The mitochondrial D-loop is a robust maternal-species identifier in gibbons (Hylobatidae)

doi:10.21203/rs.3.rs-5039233/v1

Download PDF

Research Article

The mitochondrial D-loop is a robust maternal-species identifier in gibbons (Hylobatidae)

https://doi.org/10.21203/rs.3.rs-5039233/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Some gibbon species are difficult to distinguish phenotypically. This issue is compounded by recent changes in the gibbon taxonomy, which have sometimes occurred long after individuals were taken from the wild for captive breeding programmes. Furthermore, gibbon species within the Hylobates and Nomascus genera have been documented to hybridize to produce viable, fertile offspring, both in captivity and in the wild. This raises the possibility of cryptic hybrids within captive populations, threatening the genetic integrity of these genera. Phylogenetic methods using the mitochondrial D-loop and cox1 gene were compared to genetically identify gibbon species. Both loci identified lineages with cryptic hybrid ancestry. However, the D-loop outperformed the cox1 gene, providing higher resolution, particularly for Nomascus spp., which for some species were not monophyletic when using the cox1 gene. The D-loop also revealed a significantly higher number of unique sequences, making it more suitable for investigations into relatedness, such as female dispersal patterns.

Gibbons

Genetics

Species

Hybrids

Phylogenetics

Taxonomy and Evolution

Gibbons (Hylobatidae) are a family in the order Primates, known as the small or lesser apes, that with the great apes (Hominidae) form the superfamily of apes (Hominoidea) (Groves, 2001; Roos, 2016). It has been estimated that the gibbons last shared a common ancestor with the great apes 16–20 Million Years Ago (MYA) (Roos, 2016). The gibbon family is divided into four genera (Hoolock, Hylobates, Nomascus, Symphalangus), on the basis of chromosome complement and karyotype (ranging from 2n = 38 to 2n = 52) (Roos, 2016).Within these genera 20 species are currently recognised (Mittermeier, Rylands and Wilson, 2013; IUCN, 2020). It has been estimated that the current species diversity within the gibbons arose during the last ~ 15MYA (Matsudaira and Ishida, 2010; Thinh, Mootnick, Geissmann, et al., 2010). In addition, there is genetic evidence that some radiations occurred as recently as 2MYA (Thinh, Mootnick, Geissmann, et al., 2010; Carbone et al., 2014). All gibbon species are currently listed as Vulnerable, Endangered, or Critically Endangered by the International Union for Conservation of Nature (IUCN) (IUCN, 2020).

Resolving the evolutionary history of these taxa has proven complex and controversial. Numerous phylogenetic studies of the Hylobatidae have been conducted using mitochondrial DNA (mtDNA) genes and regions, such as Cytochrome b (Chatterjee, 2006; Thinh, Mootnick, Geissmann, et al., 2010; Thinh, Rawson, et al., 2010), ND3-ND4 (Takacs et al., 2005), the D-loop/Control Region (Roos and Geissmann, 2001; Monda et al., 2007; Whittaker, Morales and Melnick, 2007), and the whole mitochondrial genome (Chan et al., 2010; Matsudaira and Ishida, 2010, 2021; Fan et al., 2017). However, these studies have failed to resolve the gibbon phylogeny, with different studies producing conflicting results. Whilst the four genera are consistently monophyletic, their branching order varies, and relationships between species remain unclear. For example, using mitochondrial D-loop sequence data, Whittaker, Morales and Melnick (2007) found Hoolock to be basal, followed by Hylobates, with Symphalangus and Nomascus being the youngest sister genera. However, Roos and Geissmann (2001) found Nomascus to be basal, with Hoolock and Hylobates being the youngest and sister genera. Takacs et al. (2005), also found Hoolock to be basal, based on mitochondrial ND3-ND4 sequence data, but Hylobates and Symphalangus were sister genera, and the youngest. Analyses of nuclear DNA have, to date, likewise failed to produce a single phylogeny (Wall et al., 2013; Carbone et al., 2014; Veeramah et al., 2015; Shi and Yang, 2018).

Hybrids

Fertile intra-genera gibbon hybrids have been recorded both in captivity and in the wild (Brockelman and Schilling, 1984; Brokelman and Gittins, 1984; Tenaza, 1985; Marshall and Brockelman, 1986; Marshall and Sugardjito, 1986; Couturier and Lersould, 1991; Van Tuinen et al., 1999; Matsudaira et al., 2013; Nie et al., 2018). In addition, viable inter-genera hybrids have been documented in captivity (Myers and Shafer, 1979; Hirai et al., 2007). Such successful hybridisation events within genera demonstrate a low level of biological reproductive isolation. These observations indicate how closely related the different species are, particularly within the same genus. This level of recent relatedness is likely contributing to the difficulty in resolving the gibbon phylogeny, despite consistent effort.

Furthermore, some gibbon species appear phenotypically so similar to one another, that they can be difficult to distinguish, even for specialists (Geissmann, 1995; Mootnick, 2006; Thinh, Rawson, et al., 2010; Nie et al., 2018). This represents a particular problem for conservation centers like zoos, who may unknowingly pair individuals of different species, creating hybrids, which then continue to breed, as the institutions believe them to be of the same species (Geissmann, 1995). Changes in the recognised gibbon taxonomy since the founding of captive populations may exacerbate this issue. Furthermore, as hybrids occur in the wild, it is also possible for zoos to import wild-caught individuals without realising they are hybrids, particularly if the provenance of the individual is unknown (Mootnick, 2006).

