‐on2,9, CandC‐‐2,9,2, CandC‐‐14,9,9, Candand ‐14, CC ‐‐14,14, 169.4, and 170.1, placed the acetoxy groups on C-2, C-9, and C-14, respectively. Ten of the 12 oxygen respectively.respectively.respectively.respectively. Ten ofTen the ofTenTen 12the ofoxygenof 12 thethe oxygen 1212 atoms oxygenoxygen atoms in atomstheatoms in molecularthe inin molecular thethe molecularmolecular formula formula of formulaformula 1 ofcould 1 couldofof be 11 couldaccountedcould be accounted bebe accountedaccounted for the for the forfor thethe atoms in the molecular formula of 1 could be accounted for the presence of a γ-lactone, three esters, presencepresence presenceofpresence a γ‐of lactone,a γ‐ ofof lactone, aa γ‐ γ‐ threelactone,lactone, three esters, three threeesters, an esters, esters,epoxide, an epoxide, anan epoxide,andepoxide, anda hydroxy aandand hydroxy aa hydroxygroup.hydroxy group. Thus, group.group. Thus, the Thus, Thus, remainingthe remaining thethe remainingremaining two two twotwo an epoxide, and a hydroxy group. Thus, the remaining two oxygen atoms had1 to1 be13 positioned13 at C-4 oxygenoxygen atomsoxygen atoms had atoms tohad be tohad positioned be to positioned be positioned at C ‐at4 asC ‐ at4an asC acetoxy ‐an4 as acetoxy an group, acetoxy group, as group, indicated as indicated as indicated by its by 1H its byand H its 13and C1H NMR andC NMR 13C NMR oxygen atoms had to be positioned at C‐4 as1 an acetoxy13 group, as indicated by its H and C NMR as an acetoxy group, as indicated by its H and C NMR chemical shifts (δH 5.91, 1H, br s; δC 70.6, chemicalchemical chemicalshiftschemical shifts (H shiftsshifts5.91, (H 5.91, 1H,((HH 5.91,br5.91,1H, s; br 1H,1H,C s;70.6, brbrC s;70.6, s;CH), CC 70.6,70.6,CH), although CH), CH),although althoughalthoughno HMBC no HMBC nono was HMBCHMBC wasobserved wasobservedwas observedobservedfrom from H‐4 from fromtoH‐ 4any toHH ‐‐any44 toto anyany CH), although no HMBC was observed from H-4 to any acetate carbonyl. acetateacetate carbonyl.acetateacetate carbonyl. carbonyl. carbonyl.

FigureFigure 2. TheFigureFigure 2.Figure correlationThe 2.2. correlation TheThe 2. The correlationcorrelation spectroscopy correlation spectroscopy spectroscopyspectroscopy spectroscopy (COSY) (COSY) ( (COSY)(COSY) (COSY)) (correlations,) (correlations,( )) correlations, correlations, selective selective heteronuclear selective selectiveselective heteronuclear heteronuclear heteronuclearheteronuclear multiple multiple multiple multiplemultiple bond bondbond correlation bondbondcorrelationcorrelation correlationcorrelation (HMBC) (HMBC) (HMBC) experiment(HMBC)(HMBC) experiment experiment experimentexperiment ( ),( ( and ),(( andandprotons),), protons protonsandand with protonsprotons with withkey key withwithnuclearkey nuclear nuclearkeykey Overhauser Overhausernuclearnuclear Overhauser OverhauserOverhauser effect effect effect spectroscopy effect effect spectroscopyspectroscopyspectroscopyspectroscopy (NOESY) (NOESY) ( (NOESY)(NOESY) () correlations ()( correlations)) correlationscorrelations of 1. of 1. ofof 11..

13 BasedBased onBased Baseda on summaryBased a ononsummary onaa summary summary aof summary the of 13theC ofof chemical 13 the oftheC the chemical 1313CC chemicalchemicalshiftsC chemical shifts of 11,20shiftsshifts of shifts 11,20‐ epoxyofof 11,20 of‐11,20epoxy 11,20-epoxygroup‐‐epoxyepoxy group in groupgroupnaturally in group naturally inin naturally innaturallyoccurring naturally occurring occurring occurring occurring 13 briarane analogues,13 13 with13 C NMR data for C-11 and C-20 atC δ 62–63 and 58–60 ppm, the epoxide briaranebriarane analogues,briaranebriarane analogues, analogues,analogues, with with C NMR withwithC NMR 13dataCC NMRNMR datafor C datafordata‐11 C and forfor‐11 CCand‐‐‐111120 C andandat‐20  C Cat 62–63‐‐20 20C atat62–63 andC 62–6362–63 Cand58–60 58–60andand ppm, 58–6058–60 ppm, the ppm,ppm, epoxidethe epoxide thethe epoxideepoxide groupgroup wasgroupgroup β‐wasgrouporiented β‐ waswasoriented was β‐ β‐ andorientedorientedβ-oriented andthe cyclohexane theandand cyclohexane and thethe cyclohexanecyclohexane the ring cyclohexane ringexisted ringexistedring in existed ringexisted a intwist existeda twistin inboat aa twist twist boat inconformation a boat twistconformationboat conformationconformation boat [16]; conformation [16];thus, [16];thus,[16]; the thus, thus,the [16 ]; thethe thus, configurationconfigurationconfigurationconfigurationthe of configuration the of 11,20the ofof 11,20 ‐thetheepoxy of11,20‐11,20epoxy the group‐‐epoxy 11,20-epoxyepoxy group in groupgroup1 inshould 1 group inshouldin 11be shouldshould in β‐ 1beorientedshould β‐ bebeoriented β‐ β‐ beorientedandorientedβ-oriented andthe cyclohexaneandtheand cyclohexane and thethe thecyclohexanecyclohexane cyclohexanering ringwas ringwasring ring waswas was 13 found to be in a twist boat conformation13 13 for the13 C chemical shifts atC δ 62.5 (C-11) and 59.1 (CH2 - foundfound to foundbefound to in be a to totwistin be bea twistin inboat aa twist twist boatconformation boatconformationboat conformationconformation for the for the C forfor chemical thetheC chemical 13CC chemicalshiftschemical shifts at  shiftsshiftsC at 62.5 C atat62.5(C ‐11)C (C62.562.5C and‐11) (C(C 59.1and‐‐11)11) 59.1(CHandand 2 (CH59.1‐ 59.1 2 (CH‐ (CH2‐ ‐ 2 20). The20). relativeThe20).20).20). relativeTheThe Thestereochemistry relativerelative relativestereochemistry stereochemistrystereochemistry stereochemistry of 1 ofwas 1 wasofestablishedof 11 of wasestablishedwas1 was establishedestablished by established the by analysisthe byby analysis the bythe theof analysisanalysis correlations analysis of correlations ofof correlationscorrelations of correlationsobserved observed observedobservedin observeda in a inin aa in a nuclearnuclear Overhausernuclearnuclear nuclearOverhauser OverhauserOverhauser Overhausereffect effect spectroscopy effecteffect spectroscopy effect spectroscopyspectroscopy spectroscopy (NOESY) (NOESY) (NOESY) (NOESY)experiment (NOESY) experiment experimentexperiment experiment(Figure (Figure 