Collection of native <scp> <i>Theobroma cacao</i> </scp> L. accessions from the Ecuadorian Amazon highlights a hotspot of cocoa diversity

“Fine flavor” cocoa, known for its superior flavor and aroma, commands a higher price farmers than “bulk” cocoa produced market. These fine varieties make an important contribution to the agricultural sector in Ecuador. However, diversity is threatened by deforestation. The effective preservation, characterization, use of tree are therefore essential future development this We carried out participatory collection surveys with local communities Ecuadorian Amazon Forest, evaluate genetic native trees protect as resource communities. Accessing wealth will aid safeguard against climate change develop new El cacao “fino de aroma,” conocido por su sabor y aroma superiores, genera un beneficio más alto para los agricultores que el “corriente” en mercado. Estas variedades fino contribuyen forma importante al agrícola Sin embargo, la diversidad del está amenazada deforestación. Por lo tanto, preservación, caracterización utilización eficaz árboles es esencial desarrollo futuro este Llevamos cabo estudios recolección participativa con las comunidades locales selva amazónica ecuatoriana, evaluar genética nativos proteger como recurso locales. acceso esta riqueza ayudará salvaguardar contra cambio climático desarrollar nuevas Le “fin aromatique,” connu pour sa saveur et son arôme supérieur, rapporte aux agriculteurs prix plus élevé le « courant» produit marché. Ces variétés à apportent une au secteur agricole Équateur. Cependant, diversité du est menacée par déforestation. La préservation, caractérisation l'utilisation efficaces des cacaoyers sont donc essentielles développement futur ce Nous avons mené collectes participatives avec les communautés dans forêt amazonienne équatorienne, afin d'évaluer génétique indigènes protéger arbres tant ressource L'accès cette richesse aidera contre changement climatique développer nouvelles Theobroma L. diploid (2n = 2x 20) perennial species Malvaceae family (Whitlock al., 2001). classification T. accessions according their has evolved significantly over last century. First, two distinct species, Forastero Criollo, were proposed Pittier (1930). Then, Cuatrecasas (1964) suggested subspecies, ssp. sphaerocarpum. Accessions also classified morphogeographical traits Cheesman (1944). More recently, study using simple sequence repeat (SSR) markers allowed more precise into 10 groups (Motamayor 2008). understory tropical upper basin. This region contains widest (Thomas 2012). Their distribution extends throughout Mesoamerica, part due past human-mediated dispersal. history expansion remains be elucidated better understand role human selection shaping diversity. rainforests have lost 20% area 60 years (Nobre 2016), 27% IUCN classification. Future projections show that deforestation could lead decline up 58% Amazonian richness. Species lose average 65% original environmentally adapted (Gomes 2019). continued loss habitat irreversible diversity, which breeders. Indeed, wild constitutes reservoir variability varietal improvement cultivated species. ability farming adapt changing environments or sustainable resilient systems greatly depend on preservation such agrobiodiversity (Satori 2021). Conservation these requires access appropriate land maintain living collections long-term accessibility breeding process. There urgent need rescue, conserve, characterize resources close collaboration processes demand-driven (Kline 2020). A overall characterization would help identify best candidates disease resistance, quality cadmium accumulation bean, there natural variation within Historically, economy Ecuador been closely linked production export particularly based aromatic variety Nacional. recently become top producer South America, 322,000 tons estimated 2018–2019, 6.7% world (ICCO, Nacional variety, called “arriba” when traditionally grown Pacific coast, floral aromas. Over century, gradual introduction foreign material resistant various fungal diseases led mixing (Loor 2009) and, consequently, alterations flavor. Extensive monoclonal plantings, spread nonflavored clone CCN51, used high productivity (Boza 2014), partial downgrading Ecuador's production. meteoric now accounts 70% (Jaimez 2022). current plantations “modern Nacional” correspond mix between ancestral introduced Trinitario 2009). nonintrogressed genotypes, corresponding representatives Nacional, identified hybrid population. analysis large from Upper covering wide geographic (Allen & Lass, 1983; Loor 2012; Pound, 1938, 1945) southern putative In recent years, several studies conducted SSR SNP markers, including those Peru (Arevalo-gardini 2019; Zhang 2006), Brazil (Sereno Bolivia (Zhang 2012), Guyana (Lachenaud 2016; Lachenaud Zhang, 2008), Colombia (Osorio-Guarín 2017). Several hotspots northern Colombia, To enhance rescue context progressive disappearance forest (Figure 1), expeditions organized Amerindian collect trees. structure collected safeguarded compared global microsatellite reported study. Four 2010, 2013, 2017, 2019 provinces Zamora-Chinchipe, Morona-Santiago, Pastaza. sampling design performed 2010 2013 already described al. (2015). 