Reach Us +441704 335730
Genetic Variation and Structure in Remnant Population of Critically Endangered Melicope zahlbruckneri | OMICS International
ISSN: 2332-2543
Journal of Biodiversity & Endangered Species
Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on Medical, Pharma, Engineering, Science, Technology and Business
All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Genetic Variation and Structure in Remnant Population of Critically Endangered Melicope zahlbruckneri

Raji JA1,2* and Atkinson CT3

1Hawai‘i Cooperative Studies Unit, University of Hawaii at Hilo, HI, USA

2U.S. Geological Survey, Pacific Island Ecosystems Research Center, HI, USA

3Agronomy Department, Iowa State University, Ames, IA, USA

Corresponding Author:
Raji JA
Hawai‘i Cooperative Studies Unit, University of Hawaii at Hilo, 200 W. Kawili St. Hilo, HI, USA
Tel: +8089330759
E-mail: [email protected]

Received Date: November 16, 2016; Accepted Date: December 13, 2016; Published Date: December 19, 2016

Citation: Raji JA, Atkinson CT (2016) Genetic Variation and Structure in Remnant Population of Critically Endangered Melicope zahlbruckneri. J Biodivers Endanger Species 4:175. doi: 10.4172/2332-2543.1000175

Copyright: © 2016 Raji JA, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at Journal of Biodiversity & Endangered Species


The distribution and amount of genetic variation within and between populations of plant species are important for their adaptability to future habitat changes and also critical for their restoration and overall management. This study was initiated to assess the genetic status of the remnant population of Melicope zahlbruckneri–a critically endangered species in Hawaii, and determine the extent of genetic variation and diversity in order to propose valuable conservation approaches. Estimated genetic structure of individuals based on molecular marker allele frequencies identified genetic groups with low overall differentiation but identified the most genetically diverse individuals within the population. Analysis of Amplified Fragment Length Polymorphic (AFLP) marker loci in the population based on Bayesian model and multivariate statistics classified the population into four subgroups. We inferred a mixed species population structure based on Bayesian clustering and frequency of unique alleles. The percentage of Polymorphic Fragment (PPF) ranged from 18.8 to 64.6% for all marker loci with an average of 54.9% within the population. Inclusion of all surviving M. zahlbruckneri trees in future restorative planting at new sites are suggested, and approaches for longer term maintenance of genetic variability are discussed. To our knowledge, this study represents the first report of molecular genetic analysis of the remaining population of M. zahlbruckneri and also illustrates the importance of genetic variability for conservation of a small endangered population.


Endangered species; Genetic conservation; Melicope zahlbruckneri ; Molecular markers


Rare and endangered species are generally vulnerable to environmental changes and genetic risks due to their often fragmented and small population sizes [1]. They are more likely to have limited fitness and fixation of recessive mutations that are deleterious which can increase the risk of genetic variability loss through random genetic drift than larger and more diverse populations [2]. Preservation of natural genetic diversity levels and structure is thus critical for the long term management of these endangered and threatened species to prevent further loss of diversity that could arise from unexpected fluctuations in their habitat and the environment [3].

Melicope zahlbruckneri is endemic to the Island of Hawai’i with a small and declining population. It belongs to the Rutaceae family and listed as federally endangered. This species currently occurs as a single population at the Hawaii Volcanoes National Park with less than 30 individuals left in the wild. While there is little published information on the reproductive ecology of M. zahlbruckneri , it is believed to be outcrossing, with pollination and seed dispersal by birds. Species in this family generally have flowers that divide into four or five parts. In terms of size, individual plants range from herbs to shrubs and small trees. M. zahlbruckneri grows up to about 10-12 m tall with mature leaves that are about 6-24 cm long and 4-12.5 cm wide, with welldefined lateral veins [4]. There is currently no information available on the genetics of the species, also, very little is known on their historic distribution. Thus, it is important to determine the current level of genetic variability and structure in the remnant population to aid development of sampling strategies for population restoration and propose informed recommendations to assist their ongoing management.

