Annals of Botany 95: 601–608, 2005 doi:10.1093/aob/mci062, available online at www.aob.oupjournals.org

Inter- and Intra-specific Variation among Five Erythroxylum Taxa Assessed by AFLP

EMANUEL L. JOHNSON*, DAPENG ZHANG and STEPHEN D. EMCHE 1USDA ARS PSI ACSL, 10300 Baltimore Avenue, BARC-W, Beltsville, MD 20705, USA

Received: 5 March 2004 Returned for revision: 23 September 2004 Accepted: 15 November 2004 Published electronically: 13 January 2005

Background and Aims The four cultivated Erythroxylum taxa (E. coca var. coca, E. novogranatense var. novogranatense, E. coca var. ipadu and E. novogranatense var. truxillense) are indigenous to the Andean region of South America and have been cultivated for folk-medicine and, within the last century, for illicit cocaine pro- duction. The objective of this research was to assess the structure of genetic diversity within and among the four cultivated alkaloid-bearing taxa of Erythroxylum in the living collection at Beltsville Agricultural Research Center. Methods Amplified fragment length polymorphism (AFLP) fingerprinting was performed in 86 Erythroxylum accessions using a capillary genotyping system. Cluster analysis, multidimensional scaling (MDS) and analysis of molecular variance (AMOVA) were used to assess the pattern and level of genetic variation among and within the taxa. Key Results A clear distinction was revealed between E. coca and E. novogranatense. At the intra-specific level, significant differentiation was observed between E. c. var. coca and E. c. var. ipadu, but the differentiation between E. n. var. novogranatense and E. n. var. truxillense was negligible. Erythroxylum c. var. ipadu had a significantly lower amount of diversity than the E. c. var. coca and is genetically different from the E. c. var. ipadu currently under cultivation in Colombia, South America. Conclusions There is a heterogeneous genetic structure among the cultivated Erythroxylum taxa where E. coca and E. novogranatense are two independent species. Erythroxylum coca var. coca is most likely the ancestral taxon of E. c. var. ipadu and a founder effect may have occurred as E. c. var. ipadu moved from the eastern Andes in Peru and Bolivia into the lowland Amazonian basin. There is an indication of artificial hybridization in coca grown in Colombia. ª 2005 Annals of Botany Company

Key words: Erythroxylum coca var. coca, Erythroxylum coca var. ipadu, Erythroxylum novogranatense var. novogranatense, Erythroxylum novogranatense var. truxillense, AFLP markers, genetic variation, cultivated coca, DNA fingerprinting, cocaine, tropical plants.

