CHEMMEDCHEM COMMUNICATIONS DOI: 10.1002/cmdc.201400051
Stereoselective Reduction of 1-O-Isopropyloxygenipin Enhances Its Neuroprotective Activity in Neuronal Cells from Apoptosis Induced by Sodium Nitroprusside Rikang Wang,[a] Jian Yang,[b] Sufen Liao,[a] Gaokeng Xiao,[b] Jun Luo,[b] Lang Zhang,[a] Peter J. Little,[a] Heru Chen,*[b, c] and Wenhua Zheng*[a] Genipin is a Chinese herbal medicine with both neuroprotective and neuritogenic activity. Because of its unstable nature, efforts have been to develop more stable genipin derivatives with improved biological activities. Among the new compounds reported in the literature, (1R)-isopropyloxygenipin (IPRG001) is a more stable but less active compound compared with the parent, genipin. Here, two new IPRG001 derivatives generated by stereoselective reduction of the C6 = C7 double bond were synthesized. The 1R and 1S isomers of (4aS,7S,7aS)methyl-7-(hydroxymethyl)-1-isopropoxy-1,4a,5,6,7,7a-hexahydrocyclopenta[c]pyran-4-carboxylate (CHR20 and CHR21) were shown to be very stable both in high-glucose cell culture medium and in mice serum at 37 8C. Evaluation using an MTT assay and Hoechst staining showed that CHR20 and CHR21 promote the survival of rat adrenal pheochromocytoma (PC12) and retinal neuronal (RGC-5) cells from injury induced by sodium nitroprusside (SNP). The neuroprotective effects of CHR20 and CHR21 were greater than both isomers of IPRG001, the parent compounds. These results indicate that reduction of 1-O-isopropyloxygenipin enhances its neuroprotective activity without affecting its stability.
Genipin, a herbal iridoid, has been reported to have neuroprotective activity in the PC12h and Neuro2a cell lines impaired by 6-hydroxydopamine, hydrogen peroxide, and under serumfree conditions.[1] However, the dehydropyran ring in genipin is easily broken resulting in an unstable dialdehyde that reacts easily with amino and/or thiol residues in physiological conditions; this reactivity renders genipin difficult to use in prac[a] R. Wang,+ S. Liao,+ L. Zhang, P. J. Little, Prof. W. Zheng State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center Sun Yat-Sen University Guangzhou, 510060 (P.R. China.) E-mail: [email protected] [b] J. Yang,+ G. Xiao, J. Luo, Prof. H. Chen Institute of Traditional Chinese Medicine & Natural Products College of Pharmacy, Jinan University Guangzhou 510632 (P.R. China) E-mail: [email protected] [c] Prof. H. Chen Guangdong Province Key Laboratory of Pharmacodynamic Constituents of TCM & New Drugs Research, Jinan University Guangzhou 510632 (P.R. China) [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cmdc.201400051.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
tice.[2] Efforts have been made to develop more stable genipin derivatives with retained or improved biological activities. Among the compounds reported, gardenamide A was identified as a more stable and active candidate, while (1R)isopropyloxygenipin (IPRG001) was found to be more stable but a little less active compared with genipin (Scheme 1).[3]
Scheme 1. Structures of genipin and its stable derivatives.
From a structural point of view, on the one hand, the iridoid and its similar scaffold in genipin or derivatives are the key structural elements for maintaining their biological functions; in contrast, the C6=C7 double bond might possibly bring about rigidity that hinders the possible combination of the C7 hydromethyl with target enzyme(s) through hydrogen bonds, and it is possible that stereoselective reduction of the C6=C7 double bond in IPRG001 will enhance its neuroprotective activity. To the best of our knowledge, no information about this strategy has been reported to date. Genipin and its derivatives are structurally similar to tetrahydrobiopterin, a co-factor of nitric oxide synthase (NOS).[3a, 4] They activate NOS and increase the production of nitric oxide (NO).[5] NO is known as a Janus-faced molecule that is physiologically produced through the l-arginine/NOS pathway. It plays various physiological roles in the central nervous system, including neuromodulation, neurotransmission, and synaptic plasticity.[6] However, the overproduction of NO is also implicated in the pathogenesis of neurodegenerative disorders, such as Parkinson’s disease, Alzheimer’s disease, Huntington’s disChemMedChem 2014, 9, 1397 – 1403
1397
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org
ease, and retinal neurodegenerative disorders like age-related (Scheme 1). The configuration at C7 was identified as R. The abmacular degeneration (AMD) and glaucoma.[7] In biological syssolute configurations at C1 and C7 of CHR21 were also contems, excessive NO reacts with superoxide anions (O2 C) resultfirmed by the same method. Two pairs of diasteromers CHR01a/CHR01b and CHR20/CHR21 were separated by reing in the formation of peroxynitrite (ONOO ), which induces verse-phase high-performance liquid chromatography (RPlipid peroxidation (LPO) and the disruption of cell membranes HPLC). leading to the release of cell organelles. Overproduction of NO The reduction of the C6=C7 double bond in IPRG001 increasis known to cause neurotoxicity, including oxidative impairment.[8] Taken together, an interesting question is put forward: es the flexibility of the new genipin derivatives (CHR20/21). As shown in Figure 1, the overlap diagram of 50 conformers with As chemicals involved in NOS activation, do genipin and its lowest energy for IPGR001 is more focused than those of stable derivatives suppress NO toxicity? To answer this quesCHR20 and CHR21. This indicates that the C6=C7 double bond tion, in the present study, CHR20 and CHR21 were designed and synthesized for the purpose of releasing the rigidity in in IPGR001 limits the movement of the molecular skeleton. ReIPRG001. Their capacities to protect PC12 and RGC-5 cells from duction of the C6=C7 double bond in both CHR01a and sodium nitroprusside (SNP)-induced injury were carefully examined. The syntheses of CHR20 and CHR21 is outlined in Scheme 2. Firstly, acetylation of genipin at C1 was carried out by reaction of genipin with isopropanol in the presence of para-toluene sulfonic acid led to mixtures of (1R)-1isopropyloxygenipin (CHR01a) and (1S)-1-isopropyloxygenipin (CHR01b) in high yield (95 %). Figure 1. Flexibility comparison of a) IPRG001, R isomer, b) CHR20, and c) CHR21. The lowest-energy conformation The CHR01a/CHR01b mixture was identified using OMEGA version 2.3 (for details, see the Supporting Information). was then reacted with sodium borohydride in the presence of
Scheme 2. The synthesis of CHR20 and CHR21. Reagents and conditions: a) iPrOH, p-TsOH, 80 8C, 3 h, 95 %; b) NiCl2·6H2O, NaBH4, MeOH, 15 8C, 2 h, 79 %.
