SYSTEMATICS, MORPHOLOGY AND PHYSIOLOGY

DNA Barcoding of an Assembly of Montane Andean Butterflies(Satyrinae): Geographical Scale and Identification Performance

MA MARÍN1,2 , IC CADAVID2, L VALDÉS2, CF ÁLVAREZ

2,3, SI URIBE2, R VILA4, TW PYRCZ5

1Depto de Biologia Animal, Instituto de Biologia, Univ Estadual de Campinas – UNICAMP, Campinas, SP , Brazil2Univ Nacional de Colombia, Sede Medellín, Grupo de Investigación en Sistemática Molecular, Medellín, Colombia3Corporación Universitaria Lasallista, Antioquia, Colombia4Instituto de Biología Evolutiva (CSIC-UPF), Barcelona, Spain5Zoological Museum, Jagiellonian Univ, Kraków, Poland

AbstractKeywords

Pronophilina, Morpho, Forsterinaria,mitochondrial DNA, molecular taxonomy,andean cloud forest, community ecology

CorrespondenceMA Marín, Depto de Biologia Animal,Instituto de Biologia, Univ Estadual deCampinas – UNICAMP, Rua Monteiro Lobato,255 - Cidade Universitária Zeferino Vaz - BarãoGeraldo, CEP, Campinas, SP 13083-970, Brazil;mamarin0@unal.edu.co

Edited by Alberto S Corrêa – ESALQ/USP

Received 15 June 2016 and accepted 28December 2016

* Sociedade Entomológica do Brasil 2017

DNA barcoding is a technique used primarily for the documentation andidentification of biological diversity based on mitochondrial DNA se-quences. Butterflies have received particular attention in DNA barcodingstudies, although varied performance may be obtained due to differentscales of geographic sampling and speciation processes in various groups.The montane Andean Satyrinae constitutes a challenging study group fortaxonomy. The group displays high richness, withmore of 550 species, andremarkable morphological similarity among taxa, which renders their iden-tification difficult. In the present study, we evaluated the effectiveness ofDNA barcodes in the identification of montane Andean satyrines and theeffect of increased geographical scale of sampling on identification perfor-mance.Mitochondrial sequences were obtained from 104 specimens of 39species and 16 genera, collected in a forest remnant in the northwestAndes. DNA barcoding has proved to be a useful tool for the identificationof the specimens, with a well-defined gap and producing clusters withunambiguous identifications for all the morphospecies in the study area.The expansion of the geographical scale with published data increasedgenetic distances within species and reduced those among species, butdid not generally reduce the success of specimen identification. Only inForsterinaria rustica (Butler, 1868), a taxon with high intraspecific varia-tion, the barcode gap was lost and low support for monophyly was ob-tained. Likewise, expanded sampling resulted in a substantial increase inthe intraspecific distance in Morpho sulkowskyi (Kollar, 1850);Panyapedaliodes drymaea (Hewitson, 1858); Lymanopoda obsoleta(Westwood, 1851); and Lymanopoda labda Hewitson, 1861; but for thesespecies, the barcode gap was maintained. These divergent lineages arenonetheless worth a detailed study of external and genitalic morphologyvariation, as well as ecological features, in order to determine the poten-tial existence of cryptic species. Even including these cases, DNA barcodingperformance in specimen identification was 100% successful based onmonophyly, an unexpected result in such a taxonomically complicatedgroup.

Neotrop EntomolDOI 10.1007/s13744-016-0481-z

Introduction

Andean Satyrinae

In the Neotropics, the subfamily Satyrinae accounts for morethan 1200 known species in five tribes (Morphini, Brassolini,Haeterini, Melanitini, and Satyrini). Two Satyrini subtribes en-compass most of the species: Euptychiina, with more than 400species, is a group of mostly lowland or premontane butterflies(Viloria 2003), particularly diversified in the basin of the Amazon,Atlantic Forest, and the foothills of the Andes, and Pronophilina,more diverse in the highlands, is the richest group of montanebutterflies with over 550 described species (Lamas et al 2004),and presents the greatest diversity in the tropical Andes.

Montane satyrines display high morphological similarityamong congeneric species (Forster 1964). For example, inEuptychiina, the genus Forsterinaria Gray, 1973 the onlystrictly montane within the subtribe, presents a demandingtaxonomy (Zubek et al 2014). Its species show little differ-ence in wing pattern and even the genitalia provide fewdiagnostic characters for the immediate segregation of someallopatric or sympatric species (Zubek & Pyrcz 2011). In somePronophilina, the situation is similar, particularly notable indiverse genera as Pedaliodes Butler, 1867; Eretris Thieme,1905; and Manerebia Staudinger, 1897. These exhibit a re-markable external similarity among many species, which canonly be identified with certainty through dissection, and thisfact has led to much confusion in the literature and resultedin a significant underestimation of their taxonomic diversity(Pyrcz et al 2006).

Additionally, montane satyrines present an intricate alti-tudinal distribution. Pronophilina butterflies occur predomi-nantly in high elevation cloud forests near or within theforest-paramo ecotone, but they are distributed in narrowbands of altitude (Adams 1985, Pyrcz & Wojtusiak 2002,Pyrcz et al 2009, Casner & Pyrcz 2010, Viloria et al 2010,Pyrcz & Garlacz 2012, Marín et al 2015) and demonstrate alarge percentage of endemic species, frequently confined tosingle massifs (Pyrcz & Rodríguez 2007, Pyrcz et al 2013,2016).Within each part of the Andes, lower-elevation specieswith broad geographic ranges are replaced by higher-eleva-tion, narrowly distributed congeners. The result is a stair-stepdistribution in mountain sides with unique species composi-tion in adjacent regions of an extended mountain chain or innearby cordilleras (Adams 1985, Pyrcz et al 1999, Casner &Pyrcz 2010).

