Exploring the diversity of andean berries from northern Peru based on molecular analyses

Abstract

More than 12,000 species have been listed under the category of berries, and most of them belong to the orders Ericales and Rosales. Recent phylogenetic studies using molecular data have revealed disagreements with morphological approaches mainly due to diverse floral arrangements, which has proven to be a problem when recognizing species. Therefore, the use of multilocus sequence data is essential to establish robust species boundaries. Although berries are common in Andean cloud forests, diversity of these taxa has not been extensively evaluated in the current context of DNA-based techniques. In this regard, this study characterized morphologically and constructed multilocus phylogenies using four molecular markers, two chloroplast markers (matK and rbcL) and two nuclear markers (ITS and GBSSI-2). Specimens did not show diagnostic features to delimit species of berries. A total of 125 DNA-barcodes of andean berries were newly generated for the four molecular markers. The multilocus phylogenies constructed from these markers allowed the identification of 24 species grouped into the order Ericales (Cavendishia = 1, Clethra = 2, Disterigma = 2, Gaultheria = 4, Thibaudia = 4, Vaccinium = 3) and Rosales (Rubus = 8), incorporating into the Peruvian flora four new records (Disterigma ecuadorense, Disterigma synanthum, Vaccinium meridionale and Rubus glabratus) and revealing the genus Rubus as the most diverse group of berries in the Amazonas region. The results of this study showed congruence in all the multilocus phylogenies, with internal transcribed spacer (ITS) showing the best resolution to distinguish the species. These species were found in coniferous forests, dry and humid forests, rocky slopes, and grasslands at 2,506–3,019 masl from Amazonas region. The integration of morphological and DNA-based methods is recommended to understand the diversity of berries along the Peruvian Andean cloud forest.

Abstract in Quechua language

Qhawarqan astawan chunka iskayniyuq waranqa especiekuna bayasmanta huch’uy mit’a maypichus hatun rak’i chayaqi ordenkunata Ericaleswan Rosaleswan. Chayraqpi Khuski filogeneticamanta rurachiy allincharqan chanikuna molecularkuna willarqan ayñi rikunawanta morfologicokunamanta, qaylla llapan rantichay t’ika tiktutaywan ñawray, ima kay kaqta qhawacgirqan kay huk champay pachaman riqsiypa especiekunamanta. Hina kaqtintaq, chanikuna qatikipaykunamanta multilocus hat’alliy tiksipmi takyachiypaq saywakuna sinchikuna especiekunamanta. Pana bayaskuna kanku allatinkuna sach’a-sach’api phuyusqa anti runap, ñawran manan karqan achka kamaykuy kunan pacha allwiyaraykupi takyasqakuna ADN. Chayrayku, Noqanchispa taqwi allincharqan huk filogenia multilocus, rarachikupúnmi tawa molecular marcadorkuna, caspa iskay markadorkunawan cloroplastomanta (matK, rbcL) iskay markadorkunawan nuclearkunamanta (ITS, GBSSI-2). Kaykunawan filogeniamanta huniqamuran kikinchay iskay chunka tawayoq especies ima tantaqamuran q'anchis generospi (Cavendishia=1, Clethra=2, Disterigma=2, Gaultheria=4, Thibaudia=4, Vaccinium=3, Rubus=8), kaykunata huñuyqamuranta piruwanu llacha kamay tawa musuq quillqakamachikuta (Disterigma ecuadorense, Disterigma synanthum, Vaccinium meridionale, Rubus glabratus). Nocaykuq lluqsisqan kuwirinti rikuchirurqan llapankuna filogeniaspi multilocusmanta, kaspa espaciador transcrito interno (ITS) pi rikuchina kutuwi mihur rantichay riqsiypaq especiekunata.

Abstract in Awajun language

Dekanauwai juú weantug 12000 sag nagkaikiut, júna nejég tente ainawai nuintushkam kuashtai Ericales nuigtu Rosales weantui. Molecularesjai takasmaug juki filogeneticos augtus yamá dekai antugnaiñasmauwa nuna Morfologicosjai disa umikmaug, juka waignawai kuashag yagkunum, juwai dekaata tamanum kuashat utugchata ama nunuka. Nunui asamtai multilocus takasmauwa nujai dekanui wajukut ainawa pipish tumaig aidaush. Tujashkam kuashtai tentee nejég ainaug ikam naig yujagkim amuamua nunuig, wajupá kuashtakit tusajig ashi dekapasjig ADNjain dischamui. Nuni tamaugmak, ii augtusag duka takasé filogenia multilocus dekamua nujai, takasji ipák usumat marcadores molecularesjai, jimag marcadores cloroplastosjai (matK nuigtu rbcL) nuigtu jimag marcadores nuclearesjai (ITS nuigtu GBSSI-2). Juu filogenias dekaji 24 sag nagkaikiut tuwaka 7 generosnug tuwaka awa nunu (Cavendishia=1, Clethra=2, Disterigma=2, Gaultheria=4, Thibaudia=4, Vaccinium=3, Rubus=8), juui dekanai yamajam ipák usumat ajag perunum awanunu (Disterigma ecuadorense, Disterigma synanthum, Vaccinium meridionale nuigtu Rubus glabratus).

Amazonas, Andes, berries, DNA barcoding, Ericales, Rosales.

1. Introduction

Botanically, berries refer to small, rounded, shiny, sweet, sour, multiseeded fruits from different ovaries of a single flower (Mazzoni et al., 2017). In common usage, the meaning of 'berry' certainly differs from this scientific definition. For instance, strawberries, raspberries, and blackberries are considered berries, but these are excluded by botanical circumscription since they are aggregate fruits (Xiang et al., 2017). Berries are consumed worldwide mainly because of the high concentrations of various phytochemicals, such as phenolic compounds, anthocyanins, and flavonoids (Mazzoni et al., 2017; López et al., 2021; Hotchkiss et al., 2021).

Regarding diversity, approximately 12,000 species have been listed under the category of berries (Rose et al., 2018), and most of them belong to the orders Ericales and Rosales (Phipps, 2014; Chase et al., 2016). The larger genera includes species of berries in Ericales are Impatiens (∼1000 spp.), Rhododendron (∼1,000 spp.), Diospyros (∼700 spp.), Erica (∼700 spp.), Vaccinium (∼500 spp.), and Primula (∼400 spp.) (Bouchenak-Khelladi et al., 2015; Schwery et al., 2015). These genera include many well-known tropical and temperate groups that are biogeographically widespread as pantropical and cosmopolitan (Chartier et al., 2017), mainly due to long-distance dispersal or vicariance scenarios (Thomas et al., 2015). In Rosales, the highest diversity of berries includes genera within the family Rosaceae (3,000 species), such as Rubus (∼750 spp.), Potentilla (∼400 spp.), Alchemilla (∼400 spp.), and Prunus (∼200 spp.) (Focke, 1911; Kalkman, 1993; Phipps, 2014). The great diversity of this group is due to polyploidy, agamospermy and constant hybridization of closely related species (Song and Hancock, 2011; Pedraza-Peñalosa and Luteyn, 2011; Mimura and Suga, 2020). Berry diversity in Rosaceae has a wide distribution, particularly in the temperate forests of the Northern Hemisphere (Hummer and Janick, 2009).

The Andean orogeny is considered one of the most significant events for radiation of vascular plants and the biogeographic history of neotropical species (including andean berries) in South America (Barthlott et al., 2011; Luebert and Weigend, 2014). This is clearly observed in Andean cloud forest ecosystems where high levels of biodiversity and endemism have been reported among different group of vascular plants (Ledo et al., 2012). Additionally, in the Peruvian Andes, the highest number of endemic plants has been found on slopes between 2,500 and 3,000 masl (Van der Werff and Consiglio, 2004). Currently, 113 species of berries in Rosaceae (Rosales) and 385 in Ericales have been reported from Peru (Ulloa-Ulloa et al., 2004; León, 2006a), and most of these species have been reported on the basis of morphological analyses alone (Coico et al., 2016). Many of these species are considered endemic and distributed in meso-andean and montane forest regions and others in natural areas (León, 2006a). However, this diversity has not been confirmed molecularly since phenomena such as high phenotypic plasticity or crypticism might over or under represent diversity, respectively (Calderon et al., 2021).

Recent phylogenetic studies of berries within Ericales have revealed disagreement with morphological approaches, such as those reporting wide floral diversity among species (Schönenberger et al., 2010; Rose et al., 2018). Accordingly, the proper taxonomic positions of these species and families were mainly resolved using molecular data (Chartier et al., 2017). The initial classification in Rosales (Rosaceae) was based on morphology and the number of chromosomes (Potter et al., 2002, 2007); however, intergeneric hybridization occurring within subfamilies and tribes has proven to be problematic when delimiting species (Hummer and Janick, 2009). To correct this incongruence, molecular analyses (e.g., plastidial markers, plastomes) are an effective tool for examining systematics (Soltis et al., 2011; Rose et al., 2018; Diaz-Garcia et al., 2021). The molecular markers for phylogenetic analyses that have been most commonly used in berries (Rosales and Ericales) are those corresponding to plastidial regions (matK, ndhF, and rbcL), plastidial intergenic spacers (trnL-trnF, trnS-trnG, psbA-trnH), and nuclear regions (nrITS, GBSSI-2) (Kron et al., 2002; Powell and Kron, 2003; Potter et al., 2007; Soltis et al., 2011; Wang et al., 2016). These barcodes have provided information for testing hypotheses on morphology, genetics and evolutionary relationships in phenotypically diverse groups of Ericales and Rosales (Xiang et al., 2017; Okada et al., 2020).

Although berries are common in Andean cloud forests, the diversity of berries has not been extensively evaluated in the current context of DNA-based techniques. Only few reports on the basis of anatomical observations has been presented. Accordingly, the novelty of this study is to characterize molecularly and determine the phylogenetic positions of berries collected from northern Peru, analyzing the evolutionary relationships of these taxa based on two chloroplast markers (matK and rbcL), the internal transcribed spacer (ITS) region, and nuclear granule-bound starch synthase (GBSSI-2). This is the first integrated study using morphology and the generation of DNA-barcodes to explore the diversity of andean berries from Amazonas region.