Molecular markers have proved to be efficient means of investigating population structure, genetic diversity, and are important components in genetic conservation studies of different species [5,6]. Molecular markers are able to provide a precise understanding of genetic diversity through the identification of genomic segments that differentiate individuals or population without the need for genetic information about the genome [7,8]. Neutral markers in combination with geographic information and appropriate statistical tools have been used frequently for the study of genetic diversity in a wide range of plant species [9-12] and have been especially useful for studies involving intra and inter-species genotypic variation in both large and small populations [13].

Several molecular markers are available with the choice often determined by availability of information for the species, the population type and size, and available resources. Microsatellites and Single Nucleotide Polymorphic (SNP) markers are recognized as the most efficient at revealing genetic diversity within and between species and with the ability to distinguish homozygous and heterozygous individuals when pedigree information is desired.

However, the use of these co-dominant markers requires the availability of genomic sequence information about the species as well as high cost consideration for the initial marker development particularly in studies involving natural populations where several species families are involved. Amplified Fragment Length Polymorphism (AFLP) on the other hand has the potential to screen a large number of genetic loci and does not require prior information about the genome of the species under investigation. Here we report an AFLP-based molecular assessment of M. zahlbruckneri to evaluate the genetic structure of the remnant population actively managed at the Hawai’i Volcanoes National Park.

For each individual, a small sized healthy leaf was chosen and stored on ice in a sealed plastic bag. Leaf samples were taken to the laboratory for immediate processing or stored at -70ºC until DNA extraction. A total of 28 samples were collected from the Kipukapuaulu (Table 1), the only remaining site where the species is known to be extant [except for a previously reported individual tree back in 1997 at Laupahoehoe [14].

Selective Sequence E/M Pairs Total No of Loci Polymorphic Loci Polymorphic Fragments (%) Size Range (Base pair)
E-AAC/M-CTC 130 84 64.6 50-480
E-ACT/M-CTT 189 114 60.3 50-500
E-AAG/M-CAC 145 85 58.6 45-450
E-ACA/M-CAC 172 96 55.8 50-500
E-ACC/M-CAC 136 75 55.1 65-480
E-ACC/M-CAT 130 80 61.5 55-500
E-AGC/M-CTA 142 81 57.0 50-480
E-AGC/M-CAC 106 66 62.3 50-500
E-AGG/M-CTA 132 80 60.6 60-500
E-AGG/M-CAG 135 73 54.0 50-500
E-ACT/M-CAG 134 72 53.7 50-490
E-ACG/M-CTC 167 95 56.8 50-475
E-ACA/M-CTG 155 87 56.1 55-480
E-AAC/M-CTT 102 59 57.8 50-500
E-ACG/M-CTT 103 39 37.9 40-500
E-ACT/M-CTC 150 85 56.6 50-500
E-AAG/M-CTC 137 78 56.9 45-500
E-AAC/M-CAG 140 67 47.9 40-500
E-ACT/M-CTG 185 110 59.4 50-500
E-ACA/M-CTC 157 92 58.5 60-500
E-ACC/M-CAG 127 65 51.8 64-500
E-AGG/M-CTT 101 40 39.6 50-500
E-ACC/M-CTG 100 62 62.0 45-490
E-AAG/M-CTT 85 16 18.8 40-500
E-AGG/M-CTG 97 61 62.9 55-480
E-ACA/M-CTT 96 25  26.0 50-500
E-AAC/M-CAT 145 81 55.0 40-500
E-AAC/M-CTA 135 77 57.0 40-500
E-ACT/M-CAT 168 101 60.2 40-490
E-ACA/M-CTA 167 102 61.0 50-500
E-ACA/M-CAT 150 92 61.3 50-500
E-AAC/M-CAC 120 64 53.3 50-500
E-ACT/M-CTA 165 102 61.8 55-500
E-ACG/M-CAT 156 93 59.6 50-500
E-ACG/M-CAG 163 100 61.3 55-500
E-AAG/M-CTA 145 81 55.9 55-500
Mean 137.97 77.2 54.9 -

Table 1: AFLP marker pairs, characteristics and average variation in selective amplification of M. zahlbruckneri.