DNA Barcoding in Butterflies

Butterflies have been widely used in DNA barcoding studies(Hebert et al 2004, Elias et al 2007, Janzen et al 2009,Linares & Soto-Calderón 2009, Lukhtanov et al 2009, Silva-Brandão et al 2009, Dasmahapatra et al 2010, Prado et al

2011, Ashfaq et al 2013, Wilson et al 2013, Dincă et al 2011,2015, Yang et al 2015), and a total of 148,306 DNA barcodesare available for named butterflies specimens (Papilionoidea)in the Barcode of Life Database (BOLD Systems v3, accessedon October 6, 2016). However, some studies have shownthat DNA barcodes are not completely accurate when iden-tifying specific taxonomical groups. Elias et al (2007), report-ed in Ithomiini (Danainae) that only 77% of species can beprecisely identified using the barcode data at local scale(21 km2), while Wiemers & Fiedler (2007) reported 84% ofaccurately identified Polyommatinae (Lycaenidae) at a conti-nental scale.

Dasmahapatra et al (2010) demonstrated that deep ge-netic divergences detected with mtDNA barcoding in thegenusMechanitis Fabricius, 1807 (Ithomiini), were not alwaysreflected in the nuclear genome and suggested the use ofnuclear genes as a check step when mtDNA barcoding givesunexpected results. However, other studies supported theeffectiveness of DNA barcoding for attribution of specimensto species, with an accuracy between 90–100% (Lukhtanovet al 2009, Prado et al 2011, Ashfaq et al 2013, Wilson et al2013, Dincă et al 2011, 2015), and its usefulness in the studyof species complexes (Hebert et al 2004, Cong & Grishin2014, Seraphim et al 2014). The variable performance ofDNA barcoding has been attributed to factors such as sam-pling, geographic scale, time divergence among species, mi-tochondrial introgression, and dominant mode of speciation(Moritz & Cicero 2004, Elias et al 2007, Wiemers & Fiedler2007, Bergsten et al 2012, Talavera et al 2013). Interestingly,DNA barcode libraries based on regional sampling may stillbe effective at much wider geographic scales (Huemer et al2014). Without a cause that fully explains the variability inthe performance of DNA barcoding, it is necessary to test itsutility for different taxa, and under different ecological andsampling conditions.

Neotropical Satyrinae constitute an interesting modelgroup for testing the performance of DNA barcoding becauseof high diversity and notable morphological stasis. However,and in contrast with other butterflies groups, montaneAndean satyrines have received little attention in molecularstudies (e.g., Casner & Pyrcz 2010). Accordingly, in the pres-ent study, we evaluated the effectiveness of DNA barcodeson the identification of montane Andean satyrines and theeffect of increased geographical sampling scale on the per-formance of this methodology.

Material and Methods

Specimens sampling

Butterflies were collected in a forest remnant of 51.71 km2

close to Medellín city in the Colombian Central Cordillera

Marín et al

of the Andes (6°05′–6°13′N/75°39′–75°43′W), fromMarch 2011to February 2012 at 2500–2890 m. In this area, maximumannual rainfall is approximately 3000 mm (Corantioquia2004, Hermelin 2007). According to Holdridge (1967), this lifezone corresponds to montane rain forest and lower montanewet forest.

External morphology was used to examine and iden-tify collected specimens, in addition for Pedaliodes spe-cies, internal genital morphology was examined at theLaboratorio de Biología y Sistemática de Insectos(Insectario) of the Universidad Nacional de Colombia,Sede Medellín and the Zoological Museum of theJagiellonian University (Kraków, Poland). Each specimenstudied was labeled, assigned a code number, and de-posited in the Museo Entomológico Francisco LuisGallego (MEFLG, Medellín, Colombia). All sequencesgenerated in this study were deposited in GenBankunder the following accession numbers: KU359838–KU359939, and KU588156–KU588157 with images ofvouche r spec imens in Ba r code o f L i f e Da taSystems—BOLD (http://boldsystems.org/). To test theeffect of expanded geographic sampling on barcodeperformance, sequences for an addit ional 157specimens of 83 species and 18 genera, sampled fromPeru to Venezuela were obtained from GenBank andincluded in the analysis (Table S1). The identificationof these specimens was verified, with the photos de-posited in the NSG Voucher Specimen Databasehttp://www.nymphalidae.net/all_species.php (accessedon September 14, 2016) (Peña & Malm 2012).

DNA extraction and PCR amplification

Two legs were removed from each specimen with ster-ile forceps and transferred to microtubes of 1.5 μl andpreserved at −20°C. Total genomic DNA was extractedusing the grind buffer method described by Rosero et al(2010). The PCR reaction had a final volume of 50 μland included: 2 mM MgCl2, 1× Buffer, 0.2 mM dNTPs,0.2 μM of each primer, 0.04 U/μl of Taq polymerase(Fermentas, St. Leon, Germany), and 4 μL of DNA. Theprimers used were LCO1490 5′ GGTCAACAAATCATAAAGATATTGG 3′ and HCO2198 5′ TAAACTTCAGGGTGACCAAAAAATCA 3′ (Bogdanowicz et al 1993, Folmeret al 1994). Amplification was verified in gel electro-phoresis with 1.2% agarose, TBE and 80 V, and visual-ized using Gel Star™ (1: 100). The PCR products werepurified on Millipore filtration plates and sequencedusing the service Macrogen Inc., Korea. The ampliconswere sequenced in both directions. Sequence quality wasevaluated with the program FinchTV 1.3.1 (Geospiza Inc.),and aligned using the software MUSCLE (Edgar 2004).