Utility of mtDNA

Whilst the evolutionary history and relationships of the gibbon species may not yet have been determined, mtDNA phylogenetic studies do demonstrate that the genera and species frequently group in a monophyletic manner (though precise relationships vary depending on the marker analysed), making such analyses potentially useful for species identification. Mitochondrial DNA is especially suited to such studies because of its high mutation rate relative to nuclear DNA, providing a higher level of phylogenetic resolution at or below the species level (Wilson et al., 1985). Other practical advantages of mtDNA are that it can be recovered from degraded genetic samples such as hair, faecal samples or stomach contents, allowing use in field studies and forensics applications (Johnson, 2010; Spencer, Schmidt and Hummel, 2010).

Mitochondrial cox1 gene

The mitochondrial cox1 gene has been proposed as a universal genetic marker (‘DNA barcode’) to identify taxonomic groups (Hebert et al., 2003; DeSalle, Egan and Siddall, 2005), including primates (Lorenz et al., 2005; Hajibabaei, Singer and Hickey, 2006; Nijman and Aliabadian, 2010). However, to date, it has not yet been tested as a DNA barcode to specifically identify gibbon species.

Concerns have been raised as to the suitability of cox1 as a truly universal species identifier, as it is an essential gene, imposing limitations upon how many functional variations can exist (DeSalle, Egan and Siddall, 2005). The reliability of DNA barcoding, utilising cox1, for species across the primate order has been assessed by Hajibabaei, Singer and Hickey (2006). These authors concluded that while cox1 may provide adequate data to assign samples to known species, it was insufficient to assign novel species or to resolve deeper phylogenetic relationships (Hajibabaei, Singer and Hickey, 2006). The authors determined that cox1 is most suitable in cases without taxonomic ambiguity but maybe inadequate in cases of historical hybridisation, a point of view shared by Krishnamurthy and Francis (2012). As noted by Moritz and Cicero (2004) closely related taxa have not yet been sufficiently sampled to determine whether cox1 can differentiate between species that have recently diverged. These concerns may render cox1 an unreliable marker for gibbon species identification. One of the aims of the research reported here was to address this question.

Mitochondrial D-loop

The non-coding mitochondrial D-loop may be more suited to species identification for gibbons as it is a more rapidly evolving part of the mitochondrial genome (Garza and Woodruff, 1992; Avise, 2000; Roos and Geissmann, 2001; Whittaker, Morales and Melnick, 2007). Importantly, as discussed above, gibbon species are apparently very recently diverged, and therefore may require a more variable locus to provide sufficient phylogenetic resolution for species identification, which the mitochondrial D-loop could offer (Whittaker, Morales and Melnick, 2007).

The aim of this work was to use the mitochondrial genome as a species identifier by conducting phylogenetic analyses of sequence data. Additionally, we wish to test whether the ‘standard’ barcode, the cox1 gene, provides sufficient resolution for gibbon species identification, or whether the more variable D-loop is required in some cases.

Samples

A total of 140 blood, tissue, hair, and FTA blood-spot samples were acquired from 20 institutions within the European Association of Zoos and Aquaria (EAZA) (see Table 1). All invasive samples were residual material collected during medical procedures, or from autopsies. 73 samples from cultured gibbon cell lines, held at the Department of Biological Sciences, Graduate School of Science, at the University of Tokyo, were also analysed for the following species: Symphalangus syndactylus (n = 11), Nomascus leucogenys (n = 2), Nomascus siki (n = 1), Hylobates muelleri (n = 3), Hylobates abbotti (n = 2) Hylobates lar (n = 35), Hylobates pileatus (n = 9), Hylobates moloch (n = 1), Hylobates albibarbis (n = 4), Hylobates agilis (n = 5).

Table 1

Number of samples for each species by sample type from EAZA institutions
Species	No. of samples by type
Species	Extracted DNA	Blood	Tissue	Hair	FTA blood-spots	TOTAL
Symphalangus syndactylus	1	23	1	5	0	30
Nomascus leucogenys	1	9	0	9	0	19
Nomascus gabriellae	0	5	0	3	0	8
N. leucogenys x N. gabriellae	0	1	1	0	0	2
Nomascus siki	0	0	0	4	0	4
Hylobates agilis	0	8	0	0	0	8
Hylobates lar	0	9	1	3	0	13
Hylobates pileatus	0	11	1	4	0	16
Hylobates moloch	0	0	0	0	40	40
TOTALS	2	66	4	28	40	140

DNA Extraction

For blood and tissue samples, DNA was extracted using the Maxwell® 16 Automated Nucleic Acid and Protein Extraction System (Promega, Madison WI, USA), with the Maxwell® 16 Blood DNA Purification Kit, following the manufacturer’s instructions. Samples were loaded into the extraction cartridges in a dedicated pre-PCR Class II laminar flow hood (Walker Safety Cabinets, Glossop, UK). For hair samples DNA extractions were performed in a dedicated pre-PCR clean room, in a laminar flow cabinet (Starlab, Milton Keynes, UK), which was UV irradiated prior to each use. All pipettes and consumables were kept exclusively in the pre-PCR laminar flow cabinets, and so were also regularly UV irradiated. Nitrile gloves were UV irradiated prior to use. DNA was extracted using the QIAamp (Qiagen, Manchester, UK) DNA Investigator Kit, following the manufacturer’s instructions, with the following modifications: lysis at 56^oC took place overnight (approximately 18 hours), and the elution step was performed twice to increase yield. Samples were then concentrated using Amicon Ultra-0.5 mL Centrifugal Filters (Merck Millipore. Watford, UK), washing with 150µl TMT (10mM Tris-HCl, pH8.0). For the FTA blood-spot cards DNA was extracted using the Qiagen DNA Investigator kit, following the manufacturer’s instructions, in a dedicated pre-PCR Class II laminar flow hood.

