Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Genetic analysis of African lions (Panthera leo) in Zambia support movement across anthropogenic and geographical barriers

  • Caitlin J. Curry,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Interdisciplinary Program of Genetics, Department of Veterinary Pathobiology, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, Texas, United States of America

  • Paula A. White,

    Roles Conceptualization, Funding acquisition, Resources, Writing – review & editing

    Affiliation Center for Tropical Research, Institute of the Environment and Sustainability, University of California, Los Angeles, Los Angeles, California, United States of America

  • James N. Derr

    Roles Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing

    jderr@cvm.tamu.edu

    Affiliation Interdisciplinary Program of Genetics, Department of Veterinary Pathobiology, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, Texas, United States of America

Abstract

The Luangwa Valley in eastern Zambia is a transverse offshoot of the Great Rift Valley system. This region appears to have an isolating effect as evidenced by suspected endemic subspecies, such as the Cookson’s wildebeest and Thornicroft’s giraffe. Recent mitochondrial DNA studies demonstrated that African lions in Zambia consist of two highly diverse eastern and western sub-populations. Herein, we report nuclear and mitochondrial DNA results from 409 lions that support this population substructure across Zambia but proposes only partial isolation of the Luangwa Valley with more movement between the populations than previously thought. Population assignment analysis identifies two populations with little evidence of admixture assigning lions to either the eastern or western sub-populations. A high occurrence of private alleles and clear evidence for a Wahlund effect further justify the presence of a highly structured population. But, while mitochondrial DNA analysis still shows little to no matrilineal gene flow (FST = 0.53) between sub-populations, microsatellite analysis suggests there is gene flow (FST = 0.04) with low but significant isolation-by-distance and an average of 6 migrants per generation. Evidence of isolation-by-distance is also found in factorial correspondence analysis with the Lower Zambezi National Park and eastern corridor clusters overlapping isolated clusters of the Luangwa Valley and western sub-population. From this evidence, the Luangwa Valley appears separated from the western sub-population with some dispersal through the southern regions of the eastern sub-population. Both the eastern and western sub-populations have high heterozygosity (0.68 and 0.69, respectively) and genetic diversity (0.47 and 0.50, respectively) values, indicative of genetically healthy populations.

Introduction

Zambia has one of the largest wild lion populations with a current estimate of around 1,200 individuals [1] within a range of more than 200,000 km2 [2]. While lions do exist in small numbers in outlying areas of Zambia (e.g., Liuwa Plains National Park, Mweru Wantipa National Park), lion distribution can generally be divided into two regions (eastern and western) that are separated by geographic and anthropogenic features.

Zambia is one of nine countries with over 1,000 lions and has 2 of the 10 lion strongholds [3]. Kafue National Park (NP) and the adjoining Game Management Areas (GMAs) in the western part of the country collectively form the Greater Kafue Ecosystem (GKE). The GKE is designated a potential stronghold due to heavy poaching of the prey base [3], a significant concern as a decrease in prey lowers carnivore carrying capacity [4].

The Luangwa Valley Ecosystem (LVE) is a lion stronghold [3]. It is an offshoot of the Great Rift Valley system along the Luangwa River consisting of three NPs, the North Luangwa NP, South Luangwa NP, and Luambe NP, and their surrounding GMAs. The presence of suspected endemic subspecies has been used as evidence of the Luangwa Valley’s geographic isolation. This includes Thornicroft’s giraffe (Giraffa camelopardalis thornicrofti) found to be genetically distinct by mtDNA analyses [5]. However, more recent genetic studies using nuclear loci no longer consider them as their own subspecies and, although geographically separated by 500-km, now group them with a neighboring population of the Masai giraffe (Giraffa camelopardalis tippelskirchi) subspecies [6]. Other presumed endemic subspecies include the Cookson’s wildebeest (Connochaetes taurinus cooksoni) [7], a subspecies of Blue wildebeest and the Crawshay’s zebra (Equus quagga crawshayi), which has a striping pattern unique to Zambia [8].

Zambia is one of a handful of countries that actively engages in wild lion monitoring, conservation, and management [9]. Lions are a protected species under the Zambia Wildlife Act No. 12 of 1998 making it illegal to hunt, kill, capture or be in possession of a lion specimen without a permit [9]. Trophy hunting of lion is legally conducted in many of the GMAs. However, in Zambia as in many other countries, human-lion conflict is the greatest threat to lions outside of protected areas (PAs) [9]. Increased human activity on the edges of NPs and into GMAs severely limits the movement of carnivores [10,11]. Therefore, it is relevant to consider population connectivity when formulating lion conservation strategies.

