Phylogenetic and genetic variation analysis of lesser short-nosed fruit bat Cynopterus brachyotis (Müller 1838) on Java island, Indonesia, inferred from mitochondrial D-loop

Background Cynopterus brachyotis (Müller 1838) is a generalist and widespread fruit bat species which inhabits different types of habitats in Southeast Asia. This species plays an essential role as a seed disperser and pollinator. Morphological study and phylogenetic analysis using mtDNA markers (cyt-b and D-loop) revealed that this species had two different forms in peninsular Malaysia and Borneo and six lineages in Southeast Asia that lead to new species formation. In addition, this species is also reported to have high genetic diversity in Malaysia and Thailand based on the D-loop sequence. However, a phylogenetic and genetic variation study of C. brachyotis in Indonesia has not been conducted yet. These two studies are important as additional information for taxonomic and population genetic studies of this species. Thus, we performed the phylogenetic and genetic diversity analysis of the C. brachyotis population collected from seven habitats on Java island, including open-fragmented habitats (urban, coffee and rubber plantations, pine forest, secondary forest, mangrove forest) and closed habitats (natural forest) using the mtDNA D-loop marker. Results The phylogenetic tree using the Bayesian inference (BI) and genetic distance using the Kimura-2 parameter (K-2P) demonstrated that 33 individuals of C. brachyotis from seven habitats on Java island overlapped between habitats and could not be distinguished according to their habitats and lineage. Intrapopulation and intraspecies analysis revealed high haplotype diversity of this species on Java island (Hd = 0.933–1.000). The haplotype network was split into two haplogroups, showing haplotype sharing between habitats. These phylogenetic and genetic variations analysis of C. brachyotis bats on Java island indicated that this species is widespread and adapt to different habitats. Conclusions This study of C. brachyotis on Java island collected from seven different habitats has overlapped and genetically close and has high genetic variation. Our results provide the first reported study of C. brachyotis on Java island and provide data to understand the phylogenetic and genetic diversity of this species in Indonesia.


Background
Lesser short-nosed fruit bat Cynopterus brachyotis (Müller 1838) belongs to the family Pteropodidae and is widely distributed in Southeast Asia, including Indonesia. It has an important role in the ecosystem as seed dispersal and pollinator [1][2][3][4]. C. brachyotis is generalis and eurytopic species where it can be found in many habitats (most frequently in the disturbed and fragmented area), such as the urban area, secondary forest, dipterocarp forest, gardens, mangroves, and strand vegetation [5,6]. This species in Southeast Asia has been reported to have morphological variations [7] and karyotype variation [8]. Two different forms of C. brachyotis (large and small size or then called Sunda and Forest) were revealed based on morphological study and inhabited two contrasting habitats in Peninsular Malaysia and Borneo. Moreover, the larger form (Sunda) was found to inhabit open areas (urban and plantation), whereas the smaller form (Forest) was confined to natural forests [5].
Phylogenetic analysis using mitochondrial DNA markers including cyt-b and the control region D-loop demonstrated that C. brachyotis Sunda and Forest have different clades and separated population. Moreover, this analysis also revealed that C. brachyotis is a species complex with six distinct lineages: Sunda, Forest, India, Philippines, Myanmar, and Sulawesi [9]. According to the morphological and phylogenetic analysis in Peninsular Malaysia and Borneo, the possibility of a new species formation in C. brachyotis species may occur [5,9]. However, the taxonomic recommendation for this species so far was to treat it as a single species until additional data are available [9].
Furthermore, according to that phylogenetic analysis, C. brachyotis Sunda and Forest lineage can be found both in Peninsular Malaysia and Borneo and has contrasting habitats. A C. brachyotis sample was collected from Java island (Jakarta) and is included in the Sunda lineage [9]. Unfortunately, sample data from other areas and habitats on Java island remain unknown. This species is also reported to have high genetic diversity based on mitochondrial cyt-b and D-loop markers in Borneo, Thailand, and Peninsular Malaysia [5,10].
D-loop is a hypervariable control region of mitochondrial DNA having high nucleotide variations and mutational hotspots [11]. Haplotype analysis of the D-loop region was essential for understanding the genetic diversity and species population, including bats in particular habitats [12,13]. Genetic analysis using mitochondrial D-loop can be used to estimate the impact of the anthropogenic process on an animal population. Genetic diversity can affect a species' ability to adapt to environmental changes [14]. Besides, C. brachyotis is fruit bat species closely related to habitat fragmentation [2,3]. Java island is an Indonesian island that has a very high and rapid population growth rate. This island belongs to the Greater Sunda islands alongside Peninsular Malaysia, Sumatra, and Borneo. A total area of about 128,297 km 2 with a population projection is 56.1% of the Indonesian population makes this island the most densely populated island in Indonesia [15]. Thus, the environmental changes and habitat fragmentation become urban areas, and other anthropogenic processes are increasing. However, genetic studies such as phylogenetic and genetic variation based on D-loop in C. brachyotis have not been explored in Indonesia, especially on Java island Only karyotype and hematological studies of C. brachyotis were conducted on this island [8,16,17]. Therefore, this study aims to analyze the phylogenetic and genetic variation of C. brachyotis collected from different types of habitats on Java island inferred from the mitochondrial D-loop gene sequence. The results of this study are expected to be the basis for ecological management of C. brachyotis species and genetic research on other bat populations in Indonesia.

