Abstract
Many apple varieties commonly planted in the United States a century ago can no longer be found in today's orchards and nurseries. Abandoned farmsteads and historic orchards harbor considerable agrobiodiversity, but the extent and location of that diversity is poorly understood. We assessed the genetic diversity of 280 apple (Malus ×domestica Borkh.) trees growing in 43 historic farmstead and orchard sites in Arizona, Utah, and New Mexico using seven microsatellite markers. We compared the samples to 109 cultivars likely introduced into the southwest in the late 19th and early 20th centuries. Genetic analysis revealed 144 genotypes represented in the 280 field samples. We identified 34 of these 144 genotypes as cultivars brought to the region by Stark Brothers Nursery and by USDA agricultural experiment stations. One hundred twenty of the total samples (43%) had DNA fingerprints that suggested they were representative of these 34 cultivars. The remaining 160 samples—representing 110 genotypes—had unique fingerprints that did not match any of the fingerprinted cultivars. The results of this study confirm for the first time that a high diversity of historic apple genotypes remain in homestead orchards in the U.S. southwest. Future efforts targeting orchards in the southwest should focus on conservation for all unique genotypes as a means to sustain both cultural heritage and biological genetic diversity.
The late 19th century is often referred to as the golden years of apple growing in the United States (Calhoun, 1995; Hensley, 2005). Farmstead and kitchen orchards were planted with a wide variety of fruit trees to suit diverse family needs. Historically, rural livelihoods were maintained by growing apples and pears that ripened in summer, would “keep” all winter in the cold cellar, produce desirable ciders, and those that were amenable to cooking and baking. This period of American horticultural history was preceded by an era of fruit diversification, which spanned much of the 1700s and into the early 1800s. Thousands of trees emerged from seedling orchards, planted for cider and for animal feed, that bore high-quality fruits worthy of naming, conserving, and distributing. These trees were clonally propagated from cuttings and traded and sold to become elements of the diverse orchards of the 19th century (Beach, 1905; Hedrick, 1950). The last remnants of 19th century plantings may still be alive in American landscapes and are the subject of this regional southwestern survey.
USDA pomologist W.H. Ragan undertook the task of recording the names and characteristics of every apple cultivar grown in the United States during the 19th century. In his book, The Nomenclature of the Apple (Ragan, 1905), Ragan lists 6654 unique named apple varieties that he found referenced in U.S. literature between the years 1804 and 1904. In 1980, Dan Bussey began expanding on Ragan's register to update descriptions of known cultivars and include additional apple cultivars referenced in the U.S. literature up to the year 1980. Over a decade after beginning this project, Bussey (in press) is close to releasing The Apple in North America, which lists over 14,000 named apple cultivars introduced to or selected in North America.
Modern commercial apple production requires consistency of ripening time, quality retention during processing and shipping, and long storage life, and not all varieties can meet these criteria (Goland and Bauer, 2004). Market pressures have reduced the diversity of fruit trees once grown in small family orchards—where diversity of ripening time, sizes, textures, and flavors were celebrated—to only a few handfuls of commonly planted commercial cultivars. Currently, 11 apple cultivars account for over 90% of the apples sold in the United States, with ‘Red Delicious’ constituting 41% of this figure (Dennis, 2008). In The Fruit, Berry and Nut Inventory (Whealy, 2001) Kent Whealy lists ≈1500 apple varieties currently available through U.S. nurseries, many of which have been developed through modern fruit breeding. This suggests a substantial decrease in the number of apple cultivars offered through U.S. nurseries over the past century (77% by Ragan's calculations and 89% by Bussey's), although we do not know to what extent this naming actually represented genetically distinct cultivars.
Although the loss of on-farm diversity can be lasting, fruit trees have an advantage over annual crops because these trees can live to remarkably old ages, surviving some fads in consumer demand. Single apple trees have been known to live 150 years or longer. In many areas, it is still possible to find trees of “heirloom” cultivars once abundant at the beginning of the 20th century. Remnant orchards planted before the “modern era” of fruit production (Jackson, 2003) hang on tenaciously around abandoned farmsteads and historic orchards. Although farmstead trees often persist without their original names being retained, they represent a snapshot of the diversity of fruit varieties available over a century ago during the peak of fruit tree diversification.
Morphological and taxonomic traits typically used to differentiate between apple cultivars can be ambiguous as a result of the broad phenotypic variation under different environmental influences. Furthermore, many 19th century apple cultivars are morphologically similar to one another and accurate descriptions are often lacking, making conventional identification methods difficult, if not impossible, for these century-old trees.