Analysis of genetic variability

Pairwise distances were calculated in the software MEGA 5.2(Tamura et al 2011) using the uncorrected p distance modelaccording to Collins et al (2012) and Srivathsana and Meiera(2012), remaining analyses were conducted in the R packageSpider (Brown et al 2012). For analytical purposes, a distancematrix of combined dataset (specimens local study area plusGenBank) was generated with pairwise deletion of missingsites. For each species, mean intraspecific divergence, maxi-mum intraspecific divergence, and minimum interspecificdistances were computed.

Evaluation of DNA barcoding utility for species identification

The existence of a “barcode gap” was checked, by plottingmaximum intraspecific divergences against minimum inter-specific distances, with a 1:1 slope representing the point atwhich the difference between them is zero and there is “nobarcode gap.” Two datasets were used to test for the exis-tence of a barcode gap: one dataset with only specimens ofthe local study area and the other with a combined datasetat wider geographical scale. Barcode efficacy was testedusing distance-based methods, enabling to make specimenidentifications even in some instances in the presence ofparaphyly (Collins & Cruickshank 2013). Based in the conceptthat the distances between three sequences do not have tobe equilateral and a fixed threshold value cannot be main-tained (Meier et al 2006), this defines how similar (thresholdvalue) a set of barcodes needs to be before it can be identi-fied to be the same species (Zhang et al 2012). The thresholdvalue can be estimated for a given data set by obtaining afrequency distribution of all intraspecific pairwise distancesand determining the threshold distance below 95% of allintraspecific found distances (Meier et al 2006, Zhang et al2012). We used the threshID function because it considers allsequences within the given threshold and makes an analysissimilar to the “Identify Specimen” tool provided by the BOLD(Brown et al 2012). The efficiency in specimen identificationwith the combined dataset was analyzed using an optimumthreshold value. The identification status of each specimenwas coded as: “correct” if the species within the giventhreshold were the same, “incorrect” if the species withinthe given threshold were different, “ambiguous” if therewere more than one species within the given threshold,and “no identification” if no species were within the thresh-old distance. The cumulative error based on false positivesand false negatives for each threshold value among 0 and0.04 (8% of genetic distance) was estimated to determinatean optimum threshold value (Meyer & Paulay 2005).

Additionally, in order to determine if species were recov-ered as monophyletic, a monophyly test was conducted. Thisincorporates a bootstrap resampling (1000 replicates) to

DNA Barcoding of Andean Satyrinae

determine the support for the monophyly. Species with abootstrap support lower than “thresh” (70%) were codedas “with weak support”. Two trees were obtained, aNeighbor Joining NJ tree with the software MEGA 5.2(Tamura et al 2011) using uncorrected p distances, and amaximum likelihood (ML) tree, with 200 heuristic searchesof 20,000 generations, using the model GTR + G in the pro-gram GARLI 2.0 (Zwickl 2006), both with bootstrap resam-pling of 1000 replicates.

Results

DNA barcode sequences greater than 500 base pairs wererecovered from 104 specimens, representing at least one

sequence for each of the 39 species and 16 genera studied(Table S2). For the local study area, the minimum interspe-cific divergence was 3.46% between Pronophila orcus(Latreille 1813) and Pronophila epidipnis (Thieme, 1907), andthe maximum intraspecific distance was 1.2% in Lymanopodalabda (Hewitson, 1861), resulting in a well-defined barcodegap (Fig 1a). With increased geographical sampling, the spe-cies of the local study area exhibited a minimum interspecificdivergence of 2.7% between Forsterinaria rustica (Butler,1868) and Forsterinaria guaniloi (C. Peña & Lamas, 2005),and the maximum intraspecific distances of 4.0% inMorpho su lkowsky i (Ko l l a r , 1850) and 4.5% inPanyapedaliodes drymaea (Hewitson, 1858) (Table S3). Theintraspecific distance in F. rustica was 3.5%, which is greaterthan its minimum interspecific distance, thus no barcode gap

Fig 1 Evaluation of barcode gap and cumulative error with uncorrected p distance. a Dot plot for each individual of the study site, plotted the distanceof the furthest conspecific against the distance to the nearest nonconspecific, above the 1:1 line indicate the presence of a “barcode gap.” b Dot plotfor each individual with the increased geographic dimension of sampling; black dots are specimens of the study site and white dots are specimens ofothers localities. c Box plots comparing the genetic variation between minimum interspecific distance and maximum intraspecific distance using onlyspecimens of the study site and with the inclusion of specimens of the geographical extension. d Cumulative error based on false positives plus falsenegatives for each threshold value until 0.04.

Marín et al

was detected in this case. Likewise, Lymanopoda labda and L.obsoleta (Westwood, 1851) presented large intraspecific dis-tances of 3.5 and 3.0%, and minimum interspecific distancesof 6.0 and 3.8%, respectively; although in these cases, thebarcode gap was maintained (Fig 1b). In summary, the intra-and interspecific genetic variation that was documented forthe study area did not show significant differences withthose obtained when sequences from the extended geo-graphical area were included (Fig 1c). However, there weresome outliers, namely M. sulkowskyi, F. rustica, L. labda,L. obsoleta, and P. drymaea, with intraspecific distancesgreater than 2%, that overlapped with interspecific variation.