Much of the area between the LVE and the GKE is comprised of an anthropogenic patchwork of towns and farms and is, therefore, considered uninhabitable by lions [3]. This is supported by mitochondrial DNA (mtDNA) which shows little to no matrilineal gene flow between the eastern and western sub-populations [12]. However, studies using minimal sampling from Zambia found an admixture pattern suggestive of male-mediated gene flow [13]. A larger study including both nuclear and mtDNA data was needed to capture a more accurate representation of lion movement and diversity in the region. This study used nuclear and mitochondrial markers from 409 lions across Zambia (Fig 1) to assess population structure and the potential for movement of lions between the Luangwa Valley and the Greater Kafue ecosystems.

thumbnail
Fig 1. Map of Zambia with sampling locations.

Sample locations are latitude and longitude coordinates of GPS location at sampling (181 lions at 181 locations) or a central location based on sampling area (228 lions at 33 locations). Detailed location information is in S1 Appendix.

https://doi.org/10.1371/journal.pone.0217179.g001

Materials and methods

Sample collection

The Zambia Lion Project (ZLP) provided 446 samples for analysis. Samples are in the form of hair, skin, bone and/or tissue and were collected during research conducted by ZLP in partnership with the Zambia Wildlife Authority (Research/Employment Permit No. #008872). For more details on sample collection, refer to Curry et al. [12].

Samples were collected between 2004–2012 in five of Zambia’s National Parks (North Luangwa, South Luangwa, Lower Zambezi, Kafue and Sioma Ngwezi) and more than forty GMAs, making this dataset representative of Zambia’s countrywide lion population. Each sample has an accompanying latitude and longitude sampling location. When exact sampling location could not be provided, a central latitude/longitude was given based on the sampling locale (see S1 Appendix).

DNA extraction

DNA was successfully extracted from 421 samples of which 409 were further analyzed for mtDNA and microsatellites. Of the 12 samples eliminated, ten were duplicate individuals and two were not lions. The 409 samples analyzed consist of 324 males, 83 females, and 2 unknown (S1 Appendix), including all individuals previously studied in Curry et al. [12].

DNA was extracted using standard protocols and procedures used in the DNA Technologies Core Laboratory at Texas A&M University in College Station, TX (http://vetmed.tamu.edu/dnacore). PCR amplification was done using the KAPA Biosystems KAPA2G Robust HotStart PCR Kit with protocols described in Curry et al. [12] for mtDNA and Curry & Derr [14] for microsatellites. Cycling profiles for mtDNA and microsatellite amplification can be found in S2 Appendix. PCR product was sequenced and/or genotyped on an Applied Biosystems 3130xl or 3730 Genetic Analyzer.

Mitochondrial DNA and sequencing

Samples were amplified for a continuous region from the mtDNA 12S-16S genes using primers designed by Antunes et al.[15]. These primers were designed to prevent amplification of a 12.5 kb integration of mtDNA into the nuclear genome, or numt [16]. Sequences were manually edited and assigned a haplotype using SEQUENCHER v4.8 [17].

Diversity calculations and phylogenetic analyses were carried out as in Curry et al. [12] so a direct comparison could be made with the larger sample size. The data was analyzed using Arlequin v3.5 [18] as a full Zambian population and separated into eastern and western sub-populations. Number of polymorphic sites, gene diversity, nucleotide diversity and haplotype diversity were calculated for each. The sub-populations dataset was used for intra-population calculations of the coefficient of differentiation (FST) and hierarchical analyses of molecular variance (AMOVA). Finally, pairwise differences were calculated between sampling locations.

Phylogenetic analysis included all haplotypes from Antunes et al. [15] (GENBANK Accession #s FJ151641-FJ151652) and Zambian haplotypes found in Curry et al. [12] (GENBANK Accession #s KT164799-KT164803) and this study (ENA Accession #: LR593884). Phylogenetic analysis was performed using maximum likelihood analysis in PhyML 3.0 [19] and Bayesian inference methods with Mr. Bayes [20,21]. For Bayesian analysis, samples were drawn every 1,000 steps over 50,000,000 MCMC steps with the first 10% discarded as burn-in and the tree was rooted with the tiger (Panthera tigris; GENBANK Accession #KJ508413) that was aligned and trimmed to the lion sequences. A haplotype network was also made using the median-joining option of Network v4.6.1.3 [22].