2. Materials and methods

2.1. Specimen collection

A total of 48 specimens of Andean berries were sampled from eight localities throughout the province of Chachapoyas, Amazonas, in northern Peru (Molinopampa, Granada, Levanto, Chachapoyas, Maino, Leymebamba, La Jalca, and Huancas; Figure 1). A permit for scientific research on wild flora (RDG N° D000394-2020-MIDAGRI-SERFOR-DGGSPFFS, with authorization code N° AUT-IFL-2020-061) was provided by Servicio Nacional Forestal y de Fauna Silvestre (SERFOR). Tissue samples of approximately 50 mm² were taken from leaf tips for molecular analyses and placed in prelabeled 1.5 mL Safelock Eppendorf tubes. For each site, the date, time, and GPS coordinates were recorded. Photographs were taken to record sampling locations and site features. In addition, inflorescences, leaves, and fruits were collected for morphological examination. Samples were morphologically characterized according to Focke (1910, 1911); Middleton and Wilcock (1990); Sleumer (1967); Middleton (1991); Smith (1933); Kron et al. (2002) and Vander and Dickinson (2009) and were deposited in the herbarium of Universidad Nacional Toribio Rodríguez de Mendoza (KUELAP), Peru (Table 1) (Thiers, 2016). Furthermore, the records and morphologies of berries were revised and contrasted from databases and collections such as the Global Biodiversity Information Facility (https://www.gbif.org/), Tropicos from Missouri Botanical Garden (http://www.tropicos.org), the New York Botanical Garden Steere herbarium (http://sweetgum.nybg.org/science), and JSTOR Global Plants (https://plants.jstor.org).

Figure 1.

Map showing the sampling of Andean berry specimens from Region Amazonas, northern Peru.

Table 1.

List of samples of Andean berries collected in northern Peru including genomic DNA QC using fluorometer.

Species	Code	Herbarium Voucher	Place	Date	Elevation (m.a.s.l)	Latitude (South)	Longitude (West)	DNA concen. (ng/μL)
Cavendishia punctata	IARAN006	KUELAP–267	La Palma	5/07/2017	2934	6°43′26.36″	77°50′42.30″	50.20
Cavendishia punctata	IARAN018	KUELAP–279	Olmal	13/07/2017	2506	6°10′54.69″	77°47′07.72″	56.10
Cavendishia punctata	IARAN032	KUELAP–293	Opelele	5/08/2017	2572	6°15′18.68″	77°48′05.35″	86.30
Cavendishia punctata	IARAN050	KUELAP–311	Santa Rosa	14/08/2017	2794	6°18′06.26″	77°53′51.48″	80.70
Cavendishia punctata	IARAN046	KUELAP–307	Maino	14/08/2017	2598	6°19′35.92″	77°52′36.40″	81.02
Clethra ovalifolia	IARAN021	KUELAP–282	Sonche	23/07/2017	2507	6°10′40.69″	77°47′09.79″	43.00
Clethra retivenia	IARAN034	KUELAP–295	Chachapoyas	5/08/2017	2627	6°15′35.49″	77°47′59.39″	25.00
Disterigma ecuadorense	IARAN024	KUELAP–285	La Jalca	28/07/2017	2851	6°32′13.45″	77°47′42.39″	93.30
Disterigma synanthum	IARAN014	KUELAP–275	Espadilla	13/07/2017	2536	6°13′16.64″	77°40′51.62″	21.30
Disterigma synanthum	IARAN003	KUELAP–264	La Palma	5/07/2017	2887	6°43′28.98″	77°50′45.47″	12.00
Gaultheria secunda	IARAN017	KUELAP–278	Espadilla	13/07/2017	2542	6°13′16.92″	77°40′52.56″	21.00
Gaultheria secunda	IARAN005	KUELAP–266	La Palma	5/07/2017	2912	6°43′28.99″	77°50′43.98″	8.70
Gaultheria secunda	IARAN023	KUELAP–284	Olmal	23/07/2017	2496	6°10′56.73″	77°47′08.38″	46.56
Gaultheria secunda	IARAN027	KUELAP–288	La Jalca	28/07/2017	2837	6°29′23.55″	77°48′47.56″	42.05
Gaultheria secunda	IARAN040	KUELAP–301	Opelele	5/08/2017	2827	6°15′30.71″	77°48′24.23″	94.72
Gaultheria sp. 1	IARAN041	KUELAP–302	Levanto	14/08/2017	2720	6°18′09.99″	77°53′51.59″	56.08
Gaultheria sp. 2	IARAN047	KUELAP–308	Levanto	14/08/2017	2770	6°18′11.66″	77°53′51.18″	79.00
Gaultheria sp. 3	IARAN028	KUELAP–289	La Jalca	28/07/2017	2,700	6°29′12.74″	77°49′19.11″	17.00
Thibaudia angustifolia	IARAN022	KUELAP–283	Olmal	23/07/2017	2509	6°10′44.94″	77°47′11.12″	95.01
Thibaudia moricandi	IARAN037	KUELAP–298	Opelele	5/08/2017	2625	6°16′04.98″	77°46′53.43″	76.30
Thibaudia obovata	IARAN011	KUELAP–272	Espadilla	13/07/2017	2502	6°13′11.16″	77°40′48.18″	19.03
Thibaudia ovalifolia	IARAN038	KUELAP–299	Tañapampa	5/08/2017	2372	6°13′56.99″	77°51′13.92″	22.08
Vaccinium floribundum	IARAN001	KUELAP–262	La Palma	5/07/2017	2942	6°43′31.76″	77°50′42.14″	7.40
Vaccinium floribundum	IARAN004	KUELAP–265	La Palma	5/07/2017	2906	6°43′28.99″	77°50′43.94″	21.20
Vaccinium floribundum	IARAN007	KUELAP–268	La Palma	5/07/2017	3019	6°43′21.11″	77°50′33.62″	7.30
Vaccinium floribundum	IARAN012	KUELAP–273	Espadilla	13/07/2017	2515	6°13′13.72″	77°40′49.62″	23.00
Vaccinium floribundum	IARAN016	KUELAP–277	Espadilla	13/07/2017	2544	6°13′17.18″	77°40′52.96″	38.09
Vaccinium floribundum	IARAN020	KUELAP–281	Sonche	23/07/2017	2494	6°10′47.96″	77°47′07.36″	49.86
Vaccinium floribundum	IARAN026	KUELAP–287	Leymebamba	28/07/2017	2857	6°43′03.77″	77°47′44.91″	12.43
Vaccinium floribundum	IARAN029	KUELAP–290	Huancaurco	2/08/2017	2680	6°07′59.65″	77°52′34.45″	86.09
Vaccinium floribundum	IARAN031	KUELAP–292	Huancaurco	2/08/2017	2699	6°08′10.84″	77°52′27.79″	32.00
Vaccinium floribundum	IARAN036	KUELAP–297	Opelele	5/08/2017	2626	6°15′24.92″	77°47′58.95″	43.32
Vaccinium floribundum	IARAN051	KUELAP–312	Santa Rosa	14/08/2017	2597	6°19′38.43″	77°52′36.31″	14.00
Vaccinium mathewsii	IARAN002	KUELAP–263	La Palma	5/07/2017	2906	6°43′31.46″	77°50′43.66″	64.30
Vaccinium mathewsii	IARAN025	KUELAP–286	Leymebamba	28/07/2017	2855	6°43′03.79″	77°47′45.14″	23.10
Vaccinium mathewsii	IARAN039	KUELAP–300	Opelele	5/08/2017	2827	6°15′30.87″	77°48′24.53″	65.30
Vaccinium mathewsii	IARAN030	KUELAP–291	Huancaurco	2/08/2017	2727	6°07′58.84″	77°52′39.03″	32.20
Vaccinium meridionale	IARAN009	KUELAP–270	Espadilla	13/07/2017	2399	6°13′04.40″	77°40′19.82″	24.90
Vaccinium meridionale	IARAN013	KUELAP–274	Espadilla	13/07/2017	2519	6°13′14.60″	77°40′49.79″	78.00
Vaccinium meridionale	IARAN019	KUELAP–280	Sonche	23/07/2017	2514	6°10′48.38″	77°47′07.56″	56.00
Rubus adenothallus	IR003	KUELAP–256	Granada	4/04/2019	2822	6°06′07.01″	77°38′28.79″	129.20
Rubus andicola	IR001	KUELAP–254	Izcuchaca	4/04/2019	2188	6°20′15.31″	77°31′06.41″	111.00
Rubus floribundus	IR002	KUELAP–255	Izcuchaca	4/04/2019	2156	6°20′15.30″	77°31′06.40″	130.00
Rubus glabratus	IR008	KUELAP–261	Calla Calla	4/04/2019	2887	6°43′19.29″	77°50′45.41″	143.60
Rubus lechleri	IR005	KUELAP–258	Granada	4/04/2019	2923	6°07′34.69″	77°38′59.81″	110.20
Rubus loxensis	IR006	KUELAP–259	Granada	4/04/2019	2949	6°07′34.92″	77°38′59.97″	160.00
Rubus sparsiflorus	IR007	KUELAP–260	Granada	4/04/2019	3068	6°07′56.21″	77°38′59.97″	132.10
Rubus weberbaueri	IR004	KUELAP–257	Molinopampa	4/04/2019	3251	6°08′59.48″	77°40′16.09″	120.00

2.2. DNA sequencing and alignment preparation

Genomic DNA was extracted from leaf tissue using the NucleoSpin Plant II Kit (Macherey-Nagel, Düren, Germany) following the Tineo et al. (2020). Briefly, samples were homogenized in a freeze-crush apparatus (SK-100, Funakoshi, Japan). 550 μl of lysis buffer was added and incubated at 65 °C overnight and then centrifuged at 11000 rpm for 60s. Then, 480 μl of binding buffer was added and centrifuged at 11000 rpm for 60s. Then, two washing steps of 600 μl of washing buffer was performed and centrifuged at 13000 rpm for 60s. Finally, 50 μl of elution buffer was added and centrifuged at 11000 rpm for 60s. DNA concentration was quantified by a Quantus™ Fluorometer (Promega, Madison, USA) (Table 1), and quality was measured by 1% agarose gel electrophoresis and visualized on a photodocumenter (SmartView Pro UVCI-1000, Major Science, Saratoga, USA) (Figure 2). Two chloroplast markers (matK and rbcL) and two nuclear markers (nrITS and GBSSI-2) were sequenced. Each gene was amplified using polymerase chain reaction (PCR) with MasterMix (Promega, Wisconsin, USA) in the following reaction mixture: 10 ng of DNA and 0.25–0.5 pmol of forward and reverse primers for a total volume of 10 μl. The PCR protocols followed Bustamante et al. (2021) and Tineo et al. (2020), and primer combinations are summarized in Table 2. Amplicons were purified using the NucleoSpin™ Gel and PCR Clean-up Kit protocol (Macherey-Nagel™, Düren, Germany). The sequences of the forward and reverse strands were determined commercially by Macrogen Inc. (Macrogen, Seoul, Korea). The sequences were manually edited with Chromas V.2.6.6 software. The 125 newly generated sequences (DNA-barcodes) from the four markers (matK, rbcL, nrITS and GBSSI-2) were deposited in GenBank. These sequences and others obtained from GenBank (Table 3) were initially aligned with Muscle algorithms (Thompson et al., 1994) and were adjusted manually with MEGA10 software (Kumar et al., 2018) (Figure 3).

