Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
1
Article
3
Trade in Zambian edible orchids – DNA barcoding
reveals use of unexpected orchid taxa for chikanda.
4
5
6
Veldman, S.1*, Kim, S.-J.1, van Andel, T.R.2, Bello Font, M.3, Bone, R.E.4, Bytebier, B.5, Chuba, D.6,
Gravendeel, B.2,7,8, Martos, F.5,9, Mpatwa, G.10, Ngugi, G.5,11, Vinya, R.10, Wightman, N.12, Yokoya,
K.4 and de Boer, H.J.1,2,
2
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Department of Organismal Biology, Systematic Biology, Uppsala University, Norbyvägen 18D, 75236
Uppsala, Sweden; sarina.veldman@ebc.uu.se; seoljong.kim@gmail.com;
2 Naturalis Biodiversity Center, P.O. Box 9517, 2300 RA Leiden, the Netherlands.;
Tinde.vanAndel@naturalis.nl;
3 Natural History Museum, University of Oslo, Postboks 1172, Blindern, 0318 Oslo, Norway.;
mariabellofont@hotmail.com; h.d.boer@nhm.uio.no;
4 Royal Botanic Gardens, Kew, Richmond, Surrey, TW9 3AB, UK.; R.Bone@kew.org; K.Yokoya@kew.org;
5 Bews Herbarium, School of Life Sciences, University of KwaZulu-Natal, Pr. Bag X01, 3209 Scottsville,
South Africa ; Bytebier@ukzn.ac.za;
6 Department of Biological Sciences, University of Zambia, Box 32379, Lusaka, Zambia.;
david.chuba@unza.zm;
7 Institute of Biology Leiden, Leiden University, P.O. Box 9505, 2300 RA Leiden, the Netherlands;
barbara.gravendeel@naturalis.nl;
8 University of Applied Sciences Leiden, Zernikedreef 11, 2333 CK Leiden, The Netherlands
9 Institut de Systématique, Evolution, Biodiversité (ISYEB), Muséum national d’histoire naturelle, CNRS,
Sorbonne Université, EPHE; CP50, 45 rue Buffon 75005 Paris, France; florentmartos@gmail.com;
10 School of Natural Resources, The Copperbelt University, PO Box 21692, Kitwe, Zambia;
gmpatwa@gmail.com; royd.vinya@gmail.com;
11 East African Herbarium, National Museums of Kenya, P.O. Box 40658-00100, Nairobi, Kenya;
grace.ngugi@yahoo.com;
12 Homegarden Landscape Consultants Ltd., P/Bag 30C, Chilanga, Lusaka, Zambia;
homegarden.nicholas@gmail.com;
* Correspondence: sarina.veldman@ebc.uu.se; Tel.: +46737828087
1
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Abstract: In Zambia wild edible terrestrial orchids are used to produce a local delicacy called
chikanda, which has become increasingly popular throughout the country. Commercialization puts
orchid populations in Zambia and neighbouring countries at risk of overharvesting. Hitherto, no
study has documented which orchid species are traded on local markets, as orchid tubers are
difficult to identify morphologically. In this study, the core land-plant DNA barcoding markers rbcL
and matK were used in combination with nrITS to determine which species were sold on Zambian
markets. Eighty-two interviews were conducted to determine harvesting areas, as well as possible
sustainability concerns. By using nrITS DNA barcoding, a total of 16 orchid species in six different
genera could be identified. Both rbcL and matK proved suitable to identify the tubers up to genusor family level. Disa robusta, Platycoryne crocea and Satyrium buchananii were identified most
frequently and three previously undocumented species were encountered on the market. Few
orchid species are currently listed on the global IUCN Red List. Local orchid populations and
endemic species could be at risk of overharvesting due to the intensive and indiscriminate
harvesting of chikanda orchids and we therefore encourage increased conservation assessment of
terrestrial African orchids.
46
47
Keywords: CITES; Chikanda; Conservation; DNA barcoding; Orchids; Species delimitation;
© 2018 by the author(s). Distributed under a Creative Commons CC BY license.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
48
1. Introduction
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
Terrestrial orchids have been used for medicinal and culinary purposes for centuries [1], with
the most notable example being the use of orchid tubers to make salep, a traditional Turkish creamy
drink or ice cream, consumed in Asia Minor and several countries on the Balkan peninsula [2–5]. In
south-eastern Africa terrestrial orchid tubers are mixed with peanut flour, salt, baking soda and
chili powder to make a traditional Zambian meat-like cake known as chikanda or African polony [6–
8]. Although initially not highly regarded [9], chikanda has more recently become popular
throughout the country. It is sold as a snack along the streets, on markets, in supermarkets and on
the menu of high-end restaurants [10] and recipes as well as cooking tutorial videos can be found
online [11]. Orchids used for chikanda are harvested exclusively from the wild, and although it is
unlikely that traditional village consumption poses a serious threat to orchid populations, the
increased popularity and subsequent commercialization of chikanda has led to the exhaustion of
Zambian orchid resources [8]. Collecting tubers means the end of a perennial and generally longlived orchid, since the entire plant is removed in the harvesting process.
Soweto market in the Zambian capital Lusaka is the hub of the chikanda trade. Surveys
performed on this market have shown that a large part of the chikanda tubers sold are sourced from
Tanzania and that Zambian chikanda orchids are collected from the Luwingu, Mporokoso and
Kasama districts in the Northern Province and Serenje in the Central province [7,8]. According to
local chikanda vendors, another region with a flourishing chikanda trade is the Kitwe region in the
Copperbelt Province, but so far no surveys have been performed there. Despite international
legislation (CITES) banning prohibiting cross-border trade, an estimated 2–4 million orchid tubers
are transported annually from Tanzania to Zambia [8,12]. Import from the surrounding countries of
Angola, DRC, Malawi and Mozambique is also documented [8,10]. Orchid species originally
reported as ingredients for chikanda are Disa robusta N.E.Br. and Satyrium buchananii Schltr. [12,13],
whereas recently at least 32 species belonging to the genera Brachycorythis, Disa, Eulophia, Habenaria,
Roeperocharis and Satyrium were suggested to be used for chikanda production based on collections
in the field [12,14–19] and one metabarcoding study of ready-made chikanda cakes [19]. To date,
however, no study identified the orchids traded at the local markets, since the tubers lack
sufficient morphological characters for taxonomic identification to species level [8,19]. Local
classification systems categorize the tubers based on texture, harvesting locality, soil colour and
phenology, but these are not likely to be congruent with scientific classifications [6,15].
