Exposing the illegal trade in cycad species (Cycadophyta: Encephalartos) at two traditional medicine markets in South Africa using DNA barcoding
Abstract
Species in the cycad genus Encephalartos are listed in CITES Appendix I and as Threatened or Protected Species in terms of South Africa’s National Environmental Management: Biodiversity Act (NEM:BA) of 2004. Despite regulations, illegal plant harvesting for medicinal trade has continued in South Africa and resulted in declines in cycad populations and even complete loss of sub-populations. Encephalartos is traded at traditional medicine markets in South Africa in the form of bark strips and stem sections; thus, determining the species traded presents a major challenge due to a lack of characteristic plant parts. Here, a case study is presented on the use of DNA barcoding to identify cycads sold at the Faraday and Warwick traditional medicine markets in Johannesburg and Durban, respectively. Market samples were sequenced for the core DNA barcodes (rbcLa and matK) as well as two additional regions: nrITS and trnH-psbA. The barcoding database for cycads at the University of Johannesburg was utilized to assign query samples to known species. Three approaches were followed: tree-based, similarity-based, and character-based (BRONX) methods. Market samples identified were Encephalartos ferox (Near Threatened), Encephalartos lebomboensis (Endangered), Encephalartos natalensis (Near Threatened), Encephalartos senticosus (Vulnerable), and Encephalartos villosus (Least Concern). Results from this study are crucial for making appropriate assessments and decisions on how to manage these markets.
Résumé
Les espèces appartenant au genre Encephalartos des cycadales figurent à l’Annexe I de la convention CITES en tant qu’espèces menacées ou protégées sur la base d’une loi de 2004 en Afrique du Sud (« National Environmental Management : Biodiversity Act », ou NEM : BA). En dépit de cette règlementation, des activités illégales de récolte pour des fins médicinales se sont poursuivies en Afrique du Sud et ont entraîné un déclin dans les populations de cycadales et même une perte complète de certaines sous-populations. L’Encephalartos est vendu dans les marchés de remèdes traditionnels en Afrique du Sud sous forme de bandes d’écorce et de sections de tiges; il est ainsi très difficile d’identifier les espèces vendues en raison de l’absence de parties végétales pouvant servir à l’identification. Dans ce travail, les auteurs présentent une étude de cas sur l’emploi des codes à barres de l’ADN pour l’identification des cycadales vendues dans les marchés de remèdes traditionnels Faraday et Warwick, respectivement situés à Johannesbourg et Durban. Les échantillons de ces marchés ont été séquencés pour les séquences usuelles (rbcLa et matK) ainsi que pour deux régions additionnelles (nrITS et trnH-psbA). La base de données pour les codes à barres des cycadales à l’Université de Johannesbourg a été employée pour assigner les séquences des échantillons à des espèces connues. Trois approches ont été explorées : les méthodes fondées sur les arbres, la similarité et les caractères (BRONX). Les échantillons ont été identifiés comme appartenant aux espèces suivantes : Encephalartos ferox (presque menacée), Encephalartos lebomboensis (en danger), Encephalartos natalensis (presque menacée), Encephalartos senticosus (vulnérable) et Encephalartos villosus (préoccupation mineure). Les résultats obtenus seront essentiels pour l’évaluation et la prise de décisions en ce qui a trait à la gestion de ce commerce. [Traduit par la Rédaction]
Introduction
Cycads, with a fossil record dating back to the early Permian period (Norstog and Nicholls 1997; Brenner et al. 2003), are the world’s oldest extant group of seed plants and have survived three mass extinction events in earth’s history. Furthermore, cycads are classified as the world’s most threatened plant group (Donaldson 2011). Yet, conservation of cycads presents a challenge as threats such as habitat transformation (e.g., replacement by commercial crops), illegal trade, and harvesting activities subject them to human-induced extinction (Donaldson and Bösenberg 1999; Donaldson 2008).
African cycads are represented by three genera: Cycas (only Cycas thouarsii R.Br. distributed in eastern Africa), Encephalartos Lehm., and the monotypic genus Stangeria T. Moore. Encephalartos, comprised of 68 species and subspecies, is endemic to Africa, with South Africa (SA) being a diversity hotspot, with 37 species. Of these, approximately 70% are threatened with extinction (Retief et al. 2014). Due to risk associated with international trade in wild-collected specimens, all SA Encephalartos spp. are listed in CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) Appendix I. This number includes three Extinct in the Wild (EW), 12 Critically Endangered (CR), four Endangered (EN), seven Near Threatened (NT), eight Vulnerable (VU), and three Least Concern (LS) species (http://redlist.sanbi.org). Notably, two of the three species (Encephalartos brevifoliolatus Vorster and Encephalartos nubimontanus P.J.H. Hurter) that are classified as EW acquired this Red List Status between 2003 and 2010 (Cousins et al. 2012). Furthermore, Encephalartos cerinus Lavranos & D.L. Goode, Encephalartos inopinus R.A. Dyer, Encephalartos latifrons Lehm., and Encephalartos msinganus Vorster, listed as CR, have population sizes in the wild of less than 100 individuals (Donaldson 2008). Consequently, SA is now at risk of losing 50% of its cycad species within the next 2–10 years. The South African National Biodiversity Institute (SANBI) identified this as the “South African cycad extinction crisis” (Cousins et al. 2012).
Continuous trading of South African Encephalartos spp., predominantly for the horticultural trade, has led to a rapid decline of wild populations (Donaldson and Bösenberg 1999). Wild cycad populations are particularly vulnerable to harvesting due to the slow growth rate of individuals. Habitat loss plays a minor role in depletion of populations, whereas invasive species have a direct impact on approximately 10% of cycad habitats (Donaldson 2008). Furthermore, with an estimated 27 million users of traditional plant-based treatments in SA (Mander et al. 2007), which in some instances include cycads, puts cycad populations now further at risk. Approximately two-thirds (25 species) of SA’s Encephalartos spp. (known as “isiGqiki-somkovu” in Zulu) are collected for traditional medicinal purposes and illegally traded at traditional medicine (TM) markets, mainly for the protection they offer from evil spirits (Cousins et al. 2011, 2012). Large arborescent species appear to be harvested by removing bark strips from adult individuals, whereas smaller arborescent and subterranean species are harvested by removing the entire plant, stripping off all its characteristic features (Donaldson 2008; Cousins et al. 2012). This makes identification very challenging or impossible in many cases.