catalytic nickel(II) chloride at 15 8C to give a mixture of CHR20 and CHR21. This is a chemoselective reaction in which the C3=C4 double bond remains intact. Interestingly, the reduction of the C6=C7 double bond is highly stereoselective, and only the R-configuration at C7 was observed. The mechanism might be relative to the chelation of nickel with 7-hydroxymethyl and 1-isopropyloxy. The absolute configurations at C1 and C7 were confirmed by analysis of the coupling constants of C7a H versus C1 H and C7 H, respectively. Thus, the C1 H of CHR20 displayed a doublet at a d value of 4.81 ppm, with a coupling constant of 9.3 Hz, which indicates trans to C7a H. The configuration at C1 was confirmed as R. The C7 H of CHR20 coupled with C7a H as a doublet at a d value of 4.15 ppm, with a coupling constant of 9.1 Hz, this indicates trans conformation to C7a H 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
CHR01b releases the rigidity and results in more flexible derivatives, CHR20 and CHR21, respectively. This result supports the notion that the C6=C7 double bond in IPRG001 limits the movement of hydromethyl group at C7, and the reduction of the C6=C7 double bond increases the flexibility of the hydromethyl group at the C7 of newly generated molecules CHR20 and CHR21. The significance of this change in flexibility is ambiguous. It may be in favor of the interaction of CHR20/CHR21 with target enzyme(s) (especially through hydrogen bond interactions), leading to modulation of their physiological functions. To confirm the stability of these new compounds in culture media, CHR20 and CHR21 were incubated in high-glucose Dulbecco’s modified Eagle’s medium (DMEM) and mice serum at 37 8C for 24 hours. As shown in Figure 2, CHR20 was confirmed to be very stable, with no significant formation of decomposition components found over the course of the incubation. A similar result was obtained for CHR21 (data shown in the Supporting Information). Together, these results demonstrated that the reduction of the C6=C7 double bond in IPRG001 increases the flexibility of this molecule without affecting its stability. To examine the cytotoxicity of CHR20 and CHR21, PC12 cells were treated with test compounds at concentrations ranging from 0.3 to 30 mm for 24 hours. As shown in Figure 3, CHR20 did not induce any toxicity at concentration up to 30 mm. As a comparison, CHR21 slightly decreased cell survival (about 20 %) at 30 mm, which implies weak cytotoxicity. ChemMedChem 2014, 9, 1397 – 1403
1398
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org
Figure 2. Stability of CHR20. CHR20 was incubated in either a) high-glucose Dulbecco’s modified Eagle’s medium (DMEM) or b) mice serum at 37 8C for a period of 24 h. Samples were analyzed at 0, 0.5, 1.0, 2.0, 4.0, 12.0, 24.0 h using reverse-phase high-performance liquid chromatography (RP-HPLC). Arrows point to target compounds. The experiments were repeated at least three times and compared with the control.
Figure 3. Cytotoxicity of CHR20 and CHR21 in PC12 cells. Cells were treated with test compound at concentrations ranging from 0 to 30 mm for 24 h. Cell viability was determined by using an MTT assay. Data represent the mean SD of six replicates. ** P < 0.01 vs control (CT) group.
It is reported that overproduction of NO causes neurotoxicity, including oxidative impairment.[7g] Oxidative stress is a risk factor in the development of neurodegenerative diseases, such as Alzheimer’s disease, Parkinson’s disease, and glaucomatous visual field deterioration.[8b–d, 9] Interestingly, both CHR20 and CHR21 attenuated the injury induced by SNP, which mimics the overproduction of NO, in a dose-dependent manner in PC12 cells (Figure 4). The protective effect of CHR21 was statistically significant at a concentration of 3 mm (P < 0.05) and maximal at 10 mm. At a concentration over 30 mm, CHR21 worsened the cell damage. Evidently, the protective effects of CHR20 and CHR21 (CHR21 > CHR20) are slightly more potent than their parent compounds, CHR01a/CHR01b. For example, the protective ratio of CHR01a at 10 mm was 44.4 % 2.4 %, while CHR20 at the same concentration gave a protective ratio of 66.2 % 1.4 %. CHR21 is the most potent, with an EC50 value of 5.32 mm against SNP-induced cell death. CHR20 is less 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
potent than CHR21, with an EC50 of 6.63 mm. The EC50 values of CHR01a and CHR01b are 11.40 mm and 8.85 mm, respectively. Similar results were observed in RGC-5 cells (data not shown), which is a photoreceptor cell line.[10] These results indicate that CHR20 and CHR21 are able to protect PC12 and RGC-5 cells from SNP-induced apoptosis, and the reduction of the C6=C7 double bond in IPRG001 enhances the neuroprotective effect of these compounds. Mechanisms underlying this effect are not clear at present. One explanation is that the reduction of the
Figure 4. Neuroprotection of PC12 cells by CHR01a, CHR20, CHR01b, and CHR21 from toxicity induced by sodium nitroprusside (SNP) (800 mm). Cells were pretreated with CHR21 and CHR20 at different concentrations for 2 h and then exposed to SNP. Cell survival was determined by an MTT assay. Protection (%) was calculated as 100 ((drug + SNP)OD SNPOD)/(controlOD SNPOD). Results are the mean SD of six replicates.