Optimization threshold analysis using the whole datasetreturned an optimum value of 0.016 (3.2% distances) wherethe cumulative error was minimized at 15%. Errors were keptbelow 20% when using threshold values between 0.008(1.6%) and 0.026 (5.2%) (Fig 1d), although the cumulativeerror indicates the possible presence of misidentified taxa,the concordance with independent morphological identifica-tion indicates that it is unlikely to be misidentifications. Usingthe lowest threshold of 1.6%, all species were correctly iden-tified. Ambiguity in identification started to appear at 2.8%for F. rustica and, when increasing the threshold value to4.0%, five more ambiguous identifications appeared: Eretrisporphyria (C. Felder & R. Felder, 1867); Eretris apuleja (C.Felder & R. Felder, 1867); P. epidipnis; P. orcus; andL. obsoleta. The NJ and ML tree (Fig 2, Fig S1) showed allthe specimens clustered according to morphological identifi-cation for both limited and extended datasets, although alower bootstrap support (<70%) was found for three species,M. sulkowskyi, F. rustica, and P. drymaea, and a bootstrapsupport below 50% was found for L. obsoleta.

Discussion

Specimens identification

The DNA barcode was found to be effective in the identifica-tion of montane Andean Satyrinae, showing its usefulness indifferentiating morphologically very similar species (Fig 3). Awell-defined gap was observed and clusters with unambigu-ous identifications for all 39 species from the study areaobtained, invariably matching with morphospecies attribu-tion. Based on cluster analysis, the increase of the geograph-ical area studied, retained the success in specimen identifi-cation, despite the fact that the barcode gap was lost for onespecies (F. rustica). This study contributes to the constructionof a DNA barcode reference library for Neotropical montanebutterflies, with the publication of 29 DNA barcodes for for-merly unavailable species. This is particularly relevant forPedaliodes, the most species-rich genus of Neotropical but-terflies with over 240 known species (Lamas et al 2004), for

which only five species had been DNA barcoded previously(October 6, 2016).

Identifications of congeneric specimens in the most diversegenera (Pedaliodes, Lymanopoda, and Manerebia) were suc-cessful. Similar studies with butterflies had difficulties in identi-fication of congeneric species (Elias et al 2007, Wiemers &Fiedler 2007, Dasmahapatra et al 2010, Prado et al 2011,Kreuzinger et al 2015, Dincă et al 2011, 2015). For instance,Elias et al (2007) studied Ithomiini butterflies at a similar geo-graphical scale (21 km2) and found that the efficiency of speci-men identification was much lower among closely related spe-cies whose biogeographic and evolutionary histories affectedthe barcode performance. It has also been proposed that sym-patric species likely share more barcode sequences than allo-patric species (Lukhtanov et al 2009). Discrete mtDNA clustersprovide strong evidence for independently evolving populationsor species, although apparently their evolution is suppressedeven under very low levels of dispersal (Papadopoulou et al2008). In this study, the identification success of congenericspecimens could be explained by the particular evolutionaryhistory of montane Andean Satyrinae, with independentlyevolving populations and reduced levels of dispersal, forminga speciation pattern predominantly allo- and parapatric, likelyinduced by Pleistocene glacial cycles (Adams 1985, 1986, Casner& Pyrcz 2010). This scenario is evident in the high percentage ofspecies with narrow elevation ranges, which diversified bydispersal into adjacent elevation strata, followed by special-ization and reduced expansion across elevation gradients(Casner & Pyrcz 2010).

In addition to being useful for specimen identification, DNAbarcodes revealed deep intraspecific differences in some spe-cies, which deserve to be studied in more detail. The mostremarkable is F. rustica, which lost the barcode gap, had anambiguous identification, and exhibited low bootstrap sup-port. DNA barcode data suggest that this taxon may, in fact,represent two separate species. However, it is necessary tocarry out a detailed morphological analysis to test this hy-pothesis. Similarly, a substantial increase in the intraspecificdistance was observed for some other species that neverthe-less retained the barcode gap; for example, M. sulkowskyiwith high intraspecific distance between the subspecies (3.3to 4%) and a minimum interspecific divergence from itssister-species Morpho lympharis (Butler, 1873) of 4.3%.However, Cassildé et al (2012) and Nattier et al (2016) con-firmed the separate specific status of M. sulkowskyi andM. lympharis. As our data indicate that their genetic distanceis similar to those among the subspecies of M. sulkowskyi,which may suggest the presence of more species in thegroup. Admittedly, the DNA barcoding alone is not enoughto define the limits between species. Therefore, we suggest adetailed review of subspecies of M. sulkowskyi with a widersampling, the inclusion of nuclear genes and detailed mor-phological and ecological data.

DNA Barcoding of Andean Satyrinae

The intraspecific divergence of more than 3% between geo-graphically isolated specimens of L. obsoleta, L. labda, and

P. drymaea contrasts with those frequently observed amonglowland satyrines, which rarely exceed 2% (Seraphim et al

Fig 2 Maximum likelihood (ML)tree based on the GTR + G modeland obtained using the expandedgeographic sampling matrix of261 sequences from 112 species.The nodes with multiplespecimens or multiple specieswere collapsed to a vertical lineor triangle with the horizontaldepth indicating the level ofdivergence. Bold taxa indicatestudy site specimens and simpletext taxa are specimens orspecies out of study site. Nodesmarked with a black dot areconflictive identifications in theDNA barcode analysis, the threshindicate the threshold valuewhere the identification status isvalued as “ambiguous,” the boot<70 indicate monophyly test witha bootstrap support lower than70%.