INTRODUCTION Since the 1970s, morphology, breeding systems and chemotaxonomic data were the primary descriptors used to The extensive living collection of Erythroxylum at the detail the differences between the cultivated Erythroxylum Beltsville Agricultural Research Center, Beltsville, MD, taxa (Bohm et al., 1982; Johnson et al., 1997, 1998, 2002, USA, which has been maintained since the early 1970s, 2003a; Johnson and Schmidt, 1999). With the identification contains the four cultivated alkaloid-bearing varieties: of molecular markers and their associated specificity, Erythroxylum coca var. coca Lam (E. c. var. coca); further assessment of the genetic diversity among the cult- Erythroxylum coca var. ipadu Plowman (E. c. var. ivated Erythroxylum taxa is warranted in order to accrete ipadu); Erythroxylum novogranatense var. novogranatense and refine the existing morphological and chemotaxonomic- (Morris) Hieron (E. n. var. novogranatense); Erythroxylum based classification system. While a variety of molecular novogranatense var. truxillense [Rusby] Plowman (E. n. var. assays could be used to assess the genetic diversity, each truxillense). The geographical, ecological and morpho- method differs in principle, application, the amount of poly- logical differences of these taxa were detailed as early as morphism detected, cost and time required. In a previous the 16th century (Ganders, 1979; Plowman, 1979, 1982, study, amplified fragment length polymorphism (AFLP) 1984; Rury, 1981; Schultes, 1981). However, it was not (Vos et al., 1995) was used to analyse 132 accessions of until the 1970s that cultivated coca was determined to be Erythroxylum to characterize and positively identify the four derived from two species of the genus Erythroxylum; E. c. cultivated taxa, as well as a feral taxon (Johnson et al., var. coca Lam and E. n. var. novogranatense (Morris) Hieron 2003b). The first objective of the current study was to (Plowman, 1979, 1982, 1984; Bohm et al., 1982). This clas- examine further the taxonomic status, and elucidate the sification, according to Plowman and Rivier (1983), has evolutionary relationship, of the four cultivated alkaloid- been supported multifactorially through interdisciplinary bearing Erythroxylum taxa. The second objective was to research (Johnson et al., 2003b). Furthermore, breeding detect and quantify the inter- and intra-specific genetic vari- evidence and eco-geographical data suggests that the most ation in these taxa. Eighty-five Erythroxylum accessions, likely phylogeny for the four cultivated taxa is a linear evolu- which are representative of the four cultivated taxa in the tionary sequence, wherein E. c. var. coca is the ancestral living collection, were analysed using AFLP genotyping taxon that gave rise to E. n. var. truxillense which gave rise, in combination with cluster and ordination analysis. The in turn, to E. n. var. novogranatense (Bohm et al., 1982). resulting information provided insights into the structure * For correspondence. E-mail [email protected] and pattern of genetic diversity of Erythroxylum in the Annals of Botany 95/4 ª Annals of Botany Company 2005; all rights reserved 602 Johnson et al. — Variation among Cultivated Coca as Assessed by AFLP living collection at Beltsville Agricultural Research Center ethnobotany, morphological characterization, alkaloid and the Andes region of South America. content, breeding system, chemotaxonomic data and geo- graphical distribution have been summarized previously (Ganders, 1979; Plowman, 1979, 1981, 1982, 1983, 1984; MATERIALS AND METHODS Rury, 1981; Schultes, 1981; Bohm et al., 1982; Plowman Plant material and Rivier, 1983; Johnson et al., 1997, 1998, 1999, 2002, 2003b; Johnson and Schmidt, 1999). Due to the confusion Young expanding leaf tissue was harvested from 86 samples, by investigators for the four cultivated alkaloid-bearing taken from the living collection at Beltsville Agricultural taxa it was considered necessary to show how the taxa Research Center, of E. coca var. coca Lam, E. coca var. differed using AFLP DNA analysis. The living collec- ipadu Plowman, E. novogranatense var. novogranatense tion of Erythroxylum at Beltsville Agricultural Research (Morris) Hieron, E. novogranatense var. truxillense Center was authenticated by T. Plowman in 1988 and (Rusby) Plowman and E. ulei O.E. Schulz, as well as some re-authenticated by P. M. Rury in 1993. Plants derived F1 Erythroxylum coca var. ipadu propagules (Table 1). The from the living collection and authenticated by Rury were transferred to a Hawaiian field site. The Hawaiian T ABLE 1. Sample table of 86 Erythroxylum c. var. coca, field site was located on the Island of Kauai and was Erythroxylum c. var. ipadu, Erythroxylum n. var. novograna- selected by the US Department of Agriculture, Agricultural tense, E. n. var. truxillense and Erythroxylum ulei samples Research Service and the State of Hawaii because of the from the living collection at the Beltsville Agricultural similarity of soils to those found in the coca-growing Research Center regions of Bolivia and Peru. The pH of