Figure 2.

Genomic DNA QC using standard Gel Electrophoresis for Andean berries specimens from Region Amazonas, northern Peru.

Table 2.

Sets of primer combinations for matK, rbcL, nrITS and GBSSI-2 markers used for specimens from Ericales and Rosales (listed 5′→ 3′).

Gene or spacer region	Amplified length (bp)	Primers sequence (5′–3′)	References
GBSSI–2	550	F: 5′–TGGTCTTGGTGATGTTCTTGG–3′	Rousseau–Gueutin et al., 2009
		R: 5′– GTGTAGTTGGTTGTCCTTGTAATCC–3′	Rousseau–Gueutin et al., 2009
ITS	650	F: 5′–GGAAGTAAAAGTCGTAACAAGG–3′	White et al., 1990
		R: 5′–TCCTCCGCTATATGATATGC–3′	White et al., 1990
rbcL	1600	F: 5′–ATGTCACCACAAACAGAAACTAAAGC–3′	Chase et al. (2016)
		R: 5′– CTTTTAGTAAAAGATTGGGCCGAG–3′	Chase et al. (2016)
matK	1500	F: 5′–CTATATCCACTTATCTTTCAGGAGT–3′	Ooi et al. (1995)
		R: 5′–AAAGTTCTAGCACAAGAAAGTCGA–3′	Ooi et al. (1995)

Table 3.

List of taxa used in molecular analyses along with voucher numbers followed by GenBank accession numbers. Sequences generated in the present study are in bold.