C6=C7 double bond in IPRG001 increases the flexibility of CHR20 and CHR21, which facilitates the interaction of the C7 hydromethyl group with target enzyme(s) through hydrogen bonding, leading to enhancement of their neuroprotective activities as expected. Analysis of PC12 and RGC-5 cells with Hoechst 33342 staining revealed that pretreatment with CHR21 significantly decreased the number of apoptotic cells induced by SNP. CHR21 pretreatment alone showed no effect on apoptotic bodies or the nuclear morphology of PC12 cells (data not shown). Pretreatment with CHR21 (10 mm) attenuated nuclear fragmentation and chromatin condensation, and inhibited apoptosis at the early and intermediate stages in PC12 and RGC-5 cells induced by SNP exposure (Figure 5). Interestingly, CHR21 is more potent than CHR20, while the effect of CHR01b is greater than that of CHR01a in protecting PC12 and RGC-5 cells ChemMedChem 2014, 9, 1397 – 1403
1399
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org
Figure 5. CHR21 protected PC12 and RGC-5 cells from SNP-induced apoptosis. Cells were pretreated with CHR21 (10 mm) for 2 h before exposure to SNP (800 mm). After 24 h, cells were stained with Hoechst 33342, and apoptosis was detected by a high-content screening system. The proportion of apoptosis (%) was determined as the number of apoptotic cells to total number of cells. Intracellular reactive oxygen species (ROS) levels (%) were expressed as the mean intensity of fluorescence (MIF) within the live cell and were measured by a high-content screening system after cells were stained with DCFH-DA. a) Morphologic changes, nuclear condensation, and accumulation of ROS in PC12 cells. b) Histogram showing the percentage of apoptosis in PC12 cells; data are the mean SD of three individual experiments. c) Histogram showing the percentage of apoptosis in RGC-5 cells; data are the mean SD of three individual experiments. d) Histogram showing ROS levels in PC12 cells after treatment with sodium nitroprusside (SNP) in the presence or absence of CHR21 compared with untreated groups. ## P < 0.05 vs control group; * P < 0.05, ** P < 0.05 vs SNP group.
from SNP-induced apoptosis as shown in Figures 4 and 5. These results indicate that the 1S isomer (CHR21 and CHR01b) is more potent than the 1R isomer (CHR20 and CHR01a). The phenomenon is consistent with many reports about the different physiological effects between the enantiomers. CHR21 was therefore used as the representative compound in further experiments. SNP is known to increase the levels of oxidative products, such as reactive oxygen species (ROS), in many cell types and causes apoptosis.[11] In the current investigation, exposure to SNP increased the levels of ROS in PC12 cells, while pretreatment with CHR21 significantly decreased SNP-induced production of ROS (Figure 5). The mechanism through which CHR21 eliminates ROS is not known at present. It is well established that antioxidative proteins play an important role in eliminating ROS in organisms. Overproduction of ROS usually induces the expression of antioxidative proteins, which can antagonize 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
the effects of ROS, leading to a balance between oxidants and antioxidants. Consistent with this hypothesis, it has been reported that genipin upregulates heme oxygenase-1 via the PI3-kinase-JNK1/2-Nrf2 signaling pathway in RAW264.7 macrophages.[12] On the other hand, IPRG001 induces S-nitrosylation of Keap1, causing translocation of Nrf2 to the nucleus and expression of antioxidants in retina neuronal cells.[4a] In order to determine the role of antioxidative proteins in the effect of CHR21 on ROS levels, the effects of CHR21 on the expression of antioxidative proteins like GCLC and SOD1 were determined longitudinally by reverse transcription polymerase chain reaction (RT-PCR). As indicated in Figure 6, CHR21 increased mRNA levels of GCLC and SOD1 in a time-dependent manner. Treatment with CHR21 increased mRNA of GCLC at about 4 hours, and the mRNA peaked at 6–10 hours; while the mRNA of SOD1 significantly increased at 6 hours and peaked at 8 hours. A similar result was observed for CHR20 (data not shown). The ChemMedChem 2014, 9, 1397 – 1403
1400
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org Acknowledgements This research was financially supported by the National Natural Science Foundation of China (grant no. 81172982, 30970935, and 31371088), the Guangdong Provincial Project of Science & Technology (China) (grant no. 2010A030100006 and 2011B050200005), and the Director’s Fund of the State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University (Guangzhou, China). Keywords: apoptosis · neurotoxicity · PC12 cells
Figure 6. Effect of CHR21 on mRNA levels of antioxidative proteins in sodium nitroprusside (SNP)-induced PC12 cells. a) Expression of GCLC and SOD1 determined by RT-PCR. PC12 cells were treated with CHR21 (10 mm) for different lengths of time. RPL-19 was used as an internal control. b) Relative mRNA levels of GCLC (&) and SOD1 (&) analyzed by ImageJ analysis software. mRNA levels were normalized to that of RPL-19. Results are the mean SEM of three independent experiments. * p < 0.05, # p < 0.05, ** p < 0.01, ## p < 0.01 vs control group (n = 3).
increase in mRNA levels of GCLC and SOD1 reflects an elevation of antioxidant defense mechanisms as a response to oxidative stress,[13] as such, our results indicate that CHR20/ CHR21 may improve the antioxidative capacity of PC12 cells, resulting in protection against oxidative injury. Based on this observation, it is reasonable to speculate that CHR20 and CHR21 may protect against neurodegenerative impairment by increasing the expression of antioxidative proteins. In summary, our data demonstrate that CHR20 and CHR21 ameliorated SNP-induced cell apoptosis, decreased ROS levels, and modulated antioxidant protein expression in PC12 cells. Reduction of the C6=C7 double bond in IPRG001 led to derivatives with increased flexibility and enhanced neuronal protective activity. These findings may be useful for the design of novel compounds for the treatment of neurodegenerative disorders, such as Alzheimer’s disease and glaucoma.