Knowing which orchid species are currently being collected for the expanding chikanda trade
enables the identification of species susceptible to overharvesting and can inform conservation
planning . Molecular methods such as DNA barcoding can be applied to identify samples when
morphological diagnostic characters for identification are lacking [20]. DNA barcoding and
metabarcoding has proven to be effective in the authentication of commercial wood species (Jiao,
2018), medicinal plants [21] and salep-producing orchids on Iranian markets [5]. The analysis of
ingredients in Tanzanian chikanda cake with DNA metabarcoding revealed the presence of 21
different orchid species [19], but a DNA barcoding approach has not yet been applied to individual
orchid tubers used to make this product. The aim of this study was to test to what extent species
delimitation using standard molecular markers yields robust identification of chikanda orchid tubers
traded on Zambian markets. Molecular identification can enable mapping of harvesting and trade
of specific Zambian orchid species and facilitate identification and implementation of targeted
conservation strategies. Within that framework this study also aimed to identify conservation issues
associated with the chikanda trade, and addresses the following questions: 1) Which species are used
for chikanda production in the Lusaka and Kitwe districts of Zambia and what is their geographic
origin? 2) Can chikanda tubers be identified up to species level using DNA barcoding? 3) How do
local classification systems relate to scientific species concepts? 4) What are the main conservation
issues associated with chikanda trade in the Lusaka and Kitwe districts?
97
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
98
2. Materials and Methods
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
2.1. Interviews and sample collection
Fieldwork in Zambia was conducted in 2016 in the Kitwe, Kalulushi, Luanshya, Ndola,
Mufulira, Chingola and Chililabomwe districts of the Copperbelt Province; the Kapiri Mposhi and
Serenje districts in the Central Province and in the capital Lusaka (Figure 1). Semi-structured
interviews were conducted with harvesters, middlemen and vendors to obtain insight in chikanda
commercialization, harvesting times, preferences and availability. The questionnaires consisted of
three sections, one on informant and interview characteristics, one with general questions about
chikanda posed to all informants, and a third section with questions more specifically designed for
each interviewee category: harvester, middleman and vendor. All research was conducted in
accordance with the International Society of Ethnobiology Code of Ethics [22]. Ethical clearance was
obtained from the Humanities and Social Science Research Ethics Committee of the University of
Zambia. The interviews were performed in English or Bemba, with a translator affiliated with the
University of Zambia and the Copperbelt University. Informants were selected using the snowball
technique [23], by asking people whether they could direct us to people harvesting or selling
chikanda. All informants were provided with information about the study and signed a prior
informed consent sheet. Fieldwork took place during June and July, the peak season for chikanda [8],
to ensure collection of both fresh and dried chikanda tubers on the market and in the field. A
collection was made each time a specific vernacular chikanda type was bought from a specific
vendor, and assigned a collection number (SJK1, SJK2, etc.). Each individual tuber within the
collection received a subsample number within that collection (SJK1.1, SJK1.2, SJK 1.16, etc.). The
fresh tubers were sliced and stored with silica gel in plastic bags. All chikanda samples were brought
to Sweden under a CITES inter-institutional exchange agreement between the University of Zambia
(ZM001) and the Botany Section of the Evolutionary Museum in the Evolutionary Biology Center in
Uppsala (SE009). Export permission was obtained from the Zambian CBD and Nagoya Protocol
focal point at the Ministry of Natural Resources and Environmental Protection. Upon arrival to
Sweden some of the chikanda tubers had sprouted. Those were transferred to the Uppsala Botanical
Garden for cultivation and subsequent sampling of fresh leaf tissue for DNA barcoding as well as
morphological identification.
127
128
129
130
Figure 1. Overview of interview localities and reported provenance of the chikanda tubers. Dot size
corresponds to the number of informants.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
131
132
133
134
135
136
137
138
139
140
141
142
143
144
2.2. Reference taxon sampling
Herbarium specimens were collected with associated silica-dried material for DNA extraction
and spirit collections during fieldwork in Zambia in January and February 2017. In addition,
putative orchid mycorrhizal fungi were sampled from roots and tubers for isolation, culture and
identification at RBG Kew (results not reported here). All material was collected and exchanged in
accordance with national and international legislation. The collections are deposited at the Division
of Forest Research (Kitwe, Zambia) and RBG Kew (UK) and field identifications verified at the
Bews Herbarium (South Africa). A total of 94 novel Orchidaceae reference vouchers were collected
for this study, representing 4 Brachycorythis, 9 Disa, 16 Habenaria, 6 Satyrium species and 26 species
in other orchid genera. Voucher specimens of all taxa sampled are listed in Table S1
(Supplementary Material). In addition, 88 ITS, 71 matK and 45 rbcL Habenaria sequences generated
for a forthcoming phylogenetic study [24], and 510 ITS, 522 matK and 213 rbcL sequences
corresponding to 311, 325 and 100 taxa in the previously mentioned orchid genera downloaded
from GenBank were included in the reference database.
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
2.3. From sample to sequence
Out of the 1284 individual tubers in 48 different sample collections, 304 samples were selected
for DNA extraction. A few tubers per sample (2-8) were extracted if the sample was
morphologically homogenous, whereas more (6-33) were selected if the sample was diverse. DNA
was extracted using a CTAB protocol [25] modified with 3 to 5 extra washing steps with STE buffer
(0.25 M sucrose, 0.03 M Tris, 0.05 M EDTA) [26,27], to reduce the gelatinization effect of the large
amount of polysaccharides in the starch-rich orchid tubers. Total DNA was stored in 70-100 μl
10mM Tris-HCl buffer, pH 8.0. DNA concentration was measured with a Qubit 3.0 fluorometer
(Thermo Fisher Scientific, Oakwood, USA). The core land plant barcoding markers rbcL and matK
were amplified using the primers and protocols described in Kress et al. [28] and Dunning and
Savolainen [29] respectively. The reactions were performed in a total reaction volume of 25μl with
14.725μl ddH2O, 2.5μl DreamTaq Buffer (Thermo Fisher Scientific, Oakwood, USA), 0.5μl 25mM
dNTP, 0.65μl 2% Bovine Serum Album (BSA), 0.125μl DreamTaq Polymerase, 2.5μl 5pmol forward
and reverse primer and 1.5μl template DNA. Nuclear ribosomal nrITS was amplified using the Sun
et al. [30] primers and protocol in a total reaction volume of 25μl containing 15.25μl ddH2O, 2.5μl
DreamTaq Buffer (Thermo Fisher Scientific, Oakwood, USA), 0.5μl 25mM dNTP, 0.125μl 2% BSA,
0.125μl DreamTaq Polymerase (Thermo Fisher Scientific, Oakwood, USA), 2.5μl 5pmol forward and
reverse primer and 1.5μl template DNA. For ITS an additional protocol was used with Q5 highfidelity polymerase: reactions were performed in a total reaction volume of 23.5μl including
10.875μl ddH2O, 5μl Q5 reaction buffer, 0.5μl 25mM dNTP, 5μl Q5 GC enhancer, 0.125μl Q5 highfidelity polymerase, 1.5μl 5pmol forward and reverse primer and 0.5μl template DNA. The PCR
program for the ITS primers in combination with the Q5 polymerase was an initial heating step of
30s at 98°C; 35 cycles of 10s 98°C, 30s 56°C, 30s 72°C; and a final elongation of 2min at 72°C. PCR
products were cleaned using eight times diluted ExoSAP (Thermo Fisher Scientific, Oakwood,
USA) and analysed on an ABI3730XL Sanger sequencer by Macrogen Europe (Amsterdam, The
Netherlands). The obtained trace files were assembled using Pregap4 and Gap4 [31] as
implemented in the Staden package [32]. Sequences shorter than 200 bp were discarded from the
analysis and all sequences have been deposited in NCBI GenBank (Table S2, Supplementary
Material). The NCBI BLAST algorithm was used to assess the identification of all the obtained
query sequences using the Python BLAST tabular parser script
(https://github.com/SLAment/Genomics/blob/master/BLAST/BLAST_tabularparser.py). Similarity
scores, query coverage, expect value (E-value), and max identity percentage were calculated, and all
the information of the top 5 hits were automatically mined and tabulated per marker (Table S3-S5,
Supplementary Material), and summarized per individual tuber sample (Table S6, Supplementary
Material).