Marín et al

2014). Such deep divergences are frequently interpreted as anindication of cryptic species (Prado et al 2011, Cong & Grishin2014, Seraphim et al 2014, Kreuzinger et al 2015). In montanesatyrines, greater intraspecific genetic distances coincide withthe expected isolation and low degree of dispersal of mostmontane species, generally distributed in discrete populations,geographically structured and genetically divergent across therange of the morphologically coherent lineages. Consideringthis, it is necessary to analyze more specimens including adetailed study of external morphology, genitalia, and ecologi-cal features to properly evaluate their taxonomic status.

DNA barcoding in the study and conservation of AndeanSatyrines

The results of this study show that DNA barcoding is a usefultechnique for the identification of montane AndeanSatyrinae, especially that the increased geographical scaledid not significantly reduce the success of specimen identifi-cation. Additionally, the results are consistent with the defi-nition of DNA barcoding sensu lato, with identification ofspecimens at different taxonomical level (Valentini et al2009), that help to glimpse the genetic variation in speciesand genera. For Andean satyrines, the barcodes allowed to

identify intraspecific taxonomical levels, with several speciesshowing intraspecific clusters with large genetic distances(Supporting information). These clusters recognize allopatricsubspecies for M. sulkowskyi, and evolutionary significantunits for F. rustica, Lymanopoda ionius, and L. obsoleta(Westwood, 1851). Thus, our results provide evidence fordistinguish geographic units of biodiversity, and identify ge-netic lineages that reflect phylogeographic processes.

The presence of monophyletic intraspecific clusters in mid-elevations species typical of open areas and secondary forestsuch asM. sulkowskyi, F. rustica, and L. obsoleta, indicate a localand regional structuring of the genetic diversity even in widelydistributed Andean satyrines, with lineages suggesting restrict-ed gene flow. This, added to the high proportion of endemictaxa (Pyrcz 2004, Pyrcz & Rodríguez 2007, Pyrcz et al 2016), andthe existence of unique species assemblages in each mountainsystem (Viloria et al 2010, Marín et al 2015, Pyrcz et al 2016),suggest that the assemblages of Andean satyrines may act asmetacommunities, with defined groups of interacting species inspatially isolated habitat patches. If this hypothesis is confirmed,each mountain system could be considered an independentvaluable unit of biodiversity. Despite some limitations ofsingle-marker approaches, DNA barcoding emerges as a practi-cal tool for comparative phylogeography at the community and

Fig 3 Morphologically similarspecies of Pedaliodes of the localstudy area, dorsal (left) andventral (right): a Pedaliodesrodriguezi Pyrcz & Andrade, 2013.b Pedaliodes hebena Pyrcz &Viloria, 1999. c Pedaliodestransmontana Pyrcz & Viloria,1999. d Pedaliodes phrasicla(Hewitson, 1847). e Pedaliodessimpla costipunctata Weymer,1912. f Pedaliodes obstructa Pyrcz& Viloria, 1999. g Pedaliodesmontagna Adams & Bernard,1981. h Pedaliodes pisonia(Hewitson, 1862). i Pedaliodesobstructa Pyrcz & Viloria, 1999. jPedaliodes praemontagna Pyrcz& Viloria 2007. k Pedaliodespraemontagna Pyrcz & Viloria2007. l Pedaliodes polloniaAdams, 1986.

DNA Barcoding of Andean Satyrinae

metacommunity levels, providing information on the spatialvariation and historical features of landscapes through driftand migration (Joly et al 2014).

DNA barcoding is a technique that extends beyond speci-mens identification. Applied to specific taxonomic groups it pro-vides information for understanding species boundaries and forthe prioritization of biodiversity conservation (Kress et al 2015).Intraspecific variation in DNA barcodes has a unexploited po-tential for ecological applications such as in spatial ecology, eco-evolutionary dynamics, and for investigating the genetic conse-quences of environmental change (Joly et al 2014). Likewise, it isa useful tool for the study of community assembly processesand its phylogenetic structure (Kress et al 2009, Boyle &Adamowicz 2015). Large-scale DNA barcoding allows a simulta-neous assessment of variation in community compositionacross of multiple hierarchical levels and can be used to discernthe effects of neutral and non-neutral macroecological process-es (Baselga et al 2015, Vodă et al 2016).

Acknowledgements The authors would like to thank NoemySeraphim, Leila Shirai, Rita Isabel Veléz, and the anonymous reviewersfor critically reading the manuscript. This work was supported by theCorporación Universitaria Lasallista under grant [FCAA-07 2010] andFundación BBVA (Proyecto MARIPOSA) [BIOCON08_021]. MAM ac-knowledges support of FAPESP (2014/16481-0) for the graduate fellow-ship and CFA does it to COLCIENCIAS (528/2011).

Electronic supplementary material The online version of this article(doi:10.1007/s13744-016-0481-z) contains supplementary material,which is available to authorized users.