Species	Voucher/N° Taxon	ITS	matK	rbcL
Cavendishia angustifolia	Pedraza 1749	KJ788223	KJ788254	–
Cavendishia arizonensis	Luteyn 15286	–	KP729914	–
Cavendishia bomareoides	Pedraza 1752	KJ788224	KJ788255	–
Cavendishia bracteata	Luteyn 14223	AY331867	AY331894	–
Cavendishia callista	Clarke 5241	–	KP729912	MF786429
Cavendishia capitulata	Powell 10	AY331868	AY331895	–
Cavendishia complectens	Pedraza 1749	KM209386	–	–
Cavendishia grandifolia	NY/L. 8023	AY331869	AY331896	–
Cavendishia isernii	Salinas 707	KP729959	–	–
Cavendishia leucantha	Pedraza 1768	KJ788226	–	–
Cavendishia lindauiana	Pedraza 1766	KJ788227	KJ788258	–
Cavendishia mariae	Luteyn 15198	KP729960	KP729913	–
Cavendishia martii	Luteyn 15443	AF382658	AF382747	–
Cavendishia micayensis	Pedraza 1888	KJ788228	AF382748	–
Cavendishia nobilis	Lewis 3414	KP729961	KP729916	–
Cavendishia pilosa	Pedraza 1743	KJ788229	KJ788260	–
Cavendishia pubescens	Pedraza 1038	KJ788230	KJ788261	–
Cavendishia punctata	KUELAP–267	OL361763	OL706727	OL707640
Cavendishia punctata	KUELAP–311	OL361767	OL706731	OL707644
Cavendishia punctata	KUELAP–307	OL361766	OL706730	OL707643
Cavendishia punctata	KUELAP–293	OL361765	OL706729	OL707642
Cavendishia punctata	KUELAP–279	OL361764	OL706728	OL707641
Cavendishia quereme	Pedraza 1707	KJ788231	KJ788262
Cavendishia tarapotana	Pedraza 1958	KJ788232	KP729915	–
Cavendishia zamorensis	Salina 721	KP729966	KP729917	–
Cavendishia litensis		AY331890	–	–
Thibaudia floribunda		AF382709	–	–
Thibaudia parvifolia		AF382713	–	–
Clethra acuminata	Leonard et al., 1849	AY190572	–	JQ594906
Clethra alnifolia	CCDB–20334–D03	AY190571	MF350258	MG224565
Clethra alnifolia	CCDB–20334–C04	MG220127	AJ429281	MG222185
Clethra arborea	Hedenas & Bisang s.	AY190569	–	–
Clethra arfakana	Sleumer & Vink 4380	AY190568	–	–
Clethra barbinervis	Anderberg & Lundin 11	AY190573	AB697681	AF421089
Clethra canescens	224281	AY190564	–	–
Clethra castaneifolia	S.V&D. 9109	AY190567	–	–
Clethra cubensis	Rova et al., 2248	AY190560	–	–
Clethra delavayi	Aldén et al., 1717	AY190570	–	–
Clethra fimbriata	Harling 27133	AY190563	–	–
Clethra hartwegii	H.S.Gaultheria 2135	AY190574	–	–
Clethra mexicana	C&V 1831	AY190558	–	JQ591083
Clethra ovalifolia	H&A.21905	AY190561	–	–
Clethra ovalifolia	KUELAP–282	OL361761	OL706732	OL707645
Clethra pachyphylla	Emanuelsson 261	AY190565	–	–
Clethra peruviana	S.V et al., 10006	AY190566	–	–
Clethra retivenia	KUELAP–295	OL361762	OL706733	OL707646
Clethra revoluta	Persson 515	AY190562	–
Clethra scabra	Oliveira 297	AY190559	–	MG833484
Clethra vicentina	W&M 23234	AY190557	–	–
Ternstroemia sp.	–	–	HQ437950
Franklinia alatamaha	–	–	AF380082	MF349693
Diospyros aculeata	–	–	MG201641
Disterigma acuminatum	PP1098	FJ001669	–	–
Disterigma agathosmoides	L15190	–	KC175470	–
Disterigma alaternoides	L15074	FJ001672	AY331901	–
Disterigma appendiculatum	PP1113	FJ001673	–	–
Disterigma balslevii	PP998	FJ001674	–	–
Disterigma bracteatum	PP1016	FJ001675	–	–
Disterigma chocoanum	PP1121	FJ001696	–	–
Disterigma codonanthum	L15117	FJ001677	–	–
Disterigma cryptocalyx	L14993	FJ001678	–	–
Disterigma dumontii	L15177	FJ001679	–	–
Disterigma ecuadorense	KUELAP–285	OL361760	OL706736	OL707648
Disterigma empetrifolium	CP7	FJ001680	–	–
Disterigma hiatum	PP1112	FJ001681	–	–
Disterigma humboldtii	P1075	FJ001684	–	–
Disterigma luteynii	LPP14797	FJ001687	–	–
Disterigma micranthum	PP1229	FJ001688	–	–
Disterigma noyesiae	PP1155	FJ001690	–	–
Disterigma ollacehum	PP1528	FJ001697	–	–
Disterigma ovatum	LPP15457	FJ001692	AY331902	–
Disterigma pallidum	PP1506	AF382674	–	–
Disterigma parallelinerve	JB12532	KC175459	–	–
Disterigma pentandrum	L15085	FJ001693	KC175465	–
Disterigma pernettyoides	L15441	–	AF382762	–
Disterigma pseudokillipiella	PP1143	FJ001694	KC175471	–
Disterigma rimbachii	PP1018	FJ001695	AY331903	–
Disterigma staphelioides	PP1062	FJ001698	–	–
Disterigma stereophyllum	L15206	FJ001699	–	–
Disterigma synanthum	KUELAP–264	–	OL706734	OL707647
Disterigma synanthum	KUELAP–275	OL361759	OL706735
Disterigma trimerum	L15568	FJ001700	–
Disterigma ulei	PP1515	FJ001701	–	–
Disterigma verruculatum	PP1138	FJ001703	–	–
Notopora schomburgkii		AF382683	AF382768	–
Orthaea venamensis		AF382687	AF382772	–
Orthaea apophysata		AF382685	–	–
Gaultheria acuminata	1091527	JF801586	JF801333	–
Gaultheria adenothrix	586107	FJ010595	–	–
Gaultheria antipoda	672075	JF801617	JF801372	KT626709
Gaultheria borneensis	VacciniumK.2101092, ACAD	JF801598	AF366629	JF941568
Gaultheria bracteata	1091528	JF801593	JF801341	–
Gaultheria buxifolia	1091526	–	JF801359	–
Gaultheria cardiosepala	LuLu–06–0022–1	JF976341	HM597394	JF941573
Gaultheria corvensis	1091531	JF801614	–
Gaultheria cumingiana	VacciniumK.3101092, ACAD	AF358882	–	–
Gaultheria cuneata	S.D. Z&L. Lu 031543	HM597250	–	–
Gaultheria discolor	GLGS32542	–	HM597366	JN098404
Gaultheria dolichopoda	L. Lu et al., 060005	HM597318	HM597405	–
Gaultheria domingensis	679020	JF801594	JF801342	–
Gaultheria dumicola	LuLu–GLGS20245	–	HM597346	JF941588
Gaultheria eciliata	LuLu–LL–07149–1	–	HM597421	–
Gaultheria erecta	L.13813, NY	JF801585	AF366631	–
Gaultheria eriophylla	763043, RBGE	–	U61317	L12618
Gaultheria foliolosa	L.15075, NY	JF801610	–
Gaultheria glomerata	L.15327, NY	JF801592	AF366633	–
Gaultheria gracilis	1091532	JF801587	JF801335	–
Gaultheria hapalotricha	1091533	JF801596	–	–
Gaultheria heteromera	L. Lu et al., 07316A	–	HM597358	–
Gaultheria hispidula	VacciniumK.s.n., ACAD	JF801562	AF366634	MG223840
Gaultheria hookeri	S.D. Z&W.B. Yu 009	–	HM597364	–
Gaultheria humifusa	FF132	FJ665708	JF801346	KX678317
Gaultheria hypochlora	LuLu–GLGS16817–1	JF976381	HM597410	JF941640
Gaultheria insana	672082	JF801604	JF801354
Gaultheria lanigera	L.15062, NY	JF801590	–	–
Gaultheria leucocarpa	VacciniumK. 318896, ACAD	JF976385	JF801306	–
Gaultheria macrostigma	176244	FJ665711	JF801369	–
Gaultheria megalodonta	157515	AF358890	AF366639
Gaultheria miqueliana	1636–77, AA	AF358891	–	AF124590
Gaultheria mucronata	586115	FJ010604	FJ010622	–
Gaultheria myrsinoides	L.14814, NY	AF358892	AF366640
Gaultheria notabilis	L. Lu et al., 07005	–	HM597370	–
Gaultheria nubigena	672084	JF801600	JF801350	–
Gaultheria ovatifolia	CCDB–23363–F06	JF801597	–	MG222845
Gaultheria parvula	672087	FJ665715	JF801371	–
Gaultheria praticola	861407	–	JF801383	–
Gaultheria procumbens	Powell s.n., WFU	–	AF366643	MG222887
Gaultheria prostrata	S.D.Z&W.B.Yu ZY011	JF801603	JF801348	JN098405
Gaultheria pseudonotabilis	GLGS 16565	–	HM597382	–
Gaultheria pyroloides	95633	HM597252	JF801349	–
Gaultheria reticulata	L.15077, NY	AF358897	AF366645	–
Gaultheria schultesii	586118	FJ010601	–	–
Gaultheria sclerophylla	L.5331, NY	AF358898	AF366646	–
Gaultheria secunda	KUELAP–278	–	OL706744	OL707650
Gaultheria secunda	KUELAP–284	OL361752	OL706738	OL707651
Gaultheria secunda	KUELAP–301	OL361754	OL706741	OL707654
Gaultheria secunda	KUELAP–288	OL361753	OL706739	OL707652
Gaultheria secunda	KUELAP–266	OL361751	OL706737	OL707649
Gaultheria semi–infera	L. Lu et al., 07312	–	HM597388	–
Gaultheria serrata	1091539	JF801595	JF801343	–
Gaultheria shallon	DNA 185, WFU	JF801581	JF801329	MG221678
Gaultheria sleumeriana	1091540	JF801613	–	–
Gaultheria straminea	L. Lu et al., 07306	–	HM597390	–
Gaultheria strigosa	L.15358, NY	JF801608	AF366647	–
Gaultheria suborbicularis	1045346	JF801563	–	–
Gaultheria tasmanica	1977–5050, RBGK	AF358901	JF801370	–
Gaultheria thymifolia	586120	–	HM597396	–
Gaultheria tomentosa	L.15076, NY	AF358902	AF366648	–
Gaultheria trichophylla	LuLu–ZY–013–1	–	HM597416	JF941727
Gaultheria vaccinioides	1091543	–	JF801331	–
Gaultheria viridicarpa	1842756	–	–	KU564802
Gaultheria sp. 01	KUELAP–302	OL361755	OL706742	OL707655
Gaultheria sp. 02	KUELAP–308	–	OL706743	OL707656
Gaultheria sp. 03	KUELAP–289	–	OL706740	OL707653
Leucothoe griffithiana		FJ010598	FJ010616	–
Leucothoe tonkinensis		MH558159	–	–
Leucothoe davisiae		JF801553	FJ010617	–
Thibaudia ovalifolia	KUELAP–299	OL361758	Yes	OL707660
Thibaudia moricandi	KUELAP–298	–	Yes	OL707659
Thibaudia obovata	KUELAP–272	OL361756	–	OL707657
Thibaudia costaricensis	WFU/EAP016	AY331887	AY331914	–
Thibaudia densiflora	MM001	–	AF382790	–
Thibaudia diphylla	NY/L15459	AY331888	AY331915	–
Thibaudia floribunda	NY/L15090	AF382709	–	–
Thibaudia inflata	NY/L15029	–	AY331916
Thibaudia jahnii	180744	–	AF382792	–
Thibaudia litensis	NY/L15020	AY331890	–	–
Thibaudia macrocalyx	NY/L15444	AF382711	AF382793	–
Thibaudia martiniana	NY/L15028	AY331891	AY331918	–
Thibaudia angustifolia	KUELAP–283	OL361757	OL706745	OL707658
Thibaudia pachyantha	NY/L15189	AF382712	–	–
Thibaudia parvifolia	NY/L5212	AF382713	–	–
Thibaudia tomentosa	NY/L15502	AY331892	AY331919	–
Vaccinium poasanum	–	AF382736	–	JQ594910
Disterigma trimerum	–	FJ001700	–	–
Vaccinium alvarezii	KIG, HGG, P–659–L	KM209414	–	–
Vaccinium amamianum	TI:Ohi–Toma s.n	–	LC168877	–
Vaccinium ambivalens	C.Koster BW 13699–L	KM209415	–	–
Vaccinium andersonii	RG–9104–L	KM209418	–	–
Vaccinium arboreum	FLAS:M–4609	KM209419	–	KY626810
Vaccinium arctostaphylos	ACAD/VK–23991	AF419774	AF419702	–
Vaccinium berberidifolium	FRF–51757–L	KM209424	–	–
Vaccinium boninense	1004256	–	AB623168	–
Vaccinium bulleyanum	1633929	–	LC168878	–
Vaccinium caespitosum	ACAD/VK–313887	AF419775	AF419703	KX678256
Vaccinium calycinum	ACAD/VacciniumK–630886	AF419776	AF419704	–
Vaccinium caudatifolium	RBGE 1993–4020	AF382715	AF382797	–
Vaccinium cercidifolium	RBGE 1982–0845	AF382716	–	–
Vaccinium cereum	ACAD/VacciniumK–316992	KM209431	AF419705	–
Vaccinium ciliatum	445570	AB623188	–	–
Vaccinium corymbosum	ACAD/VacciniumK–ABS7	AF419778	AF419706	MG223027
Vaccinium crassifolium	WFU/K&P–DNA208	AF382718	–	–
Vaccinium crenatum	NY/L14171	AF382719	–	–
Vaccinium cruentum	1633933	KM209436	–	–
Vaccinium cylindraceum	180753	AF382720	AF382800_	–
Vaccinium deliciosum	ACAD/VK–529879	AF419790	AF419707	KX678227
Vaccinium dentatum	RBGE 1011085	AF382721	AF382801	–
Vaccinium emarginatum	174252	–	AB623166	–
Vaccinium erythrocarpum	ACAD/VK–81981	AF419779	AF419710	–
Vaccinium exul	1633938	–	–	KU568131
Vaccinium filiforme	RBGE 1980–1411	AF382722	–	–
Vaccinium floribundum	180757	–	AF382804	–
Vaccinium floribundum	KUELAP–312			OL707676
Vaccinium floribundum	KUELAP–268		OL706747	OL707663
Vaccinium floribundum	KUELAP–273	–	–	OL707665
Vaccinium floribundum	KUELAP–281	OL360763	OL706752	OL707668
Vaccinium floribundum	KUELAP–265	OL360759	OL706746	OL405713
Vaccinium floribundum	KUELAP–277	OL360761	OL706750	OL707667
Vaccinium floribundum	KUELAP–287	–	OL707625	OL707670
Vaccinium floribundum	KUELAP–262	–	–	OL707661
Vaccinium floribundum	KUELAP–290	OL360765	OL706753	OL707671
Vaccinium floribundum	KUELAP–297	OL360768	OL707628	OL707674
Vaccinium floribundum	KUELAP–292	OL360767	OL707627	OL707673
Vaccinium fragile	ACAD/VacciniumK–128796	AF382725	AF382805	–
Vaccinium gaultheriifolium	RBGE 1992–0332	AF382726	LC168880	–
Vaccinium hirsutum	ACAD/VacciniumK–83981	AF419780	AF419709	–
Vaccinium hirtum	RBGE 1921–9886	AB623185	AB623169	–
Vaccinium horizontale	180761	–	AF382808	–
Vaccinium latissimum	1633959	KM209449	–	–
Vaccinium leucobotrys	1633944	KM209451	–	–
Vaccinium macrocarpon	13750	AF382730	U61316	MG221913
Vaccinium madagascariense	LB–11063–L	KM209442	–
Vaccinium mathewsii	KUELAP–286	OL360764	OL707624	OL707669
Vaccinium mathewsii	KUELAP–300	OL360769	OL707629	OL707675
Vaccinium mathewsii	KUELAP–263	–	–	OL707662
Vaccinium mathewsii	KUELAP–291	OL360766	OL707626	OL707672
Vaccinium membranaceum	ACAD/VK–133979	AF419782	AF419711	MH926046
Vaccinium meridionale	ACAD/VacciniumK–s.n.	AF382731	–	AF124576
Vaccinium meridionale	KUELAP–274	OL360760	OL706749	OL707666
Vaccinium meridionale	KUELAP–270		OL706748	OL707664
Vaccinium meridionale	KUELAP–280	OL360762	OL706751
Vaccinium moupinense	S.M, A.F. 079	KM209457	–	–
Vaccinium myrtillus	S/Anderberg s.n.	AF382732	AF382810	MG221208
Vaccinium nummularia	180764	–	LC168882	–
Vaccinium oldhamii	ACAD/VacciniumK–426886	AF419783	AB623174	–
Vaccinium ovalifolium	ACAD/VK–1419886	AF419784	–	KX679055
Vaccinium ovatum	ERM1383	FJ001692	–	KX678497
Vaccinium oxycoccos	HERB0230	–	LC168883	KX677905
Vaccinium padifolium	ACAD/VK–5141090	AF382734	AF382812	–
Vaccinium phillyreoides	989263–L	KM209465	–	–
Vaccinium praestans	ACAD/VK–Vacc813	AF419785	AF419714	–
Vaccinium pratense	SCBGP385_2	KP092616	–	–
Vaccinium reticulatum	ACAD/VacciniumK–324992	AF382737	AF382814	–
Vaccinium scoparium	ACAD/VacciniumK–731883	AF419787	AF419716	MG222739
Vaccinium sieboldii	TNS:175ws–20100513	AB623191	AB623175	–
Vaccinium smallii	ACAD/VK–725886	AF382739	AB623170	–
Vaccinium summifaucis	RBGE 1963–0610	AF382740	AF382817
Vaccinium tenellum	WFU/K& P–DNA209	AF382741	AF382818	–
Vaccinium uliginosum	ACAD/VK–217995	AF419788	AF419717	KX677950
Vaccinium varingifolium	229200	AY274564	–	–
Vaccinium vitis idaea	RBGE 1977–3274A	AH011361	MN150141	MG222697
Vaccinium wrightii	1004259	AB623192	–	–
Vaccinium yakushimense	1004255	AB623183	–	–
Vaccinium yatabei	ACAD/VK–419886	AF419789	AF419718	–
Z. pulverulenta (Outgroup)	AF358906	AF124571	–
A. polifolia (Outgroup)		AF358872	LC168873	–
Species	Voucher/N° Taxon	ITS	GBSSI	rbcL
Rubus acuminatus	R2007	–	–	KU881197
Rubus adenothallus	KUELAP–256	OL348471	OL707633	OL707678
Rubus amabilis	R01–14–SICUA	FJ472909	KU926726	KU881200
Rubus andicola	KUELAP–254	OL348470	OL707636	OL707677
Rubus assamensis	R0118	AH006024	KU926729	KU881203
Rubus australis	Gardner 1539, MO	H006022	–	–
Rubus biflorus	R2504	KU881063	KU926733	KU881207
Rubus bifrons	Alice 98–9, M	AF055775	–	–
Rubus bollei	Bol_col14	KM037227	–	–
Rubus caesius	Karlen 243, S/75065	AF055776	–	FN689382
Rubus calycinus	R2519	KU881065	KU926735	KU881209
Rubus canadensis	A&C 98–10, M/MOBOT.S.27940	AF055777	–	KY427303
Rubus caudifolius	R2021	KU881067	KU926737	KU881211
Rubus chamaemorus	Alice, R17, M	AF055740	–	–
Rubus chingii	R2128	KU881068	–	KU881212
Rubus corchorifolius	PDBK 2008–0160	–	–	MH593651
Rubus coreanus	321593	MT078683	KU926741	MN732644
Rubus cuneifolius	Alice 5, M/A.22485	AF055778	–	KJ773846
Rubus deliciosus	Alice, 98–1, M	AF055733	–	–
Rubus ellipticus	R2512/R0112	KU881060	KU926746	KU881223
Rubus eucalyptus	R2354	–	KU926749	KU881226
Rubus eustephanus	R2518	KU881083	KU926750	KU881227
Rubus loxensis	KUELAP–259	OL348474	OL707635	OL707679
Rubus flagellaris	Alice 61:WKU/BM 2008/273	AY083372	–	HM850313
Rubus foliosus	Fol_col08	KM037335	–	–
Rubus floribundus	KUELAP–255	OL351854	OL707632	–
Rubus geoides	Dudley et al., 1538a, MO	AF055799	–	–
Rubus glabratus	Rubus5132 QCA	HM453950	–	–
Rubus glabratus	KUELAP–261	OL348476	OL707639	OL707681
Rubus glaucus	PI 548906	AY083361	–	–
Rubus gracilis	Grac_co0l5	KM037377	–	–
Rubus gunnianus	Wells 96–1, M	AF055749	–	–
Rubus hirsutus	ODdo/R2225	AY818208	KU926758	KU881236
Rubus hypomalacus	Hma_col07	KM037395	–	–
Rubus hypopitys	R2533	KU881094	–	KU881238
Rubus idaeus	Alice, R8, MAINE	AF055755	–	JX848533
Rubus lasiococcus	Merello et al., 827, MO	AF055750	–	–
Rubus lechleri	KUELAP–258	OL348473	OL707637	OL707682
Rubus macilentus	R2501	KU881118	KU926783	KU881262
Rubus macraei	59494	AF055763	–	–
Rubus matsumuranus	PDBK 2012–0085	–	–	MH593654
R, moorrei	Streimann 8207, GH	AF055765	–	–
Rubus mesogaeus	ALTA:120202	KU881122	KU926787	KU881266
Rubus moschus	Mos01_col04	KM037437	–	–
Rubus niveus	R0101	KU881126	KU926791	KU881270
Rubus nubigenus	1257, NCGR	AF055769	–	–
Rubus odoratus	Alice, R14, M	AF055734	–	–
Rubus parviflorus	Richards, 666, M	AF055735	–	–
Rubus parvifolius	R2035	KU881132	KU926797	GU363802
Rubus parvus	Alice 97–3, M/CHR:688824	AF055766	–	KT626843
Rubus pectinellus	680–MO	AF055797	–	–
Rubus pedemontanus	Martensen s.n.	AF055783	–	–
Rubus pentagonus	R0223	–	KU926801	KU881280
Rubus phoenicolasius	Alice, 96–2, M	AF055759	KU926803	–
Rubus pinfaensis	R0102	–	KU926811	KU881291
Rubus platyphyllus	Sva_col04	KM037581	–	–
Rubus praecox	Pra01_col01	KM037481	–	–
Rubus pungens	R2337	KU881153	KU926818	KU881297
Rubus radula	Rad_col06	KM037522	–	–
Rubus reflexus	S.1033I/S.0492L	JN407524	–	JN407362
Rubus robustus	Steinbach 247, GH	AF055771	–	–
Rubus roseus	L&14402, M	AF055770	–	–
Rubus sanctus	Thibaudia Eriksson 714, S	AF055785	–	–
Rubus saxatilis	Thibaudia Eriksson 719, S	AF055746	–	–
Rubus schizostylus	TKM201536	KT634247	–	–
Rubus schleicheri	Schl_col05	KM037537	–	–
Rubus scissoides	1546356	KM037543	–	–
Rubus setosus	Alice 113, MAINE	AF055787	–	–
Rubus silvaticus	Sil_col08	KM037557	–	–
Rubus simplex	R2321	–	KU926832	KU881312
Rubus sparsiflorus	KUELAP–260	OL348475	OL707638	OL707680
Rubus sulcatus	Martensen 1325.12	AF055789	–	–
Rubus sumatranus	R2111	KU881182	KU926845	KU881326
Rubus tephrodes	Yao, 9231, MO	AF055767	–	–
Rubus thibetanus	Q186	MH711174	–	–
Rubus trifidus	C, 3.001/A, 98–2, M	AF055737	–	–
Rubus trilobus	Ruiz, 889, MO	AF055738	–	–
Rubus trivialis	Alice 55, M/Abbott 26055	AF055790	–	KJ773847
Rubus ursinus	197, NCGR/Alice 98–8, M	AF055794	–	–
Rubus vigorosus	Martensen 2518.32	AF055793	–	–
Rubus weberbaueri	KUELAP–257	OL348472	OL707634	OL707683
Fullgaria paradoxa		U90805	AM116869	U06802
Waldsteinia fragarioides			–	U90822
Geum urbanum			AM116871	U90802