PCR Amplification

The mitochondrial D-loop (nucleotides 1,319-1,898 of the NomLeu3 reference, Genbank assembly accession GCA_000146795.3) was amplified using primers GIBDLF3 (5’-CTTCACCCTCAGCACCCAAAGC-3’) & GIBDLR4 (5’-GGGTGATAGGCCTGTGATC-3’) (Andayani et al., 2001). The mitochondrial cox1 gene (nucleotides 2,001–2,832 of the NomLeu3 reference) was amplified using primers LLGIBCOX1_F (5’-CTGGTTATTCTCCACAAACC-3’) & LLGIBCOX1_R (5’-GAAGCCAATTGATATTATGGC-3’).

DNA sequencing

All PCR amplicons were directly sequenced via Sanger sequencing (Source BioScience, Nottingham, UK), using the amplifying primers.

Alignment and phylogenetics

MAFFT (v.7.490) alignments (Katoh et al., 2002; Katoh and Standley, 2013) were generated in Geneious Prime (Kearse et al., 2018) (algorithm G-INS-I, scoring matrix 200PAM/k = 2, gap open penalty = 2.8, offset value = 0.5) together with reference sequences (Genbank Accession numbers provided in figures; reference sequence for N. annamensis provided by C. Roos at the German Primate Center), and manually refined where appropriate. For D-loop, sequencing amplicons ranged from 190–596bp (mean 567bp). The final alignment length was 605bp. For the cox1 gene sequencing amplicons ranged from 791–832bp, and the final alignment length was 832bp. The alignments were used to generate Maximum-Likelihood phylogenetic trees using PhyML (Guindon, Dufayard and Lefort, 2010), in Geneious Prime (Kearse et al., 2018) with 100 bootstrap replicates of the data.

All unique sequences have been submitted to Genbank, accession numbers: PQ186079-186184.

D-loop

Samples

For this locus, 121 samples were sequenced from the EAZA collection, and 73 from the Department of Biological Sciences, Graduate School of Science, University of Tokyo.

Matrilines

All but one of the 48 documented matrilines (as recorded in the EAZA studbooks) had a unique D-loop sequence. The SIA18 and SIA23 matrilines had identical sequences. These matrilines are not recorded as being related. However, in both cases the founder parentage is unknown, so it cannot be determined whether these individuals have identical amplified D-loop haplotypes by chance or are identical by descent (IBD). Two samples from the Tokyo collection also had identical sequences: LAR_G712 and LAR_DZ10. Additionally, two samples from the EAZA collection were found to have identical D-loop sequences as two samples from the Tokyo collection: LAR05 (EAZA) and LAR_G110 (Tokyo), and PIL12 (EAZA) and PIL_KZ15 (Tokyo). As no pedigree information is available for the Tokyo samples, again, it cannot be determined whether they are IBD. One of each unique sequence was included in the alignment for the phylogenetic tree.

Phylogenetic analysis

As shown in Fig. 1. Hylobates form one clade, with Symphalangus, Hoolock, and Nomascus in the other, with Symphalangus and Nomascus clustering as sister taxa. Within the Hylobates clade H. agilis and H. albibarbis are sister to all other species. In the second major clade H. klossii is separate, H. pileatus and H. lar are sister taxa, and H. moloch forms a sub-clade with sister taxa H. muelleri and H. abbotti. Within the Nomascus clade N. concolor separates first, with N. siki, N. leucogenys, N. annamensis, N. gabriellae, and N. hainanus forming another clade. Within this clade. N. leucogenys and N. siki are sister taxa, as are N. annamensis and N. gabriellae, which form a sub-clade with H. hainanus.

Species identification

The average bootstrap support at the species nodes is 86.3%, ranging from 61% − 100% (N.B. Symphalangus syndactylus is not included in this calculation as it is the only species in the genus, nor are Hoolock spp. as the tree contains only six reference sequences from Genbank for the entire genus).

For the D-loop, 111/116 sequences grouped with their documented species. However, three purported H. agilis sequences grouped as H. lar (see Fig. 2). Bootstrap support for the H. lar clade is 89%. Additionally, two documented N. gabriellae sequences did not group as expected. The GAB02 sequence grouped as N. siki rather than N. gabriellae (see Fig. 3). Bootstrap support for this clade is 79%. The GAB09 sequence grouped with N. annamensis, with 87% support (see Fig. 3).

cox1

Samples

For this locus, 83 samples were successfully amplified sequenced from the EAZA collection.

Matrilines

Compared to the D-loop, far fewer cox1 matrilines had unique sequences. Of the 41 documented matrilines tested, only 26 had unique sequences. One of each unique sequence was included in the phylogenetic tree.

Phylogenetic analysis

As shown in Fig. 4, the branching order for cox1 was different to that obtained with the D-loop: Nomascus splits in to one clade, with the second clade separating Hoolock, and Symphalangus and Hylobates being sister taxa.