Previous studies described the trade of cycads at urban TM markets, but only morphological-based identifications were used, and in several cases plants could not be identified to species level (Cousins et al. 2012, 2013; Williams et al. 2014). Here, we used DNA barcoding to determine the diversity of cycads sold at SA’s two largest TM markets (Faraday in Johannesburg and Warwick in Durban). The value of applying DNA barcoding to cycad identification and the conservation implications of the results are discussed.
Material and methods
Sampling
Our sample strategy involved sampling all the stalls where cycads were sold to ensure all possible Encephalartos spp. present at the TM markets were collected. Only openly displayed products were purchased from the traders. The price per bark/stem fragment (160 cm × 10 cm) varied between 30 and 40 ZAR (2.60–3.47 CAD). In total, 45 unknown cycad bark specimens were obtained between January and May 2015 from 12 stores at Faraday TM market in Johannesburg, Gauteng Province, and 20 stores at the Warwick TM market, located on the periphery of Durban Central Business District, KwaZulu-Natal Province (SA; Fig. 1). Warwick is the largest TM market in the country, with nearly 500 traders (Mander 1998), whereas Faraday is the second largest, with over 400 traders and the wholesale epicenter for the trade in traditional medicine in Gauteng (Williams et al. 2007a, 2007b). All specimens collected were processed and analyzed at the African Centre for DNA Barcoding (ACDB) at the University of Johannesburg.
Fig. 1.
DNA extraction, amplification, and sequencing
DNA was extracted from 0.2–0.8 g of bark material using the 2× CTAB method described by Doyle and Doyle (1987). Polyvinyl pyrolidone (2%) was added to reduce the effect of high polysaccharide concentrations in the samples. To avoid problems of PCR inhibition, all DNA samples were purified using QIAquick purification columns (Qiagen Inc., Hilden, Germany) according to the manufacturer’s protocol.
Polymerase chain reaction (PCR) amplifications were done using the core DNA barcodes (matK and rbcLa) and the additional regions trnH-psbA and nrITS recommended to enhance the performance of the core barcodes (CBOL Plant Working Group 2009; Hollingsworth et al. 2009). Primer pairs used for the PCR amplification of matK, rbcLa, and trnH-psbA regions were GYM F1A:GYM R1A (Hollingsworth et al. 2011), rbcLa-F (Levin et al. 2003):rbcLa-R (Kress and Erickson 2007), and NY1493:NY1494 (Sass et al. 2007), respectively. The nrITS region (internal transcribed spacer 1, 5.8S ribosomal RNA, and internal transcribed spacer 2) was amplified in two overlapping pieces using two internal primers with a pair of external primers: NY90F:NY729R and 728F:514R (White et al. 1990; Sun et al. 1994). All reactions were done using Ready-Mix Mastermix (Advanced Biotechnologies, Epsom, UK), with the addition of 4.5% of dimethylsulphoxide to minimize problems with secondary structures and improve annealing (Palumbi 1996). We also added 3.2% bovine serum albumin to all PCR reactions to stabilize enzymes, reduce problems with secondary structure, and improve annealing (Savolainen et al. 1995). PCR amplifications were done using the following programs: for matK, rbcLa, and trnH-psbA, pre-melt at 95 °C for 2:30 min, denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, extension at 72 °C for 2 min (for 35 cycles), followed by a final extension at 72 °C for 10 min; for nrITS the protocol consisted of pre-melt at 94 °C for 2 min, denaturation at 94 °C for 1 min, annealing at 52 °C for 1 min, extension at 72 °C for 3 min (for 28 cycles), followed by a final extension at 72 °C for 7 min. Amplified products were purified using QIAquick columns (Qiagen Inc., Germany) following the manufacturer’s protocol.
Bidirectional cycle sequencing reactions were performed using the BigDye® Terminator V3.1 kit (Applied Biosystems, Inc. Warrington, Cheshire, UK). Cycle sequencing products were precipitated in ethanol and sodium acetate to remove excess dye terminators before sequencing on an ABI 3130xl genetic analyser. Complementary strands were assembled and edited using Sequencher v.5.1 (Gene Codes Crop., Ann Arbor, Mich., USA). All protein-coding regions were translated to verify that no stop codons were present. Alignment was performed for trnH-psbA and nrITS using Multiple Sequence Comparison by Log-Expectation (MUSCLE v.3.8.31; Edgar 2004) and adjusted manually in PAUP* (version 4.0b10; Swofford 2002). The alignments used for further analysis are available as supplementary data, Data file S12.
Encephalartos DNA barcode library
An Encephalartos DNA barcode library was prepared and used to identify the unknown TM market samples. This library was assembled from 260 taxa (excluding market samples) representing 48 southern African species of known provenance that were identified by an Encephalartos taxonomist, Piet Vorster. We have also consulted with various authorities from the Environmental Management Inspectorate (EMI) and the Green Scorpions (Environmental law enforcement) based in the different provinces to assist with and verify species identifications. Between two and six individuals per species were sampled to incorporate species-level variation into the sampling. Specimens were prepared into herbarium vouchers and archived at the South African National Biodiversity Institute (SANBI), Pretoria (PRE), and the herbarium at the University of Johannesburg (JRAU), with the collection and sequence data deposited in the Barcode of Life Data Systems (BOLD; www.boldsystems.org; Ratnasingham and Hebert 2007). Although the barcode library for this study contained 48 of the estimated 68 Encephalartos spp. and subsp. occurring in Africa, all of the 37 species occurring in SA are included, and all other Encephalartos spp. occurring in the region have been barcoded and are available on BOLD (Rousseau 2012). Voucher specimen information and GenBank accession numbers are listed in the supplementary data, Table S12.
