A revisited history of cacao domestication in pre-Columbian times

 

The following is a summary of the stated research mentioned above. The content summarized here, including the figures and tables, all belong to the researchers (unless otherwise indicated). The summary attempts to stay as close to the original paper as much as possible with some adjustments in regards to jargon, length, or to focus on bean to bar aspects.

 

Supplementary Information

Click here to download the supplementary information including by the researchers. It includes background information on the history of some cultures mentioned as well as supplementary data provided by the researchers. It is worth a skim through.

Introduction

Cacao has been domesticated for millennia, but little is known about the details of it’s domestication. Theobroma cacao originated in the Neotropics (generally southern Mexico, Central and South America). This study will focus on the residues from ceramic artifacts of various pre-Columbian cultures from both Central and South America. For the first time it is shown that the use of cacao in South America origin extends back 5000 years, supported by interactions between the Amazon and the Pacific Coast of South and Central America. The data illustrates a strong genetic mixing between geographically distant cacao populations, which occurred as early as the middle Holocene. This was driven by human activity who transported T. cacao to new environments. The purpose of better understanding the history of cacao domestication can help better manage cacao genetic resources.

The genetic diversity of crops we have today was shaped by both environmental and human interactions over millennia. Humans have transported and traded various plants to different environments, and also selectively bred based on their preferences, which over time has had a massive impact on the traits of various plants and crops today, including cacao.   Theobroma cacao L. has had both symbolic and economic importance to ancient farmers of the New World. It originated in the Northern Amazon region of South America, with hotspots of biodiversity between Colombia and Ecuador.  As of now, cacao has been classified by many in ten genetic groups (Amelonado, Contamana, Criollo, Curaray, Guiana, Iquitos, Marañon, Nacional, Nanay, and Purus) with a possible 11th called Caqueta from Colombia.

Central America and Mesoamerica is where we know the most about ancient cacao use and domestication which was tied very closely to many cultures in these regions.  Cacao and chocolate held a significant cultural importance for the Mokaya, Olmec, and Maya.  The Criollo variety is considered to be a unique variety in Mesoamerica and Central America.  Most people today consider this region to be the homeland for cacao, but it was actually introduced there from South America. 

We can currently date cacao use and domestication in South America as far back as 5300 years BP, documented in Southern Ecuador where the variety Nacional is believed to have originated. However, the modern Nacional variety used today is a hybrid which includes some Criollo and Amelonado genetics.  Amelonado originated in Brazil, and believed to have been domesticated in the 18th Century.  The origin of todays modern Nacional is believed to have occurred by introducing Nacional genotypes from southeastern Ecuador to the Pacific Coast, followed by the introduction of Trinitario cacao (a hybrid of Criollo and Amelonado) from Venezuela about 100 years ago. 

There are many uncertainties and questions in regards to the timing, the way the cacao migrated, and population diversity of Criollo and Nacional during its domestication. The goal of this work is to track the migration of T. cacao from the Amazonia region of South America, to the Pacific Coast where it was introduced and observed by the Spanish upon arrival on the Ecuadorian Coast.  Ceramics from various locations were analyzed for the presence of T. cacao by looking at aDNA (ancient DNA) which was adsorbed by the ceramics. In addition to this aDNA, the presence of various methylxanthine compounds were also identified and analyzed.

Results

Table 1 summarizes the information gathered from the 352 archaeological items, which represent 19 ancient human cultures from 6 different countries. Methylxanthine analyses were carried out for 326 archaeological items, and aDNA analyses were carried out for 157 items.

Methylxanthine Analyses

Theobroma cacao contains a few methylxanthines (methylated xanthines) which are alkaloids that stimulate heartrate and impact breathing. The two with the highest concentrations in cacao included theobromine and caffeine, with lesser amounts of theophylline. Some other South American plants also contain these methylxanthines, such as mate (Ilex paraguariensis), Guarana (Paullinia cupana), and Theobroma bicolor. However, these other plants contain much lower levels of theobroma and theophylline, so higher levels of these found in ceramics would more strongly indicate the presence of T. cacao rather than other Theobroma or Herrania (genus closely related to Theobroma) species.

A total of 326 ceramic items were tested for methylxanthines (Figure 2). Other studies have noted the fact that some environmental contamination may occur during museum storage, or due to human activity during excavations, and by exposure to water and/or micro-organisms. In order to account for this and avoid false positives, a threshold of 700pg/sample was set in order to consider a sample as positive for a methylxanthine (theobromine, caffeine, theophylline).