180
181
2.4. Phylogenetic analysis and species delimitation
Alignments for nrITS, matK and rbcL were made using AliView [33], combing the query
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
sequences and local reference databases consisting of sequences from NCBI GenBank and reference
collections from fieldwork and unpublished data from collaborators [33]. Species delimitation was
performed using the Poisson Tree Processes (PTP) model, as it has been shown to outperform the
Generalized Mixed Yule Coalescent (GMYC) approach as well as OTU-picking methods when
evolutionary distances between species are short enough [34]. For all alignments a maximum
likelihood (ML) search for the best-scoring tree was performed using the RAxML web server [35] to
generate input trees for species delimitation analysis using bPTP [34]. GTRCATI was used to
implement the CAT approximation, and the final tree was evaluated using the traditional GTR
model. The bPTP.py script settings was 100,000 MCMC chain iterations for all trees; sampling
interval thinning value of 100; burn-in of 25%; and a random seed of 1234. No outgroup was
defined. The generated convergence curve was visually confirmed. The description of each voucher
is built up as following: 'sample #_BLAST search result_identification %_reported city/region of the
origin_the country,'; for example: ‘SJK04.09_S.buchananii_97.893_Mwinilunga_Zambia’ (Figure S13, Supplementary Material). For the reference sequences, accession number and species names are
described for Disa and Satyrium. The reference Habenaria species and others are described as species
name and voucher (sample) number.
198
3. Results
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
3.1. Market surveys and interviews
We visited 25 markets in ten Zambian cities: seven cities in the Copperbelt Province, two in the
Central Province, and one in Lusaka (Figure 1). Eighty-two persons involved in chikanda trade were
interviewed of which 8 were harvesters, 44 middlemen, 29 vendors and one informant was both
harvester and middleman. The term ‘harvesters’ refers to people collecting chikanda in the field,
‘middlemen’ to people selling either dried or fresh chikanda orchid tubers on the market, and
‘vendors’ to informal street vendors selling ready-made chikanda cakes contained in a basket carried
on the head. Since one of the harvesters also acted as middleman, in total 83 individual interviews
were conducted: 72 participants were female and 10 male. The ages of the respondents varied
between 18 and 78, with an average of 41 years. The majority of the respondents belonged to the
Bemba tribe (62%), while the other the respondents (38%) belonged to smaller ethnic groups. Most
interviews were conducted in the Copperbelt Province (55), eight in the Central Province and 19 in
Lusaka.
3.2. Local classification system
Fifty-three different vernacular names for the various chikanda types were recorded during the
interviews. The most common way to distinguish between chikanda tubers was by using the terms
original (myala) and fake (mbwelenge or msekelele). In some cases it was the shape of the tuber that was
used to differentiate between the different tubers: mshilamshila means root-like in Bemba and referred
to the elongated, root-shaped tubers whereas mampanda referred to the heart-shaped tubers. It also
appeared common to use the origin of the tuber as a trade name: mwinilunga, chozi, luwingu and
kasama are for example all Zambian city names, sumbawanga and iringe are referring to Tanzanian
cities (Sumbawanga and Iringa) and angola refers to one of the countries bordering Zambia. In some
cases the chikanda tubers were sold pre-mixed, whereas other vendors marketed the different types
of tubers separately. The morphology and size of the tubers varied both within and between
collections of a certain chikanda type. Tubers could be heart-shaped, rounded, egg-shaped to
elongated and almost root-like. The largest tubers were the heart-shaped ones, which could be up to
5 cm long and 6 cm wide. The elongated tubers were up to 9 cm long and maximally 2 cm wide.
Harvesters, middlemen and vendors themselves indicated that they distinguish the tubers based on
the size of the granules inside the tubers, which can be large or small and in some tubers a concentric
ring was said to be present. Figure 2 illustrates the different tuber types and ready-made chikanda that
were encountered on the local markets, as well as some of the orchid species producing these tubers.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
233
234
235
236
237
238
239
Figure 2. Chikanda tubers, cake and orchids. (a) Myala – real chikanda; (b) Mbwelenge – fake chikanda;
(c) Mshilamshila – supposedly Brachycorytis sp.; (d) Mampanda. (e-g); Chikanda cake; (h) Disa robusta;
(i) Disa welwitschii; (j) Platycoryne crocea; (k) Satyrium carsonii; (l) Satyrium buchananii; (m) Satyrium
kitimboense; (n) Brachycorytis cf. friesii; Photographs (a-g) by Seol-Jong Kim, (h) by Robert v.
Blittersdorff, (i,k and l) by Nicholas Wightman, (j) by Warren McCleland, (m) by Ruth E. Bone and
(n) by Sarina Veldman.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
3.3. Chikanda trade and availability
Many participants had a relatively long experience (average 13.5 year) in the chikanda trade. They
indicated that they were asked by family, friends or neighbors to get involved in the chikanda
business, or simply because it seemed a profitable industry. Some of the participants, especially
harvesters, stated that they started collecting chikanda because the plants were easily accessible and
were found growing close to their areas of residence. In general, people involved in chikanda trade
indicated that this work alone did not suffice to fully support themselves and their families, and these
traders therefore supplemented their income by trading additional natural products such as fruits,
maize, groundnuts, beans, mushrooms, snacks, herbs, and kapenta (a dried fish from Lake
Tanganyika). In Lusaka, however, income generated with chikanda trade seemed to be sufficient for
subsistence, and this suggests that there are significant local differences in profits generated with
chikanda business.