Species identification

Within Hylobates bootstrap support values are not available for H. klossii, H. abbotti, or H. albibarbis as only a single reference sequence is available for each (N.B. see Discussion below regarding possible misidentification of H. abbotti/H. muelleri sequences). For the remaining species, the average support at the species nodes is 97.6% (range 91%-100%). Within Nomascus, only a single sequence is available for N. concolor, no sequences are available for N. hainanus, and N. leucogenys and N. siki do not group monophyletically (see Fig. 4). Support for N. gabriellae is 52%, and 71% for N. annamensis.

All cox1 sequences grouped as expected, with only three exceptions: consistent with the D-loop data, AG02 grouped as H. lar (see Fig. 5), GAB02 grouped within the clade containing N. siki, and GAB09 grouped with the N. annamensis reference sequence (see Fig. 6).

Identification of species and hybrids

Overall, 96% (111/116) of individuals were unambiguously assigned to their provenance asserted species, via their mitochondrial D-loop sequences. This demonstrates that the D-loop performs well as a maternal species identifier for gibbons, in the tested species and samples. Specific exceptions to species identification were limited to two species. Three H. agilis individuals clustered with H. lar; one N. gabriellae sequence clustered with N. siki, and a second N. gabriellae matriline grouped with N. annamensis. As the remaining 96% of sequences grouped as expected we are confident that these results indicate genuine mixed-species ancestry in these matrilines. Furthermore, bootstrap support at the species nodes is generally high (mean 86.3%, median 88%).

The main exception to this robust species discrimination is the clade containing H. agilis and H. albibarbis. Historically H. albibarbis was considered a sub-species of H. agilis (Lyon, 1911), but was raised to full species level by Groves (2001), and this was supported by cytogenetic and molecular genetic analyses (Hirai et al., 2005, 2009). H. albibarbis and H. agilis are currently recognised as distinct species by the IUCN (IUCN, 2020). However, the sequences reported here do not group as monophyletic taxa. The sequences are grouped together in a single clade, with 74% bootstrap support, which suggests they may be the same species, as previously recognised. The purported H. albibarbis sequences cluster together in a group within this clade, with moderate bootstrap support (61%), suggesting the existence of a sub-species. However, the chromosomal translocation between the two taxa (Hirai et al., 2005), and their geographic isolation (H. agilis is found on the Malaysian peninsular and the island of Sumatra, while H. albibarbis is found only on the island of Borneo), support separate species status. Furthermore, recent work by Matsudaira and Ishida (2021) found nuclear data also support distinction of H. agilis and H. albibarbis. It is possible that the two species have diverged so recently as to have accumulated limited phylogenetic signal. Excluding these two taxa, species-level bootstrap support across Hylobates and Nomascus is strong (mean 89.4%, median 90%).

H. agilis and H. lar have been documented to produce fertile hybrid offspring, both in the wild and in captivity (Brokelman and Gittins, 1984; Van Tuinen et al., 1999). Mixed ancestry is therefore a plausible explanation for the anomalous H. agilis results. No pedigree information is available for the two cell line samples from Tokyo (AG_G80 and AG_JMC1). However, nuclear data confirm hybrid ancestry in the case of AG_G80 (Matsudaira and Ishida, 2021). These data showed the majority of the nuclear SNVs were of H. agilis (the documented species), with some H. lar. The levels of H. lar introgression were low, suggesting the hybridisation event was not recent (i.e. in the wild, rather than recently in captivity). The data presented here confirm the H. lar introgression in this individual, and demonstrate that the introgression comes from the maternal lineage. In the case of the EAZA individual AG02, the identity of the dam is unknown. It is possible a hybrid was accidentally created in captivity, if the dam was misidentified, although as these two species are readily distinguishable phenotypically, we consider this to be unlikely. Alternatively, the founding female taken from the wild could have been an H. lar x H. agilis hybrid, with the physical appearance of H. agilis, similar to the case of AG_G80, leading to misidentification in the studbook.

Similarly, N. siki x N. gabriellae hybrids have been documented in captivity (Nie et al., 2018), and suspected in the wild (Geissmann et al., 2000; Mootnick, 2006). Male N. siki and N. gabriellae individuals may be readily distinguished by their white and buff cheek colouration respectively. Females of these species however are very similar in appearance and can be difficult to distinguish reliably (Thinh, Rawson, et al., 2010; Harding, 2012). If we assume our sequencing data to be reliable, the anomalous grouping of GAB02 with N. siki may therefore be either the result of a hybrid individual being taken from the wild, or the founding female (the dam of GAB02) being a misidentified N. siki, particularly if their geographic origin was unknown, or incorrectly documented.

Historically N. annamensis was considered a sub-species of N. gabriellae, but was raised to full species status in 2010 (Thinh, Mootnick, Thanh, et al., 2010), 13 years after the founding female of the GAB09 matriline was taken into captivity. With only N. gabriellae being recognised at that time, this female could only have been identified as such. Furthermore, females of these species are indistinguishable phenotypically (Thinh, Mootnick, Thanh, et al., 2010), so it is unlikely the misidentification would have been diagnosed later, leading to unavoidable, if accidental, hybridization.

Implications for breeding programmes

As H. agilis are no longer bred for conservation by EAZA, these results do not have significant consequences for the remaining H. agilis population in the European collection (12 individuals remaining as of September 2023). However, the data reported here show H. lar introgression in three different lineages of H. agilis (one in EAZA, and two of the samples from the University of Tokyo), indicating the EAZA case is not isolated. It would therefore be prudent for other global breeding programmes to investigate their H. agilis groups, as well as H. lar and H. pileatus, as these species are also known to hybridize in the wild (discussed above).

