Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 Contents lists available at ScienceDirect 2 3 4 5 6 journal homepage: www.elsevier.com/locate/jep 7 8 9 Research paper 10 11 12 13 14 15 Antonio Currais a,n, Chandramouli Chiruta a, Marie Goujon-Svrzic a, Gustavo Costa b,c, 16 Tânia Santos b,c, Maria Teresa Batista b,c, Jorge Paiva d, Maria do Céu Madureira d, 17 a 18 Q1 Pamela Maher a 19 The Salk Institute for Biological Studies, 10010 N. Torrey Pines Road, La Jolla, CA 92037, USA 20 Q2 b Center for Pharmaceutical Studies, Faculty of Pharmacy, University of Coimbra, 3000-548 Coimbra, Portugal c Center for Neurosciences and Cell Biology, University of Coimbra, Largo Marquês de Pombal, 3004-517 Coimbra, Portugal 21 d Centre for Functional Ecology, Department of Life Sciences, University of Coimbra, 3004-516 Coimbra, Portugal 22 23 24 art ic l e i nf o a b s t r a c t 25 26 Article history: Ethnopharmacological relevance: Alzheimer's disease (AD) neuropathology is strongly associated with the 27 Received 7 February 2014 activation of inflammatory pathways, and long-term use of anti-inflammatory drugs reduces the risk of 28 Received in revised form developing the disease. In S. Tomé e Príncipe (STP), several medicinal plants are used both for their 29 5 June 2014 positive effects in the nervous system (treatment of mental disorders, analgesics) and their antiAccepted 18 June 2014 30 inflammatory properties. The goal of this study was to determine whether a phenotypic, cell-based 31 screening approach can be applied to selected plants from STP (Voacanga africana, Tarenna nitiduloides, 32 Keywords: Sacosperma paniculatum, Psychotria principensis, Psychotria subobliqua) in order to identify natural 33 Aging compounds with multiple biological activities of interest for AD therapeutics. Alzheimer 34 Materials and methods: Plant hydroethanolic extracts were prepared and tested in a panel of phenotypic Dementia screening assays that reflect multiple neurotoxicity pathways relevant to AD—oxytosis in hippocampal 35 Inflammation nerve cells, in vitro ischemia, intracellular amyloid toxicity, inhibition of microglial inflammation and 36 Neurodegenerative disorders nerve cell differentiation. HPLC fractions from the extract that performed the best in all of the assays 37 Traditional medicine Africa were tested in the oxytosis assay, our primary screen, and the most protective fraction was analyzed by 38 mass spectrometry. The predominant compound was purified, its identity confirmed by ESI mass 39 spectrometry and NMR, and then tested in all of the screening assays to determine its efficacy. 40 Results: An extract from the bark of Voacanga africana was more protective than any other plant extract 41 in all of the assays (EC50s r2.4 mg/mL). The HPLC fraction from the extract that was most protective 42 against oxytosis contained the alkaloid voacamine (MW¼ 704.90) as the predominant compound. 43 Purified voacamine was very protective at low doses in all of the assays (EC50sr 3.4 mM). 44 Conclusion: These findings validate the use of our phenotypic screening, cell-based assays to identify 45 potential compounds to treat AD from plant extracts with ethnopharmacological relevance. Our study identifies the alkaloid voacamine as a major compound in Voacanga africana with potent neuroprotective 46 activities in these assays. 47 & 2014 Published by Elsevier Ireland Ltd. 48 49 50 51 52 Abbreviations: Aβ, amyloid beta peptide; AD, Alzheimer's disease; APP, amyloid 1. Introduction precursor protein; ATP, adenosine triphosphate; BBB, blood–brain barrier; CLogP, 53 lipophilicity; CNS, Central nervous system; DMEM, Dulbecco's modified Eagle's 54 medium; ESI, Electrospray ionization; FCS, fetal calf serum; GSH, glutathione; HBA, Alzheimer's disease (AD) is the most common form of demen55 hydrogen bond acceptor; HBD, hydrogen bond donor; HPLC, High-performance tia in the elderly. It is characterized by the presence of senile 56 liquid chromatography; HT22, mouse hippocampal nerve cells; IAA, iodoacetic plaques, neurofibrillary tangles and neuronal loss associated with 57 acid; LPS, lipopolysaccharide; MC65, human nerve cells; MTT, 3-(4, 5other age-related detrimental events such as increased oxidative dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide; MW, molecular weight; 58 N9, mouse microglial cells; NGF, nerve growth factor; NMR, nuclear magnetic stress, reduced energy metabolism and inflammation (Schubert 59 resonance; Opti-MEM, Opti-minimal essential media; PC12, rat pheochromoand Maher, 2012). Therefore, AD is multi-factorial in the sense that 60 cytoma cells; ppm, parts per million; STP, S. Tomé e Príncipe; TLC, thin layer there are a large number of mechanisms that can contribute to the 61 chromatography; TMS, tetramethylsilane; tPSA, topological polar surface area n disease and, specifically, nerve cell death. Many, if not most, of Corresponding author. Tel.: þ 1 858 453 4100x1480; fax: þ1 858 535 9062. 62 these mechanisms can be reproduced in cell culture assays. The E-mail address: [email protected] (A. Currais). 63 64 http://dx.doi.org/10.1016/j.jep.2014.06.046 65 0378-8741/& 2014 Published by Elsevier Ireland Ltd. 66