Supporting Information Experimental protocols for the synthesis and biological evaluation of the compounds described here are given in the Supporting Information together with NMR and mass spectrometry spectra, and HPLC chromatograms.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
genipin
·
neuroprotection
·
[1] a) M. Yamazaki, N. Sakura, K. Chiba, T. Mohri, Biol. Pharm. Bull. 2001, 24, 1454 – 1455; b) M. Yamazaki, K. Chiba, K. Satoh, J. Health Sci. 2008, 54, 638 – 644. [2] Y. Kawata, M. Hattori, T. Akao, K. Kobashi, T. Namba, Planta Med. 1991, 57, 536 – 542. [3] a) J. Luo, R. Wang, Z. Huang, J. Yang, X. Yao, H. Chen, W. Zheng, ChemMedChem 2012, 7, 1661 – 1668; b) H. Suzuki, M. Yamazaki, K. Chiba, Y. Uemori, H. Sawanishi, Chem. Pharm. Bull. 2010, 58, 168 – 171. [4] a) Y. Koriyama, K. Chiba, M. Yamazaki, H. Suzuki, K. Muramoto, S. Kato, J. Neurochem. 2010, 115, 79 – 91; b) H. Suzuki, M. Yamazaki, K. Chiba, H. Sawanishi, J. Health Sci. 2007, 53, 730 – 733. [5] a) M. Yamazaki, K. Chiba, T. Mohri, Br. J. Pharmacol. 2005, 146, 662 – 669; b) Y. Koriyama, Y. Takagi, K. Chiba, M. Yamazaki, K. Arai, T. Matsukawa, H. Suzuki, K. Sugitani, H. Kagechika, S. Kato, J. Neurochem. 2011, 119, 1232 – 1242. [6] a) F. X. Guix, I. Uribesalgo, M. Coma, F. J. Munoz, Prog. Neurobiol. 2005, 76, 126 – 152; b) C. M. Troy, S. A. Rabacchi, W. J. Friedman, T. F. Frappier, K. Brown, M. L. Shelanski, J. Neurosci. 2000, 20, 1386 – 1392; c) A. Contestabile, E. Ciani, Neurochem. Int. 2004, 45, 903 – 914. [7] a) V. Calabrese, C. Mancuso, M. Calvani, E. Rizzarelli, D. A. Butterfield, A. M. Stella, Nat. Rev. Neurosci. 2007, 8, 766 – 775; b) C. S. Chan, T. S. Gertler, D. J. Surmeier, Trends Neurosci. 2009, 32, 249 – 256; c) J. Emerit, M. Edeas, F. Bricaire, Biomed. Pharmacother. 2004, 58, 39 – 46; d) V. L. Dawson, T. M. Dawson, J. Chem. Neuroanat. 1996, 10, 179 – 190; e) P. Kuppusamy, S. T. Ohnishi, Y. Numagami, T. Ohnishi, J. L. Zweier, J. Cereb. Blood Flow Metab. 1995, 15, 899 – 903; f) S. C. Bondy, S. Naderi, Neurosci. Lett. 1994, 168, 34 – 36; g) C. A. Massaad, Curr. Neuropharm. 2011, 9, 662 – 673. [8] a) F. Zhang, R. M. Casey, M. E. Ross, C. Iadecola, Stroke 1996, 27, 317 – 323; b) A. H. Neufeld, Surv. Ophthalmol. 1999, 43, S129 – S135; c) S. M. Ferreira, S. F. Lerner, R. Brunzini, P. A. Evelson, S. F. Llesuy, Am. J. Ophthalmol. 2004, 137, 62 – 69; d) D. Maneesh Kumar, N. Agarwal, J. Glaucoma 2007, 16, 334 – 343. [9] C. A. Massaad, Curr. Neuropharmacol. 2011, 9, 662 – 673. [10] R. R. Krishnamoorthy, A. F. Clark, D. Daudt, J. K. Vishwanatha, T. Yorio, Invest. Ophthalmol. Visual Sci. 2013, 54, 5712 – 5719. [11] a) K. Verma, S. K. Mehta, G. S. Shekhawat, Biometals 2013, 26, 255 – 269; b) H. F. Qian, W. Chen, J. J. Li, J. Wang, Z. Zhou, W. P. Liu, Z. W. Fu, Aquat. Toxicol. 2009, 92, 250 – 257; c) N. Ning, J. F. Hu, Y. H. Yuan, X. Y. Zhang, J. G. Dai, N. H. Chen, Acta Pharmacol. Sin. 2012, 33, 34 – 40. [12] W. K. Jeon, H. Y. Hong, B. C. Kim, Arch. Biochem. Biophys. 2011, 512, 119 – 125. [13] a) J. W. Kaspar, S. K. Niture, A. K. Jaiswal, Free Radical Biol. Med. 2009, 47, 1304 – 1309; b) Y. Kurauchi, A. Hisatsune, Y. Isohama, H. Katsuki, Neuroscience 2009, 158, 856 – 866. [14] H. T. Wang, X. H. Zhou, J. C. Huang, N. Mu, Z. L. Guo, Q. Wen, R. K. Wang, S. R. Chen, Z. P. Feng, W. H. Zheng, Psychopharmacology 2013, 228, 129 – 141. [15] H. Wang, X. Duan, Y. Ren, Y. Liu, M. Huang, P. Liu, R. Wang, G. Gao, L. Zhou, Z. Feng, W. Zheng, Mol. Neurobiol. 2013, 47, 24 – 36. Received: January 22, 2014 Published online on April 6, 2014
ChemMedChem 2014, 9, 1397 – 1401
1401
Stereoselective Reduction of 1-O-Isopropyloxygenipin Enhances Its Neuroprotective Activity in Neuronal Cells from Apoptosis Induced by Sodium Nitroprusside Rikang Wang,[a] Jian Yang,[b] Sufen Liao,[a] Gaokeng Xiao,[b] Jun Luo,[b] Lang Zhang,[a] Peter J. Little,[a] Heru Chen,*[b, c] and Wenhua Zheng*[a] Genipin is a Chinese herbal medicine with both neuroprotective and neuritogenic activity. Because of its unstable nature, efforts have been to develop more stable genipin derivatives with improved biological activities. Among the new compounds reported in the literature, (1R)-isopropyloxygenipin (IPRG001) is a more stable but less active compound compared with the parent, genipin. Here, two new IPRG001 derivatives generated by stereoselective reduction of the C6 = C7 double bond were synthesized. The 1R and 1S isomers of (4aS,7S,7aS)methyl-7-(hydroxymethyl)-1-isopropoxy-1,4a,5,6,7,7a-hexahydrocyclopenta[c]pyran-4-carboxylate (CHR20 and CHR21) were shown to be very stable both in high-glucose cell culture medium and in mice serum at 37 8C. Evaluation using an MTT assay and Hoechst staining showed that CHR20 and CHR21 promote the survival of rat adrenal pheochromocytoma (PC12) and retinal neuronal (RGC-5) cells from injury induced by sodium nitroprusside (SNP). The neuroprotective effects of CHR20 and CHR21 were greater than both isomers of IPRG001, the parent compounds. These results indicate that reduction of 1-O-isopropyloxygenipin enhances its neuroprotective activity without affecting its stability.
