Isolation, Characterization and Antiproliferative Activity of New Metabolites from the South African Endemic Red Algal Species Laurencia alfredensis

The marine red algae of the genus Laurencia have been widely studied for their structurally diverse and biologically active secondary metabolites. We report here the natural product investigation of the organic extract of a newly identified South African endemic species, Laurencia alfredensis. A sequence of column chromatography, preparative TLC and normal phase HPLC resulted in the isolation of eleven compounds comprising three labdane-type diterpenes (1–3), four polyether triterpenes (4–7), three cholestane-type ecdysteroids (8–10) and a glycolipid (11). Compounds 1–3, 5–8 and 10 have not previously been reported, while compound 9 is reported here for the first time from a natural source and the known compound 11 isolated for the first time from the genus Laurencia. The structural elucidation and the relative configuration assignments of the compounds were accomplished by extensive use of 1D- and 2D-NMR, HR-ESI-MS, UV and IR spectroscopic techniques, while the absolute configuration of compound 1 was determined by single-crystal X-ray diffraction analysis. All compounds were evaluated against the MDA-MB-231 breast and HeLa cervical cancer cell lines. Compound 2 exhibited low micromolar antiproliferative activity (IC50 = 9.3 µM) against the triple negative breast carcinoma and compound 7 was similarly active (IC50 = 8.8 µM) against the cervical cancer cell line.