Ethical statement
The ethical clearance for conducting this research was taken from the Research Ethics Committee of the Faculty of Veterinary Medicine, Gadjah Mada University, Indonesia (Approval No. 0108/EC-FKH/Ex./2019). Animal handling and sampling (including tissue sampling) were carried out following their guidelines and supervision. This ethical clearance also covered other analyses, such as morphological measurements and hematological analysis of this species.

Sample collection and species identification
A total of 33 individual C. brachyotis were collected from seven habitats including urban area, coffee plantation, rubber plantation, pine forest, secondary forest, natural forest, and mangrove forest in East Java, a special region of Yogyakarta, Central Java, and West Java ( Table 1, Fig. 1). Bats were trapped using mist nets (12 × 2 m) which placed 2-3 m above the ground near potential fruit trees, such as Muntingia calabura or Ficus spp. Mist nets were set up at 05.00 pm-09.00 pm. The trapped bats were handled carefully and put in a canvas bag for further identification. Bats identification was conducted based on the identification key of bats [6,18]. Tissue samples were collected from the wings membrane using a 4 mm circular Sklar Tru-Punch sterile biopsy tool. The tissue was preserved in a 1.5 ml tube with 96% ethanol and transported to the Laboratory of Genetics and Breeding Faculty of Biology Universitas Gadjah Mada Yogyakarta, Indonesia, for further analysis.   [20]. PCR products were separated on 1% agarose gel (stained with FloroSafe (Bioline)) at 50 V for 20 min and then visualized under UV light. For sequencing, the qualified amplification products were sent to First Base Sdn Bhd Malaysia through P.T. Genetika Science Jakarta.

Data analysis Sequence editing and alignment of D-loop
The ambiguous bases of chromatograms were corrected manually in GeneStudio software (GeneStudio, Inc., Georgia) to get a consensus sequence of D-loop. Alignment of D-loop sequence was performed using Opal in Mesquite v.3.51 [21] and ClustalW in MEGA X software [22].

Nucleotide composition analysis
Nucleotide composition analysis was used to determine the nucleotide differences between samples which indicated genetic variation. Differences in D-loop sequence nucleotides between habitats (intrapopulation) and all samples (intraspecies) were analyzed using the MEGA X software [22].

Genetic variation and haplotype network analysis
Genetic variation analysis among habitat types including haplotype number (h), haplotype diversity (Hd), and nucleotide diversity (π) was performed using DnaSP 6.0 software [25]. The level of genetic diversity was calculated from the haplotype diversity value based on Nei (1987) category [26]. Haplotype networks were analyzed using the Median Joining Network method in NETWORK ver 10.1 software.

Nucleotide composition of D-loop sequences
The nucleotide composition analysis showed that the percentage of nucleotides T, C, A, and G of samples between habitats (intrapopulation) was varied. The average percentage range of T, C, A, and G nucleotides in seven habitats was 26.67-27.57%, 28.44-29.87%, 29.97-31.14%, and 12.57-13.87%, respectively. This analysis also showed AT-rich in all habitats with an average range of 57.39-58.35%. Meanwhile, the GC was 41.65-42.61%. The average nucleotide percentage of all sample (intraspecies) was T = 27.29%, C = 29.17%, A = 30.16%, and G = 13.38%. The nucleotide composition analysis of intraspecies also revealed an AT-rich (57.45%) compared to GC (42.55%).