2). (Figure(Figure In (Figure 2). the In 2).2).NOESY the2). InIn InNOESY thethe the spectrum NOESYNOESY NOESY spectrum spectrum spectrumof spectrum of of of 1, 1, NOE1, NOE correlations11,, NOE NOEcorrelations correlations correlations between between between betweenHbetween‐10/H H‐10/H‐ H-102, HH H‐‐10/H‐10/H‐2,/10/HH-2, H‐‐2,2,10/H‐ H-109, HH and‐‐10/H‐10/H/9,H-9, andH‐‐‐9, and9,10/OH Handand‐ H-1010/OH H‐H8,‐/‐ OH-8,10/OH10/OHwhile‐8, while while‐‐no8,8, whilewhilecorrelation no no correlation correlation nono correlationcorrelation was waswasseen seen was seenwas between seenseen H-10 and H -15, suggesting3 that these protons H-2, H-9, and H-10, and the hydroxy group at C-8 were betweenbetween Hbetweenbetween‐10 H and‐10 H HHand‐‐31010‐15, H and3and 3suggesting‐15, HH suggesting3‐‐15,15, suggestingsuggesting that thatthese thesethatthat protons thesethese protons H protonsprotons‐2, HH‐‐2,9, H Hand‐‐‐2,9,2, HandH‐‐‐9,10,9, H andand and‐10, HH theand‐‐10,10, hydroxy the andand hydroxy thethe grouphydroxyhydroxy group groupgroup α-oriented; meanwhile, an NOE correlation of H -15 with3 H-14 indicated that H-14 was β-oriented. at C‐at8 wereC‐at8at were α‐CC‐‐8oriented;8 were α‐wereoriented; α‐ α‐ oriented;meanwhile,oriented; meanwhile, meanwhile,meanwhile, an NOE an NOE correlation anan NOE NOEcorrelation correlationcorrelation of H 3of3‐15 H with3 ofof‐15 HH with H3‐‐1515‐14 withwithH indicated‐14 H Hindicated‐‐1414 indicated indicatedthat thatH‐14 Hthat thatwas‐14 H Hwas‐‐1414 waswas The NOESY spectrum showed a correlation from H-6 to H -16, revealing3 the Z geometry of C-5/6 β‐oriented.β‐oriented.β‐β‐ oriented.Theoriented. NOESYThe NOESY TheThe spectrum NOESYNOESY spectrum spectrum spectrumshowed showed a showed showedcorrelation a correlation aa correlationcorrelation from from H‐6 from fromtoH ‐H6 3to H‐H16, ‐3H‐66 3 revealingto‐to16, HH revealing3‐‐16,16, revealing revealingthe Z the geometry Z thethe geometry ZZ geometry geometry double bond. H -18 exhibited3 NOE correlations to OH-8 and H-9, suggesting the α-orientation of of Cof‐5/6 C of‐ofdouble5/6 CC ‐‐double5/65/6 bond.doubledouble bond. H 3bond.bond.‐ 318H 3exhibited‐ 18 HH 3exhibited‐‐1818 exhibited exhibitedNOE NOE correlations NOE NOEcorrelations correlationscorrelations to OHto ‐8 OH totoand ‐ 8OHOH andH‐‐88‐ 9, and andHsuggesting‐9, HHsuggesting‐‐9,9, suggestingsuggesting the the thethe α‐orientationα‐orientationα‐α‐orientationorientationMe-18 of Me of at‐ 18Me C-17. ofofat‐18 MeCMe H-7 ‐at17.‐‐18 18C displayedH ‐ at17.at‐ 7 CC displayedH‐‐17.17.‐7 displayedH NOEH‐‐77 displayeddisplayed correlationsNOE NOE correlations NOE NOEcorrelations with correlationscorrelations H-4 with and with H H-17,‐4 withwithandH‐4 which andHH‐‐‐17,44 Handand further which‐17, HH which‐‐17,17, confirmedfurther whichwhich further furtherfurther that these confirmedconfirmedconfirmedconfirmed thatthree thatthese protons thesethatthat three thesethese three were protons three threeprotons in β were -orientation protonsprotons were in β‐ were wereinorientation β‐ at inorientationin C-7, β‐ β‐orientationorientation C-4, at C and ‐at7, C C-17. ‐‐ at7,at4, CandC Based‐‐7,4,7, CandC‐‐4,17.4, on Candand Based‐ the17. CC aboveBased‐‐17. 17.on BasedBasedthe findings,on abovethe onon above thethe the aboveabove relative findings,findings, findings,thefindings,configurations relativethe relative thethe configurations relativerelative configurations of stereogenic configurationsconfigurations of stereogenic of carbons stereogenic ofof stereogenicstereogenic of carbons1 were carbons elucidatedof carbonscarbons 1 ofwere 1 wereof ofelucidated as 11 1were wereelucidatedR*,2 S elucidatedelucidated*,4 asS *,71 RasS*,2*, 1SR 8 as*,4asR*,2*,9 S11S*,7RR*,4S*,2*,10*,2SS*,S*,7S *,4S*,4*,11SS*,S*,7*,7 SS*,14S*,*, S* 8R*,98SR*,10*,988SSRR*,10*,11*,9*,9andSSS*,10*,11*,10*,14 17RSSS*,11*,14*.*,11* and However,SS*,14**,14 and17SSR** *. and17and However,R as*. 1717 briaranesHowever,RR*.*. However,However, as briaranes1 –as4 briaraneswere asas briaranes briaranes1 isolated–4 were1–4 were along 11isolated––44 were wereisolated with alongisolatedisolated the along knownwith alongalong with the chlorinated withknownwiththe known thethe knownknown briarane, chlorinatedchlorinatedchlorinatedchlorinated briarane, briarane, briarane,briarane,junceellin junceellin junceellinjunceellin (5) [6], (5) and[6], ((55) ) and the[6],[6], structure,andtheand structure, thethe structure, structure,including including including includingthe absolutethe absolute thethe absoluteconfigurationabsolute configuration configurationconfiguration of of ofof junceellinjunceellinjunceellinjunceellin (5) was(5) was further((55)) waswasfurther confirmed furtherfurther confirmed confirmedconfirmed by aby single a by bysingle ‐ crystalaa singlesingle‐crystal X‐‐crystal‐crystalray X ‐diffractionray XX ‐‐diffractionrayray diffractiondiffraction analysis analysis analysis[7,15].analysis [7,15]. It [7,15].[7,15].is It is ItIt isis reasonable,reasonable,reasonable,reasonable, therefore, therefore, therefore, therefore,on biogenetic on biogenetic onon biogeneticbiogenetic grounds grounds togroundsgrounds assume to assume toto that assumeassume thatbriaranes briaranesthatthat briaranes briaranes1–4 have1–4 have 11the––44 have samehavethe same thetheabsolute samesame absolute absoluteabsolute configurationconfigurationconfigurationconfiguration as that as thatof as as5 . thatofthatTherefore, 5 . ofofTherefore, 55.. Therefore,Therefore, the configurationsthe configurations thethe configurationsconfigurations of the of stereogenicthe ofof stereogenic thethe stereogenicstereogenic carbons carbons of carbonscarbons 1 ofshould 1 ofshouldof 11be shouldshould be bebe elucidatedelucidatedelucidatedelucidated as 1R as,2 S1,4R asas,2S,7 S 11,4SRR,8S,2,2,7RSS,9S,4,4,8SS,10R,7,7,9SSS,8S,8,11,10RR,9,9SSS,14S,11,10,10SSS S,14and,11,11SSS 17and,14,14RSS (Supplementary17 andandR (Supplementary 1717RR (Supplementary(Supplementary Materials, Materials, Materials,Materials, Figures Figures S1–S8). FiguresFigures S1–S8). S1–S8).S1–S8).

