Tingyu Wen, Junyan Chu, Wanqing Cheng, Yingying Fu, Feixia Hu, Ruige Yang, Yong Guo*, Yanbing Zhang, Jifeng Liu*
ABSTRACT
Natural products (NPs) are very important sources for the development of new drugs. Merrillianone and cycloparvifloralone, isolated from the roots, stems, and fruits of Illicium henryi Diels, are two natural sesquiterpene compounds. In continuation of our effort to discovery more effective neurotrophic compounds from NPs, a series of novel merrillianone/cycloparviforalone based esters 2a–i, 3a–g and 3i–q were prepared and their structures were characterized by 1H NMR, 13C NMR and IR spectral analyses. Furthermore, the spatial structure of compound 2h was unambiguously confirmed by X-ray crystallography. The neurite outgrowth-promoting activity results indicated that most of the target derivatives exhibited more potent neurite outgrowth- promoting activity than merrillianone and cycloparviforalone. Among all target derivatives, the neurite outgrowth-promoting activity of compounds 2a, 3a and 3b was about 2-fold stronger than that of their pre- cursors merrillianone and cycloparviforalone, respectively. Besides, compounds 2a and 3a displayed relatively low cytotoxicity to normal GES-1 cells. Moreover, these derivatives had good hydrolytic stability. Finally, some interesting results of the structure–activity relationships (SARs) were also discussed. This work will pave the way for the development of merrillianone/cycloparviforalone derivatives as potential neurotrophic agents.
Keywords:Natural product;Merrillianone;Cycloparviforalone;Neurite outgrowth-promoting activity;Structure–activity relationship
Currently, as populations age increases, more and more old people are suffering from the Alzheimer ’s disease (AD).1 It is a progressive neurodegenerative disease that impairs the patient ’s memory and other important cognitive functions.2,3 However, there are no effective methods to address this problem in clinical. At present, neuronal injury and synaptic loss have been considered to be the main features of the pathophysiology of AD.4,5 Therefore, regulation of neuronal survival and neurite outgrowth may be an efficient strategy for the treatment of AD, which is becoming a hot research topic.6–8 Neurotrophins such as neurotrophin-3 (NT-3), neurotrophin-4/5 (NT-4/5), and nerve growth factor (NGF) are a family of proteins that play important roles in neuron survival, outgrowth, and synaptic connectivity.9,10 Thus, they are considered to be therapeutic potential agents for the treatment of neu- rodegeneration. However, due to high molecular weight of these poly- peptide neurotrophins and easy metabolism by peptidases under physiological conditions, they can ’t cross the brain–blood barrier.11,12 Hence, small natural molecule modulators of nerve growth factor (NGF) might be promising candidates for the treatment of neurodegenerative diseases.13–15
Merrillianone (1, Fig.1) and cycloparvifloralone (2, Fig. 1), isolated from the roots, stems, and fruits of Illicium henryi Diels, are two natural sesquiterpene compounds.5,16,17 Previous research reported that com- pounds 1, 2 and their analogs had good neurotrophic activity.18-21 However, to the best of our knowledge, little attention has been paid to further structural modification of merrillianone and cycloparvifloralone for developing potential neurotrophic agents. Based on the above mentioned, and in our ongoing efforts to discover more effective neu- rotrophic compounds from NPs, herein, a series of novel merrillianone/ cycloparviforalone based esters 2a–i, 3a–g and 3i–q were prepared and their structures were characterized by 1H NMR, 13C NMR and IR. Among all target derivatives, the structure of compound 2h was further confirmed by X-ray crystallography. The neurite outgrowth-promoting activity results indicated that most of the target derivatives exhibited more potent neurite outgrowth-promoting activity than their pre- cursors. In particular, compounds 2a, 3a and 3b showed about 2-fold enhancement in neurite outgrowth-promoting activity
Fig. 1. Illicium henryi Diels and structures of merrillianone (1) and cycloparvifloralone (2).
