Introduction

Folivory in bats was discovered more than sixty years ago (Greenhall, 1957; Van der Pijl, 1957), and researchers are still trying to understand the functional and evolutionary significance of chewing leaves of specific plant species by some frugivorous bats (Kunz and Díaz, 1995). In most cases of folivory, bats chew pieces of leaves to swallow the liquid portion presumably containing micronutrients, vitamins, and proteins, usually scarce in fruits, and drop the resulting fibrous mass (i.e., pellets) formed in their mouths while chewing (Kunz and Ingalls, 1994). In some bats with folivorous habits this strategy presumably allows them to obtain essential biological components by squeezing them from leaves without modifying and extending their digestive tracts to metabolize cellulose and/or process fiber (Kunz and Ingalls, 1994).

Folivory in bats has remained poorly studied especially in terms of the chemical properties of leaves, and their meticulous selection and processing. Kunz and Ingalls (1994) proposed several questions to address an explanation for folivory besides nutrition: Do bats select specific plant leaves? Do secondary compounds from leaves stimulate or inhibit reproductive activity? In particular, these last questions have been repeatedly postulated in the past, but the last one has not been experimentally tested, confirmed, or thoroughly investigated. Regarding this poorly explored hypothesis, Kunz and Díaz (1995) postulated that the frequent use of leaves by several bat species could allow males to obtain secondary metabolites from the liquid fraction from leaves (Janeczko and Skoczowski, 2005; Krishnarathi et al., 2014) that would function as hormone precursors affecting reproductive regulatory steps, and/or females could obtain secondary metabolites from the liquid fraction from leaves that would function as critical hormone precursors for pregnancy and lactation (Wickler and Seibt, 1976; Janeczko and Skoczowski, 2005; Krishnarathi et al., 2014).

Duque-Marquez et al. (2019) determined that leaves of Aspidosperma desmanthum (Apocynaceae) (Figure 1A) were frequently consumed by individuals of the large fruit-eating bat, Artibeus amplus (Chiroptera: Phyllostomidae). However, the highest consumption of leaves from this plant species was observed in August (Duque-Marquez et al., 2019), shortly before males showed maximum size of testes, and three months before all females were observed with advanced pregnancy (Ruiz-Ramoni et al., 2017). The clear and unequivocal predominance of leaves of A. desmanthum (Ruiz-Ramoni et al., 2011; Duque-Marquez et al., 2019) previous to important changes in reproductive status of males and females (Ruiz-Ramoni et al., 2017) made us wonder whether leaves of A. desmanthum have an abundant secondary metabolite (Pagare et al., 2015; Tarkowská, 2019), candidate to be a hormone or hormonal precursor (Janeczko and Skoczowski, 2005; Tarkowská, 2019). Thus, we hypothesized that A. desmanthum contains a secondary metabolite that could potentially become a hormonal precursor to be tested in future experiments. The aim of this study was to isolate and purify the predominant secondary metabolite in A. desmanthum through chromatographic analytical procedures with the hope of finding a target chemical component to test in future controlled experiments on the regulation of reproductive phenomena in A. amplus. Here we report the isolation and purification of stigmasterol from highly consumed leaves of A. desmanthum in the search of key elements to improve our understanding of folivory in frugivorous bats.