Microsatellites and genotyping

A panel of 14 miniSTRs [14] (Leo006, Leo008, Leo031, Leo045, Leo077, Leo085, Leo098, Leo126, Leo224, Leo230, Leo247, Leo281, Leo391, and Leo506) were used for microsatellite analysis. Each sample was amplified, genotyped, and scored a minimum of two times at each locus to determine a consensus genotype. Samples were scored using STRand [23]. Samples scored at 10 or more loci were retained for further analysis.

Molecular diversity indices of expected and observed heterozygosity, deviation from Hardy-Weinberg equilibrium (HWE), conventional F-statistics, percentages of molecular variation, and number of migrants were calculated using Arlequin [18], GenAlEx [24], and GenePop [25]. Isolation-by-Distance (IBD) was calculated for females, males, and all individuals using the Mantel Test in GenAlEx [24]. Factorial Correspondence Analysis (FCA) and Principle Coordinate Analysis (PCoA) were carried out in GeneTix [26] and GenAlEx [24], respectively.

Number of alleles, number of private alleles and distribution of alleles were determined using GenAlEx [24] and allelic richness and private allelic richness was calculated using rarefaction in HPRare [27,28].

Effective Population Size was calculated using NeEstimator [29] using the linkage disequilibrium (LD) model. The two sub-populations have a high number of private alleles, many of which occur in low frequencies (<0.01). Because low allele frequencies can bias calculations of effective population size [30], the Pcrit was set to frequencies of 0.00, 0.01, and to exclude only singleton alleles.

STRUCTURE [31] was used to evaluate population structure. Fifteen runs were performed for K = 1 to 6 using 1,000,000 MCMC reps after 100,000 burn-in. Structure Harvester [32] was used to implement the Evanno method [33] to determine ΔK. CLUMPP [34] was then used to combine replicate runs and results were visualized using Distruct [35].

Results

A total of 409 individuals were analyzed with 398 genotype panels and 391 sequences produced. Eighteen of the 398 genotyped individuals did not produce a haplotype. This is likely due to sample degradation. Degraded samples may not allow for amplification of large amplicons [36], such as this 1800-bp sequence of 12S-16S. Eleven individuals produced a haplotype but were not genotyped because DNA was lost in an accident after sequencing. These individuals didn’t have more sample available for re-extraction of DNA.

Mitochondrial diversity

A total of 391 samples produced full sequences (Table 1 and S1 Appendix). Distribution of haplotypes is shown in Table 1. A novel haplotype, Z6, was found in three individuals in the eastern sub-population. Even with the addition of another haplotype [12], these results support the findings of Curry et al. Haplotype Z6 differs from haplotype H11 by one single nucleotide polymorphism (SNP) at position 1524 (Table 2).

thumbnail
Table 1. Distribution of haplotypes.

Number of males (♂), females (♀), and with unknown gender (?) for each haplotype is indicated for all areas sampled in Zambia along with the haplotype frequencies. Haplotypes H1, H9 and H11 were previously described by Antunes et al. [15]. Haplotypes Z1-Z6 are novel.

https://doi.org/10.1371/journal.pone.0217179.t001

thumbnail
Table 2. Nucleotide position for each polymorphic site.

All nucleotide polymorphisms in 12S-16S are shown for haplotype H1. For all other haplotypes, only nucleotides that differ from H1 are shown. Haplotypes found in Zambia are in bold.

https://doi.org/10.1371/journal.pone.0217179.t002

AMOVA analysis, run as an eastern and western sub-population, resulted in an FST of 0.53 (p-value <0.001; Table 3). Gene diversity was calculated to be 0.7237 +/- 0.0112 for one Zambian population then 0.4712 +/- 0.0226 and 0.5041 +/- 0.0382 for the eastern and western sub-populations, respectively (S9 Appendix).

thumbnail
Table 3. AMOVA results with FST.

Percent variation is given among populations (Va) and within groups (Vb). The significance of differentiation within and among populations was tested by 1,000 permutations.

https://doi.org/10.1371/journal.pone.0217179.t003

All trees resulted in similar clustering. The unrooted maximum likelihood tree is shown (Fig 2). The general configuration of the phylogenetic tree and haplotype network did not change from Curry et al. [12], even with the addition of another novel haplotype (S3 Appendix). Haplotype Z6 appears as an additional branch to the H11 cluster, the second most predominant haplotype in the eastern sub-population.

thumbnail
Fig 2. Radiated maximum likelihood tree with branch support.