Introduction
The family of red algae (Rhodomelaceae) comprises over 1000 recognized species globally.The genus Laurencia (Order Ceramiales) in this family has about 144 identified species [1].It is an abundant source of structurally diverse halogenated and non-halogenated secondary metabolites and thus the most investigated of the Rhodomelaceae [2][3][4].Halogenated compounds are involved in chemical defense from grazers and, in the genus Laurencia, they are contained in special cell vacuoles known as corps-en-cerise ("cherry bodies", in French), which can be seen using a compound microscope in surface cells in live material [5].These secondary metabolites include terpenoids, non-terpenoid C 15 acetogenins and indoles.The terpenoids comprise about nine different major carbon backbones of sesquiterpenes, six of diterpenes and, to date, three of polyether triterpenes [3].Isolated C 15 acetogenins, which arise from fatty acid metabolism [6], are grouped into 12 different classes as either linear molecules or cyclic ethers with varying size of the oxygenated rings present in the latter [3].Several brominated indoles have been reported from the L. brongniartii [7][8][9][10][11], L. decumbens [12] and L. similis [13][14][15] species only.
Laurencia alfredensis is a newly described species of the genus and is endemic to South Africa.The material investigated here was collected on a rocky intertidal seashore at Three Sisters, north of Port Alfred in the Eastern Cape Province of the Republic of South Africa in 2011 [16].Based on the interesting chemical diversity of secondary metabolites reported from the genus, we undertook a chemical investigation of the constituents of this new species and evaluated their antiproliferative effects against the MDA-MB-231 and HeLa cancer cell lines.Our study has led to the isolation and structural elucidation of 11 compounds, comprising eight newly reported compounds along with the known polyether triterpene saiyacenol B (4) [17], the synthetic compound 9 [18], and the glycolipid, 1,2-di-O-palmitoyl-3-O-(6-sulfo-α-D-quinovopyranosyl)-glycerol (11) [19] (Figure 1).Except for compound 4, none of the compounds isolated have previously been reported from the genus Laurencia.
Laurencia alfredensis is a newly described species of the genus and is endemic to South Africa.The material investigated here was collected on a rocky intertidal seashore at Three Sisters, north of Port Alfred in the Eastern Cape Province of the Republic of South Africa in 2011 [16].Based on the interesting chemical diversity of secondary metabolites reported from the genus, we undertook a chemical investigation of the constituents of this new species and evaluated their antiproliferative effects against the MDA-MB-231 and HeLa cancer cell lines.Our study has led to the isolation and structural elucidation of 11 compounds, comprising eight newly reported compounds along with the known polyether triterpene saiyacenol B (4) [17], the synthetic compound 9 [18], and the glycolipid, 1,2-di-O-palmitoyl-3-O-(6-sulfo-α-D-quinovopyranosyl)-glycerol (11) [19] (Figure 1).Except for compound 4, none of the compounds isolated have previously been reported from the genus Laurencia.
The HSQC-DEPT spectrum showed cross-peaks attributed to six methyl, seven diastereotopic methylene, four methine and five quaternary carbons.Four 1 H-1 H COSY spin systems (Figure 2a) were observed, comprising three similar methine proton→methylene protons→methylene protons coupling patterns and the ABX coupling pattern of the vinyl protons.Key HMBC correlations included the long range coupling of H-3 to C-4 (δ C 39.5), C-18 (δ C 17.7) and C-19 (δ C 30.6), and H-14 to C-13 (δ C 82.9) and C-16 (δ C 23.5) (Figure 2a).The chemical shifts of the methine carbon signals at δ C 69.7 and δ C 82.9 suggested the presence of the bromine atom and the acetate group at C-3 and C-13, respectively.Hence, the backbone structure of compound 1 was found to be consistent with that of a brominated labdane-type diterpene, similar to the known compounds Isoconncindiol and Pinnatol A that have both been previously isolated from the Laurencia genus [20][21][22].The presence of ROE enhancements between H2-1b, H-3 and H3-18 signified they occupied the same face of the cyclohexane ring, as did H3-19 and H3-20 (Figure 2b).Moreover, a ROESY correlation present between H3-20 and H-9 but not H-5 and H3-17 was indicative of the syn-ent configuration about the cyclohexane ring junction and the equatorial orientation of H3-17 in ring B. The structure of 1 was successfully assigned as the 13-acetyl derivative of the known C-3 brominated labdane diterpene Pinnatol A [22].
X-ray diffraction analysis of 1 confirmed the relative stereochemistry predicted from the ROESY data.Furthermore, the absolute configuration of 1 (Figure 3) was unequivocally determined via the anomalous X-ray scattering of the bromine atom.Both cyclohexane rings assume the more stable chair conformation with a syn-ent configuration at the ring junction.Protons H-1b, H-3, H-5 and H3-18 occupy the same face of ring A whilst H3-19 and H3-20 occupy the opposite face.The equatorial orientation of H3-17 compared to H3-20 supports the lack of ROESY enhancement between the two methyl groups.451.1824 for C22H37 79 BrO3Na) to 1, suggesting they were isomeric and this was indeed supported by the high degree of similarity in their NMR data.In fact, both 1D-and 2D-NMR data acquired for compound 2, except for the ROESY data, were congruent with the NMR spectra for 1.However, subtle differences in the proton chemical shifts of H-9 (δH 0.95, m), H-20 (δH 1.09, s), H-17 (δH 1.43, s) and H2-11 (δH 1.28, m; 1.71, m) (Table 1) were observed in the 1 H-NMR spectrum of 2 compared to 1.The presence of ROESY cross-peaks from H3-20 to H-9 and H3-17 confirmed a similar spatial orientation of these protons and the absence of such a correlation between H3-20 and H-5, implied a similar configuration about the ring junction as observed in 1 (Figure 4a).The exact same chemical The presence of ROE enhancements between H 2 -1b, H-3 and H 3 -18 signified they occupied the same face of the cyclohexane ring, as did H 3 -19 and H 3 -20 (Figure 2b).Moreover, a ROESY correlation present between H 3 -20 and H-9 but not H-5 and H 3 -17 was indicative of the syn-ent configuration about the cyclohexane ring junction and the equatorial orientation of H 3 -17 in ring B. The structure of 1 was successfully assigned as the 13-acetyl derivative of the known C-3 brominated labdane diterpene Pinnatol A [22].
X-ray diffraction analysis of 1 confirmed the relative stereochemistry predicted from the ROESY data.Furthermore, the absolute configuration of 1 (Figure 3) was unequivocally determined via the anomalous X-ray scattering of the bromine atom.Both cyclohexane rings assume the more stable chair conformation with a syn-ent configuration at the ring junction.Protons H-1b, H-3, H-5 and H 3 -18 occupy the same face of ring A whilst H 3 -19 and H 3 -20 occupy the opposite face.The equatorial orientation of H 3 -17 compared to H 3 -20 supports the lack of ROESY enhancement between the two methyl groups.The presence of ROE enhancements between H2-1b, H-3 and H3-18 signified they occupied the same face of the cyclohexane ring, as did H3-19 and H3-20 (Figure 2b).Moreover, a ROESY correlation present between H3-20 and H-9 but not H-5 and H3-17 was indicative of the syn-ent configuration about the cyclohexane ring junction and the equatorial orientation of H3-17 in ring B. The structure of 1 was successfully assigned as the 13-acetyl derivative of the known C-3 brominated labdane diterpene Pinnatol A [22].
X-ray diffraction analysis of 1 confirmed the relative stereochemistry predicted from the ROESY data.Furthermore, the absolute configuration of 1 (Figure 3) was unequivocally determined via the anomalous X-ray scattering of the bromine atom.Both cyclohexane rings assume the more stable chair conformation with a syn-ent configuration at the ring junction.Protons H-1b, H-3, H-5 and H3-18 occupy the same face of ring A whilst H3-19 and H3-20 occupy the opposite face.The equatorial orientation of H3-17 compared to H3-20 supports the lack of ROESY enhancement between the two methyl groups.451.1824 for C22H37 79 BrO3Na) to 1, suggesting they were isomeric and this was indeed supported by the high degree of similarity in their NMR data.In fact, both 1D-and 2D-NMR data acquired for compound 2, except for the ROESY data, were congruent with the NMR spectra for 1.However, subtle differences in the proton chemical shifts of H-9 (δH 0.95, m), H-20 (δH 1.09, s), H-17 (δH 1.43, s) and H2-11 (δH 1.28, m; 1.71, m) (Table 1) were observed in the 1 H-NMR spectrum of 2 compared to 1.The presence of ROESY cross-peaks from H3-20 to H-9 and H3-17 confirmed a similar spatial orientation of these protons and the absence of such a correlation between H3-20 and H-5, implied a similar configuration about the ring junction as observed in 1 (Figure 4a).The exact same chemical   1) were observed in the 1 H-NMR spectrum of 2 compared to 1.The presence of ROESY cross-peaks from H 3 -20 to H-9 and H 3 -17 confirmed a similar spatial orientation of these protons and the absence of such a correlation between H 3 -20 and H-5, implied a similar configuration about the ring junction as observed in 1 (Figure 4a).The exact same chemical shift observed for H 3 -16 (δ H 1.52, s) in compounds 1-3 indicated that all three labdane diterpenes have the same 13S configuration, consistent with the absolute configuration reported for the structurally related marine natural product, isoconncindiol [20,21].The structure of 2 was therefore elucidated as the 8S-diastereomer of 1, that is with the only difference observed at the stereogenic center C-8, thus confirming the structure of 2 as the 13-acetyl derivative of isoconcinndiol.