Materials and Methods

Plant material, collection and identification

Aspidosperma is a genus of moderate size in the family Apocynaceae, with over 50 species distributed in the Neotropical region from Mexico and the Antilles to Argentina, with a higher diversity in Brazil (Woodson, 1951;Endress et al., 2018). Aspidosperma desmanthum Benth. ex Müll. Arg. is a tree 13-25 m tall, and its exact distribution in Venezuela is unknown especially to the north of the Orinoco river. It seems to be a rare genus in the Venezuelan Andes, since it is neither included in detailed inventories (Ataroff and García-Nuñez, 2013), nor was abundant in the study area. Plant material (fresh mature leaves of A. desmanthum; Figure 1A) was collected near the cave where Artibeusamplus lives (08°00'N, 71°43'W), at an altitude of 1320masl in May 2010. The collection place is known as ‘Parque Las Escaleras’, near the village of Pregonero, Municipio Autónomo Uribante, Estado Táchira, Venezuela. A voucher specimen (J.M. Amaro-Luis, Nº 1652) was deposited at Herbarium MERF, Faculty of Pharmacy and Bioanalysis, Universidad de Los Andes (ULA), and its identity was confirmed by molecular (Ruiz-Ramoni et al., 2011) and morphological (J. Carmona Arzola, Department of Pharmacognosy and Organic Medicaments, Faculty of Pharmacy and Bioanalysis, ULA) procedures. Previous mentions of this species in Duque-Marquez et al. (2019) were made using the synonym A. cruentum, replaced here by the current valid name, A. desmanthum.

Extraction and isolation of the chemical compound from leaves

The main goal of our study was the isolation, purification, identification and structural characterization of the predominant secondary metabolite present in the leaves of A. desmanthum. To achieve this goal, we followed a methodology that included conventional procedures typical in phytochemical investigations: chromatographic techniques for isolation and purification, such as column (CC) and thin layer (TLC) chromatography; and spectroscopic methods for identification and structural characterization: infrared spectroscopy (IR), uni- (¹H- and ¹³C-NMR) and two-dimensional (¹H,¹H-COSY, HMQC, HMBC and NOESY) nuclear magnetic resonance and mass spectrometry (MS).

General experimental procedures

Melting point was determined with a Fisher-Johns apparatus and it has not been corrected. Optical activity was measured on 60Hz-Steeg & Reuter polarimeter using methanol as solvent. IR spectra were recorded on a Perkin-Elmer FT-1725X spectrophotometer as KBr pellets. ¹H-, ¹³C- and two-dimensional NMR spectra were run with a Bruker-Avance DRX-400 instrument, using CDCl. as solvent. HRMS were acquired on a VG Micromass ZAB-2F. TLC was carried out on 0.25mm layers of silica gel PF 254 (Merck); spots were visualized using UV light (254 and 365nm) and subsequently by spraying with ‘oleum’, an CH₃COOH-H₂O-H₂SO₄ mixture (20:4:1 v/v) and then heating with air flow at 100°C for a few minutes.VCC was performed with silica gel 60 (70-230 mesh).

Extraction and separation

Partially consumed mature leaves of A. desmanthum were always observed below the colony of bats. Mature leaves were sampled and analyzed. Air dried crushed leaves (≈390g) were extracted with methanol in a Soxhlet to give, after evaporation under reduced pressure at 40ºC in a rotary evaporator, 26.5g of crude extract. This extract was pre-adsorbed on silica gel 60 and chromatographed on the same adsorbent, using the Coll and Bowden (1986) vacuum liquid chromatography technique. The column was eluted with hexane, dichloromethane and methanol in binary mixtures of increasing polarity. Fractions of 0.5L were collected, concentrated in vacuo, and combined according to TLC similarity to afford twelve major fractions (A-L; Table I and Figure 2A). The combined major fraction ‘D’ (≈5.18g) was re-chromatographed on a silica gel column using hexane-dichloromethane (3:2; Figure 2B) to furnish an impure crystalline solid (≈0.418g; 1.57%), that was purified by preparative TLC (silica gel, three developments in hexane-ethyl acetate 4:1; Figure 2C). Crystallization from methanol provided pure white needles (≈78mg; 0.29%) detected in TLC plates as a red-brown spot. These yields, from the phytochemical point of view, are acceptable for a metabolite that can be considered abundant.