Genipin, a herbal iridoid, has been reported to have neuroprotective activity in the PC12h and Neuro2a cell lines impaired by 6-hydroxydopamine, hydrogen peroxide, and under serumfree conditions.[1] However, the dehydropyran ring in genipin is easily broken resulting in an unstable dialdehyde that reacts easily with amino and/or thiol residues in physiological conditions; this reactivity renders genipin difficult to use in prac[a] R. Wang,+ S. Liao,+ L. Zhang, P. J. Little, Prof. W. Zheng State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center Sun Yat-Sen University Guangzhou, 510060 (P.R. China.) E-mail: [email protected] [b] J. Yang,+ G. Xiao, J. Luo, Prof. H. Chen Institute of Traditional Chinese Medicine & Natural Products College of Pharmacy, Jinan University Guangzhou 510632 (P.R. China) E-mail: [email protected] [c] Prof. H. Chen Guangdong Province Key Laboratory of Pharmacodynamic Constituents of TCM & New Drugs Research, Jinan University Guangzhou 510632 (P.R. China) [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cmdc.201400051.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
tice.[2] Efforts have been made to develop more stable genipin derivatives with retained or improved biological activities. Among the compounds reported, gardenamide A was identified as a more stable and active candidate, while (1R)isopropyloxygenipin (IPRG001) was found to be more stable but a little less active compared with genipin (Scheme 1).[3]
Scheme 1. Structures of genipin and its stable derivatives.
From a structural point of view, on the one hand, the iridoid and its similar scaffold in genipin or derivatives are the key structural elements for maintaining their biological functions; in contrast, the C6=C7 double bond might possibly bring about rigidity that hinders the possible combination of the C7 hydromethyl with target enzyme(s) through hydrogen bonds, and it is possible that stereoselective reduction of the C6=C7 double bond in IPRG001 will enhance its neuroprotective activity. To the best of our knowledge, no information about this strategy has been reported to date. Genipin and its derivatives are structurally similar to tetrahydrobiopterin, a co-factor of nitric oxide synthase (NOS).[3a, 4] They activate NOS and increase the production of nitric oxide (NO).[5] NO is known as a Janus-faced molecule that is physiologically produced through the l-arginine/NOS pathway. It plays various physiological roles in the central nervous system, including neuromodulation, neurotransmission, and synaptic plasticity.[6] However, the overproduction of NO is also implicated in the pathogenesis of neurodegenerative disorders, such as Parkinson’s disease, Alzheimer’s disease, Huntington’s disChemMedChem 2014, 9, 1397 – 1403
1397
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org
ease, and retinal neurodegenerative disorders like age-related (Scheme 1). The configuration at C7 was identified as R. The abmacular degeneration (AMD) and glaucoma.[7] In biological syssolute configurations at C1 and C7 of CHR21 were also contems, excessive NO reacts with superoxide anions (O2 C) resultfirmed by the same method. Two pairs of diasteromers CHR01a/CHR01b and CHR20/CHR21 were separated by reing in the formation of peroxynitrite (ONOO ), which induces verse-phase high-performance liquid chromatography (RPlipid peroxidation (LPO) and the disruption of cell membranes HPLC). leading to the release of cell organelles. Overproduction of NO The reduction of the C6=C7 double bond in IPRG001 increasis known to cause neurotoxicity, including oxidative impairment.[8] Taken together, an interesting question is put forward: es the flexibility of the new genipin derivatives (CHR20/21). As shown in Figure 1, the overlap diagram of 50 conformers with As chemicals involved in NOS activation, do genipin and its lowest energy for IPGR001 is more focused than those of stable derivatives suppress NO toxicity? To answer this quesCHR20 and CHR21. This indicates that the C6=C7 double bond tion, in the present study, CHR20 and CHR21 were designed and synthesized for the purpose of releasing the rigidity in in IPGR001 limits the movement of the molecular skeleton. ReIPRG001. Their capacities to protect PC12 and RGC-5 cells from duction of the C6=C7 double bond in both CHR01a and sodium nitroprusside (SNP)-induced injury were carefully examined. The syntheses of CHR20 and CHR21 is outlined in Scheme 2. Firstly, acetylation of genipin at C1 was carried out by reaction of genipin with isopropanol in the presence of para-toluene sulfonic acid led to mixtures of (1R)-1isopropyloxygenipin (CHR01a) and (1S)-1-isopropyloxygenipin (CHR01b) in high yield (95 %). Figure 1. Flexibility comparison of a) IPRG001, R isomer, b) CHR20, and c) CHR21. The lowest-energy conformation The CHR01a/CHR01b mixture was identified using OMEGA version 2.3 (for details, see the Supporting Information). was then reacted with sodium borohydride in the presence of
Scheme 2. The synthesis of CHR20 and CHR21. Reagents and conditions: a) iPrOH, p-TsOH, 80 8C, 3 h, 95 %; b) NiCl2·6H2O, NaBH4, MeOH, 15 8C, 2 h, 79 %.