Structural Elucidation of Polyether Triterpenes (4-7)
Compound 4 was isolated as a white amorphous solid.The 1 H-NMR spectrum showed signals of eight methyl groups (δH 1.07, s; 1.10, s; 1.13, s; 1.16, s; 1.18, s; 1.19, s; 1.26, s; 1.38, s), several overlapping multiplicities at δH 1.3-2.3,and six methine proton groups (Table 2).The 13 C-NMR spectrum showed 30 carbon signals and a high degree of oxygenation was inferred from 10 carbon resonances present in the downfield range δC 70-87 ppm.A signal observed at δC 59.0 suggested the presence of a bromo-methine C-3, characteristic in similar polyether triterpenes reported from the genus Laurencia [24,25].This was confirmed by the presence of the mono-isotopic bromine peaks at m/z 611/609 [M + Na] + , 571/569 [M − OH] + recorded in the LR-ESI-MS and consistent with the reported values by Cen-Pacheco and co-workers [17].Compound 4 was deduced to be a congener of this group of secondary metabolites and the structure was found be that of the known compound saiyacenol B, previously isolated from L. viridis [17].
After careful examination of the 1D NMR spectra of a fraction obtained from the same HPLC run that yielded compound 2, we identified the presence of a mixture of the two very closely related compounds 5 and 6.However, due to the paucity of the HPLC fraction obtained (0.5 mg), we were unable to separate these two compounds any further.The HR-ESI-MS spectrum obtained for compound 5 established a molecular formula of C30H50O6, requiring six degrees of unsaturation.Its 1 H-NMR spectrum exhibited signals of seven methyls (δH 1.08, s; 1.10, s; 1.13, s; 1.15, s; 1.19, s; 1.20, s; Compound 3 showed signals for five methyl singlets (δ H 0.95; 0.96; 1.05; 1.52; 2.00), a methyl doublet (δ H 0.86, d, 6.6 Hz, H 3 -17), several overlapping multiplets at δ H 1.00-2.15,the vinyl protons double doublets (δ H 5.12, dd, J = 0.7, 6.6 Hz, H-15a; 5.14, dd, J = 0.7, 13.1 Hz, H-15b; 5.90, dd, J = 11.0,17.5 Hz, H-14) and the "halo-methine" proton H-3 at δ H 4.00 (dd, J = 4.3, 12.4 Hz) in its 1 H-NMR spectrum (Table 1).Four different 1 H-1 H COSY spin systems were observed (Figure 4b), with two of them (H 2 -1→H-3 and H-14→H 2 -15) similar to those also observed in compounds 1 and 2. However, the presence of the significant H-5→H 3 -17 (Figure 4b), concomitant with the absence of the H-9→H 2 -12 COSY spin system and the splitting of H 3 -17 into a doublet, suggested that 3 was a positional isomer of 1 and 2, with its hydroxyl group present at C-9 (δ C 76.8) rather than C-8 (δ C 35.8).The relative configuration of the groups on the cyclohexane ring A was retained whilst no ROE cross-peak was observed between H 3 -17 and H 3 -20.The structure of 3 was therefore elucidated as the 13-acetyl derivative of concinndiol [23].