Stigmasterol identification

White needles; m.p.= 165-168ºC; [α]. -54.0º (MeOH, c 0.45). HR EI-MS: m/z: 478.3663 [M.] C_29.48O (requires 412.3705); LW EI-MS: m/z: 412.4 (53), 394.4 (17), 369.4 (10), 351.3 (20), 300.3 (29), 271.2 (46), 255.2 (60), 229.2 (13), 213.2 (28), 159.1 (53), 143.5 (49), 133.1 (52), 105.1 (50), 93.1(81), 65.3 (62), 55.1 (100); IR (KBr), ν_max (cm^-1): 3442 (O-H), 2938-2865 (C-H), 1605 (C=C), 1458 and 1375 (C-H), 1056 (C-O), 967 (=C-H). ¹H NMR (400MHz, CDCl.), δ. (ppm): 0.69 (3H, s, H-19), 0.79 (3H, d, J= 7.0Hz, H-26), 0.80 (3H, t, J= 7.0Hz, H-29), 0.84 (3H, d, J= 7.0Hz, H-27), 1.01 (3H, s, H-18), 1.03 (3H, d, J= 7.0Hz, H-21), 3.52 (1H, tdd, J= 4.5, 4.0 and 3.9Hz, H-3), 5.01 (1H, dd, J= 9.0 and 15.0Hz, H-23), 5.14 (1H, dd, J= 9.0 and 15.0Hz, H-22) and 5.35 (1H, d, J= 5.0Hz, H-6). ¹³C NMR (100 MHz CDCl.), δ. (ppm): 12.1 (CH₃, C-19), 12.3 (CH₃, C-29), 19.0 (CH₃, C-26), 19.5 (CH₃, C-18), 21.1 (CH₃, C-27), 21.1 (CH₂, C-11), 21.3 (CH₂, C-21), 24.4 (CH₂, C-15), 25.5 (CH₂, C-28), 29.0 (CH₂, C-16), 31.7 (CH₂, C-2), 31.7 (CH₂, C-7), 31.9 (CH, C-8), 31.9 (CH, C-25), 36.6 (>C<, C-10), 37.3 (CH₂, C-1), 39.7 (CH₂, C-12), 40.5 (CH, C-20), 42.3 (CH₂, C-4), 42.4 (>C<, C-13), 50.2 (CH, C-9), 51.3 (CH, C-24), 56.0 (CH, C-17), 56.9 (CH, C-14), 71.9 (CH-O-, C-3), 121.8 (=CH-, C-6), 129.3 (=CH-, C-23), 138.4 (=CH-, C-22), 140.8 (=C<, C-5).

Results

The followed methodology led to the isolation and purification of a single compound, whose identity was established by IR, EI-MS and 1D- and 2D-NMR studies: stigmasterol (Figure 1B), the most abundant secondary metabolite present in the leaves of A. desmanthum (Figure 1A).