However, the captive N. gabriellae population are part of active conservation breeding programmes, and so these results could have implications for these groups. The data reported here indicate N. siki and N. annamensis ancestry in two N. gabriellae lineages, and it is the stated policy of EAZA to avoid the breeding of hybrids (EAZA, 2013). It is important to note that these data are not conclusive evidence of introgression, for several reasons. Firstly, the resolution of N. leucogenys, N. siki, N. gabriellae, and N. annamensis remains uncertain. While these taxa are recognised as four distinct species by the IUCN (IUCN, 2020), it is noted that N. annamensis has been considered a sub-species of N. gabriellae, and that N. gabriellae and N. siki have historically been considered sub-species of N. leucogenys. The IUCN further notes that N. siki may not be a true species at all, but rather may be a naturally occurring hybrid of N. leucogenys and N. gabriellae. Significant overlap between these purported species was found in phylogenetic analysis of the mitochondrial Cytochrome b gene by Thinh, Rawson, et al., (2010). Our analysis of the D-loop, however, demonstrates clear distinctions between these four taxa (Fig. 3). This finding suggests that the individuals of the GAB02 and GAB09 lineages are genuinely of mixed-species ancestry.

When considering possible implications for the N. gabriellae breeding programme, it should also be noted that these data do not reveal the proportion of the current generation’s genome which might be affected. The youngest generation is three generations removed from the founding females. Theoretically, if all other breeding has been with true N. gabriellae individuals, then approximately 12.5% of the current genome would be expected to be N. siki or N. annamensis depending on the inheritance of alleles after random segregation at meiosis. Furthermore, if the founding female was a hybrid this proportion would be lower still. In the absence of sequence data from variable nuclear loci, this proportion cannot be estimated.

Cryptic relatedness

97% of documented matrilines for all species were found to have a unique sequence for the D-loop amplicon analysed here. Within the EAZA collection tested, unexpected identical sequences were limited to three cases, one for S. syndactylus, one for H. lar, and one for H. pileatus. Incomplete pedigree information prevents a definitive conclusion as to whether these matrilines are related. However, given the high number of unique haplotypes identified in this study (n = 113), we consider this possibility most likely.

D-loop vs cox1

Within Hylobates, where multiple cox1 sequences are available, bootstrap support for species clusters is high (91%-100%). However, H. abbotti does not group separately from H. muelleri, unlike in the D-loop phylogenetic tree. This may an artefact of only having a single sequence available for H. abbotti, which is therefore grouped with H. muelleri as the most closely related sequences. Alternatively, the reference sequences used may be erroneously labelled. In Chan et al. (2010), where the sequences were published, they were labelled as H. muelleri, however, the study does not document H. funereus or H. abbotti samples. It is possible therefore that some or all of the reference sequences are in fact H. abbotti. As with the D-loop phylogenetic tree, the single H. albibarbis sequence groups within H. agilis, although it does have a longer branch length (Fig. 5) suggesting greater divergence. The Nomascus clade is not clearly delineated by cox1 compared to the D-loop, in particular N. leucogenys, and N. siki, which do not group monophyletically. This makes cox1 unsuitable for species identification for these species. The results for AG02 and GAB09 are concordant with the results for the D-loop, being identified as H. lar and N. annamensis respectively. The sequence for GAB02 also groups within the clade containing N. siki. In contrast to the D-loop, matrilines are not readily distinguished using cox1 sequences. It is therefore not recommended for studies in which ascertaining relatedness is important, such as in captive populations, or studying female dispersal in the wild. Instead, the D-loop should be used in preference. Additionally, the D-loop had a higher successful amplification rate for the lower-quality hair samples compared to cox1 (68% and 59% respectively). This may be due to the smaller amplicon size of the D-loop compared to cox1 (~ 560bp and ~ 790bp respectively). This makes the D-loop a more attractive target locus for situations where only lower quality samples such as hair are available, as is often the case in wildlife rescue centers.

Based on the data reported here we recommend the D-loop should be used for gibbon maternal species identification, due to its higher resolution compared to cox1. In particular, we do not recommend the use of cox1 for species identification for Nomascus spp. as N. leucogenys and N. siki are not monophyletic when using this locus. By contrast, the D-loop sequences cluster clearly along recognised species boundaries, with strong support. The possible exception to this recommendation, is in the case of H. albibarbis, as discussed above. Two recognised species are currently absent from this work: Hylobates funereus, and Nomascus nasutus. (N.B. at the time of writing, two sequences for N. nasutus were published, but were significantly shorter than the other sequences used in these analyses, and so were excluded). Sequencing of individuals of known provenance for these species would provide the opportunity to construct a complete phylogenetic tree and most likely enable reliable maternal species identification for any individual.

A significant limitation of mitochondrial sequences is that they are maternally inherited. As a result, any introgression which has occurred in the paternal line will remain undetected. Therefore, markers from the nuclear genome which can identify species and possible hybrids from both the maternal and paternal lineages are highly desirable. Whole genome sequencing would be ideal as not only could hybrids be detected, but also the proportion of introgression quantified. However, such methods are currently impractical in terms of cost, time, data analysis, and sample requirements. One approach to this problem are nuclear markers, utilising species-specific retrotransposon insertions, which can be detected with simple PCR and gel electrophoresis technology (Lansdowne, 2022).