Clades are colored by region. Haplotypes in bold are found in Zambia. Map insert is from Curry et al. 2015 showing locations of lions sampled. Circles indicate geographic locations for populations determined by Antunes et al [15]. UGA (Uganda); KEN (Kenya); SER (Serengeti National Park, Tanzania); NGC (Ngorongoro Crater, Tanzania); KRU (Kruger National Park, South Africa); BOT-I (Southern Botswana and Kalahari, South Africa); BOT-II (Northern Botswana); NAM (Namibia); GIR (Gir Forest, India); ANG (Angola); ZBW (Zimbabwe); and MOR (Morocco). ZAM (Zambia) is denoted by a square.

https://doi.org/10.1371/journal.pone.0217179.g002

Microsatellite diversity

A total of 398 individuals were genotyped for 14 miniSTRs (S1 Appendix). Structure analysis revealed two sub-populations (ΔK = 2; Fig 3) in agreement with mtDNA structuring [12]. Luangwa Valley (LV), Corridor (CO) and Lower Zambezi (ZA) of the LVE make up the eastern sub-population and Kafue (KF) and Sioma Ngwezi (SI) of the GKE make up the western sub-population. Little admixture is evident between sub-populations. Two individuals are assigned to a sub-population contrary to the location they were sampled. Individual 2011000432 sampled in KF is assigned to the eastern sub-population (0.81) and 2011000878 from north of North Luangwa NP is almost equally assigned to the eastern (0.43) and the western (0.57) sub-populations.

thumbnail
Fig 3. Results of STRUCTURE analysis based on 14 microsatellite loci of 398 Zambian lions.

Population assignment results for ΔK = 2 showing a separation of the western and eastern lion populations in Zambia.

https://doi.org/10.1371/journal.pone.0217179.g003

Gene diversity, as established by expected heterozygosity (HE), is high (Table 4) locus by locus and as the mean across loci for the population (0.701) and within sub-populations (eastern = 0.682 and western = 0.692). Significant deviations for HWE are seen in 13 of 14 loci when the population is considered as a whole. Separated into sub-populations, the eastern sub-population deviates from HWE at 4 loci (Leo126, Leo224, Leo230, and Leo 391) while the western sub-population deviates from HWE at only 3 loci (Leo230, Leo247, Leo247). The only shared loci that deviates from HWE is Leo230. For most loci that deviate from HWE, a deficiency of heterozygotes is observed (Table 4).

Number of alleles (A) and allelic richness (AR) are high across loci (Table 4). The number of private alleles (PA) is high for both sub-populations. There are 213 individuals with 1–4 PA at 1–3 loci (Table 5). For the eastern sub-population, there are 31 PA spanning all 14 loci with 144 of 244 individuals having at least one PA. The western sub-population has 14 PA in nine of 14 loci and 69 out of 150 individuals have at least one PA.

thumbnail
Table 5. Number of individuals with private alleles (PA).

N#PA@#Loci is the number of private alleles at the given number of loci. Totals are given for the total number of individuals with any PA (NwPA), number of individuals homozygous for a PA (NHom4PA), i.e. two of the same PA at the same locus for one or more loci, and number of individuals heterozygous for a PA (NHom4PA), i.e. two different PA at the same locus for one or more loci.

https://doi.org/10.1371/journal.pone.0217179.t005

Weak but significant genetic structure is detected with an FST = 0.04 (p-value = 0.001; Table 3) attributing 4% of molecular variance among populations and 96% of molecular variance within populations. There is little evidence of inbreeding with an FIS below zero (-0.034). The number of migrants per generation (Nm) was calculated to be 5.6 using two methods (GenAlEx and GenePop).

A Mantel test was done for males and females separately to account for possible dispersal differences between sexes due to their mating system [3739]. Both males and females show a low but significant level of IBD (Male Rxy = 0.214, p-value = 0.01; Female Rxy = 0.269, p-value = 0.01; S5 Appendix). Nei’s distances are higher between western and eastern regions (S7 Appendix). Factorial Correspondence Analysis (FCA; Fig 4) and Principle Coordinate Analysis (PCoA; S6 Appendix) reveal that LV and KF cluster separately with CO and ZA as intermediaries, overlapping with both LV and KF.

thumbnail
Fig 4. Factorial correspondence analysis (FCA) based on 14 microsatellite loci of 398 Zambian lions.