Structural Elucidation of Polyether Triterpenes (4-7)
Compound 4 was isolated as a white amorphous solid.The 1 H-NMR spectrum showed signals of eight methyl groups (δ H 1.07, s; 1.10, s; 1.13, s; 1.16, s; 1.18, s; 1.19, s; 1.26, s; 1.38, s), several overlapping multiplicities at δ H 1.3-2.3,and six methine proton groups (Table 2).The 13 C-NMR spectrum showed 30 carbon signals and a high degree of oxygenation was inferred from 10 carbon resonances present in the downfield range δ C 70-87 ppm.A signal observed at δ C 59.0 suggested the presence of a bromo-methine C-3, characteristic in similar polyether triterpenes reported from the genus Laurencia [24,25].This was confirmed by the presence of the mono-isotopic bromine peaks at m/z 611/609 [M + Na] + , 571/569 [M − OH] + recorded in the LR-ESI-MS and consistent with the reported values by Cen-Pacheco and co-workers [17].Compound 4 was deduced to be a congener of this group of secondary metabolites and the structure was found be that of the known compound saiyacenol B, previously isolated from L. viridis [17].
After careful examination of the 1D NMR spectra of a fraction obtained from the same HPLC run that yielded compound 2, we identified the presence of a mixture of the two very closely related compounds 5 and 6.However, due to the paucity of the HPLC fraction obtained (0.5 mg), we were unable to separate these two compounds any further.The HR-ESI-MS spectrum obtained for compound 5 established a molecular formula of C 30 H 50 O 6 , requiring six degrees of unsaturation.Its 1 H-NMR spectrum exhibited signals of seven methyls (δ H 1.08, s; 1.10, s; 1.13, s; 1.15, s; 1.19, s; 1.20, s; 1.68, s), several overlapping multiplicities at δ H 1.2-2.2,six oxygenated methine protons double doublets and two olefinic geminal protons at δ H 4.77 (m, H-1a) and 4.98 (m, H-1b) (Table 2).The 13 C-NMR spectrum showed the presence of 30 non-equivalent carbons and the HSQC-DEPT confirmed the presence of seven methyl, 11 diastereotopic methylene, six methine and six quaternary carbons.Thorough analysis of the 1D and 2D NMR data identified compound 5 as another polyether triterpene, which lacked a C-3 bromine atom and one methyl group (δ H 1.26, s, H 3 -1) compared to 4. With evidence of only one double bond (C-1, δ H 4.77/4.98,δ C 110.2; C-2, δ C 146.2), compound 5 was deduced to be pentacyclic.
Six 1 H-1 H COSY spin systems were observed, of which four exhibited the concomitant vicinal AA'BB' coupling pattern involving two methylene groups, with one of the groups further involved in an ABX vicinal coupling with a methine proton (Figure 5a).The coupling constants recorded, especially for the methine protons supported the presence of both substituted oxolane and oxane moieties.After careful analysis of the long range 1 H- 13 C correlations, the rest of the molecule was found to be similar to 4, apart from the change in chemical shifts observed from C-1 through C-7 and C-25.An HMBC cross-peak between H-3 and C-1 and the downfield shift of C-3 to δ C 83.4 suggested the absence of a bromine atom to give the exocyclic 2-methyl ethylenyl group on the oxolane ring in compound 5, which is consistent with similar isolated compounds [26].ROESY enhancements of H-14 and H-18 by H 3 -28 and H 3 -29, respectively, led to a similar relative configuration for 5 as reported for 4 [17] (Figure 5b).Compound 5 is reported for the first time here as Alfredensinol A. Six 1 H-1 H COSY spin systems were observed, of which four exhibited the concomitant vicinal AA'BB' coupling pattern involving two methylene groups, with one of the groups further involved in an ABX vicinal coupling with a methine proton (Figure 5a).The coupling constants recorded, especially for the methine protons supported the presence of both substituted oxolane and oxane moieties.After careful analysis of the long range 1 H- 13 C correlations, the rest of the molecule was found to be similar to 4, apart from the change in chemical shifts observed from C-1 through C-7 and C-25.An HMBC cross-peak between H-3 and C-1 and the downfield shift of C-3 to δC 83.4 suggested the absence of a bromine atom to give the exocyclic 2-methyl ethylenyl group on the oxolane ring in compound 5, which is consistent with similar isolated compounds [26].ROESY enhancements of H-14 and H-18 by H3-28 and H3-29, respectively, led to a similar relative configuration for 5 as reported for 4 [17] (Figure 5b).Compound 5 is reported for the first time here as Alfredensinol A. The NMR data obtained for compound 6 were superimposable with that of 5, signifying their structures were closely related (Figure 1  The NMR data obtained for compound 6 were superimposable with that of 5, signifying their structures were closely related (Figure 1), with the exception of the olefinic geminal protons of C-1 resonating downfield at δ were observed, including the key correlations involving the olefinic proton H-16, the diastereotopic methylene protons H 2 -17 and the oxymethine H-18 (Figure 6a).Key HMBC cross-peaks observed from H-16 to C-14, C-15, C-18, and C-28 established the ∆ 15,16 olefinic functionality.HMBC correlations from H-22 ( 3 J C, H ) and H 3 -29 ( 3 J C, H ) to C-18, along with key COSY correlations between H 2 -20, H 2 -21 and H-22, confirmed the presence of an oxane ring in the right hand side chain of 7. The absence of ROE enhancement of H-16 by H 3 -28 was indicative of the trans configuration about the double bond between C-15 and C-16, which was further supported by the ROESY correlation between H-14 and H-16 (Figure 6b).The relative configuration at C-18 was established as a result of the observed ROESY cross-peak from H-18 to H-16 and the absence of one to H 3 -29.The absence of an ROE correlation between H-22 and H 3 -29, implied axial orientations for both these groups on the oxane ring.Thus, the structure of 7 was elucidated and identified as a new compound which is named here as Alfredensinol C.
the double bond between C-15 and C-16, which was further supported by the ROESY correlation between H-14 and H-16 (Figure 6b).The relative configuration at C-18 was established as a result of the observed ROESY cross-peak from H-18 to H-16 and the absence of one to H3-29.The absence of an ROE correlation between H-22 and H3-29, implied axial orientations for both these groups on the oxane ring.Thus, the structure of 7 was elucidated and identified as a new compound which is named here as Alfredensinol C.