catalytic nickel(II) chloride at 15 8C to give a mixture of CHR20 and CHR21. This is a chemoselective reaction in which the C3=C4 double bond remains intact. Interestingly, the reduction of the C6=C7 double bond is highly stereoselective, and only the R-configuration at C7 was observed. The mechanism might be relative to the chelation of nickel with 7-hydroxymethyl and 1-isopropyloxy. The absolute configurations at C1 and C7 were confirmed by analysis of the coupling constants of C7a H versus C1 H and C7 H, respectively. Thus, the C1 H of CHR20 displayed a doublet at a d value of 4.81 ppm, with a coupling constant of 9.3 Hz, which indicates trans to C7a H. The configuration at C1 was confirmed as R. The C7 H of CHR20 coupled with C7a H as a doublet at a d value of 4.15 ppm, with a coupling constant of 9.1 Hz, this indicates trans conformation to C7a H 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
CHR01b releases the rigidity and results in more flexible derivatives, CHR20 and CHR21, respectively. This result supports the notion that the C6=C7 double bond in IPRG001 limits the movement of hydromethyl group at C7, and the reduction of the C6=C7 double bond increases the flexibility of the hydromethyl group at the C7 of newly generated molecules CHR20 and CHR21. The significance of this change in flexibility is ambiguous. It may be in favor of the interaction of CHR20/CHR21 with target enzyme(s) (especially through hydrogen bond interactions), leading to modulation of their physiological functions. To confirm the stability of these new compounds in culture media, CHR20 and CHR21 were incubated in high-glucose Dulbecco’s modified Eagle’s medium (DMEM) and mice serum at 37 8C for 24 hours. As shown in Figure 2, CHR20 was confirmed to be very stable, with no significant formation of decomposition components found over the course of the incubation. A similar result was obtained for CHR21 (data shown in the Supporting Information). Together, these results demonstrated that the reduction of the C6=C7 double bond in IPRG001 increases the flexibility of this molecule without affecting its stability. To examine the cytotoxicity of CHR20 and CHR21, PC12 cells were treated with test compounds at concentrations ranging from 0.3 to 30 mm for 24 hours. As shown in Figure 3, CHR20 did not induce any toxicity at concentration up to 30 mm. As a comparison, CHR21 slightly decreased cell survival (about 20 %) at 30 mm, which implies weak cytotoxicity. ChemMedChem 2014, 9, 1397 – 1403
1398
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org
Figure 2. Stability of CHR20. CHR20 was incubated in either a) high-glucose Dulbecco’s modified Eagle’s medium (DMEM) or b) mice serum at 37 8C for a period of 24 h. Samples were analyzed at 0, 0.5, 1.0, 2.0, 4.0, 12.0, 24.0 h using reverse-phase high-performance liquid chromatography (RP-HPLC). Arrows point to target compounds. The experiments were repeated at least three times and compared with the control.
Figure 3. Cytotoxicity of CHR20 and CHR21 in PC12 cells. Cells were treated with test compound at concentrations ranging from 0 to 30 mm for 24 h. Cell viability was determined by using an MTT assay. Data represent the mean SD of six replicates. ** P < 0.01 vs control (CT) group.
It is reported that overproduction of NO causes neurotoxicity, including oxidative impairment.[7g] Oxidative stress is a risk factor in the development of neurodegenerative diseases, such as Alzheimer’s disease, Parkinson’s disease, and glaucomatous visual field deterioration.[8b–d, 9] Interestingly, both CHR20 and CHR21 attenuated the injury induced by SNP, which mimics the overproduction of NO, in a dose-dependent manner in PC12 cells (Figure 4). The protective effect of CHR21 was statistically significant at a concentration of 3 mm (P < 0.05) and maximal at 10 mm. At a concentration over 30 mm, CHR21 worsened the cell damage. Evidently, the protective effects of CHR20 and CHR21 (CHR21 > CHR20) are slightly more potent than their parent compounds, CHR01a/CHR01b. For example, the protective ratio of CHR01a at 10 mm was 44.4 % 2.4 %, while CHR20 at the same concentration gave a protective ratio of 66.2 % 1.4 %. CHR21 is the most potent, with an EC50 value of 5.32 mm against SNP-induced cell death. CHR20 is less 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
potent than CHR21, with an EC50 of 6.63 mm. The EC50 values of CHR01a and CHR01b are 11.40 mm and 8.85 mm, respectively. Similar results were observed in RGC-5 cells (data not shown), which is a photoreceptor cell line.[10] These results indicate that CHR20 and CHR21 are able to protect PC12 and RGC-5 cells from SNP-induced apoptosis, and the reduction of the C6=C7 double bond in IPRG001 enhances the neuroprotective effect of these compounds. Mechanisms underlying this effect are not clear at present. One explanation is that the reduction of the
Figure 4. Neuroprotection of PC12 cells by CHR01a, CHR20, CHR01b, and CHR21 from toxicity induced by sodium nitroprusside (SNP) (800 mm). Cells were pretreated with CHR21 and CHR20 at different concentrations for 2 h and then exposed to SNP. Cell survival was determined by an MTT assay. Protection (%) was calculated as 100 ((drug + SNP)OD SNPOD)/(controlOD SNPOD). Results are the mean SD of six replicates.