Journal of Ethnopharmacology

Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

67 Fieldwork was carried out with the collaboration of more than 50 traditional healers from both islands. Information about the Q3 68 medical ideology, including basic conceptual and philosophical Q4 69 70 aspects, classification of diseases and treatments, was recorded 71 (Madureira et al., 2002, 2008; Madureira, 2006). 72 In this study, five different species of plants—Voacanga africana, 73 Sacosperma paniculatum, Psychotria subobliqua, Psychotria princi74 pensis and Tarenna nitiduloides, previously collected, were selected 75 based on ethnopharmacological data or pre-existing biological and 76 pharmacological information. Voacanga africana, Sacosperma pani77 culatum and Psychotria subobliqua were selected for their local 78 medicinal properties in STP, namely their positive effects in the 79 nervous system (NS) and their anti-inflammatory activities 80 (Madureira et al., 2002, 2008; Madureira, 2006) (Table 1). There 81 is strong evidence from both biochemical and neuropathological 82 data for an activation of inflammatory pathways in the brains of 83 AD patients, and long-term use of anti-inflammatory drugs is 84 linked with a reduced risk of developing the disease (Wyss-Coray 85 and Rogers, 2012). Therefore, plants with anti-inflammatory properties represent good candidates for the treatment of AD, and this 86 was one of our primary criteria in selecting the above plants for 87 further study. Of these species, Voacanga africana is the only one 88 whose medicinal properties have been examined (Koroch et al., 89 2009). Voacanga africana is traditionally used to treat a wide 90 range of conditions in Africa, including leprosy, diarrhea, general91 ized edema, mental disorders and as an analgesic and anti92 inflammatory (Burkill, 1985; Olaleye et al., 2004; Koroch et al., 93 2009). In STP, Voacanga africana is also used as a hypotensive 94 and to reduce body aches and trauma (Madureira et al., 2002; 95 Madureira, 2006). Sacosperma paniculatum is used in Africa to 96 treat arthritis and rheumatism (Burkill, 1985), and in STP as a body 97 analgesic and to treat intestinal cramps, liver disease and hernias 98 (Madureira et al., 2002, 2008; Madureira, 2006). Tarenna nitidu99 loides and Psychotria principensis are two species endemic to STP 100 that are not used by the local healers but that were selected 101 because of their affinity (Rubiaceae family) to Psychotria subobli102 qua, which is locally used to treat toothaches and mouth inflam103 mation (Madureira et al., 2002, 2008; Madureira, 2006). Several 104 Rubiaceae, namely Psychotria species, present analgesic and anti105 106 inflammatory properties (Burkill, 1985) and have been studied in 107 the context of neurodegenerative diseases (Passos et al., 2013). 108 Since Tarenna nitiduloides and Psychotria principensis have no 109 reported ethnopharmacological relevance, they were chosen as 110 negative controls. 111 112 Table 1 113 Descriptions of the extracts from the five STP medicinal plants selected based on their potential positive effects in the CNS and anti-inflammatory activity used in this study. 114 115 Extract Botanical name Local name Therapeutic use Plant part used Voucher No. (COI) (family) 116 117 Leavesþstems 46 MCM/2011 A Voacanga africana Cata-manginga, Leprosy, diarrhea, generalized edema, mental disorders, 118 Stapf (Apocynaceae) Cata-kiô analgesic, anti-inflammatory (Burkill, 1985; Olaleye et al., 119 2004; Koroch et al., 2009), hypotensive, body aches/trauma (Madureira et al., 2002; Madureira, 2006) 120 B Voacanga africana Cata-manginga, Same as A Bark 46 MCM/2011 121 Stapf (Apocynaceae) Cata-kiô 122 C Tarenna nitiduloides – – Leavesþstems 19A;19B MCM/ 123 G. Taylor (Rubiacea) 2011 Leavesþstems 50 MCM/2011 Gligô-d’obô Arthritis, rheumatism (Burkill, 1985), analgesic, intestinal D Sacosperma 124 cramps, paniculatum (Benth.) 125 liver disease, hernias (Madureira et al., 2002, 2008; G. Taylor (Rubiaceae) 126 Madureira, 2006) 127 E Psychotria principensis – – Leavesþstems 20 MCM/2011 128 G. Taylor (Rubiaceae) F Psychotria subobliqua Kuako-maguita Toothaches, mouth inflammation (Madureira et al., 2002, Leavesþstems 49 MCM/2011 129 Hiern (Rubiaceae) 2008; Madureira, 2006) 130 G Psychotria subobliqua Kuako-maguita Same as F Fruits 49 MCM/2011 131 Hiern (Rubiaceae) 132

current single-target approach to the development of anti-AD therapies has been unsuccessful (Frautschy and Cole, 2010). An alternative approach is to identify small molecules that have a broad range of biological activities that are relevant to AD (Schubert and Maher, 2012). Our laboratory has developed a set of cell-based phenotypic screening assays that reflect multiple pathological features associated with AD (Schubert and Maher, 2012). Importantly, by means of these assays we have successfully identified compounds that present beneficial therapeutic efficacy in in vivo models of AD (Ishige et al., 2001; Sagara et al., 2004; Liu et al., 2008; Chen et al., 2011; Chiruta et al., 2012; Schubert and Maher, 2012; Prior et al., 2013, Currais et al., 2014). However, there is a great need for additional compounds that have a therapeutic potential for the treatment of AD. Although we have previously applied our phenotypic screening approach to test pure natural compounds and their derivatives, we have never taken advantage of the rich ethnopharmacological knowledge available from historically relevant plants to directly test crude extracts and identify active components. Traditional herbal medicines have been used for a long time in many countries all over the world as memory enhancers or to treat dementia-related disorders, and have recently become more popular. Ginkgo biloba leaf and Lycium barbarum fruit extracts are used as memory enhancers, and also have strong anti-oxidant and anti-inflammatory effects (Kim and Oh, 2012). A few clinical studies have been conducted to address the potential of herbal medicines or natural compounds to treat AD but to date none has been conclusive (Kim and Oh, 2012). S. Tomé e Príncipe (STP) is part of the Guinean Forests of West Africa Hotspot, one of the eight biodiversity hotspots in Africa (Hotspots, 2014), and is endowed with a unique endemic plant diversity. Traditional medicine plays a crucial role in STP as the population has few means of accessing modern medical treatment. Medicines prepared from plants have been used for centuries in STP and have been proven to be safe and efficient (Madureira et al., 2008). The evolution of the local traditional medicine in STP benefited from the privileged location of the islands which were an important crossroads of culture and knowledge between the Mediterranean, Africa, South America and Asia during the Age of the Discovery in the 15th century (Madureira et al., 2008). Since 1993 several ethnopharmacological surveys have been conducted in STP with the collaboration of the Ministry of Health. Vernacular names and the specialized therapeutic use of more than 325 plants were documented, including over 1500 detailed medical recipes.

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

All plant extracts were tested in a panel of five phenotypic screening assays—oxidative glutamate toxicity (oxytosis) in HT22 mouse hippocampal nerve cells, in vitro ischemia in HT22 hippocampal nerve cells, intracellular amyloid toxicity in MC65 human nerve cells, inhibition of inflammation mediated by microglial activation using N9 mouse microglial cells and differentiation of rat PC12 cells. These assays reflect multiple, age-associated neurotoxicity pathways of particular relevance to AD, such as increased oxidative stress and glutathione (GSH) depletion, reduced energy metabolism, accumulation of misfolded, aggregated proteins and inflammation (Chiruta et al., 2012; Schubert and Maher, 2012). In addition, these particular models were selected to provide a replicable, cost- and time-effective screening approach. The main objective of this study was to test whether these assays can be used to identify single neuroprotective compounds from extracts of selected plants of STP. Briefly, our workflow consisted of testing extracts from the different candidate species and selecting the extract that performed best in all of the different assays for subsequent fractionation. The resultant individual fractions were tested in the oxytosis assay and the predominant compound of the most active fraction was purified, structurally determined, and its efficacy tested in the complete screening assay panel.