References

Adams MJ (1986) Pronophiline butterflies (Satyridae) of the threeAndean cordilleras of Colombia. Zool J Linnean Soc 87:235–320

AdamsMJ (1985) Speciation in the pronophiline butterflies (Satyridae) ofthe northern Andes. J Res Lepid Supplement 1:33–49

Ashfaq M, Akhtar S, Khan AM, Adamowicz SJ, Hebert PDN (2013) DNAbarcode analysis of butterfly species from Pakistan points towardsregional endemism. Mol Ecol Resour 13:832–843

Baselga A, Gómez-Rodríguez C, Vogler AP (2015) Multi-hierarchicalmacroecology at species and genetic levels to discern neutral andnon-neutral processes. Glob Ecol Biogeogr 24:873–882

Bergsten J, Bilton DT, Fujisawa T, Elliott M, Monaghan MT, Balke M,Hendrich L, Geijer J, Herrmann J, Foster GN, Ribera I, Nilsson AN,Barraclough TG, Vogler AP (2012) The effect of geographical scale ofsampling on DNA barcoding. Syst Biol 61:851–869

Bogdanowicz SM, Wallner WE, Bell TM, Harrison RG (1993) Asian gypsymoths (Lepidoptera: Lymantriidae) in North America: evidence frommolecular data. Ann Entomol Soc Am 86:710–715

Boyle EE, Adamowicz SJ (2015) Community phylogenetics: assessing treereconstruction methods and the utility of DNA barcodes. PLoS One10(6):e0126662. doi:10.1371/journal.pone.0126662

Brown SD, Collins RA, Boyer S, Lefort MC, Malumbres-Olarte J, Vink CJ,Cruickshank RH (2012) Spider: an R package for the analysis of speciesidentity and evolution, with particular reference to DNA barcoding.Mol Ecol Resour 12:562–565

Casner KL, Pyrcz TW (2010) Patterns and timing of diversification in atropical montane butterfly genus, Lymanopoda (Nymphalidae,Satyrinae). Ecography 33:251–259

Cassildé C, Blandin P, Silvain J-F (2012) Phylogeny of the genusMorphoFabricius 1807: insights from two mitochondrial genes (Lepidoptera:Nymphalidae). Ann Soc Entomol Fr (N.S.) 48:173–188

Collins RA, Boykin LM, Cruickshank RH, Armstrong KF (2012) Barcoding’snext top model: an evaluation of nucleotide substitution models forspecimen identification. Methods Ecol Evol 3:457–465

Collins RA, Cruickshank RH (2013) The seven deadly sins of DNAbarcoding. Mol Ecol Resour 13:969–975

Cong Q, Grishin NV (2014) A new Hermeuptychia (Lepidoptera,Nymphalidae, Satyrinae) is sympatric and synchronic withH. sosybius in southeast US coastal plains, while another newHermeuptychia species - not hermes - inhabits South Texas andNortheast Mexico. ZooKeys 91:43–91

Corantioquia (2004) Programa integral de educación y comunicación alas comunidades de Angelópolis, Caldas y La Estrella, para laconservación y el manejo sostenible de los recursos naturales de lareserva forestal alto el romeral y la vía parque Angelópolis-Caldas.Fundación con Vida, Medellín, p 204

Dasmahapatra KK, Elias M, Hill RI, Hoffman JI, Mallet J (2010)Mitochondrial DNA barcoding detects some species that are real,and some that are not. Mol Ecol Resour 10:264–273

Dincă V, Montagud S, Talavera G, Hernández-Roldán J, Munguira ML,García-Barros E, Hebert PD, Vila R (2015) DNA barcode referencelibrary for Iberian butterflies enables a continental-scale preview ofpotential cryptic diversity. Sci Rep 5:12395. doi:10.1038/srep12395

Dincă V, Zakharov EV, Hebert PDN, Vila R (2011) Complete DNA barcodereference library for a country’s butterfly fauna reveals high perfor-mance for temperate Europe. Proc Biol Sci 278:347–355

Edgar RC (2004) MUSCLE: multiple sequence alignment with high accu-racy and high throughput. Nucleic Acids Res 32:1792–1797

Elias M, Hill RI, Willmott KR, Dasmahapatra KK, Brower AV, Mallet J,Jiggins CD (2007) Limited performance of DNA barcoding in a diversecommunity of tropical butterflies. Proc Biol Sci 274:2881–2889

Folmer O, BlackM, HoehW, Lutz R, Vrijenhoek R (1994) DNA primers foramplification of mitochondrial cytochrome c oxidase subunit I fromdiverse metazoan invertebrates. Mol Mar Biol Biotechnol 3:294–299

Forster W (1964) Beiträge zur kenntnis der insektenfauna Boliviens XIX.Lepidoptera III Satyridae Veroffentlichungen der ZoologischenStaatssammlung München 8:51–188

Hebert PDN, Penton EH, Burns JM, Janzen DH, Hallwachs W (2004) Tenspecies in one: DNA barcoding reveals cryptic species in the neotrop-ical skipper butterfly Astraptes fulgerator. Proc Natl Acad Sci U S A101:14812–14817

Hermelin M (2007) Valle de Aburrá: ¿Quo vadis? Gestión y Ambiente 10:7–16

Holdridge LR (1967) Life zone ecology. Tropical Science Center, San José,p 206

Huemer P, Mutanen M, Sefc KM, Hebert PDN (2014) Testing DNAbarcode performance in 1000 species of European Lepidoptera: largegeographic distances have small genetic impacts. PLoS One 9:e115774.doi:10.1371/journal.pone.0115774

