Rediscovery and Characterization of Citrus Indica Tanaka, A Wild Endangered and Progenitor Species in Dailong Forest, Manipur, and Recommendations for Its Conservation.

Citrus indica Tanaka is a wild endangered and one of the progenitor species of citrus endemic to North East (NE) India, which is reported in Nokrek Biosphere Reserve in the Garo hills of Meghalaya. A wild orange morphologically claimed to be C. indica was encountered in the Dailong forest of Tamenglong district, Manipur which is located in the Indo-Burma biodiversity hotspot region. Surveys were conducted to establish the exact identity of possible C. indica and study the natural habitat in Dailong forest area. The traditional morpho-taxonomic description for the identication of the plant is performed along with the molecular characterization using three chloroplast regions (trnL-F, psbK-I, matK-5′trnK spacer) and one nuclear (ITS) region. Phylogenetic analysis revealed that the C. indica found in Dailong Forest is monophyletic with the earlier reported C. indica sequence from Garo hills, Meghalaya. Both morphological and molecular data support the identity of C. indica that has been collected from Manipur. However, lack of knowledge and clearing of forest cover has led to an urgent need for the implementation of conservation strategies to safeguard this endangered species. This study supports the evidence that NE India is one of the origin centres for citrus and this nding can guide the conservationists to initiate priority for adopting possible methods of proper conservation of C. indica in the region.


Introduction
Citrus indica Tanaka belongs to the genus Citrus of the family Rutaceae and subfamily Aurantiodeae which was rst discovered and characterized by Tanaka (1928). It is believed to be the most primitive and progenitor of all citrus species and is endemic to NE India (Singh 1981). Presently, it is reported to be endemic to the Tura range of Garo Hills of Meghalaya (Singh and Singh 2003) and the Garo tribes used the fruits of C. indica for medicinal and religious purposes (Malik et al. 2006). As the natural population of C. indica got drastically decreased due to large scale deforestation, Citrus Gene Sanctuary was established in the Nokrek Biosphere Reserve to protect the area (Singh 1981). Conservation of this species is required as it is endemic to only a speci c region with a particular microclimate. Citrus genetic resources of NE India need appropriate conservation strategies as the wild and indigenous population is being eroded due to various anthropogenic activities like jhumming and introduction of exotic species, etc. C. indica is not only exposed to habitat destruction but also has extremely low genetic diversity, high habitat speci city, and low regeneration which pose a serious threat in its natural population (Kumar et al. of citrus have been enlisted in the digital database on the citrus biodiversity of Manipur (Sanabam et al. 2012). Bhattacharya and Dutta (1956) reported 17 species, 52 cultivars, and 7 probable natural hybrids of citrus are found growing in natural wild conditions in the NE India region. Many citrus species and progenitors are believed to have originated in this region (Malik et al. 2012). In a study to trace the proposed origin of citrus and its ancient dispersal routes, it was found that the ancestral citrus species are migrated originally from NE India, northern Myanmar, and north-western Yunnan (Wu et al. 2018).
Thus, the NE India region is considered to be one of the centres of origin of citrus. However, citrus germplasm of NE India has been minimally utilized in the improvement programs due to a lack of exploration and characterization (Sharma et al. 2004). An approach for systematic exploration, characterization, and conservation of C. indica is still lacking (Malik et al. 2006). Systematic characterization including morphological and molecular identi cation should be implemented to assess Dailong is located in Tamenglong district of Manipur, North East India (NE India) which is part of the Indo-Burma biodiversity hotspot region. Tamenglong is the least densely populated district of Manipur which has a forest cover of 88%. Dailong forest covers an area of 11.35 km2. In May 2017, the Government of Manipur declared Dailong village as the 'Biodiversity Heritage Site' under the Biological Diversity Act, 2002 Sect. 37(1) because of its rich biodiversity of ora and fauna. This village is home to the Rongmei Naga tribes of Tamenglong district. Even though there was an earlier report of the occurrence of C. indica in Manipur (Tanaka 1937;Singh 1981;Sharma et al. 2004), systematic efforts for characterization and conservation status has not been carried out in detail. The present study attempts to analyse the distribution, characterization and conservation of C. indica along with its associated local indigenous knowledge. Both the morphological and molecular analyses were employed to characterize the species as proper identi cation of species is fundamental for biodiversity conservation.