Structural Elucidation of Cholestane-Type Ecdysteroids (8-10)
Compound 8 was isolated as a white amorphous solid and HR-ESI-MS data obtained established the molecular formula as C 31 H 48 O 6 .The 1 H-NMR spectrum showed the presence of seven methyl singlets of which two were acetate methyls at δ H 2.03 (s, H 3 -30) and δ H 2.04 (s, H 3 -31), and five oxymethine protons.Several overlapping multiplets were observed in the methylene envelope between δ H 1.11-2.15(Table 3).The  3).The HSQC-DEPT data indicated the presence of seven methyl, eight diastereotopic methylene, 10 methine, and six quarternary carbons.Two 1 H-1 H COSY spin systems were observed, from H 2 -1 to H-5 and H-7 to H 3 -27 (Figure 7a).Some characteristic aspects of a steroidal skeleton became evident from the 1 H, 13 C and COSY NMR data; for example, the 1 H chemical shifts for H 3 -18 and H 3 -19, and the splitting of H 3 -21, H 3 -26 and H 3 -27 into doublets were characteristic of the side chain to C-17 for cholesterol.The chemical shifts of C-7 and C-8, together with the chemical shift of H-7 (δ H 5.75, m) and its COSY correlation to H-9 (δ H 2.25, ddd, J = 2.5, 6.9, 9.9) and H-14 (δ H 2.05, m) suggested the ketone carbonyl was adjacent to the C-7-C-8 double bond at C-6 to give the endocyclic α, β-unsaturated ketone, a structural feature of ecdysteroids.HMBC cross-peaks from  Key NOESY enhancements together with 13 C comparisons enabled the assignment of the relative configuration of 8 (Figure 7b).The chair conformation and trans-fused A/B ring junction were established from two 1,3-diaxial NOESY correlations from H3-19β to H3-30β (OAc) and H-4β (δH 1.93) respectively, and a further two 1,3-diaxial NOESY correlations from H-5 to H-3 and H-9, where the latter three protons are all in the same α-orientation.The trans-ring-fusion between rings A and B was further corroborated by the absence of a NOESY cross peak between the methine proton H-5 and H3-19.The presence of correlation between H3-18 and H3-19 but not H3-21 established the β-configuration of H3-18 and H3-19 to the steroid backbone suggesting a chair-chair-chair conformation of rings A, B and C. The relative configurations at C-17, C-20, C-21 and C-22 were assigned, by comparison of the 13 C-NMR data with that of the structurally related 22R hydroxyl cholesterol [27].The structure of 8 was determined as a new compound and named Alfredensterol.
Compound 9 was isolated as a white amorphous solid and its molecular formula was found to be C31H48O7 from the HR-ESI-MS ion peak at m/z 533.3488 (calcd.533.3478).The 1 H-and 13 C-NMR spectroscopic data recorded for compound 9 were virtually identical to compound 8 and suggested a closely related cholestane-type ecdysteroid (Table 3).The two acetate methyl singlets (δH 2.04, s, H3-30; δH 2.05, s, H3-31), together with the signals for H3-19, H3-26 and H3-27 for 9 were in close correspondence (±0.02 ppm) with the chemical shifts for 8, while a downfield shift to δH 0.76 was observed for H3-18.The HSQC-DEPT spectrum revealed one less methine proton and one additional quaternary carbon at δC 96.1, indicating the likely presence of a tertiary alcohol functionality at C-14 and therefore accounting for the extra O-atom in the molecular formula of 9.Moreover, the 1 H-1 H COSY spectrum exhibited three spin systems from H2-1 to H-5, H-7 to H2-12 and H2-15 to H3-27, thus revealing the absence of H-14.The relative configuration of compound 9 was assigned primarily using 13 C and ROESY NMR data, and were found to be similar to those observed in 8. To the best of our knowledge, there is only one report of 9 in the literature as a by-product in the synthesis of ecdysone related compounds, in which no extensive NMR assignments could be found, including  Key NOESY enhancements together with 13 C comparisons enabled the assignment of the relative configuration of 8 (Figure 7b).The chair conformation and trans-fused A/B ring junction were established from two 1,3-diaxial NOESY correlations from H 3 -19β to H 3 -30β (OAc) and H-4β (δ H 1.93) respectively, and a further two 1,3-diaxial NOESY correlations from H-5 to H-3 and H-9, where the latter three protons are all in the same α-orientation.The trans-ring-fusion between rings A and B was further corroborated by the absence of a NOESY cross peak between the methine proton H-5 and H 3 -19.The presence of correlation between H 3 -18 and H 3 -19 but not H 3 -21 established the β-configuration of H 3 -18 and H 3 -19 to the steroid backbone suggesting a chair-chair-chair conformation of rings A, B and C. The relative configurations at C-17, C-20, C-21 and C-22 were assigned, by comparison of the 13 C-NMR data with that of the structurally related 22R hydroxyl cholesterol [27].The structure of 8 was determined as a new compound and named Alfredensterol.
Compound 9 was isolated as a white amorphous solid and its molecular formula was found to be C 31 H 48 O 7 from the HR-ESI-MS ion peak at m/z 533.3488 (calcd.533.3478).The 1 H-and 13 C-NMR spectroscopic data recorded for compound 9 were virtually identical to compound 8 and suggested a closely related cholestane-type ecdysteroid (Table 3).The two acetate methyl singlets (δ H 2.04, s, H 3 -30; δ H 2.05, s, H 3 -31), together with the signals for H 3 -19, H 3 -26 and H 3 -27 for 9 were in close correspondence (±0.02 ppm) with the chemical shifts for 8, while a downfield shift to δ H 0.76 was observed for H 3 -18.The HSQC-DEPT spectrum revealed one less methine proton and one additional quaternary carbon at δ C 96.1, indicating the likely presence of a tertiary alcohol functionality at C-14 and therefore accounting for the extra O-atom in the molecular formula of 9.Moreover, the 1 H-1 H COSY spectrum exhibited three spin systems from H 2 -1 to H-5, H-7 to H 2 -12 and H 2 -15 to H 3 -27, thus revealing the absence of H-14.The relative configuration of compound 9 was assigned primarily using 13 C and ROESY NMR data, and were found to be similar to those observed in 8. To the best of our knowledge, there is only one report of 9 in the literature as a by-product in the synthesis of ecdysone related compounds, in which no extensive NMR assignments could be found, including 13 C-NMR data [18].Therefore, this is the first report of 9 from a natural source, which we have named 14α-hydroxy Alfredensterol.
Compound 10 was isolated as a white amorphous solid.It recorded a HR-ESI-MS ion peak of m/z 475.3432 [M + H] + (calcd.475.3423) for the molecular formula C 29 H 46 O 5 .The 1 H-and 13 C-NMR data for 10 were very similar to both compounds 8 and 9 described above.However, key noticeable differences in the 1 H-NMR data of 10 were observed, including the absence of an acetate methyl singlet and the upfield chemical shifts of H-2 (δ H 4.88, m) and H-3 (δ H 4.01, m) compared to 8. The 13 C-NMR data of 10 showed corresponding changes (δ C 71.3, C-2; 66.4, C-3) consistent with its 1 H-NMR.The 1 H-1 H COSY spin systems were similarly consistent with those observed in compound 8.The position of the single acetate group on ring A was assigned to C-2, on the basis of the chemical shifts of C-2 and C-3 (Table 3), and HMBC cross-peaks from H-2 to C-3, C-4, C-10 and C-28, and from H-3 to C-1, C-2 and C-5.Analysis of the ROESY data revealed the same relative configuration of 10 as observed in 8 and 9. Compound 10 has not previously been reported and is named here as 3-deacetoxy Alfredensterol.
The position of the single acetate group on ring A was assigned to C-2, on the basis of the chemical shifts of C-2 and C-3 (Table 3), and HMBC cross-peaks from H-2 to C-3, C-4, C-10 and C-28, and from H-3 to C-1, C-2 and C-5.Analysis of the ROESY data revealed the same relative configuration of 10 as observed in 8 and 9. Compound 10 has not previously been reported and is named here as 3-deacetoxy Alfredensterol.