Pink = Western Zambia (KF & SI), Blue = Luangwa Valley (LV), Yellow = Lower Zambezi NP and eastern corridor (ZA & CO), Grey = Outside PAs. Axe 1, 2, and 3 represent 86.82% of the genetic variation observed.

https://doi.org/10.1371/journal.pone.0217179.g004

Effective population size (Ne) calculations with upper 95% confidence intervals are shown in Fig 5 for the eastern and western sub-populations.

thumbnail
Fig 5. Effective population size (Ne) calculations with upper 95% confidence intervals.

Estimates of Ne with Pcrit set to frequencies of 0.00 (0+), 0.01, and to exclude only singleton alleles (No S).

https://doi.org/10.1371/journal.pone.0217179.g005

Discussion

Increasing the sample size did not change the gene diversity of lions within the western sub-population of Zambia, nor did it change gene diversity at the country-wide scale [12]. Gene diversity of the eastern sub-population was lowered slightly, but still remains high (S9 Appendix). The addition of microsatellite analysis supports earlier findings that Zambia’s lion population is highly diverse (Table 4). Furthermore, with an FIS value close to zero, Zambian lions do not show any evidence of inbreeding.

The addition of another novel haplotype (Z6) did not change the configuration of the phylogenetic tree or haplotype network. Haplotype Z6 is one SNP from H11, clustering with H11 and H12 within the East/Southern Africa cluster. H11 is the most wide-spread haplotype, found in individuals as far north as Uganda and south as Kruger NP [15]. Most Zambian haplotypes cluster together within two branches of the East/Southern Africa cluster. Haplotype Z3, found only in Kafue NP, is in a different cluster with other southern haplotypes. This clustering suggests connectivity of the western sub-population southwest while the eastern sub-population has connectivity both north and south but remains to the east.

Mitochondrial analysis shows minimal gene flow between populations (FST = 0.53) while microsatellite analysis suggests greater gene flow (FST = 0.04). FST for microsatellites is the measure of the heterozygote deficit relative to its expectation under HWE [40] while FST for mtDNA is a function of the number of mutations between molecular haplotypes as measured by haploid diversity [18]. These differences in concept could also be a reason for the vast differences between the two FST values. However, this pattern could also be a result of the mating system where males disperse across farther distances than females [37]. Males are more likely to disperse, passing along their nDNA but unable to pass on mtDNA genes.

Pairwise differences (S7 Appendix) for nDNA and mtDNA also show evidence of the mating system. For nDNA distances are lower between ZA in the east and KF in the west, indicating this is likely the region genetic movement between the eastern and western sub-populations occur. Further evidence of this movement is shown in FCA analysis (Fig 4).

Migration is the transfer of genetic variation from one population to another [41]. Mutation cannot be disentangled from differences introduced by migration so FST can underestimate differentiation in a highly structured population [40], particularly when other metrics support strong structure. Migration is evident within Zambia (Nm = 5.6), therefore, FST calculated for microsatellites, could be an underestimation based on high migration coupled with the distinct structure shown in other analyses (Figs 3 and 4). However, as migration increases, the proportion of private alleles will decrease. If gene flow is low, there are more private alleles, and if gene flow is high, private alleles are more rare [41]. Zambia has a high number of private alleles with 31 private alleles in the eastern sub-population and 13 in the western sub-population found in 213 individuals. A majority of these alleles are in low frequency, however, 13 appear in frequencies greater than 1% of the sampled population.

Structure analysis shows two distinct sub-populations with admixture present in only a few individuals (Fig 3). 2011000878, sampled in a GMA north of North Luangwa NP, is the most admixed individual, assigned almost equally to each sub-population, implying it is the offspring of a resident and migrant mating. A possible migrant, 2011000432, was sampled in the western sub-population but is assigned to the eastern sub-population. This same individual was previously flagged as a possible migrant based on its mtDNA haplotype [12].

Further support for substructure is the presence of a Wahlund effect. This is when the subdivision of genetically distinct demes causes a deviation from HWE at the population level resulting in the appearance of a deficit of heterozygotes [4245]. Across Zambia, mean HO versus HE shows a heterozygote deficiency at the population level, however, when separated into sub-populations, mean values no longer deviate from HWE (Table 4). This same pattern is present across loci with a deviation from HWE at the population level and being in HWE at sub-population level. Leo230 is the only locus that remains out of HWE at the sub-population level. This may be a result of issues with the locus, as this was the only locus that exhibited problems with amplification.