To test the efficiency of the Encephalartos DNA barcode library, we calculated species discrimination using BRONX version 2.0 (Barcode Recognition Obtained with Nucleotide eXposé’s; Little and Stevenson 2007; Little 2011), which is an alignment-free algorithm that relies on the presence/absence of particular characters in the barcode sequences for identification, instead of using them all (Van Velzen et al. 2012), thereby accounting for within-taxon variation (Little 2011). For this, a unified reference database was constructed for southern African Encephalartos spp. using all four plant DNA markers. The generated markers in the dataset were combined using 18 “N” codes insulation between the marker sequences. No context combinations that included IUPAC ambiguity codes were used. The combined Encephalartos dataset was reduced to reference sequences of a series of characters defined by flanking pretext and postext of the default six base lengths using the fasta2bdb.pl function. This sidelines the alignment difficulties assigned to variable markers (i.e., trnH-psbA and nrITS). The resulting BRONX character database comprised of a list of terminals consisting of all observed sequence fragments and their flanking context (pretext and postext). To calculate species discrimination, a sequence of each complete sample was queried against this reference database. We employed the scoring function of one for each matching pretext/text/postext combination. Thus, the score for each query could vary between 0 and 1, where the reference terminal with the highest score is considered the correct identification (Little 2011).
Data analysis of query market samples
Three approaches were followed to assign query market samples to known species, namely tree-based method, similarity-based method, and character-based method. In the tree-based method, query samples were assigned to species if they form a monophyletic cluster with identified specimens in the Encephalartos DNA barcode tree. For this approach we used maximum parsimony (MP; Farris 1972) and Bayesian inference (BI; Huelsenbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003). Lepidozamia Lehm. was used as the outgroup in the analysis as its phylogenetic placement has consistently been resolved as sister to Encephalartos (Treutlein and Wink 2002; Hill et al. 2003; Bogler and Francisco-Ortega 2004; Chaw et al. 2005; Sangin et al. 2008; Zgurski et al. 2008). Maximum parsimony using PAUP* v. 4.0b1 (Swofford 2002) was implemented to analyze the combined dataset (matK, rbcLa, trnH-psbA, and nrITS). Tree searches were conducted using 1000 replicates of random taxon addition, retaining 10 trees at each step, with tree-bisection-reconnection (TBR) branch swapping and MulTrees in effect. Support for clades was estimated using bootstrap analysis (Felsenstein 1985) with 1000 replicates, simple sequence addition, TBR swapping, with MulTrees in effect, but saving a maximum of 10 trees per replicate. The following categories were used to classify bootstrap support (BS): low (50%–74%), moderate (75%–84%), and high/strong (85%–100%). Delayed transformation character optimization (DELTRAN) was used to calculate branch lengths, due to reported errors with accelerated transformation optimization (ACCTRAN) in PAUP* v.4.0b1 (http://paup.csit.fsu.edu/problems.html).
Bayesian inference (Huelsenbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003) was performed using MrBayes v.3.1.3, as implemented on the Cyber infrastructure for Phylogenetic Research (CIPRES) Science Gateway v.3.1 (Miller et al. 2009; www.phylo.org) on the combined data set. Prior to BI, a hierarchical likelihood ratio test implemented in MODELTEST v.3.06 (Posada and Crandall 1998) was used to determine the appropriate substitution model for each of the four gene regions sequenced based on the Akaike information criterion (AIC; Sugiura 1978). The following models were selected: rbcLa = HKY + Y, matK = TVM + G, trnH-psbA = HKY + I + G, nrITS = K81uF + I + G. A partitioned analysis was run since different models were selected. Analyses were run with 2 000 000 generations with a sampling frequency of 200. The log-likelihood scores were plotted against their likelihoods to determine the point where likelihoods converged on a maximum value, and the trees before the convergence were discarded as burn in (1000 trees). The remaining trees were imported into PAUP*v.4.0b10 to generate a majority-rule consensus tree showing posterior probabilities (PP) of all the observed bi-partitions. The following scale was used to evaluate the posterior probability values (PPs): well supported (0.95–1.0) and weakly supported (<0.95).
The NCBI’s MegaBlast algorithm (www.blast.ncbi.nlm.nih.gov/Blast.cgi), using the default setting, was used to assign query barcodes to species. Identification was made taking both the E-value and maximum identity into consideration.
Lastly, we used the character-based method BRONX (version 2.0) of Little (2011) to assign query market samples to known species. The function Bronx.pl was used to match the context of the query sequence with the corresponding invariant flanking context in the reference database. On the basis of these flanking regions, the variable regions were scored in the query sequence integrating the variability observed within the taxon in the scoring procedure. The technique in effect imitates the procedure used in conventional morphological systematics, where each composite exposé is equivalent to a taxonomic morphological description and where some characters provide context for others (Little 2011).
Results
DNA barcode success and efficacy of Encephalartos DNA barcode library
Recovery of DNA barcodes from TM market products was effective, with sequencing success for rbcLa, matK, trnH-psbA, and nrITS being 100%, 100%, 89%, and 89%, respectively. Of the 46 Encephalartos spp. represented by two or more individuals in the complete sample (Reference library, with 2 rbcLa and 10 trnH-psbA sequences lacking in the dataset), BRONX could distinguish 40 species (87%) from all other species using a combination of the four above-mentioned markers (Table S22).
Tree-based results
The parsimony analysis of the combined dataset (rbcLa, matK, trnH-psbA, and nrITS) resulted in four equally parsimonious trees (tree length, 675 steps; CI = 0.75; RI = 0.95). Of the 4068 included characters, 3604 (88.6%) were constant, and 464 (11.4%) were variable. Of the variable sites, 305 (7.5% of the total) were potentially parsimony informative. The combined MP analysis is congruent with the BI analysis, and therefore results are displayed on the BI tree (Fig. 2). Market samples are retrieved in four major clades: all samples from Warwick are included in a clade containing Encephalartos aplanatus Vorster and Encephalartos villosus Lem. (BI = 1.0); sample St104_1 grouped with Encephalartos ferox G. Bertol (BI = 1.0; BP = 94); samples St004_1, St004_2, St004_3, St014_1, St014_2, and St014_3 clustered with Encephalartos natalensis R.A. Dyer & I. Verd. (BI = 0.85); the remaining samples (St008_1, St032_1, St032_2, St032_3, St039a_1, St039b_1, St101_1, St101_1, St102_1, St102_3, St103_1, St103_2, St117a_1, and St117b_1) are allied with Encephalartos aemulans Vorster, Encephalartos lebomboensis I. Verd., and Encephalartos senticosus Vorster (BI = 0.73).