Mar. Drugs 2019, 17, 706 5 of 9

Mar. Drugs 2019, 17, x 4 of 9 junceellin (5)[Mar.6], and Drugs the 2019 structure,, 17, x including the absolute configuration of junceellin (5) was further 4 of 9 Mar. Drugs 2019, 17, x confirmed by a single-crystal X-ray diffraction analysis [7,15]. It isH reasonable,‐7, H43 ‐of16/C 9 ‐ therefore,5; H‐9, H on‐10, biogenetic H3‐18/C‐8; H‐9, H‐10, H2‐12, H‐13α, H2‐20/C‐11; and H‐17, H3‐18/C‐19, grounds to assumeH‐7, H that3‐16/C briaranes‐5; H‐9, 1H–4‐10,have H3 the‐18/C same‐8; H absolute‐9, H‐10, configuration permittedH2‐12, H‐ 13elucidationα as, H that2‐20/C of of‐511;. the Therefore, and carbon H‐17, skeletonH3‐18/C ‐of19, 1. A vinyl methyl at C‐5 was confirmed by the H‐7, H3‐16/C‐5; H‐9, H‐10,the H configurations3‐18/C‐8; permittedH‐9, of H the‐10, stereogenic elucidationH2‐12, H‐13 carbons ofα, theH2‐ 20/C ofcarbon1 should‐11; skeletonand be H elucidated‐17, of HHMBC1.3 ‐18/CA as vinyl 1‐betweenR19,,2 Smethyl,4S,7 SH,8 3‐atR16/C,9 CS,10‐‐54, Swas ,11C‐5,S confirmed,14 C‐S6 and H by‐6/C the‐16; and further supporting by an allylic coupling Mar. Drugs 2019, 17, x 5 of 9 permitted elucidation of andthe 17carbonR (Supplementary skeletonHMBC betweenof 1 Materials,. A vinylH3‐16/C Figuresmethyl‐4, C S1–S8).‐5,at CC‐‐65 andwas Hconfirmed‐6/C‐16;between and by furtherthe H ‐6 and supporting H3‐16 (J =by 1.6 an Hz). allylic The coupling methyl group on C‐1 (Me‐15) was substantiated by the HMBC between H3‐16/C‐4, COur‐5, C present‐6 andbetween H study‐6/C H‐ has16;‐6 alsoand ledHfurther3‐16 to the( J supporting= isolation 1.6 Hz). of Theby a newan methyl allylic briarane,HMBC group coupling fragilide fromon C ‐H1 3 V‐(Me15/C (2).‐15)‐1, The Cwas‐ molecular2, Csubstantiated‐10, C‐14; and by H ‐the2, H ‐10, H‐14/C‐15. The epoxy group at C‐11/20 was Our present study has also led to the isolation of a new briarane, fragilide V (2). The molecular HMBC from H3‐15/C‐1, C‐2, C‐10, C‐14; and H‐2, H‐10, H‐14/C‐15. The epoxy35 group at C‐11/20 was between H‐6 and H3‐16 (formulaJ = 1.6 Hz). of C 26TheH35 methylClO11 was group deduced on C from‐1 (Me (+‐)-HRESIMS15) was substantiated at m/z 581.17589confirmed by the (calcd. by the for HMBC C26H35 betweenClO11 +H2‐20/C‐10, C‐11; and H‐20a/C‐12; and further supporting by a formula of C26H35ClO11 was deduced from (+)‐HRESIMS at m/z 581.17589 (calcd. for C26H3535ClO11 + HMBC from H3‐15/C‐1, C‐2, C‐10, C‐14;confirmed and H‐2, Hby‐10, the H HMBC‐14/C‐15. between The13 epoxy H2‐20/C group‐10, at C C‐long11;‐11/20 and range was H‐ 20a/C 4J‐1H–‐112;H correlationand further between supporting H‐10 by ( Ha 2.36) and H‐20b (H 2.98) (J = 1.6 Hz). A hydroxy Na, 581.17601). Carbonyl resonances in the 13C NMR spectrum of 2 (Table1) at δC 175.2, 171.0, 170.1, Na, 581.17601). Carbonyl4 resonances1 1 in the C NMR spectrum of 2 (Table 1) at C 175.2, 171.0, 170.1, confirmed by the HMBCand between 169.7 revealedH2‐20/Clong‐ the10,range presenceC‐ 11;J‐ H–and ofH H acorrelation‐γ20a/C-lactone‐12; and andbetween three further esters.H‐ 10supporting ( InHgroup 2.36) the 1 H andbyattaching NMR aH ‐20b spectrum at( HC ‐2.98)8 ofwas 2(J (Table to= 1.6infer 2Hz).), that A anhydroxy HMBC of a hydroxy proton at H 4.85 to C‐7, C‐8, and and 169.7 revealed the presence of a γ‐lactone and three esters. In the 1H NMR spectrum of 2 (Table 4 1 1 group attaching at C‐8 was to infer that an HMBC of a hydroxy proton at H 4.85 to C‐7, C‐8, and long range J‐ H– H correlationthe signals between for three H‐10 acetate (H 2.36) methyls and H were‐20b observed (H 2.98) at(J δ=H 1.62.23, Hz). 2.03, CA‐ 9.hydroxy and Moreover, 1.99. It was HMBC found from that the the oxymethine 1D protons at H 5.03 (H‐2), 5.65 (H‐9), and 4.71 (H‐14) to 2), the signals for three acetate methyls were observed at H 2.23, 2.03, and 1.99. It was found that group attaching at C‐8 was(Tables to infer1 and that2)C and ‐an9. Moreover, 2DHMBC NMR of (Figure HMBCa hydroxy3) datafrom proton of the2 oxymethinewere at H similar 4.85 to protons to C those‐the7, C acetate‐at of8,  aandH known 5.03 carbonyls (H briarane,‐2), 5.65 at  robustolideC(H 169.0,‐9), and 169.4, 4.71 and (H ‐170.1,14) to placed the acetoxy groups on C‐2, C‐9, and C‐14, the 1D (Tables 1 and 2) and 2D NMR (Figure 3) data of 2 were similar to those of a known briarane, C‐9. Moreover, HMBC fromF(6 )[the17 oxymethine,18] (Figurethe acetate1 ),protons except carbonyls thatat H the 5.03 at signals  (HC 169.0,‐2), corresponding 5.65 169.4, (H ‐and9), and 170.1, to 4.71 a hydroxy respectively.placed (H‐14) the group to acetoxy Ten in of2 were groupsthe 12 replaced oxygenon C‐2, by atomsC‐9, and in theC‐14, molecular formula of 1 could be accounted for the robustolide F (6) [17,18] (Figure 1), except that the signals corresponding to a hydroxy group in 2 the acetate carbonyls at signalsC 169.0, for 169.4, a protonrespectively. and 170.1, in 6. In placed theTen NOESY of the the acetoxy 12 spectrum, oxygen groups oneatoms on of the Cin‐ 2,the C-3 C ‐molecular9, methylenepresence and C‐14, offormula protons a γ‐lactone, (ofδ 12.38) could three showed be esters, accounted a an epoxide, for the and a hydroxy group. Thus, the remaining two were replaced by signals for a proton in 6. In the NOESY spectrum, one of theH C‐3 methylene respectively. Ten of the 12correlation oxygen atoms topresence H-7 in and the notofmolecular a with γ‐lactone, H-2, formula suggesting three of esters, 1 could the anβ -orientationbe epoxide, accountedoxygen and of for thisa hydroxyatomsthe proton had group. by to modeling be Thus,positioned studythe remaining at C‐4 as antwo acetoxy group, as indicated by its 1H and 13C NMR protons (H 2.38) showed a correlation to H‐7 and not with H‐2, suggesting the β‐orientation of this oxygen atoms had to be positioned at C‐4 as an acetoxy group, as indicated by its 1H and 13C NMR presence of a γ‐lactone, andthree the esters, other an was epoxide, assigned and as H-3a hydroxyα (δH 1.75). group. A correlation Thus, the from remainingchemical H-4 to H-3two shiftsα , suggested (H 5.91, that 1H, H-4 br s; was C 70.6, CH), although no HMBC was observed from H‐4 to any proton by modeling study and the other was assigned as H‐3α (H 1.75). A correlation from H‐4 to oxygen atoms had to be positionedα-oriented at according Cchemical‐4 as an to shiftsacetoxy modeling ( Hgroup, 5.91, analysis. 1H,as indicated Therefore,br s; C 70.6, by the its CH), configuration 1H andalthoughacetate 13C NMR of carbonyl.no the HMBC stereogenic was observed carbons of from2 H‐4 to any H‐3α, suggested that H‐4 was α‐oriented according to modeling analysis. Therefore, the chemical shifts (H 5.91, 1H,were br elucidated s; C 70.6,acetate asCH), 1R carbonyl.,2 althoughS,4R,6S ,7 noR,8 HMBCR,9S,10 Swas,11 Sobserved,14S, and from 17R (Figure H‐4 to3 any) (Supplementary Materials, configuration of the stereogenic carbons of 2 were elucidated as 1R,2S,4R,6S,7R,8R,9S,10S,11S,14S, acetate carbonyl. Figures S9–S16). and 17R (Figure 3) (Supplementary Materials, Figures S9–S16).