C6=C7 double bond in IPRG001 increases the flexibility of CHR20 and CHR21, which facilitates the interaction of the C7 hydromethyl group with target enzyme(s) through hydrogen bonding, leading to enhancement of their neuroprotective activities as expected. Analysis of PC12 and RGC-5 cells with Hoechst 33342 staining revealed that pretreatment with CHR21 significantly decreased the number of apoptotic cells induced by SNP. CHR21 pretreatment alone showed no effect on apoptotic bodies or the nuclear morphology of PC12 cells (data not shown). Pretreatment with CHR21 (10 mm) attenuated nuclear fragmentation and chromatin condensation, and inhibited apoptosis at the early and intermediate stages in PC12 and RGC-5 cells induced by SNP exposure (Figure 5). Interestingly, CHR21 is more potent than CHR20, while the effect of CHR01b is greater than that of CHR01a in protecting PC12 and RGC-5 cells ChemMedChem 2014, 9, 1397 – 1403
1399
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org
Figure 5. CHR21 protected PC12 and RGC-5 cells from SNP-induced apoptosis. Cells were pretreated with CHR21 (10 mm) for 2 h before exposure to SNP (800 mm). After 24 h, cells were stained with Hoechst 33342, and apoptosis was detected by a high-content screening system. The proportion of apoptosis (%) was determined as the number of apoptotic cells to total number of cells. Intracellular reactive oxygen species (ROS) levels (%) were expressed as the mean intensity of fluorescence (MIF) within the live cell and were measured by a high-content screening system after cells were stained with DCFH-DA. a) Morphologic changes, nuclear condensation, and accumulation of ROS in PC12 cells. b) Histogram showing the percentage of apoptosis in PC12 cells; data are the mean SD of three individual experiments. c) Histogram showing the percentage of apoptosis in RGC-5 cells; data are the mean SD of three individual experiments. d) Histogram showing ROS levels in PC12 cells after treatment with sodium nitroprusside (SNP) in the presence or absence of CHR21 compared with untreated groups. ## P < 0.05 vs control group; * P < 0.05, ** P < 0.05 vs SNP group.
from SNP-induced apoptosis as shown in Figures 4 and 5. These results indicate that the 1S isomer (CHR21 and CHR01b) is more potent than the 1R isomer (CHR20 and CHR01a). The phenomenon is consistent with many reports about the different physiological effects between the enantiomers. CHR21 was therefore used as the representative compound in further experiments. SNP is known to increase the levels of oxidative products, such as reactive oxygen species (ROS), in many cell types and causes apoptosis.[11] In the current investigation, exposure to SNP increased the levels of ROS in PC12 cells, while pretreatment with CHR21 significantly decreased SNP-induced production of ROS (Figure 5). The mechanism through which CHR21 eliminates ROS is not known at present. It is well established that antioxidative proteins play an important role in eliminating ROS in organisms. Overproduction of ROS usually induces the expression of antioxidative proteins, which can antagonize 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
the effects of ROS, leading to a balance between oxidants and antioxidants. Consistent with this hypothesis, it has been reported that genipin upregulates heme oxygenase-1 via the PI3-kinase-JNK1/2-Nrf2 signaling pathway in RAW264.7 macrophages.[12] On the other hand, IPRG001 induces S-nitrosylation of Keap1, causing translocation of Nrf2 to the nucleus and expression of antioxidants in retina neuronal cells.[4a] In order to determine the role of antioxidative proteins in the effect of CHR21 on ROS levels, the effects of CHR21 on the expression of antioxidative proteins like GCLC and SOD1 were determined longitudinally by reverse transcription polymerase chain reaction (RT-PCR). As indicated in Figure 6, CHR21 increased mRNA levels of GCLC and SOD1 in a time-dependent manner. Treatment with CHR21 increased mRNA of GCLC at about 4 hours, and the mRNA peaked at 6–10 hours; while the mRNA of SOD1 significantly increased at 6 hours and peaked at 8 hours. A similar result was observed for CHR20 (data not shown). The ChemMedChem 2014, 9, 1397 – 1403
1400
CHEMMEDCHEM COMMUNICATIONS
www.chemmedchem.org Acknowledgements This research was financially supported by the National Natural Science Foundation of China (grant no. 81172982, 30970935, and 31371088), the Guangdong Provincial Project of Science & Technology (China) (grant no. 2010A030100006 and 2011B050200005), and the Director’s Fund of the State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University (Guangzhou, China). Keywords: apoptosis · neurotoxicity · PC12 cells
Figure 6. Effect of CHR21 on mRNA levels of antioxidative proteins in sodium nitroprusside (SNP)-induced PC12 cells. a) Expression of GCLC and SOD1 determined by RT-PCR. PC12 cells were treated with CHR21 (10 mm) for different lengths of time. RPL-19 was used as an internal control. b) Relative mRNA levels of GCLC (&) and SOD1 (&) analyzed by ImageJ analysis software. mRNA levels were normalized to that of RPL-19. Results are the mean SEM of three independent experiments. * p < 0.05, # p < 0.05, ** p < 0.01, ## p < 0.01 vs control group (n = 3).
increase in mRNA levels of GCLC and SOD1 reflects an elevation of antioxidant defense mechanisms as a response to oxidative stress,[13] as such, our results indicate that CHR20/ CHR21 may improve the antioxidative capacity of PC12 cells, resulting in protection against oxidative injury. Based on this observation, it is reasonable to speculate that CHR20 and CHR21 may protect against neurodegenerative impairment by increasing the expression of antioxidative proteins. In summary, our data demonstrate that CHR20 and CHR21 ameliorated SNP-induced cell apoptosis, decreased ROS levels, and modulated antioxidant protein expression in PC12 cells. Reduction of the C6=C7 double bond in IPRG001 led to derivatives with increased flexibility and enhanced neuronal protective activity. These findings may be useful for the design of novel compounds for the treatment of neurodegenerative disorders, such as Alzheimer’s disease and glaucoma.