Fig. 2.
Blast results
The identification success was dependent on the marker (Table S32). The highest-scoring hit from each query on the basis of E-value and maximum identity was taken as the barcoding identification. All query samples could be identified to genus level, whereas species identification was ambiguous in 54% of the cases as two or more equally high-scoring hits were assigned per query. The trnH-psbA region did not show promise as a barcoding locus for Encephalartos because of its inability to provide specific identification for taxa; in all cases when using trnH-psbA, a different species was identified that did not match what was found when using rbcLa, matK, and (or) nrITS. Furthermore, trnH-psbA also occurs in more than one copy in cycads (Sass et al. 2007).
BRONX results
The combined Encephalartos dataset displays an average sequence length of 3474 base pairs (bp) with the shortest fragment at 1785 bp and the longest at 3784 bp, representing 48 species of the genus Encephalartos. The 45 combined (rbcLa, matK, trnH-psbA, and nrITS) query market samples displayed a mean sequence length of 3098 bp with the minimum at 1269 bp and the maximum at 3631 bp. Overall, no individual query sequences matched the length of the longest sequences in the reference database. Plant material poorly conserved is routinely expected at TM markets and greatly influence the fragment size generated. However, for this specific reason the BRONX analysis is explicitly designed to use only unambiguous context/text in the scoring, which caters for the imperfect overlaps between the query and reference database (Little 2011). This is adequately reflected in the results (Fig. 3; Table S42), as genus-level tests of identification were 100% successful, with a species-level identification success rate of over 90%. Query sequences JW4, JW5, JW8, and JW24 yielded ambiguous species-level identifications. This can be attributed to the lack of nrITS sequences for these samples.
Fig. 3.
Encephalartos spp. identified at Faraday and the Warwick TM markets
Utilizing the barcoding database for cycads at the University of Johannesburg and combining the tree-based, blast, and BRONX results, we positively identified five Encephalartos spp. currently traded at Faraday and the Warwick TM markets, viz., E. lebomboensis (EN; Lebombo cycad), E. senticosus (VU; Jozini cycad), E. natalensis (NT; Natal giant cycad), E. ferox (NT; Tongaland or Kozi cycad), and E. villosus (LC; poor man’s cycad).
Discussion
As stated by Douglas Goode, considered to be the world’s finest cycad artist and a renowned expert on the group, cycads could be referred to as the “rhino horn” of the plant kingdom (Potgieter and Cresswell 1989). Very similar to the rhino poaching crises that SA is currently facing is the devastating loss of cycads—a tragedy largely unnoticed. Encephalartos spp. face several threats, with the largest being the illegal removal of adult plants from the wild for private cycad collections, horticultural purposes, and as parental stock for seedling propagation for both domestic and international trade (Donaldson 2003, 2008). Cycads have also been poached for use in TM markets where these plants are used for traditional purposes (Cousins et al. 2012, 2013; Williams et al. 2014). Harvesting of plant species from the wild for traditional medicine is having an adverse effect on biodiversity (Dold and Cocks 2002; Botha et al. 2004) and is mostly ignored by provincial and national conservation agencies, even though results from several studies have shown that there is currently a lucrative trade in cycad species at TM markets in SA. For example, Cunningham and Davis (1997) found that 30% of the 54 TM shops investigated in KwaZulu-Natal Province sold Encephalartos, averaging a 50 kg size bag annually; 6% of the 50 TM shops surveyed in the Johannesburg area (Williams 2003) sold Encephalartos specimens. Cousins et al. (2011) reported that Encephalartos spp. were sold by 26.4% and 13.2% of traders at Faraday and Warwick TM markets, respectively, with an estimated nine metric tons traded at Warwick in 2009.
Correctly identifying species at TM markets is a major challenge since plants are normally traded as plant parts and plant-derived products, either separately or in mixtures. The emergence of DNA barcoding now provides a tool to tackle this challenge. DNA barcoding involves using a short, agreed-upon region of a genome as a unique identifier for species and provides identification of a specimen to species even if only a small fragment of plant material is available. Furthermore, the power and strength of DNA barcoding has been demonstrated in many botanical studies ranging from the discovery of unknown species, resolving taxonomic problems, species adulteration in herbal products (Srirama et al. 2010; Baker et al. 2012; Newmaster et al. 2013), and TM market surveys (Kool et al. 2012; Arun Dev et al. 2014); DNA barcoding has also proven to be of great utility in scrutinizing the illegal trade of endangered plant species (Pryer et al. 2010; Subedi et al. 2013). However, the taxonomic value of DNA barcoding has rarely been tested on phylogenetically closely related species, for example Encephalartos. Sass et al. (2007) assessed the efficacy of the approach across cycad genera and recognized that none of the proposed markers (rbcLa and matK) provided unique identifiers for all species tested and concluded that nrITS showed the most promise in terms of variability. Nicolalde-Morejón et al. (2011) evaluated successful species-level molecular identification in the three extant cycad genera occurring in Mexico (Ceratozamia, Dioon, and Zamia) and found that at least three chloroplast regions and one nuclear region were needed to achieve >70% unique species identification. Also, the number of species within genera with horticultural appeal is often over-estimated. For instance, 894 species were initially described for the genus Viburnum; this number has now been drastically reduced to only 172 (Clement and Donoghue 2012). Such over-estimation of species might also be possible within Encephalartos.