Antiproliferative Activity Results
Compounds 1-11 were evaluated for their antiproliferative activity against MDA-MB-231 triple negative human breast carcinoma and HeLa human cervical carcinoma (Figure 9).Generally, all the compounds displayed antiproliferative activity in the mid-to-low micromolar range against the two cells lines tested.
To the best of our knowledge, brominated labdane diterpenes previously isolated from the genus Laurencia have not been investigated for their antiproliferative activity.However, non-halogenated labdanes have been reported to exhibit moderate to weak cytotoxic activity against various cancer cell lines [28][29][30].In this study, the brominated labdane-type diterpenes 1-3 were all found to be moderately active against both cancer cell lines tested, with the exception of Isoconcinndiol 13-acetate 2, the 8S-diastereomer of 1, which showed the highest antiproliferative activity against the HeLa cancer cell line (IC50 9.3 ± 1.3 µM), not only in this class of compounds but amongst all ten compounds tested in this study (Figure 9), indicating that the stereochemistry at C-8 in 2 could be playing a role in the observed biological activity.
Brominated polyether triterpenes of the [4.4.0] class have been reported to have good to moderate cytotoxic activity [31].The antiproliferative activity against the HeLa cell line recorded for the known compound 4, saiyacenol B, in our study was found to be within the previously reported values [17].To the best of our knowledge, this current work is the first report of the antiproliferative effects of this class of polyether triterpenes against the MDA-MB-231 breast cancer cell line, with the