Supporting Information Experimental protocols for the synthesis and biological evaluation of the compounds described here are given in the Supporting Information together with NMR and mass spectrometry spectra, and HPLC chromatograms.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
genipin
·
neuroprotection
·
[1] a) M. Yamazaki, N. Sakura, K. Chiba, T. Mohri, Biol. Pharm. Bull. 2001, 24, 1454 – 1455; b) M. Yamazaki, K. Chiba, K. Satoh, J. Health Sci. 2008, 54, 638 – 644. [2] Y. Kawata, M. Hattori, T. Akao, K. Kobashi, T. Namba, Planta Med. 1991, 57, 536 – 542. [3] a) J. Luo, R. Wang, Z. Huang, J. Yang, X. Yao, H. Chen, W. Zheng, ChemMedChem 2012, 7, 1661 – 1668; b) H. Suzuki, M. Yamazaki, K. Chiba, Y. Uemori, H. Sawanishi, Chem. Pharm. Bull. 2010, 58, 168 – 171. [4] a) Y. Koriyama, K. Chiba, M. Yamazaki, H. Suzuki, K. Muramoto, S. Kato, J. Neurochem. 2010, 115, 79 – 91; b) H. Suzuki, M. Yamazaki, K. Chiba, H. Sawanishi, J. Health Sci. 2007, 53, 730 – 733. [5] a) M. Yamazaki, K. Chiba, T. Mohri, Br. J. Pharmacol. 2005, 146, 662 – 669; b) Y. Koriyama, Y. Takagi, K. Chiba, M. Yamazaki, K. Arai, T. Matsukawa, H. Suzuki, K. Sugitani, H. Kagechika, S. Kato, J. Neurochem. 2011, 119, 1232 – 1242. [6] a) F. X. Guix, I. Uribesalgo, M. Coma, F. J. Munoz, Prog. Neurobiol. 2005, 76, 126 – 152; b) C. M. Troy, S. A. Rabacchi, W. J. Friedman, T. F. Frappier, K. Brown, M. L. Shelanski, J. Neurosci. 2000, 20, 1386 – 1392; c) A. Contestabile, E. Ciani, Neurochem. Int. 2004, 45, 903 – 914. [7] a) V. Calabrese, C. Mancuso, M. Calvani, E. Rizzarelli, D. A. Butterfield, A. M. Stella, Nat. Rev. Neurosci. 2007, 8, 766 – 775; b) C. S. Chan, T. S. Gertler, D. J. Surmeier, Trends Neurosci. 2009, 32, 249 – 256; c) J. Emerit, M. Edeas, F. Bricaire, Biomed. Pharmacother. 2004, 58, 39 – 46; d) V. L. Dawson, T. M. Dawson, J. Chem. Neuroanat. 1996, 10, 179 – 190; e) P. Kuppusamy, S. T. Ohnishi, Y. Numagami, T. Ohnishi, J. L. Zweier, J. Cereb. Blood Flow Metab. 1995, 15, 899 – 903; f) S. C. Bondy, S. Naderi, Neurosci. Lett. 1994, 168, 34 – 36; g) C. A. Massaad, Curr. Neuropharm. 2011, 9, 662 – 673. [8] a) F. Zhang, R. M. Casey, M. E. Ross, C. Iadecola, Stroke 1996, 27, 317 – 323; b) A. H. Neufeld, Surv. Ophthalmol. 1999, 43, S129 – S135; c) S. M. Ferreira, S. F. Lerner, R. Brunzini, P. A. Evelson, S. F. Llesuy, Am. J. Ophthalmol. 2004, 137, 62 – 69; d) D. Maneesh Kumar, N. Agarwal, J. Glaucoma 2007, 16, 334 – 343. [9] C. A. Massaad, Curr. Neuropharmacol. 2011, 9, 662 – 673. [10] R. R. Krishnamoorthy, A. F. Clark, D. Daudt, J. K. Vishwanatha, T. Yorio, Invest. Ophthalmol. Visual Sci. 2013, 54, 5712 – 5719. [11] a) K. Verma, S. K. Mehta, G. S. Shekhawat, Biometals 2013, 26, 255 – 269; b) H. F. Qian, W. Chen, J. J. Li, J. Wang, Z. Zhou, W. P. Liu, Z. W. Fu, Aquat. Toxicol. 2009, 92, 250 – 257; c) N. Ning, J. F. Hu, Y. H. Yuan, X. Y. Zhang, J. G. Dai, N. H. Chen, Acta Pharmacol. Sin. 2012, 33, 34 – 40. [12] W. K. Jeon, H. Y. Hong, B. C. Kim, Arch. Biochem. Biophys. 2011, 512, 119 – 125. [13] a) J. W. Kaspar, S. K. Niture, A. K. Jaiswal, Free Radical Biol. Med. 2009, 47, 1304 – 1309; b) Y. Kurauchi, A. Hisatsune, Y. Isohama, H. Katsuki, Neuroscience 2009, 158, 856 – 866. [14] H. T. Wang, X. H. Zhou, J. C. Huang, N. Mu, Z. L. Guo, Q. Wen, R. K. Wang, S. R. Chen, Z. P. Feng, W. H. Zheng, Psychopharmacology 2013, 228, 129 – 141. [15] H. Wang, X. Duan, Y. Ren, Y. Liu, M. Huang, P. Liu, R. Wang, G. Gao, L. Zhou, Z. Feng, W. Zheng, Mol. Neurobiol. 2013, 47, 24 – 36. Received: January 22, 2014 Published online on April 6, 2014
ChemMedChem 2014, 9, 1397 – 1401
1401