Stigmasterol was obtained as white needles (m.p.= 165-168 ºC; [α] .: - 54.0°). It showed in TLC on silica gel GF₂₅₄ plates revealed with ‘oleum’ a red-brown spot typical of an isoprenoid derivative. It is evident from the molecular formula deduced by EI-MS, C_29.48O [m/z: 412 (M.)] and from the NMR data, that the product is a phytosterol. The presence in the molecule of six methyl groups, assigned through the HMQC and HMBC spectra (Figures 3A, B) {[δ.: 1.01 (H-18) ↔ δ.: 19.5 (C-18)]; [δ.: 0.69 (H-19) ↔ δ.: 12.1 (C-19); [δ.: 1.03 (H-21) ↔ δ.: 21.3 (C-21)]; [δ.: 0.79 (H-26) ↔ δ.: 19.0 (C-26)]; [δ.: 0.84 (H-27) ↔ δ.: 21.1 (C-27)]; [δ.: 0.80 (H-29) ↔ δ.: 12.3 (C-29)]}, unequivocally confirms this assumption. The fact that, according to .H NMR spectrum data, two methyl signals are singlets (angular methyl C-18 and C-19), three correspond to doublets (secondary methyl C-21 and isopropyl group C-26 and C-27), and one a triplet (primary methyl C-29), confirms that phytosterol belongs to the stigmastane series; this is also consistent with the presence in the molecule of seven sp. non-oxygenated methines (>CH, C-8, C-9, C-14, C-17, C-20, C-24 and C-25) and two sp. quaternary carbons (>C<, C-10 and C-13) as it is confirmed by the DEPT-135 and DEPT-90 traces in the ¹³C NMR spectrum (Figure 3A). The functional groups present in the molecule were identified from the spectral data as a trisubstituted double bond [δ.: 5.35, d, J= 5.0Hz, =C.-(H-6) ↔ δ.: 121.8, =.H-(C-6); δ.: 140.8, =C<(C-5)], an 1,2-disubstituted ‘trans’ double bond [δ.: 5.01 and 5,14, dd, J= 9.0 and 15.0Hz, =C.-(H-23 and H-22) ↔ δ.: 129.3 and 138.4, =CH (C-23 and C-22)], and a secondary alcohol [IR, ν_max: 3442 (OH) and 1056 (C-O); EI-MS, m/z: 394 [M.-H.O], RMN, δ.: 3.52, tdd, J= 4.5, 4.0 and 3.9Hz, >C.-O-(H-3) ↔ δ.: 71.9, >.H-O-(C-3)]. The configuration of the hydroxyl group (3βec.) of the secondary alcohol was deduced from the multiplicity (tdd) of the signal assigned to its geminal proton (>.H-O-) and from the values of its coupling constants (J_{3αax., 2βax.= .3αax., 4βax.}= 4.5Hz; J_{3αax., 2αec.}= 4.0Hz and J_{3αax., 2αec.}= 3.9Hz). Similarly, the ‘trans’ stereochemistry of the double bond Δ^22,23 was consistent with the value of its highest coupling constant (J_22,23= 15.0Hz).

The position of the three functional groups in the steroidal skeleton was unequivocally assigned by C-H interactions in the HMBC spectrum (Figure 3B). Thus, the H-3 ↔ C-4 correlation places the secondary hydroxyl group at C-3, which is consistent with biogenetic considerations; the position Δ^5,6 of the trisubstituted double bond was secured by the crossed peaks of H-6 with C-4, C-8 and C-10 and H-4 with C-5 and C-6; the location Δ^22,23 of the 1,2-disubstituted double bond, is consistent for the following correlation sequences: H-22 ↔ C-24 ↔ H-23 ↔ C-20 ↔ H-22 ↔ C-23 and H-21 ↔ C-22 ↔ H-23. The position in the molecule of all methyl groups was also fixed through the HMBC spectrum (Figure 3B).

Figure 3. Heteronuclear Multiple Quantum Coherence (HMCQ; A) and Heteronuclear Multiple Bond Correlation (HMBC; B) spectra of stigmasterol.

Discussion

The preceding analysis indicates that the structure of the compound found in leaves of A. desmanthum corresponds to 3β, 22.-stigmasta-5,22-dien-3-ol, named in the literature as stigmasterol. This compound is widely distributed in plants, it is efficiently used as a starting material for the synthesis of sex hormones, and has a wide range of biological and pharmacological activities (Janeczko and Skoczowski, 2005; Kaur et al., 2011;Tarkowská, 2019). Stigmasterol is often present in plants mixed with β-sitosterol and campesterol, but there are numerous reports in the literature (Kaur et al., 2011) of detection and isolation of only one of these phytosterols. In TLC these three phytosterols have very similar Rf and our NMR data indicate that there are no traces of β-sitosterol or campesterol in our isolated sample. The presence of this phytosterol in species of the genus Aspidosperma has not been well studied as it has only been reported in two species, A. nitidum Benth ex Müll. Arg. (Pereira et al., 2006) and A. parvifolium A. DC. (Jacome et al., 2004).