Of the 311 items analyzed from South America, 37% were positive for theobromine, 7% for theophylline, and 39% for caffeine (Table 1). Theobromine and caffeine presence were observed in all South American items except for the 7 Marañon items (Fig. 2).

Ancient DNA analyses and Methylxanthine presence

Figure 2

Of the 157 samples analyzed for aDNA, 99 were positive for T. cacao and 31 for Herrania. Of the 118 ceramic samples from South America, 73 had T. cacao aDNA and 27 had Herrania aDNA. The presence of theobromine was the first selection criteria of which samples to analyze. A total of 120 items were analyzed for both methylxanthines and aDNA, and positive for at least one of them. Of these 120 items:

  • 96 (80%) were positive for theobromine

    • Of these 96, 62 (65%) were positive for T. cacao aDNA

  • 81 (66%) were positive for caffeine

  • 62 (76%) were positive for T. cacao aDNA

  • Of the 26 items positive for theophylline, 21 (81%) contained T. cacao aDNA

Therefore a high percentage of T. cacao presence can be predicted by analysing the methylxanthine levels, despite these methylxanthines being found in other South American plants.

Of the 85 items positive for T. cacao aDNA:

  • 62 (73%) are positive for theobromine

  • 62 (75%) are positive for caffeine

  • 21 (25%) are positive for theophylline (this low rate may be due to the fact that theophylline levels exist at much lower levels (5-10000 times less) in cacao beans making them a less sensitive predictor in ceramics).

Diversity and ancestries of cacao ancient DNA

In order to identify the genetic origins of the aDNA found in the various items, it was important to explain and clarify their genetic ancestry, and possible routes from Amazonia. As a reference, 76 modern T. cacao accessions (accessions are specific a cultivar, breeding line, or population) were selected to represent what we know of T. cacao’s diversity. A phylogenetic tree with 460 SNP’s was constructed (Supplementary Figure 5), confirming previous results of eleven genetic groups. The tree in Sup. Fig. 5 illustrates how close the genetic distance is between Criollo and the newly establish Caqueta (from Southern Colombia - close to the Aracuara archaeological site). This phylogenetic tree also illustrates the close genetic relationship between modern Nacional ancestors and genotypes of the Nacional group located at PAL (Palanda archaeological site).

The tree also illustrates the low genetic distance between the Guyanese and Peruvian genetic groups (Marañon, Guiana, Iquitos, Nanay, Amelonado). There is also a close genetic distance between Nacional and Contamana, contrasting some previous results. Representatives of 5 wild relative species were also included in the study.

Table one shows 157 archaeological items analyzes for the presence of T. cacao/Herrania aDNA (independently of methylxanthines analyses). Of these 157, 99 contained T. cacao aDNA. After SNP extraction of these aDNA sequences, genetic analysis could be carried out on 61 archaeological items, and genetic distance analyses on 66 of them.

Supplementary Table 3 (not shown here) displays the aDNA and reference collection. All items had a minimum of 20 SNP markers. They identified a high degree of diversity of cacao ancestries among the human cultural groups. For instance, In the Valdivia culture of coastal Ecuador (spanning 3800-1450 BC), there were samples found related to the Marañon and Contamana (Peru origin) genetic groups, as well as with samples related to Nacional, Criollo, and Amelonado groups. A similar diversity of origins was found in more recent Ecuadorian cultures such as Machalilla and Chorrera where cacao samples were either pure Nacional or a hybrid of Nacional and Amelonado (such as with modern Nacional variety). The samples from Marañon culture in Amazonian Peru show presence of Criollo and Curaray genetic groups.

Genetic diversity was also observed in ancient cacao samples from Colombia, with some highly related to Criollo in samples from Puerto Hormiga and Calima-Ilama cultures. There was also diverse genetic origins found at sites from the Olmec and Maya of Central America, where it was thought Criollo was the only variety consumed. There was evidence that Amelonado, Nacional and/or Iquitos were present by the Olmec period. The genetic distances (Nei’s standard genetic distance) calculated between the ancient DNA and the modern genetic groups confirmed the results (Supplementary Table 4 - not shown here).

It was also found that wild species of Theobroma or Herrania genus were consumed by almost all the cultures analyzed. Aside from a few, it was possible to identify Herrania species used by human populations. Theobroma speciosa was also identified in a sample in an item from Calima Ilama.

Discussion

The domestication of T. cacao dates back to about 5300 years ago in the Ecuadorian Amazon region (Zarrillo et al., 2018). The findings here show a larger landscape of cacao domestication out from its area of origin and to the Pacific coast of South America. Almost 30% of the ceramic items (which were used both in domestic and ritual activities) tested revealed T. cacao ancient DNA.