2. Materials and methods 2.1. Plant material The fresh leaves, stems, fruits and bark of Voacanga africana, Tarenna nitiduloides, Sacosperma paniculatum, Psychotria principensis and Psychotria subobliqua were collected in S. Tomé and in Príncipe islands (S. Tomé e Príncipe, Gulf of Guinea) between January and February 2011, under the guidance of the local healers. The identification and authentication were done by professor Jorge Paiva from the Instituto Botânico da Universidade de Coimbra (COI), Portugal, where voucher specimens were deposited (Table 1). The nomenclature follows Figueiredo et al. (2011) and AFPD (2009). 2.2. Preparation of extracts Dried material was powdered and hydroethanolic extracts were prepared with a 70% aqueous ethanol solution, in a proportion of 1:10 (w/v), by maceration for 10 days at room temperature. The extracts were then filtered under vacuum and the marc of each material was submitted to the same extraction process two times more. The three filtered extracts were mixed and concentrated to a small volume on a rotary evaporator at reduced pressure and then lyophilized. Samples were frozen at  20 1C in a N2 atmosphere for long-term storage. The extraction yields obtained were: 12.32%, 15.83%, 27.78%, 13.46%, 12.42%, 37.03% and 23.11% expressed in dry weight, for extracts A, B, C, D, E, F and G, respectively. Powdered extracts were resolubilized in ethanol prior to testing in the cell-based assays. 2.3. Phenotypic screening assays All reagents were obtained from Sigma-Aldrich (St. Louis, MO, USA), unless otherwise stated. 2.3.1. Oxytosis This assay, also called oxidative glutamate toxicity, tests the ability of compounds to rescue cells from oxidative stress-induced programmed cell death caused by GSH depletion after treatment with glutamate (Tan et al., 2001). A reduction in GSH is seen in the aging brain and is accelerated in AD (Currais and Maher, 2013). The

3

depletion of GSH from cells leads to lipoxygenase activation, reactive oxygen species production and calcium influx which initiates a form of programmed cell death with features similar to those implicated in the nerve cell damage seen in AD (Sonnen et al., 2008). Because of the generality of the toxicity pathway in oxytosis and its mechanistic association with aging and AD, it is used as our primary screen. In this assay, 5  103 HT22 mouse hippocampal nerve cells, grown in highglucose Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal calf serum (FCS) (Hyclone, Logan, UT, USA), were plated in 96-well plates. After 24 h of culture, the medium was exchanged with fresh medium and 5 mM glutamate and the indicated concentrations/dilutions of extracts/ fractions/compounds were added. After 24 h of treatment, viability was measured by the 3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide (MTT) assay as previously described (Tan et al., 2001). Results were confirmed by visual inspection of the wells. 2.3.2. In vitro ischemia A breakdown in neuronal energy production leading to adenosine triphosphate (ATP) loss is associated with nerve cell damage and death in AD (Saxena, 2012). In order to induce ATP loss, we used the compound iodoacetic acid (IAA) in combination with the HT22 hippocampal nerve cells. IAA has been used in a number of other studies to induce in vitro ischemia and causes a rapid loss of ATP (Winkler et al., 2003; Maher et al., 2007). HT22 cells were seeded onto 96 well plates as described in the oxytosis assay. The medium was exchanged 24 h later with fresh medium and the cells were treated with 20 mM IAA alone (which results in 90–95% cell death) or in the presence of extracts/compounds. After 2 h the medium in each well was aspirated and replaced with fresh medium without IAA but containing the extracts/compounds. After 24 h of treatment, viability was measured by the MTT assay. Results were confirmed by visual inspection of the wells. 2.3.3. Intracellular amyloid toxicity Accumulation of intracellular amyloid beta peptide (Aβ) is considered by many as being a primary toxic event in AD. The human nerve cell line MC65 conditionally expresses the C99 fragment of the amyloid precursor protein (APP) leading to the accumulation of intracellular Aβ. The MC65 cells are routinely grown in the presence of tetracycline and, following its removal, the expression of C99 is induced and the cells die within 4 days because of the accumulation of intracellular, toxic protein aggregates (Sopher et al., 1996; Liu et al., 2008). Briefly, MC65 cells were regularly grown in high-glucose DMEM supplemented with 10% FCS. For the assay, cells were dissociated and plated at 4  105 cells per 35 mm tissue culture dishes in Opti-minimal essential media (Opti-MEM) in the presence (no induction) or absence (APP-C99 induced) of 1 μg/mL tetracycline in the presence or absence of the indicated extracts/compounds. At day 4, the control cells in the absence of tetracycline were dead, and cell viability was determined by the MTT assay, as previously described (Liu et al., 2008). 2.3.4. Inhibition of microglial activation Inflammation is a major feature of AD (Wyss-Coray and Rogers, 2012). Microglia are the resident immune cell population of the CNS and activated, pro-inflammatory microglia are implicated in the pathogenesis of AD (Wyss-Coray and Rogers, 2012). Activated microglia produce a wide array of pro-inflammatory and cytotoxic factors including cytokines, free radicals, excitatory neurotransmitters and eicosanoids that may work in concert to promote neurodegeneration (Wyss-Coray and Rogers, 2012). Thus, inhibiting the generation of activated pro-inflammatory microglia is another important therapeutic target for AD. Briefly, mouse N9 microglial cells were grown in high-glucose DMEM supplemented