O O OCCH 3 OH CH3CO 15 4 16 14 5 1 6 10 OH 11 Cl 9 O 20 8 O CH3CO 17 19 O 18 Figure 2. The correlation spectroscopy (COSY) ( ) correlations, selective heteronuclear multiple O Figure 2. The correlation spectroscopy (COSY) ( ) correlations,bond correlation selective (HMBC) heteronuclear experiment multiple ( ), and protons with key nuclear Overhauser effect Figure 2. The correlation spectroscopyFigureFigure 3. 3.The The (COSY) COSY bondCOSY ( correlation ( ) correlations,) correlations, (HMBC) selectiveselective selective experiment HMBCheteronuclear HMBC ( ( ( ),), andand multipleand protons protonsspectroscopy with withwith key key key(NOESY) NOESY NOESY nuclear ( ( Overhauser) correlations) effect of 1. bond correlation (HMBC) correlationsexperimentcorrelations of( of2spectroscopy. 2.), and protons (NOESY) with ( key ) nuclearcorrelations Overhauser of 1. effect spectroscopy (NOESY) ( ) correlations of 1. Based on a summary of the 13C chemical shifts of 11,20‐epoxy group in naturally occurring 13 26 33 10 BriaraneBriarane3 3(fragilide (fragilideBased W)on W) wasa wassummary found found to ofto have thehave a molecular Ca molecularchemical formula shiftsformulabriarane of Cof11,20 26analogues, CH33‐HepoxyClOClO10 withgroupbasedbased 13 onCin on NMR itsnaturally its (+ (+) )-data‐ occurringfor C‐11 and C‐20 at C 62–63 and 58–60 ppm, the epoxide 3535 Based on a summaryHRESIMSHRESIMS of the 13 atC at mchemicalbriarane /m/zz 563.16554 563.16554 analogues,shifts (calcd. (calcd.of 11,20 forwith for‐ Cepoxy 26 13CCH26 33HNMR 33groupClOClO data10 10in+ + for Na,naturallyNa, C 563.16545).563.16545).‐11 andgroup occurring C‐ 20was ItsIts at absorption β‐absorption Coriented 62–63 and peaks peaksand 58–60 the in in the cyclohexane theppm, IR IR the epoxide ring existed in a twist boat conformation [16]; thus, the 1 –1 briarane analogues, with spectrum13spectrumC NMR showed data showedgroup for ester, C was‐11ester,γ and β‐-lactone, γ‐oriented Clactone,‐20 andat andC broad and62–63 the OHcyclohexanebroad and stretching 58–60 OH ppm, stretchingring at 1738, existedtheconfiguration epoxide 1777, at in 1738, anda twist 3459 of1777, boatthe cm − 11,20andconformation, respectively. ‐3459epoxy cm group [16];, thus,in 1 shouldthe be β‐oriented and the cyclohexane ring was 13 13 group was β‐oriented andThe respectively.the Ccyclohexane NMRconfiguration spectrum The ringC NMR indicatedexisted of spectrum thein that a 11,20 twist three indicated‐ epoxyboat esters conformation group and that a γthreein-lactone 1 should [16];estersfound were thus, andbe present, β‐to the abeoriented γ‐ inlactone as a carbonyltwist and were boat the resonances present, conformationcyclohexane as ringfor thewas 13 C chemical shifts at C 62.5 (C‐11) and 59.1 (CH2‐ 1 1 configuration of the 11,20werecarbonyl‐epoxy observed group resonancesfound at inδ C1 174.8,toshould werebe in 170.4, observed abe twist β‐ 169.7,oriented boat at 169.3  conformationC 174.8,and (Table the 170.4,1). cyclohexane The for169.7, theH NMR169.3 1320).C ring chemical spectrumThe(Table was relative 1). shifts indicatedThe stereochemistry atH  NMRC 62.5 the presence spectrum(C‐11) of and 1 was 59.1 established (CH2‐ by the analysis of correlations observed in a 1 13 1 found to be in a twist boatofindicated conformation three acetate the20). presence methyls for The the relative 13 (ofδCH threechemical2.27, stereochemistry acetate 2.25, shifts 2.11, methyls at each  Cof 3H62.5( 1H was2.27, (Cs)‐ 11) (Table established2.25, and 2.11,2). 59.1 The each by(CH H the3H2‐ and analysis s) (TableC NMR of 2).correlations spectra The H of and3 observed in a × nuclear Overhauser effect spectroscopy (NOESY) experiment (Figure 2). In the NOESY spectrum of 20). The relative stereochemistrywas13C found NMR of to1spectra was benuclear similar established of Overhauser3 withwas those foundby the effect of to analysis a knownbe spectroscopy similar of briarane, correlations with (NOESY) those juncenolide observedof 1experimenta, NOEknown M (in= correlations frajunolide abriarane, (Figure 2). juncenolide S)between In (7) the (Figure NOESYH ‐M10/H1), (= ‐spectrum2, H‐10/H of‐9, and H‐10/OH‐8, while no correlation was seen nuclear Overhauser effectisolated spectroscopyfrajunolide from S)1J. (NOESY), juncea(NOE7) (Figure correlationsand experiment J.1), fragilis isolated between [(Figure19, 20from], H except2). ‐J.10/H Injuncea the that‐2, NOESY Hand the‐10/H signalsJ. ‐fragilisspectrum9,between and corresponding [19,20],H ‐Hof10/OH‐ 10 exceptand‐8, H towhile 3‐ thethat15, 13-acetoxy nosuggestingthe correlation signals that was these seen protons H‐2, H‐9, and H‐10, and the hydroxy group 1, NOE correlations betweenandcorresponding 3(HZ‐10/H)-ene‐2, moietiesbetween H to‐10/H the inH ‐139,‐710 ‐andacetoxywere and H disappearedH‐10/OH 3and‐15, suggesting3(‐8,Z )while‐ene and moieties replaced nothat correlation these in by protons7 a wereat proton wasC‐ 8 disappeared Hwereseen and‐2, H α‐ an‐9,oriented; ( andE)-ene and H‐ 10, moieties replacedmeanwhile, and the inby hydroxy3 ,ana NOE group correlation of H3‐15 with H‐14 indicated that H‐14 was between H‐10 and H3‐15,respectively. suggestingproton and that The atan C these locations(‐E8 )were‐ene protons α‐moieties oforiented; the H functional‐2, inH ‐meanwhile,39,, andrespectively. groups H‐10, an were and NOE The confirmedthe correlation hydroxylocationsβ‐oriented. by group 2D-NMR ofof Hthe 3 The‐ 15functional correlations withNOESY H‐14 spectrumgroups indicated (Figure were 4showed), that H ‐a14 correlation was from H‐6 to H3‐16, revealing the Z geometry at C‐8 were α‐oriented; meanwhile,andconfirmed hence thean by β‐NOE structure 2Doriented.‐ NMRcorrelation of Thecorrelations fragilide NOESY of H3 W‐ 15 (Figurespectrum was with assigned H4),‐ 14showed and indicated hence as 3a, correlation and the that theofstructure H configurationsC‐14‐ 5/6from was ofdouble H fragilide‐6 to H ofbond.3‐ theW16, was stereogenicrevealingH3‐ assigned18 exhibited the Z geometry NOE correlations to OH‐8 and H‐9, suggesting the β‐oriented. The NOESY spectrumcarbonsas 3, and were showed the elucidatedof configurations Ca ‐correlation5/6 double as 1R,2 fromof Sbond.,7 theS,8 H Rstereogenic‐ ,96H Sto3,10‐ 18HS3 ‐,1116,exhibitedR carbonsrevealing, 14S, and NOEwere the 17R Zelucidatedα‐correlations(Figure geometryorientation4) (Supplementary as toof1R Me,2OHS,7‐‐188S ,8 atandR ,9C Materials,S ‐17.,10H‐ S9,H,11 ‐suggesting7R displayed, theNOE correlations with H‐4 and H‐17, which further of C‐5/6 double bond. FiguresH143‐S,18 and exhibited S17–S24). 17Rα‐ (Figureorientation NOE 4) correlations(Supplementary of Me‐18 atto C OH‐Materials,17.‐ 8H ‐and7 displayed FiguresH‐9, suggesting S17–S24).NOEconfirmed correlations the that withthese Hthree‐4 and protons H‐17, werewhich in further β‐orientation at C‐7, C‐4, and C‐17. Based on the above α‐orientation of Me‐18 at C‐17.The H‐ known7 displayedconfirmed compound NOE that correlations4 thesewas found three with toprotons be H identical‐4 wereand Hin with‐ β‐17,orientation thewhichfindings, known further junceellonoidat the C ‐ 7,relative C‐4, and configurations D, C on‐17. the Based basis ofon stereogenicthe above carbons of 1 were elucidated as 1R*,2S*,4S*,7S*, confirmed that these threeof protons the comparison werefindings, in β‐ oforientation its the physical relative at and Cconfigurations‐7, spectroscopic C‐4, and C of‐17. stereogenic data Based with on8 those Rthecarbons*,9 aboveS*,10 of reportedSof *,11 1 wereS*,14 previously Selucidated* and 17R [ 5*.as, 21 However,1]R*,2S*,4S as*,7 Sbriaranes*, 1–4 were isolated along with the known findings, the relative configurations(Supplementary of 8stereogenicR*,9 Materials,S*,10S*,11 carbons FiguresS*,14S of S25–S32).* and 1 were 17R elucidated*. However, as as1R briaranes*,2chlorinatedS*,4S*,7 1S–*,4 briarane,were isolated junceellin along (5with) [6], the and known the structure, including the absolute configuration of 8R*,9S*,10S*,11S*,14S* and 17R*. However,chlorinated as briaranes briarane, 1– 4junceellin were isolated (5) [6], along and withthe structure,junceellinthe known including( 5) was furtherthe absolute confirmed configuration by a single of ‐crystal X‐ray diffraction analysis [7,15]. It is chlorinated briarane, junceellin (5) [6],junceellin and the structure,(5) was furtherincluding confirmed the absolute by a configuration singlereasonable,‐crystal of X therefore,‐ray diffraction on biogenetic analysis grounds [7,15]. Itto isassume that briaranes 1–4 have the same absolute junceellin (5) was further confirmed reasonable,by a single therefore,‐crystal X on‐ray biogenetic diffraction grounds analysis to assume[7,15].configuration Itthat is briaranes as that of1– 45 .have Therefore, the same the absoluteconfigurations of the stereogenic carbons of 1 should be reasonable, therefore, on biogenetic groundsconfiguration to assume as that that of briaranes 5. Therefore, 1–4 have the configurationsthe sameelucidated absolute of theas 1 stereogenicR,2S,4S,7S,8 carbonsR,9S,10S of,11 1S ,14shouldS and be 17 R (Supplementary Materials, Figures S1–S8). configuration as that of 5. Therefore, theelucidated configurations as 1R,2 Sof,4 Sthe,7S ,8stereogenicR,9S,10S,11 carbonsS,14S and of 171 Rshould (Supplementary be Materials, Figures S1–S8). elucidated as 1R,2S,4S,7S,8R,9S,10S,11S ,14S and 17R (Supplementary Materials, Figures S1–S8).

Mar. Drugs 2019, 17, x 5 of 9

Our present study has also led to the isolation of a new briarane, fragilide V (2). The molecular formula of C26H35ClO11 was deduced from (+)‐HRESIMS at m/z 581.17589 (calcd. for C26H3535ClO11 + Na, 581.17601). Carbonyl resonances in the 13C NMR spectrum of 2 (Table 1) at C 175.2, 171.0, 170.1, and 169.7 revealed the presence of a γ‐lactone and three esters. In the 1H NMR spectrum of 2 (Table 2), the signals for three acetate methyls were observed at H 2.23, 2.03, and 1.99. It was found that the 1D (Tables 1 and 2) and 2D NMR (Figure 3) data of 2 were similar to those of a known briarane, robustolide F (6) [17,18] (Figure 1), except that the signals corresponding to a hydroxy group in 2 were replaced by signals for a proton in 6. In the NOESY spectrum, one of the C‐3 methylene protons (H 2.38) showed a correlation to H‐7 and not with H‐2, suggesting the β‐orientation of this proton by modeling study and the other was assigned as H‐3α (H 1.75). A correlation from H‐4 to H‐3α, suggested that H‐4 was α‐oriented according to modeling analysis. Therefore, the configuration of the stereogenic carbons of 2 were elucidated as 1R,2S,4R,6S,7R,8R,9S,10S,11S,14S, and 17R (Figure 3) (Supplementary Materials, Figures S9–S16).