While phylogenetics using the mitochondrial D-loop is limited to maternal lineages, it is nevertheless a robust method, which can assign the maternal species origin of an individual with a high degree of confidence. Where introgression has occurred on the maternal side, it can also highlight cases of cryptic hybrids, as demonstrated here. We recommend use of the D-loop for gibbon species identification, rather than cox1, due to the greater resolution provided, the higher amplification rates for low quality samples, and the higher number of observed haplotypes which enable lineage determination.

Funding

Funding was provided by the BBSRC, the University of Leicester, and Twycross Zoo.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Compliance with Ethical Standards

All invasive samples were residual material collected either during medical procedures or autopsies, in compliance with each institutions’ standards, and with support from the EAZA Gibbon Taxon Advisory Group.

Author Contribution

The study was designed by L.L. and R.B. Data collection was performed by L.L., K.M., and S.M. Data analysis was performed by L.L. Samples at the University of Tokyo were provided by T.I. The first draft of the manuscript was written by L.L. and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Acknowledgement

We would like to thank Twycross Zoo (UK) for their support and collaboration, and also the other institutions who generously provided samples: Apenheul Primate Park (Netherlands), Howletts and Port Lympne (UK), Borås Zoo (Sweden), Bristol Zoo (UK), Cotswold Wildlife Park (UK), German Primate Center (Germany), Marwell Zoo (UK), Mulhouse Zoological and Botanical Park (France), Paignton Zoo (UK), Parc Animalier d'Auvergne (France), Reserve Africaine de Sigean (France), Selwo Aventura (Spain), Serengeti Park (Germany), Zoo Boissiere (France), Zoo de Cerza (France), Zoo de la Besancon Citadelle (France), Zoopark Erfurt (Germany), ZSL (UK), Zurich Zoo (Switzerland). Funding was provided by the BBSRC, the University of Leicester, and Twycross Zoo. We would also like to thank Tony King and Jane Hopper at The Aspinall Foundation for the provision of additional DNA sequences.

Data Availability

Sequences generated during the current study may be found on Genbank (Accession numbers: PQ186079-186184).