Survey and morphological characterization
During the surveys undertaken for citrus germplasm collection and citrus greening disease in Tamenglong district of Manipur, few patches of a type of wild citrus (suspected to be Indian wild orange or Citrus indica) were encountered in a moist tract of Dailong forest. Further exploration was continued from 9th to 13th April 2018 to record the presence of any other population inside the forest. Information was collected from people who dwelled near the forests and subsequently two other sites were detected inside the forest (Table 1). A possible threat associated with the natural habitat and conservation issues had been analyzed through interaction with the community of Dialong village. Indigenous and traditional information of the species was collected by interviews with the local tribes inhabiting the area.
Meteorological data of the natural habitat of this species was recorded to know the region's microclimate. Photographs of the plants were taken in its natural habitat area (Fig. 1). The coordinates and elevations of the locations were recorded using a GPS unit (Garmin) ( Table 1). The survey was undertaken in such a manner that involved minimal disturbance and harm to the target plant and also the overall forest environment in Dailong area. General eld observations along with the natural population were brie y analyzed. All the individual plants were critically studied to determine their correct taxonomic identity and treated as a representative specimen. Young leaf tissues and fruits from all the samples were immediately stored in silica gel for subsequent DNA isolation in the laboratory. One representative voucher specimen of the collected samples is authenticated, and stored in the herbarium of the Institute of Bioresources and Sustainable development (IBSD), Imphal with the herbarium accession number 263.
Morphological characters of the citrus were recorded using 'Descriptors for citrus' by the International Plant Genetic Resources Institute (IPGRI) Rome, Italy. 14 quantitative and 27 qualitative characters were characterized (Table 2a and   The leaf samples were taken for total plant genomic DNA isolation. About 100 mg of each sample was ground in liquid nitrogen using a pre-cooled mortar and pestle. Total DNA extraction from this powdered tissue was further performed with the DNeasy Plant Mini Kit (Qiagen, Germany) by following the user's instructions. Nucleic acid was quanti ed by spectrophotometry (Nanodrop 2000, Thermo Scienti c) and it is quality checked by electrophoresis on 0.8% agarose gel.
PCR ampli cation of three chloroplast regions (matK-5′trnK spacer, psbI-psbK, and trnL-F) and one nuclear region ITS (ITS1, 5.8S, and ITS2) was performed by using the respective primers ( Table 3). The PCR reaction was performed in a C1000 Touch™ Thermal cycler (Biorad) with 25µL reaction volume. The reaction mixture consists of 10X PCR buffer, 2-25 mM of MgCl 2, 5-20 pmol of each primer, 0.125 mM-1.25M for each dNTPs, 0.5-1U Taq DNA polymerase (Sigma-Aldrich), and genomic DNA of 10-50 ng. PCR pro le followed for different primer pairs was given in Table 3. The PCR products obtained were electrophoresed in 1.2% agarose gel for 1 hour and 30 minutes at a constant volt of 60V.

Sequencing and phylogenetic analysis
The bands obtained were excised and eluted using Qia quick gel extraction kit (Qiagen) following the manufacturer's instructions. The eluted products were sent for direct sequencing at Bioserve Technologies, Hyderabad. Using BLAST analysis of the NCBI database (Altschul et al. 1997), the identity of sequences was con rmed. The consensus sequence of the four genomic regions (three chloroplast and one nuclear) was obtained using Bioedit software 7.05 (Hall, 1999). Closely related sequences were retrieved from GenBank database and used for the construction of phylogeny (Supplementary Table 1). Multiple alignments of sequences in this study and sequences from the GenBank database were performed with the ClustalW program incorporated in MEGA X (Kumar et al., 2018). One representative sequence of each of the marker sequences (i.e., four sequences) used in this study was annotated and deposited in the National Centre for Biotechnological Information (NCBI) GenBank database.
Phylogenetic analysis of the ITS, trnL-F, psbK-psbI and matK-5′trnK spacer regions were generated using the Maximum Likelihood method incorporated in the software IQ-TREE version 1.