Genetic fingerprinting, including microsatellites, have become powerful and accurate tools for analyzing genetic diversity (Hammer et al., 2003). Microsatellite loci, or simple sequence repeats, are short nucleotide sequences of up to six basepairs repeated in tandem, head to tail, without interruption. They are highly polymorphic, codominant markers that have been detected in every organism thus far studied (Hancock, 1999). Gianfranceschi et al. (1998), Hokanson et al. (1998), and Hemmat et al. (2003) were instrumental in developing Malus-specific microsatellites in the United States. In Europe, Silfverberg-Dilworth et al. (2006) and the High-Quality Disease Resistant Apples for Sustainable Agriculture have been forefront in developing microsatellites for Malus. In a similar study in Spain, Pereira-Lorenzo et al. (2007) evaluated the genetic diversity of 114 Spanish apple cultivars comparing the local Spanish landrace genotypes with 26 commercial apple cultivars found in the region. Similar to this study, Pereira-Lorenzo et al. (2007) found high levels of genetic diversity in apple trees in northern Spain based on observed heterozygosity.
In this study, we sampled tissues from 280 apple trees from 43 historic sites within Arizona, Utah, and New Mexico to assess their diversity. Genotypes identified using seven highly variable markers were compared with reference genotypes of known cultivars. The term “historic” is used in this text to refer to farmstead and orchard sites planted during the late 19th and early 20th centuries, and the term “heirloom” refers to cultivars introduced during the 19th century and before as opposed to recent introductions developed through modern fruit breeding programs.
Materials and Methods
Field collections.
We collected leaf samples from 280 apple trees located in 43 historic orchard sites on public and private (but not tribal) lands in Arizona, New Mexico, and Utah (Fig. 1). Sampling took place from June through September of 2007. We targeted places with presumed historic orchards and trees of visually differing morphologies. Local experts provided site location information. Sampling permission was obtained and samples were used only for this study, not for gene banking or crop improvement.
Historic apple collection sites (numbered and noted by diamonds) in Utah, Arizona, and New Mexico were sampled. Cities are provided for reference purposes (star designation).
Citation: HortScience horts 44, 3; 10.21273/HORTSCI.44.3.589
We focused sampling efforts primarily on historic farmstead and orchard sites dating back to the 1930s and earlier with priority given to trees planted before 1920. For a handful of the orchard sites such as Capitol Reef National Park, UT, and Slide Rock State Park, AZ, orchard planting dates could be found in historical documentation. Most of the orchard sites lacked written documentation, however, and we had to rely on oral history or visual determination of tree size and locality for approximate ages. We avoided sampling from seedling trees and rootstock trees where it appeared the original grafted top had died. The dry southwest climate limits establishment of naturalized seedling apple trees and many farmstead orchards in the southwest date to the late 19th to early 20th centuries, a time when few seedling orchards were planted in the United States. Leaf samples were collected only from trees that were visibly different from other trees at the same site to avoid repeat sampling of cultivars. However, this was not always possible for trees without fruit.
Global positioning system locations for the sample trees (Garmin eTrex-Vista handheld unit; Garmin, Olathe, KS) in UTM (Nad1983) coordinates were recorded, and small aluminum tags were nailed to the trees with a numerical identifier. For each sample, ≈50 mg of fresh leaf tissue was placed into a 96-well plate and frozen at –20 °C until extraction. Known cultivars were obtained from the USDA-ARS-Plant Genetic Resources Unit, Geneva, NY (PGRU). Varieties that were not available through the Geneva facility were obtained from Lee Calhoun of Calhoun's Nursery in Pittsboro, NC; Ram Fishman of Greenmantle Nursery in Garberville, CA; and Gordon Tooley of Tooley's Trees in Truchas, NM. Leaf samples were processed in the same manner as the unknown samples.
Microsatellite analysis.
The genetic analysis of the 280 samples and 109 cultivars was performed following procedures described in Volk et al. (2005). We extracted genomic DNA from the leaf samples using Qiagen DNeasy 96 plant kits (Qiagen, Valencia, CA). Two separate sets of DNA were extracted from each sample and run independently. Malus-specific microsatellites were amplified using unlinked primers (GD12, GD15, GD96, GD100, GD142, GD147, and GD162) as described by Hokanson et al. (1998) and by Hemmat et al. (2003). Forward primers were labeled with either IRD700 or IRD800 infrared florescent dyes (MWG-Biotech, High Point, NC). Unlabeled reverse primers were obtained from IDT (Coralville, IA).