DNA was extracted using a modified plant DNA isolation protocol by Dellaporta et al. with minor modifications [15]. The concentration and purity of DNA was measured using the nanodrop ND-1000 spectrophotometer and quality was assessed on agarose gel.

SSR marker transferability test

Simple Sequence Repeats (SSRs) have had limited use in studies involving native plant populations in Hawaii because they have to be developed specifically for each plant species. However, because several studies have demonstrated the successful transferability of gene based microsatellites across related genera in crop plants [11,16,17], we decided to test available genomic or gene based microsatellites that have been developed for other taxa close to M. zahlbruckneri to explore the prospects of transferability of such markers to genotype M. zahlbruckneri population.

Primer sequences of seven compound microsatellite loci developed for Melicope quadrilocularis [18] were synthesized (Integrated DNA Technologies, Inc. Corvallis, IA) for testing. We optimized PCR reaction components and cycling parameters for each primer pair and tested them for successful amplification and polymorphism detection in Melicope zahlbruckneri population. PCR reactions for SSR marker assay were performed in a Biorad DNA Engine (Peltier Thermal Cycler); PCR products were first separated on 1.5% TAE agarose gels and visualized under UV after ethidium bromide staining and then analyzed by capillary electrophoresis using an automated Genetic Analyzer ABI 3730.

AFLP restriction digestion and amplification

The AFLP procedure followed the original method described by Vos et al. with a few modifications, and unless otherwise indicated, Life Technologies (Invitrogen and Applied Biosystems) reagents were used in all reaction procedures [19]. Total genomic DNA (500 ng) was digested with 4.0 units of EcoRI and MseI restriction enzymes at 37ºC for 4 h in a 25 μl reaction volume that included 2.5 μl of 5X reaction buffer and 0.2 μl of 10 mg/ml bovine serum albumin (BSA). After digestion, a 5 μl aliquot of digested DNA from each sample was run on TAE 1% agarose gel to check for complete and uniform digestion. Ligation reaction was performed in a 40 μl solution and contained 20 μl of the digested DNA, 5 pmol of the EcoRI adapter, 50 pmol of the MseI adapter, 1 units T4 DNA ligase, 0.10 μl of 10 mg/ml BSA (1 μg), and 4 μL DNA Ligase buffer. The reaction was left to incubate for 3 h at 20ºC.

Materials and Methods

Sample collection and DNA isolation

M. zahlbruckneri population was sampled by collecting leaf tissue from all known and accessible individual trees of the species.

Pre amplification reaction was performed in a 30 μl volume using 1 μl of the ligation reaction as the template, and combined with 21.5 ng each of the EcoRI+A and MseI+A primers, 3.0 μl Amplitaq 360 DNA buffer (500 mMKCl), 15 mM MgCl2, 10 mM dNTPs (each at 2.5 mM), and 1 unit of Taq DNA polymerase. The reaction was processed in a thermal cycler using the following cycling parameters: an initial denaturation step at 94ºC for 60 s, 20 cycles of 94ºC/30 s, 56ºC/60 s, 72ºC/60 s and a final extension at 72ºC for 10 min. Selective amplification was performed using 36 EcoRI/MseI primer combinations to genotype the population. The primer pairs were chosen based on the number, clarity, and reproducibility of fragments produced in a preliminary screening of 60 primer combinations. Selective amplification was performed in a 10 μl reaction volume containing 2.5 μl of diluted pre amplification reaction (diluted 1:20 in 1× Tris–ethylenediaminetetraacetic acid buffer), 5 ng of EcoRI+3 and 15 ng of MseI+3 selective primers, 1 μl Amplitaq 360 buffer, 0.24 μl dNTPs (10 mM), and 0.2 units of Amplitaq 360 DNA polymerase. All EcoRI+3 selective primers were fluorescently labeled with FAM, NED and JOE using the DS-30 Matrix standard (Applied Biosystems, Foster City, CA). Amplification was performed with a touch down cycling procedure as follows: An initial denaturation step at 94°C for 2 min, 1 cycle of 30 s at 94ºC, 30 s at 65ºC, and 60 s at 72°C, followed by 11 cycles in which the annealing temperature decreases 0.7ºC per cycle, and 22 cycles of 30 s at 94ºC, 30 s at 56ºC and 60 s at 72ºC, and a final extension step at 72ºC for 5 min.