The chikanda trade was structured in several different ways. Some of the middlemen loaded the
orchid tubers in trucks directly from the harvesting areas and brought them to larger cities such as
Lusaka and Kitwe, whereas others relied on agents to gather a certain amount of chikanda tubers,
which they paid for through the East-African mobile payment system M-Pesa and received as cargo
from one of the local buses. In addition to the Tunduma and Nakonde markets that are the tradehubs between Tanzania and Zambia, other centers of trade were identified on the border with the
Democratic Republic of Congo (Chililabombwe and Kasumbalesa) and Angola (Mwinilunga), where
both chikanda tubers and ready-made chikanda cake was sold. An overview of all interview localities
and the reported provenance of the chikanda tubers is given in Figure 1. Most of the participants
indicated that chikanda plants are becoming locally rare. Middlemen emphasized the decrease of
quantity, whereas harvesters were concerned about the decline in both quantity and quality (size and
preferred chikanda type), which may also depend on the species. The participants from urban areas
stated that access to chikanda tubers was managed by the chief of each tribe, who seasonally
designated the harvestable dambo (wet meadow) within the chiefdom so that the collectors could
maintain the quality of the harvests. Nevertheless, interviewed harvesters encountered in the dambo
areas of Serenje (Central Province) claimed to be free to harvest tubers whenever available.
3.4. Molecular identification of traded orchids
During the DNA extraction many samples formed a thick jelly-like layer in the extraction tubes,
despite repeated washing steps with STE buffer. This resulted in a very small water phase and likely
negatively influenced the downstream steps of the extraction process. The average DNA
concentration of the chikanda tuber samples was 4.96 ng/μl, while the average DNA concentration
from leaf samples from chikanda orchids was 28.5 ng/μl. Out of the 304 samples selected for DNA
extraction, 232 samples produced detectable DNA. Amplification was attempted for each barcoding
marker for all of the samples. A nrITS sequence was obtained for 159 samples, rbcL for 117 samples
and matK only for 45 samples. Sequences from all three markers were obtained for 40 samples,
sequences from two markers for 58 samples, and 55 samples only yielded sequences for a single
marker. Analysis of the inter- and intraspecific variation of nrITS was performed for 141 sequences
representing 124 Disa species, 73 sequences representing 59 Habenaria species and 110 sequences
representing 67 Satyrium species. In the case of matK 135 sequences belonging to 122 Disa species were
included, 507 Habenaria sequences belonging to 239 species and 116 sequences belonging to 60
Satyrium species. For rbcL the available reference material was quite limited: 45 sequences for 37
Habenaria species, eight Satyrium sequences for two species and four Disa sequences for four species.
A graphical overview of the inter- and intraspecific variation per genus and per marker can be found
in Figure 3. The interspecific variation for nrITS was significantly higher as compared with matK and
rbcL in all three genera: on average 10.2% in Disa, 9.51% in Satyrium and 8.75% in Habenaria. For matK
the interspecific variation was 2.39% for Disa, 2.79% for Habenaria and 1.39% for Satyrium. The
intraspecific distances for nrITS and matK were respectively 1.37% and 0.61% for Disa, 1.98% and
0.36% for Habenaria and 1.45% and 0.48% for Satyrium, The limited reference sequences that were
available for rbcL showed little pairwise interspecific distances (1.44% for Disa, 0.82% for Habenaria
and 1.13% for Satyrium) and only allowed for the intraspecifc distance calculation of Habenaria
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
(0.38%), indicating that rbcL is unsuitable as barcode for species level identification of chikanda
orchids. The calculated thresholds were subsequently used to evaluate the identifications made with
blastn.
Figure 3. Boxplots showing the inter- and intraspecific variation for Disa (a), Habenaria (b) and
Satyrium (c) based on genetic diversity.
In total 15 orchid species were identified using DNA barcoding: Brachycorytis sp. SJK7, Disa caffra
Bolus, Disa celata Summerh., Disa robusta N.E.Br., Disa satyriopsis Kraenzl., Disa welwitschii Rchb. f.,
Disa sp. SJK4.1., Habenaria sp. SJK31.15, Habenaria aff. helicoplectrum Summerh., Habenaria cf. sp.
DO112, Platycoryne crocea Rolfe, Satyrium buchananii Schltr., Satyrium carsonii Rolfe, Satyrium
kitimboense Kraenzl. as well as one species in an unidentified genus, which seems to be closely related
to Habenaria based on the similarity-based BLAST identification. Additionally, one orchid that
flowered from a sprouting tuber was identified based on morphology as Brachycorythis cf. friesii
(Schltr.) Summerh.
bPTP analysis for matK and rbcL showed lumping of several supposedly different species within
one clade on several occasions (Figure S1 and S2, Supplementary Material). In case of nrITS the bPTP
outcome tree often reflected the expected species boundaries, although the posterior probabilities on
the nodes were often too low to determine with confidence if species delimitation had been
performed correctly (Figure S3, Supplementary Material). Something that could be observed from
both the nrITS BLAST identification, as well as the bPTP analysis, was that some of the identifications
showed ambiguity: in collection SJK41, SJK44 and SJK46 some samples were identified
unambiguously as Platycoryne crocea, whereas others showed a mix of Habenaria and Platycoryne top
hits. The matK BLAST identification showed an unambiguous Platycoryne crocea identification for
these samples, despite the presence of several Habenaria species in the reference database and despite
the fact that matK shows a lower level of interspecific variation. A similar observation could be made
for several Disa samples, but here a geographic pattern could be observed. The samples that could
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
unambiguously be identified as Disa robusta were all collected from Tanzania, where the reference
sequence originated from. Some of the Zambian samples also showed Disa robusta as the closest
relative, but not with a high enough percentage identity match to confirm this identification. The
bPTP result in this case shows a lumped and mixed clade with several Disa species and several other
samples included, which confirms that sequence divergence is too limited to resolve the relationship.
These samples were identified as Disa sp. 1, to reflect the fact that they all grouped together in the
same clade. In other cases, such as for certain Habenaria species, we named the closely related
sequences in the identification, whereas in this case the clade was too large to allow for this, since it
contained the following sequences: Disa engleriana Kraenzl., D. erubescens Rendle., D. miniate
Summerh., D. ochrostachya Rchb.f., D. satyriopsis, D. ukingensis Schltr., D. verdickii De Wild., D.
welwitschii, D. zombiaca N.E.Br., an unidentified Disa species and several samples that showed a
highest percentage identity match with D. robusta, D. satyriopsis and D. welwitschii. The posterior
probability for this clade is only 0.25, and combined with the posterior probabilities present on the
within-species nodes, indicates a lot of uncertainty for this identification.
Overall, the most frequently encountered species were Satyrium buchananii, with 41 samples in
eleven different collections, Platycoryne crocea with 19 samples in three different collections and Disa
robusta with 4-16 samples in two to seven different collections. Myala or original chikanda seems to
contain Disa robusta, D. welwitschii and Satyrium buchananii. Mbwelenge or fake chikanda seems to
correspond only to Satyrium buchananii and mshilamshila to one or several Brachycorythis species.
Kasebelele and kapapa referred to Habenaria spp., Platycoryne crocea, Satyrium carsonii and Satyrium
kitimboense. The mixed collections contained all of the above-mentioned species as well as an
unidentified Habenaria species. An overview of the local chikanda classification types, their collection
numbers, identifications and number of samples can be found in Table 1.