Figure 3.

Experimental procedures for sampling, identification, DNA extraction, amplification, purification and data analysis for Andean berries specimens from Region Amazonas, northern Peru.

2.3. Phylogenetic analysis of concatenated sequence data

The phylogenies were based on concatenated data of the four molecular markers (Table 2). An exploratory phylogeny consisting of Ericales and Rosales (340 sequences) was performed to identify the main lineages where Andean berries were embedded. Additionally, separate phylogenies for each lineage were evaluated. Selection of the best-fitting nucleotide substitution model was conducted using PartitionFinder (Lanfear et al., 2012) for exploratory analysis (using the four partitions matK, rbcL, nrITS and GBSSI-2) and for separate phylogenies (using three partitions each) (Table 4). The best partition strategy and model of sequence evolution were selected based on the Bayesian information criterion (BIC) for each phylogeny (Table 4). Maximum likelihood (ML) analyses were conducted with the RAxML HPC-AVX program (Stamatakis, 2014), implemented in the raxmlGUI 1.3.1 interface (Silvestro and Michalak, 2012) using Table 4 models with 1000 bootstrap replications. Bayesian inference (BI) was performed with MrBayes v. 3.2.6 software (Ronquist et al., 2012) using Metropolis-coupled MCMC and the Table 4 models. Two runs each with four chains (three hot and one cold) were conducted for 10,000,000 generations, sampling trees every 1,000 generations.

Table 4.

Evolutionary models for phylogenetic analyses of specimens from Ericales and Rosales.

	Group	Bayesian inferences	Maximum likelihood
Exploratory phylogeny	Figure S1	GTR + I+G	GTRGAMMAI
Separate phylogenies	Figure 2	GTR	K81
	Figure 3	TRNEF+G	TRNEF+G
	Figure 4	GTR	K81UF+I+G
	Figure 5	GTR+I+G	GTR+I+G
	Figure 6	GTR+I+G	GTR+I+G
	Figure 7	GTR	K81UF+G
	Figure 8	GTR+I+G	GTR+I+G

3. Results

A total of 125 DNA-barcodes of andean berries were newly generated for the four molecular markers that allowed the construction of multilocus phylogenies. In the exploratory phylogeny, the analyzed data matrix included a total of 3,324 base pairs (bp) (1,487 bp for matK, 666 bp for rbcL, 716 bp for ITS, and 455 bp for GBSSI) from 340 individuals (Table 3). This multilocus phylogeny obtained from the ML and BI analyses molecularly confirmed 24 species from 48 specimens embedded in the order Ericales and Rosales. This exploratory phylogenetic tree showed six monophyly lineages belonging to Ericales [Cavendishia Lindl., Clethra L., Disterigma (Klotzsch) Nied, Gaultheria L., Thibaudia Ruiz & Pav., and Vaccinium L.] and one belonging to Rosales (Rubus L.) (Figure S1).

3.1. Cavendishia

The phylogeny of Cavendishia included concatenated data (1,265 bp for matK, 551 bp for rbcL, and 631 bp for ITS) from 25 individuals. The specimens KUELAP-211, KUELAP-267, KUELAP-279, KUELAP-293, and KUELAP-307 were recognized as Ca. punctata (Ruiz & Pav. ex J.St.-Hil.) Sleumer. This species is characterized by pink peduncles, dark-red pedicels, pinkish-red calyx, and pale green flowers (Figure 9A, Table 5). This species was placed in sistership with Ca. bracteata (Ruiz & Pav. ex A.St.-Hil.) Hoerold. The genetic divergences between these species were over 0.9% for matK and 0.4% for ITS (Figures 4, S2, S3). The intraspecific divergences of Ca. punctata were 0.7% for matK, 0.3% for rbcL, and 0.2% for ITS.

Figure 9.

Phylogenetic tree of the Vaccinium lineage based on maximum likelihood inference of combined matK, rbcL, and ITS data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above branches. Values lower than 50% (BS. or 0.90 (BPP. are indicated by hyphens (-)). The scale bar indicates the number of nucleotide substitutions per site.

Table 5.

Morphological comparisons among species of the genus Cavendishia.

Species	Habitat	Altitude (masl)	Height (m)	Immature fruit	Mature fruit	Flowers	Corolla	References
Cavendishia bracteata	Shrub	1400–3500	1–3	Green	Black	Lilac	Pink	Luteyn (1983)
								WCVP (2021)
Cavendishia isernii	Shrub terrestrial	660–1200	1.8–3	Reddish–green	Greenish–white	Lilac	–	Luteyn (1983)
Cavendishia punctata	Shrub	2000–3000	2–3.5	Reddish–green	Purple	Pedicel and garnet calyx	Greenish	Luteyn (1983), this study
Cavendishia sirensis	Shrub hemi–epiphyte	600–1700	1.5–3	Green	Purple	White, red calyx	Tubular	Luteyn (1983)
Cavendishia tarapotana	Shrub	1200–1500	2.5–5	–	Lilac	Fuchsia red	Rose–yellow, white–yellow	Luteyn (1983)

Figure 4.