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

with 10% FCS. For the assay the cells were plated at 5  105 cells in 35 mm tissue culture dishes. After growth overnight, the cells were then treated with 10 mg/mL bacterial lipopolysaccharide (LPS) alone or in the presence of the extracts/compounds. After 24 h the medium was removed, spun briefly to remove floating cells and 100 mL assayed for nitrite using 100 mL of the Griess Reagent in a 96 well plate. After incubation for 10 min at room temperature the absorbance at 550 nm was read on a microplate reader, as described previously (Chiruta et al., 2012). 2.3.5. PC12 differentiation Connections between nerve cells are altered in AD. Thus, compounds that can promote the regeneration of these connections might be of particular benefit, thereby promoting the recovery of higher neuronal function. As a model for this property, we used neurite outgrowth in rat PC12 cells, a well-studied model system of neuronal differentiation. In response to neurotrophic factors such as nerve growth factor (NGF), PC12 cells undergo a series of physiological changes culminating in a phenotype resembling that of sympathetic neurons (Keegan and Halegoua, 1993). In this assay, PC12 cells (originally obtained from Greene and Tischler (1976)) grown in high-glucose DMEM supplemented with 10% FCS and 5% horse serum (Invitrogen, Carlsbad, CA, USA) were plated in 35 mm tissue culture dishes. After 3 days of growth, the medium was replaced with serum-free N2 medium (Invitrogen, Carlsbad, CA, USA) and the cells were treated with the extracts/ compounds. After 24 h the cells were scored for the presence of neurites. PC12 cells produce neurites much more rapidly when treated in N2 medium than when treated in regular growth medium. For each treatment, 100 cells in each of three separate fields were counted. Cells were scored positive if one or more neurites 41 cell body diameter in length were observed (Sagara et al., 2004). 2.4. High-performance liquid chromatography (HPLC) fractionation Ethanol extracts were diluted 1/100 and 80 mL was injected into an HPLC system equipped with a Targa C18 5 mm 250  4.6 mm2 reverse phase column and a UV detector set at 330 nm and a flow rate of 1 mL/min. Samples were eluted with a linear gradient of 10–100% acetonitrile in 0.1% trifluoroacetic acid and fractions were collected manually. Fractions were dried in a Speed-Vac vacuum concentrator, reconstituted in 20 mL ethanol, and tested at 1/100 in the oxytosis assay. 2.5. Electrospray ionization (ESI) mass spectrometry and nuclear magnetic resonance (NMR) analyses ESI mass analysis was carried out with a Thermo Scientific LTQ Orbitrap-XL spectrometer at the Salk Institute (La Jolla, CA, USA). 1 H NMR and 13C NMR were recorded at 500 and 125 MHz, respectively, on a Bruker DMX-500 spectrometer at NuMega Resonance Lab (San Diego, CA, USA) using the indicated solvents. Chemical shift (δ) is given in parts per million (ppm) relative to tetramethylsilane (TMS) as an internal standard. Coupling constants (J) are expressed in hertz (Hz), and conventional abbreviations used for signal shape are as follows: s ¼ singlet; d ¼doublet; t¼ triplet; m ¼multiplet; dd ¼doublet of doublets; br s ¼ broad singlet. The structure of voacamine was characterized and confirmed by ESI mass spectrometry (Fig. 3b), 1H NMR and 13C NMR (Fig. 4a and b), in agreement with the literature (Medeiros et al., 1999). 1H NMR and 13C NMR spectral data of voacamine: 1H NMR (CDCl3, 500 MHz) δ ppm 0.87 (t, J¼ 7.5 Hz, 3H-18), 1.09 (brdd, 1Hβ-15), 1.27 (m, 1H-20), 1.41 (m, 1H-19), 1.53 (m, 1H-19), 1.67 (d, J¼6.5 Hz, 3H-180 ), 1.68 (1Hα-15), 1.76 (brd, 1Hβ-17), 1.80

(m, 1H-170 ), 2.01 (m, 1H-14), 2.46 (s, 3H–OMe-220 ), 2.48 (1Hα-17) 2.48 (1Hα-170 ), 2.66 (s, 3H–NMe-40 ), 2.73 (brd, 1Hα-3), 2.78 (brs, 1H30 ), 2.86 (m, 1Hβ-3), 2.96 (m, 2Hβ-6, 210 ), 3.09 (m, 1Hα-6), 3.17 (m, 1Hα-5), 3.33 (m, 1Hβ-60 ), 3.36 (m, 1Hβ-5), 3.50 (s, 2H-60 , 21), 3.65 (s, 3H–OMe-22), 3.79 (m-2H-140 , 210 ), 3.99 (brs, 3H–OMe-10), 4.13 (m, 1H-50 ), 5.12 (brd, 1H-160 ), 5.37 (brs, 1H-190 ), 6.74 (brs, 1H-12), 6.92 (s, 1H-9), 7.05–7.07 (m, 3H-100 ,110 ,120 ), 7.47 (brs, 1H–NH-1), 7.56 (m, 1H-90 ), 7.70 (brs, 1H–NH-10 ); 13C NMR (CDCl3, 125 MHz) δ ppm 11.82, 12.61, 2.45, 26.98, 27.57, 29.92, 32.20, 33.59, 36.45, 36.73, 39.23, 42.20, 50.25, 52.06, 52.51, 52.66, 53.34, 53.62, 55.18, 56.35, 57.44, 60.27, 99.51, 110.10, 110.23, 110.49, 117.60, 119.30, 121.90, 129.86, 130.50, 136.09, 137.54, 138.28, 175.41; MS (ESI): m/z calculated for C43H52N4O5 ([Mþ H] þ ) 705.4010; found 705.3962 ([Mþ H] þ ). 2.6. Voacamine purification Solvents used for chromatographic analysis were HPLC or ACS reagent grade and were purchased from Fisher Scientific Co (Pittsburg, PA, USA). Thin layer chromatography (TLC) used EMD silica gel F-254 plates (thickness of 0.25 mm). Flash chromatography used EMD silica gel 60, 230–400 mesh, which was purchased from EMD Chemicals (San Diego, CA, USA). Extract B from Voacanga africana bark (400 mg) was dissolved in a minimal amount of 1:1 MeOH/CH2Cl2, adsorbed on silica gel, then loaded onto a wet (CH2Cl2) column and purified by flash chromatography using 2–10% MeOH/CH2Cl2 as an eluent. This gave 30 mg (7.5%) of pure voacamine and 150 mg (37.5%) of an inseparable mixture of voacamine with other compounds. 2.7. Statistical analysis The EC50s were determined from sigmoidal dose response curves using GraphPad Prism 4. All of the experiments were done at least in triplicate and repeated at least three times. Multiple groups were compared using one-way ANOVA followed by Newman–Keuls post-hoc test. GraphPad Prism 4 was used for the statistical analyses. Data are expressed as mean 7SD, and significance of difference is indicated as nP o0.05, nnP o0.01 and nnn P o0.001.