the soil at the Hawaiian field sites ranged from 40to57, which was Species Accession tag Origin Species Accession tag Origin ideal for coca growth. The harvested leaf tissues were sepa- rately placed in labelled Zip-Loc bags, immediately stored coca B102LS Bolivia novo B292LS Bolivia at 0 C and transported to the laboratory for DNA extraction coca B85LS Bolivia novo B253LS Bolivia coca B88LS Bolivia novo B205LS Bolivia and analysis. Erythroxylum ulei in the current study was coca B14LS Bolivia novo B300LS Bolivia only used to verify the consistency of the AFLP analysis and coca B104LS Bolivia novo B201-1LS Bolivia was not part of the quantitative analysis. coca B110LS Bolivia ipadu F-1 B508 Beltsville coca B56LS Bolivia ipadu F-1 B508 Beltsville coca B96LS Bolivia ipadu F-1 B501SS Beltsville Isolation of DNA from leaf tissue coca B180LS Bolivia ipadu F-1 B503SS Beltsville coca B105LS Bolivia ipadu F-1 B508 Beltsville Genomic DNA (i.e. total DNA) was extracted from leaf coca B63LS Bolivia ipadu F-1 B501SS Beltsville tissue of the five taxa using a modification of the Qiagen coca B127LS Bolivia ipadu F-1 B504 Beltsville DNA Stool Mini KitTM protocol (Qiagen Inc., Valencia, CA, coca B31LS Bolivia ipadu F-1 B503SS Beltsville coca B60LS Bolivia ipadu B501SS Bolivia USA). One hundred milligrams f. wt (i.e. 20 mg d. wt.) of coca B150LS Bolivia ipadu B503SS Bolivia leaf tissue were weighed and placed into a 2-mL lysing coca B94LS Bolivia ipadu B507 Bolivia matrix cylinder containing two 053-cm ceramic spheres coca B98LS Bolivia ipadu B504 Bolivia (QBiogene, Inc., Carlsbad, CA, USA) with 35 mg of poly- coca B9LS Bolivia ipadu B505 Bolivia vinylpolypyrrolidone (Sigma Chem., Co., St Louis, MO, coca B80LS Bolivia ipadu B508 Bolivia coca B147RLS Bolivia ipadu F-1 B504 Beltsville USA). Then 14 mL of the buffer ASL was added and coca B50LS Bolivia ipadu F-1 B501SS Beltsville the tissue homogenized with a FastPrep 120 tissue homo- coca B64LS Bolivia ipadu F-1 B504 Beltsville genizer (Savant Instruments, Holbrook, NY, USA) at a coca B91LS Bolivia ipadu F-1 B505 Beltsville speed setting of 65 for 45 s. The sample was then incubated novo B220LS Bolivia ipadu F-1 B505 Beltsville novo B276LS Bolivia ipadu F-1 B506 Beltsville for 5 min in a 70 C water bath (Lab Line, Barnstead novo B228-1LS Bolivia ipadu F-1 B505 Beltsville International, Dubuque, IA, USA). During incubation, novo B216LS Bolivia ipadu BJ IP-23 Bolivia the sample was mixed several times by inverting the cylin- novo B245LS Bolivia ipadu BJ IP-24 Bolivia der and then centrifuged at 16 100 g (23 C) for 10 min novo B294LS Bolivia ipadu BJ IP-25 Bolivia (Eppendorf 5415D/R Centrifuge, Hamburg, Germany). The novo B225LS Bolivia trux B315SS Bolivia supernatant was transferred to a 15-mL microfuge tube novo B201LS Bolivia trux B319LS Bolivia TM novo B242LS Bolivia trux B322LS Bolivia containing one InhibitEX tablet, vortexed for 1 min novo B223LS Bolivia trux B314SS Bolivia and then incubated at 23 C for 1 min. The sample was novo B286LS Bolivia trux B318SS Bolivia centrifuged at 16 100 g (23 C) for 3 min and the supernatant novo B251LS Bolivia trux B312LS Bolivia novo B206LS Bolivia trux B302LSS Bolivia transferred to a 15-mL microfuge tube and the pellet novo B236LS Bolivia trux B317SS Bolivia discarded. The sample was then centrifuged for 10 min at novo B272LS Bolivia trux B323LS Bolivia 16 100 g (23 C) and 200 mL of the supernatant transferred novo B295LS Bolivia trux B316 Bolivia to a 15-mL microfuge tube containing 15 mL of proteinase novo B244LS Bolivia trux B303LS Bolivia K. To this tube, 200 mL of AL lysis buffer were added, the novo B233LS Bolivia trux B321LS Bolivia novo B278LS Bolivia trux B305LS Bolivia microfuge tube was vortexed, and then incubated at 70 C novo B-205-1LS Bolivia ulei #1 IIBC for 10 min. Following incubation, 200 mL of 100 % ethanol were added, the sample vortexed and centrifuged as above Accession identifications correspond to the labels in the MDS (Fig. 2). for 1 min. The lysate was transferred to a QIAamp spin Johnson et al. — Variation among Cultivated Coca as Assessed by AFLP 603 column in a 2-mL microfuge tube and centrifuged (16 100 g Capillary electrophoresis and genotyping (23 C) for 1 min. The filtrate was discarded and 500 mLof Capillary electrophoresis of all samples was performed AW1 wash buffer added. It was then centrifuged for 1 min TM on a MegaBACE 500, 48-capillary system (Amersham (16 100 g (23 C) and the QIAamp spin column transferred Biosciences, Piscataway, NJ, USA) equipped with to a 2-mL microfuge tube. This wash step was repeated TM MegaBACE Instrument Control software Ver. 25 twice using 500 ml of AW2 wash buffer. The spin column (Amersham Biosciences). Samples were run under the was transferred to a 1 5-mL microfuge tube and 200 mLof following genotyping parameters: injection voltage, 3 kV; AE elution buffer added. The sample was allowed to stand injection time, 45 s; run voltage, 10 kV; run time, 105 min; for 1 min and then centrifuged at 16 100 g (23 C) for 1 min dyes, GT dye set 2 [ET-Rox (900), FAM, NED and HEX]. to elute the DNA. All