Several biological activities reported for stigmasterol would specifically affect the mammalian reproductive system. For example, it is a precursor of the sex hormones (Janeczko and Skoczowski, 2005;Kaur et al., 2011;Wang et al., 2011; Janeczko, 2012; Tarkowská, 2019), it has shown androgenic activity as it can regulate the metabolism, secretion, and growth of the prostate under specific concentrations (Hirano et al., 1994), and it also affects the ability to regenerate spermatogenic cells in mammalian testes, so that spermatogenesis could occur earlier, and sperm production is promoted (Alfiah, 2011). Stigmasterol was also found to be a key factor in increased levels of prolactin, critical for the initiation and maintenance of breast milk production in mammals (Anjaria et al., 1975). Wickler and Seibt (1976) were the first researchers that suggested leaves could be providing bats hormonal precursors, although this field (Tarkowská, 2019) remained neglected for years. These authors found that leaves of Balanites wilsoniana were the source of steroidal sapogenins for the African fruit bat, Epomophorus wahlbergi (Wickler and Seibt, 1976), although this plant species also contains stigmasterol (Sofowora and Hardman, 1973). Since steroid hormones play a vital role as androgenes, gestagenes, glucocorticoids, and as mineralocorticoids (Wickler and Seibt, 1976;Krishnarathi et al., 2014), the complete understanding of ecological and evolutionary implications of folivory in bats will be based on further field and laboratory studies.

Several functions of stigmasterol reported in the available literature support the idea of its potentially important role in mammalian reproductive systems (Tarkowská, 2019). Thus, we would suggest testing stigmasterol as a potential regulator of reproduction in the Neotropical frugivorous bat A. amplus, based on the highest consumption of leaves of A. desmanthum (Ruiz-Ramoni et al., 2011;Duque-Marquez et al., 2019) before the exhibition of unequivocal signs of reproductive activity in both sexes, such as maximum testes size and advanced pregnancy (Ruiz-Ramoni et al., 2017). The identification of stigmasterol in this study provides a target chemical compound that could function as potential regulator of individuals’ reproductive condition, which allows its experimental testing in future follow up studies.

ISOLATION AND CHARACTERIZATION OF STIGMASTEROL FROM HIGHLY CONSUMED LEAVES BY THE LARGE FRUIT-EATING BAT, Artibeus amplus (Chiroptera: Phyllostomidae)

Alfiah S (2011) Effect Of Tapak Liman (Elephantophus scaber L) Extract on Quantity and Quality of Spermatogenic Cells in Subfertile Male Mice (Mus musculus). Thesis. Universitas Airlangga. Surabaya, Indonesia.

Anjaria JV, Varia MR, Janakiraman K, Gulati OD (1975) Studies on Leptadenia reticulata: lactogenic effect on rats. Indian J. Exp. Biol. 13: 448-449.

Ataroff M, García-Nuñez C (2013) Selvas y bosques nublados de Venezuela. In Medina E, Hubber O, Nassar J (Eds) Recorriendo el Paisaje Vegetal de Venezuela. Homenaje a Volkmar Vareschi. Instituto Venezolano de Investigaciones Científicas. Caracas, Venezuela. pp. 125-155.

Coll JC, Bowden BF (1986) The application of vacuum liquid chromatography to the separation of terpene mixtures. J. Nat. Prod. 49: 934-936.

Duque-Márquez A, Ruiz-Ramoni D, Ramoni-Perazzi P, Muñoz-Romo M (2019) Bat folivory in numbers: how many, how much, and how long? Acta Chiropterol. 21: 183-191.

Endress ME, Meve U, Middleton DJ, Liede-Schumann S (2018) Apocynaceae-6 Aspidosperma Mart. & Zucc. In Kubitzki K (Ed.) The Families and Genera of Vascular Plants XV. Kadereit JW, Bittrich V (Eds.) Flowering Plants-Eudicots Apiales, Gentianales (Except Rubiaceae). Springer. Cham, Switzerland. pp. 256-257.