Phylogenetic and genetic distance
The alignment of the D-loop sequence of 33 samples in this study and GenBank data yielded a fragment length of 412 bp. The phylogenetic tree demonstrated the D-loop sequence of C. brachyotis from seven habitats in this study, and GenBank data formed a single clade (Fig. 2). Genetic distance analysis using the Kimura-2 parameter method showed the genetic distance value of C. brachyotis among habitat types ranged from 0.009-0.039, while the genetic distance between samples within a habitat ranged from 0.010-0.047 (Table 2). Samples in the secondary forest and natural forest had the highest genetic distances compared to genetic distances within and between other habitats of 0.047 (within habitats) and 0.034 (between habitats) and also 0.042 (within habitats) and 0.039 (between habitats), respectively (Table 2). Furthermore, the average value of the genetic distance between all samples in this study was 0.025, and if compared to GenBank data, the genetic distance of C. brachyotis in this study ranged from 0.009 to 0.179.

Genetic variation
Analysis of genetic variation showed that C. brachyotis among seven habitats (intrapopulation) had haplotypes ranging from 3 to 5 haplotypes with the number of samples of 3-6 individuals per habitat. The haplotype diversity and nucleotide diversity of seven habitats were 0.933-1.000 and 0.00980-0.01115, respectively ( Table 3). The C. brachyotis samples from the natural forest in this study had high haplotype diversity and nucleotide diversity compared to other habitats (Table 3). All samples (intraspecies) showed haplotype diversity of 0.955 and nucleotide diversity of 0.03797. The polymorphic site was 86 sites with 69 parsimony informative (Table 3).  Furthermore, the number of haplotypes was 22 haplotypes from a total sample of 33 individuals (Table 4).

Haplotype network
Haplotype network analysis demonstrated that the C. brachyotis of this study was divided into two groups of haplotypes (haplogroups). The first haplogroup consisted of H1-H22 (excluding H16 and H18) haplotypes originating from seven habitats (urban, coffee plantation, rubber plantation, pine forest, secondary forest, natural forest, and mangrove forest). The second haplogroup consisted of H16 and H18 haplotypes from the secondary forest and natural forest habitats. Haplotype sharing in the first haplogroup, namely H1, H2, H4, and H11 haplotypes, was overlapping between habitats in different sampling locations. Meanwhile, other haplotypes were unique haplotypes (Fig. 3).

Phylogenetic of Cynopterus brachyotis on Java island
The phylogenetic tree reconstruction and genetic distance analysis using D-loop sequences in this study indicated that C. brachyotis bats on Java island overlapped between habitats and were genetically close (the average of the genetic distance was 0.025). No previous studies explained the genetic distance threshold based on D-loop sequences in bats (Chiroptera). However, the genetic distance between samples in cattle (mammals) was less than 0.5 considered to have a close relationship [27]. The higher genetic distance within a habitat population than between populations in this study may indicate that genetic variation is more related to differences between samples in a habitat [13]. In addition, Miniopterus pallidus bat in Iran also showed a higher genetic distance within a population [13]. Thus, higher genetic distances within the sample in the secondary and the natural forest populations may reflect a high genetic variation among samples in these two habitats. It indicated from the phylogenetic tree that HS-4 and HP-2 samples separated from others (Fig. 2).
The results showed that C. brachyotis bats on Java island could not be grouped into their habitat type and lineage based on the D-loop sequence marker. This finding is supported by the external morphology  Further analysis using the cytochrome b (cyt-b) marker is needed to support species identification. The average percentage differences in nucleotide composition between habitats in this study also indicated nucleotide variations of intrapopulation and may reflect the genetic variation. Nucleotide composition is a modest way to characterize the genome. Furthermore, the AT-rich of C. brachyotis D-loop sequence from seven habitats on Java island may describe the rapid evolutionary adaptation of this species. The higher nucleotide pair AT of the D-loop may cause the structure of C. brachyotis mitogenome to become less stable (due to the AT pair have two hydrogen bonds) and accelerate its evolutionary adaptation [28]. Evolutionary adaptation is an adjustment of structure or behavior derived by a species or individual to increase survival ability and inherits genes related to the environment [29]. Thus, it was suspected that C. brachyotis would be able to survive and adapt to different types of habitats on Java island.