343
4. Discussion
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
4.1. Species used for chikanda
Using DNA barcoding as an identification tool for chikanda tubers sold on local Zambian markets
has allowed us to determine for the first time what orchid species are sold on local markets. Previous
studies identifying orchids used for chikanda relied on voucher collections made with collectors in the
field and their morphological identification, which requires a qualified orchid taxonomist
[7,12,14,17,36]. Additionally, relying on local harvesters for details on chikanda collection might not
always lead to collection in the areas where actual intensive harvest is taking place, since some
harvesters do not like to divulge where the best places to harvest are and in some initial harvesting
areas such as the Kitulo Plateau in Tanzania’s Southern Highlands, collection is now prohibited [37].
The current study identified 16 orchid species present on the markets, including at least three
previously undocumented ones: Brachycorythis cf. friesii, Platycoryne crocea and an unidentified species
in a genus, which appears to be closely related to Habenaria, but is not present in our reference
database.
Orchids used for chikanda seem to be harvested from several provinces in Zambia, as well as at
least two regions in Tanzania. Moreover, three international chikanda trade-hubs in towns on or close
to the border with the DRC (Chililabombwe and Kasumbalesa) and Angola (Mwinilunga) were
identified, in addition to the already known trade-hub Tunduma-Nakonde on the Tanzanian border
[8,12]. However, unlike reported in other studies, no harvest from Malawi was mentioned by the
people interviewed in this study, which could mean that trade from this country is currently not
taking place. Another explanation is that this information is lost on the way and that only Tanzania,
being geographically closer and thus more easily accessible for the Bemba people, is mentioned as a
region of origin for chikanda traded in Zambia.
4.2 DNA barcoding performance
Since the term DNA barcoding was coined in 2003 [20], a plethora of studies applying DNA
(meta)barcoding has been performed ranging from retrieving orchids from paleoenvironments [38],
preserved in mammoth dung [39] to the identification of Iranian orchid tubers used for salep [5]. In
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
this study a combined use of the core plant markers matK and rbcL and the nuclear ribosomal ITS
region was used to attempt species level identification of tubers traded on Zambian markets.
Although genetic distance calculations showed limited interspecific distances between closely related
species for all three barcoding markers, DNA barcoding allowed for species-level identification for
several of the frequently sold chikanda species. The data shows that the core land-plant DNA
barcoding markers rbcL and matK were not suitable because of limited variability between species
(matK and rbcL), amplification problems (matK), and/or a limited sequence reference database (rbcL).
nrITS was shown to be more suitable as a barcode marker to distinguish between different chikanda
species, but is not discriminative enough to enable reliable species level identification in certain
orchid clades, such as the clade with Platycoryne crocea, P. buchananiana and Habenaria buchananii; the
clade with H. schimperiana, H. kyimbilae and H. microsaccos and the clade containing Disa sp. 1. Another
drawback of nrITS are the multiple ITS paralogs present in the ribosomal genome. Usually these
copies would show high similarities due to concerted evolution [40,41], but this is not always the case
[42–45]. In our case potentially different nrITS ribotypes became fixed in different orchid populations
and having only one of them in our reference database could lead to the unresolved identifications
observed. Our bPTP results show that even nrITS has too little resolution to reliably delimit species
with high posterior probability support using this method. It does demonstrate, however, how
valuable the use of tree-based methods can be, since it shows the relations between the sequences
and can be used to determine if some of the unidentified samples are likely to belong to the same
species or different species within the same genus. Even if no species-level identification can be made
for these samples, it is possible to use the clustering to determine the diversity of species used.
Although several samples can only be reliably identified as Habenaria sp., we find that they are likely
to belong to at least three different species (Habenaria aff. helicoplectrum, Habenaria cf. sp. DO122 in the
clade with H. schimperiana, H. kyimbilae and H. microsaccos and lastly the Platycoryne sp./Habenaria sp.,
which group together with H. buchananii, P. buchananiana and P. crocea. Expansion of the reference
database, by including at least one individual per species, and preferably multiple individuals per
species from different populations and countries, could ultimately solve remaining challenges, and
this seems the way forward in identification of the traded chikanda tubers, as well as other species
unidentifiable based on morphology. Similar studies using DNA barcoding for the identification of
unknown samples show comparable results for the employed barcoding markers. In a study on the
identification of orchids used for salep, nrITS showed an over three-time higher sequencing success
than matK as well as a two-times higher species-level identification success [5]. Moreover, the
similarity-based approach seemed to outperform the tree-based identification method (ML) in this
study as well with 57% and 39% species-level identifications, respectively. nrITS also shows the
highest identification performance in studies on the identification of medicinal plants [46,47], which
supports the idea that it is recommendable to add a more discriminative marker to the two core landplant barcodes in studies where it is needed to distinguish between closely-related species. Moreover,
our results stress the need for a phylogenetically underpinned taxonomic framework, which is
currently available for Disa [48] and Satyrium [49], but not yet for Habenaria and related genera.
4.3. Local versus scientific classification of chikanda
The results of our identifications of chikanda orchids traded in the Zambia show that the local
classification systems for chikanda are not in congruence with the botanical classification of orchid
species. The orchids sold on the markets were grouped according to area of origin, tuber consistency
preference, or shape of the tuber, but often the tubers offered for sale were mixtures. The different
local types of chikanda sometimes show a variation in orchid species that are identified within these
local grouping. When we look at the chikanda type known as kapapa for example, SJK44 contains
Platycoryne crocea, whereas SJK11, which is supposed to be a mix of kasebelela and kapapa, only seems
to contain Satyrium species. In case of other chikanda types there seems to be more consistency: myala
or real chikanda referred to Disa robusta, D. welwitschii and Satyrium buchananii, mshilamshila samples
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
Table 1. Overview of the different local chikanda classification types, their collections and the identified scientific species.
#
Vernacular name
Collections
Reported origin
Fungulwe
SJK16
unknown
Iringe
SJK17
Tanzania
John White
SJK39
Mporokoso, Zambia
Kabula seke
SJK46
Serenje, Zambia
Kapapa
SJK44
Mporokoso, Zambia
Kasebelela, John White
and Myala
Kasebulela and Kapapa
Mbwelenge
Mshilamshila
SJK41
Chinsali and Mporokoso,
Zambia and Tanzania
SJK11
Luwingu, Zambia
SJK5
Luwingu, Zambia
SJK32
Serenje, Zambia
SJK7
Luwingu, Zambia
SJK12
Kawamba, Zambia
SJK4
Mwinilunga, Zambia;
Myala
SJK18
Sumbawanga, Tanzania
barcoding IDs
samples
Disa robusta
1
Satyrium buchananii
1
Satyrium carsonii
1
Satyrium buchananii
1
Habenaria sp. (Clade H. schimperiana, H. kyimbilae, H. microsaccos
1
Platycoryne crocea Rolfe
7
Platycoryne crocea Rolfe
6
Habenaria cf sp. DO122 (Clade H. schimperiana, H. kyimbilae, H. microsaccos
4
Platycoryne crocea Rolfe
4
Platycoryne sp./Habenaria sp.