O O OCCH 3 OH CH3CO 15 4 16 14 5 1 6 10 OH 11 Cl 9 O 20 8 Mar. Drugs 2019, 17, x 4 of 9 O CH3COMar. Drugs 2019, 17, x 4 of 9 17 19 O 3 3 2 2 3 Mar. Drugs 2019, 17, x 18 H‐7, H4 ‐of16/C 9 ‐5; H‐9, H‐10, H ‐18/C‐8; H‐9, H‐10, H ‐12, H‐13α, H ‐20/C‐11; and H‐17, H ‐18/C‐19, H‐7, H3‐16/C‐O5; H‐9, H ‐10, H3‐18/C‐8; H‐9, H‐10, permittedH2‐12, H‐ 13elucidationα, H2‐20/C of‐11; the and carbon H‐17, skeletonH3‐18/C ‐of19, 1. A vinyl methyl at C‐5 was confirmed by the H‐7, H3‐16/C‐5; H‐9, H‐10, H3‐18/C‐8; permittedH‐9, H‐10, elucidationH2‐12, H‐13 ofα, theH2‐ 20/Ccarbon‐11; skeletonand H‐17, of H1.3 ‐18/CA vinyl‐19, methyl 3 at C‐5 was confirmed by the Figure 3. The COSY ( ) correlations, selective HMBC ( ), and HMBCprotons withbetween key NOESYH ‐16/C ( ‐4, C) ‐5, C‐6 and H‐6/C‐16; and further supporting by an allylic coupling HMBC between H3‐16/C‐4, C‐5, C‐6 and H‐6/C‐16; and further supporting by an allylic coupling permitted elucidation of thecorrelations carbon skeleton of 2. of 1. A vinyl methyl at C‐5 was confirmedbetween by the H ‐6 and H3‐16 (J = 1.6 Hz). The methyl group on C‐1 (Me‐15) was substantiated by the HMBC between H3‐16/C‐4, C‐5, C‐6 andbetween H‐6/C H‐16;‐6 and Hfurther3‐16 ( J supporting= 1.6 Hz). Theby an methyl allylicHMBC group coupling fromon C ‐H1 3‐(Me15/C‐15)‐1, Cwas‐2, Csubstantiated‐10, C‐14; and by H ‐the2, H ‐10, H‐14/C‐15. The epoxy group at C‐11/20 was HMBC from H3‐15/C‐1, C‐2, C‐10, C‐14; and H‐2, H‐10, H‐14/C‐15. The epoxy group at C‐11/20 was between H‐6 and H3‐16 (J =Briarane 1.6 Hz). 3 The (fragilide methyl W) group was foundon C‐ 1to (Me have‐15) a molecularwas substantiated formulaconfirmed ofby C the26H by 33ClO the10 HMBC based on between its (+)‐ H2‐20/C‐10, C‐11; and H‐20a/C‐12; and further supporting by a confirmed by the HMBC between H2‐20/C‐10, C‐11; and H‐20a/C4 1 ‐112; and further supporting by a HMBC from H3‐15/C‐1,HRESIMS C‐2, C‐10, atC ‐m/z14; and563.16554 H‐2, H (calcd.‐10, H‐ 14/Cfor C‐2615.H 33The35ClO epoxy10 + Na, group 563.16545). at Clong‐11/20 Itsrange wasabsorption J‐ H– H peaks correlation in the betweenIR H‐10 (H 2.36) and H‐20b (H 2.98) (J = 1.6 Hz). A hydroxy 4 1 1 H H confirmed by the HMBCspectrum between showed H2‐20/Clong ‐ ester,10,range C γ‐‐ 11;J‐lactone, H–andH H correlation ‐20a/Cand ‐broad12; andbetween OH further stretching H‐ 10supporting ( group 2.36)at 1738, andbyattaching a H 1777,‐20b at (and C ‐2.98)8 was3459 (J to=cm 1.6infer–1 ,Hz). that A anhydroxy HMBC of a hydroxy proton at H 4.85 to C‐7, C‐8, and 4 1 1 long range J‐ H– H correlationrespectively. between Thegroup H13C‐10 NMRattaching (H 2.36) spectrum atand C ‐H8 indicated‐was20b (toH infer 2.98) that that (threeJ = an1.6 esters HMBCHz). C Aand ‐ of9.hydroxy Moreover, aa γ‐hydroxylactone HMBCproton were frompresent,at H the4.85 asoxymethine to C‐7, C‐8, protonsand at H 5.03 (H‐2), 5.65 (H‐9), and 4.71 (H‐14) to C‐9. Moreover, HMBC from the oxymethine protons at H 5.03 (H‐2), 5.65 (H‐9), and 4.71 (H‐14) to group attaching at C‐8 carbonylwas to infer resonances that an HMBCwere observed of a hydroxy at C 174.8,proton 170.4, at  H169.7, 4.85 to169.3 C‐the7, (Table C acetate‐8, and 1). Thecarbonyls 1H NMR at  spectrumC 169.0, 169.4, and 170.1, placed the acetoxy groups on C‐2, C‐9, and C‐14, the acetate carbonyls at C 169.0, 169.4, and 170.1, placed the acetoxy groups on C‐2, C‐9, and C‐14, C‐9. Moreover, HMBC indicatedfrom the oxymethinethe presence protons of three at acetate H 5.03 methyls (H‐2), 5.65(H 2.27, (H‐9), 2.25, and 2.11, 4.71respectively. each (H‐ 14)3H to s) Ten(Table of 2).the The 12 oxygen1H and atoms in the molecular formula of 1 could be accounted for the the acetate carbonyls at13 CC 169.0,NMR 169.4,spectrarespectively. and of 170.1,3 was placedfound Ten of theto the beacetoxy 12 similar oxygen groups with atoms thoseon Cin‐ 2,ofthe C a ‐ molecular9,knownpresence and C briarane,‐14, offormula a γ‐ lactone,juncenolide of 1 could three beM esters, accounted(= an epoxide, for the and a hydroxy group. Thus, the remaining two respectively. Ten of thefrajunolide 12 oxygen atomsS) (presence7) (Figurein the ofmolecular 1), a γ‐ isolatedlactone, formula from three ofJ. esters, juncea1 could anand be epoxide, accountedJ. fragilisoxygen and [19,20], for a hydroxyatomsthe except had group. thatto be the Thus,positioned signals the remaining at C‐4 as antwo acetoxy group, as indicated by its 1H and 13C NMR 1 13 presence of a γ‐lactone,corresponding three esters, anoxygento epoxide,the 13 atoms‐acetoxy and had a and hydroxyto be 3( positionedZ) ‐enegroup. moieties Thus,at C‐ 4 inthe as 7 an remainingwere acetoxychemical disappeared group,two shifts as (and indicatedH 5.91, replaced 1H, by br byits s; aH C and70.6, CCH), NMR although no HMBC was observed from H‐4 to any oxygen atoms had to beproton positioned and atan C chemical(‐E4) ‐asene an moietiesshiftsacetoxy ( Hgroup,in 5.91, 3, respectively. 1H,as indicated br s; C 70.6, Theby its CH),locations 1H andalthoughacetate 13ofC NMRthe carbonyl.no functional HMBC was groups observed were from H‐4 to any chemical shifts (H 5.91,confirmed 1H, br s; byC 70.6,2Dacetate‐NMR CH), carbonyl. correlationsalthough no (Figure HMBC 4), was and observed hence the from structure H‐4 to of any fragilide W was assigned Mar. Drugs 2019, 17, 706 6 of 9 acetate carbonyl. as 3, and the configurations of the stereogenic carbons were elucidated as 1R,2S,7S,8R,9S,10S,11R, 14S, and 17R (Figure 4) (Supplementary Materials, Figures S17–S24).