Andayani N et al (2001) ‘Genetic variability in mtDNA of the silvery gibbon: Implications for the conservation of a critically endangered species’, Conservation Biology, 15(3), pp. 770–775. https://doi.org/10.1046/j.1523-1739.2001.015003770.x
Avise JC (2000) Phylogeography: The History and Formation of Species. Harvard University Press
Brockelman WY, Schilling D (1984) ‘Inheritance of stereotyped gibbon calls’, Nature, 312(13), pp. 634–636. https://doi.org/10.1038/312634a0
Brokelman W, Gittins S (1984) Natural hybridisation in the Hylobates lar species group: implications for speciation in gibbons. In: Preuschoft H, Chivers D, Brockelman W (eds) The lesser apes: evolutionary and behavioural biology. Edinburgh University, Edinburgh, UK, pp 498–532
Carbone L et al (2014) ‘Gibbon genome and the fast karyotype evolution of small apes’, Nature, 513(7517), pp. 195–201. https://doi.org/10.1038/nature13679
Chan YC et al (2010) ‘Mitochondrial genome sequences effectively reveal the phylogeny of Hylobates gibbons’, PLoS ONE, 5(12). https://doi.org/10.1371/journal.pone.0014419
Chatterjee HJ (2006) ‘Phylogeny and biogeography of gibbons: A dispersal-vicariance analysis’, International Journal of Primatology, 27(3), pp. 699–712. https://doi.org/10.1007/s10764-006-9044-1
Couturier J, Lersould J (1991) ‘Karyotypic study of four gibbon forms provisionally considered as subspecies of Hylobates (Nomascus) concolor (Primates, Hylobatidae)’, Folia Primatologica, 56, pp. 95–104. https://doi.org/10.1159/000156533
DeSalle R, Egan MG, Siddall M (2005) ‘The unholy trinity: Taxonomy, species delimitation and DNA barcoding’, Philosophical Transactions of the Royal Society B: Biological Sciences, 360(1462), pp. 1905–1916. https://doi.org/10.1098/rstb.2005.1722
EAZA (2013) The modern Zoo: Foundations for Management and Development, EAZA Executive Office. https://doi.org/10.1038/119675a0
Fan P et al (2017) ‘Description of a new species of Hoolock gibbon (Primates: Hylobatidae) based on integrative taxonomy’, American Journal of Primatology, 79(5), p. e22631. https://doi.org/10.1002/ajp.22631
Garza JC, Woodruff DS (1992) ‘A phylogenetic study of the gibbons (Hylobates) using DNA obtained noninvasively from hair’, Molecular Phylogenetics and Evolution, 1(3), pp. 202–210. https://doi.org/10.1016/1055-7903(92)90016-A
Geissmann T (1995) ‘Gibbon Systematics and Species Identification’, International Zoo News, 42(8), pp. 467–501. https://doi.org/10.1017/CBO9781107415324.004
Geissmann T et al (2000) Vietnam Primate Conservation Status Review 2000 - Part 1: Gibbons. Fauna & Flora International, Hanoi, Vietnam
Groves C (2001) Primate Taxonomy. Smithsonian Institute, Washington DC, USA
Guindon S, Dufayard J, Lefort V (2010) New Algorithms and Methods to Estimate Maximum-Likelihood Phylogenies: Assessing the Performance of PhyML 3. 0. Syst Biol 59(3):307–321
Hajibabaei M, Singer GAC, Hickey DA (2006) ‘Benchmarking DNA barcodes: An assessment using available primate sequences’, Genome, 49(7), pp. 851–854. https://doi.org/10.1139/G06-025
Harding LE (2012) ‘Nomascus leucogenys (Primates: Hylobatidae)’, Mammalian Species, 44(890), pp. 1–15. https://doi.org/10.1644/890.1
Hebert PDN et al (2003) ‘Biological identifications through DNA barcodes’, Proceedings of the Royal Society B: Biological Sciences, 270(1512), pp. 313–321. https://doi.org/10.1098/rspb.2002.2218
Hirai H et al (2005) ‘A whole-arm translocation (WAT8/9) separating Sumatran and Bornean agile gibbons, and its evolutionary features’, Chromosome Research, 13(2), pp. 123–133. https://doi.org/10.1007/s10577-005-7475-8
Hirai H et al (2007) ‘A most distant intergeneric hybrid offspring (Larcon) of lesser apes, Nomascus leucogenys and Hylobates lar’, Human Genetics, 122(5), pp. 477–483. https://doi.org/10.1007/s00439-007-0425-0
Hirai H et al (2009) ‘The Gibbons’, The Gibbons, pp. 37–49. https://doi.org/10.1007/978-0-387-88604-6
IUCN (2020) The IUCN Red List of Threatened Species. Version 2020-2. http://www.iucnredlist.org
Johnson RN (2010) ‘The use of DNA identification in prosecuting wildlife-traffickers in Australia: Do the penalties fit the crimes?’, Forensic Science, Medicine, and Pathology, 6(3), pp. 211–216. https://doi.org/10.1007/s12024-010-9174-9
Katoh K et al (2002) ‘MAFFT: A novel method for rapid multiple sequence alignment based on fast Fourier transform’, Nucleic Acids Research, 30(14), pp. 3059–3066. https://doi.org/10.1093/nar/gkf436
Katoh K, Standley DM (2013) ‘MAFFT multiple sequence alignment software version 7: Improvements in performance and usability’, Molecular Biology and Evolution, 30(4), pp. 772–780. https://doi.org/10.1093/molbev/mst010
Kearse M et al (2018) ‘Geneious version 11.1.4&#8217
Krishnamurthy PK, Francis RA (2012) ‘A critical review on the utility of DNA barcoding in biodiversity conservation’, Biodiversity and Conservation, 21(8), pp. 1901–1919. https://doi.org/10.1007/s10531-012-0306-2
Lansdowne L (2022) Genetic Tools for the Captive Conservation Management of Gibbons (PhD Thesis). University of Leicester, UK. https://figshare.le.ac.uk/articles/thesis/Genetic_Tools_for_the_Captive_Conservation_Management_of_Gibbons/20348694
Lorenz JG et al (2005) ‘The problems and promise of DNA barcodes for species diagnosis of primate biomaterials’, Philosophical Transactions of the Royal Society B: Biological Sciences, 360(1462), pp. 1869–1877. https://doi.org/10.1098/rstb.2005.1718
Lyon MW (1911) ‘Mammals Collected by Dr. W. L. Abbott on Borneo and Some of the Small Adjacent Islands.’, Proceedings of the United States National Museum, 40(1809)
Marshall JT, Brockelman WY (1986) Pelage of hybrid gibbons (Hylobates lar x H. pileatus) observed in Khao Yai National Park, Thailand. Nat History Bull Siam Soc 34:145–157
Marshall JT, Sugardjito J (1986) Gibbon systematics. In: Swindler DR, Erwin J (eds) Comparative Primate Biology. Volume 1: systematics, evolution, and anatomy. Alan R. Liss, New York, pp 137–185