2
Satyrium kitimboense
6
Satyrium carsonii
5
Satyrium buchananii
11
Satyrium sp.
1
Satyrium buchananii
6
Brachycorythis sp.
1
Brachycorythis sp.
1
Brachycorythis cf. friesii
1
Disa robusta
4
Disa welwitschii
1
Satyrium buchananii
4
Disa robusta
4
Satyrium buchananii
1
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
Myala
SJK37
Myala and nampanda
SJK21
Luapula, Zambia
Ntonkonshi
SJK25
Democratic Republic of Congo
Sumbawanga
SJK20
mixed
unknown-mixed
SJK31
SJK19
SJK8
unknown
SJK9
SJK13
421
Kawambwa, Zambia
Sumbawanga, Tanzania
Serenje, Zambia
Luwingu, Zambia
Mwinilunga, Zambia;
Luwingu, Zambia
Kawamba, Zambia
Satyrium buchananii
1
Disa welwitschii
2
Satyrium buchananii
1
Disa robusta
1
Disa satyriopsis
1
Disa caffra
1
Disa robusta
1
Habenaria cf sp. DO122 (Clade H. schimperiana, H. kyimbilae, H. microsaccos
1
Satyrium buchananii
6
Satyrium carsonii
1
Habenaria aff. helicoplectrum (BB3151)
1
Disa miniata
1
Disa robusta
2
Disa welwitschii
1
Satyrium buchananii
2
Satyrium carsonii
1
Disa celata
1
Disa welwitschii
1
Satyrium buchananii
3
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
were identified as Brachycorythis species and mbwelenge or fake chikanda was made of Satyrium
buchananii. However, our previous study on the analysis of Tanzanian chikanda cakes showed that the
cake made with fake chikanda tubers also contained Disa miniata, Satyrium anomalum, S. comptum, S.
elongatum, S. riparium, S. shirense and S. volkensii, indicating that there might be some differences
between fake chikanda samples as well [19]. It is well-known from literature that local species concepts
are not necessarily congruent with scientific classifications and that species might be subject to overor underdifferentiation [50,51]. In this case, the grouping of the orchids according to the area of origin,
shape or consistency preference or plainly under the general term chikanda is clearly a case of
underdifferentiation as a much higher diversity was retrieved when employing DNA barcoding. In
order to more reliably identify the orchid species used for a particular chikanda type more samples
per local classification need to be analyzed.
4.4. Orchid availability and conservation
Througout recent decades chikanda has made a remarkable leap in popularity. The first record of
chikanda use made by Audrey Richards [9], described the relish as a poor man’s food, eaten in times
of famine. Recent studies, from 2002 onwards, show that chikanda has emerged as a Zambian snack
popular throughout the country. Studies on chikanda report that with this rise in popularity, the
orchid harvest has escalated and is pressuring local Zambian orchid populations as well as those in
neighboring countries [52,12,10,53]. Many people involved in chikanda trade indicated that chikanda
plants were becoming scarce, and many were concerned about both the quality as well as the quantity
of the orchids available. Our study also confirmed a significant international trade network for
chikanda sources in several regions in Tanzania, as well as in the Democratic Republic of Congo and
Angola. In our current study, we found at least 16 different orchid species sold as chikanda on the
Zambian markets and an overview of previous studies contains 46 species reported to be used for
chikanda (Table S7). This brings us from the use of an initial two orchid species reported for chikanda
[13] to a total of 49 species in eight different genera. The increased harvesting pressure in combination
with the indiscriminate harvesting and use of many more species than earlier assumed pose a threat
to nearly half of the terrestrial orchids occurring in these regions. Despite the establishment of Kitulo
National Park in Tanzania, with orchid conservation as a prime concern, it seems that harvesting
continues even there, since iringe tubers found in this study come specifically from this region [12,53].
Currently there are only seven Disa species from Zambia and surrounding countries registered on
the global IUCN Red List and no species from other genera used for chikanda [54]. Most of the orchid
species used for chikanda seem to have a widespread distribution, but local populations as well as
endemic species could be at risk of overharvesting, and we urge to add the most frequently traded
chikanda species, such as Disa robusta and Satyrium buchananii to the IUCN Red List. Although there
seems no stopping to commerce, people involved in chikanda trade seem genuinely concerned about
welfare of local orchids and interested in exploring other options. Since especially the chikanda
harvesters seem to be in a vulnerable position, where they have to rely on surrounding natural
resources to secure their livelihoods [16], it is essential that when trying to protect orchids used for
chikanda, the situation of the people dependent on the trade is taken into account as well. Currently
the development of sustainable cultivation or chikanda orchids is attempted in collaboration with the
Cape Institute of Micropropagation (Barrydale, South Africa) and possible alternative sources of
income for the people involved in chikanda trade, such as honey production, are being explored [55].
Alternatively, since the purpose of the chikanda orchids mainly is to bind and create an elastic
structure to the cake, it might be possible to encourage the use of an alternative source of starch to
replace the tubers.
5. Conclusions
DNA barcoding using the nuclear ribosomal ITS marker proved to be useful in identifying
terrestrial orchid species traded as chikanda on local Zambian markets and outperformed
identification using the core land-plant barcoding markers matK and rbcL. Sixteen orchid species, of
which three previously undocumented, were identified from marketed chikanda tubers, bringing the
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
475
476
477
478
479
480
481
482
483
484
485
486
total number of orchid species used for chikanda to at least 49. The species most frequently found on
the markets were Disa robusta, Satyrium buchananii and Platycoryne crocea. However, the results are
only as good at the reference material is and an increased reference database in combination with an
underpinned phylogenetic framework for Habenaria and related genera would likely ameliorate the
reliability of the identifications. Tubers are harvested from various regions in Zambia and Tanzania,
and additional international chikanda trade-hubs have been identified on the border with the
Democratic Republic of Congo and Angola. People involved in chikanda trade indicate that both
orchid quality as well as quantity are decreasing and are willing to consider alternatives to chikanda
trade to secure their income.
Authors should discuss the results and how they can be interpreted in perspective of previous
studies and of the working hypotheses. The findings and their implications should be discussed in
the broadest context possible. Future research directions may also be highlighted.
487
488
Supplementary Materials: The following supplementary figures and tables are available online at
www.mdpi.com/xxx/s1
489
490
491
492
493
494
495
496
497
498
499
500
Figure S1. bPTP analysis of all chikanda matK query and reference samples.
Figure S2. bPTP analysis of all chikanda rbcL query and reference samples.
Figure S3. bPTP analysis of all chikanda nrITS query and reference samples.
Table S1. Brahms RDE file of novel voucher specimens of orchid taxa samples in this study.
Table S2. Genbank accession numbers of the chikanda tuber sequences.
Table S3. Hit table with the first 5 BLAST top hits per sample for nrITS.
Table S4. Hit table with the first 5 BLAST top hits per sample for matK.
Table S5. Hit table with the first 5 BLAST top hits per sample for rbcL.