Mar. Drugs 2019, 17, x 6 of 9

Figure 4. The COSY ( ) correlations, selective HMBC ( ), and protons with key NOESY ( ) Figure 2. The correlation spectroscopy (COSY) ( ) correlations, selective heteronuclear multiple correlations of 3. Figure 2. The correlation spectroscopy (COSY) ( ) correlations,bond correlation selective (HMBC) heteronuclear experiment multiple ( ), and protons with key nuclear Overhauser effect Figure 2. The correlation spectroscopyFigure 4. The (COSY) COSYbond ( correlation) correlations, (HMBC) selectiveselective experiment HMBCheteronuclear ( ( ), andandmultiple protons protonsspectroscopy with with key key(NOESY) NOESY nuclear ( Overhauser) correlations effect of 1. The known compound 4 was found to be identical with the known junceellonoid D, on the 3spectroscopy (NOESY) ( ) correlations of 1. bond correlation (HMBC)basis correlationsexperiment of the comparison of( . ), and ofprotons its physical with keyand nuclearspectroscopic Overhauser data witheffect those of reported previously spectroscopy (NOESY) ( ) correlations of 1. Based on a summary of the 13C chemical shifts of 11,20‐epoxy group in naturally occurring [5,21] (Supplementary Materials, Figures S25–S32).13 Using an in vitroBasedpro-inflammatory on a summary suppression of the C assay,chemical the activitiesshifts of of11,20 briaranes‐epoxy1 –group5 on13 the in releasenaturally occurring C Using an in vitro pro‐inflammatory suppression assay, thebriarane activities analogues, of briaranes with 1 –5C on NMR the data for C‐11 and C‐20 at 62–63 and 58–60 ppm, the epoxide of iNOS and13 cyclooxygenase-2briarane analogues, (COX-2) with protein13C NMR from data lipopolysaccharides for C‐11 and C‐20 (LPS)-stimulated at C 62–63 and 58–60 RAW264.7 ppm, the epoxide Based on a summary ofrelease the ofC iNOSchemical and shiftscyclooxygenase of 11,20‐epoxy‐2 (COX group‐2) protein in naturally from lipopolysaccharidesgroup occurring was β‐ oriented (LPS) and‐stimulated the cyclohexane ring existed in a twist boat conformation [16]; thus, the were13 assayedgroup (Figure was5 and β‐oriented Table3). Theand resultsthe cyclohexane showed that ring briaranes existed in 3aand twist5 reducedboat conformation the release [16]; thus, the briarane analogues, with CRAW264.7 NMR data were for Cassayed‐11 and (Figure C‐20 at5 andC 62–63 Table and3). The 58–60 results ppm, showed theconfiguration epoxide that briaranes of the 3 and11,20 5 ‐reducedepoxy group in 1 should be β‐oriented and the cyclohexane ring was of iNOS to 28.55 and 33.72% at a concentration of 10 µM. Briarane 4 was found to be more weak in 13 group was β‐oriented and thethe cyclohexane release configurationof iNOS ring to existed 28.55 of theandin a 11,20 33.72%twist‐ epoxyboat at a conformation concentrationgroup in 1 should [16];of found10 thus,  beM. β‐to Briaranethe beoriented in a 4twist wasand boat foundthe conformationcyclohexane to be ringfor thewas C chemical shifts at C 62.5 (C‐11) and 59.1 (CH2‐ 13 4 C 5 2 configuration of the 11,20terms‐epoxymore of weak reducinggroupfound in in terms the1 toshould expression beof reducingin abe twist β‐ of orientedthe iNOS,boat expression conformation indicatingand the of iNOS,cyclohexane that for theindicatingthe activity 20).C ring chemical The that ofwas relative briaranesthe shifts activity stereochemistry at andof 62.5briaranesis (C largely‐11) 4of and 1 was 59.1 established (CH ‐ by the analysis of correlations observed in a found to be in a twist boatdependent conformationand 5 is onlargely20). the for functionalThe dependent the relative 13C chemical groups onstereochemistry the at shiftsfunctional C-2 andat  C C-3. ofgroups 62.5 1 Itwas (C is at‐ interesting 11)established C ‐and2 andnuclear 59.1 C to ‐ 3.by(CH note It Overhauserthe 2is‐ that interestinganalysis briarane effect of tocorrelations4 wasnotespectroscopy foundthat observed (NOESY) in a experiment (Figure 2). In the NOESY spectrum of 20). The relative stereochemistryto enhancebriarane of the14 wasnuclear expression foundestablished Overhauser to ofenhance COX-2 by the effectthe to analysis expression 130.88%, spectroscopy of at correlationsof a COX concentration (NOESY)‐2 to 130.88%,observed 1experiment, ofNOE 10 at in µacorrelationsM. concentrationa (Figure 2). between Inof 10the  M.NOESYH ‐10/H ‐spectrum2, H‐10/H of‐9, and H‐10/OH‐8, while no correlation was seen nuclear Overhauser effect spectroscopy1 (NOESY), NOE correlations experiment between (Figure H2).‐10/H In the‐2, NOESYH‐10/H ‐spectrum9,between and H ‐Hof10/OH‐ 10 and‐8, Hwhile3‐15, nosuggesting correlation that was these seen protons H‐2, H‐9, and H‐10, and the hydroxy group 1, NOE correlations between H‐10/H‐2,between H‐10/H H‐9,‐10 and and H H‐10/OH3‐15, suggesting‐8, while nothat correlation these protonsat wasC‐8 Hwereseen‐2, H α‐‐9,oriented; and H‐10, meanwhile, and the hydroxy an NOE group correlation of H3‐15 with H‐14 indicated that H‐14 was between H‐10 and H3‐15, suggesting thatat Cthese‐8 were protons α‐oriented; H‐2, H ‐meanwhile,9, and H‐10, an and NOE the correlation hydroxyβ‐oriented. group of H 3 The‐15 withNOESY H‐14 spectrum indicated showed that H ‐a14 correlation was from H‐6 to H3‐16, revealing the Z geometry at C‐8 were α‐oriented; meanwhile, an β‐NOEoriented. correlation The NOESY of H3‐15 spectrum with H‐ 14showed indicated a correlation thatof H C‐14‐ 5/6from was double H‐6 to Hbond.3‐16, revealingH3‐18 exhibited the Z geometry NOE correlations to OH‐8 and H‐9, suggesting the β‐oriented. The NOESY spectrum showedof Ca ‐correlation5/6 double from bond. H‐ 6H to3‐ 18H3 ‐16,exhibited revealing NOE the Zα‐correlations geometryorientation toof MeOH‐‐188 atand C ‐17.H‐ 9,H ‐suggesting7 displayed theNOE correlations with H‐4 and H‐17, which further of C‐5/6 double bond. H3‐18 exhibitedα‐orientation NOE correlations of Me‐18 atto C OH‐17.‐ 8H ‐and7 displayed H‐9, suggesting NOEconfirmed correlations the that withthese Hthree‐4 and protons H‐17, werewhich in further β‐orientation at C‐7, C‐4, and C‐17. Based on the above α‐orientation of Me‐18 at C‐17. H‐7 displayedconfirmed NOE that correlations these three with protons H‐4 wereand Hin‐ β‐17,orientation whichfindings, further at the C ‐ 7,relative C‐4, and configurations C‐17. Based ofon stereogenicthe above carbons of 1 were elucidated as 1R*,2S*,4S*,7S*, confirmed that these three protons werefindings, in β‐orientation the relative at Cconfigurations‐7, C‐4, and C of‐17. stereogenic Based on8 Rthecarbons*,9 aboveS*,10 Sof *,11 1 wereS*,14 Selucidated* and 17R *.as However,1R*,2S*,4S as*,7 Sbriaranes*, 1–4 were isolated along with the known findings, the relative configurations of 8stereogenicR*,9S*,10S*,11 carbonsS*,14S of* and 1 were 17R elucidated*. However, as as1R briaranes*,2chlorinatedS*,4S*,7 1S–*,4 briarane,were isolated junceellin along (5with) [6], the and known the structure, including the absolute configuration of 8R*,9S*,10S*,11S*,14S* and 17R*. However,chlorinated as briaranes briarane, 1– 4junceellin were isolated (5) [6], along and withthe structure,junceellinthe known including( 5) was furtherthe absolute confirmed configuration by a single of ‐crystal X‐ray diffraction analysis [7,15]. It is chlorinated briarane, junceellin (5) [6],junceellin and the structure,(5) was furtherincluding confirmed the absolute by a configuration singlereasonable,‐crystal of X therefore,‐ray diffraction on biogenetic analysis grounds [7,15]. Itto isassume that briaranes 1–4 have the same absolute reasonable, therefore, on biogenetic grounds to assume that briaranes 1–4 have the same absolute junceellin (5) was further confirmedFigureFigure 5. Activities5.by Activities a single of of briaranes ‐briaranescrystal1 –X15–‐5onray on the thediffraction expression expression analysis of of inducibleinducible [7,15].configuration nitric Itoxide is synthase synthase as that (iNOS) (iNOS) of 5. andTherefore, and the configurations of the stereogenic carbons of 1 should be reasonable, therefore, on biogeneticcyclooxygenase-2cyclooxygenase groundsconfiguration (COX-2)to‐2 (COXassume proteinsas‐2) that proteinsthat inof briaranes LPS-treated 5 in. Therefore, LPS‐ treated1– murine4 have themurine RAW264.7 configurationsthe RAW264.7sameelucidated macrophage absolute macrophage of theas cells. 1 stereogenicR,2 Westerncells.S,4S ,7WesternS blotting,8 carbonsR,9 S,10S of,11 1S ,14shouldS and be 17 R (Supplementary Materials, Figures S1–S8). blotting showed that briaranes 3 and 5 reduced the expression of iNOS. Data were normalized to configuration as that of 5. Therefore,showed thattheelucidated briaranesconfigurations as3 1andR,2 5Sof,4reduced Sthe,7S ,8stereogenicR the,9S expression,10S,11 carbonsS,14 ofS iNOS.and of 171 DataRshould (Supplementary were be normalized Materials, to the cells Figures S1–S8). the cells treated with LPS only, and cells treated with dexamethasone were used as a positive elucidated as 1R,2S,4S,7S,8R,9treatedS,10S,11 withS ,14 LPSS and only, 17 andR (Supplementary cells treated with Materials, dexamethasone Figures were S1–S8). used as a positive control. Data are control. Data are expressed as the mean  SEM (n = 3). * Significantly different from cells treated expressed as the mean SEM (n = 3). * Significantly different from cells treated with LPS (p < 0.05). with LPS (p < 0.05). ± Table 3. Effects of briaranes 1–5 on LPS-induced pro-inflammatory inducible nitric oxide synthase (iNOS)Table and 3. cyclooxygenase-2Effects of briaranes (COX-2) 1–5 on proteinLPS‐induced expression pro‐inflammatory in macrophages inducible at a concentration nitric oxide synthase of 10 µM. (iNOS) and cyclooxygenase‐2 (COX‐2) protein expression in macrophages at a concentration of 10 M. iNOS COX-2 β-Actin Compound Expression (% of LPS) iNOS COX‐2 β‐Actin Lipopolysaccharides 100.01 17.81 100.00 3.04 100.00 6.05 Compound ± Expression± (% of LPS) ± 1 55.88 11.42 64.73 4.07 90.05 7.11 Lipopolysaccharides 100.01±  17.81 100.00±  3.04 ±100.00  6.05 2 77.58 8.82 66.23 6.59 99.29 5.24 ± ± ± 13 28.5555.88  11.425.35 86.4664.738.46  4.07 101.69 90.056.46  7.11 ± ± ± 24 67.2577.58  4.278.82 130.8866.2313.94  6.59 103.20 99.292.56  5.24 ± ± ± 35 33.7228.55  2.575.35 64.1586.463.03  8.46 84.52 101.697.78  6.46 ± ± ± Dexamethasone 12.11 0.03 2.11 0.44 123.86 2.99  4 67.25± 4.27 130.88± 13.94 103.20± 2.56 Data were normalized5 to those of cells treated33.72 with  2.57 LPS alone, and cells64.15 treated  3.03 with dexamethasone84.52 were  7.78 used as a positive control (10 µM). Data are expressed as the mean SEM (n = 3). Dexamethasone 12.11  0.03 ± 2.11  0.44 123.86  2.99 Data were normalized to those of cells treated with LPS alone, and cells treated with dexamethasone were used as a positive control (10 M). Data are expressed as the mean  SEM (n = 3).