In the current study, using DNA barcoding, five Encephalartos spp. were identified at Faraday and the Warwick TM markets, viz., E. ferox, E. lebomboensis, E. natalensis, E. senticosus, and E. villosus. All samples collected at Warwick TM market were identified as E. villosus (poor man’s cycad), which coincides with the findings of Williams et al. (2014), i.e., that it is the most abundant species sold as bark fragments at Warwick. Encephalartos villosus, presently categorized as LC (Donaldson 2010a), is one of the most common cycads in SA, with numerous large subpopulations throughout KwaZulu-Natal in easy access from/to Warwick medicinal market. It is evident that many hundreds of plants have been taken from their habitats; in the Eastern Cape, large habitat areas have been cleared for pineapple plantings, and in KwaZulu-Natal many E. villosus habitats have been converted to banana plantations. It should be noted that relative abundance in a cycad species should never compromise conservation efforts; thus, care must be taken that the incessant removal of E. villosus from the wild be monitored to ensure that the species does not become threatened in the near future.
The extensive trade in E. natalensis fragments is concerning. Williams et al. (2014) also identified E. natalensis as one of the cycads traded in the largest quantities at Faraday. Encephalartos natalensis has declined in certain parts of its range, which includes the Qumbu and Tabankulu areas of the northern part of the Eastern Cape, through most of KwaZulu-Natal up to the upper catchment areas of the Mkuze and Umfolozi rivers near Vryheid in SA. The overall population decline is estimated to be <30% over the past 60 years (Donaldson 2010b). If this decline continues and the overall population numbers drop below 10 000, this species could be listed as Vulnerable (Donaldson 2010b).
One sample at Faraday was identified as E. ferox. It is a widespread species occurring in northern KwaZulu-Natal Province of SA and southern Mozambique. However, according to Donaldson (2010c), E. ferox needs monitoring as numerous plants have been removed and sold alongside roads in Mozambique. Donaldson (2010c) estimated that population reduction could increase to >30% over three generations in which case it would qualify as VU under criterion A2 (Donaldson 2010c).
It is of specific concern that E. lebomboensis and E. senticosus are also sold at Faraday. Encephalartos lebomboensis occurs in Swaziland and Mozambique as well as in northeastern KwaZulu-Natal and the Lebombo Mountains of southern Mpumalanga Province of SA. The E. lebomboensis population has declined by at least 50% in the past 90 years, with only 5000 mature plants recorded in the wild in 2010 by Donaldson (2010d). It qualifies as EN due to the extent of past decline and its narrow distribution range. The E. senticosus population has declined by >30% at least in parts of its range (near Sitegi in Swaziland and Mkuze in SA) during the previous 60 years. There is evidence of continuing decline, which means that it also qualifies as VU under criterion C1 (Donaldson 2010e).
Both Crouch et al. (2003) and Cousins et al. (2012) identified E. ghellinckii Lem. at the Warwick market, and this species was also identified at Faraday along with E. ngoyanus I. Verd. by Cousins et al. (2013). We did not encounter any other Encephalartos spp. at these two markets other than the five species mentioned above. Moreover, we identified an additional species sold at the Faraday market, E. lebomboensis, which was not encountered by Cousins et al. (2013). This might be due to the fact that TM traders are aware that the trade in cycads is illegal and therefore tend to hide their stock of cycad fragments. Also, Encephalartos spp. sold at Faraday and Warwick could have been overlooked or missed as our collections at the two markets, especially at Warwick, were conducted as snap shots and not over a longer period. Finally, Cousins et al. (2012, 2013) based their identification on stem- and leaf-based morphological characters, together with harvesting localities records. The authors interviewed TM traders regarding harvesting locality, and these data were then use to distinguished species by plotting harvesting localities on a map that was overlaid with the geographical distribution of the 14 Encephalartos spp. occurring in KwaZulu-Natal. Although this approach to identify Encephalartos spp. in the TM trade is useful, numerous Encephalartos spp. are morphologically very similar, and closely related species are often difficult to distinguish from one another (Golding and Hurter 2003; Grobbelaar 2004). Therefore, morphological and harvesting recollections alone will unlikely provide conclusive identifications that could be used in law enforcement programs and prosecutions. Also, plants traded at TM markets are mainly harvested from wild resources by specialized collectors and reach the market via a middle man; thus, exact harvesting localities are not always available.
To summarize, all Encephalartos spp. (E. ferox, E. lebomboensis, E. natalensis, and E. senticosus) identified at Faraday and E. villosus identified at Warwick have distribution ranges extending into KwaZulu-Natal. This coincides with the demographics of traders and customers at these two TM markets. Traders from Warwick are mainly Zulu speaking from KwaZulu-Natal and neighbouring countries such as Swaziland and Mozambique, whereas the majority of the traders at Faraday are migrants from KwaZulu-Natal, with 90% speaking Zulu. Two reasons proposed by Williams (2003) for the movement of traders from Warwick to Faraday are, first, that many traders realized that the market in Durban was becoming overcrowded with traders, and hence concluded that there was comparatively more business at Faraday. Second, many traders are of the opinion that most plants are easily accessible in the mountains of KwaZulu-Natal, and consequently local people can harvest what they need and do not need to buy it, therefore resulting in fewer customers. The opposite is true for Faraday; the urban population in Gauteng does not have free access to medicinal plants and need to buy it from TM markets. Finally, both TM markets sell stock harvested primarily in KwaZulu-Natal, with traders at Faraday obtaining most of their stock from Warwick TM market. The explanation why we identified different Encephalartos spp. at Faraday than at Warwick could be attributed to the fact that at Warwick traders are more aware that trading in cycads is illegal and thus hide them under more common species. The opposite is true for Faraday, where we observed several heaps of Encephalartos fragments.
Conclusion
This study presents yet another example of the applicability of DNA barcoding in the identification of medicinal plants and highlights the undercover trade in Encephalartos spp. at TM markets, which is not adequately recognized in official government documents or databases. Five Encephalartos spp. (Fig. 4) were identified as currently being traded illegally at Faraday and Warwick. All of these species were harvested in the form of bark strips and whole stem sections, thus making morphological identification difficult. The combined power of rbcLa, matK, trnH-psbA, and nrITS to discriminate among southern African Encephalartos spp. has proven to be relatively high (86%), thus validating the use of DNA barcoding to identify Encephalartos spp. rapidly when morphological evidence is lacking. Furthermore, the technique can be of great assistance in monitoring the trade in Encephalartos spp., specifically in relation to conservation and law enforcement interventions that may be implemented in the future.