Sample genotyping and pre-analysis of data

AFLP fragment analysis of all individuals was performed by capillary electrophoresis using automated Genetic Analyzer ABI 3730. ROX labeled size standard that range in size from 35-500 bp was used for fragment sizing. AFLP fragment data were normalized based on local southern size-call algorithm, peak saturation, baseline saturation, pull-up correction, and spike removal corrections. Following normalization, allele sizing and call procedures were performed using the GENEMAPPER (ver. 4.1, Applied Biosystems) software program. Individual peaks were called on the basis of the total signal intensity and a peak was scored only if its intensity exceeded a fixed threshold. Data from a standardized template was used for each primer combination to ensure that peak sizing and position was precise and reproducible in all samples.

Data analysis

Marker fragments were scored as either present or absent to create a binary matrix for each individual. The resulting binary matrix was used as input data to generate Jaccard’s similarity coefficient and then transformed into a dissimilarity measure as (1–Jaccard’s similarity) using the SIMQUAL module of NTSYSpcver. 2.20d program [20-21]. Cluster analysis was performed using the unweight pair group method with arithmetic averages (UPGMA). The resulting tree branching pattern was evaluated by bootstrapping based on 1000 replicates (PHYLIP software package v.3.66) [22]. Genetic variation was further assessed by using the TREECON program to generate Neighbor joining cladograms. Additionally, genetic diversity and allele frequency measures for the population were calculated using the AFLP-SURV program [23]. Hardy-Weinberg equilibrium (FIS=0) was assumed for the estimation of allele frequencies as an essential requirement for dominant markers where the absent fragment frequency is q2 (q being the null allele frequency). Furthermore, Nei’s unbiased expected heterozygosis was used to calculate genetic diversity within the population as the percentage of within-population polymorphic loci relative to overall polymorphic loci. This is based on the assumption that the presence of a fragment indicates either a heterozygote or homozygote for the dominant allele with frequencies estimated based on the null allele frequency. To further examine the genetic structure of M. zahlbruckneri population, a model based Bayesian analysis was performed using the program STRUCTURE ver. 2.4 [24-26]. A nonuniform prior distribution of allelic frequencies was assumed with its parameters derived from the observed distribution of fragment frequencies among AFLP loci [27]. Log likelihood and estimates for allele frequency divergence as a result of random drift (FST) were computed using the admixture model and correlated allele frequencies options with a burn-in period of 80,000 and 100,000 Markov Chain Monte Carlo (MCMC) iterations after burn in. Likely genetic clusters (K) groups were set over a range from K=1-5 with 10 replicate runs at each K. The posterior probability P (X|K) was estimated to give an indication of the true likely number of groups without prior information about the population. Results from STRUCTURE were further analyzed and collated with the structure harvester program [28]. The CLUMPP program was used for permutations of the most likely observations among replicate runs for each K, and results were visualized in the DISTRUCT program [29,30].

Principal Coordinate Analysis (Pcoa) was used to explore genetic relatedness of individuals based on the modified Roger’s distance matrices generated from the binary data populations. PCoA computations were performed with the MVSP (Ver.3.21) program. A plot of the population was produced to show distances among samples in the plot as likely illustration of genetic distances within the population. The Mantel statistics was used to assess correlation between observed spatial and genetic distances among individuals.


Microsatellites marker analysis

Twenty-eight plant samples from the M. zahlbruckneri population were genotyped using the seven compound microsatellite primers developed for Melicope quadrilocularis [18]. Five of the seven SSR’s amplified expected allele sizes for each primer pair as described for M. quadrilocularis . Although transferability of the SSRs was confirmed in M. zahlbrukneri , the markers were unable to detect polymorphism in M. zahlbruckneri samples. Consequently, we decided to focus the rest of the study on the AFLP marker analysis results.