samples were stored at 80 C. Some samples were prepared for analysis by diluting the final amplified product 1 : 30 (v/v) in 01 % Tween-20. All DNA quantification samples included 1 % (v/v) Amersham Bioscience ET-Rox 900 bp DNA size standard. Electropherograms were ana- DNA was quantified by fluorimetric analysis using a TM lysed with Fragment Analyzer 1.2 Software (Amersham Fluoroskan Ascent microplate reader using 485/538 nm Biosciences) and all dendrograms with bootstrap values excitation/emission filter settings (LabSystems, Helsinki, (1000 repetitions) were produced using Free Tree Finland). All samples were diluted 1 : 20 with Tris-buffered (Pre-released Version 0.9.1.50; ª Adam Pavlieek, Tomal EDTA (TE) and analysed using the PicoGreen dsDNA quan- Pavlieek and Jaroslav Flegr, 1998–1999) and Tree View titation kit (Molecular Probes, Eugene, OR, USA) in a 96- (Version 1.6.6; ª Roderic D. M. Page, 2001). well platform plate (Greiner Bio-one, Longwood, FL, USA). Twenty primer pairs were evaluated for the current Fifty microlitres of PicoGreen solution (1 : 200 dilution), research. Sixteen primer pairs were successful; however, 2 5 L of diluted DNA (1 : 20) and 47 5 L of de-ionized m m only the four primer pairs with optimal resolution were water were added to each well. A standard curve was gen- selected for this study. A previous study with Erythroxylum erated from DNA standards (PicoGreen kit) ranging from revealed that three primer pairs were sufficient to determine 10 to 500 ng mL1 on the same plate. Final sample dilution relationships in a population containing up to 132 samples was 1 : 800 and all measurements were repeated three times. (Johnson et al., 2003b). The fragment data (100–550 nt range) were analysed AFLP analysis using Amersham’s Fragment AnalyzerTM 1.2 Software (Amersham Biosciences). For AFLP analysis, the maximum DNA fragments were amplified using a modification of bin width was 100 nt, and no further Y threshold was the procedure by Vos et al. (1995). The modification was as applied. Each sample was scored for each bin as ‘1’ if a follows: template DNA (500 ng) was digested with EcoRI fragment of that size was present, and ‘0’ if not. Fully and MseI (New England BioLabs, Beverly, MA, USA) and populated and unpopulated bins were excluded (i.e. peaks ligated in a single step to commercial EcoRI and MseI present in all or no accessions). In rare cases where two oligonucleotide adapters (Applied Biosystems, Foster City, fragments were present in one bin, the bin was scored as ‘1’ CA, USA) by incubation overnight at room temperature. for that sample. Solutions were prepared as previously described, (Johnson et al., 2003b), except as noted below. The first, preselective amplification of the restricted and Data analysis ligated fragments utilized commercial EcoRI and MseI AFLP preselective primers and AFLP core mix (both For cluster analysis, matrix of pairwise distances (Nei and from Applied Biosystems). The thermocycling programme Li, 1979) between all pairs of individuals, were calculated for this amplification was: 94 C for 3 min, followed by using Free Tree (Pre-released Version 0.9.1.50; ª Adam 20 cycles of the following profile: 94 C for 20 s, 56 C for Pavlieek, Tomal Pavlieek and Jaroslav Flegr, 1998–1999). 30 s and 72 C for 2 min with a final hold of 60 C for 45 min A neighbour-joining tree was constructed and the bootstrap (PTC-200 Peltier Thermal Cycler, MJ Research, Waltham, value for internal branches of the tree was computed with MA, USA). For the second, selective amplification, primers 1000 repetitions. The software Tree View (Version 1.6.6; with a FAM, HEX or NED active ester dye attached to the 50 ª Roderic D. M. Page, 2001) was used to draw the tree. end of each EcoRI primer, and non-tagged MseI primers Erythroxylum ulei was included as a check in the cluster (Applied Biosystems) were used. The products from the analysis and is a taxon of Erythroxylum which does not preselective amplification were diluted as described pre- contain the cocaine alkaloid. viously (Johnson et al., 2003b) and used as templates for For the assessment of within taxon genetic variation, the selective amplification. The labelled EcoRI primer and the mean distance (Nei and Li, 1979) of each taxon was unlabelled MseI primer were both used at a concentration of calculated using the SAS program of Dubreuil et al. (2002). 010 mM. The thermocycling profile was: 94 C for 2 min, The mean distance was defined as the average of all pair- followed by 10 cycles of 94 C for 20 s, 1 degree per cycle wise distances between individuals within each taxon step-down of annealing temperature from 66 C held for 30 s (Nienhuis et al., 1994). The mean distances then were com- and 72 C for 2 min. This was followed by 25 cycles of 94 C pared using the T-TEST procedure of SAS (SAS, 1999). for 20 s, 56 C for 30 s, 72 C for 2 min and a final hold at For ordination analysis, a matrix of Euclidian distances 60 C for 45 min. was calculated using a program written in Visual Basic 604 Johnson et al. — Variation among Cultivated Coca as Assessed by AFLP