3. Results 3.1. Phenotypic screening of the plant extracts In order to identify plants with multiple medicinal properties relevant to AD, we selected five species of plants, three of which are used by the local traditional healers in STP to treat conditions associated with mental disorders and inflammation (Voacanga africana, Sacosperma paniculatum, Psychotria subobliqua), and two endemic species not used locally (Tarenna nitiduloides, Psychotria principensis), belonging to the Rubiaceae family served as negative controls. Extracts from these plants were prepared and tested in our panel of cell-based screening assays: oxytosis, in vitro ischemia, intracellular amyloid toxicity, inhibition of inflammation mediated by microglial activation and differentiation of PC12 cells. The results are presented in Table 2 and show that extracts A and B from Voacanga africana demonstrated the lowest EC50s in the first four assays. With the exception of the inhibition of microglial activation, extract B from Voacanga africana bark was significantly more protective than extract A from Voacanga africana leaves and stems. In addition, extract B also induced PC12 differentiation, which was not the case for extract A. Extract G, prepared from Psychotria subobliqua fruits, was the least active in all of the assays. Extract F from the leaves and stems

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

5

Table 2 Biological activities of the plant extracts in the five phenotypic screening assays relevant to AD: oxytosis, in vitro ischemia, intracellular amyloid toxicity, inhibition of microglial activation and ability to promote differentiation of PC12 cells. Extract Oxytosis EC50 (lg/mL)a

1.2 7 0.2 0.34 7 0.2 15 72 3.8 7 1.2 10 74 10.6 7 3 1007 10

A B C D E F G a b

In vitro ischemia EC50 (lg/mL)a

Intracellular amyloid toxicity EC50 (lg/mL)a

Inhibition of PC12 cells microglial differentiation activation a EC50 (lg/mL)

1 70.2 0.17 0.01 o2 o2 107 2 2 70.24 NDb

0.1147 0.06 o 0.02 0.0727 0.06 0.168 7 0.08 0.228 7 0.06 o 0.02 NDb

1.6 7 0.16 2.4 70.1 17.6 7 4 22.6 74 10 71.2 22.6 71.6 NDb

No Yes Yes No No Yes NDb

EC50 (mg/mL)—half maximal effective concentrations. ND—not determined. Fig. 2. Biological activity of the fractions prepared by HPLC in the oxytosis assay, our primary screening model. Fraction 13 showed the highest level of protection in the oxytosis assay. nPo 0.05, nnnP o0.001, one-way ANOVA followed by Newman– Keuls post-hoc test (n¼ 3). All data are mean 7 SD.

dilution in the oxytosis assay, as described in Section 2.4. Fraction 13 was identified as the one that protected the most with 94% cell survival (Fig. 2). To a much lesser extent, fractions 6, 7 and 11 also showed some minor protection, 35.68%, 32.86% and 29.50% respectively. Therefore, fraction 13 was selected for further study. 3.3. Identification and purification of the natural compound voacamine

Fig. 1. HPLC profile of extract B prepared from the bark of Voacanga africana and the respective fractions that were collected (1–19).

of the same plant strongly protected against the toxicity of intracellular amyloid and induced PC12 differentiation; however, it was poorly effective in the other assays. Extract C from Tarenna nitiduloides also induced PC12 differentiation but was not particularly protective in the other assays. Extract D from Sacosperma paniculatum slightly protected HT22 cells in the oxytosis model and MC65 cells against amyloid toxicity. Finally, extract E from Psychotria principensis did not reveal any major effect in any of the assays, other than slightly preventing microglial activation. Overall, extract B performed better than any other extract in all of the assays. For this reason, it was selected for further analysis. 3.2. HPLC fractionation of Voacanga africana and screening in the oxytosis assay Given the complex composition of compounds in a crude extract, extract B was fractionated by HPLC (Fig. 1). Each fraction was then tested in the oxytosis assay (Fig. 2). Because of the generality of the toxicity pathway in oxytosis and its mechanistic association with both aging and AD, it is a good assay to quickly and simply screen the different fractions from the HPLC analysis of extract B. Multiple fractions corresponding to peaks in the HPLC profile (Fig. 1) were prepared identically and used at the same

In order to determine the composition of fraction 13 and possibly identify single biologically active compounds, the fraction was analyzed by ESI mass spectrometry (Fig. 3a). A strong peak was identified with a molecular weight (MW) identical to the dimeric indole alkaloid voacamine (MW¼704.90) (Fig. 3a), described previously in the literature (Koroch et al., 2009; Hussain et al., 2012). We have also identified two smaller peaks in the same fraction corresponding to voacafrine (MW¼382.45) and voacangine (MW¼368.47) (Koroch et al., 2009; Hussain et al., 2012). Note that the MW values in the mass spectrum correspond to [MþH] þ (Section 2.5). Because voacamine was the predominant compound in fraction 13, we proceeded to purify it as described in Section 2.6. The structure of voacamine was characterized and confirmed by mass spectrometry (Fig. 3b) and 1H NMR and 13C NMR (Fig. 4a and b), which were all in agreement with the literature (Medeiros et al., 1999). The 1H NMR and 13C NMR spectral data of voacamine are presented in Section 2.5. Given that fraction 13 showed by far the strongest biological activity we focused our analysis on this fraction. Analysis of the other fractions was beyond the scope of the current study but it is addressed in Section 4. 3.4. Validation of the biological activity of voacamine in the screening assays In order to determine whether the potent biological activity of extract B could be explained, at least in part, by the presence of voacamine, the biological activity of the purified compound was tested in all of the different screening assays (Table 3). As a comparison, the EC50 values of fisetin are also included. Fisetin is a neuroprotective and cognition-enhancing molecule which our laboratory has identified using the same screening assays (Ishige et al., 2001; Sagara et al., 2004) and that we have recently shown to protect in an in vivo model of AD (Chiruta et al., 2012; Currais et al., 2014). The results show that voacamine was indeed a

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 3. (a) Mass spectrometric analysis of the candidate HPLC fraction 13 identified the alkaloid voacamine (MW ¼ 704.90) as the predominant compound. The alkaloids voacafrine (MW ¼382.45) and voacangine (MW ¼ 368.47) were also identified. (b) Mass spectrometric analysis confirming the purification of voacamine (MW ¼ 704.90) from the bark extract of Voacanga africana.

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Fig. 4.

7

13

C NMR (a) and 1H NMR (b) spectra confirming the structure of voacamine.

very effective compound in all of the assays. Furthermore, with the exception of inhibition of microglial activation, the EC50 values of voacamine were substantially better than those of fisetin.

