Grapevine domestication: A tale of two centers and selected traits

Dong, et al., (Dual domestications and origin of traits in grapevine evolution, Science, 3/3/23) have postulated, based on their study results, a dual origin for vitis vinifera stemming from simultaneous domestications of distinct vitis sylvestris pairs in separate geographic regions. I have previously described the study origin, the history of the progenitor species vitis sylvestris, and the case for dual domestication. In this post I will provide a contemporaneous look at the proposed domestication centers and identify traits selected during the grapevine cultivation process.

The Domestication Centers

The figure below locates the Western Asia (Syl-E1 to CG1) and Caucasus (Syl-E2 to CG2) domestication centers. 

Grape domestication centers (as per Dong, et al.)

According to the authors, the dual domestications occurred 11,000 years ago. The Little Ice Age had ended by this time and the world was transitioning from the Pleistocene to the Holocene Epoch. 

The Western Asia physical environment was varied, inclusive of highlands, mountain ranges, fertile river valleys, and arid and semi-arid landscapes. Notable geographic features included the Taurus and Zagros Mountains, the Anatolian Plateau, the Levant, the Mesopotamian Plain, and the Jordan River Valley. The Caucasus had a similar distribution of environments but with more prominent mountain ranges. River valleys, such as those hosting the Aras and Kura Rivers, provided conditions suitable for early agricultural development.

The populace of the two centers were mainly hunter-gatherers (hunting, fishing, and gathering), with some early agriculturists. Early farmers likely cultivated wheat, barley, legumes, and flax and domesticated animals such as goats, sheep, and cattle.

It is believed that the Caucasus was inhabited by several indigenous groups but the demographic composition is difficult to determine with any precision. Western Asia, on the other hand, was home to some notable civilizations to include the Natufians (Levant) and the pre-pottery Neolithic cultures of Hassuna, Halaf, and Samarra (Mesopotamia).

Domestication Intent and Thrusts

The authors investigated domestication signatures on the cultivars and found both shared and unique manifestations on the cultivars. Most of the shared signals are on chromosomes 2 and 17, "confirming previous findings that that the selection on flower sexual morphs (sex determination region, SDR), berry skin color (VvMybA gene cluster), and berry development (SDH gene cluster) were of great importance during grapevine domestication." The authors further identified shared domestication genes that possibly serve as foundations for grapevine growth, physiology, fruit set, and resistance to biotic/abiotic stress.

The unique manifestations are exhibited in cases where "the same goals were advanced  by using different aspects of common domestication traits." Some examples of the latter are as follows:

• Improvement of berry palatability through the reduction of alkaloid biosynthesis (MecghR gene cluster in CG1 and TR2 and SSL gene clusters in CG2)
• Enhancement of carbohydrate metabolism (SWEET17 in CG1 and PFKFB1 in CG2)
• Perceived berry desirability (BEAT gene cluster for floral scent in CG1 and UFGT gene cluster for berry color in CG2)
• Response to environmental stresses (UPL6 in CG1 and WAK in CG2).

According to the authors, "... their findings suggest that the initial cultivations of CG1 and CG2 may have been to serve early humans' caloric and micronutrient needs." And such an observation fits with the economic environment described previously; that is, a landscape populated primarily by hunter-gatherers and fledgling agriculturists.

In that both cultivars were domesticated as food sources, the selection of factors suitable for winemaking exhibited in CG2 "could have been serendipitous, and the practice of winemaking with CG2 ... possibly postdates grapevine domestication."

In my next post I will examine the spread of these cultivars outward from their domestication centers.

©Wine -- Mise en abyme

The case for dual domestication of vitis vinifera, as per Dong, et al.

The recent Dong, et al., study has upended the conventional wisdom of a single-point vitis vinifera domestication somewhere in the Caucasus over 8000 years ago. In a series reporting on the study, I have specified the study origin and the history of vitis sylvestris, the accepted progenitor of vitis vinifera. In this post I cover the study findings as regards the domestication of vitis vinifera.