Phylogenetic tree of the Cavendishia lineage based on maximum likelihood inference of combined matK, rbcL, and ITS data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above branches. Values lower than 50% (BS) or 0.90 (BPP) are indicated by hyphens (-). The scale bar indicates the number of nucleotide substitutions per site.

3.2. Clethra

The multilocus phylogeny of Clethra (1,323 bp for matK, 532 bp for rbcL, and 716 bp for ITS) included 21 individuals. Two species were identified among the specimens, Cl. ovalifolia Turcz (KUELAP-282) and Cl. retivenia Sleumer (KUELAP-295) (Figure 5). Cl. ovalifolia was characterized by oval leaves with stipules in the edges (Figure 9B), while Cl. retivenia was diagnosed with pubescent leaves and ferruginous back side leaves (Figure 9C, Table 6). Cl. retivenia resolved sistership to the clade composed of Cl. fimbriata Kunth, Cl. ovalifolia and Cl. revoluta (Ruiz & Pav.) Spreng., and genetic divergences were over 0.5% for ITS (Figure S4).

Figure 5.

Phylogenetic tree of the Clethra lineage based on maximum likelihood inference of combined matK, rbcL, and ITS data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above branches. Values lower than 50% (BS) or 0.90 (BPP) are indicated by hyphens (-). The scale bar indicates the number of nucleotide substitutions per site.

Table 6.

Morphological comparisons among species of the genus Clethra.

Species	Habitat	Altitude (masl)	Height (m)	Leaf shape	Immature fruit	Mature fruit	Flowers	Corolla	References
Clethra fimbriata	subshrub	2800–3600	2–3	Coriaceous	Brown	Brown	White	White	León, 2006a, León, 2006b, Sleumer (1967)
Clethra ovalifolia	Shrub	2000–3100	1–3	–	–	Brown	Cream	White	Sleumer (1967)
Clethra retivenia	Shrub	1500–3200	3	Coriaceous	–	–	White	White	León, 2006a, León, 2006b, this study
Clethra revoluta	Tree	2350	10–16	Coriaceous	–	–	White	White	Sleumer (1967)
Clethra scabra	Tree	1700–2000	4–8	–	Green	Brown– Reddish	Pink	Pink	Sleumer (1967)

3.3. Disterigma

The phylogeny of Disterigma included concatenated data (1274 bp for matK and 686 bp for ITS) from 32 individuals. Based on the multilocus tree obtained from the ML and BI analyses (Figure 6), the specimens were identified as D. synanthum Pedraza (KUELAP-264, KUELAP-275) and D. ecuadorense Luteyn (KUELAP-285). The former species was characterized by pale green floral bracts and a white corolla (Figure 9D). This species was sister to D. alaternoides (Kunth) Nied (BS/BI = 85/1.0), differing by 0.2% for the ITS. Additionally, D. ecuadorense was characterized by a green calyx, pink corolla, and white berry (Figure 9E, Table 7). This species was sister to D. ulei Sleumer, differing by 2.3% for ITS (Figures S5, S6).

Figure 6.

Phylogenetic tree of the Disterigma lineage based on maximum likelihood inference of combined matK and ITS data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above branches. Values lower than 50% (BS) or 0.90 (BPP. are indicated by hyphens (-)). The scale bar indicates the number of nucleotide substitutions per site.

Table 7.

Morphological comparisons among species of the genus Disterigma.

Species	Habitat	Altitude (masl)	Height (m)	Calyx	Immature fruit	Mature fruit	Flowers	Corolla	References
Disterigma synanthum	Shrub–epiphyte	2500–3000	0.5–1	Green pale	Green pale	Brown	White–pink	white, style white	Pedraza–Peñalosa (2008), this study
Disterigma alaternoides	Shrub terrestrial	1500–2960	1.5–2	Green	Whitish–green	Brown	White–pink	White tubular	León, 2006a, León, 2006b, Pedraza–Peñalosa (2008)
Disterigma ecuadorense	Shrub terrestrial	2500–3000	1	Green	–	White	White	Whitish–pink	Smith (1933), León, 2006a, León, 2006b, this study
Disterigma ulei	Epiphytic grass, Shrub terrestrial	2000–2800	0.5–1	Light green	Light green	Ocbonicos lilac	Greenish		Pedraza–Peñalosa (2008), Smith (1933)

3.4. Gaultheria

The multilocus phylogeny of Gaultheria (1487 bp for matK, 550 bp for rbcL and 659 bp for ITS) included 67 individuals (Figure 7). The materials comprise four species within Gaultheria. One of this species was identified as G. secunda J. Rémy (KUELAP-266, KUELAP-278, KUELAP-284, KUELAP-288, KUELAP-301) based on the red calyx and pale-white corolla (Figure 9F, Table 8). This species was resolved in sistership to the clade composed of G. foliolosa Benth and G. mucronata (L. fil.) J.Rémy. The intraspecific divergences of G. secunda were 0.2% for matK, 0.2% for rbcL, and 2.4% for ITS (Figures S7, S8, S9). The other three species remained unidentified. Gaultheria sp. 1 (KUELAP-289) (Figure 9G) resolved sister to G. myrsinoides Kunth. Additionally, Gaultheria sp. 2 (KUELAP-308) (Figure 9H) and Gaultheria sp. 3 (KUELAP-302) (Figure 9I) were sister species, and both were sister to G. glomerata (Cav.) Sleumer.

Figure 7.

Phylogenetic tree of the Gaultheria lineage based on maximum likelihood inference of combined matK, rbcL, and ITS data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above branches. Values lower than 50% (BS. or 0.90 (BPP. are indicated by hyphens (-)). The scale bar indicates the number of nucleotide substitutions per site.

Table 8.

Morphological comparisons among species of the genus Gaultheria.

Species	Habitat	Altitude (masl)	Height (m)	Calyx	Immature fruit	Mature fruit	Flowers	Corolla	References
Gaultheria foliolosa	Shrub terrestrial	2000–3000	0.5–3	Cream	-	Blue–black	White	Cream	Middleton (1990, 1991)
Gaultheria glomerata	Shrub terrestrial	1000–3000	0.5–3	Red–rose	Greenish	Black	Lilac, red	–	Middleton (1990, 1991)
Gaultheria myrsinoides	Shrub terrestrial	2000–2800	1–2	Green	Green	Purple	White	White	Middleton (1990, 1991)
Gaultheria mucronata	Shrub terrestrial	2000–3120	1–2	Green	Green	Lilac	Pink	Cream at base rose distally	Middleton (1990, 1991)
Gaultheria secunda	Shrub, half–terrestrial	2500–3500	1–2	Rose–red	Green	Red–rose	Rose–pink	Pale pinkish–white	Middleton (1990, 1991), WCVP (2021), this study
Gaultheria sp 1	Shrub	2000–2720	1–2	Rose–red	Green	Red–rose	Rose–pink	Pale pinkish–white	This study
Gaultheria sp 2	Shrub	2000–2700	1–2	Rose–red	Green	Red–rose	Rose–pink	Pale pinkish–white	This study
Gaultheria sp 3	Shrub	2000–2700	1–2	Rose–red	Green	Red–rose	Rose–pink	Pale pinkish–white	This study

3.5. Thibaudia

The phylogeny of Thibaudia included concatenated data (1,262 bp for matK, 551 bp for rbcL and 650 bp for ITS) from 18 individuals. The materials comprised four species in Thibaudia (Figure 8). T. ovalifolia A.C.Sm. (KUELAP-299) and T. moricandi Dunal (KUELAP-298) were recognized as sister species, and both differed by 0.1% for matK and 0.1% for rbcL. T. ovalifolia was characterized by glabrous flowers and rugose calyxes (Figure 9L), whereas T. moricandi was characterized by a pubescent corolla (Figure 9K, Table 9). These two species were sister to T. obovata A.C.Sm. (KUELAP-272), and both differed from the latter by over 0.1% for rbcL and 0.1% for ITS (Figures S10, S11). T. obovata was characterized by obovate-oblong leaves, pilose calyx and pedicels, and tomentose corolla (Figure 9M). Moreover, the clade composed of these three species and T. nutans Klotzsch ex Mansf. was closely related to T. angustifolia Hook (KUELAP-283). T. angustifolia was diagnosed by the presence of a bright red corolla and purple berries (Figure 9J).

Figure 8.

Phylogenetic tree of the Thibaudia lineage based on the maximum likelihood inference of combined matK, rbcL, and ITS data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above branches. Values lower than 50% (BS) or 0.90 (BPP. are indicated by hyphens (-)). The scale bar indicates the number of nucleotide substitutions per site.

Table 9.

Morphological comparisons among species of the genus Thibaudia.

Species	Habitat	Altitude (masl)	Height (m)	Calyx	Flowers	Corolla	References
Thibaudia angustifolia	Shrub	2000–2800	0.5–2	Red–rose	Bright red	Bright red	León, 2006a, León, 2006b, WCVP (2021), this study
Thibaudia diphylla	Shrub	1000–2400	2–4	Pale–pink	White pink	Rich pink	León, 2006a, León, 2006b, WCVP (2021)
Thibaudia moricandii	Shrub	1500–2600	0.5–2	Red	Red	Pubescent, throat, lobes white	León, 2006a, León, 2006b, this study
Thibaudia nutans	Herbaceous–bush	1000–1900	0–5 – 3	Green	Cauliflorous	Dark–red	León, 2006a, León, 2006b
Thibaudia ovalifolia	Shrub	2100–2700	0.5–2	Rugose	glabrous flowers	–	León, 2006a, León, 2006b, WCVP (2021), this study
Thibaudia obovata	Shrub	1600–2200	1–3	Pilose pedicels		Tomentose	León, 2006a, León, 2006b, WCVP (2021), this study
Thibaudia tomentosa	Shrub terrestrial	–	0.5–1.5	Globose bell–shaped	Bright red, curved	Orange	León, 2006a, León, 2006b

3.6. Vaccinium

The multilocus phylogeny of Vaccinium (1,253 bp for matK, 551 bp for rbcL and 673 bp for ITS) included 79 individuals. In this collection, three species were recognized in this genus, namely, V. meridionale Sw, V. mathewsii Sleumer, and V. floribundum Kunth. The former species (KUELAP-270, KUELAP-274, KUELAP-280) was characterized by dark berries and a bitter taste (Figure 9N). V. meridionale was sister to V. arboreum Marshall and genetically differed by 0.3% in rbcL and by 3.5% in the ITS. V. mathewsii (KUELAP-263, KUELAP-286, KUELAP-291, KUELAP-300) was morphologically characterized by a pinkish-white corolla and blue–black fruit (Figure 9O, Table 10). V. mathewsii was closely related to V. crenatum (G. Don) Sleumer, and genetic divergence of these taxa was 5.7% for ITS. V. floribundum (KUELAP-262, KUELAP-265, KUELAP-267, KUELAP-268, KUELAP-273, KUELAP-277, KUELAP-281, KUELAP-287, KUELAP-290, KUELAP-292, KUELAP-312) was diagnosed by having leathery leaves with pinkish-white flowers and dark berries (Figure 9P) and showed high intraspecific divergences (1.2% for matK, 0.3% for rbcL, and 3.7% for ITS), while the general appearance remained identical among all specimens of this study, suggesting cryptic diversity. V. floribundum was sister to V. ovatum Pursh, differing by 0.7% for rbcL (Figures S12, S13, S14).