Scheme 1. Synthetic route of merrillianone/cycloparviforalone based esters (2a–i, 3a–g and 3i–q)merrillianone and cycloparviforalone, respectively.Some interesting results of the structure–activity relationships(SARs)were also discussed.As displayed in Scheme 1, merrillianone/cycloparviforalone reacted with different carboxylic acids in the presence of N, N ’-dicyclohex- ylcarbodiimide (DCC) and Chidamide chemical structure 4-dimethylaminopyridine (DMAP) to obtain title derivatives 2a–i, 3a–g and 3i–q. To our delight, we found that hydroxyl groups in different positions of merrillianone/cycloparvifor- alone have regioselectivity, and the carboxylic acids only can react with specific hydroxyl groups. For example, when merrillianone reacted with the corresponding acid, only hydroxyl group at C-10 position of mer- rillianone can react with acids; while cycloparviforalone reacted with the corresponding acid, only hydroxyl group at C-7 position of cyclo- parviforalone can react with acids, but other hydroxyl groups can ’t react
with the corresponding acid. This result was proved by comparison of partial 1H NMR spectra of compounds 1, 2, 2h and 3k. As shown in Fig. 2, when the acyloxy was introduced at C-10 position of compound 1, the chemical shifts of H-10 and H-11 were faecal immunochemical test increased from 3.77 and 5.14 ppm in compound 1 to 4.88 and 5.53 ppm in compound 2h, respec- tively, because of the electron withdrawing effect of acyloxy group. Whereas, compared to the compound 1, due to the steric effects of H-1′′ and H-2′′ to H-15 of compound 2h, the chemical shift of H-15 of 2h was shifted from 1.19 ppm to 0.90 ppm. But for hydrogen at other positions of compound 2h, there is no significant change for their chemical shifts. Similarly, when the acyloxy introduced at C-7 position of compound 2, the chemical shifts of H-8 and H-11 were shifted to the low field. Moreover, the X-ray crystal structure (Fig. 3) of compound 2h further demonstrated this conclusion. Finally, the structures of all target
Fig. 2. Comparison of partial 1H NMR spectra of compounds 1, 2, 2h and 3k.
Fig. 3. X-ray crystal structure of compound 2h.
merrillianone/cycloparviforalone derivatives were characteried by various spectrum analyses, such as IR, 1H NMR, 13C NMR and ESI-MS, and spectra data of all target derivatives 2a-i, 3a-g and 3i-q were lis- ted in the Supplementary Material.The crystal structure data for 2h: C24H30O6,Mr= 414.48, orthorhombic, 0.18 mm X 0.15 mm X 0.1 mm, space group I121 (no. 5), a = 8.9722(4) Å, b = 9.7642(5) Å, c = 24.0808(10) Å, α = β = γ = 90o, V = 2084.39(17) Å3, Z = 4, D calc = 1.321 g.cm – 3, F000 = 888, Xcalibur, Eos, Gemini, Cu Kα radiation, λ = 1.54184 Å, T = 293(2) K, 2θmax = 134.0o, 14,672 reflections collected, 8689 unique (Rint = 0.0287).The concentration of each test compound was 10 μmol/L plus 5 ng/mL NGF. b PC12 cells were induced by 5 ng/mL NGF and the cells have the ability to differentiate (or have a tendency to differentiate) but do not differentiate.PC12 cells were induced by 50 ng/mL NGF and the cells differentiated well.GooF = 1.013, R1 = 0.0406, wR2 = 0.1024, R indices based on 3209 reflections with I > 2c(I) (refinement on F2), 277 parameters, 1 restraint. Lp and absorption corrections applied, “ = 0.768 mm− 1. Flack param- eter is − 0.12(13). Crystallographic data of compound 2h were deposited at the Cambridge Crystallographic Data Centre (CCDC) with the No. 1528503.All target merrillianone/cycloparviforalone derivatives 2a一i, 3a一g and 3i一q were evaluated for their neurite outgrowth-promoting activ- ities in PC12 cells according to the previously reported procedures.22,23 The neurite outgrowth-promoting activity was expressed as the differ- entiation rates (DRs) of PC12 cells. As shown in Table 1, the lead com- pounds 1 and 2 showed weak neurite outgrowth-promoting activity in comparison with negative control.