Vitis vinifera sylvestris (Source: floredusud.com)

According to the Dong, et al., study, the wet climate of the Early Holocene (11,700 - 8300 years ago) expanded the geographic spaces capable of sustaining Syl-E (identified in my prior post as the eastern ecotype of v. sylvestris) and the species rose to the occasion by moving westward, occupying a "large geographic space from Central Asia to the Iberian Peninsula."

I have identified six genetic ancestries for cultivated grapes reported out in the Dong, et al., study -- CG1 to CG6 -- each associated with a specific geographic area. For example, CG1 is associated with Western Asian table grapevines and CG2 with Caucasian wine grapevines. CG1, according to the study, "shares the main ancestral components with Syl-E1" while CG2 shares its main components with Syl-E2. This state of affairs suggests the possibility of two domestication events. Further:

• CG1 and CG2 maintain the highest genetic diversity of all the CG groups
• CG1 and CG2 manifest the greatest Linkage Disequilibrium decay among all CG groups
• CG1 and CG2 are less differentiated from their corresponding wild ecotypes
• The Akaike information criterion-based phylogenetic selection prefers a dual origin tree model which agrees that CG1 and CG 2 are genetically closer to Syl-E1 and Syl-E2, respectively
• The population split lines of CG1/Syl-E2 and CG2/Syl-E1 pairs resemble that of Syl-E1/Syl-E2 and differ from those of CG1/Syl-E1 and CG2/Syl-E2 pairs.

These data, according to the study, "collectively support a dual origin of v. vinifera and reject the theory of a single primary domestication center."

Both sylvestris/cultivar pairs (CG1/Syl-E1 and CG2/Syl-E2) separated quickly which, according to the authors, is compatible with a "clean-split" scenario. The authors estimate the median population split time to be ~ 11,000 years ago for both pairs "suggesting that the domestication events took place concurrently around the advent of agriculture."

In that CG1 represents the Western Asian table grapevine, and CG2 the Caucasian wine grape vine, the dual-origin schema advanced herein rejects earlier assumptions that wine grape domestication predated table grape domestication. To reiterate, the data suggest vitis vinifera simultaneous origins in Western Asia and the Caucasus, covering both wine grapes and table grapes.

The implications of the foregoing are mind-blowing. It suggests that Syl-E1 and Syl-E2 had self-selected for characteristics that rendered one more palatable as a food source and the other as a beverage. Further, it means that hunter-gatherers within the regions had been interacting with these ecotypes in their natural habits and habitats for many years prior to the instance of dual selection and propagation of the characteristics embodied in CG1 and CG2.

My next post will highlight how domesticated grapes spread from these points of origin to the major wine regions.

©Wine -- Mise en abyme

Rewriting the story of ancient grapes: Establishing genetic populations and the story of V. sylvestris

A recently published grapevine genetic study (Y. Dong, et al., Dual domestications and origin of traits in grapevine evolution, Science 379 (6635), pp. 892 - 900, 3/3/23) "has upended the history of how humans first domesticated grapes for winemaking ..." (Melanie Lidman, It's in the DNA, Times of Israel, 3/25/23). I discussed the factors driving this study, as well as its objectives, in a prior post. In this post I describe study preparation efforts, the steps taken to define ancient vine genetic populations, and the stories hidden in the genes of V. sylvestris, the accepted progenitor of V. vinifera.

Data Collection

As discussed in my lead-in post, in 2019 Dr Chen and his lab reached out to colleagues around the world asking them to contribute material towards the study. A total of 3525 samples of genetic material were received at the State Key Laboratory for Conservation and Utilization of Bio-resources of Yunnan Agricultural University. The samples were distributed as follows:

• 2503 Vitis vinifera, 1022 Vitis sylvestris
• 3186 from Eurasian germplasm and private collections
• 2237 V. vinifera, 949 V. sylvestris
• 339 from previously sequenced samples

According to the authors, the sample population “deferentially included old, autochthonous, economically important varieties to maximize the spectrum of genetic diversity." Lidman asserts that Israel's sample submission "... constituted the largest contingent of wild grapes from a single country, or about 10 percent of the total wild grapes sequenced for the study."

