The Domestication History of Apples Revealed by Genomic Analysis

By Kevin E. Noonan --

A major conceit of the "genomics" revolution, involving the various species-specific genome projects epitomized by the one for Homo sapiens was the idea that decoding a genome would tell us everything there was to know about the species.  In the first blush of acquiring this detailed sequence-based knowledge was forgotten what Stephen Jay Gould termed the "contingent" nature of natural history (although that view of life has recently been challenged due to more recent genomic research; see Losos, Improbable Destinies: Fate, Chance, and the Future of Evolution, Riverhead Books: August 8, 2017).  In Gould's view it was the "history" portion that was preeminent, because for any species its very existence was exquisitely determined by its genetic and evolutionary path.

Last month, a cohort of China- and America-based researchers* published a paper illuminating the effect of human crop domestication on apples, entitled "Genome re-sequencing reveals the history of apple and supports a two-stage model for fruit enlargement," Nature Communications (at doi:10.1038/s41467-017-00336-7).  These scientists sequenced 117 cultivars of the domestic apple, Malus domestica Borkh.  Their data support the hypothesis that domesticated apples arose from Malus sieversii in the Tian Shan Mountains bordering China and Kazakhstan with "intensive introgressions" from M. sylvestris over the past 4,000-10,000 years.  (Interestingly, Malus sieversii from nearby Xinjiang China appears not to have contributed to apple domestication.)  From these origins, the apple dispersed to Western Europe along the Silk Road with additional introgressions with wild crab apples from Siberia (M. baccata (L.) Borkh.), the Caucasus (M. orientalis Uglitz.) and Europe (M. sylvestris Mill.).

Today, there are thousands of apple cultivars that are famously diverse in fruit size, sweetness, acidity, and skin color; the authors attribute large fruit size to inheritance from M. sieversii and firm texture and appetizing flavor coming from M. sylvestris.

Apples differ from other domesticated fruits and grain, being reproductively self-incompatible and typically being propagated by grafting to heterologous rootstock.  Native apple trees grow disadvantageously tall (~40 meters) for example.  The researchers undertook their study to better elucidate the effects of human domestication on the apple at least in part due to these differences from other domesticated fruit species.

These authors sequenced genomic DNA from 117 cultivars from 24 species ("35 M. domestica (24 scion and 11 rootstock cultivars), 10 M. sylvestris, 29 M. sieversii, 9 M. robusta, 6 M. baccata, 4 M. asiatica, 4 M. hupehensis, and 20 in the remaining 17 wild species with one or two accessions per species"); for context, there are more than 7,500 apple cultivars known.  Genomic sequence information (1060 Gb) was produced (average 9.06 Gb/cultivar accession, about a 12-fold sequence of the apple genome).  Prior work had identified ~57,000 genes in the apple genome; humans have about 30,000).  From these the researchers discerned 7,218,060 single nucleotide polymorphisms (SNPs), useful as genetic markers between the cultivar genomes; 431,597 indels (insertions or deletions) were also identified.  Comparison to the pear genome showed North American-derived apple species (M. ioensis, M. angustifolia, M. fusca, and M. coronaria) were most closely related to pear, while M. domestica and M. sylvestris formed a distinct subclade within the clade further comprising M. sieversii; the paper reports that the introgression of M. sylvestris into M. domestica was so intense that M. domestica DNA has closer similarity to M. sylvestris than to the progenitor M. sierversii species.

Genetic evidence of domestication was exhibited by an analysis of genome-wide nucleotide diversity, which was lower in M. domestica than in M. sieversii, M. sylvestris or other wild species (but higher than peach).  The genomic evidence was also consistent with a weak domestication bottleneck.  An estimated 46% of the M. domestica genome was derived from M. sieversii, with another 21% inherited from M. sylvestris; the origin of the remaining third of the apple genome was reported to be "uncertain."  At the gene level, 840 genes were characterized as having been selected for during initial domestication and 1,089 additional genes selected during introgression; ~29% (246) and ~31% (336) of these two sets of genes showed differential expression during apple fruit development.  Regions of "selective sweeps" (where linkage disequilibrium is suppressed, indicating section for genes involved in related phenotypes) were detected for genes associated with "fruit sugar content, firmness, color, hormone, and secondary metabolism" for the first set of such genes, while genes associated with fruit acidity were associated with the second (introgression) set of genes.  Specific genes so identified included six sugar transporters, genes for several important enzymes in the glycolysis/ gluconeogenesis pathway, and several sucrose and cellulose synthases, in addition to two aluminum-activated malate transporters, a malate dehydrogenase, a citrate synthase (from introgression genes) and a pyruvate decarboxylase in both sets.  These findings are consistent, according to the authors, with " the constant selection of sweet and firm fruits in the history of apple domestication" and "different selective forces for improving different agronomic traits from the two wild contributors during domestication."  Also significant but not subject to extensive explication in this paper were patterns of highly divergent SNPs in M. domestica and other wild species (which the paper states "provided ample information for broadening our understanding of apple speciation, differentiation, and evolution.")