Genetic distance and population structure

Thirty-six AFLP EcoRI/MseI combination primer pairs generated 2780 fragments (Table 2). The percentage of polymorphic fragments per AFLP locus ranged from 18.8% to 64.6% with an average of 54.9% (Table 2). Of the total fragments scored, 620 unique fragments (22.3%) were detected in a few samples within the population. Coefficients of similarity for all possible pair-wise comparisons ranged from 0.621-1.171.

Cluster Average distance* Mean Fst value
1 0.3762 0.0117 (p>0.05)
2 0.3661 0.0150 (p>0.05)
3 0.2271 0.3471 (p<0.01)
4 0.2317 0.9157 (p>0.01)
Parameter Value (clusters 1-4)
Estimated Ln Prob of Data (Ln P(D)) -33461.6
Mean ln likelihood -31309.1
Variance of ln likelihood 4305
Mean value of alpha 0.1453
*Average distance (between individuals within a cluster).
Ln Likelihood: log likelihood of the data given values of P (estimated allele frequencies) and Q (the estimated membership coefficient for each individual in a cluster).
Est. Ln P (D): current estimate of ln (PX|K)), averaging over all iterations after burn in period.

Table 2: Average distances between individuals and estimated mean value of Fst for each cluster of M. zahlbruckneri.

Hierarchical trees with bootstrap values revealed low to moderate sub structuring in the M. zahlbruckneri population. Main clusters that are subdivided into smaller groups are shown as well as 2 individual entries located on separate branches away from the rest of the groups (Figure 1). Neighbour joining tree derived from Jaccard’s coefficient appeared to be less differentiated than expected considering the percentage of unique alleles detected in a small population. Many of the samples that accounted for the private alleles found in the M. zahlbruckneri population showed genetic profiles that varied substantially from the rest of the population. MZ23 and MZ27 in particular were noticeably the most diverged from the rest of the population (Figure 2). Cluster analysis of the population was reconstructed without MZ23 and MZ27 to examine possible outlier effect but no significant effect was detected in the overall topology of the trees.


Figure 1: Hierarchical cluster analysis of M. zahlbruckneri using UPGMA (unweighted pair group method with arithmetic mean). Bootstrap values (based on 1000 permutations) are indicated at each tree node.


Figure 2: Unrooted neighbour joining tree illustrating genetic relationship of 28 M. zahlbruckneri individuals based on genetic similarities derived from AFLP marker.

Bayesian analysis using the STRUCTURE program partitioned the observed population genetic variation into four clusters (K) that are identical to those obtained from PCoA and unrooted NJ trees (Figures 2 and 3). A correlated allele frequency with admixture model was indicated for the analysis. The optimal convergence of the MCMC algorithm was achieved by using a burn-in period of 80,000 steps, followed by 100,000 steps of data collection and 10 replicated runs to ensure efficiency in clustering computation and accurate estimates of the posterior distribution. The optimal sample resolution in each cluster was obtained at K=4. The graphical output showing the distribution and proportion of shared AFLP genomic segments is presented in Figure 4. The plot showed the assignment of individuals into different cluster groups denoted by vertical bars with different colours.


Figure 3: Principal coordinate analysis of M. zahlbruckneri population showing estimated genetic relationship among individuals based on AFLP loci.

Two individuals (MZ23 and MZ27) were significantly diverged from the rest of the population. Similarly, eight individuals (MZ3, MZ13, MZ14, MZ16, MZ24, MZ25, MZ26 and MZ28) had significant numbers of rare alleles but also shared a significant number of alleles with the rest of the population. These 8 individuals are suspected to have mixed ancestry based on the combination of unique allelic profiles leaf characteristics that seemed unusual in shape and outside of the expected size range for M. zahlbruckneri (McDaniels S., Pers. Comm). They also showed substantial degree of genetic similarities to both MZ23 and M27.

