Determining the Dietary Preferences of Wild Asian Elephants (Elephas maximus) in Taman Negara National Park, Malaysia Based on Sex and Age using trnL DNA Metabarcoding Analysis

© 2022 Academia Sinica, Taiwan Open Access Determining the Dietary Preferences of Wild Asian Elephants (Elephas maximus) in Taman Negara National Park, Malaysia Based on Sex and Age using trnL DNA Metabarcoding Analysis Nurfatiha Akmal Fawwazah Abdullah-Fauzi1, Kayal Vizi Karuppannan2, Nor Hafisa Syafina Mohd-Radzi1, Millawati Gani1,2 , Abd Rahman Mohd-Ridwan1,3 , Nursyuhada Othman4,5 , Hidayah Haris4,5 , Nur Hartini Sariyati4,5 , Nor Rahman Aifat1,6 , Muhammad Abu Bakar AbdulLatiff4,5 , Mohd Firdaus Ariff Abdul-Razak2, and Badrul Munir Md-Zain1,* 1Department of Biological Sciences and Biotechnology, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia. *Correspondence: E-mail: [email protected] (Md-Zain). E-mail: [email protected] (Abdullah-Fauzi); E-mail: [email protected] (Mohd-Radzi) 2Department of Wildlife and National Parks (PERHILITAN), KM 10 Jalan Cheras, 56100 Kuala Lumpur, Malaysia. E-mail: [email protected].my (Karuppannan); [email protected].my (Gani); [email protected].my (Abdul-Razak) 3Centre for Pre-University Studies, Universiti Malaysia Sarawak, 94300 Kota Samarahan, Sarawak, Malaysia. E-mail: [email protected] (Mohd-Ridwan) 4Faculty of Applied Sciences and Technology, Universiti Tun Hussein Onn Malaysia, 84600 Johor, Malaysia. E-mail: [email protected] (Othman); [email protected] (Haris); [email protected] (Sariyati); [email protected] (Abdul-Latiff) 5Oasis Integrated Group (OIG), Institute for Integrated Engineering (I2E), Universiti Tun Hussein Onn Malaysia, 86400 Parit Raja, Johor, Malaysia 6Faculty of Tropical Forestry, Universiti Malaysia Sabah (UMS), Jalan UMS, 88400 Kota Kinabalu, Sabah, Malaysia. E-mail: [email protected] (Aifat) Received 10 October 2021 / Accepted 11 August 2022 / Published 29 November 2022 Communicated by Teng-Chiu Lin The world’s largest terrestrial mammal, Asian elephants, are known to have enormous feeding needs. Several factors such as season, sex, age, and daily activities influence the amount of food required by an individual. Generally, captive elephants have a limited choice of food on a daily basis compared with that of elephants in the wild. Elephants in captivity are fed according to a prepared feeding schedule, whereas wild elephants are free to choose the type of plants that they consume in their natural habitat. In the past, ecological observations have been widely used to determine the diet of wild elephants. However, the molecular approach has never been carried out. In the present study, we aimed to; 1) identify the plant diet of wild Asian elephants in Taman Negara National Park (TNNP) according to their sex and age using high-throughput DNA metabarcoding; and 2) determine the dietary formulation of captive elephants based on the generated plant metabarcoding database. DNA was extracted from 24 individual fecal samples collected using noninvasive sampling techniques from TNNP and the National Elephant Conservation Centre (NECC) Kuala Gandah. Seven pooled samples from male adult, female adult, male subadult, female subadult, male juvenile, female juvenile, and captive elephants were amplified and sequenced targeting the trnL region (50–150 base pairs). The CLC Genomic Workbench and PAST 4.02 software were used for data analysis. In total, 24 orders, 41 families, 233 genera, and 306 species of plants were successfully detected in the diet of the Asian elephants. The most abundant plant genera consumed were Sporobolus (21.88%), Musa (21.48%), and Ficus (10.80%). Plant variation was lower in samples from male elephants than in those from female elephants. The plant species identified were correlated with the nutrient benefits required by elephants. Adults and subadults consumed more plant species than were Citation: Abdullah-Fauzi NAF, Karuppannan KV, Mohd-Radzi NHS, Gani M, Mohd-Ridwan AR, Othman N, Haris H, Sariyati NH, Aifat NR, AbdulLatiff MAB, Abdul-Razak MFA, and Md-Zain BM. 2022. Determining the dietary preferences of wild asian elephants (Elephas maximus) in Taman Negara National Park, Malaysia based on sex and age using trnL DNA metabarcoding analysis. Zool Stud 61:60. doi:10.6620/ZS.2022.61-60. Zoological Studies 61:60 (2022) doi:10.6620/ZS.2022.61-60 1 © 2022 Academia