Table 10.

Morphological comparisons among species of the genus Vaccinium.

Species	Habitat	Altitude (masl)	Height (m)	Immature fruit	Mature fruit	Flowers	Corolla	References
Vaccinium arboreum	Shrub	-	2–5	Green	Black	White	White	Bracko and Zurucchi (1993), León et al. (2017)
Vaccinium crenatum	Shrub terrestrial	1000–2800	–	Green	Reddish, blue–black	White–pink	Rose–red	Vander Kloet and Dickinson (2009), León et al. (2017)
Vaccinium floribundum	Shrub	2000–3000	0.5–1	Green	blue–black	White, red tips	white	Bracko and Zurucchi (1993), León et al. (2017), this study
Vaccinium mathewsii	Shrub	2000–3000	1–2	Green	blue–black	pinkish–white	Pinkish–white	Bracko and Zurucchi (1993), León et al. (2017), this study
Vaccinium meridionale	Shrub	1800–2800	0.5–2	Green–reddish	dark	White–pink	White	Vander Kloet and Dickinson (2009), this study
Vaccinium ovatum	Shrub	-	-	Green	red	White–pink	Pink	Bracko and Zurucchi (1993), León et al. (2017)

3.7. Rubus

The phylogeny of Rubus (666 bp for rbcL, 455 bp for GBSSI, and 639 bp for ITS) included 81 individuals. In this collection, eight species were recognized in Rubus and grouped into two subgenera (Orobatus ans Rubus, Figure 10). The subgenus Orobatus consisted of R. andicola Focke, R. glabratus Kunth, R. lecheri Focke, R. sparsiflorus J.F. Macbr, and R. weberbaueri Focke, whereas the subgenus Rubus consisted of R. adenothallus Focke, R. floribundus J.F.Macbr. and R. loxensis Benth. These two subgenera were distinguished by glands without flexible bristles (Orobatus) and inflorescences in panicle or subracemose forms (Rubus). In the subgenera Orobatus (Figure 11); R. andicola (KUELAP-254) was characterized by elongated branches, spines with short, curved and compressed trichomes, with leaves pubescent on the underside (Figure 11C); R. glabratus (KUELAP-261) by having pink-rose petals and reddish-orange immature fruits (Figure 11A); R. lechleri (KUELAP-258) by its bristly pubescence on the back sides of leaves and purple petals (Figure 11E); R. sparsiflorus (KUELAP-260) by the presence of flowers in dense clusters, crepe-like petals and a pink corolla (Figure 11B); and R. weberbaueri (KUELAP-257) by having veins and spines on the back sides of the leaves, magenta flowers, and black fruits (Figure 11D, Table 11). Genetically, R. weberbaueri and R. lechleri were sister species, differing by 1.0% for rbcL and 0.2% for ITS. These two species were closely related to R. roseus, and the three species differed by over 0.2% for ITS. R. andicola was sister to the clade composed of these tree species, differing over 0.9% for rbcL and 0.2% for ITS. R. sparsiflorus was closely related to the clade composed of these four species and diverged over 0.9% for rbcL and 0.3% for ITS. R. glabratus was closely related to the clade composed of six species of the sugenus Orobatus and differed over 0.9% for rbcL and 0.8% for ITS. Conversely, in the subgenus Rubus, R. floribundus (KUELAP-255) was recognized as a sister species to R. robustus, and both differed by 0.9% for the ITS. R. floribundus had dense inflorescences with pyramidal-shaped paniculata extraaxillaris that tapered toward the lower branches (Figure 11F). These species were sister to the clade composed of R. adenothallus (KUELAP-256) and R. loxensis (KUELAP-259). R. loxensis had creeping-climbing stems and slightly ovate petals and sepals (Figure 11G), whereas R. adenothallus was characterized by small greenish-white flowers and elongated red–black baya (Figure 11H). R. adenothallus and R. loxensis differed by 0.3% for rbcL (Figures S15, S16, S17).

Figure 10.

Phylogenetic tree of the Rubus lineage based on maximum likelihood inference of combined rbcL, GBSSI, and ITS data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above branches. Values lower than 50% (BS) or 0.90 (BPP) are indicated by hyphens (-). The scale bar indicates the number of nucleotide substitutions per site.

Figure 11.

Diversity of Andean berries belonging to Ericales. A. Cavendishia punctata. B. Clethra ovalifolia. C.Clethra retivenia. D.Disterigma synanthum. E.Disterigma ecuadorense. F.Gaultheria secunda.G.Gaultheria sp. 01. H.Gaultheria sp. 02. I.Gaultheria sp. 03. J.Thibaudia angustifolia.K.Thibaudia moricandi.L.Thibaudia ovalifolia. M.Thibaudia obovata. N. Vaccinium meridionale. O. Vaccinium mathewsii. P.Vaccinium floribundum.

Table 11.

Morphological comparisons among species of the genus Rubus.

Species	Habit	Altitude (masl)	Height (m)	Calyx	Immature fruit	Mature fruit	Flowers	Corolla	References
Rubus andicola	Shrub	800–2500	2–3	Green	Green whit trichomes	Red	White–pink	Pink	Focke (1910, 1911), Mendoza and León (2006), this study
Rubus adenothallus	Shrub terrestrial	2000–3500	2–5	Green	Light green	Red–black, red	greenish–white	Greenish–white	Focke (1910, 1911), Mendoza and León (2006), this study
Rubus floribundus	Shrub	1500–2000	1–3	Green	Green	Becoming black	White	Pink	Focke (1910, 1911), Mendoza and León (2006), this study
Rubus glabratus	Trailing–shrublet	2000–3400	0.5–1	Green	Reddish–orange	-	Pink, red	Pink–rose	Focke (1910, 1911), Mendoza and León (2006), this study
Rubus lechleri	Shrub	2000–3600	2–4	Green	Green	Red	White–purple	Purple	Focke (1910, 1911), Mendoza and León (2006), this study
Rubus loxensis	Shrub	2000–3000	2–3.5	Green–reddish	Green, red	Red	Greenish, lilac	Slightly ovate	Focke (1910, 1911), Mendoza and León (2006), this study
Rubus nubigenus	Supporting– shrub	2000–3500	2–3	Green	Green–reddish	Dark fruits	White–greenish, pink	White	Focke (1910, 1911), Mendoza and León (2006)
Rubus robustus	Shrub	1000–3000	1–2	-	–	Black	White–pink	White–pink	Focke (1910, 1911), Mendoza and León (2006)
Rubus roseus	climbing shrub	1600–3000	1–3	Green, purplish tint	–	Red	White–pink	Reddish–violet	Focke (1910, 1911), Mendoza and León (2006)
Rubus sparsiflorus	Shrub	2000–3500	1–4	Green, red–brown	–	Red–purple, black	Black	Crepe–linke pink, style red	Focke (1910, 1911), Mendoza and León (2006), this study
Rubus weberbaueri	Shrub terrestrial	2400–3600	1.5–2.5	Lead–green	–	Black	Pink–reddish	Magenta	Focke (1910, 1911), Mendoza and León (2006), this study

4. Discussion

Most berries from mountainous habitats tend to be more diverse than those from lowland habitats (Powell and Kron, 2003) due to the interactions of UV radiation with environmental (climate) and geographic (relief) factors, which evoke species-specific responses leading to adaptation and diversification (Sedej et al., 2020). Using molecular markers, this study identified 24 species of andean berries (Cavendishia = 1, Clethra = 2, Disterigma = 2, Gaultheria = 4, Thibaudia = 4, Vaccinium = 3, Rubus = 8) from the Amazonas region. The majority of these species were found in coniferous forests, dry and humid forests, rocky slopes, and grasslands at 2,506–3,019 masl (Figure 12).

Figure 12.

Diversity of andean berries belonging to Rosales. A. Rubus glabratus.B. Rubus sparsiflorus.C. Rubus andicola. D. Rubus weberbaueri. E. Rubus lechleri.F. Rubus floribundus. G.Rubus loxensis.H. Rubus adenothallus.

The genus Cavendishia has approximately 130 species distributed throughout the Andes of South America, and most of these species are endemic to Colombia (Pedraza-Peñalosa et al., 2015; WCVP, 2021). Only nine species of Cavendishia have been reported from Peru (León, 2006a; Pedraza-Peñalosa et al., 2015; WCVP, 2021), and two of these were from the Amazonas region. In addition to Ca. isernii Sleumer and Ca. sirensis Luteyn (León, 2006a; Salinas, 2015), this study confirms the presence of Ca. punctata (KUELAP-267, KUELAP-279, KUELAP-293, KUELAP-307, KUELAP-311) in cold and humid habitats in the Amazonas region. Ca. punctata was already recorded from central (Junín and Pasco) and southern Peru (Cusco) at 1,800–2,360 masl, forming sympatric populations with C. bracteata (Pedraza–Peñalosa and Luteyn, 2011). This study also confirms the wider distribution of Ca. punctata along the Peruvian Andes (Table 5).

The genus Clethra consists of 85 species distributed in Africa, America, and Asia (Sleumer, 1967; Fior et al., 2003; WCVP, 2021). Six of the 12 species reported from Peru were registered in the Amazonas region (Cl. castaneifolia Meisn., Cl. ovalifolia, Cl. pedicellaris Turcz., Cl. peruviana Szyszyl, Cl. retivenia and Cl. revoluta Ruiz & Pav) (WCVP, 2021; León, 2006b). This study confirmed the presence of Cl. ovalifolia (KUELAP-282) and Cl. retivenia (KUELAP-295) in Amazonas using molecular data (matK, rbcL and ITS). These species were previously recorded from Cajamarca (northern Peru) and Ucayali (southern Peru) (León, 2006b), and this study found that they occurred in similar habitats (i.e., temperate to humid tropical environments, 2,507–2,800 masl) coexisting with V. floribundum and V. meridionale (Table 6).