Macro language embedded in MS Excel. The distances T ABLE 2. AFLP markers generated among 85 Erythroxylum between the 85 accessions were then presented in a two- accessions representing four taxa using four EcoRI + MseI dimension scaling plot using the Multi-Dimensional Scaling primer pairs (MDS) procedure of SAS (SAS, 1999). The analysis of molecular variance (AMOVA) procedure, based on the Erythroxylum Total no. No. polymorphic % polymorphic squared Euclidian distance was used to estimate variance species of markers markers markers components for AFLP genotypes (Excoffier et al., 1992). . Variation was partitioned between species, within species E. c. var. ipadu 427 403 94 4 E. c. var. coca 399 393 98.5 and among varieties, within variety and among individuals. E. n. var. truxillense 418 386 92.3 The variance components of interest were extracted and E. n. var. novogranatense 423 403 95.3 tested using nonparametric permutation procedures. Total 1667 1585 94.4 Variation between taxa was then partitioned into pair- wise distances between taxa to examine their relative contribution to the total molecular diversity (Excoffier distinction between E. c. var. coca and E. n. var. and Smouse, 1994). novogranatense (Fig. 2). Within the species of E. coca, the pattern of differentiation between E. c. var. coca and E. c. var. ipadu is clearly shown, although one outlier RESULTS accession of E. c. var. ipadu, BJIP-23, was closely grouped AFLP polymorphisms and cluster analysis with E. c. var. coca (Table 1 and Figs 1 and 2). In addition, three accessions of E. c. var. ipadu, BJ IP-24, BJ IP-25 and The four primer pair combinations used for AFLP analysis B507, were located intermediately between E. c. var. ipadu generated 1667 fragments for the 85 accessions of which and E. c. var. coca. In contrast to E. coca, the two varieties 1585 were polymorphic (944 %; Table 2). Between the in E. novogranatense are not distinguishable (Figs 1 and 2). two species studied, the polymorphic fragments accounted One outlier of E. coca var. coca, accession B147RLS, was for 985%inE. c. var. coca and 953%inE. n. var. close to E. novogranatense and will be discussed later novogranatense (Table 2). Within each individual taxon, (Table 1 and Figs 1, 2 and 3B). the number of polymorphic fragments was significantly The results of AMOVA revealed significant genetic smaller than at the inter-specific level, ranging from 386 variation hierarchically at the level of inter-species, inter- for E. n. var. truxillense to 403 for E. c. var. ipadu and E. n. variety and inter-individual (Table 4). Variation from the var. novogranatense (Table 2). The large number of poly- three sources accounts for 357%,130 % and 513 % of the morphic fragments at the inter- and intra-specific level total molecular variance, respectively. Partitioning the inter- demonstrated that there is a high level of genetic variation variety variability into pair-wise distances showed the across the four cultivated Erythroxylum taxa. This variation amount that each variety contributes to the total molecular is sufficient to permit the assessment of inter- and intra- diversity (Table 5). Pair-wise inter-variety distances differed specific diversity in Erythroxylum using AFLP analysis. greatly. Within the species of E. novogranatense, the dis- Clustering of the 85 accessions based on Nei’s genetic tance between the two varieties, E. n. var. novogranatense distance (Nei and Li, 1979) resulted in two distinct clusters and E. n. var. truxillense, was negligible (distance = 0060), at the coefficient level around 036 (Fig. 1). The first cluster whereas the distance between the two varieties in E. coca includes most accessions in E. coca, while the second clus- was substantial (distance = 0299; Table 5). This demon- ter includes most accessions in E. novogranatense. Within strates that the differentiation between E. c. var. coca the cluster of E. coca, almost all accessions from the two and E. c. var. ipadu contributed most to the inter-variety E. coca varieties grouped within their own taxon, which variability in this analysis. The inter-individual variation supports their current systematic status as two different also differed among the four taxa. E. n. var. novogranatense varieties. However, in the cluster of E. novogranatense, had the largest mean squared deviation (287), whereas a high frequency of overlapping occurred between E. n. E. c. var. ipadu had the smallest one (237). E. c. var. coca var. novogranatense and E. n. var. truxillense (Fig. 1). and E. n. var. truxillense had intermediate mean squared deviation levels of 261 and 264, respectively (Table 4). Genetic diversity within and among the four Erythroxylum taxa DISCUSSION Mean genetic distance, as another measure of internal genetic diversity within each taxon, varied significantly The neotropical taxa of Erythroxylum have been cultivated across the four varieties (Table 3). While there were no in South America for at least 5000 years. While intensive significant differences between E. c. var. coca, E. n. var. ethnobotanical studies have been conducted on these taxa novogranatense, and E. n. var. truxillense, the accessions of (Plowman, 1984), the taxonomic status, evolutionary rela- E. c. var. ipadu had a significantly smaller mean distance tionship, and organization of genetic diversity have been than the rest of the three taxa. poorly understood. Various taxonomic treatments have been The MDS plot, based on Euclidian distance, divided proposed for the cultivated cocaine-bearing Erythroxylum, the 85 accessions (Erythroxylum ulei was excluded in the ranging from one to three separate species (Plowman, 1979, ordination analysis) into two main groups, showing a clear 1981, 1982, 1984; Bohm et al., 1982). The recent treatment Johnson et al. — Variation among Cultivated Coca as Assessed by AFLP 605