3. Materials and Methods

3.1. General Experimental Procedures Melting points were determined using a Fargo apparatus and the values were uncorrected. 1D and 2D NMR spectra were recorded on a 600 MHz Jeol NMR (model ECZ600R, Tokyo, Japan) or on

Mar. Drugs 2019, 17, 706 7 of 9

3. Materials and Methods

3.1. General Experimental Procedures Melting points were determined using a Fargo apparatus and the values were uncorrected. 1D and 2D NMR spectra were recorded on a 600 MHz Jeol NMR (model ECZ600R, Tokyo, Japan) or on a 400 MHz Jeol NMR (model ECZ400S) spectrometers using the residual CHCl3 signal (δH 7.26 ppm) and 1 13 CDCl3 (δC 77.1 ppm) as internal standards for H and C NMR, respectively. ESIMS and HRESIMS were obtained from the Bruker mass spectrometer with 7 Tesla magnets (model: SolariX FTMS system, Bremen, Germany). Column chromatography, HPLC, IR, and optical rotation were performed according to our earlier research [15].

3.2. Animal Material Specimens of J. fragilis used for this study were collected in April 2017 by self-contained underwater breathing apparatus (SCUBA) at depths of 10 15 m off the coast of South Bay, Kenting, Taiwan. − The samples were stored in a 20 C freezer until extraction. A voucher specimen was deposited in − ◦ the NMMBA (voucher no.: NMMBA-TW-GC-2017-022). Identification of the species of this organism was performed by comparison as described in previous studies [1–3].

3.3. Extraction and Isolation Sliced bodies (wet/dry weight = 795/313 g) of the coral specimen were prepared and extracted with a 1:1 mixture of methanol (MeOH) and dichloromethane (CH2Cl2) (1:1) to give a crude extract (19.0 g) which was partitioned between ethyl acetate (EtOAc) and H2O. The EtOAc extract (8.0 g) was then applied to a silica gel column chromatograph (C.C.) and eluted with gradients of n- hexane/acetone (stepwise from 50:1 to 1:2; volume ratio) to furnish 8 fractions (fractions: A H). − Fraction F was chromatographed on silica gel C.C. and eluted with gradients of n-hexane/EtOAc (2:1 to 1:2, stepwise) to furnish 4 fractions (fractions F1 F4). Fraction F3 was washed with a mixture of − n-hexane/acetone (30:1) and the undissolved 5 (23.5 mg) was obtained. Fraction G was purified by normal-phase HPLC (NP-HPLC) using a mixture of n-hexane and EtOAc (4:1 to 1:1, stepwise) to afford 16 fractions (fractions G1 G16). Afterward, fraction G15 was separated by NP-HPLC using a mixture − of CH2Cl2 and acetone (v:v = 10:1; at a flow rate = 2.0 mL/min) to yield 1 (1.7 mg). Fraction G14 was separated by NP-HPLC using a mixture of n-hexane/acetone (2:1; volume ratio) to yield 6 fractions (fractions: G14A G14F). Fractions G14C and G14D were combined and purified by reverse-phase − HPLC (RP-HPLC) using a mixture of MeOH and H2O (v:v = 65:35; at a flow rate = 4.0 mL/min) to afford 2 (0.8 mg). Fraction G9 was purified by RP-HPLC using a mixture of MeOH and H2O (v:v = 60:40; at a flow rate = 4.0 mL/min) to yield 16 fractions (fractions G9A G9P), including compound 3 (0.8 mg, − G9D). Fraction G9M was separated by RP-HPLC using a mixture of acetonitrile and H2O (v:v = 55:45; at a flow rate = 4.0 mL/min) to obtain 4 (0.8 mg).

24 1 Fragilide U (1): Amorphous powder; [α]D 12 (c 0.09, CHCl3); IR (KBr) νmax 3445, 1780, 1733 cm− ; 13 1 − C (100 MHz, CDCl3) and H (400 MHz, CDCl3) NMR data, see Table1; Table2; ESIMS: m/z 589 + [M + Na] ; HRESIMS: m/z 589.22583 (calcd. for C28H38O12 + Na, 589.22555). 23 1 Fragilide V (2): Amorphous powder; [α]D 26 (c 0.04, CHCl3); IR (KBr) νmax 3444, 1779, 1738 cm− ; 13 1 − C (100 MHz, CDCl3) and H (400 MHz, CDCl3) NMR data, see Table1; Table2; ESIMS: m/z 581 + + 35 [M + Na] , 583 [M + 2 + Na] ; HRESIMS: m/z 581.17589 (calcd. for C26H35 ClO11 + Na, 581.17601). 23 1 Fragilide W (3): Amorphous powder; [α]D 21 (c 0.04, CHCl3); IR (KBr) νmax 3459, 1777, 1738 cm− ; 13 1 − C (150 MHz, CDCl3) and H (600 MHz, CDCl3) NMR data, see Table1; Table2; ESIMS: m/z 563 + + 35 [M + Na] , 565 [M + 2 + Na] ; HRESIMS: m/z 563.16554 (calcd. for C26H33 ClO10 + Na, 563.16545). 24 Junceellonoid D (4): Amorphous powder; [α]D 18 (c 0.04, CHCl3) (ref. [5] [α]D 44.8 (c 0.10, 23 − 1−13 1 CHCl , MeOH); ref. [21] [α] 31 (c 0.05, CHCl )); IR (ATR) νmax 3509, 1793, 1733 cm ; C and H 3 D − 3 − Mar. Drugs 2019, 17, 706 8 of 9

NMR data were found to be in full agreement with those reported previously [21]; ESIMS: m/z 521 + + 35 [M + Na] , 523 [M + 2 + Na] ; HRESIMS: m/z 521.15469 (calcd. for C24H31 ClO9 + Na, 521.15488). 23 Junceellin (5): Colorless crystals; mp 277–278 ◦C (ref. [15] mp 272–275 ◦C); [α]D 3 (c 1.18, 25 1 13 1 − CHCl ) (ref. [15] [α] 2 (c 0.89, CHCl )); IR (KBr) νmax 1794, 1743 cm ; C and H NMR data were 3 D − 3 − found to be in full agreement with those reported previously [6,8,9,12]; ESIMS: m/z 605 [M + Na]+, + 35 607 [M + 2 + Na] ; HRESIMS: m/z 605.17612 (calcd. for C28H35 ClO11 + Na, 605.17601).

3.4. In Vitro Anti-Inflammatory Assay The anti-inflammatory assay was employed to evaluate the activities of briaranes 1–5 reduce the release of iNOS and COX-2 from macrophage cells as the literature reported [22–25].

4. Conclusions J. fragilis was proven to be a rich source to produce a wide structural diversity of briarane-type diterpenoids that possess various biomedical properties, particularly in anti-inflammatory activity [26,27]. In our continued study on J. fragilis, three previously unreported 11,20-epoxybriaranes, fragilides U–W (1–3), along with two known briaranes, junceellonoid D (4) and junceellin (5), were isolated. The exocyclic 11,20-epoxy group was proven to be a chemical marker for briarane-type natural products from the gorgonian corals belonging to the family Ellisellidae [28]. In the present study, the anti-inflammatory activity of 1–5 was assayed using inhibition of iNOS and COX-2 and the results indicated that fragilide W (3) and junceellin (5) showed the most potent suppressive effect on iNOS release.

Supplementary Materials: The Supplementary Materials are available online at http://www.mdpi.com/1660-3397/ 17/12/706/s1. HRESIMS, IR, 1D and 2D NMR spectra of compounds 1–4. Author Contributions: Conceptualization, Z.-H.W.; investigation, Y.-M.J., B.-R.P. and Y.-J.W.; writing—original draft preparation, T.-P.S. and C.-H.Y.; writing—review and editing, T.-Y.W., H.-W.L. and P.-J.S. Funding: This research was granted from the NMMBA; the NDHU; the Kaohsiung Armed Forces General Hospital (Grant No. 108-33); and the Ministry of Science and Technology, Taiwan (Grant Nos: MOST 108-2320- B-276-001, 104-2320-B-291-001-MY3, and 107-2320-B-291-001-MY3) awarded to Yu-Jen Wu and Ping-Jyun Sung. Conflicts of Interest: The authors declare no conflicts of interest.