Table 4 presents the physicochemical properties—MW, lipophilicity (CLogP), topological polar surface area (tPSA) and hydrogen bond donor (HBD) and acceptor (HBA)—of fisetin, voacamine,

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Table 3 Biological activities of voacamine in the five phenotypic screening assays relevant to AD: oxytosis, in vitro ischemia, intracellular amyloid toxicity, inhibition of microglial activation and the ability to promote differentiation of PC12 cells. Sample

Oxytosis EC50 (lM)a

In vitro ischemia EC50 (lM)a

Intracellular amyloid toxicity EC50 (lM)a

Inhibition of microglial activation EC50 (lM)a

PC12 cells differentiation

Voacamine Fisetin

0.73 70.05 3 71

0.177 0.02 37 0.5

0.25 7 0.03 3.28 7 0.5

3.4 7 1 2.5 7 1

Yes Yes

Fisetin was used as a positive control. a

EC50 (mM)—half maximal effective concentrations.

Table 4 Physiochemical properties of fisetin, voacamine, voacafrine and voacangine. Compound

MW

CLogP

tPSA

HBD (n.OH,NH)

HBA (n.O,N)

Successful CNS drugs

r 400 286

r5 1.97

r 90 111.12

r3 4

r7 6

704

7.71

100

2

9

382

2.54

82.63

2

6

368

4.20

54.50

1

5

Successful CNS drugs should present MW r400, CLogP r 5, tPSAr90, HBD r3 and HBAr 7 to improve their penetration of the BBB (Pajouhesh and Lenz, 2005; Hitchcock and Pennington, 2006). The physicochemical properties were predicted using Molinspiration Chemoinformatics (http://www.molinspiration.com/cgi-bin/properties).

voacafrine and voacangine. Successful CNS drugs should present MW r400, CLogP r5, tPSAr90, HBDr3 and HBAr 7 to improve their penetration of the blood–brain barrier (BBB) (Pajouhesh and Lenz, 2005; Hitchcock and Pennington, 2006). With the exception of HBD, the physicochemical properties of voacamine were above the recommended values for a CNS-active drug, which indicates that voacamine is likely to have low CNS penetration. On the other hand, the voacamine metabolites, voacafrine and voacangine have significantly better physicochemical properties, falling in the range recommended for CNS drugs. However, we have not yet fully characterized their properties in our assays. Given that extract B from Voacanga africana bark was significantly more protective than extract A from leaves and stems of the same plant, we asked whether this could be explained by different amounts of voacamine in the two extracts. ESI mass spectrometry analysis of the two extracts A and B (Fig. 5a and b) identified voacamine (MW ¼ 704.90) only in extract B and not in extract A. This observation supports the idea that voacamine is likely to be the major compound that confers better neuroprotective properties to extract B in comparison to extract A (Table 2).

4. Discussion The present study addresses whether a cell-based phenotypic screening approach could be used to identify natural compounds with multiple protective properties of interest to AD therapeutics from plant extracts of STP. Our laboratory has extensive experience in drug discovery in the field of AD and we have screened and identified compounds from libraries of natural compounds and derivatives using this approach (Ishige et al., 2001; Sagara et al., 2004; Liu et al., 2008; Chen et al., 2011; Chiruta et al., 2012; Schubert and Maher, 2012). Specifically, we have developed a set of complementary, cell-based assays that mimic various aspects of AD, which led to the identification of the natural compound fisetin as a potential treatment for AD. This was first demonstrated both in vitro and in in vivo models of neurodegeneration (Maher et al., 2007, 2011; Maher, 2009; Gelderblom et al., 2012) and more recently in a mouse model of AD (Currais et al., 2014). Therefore, we applied here the same experimental design to determine if neuroprotective compounds with potential interest to AD could be identified directly from plant extracts used in the traditional

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

9

Fig. 5. Mass spectrometric analysis of extract A from leaves and stems (a) and extract B from bark (b) of Voacanga africana identified voacamine (MW ¼704.90) only in extract B.

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

medicine of STP. In order to provide a replicable, cost- and timeeffective screening approach, we chose models based on cell lines rather than primary neurons. Primary neuronal cultures add significantly more cost and, in our hands, are less time-effective and show more variability than models based on cell lines. In this study, we have successfully identified Voacanga africana as a potential source of biologically active compounds relevant to AD. To our knowledge, our study is the first one to use a cell-based, phenotypic screening approach to screen and identify neuroprotective compounds from plant extracts. Specifically, we have identified the alkaloid voacamine as a major active molecule in Voacanga africana (Burkill, 1985) with potent neuroprotective properties. The alkaloids account for the wide range of pharmacological activities and thus there are medicinal uses of Voacanga africana (Koroch et al., 2009). Voacamine has been shown to have anti-cancer activity, to work as an immunostimulant reinforcing the immune system and to exhibit antiplasmodial activity against malaria (reviewed in Koroch et al. (2009)). That voacamine which could also act as an important neuroprotective compound has not been reported. However, further conclusions must be made with caution when considering the physicochemical properties of voacamine. In addition to protection in in vitro assays, fundamental physicochemical features of CNS drugs must be taken into account regarding their ability to penetrate the BBB in vivo (Pajouhesh and Lenz, 2005; Hitchcock and Pennington, 2006). CNS drugs should present values of MW, CLogP, tPSA and HBD and HBA that have a smaller range than general therapeutics (Table 4) (Pajouhesh and Lenz, 2005; Hitchcock and Pennington, 2006). Voacamine presents poor physicochemical properties as a CNS drug, suggesting a low penetration of the BBB into the brain. In this regard, it is worth noting that smaller secondary metabolites of voacamine, voacafrine (MW ¼382.45) and voacangine (MW¼368.47) were also identified in the same fraction and could be protective as well. Both of these metabolites, voacafrine and voacangine, have significantly better physicochemical properties, falling in the range recommended for CNS drugs. Although we have not isolated and tested these molecules in our screening assays, we plan to pursue this in the future in order to determine their biological potential. The overall goal is to identify compounds with good potency in the multiple neuroprotective pathways while at the same time having physicochemical properties consistent with those of successful CNS drugs (Pajouhesh and Lenz, 2005; Hitchcock and Pennington, 2006). Extract A from Voacanga africana displayed higher EC50 values in the oxytosis, in vitro ischemia and intracellular amyloid toxicity assays and no differentiation of PC12 cells, when compared to extract B from a different part of the same plant. This could be explained by a different array of compounds in the two extracts given that extract A was prepared from leaves and stems and extract B from bark. Indeed, there are quantitative and qualitative variations in the alkaloid content and composition between different plant tissues and individual samples of Voacanga africana from different regions of West Africa. For instance, the total amount of alkaloids in root bark is 5–10%, in trunk bark 4–5%, in leaves 0.3–0.45% and in seeds 1.5–3.5% (Bisset, 1985). Voacamine is reported to be found in both stems and bark of Voacanga africana, although predominantly in bark (Koroch et al., 2009), so the differences between extracts A and B could be explained by different amounts of voacamine in these two extracts. The leaves contain a fairly complex mixture, with voaphylline, vobtusine and voalfolidine as the major alkaloids (Koroch et al., 2009). However, some authors also reported the presence of voacamine in leaf extracts of Voacanga africana from Ivory Coast and Zambia (Bisset, 1985). To clarify this point we carried out ESI mass spectrometry analysis of leaves and stems (extract A) and trunk bark (extract B) of Voacanga africana collected in STP. We have identified the