Table S6. Successfully sequenced samples with their vernacular name, reported origin, identification
based on sequence-similarity for nrITS and matK, identification based on the tree-based bPTP analysis
and the consensus ID.
Table S7. Orchid species used for chikanda according to literature.
501
502
503
504
505
506
507
508
509
Author Contributions: S.V., R.B. and H.d.B. conceived and designed the study in collaboration with
S.J.K., D.C., R.V. and N.W.; S.J.K. performed the chikanda collection and interviews in collaboration
with D.C., G.M., N.W. and R.V; R.B., N.W. and K.Y. collected reference vouchers in the field, which
were identified in collaboration with B.B.; B.B, G.N and F.M. provided the reference sequence
database for Habenaria species and some species in closely related genera; S.J.K. performed the
labwork for the chikanda collections and M.B.F. for the reference collections; S.J.K. performed the
data analysis under supervision of S.V. and H.d.B.; S.V. and S.J.K. wrote the manuscript in
consultation with the other authors and the manuscript was edited and reviewed in detail by B.G.,
T.v.A. and H.d.B.
510
511
Funding: This research was funded by the Darwin Initiative (UK Government) Grant No. 23034 and
NWO-SIDA-COSTECH TASENE Grant W 02.29.102.
512
513
514
Acknowledgments: The authors would like to thank the participants in the study for sharing their
knowledge and Simon Hultby from the Uppsala Botanical Garden for growing the sprouted chikanda
orchid tubers.
515
516
517
518
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or
in the decision to publish the results.
519
References
520
521
522
1.
Hinsley, A.; de Boer, H. J.; Fay, M. F.; Gale, S. W.; Gardiner, L. M.; Gunasekara, R.
S.; Kumar, P.; Masters, S.; Metusala, D.; Roberts, D. L. A review of the trade in orchids
and its implications for conservation. Bot. J. Linn. Soc. 2017.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Kasparek, M.; Grimm, U. European trade in Turkish Salep with special reference to
Germany. Econ. Bot. 1999, 53, 396–406.
Ece Tamer, C.; Karaman, B.; Utku Copur, O. A traditional Turkish beverage: Salep.
Food Rev. Int. 2006, 22, 43–50.
Kreziou, A.; de Boer, H.; Gravendeel, B. Harvesting of salep orchids in north-western
Greece continues to threaten natural populations. Oryx 2015, 1–4.
Ghorbani, A.; Gravendeel, B.; Selliah, S.; Zarré, S.; Boer, H. DNA barcoding of
tuberous Orchidoideae: a resource for identification of orchids used in Salep. Mol. Ecol.
Resour. 2017, 17, 342–352.
Davenport, T. R. B.; Ndangalasi, H. J. Orchid harvest - An assessment of the harvesting
and trade of orchid tubers across Tanzania’s Southern Highlands; Wildlife
Conservation Society, 2001; p. 23;.
Bingham, M. G. Chikanda trade in Zambia. Orchid Conserv. News 2004, 4, 22–25.
Veldman, S.; Otieno, J. N.; Andel, T. van; Gravendeel, B.; de Boer, H. J. Efforts urged
to tackle thriving illegal orchid trade in Tanzania and Zambia for chikanda production.
TRAFFIC Bull. 2014, 26, 47–50.
Richards, A. I. Land, labour and diet in Northern Rhodesia: an economic study of the
Bemba tribe; Oxford Univ Press: Oxford, UK, 1939;
Bingham, M. G. The Lowdown Magazine. 2007,.
Temzie Bites Zambian Food | Chikanda African Ham (Polony) Recipe. Retrieved from:
https://www.youtube.com/watch?v=mDGKx-GOLoA; 2017;
Davenport, T. R. B.; Ndangalasi, H. J. An escalating trade in orchid tubers across
Tanzania’s Southern Highlands: assessment, dynamics and conservation implications.
Oryx 2003, 37, 55–61.
Cribb, P. J.; Leedal, G. P. The mountain flowers of southern Tanzania: a field guide to
the common flowers; AA Balkema: Rotterdam, NL, 1982;
Nyomora, A. M. S. Distribution and abundance of the edible orchids of the Southern
Highlands of Tanzania. Tanzanian J. Sci. 2005, 31, 45–54.
Mapunda, L. N. D. Edible orchids in Makete district, the Southern Highlands of
Tanzania: distribution, population and status, Master thesis: Swedish Biodiversity
Centre, Uppsala Universitet (Sweden), 2007.
Challe, J. F.; Price, L. L. Endangered edible orchids and vulnerable gatherers in the
context of HIV/AIDS in the Southern Highlands of Tanzania. J. Ethnobiol.
Ethnomedicine 2009, 5, 41, doi:10.1186/1746-4269-5-41.
Kasulo, V.; Mwabumba, L.; Munthali, C. A review of edible orchids in Malawi. J.
Hortic. For. 2009, 1, 133–139.
Hamisy, C. W. Development of conservation strategies for the wild edible orchid in
Tanzania. Prog. Rep. Rufford Small Grants Found. 2010.
Veldman, S.; Gravendeel, B.; Otieno, J. N.; Lammers, Y.; Duijm, E.; Nieman, A.;
Bytebier, B.; Ngugi, G.; Martos, F.; van Andel, T. R.; others High-throughput
sequencing of African chikanda cake highlights conservation challenges in orchids.
Biodivers. Conserv. 2017, 1–18.
Hebert, P. D. N.; Cywinska, A.; Ball, S.; de Waard, J. Biological identifications through
DNA barcodes. Proc. R. Soc. B 2003, 270, 313–322.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
21. Raclariu, A. C.; Heinrich, M.; Ichim, M. C.; Boer, H. Benefits and limitations of DNA
barcoding and metabarcoding in herbal product authentication. Phytochem. Anal. 2017.
22. ISE International Society of Ethnobiology Code of Ethics (with 2008 additions).
Available online: http://ethnobiology.net/code-of-ethics/.
23. Goodman, L. A. Snowball sampling. Ann. Math. Stat. 1961, 148–170.
24. Martos, F.; Le Péchon, T.; Ngugi, G.; Pailler, T.; Bellstedt, D. U.; Linder, H. P.;
Bytebier, B. Phylogenetic relationships amongst the African genera of the tribe
Orchideae (Orchidaceae). Manuscript in prep. 2018.
25. Yoon, C. S.; Glawe, A.; Shaw, P. D. A method for rapid small-scale preparation of
fungal DNA. Mycologia 1991, 83, 835–838.
26. Shepherd, L. D.; McLay, T. G. Two micro-scale protocols for the isolation of DNA
from polysaccharide-rich plant tissue. J. Plant Res. 2011, 124, 311–314.
27. Ghorbani, A.; Gravendeel, B.; Selliah, S.; Zarre, S.; de Boer, H. J. DNA barcoding of
tuberous Orchidoideae: A resource for identification of orchids used in Salep. Mol Ecol
Resour 2016, doi:10.1111/1755-0998.12615.