Polymerase chain reactions (PCRs) were carried out in 15 μL total volume. Each 15 μL reaction contained: 0.3 μL GoTaq® Flexi Taq Polymerase (Promega, Madison, WI; 5 units/μL), 3 μL Promega 5× Colorless GoTaq® Flexi Buffer (10 mm Tris-HCl, 50 mm KCl, and 0.5% Triton X-100), 1.5 μL of 0.25 mm MgCl2, and 1.5 μL of 0.25 mm dNTPs. Forward and reverse primers were added to a final concentration of 0.25 pM/reaction except for GD12 at 0.3 pM/reaction and GD100 at 0.5 pM/reaction.
Genomic DNA, isolated as described previously, was added at 0.5 ng to 5 ng/reaction. Reaction volumes were adjusted to 15 μL using sterile distilled H20. PCR reactions were multiplexed with the following primer sets: GD12, GD100; GD142, GD147, GD162; and GD15, GD96 run together. PCR was carried out using MJ Research PTC 200 Thermocycler (Reno, NV). Amplifications were done using touchdown PCR, in which the thermocycler reduced the annealing temperature 1° every cycle, starting at 63 °C and ending at 54 °C, followed by an annealing temperature of 55 °C for 18 cycles and ending with a 2 min 72 °C extension.
PCR products were diluted 1:1 with a loading buffer of formamide bromophenol blue loading buffer and were denatured at 95 °C for 5 min. Denatured products were diluted 1:10 with additional loading buffer. Diluted products were loaded on gels (6.5% KB Plus acrylamide; LI-COR, Lincoln, NE) and run in 1× TBE buffer (89 mm Tris, 89 mm boric acid, and 20 mm EDTA) for 1 h, 45 min at 1500 V, 40 W, 40 mA, and 45 °C in a LI-COR 4200 DNA Sequencer. Digital images of the gels collected by LI-COR Saga Generation2 software were manually analyzed using Saga software. Each allele at each locus was manually scored in Saga before being compared with the duplicate sample.
Ploidy was determined using flow cytometry by Gerard Geenen of the Plant Cytometry Services, Schijndel, The Netherlands.
Microsatellite data analysis.
Genotypes for the 280 samples and the 109 cultivars were compared manually in Microsoft Excel 2004 for Mac, Version 11.3.7 (Microsoft, Redmond, WA). Allele frequencies and observed and expected heterozygosities were computed using GenAlEx version 6 (Peakall and Smouse, 2006). Principal component analysis (PCA) ordinations were preformed using PC-ORD version 4.0 (MjM Software Design, Gleneden Beach, OR).
Results and Discussion
We found considerable genetic diversity in historic southwest orchard and farmstead sites. The “unknown” apple trees were compared with 109 known cultivars introduced into the southwest in the late 19th century by USDA agricultural experiment stations and by Stark Brother's Nursery, the largest mail order nursery during the late 19th and early 20th centuries. DNA fingerprints revealed that 120 of the 280 sample trees were indistinguishable from the fingerprints of 34 cultivars (Tables 1 and 2). The remaining 160 historic tree samples did not match any of the reference cultivars. These 160 samples represented 110 unique genotypes. These unknown genotypes could be regionally unique cultivars, local seedlings, cultivars extinct from the nursery trade, or extant cultivars originally from other regions that have not been recorded as being introduced into the southwest. For two samples, the duplicate genotypes did not match, suggesting different DNA source material. This may have been a result of leaves being collected from vegetative rootstock and from the grafted tree or a result of human error. Both samples were discarded from the study. In total, the 280 historic trees represented 144 distinct genotypes.
Ploidy, origin, and probable date of release are provided for 109 apple cultivars introduced to the southwestern United States.z
Identification of cultivars identified in historic farmsteads in the southwestern United States.
Only five of the 34 identified cultivars appeared to be commonly distributed. ‘Ben Davis’, ‘Delicious’, ‘Grimes Golden’, ‘Jonathan’, and ‘Winesap’ each represented more than four trees in the study found at multiple locations. A number of trees matched genotypes of named cultivars for all but one or two alleles at the locus GD100 (Table 2). It is not known if these 1- or 2-bp shifts represent morphologically different varieties or were a result of error during allele scoring.