With evidence of clusters, substructure, and Wahlund effects, it is suggested to remove migrants from the population before estimating Ne [46]. Therefore, migrants and individuals found outside PAs were removed before calculating Ne. The LD method was used because it is robust and mostly unbiased at a sub-population level [47]. In simulations of single-sample Ne estimators, the LD method performed best producing estimations of Ne closest to the true value of Ne [4749]. Heterozygote excess and molecular co-ancestry methods often have poor precision in comparison [46,49].

The population size of lions in Zambia during the sampling period was estimated to be as low as 700 lions with 250–500 in Kafue NP, 400–750 in the Luangwa Valley, and <50 in the corridor and Lower Zambezi NP [50]. Ne for the eastern and western sub-populations is calculated to be between 100–200, a lower value than what would be expected for a large and diverse lion population [51]. Lion prides typically have multiple related females mating with 1–7 males that originated from a different pride or prides [37]. This type of polygynous mating system can lower the value of Ne depending on the number of males breeding within each pride [51].

When loci are physically unlinked, LD is caused by drift, migration, or selection. Assuming neutral loci in an isolated population with random mating, LD would be a result of drift alone and can be used to calculate Ne [52]. However, LD calculated from a sample from a sub-population can lead to an underestimation of local Ne when the migration rate is low [47,49]. The number of migrants between the eastern and western sub-populations is calculated to be less than six individuals per generation. This is a migration rate well below 5–10% [47]. Therefore, this could be an underestimation of Ne (Fig 5) as a result of migration between sub-populations.

Overall, Zambia has a genetically diverse population of lions, although effective population size appears to be lower than expected. Previously thought to be isolated via anthropogenic and geographic barriers, the eastern and western sub-populations do exhibit isolation-by-distance, with a low level of migrants per generation. This migration may cause an underestimation of effective population size but is maintaining and introducing diversity across Zambia.

Translocation is a well‐practiced technique to prevent inbreeding [53]. Zambia does not appear to be in need of using translocation as a management strategy. While maintaining genetic diversity throughout the entire population should be considered, the high number of private alleles present within each sub-population and the level of population substructure found suggests there should be a more narrowed focus to prevent the loss of genetic diversity within sub-populations. Maintenance of diversity across Zambia will still occur through gene flow of lions between sub-populations, as it has been occurring already without intervention. This is assuming that future connectivity between sub-populations stays the same, or improves, rather than decreasing.

Range-wide studies have proposed lions are mostly structured by region due to restricted widespread movement of lions across the landscape [15,54,55]. Findings in this study agree, though movement outside protected areas is occurring. To further augment gene flow, historic or present-day corridors would need to be created and protected to help ensure continued natural dispersal of lions between the eastern and western sub-populations.

Acknowledgments

A special thank you goes to the laboratory personnel at the DNA Technologies Core Laboratory at Texas A&M University for their assistance during laboratory procedures and analysis.