According to Karlsruhe Institute of Technology (KIT),  (scitechdaily.com, Scientists Determine the Origins of One of the World's Oldest Crops, 4/17/23), it contributed "... its globally unique collection of European wild vines and very old medieval species" to the study. It also connected the research team with the Ukraine researchers who had fled Crimea -- along with the vines from the Magarach collection -- after the 2014 Russian annexation.

Data Cleanup

In this phase of the study the team "weeded out clones, mutants, synonyms, homonyms and duplicates," ending up with 2448 grapevines (1604 V. vinifera and 844 V. sylvestris) and 498 distinct genotypes.

Categorization

The team first utilized Principal Component Analysis as a mechanism for determining whether "viticultural region" was a key element in defining grapevine diversity; according to the results of the analysis, it was not.

The team subsequently "leveraged genetic ancestry information from an unsupervised ADMIXTURE analysis" (ed., a method of inferring geographical origins based on an analysis of genetic ancestry) to categorize the core accessions. The findings from this analysis were as follows:

• The hierarchical clustering of ancestry components identifies four V. sylvestris groups from distinct geographic regions
•  Western Asia
•  Caucasus
•  Central Europe
• Iberian Peninsula
• V. sylvestris accessions collected from other regions show admixed genetic structures
• For cultivated grapevines, six genetic ancestries could designate six distinctive groups, all covering a broad range of geographic regions
• Accessions with pure ancestries helped to "ascribe" names to these groups
• CG1 -- Western Asian table grapevines
• CG2 -- Caucasian wine grapevines
• CG3 -- Muscat grapevines
• CG4 -- Balkan wine grapevines
• CG5 -- Iberian wine grapevines
• CG6 -- Western European wine grapevines

"The four sylvestris and six vinifera groups formed identifiable clusters in the PCA plots and were thus suitable for population genomic investigations."

The V. sylvestris Story

Based on “genetic ancestries and the occupied ecological niches in the western Eurasia continent, the team designated the V. sylvestris accession from Western Asia and the Caucusus as the eastern ecotype and the accession in Central Europe and the Iberian Peninsula as the western ecotype. The similarities and differences between accessions drive this designation. For example, “both nucleotide diversity and individual heterozygosity show that the western ecotype has significantly reduced variation compared with its eastern counterpart.”

In tracing the sylvestris genetic history, the authors show that the eastern sylvestris (Syl-E in their parlance) flourished in its range from approximately 1.5 million years to 800 000 years ago. The Pleistocene was a time of changing climate cycles, however, and the population experienced a bottleneck between 800,000 and 400,000 years ago.

Illustration of a population bottleneck (Source: alevelbiology.co.uk)

Somewhere between 400,000 and 200,000 years ago there was a divergence in Syl-E, giving rise to Syl-W. According to the analysis, the geographical split which drove the divergence was gradual. The Syl-W grouping experienced its own population bottleneck 400,000 - 150,000 years ago.

The last glacial cycle ran from 115,000 to 11,700 years ago and was characterized by ”global climate trending towards drier and colder conditions.” It was during this period — approximately 56,000 years ago -- that the Syl-E ecotype split into the Syl-E1 and Syl-E2 designates. It was also during this period (approximately 40,000 years ago) that there was a system-wide bottleneck which drove vine population to between 10,000 and 40,000. Syl-W experienced its own divergence, designated by the study team as Syl-W1 and Syl-W2.

In my next post I will take up the tale of V. vinifera, as told by the genes.

©Wine -- Mise en abyme