The FST values of allele distribution provided additional information on allele frequency divergence estimate from possible ancestral frequencies as a result of random drift for each cluster groups. The FST values for the rarest alleles represented in the M. zahlbruckneri population ranged from 0.87-0.96 and were restricted to only a few individuals within the population. The FST values also showed the highest distance between clusters in the group comprising MZ23 and MZ27 (FST 0.92, P<0.01) and cluster 1 the lowest at FST 0.012, P>0.05. The average distance between individuals within each clusters varied among the 4 clusters. The highest distance score was found within individuals of cluster 1 (0.3762) and the lowest with the members of cluster 3 (0.2271). The mean value of log likelihood for each cluster (K) is -31309, the mean value of alpha is 0.1453 and the estimated value of Ln probability of data is -33461.

Genetic relatedness and population variation

Principal coordinates analysis scores from the first three coordinates in the M. zahlbruckneri population showed that eigenvectors accounted for 65.82% cumulative variation of the total variance observed, with the first coordinate accounting for the most variation at 29.9%. The plot of genetic distances estimated from the coordinates showed genetic grouping and relationships between individual samples that is reflective of a high degree of common alleles in more than 60% of the individuals. It also highlights the effect of unique alleles on grouping along the PCO2 axis (Figure 4).


Figure 4: STRUCTURE analysis of 28 M. zahlbruckneri individuals with K=4 clusters (blue, green, yellow, purple). The proportions of membership into the various clusters are shown on the y-axis shows. Individuals are represented by colored vertical bars.

Two individuals, MZ 23 and MZ 27 varied significantly and formed a cluster away from the rest of the population. This result is consistent with the pattern of variation observed from Bayesian analysis, NJ cladogram and UPGMA trees constructed for the population. In addition, Mantel statistics with 1000 random permutations did not indicate a significant evidence of correlation between genetic differentiation (FST) and geographical distances among individuals within the location (R2=0.0333, P=0.708) (Figure 5).


Figure 5: Mantel correlation of M. zahlbruckneri population computed from matrices of within population pairwise genetic differentiation (Nei and Li (1979)) and the natural logarithm of pairwise geographic distances.


The number of polymorphic AFLP loci analyzed in this study provided reliable estimates of the overall genetic diversity, and a reasonable representation of the M. zahlbruckneri genetic structure despite the small population size. The higher the number of marker loci assessed in a population genetic study, the better the accuracy of results [31,32]. Studies based on a few molecular marker loci have been reported to have a tendency to be biased [33]. In order to compensate for the dominant nature of AFLP markers, the presence and absence binary data were converted to expected heterozygosis through the assumption of Hardy-Weinberg equilibrium. This approach allowed us to obtain estimates that are almost as informative as those generated from co dominant markers. Furthermore, the criteria described by Lynch and Milligan were applied to the data to attain an unbiased estimate of heterozygosity.

Analyses of AFLP data revealed moderate levels of genetic variation among individuals in our study population. Identical patterns of genetic clustering of M. zahlbruckneri were obtained from both the UPGMA and NJ cladograms. Despite low bootstrap values at some of the UPGMA tree nodes, there was consistency in the pattern of clustering as shown by other methods used in this study with several individuals assigned to the same clusters with only minor overlap of samples between clusters. Model based Bayesian analysis provided additional information about the genetic structure of the population. Bayesian assignment and patterns of genetic variation suggested the presence of a moderately varied gene pool and the possibility of hybridization with closely related Melicope species based on the proportion of shared versus unique alleles across individuals in different cluster groups. In addition, the significant average FST values shown by two out of four clusters in the population further increased the likelihood of some external gene flow and the presence of individuals from other unidentified closely related Melicope species. The observed pattern was quite unexpected given the small population size and endangered status of the population. A plausible explanation for this observation is the high percentage of unique alleles detected in the eight individuals that showed the possibility of having mixed ancestry and a few others (particularly, MZ 23 and MZ 27) that may possibly belong to another species. It is noteworthy to mention that samples that showed the presence of unique alleles also had distinctly different morphological characteristics compared to the rest of the population. Additional genetic analysis is recommended to clarify the identity of these samples to determine their pedigree and species assignment.