2017 (Dataset S1) following approach involving populations four successive steps (Figures 2 S1). During 2019, 150 collected, bringing total number surveyed 283. geographical areas diverse topology grouped eight zones. Zones 1 8 south north representative different ecological conditions 3; Table 1a,b). compare cacao, previously 2008) included analysis; referred control addition, originating (referred “Morona” population) “Caquetá” (Allen, 1988). 418 samples analyzed work. Leaf stored under vacuum immediately after avoid degradation during travel. Genomic DNA was extracted leaves Risterucci (2000) purified Allegre (2011). 48 selected genomic (Pugh 2004) expressed tag sequencing data (Fouet 2011) dissimilarities individuals. SSRs distributed across linkage (Table PCR amplifications raw Genemapper 4.0 software (Applied Biosystems). our results other studies, individuals matched similar analysis. study, 119 (2012) Amazon, 64 published Motamayor (2008). alleles per locus (Na), expected heterozygosity (He) (Nei, 1978), observed (Ho), private (PA) specific each group calculated GenAlEx 6.503 program (Peakall Smouse, Allelic richness (Ar) allele (PAr) rarefaction HP-rare (Kalinowski, 2004, 2005) correct unequal sample sizes; value g equal twice size smallest organism (el Mousadik Petit, 1996). Unbiased distance (D) Nei's identity index 1972) assess similarity all reference allelic frequencies. Wright's F-fixation indices (Weir Cockerham, 1984). fixation (FIS) inbreeding coefficient GENETIX (Belkhir estimate differentiation FST pairs GENEPOP web version 4.2 (Raymond Rousset, 1995; Two methods infer population investigated model dissimilarity matrix neighborhood assembly method (Perrier 2003), implemented DARWIN 6.0.14 (https://darwin.cirad.fr/). inferred 135 first evaluated concomitance phylogenetic both constructed neighbor-joining 500 bootstrap repetitions uncertainty structure. we Bayesian STRUCTURE 2.3.4 (Pritchard 2000). K 20 tested burn-in period 100,000 iterations, 500,000 Markov chain Monte Carlo repetitions, at least K. optimal K, is, assumed level, determined Evanno (2005) StructureSelector website (Li Liu, 2018). incorporates CLUMPAK (Kopelman 2015), combines many features existing tools postprocessing incorporating calls CLUMPP (Jakobsson Rosenberg, 2007) DISTRUCT (Rosenberg, 2004). intragenetic intergenetic assessed parameters 2). 369 detected markers. (Na) varied 19, Na 7.69 whole found vary 4.56 (Zone 7) 2.92 2) zone, latter having size. (6.15) set (6.69). After correcting size, (Ar[g]) confirmed (5.82) only 12.5% lower (6.65). Each one six (PA), may seem considering proximity surveyed. values comparable groups, where PA 0 6, except Purús (PA 14) Caquetá 15) groups. Considering material, accounted 13.3% alleles. result partly related but greater difference (PAr 0.85) 1.68). FIS zones (0.324) (0.683). majority Hardy–Weinberg equilibrium near zero. some negative showing excess observed, notably Iquitos (−0.360). contrast, positive homozygosity Criollo (0.588) Amelonado (0.753) ranged 0.093 3) 0.309 thus closer Differentiation standardized variance (FST) showed very significant (FST > 0.25) S2). lowest Iquitos–Amelonado 0.3678), Iquitos–Nanay 0.3577), Iquitos–Marañon 0.3825) aligned Moderate (0.15 < mainly zones, closest geographically (Zones 6 7 3 4). Some moderate low 1–3 (0.1330, 0.1314, 0.1517, respectively) Curaray (0.2352 0.1984, respectively). 7, located upstream Morona River, similarities (0.1918 0.2284, surprisingly, (0.1906 0.1771, Marañon Guiana (D 0.274), followed 0.295). highest 2.295). distances fluctuated D 0.05 (between 4) 0.482 5). Small regions 1–3), especially Zone 0.068). 