Sampling and Morphological characterization:
The Indian wild orange or C. indica was found distributed in Dailong village of Tamenglong district, Manipur (Fig. 2). C. indica is locally known as 'Garuan-thai' ('Garuan' means cane, 'thai' means fruit) among the Rongmei tribes. The word 'cane' is given because the fruit stem is woody, slender, exible, and sometimes climber. From interactions with senior members of the community, it was learnt that the particular citrus specimen had a distinct local name. The existence of a local name further indicated that the particular genotype/specimen has been associated with the community for a long time. During the survey, a total of thirty-two specimens of wild C. indica were identi ed from three spots in the forest area of Dialong village (Table 1). In one spot, 20 specimens were recorded from, where C. indics is found maximally concentrated while the other two spots had a sparse and scattered occurrence. The type of plants varied from a thorny bush to a climber of 3 to 4 meters in height. Among the specimen, only ve of them had a thick and bushy canopy and certainly bore fruit. Plants from the same spot were separated at least by 5 meters. The natural population associated with the wild orange in the forest includes Xanthoxyllum sp, Melocanna bambusoides, Melocanna baccifera, Castapnosis spp., Medusferrea sp, Albizzia spp., Phoebe hainesiana, Albizia lebbeck, Tectona grandis, Quercus spp., orchids species like Dendrobium sp, Vanda sp, and several edible wild mushrooms.
The village of Dailong enjoyed high rainfall conditions of 3,135 mm with humidity of 76% and average temperature ranging from 4 to 31°C (Directorate of Environment & Climate Change, Govt of Manipur).
Preference for a unique shady microclimate comes to light as all the three locations come within the canopy of tall and dense forest plants. Morphological observation of the specimens does not reveal any distinctive character among them. However, variation in their habitat is quite signi cant in that some were small bushy shrubs, others straight tall to climbers type with spread canopy. Despite existing in wild conditions, the citrus plants were relatively free from any symptoms of disease.
The quantitative and qualitative morphological characters of fruit, leaf, and seed were recorded (Fig. 3) and given in Table 2a and b respectively. Fruits were obloid which weighed ranging from 21.60 to 30.71 gm. The fruit rind is whitish with a thickness of 0.17 to 0.25 cm. The outer dark orange fruit skin was smooth and shiny consisting of conspicuous oil glands with density 34.30 to 44.23 cm 2 . The number of fruit segments ranges from 9.00 to 10.00. Leaf division is simple with elliptic lamina length 8.35 to 12.00 cm, tapering at both ends. Seeds are ovoid, smooth, and creamish ranging from 7.00 to 10.00 per fruit. They are monoembryonic with green cotyledon and brown chalazal spot color.

Molecular Characterization
The four regions (three chloroplast and one nuclear) of citrus were utilized to con rm the taxonomic identity of the Indian wild orange 'Garuanthai' and also inferred their phylogenetic a nities to related taxa. Using the four pairs of primers for the three chloroplast sequences psbK-psbI intergenic sequence, trnL-F intergenic spacer region, matK-5′ trnK spacer region, and one nuclear region ITS region, we obtained sequences from multiple samples from the three locations. The sequence alignment revealed sequence identity in all of the samples (n = 32). Based on the sequence analysis for each marker genes, all the specimens collected were identical. So, only one representative sequence was taken for each gene to annotate and submit in the NCBI GenBank with accession numbers KY963832, MN447519, MN443629, and MT782064 and further used for phylogenetic analysis. The phylogenetic analysis revealed that the sequence of 'Garuanthai' grouped with the C. indica in all the Maximum Likelihood trees consistently with all the four sequences ( Fig. 4a, b, c, and d), thereby con rming the species to be C. indica. Zanthoxylum monophyllum, Poncirus trifoliata, Atalantia monophylla and Koelreuteria paniculata remained as outgroups in the respective phylogenetic trees in which they are incorporated.

Sites and morphological characterization
A recent survey to Dailong Forest Tamenglong district reported the natural wild population of C. indica, locally known as 'Garuanthai' among the Rongmei tribes. In Fig. 2, the collection site of the C. indica in the present study is shown in the red triangle while the sites of earlier explorations are depicted in red circle. Among the earlier exploration, the presence of the species were recorded only in the Garo hills of Meghalaya (Singh and Singh 2003). Our present study reports the re-occurrence of this species in Manipur in Tamenglong District where it was said to be found in earlier studies. Such ndings show that detailed research and studies should be carried out in various parts of NE India which remain largely unexplored because of its inaccessibility and di cult topographic nature. The meteorological data of Dailong forest area also support the microclimate of C. indica with low temperature, high rainfall, and humidity which is similar to that of the speci c microclimate of Garo hills, which may be the probable reason for occurrence of the plants in this forest.
Morphological characterization of citrus often creates confusion among the taxonomists as there are innumerable hybrids and mutants of citrus due to frequent hybridization, mutation, and polyploidy among the citrus species throughout the world (Susheel Kumar et al. 2013). In the present study, the morphological characters are characterized based on the Descriptors for citrus (IPGRI 1999) protocol which is a standardized tool for the characterization system. In earlier works, morphological classi cation has relied on a combination of few characters of leaf, seed, ower, and fruit (Swingle 1943