Sinica, Taiwan BACKGROUND Family Elephantidae consists of two remaining genera, which are the African elephant (Loxodonta) and the Asian elephant (Elephas). The population of Asian elephants is distributed across 13 countries spanning South and Southeast Asia (IUCN/SSC Asian Elephant Specialist Group 2017). This species has been categorized as endangered, as listed in the International Union for Conservation of Nature (IUCN) Red List (IUCN 2020). According to Saaban et al. (2011), the population of Asian elephants in Peninsular Malaysia was estimated at 1,223–1,677 individuals and can only be found in Pahang, Terengganu, Kelantan, Kedah, Perak, and Johor. Taman Negara National Park (TNNP) has an extensive area of primary rainforest and is home to the world’s largest population of wild Asian elephants (Karuppannan et al. 2020; Saaban et al. 2020). With a total area of 4,343 km2, TNNP has also become one of the main wild elephant translocation sites in Peninsular Malaysia (Saaban et al. 2011; Jambari et al. 2019). Using the molecular sexing method, Karuppannan et al. (2020) identified 86 males and 131 females in TNNP. Of these individuals, 15, 78, and 124 were juveniles, subadults, and adults, respectively (Karuppannan et al. 2020). According to Kumar et al. (2019), 14,000– 16,000 Asian elephants are currently in captivity. In Peninsular Malaysia, captive elephants are distributed in two elephant conservation centers and seven zoos: the National Elephant Conservation Centre (NECC) Kuala Gandah in Pahang, Elephant Conservation Centre Sungai Deka in Terengganu, A’Famosa Safari Wonderland, Kenyir Elephant Conservation Village, Taman Rekreasi & Zoo Kemaman, Zoo Melaka, Zoo Negara, Zoo Taiping & Night Safari, and Zoo Negeri Johor. In each of these locations, the number of captive elephants ranges from 7–40 individuals. Due to the alarmingly low population numbers of Asian elephants, they were upgraded to a fully protected species in Peninsular Malaysia under the Wildlife Conservation Act of 2010 (Act 716). The greatest threat to the population of wild Asian elephants comes from human activities and agricultural growth, which have led to human-elephant conflict (Choudhury et al. 2008; Karuppannan et al. 2019a). In addition, poaching and illegal trade of wild elephants is a major concern in many countries (Douglas-Hamilton 2009; Hedges 2012). Sukumar et al. (1998) stated that male elephants are more likely to be hunted for their valuable tusks. However, elephants are also hunted for their skin, which is used as an ingredient in Chinese medicine (Apinya 2018). For elephants in captivity, common concerns include nutritional intake and weight problems (Vanitha et al. 2008). Furthermore, economic issues can be a significant constraint when attempting to implement wildlife conservation plans (Bandara and Tisdell 2003). Given these issues, the Department of Wildlife and National Parks (PERHILITAN) together with stakeholders has established the National Elephant Conservation Action Plan (NECAP) for Asian elephants (DWNP 2013). Elephas maximus are considered mega-herbivores with tremendous dietary requirements. Sukumar (2006) stated that elephants are able to consume 10% of their body weight and spend 12 to 18 hours a day just eating. In previous research, conventional approaches such as field observations (Sukumar 1990; Godagama et al. 1999; Borah and Deka 2008; Joshi and Singh 2008), in-depth interviews (Godagama et al. 1999), and microhistological methods (Yamamoto-Ebina et al. 2016; Koirala et al. 2019) have been used to investigate the diet of the Asian elephant (Magintan et al. 2016). However, the practicality of these techniques has been questioned due to the limited results they produce (Suba et al. 2017). Asian elephants are mixed feeders that graze and forage for food in the wild (Chen et al. 2006). Compared with wild elephants, captive elephants assimilate fewer nutrients, as their food choice is restricted. Generally, 68% of plants ingested by wild Asian elephants were reportedly made up from seven types of families, namely Fabaceae, Poaceae, Malvaceae, Sterculiaceae, Tiliaceae, Palmae and Cyperaceae (Sukumar 1990; Harich et al. 2016; Koirala 2017). In tropical rainforests, most Asian elephant diets are dominated by woody and fruiting plants such as grasses, shrubs, herbs and even roots (Sukumar 2006; Ullrey et al. 1997). Meanwhile