Fig. 4.
Acknowledgements
We would like to thank the Government of Canada through Genome Canada and the Ontario Genomics Institute (2008-OGI-ICI-03), the Consortium for the Barcode of Life (CBOL) through the Google’s Global Impact Award Programme, and the University of Johannesburg (Analytical Facility) for financial support and various local authorities who granted plant collection permits (Cape Nature 0028-AAA008-00171; Department of Environmental Affairs, ToPs Certificate Number 07138 and ToPs permit number 03009; Eastern Cape ANO 3/5/4 – M. van der Bank; Gauteng 0039; Limpopo 0090-MKT001-00009; Mpumalanga MPB.1305). Zandisile Eugene Shongwe is thanked for technical assistance in the laboratory and the late Philip Rousseau for providing sequence data for the Encephalartos DNA reference library.
Footnote
2
Supplementary data are available with the article through the journal Web site at Supplementary Material.
References
Arun Dev S., Muralidharan E.M., Sujanapal P., and Balasundaran M. 2014. Identification of market adulterants in East Indian sandalwood using DNA barcoding. Ann. For. Sci. 71(4): 517–522.
Baker D.A., Stevenson D.W., and Little D.P. 2012. DNA barcode identification of black cohosh herbal dietary supplements. J. AOAC Int. 95: 1023–1034.
Bogler D.J. and Francisco-Ortega J. 2004. Molecular systematic studies in cycads: Evidence from trnL intron and ITS2 rDNA sequences. Bot. Rev. 70: 260–273.
Botha J., Witkowski E.T.F., and Shackleton C. 2004. Market profiles and trade in medicinal plants in the Lowveld, South Africa. Environ. Conserv. 31(1): 38–46.
Brenner E.D., Stevenson D.W., and Twigg R.W. 2003. Cycads: evolutionary innovations and the role of plant-derived neurotoxins. Trends Plant Sci. 8: 446–452.
CBOL Plant Working Group. 2009. A DNA barcode for land plants. Proc. Natl. Acad. Sci. U.S.A. 106(31): 12794–12797.
Chaw S.M., Walters T.W., Chang C.C., Hu S.H., and Chen S.H. 2005. A phylogeny of cycads (Cycadales) inferred from chloroplast matK gene, trnK intron and nuclear rDNA ITS region. Mol. Phylogenet. Evol. 37: 214–234.
Clement W.L. and Donoghue M.J. 2012. Barcoding success as a function of phylogenetic relatedness in Viburnum, a clade of woody angiosperms. BMC Evol. Biol. 12: 73–86.
Cousins S.R., Williams V.L., and Witkowski E.T.F. 2011. Quantifying the trade in cycads (Encephalartos species) in the traditional medicine markets of Johannesburg and Durban, South Africa. Econ. Bot. 65(4): 356–370.
Cousins S.R., Williams V.L., and Witkowski E.T.F. 2012. Uncovering the cycad taxa (Encephalartos species) traded for traditional medicine in Johannesburg and Durban, South Africa. S. Afr. J. Bot. 78: 129–138.
Cousins S.R., Williams V.L., and Witkowski E.T.F. 2013. Sifting through cycads: A guide to identifying the stem fragments of six South African medicinal Encephalartos species. S. Afr. J. Bot. 84: 115–123.
Crouch N.R., Smith G.F., Lötter M., and Symmonds R. 2003. Encephalartos woodii — the first historically documented ethnomedicinal plant casualty in southern Africa. Encephalartos, 73: 25–28.
Cunningham, A.B., and Davis, G.W. 1997. Human use of plants. In Vegetation of southern Africa. Edited by R.M. Cowling, D.M. Richardson, and S.M. Pierce. University Press, Cambridge. pp. 474–501.
Dold A.P. and Cocks M.L. 2002. The trade in medicinal plants in the Eastern Cape Province, South Africa. S. Afr. J. Sci. 98: 589–597.
Donaldson, J.S. (Editor) 2003. Cycads Status Survey and Conservation Action Plan. IUCN, Gland, Switzerland.
Donaldson, J.S. 2008. South African Encephalartos species. NDF Workshop Case Studies: Case Study 4: Encephalartos. Mexico.
Donaldson, J.S. 2010a. Encephalartos villosus. The IUCN Red List of Threatened Species 2010: e.T41947A10608666. [accessed 30 January 2016.]
Donaldson, J.S. 2010b. Encephalartos natalensis. The IUCN Red List of Threatened Species 2010: e.T41937A10605143. [accessed 30 January 2016.]
Donaldson, J.S. 2010c. Encephalartos ferox. The IUCN Red List of Threatened Species 2010: e.T41943A10607271. [accessed 30 January 2016.]
Donaldson, J.S. 2010d. Encephalartos lebomboensis. The IUCN Red List of Threatened Species 2010: e.T41907A10589133. [accessed 20 January 2016.]
Donaldson, J.S. 2010e. Encephalartos senticosus. The IUCN Red List of Threatened Species 2010: e.T41925A10600018. [accessed 29 January 2016.]
Donaldson, J.S. 2011. An overview of cycad conservation based on the 2010 Global Cycad Assessment and its implications for the IUCN/SSC Cycad Action Plan. Conference Abstract from the 9th International Conference on Cycad Biology, Shenzhen, China Journal of the Fairylake Botanical Garden, 10(3–4): 47.
Donaldson, J.S., and Bösenberg, J.D. 1999. Changes in the abundance of South African cycads during the 20th century: preliminary data from the study of matched photographs. In Biology and conservation of cycads. Edited by C.J. Chen. Proceedings of the Fourth International Conference on Cycad Biology (Panzhihua, China), International Academic Publishers, Beijing.