Disterigma includes 37 species distributed along cold mountain ecosystems of Central and South America (Pedraza-Peñalosa, 2008, 2009). Of these, 11 species were reported from Peru (Pedraza-Peñalosa, 2008, 2009; WCVP, 2021), and only three species were reported in the Amazonas region (i.e., D. baguense Pedraza; D. ulei Sleumer and D. weberbaueri Hoerold) (Pedraza-Peñalosa, 2009; WCVP, 2021). This study found two new reports of Disterigma for the Peruvian flora, namely, D. ecuadorense (KUELAP-285) and D. synanthum (KUELAP-264, KUELAP-275) (Figure 6). Although D. ecuadorense was considered endemic to Ecuador and D. synanthum to Colombia (Pedraza-Peñalosa, 2008), the analyses of this study confirmed the wider distribution of these species. D. ecuadorense and D. synanthum were found in cold to humid tropical environments at 2,500–3,000 masl and coexisting with V. floribundum (Table 7).

The genus Gaultheria is composed of 130 species from America and Asia (Middleton, 1991; Powell and Kron, 2001; WCVP, 2021). Sixteen species of Gaultheria have been reported from the tropical Andes of Peru (Middleton, 1991; Powell and Kron, 2001). In the Amazonas region, only three species of Gaultheria have been recorded (i.e., G. erecta Vent., G. rigida Kunth, G. secunda J. Rémy) (León, 2006a). Using molecular markers, the presence of G. secunda (KUELAP-278, KUELAP-284, KUELAP-301, KUELAP-288, KUELAP-266) was confirmed from Amazonas. Compared with the average intraspecific divergence observed in other species of the genus (as 0.3% for ITS in G. leucarpa and 0.3% for matK in G. appressa) (Fritsch et al., 2011; Lu et al., 2010), this taxon showed high intraspecific genetic divergence (2.4% for ITS), suggesting the presence of a species complex. Phenotypic plasticity of leaf anatomy (i.e., ovate to elliptic, leaf margins with sharp to rounded apex) among specimens of G. secunda was also observed. These phenomena have been previously reported in Gaultheria under scenarios of a high rate of reticulate evolution and hybrid speciation (Lu et al., 2010; Fritsch et al., 2011; Ocaña-Pallarés et al., 2019). G. secunda was found in wet grasslands and coniferous forest at 2,500–3,500 masl, coexisting with D. synanthum, T. obovata, V. floribundum, and V. mathewsii. This species has also been reported in Cusco, Pasco, Puno, Junin, and Ayacucho (central Peru). Additionally, another three species of Gaultheria (KUELAP-289, KUELAP-302, KUELAP-308) were found and this was not able to assign a species name because only one specimen was found and the diagnostic features of each species were not in good condition (Table 8). These unidentified species need further analyses with additional sampling and molecular markers to confirm their taxonomic status.

The genus Thibaudia consists of 73 species distributed in cloud forests from North to South America (Kron et al., 2002; Powell and Kron, 2003; WCVP, 2021). Approximately 29 species are distributed along areas of grass and shrubs (locally referred to as “pajonales”) and montane forests of the Peruvian Andes (2,500–4,000 masl) (León, 2006a, León, 2006b; WCVP, 2021). Although six of these species have been previously reported from the Amazonas region (Powell and Kron, 2003; León, 2006a, León, 2006b; WCVP, 2021), this study confirmed T. angustifolia (KUELAP-283), T. moricandi (KUELAP-298), T. obovata (KUELAP-272), and T. ovalifolia (KUELAP-299). The latter species was considered endemic to Junin (Central Peru) (León, 2006a). Ecologically, these species inhabit montane forests (2,000–2,800 masl), coexisting with C. punctata, G. secunda, V. floribundun, V. meridionale, and V. mathewsii (Table 9).

The genus Vaccinium consists of ∼400 species distributed worldwide, except Australia (Asturizaga et al., 2006; Vander and Dickinson, 2009). Fifteen species have been reported from the tropical Andes and humid forests of Peru (Pedraza-Peñalosa and Luteyn, 2011; Coico et al., 2016; León et al., 2017; Mostacero et al., 2017; WCVP, 2021). Nine of these species have been recorded in the Amazonas region along steep rocky slopes and montane forests (León, 2006a; Coico et al., 2016). This study confirms the presence of V. floribundum and V. mathewsii and adds one new record of Vaccinium (i.e., V. meridionale) to the Peruvian flora. Although V. meridionale was originally reported as an endemic species from Colombia (Pedraza-Peñalosa and Luteyn, 2011), this study found it in montane forests from northern Peru (2,000–2,800 masl), suggesting that this species has a wider distribution along the Andes. Ecologically, V. meridionale shares the same habitat and coexists with C. punctata, D. synanthum, G. secunda, V. florifundun, and V. mathewsii. Previous intraspecific divergence reported on Vaccinium ranged from 0-0.1% for ITS in V. reticulatum (Kron et al., 2002), while molecular analyses of this study revealed higher distance values within V. floribundum (3.7% for ITS). This could suggest high cryptic genetic diversity, although no morphological differences were found among specimens (Table 10). Sequencing additional markers or plastid genomes might reveal hidden taxa or overlooked interspecific introgression, which has been commonly reported in Vaccinium (Tsutsumi, 2011).

The genus Rubus encompasses ∼700 species distributed worldwide (Focke, 1914; Thompson, 1995; Lu and Boufford, 2003; WCVP, 2021), and only 20 species have been reported from the montane and humid rainforests of Peru (Mendoza and León, 2006; WCVP, 2021). Previously, six species have been recorded from the Amazonas region (Mendoza and León, 2006). This study confirmed R. adenothallus and R. weberbaueri and reported the addition of six species of Rubus from the Amazonas region. Although R. andicola, R. floribundus, R. glabratus, R. lechleri, R. loxensis, and R. spasiflorus were reported from distant regions such as Ayacucho (Central Peruvian Andes), Cusco (South Peruvian Andes) and San Martin (East Peruvian Andes), they inhabited Amazonas. Additionally, R. glabratus was originally described from Ecuador (Mendoza and León, 2006), and this study confirmed this species as having a wider distribution along the Andes. R. adenothallus, R. lechleri, R. loxensis, and R. spasiflorus share the same habitat in humid forests above 3,300 masl. R. andicola, R. floribundus, and R. weberbaueri occur in the mountain undergrowth (sotobosque) at 1,800–2,500 masl. R. glabratus, R. glaucus, and R. robustus are found from montane rainforests to moorlands. The findings of this study reveal the genus Rubus as the most diverse group of berries in the Amazonas region (Table 11).

In the last decade, several of these species have been threatened by the high rate of deforestation, a serious concern that will eventually result in loss of biodiversity and uncontrolled genetic erosion of species with economic and ecological importance (Montesinos-Tubée, 2020; Walker et al., 2021). This study highlights not only the importance of sequencing several molecular markers in applying and validating the names of Andean berries, but also the need to integrate morphological and DNA-based methods to understand the diversity along the Peruvian Andean cloud forest (Bustamante et al. 2021, 2021, 2021; Tineo et al., 2020). The characterization of berries biodiversity is an important element in any future strategy to develop ambitious commitments and tackle research, monitoring and protection programs across the Amazonas region (Sánchez et al., 2021).

5. Conclusions

This study reported 24 species of andean berries distributed in coniferous forests, dry and humid forests, rocky slopes, and grasslands at 2,506–3,019 masl from the Amazonas region. These species are grouped into seven genera and included four new reports on the Peruvian flora. A total of 125 DNA-barcodes of andean berries were generated for four molecular markers (i.e., GBSSI-2, ITS, matK, rbcL). The results of this study suggest that the genetic marker ITS showed better resolution to distinguish species of the genera Clethra, Disterigma, Thibaudia, and Rubus, whereas the combination of the plastidial marker matK and the ITS properly resolved the relationships among species of the genera Cavendishia, Gaultheria, and Vaccinium. Accordingly, an initial screening regarding the diversity of andean berries should include amplification of these markers. This study also confirmed that morphological observations and mainly multilocus phylogeny are needed to reveal diversity of andean berries.

Declarations

Author contribution statement

Daniel Tineo, Danilo E. Bustamante & Martha S. Calderon: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Eyner Huaman: Contributed reagents, materials, analysis tools or data.

Funding statement

This work was supported by SNIP (312252 - FISIOVEG).

Data availability statement

Data associated with this study has been deposited at https://www.ncbi.nlm.nih.gov/genbank/.

Declaration of interests statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.

Appendix A. Supplementary data

The following is the supplementary data related to this article:

Figure S1

Phylogenetic tree of Ericales and Rosales based on maximum likelihood inference of combined matK, rbcL, nrITS, and GBSSI-2 data. Maximum likelihood bootstrap values (BS; ≥ 50%)/Bayesian posterior probabilities (BPP; ≥ 0.9) are indicated above the branches. Values lower than 50% (BS) or 0.90 (BPP) are indicated by hyphens (-). The scale bar indicates the number of nucleotide substitutions per site.

Figure S2

Heatmap showing percentage of genetic identity among species of the genus Cavendishia for ITS marker.

Figure S3

Heatmap showing percentage of genetic identity among species of the genus Cavendishia for matK marker.

Figure S4

Heatmap showing percentage of genetic identity among species of the genus Clethra for ITS marker.

Figure S5

Heatmap showing percentage of genetic identity among species of the genus Disterigma for ITS marker.

Figure S6

Heatmap showing percentage of genetic identity among species of the genus Disterigma for matK marker.

Figure S7

Heatmap showing percentage of genetic identity among species of the genus Gaultheria for ITS marker.

Figure S8

Heatmap showing percentage of genetic identity among species of the genus Gaultheria for matK marker.

Figure S9

Heatmap showing percentage of genetic identity among species of the genus Gaultheria for rbcL marker.

Figure S10

Heatmap showing percentage of genetic identity among species of the genus Thibaudia for ITS marker.

Figure S11

Heatmap showing percentage of genetic identity among species of the genus Thibaudia for matK marker.

Figure S12

Heatmap showing percentage of genetic identity among species of the genus Vaccinium for ITS marker.

Figure S13

Heatmap showing percentage of genetic identity among species of the genus Vaccinium for matK marker.

Figure S14

Heatmap showing percentage of genetic identity among species of the genus Vaccinium for rbcL marker.

Figure S15

Heatmap showing percentage of genetic identity among species of the genus Rubus for GBSSI-2 marker.

Figure S16

Heatmap showing percentage of genetic identity among species of the genus Rubus for ITS marker.

Figure S17

Heatmap showing percentage of genetic identity among species of the genus Rubus for rbcL marker.