Antiproliferative Activity Results
Compounds 1-11 were evaluated for their antiproliferative activity against MDA-MB-231 triple negative human breast carcinoma and HeLa human cervical carcinoma (Figure 9).Generally, all the compounds displayed antiproliferative activity in the mid-to-low micromolar range against the two cells lines tested.
To the best of our knowledge, brominated labdane diterpenes previously isolated from the genus Laurencia have not been investigated for their antiproliferative activity.However, non-halogenated labdanes have been reported to exhibit moderate to weak cytotoxic activity against various cancer cell lines [28][29][30].In this study, the brominated labdane-type diterpenes 1-3 were all found to be moderately active against both cancer cell lines tested, with the exception of Isoconcinndiol 13-acetate 2, the 8S-diastereomer of 1, which showed the highest antiproliferative activity against the HeLa cancer cell line (IC 50 9.3 ± 1.3 µM), not only in this class of compounds but amongst all ten compounds tested in this study (Figure 9), indicating that the stereochemistry at C-8 in 2 could be playing a role in the observed biological activity.
Brominated polyether triterpenes of the [4.4.0] class have been reported to have good to moderate cytotoxic activity [31].The antiproliferative activity against the HeLa cell line recorded for the known compound 4, saiyacenol B, in our study was found to be within the previously reported values [17].To the best of our knowledge, this current work is the first report of the antiproliferative effects of this class of polyether triterpenes against the MDA-MB-231 breast cancer cell line, with the new compound 7, alfredensinol C, exhibiting the highest activity (IC 50 8.8 ± 5.6 µM) in this particular class of compounds and also amongst all compounds tested (Figure 9).S1).
The antiproliferative activity recorded for the ecdysteroids 8-10 was consistent with biological activity data previously reported for both natural and semi-synthetic congeners which have been shown to exhibit moderate antiproliferative or cytotoxic activity [32].In this particular class of compounds, the new alfredensterol 8 was found to be the most active against the HeLa cervical cancer cell line (IC50 25.6 ± 1.2 µM), whilst its 14α-hydroxyl derivative 9 exhibited the best antiproliferative effect (IC50 15.8 ± 1.1 µM) against the MDA-MB-231 breast cancer cell line (Figure 9).
In summary, the new compounds 2 and 7 exhibited the best antiproliferative activity against the HeLa and MDA-MB-231 cancer cell lines, respectively, while 11 was found to be non-toxic to either cell line.
The antiproliferative activity recorded for the ecdysteroids 8-10 was consistent with biological activity data previously reported for both natural and semi-synthetic congeners which have been shown to exhibit moderate antiproliferative or cytotoxic activity [32].In this particular class of compounds, the new alfredensterol 8 was found to be the most active against the HeLa cervical cancer cell line (IC 50 25.6 ± 1.2 µM), whilst its 14α-hydroxyl derivative 9 exhibited the best antiproliferative effect (IC 50 15.8 ± 1.1 µM) against the MDA-MB-231 breast cancer cell line (Figure 9).In summary, the new compounds 2 and 7 exhibited the best antiproliferative activity against the HeLa and MDA-MB-231 cancer cell lines, respectively, while 11 was found to be non-toxic to either cell line.