But after structural modification of compounds 1 and 2, most of target derivatives exhibited more potent neurite outgrowth-promoting activity than their precursors, such as compounds 2a, 2c, 2g, 2h, 2i, 3a一e and 3l一o. Among all the merrillia- none derivatives 2a一i, compound 2a displayed the most potent neurite outgrowth-promoting activity with the DR of 11.11%; among all the cycloparvi foralone derivatives 3a一g and 3i一q, compound 3a displayed the best neurite outgrowth-promoting activity with the DR of 11.74%, which is more than 2-fold improvement compared with the precursor cycloparviforalone. In addition, compound 3b also exhibited potent neurite outgrowth-promoting activity with the DR of 10.41%. Effects of compounds 2a, 3a and 3b on the NGF-induced neurite outgrowth in PC12 cells were validated through morphological observations (Fig. 4). When PC12 cells were stimulated with a concentration of50 ng/mL NGF (positive group), they ceased growth and began to grow neurites, eventually differentiating into a neuron-like phenotype. At a concen- tration of 5 ng/mL of NGF (negative group), cells were unchange in morphology as in the blank group. However, in the presence of 2a, 3a or 3b at a concentrations of 10 μM, obviously, they had a promoting effect on neurite outgrowth of PC12 cells, inducing longer average neurite length (Fig. 4).
In addition, the cytotoxicity of two potent compounds 2a and 3a against normal human gastric epithelial (GES-1) cells was assessed via a cell counting kit-8 (CCK-8) method.24 Compounds 2a and 3a exhibited relatively low cytotoxicity to normal GES-1 cells with the IC50 values of 124.5 ± 5.85 and 329.5 ± 14.5 μg/mL, respectively, which suggests that compounds 2a and 3a had good selectivity between PC12 and GES-1 cells. Moreover, we further determined the hydrolytic stability of a representative compound 3q by HPLC under the condition of simulating the pH of gastric juice after a meal.25 In brief, 3q was dissolved in a phosphate buffer solution with pH of 3.5, and the mixture was then stirred at 37 ◦ C and detected by HPLC every 2 h. As shown in Fig. 5, after hydrolysis for 6 h, the retention time of compound 3q did not change much and there were no other impurity peaks, which indicated that 3q had good hydrolytic stability and could betaken orally.Finally, some interesting results of SARs of these merrillianone/ cycloparviforalone derivatives were also observed. When introduction of hexa-2,4-dienoyloxy on merrillianone/cycloparviforalone, the corre- sponding derivatives 2aand 3a exhibited more potent neurite outgrowth-promoting activity than that of other acyloxy groups.
In addition, whether the precursor is merrillianone or cycloparviforalone, introductionof 4-fluoro/chloro/bromo-phenylpropionyloxy could obtain more active compounds than that of 3-fluoro/chloro/bromo-phe- nylpropionyloxy(Fig.6).For example,2b (R = 4-fluoro-phenyl- propionyloxy) v.s. 2e (R = 3-fluoro-phenylpropionyloxy), 2c (R = 4- chloro-phenylpropionyloxy) v.s. 2f (R = 3-chloro-phenylpropionyloxy), 2d (R = 4-bromo-phenylpropionyloxy) v.s. 2g (R = 3-bromo-phenyl- propionyloxy), 3b (R = 4-fluoro-phenylpropionyloxy) v.s. 3e (R = 3- fluoro-phenylpropionyloxy), 3c (R = 4-chloro-phenylpropionyloxy) v.s. 3f (R = 3-chloro-phenylpropionyloxy), and 3d (R = 4-bromo-phenyl- propionyloxy) v.s. 3g (R = 3-bromo-phenylpropionyloxy).
In conlusion, we have prepared a series of novel merrillianone/ cycloparviforalone based esters 2a一i, 3a一g and 3i一q and determined their structures bio-based polymer by 1H NMR, 13C NMR and IR spectral analyses. More- over, we further confirmed the spatial structure of compound 2h by X- ray crystallography. The neurite outgrowth-promoting activity results indicated that most of the target derivatives exhibited more potent neurite outgrowth-promoting activity than merrillianone and cyclo- parviforalone. Among all target derivatives, the neurite outgrowth- promoting activity of compounds 2a, 3a and 3b was about 2-fold stronger than that of their precursors merrillianone and cycloparvifor- alone, respectively. In addition, compounds 2a and 3a also displayed relatively low cytotoxicity to normal GES-1 cells.