Theobroma cacao originated in Amazonia, so it’s presence along the Pacific coast reveals interactions between the people in these two regions. The interactions may have included human migration and trade exchanges. Many authors have reported intense commercial exchange along the vast river networks throughout Amazonia (Lathrap et al., 2013; Ortiz et al., 2013). The people here exchanged tools, food, as well as knowledge and ideas.

Amazonia was a center of plant domestication beginning in the Late Pleistocene to Early Holocene, and introducing new diversity by genetic mixing of T. cacao from different origins. Within Amazonia, an important center of resources was Iquitos region of Peru, where several T. cacao groups originated (Marañon, Nanay, Iquitos, Contamana). In Colombia, an independent center of domestication was reported (Piperno & Pearsall, 1998; Piperno, D.R. 2011). The interactions of people from the Pacific Coast and Amazonia likely occurred during the earliest stages of agriculture. This is based on the observation that cacao which originated from several genetic groups located in the Peruvian Amazonia, was observed in the oldest Pacific coast cultures of Valdivia (Ecuador) and Puerto Hormiga & San Jacinto in Colombia, all dating back 5000 years ago. Samples from the Valdivia culture sites Phase III (2950-2600) contained T. cacao genotypes of Peruvian Marañon and Nanay, suggesting people in Valdivia region had contact with those in the Peruvian Amazon. Genotypes related to Nacional were also observed in Valdivia ceramics, suggesting ties between the Mayo-Chinchipe-Marañon culture with those of the coastal Valdivia culture. The Mayo-Chinchipe-Marañon were a mobile people, and navigated the many tributaries and rivers, allowing for rapid spread of plants and products in the region.

As well, samples originated from the Caribbean coast of Colombia (Puerto Hormiga and San Jacinto) where genotypes related to Marañon, Contamana, and Iquitos (all originating in Peru) were observed and links early contact with people of the Peruvian Amazon either directly or indirectly.

Theobroma cacao was introduced to Central America from the Amazon. However, the direction it took to get there (either northward overland, or by the Pacific coast) still raises many questions. Many authors suggest exchanges along the Pacific coast of Ecuador to the Pacific coast of Guatemala based on similarities between ceramics (37, 48-50). Research by Beekman & McEwan (2002) and others points to a long-distance Pacific coastal interactions spanning over 4000 km over 4000 years. Others have pointed out that maritime navigation was possible at these early times (17), and supported cacao dispersal from Ecuador to Mesoamerica through interconnected political-economic networks. More recent work based on DNA analyses have shown that early dispersal of maize (first domesticated in Mexico) all the way to Peru about 6700-5000 cal years Before Present via a rapid coastal migration route from the Pacific lowlands (53). All these findings demonstrate possible rapid exchange between Mesoamerica and the Pacific Coast of South America, and possibly involving cacao from Ecuador.

The archaeogenomic results here have shown a diversity of genetic origins of cacao, and call into question previously proposed patterns of cacao introductions on the Pacific coast of Ecuador and Central America. The modern Nacional variety today has 3 ancestors: Criollo, Ameolando, and Nacional. These already existed on the Pacific coast during Valdivia times (ranging from 3800-1800 BC), contrary to what was originally thought.

This study also suggests the domestication of Criollo is likely even older than previously thought. The origin of Criollo is still unknown, but it shows closer genetic proximity to Caqueta, located in the Colombian Amazonia (4). Criollo was observed in archaeological samples from ancient cultures of Ecuador (Valdivia) and Colombia (Puerto Hormiga, Calima Ilama). Criollo is characterized by many private alleles (meaning unique to specific species/population), suggesting Criollo is likely the result of a long domestication process. Criollo has the fewest genes compared to other genetic groups (25). Many believed Criollo to be the unique variety cultivated by Olmec and Maya in Central America, but results here indicate otherwise. Criollo genes along with other genetic groups from different regions were all present in ancient Central American cultures (Nacional and Amelonado originated from Ecuador and Brazil).

The results of this study indicate how complex the early history of cacao domestication is, and are linked to long-distance trade starting at at least mid Holocene times (7000-5000 years ago). The study also demonstrates the effectiveness of archaeogenomic approaches in tracing the history of plant domestication. The history of cacao is a complex series of events including cacao being brought to new environments encouraging gene-flow, hybridization, and new cacao populations and genetic mixes shaped by their adaptiveness to the new environments and by selective breeding and other human interventions.

Today, cacao is a multi-billion dollar industry and supports economies all around the world. A better understanding of it’s genetics and genetic history may help in combating the challenges it faces today in regards to environment and disease.


References

HistoryGeosephResearch