References

  1. 1. Southern Africa Lion Indaba Meeting Minutes. ALWG Workshop. Skukuza, South Africa; 2018. Available: www.africanliongroup.org/
  2. 2. Lindsey PA, Balme GA, Funston P, Henschel P, Hunter L, Madzikanda H, et al. The trophy hunting of African lions: scale, current management practices and factors undermining sustainability. PLoS One. 2013;8: e73808. pmid:24058491
  3. 3. Riggio J, Jacobson A, Dollar L, Bauer H, Becker M, Dickman A, et al. The size of savannah Africa: a lion’s (Panthera leo) view. Biodivers Conserv. 2012;22: 17–35.
  4. 4. Hayward MW, O’Brien J, Kerley GIH. Carrying capacity of large African predators: Predictions and tests. Biol Conserv. 2007;139: 219–229.
  5. 5. Fennessy J, Bock F, Tutchings A, Brenneman R, Janke A. Mitochondrial DNA analyses show that Zambia’s South Luangwa Valley giraffe (Giraffa camelopardalis thornicrofti) are genetically isolated. Afr J Ecol. 2013;51: 635–640.
  6. 6. Fennessy J, Bidon T, Reuss F, Kumar V, Elkan P, Nilsson MA, et al. Multi-locus Analyses Reveal Four Giraffe Species Instead of One. Curr Biol. Elsevier Ltd; 2016;26: 2543–2549. pmid:27618261
  7. 7. Cotterill F. Reduncine Antelope of the Zambezi Basin. Zambezi Basin Wetlands Volume II. 2000. pp. 145–199.
  8. 8. Pocock RI. V.—The species and subspecies of Zebras. Ann Mag Nat Hist Ser 6. 1897;20: 33–52.
  9. 9. Kasiki, S. and Hamunyela E, Kasiki SM, Hamunyela E. CITES Periodic Review of the Status of African Lion Across its Range (Kenya & Namibia). Cites. 2014.
  10. 10. Rosenblatt E, Creel S, Becker MS, Merkle J, Mwape H, Schuette P, et al. Effects of a protection gradient on carnivore density and survival: An example with leopards in the Luangwa valley, Zambia. Ecol Evol. 2016;6: 3772–3785. pmid:27231529
  11. 11. Lindsey P a, Nyirenda VR, Barnes JI, Becker MS, McRobb R, Tambling CJ, et al. Underperformance of African protected area networks and the case for new conservation models: insights from Zambia. PLoS One. 2014;9: e94109. pmid:24847712
  12. 12. Curry CJ, White PA, Derr JN. Mitochondrial Haplotype Diversity in Zambian Lions: Bridging a Gap in the Biogeography of an Iconic Species. PLoS One. 2015;10: e0143827. pmid:26674533
  13. 13. Bertola LD. Genetic diversity in the lion (panthera leo (Linnaeus 1758)): unravelling the past and prospects for the future. Universiteit Leiden. 2015.
  14. 14. Curry CJ, Derr JN. Development of lion miniSTRs for use with modern and historical DNA samples. Manuscr Submitt Publ. 2019;
  15. 15. Antunes A, Troyer JL, Roelke ME, Pecon-Slattery J, Packer C, Winterbach C, et al. The evolutionary dynamics of the lion Panthera leo revealed by host and viral population genomics. PLoS Genet. 2008;4: e1000251. pmid:18989457
  16. 16. Kim JH, Antunes A, Luo SJ, Menninger J, Nash WG, O’Brien , et al. Evolutionary analysis of a large mtDNA translocation (numt) into the nuclear genome of the Panthera genus species. Gene. 2006;366: 292–302. pmid:16380222
  17. 17. Sequencher version 4.8 sequence analysis software [Internet]. Ann Arbor, MI USA: Gene Codes Corporation; p. 8. Available: http://www.genecodes.com/
  18. 18. Excoffier L, Laval G, Schneider S. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evol Bioinform Online. 2005;1: 47–50. pmid:21565059
  19. 19. Guindon S, Dufayard J-F, Lefort V, Anisimova M. New Alogrithms and Methods to Estimate Maximum- Likelihoods Phylogenies: Assessing the performance of PhyML 3.0. Syst Biol. 2010;59: 307–321. pmid:20525638
  20. 20. Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics. 2001;17: 754–755. pmid:11524383
  21. 21. Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003;19: 1572–1574. pmid:12912839
  22. 22. Bandelt HJ, Forster P, Röhl a. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol. 1999;16: 37–48. pmid:10331250
  23. 23. Locke M, Baack E, Toonen R. Quick introduction to STRand Software. 2000. pp. 1–15.
  24. 24. Peakall R, Smouse PE. GenALEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research-an update. Bioinformatics. 2012;28: 2537–2539. pmid:22820204
  25. 25. Rousset F. GENEPOP’007: A complete re-implementation of the GENEPOP software for Windows and Linux. Mol Ecol Resour. 2008;8: 103–106. pmid:21585727
  26. 26. Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F. GENETIX 4.05, logiciel sous Windows TM pour la génétique des populations. Laboratoire Génome, Populations, Interactions, CNRS UMR 5000, Université de Montpellier II, Montpellier (France). 2004.
  27. 27. Kalinowski ST. HP-RARE 1.0: A computer program for performing rarefaction on measures of allelic richness. Mol Ecol Notes. 2005;5: 187–189.