Ulei Coca B110LS Coca B102LS Ipadu BJ IP-24 Ipadu BJ IP-25 61 Ipadu B507 74 68 Ipadu F-1 B508 79 Ipadu F-1 B505 Ipadu F-1 B508 71 Ipadu B505 100 13 Ipadu B508 62 11 Ipadu F-1 B504 22 Ipadu F-1 B505 17 34 Ipadu F-1 B505 100 97 Ipadu F-1 B506 51 9 Ipadu B501SS 72 Ipadu B503SS 53 Ipadu F-1 B504 26 14 Ipadu B504 54 Ipadu F-1 B501SS 20 Ipadu F-1 B503SS 53 Ipadu F-1 B501SS Ipadu F-1 B504 58 11 Ipadu F-1 B503SS Ipadu F-1 B508 70 Ipadu F-1 B501SS Ipadu BJ IP-23 Coca B88LS 98 Coca B96LS Coca B80LS 25 Coca B14LS 3 Coca B94LS 54 Coca B63LS 32 1 22 Coca B127LS 20 1 Coca B31LS 49 4 Coca B9LS 34 Coca B56LS 2 9 Coca B150LS 46 Coca B98LS 38 Coca B180LS 18 Coca B105LS Coca B85LS 30 Coca B104LS 2 Coca B60LS 22 Coca B50LS Coca B64LS 61 Coca B91LS Trux B305LS Coca B147RLS Novo B276LS Novo B228-1 LS 100 Novo B205-1 LS Novo B216LS Novo B294LS 25 Novo B295LS 92 22 Novo B223LS 41 Novo B286LS 4 57 6 Novo B236LS Novo B272LS 70 10 Novo B244LS 3 Trux B312LS 65 78 20 Trux B316 Novo B300LS 12 15 Novo B251LS 16 47 5 Novo B278LS Novo B242LS 23 Novo B253LS 18 Novo B233LS 63 Novo B292LS Trux B322LS 2 Trux B303LS 26 25 Trux B317SS 60 19 Trux B321LS 15 61 Trux B319LS 12 Trux B318SS 6 Trux B315SS 25 Trux B314SS Novo B206LS 14 47 Novo B201LS 2 Novo B201-1 LS Trux B302LS 85 Trux B323LS Novo B245LS 68 Novo B225LS Novo B220LS 65 Novo B205LS