Janzen DH, Hallwachs W, Blandin P, Burns JM, Cadiou JM, Chacon I,Dapkey T, Deans AR, Epstein ME, Espinoza B, Franclemont JG,Haber WA, Hajibabaei M, Hall JP, Hebert PD, Gauld ID, HarveyDJ, Hausmann A, Kitching IJ, Lafontaine D, Landry JF, Lemaire C,Miller JY, Miller JS, Miller L, Miller SE, Montero J, Munroe E,Green SR, Ratnasingham S, Rawlins JE, Robbins RK, RodriguezJJ, Rougerie R, Sharkey MJ, Smith MA, Solis MA, Sullivan JB,Thiaucourt P, Wahl DB, Weller SJ, Whitfield JB, Willmott KR,Wood DM, Woodley NE, Wilson JJ (2009) Integration of DNAbarcoding into an ongoing inventory of complex tropical biodi-versity. Mol Ecol Resour 9:Suppl s1 1–26

Marín et al

Joly S, Davies TJ, Archambault A, Bruneau A, Derry A, Kembel SW, Peres-Neto P, Vamosi J, Wheeler TA (2014) Ecology in the age of DNAbarcoding: the resource, the promise and the challenges ahead.Mol Ecol Resour 14(2):221–232

KressWJ, García-Robledo C, UriarteM, Erickson DL (2015) DNA barcodesfor ecology, evolution, and conservation. Trends Ecol Evol 30(1):25–35

Kress WJ, Erickson DL, Jones FA, Swenson NG, Perez R, Sanjur O,Bermingham E (2009) Plant DNA barcodes and a community phylog-eny of a tropical forest dynamics plot in Panama. Proc Natl Acad Sci US A 106:18621–18626

Kreuzinger AJ, Fiedler K, Letsch H, Grill A (2015) Tracing the radiation ofManiola (Nymphal idae) butterf l ies: new ins ights fromphylogeography hint at one single incompletely differentiated speciescomplex. Ecol Evol 5:46–58

Lamas G, Viloria AL, Pyrcz TW (2004) Nymphalidae. Satyrinae. TribeSatyrini. Subtribe Pronophilina. In: Heppner JB (ed), Checklist: Part4A. Hesperioidea – Papilionoidea Atlas of the NeotropicalLepidoptera, Volume 5A. Scientific Publisher, Gainesville, pp 206–217

Linares MC, Soto-Calderón ID, Lees DC, Anthony NM (2009) High mito-chondrial diversity in geographically widespread butterflies ofMadagascar: a test of the DNA barcoding approach. Mol PhylogenetEvol 50:485–495

Lukhtanov VA, Sourakov A, Zakharov EV, Hebert PDN (2009) DNAbarcoding central Asian butterflies: increasing geographical dimen-sion does not significantly reduce the success of species identification.Mol Ecol Resour 9:1302–1310

MarínMA, Giraldo CE,Marín AL, Álvarez CF, Pyrcz TW (2015) Differencesin butterfly (Nymphalidae) diversity between hillsides and hilltop for-est patches in the northern Andes. Stud. Neotrop Fauna Environ 50:194–203

Meier R, Shiyang K, Vaidya G, Ng PKL (2006) DNA barcoding and taxon-omy in Diptera: a tale of high intraspecific variability and low identi-fication success. Syst Biol 55:715–728

Meyer CP, Paulay G (2005) DNA barcoding: error rates based on com-prehensive sampling. PLoS Biol 3(12):e422. doi:10.1371/journal.pbio.0030422

Moritz C, Cicero C (2004) DNA barcoding: promise and pitfalls. PLoS Biol2:e354. doi:10.1371/journal.pbio.0020354

Nattier R, Capdevielle-Dulac C, Cassildé C, Couloux A, Cruaud C,Lachaume G, Lamas G, Silvain J-F, Blandin P (2016) Phylogeny anddiversification of the cloud forest Morpho sulkowskyi group(Lepidoptera, Nymphalidae) in the evolving Andes. Zool Scr:1–14.doi:10.1111/zsc.12226

Papadopoulou A, Bergsten J, Fujisawa T, Monaghan MT, BarracloughTG, Vogler AP (2008) Speciation and DNA barcodes: testing the ef-fects of dispersal on the formation of discrete sequence clusters.Philos Trans R Soc Lond Ser B Biol Sci 363:2987–2996

Peña C, Malm T (2012) VoSeq: a voucher and DNA sequence web appli-cation. PLoS One 7(6):e39071. doi:10.1371/journal.pone.0039071

Prado BR, Pozo C, Valdez-Moreno M, Hebert PDN (2011) Beyond thecolours: discovering hidden diversity in the Nymphalidae of theYucatan peninsula in Mexico through DNA barcoding. PLoS One6(11):e27776. doi:10.1371/journal.pone.0027776

Pyrcz TW (2004) Pronophiline butterflies of the highlands ofChachapoyas in northern Peru: faunal survey, diversity and distribu-tion patterns (Lepidoptera, Nymphalidae, Satyrinae). Genus 15(4):455–622

Pyrcz TW, Garlacz R (2012) The presence–absence situation and its im-pact on the assemblage structure and interspecific relations ofPronophilina butterflies in the Venezuelan Andes (Lepidoptera:Nymphalidae). Neotrop Entomol 41:186–195

Pyrcz TW, Rodríguez G (2007) Mariposas de la tribu Pronophilini en laCordillera Occidental de los Andes de Colombia (Lepidoptera:Nymphalidae, Satyrinae). SHILAP Soc Hispano Luso Am Lepid 35:455–489