in captivity, elephants are given food in the form of pellets, fruits, and vegetables, as well as supplements such as vitamins and minerals (Benz 2005). However, to date, no studies have been conducted on the plant diet of Asian elephants according to their sex and age specifically. consumed by juvenile elephants. However, there was no significant difference between ages and sexes. The findings of this study can be used as guidance by the Department of Wildlife and National Parks for the management of captive elephants, especially in NECC Kuala Gandah. Key words: Fecal, Captivity, Peninsular Malaysia, Next-generation sequencing. page 2 of 12 Zoological Studies 61:60 (2022) © 2022 Academia Sinica, Taiwan Generally, previous studies involving herbivorous diets by sex and age have been very minimal and poorly conducted. Due to different physiological states such as pregnancy and lactation processes in female animals, female elephants are found to require higher levels of nutrients than males (Stokke 1999). Apart from that, size of an animal also plays a part in the selection of dietary preferences. This was proven in a study on ruminant and non-ruminant herbivores, which indicated that fiber intake of an animal correlated with their body size (Demment and Van Soest 1985). Differences in dietary preferences among age groups can be seen in red kangaroos, Osphranter rufus, where the results stated that juveniles require more sustenance in their diet compared to other age classes. The main diet of this herbivore is also said to be dominated by grasses (Dawson et al. 2021). The DNA metabarcoding method is a powerful tool for identifying a wide variety of plants that have been consumed by animals (Soininen 2012); indeed, it is known to be more accurate than the observation method. It is often used to analyze animal diet from fecal samples because it avoids the handling of animals and can provide reliable results (Elliza et al. 2015; Aifat et al. 2016a; Hawlitschek et al. 2018; Md-Zain et al. 2018; Karuppannan et al. 2019b). The trnL (UAA)-P6 marker from chloroplast DNA is widely used as a marker in dietary analysis (Taberlet et al. 2007; Pompanon et al. 2012; Reese et al. 2019); it has a short length, conserved primer sites, and interspecific variation (Taberlet et al. 2007; Pompanon et al. 2012; Reese et al. 2019). Despite the reliability of the DNA metabarcoding approach, it has yet to be tested on Asian elephants. In Peninsular Malaysia, the dietary intake of elephants has yet to be investigated using DNA metabarcoding analysis. In addition, knowledge of Asian elephant diet according to their sex and age is poorly understood. Furthermore, captive elephants in National Elephant Conservation Centre (NECC) Kuala Gandah require management of a balanced diet that is appropriate to their sex and age, similar to in their natural habitat. Thus, the present study aims to identify the plant diet preferences of wild Asian elephants in TNNP according to their sex and age using highthroughput DNA metabarcoding. In addition, it was predicted the diets consumed by wild elephants through DNA metabarcoding analyses differ in terms of plant diversity. Later, the generated plant database would help to determine the optimal diet formulation for elephants in captivity. Such a diet would promote the growth of captive elephants because it would allow the elephants to receive the complete range of nutrients required from an early age. Overall, the findings of this study help distinguish the diet of wild Asian elephants based on sex and age. Moreover, the study is conducted in line with the NECAP strategy according to its long-term goals, i.e., goal number seven. This goal is to ensure that the captive elephant population of Malaysia is wellmanaged and contributes to wild elephant conservation (DWNP 2013). MATERIALS AND METHODS Sample Collection In total, 24 fecal samples collected in a noninvasive manner were used in this study: 18 samples from TNNP and six samples from the NECC Kuala Gandah (Table 1). Samples from the wild (TNNP) were obtained from the Wildlife Genetics Research Bank, PERHILITAN which were utilized from previous collected fecal samples. In contrast, the samples from captivity (NECC) were fresh and act as a positive control in this study, where the diet of captive elephants was already known. The collected fecal samples were preserved