Doyle J.J. and Doyle J.L. 1987. A rapid isolation procedure for small amounts of leaf tissue. Phytochem. Bull. 19: 11–15.
Edgar R.C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucl. Acids Res. 32: 1792–1797.
Farris J.S. 1972. Estimating phylogenetic trees from distance matrices. Am. Nat. 106(951): 645–668.
Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution, 39: 783–791.
Golding J.S. and Hurter P.J.H. 2003. A Red List account of Africa’s cycads and implications of considering life-history and threats. Biodivers. Conserv. 12: 507–528.
Grobbelaar, N. 2004. Cycads: with special reference to the southern African species. [Published by the author, Pretoria.]
Hill K.D., Chase M.W., Stevenson D.W., Hills H.G., and Schutzman B. 2003. The families and genera of cycads: a molecular phylogenetic analysis of Cycadophyta based on nuclear and plastid DNA sequences. Int. J. Plant Sci. 164: 933–948.
Hollingsworth M.L., Clark A., Forrest L.L., Richardson J., Pennington R.T., Long D.G., et al. 2009. Selecting barcoding loci for plants: evaluation of seven candidate loci with species-level sampling in three divergent groups of land plants. Mol. Ecol. Resour. 9: 439–457.
Hollingsworth P.M., Graham S.W., and Little D.P. 2011. Choosing and using a plant DNA barcode. PLoS ONE, 6: e19254.
Huelsenbeck J.P. and Ronquist F. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics, 17: 754–755.
Kool A., de Boer H.J., Kruger Å., Rydberg A., Abbad A., Björk L., and Martin G. 2012. Molecular identification of commercialized medicinal plants in southern Morocco. Plos ONE, 7: e39459.
Kress J. and Erickson D.L. 2007. A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH-psbA spacer region. PLoS ONE, 6: e508.
Levin R.A., Wagner W.L., Hoch P.C., Nepokroeff M., Pires J.C., Zimmer E.A., and Sytsma K.J. 2003. Family-level relationships of Onagraceae based on chloroplast rbcL and ndhF data. Am. J. Bot. 90: 107–115.
Little D.P. 2011. DNA barcode sequence identification incorporating taxonomic hierarchy and within taxon variability. PLoS ONE, 6: e20552.
Little D.P. and Stevenson D.W. 2007. A comparison of algorithms for identification of specimens using DNA barcodes: examples from gymnosperms. Cladistics, 23: 1–21.
Mander, M. 1998. Marketing of indigenous medicinal plants in South Africa: a case study in KwaZulu-Natal. Food and Agriculture Organization of the United Nations, Rome.
Mander, M., Ntuli, L., Diederichs, N., and Mavundla, K. 2007. Economics of the traditional medicine trade in South Africa. In South African Health Review 2007. Edited by S. Harrison, R. Bhana, and A. Ntuli. Health Systems Trust, Durban.
Miller, M.A., Holder, M.T., Vos, R., Midford, P.E., Liebowitz, T., Chan, L., et al. 2009. The CIPRES portals. Available from www.phylo.org/sub_sections/portal.
NEM:BA. 2004. National Environmental Management: Biodiversity Act (No. 10 of 2004). 471 Government Gazette No. 32090, 3 April 2009. Department of Environmental Affairs and Tourism, Pretoria.
Newmaster S.G., Grguric M., Shanmughanandhan D., Ramalingam S., and Ragupathy S. 2013. DNA barcoding detects contamination and substitution in North American herbal products. BMC Med. 11: 222–235.
Nicolalde-Morejón F., Vergara-Silva F., González-Astorga J., Stevenson D.W., Vovides A.P., and Sosa V. 2011. A character-based approach in the Mexican cycads supports diverse multigene combinations for DNA barcoding. Cladistics, 27: 150–164.
Norstog, K.J., and Nicholls, T.J. 1997. The biology of the cycads. Cornell University Press, Ithaca.
Palumbi, S.R. 1996. Nucleic acids II: The polymerase chain reaction. In Molecular systematics. 2nd ed. Edited by D.M. Hillis, C. Moritz, and B.K. Mable. Sinauer, Sunderland, Mass. pp. 241–246.
Posada D. and Crandall K.A. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics, 14: 817–818.
Potgieter, D., and Cresswell, R. 1989. Eskom slammed as cycad war hots up. Press release. Sunday Times, October 15th: 4.
Pryer K.M., Schuettpelz E., Huiet L., Grusz A.L., Rothfels C.J., Avent T., et al. 2010. DNA barcoding exposes a case of mistaken identity in the fern horticultural trade. Mol. Ecol. Resour. 10(6): 979–985.
Ratnasingham S. and Hebert P.D.N. 2007. BOLD: The Barcode of Life Data System (www.barcodinglife.org). Molecular Ecology Notes, 7: 355–364.
Retief K., West A.G., and Pfab M.F. 2014. Can stable isotopes and radiocarbon dating provide a forensic solution for curbing illegal harvesting of threatened cycads? J. Forensic Sci. 59(6): 1541–1551.
Ronquist F. and Huelsenbeck J.P. 2003. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19: 1572–1574.
Rousseau, P. 2012. A molecular systematic study of the African endemic cycads. M.Sc. thesis, University of Johannesburg, Johannesburg, South Africa, 176 p.
Sangin P., Forster P.I., Mingmuang M., and Kokubugata G. 2008. A phylogeny for two cycad families (Stangeriaceae and Zamiaceae) based on chloroplast DNA sequences. Bull. Natl. Mus. Nat. Sci. Ser. B Bot. 34: 75–82.
Sass C., Little D.P., Stevenson D.W., and Specht C.D. 2007. DNA barcoding in the cycadales: testing the potential of proposed barcoding markers for species identification of cycads. PLoS ONE, 2: e1154.
Savolainen V., Cuenoud P., Spichiger R., Martinez M.D.P., Crevecoeur M., and Manen J.F. 1995. The use of herbarium specimens in DNA phylogenetics — evaluation and improvement. Plant Syst. Evol. 197: 87–98.