High genetic variation of Cynopterus brachyotis on Java island
According to Nei's (1987) category, C. brachyotis on Java island has high haplotype diversity (Hd = 0.8-1.00). A previous study showed C. brachyotis population from Peninsular Malaysia and southern Thailand also has high haplotype diversity [10]. C. brachyotis from the natural forest habitat in this study also have a higher haplotype diversity and nucleotide diversity than in other habitats. Those results are in line with this study. C. brachyotis Forest that inhabits closed habitat (secondary forest and natural forest) has higher haplotype diversity (0.995 ± 0.0023) than C. brachyotis Sunda that inhabit the fragmented area (0.982 ± 0.0003). In addition, the nucleotide diversity of C. brachyotis Forest is about three times higher compared to C. brachyotis Sunda [10]. High haplotype diversity could describe high genetic diversity. The genetic diversity of a population is essential for the adaptation process and long-term survival to environmental changes. Moreover, high genetic diversity in a population will degrade the risk of species extinction because they will be able to survive or adapt to environmental changes [30]. Therefore, high haplotype diversity in the C. brachyotis population that inhabits different types of habitat, especially the natural forest on Java island, may represent a high genetic diversity and is considered capable of surviving environmental changes in any habitat types.
A haplotype network analysis is used to analyze and describe the relationship between haplotypes or DNA sequences in populations (intrapopulation) or species (intraspecies) [31,32]. Analysis of the haplotype network in this study revealed that C. brachyotis on Java island overlapped between habitats. This result was congruent with the phylogenetic tree. The samples of C. brachyotis in the haplotype sharing H1, H2, and H4 originated from different types of habitats in the East Java, Yogyakarta, Central Java, and West Java regions. The distance between these locations is more than 500 km, and geographical barriers such as mountains have existed. Therefore, sample connection or gene flow between the samples should not occur and lead to unique haplotype formation [33,34]. However, our results demonstrated otherwise.
The haplotype sharing formation in this study is still unknown. However, according to previous studies, the haplogroups and haplotype sharing formations between fruit bat populations within long population area distances are related to demographic history during Pleistocene refugium. Furthermore, the formation of haplogroup and haplotype sharing in fruit bat species Penthetor lucasi between the Miri and Kuching populations in Sarawak, Malaysia (the distance between populations of 348 km), based on the D-loop indicates that these two populations became the source of the P. lucasi population in Sarawak during the Pleistocene refugium [35,36].
The Pleistocene refugium theory explained that habitat fragmentation due to declining sea levels caused the isolation and diversification of taxons during the Pleistocene era, including bats [10,37]. It also occurred in other mammals such as Rattus rattus and Mus musculus. Several haplotypes of R. rattus and M. musculus in the Indochina region (including Java) were also shared based on the D-loop and cyt-b markers. It is due to the long-distance distribution that began in the mid-Pleistocene and is related to human activities in the modern era [38,39]. These possibilities may also occur in C. brachyotis on Java island, but it is necessary to be studied further.

Conservation implications
Java island is one of the islands in Indonesia that have the highest human population and massive habitat fragmentation [40]. The forest area on Java island has a crucial role in the ecosystems and as a natural habitat for wild-life populations, including bats. Unfortunately, the remaining forest area on Java island is less than 9% [41,42]. Habitat fragmentation and deforestation into urban areas and plantations can reduce or lose genetic variation in a wild-life population [43]. The phylogenetic results and high genetic variations of C. brachyotis bats on Java island may indicate that this species is a generalist and widespread species and suspected can adapt to different types of habitats. Nevertheless, C. brachyotis species rely on their foraging activities and roosting habitat in the trees or forest stands [44,45]. Therefore, the loss of the trees or forest stands in their habitats can threaten this species.
Although C. brachyotis have been categorized as the least concern on the IUCN Red List, its existence is essential for the ecosystem's sustainability and plant regeneration, especially for several trees in the forest and fragmented habitats. Previous studies showed that these species have a role in the pollination and seed dispersal of more than 16 plant species [2,3]. Hence, ecological management in various areas ranging from urban areas, plantations, and forests as foraging and roosting habitats of C. brachyotis on Java island is needed.

Conclusions
According to phylogenetic and genetic variation analysis using mtDNA D-loop sequence, C. brachyotis collected from seven habitats on Java island formed a single clade, overlapped between habitats, and was genetically close. This species on Java island also had high haplotype diversity (Hd = 0.933-1.000). A total of 22 haplotypes from 33 samples were found and split into two haplogroups. Furthermore, the haplotypes shared among habitats were also found in this study. This study result provides D-loop sequence data of C. brachyotis from Indonesia inferred from the mitochondrial DNA D-loop.