References

1. Bayer, F.M. Key to the genera of octocorallia exclusive of Pennatulacea (Coelenterata: Anthozoa), with diagnoses of new taxa. Proc. Biol. Soc. Wash. 1981, 94, 902–947. 2. Bayer, F.M.; Grasshoff, M. The genus group taxa of the family Ellisellidae, with clarification of the genera established by J.E. Gray (Cnidaria: Octocorallia). Senckenb. Biol. 1994, 74, 21–45. 3. Chen, C.-C.; Chang, K.-H. Gorgonacea (Coelenterata: Anthozoa: Octocorallia) of Southern Taiwan. Bull. Inst. Zool. Acad. Sin. 1991, 30, 149–181. 4. Chung, H.-M.; Wang, Y.-C.; Tseng, C.-C.; Chen, N.-F.; Wen, Z.-H.; Fang, L.-S.; Hwang, T.-L.; Wu, Y.-C.; Sung, P.-J. Natural product chemistry of gorgonian corals of genus Junceella—Part III. Mar. Drugs 2018, 16, 339. [CrossRef][PubMed] 5. Qi, S.-H.; Zhang, S.; Wen, Y.-M.; Xiao, Z.-H.; Li, Q.-X. New briaranes from the South China Sea gorgonian Junceella fragilis. Helv. Chim. Acta 2005, 88, 2349–2354. [CrossRef] 6. Lin, Y.; Long, K. Studies of the chemical constituents of the Chinese gorgonia (IV)-Junceella, a new chlorine-containing diterpenoid from Junceella squamata. Acta Sci. Nat. Univ. Sunyatseni 1983, (2), 46–51. 7. Yao, J.; Qian, J.; Fan, H.; Shih, K.; Huang, S.; Lin, Y.; Long, K. The structure of crystal and molecule of junceellin. Acta Sci. Nat. Univ. Sunyatseni 1984, (1), 83–87. 8. Shin, J.; Park, M.; Fenical, W. The junceellolides, new anti-inflammatory diterpenoids of the briarane class from the Chinese gorgonian Junceella fragilis. Tetrahedron 1989, 45, 1633–1638. [CrossRef] 9. Subrahmanyam, C.; Kulatheeswaran, R.; Ward, R.S. Briarane diterpenes from the Indian Ocean gorgonian Gornonella umbraculum. J. Nat. Prod. 1998, 61, 1120–1122. [CrossRef] Mar. Drugs 2019, 17, 706 9 of 9

10. García, M.; Rodríguez, J.; Jiménez, C. Absolute structures of new briarane diterpenoids from Junceella fragilis. J. Nat. Prod. 1999, 62, 257–260. [CrossRef] 11. Qi, S.-H.; Zhang, S.; Huang, H.; Xiao, Z.-H.; Huang, J.-S.; Li, Q.-X. New briaranes from the South China Sea gorgonian Junceella juncea. J. Nat. Prod. 2004, 67, 1907–1910. [CrossRef][PubMed] 12. Sung, P.-J.; Fan, T.-Y.; Fang, L.-S.; Wu, S.-L.; Li, J.-J.; Chen, M.-C.; Cheng, Y.-M.; Wang, G.-H. Briarane derivatives from the gorgonian coral Junceella fragilis. Chem. Pharm. Bull. 2003, 51, 1429–1431. [CrossRef] [PubMed] 13. Sung, P.-J.; Fan, T.-Y.; Chen, M.-C.; Fang, L.-S.; Lin, M.-R.; Chang, P.-C. Junceellin and praelolide, two briaranes from the gorgonian corals Junceella fragilis and Junceella juncea (Ellisellidae). Biochem. Syst. Ecol. 2004, 32, 111–113. [CrossRef] 14. Liaw, C.-C.; Shen, Y.-C.; Lin, Y.-S.; Hwang, T.-L.; Kuo, Y.-H.; Khalil, A.T. Frajunolides E–K, briarane diterpenes from Junceella fragilis. J. Nat. Prod. 2008, 71, 1551–1556. [CrossRef][PubMed] 15. Lin, C.-C.; Chen, W.-F.; Lee, G.-H.; Wen, Z.-H.; Fang, L.-S.; Kuo, Y.-H.; Lee, C.-Y.; Sung, P.-J. Fragilides M–O, new triacetoxybriaranes from the gorgonian coral Junceella fragilis (Ellisellidae). Heterocycles 2019, 98, 984–993. 16. Sheu, J.-H.; Chen, Y.-P.; Hwang, T.-L.; Chiang, M.Y.; Fang, L.-S.; Sung, P.-J. Junceellolides J–L, 11,20- epoxybriaranes from the gorgonian coral Junceella fragilis. J. Nat. Prod. 2006, 69, 269–273. [CrossRef] [PubMed] 17. Tanaka, C.; Yamamoto, Y.; Otsuka, M.; Tanaka, J.; Ichiba, T.; Marriott, G.; Rachmat, R.; Higa, T. Briarane diterpenes from two species of octocorals, Ellisella sp. and Pteroeides sp. J. Nat. Prod. 2004, 67, 1368–1373. [CrossRef] 18. Sung, P.-J.; Chiang, M.Y.; Tsai, W.-T.; Su, J.-H.; Su, Y.-M.; Wu, Y.-C. Chlorinated briarane-type diterpenoids from the gorgonian coral Ellisella robusta (Ellisellidae). Tetrahedron 2007, 63, 12860–12865. [CrossRef] 19. Chang, J.-Y.; Liaw, C.-C.; Fazary, A.E.; Hwang, T.-L.; Shen, Y.-C. New briarane diterpenoids from the gorgonian coral Junceella juncea. Mar. Drugs 2012, 10, 1321–1330. [CrossRef] 20. Liaw, C.-C.; Lin, Y.-C.; Lin, Y.-S.; Chen, C.-H.; Hwang, T.-L.; Shen, Y.-C. Four new briarane diterpenoids from Taiwanese gorgonian Junceella fragilis. Mar. Drugs 2013, 11, 2042–2053. [CrossRef] 21. Sung, P.-J.; Wang, S.-H.; Chiang, M.Y.; Su, Y.-D.; Chang, Y.-C.; Hu, W.-P.; Tai, C.-Y.; Liu, C.-Y. Discovery of new chlorinated briaranes from Junceella fragilis. Bull. Chem. Soc. Jpn. 2009, 82, 1426–1432. [CrossRef] 22. Jean, Y.-H.; Chen, W.-F.; Sung, C.-S.; Duh, C.-Y.; Huang, S.-Y.; Lin, C.-S.; Tai, M.-H.; Tzeng, S.-F.; Wen, Z.-H. Capnellene, a natural marine compound derived from soft coral, attenuates chronic constriction injury-induced neuropathic in rats. Br. J. Pharmacol. 2009, 158, 713–725. [CrossRef][PubMed] 23. Jean, Y.-H.; Chen, W.-F.; Duh, C.-Y.; Huang, S.-Y.; Hsu, C.-H.; Lin, C.-S.; Sung, C.-S.; Chen, I.-M.; Wen, Z.-H. Inducible nitric oxide synthase and cyclooxygenase-2 participate in anti-inflammatory and analgesic effects of the natural marine compound lemnalol from Formosan soft coral Lemnalia cervicorni. Eur. J. Pharmacol. 2008, 578, 323–331. [CrossRef][PubMed] 24. Chen, L.-C.; Lin, Y.-Y.; Jean, Y.-H.; Lu, Y.; Chen, W.-F.; Yang, S.-N.; Wang, H.-M.D.; Jang, I.-Y.; Chen, I.-M.; Su, J.-H.; et al. Anti-inflammatory and analgesic effects of the marine- derived compound comaparvin isolated from the crinoid Comanthus bennetti. Molecules 2014, 19, 14667–14686. [CrossRef] 25. Oliveira, T.; Figueiredo, C.A.; Brito, C.; Stavroullakis, A.; Prakki, A.; da Silva Velozo, E.; Nogueira-Filho, G. Effect of Allium cepa L. on lipopolysaccharide-stimulated osteoclast precursor cell viability, count, and

morphology using 40,6-diamidino-2-phenylindole-staining. Int. J. Cell Biol. 2014, 2014, 535789. [CrossRef] 26. Cheng, W.; Li, X.; Yin, F.; van Ofwegen, L.; Lin, W. Halogenated briarane diterpenes with acetyl migration from the gorgonian coral Junceella fragilis. Chem. Biodiver. 2017, 14, e1700053. [CrossRef] 27. Cheng, W.; Ji, M.; Li, X.; Ren, J.; Yin, F.; van Ofwegen, L.; Yu, S.; Chen, X.; Lin, W. Fragilolides A–Q, norditerpenoid and briarane diterpenoids from the gorgonian coral Junceella fragilis. Tetrahedron 2017, 73, 2518–2528. [CrossRef] 28. Su, Y.-M.; Fan, T.-Y.; Sung, P.-J. 11,20-Epoxybriaranes from the gorgonian coral Ellisella robusta (Ellisellidae). Nat. Prod. Res. 2007, 21, 1085–1090. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).