Srirama R., Senthilkumar U., Sreejayan N., Ravikanth G., Gurumurthy B.R., Shivanna M.B., et al. 2010. Assessing species admixtures in raw drug trade of Phyllanthus, a hepato-protective plant using molecular tools. J. Ethnopharmacol. 130: 208–215.
Subedi A., Kunwar B., Choi Y., Dai Y., van Andel T., Chaudhary R.P., et al. 2013. Collection and trade of wild-harvested orchids in Nepal. J. Ethnobiol. Ethnomed. 9(1): 64.
Sugiura N. 1978. Further analysis of the data by Akaike’s information criterion and the finite corrections. Commun. Stat. 7: 13–26.
Sun Y., Skinner D.Z., Liang G.H., and Hulbert S.H. 1994. Phylogenetic analysis of Sorghum and related taxa using internal transcribed space of nuclear ribosomal DNA. Theor. Appl. Genet. 89: 26–32.
Swofford, D.L. 2002. PAUP*: phylogenetic analysis using parsimony (* and other methods), version 4.0b10. Sinauer Associates, Sunderland, Mass.
Treutlein J. and Wink M. 2002. Molecular phylogeny of cycads inferred from rbcL sequences. Naturwissenschaften, 89: 221–225.
Van Velzen R., Weitschek E., Felici G., and Bakker F.T. 2012. DNA barcoding of recently diverged species: relative performance of matching methods. PLoS ONE, 7(1): e30490.
White, T.J., Bruns, T., Lee, S., and Taylor, J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR Protocols: a guide to methods and applications. Edited by M.A. Innis, D.H. Gelfand, J.J. Sninsky, and T.J. White. Academic Press, New York. pp. 315–322.
Williams, V.L. 2003. Hawkers of Health: The Faraday Street Medicinal Plant Market in Johannesburg. Available from http://wiredspace.wits.ac.za/bitstream/handle/10539/5313/Faraday%20Report.pdf.
Williams V.L., Balkwill K., and Witkowski E.T.F. 2007a. Size-class prevalence of bulbous and perennial herbs sold in the Johannesburg medicinal plant markets between 1995 and 2001. S. Afr. J. Bot. 73: 144–155.
Williams V.L., Witkowski E.T.F., and Balkwill K. 2007b. Volume and financial value of species traded in the medicinal plant markets of Gauteng, South Africa. Int. J. Sust. Dev. World Ecol. 14: 584–603.
Williams V.L., Cousins S.R., and Witkowski E.T.F. 2014. From fragments to figures: estimating the number of Encephalartos stems in a muthi market. S. Afr. J. Bot. 93: 242–246.
Zgurski J.M., Rai H.S., Fai Q.M., Bogler D.J., Francisco-Ortega J., and Graham S.W. 2008. How well do we understand the overall backbone of cycad phylogeny? New insights from a large, multigene plastid data set. Mol. Phylogenet. Evol. 47: 1232–1237.
Supplementary Material
Supplementary data (gen-2016-0032suppla_data file s1.pdf)
- Download
- 604.45 KB
Supplementary data (gen-2016-0032supplb_table s1.docx)
- Download
- 32.68 KB
Supplementary data (gen-2016-0032supplc_table s2.docx)
- Download
- 13.88 KB
Supplementary data (gen-2016-0032suppld_table s3.docx)
- Download
- 25.35 KB
Supplementary data (gen-2016-0032supple_table s4.docx)
- Download
- 12.82 KB
Information & Authors
Information
Published In
Genome
Volume 59 • Number 9 • September 2016
Pages: 771 - 781
Editor: Sarah Adamowicz
History
Received: 2 February 2016
Accepted: 13 May 2016
Accepted manuscript online: 13 July 2016
Version of record online: 13 July 2016
Notes
This paper is part of a special issue entitled Barcodes to Biomes.
Copyright
© 2016.
Key Words
Mots-clés
Authors
Metrics & Citations
Metrics
Other Metrics
Citations
Cite As
J. Williamson, O. Maurin, S.N.S. Shiba, H. van der Bank, M. Pfab, M. Pilusa, R.M. Kabongo, and M. van der Bank. 2016. Exposing the illegal trade in cycad species (Cycadophyta: Encephalartos) at two traditional medicine markets in South Africa using DNA barcoding. Genome.
59(9): 771-781. https://doi.org/10.1139/gen-2016-0032
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
1. Encephalartos natalensis
association with non-cyanobacterial species increased reliance on atmospheric nitrogen in acidic and nutrient-deficient savanna woodland ecosystem soils
2. Effects of trade and poaching pressure on extinction risk for cacti in the Atacama Desert
3. Sex determination of South Africa's Encephalartos - A conservation perspective
4. Leafing through genetic barcodes: An assessment of 14 years of plant DNA barcoding in South Africa
5. Automated ISSR Fingerprinting Is a Cost-Effective Way to Assess Genetic Diversity and Taxonomic Differentiation—A Case Study on the Encephalartos eugene-maraisii Species Complex
6. The Greater Midlands—A Mid-Elevation Centre of Floristic Endemism in Summer-Rainfall Eastern South Africa
7. Encephalartos villosus associated bacterial communities and extracellular enzymes improve soil nutrition in rhizosphere soils in forest ecosystem soils
8. Assessment of population, habitat, and threats to Cycas pectinata Buch.-Ham. (Cycadaceae), a vulnerable cycad in Bhutan
9. Species relationships and phylogenetic diversity of the African genus Encephalartos Lehm. (Zamiaceae)
10. Contemporary Cultural Trade of Lion Body Parts
11. Retracing origins of exceptional cycads in botanical collections to increase conservation value
12. Progress in the use of DNA barcodes in the identification and classification of medicinal plants
13. Advancement of analytical techniques in some South African commercialized medicinal plants: Current and future perspectives
14. Ten years of barcoding at the African Centre for DNA Barcoding
15. From Barcodes to Biomes: Special Issues from the 6th International Barcode of Life Conference
16. From Barcodes to Biomes: Special Issues from the 6th International Barcode of Life Conference
View Options
View options
Login options
Check if you access through your login credentials or your institution to get full access on this article.