Molecular characterization
In the four DNA regions analysed (trnL-trnF, psbI-psbK, matK-5′trnK spacer, and ITS), we detected the sequences obtained from the present study clustered with the C. indica that is found in Garo hills of Meghalaya. The emergence of the DNA barcoding technique has helped in calibrating current taxonomic resolution and provided a straightforward identi cation system when there exist troubleshooting in morphology-based taxonomy ( The chloroplast psbK-psbI intergenic spacer was proven not only to be the most reliable barcode (Suzuki et al., 2014) but it was also useful for phylogenetic relationships analysis at the taxonomic level (Enan and Ahmed 2016). Moreover, chloroplast intergenic spacer psbK-psbI was included as a potential candidate locus for DNA barcode because of its evolutionary rates and high capability to distinguish independently evolving beings corresponding to taxonomic species (Lahaye et al. 2008).

Conservation
Citrus indica is unique citrus within the true citrus species. Known for its poor regeneration and endemic nature, this citrus genotype preferred a distinct microclimate for its survival and existence. Though the presence of C. indica has earlier been reported from different restricted pockets in NE India, its existence and conservation cannot be ascertained in many of these locations now. After a long period, the prevalence of C. indica was only recently reported from Behali Reserve Forest of Assam (Dipankar et al. 2018) and thus it is scantily reported even from NE India though the region is regarded as a natural home for Citrus. An assessment of ecological conditions, habitat, and interaction with people inhabiting the locality will enable us to give a clear view for the survival of such rare and endemic citrus genotype. Apart from this, morphological data and clear molecular analysis remain essential for correct identi cation and further conservation efforts.
Various wild and semi-wild citrus grow in the NE India forest regions which still need to be explored for their proper identi cation, collection, and conservation (Sharma et al. 2004). Dailong is one such area, which is endowed with a rich biodiversity of ora and fauna. The forest consists of montane wet temperate, tropical evergreen, semi-evergreen and riparian forest. C. indica is one of the seven species of Indian citrus enlisted in the endangered list, which requires urgent and special consideration for conservation because of its endemic nature and also the degree of threat perception is high (Malik et al. 2006). In situ conservation seems to be the most suitable option for C. indica as the previous study by Malik et al. 2006 reported that the plants growing outside its habitat did not show owering fruiting. This can be related to the complex phenomena of the lack of adaptation to the man-made environment (Zeven and de Wet 1982). Kumar et al. 2010 also observed that after propagation of C. indica from seeds, its growth and survival outside its natural environment became di cult. A study in Canary Island, Spain recommended reintroduction, habitat management, and protection as the most important tools for conservation and management of endangered plants (Marrero-Gómez et al. 2003). However ex situ methods of conservation are also given importance to protect the endangered species (Laskar et al. 2009). National Bureau of Plant Genetic Resources (NBPGR), Delhi has preserved the pollens and seeds of C. indica in Cryogene Bank for long term preservation (Malik et al. 2006). Such an initiative for ex-situ germplasm conservation of citrus can help in the rehabilitation of endangered species (Kumar et al. 2010). Cryopreservation technique and other ex situ conservation actions have been developed for longterm conservation of critically endangered species Rubus humulifolius in Finland (Edesi et al. 2020). Sharma et al. 2004 also suggested the need for more nature reserves, gene sanctuaries, and parks with the inclusion of citrus species. In Burkina Faso, the conservation strategy and survival rate of Zanthoxyllum zanthoxyloides seedlings were suggested to be managed by establishing nurseries, botanical gardens, and target places for safeguarding (Ouédraogo et al. 2019). Those sites were entrusted to the local communities thereby facilitating nancial support for the local communities to encourage them to actively participate in the conservation of the species. Likewise, the conservation strategy for C. indica can also be implemented in the present study area. Moreover, any conservation strategies related to natural resources needs public education and proper awareness to the people residing in the area for maintaining a healthy environment (Kaya and Raynal 2001). Thus a holistic approach including both the ex-situ and in situ conservation methods should be developed to contribute to Citrus improvement programs (Malik et al. 2006). Singh and Singh (2003) also stated that the anthropogenic activities are causing a serious threat to the citrus biodiversity in the NE India region for which, it has been classi ed as a hotspot. C. indica is seldom cultivated and commercialized as the fruits are small, rarely edible (Bhattaccharcharya and Dutta 1956; Tanaka 1937). The present exploration in Dailong forest area also indicated no domestication and no utility of the species known to the local tribes. It remained as ordinary wild citrus to them with no economic importance. So there was no speci c conservation policy taken up in that