28. Kress, W. J.; Wurdack, K. J.; Zimmer, E. A.; Weigt, L. A.; Janzen, D. H. Use of DNA
barcodes to identify flowering plants. Proc. Natl. Acad. Sci. 2005, 102, 8369–8374.
29. Dunning, L. T.; Savolainen, V. Broad-scale amplification of matK for DNA barcoding
plants, a technical note: AMPLIFICATION OF matK FOR DNA BARCODING
PLANTS. Bot. J. Linn. Soc. 2010, 164, 1–9, doi:10.1111/j.1095-8339.2010.01071.x.
30. Sun, Y.; Skinner, D. Z.; Liang, G. H.; Hulbert, S. H. Phylogenetic analysis of Sorghum
and related taxa using internal transcribed spacers of nuclear ribosomal DNA. Theor.
Appl. Genet. 1994, 89, doi:10.1007/BF00226978.
31. Bonfield, J. K.; Smith, K. f; Staden, R. A new DNA sequence assembly program.
Nucleic Acids Res. 1995, 23, 4992–4999, doi:10.1093/nar/23.24.4992.
32. Staden, R. The Staden sequence analysis package. Mol. Biotechnol. 1996, 5, 233–241.
33. Larsson, A. AliView: a fast and lightweight alignment viewer and editor for large
datasets. Bioinformatics 2014, 30, 3276–3278, doi:10.1093/bioinformatics/btu531.
34. Zhang, J.; Kapli, P.; Pavlidis, P.; Stamatakis, A. A general species delimitation method
with applications to phylogenetic placements. Bioinformatics 2013, 29, 2869–2876,
doi:10.1093/bioinformatics/btt499.
35. Stamatakis, A.; Hoover, P.; Rougemont, J. A rapid bootstrap algorithm for the RAxML
web servers. Syst. Biol. 2008, 57, 758–771, doi:10.1080/10635150802429642.
36. Hamisy, C. W. Development of conservation strategies for the wild edible orchid in
Tanzania. Prog. Rep. Rufford Small Grants Found. 2010.
37. Davenport, T. R. B.; Bytebier, B. Kitulo Plateau, Tanzania-a first African park for
orchids. Orchid Rev. 2004, 112, 161–165.
38. Boessenkool, S.; Mcglynn, G.; Epp, L. S.; Taylor, D.; Pimentel, M.; Gizaw, A.;
Nemomissa, S.; Brochmann, C.; Popp, M. Use of Ancient Sedimentary DNA as a Novel
Conservation Tool for High-Altitude Tropical Biodiversity. Conserv. Biol. 2014, 28,
446–455, doi:10.1111/cobi.12195.
39. van Geel, B.; Aptroot, A.; Baittinger, C.; Birks, H. H.; Bull, I. D.; Cross, H. B.;
Evershed, R. P.; Gravendeel, B.; Kompanje, E. J.; Kuperus, P. The ecological
implications of a Yakutian mammoth’s last meal. Quat. Res. 2008, 69, 361–376.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 October 2018
doi:10.20944/preprints201810.0587.v1
Peer-reviewed version available at Genes 2018, 9, 595; doi:10.3390/genes9120595
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
40. Elder, J. F.; Turner, B. J. Concerted Evolution of Repetitive DNA Sequences in
Eukaryotes. Q. Rev. Biol. 1995, 70, 297–320, doi:10.1086/419073.
41. Ganley, A. R. D.; Kobayashi, T. Highly efficient concerted evolution in the ribosomal
DNA repeats: Total rDNA repeat variation revealed by whole-genome shotgun
sequence data. Genome Res. 2007, 17, 184–191, doi:10.1101/gr.5457707.
42. Harpke, D.; Peterson, A. Non-concerted ITS evolution in Mammillaria (Cactaceae).
Mol. Phylogenet. Evol. 2006, 41, 579–593, doi:10.1016/j.ympev.2006.05.036.
43. Feliner, G.; Rosselló, J. Better the devil you know? Guidelines for insightful utilization
of nrDNA ITS in species-level evolutionary studies in plants. Mol. Phylogenet. Evol.
2007, 44, 911–919.
44. Zheng, X.; Cai, D.; Yao, L.; Teng, Y. Non-concerted ITS evolution, early origin and
phylogenetic utility of ITS pseudogenes in Pyrus. Mol. Phylogenet. Evol. 2008, 48, 892–
903, doi:10.1016/j.ympev.2008.05.039.
45. Xu, B.; Zeng, X.-M.; Gao, X.-F.; Jin, D.-P.; Zhang, L.-B. ITS non-concerted evolution
and rampant hybridization in the legume genus Lespedeza (Fabaceae). Sci. Rep. 2017,
7, 40057, doi:10.1038/srep40057.
46. Chen, S.; Yao, H.; Han, J.; Liu, C.; Song, J.; Shi, L.; Zhu, Y.; Ma, X.; Gao, T.; Pang,
X.; Luo, K.; Li, Y.; Li, X.; Jia, X.; Lin, Y.; Leon, C. Validation of the ITS2 region as a
novel DNA barcode for identifying medicinal plant species. PLoS ONE 2010, 5, 1–8,
doi:10.1371/journal.pone.0008613.
47. Mezzasalma, V.; Bruni, I.; Fontana, D.; Galimberti, A.; Magoni, C.; Labra, M. A DNA
barcoding approach for identifying species in Amazonian traditional medicine: The case
of Piri-Piri. Plant Gene 2017, 9, 1–5, doi:10.1016/j.plgene.2016.11.001.
48. Bytebier, B.; Dirk U. Bellstedt; H. Peter Linder A New Phylogeny-Based Sectional
Classification for the Large African Orchid Genus Disa. Taxon 2008, 57, 1233–1251.
49. van der Niet, T.; Linder, H. P. Dealing with incongruence in the quest for the species
tree: A case study from the orchid genus Satyrium. Mol. Phylogenet. Evol. 2008, 47,
154–174.
50. Cunningham, A. B. Applied ethnobotany: people, wild plant use and conservation;
Earthscan: London, 2001; ISBN 1853836974.
51. Martin, G. J. Ethnobotany: a methods manual; Earthscan: London, UK, 2004;
52. Bingham, M. G.; Smith, P. P. Southern African plant red data lists: Zambia; Golding,
J. S., Bandeira, S. O., Eds.; SABONET: Pretoria, South Africa, 2002;
53. Veldman, S.; Otieno, J. N.; Andel, T. van; Gravendeel, B.; de Boer, H. J. Efforts urged
to tackle thriving illegal orchid trade in Tanzania and Zambia for chikanda production.
TRAFFIC Bull. 2014, 26, 47–50.
54. IUCN IUCN Red List of Threatened Species. Version 2017-3.; 2018;
55. Bone, R. E.; Wightman, N.; Vinya, R.; Veldman, S.; Yokoya, K.; Hargreaves, S.;
Kendon, J.; Crous, H. Edible wild orchid trade: sustaining livelihoods and biodiversity
in Zambia. Darwin Initiative Main Project Annual Report.; 2017;