in absolute ethanol and maintained at room temperature. Sex and Age Determination of Wild Elephants In this study, male and female elephants were distinguished using a non-invasive molecular sexing technique, targeting two Y-specific and one X-specific primers. This approach has been done previously by Karuppannan et al. (2019c) using fecal samples from TNNP. Meanwhile, the age of wild elephants in TNNP was determined using their mean boli circumference (Reilly 2002; Tyson et al. 2002; Karuppannan et al. 2020). The measurements are listed in table 2. DNA Extraction, Quantification, and Amplification DNA was extracted from ~200-mg dung samples using a QIAGEN QIAamp Fast DNA Stool Mini Kit. Extracted DNA was then quantified using a NanoDrop spectrophotometer. All 24 samples were pooled to form 7 samples as follows: male adult, female adult, male subadult, female subadult, male juvenile, female juvenile, and captive elephant samples. Polymerase chain reaction (PCR) was performed targeting the trnL region of the chloroplast gene (Taberlet et al. 2007). The PCR mixture contained 10 μL of Promega GoTaq Green Master Mix, 1 μL of each forward and reverse primer (10 µM), 6 μL of nuclease-free water, and 2 μL of DNA (final volume: 20 μL). The extracted samples were amplified using page 3 of 12Zoological Studies 61:60 (2022) © 2022 Academia Sinica, Taiwan a Bio-Rad T100 Thermal Cycler under the following conditions: 95°C for 5 min, followed by 35 cycles of 95°C for 1 min, 56°C for 30 s, and 72°C for 1 min, with a final extension step at 72°C for 5 min. The PCR products were visualized using gel electrophoresis with 1% agarose gel in 1x TAE buffer to measure amplicon size. The laboratory work conducted in this study was performed at the National Wildlife Forensic Laboratory, PERHILITAN, Peninsular Malaysia. Library Preparation and Sequencing All PCR products of seven pooled samples with a volume of 10 μL each were sent to GeneSeq Sdn. Bhd for library preparation and sequencing. They were purified using SPRI Beads (Oberacker et al. 2019), followed by index PCR to incorporate Illumina dual index barcodes. The barcoded libraries were then pooled and gel-purified using a WizPrep Gel/PCR Purification Mini Kit (WizBio, Korea) according to the manufacturer’s instructions. The pooled libraries were quantified using a Denovix dsDNA High Sensitivity Assay and an appropriate amount of the libraries was loaded onto an iSeq100 (Illumina, San Diego) for 2 × 150 paired-end sequencing. Data Analysis Quality filtering and demultiplexing of sequences was conducted using CLC Genomic Workbench software (CLC) (Qiagen, USA). A preliminary assessment of quality scores across the Illumina data was performed using FASTQ files. The operational taxonomic units (OTUs) were clustered at 97% similarity and represented by a single sequence, which was aligned using the MUSCLE tool in CLC. The phylogenetic classification of the OTUs was conducted against a trnL database downloaded from GenBank. The alpha diversity indices (Shannon and Chao-1 index estimators) assessed the species richness in the elephants’ diet and were generated using PAST 4.02 software (Hammer et al. 2001). The beta diversity was used to describe the dissimilarities in dietary diversity among individual elephants and was measured based on a UniFrac distance of 0.5. A Venn diagram was produced to determine the shared and unique OTUs among the elephant samples at 97% similarity. To assess dietary diversity relationships among elephant samples, a heatmap was constructed using 1000 bootstrap replications following the Bray-Curtis distance. The correlation in diet (plant genera) among individual elephants from different Table 1. List of samples used in this study No Sample ID Pooled samples Locality 1 EM616 Male adult (MA) Taman Negara National Park 2 EM705 Male adult (MA) Taman Negara National Park 3 EM594 Male adult (MA) Taman Negara National Park 4 EM707 Female adult (FA) Taman Negara National Park 5 EM618 Female adult (FA) Taman Negara National Park 6 EM640 Female adult (FA) Taman Negara National Park 7 EM628 Male subadult (MSA) Taman Negara National Park 8 EM729 Male subadult (MSA) Taman Negara National Park 9 EM561 Male subadult (MSA) Taman Negara National Park 10 EM566 Female subadult (FSA) Taman Negara National Park 11 EM696 Female subadult (FSA) Taman Negara National Park 12 EM677 Female