The seven microsatellites were sufficient to differentiate between most samples and cultivars in this study. However, several of the cultivars had identical fingerprints. This suggests that these cultivar names are synonyms of the same cultivar, are close sport mutations not differentiable by these microsatellites (see Hokanson et al., 1998), or are mislabeled at their source nursery or genebank location. Named sets of cultivars Albemarle Pippin and Yellow Newtown Pippin, Early Strawberry and Yates, Fameuse and Canada, and Maiden Blush and Chenango Strawberry were indistinguishable from each other.
Ploidy results revealed 24 of the 280 samples (8.6%) were triploid (3x = 51), whereas the remaining 91.4% were diploid (2x = 34). Triploids arise spontaneously in 2×-by-2× crosses and typically have larger fruit than diploid apple trees (Ferree and Warrington, 2003). Based on field observations, many of the triploids in this study appear to be late-ripening, large-fruited winter apples.
High levels of observed heterozygosity (0.92 for GD142, 0.90 for GD162, and 0.88 for GD147 and GD96) in the samples suggest that relatively high levels of genetic diversity are represented (Table 3). Heterozygosity was calculated from a sample size of 144 individuals, representing each of the 144 distinct genotypes found growing in the southwest. Observed heterozygosity was calculated by dividing the number of heterozygotes at a locus by the number of individuals surveyed. Expected heterozygosity assumes Hardy-Weinberg equilibrium but is included as a reference. The multiplicative probability of a multilocus genotype determines the power of discrimination; the high heterozsygosity of the sampled loci therefore makes the probability of distinctly different genotypes being identical at all seven microsatellite loci very slight. Microsatellite GD100 had 6% missing data as a result of poor amplification during PCR for several samples. We chose not to score this allele in these cases to avoid potential scoring error. Heterozygosity and allele frequencies should be interpreted with dubiety for this allele.
Sample size, number of alleles, observed heterozygosity, and expected heterozygosity were calculated for samples with unique genotypes.
PCA was performed on the samples to visualize the genetic difference between apple genotypes (Fig. 2). In the PCA ordination, the two distinct clusters appear to be the result of the presence of two larger alleles at marker GD12 (182 and 190 bps) instead of the more common 148- to 162-bp length alleles. Such clustering could be indicative of a shared genetic heritage between genotypes in the upper left cluster, although there does not appear to be any morphological association with the groups.
Principal component analysis of unique apple genotypes. Sample numbers and state of origin are provided for unknown varieties and names are provided for known varieties.
Citation: HortScience horts 44, 3; 10.21273/HORTSCI.44.3.589
Unknown genotypes were labeled in the PCA with a state prefix attached to the end of the sample number to show possible associations of geographic origin to genotype. Geographic separation of samples would indicate that different sources of apple trees were grown in the various regions. Spanish priests, explorers, and settlers introduced apple trees to New Mexico as early as the 17th century (Dunmire, 2004) as did the Archbishop Lamy of Santa Fe in the 19th century (Horgan, 1975). Geographic differentiation could also imply different apple preferences associated with different regions. However, there was no apparent genetic separation by geographic origin among samples. The historic trees in the southwest might all share the same recent parents, or the parent diversity could have been obscured by small sample size. Efforts were made to avoid seedling apple trees during sampling. As mentioned in the “Materials and Methods” of this article, we focused sampling efforts on grafted as opposed to seedling trees; however, a small percentage of the samples are likely to be of seedling origin and not named cultivars. Seedling trees would not share the same “heirloom” status as named 19th century cultivars but may still possess useful traits or local adaptations.
For this study, we fingerprinted cultivars from the USDA National Germplasm Collection and from private nurseries to compare with the unknown samples. We restricted the number of reference cultivars to 109 likely introduced into the southwest. The existence of a DNA fingerprint database for correctly identified fruit tree cultivars such as those at the PGRU would allow studies such as this one to answer more questions about the identities of the many unknown apple trees growing on abandoned farms and in parks and forests across the country.
Apple plantings in a number of orchards and farmsteads cultivated in the 19th and early 20th centuries in the U.S. southwest still survive, although many have been abandoned. From 43 historic sites, 280 apple trees were sampled and compared with 109 cultivars at the PGRU and in private nurseries using microsatellite analysis. The 280 samples yielded 34 named cultivars and 110 unique genotypes. These results suggest that the historic orchards in the southwest had a high diversity of genotypes. Additional genetic fingerprinting of apple cultivars in the USDA PGRU will potentially enable us to identify the unknowns in this survey. Until such work is undertaken, these unknown genotypes should be conserved and analyzed for useful traits.