The authors identify fruit size as an important aspect of domestication for food crops generally, and report that two quantitative trait loci (QTL) genes (fw1 on chromosome 15 comprising 11 genes in sweeps from M. sierversii and 8 genes in sweeps from M. sylvestris, and fw2 on chromosome 8, comprising 7 genes in sweeps from M. sierversii and 21 genes in sweeps from M. sylvestris) were found to co-localized in the selective sweep loci.  Specific examples of such genes were genes for two β-galactosidase genes and a cell division regulatory gene, these genes showing expression patterns consistent with "their potential contribution to the increase of fruit size during apple domestication."

The authors synthesize these experimental results to provide a "two-step" model for evolution of the apple with regard to fruit size.  First, and in contrast to species such as corn and tomato, fruit size in ancestral M. sieversii was already relatively large and thus there was less need to apply selective pressure than in other crops.  Accordingly, for apples "a weak selection in a highly heterozygous perennial crop" was able to yield large fruit.  Over thousands of years of domestication, they write, the relatively small increase in fruit size, as well as association of genetic markers for fruit weight, implies that it may be possible to use this knowledge for more selective breeding of larger apples in future.

Another aspect of apple phenotype is firm fruit flesh texture, which imparts a crispy taste, increases shelf life, reduces bruising post-harvest and improves resistance to diseases.  It has been determined that fruit firmness is "directly linked to enzyme-mediated cell wall modification," citing Brummell, "Cell wall disassembly in ripening fruit," Funct. Plant Biol. 33, 103–19 (2006).  Using the selective sweep method, the authors report finding a region of chromosome 16 which both was the subject of "intensive human selection" and encoded cell wall-modifying enzymes, including three polygalacturonases and one glucan endo-1,3-beta-glucosidase.  Additional regions were found on chromosome 17 (having one sweep region encoding three cellulose synthase gene and another encoding a pectate lyase, a glucan endo-1,3-beta-glucosidase and an aldose 1-epimerase.  A sweep region on chromosome 12 was found to encode one endo-beta-1,4-mannase and two pectinesterase genes.  In many cases the authors reported that these genes were found to be differentially expressed during apple fruit development.

Finally, the article reports on genetic evidence for selection for flavor; for apples, this represents a balance between sweetness and tartness (i.e., acidity).  Both traits had undergone human selective pressure during domestication.  For example, a region on chromosome 12 was identified that co-localized with a quantitative trait locus for sorbitol; genes at this locus included four sorbitol transporters and two sugar transporters, with the sorbitol transporter genes showing differential expression during apple fruit development.  Similar evidence was found for genes encoding enzymes involved in sugar metabolism and located on chromosome 13.  With regard to acidity, the paper reports that there is a region ~1.5kb upstream from an aluminum-activated malate transporter at a quantitative trait locus for tart taste.  This region was found to have "substantially reduced nucleotide diversity in M. domestica compared to M. sylvestris," as well as another region of reduced nucleotide diversity located ~750bp downstream from the coding sequence of this gene in M. siervesii.  M. sylvestris introgression also apparently contributed to reduced acidity in M. domestica related to "two other aluminum-activated malate transporter genes and a gene encoding malate dehydrogenase were identified in selection sweeps."

* From State Key Laboratory of Crop Biology, Shandong Agricultural University; Shandong Centre of Crop Germplasm Resources, Shandong Academy of Agricultural Sciences; State Key Laboratory of Crop Stress Biology in Arid Areas, College of Horticulture, Northwest A&F University; The Institute of Pomology, Chinese Academy of Agricultural Sciences; College of Forestry and Horticulture, Research Centre of Specialty Fruits, Xinjiang Agricultural University; Mudanjiang Branch of Heilongjiang Academy of Agricultural Science; Key Laboratory of Biogeography and Bioresource in Arid Land, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Peoples' Republic of China; Boyce Thompson Institute and Section of Horticulture, School of Integrative Plant Science, Cornell University; USDA-Agricultural Research Service, Plant Genetic Resources Unit and USDA-Agricultural Research Service, Robert W. Holley Center for Plant and Health.

Comments

[Paul Cole]

Very interesting and tasty article. Many thanks for this insight into human and plant history.

[Skeptical]

As human history is so often woven with retelling of "facts" and interjections of philosophical spin, the "history" here of the apple makes me wonder at what point in time did the biblical forbidden fruit become symbolized by the apple**, and whether or not that timing coincides with any zenophobic tendency of this fruit appearing in Christian lands from the lands of non-Christians.

**my recollection is that the original forbidden fruit was actually a fig, but I fully permit that this may be but a figMENT of my own imagination... ;-)