Matsudaira K et al (2013) ‘Molecular evidence for the introgression between Hylobates lar and H. pileatus in the wild’, Primates, 54(1), pp. 33–37. https://doi.org/10.1007/s10329-012-0323-5
Matsudaira K, Ishida T (2010) ‘Phylogenetic relationships and divergence dates of the whole mitochondrial genome sequences among three gibbon genera’, Molecular Phylogenetics and Evolution, 55(2), pp. 454–459. https://doi.org/10.1016/j.ympev.2010.01.032
Matsudaira K, Ishida T (2021) ‘Divergence and introgression in small apes, the genus Hylobates, revealed by reduced representation sequencing’, Heredity, (June 2021), pp. 1–11. https://doi.org/10.1038/s41437-021-00452-7
Mittermeier R, Rylands A, Wilson D (eds) (2013) Handbook of the Mammals of the World, Volume 3: Primates. Lynx Edicions
Monda K et al (2007) ‘Mitochondrial DNA Hypervariable Region-1 Sequence Variation and Phylogeny of the Concolor Gibbons, Nomascus’, American Journal of Primatology, 69, pp. 1285–1306. https://doi.org/10.1002/ajp
Mootnick AR (2006) ‘Gibbon (Hylobatidae) Species Identification Recommended for Rescue or Breeding Centers’, Primate Conservation, 21, pp. 103–138. https://doi.org/10.1896/0898-6207.21.1.103
Moritz C, Cicero C (2004) ‘DNA barcoding: Promise and pitfalls’, PLoS Biology, 2(10). https://doi.org/10.1371/journal.pbio.0020354
Myers RH, Shafer DA (1979) ‘Hybrid Ape Offspring of a Mating of Gibbon and Siamang’, Science, 205, pp. 308–310. https://doi.org/10.1126/science.451603
Nie WH et al (2018) ‘Species identification of crested gibbons (Nomascus) in captivity in China using karyotyping- and PCR-based approaches’, Zoological research, 39(5), pp. 356–363. https://doi.org/10.24272/j.issn.2095-8137.2018.036
Nijman V, Aliabadian M (2010) ‘Performance of distance-based DNA barcoding in the molecular identification of Primates’, Comptes Rendus - Biologies, 333(1), pp. 11–16. https://doi.org/10.1016/j.crvi.2009.10.003
Roos C (2016) ‘Phylogeny and Classification of Gibbons (Hylobatidae)’, in The Evolution of Gibbons and Siamang, pp. 3–41. https://doi.org/10.1007/978-1-4939-5614-2_1
Roos C, Geissmann T (2001) ‘Molecular phylogeny of the major hylobatid divisions’, Molecular Phylogenetics and Evolution, 19(3), pp. 486–494. https://doi.org/10.1006/mpev.2001.0939
Shi CM, Yang Z (2018) ‘Coalescent-Based Analyses of Genomic Sequence Data Provide a Robust Resolution of Phylogenetic Relationships among Major Groups of Gibbons’, Molecular Biology and Evolution, 35(1), pp. 159–179. https://doi.org/10.1093/molbev/msx277
Spencer PBS, Schmidt D, Hummel S (2010) ‘Identification of historical specimens and wildlife seizures originating from highly degraded sources of kangaroos and other macropods’, Forensic Science, Medicine, and Pathology, 6(3), pp. 225–232. https://doi.org/10.1007/s12024-009-9119-3
Takacs Z et al (2005) ‘A complete species-level phylogeny of the Hylobatidae based on mitochondrial ND3-ND4 gene sequences’, Molecular Phylogenetics and Evolution, 36(3), pp. 456–467. https://doi.org/10.1016/j.ympev.2005.03.032
Tenaza R (1985) ‘Songs of hybrid gibbons (Hylobates lar × H. muelleri)’, American Journal of Primatology, 8(3), pp. 249–253. https://doi.org/https://doi.org/10.1002/ajp.1350080307
Thinh VN, Mootnick AR, Thanh VN et al (2010) A new species of crested gibbon, from the central Annamite mountain range. Vietnamese J Primatol 1(4):1–12
Thinh VN, Mootnick AR, Geissmann T et al (2010) ‘Mitochondrial evidence for multiple radiations in the evolutionary history of small apes’, BMC Evolutionary Biology, 10(1). https://doi.org/10.1186/1471-2148-10-74
Thinh VN, Rawson B et al (2010) ‘Phylogeny and distribution of crested gibbons (genus Nomascus) based on mitochondrial cytochrome b gene sequence data’, American Journal of Primatology, 72(12), pp. 1047–1054. https://doi.org/10.1002/ajp.20861
Van Tuinen P et al (1999) ‘Complex, compound inversion/translocation polymorphism in an ape: Presumptive intermediate stage in the karyotypic evolution of the agile gibbon Hylobates agilis’, American Journal of Physical Anthropology, 110(2), pp. 129–142. https://doi.org/10.1002/(SICI)1096-8644(199910)110:2<129::AID-AJPA2>3.0.CO;2-M
Veeramah KR et al (2015) ‘Examining phylogenetic relationships among gibbon genera using whole genome sequence data using an approximate bayesian computation approach’, Genetics, 200(1), pp. 295–308. https://doi.org/10.1534/genetics.115.174425
Wall JD et al (2013) ‘Incomplete Lineage Sorting Is Common in Extant Gibbon Genera’, PLoS ONE, 8(1), pp. 1–5. https://doi.org/10.1371/journal.pone.0053682
Whittaker DJ, Morales JC, Melnick DJ (2007) ‘Resolution of the Hylobates phylogeny: Congruence of mitochondrial D-loop sequences with molecular, behavioral, and morphological data sets’, Molecular Phylogenetics and Evolution, 45(2), pp. 620–628. https://doi.org/10.1016/j.ympev.2007.08.009
Wilson AC et al (1985) Mitochondrial DNA and two perspectives on evolutionary genetics. 26:375–400

No competing interests reported.

Download PDF

Reviewers invited by journal
28 Oct, 2024
Editor assigned by journal
06 Sep, 2024
Submission checks completed at journal
06 Sep, 2024
First submitted to journal
05 Sep, 2024

You are reading this latest preprint version

The mitochondrial D-loop is a robust maternal-species identifier in gibbons (Hylobatidae)

Status:

Version 1

Abstract

Figures

Introduction

Taxonomy and Evolution

Hybrids

Utility of mtDNA

Mitochondrial D-loop

Materials and Methods

Samples

DNA Extraction

PCR Amplification

DNA sequencing

Alignment and phylogenetics

Results

D-loop

Samples

Matrilines

Phylogenetic analysis

Species identification

cox1

Samples

Matrilines

Phylogenetic analysis

Species identification

Discussion

Identification of species and hybrids

Implications for breeding programmes

Cryptic relatedness

D-loop vs cox1

Conclusions and future work

Declarations

Funding

Competing Interests

Compliance with Ethical Standards

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Status:

Version 1