  28. 28. Kalinowski ST. Counting alleles with rarefaction: Private alleles and hierarchical sampling designs. Conserv Genet. 2004;5: 539–543.
  29. 29. Do C, Waples RS, Peel D, Macbeth GM, Tillett BJ, Ovenden JR. NeEstimator v2: Re-implementation of software for the estimation of contemporary effective population size (Ne) from genetic data. Mol Ecol Resour. 2014;14: 209–214. pmid:23992227
  30. 30. Waples RS, Do C. Linkage disequilibrium estimates of contemporary Ne using highly variable genetic markers: A largely untapped resource for applied conservation and evolution. Evol Appl. 2010;3: 244–262. pmid:25567922
  31. 31. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155: 945–59. Available: https://www.ncbi.nlm.nih.gov/pmc/articlerender.fcgi?artid=1461096&tool=pmcentrez&rendertype=abstract pmid:10835412
  32. 32. Earl DA, vonHoldt BM. STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv Genet Resour. 2012;4: 359–361.
  33. 33. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Mol Ecol. 2005;14: 2611–2620. pmid:15969739
  34. 34. Jakobsson M, Rosenberg NA. CLUMPP: A cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics. 2007;23: 1801–1806. pmid:17485429
  35. 35. Rosenberg NA. DISTRUCT: A program for the graphical display of population structure. Mol Ecol Notes. 2004;4: 137–138.
  36. 36. Budowle B, van Daal A. Extracting evidence from forensic DNA analyses: future molecular biology directions. Biotechniques. 2009;46: 339–350. pmid:19480629
  37. 37. Pusey AE, Packer C, Erhoff-Mulder MB. The Evolution of Sex-biased Dispersal in Lions. Behaviour. 1987;101: 275–310.
  38. 38. Bertola LD, Vrieling K, de Iongh HH. Chapter: Conservation genetics of the lion: new approaches to species conservation. Genetic Diversity: New Research. 2012.
  39. 39. Nowell K, Jackson P. Status Survey and Conservation Action Plan: Wild Cats (IUCN/SSC Cat Specialist Group). 1996.
  40. 40. Balloux F, Lugon-Moulin N. The estimation of population differentiation with microsatellite markers. Mol Ecol. 2002;11: 155–165. pmid:11856418
  41. 41. Allendorf FW, Luikart G, Aitken SN. Conservation and the Genetics of Populations. 2nd ed. The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK: John Wiley & Sons, Ltd; 2013.
  42. 42. Zhivotovsky LA. Relationships between wright’s FST and FIS statistics in a context of wahlund effect. J Hered. 2015;106: 306–309. pmid:25888609
  43. 43. Garnier-Géré P, Chikhi L. Population Subdivision, Hardy-Weinberg Equilibrium and the Wahlund Effect. eLS. 2013;
  44. 44. Waples RS. Testing for hardy-weinberg proportions: Have we lost the plot? J Hered. 2015;106: 1–19. pmid:25425676
  45. 45. Wahlund S. Zusammensetzung von populationen und korrelation erscheinungen vom standpunk t der vererbungslehre aus betrachtet. Hereditas. 1928; 65–106.
  46. 46. Luikart G, Ryman N, Tallmon DA, Schwartz MK, Allendorf FW. Estimation of census and effective population sizes: The increasing usefulness of DNA-based approaches. Conserv Genet. 2010;11: 355–373.
  47. 47. Waples RS, England PR. Estimating contemporary effective population size on the basis of linkage disequilibrium in the face of migration. Genetics. 2011;189: 633–44. pmid:21840864
  48. 48. Waples RS. Making sense of genetic estimates of effective population size. Mol Ecol. 2016;25: 4689–4691. pmid:27671356
  49. 49. Wang J, Santiago E, Caballero A. Prediction and estimation of effective population size. Heredity (Edinb). Nature Publishing Group; 2016;117: 193–206. pmid:27353047
  50. 50. IUCN. Conservation Strategy for the Lion in Eastern and Southern Africa. IUCN SSC Cat Specialist Group. 2006.
  51. 51. Nunney L. The Influence of Mating System and Overlapping Generations on Effective Population Size. Evolution (N Y). 1993;47: 1329–1341.
  52. 52. Hill WG. Estimation of effective population size from data on linkage disequilibrium. Genet Res (Camb). 1981;
  53. 53. Trinkel M, Ferguson N, Reid a, Reid C, Somers M, Turelli L, et al. Translocating lions into an inbred lion population in the Hluhluwe-iMfolozi Park, South Africa. Anim Conserv. 2008;11: 138–143.
  54. 54. Dubach JM, Briggs MB, White PA, Ament BA, Patterson BD. Genetic perspectives on “Lion Conservation Units” in Eastern and Southern Africa. Conserv Genet. 2013;14: 741–755.
  55. 55. Dubach J, Patterson BD, Briggs MB, Venzke K, Flamand J, Stander P, et al. Molecular genetic variation across the southern and eastern geographic ranges of the African lion, Panthera leo. Conserv Genet. 2005;6: 15–24.