Plant Material
The species studied here has only recently been described [33].It was previously known in South Africa as Laurencia elata (C.Agardh) Hooker & Harvey.The latter is an Australian species, which has recently been placed in a new genus as Coronaphycus elatus (C.Agardh) Metti [34].The South African material has been described as a new endemic species Laurencia alfredensis [33].It is the same material previously described as "Laurencia cf.elata", and used in 1 H-NMR profiling of crude organic extracts as an identification tool for nine species of South African Laurencia [16].

X-ray Crystallographic Data
The crystal for compound 1 was obtained by recrystallization from methanol and its structure was solved by direct methods from intensity data collected from the crystal specimen at 173(2) K on a Bruker Apex II Duo diffractometer and refined by full-matrix least-squares.The Flack x parameter value of −0.013 (6), indicating that the correct absolute configuration had been assigned, was determined using Further technical details of the structure determination and molecular parameters are provided in the CIF file (Supplementary Materials).

Figure 3 .
Figure 3. Stereoscopic view of 1 (absolute configuration).Non-H atoms are drawn as thermal ellipsoids at the 40% probability level and H atoms as spheres of arbitrary size.(Color code: C gray, H white, O red, Br yellow.)

Figure 3 .
Figure 3. Stereoscopic view of 1 (absolute configuration).Non-H atoms are drawn as thermal ellipsoids at the 40% probability level and H atoms as spheres of arbitrary size.(Color code: C gray, H white, O red, Br yellow.)

Figure 3 .
Figure 3. Stereoscopic view of 1 (absolute configuration).Non-H atoms are drawn as thermal ellipsoids at the 40% probability level and H atoms as spheres of arbitrary size.(Color code: C gray, H white, O red, Br yellow.)
H 5.23 (m, H-1a) and δ H 5.30 (m, H-1b) due to the presence of the hydroxyl group on the methylene carbon C-25 (δ C 44.7).Moreover, C-1 occurred downfield at δ C 114.7, while C-3 moved upfield to δ C 80.7 (Table 2).The HR-ESI-MS ion peak at m/z 545.3457 [M + Na] + (calcd.545.3454) obtained for 6 established a molecular formula of C 30 H 50 O 7 , implying an additional O-atom compared to 5. Therefore, compound 6 is reported here as the new C-25 hydroxyl derivative of compound 5, Alfredensinol B. Compound 7 was isolated as a white amorphous solid.It recorded a HR-ESI-MS ion peak of m/z 507.3693 [M + H] + (calcd.507.3686) for the molecular formula C 30 H 50 O 6 and was identified as another polyether triterpene from its 1 H-and 13 C-NMR spectra.Six oxygenated methine doublet of doublets including an olefinic proton (δ H 5.52, dd, J = 2.7, 13.8 Hz, H-16) and similar olefinic geminal protons (δ H 4.77, m, H-1a; 4.98, m, H-1b) to compound 5 (Table 2) were evident in the 1 H-NMR spectrum.With six degrees of unsaturation inferred from the molecular formula, compound 7 was deduced to be a tetracyclic molecule with the presence of two double bonds.Six 1 H-1 H COSY spin systems Molecules 2017, 22, 513 7 of 16

Molecules 2017, 22 , 513 11 of 16 Figure 9 .
Figure 9. Results from the antiproliferative activity evaluation of compounds 1-10 against human breast (MDA-MB-231) and cervical (HeLa) cancer cell lines.(The standard errors of the mean IC50 values and correlation coefficients can be found in Supplementary Materials, TableS1).

Figure 9 .
Figure 9. Results from the antiproliferative activity evaluation of compounds 1-10 against human breast (MDA-MB-231) and cervical (HeLa) cancer cell lines.(The standard errors of the mean IC 50 values and correlation coefficients can be found in Supplementary Materials, TableS1).

Table 1
established a molecular formula of C 22 H 37 BrO 3 , signifying the presence of four degrees of unsaturation.The 1 H-NMR spectrum exhibited signals of six sharp methyl groups (δ H 0.95, s; 1.06, s; 1.20, s; 1.26, s; 1.52, s) including a characteristic acetate methyl singlet at δ H 2.00 (H 3 -22), several overlapping multiplets at δ H 1.00-2.35characteristic of a terpenoid backbone, three methine signals with a coupling pattern indicative of the presence of a vinyl group (δ H 5.13, dd, J = 0.9, 17. data