Greenhall AM (1957) Food preference of Trinidad fruit bats. J. Mammal. 38: 409-410.

Hirano T, Homma M, Oka K (1994) Effects of stinging nettle root extracts and their steroidal components on the Na+, K+-ATPase of the benign prostatic hyperplasia. Planta Med. 60: 30-33.

Jacome RLRP, Oliveira AB, Raslan DS, Wagner H (2004) Estudo químico e perfil cromatográfico das cascas de Aspidosperma parvifolium A. DC. (“paupereira”). Quim. Nova 27: 897-900.

Janeczko A (2012) The presence and activity of progesterone in the plant kingdom. Steroids 77: 169-173.

Janeczko A, Skoczowski A (2005) Mammalian sex hormones in plants. Folia Histochem. Cytobiol. 43: 71-79.

Kaur N, Chaudhary J, Jain A, Kishore L (2011) Stigmasterol: A comprehensive review. Int. J. Pharm. Sci. Res. 2: 2259-2265.

Krishnarathi A, Suthakar IS, Kannan M (2014) Phytochemical analysis of selected plant leaves consumed by the folivory bats in Tirunelveli. J. Environ. Appl. Biores. 31: 14-18.

Kunz TH, Diaz CA (1995) Folivory in fruit-eating bats, with new evidence from Artibeus jamaicensis (Chiroptera: Phyllostomidae). Biotropica27: 106-120.

Kunz TH, Ingalls KA (1994) Folivory in bats: an adaptation derived from frugivory. Funct. Ecol. 8: 665-668.

Pagare S, Bhatia M, Tripathi N, Pagare S, Bansal YK (2015) Secondary metabolites of plants and their role: overview. Curr. Trends Biotechnol. Pharm. 9: 293-304.

Pereira MM, Souza-Júnior SN, Alcântara AFC, Piló-Veloso D, Alves RB, Machado PO, Azevedo AO, Moreira FH, Castro MSA, Raslan DS (2006) Constituintes químicos e estudo biológico de Aspidosperma nitidum (Apocynaceae). Rev. Bras. Plantas Med. 8: 1-8.

Ruiz-Ramoni D, Muñoz-Romo M, Ramoni-Perazzi P, Aranguren Y, Fermin Y (2011) Folivory in the giant fruit-eating bat Artibeusamplus (Phyllostomidae): a non-seasonal phenomenon. Acta Chiropterol. 13: 195-199.

Ruiz-Ramoni D, Ramoni-Perazzi P, Muñoz-Romo M (2017) Reproductive pattern of the giant fruit-eating bat, Artibeus amplus (Phyllostomidae) in the Venezuelan Andes. Rev. Biol. Trop. 65: 335-344.

Sofowora EA, Hardman R (1973) Steroids, phthalyl esters and hydrocarbons from Balanites wilsoniana stem bark. Phytochemistry 12: 403-406.

Tarkowská D (2019) Plants are capable of synthesizing animal steroid hormones. Molecules 24: 2585.

Van der Pijl L (1957) The dispersal of plants by bats (Chiropterochory). Acta Bot. Neerl. 6: 291-315.

Wang FQ, Yao K, Wei DZ (2011) From soybean phytosterols to steroid hormones. In El-Shemy H (Ed.) Soybean and Health. InTech. Rijeka, Croatia. pp. 231-252.

Wickler W, Seibt U (1976) Field studies on the African fruit bat Epomophoruswahlbergi (Sundevall), with special reference to male calling. Z. Tierpsychol. 40: 345-376.

Woodson Jr RE (1951) Studies in the Apocynaceae VIII. An interim revision of the genus Aspidosperma Mart. & Zucc. Ann. Missouri Bot. Gard. 38: 119-204.

Notas de autor

munozromo@gmail.com