area particularly for the citrus species. However, in the Nokrek Biosphere reserve buffer zone area, preliminary domestication and conservation measures have been taken up. C. indica locally known as Memang Narang in Garo language is well known to the local tribes of this region and is used for various medicinal and ritual purposes. So the local tribes had cultivated in the area (Singh 1981;Malik et al. 2006). Maybe we can assume that the species had been conserved in Garo hills but not in other parts of the NE India because of its conservation by domestication by Garo tribes. In the other parts of NE India, there are no speci c uses and medicinal values known to them. Also, the Rongmei tribes of Dailong Forest have not used the part so far. They know its existence in the forest as wild form; however, there is no domestication history of the species in the place so the conservation motive is very low to the local people as domestication is one of the important ways to conserve the germplasm of species. During the survey, C. indica was observed to be healthy and free from any kind of disease. Such property of C. indica can provide rootstock of high quality for other commercial citrus varieties in the future (Malik et al. 2006). Conservation of C. indica was already initiated around ten decades ago in the Garo hills of Meghalaya by establishing the Citrus Gene Sanctuary which is a part of Nokrek Biosphere Reserve (Singh 1981). Likewise, the protection and rehabilitation of germplasm of the present study area should be proposed and Dailong forest should be recognized as an in situ 'Citrus Sanctuary'.
Detailed interviews with the local tribes and elders of Dailong area revealed that they have been practicing conservation activities for many generations in the form of protected sacred groves called 'Raengan'. C. indica was reported to be found in abundant according to the elders of Dailong village but now the plant has shrunk its population due to earlier destruction of forest for jhum cultivation. However, an initiative was taken way back in 1978 by Rongmei tribes of Dailong Village by preventing jhum cultivation and other exploitation from the forest. Further discussion was also continued with the members of Dailong Ecology & Environment Preservation Society (DEEPS; local NGO since 1978), we came to know that they played a critical role in spreading awareness among the Rongmei tribes for forest conservation. Despite all the efforts of the local tribes to protect the forest, recent development has hampered the forest ecosystem. There are road and railway construction, and extensive illegal timber logging which led to the loss of large forest area and ultimately leading to the loss of rare species. Even though the local tribes are protecting the forest area, adequate conservation measures have not been implemented so far from the government or other related organizations. There is an urgent need for identifying the priority areas for protection. In southern China, priority protected areas (PPAs) planning methods have been proposed to conserve endangered and threatened plant species in the identi ed locations (Wang et al. 2015).
Efforts should be given in collection, characterization, and conservation of the wild and semi-wild species of citrus in NE India. The results of our present study try to give insight to develop a conservation management policy for such wild and endangered species. Future studies including these species will give deeper insights into their evolution and geographic origin. It is reported that C. indica is inferred to have originated from NE India region, further studies can be explored whether the ancestor species colonized which location rst (likely from Nokrek, Meghalaya) followed by dispersal to other parts of other NE India. In the present study, results derived from morphology and sequence data indicated that 'Garuanthai' is identi ed as C. indica which is grown in wild habitat in Dailong forest, Manipur. The present work also supports the evidence of the origin of citrus in NE India thereby locating a new habitat of the ancestral citrus species. Both the in situ and ex situ conservation strategies should be approached to protect and conserve this species. These measures should also involve the local population by giving them proper information and awareness campaigns about the importance of habitat conservation and management to protect this precious habitat of C. indica.

Declarations Acknowledgments
The villagers of Dailong granted access to the forest area and provided accommodation for the survey team. The authors extended their special thanks to Brajesh (wildlife activist and photographer) and the villager team for their help extended during the exploration trips. We also thank the Director, Institute of Bioresources and Sustainable Development, Imphal for providing nancial assistance and research facilities. The IBSD communication number of the manuscript is IBSD/2020/01/027.

Compliance with ethical standards
This work does not contain any studies with human participants or animals performed by any of the authors.

Informed Consent
Informed consent was obtained from all individual participants included in the study.

Declaration of competing interest
The authors declare that they have no known competing nancial interests or personal relationships that could have appeared to in uence the work reported in this paper.  Table 3 List of primer sequences used for phylogenetic analyses and PCR pro le.