presence of the alkaloid voacamine only in the trunk bark extract and confirmed that this alkaloid is absent from the leaves and stems extract from STP Voacanga africana, thus supporting the idea that voacamine is an important contributor to the neuroprotective properties of extract B. Furthermore, both voacafrine and voacangine appear to be predominantly found in the bark (Koroch et al., 2009) and could also contribute to the differences between the extracts, along with other metabolites. Voacangine has psychoactive, anesthetic and anti-angiogenic properties (Kim et al., 2012), while voacafrine has never been studied. Despite the lack of good physicochemical properties of voacamine as a CNS drug candidate, our results demonstrate that voacamine is very protective in our in vitro assays and suggest that perhaps metabolites of voacamine, which share similar structural properties, might also present biological and pharmacokinetic properties of interest to AD. Similarly, other fractions of extract B could contain metabolites in minor amounts with potential protective character that only exist in the bark. In fact, other fractions of the bark extract also conferred some protection against oxytosis although to a much lesser extent than voacamine. We plan on addressing this in the future by characterizing some of the other fractions and by purifying and testing other metabolites in our screening assays. Regarding the other plant species that we investigated, it might be possible that they contain metabolites of interest as well, but if they are present in low concentrations they might be missed by our approach. However, we believe that our experimental design provides a practical and time-efficient way of screening for compounds, and our data here show that such approach can be successful. In addition, our study supports the notion that a selection of plants based on their individual ethnopharmacological information (local medicinal uses) provides higher chances of positive findings, as with Voacanga africana, rather than an indirect selection approach based on criteria related to biological and pharmacological information at the family and genus levels, as was the case for the negative controls Tarenna nitiduloides and Psychotria principensis. It is also important to mention that our workflow was facilitated by the existence of a vast descriptive literature on Voacanga africana and its metabolites (Koroch et al., 2009). In the case of new species or unstudied species, techniques such as NMR and X-ray crystallography would be required to clarify the identity of a given compound or describe its novelty. In conclusion, the present study validates the use of our approach using cell-based phenotypic screening assays to identify compounds of interest for the treatment of AD directly from plant extracts with ethnopharmacological relevance. It also identifies, for the first time, the alkaloid voacamine as a major compound in Voacanga africana with potent biological activity in several cell-based assays relevant to AD. Although voacamine does not have ideal physicochemical properties as a CNS drug, our data warrant further investigation to identify related metabolites with optimal medicinal and chemical properties. Once identified, the protective potential of these candidate compounds can be tested in animal models of AD.

Acknowledgments We would like to thank the traditional healers of S. Tomé e Príncipe, in particular to Sum Pontes and Sum Costa, and the Ministry of Health of STP. Team researchers have worked in full partnership with local institutions, traditional healers and communities in order to respectfully conduct research in the area of Indigenous Knowledge, assuring the intellectual property rights and the sharing of benefits that may

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

A. Currais et al. / Journal of Ethnopharmacology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

arise as a result of the study of STP natural resources (medicinal plants). References AFPD, 2009. African flowering plants database. 〈http://www.ville-ge.ch/musinfo/ bd/cjb/africa/〉. Bisset, N.G., 1985. Phytochemistry and pharmacology of voacanga species. Agricultural University Wageningen Papers, 85, pp. 83–122. Burkill, H.M., 1985. The Useful Plants of West Tropical Africa, volume 4: Families M-R. Royal Botanic Gardens, Kew. Chiruta, C., Schubert, D., Dargusch, R., Maher, P., 2012. Chemical modification of the multitarget neuroprotective compound fisetin. Journal of Medicinal Chemistry 55, 378–389. Currais, A., Maher, P., 2013. Functional consequences of age-dependent changes in glutathione status in the brain. Antioxidants & Redox Signaling 19, 813–822. Currais, A., Prior, M., Dargusch, R., Armando, A., Ehren, J., Schubert, D., Quehenberger, O., Maher, P., 2014. Modulation of p25 and inflammatory pathways by fisetin maintains cognitive function in Alzheimer's disease transgenic mice. Aging Cell 13, 379–390. Chen, Q., Prior, M., Dargusch, R., Roberts, A., Riek, R., Eichmann, C., Chiruta, C., Akaishi, T., Abe, K., Maher, P., Schubert, D., 2011. A novel neurotrophic drug for cognitive enhancement and Alzheimer's disease. PLoS One 6, e27865. Figueiredo, E., Paiva, J., Stevart, T., Oliveira, F., Smith, G.F., 2011. Annotated catalogue of the flowering plants of São Tomé and Príncipe. Bothalia 41, 41–82. Frautschy, S.A., Cole, G.M., 2010. Why pleiotropic interventions are needed for Alzheimer's disease. Molecular Neurobiology 41, 392–409. Gelderblom, M., Leypoldt, F., Lewerenz, J., Birkenmayer, G., Orozco, D., Ludewig, P., Thundyil, J., Arumugam, T.V., Gerloff, C., Tolosa, E., Maher, P., Magnus, T., 2012. The flavonoid fisetin attenuates postischemic immune cell infiltration, activation and infarct size after transient cerebral middle artery occlusion in mice. Journal of Cerebral Blood Flow & Metabolism 32, 835–843. Greene, L.A., Tischler, A.S., 1976. Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proceedings of the National Academy of Sciences of the United States of America 73, 2424–2428. Hitchcock, S.A., Pennington, L.D., 2006. Structure–brain exposure relationships. Journal of Medicinal Chemistry 49, 7559–7583. Hotspots. The biodiversity hotspots. 〈http://www.biodiversityhotspots.org〉 (accessed 04.02.2014). Hussain, H., Hussain, J., Al-Harrasi, A., Green, I.R., 2012. Chemistry and biology of the genus Voacanga. Pharmaceutical Biology 50, 1183–1193. Ishige, K., Schubert, D., Sagara, Y., 2001. Flavonoids protect neuronal cells from oxidative stress by three distinct mechanisms. Free Radical Biology and Medicine 30, 433–446. Keegan, K., Halegoua, S., 1993. Signal transduction pathways in neuronal differentiation. Current Opinion in Neurobiology 3, 14–19. Kim, H.G., Oh, M.S., 2012. Herbal medicines for the prevention and treatment of Alzheimer's disease. Current Pharmaceutical Design 18, 57–75. Kim, Y., Jung, H.J., Kwon, H.J., 2012. A natural small molecule voacangine inhibits angiogenesis both in vitro and in vivo. Biochemical and Biophysical Research Communications 417, 330–334. Koroch, A.R., Juliani, H.R., Kulakowski, H.A., Asante-Dartey, J., Simon, J.E., 2009. Voacanga africana: chemistry, quality and pharmacological activity (ACS Symposium Series). African Natural Plant Products: New Discoveries and Challenges in Chemistry and Quality, 1021, pp. 363–380. Liu, Y., Dargusch, R., Maher, P., Schubert, D., 2008. A broadly neuroprotective derivative of curcumin. Journal of Neurochemistry 105, 1336–1345.