0·20 0·360·52 0·68 0·84 1·00 Coefficient

F IG. 1. Dendrogram of 86 Erythroxylum c. var. coca, Erythroxylum c. var. ipadu, Erythroxylum n. var. novogranatense, E. n. var. truxillense and Erythroxylum ulei samples from the living collection at the Beltsville Agricultural Research Center. Accession labels correspond to the sample list in Table 1. recognizes two species (Erythroxylum coca var. coca Lam 1982; Johnson, 1997, 1998, 2002, 2003b). All four of these and Erythroxylum novogranatense var. novogranatense taxa have the same chromosome number, n = 12 and (Morris) Hieron) each with a variety (Erythroxylum coca are known only in cultivation, or as semi-wild escapees var. ipadu Plowman and Erythroxylum novogranatense var. from cultivation (Plowman, 1979). Based on the results truxillense (Rusby) Plowman (Plowman, 1981; Bohm et al., of artificial hybridization and the comparison of flavonoids 606 Johnson et al. — Variation among Cultivated Coca as Assessed by AFLP

T ABLE 3. Comparison of mean genetic distances of four T ABLE 4. Analysis of molecular variance (AMOVA) for AFLP Erythroxylum taxa variation among and within four Erythroxylum taxa

Mean distance Variance among Source of variation d.f. SSD* MSDy component % Totalz P valuex No. of individuals Taxa individuals within taxa Between species 1 1659.4 1659.432.235.7 <0.001 Among variety 2 569.8 284.911.813.0 <0.001 Erythroxylum coca var. coca Lam 23 0.347a within species Erythroxylum novogranatense var. 25 0.358a Individual within 81 3743.346.227.551.3 <0.001 novogranatense (Morris) Hieron variety Erythroxylum novogranatense var. 13 0.331a E. c. var. coca 22 959.126.1 truxillense (Rusby) Plowman E. c. var. ipadu 23 928.323.7 Erythroxylum coca var. ipadu Plowman 24 0.302b E. n. var. 24 1215.228.7 novogranatense . . a,bValues in the columns followed by the same letter are not significantly E. n. var. 12 640 6264 different at the 005 level from the other values in the same column. truxillense Total 84 5972.6

10 E. n. var. novogranatense *Sum of squared deviations. E. n. var. truxillense y Mean squared deviations. E. c. var. coca z Percentage of total molecular variance. E. c. var. ipadu x Probability of obtaining a larger component estimate. Number of BJ IP-23 5 B147RLS permutations = 1000.

0·0 T ABLE 5. Genetic distances between the cultivated BJ IP-24 B507 Erythroxylum taxa BJ IP-25 Dimension 2 −5 Name of taxa Genetic distance

E. c. var. coca vs. E. c. var. ipadu 0.299 −10 E. c. var. coca vs. E. n. var. novogranatense 0.451 E. c. var. coca vs. E. n. var. truxillense 0.459 −10 −5 0·0 5 10 E. c. var. ipadu vs. E. n. var. novogranatense 0.510 Dimension 1 E. c. var. ipadu vs. E. n. var. truxillense 0.519 E. n. var. novo. vs. E. n. var. truxillense 0.060 F IG. 2. Multidimensional scaling plot of 85 Erythroxylum accessions based on Euclidian distance calculated from AFLP data (MDS badness of fit = 0229). All accession identifications correspond to sample list in Table 1. and genetic separation from E. coca (Fig. 2) suggests that E. novogranatense may not be derived from E. coca. The in the four taxa, Bohm et al. (1982) proposed a hypothesis geographical distribution of these two taxa is completely of a linear evolution series suggesting that (a) E. n. var. allopatric (Plowman, 1979). The ecological habitat of novogranatense and E. n. var. truxillense are two varieties E. novogranatense is predominately in northern Peru, of a species distinct from E. coca;(b) E. c. var. ipadu was Ecuador, Colombia, and Venezuela, ranging from desert independently derived from E. c. var. coca;(c) E. c. var. to moist forest, whereas E. coca inhabits the pre-montane´ coca is the ancestral taxon from which E. n. var. truxillense wet forest of Ecuador, Peru and Bolivia, as well as the was derived; and (d ) E. n. var. novogranatense was derived lowland rain forest of the Amazonian Basin (Ganders, from E. n. var. truxillense. 1979; Schultes, 1981; Plowman, 1982, 1984). This geo- In the present study, 86 accessions were genotyped from graphical barrier between the two taxa suggests that the living collection of Erythroxylum at the Beltsville E. novogranatense may have an independent origin from Agricultural Research Center using AFLP, and this result E. coca, as suggested by Bohm et al. (1982), but later provided insight into the taxonomic relationships and evolu- refuted in favour of the E. coca lineage. tion. Overall, the pattern of genetic variation in these taxa At the inter-variety level, a significant differentiation agrees in principle with known geographic and phenotypic was detected between E. c. var. ipadu and E. c. var. coca data. There is a clear separation between E. coca and (Table 5 and Fig. 2). Erythroxylum c. var. ipadu is consid- E. novogranatense (Table 5 and Figs 1 and 2), thus support- ered to be a cultigen of E. c. var. coca based on their ing the current taxonomic treatment recognizing E. coca morphological similarities and identical flavonoid profiles and E. novogranatense as two different species. However, (Bohm et al., 1982). The significantly lower level of genetic the level of genetic diversity present in E. novogranatense diversity in E. c. var. ipadu, relative to that in E. c. var. coca, is as high as that of E. coca, as quantified by the mean supports the current view that E. c. var. ipadu is a derived genetic distance and the mean squared deviation from cultigen of E. c. var. coca. Its lower diversity, in conjunction AMOVA (Tables 3 and 4). This high level of diversity in with its clear separation from E. c. var. coca, is most likely E. novogranatense, together with its significant geographic to be the result of a founder effect as E. c. var. ipadu moved Johnson et al. — Variation among Cultivated Coca as Assessed by AFLP 607