Pyrcz TW, Wojtusiak J (2002) The vertical distribution of pronophilinebutterflies (Nymphalidae, Satyrinae) along an elevational transect inMonte Zerpa (Cordillera deMérida, Venezuela) with remarks on theirdiversity and parapatric distribution. Glob Ecol Biogeogr 11:211–221

Pyrcz TW, Clavijo A, Uribe S, Marín MA, Álvarez, CF, Zubek A (2016)Páramo de Belmira as an important centre of endemism in the north-ern Colombian Andes: new evidence from Pronophilina butterflies(Lepidoptera: Nymphalidae, Satyrinae, Satyrini). Zootaxa in press

Pyrcz TW, Wojtusiak J, Garlacz R (2009) Diversity and distribution pat-terns of Pronophilina butterflies (Lepidoptera: Nymphalidae:Satyrinae) along an altitudinal transect in North-Western Ecuador.Neotrop Entomol 38:716–726

Pyrcz TW, Prieto C, Viloria ÁL, Andrade-C G (2013) New species of highelevation cloud forest butterflies of the genus Pedaliodes Butler fromthe northern Colombian Andes (Lepidoptera, Nymphalidae,Satyrinae). Zootaxa 3716(4):528–538

Pyrcz TW,Willmott KR, Hall JPW, Viloria ÁL (2006) A review of the genusManerebia Staudigner (Lepidoptera: Nymphalida: Satyrinae) in thenorthern Andes. J Res Lepid 39:37–79

Rosero DA, Gutiérrez LA, Cienfuegos AV, Jaramillo LM, Correa MM(2010) Optimización de un procedimiento de extracción de ADN paramosquitos anofelinos. Rev Colomb Entomol 36:260–263

Seraphim N, Marín MA, Freitas AVL, Silva-Brandão KL (2014)Morphological and molecular marker contributions to disentanglingthe cryptic Hermeuptychia hermes species complex (Nymphalidae:Satyrinae: Euptychiina). Mol Ecol Resour 14:39–49

Silva-Brandão KL, Lyra ML, Freitas AVL (2009) Barcoding Lepidoptera:current situation and perspectives on the usefulness of a contentioustechnique. Neotrop Entomol 38:441–451

Srivathsana A, Meiera R (2012) On the inappropriate use of Kimura-2-parameter (K2P) divergences in the DNA-barcoding literature.Cladistics 28:190–194

Talavera G, Lukhtanov VA, Rieppel L, Pierce NE, Vila R (2013) In theshadow of phylogenetic uncertainty: the recent diversification ofthe Lysandra butterflies through chromosomal changes. MolPhylogenet Evol 69:469–478

Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011)MEGA5: molecular evolutionary genetics analysis using maximumlikelihood, evolutionary distance, and maximum parsimony methods.Mol Biol Evol 28:2731–2739

Valentini A, Pompanon F, Taberlet P (2009) DNA barcoding for ecolo-gists. Trends Ecol Evol 24(2):110–117

Viloria AL (2003) Historical biogeography and the origins of the satyrinebutterflies of the tropical Andes (Lepidoptera: Rhopalocera). In:Morrone JJ, Llorente J (eds) Una perspectiva latinoamericana de labiogeografía. Universidad Nacional Autónoma de México, México, pp247–261

Viloria A, Pyrcz T, Orellana A (2010) A survey of the Neotropicalmontane butterflies of the subtribe Pronophilina (Lepidoptera,Nymphalidae) in the Venezuelan cordillera de la costa. Zootaxa41:1–41

Vodă R, Dapporto L, Dincă V, Shreeve TG, Khaldi M, Barech G, Rebbas K,Sammut P, Scalercio S, Hebert PDN, Vila R (2016) Historical and con-temporary factors generate unique butterfly communities on islands.Scientific Reports 6:28828. doi:10.1038/srep28828

Wiemers M, Fiedler K (2007) Does the DNA barcoding gap exist?—acase study in blue butterflies (Lepidoptera: Lycaenidae). Front Zool4:8. doi:10.1186/1742-9994-4-8

Wilson J-J, Sing K-W, Sofian-Azirun M (2013) Building a DNA barcodereference library for the true butterflies (Lepidoptera) of peninsulaMalaysia: what about the subspecies? PLoS One 8:e79969.doi:10.1371/journal.pone.0079969

Yang M, Zhai Q, Yang Z, Zhang Y (2015) DNA barcoding satyrine butter-flies (Lepidoptera: Nymphalidae) in China. Mitochondrial DNA 27(4):2523–2528

DNA Barcoding of Andean Satyrinae

Zhang A-B, Muster C, Liang H-B, Zhu C-D, Crozier R, Wan P, Feng J, WardRD (2012) A fuzzy-set-theory-based approach to analyse speciesmembership in DNA barcoding. Mol Ecol 21:1848–1863

Zubek A, Pyrcz TW (2011) The synonymy of Forsterinaria difficilis ( Forster )and F. Anachoreta Pulido & Andrade (Lepidoptera: Nymphalidae:Satyrinae). Caldasia 22(2):217–225

Zubek A, Pyrcz TW, Boyer P (2014) Description of a new species of theAndean butterfly genus Forsterinaria gray (Lepidoptera: Nymphalidae)

with considerations on an apparently new structure in male genitalia.Neotrop Entomol 43:68–77

Zwickl DJ (2006) Genetic algorithm approaches for the phylogeneticanalysis of large biological sequence datasets under the maximumlikelihood criterion. Ph.D. thesis, The University of Texas at Austin,Austin, EUA, p 115

Marín et al