5, 8). An initial clones. It resulting far Furthermore, supplementary include hand Caquetá. tree, it appears group. same then dataset us observe (>70%) three clusters Morona, cluster comprised 1–3) second 5–8) third 4. Most remaining (<30%) irregularly excepted complex admixed base involved defined populations. precisely determine ancestry partition accessions. hierarchical level partitioning revealed 4 S2 S3). hierarchy yielded major subdivisions groups: (i) Curaray, (ii) (iii) (iv) Amelonado-Nanay-Marañon-Iquitos-Guiana-Purús Contamana. At appeared subdivisions. Further values, differentiating Contamana separated 7; 8; 11. dissociated 14. Finally, did not dissociation Nanay ancestries ancestries, even generally, taken consideration, newly distinctly depending zone site. 2, group, further north, 8, Surprisingly, starting mostly nearby Pangui formed 283 characterized chosen flexibility testing small numbers discriminatory power (van Inghelandt 2010). Our increased surveys, highlighting hotspot present 19 (average 7.69), assign zone. banks River village San José 6) 8) being Curaray-type altitudes 200 350 m) climatic (cloud cover, rainfall, temperature). Nacional-type 800 1,200 m altitude mountainous (cordilleras). genetically 0.695) represent constituting presence reflects short most distant (approximately 300 km straight line). Adaptation strong geo-climatic variations partially explain distance. 49) set, although they well represented regional (Curaray, Caquetá). Compared (320), gain 15.3% Different authors examined centers hypothesized origin Equator (Cheesman, 1944; 2008; 1945). Taken together, Thomas (2012). Population analyses main NJ clustering methods; differentiated Nacional-Morona, containing formerly “Forasteros,” Amelonado, Purús, Guiana, (Nanay, Marañon, Iquitos, Contamana). four-group highlighted LCT-EEN 2012) 1980 1985 (K ≥10, possible separate (2008), suggesting enough applied if had >0.90. explanation sufficient discriminate before around (north Zamora-Chinchipe region), 4, 5. equivalent proportion representing 1) relatively (Ho 0.397) suggest seemed rather substructure ancestry, considered subpopulation. Approximately 30% Nacional–Morona, sometimes villages served road connections; certainly imported like TIW063 (Amelonado) hybrids previous paternity Amazonia countries genotypes LCT-EEN85, LCT-EEN86, LCT-EEN91, Yacuambi Nangaritza rivers 1988), likely ancestors among studied. vicinity Shaïme 2), LCTEEN86 (SHAM001, SHAM002, SHAM003, NANK006) LCTEEN91 vast Palanda 1). approximately 50 Palanda, 44 49 old ceremonial archeological site dating back 5,500 discovered Santa Ana-La Florida Palanda; degree societal time (Valdez 2005). starch grains, theobromine, characteristic ancient ceramic (Zarrillo Trees controls. All Mayo-Chinchipe people played domestication variety. found, watershed connected paths still Evidence interzone connections (Valdez, facilitated northerly 3). hypothesis trees, once coast 2009), originated supported historical data. Guayas region, crops coast. Marine shells Ana–La attest coastal regions. As work qualities importance potential currently (Colonges, Solorzano, 2022; Colonges, Seguine, enrichment should provide, framework programs, growing Participatory programs initiated promote increase livelihood. basal position phylogeny reinforced derived highly (Cornejo came subset With population, 0.835), visualized Among observations frontier Ecuador, strongly rapid emergence trade globalization. (2010 2019) colleges constitute enhanced can create Amazonia. discovered. Given allow identification niches, providing income while contributing maintenance research clarified gives clues about another coming safeguarding play adaptation regarding change. thanks INIAP CIRAD. thank Agropolis Foundation, MUSE (Montpellier Université d'Excellence), Valrhona financial support project. work, Amazcacao project, publicly funded through ANR (the French National Research Agency) “Investissement d'avenir” (Agence Nationale Recherche) ANR-16-IDEX-0006. grateful GPTR (Great technical platform) Montpellier core facility support. populations, Shuar, Achuar, Kichwa communities, Pangui, Real Audiencia (San Jose), Raime Roldos (Santiago), Los Angeles (Taïsha), Tuna (Kapawi) invaluable surveys. declare no conflict interest. R.G.L.S., P.C. C.L. conceived coordinated project; C.S., F.F. established formalities government authorities enter Indian territories; R.G.L.S, D.C., F.F., seminaries colleges; O.F., F.A. B.Y., I.S. multiplicated germplasm collections; designed studies; R.R., K.C., H.V. genotyping; B.R., X.A. data; X.A., wrote manuscript. findings available author upon request. Data S1. passport 2019. (INIAP: Instituto Investigaciones Agropecuarias, PNCC: Programa Cacao Café, NA: Not Available) Dataset S2. Genetic (fixation FST, right panel, bold numbers) (D, Nei 1972, bottom left panel) estimates (control accessions). Fig. S1 Map (Instituto Agropecuarias) stations localities progenies clones implanted. Determination ΔK (Evanno S3 bar plots Description polymorphic Please note: publisher responsible content functionality any supporting information supplied authors. Any queries (other missing content) directed article.