subadult (FSA) Taman Negara National Park 13 EM613 Male juvenile (MJ) Taman Negara National Park 14 EM739 Male juvenile (MJ) Taman Negara National Park 15 EM603 Male juvenile (MJ) Taman Negara National Park 16 EM758 Female juvenile (FJ) Taman Negara National Park 17 EM573 Female juvenile (FJ) Taman Negara National Park 18 EM735 Female juvenile (FJ) Taman Negara National Park 19 EM1540 Captivity (C) NECC Kuala Gandah 20 EM1534 Captivity (C) NECC Kuala Gandah 21 EM1524 Captivity (C) NECC Kuala Gandah 22 EM1525 Captivity (C) NECC Kuala Gandah 23 EM1530 Captivity (C) NECC Kuala Gandah 24 EM1535 Captivity (C) NECC Kuala Gandah page 4 of 12Zoological Studies 61:60 (2022) © 2022 Academia Sinica, Taiwan populations was assessed using Pearson correlation coefficients. Correlations were visualized using cold-hot plots generated in PAST 4.02. Statistical significance was set at P < 0.05. Data analysis was conducted at the Evolutionary and Conservation Genetic Laboratory of Department of Technology and Natural Resources, Universiti Tun Hussein Onn Malaysia. RESULTS NGS Data Analysis The concentration of extracted DNA was 1.4–10.5 ng/µL and the amplification products were 100–150-bp long (gained from trnL fragments). High-throughput DNA metabarcoding was used to successfully determine the diet of wild Asian elephants in TNNP according to sex and age; 371,556 plant sequences were obtained, ranging from 135,035 to 236,521 and from 63,361 to 202,372, respectively. After clustering at the 97% cut-off value, female elephants had more OTUs (757) when compared with those of male elephants (Table 3). Among the age categories, subadult has the most OTUs (594), followed by adult (412), and juvenile (336) (Table 4). Plant Species Identification The plants consumed by Elephas maximus were successfully classified by taxonomic level into 24 orders, 41 families, 233 genera, and 306 species (Table 5). Figures 1–3 show the relative abundance of dietary plants at the genus level according to analysis of wild Asian elephant samples from TNNP with categorization by sex and age. Overall, Sporobolus (21.88%), Musa (21.48%), Ficus (10.80%), Laccosperma (8.22%), Coccothrinax (6.15%), Athrixia (2.11%), Colpothrinax (1.73%), and Mauritia (1.26%) were the most dominant plant genera found in the diet of Asian elephants (Fig. 1). Considering sex and age, female and subadult Asian elephants consumed more plants than those in the other categories. Alpha Diversity Indices, Heatmap, Rarefaction Curve, and Venn Diagram The alpha diversity (Shannon and Chao-1 indices) indicated that the diets of Asian elephants varied greatly depending on the analyzed parameters (Table 6 and 7). According to sex, males had a higher Shannon index value (H = 2.844) than females (H = 2.625). In contrast, females had a substantially greater Chao-1 value (785.5) than males (588.9). Among age categories, adults had the highest Shannon index value (H = 2.607), followed by subadults (H = 2.326), and juveniles (H = 2.225). However, the pattern of Chao-1 values was different: subadults had the highest value (624.7), followed by adults (459.4) and juveniles (396.6). Figures 4 and 5 show the 20 most abundant plant genera present in male, female, adult, subadult, and juvenile Asian elephant samples. Darker colors indicate the more dominant genera. Table 2. Measurements of mean boli circumference (mbc) Measurements Category mbc ≤ 30 cm juvenile 30 cm < mbc ≤ 42 cm subadult mbc > 42 cm adult Table 3. Number of sequences, OTUs, and unique OTUs of plants consumed by male and female Asian elephants Samples Sequences OTUs Unique OTUs Male 135,035 504 251 Female 236,521 757 504 Total 371,556 1,261 Table 4. Number of sequences, OTUs, and unique OTUs of plants consumed by adult, subadult, and juvenile Asian elephants Samples Sequences OTUs Unique OTUs Adult 105,823 412 222 Subadult 202,372 594 384 Juvenile 63,361 336 142 Total 371,556 1,342 Table 5. Total number of plants identified at different taxonomic levels according to trnL gene analysis Taxonomic level Total number Order 24 Family 41 Genus 233 Species 306 page 5 of 12Zoological Studies 61:60 (2022) © 2022 Academia Sinica, Taiwan Fig. 1. Distribution (%) of plants consumed by wild Asian elephants in Taman Negara National Park (TNNP) at the genus level (> 1% abundance). Fig. 2. Distribution (%) of plants consumed by male and female Asian elephants in