The extent of overall genetic diversity detected in the population could also be attributed to historic gene flow that may have occurred with other closely related taxa when populations were more abundant and continuous, or it might be a result of more recent gene flow events with other Melicope species. It remains unknown how wide spread the observed unique alleles are with respect to other existing Melicope species. Geographic information showed that M. radiata is the closest population to M. zahlbruckneri in terms of physical distance. Although the present work did not examine M. radiata by molecular methods, there are strong indications based on morphological information that some level of gene flow may be occurring between these two species. It would be informative in future studies to obtain the nucleotide sequence of these unique alleles and assess their fixation in other populations.

The results presented here demonstrate that AFLP analyses in combination with appropriate statistics can detect genetic variability and provide genetic information that are valuable for conservation purposes in natural populations. Similar observations of relatively high to moderate genetic diversity in small populations have been reported in other tropical species e.g. Swietenia humilis [34] and Eucalyptus albens [35] and endangered populations [36,37]. A number of the authors attributed the observed level of genetic diversity to the outcrossing mating system of the species, which enabled pollination across isolated and small populations to influence gene flow. This is mostly characteristic of tree species with long distance seed and pollen dispersal [38]. Also, long term adaptation of small remnant populations to local conditions is a factor that can influence genetic diversity. There are clearly some limitations with using dominant molecular markers for genetic analysis compared with co-dominant markers; for example, their inability to provide locus information and distinguish between hybrids is apparent drawbacks [39]. Nonetheless, their ease of application across taxa in natural populations, ability to detect small genetic differences and genome-wide coverage has made them suitable and highly recommended as the markers of choice in circumstances where resources are limited to develop co dominant markers and where the study objective is to obtain initial information about overall genetic variation of a population. Moreover, their compatibility with the model based Bayesian approach for inferring genetic structure without prior information about the population has demonstrated that unbiased estimates of heterozygosity and genetic differentiation can be obtained.

Given the amount of variability detected in the M. zahlbruckneri population, it may be safe to reason that the genetic risk for this species is relatively moderate at present, but considering the small population size of the single remaining population, additional studies may be needed to examine the sustainability of the observed genetic diversity regarding longer term ecological impacts on the population distribution and habitat preservation. Furthermore, it is essential to understand that using only a few samples as founding individuals can considerably impact allelic diversity and result in higher genetic drift in the long run, therefore, representation of all identified M. zahlbruckneri individuals, including their putative hybrids in restorative planting at new sites would be beneficial to restoration success. Also, MZ23 and MZ27 may be excluded pending confirmation of their species identity.

Future studies to determine the level of available genetic variability across the species range would be desirable. This could be done by including other related taxa in the family Rutaceae for a comparative genetic analysis. Additionally, developing co-dominant marker systems such as Single Nucleotide Polymorphic (SNP) markers via the application of genotyping by sequencing technique as well as microsatellite markers may be useful for finer scale genetic studies of the endangered population to obtain phylogenetic information and comparative analysis of genomes between Melicope sp. and other closely related taxa. In addition, it would be beneficial to know genome sizes and ploidy levels of these species for a better understanding of the structure, organization and evolution of the species’ genome. This information will better assist resource managers in planning conservation of this critically endangered species.


The results presented in this study illustrates the current genetic variation status of federally endangered M. zahlbruckneri and provides useful baseline information based on genetic and geographic population evaluation that can assist ongoing conservation and management priorities for the only known remaining population.


This work was supported in part by the U.S. Geological Survey Natural Resource Preservation (NRPP) and Wildlife and Terrestrial Resources Programs. The authors are grateful to Linda Pratt, Josh Van Demark, and Sierra McDaniel for assistance with field sample collections and for providing location information; and USGS Pacific Island Ecosystems Research Center for additional funds to support completion of the project. Any use of trade, product, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government.


Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Relevant Topics

Article Usage

  • Total views: 1437
  • [From(publication date):
    December-2016 - Nov 13, 2019]
  • Breakdown by view type
  • HTML page views : 1318
  • PDF downloads : 119