11

48 Madureira, M.C., 2006. Etnofarmacologia e Estudo de Espécies com Actividade Biológica da Flora de S. Tomé e Príncipe (Doctoral thesis) in Pharmacognosy and 49 Phytochemistry, Faculdade de Farmácia, Universidade de Coimbra. 50 Madureira, M.C., Martins, A.P., Salgueiro, L., Paiva, J., Proença da Cunha, A., 2002. 51 Medicinal Plants and Traditional Medicine in the Gulf of Guinea – S. Tomé and Príncipe Islands, in Medicinal Plants. Ethnomedicine and Pharmacognosy Part 52 II, 7. Researchco Book Centre, New Delhi, India, pp. 361–381. 53 Madureira, M.C., Paiva, J.A., Fernandes, A.F., Gonçalves, A., Catalão, C., Fernandes, C., 54 Atalaia, J., Vieira, J., Gaspar, V., Pontes, S., Gino, S., Costa, S., 2008. Estudo Etnofarmacológico de Plantas Medicinais de S. Tomé e Príncipe. Ed. Ministério Q5 55 da Saúde STP/Fundação Calouste Gulbenkian. 56 Maher, P., Dargusch, R., Bodai, L., Gerard, P.E., Purcell, J.M., Marsh, J.L., 2011. ERK 57 activation by the polyphenols fisetin and resveratrol provides neuroprotection 58 in multiple models of Huntington's disease. Human Molecular Genetics 20, 261–270. 59 Maher, P., 2009. Modulation of multiple pathways involved in the maintenance of 60 neuronal function during aging by fisetin. Genes & Nutrition 4, 297–307. 61 Maher, P., Salgado, K.F., Zivin, J.A., Lapchak, P.A., 2007. A novel approach to screening for new neuroprotective compounds for the treatment of stroke. 62 Brain Research 1173, 117–125. 63 Medeiros, W.L.B., Vieira, I.J.C., Mathias, L., Braz-Filho, R., Leal, K.Z., Rodrigues-Filho, E., 64 Schripsema, J., 1999. Two known bis-indole alkaloids isolated from Tabernae65 montana laeta: complete 1H and 13C chemical shift assignments. Magnetic Resonance in Chemistry 37, 676–681. 66 Olaleye, S.B., Oke, J.M., Etu, A.K., Omotosho, I.O., Elegbe, R.A., 2004. Antioxidant and 67 anti-inflammatory properties of a flavonoid fraction from the leaves of 68 Voacanga africana. Nigerian Journal of Physiological Sciences 19, 69–76. Pajouhesh, H., Lenz, G.R., 2005. Medicinal chemical properties of successful central 69 nervous system drugs. NeuroRx 2, 541–553. 70 Passos, C.S., Simoes-Pires, C.A., Nurisso, A., Soldi, T.C., Kato, L., de Oliveira, C.M., 71 de Faria, E.O., Marcourt, L., Gottfried, C., Carrupt, P.A., Henriques, A.T., 2013. Indole alkaloids of Psychotria as multifunctional cholinesterases and mono72 amine oxidases inhibitors. Phytochemistry 86, 8–20. 73 Prior, M., Dargusch, R., Ehren, J.L., Chiruta, C., Schubert, D., 2013. The neurotrophic 74 compound J147 reverses cognitive impairment in aged Alzheimer's disease 75 mice. Alzheimer's Research & Therapy 5, 25. Sagara, Y., Vanhnasy, J., Maher, P., 2004. Induction of PC12 cell differentiation by 76 flavonoids is dependent upon extracellular signal-regulated kinase activation. 77 Journal of Neurochemistry 90, 1144–1155. 78 Saxena, U., 2012. Bioenergetics failure in neurodegenerative diseases: back to the future. Expert Opinion on Therapeutic Targets 16, 351–354. 79 Schubert, D., Maher, P., 2012. An alternative approach to drug discovery for 80 Alzheimer's disease dementia. Future Medicinal Chemistry 4, 1681–1688. 81 Sonnen, J.A., Breitner, J.C., Lovell, M.A., Markesbery, W.R., Quinn, J.F., Montine, T.J., 2008. Free radical-mediated damage to brain in Alzheimer's disease and its 82 transgenic mouse models. Free Radical Biology and Medicine 45, 219–230. 83 Sopher, B.L., Fukuchi, K., Kavanagh, T.J., Furlong, C.E., Martin, G.M., 1996. Neurode84 generative mechanisms in Alzheimer disease. A role for oxidative damage in 85 amyloid beta protein precursor-mediated cell death. Molecular and Chemical Neuropathology 29, 153–168. 86 Tan, S., Schubert, D., Maher, P., 2001. Oxytosis: a novel form of programmed cell 87 death. Current Topics in Medicinal Chemistry 1, 497–506. 88 Winkler, B.S., Sauer, M.W., Starnes, C.A., 2003. Modulation of the Pasteur effect in retinal cells: implications for understanding compensatory metabolic mechan89 isms. Experimental Eye Research 76, 715–723. 90 Wyss-Coray, T., Rogers, J., 2012. Inflammation in Alzheimer disease—a brief review 91 of the basic science and clinical literature. Cold Spring Harbor Perspectives in 92 Medicine 2, a006346. 93

Please cite this article as: Currais, A., et al., Screening and identification of neuroprotective compounds relevant to Alzheimer's disease from medicinal plants of S. Tomé e Príncipe. Journal of Ethnopharmacology (2014), http://dx.doi.org/10.1016/j.jep.2014.06.046i