Taman Negara National Park (TNNP) at the genus level (20 most abundant genera). Fig. 3. Distribution (%) of plants consumed by adult, subadult, and juvenile Asian elephants in Taman Negara National Park (TNNP) at the genus level (20 most abundant genera). page 6 of 12Zoological Studies 61:60 (2022) © 2022 Academia Sinica, Taiwan Table 6. Alpha diversity indices: Shannon and Chao-1 values for male and female Asian elephants Samples Shannon_H Chao-1 Male 2.844 588.9 Female 2.625 785.5 Table 7. Alpha diversity indices: Shannon and Chao-1 values for adult, subadult, and juvenile Asian elephants Samples Shannon_H Chao-1 Adult 2.607 459.4 Subadult 2.326 624.7 Juvenile 2.225 396.6 Fig. 5. Heatmap with a dendogram showing dietary plant abundance at the genus level for adult, subadult, and juvenile Asian elephants. Gradient heatmap shows the 20 most abundant genera. Fig. 4. Heatmap with a dendogram showing dietary plant abundance at the genus level for male and female Asian elephants. Gradient heatmap shows the 20 most abundant genera. page 7 of 12Zoological Studies 61:60 (2022) © 2022 Academia Sinica, Taiwan Fig. 7. Rarefaction curves for adult, subadult, and juvenile Asian elephant samples. Fig. 8. Venn diagram showing the number of shared operational taxonomic units (OTUs) between male and female Asian elephants at 97% similarity. Fig. 9. Venn diagram showing the number of shared operational taxonomic units (OTUs) among adult, subadult, and juvenile Asian elephants at 97% similarity. Fig. 6. Rarefaction curves for male and female Asian elephant samples. The rarefaction curves in figures 6 and 7 show an increasing pattern for all samples used in this study, which indicates that plant richness was not adequately sequenced to facilitate identification of plant species. Indeed, the curves indicate that more accurate results would be possible if additional sampling were conducted. The Venn diagrams in figures 8 and 9 show that 253 unique OTUs were shared between male and female Asian elephants (251 and 504 OTUs for male and female, respectively), whereas adult, subadult, and juvenile shared 74 OTUs. DISCUSSION This study is the first to identify the diet of wild Asian elephants in TNNP according to sex and age, with gene analysis focused on the trnL region. The feces of wild elephants used in this study were collected from 2016 to 2017, which may have affected the sample quality and sequences generated from the samples. This has been demonstrated in previous studies where fresh fecal samples had high DNA quality and a higher percentage of sequencing reads (Hedges 2012; SyedShabthar et al. 2013; Aifat et al. 2016b; Hawlitschek et al. 2018; Karuppannan et al. 2019b). Furthermore, the number of samples available for each parameter in this study was restricted. For individual samples with low DNA concentration, samples representing the same parameters and having low heterogeneity can be pooled, which also reduces costs and labor (Ray et al. 2019). Of the plant species identified in the analysis, some species, such as Coccothrinax litoralis and Colpothrinax wrightii, did not exist in Peninsular Malaysia—these were palm species that were endemic to Cuba (Leiva and Verdecia 2007; Henderson et al. 2019). This is because the trnL database used in the current study page 8 of 12Zoological Studies 61:60 (2022) © 2022 Academia Sinica, Taiwan mainly includes plants from tropical rainforests in Africa and Madagascar. Thus, these detected plant species could be close relatives of the actual plant species ingested by Asian elephants in TNNP (Osman et al. 2020). TNNP is a tropical evergreen rainforest with diverse flora and fauna. It is home to more than 3,000 plant species and 150 mammal species (UNESCO 2021). Jambari et al. (2019) discovered that wild Asian elephants were dominant in the lowland forest compared with the highland forest of TNNP, which possesses the largest lowland tropical rainforests in Peninsular Malaysia. This finding relates to the number of plant species detected in the current study. By targeting the chloroplast trnL intron region, it was possible to identify 24 orders, 41 families, 233 genera, and 306 species of plants consumed by Asian elephants. The plant genera most abundantly consumed by wild Asian elephants in TNNP were Sporobolus (21.88%), Musa (21.48%), Ficus (10.80%), Laccosperma (8.22%), Coccothrinax (6.15%), Athrixia (2.11%), Colpothrinax (1.73%), and Mauritia (1.26%) (Fig. 1). While at family level, the diet of wild Asian elephants is dominated by the family Poaceae, Arecaceae, Moraceae, and Musaceae. Sporobolus diandrus and Sporobolus indicus were previously found in 10 states in Peninsular Malaysia including Pahang, Kelantan, and Terengganu, which are included in the TNNP area (Izzati et al. 2009). Furthermore, Sporobolus is a genus of grasses that is thought to be a primary food source for elephants (Dhairykar and Singh 2020). Overall, the diet selection of wild Asian elephants in TNNP is general. This study found that female elephants have more plant homogeneity in their diet compared with that in the diet of male elephants. This is consistent with a study conducted on African elephants by Stokke (1999), who concluded that female elephants are more particular in foraging when compared with male elephants. These findings could be due to sexual dimorphism, i.e., the body size hypothesis, which states that the larger an animal’s size, the less selective it is in terms of foraging behavior (Stokke 1999; Woolley et al. 2011). In the present study, female Asian elephants in TNNP were found to have higher plant species richness in their diet relative to that in male diets. This may be related to the increased nutritional demands during pregnancy and lactation, i.e., additional nutrients from a variety of plant species could be essential to these processes (Dierenfeld et al. 2020). Nevertheless, the diets of male Asian elephants were higher in plant diversity (H = 2.844) than those of females. This may be due to their solitary existence, which involves foraging over a relatively larger area than that used by female elephants (Eisenberg 1980). Generally, both sexes are likely to share similar dietary preferences, but they differ in terms of the quantity of food consumed. As reported by Koirala et al. (2019), the nutrient composition in the diet of elephants varies depending on their sex and age. However, the plant diet of wild elephants had not previously been analyzed according to age. In the current study, the number of OTUs detected in the samples showed the differences in diet among the age groups. Subadult Asian elephants were found to ingested more plant OTUs than adult and juvenile elephants. This is consistent with the requirement for growing elephants to consume relatively more food given the former’s increased need for protein and other nutrients (Dhairykar and Singh 2020). Among the three age classes investigated here, adult elephants appeared to consume the most diverse range of plants, yet the plant richness is lower than in the subadults. We can conclude that, as Asian elephants grow older and begin to mature, they choose fewer types of plants, and their diet becomes more specific. Nevertheless, most of the plant genera were shared among three age classes, although Sporobolus and Oplismenus were only present in subadults and juveniles. Even though elephants can be unspecialized feeders, they can also be quite particular in terms of selecting their preferred dietary plants (Swit 2016). According to the captive elephants’ caretaker, known as the mahout, captive elephants in NECC were fed with horse pellets in the morning. This is because their digestive system somewhat resembles that of a horse (Hatt and Clauss 2006). The elephants also are given a variety of plant species in the evening including Musa sp. (banana), Cenchrus purpureus (napier grass), Citrullus lanatus (watermelon), Carica papaya (papaya), and Saccharum officinarum (sugarcane). However, there was flexibility in the plant species provided that was entirely dependent on the availability of the stock. Similar to the plant species provided by the mahout, DNA metabarcoding analysis in this study identified an abundance of Cenchrus purpureus, Musa sp., and Saccharum officinarum in the diet of captive elephants in NECC. Plant species that are not given such as Sporobolus diandrus (tussock grass), Ficus superba (sea fig) and Panicum auritum (millet) also generated by the DNA metabarcoding, may have been eaten during walking hours around the NECC Kuala Gandah. CONCLUSIONS The final results of this study demonstrate that DNA metabarcoding can detect the plants consumed by wild Asian elephants up to the species level. Despite page 9 of 12Zoological Studies 61:60 (2022)