Literature Cited
Beach, S.A. 1905 The apples of New York. Volumes I & II. Report of the New York Agriculture Experiment Station for the year 1903 J.B. Lyon Company Albany, NY
Bussey, D. The apple in North America: A register of apple varieties listed in North American publications from 1802–1980, based on nomenclature of the apple. U.S. Department of Agriculture Division of Plant Industry Bulletin #56. Compiled by W.H. Ragan, 1905 (in press).
Calhoun, C.L. 1995 Old southern apples McDonald and Woodward Publishing Company Blacksburg, VA
Dennis F.G. Jr 2008 Malus ×domestica: Apple 661 674 Janick J. & Paull R.E. The encyclopedia of fruit and nuts CABI Publishing Cambridge, MA
Dunmire, C.C. 2004 Gardens of new Spain: How Mediterranean plants and foods changed America University of Texas Press Austin, TX
Ferree, D.C. & Warrington, I.J. 2003 Apples: Botany, production and uses CABI Publishing Cambridge, MA
Giafranceschi, L., Seglias, N., Tarchini, R., Komjane, M. & Gessler, C. 1998 Simple sequence repeats for the genetic analysis of apple Theor. Appl. Genet. 96 1096 1076
Goland, C. & Bauer, S. 2004 When the apple falls close to the tree: Local food systems and the preservation of diversity Renewable Agricultural Food Systems 19 228 236
Hammer, K., Arrowsmith, N. & Gladis, T. 2003 Agrobiodiversity with emphasis on plant genetic resources Naturwissenschaften 90 241 250
Hancock, J.M. 1999 Microsatellites and other simple sequences: Genomic context and mutational mechanisms 1 23 Goldstein D.B. & Schlötterer C. Reprinted in 2003. Microsatellites: Evolution and applications Oxford University Press Oxford, UK
Hedrick, U.P. 1950 A history of horticulture in America to 1860 Oxford University Press Oxford, UK
Hemmat, M.S.K.B. & Weeden, N.F. 2003 Mapping and evaluation of Malus ×domestica microsatellites in apple and pear J. Amer. Soc. Hort. Sci. 128 515 520
Hensley, T. 2005 A curious tale: The apple in North America 5 15 Hanson B. The best apples to buy and grow Brooklyn Botanic Garden, Inc Brooklyn, NY
Hokanson, S.C., Szewc-McFadden, A.K., Lamboy, W.F. & McFerson, J.R. 1998 Microsatellite (SSR) markers reveal genetic identities, genetic diversity and relationships in Malus ×domestica Borkh. core subset collection Theor. Appl. Genet. 97 671 683
Horgan, P. 1975 Lamy of Santa Fe: His life and times, a biography Farrar, Straus & Giroux New York, NY
Jackson, J.E. 2003 Biology of apples and pears Cambridge University Press Cambridge, UK
Peakall, R. & Smouse, P.E. 2006 GENALEX 6: Genetic analysis in Excel. Population genetic software for teaching and research. Mol. Ecol. Notes 6: 288–295 14 Nov. 2007 <http://www.anu.edu.au/BoZo/GenAlEx/>.
Pereira-Lorenzo, S., Ramos-Cabrer, A.M. & Díaz-Hernández, M.B. 2007 Evaluation of genetic identity and variation of local apple cultivars (Malus ×domestica Borkh.) from Spain using microsatellite markers Genet. Resources Crop Evol. 54 405 420
Ragan, W.H. 1905 Nomenclature of the apple: A catalogue of the known varieties referred to in American publications from 1804 to 1904. USDA-BPI Bulletin # 56 Government Printing Office Washington, DC
Silfverberg-Dilworth, E., Matasci, C.L., Van de Weg, W.E., Van Kaauwen, M.P., Walser, M., Kodde, L.P., Soglio, V., Gianfranceschi, L., Durel, C.E., Costa, F., Yamamoto, T., Koller, B., Gessler, C. & Patocchi, A. 2006 Microsatellite markers spanning the apple (Malus ×domestica Borkh.) genome Tree Genetics & Genomes 2 202 224
Smith, M.W.G. 1971 National apple register of the United Kingdom London Ministry of Agriculture, Fisheries and Food. 2002 reprint by Castlepoint Press Colvend, UK
Volk, G.M., Richards, C.M., Reilley, A.A., Henk, A.D., Forsline, P.L. & Aldwinckle, H.S. 2005 Ex situ conservation of vegetatively propagated species: Development of a seed-based core collection for Malus sieversii J. Amer. Hort. Sci. 130 203 210
Whealy, K. 2001 Fruit, berry, and nut inventory 3rd Ed Seed Savers Exchange Decorah, IA