AB C

F IG. 3. Three Erythroxylum accessions from the multidimensional scaling plot: (A) E. n. var. truxillense (B305LS); (B) E. c. var. coca (B147RLS) and (C) E. c. var. coca (B14LS). Note the phenotypical similarity between (A) and (B) and the dissimilarity to (C). B147RLS was shown to be more distanced from B14LS, but taxonomically identified as E. c. var. coca in the living collection at the Beltsville Agricultural Research Center. from the eastern Andes in Peru and Bolivia into the lowland and BJIP-25 where they appear as intermediates between Amazonian basin. Since E. c. var. ipadu can only be pro- E. c. var. coca and E. c. var. ipadu (Table 1 and Figs 1 pagated vegetatively, it is unable to reproduce in nature and 2). According to Bohm et al. (1982), where there are (Plowman, 1979, 1982, 1984). This cultigen was tradition- barriers to out-crossing, hybrid weakness and reduced ally cultivated under conditions of shifting ‘slash and burn’ hybrid fertility have occurred and three of the four culti- agriculture by Amazonian tribes, where cultivation plots vated coca taxa have become genetically isolated from each were shifted annually or biennially (Plowman, 1982). other (E. c. var. coca, E. n. var. novogranatense and E. n. var. This continued artificial breeding and migration is likely truxillense). Bohm (1982) and Johnson et al. (1997, 1998, to result in future genetic drift and population isolation. 2002, 2003a; Johnson and Schmidt, 1999) characterized the Within E. novogranatense, the rational of recognizing leaf flavonoids of the four cultivated Erythroxylum taxa. E. n. var. truxillense as a different variety is questionable. Johnson et al. (1997, 1998, 2002, 2003a) and Johnson As shown in the MDS plot and the result of AMOVA and Schmidt (1999) showed that each taxon possessed (Table 5 and Fig. 2), the genetic distance between E. n. distinct leaf flavonoids, as well as unique morphological var. novogranatense and E. n. var. truxillense is negligible. features. Johnson et al. (2002) further compared the leaf The two varieties are known to have similar morphological flavonoids in E. c. var. ipadu (collected about 1970) characters and crosses between the two varieties are capable from the living greenhouse collection at the Beltsville of producing morphologically normal and fully fertile F1 Agricultural Research Center with those of E. c. var. hybrids (Bohm et al., 1982). Therefore, main genetic ipadu currently under cultivation in Colombia and found differences that may justify the separation of E. n. var. that they differed greatly. The flavonoids of the greenhouse- truxillense from E. n. var. novogranatense is that the grown E. c. var. ipadu were similar to those present in former has better adaptability to drier ecological habitats E. c. var. coca (Johnson et al., 1998), indicating that the in northern Peru and southern Ecuador, as well as the dis- former should be derived from the latter, whereas the tinct leaf flavonoids (Johnson et al., 1997, 1998, 2002, E. c. var. ipadu currently under cultivation in Colombia 2003a). contained a mixture of the flavonoids found in both While there are distinguishable AFLP profiles for the E. c. var. coca and E. n. var. truxillense, as well as a cocaine-bearing Erythroxylum, both at species and variety new found flavonoid not present in E. c. var. coca or level, it is noteworthy that some outlier accessions do not fit E. c. var. ipadu in the living collection at the Beltsville into the typical variety or species proximity. This happened Agricultural Research Center. This suggested that the mostly in the two varieties of E. coca, which may have material presently in cultivation has resulted in a hybrid practical implications for monitoring the coca production from a cross between the E. c. var. coca and E. n. var. in the Andes. The E. c. var. coca accession B147RLS is truxillense (Johnson et al., 2002, 2003a). The leaf flavonoid grouped toward E. novogranatense (Figs 1, 2 and 3B). This data coincides with the genetic distance data in the current is of interest because, in their breeding experiment, Bohm study (Table 5 and Fig. 2), wherein B147RLS is represent- et al. (1982) showed that crosses between E. c. var. coca and ative of E. c. var. ipadu currently under cultivation in E. n. var. novogranatense were incompatible, while those Colombia (Table 1 and Figs 1 and 2). The morphology between E. c. var. coca and E. n. var. truxillense were of B147RLS is more similar to that of E. n. var. truxillense compatible. The accession B147RLS, therefore, is likely than to that of E. c. var. coca (Fig. 3) which would account a hybrid between E. c. var. coca and E. n. var. truxillense for its genetic intermediacy (Fig. 2). The separation from (Table 1 and Figs 1 –3). other accessions of the taxon may not be explained by A similar outlier pattern, as described above, is shown population differentiation and it is likely that this accession for E. c. var. ipadu accessions B507, BJIP-23, BJIP-24 represents a hybrid between E. c. var. coca and E. n. var. 608 Johnson et al. — Variation among Cultivated Coca as Assessed by AFLP truxillense, as was also suggested by the authors’ previous Johnson EL, Schmidt WF. 1999. Flavonoids as chemotaxonomic markers flavonoid data (Johnson et al., 2002). for Erythroxylum ulei. Zeitschrift fuur€ Naturforschung 54c: 881–888. It was stated above that BJIP-23, BJIP-24 and BJIP-25 are Johnson EL, Schmidt WF, Cooper, D. 2002. Flavonoids as chemotaxo- nomic markers for cultivated Amazonian coca. Plant Physiology and intermediates between E. c. var. coca and E. c. var. ipadu Biochemistry 40: 89–95. (Table 1 and Figs 1 and 2). However, BJ IP-23 is separated JohnsonEL, SchmidtWF, EmcheSD, MossobaMM, Musser SM. 2003a. from BJIP-24 and BJIP-25 and grouped within E. c. var. coca Kaempferol (rhamnosyl) glucoside, a new flavonol from Erythroxylum (Fig. 2). Hitherto, there are no reported crosses between E. c. coca var. ipadu. Biochemical Systematics and Ecology 31: 59–67. Johnson EL, Schmidt WF, Norman HA. 1997. Leaf flavonoids as var. coca and E. c. var. ipadu, and Bohm et al. (1982) con- chemotaxonomic markers for two Erythroxylum taxa. Zeitschrift f€uur cluded that preliminary crosses between E. c. var. ipadu were Naturforschung 52c: 577–585. self-incompatible. At the Hawaiian field site, it was observed Johnson EL, Schmidt WF, Norman HA. 1998. Flavonoids as markers for Erythroxylum taxa: E. coca var. ipadu and E. novogranatense that E. c. var. ipadu F1 progeny clustered (AFLP) well with E. n. var. novogranatense (Johnson et al., 2003b). This may var. truxillense. Biochemical Systematics and Ecology 26: 743–759. Nei M, Li W. 1979. Mathematical model for studying genetic variation indicate that BJ IP-23 is a cross between E. c. var. ipadu in terms of restriction endonucleases. Proceedings of the National and E. n. var. novogranatense. Therefore it is distanced Academy of Sciences of the USA 76: 5269–5273. between E. c. var. coca and E. n. var. novogranatense Nienhuis J, Tivang J, Skroch P. 1994. Analysis of genetic relationships (Fig. 2). In future research it is intended to examine the among genotypes based on molecular marker data. In: American Society for Horticultural Science, and American Genetic Association relationship between the parental, F1 and F2 progeny in (Eds.), Analysis of Molecular Marker Data. Joint Plant Breeding crosses between the four cultivated Erythroxylum taxa. Symposia Series, Crop Science Society of America. Corvallis, OR, 8–14. Plowman T. 1979. The identity of Amazonian and Trujillo coca. Botanical Museum Leaflets 27: 45–68. LITERATURE CITED Plowman T. 1981. Amazonian coca. Journal of Ethnopharmacology 3: Bohm BA, Ganders FR, Plowman T. 1982. Biosystematics and evolution 195–225. of cultivated coca (Erythroxylaceae). Systematic Botany 7: 121–133. Plowman T. 1982. The identification of coca (Erythroxylum species): Dubreuil P, Dillman C, Crossa J, Warburton M. 2002. LCCMV software 1860–1910. Botanical Journal of the Linnean Society 84: 329–353. for the calculation of molecular distances between varieties (1st edn). Plowman T. 1984. The ethnobotany of coca (Erythroxylum spp., Erythrox- Mexico D.F. CIMMYT. ylaceae). Advances in Economic Botany 1: 62–111. Excoffier L, Smouse PE. 1994. Using allele frequencies and geographic Plowman T, Rivier L. 1983. Cocaine and cinnamoylcocaine content of subdivision to reconstruct gene trees within a species: molecular Erythroxylum species. Annals of Botany 51: 641–659. variance parsimony. Genetics 136: 343–359. Rury RP. 1981. Systematic anatomy of Erythroxylum P. Browne: practical Excoffier L, Smouse PE, Quattro JM. 1992. Analysis of molecular and evolutionary implications for the cultivated cocas. Journal of variance inferred from metric distances among DNA haplotypes: Ethnopharmacology 3: 229–263. applications to human mitochondria DNA restriction data. Genetics SAS 1999. SAS Version 8.02: SAS/STAT Software: Changes and Enhance- 131: 479–491. ments through Release 8.02. SAS Institute Inc. Cary, NC. Ganders FR. 1979. Heterostyly in Erythroxylum coca (Erythroxylaceae). Schultes RE. 1981. Coca in the Northwest Amazon. Journal of Ethnophar- Botanical Journal of the Linnean Society 78: 11–20. macology 3: 173–194. Johnson EL, Saunders JA, Mischke S, Helling CS, Emche SD. 2003b. Vos P, HogersR, Bleeker M, ReijansM, vande Lee T, Hornes M, Frijters Identification of Erythroxylum taxa by AFLP DNA analysis. Phyto- A, Pot J, Peleman J, Kulper, M, et al. 1995. AFLP: a new technique chemistry 64: 187–197. for DNA fingerprinting. Nucleic Acids Research 23: 4407–4414.